SnO2

Chemical Physics Letters 619 (2015) 1–6

Contents lists available at ScienceDirect

Chemical Physics Letters journal homepage: www.elsevier.com/locate/cplett

Fabrication of ␣-Fe2 O3 hexagonal disc/SnO2 nanoparticle semiconductor nanoheterostructures and its properties C.S. Biju a , D. Henry Raja b , D. Pathinettam Padiyan a,∗ a b

Department of Physics, Manonmaniam Sundaranar University, Tirunelveli 627012, India Department of Physics, Scott Christian College, Nagercoil 629003, India

a r t i c l e

i n f o

Article history: Received 23 September 2014 In final form 20 November 2014 Available online 25 November 2014

a b s t r a c t Semiconductor nanoheterostructures with different weight ratios were successfully fabricated by hydrothermal method. X-ray diffraction analysis indicates that the samples were composed of ␣-Fe2 O3 and SnO2 . HRTEM images confirmed the immobilization of SnO2 nanoparticles on the surface of ␣-Fe2 O3 hexagonal discs. Optical studies reveal that the band gaps of the samples enlarge from 1.862(2) eV to 1.9931(7) eV with increase in SnO2 weight ratio. Hysteresis behavior clearly displayed a shift of loops toward field axis for nanoheterostructures. From the ZFC curve an increase of Morin transition from 230 K to 240 K was noticed when the SnO2 weight ratio increases. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Semiconductor nanoheterostructures (SNH’s), which consists of distinct components, have demonstrated great potential applications in photocatalysis [1–4], sensors [5–8], lithium-ion batteries [9–11] and thermite membranes [12]. Growth of complex nanostructures with controllable dimensions has attracted much because of their diverse properties generated by tailoring the morphology, composition and assembling of nanostructures on to the parent material. For example, Zhou et al. [11] fabricated a six-fold symmetry nanorod branches of ␣-Fe2 O3 onto the SnO2 nanowires and observed a superior electrochemical performance. Nu et al. [3] grew SnO2 nanorods epitaxially on the surface of ␣Fe2 O3 nanospindles and noticed an enhanced photocalytic activity toward methylene blue. Xu et al. [4] fabricated SnO2 nanorods hierarchically on the inner and outer surfaces of hollow ␣-Fe2 O3 nanoprisms and studied their optical and photocatalytic properties. Sun et al. [8] assembled short nanorod branches on the basal surface of SnO2 nanosheets and demonstrated an enhanced gas sensing performance toward acetone. Wang et al. [9] grew ␣Fe2 O3 nanorods on the surface of SnO2 nanosheets and showed an enhanced lithium ion storage capacity and cycling stability. Zhu et al. [1] fabricated SnO2 /␣-Fe2 O3 heterojunction nanotubes and showed higher visible light photocatalytic activity for the degradation of Rhodamine B dye. Wu et al. [2] prepared ␣-Fe2 O3 /SnO2 core–shell heterostructures and demonstrated an enhanced

∗ Corresponding author. E-mail address: [email protected] (D.P. Padiyan). http://dx.doi.org/10.1016/j.cplett.2014.11.034 0009-2614/© 2014 Elsevier B.V. All rights reserved.

photocatalytic performance of Rhodamine B. All the above works are devoted to check the application performance of the material. To the best of our knowledge, there have been no reports on the fabrication of SnO2 nanoparticles (NP’s) on the ␣-Fe2 O3 hexagonal discs (HD’s) and their detailed investigations on the magnetic properties viz.; hysteresis behavior and Morin transition temperature. The motivation of this work is generated based on the results obtained by Sytnyk et al. [13], Ma et al. [14] and Chen et al. [15], which are given below. Sytnyk et al. [13] tuned the magnetic parameters of colloidal magnetite nanocrystal heterostructures by cation exchange. They noticed a significant enhancement in coercivity and blocking temperature when magnetite was modified with Co2+ ions. Ma et al. [14] studied the magnetic characteristics of Fe3 O4 /␣-Fe2 O3 hybrid cubes and inferred a disappearance of Morin transition temperature. Chen et al. [15] prepared Fe3 O4 /SnO2 core–shell nanorods and observed a significant enhancement of coercivity and reduction of saturation magnetization. They concluded that the effective complementarities between the dielectric and magnetic loss tangents played crucial role in electromagnetic wave absorption. Therefore integrating diverse compositions and functionalities with enhanced properties could be attributed to the synergetic strengths owing to the unique interfacial interaction of the binary components. Moreover, Wu et al. [16] synthesized well defined ␣-Fe2 O3 nanorombohedra by dissolving FeCl3 and CH3 COONa in a medium of distilled water and in this reaction CH3 COONa served as a structure directing agent. However, there have been no reports on the synthesis of ␣-Fe2 O3 hexagonal disc (HD) employing Fe(NO3 )3 so far. In this work, we synthesized ␣-Fe2 O3 HD’s using Fe(NO3 )3 as an iron source instead of FeCl3 by a facile hydrothermal strategy. Subsequently SnO2 , with

2

C.S. Biju et al. / Chemical Physics Letters 619 (2015) 1–6

different weight ratios were immobilized on the surface of HD’s by the same hydrothermal method to form SNH’s. We investigated their structural, formation mechanism, optical and detailed magnetic properties like hysteresis behavior and Morin transition temperature. 2. Experimental 2.1. Synthesis of bare ˛-Fe2 O3 HD’s ␣-Fe2 O3 HD’s were synthesized by hydrothermal method. In a typical method, 0.1 M Fe(NO3 )3 and 0.3 M CH3 COONa were dissolved in 40 mL of double distilled water under magnetic stirring. The above solution was transferred into a Teflon-lined stainless steel autoclave and heated at 160 ◦ C in an electric oven for 24 h. The suspensions were centrifuged with double distilled water several times to collect the reddish brown precipitates. Then the precipitates were dried in an electric oven at 80 ◦ C overnight to obtain ␣-Fe2 O3 HD’s and labeled as S1 . 2.2. Fabrication of SNH’s Initially SnO2 was prepared using 40 mL of 0.015 M SnCl4 and 0.045 M NaOH by hydrothermal method at 160 ◦ C for 3 h and the yield was 0.0912 g of SnO2 . SNH’s were also prepared by the simple hydrothermal strategy. Typically, 0.1824 g of ␣-Fe2 O3 HD’s were dispersed in 40 mL of double distilled water containing 0.015 M SnCl4 and 0.045 M NaOH to set 1:0.5 ratio of ␣-Fe2 O3 : SnO2 . The solutions were mixed thoroughly under magnetic stirring and hydrothermally treated in an autoclave at 160 ◦ C for 3 h. The precipitates were centrifuged with double distilled water several times and dried in an electric oven at 80 ◦ C overnight to obtain SNH’s and labeled as S2 . Similarly the weight ratio of ␣-Fe2 O3 to SnO2 was varied in the ratio 1:1 and 1:2, by mixing 0.0912 and 0.0456 g of ␣-Fe2 O3 in the solution, labeled as S3 and S4 respectively. 2.3. Characterizations X-ray diffraction (XRD) patterns were recorded to confirm the phase purity of the samples by an Xpert’o Panalytical X-ray diffractometer. The morphologies, formation of SNH’s and elemental composition of the samples were studied with the assistance of a TECNAI high resolution transmission electron microscope (HRTEM), equipped with the energy dispersive X-ray spectroscopy (EDS) system. Fourier transform infrared (FTIR) spectroscopy was performed using a Perkin Elmer Spectrum 2 spectrophotometer by KBR method, to confirm the sample formation and to know the surface adsorbed residues. Reflectance spectra were acquired by means of a Shimadzu UV-2700 diffuse reflectance UV–Vis spectrophotometer (DRUV). Magnetic measurements were carried out by a Lakeshore vibrating sample magnetometer (VSM). 3. Results and discussion 3.1. XRD The XRD patterns of the hydrothermally synthesized samples (S1 –S4 ) are shown in Figure 1. All diffraction peaks are in good agreement with those of standard patterns for hexagonal ␣-Fe2 O3 (JCPDS 87-1164). The diffraction peaks of the ␣-Fe2 O3 are sharp and intense, revealing the highly crystalline character of the ␣Fe2 O3 , while the diffraction peaks of the SnO2 are broad and weak, indicating a small crystallite size or semi crystalline nature of the samples [17]. No other characteristic peaks are detected, indicating that the compositions are ␣-Fe2 O3 and SnO2 . Thus the formation

Figure 1. XRD patterns of S1 , S2 , S3 and S4.

of SNH’s is unambiguously confirmed. For the sample S1 the relative intensity of the peak (1 1 0) is 83.27%. With increasing SnO2 wt. ratio, the relative intensity of the peak (1 1 0) decreases to 65.17%. These intensity variations are also reflected in the lattice mismatch values and optical properties. Lattice mismatch (LM) is calculated [18] from the difference between the lattice parameters of ␣-Fe2 O3 and SnO2 along a-axis divided by the lattice parameter of SnO2 . Lattice mismatch and strain type (Tensile strain (LM −ive)), compressive strain (LM +ive)) of the samples are shown in Table 1. It is clear that, with increasing SnO2 wt. ratio the compressive strain increases slightly from S2 to S3 , while further increasing the wt. ratio, decreases the compressive strain. 3.2. Possible formation mechanism In order to track the formation of SNH’s, HRTEM images are recorded for different magnification. Figure 2(a) and (b) are HRTEM images of the samples with and without SnO2 , which shows the morphological evolution from ␣-Fe2 O3 hexagonal discs (HD’s) to SNH’s. In bare ␣-Fe2 O3 (S1 ) only HD’s with smooth surfaces are observed (Figure 2(a)). In the sample S3 many SnO2 NP’s of size 2–5 nm (Figure 2(b)) are immobilized on the surface of ␣-Fe2 O3 HD’s as a result of dehydration of Sn(OH)4 . It is observed that some NP’s are heteronucleated on the surface of the HD’s, while others homonucleated in the solution. Magnified HRTEM image in the inset of Figure 2(b) displays the detailed structures of the heteronucleated and homonucleated NP’s. Nu et al. [19] observed a similar simultaneous hetero and homonucleation, while fabricating SnO2 nanorods on ␣-Fe2 O3 nanohexohedrons. Alternate bright and dark regions seen in the HRTEM images of S3 gives the interplanar spacings of 0.27 nm and 0.33 nm, which are indexed to the (1 0 4) and (1 1 0) planes of HD’s and NP’s respectively. These results reveal the formation of SNH’s. In Figure 2(c), the EDS spectrum taken from the HD’s (S1 ) shows the presence of the elements Fe and O, indicating that the HD’s are made of pure ␣-Fe2 O3 , while the sample S3 (Figure 2(d)) shows the signal pertaining to Sn, Fe and O. On the basis of the above results it can be unambiguously deduced that the hydrothermally synthesized SNH’s is an assembly constructed


3

Table 1 XRD, optical and magnetic parameters of S1 , S2 , S3 and S4 . Sample

S1 S2 S3 S4

XRD

DRUV

VSM

Rel. int. (%) of (1 1 0) peak

Lattice mismatch (%)

Strain type

Band gap (eV)

Coercivity (Hc ) (Oe)

Remanance (Mr ) (emu/g)

Saturation magnetization (Ms ) (emu/g)

Morin transition temperature (TM ) (K)

83.27 71.21 76.72 65.17

– 6.29 6.32 6.24

– Compressive Compressive Compressive

1.862 (2) 1.9640 (6) 1.967 (1) 1.9931 (7)

76.9 – 112.4 538.9

0.0898 – 0.0527 0.0068

1.5632 – 0.4971 0.0239

230 – 236 240

with HD’s and NP’s. These results are consistent with XRD results. The possible reaction mechanism for the formation of bare ␣-Fe2 O3 HD’s (S1 ) in the CH3 COONa system may be illustrated as follows [20]. CH3 COO− + H2 O ↔ CH3 COOH + OH−

(1)

Hydrothermal (160 ◦ C)

Fe3+ + 3OH− → Fe(OH)3 /FeOOH 3+

␣-Fe2 O3 + 6CH3 COOH ↔ 2Fe

−→

−

␣-Fe2 O3

+ 6CH3 COO + 3H2 O

(2)

Scheme 1. Formation of SNH’s during hydrothermal process.

(3)

As shown above, Fe3+ reacted with OH− produced by hydrolysis and formed Fe(OH)3 /FeOOH suspension. During hydrothermal process, Fe(OH)3 /FeOOH powders are easily dehydrated to form ␣Fe2 O3 HD’s. CH3 COONa is used as a structure directing agent [16] and leads to the formation of HD’s due to the different adsorption ability of CH3 COO− on different planes of ␣-Fe2 O3 . The CH3 COO− stemming from the added CH3 COONa can charge the surfaces of as-obtained nanoparticles negatively. As a result, well dispersed ␣-Fe2 O3 HD’s are obtained with the addition of CH3 COONa. Also the hydrothermal route and formation mechanism for fabricating SNH’s is shown in Scheme 1. Growth of NP’s occur by a solutionsolid process due to the dissolution of precipitate and homogeneous

nucleation [21]. The chemical mechanism for the formation of SnO2 NP’s during hydrothermal process is given below [22]. SnCl4 + 4NaOH → Sn(OH)4 ↓ +4NaCl

Sn(OH)4 −→SnO2 + 2H2 O

(4) (5)

In seed-mediated growth method, the shell formation involves two steps [23]. First, SnCl4 reacts with NaOH and forms Sn(OH)4 , according to reaction (4). When the HD seeds are dispersed into the solution containing Sn(OH)4 , it is adsorbed on the surface of ␣-Fe2 O3 via electrostatic effects [2]. When the surface energy decreases, SnO2 NP’s are randomly heteronucleated on the [12]. Second, surface of HD’s through dehydration of Sn(OH)2− 6

Figure 2. (a, b) HRTEM images of S1 and S3 ; the insets are magnified views and SAED patterns; (c, d) EDS spectra of S1 and S3 .

4


Figure 4. Absorbance spectra of the samples S1 , S2 , S3 and S4 and the inset shows the band gap energies.

Figure 3. FTIR spectra of bare ␣-Fe2 O3 HD’s and SNH’s.

during hydrothermal process, with slow heating rate, the adsorbed Sn(OH)4 decompose and form SnO2 NP shell layer (reaction (5)) on the surface of ␣-Fe2 O3 HD seeds, and finally forming core–shell SNH’s. 3.3. FTIR FTIR spectra of the samples recorded in the region 400–1200 cm−1 are shown in Figure 3. In S1 , two broad peaks centered at 450 and 576 cm−1 are assigned to stretching vibrations of Fe3+ –O2− bond in the FeO6 octahedron and FeO4 tetrahedron structures [24]. With the addition of SnO2 (S2 ), the width of the peak at 576 cm−1 decreases slightly. In S3 , the broad peaks of ␣-Fe2 O3 collapse slightly and some minor peaks pertaining to SnO2 appears in the region 400–500 cm−1 , as shown in the inset. This is a signature for the presence of SnO2 and the formation of SNH’s. In S4 , the peaks of ␣-Fe2 O3 completely disappeared and three distinct peaks appeared at 500, 675 and 832 cm−1 . In addition to that, number of minor peaks appeared in the region 400–550 cm−1 , as shown in the inset. The distinct peaks along with the minor peaks are assigned to Sn–O stretching vibrations of SnO2 [25]. Thus the FTIR results clearly verify the formation of SNH’s and they are consistent with HRTEM and XRD results. 3.4. DRUV Band gap of the samples are studied by UV–Vis diffuse reflectance spectroscopy (DRUV). Kubelka–Munk function [26], F(R) = (1 − R∞ )2 /2R∞ is used to determine the band gap by analyzing the DRUV data. The reflectance data are converted to equivalent absorption coefficient ˛, proportional to the Kubelka–Munk function, F(R). The absorption spectra of the samples are shown in Figure 4. It can be seen that, after coupling SnO2 shells on the surface of the ␣-Fe2 O3 cores, and when the wt. ratio increases, the absorption band edge shifts toward the blue wavelength region of the absorption spectrum. This enlarges the band gap energies

from 1.862(2) to 1.9931(7) eV (inset of Figure 4). These results are consistent with the lattice mismatch values obtained from the XRD results. This can be explained by the interactive strain between the core and shell that alters the optical properties [27]. Strain effect is caused by the variation in lattice parameter val˚ c = 3.1864 A), ˚ smaller than those of ues of SnO2 (a, b = 4.7373 A, ˚ c = 13.7458 A). ˚ As a result lattice mismatch ␣-Fe2 O3 (a, b = 5.0334 A, occurs, where the core is subjected to compressive strain, while shell experiences tensile strain. From the XRD results, shown in Table 1, compressive strain increases with increase of SnO2 wt. ratio, thereby enlarging the band gap slightly (weak blue shift) from 1.9640(6) (S2 ) to 1.967(1) (S3 ). However, when the compressive strain decreases slightly (S4 ), the band gap gets blue shifted to 1.9931(7). According to Bao et al. [28], the reason for blue shift can be explained as follows. As the SNH’s are excited with UV-light, electrons and holes are spatially separated across the interface, bends the band and modifies the electron and hole energy levels via stark effect. This leads to the weak blue shift of the samples from S2 to S3 . At high SnO2 wt. ratio (S4 ), photoexcited electrons and holes occupy the higher energy sub bands and a band filling effect takes place. This is responsible for the blue shift from S3 to S4 . Another possible reason for the blue shift is the antiferromagnetic ordering in the samples, which can be interpreted from ZFC curves of the samples. With increase of SnO2 wt. ratio, the Morin transition temperature increases, which increase the magnetic ordering, thereby increase the band gap energies. We therefore concluded that the enlargement of band gap energies is due the combined effects of compressive strain of the core and magnetic ordering. 3.5. Magnetic properties Magnetic hysteresis loops recorded at 300 K for the samples S1 , S3 and S4 are shown in Figure 5(a). The detailed magnetic parameters obtained in this study are shown in Table 1. The hysteresis loops of the samples shift toward the field axis with increase in SnO2 wt. ratio. It is also observed that the samples saturate in less applied magnetic field with increasing SnO2 contact. Magnetic cores have magnetic interaction mediated by the electrons in their neighboring non-magnetic shells, which is crucial than their direct exchange interactions known as super exchange interaction [29]. Immobilization of NP’s at the surface leads to the breakage of super exchange bonds. This leads to the decrease of saturation magnetization in the sequence S1 = 1.5632 < S3 = 0.4976 < S4 = 0.0239 emu/g. The magnetic contacts decreases with SnO2 wt. ratio and result in decrease of the net magnetization [29]. Figure 5(b)–(d) shows the FC and ZFC magnetization curves of S1 , S3 and S4 as a function of temperature from 80 K to 300 K with an applied field of 1000 Oe.


5

Figure 5. (a) Hysteresis loops of S1 , S3 , and S4 ; (b–d) FC, ZFC curves and Morin transition temperature (inset) of samples S1 , S3 and S4 .

The ZFC and FC magnetizations are unchanged below 225 K for S1 . With the addition of NP’s a concave nature is observed for the samples S3 and S4 in the temperature region 100–225 K. The concave nature observed is clearly suggesting a strong localized nature of the magnetic ions and a low value carrier density insulating system [30]. This inference clearly verifies the formation of SNH’s. Both FC and ZFC curves increases with temperature above 225 K and reveal a maximum. This behavior with increasing temperature can be attributed to the rapid decrease of unidirectional exchange anisotropy (UEA) [31]. UEA was first discovered in Co/CoO particles, caused by the strong exchange coupling between Co core and the CoO layer [32]. At high temperatures the thermal energy of the nanoparticles overcomes the magnetic anisotropic energy barrier between different magnetic easy axes and between the spin up and spin down direction within the same easy axis respectively [13]. As a result, Morin transition temperature TM , appears in ZFC dM/dT curves at 230 K, 236 K and 240 K for the samples S1 , S3 and S4 respectively. This causes a slight increase of coercivity and a slight decrease of remanance and saturation magnetizations in the hysteresis loops. For bulk ␣-Fe2 O3 the Morin transition temperature (TM ) is reported as 263 K, since it undergoes a first order magnetic transition from weakly ferromagnetic to antiferromagnetic phase [33]. The obtained TM value for the sample S1 is lower than the bulk counterparts. The immobilization of SnO2 NP’s results in an overall increase of coercivity and Morin transition temperature due to increased magnetocrystalline anisotropy. Thus the observed variations of the magnetic properties are solely due to cationic exchange of the ions of core and shell. It is noted that cationic

exchange reaction preserves the core–shell structures, sizes and shapes [13]. These results are also consistent with DRUV results, where the absorption band edge shifts toward the blue region, which leads to an increase of band gap energies with increase of SnO2 wt. ratio. High coercivity and low saturation magnetization produces high dielectric tangent loss (tan ıe ) and magnetic tangent loss (tan ım ) [15], thereby enhancing the electromagnetic wave absorbing property. Since the low Ms and high Hc obtained in this work are consistent with the above report, we speculate that the SNH’s could be an ideal material for electromagnetic wave absorption.

4. Conclusions We have successfully synthesized the novel ␣-Fe2 O3 /SnO2 SNH’s by a facile two step hydrothermal strategy. It is found that CH3 COONa served as a structure directing agent for preparing ␣Fe2 O3 hexagonal discs. XRD, HRTEM and FTIR characterizations clearly verify the formation of SNH’s. DRUV results reveal that the band gap energies of SNH’s enlarge with increase of SnO2 wt. ratio due to strain effects. Hysteresis behavior indicates that saturation magnetization reduces due to the breakage of super exchange bonds between the core and the shell. ZFC curves of the samples show that the Morin transition temperature increases with increase of SnO2 wt. ratio. The modifications of the overall properties of the SNH’s could be attributed to the synergetic effect of ␣-Fe2 O3 core and SnO2 shell.

6


Acknowledgments We acknowledge SAIF, S.N. Bose National institute for basic sciences, Kolkata, for HRTEM and VSM measurements. We also acknowledge Mr. Samik Roy Mallik and Mr. Dipankar Roy for data recording/technical assistance. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]

C. Zhu, et al., J. Alloys Compd. 575 (2013) 333. W. Wu, S. Zhang, F. Ren, X. Xiao, J. Zhou, C. Jiang, Nanoscale 3 (2011) 4676. M. Nu, F. Huang, L. Cui, P. Huang, Y. Yu, Y. Wang, ACS Nano 4 (2010) 681. J. Xu, F. Huang, Y. Yu, A. Yang, Y. Wang, CrystEngComm 13 (2011) 4873. P. Sun, et al., Sens. Actuators B 182 (2013) 336. S. Zhang, et al., Phys. Chem. Chem. Phys. 15 (2013) 8228. Y. Li, Y. Hu, H. Jiang, X. Hou, C. Li, RSC Adv. 3 (2013) 22373. P. Sun, X. Zhou, C. Wang, K. Shimanoe, G. Lu, N. Yamazoe, J. Mater. Chem. A 2 (2014) 1302. Y. Wang, J. Xu, H. Wu, M. Xu, Z. Peng, G. Zheng, J. Mater. Chem. 22 (2012) 21923. Y. Li, Y. Hu, H. Jiang, X. Hou, C. Li, CrystEngComm 15 (2013) 6715. W. Zhou, et al., Adv. Funct. Mater. 21 (2011) 2439. Y. Yang, et al., Inorg. Chem. 52 (2013) 9449.

[13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]

M. Sytnyk, et al., Nano Lett. 13 (2013) 586. J. Ma, K. Chen, X. Zhang, J. Appl. Phys. 111 (2012) 033925. Y. Chen, et al., J. Phys. Chem. C 113 (2009) 10061. X. Wu, Y. Guo, L. Wan, C. Hu, J. Phys. Chem. C 112 (2008) 16824. L. Zheng, et al., Inorg. Chem. 48 (2009) 1819. J. Zhang, H. Tanaka, T. Kanki, J. Choi, T. Kawai, Phys. Rev. B 64 (2001) 184404. M. Nu, Y. Cheng, Y. Wang, L. Cui, F. Bao, L. Zhou, Cryst. Growth Des. 8 (2008) 1727. J. Lian, X. Duan, J. Ma, P. Peng, T. Kim, W. Zheng, ACS Nano 3 (2009) 3749. W. Wu, S. Zhang, J. Zhou, X. Xiao, F. Fen, C. Jiang, Chem. Eur. J. 17 (2011) 9708. H. Zhu, D. Yang, G. Yu, H. Zhang, K. Yao, Nanotechnology 17 (2006) 2386. W. Wu, et al., ACS Appl. Mater. Interfaces 4 (2012) 3602. G. Xia, et al., ACS Appl. Mater. Interfaces 5 (2013) 8607. X. Xu, J. Zhuang, X. Wang, J. Am. Chem. Soc. 130 (2008) 12527. P. Kubelka, F. Munk, Z. Tech. Phys. 12 (1931) 593. W. Chiu, et al., J. Phys. Chem. C 114 (2010) 8212. J. Bao, D. Bell, F. Capasso, Nano Lett. 8 (2008) 836. R.S. Devan, B.K. Chougule, J. Appl. Phys. 101 (2007) 014109. R. Adhikari, A.K. Das, D. Karmakar, T.V. Chandrasekhar Rao, J. Ghatak, Phys. Rev. B 78 (2008) 024404. D.L. Peng, K. Sumiyama, T. Hihara, S. Yamamuro, T.J. Konno, Phys. Rev. B 61 (2000) 3103. W.H. Meiklejohn, C.P. Bean, Phys. Rev. 102 (1956) 1413. N. Amin, S. Ajras, Phys. Rev. B 35 (1987) 4810.