J Clust Sci DOI 10.1007/s10876-013-0562-7 ORIGINAL PAPER
A Synergy of ZnO and ZnWO4 in Composite Nanostructures Deduced from Optical Properties and Photocatalysis Tatjana D. Savic´ • Ivana Lj. Validzˇic´ • Tatjana B. Novakovic´ • Zorica M. Vukovic´ ˇ omor Mirjana I. C
•
Received: 20 November 2012 Ó Springer Science+Business Media New York 2013
Abstract ZnO/ZnWO4 composite rod-like nanoparticles were synthesized by lowtemperature soft solution method at 95 °C with different reaction times (1–120 h), in the presence of non-ionic copolymer surfactant Pluronic F68. Obtained nanoparticles had diameters in the range around 10 nm and length of 30 nm. Optical properties such as reflection and room temperature photoluminescence of obtained samples showed strong dependence on their crystallinity and composition. Photocatalytic activity of ZnO/ZnWO4 nanopowders was checked using photodegradation of selected dyes as model system. Obtained results were correlated with specific surface area, particle sizes, crystallinity and ZnO/ZnWO4 ratio of the samples. As crystallinity of ZnWO4 component in the ZnO/ZnWO4 increase, photocatalytic activity also increases. The main findings can be explained by charge transfer reactions that follow light absorption by ZnO and ZnWO4 in nanocomposite. Keywords Optical materials Chemical synthesis Interfaces X-ray diffraction Photocatalysis
Introduction The absorption of light by a semiconductor results in creation of charge carriers (electrons and holes) through a process of electronic excitation between the valence and conduction band. Once created, these charge carriers are able to migrate to the surface of the semiconductor and undergo redox reactions with adsorbed molecules. T. D. Savic´ I. Lj. Validzˇic´ M. I. Cˇomor (&) Vincˇa Institute of Nuclear Sciences, University of Belgrade, P.O. Box 522, 11001 Belgrade, Serbia e-mail:
[email protected] T. B. Novakovic´ Z. M. Vukovic´ IChTM-Department of Catalysis and Chemical Engineering, University of Belgrade, Njegosˇeva 12, 11000 Belgrade, Serbia
123
T. D. Savic´ et al.
This process, known as heterogenous photocatalysis, is currently of significant research interest as a method for destruction of chemical pollutants [1]. From previous experimental studies, it is well known that there exists an optimal particle size and shape for which the activity of particulate photocatalyst is maximized. As a consequence, an ability to control particle size and shape provides a means of tailoring the photocatalytic activity for specific commercial application [2–5]. ZnO and ZnWO4 are photocatalysts that draw attention of scientists lately. It is known that ZnO shows good photocatalytic effect and high quantum activity for degradation of environmental polutants [2, 3]. ZnO has been characterized with its wide bandgap (3.37 eV) and relatively large exciton binding energy (60 meV) at room temperature [4, 5]. ZnWO4 has been also used for water splitting and mineralization of organic pollutants under UV irradiation [6–9]. Its commercial application is modest because the photocatalytic activity of ZnWO4 is not high enough for the requirements of practical application [8]. Literature data regarding band gap energy are different: ranging from 3.8 eV [10], 4 eV [7] to 4.6 eV obtained by quantum chemistry calculations [11]. Previous experimental studies have shown that wet chemical techniques can successfully produce nanoparticulate photocatalysts in the ZnO–ZnWO4 system, including single phase ZnWO4 [12] and dual phase ZnO/ZnWO4 [13, 14]. Here, we report low temperature (95 °C), one-pot method for preparation of ZnO/ZnWO4 rod-like nanopowders that include the assistance of non-ionic block copolymer (Pluronic F68). The obtained materials were characterized using morphological, texture and optical techniques. The main subject of this paper is detailed correlation between structure (morphology) and texture properties of rodlike ZnO/ZnWO4 composite nanoparticles and their optical properties and photocatalytic activity regarding degradation of selected dyes: synergistic effect of ZnO and ZnWO4 in nanocomposite. We will show that close junction between ZnO and ZnWO4 and charge transfer processes define optical properties, especially photoluminescence, of composite as well as its photocatalytical efficacy.
Materials and Methods All chemicals: Na2WO42H2O (99 % Riedel-de Hae¨n), ZnCl2 (99 % Merck), nonionic copolymer surfactant Pluronic F68 (Polyoxyethylene-polyoxypropylene block copolymer, Mn * 8,400 (Aldrich)), NaOH (98 % Fluka) and ZnO (C99 % SigmaAldrich), were of the highest purity available and they were used without further purification. Synthesis ZnO/ZnWO4 rod like nanoparticles were prepared as described in literature [14]. Briefly, 0.1 M ZnCl2 solutions were mixed with 100 mL of copolymer solution (10 g/L). The pH of the solution was adjusted to 8 using 0.1 M NaOH. Under vigorous stirring, 0.1 M Na2WO42H2O was added drop by drop, and the mixture was refluxed at 95 °C for 1, 5, 48 and 120 h, assigned as samples A, B, C, D
123
A Synergy of ZnO and ZnWO4
respectively. During the reflux, precipitation of ZnO and ZnWO4 took place. The obtained ZnO/ZnWO4 nanoparticles were separated from solvent containing copolymer immediately after synthesis by using ultra-centrifugation. Synthesized ZnO/ZnWO4 nanoparticles were washed several times with ethanol and distilled water using centrifugation in every washing step and annealed at 95 °C for 18 h. The chemical composition was checked using JEOL JSM 6460 LV (SEM) equipped with the Energy Dispersive Spectrometer (EDS INCA x-sight Oxford Instruments). The used acceleration potential was 25 kV. ZnO/ZnWO4 pellets with thickness around 2 mm were coated with carbon. Only Zn, W and O were detected in our samples. From atomic and weight present ratio of Zn and W we estimated that ZnO:ZnWO4 ratio was about 1:2, in moles, for all samples. Structural Characterization Microstructural characterization of ZnO/ZnWO4 rod-like composite nanoparticles was performed by transmission electron microscopy (TEM) using JEOL 100 CX microscope operated at an accelerating voltage of 100 kV. The samples for TEM measurements were prepared by dispersing a powder in water using ultrasonication. The obtained samples were placed on C-coated Cu grids. The electron diffraction (ED) patterns were recorded with a plate camera. Optical Properties Diffuse reflectance spectra (DRS) of the ZnO/ZnWO4 pellets were recorded using Perkin Elmer Lambda 9 UV–Vis-NIR Spectrophotometer. Photoluminescence (PL) spectra were obtained using Perkin Elmer LS 45 Luminescence spectrometer. Textural Properties Textural properties were determined via N2 adsorption–desorption isotherms, which were obtained at 77 K with an automatic adsorption apparatus (Sorptomatic 1990 Thermo Finningen). The specific surface area of the samples was calculated by fitting the adsorption data to the linear range of Brunauer–Emmet–Teller equation (relative to p/p0 (0.05–0.35)) [15]. The pore size distribution was calculated according to the Barret–Joyner–Halenda method using the N2 desorption isotherms, while the micropore volume was calculated according to the Dubinin and Raduskevich method [16]. Photocatalytical Testing Methylene blue (Fluka) and Rhodamin B (Sigma) were used as model dyes to evaluate the photocatalytic activity of the commercial ZnO and synthesized ZnO/ ZnWO4 samples. In a typical experiment 0.5 mg/mL of catalyst was dispersed in a water solution containing 1 9 10-5 M of disolved dye. The obtained dispersions were thoroughly mixed and left in a dark over night, in order to obtaine equilibrium between free dye and dye adsorbed on particle’s surface. The photocatalytic
123
T. D. Savic´ et al.
experiments were conducted at room temperature (*18 °C) under UV light of an 17 W low-pressure lamp with max at 254 nm (Model UV17F Helios Italquartz), in the streem of O2. Volume of typical experimental setup was 20 mL. Aliquots were extracted during irradiation and centrifuged at 5,000 rpm for 10 min. The decomposition of dyes was monitored by measuring the absorbance of the supernatant at k = 554 nm (RhB) or k = 664 nm (MB) using UV/Vis spectrophotometer (Evolution 600 Thermo EC) in quartz cuvette. For comparison, photodegradation of RhB and MB in the presence of ZnO commercial photocatalyst and photolysis of dyes under UV irradiation were also performed.
Results and Discussion TEM and ED of ZnO/ZnWO4 Nanoparticles Powders obtained after heating and annealing were morphologically and structurally characterized using TEM. Results are presented in Fig. 1, were typical TEM images and ED patterns are presented. As can be seen sample A (Fig. 1a) is consisted of irregularly shaped nanoparticles with fade and undistinguishable diffraction rings referring that particles are amorphous or of very low crystallinity. Samples from B to D (Fig. 1b, c and d, respectively) have nanoparticles with increasing definition of rodlike shape and ED patterns are more and more easily distinguishable. For sample D, ED rings are assigned to corresponding ZnWO4 planes (JCPDS 00-015-0774). No diffraction that can be assigned to ZnO was obtained with this technique, although ZnO is present in all samples as will be shown in optical characterization part of this manuscript. Zhang et al. [17] also showed that ZnO nanoparticles can have small crystalline domains, almost undetectable by XRD technique, but with pronounced photoluminescence. It can be concluded that crystallinity of the samples increase with increasing the reaction time at 95 °C from 1 to 120 h. From the TEM images with higher magnification, sizes of ZnO/ZnWO4 rod-like nanoparticles were estimated to be in the range from 10 to 30 nm (diameter *10 nm and length *30 nm). Optical Properties of ZnO/ZnWO4 Nanoparticles Optical absorption of obtained samples was investigated by diffuse reflectance spectroscopy. Obtained data were converted in absorption units (Fkm) using Kubelka–Munk equation and presented in Fig. 2. Two regions of the wavelengths can be observed, region which cover ZnO (*3.37 eV) absorption, k C 350 nm, and region that corresponds to band gap of both ZnO and ZnWO4 (C3.8 eV), k B 350 nm. The part of the spectra that corresponds to ZnO is enlarged in the inset of Fig. 2. As can be seen, sample A (1 h reaction time) shows only one shoulder peaking at about 370 nm, which corresponds well with band gap of ZnO. After that shoulder, absorption increases monotonically with decreasing wavelengths. Most probably ZnWO4 crystall phase was not formed yet. According to chemical reactions that lead to precipitation of ZnO/ZnWO4 [14], ZnO phase is formed before ZnWO4:
123
A Synergy of ZnO and ZnWO4
Fig. 1 Typical TEM images of the as-synthesized ZnO/ZnWO4 nanoparticles obtained for sample A (a), B (b), C (c) and D (d) reaction time and corresponding ED patterns. Pattern d contains corresponding ZnWO4 planes (JCPDS 00-015-0774)
ZnCl2 þ 2H2 O ! ZnðOHÞ2 þ 2HCl ! ZnO þ H2 O þ 2HCl
ð1Þ
ZnðOHÞ2 þ Na2 WO4 ! ZnWO4 þ 2NaOH
ð2Þ
In spectra of samples B, D and C a shoulder at the same position (Fig. 2, inset *370 nm) only less stressed, and broad peak at about 270 nm can be seen. This broad peak can be assigned to superposition of ZnO and ZnWO4 absorption, although its maximum corresponds well with literature data for band gap of ZnWO4 (4.6 eV). Sample D (the longest reaction time) showed the lowest intensity of absorption in the region of ZnO absorption, probably because ZnO phase is mostly spent on the ZnWO4 synthesis. Sample D had crystalline ZnWO4 phase (Fig. 1d) with sharp increase of absorption at about 320 nm. It should be mentioned that small peak at 290 nm can be seen in the spectra of samples B and C is not detectable in spectra of samples A and D. This peak is most probably characteristic for mixed
123
T. D. Savic´ et al.
Fig. 2 Absorption spectra of ZnO/ZnWO4 powders calculated from DRS by Kubelka–Munk transformation
ZnO/ZnWO4 phases, where neither of oxides is dominant. We presumed that sample A is mainly ZnO and sample D is mainly ZnWO4 with traces of ZnO which can be detected due to its PL properties. PL spectra of obtained samples, using two different excitation wavelengths, are presented in Fig. 3. Bearing in mind our previous results obtained with similar ZnO/ ZnWO4 rod-like nanoparticles [14] we assigned PL properties to ZnO phase of our composite. We used two excitation wavelengths, kexc = 270 nm (4.6 eV) and kexc = 330 nm (3.75 eV), Fig. 3a and b respectively, in order to compare PL spectra obtained when both materials are excited (band gap of ZnWO4 is between 3.8 and 4.6 eV and of ZnO is about 3.37 eV) with PL spectra when only ZnO is excited. Firstly, it can be seen that PL differences between samples obtained after different reaction times are more prominent when grater energy of excitation is used (Fig. 3a). In PL spectra of all samples (Fig. 3a) peak at about 380 nm can be observed with maximum intensity for sample A and minimum intensity for sample B. This peak can be assigned to ZnO band edge emission. Prolonging reaction time from 5 h (sample B), 48 h (sample C) to 120 h (sample D) induced increase of intensity of 380 nm peak as well as green emission intensity (k C 400 nm). These results indicated that prolonged reaction time gives rise to formation of near-surface oxygen vacancies in ZnO which cannot be blocked by formation of crystaline ZnWO4 layer. Most probably, composites with pronounced green PL (induced by surface defects) can also have ferromagnetic properties as stated in literature [18]. We plan these studies in the near future. Similar PL spectra measured Kim et al. [19] when nanorod arrays of ZnO were surface covered by different oxides (TiO2, Y2O3, CeO2, and Er2O3). Obviously surface layer of ZnWO4 favorably induced the
123
A Synergy of ZnO and ZnWO4
Fig. 3 PL spectra of ZnO/ZnWO4 powders for two excitation wavelengths: 270 nm (a) and 330 nm (b)
formation of oxygen vacancies in ZnO nanoparticles responsible for green emission. Amorphous ZnO (Fig. 3a curve A) showed only band edge emission that is typical for small ZnO clusters [17]. On the other hand, excitation with photons of lower energy (3.75 eV, Fig. 3b) gave us PL spectra of almost no difference regarding of reaction time. Only peak that can be assigned to ZnO band edge emission can be seen (*380 nm). The differences that can be observed comparing spectra in Fig. 3a and b, gave us insight in complicated mechanism of filling the near-surface oxygen vacancies in ZnO nanoparticles. Textural and Photocatalytical Properties of ZnO/ZnWO4: Interfacial Charge Transfer Textural properties of our samples and corresponding pseudo-first order rate constants of dye photodecomposition (MB-Methylene blue and RhB- Rhodamin B) are presented in Table 1 and in Fig. 4. We compared results obtained using our samples with commercial crystalline ZnO (Sigma-Aldrich, d * 30 nm). Commercial nanopowder has the best photocatalytic activity. As can be seen there is no liner dependence between specific surface area (SBET) and photocatalytic activity of our samples. The highest value for SBET has been measured for sample B and highest pseudo-first order rate constants for photodecomposition of both dyes (MB-Methylene blue and RhB- Rhodamin B) have been established for sample C. As reaction time for synthesis of our samples increases from sample A to sample C, photocatalytic activity increases, as well as crystallinity, but for the most crystalline sample D photocatalytic activity slightly decrease. This effect can be explained by
123
T. D. Savic´ et al. Table 1 Texture properties of ZnO/ZnWO4 nanopowders and corresponding pseudo-first order rate constants of dye photodecomposition Sample
SBET (m2g-1)
Vp (cm3g-1)
Vmicro (cm3g-1)
D (nm)
A
30
0.215
0.013
24.5
0.05
0.04
B
94
0.162
0.040
6.0
0.15
0.13
C
82
0.155
0.034
7.3
0.51
0.38
D
73
0.130
0.029
7.7
0.30
0.22
ZnO
50
/
/
2.00
0.73
30
MB, k(min-1)
RhB, k(min-1)
Vp-total pore volume (Gurvich) at p/p0 0.99, d mean pore diameter. SBET of commercial ZnO (SigmaAldrich) was calculated from data obtained from the producer
Fig. 4 Specific surface area and photodegradation rate constants (for degradation of MB and RhB) of ZnO/ZnWO4 powders and commercial ZnO powder
core/shell structure of rod-like nanoparticles, where ZnO is core and ZnWO4 exist as a shell which partially cover the surface of ZnO (as proposed by chemical reactions (1) and (2) in Section ‘‘Optical Properties of ZnO/ZnWO4 Nanoparticles’’). In sample C, most probably, a significant quantity of ZnWO4 is present at the surface of ZnO but also part of ZnO surface is still free and available for photocatalytic degradation of dyes. Figure 5 depicts energetic positions of the conduction and valence bands (cb, vb) of ZnO and ZnWO4 and the values of their band gap energies [20]. As presented in a part ‘‘Photocatalytical Testing’’ we used for photocatalysis UV light with k = 254 nm (4.88 eV). Used photons have energy greater than band gap energies of both constituents of composite and electron–hole pairs were generated in both materials in our composite. The cb edge of ZnWO4 is more negative than cb edge of ZnO and vb of ZnWO4 is more positive than vb edge of ZnO (Fig. 5); as a result of these facts charge carriers from ZnWO4 tend to ‘‘escape’’ to ZnO through close junction of composite in order to lower their energies (thermodynamically more stable positions). So, after light absorption, all
123
A Synergy of ZnO and ZnWO4 Fig. 5 Electronic correlation diagram of the ZnO/ZnWO4 nanoparticles: cb conduction band, vb valence band, band gaps in the brackets
photogenerated holes and electrons are transferred to ZnO part of the composite due to band edge positions. This process is thermodynamically allowed (see Fig. 5) and favored. This is the cause of synergistic effect of ZnO/ZnWO4 close junction. These charge carriers can now react with adsorbed dye molecules, water and oxygen which explain the above mentioned necessity to maintain free surface of ZnO core for efficient photocatalysis. Obviously, sample C has optimal ratio between ZnO and ZnWO4 for efficient absorption of light and enough free ZnO surface where photodegradation can take place. Further growth and crystalization of the ZnWO4 shell is not beneficial for photocatalytic activity. Energetic diagram presented in Fig. 5 also explain why all room-temperature photoluminescence characteristics of ZnO/ZnWO4 nanoparticles belong entirely to ZnO part of composite. After photogeneration of charge carriers in ZnWO4 shell, they undergo fast extraction in ZnO core where they can be recombined with emission of photon or lead useful chemical reactions if reactive species are adsorbed on its surface.
Conclusions A low temperature method (95 °C), which involves non-ionic copolymer surfactant, was used for preparation of ZnO/ZnWO4 rod-like composite nanoparticles. TEM measurements showed that particles were of nanodimensions and with rod-like shapes. The development of ZnO and ZnWO4 phase was followed by optical absorption measurements. Room-temperature PL was observed for all samples, using two excitation wavelengths (270 and 330 nm). They were characterized by band edge emission of ZnO (380 nm) for both excitations, and green emission from oxygen vacancies present in ZnO when kexc = 270 nm (4.88 eV) was used. Photocatalytical and optical properties were correlated and explained by conduction and valence band positions in our composite nanoparticles. The most effective synergy between ZnO and ZnWO4 was clearly observed for nanoparticles obtained after 48 h reaction time. Acknowledgments Financial support for this study was granted by the Ministry of Education, Science and Technological Development of the Republic of Serbia, Project No. III45020, III45001, ON172015 and ON172056.
123
T. D. Savic´ et al.
References 1. M. Hoffman, S. Martin, W. Choi, and D. Bahnemann (1995). Chem. Rev. 95, 69. 2. S. Sakthivel, B. Neppolian, M. V. Shankar, B. Arabindoo, M. Palanichamy, and V. Murugesan (2003). Energy Mater. Sol. Cells 77, 65. 3. H. H. Wang, C. S. Xie, W. Zhang, S. Z. Cai, Z. H. Yang, and Y. H. Gui (2007). J. Hazard. Mater. 141, 645. 4. W. I. Park, G. C. Yi, M. Kin, and S. J. Pennycook (2003). Adv. Mater. 15, 526. 5. P. Li, Y. Wei, H. Liu and X. Wang (2004). Chem. Commun. 2856. 6. X. Zhao and Y. F. Zhu (2006). Environ. Sci. Technol. 40, 3367. 7. X. Zhao, W. Q. Yao, Y. Wu, S. C. Zhang, H. P. Yang, and Y. F. Zhu (2006). J. Solid State Chem. 179, 2562. 8. H. B. Fu, J. Lin, L. W. Zhang, and Y. F. Zhu (2006). Appl. Catal. A 306, 58. 9. G. L. Huang, C. Zhang, and Y. F. Zhu (2007). J. Alloys Compd. 432, 269. 10. J. Bi, L. Wu, Z. Li, Z. Ding, X. Wang, and X. Fu (2009). J. Alloys Compd. 480, 684. 11. A. Kalinko, A. Kuzmin, and R. A. Evarestov (2009). Solid State Commun. 149, 425. 12. H. Fu, P. Chengesi, L. Zhang, and Y. Zhu (2007). Mater. Res. Bull. 42, 696. 13. A. Dodd, A. McKiley, T. Tsuzuki, and M. Saunders (2009). J. Eur. Ceram. Soc. 29, 139. 14. I. Lj. Validzˇic´, T. D. Savic´, R. M. Krsmanovic´, D. J. Jovanovic´, M. M. Novakovic´, M. Cˇ. Popovic´ and M. I. Cˇomor (2012). Mat. Sci. Eng. B 177, 645. 15. K. Sing, D. Everett, R. Haul, L. Moscou, R. Pierotti, J. Rouquerol, and T. Siemieniewska (1985). Pure Appl. Chem. 57, 603. 16. E. P. Barrett, L. G. Joyner, and P. H. Halenda (1951). J. Am. Chem. Soc. 73, 373. 17. H. Zhang, G. Chen, G. Yang, J. Zhang, and X. Lu (2007). J. Mater. Sci: Mater. Electron. 18, 381. 18. L. S. Panchakarla, Y. Sundarayya, S. Manjunatha, A. Sundaresan, and C. N. R. Rao (2010). ChemPhysChem. 11, 1673. 19. Y. Kim, Y. Kim, and S. Kang (2010). J. Phys. Chem. C 114, 17894. 20. M. Bonnani, L. Spanhel, M. Lerch, E. Fu¨glein, and G. Mu¨ller (1998). Chem. Mater. 10, 304.
123
