Reduced Nickel Cobalt Tungstate as an Efficient Electrocatalyst for Urea-Assisted Hydrogen Production

Abstract

Urea electrolysis has emerged as a promising alternative to conventional water electrolysis for hydrogen production, owing to low electrical energy consumption as well as organic wastewater. However, the practical implementation of this approach is primarily constrained by the lack of cost-effective and efficient electrocatalysts. Thus, the development of earth-abundant, non-precious metal-based bifunctional electrocatalysts toward both the hydrogen evolution reaction (HER) and the urea oxidation reaction (UOR) is of critical importance. In this context, nanostructured, reduced nickel-cobalt tungstate supported on Ni foam is fabricated as a binder-free, freestanding electrode via a two-step hydrothermal process followed by partial thermal reduction. By systematically tuning the precursor concentrations of Ni, Co, and W, the morphology and electronic structure of the material are effectively modulated. The introduction of oxygen vacancies through partial thermal reduction plays a key role in enhancing charge transport properties. The optimized NiCo@W_0.5/NF electrode exhibits a porous, flower-like architecture and demonstrates excellent bifunctional electrocatalytic activity toward both UOR and HER, accompanied by improved mass transport behavior. When employed as both the anode and cathode for overall urea electrolysis, NiCo@W_0.5/NF requires a low cell voltage of only 1.68 V to achieve a current density of 100 mA cm⁻² and delivers impressive operational stability in an optimized electrolyte composed of 3 M KOH and 0.33 M urea. These results indicate that NiCo@W_0.5/NF is a highly promising and efficient bifunctional electrode material for urea assisted hydrogen production.

1. Introduction

Perhaps two of the key concerns to be addressed by society today are the identification and construction of a sustainable energy system. One of the most important pieces in this system is to replace our existing energy carrier with a sustainable fuel [1]. Hydrogen, as a renewable energy source, has zero carbon emissions and a high energy density, allowing it to partially replace the excessive use of fossil fuels and therefore reduce pollution [2,3]. Hydrogen production via water electrolysis is regarded as a green technology, which is constrained by the oxygen evolution reaction (OER) at the anode and the hydrogen evolution reaction (HER) at the cathode [4,5]. More specifically, there is a large kinetic overpotential for water splitting at the anode due to the sluggish OER kinetics [6,7]. Even though water splitting only requires a theoretical minimum voltage of 1.23 V, practical electrolyzers generally run at a cell voltage of 1.8–2.0 V due to the uphill nature of the process [8]. Substituting the OER with the oxidation of more easily oxidized molecules offers a viable strategy to overcome this limitation. Among these alternatives, urea oxidation has been investigated to lower the anode overpotential in electrolysis cells. Overall, urea electrolysis consists of two half-reactions: the urea oxidation reaction (UOR) at the anode and the hydrogen evolution reaction (HER) at the cathode. This process enables the simultaneous production of hydrogen and the remediation of nitrogen-containing effluents. Furthermore, the theoretical cell voltage is significantly reduced from 1.23 V for conventional water splitting to 0.37 V for urea electrolysis. As a result, the energy usage for H₂ generation may be efficiently reduced. Unfortunately, due to a six-electron transfer mechanism of UOR, it still has slow kinetics, necessitating the development of extremely efficient catalytic materials for overall urea oxidation. In addition, owing to their high cost and acute scarcity of materials, the most advanced Pt-, Ru-, and Ir-based composite catalysts cannot be widely used for HER, OER and UOR. Therefore, numerous studies have been conducted on non-noble metal-based bifunctional catalysts for overall urea electrolysis, such as transition-metal carbides [9,10,11,12], nitrides [13,14,15], phosphides [16,17,18,19], sulfides [20,21,22], oxides [23,24], and others. Furthermore, the use of bifunctional catalysts for both reactions simplifies the system architecture while decreasing overall cost. As a result, urea-assisted water electrolysis has emerged as a promising route for the large-scale production of high-purity hydrogen. Accordingly, the development of efficient bifunctional catalysts is critically important for overall urea electrolysis, and significant progress has been made in this area in recent years.

Nickel- (Ni)-based catalysts are among the most studied UOR electrocatalysts, owing to the crucial role of dynamic Ni³⁺ species in driving the reaction [25]. In view of this, Ni-based transitional metal oxides are promising non-noble metal catalysts but remain a great challenge due to their poor electronic conductivity [26]. Tungstate materials possess many advantages, such as simple synthesis processes, low cost, low toxicity and stable multifunctional properties [27]. Most importantly, tungstate materials have higher electronic conductivity than most binary and some ternary metal oxides [28,29]. In addition, the partial reduction of transition metal oxides increases the number of catalytic active sites due to oxygen vacancies within its crystal structure [23,30].

Motivated by these considerations, we hydrothermally grew nickel-cobalt tungstate (NiCoWO_x) directly on Ni-foam (NF) to form a binder-free, self-supported, and freestanding electrode. The dissolution–recrystallization process used to incorporate tungstate promotes well-defined morphologies with an optimized Ni:Co:W stoichiometric ratio, while subsequent thermal reduction introduces oxygen vacancies that enhance electrocatalytic activity. We systematically investigated the role of tungstate in promoting UOR and evaluated the fabricated catalyst as a bifunctional catalyst for urea-assisted hydrogen evolution. Furthermore, full-cell urea electrolysis performance was maximized by optimizing the concentrations of KOH and urea in the electrolyte. The optimized NF-supported NiCo tungstate exhibits stable operation for 140 h at a current density of 50 mA cm⁻² during overall urea-assisted water splitting. This binder-free bifunctional electrode demonstrates strong potential for cost-effective and energy-efficient hydrogen production while simultaneously enabling the valorization of organic waste streams.

2. Materials and Methods

2.1. Materials

In this study, all chemicals were obtained commercially and used as received without further purification. Nickel nitrate hexahydrate (Ni(NO₃)₂∙6H₂O), cobalt nitrate hexahydrate (Co(NO₃)₂∙6H₂O), sodium tungstate dihydrate (Na₂WO₄∙2H₂O), ammonium fluoride (NH₄F), urea (CO(NH₂)₂), and potassium hydroxide (KOH) were purchased from Sigma-Aldrich (Milwaukee, WI, USA). Nickel foam was procured from MTI Corporation (Richmond, CA, USA).

2.2. Materials Synthesis

The reduced NiCoWO_x materials were prepared by a two-step hydrothermal method. Typically, 1.5 mM of Ni(NO₃)₂.6H₂O and 1.5 mM Co(NO₃)₂.6H₂O were dissolved into a beaker containing 80 mL of deionized water. The mixture solution was stirred for about 10 min before adding 1.75 mM NH₄F and 5.25 mM urea under continuous stirring condition at room temperature. Pre-cleaned nickel foam with a 1 cm² active area were prepared and placed inside a 200 mL capacitive Teflon-lined stainless-steel autoclave. Subsequently, the reaction mixture was placed inside the autoclave and treated at 110 °C for 12 h. After the autoclave was cooled down to room temperature, the Ni foams (NFs) were collected and rinsed with deionized water and ethanol several times and dried overnight at 60 °C inside an oven. In the second step, dried NFs from the first-step hydrothermal experiment were placed inside the autoclave and a solution containing 3.0 mM Na₂WO₄ in 40 mL H₂O was poured inside the autoclave and reacted for 12 h at 180 °C. The overall Ni:Co:W molar ratio was ~1:1:2. After the autoclave was cooled down to room temperature, coated NFs were collected and rinsed with deionized water and ethanol several times and dried overnight. The dried sample (NiCo@W/NF-hydrothermal) was further annealed in 3% H₂/Ar at 450 °C for 2 h in a tube furnace and referred as NiCo@W/NF. A schematic illustration of the synthetic procedure is shown in Figure 1. Controlled samples, NiCo@W_0.5/NF, NiCo_0.5@W_0.5/NF, and Ni_0.5Co@W_0.5/NF, were fabricated in the same manner with varying Ni:Co:W molar ratios of 2:2:1, 2:1:1, and 1:2:1, respectively.

2.3. Characterization Techniques

The materials were characterized with X-ray diffraction by a PANalytical X’Pert PRO Theta/Theta Powder X-ray Diffraction System with Cu Kα radiation (λ = 0.15406 nm) at a scan rate of 10° min⁻¹. The morphologies of the prepared samples were observed by a Field Emission Scanning Electron Microscopy (FE-SEM, Zeiss Gemini SEM 500(Zeiss, Oberkochen, Germany)). The high-resolution transmission electron microscopy (HRTEM) images and energy dispersive spectra were collected using aTalos F200C G2 Transmission Electron Microscope(Thermo Fisher Scientific (formerly FEI), Brno, Czech Republic) operating at 200 kV. X-ray photoelectron spectroscopy (XPS) is carried out using monochromated, micro-focused, low power Al Ka X-ray on a Nexus X-Ray Photoelectron Spectrometer (Thermo Fisher Scientific, Hillsboro, OR, USA).

2.4. Electrochemical Characterizations

Electrochemical tests were conducted with Autolab Metrohm PGSTAT128N potentiostat/galvanostat at room temperature and ambient pressure. Catalysts supported NF (1.0 × 1.0 cm²) as a working electrode and Pt mesh as a counter electrode, and a Hg/HgO in 1 M KOH electrode was used as the reference electrode. To minimize the possibility of Pt contamination, several precautions and control experiments were implemented. First, all electrochemical measurements were conducted under carefully controlled conditions, and the electrolyte was freshly prepared for each experiment. The Pt mesh counter electrode was thoroughly cleaned prior to use, following standard electrochemical cleaning protocols to reduce surface impurities and loose Pt species. Cyclic voltammetry (CV), linear sweep voltammetry (LSV), electrochemical impedance spectroscopy (EIS) and chronoamperometric measurements were performed. The electrocatalytic HER was studied by linear sweep voltammetry in N₂-saturated 1 M KOH solutions. The scan rate for LSV was kept at 5 mV s⁻¹. LSV was also used to analyze the catalytic performance of the catalysts in an aqueous solution of 1 M or 3 M KOH with a desired concentration of urea (0.33 M or 0.5 M) at a scan rate of 10 mV s⁻¹. The Nernst equation was used to convert the potentials to the RHE scale: E(RHE) = E(Hg/HgO) + 0.059 pH + 0.098 V. Electrochemical impedance spectroscopy (EIS) measurements were carried out at 0.5 V vs. Ag/AgCl with an amplitude of 5 mV from 10⁵ to 0.01 Hz. The overall urea-assisted water electrolysis experiment was done in two electrode configurations, with the synthesized electrodes serving as both the UOR and HER electrodes, and linear sweep voltammetry was employed to measure from 1 to 2 V at 10 mV s⁻¹. The catalysts were evaluated in 0.33 M urea with 1.0 M or 3 M KOH electrolyte at a current density of 50 mA cm⁻² for catalytic durability.

3. Results and Discussion

3.1. Physicochemical Characterization

To analyze the crystal structure of the prepared NiCo@W/NF samples, the coated NF samples were gently scratched to collect the corresponding powder, which was subsequently characterized by X-ray diffraction (XRD). Figure 2 shows the XRD patterns of the NiCo@W after hydrothermal (Figure 2a) and after reductive thermal treatment in 3% H₂/Ar at 450 °C for 2 h (Figure 2b). The XRD patterns of hydrothermally synthesized and thermally reduced NiCo@W reveal the formation of monoclinic crystal structures, in good agreement with the standard diffraction patterns of NiWO₄ (ICDD No. 01-072-1189) and CoWO₄ (ICDD No. 00-015-0867), suggesting the successful formation of bimetallic NiCo tungstate [31]. Due to the close proximity of the characteristic diffraction peaks of NiWO₄ and CoWO₄ in their respective ICDD references, unambiguous phase identification between the two components is challenging. Therefore, the observed diffraction peaks were indexed accordingly. Comparable results have been reported by Kumar et al. [32] for hydrothermally synthesized NiCoWO₄. In particular, the XRD pattern of the NiCo@W/NF-hydrothermal sample is in good agreement with the diffraction pattern of NiCoWO₄ reported in their study. The standard diffraction pattern of monoclinic WO₃ (ICDD 01-072-1465) was used as a reference to examine the possible presence of WO_x phases, as shown in Figure 2. The diffraction patterns of the fabricated NiCo@W samples do not match this reference pattern, and the absence of any additional peaks indicates the phase purity of the prepared samples. Further, the intensity of diffraction peaks of the reduced NiCo@W are lower than those of the NiCo@W-hydrothermal, which can be attributed to the increased defect density induced by thermal reduction in 3% H₂/Ar atmosphere [33]. It is well known that defects/vacancies play a crucial role in enhancing electrocatalytic activities by modulating the electronic structure and increasing the number of active sites [34]. Consequently, the reduced NiCo@W/NF samples are expected to exhibit improved electrochemical performance. In addition, thermal reduction promotes a preferred crystallographic orientation along the (002) plane, which is evident from its higher intensity compared to the (−111) plane.

The preparation process of NiCo@W/NF samples involved two steps of hydrothermal followed by annealing in H₂/Ar atmosphere to introduce defects or vacancies in the self-supported catalysts. The first hydrothermal step in all probability led to the formation of NiCo hydroxy carbonate nanowires on the NF surface (NiCo/NF), as displayed in Figure 3a [35,36]. In the second step hydrothermal, this NiCo hydroxy-carbonate-coated NF was treated with aqueous tungstate solution for an anion exchange reaction. The resultant product was characterized as NiCo tungstate, supported on NF. The formation of NiCo tungstate in the second hydrothermal step was assumed to have undergone a dissolution and recrystallization process. This speculation is supported by analyzing the surface morphology, as the nanowire morphology of NiCo/NF in first hydrothermal was changed completely to button-shaped nanostructured morphologies after the second step (Figure 3b).

Furthermore, the effect of composition on the surface morphology was also investigated and the corresponding images are shown in Figure 4. NiCo@W_0.5/NF showed a nanostructured, porous, petal-like morphology. On the other hand, two other compositions, NiCo_0.5@W_0.5/NF and Ni_0.5Co@W_0.5/NF, exhibited dense/close packed structures, which seemed to have limited porosity. The porous flowerlike morphology of NiCo@W_0.5/NF (Figure S1) would be beneficial in electrolyte penetration, leading to an increase in the electroactive surface area and improved electrocatalytic performance. Thus, composition has a significant effect on the morphology of the samples.

NiCo@W_0.5/NF was further selected to analyze by transmission electron microscopy (TEM), owing to its distinctive morphology compared to the other compositions. Figure 5a,b show the low and high magnification images of NiCo@W_0.5/NF, which further affirmed the porous petal-like shape. The microstructure was analyzed by high-resolution TEM (HRTEM), as shown Figure 5c. The findings indicated the formation of (200), (020), (010), and (−111), corresponding to the lattice fringe spacings of 0.232, 0.277, 0.592, and 0.290 nm, respectively. Furthermore, the scanning transmission electron microscopy (STEM)-energy dispersive X-ray (EDX) mapping showed the presence of Ni, Co, W, and O in Figure 5d–h, suggesting the successful formation of NiCoWO_x.

The surface chemical states of NiCo@W_0.5/NF were analyzed by X-ray photoelectron spectroscopy (XPS), and Figure 6 indicates the presence of Ni, Co, W, and O as constituent elements. The high-resolution Ni 2p (Figure 6a) showed the appearance of strong peaks at 852.5 and 869.8 eV, corresponding to Ni⁰, which could have originated from the reductive thermal treatment as well as NF support [37]. The Ni ${2\mathrm{p}}_{3/2}$ and ${2\mathrm{p}}_{1/2}$ peaks were separated by a spin-orbit splitting of ~17.5 eV, which is a characteristic of Ni²⁺ [38]. In addition, the deconvoluted Ni ${2\mathrm{p}}_{1/2}$ and ${2\mathrm{p}}_{3/2}$ peaks indicated the co-existence of Ni²⁺ and Ni³⁺ [39], although relatively higher intensity suggested Ni²⁺ as the major oxidation state. The Co 2p spectrum also indicated the presence of Co⁰ (778.0 and 793.0 eV) and co-existence of Co³⁺ (781.0 and 796.8 eV) and Co²⁺ (783.0 and 798.7 eV) in Figure 6b [40,41,42]. The spin-orbit splitting of ~16 eV suggested Co²⁺ as the major oxidation state [43]. The deconvoluted W 4f spectrum showed two peaks at 31.3 and 33.4 eV owing to the metallic W and W⁴⁺ states, respectively [44]. In addition, two spin-orbit doublets appeared at 35.2 and 37.3 eV, which could be assigned to the W⁶⁺ states [45]. The deconvoluted O 1s peaks at 530.5, 531.6, and 532.4 eV could be attributed to the lattice oxygen or M–O bond (M is Ni or Co), oxygen defect sites, and physiosorbed water molecules [46]. Further, the relative oxygen vacancy concentration was estimated from the ratio of the fitted peak areas corresponding to deconvoluted lattice oxygen (A_lattice-O) and oxygen vacancies (A_vacancy-O), following the method reported by Zhang et al. [47]:

$${relative C}_{{\mathrm{O}}_{\mathrm{v}}}=\frac{A_{\mathrm{v}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y}\mathrm{-}\mathrm{O}}}{A_{\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{-}\mathrm{O}}+A_{\mathrm{v}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y}\mathrm{-}\mathrm{O}}}$$

(1)

It should be noted that this peak area ratio does not represent the absolute oxygen vacancy content; rather, it provides a qualitative measure of the oxygen vacancy concentration in the near-surface region of NiCo@W_0.5/NF. Based on the deconvolution of the O 1s XPS spectrum, the relative oxygen vacancy fraction was calculated to be approximately ~29.6%.

3.2. Electrocatalytic Properties

The electrocatalytic urea oxidation performance of the prepared NF-supported NiCoWO_x samples were investigated in 1 M KOH in the presence of 0.33 M urea, and the corresponding findings are displayed in Figure 7. The polarization curves (Figure 7a) indicated that NiCo@W_0.5/NF delivered the highest current density of ~202.54 mA cm⁻² for UOR at 1.625 V vs. RHE compared to the other prepared samples. The required potentials to reach 100 mA cm⁻² current density were found to be ~1.556, 1.550, 1.473, 1.522, and 1.484 V vs. RHE for NiCo/NF, NiCo@W/NF, NiCo@W_0.5/NF, Ni_0.5Co@W_0.5/NF, and NiCo_0.5@W_0.5/NF, respectively. This clearly indicates the superior UOR activity of NiCo@W_0.5/NF among the compositionally different NiCoWO_x samples grown on NF by two-step hydrothermal. In addition, this sample showed significant improvement in the UOR activity in comparison to the NiCo/NF, which suggested the beneficial role of W incorporation. Notably, the optimal composition and dissolution–recrystallization synthetic approach were advantageous to enhance the electrocatalytic performance of NiCoWO_x. The bare NF support was unable to reach 100 mA cm⁻² current density for UOR, which suggested NiCo tungstate as an effective electrode material for electrocatalytic urea oxidation. These findings inferred NiCo@W_0.5/NF as the best sample for UOR among the prepared samples. The kinetic aspects of UOR for the prepared samples were also analyzed from the Tafel plot, derived from the LSV curves of the corresponding samples, as shown in Figure 7b. The Tafel slope values of NiCo/NF (155 mV dec⁻¹), NiCo@W/NF (189 mV dec⁻¹), NiCo@W_0.5/NF (132 mV dec⁻¹), Ni_0.5Co@W_0.5/NF (225 mV dec⁻¹), and NiCo_0.5@W_0.5/NF (144 mV dec⁻¹) indicated the lowest Tafel slope for NiCo@W_0.5/NF, suggesting a relatively faster reaction rate of urea oxidation. To further confirm the high UOR activity of NiCo@W_0.5/NF, the polarization curves were recorded in 1 M KOH with and without 0.33 M urea in Figure 7c. It was observed that the current density at 1.625 V was increased by ~3.34 times in the presence of urea, which reaffirms the efficient UOR performance of NiCo@W_0.5/NF.

Furthermore, the charge transfer characteristics of the prepared samples were evaluated by Electrochemical Impedance analysis at 1.374 V vs. RHE in 1 M KOH with 0.33 M urea. Figure 7d shows the corresponding Nyquist plots, which were fitted to the equivalent circuit, LR_s (R₀ C) (R_ct CPE), shown at the inset. In this model, L, R_s, R₀, and R_ct correspond to the inductance, ohmic resistance, intrinsic resistance of the oxide catalyst, and the charge-transfer resistance, respectively [47]. C and constant phase element (CPE) represent the capacitive contributions associated with the catalyst layer and the electrode-electrolyte interface [36]. CPE is commonly introduced to account for depressed semicircles in Nyquist plots arising from surface heterogeneity and electrode porosity [48]. The impedance of CPE is generally defined as follows [49]:

$$Z_{CPE}={\displaystyle \frac{1}{\left[Q\right(iw{)}^{\alpha }]}}$$

(2)

where Q, w, i, and α or n is a frequency-independent constant, angular frequency of the AC signal, imaginary unit, and factor varies between 0 to 1. For n = 1, the CPE corresponds to an ideal capacitive element, whereas, for n ≠ 1, Q has units other than those of capacitance [50].

The EIS fitted parameters were listed in Table S1 and the R_ct values followed the order NiCo@W_0.5/NF (0.504 Ω) < NiCo_0.5@W_0.5/NF (0.581 Ω) < NiCo@W/NF (1.151 Ω) < Ni_0.5Co@W_0.5/NF (1.073 Ω) < NiCo/NF (1.385 Ω). The lowest R_ct and total resistance (R_o + R_ct) indicated faster reaction kinetics and enhanced charge transfer for UOR, when catalyzed by NiCo@W_0.5/NF compared to the other reduced NiCoWO_x samples. In addition, electrical double layer capacitance (C_dl) was calculated from the CPE associated with the R_ct using the EIS fitted parameters from Table S1 following the Brug relation [51]:

$$C_{\mathrm{d}\mathrm{l}}={CPE}^{1/\alpha }{\left(\frac{1}{R_{\mathrm{s}}}+\frac{1}{R_{\mathrm{c}\mathrm{t}}}\right)}^{(\alpha -1)/\alpha }$$

(3)

The total electrochemically accessible surface area (electrode level) of the catalyst-coated Ni foam electrode was estimated from the double-layer capacitance (C_dl) measurements. It should be noted that the calculated total electrochemically accessible surface area (electrode level) values represent the total electrochemically accessible electrode surface (catalyst deposited on Ni foam), rather than the intrinsic catalyst-only surface area, due to the capacitive contribution and possible surface transformations of the Ni substrate. C_dl values were used to determine the total electrochemically accessible surface area (electrode level) using the following relation [52]:

$$\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{l}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{y} \mathrm{a}\mathrm{c}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{b}\mathrm{l}\mathrm{e} \mathrm{s}\mathrm{u}\mathrm{r}\mathrm{f}\mathrm{a}\mathrm{c}\mathrm{e} \mathrm{a}\mathrm{r}\mathrm{e}\mathrm{a}(\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{e} \mathrm{l}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{l})=\frac{C_{\mathrm{d}\mathrm{l}}}{C_{\mathrm{s}}}$$

(4)

Ni foam was used as the substrate to fabricate the free-standing electrode materials. Owing to its highly porous architecture, the use of conventional C_s value, such as 0.04 mF cm⁻² for an atomically smooth flat surface, would lead to a significant overestimation of the total electrochemically accessible surface area (electrode level). Therefore, the experimentally determined C_dl value of bare Ni foam (~17.18 mF for a geometric area of 1 cm²) was used as the C_s in this work. Accordingly, the relative total electrochemically accessible surface area (electrode level) was calculated as follows:

$$\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{l}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{y} \mathrm{a}\mathrm{c}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{b}\mathrm{l}\mathrm{e} \mathrm{s}\mathrm{u}\mathrm{r}\mathrm{f}\mathrm{a}\mathrm{c}\mathrm{e} \mathrm{a}\mathrm{r}\mathrm{e}\mathrm{a}(\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{e} \mathrm{l}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{l})=\frac{C_{\mathrm{d}\mathrm{l}}}{C_{\mathrm{d}\mathrm{l}\mathrm{,}\mathrm{ }\mathrm{N}\mathrm{i}\mathrm{ }\mathrm{f}\mathrm{o}\mathrm{a}\mathrm{m}}}$$

(5)

The estimated total electrochemically accessible surface area (electrode level) values were ~8.07, 6.09, 30.50, 10.58, and 5.86 cm² for NiCo/NF, NiCo@W/NF, NiCo@W_0.5/NF, NiCo_0.5@W_0.5/NF, and Ni_0.5Co@W_0.5/NF, respectively. NiCo@W_0.5/NF exhibited the highest total electrochemically accessible surface area (electrode level), suggesting a porous nature and enhanced electrolyte penetration. Thus, incorporation of W in optimized concentration was crucial to achieve remarkably low R_ct and high total electrochemically accessible surface area (electrode level)

To investigate the bifunctional behavior of the reduced NiCoWO_x/NF samples, activity for hydrogen evolution reaction (HER) was evaluated in 1 M KOH. The LSV curves (Figure 8a) exhibited that Ni_0.5Co@W_0.5/NF showed better alkaline HER initially compared to other samples; however, at a higher potential region, it failed to retain its activity. The maximum current density obtained for alkaline HER with Ni_0.5Co@W_0.5/NF was only ~64.12 mA cm⁻² at 0.36 V vs. RHE, whereas NiCo@W_0.5/NF delivered ~114.68 mA cm⁻² at 0.37 V vs. RHE. Thus, it is evident that NiCo@W_0.5/NF showed improved mass transport properties compared to the other reduced NiCoWO_x/NF samples and NiCo/NF. Further, the Tafel slopes derived from the polarization curves, displayed in Figure 8b, exhibited the lowest value for Ni_0.5Co@W_0.5/NF, suggesting better reaction kinetics than that catalyzed by other samples. However, Ni_0.5Co@W_0.5/NF exhibited superior performance in the kinetic region. As observed from the Tafel plot, it showed very poor mass transfer characteristics and failed to deliver high current densities of HER compared to NiCo@W_0.5/NF, as observed from the polarization curves. This could be ascribed to its dense morphology, which limited the electrolyte penetration and accessibility of the active metal centers, resulting in a very low total electrochemically accessible surface area (electrode level) (~5.86 cm²). Therefore, NiCo@W_0.5/NF could be considered as the preferred electrode material for the HER among the prepared samples, demonstrating bifunctional electrocatalytic activity for both UOR and HER in alkaline media.

3.2.1. Role of Oxygen Vacancy

The superior UOR and HER activities of NiCo@W_0.5/NF can be rationalized by the synergistic effects of oxygen vacancy, electronic structure modulation, and electrode architecture. Oxygen vacancies create under-coordinated Ni/Co sites and redistribute local charge, which promotes the adsorption and activation of urea molecules and ${\mathrm{O}\mathrm{H}}^{-}$ ions, thereby facilitating the first dehydrogenation step of UOR [30,53]. This effect is reflected in the significantly lower potential required to reach 100 mA cm⁻² and the smallest Tafel slope (132 mV dec⁻¹) observed for NiCo@W_0.5/NF, indicating accelerated UOR kinetics.

In parallel, the presence of oxygen vacancies is accompanied by the partial reduction of neighboring Ni/Co centers, which modifies their oxidation states and makes them more readily oxidizable to catalytically active high valent species (Ni³⁺/Co³⁺) under anodic polarization. Such electronic tuning promotes C–N bond activation during UOR [54,55] and contributes to the remarkably lower R_ct (~0.504 Ω) measured for NiCo@W_0.5/NF, demonstrating more efficient interfacial electron transport. Although oxygen vacancies also introduce donor states that improve electrical conductivity, the strong correlation between vacancy-rich composition, reduced Tafel slope, and lower R_ct suggests that their dominant role is to enhance reaction kinetics through adsorption and electronic structure modulation rather than conductivity alone.

For HER, oxygen vacancies similarly influence activity by adjusting the electron density of adjacent Ni/Co sites and optimizing metal–hydrogen binding strength [56]. While Ni_0.5Co@W_0.5/NF exhibits a lower Tafel slope, its dense morphology and low total electrochemically accessible surface area (electrode level) (~5.86 cm²) limit electrolyte diffusion and mass transport, preventing it from sustaining high HER current densities. In contrast, NiCo@W_0.5/NF combines an appropriate oxygen vacancy concentration with a highly porous structure and the highest total electrochemically accessible surface area (electrode level) (~30.50 cm²), enabling sufficient exposure of vacancy-adjacent active sites and efficient transport of reactants and products. This balance explains why NiCo@W_0.5/NF delivers high HER current density, outperforming the other reduced NiCoWO_x/NF samples.

Overall, these results demonstrate that oxygen vacancies play a key role in enhancing both UOR and HER by (i) creating adsorption-active metal sites, (ii) modulating Ni/Co oxidation states and redox accessibility, and (iii) facilitating charge transfer. However, optimal catalytic performance is achieved only when the vacancy concentration is balanced with favorable morphology and high total electrochemically accessible surface area (electrode level), as excessive densification or poor mass transport can offset the intrinsic kinetic advantages provided by oxygen vacancies.

3.2.2. Urea Electrolysis Performance in Full Cell

To check the performance of urea electrolysis in a full cell, the NiCo@W_0.5/NF was taken as both anode and cathode in a H-cell configuration. The NiCo@W_0.5/NF‖NiCo@W_0.5/NF cell was tested in 1 M KOH with and without 0.33 M urea, combining the UOR and HER. The corresponding performances of NiCo@W_0.5/NF‖NiCo@W_0.5/NF cell are also compared with that of conventional water electrolysis (OER and HER) in the absence of urea in the electrolyte. The polarization curves (Figure 9a) showed that NiCo@W_0.5/NF‖NiCo@W_0.5/NF cell required ~130, 220, and 210 mV less potentials to deliver 10, 50, and 100 mA cm⁻² current densities for urea electrolysis compared to water electrolysis, respectively. This suggested that replacing OER with UOR to couple with HER for hydrogen production is effective and energy efficient.

Furthermore, at lower concentrations of KOH in urea electrolysis, deviation due to ${\mathrm{O}\mathrm{H}}^{-}$ deprivation could be a reason for competitive OER to become more dominant than UOR, which could alter the reaction pathway and affect overall hydrogen generation efficiency [57,58]. To alleviate this, the concentration of the electrolyte should be optimized. In this regard, the effect of electrolyte concentration was also studied in the NiCo@W_0.5/NF‖NiCo@W_0.5/NF cell by using varied combinations of KOH and urea concentrations, as shown in Figure 9b,c. The choice of KOH and urea concentrations was guided by a balance between ${\mathrm{O}\mathrm{H}}^{-}$ ion availability, which governs UOR kinetics through catalyst activation, and urea concentration, which influences reactant availability and mass transport. As discussed in our previous work [59], increasing KOH concentration reduces UOR transport resistance in both the kinetic and mass transport regions. The presence of hydroxide ions promotes the deprotonation of Ni-based catalysts, facilitating the formation of catalytically active Ni³⁺ species, which are responsible for urea oxidation reactions as follows [60]:

$${6\mathrm{N}\mathrm{i}}^{2+}{\left(\mathrm{O}\mathrm{H}\right)}_{2\left(\mathrm{s}\right)}+{6\mathrm{O}\mathrm{H}}^{-}\to {6\mathrm{N}\mathrm{i}}^{3+}\mathrm{O}\mathrm{O}{\mathrm{H}}_{\left(\mathrm{s}\right)}+{6\mathrm{H}}_2\mathrm{O}+{6\mathrm{e}}^{-}$$

(6)

$${6\mathrm{N}\mathrm{i}}^{3+}\mathrm{O}\mathrm{O}{\mathrm{H}}_{\left(\mathrm{s}\right)}+\mathrm{C}\mathrm{O}(\mathrm{N}{\mathrm{H}}_2{)}_2+{6\mathrm{O}\mathrm{H}}^{-}\to {6\mathrm{N}\mathrm{i}}^{2+}{\left(\mathrm{O}\mathrm{H}\right)}_{2\left(\mathrm{s}\right)}+{\mathrm{N}}_{2\left(\mathrm{g}\right)}+{5\mathrm{H}}_2\mathrm{O}+{\mathrm{C}\mathrm{O}}_{2\left(\mathrm{g}\right)}$$

(7)

Higher ${\mathrm{O}\mathrm{H}}^{-}$ concentration enables faster and more efficient generation of Ni³⁺, thereby enhancing UOR kinetics and shifting the onset potential to lower values, consistent with prior reports on Ni-based catalysts in alkaline media [61,62,63,64].

At lower KOH concentrations (0.5–1 M), the UOR transport resistance in the kinetic region is significantly higher than that in the mass transport region, indicating that insufficient ${\mathrm{O}\mathrm{H}}^{-}$ availability limits catalyst activation and charge-transfer kinetics. In contrast, at higher KOH concentration (3 M), the transport resistances in the kinetic and mass transport regions become comparable, suggesting that ${\mathrm{O}\mathrm{H}}^{-}$ availability is no longer the rate-limiting factor and that the reaction proceeds under more balanced kinetic and transport conditions. Additionally, the variation in UOR transport resistance with catalyst loading is minimized at 3 M KOH, further indicating a stable and optimal reaction environment. With respect to urea concentration, increasing urea content enhances reactant availability but can also introduce diffusion limitations and competitive adsorption effects at excessively high concentrations. The selected urea concentration (0.33 M) represents an optimal compromise, providing sufficient urea flux to sustain high UOR rates without inducing mass transport penalties or suppressing ${\mathrm{O}\mathrm{H}}^{-}$ accessibility at the catalyst surface.

Thus, deviation in the polarization curves were observed with KOH and urea concentrations of 1 M and 0.5 M, respectively. This suggested that lower KOH and higher urea concentrations are unfavorable for urea electrolysis with the NiCo@W_0.5/NF‖NiCo@W_0.5/NF cell. Based on these, the 3 M KOH with a 0.33 M urea electrolyte composition offered an optimal balance between catalyst activation (via ${\mathrm{O}\mathrm{H}}^{-}$ driven Ni³⁺ formation) and favorable mass transport, leading to the best urea electrolysis performance, requiring much lower potentials of only 1.68 and 2.28 V to reach 100 and 500 mA cm⁻² current densities than that for water electrolysis (Figure 9c).

To evaluate the practical applicability of the NiCo@W_0.5/NF‖NiCo@W_0.5/NF cell for urea electrolysis, its durability was tested galvanostatically at 50 mA cm⁻² current density for 140 h in 3 M KOH with 0.33 M urea. The Voltage vs. Time plot (Figure 9d) showed that the potential increased initially to 1.70 V in ~8.8 h and, after, that it was almost stable with a slight change until 140 h, indicating the promising stability of the NiCo@W_0.5/NF‖NiCo@W_0.5/NF cell. Therefore, NiCo@W_0.5/NF can be considered as an effective and stable bifunctional electrode material for urea-assisted hydrogen production.

Post-mortem characterization of the NiCo@W_0.5/NF anode was carried out using SEM–EDX elemental mapping and XPS to evaluate compositional and structural changes after 140 h of urea electrolysis in 3 M KOH containing 0.33 M urea. Figure S2 presents the SEM image and corresponding EDX elemental maps of the NiCo@W_0.5/NF anode, confirming the presence of the constituent elements Ni, Co, W, and O, thereby indicating the good compositional stability of the electrode. The surface chemical states of the post-UOR NiCo@W_0.5/NF anode were further examined by XPS, and the results are shown in Figure S3. Deconvolution of the Ni 2p spectrum revealed a pronounced increase in the proportion of Ni³⁺ species, as reflected by a change in the Ni²⁺/Ni³⁺ atomic ratio (based on the 2p_3/2 peak) from 5.71 in the pristine sample to 0.39 after UOR. Similarly, analysis of the high-resolution Co 2p_3/2 spectrum showed that the Co²⁺/Co³⁺ atomic ratio decreased from 2.42 to 1.02. In addition, the Ni⁰ peak completely disappeared after UOR, whereas a residual Co⁰ signal was still observed. These findings suggest that Ni sites act as the primary active centers and undergo oxidation to Ni³⁺ species, which are known to be highly active toward the UOR. The diminished intensity of the W 4f signal implies partial dissolution of tungstate species $\left({WO}_4^{2-}\right)$ in the strongly alkaline electrolyte (3 M KOH), accompanied by electrooxidation of the metal ions [39,65]. Despite the partial leaching of tungsten, the NiCo@W_0.5/NF electrode maintained stable urea electrolysis performance, as demonstrated in Figure 9d.

3.2.3. Commercial Feasibility of Urea-Assisted HER

In the context of urea-assisted HER, the sustainability of the process does not rely on synthetically produced, fertilizer-grade urea, but, rather, on the utilization of urea-rich waste streams that are already generated in large quantities. Notably, urea is abundantly present in human urine, animal urine, and agricultural and livestock wastewater, as well as in effluents from urea and fertilizer manufacturing plants. These waste streams typically contain urea concentrations ranging from tens to several hundred millimolar, making them directly relevant to the concentration range explored in this study. Accordingly, wastewater electrolysis herein refers to urea-assisted electrolysis conducted in a well-defined electrolyte under controlled laboratory conditions, serving as a representative system for urea-rich wastewater streams.

From an industrial and environmental perspective, the integration of urea-assisted HER with wastewater treatment offers a dual benefit: (i) valorization of urea-containing waste as a feedstock for hydrogen production, and (ii) simultaneous mitigation of nitrogen-rich effluents, which are otherwise energy-intensive, to be treated using conventional biological or chemical processes. This coupling has been widely proposed as a sustainable pathway for decentralized hydrogen production while addressing water pollution challenges. Therefore, the commercial feasibility of urea-assisted HER is envisioned primarily in applications linked to wastewater treatment facilities, agricultural operations, and industrial effluent streams, rather than relying on purpose-made urea.

4. Conclusions

In summary, a partially reduced NiCoWO_x/NF electrode was developed via a two-step hydrothermal synthesis followed by thermal reduction, yielding a binder-free, self-supported, non-precious electrocatalyst for urea-assisted hydrogen production. The dissolution–recrystallization strategy employed during the second hydrothermal step for tungstate formation proved effective in generating a porous electrode morphology with an optimized elemental composition. Subsequent reductive thermal treatment under a controlled atmosphere further enhanced charge-transport properties by introducing oxygen vacancies. The optimized NiCo@W_0.5/NF electrode, with an optimal Ni:Co:W molar ratio of 2:2:1, demonstrated pronounced bifunctional activity, delivering high catalytic performance for both UOR as well as HER, along with improved mass-transport characteristics in alkaline media. Importantly, as a bifunctional electrode, NiCo@W_0.5/NF demonstrated promising performance in a full-cell urea electrolysis configuration. By optimizing the electrolyte composition, the electrolyzer delivered a current density of 100 mA cm⁻² at a low cell voltage of 1.68 V and maintained stable operation for over 140 h at 50 mA cm⁻². These results establish NiCo@W_0.5/NF as an efficient and durable electrode material for urea electrolysis. This study provides valuable insights for the design of low-cost, energy-efficient electrolyzers with freestanding bifunctional electrodes, enabling sustainable hydrogen production while simultaneously valorizing low-grade and agricultural wastewater streams.