Electronic Modulation of Nickel Cobalt Phosphide Nanosheets by Ce Doping for Efficient Overall Water Splitting

Abstract

Efficient hydrogen (H₂) generation from electrochemical overall water splitting (OWS) is key to a sustainable H₂ economy. Low‐cost transition metal‐based catalysts, such as Ni‐ and Co‐based phosphides, have gained attention for the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) due to their excellent corrosion resistance and high electrical conductivity. In particular, bimetallic Ni and Co‐based phosphide catalysts are considered highly efficient electrocatalysts for OWS due to their abundant adsorption sites and low adsorption energy for hydrogen species. However, improving their stability and activity remains challenging. Herein, a Ce doping NiCo phosphide catalyst is presented with vary Ce amount (Ce_x‐NiCoP) supported on nickel foam (NF) with multi‐site functionality, achieving highly efficient HER performance comparable to benchmark platinum catalysts support carbon fiber or NF (Pt/C or Pt/NF). Comprehensive characterization results show that the optimal amount of Ce doping significantly influences the electronic structure of the catalyst, preventing the formation of Ni₅P₄ and CeO₂, promoting the dominant NiCoP phase. This modification enhances the catalyst's hydrophilicity, improving the HER activity significantly. Remarkably, the catalyst also demonstrates exceptional OER performance, making it a highly active and stable bifunctional catalyst for OWS, with the highest energy efficiency of 96.7%.

This study addresses the insufficient activity of transition metal phosphide catalysts, a Ce‐doped NiCoP bifunctional catalyst is designed. By modulating the electronic structure to suppress impurity phase formation, this catalyst exhibits significantly enhanced hydrophilicity, HER/OER activity, and stability. The overall energy efficiency for water splitting achieves 96.7%, surpassing many reported advanced non‐noble metal catalysts.

1. Introduction

Electrochemical overall water‐splitting (OWS) has been considered as one of the most promising approaches to store renewable electricity in the form of hydrogen fuel.^{[ 1} , ^{2 ]} To achieve efficient water splitting, it is crucial to design electrocatalysts that exhibit both high activity and stability, effectively driving the anodic oxygen evolution reaction (OER) and the cathodic hydrogen evolution reaction (HER), thereby enhancing the overall energy conversion efficiency in electrochemical water electrolysis.^{[ 3 ]} However, the high half‐reaction energy barrier in water electrolysis presents a significant challenge for the large‐scale application of hydrogen energy. Currently, nanoparticles based on Pt, Ru, Ir and their derivatives^{[ 4} , ⁵ , ^{6 ]} are considered the most effective electrocatalysts for both HER and OER. However, their scarcity severely limits their feasibility for large‐scale application. Moreover, the use of two distinct electrode materials not only increases costs but also complicates the design and operation of electrolysis devices.^{[ 7} , ^{8 ]} Consequently, there has been growing research into development of cost‐efficient bifunctional electrocatalysts with excellent catalytic activity and stability, aiming to simultaneously enhance the kinetics of both the HER and OER, which enables efficient and low‐cost water splitting for hydrogen production.^{[ 9 ]}

In recent years, transition metal phosphides (TMPs) have emerged as one of the most promising materials for water electrolysis catalysts due to their excellent corrosion resistance, good electrical conductivity, and low cost.^{[ 10 ]} A wide range of transition metals can be phosphorylated to form TMPs, where the flexible phosphorus skeleton between metal and phosphorus atoms creates numerous M‐M, P‐P, and M‐P bonds, resulting in a variety of compositions and structures. Among these, NiP^{[ 11 ]} and CoP^{[ 12 ]} stand out in the field of HER due to it facilitates hydrogen adsorption and conjugation. However, their OER activity remains less satisfactory. This discrepancy is largely due to the higher energy barrier associated with the four‐electron transfer reaction in the OER, compared to the two‐electron transfer reaction in the HER.^{[ 13 ]} Previous studies^{[ 14 ]} have demonstrated that the bimetallic NiCo phosphide catalyst is considered as highly efficient electrocatalysts for HER and OER, due to its rich adsorption sites and a low adsorption energy for H species. Nevertheless, the catalytic activity and stability of these inexpensive catalysts still require significant improvement. Typically, the activity of a catalyst is closely related to its surface electronic structure. Elemental doping is an effective method for modulating the surface electronic structure of TMPs, boosting their HER and OER activities as well as stability. Furthermore, doping can adjust the covalent bonds between metals and phosphorus, optimizing the adsorption process of reaction intermediates and further enhancing the intrinsic activity and stability of the catalyst.

Cerium (Ce), a rare earth element, can regulate the d‐electrons of transition metal in TMPs via its 4f and 5d orbital occupation. This unique property enhances electrocatalytic processes. Moreover, the incorporating Ce into TMPs enhances the structural stability of the catalyst due to the strong interaction between the 3d orbitals of transition metals and the 5d orbitals of Ce.^{[ 15 ]} Beyond structural stability, the flexible oxidation states transition between the Ce³⁺ and Ce⁴⁺ also increases the number of active sites, further improving catalytic performance. Notably, studies have shown that Ce‐doped CoP facilitates electron transfer from Ce to Co, significantly reducing the energy barrier associated with the rate‐limiting step in electrochemical water splitting and thereby optimizing the overall water splitting performance.^{[ 16 ]} Additionally, among all rare earth elements, Ce exhibits exceptional oxygen affinity,^{[ 17 ]} enabling to adsorb intermediates effectively during the OER process and further facilitating the OER performance. Based on these properties, Ce doping in TMPs introduces a regulatory mechanism that optimizes both OER and HER performances in alkaline media.

Herein, Ce‐doped P‐based nickel‐cobalt phosphide catalysts (Ce‐NiCoP) supported on nickel foam (NF) were synthesized via a simple solvothermal method followed by low‐temperature phosphorization. Ce doping plays a crucial role in tailoring the lattice structure and electronic configuration while simultaneously facilitating the in situ formation of more hydrophilic NiCoP phases. By tuning the Ce concentration, the impact of Ce on the structure and activities of Ce‐NiCoP catalysts was systematically investigated. The significantly enhanced catalytic kinetics of both the HER and OER and the OWS after Ce doping could be attributed to improved coverage of key intermediates, optimized lattice and electronic structures, and a unique microscopic morphology.

2. Results and Discussions

2.1. Synthesis and Characterizations

As shown in Figure 1a, these Ce_x‐NiCoP catalysts with varied Ce amount (5, 10, and 15 mol%), denoted as Ce_0.05‐NiCoP, Ce_0.10‐NiCoP and Ce_0.15‐NiCoP, were synthesized through a facile hydrothermal method followed by a low‐temperature phosphating treatment process at 300 °C. For comparison, NiCoP without Ce doping was also prepared using same procedure. The NiCoP exhibited a prominent blocky, stacked sheet morphology with an average diameter of ≈200 nm (Figure 1b). After Ce doping, the nanosheet size decreased (Figure 1c), and with the increase of Ce doping from 5 to 10 mol%, the nanosheets structure was transformed into nanorod structures (Figure 1c,d). Further increasing the Ce amount to 15 mol% (Figure 1e) resulted in a rougher catalyst surface. High‐resolution transmission electron microscopy (HRTEM) images of Ce_0.10‐NiCoP reveal the coexistence of crystalline (Figure 1g) and amorphous structures (Figure 1f highlighted with yellow circles) in the catalyst. Lattice fringes with a spacing of 0.22 nm, corresponding to the (111) plane of NiCoP, were observed in crystalline regions (Figure 1g,h). In addition, further analysis through fast Fourier transform (FFT) indicated the presence of lattice distortions and curves in localized regions (Figure 1i,j). These structural modifications, likely induced by Ce incorporation (Figure 1i), could increase defect density, thereby enhancing electrocatalytic performance.^{[ 18 ]} The selected‐area electron diffraction (SAED) pattern further confirms the presence of NiCoP^{[ 19 ]} (Figure 1k). Next, the HADDF image of Ce_0.10‐NiCoP (Figure 1l) further exhibits its mixed structures of rod‐like and sheet‐like, which is consistent with the SEM result (Figure 1d). Additionally, EDS elemental mapping confirmed the homogeneous distribution of all elements (Ni, Co, Ce, P, and O) in Ce_0.10‐NiCoP catalyst.

Figure 1.

a) The synthesis procedure illustration of Ce_x‐NiCoP catalysts supported on a clean nickel foam (NF) substrate; b–e) SEM images of NiCoP, Ce_0.05‐NiCoP, Ce_0.10‐NiCoP, and Ce_0.15‐NiCoP catalysts, respectively; f) TEM image of Ce_0.10‐NiCoP, g) HRTEM image of Ce_0.10‐NiCoP showing lattice fringes, h) interplanar spacing corresponding to the NiCoP (111) plane, i) normal lattice plane after Fourier transform, j) distorted lattice plane in a local region, k) SAED pattern of Ce_0.10‐NiCoP, and l) HDDAF image and EDS elemental mapping images of Ce_0.10‐NiCoP.

The crystal structures of these catalysts were then characterized by X‐ray diffraction (XRD) analysis. As shown in Figure 2a, the main diffraction peaks for NiCoP and Ce doping samples(x = 0.05–0.15) correspond to the characteristic peaks of pure NiCoP phase (JCPDS #71‐2336) and NF (JCPDS #70‐0989). This suggests that the dominant phase in all samples is NiCoP. Notably, the obvious phase of Ni₅P₄ (JCPDS #71‐2336) was observed on NiCoP samples without Ce doping. With the increase of Ce doping amount, the intensity of these characteristic peaks of Ni₅P₄ decreased and vanished when Ce amount was 10 mol%. This suggests that a certain amount of Ce could be incorporated into the crystal lattice of sample, changing the crystal structure. The absence of CeO₂ phase (JCPDS #71‐2336) on Ce_0.10‐NiCoP indicates that Ce is uniformly dispersed in the sample. However, further increasing the Ce amount led to the emerging of two distinct peaks at 28.55° and 33.08°, which can be attributed to the (111) and (200) planes of CeO₂ (JCPDS #34‐0394). This is because the high amount of Ce doping results in the formation of CeO₂ particles on the catalyst. Besides, the peak intensities of the corresponding NiCoP phases after Ce doping were significantly reduced compared to the sample without Ce doping, suggesting that Ce doping effectively decreases the crystallinity of the catalyst.^{[ 20 ]} This crystallinity reduction agrees with HRTEM results showing amorphous regions (Figure 1g).

Figure 2.

a) XRD patterns of NiCoP and Ce_x‐NiCoP catalysts; b) full XPS spectrum of Ce_0.10‐NiCoP; c–e) XPS spectra of Ni 2p, Co 2p, and P 2p, respectively, for NiCoP and Ce_0.10‐NiCoP; f) XPS spectrum of Ce 3d for Ce_0.10‐NiCoP.

X‐ray Photoelectron Spectroscopy (XPS) was utilized to analyze the surface composition and chemical state of materials. The XPS survey spectrum of Ce_0.10‐NiCoP (Figure 2b) confirms the coexistence of Ce, Ni, Co, P, and O, consistent with the EDS mapping (Figure 1l). The XPS spectra of Ni 2p for NiCoP and Ce_0.10‐NiCoP samples (Figure 2c) show three prominent sets of peaks, which are corresponded to Ni^δ+ (0 < δ < 1) (853.5 and 870.9 eV), Ni²⁺ (857.3 and 875.1 eV), and satellite peaks (861.4 and 880.4 eV).^{[ 21 ]} Compared to NiCoP, the Ni²⁺ peak on Ce_0.10‐NiCoP shifted to a lower binding energy by ≈0.8 eV. In the Co 2p spectra (Figure 2d), the peaks at 782.1 and 798.1 eV are ascribed to Co²⁺, while these peaks at 778.6 and 792.7 eV are assigned to Co‐P bonds.^{[ 22} , ^{23 ]} Similarly, the main Co²⁺ peak in Ce_0.10‐NiCoP also shifted by 0.2 eV toward a lower binding energy. A similar trend was observed in the P 2p spectra, where the peak attributed to P‐O bond shifted to a lower binding energy after Ce doping (Figure 2e). This shift may originate from charge transfer between Ce and Ni and/or Co, forming Ni^δ+ and Co^δ+ species. After deconvolution, The Ce 3d XPS spectrum of Ce_0.10‐NiCoP sample (Figure 2f) shows the distinct peaks at 881.7, 886.7, 891.9, 895.5, 902.7, and 904.9 eV that are attributed to Ce⁴⁺, whereas the peaks at 884.5 and 900.1 eV ascribed to Ce³⁺.^{[ 24 ]} In summary, the shifts of the Ni, Co, and P peaks toward lower binding energies can be attributed to the electron‐donating behavior of Ce, which redistributes charges in the TMPs.^{[ 25 ]} As a result, the potential around Ni, Co, and P elements changed, causing their binding energies to shift. This alteration in the electronic structure is beneficial for the water electrolysis reaction.^{[ 26 ]}

2.2. Electrochemical HER Performance

Figure 3a shows the HER activity of NiCoP and Ce_x‐NiCoP obtained in 1.0 M KOH by using linear sweep voltammetry (LSV) at a scan rate (v) of 5 mV s⁻¹. NF was used as substrate due to its excellent electrochemical properties, e.g. corrosion resistance, electrical conductivity, 3‐D porous structure, and robust mechanical strength.^{[ 27} , ^{28 ]} The Ce doping significantly enhances the HER performance, achieving an onset potential close to 0 V (vs RHE), accompanied by the visible formation of hydrogen bubbles on the electrode surface. Notably, Ce_0.10‐NiCoP exhibited an exceptionally low HER overpotential (η), requiring only 96 mV to attain a current density (j) of 10 mA cm⁻². In contrast, Ce_0.05‐NiCoP (154 mV), Ce_0.15‐NiCoP (144 mV), and Ce‐free NiCoP (239 mV) require significantly higher overpotentials to reach the same current density (Figure 3a), highlighting the superior catalytic activity of Ce_0.10‐NiCoP. Even at elevated current densities of j ₅₀ and j ₁₀₀, Ce_0.10‐NiCoP maintained relatively low overpotentials of 166 and 192 mV, respectively (Figure 3b). This trend was further supported by electrochemical active surface area (ECSA)‐normalized LSV curves (Figure S1, Supporting Information) and a 3.48‐fold higher double‐layer capacitance (C_dl) compared to pristine NiCoP. Moreover, under identical catalyst loading conditions, Ce_0.10‐NiCoP demonstrated HER activity comparable to that of the benchmark Pt catalyst (20% Pt/C) supported on NF and surpassed various control catalysts prepared under the same conditions but with different compositions, including Ce‐NiCoO, Ce‐NiP, Ce‐CoP, CeP, CoP, NiP, and CeO₂ (Figure S2, Supporting Information). These findings underscore the unique catalytic properties of Ce_0.10‐NiCoP with dominant NiCoP phase, and its optimized composition and electronic structure. To quantify the activity enhancement, we evaluated mass activity (MA, Figure 3c) and turnover frequency (TOF, Figure 3d). Besides, as shown in Figure 3e, the TOF value of Ce_0.10‐NiCoP at 150 mV is 9 times higher than that of the Ce‐free NiCoP counterpart, indicating that Ce doping enhanced HER activity significantly.

Figure 3.

The HER performance of the prepared catalysts in 1 m KOH electrolyte is presented as follows: a) LSV curves for Pt/C, Ce_0.05‐NiCoP, Ce_0.10‐NiCoP, Ce_0.15‐NiCoP, and NiCoP catalysts; b) Corresponding overpotential values at 10, 50, and 100 mA cm⁻²; and c) Massic activity; d) the TOF values of Ce_0.10‐NiCoP and NiCoP; e) the TOF values at overpotentials of 50, 100, and 150 mV; f) Tafel slopes; g) EIS curves; h) C_dl values; i) The stability test (i‐t test) of Ce_0.10‐NiCoP during HER operation for 100 h at 20 mA cm⁻².

The Tafel plot (Figure 3f) also demonstrated the superior HER performance of Ce_0.10‐NiCoP with the Tafel slope of 94.75 mV dec⁻¹, which is significantly lower than those of Ce‐free NiCoP, Ce_0.05‐NiCoP, Ce_0.15‐NiCoP, and Pt/C. This indicates that Ce doping accelerates the HER reaction kinetics and follows the Volmer‐Heyrovsky mechanism during the HER process.^{[ 29 ]} The charge transfer kinetics of the catalysts were investigated through electrochemical impedance spectroscopy (EIS) measurements. The lowest charge transfer resistance (Rct) was observed on Ce_0.10‐NiCoP compared to other catalysts (Figure 3g), indicating the fastest charge transfer rate, which is consistent with the trend of high activity. Figures S3 and S4 (Supporting Information) display cyclic voltammetry (CV) curves recorded within specific voltage ranges, which were utilized to determine the electrochemical double‐layer capacitance (C_dl) of each sample as an indicator of their ECSA. As illustrated in Figure 3h, Ce_0.10‐NiCoP exhibited the highest C_dl value of 4.52 mF cm⁻², substantially surpassing those of Ce_0.05‐NiCoP (1.72 mF cm⁻²), Ce_0.15‐NiCoP (2.18 mF cm⁻²), and NiCoP (1.30 mF cm⁻²). These results suggest that Ce_0.10‐NiCoP possesses a significantly enhanced electrochemically active surface, which is expected to contribute to its superior catalytic performance. This higher C_dl of Ce_0.10‐NiCoP implies that the catalyst possesses a larger active surface area with more exposed active sites, which could be attributed to its rougher and more hydrophilic surface, further confirming the structure‐activity relationship.^{[ 30 ]} Catalyst stability was also evaluated through an i‐t test at j _HER = –20 mA cm⁻² (Figure 3i), revealing negligible decay in current density within the first 50 h and an observed decay of ≈8.25% after 100 h of reaction. The inset in Figure 3i displays the HER LSV curves before and after the stability test, showing an increase in overpotential by only 12 mV at j ₁₀. These findings indicate that an appropriate amount of Ce doping enhanced electrochemical stability significantly in HER.

2.3. Electrochemical OER Performance

The OER catalytic performance of the samples was evaluated following the same testing protocol used for HER. However, to eliminate interference from oxidation peaks that significantly affect overpotential measurements at low current densities, reverse scanning was conducted within the voltage range during LSV testing. In the OER process, the oxidation peak is typically attributed to the single‐electron transfer process between Ni³⁺ and Ni²⁺, while the reduction peak corresponds to the reverse transition.^{[ 31} , ^{32 ]} As shown in Figure 4a, the LSV curve of Ce_0.10‐NiCoP exhibited a significantly enhanced reduction peak, with an integrated peak area ∼10‐fold larger than that of Ce‐free NiCoP. Additionally, the reduction peak was shifted by 0.17 V, indicating a higher density of active sites and an accelerated electron transfer rate in the Ce‐doped sample. The enhanced redox kinetics and electron transfer rate directly contribute to its superior OER performance, with Ce_0.10‐NiCoP achieving a current density of 10 mA cm⁻² at a low overpotential of 218 mV. Furthermore, higher current densities of 50 mA cm⁻² and 100 mA cm⁻² were attained at overpotentials of 271 and 310 mV, respectively (Figure 4b). These values surpassed those of Ce‐free NiCoP, RuO₂ (Figure 4b), and other control catalysts, including Ce‐NiCoO, Ce‐NiP, Ce‐CoP, CeP, CoP, NiP, and CeO₂ (Figure S5, Supporting Information). The trend in oxygen evolution mass activity across these catalysts (Figure 4c) aligned well with the LSV results presented in Figure 4a. Additionally, TOF of Ce_0.10‐NiCoP at an overpotential of 350 mV was calculated to be 0.0204 s⁻¹, ≈14 times higher than that of Ce‐free NiCoP (Figure 4d,e), further demonstrating the significant catalytic enhancement induced by Ce doping.

Figure 4.

The OER performance of the prepared catalysts in 1 M KOH electrolyte. a) LSV curves, b) Corresponding overpotential values at 10, 50, and 100 mA cm⁻², and c) Massic activity; d) The TOF values of Ce_0.10‐NiCoP and NiCoP;e) the TOF values at overpotentials of 50, 100, and 150 mV; f) Tafel slopes; g) EIS curves; h) C_dl values; i) The stability of Ce_0.10‐NiCoP during OER operation for 50 h is tested using the i‐t method.

In terms of reaction kinetics, Ce_0.10‐NiCoP had a Tafel slope of 72.1 mV dec⁻¹ (Figure 4f), which is lower than that of RuO₂ (80.8 mV dec⁻¹), Ce_0.05‐NiCoP (151.9 mV dec⁻¹), Ce_0.15‐NiCoP (99.1 mV dec⁻¹), and NiCoP (102.6 mV dec⁻¹). This indicates fast electron transfer and efficient mass transport of the catalyst.^{[ 33 ]} EIS measurements further confirmed the improvement in its OER kinetics, as the Rct value of Ce_0.10‐NiCoP was significantly smaller than that of other catalysts (Figure 4g), suggesting optimized charge transfer kinetics at the electrode/electrolyte interface. Additionally, the C_dl value of Ce_0.10‐NiCoP (3.76 mF cm⁻²) was much higher than that of the other control groups, correlating with its high OER activity (Figure 4h). Lastly, excellent stability of the Ce_0.10‐NiCoP for prolonged OER was illustrated in Figure 4i. There is no significant change to OER activity (insert in Figure 4i), compared to fresh Ce_0.10‐NiCoP catalyst.

2.4. Using Methanol Molecules as Probes to Detect the OH* Intermediate in the OER

It is widely accepted that OH* is the first intermediate during the OER process in KOH and oxidizing active species such as OH*, O*, and OH*(O^I) are formed step by step on the catalyst surface. As shown in Figure 5a, for TMPs, the deprotonation of OH* to O* (Step 1) and the formation of OOH* (Step 2) typically involve higher energy barriers compared to other reaction steps, which results in the slow reaction kinetics of OER.^{[ 34 ]} Therefore, a high, coverage of OH* on the catalyst surface could facilitate the OER reaction, and the oxygen atoms in OH* are electron‐deficient, endowing the species with electrophilic properties. Due to the nucleophilic nature of methanol molecules and their significantly lower oxidation potential compared to water, OH* preferentially binds with methanol, initiating methanol as the probe molecule to detect the active intermediate OH* (Figure 5b). A high concentration of OH* on the catalyst surface significantly enhances the MOR reaction, resulting in a clear distinction between the MOR and OER curves.

Figure 5.

a) Schematic diagram of the energy levels of intermediates during the OER process, b) Typical diagrams illustrating the reaction pathways of OER and MOR, c) and d) the CV curves of Ce_0.10‐NiCoP and NiCoP, respectively, in 1 M KOH and 1 M KOH + 0.01 M CH₃OH, e) and f) show the LSV curves and Tafel plots of Ce_0.10‐NiCoP and NiCoP, respectively, in 1 M KOH + 0.01 M CH₃OH. All of the above curves have not been corrected for iR.

Figure 5c,d show the CV curves of NiCoP and Ce_0.10‐NiCoP catalysts tested with and without 0.01 M methanol, respectively. The green‐shaded areas corresponding to the difference of current density reflect the OH* accumulation during OER reaction. Notably, the potential difference between the anodic peak and the onset potential decreased significantly from 84 to 39 mV after Ce doping, indicating intensified OH* consumption and improved OER performance_. ^{[ 35} , ^{36 ]} The Tafel slope of Ce_0.10‐NiCoP is lower than that of NiCoP in both the low‐current region for MOR and the high‐current region for OER (Figure 5e,f), further confirming the superior kinetic performance of NiCoP after Ce doping. Collectively, these results suggest that Ce doping effectively enhances the adsorption capacity of active intermediates OH* and accelerates their subsequent reaction steps, which play a crucial role in OER performance.^{[ 37 ]}

2.5. Electrochemical OWS Performance

In water electrolysis, the hydrophilicity of the catalyst surface plays a critical role in enhancing catalytic performance. A hydrophilic surface facilitates the diffusion of the electrolyte, thereby improving the accessibility of active sites to OH⁻ and H⁺, which is essential for efficient electrochemical reactions.^{[ 38} , ^{39 ]} To assess the hydrophilicity of the catalysts, contact angle (CA) measurements were conducted, as shown in Figure 6a. The measured contact angle for NiCoP was 100.3°, indicating a hydrophobic surface interface, while Ce_0.10‐NiCoP showed a significantly lower contact angle (60.3° vs 100.3°), confirming hydrophilicity. This indicates that Ce doping enhances surface hydrophilicity, likely due to the progressive roughening of the catalyst surface. Improved hydrophilicity promotes better wettability, accelerates bubble detachment, and enhances the exposure of active sites, thereby contributing to superior catalytic efficiency.^{[ 40 ]}

Figure 6.

a) Contact angle measurements of the samples; b) LSV curves of Ce_0.10‐NiCoP ||Ce_0.10‐NiCoP, NiCoP ||NiCoP, and Pt/C || |RuO₂; c) Comparison of the overall water splitting overpotentials at j₁₀ and j₅₀ between Ce_0.10‐NiCoP and other transition metal phosphide catalysts reported in the literature; d) Stability test curves of Ce_0.10‐NiCoP||Ce_0.10‐NiCoP and NiCoP || NiCoP; e) H₂ and O₂ production rates and Faraday efficiency as a function of electrolysis time.

Furthermore, Ce_0.10‐NiCoP was evaluated as both the anode and cathode in a full‐cell water electrolyzer operating in a 1 M KOH. The electrolyzer setup and a magnified view of the electrode are illustrated in Figure S6 (Supporting Information). The polarization curve of the Ce_0.10‐NiCoP‐based cell was recorded and compared to that of a benchmark system employing RuO₂ for OER and 20% Pt/C for HER. As depicted in Figure 6b, the Ce_0.10‐NiCoP cell outperforms the Pt/C||RuO₂ cell, achieving a current density of 10 mA cm⁻² at a cell voltage of only 1.536 V—lower than the 1.564 V required for Pt/C||RuO₂ counterpart. Additionally, Ce_0.10‐NiCoP cell exhibited exceptional OWS performance, surpassing most recently reported bi‐functional catalysts for alkaline water electrolysis(Figure 6c; Table S1, Supporting Information), highlighting its potential for practical applications in sustainable hydrogen production.

To further evaluate the practicality of Ce_0.10‐NiCoP for OWS, its long‐term stability was systematically investigated. As shown in Figure 6d, the Ce_0.10‐NiCoP cell showed excellent stability over 40 h of continuous operation in alkaline water electrolysis at current densities of 10 and 50 mA cm⁻².^{[ 41} , ^{42 ]} In contrast, the Ce‐free NiCoP||NiCoP cell required a higher cell voltage of 1.786 V to achieve 10 mA cm⁻², and experienced noticeable performance degradation after 20 h. Subsequently, we conducted SEM, XRD, and XPS analyses on Ce_0.10‐NiCoP after the stability test (Figure S10, Supporting Information). The characterization results indicated that after prolonged electrochemical testing, the catalyst maintained its morphology, crystalline structure, and elemental valence states well, without significant signs of degradation or aggregation. The enhanced stability of Ce_0.10‐NiCoP is attributed to the protective role of Ce, which mitigates catalyst degradation and enhances its durability in prolonged alkaline overall water splitting.^{[ 43 ]}

Additionally, the Faradaic efficiency (FE) of the Ce_0.10‐NiCoP electrode was assessed using a water displacement gas collection method (Figures S7–S9, Supporting Information), with detailed procedures provided in Supporting Information. The experimentally measured H₂ and O₂ production rates were plotted against time (Figure 6e), revealing that the H₂/O₂ ratio was close to the theoretical value of 2:1. Moreover, the calculated FE for HER and OER were ≈96.3% and 97.1%, respectively, at j = 100 mA cm⁻², indicating high electron utilization efficiency.^{[ 44 ]} Subsequently, a cost analysis was conducted (Note S1, Supporting Information), and it was found that the hydrogen production cost using Ce_0.10‐NiCoP is only $1.448/kg H₂, which is significantly lower than the 2026 target set by the United States ($2/kg H₂ for 2026). These results confirmed that the Ce_0.10‐NiCoP/NF electrode effectively facilitates water electrolysis with minimal energy loss, demonstrating its strong potential for practical H₂ and O₂ production.

3. Conclusion

In summary, a series of bifunctional Ce‐NiCoP electrolysis catalysts were successfully synthesized via an in situ growth strategy combined with a low‐temperature phosphating process. The optimal Ce_0.10‐NiCoP(10 mol% Ce) exhibited enhanced hydrophilic and electronic modulation, boosting HER and OER activity. In alkaline media, it achieved low overpotentials of 96 mV for HER and 218 mV for OER at 10 mA cm⁻², with stable activity retention over 50 h. For OWS, the Ce_0.10‐NiCoP||Ce_0.10‐NiCoP electrolyze required only 1.536 V at 10 mA cm⁻² and maintained stably for over 40 h. Furthermore, mechanistic studies revealed Ce doping promotes OH* adsorption and accelerates its conversion on Ce_0.10‐NiCoP, significantly lowering the OER reaction barrier. This work offers a fundamental understanding of the synergistic interactions between rare earth elements and transition metal phosphides, providing a promising strategy for the rational design of advanced bifunctional electrocatalysts for efficient water splitting.

4. Experimental Section

Catalysts Synthesis

The nickel foam (with a thickness of 1.0 mm) was purchased from Saibo Electrochemical Materials Company. Cobalt(II) nitrate hexahydrate (Co(NO₃)₂·6H₂O), nickel(II) nitrate hexahydrate (Ni(NO₃)₂·6H₂O), cerium(III) nitrate hexahydrate (Ce(NO₃)₃·6H₂O), potassium hydroxide (KOH), sodium hypophosphite anhydrous (NaH₂PO₂), and urea (CO(NH₂)₂) were all obtained from Aladdin Reagent Co., Ltd. Hydrochloric acid (HCl, ≈36.0%–38.0% aqueous solution) was sourced from Xilong Scientific Co., Ltd., while ethanol (CH₃CH₂OH), methanol (CH₃OH), and acetone were purchased from Tianjin Jindong Tianzheng Fine Chemical Reagent Factory. Deionized water was acquired from Hangzhou Wahaha Group Co., Ltd. RuO₂ (70% purity) is sourced from Energy Chemical Co., Ltd., while Pt/C (20% platinum content) is obtained from Macklin Biochemical Co., Ltd. All chemicals are analytical reagents without further purification.

Synthesis of Ce_x‐NiCoO Catalyst

Initially, a piece of nickel foam (NF, 1*4.5 cm²) was cleaned by ultrasonic treatment in an acetone solution for 20 min, followed by successive treatment 3 M HCl, deionized water, and ethanol solution, each performed three times. Typically, a facile strategy was employed to synthesize the Ce_0.10‐NiCoO catalyst through a solvothermal method followed by a low‐temperature phosphorylation process. In detail, 0.292 g of Ni(NO₃)₂⋅6H₂O, 0.146 g of Co(NO₃)₂⋅6H₂O, 0.065 g of Ce(NO₃)₃⋅6H₂O, and 0.276 g of CO(NH₂)₂ were dissolved in 70 mL of deionized water, and the mixture was stirred for 1 h. The pre‐treated NF and the resulting solution were then transferred to a 100 mL Teflon‐lined stainless‐steel autoclave, which was placed in an oven at 160 °C for 14 h. The as‐obtained product was subsequently washed several times with absolute ethanol and purified water, followed by drying at 60 °C for 12 h in a vacuum oven. The final product was denoted as Ce_0.10‐NiCoO, where 0.10 represents the molar ratio of Ce to Ni and Co. Other catalysts, named as Ce_0.05‐NiCoO, Ce_0.15‐NiCoO, and NiCoO were synthesized following the same procedure, with the Ce(NO₃)₃⋅6H₂O amount, adjusted to 0.033 g, 0.098 g, and 0 g, respectively.

Synthesis of Ce_x‐NiCoP Catalyst

The Ce_0.10‐NiCoP catalyst was prepared through a low‐temperature phosphidation method. Specifically, the Ce_0.10‐NiCoO catalyst and 1.2 g of anhydrous sodium hypophosphite were placed at the downstream and upstream of the tube furnace, respectively. Under a N₂ atmosphere at a flow rate of 20 sccm and a vacuum pressure of 24 Pa, the temperature was ramped up to 300 °C at a rate of 5 °C min⁻¹ and maintained for 2 h, yielding the Ce_0.10‐NiCoP catalyst. All other phosphorus‐containing catalysts used in this study were prepared following the same procedure.

Synthesis of Pt/C and RuO₂ Catalyst

Both the Pt/C and RuO₂ electrodes were fabricated using a drop‐casting technique. A mixture of 5 mg of commercial 20% Pt/C, 20 µL of Nafion, and 480 µL of ethanol was ultrasonicated for 1 h. Subsequently, 200 µL of the prepared Pt/C dispersion was drop‐cast in multiple portions onto a 1 cm² NF substrate and a dried at room temperature. The loading mass of Pt/C on the NF was ≈2 mg cm⁻². The RuO₂ catalyst was synthesized using the same method, with Pt/C being replaced by RuO₂.

Materials Characterization

The catalytic materials were characterized using high‐angle X‐ray diffraction (XRD) on a Bruker D8 ADVANCE instrument (Germany), utilizing Cu‐Kα radiation (λ = 1.5444 nm) with a 2θ scan range from 10° to 80° at a scan rate of 20° min⁻¹. Morphological analysis of the materials was carried out using a Czech TESCAN MIRA LMS scanning electron microscope (SEM) and a transmission electron microscope (TEM) system (FEI Talos F200x, USA). X‐ray photoelectron spectroscopy (XPS) measurements were performed on a Thermo Scientific K‐Alpha spectrometer (USA) with Al Kα radiation (E = 1486.6 eV), operating at a working voltage of 12 kV and a filament current of 6 mA. The survey scans were acquired with a pass energy of 150 eV and a step size of 1 eV, while narrow scans were conducted with a pass energy of 50 eV and a step size of 0.1 eV. The binding energies of photoelectrons were referenced to the C 1s contamination peak at 284.8 eV for charge correction, and the experimental data were fitted using Avantage software. The wettability of the materials was evaluated using a Dataphysics OCA20 contact angle meter (Germany).

Electrochemical Tests

The electrochemical performance of the catalysts was evaluated using a typical three‐electrode cell configuration. An Hg/HgO electrode served as the reference electrode, and a graphite rod as the counter electrode. These electrochemical measurements were conducted in 1 M KOH solution using a CHI 760E electrochemical workstation.

All recorded potentials were converted to the reversible hydrogen electrode (RHE) scale using Nernst equation: RHE = E_Hg/HgO + 0.098 + 0.059 × pH. LSV was employed to investigate the catalytic activity of the catalysts toward the HER and OER at 25 °C, with a scan rate of 5.0 mV s⁻¹. EIS test were carried out within a frequency range of 0.01 to 105 Hz, using an AC. The EIS data were subsequently fitted using Zview2 software. The C_dl at the electrode‐electrolyte interface of was obtained through CV at scan rate ranging from 20 to 120 mV s⁻¹ to estimate the ECSA of the catalyst. The long‐term stability of the electrocatalyst was investigated through i‐t measurements conducted for 100 and 50 h for HER (j₂₀) and OER (j₅₀), respectively. TOF plots are obtained based on Table S2. (Supporting Information) The Faradaic efficiency test is conducted in an H‐type electrolytic cell, with AMI‐7001S used as the separator. The gases generated are collected using the water displacement method. The specific calculation formulas are as follows: FE = [n×F×(V/V_m)]/I×t, in this context, n represents the number of electrons transferred during the HER or OER. F denotes the Faraday constant (96485 C·mol⁻¹). V stands for the volume of H₂ or O₂ collected during the electrolysis process of HER or OER. V_m is the molar volume of gas under standard temperature and pressure, i indicates the applied current, and t represents the reaction time for the HER electrocatalytic process.

Conflict of Interest

The authors declare no conflict of interest.

Supporting information

Supporting Information

Zhang M., Xu H., Yang H., Shang X., Yuan M., Fu Y., Xiao Y., Wang S., Wang X., Jia B., Li S., Ma T., Electronic Modulation of Nickel Cobalt Phosphide Nanosheets by Ce Doping for Efficient Overall Water Splitting. Small 2025, 21, 2504837. 10.1002/smll.202504837

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.