Open Access Article
Yongli Tong
*a,
Xuan Zhaoa,
Yu Donga and
Ende Wang*b
aSchool of Science, Shenyang Ligong University, Shenyang 110159, China. E-mail: tyl.tongyongli@163.com
bSchool of Materials Science and Engineering, Shenyang University of Technology, Shenyang 110870, China. E-mail: ende_wang@163.com
First published on 17th March 2026
The development of highly efficient and economically viable bifunctional electrocatalysts is essential for overall water splitting in alkaline environments. Therefore, we synthesize V-doped NiCoP one-dimensional nanowire catalysts using hydrothermal and chemical vapor deposition methods. The as-obtained NCP-2 sample exhibits superior electrocatalytic performance, which might be attributed to the moderate vanadium doping that increases the density of electrochemically active sites and modulates the electronic structure of NiCoP. The NCP-2 sample shows an overpotential of 74.3 mV at 10 mA cm−2 with a small Tafel slope of 83.8 mV dec−1 during HER activity. In the OER process, NCP-2 exhibits an overpotential of 280 mV at 20 mA cm−2 (79.2 mV dec−1). Meanwhile, the NCP-2 nanowire arrays possess a low cell voltage of 1.55 V at 10 mA cm−2. This material shows significant potential for applications in sustainable energy systems.
A variety of electrocatalysts based on abundant and inexpensive elements have emerged as viable candidates to replace noble-metal catalysts, including transition metal oxides,7 metal sulfides,8 and metal phosphides.9 Among these candidates, NiCoP, a representative transition metal phosphide (TMP), has drawn considerable interest due to its metallic characteristics and the strong synergistic interaction between Ni and Co.10–12 NiCoP can undergo surface reconstruction to generate metal oxyhydroxide (M–OOH) species that are highly active toward the OER, while the metal–phosphorus (M–P) bonds facilitate hydride and proton adsorption, thereby enhancing HER performance.13–15 Despite these advantages, its practical catalytic efficiency remains limited by insufficient electrical conductivity and a scarcity of accessible active sites, which impede its application in efficient overall water splitting.
In general, the electrocatalytic activity of a catalyst is governed by number of accessible active sites and the inherent activity of each site. Doping strategies and structural morphology design are both effective approaches to enhance catalytic performance. Introducing metal cations as dopants can induce lattice distortion and modulate the electronic structure of catalysts, thereby increasing the density of exposed active sites, optimizing adsorption behavior, and ultimately enhancing electrocatalytic activity.16,17 For example, Kim et al.18 reported the fabrication of Ru-doped NiCoP nanoparticles via a one-step synthesis strategy, which demonstrated outstanding OER activity with an overpotential of only 281 mV to reach 10 mA cm−2. The introduction of Ru atoms generated a high density of Ni and Co vacancies within the NiCoP lattice, resulting in a more defect-rich structure and a large number of active sites compared to the pristine NiCoP counterpart. Zhang's group19 constructed two-dimensional vanadium-doped NiCoP nanosheets, which exhibited excellent HER performance. At a current density of 100 mA cm−2, the overpotential reached 332 mV.
Herein, we synthesize V-doped NiCoP nanowires using hydrothermal and chemical vapor deposition methods. Vanadium is particularly well suited as a dopant for vacancy engineering in TMPs, owing to its ability to adopt multiple high oxidation states (V3+/V4+/V5+) in phosphide-based materials.20 Moreover, under alkaline conditions, V species tend to form thermodynamically unstable and partially soluble vanadium-containing intermediate, which can promote the generation of structural defects and increase the number of catalytically active sites during electrochemical operation. Therefore, the incorporation of V dopants during TMP synthesis can effectively modulate both the electronic structure and crystal lattice, creating abundant vacancies and active site.21 The NiCoP-2 sample shows excellent electrocatalytic performance. The overpotential of NCP-2 sample is 74.3 mV at 10 mA cm−2 during HER process; in the oxygen evolution reaction (OER), NCP-2 product exhibits an overpotential of 280 mV at 20 mA cm−2.
| ERHE = E (Hg/HgO) + 0.098 + 0.591 × pH (∼13.7) | (1) |
![]() | (2) |
| Co2+ + 2PH3 → Co–P + 6H+ + P↓ | (3) |
| Ni2+ + 2PH3 → Ni–P + 6H+ + P↓ | (4) |
XRD is used to describe the structure and phase composition of the prepared sample in Fig. 2a. Peaks at 44.6, 51.9 and 76.5° correspond to characteristic peaks of nickel foam (JCPDS no. 04-0850). The peaks at 30.6°, 35.5°, 40.9°, 47.6°, 54.7°, 66.7° and 75.2° are consistent with (110), (200), (111), (210), (002), (202) and (400) crystal planes NiCoP (JCPDS: 71-2336), respectively. These diffraction peaks at 24.8°, 35.6°, 38.8°, 48.4°, 58.5°, 61.4°, 67.1°, 67.6°, and 78.4° are assigned to the (−111), (200), (012), (022), (131), (−313), (113), (311), and (042) reflections of CoP2 in Fig. S1, respectively, which are in good agreement with the standard JCPDS card (No. 77-0263). With the vanadium content increasing from 0.05 mmol to 0.2 mmol, the (111) and (210) crystal planes of NiCoP shift gradually to the left and no other new diffraction peaks are produced. Upon vanadium doping, the diffraction peaks shift to lower angles, evidencing successful incorporation and lattice expansion. This geometric distortion, driven by the disparate atomic dimensions and electronic properties of V relative to Ni/Co, creates localized strain fields that modify the d-band electronic structure of metal sites, potentially optimizing their catalytic activity through altered adsorbate binding strengths.
![]() | ||
| Fig. 2 Structural characterization (a) XRD patterns (b) XPS patterns of survey spectra (c) Ni 2p (d) Co 2p (e) P 2p (f) V 2p (g) HRTEM of NiCoP-2. | ||
XPS is carried out to further determine the composition and electronic structure of the material. The survey spectrum shows the presence of Ni, Co, P, V, C and O elements in Fig. 2b. Among these, Ni, Co, P, and V are derived from the prepared sample V-NiCoP. Fig. S2 shows the atomic percentage of each element, with C, O, P, Co, Ni, and V accounting for 60.7%, 25%, 11%, 2.5%, 0.7% and 0.2%, respectively. Ni 2p spectra (Fig. 2c) reveal four prominent peaks.22,23 The binding energy at 855.3 eV is associated with Ni 2p3/2 peaks, while the peaks at 872.6 eV correspond to Ni 2p1/2 peaks. Additionally, there are two satellite peaks at 859.5 and 877.8 eV, indicative of higher oxidation states. The peaks at 855.6 and 873.9 eV are assigned to Ni2+, whereas the peaks at 857.6 and 875.8 eV are linked to Ni3+. Compared to the Ni 2p peak in pristine NiCoP, the binding energy exhibits a negative shift of approximately 0.3 eV.24 This phenomenon arises because nickel possesses greater electronegativity than vanadium, facilitating electron transfer from vanadium to nickel. The resulting increase in electron density and the lowering of the oxidation state leads to a decrease in binding energy.25 The Co 2p XPS spectra (Fig. 2d) exhibit a spin–orbit doublet with Co 2p3/2 and Co 2p1/2 components centered at 781.7 eV and 797.6 eV, respectively, yielding a spin–orbit splitting energy of ∼15.9 eV, consistent with reported values for cobalt phosphides.26,27 The binding energies at 781.5 and 797.5 eV are ascribed to Co3+, while those at 784.1 and 799.9 eV are linked to Co2+. Likewise, given that cobalt is more electronegative than vanadium, it accepts electrons from vanadium, leading to a decreased oxidation state and enhanced electron density, with the binding energy shifting negatively by approximately 0.3 eV as well.24,25 Fig. 2e shows two peaks of P 2p1/2 and P 2p3/2, located at 129.5 and 130.5 eV, respectively. The peak at 135.3 eV results from the oxidation of phosphide by the superficial oxidation of metal phosphide due to the direct air contact of the sample.28 From Fig. 2f, the peak at 530.3 eV can be attributed to the V–O bonding, which is a consequence of the oxidation of vanadium.29 The V 2p XPS spectrum confirms vanadium exists in mixed oxidation states. The V 2p envelope is deconvoluted into three components at 514.2 eV (V3+), 515.3 eV and 516.7 eV (V4+), with the corresponding V 2p peak at 523.35 eV.20 Notably, the 516.7 eV binding energy aligns with reported values for V4+. This V3+/V4+ redox couple endows NiCoP with dynamic electronic flexibility, enabling adaptive modulation of active metal centers under varying electrochemical conditions.21 This multivalent-induced charge redistribution optimizes hydrogen adsorption/desorption during HER and stabilizes high-valence Ni/Co active species for OER, thereby delivering enhanced bifunctional electrocatalytic performance. The HRTEM image in Fig. 2g reveals lattice fringes with spacings of 0.2224 and 0.2815 nm, which can be indexed to the (111) and (101) planes of NiCoP, respectively, confirming the successful formation of the NiCoP structure. Notably, both values exceed the corresponding standard spacings (0.2200 and 0.2793 nm), indicating lattice expansion associated with the incorporation of vanadium into the NiCoP framework. Lattice expansion and local lattice disorder, together with XPS peak shifts may collectively indicate the generation of defects.
SEM is employed to examine the surface structure of the prepared samples. Fig. 3a shows the morphology of NiCoP sample, where nanowires are uniformly grown on the surface of nickel foam. The nanowires show a diameter of approximately 10–20 nm and a length of several micrometers. Fig. 3b–d display the morphologies of NCP-1, NCP-2 and NCP-3. It is evident that they all retain the nanowire structure, with no significant morphological changes observed due to the doping of vanadium. However, a closer examination reveals that as the vanadium content increases, the diameter of the nanowires slightly thickens. Vanadium doping may influence the growth mechanism of the nanowires, leading to the enlargement or aggregation of the crystalline grains. Moreover, TEM tests are conducted on the morphology before and after phosphidation to further observe the morphological changes. Fig. 3e shows the morphology of the V-NiCo precursor before phosphidation, where the nanowires appear smooth. After phosphidation, the V doped NiCoP-2 nanowires become uneven (Fig. 3f), with a significantly increased surface area, indicating enhanced catalytic performance. As shown in Fig. 3g, elemental mapping reveals a homogeneous distribution of Co, Ni, P, V elements within the NiCoP sample.
![]() | ||
| Fig. 3 SEM images (a) NCP sample (b) NCP-1 sample (c) NCP-2 sample (d) NCP-3 sample. TEM image of (e) V doped NiCo-precursor (f) NCP-2 sample (g) elemental mapping of NiCoP-2. | ||
The HER reaction in an alkaline solution involves the following two processes:35,36
| H2O + e− → Hads + OH− (Volmer reaction) |
| H2O + Hads + e− → H↑ + OH− (Heyrovsky reaction) | (5) |
| H2O + e− → Hads + OH− (Volmer reaction) |
| Hads + Hads → H↑ (Tafel reaction) | (6) |
The intrinsic activity of HER can be characterized by the Tafel slope in Fig. 4c. Among all the samples, NCP-2 shows the lowest Tafel slope of 83.8 mV dec−1, which is lower than the NCP sample (142.8 mV dec−1), NCP-1 (88.8 mV dec−1), and NCP-3 (98.3 mV dec−1). It indicates that the HER follows a Volmer–Heyrovsky mechanism with mixed kinetic control. The incorporation of V modulates the electronic structure of NiCoP, facilitating water dissociation and optimizing H* adsorption, thereby accelerating the Volmer step while enabling the Heyrovsky step to participate in the rate-determining process.
To assess the inherent activity of catalysts, the electrochemical active surface area (ECSA) is evaluated through the electrochemical double-layer capacitance method (Cdl). The Cdl is measured using a straightforward cyclic voltammetry (CV) technique at varying scan rates (10, 20, 30, 40, 50 mV s−1) within the potential range of 0.67–0.87 V (vs. RHE), with no faradaic current features observed. Fig. 4d shows NCP-2 sample possesses the ECSA of 0.49 mF cm−2, meaning a high specific surface area with abundant active sites. It can significantly enhance the electrochemical process. EIS characterization is employed to analyze the charge transfer kinetics. The EIS measurements are fitted using an equivalent circuit model, where the solution resistance (Rs) is connected in series with a parallel arrangement of the charge-transfer resistance (Rct) and a constant phase element (CPE). The CPE accounts for the non-ideal behavior of the double-layer capacitance at the electrode/electrolyte interface. The equivalent circuit is illustrated in Fig. 4e. The curve fitted with the equivalent circuit model closely coincides with the experimental data, demonstrating the reliability of the chosen model. The fitting results indicate that the equivalent resistance of the NiCoP-2 sample is 2.5 Ω. Fig. 4f presents a comparison of the impedance spectra of samples with different vanadium doping contents. The NiCoP-2 sample exhibits the smallest semicircle radius, indicating the lowest charge-transfer resistance (Rct), the fastest HER kinetics, and the best catalytic performance. With increasing vanadium content, the semicircle radius first decreases and then increases. This suggests that an appropriate amount of vanadium doping can reduce the charge-transfer resistance, whereas excessive vanadium doping hinders charge-transfer pathways and leads to increased resistance. The long-cycle performance of the NCP-2 material is tested to explore its stability and durability. As shown in the Fig. 4g, there is no significant decrease in current density after 100 h of cycling. The excellent cycling stability stems from the synergistic stabilization of the chemically bonded interface, V-doping, and 1D nanostructure. The in situ grown nanowires prevent delamination and dissipate mechanical stress from gas bubble evolution, while V-doping strengthens the lattice, suppresses phosphorus leaching, and inhibits metal dissolution via charge compensation. Furthermore, surface reconstruction forms a stabilized core–shell architecture where V-anchored (oxy)hydroxide active layers protect the conductive phosphide core, and the 3D Ni foam backbone ensures persistent electron transport throughout long-term redox cycling.
Fig. 5b presents a comparison of the overpotentials of NCP-2 and the other samples. At a current density of 20 mA cm−2, the overpotential of NCP-2 is lower than that of NCP (290.3 mV), NCP-1 (301.2 mV), NCP-3 (322.3 mV). At a current density of 50 mA cm−2, the overpotential of NCP-2 is still superior to that of the other samples.Table 2 shows that the overpotential of NCP-2 outperforms some of the previously reported data at a current density of 100 mA cm−2.37–40 Vanadium shows a strong electron-accepting ability, and it can optimize the catalytic performance of NiCoP by altering its electronic structure. When vanadium is incorporated into NiCoP, the electronic structure of vanadium may interact with the electronic structures of Ni and Co, thereby affecting the electron density distribution of the catalyst. This change can facilitate the adsorption and desorption of oxygen molecules, thereby lowering the energy barrier for the OER reaction and, in turn, reducing the overpotential. Meanwhile, as a transition metal, vanadium can alter the redox behavior on the catalyst surface. The incorporation of vanadium may help optimize the surface oxidation state transitions of Ni and Co, making it easier for NiCoP to form intermediate species with lower activation energies (such as *O, *OOH, etc.) during the catalytic process. The formation of these intermediate species helps reduce the overpotential for OER.
From Fig. 5c, NCP-2 exhibits the lowest Tafel slope of 79.2 mV dec−1, outperforming NCP (93.6 mV dec−1), NCP-1 (96.8 mV dec−1), NCP-3 (98.6 mV dec−1). This means that an appropriate amount of vanadium doping helps accelerate the kinetic reaction rate. This is because the appropriate introduction of vanadium may improve the interaction between the catalyst and reaction intermediates (such as *O, *OOH, etc.), leading to a faster OER reaction rate. The oxidation state transitions of Ni and Co on the NiCoP surface may occur more smoothly, thus accelerating the OER reaction. Additionally, the incorporation of vanadium may result in the formation of more active sites on the NiCoP surface, or optimize the electronic state of existing active sites, making it easier for reaction intermediates to react with the catalyst surface and reducing the rate-limiting steps of the reaction. These factors could all contribute to the decrease in the Tafel slope.
NCP-2 exhibits a Cdl of 7.10 mF cm−2, the highest among all the samples (Fig. 5d). It implies that the appropriate incorporation of vanadium can facilitate greater electron accumulation at the electrode surface, enhancing its charge transfer capabilities. From Fig. 5e, the NCP-2 sample still maintains excellent stability after 100 h of extended cycling. This sustained stability can be attributed to the robust structural integrity and high durability of NCP-2 under prolonged electrochemical stress, which allows it to effectively withstand the repeated charging and discharging cycles without significant loss of catalytic performance. The radar chart in Fig. 5f clearly demonstrates that the dual-functional electrocatalytic performance of NCP-2 surpasses that of the other five materials. The enhanced performance could stem from its optimal balance between conductivity, stability, and surface area, which allows it to facilitate multiple reaction pathways simultaneously. Such a multifaceted performance further underscores the potential of NCP-2 in practical applications.
After the OER cycling, the sample is further characterized by XRD, TEM, and XPS to elucidate the OER mechanism. As shown in Fig. S3, the diffraction peaks located at 37.5°, 43.4°, 54.2°, and 58.2° can be indexed to the (011), (210), (211), and (220) crystal planes of NiOOH (JPCDS no. 27-0956), respectively. Meanwhile, the peaks at 47.9°, 53.9°, 61.2°, and 63.0° are consistent with the (131), (211), (231), and (151) planes of CoOOH (JPCDS no. 26-0480), respectively. Two new phases are formed. The HRTEM images further illustrate this point in Fig. S4. The lattice spacing of 1.901 nm corresponds to the (131) crystal plane of CoOOH, while 1.987 nm corresponds to the (210) crystal plane of NiOOH, confirming that new CoOOH and NiOOH phases are indeed formed on the sample surface. This is because the V-doped NiCoP surface is no longer thermodynamically stable. OH− nucleophilically attacks the P sites, oxidizing P to soluble phosphate and leaving behind P-vacancies. Simultaneously, the V dopant (V3+/V4+) acts as an electron pump, facilitating the oxidation of Ni2+ to Ni3+ and Co2+ to Co3+ by lowering the associated energy barriers. The resulting high-valent Ni3+ and Co3+ ions immediately coordinate with OH− in the alkaline electrolyte, forming NiOOH and CoOOH phases. These oxyhydroxides are the true active layers for water oxidation.
Relative to the pre-cycling XPS spectrum, the Ni2+ contribution decreases, the Ni3+ fraction increases, and the satellite peak at 863 eV gains intensity (Fig. S5a). This evolution aligns with the preceding XRD and HRTEM findings, corroborating that the NiOOH phase has indeed formed. Likewise, the Co 2p peak shifts to lower binding energy, the Co2+ content decreases while Co3+ increases, all evidencing the formation of CoOOH (Fig. S5b). From Fig. S5c, the characteristic P 2p3/2 and P 2p1/2 peaks associated with metal–phosphide species nearly disappear, indicating the depletion of P3− due to oxidative transformation. Meanwhile, the P–O peak becomes significantly broadened, suggesting the formation of multiple oxidized phosphorus species with diverse coordination environments. The V 2p3/2 peak show an obvious positive shift from 516.3 eV to 517.4 eV after OER cycling, accompanied by a significant increase in the V5+ species and a decrease in low-valence V3+/V4+ components, indicating the oxidation of vanadium during electrochemical activation. This suggests that the surface undergoes in situ reconstruction to form V-incorporated Ni (Co) oxyhydroxide as the real active phase for OER (Fig. S5d).
![]() | ||
| Fig. 6 (a) Schematic illustration of overall water splitting (b) polarization curves at a scan rate of 5 mV s−1 (c) chronopotentiometry measurements. | ||
Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d5ra09888b.
| This journal is © The Royal Society of Chemistry 2026 |