Efficient Redox-Mediator-Free Z-Scheme Water Splitting Employing

Efficient Redox-Mediator-Free Z-Scheme Water Splitting Employing Oxysulfide Photocatalysts under Visible Light ... Publication Date (Web): January 12,...
0 downloads 0 Views 2MB Size
Research Article Cite This: ACS Catal. 2018, 8, 1690−1696

pubs.acs.org/acscatalysis

Efficient Redox-Mediator-Free Z‑Scheme Water Splitting Employing Oxysulfide Photocatalysts under Visible Light Song Sun,†,‡ Takashi Hisatomi,† Qian Wang,† Shanshan Chen,† Guijun Ma,† Jingyuan Liu,† Swarnava Nandy,† Tsutomu Minegishi,† Masao Katayama,† and Kazunari Domen*,† †

Department of Chemical System Engineering, School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan ‡ National Synchrotron Radiation Laboratory, Collaborative Innovation Center of Chemistry for Energy Materials, University of Science & Technology of China, Hefei, Anhui 230029, People’s Republic of China S Supporting Information *

ABSTRACT: Oxysulfides, which exhibit intense visible light absorption and high photocatalytic activity during hydrogen evolution, are promising photocatalysts for Z-scheme water splitting. However, the realization of efficient Z-scheme water splitting using oxysulfides as hydrogen evolution photocatalysts (HEPs) has been hampered by backward reactions involving reversible redox couples, the low efficiency of interparticle electron transfer, and a lack of knowledge regarding the means of promoting hydrogen evolution activity under nonsacrificial conditions. In this study, La5Ti2CuS5O7 was found to realize water splitting without any redox mediator by constructing Zscheme-type photocatalyst sheets with BiVO4 as the oxygen evolution photocatalyst. It is also demonstrated that p-type doping and the formation of a La5Ti2Cu0.9Ag0.1S5O7 solid solution effectively enhance the water-splitting activity of the photocatalyst sheet. An apparent quantum yield of 4.9% at 420 nm and a solar to hydrogen energy conversion efficiency of 0.11% were obtained, both of which are among the highest values reported for water-splitting systems employing a particulate photocatalyst absorbing visible light up to ca. 700 nm. This work shows the potential for oxysulfides to serve as HEPs for Z-scheme water splitting and provides insights toward the development of efficient photocatalyst sheets. KEYWORDS: photocatalysis, water splitting, oxysulfides, Z-scheme, photocatalyst sheet efficiencies (STH) of 3 × 10−3% at most.20 The main obstacles to further improvements are backward reactions, that is, reduction and oxidation of redox couples on the HEP and OEP, respectively, and inefficient interparticle charge transfer between the HEP and the OEP.10,11 The concept of particulate photocatalyst sheets represents a new approach to addressing the aforementioned challenges associated with conventional Z-scheme systems.21 In these sheet structures, a thin particulate layer consisting of both HEP and OEP particles is embedded in a conductive layer, allowing intimate mechanical and electrical contact between particles.21−23 The conductive layer results in efficient electron transfer between the HEP and OEP. As an example, a sheet based on La- and Rh-codoped SrTiO3 as the HEP achieved a benchmark STH exceeding 1.1%.21,22 However, this sheet only utilized visible light up to 520 nm due to the short absorption

1. INTRODUCTION Direct water splitting into H2 and O2 over particulate semiconductor photocatalysts is one of the simplest models of artificial photosynthesis.1−5 The development of narrowband-gap semiconductors, such as (oxy)nitrides and (oxy)sulfides, as a means of effectively utilizing solar energy has attracted considerable attention.6−8 However, particulate semiconductors with narrow band gaps can rarely drive both the H2 and O2 evolution reactions simultaneously. Therefore, Zscheme water splitting based on the two-step excitation of a hydrogen evolution photocatalyst (HEP) and an oxygen evolution photocatalyst (OEP) has become a popular approach for the efficient splitting of water into H2 and O2.9−14 (Oxy)sulfides have been applied to Z-scheme water splitting systems in which electron transfer is mediated by a solid-state electron mediator (such as Au15 or reduced graphene oxide16,17) or an aqueous redox mediator (including Co complexes,18 Fe(CN)64−,19 or I3−/I− 20). However, existing visible-light-driven Z-scheme systems involving (oxy)sulfide photocatalysts exhibit solar to hydrogen energy conversion © XXXX American Chemical Society

Received: November 14, 2017 Revised: January 9, 2018 Published: January 12, 2018 1690

DOI: 10.1021/acscatal.7b03884 ACS Catal. 2018, 8, 1690−1696

Research Article

ACS Catalysis

corresponding amount of Ag2S was added. In the present work, the value of x was 0.1 unless otherwise noted. Ga-doped LTCA (Ga-LTCA) was obtained by adding an appropriate amount of Ga2O3 while the amount of TiO2 was reduced. BiVO4 powder was synthesized by a solid−liquid reaction previously reported in the literature.30 First, K3V5O14 having a layered structure was obtained for use as a precursor via the calcination of a mixture of K2CO3 (12 mmol, Kanto Chemical, 99.5%) and V2O5 (20 mmol, Wako Pure Chemical, 99.0%) in air at 723 K for 5 h. The resulting K3V5O14 powder (4 mmol) was subsequently added to an aqueous solution of Bi(NO3)3· 5H2O (20 mmol, Kanto Chemical, 99.9%) with stirring at 343 K for 10 h to form BiVO4. The BiVO4 product was collected by filtration and washed with distilled water. 2.2. Preparation of Photocatalyst Sheets. The asprepared particulate photocatalysts LTC, LTA, and LTCA, acting as HEPs, and the BiVO4, acting as an OEP, were embedded in a Au layer using a particle transfer method to fabricate photocatalyst sheets.21 Taking LTC/Au/BiVO4 as an example, LTC powder (10 mg) and BiVO4 powder (10 mg) were dispersed in isopropyl alcohol (0.5 mL) by ultrasonication and subsequently deposited on a glass substrate (3 × 3 cm) that had been previously washed with acetone and distilled water. The substrate coated with this suspension was dried naturally at room temperature, after which a thin Au layer (370 nm) was deposited by vacuum evaporation (VFR-200 M/ERH, ULVAC KIKO, Inc.) at a rate of approximately 8 nm s−1. For crosssection observation, a thicker Au layer was deposited for handling. A second glass substrate covered with adhesive carbon tape (Nisshin EM Co., Ltd.) was then used to lift the Au layer holding the photocatalyst particles from the first glass substrate. Excess particles on top of the Au layer on the photocatalyst sheet were subsequently removed by twice ultrasonicating the sheet in distilled water, using an initial duration of 2 min followed by an exchange of fresh distilled water and a second ultrasonication for 1 min. 2.3. Photodeposition of Cocatalysts. Rh species were initially photodeposited on the photocatalyst sheets using solutions of RhCl3·3H2O in distilled water (40 mL) over a span of 3 h. A 0.4 μmol quantity of RhCl3 was employed in conjunction with a photocatalyst sheet having a geometric area of 8 cm2. The photodeposition of Ru and Pt species was also examined respectively in some experiments. Cr2O3 layers intended to suppress the backward reactions were subsequently photodeposited in the same manner, using a 0.2 μmol K2CrO4 solution for 2 h.31 The photodeposition reactions were carried out under visible light irradiation (λ >420 nm) from a Xe lamp using a closed circulation system. This same system was also used to perform the photocatalytic water splitting reaction trials described in detail below. 2.4. Photocatalytic Overall Water Splitting. Watersplitting reactions over the photocatalyst sheets were carried out using a Pyrex top-irradiation reactor connected to a glass closed gas circulation system. In each trial, a photocatalyst sheet sample (3 × 3 cm, effective area ∼8 cm2) was placed at the bottom of the reactor, which held 40 mL of distilled water. Prior to the reaction, the reactor was thoroughly evacuated to remove air and Ar was introduced into the system to adjust the initial pressure. The reaction solution was irradiated using a 300 W Xe lamp equipped with a cutoff filter (λ ≥420 nm). Each reaction was carried out while the solution temperature was maintained at approximately 288 K using a flow of cooling water. The evolved gases were analyzed by gas chromatography

edge of the photocatalysts. In contrast, other Z-scheme photocatalysts incorporating the oxynitride LaMg1/3Ta2/3O2N (with an absorption edge at 600 nm) and ZnRh2O4 (response to the wavelength up to 740 nm) as the HEP only give an STH of approximately 1 × 10−3% and an apparent quantum yield (AQY) value of 0.07% at most in the visible light region, respectively.24,25 Therefore, the development of efficient HEPs for photocatalyst sheets is an essential step toward the realization of higher performances during the water-splitting reaction. La5Ti2CuS5O7 (LTC) has an absorption edge wavelength of 660 nm, and its photocatalytic and photoelectrochemical (PEC) H2 evolution activities can be significantly changed by doping and forming a La5Ti2Cu0.9Ag0.1S5O7 (LTCA) solid solution.26−28 Moreover, both LTC and LTCA have been applied as HEPs for Z-scheme water splitting, either as powder suspensions or in a p/n PEC cell.20,29 Therefore, the use of these oxysulfides as HEPs in particulate photocatalyst sheets offers the opportunity to both harvest long-wavelength visible light and elucidate the properties governing the performance of such sheets. Herein, we report the application of photocatalyst sheets based on LTC as the HEP and BiVO4 as the OEP to Zscheme water splitting without a redox mediator. A mature BiVO4 is adopted as the OEP to identify the reaction properties of oxysulfide-based HEPs in the Z-scheme photocatalyst sheet system because BiVO4 has been used as an authentic and active OEP in many Z-scheme systems. The activity of such sheets is shown to be boosted by enhancing the activity of the LTC as a photocathode through p-type doping and the formation of an LTCA solid solution, yielding an AQY of 4.9% at 420 nm and a STH of 0.11% during Z-scheme water splitting.

2. EXPERIMENTAL SECTION 2.1. Preparation of Photocatalysts. LTC was prepared by a solid-state reaction on the basis of our previous studies.26 Initially, La2O3 (99.99%, Kanto Chemical Co., Inc.), La2S3 (99.9%, Kojundo Chemical Laboratory Co., Ltd.), TiO2 (99.99%, Rare Metallic Co., Ltd.), Cu2S (99%, Kojundo Chemical Laboratory Co., Ltd.), and sulfur (99.99%, Kojundo Chemical Laboratory Co., Ltd.) were mixed at a La2O3:La2S3:TiO2:Cu2S:S molar ratio of 8:12:16:4:1 in a glovebox under nitrogen. The La 2 O 3 and TiO 2 were respectively calcined at 1273 K for 10 h and 1073 K for 1 h just prior to mixing, and an excess of sulfur was added to suppress the generation of oxide impurities. This precursor mixture was ground for 30 min and then sealed in an evacuated quartz tube. The tube was subsequently heated from room temperature to 473 K in 9 min, from 473 to 673 K in 100 min, and from 673 to 1273 K in 50 h and then maintained at 1273 K for 48 h. After the sample was allowed to naturally cooled to ambient temperature, it was ground into a powder. La5Ti2AgS5O7 (LTA) was prepared in the same manner but using Ag2S (99%, Kojundo Chemical Laboratory Co., Ltd.) instead of Cu2S. For the synthesis of Ga-, Al-, Sc-, and Mgdoped LTC, appropriate amounts of Ga2O3 (99.99%, Soekawa Chemicals), Al2O3 (99.99%, Soekawa Chemicals), Sc2O3 (99.95%, Kanto Chemical Co., Inc.), and MgO (99.9%, Wako Pure Chemical Industries, Ltd.) were respectively added and the corresponding amount of TiO2 was reduced.26 LTCA solid solution powders were synthesized by a solidstate reaction similar to the process described above, but the final heating at 1273 K was extended to 96 h.28 In these syntheses, the amount of Cu2S was reduced and the 1691

DOI: 10.1021/acscatal.7b03884 ACS Catal. 2018, 8, 1690−1696

Research Article

ACS Catalysis

Figure 1. (a) Top view and (b) cross-sectional SEM images of LTC/Au/BiVO4 particulate sheet and SEM-EDX mapping images for (c) superimposition of (d) La, (e) Bi, and (f) Au.

where R(H2), ΔG, P, and S are the rate of hydrogen evolution from the Z-scheme water splitting system, the change in Gibbs free energy that accompanies water splitting, the energy intensity of the solar light irradiation, and the effective irradiation area (∼8 cm2), respectively. The values of ΔG were 227, 226, and 220 kJ mol−1 at 275, 288, and 313 K, respectively, at a background pressure of 5 kPa. The energy intensity of the solar light irradiation was 100 mW cm−2. 2.7. Characterization. Scanning electron microscopy (SEM) and energy dispersive X-ray fluorescence spectroscopy (EDX) images were obtained with Hitachi S-4700 and JEOL JSM-7001F instruments. X-ray photoelectron spectroscopy (XPS) data were acquired with a JEOL JPS-9000 apparatus using a monochromatic Mg Kα source operating at 10 kV and 10 mA. The analysis chamber pressure was on the order of 10−6 Pa. The binding energies were corrected using the C 1s peak of surface adventitious carbon at 284.8 eV. X-ray diffraction (XRD) patterns were obtained using a Rigaku Ultima III with Cu Kα radiation, operating at 40 kV and 40 mA. Diffuse reflectance spectroscopy (DRS) was performed with a UV/vis− infrared diffuse reflectance spectrometer (V-670, JASCO).

(Shimadzu GC-8A) using a thermal conductivity detector, a 5 Å molecular sieve column, and Ar as the carrier gas. When the powder suspensions were evaluated, a mixture of 0.1 g of one of the LTC-based materials (as the HEP) and 0.1 g of BiVO4 (as the OEP) was dispersed in 150 mL of distilled water. In the case of a half-reaction, 0.15 g of the photocatalyst was used. The redox mediator and sacrificial reagents are noted in the figure captions or table footnotes. The irradiation area was approximately 48 cm2. 2.5. Quantum Yield Measurements. AQY obtained during Z-scheme water splitting over each photocatalyst sheet was determined using the same experimental setup but with regulation of the light irradiation wavelength using a series of band-pass filters. The number of photons received by the photocatalyst sheet was measured using a grating spectroradiometer (EKO Instruments Co., Ltd., LS-100), and the AQY was calculated using the equation AQY (%) = [4 × n(H 2)]/n(photons) × 100

where n(H2) and n(photons) denote the number of H2 molecules generated and the number of incident photons, respectively. The coefficient of 4 reflects the two-step excitation involved in this process. 2.6. STH Measurements. The STH values obtained from photocatalyst sheets during water splitting were assessed using the same experimental apparatus but with illumination from a solar simulator (Asahi Spectra Co., Ltd., HAL-320). The STH was calculated as

3. RESULTS AND DISCUSSION The XRD patterns of the synthesized powders are shown in Figure S1 in the Supporting Information. LTC-based samples and BiVO4 were obtained as the major phases reported in our previous work27,28 and the literature,30 respectively. The major diffraction peaks for LTCA and Ga-LTCA are similar to those for the LTC phase but are shifted toward lower angles because Ag+ has a larger ionic radius than Cu+. These observations match well with our previous work.28 Particulate photocatalyst

STH (%) = (R(H 2) × ΔG)/(P × S) × 100 1692

DOI: 10.1021/acscatal.7b03884 ACS Catal. 2018, 8, 1690−1696

Research Article

ACS Catalysis sheets were fabricated by a particle transfer method using these particles. A top-view SEM image (Figure 1a) demonstrates that LTC and BiVO4 particles covered the conductive layer, with only the occasional void between particles. A cross-sectional SEM image (Figure 1b) and the corresponding EDX mapping images (Figure 1c−f) show that the two types of particles were both distributed over and anchored by the Au layer. Direct and rigid contact of particles with the Au layer in the LTC/Au/ BiVO4 sheets is crucial for efficient charge transfer between the BiVO4 and LTC through the Au layer. The photocatalytic activities of the LTC/Au/BiVO4 sheets under visible light irradiation are plotted in Figure 2. The sheet

Table 1. Water Splitting Activities of Z-Scheme Systems Based on LTC, Ga-LTCA, and BiVO4 in Sheet or Powder Suspension Form with Different Redox Mediators photocatalyst redox mediator

entry

type

OEP

H2

O2

1

pure water

LTC

BiVO4

ndb

ndb

Fe2+/Fe3+ c

LTC

BiVO4

0.7d

ndb

I3−/I−e

LTC

BiVO4

0.9d

ndb

Fe2+/Fe3+ c

BiVO4

0.2d

ndb

BiVO4

0.1d

ndb

BiVO4

0.4d

0.1c

pure water

GaLTCA GaLTCA GaLTCA LTC

BiVO4

ndb

ndb

pure water

LTC

BiVO4

4.7

2.3

pure water

GaLTCA GaLTCA

BiVO4

22.0

11.0

10

powder suspensiona powder suspensiona powder suspensiona powder suspensiona HEP/Au/OEP sheetf HEP/Au/OEP sheetf HEP/OEP sheetf HEP/Au/OEP sheetf HEP/Au/OEP sheetf HEP/Au sheetf

d

ndb

11

OEP/Au sheetf

pure water

ndb

ndb

2 3 4 5 6 7 8 9

Fe2+/Fe3+ c I3−/I− e

pure water

HEP

activity (μmol h−1)

0.3 BiVO4

a

Reaction conditions: photocatalysts, 10 mg each; 150 mL distilled water; light source, 300 W Xe lamp equipped with a visible light filter (λ >420 nm); irradiation area, 48 cm2; cocatalyst, Rh/Cr2O3. bnd denotes not detected. cFeCl2 solution (2 mM) adjusted to pH 4 using H2SO4(aq). dGas evolution was observed only in the initial few hours and then stopped. eNaI solution (2.5 mM) without adjustment of pH. f Reaction conditions: 40 mL of distilled water; light source, 300 W Xe lamp equipped with a visible light filter (λ >420 nm); irradiation area, 8 cm2; cocatalyst, Rh/Cr2O3.

Figure 2. Time courses of overall water splitting over unmodified LTC/Au/BiVO4 sheets (triangles), sheets loaded with Rh (squares), and sheets loaded with Cr2O3/Rh (circles). Closed and open symbols represent H2 and O2, respectively. Reaction conditions: distilled water (pH 6.9, 40 mL); light source, 300 W Xe lamp equipped with a visible light filter (λ >420 nm); irradiation area, 8 cm2.

without a cocatalyst did not evolve H2 or O2 because of the lack of H2 evolution sites on the LTC. In fact, unmodified LTC is hardly active for H2 evolution, while bare BiVO4 is active for O2 evolution in the presence of sacrificial reagents.30,32 Sheets were therefore loaded with noble-metal cocatalysts followed by the addition of Cr2O3 (denoted as Cr2O3/M; M = Rh, Pt, Ru) by photodeposition. In the absence of Cr2O3 loading, the watersplitting activity was lower than that with Cr2O3/M cocatalysts (Figure S2 in the Supporting Information), because a Cr2O3 shell over noble metals could inhibit the formation of water from the H2 and O2 reaction products as well as limit the oxygen reduction reaction (ORR) on the noble-metal cocatalysts.31 This feature is similar to the results obtained with photocatalyst sheets in our earlier study.21 For all of the trials after the loading of the Cr2O3/M cocatalysts, the LTC/ Au/BiVO4 sheets evolved H2 and O2 at the expected stoichiometric molar ratio of 2 under visible light irradiation (Figure S3 in the Supporting Information). Cr2O3/Rh was found to be the most effective cocatalyst for enhancing the water splitting (Figure 2). In contrast, a powder suspension composed of LTC and BiVO4 did not show water-splitting activity comparable to that of the sheet system, even when aqueous redox mediators were employed (entries 1−3, Table 1). In addition, sheets without a conductive layer showed negligible photocatalytic activity (entry 7, Table 1) because an electron transfer path from BiVO4 to LTC was missing. Therefore, the electron transfer via the underlying Au layer is essential for the realization of Z-scheme water splitting. Moreover, according to X-ray photoelectron spectra shown in Figure 3, the peaks at 306.9 and 309.4 eV represent the 3d5/2

Figure 3. Rh 3d XPS spectra obtained from (a) Cr2O3/Rh/LTC/Au/ BiVO4, (b) Rh/LTC/Au, and (c) Rh/BiVO4/Au sheets. The signals at higher binding energies originate from Rh 3d3/2 orbitals but are too weak to fully deconvolute.

orbitals of metallic and trivalent Rh species. The trivalent Rh species can possibly be assigned to Rh2O3.33 It suggested that metallic and trivalent Rh species were respectively dominant on the LTC/Au and BiVO4/Au sheets, while both species were present on the LTC/Au/BiVO4 sheet. The Rh species present on the LTC would be expected to function as a hydrogen evolution cocatalyst, while the RhOx species on the BiVO4 1693

DOI: 10.1021/acscatal.7b03884 ACS Catal. 2018, 8, 1690−1696

Research Article

ACS Catalysis

(Figure 4), implying that it is more active and more resistant to self-oxidation during water splitting than conventional powder suspension systems. This is due in part to the absence of redox couples in the aqueous phase, which is indeed a key feature of the sheet configuration.21 Adding redox mediators such as Fe3+/Fe2+ and I3−/I− lowers the water-splitting activity of the sheets (entries 5 and 6, Table 1) as a result of the backward reactions involving the reduction and oxidation of the redox couples at the HEP and OEP, respectively.11,20 Interestingly, the original level of activity was not fully recovered even after the sheet was rinsed and used once more in pure distilled water. This effect was especially evident when an aqueous Fe3+/Fe2+ solution was used, suggesting degradation of the oxysulfide photocatalyst in aqueous solutions containing redox couples (Figure S7 in the Supporting Information). These observations highlight the challenges associated with backward reactions and the instability of (oxy)sulfide photocatalysts in Z-scheme watersplitting systems employing redox couples. A sheet configuration functioning in water without any redox mediators is therefore particularly advantageous when the (oxy)sulfide photocatalyst is prone to backward reactions and/or degradation in solutions containing redox couples. However, it should also be pointed out that the gas production rate over Cr2O3/ Rh/Ga-LTCA/Au/BiVO4 became lower over time when the system was not evacuated periodically (Figure S8 in the Supporting Information) because of an increase in the partial pressures of H2 and O2 in the reaction system. It is suspected that backward reactions still occurred on the regions of the Au conductor layer directly exposed to the reaction solution. Such an undesirable contribution of the underlying conductor layer can be avoided by utilizing inert conductors such as carbon.21 Figure 5 plots the AQY for the water-splitting reaction obtained from a Cr2O3/Rh/Ga-LTCA/Au/BiVO4 sheet as a

evidently did not serve as an oxygen evolution cocatalyst, judging from the results of the sacrificial oxygen evolution reaction (Figure S4 in the Supporting Information). These results indicate that, in the present Z-scheme photocatalyst sheet system, electron transfer occurs from the conduction band of BiVO4 to the valence band of oxysulfides via the Au layer in distilled water: in other words, without any redox mediator. No redox reaction occurs during the charge transfer through the Au layer. Notably, the activity of the sheets was greatly enhanced following p-type doping with lower valence cations such as Ga3+, Al3+, Sc3+, and Mg2+ at the Ti sites of the LTC and forming LTCA solid solutions (Figure 4 and Tables S1 and S2

Figure 4. A Cr2O3/Rh-loaded Ga-LTCA/Au/BiVO4 sheet under visible light. Closed and open symbols represent H2 and O2, respectively. Reaction conditions: distilled water (pH 6.9, 40 mL); light source, 300 W xenon lamp equipped with a visible light filter (λ >420 nm); irradiation area, 8 cm2.

in the Supporting Information). The performance tendency of Cr2O3/M cocatalysts over Ga-LTCA was similar to that over undoped LTC (Figure S5 in the Supporting Information). Loading of Rh as the core of the Cr2O3/M cocatalyst was the most effective. A Cr2O3/Rh/Ga-LTCA/Au/BiVO4 sheet evolved H2 and O2 at the anticipated H2/O2 ratio of 2. In our previous works, it has been shown that p-type doping boosts the photocathodic current of the LTC(A) system, because the internal electric field in the depletion layer needed for charge separation is enhanced and the resistivity for hole conduction is reduced.26−29 Conversely, p-type doping lowers the photocatalytic H2 evolution activity of an LTC powder suspension in an aqueous Na2S and Na2SO3 solution.27 The present study also found that the use of LTA in place of LTC lowered the water splitting activity of the sheet (Figure S6 in the Supporting Information). LTA exhibits twice the H2 evolution rate of LTC in the form of a powder suspension but functions less efficiently as a photocathode.28,32 These observations indicate that the activity of the present sheet system was strongly correlated with the activity of the HEP as a photocathode rather than as a powder suspension. Therefore, a Z-scheme particulate photocatalyst sheet can be regarded as the integration of a number of miniaturized, parallel p/n PEC cells. However, this system functions efficiently in the absence of supporting electrolytes because H2 and O2 are evolved in close proximity to one another such that the solution resistance is minimized. Furthermore, the Cr2O3/Rh/Ga-LTCA/Au/BiVO4 system maintained its activity during prolonged light irradiation

Figure 5. AQY of a Cr2O3/Rh/Ga-LTCA/Au/BiVO4 sheet as a function of the incident light wavelength. Reaction conditions: distilled water (pH 6.9, 40 mL); light source, 300 W Xe lamp equipped with various band-pass filters; temperature, 288 K. The AQY increases to 4.9% at 420 nm (red dot) when the reaction temperature is raised to 313 K.

function of the irradiation wavelength, along with the diffuse reflectance spectra of Ga-LTCA and BiVO4. An AQY of 3.2% was obtained at 420 nm. This value is remarkably high in comparison to those normally associated with photocatalytic water-splitting systems involving a material with an absorption edge extended up to 710 nm.6 The AQY value decreased with 1694

DOI: 10.1021/acscatal.7b03884 ACS Catal. 2018, 8, 1690−1696

Research Article

ACS Catalysis

and an STH of 0.11% were obtained when using Ga-LTCA as the HEP, and this STH value represents one of the highest yet reported for a water-splitting system based on a particulate photocatalyst that harvests light up to ca. 700 nm. This work not only provides additional insights into the factors governing the water-splitting activity of photocatalyst sheets but also expands the range of possibilities for the development of sheets capable of functioning under irradiation by long-wavelength photons. We believe that this visible-light-driven water splitting system could be further enhanced through replacing BiVO4 with an OEP having a narrower band gap so as to fully take advantage of the ability of the Ga-LTCA to remain active up to 710 nm.

increases in the incident light wavelength, and the longest wavelength that could be reasonably utilized for water splitting was 520 nm. This wavelength is consistent with the absorption edge wavelength of BiVO4. At present, the photon wavelength available for Z-scheme water splitting is therefore limited by the use of BiVO4. At wavelengths shorter than 520 nm, it is considered that the activity of the Ga-LTCA/Au/BiVO4 sheet is capped by Ga-LTCA showing lower hydrogen evolution activity in comparison to SrTiO3:La,Rh from the comparison of the AQY values.34 However, these results do demonstrate that a sheet based on Ga-LTCA as the HEP has the potential to harvest longer-wavelength visible light when it is used with an OEP having a longer-wavelength response. Figure 6 summarizes the water-splitting activity of a Cr2O3/ Rh/Ga-LTCA/Au/BiVO4 sheet at different temperatures under



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscatal.7b03884. Sample preparation procedures, XRD patterns, and photocatalytic activity (PDF) Video showing bubbles evolving on the photocatalyst sheet during illumination (AVI)



AUTHOR INFORMATION

Corresponding Author

*E-mail for K.D.: [email protected]. ORCID

Takashi Hisatomi: 0000-0002-5009-2383 Tsutomu Minegishi: 0000-0001-5043-7444 Kazunari Domen: 0000-0001-7995-4832

Figure 6. Temperature (T) dependence of the water-splitting activity and STH over a Cr2O3/Rh/Ga-LTCA/Au/BiVO4 sheet at a background pressure of 5 kPa under simulated sunlight (AM 1.5G).

Notes

The authors declare no competing financial interest.



irradiation with a solar simulator (AM 1.5 G). The reaction was enhanced monotonically with increasing temperature from 275 to 313 K at a background pressure of 5 kPa. The apparent activation energy of the overall water-splitting reaction was estimated to be 14 kJ mol−1. The STH of the water splitting reaction was 0.11% at 313 K, and the AQY was increased to 4.9% at 420 nm under these conditions, as highlighted in Figure 5. The vigorous evolution of bubbles of H2 and O2 was observed over the photocatalyst sheet during the reaction (see movie 1 in the Supporting Information). The STH value obtained in these trials is 2 orders of magnitude greater than that observed in our previous work with oxysulfide-based Zscheme systems using an aqueous I3−/I− redox mediator. In fact, this value is the highest yet reported for a Z-scheme system involving a particulate photocatalyst absorbing light up to ca. 700 nm.6

ACKNOWLEDGMENTS This work was financially supported by the Artificial Photosynthesis Project of the New Energy and Industrial Technology Development Organization (NEDO) and by Grants-in-Aid for Scientific Research (A) (No.16H02417) and for Young Scientists (A) (No. 15H05494) from the Japan Society for the Promotion of Science (JSPS). A part of this work was conducted at the Advanced Characterization Nanotechnology Platform of the University of Tokyo, supported by the “Nanotechnology Platform” of the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan. S.S. acknowledges the support of the China Scholarship Council (CSC) and the National Natural Science Foundation of China (No.U1632273). The authors also wish to thank Prof. Shibata and Ms. Mamiko Nakabayashi of The University of Tokyo for their assistance in acquiring SEM-EDX data.

4. CONCLUSIONS Water splitting in the absence of a redox couple was achieved by applying LTC as the HEP and BiVO4 as the OEP to construct Z-scheme-type sheets. The successive photodeposition of Rh and Cr2O3 onto these particulate sheets was found to be essential for effective gas evolution. These sheets permit the use of redox-mediator-free water and so mitigate the problem of backward reactions involving redox mediators and severe corrosion and thereby enable steady Z-scheme water splitting using oxysulfides. The activity of the sheets was strongly correlated to the activity of the HEP as a photocathode rather than as a powder suspension. An AQY of 4.9% at 420 nm



REFERENCES

(1) Lewis, N. S. Science 2016, 351, aad1920. (2) Chen, X.; Liu, L.; Yu, P. Y.; Mao, S. S. Science 2011, 331, 746− 750. (3) Walter, M. G.; Warren, E. L.; Mckone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.; Lewis, N. S. Chem. Rev. 2010, 110, 6446−6473. (4) Hisatomi, T.; Kubota, J.; Domen, K. Chem. Soc. Rev. 2014, 43, 7520−7535. (5) Chen, S.; Takata, T.; Domen, K. Nature Rev. Mater. 2017, 2, 17050. (6) Kudo, A.; Miseki, Y. Chem. Soc. Rev. 2009, 38, 253−278.

1695

DOI: 10.1021/acscatal.7b03884 ACS Catal. 2018, 8, 1690−1696

Research Article

ACS Catalysis (7) Chen, X.; Chen, S.; Guo, L.; Mao, S. Chem. Rev. 2010, 110, 6503−6570. (8) Osterloh, F. Chem. Soc. Rev. 2013, 42, 2294−2320. (9) Zhou, P.; Yu, J.; Jaroniec, M. Adv. Mater. 2014, 26, 4920−4935. (10) Abe, R. J. Photochem. Photobiol., C 2010, 11, 179−209. (11) Maeda, K. ACS Catal. 2013, 3, 1486−1503. (12) Tachibana, Y.; Vayssieres, L.; Durrant, J. R. Nat. Photonics 2012, 6, 511−518. (13) Li, H.; Tu, W.; Zhou, Y.; Zou, Z. Adv. Sci. 2016, 3, 1500389. (14) Che, W.; Cheng, W.; Yao, T.; Tang, F.; Liu, W.; Su, H.; Huang, Y.; Liu, Q.; Liu, J.; Hu, F.; Pan, Z.; Sun, Z.; Wei, S. J. Am. Chem. Soc. 2017, 139, 3021−3026. (15) Tada, H.; Mitsui, T.; Kiyonaga, T.; Akita, T.; Tanaka, K. Nat. Mater. 2006, 5, 782−786. (16) Iwashina, K.; Iwase, A.; Ng, Y. H.; Amal, R.; Kudo, A. J. Am. Chem. Soc. 2015, 137, 604−607. (17) Iwase, A.; Yoshino, S.; Takayama, T.; Ng, Y. H.; Amal, R.; Kudo, A. J. Am. Chem. Soc. 2016, 138, 10260−10264. (18) Kato, T.; Hakari, Y.; Ikeda, S.; Jia, Q.; Iwase, A.; Kudo, A. J. Phys. Chem. Lett. 2015, 6, 1042−1047. (19) Wang, W.; Chen, J.; Li, C.; Tian, W. Nat. Commun. 2016, 5, 4647. (20) Ma, G.; Chen, S.; Kuang, Y.; Akiyama, S.; Hisatomi, T.; Nakabayashi, M.; Shibata, N.; Katayama, M.; Minegishi, T.; Domen, K. J. Phys. Chem. Lett. 2016, 7, 3892−3896. (21) Wang, Q.; Hisatomi, T.; Jia, Q.; Tokudome, H.; Zhong, M.; Wang, C.; Pan, Z.; Takata, T.; Nakabayashi, M.; Shibata, N.; Li, Y.; Sharp, I. D.; Kudo, A.; Yamada, T.; Domen, K. Nat. Mater. 2016, 15, 611−615. (22) Wang, Q.; Hisatomi, T.; Suzuki, Y.; Pan, Z.; Seo, J.; Katayama, M.; Minegishi, T.; Nishiyama, H.; Takata, T.; Seki, K.; Kudo, A.; Yamada, T.; Domen, K. J. Am. Chem. Soc. 2017, 139, 1675−1683. (23) Minegishi, T.; Nishimura, N.; Kubota, J.; Domen, K. Chem. Sci. 2013, 4, 1120−1124. (24) Pan, Z.; Hisatomi, T.; Wang, Q.; Chen, S.; Iwase, A.; Nakabayashi, M.; Shibata, N.; Takata, T.; Katayama, M.; Minegishi, T.; Kudo, A.; Domen, K. Adv. Funct. Mater. 2016, 26, 7011−7019. (25) Kobayashi, R.; Yakashima, T.; Tanigawa, S.; Takeuchi, S.; Ohtani, B.; Irie, H. Phys. Chem. Chem. Phys. 2016, 18, 27754−27760. (26) Liu, J.; Hisatomi, T.; Katayama, M.; Minegishi, T.; Kubota, J.; Domen, K. J. Mater. Chem. A 2016, 4, 4848−4854. (27) Liu, J.; Hisatomi, T.; Ma, G.; Iwanaga, A.; Minegishi, T.; Moriya, Y.; Katayama, M.; Kubota, J.; Domen, K. Energy Environ. Sci. 2014, 7, 2239−2242. (28) Hisatomi, T.; Okamura, S.; Liu, J.; Shinohara, Y.; Ueda, K.; Higashi, T.; Katayama, M.; Minegishi, T.; Domen, K. Energy Environ. Sci. 2015, 8, 3354−3362. (29) Higashi, T.; Shinohara, Y.; Ohnishi, A.; Liu, J.; Ueda, K.; Okamura, S.; Hisatomi, T.; Katayama, M.; Nishiyama, H.; Yamada, T.; Minegishi, T.; Domen, K. ChemPhotoChem. 2017, 1, 167−172. (30) Kudo, A.; Omori, K.; Kato, H. J. Am. Chem. Soc. 1999, 121, 11459−11467. (31) Maeda, K.; Teramura, K.; Lu, D.; Saito, N.; Inoue, Y.; Domen, K. Angew. Chem., Int. Ed. 2006, 45, 7806−7809. (32) Suzuki, T.; Hisatomi, T.; Teramura, K.; Shimodaira, Y.; Kobayashi, H.; Domen, K. Phys. Chem. Chem. Phys. 2012, 14, 15475−15481. (33) Vallet, C. E.; Choudhury, A.; Sobol, P. E.; White, C. W. Electrochim. Acta 1993, 38, 1313−1320. (34) Wang, Q.; Li, Y.; Hisatomi, T.; Nakabayashi, M.; Shibata, N.; Kubota, J.; Domen, K. J. Catal. 2015, 328, 308−315.

1696

DOI: 10.1021/acscatal.7b03884 ACS Catal. 2018, 8, 1690−1696