UV-Visible-Near-Infrared-Driven Photoelectrocatalytic Urea Oxidation and Photocatalytic Urea Fuel Cells Based on Ruddlensden–Popper-Type Perovskite Oxide La 2 NiO 4

: Photocatalysis and photoelectrocatalysis, as green and low-cost pollutant treatment tech-nologies, have been widely used to simultaneously degrade pollutants and produce clean energy to solve the problems of environmental pollution and energy crisis. However, the disadvantages of photocatalysts in a narrow absorption range and low utilization rate of solar energy still hinder the practical application. Here we fabricate two-dimensional porous Ruddlensden–Popper type nickel-based perovskite oxide La 2 NiO 4 as a noble metal-free photoanode for photoelectrocatalytic urea oxidation under full spectrum sunlight irradiation. The transient photocurrent density under near infrared (NIR) light ( λ > 800 nm) can reach 50 µ A cm − 2 . Urea wastewater was used as the fuel to obtain low-energy hydrogen production, and round-the-clock hydrogen production was achieved with the optimal yield of 22.76 µ mol cm − 2 h − 1 . Moreover, a photocatalytic urea fuel cell (PUFC) was constructed with La 2 NiO 4 as the photoanode. The power density under UV-vis-NIR was 0.575 µ W cm − 2 . Surprisingly, the ﬁlling factor (FF) under NIR light was 0.477, which was much higher than those under UV-vis-NIR and visible light. The results demonstrated that PUFCs constructed from low-cost nickel-based perovskite oxides have potential applications for low-energy hydrogen production and efﬁcient utilization of sunlight.


Introduction
With fast industrialization and urbanization, urea, one of the most common pollutants in wastewater, inevitably produces a mass of ammonia and nitrate under natural degradation, which poses severe risks to the environment and public health [1]. Thus, it is of great significance to develop cost-effective ways for urea removal in wastewater [2]. Meanwhile, urea is widely recognized as an efficient material for H 2 transportation and storage due to its high energy density (16.9 MJ L −1 ) [3]. Compared with electrolytic water (1.23 V vs. RHE), electrolytic urea solution shows a relatively lower theoretical electrolytic voltage (0.37 V vs. RHE), which is beneficial for electrolytic H 2 production from wastewater containing urea [4][5][6][7]. The largest source of urea is urine discharge from humans and livestock. The high concentration of urea in human urine (~0.33 M) has been extensively utilized to produce hydrogen and to directly construct urine fuel cells [8][9][10][11][12].

of 14
Li et al. reported the photoelectrocatalytic oxidation of urea with Ni(OH) 2 modified TiO 2 as the photoelectrode, for the first time, in 2012, which aroused great interest in solar-driven hydrogen released from urea and human urine [22]. Xie et al. further improved the photoelectrochemical oxidation activity of urea by using a Ni(OH) 2 -modified Ti-doped α-Fe 2 O 3 photoanode [23]. However, conventional photocatalysts (TiO 2 , α-Fe 2 O 3 , BiVO 4 etc.) have a narrow solar absorption band, which greatly restricts utilization of solar energy [24,25]. Currently, there are only a few reports on urea oxidation reaction (UOR) by photoelectrocatalysts that utilizes the entire solar spectrum (from UV to NIR region), especially NIR light [26][27][28]. Therefore, the development of a cheap transition metal-based photoelectrode, with broad spectrum response, is crucial for cost-effective urea wastewater treatment and clean energy conversion via photoelectrocatalytic technology.
It is well known that the efficiency of urea oxidation depends on the rate of the anodic urea oxidation reaction [11,29,30]. However, UOR is a kinetically slow process which involves 6e-transfer [31,32]. Therefore, using a high-performance photoelectrocatalyst to reduce the UOR overpotential is the key to accelerate the kinetics. Nickel-based catalysts have been extensively developed for urea electrolysis, since they can effectively reduce the active potential of UOR [3,[32][33][34]. Among nickel-based catalysts, perovskite oxides are the most attractive, due to their adjustable composition, as well as excellent light absorption ability [5,[35][36][37][38][39]. Perovskite oxides are generally prepared under high temperature conditions, resulting in less active sites on the surface and low activity [40][41][42][43]. Thus, improving the active sites on the surface of nickel-based perovskite oxides is very important to enhance the performance of solar-driven urea oxidation, and realize the treatment and energy conversion of urea wastewater.
Herein, we employed rGO as a template to successfully prepare a two-dimensional porous Ruddlensden-Popper La 2 NiO 4 perovskite with abundant active sites on the surface for photoelectrocatalytic urea oxidation [44]. The photocurrent densities of the obtained La 2 NiO 4 photoanode, at a potential of 1.47 V vs. RHE, were 2.94, 2.50 and 2.00 mA cm −2 under UV-vis-NIR, visible, and NIR light irradiation, respectively. Moreover, the corresponding photocatalytic urea fuel cells (PUFCs) were adopted to show the great potential of the two-dimensional porous La 2 NiO 4 for practical urea wastewater treatment. Under NIR irradiation, both the urea solution and real human urine showed a large filling factor (FF) of 0.477 and 0.411 with maximum power densities (Pmax) of 0.129 µW cm −2 and 0.041 µW cm −2 , respectively. The results demonstrated that the La 2 NiO 4 photoanode showed full spectral response to efficiently oxidize urea solution.

Preparation and Characterizations of Catalysts
The holey two-dimensional La 2 NiO 4 nanosheets were fabricated using rGO as a sacrificial template through a general two-step strategy, as schematically illustrated in Figure 1. The starting La 2 NiO 4 precursor, grown on rGO, was obtained through the chemical impregnation treatment of the starting materials [44]. The X-ray diffraction (XRD) pattern of the La 2 NiO 4 precursor/rGO revealed an amorphous structure, since no diffraction peaks were found ( Figure S1). Subsequently, we annealed the La 2 NiO 4 precursor/rGO under air at 400 • C to remove the rGO template, and, further, at 800 • C to give pure La 2 NiO 4 ( Figure 1). As shown in Figure 2a, both a Ruddlensden-Popper La 2 NiO 4 phase (PDF#34-0314) and a small amount of La 2 O 3 (PDF#05-0602) were observed. The general formula for the Ruddlesden-Popper phase is described as A n+1 B n O 3n+1 or A 2 'A n-1 B n O 3n+1 , which is constructed of a LaNiO 3 perovskite unit and a LaO rock salt unit (top right inset of Figure 2a) [44]. The morphology was of two-dimensional nanosheets with a holey feature inherited from rGO (Figure 2b-e and Figure S2). The diffused concentric rings in the selected area electron diffraction (SAED) pattern (inset of Figure 2d) indicated a polycrystalline feature. The clear lattice fringes were determined to be 0.273 nm, which corresponded well to the (110) facet of the tetragonal La 2 NiO 4 ( Figure 2e). The SEM-EDS mapping images revealed that the obtained La 2 NiO 4 nanosheets were principally composed of La, Ni, and O, with a uniform spatial distribution (Figure 2c). The Brunauer-Emmett-Teller (BET) surface area of the holey La 2 NiO 4 nanosheets (14.0 m 2 g −1 ) was much higher than that of the sample obtained without the rGO template (5.6 m 2 g −1 ) ( Figure S3). These results demonstrated that the application of rGO as a template could dramatically increase the BET surface area of La 2 NiO 4 , and, thus, increase the reactive sites. The holey two-dimensional nanosheet structure could also enhance light reflection and improve the reactivity of UOR ( Figure S4). holey feature inherited from rGO (Figures 2b-e and S2). The diffused concentric rings in the selected area electron diffraction (SAED) pattern (inset of Figure 2d) indicated a polycrystalline feature. The clear lattice fringes were determined to be 0.273 nm, which corresponded well to the (110) facet of the tetragonal La2NiO4 (Figure 2e). The SEM-EDS mapping images revealed that the obtained La2NiO4 nanosheets were principally composed of La, Ni, and O, with a uniform spatial distribution (Figure 2c). The Brunauer-Emmett-Teller (BET) surface area of the holey La2NiO4 nanosheets (14.0 m 2 g −1 ) was much higher than that of the sample obtained without the rGO template (5.6 m 2 g −1 ) ( Figure S3). These results demonstrated that the application of rGO as a template could dramatically increase the BET surface area of La2NiO4, and, thus, increase the reactive sites. The holey two-dimensional nanosheet structure could also enhance light reflection and improve the reactivity of UOR ( Figure S4).  holey feature inherited from rGO (Figures 2b-e and S2). The diffused concentric rings in the selected area electron diffraction (SAED) pattern (inset of Figure 2d) indicated a polycrystalline feature. The clear lattice fringes were determined to be 0.273 nm, which corresponded well to the (110) facet of the tetragonal La2NiO4 (Figure 2e). The SEM-EDS mapping images revealed that the obtained La2NiO4 nanosheets were principally composed of La, Ni, and O, with a uniform spatial distribution (Figure 2c). The Brunauer-Emmett-Teller (BET) surface area of the holey La2NiO4 nanosheets (14.0 m 2 g −1 ) was much higher than that of the sample obtained without the rGO template (5.6 m 2 g −1 ) ( Figure S3). These results demonstrated that the application of rGO as a template could dramatically increase the BET surface area of La2NiO4, and, thus, increase the reactive sites. The holey two-dimensional nanosheet structure could also enhance light reflection and improve the reactivity of UOR ( Figure S4).   The UV-vis-NIR diffuse reflectance spectra showed that La 2 NiO 4 exhibited broad spectral absorption that covered near-infrared (NIR) light (λ = 200-1200 nm) (Figure 2f), which indicated that the material had potential photocatalysis activities under NIR light irradiation. The absorbance decreased with increasing of the annealing temperature, which might be caused by the reduction of light reflection, due to the reduction of pores ( Figure S5a). Typical diffraction peaks of perovskite oxide La 2 NiO 4 , with a spinel structure (PDF#34-0314), and a small amount of La 2 O 3 (PDF#05-0602) were observed for the samples annealed at 700 • C, 800 • C and 900 • C. However, the diffraction peak of the spinel structure was obviously enhanced and became sharper, and no peak of impurities was found when the annealing temperature was 1000 • C, indicating that the sample obtained was composed of pure La2NiO4 with a perfect crystalline phase ( Figure S5b). In the conventional route of synthesis of La 2 NiO 4 , similar results were obtained for its composition and crystalline transformation as the annealing temperature increased [38,45]. The BET surface area and pore diameter decreased with increase in the annealing temperature ( Figure S5c,d). The SEM images showed that sintering occurred on the surface with particle agglomeration when the annealing temperature increased ( Figure S6). We then evaluated the UOR activities of La 2 NiO 4 through a three-electrode configuration in 1 M KOH with 0.5 M urea under NIR light irradiation. La2NiO4 annealed at 800 • C showed the best UOR activity ( Figure S7), due to its large surface area and lack of sintering caused by high temperature calcination.
Since catalytic activities are commonly sensitive to valence states and the coordination environment of metal centers, X-ray photoelectron spectroscopy (XPS) was carried out to explore the chemical states of La, Ni and O in La 2 NiO 4 . As shown in Figure S8, the survey XPS spectrum clearly confirmed elemental composition of La 2 NiO 4 . As shown by the O 1s regions in Figure 2g, there were three peaks of La 2 NiO 4 that could be clearly identified, among which the peaks of 528.5 and 531.2 eV could be ascribed to metal oxygen bonds and surface-adsorbed water molecules, respectively. The peak located at 529.7 eV could be attributed to O atoms in the vicinity of an O vacancy. Figure 2h shows the Ni 2p spectra of La 2 NiO 4 , wherein the peak values of 851.6 eV and 854.7 eV in La 2 NiO 4 corresponded to the Ni 2+ 2p 3/2 and Ni 3+ 2p 3/2 , indicating the coexistence of Ni 2+ and Ni 3+ [36]. Nevertheless, due to the overlap of Ni 2p 3/2 and La 3d 3/2 , the ratio of Ni 2+ and Ni 3+ was still hard to calculate. However, the intensity of Ni 3+ in La 2 NiO 4 was higher compared to that of Ni 2+ , indicating that the prepared Ruddlensden-Popper La 2 NiO 4 would introduce partial oxygen vacancies, and more Ni 2+ would be converted into Ni 3+ . The Ni 3+ was the active site that affected UOR activity [5], and, thus, would enhance the activity of UOR. For the La 3d (Figure 2i), there was no obvious peak position shift.

Photoelectrochemical Analysis of Catalysts
Photoelectrochemical measurements were conducted to evaluate the photoelectric properties of La 2 NiO 4 . In order to eliminate the current change caused by temperature, the photocurrent test under NIR light irradiation was carried out under low-power density light (100 mW cm −2 ) and in ice water bath ( Figure S9). The results showed that NIR light could excite electrons without temperature inducing current change. The photocurrent density of the obtained La 2 NiO 4 was much higher than that of the sample prepared without the rGO template, indicating that the holey nanosheet structure of La 2 NiO 4 inherited from the rGO template was conducive to full contact between the electrode and the electrolyte and could enhance the absorption of light ( Figure S4a). The active surface areas were further evaluated by electrochemical double-layer capacitance (Cdl) measurements ( Figure S10). The Cdl value of La 2 NiO 4 electrode material prepared with the rGO template was calculated to be 2.40 mF cm −2 , which was much higher than that of the electrode material prepared without the rGO template (Cdl = 0.28 mF cm −2 ). The higher Cdl value of La 2 NiO 4 suggested a larger electrochemical active surface area of the porous and two-dimensional structure inherited from the rGO template. On the basis of the above results, the obtained La 2 NiO 4 was tested to be a superior catalyst for oxidation of urea.

UOR Activity
Given that the thermodynamic equilibrium potential shift from the 1.23 V vs. RHE of OER to the 0.37 V vs. RHE of the UOR favors the mass-production of hydrogen with a theoretical energy saving of~70%, replacing the dreaded OER with UOR can result in energy efficient H 2 production [12]. We then investigated the photoelectrocatalytic activities of La 2 NiO 4 in 1 M KOH solution in the absence (OER) and presence of 0.5 M urea (UOR) by linear sweep voltammetry (LSV). As shown in Figure 3a, the remarkable peaks at about 1.42 V vs. RHE for the OER curves could be ascribed to the redox pair of Ni 2+ /Ni 3+ . A redox pair of symmetrical Ni 2+ /Ni 3+ was also found in the cyclic voltammetry (CV) curves ( Figure S11) [7,31]. Meanwhile, the LSV curve showed that the UOR onset potential for La 2 NiO 4 was~1.35 V vs. RHE. In comparison to the OER curve, recorded in the KOH electrolyte, the anodic current density of La 2 NiO 4 increased rapidly after adding urea, indicating that La 2 NiO 4 exhibited a high catalytic response activity for oxidation of urea. The NIR light irradiation could obviously accelerate the current density of UOR, with the application potential of 1.47 V vs. RHE current density of 2.00 mA cm −2 under the NIR light irradiation, while the promotion effect on the OER curve was small, with a current density of 0.38 mA cm −2 under NIR light and in the dark (Figure 3a). The photocurrent responses of the La 2 NiO 4 catalysts were also evaluated in 1 M KOH solution for OER and UOR processes under NIR light irradiation (λ > 800 nm). As shown in Figure 3c, the photocurrent intensity of La 2 NiO 4 for the UOR process was around 50 µA cm −2 , which was much higher than that of the OER process (25 µA cm −2 ). Furthermore, the UOR and OER kinetics of La 2 NiO 4 were investigated by the corresponding Tafel plots (Figure 3b). La 2 NiO 4 had a small Tafel slope of 79 mV dec −1 for the UOR process. This value was substantially lower than that of the 165 mV dec −1 needed for the OER process, demonstrating more beneficial catalytic kinetics of UOR than OER. Further studies of the La 2 NiO 4 electrode for UOR and OER kinetics by electrochemical impedance spectroscopy (EIS) were conducted. As shown in Figure 3d, the UOR process also had lower charge transfer resistance (Rct) values compared to OER, due to the increased rate of electron transfer between the electrode interface and the solute. The above results showed that replacing OER with UOR could greatly reduce the overpotential, and using La 2 NiO 4 as the photoanode could save energy to produce hydrogen.
Hydrogen production performances of the cathode were analyzed from the two anodes' half-reactions, UOR and OER, through a three-electrode system at an applied potential of 1.47 V vs. RHE. The current density was 0.96 mA cm −2 for the UOR process, while only 0.08 mA cm −2 for the OER process ( Figure S12a). It was remarkable that the current densities of the two half-reactions were very stable, which meant that the La 2 NiO 4 catalyst showed excellent stability. The hydrogen production rates of the UOR and OER process were 9.75 and 0.55 µmol cm −2 h −1 , respectively ( Figure S12b). We integrated this electrolyzer with a commercial silicon solar cell to construct a solar-driven overall urea splitting system ( Figure S13a). The J-V curves of 1 M KOH solution presented a high solar-to-hydrogen efficiency of 14.9% with 0.5 M urea, and 13.3% without urea [46][47][48][49], implying that a mild UOR, instead of the dreaded OER, could reduce the overpotential of hydrogen production at a smaller electrolytic voltage ( Figure S13b). Similar measurements were performed using a Pt/C coated FTO electrode for comparison. The Pt/C coated FTO electrode showed negligible catalytic activity for UOR in the dark, while a significantly enhanced electrocatalytic performance could be observed after La 2 NiO 4 was coated on FTO (Figures S14a and S15). A smaller onset potential, higher current density and smaller Tafel slope of La 2 NiO 4 coating electrode were also observed for La 2 NiO 4 -coated FTO ( Figure S14b). The hydrogen production of Pt/C electrode was negligible, which indicated that La 2 NiO 4 as the photoanode could replace noble metal to achieve efficient hydrogen production through UOR [3]. Catalysts 2022, 12, x FOR PEER REVIEW 6 of 15 Hydrogen production performances of the cathode were analyzed from the two anodes' half-reactions, UOR and OER, through a three-electrode system at an applied potential of 1.47 V vs. RHE. The current density was 0.96 mA cm −2 for the UOR process, while only 0.08 mA cm −2 for the OER process ( Figure S12a). It was remarkable that the current densities of the two half-reactions were very stable, which meant that the La2NiO4 catalyst showed excellent stability. The hydrogen production rates of the UOR and OER process were 9.75 and 0.55 μmol cm −2 h −1 , respectively ( Figure S12b). We integrated this electrolyzer with a commercial silicon solar cell to construct a solar-driven overall urea splitting system ( Figure S13a). The J-V curves of 1 M KOH solution presented a high solarto-hydrogen efficiency of 14.9% with 0.5 M urea, and 13.3% without urea [46][47][48][49], implying that a mild UOR, instead of the dreaded OER, could reduce the overpotential of hydrogen production at a smaller electrolytic voltage ( Figure S13b). Similar measurements were performed using a Pt/C coated FTO electrode for comparison. The Pt/C coated FTO electrode showed negligible catalytic activity for UOR in the dark, while a significantly enhanced electrocatalytic performance could be observed after La2NiO4 was coated on FTO ( Figures S14a and S15). A smaller onset potential, higher current density and smaller Tafel slope of La2NiO4 coating electrode were also observed for La2NiO4coated FTO ( Figure S14b). The hydrogen production of Pt/C electrode was negligible, which indicated that La2NiO4 as the photoanode could replace noble metal to achieve efficient hydrogen production through UOR [3].
The remarkable catalytic abilities observed above for the UOR under NIR light irradiation prompted us to explore the activities of La2NiO4 as the electrode material under full spectrum light irradiation (Figure 4a), hence moving a step closer to practical The remarkable catalytic abilities observed above for the UOR under NIR light irradiation prompted us to explore the activities of La 2 NiO 4 as the electrode material under full spectrum light irradiation (Figure 4a), hence moving a step closer to practical applications. As shown in Figure 4b, the LSV curves of the La 2 NiO 4 electrode showed the same onset potential under different light source irradiations (UV-vis-NIR, visible, and NIR) in 1 M KOH with 0.5 M urea, and the current density under UV-vis-NIR irradiation was the highest. Furthermore, the UOR kinetics of La 2 NiO 4 were investigated through the corresponding Tafel plots and electrochemical impedance spectroscopy (EIS). The Tafel plot and charge transfer resistance (Rct) under UV-vis-NIR irradiation were comparatively lower than those under visible and NIR light (Figure 4c and Figure S16). Moreover, the photocurrent response plots ( Figure S17) suggested that the photoanode excited more electrons for HER under UV-vis-NIR irradiation. The H 2 generation rate and ∆TOC of urea degradation under different light source irradiations were then recorded ( Figure S18). Both the H 2 generation rate and ∆TOC of urea degradation were highest under UV-vis-NIR irradiation, compared to another light source irradiation, at 22.76 µmol cm −2 h −1 and 1750 mg/L, respectively. The material stability was also important to estimate the long-term UOR activity of the photoelectrocatalyst. To assess the stability of the La 2 NiO 4 electrode, chronoamperometry (J-t) measurements were performed at an applied voltage of 1.47 V vs. RHE. As shown in Figure 4d, superior catalytic stabilities, with negligible changes in current density, were maintained after 8h testing under various light irradiations, indicating the excellent durability of La 2 NiO 4 for the UOR process. 22.76 μmol cm −2 h −1 and 1750 mg/L, respectively. The material stability was also important to estimate the long-term UOR activity of the photoelectrocatalyst. To assess the stability of the La2NiO4 electrode, chronoamperometry (J-t) measurements were performed at an applied voltage of 1.47 V vs. RHE. As shown in Figure 4d, superior catalytic stabilities, with negligible changes in current density, were maintained after 8h testing under various light irradiations, indicating the excellent durability of La2NiO4 for the UOR process. Direct urea fuel cell (DUFC), as a device to produce electricity and purify urea/urinerich wastewater at the same time, is of great significance for the integration of energy conversion and environmental treatment [9,11]. We evaluated the performance of a pho- tocatalytic urea fuel cell in a two-electrode system. The process of continuous energy generation from PUFC is shown in Figure 4e. The La 2 NiO 4 photoanode generated electron hole pairs under exposure to light. The holes moved to the surface of La 2 NiO 4 to oxidize Ni 2+ to Ni 3+ , the active Ni 3+ species then oxidized urea and was chemically reduced back to inactive Ni 2+ [4]. The electrons that were transferred to external circuits providing a sustainable output for oxygen reduction reaction (ORR) at the cathode. The whole process is described as following: Photoanode: Cathode: Overall: In order to further study the photocatalytic performances, the polarization curves of the PUFC under various light irradiations were collected (Figure 4f). It can be clearly observed that the PUFC showed the highest open-circuit voltage of 0.32 V and short-circuit current density of 9.00 µA cm −2 under UV-vis-NIR irradiation, which displayed a higher energy compared to other light sources (Table S1). The same trends were found in the polarization and power curves (Figure 4f,g), with the maximum photovoltage response and output power density of 0.575 µW cm −2 under the UV-vis-NIR light irradiation. Filling factor (FF) is commonly used to evaluate the performance of photocatalytic fuel cells, the FF value under NIR irradiation was 0.477 (Table S1), which was significantly larger than those of the other light sources [50]. The experimental results showed that the La 2 NiO 4 photoanode could process urea round-the-clock, which meant that it could not only efficiently convert urea into electricity in the daytime, but also continue the process under weak NIR light irradiation at night. The photoelectrochemical properties of the designed PUFC in urea solution with different concentrations were also investigated. Under NIR light irradiation, the current density, corresponding to the UOR polarization curves at a potential of 1.42 V vs. RHE, increased with increase of urea concentration, ranging from 0.1 to 0.5 M ( Figure S19a). However, the relationship between the current density and urea concentration was nonlinear ( Figure S19b), implying that excessive urea could not be oxidized in time by the La 2 NiO 4 photoanode [34]. The polarization curves also exhibited a similar trend under urea concentration ranging from 0.1 to 0.5 M, and the power curves demonstrated the corresponding output power density ( Figure S20), and the short circuit current and maximum power density decreased with increase of urea concentration. Note that the UOR polarization curves ( Figure S21) in urine were similar to the onset potential in urea solution, but the current density was reduced by an order of magnitude due to the lack of electrolytes. To further evaluate the PUFC performance of urine, the polarization curves ( Figure S20k,i) showed open circuit voltage similar to that in urea solution under different light sources, which was still due to the influence of electrolytes, leading to a greatly reduced current density. However, the FF value in urine maintained a relatively high value of 0.411 under NIR light irradiation (Table S2), indicating the potential of using urine as fuel directly for PUFCs [10]. The durability of the PUFC was evaluated in 1 M KOH with urea and urine at 0.1 V output voltages, respectively (Figure 4h). The PUFC exhibited excellent stability within 240 min, without any major loss in performance in the urea solution. However, the current density in urine decreased sharply in the first 120 min and then tended to stabilize, possibly due to the interference of the complex components in urine [51].

UOR Activity
The mechanism of La 2 NiO 4 's remarkable catalytic performance was also investigated by the adsorption and activation of urea on the surface of the La 2 NiO 4 catalyst ( Figure 5). As shown in Figure 5a,b, the adsorption energy of urea on La 2 NiO 4 was −2.185 eV, which indicated that the unique nanosheet structure of La 2 NiO 4 significantly enhanced the adsorption of urea on the catalyst surface. According to the electron localization function (ELF), there was a strong interaction between the N atom in urea and the La atom in La 2 NiO 4 (Figure 5c). Bader charge was analyzed to illustrate the electron transfer during the activation of the urea molecule. Urea transferred 0.31 e to La 2 NiO 4 , indicating that urea had a strong electron transfer effect with the catalyst surface. The theoretical results aligned well with our experimental evidence in the fact that the La 2 NiO 4 could efficiently adsorb and activate urea on the surface, resulting in excellent UOR activity.
value in urine maintained a relatively high value of 0.411 under NIR light irradiation (Table S2), indicating the potential of using urine as fuel directly for PUFCs [10]. The durability of the PUFC was evaluated in 1 M KOH with urea and urine at 0.1 V output voltages, respectively (Figure 4h). The PUFC exhibited excellent stability within 240 min, without any major loss in performance in the urea solution. However, the current density in urine decreased sharply in the first 120 min and then tended to stabilize, possibly due to the interference of the complex components in urine [51].

UOR Activity
The mechanism of La2NiO4's remarkable catalytic performance was also investigated by the adsorption and activation of urea on the surface of the La2NiO4 catalyst ( Figure 5). As shown in Figure 5a,b, the adsorption energy of urea on La2NiO4 was −2.185 eV, which indicated that the unique nanosheet structure of La2NiO4 significantly enhanced the adsorption of urea on the catalyst surface. According to the electron localization function (ELF), there was a strong interaction between the N atom in urea and the La atom in La2NiO4 (Figure 5c). Bader charge was analyzed to illustrate the electron transfer during the activation of the urea molecule. Urea transferred 0.31 e to La2NiO4, indicating that urea had a strong electron transfer effect with the catalyst surface. The theoretical results aligned well with our experimental evidence in the fact that the La2NiO4 could efficiently adsorb and activate urea on the surface, resulting in excellent UOR activity.

Synthesis of La 2 NiO 4
The La 2 NiO 4 -precursor was deposited on the rGO by a simple chemical impregnation method. The La(NO 3 ) 3 •6H 2 O (0.5 mmol), Ni(NO 3 ) 2 •6H 2 O (0.25 mmol) and citric acid (2 mmol) were dissolved in 10 mL of deionized water and 20 mL ethanol, and then transferred into a 100 mL glass beaker. Subsequently, a certain amount of rGO was added into the mixture. After vigorous stirring and ultrasonic for 15 min, the mixture was maintained at 70 • C for 12 h under constant stirring. Excess NH 4 HCO 3 (2 g) was then added into the above mixture, and it placed in a 50 • C water bath, heated and dried to obtain the La 2 NiO 4 precursor. The La 2 NiO 4 precursor/rGO was treated at 400 • C for 2 h (heating rate: 4 • C min −1 ) and 800 • C for 2 h (heating rate: 10 • C min −1 ) under air atmosphere to obtain La 2 NiO 4 . All reagents were purchased from Aladdin (Shanghai, China) and were not further purified.

Characterization
Powder XRD patterns were obtained with Rigaku Dmax-3C Advance (Rigaku, Tokyo, Japan) X-ray diffractometer using Cu Ka radiation at a wavelength of 1.5406 Å. Field emission scanning electron microscopy (FESEM) was conducted using a Hitachi S-4800 SEM (Hitachi, Tokyo, Japan). Transmission electron microscopy (TEM) images and selected area electron diffraction (SAED) images were collected by using a JEOL 2010 TEM (JEOL, Tokyo, Japan). The XPS was performed with Perkin-Elmer PHI 5000 (Perkin-Elmer, Waltham, MA, USA) X-ray photoelectron spectroscopy. All of the values of binding energy were calibrated using C 1s = 284.8 eV as a reference. UV-vis diffuse reflectance spectra (UV-vis DRS) was obtained with an UV-2450 instrument (Shimazu, Tokyo, Japan). The N 2 adsorptiondesorption isotherms were collected by a TriStar II 3020 (micromeritics, Norcross, GA, USA) at 77 K, and all samples degassed at 423 K and 10 −6 Torr for 6 h. The Brunauer-Emmett-Teller (BET) approach was used to calculate the surface area. The zeta potential was measured on a Zetasizer Nano ZS90 (Malvern, UK).

Photoelectrochemical Measurement
All the photoelectrochemical measurements were performed on a CHI660E (Chenhua, Shanghai, China) electrochemical workstation using a three-electrode system. FTO-coated glass substrates (2 × 3 cm 2 ) were cleaned sequentially with acetone, deionized water, and ethanol via sonication for 15min. Subsequently, samples were loaded on the FTO through the doctor blade method, and the as-coated films were air-annealed at 200 • C for 2 h to form the working electrode. The mass loading of samples on FTO was about 1 mg cm −2 . Hg/HgO (1 M KOH) and a commercial platinum foil (99.99%, 4 cm 2 ) were used as the reference and counter electrodes, respectively. An airtight single cell was used as the reactor for PEC measurement. The electrolyte was purged with nitrogen (99.999%) to remove dissolved O 2 before testing. A 300 W xenon lamp was used as the UV-visible-NIR light (AM 1.5G, 1100 nm > λ > 300 nm), visible light (780 nm > λ > 420 nm) and near infrared light (λ > 800 nm) sources by using filters. The photocurrent density was detected in an aqueous solution containing 1 mol/L of KOH, 1 mol/L of KOH with 0.

Photoelectrocatalytic Activity Test
An H-cell and the electrolyte of 1 M KOH and 0.5 M urea were used for the UOR and HER tests. The H-cell was separated by an anion exchange membrane (AMI-7001). The cathode chamber electrolyte was purged with nitrogen (99.999%) to remove dissolved O 2 before testing. During the photoelectrocatalytic process, a constant potential (1.47 V vs. RHE) was applied on the working electrode. A 300 W xenon lamp was used as the UV-visible-NIR light (AM 1.5 G, 1100 nm > λ > 300 nm), visible light (780 nm > λ > 420 nm) and near infrared light (λ > 800 nm) sources. UOR and HER under UV visible light, visible light, near-infrared light and dark conditions were tested for 8 h, and the change of photocurrent within 8 h was recorded. An amount of 0.5 mL of sampled gas was used to analyze the hydrogen production rate using a gas chromatograph (GC9800 (N), Shanghai Ke Chuang Chromatograph Instruments Co. Ltd., Shanghai, China, TCD, with nitrogen as a carrier gas and a 5 A molecular sieve columns). The urea degradation rate was measured using Total organic carbon (TOC) content (vario TOC cube, elementar), and all solutions were diluted 10,000 times to obtain ∆TOC (urea) before testing.

Photocatalytic Fuel Cell Test
The PUFC performance was tested in a single cell with a two-electrode system using a CHI660E electrochemical workstation. The catalyst-coated FTO was used as the working electrode, and the Pt foil was used as the auxiliary and reference electrode. The distance between the two electrodes was fixed at approximately 4 cm. A 300 W xenon lamp was used as the UV-visible-NIR light (AM 1.5 G, 1100 nm > λ > 300 nm), visible light (780 nm > λ > 420 nm) and near infrared light (λ > 800 nm) sources. An aqueous solution of 1 mol/L KOH, 1 mol/L KOH with 0.5 mol/L urea or healthy people's fresh urine were used as the electrolyte, respectively. The change curve of open circuit voltage under different light sources was measured (Voc-t plots). The current-voltage (J-V) plots were collected to characterize the PFC performance with a scan rate of 1 mV/s. Each J-V plot was collected when the open circuit voltage was stabilized. A discharge test was conducted at 0.1 V for 4h, and the urea degradation rate was measured to obtain ∆TOC (urea).

Computational Methods
All the spin theoretical simulations in our work were carried out on the Vienna ab initio Simulation Package (VASP) with version 5.4.1. The Generalized gradient approximation (GGA) with the Perdew-Burke-Emzerhof (PBE) functional form was employed to evaluate the electron-electron exchange and correlation interactions while the projector augmentedwave (PAW) methods were implanted to represent the core-electron (valence electron) interactions. The GGA + U calculation was performed with the on-site Coulomb Repulsion U term on the Ni 3d and the Ueff (Ueff = U-J) value is 6.3 eV. Plane-Wave basis function was set with a kinetic cut-off energy of 550 eV. The ground-state atomic geometries were optimized by relaxing the force below 0.02 eV/Å and the convergence criteria for energy was set with the value of 1.0 × 10 −5 eV/cell. A Monkhorst-Pack mesh, with the size of 3 × 3 × 1, was employed to sample the surface Brillouin zone. The Gaussian method was employed for the both electronic structures and total energy of our models and stress/force relaxations. In order to simulate the interactions between molecules, van der Waal (vdw) interactions were included described by the DFT-D3 method of Grimme.

Conclusions
In summary, a two-dimensional porous Ruddlensden-Popper type nickel-based perovskite oxide La 2 NiO 4 was prepared using rGO as a template. Subsequent studies on UOR activity showed that the La 2 NiO 4 photoanode exhibited superior UOR performance under full spectrum solar irradiation, and, especially, the transient photocurrent could reach 50 µA cm −2 in urea solution under NIR light irradiation. The La 2 NiO 4 photoanode onset potential of UOR was only~1.35 V vs. RHE, which greatly increased the hydrogen production efficiency, compared to OER, which was attributed to the partial conversion of Ni 2+ to Ni 3+ caused by the oxygen deficiency in La 2 NiO 4 . The PUFC, constructed with La 2 NiO 4 as the photoanode, also showed excellent UOR activity and effective energy conversion under full spectrum solar irradiation. Moreover, the filling factor (FF) under NIR light was 0.477. This work shows that urea electrolyzers and photocatalytic fuel cells constructed from low-cost nickel-based perovskite oxides have potential for round-the-clock low energy hydrogen production and power generation by efficient utilization of solar energy.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/catal13010053/s1, Figure S1: The XRD pattern of the La 2 NiO 4 precursor/rGO; Figure S2: SEM of La 2 NiO 4 with rGO (a) and without rGO (b); Figure S3: (a) Nitrogen adsorption-desorption isotherms and (b) pore size distribution of La 2 NiO 4 with and without rGO added; Figure S4: (a) The UV-vis-NIR diffuse reflectance spectra and (b) UOR activity under NIR light of La 2 NiO 4 with and without rGO; Figure S5: (a) The XRD pattern (b) The UV-vis-NIR diffuse reflectance spectra (c) Nitrogen adsorption-desorption isotherms and (d) pore size distribution of La 2 NiO 4 annealed at 700 • C, 800 • C, 900 • C and 1000 • C, respectively; Figure Table S1: Performance parameters of PUFCs in 1 M KOH with 0.5 M urea exposed to different light sources irradiation; Table S2: Performance parameters of PUFCs under urea concentration ranging from 0.1 to 0.5 M and human urine exposed to NIR light irradiation.