Photocatalytic water splitting

PHOTOCATALYTIC WATER SPLITTING TEO YU HAN (B.Eng.(Hons.), NUS) A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF ENGINEERING DEPARTMENT OF ELECTRICAL & COMPUTER ENGINEERING NATIONAL UNIVERSITY OF SINGAPORE 2013 Declaration I hereby declare that this thesis is my original work and it has been written by me in its entirety. I have duly acknowledged all the sources of information which have been used in the thesis. This thesis has also not been submitted for any degree in any university previously. ____________________________________ Teo Yu Han 19 August 2013 i Acknowledgements First of all, I would like to express my deepest appreciation to Dr Ho Ghim Wei for giving me the chance to carry out my master studies under her supervision. She has been an excellent mentor who has displayed great patience and foresight, and her invaluable guidance and insightful suggestions has helped me a great deal during the course of my work. I would also like to thank Mr Thomas Ang, the lab officer in charge of our lab. His high standards of lab safety and logistics have enabled me to carry out my work smoothly without any serious logistics issues. My sincere thankfulness also goes out to my fellow colleagues: Connor, Franco, Gah Hung, Kevin, Minmin and Wei Li. All of you have made the time spent in lab enjoyable and also gave me a lot of ideas and insight during the many discussions that we had. I would also like to thank Prof Sow Chorng Haur from the Physics Department for giving me access to some of his equipment and Mr Chen for many of his logistical help. Finally, and most importantly, my profound gratitude goes to my family especially my fiancée for her love and all the help she has provided me throughout the course of my work. Without her constant support, motivation and love, I would not have been able to finish this work. Thank you for everything. ii Table of contents Declaration i Acknowledgements ii Table of contents iii Summary v List of tables vii List of figures viii List of symbols xi Chapter 1 Introduction to photocatalytic water splitting 1 1.1 Introduction 1 1.2 Mechanism of water splitting 3 1.3 Methods in enhancing photocatalytic efficiency 4 1.3.1. Sensitization 4 1.3.2 Morphology modification 5 1.3.3 Doping 6 1.3.4 Co-catalyst loading 7 1.3.5 Thermal treatment 7 1.3.6 Utilization of localized surface plasmon resonance (LSPR) effect 8 1.4 Z-Scheme 9 1.5 TaON and WO3 as Z-scheme photocatalysts 14 References 16 Chapter 2 Synthesis of TaON via urea route for the photocatalytic reduction of water 23 2.1 Introduction 23 2.2 Experimental procedures 25 2.2.1 Synthesis conditions of TaON samples 25 2.2.2 Photocatalytic reactions 27 2.3 Results and discussion 28 2.3.1 Characterization of photocatalyst 28 2.3.2 Process of elemental nitrogen incorporation in Ta2O5 in the synthesis of TaON 34 2.3.3 Photocatalytic H2 evolution performance 37 2.3.3.1 Effect of in-situ synthesis vacuum calcination duration 37 2.3.3.2 Effect of post-synthesis calcination 40 iii 2.3.3.3 Effect of pre-synthesis calcination 40 2.4 Conclusions 42 References 43 Chapter 3 Loading of CuO nanoparticles on WO3 for enhanced visible light response for photocatalytic oxidation of water 47 3.1 Introduction 47 3.2 Experimental procedures 49 3.2.1 Synthesis of CuO-loaded WO3 composite photocatalyst 49 3.2.2 Photocatalytic reactions 51 3.3 Results and discussion 52 3.3.1 Materials characterization 52 3.3.2 Synthesis conditions of CuO nanoparticle and its loading process on WO3 58 3.3.3 Photocatalytic O2 evolution performance 61 3.3.3.1 Effect of CuO nanoparticle loading amount 62 3.3.3.2 Effect of post-synthesis calcination 67 3.4 Conclusions 72 References 73 Chapter 4 Loading of AgCl/Ag hybrid nanostructure on WO3 as electron-accepting co-catalyst on WO3 78 4.1 Introduction 78 4.2 Experimental procedures 81 4.2.1 Synthesis of AgCl/Ag-WO3 composite photocatalyst 81 4.2.2 Photocatalytic reactions 84 4.3 Results and discussion 85 4.3.1 Materials characterization 85 4.3.2 Synthesis process of AgCl/Ag nanoparticle and its function 89 4.3.3 Photocatalytic O2 evolution rate 92 4.3.3.1 Effect of calcination on AgCl/Ag-WO3 93 4.3.3.2 Effect of AgCl/Ag co-catalyst loading amount 99 4.4 Conclusions 101 References 102 Conclusion 105 iv Summary Photocatalytic water splitting reaction is a chemical process that involves direct solar energy conversion of H2O to H2 and O2 on heterogeneous photocatalysts. Upon absorbing photons, the photocatalyst would generate electron-hole pairs whereby the electrons would reduce the H+ ions to H2 and the holes would oxidize H2O to O2. However, it is challenging to achieve overall water splitting by using a single photocatalyst. One of the alternatives is to utilize the Z-scheme system consisting of two separate H2 and O2producing photocatalysts. Two Z-scheme photocatalysts will be the main study subject in this work, namely the H2-producing TaON and O2-producing WO3. In general, photocatalysts face the problem of large bandgap which renders the absorption of visible light. Besides, recombination of electron-hole pair is also limiting the photocatalytic performance by reducing the count of charge carrier available for photocatalytic reactions. There are many ways to enhance the visible light absorption of a photocatalyst, such as narrowing the photocatalyst bandgap through nitrogen doping. TaON is a nitrogen-doped photocatalyst synthesized via the conventional nitridation method by exposing Ta2O5 to flowing NH3 at high temperature. However, the nitridation process is energy intensive and uses hazardous NH3. In this work, an alternative approach to synthesize TaON will be introduced, namely the urea route which utilizes nitrogen-rich urea and requires a significantly shorter annealing duration. The effects of vacuum calcination duration and the pre-synthesis calcination on the Ta2O5 precursor on TaON’s photocatalytic H2 evolution performance were investigated. Another approach to enhance the utilization of visible light is through the loading of sensitizing agent such as CuO nanoparticles. CuO has a narrow bandgap which allows greater visible light absorption. Upon loading CuO on WO3, CuO’s sensitizing effect would enhance the photogeneration rate of hole and increase the photocatalyst’s O2 evolution rate. Furthermore, the formation of p-n junction between CuO and WO3 also enhances the separation of electron-hole pairs, thus reducing charge recombination. In this work, the feasibility of loading CuO nanoparticle on WO3 in enhancing the photoactivity v of the composite photocatalyst was investigated. The effects of CuO nanoparticle loading amount and the post-synthesis annealing process on the photoactivity were also studied. The loading of co-catalyst is also effective in reducing electron-hole pair recombination. The studies on the effects by AgCl/Ag co-catalyst with a hybrid nanostructure on the O2 evolution performance of WO3 were performed. The metallic Ag section acts as an electron sink, thereby allowing the holes in WO3 to oxidize H2O more efficiently. The rationale behind the use of such hybrid structure is that AgCl-loaded WO3 photocatalyst could be calcined prior to the partial photoreduction of AgCl surface to Ag. Thermal annealing is routinely employed to enhance bonding between a co-catalyst and its host photocatalyst. Pure Ag nanoparticle-loaded WO3 is not suitable to undergo calcination process as it would be oxidized, thus rendering its capability as an electron sink. In this work, the effect of the co-catalyst loading amount and the post-synthesis calcination conditions were investigated to obtain the optimal photoactivity by the AgCl/Ag-loaded WO3 photocatalyst. vi List of tables Table 2.1 Synthesis conditions of samples T1 to T7. 26 Table 2.2 Summary of synthesis conditions for samples T1 to T7. 27 Table 2.3 Photocatalytic activities for samples T1 to T7. 37 Table 2.4 EDX readings for the average atomic % of nitrogen content in samples T1, T2 and T3. 38 Table 3.1 Photocatalytic O2 evolution rate of pristine WO3 and the CuO- 61 WO3 composite photocatalyst samples loaded with 0.5, 1, 2, 4, 6 and 8 wt. % CuO nanoparticles. Table 3.2 Photocatalytic O2 evolution rate of CuO-WO3 composite photocatalyst samples calcined at various temperatures and atmospheres. 62 Table 4.1 EDX analysis on the Ag and Cl elemental atomic % of sample A and B. 87 Table 4.2 Photocatalytic O2 evolution rate for the 10 samples synthesized with various calcination temperatures of AgCl loading wt. %. 92 vii List of figures Figure 1.1 Conduction and valence band edge positions of several photocatalysts. 10 Figure 1.2 Photocatalytic reactions involved in the Z-scheme system. 13 Figure 2.1 XRD spectra of (a) sample T1 and (b) Ta2O5. 29 Figure 2.2 (a) UV–visible diffuse reflectance spectra for Ta2O5 and TaON and (b) colour difference between Ta2O5 and TaON (sample T7). 30 Figure 2.3 Tauc plot of the synthesized TaON sample. 32 Figure 2.4 X-ray photoelectron spectra for (a) Ta 4f, (b) O 1s and (c) N 1s regions of Ta2O5 and TaON. 33 Figure 2.5 SEM images of (a) Ta2O5, and (b) TaON. 33 Figure 2.6 Colour of various oxynitride materials synthesized at the initial vacuum calcination temperature of (a) 725, (b) 700 and (c) 650 ˚C. 35 Figure 2.7 Schematic representation of the energy band structures of Ta2O5 and TaON. 36 Figure 2.8 SEM images of samples T1, T2 and T3 showing various degree of particle agglomeration. 39 Figure 2.9 XRD spectrum of sample T4. 40 Figure 2.10 PL spectra of samples T1, T6, and T7. 41 Figure 3.1 XRD spectra for (a) CuO nanoparticles and (b) CuO-WO3 composite photocatalyst. 54 viii Figure 3.2 UV-Vis absorbance of WO3, CuO nanoparticles and the CuO-WO3 composite photocatalysts loaded with various wt. % of CuO nanoparticles. 55 Figure 3.3 SEM images of (a) pristine WO3, (b) CuO nanoparticles and CuO-WO3 composite photocatalyst at (c) 30 k and (d) 50 k magnification. 56 Figure 3.4 EDX spectrum for the CuO-WO3 composite photocatalyst sample indicating the presence of Cu in the sample. 57 Figure 3.5 TEM images of the CuO-WO3 composite photocatalyst loaded with 4 wt. % CuO nanoparticle and the obtained lattice spacing value of CuO. 58 Figure 3.6 Colour of WO3 suspension in solution with the pH value at (a) neutral, (b) 8 and (c) 10. 60 Figure 3.7 Comparison in photocatalytic O2 evolution rate between pristine WO3 and the CuO-WO3 composite photocatalyst samples loaded with 0.5, 1, 2, 4, 6 and 8 wt. % CuO nanoparticles. 62 Figure 3.8 Charge carrier transfer mechanism between the CuO and 65 WO3 due to (a) sensitization effect by CuO nanoparticle, and (b) the role of CuO in separating the photoinduced electrons and holes. Figure 3.9 Photocatalytic O2 evolution activity of sample C6 for a duration of 6 h. 66 Figure 3.10 Comparison in photocatalytic O2 evolution rate between pristine sample C4 and the CuO-WO3 composite photocatalyst samples annealed at various temperatures and atmospheres. 68 Figure 3.11 XRD spectra of (a) sample D4 annealed in N2, (b) unannealed sample C4 and (c) sample D3 annealed in air. 72 Figure 4.1 Colour of the composite photocatalyst suspension at 6 different stages: (a) prior to irradiation, (b) 5 s, (c) 2 min, (d) 4 min, (e) 5 min into irradiation and (f) after irradiation process. 83 ix Figure 4.2 Flowchart of the systhesis process for the AgCl/Ag-WO3 composite photocatalyst. 84 Figure 4.3 Images of (a) pristine WO3 and the composite photocatalyst at stage (b) 1, (c) 2 and (d) 3. 84 Figure 4.4 SEM images of AgCl/Ag-WO3 composite photocatalyst of (a), (b) sample A, and (c), (d) sample B. 86 Figure 4.5 XRD spectrum of (a) composite photocatalyst sample with 10 wt. % AgC/Ag nanoparticle loading and (b) pristine WO3. 89 Figure 4.6 Photoreduction of AgCl nanoparticle loaded on WO3 to the AgCl/Ag hybrid nanostructure with and without ethanol. 92 Figure 4.7 Comparison in O2 evolution rate among prisinte WO3 and samples A1 to A5 post-calcined at various temperatures. 94 Figure 4.8 Width of AgCl diffraction peaks for samples A1, A2 and A3 at around 2θ = 32.2 °. 95 Figure 4.9 Images of PVP at its (a) pristine state, and calcined at (b) 350 ˚C, (c) 450 ˚C and (d) 550 ˚C. 97 Figure 4.10 Images of (a) pristine WO3 and samples (b) A1, (c) A2, (d) A3, (e) A4 and (f) A5. 99 Figure 4.11 Comparison in O2 evolution rate among samples B1 to B5. 100 x List of symbols Ag Silver Ag+ Silver ion Ag2O Silver(I) oxide AgCl Silver chloride AgNO3 Silver nitrate Al Aluminium AgO Silver(II) oxide Au Gold B Boron BiVO4 Bismuth vanadate Bi2O3 Bismuth(III) oxide C Carbon C3N4 Carbon nitride CaFe2O4 Calcium Ferrite CaTiO3 Calcium titanate CdS Cadmium sulfide CdSe Cadmium selenide CdTe Cadmium telluride Ce3+ Cesium(III) ion Ce4+ Cesium (IV) ion CH3COOH Acetic acid CoOx Cobalt oxide Cr Chromium Cu2+ Copper(II) ion CuO Copper oxide Cu(OH)2 Copper hydroxide Cu(NO3)2 Copper nitrate DI Deionized EDTA ethylenediaminetetraacetic acid EDX Energy-dispersive X-ray spectroscopy eV Electronvolt xi F Fluorine Fe2+ Iron(II) ion Fe3+ Iron(III) ion Fe2O3 Iron(III) oxide FeCl3 Iron(III) chloride g-C3N4 Graphitic carbon nitride h Hour H+ Hydrogen ion H2 Hydrogen H2O Water HOMO Highest occupied molecular orbital HOQ 8-hydroxyquinoline I- Iodide ion IO3- Iodate ion IrOx Iridium oxide LSPR Localized surface plasmon resonance LUMO Lowest unoccupied molecular orbital MEMS Microelectromechanical system MgPc Magnesium phthalocyanine Min Minute N Nitrogen N2 Nitrogen NaOH Sodium hydroxide NaNO3 Sodium nitrate NHE Normal hydrogen electrode nm nanometer NO2 Nitrogen dioxide O Oxygen O2 Oxygen OH Hydroxide OH- Hydroxide ion PL Photoluminescence Pt Platinum xii PVP Polyvinylpyrrolidone rGO Reduced graphene oxide RuOx Ruthenium oxide s Second S Sulfur S2- Sulfide ion SEM Scanning electron microscopy SERS Surface Enhanced Raman Scattering SrTiO3 Strontium titanate Ta Tantalum Ta2O5 Tantalum oxide Ta3+ Tantalum(III) ion TEM Transmission electron microscopy TG-DTA Thermal Analysis - Differential Thermal Analysis Ti 3+ Titanium(III) ion TiO2 Titanium oxide UV Ultraviolet UV-Vis UV-Visible spectrometer W Watt WO3 Tungsten oxide / tungsten trioxide wt. % Weight percentage XRD X-ray diffraction spectroscopy ZnO Zinc oxide xiii Chapter 1 Introduction to photocatalytic water splitting 1.1 Introduction Global demand and consumption for energy has been rising significantly due to the rapid expansion in economy by both developed and developing nations. Traditional fossil fuel sources such as petroleum, natural gas and coal are still the primary source for power which account for about 81 % globally in the year 2008. However, fossil fuel reserve is finite and the continual staggering usage of fossil fuels will gradually result in its depletion. The dwindling of fossil fuel reserve will give rise to volatility in fuel prices as well as its supply, thus resulting in economy instability. Besides, the combustion of fossil fuels also emits polluting greenhouse gasses strongly believed to be responsible for global warming.1,2 Due to these reasons, generation of cheap, clean and renewable energy as an alternative energy source to fossil fuel is becoming more viable and gradually gaining attention from various stakeholders . Solar energy, on the other hand is clean and environmentally friendly. Its source is virtually inexhaustible and has potential of fulfilling the ever increasing global energy demand. Silicon as well as organic-based photovoltaics have been heavily used to tap the potential of solar energy. Another method of harvesting solar energy is through the use of photocatalysts to undergo water splitting reaction, discovered by Fujishima and Honda in the year 1972.3 Both Fujishima and Honda first demonstrated the overall water splitting reaction (i.e., simultaneous generation of both H2 and O2 gases) by using a photoelectrochemical cell consisting of a single-crystalline rutile TiO2 anode and a Pt cathode under ultraviolet (UV) irradiation with an external bias. Photocatalytic water splitting is also commonly known as artificial photosynthesis due to the fact that water splitting process involves direct solar energy conversion to chemicals on heterogeneous photocatalysts. Upon absorbing photons, photocatalysts are able to chemically split water molecules (H2O) to H2 and O2 gases. The significance of photocatalytic water splitting is the generation of H2 gas via a cheap and clean manner. As a matter of fact, H2 in its atomic form, H is the most abundant chemical element 1 in the universe and found naturally in fossil fuels, water and most organic compounds. H2 can be used to generate energy in hydrogen combustion engines or fuel cells which do not result in any toxic or greenhouse gas emission but produces water instead.4,5 However, H2 does not occur naturally and needs to be artificially produced. One of the many methods commonly used to produce H2 is through the conversion of biomass via the following reactions:6,7 (1) (2) Another approach to generate H2 is through steam reforming of methanol or ethanol at high temperatures whereby this process accounts for approximately 95 % of the worldwide H2 generation processes.8-10 The steam reforming method from methanol and ethanol are represented by Eq. (3), (4) and (5), (6), respectively as shown below. (3) (4) (5) (6) Unfortunately, these methods will result in the generation of unwanted CO and CO2 greenhouse gases which are the primary cause for global warming. Besides, the above-mentioned methods are energy-intensive processes whereby the energy is normally produced from fossil fuel sources. Thus the produced H2 gas may not be suitable for powering fuel cells after all. Electrolysis method is also being used for the production of H2. Even though such approach does not produce any greenhouse gas as side-product, it is energy intensive and hence costly. Based on these reasons, the alternative approach for H2 production through the photocatalytic splitting of water reactions seems to be a more feasible option for the purpose of clean energy generation. 2 1.2 Mechanism of water splitting The main processes in photocatalytic water splitting consist of three steps: (1) light/photon absorption with energies larger than the bandgap of the semiconductor photocatalyst to generate electron-hole pairs, (2) charge separation followed by the diffusion of the photoinduced charge carriers to the surface of the photocatalyst, and finally (3) undergoing chemical reactions between the charge carriers and foreign compounds such as H2O on the surface of the photocatalyst.11,12 In the first step, the photocatalyst absorbs photons with energy greater than its energy bandgap in order to excite electrons from its valence band to the conduction band, while producing holes in the valence band at the same time. The photoexcited electron-hole pairs have to remain separated and the free charge carriers will then diffuse or migrate to the surface to undergo reactions. The electrons are responsible in reducing H2O to H2 whereas the holes would oxidize H2O to O2 via the following reactions:13 (7) (8) (9) (10) The overall water splitting reaction can thus be represented as: (11) Eq. (11) can also be explained as such: upon absorbing four photons, the photocatalyst is able to chemically split two H2O molecules to a single O2 molecule and two H2 molecules. There are several basic criteria that a particular photocatalyst has to fulfil in order to undergo the reaction of photocatalyst water splitting. First and foremost, the bottom of its conduction band must be positioned at a more negative potential than the reduction potential of H+ to H2, which is at 0 V vs normal hydrogen electrode (NHE) at pH = 0. On the other hand, the top of the valence band of the photocatalyst has to be positioned more positive than the oxidation potential of H2O to O2 with the potential positioned at 1.23 V vs 3 NHE. This implies that the energy bandgap of a photocatalyst has to have a minimum value of 1.23 eV which corresponds to the photon wavelength of approximately 1100 nm in the near infrared region, before the photocatalytic water splitting reaction can occur. The overall water splitting reaction, as simple as it may seem, is in fact a thermodynamically uphill reaction with a large positive change in the Gibbs free energy (ΔG˚) of +238 kJ/mol which essentially corresponds to 1.23 eV per electron transferred.14,15 As a result, there exist an activation barrier in the charge-transfer process between the photocatalyst and the water molecules. Thus, photon with energy greater than the bandgap value of the photocatalyst is normally necessary in order to enable and drive the overall photocatalytic water splitting process. Apart from requiring a suitable energy bandgap value and appropriate valence and conduction band positions, it is also crucial to ensure that defects in a photocatalyst are minimized as defects commonly act as electron-hole pair recombination 16,17 vacancies, centres. Some OH-related examples defects, 18 of defects charge include carriers-trapping oxygen grain boundaries,19,20 reduced species (e.g. Ti3+, Ta3+)21,22 as well as impurities present within the lattice system of photocatalysts.22,23 Nevertheless, it is possible to overcome the abovementioned challenges by modifying the chemical and physical structures of a photocatalyst appropriately in order to produce a highly efficient photocatalyst, which will be discussed in Section 1.3. 1.3 Methods in enhancing photocatalytic efficiency There are several commonly applied techniques to enhance the photocatalytic efficiency of a photocatalyst, such as through (1) sensitization, (2) morphology modification, (3) doping, (4) co-catalyst loading, (5) thermal annealing and (6) utilization of plasmonic nanoparticles. More details on the abovementioned techniques will be discussed from Sections 1.3.1 to 1.3.6. 1.3.1. Sensitization Sensitization is commonly applied to expand the light absorption range of a photocatalyst, especially towards the visible and infra-red regions.24 4 Organic materials such as 8-hydroxyquinoline,25 magnesium phthalocyanine (MgPc),24 ethylenediaminetetraacetic acid (EDTA)26 and Eosin Y27 are examples of sensitizing dyes. The function of such dyes is to facilitate the absorption of photons of longer wavelengths beyond the absorption range of some photocatalysts. Upon absorbing photons of appropriate wavelength, electrons will be excited from the highest occupied molecular orbital (HOMO) level to lowest unoccupied molecular orbital (LUMO) level of the dye before transferring to the host photocatalyst to undergo photocatalytic reactions.28 The photoexcited electrons will then be transferred to the conduction band of the photocatalyst and such process helps to increase the number of free electrons to undergo photocatalytic reactions. Apart from dye, semiconductor materials are also commonly loaded onto photocatalysts to function as sensitizers. For instance, semiconductor materials such as CuO,29 CdSe30 and CdS31 which usually has narrow energy bandgap values as compared to their host photocatalyst are usually used as a sensitizer. Similar to the dye sensitizers mentioned above, these semiconductor materials are able to absorb photons with shorter wavelength, typically in the visible light region due to their narrow energy bandgap values. As a result, when one of these materials is loaded onto a photocatalyst, it is able to utilize visible light to generate free charge carriers which will then be injected into its conduction band of its host photocatalyst, thus enhancing the photocatalytic efficiency by increasing the amount of free charge carriers available to undergo more photocatalytic reactions. 1.3.2 Morphology modification Morphology modification is one of the most studied methods for the enhancement in photocatalytic efficiency. Some of the morphology types used by photocatalysts in various photocatalytic applications are nanoparticle, nanorod, mesoporous structure, 2-dimensional planar structure as well as aerogel nanostructure in which the advantages of some of the morphology structures are summarized as below. First of all, nanoparticle structure is usually preferred over bulk particle type. This is because in nanoparticles the photoexcited electrons and holes would experience shorter distance migrating to the reaction sites on the 5 photocatalyst surface, thus minimizing the probability of electron-hole pair recombination.14 Besides, nanoparticles have highly crystalline structure for higher charge carrier mobility and less boundary defects acting as electronhole pair recombination centres.20,23,32,33 As for nanorod, such 1-dimensional nanostructure shows superior charge transport properties that could accelerate the photoactivity of the photocatalyst, hence leading to a higher photocatalytic efficiency.34,35 A common feature among the abovementioned morphologies is that they possess a larger surface area to volume ratio as compared to bulk particles.36,37 This allows higher photocatalytic activity as a result of higher photoexcitation rate of electron-hole pair and the enhancement in the light harvesting capability.38 Another advantage of a high surface area to volume ratio is the increase in the number of active sites available.39,40 For instance, C3N4, which can be used for photocatalytic H2 evolution, has a surface area of 10 m2 g-1 in its bulk particle form.41 On the other hand, C3N4 nanosheet has a significantly higher surface area of 84.2 m2 g-1 and such characteristic would naturally lead to higher photocatalytic water splitting rate, as reported by Wang, et al. and Chen et al..41,42 1.3.3 Doping Doping process is commonly employed to alter or modify the chemical structure of the photocatalyst to suit certain types of photocatalytic reactions. There are several examples of doping materials, with the cation dopants usually originating from transition metals such as Fe,43 Al44 and Cr45 and the commonly used anion dopants are N30,46 S,47,48 C,49 F44,50 and B.51 In many occasions, transition metal dopant species are used for various enhancement purposes such as to inhibit the recombination between the electron-hole pairs,52,53 to increase the minority carrier diffusion length54,55 as well as to extend the spectral response of the photocatalyst into visible region by inducing optical transitions from d electrons of the dopant metal to the conduction band of the photocatalyst.56,57 On the other hand, anionic dopant species are better used for bandgap narrowing to allow absorption of photons within the visible region.44,47,49,58 However, the doping effect may not necessarily prove to be satisfactory in some cases. For example, in certain 6 cation doping process, the metal ion dopants may introduce deep impurity levels within the forbidden bandgap energy levels of the photocatalyst. These impurity levels may act as recombination centers for the photoinduced electron and hole charge carriers, thus impairing the photocatalytic efficiency of the photocatalyst.59 Another example is the reduction in the hole diffusion length in TiO2 photocatalyst caused by certain doping species such as Cr.55 Hence, proper doping procedures including choice of dopant and suitable doping process are essential to ensure enhancement in the photocatalytic efficiency of the photocatalyst. 1.3.4 Co-catalyst loading There are several materials, be it metal or non-metal, that function as co-catalysts in order to improve the photocatalytic activity of photocatalysts. For example, reduced graphene oxide (rGO) is currently one of the most active materials under research to provide enhancement for photocatalytic activities such as photodegradation of organic compound60 as well as water splitting process.61,62 rGO has a unique structure and properties such as high electron mobility of 2.5 x 105 cm2 V-1 s-1 with a high surface area to volume ratio of 2630 m2 g-1.63,64 These properties make rGO an excellent material as an electron acceptor. By loading rGO onto photocatalysts such as TiO2,65,66 ZnO, 67 Cu2O68 and WO361 the photoinduced electrons could be transferred to rGO which then minimizes the recombination rate between the electron and hole pairs.69 Apart from rGO, noble metal nanoparticles such as Au, Ag and Pt are commonly used as electron-accepting co-catalysts due to the lower Fermi energy levels of these metals as compared to the conduction band edge potential of the photocatalyst.70-73 Besides electron-accepting co-catalysts, semiconductor nanoparticles such as RuOx, IrOx and CoOx nanoparticles are also commonly loaded on photocatalyst in order to act as a hole sink to scavenge for photoinduced holes due to a more energetically-favourable valence band.74 This is because photoinduced holes in the valence band edge of the photocatalyst would be transferred to the v 1.3.5 Thermal treatment 7 The presence of defect is almost unavoidable in most photocatalysts. Examples of defects such as oxygen vacancies,75,76 OH-related defects,77 reduced species (e.g. Ti3+, Ta3+),21,22 charge carriers-trapping grain boundaries18,19 as well as impurities in photocatalyst acting as electron-hole pair recombination centres.20 Under certain conditions, thermal annealing can be employed to reduce the defect density of a photocatalyst which would help to improve the photocatalytic efficiency.78 There are various types of atmosphere in which thermal treatment is usually carried out, for example the gaseous atmosphere of H2,21,79 O2,80,81 plasma,81-83 as well as inert environment83,84 or in a solution environment as in the case of hydrothermal treatment.36 Besides, thermal annealing could also lead to the increase in crystallinity of the photocatalyst which would then reduce grain boundaries and facilitate in the transfer of charge carriers, thus resulting in the enhancement of the photocatalytic efficiency.23,23,33,83 Besides, thermal annealing also helps to enhance bonding or interfacial contact between two different materials, such as between a co-catalyst and its host photocatalyst which would enable improved photocatalytic activities.85,86 1.3.6 Utilization of localized surface plasmon resonance (LSPR) effect The LSPR effect occurs due to the loading of certain types of plasmonic metal nanoparticles on the surface of a photocatalyst. Similar to the function of dye-sensitization, plasmonic metal nanoparticles such as Au,87 Ag,88 Cu89 and Pt90 are employed to harvest visible light. Such approach is especially effective in expanding the absorption range of photocatalysts with large energy bandgap level such as TiO2,91 ZnO,92 SrTiO393 and CrTaO4.94 These large-bandgap photocatalysts are only limited to absorption of UV light or photons with shorter wavelengths, thus rendering the photocatalysts less effective due to their inability to harvest majority of the solar spectrum, especially wavelengths in the visible region. In LSPR, the incident light spatially interacts with the surface of the photocatalyst and temporarily confines the electromagnetic waves. Such effect can be attained through the localization of the electromagnetic waves in an extremely small area of nanometer size region that exceeds the diffraction limit and thus confining the radiation within the space for a certain period.95 The localization of the near8 field light irradiation from the surface of the plasmonic metal nanoparticle stays until the phase relaxation of the LSPR before a large electromagnetic field enhancement is induced.96 As a result, the section of the photocatalyst surface within the vicinity of the plasmonic metal nanoparticle has a high probability of photoexcitation as the surface experiences a strong or enhanced electromagnetic field. The photoexcitation would result in the generation of electrons and holes in the metallic nanoparticle. The free electrons would then travel to the conduction band of the host photocatalyst and diffuse to the photocatalyst surface to undergo photocatalytic reaction. In short, the LSPR effect enables the photogeneration of electron-hole pairs through the interaction between visible light and the plasmonic nanoparticles-loaded photocatalyst. However, the size and morphology of the plasmonic metal nanoparticle play an influential role in affecting the particular electromagnetic or incident light wavelength that the plasmonic metal nanoparticle may respond to. For instance, Au nanoparticle of 5 nm in diameter has a plasmon resonance wavelength of 532 nm97 whereas the plasmon resonance of a Ag nanorod of 200 nm in length is at about 420 nm.98 1.4 Z-Scheme As mentioned in section 1.2, one of the criteria for a successful overall photocatalytic water splitting reaction to occur is the bandgap of the photocatalyst has to be sufficiently large (>1.23 eV). Besides, the bottom of the conduction band of a photocatalyst has to be more negative in potential than the reduction potential of H+ to H2 (at 0 V vs NHE) whereas the top of its valence band has to be more positive than the oxidation potential of H2O to O2 (at +1.23 V vs NHE). However, only a handful of photocatalysts fulfil the abovementioned criteria, such as TiO2, ZnO and Ta2O5. Unfortunately, these photocatalysts have a relatively large bandgap (Eg (TiO2) ≈ 3.2 eV,99 Eg (ZnO) ≈ 3.2 eV,99 Eg (Ta2O5) ≈ 4.1eV100) which only absorb light in the UV region (λ < 400 nm). Other commonly used semiconductor photocatalysts, for example WO3 (Eg ≈ 2.8 eV),99,100 and CdSe (Eg ≈ 1.7 eV,14,30) have a narrower bandgap which allows for the absorption of visible light. However, these photocatalysts lack appropriate conduction or valence band energy level position which 9 allows them to perform either the photoreduction process of H2O to H2 or photooxidizing H2O to O2. For WO3, the bottom of its conduction band and the top of its valence band are positioned at +0.5 and +0.32 V vs NHE at pH = 0, respectively. Since the conduction band potentials are more positive than the reduction potential of H+ to H2 which is at 0 V vs NHE, WO3 is not a suitable H2-generating photocatalyst.101 On the other hand, WO3 is suitable for the photogeneration of O2 due to the fact that the oxidation potential of H2O to O2 is at 1.23 V vs NHE. As for CdSe, the bottom of its conduction band is located at approximately -0.8 V with the valence band top at +0.9 V (vs NHE at pH = 0).14,102 As a result, CdSe photocatalyst is usually used for the photocatalytic H2 evolution reaction. Figure 1.1 below shows the conduction and valence band edge positions of several photocatalysts: Potential / vs NHE (pH = 0) -2 CdS 3 0 (H /H2) TiO2 WO3 Fe2O3 3.0 eV 3.2 eV 3.4 eV 5.0 eV +3 2.4 eV + +1.23 +1 (H2O/O2) +2 +4 CdS 2.8 eV KTaO3SrTiO 1.7 eV -1 2.3 eV ZrO2 Conduction band edge Valence band edge Figure 1.1 Conduction and valence band edge positions of several photocatalysts. Apart from requiring a photocatalyst with suitable bandgap value together with appropriate conduction and valence band positions, it is also crucial that the free electron and hole charge carriers do not get trapped by 10 defects or recombine with each other. Defects can normally be repaired or minimized through thermal treatment, as explained in section 1.3.5. As for the issue of electron-hole pair recombination, such problem can be reduced by employing the use of sacrificial agent to scavenge for one of the two charge carrier types (either electron or hole), thus preventing it from recombining with its counterpart. As a matter of fact, for the majority of photocatalytic water splitting research, be it H2 or O2 evolution process, the use of sacrificial agent is a common practice. For example, methanol103,104 and ethanol105 are the two most common hole-scavenging sacrificial agents, apart from Ce3+,106 Fe2+,107 I-108,109 and S2-.13,30 As for the scavenging of electrons, usually Ce4+,106 Fe3+ 61,107,110 or IO3- 109,110 ions are used. In order to solve the abovementioned challenges, a strategy commonly employed is the application of a two-step photoexcitation mechanism between two different photocatalysts in order to achieve overall water splitting process known as Z-scheme. In fact, Z-scheme was inspired by photosynthesis, a process that occurs naturally in green plants and several other organisms to convert light energy into chemical energy to fuel the organisms’ metabolic activities.111 With the Z-scheme system, it is possible to use two different photocatalysts with each only requiring the capability of producing either H2 or O2. As a result, photocatalysts employed in the Z-scheme system do not require a bandgap that covers both the water reduction and oxidation potentials, but one that covers either one of the two potentials. Another crucial component of Z-scheme is the redox mediators, comprising of a reductant and oxidant such as Fe3+/Fe2+ and IO3-/I –.112 The primary role of the redox mediators is to link the two photocatalysts by shuttling electrons and holes between the photocatalysts, thus enabling the full photocatalytic water splitting reaction via Z-scheme. However, the redox potential of the electron acceptor is necessary to be more positive than the conduction band edge of the O2-production photocatalyst whereas the redox potential of the electron donor is required to be more negative than the valence band edge of the H2producing photocatalyst in order for both the redox mediators to function as efficient electron and hole acceptors. Recently Iwase et al. proposed the use of rGO as a solid-state mediator in Z-scheme photocatalytic water splitting.10 The high electron mobility property of rGO enables an effective and efficient 11 transferring mechanism of the photogenerated electrons from the O2producing photocatalyst to the H2-producing photocatalyst, thus contributing to an enhanced photoactivity. However, the research in the use of rGO as a solid-state mediator is still in a relatively early stage and more experiments have to be conducted to prove its suitability as a Z-scheme charge mediator. As for a brief overview of the Z-scheme process, the photocatalyst responsible for H2 evolution absorbs photon and generate electron-hole pairs as a result. The free electrons in the conduction band would then photoreduce H2O to H2, with the free holes in the valence band oxidizing the reductants (Fe2+ or I-) to oxidants (Fe3+ or IO3-). On the contrary, the holes photoexcited by the O2-generating photocatalyst would oxidize the H2O molecules to O2, whereas the electrons would react with the oxidants (Fe3+ or IO3-) and reduce them to reductants (Fe2+ or I-). The design of such photocatalytic system reduces the energy required to undergo overall water splitting reaction, which enables the photoreduction and phooxidation of H2O to H2 and O2 simultaneously. In other words, photocatalysts with small bandgap values that only allows either H2 or O2 generation can be used in Z-scheme, thus allowing a wider range of visible light to be utilized more efficiently due to less amount of energy needed to run each of the photocatalyst. The schematic representation of the photocatalytic reactions involved in Z-scheme is depicted in Figure 1.2. 12 Potential / vs NHE (pH = 0) - CB 0 + (H /H2) e - e H2 + - CB Red/Ox e H Reductant - e - e Oxidant H2O +1.23 (H2O/O2) - e O2 + h + h VB H2-producing photocatalyst VB Conduction band edge Valence band edge O2-producing photocatalyst Figure 1.2 Photocatalytic reactions involved in the Z-scheme system. The following reactions represent the charge transfer processes involved in a typical Z-scheme system for the photocatalytic water splitting reaction using Fe3+/Fe2+ as the redox mediators: At H2-producing photocatalyst: (9) (10) At O2-producing photocatalyst: (11) (12) Despite the many advantages of Z-scheme process in photocatalytic water splitting reaction, it also suffers from several drawbacks. First of all, the co-existence of H2 and O2 gases produced via Z-scheme system may result in 13 the two gases recombining to form H2O. Secondly, the reversibility of the redox mediators may result in the backward reactions of the redox mediators to proceed more readily, thus suppressing the H2 and O2 evolution reactions. For example, at the H2-producing photocatalyst, reduction of Fe3+ to Fe2+, which is a thermodynamically favourable process may proceed in preference over the reduction of H+ to H2. The same goes to O2-producing photocatalyst whereby the oxidation of Fe2+ to Fe3+ may readily proceed over the oxidation of H2O to O2. Thirdly, the redox mediators may also react with the evolved H2 and O2 gases, thus promoting backward reactions represented as: (13) (14) As shown in Eq. (13) and (14), it is possible for the redox mediator Fe3+ ions to revert H2 gas trapped in the solution to H+. Similarly, there is a tendency of Fe2+ ions, together with H+ ions to react with O2 gas to form H2O.113 Fourthly, the concentration of the redox mediators also plays a crucial role in determining the success of the Z-scheme system. Unsuitable amount of either the oxidant or reductant may affect the photocatalytic H2 and O2 rate by the photocatalysts.13,107,114,115 Kato et al.116 observed that the H2 evolution rate of the photocatalyst increased when the Fe3+ ion concentration is higher than the concentration of Fe2+ ion. This could be due to the suppression of the backward reaction between H2 and Fe3+ as shown in Eq. (13). Due to these problems, it is difficult to build an efficient Z-scheme system which could demonstrate a highly efficient simultaneous evolution of H2 and O2 process. Hence, it is critical that challenges should be overcome or to have their effects minimized in order to optimize the H2 and O2 evolution efficiency. 1.5 TaON and WO3 as Z-scheme photocatalysts There are several candidates commonly used as Z-scheme photocatalysts, among them are SrTiO3 (Cr–Ta-doped),93 anatase TiO2,108 CaTiO3 (Cr–Ta-doped) and CrTaO494 which are used as H2-producing photocatalyst whereas BiVO4,108,117 rutile TiO2,108 In2O3,94 Bi2O3,94 and Fe2O394 photocatalysts are known for their O2-producing capability. Apart 14 from these photocatalysts, TaON and WO3 are also widely used in the Zscheme system.118,119 TaON is frequently used as a H2-producing photocatalyst because it has a suitable conduction edge position at -0.3 V (vs NHE at pH = 0), which is more negative than the reduction potential of H+ to H2 at 0 V (vs NHE at pH = 0). Such characteristic would allow the reduction of H+ ions by the photoexcited electrons to H2.118,120 Furthermore, TaON has a sufficiently small bandgap with a value of 2.5 eV which enables visible light response.118,120 Apart from that, it is relatively stable in aqueous solution with negligible rate of nitrogen anion self-oxidative decomposition to N2 by the ) during photocatalytic reaction.120- photogenerated holes ( 122 On the other hand, WO3 photocatalyst is often used for O2 evolution reaction. This is because its valence band edge can provide enough potential for O2 production since it is located at approximately +3.0 V vs NHE at pH = 0.123,124 Similar to TaON, WO3 also has a high chemical stability in aqueous solution under O2 evolving conditions.125 Due to their suitability as a Z-scheme photocatalyst, numerous modifications have had been performed on TaON and WO3 in order to enhance their photoactivity. For example, TaON has had been loaded with several types of semiconductor nanoparticle such as CoOx and IrO2 which serve as hole-scavenging co-catalysts.126,127 Effects of metallic nanoparticle loading such as Pt and Ag acting as electron sink to TaON have also been investigated.118,128,129 Besides that, the loading of CaFe2O4 as well as CdS with conduction and valence band potentials which are more negative than those of TaON has also led to the enhancement in the separation of photogenerated electron-hole pairs.31,121 These two materials also have a narrower bandgap (CaFe2O4: 1.9 eV, CdS: 2.45 eV) which allows wider absorption range of the visible light spectrum and such characteristic enables them to act as a sensitizer.31,121 Morphology modification is also another common approach used to enhance the photocatalytic efficiency of TaON. For example, TaON in the form of urchin-like hierarchical nanostructures,130 nanotube131 and nanoparticle,132 have been reported to show higher photocatalytic efficiency as compared to its bulk particle form. As for WO3, some of the reported modifications made on this photocatalyst are such as morphology alteration in the form of mesoporous81,133,134, nanorod,135 nanobelt and nanoplatelet 15 structures,136 as well as nanoparticle loading such as Pt and RuO2 as electron and hole-scavenging co-catalysts, respectively.137 Besides, the effect of the loading of materials with narrow bandgap such as g-C3N4 and Co3O4 as sensitizers on WO3 have also been investigated and found to improve the photcatalytic performance of WO3.138,139 In the coming chapters, the two Z-scheme photocatalysts will be subjected to further discussion. Specifically, chapter 2 will touch on an alternative approach for the synthesis of TaON. In chapter 3, the synergetic effect and the disadvantages of the loading of CuO nanoparticles as sensitizer on WO3 towards the composite photoactalyst’s photocatalytic O2 evolution efficiency will be discussed. Finally, a AgCl-Ag core-shell structure in the nanoparticle form loaded on WO3 functioning as an electron-scavenging cocatalyst will be introduced as well as studying its influence in the O2 evolution efficiency of the WO3 photocatalyst. References [1] M. Steinberg, Int. J. Hydrogen Energ., 1999, 24, 771. [2] A. Zecca and L. Chiari, Energy Policy, 2010, 38, 1 [3] A. Fujishima and K. Honda, Nature, 1972, 238, 37. [4] F. Haas, A. Jain, J. Lehmann, O. Luschtinetz and R. Schefler, Int. J. Hydrogen Energ., 2005, 30, 555. [5] D. Hotza and J.C. Diniz da Costa, Int. J. Hydrogen Energ., 2008, 33, 4915. [6] B. D. Solomon and A. Banerjee, Energy Policy, 2006, 34, 781. [7] J. Guan, Q. Wang, X. Li, Z. Luo and K. Cen, Renew. Energ., 2007, 32, 2502. [8] A.C.C. Souza, J. Luz-Silveira and M.I. Sosa, J. Fuel Cell Sci. Technol., 2006, 3, 346. [9] C. Song, Catal. Today, 2002, 77, 17. [10] A. Iwase, Y. H. Ng, Y. Ishiguro, A. Kudo and R. Amal, J. Am. Chem. Soc., 2011, 133, 11054. [11] A. Kudo, H. Kato and I. Tsuji, Chem. Lett., 2004, 33, 1532. [12] R. Abe, J. Photochem. Photobiol. C, 2010, 11, 179. [13] A. Galińska and J. Walendziewski, Energy Fuels, 2005, 19, 1143. 16 [14] A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253. [15] M. R. Pai, A. M. Banerjee, A. K. Tripathi and S. R. Bharadwaj, Functional Materials: Preparation, Processing and Applications, 2012, 579. [16] M. B. Radoicic, I. A. Jankovic, V. N. Despotovic, D. V. Sojic, T. D. Savic, Z. V. Saponjic, B. F. Abramovic and M. I. Comor, Appl. Catal., B, 2013, 138-139, 122. [17] H. Irie, Y. Watanabe and K. Hashimoto, J. Phys. Chem. B, 2003, 107, 5483. [18] C. Pan, J. Xu, Y. Chen and Y. Zhu, Appl. Catal., B, 2012, 115 – 116, 314. [19] M. R. Bayati, J. Ding, Y. F. Lee, R. J. Narayan, J. Narayan, H. Zhou and S. J. Pennycook, J. Phys.: Condens. Matter, 2012, 24, 395005. [20] Y. Li, J. Du, S. Peng, D. Xie, G. Lu and S. Li, Int. J. Hydrogen Energ., 2008, 33, 2007. [21] H. Liu, H. T. Ma, X. Z. Li, W. Z. Li, M. Wu and X.H. Bao, Chemosphere, 2003, 50, 39. [22] Y. Moriya, T. Takata and K. Domen, Coord. Chem. Rev., 2013, 257, 1957. [23] H. Jeong, T. Kim, D. Kim and K. Kim, Int J. Hydrogen Energ., 2006, 31, 1142. [24] K. Takanabe, K. Kamata, X. Wang, M. Antonietti, J. Kubota and K. Domen, Phys. Chem. Chem. Phys., 2010, 12, 13020. [25] V. H. Houlding and M. Gräzel, J. Am. Chem. Soc., 1983, 105, 5695. [26] R. Abe, K. Sayama and H. Arakawa, J. Photochem. Photobiol., A, 2004, 166, 115. [27] R. Abe, K. Hara, K. Sayama, K. Domen and H. Arakawa, J. Photochem. Photobiol. A, 200, 137, 63. [28] Y. Wang, J. Hong, W. Zhang and R. Xu, Catal. Sci. Technol., 2013, 3, 1703. [29] N. Helaïli, Y. Bessekhouad, A. Bouguelia and M. Trari, J. Hazard. Mater., 2009, 168, 484. [30] J. Hensel, G. Wang, Y. Li and J. Z. Zhang, Nano Lett., 2010, 10, 478. [31] J. Hou, Z. Wang, W. Kan, S. Jiao, H. Zhu and R. V. Kumar, J. Mater. Chem., 2012, 22, 7291. [32] F. Amano, O. O. Prieto-Mahaney, Y. Terada, T. Yasumoto, T. Shibayama and B. Ohtani, Chem. Mater., 2009, 21, 2601. [33] Y. Bu, Z. Chen, W. Li and J. Yu, ACS Appl. Mater. Interfaces, 2013, 5, 5097. [34] I. S. Cho, Z. B. Chen, A. J. Forman, D. R. Kim, P. M. Rao, T. F. Jaramillo and X. L. Zheng, Nano Lett., 2011, 11, 4978. 17 [35] H. Wang, X. Quan, Y. B. Zhang and S. Chen, Nanotechnology, 2008, 19, 065704. [36] H. T. Chang, N. M. Wu and F. Zhu, Water Res., 2000, 34, 407. [37] M. I. Arabatzis and P. Falaras, Nano Lett., 2003, 3, 249. [38] J. H. Pan, Z. Cai, Y. Yu and X. S. Zhao, J. Mater. Chem., 2011, 21, 11430. [39] P. Niu, L. Zhang, G. Liu and H.M. Cheng, Adv. Funct. Mater., 2012, 22, 4763. [40] N. Bao, L. Shen, T. Takata and K. Domen, Chem. Mater., 2008, 20, 110. [41] X. Chen, Y. S. Jun, K. Takanabe, K. Maeda, K. Domen, X. Fu, M. Antonietti and X. Wang, Chem. Mater., 2009, 21, 4093. [42] X. Wang, K. Maeda, X. Chen, K. Takanabe, K. Domen, Y. D. Hou, X. Fu and M. Antonietti, J. Am. Chem. Soc., 2009, 131, 1680. [43] A. Sclafani, L. Palmisano and M. Schiavello, Res. Chem. Intermed., 1992, 18, 211. [44] J. C. Yu, J. Yu,W. Ho, Z. Jiang and L. Zhang, Chem. Mater., 2002, 14, 3808. [45] E. Borgarello, J. Kiwi, M. Gratzel, E. Pelizzetti and M. Viscald, J. Am. Chem. Soc., 1982, 104, 2996. [46] S. U. M. Khan, M. Al-Shahry and W. B. Ingler Jr., Science, 2002, 297, 2243. [47] T. Umebayashi, T. Yamaki, H. Itoh, and K. Asai, Appl. Phys. Lett., 2002, 81, 454. [48] G. Liu, P. Niu, C. Sun, S. C. Smith, Z. Chen, G. Q. Lu and H. M. Cheng, J. Am. Chem. Soc., 2010, 132, 11642. [49] N. T. Q. Hoa, Z. Lee and E. Kim, J. Electrochem. Soc., 2012, 159, K42. [50] Y. Wang, Y. Di, M. Antonietti, H. Li, X. Chen and X. Wang, Chem. Mater., 2010, 22, 5119. [51] S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2010, 26, 3894. [52] G. N. Schrauzer and T. D. Guth, J. Am. Chem. Soc., 1977, 99, 7189. [53] J. Soria, J. C. Conesa, V. Augugliaro, L. Palmisano, M. Schiavello and A. Sclafani, J. Phys. Chem., 1991, 95, 274. [54] J. Augustynski, J. Hinden and C. Stalder, J. Electrochem. Soc., 1977, 124, 1063. [55] A. Monnier and J. Augustynski, Ibid., 1980, 127, 1576. [56] A. Mackor and G. Blasse, Chem. Phys. Lett., 1981, 77, 6 [57] H. P. Maruska and A. K. Ghosh, Sol. Energy Mater., 1979, 1, 237. [58] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 18 2001, 293, 269. [59] J.G. Highfield and P. Pichat, New. J. Chem., 1989, 13, 61. [60] P. Wang, J. Wang, T. Ming, X. Wang, H. Yu, J. Yu, Y. Wang and M. Lei, ACS Appl. Mater. Interfaces, 2013, 5, 2924. [61] J. Guo, Y. Li, S. Zhu, Z. Chen, Q. Liu, D. Zhang, W. Moon and D. Song, RSC Adv., 2012, 2, 1356. [62] J. Lin, P. Hu, Y. Zhang, M. Fan, Z. He, C. K. Ngaw, J. S. C. Loo, D. Liao and T. T. Y. Tan, RSC Adv., 2013,3, 9330. [63] K. Novoselov, Rev. Mod. Phys., 2011, 83, 837. [64] M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010, 110, 132. [65] W. Fan, W. Lai , Q. Zhang and Y. Wang, J. Phys. Chem. C, 2011, 115, 10694. [66] B. Wang, Q. Sun, S. Liu and Y. Li, Int. J. Hydrogen Energy, 2013, 38, 7232. [67] T. Xu, L. Zhang, H. Cheng and Y. Zhu, Appl. Catal., B, 2011, 101, 382. [68] P. D. Tran, S. K. Batabyal, S. S. Pramana, J. Barber, L. H. Wong and S. C. J. Loo, Nanoscale, 2012, 4, 3875. [69] H. Wu, M. Xu, P. Da, W. Li, D. Jia and G. Zheng, Phys. Chem. Chem. Phys., 2013. [70] M. Qamar, Z.H. Yamani, M.A. Gondal and K. Alhooshani, Solid State Sci., 2011, 13, 1748. [71] C. Ren, B. Yang, M. Wu, J. Xu, Z. Fu, Y. lv, T. Guo, Y. Zhao and C. Zhu, J. Hazard. Mater., 2010, 182, 123. [72] J. Yu, L. Qi and M. Jaroniec, J. Phys. Chem. C, 2010, 114, 13118. [73] S. Sato and J.M. White, Chem. Phys. Lett., 1980, 72, 83. [74] J. Yang, D. Wang, H. Han and C. Li, Acc. Chem. Res., Article ASAP. [75] T. Sameshima, K. Sakamoto, K. Asada, M. Kondo, A. Matsuda and S. Higashi, Sol. Energ. Mat. Sol., 2001, 65, 577. [76] X. Zhang, H. Li, X. Cui and Y. Lin, J. Mater. Chem., 2010, 20, 2801. [77] J. Wang, D. N. Tafen, J. P. Lewis, Z. Hong, A. Manivannan, M. Zhi, M. Li and N. Wu, J. Am. Chem. Soc., 2009, 131, 12290. [78] N. Wu, M. Lee, Z. Pon and J. Hsu, J. Photochem. Photobiol. A: Chem., 2004, 163, 277. [79] T. Huang and T. Yu, Appl. Catal., 1991, 71, 275. [80] S. Poulston, P. M. Parlett, P. Stone and M. Bowker, Surf. Interface Anal., 1996, 24, 811. [81] M. de Respinis, G. De Temmerman, I. Tanyeli, M. C. M. van de Sanden, R. P. Doerner, M. J. Baldwin and R. van de Krol, ACS Appl. Mater. Interfaces, Article ASAP. 19 [82] J. Han, X. Wang, N. Wang, Z. Wei, G. Yu, Z. Zhou and Q. Wang, Surf. Coat. Technol., 2006, 200, 4876. [83] H. Hsi and C. Tsai, Chem. Eng. J., 2012, 191, 378. [84] A. Sarkar, A. Shchukarev, A. Leino, K. Kordas, J. Mikkola, P. O. Petrov, E. S. Tuchina, A. P. Popov, M. E. Darvin, M. C. Meinke, J. Lademann and V. V. Tuchin, Nanotechnology, 2012, 23, 475711. [85] Z. Liu, S. Chai, A. Binder, Y. Li, L. Ji and S. Dai, Appl. Catal., A, 2013, 451. 282. [86] S. S. Kalanur, S. H. Lee, Y. J. Hwang and O. Joo, J. Photochem. Photobiol. A, 2013, 259, 1. [87] S. Palomba, M. Danckwerts and L. Novotny, J. Opt. A-Pure Appl. Op., 2009, 11, 114030. [88] J. Jiang, H. Li and L. Zhang, Chem. Eur. J., 2012, 18, 6360. [89] D. B. Ingram and S. Linic, J. Am. Chem. Soc., 2011, 133, 5202. [90] T. Ahmadi, Z. L. Wang, T. C. Green, A. Henglein and M. El-Sayed, Science, 1996, 272, 1924. [91] A. K.P.D. Savioa, D. Starikovb, A. Bensaoulab, R. Pillaib, L. L. de la Torre Garcíad and F. C. Robles Hernández, Ceram. Int., 2012, 38, 3529. [92] C. Yang, X. M. Li, Y. F. Gu, W. D. Yu, X. D. Gao, and Y. W. Zhang, Appl. Phys. Lett., 2008, 93, 112114. [93] K. Sayama, K. Mukasa, R. Abe, Y. Abe and H. Arakawa, Chem. Commun., 2001, 2416. [94] K. Sayama, K. Mukasa, R. Abe, Y. Abe and H. Arakawa, J. Photochem. Photobiol. C, 2002, 148, 71. [95] K. Uenoa and H. Misawa, J. Photochem. Photobiol. C, 2013, 15, 31. [96] K. Ueno and H. Misawa, J. Photochem. Photobiol. A, 2011, 221, 130. [97] W. Hou, W. H. Hung, P. Pavaskar, A. Goeppert, M. Aykol and S. B. Cronin, ACS Catal., 2011, 1, 929. [98] H. Wei, A.Reyes-Coronado, P. Nordlander, J. Aizpurua and H. Xu, ACS Nano, 2010, 4, 2649. [99] J. Tao, T. Luttrell and M. Batzill, Nature Chem., 2011, 3, 296. [100] N. Ndiege, R. Chandrasekharan, A. D. Radadia, W. Harris, E. Mintz, R. I. Masel and M. A. Shannon, Chem. Eur. J., 2011, 17, 7685. [101] D.E. Scaife, Solar Energy, 1980, 25, 41. [102] C. J. Wang, M. Shim and P. Guyot-sionnest, Science, 2001, 291, 2390. [103] C. Santato, M. Ulmann and J. Augustynski, Adv. Mater., 2001, 13, 511. [104] C. Santato, M. Odziemkowski, M. Ulmann and J. Augustynski, J. Am. 20 Chem. Soc., 2001, 123, 10639. [105] A. Pérez-Lariosa, R. Lopez, A. Hernández-Gordillo, F. Tzompantzi, R. Gómez and L.M. Torres-Guerra, Fuel, 2012, 100, 139. [106] G. R. Bamwenda, T. Uesigi, Y. Abe, K. Sayama and H. Arakawa, Appl. Catal., A, 2001, 205, 117 [107] K. Maeda, M. Higashi, D. Lu, R. Abe and K. Domen, J. Am. Chem. Soc., 2010, 132, 5858. [108] R. Abe, K. Sayama and H. Sugihara, J. Phys. Chem. B, 2005, 109, 16052. [109] M. Tabata, K. Maeda, M. Higashi, D. Lu, T. Takata, R. Abe and K. Domen, Langmuir, 2010, 26, 9161. [110] H. Arakawa and K. Sayawa, Res. Chem. Intermed., 2000, 26, 145. [111] C. L. Beadle and S.P. Long, Biomass, 1985, 8, 119. [112] K. Yoshihara and S. Kumazaki, J. Photochem. Photobiol. C, 2000, 1, 22. [113] Y. Sasaki, A. Iwase, H. Katoa and A. Kudo, J. Catal., 2008, 259, 133. [114] P. Chen and H. Y. He, Res. Chem. Intermed., 2011, 37, 1057. [115] G. R. Bamwenda, K. Sayama and H. Arakawa, J. Photochem. Photobiol., A, 1999, 122, 175. [116] H. Kato, Y. Sasaki, A. Iwase and A. Kudo, Bull. Chem. Soc. Jpn., 2007, 80, 2457. [117] Y. Park, K. J. McDonald and K. Choi, Chem. Soc. Rev., 2013, 42, 2321. [118] R. Abe, M. Higashi and K. Domen, J. Am. Chem. Soc., 2010, 132, 11828. [119] R. Abe, T. Takata, H. Sugihara and K. Domen, Chem. Commun., 2005, 3829. [120] G. Hitoki, T. Takata, J. N. Kondo, M. Hara, H. Kobayashi and K. Domen, Chem. Commun., 2002, 16, 1698. [121] E. S. Kim, N. Nishimura, G. Magesh, J. Y. Kim, J.k Jang, H. Jun, J. Kubota, K. Domen and J. S. Lee, J. Am. Chem. Soc., 2013, 135, 5375. [122] T. Takata, G. Hitoki, J. N. Kondo, M. Hara, H. Kobayashi and K. Domen, Res. Chem. Intermed., 2007, 33, 13. [123] R. van de Krol, Y. Q. Liang and J. Schoonman, J. Mater. Chem., 2008, 18, 2311. [124] F. Amano, M. Tian, G. S. Wu, B. Ohtani and A. C. Chen, ACS Appl. Mater. Interfaces, 2011, 3, 4047. [125] V. Cristino, S. Caramori, R. Argazzi, L. Meda, G. L. Marra and C. A. Bignozzi, Langmuir, 2011, 27, 7276. [126] M. Higashi, K. Domen and R. Abe, J. Am. Chem. Soc., 2012, 134, 21 6968. [127] K. Maeda, D. Lu and K. Domen, Chem. Eur. J., 2013, 19, 4986. [128] K. Sekizawa, K. Maeda, K. Domen, K. Koike and O. Ishitani, J. Am. Chem. Soc., 2013, 135, 4596. [129] J.H. Hsieh, C.C. Chang, Y.K. Chang and J.S. Cherng, Thin Solid Films, 2010, 518, 7263. [130] Z. Wang, J. Hou, C. Yang, S. Jiao, K. Huang and H. Zhu, Environ. Sci., 2013, 6, 2134. [131] S. Banerjee, S. K. Mohapatra and M. Misra, Chem. Commun., 2009, 46, 7137. [132] Q. Gao, S. Wang, Y. Ma, Y. Tang, C. Giordano and M. Antonietti, Angew. Chem. Int. Ed., 2012, 51, 961. [133] L. Li, M. Krissanasaeranee, S. W. Pattinson, M. Stefik, U. Wiesner, U. Steiner and D. Eder, Chem. Commun., 2010, 46, 7620. [134] Z. Wen, W. Wu, Z. Liu, H. Zhang, J. Li and J. Chen, Phys. Chem. Chem. Phys., 2013, 15, 6773. [135] S. S. Kalanur, Y. J. Hwang, S. Y. Chae and O. S. Joo, J. Mater. Chem. A, 2013, 1, 3479. [136] L. Xien, W. F. Ying and W. Qing, Phys. Chem. Chem. Phys., 2012, 14, 7894. [137] S. S. K. Ma, K. Maeda, R. Abec and K. Domen, Energy Environ. Sci., 2012, 5, 8390. [138] Y. Zang, L. Li, Y. Zuo, H. Lin, G. Li and X. Guan, RSC Advances, 2013, 3, 13646. [139] A. K. Chakraborty, M. S. Akter, M. A. Haque, G. M. A. Khan and M. S. Alam, J. Clust. Sci., 2013. 22 Chapter 2 Synthesis of TaON via urea route for the photocatalytic reduction of water Abstract In this chapter, a novel approach for the synthesis of TaON photocatalyst for the photocatalytic reduction of H2O to generate H2 gas in the visible light region will be discussed. Instead of the conventional nitridation method which uses NH3 as the nitrogen source, the synthesis of TaON via the urea route method employs the use of urea as the nitrogen source for the alloying of Ta2O5 to TaON. The effect of pre- and post-synthesis calcination on the photocatalyst will also be studied and discussed in this chapter. 2.1 Introduction The issue of large bandgap has always been a bottleneck to photocatalytic water splitting reactions due to its inability to utilize the visible light section of the solar spectrum to generate charge carriers. Several of the commonly used semiconductor photocatalysts such as TiO2 (3.0 eV for anatase and 3.75 eV for rutile form)1 and ZnO (3.37 eV)2 have a wide bandgap and such undesirable effect is limiting their photocatalytic performances. However, there are several methods to overcome such limitation, such as through dye-sensitization,3-6 incorporation of nitrogen into photocatalyst7,8 and utilization of surface plasmonic resonance effect by loading noble metals such as Au, Ag, Cu and Pt.9-12 The introduction of nitrogen in photocatalyst such as NbO3,13 Bi2WO6 ,14 ZnO15 as well as TiO216,17 helps to increase the upper absorption limit of the photocatalyst from UV to visible region. In this chapter a particular type of photocatalyst, namely tantalum oxynitride (TaON) will be the center of discussion. TaON is a nitrogen-doped photocatalyst which has a visible absorption range of up to 500 nm.18 The process of nitrogen incorporation is to primarily reduce the bandgap of its Ta2O5 precursor which has a bandgap of 3.9 eV to approximately 2.5 eV.7,19 The reason for such bandgap reduction is because of the introduction of nitrogen to the anionic (oxygen) network of Ta2O5,8 thus resulting in new N2p 23 atomic orbitals with a higher potential energy relative to the O2p atomic orbital into the Ta2O5. Consequently, new orbitals with a higher bound state energy are generated, hence leading to a decrease in the bandgap of Ta2O5 upon introducing nitrogen which could then lead to the absorption of visible light by the nitrogen-doped Ta2O5 to generate electron-hole pairs.18,20,21 Despite the decrease, the bandgap of TaON is still sufficiently large to achieve overall water splitting, namely the oxidation and reduction of water to oxygen and hydrogen gasses, respectively. TaON has a conduction and valence band edge potentials of -0.3 and +2.2 V, respectively (vs NHE at pH = 0).22 Such characteristics are suitable for the photoreduction of H+ to H2 with a reduction potential of 0 V, as well as photooxidation of H2O to O2 with an oxidation potential of 1.23 V.23 Apart from having appropriate bandgap values and suitable conduction and valence band edge positions for photocatalytic water splitting reactions, TaON is also non-toxic and relatively stable during photooxidation and photoreduction of water.7 Apart from photocatalytic water splitting reactions, TaON has also been reported to have the capability of photodegrading organic contaminants such as Rhodamine B (RhB) and atrazine,24-26 which further demonstrates the versatility of TaON as a photocatalyst. Conventionally, TaON is prepared through the nitridation of Ta2O5 precursor in NH3 flow at high temperature for several hours.19,22,27-29 However, the drawbacks of such method are the consumption of hazardous NH3 gas as well as long duration of energy-intensive, high-temperature annealing. In the chapter, a novel approach in synthesizing TaON of bulk particle form without the need for NH3 gas as the source for introducing nitrogen into the Ta2O5 precursor will be discussed. Instead, nitrogen-rich urea is used as the primary nitrogen source. Another advantage of such approach over the conventional method is that it does not require long hours of high-temperature annealing. It should be noted that this work does not represent the pioneer use of urea as the nitrogen source. As a matter of fact, several other photocatalysts such as TiO2,30 ZnO31 and Ta2O532 have been doped with nitrogen via the urea route. However, the synthesis method for TaON via the urea route as reported by Gao et al.32 differs from the approach reported in this work and has not been tested for its photocatalytic reduction of water. Apart from reporting a 24 novel approach for the synthesis of TaON, we also attempted to investigate the effect of calcination in enhancing the TaON photocatalytic performance such as the duration of vacuum calcination during the synthesis of TaON, the effect of post-synthesis calcination of TaON in air and Ar atmosphere as well as calcinating the Ta2O5 precursor in air and Ar environment prior to the synthesis of TaON. 2.2 Experimental procedures 2.2.1 Synthesis conditions of TaON samples In this work the materials used for the synthesis of TaON samples are Ta2O5 (Inframat Advanced Materials, 99.99%) and urea (Sigma-Aldrich, 99.99%). Both chemicals were used as-purchased without further purification. The synthesis of TaON photocatalyst via the urea route basically involves the thorough mixing of 0.6 g urea with 0.2 g Ta2O5 powder followed by grinding using a mortar and pestle set prior to vacuum calcination. A total of seven different TaON samples were prepared from various synthesis conditions, which will be explained as followings: 25 Table 2.1 Synthesis conditions of samples T1 to T7. Sample Synthesis method T1 Powder mixture of urea and Ta2O5 precursor were subjected to vacuum calcination in a tube furnace at 725 ˚C followed by increasing the furnace temperature to 950 ˚C for a duration of 20 min during each annealing temperature, followed by natural cooling of the furnace to room temperature in vacuum. T2 Powder mixture of urea and Ta2O5 precursor were subjected to vacuum calcination in a tube furnace at 725 ˚C followed by increasing the furnace temperature to 950 ˚C for a duration of 60 min during each annealing temperature, followed by natural cooling of the furnace to room temperature in vacuum. T3 Powder mixture of urea and Ta2O5 precursor were subjected to vacuum calcination in a tube furnace at 725 ˚C followed by increasing the furnace temperature to 950 ˚C for a duration of 180 min during each annealing temperature, followed by natural cooling of the furnace to room temperature in vacuum. T4 Powder mixture of urea and Ta2O5 precursor were subjected to vacuum calcination in a tube furnace at 725 ˚C followed by increasing the furnace temperature to 950 ˚C for a duration of 20 min during each annealing temperature, followed by natural cooling of the furnace to room temperature in vacuum. This is followed by treating the sample to postsynthesis calcination in air at 700 ˚C for 1 h. T5 Powder mixture of urea and Ta2O5 precursor were subjected to vacuum calcination in a tube furnace at 725 ˚C followed by increasing the furnace temperature to 950 ˚C for a duration of 20 min during each annealing temperature, followed by natural cooling of the furnace to room temperature in vacuum. This is followed by treating the sample to postsynthesis calcination in Ar environment at 700 ˚C for 1 h. T6 Prior to vacuum calcination, the Ta2O5 powder was first allowed to undergo pre-synthesis thermal treatment in Ar environment at 700 ˚C for 1 h. This is followed by mixing the calcined Ta2O5 with urea before allowing the powder mixture to undergo vacuum calcination similar to the condition as sample T1. T7 Prior to vacuum calcination, the Ta2O5 powder was first allowed to undergo pre-synthesis thermal treatment in air at 700 ˚C for 1 h. This is followed by mixing the calcined Ta2O5 with urea before allowing the powder mixture to undergo vacuum calcination similar to the condition as sample T1. 26 The seven samples prepared via various synthesis procedures are as tabulated in Table 2.2: Table 2.2 Summary of synthesis conditions for samples T1 to T7. Sample Pre-synthesis calcination ambient of Ta2O5 precursor In-situ vacuum calcination time at 725 and 950 ˚C Post-synthesis calcination ambient T1 − 20 − T2 − 60 − T3 − 180 − T4 − 20 Air T5 − 20 Ar T6 Ar 20 − T7 Air 20 − After the synthesis process, the samples were then characterized and evaluated for their photocatalytic performances. 2.2.2 Photocatalytic reactions Photocatalytic reduction of H2O to H2 was used as a test reaction to evaluate the photocatalytic capabilities of the various TaON samples. The reactions were carried out in 25 mL-glass tubes with each tube containing 20 mg of TaON sample dispersed in a 10 mL aqueous solution with 10 vol. % methanol as sacrificial electron donor.33,34 The glass tubes were sealed with rubber septa to prevent gasses from entering or leaving the air space within the glass tubes. Next the air space was purged with Ar to remove any traces of foreign gasses while maintaining the pressure within the air space at atmospheric level. This was followed by irradiating the glass tubes with a 300 W xenon lamp (1000 W/m2) equipped with a 400 nm longpass filter for up to 5 h. A magnetic stirrer was used to maintain the photocatalyst powder in a constant suspended state. A 100 μL gas-tight syringe was used to draw the 27 evolved H2 gas hourly to determine the H2 concentration by a gas chromatograph (Shimadzu GC-2014). 2.3 Results and discussion 2.3.1 Characterization of photocatalyst The various TaON samples were examined and studied under several characterization tools. The study on the absorbance of the TaON samples was performed using the UV-visible diffuse reflectance spectroscopy (UV-Vis) whereas the morphology of the samples was examined using the Scanning Electron Microscopy (SEM). X-ray diffraction spectroscopy (XPS) was also used to examine the surface composition of Ta, N, and O elements of the samples whereas X-ray diffraction (XRD) analysis technique was performed to study the crystallinity and the crystal orientation of the samples. Apart from studying the physical and chemical properties of the samples, such characterization work were also necessary to validate that the synthesized compounds were indeed TaON. First and foremost, the XRD spectrum of Ta2O5 as well as sample T1 are shown in Figure 2.1 below: 10 15 20 25 30 35 2θ (°) 28 40 45 50 (202) (113) (130) (022) (220) (122) (112) (202) (121) (211) (002) (020) (200) (111) (021) (100) (011) (110) Intensity (a.u.) (111) (a) 55 60 10 15 20 25 30 35 40 45 (020) (002) (021) Intensity (a.u.) (111) (001) (110) (b) 50 55 60 2θ (°) Figure 2.1 XRD spectra of (a) sample T1 and (b) Ta2O5. In general, the XRD spectra for both materials appeared to have sharp and well-defined diffraction peaks. Such characteristics showed that both Ta2O5 and sample T1 were highly crystalline. Apart from that, the diffraction peaks of sample T1 can be indexed to β-TaON (ICDD No. 04-014-7350). Furthermore, the diffraction spectra of these samples were highly similar to the diffraction spectrum of TaON synthesized via the conventional nitridation method.7,19 Such finding suggests that TaON can indeed be synthesized via the urea route as described in section 2.2.1. Nevertheless, more evidence is necessary in order to confirm that the concerned samples are indeed TaON. In order to understand the absorbance of the samples, UV-visible diffuse reflectance spectroscopy was used to study the absorption range, peak and band edge of the effect of allo Ta2O5. The resulting UV-Vis spectra of the TaON sample as well as its Ta2O5 precursor are shown in Figure 2.2 (a). 29 (a) TaON Ta O 5 Intensity (a.u.) 2 200 300 400 500 600 700 800 Wavelength (nm) Figure 2.2 (a) UV–visible diffuse reflectance spectra for Ta2O5 and TaON and (b) colour difference between Ta2O5 and TaON (sample T7). For the Ta2O5 precursor, it had an absorption peak at approximately 260 nm with an approximate absorption band edge at 320 nm. On the other hand, TaON showed an absorption peak at 400 nm with an absorption band edge at around 550 nm. The latter showed a shift in absorption peak as well as absorption edge by approximately 140 nm and 180 nm, respectively. The alloying of Ta2O5 with nitrogen in the synthesis of TaON enabled the redshifting of both the absorption peak and edge from UV to visible light region. Another approach to determine the possibility of a photocatalyst absorbing visible light is by observing the photocatalyst colour. The observed change in colour from the white Ta2O5 to yellow TaON as shown in Figure 2.2 (b) also indicated the ability of TaON in absorbing light in the visible region. Besides, the absorption peak intensity of TaON had also been markedly increased 30 relative to its Ta2O5 precursor, with an approximate increase of 50 %. The characteristic of having a higher absorbance is linked to the ability of the photocatalyst to absorb more photon for the generation of electron-hole pairs in greater amount, hence a higher photocatalytic efficiency. As for the bandgap of the synthesized TaON sample, the value can be estimated by taking into consideration its absorption band edge value and apply with the following formula: (1) where is the bandgap value, speed of light and is the Planck's constant, representing the is the wavelength of the light, which can also be represented by the value of the absorption band edge. With = 550 nm, the bandgap value was estimated to be approximately 2.3 eV which is fairly close to the reported value of 2.5 eV.7,18-20 Another convenient and widely used method to estimate the bandgap value is through the plotting of the Tauc plot using the following formula.35,36 ( where is the photon energy, relative to the material and the material whereby ) (2) is the absorption coefficient, is a constant is determined by the type of optical transition of = 1 for direct transition and = 4 for indirect transition. As for the case of TaON, it has an indirect bandgap and this gives it the value of 4.8 Figure 2.3 shows the Tauc plot for the synthesized TaON sample obtained by applying Eq. (2). The estimated energy bandgap value, which was where the straight line in Figure 2.3 intercepted the x-axis, was approximately 2.08 eV. This value appeared to be in contradiction to the typical TaON bandgap value of 2.5 eV as reported by others.7,19 31 1.2 y = 0.925x - 1.9225 (αhv) 1/2 (eV.m ) -1 1/2 1 0.8 0.6 0.4 0.2 0 1 2 3 4 5 6 hv Figure 2.3 Tauc plot of the synthesized TaON sample. Indeed, Fang et al. performed the band structure calculations for TaON via two approaches, namely the Vienna Ab initio Simulation Program (VASP) and the Full Potential Linearized Augmented Plane Waves method (FPLAPW) using the WIEN97 program.8 The calculated bandgap value for TaON via the VASP and WIEN97 calculations were 1.8 eV and 2.0 eV, respectively, which also did not agree well with the experimental value of 2.5 eV.7,18-20 A possible explanation for such discrepency is that the bandgap value of TaON is not suitable to be estimated via the numerical calculation approach. XPS measurements were also performed to examine the surface composition of Ta, N, and O as well as the binding energies of the TaON and Ta2O5 samples which were corrected by reference to the Cls peak (284.6 eV). Figure 2.4 shows the resulting XPS spectra of TaON and its Ta2O5 precursor: 32 Figure 2.4 X-ray photoelectron spectra for (a) Ta 4f, (b) O 1s and (c) N 1s regions of Ta2O5 and TaON. It can be seen that the Ta 4f7/2, Ta 4f5/2, and O 1s peaks of Ta2O5 are positioned at 26.1 eV, 27.9 eV and 530.4 eV, respectively whereas the peaks positions Ta 4f7/2, Ta 4f5/2, O 1s and N 1s for TaON appear at 24.8 eV, 26.6 eV, 530.6 eV and 396.6 eV, respectively. The primary objective of XPS measurement was aimed at examining the presence of N 1s peak in the XPS spectrum of the TaON sample which was absent from the Ta2O5 precursor. Such results presented an additional proof to the successful incorporation of nitrogen element onto Ta2O5, which resulted in the synthesis of TaON. SEM imaging was also performed to understand whether the synthesis of TaON via the urea route had a profound effect on the morphology of the TaON sample. The SEM images of the TaON sample and its Ta2O5 precursor are as shown in Figure 2.5. Figure 2.5 SEM images of (a) Ta2O5, and (b) TaON. 33 Firstly, the TaON particles appeared to be agglomerated whereby such characteristic is absent from the Ta2O5 sample. The occurrence of such agglomeration could most likely be due to the sintering effect during the hightemperature synthesis process. Secondly, the surface of TaON appeared to be porous whereas Ta2O5’s surface was relatively smooth. This could be due to the etching effect by the corrosive NH3 gas released by the decomposition of urea. Urea releases NH3 gas as one of its decomposition by-products upon being heated at 210 ˚C or higher.37 The NH3 gas not only acts as a source for nitrogen but also as a strong etching agent. Since both Ta2O5 and urea were thoroughly mixed prior to thermal treatment, the NH3 gas released at the time of calcination instantly reacted with the oxide particles and thus etching the surface. NH3 is corrosive in nature and it is known for causing chemical etching of materials such as GaAs(100)38 and carbon nanotubes. In the latter case the NH3 gas specifically attacks the weaker C–C bonds of small chiral angle tubes.39,40 In the case of TaON, it could be deduced that one or more of the chemical bonds in TaON could have been attacked by the NH3 gas which then led to the corroded surface as seen in Fig. 2.5 (b). 2.3.2 Process of elemental nitrogen incorporation in Ta2O5 in the synthesis of TaON Conventionally, TaON is synthesized via nitridation process in which the Ta2O5 precursor is exposed to flowing NH3 gas at high temperature. At a sufficiently high temperature, the N element of NH3 would be introduced into the anionic (oxygen) network of Ta2O5. The partial replacement of oxygen by nitrogen within Ta2O5 will result in the synthesis of oxynitride materials with different physical and chemical properties, such as TaON which has a contrasting difference in physical colour appearance and bandgap properties. An example of the synthesis parameters used in the synthesis process of TaON via the nitridation reaction is the heating Ta2O5 in an atmosphere of flowing NH3 with a flow rate of 20 ml min-1 at 850 ˚C for up to 15 h.7,19 However, the synthesis of TaON via urea route is different in which the elemental nitrogen dopant source does not come directly from free-flowing NH3 gas. Instead, the NH3 gas is obtained by decomposing the nitrogen-rich urea. In the urea route, 34 Ta2O5 is mixed with sufficient amount of urea powder (ratio of Ta2O5/urea at 3:1) before being annealed in vacuum. This would result in the urea powder being instantly decomposed to its various by-products which include NH3 gas.37 At 725 ˚C, ammonia as a volatile gas breaks down to provide free atomic nitrogen. The energetic environment also provides the free atomic nitrogen sufficient energy to be incorporated into Ta2O5 which would result in the synthesis of an oxynitride material. However, at temperature below 725 ˚C such as 650 and 700 ˚C, the synthesized oxynitride material appeared to be lighter in yellow in comparison to TaON’s darker yellow hue, as seen in Figure 2.6. Such effect could be attributed to the significantly lower nitrogen over oxygen ratio in those oxynitride materials synthesized at 650 and 700 ˚C. After the annealing step at 725 ˚C, the temperature was raised 950 ˚C. This is because in order to convert the sample to TaON, the incorporated nitrogen dopants need to be further activated which can only be achieved by ramping up the calcination temperature to 950 ˚C while maintaining the calcination process in a vacuum environment. Figure 2.6 Colour of various oxynitride materials synthesized at the initial vacuum calcination temperature of (a) 725, (b) 700 and (c) 650 ˚C. 35 The introduction of nitrogen into Ta2O5 primarily resulted in the reduction of its bandgap, from 3.9 eV to approximately 2.5 eV, estimated from the absorption band edge of TaON as seen in Figure 2.2 (a). The conduction bands of both Ta2O5 and TaON consist of Ta5d atomic orbitals. On the other hand, the valence band of Ta2O5 consists of O2P atomic orbitals. As a result of nitrogen being incorporated into the anionic (oxygen) network of Ta2O5, the resulting valence band of TaON will consist of combination between O2p and N2P atomic orbitals.8 Since the new N2p atomic orbitals have a higher potential energy relative to the O2p atomic orbitals, the increase in nitrogen consequently results in the higher negative potential of the valence band of TaON, hence the narrowing of the energy bandgap of TaON.8,22,41 A schematic representation of the energy band structures of Ta2O5 and TaON is shown in Figure 2.7: Potential / vs NHE (pH = 0) Ta5d orbital Ta5d orbital ECB 0 (H /H2) + EGB 2.6 eV EGB 3.9 +1.23 (H2O/O2) O2p + N2p orbital EVB O2p orbital Ta2O5 TaON Figure 2.7 Schematic representation of the energy band structures of Ta2O5 and TaON. Another interesting aspect of TaON is that its energy bandgap is sufficiently large to allow overall photocatalytic splitting of water in theoretical terms, as shown in Figure 2.7. Furthermore, the narrowing in the 36 energy bandgap allows TaON to absorb light not only in the UV but also in the visible region as well, with the absorption band edge at approximately 550 nm. 2.3.3 Photocatalytic H2 evolution performance Table 2.3 shows the time course of H2 evolution under visible light irradiation (λ > 400 nm) for samples T1 to T7 synthesized under various conditions, namely different vacuum calcination durations, post-synthesis calcination conditions and conditions for the heat treatment of Ta2O5 precursor during the pre-synthesis step. Detailed explanation for reasons of various H2 evolution rate shown in Table 2.3 will be further discussed in Sections 2.3.3.1 to 2.3.3.3. Table 2.3 Photocatalytic activities for samples T1 to T7. Sample Pre-synthesis calcination ambient of Ta2O5 precursor In-situ vacuum calcination time at 725 and 950 ˚C Post-synthesis calcination ambient Activity -1 -1 (μmol g h ) T1 − 20 − 45.1 T2 − 60 − 35.6 T3 − 180 − 21.4 T4 − 20 Air 0 T5 − 20 Ar 49.1 T6 Ar 20 − 60.8 T7 Air 20 − 92.4 2.3.3.1 Effect of in-situ synthesis vacuum calcination duration Samples T1, T2 and T3 in Table 2.3 represented the TaON samples synthesized via 20, 60 and 180 min of vacuum calcination at both 725 and 950 ˚C. Among the three different durations, sample T1 with 20 min of vacuum calcination duration produced the highest photocatalytic activity of 45.1 μmol g-1 h-1. However, as the vacuum calcination duration progressively increased 37 to 60 and 180 min, the photocatalytic activities dropped dramatically to 35.6 -1 -1 and 21.4 μmol g h , respectively. The reason for the decrease in activity by samples T2 and T3 could be attributed to the loss of nitrogen content.30 EDX elemental analysis on nitrogen content was conducted on samples T1 to T3 with each sample scanned for 10 times before calculating the average values. The resulting average values of the atomic percentage for the 3 samples’ nitrogen content are tabulated as shown in Table 2.4: Table 2.4 EDX readings for the average atomic % of nitrogen content in samples T1, T2 and T3. Sample Average atomic % of nitrogen content T1 34.9 T2 34.2 T3 33.8 It can be seen that with increasing vacuum calcination duration from 20 to 180 min, the nitrogen content of sample T3 dropped by around 1 % as compared to sample T1. Since nitrogen is responsible for introducing N2p atomic orbitals into the valence band of TaON which subsequently help to reduce the energy bandgap, a lower nitrogen content may result in less reduction in the bandgap value, hence a lower photocatatlytic efficiency. However, EDX analysis may not give an accurate result on the atomic % of samples in general. As a result, the obtained values of the nitrogen atomic % shown in Table 2.4 may not be sufficiently accurate, and hence the results have to be interpreted cautiously. Secondly, an extended duration of vacuum calcination duration could also lead to the gradual agglomeration and fusion of particles, thus resulting in the formation of larger size particles. Particle agglomeration has always been a downside of thermal annealing, especially at high temperature at an extended duration.42-44 Such effect was captured and shown in Figure 2.8. Particles of sample T1 and T2 which were calcined for 20 and 60 min respectively show 38 lesser degree of particle agglomeration as compared to sample T3 which was calcined for up to 180 min, thus resulting in much more particle agglomeration. This in turn results in the formation of grain boundaries within the newly-formed larger particles which acted as recombination sites for the photogenerated electrons and holes.45 On the other hand, an increase in size also led to the decrease in the photocatalyst’s effective surface area and thus redueced the amount of photon that was able to reach the surface of the photocatalyst to generate electron-hole pairs for the photocatalytic reduction of H2O to form H2. (a) (b) 400 nm 400 nm (c) 400 nm Figure 2.8 SEM images of samples T1, T2 and T3 showing various degree of particle agglomeration. Thirdly, prolonged vacuum calcination in an oxygen-poor environment may also induce the increase of oxygen vacancy density within the sample.46 Oxygen vacancies have been described as acting as recombination centres for photoinduced electron and hole charge carriers. This inevitably contributes to the decrease in carrier concentration and slower charge transport.45,47,48 As a result, less free electrons are available to undergo photoreduction of water to H2, hence a lower photocatalytic efficiency with increasing oxygen vacancy density within the photocatalyst. 39 2.3.3.2 Effect of post-synthesis calcination Attempts were also made to increase the photocatalytic efficiency of the samples by allowing them to undergo post-synthesis calcination, either in air or in inert Ar ambient at 700 ˚C with the resulting photocatalytic activities shown in Table 2.3 (samples T4 and T5). Sample T4 which had undergone calcination in air appeared to be white in colour and did not exhibit any photocatalytic activity. This observation could be attributed to the oxidizing effect by air, especially at high temperature which could have driven the nitrogen content out of the sample, thus resulting in the TaON being reverted back to its original Ta2O5 form. Analysis of sample T4 with XRD showed that its spectrum is highly similar to the XRD spectrum of Ta2O5, as shown in (111) 10 15 20 25 30 35 40 45 (020) (002) (021) Intensity (a.u.) (001) (110) Figure 2.9 below. 50 55 60 2θ (°) Figure 2.9 XRD spectrum of sample T4. As for sample T5, its recorded photocatalytic activity was slightly higher as compared to sample T1. Here the thermal treatment of the sample in an inert environment could have likely contributed to the reduction of defects acting as electron-hole pair recombination sites, hence the slight improvement in the photocatalytic efficiency. 2.3.3.3 Effect of pre-synthesis calcination 40 Thermal treatment of the Ta2O5 precursor was also found to contribute to the increase in photocatalytic efficiency of the photocatalysts. The Ta2O5 precursors of samples T6 and T7 both underwent pre-synthesis thermal treatment in Ar and air, respectively. The former showed a photocatalytic activity of 60.8 μmol g-1 h-1 whereas the latter 92.4 μmol g-1 h-1. In order to further understand the effect of calcination, samples T1, T6 and T7 were analyzed under photoluminescence (PL). Identical amount of powder for all three samples was spread across an approximately same surface area to ensure uniformity in thickness across the samples. The differences in the PL emission spectrum of the three samples are as shown in Figure 2.10. Intensity (a.u.) Sample T1 500 Sample T6 Sample T7 550 600 650 700 750 800 Wavelength (nm) Figure 2.10 PL spectra of samples T1, T6, and T7. In a typical PL emission spectrum, a higher intensity signifies a higher degree of recombination between electrons and holes, and vice versa.49,50 Besides that, trapping of excitons by defects also contributes to the PL emission peak.49-52 It is notably clear that samples T6 and T7, in which their Ta2O5 precursors had undergone thermal treatment resulted in lower PL peak intensity as compared to sample T1 with its Ta2O5 precursor not being subjected to any thermal treatment. Between samples T6 and T7, the latter displayed a more significant drop in the PL intensity. The PL peak intensity of certain materials have been associated with the presence of oxygen vacancies.53-55 Oxygen vacancies are known to act as electron-hole pair 41 recombination centres, which are responsible for the decrease in carrier concentration and slower charge transport.16,20,21 As a matter of fact, Devan et al. reported that oxygen vacancies exist within the bandgap of Ta2O5.56,57 Ta2O5 crystalline structures are built on a network of only two polyhedral building blocks, TaO6 octahedra and TaO7 pentagonal bipyramids with shared oxygen atoms.58,59 The oxygen anions are located only at the in-plane and cap sites of Ta2O5, thus causing the anions to be easily volatilized to yield oxygen vacancies.58,59 This is because whenever the oxygen anions leave from the two in-plane and cap sites of Ta2O5 to form oxygen vacancies, each oxygen vacancy can provide a reduction of two electrons at two dangling bonds. The consequence of the formation of such dangling bonds is that they can give rise to trap levels within the bandgap of Ta2O5 which are capable of trapping excited electrons.56 By exposing the Ta2O5 precursor to thermal annealing in air, the oxygen vacancies can be minimized or passivated.54,60 This in turn resulted in lower recombination frequency between the electron-hole charge carriers or trapping of free electrons by defects, hence the enhanced photocatalytic efficiency by sample T7. As for the effect of thermal annealing in an inert Ar environment, the reduction of defect density is likely to be less significant than air calcination, thus resulting in a marginal increase in the photocatalytic efficiency by sample T6. It could also be due to the inability of inert gas to repair oxygen vacancies within the TaON sample, hence the inferior photocatalytic performance by sample T6 in comparison to sample T7. 2.4 Conclusions In conclusion, the TaON photocatalyst for the photocatalytic water reduction assisted by methanol as sacrificial agent has been successfully synthesized through an alternative approach via the urea route. In the urea route, urea was decomposed at high temperature which resulted in the release of NH3 gas. NH3 provides Ta2O5 with the nitrogen source for the conversion of Ta2O5 to TaON. The incorporation of nitrogen gave rise to new N2p atomic orbitals which resulted in a higher negative potential of the valence band for TaON. This in turn contributed to a reduced energy bandgap value for TaON as compared to its Ta2O5 precursor. The XRD and UV–visible diffuse 42 reflectance spectra of the TaON synthesized via urea route matched relatively well to those synthesized via the conventional method, supplemented by XPS study which showed the presence of N element in the sample. The TaON sample with the best photocatalytic performance based on the methanolassisted photoreduction of water can be synthesized with the shortest vacuum calcination duration of 20 min each at 725 and 950 ˚C with its Ta2O5 precursor pre-calcined in air for 1 h in order to reduce its oxygen vacancies. To the best of our knowledge, such TaON synthesized via urea route has resulted in one of the highest efficiencies in terms of photocatalytic water reduction with the methanol as sacrificial agent as well as a higher UV-visible light absorption. References [1] A. K.P.D. Savioa, D. Starikovb, A. Bensaoulab, R. Pillaib, L. L. de la Torre Garcíad and F. C. Robles Hernández, Ceram. Int., 2012, 38, 3529. [2] C. Yang, X. M. Li, Y. F. Gu, W. D. Yu, X. D. Gao, and Y. W. Zhang, Appl. Phys. Lett., 2008, 93, 112114. [3] K. Takanabe, K. Kamata, X. Wang, M. Antonietti, J. Kubota and K. Domen, Phys. Chem. Chem. Phys., 2010, 12, 13020. [4] V. H. Houlding and M. Gräzel, J. Am. Chem. Soc., 1983, 105, 5695. [5] R. Abe, K. Sayama and H. Arakawa, J. Photochem. Photobiol., A, 2004, 166, 115. [6] R. Abe, K. Hara, K. Sayama, K. Domen and H. Arakawa, J. Photochem. Photobiol. A, 200, 137, 63. [7] M. Hara, G. Hitoki, T. Takata, J. N. Kondo, H. Kobayashi and K. Domen, Catal. Today, 2003, 78, 555. [8] C. M. Fang, E. Orhan, G. A. de Wijs, H. T. Hintzen, R. A. de Groot, R. Marchand, J.-Y. Saillardd and G. de With, J. Mater. Chem., 2001, 11, 1248. [9] S. Palomba, M. Danckwerts and L. Novotny, J. Opt. A-Pure Appl. Op., 2009, 11, 114030. [10] J. Jiang, H. Li and L. Zhang, Chem. Eur. J., 2012, 18, 6360. [11] D. B. Ingram and S. Linic, J. Am. Chem. Soc., 2011, 133, 5202. [12] T. Ahmadi, Z. L. Wang, T. C. Green, A. Henglein and M. El-Sayed, Science, 1996, 272, 1924. 43 [13] R. Wang, Y. Zhu, Y. Qiu, C. Leung, J. He, G. Liu and T. Lau, Chem. Eng. J., 2013, 226, 123. [14] K. Lai, Y. Zhu, J. Lu, Y. Dai and B. Huang, Comput. Mater. Sci., 2013, 67, 88. [15] X. Yang, A.m Wolcott, G. Wang, A. Sobo, R. C. Fitzmorris, F. Qian, J. Z. Zhang and Y. Li, Nano Lett., 2009, 9, 2331. [16] H. Wu and Z. Zhang, Int. J. Hydrogen Energ., 2011, 36, 13481. [17] J. Cao, Y. Zhang, H. Tong, P. Li, T. Kako and J. Ye, Chem. Commun., 2012, 48, 8649. [18] K. Maeda and K. Domen , J. Phys. Chem. C, 2007, 111, 7851. [19] G. Hitoki, T. Takata, J. N. Kondo, M. Hara, H. Kobayashi and K. Domen, Chem. Commun., 2002, 16, 1698. [20] W. J. Chun, A. Ishikawa, H. Fujisawa, T. Takata, J. N. Kondo, M. Hara, M. Kawai, Y. Matsumoto and K. Domen, J. Phys. Chem. B, 2003, 107, 1798. [21] T. Mishima, M. Matsuda and M. Miyake, Appl. Catal. A – Gen., 2007, 324, 77. [22] R. Abe, T. Takata, H. Sugihara and K. Domen, Chem. Commun., 2005, 30, 3829. [23] R. Abe, J. Photochem. Photobiol. C, 2010, 11, 179. [24] S. Bae, H. Shin, S. Lee, D. W. Kim, H. S. Jung and K. S. Hong, React. Kinet. Mech. Cat., 2012, 106, 67. [25] J. Hou, Z. Wang, R. Cao, S. Jiao and H. Zhu, Dalton Trans., 2011, 40, 4038. [26] Y. Du, L. Zhao and Y. Su, J. Hazard. Mater., 2011, 195, 291. [27] M. Higashi, K. Domen and R. Abe, J. Am. Chem. Soc., 2012, 134, 6968. [28] A. Dabirian, H. van’t Spijker, R. van de Krol, Energy Procedia, 2012, 22, 15. [29] J. H. Swisher and M. H. Read, Metall. Trans., 1972, 3, 493. [30] J. Yuan, M. Chen, J. Shi and W. Shangguan, Int. J. Hydrogen Energ., 2006, 31, 1326. [31] A. P. Bhirud, S. D. Sathaye, R. P. Waichal, L. K. Nikama and B. B. Kale, Green Chem., 2012, 14, 2790. [32] Q. Gao, C. Giordano and M. Antonietti, Small, 2011, 7, 3334. [33] T. Kawai and T. Sakata, J. C. S. Chem. Commun., 1980, 694. [34] L. Millard and M. Browker, J. Photochem. Photobiol. A, 2002, 148, 91. [35] R. Jiang, H.Y. Zhu, X.D. Li and L. Xiao, Chem. Eng. J., 2009, 152, 537. 44 [36] J. Tauc, Mater. Res. Bull., 1968, 3, 37. [37] P. M. Schabera, J. Colsonb, S. Higginsb, D. Thielenb, B. Anspachb and J. Brauerb, Thermochim. Acta, 2004, 424, 131. [38] M. V. Lebedev, E. Mankel, T. Mayer and W. Jaegermann, J. Phys. Chem. C, 2010, 114, 21385. [39] O. Gu¨lseren, T. Yildirim and S. Ciraci, Phys. Rev. B, 2002, 65, 153405. [40] U. Borstnik, M. Hodoscek, D. Janezic, and I. Lukovits, Chem. Phys. Lett., 2005, 411, 384. [41] M. Yashima, Y. Lee and K. Domen, Chem. Mater., 2007, 19, 588. [42] P. K. Narayanam, P. Soni, P. Mohanta, R.S. Srinivasa, S.S. Talwar and S.S. Major, Mater. Chem. Phys., 2013, 139, 196. [43] A. Mills and S. Morris, J, Photochem. Photobiol. A: Chem., 1993, 71, 285. [44] E. I. Solomon, A. J. Augustine and J. Yoon, Dalton Trans., 2008, 30, 3921. [45] J. Wang, D. N. Tafen, J. P. Lewis, Z. Hong, A. Manivannan, M. Zhi, M. Li and N. Wu, J. Am. Chem. Soc., 2009, 131, 12290 [46] Z. Zhang, Y. Zhao and M. Zhu, Appl. Phys. Lett., 2006, 88, 033101. [47] P. H. Tsai, T. Norby, T. T. Tan, R. Donelson, Z. D. Chen and S. Li1, Appl. Phys. Lett., 2010, 96, 141905. [48] Md. K. Parvez, G. M. Yoo, J. H. Kim, M. J. Ko and S. R. Kim, Chem. Phys. Lett., 2010, 495, 69. [49] Y. P. Xie, G. Liu, G. Q. Lu and H. Cheng, Nanoscale, 2012, 4, 1267. [50] X. Meng, Z. Shi, X. Chen, X. Zeng and Z. Fu, J. Appl. Phys., 2010, 107, 023501. [51] H. S. Kang, J. S. Kang, J. W. Kim, and S. Y. Lee, J. Appl. Phys., 2004, 95, 1246. [52] T. B. Hur, G. S. Jeen, Y. H. Hwang, and H. K. Kim, J. Appl. Phys., 2003, 94, 5787. [53] K. Lee, W. S. Seo and J. T. Park, J. Am. Chem. Soc., 2003, 125, 3408. [54] M. Zhu, Z. Zhang and W. Miao, Appl. Phys. Lett., 2006, 89, 021915. [55] M. Herrera, D. Maestre, A. Cremades and J. Piqueras, J. Phys. Chem. C, 2013, 117, 8997. [56] R. S. Devan, C. Lin, S. Gao, C. Cheng, Y. Liou and Y. Ma, Phys. Chem. Chem. Phys., 2011, 13, 13441. [57] H. Shin, S. Y. Park, S. Bae, S. Lee, K. S. Hong and H. S. Jung, J. Appl. Phys., 2008, 104, 116108. [58] H. Sawada and K. Kawakami, J. Appl. Phys., 1999, 86, 956. [59] R. Ramprasad, J. Appl. Phys., 2003, 94, 5609. 45 [60] J. Weidmann, Th. Dittrich, E. Konstantinova, I. Lauermann, I. Uhlendorf and F. Koch, Sol. Energy Mater. Sol. Cells., 1999, 56, 153. 46 Chapter 3 Loading of CuO nanoparticles on WO3 for enhanced visible light response for photocatalytic oxidation of water Abstract The photocatalytic oxidation of water from H2O to O2 by CuO-loaded WO3 (CuO-WO3) composite photocatalysts were investigated under visible light with the use of Fe3+ ions as the sacrificial hole acceptor. The narrow bandgap of CuO with values ranging from 1.2 to 1.6 eV enables the absorption of visible light, which then allows it to act as a sensitizer to WO3 by enhancing the photogeneration rate of electron-hole pair for photocatalytic reactions. Furthermore, the formation of p-n junction between the n-type CuO and ptype WO3 semiconductor helps to promote the separation of electron-holes pairs, thus reducing the recombination rate between the two charge carriers. In this work, CuO nanoparticles with an average diameter of 5 nm were synthesized prior to loading them onto the surface of WO3 in order to form the CuO-WO3 composite photocatalyst. A CuO nanoparticle loading amount of 4 wt.% resulted in the highest photocatalytic O2 evolution rate of 75.23 μmol g1 h-1 among the various samples with CuO nanoparticle loading amount ranging from 0.5 to 8 wt. %. The CuO-WO3 composite photocatalyst samples were also subjected to post-synthesis thermal annealing in air and N2 atmospheres, whereby the former resulted in the decrease of O2 evolution efficiency whereas the latter one contributed to slight increase in the photoactivity by the N2-annealed composite photocatalyst sample. 3.1 Introduction There have been several studies on methods to increase the light absorption range of photocatalyst, especially to absorb visible light. An effective and commonly used method is the incorporation of anion dopants such as N,1 S,2 and F3 into the photocatalyst to reduce the bandgap of the photocatalyst. Another approach is the loading of noble metallic nanoparticles, for instance Au and Ag on the surface of photocatatalyst which could give rise 47 to the localized surface plasmon resonance (LSPR) effect,4,5 whereby such effect helps to extend the photoabsorbance of the photocatalyst by capturing a larger portion of the visible light region. Such effect is particularly beneficial for photocatalysts which are only limited to absorbing UV light, such as TiO2, Ta2O5 and ZnO.6-8 Apart from the abovementioned methods, the use of sensitizing materials such as organic dye in the form of magnesium phthalocyanine (MgPc), 8-hydroxyquinoline (HOQ) and Eosin Y9-12 and semiconductor nanoparticles such as CdS,13,14 CdSe15 and CdTe16 is also effective in enhancing the visible light absorbance of the photocatalyst. Even though WO3 is intrinsically a visible light-absorbing photocatalyst due to its bandgap value of 2.7 eV,17 its visible light absorption range is relatively weak. It has an absorption band edge at approximately 480 nm which is only a slight extension into the visible region from UV.18 As mentioned previously, sensitizing material can also be used to enhance the absorbance of a photocatalyst in the visible region. Apart from CdS, CdSe, CddTe and other organic sensitizing dye materials such as MgPc, HOQ and Eosin Y, CuO is also one of the commonly used sensitizing agents.19,20 Apart from acting as a sensitizing agent, CuO also forms p-n junction with certain materials which provide enhanced charge separation effect and helps to reduce recombination between electrons and holes.20,21 CuO belongs to the category of p-type transitional metal oxide semiconductor material with a monoclinic crystalline structure. It is a nontoxic material in nature, highly stable and generally requires simple and lowcost processing steps.22-24 Due to these reasons, CuO nanoparticles have found various applications in a wide variety of fields, for instance electronic and optoelectronic devices. The specific examples are microelectromechanical system (MEMS),25 solar cells,26,27 magnetic storage media28 and lithium batteries.29 Besides, CuO nanoparticles are also widely use in catalytic activities such as gas sensing,30-32 antibacterial activity33 and degradation of organic contaminant.34,35 Apart from the abovementioned applications, CuO particles is also a common study material in the research of photocatalytic water splitting, primarily due to its narrow band gap ranging from 1.2 to 1.6 eV which allows the CuO to either act as a visible light-absorbing photocatalyst or as a sensitizing agent.36-38 However, the energy bandgap value 48 is bigger in the case of CuO nanoparticle as a result of quantum confinement effect. Several works reported the value to be around 2.5 to 3 eV,39-43 depending on the size of the CuO nanoparticle. The advantage of having such a narrow band gap is that not only CuO is able to absorb UV light but visible light as well. This contributes to the increase in photon absorption, which in turn leads to higher amount of electron-hole pair generation. Apart from behaving as a sensitizer, the favourable conduction and valence energy band positions of p-type CuO also enable it to act as either electron or hole sink by forming a p-n junction with the n-type WO3 photocatalyst which further enhance charge separation between the photogenerated electrons and holes, thus minimizing the charge recombination rate.20,44-46s In this chapter, the primary objective is to study the feasibility of loading CuO nanoparticle as a sensitizing agent on WO3 by investigating and understanding its influence on the photocatalytic O2 evolution efficiency of the CuO-loaded WO3 composite photocatalyst under visible light irradiation. Various WO3 samples were loaded with varying amount of CuO nanoparticles and the consequences of different loading amount were studied and discussed. The effects of post-calcination in various atmospheres on the photocatalytic performance of the composite photocatalyst were also investigated. Lastly, the composite photocatalysts were also studied under X-ray diffraction spectroscopy (XRD), UV-Visible spectrometer (UV-Vis), scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX) and transmission electron microscopy (TEM) to further understand the chemical and physical structure of the photocatalysts. 3.2 Experimental procedures 3.2.1 Synthesis of CuO-loaded WO3 composite photocatalyst In this work, the CuO-WO3 composite photocatalyst was prepared by first synthesizing the CuO nanoparticles separately followed by depositing them on WO3. Typically, 2.1 mmol of Cu(NO3)2 and 4.2 mmol of acetic acid were added into 15 mL of absolute ethanol and stirred to ensure that the chemicals are mixed and dissolved completely. Next, the solution was heated in a water bath at a temperature of 78 ˚C for several minutes. This was 49 followed by adding 8.4 mmol of NaOH into the solution under vigorous stirring. Upon the addition of NaOH, the colour of the solution instantly changed from turquoise to dark brown, which indicated the formation of CuO nanoparticles. The solution was then left to stir at 78 ˚C for the next 1 hour. Next, the prepared dark brown solution was collected and washed with D.I. water through centrifugation in order to remove impurities, particularly NaNO3 and unreacted OH- ions. By the end of the washing process, the sample was scanned using EDX in order to ensure the absence of Na+ ions from the sample. As for the testing of OH- ions, a pH meter was used to test the pH value of the solution in the centrifuge tube containing the washed composite photocatalyst powder, whereby a neutral solution indicates the complete removal of OH- ions. Once the washing process had been completed, the brown suspension in DI water was left to dry in oven. After the drying step, 0.1 g of CuO nanoparticle powder was collected and dispersed in 15 mL of absolute ethanol to produce a CuO nanoparticle suspension with a concentration of 6.667 g/L. Next, 0.3 g of WO3 powder was added to 15 mL of DI water and dispersed through sonication for several minutes followed by stirring the WO3 suspension vigorously. At the same time, the CuO nanoparticle suspension was also subjected to sonication and vigorously stirring in order to ensure a thorough and homogeneous dispersion of the nanoparticles, as well as to ensure the concentration of CuO nanoparticles in the suspension was evenly distributed. After the vigorous stirring of WO3 nanoparticles suspension, appropriate amount of CuO nanoparticle suspension was drawn with a pipette and transferred to the WO3 suspension while under stirring. Six samples of WO3 with different amount of CuO loading weight % (wt. %) – 0.5, 1, 2, 4, 6 and 8 wt. % were prepared. The suspension containing WO3 and CuO nanoparticles were allowed to stir for 2 h in order to physically load the CuO nanoparticles onto WO3. After that, the suspension was dried in a water bath heated to the temperature of 60 °C. Finally, the dried composite photocatalysts samples with various CuO nanoparticle loading wt. % were collected and tested for their photocatalytic efficiency in terms of O2 evolution efficiency rate. 50 3.2.2 Photocatalytic reactions Photocatalytic oxidation of H2O to O2 reaction was performed on all samples as a test reaction to evaluate the photocatalytic efficiency of the various CuO-WO3 composite photocatalyst samples. Each test reaction was carried out in a 25 mL-quartz tube with containing 10 mg of sample powder dispersed in a 7 mL aqueous solution with 50 mmol Fe3+ ions as sacrificial electron acceptors. The Fe3+ ion plays several crucial roles in the photocatalytic O2 evolution reaction: (1) it acts as an electron-scavenging agent for the photoinduced electrons by the CuO-WO3 composite photocatalyst to minimize the effect of holes from recombining with the electrons, (2) Fe3+ ion prevents or minimizes the formation of superoxide radical anion (•O2 − ) which may reduce the amount of O2 molecules being detected by the gas chromatograph,47 whereby such outcome could otherwise be misinterpreted as the composite photocatalyt having a low photocatalytic O2 evolution efficiency, and (3) Fe3+ reduces the probability of water splitting backward reaction in which O2 molecules can be reverted back to H2O molecules as a result of the reaction between O2, electrons and H+ ions. Once the appropriate materials had been placed inside the quartz tube, the glass tube was sealed with a tight-fitting rubber septum and wrapped with paraffin film in order to prevent exchange of gas between the sealed quartz tube and the outer environment. After ensuring that the quartz tube had been tightly sealed, the air space within was purged with inert gas such as Ar in order to remove any trace of foreign gasses from the quartz tube. At the same time, the pressure of the air space within was maintained at atmospheric throughout and after the purging process. This is to prevent the build-up of inner pressure which may lead to the cracking of the quartz tube as well as impeding the release of evolved O2 gas from the solution. After the purging process, the quartz tube was irradiated with a 300 W xenon lamp at an intensity of 1000 W/m2 equipped with a 400 nm longpass filter for up to 3 h. During the irradiation process, the photocatalyst in the solution was stirred continuously with a magnetic stirrer in order to maintain the photocatalyst powder constantly in a suspended state. A gas-tight syringe with a 100 μL 51 volume was used to draw the evolved H2 gas hourly to determine the O2 concentration by a gas chromatograph (Shimadzu GC-2014). 3.3 Results and discussion 3.3.1 Materials characterization The CuO-WO3 composite photocatalyst samples were examined and studied under several characterization tools. For example, the study on the absorbance of the various composite photocatalyst samples was performed using the UV-visible diffuse reflectance spectroscopy (UV-Vis), whereas the morphology of the composite photocatalyst sample was studied under the scanning electron microscope (SEM). High resolution imaging of the CuO nanoparticle-loaded WO3 photocatalyst was also performed with transmission electron microscope (TEM) and finally the samples were analysed by X-ray diffraction (XRD) technique in order to determine the diffraction spectra of the CuO nanoparticles as well as the composite photocatalyst. Figure 3.1 (a) shows the XRD spectrum of the CuO nanoparticles synthesized as explained in section 3.2.1. The observed diffraction peaks can be indexed to the monoclinic CuO crystal structure (JCPDS Card No. 450937). No characteristic peaks can be observed for Cu(OH)2 or Cu2O, but the XRD peaks at 2θ = 32.6, 35.6, 38.8, 48.9, 53.5, 58.4, 65.9, 66.4 and 68.1° which correspond to the (110), (002), (11-1), (20-2), (020), (202), (11-3), (113) and (113) lattice planes can be assigned to the XRD peaks of CuO.48,49 The presence of these diffraction peaks indicate that CuO of pure phases were successfully synthesized, with the lattice constants reported to be a = 4.68 Å, b = 3.44 Å, and c = 5.15 Å.50 Besides, the diffraction peaks appear to be narrow which suggest that the CuO had a relatively high degree of crystallinity. The crystalline size of the CuO nanoparticles can be determined from its XRD spectrum by using the Scherrer equation,51 represented as: (1) where D is the calculated crystalline size of the nanoparticle concerned, K is a dimensionless constant known as the shape factor with a typical value of 0.9, λ 52 is the wavelength of the X-ray irradiation used in the XRD spectroscopy, β is the line broadening at half the maximum intensity (FWHM) of the diffraction peak in radians and θ is the diffraction angle in degrees. By considering the diffraction peak for (11-1) plane at 2θ = 39 °, the value of β would be 0.0234 radians and with λ = 1.54 Å, the calculated size of the CuO nanoparticle is approximately 7.6 nm. The CuO-WO3 composite photocatalyst sample was also characterized with the XRD spectrometry, with the resulting diffraction spectrum is as shown in Figure 3.1 (b). The diffraction peaks for WO3 are clearly visible, especially for the (001), (020), (200), (-120), (-111), (021), (201), (12-1), (221), (002), (11-2), (202), (041) and (14-1) planes. However, most of the CuO diffraction peaks could not be detected except for the peak at 2θ = 48.9° which can be assigned to the CuO (-111) lattice plane. However, its peak intensity is weak and barely noticeable, which is encircled and indicated by a dot on top of it, as shown in Figure 3.1 (b). Despite a relatively high CuO nanoparticles loading amount of 8 wt. % that were deposited on the WO3 photocatalyst, none of the diffraction peak referring to CuO was detected. Nevertheless, such behaviour is not surprising as several works on CuO nanoparticle loading also experienced similar situation whereby the XRD analysis on the CuO-loaded samples failed to indicate any presence of CuO loading.52-55 Such could most likely be due to the extremely small CuO nanoparticle size coupled with the fact that the nanoparticles were homogeneously dispersed on the surface of WO3 that resulted in the ability of XRD to detect its presence.56 Another factor is due to the low signal-to-noise ratio for the CuO XRD peak due to the low CuO loading wt. % which resulted in the (-111) CuO XRD peaks to be overlapped by the background noise during XRD scanning.57 53 20 25 30 35 40 45 50 55 60 (113) (113) (113) (202) (020) (202) (111) (110) Intensity (a.u.) (002) (a) 65 70 2θ (°) CuO (020) (200) 20 25 30 35 50 (141) (202) (002) 45 (112) (041) 40 (221) (111) (121) (021) (111) (120) (201) Intensity (a.u.) (001) (b) 55 60 65 70 2θ (°) Figure 3.1 XRD spectra for (a) CuO nanoparticles and (b) CuO-WO3 composite photocatalyst. In order to study the absorbance of the CuO-WO3 composite photocatalyst samples with varying loading amount of CuO nanoparticle, the samples were analyzed under UV-Vis. Figure 3.2 shows the absorbance of pristine WO3, CuO nanoparticle and various CuO-WO3 composite photocatalyst samples loaded with 0.5, 1, 2, 4, 6 and 8 wt. % of CuO nanoparticles from 200 to 800 nm. From Figure 3.2, it can be observed that the synthesized CuO nanoparticles appeared to have an absorbance which covers the entire visible 54 light spectrum due to their narrow bandgap.58,59 On the other hand, WO3 appeared to have slight absorbance in the visible light region with an absorption edge at approximately 480 nm. However, with the addition of CuO nanoparticles the composite photocatalyst samples showed enhanced absorbance in the visible light region. The visible light absorption range of the composite photocatalyst samples gradually increased as the CuO nanoparticles loading amount increased. In other words, the loading of CuO nanoparticles would help to better utilize the visible region, hence increasing the rate of electron-hole pair generation by the composite photocatalyst. Such effect may result in higher O2 evolution rate by the CuO-WO3 composite photocatalyst, but at the same time there may also be other factors that would influence the final photocatalytic efficiency of the composite photocatalyst. These reasons will be further discussed in Section 3.3.3.1. Absorbance (a.u.) Pristine WO3 CuO nanoparticles 0.5 wt. % CuO 1 wt. % CuO 2 wt. % CuO 4 wt. % CuO 6 wt. % CuO 8 wt. % CuO 200 300 400 500 600 700 800 Wavelength (nm) Figure 3.2 UV-Vis absorbance of WO3, CuO nanoparticles and the CuO-WO3 composite photocatalysts loaded with various wt. % of CuO nanoparticles. WO3, CuO nanoparticles and the CuO-WO3 composite photocatalyst were studied under SEM as well in order to understand their morphology. Figure 3.3(a) and (b) show the SEM images of WO3 and CuO nanoparticles, respectively whereas Figure 3.3(c) and (d) show the respective SEM images of the composite photocatalyst with a CuO nanoparticle loading of 4 wt. % at the respective magnification of 30 k and 50 k. 55 (b) (a) 500 nm 500 nm (c) (d) 100 nm 200 nm Figure 3.3 SEM images of (a) pristine WO3, (b) CuO nanoparticles and CuOWO3 composite photocatalyst at (c) 30 k and (d) 50 k magnification. From Figure 3.3(a) it can be observed that the average size of pristine WO3 particles is approximately 100 nm. On the other hand, the individual CuO nanoparticles as seen in Figure 3.3 (b) are indistinguishable due to their extremely small particle size. The relatively low electrical conductivity nature of oxide materials further complicated the SEM imaging process and thus the imaging of CuO nanoparticles would require microscopy tools of higher resolution such as TEM. Naturally, CuO nanoparticles loaded on WO3 could not be clearly observed in the SEM images of the composite photocatalyst shown in Figure 3.3 (c) and (d). However, under the analysis of EDX the presence of Cu and O elements were detected, as displayed in Figure 3.4. Another observation that can be discerned Figure 3.3(a) and (c) is that the WO3 particles belonging to the composite photocatalyst appear to be more agglomerated than the pristine WO3 particles. The cause for the agglomeration could most likely be the result of attractive capillary force between the particles during the drying process in a water bath as described in section 3.2.1. 56 Intensity (a.u.) W O Cu C 0 W Cu W W 1 2 3 4 5 6 7 Cu 8 W W W 9 10 11 12 13 Energy (keV) Figure 3.4 EDX spectrum for the CuO-WO3 composite photocatalyst sample indicating the presence of Cu in the sample. TEM analysis was also performed to further understand the degree of crystallinity, size as well as to the lattice spacing value of the synthesized CuO nanopartice. Figure 3.5 shows the TEM images of WO3 loaded with CuO nanoparticle of 20 wt. % in loading amount. The reason for using such high loading wt. % is to ease the TEM imaging process of CuO nanoparticles. This is because the synthesized CuO nanoparticles are very small in size and this may cause the CuO nanoparticles to be elusive and unable to be captured under the TEM imaging process if the loading amount is too low. Based on the TEM images, the average diameter of the CuO nanoparticles is calculated to be around 5 nm. Unfortunately, the sizes of the majority nanoparticles could not be estimated any more accurately due to the agglomerated state of the nanoparticles, which also caused difficulty in distinguishing the individual nanoparticles. As for the degree of crystallinity of the CuO, it appeared to be polycrystalline with some visible grain boundaries. Besides, the TEM image also provides us with the lattice spacing of the nanoparticle. In this case the value of the lattice spacing appears to be approximately 2.45 Å. The value is similar to the values reported by Yao et al. and Kitsomboonloha et al. and this indicates that the nanoparticle compound type belongs to CuO.60,61 57 (a) (b) WO3 CuO 50 nm 50 nm (d) (c) WO3 2.45 Å CuO 4 nm 20 nm Figure 3.5 TEM images of the CuO-WO3 composite photocatalyst loaded with 4 wt. % CuO nanoparticle and the obtained lattice spacing value of CuO. 3.3.2 Synthesis conditions of CuO nanoparticle and its loading process on WO3 There are reported methods on the synthesis of CuO nanostructures, namely flower-like CuO nanostructures through oxidation of a Cu substrate in alkaline solution,62 submicrometer CuO ribbons on a Cu foil through the process of oxidation-dehydration,63 CuO nanorods synthesized via a facile hydrothermal process64 and water-assisted synthesis of CuO nanourchins.65 In this work, CuO nanoparticles were prepared via the precipitation reaction between Cu(NO3) and NaOH. There are two different types of solvent suitable for the precipitation reaction, namely ethanol and aqueous solution. The precipitation of CuO nanoparticles in aqueous solution involved the chemical reaction between Cu2+ ions from Cu(NO3)2 and OH- ions from NaOH which resulted in the formation of Cu(OH)2 crystalline units at a temperature below 30 ˚C.24,62,66 Cu(OH)2 crystalline units would then further undergo hydrolysis process and convert to CuO nanostructures under a suitable condition such as alkaline environment at an elevated temperature of 58 60 ˚C.24,62,67 In fact, such synthesis approach is highly similar to the methods developed for the preparation of ZnO colloids.68,69 The chemical reactions that represent the synthesis process of CuO particle in an aqueous solution can be represented as follows: Cu(NO3)2 + 2 NaOH  Cu(OH)2 + 2 NaNO3 (2) Cu(OH)2 + heat  CuO + H2O (3) Apart from the abovementioned method, there are also several other methods to transform Cu(OH)2 into CuO, such as hydrothermal and annealing in a N2 atmosphere.70,71 However, the disadvantage of involving the Cu(OH)2 crystalline unit as the intermediate product is that prior to the dehydration process that converts Cu(OH)2 to CuO, the Cu(OH)2 crystalline units may aggregate and this would result in the formation of larger CuO particles.72 Besides, the morphology of Cu(OH)2 may also be affected by the concentration of nucleophile OH-,63 thus resulting in a complicated CuO nanostructure synthesis process. On the other hand, the synthesis approach via the CuO precipitation reaction in ethanol as solvent does not produce Cu(OH)2 as the intermediate compound. Such synthesis method was adopted in this work, whereby CuO nanoparticles were prepared via the precipitation reaction between Cu(NO3) and NaOH in absolute ethanol at an elevated temperature of 78 ˚C for 1 hour, as described in section 3.2.1. The precipitation reaction can be represented by the following reaction: Cu(NO3)2+ 2 NaOH + heat  CuO + 2 NaNO3 + 4H2O (4) The use of absolute ethanol as the solvent for Cu(NO3)2 prevents the Cu2+ ions from reacting with OH- ions, thus bypassing the formation of Cu(OH)2 crystalline units. By heating the ethanol solvent containing Cu2+ ions at 78 ˚C, CuO nanoparticles will be formed instantaneously upon the addition of NaOH to the boiling ethanol.73 Acetic acid is also one of the chemicals used in the synthesis of CuO nanoparticles, as explained in Section 3.2.1. The use of acetic acid has also been reported to be able to suppress excessive particle growth,68,73 hence maintaining the CuO particles in the nanoparticulate state. Kida et al.73 noted 59 the importance and significance of the amount of acetic acid used in the synthesis process of CuO nanoparticles. It was found that the addition of acetic acid in equimolar proportion to NaOH would result in an incomplete CuO synthesis reaction, indicated by the formation of blue precipitate of Cu(CH3COO)2 instead of black CuO suspension. This could be due to excessive dissolution of the synthesized CuO, or due to the consumption of NaOH by acetic acid. On the other hand, with the use of optimal amount of NaOH, it would lead to the increase in the rate of precipitation reaction, thus promoting the formation of CuO nuclei and at the same time minimizes the effect of excessive CuO particle growth rate. On the contrary, the use of stoichiometrically excess amount of NaOH would result in the sedimentation of CuO particles caused by the increase in rate of of CuO particle growth and agglomeration. Based on the discussion above, suitable amount of acetic acid and NaOH are the two essential factors in ensuring the formation of stable CuO colloids. Kida et al. reported that the use of molar ratio of 1:4:2 for Cu2+:NaOH:acetic acid would result in the formation of CuO colloid with a stable suspension. As a result, similar molar ratio was also used for the synthesis of CuO nanoparticles in this work. As for the synthesis of CuO-WO3 composite photocatalyst, the deposition-precipitation method is not suitable in this work. This is because such synthesis method requires WO3 to be mixed together with the synthesis precursor materials of CuO nanoparticles. OH- ion is a key material for the synthesis of CuO nanoparticles and OH- ions would result in the dissolution of WO3 as WO3 is unstable in alkaline solution.74,75 Figure 3.6 shows the colour of WO3 suspension in a solution of (a) neutral, (b) 8 and (c) 10 pH value. Figure 3.6 Colour of WO3 suspension in solution with the pH value at (a) neutral, (b) 8 and (c) 10. 60 3.3.3 Photocatalytic O2 evolution performance Several WO3 photocatalyst samples were loaded with different amount of CuO nanoparticles in order to study the influence of CuO nanoparticle loading amount on the photoactivity of the composite photocatalysts. Table -1 3.1 shows the photocatalytic activities, in terms of μmol g -1 h of the composite photocatalyst samples loaded with CuO nanoparticles ranging from 0.5 to 8 wt. %. Table 3.1 Photocatalytic O2 evolution rate of pristine WO3 and the CuO-WO3 composite photocatalyst samples loaded with 0.5, 1, 2, 4, 6 and 8 wt. % CuO nanoparticles. Sample CuO nanoparticles loading (wt. %) Activity (μmol g-1 h-1) Pristine WO3 ─ 19.66 C1 0.5 57.79 C2 1 64.51 C3 2 69.54 C4 4 75.23 C5 6 32.68 C6 8 27.11 Besides that, it is also interesting to study the effect of post-synthesis thermal treatment on the photocatalytic O2 evolution rate by the CuO-WO3 composite photocatalyst. The composite photocatalyst samples were also subjected to various calcination conditions, such as different temperature and atmosphere. In this study, 4 wt. % of CuO nanoparticles were loaded on several WO3 particles followed by subjecting the composite photocatalysts samples to thermal treatment under various conditions. The resulting photocatalytic activities of the samples are tabulated in Table 3.2 as shown below: 61 Table 3.2 Photocatalytic O2 evolution rate of CuO-WO3 composite photocatalyst samples calcined at various temperatures and atmospheres. Sample CuO nanoparticle loading (wt. %) Post-synthesis calcination condition Activity (μmol g-1 h-1) D1 4 200 °C, air 38.72 D2 4 350 °C, air 67.35 D3 4 550 °C, air 54.94 D4 4 350 °C, N2 81.22 3.3.3.1 Effect of CuO nanoparticle loading amount The effects of CuO nanoparticle loading as well as the amount loaded on WO3 were investigated by testing the O2 evolution rate of the samples. Figure 3.7 shows the O2 evolution rate of pristine WO3 and samples C1 to C6. -1 -1 O2 evolution rate (µmol h g ) 100 75.23 80 69.54 64.51 57.79 60 40 32.68 27.11 20 19.66 0 Pristine WO3 Sample C1 Sample C2 Sample C3 Sample C4 Sample C5 Sample C6 Figure 3.7 Comparison in photocatalytic O2 evolution rate between pristine WO3 and the CuO-WO3 composite photocatalyst samples loaded with 0.5, 1, 2, 4, 6 and 8 wt. % CuO nanoparticles. From Figure 3.7 it can be observed that the loading of 0.5 wt. % CuO nanoparticles on WO3 led to an increase in the O2 evolution rate to 57.79 μmol g-1 h-1 from the activity rate of 20.33 μmol g-1 h-1 for pristine WO3. The evolution rates further increased to 64.51 and 69.54 μmol g-1 h-1 for the 62 loading amount of 1 and 2 wt. %, respectively before the O2 evolution rate peaked at 75.23 μmol g-1 h-1 at 4 wt. % CuO nanoparticles loading. However, the photocatalytic activity dropped from a high 75.23 μmol g-1 h-1 to 32.68 and 27.11 μmol g-1 h-1 for the respective CuO nanoparticle loading amount of 6 and 8 wt. %. There are several reasons for such a trend of rising and followed by the subsequent fall in the O2 evolution rate by the composite photocatalyst as the CuO nanoparticle loading amount continuously increased. First and foremost, the use of CuO nanoparticle as a sensitizing material may have possibly allowed the absorption of photon originating from the entire visible region. This would then lead to an overall enhancement in the generation rate of electron and hole charge carriers by the composite photocatalyst whereas the pristine WO3 photocatalyst would have a much lower capability of utilizing the visible spectrum due to its absorption edge positioned at only around 480 nm, as shown in Figure 3.2. Secondly, the formation of p-n junction between p-type CuO and ntype WO3 also helped to separate the electron and hole charge carriers from one another.20,44-46 As a matter of fact, both the conduction and valence bands of CuO have a more negative potential as compared to the two energy bands of WO3.21,76 As a result, photoinduced free electrons excited to the conduction band of CuO would flow to the conduction band of the WO3 which is energetically lower in order for the electrons to achieve lowest possible energy state. On the other hand, the photoinduced free holes in the valence band of WO3 would travel to the valence band of CuO which has a more negative potential and energetically favourable for the free holes. Due to this reason, the free electron and hole carriers are separated more efficiently which would then contribute to reduction in the electron-hole pair recombination rate. Despite the mechanism of the electron-hole flow between CuO and WO3 as explained above, there could be possibilities of photoinduced free electrons in CuO be scavenged by Fe3+ and photoinduced free holes in WO3 undergo H2O oxidation reaction. Furthermore, the conduction band level of CuO also seems suitable for the photoreduction reaction of H2O to H2. However as the loading wt. % increases, agglomeration between the CuO nanoparticles would cause the bandgap to shrink. Such effect could result in the conduction band edge becoming more positive than the reduction potential of H+ to H2, thus 63 effectively preventing the direct transfer of photoinduced electrons from CuO to H+. The schematic diagram shown in Figure 3.8 illustrates the energy band positions for the conduction and valence bands of of CuO and WO3 and the charge carrier transfer process between the two materials due to (1) sensitization effect by CuO nanoparticle, and (2) the role of CuO in separating the photoinduced electron and hole charge carriers. However, the positions of both conduction and valence bands of CuO may not be accurate, as the bandgap value of a nanoparticle is considerably larger than its bulk counterpart due to quantum confinement effect.77-79 (a) CuO -1.69 +0.01 λ ≤ 480 nm CB VB CB WO3 electron hole VB +0.5 +3.2 Potential / vs NHE (pH = 0) Potential / vs NHE (pH = 0) visible light CB: Conduction band VB: Valence band (b) CuO -1.69 CB +0.0 1 VB CB WO3 electron hole λ ≤ 410 nm CB: Conduction band VB: Valence band 64 VB +0.5 +3.2 Potential / vs NHE (pH = 0) Potential / vs NHE (pH = 0) visible light Figure 3.8 Charge carrier transfer mechanism between the CuO and WO3 due to (a) sensitization effect by CuO nanoparticle, and (b) the role of CuO in separating the photoinduced electrons and holes. Thirdly, loading of CuO in nanoparticulate on WO3 would also contribute to the increase in the total surface area of the composite photocatalyst.80,81 This would lead to more available O2 evolution site, which naturally leads to the increase rate of O2 evolution by the composite photocatalyst, provided that the nanoparticle loading amount is appropriate. The synergetic effect of CuO nanoparticle loading discussed previously led to the gradual increase in O2 evolution rate by the composite photocatalyst as the CuO nanoparticle loading amount increased from 0 to 4 wt. %, as shown in Figure 3.7. However, continual increase in the loading amount would eventually lead to excessive coverage of WO3 surface by the CuO nanoparticles. Such effect is shown by the TEM images in Figure 3.5, with WO3 particles loaded with 20 wt. % of CuO nanoparticles. This would lead to several implications, such as: (1) The overcrowding of CuO nanoparticles loaded on WO3 would prevent or minimize interaction between the WO3 surface and the Fe3+ ions in the solution. Fe3+ ions are used as electron scavengers in photocatalytic O2 evolution reactions.82,83 Upon accepting a free electron photogenerated by a photocatalyst, the Fe3+ ion will be reduced to Fe2+ and this ensures minimal recombination between photogenerated electrons and holes. Failure to remove the photoinduced electrons due to excessive CuO nanoparticle coverage preventing the Fe3+ ions from scavenging the free electrons on WO3 surface would result in less free hole charge carriers available for the photooxidation of H2O molecules to O2, hence the lower O2 evolution rate by samples C5 and C6 with CuO nanoparticles loading amount of 6 and 8 wt. %, respectively. (2) Excessive CuO nanoparticles loading would lead to a higher reduction rate of O2 molecules to radical species by the electrons photogenerated by the CuO nanoparticles, particularly superoxide radical anions (∙O2-),47,76 which would then contribute to a reduced O2 count by the gas chromatograph. Such effect is possible is due to the fact that the redox 65 potential for the formation of ∙O2- from O2 has a more positive potential (-0.51 V vs NHE at pH = 0) than the conduction band of CuO.21 Besides, the high photogeneration rate of free electrons as a result of excessive CuO loading may also lead to the water splitting backward reaction whereby four electrons in the presence of H+ ions can cause the reduction of O2 to H2O.45,84 Based on these effects, the higher the loading amount of CuO nanoparticle, the more electrons will be generated which subsequently lead to the higher generation rate of O2 radical species as well as the reduction of O2 to H2O. In other words, the net O2 evolution activity of the CuO-WO3 composite photocatalyst sample will gradually decrease and eventually the net photoactivity rate simply becomes negative as more and more O2 are being photoreduced to ∙O2- instead of being generated by the composite photocatalyst. A separate experiment on the photocatalytic O2 evolution by sample C6 for a prolonged duration of 5 hours was conducted, and the graph for its gas evolution activity was plotted as shown in Figure 3.9. It can be observed that the total amount of evolved O2 gas steadily increased initially. After the third hour under irradiation, the O 2 level of the air space inside the quartz tube began to drop. By the fifth hour, the O2 amount had decreased to 0.56 μmol from 0.76 μmol μmol at third hour. Such drop in the O2 amount clearly indicates the photoreduction process of O2 molecules which resulted in the consumption of O2 molecues. 1 O2 content (µmol) 0.8 0.6 0.4 0.2 0 0 1 2 3 4 5 6 Time (h) Figure 3.9 Photocatalytic O2 evolution activity of sample C6 for a duration of 6 h. 66 (3) The higher the CuO nanoparticle loading amount, the more WO3 surface will be shielded or masked.52 Such undesirable effect would reduce the number of incident photons reaching the surface of WO3, which would subsequently lead to decrease in the photoexcitation rate of hole charge carriers and thereby decreasing the photocatalytic O2 evolution performance by the composite photocatalyst.86 Such effect is evident in samples C5 and C6 whereby a relatively higher CuO nanoparticle loading wt. % of 6 and 8 led to the decrease in O2 evolution rate, as compared to sample C4 which has a higher photoactivity due to lower CuO nanoparticle loading amount of 4 wt. %. (4) Higher CuO nanoparticle loading amount may lead to agglomeration between the nanoparticles, thus forming larger CuO particles.57 This would cause an increase in the formation of grain boundaries, which act as trapping or recombination sites for free charge carriers.47,87,88 Naturally, this would lead to the reduction in the availability of free charge carriers, especially holes for the photocatalytic oxidation of H2O moles for the generation of O2. 3.3.3.2 Effect of post-synthesis calcination Thermal treatment is commonly applied to reduce the defect density of several materials, for example carbon nanotube,89 silicon film90 as well as photocatalyst such as TiO291 and ZnO.92 Examples of thermal treatment environment are such as the gaseous atmosphere of H2,93,94 O2,95 plasma96,97 as well as inert environment6,97 or in a solution environment as in the case of hydrothermal treatment.98 Some of the common defects known to exist in photocatalyst are such as oxygen vacancies,99,100 OH-related defects,101 charge carriers-trapping grain boundaries,87,102 reduced species (e.g. Ti3+, Ta3+)93,103 as well as impurities in photocatalyst which act as electron-hole pair recombination centres.103,104 There have been studies on the effect of thermal annealing whereby such annealing process was reported to be a contributing factor to the increase in the photocatalytic reaction.105 Several studies reported that thermal annealing helps in minimizing defect density within the photocatalyst and thus resulted in enhanced photocatalytic efficiency.104,105 Besides, thermal annealing also helps to enhance bonding between two 67 different materials, such as between a co-catalyst and its host photocatalyst which would enable improved photocatalytic activities.106,107 In order to have a better understanding of the effect of thermal annealing on the photocatalytic O2 evolution by the CuO-WO3 composite photocatalyst, several WO3 powder samples were loaded with 4 wt. % of CuO nanoparticles followed by undergoing thermal treatment under various conditions as listed in Table 3.2. Figure 3.10 shows the comparison between sample C4 (which did not undergo any annealing process), D1, D2, D3 and D4 in terms of their O2 evolution rates. -1 -1 O2 evolution rate (µmol h g ) 100 81.22 80 75.23 67.35 60 54.94 38.72 40 20 0 Sample C4 Sample D1 Sample D2 Sample D3 Sample D4 Figure 3.10 Comparison in photocatalytic O2 evolution rate between pristine sample C4 and the CuO-WO3 composite photocatalyst samples annealed at various temperatures and atmospheres. There samples were subjected to calcination for 2 h in a static oxidizing environment, namely air at a temperature of 200, 350 and 550 °C which correspond to sample D1, D2 and D3 respectively whereas sample D4 was heated in the inert environment of N2 gas at 350°C with a gas flow rate of 50 sccm for a duration of 2 h. It can be seen that the thermal annealing process has a significant effect on the photocatalytic activity of the composite photocatalyst samples. Samples D1, D2 and D3 which were annealed in air were found to experience lower photocatalytic efficiency in terms of O2 evolution rate as compared to sample C4 which was not subjected to any post- 68 synthesis thermal treatment. It is surprising to discover such behaviour by the annealed composite photocatalysts (samples D1, D2 and D3) when air calcination would normally result in enhancement to the photocatalytic activity, as reported by Tytgat et al. for TiO2 photocatalyst108,109 and Wang et al. for ZnO.110 As a matter of fact, there have been reports on negative impacts of thermal annealing on certain photoctalytic activities. For example, Huang et al. reported that an air calcination temperature at above 900 °C would result in the morphology change of the alumina-supported CuO catalysts, which would in turn decrease the catalyst’s carbon monoxide oxidation rate.86 This is because in some cases, thermal annealing may induce inter-particle sintering or agglomeration84,111 or perhaps morphology change to the annealed particles whereby such effects could have been experienced by the CuO particles in this work and thus causing them to change from nanoparticles to microparticles.112 The first implication of particle agglomeration is the formation of grain boundaries which would result in the formation of defects serving as recombination sites for free charge carriers.47,87,111 Secondly, nanoparticles agglomeration to larger particles would inevitably reduce the surface area of the composite photocatalyst, therefore reducing the surface active sites for photocatalytic reactions and hence resulting in a reduced photocatalytic O2 evolution rate.113 Despite the undesirable effect of particle agglomeration, thermal annealing may on the other hand lead to increase in crystallinity of the photocatalyst. Highly crystalline structure would result in fewer lattice or boundary defects acting as recombination sites for photoinduced electrons and holes, thus improving the photocatalytic efficiency of the photocatalyst, and in some cases the improvement is more significant.114,115 Such effect can be used to explain the increase in the photoactivity from 38.72 μmol g-1 h-1 for sample D1 to 67.5 μmol g-1 h-1 for sample D2. However, when the annealing temperature was increased to 550 °C, the photocatalytic O2 evolution rate deteriorated to 54.94 μmol g-1 h-1. Such drop in the O2 evolution rate could most likely be due to the overwhelming negative effect of CuO nanoparticle agglomeration at high temperature calcination, which led to the reduction in the available active sites for photocatalytic reactions.114 Such effect explains the presence of CuO diffraction peaks at 2θ = 39° of sample D3 which 69 corresponds to the (11-1) plane, as shown in Figure 3.11 (c). The reason for being unable to detect the presence of other CuO diffraction peaks such as the (002) plane at 2θ = 35.6° could be due to it being superimposed by the WO3 diffraction peak for the (121) plane at 2θ = 35.7 °, whereas the other CuO diffraction peaks simply have a low signal-to-noise ratio which caused the peaks to be overlapped by the background noise during XRD scanning, hence their absence from the XRD spectrum of sample D3. On the other hand, the unannealed sample C4 with a similar CuO nanoparticle loading amount of 4 wt. % did not yield any CuO diffraction peak, as seen in Figure 3.11 (b) which is in contrast to sample D3 with similar CuO loading amount of 4 wt. %. Bandara et al.113 studied the diffraction patterns for various CuO-loaded TiO2 samples annealed at various temperatures ranging from 100 °C to 500 °C with a temperature increment step of 100 °C. It was found that as the annealing temperature increases, so does the CuO XRD peak intensity. This could be due to the fact that XRD peaks of agglomerated nanoparticles of larger sizes are easier to be detected by the XRD measurement as compared to homogeneously deposited nanoparticles, and hence the presence of CuO XRD peak in Figure 3.11 (b).56 As for the effect of the surrounding gaseous environment during the thermal annealing process, calcination in N2 atmosphere appeared to improve the photocatalytic O2 evolution rate of the composite photocatalyst and resulted in higher photoactivity as compared to the sample annealed in air at similar temperature. Sample D4 which was annealed in N2 recorded a photoactivity rate of 81.22 μmol g-1 h-1, which is approximately 21 % higher than sample D2 which was annealed in air. Sample D4 also shows a slight improvement of approximately 8 % as compared to the unannealed sample C4 which has a photoactivity rate of 75.23 μmol g-1 h-1. The reason for such favourable result by thermal annealing in N2 could be due to a milder degree of particle agglomeration, as compared to thermal annealing in air. By studying the XRD spectrum of sample D4 shown in Figure 3.11 (a), not a single CuO diffraction peak could not be detected, which very likely indicate that the CuO nanoparticles still remain in a finely dispersed state, unlike sample D3 which shows the presence of a CuO diffraction peak at around 2θ = 39 °, despite a weak one, as seen in Figure 3.11 (c). The presence of a single 70 CuO diffraction peak at 2θ = 39 ° could be due to the agglomeration of CuO nanoparticles to a larger size, thus enabling its detection during XRD scanning. A study on the effect of thermal annealing in either air or N2 on the photocatalytic performance of TiO2 nanotubes by Shaddad et al.116 showed that TiO2 nanotubes annealed at 450 °C in N2 appeared to have higher active surface area than the TiO2 nanotubes annealed in air. Shaddad et al. attributed the reason for the lower photoactivity by the sample annealed in air to the undesirable larger sintering effect between the nanotubes. Subsequently, the sample annealed in N2 resulted in higher photocatalytic O2 reduction and H2 evolution reactions as compared to the one annealed in air. Similar explanation can thus be used to describe the beneficial effect of N2 annealing on the CuOWO3 composite photocatalyst, coupled with the effect of crystallinity enhancement by thermal annealing which resulted in the enhanced 20 25 30 35 45 2θ (°) 71 50 (141) (002) (111) 40 (221) (201) (121) (021) (111) (120) Intensity (a.u.) (001) (020) (200) (a) (112 ) (202) (041) photocatalytic O2 evolution rate by sample D4. 55 60 65 70 Intensity (a.u.) (b) 20 25 30 35 40 45 50 55 60 65 70 2θ (°) 30 35 40 (202) (112) (002) (221) (111) (121) (021) 25 (111) (120) 20 45 (041) (141) CuO (201) Intensity (a.u.) (001) (020) (200) (c) 50 55 60 65 70 2θ (°) Figure 3.11 XRD spectra of (a) sample D4 annealed in N2, (b) unannealed sample C4 and (c) sample D3 annealed in air. 3.4 Conclusions In conclusion, a composite photocatalyst comprised of CuO nanoparticles and WO3 were introduced whereby CuO nanoparticles were loaded on WO3 in order to provide enhancement to the photocatalytic O2 evolution efficiency. The advantages of loading CuO nanoparticles on WO3 are such as increasing the visible light absorption range as well as enhancing 72 the separation of electrons and holes to minimize recombination between the free charge carriers, therefore increasing the photocatalytic O2 evolution rate of the composite photocatalyst. Despite these advantages, excessive loading of the nanoparticles would lead to decrease in the O2 evolution rate. As for the thermal annealing effect on the photoactivity of the composite photocatalyst, the annealing of the sample in a N2 atmosphere helped to further enhance the photocatalytic efficiency of the composite photocatalyst. References [1] S. U. M. Khan, M. Al-Shahry and W. B. Ingler Jr., Science, 2002, 297, 2243. [2] G. Liu, P. Niu, C. Sun, S. C. Smith, Z. Chen, G. Q. Lu and H. M. Cheng, J. Am. Chem. Soc., 2010, 132, 11642. [3] J. C. Yu, J. Yu,W. Ho, Z. Jiang and L. Zhang, Chem. Mater., 2002, 14, 3808. [4] K. Takanabe, K. Kamata, X. Wang, M. Antonietti, J. Kubota and K. Domen, Phys. Chem. Chem. Phys., 2010, 12, 13020. [5] V. H. Houlding and M. Grätzel, J. Am. Chem. Soc., 1983, 105, 5695. [6] R. Abe, K. Sayama and H. Arakawa, J. Photochem. Photobiol., A, 2004, 166, 115. [7] R. Abe, K. Hara, K. Sayama, K. Domen and H. Arakawa, J. Photochem. Photobiol. A, 2000, 137, 63. [8] D. R. Baker and P. V. Kamat, Adv. Funct. Mater., 2009, 19, 805. [9] W. T. Sun, Y. Yu, H. Y. Pan, X. F. Gao, Q. Chen and L. M. Peng, J. Am. Chem. Soc., 2008, 130, 1124. [10] J. Luo, L. Ma, T. He, C. F. Ng, S. Wang, H. Sun and H. J. Fan, J. Phys. Chem. C, 2012, 116, 11956. [11] X. F. Gao, H. B. Li, W. T. Sun, Q. Chen, F. Q. Tang and L. M. Peng, J. Phys. Chem. C, 2009, 113, 7531. [12] R. Liu, Y. Lin, L. Y. Chou, S. W. Sheehan, W. He, F. Zhang, H. J. M. Hou and D. Wang, Angew. Chem., Int. Ed., 2011, 50, 499. [13] S. Shamaila, A. K. L. Sajjad, F. Chen and J. Zhang, J. Colloid Interface Sci., 2011, 356, 465. [14] S. Harikrishnan and S. Kalaiselvam, Thermochim. Acta, 2012, 533, 46. [15] S. Kim, K. Kim and S. Ihm, Chemosphere, 2007, 68, 287. [16] W. Liu, L. Jia, Y. Wang, L. Song, Y. Zhu, X. Chen, Z. Zhong and F. 73 Su, Ind. Eng. Chem. Res., 2013, 52, 6662. [17] K. Zhang, Y. Yang, E.Y.B. Pun and R. Shen, Nanotechnology, 2010, 21, 235602. [18] P.O. Larsson, A. Andersson, R.L. Wallengerg and B. Svensson, J. Catal., 1996, 163, 279. [19] R. P. Wijesundera, Semicond. Sci. Technol., 2010, 25, 045015. [20] H. Fan, L. Yang, W. Hua, X. Wu, Z. Wu, S. Xie and B. Zou, Nanotechnology, 2004, 15, 37. [21] Y. Xiang, J. P. Tu, L. Zhang, Y. Zhou, X. L. Wang and S. J. Shi, J. Power Sources, 2010, 195, 313. [22] A. Chowdhuri, P. Sharma, V. Gupta, K. Sreenivas and K. V. Rao, J. Appl. Phys., 2002, 92, 2172. [23] A. Cruccolini, R. Narducci and R. Palombari, Sens. Actuators B, 2004, 98, 227. [24] V. R. Katti, A. K. Debnath, K. P. Muthe, M. Kaur, A. K. Dua, S. C. Gadkari, S. K. Gupta and V. C. Sahni, Sens. Actuators B, 2003, 96, 245. [25] A. El-Trass, H. ElShamy, I. El-Mehasseb and M. El-Kemary, Appl. Surf. Sci., 2012, 258, 2997 [26] T. Li, S. Chiang, B. Liaw and Y. Chen, Appl. Catal., B, 2011, 103, 143. [27] S. S. Lee, H. Bai, Z. Liu and D. D. Sun, Water Res., 2013, 47, 4059. [28] O. Akhavan, R. Azimirad, S. Safad and E. Hasan, J. Mater. Chem., 2011, 21, 9634. [29] M. A. Rafea and N. Roushdy, J. Phys. D: Appl. Phys., 2009, 42, 015413. [30] X. Q. Wang, G. C. Xi, S. L. Xiong, Y. K. Liu, B. J. Xi, W. C. Yu and Y. T. Qian, Cryst. Growth Des., 2007, 7, 930. [31] X. Zhang, D. Zhang, X. Ni and H. Zheng, Solid-State Electron., 2008, 52, 245. [32] S. K. Maji, N. Mukherjee, A. Mondal, B. Adhikary and B. Karmakar, J. Solid State Chem., 2010, 183, 1900. [33] X. D. Xu, M. Zhang, J. Feng and M. L. Zhang, Mater. Lett., 2008, 62, 2787. [34] M. Zhang, X. D. Xu, Z. H. Zhao, J. Feng and M. L. Zhang, J. Dispers. Sci. Technol., 2007, 28, 1223. [35] G. Zou, H. Li, D. Zhang, K. Xiong, C. Dong and Y. Qian, J. Phys. Chem. B, 2006, 110, 1632. [36] A. Nezamzadeh-Ejhieh and Z. Salimi, Appl. Catal., A, 2010, 390, 110. [37] M. Ma, S. Qing, S. Li, J. Zhu, L. Fu and R. Sun, Carbohydr. Polym., 74 2013, 91, 162. [38] G. Zhu, H. Xu, Y. Xiao, Y. Liu, A. Yuan and X. Shen, ACS Appl. Mater. Interfaces, 2012, 4, 744. [39] X. C. Jiang, Y. Herricks and Y. N. Xia, Nano Lett., 2002, 2, 1333. [40] M. Yin, C. Wu, Y. Lou, C. Burda, J. T. Koberstein, Y. Zhu and S. O’Brien, J. Am. Chem. Soc., 2005, 127, 9506. [41] P. Djinović, J. Batista and A. Pintar, Appl. Catal., A, 2008, 347, 23. [42] O. Akhavan, R. Azimirad, S. Safad and E. Hasan, J. Mater. Chem., 2011, 21, 9634. [43] M. A. Rafea and N. Roushdy, J. Phys. D: Appl. Phys., 2009, 42, 015413. [44] W. Yao, S. Yu, Y. Zhou, J. Jiang, Q. Wu, L. Zhang and J. Jiang, J. Phys. Chem. B, 2005, 109, 14011. [45] R. Kitsomboonloha, S. Baruah, M. T. Z. Myint, V. Subramanian and J. Dutta, J. Cryst. Growth, 2009, 311, 2352. [46] J. Liu, X. Huang, Y. Li, K. M. Sulieman, X. He and F. Sun, J. Mater. Chem., 2006, 16, 4427. [47] H. Hou, Y. Xie and Q. Li, Cryst. Growth Des., 2005, 5, 201. [48] M. Zhu and G. Diao, Catal. Sci. Technol., 2012, 2, 82. [49] S. Sun, X. Zhang, Y. Sun, S. Yang, X. Song and Z. Yang, ACS Appl. Mater. Interfaces, 2013, 5, 4429. [50] X. G. Wen, Y. T. Xie, C. L. Choi, K. C. Wan, X. Y. Li and S. H. Yang, Langmuir, 2005, 21, 4729. [51] S. Zeng, Y. Wang, B. Qin, X. Gu, H. Su, L. Lia and K. Liu, Catal. Sci. Technol., 2013, 3, 3163. [52] E. A. Meulenkamp, J. Phys. Chem. B, 1998, 102, 5566. [53] L. Spanhel and M. A. Anderson, J. Am. Chem. Soc., 1991, 113, 2826. [54] Y. Konishi, T. Nomura and D. Satoh, Ind. Eng. Chem. Res., 2004, 43, 2088. [55] X. Wen, W. Zhang and S. Yang, Langmuir, 2003, 19, 5898. [56] C. Zhang, Y. Bai, Y. Yin, J. Gu, and Y. Sun, Korean J. Chem. Eng., 2011, 28, 602. [57] T. Kida, T. Oka and M. Nagano, J. Am. Ceram. Soc., 2007, 90, 107. [58] S. Anandan, N. Ohashi and M. Miyauchi, Appl. Catal., B, 2010, 100, 502. [59] N. Helaïli, Y. Bessekhouad, A. Bouguelia and M. Trari, J. Hazard. Mater., 2009, 168, 484. [60] J. Y. Zheng, G. Song, C. W. Kim and Y. S. Kang, Electrochim. Acta, 2012, 69, 340. 75 [61] A. Shui, W. Zhu,L. Xu, D. Qin and Y. Wang, Ceram. Int., 2013, Article in press. [62] W. L. Ong, S. Natarajan, B. Kloostra and G. W. Ho, Nanoscale, 2013, 5, 5568. [63] J. Sun, L. Gao and Q. Zhang, J. Am. Ceram. Soc., 2003, 86, 1677. [64] J. Guo, Y. Li, S. Zhu, Z. Chen, Q. Liu, D. Zhang, W. Moon and D. Song, RSC Advances, 2012, 2, 1356. [65] Y. Sasaki, A. Iwase, H. Katoa and A. Kudo, J. Catal., 2008, 259, 133. [66] E. I. Solomon, A. J. Augustine and J. Yoon, Dalton Trans., 2008, 30, 3921. [67] M. Qamar, Z.H. Yamani, M.A. Gondal and K. Alhooshani, Solid State Sci., 2011, 13, 1748. [68] M. R. Bayati, J. Ding, Y. F. Lee, R. J. Narayan, J. Narayan, H. Zhou and S. J. Pennycook, J. Phys.: Condens. Matter, 2012, 24, 395005. [69] F. Su, J. Lu, Y. Tian, X. Ma and J. Gong, Phys.Chem. Chem. Phys., 2013, 15, 12026. [70] J. Park, M. F. P. Bifano and V. Prakash, J. Appl. Phys., 2003, 113, 034312. [71] T. Sameshima, K. Sakamoto, K. Asada, M. Kondo, A. Matsuda and S. Higashi, Sol. Energ. Mat. Sol., 2001, 65, 577. [72] H. Kominami, J. Kato, S. Murakami, Y. Ishii, M. Kohno, K. Yabutani, T. Yamamoto, Y. Kera, M. Inoue, T. Inui and B. Ohtani, Catal. Today, 2003, 84, 181. [73] A. Saffar-Teluri, S. Bolouk and M. H. Amini, Res. Chem. Intermed., 2012. [74] H. Liu, H. T. Ma, X. Z. Li, W. Z. Li, M. Wu and X.H. Bao, Chemosphere, 2003, 50, 39. [75] T. Huang and T. Yu, Appl. Catal., 1991, 71, 275. [76] S. Poulston, P. M. Parlett, P. Stone and M. Bowker, Surf. Interface Anal., 1996, 24, 811. [77] J. Han, X. Wang, N. Wang, Z. Wei, G. Yu, Z. Zhou and Q. Wang, Surf. Coat. Technol., 2006, 200, 4876. [78] H. Hsi and C. Tsai, Chem. Eng. J., 2012, 191, 378. [79] A. Sarkar, A. Shchukarev, A. Leino, K. Kordas, J. Mikkola, P. O. Petrov, E. S. Tuchina, A. P. Popov, M. E. Darvin, M. C. Meinke, J. Lademann and V. V. Tuchin, Nanotechnology, 2012, 23, 475711. [80] M. B. Radoicic, I. A. Jankovic, V. N. Despotovic, D. V. Sojic, T. D. Savic, Z. V. Saponjic, B. F. Abramovic and M. I. Comor, Appl. Catal., B, 2013, 138 – 139, 122. [81] H. Irie, Y. Watanabe and K. Hashimoto, J. Phys. Chem. B, 2003, 107, 76 5483. [82] C. Pan, J. Xu, Y. Chen and Y. Zhu, Appl. Catal., B, 2012, 115 – 116, 314. [83] Y. Li, J. Du, S. Peng, D. Xie, G. Lu and S. Li, Int. J. Hydrogen Energy, 2008, 33, 2007. [84] Y. Moriya, T. Takata and K. Domen, Coord. Chem. Rev., 2013, 257, 1957. [85] H. Jeong, T. Kim, D. Kim and K. Kim, Int J Hydrogen Energy, 2006, 31, 1142. [86] N. Wu, M. Lee, Z. Pon and J. Hsu, J. Photochem. Photobiol. A: Chem., 2004, 163, 277. [87] Z. Liu, S. Chai, A. Binder, Y. Li, L. Ji and S. Dai, Appl. Catal., A, 2013, 451. 282. [88] S. S. Kalanur, S. H. Lee, Y. J. Hwang and O. Joo, J. Photochem. Photobiol. A: Chem., 2013, 259, 1. [89] T. Tytgat, M. Smits and S. Lenaerts, Int. J. Appl. Ceram. Technol., 2013. [90] Z. Xiong, H. Dou, J. Pan, J. Ma, C. Xua and X. S. Zhao, CrystEngComm, 2010, 12, 3455. [91] H. Wang and C. Xie, J. Phys. Chem. Solids, 2008, 69, 2440. [92] A. Mills and S. Morris, J, Photochem. Photobiol. A: Chem., 1993, 71, 285. [93] J. Bandara, C. P. K. Udawatta and C. S. K. Rajapakse, Photochem. Photobiol. Sci., 2005, 6, 857. [94] H. Fu, J. Lin, L. Zhang and Y. Zhu, Appl. Catal., A, 2006, 306, 58. [95] H. Kominami, J. Kato, S. Murakami, Y. Kera, M. Inoue, T. Inui and B. Ohtani, J. Mol. Catal. A: Chem., 1999, 144, 165. [96] M. N. Shaddad, A. M. Al-Mayouf, M. A. Ghanem, M. S. AlHoshan, J. P. Singh and A. A Al-Suhybani, Int. J. Electrochem. Sci., 2013, 8, 2468. 77 Chapter 4 Loading of AgCl/Ag hybrid nanostructure on WO3 as electron-accepting co-catalyst on WO3 Abstract The loading of AgCl/Ag nanoparticle of a hybrid structure on O2producing WO3 photocatalyst as co-catalyst was performed in this work in order to study its effect on the photooxidation of H2O to O2 by the composite photocatalyst. Such hybrid structure is formed by first loading AgCl nanoparticles on WO3 followed by a partial photoreduction of the AgCl surface to metallic Ag. The Ag outer layer of the nanoparticle acts as an electron sink by trapping the photogenerated electrons in the WO3 host photocatalyst. Such effect helps to minimize the recombination rate between the electron-hole pairs, therefore enhancing the photocatalytic O2 evolution rate by the WO3 photocatalyst. Ag has been one of the conventional electronscavenging coc-atalyst used in the photocatalytic water splitting reaction. Unfortunately Ag could not be subjected to high-temperature thermal annealing process whereby such process is often applied to enhance bonding between a co-catalyst and its host photocatalyst. This is because Ag would be converted to oxide form such as AgO and Ag2O semiconductor, thus losing its capability to function as an electron sink. However, AgCl has a high thermal stability and such characteristic allows the AgCl-loaded WO3 to be annealed in order to enhance the bonding between the co-catalyst and its host photocatalyst, followed by partially photoreducing the AgCl surface to Ag. In this work, the effect of AgCl/Ag loading amount as well as thermal annealing condition on the photocatalytic O2 evolution efficiency of the photocatalyst were also investigated to obtain the optimal photoactivity performance. 4.1 Introduction The loading of co-catalyst has always been one of the important topics in the field of photocatalysis. There are numerous reported works on the loading of nanoparticulate co-catalyst, be it metal or semiconductor nanoparticles onto WO3. Nevertheless, the primary objective of such material 78 loading is to enhance the photocatalytic efficiency of the photocatalyst such as WO3 in order to enhance separation of charge carriers by either scavenging the photogenerated electrons or holes.1-4 Such effect helps to minimize the recombination rate between photoinduced electron-hole pairs, therefore resulting in more free charge carriers available to undergo photocatalytic activities. For instance, there has been report on the loading of Au nanoparticles on WO3 catalyst which resulted in enhanced sensitivity as well as selectivity of the sensing of NO2 and ethanol.5 Deposition of TiO2 nanoparticles on WO3 also leads to enhanced separation of photoinduced holes from the valence band of WO3 to TiO2 which helps to minimize the recombination between electrons and holes in WO3, hence resulting in improved photocatalytic activity.6 As for the photocatalytic reaction of H2O splitting, the most commonly used electron-accepting co-catalyst is Pt.7-9 Despite its high effectiveness in enhancing the photocatalytic performance of WO3, Pt is a rare and costly metal and with these reasons the mass production of Pt-loaded WO3 photocatalyst for commercial use may not materialize. In this chapter, an alternative electron-accepting co-catalyst will be introduced, which is in the form of AgCl/Ag hybrid nanostructure acting as an electron sink to the O2-producing WO3 photocatalyst. Such core-shell structure is routinely formed by irradiating AgCl with photons which will then partially reduce the Ag+ ions on AgCl surface to a layer of metallic Ag, hence forming the AgCl/Ag hybrid nanostructure. The metallic Ag outer layer on AgCl has a lower energy Fermi level as compared to the conduction band energy level of WO3. Such condition would encourage the photoinduced electrons in the conduction band of WO3 to flow to the energetically lower Fermi level of metallic Ag in order to achieve lowest possible energy state. By separating the photoinduced electrons from the holes in WO3, recombination between the two charge carriers would be minimized. As a result, the holes in the valence band of WO3 can effectively oxidize the H2O molecules to generate O2 gas with an enhanced efficiency. Secondly, apart from functioning as an electron-accepting co-catalyst for WO3, the AgCl/Ag hybrid nanostructure is also highly stable in a solution that contains Cl- ions, which can originate from FeCl3 when Fe3+ ions are required to be used in the photooxidation reaction of H2O to O2. Fe3+ ions are 79 commonly used as electron scavengers due to their efficient single-electron reduction process.10-12 Upon accepting an electron, Fe3+ will be reduced to Fe2+ and this ensures minimal recombination between photogenerated electrons and holes. Indeed, the use of sacrificial charge acceptor such as methanol and Fe2+ ions for holes and Fe3+ ions for electrons is instrumental in ensuring the success of photocatalytic water splitting research.11,13 Apart from the AgCl/Ag hybrid nanostructure, Ag nanoparticles have also been used as electron-accepting co-catalyst for WO3. However, they are unstable in the presence of Cl- ions as the surface of Ag will be quickly converted to AgCl.14,15 This would result in the creation of a AgCl layer covering the Ag nanoparticle within, and acts as a barrier by denying direct contact between the covered Ag and Fe3+ ions in the surrounding solution. This prevents the Fe3+ ions from removing the photogenerated electrons collected by the Ag cocatalyst from the conduction band of WO3. The excess accumulation of electrons in the Fermi energy level of Ag will eventually recombine with the photogenerated holes in the valence band of WO3. Thus, Ag is unsuitable in functioning as electron-accepting co-catalyst for the photocatalytic O2 evolution through the oxidation of H2O molecules by WO3 when the use of FeCl3 as electron scavenging agent is involved. Thirdly, AgCl-loaded WO3 could be subjected to thermal treatment, especially in air prior to the partial photoreduction process of AgCl to form the AgCl/Ag hybrid structure. This is because AgCl has a high thermal stability even when calcined at a temperature of 600 °C.16 On the other hand, calcination of Ag-loaded WO3, even at a relatively low temperature risks oxidizing the Ag nanoparticles to AgO or Ag2O. Thermal treatment is usually crucial because it allows the enhancement of the bonding between the cocatalyst and host photocatalyst,17 apart from repairing or reducing defects in the photocatalyst system which may act as recombination centres of photogenerated electron and hole charge carriers. As oxide semiconductor materials, AgO and Ag2O do not have the capability of functioning as an electron-accepting co-catalyst like metallic Ag nanoparticles. Awazu et al.18 developed a different approach by coating the Ag nanoparticles with silica (SiO2) in order to prevent the oxidation of Ag nanoparticles prior to loading them on TiO2. However, with the use of AgCl as the co-catalyst, the problem 80 of Ag oxidation can be avoided as AgCl can be annealed prior to the partial photoreduction of AgCl to form the AgCl/Ag hybrid structure. In this chapter, the photocatalyst O2 evolution rate of a composite photocatalyst comprising of WO3 loaded with AgCl/Ag nanoparticles of a hybrid structure will be reported. Besides that, several characterization techniques will also be applied to further understand the chemical and physical properties of the composite photocatalyst. Finally, the various stages involved in the synthesis process of the composite photocatalyst such as calcination process and loading weight % (wt. %) of AgCl/Ag nanoparticles will be investigated and discussed in terms of their effects on the O2 evolution efficiency by the composite phtoocatalyst. 4.2 Experimental procedures 4.2.1 Synthesis of AgCl/Ag-WO3 composite photocatalyst The AgCl/Ag-WO3 composite photocatalysts were prepared via the deposition-precipitation method followed by photoreduing the AgCl in order to form the AgCl/Ag hybrid nanostructure. Briefly, the synthesis method involves a three-step approach, beginning with the heterogeneous precipitation of AgCl nanoparticles on WO3, followed by calcination in air and finally undergoing photoreducing reaction to partially convert the surface of AgCl nanoparticles to metallic Ag. As for a detailed explanation of the composite photocatalyst synthesis process, firstly 0.3 g of WO3 with particle size less than 100 nm (Sigma Aldrich) were added into 10 mL of DI water and sonicated for 5 min followed by stirring the WO3 photocatalyst suspension vigorously in order to disperse the particles thoroughly. Next, 0.03 g of polyvinylpyrrolidone (PVP) (Sigma Aldrich) with an average molecular weight of 10,000 was added to the suspension before adding varying amount of AgNO3 (Sigma Aldrich, ≥ 99.0 %, ACS reagent) solution in order to obtain WO3 with AgCl loading wt. % of 1, 2.5, 5, 10 and 15. This was followed by adding HCl acid to the suspension under stirring. For every mole of AgNO3 added to the suspension, 3 moles of Cl- ion were added into the suspension. Slight excess amount of Cl- ion was added in order to ensure a successful conversion of Ag+ ions to AgCl 81 nanoparticles. Once HCl acid had been added, the suspension was allowed to stir in the dark for 1 h to allow a homogeneous dispersion of the AgCl nanoparticles before being dried in a water bath at a temperature of 60 °C under stirring in order to accelerate the drying process. After the drying process had been completed, the powder of the composite photocatalyst comprised of AgCl-loaded WO3 was collected and calcined in air at various temperatures of 350, 450, 550, 650 and 750 °C. The calcined powder was then added into a solution containing 9 mL of DI water and 1 mL of ethanol, followed by vigorous stirring in order to disperse the agglomerated particles. The agglomeration is an unwanted side effect of the drying process (prior to air calcination process) and the dried powder had to be dispersed in order to maximize the exposed surface area of the composite photocatalyst. Once the composite photocatalyst particles had been dispersed, the suspension was then irradiated for 5 min using a 300 W Xe lamp at an intensity of 1000 W/cm2 in order to partially reduce the Ag+ component on the surface of the AgCl nanoparticles to metallic Ag. The gradual darkening in colour of the suspension observed during irradiation indicated the partial reduction of Ag+ on the AgCl surface to Ag0, as shown in Figure 4.1. Once the photoreduction process had completed, the suspension was dried in a water bath under stirring before the AgCl/Ag-WO3 composite photocatalyst was ready to be tested for its photocatalytic performance. \ 82 (a) (b) (c) (d) (e) (f ) Figure 4.1 Colour of the composite photocatalyst suspension at 6 different stages: (a) prior to irradiation, (b) 5 s, (c) 2 min, (d) 4 min, (e) 5 min into irradiation and (f) after irradiation process. Figure 4.2 shows a flowchart representing each of the process involved in the synthesis of AgCl/Ag-WO3 composite photocatalysts. Three different states of the composite photocatalyst during the synthesis process can be categorized into three main stages: stage 1 which represents the completion of AgCl nanoparticle loading process on WO3, stage 2 as the completion of air calcination process and lastly stage 3 which shows the end of the composite photocatalyst synthesis process after the partial photoreduction of AgCl nanoparticles to AgCl/Ag hybrid nanostructure. 83 Figure 4.2 Flowchart of the systhesis process for the AgCl/Ag-WO3 composite photocatalyst. Figure 4.3 below shows the images of the composite photocatalyst at various stages during the synthesis process. Figure 4.3 Images of (a) pristine WO3 and the composite photocatalyst at stage (b) 1, (c) 2 and (d) 3. 4.2.2 Photocatalytic reactions Photocatalytic oxidation of H2O to O2 reaction was performed as a test reaction to evaluate the photocatalytic capabilities of the various AgCl/Ag84 WO3 composite photocatalyst samples. Each test reaction was carried out in a 25 mL-quartz tube containing 10 mg of the sample powder dispersed in a 7 mL aqueous solution with 50 mmol Fe3+ ions from FeCl3 as sacrificial electron acceptors. The Fe3+ ion plays several crucial roles in the photocatalytic O2 evolution reaction: (1) it acts as an electron-scavenging agent for the photoinduced electrons by the AgCl/Ag-WO3 composite photocatalyst to minimize the effect of holes recombining with the electrons, (2) Fe3+ ion prevents or minimizes the formation of superoxide radical anion (•O2−) which would reduce the amount of O2 molecules being detected by the gas chromatograph,19 and (3) Fe3+ reduces the probability of water splitting backward reaction in which O2 molecules can be reverted back to H2O molecules as a result of the reaction between O2, electrons and H+ ions. Once the appropriate materials had been placed inside the quartz tube, the glass tube was sealed with a tight-fitting rubber septum and wrapped with paraffin film in order to prevent exchange of gas between the sealed quartz tube and the outer environment. Once the quartz tube had been tightly sealed, the air space within was purged with inert gas such as Ar in order to remove any trace of foreign gasses from the quartz tube. At the same time, the pressure of the air space within was maintained at atmospheric throughout and after the purging process. This is to prevent the build-up of inner pressure which may lead to the cracking of the quartz tube as well as impeding the release of evolved O2 gas from the solution. After the purging process, the quartz tube was irradiated with a 300 W xenon lamp at an intensity of 1000 W/m2 for up to 3 h. During the irradiation process, the photocatalyst in the solution was stirred continuously with a magnetic stirrer in order to maintain the photocatalyst powder constantly in a suspended state. A gas-tight syringe with a 100 μL volume was used to draw the evolved O2 gas hourly to determine the O2 concentration by a gas chromatograph (Shimadzu GC-2014). 4.3 Results and discussion 4.3.1 Materials characterization 85 The various AgCl/Ag-WO3 composite photocatalyst samples were examined and studied under several characterization tools. For example, the morphology of the composite photocatalyst were studied under the scanning electron microscopy (SEM). The samples were also analysed with X-ray diffraction (XRD) technique in order to determine the crystallinity and the presence of AgCl and Ag in the sample. First and foremost, SEM imaging was performed on the composite photocatalyst sample in order to investigate the effect of the various synthesis processses on the morphology of the photocatalyst, in particular during the stages of pre- and post-calcination. The SEM images are as shown in Figure 4.4. Two samples were selected to undergo SEM imaging: WO3 loaded with 5 wt. % of AgCl prior to the calcination process, which is denoted as sample A and WO3 loaded with 5 wt. % of AgCl/Ag after the calcination and photoreduction processess, denoted as sample B. (a) (b) Sample A 200 nm Sample A (c) 100 nm (d) Sample B 200 nm Sample B 100 nm Figure 4.4 SEM images of AgCl/Ag-WO3 composite photocatalyst of (a), (b) sample A, and (c), (d) sample B. Small nanoparticles can be seen loaded on the surface of WO3, especially at a higher magnification as seen in Figure 4.4 (b) and (d). The small nanoparticles in Figure 4.4 (b) are the loaded AgCl nanoparticles, 86 whereas those seen in Figure 4.4 (d) are the photoreduced AgCl/Ag nanoparticle. EDX analysis was performed on both samples A and B with 10 runs of scan each. The resulting values of the atomic % for each Ag and Cl element are shown in in Table 4.1. Table 4.1 EDX analysis on the Ag and Cl elemental atomic % of sample A and B. Atomic % ratio of Ag to Cl Sample A Sample B Scan 1 1.0 2.1 Scan 2 0.9 1.9 Scan 3 1.1 1.9 Scan 4 1.0 1.7 Scan 5 1.0 2.4 Scan 6 1.1 2.4 Scan 7 1.0 2.3 Scan 8 1.0 1.9 Scan 9 1.3 2.0 Scan 10 1.2 2.1 Average 1.1 2.1 From Table 4.1 it can be observed that that sample B has a higher Ag to Cl atomic % ratio as compared to sample A. This shows the partial photoreduction of Ag+ on the surface of AgCl to metallic Ag0 which eliminated Cl atoms in process, forming the AgCl/Ag core-shell structure as a result. However, it should be noted that EDX reading may not necessarily provide accurate analysis on the elemental atomic % of the samples, hence the results should be interpreted with caution. XRD analysis was performed on pristine WO3 as well as the the AgCl/Ag-WO3 composite photocatalyst sample loaded with 10 wt. % AgCl/Ag co-catalyst and Figure 4.5 shows the XRD spectrum of the two materials. The composite photocatalyst sample with 10 wt. % loading of AgCl/Ag was selected as the intensities of the AgCl diffraction peaks for 87 samples with 5 wt. % loading and lower were too weak to be detected. The diffraction peaks for AgCl (JCPDS No.31-1238) were detected at 2θ = 27.8 °, 32.2 ° and 46.28 ° which can be assigned to the (111), (200) and (220) planes of the cubic phase of AgCl, respectively with the rest of the peaks belonging to WO3. However, the XRD analysis failed to detect the presence of Ag. Such finding contradicts the darkening in colour of the composite photocatalyst suspension after irradiation as observed by comparing Figure 4.1 (a) and (f), which indicates the partial photoreduction of Ag+ to Ag0 on the surface of AgCl to form the AgCl/Ag hybrid nanostructure. A possible reason could be due to the formation of extremely thin Ag layer that caused the XRD unable to detect the presence of Ag. Yin et al.20 experienced similar situation in their work on the preparation of Cu/Cu2O/CuO nanoscale system whereby their XRD analysis failed to detect the presence of a CuO thin layer, despite the XPS measurement indicating the presence of Cu2+. Besides, the XRD analysis on CuO-loaded WO3 in Chapter 3 also failed to detect the presence of CuO despite a high CuO nanoparticle loading amount of 20 wt. %. Such situation could also be attributed to the extremely small CuO nanoparticle size. Another aspect to be noted is the crystallinity of the composite photocatalyst. It can be observed that the diffraction peaks belonging to the composite photocatalyst are narrower in width than the peaks of pristine WO3. The narrowing of XRD diffraction peak can be atatributed to the increase in crystallinity of the material. The increase in photocatalyst crystallinity is especially beneficial for photocatalytic reactions and the details will be discussed in section 4.3.3.1. 88 (a) 20 25 30 (220) (200) (111) Intensity (a.u.) AgCl 35 40 45 50 55 60 65 70 60 65 70 2θ (°) 20 25 35 40 45 50 (041) (141) (202) (002) (221) (121) 30 (112) (021) (201) (200) (120) (111) Intensity (a.u.) (001) (020) (b) 55 2θ (°) Figure 4.5 XRD spectrum of (a) composite photocatalyst sample with 10 wt. % AgC/Ag nanoparticle loading and (b) pristine WO3. 4.3.2 Synthesis process of AgCl/Ag nanoparticle and its function In this work, the deposition-precipitation method was adopted in synthesis of AgCl-loaded WO3 composite material. In the depositionprecipitation process, the WO3 nanoparticles with a high surface area act as nucleating agent and this helps to avoid localized high concentration of Ag+ 89 ions.21 Strong ionic and Van der Waals interaction between WO3 and Ag+ ions help to restrict the diffusion of Ag+ ions, coupled with using of small WO3 particles (diameter < 100 nm) with high surface area further enhance the dispersion of Ag+ ions on WO3 surface. After the process of dispersing Ag+ ions, HCl acid was added to the suspension under vigorous stirring in order to synthesize AgCl nanoparticles which occurred as a result of precipitation reaction between the Ag+ and Cl- ions. Such precipitation reaction is possible due to the low solubility of AgCl, which has a solubility product constant of 1.77 10-10 at 25 °C in water.22 The objective in stirring the suspension vigorously during the AgCl nanoparticle deposition-precipitation synthesis process is to ensure a thorough and homogeneous dispersion of the AgCl nanoparticles during the in-situ precipitation reaction. By avoiding or minimizing the localization of high concentration of AgCl nuclei, the effect of Ostwald ripening process of AgCl nuclei will then be reduced, which may otherwise result in the formation of large AgCl particles. As for the role of PVP, it is commonly used in the synthesis of stable nanoparticles such as nanosized Ag colloids,23 Pt nanocrystals,24 Au nanocatalysts25 as well as Au and Ag nanoshells on silica nanoparticles.26 In this work, the primary role of PVP during the composite photocatalyst synthesis process was to serve as a capping agent for the Ag+ ions in order to disperse the ions homogeneously on the surface of WO3 through steric stabilization which would result in the homogeneous dispersion of AgCl nanoparticles on WO3 surface as well as preventing the agglomeration of AgCl nanoparticles.27,28 In addition to that, the long polymeric chains of PVP molecules also contributed to the increase in the overall viscosity of the reaction solution which helped to slow down the AgCl precipitation reactions.29 The low precipitation facilitated in the formation of AgCl of nanoparticulate state, which may bring favourable effects to the photocatalytic performance of the composite photocatalyst.28 After the deposition-precipitation synthesis process of AgCl nanoparticles loaded on WO3, the composite material was subjected to thermal annealing in air. One of the primary objectives of thermal treatment was to remove the NO3- ions resulted from the use of AgNO3 precursor by decomposing the ions at elevated temperature to H2O, N2 and O2.30 Apart from 90 the elimination of NO3- ions, the effects of thermal annealing process on the photocatalytic O2 evolution performance by the composite photocatalyst were also part of the research objectives for this work which will be discussed in detailed in section 4.3.3.1. After the thermal annealing process, the synthesis process was followed by the partial photoreduction of the AgCl surface to metallic Ag in order to produce the AgCl/Ag hybrid nanostructure. Electrons are the primary source for the reduction of Ag+ to Ag0. There are two sources for the photogeneration of electrons, which is from AgCl itself and WO3. AgCl has an indirect energy bandgap value of 3.25 eV and is able to photoexcite electrons by absorbing UV light.31 The photoexcitation of an electron to the conduction band is accompanied by the generation of a hole in the valence band of a photocatalyst. In order to minimize the effect of electron-hole pair recombination so as to increase the rate of photoreduction process from Ag+ to Ag0 by the electrons, ethanol can be used to scavenge the holes.32 AgCl alone has the capability to generate electron-hole pairs upon absorption of UV light due to its indirect energy bandgap value of 3.25 eV.31 The photoinduced electrons are able to facilitate the partial reduction of Ag+ ions on the surface of AgCl to metallic Ag0. However, such reaction is only limited to the AgCl surface exposed to the outer environment with access to photons from the UV light source. With the AgCl loaded on WO3, a AgCl/Ag semi core-shell structure will be produced with the AgCl surface in direct contact not being reduced to Ag. However, with the contributions of electrons from WO3, the side of AgCl interfacing with WO3 can also be reduced to form a layer of metallic Ag sandwiched between AgCl and WO3. This would enable a more efficient of electron transferring process from WO3 to Ag. The mechanism of the AgCl photoreduction process as described above is represented by the schematic diagram shown in Figure 4.6. 91 (a) (b) Figure 4.6 Photoreduction of AgCl nanoparticle loaded on WO3 to the AgCl/Ag hybrid nanostructure with and without ethanol. 4.3.3 Photocatalytic O2 evolution rate The photocatalytic performances of the various AgCl/Ag-WO3 composite photocatalysts synthesized under different conditions were evaluated based on their oxidation rate of H2O to O2. Table 4.2 shows the photoactivity of eleven different samples including the pristine WO3 which served as a reference for the enhancement effect by the AgCl/Ag hybrid nanostructure. Table 4.2 Photocatalytic O2 evolution rate for the 10 samples synthesized with various calcination temperatures of AgCl/Ag loading wt. %. Sample AgCl/Ag loading amount (wt. %) Calcination temperature (˚C) O2 evolution rate (µmol h-1 g-1) Pristine WO3 - - 48.99 A1 5 350 37.84 A2 5 450 63.96 A3 5 550 86.74 A4 5 650 60.48 A5 5 750 65.22 B1 1 550 77.38 B2 2.5 550 108.83 B3 5 550 86.74 B4 10 550 trace B5 15 550 trace 92 There are two types of variable in the study of photocatalytic O2 evolution efficiency of the composite photocatalyst samples, namely the calcination temperature and the co-catalyst loading wt. %. For the study on the effect of calcination temperature, the O2 evolution rate steadily rose from 37.84 µmol h-1 g-1 at 350 ˚C to 63.96 µmol h-1 g-1 at 450 ˚C and peaked at the O2 evolution rate of 86.74 µmol h-1 g-1 at 550 ˚C. Beyond 550 ˚C, the photoactivity of the composite photocatalyst samples dropped to 60.48 and 65.22 µmol h-1 g-1 at the calcination temperatures of 650 and 750 ˚C, respectively. As for the study on the effect of AgCl/Ag co-catalyst loading wt. %, the O2 evolution rate rose markedly from 77.3 µmol h-1 g-1 at a loading of 1 wt. % to 108.83 µmol h-1 g-1 at 2.5 wt. %. However, with the increase in the co-catalyst loading amount to 5 wt. %, a dramatic drop in the photocatalytic rate of the composite photocatalyst to 86.74 µmol h-1 g-1 was observed. Any further loading of AgCl resulted in almost zero photocatalytic O2 evolution activity by the composite photocatalyst. From the photocatalytic performances by the various composite photocatalyst samples shown in Table 4.2, it can be seen that the loading of AgCl/Ag hybrid nanostructure as electron sink had contributed to the increase in O2 evolution rate for all the composite photocatalyst samples except for samples B4 and B5, regardless of the synthesis condition upon comparison with the O2 evolution rate by pristine WO3. In the following sections, the effect of air calcination as well as the effect of AgCl/Ag co-catalyst loading amount on the photocatalytic performance of the samples will be discussed. 4.3.3.1 Effect of calcination on AgCl/Ag-WO3 The thermal treatment on the composite photocatalyst prior to the final synthesis process of partial AgCl photoreduction had brought enhancement to the photocatalytic O2 evolution rate by the majority of the photocatalyst samples. The O2 evolution performances of the different samples annealed at various temperatures are represented in the form of bar chart as shown in 93 Figure 4.7 in order to provide a clearer comparison between the various photocatalyst samples. 100 86.74 -1 -1 O2 evolution rate (µmol h g ) 120 80 63.96 60 60.48 65.22 48.99 37.84 40 20 0 Pristine WO3 350 °C 450 °C Sample A1 Sample 550 °C 650 °C Sample A3 Sample A4 750 °C Sample A2rate among prisinte WO3 and samples A5 Figure 4.7 Comparison in O2 evolution A1 to A5 post-calcined at various temperatures. Thermal treatment is commonly applied to reduce the defect density in materials such as carbon nanotube and silicon film.33,34 Some of the common defects known to exist in photocatalyst are such as oxygen vacancies,35,36 OHrelated defects,37 charge carriers-trapping grain boundaries38,39 as well as impurities in photocatalyst which act as electron-hole pair recombination centres.40 Studies have shown the synergetic effect of thermal annealing in the form of enhanced photocatalytic reaction.41 Several studies have reported that thermal annealing helps in reducing defects and increasing the crystallinity of the photocatalyst which would in turn reduce grain boundaries, facilitate in the transfer of charge carriers and minimize the frequency of electron-hole pair recombination.40-43 Such findings may help to explain the gradual increase in the O2 generation rate with the increasing calcination temperature which eventually peaks at 550 ˚C. From Figure 4.7 it can be seen that the O2 evolution rate initially rose together with the increase in calcination temperature. A possible explanation for such trend is the reduction in AgCl nanoparticle size as the annealing temperature increases. Yamashita et al.16 discovered that with increasing 94 calcination temperature, smaller AgCl nanoparticle size was obtained as observed by the widening of an AgCl XRD peak at 2θ = 32.2 °. In order to verify the observation made by Yamashita et al., samples A1, A2 and A3 were studied under XRD. As shown in Figure 4.8, the widening of AgCl XRD diffraction peaks at around 2θ = 32.2 ° can be observed, albeit at a less noticeable scale. This shows that the AgCl nanoparticles of sample A3 were of the smallest in size as compared to samples A1 and A2. AgCl/Ag nanoparticles of a smaller size would be more thoroughly dispersed on the surface of its host photocatalyst WO3, which would then enable the synergistic effect of AgCl/Ag nanoparticles as electron sink to be better felt across the (021) AgCl WO3 Intensity (a.u.) (201) WO3 surface. Sample A1 (200) Sample A2 Sample A3 30 31 32 33 34 35 2θ (°) Figure 4.8 Width of AgCl diffraction peaks for samples A1, A2 and A3 at around 2θ = 32.2 °. As mentioned in section 4.2.2, Fe3+ ions play a crucial role as an electron scavenger to minimize the photogenerated electrons accumulating in the Ag electron sink from recombining with the hole carriers in the photocatalyst. In order to function as an effective electron scavenger, Fe3+ ions must be able to access the accumulated electrons in the Ag as easy as possible. AgCl/Ag nanoparticles of smaller sizes are able to produce higher effective surface area to volume ratio as compared to bulk particles. This would enhance the availability of the active sites for the Fe3+ ions to scavenge the accumulated electrons, thus increasing the photocatalytic efficiency of the composite photocatalyst through the minimization of electron-hole pair recombination rate. Apart from that, well-dispersed AgCl/Ag nanoparticles 95 with narrow diameter range also aid in the scattering of absorbed photon by the WO3 host photocatalyst. Nanoparticles such as Au, Ag and SiO2 have been actively used as light scatterer in order to increase the effective optical path length of the absorbed light by confining the propagating light within the lightabsorbing material.44,45 By using AgCl/Ag nanoparticles decorated on the surface of WO3, the light-harvesting effect by WO3 would be enhanced, hence increasing the rate of electron-hole pair generation. A wide array of applications benefiting from such effect are such as quantum-well solar cell,46 dye-sensitized solar cell,45 Surface Enhanced Raman Scattering (SERS)47 and photoelectrochemical water splitting.48,49 Another plausible explanation for the optimal photocatalytic performance at a calcination temperature of 550 ˚C could be closely linked to the decomposition temperature of PVP. Failure in removing PVP would cause the polymer to blanket the surface of WO3 and enwrap the AgCl/Ag cocatalyst nanoparticles. The direct consequence of the WO3 photocatalyst being blanketed is that the O2 evolution process would be hindered, whereas the blocking of the co-catalyst surface would prevent the Fe3+ ions from effectively eliminating the electrons accumulated in Ag. In order to understand the thermal decomposition behavior of PVP, Du et al.50 applied the Thermal Analysis - Differential Thermal Analysis (TGDTA) method to investigate the weight loss of pure PVP and PVP coated on Pt nanoparticles by heating the samples in a N2 environment from room temperature up to 600 °C. It was found that pure PVP began to decompose at 380 °C and fully decomposed at approximately 460 °C. Separately, in this work an experiment was designed whereby various PVP powder samples were subjected to heat treatment in air at different temperatures, namely 350, 450 and 550 ˚C for 2 h. Figure 4.9 shows the PVP powder in its pristine state and the resulting effect of the various thermal treatment conditions on PVP. 96 (a) (b) (c) (d) Figure 4.9 Images of PVP at its (a) pristine state, and calcined at (b) 350 ˚C, (c) 450 ˚C and (d) 550 ˚C. It is clearly shown that PVP did not decompose completely at the calcination temperatures of 350 and 450 ˚C, as seen by the black residue after the process of thermal treatment. Nevertheless, the calcination temperature at 450 ˚C resulted in a higher degree of PVP decomposition than the temperature at 350 ˚C. Such statement is made based on the condition of the calcined PVP at 450 ˚C which had been reduce to the flaky state, whereas the calcination at 350 ˚C only managed to blacken or char the PVP powder. As for the calcination temperature of 550 ˚C, no trace of PVP was visible after the heating process. This indicates that 550 ˚C is sufficient in ensuring a complete decomposition of PVP. By relating these experimental observations to the photocatalytic O2 evolution by samples A1, A2 and A3, the lower O2 evolution rate for samples A1 and A2 could be due to the incomplete decomposition of PVP that remained partially enwrapping the AgCl/Ag and WO3 nanoparticles. The presence of burned PVP is especially prominent in sample A1, as seen by the colour of the sample shown in Figure 4.10 (b). The severity of not removing PVP is more prominent for sample A1 which shows 97 a dramatic dip in its O2 evolution rate as compared to pristine WO3. On the other hand, sample A3 which was calcined at 550 ˚C was completely relieved of PVP did not experience the problem of PVP polymer acting as a barrier layer in denying OH- ions access to photo-generated holes on WO3 surface. As for samples A4 and A5 which were calcined at 650 and 750 ˚C, respectively, their lower O2 evolution rates as reported contradict the beneficial effect of high temperature thermal treatment in reducing defect density and eliminating PVP through thermal decomposition as explained in prior discussion. By looking at the powder colour of samples A4 and A5 as shown in Figure 4.10 (e) and (f), it is noticeable that the powder colour of both samples was inhomogeneous with a mixture of bright yellow and green. It is possible that heat treatment at 650 and 750 ˚C may have led to possible chemical reactions that could have altered the chemical and physical properties of the composite photocatalyst. For example, high-temperature annealing could have resulted in the partial diffusion of Ag+ or Cl- ions into the lattice system of WO3. This in turn resulted in the formation of doped WO3 or a different form of solid solution with different physical and chemical properties, such as changes to the physical colour appearance as seen in Figure 4.10 (e) and (f) as well as its photocatalytic O2 generation efficiency. Lastly, thermal annealing may have also led to the agglomeration of particles, such as between the WO3 particles or the AgCl/Ag co-catalyst nanoparticles.51,52 The implications of particle agglomeration are: (2) formation of grain boundaries which would lead to the formation of defects serving as recombination sites for free charge carriers,39,51,53 and (2) reduction in the composite photocatalyst surface area which would decrease the surface active sites for photocatalytic reactions and hence a reduced photocatalytic O2 evolution rate.54 98 (a) (b) (c) (d) (e) (f) Figure 4.10 Images of (a) pristine WO3 and samples (b) A1, (c) A2, (d) A3, (e) A4 and (f) A5. 4.3.3.2 Effect of AgCl/Ag co-catalyst loading amount The loading amount of AgCl/Ag co-catalyst loaded on WO3 in terms of wt. % also considerably influences the O2 photocatalytic evolution efficiency of the composite photocatalyst. Figure 4.11 shows the O2 evolution rate of pristine WO3 and samples B1 to B5 for various amount of AgCl/Ag cocatalyst loading. 99 120 108.83 -1 -1 O2 evolution rate (µmol h g ) 140 100 86.74 77.38 80 60 48.99 40 20 0 Pristine WO3 1 wt. % 2.5 wt. % 5 wt. % Sample B1 Sample B2 Sample B3 trace 0 trace 0 10 wt. % 15 wt. % Sample Sample B5 B4 B1 to B5. Figure 4.11 Comparison in O2 evolution rate among samples It can be observed that a mere 1 wt. % of AgCl/Ag nanoparticle loading was sufficient to enhance the O2 evolution rate to 77.38 µmol h-1 g-1 from 48.99 µmol h-1 g-1 for pristine WO3. Such hike in the photoactivity represents an approximate 58 % improvement in the photocatalytic efficiency. With 2.5 wt. %, the evolution rate peaked at 108.83 µmol h-1 g-1 with an enhancement of 1.2 times. However, with further increase in loading wt. %, the O2 evolution rate dropped to 86.75 µmol h-1 g-1 for the loading amount of 5 wt. %. Small amount of trace O2 were detected for samples B4 and B5 loaded with 10 and 15 wt. % of AgCl/Ag nanoparticle, respectively. The trace O2 could be due to the O2 leakage from external atmosphere outside the quartz tube which contained the composite photocatalyst suspension. There are several plausible reasons for the drop in O2 evolution rate with increasing AgCl/Ag co-catalyst loading amount. First of all, a WO3 surface overcrowded with AgCl/Ag nanoparticles would result in the undesirable shielding or masking of the surface of WO3. Such effect reduces surface exposure to the external environment, thus hindering the incident photons from reaching WO3. As a result, the photoexcitation rate of hole charge carriers would be reduced whereby holes are necessary for the photooxidation of H2O to O2.55 100 Secondly, excessive amount of metallic AgCl/Ag co-catalyst loading on the composite photocatalyst may lead to the unwanted reduction of O2 molecules. As described earlier, the metallic Ag layer on the surface of AgCl acts as an electron sink by collecting the electrons photogenerated by WO3, leaving the holes in the valence band of WO3. This helps to separate the electrons and holes, thus minimizing the effect of electron-hole pair recombination. However, with the increase in the amount of AgCl/Ag nanoparticle loaded on WO3, the tendency of evolved O2 molecules generated on the WO3 surface coming into contact with the metallic Ag layer on AgCl co-catalyst would be increased as well. This would cause some of the O2 molecules to be reduced, thus leading to the generation of radical species such as •O2−, •O − and other reactive oxygen species56-60. Some of these reactive species may also resulting in water splitting backward reaction by reacting with H+ ions to form H2O molecules.61 These undesirable effects would inevitably result in a reduced O2 count by the GC, hence a drop in the recorded O2 evolution rate. 4.4 Conclusions In conclusion, a AgCl/Ag-WO3 composite photocatalyst was introduced for the generation of O2 gases through photocatalytic reaction. The composite photocatalyst was prepared via the deposition-precipitation method whereby AgCl co-catalyst was synthesized and immediately loaded onto the surface of the host photocatalyst WO3. This was followed by annealing the composite material before partially photoreducing the loaded AgCl nanoparticles to form the AgCl/Ag hybrid nanostructure. The metallic Ag layer on AgCl serves as an electron sink whereby it helps to minimize the rate of electron-hole pair recombination by separating the photoinduced electrons from the holes in the valence band of WO3. The composite photocatalyst was the evaluated for its performance in oxidizing H2O to O2 gases. It was found that two parameters of the synthesis process affected the photocatalytic efficiency of the composite photocatalyst considerably, namely the temperature of the annealing process and the loading amount of AgCl/Ag cocatalyst. The experimental results showed that the composite photocatalyst 101 annealed at 550 ˚C with a AgCl/Ag loading wt. % of 2.5 provided the optimal O2 evolution rate among the various experimental parameters studied in this work. References [1] M. Higashi, K. Domen and R. Abe, J. Am. Chem. Soc., 2012, 134, 6968. [2] K. Maeda, D. Lu and K. Domen, Chem. Eur. J., 2013, 19, 4986. [3] K. Sekizawa, K. Maeda, K. Domen, K. Koike and O. Ishitani, J. Am. Chem. Soc., 2013, 135, 4596. [4] J.H. Hsieh, C.C. Chang, Y.K. Chang and J.S. Cherng, Thin Solid Films, 2010, 518, 7263. [5] Y. Shim, H. G. Moon, D. H. Kim, L. Zhang, S. Yoon, Y. S. Yoon, C. Kang and H. W. Jang, RSC Adv., 2013, 3, 10452. [6] S. Anandana and M. Miyauchi, Chem. Commun., 2012, 48, 4323. [7] R. Abe, K. Sayama and H. Sugihara, J. Phys. Chem. B, 2005, 109, 16052. [8] S. S. K. Ma, K. Maeda, R. Abe and K. Domen, Energy Environ. Sci., 2012, 5, 8390. [9] X. Liu, F. Wang and Q. Wang, Phys. Chem. Chem. Phys., 2012, 14, 7894. [10] H. Arakawa and K. Sayawa, Res. Chem. Intermed., 2000, 26, 145. [11] J. Guo, Y. Li, S. Zhu, Z. Chen, Q. Liu, D. Zhang, W. Moon and D. Song, RSC Advances, 2012, 2, 1356. [12] K. Maeda, M. Higashi, D. Lu, R. Abe and K. Domen, J. Am. Chem. Soc., 2010, 132, 5858. [13] Y. Sasaki, A. Iwase, H. Kato and A. Kudo, J. Catal., 2008, 259, 133. [14] B. Ma, J. Guo, W. Dai and K. Fan, Appl. Catal., B, 2013, 130 – 131, 257. [15] G. Luo, X. Jiang, M. Li, Q. Shen, L. Zhang and H. Yu, ACS Appl. Mater. Interfaces, 2013, 5, 2161. [16] Y. Yamashita, N. Aoyama, N. Takezawa and K. Yoshida, Environ. Sci. Technol., 2000, 34, 5211. [17] Z. Liu, S. Chai, A. Binder, Y. Li, L. Ji and S. Dai, Appl. Catal., A, 2013, 451. 282. [18] K. Awazu, M. Fujimaki, C. Rockstuhl, J. Tominaga, H. Murakami, Y. Ohki, N. Yoshida and T. Watanabe, J. Am. Chem. Soc., 2008, 130, 1676. 102 [19] A. Nezamzadeh-Ejhieh and Z. Salimi, Appl. Catal., A, 2010, 390, 110. [20] M. Yin, C. Wu, Y. Lou, C. Burda, J. T. Koberstein, Y. Zhu and S. O’Brien, J. Am. Chem. Soc., 2005, 127, 9506. [21] M. Á. Centeno, I. Carrizosa and J. A. Odriozola, Appl. Catal., A, 2003, 246, 365. [22] D. R. Lide, CRC Handbook of Chemistry and Physics, 88th edn., CRC, 2008 [23] K. Chou and Y. Lai, Mater. Chem. Phys., 2004, 83, 82. [24] Y. Borodko, S. E. Habas, M. Koebel, P. Yang, H. Frei and G. A. Somorjai, J. Phys. Chem. B, 2006, 110, 23052. [25] J. Zhang, Y. Yuan, K. J. Kilpin, Y. Kou, P. J. Dyson and N. Yan, J. Mol. Catal. A: Chem., 2013, 371, 29. [26] A. M. Brito-Silva, . . Sobral- ilho, . Barbosa-Silva, C. B. de Ara o, A. alembeck and A. . Brolo, Langmuir, 2013, 29, 4366. [27] D. Chen, T. Li, Q. Chen, J. Gao, B. Fan, J. Li, X. Li, R. Zhang, J. Sun and L. Gao, Nanoscale, 2012, 4, 5431. [28] Y. Tang, Z. Jiang, G. Xing, A. Li, P. D. Kanhere, Y. Zhang, T. C. Sum, S. Li, X. Chen, Z. Dong and Z. Chen, Adv. Funct. Mater., 2013, 23, 2932. [29] S. Peng and Y. Sun, J. Mater. Chem., 2011, 21, 11644. [30] M. Outokesh, M. Hosseinpour, S. J. Ahmadi, T. Mousavand, S. Sadjadi and W. Soltanian, Ind. Eng. Chem. Res., 2011, 50, 3540. [31] J. Tejeda, N. J. Shevchick,W. Braun, A. Goldmann and M. Cardona, Phys. Rev. B, 1975, 12, 1557. [32] A. Pérez-Lariosa, R. Lopez, A. Hernández-Gordillo, F. Tzompantzi, R. Gómez and L.M. Torres-Guerra, Fuel, 2012, 100, 139. [33] C. Hu, Y. Lan, J. Qu, X. Hu and A. Wang, J. Phys. Chem. B, 2006, 110, 4066. [34] J. Park, M. F. P. Bifano and V. Prakash, J. Appl. Phys., 2013, 113, 034312. [35] T. Sameshima, K. Sakamoto, K. Asada, M. Kondo, A. Matsuda and S. Higashi, Sol. Energ. Mat. Sol., 2001, 65, 577. [36] X. Zhang, H. Li, X. Cui and Y. Lin, J. Mater. Chem., 2010, 20, 2801. [37] J. Wang, D. N. Tafen, J. P. Lewis, Z. Hong, A. Manivannan, M. Zhi, M. Li and N. Wu, J. Am. Chem. Soc., 2009, 131, 12290. [38] C. Pan, J. Xu, Y. Chen and Y. Zhu, Appl. Catal., B, 2012, 115 – 116, 314. [39] M. R. Bayati, J. Ding, Y. F. Lee, R. J. Narayan, J. Narayan, H. Zhou and S. J., J. Phys.: Condens. Matter, 2012, 24, 395005. [40] Y. Li, J. Du, S. Peng, D. Xie, G. Lu, S. Li, Int. J. Hydrogen Energ., 2008, 33, 2007. 103 [41] H. Jeong, T. Kim, D. Kim and K. Kim, Int. J Hydrogen Energ., 2006, 31, 1142. [42] F. Amano, O. O. Prieto-Mahaney, Y. Terada, T. Yasumoto, T. Shibayama and B. Ohtani, Chem. Mater., 2009, 21, 2601. [43] Y. Bu, Z. Chen, W. Li and J. Yu, ACS Appl. Mater. Interfaces, 2013, 5, 5097. [44] N. Wu, M. Lee, Z. Pon and J. Hsu, J. Photochem. Photobiol. A, 2004, 163, 277. [45] Y. H. Fu, A. I. Kuznetsov and A. E. Miroshnichenko, Nat. Commun., 2013, 4, 1527. [46] X. Yu, H. Xu, L. Xin, X. Wang, Y. Liu, X. Zhou, B. Li, W. Zhao and H. Shen, J. Mater. Chem. A, 2013, 1, 6175. [47] D. Derkacs, W. V. Chen, P. M. Matheu, S. H. Lim, P. K. L. Yu and E. T. Yub, Appl. Phys. Lett., 2008, 93, 091107. [48] Z. Chen, R. Liu, Y. Wang, H. Zhu, Z. Sun, T. Zuo, X. Chang, F. Zhao, G. Xing, H. Yuan, J. Xiang and X. Gao, J. Nanosci. Nanotechnol., 2010, 10, 8538. [49] B. Iandolo, T. J. Antosiewicz, A. Hellmana and I. Zorića, Phys. Chem. Chem. Phys., 2013, 15, 4947. [50] Y. Wei, L. Ke, J. Kong, H. Liu, Z. Jiao, X. Lu, H. Du and X. W. Sun, Nanotechnology, 2012, 23, 235401. [51] A. Mills and S. Morris, J, Photochem. Photobiol. A, 1993, 71, 285. [52] X. Xiao, R. Hu, C. Liu, C. Xing, C. Qian, X. Zuo, J. Nan and L. Wang, Appl. Catal., A, 2013, 140–141, 433. [53] A. Nezamzadeh-Ejhieh and Z. Salimi, Appl. Catal., A, 2010, 390, 110. [54] J. Bandara, C. P. K. Udawatta and C. S. K. Rajapakse, Photochem. Photobiol. Sci., 2005, 6, 857. [55] M. Qamar, Z.H. Yamani, M.A. Gondal and K. Alhooshani, Solid State Sci., 2011, 13, 1748. [56] Y. K. Du, P. Yang, Z. G. Mou, N. P. Hua and L. Jiang, J. Appl. Polym. Sci., 2006, 99, 23. [57] X. Yin, W. Que, Y. Liao, H. Xie and D. Fei, Colloids Surf., A, 2012, 410, 153. [58] S. S. Soni, M. J. Henderson, J. F. Bardeau and A. Gibaud, Adv. Mater., 2008, 20, 1493. [59] M. R. Hoffmann, S. T. Martin,W. Choi and W. Bahnemann, Chem. Rev., 1995, 95, 69. [60] J. Yu, L. Qi and M. Jaroniec, J. Phys. Chem. C, 2010, 114, 13118. [61] J. Yu, G. Dai and B. Huang, J. Phys. Chem. C, 2009, 113, 16394. 104 Conclusion A total of three different photocatalytic materials are presented in this thesis, namely TaON, CuO-WO3 nanocomposite and AgCl/Ag hybrid nanostructure loaded on WO3. Chapter 2 specifically discuss on the synthesis of visible light-absorbing TaON photocatalyst through the alloying of Ta2O5 with nitrogen via the urea route approach. This helps to enhance the absorption band edge from 320 nm of Ta2O5 to 550nm of TaON, thus allowing the photo-excitation of electronhole pairs by visible light. However, its stability in aqueous environment is crucial which enables the photocatalyst to continuous produce H in the long term. There have been reports on TaON decomposition during photocatalytic reaction, albeit at a negligible rate. Prolonged photocatalytic experiment duration could be performed to determine the suitability of TaON as a H2generating photocatalyst in a commercial scale. In Chapter 3 and 4, electron-scavenging co-catalysts such as CuO & AgCl/Ag hybrid nanostructure were introduced to WO3 with the primary objective of enhancing the photo-oxidation rate of H2O by WO3. However, the shortcoming of such design is the composite photocatalyst’s over-reliance on Fe3+ ions which help to remove excessive electrons from the electronscavenging co-catalysts, as well as minimizing the formation of superoxide radical anions and reducing the rate of water splitting backward reaction. Unfortunately, Fe3+ is vulnerable to oxidation process and may slowly react with oxygen to form oxides. This will affect the availability of Fe3+ ions, hence jeopardizing the photocatalytic performance of the composite photocatalyst. Besides, the bulkier iron oxide molecules will also serve as a light barrier, reducing the amount of photons reaching the composite photocatalysts. Further work is necessary to enhance the long-term stability of Fe3+ ions such as through modifying the condition of the surrounding aqueous solution. 105 [...]... intensive and hence costly Based on these reasons, the alternative approach for H2 production through the photocatalytic splitting of water reactions seems to be a more feasible option for the purpose of clean energy generation 2 1.2 Mechanism of water splitting The main processes in photocatalytic water splitting consist of three steps: (1) light/photon absorption with energies larger than the bandgap of... external bias Photocatalytic water splitting is also commonly known as artificial photosynthesis due to the fact that water splitting process involves direct solar energy conversion to chemicals on heterogeneous photocatalysts Upon absorbing photons, photocatalysts are able to chemically split water molecules (H2O) to H2 and O2 gases The significance of photocatalytic water splitting is the generation of... undergo water splitting reaction, discovered by Fujishima and Honda in the year 1972.3 Both Fujishima and Honda first demonstrated the overall water splitting reaction (i.e., simultaneous generation of both H2 and O2 gases) by using a photoelectrochemical cell consisting of a single-crystalline rutile TiO2 anode and a Pt cathode under ultraviolet (UV) irradiation with an external bias Photocatalytic water. .. represent the charge transfer processes involved in a typical Z-scheme system for the photocatalytic water splitting reaction using Fe3+/Fe2+ as the redox mediators: At H2-producing photocatalyst: (9) (10) At O2-producing photocatalyst: (11) (12) Despite the many advantages of Z-scheme process in photocatalytic water splitting reaction, it also suffers from several drawbacks First of all, the co-existence... of a photocatalyst has to have a minimum value of 1.23 eV which corresponds to the photon wavelength of approximately 1100 nm in the near infrared region, before the photocatalytic water splitting reaction can occur The overall water splitting reaction, as simple as it may seem, is in fact a thermodynamically uphill reaction with a large positive change in the Gibbs free energy (ΔG˚) of +238 kJ/mol... As a result, there exist an activation barrier in the charge-transfer process between the photocatalyst and the water molecules Thus, photon with energy greater than the bandgap value of the photocatalyst is normally necessary in order to enable and drive the overall photocatalytic water splitting process Apart from requiring a suitable energy bandgap value and appropriate valence and conduction band... host photocatalyst, thus enhancing the photocatalytic efficiency by increasing the amount of free charge carriers available to undergo more photocatalytic reactions 1.3.2 Morphology modification Morphology modification is one of the most studied methods for the enhancement in photocatalytic efficiency Some of the morphology types used by photocatalysts in various photocatalytic applications are nanoparticle,... of active sites available.39,40 For instance, C3N4, which can be used for photocatalytic H2 evolution, has a surface area of 10 m2 g-1 in its bulk particle form.41 On the other hand, C3N4 nanosheet has a significantly higher surface area of 84.2 m2 g-1 and such characteristic would naturally lead to higher photocatalytic water splitting rate, as reported by Wang, et al and Chen et al 41,42 1.3.3 Doping... metal or non-metal, that function as co-catalysts in order to improve the photocatalytic activity of photocatalysts For example, reduced graphene oxide (rGO) is currently one of the most active materials under research to provide enhancement for photocatalytic activities such as photodegradation of organic compound60 as well as water splitting process.61,62 rGO has a unique structure and properties such... (10) The overall water splitting reaction can thus be represented as: (11) Eq (11) can also be explained as such: upon absorbing four photons, the photocatalyst is able to chemically split two H2O molecules to a single O2 molecule and two H2 molecules There are several basic criteria that a particular photocatalyst has to fulfil in order to undergo the reaction of photocatalyst water splitting First ... positions for photocatalytic water splitting reactions, TaON is also non-toxic and relatively stable during photooxidation and photoreduction of water. 7 Apart from photocatalytic water splitting. .. through the photocatalytic splitting of water reactions seems to be a more feasible option for the purpose of clean energy generation 1.2 Mechanism of water splitting The main processes in photocatalytic. .. List of symbols xi Chapter Introduction to photocatalytic water splitting 1.1 Introduction 1.2 Mechanism of water splitting 1.3 Methods in enhancing photocatalytic efficiency 1.3.1 Sensitization