Photoelectrochemical Water Splitting and H₂ Generation Enhancement Using an Effective Surface Modification of W-Doped TiO₂ Nanotubes (WT) with Co-Deposition of Transition Metal Ions

Abstract

W-doped TiO₂ nanotube arrays (WT) were fabricated by in situ electrochemical anodization of titanium substrate. The results of the influence of different photo-deposited transition ions (Cr_xFe_1−x, 0 ≤ x ≤ 1) on the surface of WT on photoelectrochemical (PEC) water splitting and H₂ generation are presented. The crystallinities, structural, elemental, and absorption analysis were conducted by XRD, SEM, RAMAN, EDX, and UV–Vis absorption spectroscopy, which demonstrated anatase as the main crystalline phase of TiO₂, and the existence of Cr_xFe_1−x (nano)particles/film deposited on the surface of WT. The SEM images revealed that the deposition rate and morphology are highly related to the ratio of Cr and Fe ions. Under visible light illumination, the entire photoelectrodes showed a very good response to light with stable photocurrent density. PEC measurements revealed that the mixture of transition ions with a certain ratio of ions (Cr_0.8Fe_0.2–T) led to enhanced photocurrent density more than that of other modifiers due to decreasing charge recombination as well as improving the charge transfer. Moreover, PEC water splitting was conducted in an alkaline solution and the Cr_0.8Fe_0.2–T photoelectrode generated 0.85 mL cm⁻² h⁻¹ H₂, which is over two times that of pristine WT.

1. Introduction

Hydrogen (H₂) as renewable energy has attracted significant attention to be the best replacement for fossil fuels to move toward a decarbonized future owing to unique properties such as high efficiency and energy density (120–142 MJ/kg), the ability to obtain from H₂O and generate only H₂O with its combustion [1,2]. Photoelectrochemical (PEC) water splitting using sunlight is the most promising strategy to generate hydrogen. Early work on TiO₂ for water splitting and PEC H₂ generation was reported by Fujishima and Honda in 1972 [3]. Since then, extensive efforts have been invested in exploring and introducing new materials such as WO₃, MnO_x, Bismuth-based nanomaterials, and Fe₂O₃ [4,5,6,7]. Although a large number of semiconductors have been inspected over the past few decades, TiO₂ is still particularly appealing because of suitable band-edge positions and relatively good stability in different electrolytes as well as photo-corrosion resistance [8,9]. The TiO₂ nanotube (NTs) structures, as one-dimensional (1D) nanostructures, have been considered highly promising candidates with high efficiency in PEC water splitting because of high specific surface area, electron mobility, and mechanical strength [10,11]. However, pristine TiO₂ is still suffering from the utilization of less than 5% of sunlight, thanks to its wide band gap (≥3.0 eV) [12] and its modification is necessary to improve photo-response in the visible range and enhance the efficiency of chemical solar energy conversion [13,14]. To achieve this, many efforts have been made such as composites with other metal oxide and/or carbon-based materials, metal or non-metal doping, and surface modification [13,15,16,17]. For example, composite with other semiconductors such as WO₃ applied in this work, not only enhances the visible light absorption but also increases the photo-generated charge (e⁻/h⁺) separation [18]. Another simple and effective technique is through the TiO₂ NTs surface modification, especially with the transition metals. Transition metals can improve PEC activities of TiO₂ NTs through (i) decreasing the band gap by introducing new sub-levels because of partially filled D-orbitals, and (ii) decreasing charge recombination by trapping photo-generated charge (e⁻/h⁺) of TiO₂ [9,19,20]. Suligoj et al. have recently reported that deposited transition metal ion clusters on the TiO₂, via surface modification, can also act as co-catalysts and improve the photocatalytic activity of TiO₂ [21].

There are several methods to deposit transition metal (oxide) particles on the surface of TiO₂ NTs such as sputtering, chemical bath deposition, electrodeposition, and photo-deposition [13,21,22,23,24,25]. Among these, photo-deposition is one of the easiest methods that does not need an elevated temperature or applied (bias) potential. Moreover, adjusting some parameters in this method (metal precursor, pH, sacrificial reagent, time of deposition, etc.), gives the possibility to control the size and geometrical distribution, and oxidation state of the deposited (nano)particles [15,22,26].

In this work, self-organized WO₃ doped TiO₂ NTs (WT) were successfully fabricated through an optimized in situ electrochemical anodization of titanium substrate according to our previous studies [13]. Then, Fe and Cr ions were chosen for the WT surface modification via the photo-deposition method. Different mixtures of them, Cr_xFe_1-x (0 ≤ x ≤ 1), were also prepared for photo-deposition to clarify the effects of different ratios of two transition metal ions. The Cr_xFe_1-x–T (0 ≤ x ≤ 1) abbreviation has been used throughout the manuscript; x stands for the molar ratio of Cr and Fe ions used in the photo-deposition precursor for the WT surface modification. Cr_xFe_1−x–T (0 ≤ x ≤ 1) photoelectrodes were submitted for thorough material characterization, electrochemical and semiconducting properties, and the activity in PEC water splitting for H₂ evolution.

2. Materials and Methods

2.1. Materials

Titanium foils (Ti, 99%) with 1 mm thickness, sodium tungstate dihydrate (Na₂WO₄ × 2 H₂O, 99%), hydrogen fluoride (HF, 40%), nitric acid (HNO₃, 65%), dimethyl sulfoxide (C₂H₆SO, 99.5%), ethanol (C₂H₆O, 99%), potassium hydroxide (KOH, 499%), ethylene glycol (C₂H₆O₂, 99.5%), potassium chromate (K₂CrO₄, 99.5%), and iron (III) chloride hexahydrate (FeCl₃ × 6 H₂O, 99%) were purchased from commercial sources and were used as purchased and without further purification. Solutions were prepared using deionized (DI) water within this work.

2.2. In Situ Electrochemical Anodization Procedure

The TiO₂ NTs were prepared by in situ electrochemical anodization with the two-electrode system including Ti foil and platinum-coated titanium mesh electrodes as anode and cathode electrodes, respectively. Prior to anodization, the surface of Ti sheets (1 × 3 cm²) was polished with different emery types of abrasive papers and then cleaned by rinsing with DI water. Afterward, Ti sheets were chemically etched with a mixture of HF:HNO₃:H₂O in a volume ratio of 1:4:5 for 30 s at room temperature and then washed with DI water. In situ electrochemical anodization was performed at 40 V for 8 h in a solution of dimethyl sulfoxide as an electrolyte containing 12 mM of sodium tungstate dihydrate, HF (2 vol%), and H₂O (1 vol%). Finally, as-anodized amorphous W-doped TiO₂ nanotubes were annealed at 400 °C in air for 2 h, with a heating rate of 2 °C min⁻¹ (F3L-1720, AZAR Furnaces, Tehran, Iran) (Figure 1A).

2.3. Photo-Deposition Procedure

Surface modification of WT with Cr_xFe_1−x (0 ≤ x ≤ 1) was accomplished using the photo-deposition method. Briefly, the WT electrodes were soaked in 25 mL precursor of K₂CrO₄ (10 mM), or FeCl₃ × 6H₂O (10 mM), and different molar ratio of Fe and Cr as Cr_xFe_1-x, (x = 0, 0.2, 0.4, 0.5, 0.6, 0.8, and 1), containing ethanol (10 vol%). Photo-deposition was performed by illuminating the solution in which WT electrodes were immersed with a high-pressure mercury lamp (400 W) for 1 h. After the photo-deposition, photoelectrodes were washed with DI water (Figure 1B). Accordingly, obtained photoelectrodes are denoted onward throughout the manuscript as Cr_xFe_1−x–T (x = 0, 0.2, 0.4, 0.5, 0.6, 0.8, and 1), corresponding to the used molar ratio of Fe and Cr in the photo-deposition precursor.

2.4. Material Characterization

A Philips XL30 (Eindhoven, The Netherlands) field-emission scanning electron microscope equipped with an energy dispersive X-ray (EDX) mapping port was used to characterize the morphology and the elemental distribution of the microscopic region of the materials. X-ray diffraction using a PMD Philips X-Pert (Panalytical, Almelo, The Netherlands) was used for crystal structure characterization. The Raman spectra of the samples were obtained with a TakRam N1-541 (Tehran, Iran) Raman spectrometer. Moreover, the UV–vis absorption spectra were measured by a JASCO V-570 (Tokyo, Japan) UV–vis spectrophotometer.

2.5. Photoelectrochemical (PEC) Characterization and PEC Water Splitting

PEC measurements were carried out in KOH (1 M, pH = 13.9) solution containing ethylene glycol (5 vol%) with a potentiostat/galvanostat OGF 500 (Origsflex, Lyon, France). Linear sweep voltammetry (LSV), chronoamperometry (CA), and open circuit potential (OCP) measurements were conducted in a standard three-electrode configuration including as-prepared Cr_xFe_1−x–WT photoelectrodes, Ag/AgCl electrode, and platinum foil as working electrode, reference electrode, and counter electrode, respectively (Figure 1C). All of these electrochemical tests were carried out in the presence, absence, and chopped light. The photoelectrodes were illuminated by a 35 W Xenon (Xe) lamp (200 mW cm⁻²) equipped with a UV cut-off filter (λ ˃ 420 nm). The potential of the photoelectrodes was reported versus (vs.) the reversible hydrogen electrode (RHE) using Equation (1):

$${\mathrm{E}}_{\mathrm{RHE}}=0.1976\mathrm{V}+0.059\mathrm{pH}+{\mathrm{E}}_{\mathrm{Ag}/\mathrm{AgCl}}$$

(1)

The electrochemical impedance spectra (EIS) were also performed on the electrochemical workstation system Zahner Zenium Pro (Zahner-Elektrik, Kronach, Germany) in a range of 100 kHz−100 mHz under 1.22 V vs. RHE and a DC voltage of 10 mV amplitude in the dark and under illumination in the same electrolyte used in PEC measurements.

The same three-electrode configuration was used for PEC water splitting (Figure 1C). Cr_xFe_1−x–WT photoelectrodes were placed in the anodic chamber while immersed in KOH (1 M, pH = 13.9) solution containing ethylene glycol (5 vol%) and irradiated by Xe lamp (200 mW cm⁻², and λ ˃ 420 nm). A platinum-coated titanium mesh electrode as the cathode was inserted into an inverted burette, where H₂ was collected [27,28,29]. PEC water splitting was continued by applying a constant external potential of 1.23 V vs. RHE. The volume of the produced hydrogen gas was quantitatively measured via electrolyte displacement level in the inverted burette at predefined time (Figure 1C).

3. Results and Discussion

3.1. Material Characterization

Our previous results revealed that highly ordered WT was fabricated through in situ electrochemical anodization of Ti foil in an electrolyte containing Na₂WO₄. Although, the concentration of Na₂WO₄ did not show a significant influence on WT morphology. However, the PEC activity of fabricated WT can change and the maximum PEC activity for WT was obtained using ~12 mM of Na₂WO₄ as an optimum concentration [13,30]. Hence, in this work, WT photoelectrode was fabricated according to the optimal conditions established in our previous studies [13,30].

The surface morphology of pristine WT nanotubes can be seen in Figure 2. Uniform and highly ordered nanotubes were observed with an average inner diameter and wall thickness of about 110 nm and 25 nm, respectively. The surface modification of WTs was then performed with the photo-deposition of Cr_xFe_1−x (0 ≤ x ≤ 1) and their SEM images have been compared in Figure 3. In the case of Cr₀Fe₁–WT, the precursor of photo-deposition only consisted of Fe ions, particles with sizes of ≤200 nm × 200 nm were observed on the WT surface (Figure 3A). Although particles are fairly homogeneously distributed over the WT surface, some of them covered the top of the nanotubes (Figure 3A) which might decrease PEC activity.

Interesting effects can be observed regarding the different ratios of Fe and Cr ions in the photo-deposition precursor (0.2 ≤ x ≤ 0.8, Figure 3B–F), particularly when comparing with Figure 3A,G which correspond to using single metal ions, Cr₀Fe₁–WT and Cr₁Fe₀–WT, respectively. Remarkably, with the introduction of a low ratio of Cr ions in the photo-deposition precursor, in the case of Cr_0.2Fe_0.8–WT and Cr_0.4Fe_0.6–WT, smaller rounded nanoparticles (diameter ≤ 25 nm) were deposited onto the edge of the WT nanotubes (Figure 3B,C). However, with a more increasing Cr ratio, e.g., Cr_0.5Fe_0.5–WT, a well-decorated amorphous composite of Cr and Fe as a uniform film on the top of WT was observed (Figure 3D). In the case of Cr_0.6Fe_0.2–WT, a similar coated film is noticeable with an agglomeration of nanoparticles on the top of the film (Figure 3E). In the case of Cr_0.8Fe_0.2–WT, there is no trace of the above-mentioned film covering the WT surface. Nanoparticles are homogeneously deposited on the WT surface and few partial agglomerations are observed on the WT surface (Figure 3F). It can be stated that the Fe photo-deposition rate is higher than that of Cr (comparing Figure 3A,G) since, in the case of Cr₁Fe₀–WT, fewer and smaller nanoparticles can be seen all over the edge of open-top WT nanotubes.

The elemental EDS and EDS-mapping images were provided in Figure 4A–C and Figures S1 and S2 (Supplementary Materials), revealing the presence of Ti and O elements, as well as W, Fe, and Cr which confirmed the successfully fabricated Cr_xFe_1−x–WT photoelectrodes. It can be clearly seen that the Fe and Cr were homogeneously distributed on the surface of WT, as demonstrated by the elemental mapping (Figures S1 and S2 (Supplementary Materials)). The amount of Fe and Cr deposited on the surface of WT, listed in Table S1, showed that Fe has a rather higher photo-deposition rate than that of Cr; comparing the Cr₀Fe₁–WT and Cr₁Fe₀–WT. This result can explain the SEM results where bigger and more agglomerations of Fe particles were obtained rather than that of Cr (Figure 3A,G).

The X-ray diffractograms of TiO₂ NTs, WT, Cr₀Fe₁–WT, Cr_0.5Fe_0.5–WT, and Cr₁Fe₀–WT are shown in Figure 4D. Generally, similar XRD patterns were observed for all measured photoelectrodes but Cr_0.5Fe_0.5–WT. Diffraction peaks can be assigned to metallic Ti substrate and TiO₂ at its anatase phase with evidence of the peaks at 2θ of 25.3°, 37.8°, 48.1°, 53.9°, and 70.3° which could be attributed to the (101), (004), (200), (105), and (220) planes of anatase, respectively (ICDD-JCPD: 01-086-1156). The crystalline phase of tungsten (W) was not detected by XRD in the samples which can be due to the incorporation of a low concentration of W into the TiO₂ lattice since the radii of W⁺⁶ and Ti⁺⁴ are so close to each other as explained in our previous reports [13,31]. In the case of Fe_0.5Cr_0.5–WT as shown in the XRD pattern, the anatase peaks at 25.3° and 48.1° were not detected. That can be attributed to deposited amorphous Cr_0.5Fe_0.5 film which covered the top of the WT photoelectrode as mentioned in SEM results (Figure 3D). There is no sign of Fe or Cr crystalline phase in XRD patterns. However, this result does not exclude the possible deposited amorphous phase of Cr_xFe_1−x at the surface of WT.

Raman was employed to further structural analysis of Cr_xFe_1−x-WT photoelectrodes. The characteristic peaks observed at 159, 401, 520, and 637 cm⁻¹ are consistent with those of anatase TiO₂; corresponding to Eg, B1g, A1g, and Eg, respectively (Figure 4E). No crystalline phases related to other species were observed.

The optical properties of Cr_xFe_1−x–WT photoelectrodes were also studied (Figure S3 (Supplementary Materials) and the band gaps of photoelectrodes have been estimated from the Tauc plot (Figure 4F and Figure S3). Cr_xFe_1−x–WT (0 ≤ x ≤ 1) photoelectrodes exhibit better visible region absorption than that of pristine WT, especially in the case of using Cr ions for the WT surface modification. The estimated band gaps were found to be in the range of 2.16–2.65 eV after the photo-deposition of Cr_xFe_1−x, (0 ≤ x ≤ 1)) on the WT surface (Table S2). As can be seen in Figure S3, in the case of Cr₁Fe₀–WT, a peak around 550 nm was observed that belongs to the presence of Cr on the surface of WT photoelectrodes. This kind of peak, λ > 500 nm, was also observed in other literature [26,32].

3.2. PEC Characterization

LSV measurements were performed to examine the photo-sensitivity of Cr_xFe_1−x–WT by recording current density as responses of as-prepared photoelectrodes, through scanning the voltage in the dark, under illumination, and by switching the light on and off (Figure 5A,B and Figure S4). The current density of Cr_xFe_1−x–WT is near zero in the dark (Figure 5A) and significant light sensitivity was observed under light illumination for all Cr_xFe_1−x–WT photoelectrodes. In some cases, particularly Cr_0.5Fe_0.5–WT, an anodic peak centered at +0.75 V was observed which can be assigned to surface oxidation of the deposited amorphous composite Cr_xFe_1−x (0.5 ≤ x ≤ 0.6) film on the surface of WT photoelectrode (see Figure 3D,E). Cr_0.8Fe_0.2–WT showed the maximum photocurrent density. It achieved a value of 0.48 mA cm⁻² at 1.23 V vs. RHE, which corresponds to a ~2 times increase compared to the pristine WT (Figure 5B). The chopped light polarization curves revealed that Cr_xFe_1−x–WT (0 ≤ x ≤ 1) photoelectrodes have good repeatability and stability during light irradiation (Figure S5).

Chronoamperometry (CA) measurement was conducted to evaluate the stability of the photocurrent density of Cr_xFe_1−x–WT (Figure 5C). The response of Cr_xFe_1−x–WT photoelectrodes to light is rapid and after the initial spike, the photocurrent decreases due to the strong e⁻/h⁺ recombination and then stabilizes less than 60 s. Generally, the current density of pristine WT was increased after the surface modification. The photocurrent density of pristine WT was 0.43 mA cm⁻² which enhanced to 0.63 mA cm⁻² for the Cr_0.8Fe_0.2–WT photoelectrode at a bias of +1.62 V vs. RHE (Figure 5C). Highly sensitive photocurrent responses were observed upon switching the light on and off. The photocurrent was enhanced significantly under illumination and returned to near zero value as the light was turned off (Figure 5D).

Open circuit potential (OCP) of pristine WT and Cr_xFe_1−x–WT photoelectrodes were investigated to study photovoltage as well as the lifetime of photo-generated e⁻/h⁺. Characteristic behavior of an n-type semiconducting material was observed (Figure 6A); upon illumination, the potential of Cr_xFe_1−x–WT shifts to more negative values. Afterward, when the light is turned off, the potential of Cr_xFe_1−x–WT increases back to near initial values prior to the illumination and a decay profile appears. In all Cr_xFe_1−x–WT photoelectrodes, a fast change in OCP was observed after the light is switched on which correlated with the fast response to light in photocurrent measurements. The photovoltages are ≥ 400 mV, which suggests a significant concentration of photo–generated electrons (Figure 6A). The higher photovoltages for Cr_xFe_1−x–WT were obtained rather than that of pristine WT, suggesting a more remarkable photo response after the WT surface modification. Moreover, it has been reported that the decay profile of OCP after turning the light off represents the charge recombination kinetics (Figure 6B) [9,13].

Then, by fitting OCP decay profiles with a bi-exponential function (Equation (2)), the average recombination lifetime of photo charges was calculated.

$$\mathrm{V}\left(\mathrm{t}\right)={\mathrm{V}}_0+{\mathrm{A}}_1{\mathrm{e}}^{\raisebox{1ex}{$-\mathrm{t}$}\!\left/ \!\raisebox{-1ex}{${\mathsf{\tau }}_1$}\right.}+{\mathrm{A}}_2{\mathrm{e}}^{\raisebox{1ex}{$-\mathrm{t}$}\!\left/ \!\raisebox{-1ex}{${\mathsf{\tau }}_2$}\right.}$$

(2)

where V, t, V₀, A₁ and A₂ are potential decay, time, fitting constants, and τ₁ and τ₂ are exponential lifetime components for recombination processes in the bulk and surface, respectively.

The harmonic mean of decay lifetime (τ_m) is calculated by Equation (3) and the data are summarized in Table S3.

$${\mathsf{\tau }}_{\mathrm{m}}=({\mathrm{A}}_1{\mathsf{\tau }}_1\times {\mathrm{A}}_2{\mathsf{\tau }}_{2)}/\left({\mathrm{A}}_1{\mathrm{A}}_2\right)$$

(3)

The longest photo charge lifetime was obtained in the case of Cr_0.8Fe_0.2–WT photoelectrode, indicating more efficient charge recombination suppression compared with other Cr_xFe_1−x–WT photoelectrodes.

To further confirm the e⁻/h⁺ separation and transfer of photogenerated carriers, electrochemical impedance spectroscopy (EIS) was measured. EIS data under the dark and light illumination have been plotted as Nyquist plots in Figure 6C,D, fitted by the equivalent circuit model (the inset plot in Figure 6C) to simulate the Nyquist plots. R_s is the uncompensated solution resistance, R_ct denotes the charge transfer resistance, and CPE is defined as CPE-T and CPE-P, representing the constant phase elements. The fitting impedance parameter values are listed in

Table 1. EIS fitting parameters of Cr_xFe_1−x–WT photoelectrodes under illumination.

Photoelectrode	Rct (kΩcm−2)	CPE
		n	Q (μFcm−2)	C (μFcm−2)
WT	7.247	0.97	71.9	70.3
Cr0Fe1–WT	7.541	0.90	106	104
Cr0.2Fe0.8–WT	9.670	0.92	86.0	84.7
Cr0.4Fe0.6–WT	5.716	0.95	73.8	70.4
Cr0.5Fe0.5–WT	6.511	0.94	81.3	78.3
Cr0.6Fe0.4–WT	4.253	0.95	81.9	77.8
Cr0.8Fe0.2–WT	3.420	0.94	88.3	81.8
Cr1Fe0–WT	10.016	0.91	117	119

.

The diameter of the semicircle of the Nyquist plot represents the photoelectrode charge transfer resistance and the smaller arc radius means better e⁻/h⁺ separation efficiency (Figure 6C,D). The charge transfer resistances of photoelectrodes under light illumination are much smaller than those obtained in the dark (Figure 6C). Moreover, the Cr_0.8Fe_0.2–WT photoelectrode showed the smallest R_ct value, indicating better charge separation and faster charge transfer kinetics at the electrode interface.

3.3. PEC Water Splitting and H₂ Generation

PEC water splitting using Cr_xFe_1−x–WT photoelectrodes was conducted via an inverted burette setup (Figure 1C). The volume of H₂ production was then quantitatively recorded according to the variation of the electrolyte level in the burette every 5 min. As can be seen (Figure 7A), the Cr_0.8Fe_0.2–WT photoelectrode produced more H₂ than that of pristine WT and other photoelectrodes in an alkaline electrolyte. Water splitting for H₂ evolution using Cr_0.8Fe_0.2–WT conducted during four cycles of 60 min (Figure 7B) showed no significant decrease in activity during consecutive cycles, which proves the good stability of the as-prepared electrodes in the PEC water splitting. The PEC evolved H₂ in this work compared with some of the other materials used in PEC water splitting in

Table 2. Comparison of H₂ evolution using TiO₂-based nanomaterials.

Photoelectrode	Electrolyte	Light Source	Rate of H2 Evolution	Ref.
TiO2 NP	MeOH + H2O	150 W halide	0.270 µmol g−1min−1	[33]
1Pt-3WO3 -TiO2 NTs	H2O +20 vol% EtOH	UV	5 µL h−1	[34]
Cr-doped TiO2 NTs	NaOH	Vis	9.70 μL cm−2 h−1	[35]
Pd-TiO2 NTs	H2O +50 vol% MeOH	UV	70 μL cm−2 h−1	[36]
Ru-TiO2 NTs	Na2SO4 (2 M)	Vis	29 μmol cm−2 h−1	[37]
Fe-TiO2 NTs	NaOH	Vis	10 μL cm−2 h−1	[38]
Cu-TiO2	H2O +50 vol% EtOH	UV	81.1 µmol g−1h−1	[20]
Cr0.8Fe0.2–WT	KOH + 10 vol% EG	Vis	0.85 mL cm−2 h−1	This work

.

4. Conclusions

A simple and effective surface modification was applied by photo-deposition of a single ion of Fe, Cr, as well as dual ions (Cr_xFe_1−x, 0 ≤ x ≤ 1) on the surface of WT to fabricate Cr_xFe_1−x–WT photoelectrodes. The determined band gaps of the photoelectrodes using Tauc plots were in a range of 2.16–2.65 eV which is less than pristine WT. The ratio of Fe and Cr showed a significant effect on the deposition rate. In some ratios, an amorphous layer of Cr_xFe_1−x (0.5 ≤ x ≤ 0.6) covered the WT surface which is unfavorable for the separation of photo-generated e⁻/h⁺ and consequently decreases the PEC efficiency of these photoelectrodes. The Cr and Fe doping individually might improve the photocurrent by expended visible light absorbance as well as the effective separation of photo-generated e⁻/h⁺. However, the ratio of dual co-catalysts should be carefully optimized. On the other hand, Cr_0.8Fe_0.2–WT made a great improvement in PEC activities. Better PEC activity of Cr_0.8Fe_0.2–WT can be due to less charge transfer resistance as well as a longer photo charge lifetime. The maximum evolved H₂ was also obtained for Cr_0.8Fe_0.2–WT photoelectrodes which are over two times that of pristine WT.