- Email: [email protected]
Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation
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 x x x ( 2 0 1 6 ) 1 e6
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation Stavroula Sfaelou a, Lucian-Cristian Pop a, Olivier Monfort a,1, Vassilios Dracopoulos b, Panagiotis Lianos a,b,* a b
Department of Chemical Engineering, University of Patras, 26500 Patras, Greece FORTH/ICE-HT, P.O. Box 1414, 26504 Patras, Greece
article info
abstract
Article history:
WO3 photoanodes have been constructed by a simple soft chemistry procedure that pro-
Received 6 November 2015
duced efficient mesoporous nanocrystalline films. These photoanodes absorbed visible
Received in revised form
light and could be efficiently employed for photoelectrochemical hydrogen production
15 January 2016
under electric bias. The current increased and the rate of hydrogen production more than
Accepted 15 February 2016
tripled in the presence of a small quantity of ethanol showing that such photoanodes may
Available online xxx
be successfully used in alternative photoelectrochemical installations for solar fuel production by consumption of organic wastes.
Keywords:
Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
Tungsten trioxide
reserved.
Water splitting PhotoFuelCells Hydrogen production
Introduction WO3 is one of the most popular metal oxide semiconductor photocatalysts. It is being studied for several decades [1e4] as an alternative to the UVA absorbing titania. WO3 has a bandgap of 2.5e2.8 eV, therefore, it absorbs visible light up to 500 nm, which accounts for 12% of the solar radiation on the surface of the earth [4]. WO3 is an n-type indirect semiconductor. It is relatively easy to synthesize and deposit on electrodes, it has a moderate hole-diffusion length (~150 nm [4]), it exhibits resistance against photocorrosion and it
demonstrates satisfactory chemical stability at relatively low pH values. For this reason, WO3 has been studied as a photoanode material for photoelectrochemical water splitting applications [1e7]. Its valence band is located approximately at þ2.8 V vs NHE, therefore, it is well placed for water oxidation. Its conduction band is located at positive potentials (approximately þ0.2 tο þ0.3 V vs NHE), therefore, a bias is necessary in order to guide photogenerated electrons to the counter electrode and produce hydrogen by water or proton reduction. The expected theoretical solar to hydrogen efficiency for WO3 photoanodes is about 4.8%, based on its range of light absorption [7] (the corresponding value is only 2.2% for
* Corresponding author. Department of Chemical Engineering, University of Patras, 26500 Patras, Greece. E-mail address: [email protected] (P. Lianos). 1 Permanent address: Department of Inorganic Chemistry, Faculty of Natural Sciences, Comenius University in Bratislava, Mlynska Dolina, 84215 Bratislava, Slovakia. http://dx.doi.org/10.1016/j.ijhydene.2016.02.063 0360-3199/Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
2
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 x x x ( 2 0 1 6 ) 1 e6
titania [7]). For an unbiased cell, where the only input is incident radiation, the photoconversion efficiency, also frequently expressed as solar to hydrogen efficiency, STH, is calculated by the following equation [8]: hSTH ¼
JðmA$cm2 Þx1:229ðVÞ x 100% IðmW$cm2 Þ
(1)
Where J is the current density flowing in the external circuit of the photoelectrochemical cell, I is the intensity of the incident radiation and 1.229 V corresponds to the potential for water splitting at pH zero. Obviously, for solar hydrogen efficiency equal to 4.8% and I ¼ 100 mW cm2 (one sun), the maximum theoretical current density is 3.9 mA cm2. In practice, the obtained values of current density are much smaller so that a strong bias is necessary to reach substantial values, as it will be also shown in the present work. As a matter of fact, due to extensive electronehole recombination in WO3 photoanodes, biases larger than 1.3 V vs RHE are necessary in order to produce substantial currents [7,9e11] and this holds true also for the present data. In order to improve efficiency of WO3 photoanodes, a lot of effort has been spent to synthesize WO3 in a large variety of nanostructures. Comprehensive reviews of such efforts have been presented in Refs. [5] and [7]. Water oxidation by WO3 photoanodes is energetically allowed. However, what is of practical importance in photoelectrochemical water splitting is hydrogen production. In this respect, hydrogen production by oxidation of organic substances is more efficient than pure water splitting [3]. This matter, as far as WO3 is concerned, is rather neglected. For this reason, in the present work, we have investigated hydrogen production both in the absence and in the presence of ethanol, which is being used as a fuel, i.e. a sacrificial electron donor. Thus, the studied photoelectrochemical cells behave as PhotoFuelCells [12e14] set up for hydrogen production. The results showed that a satisfactory hydrogen production rate can be reached with WO3 photoanodes. Ethanol in the present work is employed as model fuel, as in previous works [12e14]. However, other organic substances, for example organic wastes may be also employed thus offering the double environmental benefit of renewable hydrogen production with simultaneous environmental remediation.
Experimental Materials Unless otherwise indicated, reagents were obtained from Sigma Aldrich and were used as received. Millipore water was used in all experiments. SnO2:F transparent conductive electrodes (FTO, Resistance 8 U/square) were purchased from Pilkington (USA) and carbon paste Elcocarb C/SP from Solaronix (Switzerland).
Preparation of electrodes WO3 photoanodes High purity tungsten powder with average particle sizes up to 10 microns (0.4 g, 99.99%) reacted with aqueous hydrogen
peroxide (3 ml, 30%) [15] under sonication for 2e3 h until a transparent colorless solution was obtained. Sonication decreases reaction time to 2e3 h while magnetic stirring necessitates 12e16 h. The excess of H2O2 was catalytically decomposed overnight using a Pt foil [16,17]. Then, 3 ml ethanol and 0.3 g of surfactant Triton X-100 were added to the solution. An FTO glass was cut in the appropriate dimensions and was carefully cleaned first with soap and then by sonication in acetone, ethanol, and water. The WO3 film was prepared by coating the FTO electrode with the above solution by doctor blading. The films were sintered for 10 min at 500 C (heating rate of 20 C/min). The obtained WO3 film had a pale yellow color. This deposition-annealing procedure was repeated five times to obtain a homogeneous film. The active area was 1 cm 1 cm in the case of photoanodes used for IeV measurements and 3.5 cm 5 cm for the films that were used for hydrogen production. The quantity of photocatalyst was calculated by weighing larger samples to be approximately 4.5 mg/cm2.
TiO2 photoanodes For reasons of comparison, some measurements were made by using nanoparticulate titania photoanodes, taking care to deposit the same quantity of photocatalyst and covering the same area of FTO electrode as in the case of WO3. Titania was deposited by the same procedure as repeatedly done in our previous works [13,14].
Construction of the counter electrode For the construction of the counter electrode, a commercial carbon paste (Elcocarb C/SP (Solaronix)) was applied on FTO by doctor blading and was annealed at 450 C. Pt was then deposited on the top using a solution of 3.4% Diamminedinitritoplatinum(II) in ammonium hydroxide diluted in ethanol and casted on warm film. The film was subsequently annealed again at 450 C. The final quantity of Pt on the film was calculated to be 0.1 mg/cm2. The active area of the thus prepared electrocatalyst film was 3.5 cm 5 cm.
Photoelectrochemical measurements JeV measurements were conducted in a small reactor made of plexiglas. The photoanode electrode played the role of reactor window. Its active area was 1 cm 1 cm. A Pt foil (2.5 cm 2.5 cm) was used as counter electrode. Photoelectrochemical measurements were conducted in a threeelectrode configuration using a Ag/AgCl reference electrode and aqueous 0.5 M NaClO4 or 0.5 M LiClO4 as electrolytes. The quantity of the added electrolyte was 10 ml. In some cases the electrolyte contained 5% v/v ethanol. In the case of hydrogen production, a large H-shaped reactor made of pyrex glass was used, which could accommodate larger photoanode and counter electrodes (3.5 cm 5 cm active area). The above Pt/Elcocarb/FTO electrode (section Construction of the counter electrode) was used as counter electrode. 0.5 M aqueous NaClO4 without or with 5% v/v ethanol was used as electrolyte. Ethanol was added only in the anode compartment. The quantity of the electrolyte was 250 ml in each compartment of the cell. The two
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
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 x x x ( 2 0 1 6 ) 1 e6
3
compartments were separated by a silica frit (ROBU, Germany, porosity SGQ 5, diameter 25 mm, thickness 2 mm). Hydrogen was monitored on line by using Ar as carrier inert gas and by applying a bias measured vs Ag/AgCl. Illumination was made in all cases using a Xe lamp providing an intensity of 100 mW cm2 at the position of the photoanode. Hydrogen was detected by using an SRI 8610C gas chromatograph. Calibration of the chromatograph signal was accomplished by comparison with a standard of 0.25% H2 in Ar. Application of electric bias and currentevoltage curves were traced with the help of an Autolab potentiostat PGSTAT128N.
Morpho-structural characterization of the WO3 films UVeVis diffuse reflectance spectra (DRS) were recorded with a Shimadzu model 2600 spectrophotometer and IPCE spectra with a home-made apparatus using a Xe lamp and a series of interference filters. The surface morphology and particle size of the samples was observed with Field-Emission Scanning Electron Microscopy (FESEM, Zeiss SUPRA 35 VP). XRD measurements were carried out with a D8 ADVANCE (Bruker AXS) diffractometer (Source/CuKa: 1.54 A, Power 1.6 W) operating in Bragg-Brentano q/q geometry and BET measurements were made with a Micromeritics Tristar 3000 apparatus.
Results and discussion
Fig. 2 e X-ray diffractogram of monoclinic WO3 nanocrystals on the FTO background.
crystallites (Fig. 2) of size about 25 nm as calculated by using Scherrer's formula. Obviously, this value matches the data obtained by FESEM measurements. The absorption spectrum of the presently made WO3 photoanode (Fig. 3) had a threshold at 465 nm, which corresponds to an energy gap of 2.7 eV.
Current density-voltage characteristics of the WO3 photoanodes
Characterization of the WO3 films Mesoporous tungsten trioxide films produced by the method described in subsection Preparation of electrodes had the structure shown in the FESEM image of Fig. 1. Nanoparticles had polydispersed sizes ranging between 20 and 50 nm. BET specific surface area SSA was 25.4 m2 g1. This SSA value is rather low, compared with, for example, commercial nanoparticulate titania P25, which is around 50 m2 g1. X-Ray diffractograms revealed the formation of monoclinic WO3
Fig. 1 e FESEM image of a typical WO3 photoanode used in the present work. The scale bar is 200 nm.
The photocurrent produced by WO3 photoanodes reached a substantial value, which almost doubled in the presence of ethanol. Indeed, as seen in Fig. 4, when the voltage was 1.6 V vs Ag/AgCl, i.e. just before water electrolytic oxidation onset (1.7 V), the current density in the absence of ethanol reached 3.5 mA cm2 but became 6.3 mA cm2 in its presence. Obviously, a strong bias is necessary to reach current density
Fig. 3 e Diffuse reflectance absorption spectrum and IPCE% values for a WO3 photoanode. IPCE data were recorded under bias of 1.6 V vs Ag/AgCl in the presence of 0.5 M aqueous NaClO4.
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
4
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 x x x ( 2 0 1 6 ) 1 e6
Fig. 4 e JeV curves obtained with a cell comprising a WO3 photoanode in the presence of 0.5 M aqueous NaClO4: (1) in the dark; (2) under illumination in the absence and (3) in the presence of 5% v/v Ethanol. The value of the pH was 7.3 throughout the cell.
values comparable with those theoretically expected. The extensive increase of current in the presence of ethanol is obviously due to current doubling phenomena [3,18e21], i.e. the injection of electrons into the conduction band of the semiconductor by unstable radicals created during ethanol photoelectrocatalytic oxidation. The onset for anodic photocurrent was around 0.3 V vs Ag/AgCl but in the presence of ethanol it appeared at lower voltage. In fact the position of the onset is obscured by the formation of an anodic peak around 0.2 V vs Ag/AgCl, which is due to capacitance current and derives from the injection of Naþ ions into the mesoporous WO3 film structure (cf. similar phenomenon with titania films in Ref. [22]). If lithium perchlorate electrolyte is used instead of sodium perchlorate this peak becomes more intense, obviously, because smaller-sized lithium ions are more easily injected in the mesostructure than sodium ions. This is clearly demonstrated by the cyclic voltametry data shown in Fig. 5. Indeed in the presence of Liþ, a more intense peak was produced. Injection of Liþ ions into WO3 films is, of course, a long known and studied phenomenon related with electrochromism [23]. The photocurrent obtained with the present WO3 photoanode is much higher than that produced by an equivalent titania photoanode under similar conditions. Indeed, a photoanode was constructed by using titania P25. Care was taken to deposit approximately the same quantity of photocatalyst in both titania and WO3 photoanodes (4.5 mg/cm2). Comparison of the photocurrents is made in Fig. 6 both in the absence and in the presence of ethanol. The higher current values observed with WO3 in both cases, is obviously due to the broader range of photon absorption by the WO3 photoanode. The obvious advantage of using WO3 has then been demonstrated both for photoelectrochemical water oxidation (absence of ethanol) and PhotoFuelCell operation (presence of ethanol). The photocurrent onset in the case of the titania
Fig. 5 e Cyclic voltametry curves recorded with WO3 as working electrode, a Pt foil as counter electrode and Ag/ AgCl as reference electrode in the presence of 0.5 M aqueous LiClO4 or NaClO4. Both curves were recorded in the dark.
photoanode appeared at more negative potential and this is the result of the fact that the titania conduction band lies at more negative potentials than the WO3 conduction band. WO3 mesoporous films then demonstrated themselves as important visible light responsive photoanodes. Visible light response was certified in the most revealing manner by the photoanode action spectrum, which is also shown in Fig. 3. Indeed, IPCE% values fitted the (diffuse reflectance) absorption spectrum thus certifying visible light response of the mesoporous WO3 photoanode.
Fig. 6 e JeV curves obtained with a cell comprising a WO3 (1,2) or a TiO2 (3,4) photoanode in the presence of 0.5 M aqueous NaClO4: (1,3) in the absence and (2,4) in the presence of 5% v/v Ethanol (pH ¼ 7.3).
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
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 x x x ( 2 0 1 6 ) 1 e6
Hydrogen production by photoelectrochemical water splitting and ethanol oxidation Photoelectrochemical cells comprising mesoporous WO3 photoanodes can be used to produce hydrogen by photoelectrocatalytic reductive reactions at the cathode electrode. Hydrogen production was indeed monitored by applying an external bias of 1.6 V vs Ag/AgCl, i.e. just before the onset potential for electrocatalytic water oxidation. Hydrogen production was monitored both in the absence or the presence of ethanol in the anode compartment. As seen in Fig. 7, a substantial quantity of hydrogen was produced by water splitting but it became much larger when ethanol was present. Thus the cell can be efficiently used both for water splitting and for PhotoFuelCell operation producing hydrogen. Ethanol was presently used as model fuel, however, other organic substances or mixtures of substances that create wastes can be used as fuel with obvious environmental benefit. Hydrogen
5
production rate in Fig. 7 demonstrated a fast rising part and then dropped a little to attain a very stable value for several hours. The reason for this small drop is not clear but it is not important either. Apparently, the studied system demonstrated a remarkable stability. The above data may be subject to skepticism concerning the relatively high bias applied to produce hydrogen. Hydrogen can, of course, be also produced at lower bias but, as expected, it is substantially lower, as can be seen by the data of Fig. 7. Indeed, hydrogen production rate dropped by more than 50% when the applied bias was 1.0 V vs Ag/AgCl. The choice of a bias of 1.6 V vs Ag/AgCl is, of course, (slightly) lower than the onset for water electrocatalytic oxidation, which appears at 1.7 V vs Ag/AgCl. Furthermore, both current values and hydrogen production rates became zero when the light was turned off. The process of hydrogen production is then purely photoelectrocatalytic. Therefore, it makes no harm to exploit the whole potential range. In addition, it must be taken into account that water electrocatalytic splitting necessitates the employment of expensive Pt electrocatalysts on both anode and cathode electrode. At least in the present case, one of the two electrodes, i.e. the photoanode, was constructed by using an inexpensive photocatalyst. Furthermore, a favorable situation would prescribe water splitting at smaller biases than the theoretical value of 1.229 V. Unfortunately, real electrocatalytic water splitting cases so far reported were supported by electric biases much higher than this theoretical value [24,25]. Such data, justify why high bias may also be necessary to make the present WO3 photoanodes effective.
Conclusions Mesoporous WO3 films on FTO electrodes make efficient visible-light-responsive photoanodes, which can be used for photoelectrochemical hydrogen production. WO3 photoanodes can approach the maximum current density value expected for their light absorption range only under strong bias, while in the presence of ethanol, employed as model fuel, the current density was doubled, indicating a clear-cut current doubling phenomenon. WO3 demonstrated a large advantage vs titania photoanodes, obviously due to its visible light response.
Acknowledgments
Fig. 7 e Photoelectrochemical hydrogen production rate (A) and cumulative hydrogen production (B) in the absence and in the presence of ethanol using a WO3 photoanode. The applied bias was measured vs Ag/AgCl.
This project is implemented under the “ARISTEIA” Action of the “OPERATIONAL PROGRAMME EDUCATION AND LIFELONG LEARNING” and is co-funded by the European Social Fund and National Resources (Project No.2275). Olivier Monfort wishes to acknowledge a grant provided by the Scientific Grant Agency of the Slovak Republic (Project VEGA 1/0276/15) and the National Scholarship Program of the Slovak Republic managed by SAIA n.o. and funded by the Ministry of Education, Sport, Science and Research of the Slovak Republic, which allowed his stay in the University of Patras.
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
6
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 x x x ( 2 0 1 6 ) 1 e6
Finally, we would like to thank Prof.E.Stathatos for his help with the BET measurements.
references
[1] Hodes G, Cahen D, Manassen J. Tungsten trioxide as a photoanode for a photoelectrochemical cell (PEC). Nature 1976;260:312e3. [2] Ashokkumar M, Maruthamuthu P. Photocatalytic hydrogen production with semiconductor particulate systems: an effort to enhance the efficiency. Int J Hydrogen Energy 1991;16(9):591e5. [3] Alexander BD, Kulesza PJ, Rutkowska I, Solarska R, Augustynski J. Metal oxide photoanodes for solar hydrogen production. J Mater Chem 2008;18:2298e303. [4] Gan J, Lu X, Tong Y. Towards highly efficient photoanodes: boosting sunlight-driven semiconductor nanomaterials for water oxidation. Nanoscale 2014;6:7142e64. [5] Zhu T, Chong MN, Chan ES. Nanostructured tungsten trioxide thin films synthesized for photoelectrocatalytic water oxidation: a review. Chem Sus Chem 2014;7:2974e97. ky C, Rajeshwar K, de Tacconi NR, Chanmanee W, [6] Jana Huda MN. Tungsten-based oxide semiconductors for solar hydrogen generation. Catal Today 2013;199:53e64. [7] Liu X, Wang F, Wang Q. Nanostructure-based WO3 photoanodes for photoelectrochemical water splitting. Phys Chem Chem Phys 2012;14:7894e911. [8] Li Z, Luo W, Zhang M, Feng J, Zou Z. Photoelectrochemical cells for solar hydrogen production: current state of promising photoelectrodes, methods to improve their properties, and outlook. Energy Environ Sci 2013;6:347e70. [9] Li W, Li J, Wang X, Ma J, Chen Q. Photoelectrochemical and physical properties of WO3 films obtained by the polymeric precursor method. Int J Hydrogen Energy 2010;35:13137e45. [10] Jiao Z, Wang J, Ke L, Sun XW, Demir HV. Morphology-tailored synthesis of tungsten trioxide (Hydrate) thin films and their photocatalytic properties. Appl Mater Interfaces 2011;3:229e36. [11] Cristino V, Caramori S, Argazzi R, Meda L, Marra GL, Bignozzi CA. Efficient photoelectrochemical water splitting by anodically grown WO3 electrodes. Langmuir 2011;27:7276e84.
[12] Lianos P. Production of electricity and hydrogen by photocatalytic degradation of organic wastes in a photoelectrochemical cell the concept of the PhotoFuelCell: a review of a re-emerging research field. J Hazard Mater 2011;185:575e90. [13] Michal R, Sfaelou S, Lianos P. Photocatalysis for renewable energy production using PhotoFuelCells. Molecules 2014;19:19732e50. [14] Antoniadou M, Lianos P. Near ultraviolet and visible light photoelectrochemical degradation of organic substances producing electricity and hydrogen. J Photochem Photobiol 2009;204:69e74. [15] Murau PC. Dissolution of tungsten by hydrogen peroxide. Anal Chem 1961;33:1125e6. [16] Orel B, Krasovec UO, Groselj N, Kosec M, Drazi G, Reisfeld R. Gasochromic behavior of sol-gel derived Pd doped peroxopolytungstic acid (W-PTA) nano-composite films. J Sol-Gel Sci Technol 1999;14:291e308. [17] Kudo T. A new heteropolyacid with carbon as a heteroatom in a Keggin-like structure. Nature 1984;312:537e8. [18] Morisson SR, Freund T. Chemical role of holes and electrons in ZnO photocatalysis. J Chem Phys 1967;47:1543e51. [19] Maeda Y, Fujishima A, Honda K. The investigation of current doubling reactions on semiconductor photoelectrodes by temperature change measurements. J Electrochem Soc 1981;178:1731e4. [20] Ohno T, Izumi S, Fujihara K, Masaki Y, Matsumura M. Vanishing of current-doubling effect in photooxidation of 2propanol on TiO2 in solutions containing Fe(III) ions. J Phys Chem B 2000;104:6801e3. [21] Kalamaras E, Lianos P. Current doubling effect revisited: current multiplication in a PhotoFuelCell. J Electroanal Chem 2015;751:37e42. [22] Pop L-C, Sfaelou S, Lianos P. Cation adsorption by mesoporous titania photoanodes and its effect on the current-voltage characteristics of photoelectrochemical cells. Electrochim Acta 2015;156:223e7. [23] Kanai F, Kurita S, Sugioka S, Li M, Mita Y. Optical characteristics of W03 electrochromic cells under heavy Li ion injection. J Electrochem Soc 1982;129:2633e5. [24] Liu Q, Gu S, Li CM. Electrodeposition of nickel-phosphorus nanoparticles film as a Janus electrocatalyst for electrosplitting of water. J Power Sources 2015;299:342e6. [25] Carmo M, Fritz DL, Mergel J, Stolten D. A comprehensive review on PEM water electrolysis. Int J Hydrogen Energy 2013;38:4901e34.
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation Stavroula Sfaelou a, Lucian-Cristian Pop a, Olivier Monfort a,1, Vassilios Dracopoulos b, Panagiotis Lianos a,b,* a b
Department of Chemical Engineering, University of Patras, 26500 Patras, Greece FORTH/ICE-HT, P.O. Box 1414, 26504 Patras, Greece
article info
abstract
Article history:
WO3 photoanodes have been constructed by a simple soft chemistry procedure that pro-
Received 6 November 2015
duced efficient mesoporous nanocrystalline films. These photoanodes absorbed visible
Received in revised form
light and could be efficiently employed for photoelectrochemical hydrogen production
15 January 2016
under electric bias. The current increased and the rate of hydrogen production more than
Accepted 15 February 2016
tripled in the presence of a small quantity of ethanol showing that such photoanodes may
Available online xxx
be successfully used in alternative photoelectrochemical installations for solar fuel production by consumption of organic wastes.
Keywords:
Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
Tungsten trioxide
reserved.
Water splitting PhotoFuelCells Hydrogen production
Introduction WO3 is one of the most popular metal oxide semiconductor photocatalysts. It is being studied for several decades [1e4] as an alternative to the UVA absorbing titania. WO3 has a bandgap of 2.5e2.8 eV, therefore, it absorbs visible light up to 500 nm, which accounts for 12% of the solar radiation on the surface of the earth [4]. WO3 is an n-type indirect semiconductor. It is relatively easy to synthesize and deposit on electrodes, it has a moderate hole-diffusion length (~150 nm [4]), it exhibits resistance against photocorrosion and it
demonstrates satisfactory chemical stability at relatively low pH values. For this reason, WO3 has been studied as a photoanode material for photoelectrochemical water splitting applications [1e7]. Its valence band is located approximately at þ2.8 V vs NHE, therefore, it is well placed for water oxidation. Its conduction band is located at positive potentials (approximately þ0.2 tο þ0.3 V vs NHE), therefore, a bias is necessary in order to guide photogenerated electrons to the counter electrode and produce hydrogen by water or proton reduction. The expected theoretical solar to hydrogen efficiency for WO3 photoanodes is about 4.8%, based on its range of light absorption [7] (the corresponding value is only 2.2% for
* Corresponding author. Department of Chemical Engineering, University of Patras, 26500 Patras, Greece. E-mail address: [email protected] (P. Lianos). 1 Permanent address: Department of Inorganic Chemistry, Faculty of Natural Sciences, Comenius University in Bratislava, Mlynska Dolina, 84215 Bratislava, Slovakia. http://dx.doi.org/10.1016/j.ijhydene.2016.02.063 0360-3199/Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
2
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 x x x ( 2 0 1 6 ) 1 e6
titania [7]). For an unbiased cell, where the only input is incident radiation, the photoconversion efficiency, also frequently expressed as solar to hydrogen efficiency, STH, is calculated by the following equation [8]: hSTH ¼
JðmA$cm2 Þx1:229ðVÞ x 100% IðmW$cm2 Þ
(1)
Where J is the current density flowing in the external circuit of the photoelectrochemical cell, I is the intensity of the incident radiation and 1.229 V corresponds to the potential for water splitting at pH zero. Obviously, for solar hydrogen efficiency equal to 4.8% and I ¼ 100 mW cm2 (one sun), the maximum theoretical current density is 3.9 mA cm2. In practice, the obtained values of current density are much smaller so that a strong bias is necessary to reach substantial values, as it will be also shown in the present work. As a matter of fact, due to extensive electronehole recombination in WO3 photoanodes, biases larger than 1.3 V vs RHE are necessary in order to produce substantial currents [7,9e11] and this holds true also for the present data. In order to improve efficiency of WO3 photoanodes, a lot of effort has been spent to synthesize WO3 in a large variety of nanostructures. Comprehensive reviews of such efforts have been presented in Refs. [5] and [7]. Water oxidation by WO3 photoanodes is energetically allowed. However, what is of practical importance in photoelectrochemical water splitting is hydrogen production. In this respect, hydrogen production by oxidation of organic substances is more efficient than pure water splitting [3]. This matter, as far as WO3 is concerned, is rather neglected. For this reason, in the present work, we have investigated hydrogen production both in the absence and in the presence of ethanol, which is being used as a fuel, i.e. a sacrificial electron donor. Thus, the studied photoelectrochemical cells behave as PhotoFuelCells [12e14] set up for hydrogen production. The results showed that a satisfactory hydrogen production rate can be reached with WO3 photoanodes. Ethanol in the present work is employed as model fuel, as in previous works [12e14]. However, other organic substances, for example organic wastes may be also employed thus offering the double environmental benefit of renewable hydrogen production with simultaneous environmental remediation.
Experimental Materials Unless otherwise indicated, reagents were obtained from Sigma Aldrich and were used as received. Millipore water was used in all experiments. SnO2:F transparent conductive electrodes (FTO, Resistance 8 U/square) were purchased from Pilkington (USA) and carbon paste Elcocarb C/SP from Solaronix (Switzerland).
Preparation of electrodes WO3 photoanodes High purity tungsten powder with average particle sizes up to 10 microns (0.4 g, 99.99%) reacted with aqueous hydrogen
peroxide (3 ml, 30%) [15] under sonication for 2e3 h until a transparent colorless solution was obtained. Sonication decreases reaction time to 2e3 h while magnetic stirring necessitates 12e16 h. The excess of H2O2 was catalytically decomposed overnight using a Pt foil [16,17]. Then, 3 ml ethanol and 0.3 g of surfactant Triton X-100 were added to the solution. An FTO glass was cut in the appropriate dimensions and was carefully cleaned first with soap and then by sonication in acetone, ethanol, and water. The WO3 film was prepared by coating the FTO electrode with the above solution by doctor blading. The films were sintered for 10 min at 500 C (heating rate of 20 C/min). The obtained WO3 film had a pale yellow color. This deposition-annealing procedure was repeated five times to obtain a homogeneous film. The active area was 1 cm 1 cm in the case of photoanodes used for IeV measurements and 3.5 cm 5 cm for the films that were used for hydrogen production. The quantity of photocatalyst was calculated by weighing larger samples to be approximately 4.5 mg/cm2.
TiO2 photoanodes For reasons of comparison, some measurements were made by using nanoparticulate titania photoanodes, taking care to deposit the same quantity of photocatalyst and covering the same area of FTO electrode as in the case of WO3. Titania was deposited by the same procedure as repeatedly done in our previous works [13,14].
Construction of the counter electrode For the construction of the counter electrode, a commercial carbon paste (Elcocarb C/SP (Solaronix)) was applied on FTO by doctor blading and was annealed at 450 C. Pt was then deposited on the top using a solution of 3.4% Diamminedinitritoplatinum(II) in ammonium hydroxide diluted in ethanol and casted on warm film. The film was subsequently annealed again at 450 C. The final quantity of Pt on the film was calculated to be 0.1 mg/cm2. The active area of the thus prepared electrocatalyst film was 3.5 cm 5 cm.
Photoelectrochemical measurements JeV measurements were conducted in a small reactor made of plexiglas. The photoanode electrode played the role of reactor window. Its active area was 1 cm 1 cm. A Pt foil (2.5 cm 2.5 cm) was used as counter electrode. Photoelectrochemical measurements were conducted in a threeelectrode configuration using a Ag/AgCl reference electrode and aqueous 0.5 M NaClO4 or 0.5 M LiClO4 as electrolytes. The quantity of the added electrolyte was 10 ml. In some cases the electrolyte contained 5% v/v ethanol. In the case of hydrogen production, a large H-shaped reactor made of pyrex glass was used, which could accommodate larger photoanode and counter electrodes (3.5 cm 5 cm active area). The above Pt/Elcocarb/FTO electrode (section Construction of the counter electrode) was used as counter electrode. 0.5 M aqueous NaClO4 without or with 5% v/v ethanol was used as electrolyte. Ethanol was added only in the anode compartment. The quantity of the electrolyte was 250 ml in each compartment of the cell. The two
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
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 x x x ( 2 0 1 6 ) 1 e6
3
compartments were separated by a silica frit (ROBU, Germany, porosity SGQ 5, diameter 25 mm, thickness 2 mm). Hydrogen was monitored on line by using Ar as carrier inert gas and by applying a bias measured vs Ag/AgCl. Illumination was made in all cases using a Xe lamp providing an intensity of 100 mW cm2 at the position of the photoanode. Hydrogen was detected by using an SRI 8610C gas chromatograph. Calibration of the chromatograph signal was accomplished by comparison with a standard of 0.25% H2 in Ar. Application of electric bias and currentevoltage curves were traced with the help of an Autolab potentiostat PGSTAT128N.
Morpho-structural characterization of the WO3 films UVeVis diffuse reflectance spectra (DRS) were recorded with a Shimadzu model 2600 spectrophotometer and IPCE spectra with a home-made apparatus using a Xe lamp and a series of interference filters. The surface morphology and particle size of the samples was observed with Field-Emission Scanning Electron Microscopy (FESEM, Zeiss SUPRA 35 VP). XRD measurements were carried out with a D8 ADVANCE (Bruker AXS) diffractometer (Source/CuKa: 1.54 A, Power 1.6 W) operating in Bragg-Brentano q/q geometry and BET measurements were made with a Micromeritics Tristar 3000 apparatus.
Results and discussion
Fig. 2 e X-ray diffractogram of monoclinic WO3 nanocrystals on the FTO background.
crystallites (Fig. 2) of size about 25 nm as calculated by using Scherrer's formula. Obviously, this value matches the data obtained by FESEM measurements. The absorption spectrum of the presently made WO3 photoanode (Fig. 3) had a threshold at 465 nm, which corresponds to an energy gap of 2.7 eV.
Current density-voltage characteristics of the WO3 photoanodes
Characterization of the WO3 films Mesoporous tungsten trioxide films produced by the method described in subsection Preparation of electrodes had the structure shown in the FESEM image of Fig. 1. Nanoparticles had polydispersed sizes ranging between 20 and 50 nm. BET specific surface area SSA was 25.4 m2 g1. This SSA value is rather low, compared with, for example, commercial nanoparticulate titania P25, which is around 50 m2 g1. X-Ray diffractograms revealed the formation of monoclinic WO3
Fig. 1 e FESEM image of a typical WO3 photoanode used in the present work. The scale bar is 200 nm.
The photocurrent produced by WO3 photoanodes reached a substantial value, which almost doubled in the presence of ethanol. Indeed, as seen in Fig. 4, when the voltage was 1.6 V vs Ag/AgCl, i.e. just before water electrolytic oxidation onset (1.7 V), the current density in the absence of ethanol reached 3.5 mA cm2 but became 6.3 mA cm2 in its presence. Obviously, a strong bias is necessary to reach current density
Fig. 3 e Diffuse reflectance absorption spectrum and IPCE% values for a WO3 photoanode. IPCE data were recorded under bias of 1.6 V vs Ag/AgCl in the presence of 0.5 M aqueous NaClO4.
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
4
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 x x x ( 2 0 1 6 ) 1 e6
Fig. 4 e JeV curves obtained with a cell comprising a WO3 photoanode in the presence of 0.5 M aqueous NaClO4: (1) in the dark; (2) under illumination in the absence and (3) in the presence of 5% v/v Ethanol. The value of the pH was 7.3 throughout the cell.
values comparable with those theoretically expected. The extensive increase of current in the presence of ethanol is obviously due to current doubling phenomena [3,18e21], i.e. the injection of electrons into the conduction band of the semiconductor by unstable radicals created during ethanol photoelectrocatalytic oxidation. The onset for anodic photocurrent was around 0.3 V vs Ag/AgCl but in the presence of ethanol it appeared at lower voltage. In fact the position of the onset is obscured by the formation of an anodic peak around 0.2 V vs Ag/AgCl, which is due to capacitance current and derives from the injection of Naþ ions into the mesoporous WO3 film structure (cf. similar phenomenon with titania films in Ref. [22]). If lithium perchlorate electrolyte is used instead of sodium perchlorate this peak becomes more intense, obviously, because smaller-sized lithium ions are more easily injected in the mesostructure than sodium ions. This is clearly demonstrated by the cyclic voltametry data shown in Fig. 5. Indeed in the presence of Liþ, a more intense peak was produced. Injection of Liþ ions into WO3 films is, of course, a long known and studied phenomenon related with electrochromism [23]. The photocurrent obtained with the present WO3 photoanode is much higher than that produced by an equivalent titania photoanode under similar conditions. Indeed, a photoanode was constructed by using titania P25. Care was taken to deposit approximately the same quantity of photocatalyst in both titania and WO3 photoanodes (4.5 mg/cm2). Comparison of the photocurrents is made in Fig. 6 both in the absence and in the presence of ethanol. The higher current values observed with WO3 in both cases, is obviously due to the broader range of photon absorption by the WO3 photoanode. The obvious advantage of using WO3 has then been demonstrated both for photoelectrochemical water oxidation (absence of ethanol) and PhotoFuelCell operation (presence of ethanol). The photocurrent onset in the case of the titania
Fig. 5 e Cyclic voltametry curves recorded with WO3 as working electrode, a Pt foil as counter electrode and Ag/ AgCl as reference electrode in the presence of 0.5 M aqueous LiClO4 or NaClO4. Both curves were recorded in the dark.
photoanode appeared at more negative potential and this is the result of the fact that the titania conduction band lies at more negative potentials than the WO3 conduction band. WO3 mesoporous films then demonstrated themselves as important visible light responsive photoanodes. Visible light response was certified in the most revealing manner by the photoanode action spectrum, which is also shown in Fig. 3. Indeed, IPCE% values fitted the (diffuse reflectance) absorption spectrum thus certifying visible light response of the mesoporous WO3 photoanode.
Fig. 6 e JeV curves obtained with a cell comprising a WO3 (1,2) or a TiO2 (3,4) photoanode in the presence of 0.5 M aqueous NaClO4: (1,3) in the absence and (2,4) in the presence of 5% v/v Ethanol (pH ¼ 7.3).
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
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 x x x ( 2 0 1 6 ) 1 e6
Hydrogen production by photoelectrochemical water splitting and ethanol oxidation Photoelectrochemical cells comprising mesoporous WO3 photoanodes can be used to produce hydrogen by photoelectrocatalytic reductive reactions at the cathode electrode. Hydrogen production was indeed monitored by applying an external bias of 1.6 V vs Ag/AgCl, i.e. just before the onset potential for electrocatalytic water oxidation. Hydrogen production was monitored both in the absence or the presence of ethanol in the anode compartment. As seen in Fig. 7, a substantial quantity of hydrogen was produced by water splitting but it became much larger when ethanol was present. Thus the cell can be efficiently used both for water splitting and for PhotoFuelCell operation producing hydrogen. Ethanol was presently used as model fuel, however, other organic substances or mixtures of substances that create wastes can be used as fuel with obvious environmental benefit. Hydrogen
5
production rate in Fig. 7 demonstrated a fast rising part and then dropped a little to attain a very stable value for several hours. The reason for this small drop is not clear but it is not important either. Apparently, the studied system demonstrated a remarkable stability. The above data may be subject to skepticism concerning the relatively high bias applied to produce hydrogen. Hydrogen can, of course, be also produced at lower bias but, as expected, it is substantially lower, as can be seen by the data of Fig. 7. Indeed, hydrogen production rate dropped by more than 50% when the applied bias was 1.0 V vs Ag/AgCl. The choice of a bias of 1.6 V vs Ag/AgCl is, of course, (slightly) lower than the onset for water electrocatalytic oxidation, which appears at 1.7 V vs Ag/AgCl. Furthermore, both current values and hydrogen production rates became zero when the light was turned off. The process of hydrogen production is then purely photoelectrocatalytic. Therefore, it makes no harm to exploit the whole potential range. In addition, it must be taken into account that water electrocatalytic splitting necessitates the employment of expensive Pt electrocatalysts on both anode and cathode electrode. At least in the present case, one of the two electrodes, i.e. the photoanode, was constructed by using an inexpensive photocatalyst. Furthermore, a favorable situation would prescribe water splitting at smaller biases than the theoretical value of 1.229 V. Unfortunately, real electrocatalytic water splitting cases so far reported were supported by electric biases much higher than this theoretical value [24,25]. Such data, justify why high bias may also be necessary to make the present WO3 photoanodes effective.
Conclusions Mesoporous WO3 films on FTO electrodes make efficient visible-light-responsive photoanodes, which can be used for photoelectrochemical hydrogen production. WO3 photoanodes can approach the maximum current density value expected for their light absorption range only under strong bias, while in the presence of ethanol, employed as model fuel, the current density was doubled, indicating a clear-cut current doubling phenomenon. WO3 demonstrated a large advantage vs titania photoanodes, obviously due to its visible light response.
Acknowledgments
Fig. 7 e Photoelectrochemical hydrogen production rate (A) and cumulative hydrogen production (B) in the absence and in the presence of ethanol using a WO3 photoanode. The applied bias was measured vs Ag/AgCl.
This project is implemented under the “ARISTEIA” Action of the “OPERATIONAL PROGRAMME EDUCATION AND LIFELONG LEARNING” and is co-funded by the European Social Fund and National Resources (Project No.2275). Olivier Monfort wishes to acknowledge a grant provided by the Scientific Grant Agency of the Slovak Republic (Project VEGA 1/0276/15) and the National Scholarship Program of the Slovak Republic managed by SAIA n.o. and funded by the Ministry of Education, Sport, Science and Research of the Slovak Republic, which allowed his stay in the University of Patras.
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063
6
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 x x x ( 2 0 1 6 ) 1 e6
Finally, we would like to thank Prof.E.Stathatos for his help with the BET measurements.
references
[1] Hodes G, Cahen D, Manassen J. Tungsten trioxide as a photoanode for a photoelectrochemical cell (PEC). Nature 1976;260:312e3. [2] Ashokkumar M, Maruthamuthu P. Photocatalytic hydrogen production with semiconductor particulate systems: an effort to enhance the efficiency. Int J Hydrogen Energy 1991;16(9):591e5. [3] Alexander BD, Kulesza PJ, Rutkowska I, Solarska R, Augustynski J. Metal oxide photoanodes for solar hydrogen production. J Mater Chem 2008;18:2298e303. [4] Gan J, Lu X, Tong Y. Towards highly efficient photoanodes: boosting sunlight-driven semiconductor nanomaterials for water oxidation. Nanoscale 2014;6:7142e64. [5] Zhu T, Chong MN, Chan ES. Nanostructured tungsten trioxide thin films synthesized for photoelectrocatalytic water oxidation: a review. Chem Sus Chem 2014;7:2974e97. ky C, Rajeshwar K, de Tacconi NR, Chanmanee W, [6] Jana Huda MN. Tungsten-based oxide semiconductors for solar hydrogen generation. Catal Today 2013;199:53e64. [7] Liu X, Wang F, Wang Q. Nanostructure-based WO3 photoanodes for photoelectrochemical water splitting. Phys Chem Chem Phys 2012;14:7894e911. [8] Li Z, Luo W, Zhang M, Feng J, Zou Z. Photoelectrochemical cells for solar hydrogen production: current state of promising photoelectrodes, methods to improve their properties, and outlook. Energy Environ Sci 2013;6:347e70. [9] Li W, Li J, Wang X, Ma J, Chen Q. Photoelectrochemical and physical properties of WO3 films obtained by the polymeric precursor method. Int J Hydrogen Energy 2010;35:13137e45. [10] Jiao Z, Wang J, Ke L, Sun XW, Demir HV. Morphology-tailored synthesis of tungsten trioxide (Hydrate) thin films and their photocatalytic properties. Appl Mater Interfaces 2011;3:229e36. [11] Cristino V, Caramori S, Argazzi R, Meda L, Marra GL, Bignozzi CA. Efficient photoelectrochemical water splitting by anodically grown WO3 electrodes. Langmuir 2011;27:7276e84.
[12] Lianos P. Production of electricity and hydrogen by photocatalytic degradation of organic wastes in a photoelectrochemical cell the concept of the PhotoFuelCell: a review of a re-emerging research field. J Hazard Mater 2011;185:575e90. [13] Michal R, Sfaelou S, Lianos P. Photocatalysis for renewable energy production using PhotoFuelCells. Molecules 2014;19:19732e50. [14] Antoniadou M, Lianos P. Near ultraviolet and visible light photoelectrochemical degradation of organic substances producing electricity and hydrogen. J Photochem Photobiol 2009;204:69e74. [15] Murau PC. Dissolution of tungsten by hydrogen peroxide. Anal Chem 1961;33:1125e6. [16] Orel B, Krasovec UO, Groselj N, Kosec M, Drazi G, Reisfeld R. Gasochromic behavior of sol-gel derived Pd doped peroxopolytungstic acid (W-PTA) nano-composite films. J Sol-Gel Sci Technol 1999;14:291e308. [17] Kudo T. A new heteropolyacid with carbon as a heteroatom in a Keggin-like structure. Nature 1984;312:537e8. [18] Morisson SR, Freund T. Chemical role of holes and electrons in ZnO photocatalysis. J Chem Phys 1967;47:1543e51. [19] Maeda Y, Fujishima A, Honda K. The investigation of current doubling reactions on semiconductor photoelectrodes by temperature change measurements. J Electrochem Soc 1981;178:1731e4. [20] Ohno T, Izumi S, Fujihara K, Masaki Y, Matsumura M. Vanishing of current-doubling effect in photooxidation of 2propanol on TiO2 in solutions containing Fe(III) ions. J Phys Chem B 2000;104:6801e3. [21] Kalamaras E, Lianos P. Current doubling effect revisited: current multiplication in a PhotoFuelCell. J Electroanal Chem 2015;751:37e42. [22] Pop L-C, Sfaelou S, Lianos P. Cation adsorption by mesoporous titania photoanodes and its effect on the current-voltage characteristics of photoelectrochemical cells. Electrochim Acta 2015;156:223e7. [23] Kanai F, Kurita S, Sugioka S, Li M, Mita Y. Optical characteristics of W03 electrochromic cells under heavy Li ion injection. J Electrochem Soc 1982;129:2633e5. [24] Liu Q, Gu S, Li CM. Electrodeposition of nickel-phosphorus nanoparticles film as a Janus electrocatalyst for electrosplitting of water. J Power Sources 2015;299:342e6. [25] Carmo M, Fritz DL, Mergel J, Stolten D. A comprehensive review on PEM water electrolysis. Int J Hydrogen Energy 2013;38:4901e34.
Please cite this article in press as: Sfaelou S, et al., Mesoporous WO3 photoanodes for hydrogen production by water splitting and PhotoFuelCell operation, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.02.063