Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2021 Mar 4;125(10):5692–5699. doi: 10.1021/acs.jpcc.0c11607

Ultrafast Charge Carrier Dynamics in CuWO4 Photoanodes

Ivan Grigioni †, Annalisa Polo †, Maria Vittoria Dozzi †, Lucia Ganzer ‡, Benedetto Bozzini §, Giulio Cerullo ‡, Elena Selli †,*
PMCID: PMC8765008  PMID: 35069964

Abstract

graphic file with name jp0c11607_0008.jpg

CuWO4 is a ternary metal oxide semiconductor with promising properties for photoelectrochemical (PEC) water splitting and solar light conversion, due to its quite low band gap (2.3 eV) and high stability in an alkaline environment. Aiming at understanding the origin of the relatively low PEC efficiency attained with CuWO4 photoanodes, we here investigate transparent CuWO4 electrodes prepared by a simple solution-based method through the combination of femtosecond transient absorption spectroscopy with electrochemical, PEC, and photochromic characterizations. The very fast recombination dynamics of the charge carriers photogenerated in CuWO4, which is the reason for its low efficiency, is discussed in relation with its PEC performance and with the recently calculated band structure of this material, also in comparison with the behavior of other semiconductor oxides employed in PEC applications, in particular Fe2O3.

1. Introduction

Solar energy conversion and storage into fuels such as H2 through water splitting form a viable way to achieve a future based on renewable energy sources.1−3 The photocatalytic approach is most promising among those allowing one to transform intermittent and abundant solar light into valuable energy vectors. In this regard, the photocatalytic water splitting technology and, in particular, photoelectrochemical (PEC) tandem cells in which two light harvesting materials with band gaps matching two different portions of the visible spectrum are coupled would in principle allow a solar light energy conversion efficiency up to 31%.4−6 For PEC water oxidation purposes, visible light active, stable, and earth abundant materials are highly desirable.7 Metal oxide semiconductors are very attractive for this application, being intrinsically more stable toward oxidation than other visible light harvesting materials.8

TiO2, WO3, Fe2O3, and BiVO4 are the most studied photoactive materials. All of them present advantages in many respects but also drawbacks.6,8 In particular, Fe2O3 has a band gap of 2.1 eV, which is low enough to allow a solar light conversion efficiency higher than 12%,9 but it is characterized by a very fast charge carrier recombination, which seriously limits its performance. BiVO4, on the other hand, has longer-lived charge carriers, compatible with an internal quantum efficiency (IQE) close to unity up to 2.75 eV (450 nm), but with its band gap of 2.4 eV, it would theoretically allow a sunlight to hydrogen conversion efficiency of ca. 8% only.10−12

Copper tungstate (CuWO4), with a band gap of ca. 2.3 eV corresponding to an absorption edge of 550 nm, represents a good candidate material for PEC devices.13 It shows good stability in prolonged PEC water oxidation experiments under simulated solar irradiation in neutral and slightly basic solutions.13 However, although the exploitation of CuWO4 as photoanode material dates back to pioneering works of the 1980s,14,15 and despite the fact that it features quantitative selectivity to water oxidation and hole collection at the electrode/electrolyte interface,13,16,17 the highest photocurrent obtained with CuWO4-based electrodes is only 0.2–0.3 mA cm–2 at 1.23 VRHE,18,19 as compared to, e.g., 6.7 and 6 mA cm–2 obtained with state of the art BiVO4 and Fe2O3 photoanodes.20,21 The reasons for such a poor PEC performance are unclear, also because the dynamics of the charge carriers photogenerated in CuWO4 is still unexplored.

Here we investigate the ultrafast dynamics of the charge carriers photogenerated in transparent CuWO4 electrodes through femtosecond transient absorption (TA) spectroscopy and discuss it in relation to the recently proposed electronic structure of CuWO4, which receives full support from the here performed PEC characterization of this ternary oxide.

2. Methods

2.1. Materials

The following chemicals, all purchased from Sigma-Aldrich, were employed in the present work: copper(II) nitrate trihydrate 99%, citric acid 99%, tetrabutylammonium hexafluorophosphate 99%, nitric acid 65%, anhydrous sodium sulfite, and boric acid ACS reagent, together with ammonium tungsten oxide hydrate 99% (Fluka). Deionized Millipore water was employed to prepare the solutions.

2.2. Photoelectrode Preparation

A 0.5 M solution of CuWO4 was prepared as follows. First 14 mmol of citric acid were dissolved in 5.3 mL of ethanol and 2.4 mL of deionized H2O. Then 5 mmol of Cu(NO3)2·3H2O and 0.4167 mmol of (NH4)6H2W12O40·xH2O were sequentially added, under stirring. A film was grown from this solution onto fluorine-doped tin oxide (FTO) glass (Pilkington Glass, TEC-7, thickness 2 mm) by spin coating at 2000 rpm for 20 s. Prior to deposition, the FTO glass was cleaned by 30 min long sonication in a soap solution, then thoroughly rinsed with water, sonicated for 30 min in ethanol, and finally dried in air. Before moving the FTO slice to the spin coater, it was soaked in 2-propanol for a few seconds. This procedure was found to increase the optical transparency of the final film. After spin coating, the CuWO4 film was preannealed at 250 °C for 10 min and then annealed at 550 °C for 1 h.

2.3. Optical, Morphological, and Structural Characterization

The absorption spectra were recorded in the transmission mode with a Jasco V-650 spectrophotometer. The spectrum shown in Figure 1A refers to a double layered CuWO4 film electrode, to minimize the presence of interference fringes. Top view and cross-section scanning electron microscopy (SEM) images were acquired by employing a LEO 1430 scanning electron microscope operating at a 10 kV accelerating voltage and 8 mm working distance. The crystalline phase of the material was determined through X-ray diffraction (XRD) analysis using a Philips PW 1830/40 instrument, equipped with a Cu sealed tube at 40 kV and 40 mA.

Figure 1.

Figure 1

Open in a new tab

(A) Absorption spectrum of a CuWO4 photoelectrode. (B) X-ray diffraction pattern of the CuWO4 thin film coated on FTO glass. The gray bars are the reference Bragg reflections of CuWO4 from JCPDF 72-0616. The FTO reflections are marked separately.

2.4. Photoelectrochemical Measurements

PEC measurements were performed in a homemade three electrode cell equipped with two quartz windows. The FTO/CuWO4 electrode was used as working electrode, a Pt wire as counter electrode, and a Ag/AgCl electrode as reference electrode. The photoanodes were tested under back (from the FTO side) irradiation. The light source was an Oriel, Model 81172 solar simulator equipped with a AM 1.5 G filter. The light intensity, measured by means of a Thorlabs PM200 power meter equipped with a S130VC power head with Si detector, was 100 mW cm–2. PEC measurements were carried out with the electrode in contact with a 0.1 M buffer borate (KBi) solution at pH 9 or with 0.1 M pH 9 KBi containing 0.5 M Na2SO3.

The KBi solution was prepared by dropwise adding a 1.0 M KOH solution to aqueous boric acid up to pH 9, followed by dilution to 0.1 M. Prior to performing the linear scans, the open circuit potential (OCP) of the electrode was measured for 10 min under irradiation. Then the scan started from the so determined OCP with a 10 mV s–1 sweep rate. The potential vs Ag/AgCl was converted to the RHE scale at pH 0 (corresponding to NHE) using the following equation: ERHE = EAg/AgCl + 0.059 pH + E°Ag/AgCl, with E°Ag/AgCl(3.0 M NaCl) = 0.210 V at 25 °C.

Incident photon to current efficiency (IPCE) scans were recorded under a 1.23 V vs RHE bias, and the current was measured with a 10 nm step, within the 300–550 nm wavelength range. A 300 W Lot-Oriel Xe lamp was employed as an irradiation source, equipped with a Lot-Oriel Omni-λ 150 monochromator and a Thorlabs SC10 automatic shutter. The incident light power was measured at each irradiation wavelength, and the IPCE was calculated as

2.4. 1

where J (A cm–2) is the photocurrent density and Pλ (W cm–2) is the power of the monochromatic light at the wavelength λ (nm). The internal quantum efficiency (IQE) was calculated by combining the IPCE curve with the absorption (A) spectrum of the photoanode, as follows:

2.4. 2

2.5. Photochromic Measurements

The photochromic experiments were performed by irradiating a 25 × 9 mm2 CuWO4 electrode, sealed inside a N2 filled glovebox in a 10 mm-thick cuvette modified to allow the insertion of the 9 mm wide electrode and containing 3 mL of ethanol previously stored overnight in the glovebox, in order to attain a completely oxygen free environment. The film was irradiated with a 300 W Xe arc lamp (Lot Oriel, LSH302), with a light intensity of ca. 150 mW cm–2, measured with a PM100A Thorlabs power meter equipped with a S302C thermal power sensor. The photochromic effect was monitored by recording the absorption spectra at fixed irradiation times with a Shimadzu model 3600 UV–vis–IR spectrophotometer. The irradiated cuvette was cooled with a flux of compressed air.

At the end of the experiment, the CuWO4 electrode in the airtight cuvette was moved to the spectrophotometer, the cuvette was opened to air, and the subsequent absorption spectrum changes were monitored. Almost complete CuWO4 back oxidation was attained after 11 h long contact with air.

2.6. Electrochemical Measurements

The Mott–Schottky plot was obtained by acquiring a series of potentiostatic electrochemical impedance spectroscopy (EIS) spectra employing a Parstat potentiostat, with a sinusoid of peak-to-peak amplitude of 10 mV in the 105–0.1 Hz frequency range. The electrolyte employed was a 0.3 M Na2SO4 solution at pH 7. The bias ranged between −0.2 and +0.6 V vs Ag/AgCl (VAg/AgCl), and the potentials were scanned in the cathodic direction, to avoid possible irreversibility due to reduction of the metallic components. Capacitance values were extracted by a nonlinear least-squares fitting with an equivalent-circuit model featuring the parallel of a faradaic resistance and a double-layer capacitance, in series with an ohmic resistance (Randle’s model).16 As is customary in EIS modeling, morphological heterogeneities of the film were accounted for by replacing the double-layer capacitance with a Constant Phase Element (CPE). Fitting was carried out with a Matlab-based program written by the authors, allowing better control over the regression statistics than the commonly used commercial packages. The CPE exponent estimates ranged in the 0.86–0.98 interval and did not exhibit any measurable potential dependence. The extrapolation to zero of the linear part of the Mott–Schottky plot22 and the subsequent correction of the intercept for kT/q ≈ 25 mV at room temperature, provided a value of 0.51 VRHE.

Cyclovoltammetry experiments were performed by polarizing a 25 × 9 mm2 electrode within the same cuvette employed in the photochromic experiments. The ohmic contact was obtained by soldering with indium a copper wire to the uncovered FTO edge; the contact was then glued with an epoxy resin (Araldite) for insulation. Two platinum wires were used as counter and quasi-reference electrodes, respectively, and the cuvette was closed with a rubber septum. The potential was modulated with an Autolab PGSTAT 12 instrument (EcoChemie, Utrecht, The Netherlands) controlled by the NOVA software. The potential of the Pt pseudoreference electrode was calibrated with respect to RHE using the ferrocene/ferrocenium redox couple (+400 mVRHE) as standard.23 The electrolyte was a 0.1 M tetrabutylammonium hexafluorophosphate solution in acetonitrile, which was deaerated through N2 bubbling for 5 min prior to the electrochemical experiments.

2.7. Transient Absorption Characterization

Femtosecond transient absorption (TA) experiments were performed using an amplified Ti:sapphire laser system (Libra, Coherent) delivering 4 mJ, 100 fs pulses at wavelength λ = 800 nm with 1 kHz repetition rate. A 400-μJ fraction of the laser energy was used for the experiment, 95% of which was frequency doubled to λ = 400 nm and used as a pump, while the remaining 5% was focused on a CaF2 crystal to generate the white light continuum probe, covering the 380–680 nm wavelength range. Alternatively, the output of an optical parametric amplifier was used to generate the pump pulses at λ = 500 nm. TA spectra were measured with an Acton Sp2150 spectrometer (Princeton Instruments) equipped with a CCD camera (Entwicklungsbro Stresing) detecting data at the full 1 kHz repetition rate of the laser. The photoexcited spot area was 6 × 10–4 cm2 with a pump energy ranging from 16 to 500 nJ, corresponding to a fluence from ca. 27 to 830 μJ cm–2.

3. Results and Discussion

3.1. Characterization of the CuWO4 Photoanodes

As shown in Figure 1A, the CuWO4 photoanodes prepared by spin coating a precursor solution on FTO, followed by annealing, are optically transparent at wavelengths longer than the absorption edge of the material. The absorbance drops above 350 nm, due to the indirect band gap and the low absorption coefficient of CuWO4;22 the broad and weak feature above 700 nm is the tail of the Cu d–d transition.24

The X-ray diffraction pattern of a polycrystalline CuWO4 film, shown in Figure 1B, coincides with the set of Bragg reflections reported in reference JCPDF 72-0616, without reflections typical of WO3, CuO, and Cu2O, indicating that CuWO4 is in pure triclinic form.15,16,22

The top view SEM image presented in Figure 2A confirms that the CuWO4 layer completely covers the conductive FTO substrate and is composed of small grains with ca. 50 nm average diameter. The CuWO4 film in the here investigated electrodes is ca. 80 nm thick, as estimated from the cross section SEM image shown in Figure 2B.

Figure 2.

Figure 2

Open in a new tab

(A) SEM top view and (B) cross section image of the CuWO4/FTO electrode employed in the PEC measurements. The scale bars are 500 (A) and 200 nm (B).

Figure 3A reports the linear sweep voltammetry recorded with the CuWO4 electrode in contact with a 0.1 M potassium borate aqueous solution buffered at pH 9, both in the presence and in the absence of 0.5 M Na2SO3 as hole scavenger. The photocurrent values obtained at 1.23 VRHE in the presence or absence of sulfite, i.e., 0.40 and 0.15 mA cm–2, respectively, are compatible with those recorded with electrodes prepared through other routes.16,19,22

Figure 3.

Figure 3

Open in a new tab

(A) Linear sweep voltammetry under AM 1.5 G simulated solar light for the CuWO4 film in contact with 0.1 M potassium borate (KBi) aqueous solutions at pH 9, (1) in the presence of 0.5 M Na2SO3 (dash dot line) and (2) in the absence of Na2SO3 (continuous line) and (3) in the absence of irradiation and of Na2SO3. (B) Incident photon to current efficiency (IPCE, black circles) and internal quantum efficiency (IQE, red squares) of the CuWO4 photoanode at 1.23 VRHE in KBi at pH 9.

As expected, in the presence of an electron donor the photocurrent onset potential shifts to more negative values.25 However, only a ca. 3-fold photocurrent density increase was observed at 1.23 VRHE upon addition of Na2SO3 as a hole scavenger, which is 2 orders of magnitude smaller than that measured, for instance, with BiVO4 at 1.23 VRHE in the presence of the same hole scavenger.26 This supports the view that surface water oxidation is almost quantitative on CuWO4, as recently outlined by Gao et al.,27 and it implies that the overall efficiency of this material is limited by poor internal charge transport or fast bulk charge carrier recombination, hampering efficient hole diffusion from the excitation site to the oxide surface.

The incident photon to current efficiency (IPCE) curve shown in Figure 3B reflects the absorption spectrum of CuWO4 (see Figure 1A), decreasing with increasing wavelength. The IQE curve, obtained by normalizing the photocurrent for the intensity of absorbed light, has a similar wavelength dependence. The photocurrent onset is at ca. 510 nm, in line with that reported for bulk CuWO416,28 and slightly red-shifted with respect to that of WO3 (absorption onset at ca. 480 nm).29

A cyclovoltammetry (CV) test was employed to investigate and localize redox states within the band gap of CuWO4. As shown in Figure 4A, a Butler–Volmer type current rise (marked with a blue arrow), starting at ca. 2.0 VRHE, indicates the oxidation of the highest occupied states of the valence band (VB) of CuWO4. At the lowest investigated potential range a quasi-reversible redox process centered at ca. 0 VRHE (cathodic and anodic peaks marked with gray arrows) is found, which might be related to the reduction and oxidation of the narrow band composed of localized Cu 3d states, as first proposed by Pyper et al.30 and confirmed by recent calculations.31

Figure 4.

Figure 4

Open in a new tab

(A) Cyclovoltammetry recorded in 0.1 M tetrabutylammonium hexafluorophosphate, with a scan rate of 100 mV s–1. (B) Mott–Schottky plot of the CuWO4 photoanode in 0.3 M Na2SO4 solution at pH 7.

The voltammetric features found in the potential range between the VB oxidation threshold (ca. 2.0–2.4 VRHE) and the Cu 3d states (ca. 0 VRHE), consisting of two relatively weak oxidation peaks (marked with green arrows in Figure 4A) and a broad cathodic feature (marked with a red arrow), can thus be interpreted as the redox response of intragap states, possibly located in the 0.5–1 VRHE range.

The Mott–Schottky plot recorded with the CuWO4 photoanode in 0.3 M Na2SO4 solution at pH 7 (Figure 4B) presents a change in slope that suggests the occurrence of Fermi level pinning.30 This can be related to the presence of midgap states, energetically located at 0.51 VRHE minimum potential, as calculated from the intercept of the Mott–Schottky plot on the abscissa axis (see Section 2.3), and perfectly in line with CV results.

The photochromic behavior of CuWO4 was investigated by irradiating a film in deaerated ethanol, followed by monitoring the absorption spectrum changes by spectrophotometric analysis. Figure 5 shows that a positive absorption variation ΔA was observed, extending over the visible spectrum with a shoulder at ca. 600 nm, which linearly increased under irradiation.

Figure 5.

Figure 5

Open in a new tab

Photochromic measurements of a CuWO4 electrode recorded in deaerated ethanol at increasing time under irradiation with a 300 W Xe arc lamp.

This signal can be assigned to electrons trapped in CuWO4, in line with the ΔA signal decrease with time observed when, after 2-h-long irradiation, the film was exposed to air (see Figure S1 in the Supporting Information). Thus, evidence is obtained that electron trapping in CuWO4 is reversible and that metastable trapped states can be back-oxidized by atmospheric oxygen.

3.2. Femtosecond Transient Absorption Spectroscopy

The ultrafast charge carrier dynamics in CuWO4 was here investigated through femtosecond TA spectroscopy. Previously, only a time-resolved microwave conductivity investigation was reported on the long-lived charge carriers in CuWO4.32 Upon excitation at 400 nm, promoting electrons from the VB of CuWO4 into the localized Cu 3d-based band, the film exhibits a broad transient absorption increase (ΔA > 0) extending over the entire visible spectral range (Figure 6A), similar to the ΔA feature recorded in photochromic measurements (Figure 5), which is thus ascribed to trapped electrons. The results are similar to those obtained with Fe2O3 oxide in the same time scale.33,34 In the case of CuWO4, the transient spectrum does not present any photobleaching (ΔA < 0).

Figure 6.

Figure 6

Open in a new tab

(A) Femtosecond TA spectra recorded at different times after photoexcitation of CuWO4 at 400 nm under vacuum and ΔA decay traces (B) at 500 nm upon excitation with different pump fluences and (C) at different wavelengths upon excitation with a 420 μJ cm–2 fluence. (D) Evolution associated spectra (EAS) obtained from Global Analysis of the ΔA(λ,Δt) data recorded upon photoexcitation at 400 nm with a 170 μJ cm–2 fluence.

In order to investigate how transient decay processes depend on the density of photogenerated charges, TA experiments were performed at different excitation fluences in the 27 to 830 μJ cm–2 range. The ΔA values monitored at different wavelengths increase linearly with the pump fluence (Figure S2), indicating that multiphoton absorption of the pump and nonlinear carrier interactions are not significantly contributing to the observed signal.

The decays of ΔA signal obtained with different pump fluences and monitored at 500 nm are reported in Figure 6B. No significant change in the dynamics can be appreciated upon varying the pump fluence, within the investigated fluence range. This indicates that higher order kinetics, like a carrier–carrier annihilation (Auger recombination) decay mechanism, does not play any relevant role.

The TA signal in CuWO4 decays much faster than in other semiconductor oxides, with a behavior similar to that observed in Fe2O3.33 The normalized ΔA time traces at different probe wavelengths, shown in Figure 6C, indicate that the signal halves in ca. 1 ps, with only a slight dependence on the detection wavelength. The ΔA decay becomes slightly faster by moving the monitoring wavelength from 425 to 600 nm, without any appreciable difference at wavelengths above 600 nm. The lack of a strong dependence of the dynamics on the probe wavelength suggests that the carriers are trapped in the Cu 3d band or in lower energy midgap states within the pump laser pulse, in a way similar to that reported in pioneering ultrafast TA studies on hematite.33

In order to shed light on the overall evolution of the ΔA signal as a function of wavelength and delay, and to better visualize individual spectral contributions over the entire probe wavelength range, Global Analysis35,36 was carried out on the data set acquired with a representative pump fluence of 170 μJ cm–2. The time-dependent TA spectra reported in form of evolution-associated spectra (EAS) in Figure 6D consist of three kinetic components with different time constants, which span over the entire investigated 400–700 nm wavelength region. The analysis of their relative amplitudes indicates that the fastest component, with a time constant τ1 = 780 ± 10 fs, dominates over the entire spectral range. The rest of the ΔA signal decays with a time constant τ2 = 9.4 ± 0.1 ps, with only a small contribution from the longest time constant τ3 = 1.18 ± 0.02 ns.

In agreement with the single decay traces reported in Figure 6C, the results of Global Analysis are in line with a slight dependence of the signal decay on the monitoring wavelength, with both EAS relative to the first two time constants τ1 and τ2 exhibiting a maximum around 500 nm and that associated with the slowest component (τ3) being slightly blue-shifted and peaking at 480 nm.

A substantially unchanged photoexcitation scenario is observed when pumping closer to the bandgap energy of the material, at 500 nm (2.48 eV). Despite the lower TA signal due to the reduced absorption cross section, the obtained dynamics, reported in Figure S3, is very similar to the one measured upon excitation at 400 nm. Global analysis of the TA data retrieves three components with time constants τ1 = 785 ± 10 fs, τ2 = 9.3 ± 0.1 ps, and τ3 = 1.97 ± 0.02 ns, very close to those obtained upon pumping at 400 nm.

Our interpretation of TA data well reflects the CuWO4 band structure recently calculated by Thang et al.,31 by treating the material as a Mott–Hubbard magnetic insulator, a feature common to both CuWO4 and Fe2O3 oxides. According to this model, shown in Scheme 1, the VB of CuWO4, consisting of O 2p states partly mixed with Cu 3d states, is separated by ca. 2.1 eV from a narrow band consisting of very localized, Cu empty 3d levels, while a wider band dominated by W 5d and Cu 3d states is located at higher energy, corresponding to a gap of about 5 eV from the top of the VB. Similarly, the recently calculated electronic structure of hematite consists of two energetically well separated conduction bands, both primarily of Fe 3d character, accounting for the detected positive TA signal in the ultrafast regime.34

Scheme 1. Ultrafast Relaxation and Recombination Pathways of Electrons Photopromoted in CuWO4 upon 400 nm Excitation.

Scheme 1

Open in a new tab

In this picture, the optical absorption in CuWO4 associated with the 2.3 eV experimental band gap implies the transition from a not fully localized mixed O 2p–Cu 3d state within the VB to localized Cu 3d states. The localized nature of these Cu 3d states is responsible for the low charge carrier mobility in excited CuWO4, while the forbidden character of this d–d transition results in a low absorption coefficient in the visible region.37 Furthermore, on the basis of the here collected results by means of different techniques, trap states experimentally detected in the 0.5–1.0 VRHE range are located just below the Cu 3d band

According to Scheme 1, upon ultrafast photoexcitation at 400 nm (3.1 eV), corresponding to an energy just higher than the first localized (Cu 3d in character) band, electrons are promoted from the VB to high energy states of the Cu 3d band, from which they quickly relax within the laser pulse to its bottom or toward midgap states (located just below the Cu 3d band edge).33,38 The observed TA signal is thus ascribed to the photoinduced absorption (ΔA > 0) of electrons trapped in such states, which are promoted to the W 5d-based high energy conduction band (CB) (Scheme 1).

The above attribution is in agreement with that recently reported for Fe2O3 investigated in the same ultrafast time scale34 and matches well with the maximum absorption energy of the involved transition, located at around 500 nm (≈2.5 eV, see TA spectra in Figure 6A), corresponding to the energy difference between the midgap trap states at around 0.5–1.0 VRHE and the highest empty CB located at ca. −2 VRHE (i.e., 4–5 eV higher than the top of the VB).31 Moreover, the ultrafast TA spectra are similar to those recorded in photochromic tests in the presence of a hole scavenger, which is able to fill the holes photogenerated in the VB, thus stabilizing the electrons trapped in midgap states. This rules out any contribution to the TA signal originated by the holes photogenerated in the VB.34

The fastest time constants τ1 and τ2 are associated with the recombination with VB holes of electrons located at a distribution of levels in the high-energy range of the midgap states, while the longest time constant τ3 is associated with the recombination of electrons trapped more in depth in these midgap states, from which the promotion to the highest W 5d CB requires higher photon energy. This is reflected by the slight blue-shift of the EAS spectrum with a τ3 time constant compared to that of the other two components (Figure 6D). This relatively longer-lived, though minority, decay component, associated with deeply trapped electrons, may significantly reduce electron/hole recombination losses favoring hole transfer in water oxidation at the electrode surface.

Massive internal recombination thus appears to be common to both CuWO4 and Fe2O3.31,39 Indeed, for both CuWO4 and Fe2O3 the TA spectrum initially decays very fast, compared to other oxides such as TiO2 or BiVO4.40−42 In early femtosecond-TA studies on Fe2O3, the photoinduced TA signal was found to disappear within 300 ps,33 and the lack of long-lived components in Fe2O3 was reflected by the extremely low photocurrent values (10 μA cm–2) reported in pioneering studies on this material.43 However, recently investigated nanostructured and doped Fe2O3-based electrodes exhibit longer-living components and higher efficiencies,41,44−46 and a similar improvement in performance might be envisaged also in the case of CuWO4.

4. Conclusions

In consideration of the higher photocurrent values measured with CuWO4 in recent studies22,27 and of the encouraging effects recently obtained, for instance, by CuWO4 nanostructuring47 and doping48 or substituting Mo6+ for W6+ ions,49 CuWO4 and other cuprates merit further attention as materials for efficient PEC water splitting and other photocatalytic applications, although several limitations need to be carefully addressed in order to enhance their efficiency. On the basis of the large similarity between CuWO4 and Fe2O3, the use of strategies such as doping, nanostructuring, and the combination with proper oxygen evolution catalysts, which largely contributed to increasing the performance of hematite photoanodes,50 might lead to enhanced efficiency also in the case of CuWO4 photoanodes.

Acknowledgments

This work received financial support from the MIUR PRIN 20173397R7 MULTI-e project, from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie Grant No. 846107, and from a Transition Grant of the University of Milano. The use of instrumentation purchased through the Regione Lombardia-Fondazione Cariplo joint SmartMatLab project (Fondazione Cariplo Grant 2013-1766) is also gratefully acknowledged.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jpcc.0c11607.

  • Reversible photochromism of photoreduced CuWO4, fluence dependence of fs-TA, and TA experiments upon pumping at 500 nm (PDF)

Author Contributions

⊥ I.G. and A.P. contributed equally.

The authors declare no competing financial interest.

Supplementary Material

jp0c11607_si_001.pdf (382.9KB, pdf)

References

  1. Ciamician G. The Photochemistry of the Future. Science 1912, 36, 385–394. 10.1126/science.36.926.385. [DOI] [PubMed] [Google Scholar]
  2. Shaner M. R.; Atwater H. A.; Lewis N. S.; McFarland E. W. A Comparative Technoeconomic Analysis of Renewable Hydrogen Production Using Solar Energy. Energy Environ. Sci. 2016, 9, 2354–2371. 10.1039/C5EE02573G. [DOI] [Google Scholar]
  3. Bushuyev O. S.; De Luna P.; Dinh C. T.; Tao L.; Saur G.; van de Lagemaat J.; Kelley S. O.; Sargent E. H. What Should We Make with CO2 and How Can We Make It?. Joule 2018, 2, 825–832. 10.1016/j.joule.2017.09.003. [DOI] [Google Scholar]
  4. Hu S.; Xiang C.; Haussener S.; Berger A. D.; Lewis N. S. An Analysis of the Optimal Band Gaps of Light Absorbers in Integrated Tandem Photoelectrochemical Water-Splitting Systems. Energy Environ. Sci. 2013, 6, 2984–2993. 10.1039/c3ee40453f. [DOI] [Google Scholar]
  5. Kim J. H.; Hansora D.; Sharma P.; Jang J. W.; Lee J. S. Toward Practical Solar Hydrogen Production-an Artificial Photosynthetic Leaf-to-Farm Challenge. Chem. Soc. Rev. 2019, 48, 1908–1971. 10.1039/C8CS00699G. [DOI] [PubMed] [Google Scholar]
  6. Hisatomi T.; Kubota J.; Domen K. Recent Advances in Semiconductors for Photocatalytic and Photoelectrochemical Water Splitting. Chem. Soc. Rev. 2014, 43, 7520–7535. 10.1039/C3CS60378D. [DOI] [PubMed] [Google Scholar]
  7. Sivula K.; van de Krol R. Semiconducting Materials for Photoelectrochemical Energy Conversion. Nat. Rev. Mater. 2016, 1, 15010. 10.1038/natrevmats.2015.10. [DOI] [Google Scholar]
  8. Sivula K. Metal Oxide Photoelectrodes for Solar Fuel Production, Surface Traps, and Catalysis. J. Phys. Chem. Lett. 2013, 4, 1624–1633. 10.1021/jz4002983. [DOI] [PubMed] [Google Scholar]
  9. Prévot M. S.; Sivula K. Photoelectrochemical Tandem Cells for Solar Water Splitting. J. Phys. Chem. C 2013, 117, 17879–17893. 10.1021/jp405291g. [DOI] [Google Scholar]
  10. Kim J. H.; Jang J.-W.; Jo Y. H.; Abdi F. F.; Lee Y. H.; van de Krol R.; Lee J. S. Hetero-Type Dual Photoanodes for Unbiased Solar Water Splitting with Extended Light Harvesting. Nat. Commun. 2016, 7, 13380. 10.1038/ncomms13380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kim T. W.; Choi K.-S. Nanoporous BiVO4 Photoanodes with Dual-Layer Oxygen Evolution Catalysts for Solar Water Splitting. Science 2014, 343, 990–994. 10.1126/science.1246913. [DOI] [PubMed] [Google Scholar]
  12. Wang S.; Chen P.; Bai Y.; Yun J.-H.; Liu G.; Wang L. New BiVO4 Dual Photoanodes with Enriched Oxygen Vacancies for Efficient Solar-Driven Water Splitting. Adv. Mater. 2018, 30, 1800486. 10.1002/adma.201800486. [DOI] [PubMed] [Google Scholar]
  13. Lhermitte C. R.; Bartlett B. M. Advancing the Chemistry of CuWO4 for Photoelectrochemical Water Oxidation. Acc. Chem. Res. 2016, 49, 1121–1129. 10.1021/acs.accounts.6b00045. [DOI] [PubMed] [Google Scholar]
  14. Benko F. A.; MacLaurin C. L.; Koffyberg F. P. CuWO4 and Cu3WO6 as Anodes for the Photoelectrolysis of Water. Mater. Res. Bull. 1982, 17, 133–136. 10.1016/0025-5408(82)90194-5. [DOI] [Google Scholar]
  15. Doumerc J.-P.; Hejtmanek J.; Chaminade J.-P.; Pouchard M.; Krussanova M. A Photoelectrochemical Study of CuWO4 Single Crystals. Phys. Status Solidi 1984, 82, 285–294. 10.1002/pssa.2210820136. [DOI] [Google Scholar]
  16. Yourey J. E.; Pyper K. J.; Kurtz J. B.; Bartlett B. M. Chemical Stability of CuWO4 for Photoelectrochemical Water Oxidation. J. Phys. Chem. C 2013, 117, 8708–8718. 10.1021/jp402048b. [DOI] [Google Scholar]
  17. Gao Y.; Hamann T. W. Trash. J. Phys. Chem. Lett. 2017, 8, 2700–2704. 10.1021/acs.jpclett.7b00664. [DOI] [PubMed] [Google Scholar]
  18. Bohra D.; Smith W. A. Improved Charge Separation via Fe-Doping of Copper Tungstate Photoanodes. Phys. Chem. Chem. Phys. 2015, 17, 9857–9866. 10.1039/C4CP05565A. [DOI] [PubMed] [Google Scholar]
  19. Hu D.; Diao P.; Xu D.; Xia M.; Gu Y.; Wu Q.; Li C.; Yang S. Copper (II) Tungstate Nanoflake Array Films: Sacrificial Template Synthesis, Hydrogen Treatment, and Their Application as Photoanodes in Solar Water Splitting. Nanoscale 2016, 8, 5892–5901. 10.1039/C5NR09210H. [DOI] [PubMed] [Google Scholar]
  20. Pihosh Y.; Turkevych I.; Mawatari K.; Uemura J.; Kazoe Y.; Kosar S.; Makita K.; Sugaya T.; Matsui T.; Fujita D.; et al. Photocatalytic Generation of Hydrogen by Core-Shell WO3/BiVO4 Nanorods with Ultimate Water Splitting Efficiency. Sci. Rep. 2015, 5, 11141. 10.1038/srep11141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jeon T. H.; Moon G.; Park H.; Choi W. Ultra-Efficient and Durable Photoelectrochemical Water Oxidation Using Elaborately Designed Hematite Nanorod Arrays. Nano Energy 2017, 39, 211–218. 10.1016/j.nanoen.2017.06.049. [DOI] [Google Scholar]
  22. Yourey J. E.; Bartlett B. M. Electrochemical Deposition and Photoelectrochemistry of CuWO4, a Promising Photoanode for Water Oxidation. J. Mater. Chem. 2011, 21, 7651–7660. 10.1039/c1jm11259g. [DOI] [Google Scholar]
  23. Gagne R. R.; Koval C. A.; Lisensky G. C. Ferrocene as an Internal Standard for Electrochemical Measurements. Inorg. Chem. 1980, 19, 2854–2855. 10.1021/ic50211a080. [DOI] [Google Scholar]
  24. Dey S.; Ricciardo R. a.; Cuthbert H. L.; Woodward P. M. Metal-to-Metal Charge Transfer in AWO4 (A = Mg, Mn, Co, Ni, Cu, or Zn) Compounds with the Wolframite Structure. Inorg. Chem. 2014, 53, 4394–4399. 10.1021/ic4031798. [DOI] [PubMed] [Google Scholar]
  25. Dotan H.; Sivula K.; Grätzel M.; Rothschild A.; Warren S. C. Probing the Photoelectrochemical Properties of Hematite (α-Fe2O3) Electrodes Using Hydrogen Peroxide as a Hole Scavenger. Energy Environ. Sci. 2011, 4, 958–964. 10.1039/C0EE00570C. [DOI] [Google Scholar]
  26. Seabold J. A.; Choi K.-S. Efficient and Stable Photo-Oxidation of Water by a Bismuth Vanadate Photoanode Coupled with an Iron Oxyhydroxide Oxygen Evolution Catalyst. J. Am. Chem. Soc. 2012, 134, 2186–2192. 10.1021/ja209001d. [DOI] [PubMed] [Google Scholar]
  27. Gao Y.; Hamann T. W. Quantitative Hole Collection for Photoelectrochemical Water Oxidation with CuWO4. Chem. Commun. 2017, 53, 1285–1288. 10.1039/C6CC09029J. [DOI] [PubMed] [Google Scholar]
  28. Hill J. C.; Ping Y.; Galli G. A.; Choi K.-S. Synthesis, Photoelectrochemical Properties, and First Principles Study of n-Type CuW1–xMoxO4 Electrodes Showing Enhanced Visible Light Absorption. Energy Environ. Sci. 2013, 6, 2440–2446. 10.1039/c3ee40827b. [DOI] [Google Scholar]
  29. Alexander B. D.; Kulesza P. J.; Rutkowska I.; Solarska R.; Augustynski J. Metal Oxide Photoanodes for Solar Hydrogen Production. J. Mater. Chem. 2008, 18, 2298–2303. 10.1039/b718644d. [DOI] [Google Scholar]
  30. Pyper K. J.; Yourey J. E.; Bartlett B. M. Reactivity of CuWO4 in Photoelectrochemical Water Oxidation Is Dictated by a Midgap Electronic State. J. Phys. Chem. C 2013, 117, 24726–24732. 10.1021/jp408434v. [DOI] [Google Scholar]
  31. Thang H. V.; Albanese E.; Pacchioni G. Electronic Structure of CuWO4: Dielectric-Dependent, Self-Consistent Hybrid Functional Study of a Mott-Hubbard Type Insulator. J. Phys.: Condens. Matter 2019, 31, 145503. 10.1088/1361-648X/aaff3e. [DOI] [PubMed] [Google Scholar]
  32. Peeters D.; Mendoza Reyes O.; Mai L.; Sadlo A.; Cwik S.; Rogalla D.; Becker H.-W.; Schütz H. M.; Hirst J.; Müller S.; et al. CVD-Grown Copper Tungstate Thin Films for Solar Water Splitting. J. Mater. Chem. A 2018, 6, 10206–10216. 10.1039/C7TA10759E. [DOI] [Google Scholar]
  33. Cherepy N. J.; Liston D. B.; Lovejoy J. A.; Deng H.; Zhang J. Z. Ultrafast Studies of Photoexcited Electron Dynamics in γ- and α-Fe2O3 Semiconductor Nanoparticles. J. Phys. Chem. B 1998, 102, 770–776. 10.1021/jp973149e. [DOI] [Google Scholar]
  34. Sorenson S.; Driscoll E.; Haghighat S.; Dawlaty J. M. Ultrafast Carrier Dynamics in Hematite Films: The Role of Photoexcited Electrons in the Transient Optical Response. J. Phys. Chem. C 2014, 118, 23621–23626. 10.1021/jp508273f. [DOI] [Google Scholar]
  35. van Stokkum I. H. M.; Larsen D. S.; van Grondelle R. Global and Target Analysis of Time-Resolved Spectra. Biochim. Biophys. Acta, Bioenerg. 2004, 1657, 82–104. 10.1016/j.bbabio.2004.04.011. [DOI] [PubMed] [Google Scholar]
  36. Snellenburg J. J.; Laptenok S.; Seger R.; Mullen K. M.; van Stokkum I. H. M. Glotaran: A Java-Based Graphical User Interface for the R Package TIMP.. J. Stat. Softw. 2012, 49, 1–22. 10.18637/jss.v049.i03. [DOI] [Google Scholar]
  37. Tian C. M.; Jiang M.; Tang D.; Qiao L.; Xiao H. Y.; Oropeza F. E.; Hofmann J. P.; Hensen E. J. M.; Tadich A.; Li W.; et al. Elucidating the Electronic Structure of CuWO4 Thin Films for Enhanced Photoelectrochemical Water Splitting. J. Mater. Chem. A 2019, 7, 11895–11907. 10.1039/C8TA12070F. [DOI] [Google Scholar]
  38. Nadtochenko V. A.; Denisov N. N.; Gak V. Y.; Gostev F. E.; Titov A. A.; Sarkisov O. M.; Nikandrov V. V. Femtosecond Relaxation of Photoexcited States in Nanosized Semiconductor Particles of Iron Oxides. Russ. Chem. Bull. 2002, 51, 457–461. 10.1023/A:1015504217815. [DOI] [Google Scholar]
  39. Huda M. N.; Al-Jassim M. M.; Turner J. A. Mott Insulators: An Early Selection Criterion for Materials for Photoelectrochemical H2 Production. J. Renewable Sustainable Energy 2011, 3, 053101–053111. 10.1063/1.3637367. [DOI] [Google Scholar]
  40. Ravensbergen J.; Abdi F. F.; Van Santen J. H.; Frese R. N.; Dam B.; Van De Krol R.; Kennis J. T. M. Unraveling the Carrier Dynamics of BiVO4: A Femtosecond to Microsecond Transient Absorption Study. J. Phys. Chem. C 2014, 118, 27793–27800. 10.1021/jp509930s. [DOI] [Google Scholar]
  41. Pendlebury S. R.; Wang X.; Le Formal F.; Cornuz M.; Kafizas A.; Tilley S. D.; Grätzel M.; Durrant J. R. Ultrafast Charge Carrier Recombination and Trapping in Hematite Photoanodes under Applied Bias. J. Am. Chem. Soc. 2014, 136, 9854–9857. 10.1021/ja504473e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Grigioni I.; Stamplecoskie K. G.; Selli E.; Kamat P. V. Dynamics of Photogenerated Charge Carriers in WO3/BiVO4 Heterojunction Photoanodes. J. Phys. Chem. C 2015, 119, 20792–20800. 10.1021/acs.jpcc.5b05128. [DOI] [Google Scholar]
  43. Bjorksten U.; Moser J.; Gratzel M. Photoelectrochemical Studies on Nanocrystalline Hematite. Chem. Mater. 1994, 6, 858–863. 10.1021/cm00042a026. [DOI] [Google Scholar]
  44. Huang Z.; Lin Y.; Xiang X.; Rodríguez-Córdoba W.; McDonald K. J.; Hagen K. S.; Choi K.-S.; Brunschwig B. S.; Musaev D. G.; Hill C. L.; et al. In Situ Probe of Photocarrier Dynamics in Water-Splitting Hematite (α-Fe2O3) Electrodes. Energy Environ. Sci. 2012, 5, 8923. 10.1039/c2ee22681b. [DOI] [Google Scholar]
  45. Shen S.; Guo P.; Wheeler D. A.; Jiang J.; Lindley S. A.; Kronawitter C. X.; Zhang J. Z.; Guo L.; Mao S. S. Physical and Photoelectrochemical Properties of Zr-Doped Hematite Nanorod Arrays. Nanoscale 2013, 5, 9867–9874. 10.1039/c3nr03245k. [DOI] [PubMed] [Google Scholar]
  46. Fitzmorris B. C.; Patete J. M.; Smith J.; Mascorro X.; Adams S.; Wong S. S.; Zhang J. Z. Ultrafast Transient Absorption Studies of Hematite Nanoparticles: The Effect of Particle Shape on Exciton Dynamics. ChemSusChem 2013, 6, 1907–1914. 10.1002/cssc.201300571. [DOI] [PubMed] [Google Scholar]
  47. Yang J.; Li C.; Diao P. Molybdenum Doped CuWO4 Nanoflake Array Films as an Efficient Photoanode for Solar Water Splitting. Electrochim. Acta 2019, 308, 195–205. 10.1016/j.electacta.2019.04.044. [DOI] [Google Scholar]
  48. Liang Q.; Guo Y.; Zhang N.; Qian Q.; Hu Y.; Hu J.; Li Z.; Zou Z. Improved Water-Splitting Performances of CuW1–xMoxO4 Photoanodes Synthesized by Spray Pyrolysis. Sci. China Mater. 2018, 61, 1297–1304. 10.1007/s40843-018-9287-5. [DOI] [Google Scholar]
  49. Polo A.; Nomellini C.; Grigioni I.; Dozzi M. V.; Selli E. Effective Visible Light Exploitation by Copper Molybdo-Tungstate Photoanodes. ACS Appl. Energy Mater. 2020, 3, 6956–6964. 10.1021/acsaem.0c01021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sivula K.; Le Formal F.; Grätzel M. Solar Water Splitting: Progress Using Hematite (α-Fe2O3) Photoelectrodes. ChemSusChem 2011, 4, 432–449. 10.1002/cssc.201000416. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

jp0c11607_si_001.pdf (382.9KB, pdf)

Articles from The Journal of Physical Chemistry. C, Nanomaterials and Interfaces are provided here courtesy of American Chemical Society

RESOURCES