A comparison of organic matters responsible for ... - Wiley Online Library

2 downloads 0 Views 726KB Size Report
Jun 5, 2012 - which was also applied for backwashing. A pressure sensor was set between the membrane module and the suction pump in order to monitor ...
ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING Asia-Pac. J. Chem. Eng. 2013; 8: 339–345 Published online 5 June 2012 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/apj.1665

Research article

A comparison of organic matters responsible for immersed ultrafiltration membranes fouling in drinking water treatment Lu Qi,1* Hong-chen Wang,1 Xiang Zheng,1 Guang-ming Zhang,1 Guang-hui Yu2 and Gui-bai Li3 1

School of Environment and Natural Resources, Renmin University of China, Beijing 100872, China Jiangsu Key Lab for Organic Solid Waste Utilization, College of Resources and Environmental Sciences, Nanjing Agricultural University, Nanjing 210095, China 3 State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, China 2

Received 28 December 2011; Revised 18 April 2012; Accepted 26 April 2012

ABSTRACT: Membrane fouling caused by natural organic matter is an important problem in drinking water treatment. In this study, bench-scale experiments using three different types of ultrafiltration (UF) membranes were carried out with the surface water of the Songhua River in China in order to investigate the effect of natural organic matter on membrane fouling. Organic matter that caused reversible and irreversible fouling in the filtration operation was desorbed from the fouled membranes and subjected to chemical fractionation and Fourier transform infrared analyses. These analyses revealed that hydrophilic organic matter accounted for the majority of both reversible and irreversible fouling, regardless of the type of membrane. Results also demonstrated that the type of organic matter responsible for fouling differed significantly and depended on the type of membrane. The main types of organic matter that caused reversible fouling were hydrophilic organic matter and hydrophobic acids, for the polyvinyl chloride (PVC) and polysulfone (PS) membranes and hydrophilic matter and weakly hydrophobic acids, for the polyvinylidene fluoride (PVDF) membrane. Hydrophobic acids were largely responsible for the irreversible fouling of PVDF membranes. The PVC membrane was more vulnerable to fouling because of the hydrophilic fraction, and the PS membrane was most easily fouled by the hydrophobic neutral fraction. © 2012 Curtin University of Technology and John Wiley & Sons, Ltd. KEYWORDS: natural organic matter (NOM); hydrophilic matter; membrane materials; immersed UF; fouling

INTRODUCTION In recent years, many microfiltration/ultrafiltration (MF/UF) membranes, which can be operated at relatively low pressure, have been considered around the world as alternatives to the conventional clarification and filtration processes.[1] Nevertheless, membrane fouling remains an important issue, as it is a major impediment to the progress of this technology. Membrane fouling usually describes the loss of membrane hydraulic permeability due to the accumulation of aquatic materials on the membrane surface during the filtration process, which results in reduction of the productivity from the membrane process and ultimately increases the cost of operation.[2] Studies on the fouling of membranes used for water treatment have shown that natural organic matter (NOM) is a major cause of fouling.[3–7] However, because of the complex and unstable nature of NOM in natural waters, the mechanisms describing how *Correspondence to: Lu Qi, School of Environment and Natural Resources, Renmin University of China, Beijing 100872, China. E-mail: [email protected] © 2012 Curtin University of Technology and John Wiley & Sons, Ltd. Curtin University is a trademark of Curtin University of Technology

NOM contributes to fouling are still under debate. Many bench-scale tests of membrane fouling have been conducted using natural waters or synthetic model waters containing isolated and/or fractionated NOM. The results from previous studies imply that when fractionated by absorbent resins, the fractions of NOM or soil-extracted commercial humic acids exhibit differences in their fouling tendencies.[8–16] It is generally suggested that the hydrophilic fraction has a higher fouling tendency, although observations to the contrary have also been documented in the literature.[17] One study using fractionated NOM samples and hollowfiber membranes demonstrated that the ‘hydrophilic neutral’ fraction of NOM had the highest fouling potential.[18] The low-pressure membranes currently used in largescale water treatment facilities are predominantly hollow-fiber membranes and are often made of polyvinylidene fluoride (PVDF) or polyether sulfone because of their superb resistance to chlorine and acid. Many manufacturers have introduced structural modifications to these membranes with the intent of reducing membrane fouling; however, the concentration and properties of NOM present in raw water can influence the

340

Asia-Pacific Journal of Chemical Engineering

L. QI ET AL.

1

3 2

10

8

6 5 4

7

9

11

Figure 1. Schematic diagram of the experimental

setup. (1) Feed water, (2) air pump, (3) air flow meter, (4) air diffuser, (5) UF membrane module, (6) UF tank, (7) overflow, (8) pressure sensor, (9) peristaltic pump, (10) computer and software and (11) effluent.

particular interactions of NOM with a membrane. Studies have shown that humic acid adsorption was greater on hydrophobic UF membranes than on hydrophilic UF membranes.[15] To date, little information is available in the literature concerning the types of NOM that govern the fouling of low-pressure and high-flux membranes composed of different materials. Therefore, the main objective of this paper is to investigate systematically the effect of NOM in natural waters on the fouling of immersed UF membranes with different materials. The Songhua River was investigated, and this river is considered to be representative of surface waters in China.

trans-membrane pressure (TMP) automatically. The bottom of the membrane tank was aerated to reduce fouling. UF filtration was operated at a constant flux of 30 L/m2 h, and changes in TMP were monitored to determine the extent of fouling. The flux of backwash was 50 L/m2 h with the time 5 min, and the intensity of aeration was 35 m3/m2 h. Backwash and aeration were simultaneously applied when a filtration cycles finished, and one filtration cycle is 24 h. UF membrane unit Three UF hollow-fiber membranes of different materials were used in this study, the characteristics of which are shown in Table 1. All of the UF membranes tested had the same molecular weight cutoff of 100 kDa but were made from different polymers, including polyvinyl chloride (PVC; Litree, Suzhou, China), PVDF (Litree, Suzhou, China) and polysulfone (PS; Tri-High, Beijing, China). By using these three different membranes, bench-scale membrane filtration tests were carried out in parallel. Raw water supply The raw water was taken from the Songhua River in northeast of China, which is considered to be typical of the surface waters observed in China. The main water-quality characteristics of the raw water used in the study are summarized in Table 2. Hydrophilic and hydrophobic fractions based on XAD resin separation

MATERIALS AND METHODS Experimental setup Three identical bench-scale immersed UF setups were employed in this study. A schematic illustration of the experimental setup is shown in Fig. 1. The UF membrane modules were made of three different materials with a nominal pore size of 0.01 mm and a total membrane area of 0.01 m2. The effluent was drawn directly from the membrane module using a peristaltic pump, which was also applied for backwashing. A pressure sensor was set between the membrane module and the suction pump in order to monitor and record the

By following the procedures of Thurman and Malcolm,[19] XAD resin was used to determine the hydrophilic and hydrophobic fractions of the organic matter in the UF influent and foulants. In brief, the XAD-8 and XAD-4 resins (Supelco) with sizes ranging from 20 to 50 meshes were cleaned with several solvents and used to fill a glass chromatography column. The resin column was thoroughly rinsed with 0.1 M NaOH, 0.1 M HCl and DI water. A water sample pre-filtered by a 0.45-mm filter was pumped through the XAD-8 column, and the column was washed with H3PO4 (0.1 mol/L). The organic substances reclaimed by

Table 1. Physical characteristics of the UF membranes.

Material Type Filtration mode Weight cutoff (kDa) Contact angle ( ) Inner diameter (mm) Outside diameter (mm)

PVC

PVDF

PS

Hollow fiber External pressure 100 67.0 0.85 1.45

Hollow fiber External pressure 100 56.5 0.85 1.45

Hollow fiber External pressure 80–100 75.5 0.50 1.00

PVC, polyvinyl chloride; PVDF, polyvinylidene fluoride; PS, polysulfone. © 2012 Curtin University of Technology and John Wiley & Sons, Ltd.

Asia-Pac. J. Chem. Eng. 2013; 8: 339–345 DOI: 10.1002/apj

Asia-Pacific Journal of Chemical Engineering

UF MEMBRANES FOULING IN DRINKING WATER TREATMENT

Parameter Turbidity (NTU) pH Dissolved organic carbon (DOC) (mg/L) Chemical oxygen demand (CODMn)(mg/L) UV254 (cm1) Total hardness (CaCO3) (mg/L) Total alkalinity (mg/L)

Results 19.5–21.2 7.62–7.72 7.72–7.83 5.03–5.27 0.088–0.096 75.5–81.0 62.3–72.0

H3PO4 were the hydrophobic bases (HoB), and the organic substances that remained within the column were hydrophobic neutral fraction (HoN). The water sample that passed through the column was acidified to pH = 2.0 with HCl and was pumped through the XAD-8 column for further separation. The organics adsorbed by the XAD-8 resin were the hydrophobic acids (HoA). The remaining water sample was pumped through the XAD-4 column, and the organic substances that passed the column without adsorption (and extraction) were determined to be the hydrophilic matter (HiM). Organics adsorbed by the XAD-4 resin were the weakly hydrophobic acids (WHoA). Concentrations of total organic carbon (TOC) were measured for all hydrophilic and hydrophobic fractions. Fourier transform infrared analysis A clean membrane, a fouled membrane without cleaning and a fouled membrane with backwash and sponge scrubbing were analyzed by attenuated total reflectance-Fourier transform infrared analyses (ATR-FTIR) spectroscopy to identify the organic matters causing reversible and irreversible fouling. As the tests were terminated, the fouled membrane module was flushed with pure water. Approximately 200-mL washed liquid was collected and placed in a dryer at 105  C for 24 h to obtain foulants. An FTIR spectrometer (EQUINOX55, Bruker, Germany) was used to characterize the major functional groups of organic matters in the membrane foulants. KBr pellets containing 0.50% (dry powder) of the sample were prepared and examined in the FTIR spectrophotometer. The spectrum was calculated as an average of 256 scans over the wave number ranging from 4000 to 400 cm1 at a resolution of 4 cm1.

(m1), Pt is the pressure in the membrane filtration of raw water (KPa) and Rt is the total resistance (m1). Membrane fouling can be assessed by the changes of SFR (SFRt/SFR0). SFR0 ¼

J 1 ¼ ΔP0 mRm

(1)

SFRt ¼

J 1 ¼ ΔPt mRt

(2)

Additional analytical methods Water-quality analyses were conducted utilizing standard methods.[20] Chemical oxygen demand (CODMn) was analyzed by potassium permanganate oxidation methods. UV254 was determined with a spectrometer (UV754, CANY, China). Dissolved organic carbon (DOC; prefiltration through 0.45-mm membrane) was measured with a TOC analyzer (TOC-VCPH, Shimadzu, Japan). Turbidity was monitored by a turbidimeter (TURBO550, WTW, Germany).

RESULTS AND DISCUSSION Evolution of SFR through the three UF membranes The evolution of the SFR through the different UF membranes is shown in Fig. 2. The amount of DOC passing through the membrane module (mgDOC/m2 membrane area) was also used to investigate the extent of fouling on the different membranes, as exerted by NOM in raw water. Results suggest that fouling of the three UF membrane types over the course of three filtration cycles was quite different. The SFR of the PVDF membrane 1.0 PVC PVDF PS 0.9

SFR

Table 2. The main characteristics of the raw water.

0.8

Standard flux ratio analysis 0.7

Standard flux ratio (SFR) was used to compare the effects of fouling on the membranes of different materials. SFR is described in Eqns [1] and [2], and the parameters are defined as follows: J is the membrane flux (L/m2 h), P0 is the pressure in the membrane filtration of pure water (KPa), m is the viscosity coefficient (Pa s), Rm is the original resistance of the UF membrane © 2012 Curtin University of Technology and John Wiley & Sons, Ltd.

0.6 0

1000

2000

3000

4000

5000

6000

7000

DOC (mg/m2)

Figure 2. Evolution of the SFR during three filtration

periods with different UF membranes.

Asia-Pac. J. Chem. Eng. 2013; 8: 339–345 DOI: 10.1002/apj

341

Asia-Pacific Journal of Chemical Engineering

L. QI ET AL.

decreased the most, and the SFR of the PS membrane flux decreased the least. This finding indicated that raw water fouled the PVDF membrane most seriously, whereas the PS membrane was better able to resist fouling than the PVDF and PVC membranes. As the molecular weight cutoff of the three different UF membranes is almost identical, the materials of the membrane have a substantial effect on the membrane fouling. Additionally, backwash recovered the filtration performance of the UF membranes to a certain extent, but the degree of recovery varies. The first backwash improved the SFR of the PVC, PVDF and PS membranes from 0.764, 0.73 and 0.78 to 0.894, 0.867 and 0.948, respectively; the second backwash improved them from 0.75, 0.667 and 0.771 to 0.875, 0.839 and 0.92, respectively. The recovery capacity of the PS and PVDF membranes from fouling was better, considering the degree of their recovery. Chemical fractionation of the organic matter causing membrane fouling Membrane fouling during filtration was usually caused by particulate clogging and organic adsorption. The organic matter removed by backwash was fractionated to analyze compounds causing reversible fouling, and the compounds absorbed by NaOH were fractionated to analyze irreversible fouling. The chemical fractionation of NOM in the raw water and in the UF effluent is shown in Fig. 3. The HoB was least abundant in the raw water, whereas HoA, WHoA, HiM and HoN accounted for 27.3%, 26.8%, 20.2% and 17.1% of the organic matter, respectively. The three different membranes were differentially able to remove hydrophilic and hydrophobic components, but overall, HoA and HiM components were most retained on the filters. The removal of HoA by the PAC, PVDF and PS UF membranes was 31.5%, 40.5% and 21.4%, respectively, and the removal of HiM was 17.7%, 47.3% and 26.7%, respectively. These results clearly suggested that the HoA and HiM components are most

likely to cause membrane fouling. Of the three membranes types, HoA and HiM were retained the most by the PVDF membrane, which could be interpreted as the PVDF membranes having the most serious decrease of SF during UF filtration. Fractions of dissolved organic matter (DOM) that caused reversible membrane fouling are shown in Fig. 4, where it can be seen that HiM was a major component of organic matter capable of inducing reversible fouling. HiM accounted for 54.1%, 42.8% and 46.6% of the total organic matter causing reversible fouling for PVC, PVDF and PS membranes, respectively. Another primary fouling substance was the HoA, which accounted for 20.7%, 16.7% and 41.8% of the total organic matter and caused reversible fouling in the PVC, PVDF and PS membranes, respectively. In addition, WHoA accounted for 36.8% of the fouling of the PVDF membrane. These results indicated that the main types of organic matter that could cause reversible fouling were HiM and HoA for PVC and PS membranes and HiM and WHoA for the PVDF membrane. The specific fractions of DOM causing irreversible membrane fouling are shown in Fig. 5. HiM was also the DOM component causing irreversible fouling, which is same as the results in other study;[21] it accounted for 62.6%, 46.7% and 45.1% of the total types of organic matter, causing irreversible fouling for the PVC, PVDF and PS membranes, respectively. HoN accounted for 13.9%, 12.2% and 23.6% and WHoA accounted for 17.9%, 16.7% and 14.9% of the matter fouling the PVC, PVDF and PS membranes, respectively. These results clearly suggested that HiM, WHoA and HoN were the main constituents of the organic matter that caused irreversible fouling. Additionally, HoA had a great effect on the fouling of the PVDF membrane. Among the different membranes, the PVC membrane was more vulnerable to fouling by HiM, and the PS membrane was more vulnerable to fouling by HoN. 70 Raw water

Raw water

2.5

PVC

60

PVC

PVDF

PVDF

2.0

PS

PS

50

Fraction(%)

DOC(mg/L)

342

1.5

1.0

40 30 20

0.5

0.0

10

HoB

HoN

HoA

WHoA

HiM

0

HoB

HoN

HoA

WHoA

HiM

Figure 3. Fractionation of DOM in raw water and UF

Figure 4. Fractionation of DOM causing reversible

effluents.

membrane fouling.

© 2012 Curtin University of Technology and John Wiley & Sons, Ltd.

Asia-Pac. J. Chem. Eng. 2013; 8: 339–345 DOI: 10.1002/apj

Asia-Pacific Journal of Chemical Engineering 70

UF MEMBRANES FOULING IN DRINKING WATER TREATMENT

Raw water PVC

60

PVDF

Fraction(%)

PS

50 40 30 20 10 0

HoB

HoN

HoA

WHoA

HiM

Figure 5. Fractionation of DOM causing irreversible

membrane fouling.

ATR-FTIR spectroscopic analysis of membrane surfaces Infrared spectroscopy has been widely used for gross characterization of humic substances and can provide valuable information on the functional properties of NOM molecules. Figures 6 show the FTIR spectroscopy of clean membrane surfaces for each of the different membrane types. The PVC membrane has its original characteristic functional groups, such as C–H (2915 cm1), –CH2 (1429 cm1), C–C (1105 cm1) and C–Cl (965 and 685 cm1). The PVC membrane used in these experiments was modified to have several hydrophilic function groups, such as carboxyl C═O (1738 cm1) and O–H (3374 cm–1). The PVDF membrane had its original characteristic functional groups, such as C–F (1170 cm1), CF2 (1070 cm1) and ═C–H (875 cm1) and several hydrophilic functional groups, including C═O (1666 cm1), C–O (1271 cm1) and O–H (1400 cm1), because of modification. The PS membrane used in these tests was also modified and had hydrophilic functional groups presented on

the membrane surface, including carboxylic C═O (1771 cm1), C–O (1237 cm1) and O–H (1408 and 1013 cm1). The FTIR spectroscopy of fouled membrane surfaces is shown in Fig. 7. After filtration, the surface of membrane was covered with caked layers. As seen in the figures, the absorption peaks of functional groups on the surface membrane nearly disappeared and were replaced by the peaks of the organic matter in the cake layers. Groups characteristics of hydrophilic organic matter existed on the fouled membrane surface, including O–H (3285 cm1) and C–O (1000 cm1). There were also some groups characteristic of HoA, including aliphatic C–H (2928 cm1) and carboxylate COO (1634 or 1403 cm1) from all three kinds of membranes, suggesting that HiM and HoA might induce reversible fouling. In addition, characteristic groups of polysaccharides C–O (1100 cm1) and of the amide proteins N–H (1540 cm1) existed on the PVDF membrane surface, indicating that weakly hydrophobic substances might cause reversible fouling of the PVDF membrane. The FTIR spectroscopy of the fouled membrane surfaces after cleaning is shown in Fig. 8. Reversible fouling was considered to be complete removal by backwash and sponge scrubbing. Organic matter causing irreversible fouling could be analyzed by comparing the functional groups of clean membranes and fouled membranes after cleaning. The results showed that characteristic groups of carbohydrate C–O (1046 or 1050 cm1) existed in the spectra of all three kinds of filters, suggesting that HiM may be the main organic substance causing irreversible fouling. Similar studies have also found that the carbohydrate-like organic matters were probably responsible for physically irreversible fouling.[21] Amides N–H (1540 cm1) existed in the spectra of PVDF membrane surface, which indicated that weakly hydrophobic substances induced irreversible fouling of the PVDF membrane. The spectra of

100 100

80

80

2928

1403

%T

%T

3285

60

40

PS

60

PS

40

PVDF

PVDF

PVC

20 4000

3500

1540 1634

1100

PVC

3000

2500

2000

1500

1000

500

cm-1

Figure 6. FTIR of functional groups on clean UF membranes surface. © 2012 Curtin University of Technology and John Wiley & Sons, Ltd.

20 4000

1000

3500

3000

2500

2000

1500

1000

500

cm-1

Figure 7. FTIR of the fouled membranes surface without cleaning. Asia-Pac. J. Chem. Eng. 2013; 8: 339–345 DOI: 10.1002/apj

343

Asia-Pacific Journal of Chemical Engineering

L. QI ET AL.

removed by conventional drinking water treatment processes, their presence must be accounted for in the application of UF filtration processes. In addition, suitable materials need to be chosen for different waterquality applications to reflect how the kind of organic matter present may affect the specific UF membranes.

100 90 80 70

%T

344

3382

1710

60 50 40 30

CONCLUSION

PS

1536

PVDF PVC

1050

20 4000

3500

1046

3000

2500

2000

1500

1000

500

cm-1

Figure 8. FTIR of the fouled membranes surface after

cleaning.

the PS membrane surface had a dominant peak near 1710 cm1, which is an indication of aliphatic ketones, showing that the PS membrane might be irreversibly fouled by HoN. Taken together, the results of these tests show that the organic matter causing reversible fouling consisted mainly of HiM and HoA, and the organic matter that could induce irreversible fouling consisted mainly of HiM. All three membranes were made from polymers containing electronegative atoms; therefore, carbohydrate-like substances were relatively easily retained at the membrane surface because of their electrostatically neutral nature and thus created the irreversible fouling. The greater the electronegativity of the membranes, the more easily hydrogen bonds can be formed.[22] The PVC and PVDF membranes used in this test contained the highly electronegative substances of Cl and F, which could easily form hydrogen bonds with carbohydrate-like substances. This finding partially explains why hydrophilic NOM was shown to be a major foulant in previous studies on fouling—hydrophilic NOM actually has a greater binding power to the membrane because of hydrogen bonding. All the materials in the membranes were hydrophobic, which could react with hydrophobic organic matters. The contact angle of the PS membrane was larger than other two membranes; therefore, the PS membrane was the most hydrophobic of the three materials, and its reaction with hydrophobic substances was strongest. HoN and WHoA were relatively easily accessed at the membrane surface because of their electrostatically neutral nature, meaning that the PVDF and PS membranes were more vulnerable to WHoA and HoN, respectively. The reaction of the different membrane materials with NOMs in source water requires further research in the future. HiM, HoN and WHoA could induce irreversible fouling of UF membranes composed of different materials because of their electrostatically neutral properties, but because they cannot be effectively © 2012 Curtin University of Technology and John Wiley & Sons, Ltd.

These experiments tested the effects of NOM on the fouling of immersed hollow-fiber UF membranes composed of different materials. Different kinds of organic matter affected the performance of the three types of UF membranes. Results of DOM fractions showed that HiM and HoA were major organic matter capable of inducing reversible fouling and HiM, WHoA and HoN were the main constituents of the organic matter that caused irreversible fouling. ATR-FTIR spectroscopic analysis also showed that the organic matter that could induce irreversible fouling consisted mainly of HiM because carbohydrate-like substances were relatively easily retained at the membrane surface by hydrogen bonds. These kinds of organic matter that cause fouling must be effectively removed before the UF filtration process is applied to surface waters, and the specific filter material needs to be chosen with respect to the quality of the source water.

REFERENCES [1] K.Y.J. Choi, B.A. Dempsey. Water Res., 2004; 38(19), 4271–4281. [2] T. Carroll, S. King, S.R. Gray, B.A. Bolto, N.A. Booker. Water Res., 2000; 34(11), 2861–28681. [3] K. Kimura, H. Yamamura, Y. Watanabe. Sep. Sci. Technol., 2006; 41, 1331–1344. [4] K. Kimura, Y. Hane, Y. Watanabe, G. Amy, N. Ohkuma. Water Res., 2004; 38, 3431–3441. [5] J.A. Leenheer, C.E. Rostad, L.B. Barber, L.B. Barber, R.A. Schroeder, R. Anders, M.L. Davisson. Environ. Sci. Technol., 2001; 35, 3869–3876. [6] M.B. Muller, W. Fritz, U. Lankes, F.H. Frimmel. Environ. Sci. Technol., 2004; 38, 1124 –1132. [7] S. Lee, J.W. Cho, M. Elimelech. J. Membr. Sci., 2005; 262, 27–41. [8] C.F. Lin, Y.J. Huang, O.J. Hao. Water Res., 1999; 33, 1252–1264. [9] C.F. Lin, T.Y. Lin, O.J. Hao. Water Res., 2000; 34, 1097–1106. [10] C.F. Lin, S.H. Liu, O.J. Hao. Water Res., 2001; 35, 2395–2402. [11] N. Lee, G. Amy, J.P. Croue, H. Buisson. Water Res., 2004; 38, 4511–4523. [12] J. Cho, G. Amy, J. Pellegrino. J. Membr. Sci., 2000; 164, 89–110. [13] N. Her, G. Amy, H.R. Park, M. Song. Water Res., 2004; 38, 1427–1438. [14] B.H. Gu, J. Schmitt, Z.H. Chen. Environ. Sci. Technol., 1994; 28, 38 – 46. [15] W. Yuan, A.L. Zydney. J. Membr. Sci., 1999; 157, 1–12. [16] N.H. Lee, G. Amy, J.P. Croue, H. Buisson. Water Res., 2004; 38, 4511–4523. [17] J.A. Nilson, F.A. DiGiano. J. Am. Waterworks Assoc., 1996; 88(5), 53–66. Asia-Pac. J. Chem. Eng. 2013; 8: 339–345 DOI: 10.1002/apj

Asia-Pacific Journal of Chemical Engineering

UF MEMBRANES FOULING IN DRINKING WATER TREATMENT

[18] T. Carroll, N.A. Booker, J. Meier-Haack. J. Membr. Sci., 2002; 203(1–2), 3–13. [19] E.M. Thurman, R.L. Malcolm. Environ. Sci. Technol., 1981; 15(4), 463 – 466. [20] APHA. Standard Methods for the Examination of Water and Wastewater, 20th edn, American Public Health Association/

© 2012 Curtin University of Technology and John Wiley & Sons, Ltd.

American Water Works Association/Water Environment Federation, Washington DC, USA, 1998. [21] H. Yamamura, K. Kimura, Y. Watanabe. Environ. Sci. Technol., 2007; 41, 6789–6794. [22] H. Yamamura, K. Kimura, T. Okajima, H. Tokumoto, Y. Watanabe. Environ. Sci. Technol., 2008; 42, 5310–5315.

Asia-Pac. J. Chem. Eng. 2013; 8: 339–345 DOI: 10.1002/apj

345