All reagents employed in this study were used as received without any additional purification. Cobalt nitrate hexahydrate (Co(NO₃)₂·6H₂O, reagent grade 99%), ammonium fluoride (NH₄F, reagent grade 96%), and hydrochloric acid (HCl) were acquired from Aladdin Industrial Corporation (Shanghai, China). Thiourea (CH₄N₂S, reagent grade 99%), tungsten (VI) chloride (WCl₆, reagent grade 99%), a 5 wt.% Nafion solution, and potassium hydroxide (KOH, reagent grade 90%) were obtained from Macklin (Shanghai, China). A commercial Pt/C catalyst (20 wt.%) was supplied by Hessen Corporation (Shanghai, China).

A one-step electrodeposition strategy was used to prepare S,W-Co(OH)₂/NF. In a standard fabrication, a fragment of NF substrate (1 × 1 cm²) was thoroughly washed in 3.0 M HCl, distilled water, and anhydrous ethanol multiple times to remove surface oxides and organic contaminants. Electrodeposition was conducted at room temperature in a two-electrode system using 50 mL of electrolyte containing 4 mM Co(NO₃)₂·6H₂O, 20 mM CH₄N₂S, 13.5 mM NH₄F, and 0.05 mM WCl₆. The pretreated NF served as the working electrode, while a graphite rod served as both the counter electrode and reference electrode. Deposition was performed at -5 V for 30 min, followed by rinsing with water to obtain the S,W_0.05-Co(OH)₂/NF electrode. Dual-doped catalysts with W_0.01 and W_0.1 were synthesized using the same method, with WCl₆ concentrations adjusted to 0.01 mM and 0.1 mM, respectively. Samples with different deposition times (10, 20, and 30 min) were also prepared for comparison. The S-Co(OH)₂/NF sample was fabricated in the same way as the dual-doped catalyst, but without the addition of WCl₆ to the electrolyte. Similarly, W-Co(OH)₂/NF was synthesized by omitting CH₄N₂S from the electrolyte. For the Pt/C/NF electrode, commercial 20% Pt/C powder was dispersed in a mixed solution of deionized water, ethanol, and Nafion binder, and then drop-cast uniformly onto the NF surface. The resulting Pt/C/NF was then dried overnight in a vacuum oven.

The morphology and structural features of the synthesized products were characterized by field emission scanning electron microscopy (FESEM; SU-8220, Hitachi, Tokyo, Japan) and Transmission electron microscopy (TEM; HT7700, JEM-2010, Talos F200X, Hitachi, Tokyo, Japan). Powder X-ray diffraction (XRD; TTR-III, Tokyo, Japan) analysis was performed to determine the crystalline structure. Raman spectra were recorded using a Renishaw inVia Raman spectrometer with a 532 nm laser as the excitation source. Energy-dispersive spectroscopy (EDS) mapping was conducted using a Talos F200X (Thermo Fisher Scientific, USA) TEM working at 200 kV. Surface chemical states were analyzed by X-ray photoelectron spectroscopy (XPS; ESCALAB 250, UK) using Al Kα radiation as the excitation source.

The electrochemical performance of the synthesized catalysts was evaluated using a three-electrode system with a Corrtest (CS-310X) electrochemical workstation at ambient temperature. The prepared samples functioned as the working electrode (testing area: 0.25 cm²), whereas Hg/HgO (1 M KOH) served as the reference electrode, and a graphite rod was used as the counter electrode in an alkaline medium. The HER and UOR measurements were conducted in 1.0 M KOH and 1.0 M KOH containing urea, respectively. The linear sweep voltammetry (LSV) curves were obtained at 5 mV s^-1, and the Tafel slope was derived by fitting the linear region of the Tafel plot using the Tafel equation, [η = b log (j) + a]. Electrochemical impedance spectroscopy (EIS) was performed over a frequency range from 100 kHz to 0.01 Hz with an amplitude of 5 mV. The electrochemically active surface area (ECSA) was calculated according to a previously reported equation: ECSA = A_geo × C_dl /C_s, where A_geo is the geometric area of the electrode, C_dl represents the electric double-layer capacitance in the non-Faradaic region, and C_s is the specific capacitance of the material (typically 0.04 mF cm^-2). C_dl was determined from cyclic voltammetry (CV) measurements at scan rates ranging from 10 to 50 mV s^-1. The current density difference (Δj = ja - jc) was graphed against the scan rate, and the slope of the linear fit corresponds to twice (C_dl). All potentials measured during the three-electrode performance tests were iR-corrected and converted to the reversible hydrogen electrode (RHE) scale using the following equation: E_(RHE) = E_(Hg/HgO) + 0.059 × pH + 0.098 - 90% × iR

This study utilized the Vienna Ab Initio Simulation Package (VASP) for theoretical computations of materials, including DFT^[28]. The Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional was adopted within the generalized gradient approximation (GGA), together with projector augmented-wave (PAW) pseudopotentials^[29]. The empirical dispersion-corrected DFT method, specifically DFT-D3, was employed to accurately account for weak long-range van der Waals interactions. A kinetic energy cutoff of 500 eV was applied for the plane wave expansion. The convergence criterion for the self-consistent field (SCF) iterations was set to 10^-5 eV. Brillouin zone integration was executed using the Gamma sampling technique, with a geometric optimization density of 2 × 2 × 1 and an electronic structure calculation density of 4 × 4 × 1. A vacuum region of 15 Å was implemented in all systems to prevent interactions between adjacent slabs. The geometric optimization was performed using the conjugate gradient method, ensuring that the force on each atom remained below 0.03 eV Å^-1. The Gibbs free energy change for adsorption or reaction processes was calculated as: ΔG = ΔE + (ZPE-TΔS), where ΔE is the adsorption or reaction energy derived from DFT calculations, ΔZPE indicates the zero-point energy correction, T represents the temperature, and ΔS is the entropy change.

[Figure 1A] schematically illustrates the facile one-step electrodeposition used to fabricate

S,W-Co(OH)₂/NF. The single-doped and pristine catalysts were synthesized under identical conditions but without the dopant sources. After deposition, the initially porous NF, as depicted in [Supplementary Figure 1], is uniformly covered by precursor nanosheets, as shown in [Supplementary Figure 2], forming a conformal and porous catalyst layer. The XRD patterns in [Figure 1B] confirm that both catalysts adopt the Co(OH)₂ phase (JCPDS No. 30-0443). Although slight peak broadening is observed after doping, the characteristic reflections of Co(OH)₂ are retained, indicating a preserved phase structure. The lattice distortion induced by W and S incorporation enhances defect density and electronic conductivity, contributing to improved electrocatalytic performance. Likewise, S-Co(OH)₂/NF and W-Co(OH)₂/NF display characteristic diffraction patterns of Co(OH)₂, as depicted in [Supplementary Figure 3]. Diffraction peaks from metallic nickel arise from the Ni foam substrate. The morphology of the samples was investigated by FESEM and TEM. As presented in [Figure 1C and D], the FESEM images show a three-dimensional network of interconnected nanosheets, which provides numerous exposed active sites. TEM further corroborates the ultrathin morphology of dual-doped S,W-Co(OH)₂/NF, as demonstrated in [Figure 1E and F]. The HRTEM image of S,W-Co(OH)₂/NF in [Figure 1G] reveals distinct lattice fringes with d-spacings of 0.238 and 0.208 nm, corresponding to the (101) and (002) planes of Co(OH)₂, respectively, consistent with the XRD results. Furthermore, the EDS mapping data depicted in [Figure 1H] demonstrate homogeneous elemental distributions of Co, oxygen (O), S, and W across the nanosheets, underscoring the efficacy of the doping process. Although the W signal in EDS appears relatively weak due to its low concentration, inductively coupled plasma (ICP) analysis confirms a W content of 0.73 at% [Supplementary Figure 4 and Supplementary Table 1], verifying successful incorporation. XPS further confirms the chemical state of W within the lattice.

Raman spectroscopy, owing to its sensitivity to subtle microstructural variations in inorganic frameworks, was employed to probe the lattice characteristics of both catalysts. As shown in [Figure 2A], both samples exhibit nearly identical peaks, indicating that W incorporation preserves the overall Co(OH)₂ lattice. Nevertheless, S,W-Co(OH)₂/NF displays discernible red shifts in the vibrational bands centered at 466, 505, and 660 cm^-1 relative to S-Co(OH)₂/NF, indicative of lattice perturbation and modified local bonding environments upon W incorporation, consistent with vibrational softening around Co sites^[30]. XPS was performed to clarify the surface composition and valence states of the pristine, S-doped, and dual-doped catalysts. The survey spectrum of the dual-doped catalyst [Figure 2B] validates the existence of Co, S, O, and W. The Co 2p spectrum was deconvoluted into two spin-orbit doublets with associated satellite peaks, yielding Co³⁺ peaks at 781.0 and 796.9 eV and Co²⁺ peaks at 783.0 and 797.8 eV, as shown in [Figure 2C]^[31]. Notably, the Co 2p binding energies of the dual-doped catalyst shift positively by 0.34 eV in comparison to the S-doped and pristine catalysts, evidencing electron redistribution and a modified Co electronic structure induced by W doping. The high-resolution S 2p spectra [Figure 2D] for both catalysts exhibit three characteristic peaks corresponding to S 2p_3/2, S 2p_1/2, and S-O^[32]. The W 4f spectrum [Figure 2E] exhibits two distinct peaks at binding energies of 35.1 eV and 37.2 eV, attributed to the W 4f_7/2 and W 4f_5/2 spin-orbit components, respectively, corroborating the existence of W^6+[33]. Although the surface W concentrations appear comparable, the catalytic performance is governed by electronic structure modulation rather than dopant content alone. The synergistic interaction between W and S induces more effective charge redistribution and optimized adsorption energetics, leading to enhanced catalytic activity. In the O 1s XPS spectra [Figure 2F], the catalysts exhibit three components located at 530.5, 531.5, and 533.0 eV, corresponding to lattice oxygen (M-O), hydroxyl species (M-OH), and adsorbed H₂O, respectively. The dominant M-OH peak indicates a hydroxyl-rich surface, while slight intensity variations after S and W doping suggest modulation of the surface oxygen environment^[34,35]. The C 1s spectrum [Supplementary Figure 5] displays three characteristic carbon peaks at 284.8, 285.4, and 288.4 eV, corresponding to C-C, C-O-C, and C=O species, respectively. All three samples exhibit similar features, which are attributed to adventitious carbon and surface residues arising from air exposure or synthesis-related residues^[36]. Collectively, the XPS findings confirm the effective incorporation of S and W into Co(OH)₂/NF and reveal associated electronic structure modulation and local coordination changes, which are beneficial for enhancing bifunctional HER and UOR electrocatalytic performance.

The HER activity was initially evaluated in 1.0 M KOH using a three-electrode setup at room temperature. LSV reveals clear performance differences among the fabricated samples, as illustrated in [Figure 3A]. The LSV results suggest that bare NF demonstrates poor HER performance, while the single-doped and pristine samples display reasonable performance levels. In contrast, S,W-Co(OH)₂/NF requires only an overpotential of 70 mV to achieve a current density of 10 mA cm^-2 (vs. RHE). The HER performance of the dual-doped catalyst was optimized by tuning the W content and electrodeposition time, which led to improved catalytic activity. This was achieved by synthesizing S,W_x-Co(OH)₂/NF with tungsten chloride-to-total metal ion molar ratios of x = 0.01, 0.05, and 0.1. As shown in [Supplementary Figure 6], the S,W_0.05-Co(OH)₂/NF electrode, with an intermediate W concentration, exhibits superior electrochemical performance compared to the other variants. In addition, LSV curves for S,W-Co(OH)₂/NF deposited for 10, 20, and 30 min are presented in [Supplementary Figure 7]. Notably, the sample deposited for 30 min exhibits substantially improved HER performance compared to those prepared at shorter deposition times. The variation in charge transfer resistance (R_ct) among different samples originates from differences in dopant concentration and structural disorder induced during electrodeposition. No additional crystalline phases were detected by XRD, and XPS confirms lattice incorporation rather than phase segregation. The optimized co-doped sample achieves a balance between defect generation and electronic conductivity, resulting in minimized R_ct. Furthermore, the Tafel slope was used to evaluate reaction kinetics across different catalysts, providing insight into their electrocatalytic efficiency and mechanisms. As shown in [Figure 3B], the S,W-Co(OH)₂/NF electrode exhibits the lowest Tafel slope (25 mV dec^-1), which is significantly lower than those of S-Co(OH)₂/NF (83 mV dec^-1), W-Co(OH)₂/NF (167 mV dec^-1), and pristine Co(OH)₂/NF (117 mV dec^-1). This considerable drop indicates that the dual-doped catalyst allows for more efficient HER kinetics. Specifically, S,W-Co(OH)₂/NF requires η = 70, 117, and 136 mV at j = 10, 50, and 100 mA cm^-2, respectively (vs. RHE), as illustrated in [Figure 3C]. These outcomes are markedly inferior to those recorded for S-Co(OH)₂/NF (134, 191, and 217 mV), W-Co(OH)₂ (241, 305, and 334 mV), and Co(OH)₂ (272, 366, and 419 mV). As summarized in [Supplementary Table 2], the dual-doped catalyst requires a lower overpotential at 10 mA cm^-2 and exhibits faster HER kinetics than many previously reported catalysts. EIS provides additional evidence of the improved interfacial charge-transfer capability of the dual-doped catalyst. As presented in [Supplementary Figure 8], this catalyst shows the smallest charge transfer resistance (R_ct = 7.4 Ω), markedly lower than those of S-Co(OH)₂/NF (15.0 Ω), W-Co(OH)₂/NF (60.1 Ω), and Co(OH)₂/NF (28.0 Ω). These findings confirm that dual doping enhances electron transport efficacy at the electrolyte -electrode interface. Beyond intrinsic kinetics, the catalytic performance also benefits from an increased density of electrochemically accessible active sites. ECSA is an important parameter for evaluating catalyst performance. In this work, ECSA was estimated from C_dl, which was derived from CV measurements at various scan rates [Supplementary Figure 9]. S,W-Co(OH)₂/NF electrode exhibits a C_dl of 111 mF cm^-2, which is much higher than those of S-Co(OH)₂/NF (61 mF cm^-2), W-Co(OH)₂/NF (4 mF cm^-2), and Co(OH)₂/NF (8 mF cm^-2). This large C_dl value suggests that S,W-Co(OH)₂/NF possesses more exposed electrochemically active sites, as shown in [Figure 3D]. To effectively compare the intrinsic HER activity across various catalysts, the current density was normalized to ECSA. The normalized LSV curves, as illustrated in [Supplementary Figure 10], show that S,W-Co(OH)₂/NF exhibits superior HER activity compared to the other catalysts.

These findings indicate that W doping not only increases the number of active sites in S-Co(OH)₂/NF but also significantly enhances its intrinsic HER activity. Moreover, long-term durability is critical for electrocatalysts in practical applications and is evaluated using various techniques. As shown in [Figure 3E and Supplementary Figure 11], the LSV curves and Nyquist plots of S,W-Co(OH)₂/NF demonstrate no significant alterations after 5,000 and 10,000 cycles, indicating excellent stability. The variation in Nyquist plots after cycling can be attributed to surface reconstruction and electrochemical activation during prolonged operation. Such structural evolution is common in TMHs under alkaline conditions. Despite changes in interfacial resistance, the polarization curves remain nearly unchanged, indicating the formation of a stable and catalytically active surface phase. Furthermore, the voltage remains stable without noticeable degradation during chronoamperometric measurement at -80 mA cm^-2 for approximately 100 h, as illustrated in [Supplementary Figure 12]. This outstanding performance underscores the potential usefulness of S,W-Co(OH)₂/NF in hydrogen production.

The electrochemical performance of UOR was assessed in 1 M KOH containing 0.5 M urea using a standard three-electrode configuration. The deposition duration and W loading of S,W-Co(OH)₂/NF were systematically optimized to enhance catalytic performance. Analysis of the LSV curves indicates that a deposition time of 30 min is optimal [Supplementary Figure 13]. Meanwhile, compositional optimization of S,Wx-Co(OH)₂/NF (x = 0.01, 0.05, and 0.1) reveals that S,W_0.05-Co(OH)₂/NF shows the best catalytic activity and was therefore selected for further investigation [Supplementary Figure 14]. The S,W-Co(OH)₂/NF catalyst achieves a current density of 100 mA cm^-2 for UOR at 1.37 V vs. RHE, as depicted in [Figure 3F]. Notably, this potential is 190 mV lower than that required for OER at the same current density, indicating that UOR is thermodynamically more favorable than OER. Furthermore, as shown in [Figure 3G], S,W-Co(OH)₂/NF achieves current densities of 200 and 500 mA cm^-2 at potentials of 1.41 and 1.45 V vs. RHE, respectively. These findings demonstrate a clear reduction compared with single-doped and pristine catalysts. As illustrated in [Supplementary Figure 15], UOR reaches current densities of 10, 50, and 100 mA cm^-2 at applied potentials of 1.02, 1.32, and 1.37 V vs. RHE, respectively. These results surpass the performance of most reported UOR catalysts, as detailed in [Supplementary Table 3]. For systematic comparison, the dual-doped catalyst was examined alongside the corresponding single-doped materials, the pristine catalyst, and the bare NF. As shown in [Supplementary Figure 16], kinetic analysis shows that S,W-Co(OH)₂/NF exhibits a Tafel slope of only 13 mV dec^-1, which is substantially lower than those of S-Co(OH)₂/NF (66 mV dec^-1), W-Co(OH)₂/NF (21 mV dec^-1), Co(OH)₂/NF (45 mV dec^-1), and bare NF (93 mV dec^-1). These results indicate that the dual-doped catalyst delivers faster electrocatalytic kinetics toward UOR. In addition, the R_ct of S,W-Co(OH)₂/NF is significantly lower than that of the other catalysts, as presented in [Figure 3H], suggesting enhanced electrode kinetics for UOR. The LSV curves of the dual-doped catalyst were also recorded in 1 M KOH with different urea concentrations, as shown in [Supplementary Figure 17]. The results demonstrate that the current density increases with increasing urea concentration from 0.1 M to 1 M. To ensure sufficient urea availability during long-term stability testing, a concentration of 0.5 M urea in 1.0 M KOH was selected. Chronoamperometric measurements reveal that S,W-Co(OH)₂/NF maintains a current density of 80 mA cm^-2 over 100 h, as shown in [Figure 3I], confirming its excellent long-term stability for UOR. The ECSA of the catalysts was evaluated in a urea-containing electrolyte by estimating the C_dl from CV at different scan rates, as depicted in [Supplementary Figure 18]. S,W-Co(OH)₂/NF demonstrates a C_dl value of 74 mF cm^-2, which is significantly higher than those of S-Co(OH)₂/NF (38 mF cm^-2), W-Co(OH)₂/NF (3 mF cm^-2), and NF (2 mF cm^-2), as shown in [Supplementary Figure 19].

Because of its excellent activity toward both HER and UOR, overall water and urea splitting devices were constructed using two identical dual-doped catalyst electrodes as the cathode and anode, respectively. In this section, “cell voltage” refers to measurements in a two-electrode configuration, whereas “potential vs. RHE” refers to three-electrode measurements. This setup is designed to assess practical applicability, as illustrated in [Figure 4A]. The HER LSV curves of S,W-Co(OH)₂/NF remain nearly identical in the absence and presence of 0.5 M urea in the electrolyte, suggesting that urea has a negligible effect on HER performance [Supplementary Figure 20]. Notably, this dual-electrode system exhibits exceptional activity for overall urea-assisted electrolysis under alkaline conditions, as demonstrated in [Figure 4B]. For urea-assisted electrolysis, the assembled device requires only 1.19 V to reach 10 mA cm^-2, whereas conventional water splitting requires 1.54 V under identical conditions. This performance is comparable to previously reported electrocatalysts for urea-aided water splitting, as summarized in [Supplementary Table 4]. Based on the cell voltage at 10 mA cm^-2, the energy consumption of urea-assisted electrolysis is reduced by approximately 23% compared with conventional water electrolysis. The detailed calculation is provided in the Supporting Information.

To enable electrolysis at higher current densities of 50, 100, and 150 mA cm^-2, as depicted in [Figure 4C], the required cell voltages for water decomposition are 1.75, 1.85, and 1.93 V, respectively. In contrast, the corresponding voltages for urea electrolysis are markedly reduced to 1.52, 1.61, and 1.68 V under the same conditions. This difference highlights the system’s potential for energy-efficient hydrogen generation and the utilization of urea-rich wastewater. While catalytic activity is important, stability over time is a critical factor for practical catalyst applications. Notably, the S,W-Co(OH)₂/NF electrode demonstrates excellent stability, maintaining effective urea electrolysis over 24 h despite a significant reduction in current density, as illustrated in [Figure 4D]. These results emphasize the robust stability of the catalyst. Collectively, these findings indicate that the dual-functional S,W-Co(OH)₂/NF displays exceptional catalytic efficacy in concurrent water and urea electrolysis, making it a promising candidate for hydrogen evolution systems.

Post-HER and post-UOR characterizations of the dual-doped catalyst were conducted using XRD, SEM, and XPS. The XRD pattern in [Supplementary Figure 21] confirms the presence of the Co(OH)₂ phase. SEM images [Supplementary Figure 22] verify the electrode's exceptional endurance. The XPS data [Supplementary Figure 23] indicate enhanced chemical stability after electrolysis. These findings demonstrate outstanding structural and chemical stability, further highlighting the superior performance of S,W-Co(OH)₂/NF for urea-assisted water splitting compared to current catalysts.

To gain deeper insight into the intrinsic relationship between the electronic structure of the doped hydroxide system and its catalytic behavior, DFT calculations were performed on representative single-doped and dual-doped model structures, as illustrated in [Figure 5A-C]. Differential charge density maps [Supplementary Figure 24] reveal the complementary electronic effects of S and W dopants in Co(OH)₂. S doping induces charge accumulation at adjacent Co sites and charge depletion around nearby O atoms, indicating electron donation that enhances Co-O covalency. In contrast, W doping causes electron depletion around the W-O unit and charge accumulation on neighboring Co atoms, reflecting electron withdrawal by high-valence W and the resulting local lattice polarization. The dual-doped S,W-Co(OH)₂ [Figure 5D] integrates these effects, producing coupled charge accumulation-depletion regions and long-range polarization. This redistributes electron density by enriching certain Co sites while creating electron-deficient Co-O-W domains, a built-in electronic asymmetry that facilitates urea activation and oxygenation steps, thereby accelerating UOR kinetics^[37]. The observed charge redistribution around Co sites after W and S co-doping modifies the local electronic structure, which directly influences adsorption energetics. For HER, the optimized electronic configuration adjusts the hydrogen adsorption free energy (ΔG_H*) toward a more favorable value, thereby accelerating reaction kinetics. For UOR, the redistributed charge enhances interactions with reaction intermediates and facilitates electron transfer. These theoretical results are consistent with the experimentally observed reduced overpotential and lower charge transfer resistance. To further elucidate the effect of S and W incorporation on the electronic configuration, the projected density of states (PDOS) of the modeled structures was examined. The corresponding d-band centers relative to the Fermi level (E_F) are calculated to be -0.258, -0.293, and -0.270 eV for the S-doped, W-doped, and dual-doped structures, respectively [Figure 5E and Supplementary Figure 25]. The S-doped sample exhibits the d-band center closest to E_F, the W-doped sample the farthest, while the dual-doped sample lies in between, indicating a finely tuned intermediate adsorption strength consistent with its balanced catalytic behavior^[17,38]. In the context of HER, it is well established that ΔG_H∗ is the pivotal parameter, with the optimal value ideally approaching zero^[39]. As depicted in [Figure 5F], S,W-Co(OH)₂ delivers a near-thermoneutral ΔG_H* of 0.37 eV, consistent with its fastest HER kinetics. In contrast, W-Co(OH)₂ binds H^∗ too weakly, while S-Co(OH)₂ exhibits intermediate binding. These trends are consistent with the DCD/PDOS results: S doping enriches Co 3d states, while W induces electron withdrawal via W-O bonds; dual doping synergistically integrates these effects to achieve near-ideal H^∗ adsorption. Free-energy profiles [Figure 5G] identify dehydrogenation as the rate-determining step (RDS). The highest energy barrier for S,W-Co(OH)₂ is 1.80 eV, lower than those of S-Co(OH)₂ (1.99 eV) and W-Co(OH)₂ (1.91 eV). This reduction is linked to dual doping-induced electronic polarization; electron-rich Co sites promote urea activation/dehydrogenation, while electron-deficient Co-O-W domains stabilize oxygenated intermediates, collectively lowering the reaction barrier. The urea oxidation pathway and corresponding optimized configurations for the single and dual-doped models are shown in [Figure 5H and Supplementary Figure 26]. Overall, the combined experimental and theoretical findings demonstrate that synergistic anion-cation co-doping is an effective electronic structure modulation strategy for developing highly active bifunctional catalysts for water and urea splitting.