Jump to ContentJump to Main Navigation
Show Summary Details
More options …

Nanophotonics

Editor-in-Chief: Sorger, Volker

12 Issues per year


CiteScore 2017: 6.57

IMPACT FACTOR 2017: 6.014
5-year IMPACT FACTOR: 7.020


In co-publication with Science Wise Publishing

Open Access
Online
ISSN
2192-8614
See all formats and pricing
More options …
Volume 5, Issue 4

Issues

Towards efficient solar-to-hydrogen conversion: Fundamentals and recent progress in copper-based chalcogenide photocathodes

Yubin Chen
  • International Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Shaanxi 710049, P. R. China
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Xiaoyang Feng
  • International Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Shaanxi 710049, P. R. China
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Maochang Liu
  • International Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Shaanxi 710049, P. R. China
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Jinzhan Su
  • International Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Shaanxi 710049, P. R. China
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Shaohua Shen
  • Corresponding author
  • International Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Shaanxi 710049, P. R. China
  • Email
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
Published Online: 2016-10-20 | DOI: https://doi.org/10.1515/nanoph-2016-0027

Abstract

Photoelectrochemical (PEC) water splitting for hydrogen generation has been considered as a promising route to convert and store solar energy into chemical fuels. In terms of its large-scale application, seeking semiconductor photoelectrodes with high efficiency and good stability should be essential. Although an enormous number of materials have been explored for solar water splitting in the last several decades, challenges still remain for the practical application. P-type copper-based chalcogenides, such as Cu(In, Ga)Se2 and Cu2ZnSnS4, have shown impressive performance in photovoltaics due to narrow bandgaps, high absorption coefficients, and good carrier transport properties. The obtained high efficiencies in photovoltaics have promoted the utilization of these materials into the field of PEC water splitting. A comprehensive review on copper-based chalcogenides for solar-to-hydrogen conversion would help advance the research in this expanding area. This review will cover the physicochemical properties of copper-based chalco-genides, developments of various photocathodes, strategies to enhance the PEC activity and stability, introductions of tandem PEC cells, and finally, prospects on their potential for the practical solar-to-hydrogen conversion. We believe this review article can provide some insights of fundamentals and applications of copper-based chalco-genide thin films for PEC water splitting.

Keywords: Copper-based chalcogenide; photocathode; water splitting; solar energy; hydrogen

1 Introduction

Solar energy is an inexhaustible natural resource, which can provide enormous power to meet the current and future energy demand of human society. However, solar energy has some drawbacks of low energy density, unstable intensity, and discontinuousness. Converting solar energy to chemical fuels seems to be promising for the efficient utilization of solar energy. Hydrogen, as a clean fuel, emits almost nothing other than water upon utilization, which neither leads to air pollution nor results in the emission of greenhouse gases. Due to its cleanliness, high energy density, storability, and transportability, hydrogen has been considered as one promising energy carrier to substitute electricity, and a so-called “hydrogen economy” can be expected [1, 2]. Meanwhile, hydrogen also plays an important role in the modern chemical industry such as the formation of methane, methanol, and even more hydrocarbons. If solar energy can be converted into hydrogen in a cost-effective and environment-friendly way, that will be of great meaning. Photoelectrochemical (PEC) water splitting using semiconductors is considered as the “Holy Grail” of solar energy conversion, which represents a promising path towards renewable and economical hydrogen generation using sunlight and water as the only inputs [38]. The solar water splitting principle can be illustrated in Figure 1A. The photoexcited electrons and holes are first generated in the photoelectrode by absorbing photons with appropriate energy, and then transferred to the interface of the photoelectrode and the electrolyte to participate in the water reduction and oxidation reactions. The free energy change for the conversion of one molecule of H2O to H2 and half O2 under standard conditions is 237.2 kJ/mol, which corresponds to a water electrolysis potential of 1.23 eV, according to the Nernst equation. To accomplish the water splitting, the bandgaps of the semiconductors should be larger than 1.23 eV, and the conduction band edge and valence band edge should straddle the reduction/oxidation potentials of water splitting. Considering the overpotentials of the hydrogen and oxygen evolution reactions, a much larger bandgap (ca. 1.7 eV) is desirable.

(A) Ideal semiconductor material for splitting water at its surface under illumination with absolute energy scale represented [left vertical axis (-) and (+)] for Ecb, and EVb, and the electrochemical potentials given by -qE°, where E° is the reduction potential for both (H+/H2) and (O2/H2O) redox couples. (B) A dual bandgap tandem configuration with n-type and p-type photoelectrodes electrically connected. Reprinted with permission from ref. [4]. Copyright 2010 American Chemical Society.
Figure 1

(A) Ideal semiconductor material for splitting water at its surface under illumination with absolute energy scale represented [left vertical axis (-) and (+)] for Ecb, and EVb, and the electrochemical potentials given by -qE°, where E° is the reduction potential for both (H+/H2) and (O2/H2O) redox couples. (B) A dual bandgap tandem configuration with n-type and p-type photoelectrodes electrically connected. Reprinted with permission from ref. [4]. Copyright 2010 American Chemical Society.

Since the pioneering work by Fujishima and Honda on TiO2 photoelectrode for water splitting [9], a number of different semiconductors have been examined in terms of their suitability as photoelectrode materials for water splitting, and the results obtained could be found in a number of excellent earlier and recent reviews [1020]. Nevertheless, up to date, it has proven to be difficult to identify a single semiconductor that can harvest substantial solar light, own appropriate conduction, and valence band levels to drive the water reduction and oxidation reactions, and exhibit adequate stability in the harsh aqueous electrolytes. Since the water splitting process inescapably includes two separate half-reactions, it seems promising to use two semiconductor photoelectrodes for the separate oxygen evolution reaction (OER) and hydrogen evolution reaction (HER), and integrated them into a tandem PEC cell for overall water splitting [21]. As illustrated in Figure 1B, upward (downward) band bending can be formed at the interface of n-type (p-type) semiconductor electrode and the aqueous electrolyte due to the equilibrium of Fermi levels. In this case, photoexcited holes in the n-type semiconductor electrode will migrate to the interface for water oxidation, while photoexcited electrons in the p-type semiconductor electrode will transfer to the interface for water reduction. As a consequence, n-type and p-type semiconductor electrodes act as photoanodes and photocathodes, respectively. Compared to overall water splitting, the thermal dynamic requirements are easier satisfied for the separate half-reactions, and only the conduction or valence band position should be considered for HER or OER. By combining suitable photoanode and photocathode together, substantial PEC water splitting without an external bias voltage can be expected in the tandem cell.

In the case of photoanodes for water oxidation, numerous n-type semiconductors have been investigated. However, comparatively, very little work has been carried out on p-type semiconductors as photocathodes for water reduction [22, 23]. The photocathodes based on p-type semiconductors can be cathodically protected from the photooxidation to some extent, because photoexcited holes would be transferred to an outer circuit before the photocorrosion due to the band bending at the photo-electrode/electrolyte interface and an applied external potential. Hence, the search for efficient photocathodes has not been limited to metal oxide semiconductors. In terms of practical applications, semiconductor photoelectrodes should necessitate high efficiency, long-term stability, low cost, and low toxicity. Copper-based chalcogenides, exhibiting p-type semiconductor characters, have been widely used for solar cells with high efficiencies (over 20%) and long-term stability due to narrow bandgaps, high absorption coefficients, and practical carrier-transport properties [2426]. The excellent properties of these materials in the photovoltaic (PV) devices have promoted the utilization for PEC water splitting. In the past decade, various copper-based chalcogenide (e.g. CuInS2, CuGaSe2, Cu(In, Ga)Se2 (CIGS), and Cu2ZnSnS4 (CZTS)) films were examined as photocathodes for PEC water splitting [2730]. High solar-to-hydrogen (STH) conversion efficiency, good stability, and relatively simple and low-cost fabrication process have been achieved for these materials, showing great potentials for efficient PEC water splitting.

This review will focus on the recent development of copper-based chalcogenides for hydrogen evolution by water splitting using solar energy. The fundamental physicochemical properties of copper-based chalco-genides are evaluated. A wide range of copper-based chalcogenides as photocathodes have been studied for PEC water splitting. Several effective strategies to enhance the PEC properties have been reviewed. The tandem PEC configuration using copper-based chalcogenides are discussed for the efficient STH conversion. It is hoped that by the thorough analysis of light harvesting, charge separation, surface chemical reaction over copper-based chalco-genides films, as well as the optimization of their tandem PEC cells, this review will pave the way for constructing an efficient PEC system.

2 Crystal structure, defect state, and optical property

Copper chalcopyrite and kesterite semiconductors have been demonstrated to be active copper-based chalco-genides for PEC water splitting. They are usually p-type semiconductors owing to the intrinsic defects such as Cu vacancies, serving as photocathodes for water splitting. The copper chalcopyrite materials for PEC water splitting generally include ternary I-III-VI2 (I = Cu; II = In, Ga; VI = S, Se) compounds, which are considered to be derived from a parent II-IV compound (such as ZnS) by replacing two group II atoms by one group I atom and one group III atom, as shown in Figure 2 [3133]. The chalcopyrite crystal structure resembles the zinc-blende structure, where each of two group I and group III cations are tetrahedrally coordinated by four group VI anions [34]. The kesterite material for PEC water splitting is quaternary Cu2ZnSnS4 (CZTS). As displayed in Figure 2, the kesterite structure is achieved from a ternary I-III-VI2 chalcopyrite system by replacing two group III atoms with one group II atom and one group IV atom [33]. For the quaternary kesterite semiconductor, similar tetrahedral crystal structure is maintained to the ternary chalcopyrite structure. The similar crystal structures from binary to quaternary semiconductors can facilitate the tuning of physicochemical properties by compositional control [35].

Relationship between binary, ternary, and quaternary semiconductors to produce CuInS2 and CZTS, starting from a II-VI parent compound. Reprinted with permission from ref. [33]. Copyright 2012 WILEY-VCH.
Figure 2

Relationship between binary, ternary, and quaternary semiconductors to produce CuInS2 and CZTS, starting from a II-VI parent compound. Reprinted with permission from ref. [33]. Copyright 2012 WILEY-VCH.

Crystal defects are unavoidable in ternary chalcopy-rite and quaternary kesterite semiconductors, and always correlated with their opto-electronic properties. Unlike the analogous II-IV binaries, chalcopyrite semiconductors appear to tolerate a large range of anion-to-cation off stoi-chiometry. It is relatively easier to form Cu vacancy (VCu) and cation antisite defects such as indium on copper site (denoted as InCu) in CuInSe2, and the large concentration of off-stoichiometry can result from the low formation energy of defect complexes such as (2VCu + InCu) [36]. Meanwhile, the structural tolerance to large off-stoichiometry can lead to the generation of stable copper-poor compounds such as CuIn5Ses and CuIn5Se8 [34]. Since there is one more element in quaternary kesterite semiconductors, a larger amount of possible intrinsic defects can be expected compared to ternary chalcopyrite semiconductors. It is reported that CuZn antisite defect as an acceptor is the dominant intrinsic defect in kesterite semiconductors due to its lowest formation energy. Other acceptor defects with low formation energy are the CuSn antisite and Cu vacancy. In general, the donor defects such as the ZnCu antisite and S vacancy have higher formation energies [37]. The CuZn antisite as the most probable acceptor can explain why kesterite semiconductors always exhibit p-type character. The situation is different for the chal-copyrite semiconductors, where Cu vacancy is the lowest formation energy defect contributing to the p-type doping [38]. For both chalcopyrites and kesterites, deep defect states may act as recombination centers for photogenerated electrons and holes. However, the intrinsic defects can undergo self-passivation due to the formation of defect complexes. For CuInSe2 and CZTS, the most important defects for carrier recombination are expected to be InCu and CuZn, which can be, respectively, passivated by the formation of (2VCu + InCu) and (CuZn + ZnCu) defect complexes.

The defect states are considered to play an important role in the performance of solar cells. For instance, the defect microstructure leads to the high efficiency of CIGS solar cells, and the best kesterite solar cell is based on the device with a Zn-rich and Cu-poor composition [33, 38]. Although the physicochemical nature of the defect states is not well understood at present, it is believed that the delicate control of the composition and defect states of copper-based chalcogenides is crucial to the PEC water splitting.

Most of copper chalcopyrite and kesterite semiconductors for PEC water splitting have narrow bandgaps such as CuInS2 (1.5 eV), CuInSe2 (1.0 eV), CuGaSe2 (1.7 eV), CZTS (1.5 eV), enabling better harvesting of visible and even near-infrared light [32, 39]. The anomalously smaller bandgaps of chalcopyrite semiconductors compared to those of binary II-VI analogs should be attributed to a p-d repulsion effect [34]. Of particular importance, the bandgaps of such copper-based chalcogenides can be easily tuned by forming different alloy compositions, which allows for delicate control of the band structures for alternative solar-to-energy conversion configurations. As shown in Figure 3, alloyed chalcopyrite semiconductors of Cu(In, Ga)(S, Se)2 own the varied bandgaps in the range of 1.0–2.5 eV by tuning the In/(Ga + In) and S/(S + Se) molar ratios [40], and the opto-electronic property is closely related to the composition. Additional significant advantages of these chalcopyrite and kesterite semiconductors are the direct bandgap and high absorption coefficient, which indicate that their films with thickness less than 1 μm can absorb a substantial fraction of the incident photons. Consequently, fewer materials are needed so as to decrease the manufacturing cost and reduce the device weight. Although most of the copper-based chalco-genide semiconductors have more negative conduction band edges for water reduction, their valence band edges are generally inappropriate for water oxidation. Therefore, a suitable photoanode should be coupled with the copper-based chalcogenide photocathode to generate a tandem system for PEC water splitting without the applied external bias.

Optical band gap energies of the complete Cu(In(1-X)Gax)(SYSe(1-Y))2 chalcopyrite system for 0 ≤ X ≤ 1 and 0 ≤ Y ≤ 1. The colors correspond to the light sensation of the human eye. Reprinted with permission from ref. [40]. Copyright 2004 American Institute of Physics.
Figure 3

Optical band gap energies of the complete Cu(In(1-X)Gax)(SYSe(1-Y))2 chalcopyrite system for 0 ≤ X ≤ 1 and 0 ≤ Y ≤ 1. The colors correspond to the light sensation of the human eye. Reprinted with permission from ref. [40]. Copyright 2004 American Institute of Physics.

3 Advancements in copper-based chalcogenides for PEC water splitting

In order to meet the practical water splitting, the copper-based chalcogenide photocathodes should necessitate high photocurrent, appropriate onset potential, and good stability. In the last decade, various copper-based chalco-genide materials have been developed for PEC water splitting. The half-cell solar-to-hydrogen (HC-STH) efficiency has been rapidly improved to 8.5%, and the most stable photocathode can last for 20 days without serious photocurrent decay. These results verify their attractive potential in PEC hydrogen production system. The following parts will be focused on the recent development of copper chalcopyrite and kesterite materials for PEC water splitting.

3.1 Copper chalcopyrite photocathodes

CIGS is one of the most examined copper chalcopyrites for PEC water splitting due to the tunable composition, structure and bandgap (1.0–1.7 eV), as well as the high efficiency over 20% in solar cell. Valderrama et al. first investigated the photocurrent and stability of CIGS thin film as a photocathode for hydrogen generation in 2004 and 2005. It was observed that under illumination, the amount of produced hydrogen was two orders of magnitude higher than the hydrogen generated in the dark. The long-term test showed morphological changes on the electrode surface, but without substantial changes in the chemical composition [29, 41]. Marsen et al. demonstrated that high photocurrents of 18–27 mA/cm2 could be achieved for CIGS and CuGaSe2 photocathodes in 0.5 M sulfuric acid solution, and no major degradation was observed during the PEC testing. Due to better conductivity, electrodes with molybdenum back contact exhibited superior performance compared to those with SnO2: F back contact [42]. Jacobsson and cowork-ers declared that the PEC property of CIGS photocathode could be significantly improved by using a solid state p–n junction for charge separation and a catalyst for surface reaction. A photocurrent of 6 mA/cm2 was demonstrated for the water reduction. However, the stability of CIGS photocathode in water turned out to be a problem [43].

The PEC reaction conditions have a great influence on the properties of CIGS photocathodes. Domen’s group compared the PEC activities of CIGS in 0.1 M Na2SO4 solution with different pH values, and the electrolyte with the pH of 9.5 was the optimal one. By depositing CdS to form a p–n junction and Pt as a HER catalyst, the activity of CIGS could be significantly improved [44]. Further, much higher photocurrent could be achieved in a phosphate buffer electrolyte as a result of the stabilized pH near the electrode surface and the reduced chemical bias. As presented in Figure 4A and 4B, the HC-STH efficiency of Pt/CdS/CIGS photocathode reached a maximum of 5.4% at 0.30 VRHE even under neutral conditions. By additional surface modification with a thin conductive Mo/Ti layer, the acquired hybrid Pt/Mo/Ti/CdS/CIGS photocathode exhibited a significantly improved photocurrent, corresponding to a HCSTH as high as 8.5% at 0.38 VRHE. This efficiency exceeded the previously reported values for photocathodes based on polycrystalline thin films. A proposed mechanism for hydrogen evolution on the surface of the irradiated photocathode with the conductor layer was illustrated in Figure 4C. On one hand, the excellent conductivity of the Mo/Ti layer could accelerate the migration of photogenerated electrons to the HER sites. On the other hand, the Mo layer and Ti layer might respectively form intimate contacts with the Pt particle and the underlying CdS/CIGS, which were also beneficial to the charge transfer. Despite a gradual decay, the hybrid Pt/Mo/Ti/CdS/CIGS photoelectrode generated an impressive photocurrent over a period of 10 days under simulated sunlight [45].

(A) Current-potential curves and (B) corresponding HCSTH values of Pt/M/CdS/CIGS electrodes (M = none, Ti, Mo, and Mo/Ti) in 0.5 M Na2SO4, 0.25 M Na2HPO4 and 0.25 M NaH2PO4 (aq.) (pH adjusted to 6.8 by NaOH addition) under AM 1.5 G irradiation. The potential was swept towards the positive direction at 10 mV s-1. (C) A proposed mechanism for hydrogen evolution on the surface of the irradiated photocathode with the conductor layer. Reprinted with permission from ref. [45]. Copyright 2015 Royal Society of Chemistry. HC-STH: half-cell solar-to-hydrogen.
Figure 4

(A) Current-potential curves and (B) corresponding HCSTH values of Pt/M/CdS/CIGS electrodes (M = none, Ti, Mo, and Mo/Ti) in 0.5 M Na2SO4, 0.25 M Na2HPO4 and 0.25 M NaH2PO4 (aq.) (pH adjusted to 6.8 by NaOH addition) under AM 1.5 G irradiation. The potential was swept towards the positive direction at 10 mV s-1. (C) A proposed mechanism for hydrogen evolution on the surface of the irradiated photocathode with the conductor layer. Reprinted with permission from ref. [45]. Copyright 2015 Royal Society of Chemistry. HC-STH: half-cell solar-to-hydrogen.

In order to realize the commercialization of CIGS thin-films based devices, seeking an economical and scalable fabrication method is indispensable. Electrodeposition, as a promising non-vacuum route, is easily amenable for achieving large-area films of high quality with high rate deposition and efficient material utilization. Mandati et al. employed a simplified sequential pulsed current electrodeposition and a separate selenization step to fabricate CIGS thin films, which could avoid the formation of impurities in CIGS samples. Characterization of annealed films revealed that a single phase chalcopyrite CIGS film with a compact morphology and well-controlled composition was prepared. A photocurrent density of 0.8 mA/cm2 at 0.4 V vs SCE was obtained for water reduction, which indicated the potential of this simplified preparation method for the efficient PEC water splitting [46].

Although CIGS thin films with a typical bandgap of approximately 1.2 eV have shown high photocurrents, substantial external voltage bias is always required to sustain the high photocurrent due to the shallow valence band maximum (VBM). Therefore, CuGaSe2 with a wider bandgap of ca. 1.65 eV and a more positive VBM has been investigated as an alternative for PEC water splitting. Marsen and coworkers fabricated polycrystalline CuGaSe2 thin films with nearly stoichiometric composition by a vacuum co-evaporation. The obtained electrode exhibited a photocurrent of 13 mA/cm2 under the outdoor 1-sun illumination. Spectral response result showed significant incident photon to current efficiencies (IPCEs) throughout the visible spectrum, with the maximum of 63% at 640 nm, and the stable photocurrent output could last for 4 h [47]. Besides the vacuum-based synthetic method, a facile particle transfer method was introduced to prepare copper gallium selenide (CGSe) films with different Ga/Cu ratios. CGSe particles with a Ga/Cu ratio of 2 yielded the highest cathodic photocurrent. Through surface modifications, both the photocurrent and onset potential were significantly increased by 12.4 times and 0.40 V (0.81 V vs RHE). The hybrid Pt/CdS/CGSe electrode could steadily contribute to the stoichiometric hydrogen evolution from water splitting for 16 h under visible light irradiation [48].

CuInS2 with the bandgap of 1.5 eV, as a typical copper chalcopyrite, has received particular interest for solar energy conversion due to the absence of highly toxic Se. It is also of great interest that CuInS2 films with high quality can be fabricated through facile non-vacuum deposition techniques. Ikeda and coworkers prepared polycrystalline CuInS2 films by sulfurization of electrodeposited Cu and In metallic precursor. PEC characterization revealed that the film was suitable for water reduction, but not for water oxidation. Upon the surface modification with CdS (ZnS) and Pt nanoparticles, appreciable H2 production was achieved with IPCE as high as 20% in the wavelength range from ca. 500-750 nm [27]. Zhao et al. used the similar method to prepare porous CuInS2 films. Modification with n-type CdS and TiO2 thin layers significantly increased the photocurrent and onset potential. A maximum solar conversion efficiency reached 1.82% at + 0.25 VRHE, and the IPCE was 62% in the 500–700 nm range at 0 VRHE [49]. Grätzel’s group introduced a low-cost solution-based method to transform Cu2O into CuInS2 film for water splitting. As shown in Figure 5A, the photocurrent of the bare CuInS2 film was very poor. After coating of CdS, aluminum-doped zinc oxide (AZO), and TiO2 overlayers, together with Pt catalyst, the as-prepared hybrid electrode demonstrated a photocurrent of 3.5 mA/cm2 (Figure 5B). IPCE measurement showed the spectral distribution of the photocurrent generation (Figure 5C), revealing a broad response of the CuInS2-based photocathode. The stability test in Figure 5D revealed that 80% photocurrent retention could be obtained under chopped illumination for 2 h, indicating the effectiveness of the protection strategy [50].

3.2 Kesterite photocathodes

Although high PEC properties have been obtained over copper chalcopyrite photocathodes for water splitting, the scarcity of indium and gallium in most examined copper chalcopyrite electrodes inevitably increases the manufacturing cost and restricts the large-scale application. Kesterite CZTS, as a p-type semiconductor, has been considered as one of the most promising light absorbing materials instead of copper chalcopyrites due to its narrow bandgap (~1.5 eV), high absorption coefficient (> 104 M-1 cm-1), low toxicity, and elemental abundance [5153]. Power conversion efficiency of 8.5% was achieved for CZTS-based solar cell, and an efficiency of 12.6% has been reached by partially replacing S with Se [24, 54]. However, related studies using CZTS photo-electrodes for water splitting were only reported in recent years. Domen’s group first reported using CZTS deposited onto Mo-coated soda-lime glass substrates by means of magnetron co-sputtering for water splitting. The obtained CZTS electrode generated cathodic photocurrents, indicating the p-type character. Although the IPCE of single CZTS was only 0.01% at 600 nm under an applied potential of -0.24 V vs RHE, the efficiency could be significantly improved to 40% at 600 nm by the modification with CdS, TiO2, and Pt. The STH conversion efficiency reached 1.2% [30].

(A) Current versus potential (J-V) curve of pure CuInS2 electrode under chopped AM 1.5 G simulated solar illumination. (B) J-V curve of CuInS2 electrode after CdS, AZO, and TiO2 coating in the dark and under AM 1.5 G simulated solar illumination. (C) IPCE spectrum of the CuInS2 electrode after CdS, AZO, and TiO2 coating at 0 V versus RHE under three-electrode configuration. (D) Stability test of the CuInS2 electrode after CdS, AZO, and TiO2 coating at 0 V versus RHE under chopped AM 1.5 G simulated solar illumination for 2 h. Reprinted with permission from ref. [50]. Copyright 2015 American Chemical Society. AZO: aluminum-doped zinc oxide; IPCE: incident photon to current efficiency.
Figure 5

(A) Current versus potential (J-V) curve of pure CuInS2 electrode under chopped AM 1.5 G simulated solar illumination. (B) J-V curve of CuInS2 electrode after CdS, AZO, and TiO2 coating in the dark and under AM 1.5 G simulated solar illumination. (C) IPCE spectrum of the CuInS2 electrode after CdS, AZO, and TiO2 coating at 0 V versus RHE under three-electrode configuration. (D) Stability test of the CuInS2 electrode after CdS, AZO, and TiO2 coating at 0 V versus RHE under chopped AM 1.5 G simulated solar illumination for 2 h. Reprinted with permission from ref. [50]. Copyright 2015 American Chemical Society. AZO: aluminum-doped zinc oxide; IPCE: incident photon to current efficiency.

In order to decrease the fabrication cost, various non-vacuum deposition techniques have been developed to prepare CZTS photocathodes. Zhang et al. developed a facile method to synthesize CZTS photocathodes via electrodeposition of metal precursors followed by sulfurization. By optimizing the sulfur partial pressure in a semi-closed system during the sulfurization process, the PEC performance of CZTS photocathodes can be significantly enhanced. The restraining of Sn volatilization and the increasing S penetration in the semi-hermetic system should account for the increased property [55]. Rovelli et al. investigated the optimization and stabilization of electrode-posited CZTS photocathodes for PEC water reduction. The CZTS photoelectrode was first modified by a CdS buffer layer to improve the charge separation, and then was protected by the overlayers of AZO and titanium dioxide. The overlayer coating not only led to the improved activity owing to favorable interface formation with n-type ZnO and TiO2, but also significantly enhanced the stability of the photoelectrode due to the deposition of a conformal, pinhole-free protection layer [56]. Chen et al. provided a facile solution-based route to prepare composition-tunable (Cu2Sn)xZn3(1-x)S3 (0 ≤ x ≤ 0.75) photoelectrodes based on CZTS materials. (Cu2Sn)xZn3(1-x)S3 nanocrystals were first prepared, and then deposited onto the substrate via an electrophoretic deposition process (Figure 6A). As observed in Figure 6B, (Cu2Sn)xZn3(1-x)S3 films with varied compositions were fabricated, showing different optical properties. The side-view FESEM image in Figure 6C indicated that a dense nanocrystal film with high quality was acquired. The PEC properties of (Cu2Sn)xZn3(1-x)S3 films with p-type character have been investigated under water-splitting conditions. As displayed in Figure 6D, the photocurrent varied as a function of the chemical composition. The obtained (Cu2Sn)0.45Zn1.65S3 (x = 0.45) film showed the highest photocurrent due to its suitable band structure, morphology, and smallest compositional deviation [57]. In addition, Wang and coworkers presented a facile spin-coating method to prepare well crystallized and compact CZTS thin films for PEC water splitting. The bare CZTS photoelectrode demonstrated outstanding PEC performance and chemical stability, which could be further enhanced by surface modification with CdS and TiO2 layers using chemical bath and atomic layer deposition. Especially, the TiO2 layer not only promoted the charge transfer to improve the photocurrent, but also protected the CZTS film by separating it from the electrolyte [58].

(A) Illustration of the electrophoretic deposition for (Cu2Sn)xZn3(1-x)S3 film. (B) UV-vis absorption spectra of (Cu2Sn)xZn3(1-x)S3 films with different x values (x = 0.25, 0.45, and 0.75). The inset shows the photographs of different (Cu2Sn)xZn3(1-x)S3 films. (C) Aside-view FESEM image of the (Cu2Sn)0.45 Zn165S3 (x = 0.45) film. (D) Current-potential curves of the prepared (Cu2Sn)xZn3(1-x)S3 (0.25 ≤ x ≤ 0.75) films with different compositions. The PEC measurements were carried out in the 0.1 M Na2SO4 solution (pH ~ 3) with a mechanical chopper to turn on/off illumination (AM 1.5). Reprinted with permission from ref. [57]. Copyright 2014 American Chemical Society.
Figure 6

(A) Illustration of the electrophoretic deposition for (Cu2Sn)xZn3(1-x)S3 film. (B) UV-vis absorption spectra of (Cu2Sn)xZn3(1-x)S3 films with different x values (x = 0.25, 0.45, and 0.75). The inset shows the photographs of different (Cu2Sn)xZn3(1-x)S3 films. (C) Aside-view FESEM image of the (Cu2Sn)0.45 Zn165S3 (x = 0.45) film. (D) Current-potential curves of the prepared (Cu2Sn)xZn3(1-x)S3 (0.25 ≤ x ≤ 0.75) films with different compositions. The PEC measurements were carried out in the 0.1 M Na2SO4 solution (pH ~ 3) with a mechanical chopper to turn on/off illumination (AM 1.5). Reprinted with permission from ref. [57]. Copyright 2014 American Chemical Society.

Recently, Jiang et al. used an In2S3/CdS double layer to modify the CZTS thin film for PEC water splitting. As shown in Figure 7A, Compared to a Pt deposited CZTS film (Pt/CZTS) and a CZTS film modified with a CdS single layer (Pt/CdS/CZTS), the Pt/In2S3/CdS/CZTS hybrid electrode showed a significantly improved cathodic photocurrent. The highest HC-STH of the prepared Pt/In2S3/CdS/CZTS electrode reached 1.63%, which is a record value for the CZTS-based photocathodes (Figure 7B). Meanwhile, the deposited In2S3 layer could effectively stabilize the CZTS-based photoelectrode against degradation induced by the photocorrosion of CdS layer [59].

(A) Current density-potential curves of Pt/CZTS, Pt/CdS/CZTS, and Pt/In2S3/CdS/CZTS photocathodes in a 0.2 mol dm-3 Na2HPO4/NaH2PO4 solution (pH 6.5) under chopped solar simulated AM 1.5 G light irradiation. (B) The corresponding HC-STH efficiency curves of Pt/CdS/CZTS and Pt/In2S3/CdS/CZTS photocathodes. Reprinted with permission from ref. [59]. Copyright 2015 American Chemical Society. CZTS: Cu2ZnSnS4.
Figure 7

(A) Current density-potential curves of Pt/CZTS, Pt/CdS/CZTS, and Pt/In2S3/CdS/CZTS photocathodes in a 0.2 mol dm-3 Na2HPO4/NaH2PO4 solution (pH 6.5) under chopped solar simulated AM 1.5 G light irradiation. (B) The corresponding HC-STH efficiency curves of Pt/CdS/CZTS and Pt/In2S3/CdS/CZTS photocathodes. Reprinted with permission from ref. [59]. Copyright 2015 American Chemical Society. CZTS: Cu2ZnSnS4.

4 Strategies to improve the PEC performance

4.1 Composition tuning

Changing the cation composition of copper-based chalco-genides is a useful route to tune their physicochemical and PEC properties. For instance, by decreasing the amount of Cu during the synthesis of CuGaSe2, Cu-deficient CuGaSe2 compounds can be obtained. Since the VBM of CuGaSe2 is composed of hybridized Cu d and Se p orbitals, more positive VBM positions are achieved for these Cu-deficient CuGaSe2 compounds compared to the stoichiometric CuGaSe2. Therefore, the required external voltage bias for efficient PEC water splitting can be effectively decreased [60]. Kim et al. examined the effects of the Ga/Cu ratio on the band structures and PEC properties of CGSe. With the increased Ga/Cu ratio, the bandgap became larger and the VBM position became more positive. The optimal photocurrent and onset potential were achieved with the Ga/Cu ratio in the range of 3-3.5. Pt-deposited CGSe electrode with the Ga/Cu ratio of 3 showed an energy conversion efficiency of 0.35% and an onset potential of ca. 1.1 V vs RHE. The relatively large onset potential indicated that only a small external bias voltage is needed for PEC water splitting, and the CGSe photocathode is feasible for a p-n tandem PEC cell under zero bias voltage [61]. Subsequently, they investigated the influence of film deposition conditions on the structural and PEC properties of CGSe thin films. By tuning the flux ratio and the partial pressure of hydrogen during vacuum co-evaporation, varied surface morphology, composition, crystallite size, and photocurrent density of the CGSe thin film could be obtained. SEM images of CGSe films grown without and with hydrogen introduction were presented in Figure 8A and 8B. Hydrogen introduction gave rise to a coarser surface, which might result from the effect of the enhancing growth of the (112) plane. Figure 8C showed the variation of the STH conversion efficiency with the introduced hydrogen pressure during CGSe film growth. It was observed that with increasing hydrogen pressure, the STH values increased remarkably after ZnS modification. Hydrogen introduction during CGSe film growth improved the crystal quality, which consequently, resulted in ZnS/CGSe photoelectrodes with a higher water splitting efficiency [62].

Plane views of CGSe films grown without (A) and with (B) hydrogen introduction. (C) Variation in the values of STH with various hydrogen pressures in Pt/CGSe (•) and Pt/ZnS/CGSe (·). Reprinted with permission from ref. [62]. Copyright 2012 Royal Society of Chemistry. CGSe: copper gallium selenide; STH: solar-to-hydrogen.
Figure 8

Plane views of CGSe films grown without (A) and with (B) hydrogen introduction. (C) Variation in the values of STH with various hydrogen pressures in Pt/CGSe (•) and Pt/ZnS/CGSe (·). Reprinted with permission from ref. [62]. Copyright 2012 Royal Society of Chemistry. CGSe: copper gallium selenide; STH: solar-to-hydrogen.

Another viable way to make the VBM of CuGaSe2 more positive is to substitute partial Cu+ with Ag+ to form AgxCu1-xGaSe2 alloys, since the VBM position of AgGaSe2 is much deeper than that of CuGaSe2 [63]. Zhang et al. successfully fabricated a series of AgxCu1-xGaSe2 (ACGSe) thin films with varied Ag concentrations for PEC water splitting. It was demonstrated that the VBM potential of ACGSe is deeper than that of CuGaSe2. With the optimal Ag concentration, the Pt/CdS/ACGSe photocathode showed a photocurrent density of 8.1 mA/cm2 at 0 VRHE and an onset potential of 0.70 VRHE. Meanwhile, no noticeable decrease could be found over 55 h of PEC reaction, indicating the good stability of Pt/CdS/ACGSe electrode [64]. Furthermore, a CuGa3Se5 thin layer was deposited between ACGSe film and CdS layer, leading to a more desirable band alignment for charge separation. The improved photocurrent and onset potential were thus achieved compared to unmodified Pt/CdS/ACGSe electrode. In particular, this Pt/CdS/CuGa3Se5/ACGSe photocathode could generate a stable photocurrent for around 20 days under visible light irradiation, which is of great importance for the practical application [65].

By the partial replacement of In in CuInS2 with Ga, alloyed Cu(In, Ga)S2 compounds can be achieved, giving rise to the composition-dependent physicochemical and opto-electronic properties [66]. Septina et al. fabricated a series of Cu(In, Ga)S2 films with different amounts of Ga using spray pyrolysis followed by sulfurization. With the increased concentration of Ga, the PEC performance of Cu(In, Ga)S2 film was first improved and then decreased. Pt-CdS/Cu(In, Ga)S2 photocathode with the Ga/(In + Ga) ratio of 0.25 showed both the maximum photocurrent density of 6.8 mA/cm2 (at 0 V vs RHE) and the highest onset potential of 0.89 V vs RHE. The best PEC activity should be ascribed to the relatively wide interface bandgap and much larger grains [67]. The composition-tunable Cu1-xAgxaS2 and (CuGa)1-xZn2xS2 electrodes were also developed as photocathodes for water splitting. It was demonstrated that the band structure and PEC property were closely related to the composition [68, 69].

4.2 Surface modification

It should be noticed that in order to achieve high PEC activity and good stability, the surface modification is indispensable for copper-based chalcogenide photocathodes. The general strategies of the surface modification include forming a p-n junction, depositing a protection layer, and loading a HER catalyst. As known, p-type copper-based chalcogenide film is always modified with an n-type semiconductor to form a p-n junction, so as to promote the charge separation. Commonly, the n-type semiconductor is CdS. For instance, Moriya et al. prepared a CuGaSe2 photoelectrode by co-evaporation and modified it with CdS by a chemical bath deposition. CdS deposition could greatly improve the photocurrent and onset potential in the CuGaSe2 photocathode, possibly owing to the increased thickness of the depletion layer at the solid-electrolyte interface, which played a principal role in charge separation. Meanwhile, the Pt/CdS/CuGaSe2 electrode could generate a stable photocurrent for more than 10 days under visible light irradiation [28]. Although high efficiencies could be achieved for CdS-modified photocathodes, substitution of CdS with environment-friendly n-type layers would be advantageous for the practical application due to the high toxicity of Cd. Gunawan et al. investigated the surface modification of CuInS2 thin film with an n-type In2S3 layer for PEC water splitting. The as-prepared Pt-In2S3/CuInS2 photoelectrode showed much higher efficiency than Pt-CdS/CuInS2, indicating that n-type In2S3 should be an ideal alternative to replace CdS [70]. In addition, n-type ZnS was also demonstrated to be efficient modifier to promote the PEC performances of CuInS2 and CuGa3Se5 photocathodes [27, 62].

The band alignment between p-type copper-based chalcogenide and n-type modifier played a vital role in the charge separation. Mali and coworkers fabricated p-type CIGS films by co-evaporation route and modified them by the deposition of an n-type CdS layer and an additional ZnO layer. It was claimed that the obtained type II cascade band structure yielded the optimal PEC performance with the highest photocurrent density of -32.5 mA/cm2 (-0.7 V vs Ag/AgCl) at a pH of 9 [71]. Sivula’s group systematically discussed the role of surface modifications in enhancing the charge separation of nanocrystalline CZTS photocathodes. It was demonstrated that ZnSe and CdSe coatings gave rise to better photocurrent compared to conventional CdS-modified CZTS thin film. The increased performance of the CZTS/CdSe electrode should be ascribed to a larger conduction band offset (CBO) between CZTS and CdSe, which increased the driving force for charge separation. As ZnSe buffer layer was used, superior charge separation might be achieved owing to the charge transfer through midgap states or bands realignment upon illumination. Therefore, higher photocurrents were achieved due to the type II band alignment at the CZTS/buffer layer interface [72].

Zou and coworkers examined the band positions and PEC properties of CZTS thin films prepared by the ultrasonic spray pyrolysis method. The determined conduction band of CZTS was 0.26 eV higher than that of CdS, which indicated that a type II band alignment and a cliff-like CBO were formed between CZTS and CdS. It was believed that compared to the spike-like CBO in CdS/CIGS, the clifflike offset would give rise to the serious charge recombination at the CdS/CZTS interface and lower the PEC property. In order to achieve high efficiencies, developing n-type semiconductor with suitable conduction band edge to couple with CZTS is thus indispensable [73]. Similarly, in the study of In2S3/CuInS2 and CdS/CuInS2 photocathodes for water splitting, spectroscopic evaluation of conduction band offsets uncovered that a cliff-type offset was formed in CdS/CuInS2 (Figure 9A), whereas In2S3/CuInS2 had a positive notch-type CBO (Figure 9B). In the case of cliff-type band alignment, the photogenerated electrons on the conduction band minimum (CBM) of CuInS2 would lose some energy when transported to CBM of CdS, and led to an increased probability of cross recombination with the holes on VBM of CuInS2 due to the relatively narrow energy. This interface recombination could be suppressed at the notch-type junction of In2S3/CuInS2 since the energy difference between CBM of In2 S3 and VBM of CuInS2 was sufficiently large. Hence, the superior PEC performance of In2S3/CuInS2 compared to CdS/CuInS2 should be ascribed to the formation of a better heterojunction [74].

Energy diagrams of (A) CdS/CuInS2 and (B) In2S3/CuInS2 heterojunctions estimated from photoabsorption and XP spectroscopy data. Reprinted with permission from ref. [74]. Copyright 2015 American Chemical Society.
Figure 9

Energy diagrams of (A) CdS/CuInS2 and (B) In2S3/CuInS2 heterojunctions estimated from photoabsorption and XP spectroscopy data. Reprinted with permission from ref. [74]. Copyright 2015 American Chemical Society.

Based on the above discussion, it can be observed that two basic mechanisms were proposed to elucidate the charge separation behavior between a p-type copper-based chalcogenide and an n-type modifier. A few of researchers believed that a type II band alignment with a cliff-like CBO would facilitate the charge transfer [71, 72]. Nevertheless, some researchers believed that the cliff-like offset would result in the serious charge recombination at the interface and lower the PEC efficiency. It was considered that the spike-like (notch-like) CBO with the CB position of n-type modifier, a little higher than that of p-type copper-based chalcogenide, would effectively avoid the interface charge recombination, leading to a better PEC performance [73, 74]. We believe that the charge separation efficiency through the p-n junction was closely influenced by both the band alignment and the interface states between the p-type copper-based chalcogenide and the n-type semiconductor. In different reported studies, the interface states could be varied, which would lead to different charge separation efficiencies. Therefore, further study should be carried out to clarify the fundamental mechanism of charge transfer through the p-n junction.

To achieve the large-scale application, excellent durability of copper-based chalcogenide photocathodes should be indispensable. Although the self-photooxidation can be partially avoided due to cathodic protection for the photocathodes, the stability of copper-based chalcogenides in water under illumination is still a major problem. The instability always occurred at the photoelectrode/electrolyte interface because the local environment at the interface tended to be rather hostile owing to the surface redox reactions [75]. One effective approach to improve the stability is depositing a stable metal-oxide layer on the surface to prevent the contact of copper-based chalcogenide with the electrolyte [76]. In addition to the stability, the protection layer should be transparent. The band edges should match those of the p-type photoabsorber so that photogenerated electrons in the photoabsorber can be easily injected to the protection layer. An efficient charge transfer through the protection layer is also needed.

It has been demonstrated that the stable TiO2 film should be a useful protection layer for the fragile copper-based chalcogenide photoelectrodes. As illustrated in Figure 10, Azarpira et al. presented a transparent conductive oxide Pt implemented TiO2 layer modified CIGS chalcopy rite thin film for efficient and stable solar-driven hydrogen generation in a three-electrode setup. Pt nanoparticles could be identified in the TiO2 thin film through TEM analysis. Under illumination, a charge carrier separation was caused in the space charge region at the composite interface. High IPCEs of more than 80% could be achieved over the full visible light range in an acidic electrolyte (pH ~0.3), and no degradation was observed over more than 24 h of testing. It was summarized that thin films of Pt implemented phase-pure anatase TiO2 layer could simultaneously optimize the conductivity of the films, the elec-trocatalytic activity, and the light guidance towards the chalcopyrite. Therefore, high photocurrent density, good stability, and anodic onset potential were achieved for TiO2:Pt-CIGS composite photoelectrodes towards PEC water splitting [77].

A schematic outline of the novel Pt-implemented TiO2-CIGS light converting composite photocathode in a three-electrode setup of the photoelectron catalytic cell with Pt counter electrode (CE) and Ag/AgCl reference electrode (left). A SEM cross section is shown of the most efficient Pt-implemented TiO2-CIGS composite photocathode device (center). As indicated by TEM analysis, nanoparticles of Pt could be identified in theTiO2 thin film (top right). Reprinted with permission from ref. [77]. Copyright 2015 WILEY-VCH.
Figure 10

A schematic outline of the novel Pt-implemented TiO2-CIGS light converting composite photocathode in a three-electrode setup of the photoelectron catalytic cell with Pt counter electrode (CE) and Ag/AgCl reference electrode (left). A SEM cross section is shown of the most efficient Pt-implemented TiO2-CIGS composite photocathode device (center). As indicated by TEM analysis, nanoparticles of Pt could be identified in theTiO2 thin film (top right). Reprinted with permission from ref. [77]. Copyright 2015 WILEY-VCH.

Neumann and coworkers introduced a niobium-doped NbxTi1-xOy film to protect the CIGS absorber. It was revealed that NbxTi1-xOy oxide films had a transparency of more than 90% and a low resistivity. A high degree of freedom in niobium atom insertion existed for this material, and it showed a good stability against water and UV light. The developed NbxTi1-xOy front contact in the CIGS tandem cell thus served as an effective protection layer and reactive interface, which allowed the efficient reduction of protons to hydrogen [78]. Besides the stable TiO2 as the protection layer, Yang et al. carried out the surface modification of polycrystalline CuInS2 thin films with photocatalytically active g-C3N4 films for the first time. Although the cathodic photocurrent was reduced, the stability of CuInS2 electrode was significantly improved by the deposition of g-C3N4. The photocurrent of g-C3N4/CuInS2 composite electrode was stable for 22 h in 0.1 M H2SO4 aqueous solution. It was considered that the network of nanoporous g-C3N4 crystallites could strongly protect the CuInS2 substrate from degradation and photocorrosion under acidic conditions [79].

For most of copper-based chalcogenide photocathodes, deposition of noble-metal Pt on the surface as the HER catalyst was indispensable to promote the PEC water reduction because it could efficiently capture the photogenerated charges, and decrease the overpotentials for the surface reaction. However, the high price and low reserve of Pt restricted the large-scale application. Therefore, developing highly efficient noble-metal-free HER catalysts (such as MoSx, WSx, NiSx) for copper-based chalcogenide photocathodes should be carried out in the future [8083].

4.3 Impurity elimination

The impurities such as binary compounds were always unavoidable in the synthesis of copper chalcopyrite and kesterite semiconductors, whichwouldhave anegativein-fluence on the PEC properties. Therefore, seeking effective routes to eliminate the impurities is critical to improve the PEC performance. Guan et al. examined the influence of impurities such as CuxS and a metastable CuAu ordering phase on the PEC properties of CuIn0.7Ga0.3S2 nano-photocathodes. It was observed that the photocurrent of a CuIn0.7Ga0.3S2 photocathode could be significantly enhanced by selective electrochemical etching of a CuAu ordering phase, but not improved after etching of CuxS. By the optimal etching and modification with CdS and Pt, a high photocurrent density of 6.0 mA/cm2 at 0 VRHE under AM 1.5 G simulated sunlight irradiation could be obtained [84]. Subsequently, they investigated the formation mechanism of ZnS impurities and their effect on PEC properties of a CZTS photocathode. It was observed that the amorphous ZnS would react with Cu2SnS3 to form CZTS during sulfur annealing at high temperature. If the Zn/Sn precursor ratio was too high, the excess ZnS could not be eliminated. A relatively pure CZTS was only achieved with the Zn/Sn precursor ratio of 0.6. ZnS impurities could impede charge transfer in CZTS due to high resistance and lower the photocurrent, and a CZTS photocathode without ZnS showed higher photocurrent than those with ZnS [85]. Rovelli et al. used a KCN solution to etch the CZTS samples after sulfurization. Raman results showed that ZnS and CuxSy impurities could be greatly reduced after the etching step, which led to a dramatic improvement of the PEC performance [56].

4.4 Morphology design

Morphology design for copper-based chalcogenide photocathodes can provide a feasible approach to tune their light absorption property, charge transport behavior, and surface reaction rate for PEC water splitting. The vertically aligned one-dimensional or two-dimensional nanostructures are of particular interest for the PEC application because they are capable of capturing more light and facilitating the charge separation and transport. Moon and coworkers used a template-assisted growth and transfer process to prepare vertically aligned CuInS2 nanorod arrays, and demonstrated the surface-modified CuInS2 nano-array as a photocathode with reliable PEC property for the first time. By implementation of CdS and ZnS surface modification, the PEC properties of CuInS2 nanorod arrays could be greatly improved [86]. Li et al. presented a scalable route for the synthesis of single crystalline CZTS nanosheet arrays by using CuS nanosheets as a sacrificial template. As illustrated in Figure 11A, the crystalline nanosheets owned high conductivity, leading to reduced charge transfer resistance and increased carrier collection efficiency. Meanwhile, the nanosheets possessed large surface area, which could reduce the minority carrier diffusion length to the electrolyte. Furthermore, the nanosheet arrays could effectively improve light absorption due to increased optical length. Consequently, a high photocurrent density of -1.32 mA/cm2 was achieved at 0 V vs RHE under illumination of AM 1.5 (Figure 11B), which is the highest value reported for the bare CZTS photoelectrode [87].

Wen et al. successfully synthesized the dense and the porous polycrystalline CZTS photocathodes by tuning the thiourea/zinc ratios via a facile metal organic decomposition method. It was revealed that the porous structure resulted from the decomposition of excess amount of thiourea. Compared to the dense electrode, the porous CZTS photocathode exhibited 3 times higher photocurrent due to the shorter transport distance of minority carriers [88]. Ma and coworker prepared a CZTS photoelectrode by sulfurizing the electroplated Zn/Sn/Cu/Mo-mesh precursor. It was found that the surface morphology and PEC property of the CZTS film could be changed by tuning the pH of the Zn electroplating solution. The CZTS thin film with a hollow-column morphology showed the highest photocurrent due to an increased surface area and easier transfer of photogenerated holes [89].

(A) Schematic illustration of (left) single crystalline CZTS nanosheet arrays generating hydrogen under illumination, and (right) light scattering, charge separation, and transportation in nanosheets. (B) The current density-potential plots of CZTS photocathodes in dark and under AM 1.5 light illumination in 0.5 M Na2SO4 solution (pH 9.5). Reprinted with permission from ref. [87]. Copyright 2015 Royal Society of Chemistry.
Figure 11

(A) Schematic illustration of (left) single crystalline CZTS nanosheet arrays generating hydrogen under illumination, and (right) light scattering, charge separation, and transportation in nanosheets. (B) The current density-potential plots of CZTS photocathodes in dark and under AM 1.5 light illumination in 0.5 M Na2SO4 solution (pH 9.5). Reprinted with permission from ref. [87]. Copyright 2015 Royal Society of Chemistry.

In summary, various copper-based chalcogenide photocathodes and modification strategies have been examined, and a rapid progress has been achieved in the last 5 years. The HC-STH efficiency reaches 8.5%, and the most stable photocathode can last for 20 days. These results verify the attractive potential of copper-based chalcogenide materials in PEC water splitting. To make a direct comparison, the PEC properties and reaction conditions of the representative copper-based chalcogenide photocathodes are listed in Table 1.

Table 1

Representative copper-based chalcogenide photocathodes for PEC water splitting.

5 Tandem PEC cells using copper-based chalcogenides for water splitting

5.1 p-n tandem cell

The above analysis revealed that copper-based chalco-genides could serve as efficient photocathodes for water reduction. However, due to the shallow VBM, a sufficient external potential should be added for overall water splitting. In order to achieve PEC water splitting without an externally applied bias, the p-type photocathode could be integrated with an n-type semiconductor photoanode to form a p-n tandem PEC cell [90, 91]. For this system to work, it is essential that the CBM of the n-type semiconductor must be higher than the VBM of the p-type copper-based chalcogenide, so that the electrons could flow in the external circuit [92].

Kudo and coworkers have combined CoOx-loaded BiVO4 photoanode with various copper-based chalco-genide photocathodes to form p-n tandem cells. First, it was demonstrated that the graphene oxide-modified CuGaS2 composite photocathode possessed electrochemically sufficient onset potential so as to fabricate a tandem cell with CoOx-loaded BiVO4 photoanode for water splitting without an external bias. It was observed that the photocurrent was gradually improved with the increased external bias between the photocathode and the photoanode.

In particular, the photocurrent without any external bias could be expected, indicating the feasibility of this p n tandem cell [93]. Further, they demonstrated PEC water splitting by coupling the Ru-loaded Cu0.8Ag0.2GaS2 photocathode with the CoOx-loaded BiVO4 photoanode in a one-pot cell without an external bias. The tandem PEC cell exhibited stable photocurrent for the reaction time longer than 15 h under simulated sunlight irradiation, and the amount of generated hydrogen agreed well with the photocurrent, indicating that an energy conversion reaction proceeded. The solar energy conversion efficiency was determined to be 0.005% [68]. To improve the PEC performance, a Ru-loaded (CuGa)0.5ZnS2 photocathode was then combined with a CoOx-modified BiVO4 photoanode to form a p n tandem cell (Figure 12A). Stoichiometric amounts of hydrogen and oxygen were evolved under simulated sunlight irradiation without any applied bias (Figure 12B). The solar energy conversion efficiencies with 0 and 0.4 V of the external bias were 0.016 and 0.029%, respectively. Meanwhile, almost 100% of the Faradaic efficiency and good stability were achieved for this tandem PEC cell [69].

(A) One-pot cell made of Pyrex for PEC solar water splitting with a photoanode and photocathode. (B) PEC solar water splitting using Ru (1 wt%)-loaded (CuGa)0.5ZnS2 and CoOx-loaded BiVO4 electrodes under simulated sunlight irradiation without bias. Area of electrodes of (CuGa)0.5ZnS2 and BiVO4: 7.4 and 2.4 cm2, respectively. Electrolyte: 0.025 mol l-1 of Na2HPO4 and 0.0025 mol l-1 of KH2PO4 (pH 8). Light source: solar simulator (AM 1.5, 100 mW cm-2). Reprinted with permission from ref. [69]. Copyright 2015 American Chemical Society. PEC: photoelectrochemical.
Figure 12

(A) One-pot cell made of Pyrex for PEC solar water splitting with a photoanode and photocathode. (B) PEC solar water splitting using Ru (1 wt%)-loaded (CuGa)0.5ZnS2 and CoOx-loaded BiVO4 electrodes under simulated sunlight irradiation without bias. Area of electrodes of (CuGa)0.5ZnS2 and BiVO4: 7.4 and 2.4 cm2, respectively. Electrolyte: 0.025 mol l-1 of Na2HPO4 and 0.0025 mol l-1 of KH2PO4 (pH 8). Light source: solar simulator (AM 1.5, 100 mW cm-2). Reprinted with permission from ref. [69]. Copyright 2015 American Chemical Society. PEC: photoelectrochemical.

Recently, Kim and coworkers presented a tandem system comprised of Pt/CdS/CuGa3 Se5/(Ag, Cu)GaSe2 photocathode and NiOOH/FeOOH/Mo:BiVO4 photoanode in a neutral phosphate buffer solution. The prepared semitransparent Mo:BiVO4 layer allowed sunlight to pass through the top photoanode and reached the bottom photocathode. Consequently, the tandem cell exhibited stoichiometric hydrogen and oxygen evolution with a STH conversion efficiency of 0.67% over 2 h without noticeable degradation. This efficiency is one of the largest values ever reported among PEC tandem cells with the similar configuration [94]. Jiang et al. introduced a two-electrode system to achieve the bias-free water splitting using a Pt/In2S3/CdS/CZTS photocathode for the first time. A two-electrode system was fabricated by simply connecting Pt/In2S3/CdS/CZTS photocathode and BiVO4 photoanode through a conducting wire, which generated stoichiometric amounts of H2 and O2 without an applied external bias. The power conversion efficiency in the present system was determined to be 0.28%. It was implied that the PEC performance could be improved by introducing a photoanode with relatively negative onset potential to combine with Pt/In2 S3/CdS/CZTS photocathode [59]. Menezes et al. claimed that Cu2Se(In2Se3)n (CISe) films could be either p- or n-doped so as to fabricate a p n tandem cell for PEC water splitting. It was believed that the p-CISe and n-CISe-based bi-hybrid photoelectrodes might be combined into side by side or tandem configurations to increase the cell voltage. The solar-matched bandgap of CISe films could potentially enhance the voltage sufficiently to realize spontaneous photoelectrolysis [95].

5.2 PV/PEC tandem cell

As discussed above, combining the copper-based chalco-genide photocathode with a suitable photoanode to construct a p–n tandem cell is a promising way to achieve PEC water splitting without any external bias. However, the reported STH efficiencies are quite low (< 1%) at present, possibly because it is hard to fabricate two photoelectrodes well-matched in the same electrolyte and the generated photovoltage in the p-n tandem cell is usually too small to drive the reaction of water splitting efficiently [96]. An alternative approach is to couple PV devices with the copper-based chalcogenide photocathode to form a PV/PEC tandem cell in which PV device could provide sufficient voltage for solar water splitting. This route provides a good compromise between device performance, stability, and complexity. A typical PV/PEC tandem cell can be constructed by simply wiring the photocathode and PV cell together, or fabricated in a monolithic geometry [21, 97, 100].

Researchers in Hawaii Natural Energy Institute proposed a monolithic PV/PEC tandem device, which was composed of a semiconductor photoelectrode, a conductive interface layer, PV cell(s), a primary substrate material, and a catalytic back-surface CE (Figure 13) [101, 102]. In the case of using p-type copper chalcopyrite photoelec-trodes, hydrogen would be evolved on the PEC electrode, providing a degree of cathodic protection for preventing photocorrosion, and oxygen would be evolved on the catalytic counter electrode. Similar photoconversion performances of CIGS thin films operated in a PEC configuration and a PV solar cell validated the formation of an efficient PEC junction at the CIGS-electrolyte interface. However, CIGS was not quite suitable for monolithically stacked PV/PEC device because its narrow bandgap of 1.4 eV led to poor photogenerated voltage and insufficient transmitted-light to drive underlying PV layers. It was suggested that the optimal bandgap of a copper chalcopyrite-based PEC absorber in a monolithic stack PV/PEC device should be between 1.7 and 2.2 eV to balance the solar absorption by each layer [101]. Therefore, various parameters regarding utilization of CuGaSe2 photoelectrodes with a bandgap of ca. 1.7 eV in the PV/PEC tandem system were examined. Constituent components of the proposed integrated hybrid PV/PEC device were first characterized, and then combined theoretically through electronic and optical models to demonstrate the feasibility of the tandem device.

HNEI-developed “hybrid photoelectrode” structure of a multijunction monolithic stack structure consisting of a PEC junction on top of a PV cell or cells to spontaneously split water under illumination. Reprinted with permission from ref. [101]. Copyright 2010 Elsevier. HNEI: Hawaii Natural Energy Institute; PEC: photo-electrochemical; PV: photovoltaic.
Figure 13

HNEI-developed “hybrid photoelectrode” structure of a multijunction monolithic stack structure consisting of a PEC junction on top of a PV cell or cells to spontaneously split water under illumination. Reprinted with permission from ref. [101]. Copyright 2010 Elsevier. HNEI: Hawaii Natural Energy Institute; PEC: photo-electrochemical; PV: photovoltaic.

The stimulated results revealed that robust CuGaSe2 photocathodes could be coupled with either CIGS or possibly CuInSe2 to generate a hybrid PV/PEC system for water splitting, and a photocurrent up to 15.87 mA/cm2 could be expected by the optimal utilization of incident light. Although the proposed tandem cell could not provide all external voltage to drive the water splitting, it could greatly decrease the required potential to make the PEC water splitting more economically feasible [103]. In addition, they successfully fabricated coplanar hybrid PV/PEC device based on copper chalcopyrite materials for water splitting. The required bias for single photocathode was offset, and a high photocurrent could be achieved. For instance, a coplanar PV/PEC device made of a CuGaSe2 photocathode and Si solar cells was prepared and tested for water splitting in a two-electrode configuration under AM 1.5 G outdoor conditions. A photocurrent density of 3.5 mA/cm2 was finally obtained corresponding to a STH benchmark efficiency of 3.7%. Although the hybrid coplanar design could allow water splitting with a high photocurrent, solar energy was not efficiently utilized in this system compared to the monolithic stack device [104, 105].

Jacobsson and coworkers proposed another kind of PV/PEC cell for water splitting. As displayed in Figure 14A, three CIGS-based PV cells were connected in series into a monolithic device, and then integrated with a catalyst. The interconnected PV cells could provide sufficient driving force for PEC water splitting. When the as-prepared PV/PEC device was immersed in water under AM 1.5 solar illumination, a STH efficiency over 10% could be achieved for unassisted water splitting. As shown in Figure 14B, vigorous bubble generation could be observed. Meanwhile, the low stability of CIGS absorber in the electrolyte could be overcome by encapsulating the absorber module into a protection setup. It was inferred that this efficient PV/PEC device could be easily developed from a CIGS-based PV/electrolysis system by only performing small changes because they owned the same fundamental processes. The obtained results indicated that the series interconnected device concepts could provide a simple approach towards highly efficient water splitting, and could be extended to other solar absorbers with good opto-electronic properties [106]. Recently, Luo and coworkers demonstrated a panchromatic PV/PEC tandem water-splitting cell comprising a semi-transparent and high open-circuit voltage perovskite solar cell and a state-of-the-art CIGS photocathode, which led to a STH conversion efficiency of 6%, which is the highest value obtained for a PV/PEC device using only one single-junction solar cell as the bias source under one sun illumination. The analysis showed that the efficiency could reach more than 20% through further optimization [107].

(A) Sketch of the monolithic PV/PEC configuration seen from above. (B) Photo of the device in action. Reprinted with permission from ref. [106]. Copyright 2013 Royal Society of Chemistry. PV: photovoltaic; PEC: photoelectrochemical.
Figure 14

(A) Sketch of the monolithic PV/PEC configuration seen from above. (B) Photo of the device in action. Reprinted with permission from ref. [106]. Copyright 2013 Royal Society of Chemistry. PV: photovoltaic; PEC: photoelectrochemical.

6 Summary and prospect

As exemplified in this review, copper-based chalcogenides mainly composed of copper chalcopyrite and kesterite semiconductors have been demonstrated to be efficient photocathode materials for PEC water splitting. A basic examination of the physicochemical properties such as crystal structure, defect state, and optical property have provided a fundamental understanding about the favorable properties of copper-based chalcogenides for water reduction. In the last decade, significant progress has been achieved in this expanding research area. The HC-STH efficiency has rapidly reached 8.5%, and the most stable photocathode can last for 20 days without the obvious photocurrent decay. Of particular interest, the modification strategies play a vital role in improving the PEC performance. Composition tuning can easily change the band structures, giving rise to the optimal photocurrent and onset potential. Surface modifications, including pn junction construction, protection layer deposition, and HER catalyst loading can effectively improve the charge separation, enhance the stability, and accelerate the surface chemical reaction. Impurity elimination is an effective route to improve the purity of copper-based chalcogenide materials for the efficient charge transfer. Morphology design can provide a feasible approach to tune their light absorption property, charge transport behavior, and surface reaction rate. However, as the valence band edges of copper-based chalcogenide materials lie above the water oxidation potential, a large external bias is always needed for the efficient PEC water splitting. In order to achieve unassisted water splitting, p n and PV/PEC tandem cells have been developed. A STH efficiency over 10% can be obtained by introducing an interconnected PV/PEC tandem configuration. These results verify the attractive potential of copper-based chalcogenides in the practical PEC hydrogen production system.

At present, several drawbacks still exist for the practical application using copper-based chalcogenides for PEC water splitting. More work should be carried out to understand the fundamental principles, improve the activity and stability, as well as reduce the cost and toxic-ity. Forming a p n junction is a useful approach to improve the activities of copper-based chalcogenide photocathodes. However, the working principle is not well un derstood, and the proposed mechanisms in the literature are not consistent. Further studies should be performed to elucidate the underlining principle of the p n junction so as to boost the PEC performance. In addition, developing more suitable n-type semiconductors instead of CdS should be essential due to the high toxicity of Cd. Although several strategies have been applied to improve the PEC performance of copper-based chalcogenides, coupling different modification strategies to systematically optimize the light absorption, charge separation, and surface chemical reaction should be carried out in the future. The stability of copper-based chalcogenide materials in the electrolyte is an important issue for PEC water splitting. Relatively good stability has been demonstrated by depositing a TiO2 film as the protection layer. Detailed examination of the optimal fabrication parameters and physicochemi-cal properties of the protection layer should be executed to further improve the stability.

With regard to the cost, most of the efficient copper-based chalcogenide photoelectrodes consisted of scarce indium and gallium, and platinum is always indispensable as HER catalyst to promote the water reduction, which inevitably restricted the practical application. Hence, more efforts should be devoted to improve the efficiencies of alternative photoelectrodes with abundant elements such as CZTS, as well as exploit appropriate noble-metal-free HER catalysts. Seeking economical and scalable non-vacuum method is also indispensable to reduce the fabrication cost. Meanwhile, investigation of the matching principles of p-type copper-based chalcogenides and n-type semiconductors in a p n tandem cell should be performed to achieve highly efficient PEC water splitting without any external bias. Overall, it is expected that this comprehensive review can provide some insights of fundamentals and recent progress of copper-based chalcogenide thin films for PEC water splitting, and help advance the research in this expanding area.

Acknowledgement

The authors thank the financial support from the National Natural Science Foundation of China (Nos. 51323011, 51236007, and 51502240), the Program for New Century Excellent Talents in University (No. NCET-13-0455), the Natural Science Foundation of Shaanxi Province (No. 2014KW07-02), the Natural Science Foundation of Jiangsu Province (No. BK20141212), the Nano Research Program of Suzhou City (Nos. ZXG201442 and ZXG2013003), the China Postdoctoral Science Foundation (Nos. 2014M560768 and 2014M560769), the Foundation for the Author of National Excellent Doctoral Dissertation of China (No. 201335), the National Program for Support of Top-notch Young Professionals, and the Fundamental Research Funds for the Central Universities.

References

  • [1]

    Turner JA. Sustainable hydrogen production. Science 2004; 305:972–4. Google Scholar

  • [2]

    Barreto L, Makihira A, Riahi K. The hydrogen economy in the 21st century: a sustainable development scenario. Int J Hydrogen Energy 2003; 28:267–84. Google Scholar

  • [3]

    Bard AJ, Fox MA. Artificial photosynthesis-solar splitting of water to hydrogen and oxygen. Acc Chem Res 1995; 28:141–5. Google Scholar

  • [4]

    Walter MG, Warren EL, McKone JR, Boettcher SW, Mi QX, San-tori EA, Lewis NS. Solar water splitting cells. Chem Rev 2010; 110:6446–73. Google Scholar

  • [5]

    Tachibana Y, Vayssieres L, Durrant JR. Artificial photosynthesis for solar water-splitting. Nat Photonics 2012; 6:511–8.Google Scholar

  • [6]

    Mao SS, Shen SH. Hydrogen production catalysing artificial photosynthesis. Nat Photonics 2013; 7:944–6.Google Scholar

  • [7]

    Shen SH, Mao SS. Nanostructure designs for effective solar-to-hydrogen conversion. Nanophotonics 2012; 1:31–50.Google Scholar

  • [8]

    Han ZJ, Eisenberg R. Fuel from water: The photochemical generation of hydrogen from water. Acc Chem Res 2014; 47:2537–44. Google Scholar

  • [9]

    Fujishima A, Honda K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972; 238:37–8. Google Scholar

  • [10]

    Chen XB, Shen SH, Guo LJ, Mao SS. Semiconductor-based pho-tocatalytic hydrogen generation. Chem Rev 2010; 110:6503–70. Google Scholar

  • [11]

    Hisatomi T, Kubota J, Domen K. Recent advances in semiconductors for photocatalytic and photoelectrochemical water splitting. Chem Soc Rev 2014; 43:7520–35.Google Scholar

  • [12]

    Osterloh FE. Inorganic nanostructures for photoelectrochemical and photocatalytic watersplitting. Chem Soc Rev2013; 42:2294320.Google Scholar

  • [13]

    Kudo A, MisekiY. Heterogeneous photocatalyst materials for water splitting. Chem Soc Rev2009; 38:253–78.Google Scholar

  • [14]

    Kronawitter CX, Vayssieres L, Shen SH, Guo LJ, Wheeler DA, Zhang JZ, Antoun BR, Mao SS. A perspective on solar-driven wa-tersplittingwith all-oxide hetero-nanostructures. Energy Environ Sci 2011; 4:3889–99.Google Scholar

  • [15]

    Sun K, Shen SH, Liang YQ, Burrows PE, Mao SS, Wang DL. Enabling silicon for solar-fuel production. Chem Rev 2014; 114:8662–719.Google Scholar

  • [16]

    Ida S, Ishihara T. Recent progress in two-dimensional oxide pho-tocatalysts for water splitting. J PhysChem Lett 2014; 5:25 33–42.Google Scholar

  • [17]

    Sivula K, Le Formal F, Grätzel M. Solar water splitting: progress using hematite (α-Fe2O3) photoelectrodes. ChemsusChem 2011; 4:432–49.Google Scholar

  • [18]

    Li Y, Zhang JZ. Hydrogen generation from photoelectrochemi-cal water splitting based on nanomaterials. Laser Photonics Rev 2010; 4:517–28.Google Scholar

  • [19]

    Ismail AA, Bahnemann DW. Photochemical splitting of water for hydrogen production by photocatalysis: A review. Sol Energy Mater Sol Cells 2014; 128:85–101.Google Scholar

  • [20]

    Li JT, Wu NQ. Semiconductor-based photocatalysts and photo-electrochemical cells for solar fuel generation: a review. Catal Sci Technol 2015; 5:1360–84. Google Scholar

  • [21]

    Prevot MS, Sivula K. Photoelectrochemical tandem cells for solar water splitting. J Phys Chem C 2013; 117:17879–93. Google Scholar

  • [22]

    Huang Q, Ye Z, Xiao XD. Recent progress in photocathodes for hydrogen evolution. J Mater Chem A 2015; 3:15824–37. Google Scholar

  • [23]

    Awad NK, Ashour EA, Allam NK. Recent advances in the use of metal oxide-based photocathodes for solar fuel production. J Renew Sustain Energy 2014; 6:022702–21. Google Scholar

  • [24]

    Green MA, Emery K, Hishikawa Y, Warta W, Dunlop ED. Solar cell efficiency tables (version 45). Prog Photovolt Res Appl 2015; 23:1–9. Google Scholar

  • [25]

    Calixto ME, Sebastian PJ, Bhattacharya RN, Noufi R. Compositional and optoelectronic properties of CIS and CIGS thin films formed by electrodeposition. Sol Energy Mater Sol Cells 1999; 59:75–84.Google Scholar

  • [26]

    Jager-Waldau A. Progress in chalcopyrite compound semiconductor research for photovoltaic applications and transfer of results into actual solar cell production. Sol Energy Mater Sol Cells 2011; 95:1509–17. Google Scholar

  • [27]

    Ikeda S, Nakamura T, Lee SM, Yagi T, Harada T, Minegishi T, Mat-sumura M. Photoreduction of water by using modified CuInS2 electrodes. ChemsusChem 2011; 4:262–8. Google Scholar

  • [28]

    Moriya M, Minegishi T, Kumagai H, Katayama M, Kubota J, Domen K. Stable hydrogen evolution from CdS-modified CuGaSe2 photoelectrode under visible-light irradiation. J Am Chem Soc 2013; 135:3733–5. Google Scholar

  • [29]

    Valderrama RC, Sebastian PJ, EnriquezJP, Gamboa SA. Photoelectrochemical characterization of CIGS thin films for hydrogen production. Sol Energy Mater Sol Cells 2005; 88:145–55. Google Scholar

  • [30]

    Yokoyama D, Minegishi T, Jimbo K, Hisatomi T, Ma GJ, Katayama M, Kubota J, Katagiri H, Domen K. H2 evolution from water on modified Cu2ZnSnS4 photoelectrode under solar light. Appl Phys Express 2010; 3:101202(1-3). Google Scholar

  • [31]

    Jiang XS, Lambrecht WRL. Electronic band structure of ordered vacancy defect chalcopyrite compounds with formula II-III2-VI4. Phys Rev B 2004; 69:035201(1-8). Google Scholar

  • [32]

    JaffeJE, Zunger A. Theory of the band-gap anomaly in ABC2 chalcopyrite semiconductors. Phys Rev B 1984; 29:1882–906. Google Scholar

  • [33]

    Walsh A, Chen SY, Wei SH, GongXG. Kesterite thin-film solar cells: advances in materials modelling of Cu2ZnSnS4. AdvEnergy Mater 2012; 2:400–9. Google Scholar

  • [34]

    Zhang SB, Wei SH, Zunger A, Katayama-Yoshida H. Defect physics of the CuInSe2 chalcopyrite semiconductor. Phys Rev B 1998; 57:9642–56. Google Scholar

  • [35]

    Wang CC, Chen SY, Yang JH, Lang L, Xiang HJ, Gong XG, Walsh A, Wei SH. Design of I2-II-IV-VI4 semiconductors through element substitution: the thermodynamic stability limit and chemical trend. Chem Mater 2014; 26:3411–7. Google Scholar

  • [36]

    Wei SH, Zhang SB. Defect properties of CuInSe2 and CuGaSe2. J Phys Chem Solids 2005; 66:1994–9. Google Scholar

  • [37]

    Chen SY, Gong XG, Walsh A, Wei SH. Defect physics of the kesterite thin-film solar cell absorber Cu2ZnSnS4 .Appl Phys Lett 2010; 96:021902(1-3). Google Scholar

  • [38]

    Chen SY, Yang JH, GongXG, Walsh A, Wei SH. Intrinsic point defects and complexes in the quaternary kesterite semiconductor Cu2ZnSnS4. Phys Rev B 2010; 81:245204(1-10). Google Scholar

  • [39]

    Steinhagen C, Panthani MG, Akhavan V, Goodfellow B, Koo B, Korgel BA. Synthesis of Cu2ZnSnS4 nanocrystals for use in low-cost photovoltaics. J Am Chem Soc 2009; 131:12554–5. Google Scholar

  • [40]

    Bär M, Bohne W, Röhrich J, Strub E, Lindner S, Lux-Steiner MC, Fischer CH, Niesen TP, Karg F. Determination of the band gap depth profile of the penternary Cu(ln(1-x)Gax)(SySe(1-y))2 chalcopyrite from its composition gradient. J Appl Phys 2004; 96:3857–60. Google Scholar

  • [41]

    Valderrama RC, Sebastián PJ, Miranda-Hernandez M, Enriquez JP, Gamboa SA. Studies on the electrochemical stability of CIGS in H2SO4. J Photochem Photobiol A Chem 2004; 168:75–80. Google Scholar

  • [42]

    Marsen B, Dorn S, Cole B, Rocheleau RE, Miller EL. Copper chalcopyrite film photocathodes for direct solar-powered water splitting. Mater Res Soc Symp Proc 2007; 974:0974-CC09–05. Google Scholar

  • [43]

    Jacobsson TJ, Platzer-Björkman C, Edoff M, Edvinsson T. CuInxGa1_xSe2 as an efficient photocathode for solar hydrogen generation. IntJ Hydrogen Energy 2013; 38:15027–35. Google Scholar

  • [44]

    Yokoyama D, Minegishi T, Maeda K, Katayama M, Kubota J, Ya-mada A, Konagai M, Domen K. Photoelectrochemical water splitting using a Cu(ln, Ga)Se2 thin ilm. Electrochem Commun 2010; 12:851–3. Google Scholar

  • [45]

    Kumagai H, Minegishi T, Sato N, Yamada T, Kubota J, Domen K. Efficient solar hydrogen production from neutral electrolytes us-ingsurface-modified Cu(In, Ga)Se2 photocathodes. J Mater Chem A 2015; 3:8300–7. Google Scholar

  • [46]

    Mandati S, Sarada BV, Dey SR, Joshi SV. Photoelectrochemistry of Cu(In, Ga)Se2 thin-films fabricated by sequential pulsed elec-trodeposition. J Power Sources 2015; 273:149–57. Google Scholar

  • [47]

    Marsen B, Cole B, Miller EL. Photoelectrolysis of water usingthin copper gallium diselenide electrodes. Sol Energy Mater Sol Cells 2008; 92:1054–8. Google Scholar

  • [48]

    Kumagai H, MinegishiT, MoriyaY, KubotaJ, Domen K. Photoelec-trochennical hydrogen evolution from water using copper gallium selenide electrodes prepared by a particle transfer method. J Phys Chem C 2014; 118:16386–92. Google Scholar

  • [49]

    Zhao J, Minegishi T, Zhang L, Zhong M, Gunawan, Nakabayashi M, Ma GJ, Hisatomi T, Katayama M, Ikeda S, Shibata N, Yamada T, Domen K. Enhancement of solar hydrogen evolution from water by surface modification with CdS and TiO2 on porous CuInS2 photocathodes prepared by an electrodeposition-sulfurization method. Angew Chem Int Ed 2014; 53:11808–12. Google Scholar

  • [50]

    Luo JS, Tilley SD, Steier L, Schreier M, Mayer MT, Fan HJ, Grätzel M. Solution transformation of Cu2O into CuInS2 for solar water splitting. Nano Lett 2015; 15:1395–402. Google Scholar

  • [51]

    Zhou HP, Hsu WC, Duan HS, Bob B, Yang WB, Song TB, Hsu CJ, Yang Y. CZTS nanocrystals: a promising approach for next generation thin ilm photovoltaics. Energy Environ Sci 2013; 6:282238. Google Scholar

  • [52]

    Yang WB, Duan HS, Cha KC, Hsu CJ, Hsu WC, Zhou HP, Bob B, Yang Y. Molecular solution approach to synthesize electronic quality Cu2ZnSnS4 thin films. J Am Chem Soc 2013; 135:6915–20.Google Scholar

  • [53]

    YuXL, ShavelA, An XQ, Luo ZS, IbánezM, Cabot A. Cu2ZnSnS4-Pt and Cu2ZnSnS4-Au heterostructured nanoparticles for photocat-alytic water splitting and pollutant degradation. J Am Chem Soc 2014; 136:9236–9. Google Scholar

  • [54]

    Wang W, Winkler MT, Gunawan O, Gokmen T, Todorov TK, Zhu Y, Mitzi DB. Device characteristics of CZTSSe thin-film solar cells with 12.6% efficiency. Adv Energy Mater 2014; 4:1301465(1-5). Google Scholar

  • [55]

    Zhang YF, OuyangSX, Yu Q, Li P, YeJH. Modulation of sulfur partial pressure in sulfurization to signiicantly improve the photo-electrochemical performance over the Cu2ZnSnS4 photocathode. Chem Commun 2015; 51:14057–9. Google Scholar

  • [56]

    Rovelli L, Tilley SD, Sivula K. Optimization and stabilization of electrodeposited Cu2ZnSnS4 photocathodes for solar water reduction. ACS Appl Mater Interfaces 2013; 5:8018–24. Google Scholar

  • [57]

    Chen YB, Chuang CH, Lin KC, Shen SH, McCleese C, Guo LJ, Burda C. Synthesis and photoelectrochemical properties of (Cu2Sn)xZn3(1-x)S3 nanocrystal films. J Phys Chem C 2014; 118:11954–63. Google Scholar

  • [58]

    Wang J, Zhang P, SongXF, Gao L. Cu2ZnSnS4 thin films: spin coating synthesis and photoelectrochemistry. RSC Adv 2014; 4:21318–24. Google Scholar

  • [59]

    Jiang F, Gunawan, Harada T, Kuang YB, Minegishi T, Domen K, Ikeda S. Pt/In2S3/CdS/Cu2ZnSnS4 thin film as an efficient and stable photocathode for water reduction under sunlight radiation. J Am Chem Soc 2015; 137:13691–7. Google Scholar

  • [60]

    Kaneko H, Minegishi T, Domen K. Chalcopyrite thin film materials for photoelectrochemical hydrogen evolution from water under sunlight. Coatings 2015; 5:293–311. Google Scholar

  • [61]

    Kim J, Minegishi T, Kobota J, Domen K. Investigation of Cu-deficient copper gallium selenide thin film as a photocathode for photoelectrochemical water splitting. Jpn J Appl Phys 2012; 51:015802(1-6). Google Scholar

  • [62]

    Kim J, Minegishi T, Kobota J, Domen K. Enhanced photoelectro-chemical properties of CuGa3Se5 thin films for water splitting by the hydrogen mediated co-evaporation method. Energy Environ Sci 2012; 5:6368–74. Google Scholar

  • [63]

    Chen SY, Gong XG, Wei SH. Band-structure anomalies of the chalcopyrite semiconductors CuGaX2 versus AgGaX2 (X=S and Se) and their alloys. Phys Rev B 2007; 75:205209(1-9). Google Scholar

  • [64]

    Zhang L, Minegishi T, Kubota J, Domen K. Hydrogen evolution from water using AgxCui1-xGaSe2 photocathodes under visible light. Phys Chem Chem Phys 2014; 16:6167–74. Google Scholar

  • [65]

    Zhang L, Minegishi T, Nakabayashi M, Suzuki Y, Seki K, Shibata N, Kubota J, Domen K. Durable hydrogen evolution from water driven by sunlight using (Ag, Cu)GaSe2 photocathodes modified with CdS and CuGa3Se5. Chem Sci 2015; 6:894–901. Google Scholar

  • [66]

    Ikeda S, Nonogaki M, Septina W, Gunawan G, Harada T, Mat-sumura M. Fabrication of CuInS2 and Cu(In, Ga) S2 thin films by a facile spray pyrolysis and their photovoltaic and photoelectro-chemical properties. Catal Sci Technol 2013; 3:1849–54. Google Scholar

  • [67]

    Septina W, Gunawan, Ikeda S, Harada T, Higashi M, Abe R, Mat-sumura M. Photosplitting of water from wide-gap Cu(In, Ga)S2 thin films modified with a CdS layer and Pt nanoparticles for a high-onset-potential photocathode. J Phys Chem C 2015; 119:8576–83. Google Scholar

  • [68]

    Kaga H, Tsutsui Y, NaganeA, IwaseA, Kudo A. An effect of Ag(I)-substitution at Cu sites in CuGaS2 on photocatalytic and photo-electrochemical properties for solar hydrogen evolution. J Mater Chem A2015; 3:21815–23. Google Scholar

  • [69]

    Kato T, Hakari Y, Ikeda S, Jia QX, Iwase A, Kudo A. Utilization of metal sulfide material of (CuGa)(1-x)Zn2xS2 solid solution with visible light response in photocatalytic and photoelectrochemi-cal solar water splitting systems. J Phys Chem Lett 2015; 6:10427. Google Scholar

  • [70]

    Gunawan, Septina W, Ikeda S, Harada T, Minegishi T, Domen K, Matsumura M. Platinum and indium sulfide-modified CuInS2 as efficient photocathodes for photoelectrochemical water splitting. Chem Commun 2014; 50:8941–3. Google Scholar

  • [71]

    Mali MG, Yoon H, Joshi BN, Park H, Al-Deyab SS, Lim DC, Ahn S, Nervi C, Yoon SS. Enhanced photoelectrochemical solar water splitting using a platinum-decorated CIGS/CdS/ZnO photocathode. ACS Appl Mater Interfaces 2015; 7:21619–25. Google Scholar

  • [72]

    Guijarro N, Prevot MS, Sivula K. Enhancing the charge separation in nanocrystalline Cu2ZnSnS4 photocathodes for photoelectrochemical application: the role of surface modifications. J Phys Chem Lett 2014; 5:3902–8. Google Scholar

  • [73]

    Huang S, Luo WJ, Zou ZG. Band positions and photoelectro-chemical properties of Cu2ZnSnS4 thin films by the ultrasonic spray pyrolysis method. J Phys D Appl Phys 2013; 46:235108(1-6). Google Scholar

  • [74]

    Gunawan, Septina W, Harada T, Nose Y, Ikeda S. Investigation of the electric structures of heterointerfaces in Pt- and In2S3-modified CuInS2 photocathodes used for sunlight-induced hydrogen evolution. ACS Appl Mater Interfaces 2015; 7:16086–92. Google Scholar

  • [75]

    Jacobsson TJ, Fjällström V, Edoff M, Edvinsson T. Sustainable solar hydrogen production: From photoelectrochemical cells to PV-electrolyzers and back again. Energy Environ Sci 2014; 7:205670. Google Scholar

  • [76]

    Jacobsson TJ, Fjällström V, Edoff M, Edvinsson T. CIGS based devices for solar hydrogen production spanning from PEC-cells to PV-electrolyzers: a comparison of efficiency, stability and device topology. Sol Energy Mater Sol Cells 2015; 134:185–93. Google Scholar

  • [77]

    Azarpira A, Lublow M, Steigert A, Bogdanoff P, Greiner D, Kaufmann CA, Kruger M, Gernert U, Krol Rvd, Fischer A, Schedel-Niedrig T. Efficient and stable TiO2:Pt-Cu(In, Ga)Se2 composite photoelectrodes for visible light driven hydrogen evolution. Adv Energy Mater 2015; 5:1402148(1-9). Google Scholar

  • [78]

    Neumann B, Bogdanoff P, Tributsch H. TiO2-protected photoelectrochemical tandem Cu(In, Ga)Se2 thin film membrane for light-induced water splitting and hydrogen evolution. J Phys Chem C 2009; 113:20980–9. Google Scholar

  • [79]

    Yang F, Kuznietsov V, Lublow M, Merschjann C, Steigert A, Klaer J, Thomas A, Schedel-Niedrig T. Solar hydrogen evolution using metal-free photocatalytic polymeric carbon nitride/CuInS2 composites as photocathodes. J Mater Chem A 2013; 1:6407–15. Google Scholar

  • [80]

    Seger B, Laursen AB, Vesborg PCK, Pedersen T, Hansen O, Dahl S, Chorkendorff I. Hydrogen production usinga molybdenum sulfide catalyst on a titanium-protected n plus p-silicon photocathode. Angew Chem Int Ed 2012; 51:9128–31. Google Scholar

  • [81]

    Morales-Guio CG, Tilley SD, Vrubel H, Grätzel M, Hu XL. Hydrogen evolution from a copper(I) oxide photocathode coated with an amorphous molybdenum sulphide catalyst. Nat Commun 2014; 5:3059(1-7).Google Scholar

  • [82]

    Huang ZP, Wang CF, Chen ZB, Meng H, Lv CC, Chen ZZ, Han RQ, Zhang C. Tungsten sulfide enhancing solar-driven hydrogen production from silicon nanowires. ACS Appl Mater Interfaces 2014; 6:10408–14.Google Scholar

  • [83]

    Qin ZX, Chen YB, WangXX, Guo X, Guo LJ. Intergrowth of cocat-alysts with host photocatalysts for improved solar-to-hydrogen conversion. ACS Appl Mater Interfaces 2016; 8:1264–72. Google Scholar

  • [84]

    Guan ZJ, Luo WJ, Feng JY, Tao QC, Xu Y, Wen X, Fu G, Zou ZG. Selective etching of metastable phase induced an efficient CuIn0.7Ga0.3S2 nano-photocathode for solar water splitting. J Mater Chem A 2015; 3:7840–8. Google Scholar

  • [85]

    Guan ZJ, Luo WJ, Zou ZG. Formation mechanism of ZnS impurities and their effect on photoelectrochemical properties on a Cu2ZnSnS4 photocathode. CrystEngComm 2014; 16:2929–36. Google Scholar

  • [86]

    Yang W, Oh Y, Kim J, Kim H, Shin H, Moon J. Photoelectrochem-ical properties of vertically aligned CuInS2 nanorod arrays prepared via template-assisted growth and transfer. ACS Appl Mater Interfaces 2016; 8:425–31. Google Scholar

  • [87]

    Li BJ, Yin PF, Zhou YZ, Gao ZM, Ling T, Du XW. Single crystalline Cu2ZnSnS4 nanosheet arrays for efficient photochemical hydrogen generation. RSC Adv 2015; 5:2543–9. Google Scholar

  • [88]

    Wen X, Luo WJ, Zou ZG. Photocurrent improvement in nanocrys-talline Cu2ZnSnS4 photocathodes by introducing porous structures. J Mater Chem A 2013; 1:15479–85. Google Scholar

  • [89]

    Ma GJ, Minegishi T, Yokoyama D, Kubota J, Domen K. Photoelec-trochemical hydrogen production on Cu2ZnSnS4/Mo-mesh thin-film electrodes prepared by electroplating. Chem Phys Lett 2011; 501:619–22. Google Scholar

  • [90]

    Nozik AJ. P-n photoelectrolysis cells. Appl Phys Lett 1976; 29:150–3. Google Scholar

  • [91]

    Ingler WB, Khan SUM. A self-driven p/n-Fe2O3 tandem photo-electrochemical cell for water splitting. Electrochem Solid-State Lett 2006; 9:G144–6. Google Scholar

  • [92]

    Wang HL, Turner JA. Characterization of hematite thin films for photoelectrochemical water splitting in a dual photoelectrode device. J Electrochem Soc 2010; 157:173–8. Google Scholar

  • [93]

    Iwase A, Ng YH, Amal R, Kudo A. Solar hydrogen evolution using a CuGaS2 photocathode improved by incorporating reduced graphene oxide. J Mater Chem A 2015; 3:8566–70. Google Scholar

  • [94]

    Kim JH, Kaneko H, Minegishi T, Kubota J, Domen K, Lee JS. Overall photoelectrochemical water splitting using tandem cell under simulated sunlight. ChemsusChem 2016; 9:61–6. Google Scholar

  • [95]

    Menezes S, Li Y. Potential of electrodeposited copper indium selenide thin-films for various solar energy conversion devices. J Electrochem Soc 2014; 161:3083–7. Google Scholar

  • [96]

    Ding CM, Qin W, Wang N, Liu GJ, Wang ZL, Yan PL, Shi JY, Li C. Solar-to-hydrogen efficiency exceeding 2.5% achieved for overall water splitting with an all earth-abundant dual-photoelectrode. Phys Chem Chem Phys 2014; 16:15608–14. Google Scholar

  • [97]

    Khaselev O, Turner JA. A monolithic photovoltaic-photoelectrochemical device for hydrogen production via water splitting. Science 1998; 280:425–7. Google Scholar

  • [98]

    Döscher H, Young JL, Geisz JF, Turner JA, Deutsch TG. Solar-to-hydrogen efficiency: shining light on photoelectrochemical device performance. Energy Environ Sci 2016; 9:74–80.Google Scholar

  • [99]

    Brillet J, Yum JH, Cornuz M, Hisatomi T, Solarska R, Augustynski J, Graetzel M, Sivula K. Highly efficient water splitting by a dual-absorber tandem cell. Nat Photonics 2012; 6:824–8.Google Scholar

  • [100]

    Abdi FF, Han L, Smets AHM, Zeman M, Dam D, Krol RVD. Efficient solar water splitting by enhanced charge separation in a bismuth vanadate-silicon tandem photoelectrode. Nat Commun 2013; 4:2195(1-7). Google Scholar

  • [101]

    Kaneshiro J, Gaillard N, Rocheleau R, Miller E. Advances in copper-chalcopyrite thin films for solar energy conversion. Sol Energy Mater Sol Cells 2010; 94:12–6. Google Scholar

  • [102]

    Miller EL, Marsen B, Paluselli D, Rocheleau R. Optimization of hybrid photoelectrodes for solar water-splitting. Electrochem Solid-State Lett 2005; 8:A247–9. Google Scholar

  • [103]

    Kaneshiro JM, Deangelis A, Song X, Gaillard N, Miller EL. I-III-VI2 (copper chalcopyrite-based) thin films for photoelectrochem-ical water-splitting tandem-hybrid photocathode. Mater Res Soc Symp Proc 2011; 1324:83–9. Google Scholar

  • [104]

    Gaillard N, Prasher D, Kaneshiro J, Mallory S, Chong M. Development of chalcogenide thin film materials for photoelectro-chemical hydrogen production. Mater Res Soc Symp Proc 2013; 1558:mrss13-1558-z02–07. Google Scholar

  • [105]

    Kaneshiro J, Chang Y, Gaillard N. Hybrid Photo-voltaic/photoelectrochemical device design using I-III-VI2 copper chalcopyrite-based photocathodes. Meeting Abstracts. The Electrochemical Society 2012; 02:1710. Google Scholar

  • [106]

    Jacobsson TJ, Fjällström V, Sahlberg M, Edoff M, Edvinsson T. A monolithic device for solar water splitting based on series interconnected thin film absorbers reaching over 10% solar-to-hydrogen eiciency. Energy Environ Sci 2013; 6:3676–83. Google Scholar

  • [107]

    Luo JS, Li Z, Nishiwaki S, Schreier M, Mayer MT, Cendula P, LeeYH, Fu K, Cao AY, Nazeeruddin MK, RomanyukYE, Buecheler S, Tilley SD, Wong LH, Tiwari AN, Grätzel M. Targeting ideal dual-absorber tandem water splitting using perovskite photo-voltaics and CuInxGa1-xSe2 photocathodes. Adv Energy Mater 2015; 5:1501520(1-8). Google Scholar

About the article

Received: 2016-01-18

Accepted: 2016-03-03

Published Online: 2016-10-20

Published in Print: 2016-09-01


Citation Information: Nanophotonics, Volume 5, Issue 4, Pages 524–547, ISSN (Online) 2192-8614, ISSN (Print) 2192-8606, DOI: https://doi.org/10.1515/nanoph-2016-0027.

Export Citation

© 2016 Y. Chen et al., published by De Gruyter Open. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License. BY-NC-ND 3.0

Citing Articles

Here you can find all Crossref-listed publications in which this article is cited. If you would like to receive automatic email messages as soon as this article is cited in other publications, simply activate the “Citation Alert” on the top of this page.

[1]
Christopher P. Muzzillo, W. Ellis Klein, Zhen Li, Alexander Daniel DeAngelis, Kimberly Horsley, Kai Zhu, and Nicolas Gaillard
ACS Applied Materials & Interfaces, 2018
[2]
Carles Ros, Teresa Andreu, Sergio Giraldo, Victor Izquierdo-Roca, Edgardo Saucedo, and Joan Ramon Morante
ACS Applied Materials & Interfaces, 2018
[3]
Yu-Xiang Yu, Linfeng Pan, Min-Kyu Son, Matthew T. Mayer, Wei-De Zhang, Anders Hagfeldt, Jingshan Luo, and Michael Grätzel
ACS Energy Letters, 2018, Page 760
[4]
Xiaoyang Feng, Yubin Chen, Menglong Wang, and Liejin Guo
International Journal of Hydrogen Energy, 2017, Volume 42, Number 22, Page 15085
[6]
Yubin Chen, Chi-Hung Chuang, Zhixiao Qin, Shaohua Shen, Tennyson Doane, and Clemens Burda
Nanotechnology, 2017, Volume 28, Number 8, Page 084002
[7]
Shaohua Shen, Sarah A. Lindley, Xiangyan Chen, and Jin Z. Zhang
Energy Environ. Sci., 2016, Volume 9, Number 9, Page 2744
[8]
Zhixiao Qin, Yubin Chen, Zhenxiong Huang, Jinzhan Su, Zhidan Diao, and Liejin Guo
The Journal of Physical Chemistry C, 2016, Volume 120, Number 27, Page 14581
[9]
Jinzhan Su, Tao Zhang, Yufeng Li, Yubin Chen, and Maochang Liu
Molecules, 2016, Volume 21, Number 6, Page 735
[10]
Yubin Chen, Zhixiao Qin, Tao Chen, Jinzhan Su, Xiaoyang Feng, and Maochang Liu
RSC Adv., 2016, Volume 6, Number 63, Page 58409

Comments (0)

Please log in or register to comment.
Log in