An Electrospun Porous CuBi2O4 Nanofiber Photocathode for Efficient Solar Water Splitting

While the CuBi2O4-based photocathode has emerged as an ideal candidate for photoelectrochemical water splitting, it is still far from its theoretical values due to poor charge carrier transport, poor electron–hole separation, and instability caused by self-photoelectric-corrosion with electrolytes. Establishing synthesis methods to produce a CuBi2O4 photocathode with sufficient cocatalyst sites would be highly beneficial for water splitting. Here, the platinum-enriched porous CuBi2O4 nanofiber (CuBi2O4/Pt) with uniform coverage and high surface area was prepared as a photocathode through an electrospinning and electrodeposition process for water splitting. The prepared photocathode material was composed of a CuBi2O4 nanofiber array, which has a freestanding porous structure, and the Pt nanoparticle is firmly embedded on the rough surface. The highly porous nanofiber structures allow the cocatalyst (Pt) better alignment on the surface of CuBi2O4, which can effectively suppress the electron–hole recombination at the electrolyte interface. The as-fabricated CuBi2O4 nanofiber has a tetragonal crystal structure, and its band gap was determined to be 1.8 eV. The self-supporting porous structure and electrocatalytic activity of Pt can effectively promote the separation of electron–hole pairs, thus obtaining high photocurrent density (0.21 mA/cm2 at 0.6 V vs. RHE) and incident photon-to-current conversion efficiency (IPCE, 4% at 380 nm). This work shows a new view for integrating an amount of Pt nanoparticles with CuBi2O4 nanofibers and demonstrates the synergistic effect of cocatalysts for future solar water splitting.


Introduction
It is imperative to find sustainable alternative energy to cope with humankind's energy source crisis [1,2]. Photoelectrochemical water splitting for hydrogen under solar irradiation is seen as the ultimate way to solve the energy crisis [3][4][5][6]. A critical challenge for photoelectrochemical water splitting is the low conversion efficiency suffering from poor charge carrier transport and poor electron-hole separation. In order to improve the efficiency of water splitting, it is necessary to explore a new type of photoelectric material with the best band gap and photocurrent starting potential [7][8][9][10]. In this regard, copperbased oxides-based photocathodes with natural p-type conductivity are a very good choice for their high photocurrents [10][11][12]. Nevertheless, the Cu 3d character in the conduction band of Cu 2 O will lead to photoelectron-induced self-reduction and poor operational stability [10,11]. Therefore, it is urgent to develop a new type of copper-based metal oxide photocathode, so that the Cu 2 O conduction band is sheared, and the photogenerated electrons are directed to the redox stable metal orbitals.
Substantial studies have revealed CuBi 2 O 4 , a multinary p-type metal oxide semiconductor that alloys Cu 2 O with Bi oxide, and its ternary alloy structure allows the photogenerated electrons to be directed toward redox stable metal orbitals [12]. Such CuBi 2 O 4 possesses a sufficiently narrow direct bandgap and exceptionally positive photocurrent onset potential (>1.0 V vs. RHE), thus improving solar energy utilization [13][14][15][16]. For example, a powder-type CuBi 2 O 4 photocatalyst can be realized by hypoxic calcination [17], hydrothermal synthesis [18], and the sol−gel method [19]. Although the photocathode prepared by powder-type CuBi 2 O 4 has various desirable properties, it has not achieved a high photoelectric conversion efficiency. Photocathode corrosion often occurs during oxygen reduction due to the poor transport property of the carrier (~1.2 × 10 −3 cm 2 /Vs) [14]. Therefore, substantial improvements in activity and stability are greatly needed.
Recently, the coupling of film-type CuBi 2 O 4 with different noble metal decorative materials has attracted widespread attention due to its synergistic effect, which can increase photoelectrochemical activity [20]. Such CuBi 2 O 4 film can be realized by hydrothermal synthesis, chemical bath deposition [21], and a template-directed method [22]. For example, Xu et al. demonstrated Au coating film-type CuBi 2 O 4 photocathodes with high photoelectrochemical activity through coupling p-type doping with Au and gradient Cu-vacancy doping [23,24]. Cao et al. fabricated CuBi 2 O 4 film decorated with Pt nanoparticles using atomic layer deposition and indicated an attractive p-type material in water splitting without concern for the corrosion problem in aqueous electrolytes [16]. Park et al. reported a CuO|CuBi 2 O 4 film coated with Pt layers for water splitting and showed more than double the photoactivity compared to the corresponding monolayer photocathode [25]. The photoelectrochemical properties of such polycrystalline thin films can vary significantly, which mainly depend on their morphological details (such as uniform coverage, surface area, and the size and number of cocatalysts) [26][27][28][29]. Although the thin-film CuBi 2 O 4 photocathode system has a high application efficiency, its application scope is limited, especially in dense thin films, which often suffer from the recombination of electron-hole pairs [20]. Therefore, designing a simple and effective synthesis method to obtain a stable extensible film-type CuBi 2 O 4 photoelectrode with sufficient active sites will be very conducive to improving water splitting.
Electrospinning provides a simple and scalable synthesis method to fabricate onedimensional nanomaterial [30] and has been proven particularly useful in the field of photocatalysis [31][32][33]. While electrospinning has been used to fabricate CuO nanofibers [34] and BiVO 4 nanotubes [35] for solar water splitting, no studies have been reported on fabricating CuBi 2 O 4 nanofibers. Here, the novel platinum-enriched porous CuBi 2 O 4 nanofiber (CuBi 2 O 4 /Pt) with uniform coverage and high surface area was prepared as a photocathode through an electrospinning and electrodeposition process for efficient water splitting. The prepared photocathode material was composed of a CuBi 2 O 4 nanofiber array, which has a freestanding porous structure, and the Pt nanoparticle was firmly embedded on the rough surface. The porous nanofiber structure makes the cocatalyst (Pt) better arranged on the CuBi 2 O 4 surface, which effectively prevents the electron-hole pair recombination at the electrolyte interface. The as-fabricated CuBi 2 O 4 nanofiber has a tetragonal crystal structure, and its band gap was determined to be 1.8 eV. The self-supporting porous structure and cocatalytic activity of Pt can effectively promote the separation of electron-hole pairs, resulting in high photocurrent density (0.21 mA/cm 2 at 0.6 V vs. RHE) and IPCE (4% at 380 nm). This study provides a new idea for the integration of Pt nanoparticles and CuBi 2 O 4 nanofibers and provides a synergistic catalyst for future solar water splitting.

Materials and Methods
The CuBi 2 O 4 /Pt nanofiber film was synthesized via a three-step process: electrospinning, annealing, and deposition, which is shown in Figure 1.

Preparation of Porous CuBi2O4/Pt Nanofiber Film
The synthesis of the precursor solution followed two steps: first of all, the Bi(NO3)3·5H2O and Cu(NO3)2·3H2O were added to a mixture of acetic acid and DMF and stirred 1 h to ensure dissolution; then the PVP was added to the above mixture and stirred 10 h to form the homogeneous precursor ( Figure S1a, Supplementary Materials). The electrospinning was carried out with a self-made apparatus [35], which was composed of a plastic syringe, a high voltage supply, and a plate collector. The homogeneous precursor was injected into the syringe with a stainless steel needle (diameter = 0.5 mm). The FTO glass (OPV-FTO-22-07, 2.5 × 3 cm 2 ) was pasted on the counter plate to collect nanofibers. The electrospinning was performed at a distance of 20 cm between the tip of the steel needle and the plate collector, at a high voltage of 20 kV, at an injection rate of 0.1 mL/h, and with an air humidity of 40%. After electrospinning for 25 min, the films collected on the FTO glass ( Figure S1b

Preparation of Porous CuBi 2 O 4 /Pt Nanofiber Film
The synthesis of the precursor solution followed two steps: first of all, the Bi(NO 3 ) 3 ·5H 2 O and Cu(NO 3 ) 2 ·3H 2 O were added to a mixture of acetic acid and DMF and stirred 1 h to ensure dissolution; then the PVP was added to the above mixture and stirred 10 h to form the homogeneous precursor ( Figure S1a, Supplementary Materials). The electrospinning was carried out with a self-made apparatus [35], which was composed of a plastic syringe, a high voltage supply, and a plate collector. The homogeneous precursor was injected into the syringe with a stainless steel needle (diameter = 0.5 mm). The FTO glass (OPV-FTO-22-07, 2.5 × 3 cm 2 ) was pasted on the counter plate to collect nanofibers. The electrospinning was performed at a distance of 20 cm between the tip of the steel needle and the plate collector, at a high voltage of 20 kV, at an injection rate of 0.1 mL/h, and with an air humidity of 40%. After electrospinning for 25 min, the films collected on the FTO glass ( Figure S1b The Pt nanoparticles were loaded onto the CuBi 2 O 4 nanofibers by electrodeposition, as the PtCl 6 2+ can be reduced to Pt nanoparticles at low potential. A three-electrode system was employed with an as-prepared nanofiber on FTO glass (working electrode), an Ag/AgCl reference electrode, and a platinum counter electrode. The electrolyte was 0.1 mM H 2 PtCl 6 •6H 2 O in 0.1 M potassium borate buffer (pH = 7.0). The electrodeposition was carried out using an electrochemical workstation (Zahner Zennium) at −0.20 V versus Ag/AgCl for 1 min.

Physical Characterization
In order to investigate the nanofiber, its morphology was measured by a scanning electron microscope (SEM, Zeiss Merlin), an energy-dispersive X-ray spectroscope (EDS), and a transmission electron microscope (TEM, JEM-2100). Its chemical element and crystallinity were characterized by X-ray photoelectron spectroscopy (XPS, ESCALAB Xi +) and X-ray diffraction (XRD, Bruker Smart-1000CCD diffractometer), respectively. The surface area of the CuBi 2 O 4 nanofiber was measured by the Brunauer-Emmett-Teller (BET, Micromeritics ASAP2460, Norcross, GA, USA). The UV-visible diffuse reflectance spectrum was characterized by a UV-vis spectrophotometer (PE lambda 750) with an integrated sphere attachment.
The photocatalytic H 2 was carried out using the CEL-PAEM-D8 photocatalytic activity evaluation system, which consisted of a gas chromatograph (AgilentTechnologies GC-7890B) and a 300 W Xe lamp (MicroSolar 300, Perfect Light). The circulation water of 25 • C was applied to maintain the reaction temperature of the solution. The photoelectrochemical experiments on the CuBi 2 O 4 /Pt nanofiber photocathode were performed on the Zahner electrochemical workstation in a three-electrode cell (the CuBi 2 O 4 /Pt nanofibers on FTO, Ag/AgCl reference electrode, and a platinum counter electrode, respectively). For photocurrent measurements, the electrolyte was 0.2 M PBS (pH 7.0). The Xe lamp (CEL-HXF300-T3, P = 100 mW/cm 2 , AM1.5) was used as the illumination source. The light intensity was adjusted by a calibrated photodetector. For incident photon-to-current efficiency (IPCE) measurements, the CIMPS TLS03 model (Zahner tunable light source system) was employed for monochromatic light excitation. The chopped photocurrentvoltammetry measurement was carried out with a scan speed of 10 mV/s and a chopped light time of 8 s.

Morphology and Structure of Nanofibers
As indicated in Figure S1b (Supplementary Materials), a typical digital photo of the as-spun film before annealing shows white. After high-temperature annealing, the photocathode shown in Figure S1c (Supplementary Materials) changes to a transparent yellow, which is similar to the spray CuBi 2 O 4 photocathode. The morphology of the CuBi 2 O 4 nanofibers was evaluated by TEM and SEM measurements. As shown in Figure 2a, the randomly oriented nanofibers inherited the one-dimensional structure, and the nonwoven film, which was composed of nanofibers, showed typically interconnected flyover-like network form. The high-magnification image of Figure 2b shows these nanofibers possess a porous fiber structure, which has sufficient surface active sites for photocatalytic reaction. As shown in Figure S3 (Supplementary Materials), the average diameter of the porous nanofiber was 225 nanometers, and its length was up to several dozen micrometers. Under low magnification, Figure 2c reveals one-dimensional morphology, and numerous connective nanoparticles make up the products, similar to the SEM results. The distance between adjacent lattice planes was measured to be 0.31 nm, which belongs to the crystal plane (211) of the CuBi 2 O 4 tetragonal phase (JCPDS 01-080-0996). Figure 2e shows the elemental mapping of the CuBi 2 O 4 nanofiber after depositing Pt. It can be seen that the Bi, Cu, and O elements were uniformly distributed inside the nanofiber. In addition, the Pt element was uniformly distributed on the surface of the nanofiber. attributed to Bi4f7/2 and Bi4f5/2, corresponding to the oxidation state of Bi 3+ . Then, the peaks at 954.0 and 934.4 eV were attributed to Cu 2p1/2 and Cu 2p3/2. Together with a satellite peak at 942.0 eV, the copper mainly existed in the form of Cu 2+ . Besides, the asymmetrical O1s peak (Figure 3e) ranging from 527 eV to 535 eV was fitted into two peaks at 529.4 and 531.0 eV, which were attributed to Cu-O and Sn-O (SnO2 phase of FTO glass) bonds, respectively. Together, both the XRD and XPS results confirmed that these nanofibers were highly crystallized.  As indicated in Figure 4a, the CuBi2O4 nanofibers exhibited strong UV-vis absorbance in both the ultraviolet and visible light regions, and their absorption cutoff wavelength was about 650 nm. By employing the linear part of (αhv) 2 vs. hv, the band gap was calculated to be 1.8 eV, which is similar to that reported in other studies (1.74 eV) [17]. From the nitrogen adsorption and desorption isotherms of CuBi2O4 nanofibers (Figure 4b), the specific surface area of the CuBi2O4 nanofibers was calculated to be 20.5 m 2 g −1 using the Brunauer-Emmet-Teller model, which was larger than the value (14 m 2 g −1 ) of nanoparticles reported in other research [36].  Figure 4a, the CuBi 2 O 4 nanofibers exhibited strong UV-vis absorbance in both the ultraviolet and visible light regions, and their absorption cutoff wavelength was about 650 nm. By employing the linear part of (αhv) 2 vs. hv, the band gap was calculated to be 1.8 eV, which is similar to that reported in other studies (1.74 eV) [17]. From the nitrogen adsorption and desorption isotherms of CuBi 2 O 4 nanofibers (Figure 4b), the specific surface area of the CuBi 2 O 4 nanofibers was calculated to be 20.5 m 2 g −1 using the Brunauer-Emmet-Teller model, which was larger than the value (14 m 2 g −1 ) of nanoparticles reported in other research [36].

Photoelectrochemical Performance
To investigate the photoelectrochemical activity of the CuBi 2 O 4 photocathode, the photocurrent-voltammetry measurement under AM 1.5 was performed. As shown in Figure 5a, the photocurrent started to appear at the initial potential of 1.0 V vs. RHE, increased rapidly when the lamp was turned on, and decreased when the lamp was turned off, indicating that the photocurrent was generated under light irradiation. Interestingly, the instantaneous photocurrent overshoot could also be observed when the lamp was switched on/off, which indicates that electrons accumulate in the space charge layer and reverse recombination occurs between electrons and holes. Moreover, the chopped photocurrent for the CuBi 2 O 4 /Pt nanofibers showed little cathodic transient spikes, which presumably were caused by surface recombination. Figure 5b indicates the photocurrentvoltage (J-V) curves under AM 1.5 irradiation. The photocurrent of CuBi 2 O 4 nanofibers rose slowly when decreasing the potential, and yielded −0.12 mA/cm 2 at 0.6 V vs. RHE.
On the contrary, after depositing the Pt nanoparticles, their photocurrent rose quickly when decreasing the potential and yielded −0.21 mA/cm 2 at 0.6 V vs. RHE. Remarkably, the photocurrent of CuBi 2 O 4 /Pt nanofibers was about 75% higher than that of the pristine nanofibers. Then, the IPCE measurement by the tunable light source TLS03 model at 0.6 V vs. RHE was performed. As indicated in Figure 5c, with the increase in illuminant wavelength, the IPCE values gradually decreased to zero at 650 nm (1.8 eV), which was consistent with its band gap energy. Significantly, compared to those of pristine nanofibers (1.8% at 380 nm), the as-prepared CuBi 2 O 4/ Pt nanofibers showed a higher IPCE reaching up to 4% at 380 nm. As shown in Table S1, the nanofibers decorated with Pt exhibit higher photoelectrochemical performance (the value of photocurrent and IPCE) than that of CuBi 2 O 4 nanofilm (0.15 mA/cm 2 at 0.6 V vs. RHE) [21]. Nevertheless, the gradient self-doping nanofilm (0.50 mA/cm 2 at 0.6 V vs. RHE) showed higher photoelectrochemical performance than that of the as-prepared CuBi 2 O 4 /Pt nanofiber due to its internal electric field promoting charge separation. [19].

Photoelectrochemical Performance
To investigate the photoelectrochemical activity of the CuBi2O4 photocathode, the photocurrent-voltammetry measurement under AM 1.5 was performed. As shown in Figure 5a, the photocurrent started to appear at the initial potential of 1.0 V vs. RHE, increased rapidly when the lamp was turned on, and decreased when the lamp was turned off, indicating that the photocurrent was generated under light irradiation. Interestingly, the instantaneous photocurrent overshoot could also be observed when the lamp was switched on/off, which indicates that electrons accumulate in the space charge layer and reverse recombination occurs between electrons and holes. Moreover, the chopped photocurrent for the CuBi2O4/Pt nanofibers showed little cathodic transient spikes, which presumably were caused by surface recombination. Figure 5b indicates the photocurrent-voltage (J-V) curves under AM 1.5 irradiation. The photocurrent of CuBi2O4 nanofibers rose slowly when decreasing the potential, and yielded −0.12 mA/cm 2 at 0.6 V vs. RHE. On the contrary, after depositing the Pt nanoparticles, their photocurrent rose quickly when decreasing the potential and yielded −0.21 mA/cm 2 at 0.6 V vs. RHE. Remarkably, the photocurrent of CuBi2O4/Pt nanofibers was about 75% higher than that of the pristine nanofibers. Then, the IPCE measurement by the tunable light source TLS03 model at 0.6 V vs. RHE was performed. As indicated in Figure 5c, with the increase in illuminant wavelength, the IPCE values gradually decreased to zero at 650 nm (1.8 eV), which was consistent with its band gap energy. Significantly, compared to those of pristine nanofibers (1.8% at 380 nm), the as-prepared CuBi2O4/Pt nanofibers showed a higher IPCE reaching up to 4% at 380 nm. As shown in Table S1, the nanofibers decorated with Pt exhibit higher photoelectrochemical performance (the value of photocurrent and IPCE) than that of CuBi2O4 nanofilm (0.15 mA/cm 2 at 0.6 V vs. RHE) [21]. Nevertheless,the gradient self-doping nanofilm (0.50 mA/cm 2 at 0.6 V vs. RHE) showed higher photoelectrochemical performance than that of the as-prepared CuBi2O4/Pt nanofiber due to its internal electric field promoting charge separation. [19]. (c) IPCE spectrum(blue and red lines represent before and after depositing Pt).
As indicated in Figure 6a, the photocurrent of the CuBi2O4/Pt nanofibers photocathode decreased obviously with the increase in time at 0.6 V versus RHE and decreased by nearly 40% after 25 min illumination. Nevertheless, the photostability of the CuBi2O4/Pt nanofibers photocathode was better than that of the pure Cu2O [37] and CuO (52% reduction of photocurrent after 25 min illumination [38]) photocathode. The photocurrent decay was mainly caused by photocorrosion, and a similar phenomenon also appeared at the CuBi2O4 photocathode consisting of open windows and struts [14]. The photocatalytic H2 production of the fabricated nanofibers was measured by the gas chromatography-mass spectrometer (GC-7890B). As shown in Figure 6b, CuBi2O4/Pt nanofibers exhibited higher photocatalytic performance (380 μmol/(g·h)) compared to the CuBi2O4 nanofibers (290 μmol/(g·h)). Nonetheless, its photocatalytic performance As indicated in Figure 6a, the photocurrent of the CuBi 2 O 4 /Pt nanofibers photocathode decreased obviously with the increase in time at 0.6 V versus RHE and decreased by nearly 40% after 25 min illumination. Nevertheless, the photostability of the CuBi 2 O 4/ Pt nanofibers photocathode was better than that of the pure Cu 2 O [37] and CuO (52% reduction of photocurrent after 25 min illumination [38]) photocathode. The photocurrent decay was mainly caused by photocorrosion, and a similar phenomenon also appeared at the CuBi 2 O 4 photocathode consisting of open windows and struts [14]. The photocatalytic H 2 production of the fabricated nanofibers was measured by the gas chromatographymass spectrometer (GC-7890B). As shown in Figure 6b, CuBi 2 O 4/ Pt nanofibers exhibited higher photocatalytic performance (380 µmol/(g·h)) compared to the CuBi 2 O 4 nanofibers (290 µmol/(g·h)). Nonetheless, its photocatalytic performance was much lower than that of Pt/TiO2 nanosheet with exposed (001) facet (8500 µmol/(g·h), [39]). The main reason was that although the one-dimensional nanostructure could reduce the recombination of photogenerated electron-hole pairs, some electron-hole pairs still recombined due to the poor transport property of the carrier (~1.2 × 10 −3 cm 2 /Vs, [14]). In addition, the photocatalytic activity of these nanofibers slightly decreased with time due to photocorrosion. The instability of CuBi 2 O 4 nanofibers presents a major challenge for solar water splitting, and protection layers using atomic layer deposition were essential in order to use them as a practical photocathode. Figure 6a, the photocurrent of the CuBi2O4/Pt nanofibers photocathode decreased obviously with the increase in time at 0.6 V versus RHE and decreased by nearly 40% after 25 min illumination. Nevertheless, the photostability of the CuBi2O4/Pt nanofibers photocathode was better than that of the pure Cu2O [37] and CuO (52% reduction of photocurrent after 25 min illumination [38]) photocathode. The photocurrent decay was mainly caused by photocorrosion, and a similar phenomenon also appeared at the CuBi2O4 photocathode consisting of open windows and struts [14]. The photocatalytic H2 production of the fabricated nanofibers was measured by the gas chromatography-mass spectrometer (GC-7890B). As shown in Figure 6b, CuBi2O4/Pt nanofibers exhibited higher photocatalytic performance (380 μmol/(g·h)) compared to the CuBi2O4 nanofibers (290 μmol/(g·h)). Nonetheless, its photocatalytic performance was much lower than that of Pt/TiO2 nanosheet with exposed (001) facet (8500 μmol/(g·h), [39]). The main reason was that although the one-dimensional nanostructure could reduce the recombination of photogenerated electron-hole pairs, some electron-hole pairs still recombined due to the poor transport property of the carrier (~1.2 × 10 −3 cm 2 /Vs, [14]). In addition, the photocatalytic activity of these nanofibers slightly decreased with time due to photocorrosion. The instability of CuBi2O4 nanofibers presents a major challenge for solar water splitting, and protection layers using atomic layer deposition were essential in order to use them as a practical photocathode.  Based on the above analysis of experimental data, the transfer process of the electronhole pair in the photocathode was shown in Figure 7. Under light illumination, the CuBi 2 O 4 nanofibers could absorb photons and excite the valence electron to the conduction band. Then the photogenerated electron moved to the interface between the photocathode and the electrolyte due to the downward band bending and injected into the electrolyte to take part in the reduction reaction of hydrogen, which was similar to the charge transfer of BiVO 4 anode [35]. As shown in Figure 5a, there are amounts of recombination of electron-hole pairs during the water splitting reaction. The photoelectrochemical test demonstrates that the CuBi 2 O 4 /Pt nanofibers show better photocatalytic activity. The main reasons are as follows: Firstly, nanofibers have a large specific surface area and porous structure, which maintain good contact with electrolytes and enrich the active sites. Moreover, the nanofibers, possessing a one-dimensional structure, can also shorten the length of hole diffusing to the FTO substrate and decrease the recombination of the electron-hole pair. In addition, the Pt cocatalyst can also efficiently extract a photogenerated electron from the space charge layer. Therefore, more electrons can be transferred into electrolytes to take part in the reduction reaction of hydrogen, and the electron-hole recombination is significantly hindered. In summary, the main reasons for the enhanced photoelectrochemical performance of CuBi 2 O 4 /Pt nanofiber photocathode are porous nanofibers and cocatalysts. substrate and decrease the recombination of the electron-hole pair. In addition, the Pt cocatalyst can also efficiently extract a photogenerated electron from the space charge layer. Therefore, more electrons can be transferred into electrolytes to take part in the reduction reaction of hydrogen, and the electron-hole recombination is significantly hindered. In summary, the main reasons for the enhanced photoelectrochemical performance of CuBi2O4/Pt nanofiber photocathode are porous nanofibers and cocatalysts.

Conclusions
In summary, the platinum-enriched porous CuBi2O4 nanofibers (CuBi2O4/Pt) with uniform coverage and high surface area were prepared as a photocathode through an electrospinning and electrodeposition process for improving the photoelectrochemical water splitting. The CuBi2O4 nanofibers showed an average diameter of 225 nanometers, and lengths up to several dozens of micrometers. The porous nanofiber structures allow the cocatalyst (Pt) to have better alignment on the surface of CuBi2O4, which can effectively hinder the electron-hole recombination at the electrolyte interface. These nanofibers have a tetragonal crystal structure, and their band gap was determined to be 1.8 eV. After depositing Pt nanoparticles, their photocurrent density was 0.21 mA/cm 2 at 0.6 V vs. RHE under AM 1.5 illumination, and the IPCE was 4% at 380 nm. The enhanced photoelectrochemical ability was mainly attributed to the porous nanofibers, large specific surface area, and the cocatalytic activity of Pt nanoparticles. This work shows a new view for integrating an amount of Pt nanoparticles with CuBi2O4 nanofibers, indicating the synergistic effect of cocatalysts for efficient storage of solar energy into hydrogen.

Conclusions
In summary, the platinum-enriched porous CuBi 2 O 4 nanofibers (CuBi 2 O 4 /Pt) with uniform coverage and high surface area were prepared as a photocathode through an electrospinning and electrodeposition process for improving the photoelectrochemical water splitting. The CuBi 2 O 4 nanofibers showed an average diameter of 225 nanometers, and lengths up to several dozens of micrometers. The porous nanofiber structures allow the cocatalyst (Pt) to have better alignment on the surface of CuBi 2 O 4 , which can effectively hinder the electron-hole recombination at the electrolyte interface. These nanofibers have a tetragonal crystal structure, and their band gap was determined to be 1.8 eV. After depositing Pt nanoparticles, their photocurrent density was 0.21 mA/cm 2 at 0.6 V vs. RHE under AM 1.5 illumination, and the IPCE was 4% at 380 nm. The enhanced photoelectrochemical ability was mainly attributed to the porous nanofibers, large specific surface area, and the cocatalytic activity of Pt nanoparticles. This work shows a new view for integrating an amount of Pt nanoparticles with CuBi 2 O 4 nanofibers, indicating the synergistic effect of cocatalysts for efficient storage of solar energy into hydrogen.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/polym13193341/s1, Figure S1: Digital micrograph of (a) solution precursor, (b) PVP-CuBi 2 O 4 nanofiber mat before annealing and (c) CuBi2O4 nanofiber mat after annealing; Figure S2: TG-DSC curve of the crystallization of CuBi2O4 nanofiber; Figure S3: CuBi 2 O 4 nanofiber (a) SEM, (b) corresponding diameter distribution; Table S1: Comparison of photocurrent data reported in the literature with the photocurrent value obtained in the present study.

Data Availability Statement:
The data presented in this study are available upon request from the corresponding author.