Improving Photoelectrochemical Activity of Magnetron-Sputtered Double-Layer Tungsten Trioxide Photoanodes by Irradiation with Intense Pulsed Ion Beams

The photoelectrochemical (PEC) activity of metal oxide photoelectrodes for water-splitting applications can be boosted in several different ways. In this study, we showed that PEC activity can be significantly improved with a double-layer (crystalline-amorphous) configuration of WO3 thin films irradiated with intense pulsed ion beams (IPIB) of a nanosecond duration. It was found that IPIB irradiation promotes the formation of crystalline and sponge-like WO3 structures on the surface. Due to an increase in the active surface and light scattering in irradiated samples, photocurrent generation increased by ~80% at 1.23 reversible hydrogen electrodes (RHE).


Introduction
At the present moment, photoelectrochemical (PEC) water splitting is considered as one of the viable methods for meeting renewable energy needs. Solar-light-based hydrogen production could offer new opportunities for a variety of industries [1,2]. However, further research for efficiency improvement is needed, as it is relatively low compared with other renewable energy sources [3,4].
Tungsten trioxide WO 3 is a metal oxide semiconductor material with a wide band gap, which can be tuned from 3.2 eV to 2.5 eV using various methods of modifications [5][6][7]. It could act as a photoanode by absorbing sunlight and generating electron-hole pairs for oxygen evolution. However, it has some drawbacks, such as photo corrosion caused by peroxo-species generation on the surface, the slow-moving kinetics of generated holes, poor charge separation [8], etc. Doping, composite material construction, heterostructured material fabrication, and surface nanostructuring may all play important roles in mitigating such disadvantages [9][10][11][12][13]. To examine the photoelectrodes associated with their morphologies, tungsten oxides in the form of nanoplates, -fibers, and -rods, and wedge-or sheet-like structures were created and tested [14][15][16][17][18]. It has also been reported that surface and morphology engineering can improve the performance of photoelectrode materials [19]. In this regard, etching and surface passivation could help to improve surface electron-hole kinetics, reduce the overpotential of thin films, and achieve an improvement in photoelectric conversion [20]. Moreover, type annealing in various atmospheres is another method that has been used to improve the photocatalytic activities of various materials because annealing has a significant impact on the structural properties of materials [21,22]. Approaches such as laser and ion irradiation may be used during the deposition or post-deposition treatments for tuning electrical, morphological, and structural characteristics [23][24][25].
In addition to the ways already stated, computational [26] and experimental strategies [27] are widely employed to increase the photoactivity of WO 3 . Recent works report on the highly enhanced photocatalytic activity of the WO 3 layer embedded in heterostructures [28][29][30]. The proper band alignment of photoanode layers with water oxidation potentials opens a route to significantly increasing photocurrent generation because of the better charge separation in multiple-layered heterostructures. Additionally, the PEC performance of photoelectrodes can be boosted by combining 2D materials with WO 3 [31,32], controlling the texturing of particular crystallographic facets [33][34][35], creating unassisted tandem cell structures [36], using plasmonic nanoparticles [37], and by creating oxygen vacancies [38].
In this study, we propose a new method to modify a WO 3 photoelectrode that combines simultaneous surface nanostructuring and annealing for enhancing PEC properties. Highcurrent-density pulsed ion beam (IPIB) irradiation promotes the formation of the spongelike porous surface of WO 3 with a simultaneous annealing process. Obtained samples were analyzed for morphological, structural, optical, and PEC characteristics. Hence, a novel method of surface engineering allows us to create crystalline porous sponge-like thin films with increased surface area and absorbance directly from amorphous films.

Sample Preparation
We prepared amorphous, crystalline, and double-layer WO 3 samples on fluorine doped tin oxide (FTO) substrates. We washed commercial FTO (~350 nm) on glass substrates in ethanol with ultrasonic bath before deposition procedure. We used the Kurt Lesker magnetron system for the reactive DC deposition of WO 3 using a tungsten target (W, Kurt J. Lesker, 99.95% purity, 2.00 diameter) and holding argon/oxygen flow rates at 33/21 sccm. During the deposition, we assigned the sputtering power as 100 W, and rotated the substrate at a rate of 20 rounds per minute. We held base pressure during the deposition at 11.4 mTorr.
Initially, we deposited amorphous WO 3 thin film samples for 45 min. It was already reported that the magnetron-sputtered WO 3 has an amorphous nature [39]. Then, we annealed half samples at 500 • C for 20 min in air on a hot plate to obtain crystalline samples. We irradiated the prepared amorphous and crystalline samples with intense pulsed ion beam (IPIB) with ion beam current densities from 10 A/cm 2 to 32 A/cm 2 . We performed IPIB irradiation by the INURA accelerator installed at Nazarbayev University (Nur-Sultan, Kazakhstan) [40]. INURA is a pulsed-power architecture ion accelerator with a total power consumption less than 10 kW. The peak power density of the beam at the target is up to 10 MW/cm 2 . Main parameters of the accelerator are as follows: accelerating voltage is in the range from 300 to 400 kV, total beam current is 10 kA, beam pulse length is 100 ns (FWHM). Irradiation by single pulse of 100 ns of ion beam induces superfast annealing process, with a hundred-nanosecond heating followed by a microseconds-scale cooling process. Next, we fabricated irradiated and non-irradiated double-layer WO 3 thin films with thicknesses approx. 380 nm (top amorphous ∼120 nm, bottom crystalline ∼260 nm). Firstly, we deposited an amorphous WO 3 layer for 30 min on two FTO substrates and annealed on a hot plate for 20 min in air to form crystalline film samples. Secondly, on the top of that crystalline film, we deposited a second layer of amorphous WO 3 films for 15 min. Both FTO substrates were in the same conditions during the annealing and deposition processes. Then, we performed irradiation with IPIB at a current density of 6 A/cm 2 on one of the prepared double-layer WO 3 .

PEC Characterization
We used a three-electrode cell configuration with double-layer WO 3 as a working electrode for the PEC testing. We measured PEC characteristics of working electrodes with Nanomaterials 2022, 12, 2639 3 of 10 Ag/AgCl (3.5 M KCl) reference electrode and Pt wire-counter electrode. We measured the photocurrent generation of samples in 0.1 M Na 2 SO 4 (pH 7.2) electrolyte using a potentiostat CHI660E in conjunction with a PLS-EM solar simulator (100 mW/cm 2 , AM 1.5). We used the following equation to convert the measured potential to the reversible hydrogen electrode (RHE) scale: where, V Ag/AgCl is the applied potential and 0.205 is the standard potential of the Ag/AgCl, KCl (3.5 M) reference electrode. We performed the electrochemical impedance spectroscopy (EIS) in a frequency range from 0.01 Hz to 10 kHz with an AC voltage amplitude of 50 mV at a DC bias of 0.68 V vs. RHE under light illumination. We used EIS Spectrum Analyzer software to obtain circuit model fitting of EIS data [41]. Figure 1 shows surface SEM images of crystalline (a-d) and amorphous (e-h) WO 3 thin films on an FTO/glass substrate after ion beam irradiation with current densities ranging from 10 A/cm 2 to 32 A/cm 2 . SEM images of crystalline samples irradiated with an ion beam current density of 10 A/cm 2 ( Figure 1a) had no visible differences with SEM images of non-irradiated samples; therefore, 10 A/cm 2 irradiation has no or very little effect on the surface morphology of crystalline WO 3 thin film. It can be seen that irradiating with ion beam current densities of 16 A/cm 2 and 26 A/cm 2 (1b and 1c) drastically changed the initial thin film morphology and transformed it into a porous thin film with empty spaces opening the substrate material underneath. A high ion beam current density of 32 A/cm 2 exposed the sample in the same way; however, the open spaces became wider. The same ion beam current densities had different effects on amorphous sample morphologies compared with crystalline samples, as seen in Figure 1e-h. After irradiation, the overall appearance of all irradiated amorphous films resembled sponge-like structures. As the ion beam current density increases from 10 to 32 A/cm 2 , the shape irregularities and sizes of the sponge-like structures become larger. It can be concluded that IPIB irradiation causes more morphological changes on amorphous surfaces than on crystalline.

PEC Characterization
We used a three-electrode cell configuration with double-layer WO3 as a working electrode for the PEC testing. We measured PEC characteristics of working electrodes with Ag/AgCl (3.5 M KCl) reference electrode and Pt wire-counter electrode. We measured the photocurrent generation of samples in 0.1 M Na2SO4 (pH 7.2) electrolyte using a potentiostat CHI660E in conjunction with a PLS-EM solar simulator (100 mW/cm 2 , AM 1.5). We used the following equation to convert the measured potential to the reversible hydrogen electrode (RHE) scale: where, VAg/AgCl is the applied potential and 0.205 is the standard potential of the Ag/AgCl, KCl (3.5 M) reference electrode. We performed the electrochemical impedance spectroscopy (EIS) in a frequency range from 0.01 Hz to 10 kHz with an AC voltage amplitude of 50 mV at a DC bias of 0.68 V vs. RHE under light illumination. We used EIS Spectrum Analyzer software to obtain circuit model fitting of EIS data [41]. Figure 1 shows surface SEM images of crystalline (a-d) and amorphous (e-h) WO3 thin films on an FTO/glass substrate after ion beam irradiation with current densities ranging from 10 A/cm 2 to 32 A/cm 2 . SEM images of crystalline samples irradiated with an ion beam current density of 10 A/cm 2 ( Figure 1a) had no visible differences with SEM images of non-irradiated samples; therefore, 10 A/cm 2 irradiation has no or very little effect on the surface morphology of crystalline WO3 thin film. It can be seen that irradiating with ion beam current densities of 16 A/cm 2 and 26 A/cm 2 (1b and 1c) drastically changed the initial thin film morphology and transformed it into a porous thin film with empty spaces opening the substrate material underneath. A high ion beam current density of 32 A/cm 2 exposed the sample in the same way; however, the open spaces became wider. The same ion beam current densities had different effects on amorphous sample morphologies compared with crystalline samples, as seen in Figure 1e-h. After irradiation, the overall appearance of all irradiated amorphous films resembled sponge-like structures. As the ion beam current density increases from 10 to 32 A/cm 2 , the shape irregularities and sizes of the sponge-like structures become larger. It can be concluded that IPIB irradiation causes more morphological changes on amorphous surfaces than on crystalline.  We measured the PEC characteristics of amorphous and crystalline WO 3 films irradiated with ion beams of current densities and found that increasing ion beam current density reduces the photocurrent generation of crystalline WO 3 thin films, as shown in Figure S1. In the case of amorphous samples, photocurrent generation was not observed at any current density of irradiation. This could be attributed to an increase in various defects in films and the formation of open spaces, where FTO is accessible to electrolytes and leads to a short circuit during the photocurrent measurement. Therefore, we propose to utilize a double-layer film, an amorphous layer on top of a crystalline layer that prevents the formation of open spaces exposing the substrate. It was found that by irradiating at an ion beam current density of 6 A/cm 2 , there were no open spaces on the crystalline bottom layer, while a good sponge-like structure was observed on the top. Figure 2 shows SEM images of the surface and cross-section of double-layer WO 3 thin films before and after irradiation with an ion beam current density of 6 A/cm 2 , i.e., confirming the nanostructuring of the top amorphous layer after irradiation.  Figure 2. Calculations revealed that a ~1600 nm ion beam range (penetration depth) in a WO3/FTO/glass multilayer structure, which indicates that the energy of the intense pulsed ion beam is absorbed mostly near the surface layer of the irradiating material. The absorbed energy of the intense pulsed ion beam leads to the super-fast heating of this surface layer with subsequent fast cooling, resulting in the modification of the surface morphology and crystal structure.

Results and Discussions
The surface of a non-irradiated sample has a granular microstructure; however, after ion beam irradiation, the morphology changes to a sponge-like structure with randomly distributed open spaces. Furthermore, as shown in Figure 2b, the formation of some clusters can be seen on the top layer of the irradiated sample. Figure 2c,d show a SEM cross-sectional view of the photoelectrodes before and after ion irradiation, and also show the sponge-like structure and formation of clusters. The irradiated double-layer WO 3 thin film in Figure 2d clearly shows that the bottom crystalline layer beneath remains unchanged after the irradiation process.  Figure 2. Calculations revealed that a~1600 nm ion beam range (penetration depth) in a WO 3 /FTO/glass multilayer structure, which indicates that the energy of the intense pulsed ion beam is absorbed mostly near the surface layer of the irradiating material. The absorbed energy of the intense pulsed ion beam leads to the super-fast heating of this surface layer with subsequent fast cooling, resulting in the modification of the surface morphology and crystal structure.
The surface of a non-irradiated sample has a granular microstructure; however, after ion beam irradiation, the morphology changes to a sponge-like structure with randomly distributed open spaces. Furthermore, as shown in Figure 2b, the formation of some clusters can be seen on the top layer of the irradiated sample. Figure 2c,d show a SEM cross-sectional view of the photoelectrodes before and after ion irradiation, and also show the sponge-like structure and formation of clusters. The irradiated double-layer WO 3 thin film in Figure 2d clearly shows that the bottom crystalline layer beneath remains unchanged after the irradiation process.
Grazing incidence X-ray diffraction (GI-XRD) was used to examine the structural properties of non-irradiated and irradiated double-layer WO 3 photoelectrodes. Figure 3 shows the corresponding GI-XRD patterns. The bottom layer of the non-irradiated sample is crystalline because it was hot-plate annealed after deposition, and the top layer is amorphous. The red pattern in Figure 3 indicates the dominance of a monoclinic WO 3 phase (PDF2, No:01-075-2072) in the bottom annealed layer. However, after irradiation, both the top amorphous layer and the bottom crystalline layer transformed into a dominating tetragonal phase (PDF2, No:01-089-1287), as shown in Figure 3 (blue pattern). The peaks of the black pattern shown on the bottom plot are from the FTO substrate and are less visible on irradiated and non-irradiated samples since the GI-XRD mode probes only a thin surface layer. Peak (020), which existed in the monoclinic phase, vanished after the ion beam irradiation. We also performed GI-XRD measurements on a one-layer irradiated amorphous sample to ensure that after a 6 A/cm 2 ion beam irradiation, it does indeed become a predominantly crystalline tetragonal WO 3 , as shown in Figure S2. Therefore, both top and bottom layers of our double-layer sample become crystalline with a dominating tetragonal phase because of the super-fast annealing by IPIB irradiation. The GI-XRD peaks seen in Figure S2 for amorphous WO 3 (blue pattern) are assigned to peaks from the FTO substrate. High-resolution GI-XRD patterns in Figure S4 confirm the amorphous nature of as-deposited WO 3 , and it is consistent with the literature [39].
Nanomaterials 2022, 12, x FOR PEER REVIEW 6 of measurements on a one-layer irradiated amorphous sample to ensure that after a 6 A/cm 2 ion beam irradiation, it does indeed become a predominantly crystalline tetragonal WO 3 , as shown in Figure S2. Therefore, both top and bottom layers of our double-layer sample become crystalline with a dominating tetragonal phase because of the super-fast annealing by IPIB irradiation. The GI-XRD peaks seen in Figure S2 for amorphous WO 3 (blue pattern) are assigned to peaks from the FTO substrate. High-resolution GI-XRD patterns in Figure S4 confirm the amorphous nature of as-deposited WO 3 , and it is consistent with the literature [39].    Figure 4a,b depict the absorption curves and Tauc plots of non-irradiated and irradiated double-layer WO 3 , respectively. Obviously, as depicted in Figure 4a, the irradiated double-layer sample has a higher absorbance than that of the non-irradiated sample. It can be explained by electron trapping in induced oxygen vacancies, which result after irradiation with IPIB [42]. Light scattering through the walls of sponge-like porous inclusions and light trapping on the irradiation-induced point defects of the interstitial oxygen sublattice result in increased absorption. The Tauc method was used to analyze the band gap change after the irradiation process [43]. Tauc plots for both samples in Figure 4b show the same 3.35 eV band gap before and after irradiation. Nanomaterials 2022, 12, x FOR PEER REVIEW 6 of 10 Electrochemical impedance spectroscopy (EIS) was used to determine the charge transfer properties for irradiated and non-irradiated double-layer samples, as shown in Figure 5. It is observed that the irradiated sample shows a smaller radius of the semicircle than a non-irradiated sample, i.e., it implies that the nanostructured sample is more beneficial for the charge transfer between a photoelectrode and electrolyte. Irradiation-induced surface nanostructuring improves the charge transport properties, i.e., sponge-like structures facilitate hole migration to the WO3/electrolyte junction and increase the surface-active sites. Furthermore, light trapping in such a structure promotes the increased formation of electron-hole pairs through the improved photon absorption. The linear sweep voltammetry (LSV) curves of non-irradiated and irradiated doublelayer WO3 films are depicted in Figure 6. Additional chopped LSV curves in Figure S3 confirm that prepared photoelectrodes show photocatalytic activity. The photocurrent of 0.32 mA/cm 2 was measured at 1.23 RHE for bare WO3, and 0.58 mA/cm 2 was measured at 1.23 RHE for irradiated WO3 photoelectrodes. This ~80% increase in the photocurrent density of an irradiated sample compared with a non-irradiated sample could be due to the following factors: Firstly, an IPIB-beam-induced sponge-like structure could act as a light-trapping structure, enhancing absorption. Additionally, more absorption improvements occurred due Electrochemical impedance spectroscopy (EIS) was used to determine the charge transfer properties for irradiated and non-irradiated double-layer samples, as shown in Figure 5. It is observed that the irradiated sample shows a smaller radius of the semicircle than a non-irradiated sample, i.e., it implies that the nanostructured sample is more beneficial for the charge transfer between a photoelectrode and electrolyte. Irradiation-induced surface nanostructuring improves the charge transport properties, i.e., sponge-like structures facilitate hole migration to the WO 3 /electrolyte junction and increase the surface-active sites. Furthermore, light trapping in such a structure promotes the increased formation of electron-hole pairs through the improved photon absorption.  Electrochemical impedance spectroscopy (EIS) was used to determine the charge transfer properties for irradiated and non-irradiated double-layer samples, as shown in Figure 5. It is observed that the irradiated sample shows a smaller radius of the semicircle than a non-irradiated sample, i.e., it implies that the nanostructured sample is more beneficial for the charge transfer between a photoelectrode and electrolyte. Irradiation-induced surface nanostructuring improves the charge transport properties, i.e., sponge-like structures facilitate hole migration to the WO3/electrolyte junction and increase the surface-active sites. Furthermore, light trapping in such a structure promotes the increased formation of electron-hole pairs through the improved photon absorption. The linear sweep voltammetry (LSV) curves of non-irradiated and irradiated doublelayer WO3 films are depicted in Figure 6. Additional chopped LSV curves in Figure S3 confirm that prepared photoelectrodes show photocatalytic activity. The photocurrent of 0.32 mA/cm 2 was measured at 1.23 RHE for bare WO3, and 0.58 mA/cm 2 was measured at 1.23 RHE for irradiated WO3 photoelectrodes. This ~80% increase in the photocurrent density of an irradiated sample compared with a non-irradiated sample could be due to the following factors: Firstly, an IPIB-beam-induced sponge-like structure could act as a light-trapping structure, enhancing absorption. Additionally, more absorption improvements occurred due The linear sweep voltammetry (LSV) curves of non-irradiated and irradiated doublelayer WO 3 films are depicted in Figure 6. Additional chopped LSV curves in Figure S3 confirm that prepared photoelectrodes show photocatalytic activity. The photocurrent of 0.32 mA/cm 2 was measured at 1.23 RHE for bare WO 3 , and 0.58 mA/cm 2 was measured at 1.23 RHE for irradiated WO 3 photoelectrodes. This~80% increase in the photocurrent density of an irradiated sample compared with a non-irradiated sample could be due to the following factors: to light capture by point defects in the oxygen sublattice [42]. Secondly, nanostructuring caused by ion bombardment may increase the active surfaces of the interface layers between the solution and photoanode. Thirdly, IPIB irradiation converts the top amorphous layer of WO3 into a crystalline phase. It was already proved that crystalline WO3 has an advantage over amorphous WO3 in PEC performance [44]. Accordingly, IPIB irradiation could be considered as one of the methods for creating porous and crystalline materials with improved PEC activity. In general, irradiation with ion beams is widely used for material modifications [45]. As a result of ion beam impact, a series of effects, such as temperature increases, ion doping, and atomic displacement, cause changes in the target material. Continuous ion beams with low current densities are widely used for tailoring the electronic, morphological, and structural properties of target materials. For example, in Ref. [24], WO3 films were irradiated with 190 keV Ar+ ion beams with current densities ranging from 1 to 2 μA cm −2 . It was shown that morphological changes (formation of pores) appear when the fluence of the Ar+ ion beam reaches 10 17 ions·cm −2 . For the IPIB, the irradiation of materials with very short and powerful pulses causes a very fast and intense energy deposition onto the material's surface. For instance, we use 100 ns-duration IPIB with a typical fluence of ions in a single pulse of about 10 12 protons·cm −2 , which provides a rate of ions to the target up to 10 19 protons cm −2 s −1 and corresponds to the power density of 1 MW/cm 2 . This high-power density is deposited into the thin surface layer, leading to a fast temperature increase. It can be used with a higher beam intensity, even in high-energy-density physics, to ignite warm-dense-matter states [46]. In practical applications, the fast-heating annealing effect can be used for the modification of the near-surface layer of bulk targets [47]. The main result in our paper is the demonstration of the applicability of IPIB with appropriate current densities for the useful structural modification of photoactive thin films based on WO3. Due to the fast injection of ions, the required fluence of IPIB (10 12 ions·cm −2 ) is significantly lower when compared with a continuous ion beam (higher than 10 16 ions·cm −2 ).

Conclusions
In summary, we introduced a method for creating porous double-layer WO3 films with improved PEC activity. Irradiation of double-layer WO3 films with a single pulse of IPIB leads to a working electrode with a sponge-like crystalline surface layer with improved light absorption and charge transport properties. We revealed that the photocurrent was increased to ~80% because of the increase in the active surface and enhancement Firstly, an IPIB-beam-induced sponge-like structure could act as a light-trapping structure, enhancing absorption. Additionally, more absorption improvements occurred due to light capture by point defects in the oxygen sublattice [42]. Secondly, nanostructuring caused by ion bombardment may increase the active surfaces of the interface layers between the solution and photoanode. Thirdly, IPIB irradiation converts the top amorphous layer of WO 3 into a crystalline phase. It was already proved that crystalline WO 3 has an advantage over amorphous WO 3 in PEC performance [44]. Accordingly, IPIB irradiation could be considered as one of the methods for creating porous and crystalline materials with improved PEC activity.
In general, irradiation with ion beams is widely used for material modifications [45]. As a result of ion beam impact, a series of effects, such as temperature increases, ion doping, and atomic displacement, cause changes in the target material. Continuous ion beams with low current densities are widely used for tailoring the electronic, morphological, and structural properties of target materials. For example, in Ref. [24], WO 3 films were irradiated with 190 keV Ar+ ion beams with current densities ranging from 1 to 2 µA cm −2 . It was shown that morphological changes (formation of pores) appear when the fluence of the Ar+ ion beam reaches 10 17 ions·cm −2 . For the IPIB, the irradiation of materials with very short and powerful pulses causes a very fast and intense energy deposition onto the material's surface. For instance, we use 100 ns-duration IPIB with a typical fluence of ions in a single pulse of about 10 12 protons·cm −2 , which provides a rate of ions to the target up to 10 19 protons cm −2 s −1 and corresponds to the power density of 1 MW/cm 2 . This highpower density is deposited into the thin surface layer, leading to a fast temperature increase. It can be used with a higher beam intensity, even in high-energy-density physics, to ignite warm-dense-matter states [46]. In practical applications, the fast-heating annealing effect can be used for the modification of the near-surface layer of bulk targets [47]. The main result in our paper is the demonstration of the applicability of IPIB with appropriate current densities for the useful structural modification of photoactive thin films based on WO 3 . Due to the fast injection of ions, the required fluence of IPIB (10 12 ions·cm −2 ) is significantly lower when compared with a continuous ion beam (higher than 10 16 ions·cm −2 ).

Conclusions
In summary, we introduced a method for creating porous double-layer WO 3 films with improved PEC activity. Irradiation of double-layer WO 3 films with a single pulse of IPIB leads to a working electrode with a sponge-like crystalline surface layer with improved light absorption and charge transport properties. We revealed that the photocurrent was increased to~80% because of the increase in the active surface and enhancement of charge transport characteristics. Future research can address the stability issues of irradiated WO 3 films. We believe that our method can be utilized and further studied to boost photoactive materials by IPIB irradiation.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nano12152639/s1. Figure S1: LSV curves of irradiated crystalline WO 3 thin films with increasing ion beam current density. Black arrows indicate that photocurrent generation reduces with increasing ion beam current densities; Figure S2: GI-XRD patterns of initially amorphous WO 3 thin film before and after irradiation with 6 A/cm 2 ion beam current density; Figure S3: chopped linear sweep voltammetry curves of double-layer WO 3 before and after the modification with IPIB of 6 A/cm 2 ion beam current density; Figure S4: high-resolution GI-XRD patterns of amorphous WO 3 and FTO substrate.