Revealing the In Situ Evolution of Tetrahedral NiMoO4 Micropillar Array for Energy‐Efficient Alkaline Hydrogen Production Assisted by Urea Electrolysis

The great promise of the combination of urea oxidation reaction (UOR) with hydrogen evolution reaction (HER) to simultaneously achieve wastewater treatment and hydrogen production calls for the rational design of high‐performance electrocatalysts. Herein, the reconstruction with the formation of Ni2P and Mo2O72− on the surface can largely enhance the alkaline HER activity of P‐NiMoO4 by on‐site electrochemical activation strategy. Systematic experimental results indicate that the reconstruction process enables the exposure of additional Ni sites and the adjustment of hydrogen adsorption to facilitate HER kinetics. Ultimately, a highly efficient alkaline HER electrode with low overpotential of −48.9 mV for 10 mA cm−2 is achieved. More importantly, a UOR electrolyzer assembled with A‐P‐NiMoO4 as the cathode and P‐NiMoO4 as the anode exhibits impressive performance with a small cell voltage of 1.363 V for 10 mA cm−2. The application of the fabricated electrodes in a practical cell driven by a commercial dry battery (1.5 V) demonstrates efficient and stable hydrogen production, making the developed strategy promising for the rational design of highly active electrocatalysts for green hydrogen production.


Introduction
Increasing energy demand along with the severe environmental crisis poses a great challenge for global sustainable development. [1]Compared to conventional carbon-based hydrogen production methods (e.g., natural gas, coal, coke oven gas to hydrogen, etc.), hydrogen production via electrochemical water splitting powered by renewable energies represents a promising energy conversion technology because of the high energy density of hydrogen, environmental friendliness, and process simplicity. [2]A large range of highly active electrocatalysts such as nitrides, [3] phosphides, [4] carbides, [5] and sulfur generics [6] have been developed to reduce the overpotentials accompanied with the energy loss for water splitting.However, the sluggish kinetics of oxygen evolution reaction (OER) with the formation of O═O double bonds makes it still the bottleneck for efficient hydrogen generation from water splitting. [7]Recently, the alternative reactions to OER have attracted great attention due to their potential to convert small molecules into high-value-added products or alleviate wastewater pollution dilemmas, as well as their high selectivity and conversion efficiency. [8]In particular, the large emission of urea from industrial and household effluents without appropriate and adequate treatment can cause severe environmental contamination and eutrophication of water resources. [9]Thus, the coupling of UOR, occurring at a much lower theoretical voltage of 0.37 V versus reversible hydrogen electrode (RHE) than OER, with HER offers great promise to achieve the sustainable pollutant control and economical hydrogen generation simultaneously. [10]evertheless, the extremely complex six-electron transfer process endows the anodic urea oxidation reaction (UOR) with slow kinetics. [11]nspired by the efficient catalytic conversion of urea to NH 3 and CO 2 by natural urease with nickel active sites, [12] various low-cost nickel-based catalysts have been explored for UOR. [13]nfortunately, previously reported Ni-based UOR catalysts still exhibit unsatisfactory activity due to the unfavorable electronic structure of Ni for the absorption of intermediates such as COO*. [14]Therefore, optimizing the electronic structure of Ni sites is vital for the development of highly efficient UOR electrocatalysts.
On the other hand, platinum-based materials are demonstrated as the most advanced electrocatalysts for HER. [15]owever, their sumptuosity and scarcity limit their large deployment for practical applications. [16]In addition, the alkaline water splitting has been commercialized as a cost-effective and mild method to compete with traditional fossil fuels, [17] while the typical HER electrocatalysts such as CoP [18] and Mo 2 C [19] working under acidic conditions display limited performance in alkaline solution.Therefore, the formation of adsorbed hydrogen (Volmer step) from the electronically coupled water dissociation becomes a key factor for efficient alkaline water splitting. [20]his necessitates that the designed catalysts possess multiple active sites with individually optimized adsorption energy (ΔG*), conferring the interface engineering strategies a unique way to design advanced electrocatalysts. [21]ecently, the dynamic reconstructions of the electrode surface usually occurring under practical operating conditions spur great research interest, which generally induces the formation of new phases with abundant interfaces. [22]For example, transition metal-based materials under the high oxidation potential of the OER process will form metal hydroxyl/oxides as one of the most efficient OER catalysts. [23]Compared with the efforts on OER electrocatalysts, few reports of self-reconstruction processes during HER have been revealed. [24]Additionally, the mass and charge transfer at the catalyst-electrolyte interface is highly susceptible to interfacial transformation and degradation. [25]lthough the activation and self-reconstruction of catalysts allow to tune the crystal structure and engineer the interface for optimizing the catalytic activity, this change imposes a great challenge to ensure the long-term durability, simultaneously. [26]herefore, the study of dynamic reconstructions is crucial to identify the true active species and understand the electrocatalytic mechanism for rationally designing advanced catalysts for green hydrogen production in an economical way. [27]erein, we report the investigation on phosphorus-doped nickel molybdate for HER and UOR, subtly revealing the activation process during alkaline HER.The results show that phosphorus doping produces β-NiMoO 4 with an active electronrich state that promotes the kinetics of water dissociation, leading to a shift in the HER rate-limiting step from the Volmer step to the Heyrovsky step.Most importantly, the phosphate in the precatalyst is reduced in situ to the metal phosphide Ni 2 P under an applied reduction voltage, which subsequently leads to the formation of a heterogeneous interface with β-NiMoO 4 .During this process, the dissolution of Mo allows for the exposure of additional Ni sites and the redeposition of Mo species as the dimer Mo 2 O 7 2À allows for the adjustment of the free energy of hydrogen adsorption to facilitate HER kinetics.The phosphorus doping and activation process successively reconstruct the electronic structure and accelerate charge transfer, facilitating the adsorption of reactant and lowering the energy barrier for water dissociation, thus increasing HER activity.Ultimately, the activated A-P-NiMoO 4 electrode exhibited excellent performance for alkaline HER with a low overpotential of À48.9 mV to achieve a current density of 10 mA cm À2 .On the other hand, the as-fabricated P-NiMoO 4 electrode exhibited an outstanding performance for UOR with only 0.978 V overpotential required to drive a current density of 100 mA cm À2 .The UOR electrolyzer assembled with A-P-NiMoO 4 as the cathode and P-NiMoO 4 as the anode displayed impressive performance with 1.363 V to deliver a current density of 10 mA cm À2 and could be efficiently driven by a commercial dry battery (1.5 V) to generate plentiful gas bubbles.
Furthermore, the surface composition and element valence state were investigated by X-ray photoelectron spectroscopy (XPS) (Figure 1c and S3, Supporting Information) and the survey spectrum confirmed the existence of Ni, Mo, O, and P elements in P-NiMoO 4 .In particular, the distinct peak at %134 eV indicates the successful doping of phosphorus elements into NiMoO 4 rather than the formation of phosphides. [31]For Ni 2p, the peaks at 874.4 and 857.1 eV can be ascribed to Ni 2þ 2p 1/2 and Ni 2þ 2p 3/2 , respectively.Compared to NiMoO 4 , the binding energy of Ni 2p is positively shifted due to P doping, which is attributed to the higher electronegativity of P compared to Ni, thus reducing the surrounding electron concentration of Ni 2þ . [32]In the region of Mo 3d, the two peaks of NiMoO 4 at 232.4 and 235.5 eV can be well indexed to Mo 6þ , in good agreement with previous reports. [33]In stark contrast, the peaks in P-NiMoO 4 can be deconvoluted into three components of Mo 4þ , Mo 5þ , and Mo 6þ with the peaks of Mo 3d 5/2 at 229.1, 230.3, and 232.3 eV, respectively (Figure S3c, Supporting Information). [34]As for O 1s, the peak was apparently shifted positively after phosphorylation.By fitting the signal, a new dominated peak at 531.2 eV assigned to oxygen vacancies (O v ) was observed due to the formation of PO 4 3À with more negative charges than MoO 4 2À , [35] which has to be balanced by the formation of oxygen vacancies, also explaining the formation of low-valence Mo (Mo 4þ , Mo 5þ ).

HER Electrochemical Performance and Activation Mechanism Analysis
The HER electrochemical performance was evaluated in a threeelectrode cell using 1 M KOH solution.Clearly, the performance of P-NiMoO 4 is substantially better than that of NiMoO 4 due to the formation of β-NiMoO 4 with more active electronic states and oxygen vacancies. [29]Interestingly, the P-NiMoO 4 electrode exhibited a progressively increasing HER activity as the linear sweep voltammetry (LSV) measurement continuously progressed (Figure 2a), leading to a significantly reduced overpotential corresponding to 10 mA cm À2 from the initial 92.9 mV to 48.9 mV at the 1000th cycle (Figure S4, Supporting Information) with negligible effect of the counter and reference electrodes (Figure S5, Supporting Information).To rule out the effect of nickel foam substrate, the catalytic activity of the powder samples on a glassy carbon electrode was evaluated and the results consistently confirm the superior intrinsic activity for HER (Figure S6a, Supporting Information).Intriguingly, the activity outperformed the state-of-the-art Pt/C at higher current densities (>100 mA cm À2 ) (Figure 2b).To study this activation process, three typical states (the 1st cycle, the 200th cycle, and the 1000th cycle) were selected and denoted as P-NiMoO 4 , P-NiMoO 4 -200, and P-NiMoO 4 -1000, respectively.In general, the Tafel slope signifies the reaction mechanism of HER at the catalyst surface and the theoretical Tafel slopes of the Volmer, Heyrovsky, and Tafel steps for HER are 120, 40, and 30 mV dec À1 , respectively. [36]As shown in Figure 2c, The Tafel slope plummeted from 164 mV dec À1 for NiMoO 4 to 74 mV dec À1 for P-NiMoO 4 , indicating that the phosphorylation treatment effectively promoted the kinetics of water dissociation and the rate-determining step of HER changed from the Volmer step to the Heyrovsky step.During subsequent electrochemical activation, the Tafel slopes gradually decreased and approached 46 mV dec À1 for P-NiMoO 4 -1000, close to that of Pt/C, demonstrating that the Heyrovsky step was significantly accelerated.Additionally, electrochemical impedance spectroscopy (EIS) was used to study the charge transfer process and the Nyquist plots were simulated with an equivalent circuit with a charge-transfer resistance (R ct ), a solution resistance (R s ), and a constant phase element (Figure 2d). [37]The results revealed that the charge transfer resistance (R ct ) of the electrode decreased significantly (Table S1, Supporting Information), contributing to the enhanced HER catalytic performance.
To understand the enhanced HER performance, the electrochemical surface area (ECSA) of the samples, which is proportional to the double-layer capacitance (C dl ), was evaluated (Figure S7, Supporting Information). [38]Unexpectedly, as the measurement proceeded, the C dl dropped significantly from 256.94 mF cm À2 of P-NiMoO 4 to 112.04 mF cm À2 of P-NiMoO 4 -200 and then slightly increased to 139.41 mF cm À2 for P-NiMoO 4 -1000, indicating that the ECSA was reduced largely during the first 200 LSVs and then approached a steady state after 200th LSV (Figure 2e).This result implies that a large amount of active sites were destroyed or inactivated during the alkaline HER process probably due to the dissolution of Ni and Mo, indicating that a considerable reconstruction occurred on the surface of the electrode.When the measured cathodic current density was normalized by the values of ECSA, the results manifested a significant increase in the intrinsic activity of the catalyst (Figure 2f ), indicative of the optimized electronic structure or newly created active sites.Additionally, the apparent exchange current density ( j 0 ) was calculated (Figure 2g), [33] j 0 increases continuously during the activation process, further indicating that the intrinsic activity of the catalyst increased significantly.The above results indicate that the activation process improved the electron transfer process, accelerated the HER kinetics, and consequently enhanced the intrinsic activity of the electrocatalyst.Promisingly, the HER performance of A-P-NiMoO 4 was superior to most of the reported catalysts (Figure 2h and Table S2, Supporting Information) and the electrocatalytic activity of A-P-NiMoO 4 remained nearly unchanged for at least 120 h (Figure 2i).Additionally, compared with the initial HER activity, the polarization curve after 2000 LSVs exhibited negligible change (the inset in Figure 2i), further demonstrating the outstanding durability as a potential electrocatalyst for commercial applications.
To better understand the enhanced intrinsic HER activity, the electrocatalysts at various stages were investigated.Surprisingly, as shown in Figure 3a, a new peak at 129.3 eV attributed to the P─Ni bonding was observed in the spectrum of P 2p, [39] which indicates that a partial conversion of the phosphate to the phosphide occurred during the electrochemical reduction process.The binding energy of the P─Ni bond is negatively shifted compared to elemental phosphorus (130.0 eV), suggesting that the P in the catalyst has a partial negative charge (δ À ). [40]ditionally, the formation of metal phosphide was further confirmed by the presence of a characteristic peak of the Ni─P bonding at 852.8 eV in the region of Ni 2p (Figure 3b). [39]Consistent with P δÀ , the binding energy of the Ni─P bond was positively shifted compared to that of metallic nickel (852.6 eV), indicating that nickel atoms in the catalyst possess a partial positive charge (δ þ ). [41]Notably, the P─O and Ni─O binding energies were negatively shifted, due to the strong coupling of the in situ-generated electron-rich Ni 2 P with NiMoO 4 , which triggered electronic interactions between the two species. [42]Exceptionally, the binding energy of Mo 3d and O 1s (Figure 3c and S8, Supporting Information) was first negatively shifted (À0.15 eV) and subsequently showed a positive shift (þ0.1 eV) after 200 LSVs, implying that the Mo-based components underwent distinct change during the activation process.Therefore, the variation of the surface composition of the catalyst was analyzed (Figure S9, Supporting Information).The atomic ratio of Mo and Ni decreased from 80.23% for P-NiMoO 4 to 51.03% for P-NiMoO 4 -200, followed by an increase to 65.50% for P-NiMoO 4 -1000.To further study the change of the surface composition of the catalyst, the elemental content of the electrolyte was monitored by inductively coupled plasma-mass spectrometry (ICP-MS).As shown in Figure 3d, all the elements of P, Ni, and Mo in the electrocatalyst were detected in the electrolyte after 200 LSV cycles, and particularly, a larger amount of Mo was dissolved into the electrolyte than Ni and P during the first 200 LSV cycles, explaining the reduced ECSA.However, during the subsequent LSV cycles, the elemental content of Ni and P remained nearly constant, while the elemental content of Mo visibly decreased.These results collectively suggest that Mo first leached out into the electrolyte and then redeposited back to the catalyst surface.
To validate this hypothesis, the structural evolution during the activation of P-NiMoO 4 was examined by Raman spectroscopy (Figure 3e).A set of strong Raman peaks can be observed at 339, 809, 881, and 931 cm À1 in both pre-and postactivation samples, which are consistent with Mo-O vibrations in NiMoO 4 . [43]In contrast, the new peaks at 706 and 190 cm À1 after activation are attributed to the asymmetric extension and bending modes of the Mo─O─Mo bond, respectively, indicating that Mo 2 O 7 2À formed on the surface of the catalyst. [44]Thus, one can speculate that the dissolution of Mo in the form of MoO 4 2À proceeded during the first hundreds of cycles, followed by the redeposition and polymerization of MoO 4 2À to Mo 2 O 7 2À on the catalyst surface.It has been revealed that the presence of Mo 2 O 7 2À on bare Ni and Ni 4 Mo can effectively modulate the hydrogen adsorption on Ni sites due to its lower d-band center. [45]Additionally, the formation of Ni-P can further tune the electronic structure of nickel sites due to the weak stabilization and reduced electron transfer from Ni to P and the P sites are also able to provide moderate bonding to the key intermediates involved during the process of HER. [46]ollectively, the formation of Mo 2 O 7 2À and Ni 2 P optimized the process of hydrogen adsorption/desorption, signified by the reduced Tafel slope from 74 mV dec À1 for P-NiMoO 4 to 46 mV dec À1 for P-NiMoO 4 -1000, thus contributing to the enhanced HER activity. [45]he changes of the morphology and crystal structure were further investigated.As shown in Figure S10, Supporting Information, the XRD patterns after activation remained nearly unchanged with a slight decrease of the β-phase signal, consistent with the preserved micropillar array structure (Figure S11 and S12, Supporting Information).Unexpectedly, no characteristic peaks of metal phosphides were observed probably due to the low content or crystallinity.Therefore, the microstructure of the micropillar array was studied by HRTEM.A more porous and rougher surface was observed for the samples of P-NiMoO 4 -200 and P-NiMoO 4 -1000, which may be caused by the leaching of Mo from the crystal structure.Furthermore, in the HRTEM image of P-NiMoO 4 -1000, three regions were observed (Figure 3f ) and the clear d-spacing of 0.618, 0.335, and 0.219 nm can be well attributed to the (1 1 0) plane of α-NiMoO 4 , the (2 2 0) plane of β-NiMoO 4 , and the (1 1 1) plane of hexagonal Ni 2 P (PDF#03-0953), respectively, consolidating that Ni 2 P in situ formed after activation and coexisted with α-NiMoO 4 and β-NiMoO 4 .Ni 2 P has been demonstrated to possess excellent HER activity by both experimental and theoretical results, which can be attributed to the P-Ni bridge site bonding to the reaction intermediate H* with moderate intensity after incorporating with P. [47] Impressive HER activity was obtained when the Ni 2 P/β-NiMoO 4 heterogeneous interface was successfully constructed by Ni 2 P with excellent hydrogen adsorption energy and β-NiMoO 4 with excellent water dissociation kinetics.

UOR Electrochemical Performance Test
The UOR is the underlying reaction that determines the performance of modern urea-based energy conversion technologies, providing more economical electrons than OER for various renewable energy-related systems owing to its lower thermodynamic barriers. [8]Thus, the UOR activity of P-NiMoO 4 was evaluated in 1 M KOH with urea.The urea concentration and phosphorus doping content showed a key effect on the performance of UOR (Figure S13, Supporting Information).As the concentration of urea increased, the availability of the active species for the adsorption of other intermediates such as OH À onto the catalyst was reduced, thus limiting the conversion of urea into CO 2 and N 2 . [48]Similarly, P doping plays a critical role in the formation of oxygen vacancies and phosphates adsorbed on the surface and stabilizes the metastable phase of β-NiMoO 4 . [29]The optimized UOR activity was achieved at the urea concentration of 0.33 M with the phosphorus source amount of 1.0 g.Coincidently, this concentration is close to the average molar concentration of urea in human urine, allowing for the economical treatment and reuse of urine, a cheap and accessible natural source. [49]Compared with OER, the polarization curve of UOR showed a significantly reduced onset potential from 1.41 V (versus RHE) for OER to 1.26 V (versus RHE) for UOR (Figure 4a).The electrode of P-NiMoO 4 needs 1.35 V (versus RHE) to drive a current density of 100 mA cm À2 under UOR conditions.The needed potential was considerably reduced by 181 mV and at the same potential of 1.35 V, the UOR current density of P-NiMoO 4 was 5.9-20.8times higher than those of NiMoO 4 , RuO 2 , and NF (Figure 4b), manifesting the superiority and high performance of P-NiMoO 4 for UOR.Additionally, a much smaller Tafel slope of 19 mV dec À1 for P-NiMoO 4 was obtained than that of NiMoO 4 (29 mV dec À1 ) and RuO 2 (66 mV dec À1 ), indicating a superior UOR catalytic kinetics (Figure 4c).The smallest value of the charge transfer resistance (R ct ) obtained from P-NiMoO 4 (28 Ω) is consistent with its smallest Tafel slope, validating its excellent UOR charge transfer kinetics (Figure 4e and Table S5, Supporting Information).The relationship between η 100 and Tafel slopes was plotted and the results indicate that the UOR catalytic activity of P-NiMoO 4 is superior to that of most reported nonprecious metal catalysts (Figure 4d and Table S4, Supporting Information).Promisingly, the UOR activity of P-NiMoO 4 remained nearly unchanged after working continuously for 24 h at 1.28 V versus RHE (Figure 4f ), and compared with the initial LSV curve, the measured LSV curve after the stability test exhibited negligible change, collectively reflecting the outstanding electrocatalytic stability of P-NiMoO 4 for UOR in an alkaline environment as a highly potential catalyst for practical industrialized applications.
To study the catalytic mechanism of UOR, the sample of P-NiMoO 4 after 24 h testing under UOR conditions was investigated by XPS (Figure S14, Supporting Information).For Ni 2p, a new pair of peaks at 857.8 and 875.4 eV attributed to the highly oxidized state of Ni 3þ were observed, indicating that the Ni site was oxidized to the true active species of NiOOH, consistent with the previous report. [50]As for Mo 3d, the binding energy of all the peaks was reduced by 0.4 eV, implying a strong electronic interaction between Mo and Ni to allow for the electron transfer from Ni to Mo. [14,51] Apparently, the peak intensity of P 2p was reduced, indicating a significant dissolution of P. According to previous reports, leaching of anions from the catalysts usually induces surface reconstruction, facilitating the transformation of metal oxides into an amorphous phase of oxyhydroxide, consistent with the formation of NiOOH. [52]

Overall Urea Splitting
Considering the high activity and durability of the designed electrodes for HER and UOR, an A-P-NiMoO 4 || P-NiMoO 4 asymmetric urea electrolyzer was constructed in 1 M KOH containing 0.33 M urea (Figure 5a).First, the effect of urea on the HER activity of A-P-NiMoO 4 was evaluated (Figure 5b), and the result indicates that the presence of urea in the electrolyte exhibited no effect on the HER activity of A-P-NiMoO 4 in 1 M KOH, enabling the construction of the urea electrolyzer.Compared with the full water splitting (Figure 5c), the potential  required to deliver a current density of 10 mA cm À2 was reduced by 173 mV from 1.536 V to 1.363 V for urea electrolysis, demonstrating the superiority of the fabricated electrodes for HER and UOR as an effective way to produce hydrogen in an energyefficient manner.More importantly, the constructed UOR electrolyzer can operate stably for at least 24 h at 1.37 V without degradation, demonstrating its long-term operational stability (Figure 5d).The quantification of the generated hydrogen gives a Faraday efficiency close to 100% (Figure 5e), indicating that the observed current was from hydrogen evolution rather than other processes.Promisingly, a commercial dry cell of 1.5 V can efficiently drive the constructed urea electrolyzer (Figure 5f ), and the fast bubble release on the electrode surface was observed, further manifesting its vast potential for practical applications (Video S1, Supporting Information).

Conclusion
In summary, we discovered that the reconstruction with the formation of Ni 2 P and Mo 2 O 7 2À on the surface could largely enhance the alkaline HER activity of P-doped NiMoO 4 by an on-site electrochemical activation strategy.By P doping NiMoO 4 , the formation of β-NiMoO 4 was achieved, which could promote the kinetics of water dissociation and shift the HER rate-limiting step from the Volmer step to the Heyrovsky step.The systematic investigations manifested that a heterogeneous interface between NiMoO 4 and the newly formed Ni 2 P was generated and the dimer Mo 2 O 7 2À on the surface formed via the dissolutionredeposition process, which allowed for the exposure of additional Ni sites and the adjustment of hydrogen adsorption to facilitate HER kinetics.As a result, the activated A-P-NiMoO 4 electrode exhibited excellent performance for alkaline HER with a low overpotential of À48.9 mV for 10 mA cm À2 .Additionally, the as-fabricated P-NiMoO 4 electrode also exhibited outstanding UOR performance with an ultralow overpotential of 0.978 V for 100 mA cm À2 .Finally, a practical UOR electrolyzer was assembled, demonstrating a small cell voltage of 1.363 V to drive a current density of 10 mA cm À2 , and promisingly the electrolyzer could be efficiently driven by a commercial dry battery (1.5 V) for stable hydrogen production, suggesting that the proposed strategy represents an intriguing way for the rational design of advanced and cost-effective electrocatalysts for economical hydrogen production.

Experimental Section
Chemicals: Nickel foam (NF), hydrochloric acid (HCl, %36.0 %38.0%solution in water), ethanol, acetone, nickel nitrate hexahydrate (Ni(NO 3 ) 2  [53] the tetrahedral NiMoO 4 •nH 2 O micropillar array on nickel foam was obtained by hydrothermal reaction.The commercial nickel foam was sonicated sequentially with acetone, ethanol, 2 M HCl aqueous solution, and deionized water for 30 min, respectively.A piece of washed nickel foam (1 Â 2 cm 2 ) was immersed in the reaction solution of 15 mL containing Ni(NO 3 ) 2 •6H 2 O (0.04 M) and (NH 4 ) 6 Mo 7 O 24 •4H 2 O (0.01 M) in a Teflon autoclave.Then, the autoclave was sealed and heated in an oven at 150 °C for 6 h.After washing with deionized water, the product of NiMoO 4 •nH 2 O nanoarray was obtained for subsequent experiments.
Preparation of Tetrahedral P-NiMoO 4 Micropillar Array and Control Samples: The obtained NiMoO 4 •nH 2 O sample was annealed at 500 °C for 4 h using NaH 2 PO 2 •H 2 O as the phosphorus source to obtain phosphorus-doped nickel molybdate (P-NiMoO 4 ) in Ar.To optimize the electrochemical performance, the amount of the phosphorus source and the reaction temperature were studied.For comparison, a sample was obtained without the addition of the phosphorus source following the same procedure.The RuO 2 electrode was prepared using a drop-casting method. [54]Specifically, 10 mg of RuO 2 powder was dispersed in a solution containing 100 μL Nafion solution and 900 μL anhydrous ethanol.The mixed solution was sonicated for at least 60 min to form a homogeneous catalyst ink.The black dispersion was drop cast onto NF and then the sample was dried overnight in air at room temperature.Similarly, the Pt/C electrode was prepared.
Materials Characterization: XRD patterns were obtained on Bruker D8 Advance Powder X-ray diffractometer with a Cu Kα X-ray source (λ = 0.15406 nm).The scanning electron microscope (SEM) images were captured on a HITACHI S-4800 field-emission scanning electron microscope.TEM and HRTEM measurements were carried out using JEOL JEM 2100.The chemical valence states of the obtained products were studied by XPS (Thermo Scientific ESCALab 250).All the XPS spectra were corrected using a C 1s peak at 284.8 eV.The chemical composition of catalysts and electrolytes was analyzed by ICP-MS using an ICP-MS Agilent 7700.Raman spectra were conducted on Raman Spectrometry (Bruker) with a 532 nm excitation laser.
Electrochemical Measurements: The electrocatalytic activity was evaluated using a standard three-electrode system on the electrochemical workstation (CHI, 660E).An Ag/AgCl electrode, a platinum sheet electrode, and the prepared catalyst electrode were used as the reference, the counter electrode and the working electrode, respectively.All the potentials were calibrated to the RHE according to the equation of E (RHE) = E (Ag/AgCl) þ 0.197 þ 0.0591 Â pH, where pH = 14 in 1 M KOH solution. [55]The geometric area of all samples was controlled to be 1 cm 2 .The scan rate was 5 mV s À1 and all the results were corrected using 85% iR compensation.EIS curves were obtained at open circuit potentials between 10 À2 and 10 5 Hz with an amplitude of 5 mV.

Figure 2 .
Figure 2. Evaluation of HER electrochemical performance for P-NiMoO 4 .a) LSV curves.b) Overpotentials of the electrocatalysts at a specific current density.c) Tafel slopes.d) Electrochemical impedance spectra (The inset shows the equivalent circuit for fitting).e) Electrochemical double-layer capacitance.f ) Polarization curves normalized by ECSA.g) The apparent exchange current density ( j 0 ).h) Comparison of A-P-NiMoO 4 with reported advanced catalysts.i) The I-T curve of A-P-NiMoO 4 at À50 mV versus RHE; the inset shows the polarisation curve of A-P-NiMoO 4 before and after 2000 LSVs.

Figure 3 .
Figure 3. Composition and structural characterization of P-NiMoO 4 at different stages.a-c) High-resolution XPS spectra of P 2p, Ni 2p, and Mo 3d.d) Changes of the elemental content in the electrolyte.e) Raman spectra.f ) A HRTEM image of P-NiMoO 4 -1000.

Figure 4 .
Figure 4. Evaluation of the electrocatalytic UOR performance for P-NiMoO 4 .a,b) Polarization curves.c) Tafel slopes.d) Comparison of P-NiMoO 4 with reported UOR electrocatalysts.e) Electrochemical impedance spectra, and the inset shows the equivalent circuit for fitting.f ) Chronoamperometry curve of P-NiMoO 4 , and the inset shows the polarization curve of P-NiMoO 4 before and after stability test.

Figure 5 .
Figure 5. a) Schematic diagram of P-NiMoO 4 || A-P-NiMoO 4 urea electrolyzer.b) LSV curves of A-P-NiMoO 4 in 1 M KOH with/without 0.33 M urea.c) LSV curves of water and urea electrolysis and the inset shows a comparison of cell voltages for urea and water electrolysis at 10 mA cm À2 .d) curves.e) Faraday efficiency of hydrogen production.f ) A photograph of the practical urea electrolyzer driven by a commercial dry cell of 1.5 V.