High-spin state electron configuration in Mn-doped Ni3Se4 for efficient methanol oxidation

Journal of Energy Chemistry 112 (2026) 720–729 Contents lists available at ScienceDirect Journal of Ener gy Chemistry journal homepag e: www.elsevier.com/locate/jechem High-spin state electron configuration in Mn-doped Ni 3 Se 4 for efficient methanol oxidation Yong Zhang a,b , a Institute for Advanced Study, Chengdu University, Chengdu 610106, Sichuan, China b School of Mechanical Engineering, Chengdu University, Chengdu 610106, Sichuan, China Yi M a a,b , Jing Yu c,d , c Catalonia Institute for Energy Research - IREC, Sant Adrià de Besòs, Barcelona 08930, Catalonia, Spain d Catalan Institute of Nanoscience and Nanotechnology - ICN2 (CSIC and BIST), Campus UAB, Bellaterra 08193, Barcelona, Spain Canhuang Li c , Jordi Arbiol d,e , e ICREA, Pg. Lluís Companys 23, Barcelona 08010, Catalonia, Spain Xiaoxi Wang f , f State Key Laboratory of Fine Chemicals, School of Chemical Engineering, Dalian University of Technology, Dalian 116012, Liaoning, China Ning Jian a,b , Huan Ge a,b , Luming Li a , Andreu Cabot c,e, ⁎ , Junshan Li a,b, ⁎ a r t i c l e info Article history: Received 16 June 2025 Revised 14 August 2025 Accepted 14 August 2025 Available online 11 September 2025 Keywords: Methanol oxidation reaction Nickel selenide Spin state, Electrocatalysis Formic acid a b s t r a c t The methanol oxidation reaction (MOR) to formic acid offers a promising alternative to the anodic oxygen evolution reaction (OER) in water electrolysis. However, the development of efficient and cost-effective catalysts remains a primary challenge. In this study, an enhancement in catalytic MOR performance is achieved through the incorporation of Mn atoms with unsaturated t 2g orbitals into Ni 3 Se 4 . Comprehensive experimental analyses and theoretical calculations reveal that substituting Ni with Mn induces strong electron-withdrawing effects, effectively modulating the local coordination environment of the metal centers. The presence of Mn also elongates Ni–Se(O) bonds, which reduces e g orbital occupancy and modifies the spin state of the material. Electrochemical measurements demonstrate that electrodes based on this optimized material exhibit a high spin state and deliver excellent catalytic activity, achieving a MOR current density up to ∼190 mA cm −2 at 1.6 V. This performance enhancement is attributed to the favorable electronic configuration and reduced reaction energy barriers associated with the high-spin state. © 2025 Science Press and Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier B.V. and Science Press. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 1. Introductio n The electro-oxidation of methanol to formic acid, accompanied by hydrogen co-generation, presents several advantages over conventional formic acid production methods [1–5]. These advantages include high energy efficiency, potentially high selectivity, operation under ambient temperature and pressure, and compatibility with renewable electricity sources [6–9]. Moreover, this approach enables rapid system activation and deactivation, making it wellsuited for intermittent use of residual energy, whenever available [10–12]. Its scalability and potential for integration with electroreduction reactions to produce hydrogen or other value-added chemicals further underscore its industrial relevance. Notably, the methanol oxidation reaction (MOR) can also be coupled with the oxygen reduction reaction (ORR) in direct methanol fuel cells, enabling the co-generation of formic acid and electricity [13–18]. ⁎ Corresponding authors. E-mail addresses: [email protected] (A. Cabot), [email protected] (J. Li). https://doi.org/10.1016/j.jechem.2025.08.085 2095-4956/© 2025 Science Press and Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier B.V. and Science Press. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Traditionally, MOR activation has relied on expensive precious metal-based electrocatalysts, which present a significant barrier to large-scale industrial applications [16,19–21]. To address this limitation, recent research efforts have focused on the development of alternative cost-effective MOR catalysts based on 3 dtransition metals. Nickel-based catalysts, in particular, have demonstrated remarkable performance in alkaline media [22– 29]. This superior performance is attributed to the formation of NiOOH on the catalyst surface, which is considered the true active species in MOR [30–32]. Current research is primarily directed towards enhancing catalyst activity by increasing the number of active sites and improving their intrinsic activity by adjusting their electronic configuration. Electronic spin states have been identified as a key, yet often overlook ed, parameter that plays a crucial role in electrocatalysis [33]. Spin states directly influence the strength of metal–ligand bonds, thereby affecting the adsorption behavior of catalytically active sites and reaction intermediates, and ultimately governing the overall catalytic performance [34]. Consequently, tuning the spin state represents a promising strategy for the rational design and optimization of advanced electrocatalysts. Y. Zhang, Y. Ma, J. Yu et al. Journal of Energy Chemistry 112 (2026) 720–729 In this context, Zeng et al. enhanced the catalytic activity of the oxygen evolution reaction (OER) by modulating e g orbital filling through particle size control in cobalt-based salts [35]. Similarly, Wang et al. improved the performance of both the hydrogen evolution reaction (HER) and the urea oxidation reaction (UOR) by tuning electron spin via oxygen vacancy regulation in amorphous/crystalline catalysts [36]. Additionally, we have demonstrated that the electronic spin state significantly influences the reaction kinetics in electrocatalytic Li-S and Na-S systems [37– 42]. Despite growing evidence of the critical role of spin states in governing various electrochemical processes, the effect of metal cation doping on the electronic configuration and spin state of MOR catalysts remains largely underexplored. In one of our previous studies, we demonstrated that Ni 3 Se 4 exhibits superior MOR activity compared to NiSe and NiSe 2 nanomaterials [43]. Building on this, the present study introduces a novel MOR catalyst by doping manganese atoms, featuring unsaturated t 2g orbitals, into Ni 3 Se 4 to form Ni 3 Se 4 -x%Mn. The resulting modifications in the electronic structure were systematically investigated using X-ray absorption spectroscopy (XAS), while the spin state was analyzed via vibrating sample magnetometry (VSM). Furthermore, density functional theory (DFT) calculations were conducted to elucidate the underlying mechanism of the methanol-to-formate conversion. These findings highlight the critical role of spin state modulation and electronic structure tuning in enhancing the catalytic performance of nickel-based electrocatalysts. 2. Results and discussion A series of Ni 3 Se 4 -x%Mn (x = 0, 5, 10, and 15) nanoparticles (NPs) was synthesized via a hydrothermal method using Ni (NO 3 ) 2 and Mn(NO 3 ) 2 as metal precursors, as illustrated in Fig. 1 (a). Briefly, in a typical procedure, the appropriate ratio of metal nitrates was dissolved in deionized water and added to a solution containing selenium powder and hydrazine hydrate. The resulting mixture was transferred to a Teflon-lined stainless-steel autoclave and heated at 180 °C for 24 h. Further details are provided in the Experimental Section of the Supporting Information (SI). X-ray diffraction (XRD) patterns of the synthesized materials (Fig. 1b) exhibit characteristic peaks at 33.3°, 45.1°, 50.8°, 60.7°, and 61.7°, corresponding to the (−112), (−314), (310), (−716), and (−422) planes of Ni 3 Se 4 (JCPDS No. 97-004-2558). It is worth noting that the minor peaks observed at approximately 35° and 55° can be attributed to residual Se, originating from the unreacted precursor remaining on the surface. No additional diffraction peaks were detected, confirming the absence of Mn-based secondary phases or other Ni-Se compounds. Moreover, the incorporation of Mn did not alter the crystal structure of Ni 3 Se 4 . As shown in Fig. S1, the actual Mn/Ni ratio determined by scanning electron microscope-energy-dispersive X-ray spectroscopy (SEM-EDS) was consistent with the nominal value. Therefore, the theoretical Mn/ Ni ratio was used to differentiate between the samples. Transmission electron microscopy (TEM) images reveal that the synthesized particles exhibit a planar morphology (Fig. 1c and Fig. S2). The average particle size increases with Mn doping, from 18±5 nm for pristine Ni 3 Se 4 to 39±5 nm for Ni 3 Se 4 -15 %Mn. Furthermore, the Brunauer-Emmett-Teller (BET) specific surface area was determined from nitrogen adsorption–desorption isotherms (Fig. 1d). The BET surface areas for Ni 3 Se 4 , Ni 3 Se 4 -5 %Mn, Ni 3 Se 4 - 10 %Mn, and Ni 3 Se 4 -15 %Mn NPs were calculated to be 4.3, 13.4, 18.3, and 21.2 m 2 g −1 , respectively. Interestingly, the specific surface area increases with higher Mn content, despite the observed increase in particle size. This trend may be attributed to a reduction in particle thickness or a lower degree of agglomeration in the Mn-doped samples [44,45]. 721 High-resolution TEM (HRTEM) analysis of Ni 3 Se 4 -10 %Mn NPs (Fig. 1e) confirms that this nanostructure has a crystal phase that can be assigned to the monoclinic crystal (space group = C12/m1) with a = 12.15 Å, b = 3.633 Å, c = 10.45 Å, a = c = 90°, and b = 149°. In addition, well-defined lattice fringes 0.267, 0.200, and 0.267 nm, at 49° and 97°, respectively, which could be interpreted as the monoclinic Ni 3 Se 4 phase, are visualized along its [201] zone axis. In the red-squared domain, the figure illustrates that Ni (purple) is intercalated within the Se (pink) layer. Due to Mn doping, the corresponding (1−1−2) lattice spacing is slightly expanded to 0.267 nm, compared to 0.265 nm in the undoped Ni 3 - Se 4 reference sample (Fig. S3), indicating that Mn substitution leads to a slight expansion of the lattice. Furthermore, EDS elemental mapping (Fig. 1f) and line profile analysis (Fig. 1g) indicate a uniform distribution of Ni, Mn, and Se throughout the sample. X-ray photoelectron spectroscopy (XPS) and X-ray absorption fine structure (XAFS) measurements were conducted to investigate the chemical valence states and electronic structure of the NPs (Fig. 2 and Fig. S4). The Ni 2p 3/2 XPS spectrum of Ni 3 Se 4 -10 %Mn (Fig. 2a) exhibits contributions at 853.7 and 856.8 eV, corresponding to Ni 2+ and Ni 3+ , respectively [46]. The feature at ∼861 eV corresponds to the well-known Ni 2+ ‘‘satellite” peak, originating from shake-up processes involving ligand-to-metal charge-transfer excitations; its reduced intensity after Mn doping reflects the decreased Ni 2+ fraction, consistent with the observed Ni 3+ enrichment. Compared to undoped Ni 3 Se 4 , the Ni 2p peaks shift toward lower binding energies upon Mn doping, accompanied by an increased Ni 3+ /Ni 2+ ratio. This Ni 3+ enrichment can be attributed to the strong electron-withdrawing effect of Mn in the local Ni 3 Se 4 structure. The Mn 2p 3/2 spectrum (Fig. 2b) shows three distinct peaks at 637.9, 642.2, and 645.5 eV, confirming the presence of Mn 2+ ,Mn 3+ , and Mn 4+ oxidation states. Fig. 2(c) presents the fitted Se 3d XPS spectrum, revealing characteristic components corresponding to Ni(Mn)–Se and Se–O species. Due to the unsaturated t 2g orbital of Mn 2+ , electronic coupling via bridging Se–O ligands facilitates partial charge transfer from Ni 2+ to Mn 2+ , oxidizing Ni 2+ to Ni 3+ . In addition, the decrease in Se content with Mn doping primarily reflects the reduction of non-lattice, surface-bound Se rather than a change in the intrinsic Se stoichiometry of the Ni 3 Se 4 lattice, which clearly supports Mn substitution at Ni sites, as confirmed by both XRD peak shifts and HRTEM measurements. Furthermore, the presence of Se–O is likely due to partial surface oxidation, which may have occurred during synthesis, handling, or exposure to ambient conditions prior to measurement [47–49]. XAFS spectra (Fig. 2d) confirm that the valence state of Ni in both Ni 3 Se 4 and Ni 3 Se 4 -10 %Mn lies between Ni 2+ and Ni 3+ , consistent with the mixed-valence nature of nickel selenides [50,51]. Mn doping leads to an increase in the average valence state of Ni, in agreement with the XPS results. Additionally, the more pronounced absorption edge observed in Ni 3 Se 4 -10 %Mn suggests a lower occupancy of Ni electronic orbitals. This implies enhanced electron delocalization and more favorable charge transfer between Ni and the surrounding Se(O) ligands, which can facilitate improved catalytic activity in the MOR process. Compared with NiO (2.50 Å), the Ni–Ni(Mn) scattering peak in Ni 3 Se 4 -10 %Mn NPs appears at 2.72 Å, indicating a clear rightward shift due to electronic interactions between Ni and Mn atoms (Fig. 2e). The peak at 1.80 Å is attributed to Ni–Se(O) bonds, suggesting that the first coordination shell consists primarily of Ni– Se(O), while the Ni–Ni(Mn) interactions belong to the second coor- dination shell. In the R-space spectra for Mn, two peaks are observed at approximately 1.86 and 2.70 Å, corresponding to Mn–O and Mn–Mn bonds, respectively, while the peak at 2.76 Å is assigned to Mn–Ni bonding. The extended X-ray absorption fine structure (EXAFS) fitting (Fig. 2f) supports these observations, showing a decrease in Ni–Ni coordination number (CN) and a shorter Ni–O bond length in Ni 3 Se 4 -10 %Mn compared to pristine Ni 3 Se 4 . Wavelet transform (WT) analysis (Fig. 2g) further confirms the coordination environment, where Ni 3 Se 4 -10 %Mn exhibits a broader distribution in R space and a reduced intensity in the Ni–Ni domain relative to Ni foil and Ni 3 Se 4 , suggesting a more disordered local structure. Collectively, these results indicate that Mn doping introduces significant local structural and electronic modulation around Ni centers, which may be correlated with improved electrocatalytic performance. Y. Zhang, Y. Ma, J. Yu et al. Journal of Energy Chemistry 112 (2026) 720–729 ◦ Fig. 1. (a) Schematic diagram of the synthesis of Ni 3 Se 4 -x%Mn particles. (b) XRD patterns of Ni 3 Se 4 -x%Mn. (c) TEM image and particle size histogram of Ni 3 Se 4 -10 %Mn. (d) N 2 adsorption-desorption isotherms of Ni 3 Se 4 -x%Mn NPs (x = 0, 5, 10, and 15). (e) HRTEM image of Ni 3 Se 4 -10 %Mn and corresponding fast fourier transform (FFT) pattern of the selected region marked with red squares. (f) Scanning TEM-EDS compositional maps. (g) EDS line profile for Ni 3 Se 4 -10 %Mn. The MOR catalytic activity of the electrodes was evaluated in 1 M KOH containing 1 M methanol (MeOH) (see SI for details). Upon the addition of MeOH, all four electrodes exhibited a sharp increase in current density (Fig. 3a). Linear sweep voltammetry (LSV) of undoped Ni 3 Se 4 nanoparticles showed a current density of 140 mA cm −2 at 1.6 V, which is notably lower than that of the Mn-doped samples. As the Mn doping level increases, the current densities achieved are 176 mA cm −2 for Ni 3 Se 4 -5 %Mn, 190 mA cm −2 for Ni 3 Se 4 -10 %Mn, and 156 mA cm −2 for Ni 3 Se 4 -15 %Mn. Among all, Ni 3 Se 4 -10 %Mn exhibits the highest MOR activity, highlighting the optimal doping concentration for catalytic performance. In addition, the Tafel slope of Ni 3 Se 4 -10 %Mn is 36.58 mV dec −1 , significantly lower than that of Ni 3 Se 4 -5 %Mn (62.09 mV dec −1 ), Ni 3 - Se 4 -15 %Mn (70.57 mV dec −1 ), and undoped Ni 3 Se 4 (75.16 mV dec −1 ), indicating improved reaction kinetics with optimal Mn doping (Fig. 3b) [52]. 722 Electrochemical impedance spectroscopy (EIS) measurements conducted in 1 M KOH and 1 M KOH + 1 M MeOH solutions (Figs. 3c and d) reveal that the Ni 3 Se 4 -10 %Mn electrode exhibits significantly faster charge transfer kinetics for both the OER and MOR compared to the other three samples [53]. Notably, the charge transfer resistance during MOR is markedly lower than that during OER, highlighting the dominant competitiveness of MOR at the corresponding voltage. Fig. 3(e and f) displays the Bode phase diagrams, further indicating that Mn doping promotes the acceleration of the rate-determining step (RDS) of the electrochemical MOR process. Moreover, Mn incorporation reduces the resistance associated with the adsorption and migration of methanol and its intermediates, contributing to the overall enhancement in the electrocatalytic performance, which is in good agreement with previous investigations [54]. Chronoamperometry (CA) tests were performed on the Ni 3 Se 4 - 10 %Mn electrode at various applied potentials (1.3, 1.4, 1.5, 1.6, 1.7, and 1.8 V), as shown in Fig. S5(a). After 1 h of testing at each potential, 0.5 mL of electrolyte was collected and analyzed via ion chromatography (IC) to quantify the formic acid content (see SI for details). As shown in Fig. S5(b), a distinct IC peak appeared at approximately 9 min, in good agreement with the characteristics of formate [55]. The formate peak intensity increased with applied potential, while the peak at 1.3 V was negligible, indicating that the formate was not detected at this potential. Based on calibration with standard formate solutions and IC curve fitting, the Faradaic efficiencies (FEs) for methanol-to-formic acid conversion at potentials from 1.4 to 1.8 V were determined to be 87.43 %, 96.76 %, 99.10 %, 80.93 %, and 47.85 %, respectively . This trend indicates that the highest selectivity toward formic acid is achieved at 1.6 V. Beyond this potential, FE decreases, suggesting increased competition from the OER, which becomes more dominant at higher voltages (Fig. 3g). Y. Zhang, Y. Ma, J. Yu et al. Journal of Energy Chemistry 112 (2026) 720–729 Å Å Å Å Å Å Å Å Fig. 2. (a) Ni 2p 3/2 , (b) Mn 2p 3/2 , and (c) Se 3d XPS spectra of Ni 3 Se 4 and Ni 3 Se 4 -10 % Mn NPs. (d) K-edge XANES of Ni in Ni 3 Se 4 and Ni 3 Se 4 -10 % Mn NPs. (e) EXAFS spectra of Ni 3 Se 4 and Ni 3 Se 4 -10 % Mn NPs. (f) XAFS fitting curves of Ni K-edge for Ni 3 Se 4 and Ni 3 Se 4 -10 % Mn NPs (K3 space). (g) WT contour plots for Ni foil, Ni 3 Se 4 , and Ni 3 Se 4 -10 % Mn NPs at Rspace. A long-time stability test was performed in this cell. To further evaluate the impact of Mn doping ratio on the long-term stability and MOR selectivity, 18-h CA tests were performed on all four electrodes at the previously determined optimal potential of 1.6 V. The catalytic activity loss rates, formic acid production, and selectivity were analyzed over the duration of the tests. As shown in Fig. 3(h), all electrodes exhibited a decline in current density during the 18h period. However, the Ni 3 Se 4 -10 %Mn electrode demonstrated relatively high stability, maintaining a current density of ∼123 mA cm −2 after 18 h, representing a 36.4 % decrease from the initial value, and second only to Ni 3 Se 4 -15 %Mn. Lower Mn doping ratios resulted in more significant activity loss, which may be attributed to the reduced ability to mitigate the poisoning effects of reaction intermediates, a function that Mn atoms appear to enhance. To elucidate the degradation mechanism of Ni 3 Se 4 -10 %Mn during MOR, the catalyst was analyzed after long-term stability testing. Representative TEM images (Fig. S6) show that the NPs are largely aggregated on the carbon black (CB), reducing the accessible electrochemical surface area. As seen in Fig. S7, SEMEDS mapping confirmed a decrease in Se content, indicating surface leaching of chalcogen species during operation. Further XPS analysis (Fig. S8) further supported this finding, showing a marked reduction in Se intensity and a shift in the Ni 2p 3/2 spectra toward higher Ni 3+ content, consistent with electrochemical reconstruction to NiOOH-like species. Combined with the operando EIS measurements at different potentials (Fig. 3e and f), it is revealed that Mn doping accelerates this reconstruction process. These changes, which are closely related to the decline in catalytic performance, are primarily attributed to the harsh alkaline environment and the sustained high external potential during the reaction. This observation is consistent with previous reports, confirming that such conditions can induce structural transformation and deactivation of the catalyst [56–58]. 723 IC analysis of the electrolyte (Fig. 3i) revealed that the Ni 3 Se 4 -10 %Mn electrode generated 4.63 mmol of formic acid, with an average FE of 93.2 % over 18 h. This performance surpassed that of Ni 3 - Se 4 -5 %Mn (91.0 %), Ni 3 Se 4 -15 %Mn (88.1 %), and undoped Ni 3 Se 4 (86.9 %). These results indicate that an optimal Mn doping level not only enhances long-term catalytic stability but also improves selectivity toward formic acid production (Fig. 3j). Among the four different Mn doping ratios in the Ni 3 Se 4 , Ni 3 Se 4 with 10 % Mn doping had the best catalytic performance. Moreover, as shown in Fig. 3(k), the catalytic activity of Ni 3 Se 4 -10 %Mn outperforms all previously reported Ni-based electrocatalysts (see Table S1 for other details), further underscoring its potential for practical MOR applications. To figure out the reasons for the enhanced electrocatalytic performance by Mn doping, we studied the electrochemically active surface area (ECSA), the surface coverage of redox species ( C *), and the proton diffusion coefficient (D ) of the catalyst. Firstly, ECSA was estimated from the double-layer capacitance (C dl ), obtained by cyclic voltammetry (CV) at various scan rates within the nonFaradaic potential region (Figs. S9 and S10, see SI for details) [59]. Among the tested samples, the Ni 3 Se 4 -10 %Mn electrode exhibited the highest ECSA of 46 cm 2 g −1 , slightly exceeding that of Ni 3 Se 4 -5 %Mn (45 cm 2 g −1 ) and Ni 3 Se 4 -15 %Mn (44 cm 2 g −1 ), and significantly higher than that of undoped Ni 3 Se 4 (35 cm 2 g −1 ). While the variation in ECSA among the Mn-doped samples is relatively small, Mn incorporation clearly enhances the electrochemically active surface area compared to the undoped counterpart. Y. Zhang, Y. Ma, J. Yu et al. Journal of Energy Chemistry 112 (2026) 720–729 Fig. 3. (a) The MOR curves of Ni 3 Se 4 -x%Mn NPs (x = 0, 5, 10, and 15) and (b) corresponding Tafel slopes. The Nyquist plot of Ni 3 Se 4 -x%Mn NPs (x = 0, 5, 10, and 15) measured in (c) 1 M KOH solution and (d) 1 M KOH and 1 M methanol solution at 1.5 V. Bode plots of (e) Ni 3 Se 4 and (f) Ni 3 Se 4 -10 % Mn. (g) FE at different potentials of the Ni 3 Se 4 -10 %Mn electrode. (h) CA curve during 18 h continuous test at 1.6 V of the Ni 3 Se 4 -x%Mn (x= 0, 5, 10, and 15) electrodes. (i) IC curves. (j) FE and formic acid content. (k) Comparison of the MOR current density obtained in this work with other reported Ni-based catalysts. CV curves recorded in 1 M KOH at 50 mV s −1 (Fig. S11) show that the peak current density for the Ni 3 Se 4 electrode appears at 1.375 V, while the three Mn-doped electrodes exhibit slightly lower peak potentials around 1.359 V during the forward scan. This peak corresponds to the oxidation of surface Ni(OH) 2 to NiOOH, which serves as the true active species for the MOR [60]. Notably, the Ni 3 Se 4 -10 %Mn electrode exhibits the highest peak intensity, indicating the most efficient formation of active NiOOH species and superior catalytic activity of MOR. 724 The peak current increases with scan rate, while the anodic peak potential shifts positively and the cathodic peak potential shifts negatively with increasing scan rate (Fig. S12). These shifts are attributed to electrochemical polarization and kinetic limitations in the formation of NiOOH species [61]. A key parameter reflecting redox behavior is the surface coverage of Ni(OH) 2 /NiOOH redox species ( C *), which was estimated by averaging the anodic and cathodic peak currents (I p ) over scan rates ranging from 10 to 50 mV s −1 in the potential window of 0.9–1.6 V, based on both forward and reverse scans (Fig. S13). The Ni 3 Se 4 -10 %Mn electrode exhibited the highest C * value of 1.11×10 −7 mol cm −2 , exceeding that of Ni 3 Se 4 -5 %Mn (1.03×10 −7 mol cm −2 ), Ni 3 Se 4 -15 %Mn (0.99×10 −7 mol cm −2 ), and undoped Ni 3 Se 4 (0.96×10 − 7 mol cm −2 ), indicating more abundant electroactive redox sites with optimal Mn doping. Y. Zhang, Y. Ma, J. Yu et al. Journal of Energy Chemistry 112 (2026) 720–729 In Ni-based electrodes operating in alkaline media, proton diffusion is generally regarded as the rate-limiting step governing the participation of Ni(OH) 2 in the NiOOH redox reaction [62]. Within the scan rate range of 60–100 mV s −1 , the anodic and cathodic peak currents (I p ) were found to vary linearly with the square root of the scan rate ( m 1/2 ), indicating that the redox process is diffusion-controlled. The proton diffusion coefficient for the Ni 3 - Se 4 -10 %Mn electrode was estimated to be 1.344×10 −8 cm 2 s −1 , approximately one order of magnitude higher than that of the other three electrodes (Fig. S14). This result highlights the enhanced proton transport kinetics enabled by optimal Mn doping. To sum up, these results confirm that proper Mn doping enhances the catalytic activity for MOR. However, the enhancement is not linear or indefinite. Since NiOOH is the true active species for MOR, partial substitution of Ni with Mn inevitably reduces the number of active Ni sites. Excessive Mn incorporation can therefore hinder catalytic performance, as reflected by the electrochemical behavior of the differently doped samples. Thus, a moderate Mn doping level, around 10 %, achieves the best balance between electronic modulation and preservation of active sites, resulting in optimal catalytic performance. Since the enhanced MOR performance by the proper Mn presence, the effect of Mn doping was further studied. Firstly, the magnetic field dependence of the magnetization (M-H) curves for Ni 3 Se 4 and Ni 3 Se 4 -10 %Mn NPs at room temperature (Fig. 4a) reveals pronounced paramagnetic behavior in both samples, with notable differences in their magnetic properties [51,63]. Specifically, the coercivity (Hc) of Ni 3 Se 4 -10 %Mn is lower than that of pristine Ni 3 Se 4 . In nanomaterials where the grain size exceeds the single magnetic domain size, coercivity typically follows the inverse relationship H=C/d, where C is a material-dependent constant and dis the grain size. As shown above, the incorporation of Mn atoms into the Ni 3 Se 4 lattice results in lattice expansion and an increase in overall particle size, which likely contributes to the observed decrease in coercivity. The electronic configurations of the 3d orbitals before and after Mn doping were analyzed through temperature-dependent magnetization measurements under zero-field cooling (ZFC) conditions (Fig. 4b and Fig. S15). The number of unpaired 3d electrons (n) in the metal centers of Ni 3 Se 4 and Ni 3 Se 4 -10 %Mn nanoparticles was estimated using the following equation, derived from the magnetic susceptibility ( v m ) in ZFC measurements. l eff 2 828 x m T n n 2 Based on the calculated values (Fig. 4c), Ni 3 Se 4 exhibits a low-spin (LS) state, while Ni 3 Se 4 -10 %Mn shows a high-spin (HS) state. We propose that the observed difference in spin states between the two samples arises from two primary factors [64–66]. Furthermore, electron paramagnetic resonance (EPR) spectra (Fig. S16) reveal an increased signal intensity for Mn10 %-Ni 3 Se 4 compared to undoped Ni 3 Se 4 , indicating a greater number of unpaired electrons and confirming the spin-state modulation induced by Mn doping [66–68]. The strong electron-withdrawing effect was introduced by Mn substitution in the local structure surrounding the central metal in Ni 3 Se 4 . This electronic interaction modulates the adjacent coordination environment, leading to a rearrangement of the d-orbital electron configuration and a subsequent change in the spin state. The electronic configurations of Ni 2+ and Mn 2+ are shown in Fig. 4(d), where there is only one lone electron in the d xy orbital in the t 2g orbital of Mn 2+ , and the orbital is in an unsaturated state. During coupling with the ligand atoms (Fig. 4e), Ni 2+ is completely occupied in the t 2g orbital, resulting in electron repulsion between Se(O) 2− and Ni 2+ . The unpaired electrons in the d orbital of Mn 2+ (t 2g ) have a strong attraction to the ligand atoms, so the electronic interaction between Mn 2+ and Ni 2+ can be enhanced by the electronic bridging of Se(O) 2− under the transition of the coordination atoms. As a result, partial charge transfer from Ni 2+ to Mn 2+ will easily occur. 725 Fig. 4(e) illustrates the electron configurations in the 3d orbitals of Mn 2+ , Ni 3+ , and Ni 2+ in both HS and LS states. In the HS state, Mn 2+ possesses five singly occupied orbitals (d xy 1d xz 1d yz 1d z 2 1d x 2 − - y 2 1), while Ni 2+ and Ni 3+ show two (d xy 2d xz 2d yz 2d z 2 1d x 2 −y 2 1) and three (d xy 1d xz 2d yz 2d z 2 1d x 2 −y 2 1) singly occupied orbits, respectively. In contrast, the LS states of Mn 2+ ,Ni 2+ , and Ni 3+ exhibit one, zero, and one singly occupied orbital(s), respectively, with more fully paired electrons in lower energy orbitals. In addition, the electronic configurations of Mn 3+ and Mn 4+ are shown in Fig. S17, with lower dorbital utilization and significantly stronger electron transfer ability than Mn 2+ . Notably, HS states exhibit higher electrophilicity than their LS counterparts, which facilitates improved charge transfer and enhanced conductivity during the MOR. This characteristic contributes to the superior catalytic activity observed for Mn-doped Ni 3 - Se 4 in the HS configuration [69–71]. Furthermore, DFT calculations were performed to study the effect of Mn doping on the MOR mechanism. As seen in Fig. 5(a) for the charge density analyses of Ni 3 Se 4 and Mn-doped Ni 3 Se 4 , it is revealed that Mn incorporation does not disrupt the original crystal structure, consistent with XRD results. As the Mn content increases, charge redistribution becomes more evident, particularly around the Ni–Se–Mn coordinatio n sites, indicating enhanced electronic interaction. This directional charge delocalization suggests that Mn doping effectively tunes the local electronic environment of the Ni sites, likely through d-band center modulation, which may contribute to improved catalytic activity [72–74]. Taking into account the surface reconstruction of the catalyst in alkaline media [75–80], two representative structural models were established: SeO x -NiOOH (M1) for the undoped system and SeO x - NiMnOOH (M2) for the Mn-doped counterpart, as shown in Fig. 5 (b). To evaluate the catalytic performance, we calculated and compared the charge density distributions, partial density of states (PDOS), d-band centers ( e d ), and adsorption energies of intermediates involved in the methanol oxidation process. The d-band centers of SeO x -NiOOH and SeO x -NiMnOOH were found to be −1.08 and −0.941 eV relative to the Fermi level (Fig. 5c), respectively. The shift of e d closer to the Fermi level in the Mn-doped model (M2) suggests stronger interactions between the catalyst surface and reaction intermediates [81–83]. As shown in Fig. 5(d), the electronic structure of SeO x -NiMnOOH exhibits a pronounced overlap between the Ni and O states, enhancing electron delocalization and breaking the electron transfer limitation seen in the undoped model. This increased electron density facilitates faster electron transfer from the catalyst to the adsorbates, thereby reducing the energy barrier and improving methanol oxidation efficiency. Fig. S18 shows the crystal orbital Hamilton population (COHP) results comparing SeO x -NiOOH and SeO x -NiMnOOH models. Specifically, the integrated COHP values quantify bond strength between metal (Ni or Mn) and oxygen atoms with −3.65 and −4.29 eV for SeO x -NiOOH and SeO x -NiMnOOH, respectively. Thus, a more negative COHP indicates stronger bonding interaction by the Mn doping, leading to improved structural robustness and enhanced catalytic performance [84–86]. DFT-calculated adsorption energies for key intermediates further support this finding. The optimized intermedi ate adsorption structure on SeO x -NiMnOOH is shown in Fig. 5(e). For M2, the rate-determining step is the hydrogenation of CO* to HCOOH*, requiring only 0.38 eV. In contrast, for M1, the rate-determining step is the dehydrogenation of CH 3 OH* to CH 3 O*, with a higher Y. Zhang, Y. Ma, J. Yu et al. Journal of Energy Chemistry 112 (2026) 720–729 Fig. 4. (a) Magnetic field dependence of magnetization curves and comparison of coercivity of Ni 3 Se 4 and Ni 3 Se 4 -10 %Mn NPs. (b) ZFC magnetization curves of Ni 3 Se 4 -10 %Mn NPs. (c) Unpaired electron calculations in the 3d orbitals of Ni 3 Se 4 and Ni 3 Se 4 -10 %Mn NPs. (d) Schematic diagram of the electron configurations of Ni 2+ and Mn 2+ and the schematic diagram of the electronic coupling between Ni and Mn in Ni 3 Se 4 -10 %Mn NPs. (e) Schematic diagram of the orbital occupation of 3delectrons in the LS and HS states of Mn 2+ ,Ni 3+ , and Ni 2+ . 726 energy barrier of 0.45 eV. These results confirm that SeO x - NiMnOOH exhibits superior catalytic performance due to its optimized electronic structure and reduced adsorption energy for key intermediates in the methanol conversion pathway. Y. Zhang, Y. Ma, J. Yu et al. Journal of Energy Chemistry 112 (2026) 720–729 Fig. 5. (a) Charge density analysis, (b) atomic models, (c) d-band center, and (d) PDOS plots of SeO x -NiOOH and SeO x -NiMnOOH. (e) Adsorption energy plot for the methanol conversion process on SeO x -NiOOH and SeO x -NiMnOOH. 3. Conclusion s In summary, a highly effective electrocatalyst based on Ni 3 Se 4 NPs was synthesized via a simple hydrothermal method, with varying Mn doping ratios achieved by adjusting the Ni/Mn precursor ratio. The MOR performance of Ni 3 Se 4 -x%Mn (x = 0, 5, 10, and 15) was systematically evaluated and optimized. Among the series, the electrode doped with 10 % Mn demonstrated the best performance, delivering a MOR current density of approximately 190 mA cm −2 at 1.6 V. After 18 h of continuous CA operation at 1.6 V, the current density remained at 123 mA cm −2 , significantly outperforming many previously reported Ni-based electrocatalysts. IC measurements confirmed nearly 100 % FE for methanolto-formic acid conversion. Mechanistic investigations revealed that Mn doping introduces unsaturated t 2g orbitals, which modify the electronic configuration, promote the Ni 2+ → Ni 3+ transition, and increase the number of singly occupied orbitals. Simultaneously, Mn incorporation increases particle size, elongates Ni–Se(O) bonds, and reduces orbital occupancy, collectively shifting the material from an LS to an HS state. This spin state transition results in a more active electronic configuration that is better suited for facilitating electron transfer during the MOR process. 727 Experimental section Experimental details can be found in the Supporting Information. Y. Zhang, Y. Ma, J. Yu et al. Journal of Energy Chemistry 112 (2026) 720–729 CRediT authorship contribution statement Yong Zhang: Writing – original draft, Data curation, Conceptualization. Yi Ma: Resources, Methodology. Jing Yu: Methodology, Investigation. Canhuang Li: Resources, Investigation. Jordi Arbiol: Resources, Methodology, Investigation. Xiaoxi Wang: Methodology, Investigation. Ning Jian: Validation, Resources. Huan Ge: Resources, Methodology, Formal analysis. Luming Li: Software, Resources, Formal analysis. Andreu Cabot: Writing – review & editing, Supervision. Junshan Li: Writing – review & editing, Supervision, Resources, Funding acquisition, Data curation. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationsh ips that could have appeared to influence the work reported in this paper. Acknowledgments This research was financially supported by the Sichuan Science and Technology Program (Grant No. 2025NSFSC0139), and the China Postdoctoral Science Foundation (Grant No. 2023MD734228). The authors extend their gratitude to Ms. Zhou Yuke (from Scientific Compass www.shiyanjia.com) for providing invaluable assistance with the XPS analysis. ICN2 acknowledges funding from Generalitat de Catalunya 2021SGR00457. This study is part of the Advanced Materials programme and was supported by MCIN with funding from European Union NextGenerationEU (PRTR-C17.I1) and by Generalitat de Catalunya (In-CAEM Project). The authors thank the support from the project AMaDE (PID2023-149158OB-C43), funded by MCIN/ AEI/10.13039/501100011033/ and by ‘‘ERDF A way of making Europe”, by the ‘‘European Union”. ICN2 is supported by the Severo Ochoa program from Spanish MCIN / AEI (Grant No.: CEX2021001214-S) and is funded by the CERCA Programme / Generalitat de Catalunya. Part of the present work has been performed in the framework of Universitat Autònoma de Barcelona Materials Science PhD program. JY has received funding from the CSC-UAB PhD scholarship program. ICN2 is founding member of e-DREAM [87]. Appendix A. Supplementa ry material Supplementary data to this article can be found online at https://doi.org/10.1016/j.jechem.2025.08.085. References [1] J. Tian, C. Cao, D.-D. Ma, S.-G. Han, Y. He, X.-T. Wu, Q.-L. Zhu, Small Struct. 3 (2022) 2100134. [2] H. Tu, Y. Zhong, Z. Yang, C. Zhang, Y. Ma, Y. Zhang, N. Jian, H. Ge, J. Li, Catal 15 (2025) 516. [3] J. Li, L. Li, X. Ma, X. Han, C. Xing, X. Qi, R. He, J. Arbiol, H. Pan, J. Zhao, J. Deng, Y. Zhang, Y. Yang, A. Cabot, Adv. Sci. 10 (2023) 2300841. [4] J. Li, R. Wei, X. Wang, Y. Zuo, X. Han, J. Arbiol, J. Llorca, Y. Yang, A. Cabot, C. Cui, Angew. Chemie Int. Ed. 59 (2020) 20826–20830. [5] X.L. Liu, Y.C. Jiang, J.T. Huang, W. Zhong, B. He, P.J. Jin, Y. Chen, Carbon Energy 5 (2023) e367. [6] G. Han, M. Li, H. Liu, W. Zhang, L. He, F. Tian, Y. Liu, Y. Yu, W. Yang, S. Guo, Adv. Mater. 34 (2022) 2202943. [7] L. He, M. Li, L. Qiu, S. Geng, Y. Liu, F. Tian, M. Luo, H. Liu, Y. Yu, W. Yang, S. Guo, Nat. Commun. 15 (2024) 2290. [8] Y. Qi, X. Xiao, Y. Mei, L. Xiong, L. Chen, X. Lin, Z. Lin, S. Sun, B. Han, D. Yang, Y. Qin, X. Qiu, Adv. Funct. Mater. 32 (2022) 2111615. [9] Z.M. Wang, Q.L. Hong, X.H. Wang, H. Huang, C. Yu, S.N. Li, Acta Phys. - Chim. Sin. 39 (2023) 2303028. [10] Y. Hao, C. Qiao, S. Zhang, Y. Zhu, L. Ji, C. Cao, J. Zhang, Energy Mater. Adv. 4 (2023) 0040. [11] Z.X. Ge, Y. Ding, T.J. Wang, F. Shi, P.J. Jin, P. Chen, B. He, S. Bin Yin, Y. Chen, J. Energy Chem. 77 (2023) 209–216. 728 [12] H. Wang, R.L. Smith, X. Qi, J. Energy Chem. 101 (2025) 535–561. [13] A.W. Cook, K.M. Waldie, ACS Appl. Energy Mater. 3 (2020) 38–46. [14] S. Sankar, N. Watanabe, G.M. Anilkumar, B.N. Nair, S.G. Sivakamiammal, T. Tamaki, T. Yamaguchi, Catal. Sci. Technol. 9 (2019) 493–501. [15] J. Liu, Q. Wang, T. Li, Y. Wang, H. Li, A. Cabot, Nano Res. 16 (2023) 2041–2048. [16] J. Liu, H. Liu, Q. Wang, T. Li, T. Yang, W. Zhang, H. Xu, H. Li, X. Qi, Y. Wang, A. Cabot, Chem. Eng. J. 486 (2024) 150258. [17] L. Zhao, Z. Zhu, J. Wang, J. Zuo, H. Chen, X. Qi, X. Niu, D.J. Blackwood, J.S. Chen, R. Wu, Adv. Mater. 37 (2025) 2502457. [18] H. Cheng, J. Wang, C. Wu, Z. Liu, Energy Mater. Adv. 4 (2023) 0067. [19] N. Kakati, J. Maiti, S.H. Lee, S.H. Jee, B. Viswanathan, Y.S. Yoon, Chem. Rev. 114 (2014) 12397–12429. [20] J. Liu, T. Li, Q. Wang, H. Liu, J. Wu, Y. Sui, H. Li, P. Tang, Y. Wang, Chem. Sci. 15 (2024) 16660–16668. [21] Q. Zhang, Q. Jiang, X. Yang, C. Zhang, J. Zhang, L. Yang, H. He, G. Ying, H. Huang, J. Energy Chem. 93 (2024) 419–428. [22] C. Liu, W. Zhou, J. Zhang, Z. Chen, S. Liu, Y. Zhang, J. Yang, L. Xu, W. Hu, Y. Chen, Y. Deng, Adv. Energy Mater. 10 (2020) 2001397. [23] C. Wan, J. Jin, X. Wei, S. Chen, Y. Zhang, T. Zhu, H. Qu, J. Mater. Sci. Technol. 124 (2022) 102–108. [24] J. Li, L. Li, J. Wang, A. Cabot, Y. Zhu, ACS Energy Lett. 9 (2024) 853–879. [25] C. Cao, D.D. Ma, J. Jia, Q. Xu, X.T. Wu, Q.L. Zhu, Adv. Mater. 33 (2021) 2008631. [26] Y. Ma, L. Li, Y. Zhang, N. Jian, H. Pan, J. Deng, J. Li, J. Colloid Interface Sci. 663 (2024) 971–980. [27] Z. Chen, N. Han, R. Zheng, Z. Ren, W. Wei, B. Ni, SusMat 3 (2023) 290–319. [28] N. Jian, H. Ge, Y. Ma, Y. Zhang, L. Li, J. Liu, J. Yu, C. Li, J. Li, Sci. Energy Environ. 2 (2025) 3. [29] Y. Liu, R. Wu, Y. Jin, J. Dong, H. Li, J. Wang, eScience (2025). https://doi.org/ 10.1016/j.esci.2025.100430. [30] V.T.T. Phan, Q.P. Nguyen, B. Wang, I.J. Burgess, J. Am. Chem. Soc. 146 (2024) 4830–4841. [31] Y. Yan, J. Zhong, R. Wang, S. Yan, Z. Zou, J. Am. Chem. Soc. 146 (2024) 4814– 4821. [32] Y.P. Liu, S.F. Zhao, S.X. Guo, A.M. Bond, J. Zhang, G. Zhu, C.L. Hill, Y.V. Geletii, J. Am. Chem. Soc. 138 (2016) 2617–2628. [33] J.Y. Zhang, Y. Yan, B. Mei, R. Qi, T. He, Z. Wang, W. Fang, S. Zaman, Y. Su, S. Ding, B.Y. Xia, Energy Environ. Sci. 14 (2021) 365–373. [34] F. Si, J. Liu, Y. Zhang, B. Zhao, Y. Liang, X. Wu, X. Kang, X. Yang, J. Zhang, X.Z. Fu, J.L. Luo, Small 19 (2023) 2205257. [35] S. Zhou, X. Miao, X. Zhao, C. Ma, Y. Qiu, Z. Hu, J. Zhao, L. Shi, J. Zeng, Nat. Commun. 7 (2016) 11510. [36] L. Li, X. Zhang, M. Humayun, X. Xu, Z. Shang, Z. Li, M.F. Yuen, C. Hong, Z. Chen, J. Zeng, M. Bououdina, K. Temst, X. Wang, C. Wang, ACS Nano 18 (2024) 1214– 1225. [37] J. Yu, C. Huang, O. Usoltsev, A.P. Black, K. Gupta, M.C. Spadaro, I. Pinto-Huguet, M. Botifoll, C. Li, J. Herrero-Martín, J. Zhou, A. Ponrouch, R. Zhao, L. Balcells, C.Y. Zhang, A. Cabot, J. Arbiol, ACS Nano 18 (2024) 19268–19282. [38] C. Huang, J. Yu, C.Y. Zhang, Z. Cui, J. Chen, W.H. Lai, Y.J. Lei, B. Nan, X. Lu, R. He, L. Gong, J. Li, C. Li, X. Qi, Q. Xue, J.Y. Zhou, X. Qi, L. Balcells, J. Arbiol, A. Cabot, Adv. Mater. 36 (2024) 2400810. [39] C. Li, J. Yu, D. Yang, H. Li, Y. Cheng, Y. Ren, X. Bi, J. Ma, R. Zhao, Y. Zhou, J. Wang, C. Huang, J. Li, I. Pinto-Huguet, J. Arbiol, H. Zhang, S. Xin, A. Cabot, J. Am. Chem. Soc. 147 (2025) 8250–8259. [40] C.Y. Zhang, C. Zhang, G.W. Sun, J.L. Pan, L. Gong, G.Z. Sun, J.J. Biendicho, L. Balcells, X.L. Fan, J.R. Morante, J.Y. Zhou, A. Cabot, Angew. Chemie - Int. Ed. 61 (2022) e202211570. [41] L. Yang, R. He, M. Botifoll, Y. Zhang, Y. Ding, C. Di, C. He, Y. Xu, L. Balcells, J. Arbiol, Y. Zhou, A. Cabot, Adv. Mater. 36 (2024) 2400572. [42] J. Li, J. Yu, Y. Zhang, C. Li, Y. Ma, H. Ge, N. Jian, L. Li, C.Y. Zhang, J.Y. Zhou, J. Arbiol, A. Cabot, Adv. Funct. Mater. 35 (2025) 2501485. [43] Y. Zhang, R. Liu, Y. Ma, N. Jian, H. Ge, H. Pan, Y. Zhang, C. Zhang, Y. Liu, J. Deng, L. Li, J. Zhao, J. Yu, A. Cabot, J. Li, Inorg. Chem. 63 (2024) 23328–23337. [44] S. Huang, Y. Wang, J. Wan, Z. Yan, Y. Ma, G. Zhang, S. Wang, Appl. Catal. B Environ. 319 (319) (2022) 121913. [45] Y. Li, S. Shu, L. Huang, J. Liu, J. Liu, J. Yao, S. Liu, M. Zhu, L. Huang, J. Colloid Interface Sci. 633 (2023) 60–71. [46] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-Ray Photoelectron Spectroscopy: A Reference Book of Standard Spectra for Identification and Interpretation of XPS Data, Perkin-Elmer Corp, Minnesota, 1992. [47] J. Li, X. Tian, X. Wang, T. Zhang, M.C. Spadaro, J. Arbiol, L. Li, Y. Zuo, A. Cabot, Inorg. Chem. 61 (2022) 13433–13441. [48] J. Ren, Y. Zhang, J. Li, J. Liu, J. Hu, C. Li, Y. Ke, J. Zhao, A. Cabot, B. Tang, Dalt. Trans. 53 (2024) 18736–18744. [49] J. Li, Y. Ma, J. Yu, L. Li, H. Yang, W. Gu, J. Shi, J. Wang, Y. Zhu, ChemSusChem 18 (2024) e202401098. [50] J. Qi, Y. Du, Q. Yang, N. Jiang, J. Li, Y. Ma, Y. Ma, X. Zhao, J. Qiu, Nat. Commun. 14 (2023) 6263. [51] C. Xia, H. Liang, J. Zhu, U. Schwingenschlögl, H.N. Alshareef, Adv. Energy Mater. 7 (2017) 1602089. [52] M. Huynh, C. Shi, S.J.L. Billinge, D.G. Nocera, J. Am. Chem. Soc. 137 (2015) 14887–14904. [53] I. Danaee, M. Jafarian, F. Forouzandeh, F. Gobal, M.G. Mahjani, Int. J. Hydrogen Energy 34 (2009) 859–869. [54] H. Xiao, W. Zhu, X. Peng, Y. Wu, Y. Wang, H. Jiang, N. Li, Y. Yang, C. Xie, J. Phys. Chem. C 128 (2024) 18273–18283.