The use of products from CO2 photoreduction for improvement of hydrogen evolution in water splitting

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 6 5 4 6 e6 5 5 2

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

The use of products from CO2 photoreduction for improvement of hydrogen evolution in water splitting Xiaoyi Yang a,*, Tiancun Xiao b, Peter P. Edwards b a b

Department of Thermal Energy Engineering, Beihang University, 37 Xueyuan Rd, Haidian, Beijing 100191, PR China Inorganic Chemistry Laboratory, University of Oxford, South Parks Rd, OX1 3QR, UK

article info

abstract

Article history:

CuO/TiO2 photocatalysts were prepared and shown to enhance the rate of CO2 photore-

Received 19 December 2010

duction and the production of total organic carbon (TOC), including HCOOH, HCHO and

Received in revised form

CH3OH. Resulting TOC could act as electron donors for enhancing visible light hydrogen

16 February 2011

evolution from Pt/TiO2 photocatalysts. The impacts on CO2 photoreduction were investi-

Accepted 21 February 2011

gated including the effect of Cu dopant, pH, irradiation time and using Na2SO3 as a sacri-

Available online 24 March 2011

ficial agent, and those on hydrogen evolution was also studied including TOC

Keywords:

CO2 reduction potentials were discussed. CuO/TiO2 and Pt/TiO2 photocatalysts were

CO2 photoreduction

characterized by X-ray diffraction, Raman spectroscopy and diffuse reflection UVevis

Hydrogen evolution

spectrophotometry. Both photocatalysts showed a visible light response in comparison

concentration and Pt doping. The CO2 photoreduction mechanisms with respect to pH and

Copper doping

with pure TiO2. The photocatalytic experiments and FT-IR spectra indicated that photo-

Pt doping

product desorption was the rate-limiting step in the CO2 photoreduction.

Sacrificial agent

Copyright ª 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

TOC

1.

Introduction

The transformation and storage of solar energy into chemical energy is a promising approach for providing renewable energy. Solar radiation can induce transitions between electronic energy levels, which can be utilized in the photocatalytic reduction of carbon dioxide and water splitting for hydrogen production. Efficient conversion of sunlight to hydrogen by splitting water through direct photocatalysts is a major milestone for a viable hydrogen economy. However, photocatalysts generally suffer from low intrinsic hydrogen production efficiency due to the facile recombination of photo-generated electron-hole pairs. In addition, water oxidation by holes is also a slower process than its reduction by electrons [1]. Therefore, organic sacrificial electron donors have been widely used to reduce recombination and enhance

hydrogen evolution [2]. However, organic sacrificial electron donors require an additional energy input, and hence water is now considered the best choice as a hole scavenger [3]. Photoreduction of CO2 has also been attracting the increasing research attention due to its potential in mitigating the greenhouse effects and fossil fuel shortages. Moreover, it is also a promising future technology since CO2 can be reduced to useful compounds by solar energy at room temperature and ambient pressure. Possible reduction products including CO, HCOOH, HCHO, CH3OH or CH4, have been obtained by photoreduction of CO2 or aqueous carbonate. Anpo [4,5] indicated that CH4, CH3OH, and CO were the major CO2 photocatalytic products with highly dispersed titanium oxide anchored onto Vycor glass. In the presence of CO2 and H2O, Ikeue et al. [6] reported that Ti-containing porous silica thin films catalysed the formation of CH4, CH3OH and CO as

* Corresponding author. Tel./fax: þ86 10 82317346. E-mail address: [email protected] (X. Yang). 0360-3199/$ e see front matter Copyright ª 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2011.02.116

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 6 5 4 6 e6 5 5 2

a minor product under UV irradiation. Tseng et al. [7] showed that copper doping on TiO2 promoted CO2 photoreduction efficiency and improved the product selectivity toward methanol. There are other factors influencing selectivity and efficiency of CO2 photoreduction, including solvent, dissolved CO2 concentration, pH and selection of sacrificial agent. In a study investigating the role of solvent [8], increasing the dissolved CO2 concentration led to the major reduction products of formate and carbon monoxide. The product ratio could be controlled by the solvent dielectric constant. In a study of pH effect, Ku et al. [9] reported that carbonate photoreduction proceeded faster in acidic solutions than in alkaline solutions. Concerning sacrificial agent, Yoneyama [10] demonstrated that negative charges favored CO2 photoreduction to formate, while positive charges favored CO production. Conditions affecting the charge of sacrificial agents influence the CO2 photoreduction behavior. Different pathways have been investigated to enhance CO2 photoreduction efficiency and the transfer of inorganic carbon to organic carbon by solar energy. How to separate the various by-products and put them to practical use remains an important question. A logical but challenging approach is the photoproduct of HCOOH, CH3OH and HCHO from CO2 photoreduction, which could be utilized as sacrificial electron donors for enhancing hydrogen evolution. This is the main theme of the current study.

2.

Material and methods

2.1.

Preparation of photocatalysts

2.1.1.

TiO2

TiO2 nanoparticles were prepared by the soleemulsionegel method [11] using tetrabutyl titanate (Ti(OC4H9)4) as the precursor, cyclohexanol as the oil phase, distilled water as water phase, cetyltrimethyl ammonium bromide (CTAB) as surfactant and triethylamine as the gelling reagent. TiO2 was calcined at 500  C for 2 h.

2.1.2.

CuO/TiO2

An aqueous TiO2 suspension containing Cu(NO3)2 was sonicated for 1 h. The temperature was increased to 95  C and the suspension stirred for a further 4 h. The powder was collected by filtration and heated to 150  C at a rate of 3  C/min. It was then calcined at 500  C with a heating rate change of 3  C/min. After 2 h calcination, the sample was cool to the ambient temperature.

2.1.3.

Pt/TiO2

An aqueous-methanol solution (H2O:MeOH ¼ 99:1 by vol) TiO2 suspension containing PtCl2 (Pt:TiO2 ¼ 0.5%) was sonicated for 20 min and then heated to 75  C [12]. After stirring for 1 h, the suspension was irradiated by a Hg lamp for 20 h under sonication [13], whereby Pt ion nanoparticle was deposited onto the TiO2 surface. The powder was collected by filtrating, washed twice with distilled water, and heated to 200  C for 2 h to remove any residual methanol.

2.2.

6547

Photocatalytic reactions

TiO2 or Cu/TiO2 photocatalyst powder (50 mg) was dispersed in 100 ml of distilled deionised water containing Na2CO3 (0.05%) and Na2SO3 (0.1%) and sonicated for 10 min. The pH was adjusted to 3 by hydrochloric acid and the suspension placed in a quartz glass reactor (210 ml) and saturated with CO2 by bubbling for a period of 30 min and then the reactor was sealed and irradiated using a 250 W Hg lamp with onesixth of the lamp irradiation being utilized, as shown in Fig. 1. All reactors with lamps were immersed in a sonication tank with circulating water for cooling purpose. After 6 h irradiation and then filtering for recovery of CO2 reduction photocatalyst, Pt/TiO2 powders (50 mg) were dispersed in filtered solution (100 ml) from CO2 photoreduction within a pyrex glass reactor (210 ml). The reactor was then thoroughly deoxygenated with argon and sealed, and the suspension irradiated using the side of a 300 W Xe lamp with a 420 l filter. Formic acid and formaldehyde concentration in the liquid were analyzed by gas chromatography (GC, DX ICS3000). Hydrogen and methane concentrations in the gas, and methanol and formaldehyde concentrations in the liquid were all analyzed by gas chromatograph (HP7890). TOC concentration was analyzed by TOC instrument (SHIMADZU TOC-VWP).

2.3.

Sample characterization

Crystalline size and crystal structure of the prepared doped TiO2 samples were determined by X-ray diffraction (XRD) using a Philips X-PERT Pro Alpha 1 diffractometer, operated with Cu Ka radiation (l ¼ 1.5406  A) at a tube current of 40 mA and a voltage of 45 kV. Data were collected over a 2q range from 20 to 80 at a speed of 1 /min. Diffuse reflectance UVevis spectra were recorded with a UVevis spectrophotometer (U-3310) equipped with an integrating sphere. Laser Raman spectra were obtained using a PerkineElmer Ramanstation 400F Raman spectrometer. FT-IR spectra were obtained on a NICOLET 560 Fourier transform infrared spectrophotometer in the range of 4000e400 cm1 at 0.1 cm1 resolution.

3.

Results and discussion

3.1.

Factors effecting CO2 photoreduction

CO2 is a thermodynamically stable molecule. Its photoreduction mechanism is quite complex with a multi-electron transfer process, and results in several by-products with one to eight electrons transferred. Potentials of possible reaction products versus normal hydrogen electron are shown in Fig. 2 [14,15]. The photo activity of TiO2 as a semiconductor is controlled by its band gap, band potentials and charge transfer efficiency. The potential of the electron acceptor is thermodynamically required to be lower than TiO2 conduction band potential, while the potential of the electron donor should be greater than that of TiO2 valence band. Moreover, the photoexcited electron-hole pairs require the rapid transfer of charge to photocatalyst active surface sites to minimize their recombination. The driving force for the electron transfer is the energy

6548

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 6 5 4 6 e6 5 5 2

A-A

1

source

A-A

H2 CO3

HCO3

-

CO3

2-

Percen t ag e, %

0.8 Reactor

UV lamp

0.6

0.4

UV la mp

0.2

0 3

5

7

9

11

13

pH

Fig. 1 e Experimental setup for CO2 photoreduction and H2 evolution.

Fig. 3 e Variation of carbonate ion species with solution pH.

difference between the conduction band of the semiconductor and the reduction potential of the acceptor. The TiO2 conduction band energy is 0.26 V (E0 (red) ¼ 0.26 V). Consequently, photoexcited electrons can be effectively utilized to reduce CO2 and carbonate species when the potential of reduction reaction is below 0.26 V. The species of carbonate ion varies with solution pH, as shown in Fig. 3. H2CO3 is the dominant ion at pH  3.0. H2CO3 and HCO3 are the dominant ion species at 3.0  pH  6.3, while CO32 species predominates at pH  10.3. By comparison with the photoreduction potential, H2CO3 could be more readily reduced through multi-electron transfer than CO32. This hypothesis was confirmed by the photo experiment results, given in Fig. 4. The products from CO2 photoreduction include HCOOH, HCHO and CH3OH in liquid phase, and the total concentration of those can be expressed as the concentration of total organic carbon (TOC). As TOC was utilized as the sacrificial electron donors for hydrogen evolution, obtaining maximum TOC was the main aim in CO2 photoreduction. Na2SO3 was chosen as a sacrificial agent to minimize the electron-hole pair recombination. The influence of initial concentrations of Na2CO3 (0e1%) and Na2SO3

(0e0.2%) on TOC yield was investigated, and optimal concentrations were observed at 0.05% Na2CO3 and 0.1% Na2SO3, respectively. The optimal pH was 3.0 and the TOC concentration was 3.52 mmol/h gcat. CuO has been previously demonstrated to enhance the efficiency of CO2 photoreduction [16,17]. Copper sites can serve as electron trappers, and suppress electron-hole pair recombination. In addition, coppers as active sites also play an important role in exciting CO2 and accordingly increasing the photo efficiency. The effect of copper loading on TOC yield is shown in Fig. 5a. The TOC concentrations increased with Cu loading up to 3 wt%, and decreased with further Cu loading, which could be explained that the shading effects reduced the TiO2 photoexcitation capacity. The effect of irradiation time on product formation is shown in Fig. 5b, from which three stages were apparent. In the initial 2 h, called a TOC fast-increase stage, TOC concentration increased quickly while methane content increased very slowly. The second stage from 2 h to 6 h, called a TOC slow-increase stage, TOC increased slowly while methane increased slightly faster. The third stage after 6 h irradiation time, called a TOC decrease stage, TOC content decreased while methane increased more rapidly. Actually, the whole photocatalytic reaction can be separated into several steps including light absorption, charge transport to photocatalyst surface, photoreaction with adsorbed reactants at the photocatalyst surface and photoproduct desorption from photocatalyst surface. As the

-0.8

-0.6

CO2 /HCOOH = -0.61 CO2 /HCHO = -0.52 CO2 /CO = -0.48

Po ten tial, V

-0.4

-0.2

CO2 /CH3 OH = -0.38 CO2 /CH4 = -0.24 H2 CO3 /HCOOH = -0.166

0

H2 CO3 /HCHO = -0.05 H2 CO3 /CH3 OH = 0.044

0.2

2-

CO3 /CH3 OH = 0.209 2-

-

CO3 /HCOO = 0.311 0.4 2-

2-

CO3 /C2 O4 = 0.478 0.6

Fig. 2 e Thermodynamic potential versus normal hydrogen electrode (NHE) for possible CO2 photoreduction products.

Fig. 4 e The effect of pH on TOC formation.

6549

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 6 5 4 6 e6 5 5 2

b

a 4

TOC Formic acid Methane

4.5

Methanol Formaldehyde

Concentration, mmol/gcat

TOC, mmol/h.gcat

4 3

2

1

0

3.5 3 2.5 2 1.5 1 0.5 0

0

1

2

3

4

5

0

2

4

CuO, wt% pH = 3

6

8

10

12

Time,h

Fig. 5 e Effects of (a) copper dosages on TOC formation and (b) irradiation time on CO2 photoreduction.

3.2.

Photocatalyst characterization

The XRD diffraction patterns of as-prepared TiO2 nanoparticles showed a well-crystallized anatase phase, as shown in Fig. 7. The pattern of Pt/TiO2 showed characteristic anatase peaks, which indicated that nanoparticle crystallinity had not been changed by doping with Pt. Ion species usually occupy vacant TiO2 surface sites undetectable by XRD at low ions loading while can form crystalline regions detectable by XRD at high loading when ion loading exceeds dispersion capacity.

Therefore, it can be inferred that Pt species were highly dispersed on TiO2 surface. For CuO/TiO2 nanoparticles, characteristic anatase peaks were also evident in the XRD diffraction patterns, but the plane peaks of TiO2 (103) and TiO2 (112) were obscured by a broadened TiO2 (004) plane peak. In general, this broadening may have been due to defect formation, lattice strain, and the formation of amorphous layers on the particle surface. Crystalline CuO was detected by XRD when the CuO loading was above 3%, where all vacant sites had been occupied. At 3% CuO loading, CuO had a diffraction peak at 2q 35.6 which was near those of TiO2 (103) and TiO2 (112). Their overlapping resulted in the broadening of peaks at 2q 36.5 e39.5 . The measured Raman spectrum shows that the asprepared TiO2 nanoparticles are well-crystallized in the anatase structure, given in Fig. 8. Such a conclusion is in good agreement with XRD results. Three Raman peaks at 143, 196, and 638 cm1 were assigned to Eg modes of anatase phase, which peak at 397 cm1 was assigned to the B1g mode. The peak at 515 cm1 was a doublet of A1g and B1g modes. The lowest frequency Eg mode at 143 cm1 was the strongest of all observed peaks for anatase TiO2 nanoparticles and arose from external vibration which indicated that the long-range order was formed. Raman spectra of as-prepared Pt/TiO2 and CuO/TiO2 nanoparticles also showed a typical well-known anatase phase. In comparison with pure as-prepared TiO2, the

2000 water TOC = 109 mg/L TOC = 180 mg/L

1500 H2, µmol/gcat

product desorption is the rate-limiting step in the photosynthetic methanol formation by CO2 and H2O [18], the experimental results can be explained that incipient methanol and other products had not desorbed from photocatalyst surface to release active sites for further CO2 photoreduction and accordingly TOC was oxidized by excited holes or further reduced to CH4 by excited electrons. At the beginning of the photoreduction, carbonate species were initially adsorbed on the photocatalyst surface and were then photoreduced to methanol and other organic compounds, resulting in TOC increase. Upon further photoreaction, formic acid and other TOC contents became adsorbed on surface active sites and consequently resulted in a decrease in CO2 photoreduction. Among the organic compounds, methanol was identified as the primary product, and a small amount of other organic compounds were also identified. Pt/TiO2 [12,19] photocatalysts have been shown to enhance the hydrogen evolution. Pt dispersion results in the formation of a Schottky barrier at the metal and metal oxide semiconductor interface, which leads to a decrease in exciton recombination and more efficient charge separation. In addition, deposition of Pt on TiO2 also increases the reaction rate by decreasing the overpotentials for hydrogen evolution. TOC species can be oxidized by TiO2 photocatalyst, thus acting as electron donors for the photocatalytic hydrogen evolution. The addition of sacrificial reagents as electron donors can prevent rapid exciton recombination and backward photoreactions. From Fig. 6, the hydrogen production rate increased with TOC concentration. The hydrogen production rates were 119.6 mmol/h gcat at TOC ¼ 109 mg/L and 128.2 mmol/h gcat at TOC ¼ 180 mg/L, while it was 69 mmol/h gcat at pure water.

1000

500

0 0

3

6

9

12

15

Time,h

Fig. 6 e Effect of TOC concentration on hydrogen evolution.

6550

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 6 5 4 6 e6 5 5 2

Fig. 7 e XRD patterns of TiO2, Pt/TiO2, CuO/TiO2.

lowest frequency Eg mode exhibited blueshift and linewidth shortening, and the intensity of the high-frequency Raman peaks reduced. Raman peak shift and linewidth change were attributed to a combined mechanism involving non-stoichiometric oxygen deficiencies, disorder induced by minor phases and phonon confinement effects [20]. Specially, the weakness of B1g mode and Eg mode at 397 cm1 indicated a lack of shortrange order in the anatase phase. Consequently, doping metals on TiO2 surface induced short-range disorder and resulted in a weakening of Raman peak. FT-IR spectra of CuO/TiO2 before and after irradiation are shown in Fig. 9. Molecular water coordinated to Ti4þ resulted

in a very broad OeH stretching absorption because of different types of isolated hydroxyl groups [21]. For original CuO/TiO2 nanoparticles, a very broad absorption occurred at 3700e2600 cm1, which was assigned to the superposition of the OeH stretching mode of interacting hydroxyl group which could be involved in hydrogen bonds. The peak at 1629 cm1 was assigned to the molecular water bending mode. A broad strong absorption at 400e850 cm1 was due to TieO and another at 1623 cm1 was assigned to TieO stretching modes [22]. A weak peak at 2342 cm1 was assigned to adsorbed CO2 [23]. Small peaks at 1300e1400 cm1 were assigned to carbonate ions adsorbed on TiO2 surface.

Fig. 8 e Raman spectra of TiO2, Pt/TiO2 and CuO/TiO2.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 6 5 4 6 e6 5 5 2

6551

Fig. 9 e FT-IR spectra of CuO/TiO2 photocatalyst (a) original; (b) after 6 h irradiation.

After 6 h irradiation, a new peak at 1109 cm1 indicated the formation of TieOeC species [22]. Strong peaks appeared at 1000e1300 cm1 and were attributed to the formation of CeO bond. A broadened peak at 1398 cm1 indicated the formation of HCOOH. CO2 was reduced to formic acid and remained adsorbed. This was in agreement with the photo experiments, which indicated that the organic carbon desorption was the limiting step in CO2 photoreduction. The CeH absorptions usually appear at 2800e3000 cm1, and thus the absorptions at 2850, 2910 and 2950 cm1 were ascribed to CeH stretching and inferred the formation of CH3OH. A new weak absorption of formaldehyde was evident at 1111 cm1. FT-IR results indicated that formation of HCOOH, HCHO, CH3OH which agreed well with GC results. Diffuse reflectance UVeVis spectra of three photocatalysts were given in Fig. 10. The spectrum of TiO2 showed strong absorption at 315e325 nm and weak absorption at 400e800 nm. Upon doping with Pt or CuO, the most intense absorption still remained in the UV region, but considerable absorption also tailed off into the visible range. The photocatalyst band gaps were determined by spectroscopy, using the (KubelkaeMunk) KM formalism and Tauc plot [18]. The original coordinates of the spectra were transformed into KubelkaeMunk function vs. photon energy. For a semiconductor, a plot of [ahg]1/2 against hg should show a linear region just above the optical absorption edge. The

band gap can then be estimated by extrapolation of the linear region, and the calculated band gap for pure TiO2 was 3.18 eV. Those of Pt/TiO2 and CuO/TiO2 were 3.15 ev and 2.98 ev, respectively. It was inferred that transition metal doping was responsible for visible light response.

4.

Conclusions

The effects of initial concentrations of Na2CO3 and Na2SO3 on TOC yield in CO2 photoreduction were investigated using CuO/ TiO2 photocatalyst. The highest accumulated TOC of 3.52 mmol/h gcat at pH ¼ 3.0 was apparent at 3 wt% Cu doping. FT-IR spectra indicated the formation of HCOOH, HCHO, CH3OH in CO2 photoreduction, which were in agreement with GC results. The TOC products as sacrificial electron donors were shown to enhance hydrogen evolution from Pt/TiO2 photocatalyst, which was 128.2 mmol/h gcat at TOC ¼ 180 mg/L and was 69 mmol/h gcat at pure water. Band gaps of Pt/TiO2 and CuO/TiO2 were calculated to be 3.15 ev and 2.98 eV, respectively.

Acknowledgments This work was financed by theRoyal Academy of Engineering and the China Scholarship Council.

references

Fig. 10 e Diffuse reflectance UVevis spectra of TiO2, Pt/TiO2, CuO/TiO2.

[1] Bandara J, Udawattab CPK, Rajapakse CSK. Highly stable CuO incorporated TiO2 catalyst for photocatalytic hydrogen production from H2O. Photochem Photobiol Sci 2005;4(11): 857e61. [2] Yang X, Salzmann C, Shi H, Wang H, Green MLH, Xiao T. The role of photoinduced defects in TiO2 and its effects on hydrogen evolution from aqueous methanol solution. J Phys Chem A 2008;112(43):10784e9. [3] Guan G, Kida T, Yoshida A. Reduction of carbon dioxide with water under concentrated sunlight using photocatalyst

6552

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 6 5 4 6 e6 5 5 2

combined with Fe-based catalyst. Appl Catal A 2003;41(4): 387e96. Anpo M, Yamashita H, Ichihashi Y, Ehara S. Photocatalytic reduction of CO2 with H2O on various titanium oxide catalysts. J Electroanal Chem 1995;396(5):21e6. Anpo M, Chiba K. Photocatalytic reduction of CO2 on anchored titanium oxide catalysts. J Mol Catal 1992;74(1e3): 207e12. Ikeue K, Nozaki S, Ogawa M, Anpo M. Characterization of self-standing Ti-containing porous silica thin films and their reactivity for the photocatalytic reduction of CO2 with H2O. Catal Today 2002;74(2):241e8. Tseng I, Chang W, Wu JCS. Photoreduction of CO2 using solegel derived titania and titania-supported copper catalysts. Appl Catal B 2002;37(1):37e48. Liu B, Torimoto T, Matsumoto H, Yoneyama H. Effect of solvents on photocatalytic reduction of carbon dioxide using TiO2 nanocrystal photocatalyst embedded in SiO2 matrices. J Photochem Photobiol A 1997;108(3):187e92. Ku Y, Lee W, Wang W. Photocatalytic reduction of carbonate in aqueous solution by UV/TiO2 process. J Mol Catal A: Chem 2004;212(1e2):191e6. Yoneyama H. Photoreduction of carbon dioxide on quantized semiconductor nanoparticles in solution. Catal Today 1997; 39(3):169e75. Zhou W, Cao Q, Tang S. Effects on the size of nano-TiO2 powders prepared with soleemulsionegel method. Powder Technol 2006;168(1):32e6. Ikuma Y, Bessho H. Effect of Pt concentration on the production of hydrogen by a TiO2 photocatalyst. Int J Hydrogen Energy 2007;30(7):687e91. Abe R, Sayama K, Arakawa H. Significant influence of solvent on hydrogen production from aqueous I3/I redox solution

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

using dye-sensitized Pt/TiO2 photocatalyst under visible light irradiation. Chem Phys Lett 2003;379(3e4):230e5. Yahaya AH, Gondal MA, Hameed A. Selective laser enhanced photocatalytic conversion of CO2 into methanol. Chem Phys Lett 2004;400(3):206e12. Pan P, Chen Y. Photocatalytic reduction of carbon dioxide on NiO/InTaO4 under visible light irradiation. Catal Commun 2007;8(10):1546e9. Slamet, Nasution HW, Purnama E, Kosela S, Gunlazuardi J. Photocatalytic reduction of CO2 on copper-doped Titania catalysts prepared by improved-impregnation method. Catal Commun 2005;6(5):313e9. Tseng I, Chang W, Wu JCS. Effects of solegel procedures on the photocatalysis of Cu/TiO2 in CO2 photoreduction. J Catal 2004;221(2):432e40.  CK, ObalovA L, MatCjovA L, PlachA D, LacnS Z, Koa JirkovskS J, et al. Effect of TiO2 particle size on the photocatalytic reduction of CO2. Appl Catal B 2009;89(2): 494e502. Kozlova EA, Korobkina TP, Vorontsov AV. Overall water splitting over Pt/TiO2 catalyst with Ce3þ/Ce4þ shuttle charge transfer system. Int J Hydrogen Energy 2009;367(1):130e7. Du YL, Deng Y, Zhang MS. Variable-temperature Raman scattering study on anatase titanium dioxide nanocrystals. J Phys Chem Solids 2006;67(11):2405e8. Megan Winter DHXY, Shyamala Rajagopalan AKJK. Defining reactivity of solid sorbents: what is the most appropriate metric. Chem Mater 2009;21(12):2367e74. Zhang J, Wang B, Ju X, Liu T, Hu T. New observations on the optical properties of PPV/TiO2 nanocomposites. Polymer 2001;42(8):3697e702. Venkov T, Hadjiivanov K. FTIR study of CO interaction with Cu/TiO2. Catal Commun 2003;4(4):209e13.