Computer Engineering and Technology

2 downloads 0 Views 47MB Size Report
The registration desk will be situated on the 4th Floor at the Kyoto Research Park during the following time: .... Nathalia Devina Widjaja, Binus International University ...... two or more minimum hop count, it selects the smallest ID as the parent.
Conference Proceeding March 29-31, 2016 Kyoto, Japan

ACEAIT Annual Conference on Engineering and Information Technology

GLSBE

Global Conference on Life Science and Biological Engineering

ISFAS International Symposium on Fundamental and Applied Sciences

ACEAIT Annual Conference on Engineering and Information Technology ISBN 978-986-89298-6-9 GLSBE Global Conference on Life Science and Biological Engineering ISBN 978-986-5654-38-2 ISFAS International Symposium on Fundamental and Applied Sciences ISBN 978-986-89298-5-2

Content General Information for Participants .................................................................................... 13 International Committees .......................................................................................................... 15 International Committee of ACEAIT .................................................................................... 15 International Committee of GLSBE ...................................................................................... 16 International Committee of ISFAS ........................................................................................ 17 Special Thanks to Session Chairs ............................................................................................. 19 Conference Venue Information ................................................................................................ 22 Conference Schedule .................................................................................................................... 24 Natural Sciences Keynote Speech(I) ....................................................................................... 30 Natural Sciences Keynote Speech(II) ....................................................................................... 2 Oral Sessions ..................................................................................................................................... 3 Biological Engineering (1) ........................................................................................................ 3 GLSBE-19 ................................................................................................................................... 5 GLSBE-49 ................................................................................................................................... 7 GLSBE-51 ................................................................................................................................ 16 GLSBE-84 ................................................................................................................................ 23 Electrical Engineering and Technology (1) ...................................................................... 26 ACEAIT-7299 ......................................................................................................................... 27 ACEAIT-7429 ......................................................................................................................... 29 ACEAIT-7558 ......................................................................................................................... 30 Biological Engineering (2) ...................................................................................................... 31 GLSBE-45 ................................................................................................................................ 32 GLSBE-60 ................................................................................................................................ 34 GLSBE-63 ................................................................................................................................ 42 Computer Engineering and Technology (1) ..................................................................... 50 ACEAIT-7423 ......................................................................................................................... 52 ACEAIT-7320 ......................................................................................................................... 62 ACEAIT-7433 ......................................................................................................................... 71 ACEAIT-7434 ......................................................................................................................... 77 ACEAIT-7375 ......................................................................................................................... 84 ACEAIT-7623 ......................................................................................................................... 90 Environmental Engineering (1) & Green Technology ................................................... 98 ACEAIT-6536 ....................................................................................................................... 100 ACEAIT-6579 ....................................................................................................................... 109 ACEAIT-7312 ....................................................................................................................... 111 ACEAIT-7359 ....................................................................................................................... 113

3

ACEAIT-7568 ....................................................................................................................... 124 ISFAS-1478 .......................................................................................................................... 125 ISFAS-1525 .......................................................................................................................... 127 Mechanic and Engineering Sciences & Mechanical Engineering and Technology (1) ....................................................................................................................................................... 130 ACEAIT-7371 ....................................................................................................................... 132 ACEAIT-7377 ....................................................................................................................... 139 ACEAIT-7381 ....................................................................................................................... 156 ACEAIT-7419 ....................................................................................................................... 171 ACEAIT-7541 ....................................................................................................................... 178 Computer Engineering and Technology (2) .................................................................. 186 ACEAIT-7291 ....................................................................................................................... 188 ACEAIT-7413 ....................................................................................................................... 197 ACEAIT-7448 ....................................................................................................................... 205 ACEAIT-7467 ....................................................................................................................... 208 ACEAIT-7524 ....................................................................................................................... 219 Information Engineering and Technology (1) .............................................................. 224 ACEAIT-5239 ....................................................................................................................... 226 ACEAIT-5240 ....................................................................................................................... 233 ACEAIT-7416 ....................................................................................................................... 241 ACEAIT-7500 ....................................................................................................................... 249 ACEAIT-7319 ....................................................................................................................... 256 Materials Science and Engineering (1) ............................................................................ 258 ACEAIT-7289 ....................................................................................................................... 259 ACEAIT-7294 ....................................................................................................................... 260 ACEAIT-7314 ....................................................................................................................... 261 ACEAIT-7455 ....................................................................................................................... 262 Mechanical Engineering and Technology (2) ................................................................ 270 ISFAS-1471 .......................................................................................................................... 271 ISFAS-1542 .......................................................................................................................... 273 ACEAIT-7363 ....................................................................................................................... 274 ACEAIT-7439 ....................................................................................................................... 284 Computer Engineering and Technology (3) .................................................................. 294 ACEAIT-7405 ....................................................................................................................... 296 ACEAIT-7464 ....................................................................................................................... 307 ACEAIT-7627 ....................................................................................................................... 314 ACEAIT-7547 ....................................................................................................................... 321 ACEAIT-7437 ....................................................................................................................... 328 ACEAIT-7567 ....................................................................................................................... 336

4

Information Engineering and Technology (2) .............................................................. 347 ACEAIT-7292 ....................................................................................................................... 349 ACEAIT-7293 ....................................................................................................................... 355 ACEAIT-7350 ....................................................................................................................... 356 ACEAIT-7364 ....................................................................................................................... 366 ACEAIT-7428 ....................................................................................................................... 377 ACEAIT-7582 ....................................................................................................................... 386 Mechanical Engineering and Technology (3) ................................................................ 387 ACEAIT-7391 ....................................................................................................................... 388 ACEAIT-7408 ....................................................................................................................... 399 ACEAIT-7468 ....................................................................................................................... 405 ACEAIT-7475 ....................................................................................................................... 406 ACEAIT-7440 ....................................................................................................................... 413 Civil Engineering (1) .............................................................................................................. 419 ACEAIT-7349 ....................................................................................................................... 421 ACEAIT-7366 ....................................................................................................................... 429 ACEAIT-7384 ....................................................................................................................... 438 ACEAIT-7454 ....................................................................................................................... 447 ACEAIT-7589 ....................................................................................................................... 456 ACEAIT-7581 ....................................................................................................................... 462 ACEAIT-7542 ....................................................................................................................... 469 Engineering Sciences ............................................................................................................. 474 ISFAS-1477 .......................................................................................................................... 476 ISFAS-1479 .......................................................................................................................... 478 ISFAS-1480 .......................................................................................................................... 484 ISFAS-1538 .......................................................................................................................... 492 ISFAS-1596 .......................................................................................................................... 501 Material Science and Engineering (2) .............................................................................. 503 ACEAIT-7287 ....................................................................................................................... 504 ACEAIT-7328 ....................................................................................................................... 505 ACEAIT-7409 ....................................................................................................................... 506 ACEAIT-5617 ....................................................................................................................... 512 ACEAIT-7403 ....................................................................................................................... 513 Life Sciences (1) ....................................................................................................................... 524 GLSBE-33 .............................................................................................................................. 525 GLSBE-41 .............................................................................................................................. 533 GLSBE-42 .............................................................................................................................. 535 GLSBE-124............................................................................................................................ 538 Electrical Engineering and Technology (2) ................................................................... 544

5

ACEAIT-7360 ....................................................................................................................... 545 ACEAIT-7382 ....................................................................................................................... 556 ACEAIT-7394 ....................................................................................................................... 564 ACEAIT-7412 ....................................................................................................................... 575 ACEAIT-7306 ....................................................................................................................... 586 Biomedical Engineering........................................................................................................ 595 ACEAIT-7303 ....................................................................................................................... 596 ACEAIT-7355 ....................................................................................................................... 599 ACEAIT-7410 ....................................................................................................................... 601 ACEAIT-7446 ....................................................................................................................... 608 Chemical Engineering ............................................................................................................ 616 ACEAIT-7302 ....................................................................................................................... 617 ACEAIT-7315 ....................................................................................................................... 619 ACEAIT-7379 ....................................................................................................................... 625 ACEAIT-7594 ....................................................................................................................... 634 Life Sciences (2) ....................................................................................................................... 642 GLSBE-44 .............................................................................................................................. 643 GLSBE-61 .............................................................................................................................. 645 GLSBE-93 .............................................................................................................................. 647 GLSBE-89 .............................................................................................................................. 649 Information Engineering and Technology (3) .............................................................. 657 ACEAIT-7301 ....................................................................................................................... 659 ACEAIT-7325 ....................................................................................................................... 671 ACEAIT-7521 ....................................................................................................................... 677 ACEAIT-7519 ....................................................................................................................... 685 ACEAIT-5247 ....................................................................................................................... 696 ACEAIT-7556 ....................................................................................................................... 706 Chemical Sciences ................................................................................................................... 708 ISFAS-1487 .......................................................................................................................... 710 ISFAS-1511 .......................................................................................................................... 712 ISFAS-1548 .......................................................................................................................... 715 ISFAS-1590 .......................................................................................................................... 717 Geology, Earth and Environmental Sciences (1) & Environmental Engineering (2) ....................................................................................................................................................... 719 ACEAIT-7615 ....................................................................................................................... 721 ISFAS-1518 .......................................................................................................................... 722 ISFAS-1523 .......................................................................................................................... 730 ISFAS-1533 .......................................................................................................................... 732 ISFAS-1537 .......................................................................................................................... 739

6

Computer Engineering and Technology (4) .................................................................. 741 ACEAIT-7336 ....................................................................................................................... 743 ACEAIT-7431 ....................................................................................................................... 752 ACEAIT-7347 ....................................................................................................................... 762 ACEAIT-7317 ....................................................................................................................... 771 ACEAIT-7318 ....................................................................................................................... 778 ACEAIT-7368 ....................................................................................................................... 786 Food, Agriculture and Veterinary Sciences ................................................................... 794 ISFAS-1575 .......................................................................................................................... 795 ISFAS-1597 .......................................................................................................................... 797 ISFAS-1598 .......................................................................................................................... 799 ISFAS-1605 .......................................................................................................................... 808 ISFAS-1580 .......................................................................................................................... 810 Geology, Earth and Environmental Sciences (2) .......................................................... 816 ISFAS-1563 .......................................................................................................................... 817 ISFAS-1582 .......................................................................................................................... 825 ISFAS-1583 .......................................................................................................................... 833 Civil Engineering (2) .............................................................................................................. 844 ACEAIT-7497 ....................................................................................................................... 846 ACEAIT-7513 ....................................................................................................................... 847 ACEAIT-7522 ....................................................................................................................... 858 ACEAIT-7564 ....................................................................................................................... 873 ACEAIT-7549 ....................................................................................................................... 874 ACEAIT-7274 ....................................................................................................................... 885 ACEAIT-8092 ....................................................................................................................... 891 Nanotechnology& Physics Sciences .................................................................................. 902 ISFAS-1472 .......................................................................................................................... 904 ISFAS-1494 .......................................................................................................................... 914 ISFAS-1565 .......................................................................................................................... 923 ISFAS-1574 .......................................................................................................................... 929 ISFAS-1617 .......................................................................................................................... 936 Poster Sessions (1) .................................................................................................................... 944 Mechanical Engineering and Technology (1) / Engineering Sciences (1) / Green Technology / Food, Agriculture and Veterinary Sciences / Chemical Sciences (1) ....................................................................................................................................................... 944 ACEAIT-7290 ....................................................................................................................... 951 ACEAIT-7309 ....................................................................................................................... 959 ACEAIT-7326 ....................................................................................................................... 965 ACEAIT-7343 ....................................................................................................................... 967

7

ACEAIT-7362 ....................................................................................................................... 969 ACEAIT-7369 ....................................................................................................................... 976 ISFAS-1459 .......................................................................................................................... 977 ISFAS-1464 .......................................................................................................................... 978 ISFAS-1489 .......................................................................................................................... 983 ISFAS-1490 .......................................................................................................................... 989 ISFAS-1492 .......................................................................................................................... 995 ISFAS-1515 ....................................................................................................................... 1002 ISFAS-1601 ....................................................................................................................... 1004 ISFAS-1615 ....................................................................................................................... 1013 ISFAS-1634 ....................................................................................................................... 1020 ISFAS-1635 ....................................................................................................................... 1024 ISFAS-1631 ....................................................................................................................... 1030 ISFAS-1513 ....................................................................................................................... 1032 ISFAS-1465 ....................................................................................................................... 1034 ISFAS-1555 ....................................................................................................................... 1036 ISFAS-1578 ....................................................................................................................... 1037 ISFAS-1624 ....................................................................................................................... 1038 ISFAS-1495 ....................................................................................................................... 1044 ISFAS-1497 ....................................................................................................................... 1053 ISFAS-1522 ....................................................................................................................... 1059 ISFAS-1530 ....................................................................................................................... 1066 ISFAS-1558 ....................................................................................................................... 1068 ISFAS-1568 ....................................................................................................................... 1070 ISFAS-1576 ....................................................................................................................... 1076 ISFAS-1633 ....................................................................................................................... 1077 ISFAS-1567 ....................................................................................................................... 1081 Poster Sessions (3) .................................................................................................................. 1087 Computer Engineering and Technology (1) / Information Engineering and Technology (1)/ Materials Science and Engineering / Biomedical Engineering (1) ..................................................................................................................................................... 1087 ACEAIT-7324 .................................................................................................................... 1094 ACEAIT-7397 .................................................................................................................... 1105 ACEAIT-7430 .................................................................................................................... 1107 ACEAIT-7493 .................................................................................................................... 1108 ACEAIT-7504 .................................................................................................................... 1110 ACEAIT-7273 .................................................................................................................... 1122 ACEAIT-7373 .................................................................................................................... 1124 ACEAIT-7385 .................................................................................................................... 1133

8

ACEAIT-7389 .................................................................................................................... 1144 ACEAIT-7520 .................................................................................................................... 1145 ACEAIT-7296 .................................................................................................................... 1150 ACEAIT-7450 .................................................................................................................... 1158 ACEAIT-7342 .................................................................................................................... 1162 ACEAIT-7351 .................................................................................................................... 1169 ACEAIT-7557 .................................................................................................................... 1171 ACEAIT-7537 .................................................................................................................... 1178 ACEAIT-7559 .................................................................................................................... 1180 ACEAIT-7284 .................................................................................................................... 1182 ACEAIT-7338 .................................................................................................................... 1183 ACEAIT-7367 .................................................................................................................... 1184 ACEAIT-7393 .................................................................................................................... 1185 ACEAIT-7418 .................................................................................................................... 1187 ACEAIT-7479 .................................................................................................................... 1189 ACEAIT-7518 .................................................................................................................... 1190 ACEAIT-7308 .................................................................................................................... 1191 ACEAIT-7333 .................................................................................................................... 1193 ACEAIT-7421 .................................................................................................................... 1195 ACEAIT-7570 .................................................................................................................... 1196 ACEAIT-7551 .................................................................................................................... 1197 ACEAIT-7527 .................................................................................................................... 1198 ACEAIT-7526 .................................................................................................................... 1200 ACEAIT-7523 .................................................................................................................... 1202 ACEAIT-7346 .................................................................................................................... 1204 Poster Sessions (5) .................................................................................................................. 1215 Biological Engineering / Life Sciences (1) / Physics Sciences / Chemical Sciences (2) / Chemical Engineering ............................................................................................... 1215 GLSBE-14 ........................................................................................................................... 1223 GLSBE-25 ........................................................................................................................... 1225 GLSBE-68 ........................................................................................................................... 1232 GLSBE-92 ........................................................................................................................... 1233 GLSBE-110......................................................................................................................... 1235 GLSBE-97 ........................................................................................................................... 1237 GLSBE-4.............................................................................................................................. 1239 GLSBE-16 ........................................................................................................................... 1240 GLSBE-20 ........................................................................................................................... 1241 GLSBE-26 ........................................................................................................................... 1248 GLSBE-31 ........................................................................................................................... 1251

9

GLSBE-35 ........................................................................................................................... 1253 ISFAS-1442 ....................................................................................................................... 1255 ISFAS-1473 ....................................................................................................................... 1263 ISFAS-1491 ....................................................................................................................... 1272 ISFAS-1506 ....................................................................................................................... 1274 ISFAS-1508 ....................................................................................................................... 1276 ISFAS-1517 ....................................................................................................................... 1284 ISFAS-1544 ....................................................................................................................... 1287 ISFAS-1606 ....................................................................................................................... 1292 ISFAS-1433 ....................................................................................................................... 1294 ISFAS-1435 ....................................................................................................................... 1296 ISFAS-1436 ....................................................................................................................... 1298 ISFAS-1437 ....................................................................................................................... 1301 ISFAS-1475 ....................................................................................................................... 1304 ISFAS-1503 ....................................................................................................................... 1307 ISFAS-1505 ....................................................................................................................... 1308 ISFAS-1551 ....................................................................................................................... 1310 ISFAS-1632 ....................................................................................................................... 1312 ACEAIT-7332 .................................................................................................................... 1314 ACEAIT-7339 .................................................................................................................... 1315 ACEAIT-7327 .................................................................................................................... 1327 Poster Sessions (6) .................................................................................................................. 1339 Life Sciences (2) / Environmental Engineering / Geology, Earth and Environmental Sciences / Engineering Sciences (2) / Biomedical Engineering (2) ..................................................................................................................................................... 1339 GLSBE-36 ........................................................................................................................... 1346 GLSBE-38 ........................................................................................................................... 1347 GLSBE-46 ........................................................................................................................... 1349 GLSBE-47 ........................................................................................................................... 1351 GLSBE-48 ........................................................................................................................... 1353 GLSBE-53 ........................................................................................................................... 1355 GLSBE-64 ........................................................................................................................... 1358 GLSBE-23 ........................................................................................................................... 1365 GLSBE-65 ........................................................................................................................... 1367 GLSBE-80 ........................................................................................................................... 1380 GLSBE-81 ........................................................................................................................... 1382 GLSBE-83 ........................................................................................................................... 1384 GLSBE-87 ........................................................................................................................... 1386 GLSBE-88 ........................................................................................................................... 1404

10

GLSBE-96 ........................................................................................................................... 1419 GLSBE-99 ........................................................................................................................... 1421 GLSBE-101......................................................................................................................... 1422 GLSBE-119......................................................................................................................... 1424 GLSBE-122......................................................................................................................... 1426 ACEAIT-7552 .................................................................................................................... 1429 ACEAIT-7571 .................................................................................................................... 1430 ISFAS-1483 ....................................................................................................................... 1432 ISFAS-1512 ....................................................................................................................... 1433 ISFAS-1535 ....................................................................................................................... 1435 ISFAS-1536 ....................................................................................................................... 1436 ISFAS-1498 ....................................................................................................................... 1438 ISFAS-1560 ....................................................................................................................... 1445 ISFAS-1534 ....................................................................................................................... 1454 ACEAIT-7482 .................................................................................................................... 1457 Poster Sessions (7) .................................................................................................................. 1459 Mechanical Engineering and Technology (2) / Mechanic and Engineering Sciences / Nanotechnology / Civil Engineering / Electrical Engineering and Technology / Information Engineering and Technology (2)/ Computer Engineering and Technology (2) ....................................................................................................................... 1459 ACEAIT-7370 .................................................................................................................... 1465 ACEAIT-7387 .................................................................................................................... 1475 ACEAIT-7478 .................................................................................................................... 1487 ACEAIT-7402 .................................................................................................................... 1488 ACEAIT-7461 .................................................................................................................... 1496 ACEAIT-7506 .................................................................................................................... 1507 ACEAIT-7538 .................................................................................................................... 1512 ACEAIT-7614 .................................................................................................................... 1523 ACEAIT-7686 .................................................................................................................... 1532 ACEAIT-7297 .................................................................................................................... 1544 ISFAS-1470 ....................................................................................................................... 1546 ISFAS-1579 ....................................................................................................................... 1557 ISFAS-1482 ....................................................................................................................... 1568 ISFAS-1577 ....................................................................................................................... 1570 ISFAS-1593 ....................................................................................................................... 1572 ACEAIT-7555 .................................................................................................................... 1575 ACEAIT-7561 .................................................................................................................... 1576 ACEAIT-7459 .................................................................................................................... 1578 ACEAIT-7606 .................................................................................................................... 1580

11

ACEAIT-7532 .................................................................................................................... 1585 ACEAIT-7560 .................................................................................................................... 1588 ACEAIT-7607 .................................................................................................................... 1597 ACEAIT-7611 .................................................................................................................... 1605 ACEAIT-7610 .................................................................................................................... 1612 ACEAIT-7543 .................................................................................................................... 1620 ACEAIT-7424 .................................................................................................................... 1623

12

General Information for Participants 

Registration

The registration desk will be situated on the 4th Floor at the Kyoto Research Park during the following time: 13:00-17:00 Tuesday, March 29, 2016 08:00-17:30 Wednesday, March 30, 2016 08:30-17:30 Thursday, March 31, 2016



Organizer Higher Education Forum (HEF)

Tel: + 886 2 2740 1498〡www.prohef.org



A Polite Request to All Participants

Participants are requested to arrive in a timely fashion for all addresses. Presenters are reminded that the time slots should be divided fairly and equally by the number of presentations, and that they should not overrun. The session chair is asked to assume this timekeeping role and to summarize key issues in each topic.



Certificate

Certificate of Presentation or Certificate of Attendance A certificate of attendance includes participant’s name and affiliation, certifying the participation in the conference. A certificate of presentation indicates a presenter’s name, affiliation and the paper title that is presented in the scheduled session. Certificate Distribution Oral presenters will receive a certificate of presentation from the session chair after their presentations or at the end of the session. Poster presenters will receive a certificate of presentation from the conference staff at the end of their poster session.

13

The certificate of presentation will not be issued, either at or after the conference, to authors whose papers are registered but not presented. Instead, the certificate of attendance will be provided after the conference.



Preparation for Oral Presentations

All presentation rooms are equipped with a screen, an LCD projector, and a laptop computer installed with Microsoft PowerPoint. You will be able to insert your USB flash drive into the computer and double check your file in PowerPoint. We recommend you to bring two copies of the file in case that one fails. You may also connect your own laptop to the provided projector; however please ensure you have the requisite connector.

Preparation for Poster Presentation Materials Provided by the Conference Organizer: 1. Poster Panel (90cm×173cm) 2. Push pins Materials Prepared by the Presenters: 1. Home-made poster(s) 2. Material: not limited, can be posted on the poster panel 3. Maximum poster size is A0

14

International Committees International Committee of ACEAIT Subhananda Chakrabarti, Indian Institute of Technology Bombay Maryam Jalalifarahani, Islamic Azad University Sunyang Chung, Konkuk University Ravi Saxena, Nirma University Mahadev Haridas Gavhane, SRTMU Nanded UGC Abbas Madandar Arani, Lorestan University Dharam Vir Mahajan, N. A. S. Post-graduate College Jagpreet Kaur, Punjabi University Ilango Ponnuswami, Bharathidasan University Aijaz Ali Wassan, University of Sindh Raj Kumar Yadav, St. Xaviers College Rajan Mathew, University of Swaziland Kamlesh Mahajan, Ismail National Girls Post-graduate College Cesar Correa Arias, University of Guadalajara Habibollah Abolhassan Shirazi, Tehran Azad University Fu Wai Chee Ada, The Chinese University of Hong Kong Mukti Kanta Mohanty, Autonomous College Raju Pandurangji Lipte, Nagpur University Asadullah Naghdi, Bu-ali sina university Eduardo Erazo Acosta, University of Nariño Mu-Song Chen, Da-Yeh University Ir. Dr. Hj. Ramli Nazir, Universiti Teknologi Malaysia Adel Hejaaji, Engineering Services Management Limited Dr P.Sanjeevikumar, Head, Research and Development, Ohm Technologies, India Jian-Horng Chen, Chung Shan Medical University, Taiwan

15

International Committee of GLSBE Ahmad Zuhairi Abdullah, School of Chemical Engineering, Universiti Sains Malaysia Fadzilah Adibah Abdul Majid, Department of Bioprocess Engineering, Universiti Teknologi Malaysia Mohd Farid bin Atan, Department of Chemical Engineering and Energy Sustainability, University of Sarawak Malaysia Sue-Joan, Chang, Department of Life Sciences, National Cheng Kung University, Taiwan Yen-Chung, Chang,Institute of Molecular Medicine, National Tsing Hua University, Taiwan Yun-Peng, Chao, Department of Chemical Engineering, Feng Chia University, Taiwan Pei-Jen, Chen, Department of Agricultural Chemistry, National Taiwan University, Taiwan H.M. El-Shora, Department of Botany, Mansoura University, Egypt I-Ming Hsing, Department of Chemical and Biomolecular Engineering, The Hong Kong University of Science and Technology, Hong Kong Shang-Da, Huang, Department of Chemical, National Tsing Hua University, Taiwan Bing Joe Hwang, Department of Chemical Engineering, National Taiwan University of Science and Technology, Taiwan Sungjee Kim, Department of Chemical Engineering, Sungkyunkwan University, Korea Li-Fen, Lei, Department of Agricultural Economics, National Taiwan University, Taiwan Mihir Kumar Purkait, Department of Chemical Engineering, Indian Institute of Technology Guwahati, India Chung-Sung, Tan, Department of Chemical Engineering, National Tsing Hua University, Taiwan Tewin Tencomnao, Department of Clinical Chemistry, Chulalongkorn University, Thailand S.K. Tripathi, Department of Water Resource, Indian Institute of Technology Poorkee, India Henry N.C. Wong, Department of Chemical, The Chinese University of Hong Kong, Hong Kong Yusri b. Yusup , School of Industrial Technology, Universiti Sains Malaysia, Malaysia Hasan Akhtar Zaidi, University School of Chemical Technology, Guru Gobind Singh Indraprastha University, India

16

International Committee of ISFAS Khalid M. Mosalam, University of California, Berkeley Chueerat Jaruskulchai, Kasetsart University M. Cheralathan, SRM University J N Bandyopadhyay, Indian Institute of Technology Kharagpur S. Dhar, University of Calcutta Poongothai Shankar, Annamalai University Amit Agrawal, Indian Institute of Technology Bombay Cheng Li, The Hong Kong Polytechnic University T.M. Indra Mahlia, University of Malaya Kunal Ghosh, Indian Institute of Technology Kanpur Narayanan Kulathuramaiyer, University of Sarawak Malaysia Arup K. Sarma, Indian Institute of Technology Guwahati Suresh K Bhargava, School of Applied Sciences Rapeepan Pitakaso, Ubonratchthani University Jie Liu, Carleton University Carlos Alejandro Figueroa, Plasmar Tecnologia Dong-Ho Ha, Konkuk University Poul Vaeggemose, VIA University College Denmark Maha M. O. Khayyat, Umm Al-Qura University Kant Kanyarusoke, Cape Peninsula University of Technology Paramita Bhattacharjee, Jadavpur University Nathalia Devina Widjaja, Binus International University C. M. Khalique, North-West University,South Africa Banerji P, Indian Institute of Technology Kharagpur P.K. Ghosh, Indian Institute of Technology Poorkee E George Dharma Prakash Raj, Bharathidasan University R.P.Bhatnagar, Birla Institute of Technology V. Vijayagopal, Annamalai University Amit Awekar, Indian Institute of Technology Guwahati Pui-In Mak, University of Macau E,Rathakrishnan, Indian Institute of Technology Kanpur Bassim H. Hameed, University of Science Malaysia Sudhirkumar Barai, Indian Institute of Technology Kharagpur S. N. Sarkar, Calcutta University Samit Bhattacharya, Indian Institute of Technology Kanpur A. P. Shashikala, Birla Institute of Technology RM. Senthamarai, Annamalai University

17

Arnab Bhattacharya, Indian Institute of Technology Kanpur Zbigniew Michalewicz, University of Adelaide B. Bhattacharya, Indian Institute of Technology Kharagpur Amin Heidarpour, Monash University Faizal Mustapha, Universiti Putra Malaysia Susanta Banerjee, Indian Institute of Technology Kharagpur M.V.L.R Anjaneyalu, Birla Institute of Technology P. Balasubramanian, Universiti Teknologi PETRONAS Hui Tong Chua, University of Western Australia Andrew Whyte, Curtin University Johnson Agbinya, La Trobe University Yuen Ka Veng, University of Macau ShahNor Basri, Universiti Putra Malaysia T. V. Gopal, SRM University Surendra Kumar, Indian Institute of Technology Poorkee K. K. Saju, Cochin University of Science and Technology

18

Special Thanks to Session Chairs Mingshan Zhang

New Jersey City University

Jen-Wei Cheng

National Taiwan University of Science and Technology

Wen-Tai Chiu

National Cheng Kung University

Michiko Miyamoto

Akita Prefectural University

Chen-Yin Chiang

National Central University

Hei Wong

City Universtiy of Hong Kong

Chi-Chun Chen

National Chin-Yi University of Technology

Wei-Mei Chen

National Taiwan University of Science and Technology

Andrei Giniatoulline

Los Andes University

Sangho Kim

Ritsumeikan Asia Pacific University

Jacqueline Kin-sang Chan

Hong Kong Institute of Education

Arpita Ghosh

IIM Calcutta

Yu-Chao Huang

National Dong Hwa University

Gwowen Shieh

National Chiao Tung University

Todd W Kenreich

Towson University

Parkpoom Sriromreun

Srinakharinwirot University

Chun-Hung Cheng

The Chinese University of Hong Kong

Kairat Jaroenrat

Kasetsart University

Paranee Sriromreun

Srinakharinwirot University

Wei-Mon Yan

National Taipei University of Technology

Yao Shieh

Chang Gung University

Chunshien Li

National Central University

19

William Chan

University of Hong Kong

Francisco DP Daez

Meycauayan College

Shey-Huei Sheu

Providence University

Naai Jung Shih

National Taiwan University of Science and Technology

Wei-Mon Yan

National Taipei University of Technology

Mei-Ya Wang

Shih-Hsin University

Lilibeth Sabejon Galvez

Davao Oriental State College of Science and Technology

Shingjiang Jessie Lue

Chang Gung University

S. Sankaralingam

Madurai Kamaraj University

Sri Issundari

Universitas Pembangunan Nasional

Pei-Jung Yang

National Chengchi University

Ching-Wen Chang

Minghsin University of Science and Technology

Jung Whan Park

Yeungnam University

Cheng-Tung Chou

National Central University

Nanta Sooraksa

National Institute of Development Administration

To-Ming Ho

City University of Hong Kong

Yeon Ho Chung

Pukyong National University

Hyoung Jin Choi

Inha University

Diane Masuo

University of Hawaii at Mānoa

Pei-Yu Lin

Yuan Ze University

Malgorzata Rozkwitalska

Gdansk School of Banking

S. Kong Wang

I–Shou University

Chen-Wei Chiu

National Taiwan Police College

Zuhaimy Ismail

Universiti Teknologi Malaysia

20

Xubing Zhang

Hong Kong Polytechnic University

D. Kathiresan

Madurai Kamaraj University

21

Conference Venue Information Kyoto Research Park Tel: +81-75-322-7800 Address: 134, Chudoji Minami-machi, Shimogyo-ku, Kyoto 600-8813, Japan Located at the ancient capital of Japan, Kyoto Research Park (KRP) is set at a convenient spot. KRP has a wide variety of networks involving universities, industries, administrative institutes, public research institutes, and economic bodies. KRP offers various of conference rooms, offices, and labs. The conference will be held at Hall 1 inside the research park. Hall 1 is near to JR Tanbaguchi-Eki.

22

Floor Map (4th Floor)

23

Conference Schedule

Tuesday, March 29, 2016 Oral Session (4th Floor) Time

Schedule

Venue

13:00-17:00

Registration

Foyer Area

14:00-15:30

Biological Engineering (1)

Room C

15:30-15:45

Tea Break & Networking

Foyer Area

15:45-17:45

Electrical Engineering and Technology (1)

24

Room C

Wednesday, March 30, 2016 Oral Session (4th Floor) Time

Schedule

Venue

08:00-17:30

Registration

Foyer Area

Biomedical Engineering (2)

Room A

08:30-10:00 Computer Engineering and Technology (1)

Room B

10:00-10:20

Environmental Engineering(1) & Green Technology

Room C

Tea Break & Networking

Foyer Area

Natural Science Keynote Speech(1): Sergei Gorlatch Topic: Future Distributed Applications Based on Mobile Cloud Computing and Software-Defined 10:20-12:00 Networks

Science Hall

Natural Science Keynote Speech (2): Kazuaki Maeda Topic: Impressive Slideshow without Presentation Software 12:00-13:00

Lunch

25

Atrium

Wednesday, March 30, 2016 Oral Session (4th Floor)

13:00-14:30

Mechanic and Engineering Sciences & Mechanical Engineering and Technology (1)

Room A

Computer Engineering and Technology (2)

Room B

Information Engineering and Technology (1)

Room C

Materials Science and Engineering (1) 14:30-14:50

14:50-16:20

16:20-16:30

16:30-18:00

Tea Break & Networking

AV Room Foyer Area

Mechanical Engineering and Technology (2)

Room A

Computer Engineering and Technology (3)

Room B

Information Engineering and Technology (2)

Room C

Tea Break& Networking

Foyer Area

Mechanical Engineering and Technology (3)

Room A

Civil Engineering (1)

Room B

Engineering Sciences

Room C

26

Wednesday, March 30, 2016 Poster Session (Foyer Area, 4th floor) Time

Schedule Poster Session (1)

09:40-10:40

Mechanical Engineering and Technology (1) / Engineering Sciences(1) / Green Technology / Food, Agriculture and Veterinary Sciences / Chemical Science (1) Poster Session (3)

13:30-14:30

Computer Engineering and Technology (1) / Information Engineering and Technology (1) / Materials Science and Engineering / Biomedical Engineering (1) Poster Session (5)

16:30-17:30

Biological Engineering / Life Sciences (1) / Physics Sciences / Chemical Sciences / Chemical Engineering (2)

27

Thursday, March 31, 2016 Oral Session (4th Floor) Time

Schedule

Venue

08:30-17:30

Registration

Foyer Area

09:00-10:30 10:30-10:40

Materials Science and Engineering (2)

Room A

Life Sciences (1)

Room B

Tea Break & Networking Electrical Engineering and Technology (2)

10:40-12:10 Biomedical Engineering Chemical Engineering 12:10-13:10

Room A Room B Room C

Lunch Life Sciences (2)

Atrium Room A

13:10-14:40 Information Engineering and Technology (3) Chemical Sciences

Room B Room C

14:40-14:50

Tea Break & Networking Geology, Earth and Environmental Sciences (1) Environmental Engineering(2) 14:50-16:20 Computer Engineering and Technology (4) Food, Agriculture and Veterinary Sciences 16:20-16:30

Foyer Area

Foyer Area Room A Room B Room C

Tea Break & Networking

Foyer Area

Geology, Earth and Environmental Sciences (2)

Room A

16:30-18:00 Civil Engineering (2)

Room B

Nanotechnology & Physics Sciences

28

Room C

Thursday, March 31, 2016 Poster Session ( Foyer Area , 4th floor) Time

Schedule Poster Session (6)

09:40-10:40 Life Sciences (2) / Environmental Engineering / Geology, Earth and Environmental Sciences/ Engineering Sciences(2) / Biomedical Engineering (2) Poster Session (7) Mechanical Engineering and Technology (2) / Mechanic 11:30-12:00 and Engineering Sciences / Nanotechnology / Civil Engineering/ Electrical Engineering and Technology/ Information Engineering and Technology (2)/ Computer Engineering and Technology (2)

29

Natural Sciences Keynote Speech(I) Science Hall 10:20-12:00

Wednesday March 30 , 2016

Topic: Future Distributed Applications Based on Mobile Cloud Computing and Software-Defined Networks Sergei Gorlatch Professor Computer Science University of Muenster Germany

Abstract We consider an emerging class of challenging networked multimedia applications called Real-Time Online Interactive Applications (ROIA). ROIA are networked applications connecting a potentially very high number of users who interact with the application and with each other in real time, i.e., a response to a user’s action happens virtually immediately. Typical representatives of ROIA are multiplayer online computer games, advanced simulation-based e-learning and serious gaming. All these applications are characterized by high performance and QoS requirements, such as: short response times to user inputs (about 0.1-1.5 s); frequent state updates (up to 100 Hz); large and frequently changing numbers of users in a single application instance (up to tens of thousands simultaneous users). This talk will address two challenging aspects of future Internet-based ROIA applications: a) using Mobile Cloud Computing for allowing high application performance when a ROIA application is accessed from multiple mobile devices, and b) managing dynamic QoS requirements of ROIA applications by employing the emerging technology of Software-Defined Networking (SDN).

30

Brief Introduction of Dr. Sergei Gorlatch Sergei Gorlatch has been Full Professor of Computer Science at the University of Muenster (Germany) since 2003. Earlier he was Associate Professor at the Technical University of Berlin, Assistant Professor at the University of Passau, and Humboldt Research Fellow at the Technical University of Munich, all in Germany. Prof. Gorlatch has about 200 peer-reviewed publications in renowned international books, journals and conferences. He was principal investigator in several international research and development projects in the field of parallel, distributed, Grid and Cloud algorithms, networking and computing, as well as e-Learning, funded by the European Commission and by German national bodies. Among his recent achievements in the area of communications and future Internet is the novel Real-Time Framework (www.real-time-framework.com) developed in his group as a platform for high-level development of real-time, highly interactive applications. In the area of networking, his group has been recently working in the pan-European project OFERTIE on an application-oriented Quality of Service approach for emerging Software-Defined Networks (SDN).

Natural Sciences Keynote Speech(II) Science Hall 10:20-12:00

Wednesday March 30 , 2016

Topic: Impressive Slideshow without Presentation Software Kazuaki Maeda Professor Department of Business Administration and Information Science Chubu University Japan

Abstract When we give a presentation at conferences, we usually use one of desktop presentation software such as Powerpoint, OpenOffice.org and Keynote. This talk shows how to create slides and run a slideshow without the presentation software. We can use open source software such as web browsers, LaTeX and JavaFX to make the slideshow. It supports running the slideshow with eye-catching transitions. Brief Introduction of Dr. Kazuaki Maeda He graduated from Graduate School of Science and Technology in Keio University. After leaving Keio University, he got an academic position at Chubu University in Japan. Now he is a professor at Department of Business Administration and Information Science at Chubu University. His research interests are Web Applications, Open Source Software, Domain Specific Languages, Compiler Construction, Object-Oriented Programming.

2

Oral Sessions Biological Engineering (1) Tuesday, March 29, 2016

14:00-15:00

Room C

Session Chair: Prof. Wen-Tai Chiu GLSBE-19 Real Time Recording of Ca2+ Mobilization and Histamine Degranulation in Living Single Mast Cell Wen-Tai Chiu︱National Cheng Kung University Yu-Chun Chang︱National Cheng Kung University GLSBE-49 The Effects of Sound and Light Headphones on Sleep Evaluated with EEG and HRV Machine Xiang-Chao Chang︱Ming Chuan University Yue Chen︱Ming Chuan University Jih-Huah Wu︱Ming Chuan University GLSBE-51 Study of Autonomic Nervous Activity and Sleep Latency of Elderly Women Treated with LED Radiation Wei-Hao Chen︱Ming Chuan University Shau-Ming Tseng︱Ming Chuan University Yang-Chyuan Chang︱Min-Sheng General Hospital Hui-Ling Chiu︱Min-Sheng General Hospital Wei Fang︱National Taiwan University Yi-Chia Shan︱National Taiwan University Jih-Huah Wu︱Ming Chuan University

3

GLSBE-84 The Effect of Mechanical Properties of Polyhydroxyalkanoates (PHAs) Three-Dimensional (3D) Scaffolds on Proliferation of Human Gingival Fibroblasts (HGFs) Seubsakul Phuegyod︱Mahidol University Nungnit Wattanavichean︱Mahidol University Rudee Surarit︱Mahidol University Nuttawee Niamsiri︱Mahidol University

4

GLSBE-19 Real Time Recording of Ca2+ Mobilization and Histamine Degranulation in Living Single Mast Cell Wen-Tai Chiua, Yu-Chun Changb a Department of Biomedical Engineering, National Cheng Kung University, Tainan 70101, Taiwan E-mail address: [email protected] b

Department of Biomedical Engineering, National Cheng Kung University, Tainan 70101, Taiwan E-mail address: [email protected]

1. Background Mast cells have been reported to play a primary role in allergic diseases. During allergy, mast cell activation initiated by FcεRI cross-linking by IgE bound antigen results in degranulation of histamine. However, other receptors expressed on mast cells, such various G protein-coupled receptors (GPCRs), can advance degranulation when they are activated. Nowadays, the spatio-temporal relationship between Ca2+ mobilization and histamine degranulation by different stimulation is little known. The specific aim of this study is to investigate the relationship between calcium oscillation patterns and histamine degranulation after stimulating by the immunological and nonimmunological pathways in rat RBL-2H3 mast cells. 2. Methods The RBL-2H3 cell line from rat mast cells was used in this study. Fluorescent dyes calcein and Nile Red with Ni2+ complex were using as the histamine probes. Real time recording of histamine degranulation and Ca2+ oscillation were performed under total internal reflection fluorescence microscopy (TIRFM), which is used for visualizing localization or interaction of fluorescent molecules in the near-membrane region (~200 nm) of living cells. For stimulation of histamine degranulation, IgE-sensitized cells were treated with TNP-BSA in activation of FcεRIs, whereas compound 48/80 was applied to activate GPCRs. On the othe hand, two fluorescence Ca2+ probes, Fluo-4/AM and Calcium Orange, were used to measure intracellular Ca2+ levels. 3. Expected Results/ Conclusion/ Contribution The following results were demonstrated in this study: (1) the central Ca2+ levels were increased first, and the peripheral Ca2+ levels were subsequently increased from compound 48/80 treatment. While the peripheral Ca2+ levels were increased, the histamine degranulation subsequently occurred due to stimulation with compound 48/80. However, activation of TNP-BSA immediately triggered the elevation of Ca2+ and the histamine degranulation events were subsequently occurred. (2) When the cells were preincubated with Ca2+ chelator (BAPTA/AM), no intracellular Ca2+ oscillation and histamine degranulation were detected with different stimulators in 0 mM Ca2+ buffer. (3) The rate of histamine degranulation in the 0 mM Ca2+ buffer was significantly slower than in normal medium using different stimulators. (4) Inhibition of store-operated Ca2+ entry (SOCE), the major form of extracellular Ca2+ influx in non-excitable cells, by the blocker 2-APB or knockdown of Orai1, then the elevation of total intracellular Ca2+ levels was significantly suppressed and the rate of histamine degranulation became significantly slower using compound 48/80 activation. 4. Conclusion 5

Taken these results together, intracellular Ca2+ is sufficient for histamine degranulation; however, extracellular Ca2+ promotes histamine degranulation. Furthermore, there are different patterns of Ca2+ oscillation near the plasma membrane by different signaling pathways. We established the platform to measure histamine degranulation and calcium oscillation patterns by stimulation with different stimulators. The established platform may be used to screen drugs for allergy disease in the future. Keywords: Ca2+, histamine degranulation, mast cell, TIRFM

6

GLSBE-49 The Effects of Sound and Light Headphones on Sleep Evaluated with EEG and HRV Machine Yue Chena, Xiang-Chao Changb, Wei-Hao Chenc, Shau-Ming Tsengd, Jih-Huah Wue* Department of Biomedical Engineering, Ming Chuan University, Taoyuan City, ROC e Department of Biomedical Engineering, Ming Chuan University, Taoyuan City, ROC *The corresponding author Email: [email protected]

a, b, c, d

Abstract To develop some physical therapies to cure insomnia is a very important issue. One of the new therapies is the combination of the music adjunctive therapy and using lower-energy LED to radiate acupuncture points. The shape of this machine is like an earphones. It was used to stimulate acupuncture points TF1, SC10, IC6, AT3, T6, and IC8 on the ear with LED light and to do vocal stimulation with the music combined α wave. The effects of this device was assessed with a simple EEG machine and a HRV device before, during and after stimulation. There were two experiments conducted in this study, 40 college students and 20 elderly women were recruited respectively. In light and sound group, the increase of SDNN has significant meaning after stimulation. On the other hand, the falling sleep time in light and light combined sound group is short than that in the placebo group. Even it doesn‘t have significant meaning, however the tendency can be seen in this study. It is worthy to do further study. Keywords: Low Power LED, Insomnia, Autonomic nervous detector, EEG, Sound and Light Headphones 1. Background Due to the advances in medicine and science, many new applications in medicine, including new diagnostic and treatment which combined photoelectric technology and photo therapy are being developed recently. Most sleep and cardiovascular studies discovered that sleep disorder patients have higher risk to suffer cardiovascular diseases [1]. Using low-energy light to stimulate the acupuncture points is an alternative treatment in traditional Chinese medicine, it has advantages not only non-invasive, non-pain, less side effect and easy to use etc., but also affect the autonomic nervous system [2]. Two studies showed that the palm stimulated by low level laser and LED can affect the brain wave of the subject [3-4]. According to the ear acupuncture therapy proposed by Dr. P. Nogier, ears like an embryos‘ shape, and its position and acupoint have close association. In addition, ears have very dense nerves plexus and acupoints, and stimulating specific acupoint can adjust physiological phenomenon proved in clinic [5]. Many studies proved that pressing specific area can help regulate physiology [6-7], but light stimulation mentioned very few. In this study, six acupoints TF1, SC10, IC6, AT3, T6 and IC8 were stimulated with LED light. In general, these points were chosen to help relax body, fall asleep easily, adjust autonomic nervous system etc. On the other hand, audible stimulation is another method to enhance specific brainwave, some articles mentioned that using specific music can reduce β wave and increase α wave [8-10]. The main goal in this study is to use low-energy LED and music stimulation on ears to make subjects relax and fall asleep easily, the effects were assessed with a simple Electroencephalogram (EEG) and autonomic nervous system (ANS) monitor. 7

2.

Methods

2.1 Experimental subject This experiment was conducted with a light headphones, assessed with an ANS monitor and an EEG device, and the collected data were analyzed with SPSS software. This experiment was divided into two parts. In the first experiment, 20 women whose age over 40 years old were recruited. They all fall asleep before 11 pm, four times a week, sustained more than a month. The second experiment, 40 students aged from 18 to 23 years old were recruited from Ming Chun University. They all fall asleep after 1 am, four times a week, sustained more than a month. Before starting the experiment, each participant was required to give written informed consent, basic document and Pittsburg Sleep Quality Index. 2.1.1 Exclusion criteria included Ⅰ.The subject was taking any medicine. Ⅱ. Patients with cardiovascular disease. Ⅲ. Patients with chronic diseases. 2.2 Experimental Procedure The ―single-blind randomized trial‖ was used in this study. Each experiment period is three days, subjects can‘t eating foods that contain caffeine before the experiment. The subjects were requested to sleep at least eight hours the night before the experiment. The subjects were randomly assigned to three groups: group A (no stimulation), group B (light stimulation) and group C (light combined sound stimulation). Every process was presented in Fig1.

2.3Experimental equipment 2.3.1 Sound and Light Headphones In this study we use the Sound and Light Headphones produced by WellMike Corp., and its LED‗s wavelength is 660nm. Six acupoints TF1, SC10, IC6, AT3, T6 and IC8 were radiated with this LED light (Fig2.).The operating frequency of LED is 10Hz in the first six minutes, then become 6Hz in the last six minutes. The music that combine α wave and normal music can make the breath slowly and decrease blood pressure [8].

8

2.3.2 Autonomic nervous system (ANS) monitor The ANS monitor (Daily-Care Biomedical Corp., Taoyuan, Taiwan) was used to measure ANS activity. The ANS monitor pastes were attached to the wrists of the subject. The ANS of the subject were monitored before and after stimulation, and 10 minutes after wake up. 2.3.3 EEG An electroencephalograph (EEG2000 serial, Wan-Zhi Corp., Taipei, Taiwan) was used in this study. The potential of bipolar electrodes were recorded on Fp1-A2 and Fp2-A1. The data obtained from bipolar recording were shown in time domain; for each session, the four band powers of valid epochs were averaged, and FFT maps of different session were performed by the software (EEG2000 ver.36TW). When the dominant brainwave of the subject changes from beta wave to alpha wave or theta wave, and it can last at least 10 seconds, the subject was determined that he or she is drowsy or falling asleep. 2.3.4 Statistical Analysis. Two-tailed paired t-test was applied to compare the difference of ANS activity as well as EEG before and after headphones stimulation. The mean and standard deviation of the calculated values were expressed as ―Mean ± SD‖. All the statistical analyses were executed with SPSS18 software, the results will be recognized as statistical significant while p-value B> A, this result was as expected. In the LF / HF, group B and C have upward trend, and the change is apparent in group C. Finally, the average time of fall asleep is shorter in group B than that in group C as shown in Fig.6.

9

10

3.2 College student In the second experiment, the college student study, the statistics before sleep and after sleep are presented in Table 2. The statistics after sleep and wake up 10 minutes are presented in Table 3, the analysis of HR, LF/HF, SDNN and the average of fall asleep time are shown in Fig.7, Fig.8, Fig.9 and Fig.10 respectively. The SDNN‘s p-value is less than 0.05 for each group, which means the SDNN of the subjects have been changed after twenty minutes rest. However, the change of LF / HF in group A have significant meaning after wake up ten minutes. In the third ANS measurement, the SDNN still have statistically significant in group B, it means the effect can be maintained after light stimulation. On the other hand, the average time of fall asleep is shorter in group B and C than group A as shown in Fig.10, even it doesn‘t have statistically significance, only light and light combined sound stimulation can make the time of fall asleep shorter.

11

12

13

We found the change of SDNN values have significant meaning in group C in both two experiments, so the light and sound stimulation can increase the activity of the ANS. The trend of fall asleep time in the group B is apparent, even it doesn‘t have significant meaning. In the future, the parameters in this study will be changed, like as using blue light or near-infrared light to replace the red light to achieve the experiment effect that we expect. 4. References [1] Taiwan Society of Sleep Medicine journal (2013) (in Chinese) [2]Jih-Huah Wu, Hsin-Yang Chen, M.S., Yu-Ju Chang, M.S., Hung-Chien Wu, M.D., Wen-Dien Chang, M.S., Yi-Jing Chu, B.E., and Joe-Air Jiang, Ph.D. (2009) Study of Autonomic Nervous Activity of Night Shift Workers Treated with Laser Acupuncture, Photomedicine and Laser Surgery Volume 27, Number 2, 2009 [3]Jih-Huah Wu,.Yang-Chyuan Chang. (2013) Effect of Low-Level Laser Stimulation on EEG Power in Normal Subjects with Closed Eyes, Evidence-Based Complementary and Alternative MedicineVolume 2013 [4]Jih-Huah Wu, Wen-Dien Chang, Chang-Wei Hsieh, Joe-Air Jiang, Wei Fang, Yi-Chia Shan, and Yang-Chyuan Chang. (2012) Effect of Low-Level Laser Stimulation on EEG, Evidence-Based Complementary and Alternative MedicineVolume 2012 [5] Li-Chun Huang. (2007) Auricular diagnostics (in Chinese) [6]Wei-Ling Chang. (2009)Effect of Auricular Acupressure on Improving Nurses' Sleep Quality (in Chinese). Graduate Institute of Nursing, Hungkuang University, Taichung, Taiwan. [7] Chien-Tzu Liu. (2012)Effect of Auricular Point Acupressure on Puerpera with Insomnia (in Chinese). Graduate Institute of Nursing, Fu Jen Catholic University, Taipei, Taiwan. [8] Peng Zhou , Fangfang Sui, Anqiong Zhang, Fang Wang, Guohui Li. (2010). Music Therapy on Heart Rate Variability. Biomedical Engineering and Informatics (BMEI), 2010 3rd International Conference. [9]Chiu, H.W, Lin, L.S., Kuo, M.C.,Chiang, H.S. ,Hsu, C.Y.,.(2003). Using Heart Rate Variability Analysis to Assess the Effect of Music Therapy on Anxiety Reduction of Patients. Computers in Cardiology, 2003 . [10]Yaw-Chern Lee,Chun-Yang Lei, Yi-Sen Shih,Wen-Chih Zhang, Hui-Min Wang ,Cheng-Lung Tseng,Hou, M.C., Hui-Ya Chiang, Sheng-Chieh Huang ,.(2011). HRV 14

Response of Vegetative State Patient with Music Therapy .Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE

15

GLSBE-51 Study of Autonomic Nervous Activity and Sleep Latency of Elderly Women Treated with LED Radiation Wei-Hao Chena, Shau-Ming Tsengb, Yang-Chyuan Changc, Hui-Ling Chiud, Wei Fange, Yi-Chia Shanf and Jih-Huah Wug * a, b, g Department of Biomedical Engineering, Ming Chuan University, Taoyuan City, ROC. a E-mail address: [email protected] b E-mail address: [email protected] g* E-mail address: [email protected] c, d

e, f

Department of Neurology, Min-Sheng General Hospital, Taoyuan city, ROC. c E-mail address: [email protected] d E-mail address: [email protected]

Department of Bio-Industrial Mechatronics Engineering, National Taiwan University, Taipei, ROC. e E-mail address: [email protected] f E-mail address: [email protected]

Abstract The aim of this study was to evaluate the impact of LED radiation on the autonomic nervous system (ANS) and sleep latency (SL) of elderly women. According to our previous studies, utilized a non-invasive array LED (wavelength: 850nm, power output: 30 J/cm2, operated at 10Hz) to stimulate the left palm of the subjects‘ nervous plexus could increase the brainwave power of alpha rhythms when opened eyes. In our other study, stimulating the palm of the closed eyes subjects with array laser can make them fall asleep easily. Due to the phenomena mentioned above, we deem it can change their autonomic nervous system (ANS) activity, and shorten the period of sleep latency (SL). We applied LED energy to the left palm of the subjects to examine the impact of LED on the ANS and SL of 20 elderly women. Participants are randomly assigned into active group (LED on) and placebo group (LED off) for double-blind trial. Sleep latency analysis results showed no statistical significance, however the changes of SDNN comparing with baseline and another data have statistical meaning in the LED group. Keyword: LED Radiation, Multiple Sleep Latency Test, MSLT, Sleep Latency, Autonomic nervous detector. 1. Background/ Objectives and Goals Elderly women often have sleep disorders than others. [1][2] According to Taiwan Society of Sleep Medicine, 19.3 percent of people and 27 percent of people over sixty-five years old have insomnias. [3][4] Statistics of Ministry of Health and Welfare show over 3.2 million sleeping pills were taken per year in Taiwan.[5] Due to the problems mentioned above, we devoted to develop a physical therapy to help sleep. According to our previous studies, using near-infrared laser array to stimulate the plexus of left palm could increase the alpha rhythms of the subjects when opened eyes [6], decrease alpha rhythms when closed eyes respectively [7]. This phenomenon are the evidences that stimulating the palm nervous plexus with light can change the brain activity, and the effect of LED is similar to laser in our third study. Thus we speculate that LED stimulation might conducive to fall asleep [8], shortening the sleep latency (SL). The main goal of this study is to evaluate the effect of LED irradiation with ANS monitor and 16

MSLT (multiple sleep latency test). 2. Methods 2.1 Subject Twenty normal healthy postmenopausal females were included (mean age 60.9 ± 4.25 years) in the present study, with the Pittsburgh sleep quality index (PSQI) score below 14 points; then they were required to implement light diet and not to drink alcohol, caffeine, and not to eat spicy food at 8 hours before the test. Exclusion criteria included (a) having sleep disorder and using hypnotics; (b) having cardiopulmonary disease; (c) receiving medication that affect heart rate or sleep. The study was approved by the Institutional Review Board on Min-Sheng General Hospital (registration number: MSIRB-2014008). 2.2 Study design Each subject received two trials: one trial with the LED stimulator being turned on (the LED group) and the other trial with the stimulator being not turned on (the placebo group). In the first trial, each subject was randomly and blindly assigned to either LED or placebo group. Several days later in the second trial, the tested subject was then arranged to enter into the LED or the placebo group, just different to that in her first trial. The experiment were conducted in a quiet closed, air-conditioned (21-23°C) room and executed in Sleep Center of Min-Sheng General Hospital. According to the standard procedure of MSLT, each subjects underwent five naps, with a 90-min interval between them. Subjects in the LED group were required to try to fall asleep for 20 min, and the LED irradiation was applied continuously to the left palm for 10 min. Next, the subjects must keep awake for 90 min after wake-up. Before and after sleep, 5 min of the non-invasive HRV were recorded on wrist to measure the heart rate, HRV parameters. All of the procedures above are one session; a complete test in this study had five sessions operated from 9 A.M. to 5 P.M, as shown in Table. 1.

2.3 LED An array of six LEDs arranged in a triangle form to was used to radiate near infrared light at the palm of the subjects, as shown in Figure. 1. The central wavelength of LED is 850 nm, output power of each LED was set 10mW, the pulse repetition rate is 10 Hz, and the duty cycle is 50%. The area of the light spot is approximately equal to 28 mm2 on the skin. The dosage would be approximately 10.7 J/cm2 for 10 minutes stimulation.

17

2.4 EEG recording and Measurement An electroencephalogram instrument (Embla N7000, Natus Medical Incorporated) was used to record the cerebral cortex potential in this study. The placements of electrodes on cerebral cortex were followed the international 10-20 standard. Using central (C3 and C4) and occipital (O1 and O2) electrodes for the montage, with auricular reference electrodes (A1 and A2), as shown in Fig. 2, two electrooculography channels, two submental electromyography channels and electrocardiography, the MSLT was performed according to standard guidelines [9].

2.5 Assessment Sleep latency is defined as the elapsed time from lights out to the first epoch scored as sleep. And sleep stages were scored by the expert sleep technologists whose work in Sleep Center of Min-Sheng General Hospital. Using 30-s epochs according to standard criteria of the American Academy of Sleep Medicine [10]. Each subjects in both group underwent five naps, with a 90-min interval between them. The heart rate variability analyzer (CheckMyHeart; Taiwan Daily Care., Taoyuan, Taiwan) was used to measure ANS activity. All of the parameters were analyzed with HRV analysis software. For the linear time domain analysis, the R-R intervals of the ECG records were analyzed and the results were expressed in the form of normal to normal (NN) intervals. Then the standard deviation of the NN intervals (SDNN) was calculated. A fast Fourier transform (FFT)-based algorithm was used for the spectral analysis of HRV to calculate the frequency domains. 18

Spectral analysis allows discrimination between sympathetic and parasympathetic activation, which are represented by the low-frequency (LF) and the high-frequency (HF) components of the HRV signal. The LF band reflects a combination of sympathetic and parasympathetic influences and baroreflex function, while the HF band reflects parasympathetic activity. The LF to HF ratio serving as a quantitative index of the sympathetic balance was calculated. As established in 1996 by the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, the relevant indices and signals were defined and obtained as follows:  HF: the frequency band from 0.15–0.4 Hz, shown as a percentage, reflecting vagal activity.  HF (%) = HF power/(HF power + LF power)  LF: the frequency band from 0.04–0.15 Hz, shown as a percentage, reflecting the combination of SNS and PNS influences as it captures baroreflex rhythms.  LF (%) = LF power/(HF power + LF power)  LF:HF ratio: the ratio of LF to HF HRV, reflecting a quantitative index of sympathetic balance.  SDNN (ms): the standard deviation of the R-R intervals seen on the ECG 2.6 Statistical analysis Statistical analysis was performed with SPSS statistical analysis program. Paired-sample t-test was applied to compare the difference of the sleep latency and HRV before and after LED stimulation. The results will be recognized as statistical significant while p-value 0.05; Table 1). The results in this experiment reveal that sleep latencies are not obvious differences in each session after the LED stimulation. In our previous studies, output power of 7mW laser stimulation at 10Hz can increase the power of alpha rhythms and theta activities in some brain regions when subject opened eyes. And the other study showed, nearly all the tested subjects were found to fall into drowsiness or light sleep easily when closed eyes. In our previous study about the effect of LLL stimulation at the palm, the set power of each laser was 7mW, the dosage was 20 J/cm2. There was a wide irradiation angle for LED than laser. For the same power, LED could have a shorter penetration depth than laser. Even though we used the output power of LED (10mW) is higher than that of laser (7mW), however the dosage of LED (10.7 J/cm2) is not enough to induce the subjects to fall into sleep. Thus, we think the power and dosage of LED in our experiment is not sufficient to help subjects to fall asleep. However, the SDNN data before sleep in session 1 was seen as a baseline to compare another data in our experiment. There were significant increases of SDNN while comparing the 19

baseline and the data in session 2, session 3, and session 5 in the LED group (p0.05; Fig. 3). Sleep latency (SL) can be regarded as a time index to fall asleep. This study focused on the changes of SL and ANS of elderly women with LED stimulation. Polysomnography and HRV analyzer were used to record the physiological signal of the subjects in this trial. Even the experiment results showed no significant difference in SL tests in both groups, however, the ANS analysis showed SDNN were mainly getting larger in the LED group. We infer that this LED stimulation could enhance the ANS activity of the elderly women. Most of the elderly women recruited are not used to stay up late in Taiwan. Therefore, the effect of sleep latency from LED stimulation is not apparent.

20

Acknowledgments The authors are especially indebted to Yue Chen and Xiang-Chao Chang (Student of Ming Chun University (MCU)) and Sleep Center of Min-Sheng General Hospital for their support to this study. The authors are also acknowledging the Minisitry of Science and Technology of the Republic of China for financial support (contract nos.: MOST-104-2221-E-130-005). 4. Reference [1] Chi-Wen Lei, Shiuh-Long Lin(2003), ‖The relationship between physical activity and sleep quality of middle-aged women.‖ Journal of Health Education, 20, 151-169。 [2] Soares, C.N. (2005). ―Insomnia in women: An overlooked epidemic?‖ Archives of Women's Mental Health, 8, 205-213. [3] Schubert, C.R., Cruickshanks, K.J., Dalton, D.S., Klein, R., & Nondahl, D.M. (2002). ―Prevalence of sleep problems and quality of life in an older population.‖ Sleep, 25, 8889-893. [4] Jia Shih Wu (2014, April). Re: 2014 Taiwan Society of Sleep Medicine the latest survey, the silver-haired insomnia per four person ! [Online forum comment]. Retrieved from http://goo.gl/qjjsvo 21

[6] Jih-Huah Wu, Wen-Dien Chang, Chang-Wei Hsieh, Joe-Air Jiang, Wei Fang, Yi-Chia Shan, Yang-Chyuan Chang* (2012) Effect of Low Level Laser Stimulation on EEG. Evidence-Based Complementary and Alternative Medicine, Article ID 951272, 11 pagesdoi:10.1155/2012/951272. [7] Jih-Huah Wu, and Yang-Chyuan Chang*, ―Effect of Low-Level Laser Stimulation on EEG Power in Normal Subjects with Closed Eyes,‖ Evidence-Based Complementary and Alternative Medicine, Vol. 2013, Article. [8] Shih-bin Yeh (2012), ―Sleep disorders.‖ Ho-Chi Publishing Co., Ltd. [9] M.A. Carskadon, W.C. Dement, M.M. Mitler, et al. Guidelines for the multiple sleep latency test (MSLT): a standard measure of sleepiness. Sleep, 9 (4) (1986), pp. 519–524. [10] R.B. Berry, R. Budhiraja, D.J. Gottlieb, et al. Rules for scoring respiratory events in sleep: update of the 2007 AASM Manual for the Scoring of Sleep and Associated Events. J Clin Sleep Med, 8 (5) (2012), pp. 597–619.

22

GLSBE-84 The Effect of Mechanical Properties of Polyhydroxyalkanoates (PHAs) Three-Dimensional (3D) Scaffolds on Proliferation of Human Gingival Fibroblasts (HGFs) Seubsakul Phuegyoda, Nungnit Wattanavicheanb, Rudee Suraritc, and Nuttawee Niamsirid a Department of Biotechnology, Faculty of Science, Mahidol University, Bangkok, Thailand. E-mail address: [email protected] b

Department of Biotechnology, Faculty of Science, Mahidol University, Bangkok, Thailand. E-mail address: [email protected]

c

Department of Oral biology, Faculty of Dentistry, Mahidol University, Bangkok, Thailand. E-mail address: [email protected]

d

Department of Biotechnology, Faculty of Science, Mahidol University, Bangkok, Thailand. E-mail address: [email protected]

1. Background/ Objectives and Goals Periodontitis is an inflammatory disease that can destroy tooth supporting tissues, including human gingival fibroblasts (HGFs). One of interesting treatments is tissue engineering (TE) that can regenerate the damage HGFs. TE requires 3D scaffolds for in vitro culturing and implantation. The mechanical properties of scaffolds also play a crucial role of appropriate cellular responses including cell growth or differentiation. Therefore, the development of effective 3D scaffold for HGFs is still needed. In this study, four different types of bacterial polyhydroxyalkanoates (PHAs) were fabricated into 3D porous scaffolds and investigated their mechanical properties on the proliferation of HGFs. 2. Methods Four different types of PHA polymers; Polyhydroxybutyrate or PHB, Polyhydroxybutyrate-co-5%hydroxyvalerate or P(HB-co-5%HV), Polyhydroxybutyrate-co-12%hydroxyvalerate or P(HB-co-12%HV) and Polyhydroxybutyrate-co-50%hydroxyvalerate or P(HB-co-50%HV) were studied. The first three types of PHAs were commercially obtained, while the P(HB-co-50%HV) was produced ―in house‖ from culturing of Cupriavidus neator H16 with mineral salt media (MS), purified and determined %HB and %HV monomers content using 1H-NMR [1]. 3D porous scaffolds were fabricated from all types of PHAs via a standard salt leaching technique [2] with pore size range between 425-500 μm. 3D PHA scaffolds were characterized for % porosity, topography by SEM, compressive stress and secant modulus at two different conditions; dry and wet (i.e. submerged in cell culture media) by a texture analyzer. Lastly, the proliferation of HGFs cultured on these 3D PHA scaffolds were evaluated for 8 days in 48-well tissue culture plate using a standard MTT assay with initial cell loading at 5x104 cells/scaffold [3], compared to the controls, which are 2D surface and 3D polycaprolactone (PCL) scaffold. After staining actin filaments and nucleus, the morphology and migration of HGFs inside the scaffolds were also observed by confocal fluorescent microscopy analysis. RT-PCR was also performed to evaluate gene expression level of collagen type I of HGFs grown on the scaffolds after 8 days [4, 5]. 3. Expected results/ Conclusion/ Contribution 23

The polymer was purified from the lyophilized bacterial cells and characterized by 1H-NMR, showing that it was P(HB-co-50%HV). 3D scaffolds were successfully fabricated from four types of PHAs with a pore size range around 425-500 μm shown by SEM. The porosity of all four PHA scaffolds was found to be at around 90%. According to mechanical testing, the secant modulus values were found to be different among PHA scaffolds. The secant modulus was found to depend strongly on %HV composition of PHA used in making 3D scaffolds, in which the secant modulus was decreased as % HV content in PHA increased. Among these four PHAs, PHB scaffold yields the highest secant modulus at both dry and wet conditions, showing that the value of 0.75 and 0.33 MPa, respectively (i.e. the most stiff scaffold), while P(HB-co-50%HV) scaffold gives the lowest secant modulus under both conditions (i.e. 0.04 MPa), resulting into a softer scaffold. The biocompatibility of 3D PHA scaffolds were then evaluated with HGFs proliferation over 8 days. Compared to the control 2D surface of polystyrene, HGFs could attach and proliferate better in all 3D scaffolds at day 8. Among four types of PHAs, P(HB-co-50%HV) was found to be the best scaffold to support the most growth of HGFs. At the final day of experiment (i.e. day 8), the amount of HGFs grown on the P(HB-co-50%HV) scaffold was 2 times higher than 2D surface and around 1.25 times higher than the control 3D PCL scaffold. Since FBS proteins adsorption on all PHA scaffolds was found to be the same, we concluded that HGFs might prefer the mechanical properties of P(HB-co-50%HV) scaffold for attachment and proliferation. Moreover, we also observed the morphology and migration of HGFs grown inside the P(HB-co-50%HV) scaffold using confocal fluorescent analysis. The results showed HGFs have a good spreading on the surface of P(HB-co-50%HV) scaffold and their morphology were suggesting healthy fibroblast with spindle or stellate shape [6]. In addition, we also confirmed that the HGFs cultured on the P(HB-co-50%HV) scaffold showed similar expression level of collagen type I gene after 8 days when compared with traditional 2D surface, suggesting that the HGFs have a normal character of collagen synthesis and no stress of cells when grown in the scaffold. In summary, the P(HB-co-50%HV) could effectively support HGFs proliferation and should be considered as a potential biomaterial for periodontal tissue engineering and regeneration. Keywords: Polyhydroxyalkanoates, 3D porous scaffold, Mechanical properties, Human gingival fibroblasts.

[1]

[2]

[3]

[4]

[5]

References Kamiya, N., Inoue, Y., Yamamoto, Y., Chujo, R., & Doi, Y. (1990). Conformational analysis of poly (3-hydroxybutyrate-co-3-hydroxyvalerate) in solution by proton NMR spectroscopy. Macromolecules, 23(5), 1313-1317. Holy, C.E., Cheng, C., Davies, J.E., Shoichet, M.S. (2001). Optimizing the sterilization of PLGA scaffolds for use in tissue engineering. Biomaterial, 22(1), 2531. Ferè, S., Valtulina, V., Petrini, P., Alessandrini, E., Pietrocola, G. et al. (2005). In vitro interaction of human fibroblasts and platelets with a shape-memory polyurethane. Biomedical Material Research Part A, 73A(1), 1-11. Mah, W., Jiang, G., Olver, D., Cheung, G., Kim, B. et al. (2014). Human gingival fibroblasts display a non-fibrotic phenotype distinct from skin fibroblasts in three-dimensional cultures. Plos one, 9(3), e90715. Wang, L., Wang, Z. H., Shen, C. Y., You, M. L., Xiao, J. F., & Chen, G. Q. (2010). Differentiation of human bone marrow mesenchymal stem cells grown in terpolyesters of 3-hydroxyalkanoates scaffolds into nerve cells. Biomaterials, 31(7), 1691-1698. 24

[6] Sachar, A., Strom, T. A., San Miguel, S., Serrano, M. J., Svoboda, K. K., & Liu, X. (2014). Cell–matrix and cell–cell interactions of human gingival fibroblasts on three‐ dimensional nanofibrous gelatin scaffolds. Journal of tissue engineering and regenerative medicine, 8(11), 862-873.

25

Electrical Engineering and Technology (1)

Tuesday, March 29, 2016

15:45-17:45

Room C

Session Chair: Prof. Hei Wong ACEAIT-7299 Characterization of Interface Structure of Multiple Stacked Dielectric Films Used in Nano CMOS Technology Hei Wong︱City Universtiy of Hong Kong Jieqiong Zhang︱City Universtiy of Hong Kong Xuan Feng︱Zhejiang University Danqun Yu︱Zhejiang University ACEAIT-7429 Design of Municipal Solid Waste Management System for a Medium Size Community in a Hot-Humid Climate Region of Thailand Thanut Boonruangsri︱King Mongkuts Institute of Technology Ladkrabang Warawat Tangsrianukula︱King Mongkuts Institute of Technology Ladkrabang Chivalai Chivalai Temiyasathit︱King Mongkuts Institute of Technology Ladkrabang Chai Chompoo-inwai︱King Mongkuts Institute of Technology Ladkrabang ACEAIT-7558 Self-Tuning Fuzzy Sliding Mode Controller for Position Control of Robot Manipulators Tsing-Iuan James Tsay︱National Cheng Kung University Yu-Jhang Zeng︱National Cheng Kung University Kai-Hung Zhan︱National Cheng Kung University

26

ACEAIT-7299 Characterization of Interface Structure of Multiple Stacked Dielectric Films Used in Nano CMOS Technology a

b

Jieqiong Zhanga, Hei Wonga,b, Xuan Fengb, and Danqun Yub Department of Electronic Engineering, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong SAR, China E-mail address: [email protected]

School of Information Science and Electronic Engineering, Zhejiang University, Zheda Road, Hangzhou, China

1. Background To maintain proper operation and to keep low standby power of advanced CMOS microchips, high dielectric constant (high-k) dielectric/metal gate stack has been adopted for the state-of-the-art nano CMOS technology. Although the high-k layer is much thicker than the conventional silicon oxide and the thickness control is much easier now, the interface issues, including the metal gate/high-k and high-k/silicon interfaces are still the great challenge for this new material integration. The critical issue is that high-k materials are less thermal stable when they are interface with the silicon substrate and the metal gate. The second issue is that in farther technology nodes half nanometer thick (in the sense of equivalent oxide thickness, or EOT) high-k film will be indispensable and the low-k interface layer may limit the minimum achievable EOT to well above half nanometer. This work, taking tungsten/lanthanum oxide/silicon (W/La2O3/Si) stack as an example, presents a detail study on the interface structures and their associated chemical reactions by using angle-resolved x-ray photoelectron spectroscopy to perform composition profiling together with peak de-convolution technique to reveal the interface bonding structures. 2. Experiment and Results In this work, the tungsten/La2O3 gate stack was deposited on n-type silicon. The La2O3 film of 5 nm thick was deposited by electron beam evaporation in an ultra-high vacuum chamber. The tungsten gate electrode of about 3 nm thick was then deposited in situ using magnetron sputtering. The samples were then rapid thermal annealed (RTA) in N2 ambient at 500°C or 600°C for 30 min. The chemical compositions as well as the bonding structures of the samples at different depths were investigated in detail by using angle-resolved x-ray photoelectron spectroscopy (ARXPS) measurements at various take-off angles ranging from 0 to 90o. The apparatus used was a Physical Electronics PHI 5802 spectrometer with Al Kα (1486.6 eV) monochromatic source. To improve the resolution of ARXPS measurements, proper film thinning was conducted by using Ar+ sputtering at a rate of about 0.67 nm/min.

27

Figure 1 shows some typical ARXPS results measured on the as-deposited samples. As can be seen from Fig. 1(a), the bulk of W layer is elementary W with W 4f twin peak with energy of 31.3 and 33.4 eV. At larger take-off angles which disclose deeper regions of the film, it shows there presents of significant oxidized tungsten with W 4f peak energies in the range of 35 to 39 eV. With Gaussian decomposition, the majority of oxidation state is found to be W6+. The W/La2O3 interface oxidation was enhanced after thermal annealing (not shown). With Ar ion sputtering, the W layer and portion of the La2O3 layer were removed. ARXPS measurements at various angles were done again. Fig. 1(b) shows the La 3d XPS spectra at different depths/angles. The highly skewed peaks, with energy shifts, of both La 3d 3/2 (peak in the range of 852-856 eV) and La 3d5/2 imply the silicate bonding. Gaussian peak decompositions were performed for La, Si 2s, and O 1s spectra. Results consistently lead to the conclusion of the existence of Si-O and La-O-Si bonding at the interface. This lower dielectric constant interface layer is responsible to the channel mobility improvement for thermally-annealed sample and the reduction of effective gate capacitance. 3. Conclusions In short, angle-resolved XPS together with spectral peak decomposition techniques are able to disclose the bonding details at both tungsten/La2O3 and La2O3/Si interfaces. Results show that both interfaces are unstable against high-temperature annealing. The tungsten metal gate electrode could be oxidized and that may introduce some instability issues to device operation. Thick silicate layer (with lower k value) was formed at the La2O3/Si interface. This layer gives rise to a significant increase in EOT. Keywords: interface, high-dielectric constant, gate dielectric, CMOS

28

ACEAIT-7429 Design of Municipal Solid Waste Management System for a Medium Size Community in a Hot-Humid Climate Region of Thailand Thanut Boonruangsria, Warawat Tangsrianukula, Chivalai Temiyasathitb, Chai Chompoo-inwaia a Department of Electrical Engineering, Faculty of Engineering, King Mongkut's Institute of Technology Ladkrabang, Thailand E-mail address: [email protected], [email protected], [email protected] b Department of Solfware Engineering, International College King Mongkut's Institute of Technology Ladkrabang, Thailand E-mail address: [email protected] Abstract According to the current Power Development Plan (PDP 2010 rev.3) of Thailand, the government tends to continuing promote the expansion of the electricity purchase from various renewable energy generations following potential of fuel source in each region of the country. Phon-city and its 5 neighboring cities, located in Northeastern region of Thailand, with a hot-humid climate, have up to 200,000 populations which generating around 200 tons of waste per day. Preliminary survey result reveals that majority of waste is municipal solid waste (MSW). The waste are composed of organic material with approximately heating value 2,000-2,200 kcal/kg and approximately 50% of moisture content. Preliminary design result shows that the quality and contents of the existing waste in the area can produce at least 2 MW of electricity and sell to Provincial Electricity Authority (PEA) to earn revenue from generating electricity. However, with the characteristic of MSW, it is considered as a low quality fuel supply for the boiler in standard combustion value (about 3000kcal/kg). Therefore, this paper aims to solve the problem by using biomass as a secondary fuel source for supporting municipal solid waste and selecting incinerator technology as a proper electricity generation. Keywords: Municipal Solid Waste , Renewable Energy , Combustion, Incinerator

29

ACEAIT-7558 Self-Tuning Fuzzy Sliding Mode Controller for Position Control of Robot Manipulators T. I. James Tsay, Y. J. Zeng and K. H. Zang Dept. of Mechanical Engineering, National Cheng Kung University, Tainan, Taiwan E-mail address: [email protected] 1. Background/ Objectives and Goals A robot manipulator is a highly nonlinear and dynamically coupled system, which is subject to the disturbances and model uncertainties. The general control algorithms, such as computed toque technique, PD control algorithm, etc., will not render the expected performance in the presence of the disturbances and model uncertainties. Moreover, suitable gains used in these algorithms are usually obtained using time-consuming trial-and-error methods. A novel nonlinear self-tuning control scheme for the position control of robot manipulators is proposed, which is robust against the disturbances and model uncertainties. 2. Methods In the proposed control scheme, a feedforward compensation is applied to undercompensate the friction torque and a fuzzy sliding mode (FSM) control algorithm is employed in the feedback loop. The FSM control algorithm is based on the principle of sliding mode control but with the application of fuzzy logic control. A fuzzifized sliding surface is utilized to overcome the chattering problem inherent with sliding mode control. To increase the robustness of the FSM controller, a neural-network-based self-tuning approach is then proposed to automatically determine scaling factors of the fuzzy logic control. 3. Expected Results/ Conclusion/ Contribution Both computer simulations and experiments were executed to verify the proposed control scheme on the 7-DOF robot manipulator of a humanoid robot developed in our laboratory. We are able to show that the proposed control scheme can be easily applied to robot manipulators. The tracking problem of manipulators can be independently solved for each joint. No exact mathematical model for the manipulators is required for the controller design. The control rules determined by the satisfaction of the sliding condition can ensure the stability of the system. The chattering problem inherent with sliding mode control can be handled effectively by the proposed fuzzifized sliding surface. The simulation and experimental results can show that the manipulator is able to follow the desired trajectory accurately in the presence of disturbances and model uncertainties. Keywords: Robot manipulator, Self-tuning, Sliding mode control, Fuzzy logic control, Tracking.

30

Biological Engineering (2)

Wednesday, March 30, 2016

08:30-10:00

Room A

Session Chair: Prof. Chi-Chun Chen GLSBE-45 An Adaptive Deceleration Training Model for Ischemic Stroke Rehabilitation in Rats Chi-Chun Chen︱National Chin-Yi University of Technology Shi-Cheng Su︱National Chin-Yi University of Technology GLSBE-60 Hi-C Chromatin Interactions Network Construction and Its Topological Hubs Identification Jingjing Zhao︱East China Univerisity of Science and Technology Hongbo Shi︱East China Univerisity of Science and Technology GLSBE-63 The Embedded System Integrated ECG Monitoring and Low Energy LED Light Stimulation Yi-Chia Shan︱National Taiwan University Jih-Huah Wu︱Ming Chuan University Shau-Ming Tseng︱Ming Chuan University Wei Fang︱National Taiwan University Wei-Hao Chen︱Ming Chuan University

31

GLSBE-45 An Adaptive Deceleration Training Model for Ischemic Stroke Rehabilitation in Rats a

Chi-Chun Chena*, Wen-Ying Changb, Shi-Cheng Sua Department of Electronic Engineering, National Chin-Yi University of Technology, Taiwan b Information and Communications Research Laboratories, Industrial Technology Research Institute, Taiwan [email protected]

1. Background/ Objectives and Goals An adaptive deceleration training model is proposed for rehabilitation in an animal ischemic stroke. Current training methods use a fixed speed for rehabilitation training in rats. However, it is difficult for the rats to keep exercising at a constant intensity. Therefore, the goal of our method is to provide a relatively steady training method for ischemic stroke rehabilitation in rats. 2. Methods This proposed technique uses infrared sensors to measure the running positions of rats to provide feedback and adjust the exercise speed for assisting stroke rehabilitation in rats. In addition, with adaptive deceleration models, rats with stroke can achieve rehabilitation based on their individual physical characteristics. This rehabilitation model not only achieves effective recovery based on the individualized physical capacity of rats but also protects rats from fall injuries. This rehabilitation model overcomes individual differences, which has not been realized in conventional rehabilitation models. 3. Expected Results/ Conclusion/ Contribution An animal ischemic stroke model will be used to test the effectiveness of this method. An inclined plane will be used daily to test the gripping strength recovery of their hind limbs. After three weeks of exercise, the triphenyl tetrazolium chloride (TTC) test will be used to obtain the cerebral infarct volume. The result will be expected that the proposed method achieved a significant recovery from the evaluation of motor function and cerebral infarct volume. The conventional fixed-speed method may result in secondary injuries during rehabilitation, which causes adverse effects in recovery. This phenomenon is also the reason why currently training animal methods have shown little effectiveness in the rehabilitation of stroke patients. This study constructs a fully automated adaptive rehabilitation model based on physical characteristics to enhance the stability of the training process for the rats. Table 1 presents a comparison between animal-related research and training methods. The study will confirm that rehabilitation training models require proper control to achieve certain treatment effects. In other words, when forcing rats to exercise, it is important to reduce their psychological burden and prevent them from suffering secondary injuries. Therefore, this model provides an objective and effective experimental platform for clinical researchers conducting sports physiology studies. Keywords: Adaptive deceleration training model, rehabilitation, ischemic stroke

32

Acknowledgment The authors gratefully acknowledges the support provided for this study by the Ministry of Science and Technology (MOST 104-2218-E-167-001-) of Taiwan.

33

GLSBE-60 Hi-C Chromatin Interactions Network Construction and Its Topological Hubs Identification Jingjing Zhaoa.b, Hongbo Shia,b* Key Laboratory of Advanced Control and Optimization for Chemical Processes of the Ministry of Education, East China University of Science and Technology, Shanghai, China b Department of Information Science and Engineering, East China University of Science and Technology, Shanghai, China *Email:[email protected] a

Abstract Background: The three-dimensional structure of the genome is an important regulator of many biological processes and cellular processes. Recently, technologies such as Hi-C provide a frequency contact measurement of chromatin properties via genome-wide 3D interaction map, which shows the ability to reveal the DNA-DNA interaction in both intra- and inter-chromosomal level. Only a few studies have been done on that analysis analyze of Hi-C chromatin interaction network (CIN) and important hub identification, they are mainly based on frequency contract map, suffering from the problem of low resolution and bias. Finding the topological hub region CIN and revealing the biological meaning has become the interest of many researchers. Method: To reduce the restriction enzyme bias and increase the resolution, we use pre-identified anchors as the node of our network. The weight of the edge of our network is given as a relative entropy score, it adjusts the counting of chromatin interaction frequency compared with the background distribution. After filtering out the weak linkage, we generate CINs on two human cell lines (hESC and IMR90). We next identify the topological hubs by calculating the weighted PageRank score, this method not only selects the anchors with high degree, it also keeps the important low-degree genes that link to other important anchors in the chromatin interaction network. Results: We found that 1266 hubs were in IMR90 and 1103 were in hESC. The hubs we found in both human cell lines are significantly close to active elements (H3K27ac, H3K4me1, H3K4me3) and far away from a repressed mark (H3K27me3), which indicates the interaction hubs mainly play a role as active enhancers in the human genome. And also we have proved that hubs are conserved within vertebrates and the shared hubs are highly overlapped with housekeeping genes. By applying our method to different cell lines, a comprehensive list of key interactions and hub active regions is likely to significantly improve the resources available to future genetic studies focused on human disease. Keywords: Hi-C, chromatin interaction network (CIN), relative entropy score, weighted PageRank score, topological interaction hubs 1. Introduction The 3D architecture of the genome plays a key role in influencing important biological processes and cellular processes including differentiation and gene regulation [1]. Genes are regulated not only by the nearby elements, but also by the elements located far away in terms of the linear genome sequence. Moreover, a lot of inter-chromosomal interactions are detected as the linkages of a specific group of functionally active loci, which are a result of indiscriminate encounters between loci that are accessible to each other[2]. 34

Chromosome conformation capture techniques, such as 3C, 4C, 5C, and Hi-C, succeed in measuring the proximity ligation of DNA to elucidate genome structure at a high resolution; thus, these provide methods to study the chromatin interactions and 3D architecture of the whole genome [3, 4]. In particular, the Hi-C technique provides a frequency contact measurement of chromatin properties via genome-wide 3D interaction map, which shows the ability to reveal the DNA-DNA interaction in both intra- and inter-chromosomal level[5]. Original studies based on Hi-C data mainly focused on partitioning the chromosomal region into a succession of large-scale topologically associating domains(TADs)[6-13] on a 1D scale, and classifying of the domains into different function groups based on their expression and regulation behavior[14]. More importantly, the comprehensive characterization of long-range chromatin interactions can be obtained by considering the chromatin interaction data as a network[15], which is called a Chromatin Interaction Network (CIN). In this network, each node represents a genomic loci and the interaction‘s normalized frequency denotes the weight of the edge between two nodes. Only a few studies have been done on analysis of Hi-C chromatin interaction network, for example, by using scale-aware of the multi-resolution CIN to predict co-expression patterns of genes[16], and construction of spatial gene interaction and regulation networks[17]. However, what interests us the most is the specific genomic loci in the CIN that are the key organizational features for genome functions of maximal network centrality and act as interconnected hubs both within chromosomes at short and long distances and in between different chromosomes. Our work generates the CIN with Hi-C chromatin interaction data at a higher resolution. We demonstrate that the weighted PageRank hubs of the network exhibit distinct functional and regulation properties. Together, our work reveals the 3D structure of the human genome and finds chromatin-level hubs that may play an active role in DNA-DNA interaction and gene regulation. 2. Materials and methods 2.1 Hi-C data from two human cell types To comprehensively identify the DNA interaction hubs, we first downloaded 2 paired-end Hi-C sequencing data from Bing Ren's lab[18] from 2 human cell lines, hESC and IMR90 (GEO accession number GSE35156). Reads were mapped to reference human (hg18) and non-mapping reads and PCR duplicates were removed. 2.2 Identifying statistically significant chromatin interaction For the human hg18, we use 518,032 anchors identified by Fulai Jin[19], with each containing one or a few restriction fragments (fragments shorter than 2 kilobases (kb) are merged.) Basically, anchors are the short regions in the genomes that can be considered as the ―unit‖ of chromatin interaction based on Hi-C technique. Next, we identified the significant interaction anchors using a relative entropy model. To do this, we gave each anchor a relative entropy; their relative entropy is identified as:

It is worth noting that we took out the reads mapped to the telemere and centermere regions 35

from our analysis. Therefore, the higher the entropy, the more significant the Hi-C interaction is. Only the significant anchors (p 99% quantile as the hub anchors and merged the hub anchors together if they are close to each other. We found 1266 hub anchors in IMR90 and 1103 in hESC. Figure 1 is the analysis work-flow that extracts high-quality interacting hub anchors from Hi-C datasets. In order to characterize the hub anchors, we examined the distance between the hub anchors and the nearest histone marks on the both sides which include active elements (H3K27ac, H3K4me1, H3K4me3) and a repressed mark (H3K27me3). The ChIP-Seq data were downloaded from UCSC genome browser/BroadHistone and Bing Ren's lab (GSE35156). As shown in Figure 2 and Table 1, we found that hub anchors have a shorter distance with the activating histone modifications but are farther away from repressed mark (we shuffled all the hub anchors 100 times as the control background). This result shows that many of the hub anchors are close to the human enhancer elements, which indicates the interaction hubs mainly play a role as active enhancers in the human genome.

*T statistics are computed comparing log-transformed distance between hubs anchors to random shuffle 100 times anchors as control. (x is the distance between hub anchors and ChIP-Seq peaks)

37

It is expected that the important 'hubs' in the interaction network structure are conserved across evolution because of the functional roles they play. To study the conservation of the hub anchors, we extended these anchors to 25kb on both sides; we chose 25kb because Hi-C reads with two ends within a short distance (< 25kb) are usually considered as uncut DNA artifacts [19]. We next investigated the enrichment of the conservation elements across the vertebrate (28-way miltiz phastcons vertebrates conserved elements [24, 25]) in extended hub anchors compared with shuffling regions. We found that these extended hub anchors are significantly enriched in conserved elements compared with the random shuffling regions (Figure 3).

There are only 71 merged hubs shared between two cell lines(Figure 4), which may indicate that hubs for each cell line tend to have tissue-specific behavior. We overlap these shared hubs with human housekeeping genes[26], and after comparing with all the anchors in the network, we find that shared hubs are significant enriched with housekeeping genes(Fisher test, p value =1.16 E-05). That matches our assumption that hubs are active loci in the genome. 38

4. Conclusions We use human Hi-C interaction data to construct a chromatin interaction network of the two human cell lines hESC and IMR90. Unlike other CIN contraction methods, we first consider the short chromatin anchor as each node of the network. It has several advantages: (1) It has a higher resolution and takes the bias from restriction enzyme into consideration. Most Hi-C contact maps have a mega-base resolution and the starting cut point of the chromosome is arbitrary. Our method not only filters out the inactive region such as telemere and centermere, but also generates the network node at a shorter length (average 5352bp). (2) Rather than use a frequency contract map, we give each interaction a relative entropy score, which is the frequency score adjusted by the background distribution. Therefore the links we picked are more statistically and biologically significant. By filtering out the weak linkage, we generate the CIN with the anchors as the node. Compared to most complex networks, such as metabolic and gene regulation networks in biology, edges in the chromatin interaction network link vertices that belong to a 1D physical object; our networks tend to reveal the 3D structure of the genome. Hubs, as the important nodes with the most connections in network topology, are found to play an important role in the biology active process. Differing from the classical degree hubs (usually the network nodes with the highest degree), we used a weighted PageRank score. Thus, this method not only selects the anchors with high degree, it also keeps the important low-degree genes which link to other important anchors in the chromatin interaction network. The hubs we found in both human cell lines are significantly close to active elements (H3K27ac, H3K4me1, H3K4me3) and far away from a repressed mark (H3K27me3). Also, we have proven that hubs are conserved within vertebrates and highly overlapped with housekeeping genes. Taken together, hubs identified from the Hi-C chromatin interaction network provide genome-wide evidence for preferential aggregation of active chromosomal regions. Applying our analysis workflow to identify DNA-DNA interaction information from Hi-C; allows identification of important hub regions in the genome, which are functionally active and conserved. In the future, comparing datasets similar to the ones provided here with findings from GWAS and CHIA-PET studies is likely to provide mechanistic insights into modeling of chromatin 3D structure and important active loci identification. By extending it to different cell lines, a comprehensive list of key interactions and hub active regions is likely to significantly improve the resources available to future genetic studies focused on human disease. 5. References 39

1. 2. 3.

4. 5. 6. 7.

8.

9. 10. 11. 12. 13. 14. 15. 16. 17.

18. 19. 20.

21. 22. 23.

Gerstein, M.B., et al., Architecture of the human regulatory network derived from ENCODE data. Nature, 2012. 489(7414): p. 91-100. Kalhor, R., et al., Genome architectures revealed by tethered chromosome conformation capture and population-based modeling. Nat Biotechnol, 2012. 30(1): p. 90-8. Dekker, J., M.A. Marti-Renom, and L.A. Mirny, Exploring the three-dimensional organization of genomes: interpreting chromatin interaction data. Nat Rev Genet, 2013. 14(6): p. 390-403. Lieberman-Aiden, E., et al., Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science, 2009. 326(5950): p. 289-93. Hu, M., et al., Understanding spatial organizations of chromosomes via statistical analysis of Hi-C data. Quant Biol, 2013. 1(2): p. 156-174. Nora, E.P., et al., Spatial partitioning of the regulatory landscape of the X-inactivation centre. Nature, 2012. 485(7398): p. 381-5. Nguyen, V.T., et al., Differential epigenetic reprogramming in response to specific endocrine therapies promotes cholesterol biosynthesis and cellular invasion. Nat Commun, 2015. 6: p. 10044. Dileep, V., et al., Large-Scale Chromatin Structure-Function Relationships during the Cell Cycle and Development: Insights from Replication Timing. Cold Spring Harb Symp Quant Biol, 2015. Remeseiro, S., A. Hornblad, and F. Spitz, Gene regulation during development in the light of topologically associating domains. Wiley Interdiscip Rev Dev Biol, 2015. Ulianov, S.V., et al., Active chromatin and transcription play a key role in chromosome partitioning into topologically associating domains. Genome Res, 2015. Cubenas-Potts, C. and V.G. Corces, Topologically Associating Domains: An invariant framework or a dynamic scaffold? Nucleus, 2015: p. 0. Dekker, J. and E. Heard, Structural and functional diversity of Topologically Associating Domains. FEBS Lett, 2015. 589(20 Pt A): p. 2877-84. Weinreb, C. and B.J. Raphael, Identification of hierarchical chromatin domains. Bioinformatics, 2015. Zhao, J., H. Shi, and N. Ahituv, Classification of topological domains based on gene expression and regulation. Genome, 2013. 56(7): p. 415-23. Sandhu, K.S., et al., Large-scale functional organization of long-range chromatin interaction networks. Cell Rep, 2012. 2(5): p. 1207-19. Babaei, S., et al., Hi-C Chromatin Interaction Networks Predict Co-expression in the Mouse Cortex. PLoS Comput Biol, 2015. 11(5): p. e1004221. Wang, Z., et al., The properties of genome conformation and spatial gene interaction and regulation networks of normal and malignant human cell types. PLoS One, 2013. 8(3): p. e58793. Dixon, J.R., et al., Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature, 2012. 485(7398): p. 376-80. Jin, F., et al., A high-resolution map of the three-dimensional chromatin interactome in human cells. Nature, 2013. 503(7475): p. 290-4. Banky, D., G. Ivan, and V. Grolmusz, Equal opportunity for low-degree network nodes: a PageRank-based method for protein target identification in metabolic graphs. PLoS One, 2013. 8(1): p. e54204. Dellavalle, R.P., et al., Refining dermatology journal impact factors using PageRank. J Am Acad Dermatol, 2007. 57(1): p. 116-9. Griffiths, T.L., M. Steyvers, and A. Firl, Google and the mind: predicting fluency with PageRank. Psychol Sci, 2007. 18(12): p. 1069-76. Heinz, S., et al., Simple combinations of lineage-determining transcription factors prime 40

cis-regulatory elements required for macrophage and B cell identities. Mol Cell, 2010. 38(4): p. 576-89. 24. Blanchette, M., et al., Aligning multiple genomic sequences with the threaded blockset aligner. Genome Res, 2004. 14(4): p. 708-15. 25. Siepel, A., et al., Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res, 2005. 15(8): p. 1034-50. 26. Lercher, M.J., A.O. Urrutia, and L.D. Hurst, Clustering of housekeeping genes provides a unified model of gene order in the human genome. Nat Genet, 2002. 31(2): p. 180-3. Acknowledgments: The work is supported by the China National Funding: 61374140. And the author thanks Professor Nadav Ahituv from University of California, San Francisco for his advice on the whole project design.

41

GLSBE-63 The Embedded System Integrated ECG Monitoring and Low Energy LED Light Stimulation a

Yi-Chia Shana, Jih-Huah Wub, Wei-Hao Chenb, Shau-Ming Tsengb, Wei Fanga* Department of Bio-Industrial Mechatronics Engineering, National Taiwan University No. 1, Sec. 4, Roosevelt Rd., Taipei106, Taiwan. b Department of Biomedical Engineering, Ming Chuan University No. 5, Deming Rd., Gweishan Township, Taoyuan333, Taiwan. [email protected]

Background Nowadays people usually have higher work pressure and faster life rhythm. In the long-term stress environment, easily lead to autonomic nervous system (ANS) disorders and lead to physiological or psychological problems. To resolve the problem, we propose an embedded system which integrates ANS monitoring and low energy light stimulation. When the ANS was detected abnormal, a recommended low energy infrared light was activated to stimulate the subject to balance ANS. Materials and Methods The experiment instruments included electrocardiogram (ECG) module, LED array and the embedded system. ECG module received ECG signal, and then transferred analog signal to digital data. The microprocessor calculated time-domain and frequency-domain data of heart rate variability (HRV), and analyzed the time-domain data, and power spectral density data. The autonomic nervous system (ANS) of the user can be monitored by this device we developed. When the calculated values are out of the criteria, the system will alert the user to turn on infrared light to stimulate and balance ANS. LED wavelength is 850 nm, the default output power is 10 mW, and the default repetition rate is 10 Hz. Result Wu et al. proposed that using low level laser to stimulate on Neiguan point (PC6) was used to adjust the ANS of the night shift workers. And in other studies, Laser array and LED array were used to stimulate the palm of the subjects with open eyes in different clinical trial, the alpha powers were both increased significantly. So the combination of ANS monitoring and light therapy was useful and feasible. Besides ECG monitoring, LED irradiation and HRV calculation, the data of the embedded system can be transmitted by UART to the remote PC for storage, display and analysis. Furthermore, the embedded system can receive the setting value from PC. For example, adjusting the irradiated dosage, LED operation rate and duty cycle to find out the influence factor of the ANS. Conclusion An embedded system with ECG monitoring and LED infrared stimulation was realized, it combined physiological state monitoring and light stimulation, it make the ANS feedback and real time adjustment possible. Keywords: Autonomic nervous system, ECG, HRV, embedded system 1. Background/ Objectives and Goals Physiological signal are measured to monitor the body state and the common application in medicine instrument. For example, heart rate variability in ECG, brain wave in EEG, blood 42

pressure, breathes rate, and blood oxygen, etc. NASA and Stanford University developed LifeGuard system, which can record ECG, body temperature, breathing rate, blood pressure, and 3-axis accelerometer [1]. Intelligent textiles are a combination of textile, electronic, information of high technological products, which integrate active electrode textile can monitor physiological signal any time [2]. Heart rate variability is the differences between heartbeat cycles. It can reflect the state of the ANS. HRV can be used as a risk judgment factor of coronary artery occlusion, to predict coronary occlusion situation [3]. Low frequency stimulation is widely used in many medical fields, such as sound, magnetic field, electric field, and light which have different treatment effect [4]. Low level laser can cause physiological change. In vitro, He-Ne laser irradiate cells whose ATP were enhanced [5]. Low level laser can help to improve the peripheral never surgery [6]. Laser also can reduce the pain of chronic disease [7]. Visual stimulation with the irradiation by white LED light (1,350 lux) can improve the seasonal affective disorder (SAD) [8]. Besides, LED is also used in dermatology [9]. In animal experiments, blue LED light can accelerate wound healing of mice [10]. Wu et al. proposed stimulating Neiguan point (PC6) with low level laser was used to adjust the ANS of the night shift workers [11]. And in other studies, Laser array and LED array were used to stimulate the palm of the subjects with open eyes in different clinical trial, the alpha powers were both increased significantly [12, 13]. So the combination of ANS monitoring and light treatment was useful and feasible. The proposed system combined with physiological signal monitoring and low power photo therapy, it is suitable in home use as a medical device. The system measured ECG signal, calculating HRV, then analyzing autonomous never state. When the calculated values are out of the criteria, the system will alert the user to turn on infrared light to stimulate and balance ANS. 2. Methods The system architecture was shown in Figure 1. There were three parts included, ECG module, LED array and embedded system. Computer is for remote control, data storage. The photo was shown in Figure 2.

2.1. ECG module An electrocardiogram module was designed in this study. The module includes instrument operational amplifier, inverter operational amplifier, low-pass filter, and the total amplifier gain is 2,100. 2.2. LED array arrangement and dose calculation An array of six LED arranged in a triangle form was used to radiate near infrared light at specific body site of the subjects. A LED (model BIR-BO13J4G, manufactured by YELLOW STONE Co., Ltd. Taiwan) was used in this study. The central wavelength of LED was 850 nm, 43

the irradiated angle is 20 degree, the default output power of each LED was set at 10 mW, the operation frequency was 10 Hz, and the duty cycle was 50%. The light spot of LED can be considered as a circle. The area of the light spot was approximately equal to 28 mm 2 at the skin. Thus, the dosage would be approximately 5.4 J/cm2 for 5 minutes treatment. Dose can be controlled by adjusting light duty cycle. 2.3. Embedded system There are processing, recording, communication and display function in the embedded system which is responsible for analyzing ECG signal and control LED irradiation. Sunplus 32-bit RISC processor was built on the development board SPCE3200, the photo shown in Figure 2. There are many interfaces included 12-bit ADC, 16-bit DAC; UART, SPI, I2C, USB, Flash, SD card; LCD; etc in the embedded processor. The system block was shown in figure 3.

2.4. Experiment procedure Each subject sat in a chair and was required to put LED array on Neiguan point (PC6). The subject was instructed to be relaxed, kept the eyes open and withhold from any movement. In the beginning, each subject was required to relax for five minutes in order to be in a stable physiological state. There were three sessions in the experiment procedure, before stimulation (5 minutes, session 1), during stimulation (5 minutes, session 2), and after stimulation (5 minutes, session 3). The ongoing ECG was recorded in first and third sessions. In the experiment, Ag/AgCl electrode pad were bond on the chest, because the quality of ECG 44

signal was better than the limbs. The signal electrode was on the left chest near the heart, the reference electrode was on the right chest, and the ground electrode was near navel. The recommend stimulation site was on Neiguan point (PC6). Low level light irradiation is noninvasive; the effect is moderate and less impact on user psychological pressure. The wavelength of LED light is 850 nm which belonged infrared light. 2.5. Introduction of HRV analysis The detection of heartbeat is from QRS complexes, and R is the peak of the complex. HRV is standard deviation of time sequence (RR-intervals). There are two types of analysis in HRV. Time domain analysis reports the activity of cardiovascular system, and frequency domain analysis reflects the sympathovagal balance of ANS. 2.5.1. Time domain method For the time domain analysis, the RR intervals of ECG records were analyzed and the results were expressed in the form of normal-to-normal (NN) intervals. Then the standard deviation of the intervals (SDNN) was calculated. SDNN (ms): the standard deviation of the RR intervals seen on the ECG 2.5.2. Frequency domain method For the frequency domain analysis, Fourier transform was applied to complex heartbeat signals. The frequency analysis is provided the spectral composition of these variations. LF band (0.04–0. 15 Hz) spectral power reflects both sympathetic and vagal influences. HF band (0.15– 0.4 Hz) spectral power reflects vagal modulation of cardiac activity [14]. The relevant indices and signal were defined and obtained as follows:

2.6. Function block Function block was shown in Figure 4. ECG signal was amplified and filtered by ECG module. Then the signal was transmitted to the embedded system. With internal ADC port, analog signal was converted into digital signal, and then ECG data will be processed and stored into the SD card. Then ECG signal was processed by the real-time QRS detection algorithm [14]. The pre-processed functions include differentiation, squaring and moving-window integration, and an adjustable threshold to decide the period of the heartbeat. The system will calculate HRV on time-domain and frequency-domain to analyze autonomous never state. On time domain, the heartbeat rate was calculated to get the standard deviation of normal to normal RR intervals (SDNN). On frequency domain, because the RR interval of the heartbeat was not equal, firstly using Berger manner [15] to convert heartbeat as equidistant sampling, and then does the fast Fourier transform (FFT), analyze LF and HF power. When the HRV values were out of the self defined criteria, the system will remind the subject to turn on LED stimulation to balance ANS.

45

2.7. Alert criteria A high risk group may be selected by the dichotomy limits of SDNN 200 bp unigenes were distinguished by NCBI Nr and Swiss-prot database. Using the BLASTx algorithm, the function analysis of unigenes were classified into several EuKaryotic Orthologous Groups of proteins (KOG) categories and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways with E-value< 10-5. Based on nr-matched unigenes, the annotation of Gene ontology (GO) was obtained by WEGO software. Gene expression profile was constructed by differential expressed gene (DEG) analysis. Results and Discussion: High quality reads were assembled into a total of 102,620 unigenes. 32.5% of the unigene were matched with NCBI nr database, 30.4% matched with swiss-prot. 15,523 unigenes were classified into 42 GO functional groups. 8,465 unigenes were distributed into 255 KEGG pathways. In DEG analysis, of 10 annotated unigenes were selected to demonstrate the expression profiles in primordial follicle recruitment and (Table 1). The down regulation of PI3K/PTEN-Akt related genes showed the primordial follicles were maintained constant recruitment in dormant state. However, the inhibitory PI3K/PTEN-Akt pathway was declined after SPH induction. Most of the ovarian follicles were developed into further stages in the ovarian follicles of female eel. Conclusion: In this study, NGS data provide us more information concerning key genes emerged from gene expression libraries in previtellogenic ovary development. These findings will facilitate us to set the initial timing of artificial hormone induction. Keyword: Transcriptome, primordial follicles, previtellogenic ovary, female eels 647

References: Adhikari D. and Liu K. (2009) Molecular mechanisms underlying the activation of mammalian primordial follicles. Endocr. Rev. 30:438-464 Kezele P. R. et al. (2005) Alterations in the ovarian transcriptome during primordial follicle assembly and development. Biol. Reprod. 72:241-255

648

GLSBE-89 Screening and Identification of Cellulase Producing Yeast from RongKho Forest, Ubon Ratchathani University Ubon Ratchathani University Jantaporn Thongekkaew* and Jarupan Kongsanthia Department of Biological Science, Faculty of Science, Ubon Ratchathani University, Thailand E-mail: [email protected] Abstract In this present study, a total of 82 yeast strains were obtained from 111 samples collected from various sources such as soil, tree barks and insect frass from RongKho forest, Ubon Ratchathani University. They were screened and identified for cellulase producing wild yeasts. Based on their capabilities of degrading carboxymethyl cellulose (CMC) as a sole carbon sources in solid medium, 45 strains performed cellulose hydrolysis which was verified by the colorless halo in the solid medium. Among them, 3 strains namely UBU-SK6, UBU-JK8 and UBU-JK9/1 were showed higher levels of enzyme activity. After the step of solid medium selection, yeast cells were grown in liquid medium containing 1% yeast extract and different concentration of carboxymethyl cellulose (0.5-1.5%) under shaking with 150 rpm at 30°C and for 168 h. The enzyme activity of these 3 strains exhibited 0.224, 0.238 and 0.26 UmL-1, respectively. Afterwards, these wild yeast strains were molecularly identified by sequencing the ITS1-5.8S-ITS2 and D1/D2 domains of the subunit (26S) ribosomal DNA. The identification of UBU-SK6, UBU-JK8 and UBU-JK9/1 showed 98-100% similarity as Candida sp. 05-7-186T, Candida easanensis and Candida sp. ST-390 respectively. Key words: screening, identification, cellulase, Candida sp. Introduction Cellulolytic microorganisms play an important role in the biosphere by reducing cellulose. During their growth on cellulosic material, cellulases are synthesized by microorganisms. Cellulases from microorganisms have attracted a great deal of attention because they are highly specific biocatalysts that act in synergy to release sugars, especially glucose which is of great industrial interest due to the potential for its conversion to, for example, in the textile and paper industries and in animal feed (Enari, 1983; Suchita & Ramesh, 2006; Zhou et al., 2008) as well as the production of bioenergy as bioethanol and biofuels (Castro and Pereira, 2010). There is a wide variety of microorganisms in nature which produce cellulases; only a few yeast strains have been seen as major producers of this enzyme. For this reason, explorations of biodiversity in the search for new biocatalysts by selecting microorganisms from nature represents a method for discovering new enzymes which may permit the development of bio-catalysis on an industrial scale. Currently, there is a great interest in finding microorganism species that are not yet known as interesting producers of inputs to industry. Moreover, biocatalysts also benefit in biotechnological processes which can replace many chemical processes, in terms of production of key products. As an example is the replacement of acid hydrolysis for enzymatic hydrolysis, as is well known in the industry (Peixoto, 2006). Traditionally, yeasts have been identified and classified by cellular morphology and their distinctive reactions on a standardized set of fermentation and assimilation tests. These methods are laborious and time-consuming. Moreover, these characteristics are influenced by 649

culture conditions and can provide uncertain results because of strain variability. Given these difficulties and the impracticality of identification most species from genetic crosses, molecular techniques, including the sequencing of rRNA genes and/or their flanking ITS regions are increasingly used for identification because the sequencing of rRNA genes and/or their flanking ITS regions have shown that many of these genptypic characteristics are strain specific for recognition of either species or genera. Moreover, these techniques are fast, easy and reasonably precise, and thus suitable for the rapid screening of isolates (Thongekkaew et al, 2012). The main goals of the present study included the screening and identification of cellulase producing wild yeasts, isolated from samples collected from RongKho forest, Ubon Ratchathani University. Materials and Methods Yeast Isolation Two grams of each sample from natural sources such as soil, mushroom, fruits (one wound per fruit) and insect frass was incubated in 5 ml of Yeast extract-Malt extract (YM) broth (3 gL-1 yeast extract, 3 gL-1 malt extract, 5 gL-1 peptone and 10 gL-1 glucose) containing 0.1gL-1 chloramphenicol and 2 gL-1 sodium propionate to suppress bacterial and mold growth, respectively at 30ºC for 2-3 days. An aliquot of 0.1 ml of each yeast sample suspension was spread on YM agar supplemented with 0.1 gL-1chloramphenicol and 2 gL-1 sodium propionate. Plates were incubated at 30ºC for 2-3 days. Colonies of yeasts were purified by repeated streaking three times on YM agar plates without the chloramphenicol and sodium propionate. Solid Medium Selection For qualitative analysis of cellulase production; all strains of yeast were incubated at 30 oC for 2 days on carboxymethyl cellulose sodium salt (CMC-Na salt) agar (1%yeast extract, 0.5% CMC-Na salt and 1.8% agar). After incubation, cellulase-producing strains were identified based on the formation of a clear halo around the colonies, which were made visible by Congo red staining. Index of Enzymatic Activity The enzymatic activities were estimated according to the method reported by Anagnostakis and Hankin (1975) who proposed an Enzymatic Activity Index (EAI), which is the ratio between the halo diameter and the colony diameter. Determination of Carboxymethyl Cellulase Production The carboxymethyl cellulase-positive isolates were evaluated for carboxymethyl cellulase activity by growing the culture in carboxymethyl cellulose-yeast extract medium containing 1% yeast extract containing and different concentrations of carboxymethyl cellulose (0.5-1.5%) under shaking with 150 rpm at 30ºC for 7 days. Subsequently, the cells were pellet by centrifugation, and the cell-free supernatant was used as the source of the enzyme. Carboxymethyl cellulase activity was determined by measuring the release of reducing sugars from CMC-Na salt using the Somogyi–Nelson method (Somogyi, 1952). Reaction mixtures contained 0.45 mL of 0.5% carboxymethyl-cellulose sodium salt in 50 mM potassium phosphate buffer, pH 7 and 0.05 ml of each enzyme fraction. Control lacked the enzyme fraction. After incubation at 50ºC for 15 min, the reaction was terminated by adding 0.5 mL of Somogyi reagent. The mixture was vortexed, placed in a boiling-water bath for 10 min, and cooled to room temperature. A 0.5 mL of Nelson reagent was added. After being vortex, the mixture was centrifuged to remove any precipitate, and the absorbance of the supernatant was measured at 660 nm. One international unit (IU) of enzyme activity was defined as the amount 650

of enzyme required to release 1 micromol of glucose from CMC-Na salt in 1 minute under the assay condition. Molecular Identification Pure isolated yeast cells from 3 mL of each 24-h culture were harvested by centrifugation and DNA extraction as described by Sambrook et al. (1989). Identification was carried out by sequencing the gene that encodes the 5.8S regions of ribosomal ribonucleic acid (RNA) and the spacer regions ITS-1 and ITS-2. To do so, the universal primers ITS4 (5′ TCCTCCGCTTA-TTGATATGC 3′) and ITS5 (5′ GGAAGTAAAAGTCGTAACAAGG 3′) were used. The D1/D2 domains of the 26S subunit were also sequenced by using the primers NL1 (5′ GCATATCAA-TAAGCGGAGGAAAAG 3′) and NL4 (5′ GGTCCGTGTTTCAAGACGG 3'), according to the methodology described by Kurtzman and Robnett (1997); Esteve-Zarzoso et al. (1999) and Leaw et al. (2006). The PCR product was purified by using QIAquick PCR Purification Kit and analyzed with 1.5% agarose gel electrophoresis. The PCR products were sequenced using the ITS4, ITS5, NL-1 and NL-4 primers using either Amersham Pharmacia ALF Express II or ABI 310 (capillary) automated DNA sequencer, following the manufacturer‘s instructions. For identification, the obtained sequences were compared with those of all known yeast species, available at the GenBank database (http://www.ncbi.nlm.nih.gov/BLAST/) (Yarrow, 1998). Results and Discussion Screening of Cellulase Producing Yeast For screening of cellulase producing yeast, a total of 82 previously isolated wild yeast strains were grown in Petri dishes in order to identify those able to degrade carboxymethyl cellulose (CMC). After incubation, the plates were revealed with a Congo red solution and halo formation was analyzed. The formation of a clear halo surrounding the colony indicates hydrolytic activity (Figure 1). Analyzing Figure1, it can be visualized that a large halo of hydrolysis formed after 48 h of incubation on agar with carboxymethyl cellulose (CMC). The enzymatic activity index (EAI) is a semi-quantitative parameter commonly used to assess enzyme production by microorganisms in solid medium (Anagnostakis and Hankin, 1975; Lopes et al., 2011). The result of cellulase producing yeast screening showed that 45 strains performed cellulose hydrolysis (Figure 2). Figure 2 shows values of enzyme activity recorded for the 45 selected strains when grown on solid medium containing carboxymethyl cellulose as a carbon source. Some authors recommend an EAI ≥ 2.5 for considering a microorganism as a producer of enzymes in solid medium (Peixoto, 2006; Oliveira et al., 2006; Lopes et al., 2011). Since colonies with the highest EAI are those with higher extracellular enzyme activity (Ceska, 1971; Ruegger and Tauk-Tornisielo, 2004; Oliveira et al., 2006). Among them, 3 strains namely UBU-SK6, UBU-JK8 and UBU-JK9/1 exhibited the EAI 3.5, 2.67 and 3, respectively. Therefore, these three were considered cellulase producing strains and were assayed in liquid medium fermentations.

651

Production of Cellulase in Liquid Medium The result of cellulase production of three strains selected (UBU-SK6, UBU-JK8 and UBU-JK9/1) showed that culture medium containing 1.0% CMC is the most effective induction of the enzyme production; this value is quite close to the results for cellulase production Pseudomonas sp. (Gautam et al., 2010) and Alternaria sp. MS28 (Sohail et al., 2011). The highest cellulase activity of UBU-SK6 was observed at 96 h of fermentation of 0.224 UmL-1 (Figure 3) while the highest enzyme activity of yeast namely UBU-JK8 and UBU-JK9/1 were observed at 120 h of fermentations of 0.238 and 0.26 UmL-1, respectively as shown in Figures 4 and 5. When comparing cellulase production of our study to different earlier yeast strains, the activities of yeast named UBU-SK6, UBU-JK8 and UBU-JK9/1) were lower than those of Candida stellata (30.6 UmL-1) (Strauss, 2001), Aureobasidium pullulans strain PJF-4b (0.44 UmL-1), Candida glabrata strain PJS-4a (0.38 UmL-1) and Candida litsaeae strain PJB-3 (0.36 UmL-1) (Thogekkaew et al, 2012). However, the yields of cellulase from our yeasts isolates were two-fold higher than those of the most active Trichosporon spp. (0.12 and 0.14 UmL-1) (Steven & Payne, 1977).

652

653

Molecular Identification Molecular identification was based on ribosomal targets: the ITS1-5.8S-ITS2 and D1-D2 domains of the RNA operon. In recent years, the number of ITS sequences available in public databases has increased significantly, and the expanding database may improve the quality and accuracy of fungal identification (Hinrikson et al., 2005; Leaw et al., 2006). The D1/D2 recombinant deoxy-ribonucleic acid (rDNA) region has been sequenced for almost all known yeasts, including non-pathogenic species. As an identification tool, the sequencing of this region is relatively simple to apply and yields clear results (Kurtzman and Robnett, 1997; Leaw et al., 2006). The BLAST searches, even as phylogenetic analysis, are used for identification by searching public databases for the closest known sequences to the unidentified yeast (Suh et al., 2006). A BLAST search in the NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi) of the ITS1-5.8S-ITS2 and D1/D2 domains of the subunit (26S) ribosomal DNA of UBU-SK6, UBU-JK8 and UBU-JK9/1 showed 98-100% sequence identity with Candida sp. 05-7-186T, Candida easanensis and Candida sp. ST-390 respectively. Conclusion From a total of 82 previously isolated wild yeast strains, 45 strains showed a halo of hydrolyzed substrate in the solid medium and three strains stood out for having high levels of enzyme activity. Considering production of cellulase from these yeast strains in liquid media, the result showed that the enzyme activity of these 3 strains, UBU-SK6, UBU-JK8 and UBU-JK9/1, exhibited 0.224, 0.238 and 0.26 UmL-1, respectively. The identification of UBU-SK6, UBU-JK8 and UBU-JK9/1 by sequencing the ITS1-5.8S-ITS2 and D1/D2 domains of the subunit (26S) ribosomal DNA showed 98-100% similarity as Candida sp. 05-7-186T, Candida easanensis and Candida sp. ST-390 respectively. Currently, little is reported in literature regarding the production of cellulases from Candida genus, which permits numerous studies on novel microorganisms in relation to production of enzymes of industrial interest. Thus, the results of this study stimulate new investigations, seeking applications in enzymatic hydrolysis for production of bioenergy. Acknowledgement 654

The authors wish to express their gratitude to Higher Education Research Promotion (HERP) belonging to office of the higher education commission for a financial support. References Anagnostakis, S.L. and Hankin, L. (1975). Use of selective media to detect enzyme production by microorganisms in food products. J. Milk Food Technol., 38: 570-572. Castro, A.M. and Pereira, N.J. (2010). Production, Properties and Application of Cellulases in the Hydrolysis of Agroindustrial Wastes. Quim. Nova., 33(1): 181-188. Ceska, M. (1971). Enzyme catalysis of solidified media. Eur. J. Biochem., 71(22): 186-192. Chen, P.J., Wei, T.C., Chang, Y.T. and Lin, L.P. (2004). Purification and characterization of carboxymethylcellulase from Sinorhizobium fredii. Bot. Bull. Acad. Sin., 45: 111-118. Enari, T.M. (1983). Microbial cellulases: In:Microbial enzymes and biotechnology, Foogarty, W.M. (Ed). Applied Science Publishers, London, pp. 183-224. Esteve-Zarzoso, B., Belloch, C., Uruburu, F. and, Querol, A. (1999). Identification of yeast by RFLP analysis of the 5.8S rRNA gene and the two ribosomal internal transcribed spacers. Int. J. Syst. Bacteriol., 49: 329-337. Gautam, S.P., Bundela, P.S., Pandey, A.K., Jamaluddin, Awasthi, M.K. and Sarsaiya, S. (2010). Cellulase production by Pseudomonas ap. Isolated from municipal solid waste compost. Int. J. Acad. Re. 2, 330-333. Hinrikson, H.P., Hurst, S.F., Lott, T.J., Warnock, D.W. and Morrison, C.J. (2005). Assessment of ribosomal large-subunit D1–D2, internal transcribed spacer 1, and internal transcribed spacer 2 regions as targets for molecular identification of medically important Aspergillus species. J. Clin. Microbiol., 43: 2092-2103. Kurtzman, C.P. and Robnett, C.J. (1997). Identification of clinically important ascomycetous yeasts based on nucleotide divergence in the 5‘ end of the large-subunit (26S) ribosomal DNA gene. J. Clin. Microbiol., 35: 1216-1223. Leaw, S.N., Chang, H.C., Sun, H.F., Barton, R., Bouchara, J.P. and Chang, T.C. (2006). Identification of medically important yeast species by sequence analysis of the internal transcribed spacer regions. J. Clin. Microbiol., 44: 693-699. Lopes, F., Motta, F., Andrade, C.C.P., Rodrigues, M.I. and, Maugeri-Filh, F. 2011. Thermo-Stable Xylanases from Non Conventional Yeasts. J. Microbial. Biochem. Technol., 3(3): 36-42. Oliveira, N.A., Oliveira, L.A., Andrade, J.S. and Chagas Júnior, A.F. (2006). Extracellular hydrolytic enzymes in indigenous strains of rhizobia in Central Amazonia, Amazonas, Brazil. Ciênc. Tecnol. Aliment. 26: 853-860. Peixoto, A.B. (2006). Study of the production of enzymes and gums by wild yeasts collected in various regions of Brazil. M.S. Thesis. Faculty of Food Engineering, University of Campinas, Brazil, p. 84. Ruegger, M.J.S. and Tauk-Tornisielo, S.M. (2004). Cellulase activity of fungi isolated from soil of Juréia-Itatins, São Paulo, Brazil. Revista Brasil Bot., 27(2): 205-211. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989). Molecular cloning: a laboratory manual. 2nd(Ed). Cold Spring Harbor Publishers Inc, New York, USA Sohail, M., Ahmad, A., and Khan S.A. (2011). Production of cellulases from Alternaria sp. MS28 and their partial characterization. Pak. J. Bot. 43(6), 3001-3006. Somogyi, M. (1952). Notes on sugar determination. J. Bio. Chem. 195, 19-23. Stevens, B.J.H. and Payne, J. 1977. Cellulase and Xylanase Production by Yeasts of the Genus Trichosporon. J. Gen. Micro. 100, 381-393. Strauss, M.L.A., Jolly, N.P., Lambrechts1, M.G. and van Rensburg, P. (2001). Screening for the production of extracellular hydrolytic enzymes by non-Saccharomyces wine yeasts. J. Appl. Micro. 91, 182-190. 655

Suh, S.O., Blackwell, M., Kurtzman, C.P. and Lachance, M.A. (2006). Phylogenetics of Saccharomycetales, the ascomycete yeasts. Mycologia, 98(6): 1006-1017. Thongekkaew, J., Khumsap, A. and Chatsa-nga, P. (2012). Yeasts in mixed deciduous forest areas of Phujong Nayoy National Park and their ability to produce xylanase and carboxymethyl cellulase. Songklanakarin J. Sci. Technol. 34 (2), 157-163. Yarrow, D. (1998). Methods for the isolation, maintenance and identification of yeasts. In: Kurtzman CP, Fell JW (eds) The yeasts, a taxonomic study, 4th ed. Elsevier, Amsterdam, The Netherlands. Zhou, J., Wang, Y., Chu, J., Zhuang, Y., Zhang, S. and Yin, P. (2008). Identification and purification of the main components of cellulases from a mutant strain of Trichoderma viride T 100-14. Bioresour. Technol., 99: 6826-6833.

656

Information Engineering and Technology (3)

Thursday, March 31, 2016

13:10-14:40

Room B

Session Chair: Prof. Yeon Ho Chung ACEAIT-7301 Using Kinect and Human Skeleton Pattern to Build an Automatic Security System Chih-Yu Jian︱Takming University of Science and Technology Yi-Ting Hong︱Takming University of Science and Technology Wan-Jiun Liau︱Takming University of Science and Technology Chi-Hua Lee︱Takming University of Science and Technology ACEAIT-7325 Emission Spectra Analysis of Copper Plasma Induced by Laser Hao-Xiao Jian︱North University of China Sang Tao︱North University of China Han-Chang Zhou︱North University of China ACEAIT-7521 Using Storyboard to Improve Video Retarget Quality by Pattern Matching Method Wen-Pinn Fang︱Yuan Ze University Jiu-Sheng Kuo︱Yuan Ze University ACEAIT-7519 Posture Imitation and Balance Learning for Humanoid Robots Kao-Shing Hwang︱National Sun Yat-Sen University Wei-Cheng Jaing︱National Sun Yat-Sen University Jin-Ling Lin︱Shih Hsin University, Taipei ACEAIT-5247 A Face Recognition System Based on Mobile Device Using Multi-Scale and Local Gradient Feature Shi-Jinn Horng︱National Taiwan University of Science and Technology Min-Shiue Hsieh︱National Taiwan University of Science and Technology

657

ACEAIT-7556 Zone-Based LED Signal Detection for Smartphone Visible Light Communications Jae Hyuc Jang︱Pukyong National University Ho Hyeong Jo︱Pukyong National University Hyeon Ho Kim︱Pukyong National University Yong Hyeon Kim︱Pukyong National University Yeon Ho Chung︱Pukyong National University

658

ACEAIT-7301 Using Kinect and Human Skeleton Pattern to Build an Automatic Security System Chih-Yu Jian Department of Information Technology, Takming University of Science and Technology, Taiwan E-mail: [email protected] Yi-Ting Hong Department of Information Technology, Takming University of Science and Technology, Taiwan E-mail: [email protected] Wan-Jiun Liau Department of Information Technology, Takming University of Science and Technology, Taiwan E-mail: [email protected] Chi-Hua Lee Department of Information Technology, Takming University of Science and Technology, Taiwan E-mail: [email protected] Abstract Nowadays, people wants to protect their own family and properties. In the past, the security system is only based on security camera. The disadvantage is that it can only record the image or video of the intruder, but can‘t do any further processing. In this paper, we using Microsoft kinect as the image and spatial depth detecting device, and using pattern analysis method to analysis human body action. When the system detects intruders, this system will warn by smart phone application, social website or SMS. Keywords: Kinect, Anti-theft system, Home Security, Human skeleton pattern. 1. Introduction Currently ongoing in social progress, break through the door into the home is not a difficult thing , but busy lives often make at home unattended . Without any security measures, when the home was found again reported stolen , may not be able to catch the thief; things may also be brought back. If your security company to install security systems in addition to the cost of installation outside, you must also pay expensive monthly rental fee, for office workers is a serious burden. From the above known, we must keep abreast of the situation at home, so we developed a system similar to the monitor. When outsiders invasion, discrepancies found patterns of behavior and record will immediately alert. To achieve this goal with the system, we use Kinect, information technology research and development , " home security monitoring system ." Hopeing to use Kinect to detect physical node way, patterns of behavior people often record at home,when discrepancies found patterns of behavior and record, can be learned at home abnormal characters appear. Let people when away from home, the system can return when it detects abnormal, users can promptly informed 659

of the situation at home, reduce possible damage to property , damage to items missing, in order to ensure the safety of home. 2. Related Work 2.1 Human Skeleton Track Originally acquired Kinect sensor just basic image data, but the focus can be applied is to how to identify the image, capture user actions. Microsoft Research Centre is to address this problem, in fact, taking into account the depth of the image data , easily filtered pixel does not belong to the user , can also be a number of factors unrelated to filter out and tracking , the only remaining spot showing the location of the user , recycling image segmentation technique to divide these spots into the human body can identify the location, and to identify those parts of the joints , from these joint position can reconstruct the skeleton user (shown as Figure.1) .

Kinect for image recognition work made after the image information, kinect will detect the depth image is converted to a skeleton tracking system , the system can simultaneously track six six human identification of human body in two movements, each body can record 20 joints (shown as Figure.2).

2.2 Somatosensory Visual of Home Security Systems Based on [ 3 ] General Monitors can only provide access to track down afterwards , unable to deal with unexpected situations immediately. The design uses the Kinect device for the face and skeleton movement design a security system, whether for face recognition for family visitors, if not family then briefing . Allowing users to keep abreast of the status of visitors , determine whether the presence or absence of endangering the safety of the home , provide a secure , real-time monitoring system to enhance home security . 660

In this study, Kinect somatosensory device identification and movement of the face ,do a home security system integration , to provide a safe home design . Benefits: The study can also be applied in addition to the security appliance on wisdom . Disadvantages: Face similar too high , it may misjudge occur. Compare: This set of systems similar to our study , but we hope to find out the characteristics of each person's skeleton , not just to judge behavior . B. Kinect + Lego = Comprehensive Robot [7] This study wanted to be the depth of the controller and Kinect voice recognition function, make comprehensive and intelligent robots and Lego were linked, making more mobile security robots to achieve a more secure home protection. Benefits: 1. In addition to the robot it can be used as preservation purposes can also be used to make home care. 2. Mobility robot can use traditional fixed monitors simplify the use of fewer resources. Disadvantages: 1. Because it is made of mobile laptop use Kinect to drive, notebook power may not allow the robot to prolonged use. 2. The bottom of the robot is to use the wheels to move, if you accidentally fall will not continue to work. Compare: So that the body can use robots to move this idea is quite distinctive, but robots are still limited by the power and terrain, with more flexibility relative to other issues. C. Kinect Home Care Security System [5] Taiwan to face the challenges of an aging society and the growing response, government units have been increasing emphasis on home care for the elderly and safety, when encountered in most of the elderly at home not feeling is the use of a similar emergency button on a hospital bed as a caregiver notice warning triggers. The study wanted to use Kinect somatosensory sensors to replace this traditional warning button, because the traditional wired warning button binding the scope of activities of the elderly, so that the elderly can not be away from the bed or in the living room needs risk warning button Shique not found, the infinite button alerts when there are doubts about whether power is needed, so take advantage of Kinect somatosensory sensors placed in the home as a video monitoring system, sources of information, you can find out the location of the body and skeleton hands by motion analysis to determine the elderly simply wave action as a warning, trigger warning sound to alert caregivers need to be concerned about.

661

Benefits: 1. Use gestures as a trigger condition does not require multi-band sensors. 2. You can use voice recognition Kinect active steering orientation. Disadvantages: 1. Only to do with sound alerts, if nobody at home no one can know. 2. There are ways to trigger only a single gesture. 3. You can not define functions to other gestures. Compare: This system can use the built-in microphone Kinect to detect the position and turn the lens, just use simple gestures can be induced, but for the sudden illness might not be quickly detected. 3. Research Methodology and Implementation 3.1 System Architecture The system consists of kinect composed image recognition subsystem,when human bidy detected, then model analysis subsystem started to identify the proportion of node registrant. If the analysis results shows that the human body is an intruder, the warning subsystem will send warning message to authorized user, and they can check the captured photos or check the record.. The system architecture is shown in Figure.3

662

3.1.1 Image Recognition Subsystem Using kinect skeleton tracking function to implement the image recognition subsystem , each person can be recorded as a group consists of 20 nodes, including the trunk, limbs, etc.

3.1.2 Model Analysis Subsystem The goal of this subsystem is to recognize detected human body and identify the authority. Because the authority may be misjudged because similar skeleton ratio, so the custom gesture will be the second-step judgment. If the skeleton ratio and custom gesture are both mismatch, the system will determine that intrude detected. 663

In our experiment, we take three-distance scale value as the main judgment. First is from HIPCENTER to HIPRIGHT, second is from HIPRIGHT to KNEERIGHT, the last is from SHOULDERCENTER to SHOULDERRIGHT. The k-Nearest Neighbors algorithm is used to judge the dynamic value of the three sections. All the three scale value should coincide with the member data stored in database, if one of them is mismatch, the human body will be judged as intruder. 3.1.3 Warning subsystem When intruder detected, the warning subsystem will send warning message through many different way. The emergency notification application (now android system only) will send warning message to the authorized user, the user can see captured picture or record through this smartphone application. The authority user can also login to the system website to see detail information if they do not have any android smartphone. 3.2 System Flow Chart The complete system flow chart is shown as figure.4.

664

4. Research Results 4.1 Skeleton Ratio Data Our system is mainly through kinect skeleton recognition function to be judged, members skeleton information has been captured and stored in the data repository for comparison, determine whether member or intruder. The distance from head node to shoulder node as a reference to compare with distances of each node, stored skeleton data, then the scale value of each node can be used to be a standard. (shown as Figure.5).

665

And the recorded value of the members is shown as fig.9, the proportion value of each registrant may contains a permissible value. Testers A data shown in Figure 6; Testers B data in Figure 7; Testers C data shown in Figure 8.

666

667

The system is based on the proportion of each human body skeleton, carried a three-stage discrimination, the flow chart is shown as Figure.10.

4.2 Accuracy of Experiments to Identify 668

The experiment result is shown as Table.1. The success rate of our system is currently about 64%, also found that the system has three errors still amendment. Error type 1: Failed to determine the members within five seconds. Error type 2: The side frame misjudgment. Error type 3: When similar skeleton detected, it takes longer time to determine.

5. Conclusion & Future Work Using our system to prevent unauthorized intruder can save man power, but the disadvantage of kinect is the motor in base can only rotate 54 degree from left to right. If the space is bigger or obstacle exists, the system needs to connect more kinect to monitor. In the next step, we will try to connect more kinect and place them in different angle to improve the correctness. In the future work, we intend to allow users can watch the real-time image on the client (ex: smartphone). Again it is extending on human nodes, according to the paper the results of the Central University Professor Pan Junming [6], everyone has a different node positions, any two human skeletons used to compare the degree of similarity is not higher than Qi Cheng, by this point, we can clearly distinguish the intruder skeleton which exists in the repository, if the presence of members of the legal representatives, it is determined that there is no intruder, in this regard, we hope to do more accurate. In response to energy conservation, Kinect also introduced a standby function, we also hope to be a client to make Kinect enters the standby state, allows the user to enter standby mode in accordance with the needs saving effect has been reached. References [1]馬詡皓,〈多方位家電控制系統〉,銘傳大學資訊傳播工程學系碩士論文,2012。 [2]劉超群,〈Kinect 體感程式探索〉,松崗,2013。 [3]賴冠羽,〈基於體感視覺之居家安全防盜系統〉,僑光科技大學資訊科技研究所碩士 論文,2013。 [4]黃正揚,〈使用 Kinect 感測資料進行物體辨識之研究〉,國立台北科技大學自動化科 技研究所碩士論文,2013。 [5]蔡超宇,〈Kinect 居家照護安全系統〉,國立高雄應用科技大學碩士論文,2013。 [6]潘俊銘,〈使用 Kinect 骨架追蹤功能辨識人員〉,中央大學通訊工程碩士論文,2014。 [7]〈Kinect+樂高=全方位機器人〉,南臺科技大學資管四丙專題報告。 [8]Kinect for Windows。http://www.microsoft.com/en-us/Kinectforwindows/。 [9]TW-HKT 程式●城市●人。http://tw-hkt.blogspot.tw/。 [10]MSDN Evaluation Center 。 http://goo.gl/nh5IvS 。 http://www.yks.com.tw/chapters/14/knowledges/299?page=3。 669

[11] techbang , 「 身 體 就 是 控 制 器 , 微 軟 Kinect 是 怎 麼 做 到 的 ? 」 , 來 源 : http://www.techbang.com/posts/2936-get-to-know-how-it-works-Kinect。

670

ACEAIT-7325 Emission Spectra Analysis of Copper Plasma Induced by Laser Hao-xiaojian,b*, Zhou Han-chang,b Ministry of Education Key Laboratory of Instrumentation Science and Dynamic Measurement, North University of China, China b Key Laboratory of Instrumentation Science & Dynamic Measurement of Ministry of Education, North University of China, China E-mail address: [email protected] a

SangTao Key Laboratory of Instrumentation Science & Dynamic Measurement of Ministry of Education, North University of China, China E-mail address: [email protected] Abstract By using of Nd:YAG laser, spectral lines of copper were experimentally studied in air atmosphere. The plasma characteristic spectral line of copper were obtained by 119.1、154.6、 201.7 mJ  Pulse1 energy. Cu 510.5nm and 521.8nm were selected as the research object. The relationship between spectral line intensity and laser energy were analyzed and found that the broadening line is Lorenz and Gauss respectively. The results show that different bands correspond to different spectral broadening mechanism for Cu plasma emission spectrum and with the increase of laser energy, spectral line intensity is increased gradually, and the full width at half maximum (FWHM) of the spectral lines is also increased. Key words: Plasma emission spectrum; characteristic line; Spectral line broadening; 1. Introduction Intense pulsed laser shock solid material will generate plasma, which a large number of electrons, atoms, molecules, atomic and molecular clusters and their positive and negative ions is sputtered from the target surface[1]. The process of the laser induced plasma is a series of non-linear process in which there are many factors will affect the ionization properties of the material directly and the nature of the analyzed spectrum. These factors include: the pulse width and the shape of the pulse laser light, laser wavelength, laser energy, chemical and physical characteristics of the substances, the composition and the pressure of the ambient gas, the characteristics of the photodetector, the observation direction and others[2]. Shaikh et al. studied the variation of the cadmium plasma characteristic parameters (spectral intensity, electron temperature and electron density) under the condition of the different wavelengths and the different laser power densities, which found that with the increasing of the laser power density, electron temperature and electron density and increased exponentially[3]. Russo et al. studied the distillation effect of three different UV wavelengths (266nm, 213nm 157nm, and) of the laser-induced plasma, which confirmed that the shorter the laser wavelength, the lower the energy threshold[4]. The published literature mainly analyzed the electron temperature and electron density of copper plasma[5], the spectrum broadening mechanism of copper atom need to be further discussed. In this paper, the broadening mechanism of Cu plasma emission spectrum is studied by changing the laser energy. 2. Plasma Atomic Emission Spectrometry Broadening Mechanism According to Bohr frequency condition and level discontinuities, electromagnetic radiation is generated by the transition of electrons between the atomic energy levels, the energy of the 671

spectral line should be single. In fact, spectral line is not a single frequency, but with a certain frequency range, which means the spectral line has a certain width. The contour of the spectrum is the distribution values of the intensity of the spectral lines by the frequency. Traditionally, the line width at half intensity peak, which is the half-width (FWHM), called spectral width. The width of the spectral line is related to the atomic structure and the temperature and the intensity of the light source, which is necessary to understand the mechanism of the relationship between the spectral lines in atomic spectrometry. Spectral line is an important information source for the study of the plasma, in terms of the laser induced plasma atomic emission line, mainly include Lorenz Line and Gauss Line[6]. Spectral line broadening is the result of a variety of physical process and the two most important mechanism are the Stark broadening and Doppler broadening. In the optics, since the movement of the atoms relative to the detection apparatus in which emerge a wave frequency shift phenomenon and called Doppler effect[7]. In terms of the radiation of a large number of the same atomics in the state of random motion in the light source, which the atomics move in different directions at different speeds, even though the same frequency of the light emitted by each atom, the frequency of the light waves received by the detector also has a certain difference, which cause line broadening that is Doppler broadening. When the light emitting atom is coming to the detector, the wave frequency is higher than the center frequency, which occurred blue shift. When the light atom is far away from the detector, the frequency of the wave is lower than the center frequency, which occurred red shift. In the thermal equilibrium, the velocity distribution of gas molecules obey Maxwell distribution, the Doppler broadening is expressed as

In the formula (1), T is the absolute temperature (K), m is the atomic mass of radiation (Kg), k B is the Boltzmann constant ( JK 1 ). In order to calculate conveniently, use atomic weight M instead of mass m by M  N A m , N A is Avogadro constant ( mol 1 ). So the broadening can be simplified into the following formula:

From the formula (2), it is obvious that the higher the temperature, the smaller the atomic weight of the element and the longer the wavelength, the more obvious of the Doppler broadening. The calculation shows that Doppler linear function belongs to Gauss line[8]. Its linear type is:

In the formula (3), m is the particle mass, T is the temperature, c is the speed of light in vacuum. k is the Boltzman constant. In the laser induced plasma, each transmitter is influenced by the electric field of the surrounding ions or the electrons. The interaction will cause interference to the emission of atoms, not only broaden the spectral line profile, but also move the center of the spectral line, which is called Stark broadening. The FWHM of the Stark broadening and the shift of the Spectral center are the formulas below:

672

In the formula (3) and (4), ne  m3  is the Plasma electron density,  (m) the electron impact parameter, d  is the ratio of line shift and width,  is the Ion broadening parameter. Its linear type is:

In the formula (6), shift is the displacement caused by electron impact, 1/2 is the FWHM of the spectral line, A0 is the intensity of continuous spectrum, A is the spectra intensity, 0 is the central wavelength of the atomic spectral line without collision. The linear function of the Stark effect is the Lorentz line[9]. 3. The Experimental Device The experimental device is shown in the Fig.1. The laser is Nd:YAG(Neodymium-doped Yttrium Aluminium Garnet), which is a solid-state lasers, the laser wavelength is 1064nm and can produce pulses or a continuous laser. 10Hz TTL square-wave signal is generated by function signal generator, which is used to trigger dual - channel spectrometer (AvaSpec-2048FT). The spectrometer takes the rising edge of each square wave as the starting point of the measurement and response to a narrow pulse signal with the same frequency and a pulse width of 10 to trigger Nd:YAG laser. The laser beams emitted by the laser change the direction from horizontal to vertical after the aluminum mirrors and incident to the surface of the sample through a convex lens with a focal length of 100mm, then excite to form a plasma at the focal point. Micro Focus put the part of the radiation of the plasma light converge to the whole silica fiber and transmit to the spectrometer. The optical signal is projected onto the CCD array after splitting by Spectrometer. CCD detect the spectrum according to the preset delay time and the integration time, the measured spectral data is transmitted to computer through the USB line for processing. Put the spectroscope and joule meter to real-time monitoring of the laser pulse energy between the mirror and the focusing lens. An inner diameter of 40mm aluminum sample cell is placed by a stepper motor driven turntable.

4. Experimental Results and Analysis In the experiment, transfer the spectrometer to real-time acquisition and the laser to the maximum energy. Then impact on the copper target until the laser is stable relatively and the surface impurities of the copper target is ablated cleanly. Set the spectrometer in the single acquisition state, the delay time is 200ns, the integration time is 20ms, adjust the laser energy of 119.1, 154.6, 201.7, 256.2 mJ  Pulse1 respectively. Impact on the Cu sample, test 100 sets of data each time and adjust the speed of stepping rotary table for 10HZ. After the experiment, the spectra are shown in the Fig.2. 673

From the Fig.2, around the 500nm there are a few of discrete spectral line, three spectra of Cu I 521.8, 515.3 and 510.5nm are shown in the Fig.3.

Choose CuⅠ510.5 nm and 521.8 nm in the characteristics of the line as the research object, Fig. 4 (a) and Fig. 4 (b) are the spectrum distribution curve of different laser energy respectively. Fig. 4 (a) is the laser spectrum of CuⅠ510.5 nm within 0.5nm corresponding to different laser energy. The peak corresponding to the energy value is the intensity. Fig.4(b) is the laser spectrum of CuⅠ521.8 nm within 1nm corresponding to different laser energy.

674

From the Fig.4, figure (a) conform to the Lorenz lines and figure (b) conform to the Gauss linear approximately. Since the spectral intensity is highest of the CuⅠ521.8nm, so the copper atom transition probability is the largest and the emission is the strongest. Under the high temperature condition of the laser shock plasma, the broadening of here is Stark and the linetype is Lorenz line. For CuⅠ510.5nm, since the electron transition energy level activity is poorer, so the spectral intensity is weak, the Dopper broadening is the main aspect, so present Lorenz line. 5. Conclusions This paper expounds the broadening mechanism of laser induced plasma emission spectrum, use Nd:YAG pulsed laser ablation of copper target sample and change the laser energy to obtain characteristic spectral lines of different energy. The experimental results show that for copper atoms, line broadening mechanism corresponding to different wavelengths is different. With the increase of laser energy, the spectral line intensity and the FWHM increases gradually, the wavelength of the peak is not change. Acknowledgements This work was financially supported by Science and Technology Agency(No.120247 -20) and The National Natural Science Fund(No.61473267 ). References [1]. B. Kearton and Y. Mattley. LASER-INDUCED BREALDOWN SPECTROSCOPY: Sparking new applications. Nature Photonics, 2008(2): 537-540. [2]. C. Aragon and J. A. Aguilera. Characterization of laser induced plasmas by optical emission spectroscopy. A review of experiments and method. Spectrochim. Acta Part B, 675

2008(63): 893-916. [3] R. E. Russo, X. L. Mao, O. V. Borisov and H. C. Liu. Influence of wavelength on fractionation in laser ablation ICP-MS[J]. Anal. Atomic Spectrometry, 2000, (15): 1115 1120. [4] N. M. Shaikh, B. Rashid, S. Hafeez, S. Mahmood, M. Saleem and M. A. Baig. Diagnostic of cadmium plasma produced by laser ablation[J]. Appl. Phys., 2006, (100): 073102-1 073102-8. [5] J. S. Cowpe, J. S. Astin, R. D. Pilkington, A. E. Hill. Temporally resolved laser induced plasma diagnostics of single crystal silicon - Effects of ambient pressure[J]. Spectrochim. Acta Part B, 2008, (63): 1066 – 1071. [6] Choi D, Gong Y, Nam SH, Han SH, Yoo J, Lee Y. Laser-induced breakdown spectroscopy (LIBS) analysis of calcium ions dissolved in water using filter paper substrates: an ideal internal standard for precision improvement[J]. Appl Spectrosc, 2014, 68(2):198-212. [7] Lee Y, Oh SW, Han SH. Laser-induced breakdown spectroscopy (LIBS) of heavy metal ions at the sub-parts per million level in water[J]. Appl Spectrosc, 2012, 66(12):1385-96. [8] Choi D, Gong Y, Nam SH, Han SH, Yoo J, Lee Y. Laser-induced breakdown spectroscopy (LIBS) analysis of calcium ions dissolved in water using filter paper substrates: an ideal internal standard for precision improvement[J]. Appl Spectrosc, 2014, 68(2):198-212. [9] Kumar R, Rai AK, Alamelu D, Aggarwal SK. Monitoring of toxic elements present in sludge of industrial waste using CF-LIBS[J]. Environ Monit Assess, 2013, 185(1):171-80.

676

ACEAIT-7521 Using Storyboard to Improve Video Retarget Quality by Pattern Matching Method Wen-Pinn Fang* and Jiu-Sheng Kuo Department Information Communication, YuanZe University, Taiwan, R.O.C. [email protected] Abstract Recently, there are many commercial products are invented to show videos. The products include television, smart phone, tablet, and computer. People use these products somewhere because different situation. However the size and aspect ratio of these devices are different. If the video is resized in width and height individually or resize according to the minimum side, the presentation result will very bad. It is necessary to retarget the video for many display devices. The author has proposed a retarget method by searching the location of face in the frames. It is not enough because there are many frames in a video without face. This paper proposed a method to retarget the frames by the storyboard. Based on color analysis, salient region detection and SURF method, the most important region can be distinguished, and then retarget the corresponding frames. The result of this paper is better than the result that just resize the video by individual frames or just crop the frames without drama. Keyword: Video Retarget, Storyboard, Feature Matching 1. Introduction Now, it is very important to design a user friendly multimedia. A user friendly multimedia will make a lot of profit with good hit rate. In this paper, the media is focus on video. There are many device used to display a video. It causes some problems. One of the problems is the size of original video is different from the displayer. Users feel uncomfortable if they watch the aspect ratio[1] of video is wrong. For example, there are many screen size for smart phone[2]. The way to fit the video into the device is very important. There are many people proposed methods to retarget the video[3-32]. The methods have a problem that, the program or algorithm cannot distinguish the important region correctly with just color distance computing. The most precise method is to retarget the video by asking the director. However, it is very expansive. This paper proposed a method to make the cost of region marking lower. The rest of this paper is organized as follows: the background knowledge is shown in Section 2; the method is proposed in Section 3; Experimental results are shown in Section 4. Finally, the discussion is represented in Section 5. 2. Background Knowledge Before describe the proposed method there are some background knowledge need to explain. The background knowledge includes image retarget, video retarget, storyboard, and pattern matching. The detail is shown as below: 2.1 Image Resizing In the decade, there are many researches discuss how to resize an image. The methods include image cropping[3] [4] [5] [6], seam craving[7] [8] [9] [10] [11], wrapping[12] [13] [14] and hybrid approaches[15] [16] [17]. The cropping method reserves the most important region and remove others. Avidn and Shamir [7] proposed a method to find out the seams, remove the region negligible. Mansfield, etc[8]、Rubinstein and Shamir[9] and Grundmann[10] improve 677

the quality of the retarget mage. In the other direction, some papers[12-14] discuss the image distortion after it has been retargeted. .Rubinstein [15] combined cropping and scaling method to retarget an image. Sun, etc[16] proposed cyclic seam caving algorithm. Don, etc. [17] discuss the relationship between reserved region and whole effect. Some paper [18] [19] [20]discussed using coarse region instead of pixel analysis. Pritch [21], proposed shift-map to get better image quality. Ding [22] designed filter to get better result. 2.2 Video Resizing As shown as Fig.1,there are an American patent that using cropping and wrapping to do video resizing. There are some key points to retarget a video. If resize the images in a video individually and ignoring the relationship between the frames, it causes many troubles. As shown in Fig. 2 and Fig. 3, The troubles include frame inconsistent, ghost contour, color inconsistent. There are some paper discuss the cropping and wrapping methods to solve this problem. For example, [23] [24] [25] [26]discuss cropping method. [27] [28] [29][30] discuss the motion vector and retargeting method. [31] [32] discuss the relationship between moving object and retarget video. 2.3 Salient Region Detection Different from traditional segmentation, salient region detection only detect the salient foreground object from the background, rather than partition an image into regions of coherent properties as in general segmentation algorithms. As shown in Fig. 4, there is an original high resolution image, the salient region detection will find out the most important object in the image. The common method of salient region detection is computing the different between every point first, and then segment the image by the color distance.

678

2.4 Story board As the definition of Wikimedia[36], a storyboard is a graphic organizer in the form of illustrations or images displayed in sequence for the purpose of pre-visualizing a motion picture, animation, motion graphic or interactive media sequence. 3. Proposed Method This paper proposed a method to assist people to retarget a video. The method has four parts: (a) 679

retrieve the salient region.(b) split the images by texture analysis to important frames.(c) find out the important region from storyboard by pattern matching (d) Using wrapping, cropping and seam craving method to retarget the video. The detail is shown as below: (a) Retrieve the salient region. In this phase, we adopt the method which proposed by Cheng, etc,, el. [33] First at all, compute the distance by equation (1)

S is the salient value in image I. D(Ik, Ii) is the color distance metric between pixels Ik and Ii in the L*a*b* space. To decrease the computation time, compute the salient value by histogram as shown in equation(2).

fj is the probability of pixel color cj in image I. (b) Split the images by texture analysis to important frames. In this phase, the method proposed by Jiang, etc, al[34] is adopted. The idea is compute the probabilistic boundary as shown in equation (3)

where yt is a neighboring pixel of xt in frame t, and nv denotes the number of valid mapping. A mapping is defined to be valid, if the projection points xt‘ and yt‘ in frame t‘are neither occluded nor out-of-view. If pb (xt yt ) is large, it is very likely that there is a boundary across pixels xt and yt. (c) Find out the important region from storyboard by pattern matching In this phase, the storyboard translated to black and white images first. In the second step, the frames of the video translated to edges. And then compute the different between the storyboard images and frames. Mark the non-overlapped region. The non-overlapped regions are the candidate region which can be eliminated. (d) Using wrapping, cropping and seam craving method to retarget the video. In this phase, the method proposed by Bhamidipatiand and Raveendran[35] has been adopted. Using the equation(4)-(6) to determine the energy and delete the pixels recursively.

680

4. Experimental Result An experiment example is shown as Fig. 5. The storyboard is shown as Fig.5(a). The original frame is shown in Fig.5 (b) and Fig.5 (c) is the retargeted frame. (video source:http://www.storyboardworld.com/rights-of-the-game-points-optician-william-simpson)

681

(c) Fig.5 An example of retargeting. (a) is the storyboard, (b) is the original frame, and (c) is the retargeted frame. 4. Discussion In this paper, a video retarget assist method has been proposed. The method is not only detecting the salient region for retarget a video, but also providing a user friendly check method for director or video producer. There are some cases that the computer program cannot understand the drama of a video. For example, the key object maybe not so obvious in the video. The key object is possible to be removed by the existing methods. This paper solves the problem and save the time to check the quality of retargeted video. References https://en.wikipedia.org/wiki/Aspect_ratio http://developer.android.com/guide/practices/screens_support.html L. Chen, X. Xie, X. Fan, W. Ma, H. Zhang, H. Zhou, A visual attention model for adapting images on small displays, ACM Multimedia System Journal, Vol. 9 (4), pp. 353–364, 2004. H. Liu, X. Xie, W.-Y. Ma, H. Zhang, Automatic browsing of large pictures on mobile devices, in: Proceedings of ACM International Conference on Multimedia, pp. 148–155 , 2003. B. Suh, H. Ling, B. Bederson, D. Jacobs, Automatic browsing of large pictures on mobile devices, in: Proceedings of UIST, pp. 95–104, 2003. A. Santella, M. Agrawala, D. Decarlo, D. Salesin, M. Cohen, Gaze-based interaction for semiautomatic photo cropping, in: Proceedings of CHI, pp. 771–780, 2006. S. Avidan, A. Shamir, Seam carving for content-aware image resizing, ACM Transaction on Graphics Vol. 26 (3) Article 10, 2007. A. Mansfield, P. Gehler, L. V. Gool, C. Rother, Scene carving: Scene consistent image retargeting, Proceedings of European Conference on Computer Vision, 2010. M. Rubinstein, A. Shamir, S. Avidan, Improved seam carving for video retargeting, ACM Transaction on Graphics Vol. 27 (3) Article 16, 2008. M. Grundmann, V. Kwatra, M. Han, I. Essa, Discontinuous seam carving for video retargeting, in: Proceedings of International Conference on Computer Vision, 2010. Y. Liang, Z. Su, X. Luo, Patchwise scaling method for content-aware image resizing, Signal Processing, Vol.92 (5), pp. 1243–1257, 2012. R. Gal, O. Sorkine, D. Cohen-or, Feature-aware texturing, Computer, Vol. 11 (5) pp. 1–7 , 2006. Y. Wang, C. Tai, O. Sorkine, T. LEE, Optimized scale-and-stretch for image resizing, ACM Transaction on Graphics Vol. 27 (5), Article 118, 2008. 682

B. Li, Y. Chen, J. Wang, L. Duan, W. Gao, Fast retargeting with adaptive grid optimization, in: Proceedings of International Conference on Multimedia & Expo, pp. 1–4, 2011. M. Rubinstein, A. Shamir, S. Avidan, Multi-operator media retargeting, ACM Transaction on Graphics Vol. 28 (3), Article 23, 2009. J. Sun, H. Ling, Scale and object aware image retargeting for thumbnail browsing, Proceedings of International Conference on Computer Vision, 2011. W. Dong, N. Zhou, J. Paul, X. Zhang, Optimized image resizing using seam carving and scaling, in: Proceedings of ACM SIGGRAPH Asia 2009, Vol. 28, 2009. T. Cho, M. Butman, W. F. S. Avidan and, The patch transform and its applications to image editing, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2008, 2008. D. Simakov, Y. Caspi, E. Shechtman, M. Irani, Summarizing visual data using bidirectional similarity, in: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2008, 2008. C. Barnes, E. Shechtman, A. Finkelstein, D. Gold-man, Patchmatch: A randomized correspondence algorithm for structural image editing, ACM Transaction on Graphicsl. 28 (3) ,Article 24, 2009. Y. Pritch, E. Kav-Venaki, S. Peleg, Shift-map image editing, in: Proceedings of International Conference on Computer Vision, pp. 151–158, 2009. Y. Ding, J. Xiao, J. Yu, Importance filtering for image retargeting, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2008. F. Liu, M. Gleicher, Video retargeting: automating pan and scan, Multimedia pp.241–250, 2006. C. Tao, J. Jia, H. Sun, Active window oriented dynamic video retargeting, Proceedings of Workshop on Dynamical Vision, ICCV, 2007. T. Deselaers, P. Dreuw, H. Ney, Pan, zoom, scan time-coherent, trained automatic video cropping, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2008. M. L. Gleicher, F. Liu, Re-cinematography: Improving the camerawork of casual video, ACM Trans. Multimedia Comput. Commun, Vol.5 (1), pp. 1–28, 2008. L. Wolf, M. Guttmann, D. Cohen-Or, Non-homogeneous content-driven video-retargeting, in: Proceedings of International Conference on Computer Vision, 2007. Y. Zhang, S. Hu, R. R. Martin, Shrinkability maps for content-aware video resizing, Computer Graphics Forum Vol. 27 (7), pp.1797–1804, 2008. P. Krhenbhl, M. Lang, A. Hornung, M. Gross, A system for retargeting of streaming video, ACM Transaction on Graphics Vol. 28 (5), 2009. Y.Wang, H. Fu, O. Sorkine, T. Lee, H.-P. Seidel, Motion-aware temporal coherence for video resizing, ACM Transaction on Graphics Vol. 28 (5), 2009. Y. Wang, H. Lin, O. Sorkine, T.-Y. Lee, Motion-based video retargeting with optimized crop-and-warp, ACM Transaction on Graphics Vol. 29 (4), 2010 Y. Wang, J. Hsiao, O. Sorkine, T. Lee, Scalable and coherent video resizing with per-frame optimization, ACM Transaction on Graphics Vol. 30 (4), 2011. M. M. Cheng, N. J. Mitra, X. Huang, P. H. S. Torr, and S. M. Hu, Global Contrast based Salient Region Detection, IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, online. H. Jiang , G. Zhang, H. Wang , and H. Bao,E, Spatio-Temporal Video Segmentation of Static Scenes and Its Applications, IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 17, NO. 1, JANUARY 2015. P. K. Bhamidipati and V. R. Raveendran, Video retargeting using seam carving, https://www.google.com/patents/US9031358 683

https://en.wikipedia.org/wiki/Storyboard

684

ACEAIT-7519 Posture Imitation and Balance Learning for Humanoid Robots Kao-Shing Hwang Electrical Engineering , National Sun Yat-sen University, Kaohsiung E-mail address: [email protected] Wei-Cheng Jaing Electrical Engineering , National Sun Yat-sen University, Kaohsiung E-mail address: [email protected] Jin-Ling Lin Information Management, Shih Hsin University, Taipei E-mail address: [email protected] Abstract In robotics, controlling the robot by imitating human posture is going to be a tendency. To accomplish the objective of the posture imitation via the robot, there are two main points: extracting useful information from the human motion and solving the falling problem, caused by the insufficient sense of balance of the robot. This paper presents a conceptual approach for human posture imitation with balance control on the humanoid robot. First, this thesis uses Kinect to capture the human motion and extract key posture by calculating dissimilarity value. Furthermore, the same key postures are clustered via Online Clustering method. Finally, the Balance Learning is used to rebuild the sense of balance of the robot. Keywords: Humanoid robot, Reinforcement learning, Posture imitation, Posture balance, Key posture 1. Introduction In recent years, the interaction between human and humanoid robots becomes popular research topics. However, how to control the robots become important issues. So, making the robots to imitate human behaviors can form research methods. In order to imitate human behaviors, two imperative concepts should to be considered. First, the robots need to retrieve useful information from human motions. Since the human motions which like walking behaviors are usually continuous and regular, the robots isn‘t easy to perform the continuous behaviors. The analysis of the human postures derives as a crucial problem. The concepts of key postures are introduced to generalize the continuous behaviors which can make robots perform. This paper proposes some methods to achieve the goals. First, an online clustering method is proposed to retrieve the important key postures from the continuous behaviors. And then, the robots perform these key postures; it sometimes fails down the ground. Second, because the robots lack the balance mechanisms, balance methods in this paper make the robots stable perform the behaviors. 2. Methods The main topics in this paper allow the humanoid robot to achieve human-like posture. In order 685

to achieve the goals, the paper is divided into three parts. First part is a motion capture method. Second part is an online clustering method. Third part is a balance method. These three methods are combined to be similar to human and solve the falling down problem. 2-1 Motion Capture Method Kinect device perceives the skeleton posture from human body. As shown in Fig. 1, the device can get twenty body joints; then the cosine theorem is used to transfer the joint positions to the Kinect coordinate system. After this phase, the joint data forms twenty joint coordinates. After obtaining a series of information, this paper uses a dissimilarity equation to extract the key postures. The dissimilarity equation is shown as follows.

where ρi is the dissimilarity value between posture i-1 and posture i+1, N means joint number, Φi,j represents the angle of the joint j of the posture i.  j means the averaged angle of all postures in joint j. σi is the standard deviation of the posture i.

2-2 Online Clustering Method In order to prevent the same key postures from being classified into different posture clusters, this paper uses the Online Clustering Method to approach it. This method compares the input with the center point of each cluster to calculate the similarity. The similarity between two vector data is calculated by Euclidean distance. It appropriately determines that the input should be classified or produces a new cluster. If the closest similarity is smaller than the threshold, the cluster a will update its moving average of the cluster center. On the contrary, the new cluster will be created. Because a posture is composed with various kinds of joint, the threshold of each joint is defined separately and depending on the experiments. 2.3 Balance Method Since the basic gaits without any information about the angles of an ankle, it may cause walking unstably. Q-Learning was performed for humanoid robots to learn stable walking. The key postures obtained from motion capture are the basic gaits to train the humanoid robots. Base on Q-Learning, the learning structure of Balance Learning mainly design on State space, 686

Action space and Reward function. 2-3-1 State Space In order to help the robot to learn its sense of balance, the force sensors on the robot‘s feet is used to estimate Center of Press (COP). Because human posture includes two phases, single support phase (SSP) and double support phase (DSP), the calculation is also different. While the robot is in single support phase, the COP point is calculated by Equation (2) and Equation (3). Assume that the left foot in parallel with the right foot, and the skew situation of respective foot does not exist. The Projection of Center of Mass (PCOM) of DSP is calculated by the COP point for respective foot. The position coordinate of COP of left foot is ( X L , YL ) and the position coordinate of COP of right foot is ( X R , YR ) . The COP point for respective foot is calculated to estimate the Projection of Center of Mass (PCOM) of double support phase with Equation (4) and Equation (5), where FL and FR is the sum of force value of respective foot. In DSP case, the PCOM point is normalized by the distance in x-direction and the distance in d y-direction between two feet. The distance in x-direction d x and the distance in y-direction y , is calculated by the posture of two legs. Human posture includes two phase, single support phase (SSP) and double support phase (DSP), shown as Fig. 2 and Fig. 3.

As Fig. 4, we transfer the COP point or PCOM point into a discrete state space. The soles of feet are distinguished into nine regions which represents the stable quality. Furthermore, if robot agent falls down or these force values are all zero, the situation belongs to state 0. With the aim of maintaining balance, we distinguish the region 5 as stable region and this state can 687

be bigger than the others.

2-3-2 Action Space Just like the way human learning balance, we adjust full-body with a minimal sway to maintain the center of gravity. This paper proposes twenty four actions which are obtained by experiments to cause state transition as shown in Fig. 5. The moving of body limb causes the position of COP or COP point or PCOM point to move, there is a rule to adjust the body:

2-3-3 Reward Function For balance learning, the ideal situation is that COP point or PCOM point fall in stable region. It means that the action in the state is good for learning, so the agent receives a positive reward. On the contrary, COP point or PCOM point out of the stable region makes the agent to acquire a negative reward. Shown as Fig. 6, the agent receives a positive reward in stable region which is calculated by Equation (6).

688

2-3-4 Learning Process This paper proposed a learning structure of Q-Learning to train the robot to maintain its balance of posture. The designs of the elements of Q-Learning (e.g. state space, action space and reward function) are defined above. During the learning process, the robot agent maintains the balance of its posture by the decision of Q-Learning. This decision lets the robot agent adjust full-body by a specific sway (action), then the current state in transit to the next state. According to the reward function, the agent receives a reward to determine the profit of this decision. The cumulative reward becomes the experience of balance learning and updates by Equation (7), called Q-function. After a long-term learning, the robot agent finds optimal decisions to maintain the posture balance during the process. Finally, the robot agent increases the progress of the adjustment of posture, and achieves the target, implement human-like posture without falling. The algorithm of balance learning is shown in Fig. 7 and the flowchart of the learning process is shown in Fig. 8.

689

3. Results The paper implements the balance learning in two experiments: single support posture and double support posture. The result of simulation and implementation shows below. 3-1. Single Support Posture In the SSP case, the target of this experiment is to support and balance under the posture of single support phase. As shown in Fig. 9, the human demonstrates a posture with only single leg. After the robot extracts the posture, the balance learning is beginning. The learning process includes two stages and collects the last COP point or PCOM point in each episode. In the early stage, there are few points in the stable region. Also, the trend of the learning can be observed because the points which closing the stable region increase. It means that the robot realizes a good way to react to the unstable situation. In the final stage, the points stay in the stable region. Since it is often to fall down, the episodes in early stage are lack of the complete adjustment. In this paper, the failure rate is calculated with the number of adjustment and the number of unstable adjustment in every episode. The number of unstable adjustment is the number that the point is not in the stable region after adjustment. Along of the experiments is accumulated, the robot reduces the failure rate during the learning process. The COP distribution is shown in Fig. 10 and Fig. 11 respectively. The failure rate of the implementation of single leg posture is shown in Fig. 12. The implement result of humanoid robot in the real world is shown as Fig. 13 690

and Fig. 14.

691

3-2. Double Support Posture In the DSP case, the target is to support and balance under the posture of two legs mode. Fig. 15 shows that the human demonstrates a striding posture in front of the Kinect device. Because the posture is supported under two legs is more stable than the single leg posture, the wide stable region is used. During this experiment, the robot experiences several times of single leg posture. If the robot falls down with single leg, it receives a negative reward; otherwise, it receives a small positive reward. It helps the robot go across the situation under the posture of single leg mode and avoid it to insist on standing with single leg. The learning process of this experiment includes two stages and collects the last COP point or PCOM point in each episode. The diagrams of Fig. 16 and Fig. 17 show the early stage this experiment. The box with dotted line on the left side represents the left foot of the robot and the box with dotted line on the right side represents the right foot of the robot. The points in these boxes dispersedly locate at the rear side of the foot. After a long-term learning, the final stage shows the convergence of the points of the foot and the PCOM point of the feet coverage to the stable region. The failure rate of the implementation of two legs posture is shown in Fig. 18. The implement result of humanoid robot in the real world is shown as Fig. 19 and Fig. 20.

692

693

In consideration of the balance of posture, the COP or PCOM point of the robot is mainly used to determine the stability. The result of our proposed method converges these points to a stable region. Along a long-term learning, the robot decreases its failure rate. In the future, it is important to apply the balance learning to a continuous motion. Because of the transition of posture, the ideal state changes simultaneously. The robot agent estimates the trend of next posture and learns the continuous motion balance along this trend. 4. References Leslie Park Kaebling, Michael L. Littman, and Andrew W. Moore, Reinforcement Learning: A survey, Journal of Artificial Intelligence Research 4, pp. 237-285, 1996 K. Y. Yu, and Y. H. Sun, Static Stable Regions According to Center of Press to Quantify Humanoid Walking Stability, IEEE International Symposium on Robotic and Sensors Environments, pp. 172-177, 2011 M. Nakada, B. Allen, S. Morishima, and D. Terzopoulos, Learning Arm Motion Strategies for Balance Recovery of Humanoid Robots, International Conference on Emerging Security Technologies, 2010 Y. Chen, Q. Wu, X. He, C. Du, and J. Yang, Extracting Key Postures in A Human Action Video, IEEE 10th Workshop on Multimedia Signal, pp. 569-573, 2008 V. V. Nguyen, & J. H. Lee, Full-Body Imitation of Human Motions with Kinect and 694

Heterogeneous Kinematic Structure of Humanoid Robot, IEEE/SICE International Symposium on System Integration, pp. 93-98, 2012 W. C. Jiang, The Applications of Model Learning and Knowledge Sharing of Multi-Agents for Dyna-Q Architecture, Ph.D dissertation, National Chung Cheng University, 2013 K. S. Hwang, W. C. Jiang, and Y. J. Chen, Model Learning and Knowledge Sharing for a Multiagent System with Dyna-Q Learning, IEEE Transactions on cybernetics, Vol. K. S. Hwang, J. L. Lin, T. C. Huang, and H. J. Hsu, Humanoid Robot Gait Imitation, IEEE SICE Annual Conference, pp. 2124 – 2128, 2014 C. L. Hwang, B. L. Chen, H. H. Huang, and H. T. Syu, Hybrid Learning Model and MMSVM Classification for On-Line Visual Imitation of a Human with 3-D Motions, Journal of Robotics and Autonomous Systems, Vol. 71, pp. 150-165, 2015 S.F. Huang and W.-J. Wang, Human’s Postures Recognition by Using Kinect, National Symposium on System Science and Engineering, pp.187-192, 2011 H. I. Lin, and C. C. Chou, Humanoid Robot Imitation Using Kinect, International Conference on Advanced Robotics and Intelligent Systems, pp. 1-4, 2015 K.D. Nguyen, I.M. Chen, S.-H. Yeo, and B.-L. Duh, Motion Control of a Robotic Puppet through a Hybrid Motion Capture Device, IEEE International Conference on Automation Science and Engineering, pp. 753-758, 2007 D. Lee, C. Ott, Y. Nakamure, and G. Hirzinger, Physical Human Robot Interaction in Imitation Learning, IEEE International Conference on Robotics and Automation, 2011

695

ACEAIT-5247 A Face Recognition System Based on Mobile Device using Multi-Scale and Local Gradient Feature Shi-Jinn Horng Department of Computer Science and Information Engineering, National Taiwan University of Science and Technology, Taiwan Email: [email protected]; Min-Shiue Hsieh Department of Computer Science and Information Engineering, National Taiwan University of Science and Technology, Taiwan [email protected] Abstract Security problem got much attention due to the developing of technologies, the traditional security facility like setting password has become less and less unsafety. By this reason, the rate of using biometric characteristics to replace password had become higher and higher. For the biometric characteristics, although faces have less uniqueness than those of finger vein and palm vein, but it can be convenient to take pictures and can finish the identification without touching the device. That is why face recognition becomes one of major recognition techniques in biometric recognitions. In this paper, we propose a face recognition system based on Android system using haar-like feature to detect human face, and remove the impact of external factor by using image pre-processing algorithm, then extracting feature from face with a novel local binary patterns combining gradient information. And at last, we compute the similarity score with chi score and also combine the gradient information to increase the accuracy. Finally, many existing databases of face images have been used for experiments, and have achieved great recognition results. Keywords – face recognition, Android, local binary pattern, Image pre-processing, Chi square statistic, Gradient 1. Introduction Because biometric characteristic has its union characteristic and makes it hard to be faked, this technology had become a primary research these years. Using biometric identification as an access system, or to get the use right of some device. And there are some majority biometric characteristics we often used, like vein recognition, fingerprint recognition, iris recognition and face recognition. We can use these characteristic to increase information safety, but to get the image of vein, fingerprint or iris usually need special devices or cams, and face recognition is more convenience to get the image we use to process and recognition one person, and that‘s why face recognition can be a major research in biometric recognition with less characteristic than vein, fingerprint and iris. But face recognition also has its own challenges, like it is easy to be influenced by external element when putting into realistic application; for example, illumination, rotation, pose, shift, expression and etc. 2. Related Work In present stage, the technique of face detection can divide into two kinds: feature matching and learning matching. Feature matching mainly uses the features of face image, like skin color 696

or geometry traits. Learning matching is a learning system through neural network; for example, Adaboost is one of the most famous methods. [1] Although learning matching needs great amount of information and training in the beginning, it has high detection speed and is not easy to be affected by external environment. Therefore, it is a common technology in face detection. Similarity calculation aims at telling similar degree of two images through the features picking up from last step and conduct the judging process. Take area analysis as an example, histogram and Euclidean distance [2] are common ways. Then, use the feature-picking image to conclude the histogram of different area with area-divided way. After that, count the distance between two histograms according to histogram of corresponded areas through Euclidean distance. The distance formula is shown as Formula 1. Final similarity degree is the points added from every square and draw up a threshold to verify.

3. Image Pre-Processing A. Face Detection & Face Normalization To implement face detection, we used Haar-Like feature based on Adaboost Algorithm which is proposed by Viola & Jones‘. The detection method uses rectangle features to be the foundation of image classification, like the rectangle features show in Picture 1. Rectangle feature means the differences between the gray scale values in the same two or more rectangles on the image. This classifier has high detection speed and high accuracy rate, which make it suitable to be equipped on mobile devices. We can use the rectangle features through Adaboost algorithm to overlay weak classifiers into a strong classifier; calculating two most meaningful (highest proportion) rectangle features in classification of human faces, like noses or eyes, like Picture 2. The first one is two rectangle feature, representing the gray scale of eye area should less than which of the lower block. The second one is third rectangle feature, representing that the gray scale of two eyes‘ areas should less than the gray scale of the nose bridge. By the two features, it proves that the algorithm that Adaboost chose is really meaningful.

697

When the face is detected, we also use rectangle features to detect eyes and noses, getting the midpoint of eyes and nose in one face. It is for implementing the later step of matching two faces, like Picture 3.

To make face images be available to calculate the similarity point, we need to make every face image in the same scale, so we use bilinear interpolation to make resize the face image into a fixed scale. Besides, we use Formula 2 to calculate the angles of two eyes, then rotate the whole image with this angle, to make face being frontal.

B. Illumination Normalization Face image always has a problem that it is easy to be influenced by light changing. Illumination normalization introduced in this chapter can effectively filter out the problem that the original image affected by changing light, and can remain basic features and appearances; however, the speed of the old algorithm of illumination normalization is slow and the subtraction method tends to produce noise. Therefore, we proposed an illumination normalization and describes the methods step by step as follows. Smooth the original image through Gaussian. This smoothing image is primarily used to get the structure of the original image and filter out of noise; therefore, we adopt a larger scale of smooth filter. Then, divide every pixel of original image and smoothing image to get the detail of the image. Since the result of the division method will show the ratio of the pixels, the image can show more correct details.

In Formula 3, I(x,y) represents every pixel of the original image, while (x y) represents every pixel of the fuzzy image. Then, divide the two images‘ pixels and put the result back to t t (x y). After normalization, the result will back to 0 to 255 and produce a new image. Later on, Gamma correction can raise the illumination and the identification of the image, Then 698

we use Gamma Correction to adjust the illumination to finish the whole illumination normalization, like Formula 4.

In Formula 4, we adjust the range of pixel value and make the edges of face clear, it will improve the effects of feature extraction, the image after this process is shown at Picture 4.Through the two steps, the influence of changing illumination can be reduced. Furthermore, different from the former algorithm, this process tends to filter out the problem of maxima and minima, and make the image quality and later feature selecting more stable.

4. Feature Extension A. Multi-scale Block Local Binary Patterns, (MB-LBP) Local Binary Pattern is used to calculate single pixel; therefore, it belongs to texture descriptor of local subtle feature. If the image content belongs to large scale object, picking up subtle feature couldn‘t pick up texture features effectively; face image is an example. Therefore, Multi-scale Block Local Binary Patterns (MB-LBP) [4] [5] is one of the methods to solve this problem. This algorithm is different from primitive LBP, which deals with single pixel; instead, it focuses on areas and nearby areas to be the basis of judgment. The calculation way is to compare big or small of the average pixel value of middle area ( ) and the average pixel value of nearby area ( ). Areas that are larger or equal each other are set to be 1, while smaller set to be 0. The 8 bites binary value got from the same way is used to be the feature of this area. We use the size of the mask m×m to define the operator of MB-LBP, and take 9×9 MB-LBP as an example, like we can see in Picture 5.

699

B. Improve Completed Local Binary Patterns,(CLBP) CLBP [6] is a enhanced textural descriptor from Local Binary Patterns, it is proposed to improve the defect of LBP about not describing the global-compared texture, so it used 3 formulas to describe the texture of one image, like Formula 5~7. In these formulas, Formula 5 is traditional LBP, Formula 6 is to describe the magnitude LBP texture, and Formula 7 is binarization image with mean pixel value of one image.

To improve this method, we proposed a new method based on CLBP, combining Gradient and multi-scale block LBP to make the feature more detail and diversify the type of feature in one place, to enhance the discrimination of the feature, Formula 8~9 shows the improved CLBP descriptor, In these formulas, Formula 8 is used to replace Formula 5, it means we replaced traditional LBP with MBLBP(9X9), to make the texture more stable. In Formula 9, we use Histogram of Oriented Gradient [7]to replace Formula 7, because Formula 7 seems meaningless in calculating the similarity score, and we can use the difference of local gradient to make the features has the meaning of gradient, improving recognition ability.

700

5. Feature Matching A. Eye-Nose Scale Comparison After finishing feature extraction, we can use the features from faces we get to calculate the similarity score, but when we meet the situation that there are many samples to match, it will take many times to finish the calculating. To solve this problem, we proposed an method based on the distance from eyes and distance from nose to the middle point of left eye and right eye. Like Formula 10, we compare the scale value of eye-nose, and to evaluate whether the scale is similar to one sample face, St is the scale of input face, and Ss is the scale of sample face. Because the scale will not be infected with the changing of face image size, so we can use to judge whether this face is the user we want to recognize at first, to decrease the calculating time and improve the efficacy. Picture 5 shows an example of Eye-Nose Scale comparison

B. Chi-Square Statistic After the last step, we can choose he candidate faces that might be the same owner of input face image, so we are going to calculate the similarity score, then we can get the final recognition result. In this step , we use Chi-Square Statistic[8] as our formula to get the similarity score, because it can get evident score when calculating the difference of two histogram data. Formula 11 shows the Chi-Square Statistic formula, first we arrange the CLBP_S and CLBP_M feature into a histogram with 58 bins, then we use Formula 11 to calculate the similarity score. When we get a higher score, it means that these two data are more difference from each other, so we can know that this two face is less similarity, and vice versa. The advantage of using Chi-Square Statistic instead of Euclidean distance is that the score is stable, so we can set a stable threshold to evaluate whether these two face is the same person.

C. Using Gradient to adjust the Similarity Score Though using Chi-Square Statistic to calculate the similarity score can get good efficacy of evaluating the recognition result, it may lost some feature when using CLBP_M and CLBP_S, these two feature descriptor based on local binary pattern. 701

Therefore, we have to combine more features which have different meaning from local binary patterns. After research, we found that feature descriptors based on local binary patterns is less of the information of gradient. In the other hand, if we can get the gradient information of face image, we can complement the meanings of face features. To implement this idea, we first propose the new CLBP_C, replacing the original CLBP_C with Histogram of Oriented Gradient. Then we propose an method to combine the gradient information with CLBP_S and CLBP_M. In our method, we participate one face image into 48 regions, and arrange the feature as histogram separately, when calculating the similarity score, we compare these pieces in the same place of two face images, and add these 48 scores into one scores. This method can increase the accuracy of Chi-Square Statistic calculating. And we also collect the gradient separately with the same 48 regions, and choose the most gradient in one pieces on behalf of this region, this like Picture 6. When calculating the similarity score of one pieces, we also compare the gradient of the same region, to evaluate the different of gradient from two regions in the same place of two faces. Formula 12 shows the compare method, Mi means the gradient number on the ith region of input face image, and Ni means the gradient on the ith region of the sample face image. When we get the value of , we multiple it with the score of this region as an punishment, and use this method to increase the accuracy rate of our face recognition algorithm. Picture 7 gives an example of this method.

6. Experiments In this research paper, we use Faces94, FERET, YALE and self-constructed face database to conduct experiments. The experimental environment of the recognition rate is to pick an image from everyone‘s resources to be training set and the rest of them are test set. Every picture is used to perform recognition test with all registered users to calculate the rate of recognition and 702

adopt various kinds of ways to conduct the comparison of recognition. Picture 8-10 shows some samples of FERET, YALE and Faces94 face database.

Picture 8: Some samples of FERET

The calculation of false acceptance rate is to register a person, and conduct recognition to registered person using others‘ images. If the recognition is successful, it means that the system is invaded. Supposed that the totally number of people is n and the number of respective images is m, the total process times is × ( 1) × m. The calculation of false rejection rate is to register a person, and conduct recognition to him/her using the rest pictures. (Other people are not included) If the recognition is failed, it means to be rejected. Supposed that the totally number of people is n and the number of respective images is m, the total process times is (m 1) × .

703

7. Conclusion In this research paper, we successfully build a face recognition guarded system through the combination of mobile device and Arduino. In the algorithm, we adopt multi-scaled methods of feature selection, which can effective avoid the problem that when the face image was partial damaged, the rate of recognition will decrease. To improve the problem that the process speed is slower in mobile device, we first improve the algorithm of illumination normalization in pretreatment and it accelerates the calculation of the original algorithm. Furthermore, it can filter out the light changing more effectively to decrease the noise. As to the calculation of similarity, we combine MBLBP, CLBP and Histogram of Oriented Gradient. It not only improves the process that needs information discretion when conducting recognition, but also make algorithm consider the distribution of histograms and spatial distribution

704

simultaneously. At last, we proposed the Eye-Nose Scale comparison method to speed up the calculating time and avoid unnecessary process, increase the efficiency of the algorithm. 8. References P. Viola and M. Jones, ―Rapid Object Detection Using a Boosted Cascade of Simple Features‖, Computer Vision and Patterns Recognition, IEEE Computer Society Conference, vol. 1, pp. I-511-518, 2001. Quanxue Gao, Feifei Gao, Hailin Zhang, Xiu-Juan Hao, and Xiaogang Wang , ―Two-Dimensional Maximum Local Variation Based on Image Euclidean Distance for Face Recognition‖, IEEE Transactions on Image Processing, Vol.22, Issue 10, pp.3807-3817, 2013. Mohammad Reza Faraji and Xiaojun Qi, ―Face recognition under varying illumination based on adaptive homomorphic eight local directional patterns ‖, IET Computer Vision, Vol.9 , Issue 3, pp.390-399, 2015. MB-LBP S. Liao, X. Zhu, Z. Lei, L. Zhang, and S. Z. Li, ―Learning Multi-scale Block Local Binary Patterns for Face Recognition‖, ICB2007, LNCS 4642, pp. 828-837, 2007. 施佑駿, Face Recognition System Based on Local Regions and Multi-resolution Analysis”, 國立台灣科技大學資訊工程系, 2010. Zhenhua Guo, Lei Zhang, and David Zhang ,‖ A Completed Modeling of Local Binary Pattern Operator for Texture Classification‖, IEEE Transactions on Image Processing, Vol.19, Issue 6, pp.1657-1663, 2010. Ching-Yao Su and Jar-Ferr Yang, ―Histogram of gradient phases: a new local descriptor for face recognition‖, IET Computer Vision, Vol. 8, Issue 6, pp.556-567, 2014. Donagh Horgan and Colin C. Murphy ,‖ On the Convergence of the Chi Square and Noncentral Chi Square Distributions to the Normal Distribution‖, IEEE Communications Letters, Vol. 17, Issue 12, pp.2233-2236, 2014.

705

ACEAIT-7556 Zone-Based LED Signal Detection for Smartphone Visible Light Communications Jae Hyuc Jang, Ho Hyeong Jo, Hyeon Ho Kim, Yong Hyeon Kim and Yeon Ho Chung* Department of Information and Communications Engineering, Pukyong National University, Republic of Korea E-mail address: [email protected] 1. Introduction Information and communication technology has rapidly grown due to heavy demand on short range wireless communications. The solutions to this demand have mostly been based on radio frequency (RF). As smartphones become a popular communication device, attractive technologies have evolved using the smartphones. This paper considers a smartphone based data reception from the LEDs, using zone detection (ZD) method together with a camera‘s exposure lock (EL) feature. The experimental demonstration shows that the proposed method is efficient in detecting transmitted signals on the basis of LED colors. 2. Proposed scheme and experiments In the proposed scheme, the transmission is carried out using RGB LEDs. At the receiver side, an image sensor (camera) of smart phone is used for receiving the image data. In this smartphone VLC, attention must be paid to the following two issues. First, when we communicate using LEDs and a camera, we need to consider the capture rate of the camera, which is frames per second (fps). Since this rate is relatively very low, this will limit the blinking rate of the LED for transmission. Second, the widely implemented white LEDs for transmission are easy to implement with low complexity. This white color is, however, somewhat limited for increasing the data rate, since it beams a single color only. In the proposed scheme, RGB LEDs are applied for improving the data rate at the existing frame rate. With the RGB LEDs, we generate five combinations of color, i.e., red, green, blue, white and black. Note that the white color can be generated by turning RGB LEDs on simultaneously, whereas the black color can be obtained by turning off the RGB LEDs. In this way, the data rate is increased at least four times compared to a single white LED based scheme. Figure 1 shows the division of the entire frame into four equal zones and the LED signal detection using the ZD and EL. Each zone is divided again into nine equal smaller segments. The segments are made to compensate for camera shakes caused by hands. The ZD chooses the middle segment for each zone to represent the intended LED color for data transmission, while the other segments are ignored. The ZD compensates in a way that it selects the middle segment containing the intended color only as an exposure reference. When the LED blinks, it will automatically adjust the exposure value in the smartphone camera applications. Thus, the whole capture process uses the same exposure value.

706

In connection with the signal detection using smartphone cameras, it often encounters overexposure as shown in Fig. 1(a). In order to solve this problem, we employ the EL method, namely, the exposure value is fixed to an exposure reference value using the ZD. Figure 1 (a) shows an overexposed state, when the exposure value becomes too high after the LEDs are turned on. In the experiment, a smartphone was used as the receiver and the RGB color LEDs were employed as the transmitter. The communication distance between the transmitter and the receiver was set to 10 cm. The four blinking RGB LEDs transmit the data in four colors. Each LED was located in the center of each quadrant of the entire transmission zone. The LED color was designed to change, according to a certain cycle. The smartphone used in the experiment was Samsung SM-G850. The experiment was also conducted for compensating the shakes caused by hand tremors and shadowing. Figure 1(b) shows the image when the EL is applied. It can be seen that the original intended LED color is clearly observed. 3. Conclusion A smartphone based visible light communication using RGB LEDs was considered. To increase the data rate, the zone allocation method together with the exposure lock feature was applied. The experiments show that the proposed methods improve the data rate and the LED signal detectability. Acknowledgements: This research is funded by the Research Grant of BB (Brain Busan) 21 project of 2016. Keywords: Visible light communication (VLC), smartphone visible light communication, optical camera communication.

707

Chemical Sciences

Thursday, March 31, 2016

13:10-14:40

Room C

Session Chair: Prof. Hyoung Jin Choi ISFAS-1487 Fabrication and Suspension Characteristics of Silica Coated Soft-Magnetic Composite Particles Hyoung Jin Choi︱Inha University ISFAS-1511 Implementation of Flipped Classroom Methods to Build up a Student‟s Chemistry Background Li Ling Apple Koh︱Singapore University of Technology and Design Patricia D. Christie︱Massachusetts Institute of Technology Mei Chee Tan︱Singapore University of Technology and Design Gim-Yang Pee︱Singapore University of Technology and Design Aikaterini Bagiati︱Massachusetts Institute of Technology John G. Brisson︱Massachusetts Institute of Technology ISFAS-1548 Graphene/Calixarene-Glucamine Modified Magnetic Nanoparticles Adsorbent for Chlorinated Pesticides Removal from Aqueous Media Wan Aini Wan Ibrahim︱Universiti Teknologi Malaysia Hamid Rashidi Nodeh︱Universiti Teknologi Malaysia Muhammad Afzal Kamboh︱Universiti Teknologi Malaysia Mohd Marsin Sanagi︱Universiti Teknologi Malaysia

708

ISFAS-1590 Characterization of CoMo Catalysts Supported on Mesoporous Carbon from Cow‟s Bone Gelatin for Hydrocracking of Waste Lubricating Oils Marthinus Pongsendana︱Universitas Gadjah Mada Farin Windy Artanti︱Universitas Gadjah Mada Wega Trisunaryanti︱Universitas Gadjah Mada Iip Izul Falah︱Universitas Gadjah Mada

709

ISFAS-1487 Fabrication and Suspension Characteristics of Silica Coated Soft-Magnetic Composite Particles Hyoung Jin Choi Department of Polymer Science and Engineering, Inha University, Korea Email: [email protected] Background Micron-sized spherical soft-magnetic carbonyl iron (CI) particles have been extensively adopted for various industrial applications including magnetorheological (MR) materials. The MR fluids are in general a kind of smart and intelligent materials with controllable and reversible phase transition behavior from a liquid-like to a solid-like state by an external magnetic field strength applied [1], analogous to electrorheological fluids under an applied electric field [2] [3]. Nonetheless they suffer from chemical corrosion for their long-term usage as dispersed particles for MR fluids. To improve their anti-acid and anti-oxidation resistance as for MR fluids, they are coated with silica layer via a sol-gel fabrication, and then applied for magneto-responsive MR fluids. Methods Both silica and multi-walled carbon nanotube (MWNT) dual-coated soft-magnetic CI particles were fabricated. Initially, once the CI particles were firstly grafted with vinyltrimethoxysilane, they were dispersed in a mixed solution of tetraethyl orthosilicate and ethanol with high speed stirring for the purpose of uniform coating of silica shells. Following Stöber method, an ammonia solution was added into the particle suspension. On the otherhand, for the MWNT dual-coated soft-magnetic CI particles, the first layer of silica was coated onto the silane-grafted CI microspheres by a sol-gel method, and then the second layer of MWNT was sequentially wrapped by a layer-by-layer process using both positive and negative polyelectrolytes. The morphologies of the pristine and silica coated CI particles were observed by SEM. To observe the anti-acid ability of the coated sample, same amount of the two samples were steeped in diluted HCl solution separately and the pH value of the solutions were recorded according to time. Each MR fluid was prepared by dispersing pristine, silica-coated, and MWNT dual-coated CI particles in silicone oil. MR performance of these MR fluids including flow curve, yield stress and viscoelastic characterisitcs were measured by a rotational rheometer, which was equipped with a magnetic field supplier and a parallel plate measuring system. Results At first, the anti-acid corrosion and anti-oxidation properties of both pristine and silica-coated CI particles were measured and compared. The pH value of the pure CI contained solution was observed to be increased rapidly due to the reaction between the CI powder and the acid. However, the silica coated CI based solution keeps a stable pH value, demonstrating that its silica coating improved the anti-acid corrosion and anti-oxidation properties. In addition, through the observation of MR performance in the carrier fluid of silicone oil, silica coated CI particle-based MR fluid showed lower yield shear stress at magnetic field but also lower

710

off-state shear viscosity as compared to those of pure CI based MR fluid due to a better compatibility between a silicone oil and silica-coated CI particles. On the other hand, as for dual-coated magnetic CI particles by silica and MWNT, the coated particles with a reduced density exhibited solid-like properties but lower storage and loss moduli than those exhibited by the raw CI particles measured from both amplitude and frequency sweep. The relaxation modulus was also obtained using the Schwarzl equation, exhibiting stable plateau behavior. Flow curves and yield stresses of the MR fluid were analyzed using our Cho-Choi-Jhon constitutive model and universal yield stress equation, respectively, which all showed good fitting results on the experimental data. Keywords: Carbonyl iron, magnetorheological, silica coating, viscosity, suspension References Ulicny JC, Snavely KS, Golden MA, Klingenberg DJ. Enhancing magnetorheology with nonmagnetizable aprticles. Appl Phys Lett. 2010:96: 231903. Park BJ, Fang FF, Choi HJ. Mangetorheology: materials and application. Soft Matter. 2010:6: 5246-53. Liu YD, Fang FF, Choi HJ. Core–shell-structured silica-coated magnetic carbonyl iron microbead and its magnetorheology with anti-acidic characteristics. Colloid Polym Sci. 2011:289:1295-8. Liu YD, Choi HJ. Magnetorhoelogy of core-shell typed dual-coated carbonyl iron particle fabricated by a sol-gel and self-assembly process. Mater Res Bull. 2015:69:92-7. Sedlacik M, Pavlinek V, Lehocky M, Mracek A, Grulich O, Svrcinova P, Filip P, Vesel A. Plasma-treated carbonyl iron particles as a dispersed phase in magnetorheological fluids. Colloid Surf A-Physicochem Eng Asp. 2011:387:99-103.

711

ISFAS-1511 Implementation of Flipped Classroom Methods to Build up a Student‟s Chemistry Background Li Ling Apple Koh Engineering Product Development, Singapore University of Technology and Design, Singapore [email protected] Patricia D. Christie Experimental Study Group, Massachusetts Institute of Technology [email protected] Mei Chee Tan Engineering Product Development, Singapore University of Technology and Design, Singapore [email protected] Gim-Yang Pee Engineering Product Development, Singapore University of Technology and Design, Singapore [email protected] Aikaterini Bagiati Office of Digital Learning, Massachusetts Institute of Technology [email protected] John G. Brisson Department of Mechanical Engineering, Massachusetts Institute of Technology [email protected] 1. Background/Objectives and Goals At Singapore University of Technology and Design (SUTD), it is observed that students lack certain basic concepts and fundamentals in general Chemistry. As Chemistry is not a prerequisite subject in order to be admitted into SUTD, students enrol with different levels of Chemistry knowledge. In addition, Chemistry is one of the required subjects which must be taken in the first year of the degree program. As such, a proportion of students find studying Chemistry at SUTD challenging and lack the confidence in learning and studying Chemistry. To help students, educational video lessons on selected Chemistry topics were produced and made available online. These online lessons, modelled after the flipped classroom concept, provide the students the groundwork for the concepts to be taught later in the classroom in addition to improving their Chemistry knowledge [1, 2]. Here, we aim to discuss about the implementation of these online video lessons and the learning experience from the student‘s perspective. 2. Methods An educational video lesson consists of a series of short videos followed by a quiz after each video. These videos were created using ScreenflowTM (http://www.telestream.net/) which recorded the instructor‘s narration on top of a Powerpoint presentation [3]. Each video (less 712

than 12 min long) illustrates a concept and shows the steps to approaching and solving a problem. Additional online problems and solutions were also made available to the students for extra practice. For every topic that was taught through the use of video lessons, students were given at least one week to complete these video lessons before attending a class where the emphasis was on real world examples and applications and, in-depth problems. This flipped classroom method was first incorporated into the subject in Summer 2014 and developed for Lewis dot diagrams to provide the necessary knowledge and background essential for Valence Shell Electron Pair Repulsion (VSEPR), which was later taught in class. At the start of the VSEPR lesson, students completed a quiz on Lewis dot diagrams before moving on to the actual lesson on VSEPR. This quiz acts as a feedback to how well the students understood Lewis dot diagrams by watching the videos. Students then worked in small groups and built the shapes of simple molecules with no lone pairs using Playdoh TM and toothpicks. In the second part of this active-learning lesson, students had to draw the Lewis dot diagrams of various chemicals and determine the shape of these molecules by modifying the basic PlaydohTM structure created previously. Looking at the built PlaydohTM structures, students had to name the shape of the molecules with lone pairs using concepts learnt in VSEPR. To determine the effectiveness of such video lessons in this Chemistry course, students were encouraged to participate in a survey in the form of an interview. This was approved by SUTD IRB (#14-037) and MIT IRB (#1403006273) [3]. 3. Expected Results/Conclusion/Contribution After two years of running these educational video lessons in the Summer of 2014 and 2015, a total of fourteen students participated in the exit interviews. Preliminary results showed that the use of video lessons was helpful. To students with ‗A‘ Level Chemistry background, the videos serve as a good method in reviewing a concept and practicing worked problems outside of class time [3]. Students also enjoyed the ability to pause, rewind and replay the videos, allowing them to work at a pace comfortable to their learning habits. However, a common comment across all students was the reliability of the e-learning platform and the amount of time students have to spend outside class on video lessons. Specifically to the Lewis dot diagram and VSEPR flipped classroom lesson, a couple of students enjoyed the gradual build-up of concepts, which was done in a step-by-step manner in the videos, but they would like more worked examples and detailed explanations. More importantly, most students managed to make a connection between the videos and the classroom activity. The use of PlaydohTM allowed students with little Chemistry background to better see and understand electronic repulsion and name the shapes of the molecules. However, it was also found to be distracting as a few groups in the class were playing with the Playdoh TM instead of building models. A student hence suggested the use of molecular kits instead. From the students‘ feedback in the first run, additional videos, covering the concepts of stoichiometry, discovery of atoms and solutions making, were created for the Summer 2015 class. In general, the videos were well-received but it was also shown that the student-instructor interaction should not be compromised as students preferred immediate feedback from the instructors. These videos also free up class time for activities and problem solving, which is necessary for students to master the material. More importantly, the use of 713

videos on basic Chemistry concepts provides all students with the opportunity to achieve a common background before coming to class. 4. References [1] Leontyev, A., Baranov, D. (2013), Massive Open Online Courses in Chemistry: A Comparative Overview of Platforms and Features, Journal of Chemical Education, 90, 1533-1539. [2] Arnaud, C.H. (2013), Flipping Chemistry Classrooms, Chemical and Engineering News, 91, 41-43. [3] Christie, P.D., Tan, M.C., Pee, G.-y., Koh, L.L.A., Bagiati, A., Brisson, J.G. (2015), Using On-line Video Lessons to Teach Concepts in Chemistry, International Scientific Conference on Engineering and Applied Sciences, Okinawa, Japan, 29 to 31 July. Keywords: flipped classroom, education, Chemistry

714

ISFAS-1548 Graphene/Calixarene-Glucamine Modified Magnetic Nanoparticles Adsorbent for Chlorinated Pesticides Removal from Aqueous Media Wan Aini Wan Ibrahim, Hamid Rashidi Nodeh, Muhammad Afzal Kambohc, Mohd Marsin Sanagid Department of Chemistry, Faculty of Science, Universiti Teknologi Malaysia, Malaysia [email protected]; [email protected] 1. Background/ Objectives and Goals Sample preparation is an important step in analysis and is the most time consuming step in the analytical process. Recently magnetic solid phase extraction (MSPE) has received much attention as the process is simple yet effective. MSPE is based on the extraction of different compound from the sample using solids with magnetic properties. MSPE avoids steps such as centrifugation and filtration which decreases the manipulation of the sample. Many magnetic materials have been used in MSPE and most are based on Fe. This study focus on the synthesis, characterization and application of a new adsorbent derived from Fe3O4 magnetic nanoparticles (MNPs), modified with graphene (G) and calixarene-glucamine (Calix-Glu), abbreviated as Fe3O4/G-Calix-Glu, for the carcinogenic pesticides removal of hexaconazole and chlorpyrifos from tap water and industrial wastewater samples. 2. Methods This technique consists in the synthesis of particles with magnetic properties (generally Fe3O4) followed by modification of the magnetic phase with diverse organic compounds. The modified solids are applied as adsorbents during isolation, separation, and preconcentration or removal of target analytes form matrix of interest. In our current work, the newly synthesised magnetic material, Fe3O4/G-Calix-Glu is used for the removal of two chlorinated pesticides from aqueous media. The MSPE is based on the dispersion of a magnetic adsorbent in solution. In MSPE process, a fixed mass of the adsorbent (mads) is dispersed in the sample solution of fixed volume (Vs) for a fixed contact time, t for removal process. Pesticides removal efficiency of the MNPs/G-Calix-Glu was analyzed through batch-wise adsorption prior to gas chromatography analysis using micro electron capture detector (GC-µECD). The adsorption affecting MSPE parameters i.e., adsorbent dosage, sample pH, salt effect, pesticides concentration and contact time were studied separately. 3. Expected Results/ Conclusion/ Contribution Adsorption equilibrium study of the newly synthesised MNPs/G-Calix-Glu adsorbent revealed high pesticides removal efficiency (> 80% for waste water) at pH = 5.5 and contact time of 10 min. Kinetic studies showed that the fast pesticides adsorption observed followed the pseudo-second-order kinetic model (R2 = 0.99). Among the Langmuir, Freundlich and Dubinin-Radushkevich isotherm models studied, the Langmuir isotherm with high values of R2 was found to fit well with the results and showed that the two pesticides adsorption on the newly synthesised MNPs/G-Calix-Glu followed the monolayer adsorption pattern. Pesticides contaminated water, industrial waste water and tap water samples were used to assess the field applicability of the MNPs/G-Calix-Glu. The results obtained demonstrated that the newly synthesised MNPs/G-Calix-Glu adsorbent can be employed as an efficient, environmental friendly and highly reusable sorbent for removal of the two selected chlorinated pesticides from aqueous media. 715

Keywords: Magnetic nanoparticles, Graphene-based adsorbent, Magnetic solid phase extraction, chlorinated pesticides

716

ISFAS-1590 Characterization of CoMo Catalysts Supported on Mesoporous Carbon from Cow‟s Bone Gelatin for Hydrocracking of Waste Lubricating Oils Marthinus Pongsendana*, Farin Windy Artanti, Wega Trisunaryanti, Iip Izul Falah Department of Chemistry, Universitas Gadjah Mada, Bulaksumur, Indonesia E-mail address: [email protected] Abstract Fossil fuels is finite and getting decrease over time due to the world consumption in modern society, whereas waste lubricating oils from various application has dramatically increased. Mesoporous carbon made from cow‘s bone gelatin (MCG) has high thermal stability (up to 1400 oC), specific surface areas and pore volumes. These special characteristic make it a promising material to be applied as supporting catalysts for CoMo catalysts. The objectives of this study were to investigate the interactions and synergetic effects between Co and Mo at different ratio content in the catalyst. The activity of these catalysts was studied in hydrocracking of waste lubricating oils into fuel fractions. Mesoporous carbons were synthesized by using gelatin and SBA-15 as carbon source and silica template respectively. In a typical synthesis, gelatin, H2SO4, and H2O were homogenized and added to SBA-15 silica dropwisely. The mixture was first heated in an oven for 6 h at 100 oC, then at 160 oC for 6 h. The step was repeated to produce brown solid. The polimer-template composites were then carbonized in a furnace in an argon gas flow at 900 oC and kept under these condition for 4 h. Finally, the silica templates were removed by treatment with a 10 % HF solution (etanol-water) to produce mesoporous carbon. CoMo bimetallic catalysts supported on MCG were prepared by sequential wetness impregnation technique, where Mo was loaded first then followed by loading of Co. The total number of metal loaded was 2 %, with Co/Mo rasio 3:1 and 1:1. The Mo was loaded first, follow by Co. After each impregnation, the catalysts were then dried in a oven at 120 oC for 12 h. The catalysts were calcined at 500 o C for 3 h in the presence of air, followed by reducing under hydrogen gas stream at 400 oC for 3 h. The number of acid sites of the catalysts was determined gravimetrically using ammonia adsorption. The characteristics of the catalysts were studied by using FTIR, TEM, and BET. Catalytic activity test was carried out in hydrocracking process of waste lubricating oils. The process was conducted in a semi flow microreactor system under the hydrogen gas stream (20 mL/min) for 60 min; reaction temperature 450 oC, feed/catalyst weight ratio = 50 (the feed was a waste lubricant oil). The liquid products were analyzed by Gas Chromatography-Mass Spectrometry. The characterization results showed that the ordered mesoporous carbons were obtained after dehydration, pyrolisis, and silica removal process. Its hexagonal nanorod structures and uniform pores were confirmed by the Transmission Electron Microscopy. The ordered mesoporous carbon has specific surface area 404,471 m2/g, pore diameter 3,051-3,027 nm, and total pore volume 0,76 cc/g. Co and Mo were successfully loaded into the mesoporous carbon. It has confirmed by the FTIR and SEM. Total number of the acid sites in the catalysts after loading of Co and Mo is expected to be increased, whereas the best ratio content of the Co and Mo in the catalyst will be studied base on the total liquid product conversion and its selectivity toward gasoline and diesel fractions.

717

Keywords: hydrocracking, waste lubricating oils, CoMo catalyst, mesoporous carbon, liquid fuel.

718

Geology, Earth and Environmental Sciences (1) & Environmental Engineering (2)

Thursday, March 31, 2016

14:50-16:20

Room A

Session Chair: Prof. Dwi Astiani ACEIAT-7615 Periodic Change of Sea Level and Its Impact on Trace Metal Mobilization in Coastal Aquifers of Saijo City, Japan Pankaj Kumar︱United Nations University Kazuhiko Takeuchi︱United Nations University Srikantha Herath︱United Nations University ISFAS-1518 Carbon Stocks in Recent Land Cover Conditions of Tropical Lowland, Swamp, and Peatland Forests in West Kalimantan Dwi Astiani︱Tanjungpura University Mujiman︱Lembaga Landscape Livelihood Indonesia (LLI) Andjar Rafiastanto︱Fauna and Flora International-Indonesia Programme ISFAS-1523 Economic Value of Tembawang Ecosystem in Sanggau Subdistrict West Kalimantan Province, Indonesia Emi Roslinda︱Tanjungpura University Siti Masitoh Kartikawati︱Tanjungpura University Rabudin︱Tanjungpura University

719

ISFAS-1533 Trail-Following and Odor Effects of Plant Leaf Extracts on Subterranean Termite Coptotermes Formosanus Shiraki Yuliati Indrayani︱Tanjungpura University Musrizal Muin︱Hasanuddin University Cynthia Adilla︱Tanjungpura University Tsuyoshi Yoshimura︱Kyoto University ISFAS-1537 Role of Rhizosphere Microbes in Increasing Plant Growth of Jatropha Curcas in Mercury Contaminated Soil Wiwik Ekyastuti︱Tanjungpura University Hanna Artuti Ekamawanti︱Tanjungpura University

720

ACEAIT-7615 Periodic Change of Sea Level and Its Impact on Trace Metal Mobilization in Coastal Aquifers of Saijo City, Japan Pankaj Kumara*, Kazuhiko Takeuchia, Srikantha Heratha a United Nations University, Institute for the Advanced Study of Sustainability (UNU-IAS), Japan * E-mail: [email protected] Abstract Introduction/Objective Considering the limited knowledge about the dynamic nature of coastal aquifers, this study emphasized on the interaction between groundwater and seawater and its impact on spatio-temporal tidal on trace element mobilization in Saijo City, Japan. Methodology Concentrations for all eight trace elements were determined with inductively coupled plasma-mass spectrometry (ICP-MS) (AGILENT-7500) with a precision of 1 %. Saturation index for different minerals related to these trace elements was calculated using PHREEQC, a USGS numerical code. This was used to evaluate the influence of periodic changes in sea level on eight selected trace elements concentration (As, Fe, Cd, Pb, Zn, Cu, Ni, Cd and Cr). Results Time series analysis of different elements suggests that amplitude of even diurnal tides do have significant effect on trace elements and effect is more prominent for unconfined aquifer. Fluctuation of trace metals was found to be directly proportional to the hydraulic conductivity of the aquifer matrix and screen depth of the monitoring wells while inversely proportional to the amplitude of the tidal waves. The main driving mechanism for trace metal mobilization might be oxidation-reduction process because of the recirculated submarine groundwater discharge. Here periodic tides results in fluctuation in hydraulic pressure from sea towards inland when infiltrate into an aquifer and produce the fluctuation of hydraulic head in a coastal aquifer ultimately changing redox condition. Significant relation between fluctuation pattern of trace elements (analyzed) and oxidation-reduction potential (ORP) (observed) firmly supports and validates the results from chemical analysis. Keywords: Submarine groundwater discharge; Saturation index; Saijo City; Tide aquifer interaction

721

ISFAS-1518 Carbon Stocks in Recent Land Cover Conditions of Tropical Lowland, Swamp, and Peatland Forests in West Kalimantan Dwi Astiania*, Mujimanb, Andjar Rafiastantoc a Fakultas Kehutanan Universitas Tanjungpura, b Lembaga Landscape Livelihood Indonesia (LLI), Pontianak, Indonesia c Fauna and Flora International-Indonesia Programme, Jakarta, Indonesia Email: [email protected] Abstract Tropical forests constitute for a large concentrated carbon pools, ultimately in tropical peatland forest since this forest type sink carbon both in the vegetation and its underlying peat. However, these forests recently experienced a lot of pressures from anthropogenic disturbances. A study was conducted to estimate carbon stocks of degraded tropical, swamp, and peatland forests in Kayong Utara West Kalimantan. Above ground survey was conducted using stratified sampling based on the differences in spectra of Landsat 7 ETM+ according to the land cover gradation or vegetation formations. The study area was classified based on the canopy closures and land cover types and grouped into low and high degraded lowland forest, low and high degraded peat forest, low and high degraded swamp forest, shrub land, and mixed agricultural land. Aboveground carbon in each group was estimated by purposively measuring trees diameter >5cm within 5-11 plots of 20m x 100m area. Results show that among the groups, the highest to the lowest order of aboveground carbon sink consecutively are low degraded swamp, low degraded lowland, low degraded peatland, high degraded swamp, high degraded peatland, high degraded lowland forests, mixed agricultural land, and shrub land (255.9 to 36.9 ton C/ha) plus other biomass recruited from necromass, litter, and ground storey vegetation. It is demonstrated that forest degradation and land cover changes reduce amount of above ground carbon stocks and thus could result large amount of carbon loss from forests. Surprisingly, our results demonstrated that 0.5-5.2m belowground carbon contribute to large amount of carbon Each meter depth sinks ~634 ton C/ha. It is estimated that the 22,600 ha area of variety of land covers sink 2.5 million of aboveground and 5,570 ha peatland area hold ~9.2 million of below ground carbon. This amount of carbon could be potential sources of carbon losses if land cover changes continued. Keywords: aboveground carbon, belowground carbon stocks, degraded forest, land cover change, peatland Introduction Tropical forests provide an enormous roles to environment ecosystem not only to local people but also to global population such as habitat biodiversity, hydrology regulator, and carbon storage and the ecosystem recently deserve special attention because of continuing considerable destruction 1. The deforestation rates of intact forest in Southeast Asian tropical peatlands – concentrated in Sumatra and Kalimantan, Indonesia – has been reported as 3.4% y-1 from 1990 – 2010 2, 3. Similar to the global condition, tropical forests in Indonesia are in a lot of pressure mainly due to anthropogenic activities such as logging causing forest disturbances and degradation 4 . Forest degradation could bring the forests into a temporary or might be permanent destruction not only in forest vegetation density and structure, but also in species composition, and the condition could lead to reduce their productivities 5, 6. Since tropical forests play a major role in 722

regulating global carbon, fluxes and stocks-especially in peatland- 7, 8, 9, 10, even a very small alteration to carbon balance in a biome, could have a significant effect on atmospheric greenhouse gasses especially carbon dioxide 11, 12, 13, 14 . Globally, however, deforestation and forest degradation is threatening forest CO2 sequestering function of forest 15, 16, and deforestation and degradation account for approximately 12% of global GHG emission 17. To improve estimates of forest role in maintaining carbon stocks of recent West Kalimantan forest condition, our study was designed to estimate current carbon stocks of variably land covers accupying a large landscape of mixed forest types and land covers and concurrently with belowground carbon stocks of peatland in Kayong Utara in West Kalimantan. Material and Method Site Selection The areas of carbon assessment located in between 1100 10' 0‖ E to 1100 22' 80‖ E and 00 65' 0‖ S to 10 65' 0‖ S in S Matan, Kabupaten Kayong Utara, West Kalimantan, extending ~ 22,600 ha within PT Citra Usaha Lestari (CUS) oilpalm concession, covered with mixed forest type, oilpalm area. shrub land and open area. Below ground carbon assessment was located in the middle northern block and Middle East corner block of concession area which covered by peatland forest (5,576 ha). This area was chosen based on the indication of peat land area derived from satellite image interpretation and soil map. Above ground carbon assessment was distributed throughout the area based on the stratified land cover/forest types, where below ground carbon assessment was focused in peatland area. Sampling Approaches for Collecting Data Above ground survey was conducted using stratified sampling based on the differences in spectra interpretation of Landsat Image according to the land cover or forest types. The area was classified and refined based on land cover types in the area: high density lowland (HDLF), low density lowland (LDLF), high density peatland (HDPF), low density peatland forest (LDLF), high density swamp forest (HDSF), low density swamp forest (LDSF), and also shrub lands (SH) and mixed agriculture lands (AG). Within each land cover stratification, we purposively measured 9-12 plots of a 20m x 100m size for aboveground carbon estimates and, under peatland area, we also assessed below carbon stocks. In each plot we measured 5 carbon pools: tree diameter >20cm within 20m x 100m area; tree diameter >5cm>20cm within 5m x 40m nested plot; necromass (standing woody debris, CWD, stumps) within nested plot of 5 x 40 m; ground storey 4-6 1m x 1m nested plots; and litter, 4-6 0.5m x 0.5m nested plots. Except for ground storey vegetation and litter sampling which using destructive technique, all sampling measurement on forest vegetation were using non destructive methods. To estimate tree and necromass biomass, we used allometric equations generated by Chave, Brown, and Puhlin 18, 19, 20 . Based on peat area indication maps, we produced a sketch map describing planned transects to be measured. Within each transects we took measurement using systematic sampling and the points sampled were applied to estimate peat biomass. We measured 85 peat profile and depth and 12 peat cores were sampled to represent peat characteristics and brought to laboratories for further analysis. Throughout each plot, we measured tree diameters, tree heights, and identified tree species names (for tree wood density identification). For Necromass (CWD), we measured log's diameter, length, and level of decomposition (1, 2, and 3). Then, necromass volume will be quantified by cylinder equation. We brought 5 of each class wood samples to classified wood density of class 1, 2, and 3 and their dry bulk masses at laboratory. Necromass' biomass is volume x wood density. We also sampled leaf litters and understorey 723

vegetation and took about 300 g samples of each measurement for oven dry weight and C content at laboratory. For assessing below ground peat assessment, peat area indication was derived from overlaying PT CUS area, soil and vegetation maps. Based on the indication maps, a sketch map describing planned transects to be measured were planned and produced. Along main transect that approximately crossing indicated peat dome, we designed East-West or North-South orientation branch transects which were perpendicular to the main transect within 2 km distance, so called 'fish bone' transects. Within each transects we took measurement using systematic sampling and the points sampled were applied to estimate peat biomass. For this study, we measured peat depth systimatically within 500m distances. Peat soil core samples were taken systematically on each 2 km distance for further laboratory analysis (pH (H2O), bulk density, water and ash contents, and Organic Carbon). We measured 85 peat profile and depth and 12 peat cores were sampled to represent peat characteristics and brought to laboratories for further analysis.

Results and Discussion Above Ground Biomass and Carbon stocks The distribution of biomass measured on each carbon pool for each land cover type is presented in Fig 2.

Biomass distribution of each plot was dominated mostly by tree biomass (especially diameter >20cm), with exception in agriculture and shrub areas, where less or none of living tree found. Necromass (CWDs and litters) significantly determined biomass stocks in agriculture and 724

shrub areas. Estimated carbon stocks is using default carbon value of 50% of biomass (Brown, 1997). Higher carbon stocks were in High Density Swamp Forest, High Density Lowland Forest, and High Density Peat Swamp Forest (219, 207, and 193 ton C/ha consecutively) and the least carbon stocks is found in Shrub Land areas (31 t C/ha). Most of living biomass of the eight landcover/forest classes are potentially hold significant percentage of carbon (Mean ± SE = 90% ± 1.6%). In general about 87% ±1.3% of carbon stocks are sequestered in living trees (diameter 5 cm and above). Forest degradation is reducing carbon stock of each forest type. Lowland, peatland, and swamp forest reduced their stand biomass by 58%, 31%, and 26% consecutively. The biomass quantity losses approximately depend on degrees of forest degradation. Forest land use changes to oilpalm (3 years old) and shrub land decreased aboveground biomass by ~75-86%. On agriculture and shrub land, however, the highest proportion of carbon are stocked in CWD pools (above 85 %), where the stocks are potentially decomposed and be emitted to the atmosphere. Roots carbon stocks estimate approaches In addition to aboveground biomass, we estimate root carbon stock by presenting above ground biomass data using allometric equation following 19 The approaches resulted that there are additionally 17% to 20% of carbon stocks are sequestered below ground in the vegetation roots.

Total carbon stocks (both above ground and root stocks) in this area are much more higher than estimates of forests in tropical Asia (151 t C/ha), except for Low Density Lowland Forest, Agriculture and Shrubs land and also in the range by 21 estimates of carbon stocks in tropical equatorial forest (182-225 t C/ha) 22. These measures are reliable since we have stratified the mixed forests into smaller similar forests or land cover conditions. Even though agriculture and shrubs areas have less total carbon, yet they also contribute to carbon sequestration and cycling. To make projection of the carbon stocks on a larger landscape scale of the area, we present the estimated carbon for overall coverage area by extrapolating carbon stock per unit are to the overall area of each land cover type by GIS map interpretation. The estimate is presented in Table 3.

725

Belowground Carbon Stocks Peat area was vary with depth from 0 to 5m and mostly classed as fibrist peat. Peatlands are commonly interspersed in small hill of lowland area, which sometimes found abrupt. This terrain condition is especially around the perimeter of peat land in the southern of assessment area. Deeper peat depths ranges found in northern area of the concession landscape. Upon laboratory determination of peat bulk density of peat samples, the mean analysis results of bulk density and mean peat carbon per 20 cm depths samples is presented in Fig. 3a and 3b.

Spatial vertical distribution of peat bulk density at the landscape was relative constant which did not vary at 20 cm ranges, or even at 100 cm ranges of the peat depth, except for the last 60-80 cm at the peat bottom, where peat material mixed with gley humus and mineral soil, that 726

could increase the bulk density. The mean bulk density at this area was 0.11 ± 0.012 g/cm3 and the mean of 60cm of peat layers at peat bottom is 0.37 g/cm3. The distribution of vertical bulk density will dictate the carbon content in each range of depth 24, 25 . Mean peat carbon per 20 cm depth range and their accumulation along the depth are presented in Fig 3b. Result on the peat carbon stocks analysis show that within each 20 cm depth, the mean carbon stocks is 105 t/ha or within each 1 m depth, the carbon stock mean is 625 t/ha. The 5.2 m depth peat was estimated to keep ~3,250 t carbon/ha (Fig. 4). Peat landscape area of 5,570 ha estimated contain 9.2 million ton of carbon.

Conclusions Among forest types, above ground carbon stocks on tropical forest type in an extend of a landscape indicates that each forest type sequestered various carbon stocks from the lowest of 37 t/ha (shrub land) to the highest of 256 t/ha . Higher degraded forest of the 3 forest types maintain various carbon stock quantities. the important point is forest degradation reduced the role of tropical forest to maintain carbon stocks in their stand biomass per unit area by 58%, 31%, and 26% respectively for Lowland, peatland, and swamp forests. Agricultural land ( 1-3 year old) sequestered an amount equal to shrubs lands (3 year old = 48 t/ha carbon). Mean annual growth level add about 14 -18 ton/ha carbon to the oil palm plants, while shrub land maintain 37 ton/ha. Compared to forests, these land covers are a lot less in holding biomass stocks. Projected estimate of total aboveground carbon within overall landscape was about 2.5 x 106 ton. Peatlands in the landscape were found scattered and interspersed in small hill of lowland area, which sometimes found abrupt. The peat depth varies between 0.3 to 5.2 m. For below carbon stock of peatlands in the area is ~625 ton/ha/m depth, and total ~9.2 million ton C/ha maintained in 5,570 ha peatland area. This total carbon stock in peatlanf is relatively moderate for fibrist peat. The amount of carbon stock in peat layers may vary depend on the peat decomposition level, which dictate peat bulk density, proportion of carbon in organic matters amount per unit weight, and peat depth. It is demonstrated that belowground carbon in peatland hold more than 10 times of above ground C in above ground stand biomass. It is a must to maintain this ecosystem for its 727

carbon stocks, as well as above ground forest vegetation, to avoid CO2 emission/carbon loss from peat and additional sources GHGs to the atmosphere that could enhance global warming. Acknowledgements First, we would like to thank people of villages surround PT CUS landscape, who incredibly tiredless accompanying and supporting us in doing field assessment. We would like to thanks PT CUS, for their supports and facility, these companies allow us to conduct part of our field works within their oil palm concession. We are thankful to Lembaga Ilmu Pengetahuan Indonesia Jakarta, Balai Pengkajian dan Penerapan Teknologi Bogor, Universitas Tanjungpura, Balai Pengkajian Industri Departemen Perdagangan Pontianak in Indonesia for laboratory facilities and works. With daily field works tasks, we are eternally in debts people who works and share with me outstanding experiences. Special thanks to Fauna Flora International (FFI)-Indonesian Programme that facilitated and provided us funding to make this works possible. We express my deepest gratitude to all fellows in FFI and Lembaga Landscape and Livelihood Indonesia who were so supportive during the process of this study. References Celine E, Philippe M, Astrid V, Catherine B, Musampa C, Pierre D (2013) Nasional forest cover change in Congo Basin: deforestation, reforestation, degradation and regeneration for the years 1990, 2000 and 2005. Global Change Biology, 19:1173-1187. 2. Miettinen J, Shi C, Liew SC (2011) Influence of peatland and land cover distribution on fire regimes in insular Southeast Asia. Regional Environment Change 11, 191–201. 3. Achard F, Eva HD, Stibig HJ, Mayaux P, Gallego J, Richards T, Malingreau JP (2002) Determination of deforestation rates of the world‘s humid tropical forests. Science, 297 (5583), 999–1002. 4. Margono BA, Turubanova S, Zhuravleva, Potapov P, Tyukavina A, Baccuni A, Goetz S, Hansen MS (2012) Mapping and monitoring deforestation andforest degradation in Sumatera (Indonesia) using Landsat time series data sets from 1990 to 2010. Environment Research Letter. 7 (034010): 1-16 5. Lambin, EF (1999) Monitoring forest degradation in tropical region by remote sensing: some methodological issues. Global Ecology and biogeography, 8: 191-198. 6. Grainger A (1993) Controlling Tropical Deforestation. p310. Earthscan Publication Ltd. London. 7. Hoojiers A, Page S, Canadell G, Silvius M, Kwadijk J, Wosten H, Jauhiainen J (2010) Current and future CO2 emissions from drained peatlands in Southeast Asia. Biogeosciences, 7, 1505–1514. 8. Brown S (2002) Measuring carbon in forests: current status and future challenges Environmental Pollution 116: 363–72. 9. Maltby E and Immirzi P (1993) Carbon dynamics in peatlands and other wetland soils: regional and global perspectives. Chemosphere 27: 999–1023. 10. Anderson JA.R (1983). The tropical peat swamps of Western Malaysia. In: Gore A.J.P. (Ed.) Ecosystems of the World 4B, Mires: Swamp, Bog, Fen and Moor, Regional Studies. Elsevier Scientific Publishing, Amsterdam, The Netherlands, pp. 181– 199. 11. Wood TE, Cavaleri MA, Reed SC (2012) Tropical forest carbon balance in warmer world: a critical review spanning microbial- to ecosystem- scale processes. Biological Review, 87. 912-927 12. Uryu Y, Mott C, Foead N, Yulianti K, Setiabudi, Takakai F, Nursamsu, Sunarto,Purastuti E, Fadhil N, Hutajulu CMB, Jaenicke J, Hatano R, Siegert F, Stuwe M (2008). Deforestation, forest degradation, biodiversity loss, and CO2 emission in Riau Sumatra, Indonesia. WWF Indonesia technical Report, Jakarta indonesia, 74 pp. 1.

728

13. Hooijer A, Silvius M, Wösten H, Page SE (2006) PEAT-CO2, Assessment of CO2 emissions from drained peatlands in SE Asia. Delft Hydraulics report Q3943. 14. Rieley JO, Page SE (eds) (2005) Wise Use of Tropical Peatland: Focus of Southeast Asia, ALTERRA-Wagenigen University and Research Center and the EU INCO-STRAPEAT and RESTORPEAT Partership, Wagenigen. 15. Jaenicke J, Rieley JO, Mott C, Kimman P, Siegert F (2008) Determination of the amount of carbon stored in Indonesian peatlands. Geoderma 147, 151 − 158. 16. Saatchi SS, Harris NL, Brown S (2011) Benchmark map of forest carbon stocks in tropical region accross three continents. Proceedings of the National Academy of Science, 108: 9899-9904. 17. Van der Werf GR, Morton DC, DeFries RS, Olivier JGJ, Kasibhatla PS, Jackson RB, Collatz GJ, Randerson JT (2009) CO2 emission from forest loss. Nature Geoscience 2, 737–738. 18. Chave J, Andalo C, Brown S, Cairns MA,Chambers JQ, Eamus D Foster H, Fromard F, Higuchi N, Kira T, Lescure JP, Nelson BW, Ogawa H, Puig H, Rie´ra B, Yamakura T (2005) Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia 145(1): 87-99. 19. Brown S (1997) Estimating biomass and biomass change of tropical forests: a primer. FAO Forestry Paper no. 134. Rome 20. Puhlin H (2009) Methods in Assessing Carbon Stocks. Forest Carbon Project Development Training Program. World Agroforestry Centre 21. IPCC. 2011: Summary for Policymakers. In: Edenhofer O, Pichs-Madruga R, Sokona Y, Seyboth K, Matschoss P, Kadner S, Zwickel T, Eickemeier P, Hansen G, Schlomer S, von Stechow C, (eds). IPCC Special Report on Renewable Energy Sources and Climate Change Mitigation. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. 22. Gibbs HK, Brown S, Niles JO, Foley JA (2007) Monitoring and estimating tropical forest carbon stocks: making REDD a reality. Environmental Research Letter, 2, DOI:10.1088/1748-9326/2/4/045023. 23. Don A, Schumacher J, Freibauer A (2011) Impact of tropical landuse change on soil organic carbon stocks-a meta analysis. Global Change Biology 17: 1658-1670. 24. Wellock M, Reidy B, Laperle CM, Bolger T, Kiely G (2011) Soil organic carbon stocks of afforested peatland in Ireland. Forestry 04(4) : 441-451 doi.10.1093/forestry/cpr046.

729

ISFAS-1523 Economic Value of Tembawang Ecosystem in Sanggau Subdistrict West Kalimantan Province, Indonesia Emi Roslinda, Siti Masitoh Kartikawati, Rabudin Faculty of Forestry Tanjungpura University, Indonesia E-mail: [email protected] 5. Background/ Objectives and Goals The Tembawang Ecosystem in West Kalimantan (Indonesia) spreads over especially in Sanggau Subdistrict region. Tembawang is a system of land use by customary community of Dayak ethnic which traditionally managed and used communally. Tembawang has a relationship with the shifting cultivation, which has been practiced since hundred years ago. In the past, the Dayak settlements moved to follow the movement on their upland farm location. They usually cultivated various kinds of crops and fruits in their settlement. When they moved, the old settlement will be abandoned and become an agroforest which then called tembawang. Tembawang often referred to as an agroforest because it has a variety of plants ranging from trees with a diameter over 100 cm to understory plants. Tembawang is regarded as a unique ecosystem for storing high values in terms of biodiversity, economic, and conservation moral. Based on tembawang description, tembawang likes forest ecosystem which supports almost all of the needs of human being through its contribution of tangible and intangible values. This has resulted in the underestimation of the total value of the tembawang leading to the conversion of tembawang areas into other more tangibly profitable usages, and miss management of the tembawang. The objective of this study are to determine the total economic value (TEV) of the Tembawang ecosystem and evaluate the consequences of decentralization. 6. Methods Research was conducted in Tembawang Ampar Sanggau Sub-district of West Kalimantan Province. Tembawang Ampar covers 6,69 hectares is still covered by forest, while most of it has been deforested and converted to community farms and stone quarries. Research was conducted from Mei –June 2015. Survey methods was used in this research. 70 respondents selected by census. Data collection was undertaken through general field observation, key informant personal interviews & respondent interview with questionare. Total economic value in this research restrited by direct use value, indirect use value and option value. Direct use value and option value were categorized in tangible products and indirect use value was categorized in intangible product. The tangible and intangible products of the tembawang being valuated were grouped into biomass and hydrological role. While the economic values were calculated based on market price (biomass) and opportunity cost (hydrological role). 7. Expected Results/ Conclusion/ Contribution Ampar populations sample are aged between 25 and 77 years with an average of 45 years. The gender and marital status are not balanced, dominated by man. Professional occupation covered a not wide ranged of employment categories (concenterated in the agricultural sector) and unemployment. The average household size is 5. The education are averaged pass in elementary school (SD). In Tembawang Ampar, there are 84 plant species were utilized were utilized by Dayak community. More of these plants are fruit tree species. Benefit from Tembawang for community were material for energy (firewood), food and feed, resin, constrution material, 730

medicinal plants, material for religious ceremony, wildlife (for hunting), hydrological function, air and climate regulation function, cultural/religious ritual, farming location, and nutrient cycling. The results showed that total economic value of Tembawang Ampar was Rp 263,837,082/year, equivalent to Rp 36,900,291/year/ha. They are Rp 33,545,791/year/ha from direct use value, Rp 3,072,738/year/ha from indirect use value, and Rp 281,762/year/ha from option value. Total economic value of Tembawang rather low. The low of economic value due to limited benefits assessed in this study. Whereas in fact, there are many benefits from Tembawang felt by the community. Conserving Tembawang ecosystem will not be achieved by simply considering the conservation of species in Tembawang, but the preservation will need to encompass a wide range of natural resources and the potential services provide by them. Our samples appear not to entirely represent the Sanggau population and so more detailed surveys are neede. Nevertheless they provide information that is vastly superior to any other available on these forests in the subdistrict. They represent a tangible set of data pointing to the values and potential that exist. The result obtained from the real price market for the proposed economic program highlighted the real value of tembawang for Sanggau population. The importance of awareness campaign to give information about the value of though the amount deducted from amenity services. Keywords: Tembawang, economic valuation, direct use value, indirect use value, option value.

731

ISFAS-1533 Trail-Following and Odor Effects of Plant Leaf Extracts on Subterranean Termite Coptotermes Formosanus Shiraki Yuliati Indrayania, Musrizal Muinb, Cynthia Adillac, Tsuyoshi Yoshimurad a Faculty of Forestry ,Tanjungpura University, Indonesia E-mail: [email protected] b

c

Faculy of Forestry ,Hasanuddin University, Indonesia E-mail: [email protected]

Faculty of Mathematic and Fundamental Science ,Tanjungpura University, Indonesia E-mail: [email protected] d

Research Institute for Sustainable Humanosphere, Kyoto University, Japan E-mail: [email protected]

Abstract The abundance of plant-derived compounds with different characteristics has caused plenty of interesting researches in many scientific areas. Plant extract with its chemical compounds has many potential uses. One of these uses is in agriculture to manage pests with less risk than with synthetic compounds that are toxicologically and environmentally undesirable. The present paper focuses on researches regarding with the potential use of leaf extracts of some tropical plants for controlling subterranean termites. Leaves of five plant species [clove (Syzygium aromaticum), cajuput (Melaleuca leucadendra L.), cinnamon (Cinnamomum burmannii), basil (Ocimum sanctum L.), bay (Syzygium polyanthum Wight)] were extracted with 70% ethanol solution, and the extracts separately were evaluated for their attractant properties against Coptotermes formosanus Shiraki using olfactory and Y-line trail following tests, respectively. The solutions were diluted at 1% concentration based on their eugenol content in each extract. For each test, a single worker was introduced to the test unit with 15 replicates. The results shows that M. leucadendra crude extract had the most attractive among the five plant crude extracts, with the olfactory behaviour of termites by mean of olfactory response test and trail following test. While, S. aromaticum crude extract could also be considered as attractive for the termites especially through trail following test. The rest crude extract such as C. burmannii, O. sanctum, and S. polyanthum were not attractive. The attracting rate were 40%, 80%, 53.33%, 60% and 46.67% for S. aromaticum, M. leucadendra, C. burmannii, O. sanctum and S. polyanthum extract, respectively. While, trail following rates were 86.67% for clove extract, 66.67% for cajuput extract, 53.33% for cinnamon extract, 20% for basil extract as well as 26.67% for bay extract. Keywords: Attractants, Coptotermes formosanus, plant leaf extracts, attractive properties, trail following test. 1. Background As subterranean termites are well-known as one of the main organisms causing damages to wooden structure, many efforts have been made for their control. Subterranean termite treatment has changed dramatically in terms of systems, application techniques and products over the last two decades. Nowadays, the use of bait systems with various methods of application has become a commercially developed approach. Baiting systems have been widely 732

searched for termite control (Myles 1996; Barbara and Forschler 2000; Grace and Su 2001; Getty et al 2000; Kubota 2011). Basically, the important point of bait systems is the creation of condition where termite‘s workers can get to forage for food, bring it back to the colony, and feed the whole colony members. In baiting technology, a food source being a part of the system should be viable and accepted by worker termites. They then feed other colony members with this bait, leading to the elimination of termite colony because of their eventual death. Therefore, for a successful baiting system, it is necessary to provide the termites with food sources containing attractants for bait matrix. Many bait toxicants are considered to be potential as active ingredients in termite baits (Eger et al. 2012). However, due to the economic and environmental reasons, efforts have been made to search the potential of using bio-based materials for termite control (Sajap and Aloysius 2000, Sajap et al. 2006, Safian et. al. 2011). Plant extracts offer a variety of chemical compounds that have many potential uses. One of these uses is in agriculture to manage insect pests with less risk than with synthetic compounds that are toxicologically and environmentally undesirable. Utilization of plant extract for termite control has been conducted by many researchers. Singh et al., (2001) and Ding and Hu, (2010), reported plant extract having anti-termite properties, and that termite-resistant formulations. Unfortunately, usages of plant extract as bio-based attractant for termites bait formulation for has not been reported so far. Waller et al. (1999) indicated the important of chemical directional cues as one of the factors affecting termite recruitment to the location of the bait. In fact, crude extracts contains a mixture of some compounds that likely to have different performance compared to that of single compound. Therefore, besides its attractant effects, the condition of crude extracts may also result in repellent effects to termites. Study had been conducted to explore the performance of crude plant extract against insects in term of calling behaviour, but they were limited on Sogatella furcifera (Homoptera: Delphacidae) (Khan & Saxena, 1986). This current study aimed at examining the direct response of termites C. formosanus against crude extracts of leaves from five tropical plant species in terms of their attractiveness and repellence. 2.

Methods

Termites Matured workers of t h e subterranean termite Coptotermes formosanus Shiraki were obtained from the termitarium of the Deterioration Organisms Laboratory (DOL), Research Institute for Sustainable Humanosphere, Kyoto University, Japan. Extracts and treatments Leaf extracts of five tropical plant species [clove (Syzygium aromaticum), cajuput (Melaleuca leucadendra L.), cinnamon (Cinnamomum burmannii), basil (Ocimum sanctum L.), Bay (Syzygium polyanthum Wight)] species were prepared according to Anas et al (2013). Prior to the extraction, the plant leaves were air-dried for 3 days and crosscut into small segments. The extraction was conducted with 70% ethanol solution for 3 x 24 hours and then rotary evaporated at 60°C for 2 x 24 hours until the extract changed to gel condition. For treatments, all of the leaf extracts were diluted with 70% ethanol solution at the same and equal concentration of 1% eugenol content. This concentration was calculated based on the eugenol content of each extract shown by GC-MS analyses in the preliminary test. Olfactory response test 733

Response of termites to the extract odors was determined by using a test unit of olfactometer modified from Wang et al (2012) as shown in Figure 1. The unit mainly consisted of distilled water bottle, molecular sieve 3A bottle, molecular sieve 5A bottle, charcoal bottle, circulating air pump, air flow meters, and Y-tube boom. The series of distilled water, sieves, and charcoal bottles were connected with odorless silicon tube (Sanplatec, Japan) and functioned to provide clean air being used to blow the odor source. The Y-tube boom has the length of 7 cm and 3 cm for the main and side arms, respectively, with the inside diameter of 1 cm. A 25µl extract was pipetted to a 1.3 mm paper disk (Advantec, Toyo Roshi Kaisha, Ltd.), air-dried, and put into one of the Y-tube side arm. Another paper disk with ethanol solvent was also put into the other Y-tube side arm as control. The two Y-tube side arms were then ventilated with two lines of air pump at the same flow rate of 0.6 L min-1. One termite worker was introduced to the main arm and allowed to move towards any of the side arms and it replicated 15 times. The test unit was covered with a sheet of red transparent plastic wrap (Rengo, Osaka, Japan) to protect the termite from the direct light. The number of workers which went through any of the side arms and stayed for more than 30 seconds was recorded. The Y-tube boom was cleaned with ethanol and dried and after each replicate, the two side arms were exchanged in turn for treatment and control.

Trail following test Trail test was performed according to the method developed by Hall and Traniello (1985) with a modification in Y line measurement. This study employed a Y-shaped pencil line which was drawn onto a 12.5 cm diameter filter paper (Whatman No. 2, United Kingdom) with 3 cm and 7 cm lengths of main stem and arms, respectively. The stem and one arm of each Y were treated with 75 mm3 extract of each plant species, while the other arm was treated with ethanol solvent as control. After drying about 2 minutes, one termite worker was placed at the base of the Y stem using a 1-cm diameter plastic cylinder and allowed for one minute prior to releasing. The test unit was covered with a 12.5 diameter Petri dish coated with a red transparent plastic wrap (Rengo, Osaka, Japan) to protect the termite from the ambient environment. The termites were allowed to move and select any of the Y arms line direction. The test was conducted with fifteen replicates for each extract. Number of individual termites went through the line with extracts were recorded and considered as the positive response of the termites. Data analysis Repellency rate of each of the crude extracts against the termites was calculated based on the percentage of the number of termites that selected control to the total number of tested termites. Whilst, the positive response of termites to each of the crude extracts was calculated 734

based on the percentage of the number of termites that went through the extract lines to the total amount of tested termites. All data were analyzed by using Chi-Square Test to compare between treatment and control. 3. Results and Discussions Olfactory response In the olfactory test, the termites responded differently towards treated and untreated paper disks. Compared to untreated control, the significant attractive rates were found only at the paper disk treated with M. leucadendra extract (significant at P < 0.01). There were no significant response found at the other tested extracts. The odor attractive rates of extracts were 40% for S. aromaticum, 80% for M. leucadendra, 53.33% for C. burmannii, 60% for O. sanctum, and 46.67% for S. polyanthum. This results also indicated that treatment with M. leucadendra extract demonstrated significant attractant effect to the termites with the repellent rate of 20%.

Trail following In the following trail test, positive responses were found at the S. aromaticum and M. leucadendra extracts as shown in Figure 2. The number of termites went through the extract line of S. aromaticum was significantly greater than that of the number of termites went through the other ways (significant at P < 0.05). The same significant response of termites was also found at the M. leucadendra extract (significant at P < 0.05). There was no significant different at the extract line of C. burmannii (not significant at P < 0.05). On the other hand, negative response was found at the O. sanctum and S. polyanthum extracts. The number of termites went through the extract line of O. sanctum was significantly less than that of the number of termites went through the other ways (significant at P < 0.05). The same negative significant response of termites was also found at the extract line of S. polyanthum (significant at P < 0.05). Hence, the attractant rates of O. sanctum and S. polyanthum extracts were lower than those of S. aromaticum and M. leucadendra extracts. The trail following rates of extracts were 86.67% for S. aromaticum, 66.67% for M. leucadendra, 53.33% for C. burmannii, 26.67% for O. sanctum, and 20% for S. polyanthum.

735

According to our bioassay, M. leucadendra crude extract had the most attractive among the five plant crude extracts, with the olfactory behaviour of termites by mean of olfactory response test and trail following test. While, S. aromaticum crude extract could also be considered as attractive for the termites especially through trail following test. The rest crude extract such as C. burmannii, O. sanctum, and S. polyanthum were not attractive. It is understandable since many plants have been recognized to have anti-termitic activities such as clove (Khoul et al., 2008; Zhu et al., 2001a, b), cinnamon (Sbeghen et al., 2002; Khoul et al., 2008), basil are also known for their varied pest control properties (Khoul et al., 200 8), bay leaves possesses repellent properties for controlling structural pests as termite (Onyambu et al., 2014). Earlier studies have suggested that anti-termitic activities of many plants caused by the presence of eugenol compound. Cornelius et al. (1997) indicated that eugenol compound caused 100% subterranean termite C. formosanus mortality at concentration as low as 0.06 µl/g of sand. Furthermore, previous data has shown that the three crude extracts (S. aromaticum, M. leucadendra, and C. burmannii) were contain eugenol compound which GCMS analyze shows percentage of eugenol compound for each crude extract used in this study as follows: S. aromaticum (7.92%), C. burmannii (4.64%), M. leucadendra, (3.36%), while the rest two crude extract contain derivate of eugenol compound such as 2.24% of methyleugenol in O. sanctum and 6.54% of aceteeugenol in S. polyanthum (Indrayani et al, 2015, unpublished data). Results from this study confirm that M. leucadendra crude extract have the lowest eugenol compound which showing most attractive against C. formosanus. The above results also demonstrated the potential of S. polyanthum extracts as either attractant or repellent substances as it showed positive and negative effects termites with the repellent and trail following rates of 60% and 86.67%, respectively. These effects should be carefully considered in its application as it was found to have high content of eugenol (Indrayani et al, 2015, unpublished data). A previous study showed that eugenol at the concentration of 0.06 µl/g of sand was toxic to termites (Cornelius et al. (1997). In contrast, a study using 10% crude extract of S. aromaticum (equal to eugenol content of 0.00792 %) was found to attract termites (Indrayani et al, 2015, unpublished data). The reason of these different termite responses is unclear and need to be further explored, particularly related to the appropriate concentration and synergic interaction of active compound present in the crude extract of S. aromaticum.

736

The results will be useful in determining the attractive compound of different crude extract against subterranean termite C. formosanus and allow us to detect whether the attractive crude extract have promising to be bait matrix formulation particularly in the development of baiting systems technologies for termite control. 4. Conclussions The tested leaf extracts have different performances in terms of their effects on termite C. formosanus. Positive responses were performed by S. aromaticum and M. leucadendra extracts as the termites preferred to go to the extract lines in the following trial tests. However, the treatment of S. aromaticum extract also demonstrated negative reaction as the termites went to the other way (untreated control) in the olfactory response test. There were no significant response of termites on the treatments using C. burmannii, O. sanctum, and S. polyanthum extracts. The M. leucadendra extract was found to be potential to meet the need of termite control with its high attractant and low repellent rates. Acknowledgments The authors gratefully acknowledges to the DOL/LSF Collaborative Research Program at Research Institute for Sustainable Humanosphere (RISH), Kyoto University and to Directorate General of Higher Education of Indonesia for financial support in Fiscal Year 2015 under Program ―International Research Collaboration and Scientific Publication‖ 5. References Anas, Y., Puspitasari, N., & Nuria, M.C.(2013). Activity of ethanol extract from flower and leaf of clove (syzygium aromaticum (L) Merr. & Perry.) on Barbara, L.T., & Forschler, B.T.(2000). Criteria for assessing efficacy of stand-alone termite bait treatments at structure. Sociobiology 36(1), 245-255. Cornelius, M. L., Grace, K. J., & Yates J.R.(1997). Toxicity of monoterpenoids and other natural products to the formosan subterranean termite (Isoptera: Rhinotermitidae). J. Econ. Entomol 90(2), 320-325. Eger, J. E., Lees, M. D., Neese, P. A., Atkinson, T. H., Thoms, E. M., Messenger, M. T., Demark, J. J., Lee, L. C., Vargo, E. L. & Tolley, M. P.(2012). Elimination of subterranean termite (Isoptera: Rhinotermitidae) colonies using a refined cellulose bait matrix containing novifumuron when monitored and replenished quarterly. J. Econ. Entomol. 105(2), 533-539. Ding, W. & Hu, X. P.(2010). Antitermitic effect of the Lantana camara plant on subterranean termites (Isoptera: Rhinotermitidae). Insect Science 17(5), 427–433. Grace, J.K., & Su, N. Y.(2001). Evidence supporting the use of termite baiting systems for long- term structural protection (Isoptera). Sociobiology 37(2), 301-310. Getty, G. M., Haverty, M. I., Copren, K. A., and Lewis. V. R.(2000). Response of Reticulitermes spp. (Isoptera: Rhinotermitidae) in Northern California to baiting with Hexaflumuron with sentricon termite colony elimination system. Journal of Economic Entomology 93(5), 1498-1507. Hall, P., & Traniello, J. F.(1985). Behavioral bioassays of termite trail pheromones: Recruitment and orientation effects of cembrene-A in Nasutitermes costalis (Isoptera: Termitidae) and discussion of factors affecting termite response in experimental contexts. J. Chem. Ecol. 11(11),1503-1513. Khan, Z. R. & Saxena, R. C.(1986). Effect of steam distillate extracts of resistant and susceptible rice cultivars on behaviour of Sogatella furcifera (Homoptera: Delphacidae). Journal of Economic Entomology 79(4), 928-935.

737

Koul, O., Walla, S. & Dhaliwal, G.S.(2008). Essential Oils as Green Pesticides: Potential and Constraints. Journal of Biopesticide international 4(1), 63-84. Kubota, S.(2011). Colony Elimination of Subterranean Termites by Bait Application Using Benzoylphenylurea Compounds, with Special Referenceto Bistrifluron, Pesticides in the Modern World – Risks and Benefits, Dr. Margarita Stoytcheva (Ed.), ISBN:978-953-307-458-0, Retrieved from:http://www.intechopen.com/books/pesticides -in-the-modern-world-risks-and-benefits/colony-elimination-of-subterranean-termites-bybait-application-using-benzoylphenylurea-compounds-wi Myles, T. G.(1996). Development and evaluation of a transmissible coating for control of subterranean termites. Sociobiology 28(3), 373-458. Safian, A., Sajap, A. S, Sukari, M. A., Harris, F. A., Bakar, N. H. A., & Kassim. R. M.(2011). Effects of Leaf Extracts of Murraya koenigii (Rutaceae) on Coptotermes curvignathus (Isoptera: Rhinotermitidae). Sociobiology 57(2), 291-300. Sajap, A. S. & Aloysius, F.(2000). Effects of Leaf Extracts of Azadirachta excelsa (Jack) Jacobs on Coptotermes curvignathus Holmgren (Isoptera: Rhinotermitidae). Sociobiology 36(3), 497-503. Sajap, A. S., Lardizabal, M. L. T., Fauzan, A. B. & Sahri, H. M.(2006). Feeding response of subterranean termites, Coptotermes curvignathus (Isoptera: Rhinotermitidae) to Azadirachta excelsa (Meliaceae) extractives and its timber. Sociobiology 48(2), 447-455. Sbeghen, A. C., Dalfovov, V., Serafini, L. A. & De-Barros, N. M.(2002). Repellence and toxicity of basil, citronella, ho-sho and rosemary oils for the control of the termite, Cryptotermes brevis (Isoptera: Kalotermitidae). Sociobiology 40(3), 585-594. Singh, Y., Ranawat, B. S., Verma, R. K. & Nayal, S. S.(2001). Termite control with medicinal plant products. Journal of Medicinal and Aromatic Plant Sciences, 22(23), 151-153. Waller, D. A., Morlino, S. E. & Matkins, N. (1999). Factors affecting termite recruitment to baits in Laboratory and field studies. Proceedings of the 3rd International Conference on Urban Pests, p.507-600. Wang, J., Zhang, H., Zeng, L. & Xu, Y. J.(2012). Repellent effects of five plant essential oils on the red imported fire ant, Salenopsis invicta. Sociobiology 59(3), 695-701. Zhu, B. C. R., Henderson, G., Chen, F., Fei, H. & Laine, R. A.(2001a). Evaluation of vetiver oil and seven insect active essential oils against the Formosan subterranean termite. Journal of Chemical Ecology 27(8), 1617-1625. Zhu, B. C. R., Henderson, G., Chen, F., Maistrello, L. & Laine, R. A.(2001b). Nootkatone is a repellent for Formosan subterranean termite (Coptotermes formosanus). Journal of Chemical Ecology 27(3), 523-531. Onyambu, G.K., Maraga, R., Ndungu, M., Mkoji, G., Patrick, K., Wanjoya, A.(2014). Toxicity and repellent effects of leaf extracts of four Kenyan plants against nymphs and adults of Rhipicephalus appendiculatus (Neuman 1901). Journal of Applied Chemistry 7(9), 33-39.

738

ISFAS-1537 Role of Rhizosphere Microbes in Increasing Plant Growth of Jatropha Curcas in Mercury Contaminated Soil Wiwik Ekyastuti*, Hanna Artuti Ekamawanti Fakultas Kehutanan Universitas Tanjungpura, Jl. Imam Bonjol Pontianak Kalimantan Barat Email: [email protected] Soil in the area of the ex-illegal gold mining, have a chemical physical constraints to the growth of plants. These constraints are to have low organic matter content, nutrient-poor, acidic pH, CEC is very low and the soil texture is dominated by sand. In addition, the soil is contaminated with mercury (Ekyastuti 2013). Mercury in plant tissue may interfere with the process of photosynthesis and metabolism of plants so that the plants become stunted growth (Rugh et al. 2000). Plants have a different degree of resistance to the mercury in the environment. Phytoremediation is defined as the leaching of pollutants or contaminants that involves the use of plants to remove, move, stabilizing and / or degrade contaminants in soil, sediment and water (Mudgal et al. 2010). Previous research has been conducted by Ekamawanti and Ekyastuti (2009) obtain results that jatropha (Jatropa curcas) including a plant that mercury tolerant, so it could potentially be used as a phytoremediator of mercury. On the other hand, some types of rhizosphere microbes such as arbuscular mycorrhizal fungi (AMF) and mercury reducing bacteria (MRB) also has the ability to reduce heavy metals including mercury. Symbiosis between the rhizosphere microbial with jatropha is not yet known to improve plant growth. Therefore, this study was conducted in order to determine the role of rhizosphere microbial namely AMF and BRM in enhancing the growth of jatropha in mercury contaminated soil. The study was conducted in two places, namely in the greenhouse and in the field (the tailings ex-illegal gold mining). The research was conducted using factorial experiment with a completely randomized design. In both locations of planting, treatment factors consist of two types of microbes that were AMF consisting of 4 levels of factors and MRB consisting of 3 levels of factors. So, the total number of treatments were 12, namely: kontrol, Glomus SS11, Glomus SS15, Glomus SS18, Pseudomonas sp, Bacillus sp, Glomus SS11 + Bacillus sp, Glomus SS11 + Pseudomonas sp, Glomus SS15 + Bacillus sp, Glomus SS15 + Pseudomonas sp, Glomus SS18 + Bacillus sp dan Glomus SS18 + Pseudomonas sp. Each combination treatment was repeated four times in the greenhouse and ten times in the field. Mercury concentrations in the media were 10 ppm. Observations were made on plant growth response in the form: increase of height, diameter and number of leaves, total dry weight, N and P uptake and root shoot ratio. At the end of the study carried out calculations on: translocation factor, tolerance index, the percentage of infected root of AMF and search back (re-isolation) MRB. The results showed that in general, the rhizosphere microbial interactions between AMF and MRB are simultaneously able to increase the growth of jatropha not only in the greenhouse but also in the field. This is presumably due to the five isolates both AMF (Glomus SS11, Glomus SS15 & Glomus SS18) and MRB (Pseudomonas sp and Bacillus sp) capable to symbiosis with jatropha. In the nursery (greenhouse) it is known that isolates of Bacillus sp, Bacillus sp + Glomus SS11 and Bacillus sp + Glomus SS18 are an inoculum that very effective in enhancing the growth of seedlings of jatropha. Meanwhile after planting in the field (the tailing ex-illegal gold mining) turned out there was little change. Combination treatment between the bacteria Bacillus sp + AMF Glomus SS18 is still provide a good influence (very effective), while for the other combination treatments are to be less good (not effective). This condition illustrates 739

that isolates of Bacillus sp when applied singly or in combination with Glomus SS11 have a poor competitiveness in the field of tailings in the environment. Göhre & Paszkowski (2006) explains that there are several mechanisms of symbiosis between the plant and the rhizosphere microbial (AMF and MRB) in managing heavy metals including mercury. These mechanisms are as follows: as chelating agents, binding metals to the cell wall, the plasma membrane acts as a barrier selective, metal-specific and non-specific in with active and passive, chelates in the cytosol, eliminating the metals are actively or passively, alienating metal in the vacuole cell, and transporting the metal both in the hyphae and in arbuscules of fungi. This is indicates that after being transferred to the field (moved from the greenhouse) the mechanism cannot run well, so that the effectiveness of the symbiosis between plants with rhizosphere microbial (Bacillus sp and Glomus SS11) are decrease. The conclusion that can be drawn from this research is for application on Jatropha, both types of rhizosphere microbial namely AMF Glomus SS18 and MRB Bacillus sp highly recommended to be inoculated simultaneously to tailings (soil) contaminated with mercury. Furthermore, the results of this study can be used for reclamation activities, especially in the tailing of ex-illegal gold mining. Keywords: illegal gold mining, jatropa curcas, rhizosphere microbes, tailing.

1.

2. 3. 4.

5.

References Ekamawanti HA, W Ekyastuti. Uji Efektivitas Isolat-isolat Mikrob Rizosfer Terhadap Pertumbuhan Beberapa Jenis Tanaman di Media Tailing yang Tercemar Merkuri. Laporan akhir Hibah Kompetitif Penelitian Sesuai Prioritas Nasional. Fakultas Kehutanan UNTAN; Pontianak; 2009. Ekyastuti W. Akselerasi suksesi vegetasi untuk mitigasi cemaran merkuri akibat penambangan emas skala kecil. Disertasi Fakultas Kehutanan UGM; Yogyakarta; 2013. Göhre V & U Paszkowski. 2006. Contribution of The Arbuscular Mycorhizal Symbiosis to Heavy Metal Phytoremediation. Planta (223): 1115-1122. Mudgal, Varsha, N Madaan and A Mudgal. 2010. Heavy metals in plants: phytoremediation: Plants used to remediate heavy metal pollution. Agriculture and Biology Journal of North America. ISSN Print: 2151-7517, ISSN Online: 2151-7525. Rugh CL, SP Bizily & RB Meagher. 2000. Phytoreduction of environmental mercury pollution, p. 151-169. Editors: I Raskin dan BD Ensley. Phytoremediation of Toxic Metals: Using Plants to Clean Up the Environment. Wiley-Interscience Publication. John Wiley & Sons, Inc. New York.

740

Computer Engineering and Technology (4)

Thursday, March 31, 2016

14:50-16:20

Room B

Session Chair: Prof. Pei-Yu Lin ACEAIT-7336 Design of a Brainwave Controlled Quadcopter Hui-Wen Yang︱National University of Tainan Pei-Wen Cheng︱National University of Tainan Tien Tsai︱National University of Tainan Koun-Tem Sun︱National University of Tainan ACEAIT-7431 Automatic Navigation of Robot by Brainwaves Pei-Wen Cheng︱National University of Tainan Hui-Wen Yang︱National University of Tainan Koun-Tem Sun︱National University of Tainan ACEAIT-7347 High Quality Dual Imaging Technique Chi-Shiang Chan︱Asia University Wen-Chuan Wu︱Aletheia University Jen-Ho Yang︱Kainan University Pei-Yu Lin︱Yuan Ze University ACEAIT-7317 Comments on an Authentication Model of IoT Control System Ya-Fen Chang︱National Taichung University of Science and Technology Ya-Ling Lo︱National Taichung University of Science and Technology Wei-Liang Tai︱Chinese Culture University ACEAIT-7318 Comments on Kuo et al.‟s Anonymous Mobility Network Authentication Scheme Ya-Fen Chang︱National Taichung University of Science and Technology Min-How Hsu︱National Taichung University of Science and Technology Wei-Liang Tai︱Chinese Culture University

741

ACEAIT-7368 Using Kalman Filter for Improved Ego-Motion Estimation Yu-Hsiang Chang︱National University of Kaohsiung Ting-Hsiang Huang︱National University of Kaohsiung Chen-Chi Chuang︱National University of Kaohsiung Chia-Yen Chen︱National University of Kaohsiung Chia-Chen Kuo︱National Center for High-performance Computing

742

ACEAIT-7336 Design of a Brainwave Controlled Quadcopter a

Hui-Wen Yanga,*, Pei-Wen Cheng, Tien Tsai, Koun-Tem Sun Department of Information and Learning Technology, National University of Tainan, Taiwan E-mail: [email protected]

Abstract This paper proposes a brain-computer interface (BCI) of quadcopter to help severe physical disabilities explore the outside world freely. The real-time video of the camera of quadcopter transmits to the system and displays in the interface. Users can watch the scenery around the quadcopter by using the brainwave to operate the functions on the interface. Experimental results show the feasibility and practicability of the proposed system. Five merits were included in the research: (1) design a BCI of quadcopter, (2) embed quadcopter's real-time video in user interface, (3) design an experimental scene and conduct the experiment for 20 people, (4) evaluate the accuracy and efficiency of the experiments, and (5) evaluate the practicability and give suggestions for future works. Keywords: quadcopter, severe physical disabilities, brain-computer interface (BCI), real-time video 1. Background/ Objectives and Goals There are some severe physically disabled people such as cerebral palsy, brain stem stroke, muscular dystrophy and amyotrophic lateral sclerosis (ALS), etc. hope to see the outside world freely. These people are often subjected to physical barriers so that unable to freely look around the environment. Although the current technology has developed a variety of auxiliary tools to assist severe physical disabilities in maintaining daily life, most auxiliary tools still have many limitations on the operations. Most of the operations have to use body movement to complete the action such that can not apply to severe physically disabled people. However, they can control facilities with brainwaves, if the functions of their brains are normal. Brainwave controlled technology is a kind of control technologies without body movement. It allows users directly control hardware devices by brainwave (Tonin et al., 2009). By using EEG to collect the brainwave reaction is a non-invasive technique which can be effectively applied to the physically disabled people interact with the environment (Carlson & Millan, 2013; Singla & Haseena, 2013). Over the past brain-computer interface is mostly used for medical aids to obtain various control ability, for example, brainwave controlled wheelchair (Khare, Santhosh, Anand, & Bhatia, 2011; Carlson, & Millán, 2013), prosthetics (Guger, Harkam, Hertnaes, & Pfurtscheller, 1999; Hochberg et al., 2006; Müller-Putz & Pfurtscheller, 2008), biped robot (Li, Jaramillo, & Li, 2011; Bell, Shenoy, Chalodhorn, & Rao, 2008), wheeled robot (Belluomo, Bucolo, Fortuna, & Frasca, 2012), and play chess games and puzzle games (Kaplan, Shishkin, Ganin, Basyul, and Zhigalov, 2013), etc. Therefore, this research combines brainwave technique and the quadcopter controlled technology to design a brain-computer interface system to control the quadcopter such that users can explore outside world freely. 2. Methods 2.1 Event-Related Potentials (ERPs) The signals of brainwave can be observed by the electroencephalography which is a 743

non-invasive technique. In 1965, Sutton first proposed the study of event related potentials (Sutton, Braren, & Zubin, 1965) that provided an objective, feasible and simple method for observing the brainwave after stimuli. After then, event related potentials have become a popular and widely used technique to observe brain activities, and also widely used in various cognitive sciences (Kutas & Hillyard, 1980). The design of experiment of event-related potentials need some particular stimulus (e.g., text, pictures, etc.) to flicker on the computer screen. Each stimulus repeats flashes many times, and the response of brainwave is collected after the stimulus. Accumulate the brainwaves with the same stimulus such that we can obtain the effect of averaging signal and inhibiting noise. After accumulating the brainwaves, we can observe the characteristics of the stimulus. Event-related potentials consist of a series of positive and negative voltage deflections. It usually uses P to express positive wave and uses N to express negative wave. Named with the latent time (the time gap between the stimulus and the waves appeared, unit: ms) such as P300 (it means the positive wave about 300ms after stimulus), and N200 (the negative wave about 200ms after stimulus). Figure 1 shows an example of three different event-related potentials for three control buttons in which the ―Target‖ stimuli evoke the greatest P300 (Huang, Tai, Tian, & Sun, 2013). 2.2 Brain-Computer Interface (BCI) Brain Computer Interface (BCI) first proposed by Jacques J. Vidal (Vidal, 1973). In 1977, he cooperated with UCLA BCI laboratory team to develop a brain-computer interface, which used visual stimulation to control cursor (Vidal, 1977). In the ensuing decades, the result deeply affected the biological cybernetics study. Brain-computer interface systems interpret the brain signals from external stimuli that provide a communication method to people who is physically disabled.

The physically disabled person can select a target from a number of control buttons to achieve the purpose of communication (Hong, Guo, Liu, Gao, Gao, 2009). The basic process of BCI divided into four parts. Figure 2 shows the basic operation flow of BCI: (1) Signal capture: capture the brain signals from the electrode on scalp. (2) Digitization: amplify and digitize the brain signals. (3) Signal processing: extract and classify the features of brain signals, and generate a final result. 744

(4) Issue a command: convert the classified result into a command for controlling the external facilities.

2.3 Brainwave Controlled Quadcopter System Design In this study, we design a brainwave controlled quadcopter system to explore outside for severely physically disabled. The system includes six direction of movement (―Upper‖, ―Down‖, ―Go Left‖, ―Go Right‖, ―Forward‖, and ―Back‖) and two switches to turn on or turn off the system (―Start‖ and ―Shutdown‖). Table 1 shows eight control icons. Each icon corresponds to a function. Each icon has rectangular stimuli which flicker from left to right. Users only need to watch the desired function (icon). After six times stimulation, the system can determine the user‘s demand and execute the command. In this study, the quadcopter was developed by ARDrone2.0 (Fig. 3). ARDrone2.0 is a flight control board that can be programmed to perform some actions. The features of ARDrone2.0 include (1) support pressure measurement that can execute ―Alt Hold‖ and automatic landing, (2) have 3-axis gyroscope to automatically calibrate balance, (3) provide C/C++ programming language environment, (4) provide extra functions such as GPS.

745

2.4 System development and interface design We use Borland C++ Builder to develop the brainwave controlled quadcopter system. In the user interface (Fig. 4), the center embeds a real-time video image, and there are eight control icons around the interface to control the movement of quadcopter (Forward, Back, Go Left, Go Right, Upper, Down, Start, and Shutdown). When the system starts, each icon appears rectangular stimuli that flicker from left to right at various time. Initially, quadcopter stopped at the site boundary. The user watches the ―Start‖ icon to make the quadcopter take off and maintain a fixed height (one meter) of the ground. User can give a ―Forward" command to make the quadcopter fly forward, or a ―Go Left" command to make the quadcopter fly left, and so on. The quadcopter will temporarily stop after moving 70 centimeters. Each command executed, the user can rest for a few seconds before executing the next command (Fig. 5).

2.5 EEG signal control Based on our previous research results (Sun, Huang, & Chen, 2011), the brain signal located at the electrode O1 (visual cortex area) is used to control the quadcopter by the generated visual evoked potentials. For producing visual event-related potentials to be more quickly, easily and higher accuracy, we design a stimulation schedule as shown in Fig. 6. A round of flicker of stimuli is 0.96 s, and the number of round is six. Thus the time of stimuli for generating one command needs 5.76s. After six round flickers, the computer sends the command to quadcopter. 746

Then user can rest a few seconds (1 ~ 10s), the system proceed to the next round.

2.6 Experiment Our experiment had passed the review of institutional review board (IRB) of National Cheng Kung University in Taiwan and participants had signed the consent before experiment. Twenty healthy individuals between the ages of 21 and 24, no brain disease, relevant medical history or drug abuse, and had normal or corrected-to-normal vision, voluntarily participated in the experiment. Before the experiment, subjects recognized brainwave controlled quadcopter system interface (Fig. 7) and the functions of eight icons. After explain the experimental scene and purpose, start the brainwave controlled quadcopter system. The system will calculate the ERPs of each icon, and one of icons with the maximum N2P3 ERPs would be the target and execute the corresponding command (Fig. 8). The values of N2P3 (P300 minus N200) were used to evaluate users‘ targets and non-targets. N200 was defined as the minimum negative potential of ERPs between 180 and 210ms, and P300 was defined as the maximum positive potential between 280 and 310ms.

747

2.7 Experimental Scene Design Experimental scene of our research is an empty space with 750 centimeters long and 500 centimeters wide. The mission of the subject would use the brainwave to control the quadcopter flew from start point and landed in the destination. The subject stays in the brainwave control room that cannot see the experimental scene directly but view the scene by the camera of quadcopter. Initially, the subject needs to start the quadcopter, and control the movement of quadcopter by watching real-time video and selecting the demand command by brainwave on the system. The quadcopter needs to fly and land in the destination within five minutes from starting the system. If the operating time exceeds five minutes or the quadcopter flies out of experiment scene, the mission fails, and stops the experiment. 3. Results Experimental results showed that all subjects can control the quadcopter to the destination within five minutes. The averaged time for completing the task is 171.25 seconds, and the fastest time is 142 seconds and the slowest time is 194 seconds with the standard deviation 13.29. The correct rate of brainwave controlled system is between 70% to 100%, and the averaged correct rate is 83.95% with the standard deviation 8.96%. Based on the experimental results, we can obtain that the average of flying speed is 4.786 cm/s, the fastest flying speed is 5.7 cm/s, the slowest flying speed is 4.18 cm/s and the standard deviation is 0.39 cm (Table 2 and Fig. 9).

748

749

4. Conclusions The BCI has been a popular research topic for several years. Previous researches had succeeded to control wheelchairs, prosthesis, robot, and quadcopter. But still now, most of BCI researches only applied in the virtual environment. There are many differences between the real environment and the virtual environment, for example, the time lag between the quadcopter and the BCI system, and the interference of communication between them. In addition, the flying of quadcopter easily affected by the wind or hardware problems which are not existed in the computer simulation system and it must need extra technique to fix these problems. Our system has already controlled quadcopter by using BCI in the field. Based on N2P3 of ERPs at electrode O1, subject can easily control the quadcopter and it doesn‘t need train before the experiment. It causes fewer burdens than using imagine-based BCI. Moreover, by using our method, it could easily expand more functions. Experiments in field, all subjects complete the task within 5 minutes. The total time of completing the mission was between 142 seconds to 194 seconds, and the correct rate is between 70% to 100%. The preliminary results show that the proposed BCI for controlling quadcopter can apply to severe disability people to view the outside environment free. Certainly, our system could be also applied to other equipments such as wheelchairs, prosthesis, or robots. 5. References Bell, C. J., Shenoy, P., Chalodhorn, R., & Rao, R. P. N. (2008). Control of a humanoid robot by a noninvasive brain-computer interface in humans. J. Neural Eng., 5, 214–220. Belluomo, P., Bucolo, M., Fortuna, L., & Frasca, M. (2012) Robot Control through Brain-Computer Interface for Pattern Generation, Complex Systems, 20(3), 243–251. Carlson, T. & Millan, J. R. (2013). Brain–Controlled Wheelchairs: A Robotic Architecture. IEEE Robotics and Automation Magazine, 20(1), 65–73. Guger, C., Harkam, W., Hertnaes, C., & Pfurtscheller, G. (1999). Prosthetic control by an 750

EEG-based brain-computer interface (BCI), Proc. AAATE', 99, 590–595. Hochberg, L. R., Serruya, M. D. , Friehs, G. M. , Mukand, J. A. , Saleh, M. , Caplan, A. H. , Branner, A. , Chen, D. , Penn, R. D. , & Donoghue, J. P. (2006). Neuronal ensemble control of prosthetic devices by a human with tetraplegia, Nature, 442,164–171. Hong, B., Guo, F., Liu, T., Gao, X., & Gao, S. (2009). N200-speller using motion-onset visual response. Clinical Neurophysiology, 120, 1658–1666. Huang, T. W., Tai, Y. H., Tian, Y. J., & Sun, K. T. (2013). The fastest BCI for Writing Chinese Characters Using Brain Waves. Global Congress on Intelligent Systems, 346-349, Hong Kong, December 3-4. Kaplan, A. Y., Shishkin, S. L., Ganin, I. P., Basyul, I. A., & Zhigalov, A. Y. (2013). Adapting the P300-based brain–computer interface for gaming: a review. IEEE Trans. Computer Intel. AI in Games, 5(2), 141–149. Khare, V., Santhosh, J., Anand, S., & Bhatia, M. (2011). Brain Computer Interface Based Real Time Control of Wheelchair Using Electroencephalogram, International Journal of Soft Computing and Engineering, 1(5), 41–45. Kutas, M. & Hillyard, S. A. (1980). Reading senseless sentences: Brain potentials reflect semantic incongruity. Science, 207, 203–205. Li, W., Jaramillo, C., & Li, Y. Y. (2011). A Brain Computer Interface Based Humanoid Robot Control System. IASTED International Conference on Robotics (Robo 2011), Pittsburgh, 390–396. Müller-Putz, G. R. & Pfurtscheller, G. (2008). Control of an electrical prosthesis with an SSVEP-based BCI, IEEE Transactions on Biomedical Engineering, 55(1), 361-364. Singla, R. & Haseena, B. A. (2013). BCI Based Wheelchair Control Using Steady State Visual Evoked Potentials and Support Vector Machines. International Journal of Soft Computing and Engineering (IJSCE), 3(3), 46-52. Sun, K. T., Huang, T. W., & Chen, M. C. (2011). Design of Chinese spelling system based on ERPs. 11Th IEEE International Conference on Bioinformatics & Bioengineering .310-313, Oct. 24-26, Taichung, Taiwan. Sutton, S., Braren, M., & Zubin, J. (1965). Evoked-potential correlates of stimulus uncertainty. Science, 150, 1187-1188. Tonin, L., Menegatti, E., Cavinato, M., D‘Avanzo, C., Pirini, M., Merico, A., Piccione, F. (2009). Evaluation of a robot as embodied interface for brain computer interface systems. International Journal of Bioelectromagnetism, 11(2), 97–104. Vidal, J. J. (1973). Toward direct brain-computer communication. Annu. Rev. Biophys. Bioeng, 2, 157–80. Vidal, J. J. (1977). Real-time detection of brain events in EEG. Proc. of the IEEE, 65(5), 633– 641. Acknowledgment This work was partially supported by the Ministry of Science and Technology of Taiwan, ROC (Project ID: 103-2815-C-024-015-U).

751

ACEAIT-7431 Automatic Navigation of Robot by Brainwaves Pei-Wen Cheng, Hui-Wen Yang, Koun-Tem Sun Department of Information and Learning Technology, National University of Tainan, Taiwan Email: [email protected] Abstract This research proposed an automatic navigation system of robot by using event-related potentials (ERPs) of brainwaves. Based on the specific components N2P3 of ERPs, the path planning and the collision avoidance system embedded in robot can effectively control the movement of robot and reduce the burden of the user. The user can browse the scene with a relaxing and easy manner. According to the experimental results, the accuracy of this automatic navigation robot system is within the range of 66 to 100 percent for twenty volunteers in field. All volunteers completed the mission within fifteen minutes and said that the integration of the brain control technology and the navigation system into the robot control system was feasible that can greatly reduce the burden of operation by brainwaves. The integration of two technologies into brainwave control system makes operation easier and more user-friendly that would be a new trend for the development of brain control interface (BCI). Keywords: robot control, event-related potentials (ERPs), automatic navigation system, collision avoidance, brain control interface (BCI). 1. Background In the society of nowadays, some severe physically disabled people such as cerebral palsy, brain stem stroke, muscular dystrophy and amyotrophic lateral sclerosis (ALS), etc. often subjected to physical barriers so that unable to freely look around the environment. However, there is no difference between their cognitive abilities with the normal limb people. If they can control things by brainwaves, they will be able to overcome the inconvenience of physical disabilities. Brainwave control technology is a kind of control technology without body movement. It allows users directly control hardware devices by brainwaves (Tonin et al., 2009). By using EEG, one method of the brainwave reaction is a non-invasive technique (Carlson & Millan, 2013) that can be effectively used in the physically disabled people interact with the environment (Singla & Haseena, 2013). The navigation technology had also developed in unmanned vehicles. Therefore, this research combines brainwave technique, automatic navigation system, and robot control to develop a brain-computer interface system to control the robot. In this way, the proposed system has a real-time video which shows the view from the camera of robot and an automatic navigation system embedded in the robot which can help robot to avoid the collision during movement. Users can control the robot by brainwave to browse the outside world without worrying about the obstacles on the way. Therefore, the integration of two technologies into brainwave control system makes operation easier and more user-friendly. People can capture the signals of brainwave with electroencephalography (EEG) which is a non-invasive method. EEG can record the change of brain potentials from the cortex. In 1965, Sutton first proposed the study of event related potentials (Sutton, Braren, & Zubin, 1965) and provided an objective, feasible and simple method for brainwave study. After then, event related potentials have become a very popular and widely used technique in observing brain 752

activity. Event related potentials also have been widely used in various cognitive neuroscience (Kutas & Hillyard, 1980). The experiment design of event-related potentials requires particular stimulus (e.g., text, pictures ... etc.) to flicker on the computer screen and repeats many times for each stimulus. After the stimulation, the reaction of brainwave with the same (type) stimulus was accumulated within a short time. The accumulation of the signal of the irritant obtains the effect of averaging signal and inhibiting noise. In this way, the accumulated brainwave is called the event-related potentials (ERPs) which can accurately reflect the characteristics of the stimulus. The ERPs consist of a series of positive and negative voltage deflections. It usually uses P to express positive wave and uses N to express negative wave, and is named with the latent time (the time gap between the stimulus appeared and the waves appeared, unit: ms) such as P300 (the positive wave about 300 ms after stimulus), and N200 (the negative wave about 200 ms after stimulus). Fig. 1 shows three different stimuli: target, non-target1 and non-target2. The subject focuses on ―Target‖ stimuli. After several rounds of stimulation, three event-related potentials of ERPs were evoked (Huang, Tai, Tian, & Sun, 2013).

Brain Computer Interface (BCI) first appeared in Jacques J. Vidal (Vidal, 1973), and cooperated with UCLA BCI laboratory to develop a brain-computer interface, which successfully used the visual stimulation to control cursor (Vidal, 1977). In the ensuing decades, their researches deeply affected biological cybernetics studies. Brain-computer interface systems interpret the brain signals from external stimuli that provide a communicate method to people with severe physically disabled. The subject selects a target from a number of objects to issue a command and achieve the purpose of communication (Hong, Guo, Liu, Gao, & Gao, 2009). As with other used to communicate or control man-machine interface, the brain-computer interface system also has input and output. But the difference is the input of brain-computer interface system is the user's brain physiology signal, and the output is the command for controlling devices. The basic process between input and output can be divided into four parts. Fig. 2 shows the basic operation flow of BCI: (1) Signal capture: capture the brain signals from the electrodes on scalp. (2) Digitization: amplify, filter and digitize the brain signals. (3) Signal processing: extract and classify the features of brain signals, and generate the final result. (4) Issue a command: convert the final result into a command for controlling the external facilities.

753

The brain-computer interface is mostly used for medical aids to help amyotrophic lateral sclerosis (ALS) and severely physically disabled to obtain various control ability such as brainwave controlled wheelchair (Khare, Santhosh, Anand, & Bhatia, 2011; Carlson, & Millán, 2013), brainwave controlled prosthetics (Guger, Harkam, Hertnaes, & Pfurtscheller, 1999; Hochberg et al., 2006; Müller-Putz & Pfurtscheller, 2008), brainwave controlled home environment (Gao, Xu, Cheng, & Gao, 2003), brainwave controlled biped robot (Li, Jaramillo, & Li, 2011; Bell, Shenoy, Chalodhorn, & Rao, 2008), brainwave controlled wheeled robot (Belluomo, Bucolo, Fortuna, & Frasca, 2012), and brainwave played chess games and puzzle games (Kaplan, Shishkin, Ganin, Basyul, & Zhigalov, 2013), etc. Robots have been used in many varied areas and developed multiple functions for different applications. Currently robots can perform weightlifting, grabbing the horizontal bar and pan, balancing, walking the ramp to aim and, shooting balloons and other tasks, the robot fully demonstrated the robustness of the balance (Yi, 2013). When the robot embedded the vision system, the immediate image would be transmitted to a remote computer, and provided users browsing the outside world and the abilities of remote control (Kelly et al., 2011). Robot technology has now met most needs of daily life, especially physical disabilities. However, there are still many problems for robot to browse outside for physical disabilities. Path planning is a fundamental problem of robot automatic navigation. Path planning can be divided into global and local. Local path planning is mainly applied to unknown environments, such as undersea, jungle, outer space, etc. The Tangent Bug algorithm (TBA) (Kamon, Rivlin, & Rimon, 1996) uses detecting the visual system on a car to find the shortest path of travel destination in the ring around the current location. If the robot encounters obstacles, it uses bypass mode to avoid obstacles. After reaching a local border points, the robot repeats doing environmental identification and determines the path until it reaches the destination. In this study, we combine EEG and automatic navigation system, providing multiple paths allow users to choose by EEG and letting the robot to walk on the selected path, and browsing the scene on the way. If the robot encountered obstacles during the path, the robot uses the TBA algorithm to navigate without colliding to the obstacles. In this way, the user can reduce the control burden on the robot, and easily browse the scene. 2. Methods 2.1 Brainwave Controlled Automatic Navigation Robot System

754

In this study, we design a brainwave controlled automatic navigation robot system. We applied the Borland C++ Builder to develop the brainwave controlled automatic navigation system. By using bluetooth, the system transfer the command generated by evoked potentials of brainwaves to the robot. The commands displayed on the brain-computer interface (BCI) which provides the user to issue one of commands by brainwaves, and the Tangent Bug algorithm (TBA) used in the collision avoidance system to keep the robot avoiding the obstacle on the way. The robot we used in this study is LEGO Mindstorms NXT. The Mindstorms NXT is a programmable robotics kit released by LEGO in late July 2006. Users can assemble NXT robot with different sensors to reach different missions and program it by using NXT–G. The NXT-G is a graphical programming environment that comes bundled with various programming languages such as NXT, visual C++, visual C#, Borland C++, LabVIEW, and so on. 2.2 System Design There are three user interfaces (Fig. 4) designed in the system, including path selection, movement control, and camera control. There are different icons in the three interfaces, and each icon corresponds to a function. Table 1 shows all icons and the corresponding functions. Each block of icon has a rectangular bar which flickers from left to right. Users only need to watch the block of the target function. After six rounds of stimulation, the system can determine the user‘s expectation with the largest N2P3 of ERPs and execute the corresponding command. In the center of the first interface is a real-time video image, and there are three (or more) different paths at the upper of interface. An user can use brainwave to choose the path they want to browse, and the system can automatically navigate on the way. After one path is selected by the user, the robot will move along the path. The system transfers to movement control interface (state) and user can watch the video image from the robot‘s camera without controlling the movement of robot. If the user was interested in something presented in the image, he/she can stop the movement of robot by using the brainwave. Then, the system changes the interface into the camera control interface (state). In this interface, user can control the camera to watch the desired image by the brainwave. After watching the image, user can change the state of robot into the movement control interface and let robot continue to move along the designated path until it reaches the destination. When the robot meets an obstacle on his way along the path, the collision avoidance system will automatically start to bypass the obstacle. In this way, the system can effectively help user to control the movement of robot and reduce the burden of the user. The user can browse the scene along the path with a relaxing and easy manner. However, when a user browses the video image, the stimuli of control buttons still continue to produce. For distinguishing the control signal generated by brainwaves whether it makes senses. We toke an exam of twenty people to collect the value of N2P3 when uses focus on a control button or not. The averaged value of N2P3 is 4.1±1.8(μV) when people focus on a control icon, and the value of N2P3 is -0.2±1.4 (μV) when users watch the video images but focus on the control button. Fig. 5 shows the different N2P3s between focus and unfocus situation. So we define a separated value 1.9(μV) between these two states. When the value of N2P3 of ERPs exceeds this threshold 1.9(μV), the system executes the command generated by brainwaves. Otherwise, it does nothing. Figure 6 shows the control flow of the brainwave controlled automatic navigation system.

755

756

2.3 EEG signal control Based on our previous research results (Huang, Tai, Tian, & Sun, 2013), the brain signal located at the electrode O1 (visual cortex area) can effectively generate visual evoked potentials. Producing visual event-related potentials needs to consider the number of stimuli and the stimulation schedule of control buttons. An appropriate stimulation schedule provides users quickly, easily and comfortably operate the system with a higher accuracy. Fig. 7 shows the stimulation timing of our design for controlling the camera. A flicker of stimuli needs 0.2 seconds, and one round of stimulation needs to flicker six times and requires 4.8 seconds. After six rounds of flickers, the computer sends a command to the robot.

2.4 Experimental Scene Design We built a miniature museum with size one square meter as the experiment scene (Figure 8), and let the robot browse scenery by using brainwaves in the museum. There are different paths to visit the museum from the starting point to the exit. Some paintings hanged on the walls along the path, so the uses can browse the paintings through the video image of the robot system. During visiting the path, there may be some obstacles appear on the path, and the robot will bypass the obstacle by the collision avoidance system. Each participant would select a path, and browse the paintings along the path until the robot reaches the exit and completes the experiment. 757

3. Experimental Design Our experiment had passed the review of institutional review board (IRB) and participants had signed the consent before experiment. Twenty healthy individuals between the ages of 21 and 29, no brain disease, relevant medical history or drug abuse, and had normal or corrected-to-normal vision, voluntarily participated in the experiment. Before the experiment, subjects practiced the automatic navigation robot system (Fig. 9) by using brainwaves for five minutes. When the automatic navigation robot system started, the system received the ERPs for each of control buttons. If the maximum N2P3 of ERPs exceeded the threshold, the corresponding control button would be executed (Fig. 10).

758

4. Results After completing the experiment, we evaluated the effectiveness of the proposed automatic navigation robot system in accordance with the participants‘ performance which including the time of completing task and the accuracy of brainwave control. As the experimental results (Table 2), most of the subjects can finish the navigation smoothly. The accuracy of our system is between 66% to 100%, and the average of correct rate is 82.7% with the standard deviation 10.96%.

759

5. Conclusions The BCIs have been developed more than forty years, and have applied to various applications such as wheelchairs, prosthesis, robot, and quadcopter control. In this research, we designed an automatic navigated robot by using BCI in field. By using the N2P3 of ERPs at visual cortex (O1) as the control signal, it can effectively control the robot in field. The subject can easily and freely operate the camera to browse the paintings in the miniature museum, and the navigation control and collision avoidance are embedded in the robot system to control the movement of robot. It causes fewer burdens than completely using brainwave to control the robot. After experimental testing by 20 subjects in field, all of the subjects can complete the task in 3 minutes, and the correct rate of brain control is between 66% to 100%, with the mean 82.7%. These preliminary results show that the designed automatic navigation robot system can effectively apply to people to browse the scene outside by robot. In the future, we will apply the system to severe disability to verify the practicality of the designed system.

1. 2. 3. 4.

5. 6.

7.

8. 9.

10.

11.

12.

13. 14.

6. References Bell, C. J., Shenoy, P., Chalodhorn, R., & Rao, R. P. N. (2008). Control of a humanoid robot by a noninvasive brain-computer interface in humans. J. Neural Eng., 5, 214–220. Belluomo, P., Bucolo, M., Fortuna, L., & Frasca, M. (2012). Robot Control through Brain-Computer Interface for Pattern Generation. Complex Systems, 20(3), 243-251. Carlson, T. & Millan, J. R. (2013). Brain–Controlled Wheelchairs: A Robotic Architecture. IEEE Robotics and Automation Magazine, 20(1), 65–73. Gao, X., Xu, D., Cheng, M., & Gao, S. (2003) A BCI-based environmental controller for the motion-disabled, IEEE Transactions on Neural Systems and Rehabilitation Engineering, 11, 137-140. Guger, C., Harkam, W., Hertnaes, C., & Pfurtscheller, G. (1999). Prosthetic control by an EEG-based brain-computer interface (BCI), Proc. AAATE', 99, 590–595. Huang, T. W., Tai, Y. H., Tian, Y. J., & Sun, K. T. (2013). The fastest BCI for Writing Chinese Characters Using Brain Waves. Global Congress on Intelligent Systems, 346-349, Hong Kong, December 3-4. Hochberg , L. R., Serruya, M. D. , Friehs, G. M. , Mukand, J. A. , Saleh, M. , Caplan, A. H. , …Donoghue, J. P. (2006). Neuronal ensemble control of prosthetic devices by a human with tetraplegia, Nature, 442,164–171. Hong, B., Guo, F., Liu, T., Gao, X., & Gao, S. (2009). N200-speller using motion-onset visual response. Clinical Neurophysiology, 120(9), 1658–1666. Kamon, I. Rivlin, E. Rimon. E. (1996). A new range-sensor based globally convergent navigation algorithm for mobile robots. Proceedings of IEEE International Conference on In Robotics and Automation, 429-435. Kaplan, A. Y., Shishkin, S. L., Ganin, I. P., Basyul, I. A., & Zhigalov, A. Y. (2013). Adapting the P300-based brain–computer interface for gaming: a review. IEEE Trans Comput Intel AI in Games, 5(2), 141–149. Kelly, A., Chan, N., Herman, H., Huber, D., Meyers, R., Rander, P., & Capstick, E. (2011). Real-time photorealistic virtualized reality interface for remote mobile robot control. The International Journal of Robotics Research, 30(3), 384-404. Khare, V., Santhosh, J., Anand, S., & Bhatia, M. (2011). Brain Computer Interface Based Real Time Control of Wheelchair Using Electroencephalogram, International Journal of Soft Computing and Engineering, 1(5), 41–45. Kutas, M. & Hillyard, S. A. (1980). Reading senseless sentences: Brain potentials reflect semantic incongruity. Science, 207, 203–205. Li, W., Jaramillo, C., & Li, Y. Y. (2011). A Brain Computer Interface Based Humanoid Robot Control System. IASTED International Conference on Robotics (Robo 2011), 760

15. 16.

17. 18.

19. 20. 21.

Pittsburgh, 390–396. Müller-Putz, G. R. & Pfurtscheller, G. (2008). Control of an electrical prosthesis with an SSVEP-based BCI, IEEE Transactions on Biomedical Engineering, 55(1), 361-364. Singla, R. & Haseena, B. A. (2013). BCI Based Wheelchair Control Using Steady State Visual Evoked Potentials and Support Vector Machines. International Journal of Soft Computing and Engineering (IJSCE), 3(3), 46-52. Sutton, S., Braren, M., Zubin, J., & John, E. R. (1965). Evoked-potential correlates of stimulus uncertainty. Science, 150(3700), 1187-1188. Tonin, L., Menegatti, E., Cavinato, M., D‘Avanzo, C., Pirini, M., Merico, A.,…Piccione, F. (2009). Evaluation of a robot as embodied interface for brain computer interface systems. International Journal of Bioelectromagnetism, 11(2), 97–104. Yi, S. (2013). Stable Walking of Qauadruped Robot by Impedance Control for Body Motion. International Journal of Control & Automation, 6(2), 99-110. Vidal, J. J. (1973). Toward direct brain-computer communication. Annu. Rev. Biophys. Bioeng, 2, 157–80. Vidal, J. J. (1977). Real-time detection of brain events in EEG. Proc. IEEE, 65(5), 633– 641.

761

ACEAIT-7347 High Quality Dual Imaging Technique Chi-Shiang Chana,b, Wen-Chuan Wuc, Jen-Ho Yangd and Pei-Yu Line,* Department of M-Commerce and Multimedia Applications, Asia University (Taiwan) b Department of Medical Research, China Medical University Hospital, China Medical University c Department of Computer Science and Information Engineering, Aletheia University d Department of Multimedia and M-Commerce, Kainan University e Department of Information Communication, and Innovation Center for Big Data and Digital Convergence, Yuan Ze University *Corresponding Author: [email protected] a

Abstract The proposed method produces two high quality camouflage images after embedding secret data into one original image. By referring two camouflage images, not only the secret data can be extracted but also the original image can be recovered. In this paper, LSB matching revisited is applied to embed two secret bits into one original pixel so as to produce two pixels for two camouflage images. Then, the produced pixels are checked to see whether the original pixels can be recovered. If not, the proposed method provides modifying rules that make fewer modifications to the original pixels. According to experimental results, the proposed method can produce camouflage images with high image quality. Keyword: Dual imaging technique, Reversible data hiding, LSB matching revisited 1. Introduction Data hiding embeds secret data into digital media, such as text, images, and videos. The simplest way to achieve data hiding is Least-Significant Bit (LSB) substitution. It replaces the least-significant bits of the cover pixels with secret data. However, the images gone through LSB substitution may be detected by attackers [7]. And, the images may be degraded largely. Many researchers [1, 3, 6, 7, 11] have been proposed to solve the problems. The recent method is LSB matching revisited [7]. The most brilliant point of this method embeds two secret bits to each pixel pair, but at most only one cover pixel needed to be modified by adding/subtracting one to/from its value. Owning to this property, LSB matching revisited can reduce the pixel modifications. All above-mentioned methods belong to non-reversible (irreversible) data hiding (non-RDH). The meaning of non-reversible data hiding is that the images are destroyed after embedding secret data. However, for some special purpose, cover images are expected to recover without any lost after extracting the secret data. To achieve this goal, reversible data hiding (RDH) was developed. RDH techniques can be divided into two main categories. The first one is Difference Expansion and the second one is the Histogram Shifting. The Difference Expansion proposed by Tian [10] expanded the distance difference between two pixels and embedded the secret bit into the difference. On the other hand, Histogram Shifting proposed by Ni [8] used statistic to analyze pixel distribution and the secret data was embedded into frequently occurring pixels. In the recent years, RDH technique has been moved to another area called Dual Image Reversible Data Hiding (DIRDH). It produces two similar camouflage images after embedding secret data into one original image. In the reversing phase, the secret data can be extracted from 762

two camouflage images and the original image can also be recovered from two camouflage images. DIRDH was first proposed by Chang et al. [4] to build a matrix through modulus function and Exploiting Modification Direction (EMD) technique. Then, left diagonal pixels and right diagonal pixels of the matrix were treated as camouflage pixels. In order to improve image quality, Chang et al. [2] took horizontal and vertical pixels as camouflage pixels. After that, Qin et al. [9] duplicated the original image into two copies. The first copy embedded secret data by EMD and the second copy embedded secret data according to the embedding ways used in the first copy. All mentioned DIRDH methods need to build a matrix (or a rule table) and the matrix will be needed to embed secret data into an original image. It goes without saying that the matrix will significantly affect the quality of camouflage images. In order to further improve the quality of the camouflage images, Lu et al. [5] eliminated the matrix. Their method recovered original pixels by calculating the average value of two camouflage pixels. In their embedding phase, LSB matching revisited is applied to embed the secret data into two copies of the original image. After embedding, the average value of two camouflage pixels is checked to see whether the value is equal to the original pixel. If not, ―modifying rules‖ are needed to make the average value of two camouflage pixels equal to original pixels. Because of these rules, Lu et al.‘s method is highly superior to the previous-mentioned DIRDH methods. However, after inspecting their ―modifying rules,‖ we found that there exists better way to revise these modifying rules. According to the revision, the quantity of the pixel modifications becomes less. Therefore, the proposed method can produce two camouflage images with high image quality. The rest of paper is organized as follows. In Section 2, LSB matching revisited method and Lu et al.‘s method are introduced. In Section 3, the detailed descriptions of the proposed method are presented. Section 4 shows the experimental results. Finally, conclusions are made in Section 5. 2. The Related Work Firstly, LSB matching revisited is introduced in Subsection 2.1. After that, Lu et al.‘s method applied LSB matching revisited to develop dual imaging-based reversible hiding technique. The technique was described in Subsection 2.2. 2.1 LSB Matching Revisited In this section, LSB matching revisited [7] is introduced. This method embeds two secret bits into an original pixel pair. Assume the original pixel pair is (O1, O2) and two secret bits are d1, d2. In this method, the first secret bit d1 is embedded into the least-significant bit (LSB) of the first original pixel O1. The second secret bit d2 comes from calculating the output of the function called Binary Function by feeding two original pixels (O1, O2) as its parameters. Binary Function is defined as follows:

The function LSB(.) represents the least-significant bit of input parameter. Two parameters x1 and x2 are two input pixels. To embed two secret bits, two original pixels (O1, O2) are modified into two camouflage pixels (C1, C2) such that LSB(C1) = d1 and F(C1, C2) = d2. The details of modifying two original pixels into two camouflage pixels (C1, C2) are drawn in a flowchart as shown in Fig. 1. According to Fig. 1, it shows that at most only one original pixel is needed to be modified by adding/subtracting one to/from its value and two secret bits are embedded. 763

2.2 Lu et al.‟s method This subsection introduces Lu et al.‘s dual imaging-based reversible hiding technique. In the embedding phase, two camouflage images are produced from an original image. Firstly all pixels in an original image are divided into non-overlapping pixel pairs. Assume a pixel pair in the original image is (O1, O2). The two-bit secret data S1 is embedded into (O1, O2) to form a pixel pair (C11, C12) for the first camouflage image. Meanwhile, the next two-bit secret data S2 is embedded into (O1, O2) to form another pixel pair (C21, C22) for the second camouflage image.

764

Among all possible modifications, the most important way in Lu et al.‘s method is that if C11 comes from subtracting one from O1, then Lu et al.‘s modifying rule alters C21 by adding two to C21 has been altered, the second camouflage pixel C22 should also be altered to make the extracted secret bits correct. The same situation occurs when C21 comes from subtracting one from O1. In this situation, C11 is altered by adding two to O1 and C12 should also be altered. The details of Lu et al.‘s modifying rules are listed in Table 2.

The final step checks whether any camouflage pixel is out of range, that is, smaller than 0 or larger than 255. If any one of the four camouflage pixels is out of range, no secret data would be embedded in them. To indicate this case, if O1 is smaller than or equal to 251, C11, C12, C21 and C22 would be set as O1 +4, O2 , O1 and O2, respectively. On the contrary, if O1 is larger than 251, C11, C12, C21 and C22 would be set as O1 -4, O2 , O1 and O2, respectively. The reason is that the largest distance between two camouflage pixels is 3. For example, the distance between C11 and C21 in case 7 in Table 2 is 3. Therefore, the pixel that can‘t be embedded secret bits is set O1 +4 or O1 -4. Fig. 2 shows an instance of Lu et al.‘s method.

765

In the recovering phase, if the distance between C11 and C21 is equal to 4, this means there is no secret data in these four camouflage pixels. To recover original pixel pair (R1, R2), it simply assigns C21 and C22 to R1 and R2, respectively. Then, the original pixel pair can be recovered. 3. The Proposed Method The proposed method not only increases the number of pixels that can be embedded secret bits but also provides high quality camouflage images. However, it doesn‘t divide original pixels into pairs but processes one pixel each time. Two secret bits d1 and d2 are embedded into an original pixel O to produce two camouflage pixels C1 and C2. The first camouflage pixel C1 is assigned to the first camouflage image while the second camouflage pixel C2 is assigned to the second camouflage image. The embedding procedure in this paper is the same as LSB Matching Revisited. The details are shown in Fig. 3.

In Fig. 3, the secret bits (11)2 are embedded into the first original pixel with value 0 by using LSB Matching Revisited. According to the embedding procedure, C1 and C2 are set as O1-1 and O1, respectively. It is trivial that O1-1 is smaller than 0. Therefore, the secret bits can not be embedded into O1. In this case, the secret bits are embedded into the next original pixel, that is, O2. More precisely, the proposed method checks each original pixel one by one to see whether the secret bits can be embedded into the current original pixel. If so, the secret bits are embedded. If not, the special values are assigned to two camouflage pixels. The assigned values will be described later. Owing to the above property, the proposed method can embed more bits than 766

Lu et al.‘s method. For example in Fig. 3, the second pixel of the original pixel can be used to embed secret bits in the proposed method. However, that cannot be embedded secret bits in Lu et al.‘s method in Fig. 2. Another point is that two camouflage pixels can be obtained through LSB Matching Revisited. However, the average value of two camouflage pixels may or may not the same as the original pixels. For example in Fig 3, if the secret bits (11)2 are embedded into the second pixel, C1 and to O2. To make them equal, some camouflage pixels should be modified. Table 3 lists all possible modification cases of camouflage pixels, their corresponding recovered pixel values and the proposed ways to modify those camouflage pixels.

Note that the cases in Table 3 are exactly the same as those in LSB Matching Revisited in Fig. 1. According to Table 3, the original pixel can be recovered among all cases except for Case 3. In order to recover original pixel correctly, the proposed method replaces C1 and C2 with O+1 and O-1, respectively. To demonstrate this, we should prove that two pixel pairs (O-1, O) and (O+1, O-1) can extract the same secret bits. It is trivial that LSB(O-1) is equal to LSB(O+1). It means two pixel pairs can extract the same first secret bit. Moreover, if the second secret bit extracted from (O+1, O-1) is exactly the same as that from (O-1, O), then the following equation should be true. where F represents the Binary Function in Formula (1), and parameters x1 and x2 are two input pixels. Equation (2) can also be rewritten as Equation (3) when the definition of the Binary Function is brought in.

From Equation (4) and (5), left part and right part of Equation (3) can derive the same results. That means Equation (3) is correct. 767

Since Equation (3) is correct, the proposed method replaces camouflage pixels C1 and C2 in is equal to the original pixel value O. Moreover, the most valuable point in Table 3 is that the largest modifying distance for a pixel is 1. However, in Lu et al.‘s method, the largest modifying distance for a pixel is 2. Because of less pixel modification, the proposed method has better image quality of the camouflage images. The remained steps are the same as Lu et al.‘s method. That is checking whether any camouflage pixel value is out of range. By observing the distance between any two camouflage pixels, the largest distance is two. For example, the distance between C1 and C2 in Case 3 is 2. Therefore, if a camouflage pixel is out of range, C1 and C2 would be set as O +3 and O when original pixel value O is smaller than or equal to 252. On the other hand, C1 and C2 are set as O -3 and O when original pixel value O is larger than 252. The recovering phase of the proposed method is exactly the same as Lu et al.‘s method.

4. Experimental results The experimental results are demonstrated in this section. In the experiment, real images were used to perform the Lu et al.‘s method and the proposed method. The original images were Lena, Pepper, Barb, Boat, Plane, Tiffany, Toys and Baboon as shown in Fig. 4. The secret data were a sequence of random bit string generated from random number generator. The way to estimate the quality of camouflage images was the peak signal to noise ratio (PSNR), calculated from the following formula:

MSE represents the mean squared error. It comes from calculating the square errors of all pixels. The symbols α(I, J) and β(I, J) represent the pixel values at the position (I, J) in the camouflage image and the original image, respectively. The symbols w and h represent the pixel numbers for the width and the height of the image, respectively. Table 4 shows the experimental results. The values in Table 4 are the PSNR values between the original images and camouflage images. Note that both Lu et al.‘s method and the proposed method produce two camouflage images from an original image. Therefore, PSNR(1) 768

represents the PSNR values between the original images and the first camouflage images. And, PSNR(2) represents the PSNR values between the original images and the second camouflage images. According to experimental results in Table 4, the PSNR values in the proposed method are significantly higher than those in Lu et al.‘s method. These experimental results can be expected because the modifying quantity of the proposed method is less than that of Lu et al.‘s method as shown in Table 2. Moreover, both Lu et al.‘s method and the proposed method can recover identical original images. Therefore, there are no PSNR values of the recovered images in Table 4.

5. Conclusions In this paper, the proposed method provides a revision of Lu et al.‘s modifying rules. Through the revision, the largest modifying distance for a pixel becomes only one. However, the largest modifying distance for a pixel is two in Lu et al.‘s method. This is the reason why the proposed method has better performance than Lu et al.‘s method. According to the experimental results, the PSNR values of the proposed method are larger 1 dB than those of Lu et al.‘s method. That means the quality of the camouflage images of the proposed method is better than that of Lu et al.‘s method. All in all, the proposed method is superior to Lu et al.‘s method. Acknowledgment This research was supported by the National Science Council, Taiwan, under contract No. MOST 104-2410-H-468-011 and No. NSC102-2221-E-155-035-MY3.

[18] [19]

[20]

[21]

[22] [23]

[24] [25]

References Chan, C. S. 2009. ―On Using LSB Matching Function for Data Hiding in Pixels, Fundamenta Informaticae, 96, 49-59. Chang, C. C., Chou, Y. C., and Kieu, T. D., Information Hiding in Dual Images wit Reversibility, Proceedings of the Third International Conference on Multimedia and Ubiquitous Engineering, 145-152, 2009. Chang, C. C., Hsiao, J. Y. and Chan, C. S. 2003. Finding Optimal Least-significant-bit Substitution in Image Hiding by Dynamic Programming Strategy, Pattern Recognition, 36 (7), 1583-1595. Chang, C. C., Kieu, T. D. and Chou, Y. C., Reversible Data Hiding Scheme using Two Steganographicimages, Proceedings of the 2006 International Conference on Business and Information, 1-4, 2007. Lu, T. C., Tseng, C. Y. and Wu, J. H. 2015. Dual Imaging-based Reversible Hiding Technique using LSB matching, Signal Processing, 108, 77-89. Liao, X., Wen, Q. Y. and Zhang, J. 2011. A Steganographic Method for Digital Images with Four-pixel Differencing and Modified LSB Substitution, Journal of Visual Communication and Image Representation, 22 (1), 1-8. Mielikainen, J. 2006. LSB Matching Revisited. IEEE Signal Processing Letters, 13 (5), 285-287. Ni, Z., Shi, Y. Q., Ansari, N. and Su, W. 2006. Reversible Data Hiding, IEEE 769

Transaction Circuits System Video Technology, 16 (3), 354-362. [26] Qin, C., Chang, C. C., Chou and Hsu, T. J. 2014. Reversible Data Hiding Scheme based on Exploiting Modification Direction with Two Steganographic Images, Multimedia Tools and Application, http://dx.doi.org/10.1007/s11042-014-1894-5. [27] Tian, J. 2003. Reversible Data Hiding using a Difference Expansion, IEEE Transaction Circuits System Video Technology, 13 (8), 890-896. [28] Zhang, T., Li, W., Zhang, Y., Zhang, E. and Ping, X. 2010. Steganalysis of LSB Matching based on Statistical Modeling of Pixel Difference Distributions, Information Sciences, 180 (23), 4685-4694.

770

ACEAIT-7317 Comments on an Authentication Model of IoT Control System Ya-Fen Changa, Ya-Ling Loa, Wei-Liang Taib,* Department of Computer Science and Information Engineering, National Taichung University of Science and Technology, Taiwan b,* Department of Information Communications, Chinese Culture University, Taiwan E-mail: [email protected] a

Abstract Recently, Yang et al. enhanced the IoT control system by adding the backup terminal device, the status monitor device and the alarm module for availability and security. Service is provided by the backup terminal device when the main terminal device malfunctions or its status is not correct. Their authentication model uses hash function to protect identities of terminal devices and the challenge-response mechanism to resist synchronization attack, and parameters for authentication are stored in the distributed database for reliability. They claimed that their model ensured mutual authentication. After analyzing their model, we find that their authentication model possesses the following drawbacks: (1) In their model, mutual authentication is between the database and the terminal device. The reader device is not authenticated. (2) Default search is needed when the database wants to find the matched terminal device‘s identity. This approach places a heavy burden on the database. (3) An attacker can send a forged device-switching request because the origin of this request is not authenticated. In this paper, we implicitly indicate the found drawbacks that Yang et al.‘s authentication model of IoT control system suffers from in detail. Keywords: the Internet of Things (IOT), authentication, authentication model 1. Introduction The Internet of Things (IoT) brings a conceptual breakthrough, where devices are connected to realize various applications. To make management efficient and apply IoT to automatic control such as industrial production and smart living, the IoT control system is proposed. Because of IoT‘s properties, how to ensure the security of the IoT control system becomes an urgent and tough issue. To provide a high security level, the origin and integrity of transmitted data need to be verified at first. Many cryptographic authentication schemes are proposed for a high security level. Fortunately, terminal devices in the IoT control system commonly possess stronger computational and storage abilities such that the IoT control system possesses features different from other applications. In December 2013, in order to ensure the availability of service, Yang et al. added the backup terminal device, the statues monitor device and an alarm mechanism in the IoT control system (Yang, Pang & Zhang, 2013). In their system, when the main terminal device malfunctions or its status is incorrect, service is provided by the backup terminal device. Yang et al. also proposed an authentication model for the IoT control system to ensure the security of service. Before proposing their authentication model, they analyzed RFID (radio-frequency identification) authentication schemes that are computationally efficient for computational efficiency and classified them into two categories: one-way authentication and mutual authentication. One-way authentication protocols contain hash-lock protocol (Dixit, Verma & Singh, 2011), randomized hash-lock protocol (Weis, Sarma, Rivest & Engels, 2004) and 771

hash-chain protocol (Henrici & Muller, 2004), which are based on hash function. These one-way authentication protocols only authenticate the server-side, the card reader, for RFID devices. Because RFID devices are not authenticated by the server side, an illegal card reader can control the authentication process and further get confidential information. Mutual authentication protocols contain the hash-based ID variation protocol (Lee, Hwang, Lee & Lim, 2005), link aggregation control protocol (LACP) protocol (Juels & Pappu, 2003) and re-encryption protocol (Golle, Jakobsson, Juels & Syverson, 2004). Though these protocols authenticate both sides, they still have some security threats such as synchronization problem and information leakage. In the hash-based ID variation protocol and the LACP protocol, the security threats result from their own characteristics; in re-encryption protocol, the security threats occur because the fixed identity is used. Later, Rhee et al. proposed a challenge-response-based RFID authentication protocol for the distributed database environment (Rhee, Kwak, Kim & Won, 2005) to overcome the security flaws which the above protocols suffer from. In 2012, Cho et al. proposed a hash-based RFID tag mutual authentication protocol (Cho, Jeong & Park, 2015). In 2013, Safkhani et al. showed that Cho et al.‘s scheme suffers from three attacks: de-synchronization attack, tag impersonation attack and reader impersonation attack (Safkhani, Peris-Lopez, Hernandez-Castro & Bagheri, 2014). As a result, Yang et al. proposed an efficient authentication model of IoT based on Rhee et al.‘s RFID authentication scheme (Yang, Pang & Zhang, 2013) even though there are still some RFID authentication schemes are proposed after Rhee et al.‘s scheme. In Yang et al.‘s authentication model, the terminal device identity is protected by the hash function. Moreover, the challenge-response mechanism is used to resist the synchronization attack, and verified parameters are stored in the distributed database. Yang et al. also claimed that their authentication model provided mutual authentication. After analyzing their authentication thoroughly, we find their authentication model has the following weaknesses: (1) In their model, mutual authentication is between the database and the terminal device rather than between the reader device and the terminal device, where the reader device is the service access port. Thus the reader device is not authenticated in their model. (2) When the database wants to find the matched terminal device‘s identity, the database needs to check and compute the parameters with all terminal devices‘ identities. This approach places a heavy burden on the database. (3) In their authentication model, an attacker can send a forged device-switching request to make the IoT control system malfunction because the status monitor device does not authenticate the origin of the received device-switching request in the alarm phase. In this paper, we implicitly indicate the found drawbacks in detail. The rest of this paper is organized as follows. Section 2 reviews Yang et al.‘s authentication model of IoT control system. The analysis of Yang et al.‘s authentication model of IoT control system is made in Section 3. At last, some conclusions are drawn in Section 4. 2. Review of Yang et al.‟s authentication model of IoT control system Yang et al.‘s authentication model of IoT control system, illustrated in Figure 1, is composed of five phases: initialization phase, main device authentication phase, main device alarm phase, device switching and backup device authentication phase, and backup device alarm phase. Notations used in Yang et al.‘s authentication model are listed in Table 1, and the details are as follows.

772

2.1 Initialization phase Every terminal device in the IoT control system has its own unique identity. This device identity is stored in the device itself and the distributed database at the server. Operation status information of all terminal devices is stored in the server database for the server to verify them. Both the main terminal device and the backup terminal device are in hot standby status, and the status monitor device has a connection to the main terminal device. After initialization, the reader device can send an authentication request. 2.2 Main device authentication phase When the reader device attempts to access the terminal device, this phase will be performed. The details are as follows: Step 1: R sends an authentication request to Dc. 773

Step 2: R generates a random number Rr and sends an authentication request {Rr, request} to Dm. Step 3: Dc gets the main terminal device Dm‘s status information S and computes Hs(S). Step 4: Dm generates a random number RD, computes Hi(I||Rr||RD) with its identity I, and sends {Hi(I||Rr||RD), RD} to R. Step 5: The status monitor device Dc sends {Hs(S)} to R. Step 6: R sends {Hi(I||Rr||RD), RD, Hs(S), Rr}to BD. Step 7: BD computes Hi(Ii||Rr||RD) and Hs(Si) for all (Ii, Si)‘s stored in its database and checks if there exists (Ii, Si) satisfying both Hi(Ii||Rr||RD)=Hi(I||Rr||RD) and Hs(Si)=Hs(S). If (Ii, Si) does exist, BD computes Hi(Ii||RD); otherwise, the main terminal device authentication fails, and main device alarm phase is executed. Step 8: BD sends {Hi(Ii||RD)} to R. Step 9: When R receives {Hi(Ii||RD)} from BD, R sends {Hi(Ii||RD)} to Dm. Step 10: After Dm receives {Hi(Ii||RD)} from R, Dm computes H(I||RD) and checks if the computation result is equal to Hi(Ii||RD). If they are equal, the main terminal device and the back-end server authenticate each other successfully. . 2.3 Main device alarm phase Main device alarm phase will be executed when the main terminal device authentication fails. It denotes that no (Ii, Si) satisfying both Hi(Ii||Rr||RD)=Hi(I||Rr||RD) and Hs(Si)=Hs(S) is found after BD computes Hi(Ii||Rr||RD) and Hs(Si) for all (Ii, Si)‘s stored in its database. If Hi(Ii||Rr||RD)Hi(I||Rr||RD), it denotes Dm is illegal and BD sends the alarm signal A1 to the system alarm module A. If Hi(Ii||Rr||RD)=Hi(I||Rr||RD) and Hs(Si)Hs(S), it denotes that the main terminal device Dm is legal, and its status is probably incorrect because of some physical damages. Under the second condition, BD sends the alarm signal A2 to the system alarm module A. 2.4 Device switching and backup device authentication phase After sending the main device alarm signal A2, BD sends a device-switching signal to R. Then, R sends the device-switching signal to the status monitor device Dc to switch devices. After getting the device-switching signal sent from BD, R immediately terminates the connection with the main terminal device and switches the connection to the backup terminal device. After devices are switched, the remaining authentication process of device switching and backup device authentication phase is the same as that of main device authentication phase except the involved terminal device is Db and backup device alarm phase is executed when no (Ii, Si) satisfying both Hi(Ii||Rr||RD)=Hi(I||Rr||RD) and Hs(Si)=Hs(S). 2.5 Backup device alarm phase Backup device alarm phase will be executed when the backup terminal device authentication fails. It denotes that no (Ii, Si) satisfying both Hi(Ii||Rr||RD)=Hi(I||Rr||RD) and Hs(Si)=Hs(S) is found after BD computes Hi(Ii||Rr||RD) and Hs(Si) for all (Ii, Si)‘s stored in its database. If Hi(Ii||Rr||RD)Hi(I||Rr||RD), it denotes Db is illegal and BD sends the alarm signal A3 to the system alarm module A. If Hi(Ii||Rr||RD)=Hi(I||Rr||RD) and Hs(Si)Hs(S), it denotes that the main terminal device Db is legal, and its status is probably incorrect because of some physical damages. Under the second condition, BD sends the alarm signal A4 to the system alarm module A.

774

3. Analysis of Yang et al.‟s authentication model of IoT control system Yang et al. enhanced the IoT control system by adding a backup terminal device, the status monitor device and the alarm module for availability and security. In their system, service is provided by the backup terminal device when the main terminal device malfunctions or its status is not correct. Their authentication model uses hash function to protect identities of terminal devices and the challenge-response mechanism to resist synchronization attack, and parameters for authentication are stored in the distributed database for reliability. They claimed that their model ensured mutual authentication. After analyzing their authentication model, we find that it possesses the following three drawbacks: (1) The reader device is not authenticated. As shown in main device authentication phase and backup device authentication phase, the reader device R only relays transmitted data between the terminal device and the back-end database. Though the random number Rr is generated by R, R is not capable of authenticating other entities. In their authentication model, mutual authentication is between the database and the terminal device. The reader device is not authenticated. This drawback makes terminal devices might be accessed by an unauthorized reader device such that anyone can access terminal devices at will. (2) Default search is needed when the back-end database BD attempts to find the matched (Ii, Si). As shown in main device authentication phase and backup device authentication phase, the terminal device‘s identity is not transmitted for security. As a result, BD needs to compute Hi(Ii||Rr||RD) and Hs(Si) for all (Ii, Si)‘s stored in its database and checks if there exists (Ii, Si) satisfying both Hi(Ii||Rr||RD)=Hi(I||Rr||RD) and Hs(Si)=Hs(S). This approach places a heavy burden on BD. (3) The origin of the device-switching request is not authenticated. As shown in main device alarm phase and backup device alarm phase, BD sends a device-switching signal to R. Then, R sends the device-switching signal to the status monitor 775

device Dc to switch devices. After getting the device-switching signal sent from BD, R immediately terminates the connection with the main terminal device and switches the connection to the backup terminal device. Dc does not authenticate the origin of the received device-switching request. That is, an attacker can make the system malfunction by sending a forged device-switching signal. 4. Conclusion In this paper, we review Yang et al.‘s authentication model of IoT control system and analyze it thoroughly. We find it suffers from three drawbacks. First, mutual authentication is between the database and the terminal device, and the service access port, the reader device, is not authenticated. Second, default search is needed when the database wants to find the matched terminal device‘s identity. Third, an attacker can send a forged device-switching request because the origin of this request is not authenticated. How to overcome the found drawbacks and design a secure and efficient authentication model for the IoT control system becomes an urgent issue. Acknowledgement This work was supported in part by Ministry of Science and Technology under the Grants MOST 104-2221-E-034-004- and MOST 104-2221-E-025-006-. References Yang, J. C., Pang, H. & Zhang, X. (2013). Enhanced mutual authentication model of IoT. The Journal of China Universities of Posts and Telecommunications, 20, 69-74. Dixit, V., Verma, H. K. & Singh, A. K. (2011). Comparison of various Security Protocols in RFID. International Journal of Computer Applications, 24(7), 17-21. Weis, S. A., Sarma, S. E., Rivest, R. L. & Engels, D. W. (2004). Security and privacy aspects of low-cost radio frequency identification systems. Lecture Notes in Computer Science of Security in Pervasive Computing, 2802, Springer-Verlag, Boppard, 201-212. Henrici, D. & Muller, P. (2004). Hash-based enhancement of location privacy for radio-frequency identification devices using varying identifiers. Proceedings of the Second IEEE Annual Conference on Pervasive Computing and Communications Workshops, 149-153. Lee, S. M., Hwang, Y. J., Lee, D. H. & Lim, J. I. (2005). Efficient authentication for low-cost RFID systems. Lecture Notes in Computer Science of Computational Science and Its Applications–ICCSA 2005, 3480, Springer-Verlag, 619-627. Juels, A. & Pappu, R. (2003). Squealing Euros: privacy protection in RFID-enabled banknotes. Lecture Notes in Computer Science of Financial Cryptography, 2742, Springer-Verlag, 103-121. Golle, P. M., Jakobsson, Juels, A. & Syverson, P. (2004). Universal re-encryption for mixnets. Lecture Notes in Computer Science of Topics in Cryptology–CT-RSA 2004, 2964, Springer-Verlag, 163-178. Rhee, K., Kwak, J., Kim, S. & Won, U. (2005). Challenge-response based RFID authentication protocol for distributed database environment. Lecture Notes in Computer Science of Security in Pervasive Computing, 3450, Springer-Verlag, 70-84. Cho, J.-S., Jeong, Y.-S. & Park, S. O. (2015). Consideration on the brute-force attack cost and retrieval cost: A hash-based radio-frequency identification (RFID) tag mutual authentication protocol. Computers & Mathematics with Applications, 69, 58-65. Safkhani, M., Peris-Lopez, P., Hernandez-Castro, J. C. & Bagheri, N. (2014). Cryptanalysis of the Cho et al. protocol: A hash-based RFID tag mutual authentication protocol. Journal of Computational and Applied Mathematics, 256, 571-577. 776

777

ACEAIT-7318 Comments on Kuo et al.‟s Anonymous Mobility Network Authentication Scheme a

Ya-Fen Changa, Min-How Hsua, Wei-Liang Taib,* Department of Computer Science and Information Engineering, National Taichung University of Science and Technology, Taiwan b,* Department of Information Communications, Chinese Culture University, Taiwan E-mail: [email protected]

Abstract With the rapid advances of the network technologies and mobile devices, users can easily use their mobile devices to access Internet services through wireless networks. In 2014, Kuo et al. proposed a mobility network authentication scheme and claimed their scheme was efficient and secure. Unfortunately, we find that their scheme has two weaknesses. First, it cannot resist man-in-the-middle attack. Second, it suffers from the synchronization problem in password change phase. In this paper, the found weaknesses will be shown in detail. Keyword: authentication, mobile networks, elliptic curve, man-in-the-middle attack, synchronization problem 1. Introduction Mobile devices and mobile networks play an important role in our daily life. There exist three entities in mobility networks, mobile user, home agent and foreign agent. A mobile user needs to register with the home agent at first. Later, the registered mobile user can access the network service provided by the home agent directly or a foreign agent with the home agent‘s help. Although mobile devices and mobile networks provide great convenience, security becomes a tough and urgent issue because information is transmitted via a public but insecure channel. That is, an attacker can easily eavesdrop or intercept the transmitted data. To ensure security, many authentication protocols for mobility networks are proposed. In 2004, Zhu and Ma (Zhu & Ma, 2004) proposed a new authentication scheme with anonymity for wireless environments based on the hash function and smart cards. After Lee et al. analyzed Zhu and Ma‘s scheme, Lee et al. (Li & Chang, 2012) found that Zhu and Ma‘s scheme does not provide mutual authentication and cannot resist the forgery attack. In 2006, Lee et al. (Lee, Hwang & Liao, 2006) proposed an enhancement to improve Zhu and Ma‘s scheme for wireless networks. Unfortunately, Chang et al. analyzed Lee et al.‘s scheme (Lee, Hwang & Liao, 2006) and pointed out that Lee et al.‘s scheme still suffers from the forgery attack and proposed an improved scheme (Chang, Lee & Chiu, 2009). In 2014, Kuo et al. (Kuo, Wei & Cheng, 2014) pointed out that Chang et al.‘s scheme cannot ensure anonymity for mobile users and proposed an improved scheme. They claimed that their scheme could ensure efficiency and security in mobility networks and provide anonymity for mobile users. After analyzing Kuo et al.‘s scheme, we find their scheme suffers from two weaknesses. First, the scheme cannot resist the man-in-the-middle attack when the mobile user and the foreign agent negotiate the session key in authentication and establishment of the session key phase. An attacker can easily impersonate the mobile user and negotiate the session key with the foreign agent. Second, their scheme is vulnerable to the synchronization problem. The attacker can make a legal unable to be authenticated by simply modifying the transmitted data in password change phase. 778

The rest of this paper is organized as follows. Section 2 reviews Kuo et al.‘s scheme. The found security weaknesses which Kuo et al.‘s scheme suffers from are shown in Section 3. Finally, conclusion is given in Section 4. 2. Review of Kuo et al.‟s scheme In this section, we review Kuo et al.‘s scheme that consists of four phases: (1) registration phase, (2) authentication and establishment of the session key phase, (3) update session key phase, and (4) password change phase. Before reviewing Kuo et al.‘s scheme, the used notations are listed in Table 1. The details are as follows.

2.1 Registration phase In this phase, in order to access the roaming service, MU needs to register with HA at first. This phase is depicted in Figure 1, and the details are as follows. Step 1: MU selects his/her secret key pMU and identity IDMU. Step 2: MU computes PWMU=h(IDMU||pMU). Step 3: MU sends IDMU and PWMU to HA via a secure channel. Step 4: When HA receives {PWMU, IDMU} from MU, HA checks if IDMU does not exist. If it does hold, HA generates a random nonce RMU and the secret key pHA-MU for MU. Step 5: HA computes U=h(pHA-MU||RMU), W=PWMURMU and V=RMUpHA-MU. Step 6: HA stores {IDHA, W, V, h(·)} into a smart card and sends it to MU via as secure channel. Step 7: HA stores {U, PWMU, pHA-MU} into HA‘s database.

779

2.2 Authentication and establishment of the session key phase After registering with HA, MU can be authenticated anonymously via FA while roaming and negotiates a session key with FA in this phase. This phase is depicted in Figure 2, and the details are as follows.

780

2.3 Update session key phase After being authenticated by HA via FA, MU can update the session key with FA while staying in the same FA continuously. For clarity, we assume that MU has already updated the session key i times. This phase is depicted in Figure 3, and the details are as follows.

781

2.4 Password change phase MU can change his/her password with HA via a public channel. This phase is depicted in Figure 4, and the details are as follows.

782

3. Analysis on Kuo et al.‟s scheme After analyzing Kuo et al.‘s scheme, we find that two security drawbacks exist in this scheme. The details are as follows. 3.1 Man-in-the-middle attack In authentication and establishment of the session key phase, an attacker can mount the man-in-the-middle attack to impersonate MU and establish the session key with FA. The process of the man-in-the-middle attack is shown in Figure 5, and the details are as follows.

783

As a result, the attacker can mount the man-in-the-middle attack successfully and negotiate the session key with FA.

3.2 The synchronization problem

784

4. Conclusions In this paper, we point out that Kuo et al.‘s scheme suffers from two weaknesses. First, it cannot resist the man-in-the-middle attack when FA and MU negotiate their session key in authentication and establishment of the session key phase. Via the given approach, the attacker can easily impersonate MU and negotiate the session key with FA. Second, it is vulnerable to the synchronization problem. The attacker can make a legal unable to be authenticated by simply modifying the transmitted h3 in password change phase. According to the found security weaknesses, how to design a mobility network authentication scheme ensuring security and efficiency at the same time is still an urgent issue. Acknowledgement This work was supported in part by Ministry of Science and Technology under the Grants MOST 104-2221-E-034-004- and MOST 104-2221-E-025-006-. References Suzukiz, S. & Nakada, K. (1997). An authentication techinque based on distributed security management for the global mobility network. IEEE Jounal Selected Areas in Communications, 15(8), 1608-1617. Buttyan, L., Gbaguidi, C., Staamann, S., & Wilhelm, U. (2000). Extensions to an authentication technique proposed for the global mobility network. IEEE Transactions on Communications, 48(3), 373-376. Tzeng, Z. J. & Tzeng, W. G. (2001). Authentication of mobile users in third generation mobile systems. Wireless Personal Communications, 16(1), 35-50. Hwang, K.F. & Chang, C.C. (2003). A self-encryption mechanism for authentication of roaming and teleconference services. IEEE Transactions on Wireless Communications, 2(2), 400-407. Zhu, J. & Ma, J. (2004). A new authentication scheme with anonymity for wireless environments. IEEE Transactions on Consumer Electronics, 50(1), 230-234. Li, Y. F. & Chang, Y. F. (2012). A security flaw of abilinear-pairing-based electronic cash scheme with trustee-based anonymity revocation. In: Proceedings of the Sixth International Conference on Genetic and Evolutionary Computing, 25-28. Lee, C. C., Hwang, M. S., & Liao, I. E. (2006). Security enhancement on a new authentication scheme with anonymity for wireless environments. IEEE Transactions on Industrial Electronics, 53(5), 1683-1687. Chang, C.C., Lee, C.Y., & Chiu, Y.C. (2009). Enhanced authentication scheme with anonymity for roaming service in global mobility networks. Computer Communications, 32(4), 611-618. Kuo, W. C., Wei, H. J., & Cheng, J. C. (2014). An efficient and secure anonymous mobility network authentication scheme. Journal of Information Security and Applications, 19(1), 18-24.

785

ACEAIT-7368 Using Kalman Filter for Improved Ego-Motion Estimation Yu-Hsiang Chang, Ting-Hsiang Huang, Chen-Chi Chuang, Chia-Yen Chena, Chia-Chen Kuob a Dept. of Computer Science and Information Engineering National University of Kaohsiung, Taiwan E-mail address: [email protected] b

National Center for High-performance Computing, National Applied Research Laboratories, Taiwan

Abstract In this paper, we propose a method to improve ego-motion estimation. In the proposed method, feature points from input binocular stereo image pairs are detected using the Speeded Up Robust Features method and descriptor vectors are generated for the features. Semi-Global Matching is then performed to ensure the quality and quantity of the matching features, which are used in conjunctions with camera parameters to calculate the 3D coordinates for the image features. Kalman filter and adaptive matching are the used to perform feature tracking. The calculated 3D coordinates, as well as the 2D projections and camera parameters are used in Efficient Perspective-n-Point Camera Pose Estimation to find the camera motions. Finally, Sparse Bundle Adjustment is used to perform optimization of the estimated motions. Keywords: Ego-Motion Estimation, Visual Odometry, Feature Tracking, Epipolar Geometry, Outlier Removal, Kalman filter, SURF, SBA 1. Introduction In recent years, 3D related technologies and researches have received increased attention in a multitude of disciplines and applications, such as engineering, medical, archaeology, entertainment industries and so on. More and more fields have developed close associations with 3D related technologies. Among the 3D technologies, computer vision plays an important role in facilitating to understand the 3D world from 2D images. In particular, the determination of the system's own movements, or often referred to as ego-motion estimation, is a key step in understanding the location of the system with respect to the environment. This work is aimed at using the binocular stereo vision setup to obtain accurate ego-motion estimation. When the proposed method is applied to 3D data acquisition devices mounted on mobile platforms, the system will be able to estimate its own path, such that partial 3D data acquired at various positions can be aligned and merged to form a textured 3D model. In this manner, the system will be able to obtain more accurate 3D model in a more effective manner. This paper is structured as follows; section 2 provides related literature on ego-motion estimation, section 3 describes the proposed method, experimental results are presented in section 4, followed by the conclusions and future work in section 5. 2. Related Work The goal of this research is to improve the accuracy in ego-motion estimation. The method is developed to assist various 2D or 3D data acquisition devices which lack motion sensors. By incorporating the proposed method with the acquired data, the devices are able to determine their own positions within the environment. The acquired data can also be aligned or registered 786

using the calculated ego-motion. In this work, we use binocular stereo vision to acquire input images for ego-motion estimation. The steps for ego-motion estimation are shown in Fig. 1. First of all, image pairs are acquired sequentially by the cameras as the system moves within the scene. Features are then detected and matched in image pairs and then tracked temporally in subsequent images. From the displacement of corresponding feature points and camera parameters, transformations are calculated between subsequent image pairs. The transformations are collected and optimized to obtain the overall ego-motion estimation of the system.

One of the earliest approaches for ego-motion estimation uses 3D data for alignment. The Iterative Closest Point, or ICP algorithm is one method for aligning 3D points (Besl & McKay, 1992), and became popular after D. Nister et al. proposed the idea of visual odometry in 2004 (Nister & Bergen, 2004). In 2007, A. Comport et al. proposed an approach for ego-motion estimation based on 3D data points and 2D images (Comport, Malis, & Rives, 2007). In addition, many recent approaches integrate data from depth or location sensing devices, such as LiDAR or GPS, and computer vision towards ego-motion estimation. For example, in 2011, L. Wei et al. incorporate GPS and image data to perform ego-motion estimation of vehicles and mapping. Integrating various ego-motion estimation related research work, D. Scaramuzza and F. Fraundorfer provided a comprehensive survey on ego-motion estimation in 2010 (Scaramuzza & Fraundorfer, 2011, 2012). In 2014, J. H. Zhang proposed a method to improve the accuracy in feature matching based on epipolar geometry and adaptive matching (Zhang, 2014). This work investigates the application of Zhang's paper and improves the method by using Kalman filter to predict the intersection of subsequent feature tracking segment, such that feature matching and tracking can be further improved, resulting in more accurate ego-motion estimation of the system. 3. Proposed Method This section describes the stages of ego-motion estimation, as well as describes the proposed method of using adaptive constraints and Kalman filter to predict tracking direction and improve the accuracy of ego-motion estimation. The stages of ego-motion estimation are shown in Fig.2.

787

Once the features have been matched, Kalman filter and adaptive constraints are applied to feature point tracking for adjustment. The tracked matches are used to project 3D points onto 2D images to estimate the motion. Finally, bundle adjustment is used to obtained the optimized motion. 3.1 Adaptive feature tracking Feature tracking is also referred to as temporal feature matching. In this work, we propose a adaptive matching method (Zhang, 2014) using Kalman filtering to improve the accuracy, the framework of the proposed method is provided in Fig. 3.

In Fig. 3, Kalman filter is used to predict Track(t-n) to Track(t-1) and obtaining the intersection of the tracking segments. Using the intersection, we can calculate the angular constraint and the length constraint for feature matching. These two constraints are combined to provide adaptive constraint and obtain Track(t). 3.1.1 Adaptive constraint coefficient 788

There are many subtle changes between subsequent images which may influence the result of prediction. Therefore, when we are determining the constraint, we should not be overly strict in case the harsh constraint causes failed matches. The flexibility of constraint can also help to maintain the number of possible matches to keep tracking stable.

3.1.2 Angular and length constraints During the system's movement, if the Z axis changes, that is, the system moves forward or backward, then the epipolar line extending from the features tracked in the images will

The length of temporal feature matching is often associated with the depth of the feature points. The closer the feature points to the camera, the longer the matching length, inversely, the when the feature points are further away, the matching length is shorter. Therefore, we use this characteristic and use the feature points' disparities as the second constraining condition in feature matching. To determine the constraints for feature matching, we combine the above two constraints. First of all, find the standard values for the direction and length of the feature points to be matched, then obtain the difference tolerance for the two parameters and determine if the current match is within the standard values' tolerated difference.

789

3.1.3 Feature matching and outlier removal During temporal feature matching, feature matching calculates the relationship between feature vectors and perform matches accordingly. The search approach for feature matching is analogous to searching over a fan-shaped region. First of all, the sum of squared differences (SSD) is used to calculate the cost of matching then the cost is maximized for features which do not meet the criteria with respect to the angular and length constraints. The cost function is given below.

The SSD of the distance between feature vectors for time t-1 and t is calculated, the cost is written to 𝐶𝐹 based on the following conditions.

Matches minimizing the selection cost of the feature points are used as preliminary matches. 3.2 Kalman Filter Kalman filter is proposed by R. E. Kalman in 1960 (Kalman, 1960), the filter has been used widely in many fields and recently, it has been used for image processing (Fischler & Bolles, 1981). Kalman filter has two applications; one is filtering and the other, prediction. The later in particular has been used in applications requiring both prediction and correction. Prediction is made using the variance matrix constructed from previous state's estimated values and errors, which is used to calculate estimated values of the new state. The new state's estimations can be obtained and optimized using the Kalman gain and measured value calculated from the variance matrix. As Kalman filtering uses the previous state's model to predict the current state's model, in this case, the main input parameters are the coordinates of the prediction. Fig. 4 shows the main stages for the Kalman filter. 790

4. Results This work improves upon the work done by J. H. Zhang in (Zhang, 2014), which performs ego-motion estimation using adaptive matching based on epipolar geometry and disparity. In the experiment we use window size of 8 to compare with prior work, in particular, we focus on cases where the previous work was not able to perform satisfactorily. The experiment also includes bundle adjustment to optimize the results as was in Zhang's work. The results are compared in Table 1. 791

The data in Table 1 are from the database in KITTI Vision Benchmark Suite. In Zhang's method, the results from datasets 0005, 0091, and 0079 were unsatisfactory. In addition, Zhang's method could not perform for dataset 0035, since the temporary stop of the vehicle's motion caused the method to collapse and fail. In the comparison, it can be seen that the proposed method is able to perform with lower translation and rotation errors than Zhang's method most of the time. In particular, where Zhang's method was unable to perform for dataset 0035, our method was able to obtain results with quite low translation errors. For other datasets with more complex scenarios, the proposed method was also able to perform with satisfactory results. In the next comparison, the GPS+IMU data provided in the KITTI Vision Benchmark Suite is used as the ground truth and compare the estimated motion with results from our method. The comparison is shown in Fig. 5, where the red line shows the motion obtained by the proposed method and the green line represents ground truth given by KITTI. Zhang's method was unable to calculate the ego-motion for this particular dataset due to the turning motion in the sequence.

5. Conclusions In this work, we propose an improved ego-motion estimation approach. The proposed approach uses binocular stereo images as input, detects and generates feature descriptors, performs semi-global matching, and uses camera parameters to calculate the feature points' 3D coordinates. The feature points are matched temporally between image frames, and the 792

matching process uses Kalman filtering and adaptive matching for feature tracking. The calculated 3D coordinates, 2D projections and camera parameters are used to obtain the camera's motion. Finally, bundle adjustment is used to optimize the ego-motion. The method improves adaptive matching constraints with Kalman filtering, and uses Kalman filters to predict the intersection of tracked feature points' segments, which are used to obtain the angular and length constraints; the two constraints are combined to improve ego-motion estimation. The experimental results demonstrate that the proposed method is able to perform in complex scenarios and obtains better results than previous method. In the future, the proposed method can be used to determine the ego-motion of systems equipped with vision devices and used for applications such as 3D reconstruction, unmanned vehicle navigation. 6. References Besl, P. J., & McKay , N. D.(1992). A Method for Registration of 3-D Shapes, IEEE Transaction on Pattern Analysis and Machine Intelligence, 14(2), 239-256. Nister, D., Naroditsky, O. & Bergen, J.(2004). Visual Odometry. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2004, 1, 652-659. Comport, A., Malis, E. & Rives, P.(2007). Accurate Quadrifocal Tracking for Robust 3d Visual Odometry. Proceedings of the IEEE International Conference Robotics and Automation 2007, 40-45. Cappelle, L. Wei, C., Ruichek, Y. & Zann, F.(2011).GPS and Stereovision-Based Visual Odometry: Application to Urban Scene Mapping and Intelligent Vehicle Localization. International Journal of Vehicular Technology, 2011. Scaramuzza, D. & Fraundorfer, F.(2011).Visual Odometry: Part I: The First 30 Years and Fundamentals. IEEE Robotics and Automation Society, 18(4), 80-92. Scaramuzza, D. & Fraundorfer, F.(2012).Visual Odometry: Part II: Matching, Robustness, Optimization, and Applications. IEEE Robotics & Automation Magazine, 19(2), 78-90. Zhang, J. H.(2014).Adaptive Feature Tracking Based on Epipolar Geometry and Disparity for Ego-Motion Estimation. M. S. Thesis, Department of Computer Science and Information Engineering, National University of Kaohsiung. Kalman, R. E.(1960).A New Approach to Linear Filtering and Prediction Problems. Journal of basic Engineering, 82(1), 35-45. Fischler, M. A. & Bolles, R. C.(1981).Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography. Communications of the ACM, 24(6), 381-395.

793

Food, Agriculture and Veterinary Sciences

Thursday, March 31, 2016

14:50-16:20

Room C

Session Chair: Ms. Reisky Megawati Tammu ISFAS-1575 Monte Carlo Simulation of the Spread of Mealybugs in a Cassava Field Chontita Rattanakul︱Mahidol University ISFAS-1597 Phenotypic and Genotypic Characters of the „Katokkon‟ Pepper (Capsicum Annuum L.) from Toraja-Indonesia, Polyploidy by Colchicine Treatment Reisky Megawati Tammu︱Gadjah Mada University Budi Setiadi Daryono︱Gadjah Mada University Tri Rini Nuringtyas︱Gadjah Mada University ISFAS-1598 The Development of Yanang (Tiliacora Triandra) Leather Product Wisutthana Samutsri︱Phranakorn Rajabhat University Tippawan Po-Eang︱Phranakorn Rajabhat University ISFAS-1605 The Phenotype and Molecular Characters of Polyploid Soybean (Glycine max (L.) Merr.) Induced by Colchicine Irma Nofitahesti︱Gadjah Mada University Budi Setiadi Daryono︱Gadjah Mada University ISFAS-1580 Feasibility Study of Lipase Immobilized on Chicken Eggshell for Synthesis of Sugar Ester Nur Suhaila Mohamed Salleh︱Universiti Sains Malaysia Noor Aziah Serri︱Universiti Sains Malaysia Sufia Hena Ali︱Universiti Sains Malaysia Husnul Azan Tajarudin︱Universiti Sains Malaysia Azlina Harun@Kamaruddin︱Universiti Sains Malaysia

794

ISFAS-1575 Monte Carlo Simulation of the Spread of Mealybugs in a Cassava Field Chontita Rattanakula,b Department of Mathematics, Faculty of Science, Mahidol University, Thailand b Centre of Excellence in Mathematics, Commission on Higher Education, Thailand E-mail address: [email protected] a

1. Background/ Objectives and Goals Cassava is one of Thailand‘s agriculture crops of importance. The share on the world‘s export market is approximately 60.72% in 2011 according to the Office of Agricultural Economics, Ministry of Agriculture and Cooperatives, Thailand. The spread of mealybugs in cassava fields might cost a dramatic loss in crop yields as it has been recorded in 2010 that the total cassava yield reduced from 30 million tons per year to 22 tons per year according to the information from the Office of Agricultural Economics, Thailand. Therefore, the efficient control of the spread of mealybugs is essential. The spread of mealybugs might be controlled by using insecticide, biological control or the mixed of insecticide and biological control approaches. With biological control, green lacewing is considered as one of the major natural enemies of mealybugs and hence, it is often used as a biological control agent. There are various recommended amount of green lacewings to be released in the field such as 200 green lacewings per rai, 800 green lacewings per rai or 1,000 green lacewings per rai. In this study, we then investigate the effect of the different amount of green lacewings released in the field. 2. Methods In order to investigate the spread of mealybugs in a cassava field Monte Carlo Simulation and Cellular Automata techniques are employed. In the simulations, we use a 40 × 40 lattice to represent a cassava field with the planting area of 1 rai (40 m × 40 m). The planting distance between two cassava plants is 1 m. The total number of cassava plants in the field is then 1,600 plants. The states of every cell in the lattice will be updated in parallel at each time step (1 time step (Δt) = 1 day). Each cell in the lattice represents a state of the cassava planted in the cell which will be updated at each time step according to given rules. The possible states for each cell in the lattice are S, I and E representing susceptible cassava, infested cassava and removed cassava, respectively. After a month of planting, the cassava field will be surveyed every two weeks. If mealybugs are found on the surveyed cassava plants, green lacewings at the larva stage will be released randomly on the infested cassava plants every two months until there are no mealybugs on the surveyed cassava plants. Simulations for each of the different amount of green lacewings released in the field (200 green lacewings per rai, 800 green lacewings per rai, 1,000 green lacewings per rai, 1,200 green lacewings per rai and 1,500 green lacewings per rai) are carried out. 3. Expected Results/ Conclusion/ Contribution The simulations of the spread of mealybugs for each of the different amount of green lacewings released in the field are carried out using parametric values that were estimated from the available reported data at 30 °C. The results are as shown in Fig.1.

795

The results indicate that the release of 1,200 of green lacewings per rai in the field when the infestation are detected gives the maximum average estimated crop yields comparing between 200/800/1,000/1,200/1,500 green lacewings released in the field. Therefore, the more green lacewings does not yield the better control of the spread of mealybugs in a cassava field and more insight study is needed. Keywords: Monte Carlo simulation, cassava, mealybug, Green lacewings

796

ISFAS-1597 Phenotypic and Genotypic Characters of the „Katokkon‟ Pepper (Capsicum Annuum L.) from Toraja-Indonesia, Polyploidy by Colchicine Treatment Reisky Megawati Tammu1*, Budi Setiadi Daryono2, Tri Rini Nuringtyas3 Laboratory of Genetics and Breeding, Faculty of Biology, Gadjah Mada University, Indonesia 2 Laboratory of Genetics and Breeding, Faculty of Biology, Gadjah Mada University, Indonesia 3 Laboratory of Biochemistry, Faculty of Biology, Gadjah Mada University, Indonesia *E-mail: [email protected] 1

Katokkon is a cultivar of red pepper origins from Toraja, which has a good potential to be developed as high-quality pepper because of its spicy flavor and unique shapes like small paprika. Polyploidy by colchicine treatment is a plant breeding technique that had been applied in a variety of plants for improving their productivity and quality. This study aims to determine the phenotypic and genotypic characters of Katokkon poliploid pepper which is obtained by colchicine treatment. This study was conducted by using the factorial design. The first factor is concentration of colchicine, which consists of the control and treatment of four concentration variation; 0,025%; 0,05%; 0,075% and 0,1%. The second factor is soaking time, which consists of 6, 12 and 24 hours. The parameters which will be measured as a limitation of the study are phenotype characters (the plant height, girth, length and width of leaves, number of fruit/plant, length and diameter of fruit, fresh weight per fruit, the flesh‘s thickness of the fruit, the content of capsaicin in fruit, ascorbic acid level and total carotenoids in fruit) and the ploidy degree as genotype character. The study has been held since October 2015 and expected to be finished on May 2016 at several locations; Laboratory of Genetics and Breeding UGM, green house UGM, Laboratory of Mathematics UGM, Laboratory of FALITMA UGM, LIPI Cibinong and Laboratory of Chem-mix Pratama, Bantul. Plant ploidy degree is analyzed by flow cytometer with propidium iodide dye. Capsaicin content will be analyzed by HPLC method with a mobile phase mixture of acetonitrile and water (50:50). Ascorbic acid level of fruits will be measured by 2,6-Dichlorofenolindofenol titration method. Total carotenoids level of fruit will be measured by spectrophotometer. The result of this study is expected to be a source of information as well as an alternative way to increase the productivity and quality of local cultivars pepper in Indonesia. A high-quality pepper has a high productivity, pests and diseases resistance, rapid fertilization period, high-level spiciness, and uniform shape (Harpenas and Dermawan, 2011). The pepper fruit‘s quality is determined by fruit size, seed number, fruit pulp thickness, fruit color, and chemical contents such as capsaicin, sugars, protein, vitamins A and C (Kallo, 1988 cit. Darsina, 2000). Plant breeding is a program to obtain cultivars or varieties in various way such as crossing/breeding, mutation, or genetic engineering of plants (Anggraito, 2004) include polyploidy. Polyploidy treatment is commonly used in plant breeding because it can increase the yields. Polyploidy is a condition which an individu has more than two sets of chromosomes. If polyploidy occurred in Katokkon pepper, the number of chromosomes will be more than 24. The multiplication of this chromosome can create larger plant parts include roots, stems, leaves, flowers and fruits (Escadon et al., 2003). Polyploidy can be achieved by colchicine treatment (Haryanti et al., 2009) and it has been applied to various crops such as potatoes (Alam et al., 2011), papaya (Angkasa, 2006), soybean (Sofia, 2007), etc. Polyploidy by colchicine had been applied also to the red pepper varieties of Lado F1 for increasing its growth and productivity. 797

Colchicine concentration of 15 ppm with 24 hours of soaking time of seeds provide an optimal result for the growth and productivity of Lado F1 pepper. The comparison between the treatment and control in plant height is 70.2 cm and 58.3 cm, while the fresh weight is 24.76 g and 3.76 g (Syaifudin et al., 2013). Colchicine with a particular concentration and soaking time can be applied in Katokkon pepper seed to induce the polyploid plant. The polyploid Katokkon pepper (Capsicum annuum L.) is expected to have phenotypic characters which is higher than the control and its ploidy degree will be multiplied (more than 24 chromosome numbers). Keywords: Katokkon pepper, colchicine, polyploidy, capsaicin

1.

2.

3.

4.

5. 6.

7.

8.

References Alam, M.M., Karim, M.K., Aziz, M.A., Hossain, M.M., Ahmed, B., and Mandal, A. Induction and evaluation of polyploidy in some local potato varieties of Bangladesh. Journal of Biodiversity and Environmental Sciences. 2011,(1):16-21. [INNSPUB] [AJMS] Angkasa, B. Induksi poliploidi pada papaya solo (Carica papaya) dengan kolkhisin. Jurusan Biologi Fakultas Matematika dan Ilmu Pengetahuan Alam. Universitas Udayana, Denpasar. 2006. [Thesis, Udayana University - Indonesia] Anggraito, Y.U. Identifikasi Berat, Diameter, dan Tebal Daging Buah Melon (Cucumis melo, L.) Kultivar Action 434 Tetraploid Akibat Perlakuan Kolkhisin. Berkala Penelitian Hayati, 2004, 10 (1): 37-42. [Journal of Biological Researches, Indonesia] Darsina. Evaluasi karakter holtikultura 10 hibrida hasil persilangan 10 genotip tetua cabai merah (Capsicum annum L.) yang ditanam tanpa naungan di musim hujan. Skripsi. Fakultas Pertanian. Bogor: Institut Pertanian Bogor. 2000. [Thesis, Bogor Agricultural University, Indonesia] Escandon, A.S., Hagiwara, J.C. and Alderete, L.M. A new of Bacopa monnieri obtained by in vitro polyploidization. Electronic Journal of Biotechnology. 2006,(9):181-186. Haryanti, S., Hastuti, R.B., Setiari, N., dan Banowo, A. 2009. Pengaruh kolkisin terhadap pertumbuhan, ukuran sel metafase dan kandungan protein biji tanaman kacang hijau (Vigna radiata (L) Wilczek). Jurnal Penelitian Sains Teknologi. 10:112-120. [Scientific publication, Journal of Science and Technology, Indonesia] Sofia, D. Pengaruh konsentrasi dan lama waktu pemberian kolkhisin terhadap pertumbuhan dan poliploid pada biji muda kedelai yang dikultur secara in vitro. Karya Tulis. Fakultas Pertanian. Universitas Sumatera Utara. 2007. [Scientific work, Sumatera Utara University, Indonesia] Syaifudin, A., Ratnasari, E. dan Isnawati. Pengaruh pemberian berbagai konsentrasi kolkhisin terhadap pertumbuhan dan produksi tanaman cabai varietas Lado F1. LenteraBio. 2013,2 (2): 167-171. [Scientific electronic journal, State University of Surabaya, Indonesia]

798

ISFAS-1598 The Development of Yanang (Tiliacora Triandra) Leather Product a

Wisutthana Samutsria,*, Tippawan Po-Eanga Faculty of Science and Technology, Pranakorn Rajabhat University, Thailand *E-mail: [email protected]

Abstract Yanang is a climbing plant of Tiliacora triandra (Colebr.) Diels of the Menispermaceae family. Yanang leaves can form viscous in water. The extract from the leaves of this plant is used as food ingredient particularly in many cuisines of the northeast of Thailand. The residue by-product from Yanang extraction could be another source of natural fiber. The aims of this study were to develop the leather from Yanang leaves which was a by-product from Yanang extraction and characterized its properties and study the shelf-life of selected product for a period of 3 week. The ratios between Yanang puree and wheat flour at 60:40, 55:45, 50:50, 45:55, and 40:60 (w/w) were investigated. The color of Yanang leather before frying was greenish with Lightness (L*), red color (a*), and yellow color (b*) value at rangs of 40.05-41.59, 1.63-2.11, and 8.24-8.73, respectively and the color of these Yanang leather after frying was lighter than those of before frying with L*, a*, and b* value of 45.35-46.61, 2.47-2.92, and 11.31-12.53, respectively. The water activities and moisture contents of Yanang leather before frying were 0.83-0.85 and 4.47-4.52, respectively, whereas the water activities and moisture contents of those Yanang leather after frying were 0.63-0.64 and 2.21-2.40, respectively. The hardness and fracturability of fried Yanang leather at the ratios of 60:40, 55:45, 50:50, 45:55, and 40:60 (w/w) were 429.49-576.56 g and 1615.78-1845.14 g, respectively. The fat and fiber contents of those Yanang leather were 27.78-33.30% and 10.24-13.45%, respectively. The study ratio between wheat flour and Yanang puree for production of Yanang leather found that the appropriate proportion was 55:45. These product had the highest preference score for over all liking of 7.35 from 9. For the study of shelf-life of the best ratio of Yanang leather for 3 weeks in oriented polypropylene (OPP) bags at 30 °C, it was found that the storage fried Yanang leather had less hard and fracture than those of fresh fried Yanang leather with the values of 218.19-605.48 g and 1570.29-1916.66 g, respectively. The moisture content and TBA value were 1.72-5.52% and 1.40-1.95 mg.malonaldehyde/1kg, respectively. The sensory evaluation showed that the panelists rated this product for all attributes lower score when storage time was increased. Keywords: vegetables leather, Yanang (Tiliacora triandra (Colebr.) Diels), crispy, sensory evaluations, shelf life 1. Background/ Objectives and Goals Yanang, Tiliacora triandra (Colebr.) Diels, is in the family of Menispermaceae. It is a species of flowering plant native to mainland Southeast Asia, and is widespread in the Northeast of Thailand. It is a climbing plant with deep green leaves and yellowish flowery. Chemical analysis revealed that Yanang leaves contain high levels of β-carotene and minerals, such as calcium and iron (Mahidol et al., 1994). Yanang leaves can form viscous in water. Yanang extract has been used as a herbal medicine for fever relief, malaria remedy, anti-inflammation, antioxidant, anticancer, antifungal and antibacterial properties (Singthong, Ningsanond, and Cui, 2009; Sireeratawong et al., 2008; Phomkaivon and Areekul, 2009; Singthong et al., 2009; Naibaho, Laohankunjit, & Kerdchoechuen, 2012) It is used particularly in the cuisines of the Northeast of Thailand and Loa especially in bamboos shoot soup. More recently, pasteurized Yanang juice has been developed and promoted as a recommended product of the One Tumbol 799

One Product (OTOP) program by the Thai government. The residue by-product from Yanang extraction could be another source of natural fiber. Leathers are made by pouring purees on a flat surface for drying. Leathers can be made from fresh, frozen, or drained canned fruit or vegetable purees. Vegetables can be preserved by drying because they contain less acid than fruits, vegetable should be drained until they are brittle or crisp. Therefore this research aims to develop the leather from Yanang leaves which was a by-product from Yanang extraction and characterized its properties and study the shelf-life of selected product for a period of 3 week. 2.

Methods

a. Preparation of Yanang puree Yanang leaves was purchased from Sapaan-mai market, Bangkok, Thailand, cleaned with water to reduce dust and infected leaves, and then squeezed with water. The residue leaves form discarding of Yanang juice extraction were blended for 10 min (Buono® TSK-9355B), then squeezed out the water. The Yanang puree was sealed in a plastic bag until Yanang leathers were made. b. Development of Yanang leather Yanang leather recipe containing 100 g of composite puree that was prepared by replacing the Yanang puree by replacing the Yanang puree by wheat flour. The study of product development with oven dryer, the appropriate proportion between Yanang puree and wheat flour at the ratios of 60:40, 55:45, 50:50, 45:55, and 40:60 (w/w) were prepared. 2 g salt, 2 g sugar, 1 g pepper powder, 0.5 g chili powder, 5 g soy sauce, 45 g water, and the composite puree were mixed and thoroughly kneaded for 15 min, then poured the mixture and use a offset spatula to evenly spread it over the paper with thickness of 0.2-0.3 cm. The mixture were dried by hot air at 60 °C for 1 h, remove the parchment paper off and more heated for 10-15 min. After drying, the leather were cut into a strip of 3x9 cm2 size, and were sealed in a OPP bag or airtight container until frying were made. c. Measurement of Yanang leather properties Color of the Yanang leathers were measured with a Color Reader (CR-10, Konica Minolta Optics, Inc., Osaka, Japan) in the CIE L*a*b* system. An average values of three replications were reported. TPA of the Yanang leathers were performed by using Texture analyzer (TA.XTPlus, Stable Micro Systems Ltd., Surrey, UK). A set of at least five pieces of the Yanang leathers was placed on a flat metal plate. Instrument settings with a cylindrical 36 mm diameter probe were compression mode, trigger type, pre-test speed 1.0 mm/sec; post-test speed, 10.0 mm/sec; test speed 0.5 mm/sec; Distance 2 mm. From the force-distance curves generated, five texture parameters can be obtained but the highest and lowest values were discarded, thus only three strands were used for data analysis: hardness (g) and fracturability (g). Water activities, moisture, fat and fiber contents were analyzed by AOAC method (5) and Thiobarbituric acid value was measured by Egan et al. (1981) method. Studied the shelf-life for a period of 3 week, comes packing Yanang leather in Oriented Polypropylene (OPP) bags at 30 °C, shelf-life including hardness, fracturability, moisture content, TBA values, and sensory quality was checked at 0, 3, 7, 10, and 21 days. d. Sensory evaluation Sensory evaluations were conducted in the sensory laboratory of the Department of Food Science and Technology, Faculty of Science and Technology, Pranakorn Rajabhat University. The sensory attributes including appearance, color, taste, texture, and overall liking were 800

evaluated by untrained 20 panelist, using a 9-point hedonic scale. The scale ranging from 1 to 9 with 1 representing the least score (dislike extremely), 5 the middle score (neither dislike nor like), and 9 the highest score (like extremely). One piece of the Yanang leather of 3x9 cm2 size from each treatment was placed in plastic cups with lids code with three-digit random number and served in a randomized order. Panelist were seated in individual booths under red lights and provided with room temperature water. e. Statistical analysis The measurements were performed using three replications. The results were reported as mean value with standard deviation. The data were subjected to analysis of variance (ANOVA) using the SPSS statistical software package. Duncan‘s multiple range test (DMRT) was applied to determined significant differences at the 5% level of significance (p≤0.05) between the treatment parameter associated with the properties of the Yanang leather. 3. Results 3.1 Properties of Yanang leather before frying The physical and chemical properties of Yanang leather before frying are summarized in Table 1.

L* (Lightness-darkness), a* (redness-greenness), and b* (yellowness-blueness) values of all Yanang leather were significantly different (p≤0.05). An increase in Yanang puree from 40 to 60/100 total ratio into the mixture decreased L*-value from 41.59 to 40.05, a*-value from 2.11 to 1.63, and b*-value from 8.73 to 8.24 of Yanang leather before frying because of greenish color of Yanang puree (Mahidol et al., 1994). That could promote darker product when the proportion was increased. On the other hand, water activities and moisture contents of all Yanang leathers were significantly different (p≤0.05). An increase in Yanang puree from 40 to 60/100 total ratio into the mixture increased water activities and moisture contents from 0.83 to 0.85 and from 4.47 to 4.52%, respectively. This was the increasing of Yanang puree increased water proportion of Yanang puree in the formulas thus the moisture content were increased. 3.2 Properties of Yanang leather after frying The physical and chemical properties of Yanang leather after frying are summarized in Table 2 and the characteristics of Yanang leather before and after frying are showed in Fig. 1.

801

L*, a*, and b* values of Yanang leather after frying was lighter than those of before frying (Table 1 and Fig. 1). This might be possibility due to dissolving pigment in Yanang during process (Naibaho et al., 2012). Regarding to L*, a*, and b* values of Yanang leather after frying, when Yanang was increased in the formula from 40 to 60/100 total ratio, L*, a*, and b* values of all Yanang leather were significantly decreased (p≤0.05). It was decreased from 46.61 to 45.85, 2.92 to 2.47, and 12.53 to 11.31, respectively. This was coincided with the result of color of Yanang leather before frying (Table 1), presenting that increasing of Yanang puree decreased the L*, a*, and b* values of the product. However, an increase in proportion of Yanang puree from 40 to 60/100 total ratio in the recipe significantly increased (p≤0.05) water activities and moisture contents of Yanang leather after frying from 0.63 to 0.64 and 2.21 to 2.40%, respectively. This was might be due to high moisture proportion of Yanang puree. The textural and chemical properties of Yanang leather after frying are summarized in Table 3. Hardness of the Yanang leather after frying at the ratios of 60:40, 55:45, 50:50, 45:55, and 802

40:60 (w/w) of wheat flour and Yanang puree was decreased from 576.56 to 429.49 g, respectively. Fracturability of Yanang leather of all sample was significantly different (p≤0.05). The increased Yanang puree from 40 to 60/ 100% ratio in the recipe decreased their fracturability from 1845.14 to 1615.78 g. On the other hand, fat and fiber contents were significantly increased (p≤0.05) from 27.78 to 33.30 and 10.24 to 13.45 %, respectively, when the proportion of Yanang puree increased. Hardness and fracturability of Yanang leather after frying with high proportion of Yanang puree were higher than those of low content of Yanang puree but the fat contents were increased. It was possible that high fiber contents led to crispy product of Yanang leather after frying but high excessive fiber contents led to high oil uptake of Yanang leather during frying. Similar results were reported by Brannan, Myers, and Herrick (2013) with the addition of low fiber content was successful in reducing lipid absorption but the addition of high fiber content did not produce a synergistic effect to reduce fat content. However, Apart from health benefits of fibers, it has also been reported to reduce oil uptake during frying especially soluble fibers (Yadav and Rajan, 2012).

3.3 Sensory evaluation of Yanang leather after frying Sensory evaluation of Yanang leather after frying was presented in Fig. 2. Preference score of Yanang leather for appearance, color, flavor, crispy, and overall liking were ranges of 5.60-6.30, 5.65-6.75, .60-7.00, 5.30-7.10, and 5.55-7.35 from 9, respectively. Yanang lather at the ratio of wheat flour and Yanang puree of 45:55 had the highest preference score for all attributes.

803

3.4 Properties of Yanang leather during storage From the study, the appropriate proportion between wheat flour and Yanang puree was 45:55. The sensory evaluation show that the panelists rated this product for over all liking score of 7.53 from 9. Thus, this product was selected to study shelf life for a period of 3 week. Yanang leather in Oriented Polypropylene (OPP) bags at 30 °C, shelf-life including hardness, fracturability, moisture content, and sensory quality was checked at 0, 3, 7, 10, and 21 day. The ability of the spectrophotometric Thiobarbituric acid test to determine malondialdehyde in Yanang leather was aslo evaluated during storage. Regarding to study of shelf life, Hardness and fracturability of Yanang leather was significantly decreased when storage time increased (p≤0.05). Yanang leather had hardness values between 2.18-6.05 g and fracturabiity values between 1570.29-1916.66 g during storage time of 21 (Fig. 3a and 3b). Whereas moisture contents and TBA values were significantly increased when storage time increase (p≤0.05). During storage time of 21 days, Yanang leather had moisture contents between 1.72-5.52% and TBA values between 1.40-1.95 mg.malondialdehyde/1kg (Fig. 3c and 3d). This changed in texture properties was due to the increase in moisture content. The increasing of TBA values during storage was because of an oxygen permeability of OPP bag led to an increase in lipid peroxidation during storage (Kanner and Rosenthal, 1992).

804

805

After storage of Yanang leather at 30 °C in OPP bags for 21 days, Yanang leather had preference score lower from 7.25 to 5.95, 7.00 to 5.70, 7.00 to 3.25, 7.00 to 4.75, and 7.30 to 4.40 for appearance, color, flavor, crispy, and overall liking, respectively. The preference score was decrease rapidly from 0 day to 10 days of storage from 7.00-7.30 to 3.95-5.95, after that the preference score was stable. The preference score of Yanang leather for flavor during storage was the lowest score comparing with other attributes and it was related with over all liking preference score. This was coincided with TBA value result (Fig 3d) which was highly increased in 10 days of storage. 3.3 Acknowledgments and Legal Responsibility The authors gratefully acknowledge use of facilities and instrumentations supported by Food Science and Technology program, Faculty of Science and Technology, Pranakorn Rajabhat University. 4. References AOAC.(2000). Official Method of Analysis. 15th Ed. The Association of Official Analytical Chemists, Arlington, Verginia. p. 1985. Brannan, R. G., Myers, A. D. & Herrick, C. S.(2013). Reduction of fat content during frying using dried egg white and fiber solutions. Europien Journal of Lipid Science and Technology, 115, 946-955. doi:10.1002/ejlt.201200356. Egan, H., Kirk R.S. & Sawyer, R.(1981). Pearson's Chemical Analysis of Food. 8th Ed., Churchill Livingstone, Edinburgh, UK. Kanner, J. & Rosenthal, I.(1992). Assessment of lipid oxidation in foods. Pure and Applied Chemistry, 64(12), 1959-1964. Mahidol, C., Sahakitpichan, P. & Ruchirawat, S.(1994). Bioactive natural products from Thai plants. Journal of Pure and Applied Chemistry, 66, 2353–2356. Naibaho, N. M., Laohankunjit, N. & Kerdchoechuen, O.(2012). Physico-chemical properties of plant extracts from Yanang (Tiliacora triandra) leaves. Agricultural Science Journal, 43(2), 533–536. Phomkaivon, N. & Areekul, V.(2009). Screening for antioxidant activity in selected Thai wild plants. Asian Journal of Food and Agro-Industry, 2(4), 433–440.

806

Singthong, J., Ningsanond, S., & Cui, S. W.(2009). Extraction and physicochemical characterisation of polysaccharide gum from Yanang (Tiliacora triandra) leaves. Food Chemistry, 114, 1301–1307. Sireeratawong, S., Lertprasertsuke, N., Srisawat, U., Thuppia, A., Ngamjariyawat, A., Suwanlikhid, N. & Jaijoy, K.(2008). Acute and subchronic toxicity study of the water extract from Tiliacora triandra (Colebr.) Diels in rats. Songklanakarin Journal of Science and Technology, 30(5), 611–619. Yadav D. N., & Rajan A.(2012) Effect of fibre incorporation on rheological and chapatti making quality of wheat flour. Journal of Food Science and Technology, 47(6), 767–773. doi: 10.1007/s13197-010-0218-7.

807

ISFAS-1605 The Phenotype and Molecular Characters of Polyploid Soybean (Glycine max (L.) Merr.) Induced by Colchicine Irma Nofitahesti1*, Budi Setiadi Daryono2 1, 2 Laboratory of Genetic, Faculty of Biology, Gadjah Mada University, Indonesia *Email : [email protected] Soybean (Glycine max (L.) Merr) is one of the most important food commodities to fulfill the protein necessity in Indonesia. Although Indonesia has had many superior soybean varieties, Indonesia cannot fulfill the whole need of soybean and always rely on soybean import (BPS, 2014). This problem can be solved by improving the quality of the superior soybean seed and making the new superior soybean seed with higher productivity which can be made by polyploidization with colchicine. This research aims to study the phenotype and molecular characters of ‗Anjasmoro‘ soybean which is induced by colchicine. Besides, this research also aims to know the impact of polyploidization in soybean productivity enhancement. ‗Anjasmoro‘ soybean is one of the superior soybean varieties in Indonesia with big seed (14,8-15,3 g/100 seeds) (Suhartina, 2005). The research is done in January 2016 until May 2016. The colchicine concentration which is used in this research are 0.01% and 0.02%. The control soybean seed (without colchicine induction) and the treatment soybean seed which is soaked in each concentration of colchisine for 10 hour are planted in Gadjah Mada University greenhouse Yogyakarta, each planted with 10 replicates. The growth parameters as the phenotype character which is measured are the stomata size, the plant height, the total pod in one plant, the total seed in one plant, the weight of 100 seeds, the flowering time, and the ripening time of soybean. The ploidy degree of soybean as the molecular character is analyzed with flowcytometry method. The total protein content of soybean is measured with Kjehldal method, while the isoflavon content of soybean is measured with UV-VIS Spectrophotometre. The data is analyzed with ANOVA (α=5%) and DMRT test in SPSS program. The expected results from this research are the ‗Anjasmoro‘ soybean which is induced by colchicine will become polyploid soybean (soybean is a diploid plant with 40 chromosomes (Shi et al., 1996)) with more pods and bigger seed than the normal (diploid) soybean, the protein and isoflavon content will increase, and the productivity will be higher than before. Polyploidization has been used in plant breeding from long time ago. The polyploid plant has more chromosomes than the normal (diploid) plant, and it causes changes in many parts of the polyploid plant, such as root, stem, leaf, flower, and fruit which become bigger and stronger than the normal plant. Furthermore, the protein and vitamin content of the polyploid plant will increase (Suryo, 1995). There are many researches about the using of colchicine in other plants, for example in melon plant. Daryono (1998) said that the polyploidization technique with colchicine induction at 0.1% in 6 hour of soaking time can result in tetraploid melon. Keywords: Soybean, polyploidization, colchicine, protein, isoflavon References 1. BPS (Badan Pusat Statistik - Statistics Indonesia). Produksi kedelai indonesia. www.bps.go.id. 2014. 2. Daryono, B. S. 1998. Pengaruh kolkisin terhadap pembentukan sel-sel melon tetraploid. Buletin Agro Industri. 1998;5:2-11. (In Indonesian Language) 808

3. Shi, L., Zhu, T., Morgante, M., Rafalski, J. A., & Keim, P. Soybean chromosome painting : a strategy for somatic cytogenetics. Journal of Heredity. 1996;87(4):308-313. 4. Suryo. Sitogenetika. Gadjah Mada University Press. Yogyakarta. 1995:217-223. 5. Suhartina. Deskripsi varietas unggul kacang-kacangan dan umbi-umbian. Balai Penelitian Tanaman Kacang-kacangan dan Umbi-umbian - Indonesian Legumes and Tuber Crops Research Institute. 2005:52. (In Indonesian Language)

809

ISFAS-1580 Feasibility Study of Lipase Immobilized on Chicken Eggshell for Synthesis of Sugar Ester Suhaila Salleha, Noor Aziah Serri b*, Sufia Henac, Husnul Azan Tajarudind, eAzlina Harun@Kamaruddin abcd School of Industrial Technology, Universiti Sains Malaysia, Malaysia e School of Chemical Engineering, Universiti Sains Malaysia, Malaysia Email: [email protected], b*[email protected], [email protected], d [email protected], [email protected] Abstract 1. Background/ Objectives and Goals Enzyme immobilization is widely applied in research and industry as the enzyme will be easily removed from the reaction medium. Immobilized enzyme also offer other good advantages such as improved stability against fluctuation of environment conditions, reusability and recovery as well as the ease of controlling and handling in operation. Chicken eggshell that was considered as a solid waste which can be obtained at zero cost, made up from calcium carbonate was used in present study as a carrier for immobilization. In this research, parameters affecting immobilization as well as parameters affecting synthesis of sugar ester were investigated. 2. Methods Lipase from Rhizomucour Miehei with enzyme activity 360.8mol/min/mL was chosen and used for the immobilization process. Parameters such as enzyme loading (0.2mL-0.6mL), size of eggshell (45μ,63μ and 125 μ), weight of eggshells (0.2g-1.0g) were investigated for the immobilization process and simultaneously studied the parameters affected for the synthesis of sugar ester such as concentration of substrates and reaction time. Lipase was adsorped on the eggshells by incubating for 2 hours at 200 rpm shaking rate. To test the efficiency of immobilized lipase, synthesis of sugar ester using fructose and stearic acid as substrates were done. 3. Expected Results/ Conclusion/ Contribution Highest yield (77.78%) was obtained at 36 hour of reaction time when 0.4mL of enzyme loading was used at 37°C using 63µ size of eggshells. As a conclusion, lipase immobilized on chicken eggshells is considered feasible for the synthesis of sugar ester. Keywords: Immobilization, Eggshell, Esterification, Lipase 1. Introduction Enzyme immobilization is a method to attach free lipase to a support material which has been used widely in industry (Ali Khan & Alzohairy 2010)(Joshi et al. 2010). According to (Tischer & Wedekind 1999), the first immobilized enzyme was prepared in 1950s by binding the enzymes onto solid materials. One of the main concern in enzyme immobilization is the carrier itself as it have to follow certain properties in order to be chosen. A good carrier should be able to resist microbial attack, biocompatible, and the most important is to be able to obtained at lower cost (Brena et al. 2013). As we know, egg is one of the common food consumed by humans daily. The eggshells is assumed to be solid waste as the yolk is the only part that can be eaten. Chicken eggshells has three layered structured, an inner lamellar layer, spongy calcerous layer and also cuticle on outer surface (Tsai et al. 2006). Eggshells have most of the 810

properties required to be as a carrier such as ability to resist microbial attack and available at low cost (Makkar & Sharma 1983). Another concern for enzyme immobilization is the method of the binding. Adsorption, cross linking, entrapment and encapsulation method has been discussed by (Datta et al. 2013) for immobilization method. The most common method used for immobilization is adsorption because simple procedure and further treatment is not required. On top of that, adsorption reduce the probability of denaturation enzyme compared to another method of immobilization (Zhao et al. 2006). A few decades ago, a method for esterification of sugar ester with the help of lipase was studied (Okumura et al., 1979)(Stergiou et al. 2013). Synthesis of sugar ester required sugar as a substrate and fatty acid with the help of lipase (catalyst). In industry of food, cosmetics and pharmaceuticals, sugar esters are very important because it works as surfactants (Flores and Halling, 2002). 2.

Experimental

2.1 Materials Chicken eggshells were collected from hawker stall in Universiti Sains Malaysia. Lipase from Rhizomucour Miehei, used in this study was purchased from Sigma. All chemicals used were analytical grade. 2.2 Methodologies 2.2.1 Lipase Immobilization Chicken eggshells were pretreated first before being used in this study. The method were modified from previous study by (Vemuri et al. 1998)(Chattopadhyay & Sen 2012). 0.1% of SDS solution was prepared and eggshells were boiled in it for 15 min. Then, to remove SDS from the eggshells, the eggshells were washed three times with distilled water followed by acetone to remove water. Acetone were removed after that by drying at 60° for 5 hours. The dried eggshells were then sieved using sievers shakers with sizes 45µ, 63µ and 125 µ. After pretreatment, 1g of eggshells were mixed with amount of lipase required together with 50mM phosphate buffer and were incubated at 37°c for 2 hours. After incubation, the immobilized eggshells were stored in 4°c for 16 hours. Before the immobilized eggshells being used, it were washed with 50mM phosphate buffer to remove any unbound enzyme. 2.2.2 Variation of enzyme loading, size of eggshells and weight of eggshells To study the effect of enzyme loading, 0.2mL, 0.4mL and 0.6mL of lipase was loaded into the eggshells while other parameters were kept constant. Then, size of eggshell (45µ, 63µ and 125 µ) and weight of eggshells (0.2g to 1.0 g) were studied as well. 2.2.3 Synthesis of Sugar Ester Fructose and stearic acid was used as substrates for production of sugar ester. Synthesis of sugar ester was done in a 25mL shake flask where stearic acid, fructose and immobilized lipase were added into the shake flask. 6mL of ethanol was added as it was used as solvent. The mixture was incubated for 72 hours at condition 200 rpm, 37°C. Another set of mixture that will act as a control was prepared simultaneously but without lipase. This method was done according to (Neta et al. 2012). To determine the content of sugar ester, volumetric method was used where it was calculated depending on amount of fatty acid left in the reaction mixture. 0.1g of sample was taken from the reaction mixture and was added into 20mL of ethanol. Phenolphtalein was used as indicator. The sample was then titrated with NaOH (0.01M). The yield was calculated based on equation 1-3.

811

3. Results 3.1 Effect of enzyme loading Different enzyme loading were loaded into 1g of eggshell for immobilization. Based on Figure 1, when 0.2mL of enzyme loading was used, the highest yield obtained was only 33.33% but when the enzyme loading increased to 0.4 mL, the highest yield (78.13%) of sugar ester was obtained when esterification of fructose and stearic acid were done. When the enzyme loading was increased to 0.6 mL, the highest yield obtained was only 57.89% at 36 hour. According to (Šabeder et al. 2006), as we increase the concentration of lipase, water as the byproduct of esterification will be produced excessively and it will cause a reversible reaction which will affect the conversion.

3.2 Effect on size of eggshells There were 3 sizes of eggshells been tested in this study that were 45µ, 63µ and 125µ.The eggshells were sieved using sievers shaker to get uniforms size. Figure 2 displayed the bar chart showing the yield of sugar ester at 36 hour for different size of eggshells. 0.2 g of immobilized lipase was used in this parameter study. Size 63µ gave the highest yield of sugar ester compared to 45µ and 125µ. Size 45µ required longer time during crushing and it may cause changes in the structure due to crystallinity reduction. Crystallinity reduction changed the physical structure of the eggshells and affect the adsorption process (Tsai et al. 2008). Meanwhile, for size 125µ, the surface area was less than 63µ and surface area played an 812

important role in adsorption. According to (Zhao et al. 2006), high surface area will results in high enzyme loaded to the carrier.

3.3 Effect weight of eggshells Weight of immobilized eggshells were varied from 0.2 g to 1.0 g. When 1.0 g of immobilized eggshells was used, highest yield of sugar ester was obtained (77.78%) as shown in Figure 3. By increasing the weight of the immobilized eggshells, it means that amount of lipase added to the reaction is higher. High amount of lipase will increase the yield of sugar ester.

3.4 Effect concentration of substrates on synthesis of sugar ester Six different concentration of substrate were tested with ratio of 1:1 fructose and stearic acid. Free lipase was used in this study as a catalyst for the production of sugar ester. Based on Figure 4, 0.9 mmol of substrates concentration showed the highest yield (77.14%) and after 1 mmol of substrates concentration was used, the yield has dropped to 76.92%. 813

3.5 Reaction time synthesis of sugar ester Synthesis of sugar fatty acid ester were run for 72 hours and the time profile was shown in Figure 5. As displayed in Figure 5, 36 hour showed the highest yield that was 77.78%. After 40 hour, the yield of fatty acid sugar ester started to drop and maintained at 60 hour and 72 hour. Previous study by (Neta et al. 2011), proved that the highest yield of sugar fatty acid ester was produced at 37.8 hour.

4. Conclusion Choosing the right carrier for enzyme immobilization is vital and eggshells could act as a good carrier for lipase immobilization. Highest yield (77.78%) was obtained at 36 hour of reaction time when 0.4mL of enzyme loading was used at 37°C using 63µ size of eggshells. Acknowledgement 814

This work was supported by Kementerian Pendidikan Malaysia (KPM) FRGS research grant (6711374), TRGS research grant (6762003) and KPM MyMaster. References Ali Khan, A. & Alzohairy, M. a., 2010. Recent Advances and Applications of Immobilized Enzyme Technologies: A Review. Research Journal of Biological Sciences, 5(8), pp.565– 575. Brena, B., González-Pombo, P. & Batista-Viera, F., 2013. Immobilization of enzymes: a literature survey. Immobilization of Enzymes and Cells, 1051, pp.15–31. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23934795. Chattopadhyay, S. & Sen, R., 2012. A comparative performance evaluation of jute and eggshell matrices to immobilize pancreatic lipase. Process Biochemistry, 47(5), pp.749– 757. Available at: http://dx.doi.org/10.1016/j.procbio.2012.02.003. Datta, S., Christena, L.R. & Rajaram, Y.R.S., 2013. Enzyme immobilization: an overview on techniques and support materials. 3 Biotech, 3(1), pp.1–9. Available at: http://link.springer.com/10.1007/s13205-012-0071-7. Joshi, P. et al., 2010. Immobilization of monoamine oxidase on eggshell membrane and its application in designing an amperometric biosensor for dopamine. Microchimica Acta, 169(2), pp.383–388. Available at: http://dx.doi.org/10.1016/j.msec.2012.11.010. Makkar, H.P. & Sharma, O.P., 1983. Egg shell as a carrier for enzyme immobilization. Biotechnology and bioengineering, 25(2), pp.595–597. Neta, N. do A.S. et al., 2012. Enzymatic synthesis of sugar esters and their potential as surface-active stabilizers of coconut milk emulsions. Food Hydrocolloids, 27(2), pp.324– 331. Available at: http://www.sciencedirect.com/science/article/pii/S0268005X11002864. Neta, N.S. et al., 2011. Maximization of fructose esters synthesis by response surface methodology. New Biotechnology, 28(4), pp.349–355. Šabeder, S., Habulin, M. & Knez, Ž., 2006. Lipase-catalyzed synthesis of fatty acid fructose esters. Journal of Food Engineering, 77(4), pp.880–886. Stergiou, P.-Y. et al., 2013. Advances in lipase-catalyzed esterification reactions. Biotechnology Advances, 31(8), pp.1846–1859. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0734975013001407. Tischer, W. & Wedekind, F., 1999. Immobilized enzymes: methods and applications. In Biocatalysis-from discovery to application. Springer, pp. 95–126. Tsai, W.T. et al., 2006. Characterization and adsorption properties of eggshells and eggshell membrane. Bioresource technology, 97(3), pp.488–93. Available at: http://www.sciencedirect.com/science/article/pii/S0960852405001689. Tsai, W.T. et al., 2008. Utilization of ground eggshell waste as an adsorbent for the removal of dyes from aqueous solution. Bioresource Technology, 99, pp.1623–1629. Vemuri, G., Banerjee, R. & Bhattacharyya, B.C., 1998. Immobilization of lipase using egg shell and alginate as carriers: Optimization of reaction conditions. Bioprocess Engineering, 19(2), pp.111–114. Zhao, X.S. et al., 2006. Immobilizing catalysts on porous materials. Materials Today, 9(3), pp.32–39. Available at: http://dx.doi.org/10.1016/S1369-7021(06)71388-8.

815

Geology, Earth and Environmental Sciences (2)

Thursday, March 31, 2016

16:30-18:00

Room A

Session Chair: Prof. S. Kong Wang ISFAS-1563 Management of Greater-than-C-Class Waste Generated During Nuclear Power Plant Decommission of TaiPower Company S. Kong Wang︱I – Shou University ISFAS-1582 The Estimations of Global and Diffused Solar Radiation Using Meteorological Data and Atmospheric Modeling in Bangkok, Thailand Keerati Vongpadai︱Srinakharinwirot University Siriluk Ruangrungrote︱Srinakharinwirot University ISFAS-1583 Determination of Direct Solar Radiation Using Atmospheric Physics Models in Thailand Pattra Pukdeekiat︱Srinakharinwirot University Siriluk Ruangrungrote︱Srinakharinwirot University

816

ISFAS-1563 Management of Greater-than-C-Class Waste Generated During Nuclear Power Plant Decommission of TaiPower Company S. Kong Wang Professor, Dept. of Mechanical & Automation Engineering, I – Shou University Email: [email protected] Abstract Greater Than Class C (GTCC) waste management is not a new problem, but the issue of spent, or used, nuclear fuel has been the main focus of radioactive waste management for many years. Relatively small volumes of GTCC waste and the lower longevity and activity levels of GTCC have caused used nuclear fuel management to take precedence. Since decommission projects of Tai Power company are expected to embark in a few years, and the currently allowed storage space practically all accounted for, a new way to manage GTCC waste may be worth investigating. The goal of this research is to provide a new look at GTCC waste management strategies. GTCC waste is modeled as entirely activated metals for this study. Currently used treatment methods for spent fuel, mixed waste, and other low level waste will be studied and applied to GTCC waste, and alternative disposal options will also be explored. GTCC activated metals are internal reactor components which have been activated throughout by the neutron flux in operating reactors. The high neutron flux creates activation products throughout the material, and concentrations of carbon-14, nickel-59, nickel-63, and niobium-94 cause these components to be classified as GTCC waste. The goal of treatment application to GTCC waste is to reduce the impact of the waste by lowering the volume of waste for disposal. The treatments which will be studied as possibilities in this study include geologic repository, enhanced near surface burial, and intermediate depth boreholes. The safety of these potential alternatives will be examined through conservative risk assessment calculations for hypothetical reference facilities, which are chosen based on their ability to represent a variety of areas throughout TaiPower‘s nuclear power plants. A computer code PRESTO-EPA is used to perform risk assessment calculations of hypothetical reference facilities. Results from this conservative modeling approach and a sensitivity analysis are expected to show which disposal method results in the expected public dose falling well below the government limits for radionuclides in drinking water. Keywords: Greater-than-Class-C waste, geologic repository, enhanced near surface burial, intermediate depth borehole, safety assessment. 1. Background The question of how to manage Greater Than Class C (GTCC) Low Level radioactive waste has been around for a number of years. Much of the research done on GTCC waste management was done in the 1980s, along with high level waste (HLW) studies, but soon the issue of HLW and spent (or used) nuclear fuel disposal took precedence. One reason for this is the small volumes of GTCC waste when compared to HLW or even other classes of low level waste. Another is the lower longevity and activity levels of GTCC waste as compared to HLW. For example in the USA, as the future of the Yucca Mountain Project becomes less certain, and the currently allowed storage space practically all accounted for, a new way to manage GTCC waste may be worth investigating, including the GTCC waste generated during decommissioning of Taipower nuclear power plants. 817

The goal of this research is to provide GTCC waste management strategies and to propose alternatives to deep geologic disposal. This goal will be met through three major Objectives: Objective one – the determination of the composition and constituents of GTCC waste, Objective two – the application of possible treatment methods to GTCC waste and the determination of their feasibility and practicality, and Objective three – the study of disposal options for GTCC waste which could replace the default geologic repository disposal. The four defines alternative disposal options which are currently being considered by the US Department of Energy are evaluated through the analysis of potential risk to the public and expected radiation exposure. Overall, this study will be a useful addition to GTCC waste management strategies, as it will combine data from previous studies with information from new calculations to help aid in GTCC waste management decisions. Activated metal waste is produced as a result of exposure of the nuclear reactors‘ internal components to a neutron flux over the lifetime of the reactor plant [1]. Both pressurized water reactors (PWRs) and boiling water reactors (BWRs) are used in Taiwan, and each type of reactor uses slightly different components which will become GTCC waste. Typical BWR components which become GTCC waste at decommissioning are the top fuel guide, the core shroud, fuel support castings, and the core support plate. PWR components generally considered GTCC waste are the upper core support plate, the core shroud (baffle), the lower core barrel, and the lower core support plate. The location of these components in the reactor vessel can be seen in Figure 1 [1,2].

2. Methods Even when treatments may be technologically feasible, disposal options could still prove to be more cost effective and/or safe. Because none of the treatment methods appeared promising, disposal is the only option for activated metal GTCC waste. Currently the DOE is considering three different disposal possibilities for GTCC waste. These possible facilities include enhanced near surface burial, intermediate depth boreholes, and deep geologic repository [3]. Enhanced Near Surface Burial Enhanced near surface burial is based on low level waste disposal practices which are currently in use, where the waste is disposed of in shallow trenches or holes and covered with a backfill. For higher classes of waste, barriers such as concrete are used between the waste and the outside environment. GTCC waste will require even more monitoring and containment than any other class of LLW, which would most likely be increased thicknesses of the engineered barriers, greater depths of waste disposal, and longer observation periods after closure of the 818

disposal area. Waste packaging may also be enhanced. A conceptual enhanced near surface facility is shown in Figure 2 [3].

Intermediate Depth Boreholes Intermediate depth boreholes were used for GTCC waste disposal in the late 1980s [4]. Boreholes are created by digging deep holes into the ground, filling them with radioactive waste to a certain predetermined height, then using a backfill to completely fill the hole. The boreholes used in the past for GTCC waste were about 3 meters in diameter and 36 meters deep, with the bottom 15 meters were used to hold the waste. Current intermediate depth boreholes restrictions are expected to allow the boreholes to hold waste no closer to the surface than 30 meters [3]. In order to take advantage of the previously used 15 meters of waste disposal space, the boreholes of current consideration would need to be 45 meters deep. A conceptual intermediate depth borehole facility under consideration is shown in Figure 3 [3].

Deep Geologic Repository According to the Nuclear Waste Policy Act, the default for GTCC waste is a deep geologic repository, which is in place unless another method can be proven to be effective. Geologic repositories are considered a good method of isolating waste, because their depths help to delay groundwater access to the waste and slow the dissolution and sorption of radionuclides, along with other desirable properties of a waste storage and disposal facility. Typical depths of a deep geologic repository are from 300 to 1,000 meters. In the United States, salt, basalt, tuff, and 819

granite have been considered for repository sites in the past, and other countries have considered primarily salt, clay, and granite, with the most commonly studied host rock being granite. Each rock type has its own set of advantages and disadvantages. Geologic repositories can also be a variety of designs, but the most common designs usually include excavated tunnels and sometimes boreholes (either vertical or horizontal) extending out from the tunnels where the waste is buried so the boreholes can be sealed individually. Characterization of Risk from GTCC Waste Disposal In studying potential repository sites, an important consideration is the impact the site will have on the surrounding public. To determine the impact of these potential sites, the risks and possibilities of adverse effects are determined. Risks are typically measured by using data gathered from environmental monitoring or modeling which can be translated into exposure or dose levels to the public. From these exposure calculations, the expected adverse effects can be estimated, typically in terms of increased chance of cancer or lost years of life. Useful environmental modeling data for radioactive waste repository sites will generally consist of expected concentration levels of the radionuclides of concern in the environment. Radionuclide concentration data for GTCC reactor components immediately after shutdown was generated in the 1994 by the DOE, from compositions of reactor components and the operating neutron flux, and is the data used in this study. From these initial concentrations, predicted radionuclide concentrations at a site nearby the repository in contact with the public can be determined. Regulatory limits on radionuclide intake have been established in order to protect individuals in the general public from receiving a radiation dose which could result in a greater possibility of cancer or death. Radiation doses can be obtained through a variety of different means. Humans are exposed to radionuclides primarily from their transportation through water. Radionuclides will be leached from their containment in the disposal facility by water, and will then reach humans through a variety of means, shown in Figures 4 [5].

3. Results Groundwater Transport Model As water enters the disposal facility by infiltration, it causes radioactivity to be released through corrosion of the waste. These radioactive contaminants then leach into the groundwater, moving with it. As groundwater moves through the soil, it disperses in all directions. The transport and dispersion of radionuclides is conservatively neglected in unsaturated region and only the saturated region (aquifer zone) is considered, as shown in Figure 5. Thus, the general three-dimensional mass balance equation given can be applied to calculations for transport in 820

an aquifer containing radionuclides and is shown in Equation (1).

subject to the boundary conditions: C = 0 for all x, y, and z when t = 0 C = C0(t) at x = 0 for t > 0 C = finite at x = ∞ for t > 0

These variables are defined as: R is the retardation factor, C is the concentration of radionuclide in the groundwater, D is the dispersivity tensor, V is the interstitial velocity, and λd is the radionuclide decay constant [6]. The interstitial velocity is defined as V = v/ε, where v is the velocity vector and ε is the porosity of the soil. The retardation factor is given by R = 1 + (Kdρb)/ε, where ρb is the bulk density of the soil, and Kd is the soil distribution coefficient. Radionuclides are assumed to be at equilibrium in the groundwater and the soil, meaning that the linear Kd model for soil absorption is valid. The EPA‘s PRESTO code for modeling LLW movement in groundwater applies this equation in one-dimensional form, which is valid when assuming that the flow of groundwater is steady and uniform and no heat is being lost or absorbed [7]. This one-dimensional model is safe to assume because it represents a conservative approximation as opposed to a three-dimensional model. The 1-D model concentrates the radionuclides in the groundwater into a uniform, narrow stream, as opposed to the dispersed plume of the 3-D model, making a higher concentration of radionuclide reach the well in a shorter amount of time. This will make dose estimations higher than those which will actually occur, making the method conservative. Thus, the three-dimensional Equation (1) simplifies to the one-dimensional Equation (2).

subject to the boundary conditions: C(x,0) = 0, C(y,0) = 0, and C(z,0) = 0 C(0,t) = C0(t) for t > 0 C(∞,t) = finite for t > 0 As depicted in Figures 6 and 7, the flux model [8] is used to calculate the discharge rate of a radionuclide entering a surface-water body that has intercepted the aquifer containing the transported material. It is assumed that all material entering the aquifer eventually enters the surface water except for the part lost through radioactive decay. The assumptions that apply to the point concentration model also apply to this model. The model provides only the rate of 821

input to the surface water at an average distance x downgradient from the surface. In the unidirectional flow field assumed, the flux, Q (curie/year), of material crossing an area dA = dy dz perpendicular to the x-axis is described by the equation

where c is the concentration in the dissolved phase. The total flux across the plane would be

As described by Eq. (4), if ci is the concentration from an instantaneous release of 1 curie from a L x L area source at x = 0 and time t = 0, then the resulting flux at distance x downgradient would be

The above equation is tested as one curie of a radioactive pollutant leaks quickly into a water table aquifer through a highly permeable ground cover over a square surface area L m on a side. The pollutant has a half-life of 30 years. The groundwater is moving in the direction toward a 822

water reservoir or a river at a speed, U, of 1.5 m/day and that the dispersivity is 30 m2/day. The saturated thickness of the water table aquifer, b, is 50 m and has an effective porosity, ε, of 0.2. The pollutant has been determined to have a retardation coefficient, Rd, of 20 in the aquifer. Assuming the water source is located 2000 m away from the radionuclide source. The concentration of the pollutant in the water source with downgradient coordinates that is x = 2000 m with respect to the center of the source area is shown in Figure 8 in which different areas represent the sizes of the radionuclide sources which have gone through the dispersion through the unsaturated zone. Thus, a deeper disposal site will yield a smaller radionuclide source on top of the saturated region, and furthermore, a stronger radionuclide concentration.

The safety of DOE‘s potential alternatives was examined through risk assessment calculations of hypothetical reference facilities using PRESTO-EPA code. These reference facilities were chosen based on their ability to represent a variety of areas throughout the locations of nuclear power plants in Taiwan. These three facilities were differentiated by their climate and soil properties. The sites were chosen based on their success when used by the EPA for LLW disposal, and are humid permeable (loam), humid impermeable (clay), and arid permeable (sand). The EPA determined the use of these reference facilities to be an adequate method of estimating expected public doses due to drinking groundwater downstream of a potential disposal facility. Given the complexity of groundwater hydrology and radionuclide transport in the groundwater, a conservative approach was used for the estimation of models and parameters. A one-dimensional groundwater transport model was used for this reason, as well as conservative values for the soil distribution coefficients for the GTCC radionuclides of concern. Based on this conservative modeling, preliminary estimations for the expected public dose showed that engineered barriers are a necessary part of the waste disposal facility. Shallow burial without engineered barriers showed that the humid impermeable site will result in expected public dose one to two orders of magnitude above the EPA‘s limit for carbon-14 in drinking water. Nickel-59 was found to be the important radionuclide for the humid permeable site, causing expected public doses to be very close to the EPA‘s limit. Engineered barriers in the form of a waste cap between the surface of the facility and the waste disposal location were studied to show that shallow burial is a feasible option with barriers to decrease the groundwater flow through the waste. The values expected dose to the public from a total GTCC activated metals waste volume with engineered barriers are summarized in Table 1. 4. Conclusions The goal of this research was to provide new GTCC waste management strategies. This goal 823

was pursued through a major objective which was identified as studying disposal options for GTCC waste which could replace the default geologic repository disposal. Therefore, while the deep geologic repository disposal option will always be the safest, as it has the lowest expected public dose associated with it for each site location, the process of establishing the facility will certainly be both lengthy and costly, as it has been in the past. Enhanced near surface burial and intermediate depth boreholes were both shown in this study to be cheaper options for disposal while near surface burial has a lower radionuclide concentration at water source, and the expected public dose from these disposal facilities is estimated to be well within the EPA limits, making them a feasible alternative to deep geologic repositories for GTCC waste disposal.

5. References 1. Greater-Than-Class C Low-Level Radioactive Waste Characterization: Estimated Volumes, Radionuclide Activities, and Other Characteristics, Revision 1; DOE/LLW-114; Idaho National Engineering Laboratory for Greater-Than-Class C Low-Level Waste Management Program, U.S. Department of Energy: Idaho Falls, ID, September 1994. 2. Greater-Than-Class-C Low-Level Radioactive Waste and DOE Greater-Than-Class-C-Like Waste Inventory Estimates; Sandia National Laboratory for U.S. Department of Energy: Washington, DC, July 2007. Technical Study of Spent Fuel Pool Accident Risk at Decommissioning Nuclear Power Plants, NUREG-1738, US NRC. 3. Greater-Than-Class C Low-Level Radioactive Waste Environmental Impact Statement. Deparement of Energy, Office of Environmental Management GTCC EIS Website. http://www.gtcceis.anl.gov. 4. Shropshire, D.E.; Williams, K.A.; Boore, W.B.; Smith, J.D.; Dixon, B.W.; Dunzik-Gougar, M.; Adams, R.D.; Gombert,D. Advanced Fuel Cycle Cost Basis; INL/EXT-07-12107; Idaho National Laboratory; Idaho Falls, ID, 1995. 5. User‘s Guide for PRESTO-EPA-CPG/POP Operation System, Version 4.2 [online]; EPA 402-R-00-007; office of Radiation and Indoor Air, U.S. Environmental Protection Agency,2000. http://www.epa.gov/nscep/index.html 6. Hung, C.Y. Evaluation of Application of One-dimensional Groundwater Transport Model to PRESTO-EPA Model. Pract. Period. Hazard., Toxic, Radioact. Waste Manage., 2002, 6 (1), 49-55. 7. Hung, C.Y. An Optimum Groundwater Transport Model for Application to the Assessment of Health Effects due to Land Disposal of Radioactive Waste. Nucl. Chem. Waste Manage. 1986, 6, 41-50. 8. John E. Till and Helen Grogan , Radiological Risk Assessment and Environmental Analysis,Print publication date: 2008 Print ISBN-13: 9780195127270.

824

ISFAS-1582 The Estimations of Global and Diffused Solar Radiation Using Meteorological Data and Atmospheric Modeling in Bangkok, Thailand Keerati Vongpadaia,*, Siriluk Ruangrungroteb a,b Department of Physics,F , Faculty of Science, Srinakharinwirot University, Thailand a E-mail address: [email protected] b E-mail address: [email protected] Abstract Several Mathematical assumptions and models were analytically and numerically carried out in an attempt to estimate global and diffuse solar radiation by applying the actual meteorological data at Bangna meteorological station (13.67 ° N, 100.61° E) in Bangkok, Thailand. Clearness index (Kt) was firstly calculated from the year 2009 dataset of sunshine hours, temperature and relative humidity, yielding a basis mathematical relation as Angstrom-Page equation and also global solar radiation data. Next, various models were considered for calculating the average diffuse solar radiation for clear sky conditions, being classified in 4 models with different relation of each parameters. Statistical tests such as Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE) and Coefficient of determination (R2) were consequently done for every model verification apart from the comparison with available measurement of solar radiation from Pyranometer (MS 802). Results were considered in two cases: global solar radiation, model (4) provided best estimation in the range of 16.14 – 22.20 MJ/m2 with MAPE of 3.13, RMSE of 0.58 and R2 of 0.93, respectively while diffuse solar radiation could be seasonally identified for winter and summer cases. For winter, the diffuse radiation was optimally obtained by model (8) at 4.90 – 5.19 MJ/m2 with MAPE of 29.91, RMSE of 1.16 and R2 of 0.99, while model (9) provided good values at 4.65 – 6.01 MJ/m2 with MAPE of 22.20, RMSE of 1.50 and R2 of 0.66 for summer. Therefore, this work was successfully provided global and diffuse solar radiation in practical, particularly for the absence of available sunshine measurement subject to temporal topography and climate. Keywords: Diffuse fraction, diffuse coefficient, clearness index, sunshine fraction Introduction Knowledge of the atmospheric data and solar-radiation components (global and diffuse) has played significantly role and extremely in demand for both evaluation and assessment of the performance and proficiency of solar-energy system. In general, these data is collected by instrumental measurement via ground-based meteorological stations located in the specific area. Considerably, the financial, technical supports and expertise for running and maintaining the measurement system are sporadically unavailable. While, the necessity of accurate information and analysis about solar radiation budget is highly required for any industrial and domestic applications. Most case study are only focused on bright sunshine hours as clear or partly cloudy sky conditions as long as some solar irradiance components are availble. Therefore, in many cases, solar radiation components such as global and diffuse were estimated and derived using empirical formulae together with mathematical models. This work presents the analytical and numerical method for the estimation of global and diffuse solar radiation in Bangkok in a trial to provide precise solar radiation information without mechanical process. Methodology 825

Thai Meteorological Department (TMD) is located at Bangna (13.67 N and 100.61 E), Bangkok where pyronometer MS 802 and Campbell-Stokes sunshine recorder are equipped for solar radiation component and sunshine hour measurements. The daily relative humidity, temperature and sunshine hours were measured and applied for this study. By mathematical assumptions and modelling, analytic result of global and diffuse solar radiation for 12 months in 2009 were estimated and validated in an attempt to predict the incoming radiative energy distribution. 2.1 Extraterrestrial radiation Monthly average extraterrestrial radiation ( parameters with the mathematical relation:

) in MJ/m2, has to be calculated by astronomical

2.2 Global solar radiation Since the meteorological data from TMD i.e., sunshine hour, temperature and relative humidity were available and applicable, six mathematical models were selected as details:

826

Here model 1 was Angstrom-Page linear equation which related to the clearness index to

b were the derived coefficients, being compulsory to be evaluated from each models. 2.3 Diffuse solar radiation Four mathematical model were further applied for estimating diffuse solar radiation as follows:

2.4 Statistical validation To validate the obtainable result, the statistical tests were done in terms of Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE) and Coefficient of determination (R2) as given by:

Results and discussions 3.1 Correlation of global solar radiation 827

The parameters a and b were completely derived which implied correlation constants of each models as shown in Table 1.

Figure 1 and Table 2 show the performances of estimated global solar radiation from 6 models and their correlations with the actual measurement for nine months from March – December 2009 are in Figure 2. The first three months were neglected due to being unsatisfied with the assumption for clear sky condition.

828

Plus, the statistical validation are presented in Table 3.

3.2 Correlation of diffuse solar radiation The correlation constant of diffuse solar radiation calculated by model 7-10 were presented in Table 4. 829

Figure 3 and Table 5 show the result of estimated diffuse solar radiation from 4 models and their correlations with the actual measurement presents in Figure 4

830

Also, the statistical validation are presented in Table 6.

The obtainable results of global and diffuse solar radiations from 10 mathematical models together with few astronomical parameters and meteorological data in 2009 were estimated and compared with the actual measured data from Pyranometer, showing good correlation with similar trend line in overall. To be more specific, results can be classified in two cases: global solar radiation, model 4 provided best estimation in the range of 16.14 – 22.20 MJ/m2 with MAPE of 3.13, RMSE of 0.58 and R2 of 0.93, respectively while diffuse solar radiation could be seasonally identified for winter and summer cases. For winter, the diffuse radiation was optimally obtained by model 2 at 4.90 – 5.19 MJ/m2 with MAPE of 29.91, RMSE of 1.16 and R2 of 0.99, while model 3 provided good values at 4.65 – 6.01 MJ/m2 with MAPE of 22.20, RMSE of 1.50 and R2 of 0.66 for summer. Conclusion The estimation of global and diffuse solar radiation was successfully done by empirical equations and mathematical modelling, yielding the practical procedure and reliable result without any physical and instrumental constraints. Moreover, the computer program development will be stimulating and curious for a future task in order to provide solar radiation database for any site in the country. Acknowledgements

831

The authors are grateful to Thai Meteorological Department, Bangna, Bangkok for supporting raw data. And this work is sponsored partially by graduate school of Srinakharinwirot University. References C.N. Hewitt and A.V. Jackson. (2009). Atmospheric science for environmental scientists. A John Wiley and son Ltd., United Kingdom. P.J. Robinson and A. Henderson-seller. (1986). Contemporary Climatology. Wesley Longman Singapore (Pte) Ltd., Singapore. J. Almorox, M. Benito and C. Hontoria. (2005). Estimation of monthly Angstrom–Prescott equation coefficients from measured daily data in Toledo, Spain. Renewable Energy 30: 931–936 M.S. Adaramola (2012). Estimation global solar radiation using common meteorological data in Akure, Nigeria. Renewable Energy 47: 38-44. K. Ulgen and A. Hepbasli. (2009). Diffuse solar radiation estimation models for Turkey‘s big cites. Energy conversion and Management 50: 149-156. I. Muhammad. (1983). An introduction to solar radiation. Department of Mechanical Engineering, The University of British Columbia, Vancouver. British Columbia. Canada. G H Hargreaves and R G Allen. (2003).History and evaluation of Hargreaves evapotranspiration equation. Journal of Irrigation and Drainage Engineering 129: 53 63. H. Aras, O. Balli and A. Hepbasli. (2006). Estimating the horizontal diffuse solar radiation over the Central Anatolia Region of Turkey. Energy conversion and Management 47: 2240-2249. S. Janjai, P. Praditwong and C. Moonin. (1996). A new modeling for computing monthly average daily diffuse radiation for Bangkok. WREC: 1283-1286.

832

ISFAS-1583 Determination of Direct Solar Radiation Using Atmospheric Physics Models in Thailand a,b

Pattra Pukdeekiata,*, Siriluk Ruangrungroteb Department of Physics, Faculty of Science, Srinakharinwirot University, Thailand a E-mail address: [email protected] b E-mail address: [email protected]

Abstract This work was originated to precisely determine the direct solar radiation using atmospheric physics models. Since the accurate prediction of the solar radiation is necessary and useful for solar energy applications including atmospheric research. The possible models and techniques for a calculation of the regional direct solar radiation are challenging and compulsory for the case of unobtainable instrumental measurement. The investigation was mathematically governed by six astronomical parameters i.e. declination angle (), hour angle (), solar time, solar zenith angle (z), extraterrestrial radiation (Iso) and eccentricity (E0) along with two atmospheric parameters i.e. air mass (mr) and dew point temperature at Bangna meteorological station (latitude 13.67° N, longitude 100.61° E) in Bangkok, Thailand. Five models of solar radiation determination with the assumption of clear sky were performed together with three statistical tests: Mean Bias Difference (MBD), Root Mean Square Difference (RMSD) and Coefficient of determination (R2) in order to validate the accuracy of obtainable results. The calculated direct solar radiation was in a range of 403-504 Watt/m2 in winter and 489-651 Watt/m2 in summer 2014. Additionally, the dataset of seven continuous days representing both seasons, were examined with the MBD, RMSD and R2 of -0.04, 0.27, 0.93 and 0.02, 0.42, 1.97 respectively, which belonged to Kumar model for winter and Bird model for summer. In summary, the determination of direct solar radiation based on atmospheric models and empirical equations could advantageously provide instantaneous and reliable values of solar components for any location in the region without a constraint of actual measurement Keywords: Atmospheric physics models, astronomical parameters, atmospheric parameters, clear sky condition 1. Introduction The solar radiation reaching the Earth surface has been comprehensively discussed during the last decades because of its direct and indirect effects on biological and photochemical activity which plays a key role in energy system of the Earth and its atmosphere. It is currently recognized that human are the major cause of changes on climate especially the surface energy balance in urban region with the naturally high population density, materials and social activity. Clear evidences confirm that the concentrations of a atmospheric greenhouse gas have increased over the past 100 years as a result of human-related activity which affects the radiation. Although, the solar radiation and energy study has been done for more than 40 years in Thailand, there are some constraints on the measuring instrument, scarce meteorological station and limited data acquisition due to lacking of financial and technical support. Therefore, most assessment of solar resource and technology is apparently based on the available measured data. However, the sparse distribution of data measurement in space and time, leads to the use of solar radiation modelling as the basis for many engineering and economic decisions. Many solar radiation models could be calculated from atmospheric measured data for either development or validation. Consequently, the uncertainty or accuracy of obtainable data is indeterminate. In comparison with diffuse and global radiation, direct solar radiation is 833

the primary beginning for solar radiation modelling. According to the physics principles and atmospheric modelling, there are two approaches for calculating the direct radiation derived from the conversion of astronomical parameters and atmospheric parameters. 2. Astronomical and atmospheric conversion 2.1 Astronomical parameter Declination angle is the angle between the earth–sun line and the horizontal plane of site. It varies over the year from -23.45 to +23.45 degree which can be calculated by the following equation :

Solar time The solar time of a given site is estimated by the difference between the site‘s longitude and the meridian of its time zone, and the yearly perturbations in the earth‘s rate of rotation around the sun. The relation between the local standard time and the solar time is given by:

Hour angle is the angular displacement of the sun east or west of the local meridian due to rotation of the earth on its axis at 15 per hour which is given by:

Solar zenith angle Solar zenith angle  z  is the angle between the vertical and the line connecting to the sun

834

Eccentricity correction-factor of the Earth’s orbit The earth‘s eccentricity induces a variation in the earth–sun distance of 1.7%. The eccentricity correction factor of the earth‘s orbit can be obtained using the following formula :

Extraterrestrial radiation is the amount of solar energy that would be received in the absence of the atmosphere. The change in the extraterrestrial radiation is caused by two sources, i.e., variation in radiation emitted by the sun and variation of earth–sun distance. This change is considerably taken into account the astronomical factors according to the following relation :

2.2 Atmospheric Parameter Air mass Air mass is the optical path length through the atmosphere where radiation is scattered and absorbed which is the reason why objects closer to the horizon and with larger zenith angles appear less bright than when they are directly overhead. To determine the solar resources necessary for solar energy applications, various models have been developed to predict the air mass. Kasten has also suggested a new formula:

Dew point temperature is the temperature below which the water vapor in air at constant barometric pressure condenses into liquid water at the same rate at which it evaporates. The condensed water is called dew when it forms on a solid surface. There is a simple approximation that allows conversion between the dew point:

835

3. Atmospheric physics model 3.1 Bird model The Bird model was originally developed in 1983 and updated in 1986. It includes transmittance of atmosphere for Rayleigh scattering, aerosol attenuation, water vapor absorption, ozone absorption, and uniformly mixed gas absorption. This former model has also been applied to assess the solar resource in many regions throughout the world. The direct solar radiation is given by

836

3.2 CSR model The CSR model has been used in many countries for solar radiation mapping, especially in the US. This model has a mathematical relation as shown in Eq. (18)

3.3 Kumar model The Kumar model is used to provide the backbone for the Solrad and Shortwave routines of the ArcInfo GIS software which have been developed for GIS applications in agriculture and ecology. It is partially based on the Kreith–Kreider and HJL. In this model, the incident DI is estimated by

On a horizontal surface its value is subject to the variation of zenith angle as

3.4 Daneshyar-Paltridge-Proctor (DPP) model 837

DPP model is developed from the model of Paltridge and Proctor, with corrected unit for zenith angle proposed by Daneshyar. As shown in Eq. (22), the model requires only the zenith angle for predicting the direct fraction of solar radiation. It is then one of the simplest models.

3.5 Meinel model This is also another simple model which converts the solar zenith angle and the air mass to obtain the direct solar radiation as

4. Statistical validation 4.1 Mean Bias Difference ( MBD ) In the present study, MBD is applied for validating the long-term performance of every models. The most accurate model has an MBD value closed to zero; a negative value means that the model underestimates the DI intensity while a positive value means that it overestimates. MBD is given by:

The mean values of these measured and predicted values are defined as.

4.2 Root Mean Square Difference ( RMSD ) In RMSD test, the predicted and measured data are compared to provide the short term performance of the models. The smaller the value of RMSD is, the better is the model‘s performance.

4.3 Coefficient of determination ( R 2 )

838

5. Results and discussion The results from 5 atmospheric physics modelling, 6 astronomical parameters and 2 atmospheric parameters were compared with the actual measured data from Pyrheliometer which showed a good correlation. The clear sky assumption was applied in order to analyze the determined direct solar radiation from five models. Ten hours from 07.00 – 17.00 LCT were considered as shown in Figure 1; time profile on 9-17 January 2014 representing winter and Figure 2; time profile on 8-9, 11-14 March 2014 representing summer. The calculated direct solar radiations revealed fine bell-shaped in a range of 403-504 Watt/m2 in winter and 489-651 Watt/m2 in summer 2014, correspondingly.

839

840

In addition, the dataset of continuous days from both seasons were validated with the MBD, RMSD and R2 of -0.04, 0.27, 0.93 and 0.02, 0.42, 1.97 respectively, which belonged to Kumar model for winter and Bird model for summer, detailed shown in Table 1.

841

In summary, Kumar model shows the best performance and simplicity with few governing parameters for direct solar radiation determination in winter. While Bird model with complexity and many converting parameters provided the best result in summer. 6. Conclusion The solar radiation typographical models and errors is challenging in terms of the uncertain reduction in measured data, as well as the evolution of the more effective and practical models without any difficulty and user friendly, such as using less input parameters, various conditions compatibility, etc. Therefore, the estimation of a direct solar radiation established by atmospheric physics models and empirical equations in this study will therefore be able to advantageously provide instantaneous and reliable values of the solar components for any site in the region without a constraint of actual measurement. In conclusion, on the basis of existing measured data, improving and validating model are still crucial and compulsory for atmospheric modeling which provides the solar radiation records and profile in Thailand. 7. Acknowledgements The authors are grateful to Thai Meteorological Department, Bangna, Bangkok for supporting raw data. And this work is sponsored partially by the graduate school of Srinakharinwirot University. 8. References Hewitt, C. N., & Jackson, A. V.(2009). Atmospheric science for environmental scientists. John Wiley & Sons. Behar, O. & Khellaf, A. & Abdallah, K.(2015). Comparison of solar radiation models and their validation under Algerian climate – The case of direct irradiance. Energy Conversion and Management, 98, 236–251. Sen, Z. (2008). Solar energy fundamentals and modeling techniques. London: Springer-Verlag. Duffie, J. A. & Beckman, W. A.(1991). Solar engineering of thermal processes. New York: John Wiley & Sons. Reno, M. J. & Hansen, C. W. & Stein, J. S.(2012). Global horizontal irradiance clear sky models: implementation and analysis. SANDIA report. SAND 2012-2389. Lawrence, M. G.(2005). The relationship between relative humidity and the dew point temperature in moist air – a simple conversion and applications. American Meteorological Society. BAMS, 225–233.

842

Gueymard, C. A.(2012). Clear-sky irradiance predictions for solar resource mapping and large-scale applications: Improved validation methodology and detailed performance analysis of 18 broadband radiative models. Solar Energy, 86, 2145–69. Geuymard, C. A.(2003). Direct solar transmittance and irradiance predictions with broadband models. Part I: detailed theoretical performance assessment. Solar Energy, 74, 381–395. Chow, W. K. & Wong, L. T.(2001). Solar radiation model. Applied Energy, 69, 191–224. Badescu, V.(1997). Verification of some very simple clear and cloudy sky models to evaluate global solar irradiance. Solar Energy, 61,251–64. Bird, R. E. & Hulstrom, R. L.(1981). A simplified clear sky model for direct and diffuse insolation on horizontal surfaces. Solar Energy Research Institute (now NREL), Golden, CO, SERI/TR-642-761. Rapp-Arraras, I. & Domingo-Santos, J. M.(2011). Functional forms for approximating the relative optical air mass. J Geophys Res-Atmosph , 116.

843

Civil Engineering (2)

Thursday, March 31, 2016

16:30-18:00

Room B

Session Chair: Prof. Chen-Wei Chiu ACEAIT-7497 Analysis of Mooring Effects on Floating Structures Po-I Chen︱National Cheng Kung University Jaw-Fang Lee︱National Cheng Kung University ACEAIT-7513 Analysis of Modified Double-Tee Slabs by Strut-Tie Model Ho-Seong Ju︱Dongguk University Sung-Yong Yu︱Dongguk University ACEAIT-7522 Analysis of Fire Extinguishing Performance of Backpack Water Mist System Chen-Wei Chiu︱National Taiwan Police College Yi-Liang Shu︱Central Taiwan University of Health Sciences and Technology ACEAIT-7564 An Experimental Study on the Lap Details Using Headed Bars and Hooked Bars for Flexural Members with Different Depths Kyu-Seon Lee︱Hanbat National University Se-Hun Jin︱Hanbat National University Seung-Hun Kim︱Hanbat National University ACEAIT-7549 Analysis of the Vehicles CO2 Emission by Using the Freight Vehicle Probe Data in the Tokyo Metropolitan Area Hsin-Yi Wei︱Tokyo Institute of Technology Daisuke Fukuda︱Tokyo Institute of Technology

844

ACEAIT-7274 Seismic Performance Evaluation of Hybrid Coupled Walls Chung-Chan Hung︱National Cheng Kung University Wei-Ting Lu︱National Cheng Kung University ACEAIT-8092 Improving Accuracy of Preliminary Cost Estimation Using Support Vector Regression Jieh-Haur Chen︱National Central University Indradi Wijatmiko︱University of Brawijaya Ragil Purnamasari︱National Central University

845

ACEAIT-7497 Analysis of Mooring Effects on Floating Structures a, b

Po-I Chena, Jaw-Fang Leeb Department of Hydraulic and Ocean Engineering, National Cheng Kung University, Taiwan *Corresponding Author: [email protected]

Abstract In this paper, mooring effects on motions of a floating structure subjected to incident waves are investigated. Various configurations of floating structures are considered including free floating structures without moorings, structures with simulated spring moorings, and structures with cable moorings. The geometry of the floating structure is considered as a rectangle, and two-dimensional structural dynamics with three degrees of freedom are used in solution computation. Coupled analytic solutions describing scattering and radiation wave fields, structure motions, as well as mooring Morison forces are presented. The present study indicates that (1) floating structures with moorings are subjected to motion constraint to some degree, and considerations of catenary cables have higher restriction on structure motions than using simulated spring moorings, (2) with considerations of wave forces acting on mooring lines, the added damping effects are obviously shown especially at resonant frequency, however, for cases of mooring cables, on the other hand, have less effects on structure motions than that of simulated springs due to their flexible nature. It is interesting to show comparisons of structure motions for different deployments of floating structures considered in this study, and the results can provide useful information for related study in the future. Keyword: floating structure, spring mooring, catenary cable, Morison force

846

ACEAIT-7513 Analysis of Modified Double-Tee Slabs by Strut-Tie Model a

Ho-Seong Jua and Sung-Yong Yub Department of Architectural Engineering, Dongguk University, Korea E-mail address: [email protected]

b

Department of Architectural Engineering, Dongguk University, Korea E-mail address: [email protected]

Abstract Modified dapped end, precast prestressed double-tee slabs were considered in this research. It can facilitate insertion of service ducts at the dapped ends. The total depth of the floor slab may be reduced. In addition, the underside of the double-tee slab showed simpler appearance. Three modified single-tee slabs were analyzed by strut-tie model method. Static three-point shear loading test was performed on full-scale specimens. The specimen failed during the test in the same location as predicted by the strut-tie analytical model. The analysis of experimental results in comparison to the analysis results revealed that the experimental failure loads manifested 108% of analytical failure loads on the average. Keywords: strut-tie model, modified double-tee, dapped end, slab, precast concrete, prestressed 1. Introduction Double-tee slabs are widely used as the major precast concrete slab material for supermarket buildings in South Korea and as large loading and long-span slabs of parking lots in overseas. When service ducts are installed at the bottom of the double-tee slabs, the total depth of the structure is increased, and it does not appear good. Two precast prestressed and one precast non-prestressed double-tee slabs are designed by using the strut-tie model method. Full-scale experimental shear tests were also performed on the three precast concrete single-tee specimens in this study. The experimental results were compared with the analytical results of the proposed strut-tie model. Then, the final failure load and pattern were also compared.

A new modified dapped ended double-tee slab model is proposed in this study as shown in Fig. 1. This double-tee slab is to accommodate the service facility ducts at the ends of the nibs. The double-tee slab assured that there is no interference in the legs of the double-tee slab when placing strands owing to the provision of an opening at the ends. The slab can have higher flexural strength at the center for uniform loadings. Moreover, it also resulted in increasing of the shear force due to the vertical loadings at each end with thicker concrete flange than that of the original double-tee slab as shown Fig. 2. 847

2 Modified Single-tee specimen 2.1 Specimen ST-1 This single-tee specimen of ordinary reinforcement without prestressing strands was designed1) as shown in Fig. 3. The length of the nib and the extended nib were 1,500 mm and 1,200 mm, respectively. Additionally, the depth of the nib including topping concrete of 100 mm was 250 mm. The depth of the nib decreased at 1,200mm from the dapped end. 4-U-bar D10 was used for shear reinforcement at the nib. 16-D19 was used for flexural reinforcement and welded to PL-60×1,200×10. 2-D29 bar for hanger reinforcement at the dapped end was inserted and welded to an anchor steel plate on the top and bottom of them. 4-U-bar D10 was installed below the loading point. In addition, main tension bars of 2-D35 were welded to steel angle (L-300×250×125×20) at the dapped end. The detail of the reinforcement at the section h-h of Fig. 3 is shown in Fig. 4.

2.2 Specimen ST-2 The Specimen ST-2 of Fig. 5 is basically the same as specimen ST-1 except that prestressing strands of 4-ø12.7 mm with 2-D35 was used as main tension bar in the leg.

2.3 Specimen ST-3 2-U-bar D13 + 2-U-bar D10 was inserted for hanger reinforcement of the specimen ST-3 as shown in Fig. 6. The length of the nib of specimen ST-3 was 1 m. However, the length of the extended nib of 225mm was very short in comparison to that of other specimens. The loading point was located 1 m from the dapped end.

848

2.4 Comparison of specimens Different hanger reinforcements and main bars were used for each specimen. The detail on the concrete strength and arrangement of the reinforcement of the three specimens are provided in Table 1. The hanger reinforcements of specimen ST-2 were the largest one among the specimens and used smaller size bar than that of the specimen ST-1. The specimen ST-3 had the smallest hanger reinforcement than the other specimens.

3. Experiment 3.1 Test method Three shear tests were performed on the single-tee slab specimen as depicted in Fig. 7. A hydraulic load tester was used at the rate of 0.5∼2.0 mm/min in displacement control mode to apply the load until failure. The loading point was located 700 mm from the dapped end. The LVDT was placed directly under the loading point.

The critical section of the single-tee slabs can be classified into two configurations according to the extended nib portion as follows. 1) Loading point acting on the inner part of the extended nib (Fig. 8) 2) Loading point acting on the outer part of the extended nib (Fig. 9) 849

3.2 Experimental result 3.2.1 Specimen ST-1 The first flexural cracks was observed at the bottom of the leg about 0.5 m away from the dapped end at the load of 35 kN. As the load increased, a number of new flexural cracks occurred at the bottom of the nib portion. Upon loading of 46 kN, small horizontal cracks occurred at the crossing point of the horizontal nib bottom line and the vertical dapped end line. At the load of 99 kN, diagonal web shear cracks occurred close to the dapped end and started to advance. Because the end of the extended nib became a plastic hinge, all cracks occurring in the web advanced toward this plastic hinge. At the load of 200 kN, the end of the extended nib showed a final failure as Fig. 10. The ultimate failure occurred simultaneously in the form of horizontal cracks between topping concrete and precast concrete. At the failure, the widths of the cracks, which was larger than 20 mm, were propagating diagonally in the direction of the loading point, where the end of the extended nib (45° area ( ) where nib section decreases) was located.

3.2.2 Specimen ST-2 850

Small horizontal cracks started to occur in the dapped end corner area at the load of 37 kN. At the load of 195 kN, the end of the extended nib manifested final failure as shown in Fig. 11. The other cracking patterns were similar to those of specimen ST-1.

3.2.3 Specimen ST-3 In the case of specimen ST-3 at the reaction of 48 kN, Small cracks occurred between topping concrete and precast concrete at the end of the extended nib as shown in Fig, 12. The final failure occurred at the end of the extended nib at the load of 68 kN while diagonal cracks were not observed.

1. Analysis 4.1 Truss Modeling The modified single-tee was divided into nib element and dapped end element as shown in Fig. 13. The two truss elements are described as follows. The dotted lines represent strut while solid lines represent tie.

4.1.1 Model for the Nib 851

Fig. 14 illustrates the strut-tie model for the nib. The dotted lines represent strut while solid lines represent tie. The horizontal struts including strut B-D at the upper part of the nib were placed at the center of the compression reinforcement. The inclined struts were to approach near 45 degree of the compressive force exertion in accordance with the shear provisions as specified in section 11 of ACI 3318-112). The vertical ties including tie B-C were not positioned at center of the shear reinforcement while the horizontal ties at the nib including tie A-C were placed at center of the steel plate.

4.1.2 Model for the Dapped End The depth of the compression block at the dapped end was computed to be 106 mm on the basis of equilibrium with compression forces at the top of the specimen as in Eq. (1). The upper horizontal struts (strut P-V and strut V-B') in Fig. 15 were placed at the center of the resultant. The area of vertical and horizontal ties at dapped end was the same as the section area of reinforcement. They were located at the center of reinforcement.

4.1.3 Superposition of Two Truss Models The truss model of Fig. 16 was designed by the superposition of two truss models, the nib model of Fig. 14 and the dapped end model of Fig. 15. The two truss models were connected only at the three upper nodes, P, V, and B'. The nodes in Fig. 16 can be categorized into two types. The black nodes were common pin nodes, and all member forces exerting to these nodes were in the state of equilibrium. Some members did not share white nodes. For example, tie O-Q shares the node Q with tie S-Q and 852

tie P-Q; however, the tie P-D', which represents hanger reinforcement, crossed without sharing the white node Q.

4.2. Determination of the strut width 4.2.1 Nib Strut The nib portion was analyzed by using the strut width shown in Fig. 173),4),5). The angle of inclined struts approached to 45°. Accordingly, the location of vertical ties was different from that of actual reinforcement, but the total section area of ties was the same as that of actual reinforcement. In other words, not actual value but proportional value was used for the spacing and area of the vertical ties in the nib portion for the analytical model. The width of struts was not constant along the length. The smallest width was considered for the conservative result.

4.2.2 Dapped End Strut Fig. 18 shows the strut-tie model at the dapped end. The dapped end portion was analyzed with the width shown in Fig. 18.6),7). The size of node D‘ zone was determined by the size of the steel angle (L-300×250×125×20). Additionally, the width of 290 mm between node V and node E‘ and the width of 20 mm between node B‘ and F‘ depended on the arrangement and size of the bars.

The smallest value among the whole strut widths was used for the strength calculation of the struts. For example, the width between node V and node E‘ was based for the estimation of the width of strut V-D‘. 4.3 Analytical result 853

4.3.1 Specimen ST-1 The result of the analysis by strut-tie model of the specimens was analyzed. The commercial FEM software MIDAS Gen program was used for the analysis. The first result of Fig. 19 was obtained when one of the ties yielded first or one of the struts first reached the maximum compressive strength in the target truss model by increasing the load. The first yield occurred at the tie V-W. The yield load was read 195 kN from the actuator with the reaction force of 145 kN at the left side. Fig. 20 shows the stress distribution when the final failure occurred in the model. After the tie V-W yielded first, the rest of the ties at the extended nib yielded progressively. The yield load was read 238 kN from the actuator with the reaction force of 174 kN at the left side.

4.3.2 Specimen ST-2 The stresses of specimen ST-2 were also calculated by the same method used for the specimen ST-1. The analysis results of the model 2 (for specimen ST-2) are shown in Fig. 21 and Fig. 22. The tie V-W yielded first as the same as model 1 (for specimen ST-1), and then the vertical ties in the extended nib yielded progressively. The yield of members of model 2 occurred in the same order as model 1.

854

4.3.3 Specimen ST-3 Stress distribution of model 3 (model for specimen ST-3) at the time if ultimate failure is shown in Fig. 23. The ties (tie N-O and tie P-Q) of the extended nib yielded with 69 kN at the support.

4.3.4. Comparison Analysis of the Experimental Results All specimens failed with diagonal cracks at the end of extended nib. It is reasoned that the specimen ST-3 failed due to the cracks separating topping concrete and precast concrete at the end of the extended nib. The specimen ST-3 did not display desired strength, and its ultimate failure occurred at the load that was lower than 70 kN reaction force at supporting point. The cause of failure can be attributed to the fact that the concentrated load applied to the thin flange, bringing on separation failure between topping concrete and precast concrete. The reaction at the support point transferred to hanger reinforcement, and diagonal cracks occurred in the joint region of the dapped end and the nib. This led to the loss of supporting resistant force when the cracks propagated to the loading point. Moreover, additional failures occurred due to a tensile failure at the end of extended nib, finally leading to the ultimate failure. The load-deflection curves of the three specimens are shown in Fig. 24.

855

The experiment result was compared with analytical result in Table 2. The analytical result also showed failures due to diagonal cracks at the same location as the end of the extended nib. The ratio of the experimental value divided by analytical value was 108% on the average for the three specimens, and the location of failure was same at the extended nib.

5. Conclusions A modified double-tee that can accommodate ducts at the ends, was considered in this study. The experiment and analysis by strut-tie model for the three single-tee slabs were carried out. Then, the analysis result was compared with the experiment result. Summarizing the results obtained from the course of the experiment and analysis, the following conclusions were derived. 1) The proposed strut-tie models were of two types and three models, and they all predictably failed at the same location as the experimental result. Additionally, the experimental failure loads also showed accuracy that manifested 108% of the analytical failure loads on the average. 2) Test of specimen ST-3 with 225 mm length of the extended nib of showed a brittle failure as the load increased with reaction force lower than 70 kN at the support. 3) All of the specimens failed due to diagonal cracks at the end of the extended nib in this test. It also showed at the same location in the analytical results by the strut-tie model method. Acknowledgements This research was supported by a grant (15CTAP-C077925-02) from Infrastructure and Transportation Technology Promotion Research Program funded by the Ministry of Land, Infrastructure and Transport of Korean government. 856

1. 2. 3. 4. 5.

6. 7.

References PCI Concrete Institute (2010), PCI Design Handbook 7th Edition ACI Committee 318 (2011), Building Code Requirements for Structural Concrete and Commentary, American Concrete Institute, pp.387-401. Christopher S. Williams, Michael D. Brown, and Oguzhan Bayrak (2010), Four-Column Bent Cap, American Concrete Institute, vol.273, pp.1-18. Adam S. Lubell and Katrin Habel (2010), Transfer Girder with Asymmetric Loading, American Concrete Institute, vol.273, pp.1-20. Julio A. Ramirez and John E. Breen (1991), Evaluation of a Modified Truss-Model Approach, Julio A. Ramirez and John E. Breen, American Concrete Institute, vol.88, pp.562-572 David H. Sanders (2002), Dapped-end T-beam supported by an inverted T-beam, American Concrete Institute, vol.208, pp.91-104 William D. Cook and Denis Mitchell (1988), Studies of Disturbed Regions near Discontinuities in Reinforced Concrete Members, American Concrete Institute, vol.85, pp.206-216

857

ACEAIT-7522 Analysis of Fire Extinguishing Performance of Backpack Water Mist System Chen-Wei Chiua*, Yi-Liang Shub Department of Fire Safety, National Taiwan Police College, Taipei, Taiwan b Department of Safety Health and Environmental Engineering, Central Taiwan University of Health Sciences and Technology, Taichung, Taiwan a Corresponding author : [email protected] (Chen-Wei Chiu) a1

Abstract During the early stage of a fire disaster, fire extinguishing efforts are primarily performed with portable fire extinguishers. However, the portability of such fire extinguishers means that the volume capacity for the extinguishing agent is necessarily limited, with use typically lasting only for about 10–30 seconds, which is generally considered insufficient time to put out a fire. A water mist system, on the other hand, can last much longer (1–3 minutes), while causing no environmental damage. It can also reduce the temperature surrounding a fire to prevent any sudden resurgence, thereby protecting the safety of people nearby. This study conducted a series of onsite fire extinguishing tests designed to measure the characteristics and performance of a backpack water mist system. The study aimed to compile information on existing backpack fire extinguishing systems to analysis nozzles on self-made systems by testing flow rate, distance, spray angle and fire extinguishing capability. Actual cylinder tests show that the amounts of time it takes for backpack water mist fire extinguishers to extinguish Type A and B fires were considerably shortened, which compensates for the shorter fire fighting time of a regular portable fire extinguishing system. Experimental data show that, at a distance of one meter, the best angle for fire extinguishing is 90°, followed by 60°, 30°, and 45°. Results show that the backpack water mist fire extinguishing system does have certain advantages in fire-fighting capability. Keywords: Fire extinguishing efforts; Extinguish agent; Water mist; Backpack 1. Introduction A backpack mist fire extinguishing system affords both mobility and a relatively long fire fighting time, which can be used to control the temperature around a fire. The effectiveness of its excellent fire extinguishing capability is well suited for deployment at complex fire scenes, for it can compensate for the poor performance of conventional fire extinguishing systems resulting from insufficient reach or too short an extinguishing period. Portable fire extinguishers, because of the portability consideration, are by default limited in terms of volume capacity. Such extinguishers generally will last only for about 10–30 seconds of operating time. If keeping a fire under control is the objective, then such a short amount of operating time is a bit too short, to say the least. In contrast, a backpack water mist system can extend the operating time up to three minutes, while also being non-toxic, environmentally friendly, and leaving the fire scene easier to clean up afterwards. Although the backpack water mist extinguisher is not currently fully developed as a complete fire fighting system, its considerable mobility in quickly arriving at the fire scene is an advantage. Since water mist in general has an excellent cooling effect and can be used quite effectively to govern surrounding fire temperature, the backpack water mist extinguisher has excellent as well as promising potential for further development.

858

With the phase out of halogen-based agents for environmental reasons, extensive efforts have been made to find the replacement in active fire protection [1]. The use of water mists for fire suppression and control is taken as one of the potential methods for alternative agents. Water mist is defined as spray in which 99% of spray droplets, by flow-weighted cumulative distribution, have diameters 0.05). The Lin‘s coefficient figures demonstrate that most of the subjects were estimated within a range of 2 standard deviations from the mean, which was between 1.98 years and -1.89 years for males and between 1.91 years and -1.82 years for females. In conclusion, the reference data set of southern Chinese based on dental age assessment can accurately estimate the age of 8 years old northern Chinese children. That is, it is not necessary for the northern Chinese to develop their own reference data set for dental age assessment. The work described in this abstract was fully supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. 17122914). Keywords: Application, southern Chinese, northern Chinese

1192

ACEAIT-7333 A Feature Amount Extracting Technique from Walking Signal Generated by Electrostatic Induction Koichi Kurita Faculty of Engineering, Kinki University, Japan E-mail address: [email protected] 1. Background/ Objectives and Goals Biometrics consists of techniques for individually recognizing humans based on one or more intrinsic physical or behavioral distinctions. Biometric characteristics can be broadly classified into two main categories—physiological and behavioral. Physiological characteristics are associated with the shape of the body in a broad sense. Examples of physiological biometrics include fingerprint, palmar print, hand geometry, vein authentication, ear recognition, face recognition, iris recognition, retina recognition, odor recognition, and DNA as the shape of the information about the human genome. On the other hand, behavioral characteristics are related to the behavior of a person. In this paper, we have presented a new direction for gait recognition without the use of a camera or video images. We have developed an effective non-contact technique for the detection of human walking motion using human-generated body charge. This technique involves the detection of an electrostatic induction current of the order of approximately sub-picoamperes flowing through an electrode that is placed at a distance of 3 m from a subject. The absolute value of the electrostatic induction current depends on the type of footwear and floor material. However, we confirmed that this technique detects the electrostatic induction current generated by the walking motion in daily life, because the proposed technique allows ultrasensitive detection of the electrostatic induction current. This technique effectively explains the behavior of the waveform of the electrostatic induction current flowing through a given measurement electrode through a capacitance model for the human body. Gait waveforms of 33 healthy individuals aged between 12 and 53 years were obtained by the proposed technique. As a pretreatment step, Fourier analysis was performed to obtain the gait spectrum. Subsequently, the differentiated waveform of the gait spectrum was obtained in order to derive the subtle characteristics from the gait spectrum. The Pearson correlation coefficients with each other of the differentiated waveform of the gait spectrum were obtained. Results show that there is poor correlation between each waveform. This suggests that human identification based on the subtle differences in the walking signal can be realized using the proposed technique. 2. Methods The schematic of the measurement system for detecting the very weak electric current generated by the stepping motion is shown. The electrostatic induction current flowing through the electrode placed at a distance of 3 m from the subject was converted into voltage using an I-V converter comprising an operational amplifier. The conversion ratio of the I-V converter was 3 V/pA. A very weak electric current of the order of sub-picoamperes was measured. The induction current generated by the commercial power source used in the experiment appears as noise in the measured electrostatic induction current. To remove this, we use a filtering system with a cutoff frequency of 20 Hz. Therefore, the measurement system is unaffected by the noise from other electronic devices such as mobile phones or microwave ovens. The static electric field generated by the presence of charged material around the measurement electrode affects the absolute value of capacitance C, which is formed between the subject‘s body and a 1193

given measurement electrode. However, the static electric field has little effect on the measurement, because the instantaneous electrostatic induction current arises from the change in capacitance C according to the subject‘s walking motion. In fact, the transient current generated can be detected only when the subject‘s body is in motion. 3. Expected Results/ Conclusion/ Contribution In this paper, we have presented a new direction for gait recognition without the use of a camera or video images. The change in the electric potential of the human body that is caused by the walking motion induces an electrostatic induction current in the electrode placed at a distance of a few meters from the human body. Using this technology, we have developed an effective non-contact technique for the detection of human walking motion by detecting the change in this human-generated body charge. This technique involves the detection of an electrostatic induction current of the order of approximately sub-picoamperes flowing through an electrode that is placed at a distance of 3 m from the subject. This technique effectively explains the behavior of the waveform of the electrostatic induction current flowing through a given measurement electrode through a capacitance model of the human body. Gait waveforms of 33 healthy individuals aged between 12 and 53 years were obtained by the proposed technique. By data processing techniques such as differentiation and normalization after Fourier analysis of the obtained waveform, we obtained the Pearson correlation coefficients with each other. Results show that there is poor correlation between each waveform. This suggests that human identification based on subtle differences in the walking signal can be achieved using the proposed technique. In our future work, we intend to improve the data processing method to realize real-time recognition. Keywords: Walking signal, Electrostatic induction, Feature amount, Correlation

1194

ACEAIT-7421 Effects of Component Design on Dislocation and Stress Distribution after Prosthetic Impingement in Total Hip Replacement Jian-Horng Chena, Wei-Min Chib, Chien-Chung Linc, Hsin-Hsien Yeha, Ying-Jui Hod, Chin-Liang Kuoa a School of Physical Therapy, Chung Shan Medical University, Taiwan, ROC E-mail address: [email protected] b Graduate School of Gerontic Technology & Service Management, Nan Kai University of Technology, Taiwan, ROC c Orthopaedic Department, Armed Forces General Hospital Taichung, Taiwan, ROC d School of Psychology, Chung Shan Medical University, Taiwan, ROC Total hip dislocation, in which the femoral head moves out of the acetabular cup, is a complicated problem of total hip replacement and is second in importance only to liner-wear-related aspectic loosening. An insufficient range of motion can lead to impingement of the prosthetic neck on the acetabular component. Recurrent prosthetic impingement can cause material failure in the liner, and if the rotating torque exceeds the limit that the total hip can endure, dislocation occurs. The aim of this study was to use the finite element (FE) method to analyze the torque limits leading to dislocation, i.e. dislocation torque, and the contact stresses on the impingement site and the egress site of the acetabular liner when the femoral component sustains a torque after prosthetic impingement. The femoral component was composed of a CoCr femoral head and a titanium femoral neck, and the acetabular cup contained a UHMWPH liner, the back of which was considered perfectly fixed to a titanium shell. Articular contact was modeled as frictional sliding between the femoral head and the liner. A 942N of joint contact force along the femoral neck axis, almost equal to 1.5 times of body weight, was given. The FE meshes were automatically generated by the commercial software ANSYS. After a series of convergence tests, the FE models with the same sizing (90406 tetrahedral elements) were used to calculate the dislocation torques and the stress distributions. Dislocation in these simulations was defined as occurring at the point where the analysis had stopped to converge due to missing constraints. Here, different acetabular cup designs, the common cups and the ChenLin Cup, a newly described elongated cup with a preferred impact mode, were compared. The results showed that, among these cups, the ChenLin Cup can endure the largest torque, and can effectively disperse the stresses on the impingement site. In addition, the total hip structure with a big head can have a larger dislocation torque than a small head, and the maximal stresses on the impingement site and the egress site are less. Conversely, the hemispherical cup provides the least amount of dislocation stability. Keywords: total hip dislocation, finite element analysis, acetabular cup design

1195

ACEAIT-7570 Production of Bonded Sheet Metals Using a Hybrid Method of Rolling and Shear Deformation Ryoichi Chibaa, Kazuya Inadaa a Department of Mechanical Systems Engineering, National Institute of Technology, Asahikawa College, Japan E-mail address: [email protected] Yumika Sugab b Advanced Course of Production System Engineering, National Institute of Technology, Asahikawa College, Japan 1. Background/ Objectives and Goals It is known that sheet metals produced by accumulative roll-bonding (ARB), which is a severe plastic deformation process, have a drawback that they are likely to debond at the bonded interfaces when subjected to excessive loading. Thus, bond strength-enhancing technologies for roll-bonded sheet metals have been expected to be developed. In the present study, we propose a new method to enhance the bond strength: the equal channel angular rolling (ECAR) is performed immediately after the conventional roll bonding so that additional plastic (shear) strain is introduced to the bonded interface to enhance the bond strength. 2. Methods The proposed method was applied to 1-mm-thick AA1100-O aluminum alloy sheets. Strips of 25 mm in width and 250 mm in length were prepared from the aluminum sheets. First, the strips to be bonded were wire-brushed and then cold-rolled at reductions of 50-60%. A self-manufactured rolling machine with a pair of 150-mm diameter rolls was used under non-lubricated condition. Second, the bonded strip was ECARed using a die with 115° corner angle to introduce shear strain across the full thickness. Subsequently, the effectiveness of the proposed method for improving the bond strength was quantitatively examined through the T-peel tests. 3. Expected Results/ Conclusion/ Contribution Compared to the conventional roll bonding, the proposed method resulted in poor bond strengths for rolling reductions of ≈ 50%. However, the magnitude relation in the bond strength was reversed for reductions of ≈ 60%; the proposed method yielded about 40% higher bond strength than the conventional roll bonding. This is probably because if the roll-bonded object does not have a bond strength high enough to endure the subsequent shear deformation, the object is partially debonded when undergoing the shear deformation, leading to a decrease in the eventual bond strength. In conclusion, it was found that the hybrid method of rolling and shear deformation, which is proposed in the present work, may improve the bond strength for rolling reductions over a certain value, owing to the synergy effect of ECAR. Keywords: aluminum alloy, bond strength, cold roll bonding, equal channel angular rolling, peel test

1196

ACEAIT-7551 Application of Pyrimidine Derived Conjugated Polymers to Bulk Heterojunction Organic Solar Cell Yejin Changa, Dongjun Kima, Hongsuk Suha,* a Dept. of Chemistry, Pusan National University, Korea E-mail address: [email protected] 1. Background/ Objectives and Goals Organic solar cells (OSCs) have attracted large attention for their advantages such as their flexibility, light-weight, low cost and easy fabrication. We designed and synthesized three novel conjugated polymers, PTTICN, PTTICNR and PTTIFR containing 2-pyriminecarbonitrile and 2-fluoropyrimidine, which were applied to OSCs. 2. Methods By Stille coupling reaction using Pd(0)-catalyst, 2-pyriminecarbonitrile/2-fluoropyrimidine fragments and 6-(2-thienyl)-4H-thieno[3,2-b]indole (TTI) were coupled to synthesize PTTICN, PTTICNR and PTTIFR. The fabrication of the OSCs was conducted as the configuration of ITO/PEDOT:PSS/polymer:PC71BM/Al. After fabrication, the properties of the polymers were investigated with UV-Vis spectroscopy, cyclic voltammetry (CV) and atomic force microscopy (AFM). 3. Expected Results/ Conclusion/ Contribution The optical band gaps from UV-Vis absorption onset of PTTICN, PTTICNR and PTTIFR were about 2.0 eV. The HOMO energy levels of four polymers (PTTICN, PTTICNR and PTTIFR) were at -5.26 ~ -5.34 eV, their LUMO energy levels were at -3.47 ~ -3.50 eV, corresponding to the electrochemical band gap of around 1.8 eV. The device comprising PTTICN and PC 71BM (1:4) with DIO showed a VOC of 0.82 V, a JSC of 6.38 mA/cm2, and a fill factor (FF) of 0.54, giving a power conversion efficiency of 2.81%. The device comprising PTTICNR and PC71BM (1:4) with DIO showed a VOC of 0.81 V, a JSC of 6.38 mA/cm2, and a fill factor (FF) of 0.43, giving a power conversion efficiency of 2.22%. The device comprising PTTIF and PC71BM (1:4) with DIO showed a VOC of 0.64 V, a JSC of 7.38 mA/cm2, and a fill factor (FF) of 0.55, giving a power conversion efficiency of 2.61%. Keywords: Organic solar cell, Bulk heterojunction, Pyrimidine

1197

ACEAIT-7527 Electrochemical Deposition of Paclitaxel/Hydroxyapatite Composites on Ti Alloy for Biomedical Applications Yu-Mei Chenga, Shuei-Bin Laia, Yu-Liang Laiab, Shiow-Kang Yena* a Materials Science and Engineering , National Chung Shing University, Taiwan b Physical Medicine and Rehabilitation , China Medical University Hospital, Taiwan * E-mail address: [email protected] 1. Background/ Objectives and Goals With the average life expectancy getting longer and life habits changing, cancer is becoming one of the leading causes of death. In order to reduce the side effects of chemotherapy, many studies about the target treatment have been investigated recently. In this study, Paclitaxel (Ptx), the drug for cancer treatment, is electrochemically deposited on Ti alloy which could be applied to vascular stents for the tumor treatment by sustaining release of drug to achieve the apoptosis of cancer or the prevention of cancer metastasis. 2.

Methods

3. Expected Results/ Conclusion/ Contribution When electrolytic deposition potential is not negative enough such as -1.0 V (Ag/AgCl) to generate plenty of hydroxyl bonds, the quantity of paclitaxel deposited on the specimen is low (133±22μg/cm2). When the applied potential is more negative such as -2.0 V (Ag/AgCl), excessive hydrogen bubbles are formed on the specimen to hinder Ptx deposition. Tuning the applied potential around -1.8 V, the highest drug loading (395±95μg/cm2) is achieved. Besides, the Ptx drug loading on hydroxyapatite (HAp) post coated specimen (Ptx/HAp) is much more (572±99μg/cm2). It is speculated that porous HAp has greater surface area and previously provides OH bonds for enhancing the absorption of Ptx. The releasing profiles of paclitaxel coated and Ptx/HAp coated titanium alloy specimens in 3 ml PBS at 37 °C are shown in Fig. 1. The curves can be divided to two stages, including the initial burst (day 1) and the sustaining release (day 2-30). The Ptx coated specimen releases about 40 % of the total drug content for the first day burst. Since paclitaxel is hydrophobic insoluble in water, it can not be fully released in short term. After releasing a month, the cumulative release of paclitaxel is only 65% of all, revealing a decreasing rate. In contrast, though the initial burst of Ptx/HAp coated specimen is only 20% of all in the first day, it reveals a sustaining release for 90% of all in a month. The Ptx embedded in HAp leads to the lower burst in the first day, and the nanosized Ptx results in the higher release rate due to its 1198

more specific surface area. The MTT test of human endothelial cells on PS, uncoated, Ptx coated, and Ptx/HAp coated Ti alloy specimens are shown in Fig. 2. The amount of cell survival in medium with the Ptx coated specimen was much lower than the uncoated one. There is little difference in cell viability between Ptx coated and Ptx/HAp composite coated specimens. The cell viability for both coated specimens is only 30 % of the uncoated specimen, indicating that the medical function of Ptx is not affected during the electrochemical process.

It is concluded that paclitaxel can be successfully deposited on the titanium alloy by electrochemical method and the post HAp coated specimen with high porosity can enhance the drug loading from 395 to 572 μg /cm2 and reveal the more complete drug release. Besides, the electrochemical processing does not vary any chemical bonding or destroy any medical function of Ptx, revealing a practical method for drug loading. Keywords: Electrochemical deposition; Paclitaxel (Ptx); Hydroxyapatite (HAp); Ti alloy; Drug load and release.

1199

ACEAIT-7526 Preparation and Characterization of Pt/Carbon Black-Vivianite Composites as Catalysts for Fuel Cells Yu-Chuan Lina , Yu-Ching Chenga, Yen-Min Wanga , Shiow-Kang Yena* Materials Science and Engineering, National Chung Hsing University, Taiwan *E-mail address: [email protected] 1. Background/ Objectives and Goals According to our previous results, Pt nano-particles were successfully reduced on vivianite (Vi, Fe3(PO4)2·8H2O), assigned to Pt/Vi and then uniformly mixed with the carbon black (CB) to obtain Pt/Vi/CB bi-function catalysts. However, ferrous phosphate formed by iron (II) ammonium sulfate and diammonium hydrogen phosphate has shown poor electronic conductivity. Therefore, in this study, we attempt to prepare Vi directly deposited on acetylene carbon black, the surface of which is modified to generate more OH- radicals for strengthening the bonding between Vi and CB to improve its conductivity, finally nano-sized Pt reduced on to form Pt/CB-Vi catalysts. 2.

Methods

3. Expected Results/ Conclusion/ Contribution The electric resistance has been reduced from 268.0 for Pt/Vi to 119.0 (Ω) for Pt/CB-Vi. The platinum particle about 4.5 nm coupled with Vi on CB is observed by TEM, as shown in Fig. 1. The electrochemical surface area (ECSA) increased from 649.18 to 1016.89 (cm2/mg) after two times CB added in and assigned to Pt/CB-Vi/C, analyzed by cyclic voltammetry of hydrogen absorption / desorption reaction, as shown in Fig. 2. In methanol oxidation reaction (MOR), the mass activity increases from 14.2 for Pt/Vi to 36.1 and 73.5 (A/gPt) for Pt/CB-Vi and Pt/CB-Vi/C at 0.6V. No forward or reverse current peaks are observed for the latter two, indicating no CO poison effects, as shown Fig. 3. In membrane electrode assembly (MEA) tests, the prepared Pt/CB-Vi catalysts reveals the greater power density and open circuit voltage than Pt/C (Alfa) and Pt/Vi as shown in Fig. 4, since Fe3+ in these catalysts activate the water at low potential to yield Fe-OHads and detoxify the nearby nano-sized Pt poisoned by CO to create the opportunity for forming CO2 and H+, beside the enhanced conductivity. This argument also reasons results of MOR tests out, as shown Fig. 3.

1200

Obviously, both electric resistance and CO poisoning effect play the major role on the performance of catalysts in DMFC, and the improved Pt/CB-Vi from Pt/Vi has revealed the acceptable performance due to its enhanced conductivity and detoxifying function. Keyword: Vivianite (Vi); Electronic conductivity; Carbon black (CB); Direct methanol fuel cell (DMFC); Membrane electrode assembly (MEA)

1201

ACEAIT-7523 Preparation and Characterization of Pt/Carbon Black-Iron Contained Hydroxyapatite Composites as Catalysts for Fuel Cells a

Yung-Cheng Lina,,Yu-Ching Chenga,Yu-Wen Chena,, Shiow-Kang Yena* Department of Materials Science and Engineering, National Chung Hsing University, Taiwan E-mail address: [email protected]

1. Background/ Objectives and Goals According to the previous study in our lab, hydroxyapatite (HAp) microspheres were used not only in biomaterials as kinds of drug carriers and dental ceramics owing to its awesome bio-compatibility but also in fuel cells as novel supports of catalysts. On them, nanosized platinum (Pt) particles were successfully reduced to become bi-functional catalysts after ion exchanged with iron to form FeHAp, revealing the excellent electrochemical mass activity in methanol oxidation reaction (MOR) without obvious CO poisoning effects after mixed with carbon black (CB) to derive Pt/FeHAp/CB. However, it is known that hydroxyapatite is a sort of ceramic with very poor electronic conductivity (7×10-13 S/cm) and ionic conductivity (2×10-7 S /cm). In order to enhance its conductivity, the precipitation of HAp is carried out directly on CB, the surface of which is modified to generate more OH- radicals for strengthening the bonding between HAp and CB, to from CB-FeHAp composites. 2. Methods

3. Expected Results/ Conclusion/ Contribution The platinum particle reduced by ethanol sized from 2.64 to 4.42 nm is observed on CB-FeHAp composite to from Pt/CB-FeHAp composites by TEM as shown in Fig.1. Besides, cyclic voltammetry indicates that IR-drop decreased with the increasing ratio of CB to FeHAp and the de-sorption peak potential is obviously shifted from -0.05 V to -0.189 V (vs. Ag/AgCl/ saturated KCl). The electrochemical surface area (ECSA) without extra carbon added ranges from 429.37 to 861.50 (m2/g) while the other with extra carbon added in (Pt/CB-FeHAp/CB) ranges from 670.91 to 1149.70 (m2/g) as shown in Fig. 2. In methanol oxidation reaction (MOR), Pt/CB-FeHap illustrates the highest mass activity 233 (A/gpt) and the lowest on-set potential 0.28 V as shown in Fig.3. No forward or reverse current peaks are observed, indicating scarce CO poison effects.

1202

Fig. 2.(Left)Cyclic voltammograms Pt/CB-FeHAp and Pt/CB-FeHAp/CB catalysts carried out in N2-saturated 1.0 M H2SO4 at a scanning rate of 10mV s-1. Fig. 3.(Right) Cyclic voltammograms at the 100th cycle of methanol oxidation of Pt/CB, Pt/CB-FeHAp and Pt/CB-FeHAp/CB catalysts carried out in 1M CH3OH + 1M H2SO4 at a scanning rate of 50 mV s-1. In this study, the surface of acetylene CB has been oxidized to generate OH bonds to enhance the precipitation of HAp dispersed on them to form CB-HAp supports and then ion exchanged with ferrous ions to form CB-FeHAp, and finally nano-sized Pt are reduced on and consequently to from Pt/CB-FeHAp bi-function catalysts, detoxifying the effect of CO. Keywords: Bi-functional catalysts; Methanol oxidation reaction (MOR); CO poisoning; Electronic conductivity; Pt/CB-FeHAp

1203

ACEAIT-7346 OpenStack-based Smart Device Cloud Automation Test System Shih-Ying Chen, Hung-Ming Chen*, Feng-Ming Wu and Cheng-Feng Ke Computer Science and Information Engineering, National Taichung University of Science and Technology, Taiwan E-mail address: [email protected] Abstract Traditional software test needs to spend extensive manpower, and it also occupies considerable cost in software development. DevOps process and tool can shorten time for developing and deployment testing, as a result, this study proposes an Android-based smart device cloud automation test system with DevOps process. It is based on OpenStack cloud system; lightweight virtualization technology, Docker, solve deployment time attrition which caused by different test environment; use continuous integration server Jenkins tool and version control server GitLab combination. Cloud automation test framework that proposed with Android applied software automation test verification can apply to actual environment. Keywords: OpenStack, DevOps, Docker, Continuous Integration and Continuous Delivery, Android Automation Testing 1. Introduction Traditionally software testing is demanding in terms of cost and labor and can take up 40% 70% of the development costs (Yu, Tsai, Chen, Liu, Zhao, Tang & Zhao, 2010). One of the biggest issues is the communications between those who develop it and those who operate it. The testing is a very important part before the software is delivered. The software cannot be delivered if it fails in the testing process and that can be a catastrophe. For this, DevOps is introduced for the tight bonding between the developers and operators to reduce the development time, carry out the testing as quickly as possible, which contributes to early detection and modification of any bugs. The earlier the bugs are found and eliminated in the software, the lower the costs will be for the subsequent development and maintenance and the better the software‘s quality is. In the O‘Reilly Velocity conferences on 2009, a Flickr team presented a report saying that Flickr was successful in bringing the development and operations together and therefore capable of publishing more than a dozen of versions per day (Bang, Chung, Choh & Dupuis, 2013). At the same time, the DevOps has changed the cooperation model between departments by turning manual labors that that tedious and prone to mistakes to automatic. The automated results were digitalized to facilitate quick detection and solving of problems. At the end, departments were able to share their information and the experience people obtained while working on the project, thus making the later development works smoothly instead of spending the energy in negotiating back and forth. The applications grow increasingly sophisticated with the increasing number of smart devices in recent years. Specific development tools and frameworks are combined with appropriate testing techniques and tools to ensure that the security and quality of the applications (Amalfitano, Fasolino & Tramontana, 2011). Android is a popular mobile device platform. The reliability of applications developed for Android now becomes the center of attention. For this, a cloud automatic testing platform integrated with OpenStack (Wen, Gu, Li, Gao & Zhang, 2012) is established in this study. An open source cloud program, OpenStack, provides an 1204

infrastructure that allows users to create and provide their own cloud computing service (Yang, Liu, Liu, Chuang & Jiang, 2013). Therefore, the use of OpenStack to provide a test environment allows quick deployment of programs in OpenStack for testing. The good thing about using OpenStack is that bitmap files can be established for various environments depending on the testing needs; i.e. the dependency between the software being tested and the environment can be tested. The integration tool, Jenkins (Kawaguchi, 2011) is brought in for automatic testing and deployment. Finally, the teste results are shown on the Jenkins dashboard. Once the test results come out successful, the next step is packaging and delivery. 2. Related Work The DevOps process is introduced and integrated with smart devices testing, cloud computing and featherweight virtualization in this study for the study of OpenStack cloud computing, DevOps, their associated techniques and Docker. The literatures collected for this study will be reviewed as follows. 2.1 OpenStack The development of cloud computing evolves in the speed of light in recent year. Relevant techniques emerge almost on a daily basis. One that grabs serious attention is the OpenStack. An open-source cloud software developed jointly by NASA and Rackspace (Sefraoui, Aissaoui & Eleuldj, 2012), OpenStack is developed to provide cloud service providers or corporations the ability to establish their own IaaS environments. With major international players coming to play, OpenStack has evolved into the mainstream of cloud technology, as currently more than 20,000 people in a good number of international technology firms are involved in its development. OpenStack consists of three modules, the computing, network and storage modules. A dashboard module for centralized management is added to complete the OpenStack package for shared service. Computing resources are provided in the form of virtual machine for flexible expansion or allocation OpenStack evolves so quickly that almost a new version is delivered every 6 months. There have been 12 versions since 2010 with the latest being the Liberty. It started with only 2 packs and now there are 20. OpenStack is a could software consisting of many packs of function. The core packs are the Horizon, the dashboard for management interface, Keystone for identification, Nova for computing, Swift for object storage, Cinder for block storage, Neutron for network communications and Glance for bitmap file management, each of which is explained below (Kumar, Gupta, Charu, Jain & Jangir, 2014). 2.2 DevOps A traditional software organization breaks development, IT operations and quality assurance (QA) down into several different departments. It has become a significant issue as how to introduce new development model, such as Agile, to such an environment. If doing things the old-fashioned way, the communications between departments would be the cause that slowed the coordination process down and held it back, and that is why DevOps was born. The DevOps is the combination of ―Development‖ and ―Operations,‖ a generic name for the processes, methods and systems that are used to facilitate the communications, cooperation and integration between the departments of development IT operations and QA (Debois, 2011). By summing up the above, the purpose of DevOps is to encourage the cooperation between developers, operators and other people, such as QA, through communications, cooperation, integration and automation. 1205

DevOps is designed to make an organization better and accelerate delivery. Therefore, continuous integration and delivery are required. The introduction of automation in massive scale prepares the necessary resource for the development team, eliminates repetitive works and thus reduces manual mistakes. More importantly, the comparison of correlated data is enabled to allow early discussion and, therefore, the improvement of software glitches and accelerated publication of new versions. The continuous integration and delivery are presented in the following section. 2.2.1 Continuous Integration The difficulty to integrate program codes from different people is proportional to the codes to be integrated. That is to say, the later the integration and testing take place, the harder the integration will be and thus the more like to fail. Therefore, the purpose of the continuous integration is to integrate the codes of new functions into the existing codebase for testing as soon as possible(Karlesky, Williams, Bereza & Fletcher, 2007). This way, will contribute to the discovery of problems unexpected in integration and the solution to conflict codes. The earlier the glitches are found and fixed, the more time and money will be saved. Automation is brought in to achieve the purpose above. There are many tools for continuous integration. One that is flexible and widely used is Jenkins. Jenkins is a web-based server for continuous integration and an open-source project based on MIT License. Its design features high expandability and flexibility. Many developers have contributed to its derivative packs that are used extensively in many major projects and companies. The developers and operators only have to work on the Jenkins web-based dashboard to achieve easily automatic integration and testing. The rich visualized reports and graphics allow developers and operators for visualized testing. 2.2.1 Continuous Delivery In the past when a version was to be delivered, there was usually a deployment team to deploy and test. Most of the time, however, the results were different from deployment to deployment, as there were too many variables that might lead to deployment failure. The deployment procedure was usually in documents, but the rapid updates of software rendered these documents obsolete in an instant. That is why publishing software version was a process that is time consuming and full of risks. The automation from the deployment of software to testing and the final publication is called the continuous delivery, the purpose of which is to deliver new functions to the hands of end user as quickly as possible (Dlugi, Brunnert & Krcmar, 2015). The deployment time is reduced and human or environmental problems solved through continuous delivery, not to mention making the software delivered to end users more reliable. Jenkins can deliver the continuous delivery. Shell script is used for automatic deployment and launch of products, and Jenkins is used to turn manual deployment and launch to automatic. The process requires no more time from operators and mistakes are minimized. If there is an error in deployment and launch, it will be shown on Jenkins dashboard for quick improvement and modification. It is clear from the above that continuous delivery is the ultimate goal of every team, and continuous integration is an inevitable step toward continuous delivery. When repetitive works are required for something found in the software development and operations, it can be automated, and the automation allows the team to spend the time saved on something more important. The ―automation‖ is used to achieve the goal of ―being continuous;‖ in other words, the degree of automation has everything to do with the capability of continuous delivery. 1206

Figure 1 shows that DevOps brings continuous integration and delivery much closer together, which makes the time of every version delivery shorter and makes it easier to identify and solve problems as early as possible. DevOps also brings developers and operators to a much closer operation and therefore prevents potential problems in communications between departments. In this study, the continuous integration server, Jenkins, is employed to achieve both continuous integration and delivery. Jenkins features a rich array of plug-ins and integrated reports, and provides web-based dashboard interface to allow developers and operators to find out quickly and easily the entire picture of a project.

2.3 Docker Docker is an open-source code project that deploys lightweight operating environment very quickly (Chamberlain & Schommer, 2014). With supports for multiple Linux platforms, Docker provides an additional automatic management mechanism for virtualization of software abstract layer and operating system layer. Docker uses the resource separation mechanism (cgroupps) and name space in Linux Kernel to create an independent environment, and that environment is called Docker Container. The Container stores applications and their elements during the executions together by combining the lightweight application separations and in a bitmap-based deployment. Container packs applications, libraries and other dependent binary files all together to maintain the independence of the applications. By doing so, the conflicts between applications for relying on the same important elements in the hosting system are avoided. The Container does not have a (OS) core, making is faster and more nimble as opposed to a virtual machine. Figure 2 shows that Docker is different from traditional virtualization in that the traditional virtualization requires the installation of guest OS, while the Container works directly on host OS in local host, therefore, no need to install guest OS in Container to run an application. Since Docker no longer requires the presence of guest OS, the starting speed is now in the order of seconds instead of minutes as opposed to a virtual machine, meaning that Docker is capable of deploying many applications very quickly. Docker is attached to the core of Linux. The performance of Docker and a virtual machine was tested in literature (Felter, Ferreira, Rajamony & Rubio, 2014) and the result showed that 1207

Docker performed almost as well as the original and better than virtual machine in every aspect. That is why many companies started to introduced Docker applications and for DevOps, Docker saves the time efficiently for development, testing and deployment.

Docker is a client-server structure as shown in Figure 3. The Docker Client is in charge of communicating with the Docker Daemon of the host to execute docker commands such as docker build, docker pull and docker run, whereas Docker Registry is a warehouse for Docker Images. The three elements of Docker are:  Docker Image: it is a read-only template used to execute a Docker Container.  Docker Container: the Docker Container is an independent environment built upon Docker Images.  Dockerfile: it is a file instruction set used to describe how to create a Docker image automatically. Docker is capable of create Docker Images automatically through Dockerfile, and uses Docker Client to run a Docker Image. With this feature, it is possible to create quickly the bitmap file for packs of different dependence. For example, Dockerfile can be used to establish applications with packs of different dependence on the same host. Therefore, Docker saves a lot of deployment time for testing, and is capable of using all of the Docker features to build many containers in one container very quickly for testing. 3. System Design The OpenStack-based automatic testing system proposed in this study incorporates Jenkins for continuous integration and delivery. OpenStack is used to provide a test environment. The software to be tested is packaged with its dependent packs via Docker. The advantage of doing so is the possibility to establish multiple environments to test the dependence between software and environment. The integration of Jenkins, version control, OpenStack and Docker allows developers to test and generate graphic reports every time they develop a new version. The operators are allowed to schedule regular integration and acceptance tests via Jenkins. Packaging and delivery of new version may commence as soon as the tests are successful. The system overview and process are described below.

1208

3.1 System Overview The system consists of three parts: OpenStack, Jenkins and version control server, as show in Figure 4. The dotted lines in the figure are settings in advance. Web Hook settings are required for the version control server and Jenkins because Jenkins is triggered to perform the corresponding job when a developer uploads a version to the version control server. On the other hand, OpenStack triggers the OpenStack virtual machine and create a test environment through APIs when Jenkins receives a task, and the software to be tested is packaged with its dependent packs for testing using Docker. At the end of test, graphic reports are generated on Jenkins dashboard for developers and operators. If the test is successful, the webpage application is packed with Shell Script and sent to Docker. For the smart device Android system, it is packed into apk for final delivery.

3.2 System operation process Jenkins is used for continuous integration, as well as the final packaging and delivery in order to provide automatic testing process for developers and operators. The combination of OpenStack and Docker reduces the risks of building an environment and improves the software 1209

dependence test, giving better quality to the software. For this, this section provides a detailed description about how to use the OpenStack-based environment for continuous integration and delivery according to the structure presented in the previous section, as shown in Figure 5. The Web Hook was established in advance between the version control server and Jenkins. Therefore, as a developer or operator uploads a version of software to the version control server, Jenkins is triggered to perform this job and acquires Keystone authorization through APIs in order to establish a virtual machine. Once the virtual machine is up and running, it acquires the version uploaded from the version control server and starts the test for this run. If the test is successful, Jenkins will execute the post-establishment actions and establish the new version through Shell Script. The software tested this time will be packaged to Docker and the virtual machine provided by OpenStack is used to run multiple versions of Docker applications. Finally, the webpage application is packaged to Docker, and the package is in apk format for Android or iOS, respectively, as the final version of delivery. At the end the results of this run are shown on the Jenkins dashboard for the immediate access of developers and operators.

4. System Verification Android APPs are used and packaged into apk for testing in order to verify the automatic testing method proposed. The Monkeyrunner GUI (Graphical User Interface) is selected as the test tool to allow users to send control commands to a piece of Android equipment through instruction sets (Spreitzenbarth, Freiling, Echtler, Schreck & Hoffmann, 2013). An easy-to-program test framework is design to simplify the programming threshold for the test scripts. The details of the Monkeyrunner-based GUI test frameworks are provided below. Its framework is shown in Figure 6.  order.py: defines the execution sequence and counts of test process in an array;  button_define.py: defines name and location of buttons in an array;  testing_flow: is the test flow of GUI test that defines one or more test flows in an array. Every test flow is a test item. For the purpose of this study, three actions are provided for testing: tap, drag and delay;  runner_lib.py: analyzes the test flow sequence of order.py, identifies the actions of every 1210

test flow script and executes the corresponding actions. Currently three actions are provided for testing: tap, drag and delay;  main_init_compare.py: is the file to be executed first for the test. It starts by capturing the name of the test flow in testing_flow and build the corresponding directory to keep the logs of every test.  main_runner.py: is the test flow module where runner_lib.py, order.py, button_define.py and test flow scripts are loaded. It executes every test flow in sequence. The results are saved in the corresponding directory at the end of test.

The connection between Jenkins, OpenStack and version control server is established as the test scripts are completed. Therefore, every time when a new version is delivered, Jenkins starts to run the corresponding jobs, presents the test results and generates apk for the developers and operators. The following is how each of the settings is done in sequence. The first step is to establish the link between the version control server and Jenkins. For this study, GitLab is selected as the version control server, as shown in Figure 7. Once the Jenkins job establishment address is acquired, it is entered in the web hook of the GitLab project and the triggering criteria are selected as necessary. Once the settings are completed, Jenkins will start doing its jobs every time a version is delivered.

1211

The next step is to establish the link between Jenkins and OpenStack that allows every establishment of a Jenkins job to activate a virtual machine to test the software. At the end of establishment, the environment is recycled for efficient use of the environments without any waste, as shown in Figure 8. First, the OpenStack environment to be linked is established in the Jenkins system configuration. The user identification is done via Keystone APIs and then the virtual machine is built.

As the settings are completed, every time when a new version is delivered, Jenkins starts to run the corresponding jobs, generates results and packages the new version into apk. The results are shown in Figure 9. This proves that the framework proposed in this study can be realized integrated with OpenStack without any environment issue that may lead to inconsistent establishment of versions every time. The automation achieves continuous integration and delivery and saves lots of testing time and costs, allowing developers and operators to focus more on the development work.

1212

5. Conclusion Traditional software testing is very costly. Therefore, this study presents and realizes an OpenStack-based cloud automatic testing system. The integration with OpenStack makes software testing free from the losses of deployment time and inconsistency with every test result due to different environments. For the test scripts, Monkeyrunner is used for re-package. This makes the programming much easier and allows operators focus more on development. Finally, Android handheld devices are tested to verify that this system is applicable in real life and achieves continuous integration very quickly. 6. References Yu, L., Tsai, W.T., Chen, X., Liu, L., Zhao, Y., Tang, L., & Zhao, W. (2010). Testing as a Service over Cloud. In: IEEE International Symposium on Service Oriented System Engineering, 181-188. Bang, S. K., Chung, S., Choh, Y., & Dupuis, M. (2013, October). A grounded theory analysis of modern web applications: knowledge, skills, and abilities for DevOps. In: Proceedings of the 2nd annual conference on Research in information technology, 61-62. ACM. Amalfitano, D., Fasolino, A. R., & Tramontana, P. (2011). A GUI crawling-based technique for android mobile application testing. In: Software Testing, Verification and Validation Workshops (ICSTW), 2011 IEEE Fourth International Conference on. IEEE, 252-261. Wen, X., Gu, G., Li, Q., Gao, Y., & Zhang, X. (2012, May). Comparison of open-source cloud management platforms: OpenStack and OpenNebula. In: Fuzzy Systems and Knowledge Discovery (FSKD), 2012 9th International Conference on. IEEE, 2457-2461. Yang, C. T., Liu, Y. T., Liu, J. C., Chuang, C. L., & Jiang, F. C. (2013). Implementation of a cloud IaaS with dynamic resource allocation method using OpenStack. In: Parallel and Distributed Computing, Applications and Technologies (PDCAT), 2013 International Conference on. IEEE, 71-78. Kawaguchi, K. Jenkins CI (2011). http://jenkins-ci.org/ Sefraoui, O., Aissaoui, M., & Eleuldj, M. (2012). OpenStack: toward an open-source solution for cloud computing. International Journal of Computer Applications, 55(3), 38-42. 1213

Kumar, R., Gupta, N., Charu, S., Jain, K., & Jangir, S. K. (2014). Open source solution for cloud computing platform using OpenStack. International Journal of Computer Science and Mobile Computing, 3(5), 89-98. Debois, P. (2011). Devops: A software revolution in the making. Cutter IT Journal, 24(8). Karlesky, M., Williams, G., Bereza, W., & Fletcher, M. (2007). Mocking the embedded world: Test-driven development, continuous integration, and design patterns. In: Proc. Emb. Systems Conf, CA, USA. Dlugi, M., Brunnert, A., & Krcmar, H. (2015). Model-based performance evaluations in continuous delivery pipelines. In: Proceedings of the 1st International Workshop on Quality-Aware DevOps, 25-26. ACM. Chamberlain, R., & Schommer, J. (2014). Using docker to support eproducible research. figshare. Felter, W., Ferreira, A., Rajamony, R., & Rubio, J. (2014). An updated performance comparison of virtual machines and linux container. IBM Research Report. Spreitzenbarth, M., Freiling, F., Echtler, F., Schreck, T., & Hoffmann, J. (2013). Mobile-sandbox: having a deeper look into android applications. In: Proceedings of the 28th Annual ACM Symposium on Applied Computing, 1808-1815. ACM.

1214

Poster Sessions (5) Biological Engineering / Life Sciences (1) / Physics Sciences / Chemical Sciences (2) / Chemical Engineering

Wednesday, March 30, 2016

16:30-17:30

Foyer Area

GLSBE-14 Biotransformation of Quercetin by Recombinant Escherichia Coli Expressing O-Methyltransferase from Streptomyces Peucetius Chien-Min Chiang︱Chia Nan University of Pharmacy Chun-Ping Jen︱National Chung Cheng University Te-Sheng Chang︱National University of Tainan GLSBE-25 Autotrophic Growth of Paracoccus Denitrificans in Aerobic Condition and the Accumulation of Biodegradable Plastics from CO2 Kenji Tanaka︱Kinki University Syunya Mori︱Kinki University Mai Hirata︱Kinki University Hiromi Matsusaki︱Prefectural University of Kumamoto GLSBE-68 Virus Mimetic Nanocapsules for Tumor Therapy Shang-Hsiu Hu︱National Tsing Hua University Jen-Hung Fang︱National Tsing Hua University GLSBE-92 Electrochemical Biosensor Based on MoS2/Graphene for Highly Sensitive Detection of Human Parathyroid Hormone Hyeyoun Kim︱Korean Electronics Technology Institute Kook-Nyung Lee︱Korean Electronics Technology Institute Min-Ho Lee︱Korean Electronics Technology Institute

1215

GLSBE-110 Small Angle X-Ray and Neutron Scattering Studies of Multifunctional Superparamagnetic Iron Oxide Nanocarriers Kuen-Song Lin︱Yuan Ze University Khalilalrahman Dehvari︱Yuan Ze University Tyz-Wei Chang︱Yuan Ze University GLSBE-97 The Production of Rod-Like Cellulose Nanocrystals from Moso Bamboo Kuan-Hsuan Lin︱National Taiwan University Chia-Min Liu︱National Taiwan University Feng-Cheng Chang︱National Taiwan University GLSBE-4 Production of Angiotensin I-Converting Enzyme Inhibitory Peptides from Tilapia Protein Using Continuous Enzymatic Membrane Reactor Narin Charoenphun︱Burapha University Sakaeo Campus Wirote Youravong︱Prince of Songkla University GLSBE-16 Self-Splicing of Group I Intron of the Mitochondrial Genome of the Sponge, Cinachyrella Australiensis Chi-Hsin Hsu︱National Sun-Yat Sen University Ching-Lien Wu︱National Sun-Yat Sen University GLSBE-20 Co-Resistance and Multidrug Resistance of E. Coli Isolated from Water at Broiler Chicken Farms in Northeastern Thailand Sirinya Weluwanarak Sompao︱Khon Kaen University Apiradee Sopa︱Khon Kaen University Peerapol Sukon︱Khon Kaen University Pairat Sornplang︱Khon Kaen University Bongkot Noppon︱Khon Kaen University Takuo Sawada︱Nippon Veterinary and Life Science University Tetsuo Asai︱Gifu University

1216

GLSBE-26 Enzymatic Synthesis of Pentyl-α-Glycosides with Surfactant and Antibacterial Properties Rittichai Charoensapyanan︱Thammasat University Kazuo Ito︱Osaka City University Prakarn Rudeekulthamrong︱Phramongkutklao College of Medicine Jarunee Kaulpiboon︱Thammasat University GLSBE-31 Discovery of Potent Tyrosinase Inhibitors from Chinese Herbals and Oligopeptides Yu-Ching Lee︱Taipei Medical University Tien-Sheng Tseng︱National Changhua University of Education Wang-Chuan Chen︱I-Shou University Nai-Wan Hsiao︱National Changhua University of Education Keng-Chang Tsai︱National Research Institute of Chinese Medicine GLSBE-35 Antiproliferation Effect of Curcumin and Tetrahydrocurcumin on Cholangiocarcinoma Cells Nuntiya Somparn︱Thammasat University Veerapol Kukongviriyapan︱KhonKaen University ISFAS-1442 Mathematical Model of Diphtheria with Vaccination in Thailand Puntani Pongsumpun︱King Mongkut’s Institute of Technology Ladkrabang ISFAS-1473 Effectiveness of Therapy for Pregnant Low Back Pain - A Systematic Review C. W. Kan︱The Hong Kong Polytechnic University M.Y. Li︱The Hong Kong Polytechnic University A. S. W. Wong︱The Hong Kong Polytechnic University Y. L. Kwok︱The Hong Kong Polytechnic University J. Yip︱The Hong Kong Polytechnic University S. P. Ng︱The Hong Kong Polytechnic University T. H. T. Lao︱The Chinese University of Hong Kong

1217

ISFAS-1491 Preparation and Electrical Properties of the TmMnO3/SrTiO3: Nb Heterostructures Tai-Chun Han︱National University of Kaohsiung Yi-De Chung︱National University of Kaohsiung Yih-Tsyr Chu︱National University of Kaohsiung Wen-Lan Chen︱National University of Kaohsiung Jauyn Grace Lin︱National Taiwan University ISFAS-1506 Effect of Cobalt Doping on the Magnetic and Electric Properties of GaFeO3 Nanoparticles Yih-Tsyr Chu︱National University of Kaohsiung Tai-Chun Han︱National University of Kaohsiung Yi-De Chung︱National University of Kaohsiung Wen-Lan Chen︱National University of Kaohsiung Chun Yen︱National University of Kaohsiung ISFAS-1508 3 Dimensional Measurement by Hilbert Transform in Interferometric Fringe Microscopy Silin Na︱Jeju National University Sanghoon Shin︱Korea Plant Service & Engineering Younghun Yu︱Jeju National University ISFAS-1517 Low Cost Ultraviolet-Visible Spectroscope Using LED Light Source Surachart Kamoldilok︱King Mongkut’s Institute of Technology Ladkrabang Phachara Phongwisit︱King Mongkut’s Institute of Technology Ladkrabang ISFAS-1544 Investigation on Loss of Hexagonal Arrangement in Holey Fiber Donyawan Chittinan︱King Mongkut’s Institute of Technology Ladkrabang Surachart Kamoldilok︱King Mongkut’s Institute of Technology Ladkrabang S. Lekchaum︱King Mongkut’s Institute of Technology Ladkrabang

1218

ISFAS-1606 Fabrication of Photodiode Sensors and Measurements of Photodiode Detector Coupled with a Scintillator Hwanbae Park︱Kyungpook National University H. B. Jeon︱Kyungpook National University K. H. Kang︱Kyungpook National University B. B. Kim︱Kyungpook National University T. H. Kim︱Kyungpook National University S. C. Lee︱Kyungpook National University ISFAS-1433 Synthesis of White-Light-Emitting Molecules for White Organic Light-Emitting Diodes Ying-Hsuan Wu︱Feng Chia University Kew-Yu Chen︱Feng Chia University ISFAS-1435 Synthesis and Optical Properties of Novel Asymmetric Perylene Bisimides Fang-Yun Chien︱Feng Chia University Kew-Yu Chen︱Feng Chia University ISFAS-1436 Highly Soluble Monoamino-Substituted Perylene Tetracarboxylic Dianhydrides: Synthesis, Optical and Electrochemical Properties Yung-Cheng Hsieh︱Feng Chia University Kew-Yu Chen︱Feng Chia University ISFAS-1437 A Low-Molecular-Weight Ratiometric Chemodosimeter for Highly Selective Naked-Eye and Fluorogenic Detection of Cyanide Jiun-Wei Hu︱Feng Chia University Kew-Yu Chen︱Feng Chia University ISFAS-1475 Antibacterial Finishing of Cotton with Biostatic Approach Chi-Wai Kan︱The Hong Kong Polytechnic University ISFAS-1503 Structural and Thermal Characteristics of Alkali-Treated Graphites Seung-Won Song︱Dong-Eui University Hyunmyung Choi︱Dong-Eui University Weontae Oh︱Dong-Eui University

1219

ISFAS-1505 Mechanical and Thermal Characteristics of the Thermal Interface Materials (TIMs) Prepared by Silicone Composite Materials Hyunmyung Choi︱Dong-Eui University Seung-Won Song︱Dong-Eui University Weontae Oh︱Dong-Eui University ISFAS-1551 Mechanism of NO Photocatalytic Removal on g-C3N4 Was Changed by Pd-QDs Modification Guohui Dong︱Chinese Academy of Sciences Wingkei Ho︱The Hong Kong Institute of Education ISFAS-1632 Dyeing Characteristics and Tensile Strength of High Tenacity Polyethylene (HTPE) Fibers Beom Young Lee︱Pusan National University Jeong Hoon Lee︱Pusan National University Tae Kyung Kim︱Kyungpook National University Seung Geol Lee︱Pusan National University ACEAIT-7332 Synthesis and Properties of Thermal and Transparency of Dicyclopentadiene Modified Poly(Methyl Methacrylate) Chih-Wei Tsai︱Chung Cheng Institute of Technology, National Defense University Fang-Ning Li︱Chung Cheng Institute of Technology, National Defense University Kuo-Hui Wu︱Chung Cheng Institute of Technology, National Defense University ACEAIT-7339 Preparation of Shell Crosslinked Micelles as Drug Carriers from Poly(N-isopropyl acrylamide)-co-poly(L-lysine) Graft Copolymer You-Liang Tu︱National Cheng Kung University Cheng-Chien Wang︱Southern Taiwan University of Science and Technology Chuh-Yung Chen︱National Cheng Kung University ACEAIT-7327 Experimental Study on Hydraulic Characteristics of Detour Flood-Flow in a Blocked River and Flood Damage to House Takaaki Okamoto︱Kyoto University Hiroshi Takebayashi︱Kyoto University Keiichi Toda︱Kyoto University Michio Sanjou︱Kyoto University Yuto Shibayama︱Kyoto University

1220

GLSBE-14 Biotransformation of Quercetin by Recombinant Escherichia Coli Expressing O-Methyltransferase from Streptomyces Peucetius a

Chien-Min Chianga, Chun-Ping Jenb, Te-Sheng Changc Department of Biotechnology, Chia Nan University of Pharmacy, Taiwan, R.O.C. E-mail: [email protected] b

Department of Mechanical Engineering, National Chung Cheng University, Taiwan R.O.C. E-mail: [email protected] c

Department of Biological Sciences and Technology, National University of Tainan, Taiwan, R.O.C. E-mail: [email protected] 1.

Background

O-Methylation modification is a part of biosynthesis of some isoflavones and plays a key role in the secondary metabolism in plants [1]. Enzymatic O-methylation is carried out by O-methyltransferase (OMT), which uses S-adenosylmethionine (SAM) as a methyl group donor. Our previous study demonstrated that recombinant Escherichia coli expressing O-methyltransferase from Streptomyces peucetius catalyzed both C-7 and C-8 methylation of 8-hydroxydaidzein (7,8,4'-trihydroxyisoflavone) [2]. In the present study, the biotransformation of quercetin (2,5,7,3',4'-tetrahydroxyflavone)by the recombinant E. coli was investigated. 2. Materials and Methods Recombinant E. coli BL21 (DE3) harboring expression vector pETDuet-SpOMT was obtained from our previous study [2]. Quercetin and isopropyl- -D-thiogalactopyranoside (IPTG) were purchased from Sigma (St. Louis, MO). The recombinant E. coli harboring the expression vector were cultivated in 20 0.4% glycerol, with 200 rpm shaking at 37oC. As the optical density at 600 nm reached 0.6, 0.5 mM of IPTG and 0.1 mM of quercetin were added to induce expression of the OMT gene and start biotransformation. At indicated time intervals, aliquots of cells were harvested, extracted with MeOH/ACN (50%:50%), and analyzed by ultra performance liquid chromatography (UPLC). The operational conditions of UPLC were the same as our previous work [2].

3.

Results and Discussion

For studying biotransformation, the substrate quercetin was added together with IPTG in E. coli cultivation, and the products profile were determined by UPLC to check whether the substrate could be converted by the recombinant cells. As shown in Figure 1A, one major metabolite appeared as a new peak at retention time of 5.0 min in the profile of fermentation broth at 24 h of incubation. The result proved that the recombinant E. coli also catalyzed quercetin. Our previous study showed that the recombinant E. coli catalyzed both C-8 and C-7 methylation toward 8-hydroxydaidzin. However, the present study showed that only one metabolite appeared during the biotransformation and implied that the recombinant E. coli catalyzed methylation at single carbon position of quercetin. The purification and then resolving the chemical structure of the metabolite is in progress.

1221

The production profile of the metabolite is shown in Figure 1B. In the result, the product was accumulated rapidly within the initial 6 h after induction and then followed with a very slow rate. The possible reason for the decreased enzymatic conversion rate after 6 h of induction is insufficiency of cofactor SAM. We tried to feed SAM into the cultivation at 6 and/or 18 h of induction and did not increase the production (data not shown). It has been reported that co-expression of SAM synthetase gene could increase OMT activity. Further study should be conducted to evaluate the effect of co-expression of SAM synthetase gene on the biotransformation of quercetin by the recombinant E. coli. Keywords: Quercetin, O-Methyltransferase, Methylation, Streptomyces peucetius References [1] Kim BG, Sung SH, Chong SY, Lim Y, Ahn JH. Plant flavonoid O-methyltransferase: substrate specificity and application. J Plant Biol 2010; 53: 321–329. [2] Chiang CM, Ding HY, Tsai YT, Chang TS. Production of two novel methoxy-isoflavones from biotransformation of 8-hydroxydaidzein by recombinant Escherichia coli expressing O-methyltransferase from Streptomyces peucetius Int J Mol Sci (submitted).

1222

GLSBE-14 Biotransformation of Quercetin by Recombinant Escherichia Coli Expressing O-Methyltransferase from Streptomyces Peucetius a

Chien-Min Chianga, Chun-Ping Jenb, Te-Sheng Changc Department of Biotechnology, Chia Nan University of Pharmacy, Taiwan, R.O.C. E-mail: [email protected] b

Department of Mechanical Engineering, National Chung Cheng University, Taiwan R.O.C. E-mail: [email protected] c

Department of Biological Sciences and Technology, National University of Tainan, Taiwan, R.O.C. E-mail: [email protected]

1. Background O-Methylation modification is a part of biosynthesis of some isoflavones and plays a key role in the secondary metabolism in plants [1]. Enzymatic O-methylation is carried out by O-methyltransferase (OMT), which uses S-adenosylmethionine (SAM) as a methyl group donor. Our previous study demonstrated that recombinant Escherichia coli expressing O-methyltransferase from Streptomyces peucetius catalyzed both C-7 and C-8 methylation of 8-hydroxydaidzein (7,8,4'-trihydroxyisoflavone) [2]. In the present study, the biotransformation of quercetin (2,5,7,3',4'-tetrahydroxyflavone)by the recombinant E. coli was investigated. 2. Materials and Methods Recombinant E. coli BL21 (DE3) harboring expression vector pETDuet-SpOMT was obtained from our previous study [2]. Quercetin and isopropyl- -D-thiogalactopyranoside (IPTG) were purchased from Sigma (St. Louis, MO). The recombinant E. coli harboring the expression vector were cultivated in 20 mL of LeMaster and Richards minimal medium (LR medium) o C. As the optical density at 600 nm reached 0.6, 0.5 mM of IPTG and 0.1 mM of quercetin were added to induce expression of the OMT gene and start biotransformation. At indicated time intervals, aliquots of cells were harvested, extracted with MeOH/ACN (50%:50%), and analyzed by ultra performance liquid chromatography (UPLC). The operational conditions of UPLC were the same as our previous work [2]. 3. Results and Discussion For studying biotransformation, the substrate quercetin was added together with IPTG in E. coli cultivation, and the products profile were determined by UPLC to check whether the substrate could be converted by the recombinant cells. As shown in Figure 1A, one major metabolite appeared as a new peak at retention time of 5.0 min in the profile of fermentation broth at 24 h of incubation. The result proved that the recombinant E. coli also catalyzed quercetin. Our previous study showed that the recombinant E. coli catalyzed both C-8 and C-7 methylation toward 8-hydroxydaidzin. However, the present study showed that only one metabolite appeared during the biotransformation and implied that the recombinant E. coli catalyzed methylation at single carbon position of quercetin. The purification and then resolving the chemical structure of the metabolite is in progress.

1223

The production profile of the metabolite is shown in Figure 1B. In the result, the product was accumulated rapidly within the initial 6 h after induction and then followed with a very slow rate. The possible reason for the decreased enzymatic conversion rate after 6 h of induction is insufficiency of cofactor SAM. We tried to feed SAM into the cultivation at 6 and/or 18 h of induction and did not increase the production (data not shown). It has been reported that co-expression of SAM synthetase gene could increase OMT activity. Further study should be conducted to evaluate the effect of co-expression of SAM synthetase gene on the biotransformation of quercetin by the recombinant E. coli. Keywords: Quercetin, O-Methyltransferase, Methylation, Streptomyces peucetius References [3] Kim BG, Sung SH, Chong SY, Lim Y, Ahn JH. Plant flavonoid O-methyltransferase: substrate specificity and application. J Plant Biol 2010; 53: 321–329. [4] Chiang CM, Ding HY, Tsai YT, Chang TS. Production of two novel methoxy-isoflavones from biotransformation of 8-hydroxydaidzein by recombinant Escherichia coli expressing O-methyltransferase from Streptomyces peucetius Int J Mol Sci (submitted).

1224

GLSBE-25 Autotrophic Growth of Paracoccus Denitrificans in Aerobic Condition and the Accumulation of Biodegradable Plastics from CO2 *Kenji Tanakaa, Syunya Moria, Mai Hirataa, Hiromi Matsusakib a Department of Biological and Environmental Chemistry, Faculty of Humanity-Oriented Science and Engineering, Kindai University, Japan b

Department of Food and Health Sciences, Faculty of Environmental and Symbiotic Sciences, Prefectural University of Kumamoto, Japan *Corresponding Author: [email protected]

Abstract The cell growth on H2 and O2 as the energy source and CO2 as the sole carbon source in the autotrophic culture condition was tested for the gram-negative bacteria Paracoccus spp. The aerobic growth in the autotrophic condition was only observed in Paracoccus denitrificans NBRC13301 and P.pantotrophus NBRC102493. Both strains were sensitive to O2 in particular the growth of P.pantotrophus was completely inhibited at the concentrations above 2% O2. P.denitrificans grew until 15% O2 however the optimum O2 concentration was 5%. The growth characteristic of P.denitrificans was investigated by pH-controlled batch culture with supplying the gas mixture of H2, O2 and CO2. The specific growth rate was 0.12 h-1 at 5% O2, 30°C and pH7.0. The growth was much slower than other hydrogen-oxidizing bacteria however P.denitrificans accumulated biodegradable plastic, polyhydroxybutyrate, PHB with the content of 57.3% w/w in the cell under nitrogen limitation. Under DO limitation, the cell concentration increased to 25 g/L without accumulating PHB if NH3 solution as the nitrogen source was fed sufficiently. Keyword: Polyhydroxyalkanoates, Paracoccus, Hydrogen-oxidizing bacterium 1. Background Paracoccus spp. is the gram-negative bacteria possessing the ability for denitrification (reduction of NO3- to N2). They play an important role in the nature. Especially, P.denitrificans is employed in the biotreatment of waste water. Paracoccus spp. accumulates poly- -hydroxybutyrate (PHB) in their cells and the polyester is expected as biodegradable plastics [1]. Some strains of Paracoccus spp. grow with CO2 as carbon source, and H2 or thiosulfate as energy source in autotrophic condition. Many researchers have been reported about Paracoccus spp., however in most of the researches for the autotrophic culture using the inorganic gasses as the substrate was studied only for denitrification in anaerobic condition, and the PHB accumulation was studied only in heterotrophic culture using organic compounds as the substrate. We studied the growth characteristics of Paracoccus spp. in the autotrophic condition using the gas mixture of H2, O2 and CO2 as the substrate, and high cell density culture and the accumulation of PHB from CO2. 2. Methods The bacteria used in this study were Paracoccus denitrificans NBRC13301, P.alcaliphilus JCM7364T, P.pantotrophus NBRC102493, P.thiocyanatus NBRC14569, P.aminophilus NBRC16710, P.aminovorans NBRC 16711 and P.versutas NBRC14567. All the strains were purchased from NBRC (NITE Biological Resource Center, National Institute of Technology and Evaluation, Tokyo) or JCM (Japan Collection of Microorganisms, RIKEN, Tsukuba, 1225

Japan). These bacteria were subcultured every three weeks on BY medium or PYG medium agar plates. The composition of BY media was meat extract 5.0g, peptone 5.0g, NaCl 2.5g, K2HPO4 0.1g, MgSO4・7H2O 0.2g, yeast extract 5.0g and agar 15g per 1 L of distilled water (pH7.2) ; and that of PYG medium was peptone 5.0g, yeast extract 5.0g and glucose 5.0g per 1 L (pH7.0). Basic composition of mineral medium for autotrophic culture was (NH4)2SO4 3.0g, KH2PO4 4.0g, NaHPO4 0.8g, NaHCO3 1.0g, MgSO4・7H2O 0.2g and 1 L distilled water. The pH was adjusted to 7.0 with 1 M NaOH then it was autoclaved 20min at 120°C (Mg salt was autoclaved separate from other salts). After cooling, 0.1 ml of filter-sterilized trace elements solution was added to the medium. Composition of the trace elements solution was CoCl2 119mg, FeSO4・7H2O 16.2g, NiCl2・6H2O 118mg, CrCl2・6H2O 133mg, CuSO4・5H2O 156mg, CaCl2・2H2O 10.3g and citric acid 15.6g in 100 ml of 1 M HCl, which was originally used for a hydrogen-oxidizing bacterium, Ralstonia eutropha [2]. Autotrophic culture experiment was carried out carefully using the explosive gas mixture of H2, O2 and CO2 as described below [3]. The bacteria were refreshed on the BY or PYG agar plate, respectively then they were inoculated into 20ml of the mineral liquid medium in 300-ml Erlenmeyer flask. The flask was plugged with a rubber stopper which was inserted with a glass tube connected to a silicone tube and a sterile membrane filter (pore size, 0.2 m). The inside of the flask was vacuumed then it was filled with H2, O2 and CO2 from each gas cylinder. Basic composition of the substrate gas mixture was H2 85%, O2 10%, and CO2 10%. Cultivation was carried at a temperature of 30°C and a reciprocal shaking of 200 rpm. High cell density culture experiment was carried out using a glass jar fermenter (total volume 1000 ml; working volume 600 ml) equipped with a pH controller, PHC-2201 and a dissolved oxygen (DO) meter, DJ-1033 (Biott Co., Ltd., Tokyo). Batch culture was performed with a recycled-gas closed-circuit culture system. The substrate gas mixture in the gas chamber was supplied to the fermenter at a flow rate of 0.5vvm and the exhausted from the fermenter was returned to the chamber. The agitation speed was kept at 1200 rpm. As the cell concentration increases, the foaming of culture liquid becomes vigorous and it causes the flow out of culture liquid from the fermenter. Then, a defoaming agent (Einol, Biott Co. Ltd., Tokyo) was used in high cell density culture. Cell growth was monitored by measuring turbidity (OD600nm) of the culture liquid. The composition of the substrate gas mixture in the flask and the fermentation system was measured by a gas chromatograph (Shimadzu type GC-8A) using a column 4mm x 6m into which a molecular sieve 5A and a Porapack Q were packed. Total ammonia (NH4-N) in culture liquid was determined by indophenol blue reaction. PHB accumulated in the cells was determined according to the method using gas chromatography [4]. The lyophilized cells, a mixture of 2 ml of methanol acidified with 3%(w/v) H2SO4 and 2 ml of chloroform were added into a screw cap vial then it was heated at 100°C for 3.5 h for degradation of PHB, and esterification of hydroxybutyric acid and methanol. After cooling, 1 ml of H2O was added into the vial then the suspension was shaken well for 10min. After two phases were allowed to separate, the organic phase containing the methyl ester was applied to gas chromatography. 3. Results 3.1 Growth of Paracoccus strains Growth of Paracoccus spp. in the autotrophic condition was tested by flask culture. The head space inside the culture flask was filed with the gas mixture with a ratio H2/O2/CO2 = 8:1:1 before the start of cultivation. After 7 days cultivation, the increase in the turbidity of culture liquid and the consumption of substrate gas (a decrease in pressure of head space gas inside the 1226

flask) were observed only in P.denitrificans NBRC13301. There was no growth in the other Paracoccus spp. It is known that many strains of hydrogen-oxidizing bacteria are sensitive to O2 although they are aerobic bacteria. Then, the culture test was carried out again with lowering O2 concentration in the substrate gas mixture (Table). The cell growth was observed in P.denitrificans NBRC13301 and P.pantotrophus NBRC102493. P.denitrificans grew from 5 to 15% O2 and P.pantotrophus did only at 2% O2.

These two strains, in particular P.pantotrophus was much more sensitive to O2 than other hydrogen-oxidizing bacterium like Ralstonia eutropha and Alcaligenes latus, Ideonella dechloratans [5] and A.hydrogenophilus [6]. 3.2 Growth condition for P.dentrificans P.denitrificans NBRC13301 showed the best growth among the Paracoccus spp. in the autotrophic condition. Hence, the growth characteristics of P.denitrificans were investigated in detail. As a result, optimum temperature and pH were 30°C and pH 7.0, respectively. Among the nitrogen sources of nitrate, ammonium salts and urea, the best growth was obtained with (NH4)2SO4. There was no growth with urea (the data are not shown). Figure 1 shows a fermentation time course of P.denitrificans NBRC13301 in flask culture.

1227

The head space substrate gas in the flask was refiled several times during the cultivation however the cell concentration was not over 8.0 in OD600 with any medium composition. The OD600= 8.0 was almost equivalent to about 3.2 g/L of dry cell weight. It was considered that the cell growth was inhibited by the decrease in pH of culture liquid. 3.3 High cell density culture of P.dentrificans and PHB accumulation from CO2 High cell density culture of P.denitrificans NBRC13301 was examined using a pH-controller and a glass jar fermenter. The recycled-gas closed-circuit culture system was used because the high performance of gas utilization. The pH of culture liquid in the fermenter was maintained around 7.0±0.2 by feeding 4% w/v NH3 solution or 1M NaOH automatically. Cell concentration was determined by measuring OD600 of the culture liquid and dry weight of the centrifuged cells. The substrate gas mixture was supplied into the culture medium from a gas chamber (total volume, ca.15 L) using an airtight air pump. The exhausted gas from the fermenter was returned to the gas chamber for reuse. The inner pressure of the gas chamber was maintained at almost constant level because saturated saline water was voluntarily introduced according to the decrease in the volume of substrate gas mixture [3]. Figure 2 (a) shows the fermentation time course of the strain NBRC13301 at pH7.0 with feeding 4% w/v NH3 solution. Cultivation was started by recycling the substrate gas with the volume ratio of H2/O2/CO2= 85:5:10 within the culture system. As the fermentation proceeds, the gas composition in the culture system changed, especially O2 concentration sharply decreased, the gas mixture in the chamber was exchanged several times to reset to the initial composition. The cells multiplied at the specific growth rate of approximately 0.12 h-1 during the exponential growth phase. After 42 h from the start of cultivation, DO decreased to almost 0ppm (the data is not shown), the exponential growth ceased then the cell concentration increased in proportion to time. When the cells increased to the dry weight concentration 8.12 g/L after 120 h, we stopped the cultivation because it was too difficult to maintain O2 concentration of gas phase in the culture system at 5%. The cell concentration increased to about 2.5 times as that of the flask culture. Accumulation of PHB in the cells was slight throughout the cultivation. The culture liquid at the end of cultivation contained sufficient amount of NH4+ and PO43- for vigorous multiplication of the cell. Figure 2(b) shows the fermentation time course with feeding 1M NaOH instead of NH3 solution for pH control. When NH4+ was exhausted after 45 h, the cells began to accumulate PHB. The PHB content of the cells increased to 57.3% w/w at the end of cultivation. It is known that PHB generally accumulates in nitrogen- or DO-limited culture condition. Our results indicate that in case of P.denitrificans, PHB accumulation is not stimulated under DO limitation but nitrogen limitation.

1228

Figure 3 shows the fermentation time course with exchanging the substrate gas mixture in the gas chamber every 12 h during cultivation. The pH was controlled by feeding 4% w/v NH3 solution. The gas mixture with the volume ratio of H2/O2/CO2 = 85:5:10 was used from the start of cultivation until 50h, then the ratio was changed to H2/O2/CO2 = 80:10:10. After 80h, the volume ratio of the newly prepared gas mixture was changed to H2/O2/CO2 = 75:15:10 and the ratio was used until the end. As a result, DO was almost kept at 0ppm after 40 h (the data is not shown) however the cells continued to increase for the long DO limitation. When the cell concentration increased to about 25 g/L after 140 h, we stopped the cultivation because foaming of the culture liquid was so vigorous that we could not prevent the culture liquid from flowing out of the fermenter. The PHB accumulation was very small because the nitrogen source in the culture liquid was sufficient throughout the cultivation by feeding NH3 solution.

3.4 Discussion Our research showed that among the 7 strains of Paracoccus spp. tested, P.denitrificans NBRC13301 and P.pantotrophus NBRC102493 grow in the autotrophic culture condition using the gas mixture of H2, O2 and CO2 as the substrate. However, both strains were very sensitive to O2 especially P.pantotrophus did not grow with the gas mixture containing O2 above 2%. On the other hand, P.denitrificans grew at O2 concentrations up to 15% and the 1229

optimum O2 concentration for the cell growth was 5%. The specific growth rate of P.denitrificans in the autotrophic condition was 0.12 h-1 at 5% O2. The growth of P.denitrificans was much slower than other hydrogen-oxidizing bacteria, R.eutropha, A.latus, I.dechloratans [5] and A.hydrogenophilus [6]. The specific growth rate of R.eutropha at 5% O2 was 0.42 h-1 [3]. It is reported that in case of R.eutropha (A.eutrophus), the soluble NAD+-reducing hydrogenase was inactivated by O2 when H2 is available as electron donor [7]. PHB and the copolyester of D-3-hydroxybutric acid and other hydroxyalkanoic acid, PHA are expected as the biodegradable ―Green plastics‖. However, most of the researches for microbial production of PHAs are investigated on organic acids or fatty acid as the substrate and there are only several researches reported for the autotrophic production from CO2 [8]. We have been studied the production of PHB from CO2 employing several strains of hydrogen-oxidizing bacteria [5, 9-12]. In our study, 60g/L of the cells and 36g/L were obtained in autotrophic culture of R.eutropha with the conventional type jar fermenter in 60 h [9]. Furthermore, the use of a basket-type agitator with very high performance for oxygen transfer (kLa, 2970 h-1) enabled to produce 91.3 g/L of the cells with 61.9 g/L of PHB in 40 h in spite of supplying the substrate gas mixture with the lower O2 concentration below the ―lower explosion limit‖ was supplied [10]. PHB can be produced from CO2 even in the presence of high concentration of carbon monoxide (CO) by employing the CO tolerant hydrogen-oxidizing bacterium Ideonella sp.O-1 that we isolated [5]. Volova et al. also reported the production of a copolymer containing  -hydroxybutyrate more than 99 mol% from CO2 in model gas mixture containing CO by employing R.eutropha B5786 [13]. In this study, it was shown that the cell growth of P.denitrificans in autotrophic condition is more sensitive to O2 and slower than other hydrogen-oxidizing bacteria. However, PHB accumulated in the cells with the content of 57.3% w/w under nitrogen limitation while it did not under DO limitation. In the practical fermentation process for the production of PHB employing hydrogen-oxidizing bacteria in commercial scale, O2 concentration of the substrate gas mixture must be maintained lower than 6.9% to pretend the explosion of H 2 [10]. The condition does not exclude P.denitrificans from the candidate strains for PHB producer. It will be possible to achieve high PHB yield from CO2 employing P.denitrificans if the fermenter like the basket-type agitator to give high performance for oxygen transfer. On the other hand, some strains of this species are already used for the commercial production other useful substances for instance vitamin B12 (cyanocobalamin) from methanol. The property of P.denitrificans to continue the cell growth without PHB accumulation under DO limitation will be beneficial for production of commercially useful substances except PHAs from CO2. 4. References [29] Saharan, B.S., Grewal, A., Kumar, P. 2014. Biotechnological production of polyhydroxyalkanoates: A review on trends and latest developments. Chinese Journal of Biology, vol.2014, ArticleID:802984, 18pages, doi: 10.1155/2014/802984 [30] Siegel, R.S. and Ollis, D. F. 1984. Kinetics of growth of the hydrogen ‐oxidizing bacterium Alcaligenes eutrophus (ATCC17707) in chemostat culture. Biotechnology and Bioengineering, 26(7), 764‐770. 1984 [31] Ishizaki, A. and Tanaka, K. 1990. Batch culture of Alcaligenes eutrophus ATCC 17697T using recycled gas closed circuit culture system. J Ferment Bioeng, 69(3), 170-174. [32] Sonnleitner, B., Heinzle, E., Braunegg, G., Lafferty, R.M. 1979. Formal kinetics of poly- -hydroxybutyric acid (PHB) production in Alcaligenes eutrophus H16 and Mycoplana rubra R14 with respect to the dissolved oxygen tension in ammonium-limited batch cultures, Eur J Appl Microbiol Biotechnol, 7(1), 1-10. 1230

[33] Tanaka, K., Miyawaki, K., Yamaguchi, A., Kianoush, K.D., Matsusaki, H. 2011. Cell growth and P(3HB) accumulation from CO2 of a carbomonoxide-tolerant hydrogen-oxidizing bacterium, Ideonella sp. O-1. Appl Microbiol Biotechnol, 92(6), 1161-1169. [34] Ohi, K., Matsumoto, H., Yamada, K. 1979. Cultivation and characterization of a hydrogen bacterium, Alcaligenes hydrogenophilus sp. nov., J Ferment Technol, 57(3), 195-202. [35] Schlesier, M. and Friedrich, B. 1981. In vivo inactivation of soluble hydrogenase of Alcaligenes eutrophus, Arch Microbiol, 129(2), 150-153. [36] Kianoush, K. D., K., Mokhtari, Z. B., Amai, T., Tanaka, T. 2013. Microbial production of poly(hydroxybutyrate) from C1 carbon sources. Appl Microbiol Biotechnol, 97(4), 1407-1424. [37] Ishizaki, A. and Tanaka, K. 1991. Production of poly- -hydroxybutyric acid from carbon dioxide by Alcaligenes eutrophus ATCC 17697T. J Ferment Bioeng, 71(4), 254-257. [38] Tanaka, K., Ishizaki, A., Kanamaru, T., Kawano, T. 1995. Production of poly (D-3-hydroxybutyrate) from CO2, H2, and CO2 by high cell density autotrophic cultivation of Alcaligenes eutrophus. Biotechnol Bioeng, 45(3), 268-275. [39] Taga, N., Tanaka, K. Ishizaki, A. 1997. Effects of rheological change by addition of carboxymethylcellulose in culture media of an air-lift fermentor on poly-D-3-hydroxybutyric acid productivity in autotrophic culture of hydrogen -oxidizing bacterium, Alcaligenes eutrophus, Biotechnol Bioeng, 53(5), 529 - 533. [40] Sugimoto, T., Tsuge, T., Tanaka, T., Ishizaki, A. 1999. Control of acetic acid concentration by pH-stat continuous substrate feeding in heterotrophic culture phase of two-stage cultivation of Alcaligenes eutrophus for production of P(3HB) from CO2, H2 and O2 under non-explosive condition. Biotechnol Bioeng, 62(6), 625-631. [41] Volova, T.G., Kalacheva, G.S., Altukhova, O.V. 2002. Autotrophic synthesis of polyhydroxyalkanoates by the bacteria Ralstonia eutropha in the presence of carbon monoxide, Appl Microbiol Biotechnol, 58(5), 675-8.

1231

GLSBE-68 Virus Mimetic Nanocapsules for Tumor Therapy Shang-Hsiu Hu*, Jen-Hung Fang Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Taiwan Email: [email protected] 1.

Background/ Objectives and Goals

Advanced stimuli-responsive drug delivery nanosystems have recently received tremendous attention because they can improve therapeutic efficacy in cancer therapy.[1,2] Here, we propose a virus-mimetic nanocapsule (VN) for multiple-cargo delivery, on-demand release, and sqeuential infecting cancer cells, based on an unique type of double emulsions prepared by the self-assembly of iron oxide nanoparticles (IO) and a one-component functional protein. 2. Methods In this study, the shell of VNs is constructed by protein and stabilized by the magnetic nanopaticles (in particular iron oxide (IO)) which also serve as as the actuator that achieves drug triggered release and cell liberation. To form the structures, lactoferrin (a biocompatible protein) was utilized[23,24] as it possesses suitable hydrophilic/hydrophobic properties to achieve the above functionality that is important for the synthetic approach of the W/O/W structures. 3. Expected Results/ Conclusion/ Contribution Being liberated from dead cells by HFMF manipulation, the constructive VNs can sequential infect neighboring cancer cells and deliver sufficient therapeutic agents to next targeted sites. With high efficiency for sequential cell infections, VNs have successfully eliminated subcutaneous tumor after a combinatorial treatment. These results demonstrate that the VNs could be used for locally targeted, on-demand, magnetoresponsive chemotherapy/hyperthermia, combined with repeatedly cell infections for tumor therapy and other therapeutic applications.

Keywords: Drug delivery, nanoparticles, synergistic therapy, tumor therapy

1232

GLSBE-92 Electrochemical Biosensor Based on MoS2/Graphene for Highly Sensitive Detection of Human Parathyroid Hormone Hyeyoun Kim, Kook-Nyung Lee, Min-Ho Lee* *Medical IT research center, Korean Electronics Technology Institute, Korea Email: [email protected] Summary A self-assembled novel nanocomposite composed of MoS2-Graphene(MoS2-Gr) composites were electrostatically formed on top of gold (Au) electrode and their biochemical amplification responses were reported. The properties of the MoS2-Gr were characterized by scanning electron microscopy (SEM), high-resolution transmission electron microscopy (HR-TEM), X-ray diffraction (XRD), and Fourier transform infrared spectrometer (FT-IR). Cyclic voltammetry was performed in the presence of Parathyroid Hormone (PTH) and its linearity was obtained by measuring the anodic current signals coming from interactions between MoS2-Gr-Ab and different concentrations of PTH with the aid of substrate and enzyme conjugated secondary antibodies. Motivation Secretion of PTH is controlled by the amount of Ca2+ in serum in a negative feedback ways. Therefore, it acts as an indirect indicator for the diagnosis of osteoporosis. Due to their extremely low presence in serum, highly sensitive materials were needed for amplification in terms of electrochemistry. Many interesting and unique properties of graphene (Gr) can be researched after it is integrated into complex assemblies. Some Gr based hybrid material showed multifaced application for electrochemical sensors and biosensors. Moreover, hybrid material is strong synergistic interaction between Gr and its composite component which can be enhanced catalytic activity and stability. Moreover MoS2 has tunable band gap and utilized for the dydrodesulfurization reaction and for the hydrogen evolution reaction as a catalyst 2. For this reason, the integration of Gr and MoS2 has similar effect on the electrocatalysis and electrochemical responses. In this paper, MoS2-Gr novel composite was synthesized by L-cysteine-assisted solution phase method. The as-prepared MoS2-Gr modified on gold electrode was used to immobilize capture antibodies against PTH. Results Schematic of fabrication process of Au electrode on glass wafer is shown in Figure 1. In addition, this figure shows the schematic of Au-MoS2 on Au electrode using E-beam evaporation. Figure 2 shows characterization results of MoS2-Gr; graphene oxide solution, GO+NH2CSNH2 and resultants after centrifugation, respectively. CV measurement of electrode modification For the proof-of-principle studies, we have performed calibration studies. As can be seen in figure 3, each response was well correlated with the concentration of PTH and shows good linearity. For the clinical approaches, we have collected serum from the patients who have been with osteoporosis. All experiments using Au-electrode were performed under an approved protocol 2

Y, Li et al., “MoS2 nanoparticles grown on graphene: an advanced catalyst for the hydrogen evolution reaction,” Journal of the American Chemistry Society, Vol.133, pp.7296-7299, 2011

1233

of the Institutional Review Committee at Seoul National University Bundang Hospital (SNUBH). In addition, all patients gave their written informed consent. Each serum had been collected in the morning before breakfast. Figures

1234

GLSBE-110 Small Angle X-Ray and Neutron Scattering Studies of Multifunctional Superparamagnetic Iron Oxide Nanocarriers Kuen-Song Lin*, Khalilalrahman Dehvari, Tyz-Wei Chang Department of Chemical Engineering & Materials Science/Environmental Technology Research Centre, Yuan Ze University, Taiwan, R.O.C. E-mail: [email protected] 1. Background Superparamagnetic iron oxide nanoparticles (SION) with theranostics features have been greatly acknowledged and improved the scope of nanomedicine in cancer treatment [1-3]. Present study was aimed for developing and exploring the use of multifunctional drug delivery system composed of SION and thermo-responsive Pluronic for targeted drug delivery system and treatment of colon cancer. The SION have no significant cytotoxicity influence on the studied cell line. Therefore, they can act as safe drug delivery systems in order to facilitate drug entrance into the cells [4]. The objective of the present work was to investigate the micellar solubilization of PTX into P123 micelles as two mechanisms of Pluronic-PTX conjugates. 2. Method Pluronic solutions were prepared by simply dissolving the polymer in D 2O. The PTX-loaded P123 nanocarriers were prepared through an emulsion/solvent evaporation technique. The SANS measurements were carried out on the small and wide angle neutron scattering instrument, TAIKAN, installed on BL15 in the Materials and Life Science Experimental Facility (MLF) of J-PARC and the NGB30 SANS instrument at the National Institute of Standards and Technology (NIST) Centre for Neutron Research, Gaithersburg, MD. 3. Results SAXS results (Fig. 1) at the different concentrations of the Pluronic are consistent with a microstructure of spherical micelles composed of a PPO core and PEO shell regions. The average core radius was increased in response to the higher concentration of Pluronic in aqueous solution. Furthermore, the effect of hydrophobic drug and SION on core size was investigated. SANS was used to investigate the effects of varying drug and Pluronic concentrations on the Pluronic P123 micelles‘ structure. Subsequently, the changes in micellar structures were visualized using the Cryo-TEM and SANS techniques.

1235

Results showed that high PTX concentration leads to higher scattering intensity in the low-Q region, pointing that complex clusters of PTX and Pluronic were emerged in the samples (Fig. 2). Furthermore, SANS measurements showed that using higher Pluronic concentration for constant PTX concentration correlated with the larger core dimensions and subtle changes in shell thickness. The in vivo results showed that the magnetic system enables targeted drug delivery and simultaneously, gives higher MRI resolution (Fig. 3). In conclusion, treatment of rectal tumors with PTX loaded multifunctional SION showed antitumor effect on small tumors. Such systems enhance cancer therapy giving more hands on efficient delivery, controlled release, and higher imaging resolutions.

Keywords: Pluronic, SANS, paclitaxel, nanocarrier, drug delivery system. References 1. Mattingly, S. J.; O‘Toole, M. G.; James, K. T.; Clark, G. J.; Nantz, M. H., Langmuir 2015, 31 (11), 3326-3332. 2. Nair, M.; Guduru, R.; Liang, P.; Hong, J.; Sagar, V.; Khizroev, S., Nature Communication 2013, 4, 1707. 3. Singh, R. K.; Patel, K. D.; Kim, J. J.; Kim, T. H.; Kim, J. H.; Shin, U. S.; Lee, E. J.; Knowles, J. C.; Kim, H. W., ACS Applied Materials & Interfaces 2014, 6 (4), 2201-2208. 4. Dehvari, K.; Lin, K. S., Current Medicinal Chemistry, 2012, 19 (30), 5199-5204.

1236

GLSBE-97 The Production of Rod-Like Cellulose Nanocrystals from Moso Bamboo Kuan-Hsuan Lin, Chia-Min Liu, Feng-Cheng Chang School of Forestry and Resource Conservation, National Taiwan University, Taiwan E-mail: [email protected] 1. Background Cellulose is a sustainable bio-resource and of numerous affluence in nature. Due to its promising physical and mechanical characteristics, cellulose fiber has currently gained highly global attention in the material science field, particularly in exploring and applying cellulose nanocrystals (CNCs) for industrial use. The remarkable mechanical properties of CNCs are commonly agreed so that they have potentials to be applied as functional reinforcements to develop biocomposites. Many studies have reflected works in isolation of crystalline regions from a lignocellulosic chain; however, limited numbers of them selected bamboo as the raw material. The rich bamboo resources in Taiwan provide abundant raw materials for producing bamboo-based CNCs, which can not only significantly increase their added value but also benefit the environment by substituting other high energy-consuming and carbon emission man-made materials. Hence, in this study, the processing conditions for producing bamboo-based CNCs (B-CNCs) were investigated. 2. Methods Sulfuric acid hydrolysis was chosen to prepare B-CNCs from air-dried moso bamboo chips (Phyllostachys edulis; collected from Nantou, Taiwan), and the processing procedures were described as the following paragraphs. With a 2:1 (v/v) mixture of benzene/ethanol, dewaxed bamboo chips were obtained in a Soxhlet apparatus. Then, they were treated by acidified sodium chlorite solution at 75°C to become holocellulose fibers. Then, the bamboo holocellulose were treated by 3% (w/w) Sodium hydroxide (NaOH, Showa Chemical Co. Ltd.) at 50°C for 2 h, removing the fatty acids, the residual lignin, hemicellulose, resin, and other impurities. At the same time, the amorphous regions also swelled so that it would improve the effectiveness of acid hydrolysis later. B-CNCs were finally obtained after sulfuric acid hydrolysis. This treatment was done by using both 64% and 46% (w/w) sulfuric acid (H2SO4, Scharlau) under continuously stirring at 55°C for 2 h, and then the remains were freeze dried (freeze dryer, Kingmech Co. Ltd.) after dialyzing and ultra-sonicating. Afterward, several properties, such as the morphology, dimension, surface morphology, crystallinity and thermal stability, were measured to evaluate the relationship between the conditions of the manufacturing process and resulting properties. The instruments below are applied to measure the properties of B-CNCs. Transmission electron microscopy (TEM) is a common way to measure the outward appearance, morphology, and dimensions. Infrared spectroscopy (FTIR) shows the functional groups on B-CNCs through both diffuse reflectance infrared fourier transform (DRIFT) and FTIR Transmittance measurement. As for the structure of crystalline regions, normally, x-ray diffraction (XRD) is used to determine and calculate the crystallinity.

1237

3. Expected results At present, B-CNCs has been obtained as powder-like products after freeze dried. The first observation found that B-CNCs obtained through 46% sulfuric acid hydrolysis exhibit white color; on the other hand, those obtained through high concentration of acid show darker color than the previous one. In addition, further experiments is in progress to determine and measure various properties of B-CNCs products. Those products are expected to appear as rod-like fibers with approximate 20 nm in width and 200 in length, under TEM observation. In addition, the FTIR spectrum should be comparable to bamboo holocellulose. Besides, the crystalline index (CrI) will be calculated with the data provided by XRD, and is expected to be as high as approximately 70%, referring to other similar studies. After studying the preliminary tests of B-CNCs, more information is allowed to be considered in developing nanocellulose-based biocomposite materials. Further, the results will lead value-added applications for bamboo. Keywords: nanocrystalline cellulose, bamboo, biomaterial, acid hydrolysis

1238

GLSBE-4 Production of Angiotensin I-Converting Enzyme Inhibitory Peptides from Tilapia Protein Using Continuous Enzymatic Membrane Reactor a

Narin Charoenphuna, Wirote Youravongb,c Department of Natural Product Technology, Faculty of Agricultural Technology, Burapha University Sakaeo Campus, Thailand E-mail: [email protected]

b

Department of Food Technology, Faculty of Agro-Industry, Prince of Songkla University, Thailand c

Membrane Science and Technology Research Center, Prince of Songkla University, Thailand* E-mail: [email protected] 1.

Background

The use of continuous enzymatic membrane reactor (CEMR) to integrate a reaction vessel with a membrane separation unit is emerging as a beneficial method for producing Angiotensin-I converting enzyme (ACE) inhibitory peptides. In this study, the CEMB was employed to produce ACE inhibitory peptides from tilapia protein. 2.

Methods

A hollow fiber membrane (MWCO 1 kDa) was equipped with stirred reactor tank. The enzymatic hydrolysis and separation of peptides conditions were performed at constant temperature (50°C) and pH (8). The investigations of the effect of substrate feeding pattern, enzyme concentration, and gas-liquid two-phase flow on performance of CEMR were studied. 3.

Results

It was found that the substrate feeding pattern A which was the most suitable for producing of bioactive peptides because it gave constant flux and high ACE inhibitory activity. The enzyme concentration at 15 unit g-1protein gave high ACE inhibitory activity conversion and productivity. Addition of gas sparging into the membrane module led to increase permeate flux compared with that without gas sparging. In addition, gas injection factor up to 0.5 gave the best improvement of permeate flux. Operating using substrate feeding pattern A, at ε = 0.5, TMP = 1.3 bar, CFV = 1.5 m s -1, and enzyme concentration = 15 unit g-1protein at 720 min of operation, ACE inhibitory activity conversion and productivity were 404 % and 3.60 mg ACE inhibitory peptides unit-1 of enzyme. This result indicates that CEMR was successfully employed to produce ACE inhibitory peptides. Keywords: ACE inhibitory, reactor, ultrafiltration, tilapia protein, fouling 4.

References

Charoenphun, N., Cheirsilp, B., Sirinupong, N. & Youravong, W. 2013. Calcium-binding peptides derived from tilapia (Oreochromis niloticus) protein hydrolysate. European Food Research and Technology. 236 : 57–63. Charoenphun, N., Youravong, W. & Cheirsilp, B. 2013. Determination of reaction kinetics of hydrolysis of tilapia (Oreochromis niloticus) muscle protein for manipulating production of bioactive peptides with antioxidant activity, angiotensin-I-converting enzyme (ACE) inhibitory activity and Ca-binding properties. International Journal of Food Science and Technology. 48 : 419-428. 1239

GLSBE-16 Self-Splicing of Group I Intron of the Mitochondrial Genome of the Sponge, Cinachyrella Australiensis Chi-Hsin Hsu, Ching-Lien Wu Department of Marine Biotechnology and Resources, National Sun-yat Sen University, Taiwan E-mail: [email protected] 1. Background

The mitochondrial cytochrome oxidase I gene (cox 1) of the sponge, Cinachyrella australiensis, spans 2709 bp in length, which is much longer than cox 1 of other sponges due to a 1140 bp intron. The intron is classified as a group I self-splicing intron and is capable of encoding a putative homing endonuclease with two amino acid motifs of the LAGLIDADG family. The homing endonuclease may perform intron translocation. However, the self-splicing activity of the intron has not yet verified. 2. Methods

A 1587 bp DNA fragment containing the intron with franking partial cox 1 sequences was amplified by PCR. The fragment was purified and cloned into pGEM-T Easy Vector for further splicing studies in E. coli and in vitro transcription system. Intron splicing was judged by appearance of a 447 bp fragment after RT-PCR using the same primer set for the 1587 bp fragment.. 3. Results and Conclusions

To show that intron splicing is carried out correctly, C. australiensis total DNA and RNA were purified, respectively, for DNA PCR and RT-PCR. The results showed only a 1587 bp band was found on DNA PCR and only a 447 bp band was found for RT-PCR. The 447 bp DNA was purified for DNA sequencing. The results indicated that the intron was spliced correctly in vivo in the sponge. Whether the intron is self-splicing in nature, the pGEM clone plasmids were purified and linearized with Sca I restriction enzyme for in vitro transcription system. The RNA products were purified for intron splicing experiments. RT-PCR data showed that the intron could be self-spliced out correctly under certain conditions to produce the 447 bp band without additional enzyme or protein. The optimal splicing conditions were 50 mM MgCl2, 50 mM Tris-HCl, 1 M KCl and 1 mM GTP. Therefore, the intron is classified as type IB of self-splicing group I intron. However, when pGEM clones were introduced into E. coli, no splicing activity was observed, perhaps the RNA was preferentially degraded or some other factors are required in bacteria. The key nucleotides that are critical for the intron self-splicing are underway with site-directed mutagenesis. Keywords: Cinachyrella australiensis, group I intron, self-splicing.

1240

GLSBE-20 Co-Resistance and Multidrug Resistance of E. Coli Isolated from Water at Broiler Chicken Farms in Northeastern Thailand Sirinya Weluwanarak Sompao1, Apiradee Sopa2, Peerapol Sukon3, Pairat Sornplang2, Bongkot Noppon2*, Takuo Sawada4, Tetsuo Asai5 1 ,2

Department of Veterinary Public Health, Faculty of Veterinary Medicine, Khon Kaen University, Thailand 3

Department of Anatomy, Faculty of Veterinary Medicine, Khon Kaen University, Thailand

4

5

Nippon Veterinary and Life Science University, Japan Department of Applied Veterinary Science, Gifu University, Japan Email: [email protected]

Abstract E. coli was found contaminated in water at chicken farm to different extent. There was a limited study on co-resistance, and multidrug resistance of E. coli water isolates at chicken farm in northeastern areas of Thailand. Therefore, this study aimed at investigating the antimicrobial pattern of E. coli among 12 drugs (gentamycin, CN; cephazolin, KZ; cefotaxime, CTX; tobramycin, TOB; cefoxitin,FOX; erythromycin, E; ciprofloxacin, CPI; ofloxacin, OFX; trimethroprim, W; doxycycline, DO) from 6 groups of antibiotics (aminoglycosides, cephalosporins, macrolides, polypeptides, sulfonamide, and tetracycline). Disc diffusion technique was employed, and interpretive criteria were that of Clinical and Laboratory Standards Institute. Majority of E. coli water isolates were categorized into co-resistance to CTX-E (47/58, 81.0%), followed by KZ-CTX (42/58, 72.4%), and KZ-E (40/58, 68.9%). Only one E. coli isolate (1/58, 1.7%) was cross resistance to all 6 groups of antibiotics (KZ-CTX-TOB-CIP-CN-W-FOX-OFX-DO-E) in the present study. Nevertheless, at the most E. coli was cross resistance to 2 groups of drugs. In most nations, law, and regulation for water quality was that it must be free from E. coli. Proper monitoring of antimicrobial usages and better farm water management may play an important role in reducing risk of rendering E. coli from farm water. Keywords: chicken farm, co-resistance, E. coli, multidrug resistance, water Introduction E. coli is a gram-negative bacterium of the family Enterobacteriacae. It is a normal inhabitant of intestinal tract of birds (Singleton and Sainburg, 1981). It is one of the opportunist pathogen responsible for number of disease conditions in chicken such as enteritis, and many more (Rosenberger et al., 1985). Most strains of Escherichia coli are harmless, and they live in the intestines of healthy humans, and animals. Millions of germs can be released in a bowel movement from an infected human and animal. E. coli may be found in water sources, such as river, streams, creeks, private wells that have been contaminated with feces from infected humans, and animals. Waste can enter the water through many different ways, including sewage overflows, sewage systems that are not working properly, polluted storm water runoff, and agricultural runoff. Wells may be more vulnerable to such contamination after flooding, particularly if the wells are shallow, have been dug or bored, or have been submerged by floodwater for long periods of time. 1241

Water quality used for broiler chicken production and health is one of the most significant segments in health management. Water a vital nutrient in poultry metabolism which plays an important role in the digestion, absorption of food, transportation of nutrients in the body and elimination of waste products via urine. Physical, chemical and microbiological qualities of drinking water have a fundamental importance in poultry industry (Jafari et al., 2006). The role of water in spreading communicable diseases is much evident due to combined source of waters. Contaminated water with fecal coliform severely affects the performance of birds. The sources of water contamination are birds by feces, and secretions of sick birds, animals and humans (Desmarais et al., 2002). Salmonellae, Campylobacter spp. and Escherichia coli are the main poultry pathogens responsible for water contamination (He et al., 2007) at the present knowledge. The present study was conducted to check the antibiotic profiles of E. coli in different kinds of waters at 10 broiler chicken farms in northeastern Thailand. Material and methods Bacterial inoculum: E. coli was isolated from water at 10 chicken farms in northeastern Thailand. Freshly grown 58 isolates out of 112 E. coli isolates were standardized into 0.5% McFarland standard which is equal to 108 CFU/ml. It was then used to produce lawn of bacterial growth by spreading 0.1 ml onto the surface of Muller Hinton Agar plate (Oxoid, Hampshire, United Kingdom). Antimicrobial susceptibility testing: The 58 isolates were randomly sampling from 112 isolates for the antimicrobial susceptibility testing using disc diffusion method. The interpretive criteria were that of CLSI (2012). The 6 groups of drugs were aminoglycosides, cephalosporins, macrolides, polypeptides, sulfonamides, and tetracyclines, respectively. Among these, the 10 antimicrobial used were gentamycin (CN, Oxoid, Hampshire, United Kingdom), cephazolin (KZ, Oxoid), cefotaxime (CTX, Oxoid), tobramycin (TOB, Oxoid), cefoxitin (FOX, Oxoid), erythromycin (E, Oxoid), ciprofloxacin (CIP, Oxoid), ofloxacin (OFX, Oxoid), trimethoprim (W, Oxoid), and doxycycline (DO, Oxoid). Statistical analysis: Descriptive statistic was used to compare the different between mean diameters of inhibition zone using SPSS statistical package version 20.0 for antimicrobial susceptibility testing. ANOVA was used to compare differences in inhibition zone diameters among 6 groups of 10 antibiotics. Furthermore, Turkey HSD was also conducted to compare the differences between specific pair of drugs. Results E. coli was isolated from water at 10 chicken farms in northeastern Thailand. Later, 58 isolates were randomly sampling from 112 isolates for the antimicrobial susceptibility testing using disc diffusion method. It was evident that cefoxitin (38/58, 65.5%), and doxycycline (30/58, 51.7%) had similar best inhibition effect on E. coli. At the same time, resistance among E. coli isolates towards erythromycin (macrolides), and cefotaxime (cepharosporin) showed same resistance at 87.9% (51/58, both). Similar resistant rate was found in case of cephazolin (cephalosporins), and ciprofloxacin (polypeptides) at 74.1% (43/58). Trimethoprim (sulfonamides) and doxycycline (tetracycline) were in the intermediate resistant category (Table 1 and 2).

1242

Table 3 shows that majority of E. coli water isolates were categorized into co-resistance to CTX-E (47/58, 81.0%), followed by KZ-CTX (42/58, 72.4%), and KZ – E (40/58, 68.9%). Only one E. coli isolate (1/58, 1.7%) was resistance to all 6 groups of antibiotics (KZ-CTX-TOB-CIP-CN-W-FOX-OFX-DO-E) in the present study. Discussions and conclusions E. coli water isolates in particular areas in the northeast of Thailand had become resistance to certain group of antibiotics. It was evident that cefoxitin (38/58, 65.5%), and doxycycline (30/58, 51.7%) provided best growth inhibition on E. coli. At the same time, resistance among E. coli isolates towards erythromycin (macrolides), and cefotaxime (cepharosporin) showed same resistance at 87.9% (51/58, both). This high trend of resistance stemmed the need for proper management when prescribed these drugs due to limited inhibition. Mycek et al. (2000) encountered similar trend of resistance among their isolates and concluded that resistance to erythromycin is a severe problem from last decade until present. Similar resistant rate was found in case of cephazolin (cephalosporins), and ciprofloxacin (polypeptides) at 74.1% (43/58). The uses of these 2 drugs are of prime concern in this case. Muhammad et al. (2009) noted the higher resistant rate for ciprofloxacin compared to the present study of which 10.0% decrease in resistance was found. Add to that, trimethoprim (sulfonamides, 43.1% resistance), and doxycycline (tetracyclines, 20.7% resistance) were in the least resistant category. However, Oguttu et al. (2008) reported a much higher resistant rate of 98.2 for doxycycline in their study. The lowest resistant rate was found at 20.7% among 12 out of 58 E. coli water isolates for doxycycline. Majority of the isolates were categorized in to co-resistance, and limited number of the isolate was classified into multidrug resistance to 6 groups of drugs. However, care must be taken at all steps in administration of drug to alleviate the risk associated with E. coli.

Acknowledgements We are in deep appreciation to the research funding from Graduate school, Khon Kaen University, Khon Kaen, Thailand. Thanks also goes to Department of Veterinary Public Health, and the staff for facilitating the research work. We are appreciated by the warm welcoming of the farm owners to let us do the water sampling at their farm premises. References Clinical and Laboratory Standards Institute. (ed.) Performance Standards for Antimicrobial Susceptibility Testing: Twenty second informational supplement. PA: Clinical and Laboratory Standards Institute. 2012. Desmarais, TR, Solo-Gabriele HM, and Palmer CJ. Influence of soil on fecal indicator organisms in a tidally influenced subtropical environment. Appl Environ Microbiol, 2002;68:1165–1172. He, L-M, Lu J, and Shi W. Variability of fecal indicator bacteria in flowing and ponded waters in southern California: implications for bacterial TMDL development and implementation, Water Res. 2007;41(14):3132–3140. Jafari, RA, Fazlara A, and Govahi M. An Investigation into Salmonella and Fecal coliform contamination of drinking water in broiler farms in Iran. Int. J. Poultry Science, 2006;5:491-493. Muhammad Ali Akond, Saidul Alam, S.M.R. Hassan, Momena Shirin. Antibiotic Resistance of Escherichia coli Isolated From Poultry and Poultry Environment of Bangladesh. Internet Journal of Food Safety, 2009;11:19-23. 1243

Mycek, MJ, Harvey RA, and Champe PC. Lippincotts Illustrated Reviews; Pharmacology, 2 nd edition London: Lippincotts Williams and Wilkins. 2000. Oguttu, JW, Veary CM, and Picard JA. Antimicrobial drug resistance of Escherichia coli isolated from poultry abattoir workers at risk and broilers on antimicrobials. Journal of the South African Veterinary Association, 2008;79(4):161-166. Rosenberger, K, Fries PA, Cloud SS, and Wilson RA. In vitro and in vivo characterization of avian Escherichia coli. II. Factors associated with pathogenicity. Avian Diseases, 1985;29:1094-1107. Singleton, P, and Sainsburg D. Dictionary of Microbiology. International Ed. John Willey and Sons Ltd. Publication, New York, 1981:425-426

1244

1245

1246

1247

GLSBE-26 Enzymatic Synthesis of Pentyl-α-Glycosides with Surfactant and Antibacterial Properties Rittichai Charoensapyanana, Kazuo Itob, Prakarn Rudeekulthamrongc, Jarunee Kaulpiboond a Biochemistry and Molecular Biology Ph.D. Program, Thammasat University, Thailand E-mail: [email protected] b

Department of Biology, Osaka City University, Japan E-mail: [email protected]

c

Department of Biochemistry, Phramongkutklao College of Medicine, Thailand E-mail: [email protected] d

Department of Pre-Clinical Science (Biochemistry), Thammasat University, Thailand E-mail: [email protected], [email protected] 1.

Background/Objectives and Goals

Cyclodextrin glycosyltransferase (CGTase, E.C. 2.4.1.19) are multifunctional enzymes that can catalyze mainly intra- and inter-transglycosylation reactions (cyclization, coupling, disproportionation and hydrolysis reaction). The variety of enzymatic reactions makes its use in many applications, such as alkyl glycosides production. Alkyl glycosides are non-ionic surfactants which become dominant components in products of the food, detergent, cosmetic and pharmaceutical industries as a result of their biodegradability, emulsification and antibacterial properties. This study was to use the recombinant CGTase from Bacillus circulans A11 for the synthesis of pentyl-α-glycosides (PenGn) having more than one glucose unit from β-cyclodextrin (β-CD) as glycosyl donor and 1-pentanol as acceptor. Furthermore, we display the structure and biological properties of pentyl-α-glycosides, which has not been reported. 2.

Methods

The recombinant CGTase enzyme obtained from E. coli BL21 (DE3) cell harboring the pBC recombinant plasmid, a pET19b-based plasmid containing the CGTase gene from Bacillus circulans A11, was used to synthesize the pentyl glycosides via the 1248

transglycosylation reaction. The parameters involved in the reaction were optimized in a sequential independent variable by changing some conditions of the transglycosylation reaction. The pentyl glycoside products were prepared in large-scale and purified by preparative TLC. The structure of products was determined by mass spectrometry (MS) and nuclear magnetic resonance (NMR) analysis. Moreover, the transferred products were examined the emulsification properties by forming an oil-in-water emulsion and the antibacterial activities by investigating the inhibition zone, MIC and MBC values. 3.

Expected Results/ Conclusion/ Contribution

The optimal condition for the synthesis of pentyl glycosides consisted of an incubation of 1.5% (w/v) β-CD and 500 U/mL of CGTase in a water/1-pentanol content containing 5% (v/v) 1-pentanol at pH 6.0, 50°C for 168 h. Under these conditions, at least three products were observed by TLC and mass spectrometer. These transferred products had the molecular weight of 250.1, 412.2 and 574.4 daltons which were in accordance with pentyl glucoside (PenG1), pentyl maltoside (PenG2) and pentyl maltotrioside (PenG3), respectively. PenG1 and PenG2 were produced and prepared on a large-scale and purified by preparative TLC. The structure of transglycosylated products was confirmed to be 1 13 pentyl-α-D-glucopyranoside and pentyl-α-D-maltopyranoside by H- and C-NMR analysis. The two purified products were then checked for the emulsification properties by testing their ability in formation of n-hexadecane-in-water emulsion. The results revealed that PenG1 had emulsification activity of 36% while PenG2 had 38% of TritonTM X-100. Both PenG1 and PenG2 (5mg/disc) showed a moderately antibacterial function against S. aureus with an inhibition zone of 10 and 12 mm, and E. coli with inhibition zone of 14 and 16 mm, respectively. As a result, it showed that the PenG2 had a higher antibacterial activity than PenG1. In addition, PenG2 was more effective against inhibiting the E. coli growth which justified from MIC and MBC values. The lowest MIC and MBC values of PenG2 against S. aureus were found to be 1.88 and 1.88 mg/mL while that of E. coli were 0.94 and 1.88 mg/mL. In conclusion, the enzyme from the pBC recombinant cell containing the CGTase gene from Bacillus circulans A11 could be used to synthesize pentyl-α-D-glycosides having one-to-several glucose residues from the glycosyl donor as β-CD and from the 1-pentanol as acceptor through the transglycosylation reaction. At the optimal condition, the yield of pentyl glycosides obtained was 40% (w/w) of total products. The acceptor products of 1-pentanol were confirmed as pentyl-α-D-glucopyranoside and pentyl-α-D-maltopyranoside by ESI-TOF-MS and NMR analysis. Moreover, the 1249

products obtained from this study also exhibited an emulsification potential and antibacterial activity against E. coli and S. aureus. R. Charoensapyanan was supported by the Ph.D. Research-Assistant Scholarship from Thammasat University. The authors would like to acknowledge the National Research Council of Thailand (NRCT) pass to the Thammasat University (TU) for financial sponsorship of this research under the fiscal year 2015-2016. In addition, this work was performed under the Core-to-Core Program, which was financially supported for the CCP exchange program Year 2015-2016 by Japan Society for the Promotion of Science (JSPS), National Research Council of Thailand (NRCT), Vietnam Ministry of Science and Technology (MOST), the National University of Laos, Beuth University of Applied Sciences, Germany and Brawijaya University, Indonesia. Partial support from the Phramongkutklao College of Medicine Research Fund is also acknowledged. Keywords: Antibacterial activity, Cyclodextrin glycosyltransferase, Intermolecular transglycosylation, Non-ionic surfactant, Pentyl glycosides

1250

GLSBE-31 Discovery of Potent Tyrosinase Inhibitors from Chinese Herbals and Oligopeptides Yu-Ching Leea,b#, Tien-Sheng Tsengc,d, Wang-Chuan Chene, Nai-Wan Hsiaod, Keng-Chang Tsaib,c* a The Center of Translational Medicine, Taipei Medical University, Taiwan. b College of Medical Science and Technology, Taipei Medical University, Taiwan. c National Research Institute of Chinese Medicine, Ministry of Health and Welfare, Taiwan. d Institute of Biotechnology, National Changhua University of Education, Taiwan. e

The School of Chinese Medicine for Post Baccalaureate, I-Shou University, Taiwan. # E-mail: [email protected], *[email protected] 1.

Background/ Objectives and Goals

Tyrosinase, widely distributed in microorganisms, plants and animals, is an enzyme that plays a very important role in melanin synthesis. A variety of skin problems, such as spots and freckles, are caused by excessive pigmentation in the skin due to tyrosinase. The whitening ingredients currently available in the market are mostly chemically synthesized, and over-dosing could easily cause many side effects. A recent study has shown that arbutin is metabolized to produce hydroquinone and kojic acid, both of which are known carcinogens and are banned in Japan and other countries. In view of the trend of using natural medicines to reduce side effects, the use of herbal compounds and peptides as raw materials or additives is surely more safe and attractive. 2. Methods We have screened whitening ingredients from Chinese herbal compounds through a TCM compound database. The tyrosinase inhibitory activity of Gastrodia elata T1 molecule (bis(4-hydroxybenzyl)sulfide) (IC50 = 0.53 M) is better than that of kojic acid by about 100-fold and arbutin by about 1000-fold. Meanwhile, we comprehensively evaluated the inhibitory capability of 20 × 20 dipeptides against mushroom tyrosinase. We found that a cysteine-containing dipeptide, CE, which directly blocks the active site of tyrosinase, is highly potent in inhibition (IC 50 = 2.0 M). In addition, we used molecular docking and pharmacophore modeling to discover a novel tripeptide CRY which shows the most striking inhibitory potency against mushroom tyrosinase (IC50 = 6.16 M). We further systematically 1251

investigated the tyrosinase-inhibitory abilities of tetrapeptides by constructing a phage-display random tetrapeptide library to screen a novel tyrosinase tetrapeptide CRVI inhibitor (IC50 = 2.7 M). 3. Expected Results/ Conclusion/ Contribution This study reveals the functional role of cysteine residue positional preference and the selectivity of specific amino acids in cysteine-containing di-, tri-, and tetra-peptides against tyrosinase. Furthermore, the molecular modeling demonstrates that the sulfur atom of four inhibitors (T1 Chinese herbal compound, dipeptide CE, tripeptide CRY and tetrapeptide CRVI) coordinating with the copper ions in the active site of tyrosinase is essential for mushroom tyrosinase inhibition. These results indicate that these inhibitors are promising candidates in developing pharmacological and cosmetic agents of great potency in skin-whitening. Keywords: Tyrosinase, Oligopeptide, Melanin, Docking, Natural product

1252

GLSBE-35 Antiproliferation Effect of Curcumin and Tetrahydrocurcumin on Cholangiocarcinoma Cells Nuntiya Somparna, Veerapol Kukongviriyapanb a Department of Preclinical Science, Faculty of Medicine, Thammasat University, Rangsit Campus, Thailand. Email: [email protected] b

Department of Pharmacology, Faculty of Medicine, KhonKaen University, Thailand Email: [email protected]

1. Background/Objectives and Goals Curcumin (CUR) is a phenolic compound from Curcuma longa. A variety of pharmacological effects of CUR have been reported, including anti-inflammatory, antioxidant and anticancer activities. Recently, tetrahydrocurcumin (THC), a colorless derivative of CUR and one of the major metabolites of CUR, has been focused on its antioxidant, chemopreventive activity and application in pharmaceutical products. However, there is a lack of studies on the anticancer effect of CUR compared with THC in Cholangiocarcinoma cell lines (CCA). Therefore, this study aims to investigate the effect of THC and CUR on cell proliferation and mitochondrial transmembrane potential in CCA. 2. Methods CUR as curcuminoids purity of >99%, by HPLC and THC of purity >99%, by HPLC were from the Research and Development Institute, Government Pharmaceutical Organization, Thailand. K100 Cholangiocarcinoma cell lines were used in this study . K100 cells were seeded onto 96-well culture plates at a density of 5 x 103 cells/well. After an overnight culture, the cells were pre-treated with CUR or THC 1colorimetric assay. To measure the change in mitochondrial transmembrane potential (∆Ψm), cells were seeded in 96 black well plates at the density of 10,000 cells/well was performed by using the cationic, lipophilic dye, 5,5′,6,′-tetrachloro-1,1′,3,3′ tetraethyl-benzimidazolyl carbocyanine iodide (JC-1) (Clayman Chemical) staining. 3. Expected Results/Conclusion/Contribution 1253

The anti-proliferation effect of THC and CUR were dose and time dependence manner. The IC50 values (50% cell growth inhibitory concentrations) for the individual compounds on K100 were determined (IC50 at 24 hr were 107± 13.4 µM and 30±12.3 µM for THC and CUR respectively. These results suggest that THC and CUR have anti-proliferative effects in CCA and these effects are more pronounced in CUR treated cells. We examined the effect of THC and CUR on mitochondrial integrity by measuring the mitochondrial membrane potential ( ∆Ψm ) using JC-1 The effects of THC and CUR on ∆Ψm changes were evaluated. After 3 hr incubation, ∆Ψm dissipation was detectable in both CUR and THC treated cells. From these results, the data suggested that CUR and THC provide anti-proliferation effect in K100 cells. The mechanism of anti-proliferation effect may be attributable to dissipation of the mitochondrial electrochemical potential gradient which is known as an early event leading to apoptosis. However, these effects are more pronounced in CUR treated cells. Keywords: Curcumin, tetrahydrocurcumin, Cholangiocarcinoma cell lines

1254

ISFAS-1442 Mathematical Model of Diphtheria with Vaccination in Thailand Puntani Pongsumpun Department of Mathematics, Faculty of Science, King Mongkut‘s Institute of Technology Ladkrabang, Thailand Corresponding author: [email protected], [email protected] Abstract Diphtheria is an acute, toxin-mediated disease caused by the bacterium Corynebacterium diphtheriae. It can be transmitted from person to person by close physical and respiratory contacts. In Thailand, this disease can occur for adult and children because some persons may not obtain the efficient vaccination. In this paper, we formulate the mathematical model for describing the transmission of diphtheria with vaccination in Thailand. Analysis of this model is given and the way for reducing the outbreak of this disease is described. Keyword: Diphtheria, Mathematical model, Simulation,Vaccination 1. Introduction The infectious disease caused by the bacterium Corynebacterium diphtheriaem called as Diphtheria.[1] The symptoms of this disease vary from mild to severe. After exposure of this disease, they usually start two to five days. The symptoms starts from a sore throat and fever. In severe cases, a grey or white patch develops in the throat. This can create a barking cough as in croup. Diphtheria is an infection spread only from people to people. The people are the only known reservoir for it. It is contagious through direct physical contact of 4-6 droplets breathed out into the air. Secretions from the nose and throat. Infected skin lesions. Sometimes it can be contacted via intermediary objects such as bedding or clothes but can be rarely found.Previous infection may not prevent against future infection.[2-3] Some persons have the bacteria without showing symptoms, but he/she can spread the disease to the others. Diphtheriae cause different severities of disease. The symptoms are due to a toxin produced by the bacteria. The outbreak of diphtheria is usually found in children who have not been vaccinated or have not completed the vaccine group and in people who have not been vaccinated or incompletely vaccinated children in adults. This can make no immunity to prevent disease. The epidemic of this disease can be found in the Northeast and in the south of Thailand. Although the infectious rate of this disease is found only a few, but Diphtheria is a serious disease in children younger than 5 1255

years and in adults over age 40 years. The risk of death is higher than other groups. This disease is found in both female and male and all age groups. But this disease is not often found in infants under 6 months because children at this age have immune to this disease from their mothers. The immunity will disappear when the child is about six months. From fig.1, we will see that this disease occur every year.

Many research papers studied about the mathematical model of diseases [4-5]. In 1998, Shulgin et al [6] formulated the SIR (S = Susceptible, I = Infected, R = Recovered) model described an outbreak of measles by considering the volume of vaccination. The result of this study was the volumes of vaccination and age of people who should get the vaccination. In 2001, Kolibo et al. [7] created a mathematical model of diphtheria vaccine taking into account the amount of each person can receive the vaccination. Parameters of their model are shown to reduce the outbreak of diphtheria. In this study, we formulate and analyze model for the transmission of diphtheria in Thailand by considering the vaccination. 2. Mathematical Model Mathematical model of diphtheria with vaccination is considered in this study. The diagram for describing our model equations is described in figure 1.

1256

Our model equations are explained in equations(1) to (4):

1257

We suppose that there is the constant number of population, then the rate of change for human population equals to zero. The result is the birth rate is equivalent to the death rate or b = d. Introducing the new variables:

the normalized equations become:

3. Analysis of model The known method applied to many mathematical models is standard dynamical modeling method. We find equilibrium state and then determine its local stability. 3.1 Equilibrium point: The equilibrium point is obtained by setting (6)-(8) equal to zero. The equilibrium

1258

3.2 Numerical Solutions:

1259

1260

4. Conclusion In this paper, mathematical model of diphtheria with vaccination is considered. We will see that the vaccination effects to the people who can be diphtheria. If the person gets vaccination then that person has less probability of being diphtheria case as shown in fig.4. The numerical solutions are shown in fig.4.

1261

From the above figure, we will see that the proportion of human who got vaccination is less than the proportion of human who do not get vaccination. Therefore the vaccination effects to the diphtheria infection of people. Acknowledgment This work is supported by Faculty of Science, King Mongkut's Institute of Technology Ladkrabang, Thailand. References [1] Pan American Health Organization. Diphtheria. Control of Diphtheria, Pertussis, Tetanus, Haemophilus influenza type b, and Hepatitis B FIELD GUIDE, 2005. [2] Annual Epidemiological Surveillance Report, 2013, Division of Epidemiology, Ministry of Public Health, Royal Thai Government. [3] Centers for Disease Control and Prevention (CDC).Diphtheria. Epidemiology and Prevention of Vaccine-Preventable Disease.The Pink Book: Course Textbook – 12th edition second printing; May 2012 [cited 28 May 2014]; pp. 75-86. [online], Available: www.cdc.gov/vaccines/pubs/pinkbook/downloads/dip.pdf [4] S.Sangsawang, T.Tanutpanit, W.Mumtong and P.Pongsumpun, Local Stability Analysis of Mathematical Model for Hemorrhagic Conjunctivitis Disease, KMITL Science and.Technology Journal 2012,12(2); pp.189-197.. [5] Esteva L.and Vargas C.2000. Influence of vertical and mechanical transmission on the dynamics of dengue disease. Mathematical Bioscience, 167, 51-64. [6] Shulgin B., Stone L. and Agur Z., Pulse Vaccination Strategy in the SIR Epidemic Model, Bulletin of Mathematical Biology 1998, 60; pp.1123-48 [7] Kolibo D., Romaniuk S., Mathematical model of the infection process in diphtheria for determining the therapeutic dose of antitoxic anti-diphtheria serum], Ukr Biokhim Zh. 2001,73(2): pp.144-51. [8] Leah EK., Mathematical models in biology, Random House, 1988. [9] Hoffman JD., Numerical Methods for Engineers and Scientists, Singapore: McGraw-Hill. [10] Hoffman JD., Numerical Methods for Engineers and Scientists, Singapore: McGraw-Hill.

1262

ISFAS-1473 Effectiveness of Therapy for Pregnant Low Back Pain – A Systematic Review M.Y. Li1, C.W. Kan1, A.S.W. Wong1, Y.L. Kwok1, J. Yip1, S.P. Ng2, T.H.T. Lao3 1 Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong 2 The Hong Kong Community College, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong 3 Department of Obstetrics and Gynaecology, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong E-mail: [email protected] Abstract This study aims to review recently published clinical presentation on the effectiveness and impact of currect therapy for low back pain (LBP) during pregnancy. This would help to build an information background for future research. Keywords: Therapy, pregnant, low back pain, review Background 50–70% of pregnant women have experienced some form of LBP during pregnancy, pregnancy-related low back pain (PLBP) and/or pregnancy related pelvic girdle pain (PPGP) [1-4]. Typically the pain is located in the sacroiliac area [3, 5]. There is a large range of pain relief strategies and therapeutic therapy for pregnant low back pain [6]. This study aims to review recent clinical presetation investigating the effectivenss of different therapy. Methods Literature search was performed in May 2013 to obtain recent published research articles studying the effectiveness of different therapy to relief pregnant low back pain. Databases searched were: CINAHL database (1982+) and MEDLINE via OVIDSP. Advanced search was used. The keywords used include low back pain, pelvic pain combined with pregnant, pregnancy, treatment and therapy. Google was used to further explore obtained web links from the databases. Reference list of retrieved articles were also scanned for additional relevant articles. In order to review recent research to provide a more up-to-date information background, studies have to be published after 2000 so as to be included in this reivew. 1263

Results The literature search generated 233 results, 52 of them from CINAHL, and 181 of them from MEDLINE. All titles and abstracts were reviewed and 220 were excluded due to duplication, research on low back pain not related to pregnancy, not a clinical trial nor pilot study on any type of therapy for pregnant low back pain. After reviewing all the selected studies, it is observed that recent therapy for reducing low back pain can be categorized into three main groups: (i) maternity support belts or binders, (ii) acupuncture and (iii) water aerobics and exercises. Their effect on pregnant low back pain have been researched as a preventive or therapeutic treatment to manage or reduce pregnant LBP. Some important findings are summarised in Table 1. Discussion Maternity support garment Among the four categories of maternity support garment (belts, briefs, cradles and torso supports) [6], research data is found only for the effectiveness of belts to reduce pregnant LBP, no studies have investigated the effectiveness of the other three types of maternity support garments. The effect of reducing pregnant LBP by maternity support belt has been examined and proven [10, 15, 16]. Maternity support belts are regarded as a safe, low cost and accessible device [16] and have been recommended for the management of lumbar spine and/or pelvic pain symptoms [20-23]. The belt application is shown to have a mechanical effect of reducing the mobility, laxity and sagittal rotation of the sacroiliac joints in healthy women and in women with pregnancy-related pelvic girdle pain [17, 19, 24]. The pelvic belt is found to cause a significant decrease in the sagittal rotation in the sacroiliac joints [24]. It is found that sacroiliac joint laxity was significantly reduced when the belt was worn in a high position (just below the anterior superior iliac spines) rather than in a low position (at the level of the symphysis) [17]. In another similar study, sacroiliac joint laxity was significantly decreased in both high and low belt positions, compared to that in no-belt condition. The belt application in the high position decreased the sacroiliac joint laxity to a significantly greater degree than that in the low position [19]. These prove the use of a support belt may improve lumbo-pelvic stability. The hypothesis is that the support may either press the articular surfaces of sacroiliac joint together and/or it may place the sacroiliac joint in an extreme position to provide stability [6]. Although the results of most studies support the maternity belt effectively reduces pregnant low back pain, however, there is study reported that the use of maternity belt 1264

has no additionl effect. Researcher has evaluated the effect of combined treatment of maternity belt with other therapy. The study shows that there is reduction in pain, however, there were no statistically significant differences between groups (i.e the use of maternity support belt did not add effect to other treatment) [11]. Acupuncture The effectiveness of acupuncture is investigated and reported that it relieved pain without serious adverse effects in late pregnancy [8, 13, 18, 25]. Research suggests that it may be advantageous to begin acupuncture therapy later in pregnancy to maximize pain relief [7]. However, another finding showed that acupuncture had no significant effect on pain or on the degree of sick leave compared with non-penetrating sham acupuncture [8]. Therefore, the efficiency of reducing pregnant LBP by acupuncture remains inconclusive. Researcher also compared the effect of acupuncture in the treatment of LBP in comparison with physiotherapy [13, 18]. A researcher found that both acupuncture and stabilising exercises constitute efficient complements to standard treatment for the management of pelvic girdle pain during pregnancy, while acupuncture is superior to stabilising exercise stabilising exercises [13]. Results from another study show that the mean visual analogue scale values in the acupuncture group were significantly lower both in comparison with the values before treatment and in comparison with the values of the physiotherapy group. Exercise and water aerobics A common treatment of LBP is physiotherapy. There are evidences shown that exercise and water aerobics are effective means to prevent or reduce pregnant LPB. Researchers have reviewed the effect of exercise on pregnant LBP [13-15, 26, 27]. They found evidence to support exercise as an effective means to improve functional outcomes. Researcher examined the effect of home exercise and clinic group exercise, and there is no significance difference between them [15]. Research also shown that counseling and physical training may alleviate LBP and decrease the need for sick leave. Water aerobics is also found to diminish sick leave and it is concluded that it can be recommended to pregnant women [12]. Previous research supported that water gymnastics has recently been reported to reduce the intensity of LBP in pregnant women [28]. Conclusion In this review paper, pubications related to the effectiveness and impact of current therapy for low back pain during pregnancy were reviewed and it was observed that 1265

recent therapy for reducing low back pain could be categorized into three main groups: (i) maternity support belts or binders, (ii) acupuncture, and (iii) water aerobics and exercises. After reviewing these clinical studies, it is found that the preventive and therapeutic effects of maternity support garments and acupuncture cannot be concluded. More research data is necessary before their effectiveness can be confirmed. Future research should evaluate both the independent effect and combined effect with other treatment of them. Acknowledgement Authors would like to thank the financial support from the Research Grant Council of The Hong Kong Special Administrative region, China (Project No.: PolyU 5177/12E) and The Hong Kong Polytechnic University (RTBM).

1. 2. 3. 4.

5. 6.

7.

8.

References Berg G, Hammar M, Möller-Nielsen J, Lindén U, Thorblad J. Low back pain during pregnancy. Obstetrics & Gynecology. 1988; 71(1):71-75. Mogren IM, Pohjanen AI. Low back pain and pelvic pain during pregnancy: prevalence and risk factors. Spine. 2005; 30(8): 983-991. Ostgaard HC, Andersson GB, Karlsson K. Prevalence of back pain in pregnancy. Spine. 1991; 16(5): 549-552. Wang SM, Dezinno P, Maranets I, Berman MR, Caldwell-Andrews AA, Kain Z. Low back pain during pregnancy: prevalence, risk factors, and outcomes. Obstetrics and Gynecology. 2004; 104(1): 65-67. MacEvilly M, Buggy D. Back pain and pregnancy: a review. Pain. 1996; 64(3): 405-414. Ho SS, Yu WW, Lao TT, Chow DH, Chung JW, Li Y. Effectiveness of maternity support belts in reducing low back pain during pregnancy: a review. Journal of Clinical Nursing. 2009; 18(11): 1523-1532. Ekdahl L, Petersson K. Acupuncture treatment of pregnant women with low back and pelvic pain – an intervention study. Scandinavian Journal of Caring Sciences. 2010; 24(1): 175-182. Elden H, Fagevik-Olsen M, Ostgaard HC, Stener-Victorin E, Hagberg H. Acupuncture as an adjunct to standard treatment for pelvic girdle pain in pregnant women: randomised double ‐ blinded controlled trial comparing acupuncture with non‐penetrating sham acupuncture. BJOG: An International Journal of Obstetrics & Gynaecology. 2008; 115(13): 1655-1668.

1266

9.

Mrtensson L, Stener-Victorin E, Wallin G. Acupuncture versus subcutaneous injections of sterile water as treatment for labour pain. Acta Obstetricia et Gynecologica Scandinavica. 2008; 87(2): 171-177. 10. Kalus S, Kornman L, Quinlivan J. Managing back pain in pregnancy using a support garment: a randomised trial. BJOG: An International Journal of Obstetrics & Gynaecology. 2008; 115(1): 68-75. 11. Haugland KS, Rasmussen S, Daltveit AK. Group intervention for women with pelvic girdle pain in pregnancy. A randomized controlled trial. Acta Obstetricia et Gynecologica Scandinavica. 2006; 85(11): 1320-1326. 12. Granath AB, Hellgren MSE, Gunnarsson RK. Water aerobics reduces sick leave due to low back pain during pregnancy. Journal of Obstetric, Gynecologic, & Neonatal Nursing. 2006; 35(4): 465-471. 13. Elden H, Ladfors L, Olsen MF, Ostgaard HC, Hagberg H. Effects of acupuncture and stabilising exercises as adjunct to standard treatment in pregnant women with pelvic girdle pain: randomised single blind controlled trial. BMJ (Clinical Research Edition). 2005; 330(7494): 761. 14. Depledge J, McNair PJ, Keal-Smith C, Williams M. Management of symphysis pubis dysfunction during pregnancy using exercise and pelvic support belts. Physical Therapy, 2005; 85(12): 1290-1300. 15. Nilsson-Wikmar L, Holm K, Oijerstedt R, Harms-Ringdahl K. Effect of three different physical therapy treatments on pain and activity in pregnant women with pelvic girdle pain: a randomized clinical trial with 3, 6, and 12 months follow-up postpartum. Spine. 2005; 30(8): 850-856. 16. Carr, C.A., Use of a maternity support binder for relief of pregnancy‐related back pain. Journal of Obstetric, Gynecologic, & Neonatal Nursing. 2003; 32(4): 495-502. 17. Damen L, Spoor CW, Snijders CJ, Stam HJ. Does a pelvic belt influence sacroiliac joint laxity? Clinical Biomechanics. 2002; 17(7): 495-498. 18. Wedenberg K, Moen B, Norling Å . A prospective randomized study comparing acupuncture with physiotherapy for low‐back and pelvic pain in pregnancy. Acta Obstetricia et Gynecologica Scandinavica. 2000; 79(5): 331-335. 19. Mens JM, Damen L, Snijders CJ, Stam HJ. Mens, J.M.A., et al., The mechanical effect of a pelvic belt in patients with pregnancy-related pelvic pain. Clinical Biomechanics. 2006; 21(2): 122-127. 20. Borg-stein AJ, Dugan AS, Gruber AJ. Musculoskeletal aspects of pregnancy. American Journal of Physical Medicine & Rehabilitation. 2005; 84(3): 180-192.

1267

21. Balık G, Balık MS, Üstüner I, Kağıtcı M, Şahin FK, Güven ESG. Hand and wrist complaints in pregnancy. Archives of Gynecology and Obstetrics. 2014; 290(3): 479-483. 22. Ritchie JR. Orthopedic considerations during pregnancy. Clinical Obstetrics and Gynecology. 2003; 46(2): 456-466. 23. Perkins J, Hammer RL, Loubert PV. Identification and management of pregnancy-related low back pain. Journal of Nurse-Midwifery. 1998; 43(5): 331-340. 24. Vleeming A, Buyruk HM, Stoeckart R, Karamursel S, Snijders CJ. An integrated therapy for peripartum pelvic instability: A study of the biomechanical effects of pelvic belts. American Journal of Obstetrics and Gynecology. 1992; 166(4): 1243-1247. 25. Kvorning N, Holmberg C, Grennert L, Å berg A, Å keson J. Acupuncture relieves pelvic and low-back pain in late pregnancy. Acta Obstetricia et Gynecologica Scandinavica, 2004; 83(3): 246-250. 26. Nascimento SL, Surita FG, Cecatti JG. Physical exercise during pregnancy: a systematic review. Current Opinion in Obstetrics & Gynecology. 2012; 24(6): 387-394. 27. Richards E, van Kessel G, Virgara R, Harris P. Does antenatal physical therapy for pregnant women with low back pain or pelvic pain improve functional outcomes? A systematic review. Acta obstetricia et gynecologica Scandinavica. 2012; 91(9): 1038-1045. 28. Kihlstrand M, Stenman B, Nilsson S, Axelsson O. Water‐gymnastics reduced the intensity of back/low back pain in pregnant women. Acta Obstetricia et Gynecologica Scandinavica. 1999; 78(3): 180-185.

1268

1269

1270

1271

ISFAS-1491 Preparation and Electrical Properties of the TmMnO3/SrTiO3: Nb Heterostructures Tai-Chun Han1, Yi-De Chung1, Yih-Tsyr Chu1, Wen-Lan Chen1, Jauyn Grace Lin2 1 Department of Applied Physics, National University of Kaohsiung, Taiwan 2 Center for Condensed Matter Sciences, National Taiwan University, Taiwan E-mail: [email protected] Background Perovskite manganite TmMnO3 (TMO) has received considerable attentions because of its multiferroic properties [1]. Moreover, the multiferroic thin film may have differential magnetic and electric properties from bulk due to the strain induced by the mismatch between the film and the underlying substrate [2]. However, so far only a little work was done on the electrical transport characteristics of orthorhombic TMO (o-TMO) all-oxide epitaxial heterostructures. In this work, therefore, we prepared the a-axis of o-TMO thin films on Nb-doped (110) SrTiO3 [NSTO (110)] substrates by using pulsed laser deposition (PLD). We also reported on the growth of the high quality o-TMO multiferroic films with a-axis perpendicular to the film surface, and the electric transport properties of the epitaxial heterostructure were studied. Experimental Methods The single-phased hexagonal TMO target was prepared by the conventional solid-state reaction method using a stoichiometric high-purity Tm2O3 and MnO2 powders as the starting materials. TMO thin films were grown on the NSTO substrates by the method of the PLD, with a 248 nm KrF excimer laser operated at a repetition rate of 5 Hz. The TMO films were deposited at substrate temperature of 850 oC, under 50 mTorr of oxygen pressure. The thickness of TMO films was about 100 nm. The crystalline structures and crystallographic orientation of the TMO thin films were examined with x-ray diffraction (XRD). For I-V measurement, a gold pad with area of 1 mm2 was deposited on the TMO film as the top electrode by sputtering technique through a shadow mask and the bottom electrode was In film (1 mm2) pressed on to the NSTO substrate to minimize the contact resistance. The I-V characteristics were measured at a temperature range of 60  320 K by applying a pulsed dc bias across the heterostructure with the voltage polarity being defined at the top Au electrodes. Results The XRD diffraction experiments show that the as-prepared TMO film is epitaxially a-axis oriented on NSTO (110) substrate. The temperature-dependent I-V curves show a nonlinear diode-like current behavior in the temperature ranges investigated. The diffusion voltage, at

1272

which the current I starts to increase rapidly, increases monotonically with reducing temperature from 320 to 60 K. On the other hand, the breakdown voltage, at which the current I starts to increase at reverse bias, increasing with reducing temperature in the range of 320  260 K, but decreases with reducing temperature from 240 to 60 K. The striking result is that at constant reverse bias the current increases in the magnitude as the temperature decreases from 240 to 60 K. Rare-earth perovskite manganites show the ability to accommodate more oxygen called ―oxidation nonstoichiometry‖. The excess oxygen might act as an accepter, changing insulating o-TMO into a p-type semiconductor. In addition, NSTO substrate is reported as a degenerate n-type semiconductor. The anomalous reverse bias current in o-TMO/NSTO can be explained by tunneling through the heterojuctions. The probability is mainly determined by the relative position of the top of the TMO valence band and the bottom of the NSTO conduction band. The tunneling current is dominant over the thermionic current below 260 K, the bias voltage for reaching a given tunneling current varies with temperature (T) to be satisfied with VkBT = constant, which is evidenced by the linear temperature dependence of the voltages for reaching 10 and 20 A negative current on the reverse bias side below 260 K. In summary, all-oxide o-TMO/NSTO heterojunctions were fabricated by the PLD method. Below 260 K, the anomalous increasing reverse current with decreasing temperature is due to the strain-assisted tunneling through the junctions. Keywords: perovskite manganite, multiferroic, electrical characteristic, heterojuction References [1]. V. Yu. Pomjakushin, M. Kenzelmann, A. Donni, A. B. Harris, T. Nakajima, S. Mitsuda, M. Tachibana, L. Keller, J. Mesot, H. Kitazawa, and E. Takayama-Muromachi: Evidence for large electric polarization from collinear magnetism in TmMnO3. New J. Phys. 2009; 11: 043019. [2]. Y. Yu, X. Zhang, J. J. Yang, J. W. Wang, and Y. G. Zhao: Microstructure and strain relaxation of orthorhombic TmMnO3 epitaxial thin films. J. Cryst. Growth 2012; 338: 280–282.

1273

ISFAS-1506 Effect of Cobalt Doping on the Magnetic and Electric Properties of GaFeO3 Nanoparticles Yih-Tsyr Chu*, Yi-De Chung, Chun Yen, Wen-Lan Chen, Tai-Chun Han 1 Department of Applied Physics, National University of Kaohsiung, Taiwan E-mail address: *[email protected] Background Gallium ferrite, GaFeO3 (GFO), has ferromagnetic (FM) and piezoelectric orderings and has been intensively studied recently for its potential application as a magnetoelectric (ME) ferrimagnet [1]. Site-disorder in bulk GFO is reported to be induced by adopting different preparation conditions and methods, which are shown to affect the magnetic properties significantly [1-3]. Moreover, the physical properties of GFO at reduced grain sizes remain underexplored due to the challenges associated with their synthesis with a proper control over the crystalline phase. In addition, the origin of the doping effects on the correlation between the site-disorder and magnetism in the nanosized GFO system has not yet been completely established. Therefore, in this work, we prepare a series of GaFe1-xCoxO3 (GFCOx) (0  x  0.1) nanoparticles by a sol-gel method to systematically study the effect of Co-doping on their magnetic and electric properties. Experimental Methods The GaFe1-xCoxO3 (GFCOx, 0  x  0.1) nanoparticles were synthesized by a sol-gel method using nitrates as metal precursors. First, gallium nitrate [Ga(NO3)36H2O], iron nitrate [Fe(NO3)39H2O], and cobalt nitrate [Co(NO3)26H2O] in stoichiometric proportions were dissolved in distilled water. Citric acid (C6H8O7) in 1:1 molar ratio, with respect to the metal nitrates, was added to the solution as a complexant. The mixture was dried 120 oC to form a gel, and then the obtained gel was burned until the combustion process was completed. After that, the precursory powders were reground and sintered at 700 oC for 2 h. The temperature- and field-dependent magnetizations were measured with a Quantum Design superconducting quantum interference device magnetometer. For the dielectric and leakage current density measurements, the powders were pressed into the disk 10 mm and 1 mm in thickness under a pressure of 1.5 GPa and then coated with 100 nm thick silver layers on both the top and the bottom sides of the disk as electrodes. A capacitance bridge hooked to a probe station with a closed-cycle low temperature system was used. Results

1274

Based on the standard reference, all the observed XRD patterns of the GFCOx nanoparticles can be indexed on the basis of an orthorhombic unit cell of space group, Pc21n, suggesting that all samples are pure phases without any impurity. The mean particles sizes of the GFCOx nanoparticles are around 15 nm. The fact that the value of lattice parameters a, b, and c change systematically with increasing cobalt concentration implies that the Co-ions do replace the Fe-ions. The magnetic characterization indicates that with increasing Co-content, up to 0.1, the FM transition temperature (TC) increases from 325 to 355 K. The magnetic hysteresis curves measured below TC exhibit two-phase-like magnetic behavior consisting of soft and hard magnetic phases. The results could be explained by the site-disorder of the cations in GFCOx nanoparticles. Furthermore, thermo-magnetic irreversibility in zero-field cooled (ZFC) magnetization is observed for all the GFCOx nanoparticles, which are attributed to the inherent magnetic anisotropy. Interestingly, variable temperature dielectric measurements show an anomaly in the dielectric constant at the temperature close to the TC indicates the ME coupling of the GFCOx nanoparticles. In addition, doping GFO with magnetic Co2+-ions reduces the charge conduction. We have evidenced the Co-doping in GFO allows decreasing the leakage current density and tuning the nature of carriers simultaneously. These results open wide perspectives for the room temperature FM GFO ferrite in multiferroic and ME devices. Keywords: ferrite, multiferroic, magnetoelectric, nanoparticles References [1]. T. Arima, D. Higashiyama, Y. Kaneko, J. P. He, T. Goto, S. Miyasaka, T. Kimura, K. Oikawa, T. Kamiyama, R. Kumai, and Y. Tokura: Structural and magnetoelectric properties of Ga2-xFexO3 single crystals grown by a floating-zone method. Phys. Rev. B 2004; 70: 064426. [2]. M. B. Mohamed, A. Senyshyn, H. Ehrenberg, and H. Fuess: Structure, magnetic, dielectric properties of multiferroic GaFeO3 prepared by solid state reaction and sol-gel methods. J. Alloys Compd. 2010; 492: L20–L27. [3]. K. Sharma, V. R. Reddy, D. Kothari, A. Gupta, A. Banerjee, and V. G. Sathe: Low temperature Raman and high field 57Fe Mossbauer study of polycrystalline GaFeO3. J. Phys.: Condens. Matter 2010; 22: 146005.

1275

ISFAS-1508 3 Dimensional Measurement by Hilbert Transform in Interferometric Fringe Microscopy Silin Naa, Sanghoon Shinb, Younghun Yua* a Department of Physics , Jeju Nat. Univ., Korea E-mail: [email protected], *[email protected] b

KPS, Hawsung, Korea E-mail: [email protected] Abstract We use Hilbert microscopy (HM) as a optical technique for measuring high spatial resolution quantitative phase images of samples. Hilbert microscopy is suitable for investigating fast measurement since HM needs only one image. We use interferometric method for making fringe to get high spatial resolution and large depth of field. The pitch of the interference pattern could be controlled by angle between two interference beams and the depth of field also be controlled by varying width of beam. The potential of this technique for large depth sample is demonstrated with step-wise samples. Keywords: Fringe projection profiler, Hilbert microscopy, Depth of field. 1. Introduction Fringe projection profilometry (FPP) to retrieve the surface shape of the three dimensional objects is one of the active research area in optical metrology [1-7]. Application range of FPP from measuring micro scale objects to large engineering structures. The technique is used in material testing, quality control, reverse engineering, biomedical engineering, on-line inspection, surface roughness measurements, etc. It is whole-field, non-contact, non-scanning, flexible, inexpensive and capable of providing high resolution. The aim of fringe pattern analysis is the extraction of some physical quantities from optical phase distribution in one or several fringe patterns. Multiple images method such as phase shifting techniques which require at least three images are used for extracting phase information [8-10]. Single image method such as Fourier transform (FT) and Hilbert transform (HT) has been employed [11-15]. In FT method the image process in frequency domain whereas HT method can be processed in spatial domain. The HT method has several advantages such as simplicity in calculation and automatic implement. During the 3D shape measurement, a set of fringe patterns are projected onto the surfaces of the objects of interest. The surface height/depth information is naturally encoded into the distorted fringe patterns, which will be captured by the camera for further processing. The

1276

required features for a good projection pattern are: a perfect sinusoidal irradiance function, a very high contrast, high irradiance and a large depth of field. And high resolution patterns are required for accurate measurement [16-18]. In FPP, the depth of field (DOF) of projection lens should be larger than the height of

and large depth of field. In this paper we used interference pattern for high spatial resolution and large field of depth, and used Hilbert transform retrieving the phase distribution.

2. Theory Fig.1 shows typical fringe projection profilometry. If a sinusoidal fringe pattern is projected on the surface of a sample, the intensity distribution of the distorted fringe pattern can be expressed as

distribution to profile the height variation of the sample. To calculate the phase distribution, phase shift method is used with multiple images. And Fourier Transform (FT) and Hilbert Transform (HT) methods are used for retrieving the phase. FT and HT need only one image. In

1277

FT method the image is processed in frequency domain and HT method can be processed in spatial domain. In mathematics and in signal processing, the Hilbert transform is a linear operator which takes a function, g(t), and produce a function, HT(g(t)), in the same domain:

Based in Eq. (5), the phase distribution including carrier frequency of the fringe pattern can be acquired by

1278

3. Results Fig. 2 shows the schematic experimental set-up of Hilbert microscope with interferometric. We used laser interference pattern for fringe pattern. The pitch of the interference pattern is controlled by varying the angle between interference beams, the angle between beams can be varied by varying the distance between BS and M in Fig. 2, and depth of field could be controlled by controlling the overlapping area of beams.

We can show that the pitch of the fringe pattern decreased with increasing the angles. Fig. 3 (b) shows relation between DOF and beam width at pitch 89 m. We can show that the DOF is depends on the beam width. We can be obtained DOF in the few mm range by using this

1279

method. From Fig. 3 we can show that the pitch of fringe and DOF could be controlled by angle and width of interference beams.

distribution with (black line) and without filter (red dot line), (d) Hilbert transformed image, (e) double Hilbert transformed image, (f) intensity profile. Fig. 4 and Fig. 5 show the experimented results of step-wise sample. The reference plane is situated at distance of 9 cm. The camera, sample and interference beam are situated on a right triangles as shown in Fig. 2. When the linear interference pattern is formed on the reference plane, the fringe will be straight and regularly spaced of plane as shown in Fig. 4 (a). The pitch of interference pattern is 98.8 m. The distorted fringe pattern on a step-wise sample is shown in Fig. 4 (b). Fig. 4 (c) shows intensity distribution of pattern (Fig. 4 (b) dot line) and filtered intensity distribution. We use windowed Fourier Filtering method for filtering the noise. Basically WFF method is low-pass filter to suppress noise like speckle [5]. The step-wise sample is made by overlapping thin glass (thickness=34 m). To retrieve the phase, we have done HT. Fig. 4 (d) is the fringe pattern of single HT Eq. (3), and Fig. 4(e) is the result of double HT Eq. (4). Fig. 4 (f) is the profile of dotted line in Fig. 4 (d) and (e). From Fig. 4 (f), we can show that the background is suppressed and phase is shifted by 2 as expected in Eq. (3) and (4). We have calculated the phase distribution with Hilbert transformed image and double Hilbert transformed image. In this process we have corrected the phase error introduced in Hilbert transform [9]. Fig. 5 shows the calculated phase and height distribution. Fig. 5 (a) and (b) are phase and height distribution, and Fig. 5 (c) is the profiles of dotted line in Fig. 5 (b). The height is 36.9 4.7 m, which value is almost same with the thickness of thin glass.

1280

Fig. 6 is the result of another sample, which is made by cover glass. The thickness of cover glass is 107 m. Fig. 6 (a) is the distorted fringe pattern by sample and (b) is the calculated height distribution as like Fig. 5. Fig. 6 (c) is the profile of dotted line in Fig. 6 (b). The height is 112.8 4.2 m. From experimental results we can show that the interference fringe microscopy with Hilbert transform is useful and simple method for measuring 3D shape measurement.

4. Conclusion The fringe projection profilometry are widely used in three-dimensional shape measurement. In FPP, the required fringe pattern is high spatial resolution, high contrast and high depth of field. We have used interference pattern to meet the requirement of fringe pattern. We have demonstrated the usefulness of Hilbert transformation with single captured image for three dimensional measurements. This method requires only one image for evaluation of the surface shape. Also we have demonstrated interference pattern instead projection fringe pattern, which pitch and depth of filed could be controlled by varying angle and beam width. 5. Acknowledgement This research is supported by the small and medium business administration (SMBA). 6. References [1] S.S. Gorthi, P. Rastogi.(2010). Fringe projection techniques: whither we are?. Opt. Lasers Eng, 48(2), 133–140. doi:10.1016/j.optlaseng.2009.09.001

1281

[2] S.S Gorthi, G. Rajshekhar, P. Rastogi.(2011). Three dimensional shape measurement using

[3] [4] [5] [6] [7]

high-order instantaneous moments based fringe projection method. Opt. Laser Technol, 43(1), 40–44. doi:10.1016/j.optlastec.2010.05.002 T. Yoshizawa (Ed.).(2009). Handbook of Optical Metrology: Principles and Applications. CRC Press, Boca Raton, FL. F. Chen, G.M. Brown, M. Song.(2000). Overview of three-dimensional shape measurement using optical methods. Opt. Eng., 39(1), 10–22. doi:10.1117/ 1.602438 Q. Kemao.(2013). Windowed Fringe patteren analysis. SPIE press, Bellingham, Washington. Salas L, Luna E, Salinas J, Garcia V, Servin M.(2003). Profilometry by fringe projection. Opt. Eng., 42(11), 3307-3314. doi:10.1117/1.1607968 Tay C, Quan C, Wu T, Huang Y.(2004). Integrated method for 3-D rigid-body

displacement measurement using fringe projection. Opt. Eng., 43(5), 1152–1159. doi:10.1117/1.1687728 [8] K.G. Larkin, B.E. Oreb.(1992). Design and assessment of symmetrical phase-shifting algorithms, J. Opt. Soc. Am., 9(10), 1741–1748. doi:10.1364/ JOSAA.9.001740 [9] K. Hibino, B.E. Oreb, D.I. Farrant.(1995). Phase shifting for non-sinusoidal waveforms with phase-shifting errors. J. Opt. Soc. Am., 12(4), 761–768. doi: 10.1364/JOSAA.12.000761 [10] Jia PR, Kofman J, English CE.(2007). Multiple-step triangular-pattern phase shifting and the influence of number of steps and pitch on measurement accuracy. Appl. Opt., 46(16), 3253–3262. doi:10.1364/AO.46.003253 [11] M. Takeda, H. Ika, S. Kobayasha.(1982). Fourier transform method of fringe pattern analysis for computer based topography and interferometry. J. Opt. Soc. Am., 72(1), 156– 160. doi:10.1364/JOSA.72.000156 [12] T. Ikeda, G. Popescu, R. R. Dasari, and M. S. Feld.(2005). Hilbert phase microscopy for investigating fast dynamics in transparent systems, Opt. Lett., 30(10), 1165-1167. doi:10.1364/OL.30.001165 [13] S. Wang, L. Xue, J. Lai, Z. Li.(2013). An improved phase retrieval method based on Hilbert transform in interferometric microscopy. Optik, 124(14), 1897-1901. doi:10.1016/j.ijleo.2012.05.029 [14] U. P. Kumar, U. Somasundaram, M. P. Kothiyal, N. K. Mohan.(2013). Single frame digital fringe projection profilometry for 3-D surface shape measurement. Optik, 124(2), 166-169. doi:10.1016/j.ijleo.2011.11.030 [15] D, J, Bone, H. A. Bachor, and R. J. Sandeman.(1986). Fringe-pattern analysis using 2D Fourier transform. Appl. Opt., 25(10), 1653-1660. doi:10.1364/ AO.25.001653 [16] ZhaoyangWang, DungA.Nguyen,JohnC.Barnes.(2010). Some practical considerations in fringe projection profilometry, Opt. Lasers Eng., 48(2), 218–225. doi:10.1016/j.optlaseng.2009.06.005

1282

[17] Wang Z, Du H, Bi H.(2006). Out-of-plane shape determination in fringe projection profilometry. Optics Express, 14(25), 12122–12133. doi: 10.1364/OE.14.012122 [18] D. Li, C. Liu, and J. Tian.(2014). Telecentric 3D profilometry based on phase shifting fringe projection. Optics Express, 22(26), 31826-31835. doi:10.1364/ OE.22.031826

1283

ISFAS-1517 Low Cost Ultraviolet-Visible Spectroscope Using LED Light Source S.Kamoldiloka, P.Phongwisitb Department of Physics, King Mongkut‘s Institute of Technology Ladkrabang, Thailand a E-mail: [email protected], b [email protected] 1. Abstract and Introduction This paper was studied and setup an low cost Ultraviolet-Visible Spectroscope system by used LED light source. Because of commercial instrument was expensive cost, that performance and ability was over needed in general school and intermediate laboratory. This setup was change light source from xenon lamp to LED, specific wavelength can select by LED to reduce the cost. Result of this setup was compared with the commercial instrument by the same sample to understood how difference of performance. This setup can apply to portable system or used in education and more. Ultraviolet-Visible Spectroscopy is an instrument that used to studied transmission and absorption property for any sample. To studied the both of property, intensity of incident and transmit light were measure and compared intensity to find result in term of transmittance and absorbance. In Ultraviolet-Visible Spectroscopy system, intensity of incident light and transmit light were compared by Beer-Lambert law to find the result in term of transmittance (1) and absorbance (2).

Ultraviolet-Visible Spectroscopy system consists with 4 parts as Light Source, Monochromator, Sample Holder and Detector. This system can operate by start light source that give the light in length of 200-800 nm. (UV to Visible) This light were travel to Monochromator and change to monochromatic light when reflect from Monochromator. After that, monochromatic light were travel to Sample Holder and transmit to Detector. In this system, an expensive cost is a part of Light Source and Monochromator because of they are high quality and accuracy. In this paper, we try to reduce a cost to create low cost Ultraviolet-Visible Spectroscopy system for every classroom and intermediate laboratory. To save the cost, we change Light Source and Monochromator to LED. Because of LED have a cheap cost and give a specific wavelength. User can select wavelength in an experiment by change LED lamp.

1284

2. Experiment setup and Results In this paper, we change light source and Monochromator to use LED. This system can start with light travel from LED to Collimator len and change to collimated light. After that, collimated light were travel to Sample and transmit to Detector. In this step, user can calculate transmittance and absorbance for this sample and this wavelength by compare intensity of light.

With this setup, we measure transmission property of three trick film samples. Results from this setup show on table1.

To confirm performance and accuracy of our setup, in this paper we used the commercial Ultraviolet-Visible Spectroscope Avantes-Avaspec EDU to prove our setup‘s performance. Our prove show on table2 in term of percentage difference between our setup and the commercial system.

1285

3. Conclusion In this paper, we show an idea to setup and performance of low cost Ultraviolet-Visible Spectroscope using LED light source. This setup has performance and accuracy close to the commercial system. For more wavelengths testing, user can add more LED to select more specific wavelength. This setup can apply for more type of sample and suitable for intermediate laboratory and general education school. 4. References D.A. Skoog, Principles of Instrumental Analysis, 3rd Edition. Saunders College Publishing Japan, 1985. Keywords: Ultraviolet-Visible Spectroscope, LED, low cost

1286

ISFAS-1544 Investigation on Loss of Hexagonal Arrangement in Holey Fiber D. Chittinan1, S. Kamoldilok1, S. Lekchaum1 1 Department of Physics, King Mongkut‘s Institute of Technology Ladkrabang, Thailand. E-mail: [email protected] Abstract This paper investigated the confinement loss of hexagonal lattice by structure designed of pitch, shape of holes and air holes size of Large Mode Area (LMA) holey fiber with hexagonal arrangement, it‘s made from pure silica. By using the OptiFDTD programming based on the finite-difference time-domain (FDTD) method and its wavelength is 1.55 micron, the result of simulation shown that the confinement loss was proportional to changing pitch, shape of holes and air holes size. Keywords: Holey fiber, photonic crystal fiber, confinement loss, finite difference time domain Introduction Photonic crystal fiber (PCF) is optical waveguide of fiber based on the properties of 2-D photonic crystal which a 2-D photonic crystal is periodic along two directions. PCF is an arrangement of air hole and PCF has been used light confinement in hollow core methods to propagate light. PCF is now finding application in fiber optic communication, high power laser and nonlinear devices [1]. Moreover, PCF is also known as holey fiber. Holey Fiber (HF) are microstructures that solid core surrounded by cladding determined by array of air holes and guide light of holey fiber have core and cladding that difference effective refractive index and can be made from single-material typically pure silica by hole which are of a similar size to the wavelength of light when light of different wavelengths can be passed. As a result, Holey fibers have experienced loss of less than conventional fibers.

1287

Holey fiber is defined by parameter Λ is the lattice period, r is radius of air hole and d is diameter of air hole. As shown in Fig.1 In this paper we studied confinement loss in five rings hexagonal arrangement holey fiber by varying the pitch, shape of holes and air holes size. In addition, the solid core was designed and the simulated used the OptiFDTD programing based on the finite-difference time-domain (FDTD) method by changed difference diameter of the hole and pitch in the PCF which comparative to achieve very low confinement loss by evaluated for TE and TM of holes core. Analysis method and modelling of 2-D photonic crystal fiber Light is an electromagnetic wave which can be described by the following the Maxwell‘s equations and electromagnetic theory is summarized in the following mathematical equation. The general equations in partial differential equation are given by[2]

Where, E is the electric field and H is the magnetic field, respectively, The electric and magnetic fields are associated: D=ɛE and B=µH, where ɛ is the dielectric permittivity and µ is the magnetic permeability. Then, the wave equation for E and H can be presented at the following:

Therefore, the electric and magnetic fields of a mode can be written as follows

1288

1289

In figure 3. We have changed the normalized hole diameter in the PCF. The normalized hole diameter is varied equal to 0.15µm.

In figure 4. We have again changed the normalized hole diameter of inside 3-layer in the PCF. The normalized hole diameter of 3-layer is varied equal to 0.15 µm and outer 2-layer equal to 0.25 µm. Then, we have calculated confinement loss. The confinement loss was calculated from the imaginary part (Im), using the following equation:[3]

After simulation by using OptiFDTD programing we have found the lowest confinement loss is 0.016627 dB/km. Then, we have plotted confinement loss, as shown in Fig. 5.

Conclusion

1290

In this paper, we have proposed the design waveguide of fibers. The normalized hole diameter and pitch is varied by analyzed confinement loss. The result shows that the comparative between the normalized hole diameter in 5-layer balanced and the normalized diameter in various layers of hexagonal different, we have been found when the diameter of inside 3-layer and outer 2-layer is changed to low confinement loss. The proposed PCF is useful for the applications in non-linear, bio-sensing, optical sensing, display devices, communication systems, etc. References 1. Jeong I. Kim: Analysis and Applications of Microstructure and Holey Optical Fibers, Ph.D Thesis Virginia Polytechnic Institute and State University. 2003;105-118. 2. Jackson J. D. Classical electrodynamics 3rd edition. 1999; 2-3. 3. Poli F, Cucinotta A, and Selleri S: Photonic crystal fibers Properties and applications. 2007;223.

1291

ISFAS-1606 Fabrication of Photodiode Sensors and Measurements of Photodiode Detector Coupled with a Scintillator H. Park, H.B. Jeon, K.H. Kang, B.B. Kim, T.H. Kim, S.C. Lee1 Dept. of Physics, Kyungpook National University, Korea Email: [email protected] I. Abstract The silicon photodiode sensor coupled with a scintillator is being developed for an X-ray cargo scanner. The sensor was designed and fabricated to collect the scintillation light as much as possible considering the fill factor, the size, thickness of anti-reflective coating layer and depth of the implantation. The scintillator‘s properties (emission wavelength, light yield and decay time) of CdWO4 (CWO) crystals coupled with the photodiode sensor were measured. The relative quantum efficiency of the photodiode as a function of the wavelength and the signal-to-noise ratio of the photodiode were measured. The energy resolutions of the photodiode detector coupled with a scintillator using radioactive sources were also measured. Keywords: PIN diode, Scintillator, Photodiode, X-ray, Electrical Characteristics II. Introduction The demand for technology development for security devices to check the dangerous materials due to the risk of terrorism and national security purpose has been increased. We aim to develop a cargo scanner using a dual-beam X-ray. We want to use a combination of a scintillator and a photodiode detector module to detect X-ray. The photodiode is commonly used instead of photomultiplier (PMT) because of its higher quantum efficiency, smaller size, and lower operational voltage than PMT. The CWO crystal is used to be coupled with the photodiode sensor for X-ray scanner because of good detection efficiency and low afterglow optimized to high-speed scanner [1]. III. Experiment We measured scintillator‘s properties (emission wavelength, light yield and decay time) of CsI(Tl) and CWO crystals [2]. The relative quantum efficiency of the manufactured photodiode as a function of the wavelength was also measured. The signal-to-noise ratios of the photodiode were measured using a Sr-90 radioactive source. The photodiode detector coupled with CsI(Tl) crystal was used to measure the energy resolution at FWHM with Na-22, Cs-137, Mn-54, Zn-65, Co-60 and Am-241 sources. The pulse height spectra measured at room temperature with 100 V of an operational voltage was measured [3]. We checked the correlation between the radioactive energies irradiated the scintillator and ADC counts.

1292

IV. Summary Under X-ray excitation, the emission of the CWO crystal was measured between 400 and 700 nm peaking at 496 nm. The two decay time components of 10.1 ns (98.0%) and 0.7 ns (2.0%) were observed at room temperature under γ-rays excitation from a Cs-137 source. The signal-to-noise ratio of the photodiode using commercial electronics was measured to be better than 80. The energy resolutions at FWHM with the photodiode detector coupled with CsI(Tl) using 511 keV, 662 keV, 835 keV, 1115 keV, 1173 keV and 1333 keV γ-rays excitation energies were measured to be 18.3%, 13.6%, 9.3%, 7.3%, 8.8% and 7.5%, respectively. We also confirmed the linearity between irradiated gamma energies and ADC counts. The energy resolution measurements with the photodiode detector coupled with the CWO crystal is in progress. References 1. G. F. Knoll, Radiation Detection and Measurement, (John Wiley-Sons, 2010). 2. G.H. Chae, et al., Journal of the Korean Physical Society 63 (2013) 1455. 3. D.H. Kah, et al., Nuclear Instruments and Methods in Physics Research A 628 (2011) 256.

1293

ISFAS-1433 Synthesis of White-Light-Emitting Molecules for White Organic Light-Emitting Diodes Ying-Hsuan Wu 1, Kew-Yu Chen 1,2 Department of Chemical Engineering, Feng Chia University, Taiwan Email: [email protected] Background Organic fluorescent dyes with excited-state intramolecular proton transfer (ESIPT) characteristics have been drawing significant attention due to their unusual optical properties. This unique optical property has many practical applications, typical examples of which are probes for solvation dynamics and biological environments, fluorescence microscopy imaging, photochromic materials, chemosensors, nonlinear optical materials, near-infrared fluorescent dyes, and organic light emitting diodes(OLEDs). Taking advantage of an adiabatic ESIPT reaction, we were able to show that broad blue fluorescence from the excited starting material and broad red fluorescence from an ESIPT molecule 2 combined to generate white light as if there were two molecules present in the sample. However, due to its relatively low overall emission quantum yield, the electroluminescent performance using 2 as an active layer is only decent compared with the work mentioned early utilizing multi-ESIPT moieties. In an effort to expand the scope of the 2-based molecules available for designing systems for high performance single-molecule-based white-light-emitting OLEDs, we herein report the synthesis of a new derivative of 2 with a tert-butyl moiety attached to the phenol ring, i.e., 4-tert-butyl-1-hydroxy-11H-benzo[b]fluoren-11-one (1). Its X-ray structure, as well as optical properties and time-dependent density functional theory (TD-DFT) calculations are also studied. Methods Scheme shows the chemical structures and the synthetic routes of white-light-emitting small molecules 1 and 2. The synthesis of 1 started from a bromination of 7-methoxy-2,3-dihydro-1H-inden-1-one (6), followed by the elimination of the bromo adduct (5), giving a reactive dienophile 4. The naphthalene ring can then be fused onto the C(2)–C(3) double bond by placing 4 through a reaction with α,α,α‘,α‘-tetrabromo-o-xylene, yielding 3. Subsequently, compound 3 was subjected to the deprotection with BBr3 to give 2. Finally, the alkylation at the 4-position of 2 was executed by the reaction of 2 with tert-butyl alcohol and sulfuric acid, giving 1 with an overall product yield of 65%. The presence of a single tert-butyl

1294

group of 1 can be verified by the presence of a signal at δ 1.42 ppm (9H,singlet) and eight signals at δ 7.0–8.2 ppm (8H) in the 1 H NMR spectrum.

Results We have successfully synthesized and characterized a new ESIPT-based white-light-emitting small molecule (1) with a bulky tert-butyl group. Compound 1, as well as compound 2, undergoes an intramolecular proton transfer reaction in the excited state, resulting in a tautomer that is in equilibrium with the normal species, exhibiting a dual emission that generates white light. The introduction of the tert-butyl substituent not only increases the solubility of 1 compared with 2, but also improves the fluorescence intensity. Furthermore, analysis of the geometric structures clearly demonstrates that the intramolecular hydrogen bond length is shortened upon the photoexcitation, which is considered as a very important factor for ESIPT. Research on its application to single-molecule-based white OLEDs is currently in progress. Keywords: ESIPT, Tautomer, White-light-emitting small molecules, Stokes shift, X-ray diffraction, DFT calculations. References 1. T.C. Fang, H.Y. Tsai, M.H. Luo, C.W. Chang and K.Y. Chen: Chin. Chem. Lett. 24 (2013) 145. 2. W.T. Chuang, C.C. Hsieh, C.H. Lai, C.H. Lai, C.W. Shih, K.Y. Chen, W.Y. Hung, Y.H. Hsu and P.T. Chou: J. Org. Chem. 76 (2011) 8189. 3. K.C. Tang, M.J. Chang, T.Y. Lin, H.A. Pan, T.C. Fang, K.Y. Chen, W.Y. Hung, Y.H. Hsu and P.T. Chou: J. Am. Chem. Soc. 133 (2011) 17738. 4. K.Y. Chen, Y.M. Cheng, C.H. Lai, C.C. Hsu, M.L. Ho, G.H. Lee and P.T. Chou: J. Am. Chem. Soc. 129 (2007) 4534. 5. K.Y. Chen, C.C. Hsieh, Y.M. Cheng, C.H. Lai and P.T. Chou: Chem. Commun. 42 (2006) 4395.

1295

ISFAS-1435 Synthesis and Optical Properties of Novel Asymmetric Perylene Bisimides Fang-Yun Chien 1, Kew-Yu Chen 2 1,2 Department of Chemical Engineering, Feng Chia University, Taiwan, ROC Email: [email protected] Background Perylene bisimides (PBIs) and their related derivatives have attracted considerable attention due to their potential applications in molecular electronic and optical devices. PBIs have also been utilized as building blocks to construct supramolecular or artificial photosynthetic systems. These molecules are advantageous due to their high photochemical and optical stabilities, ease of synthetic modification and reversible redox properties. The most common products made from the above reaction were disubstituted (1,7- and 1,6-) symmetric PBIs, rather than asymmetric PBIs. Sometimes the asymmetric PBIs need to be designed for special applications. In an effort to expand the scope of the PBI-based chromophores available for designing systems for organic optoelectronic materials and fluorescent dyes, the present research reports the synthesis of a novel series of asymmetric PBIs, as well as their optical and electrochemical properties, and complementary time-dependent density functional theory (TD-DFT) calculations. Methods

Result We have successfully synthesized a novel series of cyan dyes based on asymmetric 1,7-disubstituted PBIs, i.e., 1-amino-7- nitroperylene bisimides (1a–1c). These molecules undergo an excited-state intramolecular electron transfer (ESIET) reaction, resulting in a unique

1296

charge transfer emission in the nearinfrared region, of which the peak wavelength exhibits strong solvatochromism. Upon excitation, they show larger dipole moment changes than those of the symmetric 1,7-diaminoperylene bisimides (2a–2c); the dipole moments of these compounds have been estimated using the Lippert–Mataga equation. Compounds 1a–1c display good thermal stability as well as the optical stability and reversible redox properties. In addition, the absorption and emission spectra of 1–3 calculated by quantum chemistry computations were in good agreement with the experimental results. Research on their applications to organic photovoltaic cells and biomolecular fluorescence imaging is currently in progress. Keyword: Asymmetric perylene bisimides; Intramolecular Solvatochromism; Solvatofluorism; Lippert–Mataga equation.

charge

transfer;

NIR;

References 1. B. Kaur, N. Quazi, I. Ivanov, S.N. Bhattacharya, Dyes Pigments 92 (2012) 1108. Scheme 2. One of the possible resonance structures for 1a–1c, explaining the disappearance of the charge transfer absorption band in concentrated HCl. 110 H.-Y. Tsai, K.-Y. Chen / Journal of Luminescence 149 (2014) 103–111 Author's personal copy 2. K.Y. Chen, T.C. Fang, M.J. Chang, Dyes Pigments 92 (2011) 517. 3. H.Y. Tsai, K.Y. Chen, Dyes Pigments 96 (2013) 319. [52] H. Dinçalp, Ş. KızılokS. İçli, Dyes Pigments 86 (2010) 32. [69] K.Y. Chen, T.J. Chow, Tetrahedron Lett. 51 (2010) 5959. 4. Z.J. Chen, L.M. Wang, G. Zou, L. Zhang, G.J. Zhang, X.F. Cai, M.S. Teng, Dyes Pigments 94 (2012) 410. 5. M.H. Luo, K.Y. Chen, Dyes Pigments 99 (2013) 456. [75] L. Hao, W. Jiang, Z. Wang, Tetrahedron 68 (2012) 9234.

1297

ISFAS-1436 Highly Soluble Monoamino-Substituted Perylene Tetracarboxylic Dianhydrides: Synthesis, Optical and Electrochemical Properties Yung-Cheng Hsieh 1, Kew-Yu Chen 2 1, 2 Department of Chemical Engineering, Feng Chia University, Taiwan (R.O.C.) E-mail: [email protected] Background PDIs and PTCDs suffer from serious problems, such as poor solubility and aggregation. A number of synthetic methods to prepare PDI and PTCD derivatives with improved solubility have been reported. The synthesis of highly soluble PDI and PTCD derivatives is very important for process ability and for the preparation of their thin films to be used in optoelectronics applications. Several diamino-substituted PTCDs based on this concept have been synthesized and studied so far. However, to the best of our knowledge, the molecular structures, as well as the optical and electrochemical properties of monoamino-substituted PTCDs have not been reported yet. Methods

Scheme 1 depicts the chemical structures and synthetic routes of monoamino-substituted asymmetrical PTCDs (1a–1c). In brief, the synthesis of 1a–1c started from an imidization of perylene dianhydride 6 and cyclohexylamine in the presence of acetic acid, followed by the

1298

mononitration of perylene diimide 5, giving a nitro compound 4. The reduction of 1-nitroperylene diimide (4) by tin (II) chloride dehydrate (SnCl2·2H2O) in refluxing THF afforded 1-aminoperylene diimide (3). Next, three highly soluble PDI derivatives (2a–2c) with different N-alkyl chain lengths (n = 6, 12 or 18) were prepared by the alkylation of 3 with the corresponding alkyl halides. Finally, alkylamino-substituted PDIs 2a–2c were converted to the respective PTCDs via saponification to afford 1a–1c.

It appears that the inductive effect of the alkyl groups in 1a and 2a causes an additional red shift. Interestingly, the longest wavelength absorption band of 1a is 40-nm red-shifted relative to that of 2a Results We have successfully synthesized three monoamino-substituted asymmetrical perylene tetracarboxylic dianhydrides with different n-alkyl chain lengths (1a–1c). They show a unique charge transfer emission in the near-infrared region, of which the peak wavelengths exhibit strong solvatochromism. Upon excitation, they show slightly larger dipole moment changes than those of corresponding perylene diimides 2a–2c; the dipole moments of these compounds have been estimated using density functional theory calculations and the Lippert-Mataga equation.

Keywords: perylene tetracarboxylic dianhydride; perylene diimide; solvatochromism

References 1. 1. Tsai, H.Y.; Chang, C.W.; Chen, K.Y. 1,6- and 1,7-Regioisomers of asymmetric and symmetric perylene bisimides: synthesis, characterization and optical properties. Molecules 2014, 19, 327–341.

1299

2. 2. Tsai, H.Y.; Chen, K.Y. Synthesis and optical properties of novel asymmetric perylene bisimides. J. Lumin. 2014, 149, 103–111. 3. 3. Chen, K.Y.; Fang, T.C.; Chang, M.J. Synthesis, photophysical and electrochemical properties of 1-aminoperylene bisimides. Dyes Pigment. 2011, 92, 517–523.

1300

ISFAS-1437 A Low-Molecular-Weight Ratiometric Chemodosimeter for Highly Selective Naked-Eye and Fluorogenic Detection of Cyanide Jiun-Wei Hu 1, Kew-Yu Chen 2 1, 2 Department of Chemical Engineering, Feng Chia University, Taiwan E-mail: [email protected] Background CN is widespread in industrial processes, such as gold mining, electroplating, metallurgy, and the synthesis of nylon, fibers, and resins. Thus, it is highly desirable to develop selective, sensitive, simple, low-cost, colorimetric and fluorometric chemosensors for detecting CN. Methods

Compound 1 was synthesized via stirring a mixture of solution of 2 (159 mg, 1.0 mmol), malononitrile (330 mg, 5.0 mmol), a drop of triethylamine, THF (10 mL) and MeOH (5 mL) at room temperature for 2 h.

1301

The absorption and emission spectra of 1 in solvents of varying polarity are shown in Fig. 1. Sensor 1 shows two absorption bands at ~288 and 369 nm; the former is attributed to the π-π* transition and the latter to an ICT transition.

Upon the addition of F-, Cl-, Br-, AcO-, S2- , SCN-, NO2- , NO3- , CO3- , BzO-, H2PO4-, and HSO4-, no substantial change was observed in the fluorescence spectra of sensor 1. The only signifi- cant response was observed when CN was added (Fig. 2a); a blueshifted fluorescence was observed at 438 nm and the band at 533 nm (ESICT) diminished, which was easily observed by the naked eye under the irradiation of a UV-lamp (Fig. 2c). The 438 nm emission of 1-CN (Scheme 1) is similar to that of 1-methylindole, suggesting that the ICT is prevented by the addition of cyanide to sensor 1. After the addition of 25 equiv. of CN, the peak at 533 nm was completely quenched (Fig. 2b). Results Sensor 1 exhibits high reactivity and selectivity for CN with respect to other anions in aqueous media. An ESICT mechanism plays a key role in the sensing properties, the mechanism of which is supported by DFT and TD-DFT calculations. Additionally, the 1- based test strips are easily fabricated and have a low cost, and they can be used in practical and efficient CN test kits. More importantly, the optical sensor 1 can also be used for the ultrasensitive determination of CN in real samples. Keywords: Indole, Chemodosimeter, Colorimetric, Ratiometric fluorescence, Cyanide detection. References 1. G.C. Miller, C.A. Pritsos, Cyanide: Soc., Ind. Econ. Aspects, in: Proc. Symp. Annu. Meet., TMS, 2001, pp. 73-81. 2. W.-C. Lin, S.-K. Fang, J.-W. Hu, H.-Y. Tsai, K.-Y. Chen, Ratiometric fluorescent/ colorimetric cyanide-selective sensor based on excited-state intramolecular charge

1302

transferexcited-state intramolecular proton transfer switching, Anal. Chem. 86 (2014) 4648-4652. 3. K.-S. Lee, H.-J. Kim, G.-H. Kim, I. Shin, J.-I. Hong, Fluorescent chemodosimeter for selective detection of cyanide in water, Org. Lett. 10 (2008) 49-51.

1303

ISFAS-1475 Antibacterial Finishing of Cotton with Biostatic Approach Chi-Wai Kan Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Hong Kong E-mail: [email protected] Background Recently, there has been an increase in the range of textiles with antibacterial effect to airborne bio-particles and micro-organism in daily use. Generally speaking, the inhibition of the growth of microbes in textile material surface can be achieved through biocidal approach (destroy microbes) and biostatic approach (inhibits the growth of microbes). Currently, apparels (with biocidal agent) will be disposed after use which would induce harmful effect to environment and high operation cost. The biostatic approach of treating textile materials would introduce antibacterial effect and will not induce possible mutation to pathogens leading to development of unknown new bacteria [1-5]. Method To investigate the methodologies of applying biostatic agent to cotton Cotton fabric will be treated with biostatic agent for surface modification. With the use of biostatic agent, some key process parameters will be studied because their variations may significantly affect the final characteristics of the cotton material. In this study, biostatic agent will be coated on the cotton surface by covalent grafting reaction with suitable crosslinking agent via (i) pad-dry-cure and (ii) spraying to achieve both sides or one side antibacterial effect respectively. To evaluate and characterise the biostatic properties of the modified cotton by using various instrumental and international standard testing methods. The surface properties of cotton with biostatic effect will be evaluated and characterised with different instrumental techniques. (i) Physical properties: Contact angle measurement, surface roughness measurement and sliding angle. (ii) Topographical properties: Scanning electron microscopy (SEM) with energy-dispersive X-ray (EDX) analyzer. (iii) Chemical properties: Fourier transform infrared (FTIR)

1304

(iv) Performance properties: Spray test (AATCC Test Method 22), oil resistance test (AATCC Test Method 118), water/alcohol resistance test (AATCC Test method 193), air permeability (BS 5636 method), tensile strength (ASTM D5034) and durability under normal washing condition. To study biological response of the cotton with biostatic effect An in-vitro test capable of detecting the combined effects of chemical on cotton could give useful information for the product development. Biological safety and response of skin component will be studied using cultured skin model to analyse and simulate the toxicity impacts of the biostatic agent-treated cotton to human body. Cytotoxicity test will be performed using a fibroblast cell line (NIH 3T3) which is commonly found in skin and other connective tissues. After developing the cotton fabric with biostatic effect, the biostatic property will be first evaluated by standard antibacterial tests (AATCC Test Method 100 and AATCC Test Method 147) and then the surface elemental composition will be determined with EDX and FTIR. After confirming the chemicals on the material surface, the cotton fabric with biostatic properties will be subjected to the fibroblast cell line and the extent of cell death induced by the test agents will be identified. Cisplatin, cisplatinum or cisdamminedichloridoplatinum (II) (CDDP), a well-known anti-cancer drug that can cause the death of cell will be used as the positive control in the test. Results (1) Have good results in AATCC Test Method 100 - Assessment of Antibacterial Finishes on Textile Materials and AATCC Test Method 147 - Antibacterial Activity Assessment of Textile Materials: Parallel Streak Method. (2) Good durability under normal washing condition (domestic laundering). Acknowledgement This is a research project of HKRITA funded by Innovation and Technology Fund (ITF) with title ―Development of Anti-bacterial textile Materials by Biostatic Approcah with Poly(ethylene glycol) (PEG) for Healthcare Use‖ (ITP/049/13TP). Author would like to thank the financial support by ITF. Any opinions, findings, conclusions or recommendations expressed in this material/event (or by members of the project team) do not reflect the views of the Government of the Hong Kong Special Administrative Region, the Innovation and Technology Commission or the Panel of Assessors for the Innovation and Technology Support Programme of the Innovation and Technology Fund and the Hong Kong Research Institute of Textiles and Apparel. Also, we would like to thank the support and sponsorship from Fately Fashion Industrial Limited, Lai Tak Enterprises Limited, Synergy Alliance Company and Fifield (Asia) Limited.

1305

Keywords: Antibacterial, finishing, cotton, biostatic

References 1. Hall-Stoodley L, Costerton JW, Stoodley P. Bacterial biofilms: from the natural environment to infectious diseases. Nature Review – Microbiology. 2004; 2(2); 95-108. 2. Medilanski E, Kaufmann K, Wick L, Wanner O, Harms H. Influence of surface topography of stainless steel on bacterial adhesion. Biofouling. 2002; 18: 193-203. 3. Edwards KJ, Rutenberg AD. Microbial response to surface microtopography: the role of metabolism in localized mineral dissolution. Chemical Geology, 2001; 180: 19-32. 4. Bretagnol F, Lejeune M, Papadopoulou-Bouraoui A, Hasiwa M, Rauscher H, Ceccone G, Colpo P, Rossi F. Fouling and non-fouling surfaces produced by plasma polymerization of ethylene oxide monomer. Acta Biomaterialia. 2006; 2(2): 165-172. 5. Muir BW, Tarasova A, Gengenbach TR, Menzies DJ, Meagher L, Rovere F, Fairbrother A, McLean KM, Hartley PG. Characterization of low-fouling ethylene glycol containing plasma polymer films. Langmuir. 2008; 24(8): 3828-3835.

1306

ISFAS-1503 Structural and Thermal Characteristics of Alkali-Treated Graphites Seung-Won Songa, Hyunmyung Choib, Weontae Ohc* a Department of Advanced Materials Engineering, Dong-Eui University,Busan,Korea E-mail address: [email protected], [email protected], c*[email protected] 1. Background

Graphite receives much attention because of its excellent electrical and thermal characteristics. As a result, graphite becomes a promising candidate in many advanced material application. Among some methods to prepare graphene from graphite, the chemical exfoliation is very useful for the mass-production of graphene, but deteriorate its conductivities (electrical and thermal) due to the severe structural damages. For an alternative, the non-oxidative exfoliation method was considered to supplement the weakness of oxidative exfoliation. We have modified the non-oxidative exfoliation method to prepared alkali-treated graphite (a-Gr) and characterize the a-Gr structurally and thermally. 2. Methods

A modified non-oxidative exfoliation method was used to prepare alkali-treated graphite (a-Gr). Natural graphite (particle size ~45 ㎛) of 50g was added to the NaOH (5M) aqueous solution of 1L, and was stirred for 2h at room temperature. The resulting mixture was filtered out and dried for 24h in an oven of 100oC. The structure of as-prepared a-Gr was investigated by XRD, Raman, and XPS. In addition, the thermal conductivity of the a-Gr was analyzed by using LFA447 in the compressed pellet form. 3.

Expected Results

Alkali-treated graphite in this work was simply prepared with natural graphite. There exists a little amount of polar groups such as carboxylic acid, hydroxyl, and epoxy groups on the surface of natural graphite, which are caused by the adsorption and oxidation of humid and oxygen in the air. Besides, these polar groups decline the thermal and electrical characteristics of the graphite. We have identified the structural evidences that the oxidized groups are eliminated and the structural damages are recovered on the surfaces of the graphite. The thermal conductivity of the a-Gr was improved as compared to that of natural graphite. Keywords: Graphite, Graphene, Nonoxidative Exfoliation

1307

ISFAS-1505 Mechanical and Thermal Characteristics of the Thermal Interface Materials (TIMs) Prepared by Silicone Composite Materials Hyunmyung Choia, Seung-Won Songb, Weontae Ohc* Department of Advanced Materials Engineering, Dong-Eui University,Busan,Korea E-mail: [email protected], [email protected],c*[email protected] 1. Background

As an electronic device is highly integrated and its performance correspondingly improves, the heat generated on operation causes a serious problem in the device. To improve the durability and the performance of many components, the heat generated from the integrated electronic device must be suitably transferred to the outside and the device temperature must be controlled not to increase too much. Thermal interface materials (TIMs) are widely used to eliminate the heat generated on an electronic device. We have prepared TIMs including silicone-based composites and investigated the thermal conductivities of the TIMs. 2. Methods

To prepare the thermal conductive composites, AlN and BN semiconductor powders, and graphite powder were used for thermal conductive fillers, and silicone oils (Dimethyl siloxanes, DMS-T25 and DMS-T31) were used for the binders. AlN (max. 70wt.%) and BN (max. 40wt.%) were added to silicone binders separately. The mixtures were mechanically stirred for 20 min, and then further mixed for 1 min at 500 rpm, for 1 min at 700 rpm, and for 30 sec at 1000 rpm by using a paste mixer. The thermal diffusivities, and specific heats of the as-mixed composites were measured by using LFA 467 (Netzsch, Germany) and DSC214 (Netzsch, Germany), respectively. The thermal conductivities of the TIM composites were obtained from the following equation; λ (T) = α (T) • ρ(T) • Cp (T). Here, λ, α, ρ, and Cp indicate thermal conductivity, thermal diffusivity, bulk density, and specific heat of a sample, respectively. 3.

Expected Results

The thermal diffusivities of the silicone composites with AlN and BN were measured by using a laser flash method (LFA467). The thermal conductivities of the composite samples were calculated through the products of the densities, the specific heats and the thermal diffusivities. The thermal conductivities of TIM composites (AlN 70 wt.% and BN 40 wt.% loadings) were 1.637 W/mK and 0.833 W/mK, respectively. The thermal conductivity of silicone binder without any filler was 0.165 W/mK. The thermal conductivities of the composites increase with increase of filler loadings, but the increment was considerably limited as compared to the thermal conductivities of the fillers themselves. We have further studied the thermal conductivities of the silicone composites with AlN/graphite and BN/graphite. These results

1308

were compared to the thermal conductivities of the silicone composites. In addition, we have discussed the thermal conductivities related to the compositions and morphologies of the composites. Keywords: Thermal conductivity, Thermal interface materials (TIMs), Silicone composites materials, Graphite

1309

ISFAS-1551 Mechanism of NO Photocatalytic Removal on g-C3N4 Was Changed by Pd-QDs Modification Guohui Dongb, Wingkei Hoa a Department of Science and Environmental StudiesThe Hong Kong Institute of Education, Tai Po, N.T. Hong Kong, People‘s Republic of China Email: [email protected] b

Laboratory of Environmental Sciences and Technology, Xinjiang Technical Institute of Physics & Chemistry, and Key Laboratory of Functional Materials and Devices for Special Environments, Chinese Academy of Sciences, China Email: [email protected] Over the past four decades, various photocatalysts, such as oxides, sulfides, and oxynitrides, have been developed to utilize solar energy for environmental purification and energy conversion. However, most of these materials can only be excited by ultraviolet (UV) light, which occupies only 4% of solar light. Therefore, many researchers turn to develop new photocatalysts that are active under visible light. Among these endeavors, graphitic carbon nitride (g-C3N4) can produce oxygen or hydrogen by water splitting under visible light irradiation. More importantly, this metal-free polymer can withstand acid alkalis and high temperature because of the strong covalent bonds between carbon and nitride atoms. These advantages make g-C3N4 one of the most significant catalysts in the field of photocatalysis. Quantum dot (QD) sensitization can increase the light absorption and electronic transmission of photocatalysts. However, limited studies have been conducted on the photocatalytic activity of photocatalysts after modification by noble metal QDs. In this study, we have successfully prepared g-C3N4, which was modified with palladium QDs (Pd-Qds) by chemical reduction method for the first time. The resulting materials were carefully characterized and then used for the photocatalytic removal of organic NO under visible light irradiation. A series of experiments was designed to clarify the roles of Pd-QDs on g-C3N4 photocatalysis under visible light. The reasons for the enhancement of photocatalytic activity were analyzed in detail. Results showed that the modification of Pd-QDs can improve the NO photocatalytic removal activity of g-C3N4. The modification of Pd-QDs not only favored the visible light absorption of g-C3N4, but also improved the separation efficiency of photogenerated carriers, and thus enhanced the NO photocatalytic removal activity. More interestingly, Pd-QDs modification changed the NO removal mechanism from the synergistic action of h+ and ·O2− to the single

1310

action of ·OH. We found that the main reason of the mechanism change was that Pd-QDs modification changed the molecular oxygen activation pathway from single-electron reduction to two-electron reduction. This study could not only develop a novel strategy to modify the Pd-QDs on the surface of photocatalysts, but also shed light on the deep understanding of the relationship between Pd-QDs modification and NO photocatalytic removal activity of semiconductor photocatalysts. Keywords: NO removal, photocatalytic, g-C3N4, quantum dot, palladium

1311

ISFAS-1632 Dyeing Characteristics and Tensile Strength of High Tenacity Polyethylene (HTPE) Fibers Beom Young Lee, a Jeong Hoon Lee, a Tae Kyung Kimb, Seung Geol Lee†,a a Department of Organic Material Science and Engineering, Pusan National University, Republic of Korea †

b

E-mail: [email protected]

Department of Textile System Engineering, Kyungpook National University, Republic of Korea E-mail: [email protected]

1. Background/ Objectives and Goals High tenacity polyethylene (HTPE) fiber is one of the important synthetic fibers which possess outstanding properties such as low density, excellent surface hardness and scratch resistance, superior electrical insulation and low cost. Because of these benefits, HTPE fiber has a lot of practical applications such as ropes, shock withstand materials, stiffeners, sports apparels, composite materials, etc. However, it is well known that dyeing using ordinary dip dyeing method is almost impossible due to its low melting temperature and no polar functional group in molecular structure. Thus, it is difficult to apply HTPE to specific applications that require a variety of colors for apparels, especially in sportswear. In this study, we selected three primary colors of dyes by calculating LogP and dyed the HTPE filament and fabric to investigate the dyeing characteristics and tensile strength of HTPE. 2. Methods Certain sized filaments and fabrics of HTPE were immersed into the dye solution in IR dyeing machine at the temperature range of 80℃-120℃ for 1hr. The dye uptake and tensile strength of dyed HTPE filaments were analyzed by UV-VIS spectrophotometer and tensile testing method (KSK0412). We also analyzed surface properties and crystallinity of dyed HTPE by SEM and XRD. 3. Expected Results/ Conclusion/ Contribution By calculating LogP, we selected the hydrophobic dyes for three primary colors. The molecular structure with long aliphatic group generally showed higher hydrophobicity. The tensile strength of dyed HTPE filaments and fabrics was reduced and its dye uptake was increased as temperature increases, respectively. The tensile strength of HTPE filaments and

1312

fabrics was reduced with increasing the dyeing time, but the dye uptake of HTPE filaments and fabrics was increased and became a constant after certain time. Keywords: high tenacity polyethylene (HTPE); solvent dye, tensile strength, dye uptake Acknowledgment This research was financially supported by the Ministry of Trade, Industry & Energy (MOTIE), Korea Institute for Advancement of Technology (KIAT) and DaeGu Institute for Regional Program Evaluation (DGIRPE) through the Industries of Economic Cooperation Region.

1313

ACEAIT-7332 Synthesis and Properties of Thermal and Transparency of Dicyclopentadiene Modified Poly(Methyl Methacrylate) Chih-Wei Tsaia *, Fang-Ning Lib, Kuo-Hui Wub a School of Defense Science, Chung Cheng Institute of Technology, National Defense University, Taiwan b

Department of Chemical and Materials Engineering, Chung Cheng Institute of Technology, National Defence University, Taiwan *Corresponding Author: [email protected] Abstract Dicyclopentadiene (DCPD) is frequently utilized as the starting material for the preparation of cyclic olefin monomers and the corresponding polymers or copolymers. These cyclic olefin polymer products possess satisfactory refractive, thermally stable, and transparent properties, and can be utilized in optics, electronic products, biomedical materials, and material science. In this study, dicyclopentanyl methacrylate(DCPMA) was synthesized from dicyclopentadiene via hydroxylation, hydrogenation, and esterification. The homopolymer of poly(dicyclopentanyl methacrylate) (PDCPMA) was polymerized in tetrahydrofuran by reacting DCPMA with 2,2‘-Azobis(isobutyronitrile) (AIBN) as a free-radical initiatorat at 80°C. The copolymer of poly(dicyclopentanyl methacrylate-co-methyl methacrylate) (P(DCPMA-co-MMA))was synthesized by free radical copolymerization reaction DCPMA with MMA. The chemical structures of each of the intermediates, DCPMA, and PDCPMA, (P(DCPMA-co-MMA)) were confirmed from FTIR and 1H NMR measurements. Molecular weight determination by gel permeation chromatography showed that the polymers were obtained of Mn > 19,000 and Mw > 49,000. The thermal properties of PDCPMA and P (DCPMA-co-MMA) were studied with thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). The introduction of rigid and steric bulky dicyclopentanyl moiety into the polymer chain receptors diminishes the rotational mobility of the polymethacrylate ester resulted in improvements in the thermal properties compared with PMMA and all the polymers were found to be amorphous. The transparency of UV-visible region( >80%) after thermal aging of recycled high-impact PDCPMA and P(DCPMA-co-MMA) were studied using UV-vis spectroscopy. These characteristics of the dicyclopentadiene modified poly (methyl methacrylate) polymers could be owed to the steric bulky, rigid spacer and polycyclic cage unit molecular structure of dicyclopentadiene. Keyword: dicyclopentadiene; PMMA; thermal properties; optical properties.

1314

ACEAIT-7339 Preparation of Shell Crosslinked Micelles as Drug Carriers from Poly(N-isopropyl acrylamide)-co-poly(L-lysine) Graft Copolymer You-Liang Tua, Cheng-Chien Wangb*, Chuh-Yung Chena a Department of Chemical Engineering, National Cheng Kung University, Taiwan, R.O.C. E-mail: [email protected] b

Department of Chemical and Materials Engineering, Southern Taiwan University of Science and Technology, Taiwan, R.O.C. E-mail: [email protected] (C. C. Wang) Abstract Shell crosslinked micelles were prepared via crosslinking reaction between double hydrophilic graft copolymers poly(N-isopropyl acrylamide)-co-poly(L-lysine) (PNIPAm-co-PLLys) micelles and natural crosslinking agent genipin. These shell crosslinked micelles possess spherical structures and the hydrodynamic radiuses are about 18.5 nm to 37.7 nm. Drug-loaded shell crosslinked micelles were also prepared as drug carriers. Model drug methotrexate (MTX) were loaded into polymeric micelles with different loading ratios (polymer / MTX = 10 / 0.5 and 10 / 1.0), then crosslinking agent genipin was added into micelle solution to form drug-loaded shell crosslinked micelles. Entrapment efficiency and drug loading content of the drug-loaded shell crosslinked micelles are about 12.66 wt% to 20.1 wt% and 0.84 wt% to 1.28 wt%, respectively. In-vitro drug release experiments of drug-loaded shell crosslinked micelles were carried out in pH 7.4 phosphate buffer solution at 37 oC. All of these samples possess burst release in initial 8 hours, and final accumulate MTX release amounts are about 71 % to 97 %. Keywords: shell crosslinked polymeric micelle; poly(L-lysine); drug delivery 1. Background/ Objectives and Goals

Amphiphilic block and graft copolymers simultaneously contain hydrophilic and hydrophobic segments, which could self-assemble to form nanoparticles, micelles, and vesicles due to the different solubility of these segments in aqueous solution at ambient temperature. However, the double hydrophilic block or graft copolymers also attracted great interest in recent decades1, which may form micelles as amphiphilic copolymers by varied temperature2-4, pH value3-6, ionic strength, complex with ionic compounds7, or conjugate with hydrophobic compounds8,9. These polymeric micelles usually possess hydrophobic core and hydrophilic shell, which could load hydrophobic drugs and have great potential in drug delivery system.

1315

In our previous study10, the double hydrophilic graft copolymer poly(N-isopropyl acrylamide)-co-poly(L-lysine) (PNIPAm-co-PLLys) were successfully synthesized, which possess temperature- and pH-response properties simultaneously. PNIPAm-co-PLLys could turn into amphiphilic graft copolymer and form uniform circular micelle-like structures in aqueous solution at 50 oC or at pH 12, but the micelle-like structures disappear at lower temperature and lower pH. In order to prepare stable micelles and apply the double hydrophilic graft copolymer PNIPAm-co-PLLys in drug release; the primary amine groups of side chain poly(L-lysine) (PLLys) were further crosslinked with crosslinking agent to form stable shell crosslinked micelles. The commonly used crosslinking agents have epoxy groups or aldehyde groups11-13, which were applied to crosslink with primary amine groups of polypeptide copolymers to prepare gel or shell crosslinked micelles. In which, glutaraldehyde is the famous crosslinking agent in biomaterial preparation and tissue fixing. However, the glutaraldehyde crosslinked biomaterials may possess toxicity, and cause the tissues sclerosis, fibrosis, and calcification14. Thus, Sung et al.14-22 chose the natural crosslinking agent, genipin, to substitute glutaraldehyde. Genipin is extracted from the gains of Gardenia jasminoides ELLIS, and could crosslink with primary amine group to obtain crosslinked products in navy blue. The biological toxicity of genipin is less than glutaraldehyde by 5,000 to 10,000 times, and genipin is thus more suitable using in biomaterials than glutaraldehyde. Genipin could apply to prepare hydrogels23,24, crosslinked micelles and particles17,25-27 expect for scaffolds of tissue engineering, which also possess high potentials in drug release. Therefore, genipin was best chosen as crosslinking agent to crosslink with double hydrophilic graft copolymer PNIPAm-co-PLLys. In this study, the double hydrophilic graft copolymers PNIPAm-co-PLLys with different length PLLys side chains were formed as a micellar structures at 50 oC, followed by crosslinking with genipin to obtain shell crosslinked micelles. For drug release application, the model drug MTX was preloaded into the polymeric micelles and crosslinked with genipin to obtain drug-loaded shell crosslinked micelles. Dynamic light scattering (DLS) and transmission electron microscopy (TEM) were used to analyze structures and size of shell crosslinked micelles. Then, in-vitro drug release experiments of drug-loaded shell crosslinked micelles were performed in pH 7.4 buffer solution at 37 oC. The drug delivery behavior, entrapment efficiency (EE), and drug loading content (DL) were analyzed by high-performance liquid chromatography (HPLC). 2. Methods

2.1 Materials N,N‘-Dimethyl formamide (DMF, anhydrous 99.8%, Aldrich Co. Ltd.), tetra-butylammonium hydrogen sulfate (TBA, >98%, Merck), acetonitrile (ACS grade, J.T. Baker), potassium

1316

dihydrogen phosphate (KH2PO4, 99%, SHOWA), sodium phosphate (Na2HPO4, 99.95%, Aldrich Co. Ltd.), and genipin (98%, CHALLENGE BIOPRODUCTS. Co., Ltd) were used as received. DMF, KH2PO4, Na2HPO4, and genipin were stored in a dry box before used. 2.2 Synthesis of the PNIPAm-co-PLLys graft copolymer The poly(N-isopropylacrylamide)-co-poly(Nε-benzyloxycarbonyl-L-lysine) (PNIPAm -co-PZLLys) graft copolymer was synthesized via free radical polymerization, followed by cleaving benzyloxycarbonyl groups (Z groups) to obtain the double-hydrophilic graft copolymer PNIPAm-co-PLLys. The detailed descriptions of the synthesis and characterization were reported in our previous study10. In this study, graft copolymers with different PLLys side chains (degree of polymerization, DP = 10, 15, 30) and nearly the same NIPAm units ratios (from 65 mol % to 76 mol %) were prepared. All of experimental results of the graft copolymers are tabulated in table S1. 2.3 Preparation of shell crosslinked micelles The polymer solutions and genipin solutions were prepared by dissolving PNIPAm-co-PLLys graft copolymers (3 mg) and genipin in pH 6 phosphate buffer solutions (3 mL / 1 mL), respectively. The both solutions were pre-heated in water-bath at 50 oC, then the genpin solution was dropped into the polymer solution and stirred at 50 oC for 8~12 hrs. After reaction completion, the solution was dialyzed against de-ionized water for 48 hrs (MWCO = 3,600) and stored in darkness. 2.4 Preparation of drug-loaded shell crosslinked micelles The graft copolymer PNIPAm-co-PLLys (12 mg) and methotrexate (MTX) (0.6 mg) were dissolved in 1.2 mL anhydrous DMF to prepare the polymer-drug solution in 10 mg/mL. Then the polymer-drug solution was dropped into pH 5 phosphate buffer solution at 50 oC and stirred for 30 minutes to obtain the polymer-drug micelle solution. The genipin solution was prepared by dissolving 6.5 mg genipin in pH 5 phosphate buffer solution (4 mL) and pre-heated to 50 oC, then which was dropped into the polymer-drug micelles solution at 50 oC. After stirred for 8 hrs, the solution was dialyzed against de-ionized water for 36 hrs (MWCO = 3,600) to eliminate free drug, then filtrated by filter (pore size: 0.45 μm). 2.5 In-vitro Drug Release of MTX from drug-loaded shell crosslinked micelles The above-mentioned solution (4 mL) was introduced into dialysis tube which was immersed in 20 mL pH 7.4 phosphate buffer solution. The drug release system was taken in shaking bath at 80 rpm and at 37 oC. At proper time interval, 3 mL outer solution was taken out and equivalent fresh buffer solution was replaced into the container. The drug release solution samples were characterized by HPLC, after the content of released drug trended to equilibrium, the drug-loaded polymer micelles solution was lyophilized and stored in dry box.

1317

2.6 Characterization

DLS measurements were carried out using an ALV/CGS-3 light-scattering spectrometer equipped with an ALV/LSE-5003 multiple-η digital correlator. A JDS-Uniphase solid-state He-Ne laser with an output power of ~22 mW and operating wavelength of 632.8 nm was used as the light source. The scattering angle was fixed at 90o. The sample solutions were filtrated and stored at room temperature overnight before DLS measured. Hitachi model HF-2000 field emission transmission electron microscopy (TEM) was used to observe the size and morphology of the graft copolymers micelles. The sample solution was dropped on a copper grid with carbon film and stained with 0.2 % uranyl acetate. High-performance liquid chromatography (HPLC, SHIMADZU) was performed at room temperature with 5 mM TBA and 0.02 N pH 7.4 phosphate buffer solutions / acetonitrile (82/18) as the mobile phase, and the flow rate was 1.0 mL/min. The eluted MTX solution was detected by SPD-10A VP UV-vis detector at 307 nm, and elution times were converted into concentrations with a calibration curve constructed from MTX dissolved in 0.01 N pH 7.4 phosphate buffer solutions. 3.

Results

3.1 Preparation of shell crosslinked micelles PNIPAm-co-PLLys may form micelle-like structures at 50 oC but destroy at lower temperature and lower pH, as reported by our previous study10. In order further to obtain stable micelles at room temperature and further applied in drug release, the amine group of graft copolymer side chain PLLys was crosslinked to form shell crosslinked micelles by genipin as crosslinking agent. Genipin may self-polymerize and precipitate in strongly basic solutions31. Thus the initial cross-linking reactions were proceeded in pH 6 phosphate buffer solution or pH 6 de-ionized water (adjusted with 0.01 N HCl). The colors of both solutions changed from colorless to navy blue without precipitation after crosslinked reaction, that reveal the amine groups of side chain PLLys were successfully crosslinked with genipin15. In order to prove whether the graft copolymers PNIPAm-co-PLLys were formed stable micelles at room temperature after cross-linking reaction in different solution, the result aqueous solutions were analyzed by DLS. Figure 1 shows the normalized intensity correlation function and hydrodynamic radius distribution of the graft copolymer L15-N65 cross-linking with genipin (L15-N65-G) in pH 6 buffer solution (L15-N65-G-b) and de-ionized water (L15-N65-G-w), respectively. The correlation function intensity of L15-N65-G-b is obviously higher than L15-N65-G-w, and the average hydrodynamic radius (Rh) of L15-N65-G-b is near 18.55 nm that might attribute to the aggregation of micelles-like structures. However, the average Rh of L15-N65-G-w is quite small that only possesses free chain vibration. The DLS result shows that amine groups of L15-N65 side chain PLLys may react with genipin in either solution; nevertheless which may

1318

not form stable micelles in pH 6 water. These results may attribute to the PLLys side chains of L15-N65 were only partial ionized in pH 6 water and difficult to form micelle-like structures at higher temperature. Thus the following crosslinked reactions were performed in phosphate buffer solution.

The graft copolymers L10-N70, L15-N65 and L30-N76, which possess different side chain length, were crosslinked with genipin in pH 6 phosphate buffer solution; named L10-N70-G, L15-N65-G and L30-N76-G, respectively. Figure 2 shows Rh distributions of L10-N70-G, L15-N65-G, and L30-N76-G. The results of DLS show all of these graft copolymers with different side chain length could cross-link with genipin and maintain the micelle structures at room temperature. The small and uniform shell crosslinked micelles with average diameters near 25 nm to 50 nm at room temperature were shown in Fig. 3.

1319

In order to investigate the stability of the micelles, the L15-N65-G shell crosslinked micelle aqueous solution was stored for three months and then analyzed by DLS. The PDI of L15-N65-G micelles is still low (PDI = 0.239 ± 0.029), and with average Rh near 24.81 ± 0.4 nm. The DLS results show that double hydrophilic graft copolymer PNIPAm-co-PLLys could crosslink with genipin to form uniform shell crosslinked micelles at 50 oC, and which may store at room temperature at least three months. All experimental results of the shell crosslinked polymeric micelles are tabulated in Tab. 1.

3.2 Preparation of drug-loaded shell crosslinked micelles The above experimental results show that graft copolymer PNIPAm-co-PLLys could crosslink with natural crosslinking agent genipin to form stable shell crosslinked micelles. Therefore, the hydrophobic anti-tumor drug MTX was loaded into the polymeric micelles, followed by crosslinking the shell of micelles to obtain drug-loaded shell crosslinked micelles, and further applied in in-vitro drug release. The anti-tumor drug MTX was preloaded into graft copolymers L15-N70 micelles with different polymer / drug feeding ratio (10 / 0.5, 10 / 1.0), then crosslinked by genipin to prepare drug-loaded shell crosslinked micelles, named L15-N70-Ml-G1 and L15-N70-Ml-G2, respectively. Figure 4 (a) shows Rh distribution of L15-N70-G, L15-N70-Ml-G1, and

1320

L15-N70-Ml-G2. The DLS results indicate the PDI of shell crosslinked micelles (L15-N70-G) was higher than drug-loaded shell crosslinked micelles (L15-N70-Ml-G1 and L15-N70-Ml-G2), and the Rh are nearly the same. Similarly, the graft copolymer L30-N76, which has longer PLLys side chains, was chosen to prepare drug-loaded shell crosslinked micelles with different polymer/drug feeding ratios (10 / 0.5, 10 /1.0), and analyzed by using DLS (Figure 4 (b)). All of experimental results were listed in Tab. 2. The results show the L30-N76-Ml-G1/2 have larger Rh and smaller PDI than L30-N76-G, and without aggregation. These results may due to the side chains PLLys of L15-N70 are shorter and have less space to load drug into micelles, thus the Rh of drug-loaded shell crosslinked micelles are similar to shell crosslinked micelles.

Moreover, L30-N76 has longer PLLys side chain which could form larger micelles after drug loading into polymeric micelles, and obtain more uniform shell crosslinked micelles, as shown in Fig. 5. Figure 5 shows the TEM image of drug-loaded shell crosslinked micelles L30-N76-Ml-G, which indicates the graft copolymer also could form uniform micelles with diameter about 30 nm after drug loaded into micelles and crosslinked by genipin. The micelle size in TEM image is smaller than in DLS measurement, due to the micelles in TEM image

1321

shrink after the solvent evaporates. The above results reveal the graft copolymer could form uniform and stable drug-loaded shell crosslinked micelles.

Afterwards the drug-loaded shell crosslinked micelles solution was lyophilized and obtained blue solids. The solids and DMF was mixed in sample vials and stirred in darkness for 24 hours, and then the above mixture was filtered by 0.45 μm filter. The shell crosslinked micelles were unable to dissolve in DMF, thus DMF was used to dissolve the MTX which was preloaded in micelles. Entrapment efficiency (EE(%) = Wload/Wadd × 100%) and drug loading content (DL(%) = Wload/Wtotal × 100%) were analyzed by HPLC. In which, Wload is weight of drug encapsulated in micelles, Wadd is weight of drug added as the micelles prepared, and Wtotal is total weight of the drug-loaded shell crosslinked micelles. All of the experimental results are tabulated in Tab. 3. The entrapment efficiency (EE) of the drug-loaded shell crosslinked micelles is about 12.66% to 20.1%, and which is lower with the polymer/drug loading ratio. Furthermore, the drug loading content (DL) is about 0.84% to 1.28% and is higher as the polymer / drug loading ratio is lower.

3.3 In-vitro Drug Release of MTX from drug-loaded shell crosslinked micelles In-vitro drug release was carried out in pH 7.4 buffer solution (0.02 N) at 37 oC. Figure 6 shows the drug release profile of MTX from drug-loaded shell crosslinked micelles with

1322

various polymer/drug loading ratio. The burst release amounts of L15-N70-Ml-G1 and L15-N70-Ml-G2 are about 51.47 ± 0.88 % and 42.80 ± 0.30 % in initial 8 hours (Figure 6 (a)); and which of L30-N76-Ml-G1 and L30-N76-Ml-G2 are about 24.71 ± 0.44 % and 25.45 ± 0.45 % in initial 6 hours (Figure 6 (b)), respectively. Moreover, the final MTX release of L15-N70-Ml-G and L30-N76-Ml-G reached to about 68 % ~ 97 % and about 71 % ~77 % around 48 hours, respectively. In addition, the graft copolymer L15-N65 was also used to fabricate drug-loaded shell crosslinked micelles L15-N65-Ml-G and then applied to drug release. The burst release of L15-N65-Ml-G1, and L15-N65-Ml-G2 are 36.46 ± 0.8 % and 51.71 ± 1.14 % in initial 8 hours, respectively, and the final MTX release reached to 85 % ~ 94 % around 48 hours. All of the drug release behaviors of drug-loaded shell crosslinked micelles are similar, and the greater parts of L15 series micelles are approximately total released (above 85 %). However, L30-N76-G drug-loaded shell crosslinked micelles have shorter burst release periods, larger burst release percentages, and less MTX release than L15 series. These results may due to the graft copolymer L30-N76-G has longer side chain PLLys (DP = 30), longer PLLys possess more amine groups, which could crosslink with genipin to form dense shell to retard serious drug release and reduce burst release periods.

3.4 Conclusion The double hydrophilic graft copolymer PNIPAm-co-PLLys with different length side chain (DP = 10, 15, 30) PLLys were successfully crosslinked with genipin and form stable shell crosslinked micelles. The average hydrodynamic radiuses of circular shell crosslinked micelles are about 18.5 nm to 37.7 nm, and the micelles could store at room temperature at least 3 months without serious aggregation and precipitation. The results of DLS and TEM show that MTX could load into polymeric micelles and crosslink with genipin, then obtains uniform and circular drug-loaded shell crosslinked micelles. All of the drug release behaviors of drug-loaded shell crosslinked micelles are similar, these samples have burst release and drug

1323

release period are about 48 hours. L15 series micelles reveal burst release in initial 8 hours, and greater parts of which have higher final MTX release percentage (above 85 %). The burst release period of L30 series shorten from 8 hours to 6 hours, and the final release MTX of L30 series micelles are about 71 % to 77 %. These experimental results show that drug-loaded shell crosslinked micelles were successfully prepared, and which could sustain MTX release for 48 hours. 3.3 Acknowledgments The authors would like to thank the Ministry of Economic Affairs of the Republic of China and the National Science Council of Taiwan for financially supporting this work under Contract No. TDPA 101-EC-17-A-08-S1-204, NSC 100-2221-E-006-056-MY3, NSC 102-2622-E-006-012-CC2, NSC 101-3113-E-024-001-CC2, and NSC 101-2120-M-006-009. 4. References

1. Helmut, C. (2001). Double-Hydrophilic Block Copolymers: Synthesis and Application as Novel Surfactants and Crystal Growth Modifiers. Macromolecular Rapid Communications, 22, 219-252. DOI: 10.1002/1521-3927 2. Cheng, C., Wei, H., Shi, B.-X., Cheng, H., Li, C., Gu, Z.-W., Cheng, S.-X., Zhang, X.-Z., Zhuo, R.-X. (2008). Biotinylated thermoresponsive micelle self-assembled from double-hydrophilic block copolymer for drug delivery and tumor target. Biomaterials, 29, 497-505. 3. Rao, J., Luo, Z., Ge, Z., Liu, H., Liu, S. (2007). ―Schizophrenic‖ Micellization Associated with Coil-to-Helix Transitions Based on Polypeptide Hybrid Double Hydrophilic Rod– Coil Diblock Copolymer. Biomacromoleculars. 8, 3871-3878. DOI: 10.1021/bm700830b 4. Ge, Z., Cai, Y., Yin, J., Zhu, Z., Rao, J., Liu, S. (2007). Synthesis and ‗Schizophrenic' Micellization of Double Hydrophilic AB4Miktoarm Star and AB Diblock Copolymers:  Structure and Kinetics of Micellization. Langmuir. 23, 1114-1122. DOI: 10.1021/la062719b 5. Qi Bo, Y. Z. (2006). Double-hydrophilic block copolymer for encapsulation and two-way pH change-induced release of metalloporphyrins. Journal of Polymer Science Part A: Polymer Chemistry. 44, 1734-1744. 6. Khousakoun, E., Gohy, J. F., Jerome, R. (2004). Self-association of double-hydrophilic copolymers of acrylic acid and poly(ethylene oxide) macromonomer. Polymer. 45, 8303-8310. 7. Chen, W., Chen, H. R., Hu, J. H., Yang, W. L., Wang, C. C. (2006). Synthesis and characterization of polyion complex micelles between poly(ethylene glycol)-grafted poly(aspartic acid) and cetyltrimethyl ammonium bromide. Colloids and Surfaces A: Physicochemical and Engineering Aspects. 278, 60-66.

1324

8. Tang, X., Pan, C.-Y. (2008). Double hydrophilic block copolymers PEO-b-PGA: Synthesis, application as potential drug carrier and drug release via pH-sensitive linkage. Journal of Biomedical Materials Research Part A. 86A, 428-438. 9. Dai, X.-H., Hong, C.-Y., Pan, C.-Y. (2012). pH-Responsive Double-Hydrophilic Block Copolymers: Synthesis and Drug Delivery Application. Macromolcular Chemistry and Physics. 213, 2192-2200. DOI: 10.1002/macp.201200324 10. Tu, Y.-L., Wang, C.-C., Chen, C.-Y. (2011). Synthesis and characterization of pH-sensitive and thermosensitive double hydrophilic graft copolymers. Journal of Polymer Science Part A: Polymer Chemistry. 49, 2866-2877. DOI: 10.1002/pola.24721 11. Li, L.-Y., He, W.-D., Li, J., Zhang, B.-Y., Pan, T.-T., Sun, X.-L., Ding, Z.-L. (2010). Shell-Cross-Linked Micelles from PNIPAM-b-(PLL)2 Y-Shaped Miktoarm Star Copolymer as Drug Carriers. Biomacromolcular. 11, 1882-1890. DOI: 10.1021/bm1004383 12. Ohkawa, K., Kitsuki, T., Amaike, M., Saitoh, H., Yamamoto, H. (1998). Biodegradation of ornithine-containing polylysine hydrogels. Biomaterials. 19, 1855-1860. 13. Rodríguez-Hernández, J., Babin, J., Zappone, B., Lecommandoux, S. (2005). Preparation of Shell Cross-Linked Nano-Objects from Hybrid-Peptide Block Copolymers. Biomacromolcular. 6, 2213-2220. DOI: 10.1021/bm0501499 14. Sung, H. W., Liang, I. L., Chen, C. N., Huang, R. N., Liang, H. F. (2001). Stability of a biological tissue fixed with a naturally occurring crosslinking agent (genipin). Journal of Biomedical Materials Research. 55, 538-546. 15. Sung, H.-W., Huang, D.-M., Chang, W.-H., Huang, R.-N., Hsu, J.-C. (1999) Evaluation of gelatin hydrogel crosslinked with various crosslinking agents as bioadhesives: In vitro study. Journal of Biomedical Materials Research. 46, 520-530. 16. Mi, F.-L., Sung, H.-W., Shyu, S.-S., Su, C.-C., Peng, C.-K. (2003) Synthesis and characterization of biodegradable TPP/genipin co-crosslinked chitosan gel beads. Polymer. 44, 6521-6530. doi:10.1016/S0032-3861(03)00620-7 17. Mi, F.-L., Sung, H.-W., Shyu, S.-S. (2002) Drug release from chitosan–alginate complex beads reinforced by a naturally occurring cross-linking agent. Carbohydrate Polymers. 48, 61-72. doi:10.1016/S0144-8617(01)00212-0 18. Sung, H.-W., Chang, Y., Chiu, C.-T., Chen, C.-N., Liang, H.-C. (1999). Crosslinking characteristics and mechanical properties of a bovine pericardium fixed with a naturally occurring crosslinking agent. Journal of Biomedical Materials Research. 47, 116-126. DOI: 10.1002/(SICI)1097-4636 19. Mi, F.-L., Tan, Y.-C., Liang, H.-C., Huang, R.-N., Sung, H.-W.(2002). In vitro evaluation of a chitosan membrane cross-linked with genipin. Journal of Biomaterial Science Polymer Edition, 12, 835-850.

1325

20. Mi, F.-L., Sung, H.-W., Shyu, S.-S. (2000) Synthesis and characterization of a novel chitosan-based network prepared using naturally occurring crosslinker. Journal of Polymer Science Part A: Polymer Chemistry. 38, 2804-2814. 21. Mi, F.-L., Tan, Y.-C., Liang, H.-F., Sung, H.-W. (2002). In vivo biocompatibility and degradability of a novel injectable-chitosan-based implant. Biomaterials. 23, 181-191. Doi:10.1016/S0142-9612(01)00094-1 22. Chen, S.-C., Wu, Y.-C., Mi, F.-L., Lin, Y.-H., Yu, L.-C., Sung, H.-W. (2004). A novel pH-sensitive hydrogel composed of N,O-carboxymethyl chitosan and alginate cross-linked by genipin for protein drug delivery. Journal of Controlled Release. 96, 285-300. 23. Moffat, K. L., Marra, K. G. (2004) Biodegradable poly(ethylene glycol) hydrogels crosslinked with genipin for tissue engineering applications. Journal of Biomedical Materials Research B Applied Biomaterial. 71, 181-187. 24. Stancu, I.-C.(2010) Gelatin hydrogels with PAMAM nanostructured surface and high density surface-localized amino groups. Reactives and Functional Polymer. 70, 314-324. 25. Huang, Y.-C., Arham, M., Jan, J.-S. (2011) Alkyl chain grafted poly(L-lysine): self-assembly and biomedical application as carriers. Soft Matter. 7(8), 3975-3983 26. Kamiński, K., Zazakowny, K., Szczubia ka, K., Nowakowska, M. (2008) pH-Sensitive Genipin-Cross-Linked Chitosan Microspheres For Heparin Removal. Biomacromolcular. 9, 3127-3132. DOI: 10.1021/bm800724q 27. Huang, K.-S., Lu, K., Yeh, C.-S., Chung, S.-R., Lin, C.-H., Yang, C.-H., Dong, Y.-S. (2009). Microfluidic controlling monodisperse microdroplet for 5-fluorouracil loaded genipin-gelatin microcapsules. Journal of Controlled Release. 137, 15-19. doi:10.1016/j.jconrel.2009.02.019

1326

ACEAIT-7327 Experimental Study on Hydraulic Characteristics of Detour Flood-Flow in a Blocked River and Flood Damage to House T. Okamotoa,, H. Takebayashib, M. Sanjouc, K. Todad, Y. Shibayamaa a, c, d Department of Civil Engineering , Kyoto university, Japan E-mail address: a [email protected], c [email protected], d [email protected] b

Disaster Prevention Research Institute, Kyoto University, Japan E-mail: [email protected]

Abstract A heavy rainstorm occurred in Uji city in August 2012. A woody debris jam blocked Shizugawa River and the detour flood-flow occurred around the bridge. Consequently, the high-speed flood-flow caused bed erosion in floodplain and the house along Shizugawa River was washed away. Therefore, it is important to know flood-flow characteristics in a blocked river (mainly due to a logjam) at bridges. A woody debris jam has been studied in river engineering. However, there is almost no detailed information about flood-flow characteristics and flood-damage to a house on floodplain in a blocked river.

Flow velocity in floodplain was measured by PIV techniques. The drag force exerted on the model house on the left bank and the right bank was measured using Force gage. The results revealed the occurrence condition of a detour flood-flow at a bridge. The results also indicate the significant effects of the floodplain shapes on the flood hazard area.

Keywords: Detour flood-flow around a bridge, blocked river, drag force, house in floodplain 1. Background A heavy rainstorm occurred in Uji city in August 2012. A woody debris jam blocked Shizugawa River and the detour flood-flow occurred around the bridge. Consequently, the high-speed flood-flow caused bed erosion in floodplain and the house along Shizugawa River was washed away. Therefore, it is important to know flood-flow characteristics in a blocked

1327

river (mainly due to a logjam) at bridges. A woody debris jam has been studied in river engineering. For example, Braudrick & Grant (2001) examined interactions between hydraulics, channel geometry, transport distance and deposition of floating wood in flume experiment. Bocchiola et al. (2006) investigated the motion of large woody debris in rivers as influenced by the presence of obstacles. Schmocker & Hager (2011) conducted flume experiments and evaluated the drift-blocking probability of both single logs and single root stocks at bridge decks. However, there is almost no detailed information about flood-flow characteristics and flood-damage to a house on floodplain in a blocked river. When a logjam blocks a river, flow depth increases upstream in a river and a detour flow occurs around a bridge. In August 2012, the detour flood-flow in Shizugawa River was concentrated on the left bank (the left bank height is lower than the right bank height) and the house in floodplain was washed away in a short time (Fig.1).

1328

2. Methods In the present study, flume experiments were conducted in a 10m long and 40cm wide glass-made flume (Fig.2). Floodplain models were placed on both sides of the channel. The right-bank height DR is constant in the streamwise direction ( DR =10cm). For CaseR10L8,

the left bank height is constant for

x  0 cm

( DL1 =10cm) and becomes smaller for

1329

x  0 cm

( DL 2 =8.0cm  DR ). A punctured metal sheet (15cm height, 20cm width, 3.0mm thickness and 3.0mm hole-diameter) was placed in front of the woody model bridge in Main-channel to mimic a blocked river (due to a logjam) at a bridge. The model bridge was mounted 4.0m from the upstream edge of the floodplain (x=0cm) and 9.0cm above the channel bottom. The bridge roadway is 20cm long, 2.0cm wide and 1.0cm thickness.

sheet (LLS) was projected into the water column horizontally, as shown in Fig.2. The illuminated flow pictures were taken by a high-speed CCD camera (1024  1024pixels) with 500Hz frame-rate and 60s sampling time. LLS on x-y vertical plane was projected vertically from water surface. The position of LLS on the right-bank and the left-bank are z/Bf=0.25, 3.75, respectively. The instantaneous velocity components (u~, v~) on x-y vertical plane (25cm  25cm) were calculated by the PIV algorithm. Drag force exerted on the model house in floodplain was measured by the Force gage with 10Hz sampling-rate and 180s sampling time. The Force gage has a resolution of 0.001N and can measure 2.5N for the maximum load. Table1 shows the hydraulic condition. Experiments were conducted on the basis of twenty flow scenarios, in which the floodplain shape, the blockage ratio and the flow discharge Q were changed. Before a flood (without the punctured metal sheet, see Fig.4(a)), the water in Main-channel does not overflow the banks. For Case R10L10 and Case R10L8, the bulk mean velocity is Um, main=38.0(cm/s) and the flow depth in Main-channel is Hm, main=6.8cm (see Fig.4(a)).

3. Results 3.1 Flood-flow characteristics of detour flood-flow around a bridge

1330

Figure 5 shows the contours of the time-averaged streamwise velocity U in x-z horizontal plane for CaseR10L8 (Left bank height is lower than Right bank height).

see Fig.4(a)). The x-axis and the z-axis are normalized by the main-channel width Bm. For CaseR10L10, the detour flows occur around the model bridge on both banks. The This is because the flood-flow depth Hf decreases downstream as described below. For CaseR10L8, the velocity values of the detour flow on the left bank become larger than those on the right bank. This indicates that the floodplain shape has the significant influence on the flood-flow path and the detour flow is concentrated on the left bank.

1331

Figure 6 shows the contours of the time-averaged streamwise velocity U in x-y vertical plane for CaseR10L8 (on the right bank and the left bank: z/Bm=0.125, 1.875). For CaseR10L8, the velocity difference between the left bank and the right bank is not significant upstream of the bridge. In contrast, downstream of the bridge

1332

To estimate the flood prone area in a blocked river, Figure 7 shows the contours of the flood-flow depth Hf in x-z plain for CaseR10L10 and CaseR10L8. Upstream of the direction and the difference between the left bank and the right bank cannot be observed.

The effect of the floodplain shape on the flood prone area is clearly observed for

the values of Hf

contrast, on the left bank are larger than those on the right bank for

1333

flood prone area increases. 3.2 Flood damage to a house on floodplain

1334

Figure 8 shows the time series of the drag force exerted on the model house for CaseR10L10 and CaseR10L8. The drag force values are normalized by the Wooden house demolitions criteria Fwd. A wooden house is broken when the drag force per meter reaches over 1.06 tf/m (Koshimura & Kayaba (2011)). By using scale ratios (1/80: model house), we calculated the Wooden house demolitions criteria Fwd. The streamwise distributions of the flood-flow depth Hf on the both banks are also indicated. At time t=0, the flood-flow reaches the house on the floodplain and drag force is exerted on the house. For CaseR10L10, the value of the drag force exerted on the house at x/Bm=0.25 increases as the flood-flow depth increases. The maximum value of the drag force Fmax decreases significantly at x/Bm=0.5 and increases again for x/Bm  0.75. This implies that the detour flood-flow flows around the side of the house at x/Bm=0.5 (see Fig.9: the same flood-flow behavior is observed at x/Bm=0.75 for Case R10L8). The drag force values exceed the Wooden house demolitions criteria (Fmax/Fwd>1.0) for x/Bm=0.0-1.5. This indicates that the wooden house is broken and washed away in the region of x/Bm=0.0-1.5 on the right bank and left bank. The drag force becomes negligibly small for x/Bm  2.0 and the house is at little risk of flood hazard in this region. For CaseR10L8, the effect of the floodplain shape on the flood hazard area is clearly observed. The maximum drag force values on the left bank are larger than those on the right bank. At x/Bm=1.0, the drag force value is almost F/Fwd= 5.0. The drag force values on the left bank are larger than the Wooden house demolitions criteria (F/Fwd>1.0) for x/Bm=1.0-3.0 (see Fig.9). The drag force values on the right bank exceed Fwd for x/Bm=0.0-1.5. This implies that the flood-flow is concentrated on the left bank and the flood hazard area increases. The drag force

1335

becomes

negligibly (Fig.7

(a)).

4. Conclusions

1336

In the present study, PIV measurements were conducted to examine the flood-flow characteristics in a blocked river. We changed the floodplain shapes for two patterns (CaseR10L10 (the bank height is constant for both banks) and CaseR10L8 (Left bank height is lower than Right bank height)). The results revealed the significant effect of the floodplain shapes on the flood-flow path. Main findings are as follows: 1. For CaseR10L10 and CaseR10L8, the detour flow occurs around the bridge on both banks. The results revealed that the flood velocity increases in the streamwise direction downstream of the bridge. For CaseR10L8, flood-flow is concentrated on the left bank and the velocity values of the detour flow on the left bank become larger than those on the right bank. All of the floodwater on the right bank flows into the

2. For CaseR10L10, the values of the drag force exerted on the house are larger than the Wooden house demolitions criteria for x/Bm=0.0-1.5. For CaseR10L8, the flood-flow is concentrated on the left bank and the flood hazard area increases. The drag force values exceed the Wooden house demolitions criteria for x/Bm=1.0-3.0.

Acknowledgement The present study was carried out under the financial support of the Research Project Grant-In-Aid for Scientific Research of Japanese Government (Kakenhi No.15K16311, Principle Investigator= T. Okamoto). The authors gratefully acknowledge this support. References Bocchila, D., Rulli, M.C. and Rosso, R. (2006): Transport of large woody debris in the presence of obstacles, Geomorphology, Vol.76, pp.166 – 178 Braudick, C.A., and Grant, G.E. (2001): Transport and deposition of large woody debris in streams: a flume experiment, Geomorphology, Vol.41, pp.263-283.

1337

Koshimura, S. and Kayaba, S. (2011): Tsunami fragility inferred from the 1993 Hokkaido Nansei-oki earthquake tsunami disaster, J. of Japan Association for Earthquake Engineering, 10(3). Schmocker, L., and Hager, W.H (2011): Probability of drift blockage at bridge decks, J. Hydraul. Eng., Vol.137, pp.470-479. Takemura, T. and Tanaka, N. (2007): Flow structures and drag characteristics of a colony-type emergent roughness model mounted on a flat plate in uniform flow, Fluid Dynamics Research, Vol.39 (Issues9-10), pp.694-710.

1338

Poster Sessions (6) Life Sciences (2) / Environmental Engineering / Geology, Earth and Environmental Sciences / Engineering Sciences (2) / Biomedical Engineering (2) Thursday, March 31, 2016

09:40-10:40

Foyer Area

GLSBE-36 Anti-Metastatic Potential of Myricetin against Human Cholangiocarcinoma Cells Laddawan Senggunprai︱Khon Kaen University Pattarapon Tuponchai︱Khon Kaen University Veerapol Kukongviriyapan︱Khon Kaen University Auemduan Prawan︱Khon Kaen University GLSBE-38 Hepatoprotective of Cymbopogon Citratus Stapf. Ethanolic Extract against Acetaminophen-Induced Hepatotoxicity in Rats Suphaket Saenthaweesuk︱Thammasat University Nuntiya Somparn︱Thammasat University Atcharaporn Thaeomor︱Suranaree University of Technology Janyuth Chaisakul︱Phramongkutklao College of Medicine GLSBE-46 Modulating the Structure and Function of an Aminoacyl-Trna Synthetase Cofactor by Biotinylation Chih-Yao Chang︱National Central University Arief Muammar︱National Central University Yu-Chen Chien︱National Central University Yu-Ren Chen︱National Central University Yu-Bao Wu︱National Central University Chien-Chia Wang︱National Central University

1339

GLSBE-47 Functional Substitution of an Indirect Pathway for tRNA Glutaminylation with a Direct Pathway Zong-You Hsieh︱National Central University Chia-Chen Hu︱National Central University Nancy Christina Batubara︱National Central University Meng-Hsuan Lee︱National Central University Hsiang-Hui Hung︱National Central University Chien-Chia Wang︱National Central University GLSBE-48 Deletion of a WHEP Domain Confers a Gain-of-Function Phenotype to a CMT-Associated Aminoacyl-tRNA Synthetase Hsiang-Ching Tseng︱National Central University Chia-Pei Chang︱National Central University I-An Chen︱National Central University Indra Lasmana Tarigan︱National Central University Wei-Cheng Chen︱National Central University Chien-Chia Wang︱National Central University GLSBE-53 Nuclear Factor-KappaB-Dependent Sestrin2 Induction Mediates the Antioxidant Effects of BDNF against Mitochondrial Inhibition in Rat Cortical Neurons Ding-I Yang︱National Yang-Ming University Chia-Lin Wu︱National Yang-Ming University GLSBE-64 Determination of Active Constituents and Antioxidative Activity in Schleichera Oleosa (Lour) Oken Buntarawan Turapra︱Khon Kaen University Chantana Bunyarat︱Khon Kaen University Supawadee Daodee︱Khon Kaen University GLSBE-23 Directed Expression of Cre Recombinase exclusively in Type II Alveolar Epithelial Cells of the Transgenic Mice Ning Hou︱Academy of Military Medical Sciences Xiao Yang︱Academy of Military Medical Sciences

1340

GLSBE-65 The Genetic Potential of Wheat Resistance to Heavy Metals Ravilya Alybayeva︱Kazakh National University named after Al-Farabi Valiria Kruzhaeva︱Kazakh National University named after Al-Farabi Ainash Alenova︱Kazakh National University named after Al-Farabi Indira Salmenova︱Kazakh National University named after Al-Farabi Akmaral Asylbekova︱Kazakh National University named after Al-Farabi A. Sadyrbaeva︱Kazakh National University named after Al-Farabi GLSBE-80 Screening on Plant Crude Extracts Containing CRTC3-Mediated Anti-Differentiation Activity of 3T3-L1 Adipocytes Han-Heom Na︱Kangwon National University Yoonsung Kang︱Institute for Diagnostic Markers Keun-Cheol Kim︱Kangwon National University GLSBE-81 Bacterial Fimbriae Regulated by Sugar Concentration in Klebsiella Pneumoniae Ting Sun︱Tzu Chi University Yu-Tze Horng︱Tzu Chi University Po-Chi Soo︱Tzu Chi University GLSBE-83 The Study of Interaction between KPN_00353 (EIIA Homolog) and Adenylate Cyclase (AC) in Klebsiella Pneumonia Ching-Yi Lin︱Tzu Chi University Yu-Tze Horng︱Tzu Chi University Po-Chi Soo︱Tzu Chi University GLSBE-87 Inhibition of ERK-Drp1-Mediated Mitochondrial Fragmentation by Peroxiredoxin 5 Prevents Amyloid-Beta Oligomer-Induced Cell Death Unbin Chae︱Kyungpook National University Dong-Seok Lee︱Kyungpook National University

1341

GLSBE-88 Iron Overload-Induced Calcineurin-Sensitive Signals Triggers Mitochondrial Fragmentation in HT-22 Hippocampal Neuron Cells Donggil Lee︱Kyungpook National University Dong-Seok Lee︱Kyungpook National University GLSBE-96 Oral Administration of Pineapple Glucosylceramide Improves Skin Barrier Function by Enhancement of Collagen Production in Hos:HR-1 Hairless Mice Sakiko Oka︱Kobe University Takashi Hashimoto︱Kobe University Nobuaki Ohto︱Maruzen Pharmaceuticals Co., Ltd Hiroshige Kuwahara︱Maruzen Pharmaceuticals Co., Ltd Masashi Mizuno︱Kobe University GLSBE-99 Identification of Surfactant and the Grease Degrading Bacteria Tsung-Han Chou︱Asia University Hebron C. Chang︱Asia University Su-Hua Huang︱Asia University GLSBE-101 Inhibitory Effect of Lactobacillus Plantarum 22A-3 Strain on type I Allergy Kou Fukuura︱Kobe University Takashi Hashimoto︱Kobe University Nobuaki Oto︱Maruzen Pharmaceuticals Co.,Ltd Taeko Hayashi︱Maruzen Pharmaceuticals Co.,Ltd GLSBE-119 Effects of Moxibustion on Weight Loss and Abdominal Obesity in Young Asian Females Ching Hsiu Hsieh︱Chang Gung University Of Science and Technology Chi-Chuan Tseng︱Chang Gung Memorial Hospital, Pei-Ying Chuang︱University of Texas

1342

GLSBE-122 The Antimicrobial Properties of Silver Nanoparticles in Bacillus Subtilis Yi-Huang Hsueh︱Yuan Ze University Kuen-Song Lin︱Yuan Ze University Wan-Ju Ke︱Chang Gung University Chien-Te Hsieh︱Yuan Ze University Shih-Tung Liu︱Chang Gung University ACEAIT-7552 Impacts of As, Cr, and Pb Treatments on Changes in the Concentrations of Selected Functional Ingredients in Go-deul-ppae-gi Plant Jin-Ho Jang︱Chonbuk National University Dong-Jin Kim︱Chonbuk National University Sung Eun Kim︱Chonbuk National University Young Mi Seol︱Chonbuk National University Jung-Hwon Park︱Chonbuk National University Jin-Ho Lee︱Chonbuk National University ACEAIT-7571 Photocatalytic Degradation of Trichloroethylene by Using TiO2 and ZnO Coated Optical Fiber Ching-Jung Chen︱National Pingtung University of Science and Technology Chen-Shiuan Huang︱National Pingtung University of Science and Technology Kuan-Chung Chen︱National Pingtung University of Science and Technology ISFAS-1483 Paleoenvironmental Changes of the Hupo Basin, Korean Inferred from Diatom Analysis of Piston Core (HB 13-2) Jihan Ko︱Kyungpook National University Seong-Joo Lee︱Kyungpook National University ISFAS-1512 Applications of Liquid Crystal (LC) Based Sensor for the Quantitative Identifications of H2O2 and NaClO in Aquatic Environments Ciao-Yan Chen︱National United University Lei-Xu Chen︱National United University Ruey-Fang Yu︱National United University

1343

ISFAS-1535 Control of the Optimal H2O2/Fe2+ Ratio in Photo-Fenton Wastewater Treatment Using DO and ORP Monitoring Jing-Lan Huang︱National United University Yi-Wen Hsu︱National United University Ruey-Fang Yu︱National United University ISFAS-1536 Monitoring of ORP, pH, DO in nZVI Wastewater Treatment Process for Chromium Cr(VI) Removal Jun-Xian Jiang︱National United University Hsien-Kuan Liao︱National United University Ruey-Fang Yu︱National United University ISFAS-1498 Adsorption of Soap in Biodiesel Using Magnesium Silicate Pornsawan Assawasaengrat︱King Mongkut's Institute of Technology Ladkrabang Pornrin Jintanavasan︱King Mongkut's Institute of Technology Ladkrabang

ISFAS-1560 Air Dehumidification System Using Propylene Glycol Solution Siripan Murathathunyaluk︱King Mongkut’s Institute of Technology Ladkrabang Nakarin Rueangwiriyanan︱King Mongkut’s Institute of Technology Ladkrabang Laksanaporn Poolnapol︱King Mongkut’s Institute of Technology Ladkrabang ISFAS-1534 An Ultra-Wideband (0.6-7.2 GHz) CMOS VCO Using a High Tunable Active Inductor Hsuan-Ling Kao︱Chang Gung University Cheng-Lin Cho︱National Tsing Hua University Li-Chun Chang︱Ming Chi University of Technology Ping-Che Lee︱Chang Gung University

1344

ACEAIT-7482 Development of a Monoclonal Antibody Specific to Envelope Domain III of Dengue Virus with Broad-Spectrum Detecting Capability to All Four Serotypes of Dengue Virus Yu Na Kim︱Chonbuk National University Sae-Hae Kim︱Chonbuk National University Kwang Yeop Jahng︱Chonbuk National University Sung Eun Kim︱Chonbuk National University Young Mi Seol︱Chonbuk National University Yong-Suk Jang︱Chonbuk National University

1345

GLSBE-36 Anti-Metastatic Potential of Myricetin against Human Cholangiocarcinoma Cells Laddawan Senggunprai, Pattarapon Tuponchai, Veerapol Kukongviriyapan, Auemduan Prawan Department of Pharmacology, Faculty of Medicine and Liver Fluke and Cholangiocarcinoma Research Center, Khon Kaen University, Thailand E-mail: [email protected], [email protected], [email protected], [email protected] Background Cholangiocarcinoma (CCA) is a highly metastatic malignancy with poor responses to traditional chemotherapeutic agents. The NF-B and STAT3 signaling cascades, which are the important pathways in transmitting inflammatory cytokine signals to nucleus, play critical roles in the progression and metastasis of CCA. The aim of this study was to determine the anti-metastatic activities of myricetin, a plant flavonoid, against human CCA cells. Methods The CCA KKU-100 cells were treated with pro-inflammatory cytokine mixture (interleukin-6 + interferon- + tumor necrosis factor-) for mimicking inflammatory condition. The effects of myricetin on cytokine-induced NF-B and STAT3 activation in CCA cells were determined by Western blot analysis. The real-time polymerase chain reaction was performed to determine mRNA expression. The migratory and invasive ability of KKU-100 cells were determined using wound healing assay and Transwell invasion assay, respectively. Results Cytokine mixture treatment could activate the NF-B and STAT3 signaling cascade in CCA cells demonstrated by increased levels of phosphorylated-NF-B and phosphorylated-STAT3 proteins. Treatment with myricetin suppressed the NF-B and STAT3 activation. The cytokine-mediated up-regulation of metastasis-associated genes which are downstream genes of NF-B and STAT3 including intercellular adhesion molecule-1 and matrix metalloproteinase-9 was also significantly inhibited by myricetin treatment. Consistently, myricetin could dramatically suppress the cytokine-induced migration and invasion of KKU-100 cells. These results demonstrate anti-metastasis effect of myricetin on CCA cells which is mediated partly through suppression of NF-B and STAT3 pathways. This compound would be potentially useful as a therapeutic agent against CCA. Keywords: Myricetin, cholangiocarcinoma, STAT3, NF-B, metastasis

1346

GLSBE-38 Hepatoprotective of Cymbopogon Citratus Stapf. Ethanolic Extract against Acetaminophen-Induced Hepatotoxicity in Rats Suphaket Saenthaweesuka, Nuntiya Somparn a, Atcharaporn Thaeomor b, Janyuth Chaisakulc a Department of Preclinical Science, Faculty of Medicine, Thammasat University, Rangsit Campus, Thailand. Email: [email protected], [email protected] b

School of Biology, Institute of Science, Suranaree University of Technology, Thailand Email:[email protected] c

Departments of Pharmacology, Phramongkutklao College of Medicine, Thailand. Email: [email protected]

1. Background/Objectives and Goals Cymbopogon citratus Stapf (CS), or lemongrass, is a widely used spice herb in tropical countries. In food and beverages, lemongrass is used as a flavoring agent whereas lemongrass tea is a refreshing beverage. The pharmacological properties of particular parts of CS have been reported including, free radical scavengers, antioxidants, anti-inflamatory and anti-mutagenicity. Although some studies have addressed the health benefits effects of CS, studies of prolonged consumption of CS extracts and hepatoprotective of CS extract against acetaminophen (APAP)-induced hepatotoxicity has not been investigated. Therefore, this study aims to investigate the effect of long term consumption and protective effect of CS extract on APAP- induced hepatotoxicity in rats. 2. Methods Thirty male Spraque Dawley rats were divided into 5 groups (n=6). Animal in group I reserved as normal control was administered vehicle solution (distilled water) throughout the duration of the experiment. Those in group II (APAP treated) received distilled water for 30 days followed by APAP (3 g/ kg BW, po) on day 31th. Animals in group III and IV were administered 500 and 1,000 mg/ kg BW/ day of the plant extract, respectively for 30 days. Group V and VI were administered 500 and 1000 mg/ kg BW/ day of the plant extract, respectively for 30 days followed by APAP (3 g/ kg BW/ day, po) on day 31th. All animal were sacrificed on day 32th and blood samples were collected for the determination of AST, ALT, BUN and Cr levels. Malondialdehyde (MDA) was measured in liver tissue sample. Moreover, the livers of rat were fixed and stained with hematoxylin-eosin (H&E) for examination of structure abnormality. The

1347

protocol for the present study was approved by the Animal Research Committee of Thammasat University, Thailand. All institutional and national guidelines for the care and use of laboratory animals were followed. 3. Expected Results/Conclusion/Contribution The treatment with CS extract for 30 days did not affect liver and kidney function as shown by normal serum AST, ALT, BUN and Cr levels. No structure abnormality of liver was observed in animals treated with CS for 30 day. For protective effects against APAP-induced hepatotoxicity, treatment with APAP alone showed significantly elevated of AST, ALT and MDA levels compared with normal control (P1000 µg/ml. The antioxidant activity from FRAP assay was shown as trolox equivalent which were 1172.91, 310.46µM trolox equivalent/g for peel and juice while the seed could not detect. It could be concluded that peel extract from the fruit of this plant showed the maximum content of these four active compounds which complied with its antioxidative activities. The maximum antioxidant activity from this part was shown by DPPH, ABTS and FRAP assay. This information could be used for further study about the product development of this kind of fruit and value added for consuming in the future. Keywords: Schleichera oleosa, antioxidative activity, DPPH, ABTS, FRAP

1358

1. Background/Objectives and Goals Nowadays increasing of natural product use especially natural antioxidant from various fruits and vegetable have received much attention. Many kinds of fruit have antioxidant components that are able to reduce oxidative stress. Active constituents in fruit that having antioxidant properties are carotenoids, phenolic compounds, flavonoid and anthocyanins. Oxidative stress seems to play a significant role in many human diseases, including cancers, alzheimer's disease and aging. Schleichera oleosa (Lour) Oken belongs to the family Sapindaceae. In Thailand, the fruit is called as Ta-Khro or Mark – Khro. It is one of the tropical fruit which generally found in Southeast Asia. It will bloom and the fruits can be harvested between March and July of every year. It is a medium to large size tree belongs with the height up to 30 m. The leaves are 20-40 cm long. The leaflets are 2 to 4 pairs, elliptic or elliptic-oblong, coriaceous and apex rounded. The flowers are minute, yellowish green (Palanuvej C. et al, 2008). Its fruit has round shape, size of 2-3 cm with thick skin. The seeds are brown, irregularly elliptic, slightly compressed, oily, enclosed in a succulent aril. Inside S. oleosa, there are 1 – 2 seeds and covered with light yellow, juicy sour fresh. Its bark, leaves, fruit and seed could be used in feeding animals. The wood is firewood and can be made as charcoal. The oil extracted from the seed, called ―kusum oil‖ is used for culinary and lighting purpose, cure of itching, acne, burns and hair dressing. It has been reported to possess antimicrobial, antioxidant, anticancer activity, and can be used for the production of biodiesel (Bhatia H. et al, 2013). Plants have been used as traditional medicine for the treatment of pain, inflammations, itching, wounds ulcers etc. Comparing to the other commercial fruits, the fruit of this plant is familiar to consume but some information about the value and biological activities are not completely known. Therefore, this study aimed to determine the amount of active constituents and the antioxidant activities from the fruit of this plant. This study was to determine the amount of active constituents (carotenoids, phenolic compounds, flavonoid and anthocyanins). DPPH, ABTS and FRAP assay were used for the analysis of antioxidative activity.

1359

2.

Methods

Materials and Chemicals M965+ Microplate Reader was product from metertech® , beta carotene, folin-Ciocaltea‘,s phenol regent, gallic acid, trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid), quercetin, ascorbic acid, sodium acetate, DPPH (2,2-diphenyl-1-picrylhydrazyl), ABTS (3-ethylbenzthiazoline-6-sulphonicacid) and TPTZ (2,4,6-tri(2-pyridyl)-s-triazine) were purchased from Sigma-Aldrich® . Sodium carbonate, aluminium chloride, potassium chloride, glucose, potassium persultate and ferrous sulfate were purchased from Univar® . Hydrochloric acid, methanol, ethanol and sulfuric acid were purchased from Merck® . Collection and Identification of fruit material S. oleosa fruits were collected during March - July 2014 from Ban Fang District, Khon Kaen, Thailand. The specimen of the plant material was identified and authenticated by Assist. Professor Dr.Prathan Luecha, Division of Pharmacognosy and Toxicology, Faculty of Pharmaceutical Sciences, Khon Kaen University, Khon Kaen, Thailand. Preparation of crude extract The S. oleosa was washed under running tap water to remove dirt. Each part of S. oleosa fruit were separated with a knife, grinder into pieces and dried at 50oC and was extracted with 95% ethanol for 7 day at room temperature. The extracts were filtered through whatman No.1 filter paper. The combined filtrates were evaporated using a vacuum rotary evaporator. Before the analysis, crude extracts were kept in bottles and stored in freezer at -20 oC. Determination of total carotenoid content Total carotenoids content was determined by the spectrophotometric method at 450 nm, which was adapted from Davies et al. (1976) using a beta carotene standard curve. Fruit extract (1 mg/ml) was prepared and dissolved in ethanol. The total carotenoids content was expressed

1360

based on beta carotene equivalents. Additional dilution was done if the absorbance value measured was over the linear range of the standard curve. Determination of total phenolic compounds content Total phenolic compounds content was determined by the folin – Ciocalteu,s phenol method, which was adapted from Follin et al. (1927) using gallic acid standard curve. Fruit extract (1 mg/ml, 500 µl) was mixes with 2.5 ml of folin–Ciocalteu reagent. After 5 min, then 2 ml of sodium carbonate solution was added to the mixture followed by the addition of 5 ml of deionized distilled water and mixed thoroughly. The mixture was kept in the dark for 2 h. The absorbance was measured at 760 nm using a spectrophotometer. The results were expressed in gallic acid equivalents. Additional dilution was done if the absorbance value measured was over the linear range of the standard curve. Determination of total flavonoid content Total flavonoid content was determined by aluminium chloride colorimetric assay, which was adapted from Woisky et al. (1998). Fruit extract (1 mg/ml, 500 µl) was mixes 0.4 ml of aluminium chloride and then 0.5 ml of sodium acetate solution was added to the mixture followed by the addition of 4 ml of deionized distilled water and mixed thoroughly. The mixture was kept at room temperature for 15 min. The absorbance was measured at 430 nm using a spectrophotometer. The results were expressed in quercetin equivalents. Additional dilution was done if the absorbance value measured was over the linear range of the standard curve. Determination of total anthocyanin content Total anthocyanin was determined by pH-differential, which was adapted from Lee et al. (2005). The absorbance, at 700 nm, of the fruit extract (1 mg/ml) was measured at both pH 1 and pH 4.5 using a spectrophotometer and 535 nm (to determine the anthocyanin content) against a blank of solvent mix used for extraction. The total anthocyanin content was calculated and expressed as mgcyanindin-3-glucoside/kg FW. In order to obtain the absorbance (A) related to the total anthocyanins the following equation was used:

Determination of antioxidant activity using a free radical scavenging assay (DPPH)

1361

The free radical scavenging (DPPH) method was used according to Ronald et al (2005). DPPH is one of the stable radicals. This assay is based on the measurement of the reducing ability of antioxidants toward DPPH. The DPPH reagent was prepared by DPPH with 100 ml ethanol and then stored at -20 oC prior to analysis. Fruit extract (10 mg/ml, 1,000 µl) and 100 µl of DPPH reagent were transferred into 96-well plate and keep it for 30 min. Then, the absorbance was measured at 517 nm by UV/Visible Spectrophotometer. The calibration curve was plotted at 10-50 µM concentrations of the standard antioxidants (Trolox). The inhibitory percentage of DPPH was calculated using

Where A is the absorbance Determination of antioxidant activity using ABTS radical cation decolorization assay (ABTS) The ABTS radical cation decolorization assay (ABTS) method was used according to Ronald et al (2005). ABTS is oxidized other oxidants to its radical cation. ABTS. +which is intensely colored and measured as the ability of test compounds to decrease the color reacting directly with the ABTS.+ radical. For ABTS assay, The ABTS reagent included ABTS.+ solution and potassium persulfate solution. The ABTS reagent was then prepared by mixing the two stock solutions and allowing them to react for 12 hours at room temperature in the dark. The solution was then diluted by mixing 1 ml ABTS.+ solution with 50 ml ethanol to obtain an absorbance of 0.70 + 0.02at 734 nm using the spectrophotometer. Fruit extract (10 mg/ml, 50 µl) and 100 µl of ABTS reagent were transferred into 96-well plate and keep it for 2 hr. Then, the absorbance was measured at 734 nm by UV/Visible Spectrophotometer . The calibration curve was plotted at 10 - 50 µM concentrations of the standard antioxidants (Trolox) The inhibitory percentage of ABTS was calculated using

Where A is the absorbance Determination of antioxidant activity using ferric reducing antioxidant power (FRAP) The ferric reducing antioxidant power (FRAP) method was used according to Ronald et al (2005). The fresh working solution was prepared by mixing 100 ml acetate buffer, 10 ml TPTZ solution, and 10 ml ferric chloride solution and then warmed at 37 oC before using. Fruit extract (10 mg/ml, 50 µl) and 150 µl of FRAP reagent were transferred into 96-well plate. Then, the absorbance was measured at 593 nm by UV/Visible Spectrophotometer. As a standard solution, it was used the Trolox synthetic antioxidant with a 10-50 µM concentration.

1362

Results are expressed in µM TE/g. Additional dilution was needed if the FRAP value measured was over the linear range of the standard curve. 3. Results and Discussion The present study about total active constituents and total antioxidative capacity of the peel, juice and seed of S. oleosa fruit. Fruit extract were prepared by using ethanol as the solvent. Total active constituents were shown in table 1. The peel contained the maximum content of carotenoid, phenolic, flavonoid and anthocyanin while the seed contained the minimum amount . Total carotenoid content was 343.57, 74.27 and 11.90 mg/g extract equivalents to beta carotene. Total phenolic content was 9.91, 6.01 and 4.06 mg/g extract as gallic acid equivalent. Total flavonoid content of 9.18, 5.07 and 2.19 mg/g extract as quercetin equivalent. Total anthocyanin content was 2.81, 0.23, 0.20 mg/g extract expressed as mg cyanindin-3-glucoside. The contents were shown in peel, juice and seed in respectively. Total antioxidant activity (Table 2) from each part of this fruit by DPPH assay were shown as IC50 which were 510.70, 779.72 and >1000 µg/ml for peel, juice and seed. From ABTS assay, IC50 were 784.74. 825.79 and >1000 µg/ml. The antioxidant activity from FRAP assay was shown as trolox equivalent which were 1172.91, 310.46µM trolox equivalent/g for peel and juice while the seed could not detect.

1363

It could be concluded that peel extract from the fruit of this plant showed the maximum content of these four active compounds which complied with its antioxidative activities. The maximum antioxidant activity from this part was shown by DPPH, ABTS and FRAP assay. This information could be used for further study about the product development of this kind of fruit and value added for consuming in the future. 4. Reference Bhatia, H., Kaur, J., Nandi, S., et al. (2013). A review on Schleichera oleosa Pharmacological and environmental aspects. Journal of pharmacy research; 6(1): 224 – 229. Earnest, E. (2015). Preparation of crude drugs from medicinal plants [Online]. [cited 2015 Aug 20]. Available from: http://www.academia.edu/ Kriengsak, T., Unaroj, B., Kevin, C., et al. (2006) Comparison of ABTS, DPPH, FRAP and ORAC assays for estimating antioxidant activity from guava fruit extracts. Journal of Food Composition and Analysis; 19: 669–675. Lucas, E. H., Kathryn, P., Lewis, R., et al. (1948). Preparation op crude plant extracts and their assay for presence of antibacterial substances. Journal of Food Science; (13)1: 82–88. Palanuvej, C., Vipunngeun, N. (2008). Fatty acid constituents of Schleichera oleosa (Lour) Oken seed oil. J Health Res; 22(4): 203-212. Srinivas, K., Celestin, R.V. (2013). Antioxidant activity of ethanolic extract of stem bank of Schleichera oleosa (Lour.) Oken. International Journal of Pharmacotherapy; 3(1):12-14. Meshram, N., Ojha, M., Singh, A., et al. (2015). Significance and Traditional Medicinal Properties of Schleichera oleosa. Asian Journal of Pharmaceutical Research; 5(1): 61 -64. Masisi, M., Beta, T., Moghadasian, M. (2015). Antioxidant properties of diverse cereal grains: A review on in vitro and in vivo studies. Food Chemistry; 196(-):90 -97. Thaipong, K., Boonprakob, U., Crosby, K. et al. (2006). Comparison of ABTS, DPPH, FRAP, and ORAC assays for estimating antioxidant activity from guava fruit extracts. Journal of Food Composition and Analysis ; 19(6-7): 669–675.

1364

GLSBE-23 Directed Expression of Cre Recombinase exclusively in Type II Alveolar Epithelial Cells of the Transgenic Mice

Ning Houa, Xiao Yangb Institute of Biotechnology, Academy of Military Medical Sciences, China. E-mail: a [email protected], b [email protected] 1. Background/ Objectives and Goals Type II alveolar epithelial (AT2) cells restrain the gas-liquid surface tension and participate in the innate immune response in lung[1-2]. Moreover, AT2 cells are proposed to contain tissue stem/progenitor cells, which can differentiate into AT1 cells after injury to maintain the adult lung epithelium[3-4]. Nevertheless, transgenic mouse model expressing Cre exclusively in AT2 cells are not yet available. 2. Methods Microinjection into the pronuclei of fertilized Kunming mouse oocytes was used to generate the transgenic mice. PCR analysis of genome DNA from different tissues was used to detect the location of the Cre-mediated recombination. Immunohistochemistry and flow cytometry analysis was used to confirm the cellular distribution of the Cre-mediated recombination in a certain tissue. 3. Expected Results/ Conclusion/ Contribution We developed a transgenic mouse strain in which the Cre recombinase expression was driven by the promoter of mouse SP-A gene. The SP-A-Cre-mediated recombination occurred only in lung but no other tissues and was expressed exclusively in AT2 cells, absent from the AT1 or bronchiolar Clara cells. We also witnessed an unambiguous expression of the SP-A-Cre in a subset of immunophenotypically-defined lung epithelial stem/progenitor cells. In Summary, we reported a transgenic mouse line in which the Cre recombinase was expressed exclusively in pulmonary AT2 cells, providing a valuable means for deciphering the physiological role of a gene specifically in AT2 cells and for precisely tracing the contribution of AT2 cells in various pathological conditions such as tumorigenesis and pulmonary fibrosis. Keywords: SP-A-Cre, Type II alveolar epithelial cells, transgenic mouse, conditional gene knockout 4. References

1365

[1] Weiss, D. J., Kolls, J. K., Ortiz, L. A., Panoskaltsis-Mortari, A., Prockop, D. J., 2008. Stem cells and cell therapies in lung biology and lung diseases. Proc Am Thorac Soc. 5, 637-67. [2] Osanai, K., Tsuchihara, C., Hatta, R., Oikawa, T., Tsuchihara, K., Iguchi, M., Seki, T., Takahashi, M., Huang, J., Toga, H., 2006. Pulmonary surfactant transport in alveolar type II cells. Respirology. 11 Suppl, S70-3. [3] Bertoncello, I., McQualter, J. L., 2013. Lung stem cells: do they exist? Respirology. 18, 587-95. [4] McQualter, J. L., Yuen, K., Williams, B., Bertoncello, I., 2010. Evidence of an epithelial stem/progenitor cell hierarchy in the adult mouse lung. Proc Natl Acad Sci U S A. 107, 1414-9.

1366

GLSBE-65 The Genetic Potential of Wheat Resistance to Heavy Metals Ravilya Alybayeva a, Valiria Kruzhaeva b, Ainash Alenova c, Indira Salmenova d, Akmaral Asylbekova i, A. Sadyrbaevaf a Kazakh National University named after Al-Farabi, Department of Biology and Biotechnology, Kazakhstan E-mail: [email protected] bcd

if

Kazakh National University named after Al-Farabi, Department Geography and Nature Management, Kazakhstan E-mail: [email protected]

Kazakh National University named after Al-Farabi, Department Geography and Nature Management, Kazakhstan E-mail: [email protected]

Abstract In Kazakhstan, the development of a powerful industry was based on its rich natural resources. However, the industrial centers are the areas of highest contamination by heavy metals. Sound environmental technologies are crucial to address heavy metal pollution problem. Development and use of plant varieties characterized by minimal accumulation of heavy metals can provide such environmental solution. The aim of this study was to identify wheat germplasm resistant to heavy metals (zinc, cadmium), which are priority pollutants in eastern Kazakhstan region, and identification of donors for breeding and promising forms of wheat that are resistance to heavy metals and destined for agricultural production. The objects of research are different genotypes of spring wheat from the collection of the East Kazakhstan Research Institute of Agriculture. Model experiments were performed on 10-days-old sprouts of different wheat genotypes, which were grown on nutrient mixtures containing Cd ions in concentration 40 mg/l (CdSO4) or Zn in concentration 400 mg/l (ZnSO4). In field experiment plants were grown on scientific test site, under natural environmental pollution, in the suburban area of Ust-Kamenogorsk city, East Kazakhstan region. The content of heavy metals in plants was determined by atomic absorption on the device AAnalyst 300 of "Perkin Elmer". Experiments and determination of physiological parameters were conducted by the method of field experiment. Based on the results received in 2015 the following conclusion was drawn: throughout the experiment Lyazzat and Samal spring wheat varieties could be considered as genotypes tolerant to zinc translocation to the above-ground parts of plants. In case of cadmium translocation most resistant wheat variety were Lyazzat, Omskaya-18 and Altai, while the most sensitive was Erytrospermum wheat variety. The membranes permeability for the electrolyte in a resistant variety Altai changed in a less

1367

degree than the more sensitive Erythrospermum. The result has showed that this fact in both cases is due to effect of cadmium as well as zinc. This fact testifies that stability of plants in general can be caused by resistance of their cellular membranes to the action of stressor. Research in model experiment has revealed tolerant and sensitive genotypes. The study of the accumulation of heavy metals in components of different genotypes of winter wheat in conditions of natural pollution has shown the genotypic differences in the accumulation of zinc and cadmium in plant seeds. The study of investigated heavy metals accumulation in the spring wheat seeds of East-Kazakhstan Agricultural Research Institute collection showed, that in terms of cadmium accumulation, the most resistant varieties of spring wheat are Ulbinka, Zaulbinka and Omskaya-18. Upon indicator of accumulation of zinc, the most resistant varieties are Samal and Kutulukskaya. Spring wheat varieties Ulbinka, Zaulbinka and Omskaya-18 can be recommended for further use in breeding for resistance to the accumulation of cadmium in the grain. Samal spring wheat varieties can be recommended for further use in breeding for resistance to the accumulation of zinc in the grain. The most harvest from plots has spring wheat varieties: Ulbinka-25, Glubochanka, Erythrospermum-616 and Altai. It is connected to a large number of grains per ear, high grain weight per ear, high productive tillering and good survival during the spring-summer growing season. Varieties Ulbinka-25 and Omskaya-18 can be recommended for cultivation on soils contaminated with cadmium, as these varieties of spring wheat accumulate least of cadmium in the seeds and at the same time are characterized by high yields and good survival during the spring-summer growing season in a polymetallic contamination of soil. Keywords: wheat, heavy metal resistant germplasm, membrane permeability, metal accumulation, physiological parameters, yield, promising forms. 1. Background/ Objectives and Goals In Kazakhstan, the development of a powerful industry was based on its rich natural resources. However, the industrial centers are the areas of highest contamination by heavy metals [1]. Sound environmental technologies are crucial to address heavy metal pollution. Development and use of plant varieties characterized by minimal accumulation of heavy metals can provide one such environmental solution. Plants, grown in contaminated soils, exhibit significant interspecies differences in responses to pollution. The available literature data showed that a significant positive correlation between the concentration of heavy metals in the grain and genotypes, indicating the possibility of breeding varieties with a low potential for accumulation of heavy metals [2]. Characterization of heavy metal accumulation in wheat and identification of germplasm resistance to influence of heavy metals is an important step towards the creation and use of wheat

1368

varieties resistant to heavy metal pollutions in agricultural production systems. This requires the study of the gene pool of cultivated plants and the selection of donors that accumulate minimum amount of heavy metals and varieties that are promising for cultivation in areas contaminated with heavy metals. The aim of this study was to identify wheat germplasm resistant to heavy metals (zinc and cadmium), which are important to eastern Kazakhstan region and identification of donors for breeding for promising forms of wheat that are resistance to heavy metals and destined for agricultural production. 2. Methods Different genotypes of wheat specific to East Kazakhstan were studied. Experiments were carried out using spring wheat varieties of East Kazakhstan Agricultural Research Institute collection: Altai, Ulbinka-25, Zaulbinka, Lyazzat, Glubochanka, Nargiz, Omskaya-18, Samal, Eritorospermum-606, Kutulukskaya. Model experiments were performed on 10-day-old seedlings of different wheat genotypes grown in a nutrient mixture containing Cd or Zn ions at a concentration of 40 and 400 mg/l respectively. Determined growth parameters, the index of tolerance and the content of cadmium and zinc in roots and aboveground parts of seedlings. Determination of membrane permeability for electrolytes was performed by conduct metric Dexter method [3]. Field studies carried out for the determination of physiological parameters. Plants were grown on scientific test site, under natural environmental pollution, in the suburban area of Ust-Kamenogorsk city, East Kazakhstan region. In field experiment the following physiological parameters were determined: phenological observations, yield, yield structure, the content of cadmium and zinc in seeds. Conducting the experiments and determination of physiological parameters were done by the method of field experiment [4]. Heavy metals (zinc and cadmium) in soil and plant samples were determined by atomic absorption with atomization in flame and graphite furnace on the device AAnalyst 300 of "Perkin Elmer" firm. Sample preparation was carried out using a heating unit «Hot Block» with the addition of concentrated nitric acid and hydrochloric acid at a temperature of 90 ± 5 ° C, in accordance with standard operating procedures. Weighed portion of sample was placed in disposable sample cups; 5 ml of 50% nitric acid and 0.5 ml of concentrated hydrochloric acid were added. Samples were mixed well to liquid clay condition, covered with a watch glass and placed in a heating block. The sample was heated to a temperature of 90 ± 5 ° C, and evaporated for 10-15 minutes without boiling. Then the sample was cooled, 5 ml concentrated HNO3 were added and heated again for 30 minutes. Content of capacitances were evaporated without boiling

1369

at a temperature of 90 ± 5 ° C for approximately up to 5ml during 2 hours, avoiding foaming. After that the sample was cooled and the volume was adjusted to 50 ml with deionized water. To calibrate of the device calibration blank have been used, consisting of deionized water and 1% HNO3 solution and standard samples of the company «High Purity". After calibration of devise readings of analyzed samples were taken. Accuracy of analysis performance was checked by the screening standard of the company «Merck». The content of heavy metal in the sample was calculated according to the formula:

3. Results The study of the regularities of accumulation and distribution of Zn and Cd in organs of spring wheat depending on genotypic differences in a model experiment was carried out. Investigation of the zinc content in the aerial parts of seedlings of different spring wheat genotypes showed that the smallest accumulation of zinc, at its content in the medium is 400 mg/l, is observed in sprouts of varieties Lyazzat and Samal. The highest zinc content in the aboveground organs observed for plant variety Ulbinka (Figure 1).

1370

Investigation of zinc accumulation in plant roots, when its content in the medium is 400 mg/L, showed that the least amount contained in the roots of seedlings of varieties Altai, Lyazzat, Glubochanka and Samal. The highest zinc content in the plants roots observed for variety Ulbinka-25 (Figure 2).

According to the study results, the effect of zinc on the growth parameters of the 10-day-old wheat seedlings and as well as the accumulation of zinc in the roots and above-ground parts of plants of different of spring wheat genotypes, we can distinguish the variety Altai as the genotype with the highest root resistance to zinc and varieties Lyazzat and Samal as resistant to the translocation of zinc in the aerial organs. Thus, the most resistant to zinc were spring wheat varieties Altai, Lyazzat and Samal, and the most sensitive - spring wheat genotype Ulbinka 25. Investigation of cadmium in the roots of spring wheats genotypes showed that at its dose of 40 mg/l in the growing medium the smallest amount of cadmium is accumulated by spring wheat varieties Kutulukskaya, Nargiz, Ulbinka-25 and Samal. The greatest amounts of cadmium accumulate in the plant roots of the spring wheat variety Erythrospermum. The remaining varieties occupy the intermediate position between them. Thus, the smallest accumulation of cadmium observed at the roots of spring wheat varieties Kutulukskaya, Nargiz, Ulbika-25 and Samal (Figure 3).

1371

The content of cadmium in the aerial parts of sprouts of studied spring wheat genotypes showed that at dose 40 mg/l of cadmium in the growing medium the least amount in above-ground organs accumulate sprouts of spring wheat varieties Lyazzat, Omskaya-18 and Altai. The largest amount of cadmium in the aerial organs, in these experimental conditions, accumulates seedlings of spring wheat variety Erythrospermum. The remaining varieties occupy an intermediate position between them (Figure 4). Thus, the greatest resistance to the intake of cadmium in the aerial organs showed winter wheat varieties Lyazzat, Omskaya-18 and Altay. According to the results of study of cadmium accumulation in the roots and above-ground parts of plants of different winter wheat genotypes varieties Kutulukskaya, Nargiz, Ulbinka-25 and Samal can be identified as the genotypes with the greatest root resistance to cadmium and varieties Lyazzat, Omskaya-18 and Altai as the genotypes with resistant to the translocation of cadmium in above-ground organs. Thus, the most resistant to cadmium were spring wheat varieties Nargiz, Ulbinka-25 and Altai, and the most sensitive - spring wheat genotype Erythrospermum.

1372

The testing was conducted of contrasting on resistance of spring wheat varieties on the permeability of cell membranes in the laboratory. Changes in membrane permeability are one of the manifestations of plant responses to external action. It is known that the output of the electrolyte is a function of membrane permeability, which in its turn characterizes the degree of its damage. Consequently, the genotypes are resistant to adverse environmental factors should have different physical-chemical characteristics of the membrane. Changes in membrane permeability under stress testify about restructuring of the membranes, which largely determines the potentially possible mechanisms of plants to withstand adverse environmental factors. As a result of changes in the permeability of cell membranes under the influence of various factors the amplified output of electrolytes from the cells. The intensity of the secretion of electrolytes depends on the varietal characteristics and age of the plant. Study of the effect of cadmium ions on cell membrane permeability of the aerial parts of plants of resistant - Altai and sensitive - Erythrospermum wheat genotypes demonstrated that the permeability of cell membranes for electrolytes under the action of high concentrations of cadmium (40 mg / L) increases (Table 1). In winter wheat variety Altai permeability of cell membranes of leaf tissues under the influence of cadmium is increased for 23% relatively to control. In spring wheat varieties Erythrospermum under the action of cadmium membrane permeability of leaf cells to the electrolyte rises for 75% (Table 1).

1373

It should be noted that the permeability of membranes for electrolytes in more resistant variety Altai was changing in a less degree in comparison to more sensitive variety Erythrospermum. Study of the effect of zinc ions in the cell membrane permeability of the aerial parts of plants of resistant - Altai and sensitive - Erythrospermum wheat genotypes demonstrated that the permeability of cell membranes for the electrolyte under the action of high concentrations of zinc (400 mg/l) increases (Table 2). In the variety of spring wheat Altai permeability of cell membranes of leaf tissues under the influence of zinc increased for 13% relatively to control. In the winter wheat variety Erythrospermum under the action of zinc leaf cell membrane permeability for the electrolytes was increasing for 45% (Table 2).

It should be noted that the permeability of membranes for the electrolytes in more resistant variety Altai varied in a less degree in comparison with more sensitive – Erythrospermum, as under the action of lead, as well as under the action of zinc. This fact indicates that, as a whole, resistance of the plants may be caused by the resistance of the cell membrane to the action of the stressor. The reason for the strongest suppression of the growth processes in spring wheat variety Erythrospermum, as well as the relative stability of the Altai variety to the toxic effects of heavy metals, may be different ability of cell membranes to withstand stress, in particular, to heavy metals. Also the genotypes of spring wheat were studied in the field of natural soil contamination with heavy metals. Our studies of the accumulation of heavy metals in the soil of root zone of different genotypes of winter wheat collection of East Kazakhstan Agricultural Research Institute under natural pollution showed that the content of heavy metals is different.

1374

Zinc content exceeds the maximum permissible concentration of the metal only in the root zone soil of varieties Nargis and Samal, in the zones of other varieties such excess is not observed (Figure 5). Thus, the wheat plants are not under stress from increased concentrations of zinc in the soil, except for varieties Samal and Nargiz, but in this case the excess is insignificant. Investigation of cadmium in the soil of the roots zone of different genotypes of winter wheat showed that the amount of the metal exceeds the maximum permissible concentration in nearly all varieties, except varieties Nargiz and Altay (Figure 6). The scope of excess of MPC is from 1.5 to 4.5 times. The highest excess recorded in the root zone soil of variety Omskaya-18. Thus, the wheat plants are stressed by high concentrations of cadmium in the soil, and this stress is significant. It should be noted also that the study of soil contamination has polymetallic character. The excess of the MPC also observed for metals such as copper and lead.

1375

In this connection, it was of interest to study the parameters of growth and development of these varieties to identify the tolerance to abiotic stress and agronomic resistance of varieties in the conditions of polymetallic stress. This may facilitate the identification of varieties that retain good productivity and at the same time little accumulates heavy metals in the grain. Investigation of zinc and cadmium in the seeds of winter wheat genotypes is the most important as a grain of wheat is used in the food industry. According to some researchers, the level of heavy metal cadmium in products is the problem of food security. Reduction of cadmium in the grain is one of the priorities of breeding programs [5]. Our research of accumulation of zinc in plant seeds of genotypes of East-Kazakhstan Agricultural Research Institute collection in a polymetallic soil contamination have shown that this metal accumulates in small quantities, and its content does not exceed the maximum permissible concentration in the seeds of almost all genotypes, except varieties Nargis, Altai and Glubochanka (Figure 7). This may be due to the fact that the content of the element in soil does not exceed MPC for soil or slightly exceeded. Despite this, in plant seeds of varieties Glubochanka and Altai accumulated amount of zinc is slightly above the MPC for wheat. The amount of zinc for 1.3 – 1.4 times higher than the maximum permissible concentration for grain in the seeds of varieties Altai and Zaulbinka and for 2.4 times greater than maximum permissible concentration in the seeds of variety Nargis.

The study of cadmium accumulation in plant seeds of genotypes of East-Kazakhstan Agricultural Research Institute collection in a polymetallic soil contamination have shown that this metal accumulates in small quantities, and its content is slightly higher than the MPC for the grain in case of spring wheat variety Erythrospermum (Figure 8).

1376

It should be noted that in the root zone soil content of cadmium is significant and exceeds MPC almost in all genotypes, except varieties Nargiz and Altay (Figure 6). Despite this, spring wheat seeds do not accumulate significant amounts of cadmium. The greatest amounts of cadmium accumulate seeds of variety Erythrospermum. Thus, the study of investigated heavy metals accumulation in the spring wheat seeds of East-Kazakhstan Agricultural Research Institute collection showed, that in terms of cadmium accumulation, the most resistant varieties of spring wheat are Ulbinka, Zaulbinka and Omskaya-18. In terms of accumulation of zinc, the most resistant varieties are Samal and Kutulukskaya. In respect to varieties Kutulukskaya, this fact can be seen as a disadvantage, since zinc content in soil, where the roots of this variety plants inhabit, is slight. Seeds of this wheat variety depleted of necessary for nutrition element. For variety Samal, this fact can be considered as an advantage, since the soil layer, where this variety roots inhabit, the zinc content heightened and in spite of this seeds does not accumulate significant amounts of zinc. Spring wheat varieties Ulbinka, Zaulbinka and Omskaya-18 can be recommended for further use in breeding for resistance to the accumulation of cadmium in the grain. Samal spring wheat varieties can be recommended for further use in breeding for resistance to the accumulation of zinc in the grain. Among the varieties of spring wheat of East-Kazakhstan Agricultural Research Institute collection the harvest from the plots is largest from the varieties of spring wheat Ulbinka-25, Glubochanka, Erythrospermum-616 and Altay (Figure 9).

1377

In the case of the variety Ulbinka-25 together with high yield, this variety is also characterized by high indicators, as grain weight of main ear and as well as grain weight of lateral stems. In varieties Erythrospermum and Altai, along with high yields also the highest values of grain weight of main ear and average indicators weight of grain of lateral stems revealed. Variety Glubochanka, although has an average the grain mass of the main ear and lateral stems, however, characterized by the highest rates of productive tillering, i.e. the number of productive lateral stems is big and probably therefore this variety has a higher overall crop yields. The lowest yields in conditions of soil contamination with heavy metals demonstrate varieties: Lyazzat, Zaulbinka, Samal, Nargiz. The average yield compared with other spring wheat genotypes detected in varieties Kutulukskaya and Omskaya-18 (Figure 40). Thus, the most harvest from plots have spring wheat varieties: Ulbinka-25, Glubochanka, Erythrospermum-616 and Altai. It is connected probably to a large number of grains per ear, high grain weight per ear, high productive tillering and good survival during the spring-summer growing season. Varieties Ulbinka-25 and Omskaya-18 can be recommended for cultivation on soils contaminated with cadmium, as these varieties of spring wheat accumulate least of all cadmium in the seeds and at the same time are characterized by high yields and good survival during the spring-summer growing season in a polymetallic contamination of soil. References [1] Alybaeva R.A. Evaluation of the ecological status of soils in Ust-Kamenogorsk, Bulletin of the KNU, environmental Series 2 (21) (2007) 40-44.

1378

[2] Wu Feibo, Zhang Guoping Genotypic variation in kernel heavy metal concentrations in barley and affected by soil factors, J.Plant Nutr. 25 (6) (2002) 1163-1173 [3] Koval S.F. Investigation of the cell membranes properties and the resistance of plants by elution of electrolytes // // News of the Sib. Branch of the Academy of Sciences of the USSR. Series of Biol.Sciences, 15 (3) (1974) 161-167. [4] Dospekhov B.A. (1985) Methodology of field experience - Moscow: Agro-industrial Publishing. [5] Knox R.E., Pozniak C.J., Clarke F.R., Clarke J.M., Houshmand S., and AK Singh RE. Chromosomal location of the cadmium uptake gene (Cdu1) in durum wheat, Genome 52 (9) (2009) 741 –747.

1379

GLSBE-80 Screening on Plant Crude Extracts Containing CRTC3-Mediated Anti-Differentiation Activity of 3T3-L1 Adipocytes Han-Heom Na 1, Yoonsung Kang 2, Keun-Cheol Kim1,* 1 Department of Biological Sciences, College of Natural Sciences, Kangwon National University, 2 Institute for Diagnostic Markers, Eudipia Inc, Republic of Korea * Corresponding author: [email protected] Background As the 3T3-L1 preadipocyte differentiate from fibroblasts to adipocytes in the addition of appropriate differentiative agents, it is useful to study regulatory mechanism of adipogenesis and obesity. It has been well-known that a growing number of plant crude extracts may regulate amount or velocity of adipogenesis and obesity. Recently, it was reported that CRTC3, CREB-regulated transcription coactivator 3, is associated with regulating adipocyte cell differentiation, suggesting that CRTC3 may play a role in the development of obesity. In this study, we are to explain our study on screening of plant crude extracts having CRTC3-mediated anti-differentiation activity of 3T3-L1 adipocytes. Materials and Methods 1. We purchased pGL3-CRTC3 p-Luc reporter plasmid from Origene Inc. and obtained stable cell lines via G418 selection after stable transfection. This clone was used for CRTC3 mediated activity screening on approximately 150 plant crude extracts. 2. We performed several rounds of MTT assay and luciferase assay to identify novel crude extracts having CRTC3-mediated anti-differentiation activity but no cytotoxicity of 3T3-L1 adipocytes. 3. Preadipocyte 3T3-L1 cells were maintained in DMEM containing 10% calf serum, but changed to differentiation inducing medium that contains FBS containing medium instead of CS. Adipocytic differentiation was calculated by oil Red O staining. 4. To investigate modulation of glycerol secretion by plant crude extracts, we treated crude extracts with forskolin, and measured the amount of glycerol release into the cell medium. Results and conclusion

1380

1. We performed stable transfection using pGL3-CRTC3-p-Luc reporter plasmid in 3T3-L1 cells. After G418 selection for 1 month, we selected two stable cell cells. These cells were used for selection of crude extracts that can modulate CRTC3-mediated anti-adipocyte differentiation. 2. Using MTT assay and luciferase assay, we screened approximately 150 plant crude extracts. Among them, five crude extracts (#9, #11, #23, #29, and #47) were identified as candidates having CRTC3 mediated anti-obesity activity. 3. Those five extracts were treated in differentiating 3T3-L1 cells, and also treated in post-differentiation cells. All five extracts shows anti-differentiation activity, and #23 extracts was especially shown as the best anti-differentiation activity. 4. Forskolin treatment induced glycerol secretion, glycerol release was no changed in crude extract treatment only, but it was increased by the combinatory treatment of forskolin and crude extracts. Interestingly, same combinatory treatment showed the decreased differentiation level using Oil Red O staining experiment, indicating that CRTC3-mediated anti-obesity activity is correlated with the regulation of lipid metabolism. Therefore, CRTC3 might be an important transcription factor that regulates excess energy production during lipid metabolism. In addition, 5 kinds of plant crude extracts are useful candidate materials for anti-obesity. Keywords: CRTC3, adipocyte differentiation, anti-obesity, crude extracts, glycerol release.

1381

GLSBE-81 Bacterial Fimbriae Regulated by Sugar Concentration in Klebsiella Pneumoniae Ting Suna, Yu-Tze Horngb, Po-Chi Sooc abc Department of Laboratory Medicine and Biotechnology, Tzu Chi University, Taiwan a E-mail: [email protected], [email protected], c [email protected] 1. Background Klebsiella pneumoniae is an opportunistic pathogen causing primarily urinary tract infections (UTIs), pneumonia, septicemia, and pyogenic liver abscess (PLA). Recently, it has been reported that K. pneumoniae is the most common pathogen causing pyogenic liver abscess (PLA) in diabetes patients in Taiwan. One of the virulence factors that contribute to pathogenic in Klebsiella is fimbriae. Two well-characterized fimbrial types have been described in K. pneumoniae, one is type 1 fimbriae (Fim) and the other is type 3 fimbriae (Mrk). Type 3 fimbriae are encoded by the mrk gene cluster, in which mrkA encodes the major subunit of fimbrial structure and mrkD encodes the fimbrial adhesion subunit. The type 3 fimbriae have been determined to play an important role in binding to eukaryotic cells surface or extracellular matrix proteins (ECMPs). The aim of this study is to identify the expression of fimbriae in Klebsiella pneumoniae in various sugar concentration. 2. Methods In order to check the production of fimbriae, the recombinant mrkA was constructed in the plasmid pET-30b::mrkA followed by expression and purification of His-tagged MrkA, which was used for anti-MrkA antibody production. The specificity of the antibody was confirmed by specificity test. Using anti-MrkA antibody, the expression of type 3 fimbriae in Klebsiella pneumoniae in various glucose concentration was detected. At the same time, we also changed the glucose concentration from high to low or low to high in two days to observe the effect of the alternation of glucose concentration on MrkA expression. 3. Results/Conclusion The result of enzyme digestion and electrophoretic analysis showed that the plasmid pET-30b::mrkA was confirmed. After the E. coli BL21(DE3)pLysS containing plasmid pET-30b::mrkA was induced by IPTG, using immobilized metal-ion affinity chromatography to purify His-tagged MrkA. The purified His-tagged MrkA was checked by SDS-PAGE. Furthermore, the specificity test demonstrated that anti-MrkA antibody had high specificity to recognize the native form MrkA. Using anti-MrkA antibody, the expression of type 3 fimbriae

1382

in K. pneumoniae was detected in glucose concentration test. Our result indicates that the expression of type 3 fimbriae in K. pneumoniae was regulated by glucose concentration and the effects of different type of sugars on the expression of type 3 fimbriae will be tested in the future work. Reference Podschun, R. and U. Ullmann (1998). "Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors." Clinical microbiology reviews 11(4): 589-603. Siu, L. K., K.-M. Yeh, J.-C. Lin, C.-P. Fung and F.-Y. Chang (2012). "Klebsiella pneumoniae liver abscess: a new invasive syndrome." The Lancet infectious diseases 12(11): 881-887. Tsai, S.-S., J.-C. Huang, S.-T. Chen, J.-H. Sun, C.-C. Wang, S.-F. Lin, B. Hsu, J.-D. Lin, S.-Y. Huang and Y.-Y. Huang (2010). "Characteristics of Klebsiella pneumoniae bacteremia in community-acquired and nosocomial infections in diabetic patients." Chang Gung Med J 33(5): 532-539. Thomsen, R. W., P. Jepsen and H. T. Sørensen (2007). "Diabetes mellitus and pyogenic liver abscess: risk and prognosis." Clinical infectious diseases 44(9): 1194-1201. Keywords: Klebsiella pneumoniae, pyogenic liver abscess (PLA), fimbriae, MrkA

1383

GLSBE-83 The Study of Interaction between KPN_00353 (EIIA Homolog) and Adenylate Cyclase (AC) in Klebsiella pneumonia

Ching-Yi Lina, Yu-Tze Horngb, Po-Chi Sooc abc Department of Laboratory Medicine and Biotechnology, Tzu Chi University, Taiwan a E-mail: [email protected], [email protected], c [email protected] 1. Background Most of bacteria use carbohydrates as a source of energy for growth, replication and metabolic processes. These carbohydrates could be transported through the phosphoenolpyrucate (PEP): carbohydrate phosphotransferase system (PTS) in bacteria. The PTS consists of the three parts, enzyme I (EI), histidine protein (HPr) and sugar-specific enzymes II (EII), which is comprised of EIIA, EIIB, and EIIC (EIID in some cases).When the bacteria uptake sugars through PTS, it will tranfer the phosphate group from PEP through EI, Hpr, EIIA, EIIB, EIIC, final to phosphorylate the uptaking sugars. In the previous studies, the phosphorylated EIIA can activate adenylate cyclase (AC), which can convert ATP into cAMP, well-known secondary messenger. In the previous study, we found the gene KPN_00353 encoding EIIA homolog. In this study, our aim is to figure out whether the KPN_00353will interact with AC or not. 2. Methods In order to construct fusion protein, glutathione S-transferase (GST)-tagged KPN_00353, the gene KPN_00353 was amplified and constructed in a vector containing GST gene to form the plasmid, pGEX-1::KPN_00353. Subsequently, the plasmid, pGEX-1::KPN_00353, was confirmed with the restriction enzymes BamHI and EcoRI. The fusion proteins, GST-tagged KPN_00353 and His-tagged AC were expressed from the E.coli carrying the recombinant plasmid, pGEX-1::KPN_00353 or pET::AC respectively. The cell lysates containing expressed proteins were analyzed by electrophoresis on 12%SDS-PAGE, followed by Coomassie blue staining. The interaction between GST-tagged KPN_00353 and His-tagged AC were examined by protein pull-down assay. We used the crude extract of recombinant E.coli containing pGEX-1 or pGEX-1::KPN_00353 respectively mixed with the crude extract of E.coli expressing His-tagged AC, followed by capture using glutathione sepharose beads. The proteins bound on the glutathione sepharose beads were detected by Western blotting using anti-His antibody. 3.

Results/Conclusion

1384

First, we confirmed the plasmid construction of GST-KPN_00353 by enzyme digestion and electrophoretic analysis. Then, we ran the SDS-PAGE to know the protein expressions of His-tagged AC (His-AC) and GST-tagged KPN_00353 (GST-353), respectively. By protein pull-down assay, His-tagged AC bound to GST-tagged KPN_00353, not GST alone. It is suggested that adenylate cyclase (AC) could interact with KPN_00353 (EIIA homolog). In order to realize the effect of KPN_00353 on the enzyme activity of AC, the cAMP levels in KPN_00353 mutant strain, KPN_00353-overexpressed strain, wild-type strain and wild-type strain containing vector control will be examined in the future study. 4. Reference Deutscher J, Francke C, & Postma PW (2006). How phosphotransferase system-related protein phosphorylation regulates carbohydrate metabolism in bacteria. Microbiol Mol Biol Rev 70: 939-1031. Gorke B, & Stulke J (2008). Carbon catabolite repression in bacteria: many ways to make the most out of nutrients. Nature reviews Microbiology 6: 613-624. McDonough KA, & Rodriguez A (2012). The myriad roles of cyclic AMP in microbial pathogens: from signal to sword. Nature reviews Microbiology 10: 27-38. Keywords: phosphotransferase system (PTS)、adenylate cyclase (AC)、cAMP

1385

GLSBE-87 Inhibition of ERK-Drp1-Mediated Mitochondrial Fragmentation by Peroxiredoxin 5 Prevents Amyloid-Beta Oligomer-Induced Cell Death Unbin Chae1, *, Dong-Seok Lee1, ** 1 School of Life Sciences, BK21 Plus KNU Creative BioResearch Group, Kyungpook National University, Republic of Korea E-mail: [email protected] * These authors contributed equally to work Running title: The protective role of Prx5 on AβO toxicity Keywords: Alzheimer's disease, amyloid beta, peroxiredoxin 5, oxidative stress, mitochondrial fragmentation, ERK, Drp1 Background: AβO impairs mitochondrial dynamics and induces cell death in neuronal cells Results: Peroxiredoxin 5 reduces AβO-induced ROS level, apoptotic cell death, and ERK-Drp1-mediated mitochondrial fragmentation. Conclusion: Peroxiredoxin 5 ameliorates AβO-induced cell death by decrease of mitochondrial fragmentation. Significance: Peroxiredoxin 5 may be used as therapeutic target of Alzheimer‘s disease. Abstract Alzheimer's disease (AD), a neurodegenerative disorder, is caused by amyloid-beta oligomers (AβOs). AβOs induce cell death by triggering excessive reactive oxygen species (ROS) generation and mitochondrial dysfunction. A recent study showed that AβO-induced ROS generation is associated with extracellular signal-regulated kinase (ERK) – dynamin related protein 1 (Drp1) - mediated mitochondrial fission. ROS are regulated by antioxidant enzymes, especially peroxiredoxins (Prxs) that scavenge H2O2. These enzymes inhibit neuronal cell death induced by various neurotoxic reagents. However, it is unclear whether Prx5, which is specifically expressed in neuronal cells, protects these cells from AβO-induced damage. In this study, we found that Prx5 expression was upregulated by AβO-induced oxidative stress and that Prx5 decreased ERK–Drp1-mediated mitochondrial fragmentation and apoptosis of HT-22 neuronal cells. Prx5 expression was affected by AβO, and reduction of ROS levels by N-acetyl-L-cysteine decreased AβO-induced Prx5 expression. Prx5 overexpression inhibited oxidative stress and apoptotic cell death but Prx5 knockdown did not. In addition, Prx5 overexpression ameliorated ERK–Drp1-mediated mitochondrial fragmentation but Prx5 knockdown did not. These results indicated that Prx5 expression mediated by AβO-induced oxidative stress plays a key role in inhibiting both ERK–Drp1-induced mitochondrial

1386

fragmentation and neuronal cell death by regulating ROS levels. Thus, Prx5 may be a new therapeutic agent for treating AD. Introduction Alzheimer‘s disease (AD) is the most common neurodegenerative disorder and is characterized by the progressive impairment of memory and cognition. Typical pathological hallmarks of AD include accumulation of abnormally folded proteins such as extracellular amyloid-beta (Aβ) plaques within neurons (1,2). Aβ induces neuronal dysfunctioning and cell death in AD. The most toxic form of Aβ is Aβ42, which forms different-sized self-aggregates and amyloid fibrils (3). Aβ oligomers (AβOs) are toxic to neurons and lead to their synaptic degeneration, which involves mitochondria. Aβ accumulates in the mitochondria of brains cells of patients and transgenic mice with AD (4,5). Accumulation of Aβ in mitochondria disrupts mitochondrial functions such as electron transport chain and calcium storage and mitochondrial dynamics (6,7). However, the specific molecular mechanism underlying the disruption of mitochondrial functions by AβOs remains unclear. Mitochondria are essential intracellular energy-generating organelles with various cellular functions. Neurons greatly depend on mitochondria for energy supply (8). Mitochondria are highly dynamic organelles that frequently fuse and divide (referred to as fusion and fission, respectively). Fusion and fission of mitochondria are important for many biological processes, including homeostasis of mitochondrial biogenesis, turnover, subcellular distribution, cell division, and apoptosis (9). Fusion and fission of mitochondria allow the formation of interconnected mitochondrial networks and fragmented mitochondria, respectively (10). Balanced mitochondrial dynamics is important for sustaining normal mitochondrial function (10). Therefore, it is important that therapeutic strategies for treating neurodegenerative diseases should exert protective effects on mitochondria. Mounting evidences suggest that Aβ increases the fission and decreases the fusion of mitochondria, thus promoting mitochondrial fragmentation and neuronal cell death (4,11,12). AβOs reduce mitochondrial length, which is suggested to increase mitochondrial fragmentation (11,13). In mammalian cells, mitochondrial fission is controlled by dynamin-related protein 1 (Drp1) and mitochondrial fusion is controlled by mitofusin 1 and 2 and optic atrophy protein 1 (10). Patients with AD show an abnormal balance in the levels of mitochondrial dynamics-related proteins (14). In addition, phosphorylation of Drp1 is increased in the brains of patients with AD (11). However, the mechanism underlying AβO-induced mitochondrial fragmentation is not understood. Increasing evidences suggest that mitochondrial dysfunction is an early event and a prominent feature of AD progression. The brains of patients with AD show several molecular

1387

abnormalities, including oxidative stress and defective energy metabolism (15,16). Oxidative stress precedes Aβ accumulation in the brains of mouse models of AD and other cellular models (17). Excessive oxidative stress results in immoderate mitochondrial fragmentation (18) and sustained activation of extracellular signal-regulated kinase (ERK, also known as p42/44 MAPK) (19). Several recent studies suggest that fibrillary Aβ induces ERK activation, thus leading to progressive neuronal cell death (20,21). Excessive mitochondrial fragmentation also induces apoptosis (18). However, the correlation between mitochondrial fragmentation and ERK is unclear. Oxidative stress occurs when the production of reactive oxygen species (ROS) surpasses the antioxidant activity of cellular and mitochondrial antioxidant enzymes. Peroxiredoxins (Prxs) are a family of antioxidant enzymes that scavenge H2O2 and regulate the transduction of various cellular signals through intracellular oxidative signaling pathways (22,23). Prxs eliminate free radicals to protect various cells, including neurons (24). Recombinant Prx5 exerts protective effects against excitotoxic brain lesions in newborn mice (23,25). Immunolabeling analysis performed in our previous study indicated that Prx5 was present in neurons and was evenly dispersed in the CA3 and CA2 regions of the hippocampus (26). However, the role of Prx5 in AD has been rarely studied to date. In the present study, we investigated that the role of Prx5 against AβO-induced toxicity that triggered apoptotic cell death through oxidative stress and mitochondrial fragmentation. Further, we confirmed whether Prx5 decreased mitochondrial fragmentation by regulating AβO-induced ERK activation. Our results showed that Prx5 may be a potential inhibitor of AβO-induced mitochondrial fragmentation and may exert protective effects against AβO-induced neuronal apoptosis. Experimental Procedures Materials N-Acetyl-L-cysteine (NAC) and dimethyl sulfoxide (DMSO) were obtained from Sigma-Aldrich (MO, St. Louis, USA). U0126 was obtained from Santa Cruz Biotechnology Inc. (TX, Dallas, USA). Oligomerization of Aβ42 Aβ1–42 was obtained from AnyGen (Jangseong, Korea) and was dissolved in DMSO to obtain a concentration of 5 mM. The dissolved Aβ1–42 was sonicated for 5 min and was diluted in Opti-MEM (Gibco, USA) to obtain a final concentration of 100 μM, which was stored at 37°C until further use. Cell culture and treatment

1388

HT-22 cells, which were derived from immortalized primary mouse hippocampal neuron cells (27), were maintained at 37°C in DMEM (Welgene, Daegu, Korea) supplemented with 10% fetal bovine serum (Gibco), 100 U/ml penicillin and 100 μg/ml streptomycin (Welgene), and 2 mM glutamine in an incubator (SANYO, Sakata, Japan) with a humidified atmosphere of 5% CO2. The cells were pretreated with NAC and U0126 for 30 min and were incubated with 5 μM AβO for 6, 12, and 24 h. Plasmid construction DsRed2-Mito gene encoding Discosoma red fluorescent protein (DsRed2) fused with a mitochondrial targeting sequence from subunit VIII of human cytochrome c oxidase was obtained from pDsRed2-Mito (Clontech, CA, USA). DsRed2-Mito and mouse Prx5 were amplified by performing PCR with LA Taq polymerase (TaKaRa, Shiga, Japan). These genes were cloned into a gateway entry vector pCR8/GW/TOPO (Invitrogen, CA, USA) to generate expression clones by performing LR recombination between the entry vector and gateway destination vectors pLenti6.3/V5-DEST and pLenti7.3/V5-DEST (Invitrogen). These 2 vectors have a C-terminal V5 epitope that aids in detecting recombinant proteins during immunoblotting analysis (28). The sequences of the constructed vectors were confirmed by performing restriction mapping and DNA sequencing. RNA interference assay HT-22 cells were grown to 50% confluency and were transfected with 10 pmol of siRNAs against Prx5 (siPrx5; Bioneer, Daejeon, Korea) by using Lipofectamine RNAiMAX (Invitrogen). The sequences of the siRNAs are as follows: Prx5 sense siRNA (5ʹ-GUCUGAGCGUUAAUGACGU-3ʹ) and Prx5 antisense siRNA (5ʹ-ACGUCAUUAACGCUCAGAC-3ʹ). Transfection and selection of stably expressing cells In all, 1 × 105 cells were seeded in 6-well plates. After 24 h, the cells were transfected with 2 μg of pLenti6.3-DsRed2-Mito and pLenti6.3-Prx5 by using Effectene (Qiagen, CA, USA), according to the manufacturer's instructions. After 24 h, the transfected cells were selected using 8 μg/ml blasticidin (Invitrogen). Western blotting Protein lysates were prepared using ice-cold PRO-PREP protein extraction solution (iNtRON Biotechnology Inc, Seongnam, Korea). Protein quantification was performed using Infinite F50 microplate reader (TECAN, Switzerland). In all, 20–50 μg of the protein lysates were separated on 10%–15% SDS-polyacrylamide gels. The proteins were transferred onto nitrocellulose membranes (Pall Corporation, NY, USA). The membranes were blocked with 5% skimmed milk (BD Biosciences, CA, USA) and were incubated overnight with primary antibodies

1389

against β-actin, Drp1 (Santa Cruz Biotechnology Inc.), phosphorylated Drp1 (Ser616), caspase-3, p44/42 MAPK, phosphorylated p44/42 MAPK (Cell Signaling Biotechnology, MA, USA), Prx5, cytochrome c (AbFrontier, Seoul, Korea), V5 epitope (Invitrogen), and Aβ1–16 (6E10; COVANCE, NJ, USA) at 4°C. The membranes were then washed 5 times with 10 mM Tris–HCl (pH 7.5) containing 150 mM NaCl and 0.1% Tween-20 (TBS-T) and were incubated with horseradish peroxidase-conjugated goat anti-rabbit and anti-mouse antibodies (Thermo Scientific, IL, USA) for 2 h at room temperature. After removing excess secondary antibodies, the membranes were washed 6 times with TBS-T. Specific binding was detected using Clarity Western ECL Substrate (Bio-Rad, CA, USA), according to the manufacturer's instructions. Measurement of intracellular ROS levels HT-22 cells treated with 5 μM Aβ1–42 were harvested by trypsinization (0.05% trypsin–EDTA; Welgene). For measuring intracellular ROS levels, the harvested cells were washed with phosphate-buffered saline (PBS, pH 7.4) and were incubated with 2.5 μM CM-H2DCFDA (Invitrogen) for 15 min at 37°C. The cells were then washed twice with PBS and were analyzed by flow cytometry (FACSCalibur; BD Biosciences). Analysis of cell viability Cell viability was assessed by performing 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay. Mock vectorand Prx5-expressing HT-22 cells (density, 5 × 103 cells for each cell type) were cultured in 96-well plates and were incubated for 24 h before treatment. The cells were then incubated with 5 μM AβO. Culture medium was removed carefully and was replaced with 0.5 mg/ml MTT solution dissolved in phenol red-free DMEM, and the cells were incubated for 1 h at 37°C. The medium was removed, and 100 μl DMSO was added in each well to dissolve formazan crystals. Absorbance was measured at 550 nm by using the Infinite F50 micro plate reader. Imaging and measurement of mitochondrial length DsRed2-Mito-expressing and DsRed2-Mito-/Prx5-coexpressing HT-22 cells were seeded on 0.1% poly-D-lysine-coated round coverslips (diameter, 24 mm; Marienfeld, Germany) and were incubated for 24 h. The cells were then incubated with 5 μM AβO for 24 h, washed twice with PBS, and fixed with 4% paraformaldehyde in PBS for 1 h. After washing 3 times with PBS, the coverslips were mounted on slides by using VECTASHIELD mounting medium (VECTOR Laboratories, CA, USA). Images were obtained using LSM-710 confocal microscope (Carl Zeiss, Oberkochen, Germany) equipped with a plan apochromatic 100×/1.40 oil DIC M27 objective lens. Images were processed using Zeiss LSM Image examiner, ZEN 2009 light edition (Carl Zeiss). Mitochondrial length was measured using ImageJ software (NIH, MD, USA) and was calculated using more than 50 mitochondria per cell for 20 cells.

1390

Mitochondria were divided into different categories based on their length, i.e., less than 1, 1–3, and more than 3 μm. Statistical analysis Data are represented as mean ± standard error of the mean (SEM) of ≥3 independent experiments. Statistically significant differences were determined using 1-way ANOVA with GraphPad Prism 5 (San Diego, CA, USA). A p-value of pH4 >>pH 10. It was thought that there were more hydroxyl radicals generated on the surface of TiO2 as the pH increased. But the electrostatic interaction affects the adsorption of TCE on ZnO which resulted in higher removal percentage in neutral and acidic solutions. Keywords: photocatalysis, optical fiber, trichloroethylene, titanium dioxide, zinc oxide

1431

ISFAS-1483 Paleoenvironmental Changes of the Hupo Basin, Korean Inferred from Diatom Analysis of Piston Core (HB 13-2) Jihan Koa,, Seong-Joo Leeb * a Department of Geology, Kyungpook National University, South Korea E-mail: [email protected] b

Department of Geology, Kyungpook National University, South Korea E-mail: [email protected]

Abstract A great number of diatoms were collected from a 3.23 m piston core (HB13-2) from the Hupo Basin, Korea. Bottom age of the core sediment yielded 6 kr in age, and a total of 112 diatom species belonging 52 genera were identified. The most dominant species is Paralia sulcata (11~27%), occurring throughout the whole core with some other dominant species (e.g., Actinoptychus senarius, Thalassiosira eccentrica, and Thalassionema nitzschioides). Freshwater species (e.g., Pinnularia viridis) were also obtained sporadically. The diatom assemblage were characterized by neritic to oceanic species and by abundant occurrence of warm species. Analysis of Td values indicates seawater temperature of Hupo Basin were fluctuated since 6 kr. Ratio of warm species (e.g., Shionodiscus oestrupii) and cold species (Thalassiosira trifulta and Thalassiothrix longissima) is also correlated with a Td curve. Such a flucatuation may have been resulted from the north-southward movement of water masses boundary where cold and warm currents meet.

1432

ISFAS-1512 Applications of Liquid Crystal (LC) Based Sensor for the Quantitative Identifications of H2O2 and NaClO in Aquatic Environments Ciao-Yan Chen, Lei-Xu Chen, Ruey-Fang Yu* Department of Safety, Health and Environmental Engineering, National United University, Taiwan, R. O. C. E-mail: [email protected]; [email protected] Abstract Recently, the Liquid Crystal (LC) was innovatively used as a bio-sensor or sensor (Liquid-Crystal-Based sensor) in some critical researches due to the specific optical and electrical characteristics of the LC, especially in the files of bio-chemical researches. The LC bio-sensors were investigated to detect or monitor the lipase activity, organophosphates, nitrogen dioxide, urea, l-tyrosine, glucose, cholesterol, acetylcholinesterase; some modified LC sensors were also investigated to detect the Volatile Organic Compounds (VOCs), ex. glutaraldehyde. The investigations on LC bio-sensor/sensor were innovative, high potential and valuable for bio-chemical researches, and were a high interesting and innovative study for many research areas. However, the applications of LC bio-sensor/sensor to identify the substances in aquatic environments are very rare and even can no find in the literatures. The present study was investigated to develop a LC sensor for quantitative identification of different concentrations of hydrogen peroxide (H2O2) and sodium hypochlorite (NaClO). The basic type of nematic LC (4-cyano-4_-pentylbiphenyl, 5CB) was used to construct the LC sensor in this study. DMOAP (N, N-dimethyl-N -octadecyl-3 -aminopropyl-trimethoxysilyl chloride) was used as the homeotropic alignment, then, the AEAPS (N-(2-aminoethyl)-3-(trimethoxysilyl)propylamine) was coated on the DMOAP layer as a reactive layer in the LC sensor. The color and brightness of the LC sensor were changed when the H2O2 and NaClO solutions were dropped in the sensor due to the liquid crystal molecules inside the sensor were disturbed by the dropping of the H2O2 and NaClO. Higher concentrations of the dropping H2O2 and NaClO made longer durations for the disturbances of the liquid crystal molecules inside the sensor. By analyzing the changes in brightness of LC sensor using image analysis technique, linear relationships were found between the concentrations of H2O2 and NaClO with the durations of the LC disturbance. As a result, the concentrations of H2O2 and NaClO can be identified according the changes of brightness of LC sensor. By comparing to traditional analyses, this proposed LC sensor provides an innovative, rapid, economy application for the quantitative identifications of the H2O2 and NaClO in aquatic environments.

1433

Keywords: Liquid crystals, sensor, hydrogen peroxide, sodium hypochlorite, aquatic environment

1434

ISFAS-1535 Control of the Optimal H2O2/Fe2+ Ratio in Photo-Fenton Wastewater Treatment Using DO and ORP Monitoring Jing-Lan Huang, Yi-Wen Hsu, Ruey-Fang Yu* Department of Safety, Health and Environmental Engineering, National United University, Taiwan, R. O. C. E-mail: [email protected]; [email protected] Abstract The Fenton series processes were regarded as most effective and popular processes in the advanced oxidation processes (AOPs), however, high operating costs and large sludge productions are the typical problems of the Fenton series processes, which direct correlating to the controlled dosages. On the other hand, to maintain an appropriate H2O2/Fe+2 ratio is also critical for reducing operating costs and enhancing the Fenton process performances. However, the optimal H2O2/Fe+2 ratio may various according to different treated wastewaters, Fenton processes, and Fe+2 and H2O2 dosages, and it was rarely to be discussed in the literatures. In this study, a continuous flow Photo-Fenton wastewater treatment system was used to treat the textile wastewater, the on-line monitoring Dissolved Oxygen (DO) and Oxidation Reduction Potential (ORP) was used to automatically adjust the optimum H2O2/Fe2+ ratios for dosages and sludge reductions. Experimental results have shown that the optimal H2O2/Fe+2 ratio can be achieved when DO values in Fenton reactor was controlled between 2.0 to 6.0 mg/L. In these control conditions, higher COD removals were performed, and both of the residual H2O2 and Fe2+ were maintained in relatively lower concentrations in the effluents, indicating that the appropriate H2O2/Fe2+ ratios were controlled. No significant COD removal was found in the Photo-Fenton system when a higher DO was controlled (ex. 12, 20, 30 mg/L) in the Fenton reactor. However, higher residual H2O2 and lower Fe2+ were found in the effluents, indicating an over-dosage of H2O2 in the reactor. On the other hand, the ORP value was controlled as the DO was less than 0.0 in the reactor. By comparing to the studies with DO controlled between 2.0 to 6.0 mg/L, the COD removal efficiencies reduced significantly in the ORP controlled studies. Significant residual Fe2+ and lower H2O2 concentrations were also found in the effluents, indicating the insufficiency of the H2O2 dosage. This finding is critical for the control of the optimal H2O2/Fe+2 ratios in the Fenton processes, resulting in significant dosages savings. Keywords: Photo-Fenton process, Dissolved Oxygen (DO), Oxidation Reduction Potential (ORP), H2O2/Fe+2 ratio, monitoring and control

1435

ISFAS-1536 Monitoring of ORP, pH, DO in nZVI Wastewater Treatment Process for Chromium Cr(VI) Removal Jun-Xian Jiang, Hsien-Kuan Liao, Ruey-Fang Yu* Department of Safety, Health and Environmental Engineering, National United University, Taiwan, R. O. C. E-mail: [email protected]; [email protected] Abstract The nano-Zero Valent Iron (nZVI) can be used to treat various pollutants in industrial wastewater, especially for the chromium Cr(VI) removal. Typically, The removal efficiency of Cr(VI) by the nZVI particles was controlled depending on the nZVI dosages. However, the reaction activities of nZVI were seriously affected by the surface characteristics, particle size and specific surface area etc., of the nZVI particles caused by different syntheses processes. Therefore, to control the nZVI process only based on the nZVI dosage may not reliable. In this study, the nZVI particles were synthesized in the laboratory using the sodium borohydride reduction method, however, different dropping flowrates of sodium borohydride solution (NaBH4) were controlled from 3.0 to 8.0 mL/min. According to the image from the scanning electron microscope (SEM), these nZVI particles prepared from different dropping flowrates of NaBH4 presented different particle size and specific surface area. A smallest particle size with an average diameter of 65.87 nm, and a largest surface area of 145.420 m2/g were found for the nZVI particles synthesized under dropping flowrate of 7.0 mL/min of NaBH4. On the other hand, a largest particle size (118.0 nm) and smallest surface area (69.4 m2/g) were also found for the nZVI particles synthesized under dropping flowrate of 3.0 mL/min of NaBH4. A nZVI batch reactor was used to remove chromium Cr(VI) with a retention time of 90 minutes. The nZVI particles synthesized under dropping flowrate of 7.0 mL/min (smallest particle size of 65.87 nm) performed the highest chromium removal efficiency. For an initial concentration of 40 mg/L, the removal efficiency of chromium can reach to 98.0 % with the nZVI dose of 1.067 g/L for the nZVI particles synthesized under dropping flowrate of 7.0 mL/min. However, only 60.3% of chromium was removed for the nZVI particles synthesized under dropping flowrate of 3.0 mL/min under the same operational condition. Similar removal efficiencies around 50.0% of chromium were also found for the nZVI particles synthesized under dropping flowrate of 7.0 mL/min with a dose of 0.4 g/L, and the nZVI particles synthesized under dropping flowrate of 3.0 mL/min with the dose of 0.67 g/L, respectively. This finding indicates that to control the nZVI process only based on the nZVI dosage is not reliable. Therefore, the monitoring of pH, ORP, and DO was used to evaluate the Cr(VI) removal efficiencies in this study, and the BPN (Back-Propagation Neural)

1436

model was used to construct the prediction model. The BPN models presented very precise prediction results of Cr(VI) removals by the nZVI particles with the correlation coefficients (R2) of 0.94 to 0.97. As a result, the monitoring pH, OPR, and DO have good capabilities to evaluate the Cr(VI) removal efficiencies, which have higher potential than dosage for the control of nZVI process for Cr(VI) removal in industrial wastewaters. Keywords: Dissolved Oxygen (DO), Oxidation Reduction Potential (ORP), nano zero valent iron (nZVI), chromium Cr(VI) removal, monitoring and control

1437

ISFAS-1498 Adsorption of Soap in Biodiesel Using Magnesium Silicate Pornsawan Assawasaengrat, Pornrin Jintanavasan Department of Chemical Engineering, Faculty of Engineering, King Mongkut's Institute of Technology Ladkrabang, Thailand Email: [email protected] Abstract Adsorption of soap in biodiesel using magnesium silicate was studied. The magnesium silicate adsorbent was prepared by the precipitation of magnesium sulfate and sodium metasilicate. The molar ratios of SiO2/MgO in synthetic magnesium silicate,i.e.,0.31:1, 0.76:1 and 2.04:1 were obtained by varying concentrations of Na2SiO3:MgSO4. The adsorption efficiency of synthesized magnesium silicate in biodiesel containing different amount of SiO2/MgO molar ratios by using 3 g of adsorbents for 150 g of biodiesel at 60˚C for 2 h was studied. The results showed that the magnesium silicate at 2.04:1 SiO2/MgO molar ratio showed the highest efficiency in adsorption of soap at 77.11%. While the adsorption capacity was 1.37 mg/g of soap. The Langmuir and Freundlich adsorpyion models were applied to describe the equilibrium isotherms. Based on R2, Freundlich model was more promising to represent the adsorption data than Langmuir model. Therefore the experimental data could be better described by Freundlich isotherm. Keywords: Adsorption, soap, magnesium silicate, biodiesel Introducation Biodiesel is an alternative and renewable fuel. It presents outstanding characteristics as non-toxicity, absence of sulphur, high energy content and biodegradability[1]. Transesterification is a chemical reaction relating triglycerides and alcohol using homogeneous or heterogeneous substances as catalyst to yield alkyl esters (biodiesel) and glycerol as by-product. The methyl ester has contaminants frequently that have effects on the quality of biodiesel. Glycerol is a potential cause of biodiesel instability. During transesterification, reaction conditions could result in the formation of oxidizes byproducts that cause discoloration, especially when glycerol is exposed to either high temperature or alkaline pH. Glycerol can oxidize or thermally degrade to form diols or acids that can contribute to biodiesel instability[2]. The water content in biodiesel causes corrosion of the automobile system (such as fuel tubes and injector) or might react with glycerides to form soap and glycerine as by-products[3]. In addition, soap and free fatty acids cause the deterioration of

1438

engine components. Conventionally wet washing is the most employed technique to remove contaminants such as soap, catalyst, glycerine and residual alcohol from biodiesel. However, the major disadvantages in the use of water to purify biodiesel are emulsion formation (preventing the separation of the esters), formation of free fatty acids and soap, wastewater formation; which all increase in cost and production time[4]. Adsorption technique was introduced to substitute wet washing to remove biodiesel contaminants. The use of adsorbents turns the process more rapid and there is not a formation of aqueous residues. The adsorbent such as magnesium silicate has the potential of selectively adsorbing hydrophilic materials. Magnesium silicate surface consists of partially hydrophobic and partially hydrophilic sites. The hydrophobic part includes siloxane (≡Si−O−Si≡) groups and the hydrophilic portion contains isolated hydroxy groups (−Mg−OH), individual silanol groups (≡Si−OH) and hydrogen bonds formed due to close vicinity of hydroxy groups linked to neighboring silicon atoms. In this paper, the adsorption of soap in biodiesel was studied by using magnesium silicate as an adsorbent Experimental Synthesis of magnesium silicate Three different concentration ratios of sodium silicate and magnesium sulphate (VI) were chosen for magnesium silicate synthesis. The concentration (Molar) ratios of Na2SiO4: MgSO4 were 0.5:0.5, 0.5:1.5 and 3.0: 0.5. The sodium silicate was dropped into the magnesium sulphate (VI) solution with stirring speed of 200 rpm at room temperature. Centrifugation at 2000 rpm was used to separate the precipitate and the liquid phase was removed by filtration. The precipitate was dried at 100°C for 24 h, then mashed and sieved. Characterization of magnesium silicate Chemical characterization of representative sample was undertaken by X-ray fluorescence spectrophotometer (XRFs) for major element detection. Analysis by X-ray diffractometer (XRD) revealed major crystalline phases in magnesium silicate samples. In order to determine magnesium silicate adsorptive potentials, their BET specific surface area and pore diameter were estimated using multipoint BET surface area, and meso pore volumes was determined by using BJH equation. Adsorption capacity of synthetic magnesium silicate 3g of adsorbent were added into 150 g of biodiesel to study the soap adsorption. The shaker machine (ZHWY-100B, ZHICHENG) was set at 60°C, 200 rpm and 2 h. After shaking, the adsorbent was removed and purified biodiesel was analysed for the soap. In order to determine the effectiveness of purification, the soap was measured before and after processing. Soap level was determined using AOCS Method Cc 17‐79. Adsorption efficiency was defined as percentage of fraction between adsorbed and initial soap.

1439

Adsorption isotherm The adsorption isotherm was determined by varying the amount of magnesium silicate ( 2.04:1 molar ratio). the amount of absorbents were varied at 1, 2, 3, 4, 5 and 6 g per 150 g of biodesel, adsorbed at 60 °C for 5 hours. Langmuir isotherm was plotted as mg of soap adsorbed per gram of adsorbent vs. residual concentration of soap in solution. An ln/ln plot was also prepared to observe the Freundlich constant, K and R2 was determined by using linear Regression. Results and discussion Synthesis and characterization The synthesized magnesium silicate was a fine white powder. The chemical composition of the synthesized magnesium silicates was analysed by X-ray fluorescence (XRF). Table 1 exhibited the chemical compositions of magnesium silicate which were synthesized by varying Na2SiO4: MgSO4 concentration ratios. The molar ratio of SiO2: MgO in synthetic magnesium silicate was 0.36:1, 0.76:1 and 2.04:1 according to the Na2SiO4: MgSO4 concentration ratio of 0.5:1.5, 0.5:0.5 and 3.0:0.5, respectively. Na2SiO4: MgSO4 concentration ratio of 3.0:0.5 exhibited highest molar ratio of SiO2: MgO, while the concentration ratio of 0.5:1.5 showed lowest molar ratio of SiO2: MgO. Thus, the molar ratio of synthetic magnesium silicate was in agreement with the used concentration ratio.

The average surface area, pore volume and pore size of magnesium silicate were characterized by BET method. The result is showed in the Table 2. The magnesium silicate has surface area between 15-58 m2/g and average pore size is between 50-105 A which could described that synthesized magnesium silicate was mesoporous. Higher SiO2: MgO molar ratio provided larger pore size. From the results, the 2.04:1 molar ratio of SiO2: MgO has the biggest pore size diameter, 105 A.

1440

Crystalline structure of magnesium silicate was characterized by X-ray diffraction technique. XRD patterns of magnesium silicate were demonstrated in Figure 1. The structure of the synthesized magnesium silicate was amorphous as there were no sharp peaks appearing at any 2. The XRD patterns of the samples were in an agreement with commercial magnesium silicate [5].

Adsorption results The adsorption efficiency of synthesized magnesium silicate in biodiesel containing different amount of SiO2/MgO molar ratios by using 3 g of adsorbents for 150 g of biodiesel at 60˚C for 2 h was showed in Figure 2. From the rerult, the adsorption efficiency of SiO2/MgO molar ratios 0.36:1, 0.76:1 and 2.04:1 of adsorbents was 23.83%, 46.17% and 77.11%, respectively. The adsorption capacity of aforementioned ratios was 0.37, 0.87 and 1.37 mg/g. respectively. The magnesium silicate with molar ratio of 2.04:1 has the highest adsorption capacity. This is because 2.04:1 molecular ratios of SiO2/MgO of magnesium silicate had high silica content than other concentration ratios of magnesium silicate which related to XRF result. It means that

1441

silica content in magnesium silicate effect on efficiency of soap adsorption. Physically, BET results presented the largest pore size, 105 A, at 2.04:1 SiO2/MgO molecular ratio of magnesium silicate which influenced the adsorption efficiency.

Magnesium silicate with SiO2/MgO ratio of 2.04 which had the highest adsorbed free fatty acid was used to study the adsorption time to reach equilibria. The adsorption results are shown in Figure 3. It could be concluded that the equilibrium could be reached within 4 hr.

The adsorption isotherm was determined by varying the amount of magnesium silicate (1, 2, 3, 4, 5, 6 and 7 gram) at 60 °C for 5 hours. The effect on dosage amount of absorbents were exhibited in Figure 4.

1442

Adsorption isotherms can be represented by several numerous theoretical models where Langmuir and Freundlich models have been commonly used to fit experimental data. Langmuir isotherm assumes monolayer adsorption, and is presented by the following linearized equation:

And the Freundlich linearized isotherm has the form

The model parameters are tabulated in Table 3. The Regression coefficients (R2) were 0.7137 and 0.7903 for Langmuir and Freundlich model, respectively. Based on R2, Freundlich model was more promising to represent the adsorption data than Langmuir model. Therefore the experimental data could be better described by Freundlich isotherm. However the maximum sorption capacity (Q0) and Langmuir‘s sorption affinity (b) were minus values which were physically meaningless.

1443

Conclusions The magnesium silicate adsorbent was prepared by the precipitation of magnesium sulphate and sodium metasilicate. The molar ratios of SiO2/MgO in synthetic magnesium silicate (0.31:1, 0.76:1 and 2.04:1) were obtained by varying concentrations of Na2SiO3:MgSO4. The adsorption efficiency of soap of synthesized magnesium silicate in biodiesel containing different amount of SiO2/MgO molar ratios by using 3 g of adsorbents for 150 g of biodiesel at 60˚C for 2 h was studied. The magnesium silicate at 2.04:1 SiO2/MgO molar ratio showed the highest efficiency in adsorption of soap at 77.11 % and the highest adsorption capacity at 1.37 mg/g. BET results presented the largest pore size, 105 A, at 2.04:1 SiO2/MgO molecular ratio of magnesium silicate which influenced the adsorption efficiency. The effect of dosage amount to equilibrium adsorption capacities showed decreasing in efficiency when dosage increased due to surplus amount of adsorbents. The experimental data could be better described by Freundlich isotherm than the Langmuir one because of values of maximum sorption capacity (Q0) and Langmuir‘s sorption affinity (b) were minus which were physically meaningless. Moreover, soap adsorption on magnesium silicate was both physisorption and chemisorption and well represented by the Fruenlich isotherm.

Acknownledgement This research was financially supported by Faculty of Engineering, King Mongkut's Institute of Technology Ladkrabang. Support from the Center of Advanced Instrument for Research in Engineering is also acknowledged. Reference 1. Ceclan R.E., Pop A., Ceclan M.: Biodiesel from Waste Vegetable Oils, Chemical Engineering Transactions. 2012; 29:1177-1182. 2. Sangita B., Pradeep K.V.: Effect of impurities on performance of biodiesel, Journal of Scientific & Industrial Research. 2010;69:575-579. 3. Predojevic Z.J.: The production of biodiesel from waste frying oils: A comparison of different purification steps, Fuel. 2008; 87:3522-3528. 4. Ignat R.M., Kiss A.A.: Energy efficient recovery of methanol and glycerol in biodiesel production, Chemical Engineering Transactions. 2012;29:1141-1146. 5. Clowutimon W., KItchaiya P., Assawasaengrat P.: Adsorption of Free Fatty Acid from Crude Palm Oil on Magnesium Silicate Derived From Rice Husk, Engineering Journal. 2011;15:15-26.

1444

ISFAS-1560 Air Dehumidification System Using Propylene Glycol Solution Siripan Murathathunyaluka, Nakarin Rueangwiriyanan, Laksanaporn Poolnapol a Department of Chemical Engineering, Faculty of Engineering, King Mongkut‘s Institute of Technology Ladkrabang, Thailand E-mail: [email protected] Abstract The purpose of this research is to provide a simple guide to instructors and students on how to design the air dehumidification system using propylene glycol solution. Mathematical simulation of the mass transfer in packed bed column makes use of MS Excel. Most of the previous works usually investigate the dehumidification by experimental results. Calculation by using equilibrium data was to simplify the trial design. This work will illustrate easy to use method for dehumidification system using liquid desiccant. The results of this study showed that the moisture removal rate increased with increasing inlet air and desiccant flow rate. Higher concentration and lower temperature of liquid desiccant solution lead to a higher efficiency of air dehumidification. Nevertheless, the high concentration of liquid desiccant required the high flow rate and high temperature of air to regenerate the desiccant. Therefore, the suitable desiccant concentration will significantly improve system performance. The simple calculation of performance of the dehumidification process can be used as a preliminary design of the dehumidifier. This will be the basis for the dehumidification system design for using in the future. Keywords: Desiccant dehumidification, Liquid desiccant, Packed bed column, Absorber 1. Background and Objectives Thailand is in a tropical climates area. Moisture becomes a major problem for both human comfort and air conditioning system. The high humidity of air affects the work load of air conditioning. This leads to an increase in energy consumption of air conditioning system [1-2]. Additionally, some products like food, electronic should be kept at low humidity. Desiccant dehumidification is the process of water vapor removal from moist air. It is based on the principle of moisture transfer due to the difference of vapor pressure between bulk air and air interfacing desiccant solution. Therefore, a lower vapor pressure at interface means the higher performance of dehumidification. The advantage of liquid desiccant is that regeneration can be done at low temperature with high moisture removal capacity. But the common drawbacks can be found involving corrosion (except for glycols), crystallization of salt and carryover of solution into the air stream. Many researchers have studied in physical properties and applications of liquid

1445

desiccants [3-5]. Triethylene glycol (TEG) and aqueous LiCl solution are two common desiccants for air dehumidification systems. Besides, glycols solution has low vapor pressure, with no corrosive properties, no crystallization; many experimental researches of air dehumidification using TEG have been investigated [7-8]. The vapor pressures of the glycols are lower than that of LiCl at the same temperature but the viscosities of the glycols are much higher than that of LiCl resulting in higher pumping cost [6]. Propylene glycol (PG) is an interesting desiccant. When comparing properties between PG and TEG shows that viscosity and vapor pressure of PG are lower than that of TEG and its price is also cheaper. The objective of this paper is to design air dehumidification system by using an aqueous solution of propylene glycol as a liquid desiccant at a constant temperature. The Excel spreadsheet was used as a calculating tool. 2. Methods The performance of an air dehumidification using propylene glycol (PG) as liquid desiccant under hot and humid conditions was evaluated and expressed in terms of the moisture removal rate. The experimental results of PG vapor pressure at 85 - 97%wt and temperature between 30-60 °C were used to calculate the performance of an air dehumidification. The effect of air and desiccant solution flow rate, desiccant solution temperature and concentration were investigated. 2.1 Measurement of vapor pressure The vapor pressure of a liquid desiccant solution is the pressure of water vapor at solution interface, which is a function of temperature and concentration of solution. The apparatus used for measuring the partial pressure is shown in Figure 1. The glass vessel (1) and the liquid desiccant in burette (2) were placed inside a temperature controlled box (3) to adjust to the desired temperature. After the vacuum pressure was reached using vacuum pump (4), the vacuum valve (5) was closed and the slurry valve (6) was opened at the beginning of experiment. The equilibrium vapor pressure of solution was measured by Mercury U-tube barometer (7). The temperature was measured by digital thermometer (8).

1446

2.2 Calculation of mass transfer in packed bed column. A theoretical model presented here allows mass transfer in countercurrent packed bed columns to describe in term of packing geometry and physical properties, which influence the gas-liquid or vapor-liquid systems in absorption and desorption [9]. In the case of systems with predominantly gas-phase mass transfer resistance, the height, Z of a mass transfer column is calculated from the overall number of transfer units; NOG, NOL and the overall height of a transfer unit; HOG, HOL as in equation (1) and (2).

NOG is defined by the inlet and outlet mole fraction of gas phase Y and by the ficticious vapor mole fraction Y* that is in equilibrium with the mole fraction in the bulk liquid X that can be calculated by equation (3).

And NOL is defined by equation (4).

On the other hand, HOG is determined by the actual design of column and the load of the internals such as packing geometry and physical properties of gas phase and liquid phase, regardless of whether the packing has been dumped at random or arranged in a geometric pattern; see equation (5) and (6).

1447

The moisture removal rate from the air, which was calculated from the equation (7) is the difference of moisture content in the air between outlet and inlet air.

where y and V were air humidity (gwater/kg dry air) and air flow rate (kg/h), respectively. The simulation study has been conducted to investigate the efficiency of air dehumidification process by using Excel spreadsheets. The process used in this study was absorption columns for dehumidifier and desorption (stripping) column for regenerator. The columns were defined as an isothermal countercurrent packed bed with column diameter of 12 inches (0.3159 m) and packing height of 0.5 m. The inlet air temperature of 30ºC and 70%Rh (at this condition the moisture content is 22 gwater/kgdry air) was applied to the process as a tropical climates air. The effect of independence parameters such as air flow rate, desiccant solution flow rate, desiccants solution temperature and concentration were studied. The window of the input parameters and output calculations is shown in Figure 2. There are 3 independent parameters here, which are air flow rate, desiccant solution flow rate and input concentration of desiccant solution, that need to be specified in the Excel spreadsheet along with the physical properties of packing, air and propylene glycol as input data to be able to calculate the moisture removal rate in packed bed column. The height of packed bed column, Z calculated from equation (1) is shown in Figure 3. The experimental results of PG vapor pressure was applied in plotting equilibrium line.

1448

1449

Numerical solution procedure ―Goal Seek‖ as in Figure 4 was used to calculate the moisture removal rate, on the condition that the height of packed bed is equal to 0.5 m. Follow these steps: From the Data tab > select the What if Analysis > Select Goal seek > In the Goal Seek dialog, enter the new ―what if‖ amount in the To value text box > We also need to tell Excel which cell to change. > Click OK. (This command is common for window2007 or higher.) 3. Results and Discussions To verify the present experimental procedure and apparatus of the vapor pressure measurement, the vapor pressure of water was measured and the results are presented in Figure 5. The results showed that the vapor pressure is lower at high concentration and low temperature conditions. A vapor pressure of water in the air at 30°C is 31.8 mm Hg then the vapor pressure of the air at 30ºC and 70%Rh is approximately 22 mm Hg. At low temperature (T55 ºC) because vapor pressure of PG (concentration lower than 95%wt) was higher than the vapor pressure of water in the air.

1450

Table 1 gives a calculation result of the effect of the investigated process variables in this study. A condition of inlet air was 30°C and 70%Rh and PG temperature was 30°C. It was found that the moisture removal rate increased with increasing inlet air and PG flow rate. Due to the increasing air flow rate resulted in rising amount of water. However, increasing the air flow rate caused high pressure drop in the column.

Increasing of PG flow rate slightly increase the moisture removal because the increasing of PG flow rate did not affect the area of the mass transfer. Additionally, higher concentration and lower temperature of PG solution lead to a higher efficiency of air dehumidification since absorption capacity depends on the vapor pressure difference between the surrounding air and the PG solution (the vapor pressure difference is the driving force for mass transfer and the equilibrium vapor pressure of solution depends on its temperature and concentration). At 30°C, Simulation result shows that air dehumidification using PG solution of 90 and 95 %wt gave the

1451

moisture removal rate of 3.77 and 4.87 kgwater/h, respectively. For 95%wt of PG solution, raising the temperature from 30°C to 40°C, making the moisture removal rate fell from 4.87 to 3.95 kgwater/h. PG that was fed into the stripping column was the outlet PG from the dehumidification column. The outlet PG from the stripping column will have higher concentration then it was recycled to the dehumidification column, which operates continuously. Performance of stripping column was represented in air flow rate of a hot air and pressure drop. Table 2 presents the effect of concentration and temperature of PG on the performance of stripping. We found that the high concentration of PG and low temperature of regenerated air required the high flow rate of hot air. As a result, the pressure drop in stripping column was higher. Due to the large concentration of PG will use higher air flow rate and higher temperature for the stripping. It also results in a higher viscosity of PG as well. In this study, the optimal conditions for dehumidification system using a PG were at a PG concentration of 80-90%wt at 30°C. At this condition, the stripping column can be used at 55°C resulting in energy savings. This simple calculation as shown in this study can be applied in the dehumidification design of other liquid desiccant. 4. References [1] Sanjeev Jain, P.L. Dhar, S.C. Kaushik. Experiment studies on the dehumidifier and regenerator of a liquid desiccant cooling system. Applied Thermal Engineering 2000;20:253-267. [2] A.Th. Mohammad, S.B. Mat, M.Y. Sulaiman, K.Sopian, A.A. Al-abidi. Survey of hybrid liquid desiccant air conditioning systems. Renewable and Sustainable Energy Reviews. 2013;20:186-200. [3] L.F. Chen, A.N. Soriano, M.H. Li. Vapour pressures and densities of mixed-solvent desiccants (glycols+water+salts). J. Chem. Thermodynamics. 2009;41:724-730. [4] Sanjeev Jain, P.K. Bansal. Performance analysis of liquid desiccant dehumidification systems. International Journal of Refrigeration. 2007;30:861-872. [5] Yonggao Yin, Junfei Qian, Xiaosong Zhang. Recent advancements in liquid desiccant dehumidification technology. Renewable and Sustainable Energy Reviews. 2014;31:38-52. [6] T.W. Chung, C.M. Luo. Vapor pressures of the aqueous desiccants. J. Chem. Eng. Data. 1999;44:1024-1027. [7] Y.H. Zurigat, M.K. Abu-Arabi, S.A. Abdul-Wahab. Air dehumidification by triethylene glycol desiccant in packed column. Energy Conversion and Management. 2004;45:141-155. [8] Esam Elsarrag. Moisture removal rate for air dehumidification by triethylene glycol in a structured packed column. Energy Conversion and Management. 2007;48:327-332.

1452

[9] Reinhard Billet and Michael Schultes. Predicting mass transfer in packed columns. Chem. Eng. Technol. 1993;16:1-9.

1453

ISFAS-1534 An Ultra-Wideband (0.6-7.2 GHz) CMOS VCO Using a High Tunable Active Inductor Hsuan-Ling Kao1*, Cheng-Lin Cho2, Li-Chun Chang3, Ping-Che Lee1 1 Dept. of Electronic Engineering, Chang Gung Univ., Taiwan 2 Dept. of Engineering and System Science, National Tsing Hua Univ., Taiwan 3 Dept. of Materials Engineering, Ming Chi Univ. of Technology, Taiwan *E-mail: [email protected] Integrated multiband wireless blocks in one chip IC is demanded for high-speed, high-functionality, small-size and low-cost communication systems. The general VCO circuits used spiral inductors and varactors to realize the circuit. However, the on-chip spiral inductors have a small inductance value per unit area and low Q-factor because of losses of Si-substrate. It also occupied a large chip size. Tunable active inductors (TAI) is one of the good candidates due to its small size, high Q, widely tunable inductance and large inductance compared to those of the spiral inductor. In this study, by utilizing a tunable cascade active inductor, the VCO circuit exhibits a very wide frequency tuning range and also enhances output power and phase noise. The TAI-VCO provides an output frequency ranges from 0.6 to 7.2 GHz (169%). Fig. 1shows the proposed TAI circuit consists of four transistors (M1~M5) and a current sources (I1). Two feedback loops are formed in this circuit. Transistors M1 and M2 make up the first feedback loop and M3 and M4 form the second loop. The cascode technique is used to reduce output conductance and enhance the gain for high frequency. Transistor M3 is stacked on top of transistor M1. For increasing the cascode effect, the additional gain stage was implemented by M4. The addition of transistor M4 dose not degrade the high frequency response of the inductor, because the signal path is still through M1, M2 and M3. M5 is used as a tunable current source controlled by gate voltage (Vctrl). According to equivalent circuit, the inductance (Leq) and resonant frequency (RES) of the active inductor are given by

where t1~ t4 and gm1~gm4 are the cut-off frequency and transconductance of the transistors M1-M4, respectively. Cgs and Cgd are gate-source capacitance and gate-drain capacitance of the

1454

corresponding transistors, respectively. The inductance and resonant frequency of TAI can be tuned by Vctrl. Fig. 2 shows the proposed VCO where the LC-tank is composed of a TAI and a varactor (Cv) for frequency and the negative conductance (-Gm) is employed to compensate for the loss from the LC-tank. The TAIs can be tuned over a wide range to provide coarse frequency tuning. The varactors provide a fine tuning range and LC-tank topology. Coarse tuning was achieved by TAIs and fine tuning by changing varactors. For coarse frequency tuning, TAI inductance is controlled by the controlled voltage (Vctrl). For fine frequency tuning, the varactor capacitance is controlled by the varactor voltage (Vc). The circuit is biased at VDD=1.8 V, Vg=0.7 V, Itotal=23~27 mA, Vctrl=1.4~2.7 V and Vc=0.5~2.0 V. Fig. 3 shows the tuning range of the TAI-VCO circuit. The proposed cross-coupled pair VCO was tuned, from 0.6 GHz to 7.2 GHz, which is a tuning range of 6.5 GHz (near 169 % tuning range). Fig. 4 shows the phase noise and output power of the TAI-VCO circuit. The VCO has phase noise, of -110.38 and -99.04 dBc/Hz, at a 1 MHz offset from a 0.63 and 7.2 GHz carrier, respectively. Within the VCO tuning range, the variations in output power are -7.64 ± 3.5 dBm. The fully integrated VCO with tunable active inductor on 0.18 m CMOS technology demonstrated good circuit performance, in terms of a wide tuning range, high output power and low phase noise. This TAI-VCO displayed a wide frequency range from 0.6 to 7.2 GHz, resulting in a tuning range of 169% at radio frequency. The lowest phase noise was -110.38 dBc/Hz, at a 1 MHz offset from a 630 MHz carrier and a highest output power of -3.89 dBm at 2.7 GHz under a 1.8 V power supply.

1455

Keywords: tunable active inductor, voltage-controlled oscillators References 1. Italia A., Ippolito C., and Palmisano G., A 1-mW 1.13-1.9 GHz CMOS LC VCO Using Shunt-Connected Switched-Coupled Inductors, IEEE Trans. Circuits Syst. I, 2012;59,6:1145-1155. 2. Lu L. H., Hsieh H. H., and Liao Y. T., A wide tuning-range CMOS VCO with a differential tunable active inductor, IEEE Transactions on Microwave Theory and Techniques, 2006;54,9:3462-3468. inductor with negative resistance, Proc. European Conf. on Circuit Theory and Design, 2007;990–993.

1456

ACEAIT-7482 Development of a Monoclonal Antibody Specific to Envelope Domain III of Dengue Virus with Broad-Spectrum Detecting Capability to All Four Serotypes of Dengue Virus Yu Na Kima, Sae-Hae Kima,b, Kwang Yeop Jahngc, Sung Eun Kimd, Young Mi Seold, Yong-Suk Janga,b,* a Department of Bioactive Material Sciences and Institute of Bioactive Materials, Chonbuk National University, KOREA b Department of Molecular Biology and the Institute for Molecular Biology and Genetics, Chonbuk National University, KOREA c Department of Life Sciences, Chonbuk National University, KOREA d

Department of Bioenvironmental Chemistry, Chonbuk National University, KOREA *Corresponding Author: [email protected]

Abstract Dengue virus is a mosquito-borne pathogen of human being and infects more than 390 million people in 100 countries annually. Symptoms of the virus infection include a relative weak dengue fever to severe dengue hemorrhagic fever/dengue shock syndrome which are mortal infectious diseases, however there is no commercially available vaccines or therapeutics yet. Passive immunotherapy using dengue virus-specific antibody can be considered as a strategy to treat dengue infection. We here developed the mAb, EDIIImAb-61, specific to the domain III of envelope glycoprotein (EDIII) of dengue virus with broad-spectrum detecting capability to all four serotypes of dengue virus (DENV-1~4) to use as a therapeutic antibody because although EDIII contains non-immunodominant epitopes compared to domains I and II, it plays a critical role in host receptor binding. EDIIImAb-61 exhibited the cross-reactive binding affinity to all 4 serotypes of DENV which had been isolated from infected humans. To further characterize the EDIIImAb-61, sequence of the complementarity determining regions was analyzed after cloning the full-length cDNA genes encoding the heavy and light chain of the mAb. Finally, we produced antibody from CHO-K1 cells transfected with the cloned heavy and light chain genes of EDIIImAb-61 and confirmed the binding of the antibody not only to recombinant EDIII protein but also DENV-2. Collectively, we can conclude that EDIIImAb-61 itself and the recombinant antibody produced by using the cloned heavy and light chain gene of EDIIImAb-61 can be a candidate material for a passive immunotherapy against dengue infection. (S.-H. Kim and Y. N. Kim were supported by BK21 PLUS program in the Department of Bioactive Material Sciences. This study was supported by the Korea-EU Collaborative Research Fund for GloPID-R Program NRF-2014K1B1A1073861 and LINC Project from the Ministry of Education and the National Research Foundation of Korea).

1457

Keyword: Dengue virus, Envelope glycoprotein domain III, Monoclonal antibody, Passive immunotherapy

1458

Poster Sessions (7) Mechanical Engineering and Technology (2) / Mechanic and Engineering Sciences / Nanotechnology / Civil Engineering / Electrical Engineering and Technology / Information Engineering and Technology (2)/ Computer Engineering and Technology (2) Thursday, March 31, 2016

11:00-12:00

Foyer Area

ACEAIT-7370 Mass Minimization on a Lower Extremity Exoskeleton Frame for Load-Carrying Assistive Exoskeleton Using Genetic Algorithms Warawuth Surojananon︱Mahidol University Rung Kittipichai︱Mahidol University ACEAIT-7387 Meso-Scale Modeling of the Drilling of Carbon Fiber Reinforced Plastic: Geometry and Numerical Analysis Sinan Al-Wandi︱RMIT University Wencheng Pan︱The University of Huddersfield Songlin Ding︱RMIT University John P.T. Mo︱RMIT University ACEAIT-7478 Aeroelastic Simulations of Wind Turbine Rotor Blades Using a Coupled CFD-CSD Methodology Dong Ok Yu︱Korea Advanced Institute of Science and Technology Oh Joon Kwon︱Korea Advanced Institute of Science and Technology

1459

ACEAIT-7402 The Effect of Polarization Maintaining Fiber on Tapered Fiber Sensor Employed Fiber Loop Mirror Interferometer Yi-Cheng Hsu︱National Pingtung University of Science and Technology Yu-Sheng Zhu︱National Pingtung University of Science and Technology Yeu-Chern Li︱National Pingtung University of Science and Technology Cheng-Wei Li︱National Pingtung University of Science and Technology ACEAIT-7461 Soft Computing Approaches to Pneumatic Control System: An Application to Suspension Test Bench Hsin-Han Chiang︱Fu Jen Catholic University Lian-Wang Lee︱Lunghwa University of Science and Technology Tsu-Tian Lee︱Tamkang University ACEAIT-7506 Performance Analysis of Power Steering Oil Pump Applying Al Material Yunchan Shin︱Chosun University Gwanghyun Im︱Chosun University Jongkuk Kim︱Younghwa Inc Honghyun Cho︱Chosun University ACEAIT-7538 Estimate of Mixed Convection Heat Transfer and Fluid Flow Characteristics of Plate-Fin Heat Sink in the Opening of the Enclosure Han-Taw Chen︱National Cheng Kung University Chung Yi Wu︱National Cheng Kung University ACEAIT-7614 Effect on the Boiling Characteristics in the Pool Boiling of Al2O3 Nanofluid Junggyun Ham︱Chosun University Gwanghyun Im︱Chosun University Honghyun Cho︱Chosun University ACEAIT-7686 Laser Beam Shaping Based on Genetic Algorithm Optimization Yi Chin Fang︱National Kaohsiung First University of Science and Technology Cheng Mu Tsai︱National Chung Hsing University

1460

ACEAIT-7297 An Artificial Intelligence Method to Investigate the Locations of Trader‟s Stall for Periodical Products Yi-Chih Hsieh︱National Formosa University Pei-Ju Lee︱National Chung Cheng University Yung-Cheng Lee︱WuFeng University ISFAS-1470 Characteristic and Efficiency Analysis of Storage-Type Solar PV/Thermal Hybrid Feng-Chin Tsai︱Tungnan University Jen-Ching Huang︱Tungnan University Xin-Zhi Huang︱Tungnan University ISFAS-1579 Natural Convection in Inclined Wavy-Wall Cavity Filled with Al2O3-Water Nanofluid Using Heatline Technique Ching-Chang Cho︱National Formosa University Ching-Huang Chiu︱National Formosa University Kuo-Ching Chiu︱National Formosa University Chong-You Lai︱National Formosa University ISFAS-1482 A Study on Mechanical Behaviors of Si/Ge Core-Shell Nanowires by Molecular Dynamics Simulation Yu-Shan Lin︱National Cheng Kung University Tei-Chen Chen︱National Cheng Kung University ISFAS-1577 Applications of Magnetic Particles to Adsorb Emerging Contaminants in Aqueous Solution Hua-Wei Chen︱St. Mary’s Medicine, Nursing and Management College Yo-Ping Wu︱National Ilan University Shu-Han Chang︱National Ilan University Chyow-San Chiou︱National Ilan University

1461

ISFAS-1593 Surface Enhanced Raman Scattering on Periodic Gold Nanowire Arrays Yu-Chung Chang︱National Changhua University of Education Yu-Ju Hung︱National Sun Yat-Sen University Bo-Han Huang︱National Changhua University of Education Tsung-Hsien Lin︱National Changhua University of Education Chun-Han Chen︱National Changhua University of Education ACEAIT-7555 Development of Typhoon Rainfall Forecast Model Using Satellite Data Chih-Chiang Wei︱National Taiwan Ocean University ACEAIT-7561 Flexural Behaviors of RC Non-Uniform Beam with Lap Details Using Headed Bars Seung-Hun Kim︱Hanbat National University Kyu-Seon Lee︱Hanbat National University ACEAIT-7459 Wireless Monitoring and Prediction of Rail Temperature Chen-Ming Kuo︱National Cheng Kung University Tsung-Hao Han︱National Cheng Kung University Tzu-Fan Chiang︱National Cheng Kung University ACEAIT-7606 Preliminary Study on the Monitoring System by Using UAV Imagery Jiyoung Kim︱Seoul National University Yoonsik Bang︱Seoul National University Yangsun Kim︱Mokpo National University Jae Bin Lee︱Mokpo National University ACEAIT-7532 Object Detecting LED Control System Using Doppler Sensor Sang Woo Jung︱Kyungpook National University Seung Hyun Paik︱Kyungpook National University Jun Yeong Lee︱Kyungpook National University Hong Bae Park︱Kyungpook National University

1462

ACEAIT-7560 The Green Design of VCO and Mixer Circuits for Ultra-Wide Band Applications H.-C. Chow︱Chang Gung University Y.-C. Kuo︱Chang Gung University W.-S. Feng︱Chang Gung University ACEAIT-7607 A Multi-Agent System for Microgrid Optimal Operations Wei-Tzer Huang︱National Changhua University of Education Tsai-Hsiang Chen︱National Taiwan University of Science and Technology Jhih-Siang Yang︱National Taiwan University of Science and Technology Hong-Ting Chen︱National Changhua University of Education Yung-Ruei Chang︱The Institute of Nuclear Energy Research Yih-Der Lee︱The Institute of Nuclear Energy Research Yuan-Hsiang Ho︱The Institute of Nuclear Energy Research ACEAIT-7611 Assessment of Upgrading Premise Wiring Systems from Single-Phase 2-Wire to Single-Phase 3-Wire System Rih-Neng Liao︱National Taiwan University of Science and Technology Tsai-Hsiang Chen︱National Taiwan University of Science and Technology ACEAIT-7610 Assessment of Maximum Allowable Capacity of EVs along a Distribution Feeder Tsai-Hsiang Chen︱National Taiwan University of Science and Technology Rih-Neng Liao︱National Taiwan University of Science and Technology ACEAIT-7543 Visual Imaging Technology for Urban Flood Monitoring and Warning Applications Lun-Chi Chen︱National Center for High-Performance Computing Chien-Hao Tseng︱National Center for High-Performance Computing Fang-Pang Lin︱National Center for High-Performance Computing Jyh-Horng Wu︱National Center for High-Performance Computing

1463

ACEAIT-7424 Effects of Virtual Reality Presentation Media on Operator Learning Performance in the Main Control Room of the Nuclear Power Plant Chiuhsiang Joe Lin︱National Taiwan University of Science and Technology Rou-Wen Wang︱Chung-Yuan Christian University Chih-Wei Yang︱Institute of Nuclear Energy Research

1464

ACEAIT-7370 Mass Minimization on a Lower Extremity Exoskeleton Frame for Load-Carrying Assistive Exoskeleton Using Genetic Algorithms Warawuth Surojananona, Rung Kittipichaib Department of Mechanical Engineering, Faculty of Engineering, Mahidol University, THAILAND a E-mail: [email protected]; [email protected] Abstract This paper introduces a new design method for the frame of a lower extremity exoskeleton for assist in carrying the load on the back of its user. The exoskeleton, as a mechanical structure that can be worn on human lower body, assists its user by sharing the load on the user‘s back and transmits it to the ground via its structure. Therefore, the frame of the exoskeleton requires a careful trade-off between its structure mass and its capabilities to sustain the load, so that the structure neither cause too much burden on the user‘s legs due to its mass nor break down during the operation due to failures occurred from sustaining the load. To efficiently search for the optimal point of this trade-off, the method developed in this paper combines an evolutionary optimization method ―Genetic Algorithms‖ along with the finite element method to minimize the mass of the exoskeleton frame while maintaining its structure properties to avoid structure failure. Keyword: Lower Extremity Exoskeleton, Genetic Algorithms, Finite Element 1. Introduction Technologies to enhance human abilities has become research topics in human history for a long time. Among these technologies, Lower extremity exoskeleton is an assistive device that can be worn and fits closely to the lower part of user body in order to enhance his/her abilities [1]. There are many works that utilize this type of exoskeleton in order to increase human load carrying capability [2-5]. In this type of exoskeleton, the equipment such as electrical motors, hydraulic systems as well as spring systems are used to directly add power in supporting user‘s movement. Such exoskeletons require a computer system to control these devices. Therefore, to design the frame of a lower extremity exoskeleton, it is necessary to consider its structure properties so that the frame can sustain loads from either the carrying weight or the installed equipment without structural failure. However, increasing these properties leads to the increasing in the frame mass which can result in adding more burdens to user‘s body. It is importance to carefully adjust this trade-off. An attempt to adjust this trade-off can be seen in Reference [6], where Ding et al. presents a success design method for a light weight

1465

exoskeleton of the rehabilitation training and the optimization technique is done through the investigation of the finite element results. However, this optimization process has to be done upon the assumption of the designer, the efficiency of the result may vary upon such an assumption and it is possible that there are still quite a distance between the result and the optimal point. To reduce this distance, it is necessary for the designer to walk back and forth between assumption process and analysis process multiple times, which can consume a large amount of time and effort. This paper proposes an idea for designing an exoskeleton frame by employing optimization technique to minimize the mass of the exoskeleton as well as maintain its structure analysis. Genetic Algorithms (GAs) as an optimization method used in this work because it has versatility to be used with various kinds of problem and has high probability to find the global optimal point of the problem. The structure properties included in this paper are displacement, stress and buckling which are identified via finite element method. All the procedure are done via MATLAB program. 2. Theories on structure analysis In this paper, the lower extremity exoskeleton have two mechanical legs. The structure properties of the exoskeleton frame are calculated through finite element method with space frame element [7]. Basically, the finite element method is employed by conceptually separating the target structure into minor components call elements where the interconnection points as well as the boundary points are called nodes. To deal with the structural displacement, stress and buckling of member, this section describes the analysis of structure of exoskeleton using the finite element method. To analysis the structure of exoskeleton which is treated as the space frame in the finite element model, the finite element selected in this paper is the 2-node element which each node can support nodal displacement in 6 degrees of freedom (DOFs) such as the translations along x, y, z directions and rotations around x, y and z axes which make it an element with 12 DOFs. The finite element model of an individual element can be formed as follows

where {fe} and {de} refer to the force and displacement vector which describe the directions of force and displacements on both nodes of the eth element. The matrix [ke] is called stiffness matrix. The stiffness matrix of a space frame element is written as

1466

When combining a model of an element to form an overall model of the system, it is possible that the coordinate system of an element is different from the coordinate system of overall structure. Therefore, it is necessary to transform the coordinate system of the elements from local system to global system before combining them. The transformation between local and global axis can be described as

where {q} is the vector in local coordinate, {Q} is the vector in global coordinate and the transformation matrix [T] which consists of the direction cosines between the global and local coordinate system can be written as

where [] is the direction cosine on each axis and the details can be found in Reference [7]. By transforming the force and displacement vectors in Equation (1) with Equation (3), the equation becomes

The equation can be rearranged by multiplying with [T]T to become

1467

By assembling the global element equations as given in Equation (6) from each element, the equation of equilibrium for the whole structure can be formed as follows

where {F} is the matrix of external forces, {D} is the vector of overall nodal displacements and [K] is the global stiffness matrix for the structure. Once, the element matrices of the system are formed. The unknown displacement can be found by applying external forces and boundary conditions to the model. The unknown nodal displacements for the system are obtained by solving Equation in the manner of

When all the nodal displacements are found, the stress and strain of each element can be individually determined as follows

where [Be] is the strain-displacement matrix and [De] is the constitutive matrix [7]. In the exoskeleton structure, the element which is under the compressed load may cause the structural failure due to buckling or yielding. Particularly, the element which is long and slender tends to fail by buckling rather than by yielding. Such an element called column will buckle when the compressed load reaches a critical level. It is called the critical load, Pcr. The critical buckling load can be calculated by using Euler equation [10]. Then the buckling of any element can be determined through the use of critical stress as follows

where L is the length of column, k is the radius of gyration of column, E is modulus of elasticity and C is the constant depending upon the end conditions [10]. To ensure that bucking doesn‘t occur in an element, the stress within an element must be lower than the critical stress of the element. In addition, the safety factor can be proposed to prevent no visible evidence of impending buckling. In this paper, the structural displacement, stress and buckling of member are considered as the constraints in the optimization process.

1468

3. Theories on optimization method To minimize the mass of the exoskeleton frame under a specific structure conditions, Genetic Algorithms (GAs) as a search method is selected. The GAs is an evolutionary optimization method based upon the theory of natural selection. The procedure of GAs starts by creating a group of initial solutions as the first population. Then each population is tested for its fitness and selected for creating offspring in the next generation. The process of creation and selection are repeated until the objective conditions are met or the process is repeated with the same value for a set amount of time. The advantage of this method is that GAs can search for a global optimum point and, as it can be seen as a black box method, it has versatility to be modified to match with various kinds of problems. Population in GAs composes of a set of design variable called chromosomes. These chromosomes are in the form of binary numbers which compose of ―0‖ and ―1‖. To find fitness (Objective function) of each chromosome, it is necessary to transform the binary number into decimal number xi using

As GAs is originally a method for solving unconstrained optimization problem, penalty function has to be applied so that the GAs can solve the constrained optimization problem. The static penalty function [11] has been used to modify the GAs. Static penalty function is a type of penalty function which penalizes the objective function with a static value when one of the constraints is violated. The term of objective function applied with a penalty function is as follows

1469

Selection is a genetic operator to randomly select a chromosomes as one parent string from the current population to create offspring for the next generation. The chromosome with a high fitness value typically has more chance to be selected as a parent string than the chromosome with a low fitness value. Therefore the probability of selecting the ith chromosome is given as follows

where N is a number of chromosomes in a population, fj is the fitness value of the jth chromosome and fi is the fitness of the ith chromosome. However, if one of the chromosomes has an exceeding fitness value, using Equation (15) may result in self-breeding which can stall the process at one local optimal point. To avoid self-breeding, the equation (15) is modified by first sorting the chromosomes by their fitness in ascending order, then the weight of an i th chromosome can be defined as follows [12]

By using Equation (16), the weight of the fittest chromosome is only approximately twice of the median value. This increases the chance that the others chromosomes are being selected than (15). The offspring of the selected chromosomes are created through two genetic operators. Crossover is a genetic operator used for passing out the characteristic of chromosomes in one generation to the next one. In crossover, new offspring are created by selecting two parent chromosomes using selection procedure described in the above discussion. The binary strings in both parent chromosomes are then divided into two parts where the second part are exchanged together resulted in two new chromosomes. This procedure is repeated until the number of new population meet with the size of the previous generation. Mutation is a genetic operator created to maintain the variety of the chromosomes in next generation. This is to reduce the chance that the process get struck in a local optimal point due to the lack of diversity in populations created through crossover. Mutation is set to occur within a small probability. When the mutation occurs, some of the chromosomes are randomly selected and a number within binary string of each selected chromosome is then flipped from 0 to 1 or vice versa and results in the change in decimal value of the chromosomes. 4. Method The objective of this design is to minimize the mass of a lower extremity exoskeleton frame using GAs, the other installations on the frame are considered as additional loads on the frame

1470

as shown in Figure 1. The whole exoskeleton frame is made of Aluminium Alloy. The material properties of Aluminium Alloy [9] are given in Table 1.

The exoskeleton is analyzed in standing position with two feet on the ground while the carrying load 25 kg is attached to its back. The finite element model of the exoskeleton frame contains 24 elements and 23 nodes. To deal the displacement of structure as one constraint, the maximum displacement is defined to 0.005 m. The procedure of optimization program is shown in Figure 2. To minimize the mass of the exoskeleton frame via changes in the size of cross-section of each element as shown in Figure 1, the constrained optimization problem for this case can be given as follows

1471

where νi is the nodal displacement in x, y and z-direction of the ith node, Ri is the ratio of the applied stress to the allowable stress in each element and  is the ratio of applied stress to the critical stress in the buckling case. The safety factor is set to 2. To determine the size of cross-section of each element as the design variables, The GAs is defined to 2,000 generations before termination.

5. Result and Discussion After running the optimization program for 2,000 generations, the minimum mass of the exoskeleton frame is obtained to be 16.16 kg while the value of stress and displacement are within the restriction set in the optimization program and none of the elements exhibit bucking behavior which prove the success of achieving minimum mass of the exoskeleton via GAs with finite element analysis. Figure 3 shows the behavior of GAs while searching for the minimum mass. The best fitness in each generation can be seen steadily moving toward the minimum point while the average fitness shows the varieties in each generation. Table 2 shows the results of searching for the width and thickness of each element at the optimum. The maximum stress in the structure occurs at an element no. 3 between node no. 3 and node no. 6 as well as the element no. 15 between node no. 3 and node no. 16 at the opposite side. The largest displacement occurred at the nodes of these element. Another elements with similar amount of stress is element no. 6 between node 4 and 7 and the element no. 16 between node 14 and 17 at the opposite side. Figure 4 shows the displacements occurred within the structure in scaled form for easy recognize.

1472

6. Conclusion Load carrying assistive exoskeleton is a device that can support labor works of people in many fields such as soldiers, firefighters, agriculturists and construction workers. As the exoskeleton in this type is operated by wearing on the lower body of the user and bear the load carried on the back of the user, it is important that the mass of the exoskeleton frame is minimized so that it would not cause too much burden on user‘s body as well as able to withstand the carrying load without the occurrences of structure failures. In this paper, Genetic Algorithms combining with Finite element method is developed to find the minimum mass of a lower extremity exoskeleton subjects to constrain such as structural displacement, stress and buckling. The result shows the success of searching for the minimum mass of the exoskeleton under the acceptable range of structure properties to avoid structure failures. This paper shows the ideas

1473

of designing the frame of a load carrying assistive lower extremity exoskeleton which can be used to further the development of an effective, lightweight exoskeleton. 7. References [1] Dollar, A.M. and Herr, H., Lower Extremity Exoskeletons and Active Orthoses: Challenges and State-of-the-Art, IEEE TRANSACTIONS ON ROBOTICS, vol. 24, no. 1, February 2008. [2] Zoss, A.B., Kazerooni, H. and Chu, A., Biomechanical Design of the Berkeley Lower Extremity Exoskeleton (BLEEX), IEEE/ASME TRANSACTIONS ON MECHATRONICS, vol. 11, no. 2, pp. 128-138, April 2006. [3] Cao, H., Ling, L., Zhu, J., Wang, Y. and Wang, W., Design Frame of a Leg Exoskeleton for Load-Carrying Augmentation, IEEE International Conference on Robotics and Biomimetics, pp. 426-431, December 2009. [4] Walsh, C. J., Paluska, D., Pasch, K., Grand, W., Valiente, A. and Herr, H., Development of a lightweight, underactuated exoskeleton for load-carrying augmentation, IEEE International Conference on Robotics and Automation, pp. 3485-3491, May 2006. [5] Fang, Y., Yu, Y., Chen, F., Ge, Y., Dynamic Analysis and Control Strategy of the Wearable Power Assist Leg, IEEE International Conference on Automation and Logistics, pp. 1060-1065, 2008. [6] Ding, B., Qian, J., Shen, L. and Zhang, Y., Finite element analysis and optimized design of exoskeleton for lower extremity rehabilitation training, IEEE International Conference on Robotics and Biomimetics, pp. 1397-1402, 2012. [7] Rao, S. S., The Finite Element Method in Engineering, 4th Ed., Elsevier Science & Technology Books, 2004. [8] Pachi, A. and Ji, T., Frequency and velocity of people walking, The Structural Engineer, Institution of Structural Engineers, pp. 36-40, February 2005. [9] Callister, W. D., Materials science and engineering : an introduction, New York: John Wiley & Sons, 2003. [10] Shigley, J. E., Mischke, C. R. and Budynas, R. G., Mechanical engineering design, 7th ed., New York: McGraw-Hill, 2003. [11] Yeniay, O., Penalty function methods for constrained optimization with genetic algorithms, Mathematical and Computational Applications, vol. 10, pp. 45-56, 2005. [12] Reeves, C. R., A genetic algorithm for flowshop sequencing, Computers & Operations Research, vol. 22, pp. 5-13, 1995.

1474

ACEAIT-7387 Meso-Scale Modeling of the Drilling of Carbon Fiber Reinforced Plastic: Geometry and Numerical Analysis Sinan Al-Wandia,1, *, Wencheng Panb,2, Songlin Dingc,1, John P.T. Mod,1 1 School of Engineering, RMIT University, Australia 2 School of Computing and Engineering, The University of Huddersfield, United Kingdom a Email: [email protected], [email protected], [email protected], d [email protected] Abstract This paper discussed the 3D finite element modeling (FEM) of the drilling of uni-directional Carbon Fiber Reinforced Plastic (CFRP). Most of the real life parts of CFRP are modeled with single layer shell element and can be modeled as composite by assigning a composite property to it. A Meso-Scale (Laminate Level) approach has been developed to extract displacements, overall stiffness behavior, and detailed stresses and strains. Ply-based modeling technology was implemented to model the laminates using the classical laminate theory keeping the laminates bonded together. Hashin damage mechanism and failure criterion for the composite material was defined in the FE simulation process to simulate the damage process of matrix and fiber. Keywords: Finite Element Model, drilling, Meso-Scale, CFRP. 1. Introduction The development of accurate and reliable machining process models has received considerable attention from both academic researchers and industry practitioners in recent years. Traditionally, the techniques used in industry are based on past experience, extensive experimentation, and trial-and-error. Such an approach is time consuming, expensive, and lack of a rigorous general scientific knowledge [1]. The alternative approaches are mathematical simulations where numerical methods are applied. Amongst the numerical procedures, the finite-element methods (FEMs) are the most frequently used [2]. Progressive damage modeling of polymer matrix composites has received a great deal of attention in recent years as predictive capabilities for the complex nonlinear behavior of these materials are sought [3]. Generally, the term meso-scale is used to describe the intermediate level of material description between micro-scale (modeling the composite to the scale of the constituents such as fibre filament and matrix materials) and macro-scale (modeling the composite to the scale of smeared or homogenous shell/solid, with no detailed ply stresses or strains). The micro-scale is usaually very computationally demanding whereas the meso-scale

1475

approach is treated as effective anisotropic materials and is more computationally efficient. Niezgoda et al [4] proposed a global-local (meso) modeling method in order to reduce the computational effort usually required to analyze composite structures. The number of elements and nodes used in the meso-scale approach was approximately 60% lower than in the micro-scale model which allowed analyzing a three-dimensional structure made of composite materials with different orientation of the fibers in a reasonable amount of time. Cherniaev et al [5] presented a meso-scale approach to numerical modeling of hypervelocity impact on CFRP. The material were models suitable for explicit modeling of laminate structure, including fiber-reinforced layers and resin-rich regions. Results of simulations were verified against experimental data, good correlation with experiments has been revealed in terms of predicted delamination area and qualitative representation of external damage. Stier et al [6] modeled a meso-scale representative volume element (RVE) of a twill weave CFRP ply consisted of the yarn part, and the surrounding matrix region. The material model accounted for different yield strengths in tensile and compressive direction as well as pressure dependence; verification of the meso-model was conducted through experiments. Numerical and experimental results were in good agreement at small deformations, however at large deformations, particularly at large shear strains, the results slightly differed. Lubineau et al [7] proposed a pragmatic approach for the description of the degradation of laminated composites under cyclic loading, including the effect of oxidation based on hybrid micro-scale and meso-scale modeling of the degradation which included and generalizes classical micromechanical approaches for static loading, the results obtained are promising but further research remains necessary for the model to become completely quantitative and predictive. This paper introduced the new modeling method of ply-based technology to study a 3D finite element model for explicitly simulating the drilling process to investigate the relative significance of the drilling parameters on thrust force and torque. It was significantly shown that the thrust force and torque increase with the increase of feed rate, which indicates the importance of reducing feed rate to achieve better results with respect to drilling forces. 2. Model Setup 2.1 Meso-Scale modeling Meso-scale was first introduced by Panin et al. [8] as an approach of modeling properties and the deformation and failure mechanisms of structural and functional materials at different scale levels, with different loads. In meso-scale continuum mechanics size matters, as plastic flow processes in crystalline solids are inherently size dependent over a scale that ranges from a fraction of a micrometer to 100 mm. Such formulations are intermediate between a direct atomistic and an unstructured continuum description of deformation processes [9].

1476

As pointed out by Mishnaevsky [10], the following scale levels are applied in the analysis of material behavior: • Macro-scale (or a specimen scale), of the order of more than 1 mm. Material behavior at this scale level is analyzed using continuum mechanics methods. • Micro- and meso-scale (or a microstructure scale), between 1μm and 1 mm. Material behavior at this scale level falls into the area of materials science, and is analyzed using methods of both physics and mechanics of materials, including micromechanics and fracture mechanics. • Nano- and atomistic scales, less than 1μm. Material behavior at this scale level falls into the area of the physics of materials. 2.2 Ply-Based Modeling Zone-based models and ply-based models are commonly employed to create composite models. Zone-based models discretize the composite into zones of effective properties, each zone is aligned a material property based on the ply-stacking and fabric deformation. Ply-based models are used to model each ply in a composite, and material properties and local orientations are assigned to each ply [11]. In this paper, ply-based modeling was applied. Fig. 1 shows a simple plate made by four plies. Each ply has different size and orientation, particularly layer #4 is diagonally oriented. In a zone-based approach the plate has to be divided into 8 zones with different thickness and different fiber direction across the thickness [12], a ply-based approach only requires four plies to capture the behavior of the structure. 2.3 Finite Element Setup Ply-based 3D Lagrangian formulation finite element (FE) model of CFRP composite was developed using ANSYS-EXPLICIT software, shell181 type of uni-directional CFRP composite laminate was used in the FE analysis as shown in Fig.2.

1477

The boundary conditions were enforced on the workpiece to enable simulation process, as it was fixed from the sides from moving (ux=uy=uz=0).

3. Material Modeling A meso-scale modeling approach was used for different components of composite laminate. The material model used for the workpiece was orthotropic homogeneous elastic assigned according to the fibre orientation through a defined local coordinate system. Table 1 shows the material properties of VTM264 [13].

3.1 Constitutive Material Model of Unidirectional CFRP Intralaminar and interlaminar damage mechanisms are the basis of fracture process of composites which include fiber breakage, matrix cracking and delamination. The orthotropic

1478

material model for shell layers developed can be expressed as in-plane stress-strain relationship [14]:

The stress-strain relation in terms of reduced stiffness is:

Whereas in terms of engineering terms:

The stress-strain relationship becomes:

Where E is modulus of elasticity; C is stiffness tensor; and Q is the reduced stiffness. 3.2 Interlaminar Normal Stresses and Shear Stresses In the analysis of layered composite structures, shell elements are widely used to keep the computational effort reasonable. In-plane stresses and even transverse shear stresses can be predicted with good accuracy using shells based on the first-order shear deformation theory (FSDT). As shown in Fig. 3, using a cylindrical coordinate system for describing an arbitrary doubly curved shell is the basis for computation of interlaminar normal stresses (INS).

1479

The differential equation of the through-the-thickness displacement is

P is the particular part of the differential equation.

3.3 Constitutive Damage Initiation and Evolution Different failure criteria has been formulated and implemented in finite element analysis of CFRP to predict the damage initiation and evolution. Failure means that one of the stress components reaches the yield stress, and then damage occurrences and progressive failure can be observed. Damage can progress in different directions around the weakest element in the model. Usually ―Matrix Cracking‖ is the first damage process to take place since the matrix has the lowest stress to failure [16]. Plates and panels are easily subjected to in-plane multiaxial stress states during service and, depending on the stacking sequence, the in-plane shear stress

1480

may play a significant role in the failure. Failure criteria can be grouped into two main groups [17]: • Failure criteria neglecting interactions between different stress components. • Failure criteria considering interactions between different stress components. The first group of failure criteria is sometimes referred to independent failure criteria whilst the second group is referred to polynomial failure criteria. The maximum stress criteria and the maximum strain criteria are the examples of the independent failure criteria. In these criteria, there isn‘t any dependence between the stress/strain components. This means a component in the longitudinal direction does not affect the component in the transverse direction or stacking direction. Polynomial maximum stress criterion, Polynomial maximum strain criterion, Tsai, Tsai-Hill, Tsai-Azzi, Hoffman, Tsai-Wu and Hashin criteria are the examples of the polynomial failure criteria. In all these criteria, there is dependence between the components in each direction [18,19]. Hashin damage mechanism and failure criterion for the composite material was defined in the FE simulation process to simulate the damage process of matrix and fibre, the failure criterion takes into account the interaction between the four failure modes and therefore chosen in the finite element analysis. Tensile Fibre Mode

Compressive Fibre Mode

Tensile Matrix Mode

Compressive Matrix Mode

Where Hashin‘s criteria have been fulfilled in any mode damage will be initiated.

1481

Without damage initiation criteria, the damage evolution law has no effect on the material. The Damage Evolution Law defines the way a material degrades following the initiation of damage. The stiffness reduction takes a value of 0 to 1, where 0 is no damage and 1 is completely damaged. The evolution law of the damage variable in the post-damage initiation phase is based on the fracture energy dissipated during the damage process, Gc. The damage variable can be defined from the equation below:

The failure displacement can be calculated as follows:

4. Results and Discussion The resultant workpiece model was a uni-directional CFRP composite laminate with stack sequence of [(0/90)2]s used in the FE analysis. The dimension of the sample is 20mm x20mm x1.6mm which consisted of 8 plies, as shown in Fig 4. The thickness of each ply was 0.2mm. The workpiece was meshed with shell181 four node linear quadrilateral elements. The mesh was refinement in the drilling area with 200µm elements, total number of elements used was 4,205, while the contact interfaces between the plies were meshed with four node linear quadrilateral contact & target elements, total number elements used was 3,364.

The simulation of the drilling process has generated thrust force and torque as shown in Fig 5a

1482

and Fig 5b. The data was validated with drilling experiments using spindle speed of 5000 rpm and feed rate of 500 mm/min shown in Fig 6a and Fig 6b. Experiment results show the average thrust force is 123.3 N whereas the FE result shows 118 N, the experiment torque is 0.39 N.m and FE torque result is 0.36 N.m. The accuracy between the experiment and the FE simulation results are very close which indicates the FE model is accurate. Ply-based composite model preparation and definition of layup can be shared from existing composite layup and can be updated faster when product design changes are required. While there are manufacturing effects that need to be considered when laying up the laminates such as draping (the avoidance of kinking when wrapping material around curved shapes), as shown in Fig. 8, the ply application on curved surfaces changes the theoretical fiber orientations and deformation occurs with the in-plane shear and up to certain deformation level and it‘s important to know how big this effect can be if it‘s considered. Ply-based modeling allows to evaluate the draped fiber directions with angles can be visualized and are considered in all analysis resulting in more accurate evaluations. Hence the need for draping simulation is twofold. Firstly, the manufacturability of the composite product can be assessed, areas where the reinforcement cannot follow the surface are indicated and hence measures can be taken in design to avoid this. Secondly, the draping simulation gives the actual fiber orientations at any location in the model. This information is needed for accurate finite element analysis of the structure.

1483

1484

5. Conclusions A new model was developed to simulate the drilling of CFRP based on Meso-scale modeling which is a good base material design and for analyzing mechanisms of composite material behavior and its degradation. By using shell elements, it reduces the computational of the simulation process. The model was found to be accurate through validating experiment results which show that the simulation results match the data measured in real drilling tests. References [1] Limido, J., et al. (2007). "SPH method applied to high speed cutting modelling." International Journal of Mechanical Sciences 49(7): 898-908. [2] R Izamshah R.A, J. Mo, S. Ding (2012). Hybrid deflection prediction on machining thin-wall monolithic aerospace component. Proceedings of the Institution of Mechanical Engineers, Part B, Journal of Engineering Manufacture. 226(4), 592-605. [3] Bednarcyk, B. A., et al. (2015). "Meso- and micro-scale modeling of damage in plain weave composites." Composite Structures 121(0): 258-270. [4] Niezgoda, T. and A. Derewońko (2009). "Multiscale composite FEM modeling." Procedia Engineering 1(1): 209-212. [5] Cherniaev, A. and I. Telichev (2015). "Meso-scale modeling of hypervelocity impact damage in composite laminates." Composites Part B: Engineering 74(0): 95-103. [6] Stier, B., et al. (2015). "Comparing experimental results to a numerical meso-scale approach for woven fiber reinforced plastics." Composite Structures 122(0): 553-560. [7]

Lubineau, G., et al. (2006). "Durability of CFRP laminates under thermomechanical loading: A micro–meso damage model." Composites Science and Technology 66(7–8): 983-992. [8] Panin, V. E., et al. (1998). "Physical mesomechanics of materials." Russian Physics Journal 41(9): 856-884.

1485

[9] [10] [11]

[12]

[13]

Needleman, A. (2000). "Computational mechanics at the mesoscale." Acta Materialia 48(1): 105-124. Mishnaevsky, L. (2007). Mesoscale Level in the Mechanics of Materials. Computational Mesomechanics of Composites, John Wiley & Sons, Ltd: 13-36. Mitchell C. J., et al (2014). "A Discrete Mesoscopic Finite Element Model Used as a Design Tool for Textile Composite Structures." Proceedings of the American Society for Composites: Twenty-ninth Technical Conference on Composite Materials. Stelzmann U., et al. (2011). " Ply-Based Composite Modeling with the New *ELEMENT_SHELL_COMPOSITE Keyword." Presented at the 8th European LS-DYNA Users Conference. Burns LA, Mouritz AP, Pook D, Feih S (2012) Strength improvement to composite T-joints under bending through bio-inspired design. Composites Part A: Applied Science

and Manufacturing 43 (11):1971-1980. [14] Jones R.M. (1975) Mechanics of Composite Material, Hemisphere, New York. [15] Roos, R., et al. (2007). "A post-processing method for interlaminar normal stresses in doubly curved laminates." Composite Structures 81(3): 463-470. [16] Rakesh, P. K., et al. (2011). "Delamination in Fiber Reinforced Plastics: A Finite Element Approach." Engineering 03(05): 549. [17] Nali, P. and E. Carrera (2012). "A numerical assessment on two-dimensional failure criteria for composite layered structures." Composites Part B: Engineering 43(2): 280-289. [18] Isbilir, O. and E. Ghassemieh (2012). "Finite Element Analysis of Drilling of Carbon Fibre Reinforced Composites." Applied Composite Materials 19(3-4): 637-656. [19] Lapczyk, I. and J. A. Hurtado (2007). "Progressive damage modeling in fiber-reinforced materials." Composites Part A: Applied Science and Manufacturing 38(11): 2333-2341.

1486

ACEAIT-7478 Aeroelastic Simulations of Wind Turbine Rotor Blades Using a Coupled CFD-CSD Methodology Dong Ok Yua, Oh Joon Kwona* a Department of Aerospace Engineering, Korea Advanced Institute of Science and Technology, Korea *Corresponding Author: [email protected] Abstract Horizontal-axis wind turbine rotor practically operate under the influence of atmospheric shear layer, and are also frequently exposed to off-axial flow conditions due to yaw misalignment. Under these circumstances, the rotor blades experience time-changing inflows for a significant portion of the turbine operation time. As a result, the blade aerodynamic loads vary in an unsteady dynamic manner with the rotor azimuth. These unsteady loads are one of the largest contributors of degrading the power supply quality and also causing structural failure of the turbines. Therefore, it is quite important to better understand the time-varying load behaviors at realistic operating situations, if more efficient and reliable wind turbines are to be designed in the future. In the present study, aeroelastic simulations of horizontal-axis wind turbine rotor blades are conducted using a coupled CFD-CSD methodology. The investigations are made to assess the unsteady blade aerodynamic loads and the dynamic blade responses due to yaw misalignment and non-uniform sheared wind. For this purpose, an unstructured mesh Navier-Stokes CFD solver and a FEM-based CSD beam solver are used. The CFD and CSD solvers are calculated simultaneously by exchanging the data between the two solvers in a loosely coupled manner. The present coupled CFD-CSD method is applied to the NREL 5MW reference wind turbine rotor, and the results are compared with those of CFD-only rigid blade calculations. It is found that aeroelastic blade deformation leads to a significant reduction of blade aerodynamic loads, and alters the unsteady load behaviors, mainly due to the torsional deformation. The reduction of blade aerodynamic loads is particularly significant at the advancing rotor blade side for yawed flow conditions and at the upper half of the rotor disk plane where wind velocity is higher due to wind shear. Keyword: Aeroelasticity, CFD Simulations, Coupled CFD-CSD Method, Wind Turbine Rotor Blades

1487

ACEAIT-7402 The Effect of Polarization Maintaining Fiber on Tapered Fiber Sensor Employed Fiber Loop Mirror Interferometer Yi-Cheng Hsu*, Yu-Sheng Zhu, Yeu-Chern Li, Cheng-Wei Li National Pingtung University of Science and Technology, Biomechatronics Engineering, Taiwan *Corresponding Author: [email protected] Abstract In this study, an electrical arc method was employed to fabricate taper structure on a standard single-mode-fiber (SMF-28, Corning) using a fusion splicer with a stretching mechanism. The taper fiber was fabricated with multiple fusions with single stretching (MFSS). The taper structure is treated as a sensing area and connected to a coupler to form a fiber loop mirror interferometer (FLMI) sensor. In order to obtain the effect of the polarization maintaining fiber (PMF) on the sensitivity of FLMI sensor, the fiber taper of the FLMI sensor were manufactured by MFSS process with the taper waist D of 40μm for refractive index measurements. The results show that the global spectrum shift of the FLMI of 40 μm taper waist without PMF shows the best sensing sensitivity of -3423.71 nm/RIU. Compare to FLMI with PMF, the best sensing sensitivity is degraded to -77.17 nm/RIU, however, the R2 is averagely improved and may be attributable to the superior extinction ration of interference spectrum. Keywords: Fiber sensor, Fiber taper, Polarization maintaining Fiber, Fiber loop mirror interferometer 1. Introduction Optical fiber sensors have been widely researched for excellent characteristic of optical fiber, such as low insertion loss, high flexibility, and electromagnetic immunity, etc. So far, various process and fabrication techniques have been used for optical sensor. By physical or chemical machining methods, optical fiber can increase sensitivity of detecting the change of environmental signals[1-5]. FLMI principle is achieved by the fiber sensor connected into a 2×2 directional fiber coupler as a new loop, then two waves travel within identical optical path lengths in opposite directions and interference is assured when the waves recombine at the coupling region. Thus, all light is reflected back into the input port. When external environment of the sensing element have been changes, the spectrum of the interference patterns will shift[6].

1488

Polarization maintaining fiber (PMF) and taper fiber have recently been researched as FLM of the sensor region[7-8]. PMF is expensive and some of taper fibers are fabricated by flame process is lack of mechanical strength. In this study, we report a method to fabricate TF-FLM by using fusion splicer. The FLM based on a fiber taper which detects changes in surrounding refractive index (RI) was studied. 2. Experiments In this study, an electrical arc method was employed to fabricate taper structure on a standard single-mode fiber (SMF-28, Corning) using a fusion splicer with a stretching mechanism. The taper fiber was fabricated in three ways , single fusion-stretching (SFS), multiple fusions with single stretching (MFSS) and multiple fusion(MF).The single fusion-stretching is that fabricating a taper structure by using a stretching fixture and single fusion. The Fig.1 shows the optical fiber fusion and stretching system. The stretching fixture is homemade to control the stretching distance which relative to the size of the taper waist. The taper structure is fabricated by SFS that causing the core and cladding scaling down and a small part of modified, for the reason that transmission loss is smaller and the reaction of the environment is less. The taper structure made by multiple fusions is that fabricated in multiple arc discharge without stretching. Making the fiber structure is subjected to high temperature and collapse arc form a tapered structure. The taper structure is fabricated by MFSS that using single stretching and multiple fusions. The process can control the taper waist and taper angle by adjusting the ratio of stretching and fusion. In this study, we focus on taper fiber fabricated by the MFSS. Order to avoid the taper fiber break in the production process, we fix the taper waist D is 40μm. The test of refractive index (RI) divided into two parts to discuss. First part is the FLMI sensor with the taper fabricated by MFSS. The second part is cascading a PMF to FLMI sensor with taper fiber fabricated by MFSS. Fig 2(a)-(b) are the illustration of the experimental setup for refractive index (RI) sensing measurements with the FLMI. The basic measurement setup is two ends of the taper fiber were connected to the two outputs of the coupler and C-band light source (Shenzhen Hoyatek, ASE broadband light source) and optical spectrum analyzer (OSA,YOKOGAWA- AQ6370Z) connected with the two inputs of the coupler to measurement. The Fig 2 (a)-(b) present the two measurement configurations. The main difference between two configurations is the cascading the PMF to the sensor. In the sensor region, the fiber taper was fixed with the circular fiber holder and also keep the test solution stable. To test the FLMI sensor with fiber taper, eight sucrose solution with different concentrations are used as the test samples.

1489

For the configuration (a), as shown in Fig. 2(a), the RI values of the samples were changed from 1.333 to 1.337 and the interval is 0.001. According to the refractive index of the sucrose solution of 1.333, 1.334, 1.335 and 1.337 corresponding concentration are 0%, 0.9%, 1.6%, 2.3% ,3%, respectively. For another configuration, as shown in Fig. 2(b), the RI values of the samples were changed from 1.333 to 1.341 corresponding to the concentrations 0%, 1.4%, 2.8%,4.2% and 5.5%, and the interval is 0.002. After each measurement, the sensing area was washed carefully three times with Di-water, then using blowing balloons to blow-dry.

1490

3. Results and Discussion The taper structure manufactured by MFSS process that shows in Fig 3. The taper structure is treated as a sensing area and connected to a coupler to form a fiber loop mirror interferometer (FLMI) sensor. The tapers in FLMI sensor were manufactured by MFSS process with taper waist D of 40μm for Refractive index measurements. Fig 4 shows the spectral of the configuration (a) of the RI measurement. So we can see the FLMI produce significant interference. Due to the compact spectrum dip and less local wavelength dip shift of interference spectrum of configuration (a), the global spectrum shift, the shift of envelope of interference spectrum, for interference spectrum of configuration (a) is relatively distinct and analyzed. Fig. 5 shows global spectrum shift for the interference spectrum of configuration (a) with the fiber taper fabricated by MFSS process. The global spectrum shift of the interference spectrum of configuration (a) is obviously significant. Table 1 shows the RI measurement results of global spectrum shift of the interference spectrum of configuration (a), including slope (sensitivity), and R2 (determination of coefficient) of linear and quadratic curve fitting. The relationship between RI and wavelength shift is closer to quadratic model. Fig 6 shows the interference spectrum of the configuration (b) for the RI measurement. A PMF cascaded to the sensing area that was used as the reference element. The PMF maintained the polarization of the incident light, offered a stronger and more stable polarization for light signal and improved the effect of interference pattern. Table 2 shows the RI measurement results of global spectrum shift of the interference spectrum of configuration (a), including slope (sensitivity), and R2 (determination of coefficient) of linear and quadratic curve fitting. The relationship between RI and spectrum shift is also closer to quadratic model. Finally, the results for two configurations are compared in RI range of 1.333 to 1.337 RIU. The best sensing sensitivity is 3423.71 nm/RIU and achieved by FLMI sensor of taper fiber fabricated by MFSS process with taper waist of 40 μm.

1491

In the results of configuration (b) for RI measurements, the best sensing sensitivity is -77.17 nm/RIU worse than that of configuration (a), however the R2 of linear and quadratic model are averagely better than configuration (a). It may be caused by the extinction ratio of interference pattern is superior. It indicated that the PMF improved the interference pattern with large extinction ratio and better R2 that means stable sensitivity.

1492

1493

4. Conclusion In this study, we report a method to fabricate FLMI by using fusion splicer with a designed stretching mechanism. The FLMI combined a taper fiber and FLM structure to detect RI changes in surrounding was studied. The RI measurement results for two configurations are compared in RI range of 1.333 to 1.337 RIU. The global spectrum shift of the FLMI without PMF of 40 μm taper waist shows the best sensing sensitivity of -3423.17nm/RIU. Compare to FLMI with PMF, the best sensing sensitivity is degraded to -77.17 nm/RIU, however, the R2 of linear and quadratic model have been improved averagely. It may be attributable to the superior extinction ration of interference spectrum. References [1] Ma, Lin, et al. "Fiber-optic temperature sensor based on a thinner no-core fiber." Optik-International Journal for Light and Electron Optics 126.9 (2015): 1044-1046. [2]Di, Haiting, Ying Xin, and Suping Sun. "Electric current measurement using fiber-optic curvature sensor." Optics and Lasers in Engineering 77 (2016): 26-30. [3] Zhaobing Tian,, ―Refractive Index Sensing With Mach–Zehnder Interferometer Based on Concatenating Two Single-Mode Fiber Tapers‖, IEEE, 2008. [4] De-Wen Duan, ―Fabrication and modeling of uniform-waist single-mode tapered optical fiber sensors‖, Optics Communications Volume 284 5311–5314, 2011 [5]Pilla, Pierluigi, et al. "Sensitivity Characteristics Tuning in Tapered Long-Period Gratings by Nanocoatings." Photonics Technology Letters, IEEE 19.19 (2007): 1517-1519. [6]O. Frazão, J. M. Baptista, and J. L. Santos, ―Recent advances in high-birefringence fiber loop mirror sensors,‖ J. Sensors, vol. 7, pp.2970–2983, 2007. [7]Fu, H. Y., et al. "Multiplexing of polarization-maintaining photonic crystal fiber based Sagnac interferometric sensors." Optics express 17.21 (2009): 18501-18512. [8]M. I. Zibaii, O. Frazão, H. Latifi, and P.A.S.Jorge, (2011). Controlling the Sensitivity of Refractive Index Measurement Using a Tapered Fiber Loop Mirror. IEEE Photo. Tech. Lett. 23, 1219-1221.

1494

[9] Shao, L. Y., Luo, Y., Zhang, Z., Zou, X., Luo, B., Pan, W., & Yan, L. (2015). Sensitivity-enhanced temperature sensor with cascaded fiber optic Sagnac interferometers based on Vernier-effect. Optics Communications, 336, 73-76. [10] Yi-Cheng Hsu, Hsin-Wen Wang, Wei-Jen Cheng, Shin-Hong Lin, ―All-fiber Mach-Zehnder Interferometer by Combined Single Fusion-stretching and Multiple Fusion Process‖, Taiwan, 2013-SAT-P0202-P031, December, 2013[OPTIC 2013].

1495

ACEAIT-7461 Soft Computing Approaches to Pneumatic Control System: An Application to Suspension Test Bench Hsin-Han Chianga,, Lian-Wang Leeb, Tsu-Tian Leec* a Department of Electrical Engineering, Fu Jen Catholic University, Taiwan b Department of Mechanical Engineering, Lunghwa University of Science and Technology, Taiwan c Department of Electrical Engineering, Tamkang University, Taiwan *Corresponding Author: [email protected] Abstract In this paper, the design and implementation of a soft-computing-based position controller of a pneumatic servo system, applied in the suspension test bench, is presented. Since the significant nonlinearities in pneumatic cylinder actuators are the couplings majorly between pressure and flow rate, the valve nonlinearities and the cylinder friction, the big obstacle raises in the fast and precise motion control of pneumatic actuators. To this end, a fuzzy controller is designed with a parameters self-tuner scheme by applying the neural networks (NN). To be against parametric uncertainty, load variation, and disturbances in different road conditions, the required parameters of fuzzy controller under different operation conditions can be optimized by genetic algorithms (GA). Then, these collected data are used to train the NN model as the self-tuner to adjust the controller parameters of fuzzy controller so that the good tracking control performance can be remained. Experimental studies validate the effectiveness and robustness of our approach under different tracking profiles. Keyword: Pneumatic cylinder, servo system, tracking control, soft computing, vehicle suspension. 1. Introduction For many industry automation projects, speed, reliability, repeatability together with simple structure and operation, easy maintenance, and low component cost are the most sought-after characteristics, which just happen to be among the virtues of pneumatic actuators. Compared to conventional electric and hydraulic actuators, pneumatic actuators have a high degree of inherent compliance due to the compressibility of air [1]. However, conventional pneumatic systems are also known to have poor dynamics coming from the highly nonlinearity and the large compressibility of air. Since establishing an accurate dynamic model for a pneumatic servo system in designing a model-based controller is hard, the model-free neural network and fuzzy control methods have been adopted in practice. In [2], fuzzy reasoning and the Newton-Raphson method are applied

1496

to be the learning methods to obtain good positioning accuracy and fast response. Other fuzzy practical control applications for servo pneumatics can also be found in [3-5]. In [6], the inaccuracies of the pneumatic components are compensated by a hybrid fuzzy logic algorithm which incorporates additional fuzzy pulse width modulation and small deviation control logics, so that good accuracy and repeatability at all stroke positions can be achieved. In addition, a linear quadratic integral (LQI) control method is presented for pneumatic cylinders by using an unsupervised neural network [7]. In [8], a multi-layer neural network is used to present the inverse of the conventionally controlled pneumatic positioning system that can handle the closed-loop dynamics to achieve a high tracking performance. Among the aforementioned research, high control performance in point-to-point positioning can be facilitated. However, some challenges still exist in the dynamic tracking control of pneumatic servo system while facing the uncertainties and complexities due to significant friction and payload variations. In this paper, the design and implementation of a soft-computing-based position controller of a pneumatic cylinder actuator, applied in the suspension test bench, is presented. Since the significant nonlinearities in pneumatic cylinder actuators are the couplings majorly between pressure and flow rate, the valve nonlinearities and the cylinder friction, the big obstacle raises in the fast and precise motion control of pneumatic actuators. To this end, a fuzzy controller is designed with a parameters self-tuner scheme by applying the neural networks (NN). To be against parametric uncertainty, load variation, and disturbances in different road conditions, the required parameters of fuzzy controller under different operation conditions can be optimized by genetic algorithms (GA). Then, these collected data are used to train the NN model as the self-tuner to adjust the controller parameters of fuzzy controller so that the good tracking control performance can be remained. Experimental studies validate the effectiveness and robustness of our approach under different tracking profiles.

1497

2. Scheme of the Utilized Pneumatic Servo System A schematic diagram of the utilized pneumatic servo system is shown in Fig. 1. The pneumatic cylinder, made by Mindman, is a double acting cylinder with single piston rod. To measure the position information of piston, a linear incremental encoder (PBS-HR type) made by GIVI MISURE is equipped with the pneumatic servo system with the measurement resolution of 5 m. The associated velocity of piston can be estimated by digitally differentiating the position signal using backward differencing. The chamber pressures are measured by two pressure sensors. The measured analog signals from pressure sensors and encoder are transferred by means of a 16-bit A/D interface module inserted in the embedded controller (cRIO-9074) made by National Instruments. In addition, a 16-bit D/A interface module is inserted in the embedded controller to supply the driving voltage for the servo valve according the digital control signals output from the embedded controller. LabVIEW program language is used to program the designed control algorithm and realize measurement in combination with the driving and monitoring programs of inserted interface modules. The pneumatic servo pressure valve is an open-center proportional type made by Festo (MPYE-5-3/8), and it can be controlled by a voltage signal between 0 and 10 V. The preamp is used to amplify and bias the control signal from the range 5 V to the range 0-10 V required for the pressure valve. As mentioned above, some important parameters for the pneumatic servo system are given as follows: Displacement range: 0‒150 mm; Maximum pressure: 6 bar; Input voltage range: 0‒10 V. A photograph of the pneumatic servo system set up in the suspension test bench is shown in Fig. 2.

1498

In this study, an intelligent control algorithm based on soft computing techniques is presented for the position tracking problem of a pneumatic servo positioning system, with the purpose of arriving at the set point quickly, stably, accurately and without great overshoot. Without loss of generality, some supposes for the proposed pneumatic control design are given as follows: a. For positioning control of the pneumatic cylinder, the reference zero is defined at the bottom of the cylinder. b. If the set point r(t) is greater than the real piston position y(t), then the output error e(t) is c. d.

positive. The pressure difference between two cylinder chambers is positive if the pressure of the chamber B is greater than the pressure of the chamber A. To move the piston from the bottom to the top, the pressure of the chamber B controlled by the valve must be greater than the sum of friction and the count force produced by the pressure of the chamber A of the cylinder. 3. Control Design

3.1 Fuzzy controller design In this study, a fuzzy controller is firstly designed to force piston position y(t) to track a given bounded reference signal r(t). The structure of fuzzy controller basically comprises fuzzification, control rule base, fuzzy inference engine, and defuzzification, as shown in Fig. 3.

1499

The designed fuzzy controller uses the error e(t) and the change-of-error e(t ) as fuzzy input variables regardless of the complexity of pneumatic cylinder. If each input variable is defined by five membership functions, then 25 fuzzy rules would be required in the fuzzy rule table. A drawback would result from building many fuzzy rules, thereby increasing the computation load for the implementation loop. Since the fuzzy rule base is constructed in a two-dimensional (2-D) space for using the error and error change phase-plane, most 2-D fuzzy rule bases have the so-called skew-symmetric property [9]. Thus, given e(t) and e(t ) , the switching line that represents the main hyperplane of the 2-D fuzzy rule table can be defined as a sliding manifold

where the slope  determines the convergence rate to the origin when the sliding mode occurs. In this way, the original fuzzy inputs of the error and error change can be replaced by one signed distance d, which is defined as the perpendicular distance from an operating point to the projection point on the switching line s. This new fuzzy input of the signed distance can be calculated as

The new fuzzy input ds involves position error and error change with a proportional gain and a derivative gain. Hence, a fuzzy rule table can be established on a 1-D space and the control

1500

action u is determined by ds only, and the associated fuzzy sets and linguistic variables can be defined as follows:

In this study, trapezoidal- and triangular-shaped fuzzy sets are applied to the fuzzy input and output variable, as shown in Fig. 4.

The linguistic labels use to describe the fuzzy sets are ‗negative big‘ (NB), ‗negative small‘ (NS), ‗zero‘ (ZE), ‗positive small‘ (PS), and ‗positive big‘ (PB). Table 1 shows the fuzzy rule table and the associated fuzzy rule form can be interpreted as:

The fuzzy rules in the 1-D rule table are easier to be understood and expressed from the expert experience and knowledge. Besides, the total number of fuzzy rules for the speed controller is only five, which is much fewer than the 25 fuzzy rules of the 2-D fuzzy control, such that generating and tuning rules in the initial design stage are much easier. In the fuzzy inference method, this study chooses Max-Product to determine the overall membership degree of the control action. The defuzzification operation applies the weight center method.

1501

Since the pneumatic cylinder actuator exhibits highly nonlinear characteristics which stem significantly from the complicity of friction and the nonlinearity of valves, the control performance readily gets worse while the piston changes the moving direction. More or less steady tracking error will be produced if without friction compensation. Therefore, the feedforward controller is used to take account of the working conditions of our pneumatic servo system. The input for the designed feedforward controller is the multiplication between the frequency and the amplitude of reference position signal. When the reference trajectory of piston moves up, the friction compensation adopts the control parameter UpGain; otherwise, the control parameter DownGain is adopted while piston moves down. Note that the value for UpGain and DownGain is different due to the friction varies with the different motions of piston as well as with the pressure of the cylinder chambers and the period of piston staying still. 3.2 Neural network for self-tuning scheme Since the pneumatic cylinder functions as an actuator for the various road profiles, the adaptive characteristics in the positioning control loop is required. For different reference road signals (which includes offset, amplitude, and frequency), vehicle mass, and mass flow delivered by servovalve, an controller parameters tuning scheme is proposed to adjust the fuzzy controller parameter  and the UpGain and DownGain of the feedforward controller. To this end, NN is applied to automatically adjust these controller parameters according to different operation conditions because it can potentially accomplish this task. The designed structure of the NN is shown in Fig. 5. The proposed NN is a fully connected multilayer perceptron (MLP) with five inputs and two hidden layers L1 and L2. L1 and L2 both have 10 tansig neurons, and output layer has three linear neurons.

1502

To incorporate the NN in the control system implementation, the training is conducted offline and the results are saved. By doing so, the calculation time of NN can be reduced, and the self-tuning scheme in the control process can achieve the real-time performance. The data to train the NN is initially obtained by using heuristic search method. The pneumatic control process using the fuzzy controller and the feedforward controller is operated at some arbitrary operating point, and the manually tuning process is activated to find the proper control gains, namely, , UpGain, and DownGain. This process needs to be repeated for some operating points to collect an array data of input and output data in the positioning control. Note that these chosen operating points are distributed in the entire working space of pneumatic cylinder, and thus the feasible control parameters can be stored associated with the given operating points. Subsequently, the stored data can be used train the NN with the aid of back-propagation learning algorithm. The proposed NN-based self-tuning scheme is trained with the data obtained from experimental setup, and therefore, the effect of friction and nonlinearities existing in the cylinder and valve can be included behind the weights and biases of neurons. Hence, the good dynamic positioning response of the above fuzzy controller and feedforward controller can be offered through the self-tuning process in accordance with different operating points. 3.3 Controller parameter optimization via GA The position tracking performance under different operating points can be well maintained by using the NN-based self-tuning scheme, as mentioned above. However, the tested suspension parameters will vary due to different passive suspension components (such as spring, damper, and tire) and active actuators. Besides, if there is any variation in the pneumatic servo system, the controller gains obtained from the NN-based self-tuner will not hold good. In this paper, GA method is adopted to prompt the control system to activate the genetic search for the optimized controller parameters so that the position tracking performance can be guaranteed. To date, using GA method to optimize fuzzy system parameters have become a trend since the trial time for finding optimal parameters can be greatly reduced. Conventionally, both the antecedent and the consequent parts of fuzzy rules should be optimized, but the optimization procedure via GA will significantly increase the computational complexity. In our proposed fuzzy controller and the feedforward controller, a simple and efficient method is used to only tune the three controller gains, as denoted as , UpGain, and DownGain. Note that the parameters of membership functions of the fuzzy controller are not used in the GA optimization procedure. Instead, using GA operators to tune the parameter  has the equivalent effect on the response performance with respect to membership function parameters. Compared with the conventional GA coding approaches, the evident benefit of our method is that it is computationally efficient and less time consuming.

1503

There are two important points in GA operation. The first step is to define the fitness function. To achieve good positioning performance for pneumatic actuation, the control attempts to maximize a function J including a term which penalizes the summation of tracking errors. Taking into account that the increasing value of amplitude and frequency of the reference position signal r(t) indicates the more difficulty in the tracking control, it is decided to consider the variation of reference signal. Moreover, when the higher offset is set to the operating point for the pneumatic cylinder, the resultant control actions can be considered as positive effect in pneumatic servo control. From these viewpoints, the following fitness function is determined:

where Offset denotes the operation height position of pneumatic cylinder; Amp and Freq are the amplitude and the frequency of the reference signal, respectively; SE is the absolute value of summation tracking error within a specified period; B0 denotes a constant bias value. The second importance is the code of chromosome. The applied GA parameters in this study are given as follows: generation numbers = 50, population size = 35, crossover probability = 0.8, mutation probability = 0.05, and the number of bits per parameter = 6. The selection scheme adopts Roulette based selection, and the initial population is randomly chosen from the database that has collected the valid control gains under various operating points. 4. Experimental Results To verify the effectiveness of the proposed pneumatic control system, two types of reference input trajectories are used in this study. In addition, experiments are conducted with a mismatch among suspension systems to test the controller robustness against uncertainties and external disturbances of pneumatic cylinder. Experiments are also conducted with the pneumatic servo system in the vertical motion to study the effect of different suspension types. To evaluate the repeatability of tracking performance, each experiment is repeated five times for a given reference position signal. The experiments are performed by raising the payload to the fixed offset with 6 cm, and the pneumatic servo positioning control is then activated with the given reference trajectory signal. The first type is a sine wave with the frequency 0.5 Hz and changed amplitudes. In Fig. 6, the tracking result is plotted. As can be seen, the amplitude of the reference trajectory varies from 30 mm to 50 mm at 8 second, and the suspension type switches to the active suspension system. In this test, the designed NN-based self-tuning scheme adjusts the controller parameters so that the effective tracking performance can be maintained under the changed operating condition. Again, the second type of reference trajectory is a sine wave with the 30-mm amplitude, but the frequency is changed from 0.5 Hz to 1.5 Hz at 10 second, as shown in Fig. 7. Even though the controller parameters are adopted by NN, the tracking performance is worsened and then the

1504

GA optimization procedure activates after 15 second. Taking about 5 seconds, the proper controller parameters can be obtained. Subsequently, the effective tracking performance is resumed when the optimized parameters are used.

5. Conclusion The pneumatic servo control system by using the soft-computing technique is presented in this study. To address the problems caused by the nonlinear friction forces under different operating points, the fuzzy controller with reduced rule structure is designed and the controller parameters can be adjusted by the NN which acts a self-tuner. If the complexity of pneumatic positioning control increases that leads to invalidity of control system, the GA optimization method is applied to regain the position tracking performance by self-searching the optimal controller parameters. Finally, the experimental results show that the proposed control strategy offers effectiveness in dynamic tracking performance for pneumatic actuation as the suspension test bench.

1505

References [1] J. F. Carneiro and F. G. D. Almeida, ―A neural network based nonlinear model of a servopneumatic system,‖ Journal of Dynamic Systems, Measurement, and Control, 134 (2), 024502, 2012. [2] T. Matsui, E. Ishimoto, and M. Takawaki, ―Learning position control of a pneumatic cylinder using fuzzy reasoning,‖ Journal of Fluid Control, 20 (3), pp. 7-29, 1990. [3] H. Lu, ―Pneumatic Force Control Servo System with Fuzzy Control,‖ Proceedings of the Third JCFP International Conference, Hangzhou, China, 495-498, 1993. [4] K. Araki, Y. B. Yin, and J. B. Chen, ―High speed force control of a pneumatic asymmetric valve-controlled cylinder with a fuzzy controller,‖ Proceedings of the Fifth Triennial International Symposium on Fluid Control, Measurement and Visualization, Hayama, Japan, 485-490, 1997. [5] M. Shih and C. Hwang, ―Fuzzy PWM control the positions of a pneumatic robot cylinder by high speed solenoid valve,‖ Proceedings of the Third JHPS International Symposium, Yokohama, Japan, 277-282, 1996. [6] H. Wang, J. Mo, and N. Chen, ―Hybrid fuzzy logic algorithm for position control of pneumatic actuator with 3/2-way solenoid valves,‖ Proceedings of the Institution of Mechanical Engineers, Part C (Journal of Mechanical Engineering Science), 210 (C2), 167-176, 1996. [7] K. Katsumata, A. Fujiwara, Y. Ishida, and S. Notoyama, ―An LQI control for pneumatic cylinders using unsupervised neural network,‖ Journal of the Japan Hydraulics and Pneumatic Society, 27 (3), 403-409, 1996. [8] D. C. Gross, and K. S. Rattan, ―Feed forward MNN controller for pneumatic cylinder trajectory tracking control,‖ Proceedings of IEEE International Conference on Neural Networks, 37, 1-15, 1997. [9] H. H. Chiang, Y. L. Chen, B. F. Wu, and T. T. Lee, ―Embedded driver-assistance system using multiple sensors for safe overtaking maneuver,‖ IEEE Systems Journal, 8 (3), 681-698, 2014.

1506

ACEAIT-7506 Performance Analysis of Power Steering Oil Pump Applying Al Material Yunchan Shina, Gwanghyun Ima, Jongkuk Kimb, Honghyun Cho c* a Graduate school of Mechanical Engineering, Chosun University, South Korea E-mail: [email protected] a

Graduate school of Mechanical Engineering, Chosun University, South Korea E-mail: [email protected] b

Younghwa Inc., research development team, South Korea E-mail: [email protected]

c*

Department of Mechanical Engineering, Chosun University, South Korea E-mail: [email protected]

Abstract Oil pump can reduce the weight by using the Al material. In addition, it is possible to make a low noise, high efficiency, and high power. In this study, the analytical study on the performance characteristics of power steering oil pump was carried out theoretically by using PUMPLINX. As a result, at the 1000 RPM which is idle state, the discharge flow rate and driving torque are very low. As the rotational speed increases, the fluctuation of discharge pressure increases and the total efficiency decreases. When the rotation speed is 1000, 2000, 3000 rpm, the volumetric efficiency is 62.7%, 56.8%, 47.6% and torque efficiency is 24.2%, 22.8%, 22.5, respectively. Total efficiency is 15.16%, 12.96%, and 10.7%, respectively. Keyword: Oil pump, Discharge flow rate, Volumetric efficiency, Torque efficiency, Total efficiency 1. Introduction Recently, there is a rising interest on the problem for decreasing weight of commercial vehicle. Oil pump can reduce the weight by using the Al material. In addition, it is possible to make a low noise, high efficiency, and high power. Power steering can reduce the force needed to turn the steering wheels of automotive. Also, it is mainly equipped into the midsize and full size car in order to improve steering ability. The hydraulic device is more complicated structure than the electric device and is about 4~5 kg heavier than that. However, the hydraulic device is superior on steering ability and it can make a high power.

1507

Hydraulic power steering system consists of a pressure balanced vane pump, steering gear and pulsation damping hydraulic hoses. Vane pump supplying hydraulic in power source transports the working fluid to the discharge from suction. The chamber between a cam ring, vanes and rotor causes periodically volume change due to rotational motion of rotor. Some study of oil pump using Al material has been carried out, but it is so insufficient. Especially, there has been carried out a little study for oil pump applying hydraulic power steering and discharge pressure of high pressure. In this study, the analytical study on the performance characteristics of power steering oil pump was carried out theoretically by using PUMPLINX. The result of this study is intended to provide required basic data for designing high efficiency power steering vane pump. 2. Analysis modeling and method In this study, the performance characteristic of a power steering oil pump was analyzed theoretically by extracting the flow field of the vane pump. The vane tip gap of the simulation was used with the real vane value in order to ensure the reliable result. Grid number was formed by131,793 totally and the flow field was shown in Fig. 1.

The analytical study on the performance characteristics of a power steering oil pump was carried out theoretically by using PUMPLINX and the simulation condition was shown in Table 1. Inlet pressure is assumed as atmospheric pressure. The performance characteristic of a vane pump was analyzed according to rotational speed from 1000 to 3000 rpm in the step of 1000 rpm. Working fluid used DEXRON-3 properties of general oil.

1508

As the rotation speed increases, a local low pressure area is presented inside of inlet port. Accordingly, density change is occurred rapidly. This phenomenon is called the cavitation. Cavitation is collapsed rapidly at the outlet with high pressure which causes a lot of noise and vibration. Therefore, analysis and protection for cavitation generation is very important. In this study, Full cavitation model was used to simulate the system in order to consider one dimensional factor of cavitation. It includes some concepts like below. 1) 2) 3)

Creation and transport of vapor bubbles Fluctuations caused by turbulent of pressure and velocity Size of non-condensable gas

Creation and transport of vapor bubble are designed Raleigh plesset‘s equation (1-3). And Ce and Cc of phase change rate factor are 0.02 and 0.01, respectively[1].

Fluctuation caused by turbulent of velocity was used approximately. Fluctuation of pressure was simulated to increase pressure influencing the phase change.

Mass fraction of non-condensable gas has constant value. Mass faction, density, volume fraction of non-condensable gas is calculated by equation (6-8).

3. Results and discussion

1509

Fig. 2 shows the pressure ripple, driving torque and chamber volume according to vane angle for variable rotor speeding. Pressure ripple and torque curve are not changed when rotation speed increases from 1000 rpm to 2000 rpm, while it is rapidly changed when it increases from 2000 to 3000 rpm. It can be found, as the compression starts, the pressure increases significantly. The reason for pressure drop in middle angle of compression is because of the back flow effect at the outlet[2]. Driving torque increases where vane is located and then was decreases. This tendency is repeated periodically. The average torque increases linearly with rotation speed as the rotating speed increases. Besides, the frequency of pressure increases significantly.

Fig. 3 shows the volumetric, torque and total efficiency according to rotation speed. Volume efficiency decreases continually as rotation speed increases. It is because the cavitation in the chamber increases. In case that rotation speed is 1000, 2000, and 3000 rpm, the volume efficiency is 62.7%, 56.8%, and 47.6%, respectively. In addition, torque efficiency is 24.18%, 22.81%, 22.45%, respectively. Total efficiency is 15.16%, 12.96%, 10.7% respectively. Reduction rate of torque efficiency is relatively small. The reduction of total efficiency is mainly due to the decrease of volume efficiency by generating much cavitation.

1510

4. Conclusion The analytical study on the performance characteristics of a power steering oil pump was carried out, and the detail results obtained through this study are as follows. 1) Pressure ripple and torque curve are not changed when rotation speed increases from 1000 rpm to 2000 rpm, while it is rapidly changed when it increases from 2000 to 3000 rpm. It can be found, as the compression starts, the pressure increases significantly. The reason for pressure drop in middle angle of compression is because of the back flow effect at the outlet. 2) Volume efficiency decreases continually as rotation speed increases. It is because the cavitation in the chamber increases. In case that rotation speed is 1000, 2000, and 3000 rpm, the volume efficiency is 62.7%, 56.8%, and 47.6%, respectively. In addition, torque efficiency is 24.18%, 22.81%, 22.45%, respectively. Total efficiency is 15.16%, 12.96%, 10.7% respectively. Acknowledgment This work (Grants No. 2014-0506) was supported by Business for Academic-industrial Cooperative establishments funded Korea Small and Medium Business Administration in 2014. References [1] Brennen. C. E., 1995, Cavitation and Bubble Dynamics, Oxford University Press, 47-50. [2] Ha. J. H., 2011. Latest Hydraulic Engineering, Cheongmoongak Co. Korea, 126-144.

1511

ACEAIT-7538 Estimate of Mixed Convection Heat Transfer and Fluid Flow Characteristics of Plate-Fin Heat Sink in the Opening of the Enclosure Han-Taw Chen*, Chung Yi Wu Department of Mechanical Engineering, National Cheng Kung University, Taiwan *Corresponding author: [email protected] Abstract This study uses an inverse method and commercial software FLUENT with the experimental temperature data to obtain heat transfer and fluid flow characteristics of plate-fin heat sink placed in a rectangular enclosure with two openings. A cooling fan placed on the top of the fins is blown downward. Another fan placed on the outside of the opening is blown out. The inverse method along with the experimental temperature data is applied to estimate the heat transfer coefficient on the fins. Afterward, commercial software FLUENT along with the zero-equation turbulence model, experimental temperature data and the resulting heat transfer coefficient is used to obtain heat transfer and fluid flow characteristics of plate-fin heat sink within a rectangular enclosure with openings. The heat transfer coefficient on the fin and the air temperature and velocity distributions within the enclosure are determined. The results show that more accurate results can be obtained if the resulting heat transfer coefficient and fin temperature obtained at the selected measurement locations can be as close as possible to inverse results and the experimental temperature data, respectively. Keywords: inverse method、FLUENT、experimental data、mixed convection heat transfer characteristics、plate-fin heat sink、enclosure with openings I. Introduction Natural convection heat transfer with heat sink and extended surfaces has been widely used in engineering applications. In recent years, with the rapid development of electronic technology, promoting the heat transfer rate under the working process at the desired operating temperature may play an important role to ensure reliable operation of the electronic components. The appropriate design of the heat sink has gradually become attractive for these applications because they provide a more economical solution to the above problem.

1512

Conventional electronic cooling normally used the heat sink with the cooling fan to show the superiority in terms of unit price, weight and reliability. In order to design a practical heat sink, some criterions such as a large heat transfer rate, a low pressure drop, and a simpler structure should be considered. The fins in the heat sink are always applied to increase the heat flow per unit of basic surface. However, it is necessary to note the interactions between the local heat transfer and flow distribution between two fins in designing the heat sink. It is known that most of previous works about the effect of the fin spacing in the heat sink are limited to the experiment or the numerical analysis. From Ref. [1], the measurement of the local heat transfer coefficient on the plate fin under steady-state heat transfer conditions may be very difficult to perform because the local fin temperature and local heat flux must be required. Thus, reliability can be an important concept in engineering design. The use of reliable components enables the designers to utilize more sophisticated techniques to improve the performance of the heat sinks [2]. Thus the estimation of a more accurate heat transfer coefficient on the fin may be an important task for the device of the high-performance heat sinks. The experimental and numerical studies for the heat transfer from an array of parallel rectangular finned surfaces on a horizontal surface have been studied for a long time. Such problems may exhibit the complex three-dimensional flow and thermal fields for a wider range of fin geometry. Although, most of the existing studies provide valuable results for this problem, the results are still inconclusive, especially for the comparative experimental results. On the other hand, the available experimental data can remain very limited. Thus, a more accurate predictive scheme based on the experimental data can be needed in order to obtain the reliable heat transfer characteristics in the plate-fin heat sink (PFHS). Numerical studies and some experiments for thermal performances of the plate-fin and plate-pin-fin heat sinks have been investigated by Yu et al. [3], Kim et al. [4] and Li and Chao [5]. From Refs. [3-5], the thermal resistance of the heat sink is obtained from the ratio of the temperature difference between the highest fin base temperature and the ambient air temperature to the heat dissipation power applied on the fin base. El-Sayed et al. [6] varied the fin height, fin width, fin spacing, number of fins and the distance from the fin tip to the shroud to study the performance of a PFHS. El-Sayed et al. [6] concluded that the pressure drop decreased with increasing the fin spacing. The mean Nusselt number increased with the fin spacing. Under the assumptions of the one-dimensional heat conduction model and the constant mean heat transfer coefficient, El-Sayed et al. [6] proposed the relation between the average Nusselt number and Reynolds number. The results show that the optimum geometric parameters of the fin thickness and fin spacing respectively are 0.8 mm and 2.0 mm in the PFHS.

1513

It is known that the heat transfer between the two fins is closely related to the airflow pattern. Velayati and Yaghoubi [7] applied the finite volume method along with the SIMPLE pressure-velocity coupling algorithm to solve the three-dimensional turbulent flow and heat transfer characteristics of parallel heated rectangular plates mounted over an insulated base plate. From Ref. [7] three-dimensional complex flow characteristics within parallel bluff plates can be observed. This complex flow pattern is accompanied by fluid separation and reattachment. Sparrow et al. [8] used experimental investigation to study heat transfer and pressure drop for airflow in arrays of heat generating rectangular modules deployed along one wall of a flat rectangular duct. Elshafei [9] applied theoretical and experimental studies to investigate the effects of the duct velocity, fin density and tip-to-shroud clearance on the flow bypass, the pressure drop across a longitudinal aluminum fin array and its thermal performance. Chen et al. [10] used the inverse method together with the finite difference method, the experimental temperature data and least squares method to predict the natural convection heat transfer coefficient and fin efficiency from three fins mounted on a horizontal plate for various values of fin spacing and fin height. Their estimate of the heat transfer coefficient based on the fin base temperature is in good agreement with that obtained from the existing correlation [11]. This implies that the inverse results given by Chen et al. [10] have good accuracy and reliability. Thus, this study uses the inverse method [10] and the commercial software FLUENT [12] combined with experimental temperature data to determine the heat transfer and flow characteristics of this problem for various values of the fin spacing and air speed. In order to determine more reliable numerical results, the resulting heat transfer coefficient and fin temperatures at the selected measurement locations are compared with inverse results and experimental temperature data, respectively. It is seen in Fig. 1 that a cooling fan placed on the top of the fins is blown downward. Another fan placed on the outside of the opening is blown out. The bottom side of the enclosure has a small opening to draw in cooling air. The inverse method along with the experimental temperature data is first applied to estimate the heat transfer coefficient on the fins. II. Inverse Method The two-dimensional inverse heat conduction problem is introduced to estimate the unknown heat transfer coefficient on the fins of the plate-fin heat sink for various values of the fin height. The fin temperature at the selected measurement locations and the ambient air temperature are measured from the experimental apparatus constructed in a rectangular enclosure with two opens, as shown in Fig. 1. The inverse method together with the finite difference method, the experimental temperature data and least squares method is used to predict the heat transfer coefficient for the three vertical fins placed on a horizontal heating plate. Due to the

1514

assumption of non-uniform distribution of heat transfer coefficient, the entire fin is divided into several sub-fin regions. The heat transfer coefficient in each sub-fin region is assumed to be an unknown constant. The measurement locations and sub-fin regions is shown in Fig. 2. Under the assumption of the thin fin, the temperature gradient in the z-direction (the fin thickness) is neglected. On the other hand, the fin temperature varies only in the x and y directions. Moreover, the total surface area of the edge of the fin relative to the total surface area of the fin is small enough. This implies that the actual heat transfer rate dissipated through the fin tip is much smaller than the total heat transfer rate drawn from the fin base. Thus, the boundary conditions at the edge surface of the fins may be assumed to be insulated [10, 11]. Under the assumption of steady state and constant thermal properties, the heat conduction equation for a thin fin can be expressed as

Its corresponding boundary conditions are

and

where T denotes the fin temperature. x and y are the Cartesian coordinates. L, H and t denote the length, height and thickness of the fins, respectively. h(x,y) is the unknown heat transfer coefficient. kf is the thermal conductivity of the fin. T0 and T∞ are fin base temperature and ambient air temperature, respectively. To predict the unknown heat transfer coefficient on the fins, the whole plate fin considered is divided into N sub-fin regions. The application of the finite difference method to Eq. (1) can produce the difference equation in the kth sub-fin region as

where Nx and Ny are, respectively, the number of nodes in the x- and y-directions. ℓx and ℓy are defined as ℓx = L/(Nx-1) and ℓy = L/(Ny-1). N is the number of sub-fin region.

1515

The difference equation for the node at the boundary surface and the interface of two adjacent sub-fin regions is similar to that shown in Refs. [10]. To avoid repetition, their derivative process can be found in Refs. [10]. Rearrangement of Eq. (5) and the difference equation at the boundary surface and the interface between two adjacent sub-fin regions yields the following matrix equation.

where [K] is the global conduction matrix. [T] is the matrix representing the nodal temperatures. [F] is the force matrix. The fin temperatures at the selected measurement locations can be obtained from Eq. (6) using Gauss elimination. Inverse analysis of this study can refer to Refs. [10]. In order to avoid repetition, its detailed procedures are not shown in this manuscript. T-type thermocouple is used to measure the fin temperature at selected measurement locations. The error range of T-type thermocouple is ±0.4%. The heat transfer rate dissipated from the jth sub-fin region Qj, average heat transfer coefficient on the fins h, heat transfer coefficient based on the fin base temperature hb and total heat transfer rate dissipated from the plate fin to the ambient Q can be obtained using the average heat transfer coefficient in each sub-fin region obtained. The Qj, h, hb and Q values can be expressed as

and

where the lateral surface area of the plate fin Af is defined as Af = 2LH. Aj is the lateral surface area of of the j-th sub-fin region The heat transfer coefficient based on the fin base temperature hb can be obtained from Eq. (9) The least-squares minimization method is used to minimize the sum of the squares of the deviations between the calculated and measured fin temperatures at selected locations of the fins. The details of estimating the unknown value hj can be found in Refs. [11, 12]. In order to avoid repetition, they are not shown in Refs. [11, 12]. The computational procedures of this study are repeated until the values of |(Tjmea- Tjinv) /Tjmea | for j = 1, 2,…, N are all less than 10-6. Once the hj values for j = 1, 2,…, N are determined, the values of h and hb can be obtained from Eqs. (8) and (9). Tjmea is the experimental temperature data at the j-th measurement location. Tjcal is the obtained fin temperatures from Eq. (6) at the j-th measurement location.

1516

III. Three-dimensional numerical analysis Another major objective of this study is to use FLUENT along with the appropriate flow model to determine the heat transfer and fluid flow characteristics of the present problem. FLUENT with laminar flow, zero-equation turbulence, k-ε turbulence and RNG k-ε turbulence models is used to obtain numerical results of this study. It is found that zero equation turbulence model can be the best one for the present problem. Thus, FLUENT with zero equation turbulence model is used to obtain air velocity and temperature distributions between the fins. Only the half-section of the fins is selected to perform the calculation of commercial software. Under the assumptions of the steady-state and constant thermal properties, the three-dimensional (3-D) heat conduction equation for the thin fin is expressed as

The boundary conditions at z = 0 and z = t/2 can be expressed as

and

where x, y and z are the Cartesian coordinates. The flow between two fins is modeled by a zero-equation turbulence model. The continuity, momentum and energy equations in the ambient air region can be expressed in tensor form as

and

where xi, i = 1, 2, 3, respectively indicate x, y and z. ui, p, gj and Ta are the component of velocity, pressure, acceleration of gravity in the xj direction and the air temperature, respectively. Tc denotes the average of four temperature measurements on the top surface of the enclosure. β is the volumetric thermal expansion coefficient. δj2 is the Kronecker delta. ,νeff, cp and ka are, respectively, the density, effective kinematic viscosity, specific heat and thermal conductivity of the air. They are all assumed to be constant. νeff and ka are defined as νeff = νt + ν and ka = νtcp/Prt. and νt are the laminar and turbulent kinematic viscosities,

1517

respectively. The turbulent Prandtl number Prt is given as Prt = 0.9 when the air flows on a plane wall surface. The turbulent kinematic viscosity νt is defined as

where the mixing length ℓm and S are defined as ℓm = min(0.419d, 0.09W) and S = (2SijSij)1/2. d is the distance from the wall. W is the distance from the maximum computational domain. The mean strain rate tensor Sij is defined as Sij = (əui/əxj + əuj/əxi)/2. Boundary conditions The matching condition of the temperature and heat flux at the fin-fluid interface can be written as:

At the solid surface, no-slip conditions are specified. The fin base temperature T0 is constant. Other walls are insulated. IV. Experimental apparatus This experiment is carried out in a rectangular enclosure with 0.03 m×0.03 m openings. The length, width and height of the rectangular enclosure are Le = 0.06 m, We = 0.06 m and He = 0.08 m, respectively. The test fins with 0.1 m long, 0.04 m high and 0.001 m thick is made of stainless material AISI 304. The fin temperature at the selected measurement locations, T∞ and T0, are measured using the T-type thermocouple from this experiment. The diameter of the T-type thermocouple is about 0.13 mm. The limit of its error is ±0.4% for 0°C ≤ T ≤ 350°C. The schematic diagram of three parallel rectangular fins mounted on the top surface of a horizontal heating plate is shown in Fig. 2. In order to heat three parallel fins, a square heater with 0.08 m in length is fixed on the bottom of this plate using the adhesive tapes (Nitto Denko Co., Ltd). The test fins and horizontal plate enclosed with an insulation material are heated to about 7600 seconds using 40W heater. Four thermocouples placed in a gap between the fins and the horizontal plate and are fixed to (L/5, 0), (2L/5, 0), (3L/5, 0) and (4L/5, 0). The average of the four measured temperatures is taken as the fin base temperature T0. Contact thermal resistance between the fin and the horizontal plate is assumed to be negligible. The ambient air temperature T∞ is measured at 0.3 m above the fins. V. Results and Discussion All physical properties are evaluated at the average temperature of the fin base and ambient air temperatures. The clearance parameter Cf is defined as Cf = (He-Hf)/He. All computations are performed with Le = 0.6 m, S = 0.01 m , He = 0.08 m, Hf = 0.04 m, Cf = 1, t/L = 0.01, 0.005 ≤ t/S ≤ 0.2, kf = 14.9 W/m·K and Vo = 1.0 m/s. The emissivity of the fin measured using FT-IR

1518

spectrum 100 (Perkin Elmer Corp., Ltd) is 0.15.

Nx = 21 and Ny = 17 are performed for the

inverse method and commercial software FLUENT. The number of grid points in the z direction of the fins is taken 5, respectively.The number of grid points between the two fins is taken 21, 9 in both the x and y directions and 8 in the z direction. A coarser mesh is used in the remainder of this region to save computing time. The number of grid points outside the fins are taken 76 in both the x and z directions and 8 in the y direction. 241173 grid points are used to discretize the entire computational domain. The commercial software FLUENT [12] is applied to determine the fluid flow and heat transfer characteristics between the two fins of the plate-fin heat sink. An unstructured grid system containing a non-uniform distribution of the grid points is used to determine all of the numerical results. High density grid points between the fins and the coarse grid points in the remainder of the closed enclosure are used. The initial guess of the unknown heat transfer coefficients hj is taken as unity for the inverse method. The rectangular fin is divided into eight regions, i.e., N = 8. The eight thermocouples are fixed at the selected positions on the fin, as shown in Fig. 1. Tjnum denotes the fin temperature obtained using FLUENT [12] at the j-th measurement location. A comparison of Tj, h and hiso is shown in Table 1 for Cf = 1 and various VF values. Table 1 shows the values of h and hiso obtained using FLUENT are in good agreement with the inverse results. The resulting fin temperatures at selected measurement locations also agree with the experimental data. This implies that the present results have good accuracy and good reliability. In order to determine a more accurate numerical result, the appropriate grid points may be selected such that the numerical results of h and hiso are as close as possible to the inverse results. The difference of the experimental data and numerical fin temperature is also small at the selected measurement locations. The obtained results of h and hiso increase with increasing the VF value. Fig. 3 shows the temperature contour in the x-z plane at y = 0.015 m for VF = 1.0 m/s. It is seen that the temperature in the vicinity of the cooling air inlet is lower than that in the vicinity of the heat sink and the air outlet. Fig. 4 shows the air velocity pattern in the y-z plane at y = 0.015 m with VF = 1.0 m/s. It is observed from Fig. 4 cooling air is drawn into the enclosure. Hot air between the fins can be withdrawn from the enclosure with the opening of the fan. The center of the opening with the fan is located at y = 0.6 m. Thus, the air velocity in the vicinity of the opening of the fan cannot be observed. This implies that the flow field of the present problem is 3-D. VI. Conclusion This study uses the inverse method and FLUENT in conjunction with the experimental temperature data to determine the fluid flow and heat transfer characteristics within the fins of plate-fin heat sink. The results show that the present numerical results of h and hiso using the

1519

k-ε model are in good agreement with the present inverse results. The present inverse and numerical results of hiso also agree with those obtained from Ref. [13]. This implies that the present results are reliable. This implies that the application of the zero equation turbulence model to solve this problem is appropriate. The total number of grid points may vary with the fin height in order to obtain a more accurate result. The commercial software FLUENT along with inverse results and experimental data can be applied to obtain a more accurate fluid flow and heat transfer characteristics of plate-fin heat sink. This study is useful in electronics cooling applications.

Acknowledgment The authors gratefully acknowledge the financial support provided by the National Science Council of the Republic of China under Grant No. NSC 102-2221-E006-177-MY3. We would also like to thank Professor Chin-Hsiang Cheng at National Cheng Kung University to give us the commercial software FLUENT. References [1] R.L. Webb, Principles of Enhanced Heat Transfer, Wiley, New York, 1994, pp. 125-153. [2] Ş. Yildiz, H. Yüncü, An experimental investigation on performance of annular fins on a horizontal cylinder in free convection heat transfer, Heat Mass Transfer 40 (2004) 239-251. [3] X.L. Yu, J.M. Feng, Q.K. Feng, Q.W. Wang, Development of a plate-pin fin heat sink and its performance comparisons with a plate fin heat sink, Appl. Therm. Eng. 25 (2005) 173– 82.

1520

[4] S.J. Kim, D.-K. Kim, H.H. Oh, Comparison of fluid flow and thermal characteristics of plate-fin and pin-fin heat sinks subject to a parallel flow, Heat Transfer Eng. 29 (2008) 169–177. [5] H.Y. Li , S.M. Chao, Measurement of performance of plate-fin heat sinks with cross flow cooling, Int. J. Heat Mass Transfer 52 (2009) 2949–2955 [6] S.A. El-Sayed, S.M. Mohamed, A.M. Abdel-latif, A.E. Abouda, Investigation of turbulent heat transfer and fluid flow in longitudinal rectangular- fin arrays of different geometries and shrouded fin array, Exp. Therm. Fluid Sci. 26 (2002) 879–900. [7] E. Velayati, M. Yaghoubi, Numerical study of convective heat transfer from an array of parallel bluff plates, Int. J. Heat Fluid Flow 26 (2005) 80-91. [8] E.M. Sparrow, J.E. Niethammer, A. Chaboki, Heat transfer and pressure drop characteristics of arrays of rectangular modules encountered in electronic equipment, Int. J. Heat Mass Transfer 25 (1982) 961-973. [9] E.A.M. Elshafei, Effect of flow bypass on the performance of a shrouded longitudinal fin array, Appl. Thermal Eng. 27 (2009) 2233-2242. [10] H.T. Chen, L.S. Liu, S.K. Lee, Estimation of heat-transfer characteristics from fins mounted on a horizontal plate in natural convection, CMES: Computer Modeling in Engineering & Sciences 65 (2010) 155-178. [11] G.D. Raithby, K.G.T. Hollands, Natural convection, in W.M. Rohsenow, J.P. Hartnett, E.N. Ganic (Eds.), Handbook of Heat Transfer Fundamentals, second ed., McGraw-Hill, New York, 1985. [12] FLUENT Dynamics Software, FLUENT, Lehanon, NH, 2010.

1521

1522

ACEAIT-7614 Effect on the Boiling Characteristics in the Pool Boiling of Al2O3 Nanofluid Junggyun Hama, Gwanghyun Imb, Honghyun Choc a Graduate school of Mechanical Engineering, Chosun University, South Korea E-mail: [email protected] b

Graduate school of Mechanical Engineering, Chosun University, South Korea E-mail: [email protected] c

Department of Mechanical Engineering, Chosun University, South Korea E-mail: [email protected]

Abstract In this study, the experiment was performed in order to investigate the effect on the boiling heat transfer of the characteristic of the Al2O3 nanofluid. Al2O3 nanofluid with the concentration of 0.001~0.1vol% was manufactured by dispersion with 50 nm-Al2O3 nanopowder in water. As a result, the critical heat flux (CHF) of nanofluid increased more than water but the boiling heat transfer coefficient decreased. CHF increased up to 220% with an increase of nanofluid concentration until 0.05vol% but CHF decreased over 0.05%. Besides, its tendency equally appears in nanoparticle deposition. The contact angle of water at 30%-heat flux of CHF was 77  but 0.001%-Al2O3 nanofluid decreased by 61.2  . It was found that the deposition of nano-particle during boiling affects seriously to the boiling heat transfer. Keywords: Nanofluids, Pool Boiling, Critical heat transfer, Contact angle, Wettability 1. Introduction Boiling heat transfer, unlike other representative conduction and convection of heat transfer of cooling system, which is operated by working fluids such as water and oil, is using the latent heat. During boiling, the phase change is occurred from liquid to gas. So, boiling heat transfer has been used as one of main method in the heat exchanger such as heat exchanger of nucleate reactor, chemical reactor, electronic device, HVAC systems which generate lots heat is widely used. In particular as energy consumption in industries is increased, the important of boiling as heat transfer and cooling method is enhanced (Wang and Leon, 2010). However, thermal properties of convectional working fluid (water, oil and refrigerant etc.) is bad and the technology development of their boiling heat transfer improvement reach the limitation. Nanofluid is the fluid as heat transfer performance improved by the deposition of nano-size metallic or non-metallic particles. A lot of studied (Kim et al. 2007, Shahmoradi et al. 2013) said that the boiling of nanofluid improves the Critical heat flux(CHF).

1523

You et al.(2003) performed the pool boiling test of the Al2O3 nanofluid in condition of the plate heater of square under 19.9 kPa and reported the critical heat flux was increased about 200% by adding a little of nanoparticle. Critical heat flux was known to be affect by the contact angle change which influenced on the surface wettability. Kandilkar et al.(2001) was proposed a CHF model included the influence of the contact angle. In addition, Bang et al.(2005) was observed that the surface roughness change by coating of nanoparticle compare to before the experiment throughout the measurement of the heater surface roughness. And this phenomenon was related the CHF which was associated with nanofluid increase. We investigated effects on the boiling caused by the variation of nanofluid concentration. Also, the characteristics were investigated through Al2O3 nanofluid through the contact angles of water and Al2O3 nanofluid during boiling and the variation of heated surface before and after boiling. 2. Experiment methods 2-1. Manufacture of Al2O3 Nanofluid In this study, two-step method which dispersing the nanoparticles in the base fluid was used to prepare nanofluid. Al2O3 nanoparticle with 50 nm were made by RND Korea, used in the experiment. Fig.1 shows the SEM image of Al2O3 nano-powder used in this study. Nanofluid was made by using the ultrasonic dispersion machine(SH 750S, Sonictopia Inc.) which has the ultrasound more than 20,000kHz after Al2O3-water mixture was injected. After then, dispersion was carried out more than 3hours by using an agitator. Table. 1 shows the amount of water and Al2O3 nanoparticle for making nanofluid with in the range of 0.001 to 0.05%.

1524

2-2. Experimental apparatus and methods Fig. 2 shows the Al2O3 nanofluids pool boiling experiment apparatus which consists of boiling chamber, preheating device, heater, condenser, and the data processing device. The chamber was filled by fluid, it was made of stainless steel with 220×220×330 mm3. The wall of chamber part was installed with two glasses in order to check the boiling phenomenon. By packing on the top and bottom of chamber, the leakage was prevented perfectly. Condenser was installed at the top in chamber, and inside pressure was measured by a low pressure manometer (Setra 730Model) in the chamber. Thermocouple was installed at the bottom of chamber, which is for measuring temperature. The 2 kw-coil heater was set in order to preheat the test fluid. The temperature of experimental fluid was controlled by using PID control method. During the experiment, for generating the heat flux on the heater surface, 350 W power was supplied into cartridge heater. The K-type thermocouple for measuring the temperature was installed at the connection point from the heating surface with 0.038 mm distance. Surface roughness of the heating surface was made by Polishing and it was measured by using the micro shape measurement. Fig. 3 shows the surface roughness of the heating surface of this study. The average surface roughness of the heating surface was 177.48 nm.

1525

Full the nitrogen gas into the chamber for leakage test before the pressure change was carried out. After filling the nanofluid into the chamber, the heating of the fluid was performed in order to remove dissolved gas and cooling until saturation temperature. The heat flux of the heating surface was controlled by voltage change. When the temperature was sharply increased, that point was designated as the CHF and then the experiment was terminated. 2-3. Data processing Thermocouple was installed inside the heater and temperature of the heating surface was calculated by measuring the temperature and heat flux. Heat flux was calculated by using Eq. (1).

1526

Where, q, A, V and I are the heat flux, heater surface area, voltage, and current, respectively. As shown in Eq. (2), the boiling heat transfer coefficient was calculated by using the heat flux and superheat.

And, the surface temperature was determined by using Eq. (3).

thermocouple and the surface, thermal conductivity, heat flux, respectively. The uncertainty of heat flux and boiling heat transfer coefficient of this study was 4.09% and 9.79%, respectively. 3. Result and discussion Fig. 4 shows the boiling curve for various concentrations of Al2O3 nanofluid. At early stage of boiling, the natural convection is the primary mechanism of boiling because the phase change from liquid to vapor is not active. Therefore, the slope of boiling curve at early stage of boiling is smaller than other stages due to low boiling heat transfer coefficient. In case of water, the finishing temperature of natural convection boiling was 8.1K but that of Al2O3 nanofluid is higher than water. As the nanofluid concentration increased from 0.001 to 0.05vol%, the finishing temperature of natural convection boiling increased from 8.6 to 10.56K, but after then it is slightly decreased. Besides, the boiling curve of Al2O3 nanofluid moved to right and the CHF increased. However, as the nanofluid concentration increased from 0.001 to 0.05vol%, its tendency is more and more strong. Afterward it decreased. Theoretically, the thermal properties of nanofluid increased with a rise of nanofluid concentration. However, the improvement of thermal properties of nanofluid by the increase of nanofluid concentration has the limitation due to the limitation about stability of dispersion. In addition, as reported by many previous studies, the heated surface geometry is changed during nanofluid boiling by the deposition of nanoparticle. Therefore, the increase of CHF with an increase of nanofluid has a limitation.

1527

Fig. 5 shows the variation of CHF enhancement according to the concentration of nanofluid. As the content of the nanoparticle increased to 0.05%, the CHF increased up to 722.5 kW/m2∙K which is 220% improved than water. After that the increase of CHF was decreased. The CHF enhancement was occurred when the nanofluid was used as the working fluid compared to pure water, but the CHF enhancement had the limitation in accordance with

nanofluid concentration. Reason for the decrease about the improvement of critical heat flux over nanofluid concentration of 0.05%, is related to the deposition of nanoparticle. Fig. 6 shows the change of heated surface before and after boiling. As the concentration of Al2O3 nanofluid increased from 0.001 to 0.05%, the deposition of nanoparticle is more and more active. However, the deposition of nanoparticle is weaken after 0.05%. Its tendency is similar to the improvement of CHF. In light of these, nanofluid concentration and deposition of nanoparticle at heated surface in boiling are closely related and they affect to the boiling heat transfer coefficient and CHF seriously.

1528

Fig. 7 shows the variation about 3D-surface roughness of heated surface geometry before and after 0.1% Al2O3 nanofluid boiling. The surface roughness of heated surface is 177.5 nm before boiling and heated surface geometry is smooth across the board. However, the surface roughness of heated surface after boiling increased to 530 nm due to the deposition of nanoparticle. In general, rough surface has a higher heat transfer coefficient due to the increase of cavities which generate bubbles. However, in case of nanofluid, deposited nanoparticle during boiling makes the porous nano particle layers and they generate additional heat resistance. As them, boiling heat transfer rate is decreased. Furthermore, the deposition of nano particle change of wettability condition heated surface.

To understand the characteristics of the boiling heat transfer of nanofluid, the contact angle of bubble at heated surface during boiling was measured by using a high speed camera(I-speed 2, Olympus). However, it is difficult to measure contact angle of bubble exactly during boiling

1529

because the light penetrability of nanofluid with high concentration is not a good condition. Therefore, in this study, As shown in Fig. 8, some pictures of the contact angle of water and 0.001% - Al2O3 nanofluid during boiling were taken. Because the CHFs of water and 0.001%-Al2O3 are different in boiling, we compare with the contact angles when bubble generates at the 30% heat flux of CHFs. The contact angle of water is about 77  but 0.001%-Al2O3 nanofluid decreased as about 61.2  . It means that the wettability of 0.001%-Al2O3 nanofluid is larger than water. Chinnam et al.(2015) reported the surface tension of liquid-solid according to the increase of nanofluid concentration but the contact angle is decreased. However, the contact angle slightly changed at low concentrations nanofluid like 0.001% nanofluid. The difference of his and this study causes measurement condition. He measured the contact angle of nanofluid at clean surface but this study measured at heated surface which changed by deposition of nanoparticle during boiling. When synthesize his and this study result, it can be concluded that the porous nano particle layers generated during boiling improve wettability and improved wettability causes CHF increase.

4. Conclusion In this study, we carried out the experiment to understand the effects of variation of Al2O3 nanofluid concentration on boiling heat transfer. The CHF increased but boiling heat transfer coefficient decreased in all Al2O3 nanofluid concentration more than in water. CHF of 0.05%-Al2O3 nanofluid is increased 220% more than water. It was the maximum improvement. However, CHF decreased over 0.05%-Al2O3 nanofluid. It is due to the effect caused by the deposition of nano particle during boiling. The deposition of nano particles made some change of the heated surface roughness. The deposition of nano particle generates additional thermal resistance and causes the decrease of boiling heat transfer coefficient. In addition, it improves the surface wettability of heated surface and causes the increase of CHF.

1530

Acknowledge This research was supported by Basic Science Research {Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education(2013R1A1A2062072). 5. References Wong, K.V., Leon, O. D.(2010). Applications of nanofluids : current and future, Hindawi Publishing Corporation Advances in Mechanical Engineering, Article ID 519659. Kim, S.J., Bang I.C., Buongiorno, J.(2007). Surface wettability change during pool boiling of nanofluids and its effect on critical heat flux, Int. J. Heat Mass Transf, 50, 4105-4116. Shahmoradi, Z., Etesami, N., Esfahany, M. N. (2013) Pool boiling characteristics of nanofluid on flat plate based on heater surface analysis, Int. J. Heat Mass Transf, 47 113-120. You, S. M., Kim, J. H., Kim, K. H.(2003). Effect of nanoparticles on critical heat flux of water in pool boiling heat transfer, Applied Physics letters, 83(16), 3374-3376. Kandlikar, S. G.(2001). A Theoretical Model to Predict Pool Boiling CHF Incorporationg Effects of Contact Angle and Orientation, J. Heat Transf, 123, 1071-1079. Bang, I. C., & Chang, S. H.(2005) Boiling Heat transfer performance and phenomena of Al2O3-water nanofluids from a plain surface in a pool, Int. J. Heat Mass Transf, 47, 2407-2419. Chinnam, J. Das, D., Vajjha, R., Satti, J.(2015). Measurements of the contact angle of nanofluids and development of a new correlation, Int. Commun. Heat Mass Transf, 62, 1-12 Chinnam, J. Das, D., Vajjha, R., Satti, J.(2015). Measurement of the surface tension of nanofluids and development of a new correlation, Int. J. Therm. Sci, 98, 68-80

1531

ACEAIT-7686 Laser Beam Shaping Based on Genetic Algorithm Optimization Yi-Chin Fanga,*,Cheng-Mu Tsaib a Graduate Institute of Electrical Engineering, National Kaohsiung First Univ. of Science and Technology, Taiwan, R.O.C E-mail: [email protected] b

Graduate Institute of Precision Engineering, National Chung Hsing University, Taiwan R.O.C E-mail: [email protected] Abstract This study improves the design of an aspherical lens system (APS) in a refractive beam shaping system. A flat-top distribution of arbitrary circular size on laser spots is optimized by using Genetic Algorithm (GA) with multipoint search that is used as the optimization method, together with the optical simulation software, Light Tools, to search for ideal parameters. The optimized results show that, in comparison to a two-dimensional computational method, the one-dimensional computation for laser beam shaping can search for the optimal solution approximately twice as fast. In addition, when using a divergence type laser to design the shaping system, the optimal results show that when the laser spot decreases from 3 mm to 1.07 mm, 88% uniformity is achieved, and when the laser spot increases from 3 mm to 5.273 mm, 90% uniformity is achieved. The distances between the lenses for both systems described above are less than the thickness for the first lens. This helps to meet the design objectives of a light and slim system. Keywords: Flat top, Genetic algorithm, laser beam shaping system, laser spot size 1. Introduction A laser beam has been applied in various fields such as laser welding, laser processing, and laser medical applications, which are affected by uneven energy distributions of laser beams that can cause material damage due to localized overheating. This is because the output of laser beams are always Gaussian distribution. Therefore, laser beam distribution must be reformed before using it. In recent years, laser beam shaping technology has become an important research topic due to the increasing demand from various sectors. By using optical design, the laser beam with the Gaussian distribution can be transformed into a uniform distribution. There are many recent studies published on laser shaping technologies and methods like refractive, reflective, or diffractive optical system design. Laser shaping designs are often based on derived methods [1, 2]. To simplify the derivation and to reduce the design difficulties

1532

associated with non-spherical parameters, the distance between lenses have to be large enough. As a result, the total system length is continuously increased. If the gap between the two lenses is not significantly larger than the thickness of the first lens during the derivation process, laser shaping cannot be achieved. In addition, for ease of derivation, most laser shaping systems assume that the incident beam is parallel [3]. Not many systems are designed for laser beams with a divergence light source. Even if a divergence light source is used, most systems focus on transforming the shape of the laser spots, and are less considerate towards uniformity. In order to find the optimal system solution, a GA is used for optimization, and the proposed laser beam shaping design is based on a refractive system [4-6]. Regardless of the environment, the GA can perform appropriate adjustment according to the design concept. Even though the construction of the GA could be varied, the search is based on the concept of natural selection. Our studies are divided into five sections: the introduction of the study, the GA, different calculation methods, simulation results for the laser beam shaping system, and conclusion. 2. Genetic Algorithm A system with multiple variables in multiple modal spaces can be optimized simultaneously by using GA that can solve complexity and discontinuity problem. For deterministic optimization methods, the next searching point is always according to a point within the searching domain. These point-to-point optimizations are easily trapped in a local optimal solution. Instead of focusing on a single point, the random and adaptable, and multiple points can be considered in GA for the optimization at the same time. A more complete and thorough survey can be performed to avoid becoming trapped in a local optimal solution. There are four main processes that have be used in GA, that is, fitness function, selection mechanism, crossover (mating) operation, and mutation operation. The process flow for the GA used is shown in Fig. 1.

1533

The fitness function used in the GA is to evaluate how the individual fits the requirement. Therefore, the specifications of the system should be appropriately described in fitness functions for evaluating the individual. With higher fitness value, the better fitness ability is for individual. According to the fitness function, the GA can evaluate the individual from the population to determine the most suitable group. The specifications are uniformity, size of the laser spot, and collimation. For the optimization method, we define the target value rT before the simulation starts. We then search for fitness values (r1, r2) and gradually move toward the target. The target functions for uniformity and spot size are defined as

where Mesh is the number of grids for the detection surface. We chose 51x51. r1 is the grid value detected on the front detection surface. r2 is the grid value detected on the detection surface on the back side of the second lens. rT is the target value. Q is the weight. Since indexing is used for calculation, the closer it is to the value 1, the closer it is to the ideal value. In terms of weighting, the computation required is excessive if we perform calculation in 2-D. After evaluation and verification, Q is set to 0.0001 if we are performing the calculation in 2-D, and 0.003 if we are performing the calculation in 1-D. When we set the target value rT we perform optimization. If the targeted laser spot is a flat-top laser beam with a uniformity of 1 mm, then the incident energy is 3.5 W. Moreover, the definition of the target function for collimation is

where N is the number of all beams detected on the detection surface. In order for all the output beams to be parallel to the optical axis, we set the ideal value (cos(0o) = 1) and gradually calculate the directional cosine values for all beams. The total average value is then obtained and expressed exponentially. A value closer to 1 means the beam is more parallel to the optical axis. In order to achieve the target, multiplication is used to calculate the optimized target (obj). We then use

to express the optimized target in the form of a fitness function. A smaller value for the optimized target means a larger value for the fitness function, and hence an easier target to be selected. The selection mechanism refers to selecting individuals from a group. Parental generation refers to the selected individual. The GA is used to produce offspring. The most well-known selection method is the roulette wheel Selection proposed by Holland. Each

1534

individual is treated as a region in a wheel. The size of the region is proportional to the fitness value of the individual. A higher fitness value means a larger region represented by the individual, and a higher probability of selection. In our study, the equation used for roulette wheel selection is

where M =1,2, …, Pop_size. In the case of crossover, two individuals from the GA parental generation are randomly selected. Two new individuals are produced after exchanging data from the parents. It is mainly used to select the good genes accumulated from the previous generation with the intention that these genes will produce even better individuals. It is not necessary for all individuals from the parental generation to perform crossover. Instead, it depends on the configured crossover rate. A higher crossover rate means a higher frequency for crossover to occur. In some cases the crossover rate needs to be set according to certain circumstances, for example, genes from certain groups need to be stably passed to the next generation. In our study, crossover is randomized and the crossover rate is 0.8. The exchange rate is a random value (Rnd). The crossover rate refers to the ratio between the parent generation and the next generation. If we suppose the parent generation has a value of 100, and a crossover rate of 0.8, then 80 offspring will be produced as a result of crossover. The exchange rate refers to the ratio of gene exchange. Suppose two chromosomes A and B are selected from the parent generation; each pair of genes in the chromosomes undergo crossover according to the exchange rate. Chromosome C is produced as the next generation, as

The exchange rate is a random value between 0 and 1. If we simply rely on crossover and reproduction, the individual is not able to evolve with new features. This is because the new individuals just inherit various characteristics as a result of mixing genes from the parents. It is difficult to develop individuals with new features. This not only makes evolutionary speed slow, but also limits the results to only a few particular points. Generally speaking, if the mutation rate is too high, then the speed for convergence will be slow. The mutation rate should be set within a reasonable range. We use the multipoint random mutation method, with the mutation rate set to 0.2. When the randomly generated value (Rnd) is less than 0.2, the mutation requirement is satisfied and a new gene is randomly generated within the specified range. In general the GA involves continuous computation. Without setting termination conditions, it will continue to evolve endlessly. In our study, we set the conditions to be 100 evolution generations and 100 groups. Analysis will be performed for each generation until the

1535

search reaches the generation specified. If the results do not match with the target specified, then evolution continues, otherwise the GA is terminated. 3. Operation method The GA requires excessive computation and requires a long time for laser beam shaping optimization. We propose another optimization method to resolve this. Figure 2(a) shows the original 2D calculation method for the GA. All grid values (51x51) from the detection surface are used in the calculation. This results in a long optimization time and high computation cost. To improve this, we attempt to take advantage of the rotationally symmetric property in refractive shaping systems to convert the 2D GA calculation to a 1D calculation, as shown in Figure 2(b). In other words, we only calculate the grid values for part of the detection surface (51x1) to reduce the optimization time.

4. Simulation results The simulation experiment is divided into two parts. In the first part we compare and discuss results from different calculation methods, and in the second part we design a shaping system for divergence laser beams in order to widen the scope for laser applications. 4.1 First part When we perform optimization, different calculation methods can be used. Our study is intended to verify that in addition to accelerating the optimization speed, the system also offers good shaping results. The same non-spherical lens architecture is used. The aim is to reduce the laser spot from 5 mm to 1 mm uniform flat-top laser beam. The 1D and 2D calculation methods are used to perform optimization for laser beam shaping. 4.1.2 Performing laser shaping using the 2D calculation After performing 2D GA optimization on a non-spherical lens system, the optimal value for the fitness function is 1.547. In Figure 3 the simulated trace diagram for the laser beam shaping system is shown. One million laser beams are used in the simulation. In terms of the detected optimized results on the detection surface, Figure 4 shows the laser spot irradiation for the

1536

system. After performing a 2D GA optimization on the non-spherical lens laser beam shaping system, the largest illumination value out of the 13 detection points on the detection surface was 4.221 W/mm2, and the smallest value was 3.432 W/mm2. The uniformity is therefore 81%. According to the program, the calculated average angle for all light when passing through the detection surface was 12.139 degrees. The total amount of radiation and the size of the laser spots are obtained from values on the detection surface using macros. The total amount of radiation obtained was 3.034 W. The initial incident energy was 3.5 W. The conversion efficiency was 87%. The size of the laser spot was set to be half of the height and width of the detection surface, and its value was 0.983 mm. The calculated shrinking rate for the laser spot was 20%. The simulated values for the detection surface are listed in Table 1.

1537

4.1.3 Performing laser beam shaping using the 1D calculation After performing the 2D GA optimization on the non-spherical lens system, the optimal value for the fitness function was 1.522. In Fig. 5 the simulated trace diagram for the laser beam shaping system is shown. One million laser beams are used in the simulation. In terms of the detected optimized results on the detection surface, Figure 6 shows the laser spot irradiation for the system. After performing the 1D GA optimization on the laser beam shaping system, the largest illumination value out of the 13 detection points on the detection surface was 5.174 W/mm2, and the smallest value was 4.551 W/mm2, giving a uniformity of 88%. According to the program, the calculated average angle for all light when passing through the detection surface was 11.026 degrees. The total amount of radiation and the size of the laser spots are obtained from values on the detection surface using macros. The total amount of radiation obtained was 3.264 W. The initial incident energy was 3.5 W. The conversion efficiency was 93 %. The size of the laser spot was set to be half of the height and width of the detection surface, and its value was 0.908 mm. The calculated shrinking rate for the laser spot was 18%. The simulated values for the detection surface are listed in Table 2.

1538

4.1.4 Comparison In Fig. 7 the optimization results described above are shown. This compares the results from the 1D and 2D optimization. We compare the time used for optimizing 100 evolution generation and 100 groups. Our results show that the 2D optimization requires 70 - 80 hours, while the 1D optimization requires 30 - 40 hours. 1D optimization is almost 2 times faster than 2D optimization and reduces the computational cost. It also proves that if we use 1D optimization, we can achieve better results in less time even when the sampling regions are smaller. In the following optimization design, we are using 1D optimization for the laser beam shaping system. We compare the time used for optimizing 100 evolution generation and 100 groups. The 1D optimization is almost 2 times faster than 2D optimization and reduces the computational cost. It also proves that if we use 1D optimization, we can achieve better results in less time even when the sampling regions are smaller. In the following optimization design, we are using 1D optimization for the laser beam shaping system.

4.2 Second part We attempt to design a shaping system for a divergence laser beam source. It is hoped that the output laser beam will have uniform energy distribution. It is also hoped that we can convert the laser spot to any size. In addition, optimization is performed using the multi-point searching feature offered by the GA, so that the laser beam shaping system will be light and slim. 4.2.2 Beam expander design for the laser beam shaping system The program configures GA optimization to be performed on the dual lens non-spherical lens system when the laser spot is enlarged from 3 mm to 5 mm. The optimal value for the fitness

1539

function is 1.021. The following uses this set of values. Figure 8 shows the simulated trace diagram for the laser beam shaping system. One million laser beams are used in the simulation. In terms of the detected optimized results on the detection surface, Fig. 9 shows the laser spot irradiation for the system.

After performing GA optimization on the dual lens non-spherical lens laser beam shaping system, the largest illumination value out of the 13 detection points on the detection surface was 0.134 W/mm2, and the smallest value was 0.119 W/mm2, giving a uniformity of 90%. According to the data collected, the calculated average angle for all light when passing through the detection surface was 5.797 degrees. The total amount of radiation and the size of the laser spots are obtained from values on the detection surface using macros. The total amount of radiation obtained was 2.79 W. The initial incident energy was 3.5 W. The conversion efficiency was 80%. The size of the laser spot was set to be half of the height and width of the detection surface, and its value was 5.273 mm. The calculated shrinking rate for the laser spot was 176%. The simulated values for the detection surface are listed in Table 3.

1540

4.2.3 Narrower beam design for the laser beam shaping system The program uses the GA optimization on the dual lens non-spherical lens system when the laser spot is reduced from 3 mm to 1 mm. The optimal value for the fitness function was 1.235. The following uses this set of values. In Fig. 10 the simulated trace diagram for the laser beam shaping system is shown. One million laser beams are used in the simulation. In terms of the detected optimized results on the detection surface, Figure 11 shows the laser spot irradiation for the system. After performing GA optimization on the dual lens non-spherical lens laser beam shaping system, the largest illumination value out of the 13 detection points on the detection surface was 3.202 W/mm2, and the smallest value was 2.829 W/mm2, giving a uniformity of 88 %. According to the data collected, the calculated average angle for all light when passing through the detection surface was 3.14 degrees. The total amount of radiation and the size of the laser spots are obtained from values on the detection surface using macros. The total amount of radiation obtained was 2.759 W. The initial incident energy was 3.5 W. The conversion efficiency was 78%. The size of the laser spot was set to be half of the height and width of the detection surface, and its value was 1.076 mm. The calculated shrinking rate for the laser spot was 36%. The simulated values for the detection surface are listed in Table 4.

1541

5. Conclusion The aim of our study is to widen the scope of laser applications. A GA is used during the system design to improve the results. The following are observed. 1) For the GA method used in our study, we first need to setup a set of target values for the laser spot using the program. We then search for the fitness value and sequentially input the fitness value into the configured target value in order to obtain the optimal solution. Its advantage is that it can be used in industrial fields that require precision cutting, drilling, etc. An accurate size specification is required for follow-up work. Our study can design according to specified design specifications. This also shows the practicality of a system designed using a GA. 2) We take advantage of the rotationally symmetric property in the refractive shaping system and change the original 2D calculation to a 1D calculation to improve the optimization speed. Our results have shown that the optimization time required for the 1D calculation is about 30 40 hours, while the optimization time required for the 2D calculation is about 70 - 80 hours. This is approximately two times faster. It also proves that if we use 1D optimization, we can achieve better results in less time even when the sampling regions are smaller.

1542

When a diffused light source is used, most systems focus on transforming the shape of the laser spots, and are less considerate towards uniformity. Using a conventional derivation process to perform laser beam shaping, if the gap between the two lenses is not significantly larger than the thickness of the first lens, laser shaping cannot be achieved. In our study, we use a GA with the multi-point searching feature on the diffused light source and performed laser beam shaping. Our results have shown that when the laser spot is increased from 3 mm to 5 mm, the uniformity is 90%. When the laser spot is decreased from 3 mm to 1.076 mm, the uniformity is 88%. The distance is even smaller than the thickness of the first lens. This means the system can be both light and slim. In conclusion, our design using a GA can perform laser beam shaping on different laser beams. The GA can be used in extensive searching domains. The GA optimizes the best solutions while complying with the design specifications required. In future studies relating to optimization, the GA is expected to be particularly helpful.

[1] [2] [3] [4]

6. Reference Frieden, B. Roy (1965). Lossless Conversion of a plane laser wave to a plane wave of uniform irradiance, Appl. Optics 4 (11), 1400-1403. Hoffnagle, J. A. & Jefferson C. M. (2000). Design and performance of a refractive optical system that converts a Gaussian to a flattop beam, Appl. Optics (30), 5488-5499. Zhang, S., G. Neil & Shinn, M. (2003). Single-element laser beam shaper for uniform flat-top profiles, Opt. Express 11(16), 1942-1948. Liu, C. & Zhang S. (2008). Study of singular radius and surface boundary constraints in refractive beam shaper design, Opt. Express 16(9), 6675-6682.

[5] Scott, P. W. & Southwell, W. H. (1981). ―Reflective optics for irradiance redistribution of laser beams: design,‖ Appl. Optics 20(9), 1606-1610. [6] Jiang, W. & Shealy, D. L.(2000) Development and Testing of a Refractive Laser Beam Shaping System, SPIE , (4095), 165-175.

1543

ACEAIT-7297 An Artificial Intelligence Method to Investigate the Locations of Trader‟s Stall for Periodical Products Yi-Chih Hsieh a,*, Pei-Ju Lee b, Yung-Cheng Lee c a Department of Industrial Management, National Formosa University, Taiwan E-mail: [email protected] b

Department of Information Management, National Chung Cheng University, Taiwan E-mai: [email protected] c

Department of Security Technology and Management, WuFeng University, Taiwan E-mail: [email protected]

1. Background and Goal A special market system in which people gather to trade with lowest level of trading function still exists only periodically today. Such periodic marketplaces play an important role in the integration of the regional marketing system and the spatial organization of economic activities. Unlike the classical traveling salesman problem (TSP), the mobile traders are not required to visit all marketplaces but even stay at the same marketplace during the entire periodic intervals. In addition, customers may travel to the neighboring city which is visited by mobile trader to buy periodical products such as toilet papers and fruits. In fact, periodicity is an advantage to such traders because their diverse economic roles preclude them from full time trading. However, the hardest decision problem encountered by these traders is to optimally decide how many and which marketplaces (locations) to visit and when it is more profitable to be itinerant or fixed. This study aims to investigate the location problem of trader‘s stall for periodical products such that his total net profit is maximized by selling periodical products. 2. Methods The main steps of the proposed immune based algorithm (IA) are as follows: Step 1. Step 2. Step 3. Step 4. Step 5.

Generate an initial population of strings randomly. Evaluate each individual. Select the best s individuals with best fitness values. Clone the best s individuals (antibodies) selected in Step 3. The set of the clones in Step 4 will suffer the genetic operation process.

1544

Step 6.

Calculate the new fitness values of these new individuals from Step 5. Update

individuals in the memory set. While the memory set is updated, the individuals will be eliminated while their structures are too similar. Step 7. Check the stopping criterion, if not stop then go to Step 2. Otherwise go to next step. Step 8. Stop. Report the optimal or near optimal solution(s) from the memory set. 3. Results and Conclusion The test problem with 13 districts, namely 1 to 13, is based upon Taoyuan, Taiwan. We set the following values of parameters for IA: (crossover rate, mutation rate)=(PC, Pm)=(0.96, 0.02), Population size=200, maximal iteration number=500. Numerical results of 50 experiments with

=0.1 (the exponent on distance) of IA for various T(planning horizon) and I(demand interval for the product) are reported in Table 1. From Table 1, we observe that: (a) There exists periodical cycles for the location of trader. For example, when T=7, I=2, the location of trader is 21 21 21 2 (with cycle 21). (b) With the increase of I, the periodical cycle of the location for trader increases. For example, when T=14, I=2, the location of trader is 12 12 12 12 12 12 12 (with cycle 12). However, when T=14, I=4, the location of trader is 1234 1234 1234 12 (with cycle 1234). (c) The standard deviation for the objective value (profit) of 50 experiments by IA is pretty small. For example, when T=7, I=2, the /=0.000247. (d) The mean CPU time of 50 experiments by IA is pretty small. This further implies that the adopted IA performs very efficiently. For example, when T=14, I=2, the mean CPU time by IA is 51.11 seconds.

Keywords: Optimization, Location, Algorithm, Profit

1545

ISFAS-1470 Characteristic and Efficiency Analysis of Storage-Type Solar PV/Thermal Hybrid System Feng-Chin Tsai1, Jen-Ching Huang1, Xin-Zhi Huang2 1 Department of Mechanical Engineering, Tungnan University, Taiwan 2 Institute of Mechanical Engineering, Tungnan University, Taiwan Email: [email protected] Abstract Storage-type solar PV/thermal hybrid system added a water tank below the PV panel directly is to cool the surface of PV chip and reuse the waste heat. Solar test equipment was set up to adjust irradiation amount of halogen lamp to light on solar PV board and measure the I-V curves of the PV panel simultaneously. Surface temperature of PV panel and water temperature of water tank are measured respectively by infrared thermal imager and thermocouples. By altering the irradiation of 600~900W/m2 at a 30∘inclination angle of PV panel, the experience data of storage-type solar PV/thermal hybrid system can be got, including the surface temperature of solar PV panel, the hot water temperature and the variables of I-V curves. Total efficiency transformed by photovoltaic efficiency and hot water efficiency of PV/thermal hybrid system can be analyzed. The results show that total efficiency of storage-type solar PV/thermal hybrid system is more excellent as compared to the pure photovoltaic efficiency. The hot water efficiency of PV/thermal hybrid system from recycling waste heat gets 38~40%. If the electric/heat energy conversion efficiency is about 0.38, total efficiency transformed obtains more efficiency about 14~15% from the recycling the waste heat. The total efficiency of storage-type solar PV/thermal hybrid system is almost 4 times than photovoltaic efficiency in experimental surrounding. Keywords: PV/thermal, storage-type, total efficiency Introducation Background Solar PV panel has been considered as the best renewable energy that is very developable and ecological in the future. The related application technologies matures gradually. If generating power of solar energy can promoted and applied directly to meet the extra electricity consumption in summer, power consumption peak will be soothed. It can make a positive contribution to domestic energy demand and electricity power supply. However, when solar PV panel is shined, the surface temperature of solar PV panel will increase. With the impact of heat, PV efficiency drops and lifetime cuts down also.

1546

Take advantage of V-shaped solar PV panel absorbing the reflected lights to rise its efficiency, Solenki et al. [1~2] set up a V-trough PV module to let the reflected lights fall on the PV panel inside the V-trough PV module with the V-shaped glass. The ditches inside the module act as cooling fins to make heat convection to cool down the system. The efficiency 10.57% in normal solar PV panel can be measured at the sunlight amount of 750W/m2, while the efficiency 17.1% in V-trough PV module can be presented. Akbarzadeh and Wadowski [3] set up a cooling PV chip system, which focused sunlight with its reflector and cool down the surface temperature of the chip with cooling fins and heat pipes in the back of the PV chip. There are R-22 refrigerant in the heat pipe to bring the heat inside the PV chip away. We can see that the surface temperature of the chip decreased from 84℃ to 46℃, and the output power increased from 10.6W to 20.6W. Many scholars referred to the research that solar PV panel can be cooled down with water to improve its efficiency [4~6]. Krauter [7] set a water tank above PV panel with multiple water pipes releasing water through the upper surface of the panel to cool it down. The surface temperature of the panel decreased obviously. Combining solar PV panel with solar heat collector is another concept to improve the overall efficiency [8~10]. Hollick [8] laid solar PV panels above a wave board painted with dark-color that could easily to absorb irradiation wavelength. The wave board absorbed the irradiation energy as sun shined. The PV panel above the board could converse sunlight to electric, and the water flowing under the wave board could absorb the heat from the PV panel. This is the combination of solar photoelectric conversion by PV panel and solar hot water by heat conversion, what we called PV/thermal (or PV/T) system. Tonui and Tripanagnostopoulos [9] made use of air to cool PV panel and indirectly transferred heat to liquid, constructing a PV/thermal solar energy collector. Kumar and Tiwari [10] set a solar PV panel beside the stored solar water heating tank to form a PV/thermal system. The shaped space in the back of PV panel could make the solar PV panel lead fluid passing through the space to cool the panel and recycle the heat of the panel. The heat was brought back to the stored solar hot water system. This stored PV/thermal solar system could improve the overall efficiency to more than 52%. This research is to recycle the waste heat from solar PV panel during the photoelectric conversion process. Normal solar hot water absorbs irradiation energy with heat collector, and converses solar energy to heat the water in the sink. Solar PV test equipment has constructed to adjust the irradiation amount of halogen light.

1547

When the light irradiate on the solar PV panel, voltage and current of PV panel can be measured with I-V curve measurer and drawn an I-V curve to analyze the efficiencies. This experiment was carried out with different irradiation of 600~900W/m2 and 30∘inclination angel of PV/thermal system to obtain the total efficiency and the surface temperature of the solar PV panel. During the photoelectric conversion process of solar PV panel, the waste heat on the surface will affect conversion efficiency. A storage-type solar PV/thermal hybrid system added directly a storage tank below the PV panel directly to cool the surface of PV panel was designed to distinguish from the separation-type solar PV/thermal hybrid system. As the solar PV panel generates waste heat, cooling water in the tank will take it away. With this, the solar PV panel efficiency and its lifetime can be maintained, and the waste heat can be reused. Total efficiency of a storage-type solar PV/thermal hybrid system combining photovoltaic efficiency of PV panel and hot water efficiency of solar hot water system can be obtained. Methods PV chip is always tested its voltage and electric current as absorbing different irradiation to form a I-V curve and get the efficiency. As shown in figure 2-1, the numbers of voltage and electric current were measured at different loads to form the I-V curve. There are five important parameters in the I-V curve. Isc and Voc are the short-circuit current and open-circuit voltage. The point Pmax shows the maximun power at the I-V curve, and Imax and Vmax are defined as the electric current and voltage at the Pmax position.

Figure 2-2 shows the correlation chart between I-V curves of solar PV chip with different irradiation amount. I-V curve of PV panel promotes with increases of irradiation. The short-circuit current Isc and open-circuit voltage Voc are also increase with increases of irradiation. Figure 2-3 shows the correlation chart between I-V curves and surface temperature of solar PV chip at the same irradiation. I-V curve of PV panel reduces with the increases of surface temperature at the same sunlight amount. Then, conversion efficiency of PV panel

1548

reduces also with the increases of surface temperature. It reveals that the efficiency of PV chip is affected obviously by the surface temperature of solar PV.

The idea about the surface heat generated at the irradiating solar PV panel can be cooled, and the waste heat can be used to become hot water, forms the structures of PV/thermal hybrid application. This hybrid system can not only cool the surface of the solar PV panel to lengthen the lifetime, but can recycle the waste heat without pump pressurization. Then the solar energy can be applied the most effectively. The experimental data of PV/thermal hybrid system at different irradiation can be measured, including I-V curve, surface temperature distribution of PV panel, and water temperature inside the water tank below the PV panel. Photovoltaic efficiency and hot water efficiency can be

1549

calculated by the efficiency equations. And, the best design parameters to promote the total efficiency of the PV/thermal hybrid system can be got from these experiments. The various efficiency in this research can be defined and calculated as follows. Photovoltaic efficiency of the PV panel, ηe , can be defined as:

I(t) is the irradiation amount, and Ac is the surface area of the PV panel. Hot water efficiency of the solar PV/thermal hybrid system, ηw, can be defined as:

mw and Cw are the weight and the specific heat of the water in the tank. Δt is the interval of each experiment. Ti and Tt are the average water temperature of the tank before and after each experiment. The total photo-electric conversion efficiency, ηe,total, can be defined as:

ηc is the electric/heat energy conversion that is changed into the PV panel electric/heat conversion efficiency. According report to Ji et al. [11] and Huang [12], heat energy has lower conversion energy level than electric energy, and its ηc is about 0.38. Namely, the hot water efficiency is about 38% of the photovoltaic efficiency. That is,

1550

The structure of storage-type solar PV/thermal hybrid system is shown in figure 2-4. Thermal pad is set for conducting the waste heat from the PV panel to storage water tank. Experimental setup shows as figure 2-5. I-V curve is obtained by solar I-V measuring instrument of accuracy ±1% reading value. Light source of halogen lamp is controlled by the angle inclination adjustment. Thermocouple of accuracy ±0.1℃ and infrared thermal imager of accuracy ±2% are used to measure the water temperature of water tank and the surface temperature of PV panel respectively. Water tank is set below the PV panel to receive the waste heat directly, as shown in figures. Results  Characteristic analysis of storage-type solar PV/thermal hybrid system The inclination angel of the solar PV panel is set to be 30°. The irradiation amount of 600, 700, 800 and 900 W/m2 used halogen light to simulate sunshine can be adjusted by tuner and measured by irradiation meter. Irradiation lighting for 60 minutes each experiment. I-V curves, as shown in figure 3-1, were obtained by I-V measuring instrument at the irradiation adjusted 600~900 W/m2. Closed-circuit current stably grows with the increase of irradiation amount, however, the open-circuit voltage is hold at a certain extent in spite of the different irradiation. The data of I-V curves show electric current drops to zero dramatically as output voltage is close to the open-circuit voltage at each irradiation amount. Short-circuit current is varied at different irradiation amount from 0.068A to 0.179A. Electric current drops greatly at output voltage about 15V. As output voltage rises to the open-circuit voltage about 19V, electric current drops to 0A.

1551

The correlation between output power and voltage, as shown in figure 3-2, were obtained also by I-V measuring instrument. Photovoltaic power stably grows with the increase of irradiation amount. The data of P-V curves show photovoltaic power increases to the maximum value and drops to zero dramatically as output voltage is close to the open-circuit voltage at each irradiation amount. The point of maximum power is located at the output voltage about 15V. Maximum power is varied at different irradiation amount from 0.7W to 2.2W. Surface temperature distribution of PV panel in the storage-type solar PV/thermal hybrid system was imaged by infrared thermal imager. Imaging pictures of surface temperature of PV panel was captured in the last experiment period with different irradiation of 600~900 W/m2, as shown in figure 3-3. The highest surface temperature located on the upper surface of the PV panel is 45.2~58.4 °C. As we know, surface temperature of the PV panel is almost about 60~85 °C generally when sun shines outside without cooling PV panel. The temperature of PV panel

1552

cooled in the storage-type solar PV/thermal hybrid system decreases very much obviously. The result makes sure that photovoltaic efficiency and lifetime of PV panel can maintain in the storage-type solar PV/thermal hybrid system.



Efficiency analysis of storage-type solar PV/thermal hybrid system Water temperature in the tank of storage-type PV/thermal hybrid system at irradiation lighting for 60 minutes each experiment is listed in table 3-1. The water temperature T1, T2 and T3 are measured by thermocouples respectively on top, middle and bottom of the storage tank. Tavg is the average value of T1, T2 and T3 water temperature. Average water temperatures Tavg increase with the increase of irradiation amount steadily. The data show the water temperature difference (Tt-Ti) increases with the increase of time when halogen light shining on the PV panel. Table 3-1(a)~(d) also shows the water temperature difference (Tt-Ti) increases with the increase of different irradiation amount. Photovoltaic efficiency of the PV panel, ηe, is calculated from equation (2-1) by measured the I-V curve with solar I-V measuring instrument at different irradiation amount. Hot water efficiency of the solar PV/thermal hybrid system, ηw, is calculated from equation (2-2) by the substitution of water temperature in the tank in table 3-1. Total photo-electric conversion efficiency, ηe,total, can be is calculated from equation (2-3) or (2-4).

1553

The Various efficiency of storage-type PV/thermal hybrid system is listed in table 3-2. Though, heat energy increases with the increase of irradiation amount, hot water efficiency ηw is maintained about 38~40%. Because of experimental simulated light with halogen lamp in the Lab, pure photovoltaic efficiency ηe is about 5~6%. The electric/heat energy conversion efficiency ηc is 0.38, recycling hot water gets more electric/heat conversion efficiency ηc·ηw about 14~15%. It reveals that the hot water conversion efficiency of storage-type solar PV/thermal hybrid system is almost triple than photovoltaic efficiency in experimental surrounding. Total efficiency ηe,total calculated is about 19~21%. The results show that total efficiency ηe,total of storage-type solar PV/thermal hybrid system is almost 4 times as compared to the pure photovoltaic efficiency ηe.

1554



Conclusion Storage-type solar PV/thermal hybrid system is set up to cool the surface of PV chip and reuse the waste heat. Surface temperature of PV panel cooled showed with IR thermal imager in the storage-type solar PV/thermal hybrid system decreases obviously. The result makes sure that the photovoltaic efficiency and the lifetime of PV panel can maintain. Total efficiency transformed by photovoltaic efficiency and hot water efficiency. The results show that total efficiency in experiment is more excellent as compared to the pure photovoltaic efficiency. The hot water efficiency of PV/thermal hybrid system from recycling waste heat gets 38~40%, and the electric/heat conversion efficiency gets about 14~15%. Photovoltaic efficiency of PV chip is about 10~15% at the irradiation of 600~900 W/m2 in real sunshine surrounding. It shows, though in real sunshine surrounding, the total efficiency of storage-type PV/thermal hybrid system is more twice than photovoltaic efficiency. 

Acknowledgment This research was supported by the Ministry of Science and technology, Republic of China,

1555

under Grant No. MOST 103-2221-E-236-015. Keywords: PV/thermal, storage-type, total efficiency

1.

2.

3. 4. 5. 6.

7.

References Sangani CS, Solanki CS: Experimental evaluation of V-trough (2 Suns) PV concentrator systems using commercial PV modules,‖ Sol. Energy Mater. Sol. Cells. 2007; 91: 453– 459. Solanki CS, Sangani CS, Gunashekar D, Antony G: Enhanced heat dissipation of V-trough PV modules for better performance,‖ Solar Energy Materials & Solar Cells. 2008; 49:1634-1638. Akbarzadeh A, Wadowski T: Heat pipe-based cooling systems for photovoltaic cell under concentrated solar radiation, Applied ThermalEngineering. 1996; 16(1):81-87. Abrahamyan YA, Serago VI, Aroutiounian VM, Anisimova ID et al.: The efficiency of solar cells immersed in liquid dielectrics,‖ Sol Energy Mater Sol Cells. 2002; 73:367–75. Muaddi JA, Jamal MA: Spectral response and efficiency of a silicon solar cell below water surface, Sol Energy. 1992; 49:29–33. Krauter, S., Hanitsch, R., Moreira, L., New optical and thermal enhanced PV modules performing 12% better under true module rating conditions, Proceedings of the 25th IEEE – PV - Specialists Conference, Washington DC, USA. 1996; May 13–19:1323–1326. Krauter S: Increased electrical yield via water flow over the front of photovoltaic panels, Solar Energy Materials & Solar Cells. 2004; 82:131–137.

8. 9.

Hollick JC: Solar cogeneration panels, Renew Energy. 1998; 15:195–200. Tonui JK, Tripanagnostopoulos Y: Ventilation benefit accrued from PV module installed in building, Proceedings of the international conference on passive and low energy cooling for built environmental, Santorini, Greece. 2005; May19–21:861–866, 2005. 10. Kumar S, Tiwari A: Design, fabrication and performance of a hybrid photovoltaic/thermal (PV/T) active solar still Original Research Article, Energy Conversion and Management. 2010; 51(6):1219-1229. 11. Ji J, Ping J, Lu T et al.: A sensitivity study of a hybrid photovoltaic/thermal water heating system with natural circulation, Applied Energy. 2007; 84:222-237. 12. Huang BJ: Performance evaluation of solar photovoltaic/thermal systems, Solar Energy. 2001; 70: 443-448.

1556

ISFAS-1579 Natural Convection in Inclined Wavy-Wall Cavity Filled with Al2O3-Water Nanofluid Using Heatline Technique

Ching-Chang Cho*, Ching-Huang Chiu, Kuo-Ching Chiu, Chong-You Lai Department of Vehicle Engineering, National Formosa University, Taiwan, ROC E-mail: [email protected] Abstract This study use the heatline technique to estimate the natural convection heat transfer characteristics in an inclined wavy-wall cavity filled with Al2O3-water nanofluid. In this study cavity, it is assumed that the left and right walls have a wavy surface and are maintained at a high and low temperature, respectively. In addition, the upper and lower walls are both flat and insulated. The flow and temperature fields within the cavity are governed by the continuity equation, momentum equation, energy equation and Boussinesq approximation, and are solved numerically using the finite-volume method and SIMPLE algorithm. The results show that the inclination angle of the cavity has no significant effect on the heat function contours and Nusselt number for low Rayleigh numbers. However, given high Rayleigh numbers, the effect of inclination angle on heat function contours and Nusselt number is significant. In addition, the results also show that the Nusselt number increases as the Al2O3 nanoparticles is added to theworking fluid. Keywords: Heatline, Nanofluid, Cavity, Wavy-wall, Natural convection 1. Introduction Natural convection heat transfer in inclined cavities has many important applications in engineering systems, including solar energy collectors, double-glazed windows, car batteries, electronic cooling systems, crystal growth, and so on [1]. Thus, many researchers have investigated the flow field characteristics and heat transfer performance of natural convection in inclined cavities [2-4]. Overall, the results have shown that the flow conditions (e.g., Rayleigh number and Prandtl number) and geometry parameters (e.g., inclination angle and aspect ratio of cavity) have a significant effect on the flow characteristics and heat transfer behavior in the cavity. An optimal heat transfer performance occurs at a specific inclination angle of the cavity. In many engineering applications, wavy geometry structures are often used to enhance the heat transfer performance within cavities, such as heat exchangers, solar collectors, cooling systems for microelectronic devices, underground cable systems, and so forth. Many studies have

1557

shown that the use of the wavy geometry structures can improve the heat transfer performance when wavy geometry parameters have appropriate values [5-7]. The heat transfer performance can also be improved by changing the thermophysical properties of the working fluid. Nanofluids, consisting of nanoparticles with high thermal conductivity (e.g., Al2O3 or Cu) suspended in a base fluid with low thermal conductivity (e.g., water or oil), provide an effective means of improving the thermophysical properties of the working fluid [8]. In recent years, many researchers have studied the problem of natural convection heat transfer in cavities filled with nanofluids [9-11]. Overall, the results have shown that the heat transfer performance increases with an increasing nanoparticle volume fraction. In addition, the type of nanoparticle added to the working fluid affects the heat transfer effect. To show the path of heat transfer, a useful tool is the heatline visualization technique [12, 13]. Many researchers have used the technique to study the natural convection heat transfer characteristics in cavities [14-16]. These results have shown that the use of the heatline visualization technique can show and explain the heat transfer process in detail. The present study investigates the natural convection heat transfer performance in an inclined wavy-wall cavity filled with Al2O3-water nanofluid using the heatline visualization technique. In modeling the cavity, the left and right walls are assumed to have a wavy surface and a high and low temperature, respectively, while the upper and lower walls are assumed to be flat and insulated. The flow behavior and heat transfer characteristics in the cavity are assumed to be governed by the continuity equation, momentum equation, energy equation and Boussinesq approximation. In performing the simulations, the governing equations are solved using the finite-volume method and the SIMPLE (semi-implicit method for pressure-linked equations) algorithm. The effects of the Rayleigh number, nanoparticle volume fraction and cavity inclination angle on the heat function distribution and Nusselt number in the inclined wavy-wall cavity filled with Al2O3-water are examined. 2. Mathematical formulation 2.1 Governing equations Figure 1 illustrates the studied cavity. As shown, the left and right walls have a wavy surface, while the top and bottom walls are flat. In addition, the width and height of this cavity are Wc and H c , respectively. Furthermore, the cavity is inclined at an angle  to the horizontal. The gravitational force ( g ) is assumed to act in the negative y-direction.

1558

It is assumed that the used nanofluid is Newtonian and incompressible. The fluid flow in the cavity is assumed to be two-dimensional, laminar and steady state. In addition, the base fluid and nanoparticles are in thermal equilibrium and no relative motion occurs between them. Furthermore, the thermophysical properties of the nanofluid are all constant other than the density which varies in accordance with the Boussinesq approximation. Let the following non-dimensional quantities be introduced:

conditions of ignoring thermal radiation and viscous dissipation effects, the governing equations described the flow and temperature fields have the following non-dimensionalized forms:

1559

properties of nanofluid presented in Eqs. (2)-(4) can be estimated as follows [17, 18]:

estimated the heat transfer performance can be defined in the non-dimensionalized form:

Finally, the mean Nusselt number along the left wavy-wall surface can be obtained as:

2.2 Boundary conditions

1560

As shown in Fig. 1, the left wall of the cavity is maintained at a constant high temperature, the right wall has a constant low temperature, and the top and bottom walls are both insulated. A no-slip impermeable velocity boundary condition is assumed to apply at all of the wall surfaces. The dimensionless boundary conditions can be summarized as follows:

2.3 Heat function The energy equation in a two-dimensional, incompressible, and steady-state for nanofluid-filled cavity without heat generation can be written in the following non-dimensionalized form [12, 13]:

The heat function equation in the non-dimensionalized form can be obtained as:

Note that the boundary conditions for the Eq. (9) can be obtained by integrating the Eq. (8). 2.4 Wavy-wall geometry and numerical solution procedure The non-dimensional profile of the wavy-wall was modeled as follows:

1561

The grid system of the wavy-wall cavity was generated by solving a set of Poisson equations [20]. A generalized coordinate transform technique was used to transform the governing equations from a Cartesian coordinate system to a generalized curvilinear coordinate system. The transformed governing equations have following generalized form [17-19]:

The finite-volume method [21] is used to discretize the governing equations and corresponding boundary conditions. The second-order upwind scheme is utilized to discretize the convection terms in the Eqs. (3) and (4). The SIMPLE algorithm [21] is used to couple the velocity-pressure fields. Finally, line-by-line TDMA (tri-diagonal matrix algorithm) iterative scheme is imposed to solve the discretized algebraic equations. 2.5 Grid independence evaluation and numerical validation The variations of the mean Nusselt number along the hot wavy-wall surface with the grid sizes of 51101 , 101 201 , 101 401 , and 201 801 are performed in the shown that a grid size of 101 401 has a grid-independent solution. The computational domain was thus meshed accordingly. The present numerical results for the variation of the mean Nusselt number in an inclined square cavity with Pr  100 have compared with those presented in [22]. Note that in the compared cavity, the top and bottom walls were maintained at low and high temperatures, respectively, while the left and right walls were assumed to be insulated. Table 1 presents the results. As shown, it is seen that the present results are consistent with those presented in [22].

3. Results and discussion

1562

In the current study, the Al2O3-water nanofluid is used. The thermophysical properties

Figures 2 and 3 plot the heat function contours in the cavity given Rayleigh numbers of Ra  102 and Ra  106 , respectively. Given a low Rayleigh number, since the flow velocity due to buoyancy force in the cavity is lower, the heat energy is transported slowly and more

uniformly form the high temperature wall to low temperature wall for all considered nanoparticle volume fractions and inclination angles. As a result, a more uniform heat function contours without enclosed cell are presented in the cavity (see Fig. 2). Given a high Rayleigh number, since the flow velocity is high, the heat energy can be transported fast

1563

via the fluid flow for all considered nanoparticle volume fractions and inclination angles. Consequently, the heat function contours are more complex. It also shows that enclosed cells are presented in the cavity. This result indicates that the heat energy is rotated in the enclosed cells. Comparing the heat function contours between the two Rayleigh numbers, it is shown that higher values of heat function are presented in the case of the higher Rayleigh number. This result implies that the heat transfer is better in the case of the higher Rayleigh number. When the cavity is inclined, the direction of the gravitational force acting on the nanofluid changes, and prompts a corresponding change in the buoyancy effect. At low Rayleigh numbers, since the buoyancy effect is weak and the fluid flow is slow, there is no significant change in the heat function patterns. However, at high Rayleigh numbers, the buoyancy effect is strong and is thus susceptible to changes in the orientation of the cavity. Therefore, significant changes in the distribution of heat function contours occur as the inclination angle is varied. Figure 4 shows the variation of the mean Nusselt number with the Rayleigh number as a function of the inclination angles. As discussed above, under low Rayleigh number conditions, the flow velocity and heat function have only slightly variation for all inclination angles. Consequently, an approximately constant mean Nusselt number is obtained for all inclination angles. However, as the Rayleigh number increases, the strength of the heat function is increased and enclosed cells are presented in the cavity. The results indicate that a stronger heat energy transport occurs. Therefore, the mean Nusselt number increases as the Rayleigh number increased. In addition, for high Rayleigh numbers, a significant change in the distribution of the heat function occurs when the cavity is inclined. Consequently, the mean Nusselt number have also significant variation as the cavity is inclined.

1564

Figure 5 shows the variation of the mean Nusselt number with the Rayleigh number as a function of nanoparticle volume fractions. Since the Al2O3 nanoparticles add to the water, the thermal conductivity of working fluid is added. As a result, the heat energy can be transported fast. Consequently, it is shown that the mean Nusselt number increases as the volume fraction of Al2O3 nanoparticle for all considered Rayleigh number.

4. Conclusions This study has used the heatline technique to estimate the natural convection heat transfer performance of Al2O3-water nanofluid in a cavity bounded by vertical isothermal walls with a wavy surface and straight upper and lower walls with adiabatic conditions. In performing the simulations, the governing equations have been modeled using the continuity equation, momentum equation, energy equation and Boussinesq approximation, and then solved numerically using the finite volume method and SIMPLE algorithm. The results have shown that the distribution of heat function in the cavity is strongly dependent on the inclination angle of the cavity for higher Rayleigh numbers. However, given low Rayleigh numbers, the inclination angle has no significant effect on heat function contours and Nusselt number. In addition, the results have also shown that the use of nanofluid can increase the Nusselt number. Acknowledgement The authors would like to thank the Ministry of Science and Technology, Taiwan, R.O.C. for the financial support of this study under Contract No. MOST 104-2221-E-150-045 References

1565

[1] Singh, A.K., Roy, & S., Basak, T. (2012). Analysis of Bejan‘s heatlines on visualization of heat flow and thermal mixing in titled square cavity. International Journal of Heat and Mass Transfer, 55, 2965-2983. [2] Rahman, M., & Sharif, M.A.R. (2003). Numerical study of laminar natural convection in inclined rectangular enclosures of various aspect ratios. Numerical Heat Transfer, Part A: Applications, 44, 355-373. [3] Huelsz, G., & Rechtman, R. (2013). Heat transfer due to natural convection in an inclined square cavity using the lattice Boltzmann equation method. International Journal of Thermal Sciences, 65, 111-119. [4] Omranian, A., Craft, T.J., & Iacovides, H. (2014). The computation of buoyant flows in differentially heated inclined cavities. International Journal of Heat and Mass Transfer, 77, 1-16. [5] Dalal, A., & Das, M.K. (2006). Natural convection in a cavity with a wavy wall heated from below and uniformly cooled from the top and both sides. ASME Journal of Heat Transfer, 128, 717-725. [6] Oztop, H.F., Abu-Nada, E., Varol, Y., & Chamkha, A. (2011). Natural convection in wavy enclosures with volumetric heat sources. International Journal of Thermal Sciences, 50, 502-514. [7] Hasan, M.N., Saha, S., & Saha, S.C. (2012). Effects of corrugation frequency and aspect ratio on natural convection within an enclosure having sinusoidal corrugation over a heated top surface. International Communications in Heat and Mass Transfer, 39, 368-377. [8] Yu, W., France, D.M., Routbirt, J.L., & Choi, S.U.S. (2008). Review and comparison of nanofluid thermal conductivity and heat transfer enhancements. Heat Transfer Engineering, 29, 432-460. [9] Oztop, H.F., & Abu-Nada, E. (2008). Numerical study of natural convection in partially heated rectangular enclosures filled with nanofluids. International Journal of Heat and Fluid Flow, 29, 1326-1336. [10] Abouali, O. & Ahmadi, G. (2012). Computer simulations of natural convection of single phase nanofluids in simple enclosures: A critical review. Applied Thermal Engineering, 36, 1-13. [11] Sidik, N.A.C., & Razali, S.A. (2014). Lattice Boltzmann method for convective heat transfer of nanofluids - A review. Renewable and Sustainable Energy Reviews, 38, 864-875. [12] Kimura, S., & Bejan, A. (1983). The ―Heatline‖ visualization of convective heat transfer. ASME Journal of Heat Transfer, 105, 916-919. [13] Casta, V.A.F. (2006). Bejan‘s heatlines and masslines for convection visualization and analysis. ASME Applied Mechanics Reviews, 59, 126-145.

1566

[14] Oztop, H.F., Mobedi, M., Abu-Nada, E., & Pop, I. (2012). A heatline analysis of natural convection in a square inclined enclosure filled with a CuO nanofluid under non-uniform wall heating condition. International Journal of Heat and Mass Transfer, 55, 5076-5086. [15] Basak, T., & Chamkha, A.J. (2012). Heatline analysis on natural convection for nanofluids confined within square cavities with various thermal boundary conditions. International Journal of Heat and Mass Transfer, 55, 5526-5543. [16] Sheikholeslami, M., Gorji-Bandpy, M., & Soleimani, S. (2013). Two phase simulation of nanofluid flow and heat transfer using heatline analysis. International Communications in Heat and Mass Transfer, 47, 73-81. [17] Cho, C.C., Chen, C.L., & Chen, C.K. (2013). Mixed convection heat transfer performance of water-based nanofluid in lid-driven cavity with wavy surfaces. International Journal of Thermal Sciences, 68, 181-190. [18] Cho, C.C., Chen, C.L., & Chen, C.K. (2013). Natural convection heat transfer and entropy generation in wavy-wall enclosure containing water-based nanofluid. International Journal of Heat and Mass Transfer, 61, 749-758. [19] Cho, C.C. (2014). Heat transfer and entropy generation of natural convection in nanofluid-filled square cavity with partially-heated wavy surface. International Journal of Heat and Mass Transfer, 77, 818-827. [20] Thomas, P.D., & Middlecoff, J.F. (1980). Direct control of the grid point distribution in meshes generated by elliptic equations. AIAA Journal, 18, 652-656. [21] Patankar, S.V. (1980) Numerical Heat Transfer and Fluid Flow, New York, McGraw-Hill. [22] Khezzar, L., Siginer, D., & Vinogradov, I. (2012). Natural convection of power law fluids in inclined cavities. International Journal of Thermal Sciences, 53, 8-17.

1567

ISFAS-1482 A Study on Mechanical Behaviors of Si/Ge Core-Shell Nanowires by Molecular Dynamics Simulation Yu-Shan Lin, Tei-Chen Chen Department Department of Mechanical Engineering, National Cheng Kung University, Tainan, Taiwan E-mail: [email protected] 1. Introduction One-dimensional semiconductor Si and Ge nanowires (NWs) have attracted great attention over the past decades. They are promising candidates for the realization of innovative nanodevices for electronics and photonics in near future, such as field-effect transistors (FETs), light-emitting diodes (LEDs), and nanoscale sensors. Due to the effects of size and quantum confinement of NWs, their optical, electrical, magnetic, mechanical, and thermal properties are quite different from the bulk materials. Recently, Si/Ge core-shell NWs have especially become the focus of study. Through the adjustment of heterostructure composition and interface geometry, Si/Ge core-shell NWs can be further manipulated to tune the electronic properties of NWs. To the authors‘ knowledge, the investigations on mechanical properties and deformation mechanisms of these materials are still lacking. Therefore, it is our objective to investigate the nanoscale mechanical behaviors of Si/Ge core-shell NWs. 2. Methods Molecular dynamics simulations were carried out using the program package LAMMPS in NVT ensemble. A simulation on NWs is performed on the behavior of slip system, strength, stress distribution and fracture process with different structure, orientation, length, temperature, and shell thickness under uniaxial tension. Three different single crystalline directions: (100), (110), and (111) of silicon, germanium, Si-core/Ge-shell, and Ge-core/Si-shell NWs at different temperatures were investigated. The lattice constants of Si and Ge are 0.543 nm and 0.565 nm, respectively. The size of all the NWs is 30 nm in length with 5 nm in width of square section. The inter-atomic potential function proposed by Tersoff, that considers the effect of bond angle and covalent bonds shown to be particularly feasible in dealing with IV elements and those with a diamond lattice structure such as C, Si, and Ge, is, adopted in this study to analyze the dynamic correlations between atoms. In this work, the time step is as short as 1 fs to ensure solution accuracy. A modified five-step methodology is used to incorporate

1568

Newton‘s equations of motion, so that the position and velocity of a particle can be effectively evaluated. Moreover, the mixed neighbor list is applied to enhance computational efficiency. 3. Results and Conclusion Simulation results show that the magnitude of Young's modulus and ultimate strength of four different NWs including single-crystal silicon, single crystal germanium, Si-core/Ge-shell, and Ge-core/Si-shell in the order from the largest to the smallest are (111)>(110)>(100). For instance, the values of Young's modulus of Si-core/Ge-shell in single crystalline orientations (111), (110), and (100) are 153.42, 115.62, and 85.27 GPa, respectively, while the values of ultimate strength are 30.89, 28.92, and 17.49 GPa, respectively. The magnitude of Young's modulus of four different NWs in (111) direction is Si-core/Ge-shell>single-crystal silicon> single crystal germanium>Ge-core/Si-shell, and their values are 153.42, 128.63, 117.93, and 115.68 GPa, respectively. The ultimate strength increases with decreasing temperature and increasing strain rate and length. For instance, the values of ultimate strength of (111) Si-core/Ge-shell NWs at temperature 50, 300, 600, and 1000 K are 30.89, 25.23, 20.15, and 7.14 GPa, respectively, while the discrepancy in ultimate strength is very small when the strain rate is increased from 0.33 to 33 %/ps. As the length of NWs of (111) Si-core/Ge-shell NWs increases from 7.5 to 50 nm, the values of ultimate strength are increased from 27.21 to 30.92 GPa. In comparison with two different core/shell NWs, the ultimate strength of Si-core/Ge-shell NW is higher than Ge-core/Si-shell NW with the same length and cross-sectional area. In addition, nanostructure of NWs on different orientations has different slip systems. When tensile stress is exerted on nanostructure silicon or germanium NWs, (001)-oriented has two pairs of slip-system, while (011)-oriented has only one pair slip-system perpendicular to tension-direction. When there exist more slip-systems under the action of tensile stress, more atoms have higher opportunity to slip to release stress. Consequently, its coefficient of elasticity is relatively low. On the contrary, if the orientation does not have any slip-system, its coefficient of elasticity is the greatest. On the other hand, for both Si-core/Ge-shell and Ge-core/Si-shell nanowires, the slip and dislocation always initiate near the shell surface first and then gradually propagate toward the core region during uniaxial tension. 4. Acknowledgement This research was supported by the Ministry of Science and Technology in Taiwan through Grants No. 104-2221-E-006-126.

Keywords: SiGe, core/shell nanowire, molecular dynamics, uniaxial tension

1569

ISFAS-1577 Applications of Magnetic Particles to Adsorb Emerging Contaminants in Aqueous Solution Hua-Wei Chena, Yo-ping Wub, Shu-Han Changb, Chyow-San Chiou c a Department of Cosmetic Application and Management, St. Mary‘s Medicine, Nursing and Management College, Taiwan E-mail: [email protected] b

c

Department of Chemical and Materials Engineering, National Ilan University, Taiwan E-mail: [email protected]

Department of Environmental Engineering, National I-Lan University, Taiwan E-mail: [email protected]

1. Background/ Objectives and Goals The study is the preparation of magnetic carrier surface modification containing phenyl polymer to absorb emerging contaminants (i.e. parabens such as methylparaben, ethylparaben and propylparaben) in aqueous solution. Magnetic adsorbent manufactured from magnetite (Fe3O4) can be easily recovered from treated water by magnetic force, without further downstream treatment. 2. Methods In this research, the surface of magnetite modified with styrene (St) as the functional monomer, divinylbenzene (DVB) as the cross-linking agent, polyvinyl pyrrolidone (PVP) as the stabilizer, 2,2-azodiisobutyronitrile (AIBN) as the radical initiator and ethanol as the solvent to prepare a surface coating of magnetic polymer adsorbent (St/Fe3O4), and use this phenyl compound to absorb parabens. 3. Expected Results/ Conclusion/ Contribution The parabens adsorption of St/Fe3O4 show good agreement both with the Langmuir and Freundlich adsorption isotherm, and the maximum adsorption capacity (qm) and Gibbs free energy at 298 K was methylparaben: 0.6015mg g-1, -21.443 k J mol-1; ethylparaben: 3.2862 mg g-1, -22.4749 k J mol-1; propylparaben: 3.4423 mg g-1, -21.443 k J mol-1, respectively. A pseudo-second-order model could best describe the adsorption kinetics, and the derived activation energy was methylparaben: 69.7328 kJ mol-1; ethylparaben: 68.7227 kJ mol-1; propylparaben: 10.5480 kJ mol-1. The optimum condition to desorb methylparaben from St/Fe3O4 is provided by a solution with 100% ethanol.

1570

Keywords: magnetic particles, styrene, paraben, adsorption

1571

ISFAS-1593 Surface Enhanced Raman Scattering on Periodic Gold Nanowire Arrays Yu-Chung Changa,*, Yu-Ju Hungb, Bo-Han Huanga, Tsung-Hsien Lina, Chun-Han Chena a Department of Electrical Engineering, National Changhua University of Education, Taiwan * E-mail: [email protected] b

Department of Photonics, National Sun Yat-sen University, Taiwan E-mail: [email protected]

1. Background/ Objectives and Goals It is well know that strong interaction occurs between local optical field and noble metal nanostructure through the excitation of surface plasmons. The resonant interactions result in highly enhanced local field in the close proximity of the nanostructures. The greatly enhanced local filed has been exploited in a wide range of applications, such as surface enhanced Raman scattering (SERS), surface enhanced fluorescence, enhanced nonlinear processes, and etc. In particular, SERS has attracted enormous attention due to its potential to identify trace amount of molecules without prior-label. However, it is usually difficult to achieve the high field required for SERS. Special design and sophisticated fabrication process of the nanostructure is often required. We present a simple method to fabricate gold nanowire arrays (grating) as an efficient SERS substrate in this report. The field enhancement is highly dependent on the grating period. We conducted both experiment and simulations to quantitatively evaluate the influence of the structural parameters on the enhanced Raman signal. 2. Methods The PMMA based photo-resist (PR) was first spin coated on cleaned ITO glass plates. Various periods of gratings were fabricated on the PR by e-beam lithography. Two different types of nanostructures were made, one was made through typical lithography process with PR lifted off (lift-off), and the other was made without the lift-off process (non-lift-off). The samples were then coated with a layer of gold thin film of various thickness (30-60 nm) by a sputter. Some samples were subsequently coated with a thin layer (14 nm) of PMMA as the spacer to investigate the effect of fluorescence enhancement in the meantime. Fluorescent molecules Rhodamin-6G were mixed in DI water. The R6G solution was put on the nanostructures for subsequent spectroscopy studies. The excitation wavelength was 532 nm. Spectroscopic data were taken with a confocal Raman microscope (DXR Raman, Thermo Scientific). The sample preparation procedure was illustrated in Fig. 1.

1572

3. Expected Results/ Conclusion/ Contribution The SERS signals measured on non-lift-off samples are shown in Fig. 2 (a). Although fluorescence enhancement were observed on both lift-off and non-lift-off nano-structures, efficient SERS was present only on non-lift-off samples. Similar to fluorescence enhancement, the SERS efficiency was highly dependent on the grating period as shown in Fig. 2 (b), indicating they were originated from a common factor - localized surface plasmon resonance. In addition, only TM polarization manifested strong enhancement, which was also the evidence of surface plasmon mediated field enhancement. In conclusion, we demonstrated that efficient SERS substrates can be made by the proposed method without photoresist lifted off. Therefore, sample preparation time can be significantly shortened and the sample quality can be well controlled. Highest SERS efficiency was obtained when the grating periods were 300-400 nm. Simulations were performed to elucidate the results.

1573

Keywords: surface enhanced Raman scattering, surface plasmon resonance, surface enhanced fluorescence, nanotechnology.

1574

ACEAIT-7555 Development of Typhoon Rainfall Forecast Model Using Satellite Data Chih-Chiang Wei National Taiwan Ocean University, Taiwan E-mail: [email protected] 1. Background/ Objectives and Goals The purpose of this study is to develop the typhoon rainfall forecast model by using Special Sensor Microwave Imager Sounder (SSMIS) satellite passive microwave radiometers. The island of Taiwan is situated in one of the main paths of western North Pacific typhoons and is affected by an average of four typhoons each year. Therefore, a useful scheme for quantitative precipitation forecast during typhoon periods is highly desired. 2. Methods To construct a typhoon rainfall forecast model, this study develops Bayesian network (BN) and logistic regression (LR) models using three different datasets and examines their feasibility under different rain intensities. The study area is the watershed of Tanshui River in Taiwan. The dataset includes a total of 70 typhoon events affecting the watershed from 1997 to 2008. For practicability, the three datasets used include climatologic characteristics of typhoons issued by the Central Weather Bureau, rainfall rates measured using automatic meteorological gauges in the watershed, and microwave data originated from SSMIS radiometers. Five separate BN and LR model cases, differentiated by a unique combination of input datasets, were tested, and their predicted rainfalls are compared in terms of skill scores including mean absolute error and root mean square error. 3. Expected Results/ Conclusion/ Contribution The study designed four cases, including Case 1: Prediction using only typhoon climatologic data; Case 2: Prediction using only hydrological data; Case 3: Prediction using only brightness temperatures; Case 4: Prediction using both datasets of hydrological data and brightness temperatures; and Case 5: Prediction using all datasets. The results show that the case with all three input datasets used is better than the other four cases. Moreover, LR can provide better predictions than BN, especially in flash rainfall situation. However, BN might be one of the most prominent approaches when considering the ease of knowledge interpretation. In contrast, LR describes associations, not causes, and does not explain the decision. Keywords: Rainfall, Typhoon, SSMIS, Bayesian networks

1575

ACEAIT-7561 Flexural Behaviors of RC Non-Uniform Beam with Lap Details Using Headed Bars Seung-Hun Kima, Kyu-Seon Leeb a Department of architectural engineering, Hanbat national university, Republic of Korea E-mail: [email protected] (Corresponding author) b

Department of architectural engineering, Hanbat national university, Republic of Korea E-mail: [email protected] 1. Background and Objectives In the codes of ACI318-14 and KCI 2012, it represents the equation for determining the development length of headed bars. These design codes don‘t provide guideline for lap splice length of headed bars. Lap details with headed bars can be applied to the joint of RC non-uniform beams with different depths. In this study, experimental works (NRF-2013R1A1A2013485) for RC non-uniform beam with these details were conducted to evaluate a flexural behavior as a variable of lap length and stirrups in lap zone. 2. Experimental Programs Three specimens were tested as bending test. The designed yield strengths of headed bars were 400~600MPa. The diameter (db) of headed bars is 22m. Specimens had a lap length of 20-25 times db and had the details of lap zone with and without stirrup reinforcement. The difference between the depths was 50mm. Four-point loads were applied to minimize the shear force and maintain the equivalent bending moment in lap splice zone. 3. Test Results and Conclusions For all specimens, the cracking initiated outside of the lap zone at the point where the depth changes. For specimens without stirrups in lap zone, it showed the diagonal cracks along the lap splice and the prying failure caused by the curvature of the specimen and bending moments in the lap bars at large deformations. Specimen with the lap length of 20 times db had the less maximum load than specimen with the lap length of 25 times db. The stirrups as confinement details in lap zone contributed to increase strength and deformation. The maximum test strength of specimen with stirrups was greater than the theory strength. For specimen without stirrups, on the other hand, it showed the opposite effect. These results showed the importance of the confinement details in lap zone.

1576

Keywords: Headed Bar, Lap Splice, Non-uniform Beam, Bending Test

1577

ACEAIT-7459 Wireless Monitoring and Prediction of Rail Temperature Kuo, Chen-Minga, Han, Tsung-Haob, Chiang, Tzu-Fanc Dept. of Civil Eng., Natl. Cheng Kung Univ., Taiwan E-mail: a [email protected], b [email protected], c [email protected] 1. Background/ Objectives and Goals The study developed measurement apparatus of rail temperature and compared the data with professional measurement system of rail temperature. The 24-hour non-stop measurement data was transferred via wireless equipment and linked to the web. The rail temperature data were analyzed with the meteorological data. A regression equation for the specific location was proposed to predict rail temperature via air temperature and solar radiation. 2. Methods This study aims to develop wireless monitoring system with Wireless Sensor Network (WSN) which is 2.4GHz, IEEE 802.15.4 radio to communicate with WSN measurement nodes. WSN-Thermocouple measuring system can measure data with low power and transmit data to computer. The actual transmitting distance from measuring position to NI WSN gateway is 80 meters. WSN-Thermocouple measuring system measures rail temperature and air temperature separately with two thermocouples. Because thermocouple is contact-type measurement, it must completely contact the surface of measured objects. Therefore, we use rubber pad to cover the thermocouple and add the thin sheet metal external part to ensure thermocouple measuring proper temperature. To verify WSN-Thermocouple, Measure & Detect System developed by Vortok Inc. and infrared thermometer are used to discuss the difference of three temperature-measuring methods. The result showed that the difference in temperature between WSN-Thermocouple and Measure & Detect System/infrared thermometer is about 1℃. 3. Expected Results/ Conclusion/ Contribution By observing single day rail temperature, it can be found that the correlation of rail temperature and air temperature, but the actual rail temperature could vary from many factors, such as season, latitude and incident angle of sunlight. The result is long-term monitoring of rail temperature, air temperature and amount of insolation3. Compared temperature history on 10th March and 19th March, the rail temperature on 19th March is higher than the rail temperature

1578

on 10th March. This is because the amount of insolation on 19th is five times more than on 10th. Although the air temperature is a great effect on rail temperature, amount of insolation and sunshine duration are the keys of deciding the gradient of rail temperature. The results show that the model has been used to predict rail temperature at Dajia Train Station in Aug, 2014.The correlation coefficient of prediction temperature and real temperature is 0.826 in Aug, 2014. The mean error of predict maximum temperature and actual maximum temperature is in 0.88℃. The mean error of predict minimum temperature and actual minimum temperature is in 0.48℃.The results show that this model has the ability to predict rail temperature accurately with air temperature data and amount of insolation.

1579

ACEAIT-7606 Preliminary Study on the Monitoring System by Using UAV Imagery Jiyoung Kim1, a, Yoonsik Bang 2, b , Yangsun Kim3, c , Jae Bin Lee4, c a Institute of Construction and Environmental Engineering, Seoul National University, Korea b Department of Civil and Environmental Engineering, Seoul National University, Korea c Department of Civil Engineering, Mokpo National University, Korea *Corresponding Author: [email protected] Abstract In the event of a disaster, a rapid arrival of an emergency response vehicle can reduce the impact of the disaster. However, obstructions such as road congestion, illegal parking, etc. can be an obstacle to effective response. Thus, a support system that detects ongoing road and traffic situations using Unmanned Aerial Vehicles (UAV) which move freely and guide an optimum path is useful. Hence, the UAV-based traffic monitoring system that consists of an aviation (UAV air component) and terrestrial section (UAV surface component) was designed for monitoring traffic and is described in this paper. Keyword: Ummanned Aerial Vehicle Imagery, Disaster Situation, Traffic Situation Monitoring, Civil Engineering 1. Introduction Civilian applications of UAVs have increased considerably in recent years due to their increased availability and the miniaturization of sensors, Global Positioning System (GPS), inertial measurement units, and other hardware [1,2]. The advantages of using UAVs over a piloted aircraft include lower image acquisition costs, quick and repeated deployment, and the ability to fly at low altitudes [3]. For these reasons, the use of UAVs for emergency and disaster monitoring has increased considerably in recent years. Aerostar UAVs of Aeronautics company with thermal infrared technology have been used for forest fire monitoring in the EU [4] and Aero-light UAVs have been used for monitoring the sources of pollution or radioactivity around a nuclear power plant in Israel [5]. Furthermore, UAVs with rescue equipment have been used in the disaster site. An ambulance UAV with an automated external defibrillator flew to an emergency site which an ambulance could not access and then the paramedics provided guidance in first aid using UAV images in Delft University of Technology in the Netherlands [6]. UAVs are being used for searches, emergency rescue, and monitoring, etc., which focuses on the application of UAV in the event of a disaster that people or vehicles cannot access.

1580

A single disaster resulting in a large amount of secondary damage and third party damage because of the complexity of the city where 70% of the population settles down. Disasters that occur in cities are often associated with delayed reactions caused by the delay of firefighting and emergency response vehicles caused by road obstructions [7]. A quick arrival of the emergency vehicle at the site is an important factor in effective disaster response in the urban areas. Although the capability to derive road conditions from aerial photographs or closed-circuit television (CCTV) has steadily progressed, this method has limited application for the quick reaction of an emergency vehicle. Thus, when an emergency vehicle is dispatched, rapid access to road condition information from UAV images is useful. This paper examined studies that used UAVs to gather road context data, and presents a system for emergency vehicle dispatch support. 2. Research on Road Context Monitoring using UAVs Studies on traffic monitoring including, road context data, using UAVs have been created consistently since the 2000s. In the early 2000s, research efforts focused on establishing an efficient system that could control UAVs effectively, and send and receive video information quickly. Thus, the improvement of aircraft operation mechanisms and the development of new communication systems were a major focus, but image processing and detection issues were considered secondary. A commercial system that could take and transmit images with Aerosonde UAVs for traffic monitoring was developed [8]. Additionally, navigation technology that could perform fully automated imaging through virtual simulation of UAV movement was studied [9]. A distributed control system was created that could collect real-time traffic information through UAV images and build multi-modal data [10], and detect collaborative vehicles using a number of UAVs was also developed [11]. Since the late 2000s, there have been major advances in weight reduction, photo storage and communication performance and image analyses. Studies have estimated the traffic situation on a particular section of road by extracting data along the road line using a single image [12] and by detecting individual vehicles in images though the utilization of statistical techniques [13]. In addition, studies were conducted to identify the likelihood of vehicle movement [15] by extracting information from a moving vehicle detected in consecutive video images [14] or by detecting vehicles and measuring the distance among multiple vehicles using thermal images obtained from infrared cameras. Many studies have used vehicle detection or traffic information, but quick image processing (within 5 minutes) is key to determining road conditions for effective emergency vehicle response. Thus, method of quickly and accurately extracting the road context information from UAV images is needed. 3. UAV-based Traffic Situation Monitoring System Design

1581

The Proposed system consists of an aviation section (UAVs) and a terrestrial section (surface component). The UAV component has an on-board digital camera, GPS, IMU, multi-sensor system and support system that stores and transfers data. The surface component consists of the receiver system which wirelessly receives and stores the sensor data transmitted from both the ground platform and UAV air component and remotely controls the sensor and the processing system. These data are georeferenced and provide real-time road context information from which the most effective response route is created. The overall configuration of the UAV-based traffic situation monitoring system is shown in Fig. 1. 3.1 UAV Air Component The main purpose of the UAV air component is obtaining and transferring sensor data to the surface component in a real-time. The UAV air component is largely composed of the UAV Platform, multi-sensor and support systems which control data storage and transmission. The system utilizes a low-altitude UAV which has on-board sensor systems such as a digital camera, GPS, and IMU. The support system includes the Gimbal and multi-sensor which provides stability for the imaging sensor. The on-board computer (OBC) controls the storage and transmittance of sensor data.

3.2 UAV Surface Component The surface component receives data from the UAV air component in real time and of generates orthoimages and traffic information from the received information. The terrestrial section is easily transportable and consists of the support platform, receiver, and processing system. The receiver transmits sensor control instructions and receives sensor data in real time via wireless communication. The processing system generates real-time context information such as orthoimages, road shielded area, vehicle speed, etc.. It detects the best path by conflating road context information with the road network data (Fig.2).

1582

4. Conclusion As the importance of rapid arrival of emergency response vehicles at the scene of a disaster or accident has increased, a UAV-based traffic monitoring system incorporating spatial movement and rapid data acquisition are also increasingly important. The system presented here is divided into the multi-sensor system embedded within the UAV platform (digital camera, GPS, IMU, and related support systems), and UAV surface component which includes the receiver system, which receives and stores the multi-sensor data, and the processing system that does georeferencing and generates information in real time. In the designed traffic situation monitoring system, the UAV air component uses a low-altitude UAV as a platform is built with low-cost light sensors and a self-developed support system. The UAV surface component has discrimination in providing the road context information by receiving and processing the sensor data in real time. However, the utility of the designed system should be verified in the future by establishing its function and applying it to the real-world environment. Acknowledgements This work (Grants No. C0276754) was supported by Business for Cooperative R&D between Industry, Academy, and Research Institute funded Korea Small and Medium Business Administration in 2015. References [1] A.S. Laliberte, A. Rango and J.E. Herrick, "Unmanned Aerial Vehicles For Rangeland Mapping And Monitoring: A Comparison Of Two Ssystems" in: Proceedings of the ASPRS 2007 Annual Convention, Florida, 2007 [2]G. Zhou and D. Zang, ―Civil UAV system for earth observation', in: Proceedings of the International Geoscience and Remote Sensing Symposium (IGARSS), Barcelona, 2007, p. 5319–5322 [3]A. Rango, A.S. Laliberte, C. Steele, J.E. Herrick, B. Bestelmeyer, T. Schmugge, A. Roanhorse, and V. Jenkins, "Using unmanned aerial vehicles for rangelands: current applications and future potentials", Environmental Practice Vol. 8 2006, p. 159–168 [4] Information on http://www.uvonline.com [5] Information on http://www.airforcetechno-logy.com

1583

[6] Information on http://techholic.co.kr/archives/24421 [7] Ministry of Government Administration and Home Affairs(MGAHA), "Cases: Delayed reaction of fire engines", 2002. [8] S. Srinivasan and H. Latchman, "Airborne Traffic Surveillance Systems – Video Surveillance of Highway Traffic", in: the ACM 2nd international workshop on Video surveillance & sensor networks, 2004. [9] P. Doherty, P. Haslum, T. Merz, E. Skarman, G. Conte, S. Duranti, F. Heintz, P-O. Pettersson, T. Persson and B. Wingman, "A Distributed Architecture for Intelligent Unmanned Aerial Vehicle Experimentation", in: 5th IFAC Symposium on Intelligent Autonomous Vehicles, 2004. [10] Information on http://www.tfhrc.gov/trnsptr/jan03/ [11] Information on http://www.comets-uavs.org/results/papers.shtml [12] Z. W. Kim, "Real-time road detection by learning from one example", in: 7th IEEE Workshops on Application of Computer Vision 2005, p.455-460. [13] J. Gleason, A.V. Nefian, X. Bouyssounousse, T. Fong and G. Bebis, " Vehicle detection from aerial imagery" in: IEEE Intelligent Conference on Robotics and Automation , 2011. [14] R.J. Evans and E. Turkbeyler, "Visual MTI for UAV systems, 4th EMRS DTC Conference, Edinburgh", in: 4th EMRS DTC Conference, Edinburgh , 2007. [15] F. Heintz, P. Rudol and P. Doherty, "From Images to Traffic Behavior - A UAV Tracking and Monitoring Application", in: Information Fusion, 2007 10th International Conference on, publications/IEEE, Quebec, Que, 2007, p. 1-8

1584

ACEAIT-7532 Object Detecting LED Control System Using Doppler Sensor Sang Woo Junga,, Seung Hyun Paikb, Jun Yeong Lee,c and Hong Bae Parkd School of Electronics Engineering, Kyungpook National University, Republic Korea E-mail: a [email protected], b [email protected], c [email protected], d [email protected] 1. Background The LED lighting is growing in popularity because it has high energy efficiency and long life time, and it is gradually replacing the typical lighting device such as incandescent, fluorescent lamps, halogen lamps, and so on. And the demand of LED also has been increased sharply in various fields. In the lighting industry, there are strong demands for lighting equipment that is getting smaller, smarter, and more colorful. In a broad range of applications, the colored LED lighting offers a number of advantages over traditional light sources. Therefore many researches for LED lighting control have been done to improve the efficiency of lighting and the energy-saving. Especially, for energy-saving, many researches of control system using PIR, camera, and micro phone have been studied. However there are still some problems as follows. The LED control system with PIR sensor detects only moving object such as people, and so on. And PIR sensor performance largely depends on the ambient temperature in the operating environment. A camera with LED control system has very high cost in large scale systems. Micro phone performance largely depends on ambient sound in the operating environment. On the other hand, Doppler sensor can detect motion or speed of moving objects through Doppler principle, and can continue to find the location of the object. It transmits a microwave frequency electromagnetic signal and waits for the signal to receive back and monitors the shift in frequency signal. The microwave frequency 10.525GHz has not any interference by environment. So we estimate that LED control system using Doppler sensor can solve these problems. In this paper, we propose a new object detecting LED control system using Doppler sensor, LED lighting device, network, and LED power control. Then the proposed system can detect moving object and estimate the location using triangulation. 2. Methods The object detecting LED control system consists of object power control module, LED lighting device, network port, and detecting module using Doppler sensor.

1585

The LED control system is designated as a single primary and 8 secondary in one group. In one group, the number of secondary is flexible according to the circumstance. The primary manages 8 secondary based on their status information. Experiment place is 14 meters wide and 20 meters long. 56EA LED lighting devices with a built-in LED control board are set up about 2 meters apart. 9EA LED lighting devices make a group shaped like a squares. Doppler sensor installs in center of the squares. Total 6EA Doppler sensors installs in experiment place and these sensors form total 6EA cell which the diameter of the circle is 10 meters. If three or more cells detect the object using Doppler sensor, it can enable triangulation to work. So LED control board can compute the approximate location of that object in relation to the three reference points. In the operating process, if any secondary detects object by Doppler sensor, the location of the secondary and status information of LED device are transmitted to the primary. And all primary share the location information and transmit the message about light to the secondary nearby object. On the other hand, if the secondary are far enough away from object, then they are received the power off message. So the LED lighting devices are controlled by area and are determined whether to be turned on or not.

3. Expected Results/ Conclusion/ Contribution In this work, we developed a new object detecting LED control system using Doppler sensor, LED lighting, network, and power control system. The proposed system can be found continuous approximate location of object through triangulation using Doppler sensor. So the LED lighting devices are operated by the area and the unnecessary power consumption was reduced. In addition, the proposed system remains unaffected by the effects of the environment compared with the conventional system. LED control board connected IOT(internet of things) or actuators can be applied home automatic system and smart building.

1586

Keywords: Object detecting, LED control system, Doppler sensor, Triangulation.

1587

ACEAIT-7560 The Green Design of VCO and Mixer Circuits for Ultra-Wide Band Applications H.-C. Chow, Y.-C. Kuo, W.-S. Feng Department and Graduate Institute of Electronics Engineering Chang Gung University, Taiwan, Republic of China E-mail: [email protected] Abstract The paper investigates the low-power design strategy for RF key circuits, such as VCO and mixer. The design considerations focus on both the low energy consumption and low spurious performance. The development for the design of ground-adjustable variable-inductor for high tuning-range UWB VCO is achieved in TSMC 90nm CMOS process. And the 3.1~10.6 GHz ultra-wide band down-conversion mixer is implemented by TSMC 0.18μm CMOS process. As for the UWB VCO, it operates ranging from 12.92 GHz to 17.6 GHz with phase noise of -99.98 to -106.07dBc/Hz@1MHz, tuning range of 4.68 GHz, output power of -8.7 to -9.6dBm, and low power consumption of 6.3 mW and chip area of 0.77x0.62mm2.In the design of ultra-wide band down-conversion mixer, the down conversion mixer shows a conversion gain of 7.19 to 10.72 dB and the return loss of RF port is smaller than -10dB. Power consumption is only 26.3mW, and IIP3 of 0.8 dBm. The LO to IF isolation and LO to RF isolation are less than -30dB, respectively. The chip area is 1.14x0.89 mm2. All circuits have been designed and simulated by Advanced Design System (ADS) software. Keywords: Voltage-controlled oscillator, Mixer, Ultra-wide band, CMOS RF 1. Introduction In recent years, with the communications market's rapid growth and the general applications of mobile communication products, peopole for mobile communications have become increasingly to demand high quality, larger communication bandwidth, high-speed wireless access to data transfer time, and multi-band operations of the transmission system [1-5]. Therefore, multi-band is a trend and the ultra-wideband technology is becoming increasingly important. Thus, the research of the growing field of wireless communications is expected to be widely used.

1588

Current research [6-10]focuses on UWB IEEE 802.15.3a of 3.1 ~ 10.6GHz frequency band and the relevant norms, 802.15.3a, that is the so-called ultra-wideband technology, short-range, low-power UWB (Ultra Wide Band, UWB) wireless technology, with ultra-high-speed data transfer capability, IEEE 802.15.3a of 3.1 ~ 10.6-GHz, 5 to 10 meters per second over distances of up to 480Mbps, or even higher. In other words, within a short distance transmissions of large data. Ultra-wideband communications system in the future play a very important role, and its easy affinity can not be understated. Therefore, the field of UWB technology in communication transmission applications becomes very promising. 2.

Proposed Design

1. UWB VCO The output signal of the voltage-controlled oscillator (VCO) is for the operation of 13GHz-17.5GHz. In the LC tank design, the use of substrate modulation by the number of segments to adjust a variable inductor structure, the structure unit is shown in Figure 1. Figure 2 shows the variable inductance of the cross-section with the use of switch Metal 7's position on the ground to get the changes and therefore it can get different values. When the current flows through the inductor clockwise on the Metal 7, it will change its magnetic flux, with the flux equations below

And its sense of value equation can be expressed as

where B is the magnetic flux density, I is current and L is the inductance value. According to the above two equations, we find that by changing the magnetic flux, it can change their sense of values. So, this design by changing our ground, is used to influence the magnetic flux to change the sense of values. And this structure of the original spiral inductors by further development, it does not take up extra area. And the ground point of the design can be used in different locations to obtain the required frequency of the switching effect, and the use of Vt for a minor readjustment in order to avoid changes in the offset process. The ultra-wideband voltage-controlled oscillator of the circuit architecture is shown in Figure 3.

1589

2. UWB Down-Conversion Mixer For the design of 3.1 ~ 10.6GHz down-mixer (Mixer) this frequency range signal directly lies in the two down-to 100MHz baseband signals, to avoid image frequency interference. Finally, by the digital system processing, including mixers in the wireless communication front-end

1590

circuit is very important. It's through a low noise amplifier and the oscillator. Gain and linearity in the design of mixer are both very important parameters. The proposed circuit using TSMC CMOS 0.18μm process for the mixer is shown in Figure 4.

Simulation Results 1. UWB VCO The circuit design is implemented by the use of National Chip Implementation Center (CIC) in TSMC 90nm 1P9M CMOS process, and Agilent of the Advanced Design System (ADS) as a circuit simulation tool, with the focus of the receiver structure on the design of the oscillator between the 13GHz ~ 17.5GHz. The active port of the VCO design takes the use of PMOS and NMOS composed of complementary coupling, and a combination of a variable inductor and variable capacitor to adjust the oscillation frequency. Number of turns for the inductor switch were simulated; and the use of ADS Momentum for post-sim. Circuit simulation results are DC bias 1.2V/5.25mA, power consumption 6.3mW (including buffer of power consumption), tuning range of 12.92 ~ 17.6 GHz (4.68GHz), output power of -8.7 ~-9.6dBm. The state of the spectrum, phase noise, and adjustable bandwidth output waveform are denoted in Figures 5 to Figure 10. Table 1 shows the circuit simulation summary. Figure 11 is the adjustable inductance of ultra-wideband VCO layout, with its size 0.77x0.62mm2.

1591

1592

2.

UWB Down-Conversion Mixer

1593

The circuit design is implemented by the use of National Chip Implementation Center (CIC) in TSMC 180nm 1P9M CMOS process, and Agilent of the Advanced Design System (ADS) as a circuit simulation tool to set the RF of 3.1 ~ 10.6GHz.Simulation results are: the input power of -40dBm, LO of 3.1 ~ 10.6GHz, the input power -1dBm, IF to 0.1GHz. And Figure 12 to Figure 15 show the simulation results for this circuit and the layout in Figure 16.

1594

4. Conclusion The paper investigates the low-power design strategy for RF key circuits, such as VCO and mixer. The design considerations focus on both the low energy consumption and low spurious performance. The development for the design of ground-adjustable variable-inductor for high tuning-range UWB VCO has been achieved in TSMC 90nm CMOS process. And the 3.1~10.6 GHz ultra-wide band down-conversion mixer is implemented by TSMC 0.18μm CMOS process. As for the UWB VCO, it operates ranging from 12.92 GHz to 17.6 GHz with phase noise of -99.98 to -106.07dBc/Hz@1MHz, tuning range of 4.68 GHz, output power of -8.7 to -9.6dBm, and low power consumption of 6.3 mW and chip area of 0.77x0.62mm2. In the design of ultra-wide band down-conversion mixer, the down conversion mixer shows a conversion gain of 7.19 to 10.72 dB and the return loss of RF port is smaller than -10dB. Power consumption is only 26.3mW, and IIP3 of 0.8 dBm. The LO to IF isolation and LO to RF isolation are less than -30dB, respectively. The chip area is 1.14x0.89 mm2. All circuits have been designed and simulated by Advanced Design System (ADS) software. Acknowledgements

1595

This work was partly supported by the Ministry of Science and Technology of Taiwan, the Republic of China under contract nos. MOST 104-2221-E-182-074 and MOST 104-2221-E-182-057. The software and chip fabrication support from the Chip Implementation Center is also acknowledged. References [1] A. A. Abidi, ―Direct-conversion Radio Transceivers for Digital Communications‖, IEEE Journal of Solid-State Circuits, vol.30, no.12, pp.186-187, December 1995. [2] T. T. Y. Wong, Fundamentals of Distributed Amplification. Norwood, MA: Artech, 1993. [3] K. C. Juang, ‖Low phase Noise Ka-band Oscillator‖, Institute of Communication Nation Chiao Tung University, pp. 8-9, 1999. [4] D. Weaver, ‖A Third Method of Generation and Detection of Single-Sideband Signals,‖ Proc. IRE, vol. 44, pp. 1703-1705, 1956. [5] G. Gonzalez, ‖Microwave Transistor Amplifier-Analysis and Design,‖ 2nd Ed., Prentice Hall, Inc., 1984. [6] Jri Lee, Yentso Chen, and Yenlin Huang, ‖A low-power low-cost fully-integrated 60-GHz transceiver system with OOK modulation and on-board antenna assembly‖, IEEE Journal of Solid-State Circuits, vol.45 pp.264-275, Feb. 2010. [7] Mitra Gilasgar, ‖A low power 2.4GHz double-balanced CMOS sub-harmonic mixer‖, 1st International Conference on Intelligent System,Modelling and Simulation ,pp.418-422, 2010. [8] J.A. Hou and Y.H. Wang, ―A 7.9GHz low-power PMOS colpitts VCO using the gate inductive feedback‖, IEEE Microwave and Wireless Components Letters, vol. 20 pp. 223-225, April 2010. [9] S.L. Jang, M.H. Juang, and J.J.Chen, ― A 0.6-V low-power Armstrong VCO in 0.18μm CMOS‖, Microwave and Optical Technology Letters, pp.116-119,January 2010 [10] W.S. Feng, C.C. Wei, H.C. Hsu, C.M. Tsao, H.H. Li, and C.H. Chen,‖Green Design Technique for RF Front-end Circuits‖, International Conference on Green Circuits and Systems, 2010.

1596

ACEAIT-7607 A Multi-Agent System for Microgrid Optimal Operations Wei-Tzer Huanga*, Tsai-Hsiang Chenb, Jhih-Siang Yangb, Hong-Ting Chena, Yung-Ruei Changc, Yih-Der Leec, Yuan-Hsiang Hoc a Department of Industrial Education and Technology, National Changhua University of Education, Taiwan b

Department of Electrical Engineering, National Taiwan University of Science and Technology, Taiwan c

The Institute of Nuclear Energy Research, Taiwan

*Corresponding Author: [email protected] Abstract A multi-agent system (MAS) was proposed in this study to solve for the optimal grid-tied and islanding operations of a microgrid. This microgrid was the first outdoor microgrid test bed established by the Institute of Nuclear Energy Research (INER) in Taiwan. First, the MAS hierarchical architecture was designed to realize the optimal INER microgrid operations. This system consisted of the load, distribution energy resource, topology, zone, and microgrid main agents. Second, the functions of and information exchange among each agent was defined, and the optimal reconfiguration and power flow algorithms were developed for the optimal reconfiguration of INER microgrid operations under grid-tied and islanding modes. Finally, the proposed MAS was implemented with LabVIEW and MATLAB. Simulation results demonstrate that the proposed MAS is extensible and flexible to solve for the expandable microgrid. Keyword: Microgrid, Multi-Agent, Optimal Operation, Power Flow, Reconfiguration 1. Introduction The COP 21 UN Climate Change Conference aims to maintain the rise in global temperature well below 2 oC in this century and to limit it to 1.5 oC above pre-industrial levels [1]. The main cause of the rise in global temperature is CO2 emissions. Thus, reducing CO2 emissions is the key point to retard global warming. The statistics for global CO2 emissions in 2013 indicated that the electricity and heat generation sectors produced nearly 42% of the total CO2 emissions [2]. CO2 emissions by power generation cannot be ignored. The penetration of distribution energy resources (DERs) that consist of renewable and non-renewable energy resources in power systems, instead of

1597

large thermal power plants, has an important role to reduce CO2 emissions. Generally, distribution systems with DERs have become active distribution networks. These networks consider the operations of clustered microgrids [3–6] that are integrated in DERs and multi loads that operate as a single autonomous grid under grid-tied or islanded mode from the existing distribution grid. Thus, the operations of the microgrid are more complex than those of traditional distribution systems. To solve this problem, multi-agent systems (MASs) are applied in power system simulation and control [7–9] and microgrid operation and management [10–11]. In this study, a hierarchical MAS is proposed to solve for the optimal grid-tied and islanding operations of the Institute of Nuclear Energy Research (INER) microgrid. The organization of this study is as follows: Section 2 describes the INER microgrid and the proposed MAS. Section 3 demonstrates the simulation results of the proposed MAS. Section 4 presents the conclusions. 2. Description of the INER Microgrid and Proposed MAS 2.1 INER Microgrid The INER microgrid is the first hundred-kW scale test bed in Taiwan, as shown in Figure 1. Figure 2 shows the single-line diagram of the INER microgrid, which consists of three zones that include the microturbine (μT), energy storage battery system (ESS), high-concentrated photovoltaics (HCPVs), and wind turbines (WTs). The microgrid is connected to the Taipower distribution network by a static switch. DERs consist of 65 kWμT and 100 kW ESS; twenty-one 1.5 kW HCPVs are installed in Zone 1. One 65 kWμT, one 125 kW WT, and twelve 1.5 kW HCPVs are installed in Zone 2. In addition, a 65 kWμT, a 25 kW WT, and two 1.5 kW HCPVs are installed in Zone 3. Furthermore, the load installed in this system includes a 048-office building and five single-phase and three-phase load banks.

1598

2.2 Proposed MAS MAS is a modular, flexible, extensible, and autonomous technique in power system applications that is especially used in microgrid operations. The major element of MAS is the agent that is characterized by autonomy, social ability, reactivity, and pro-activeness; that is, this agent can act flexibly and autonomously to meet its design objectives [12]. In this study, a hierarchical MAS is proposed for microgrid optimal reconfiguration under grid-tied and islanding modes. The developed MAS consists of an MG main and topology agents and several zone, load, and DER agents. The function of each agent is as follows:  MG Main Agent: The function of this agent is to request network reconfiguration to the zone agent and then wait for the optimal network topology from the latter agent.  Zone Agent: This agent requests information to the topology, load, and DER agents and then waits for the network connection matrix, load demand, and power generation reply from each corresponding agent.  Topology Agent: This agent accepts the zone agent request and then replies to the network connection matrix to reach the zone agent.  Load Agent: The function of this agent is to accept the zone agent request and reply with the forecasting load demand to the zone agent.  DER Agent: This agent accepts the zone agent request and then replies with the prediction power generation of renewable units and available power of other non-renewable units to the zone agent. As mentioned above, the time sequence diagram of the developed MAS for the optimal microgrid reconfiguration is shown in Figure 3. For the optimal reconfiguration, the objection function is expressed as

1599

which is subject to

In (1), L represents the number of lines, and Ij represents the current of the jth line. In (2) and (3), Vi, VUi, and VLi denote the voltage at bus i and its upper and lower limits, respectively. Ij,max is the conductor ampere capacity. A is the bus incidence matrix, and its determinant is equal to 1 or -1 and represents radial topology. Graph theory and the backward/forward sweep method are applied to solve for the power flow based on the network topology [13–14].

3. Simulation Results The proposed MAS was implemented using MATLAB and LabVIEW to develop a graphic user interface. The simulation results are discussed in the next section. 3.1 Grid-Tied Operation The INER microgrid consists of three zones with DERs and loads and includes a tie switcher and 11 sectionalizing switchers. Figure 4 depicts the simulation results at midnight of one day

1600

under grid-tied operation conditions. The lights turning deep green on the PV units indicates that no power generation in the PV units occurs. Similarly, the light between buses 3 and 7 turning deep green indicates that the sectionalizing switch is opened. The total power generation, power loss, load demand, line data, and bus data are shown in each zone window.

3.2 Islanding Operation

1601

The simulation result at noon of one day under islanding operation condition is shown in Figure 5. Under this condition, PV units can provide electric power. The lights between buses 2 and 3 and buses 3 and 9 turning deep green illustrates that the sectionalizing switch is open. The total power generation, power loss, load demand, line data, and bus data are shown in each zone window.

1602

4. Conclusions A hierarchical MAS was proposed to solve for the optimal operations of the INER microgrid. Each agent has its own function, receives the request from an upper agent, and executes the corresponding program to generate results to the upper agent. Furthermore, the agents are able to exchange information among one another. After all the agent functions are integrated and coordinated, a graphic user interface is developed through MATLAB and LabVIEW. Each agent is an independent module and functions as a flexible, extensible, and autonomous technique for system extension in the future. Thus, the proposed MAS is the most suitable solution to microgrid operations. 5. Acknowledgment The authors are grateful for financial support from the Ministry of Science and Technology, Taiwan, under Grant MOST 105-3113-E-042A-004-CC2.

1603

References [1] Historic Paris Agreement on Climate Change, http://newsroom.unfccc.int/unfccc-newsroom/finale-cop21/. [2] International Energy Agency, Key Trends in CO2 Emissions from Fuel Combustion (2015 Edition). [3] R. H. Lasseter, ―Microgrids,‖ in Proceedings of the IEEE Power Engineering SocietyWinter Meeting, vol. 1, pp. 305–308, 2002. [4] R. H. Lasseter and P. Paigi, ―Microgrid: a conceptual solution,‖ in Proceedings of the IEEE 35th Annual Power Electronics Specialists Conference (PESC 04), pp. 4285–4290, June 2004. [5] P. Paigi, R. H. Lasseter, ―Autonomous Control of Microgrids,‖ 2006 IEEE Power Engineering Society General Meeting. [6] F. Katiraei, R. Iravani,N.Hatziargyriou, andA.Dimeas, ―Microgrids management,‖ IEEE Power and Energy Magazine, vol. 6, no. 3, pp. 54–65, 2008. [7] S. McArthur, E. Davidson, V. Catterson, A. Dimeas, N. Hatziargyriou, F. Ponci, and T. Funabashi, ―Multi-agent systems for power engineering applications—Part I: Concepts, approaches, and technical challenges,‖ IEEE Trans. Power Syst., vol. 22, no. 4, pp. 1743– 1752, Nov. 2007. [8] S. D. J. McArthur, E. M. Davidson, V. M. Catterso D. Hatziargyriou, F. Ponci and T. Funabashi, "Mul Power Engineering Applications—Part II: Technolo Tools for Building Multi-agent Systems," IEEE Trans. system, pp.1753-1759, 2007. [9] M. Nordman and M. Lehtonen, ―An agent concept for managing electrical distribution networks,‖ IEEE Trans. Power Del., vol. 20, no. 2, pt. 1, pp. 696–703, Apr. 2005. [10] H.-M. Kim, Y. Lim, T. Kinoshita, ―An intelligent multiagent system for autonomous microgrid operation,‖ Energies 2012, 5, 3347–3362. [11] J. Oyarzabal, J. Jimeno, J. Ruela, A. Engler and C. Micro Grid Management System," IEEE Internat Future Power Systems, pp.1-6, 2005. [12] M. Wooldridge, An introduction to multiagent systems,: Wiley, 2009. [13] T. H. Chen and N. C. Yang, ―Three-phase power-flow by direct ZBR method for unbalanced radial distribution systems,‖ IET Generation, Transmission & Distribution, vol. 3, no. 10, pp. 903-910, March2009. [14] J. H. Teng, ―A network-topology based three: phase load flow for distribution systems,‖ Proceedings of National Science Council ROC (A), 2000. vol. 24, no. 4, pp. 259-264.

1604

ACEAIT-7611 Assessment of Upgrading Premise Wiring Systems from Single-Phase 2-Wire to Single-Phase 3-Wire System Rih-Neng Liaoa, Tsai-Hsiang Chenb ab Department of Electrical Engineering, National Taiwan University of Science and Technology, Taiwan, R.O.C. a E-mail: [email protected], b [email protected] Abstract Single-phase 2-wire (1ϕ2W) and single-phase 3-wire (1ϕ3W) circuit layouts are common in premises wiring. Single-phase 2-wire circuit arrangement is mainly adopted for serving lighting and small electrical appliances in Taiwan. However, the line loss of 1ϕ2W branch circuit arrangement is much high than that of 1ϕ3W circuit. In this paper, for supporting the upgrading of premises wiring from 1ϕ2W to 1ϕ3W arrangement, we evaluate and compare the total line loss of a premise wiring system with these two types circuits in a sample dwelling unit. Develop a reliable and efficient power flow program as a tool for this study is very important because the system performance of 1ϕ2W and 1ϕ3W premises wring systems can then be evaluated in detail, such as node (bus) voltages, branch (line) currents and power losses. The results indicate that upgrading the premise wiring system from 1ϕ2W to 1ϕ3W arrangement can reduce line loss up to 16.57% for a sample dwelling unit. The results of this study provide a valuable reference for planning and design an energy efficient and low carbon premises wiring system in residential, multi-dwelling Unit (MDU) and small business premises. Keywords: single-phase 2-wire, single-phase 3-wire, feeder, branch circuit, premise wiring system, energy efficient, low carbon, modelling and simulation. 1. Introduction 1.1 Background In the past two decades, economic development and social progress has been amazingly speeded. Investment in science, technology and innovation (STI) is essential for economic development and social progress. Accordingly, energy demands in all the residential, commercial, industrial, and transportation departments have grew rapidly. For improving the system energy efficiency, energy efficient solution has become an essential issue for a modern home energy management system (HEMS). Since the modern HEMS is usually required to operate real-time, developing an efficient system analysis tool is essential to provide the vital information and data, such as bus voltages, line currents, and power losses for system management and energy conservation.

1605

1.2 Literature Reviews There are many researchers trying to solve power flow problems efficiently from different viewpoints. In [1-5], the authors introduce traditional power-flow solutions for solving an ordinary electrical power system. These traditional power-flow methods can be divided into two categories, bus frame of reference and branch frame of reference methods. In bus frame of reference method, the bus (node) voltages and bus currents are the main system variables. The Z-bus method and Newton-Raphson method are in this category [1-3]. In branch frame of reference method, the branch voltages and branch currents are the main system variables. The backward/forward sweep method and loop impedance method are inside this category [4-5]. In [6], the authors develop a bus-injection-to-branch-current (BIBC) matrix and a branch-current-to-bus-voltage (BCBV) matrix to directly solve power-flow problem for a regular power distribution system. Building of the BIBC matrix and BCBV matrix is quite easy and the injection-current source and admittance matrix can be obtained directly from the input data of the system components. Moreover, to deal with the different natures of network topologies of distribution systems, some novel methods based on branch frame of reference and loop frame of reference are proposed in the latest years [7-8]. 1.3 Objectives and Goals In this paper, we applies BIBC and BCBV matrixes to solve power-flow problems of a 1ϕ2W and 1ϕ3W premises wiring system. The detailed energy consumption and line losses of a typical premises wiring system with 1ϕ2W and 1ϕ3W arrangements are summarized and compared to assess the energy efficient of a residential premises with 1ϕ2W and 1ϕ3W arrangement. The usage pattern and nature distribution along the branch circuits of the lighting and appliances loads in a residential, multi-dwelling units, and small business premises are taken into account. The daily, weekly, monthly, and annual energy consumption and system losses are evaluated in detail through a typical premise wiring system of a sample dwelling unit. 1.4 Paper Organization In this paper, Section 1 briefly introduces the background and objectives of this paper. Some related literatures to this study are also reviewed in this section. Section 2 presents a systematic approach method for the energy consumption and line losses evaluation of a premises wiring system. Section 3 demonstrates the effectiveness and major results of this study with some cases studies. The evaluation results show the superiority of upgrading premise wiring systems from 1ϕ2W to 1ϕ3W arrangement. Moreover, Section 4 draws a brief conclusions for this paper. 2. Energy Consumption and Line Losses Assessment

1606

In this paper, annual energy consumption and annual system loss are obtained step by step. First, we obtain the deterministic energy consumption and line loss solutions, and then the daily, weekly, monthly, and finally the annual energy consumption and system loss. The method and process are descripted as follows. 2.1 Deterministic energy consumption and line loss evaluation To solve accurately the deterministic energy consumption and system losses, the characteristics of discrete loads along branch circuits should be considered. Determination of load characteristics for transient performance study are introduced in [9]. The load characteristics of some common appliances such as air conditioner, duct heaters, water heater, deep fryer, clothes dryer, refrigerator, incandescent lights, fluorescent lights, etc. can be found in [9]. 2.2 Daily energy consumption and line losses calculation According to the specifications and usage patterns of individual lighting and all kind of electrical appliances, a typical daily-load curve should be developed. A proper time interval should be specified for demand calculation. In this study, the time interval is 15 minutes that are adopted by Taipower and many power companies worldwide. Hence, the assessment of daily energy consumption and losses will have 96 deterministic energy consumption solutions. 2.3 Weekly energy consumption and losses calculation The weekly energy consumption and losses can be obtained by using the arithmetic calculations of its associated daily energy consumption and losses. The holidays during a week should be taken into account. Moreover, the factor of the daily load curves may different from day to day during a week should be taken into account. In other word, at least, the different of daily load curves of weekday, weekend and holidays should be considered. 2.4 Monthly energy consumption and losses computation The monthly energy consumption and losses can be obtained by using the arithmetic calculations of its associated daily and weekly energy consumption and losses. The number of days and holidays during a month should be taken into account. 2.5 Annual energy consumption and losses computation The annual energy consumption and losses can be obtained by using the arithmetic calculations of the monthly or seasonal energy consumption and losses. The power supply for lighting and appliances, or combinations of such loads are different for month to month and season to season. This factor, as well as the effect of holidays, should be taken into account for the computation of annual energy consumption and system losses. 3. The Case Study

1607

In a dwelling unit, many kinds of lamps and appliances are connected to branch circuits at different tapped-off points, the receptacles and outlets, such as incandescent lamps, fluorescent lamps, high-intensity discharge (HID) lamps, dishwashers, hair dryers, TV, small fan, etc. The sample dwelling unit, shown in Fig. 1, is used to demonstrate the evaluations of the deterministic, daily, weekly, monthly, and annual energy consumption and losses, and verify the superior of 1ϕ3W circuits to 1ϕ2W circuits.

The total installed capacity of connected loads for each branch circuit of the sample dwelling unit is listed in Table 1. The loads connected between phase A of branch-circuit conductors to neutral and phase B to neutral are 3300VA and 3200VA, respectively. That is the loading of phase A and phase B of branch-circuit conductors are near balanced. The key parameters of branch-circuit conductors are shown in Table 2. The 1ϕ2W circuit arrangement dominates the premise wiring systems in Taiwan. The conductor size of branch circuits for lighting and for small appliances are 2.0 mm single-wire conductor, while the conductors of branch circuits for cooking appliances, fixed lighting units with heavy-duty lampholders, infrared heating units, etc. are 5.5 mm2 stranded-wire conductor.

1608

The analysis results of energy losses of 1ϕ2W system are shown in Table 3. The results show the seasonal energy losses of the sample dwelling unit are 367.58 kWh, 677.93 kWh, 453.39 kWh, and 295.12 kWh, in spring, summer, fall, and winter, respectively. The air-conditioner is the largest load in the summer season in Taiwan, therefore, the total energy loss of the summer season is much higher than that of other seasons. For improving the system energy efficiency of the premises wiring system, upgrading original 1ϕ2W system to 1ϕ3W system is a good approach. The analysis results of energy losses of 1ϕ3W system are shown in Table 4. These evaluation results verify that the corresponding system energy losses of the premises wiring system are reduced when the 1ϕ2W system is upgraded to 1ϕ3W system. For example, the seasonal energy loss in summer period is reduced from 677.93 kWh to 549.76 kWh, i.e. energy loss being reduced about 18.91%. Moreover, Tables 3 and 4 also show that the annual energy losses are reduced from 1,794.03 kWh (1ϕ2W system) to 1,496.5 kWh (1ϕ3W system), i.e. reducing about 16.57%.

1609

4. Conclusions The detailed energy consumptions and system losses of premises wiring system of a typical dwelling unit evaluated using a precise power flow approach are presented in this paper. The results verify the superior of 1ϕ3W system. For the sample dwelling unit, the annual energy loss can be reduced 16.57% from 1,794.03 kWh of 1ϕ2W system to 1,496.5 kWh of 1ϕ3W system, saving energy 297.53 kWh. Assumed 10,000 new dwelling units adopt the 1ϕ3W system instead of 1ϕ2W system per year. The annual energy loss reduction of the first year will be about 3×106 kWh. After five years, the total energy reduction will be 45×106 kWh. The results of this study provide of valuable reference for planning and design an energy efficiency and low carbon premises wiring systems for a residential, multi-dwelling units and small business. Acknowledgements The authors would like to thank the Ministry of Science and Technology, R.O.C., for financially supporting this research under Contract (No. MOST 103-2221-E-011-079-MY2). References [1] T.H. Chen, M.S. Chen, K.J. Hwang, P. Kotas, E.A. Chebli, Distribution system power flow analysis – a rigid approach, IEEE Trans. Power Delivery, 6 (3), 1146–1152, 1991. [2] W.F. Tinney, C.E. Hart, Power flow solution by Newton‘s method, IEEE Trans. Power Apparatus and Systems, 86, 1449–1460, 1967. [3] Van Amerongen RAM, A general purpose version of the fast decoupled load flow, IEEE Trans. Power Syst., 4 (2) 760–770, 1989.

1610

[4] D. Shirmohammadi, H.W. Hong, A. Semlyen, G.X. Luo, A compensation based power

[5] [6] [7] [8]

flow method for weakly meshed distribution networks, IEEE Trans. Power Syst., 3 (2) 753–762, 1988. G.X. Luo, A. Semlyen, Efficient load flow for large weakly meshed networks, IEEE Trans. Power Syst., 5 (4) 1309–1316, 1990. J.H. Teng, A Direct approach for Distribution System Load Flow Solutions, IEEE Trans. Power Delivery, 18 (3) 882–887, 2003. T.H. Chen, N.C. Yang, Three-phase power-flow by direct ZBR method for unbalanced radial distribution, IET Generation, Transmission & Distribution, 3 (10) 903–910, 2009. T.H. Chen, N.C. Yang, Loop frame of reference based three-phase power flow for unbalanced radial distribution systems, Electric Power Systems Research, 80 (7) 799–806, 2010.

[9] Energy Systems Research Center, UTA, Determining Load Characteristics for Transient Performance, Final Report on EPRI Project RP 849-3, I,II,III, December 1978.

1611

ACEAIT-7610 Assessment of Maximum Allowable Capacity of EVs along a Distribution Feeder Tsai-Hsiang Chena, Rih-Neng Liaob ab Department of Electrical Engineering, National Taiwan University of Science and Technology, Taiwan, R.O.C. a E-mail address: [email protected], b [email protected] Abstract This paper aims to explore the maximum allowable charging capacity of EVs that can be charged simultaneously by a distribution feeder. The limitations of steady-state voltage deviation (SSVD) and the feeder thermal capacity (FTC) should be met. In this paper, a program for determining the maximum allowable capacity of a distribution feeder for charging EVs simultaneously were developed by a commercial software, MATLAB. The results of this study can provide a valuable reference for power utility to properly planning, design and operate a distribution feeder with plenty of EV-charging stations. Besides, it is of value for the government to promote the electric vehicle policy. Keywords: electric vehicle, charging station, maximum allowable capacity, distribution feeder, steady-state voltage deviation, feeder thermal capacity. 1. Introduction 1.1 Background There are growing concerns over energy security and greenhouse gas emissions. Transportation electrification is considered as one of the important manners to solve these concerns. More and more electric vehicles (EVs) will therefore be integrated with electrical power system through plug-in hybrid and electric vehicles. The integrations need a lot of charging facilities at public and private places. The installation rules of charging infrastructure ae essential for the mass implementation of EVs as discussed in [1]. Moreover, the detailed requirements and conditions that should be considered for the design and construction of charging infrastructure systems for EVs are also presented in [1]. 1.2 Literature Reviews Many researches have directed on the issues of power quality impacts and power demand impacts of the charging of electric vehicles on the power distribution network [2-4], and contribution of electric vehicles for energy management in a smart house with intermittence power generation, including V2G (vehicle to grid) and V2H (vehicle to house) subjects [5]. The literatures listed above indicate that the main impacts on the power distribution network of

1612

EV charging, especially large-scale charging stations, are harmonic pollution, voltage fluctuation and feeder thermal capacity. On the other hand, some studies have proposed techniques from two points of view for controlling the charging of EVs; that are the system point of view [6–7] and the distribution network point of view [8–11]. The objective of [12] is to maximize the integration level of EVs on a medium voltage (MV) grid, by increasing the number of charging EVs in a stepwise manner until the grid became saturated. Moreover, probabilistic methods or robust optimization techniques are employed where uncertainties are a concern [13-16]. 1.3 Objectives and Goals With the widespread of electric vehicle, the impact of the charging of electric vehicles on the feeder cannot be ignored, especially the impact of large-scale EV-charging station at the downstream feeder segment. This study explored the maximum allowable capacity of a distribution feeder for charging EVs simultaneously. That determination of maximum allowable capacity is mainly based on the satisfactions of the power quality and thermal limit of the feeder connected. The proposed approach algorithm and simulation results could provide a valuable reference for power engineers involved in plan, design, and operation of distribution networks with plenty of EV-charging stations. 1.4 Paper Organization In this paper, Section 1 ―Introduction‖ briefly introduces the background and objectives of this paper. Some related literatures to this study are also reviewed in this section. Section 2 ―Solution Approach‖ describes some related standards of voltage quality, proposed solution method, and major factors affecting the steady-state voltage deviation. Section 3 ―A Case Study‖ portrays a sample feeder and its parameter, and presents a summarized simulation results. A brief discussion on the simulation results is also included in this section. Finally, Section 5 ―Conclusion‖ draws a concise conclusions of this paper. 2. Solution Approach 2.1 Related Standards The term ―power quality‖ is widely applied to describe the electromagnetic phenomena in power system. A definition of power quality is given in IEEE Std. 1159. The power quality issues based on several aspects can be cataloged as transients, shot duration variations, long duration variation, waveform distortion, voltage imbalance, voltage fluctuation and power frequency variations, etc. In a distribution system, the system operating conditions and loads change from time to time; therefore, the voltage profiles may not exactly equal to its desired value. The small variations from the desired value are called ―voltage deviation‖. The voltage

1613

sag, voltage swell and short interruption are classified as short duration variation; while the sustained interruption, under voltage and overvoltage are classified as long duration variation. Many major voltage quality problems are defined and described in IEEE Stds. 141, 519, 1100, and 1159. For different country may have different definitions and corresponding rules or regulations. For the definition and corresponding requirements of steady-state voltage deviation (SSVD) can be found in the ANSI standards. The nominal voltage rating between 100 V to 230 kV can be found in ANSI C84.1, and nominal voltage rating above 230 kV is in ANSI C92.2. Electrical power suppliers are commonly accorded to ANSI C84.1 for electrical power delivery. 2.2 Deterministic Approach Equations. (1)-(2) can be used to assess the SSVD for the charging of EVs on the distribution feeder.

where

Deterministic approaches are widely used to evaluate the SSVD caused by the charging of grid-connected EVs. However, they are too simplified to take into account all the system operating and loads conditions. More, the evaluation methods are usually not suitable for two or more EV-charging stations along a feeder. Hence, power flow calculation are commonly required for considering the network configuration, system operation conditions and load characteristics.

1614

2.3 Major Factors Affecting the Steady-State Voltage Deviation Operating a grid-connected EV-charging station may lead to SSVD on the connected system. The major factors that affect the SSVD can be categorized as impedance- and current-sensitive factors, as shown in Table 1.

3. A Case Study 3.1 Parameters A practical distribution system of the Taiwan Power Company (Taipower) is used as a sample system, as shown in Fig. 1. The feasible ranges of system parameters are listed as follows. The short circuit capacity (SCC) at the primary side of the substation transformers are typically between 400 MVA and 1200 MVA. The voltage level along the main feeder is 11.4 kV. The X/R ratio of the equivalent impedance at the primary side of the substation transformers are typically between 6.0 and 6.5. The rated capacity of substation transformers are 12.5 MVA and 25 MVA. The percent impedance of the substation transformers are typically between 5 % and 15 %. The X/R ratio of the substation transformer are typically between 10 and 20. The length of the main feeders are typically between 5 km and 20 km, in which the lengths of each line segment (L1 to L10) are generated randomly by the analysis program. The conductor of a feeder is 477 ACC overhead line and its unit length impedance is 0.131+j0.364 Ω/km. The total loads along the main feeder are between 0.6 MW and 3 MW, and the power factors for all loads are between 0.8 lagging and 1.0 (unity power factor). However, the distribution types of discrete loads along a feeder (such as: increasing, uniform, and decreasing distribution) are generated randomly by the analysis program. The total loads for other feeder are between 2 to 9 MW and they are represented by a lumped load connected to the secondary side of substation transformer, as well as the power factor is assumed to be 1.0 because of the power factor always corrected to near 1.0 (unity power factor) in Taipower distribution system.

1615

3.2 Results and Discussions The maximum allowable charging capacity of EVs along a distribution feeder is commonly restricted. The impact factors on the maximum allowable charging capacity include local standard, feeder thermal capacity (FTC), system topologies and features, and load conditions etc. All EVs are assumed to be unity power factor. The limitation of the SSVD (as a percentage of the nominal voltage) is assumed 2.5% and 5.0%, separately. The current of feeder thermal capacity is presumed 600 A. The analysis results of the maximum allowable charging capacity of EVs under above assumptions are shown in Fig. There are 2.5,000 data sets of system parameters that are randomly generated to use to obtain the solution sets, marked ―+‖ as shown in Fig. 2. The simulation results are explained as follows. When the EV-charging station is connected to a distribution feeder at a tapped-off point with system short-circuit capacity of 40 MVA, the maximum allowable charging capacity of the EV-charging station is 2.67 MW, mainly restricted by the limitation of SSVD of 2.5%. Hence, if the EV charging power of this EV-charging station is less than 2.67 MW, the charging of EVs shall be permitted, otherwise shall be rejected. Moreover, for a charging station with peak charging power of 3.5 MW should be connected to the power distribution grid at a tapped-off point with system short-circuit capacity of more than 50 MVA, as indicated in Fig. 2. For simplicity, some key results of Fig. 2 are summarized in Table 2. The system short-circuit capacity is gradually decreasing from the power substation to the end of the feeder, due to the impedance of the feeder conductor. Therefore, more EVs can be charged if the charging station near the power substation. On the contrary, if the charging station is located far away from power substation, the charging capacity or number of EVs that can be charged simultaneously will be firmly restricted. The maximum allowable charging capacity of a feeder can be expanded by using larger size feeder conductor, increasing rated

1616

capacity of the main transformer, and upgrading the system voltage level, etc. In additional, if the system parameters and operation conditions of the simulation cases are all within the predetermined ranges given by a utility, the simulation results can cover most cases for the installation of EV-charging station in this typical sample system. Therefore, power utility should set up their own system parameters and operation conditions for speeding review process of the application of installation of EV-charging station.

4. Conclusions In this paper, evaluation algorithm of the maximum allowable charging capacity of EVs on a distribution feeder is proposed for dealing with the problem of uncertainty in the operation of a power grid. The proposed algorithm also makes it possible to consider the feasible ranges of

1617

the system parameters and operation conditions throughout the entire power system or within a specific subsystem. The results of this study can provide a valuable reference for power utility to properly planning, design and operate a distribution feeder with plenty of EV-charging stations. It is of value to the government to promote the electric vehicle policy. Acknowledgements The authors would like to thank the Ministry of Science and Technology, R.O.C., for financially supporting this research under Contract (No. MOST 103-2221-E-011-079-MY2). References [1] C.E. Hatton, S.K. Beella, J.C. Brezet, et al. Charging station for urban settings: the design of a product platform for electric vehicle infrastructure in Dutch cities, World Electric Vehicle Journal, (3) 1–13, 2009. [2] J.C.Gomez, M.M Morcos, Impact of EV battery chargers on the power quality of distribution systems, Power Delivery, IEEE Transactions on , 18 (3) 975–981, 2003 [3] X.U. Lizhong, Y. Guangya, X.U. Zhao ,F. Marra, C. Trehol, Impacts of Electric Vehicle Charging on Distribution Network in Denmark, Automation of Electric Power System, 35 (14) 18–23, 2011. [4] L. Trovao, H.M. Jorge, Power demand impacts of the charging of Electric Vehicles on the power distribution network in a residential area, Energetics (IYCE), Proceedings of the 2011 3rd International Youth Conference on , 1–6, 2011. [5] Y. Ryuichi, Y. Kouichirou, M. Takahiko, K. Haruhiko, Contribution of Electric Vehicles for Energy Management in Smart House with Uncertain Power Generation, Automation [6] [7]

[8]

[9]

of Electric Power System, 35 (22) 18–24, 2011. A. Shortt, M.O. Malley, Quantifying the long-term power system benefits of electric vehicles, in Proc. IEEE PES Innovative Smart Grid Technologies (ISGT), 1–6, 2012. P. Mitra, G.K. Venayagamoorthy, Wide area control for improving stability of a power system with plug-in electric vehicles, IET Generation, Transmission & Distribution, 4 (10) 1151–1163, 2010. S. Acha, T. C. Green, and N. Shah, Effects of optimised plug-in hybrid vehicle charging strategies on electric distribution network losses, in Proc. IEEE PES Transmission and Distribution Conf. and Exposition, 1–6, 2010. S. Deilami, A.S. Masoum, P.S. Moses, M.A.S. Masoum, Realtime coordination of plug-in electric vehicle charging in smart grids to minimize power losses and improve voltage

profile, IEEE Transactions on Smart Grid, 2 (3) 456–467, 2011. [10] G.B. Shrestha, B.C. Chew, Study on the optimization of charge-discharge cycle of electric vehicle batteries in the context of Singapore, in Proc. Australasian Universities Power Engineering Conf. (AUPEC), 1–7, 2007.

1618

[11] S. Paudyal, S. Dahal, Impact of plug-in hybrid electric vehicles and their optimal

[12] [13]

[14] [15]

deployment in smart grids, in Proc. Australasian Universities Power Engineering Conf. (AUPEC), 1–6, 2011. J. Lopes, F. Soares, P. Almeida, Identifying management procedures to deal with connection of electric vehicles in the grid, in Proc. IEEE PowerTech, 1–8, 2009. L. Kelly, A. Rowe, P.Wild, Analyzing the impacts of plug-in electric vehicles on distribution networks in British Columbia, in IEEE Electrical Power & Energy Conf. (EPEC), 1–6, 2009. G. Li, X.-P. Zhang, ―Modeling of plug-in hybrid electric vehicle charging demand in probabilistic power flow calculations,‖ IEEE Trans. Smart Grid, 3 (1) 492–499, 2012. P. Papadopoulos, S. Skarvelis-Kazakos, I. Grau, L. Cipcigan, N. Jenkins, Electric vehicles‘ impact on British distribution networks, IET Elect. Syst. Transp., 2 (3) 91–102, 2012.

[16] A. H. Hajimiragha, C. A. Cañizares,M.W. Fowler, S.Moazeni, A. Elkamel, A robust optimization approach for planning the transition to plug-in hybrid electric vehicles, IEEE Trans. Power Syst. 26 (4) 2264–2274, 2011.

1619

ACEAIT-7543 Visual Imaging Technology for Urban Flood Monitoring and Warning Applications Lun-Chi Chen, Chien-Hao Tseng, Fang-Pang Lin, Jyh-Horng Wu* National Center for High-Performance Computing, National Applied Research Laboratories, Taiwan E-mail: [email protected] 1. Background Traditional video sensors operated in harsh environments are typically affected by changes in weather conditions such as haze, fog, and rain. Consequently, the video images are often blurred or degraded, thus increasing the probability of identification errors in hazard mitigation. To use visual imaging technology to improve the degraded images taken in hazy weather, this study proposes a single image dehazing method for urban flood monitoring. The study was conducted using an urban flood monitoring system with video streaming to ensure the efficiency of the proposed method in improving the accuracy of flood detection and monitoring. The results showed that identifying flooded areas and removing haze from images is crucial for applying image-based flood alarm modules in disaster-monitoring applications. 2. Methods This study investigated the implementation of an image-based flood alarm module (IFAM) that employs image dehazing and flooded detection. The IFAM is an image-based flood alert system that facilitates proactive monitoring, assessment, and early response to rising water levels and related inundation in near real-time. The IFAM combines on-site real-time video images with a backend image-processing module to analyse river overflow and ground inundation. The water overflow range is then calculated using an image-processing module, after which the system automatically issues a flood alarms if the overflow range intrudes the preset warning signs. Typically, in a hazy image, the haze density is higher in regions with greater depth and lower in regions with shallower depth. To estimate the haze density and removal, we first calculated the initial haze transmission by using the dark channel prior to generate reasonable transmission maps. Therefore, a bilateral filter was used to smooth the small-scale textures in the image. Finally, the light gain in the generated haze-free image can be increased by multiplying the albedo by the atmospheric light according to the haze-imaging model. With the generated haze-free image, region-based image identification methods become more suitable for flood region detection. The region-based image identification methods produced using a combination

1620

of mean-shift and region growing algorithms are presented in this paper to facilitate urban flood monitoring. 3. Conclusion The proposed IFAM was used to evaluate urban flood areas to facilitate selecting the optimal image processing and flood identification methods. All the surveillance images of flooded areas captured from streaming video were evaluated. To obtain measurements in an urban environment, field experiments were conducted on the Nan-Hu No. 2 Bridge. The Nan-Hu No. 2 Bridge data were recorded during Typhoon Sinlaku, which induced a rainstorm in the study area on September 14, 2008. The video images were captured between noon (when the rain began) and nightfall (flood tiding). Fig. 1 shows surveillance images of the Nan-Hu No. 2 Bridge, which were captured using the proposed method. The images show that the dehazing method was effective in removing the haze, even in the region where the haze was dense. Hence, through the use of the dark channel prior, the thickness of the haze can be directly estimated and a high-quality haze-free image can be obtained.

Fig 1. shows the dehazing results based on a single image, in which the flood identification methods were employed to evaluate the urban flood areas of the current overflow range. Fig. 2 shows surveillance images from the river data set, which were obtained using region-based image identification methods.

1621

Experimental results indicated the suitability of the proposed IFAM, which can effectively detect and remove haze artefacts from a single image, and identify flood hazard areas in urban video surveillance images.

Keywords: image dehazing, image identification, flood alarm system, urban flooding

1622

ACEAIT-7424 Effects of Virtual Reality Presentation Media on Operator Learning Performance in the Main Control Room of the Nuclear Power Plant Chiuhsiang Joe Lin a*, Rou-Wen Wang b, Chih-Wei Yang c a Department of Industrial Management, National Taiwan University of Science and Technology, Taiwan b Department of Industrial and Systems Engineering, Chung-Yuan Christian University, Taiwan c Institute of Nuclear Energy Research, Taiwan *Corresponding Author: [email protected] Abstract Nuclear power plant (NPP) systems are complex and require high reliability. Through effective training, operators can understand the nuclear power systems and further establish better interaction between human operators and the system, in order to reduce human errors and to ensure their safety in the control room of NPP. In this study, virtual control room is thus developed using the virtual reality (VR) technology to help the training process. The control room of Lung-Men NPP of Taiwan was constructed with VR technology in this study with the objective of comparing the performances of two VR representation methods (Desktop VR and Projection VR). A searching task was planned in which the operators have to find out the objects appointed by the experimenter in the virtual interface of the main control room. The time to complete the task was collected as dependent variables in this experiment. The subjects have to complete the questionnaire that was developed for evaluating the usability of the virtual interface of main control room (MCR) after finishing the experiment. The result showed that the performance of the virtual interface of NPP presented by the VR projector was better than the desktop. Keyword: Virtual Reality, Usability, Training, Desktop VR, Projector based VR 1. Introduction In today‘s civilized society, people‘s livelihood and industries both highly depend on supply of electricity. This demand is too big for researchers and technologists come with different modes of energy supplies. One of the recent and huge potential in energy development is electric power from the nuclear power plants. Its development brought several advantages such as environmental protection, economic development, and reliable supply. However, concerns exist on safety of nuclear power plants because of past fatal incidents such as The Three Mile Island and Chernobyl accidents [1]. The Three Mile Island accident and Chernobyl accident revealed that operator performance in the main control room (MCR) is directly related to the

1623

safety of the nuclear power plant.

Therefore, it is very critical to ensure the operators in the

MCR can maintain the safety of nuclear power plant. Engineering technology plays an important role in NPP that requires complex technical equipment and interfaces to achieve public security and safe working environment. With the technical development of computer hardware and software, the traditional nuclear power plants (NPP) have been replaced by digital systems. Gradually, Lung-Men NPP is building auto and semi-auto digital operation systems which can help operators to conduct surveillance, maintenance, testing, inspection, and recovery of nuclear power plants (ABB, 1989; AEC, 1989; GE Nuclear Energy, 1992; Westinghouse Electric, 1992). Even if the automatic systems help reduce manual operation needed, the operators still should strive to monitor the condition of nuclear power plant at any time. Huanga et al. [2] indicated that the error in the process of interaction with the interface might be encountered when the system change from traditional to digital, because there exists difference in operation interfaces between the traditional and digital systems. The operators‘ conception of the traditional operation interface has not been used to support the digital operation interface, such as the control system, leads to change demands of decision-making, and the configuration condition of the processing interface. Therefore, operators who were used to the traditional interface of MCR have to retrain, in order to ensure the high operation safety about digital MCR. The pre-employment training is an important stage at the NPP. Through training, operators can not only understand the nuclear power system but also establish the fit between human operators and the system, in order to reduce human errors and to ensure safety of the MCR of NPP. In addition, training can affect the operating performance of NPP. However, the operator trainings for the control room of NPP are difficult and time-consuming. In simulation-based training, learners can experience environments that would be too expensive, time-consuming, complexity or dangerous to provide through other methods [3]. Virtual MCR is thus developed using the virtual reality technology to help the training process. Virtual reality (VR) has arisen as a tool that involves computer technologies to create digital prototypes in order to provide high-quality three-dimensional images, interactivity, and increased telepresence for users. VR technologies have been shown to provide invaluable training in the performance of complex procedural tasks. Lehmann et al. [4] developed the virtual endoscopic surgery trainer (VEST), and compared with conventional video trainer (CVT). The result indicated that the learning effectiveness presented by the VEST was as good as the CVT. In addition, Lin et al. [5] also applied VR technology to train Computer Numerical Control (CNC) milling operations, and examined the differences of learning effectiveness between the VR training system and the traditional. The result showed that the operators were more familiar with operating task by

1624

virtual training system, and reduced the frequency of error. In recent years, there has been a dramatic proliferation of research concerned with the development about virtual system for NPP for the purpose of training [6-8]. For example, The training of workers for safety purpose in nuclear plants, Mól et al. [8] built the NPP with the VR technologies by which users can navigate simultaneously in this virtual environment in first or third person view. The virtual models of NPP were applied in order to better prepare nuclear inspectors and to make real-life plants safer. Generally, VR system can be classified based on input/output equipments employed as Desktop VR, Immersion VR, Projection VR, and Simulator VR, where Desktop VR and Projection VR were the mostly applied. Tyndiuk et al. [9] found difference in performance between the large displays and small display (a desktop monitor). However, the kind of VR system for the virtual MCR of NPP which can give users realistic immersive context as in the real environment and helpful in training performance is unknown. Besides, general training system was used to achieve the effectiveness of learning and memory by the particular human-computer interface (HCI) that users can interact with context. Shneiderman [10] indicated that HCI plays a role that reduces a gap between people and system in the operation environment. An adaptive interface can improve the performance of human operators by aiding in information management and other cognitive tasks in real-time. Hu et al. [11] estimated the effectiveness of six different interface of information retrieval systems and confirmed that interface design may have a significant effect on system–user concept communication. Therefore, interface design is a key for successful system, because the interface design can influences the operation performance and users level of presence in the system. The evaluation of human-computer interaction of training interface was important especially, because the operation of MCR was complicated. In this study, a desktop and projection based VR system was developed for the control room of Lung-Men NPP of Taiwan. In addition, usability of the virtual interface was evaluated based on the questionnaire developed for the study. 2. Method The objective of this study was to investigate the difference between desktop-based and projector-based VR systems on the performance of training operators in MCR and present the best adaptive system. 2.1 Subjects Twenty subjects aged between 22 and 28 took part in this study. All subjects can use computers well, but they don‘t have the experience about interface of instrumentation and control (I&C) systems in the nuclear power plant (NPP). In addition, all subjects had normal or

1625

corrected-to-normal vision, and they have no sicknesses about eyes and limbs, such as color blindness and disablement. The subjects were divided into two groups randomly. 2.2 Apparatus The control room of Lung-Men NPP was simulated as a virtual environment in this study as shown in Fig. 1. The virtual control room of NPP was built by Cinema 4D, Inventor, and Unity 3D. The virtual environment was presented by a desktop computer and a projector (ViewSonic PJD6251) respectively. The experiment ran on Intel Core 2 Quad CPU Q9400 2.66GHz with a ENGTX180 graphics card. In the desktop condition, the subjects viewed on 24 inch monitor (ViewSonic Optiquest Q241wb) with 1920ⅹ1200 resolution. Subjects use mouse to interact with the system.

2.3 Task During the experiment, subjects have to navigate the virtual control room of NPP based on their group assignment, including the desktop group and the projector group. In the desktop condition, the virtual control room of NPP was presented by the desktop, and the subjects sat 50 cm from the monitor. In the projection condition, the virtual control room of NPP was projected on the screen by the projector that was placed at 200 cm from the screen, and the subjects stood at the side of the projector. All subjects can interact with the virtual environment by a mouse, such as zoom in, zoom out, and turn right or left to rotate the image, in order to view the wide display panel (WDP) of the MCR in detail (Fig. 2). The subjects were asked to carry out a searching task. They have to find icons, buttons or information on the WDP, such as the word ―SHUT DOWN‖, the button ―TRIP‖, the data about the suppression pool. There were 15 searching test items out of which 10 items were sampled randomly for each subject in the task. An experimenter asked the subjects the test question, and recorded the complete time. The total completion time was collected as the performance. In addition, the subjects have to complete a questionnaire designed to test the interface usability to users.

1626

2.4 Procedure The experiment was conducted in a laboratory setting and began with the administration of a short survey designed to collect background information on each subject. In particular, the subjects‘ prior experiences about virtual reality or 3D interface operation. The experimenter explained the tasks and in the virtual interface and the general purpose of the study. Training and exercise were also given to each subject in the corresponding group. The experiment was begun once the subject can wander in the virtual environment well by the mouse. The pre-test that indicated the subjects moving the image to designate place was executed in order to confirm the subjects familiar with the operation. About 5 minute were allocated for subjects to pass the pre-test. During the experiment, participants were asked to find information as indicated by the experimenter, and then the total completion time time was collected. Finally, the subjects had to complete the evaluation about interface usability after the experiment. 2.5 Questionnaire The usability was measured by an indirect method by observing some particular attributes, because it can‘t be measured directly. Nielsen [12] defined usability as the level a system can satisfy users‘ requirements. In other words, it is a survey to investigate users‘ viewpoints on whether the system can help them complete a task well. The five attributes of usability that Nielson [12] presented are explained as follows: 1. Learnability: Users can easily learn how to operate a system. 2. Efficiency of use: Users can operate a system efficiently. 3. Memorability: Users can remember the operation method of a system in short time, and carry out the task at once. 4. Few and non-catastrophic error: The errors of a system were avoided as far as possible, and the system provides the functions to detect and eliminate error. 5. Subjective satisfaction: Users satisfied with operating a system subjectively.

1627

It is important to confirm whether the interactive systems conform to the requirements of usability after the system design finished. However, the virtual control room was provided as a navigation environment in this study. The subjects only wandered in the virtual environment using a mouse, and search some objects. The attribute of few and non-catastrophic error was not considered, because the error sign did not appear in this operating process. Therefore, only four attributes (Learnability, Efficiency of use, Memorability, and Subjective satisfaction) were included in the usability questionnaire. 3. Result The research focused on the comparison of two VR systems (Desktop and Projection) to examine their effects on performance. The results were divided into two parts, one part is the objective (the performance of the searching task), and the other is the subjective assessment (usability questionnaire). Parameters estimated were reported along with the width of their associated 95% confidence intervals. The results revealed that performance of the searching task in Desktop VR (M=350.883, SD=96.11) and Projection VR (M=445.63, SD=133.48) as shown in Fig. 3 and Fig. 4. . The difference between the means looked large However, the ANOVA showed the difference was not significant (F (1, 18) =3.318, P=0.085). This might be because of the limited sample size employed in the experiment. The results obtained from the usability questionnaire was divided into two part; the whole impression of the interface (Attractiveness, Satisfaction, Interest, Easiness, Helpfulness, Flexibility, Fluency, and Meticulous), and the usability evaluation (Learnability, Efficiency of use, Memorability, and Subjective satisfaction). The arithmetic method about the questionnaire was ―ratio‖, obtained from the object divided by the perfect reference point, 14points. The results revealed that the average score of whole impression was 0.6585, and the average score of the usability evaluation was 0.7226. The subjects satisfied with the design of the training interface roughly, and considered that the training system is helpful to the learning. The result showed average scores of impression in projection VR (M=0.691) looked higher than that of desktop VR (M=0.626). The scores are shown in Fig. 5, and its trend in Fig. 6. However, the difference was not significant (F (1, 18) = 2.149, P = 0.160).

1628

In addition, the results obtained from the average scores, out of the eight attributes of the whole impression, only the Easiness indicated significant differences between Desktop and Projection (F (1, 18) = 4.753, P = 0.043) systems. The Easiness score in projection VR was higher compared to desktop VR as shown in Fig. 7. Besides, the results about all subjects‘ evaluation on the training interface show that the highest score is Fluency (M=0.735), and the lowest score is Easiness (M=0.532). The other scores were 0.7285 (Interest), 0.6785 (Satisfaction), 0.6685 (Meticulous), 0.6595 (Attractiveness), 0.641 (Helpfulness), and 0.6265 (Flexibility). The results from the usability evaluation revealed that there were not significant differences between Desktop and Projector based VR systems (F (1, 18) = 0.198, P = 0.662). The average scores of Project VR (M=0.7355) was significantly higher than that of Desktop VR (M=0.7097). In addition, the results from usability evaluation show the Learnability score of the Desktop VR was higher as shown in Fig. 8. Besides, the results about all subjects‘ evaluation on the training interface show that the highest score is Subjective satisfaction (M=0.778), and the lowest score is Efficiency of use (M=0.694). The other scores were 0.712 (Memorability), and 0.7065 (Learnability).

1629

4. Conclusion In this study, the experiment was conducted to test the effect of virtual reality presentations namely desktop based and projector based VR. Participants were asked to make navigation and searching tasks where performance (completion time) and questionnaire (usability evaluation) were used to compare the two systems. Although the performance results show that there were not significant differences between desktop VR and projection VR, the performance of projection VR seemed better than that of desktop VR. The insignificant result could be small sample size used in this experiment. In addition, the result of the whole impression shows that the two systems are different with regard to easiness attribute where subjects gave higher overall easiness score to projector based VR, compared to desktop based VR. The result is consistent among the foregoing research that investigated the difference between the large displays and small display. Regardless of the whole impression or the usability evaluation, the

1630

results of the usability questionnaire show that there were not significant differences between desktop VR and projection VR. Therefore, there isn‘t effect that the training interface was presented by desktop VR and projection VR on the interface evaluation. The result about the evaluation of the MCR training interface indicated that the lowest score is ―Efficiency of Use‖. The questions of ―Efficiency of Use‖ were used to evaluate the usability of the training interface, such as ―I can find the button or the image quickly‖. Therefore, a large number of the information on the WDP of the MCR result in the performance of the subjects was low, and they might consider that the interface is inefficient. In addition, the result about the evaluation of the MCR training interface indicated that all subject felt fluent in operation and interested when they interacted with the training interface, and they satisfied with the design of the training interface. Therefore, the virtual training system can help the users in learning certainly, and bring more convenience to the effectiveness of both desktop VR and projector based VR. Acknowledgment The financial support of the presentation for this paper is from projects from Ministry of Science and Technology (MOST 104-2221-E-011-063-MY3, MOST 103-2221-E-011-100 -MY3). References [1] Mamun MMA (2013) Recent Fukushima nuclear detonation, Chernobyl nuclear fallout, three mile island nuclear accident and atomic bomb explosion – rethinking the effects of nuclear radiations over human health. International Journal of Radiation Research 11: 63-80. [2] Huanga HW, Shiha C, Yihc S, Chenb MH (2008) System-level hazard analysis using the sequence-tree method. Annals of Nuclear Energy 35(3): 353-362. [3] Granlund R, Berglund E, Eriksson H (2000) Designing web-based simulation for learning. Future Generation Computer Systems 17 171-185. [4] Lehmann KS, Ritz JP, Maass H, Cakmak HK, Kuehnapfel UG, et al. (2005) A prospective randomized study to test the transfer of basic psychomotor skills from virtual reality to physical reality in comparable training setting. Annals of surgery 241(3): 442-449. [5] Lin F, Ye L, Duffy VG, Su CJ (2002) Developing virtual environments for industrial training. Information Sciences 40: 153-170. [6] CORCUERA PA (2003) A Full Scope Nuclear Power Plant Training Simulator: Design and Implementation Experiences. SYSTEMICS, CYBERNETICS AND INFORMATICS 1(3): 12-17. [7] Yim HB, Seong PH (2010) Heuristic guidelines and experimental evaluation of effective augmented-reality based instructions for maintenance in nuclear power plants. Nuclear Engineering and Design 240: 4096-4102. [8] Mól ACA, Aghina MAC, Jorge CAF, Lapa CMF, Couto PM (2009) Nuclear plant's virtual

1631

simulation for on-line radioactive environment monitoring and dose assessment for personnel. Annals of Nuclear Energy 36: 1747-1752. [9] Tyndiuk F, Lespinet-Najib V, Thomas G, Schlick C (2004) Impact of large displays on virtual reality task performance. Proceedings of the 3rd international conference on Computer graphics, virtual reality, visualisation and interaction in Africa. Stellenbosch, South Africa: ACM. pp. 61-65. [10] Shneiderman B (1997) Designing the User Interface: Strategies for Effective Human-Computer Interaction. Boston, MA, USA: Addison-Wesley Longman Publishing Co., Inc. [11] Hu PJ-H, Ma P-C, Chau PYK (1999) Evaluation of user interface designs for information retrieval systems: a computer-based experiment. Decision Support Systems 27: 125-143. [12] Nielson J (1993) Usability engineering. Boston: Morgan Kaufmann.

1632