Tuning the oxygen-containing microenvironment of the Pt–Ni hetero-interface to accelerate alkaline hydrogen oxidation

Abstract

Heterojunction electrocatalysts composed of platinum and non-platinum group Ni metals have shown remarkable reactivity and economic advantages during the sluggish hydrogen oxidation reaction (HOR) in alkaline exchange membrane fuel cells (AEMFCs). However, little attention has been paid to the effect of the oxygen-containing microenvironment of the heterogeneous interface on the performance of HOR. Herein, Pt–Ni metallic heterostructures loaded on carbon nanotubes with different oxygen-containing microenvironments were controllably synthesized (Pt–Ni/CNT-x, where x represents different oxygen contents) through a galvanic replacement-annealing reduction-metal oxidation process, exhibiting remarkable alkaline HOR performance. Among them, the electrocatalyst Pt–Ni/CNT-p with an appropriate oxygen content exhibited a mass-specific kinetic current and exchange current density of 881.1 A g_Pt⁻¹ (at an overpotential of 50 mV) and 2.12 mA cm⁻², respectively, not only outperforming its Pt/C counterpart but also making it among the best reported Ni-based HOR electrocatalysts to date. Furthermore, this electrocatalyst exhibited negligible activity decay during long-term electrolysis and good CO tolerance capability. (Semi)quantitative analyses verified that the elevation of intrinsic activities for HOR was closely correlated with the oxygen-containing microenvironment of the Pt–Ni hetero-interface. Further experiments and theoretical calculations indicated that the slightly oxidized heterogeneous interface weakens the adsorption energy of the OH* intermediates, which can thus easily combine with H* species to form H₂O. On the contrary, an excessively oxidized heterogeneous interface will strengthen the adsorption of OH* and thus increase the reaction energy barrier of the above crucial Volmer step. This study not only promotes the development of alkaline HOR heterojunction electrocatalysts, but also provides a scheme to explore the influence of the microenvironment on their catalytic activity.

---

Di Zhao

Di Zhao is currently an associate professor at the School of Chemistry and Chemical Engineering, Beijing Institute of Technology. She received her PhD degree from the School of Chemistry and Chemical Engineering, Beijing Institute of Technology, in 2017. She worked as a postdoctoral fellow under the supervision of Prof. Yadong Li at the Department of Chemistry, Tsinghua University, from 2017 to 2020. Her research interests include the design, synthesis and characterization of single-atom materials, nanostructured materials and their applications in electrocatalytic energy conversion.

1 Introduction

Alkaline exchange membrane fuel cells (AEMFCs) are gaining increasing attention due to their ability to operate in low-corrosive alkaline environments as well as their use of more economical non-noble metal electrocatalysts, cost-effective bipolar plates, and alkaline separators.^1–3 They also represent a promising sustainable energy-conversion device that could help meet the U.S. Department of Energy's (DOE) goal of reducing the cost of fuel cell vehicles to $30 per kW_net. However, the kinetically sluggish anodic reaction (i.e., hydrogen oxidation reaction (HOR)) in alkaline conditions is a significant challenge for the commercial development of AEMFCs.⁴ As the most widely used alkaline HOR electrocatalyst, platinum (Pt) has become the representative metal among platinum group metals (PGM) due to its optimal hydrogen intermediate (H*) binding energy. However, the high use of Pt in these electrocatalysts seriously limits their industrial scale applications. Thus, it would be of fundamental and technological significance to explore viable alternatives that can exhibit high activity and cost-effectiveness for alkaline HOR.^4–6 At present, nickel (Ni)-based materials, as low-cost and abundant non-platinum group metal (non-PGM) electrocatalysts, have been reported to significantly promote HOR activity due to their high oxyphilic nature.^7,8 However, the documented Ni-based electrocatalysts still cannot match the performance of PGM electrocatalysts in terms of their intrinsic catalytic activity and stability, due to the extremely strong adsorption of hydroxy species (OH*) by pure nickel species. To sum up, there is an urgent need to develop a strategy for simultaneously solving the two most prominent bottlenecks of PGMs and non-PGMs mentioned above to promote their commercial application as alkaline HOR electrocatalysts.

In recent years, heterojunction electrocatalysts composed of different components have shown remarkable reactivity and selectivity in various electrocatalytic reactions due to their unique local electronic structures.^9,10 More importantly, the hetero-interface can provide bifunctional adsorption sites for H* and OH* in the alkaline HOR, which would be beneficial for solving the above bottleneck problems in the development of alkaline HOR electrocatalysts. So far, some studies have proposed optimizing the OH* adsorption sites by modifying the surface with metal compounds, such as metal oxides,^11,12 nitrides,^13,14 borides,¹⁵ carbides,¹⁶ or hydroxides,¹⁷ to form heterogeneous structures. It is noteworthy that oxygen-containing species have been proven to be widely present on the surface of Ni-based heterogeneous electrocatalysts due to their susceptibility to oxidation.^12–16 However, most reports have tended to focus on the regulation of the two heterogeneous phases, while little attention has been paid to the oxygen-containing microenvironment change of the heterogeneous interface.

As a common substrate for electrocatalysis, carbon nanotubes have high conductivity, large specific surface area and structural stability, which can efficiently disperse active components and accelerate electron transport, which can significantly improve the catalytic activity and stability. Motivated by these discoveries, we successfully tuned the oxygen-containing microenvironment of a Pt–Ni hetero-interface loaded on carbon nanotubes (Pt–Ni/CNT-x, where x represents different oxygen contents) though controlling the galvanic replacement-annealing reduction-metal oxidation processes at efficient synergistic HOR sites. Impressively, the Pt–Ni/CNT-p electrocatalyst with an appropriate oxygen content exhibited an excellent HOR performance and long-term stability in alkaline systems. It displayed an admirable anodic current density of 1.91 mA cm⁻² and mass-specific kinetic current of 881.1 A g_Pt⁻¹ (at an overpotential (η) of 50 mV), respectively, which were not only significantly better than those of commercial Pt/C but also among the best values previously reported for HOR electrocatalysts under equivalent conditions. Furthermore, this electrocatalyst exhibited negligible activity decay during 20 h long-term electrolysis and an excellent CO tolerance capability. Controlled experiments verified that the elevation of the intrinsic activities was positively correlated with increasing the ratio of Ni⁰/Ni²⁺. Also, O₂ temperature-programmed desorption (O₂-TPD) spectroscopy and density functional theory (DFT) simulations revealed that an appropriate oxygen content on the Ni substrate could optimize the OH* intermediates' adsorption strength at the interfacial Ni sites, thus reducing the reaction energy barrier and accelerating the alkaline HOR kinetics. This study not only promotes the development of anode HOR electrocatalysts for alkaline exchange membrane fuel cells, but also provides guidance for exploring the role of interfacial microenvironments on the electrocatalytic reactivity.

2 Results and discussion

2.1 Synthesis and structural characterizations of Pt–Ni/CNT-p

The sample with the partially oxidized Pt–Ni interface loaded on carbon nanotubes (Pt–Ni/CNT-p) was experimentally synthesized via a galvanic replacement-annealing process, as illustrated in Scheme 1 (see details in the Experimental section in the ESI†). First, Pt nanoclusters were built in nickel-supported carbon nanotubes (Ni/CNTs) (Fig. S1†) through a controlled galvanic replacement reaction in chloroplatinic (H₂PtCl₆) aqueous solution under constant heating conditions with the absence of other reductants. This gave nanoscale Pt clusters-coupled Ni nanocrystals with an overoxidized Pt–Ni interface on the carbon nanotubes (denoted as Pt–Ni/CNT-o). Subsequently, the Pt–Ni/CNT-o sample was annealed and reduced in a H₂/Ar atmosphere to eliminate excessive oxygen species at the Pt–Ni interface, and this sample was denoted as Pt–Ni/CNT-p. Based on the reduction potential of the metal with respect to a standard hydrogen electrode (SHE), Fe³⁺ can oxidize Ni and cannot oxidize Pt (Table S1†). So, to further verify the adverse effect of excessive oxygen species, the Pt–Ni/CNT-p sample was then re-oxidized with Fe³⁺ solution, and this sample was denoted as Pt–Ni/CNT-r.

Scheme 1 . Schematic illustration of the preparation process for the electrocatalyst Pt–Ni/CNT-x (x = o, p, r).

Field-emission scanning electron microscopy (FE-SEM) images (Fig. 1a, b and S2†) revealed that the as-obtained Pt–Ni/CNT-p had a uniform tube-like morphology. Transmission electron microscopy (TEM) images (Fig. 1c, d and S3†) showed that Pt–Ni/CNT-p had a highly curved structure, whose pipe radius was approximately 50 nm in size. Meanwhile, the Pt–Ni/CNT-o and -r samples displayed similar morphologies to that of Pt–Ni/CNT-p (their TEM images are shown in Fig. S4 and 5†). These features increase the accessibility to the active sites and improve the H₂ mass transport during electrolysis. In order to better understand the distribution of Pt elements in the Ni matrix, a high-resolution TEM image (HR-TEM) (Fig. 1e) was obtained and this clearly revealed a typical Pt–Ni heterogeneous interface with Pt (111) and Ni (111) features on both sides. Many Pt–Ni heterogeneous nanoparticles were loaded on nanotubes and most of their diameters were located at the range of 3–15 nm (Fig. S6†). In addition, for the Pt–Ni/CNT-r and -o samples, the Pt–Ni heterostructures were clear and similar (Fig. S7†). At the same time, the inset selected area electron diffraction (SAED) pattern in Fig. 1e confirmed the polycrystalline co-existence of the metals Pt and Ni in Pt–Ni/CNT-p.¹⁷ Finally, the EDS spectra and corresponding elemental mapping images (Fig. 1f, S8 and 9†) of Pt–Ni/CNT-p common demonstrated that the Pt and Ni elements were not only coupled to each other, but were also uniformly distributed on the nanotubes, which was exactly consistent with the HR-TEM and SAED results above. Inductively coupled plasma optical emission spectroscopy (ICP-OES) revealed a Pt and Ni element content of up to 5.88 wt% and 28.1 wt% in Pt–Ni/CNT-p, respectively. Meanwhile, the Fe content was 0.26 wt% in Pt–Ni/CNT-r, excluding the presence of adsorbed Fe species and the partially passivated role of Fe³⁺ (Table S2†).

Fig. 1 Morphology characterization. (a and b) FE-SEM images ((b) magnified image of single carbon nanotubes) and (c and d) TEM images of Pt–Ni/CNT-p. (e) High-resolution TEM image of the Pt–Ni interface (inset: corresponding SAED pattern image and crystal plane parameter). (f) HAADF-STEM images of Pt–Ni/CNT-p, and the corresponding EDS element mapping distributions of C, Ni, Pt, O.

To obtain information on the valence state and structure of the Pt–Ni/CNT-p electrocatalyst, we applied multiple characterization tools. Fig. 2a shows the X-ray diffraction (XRD) patterns of the Pt–Ni/CNT-x (x = o, p, r) samples, which all exhibited similar diffraction peaks at 2θ of 44.4°, 51.8°, and 76.2°, which could be indexed to the (111), (200), and (220) planes of the typical face-centered cubic (fcc) phase of Ni (JCPDS No. 89-7128); along with four prominent peaks corresponding to the (111), (200), (220) and (311) planes of Pt (JCPDS No. 65-2868), respectively, which indicates that Pt and Ni phases co-existed in the above sample.^17,18 Since the redox state determines the electronic structure, the surface chemical composition information of the as-prepared samples was studied by X-ray photoelectron spectroscopy (XPS). As shown in Fig. 2b, the high-resolution Ni 2p spectra could be split into six peaks, of which the peaks at 873.5 and 855.8 eV were assigned to Ni²⁺, which were probably due to surface micro-oxidation (Fig. S10†), while the peaks at 870.2 and 852.9 eV corresponded to metallic state Ni⁰, and the peaks at 880.2 and 861.3 eV corresponded to satellite peaks of Ni.^19,20 Compared with Pt–Ni/CNT-o, the peak area of Ni²⁺ for Pt–Ni/CNT-p was evidently decreased, indicating that the surface overoxidized Ni was partially reduced. However, after re-oxidizing with Fe³⁺ solution, the peak area of Ni²⁺ was evidently increased. In addition, the Pt 4f high-resolution spectrum showed mainly the Pt⁰ state in Pt–Ni/CNT-p (Fig. 2c), while a small number of Pt²⁺ sub-peaks existed due to the trace oxidation of the sample surface during exposure to air.^20,21 The Pt–Ni/CNT-x (x = o, p, r) sample had similar contents of Pt⁰ and Pt²⁺.

Fig. 2 Structural analysis. (a) XRD patterns of Pt–Ni/CNT-p and the control samples. (b–d) High-resolution Ni 2p (b), Pt 4f (c) and O 1s (d) XPS spectra of Pt–Ni/CNT-p and the contrast samples. (e) Area percentages of the fitting sub-peaks of Pt⁰ 4f and Ni²⁺ 2p in the XPS spectra.

Furthermore, the O 1s high-resolution spectrum (Fig. 2d) and oxygen contents (Table S3†) showed that the signal intensity and oxygen content of the Pt–Ni/CNT-p catalyst were both the lowest among the three samples (see the illustration in Fig. 2d), indicating the lower oxygen content on the catalyst surface, which was consistent with the above conclusion. In order to determine the variation of the oxidation states more precisely, we calculated the area percentages of Ni²⁺ and Pt⁰ from fitting the sub-peaks of Ni 2p and Pt 4f, respectively. As shown in Fig. 2e, compared with Pt–Ni/CNT-o, the integral area percentage of Ni²⁺ in Pt–Ni/CNT-p was significantly decreased. However, for Pt–Ni/CNT-r obtained the re-oxidation of Pt–Ni/CNT-p, the integral area percentage of Ni²⁺ was restoratively increased again. Meanwhile, the integral area percentages of Pt⁰ were almost the same for the Pt–Ni/CNT-x (x = o, p, r) catalysts. In addition, the carbon signal was nearly identical for all three samples (Fig. S11†), demonstrating that the reduction/oxidation treatment did not change the exposed surface groups of the CNTs. To sum up, the above results indicate that the oxidation states of Ni and Pt on the surfaces of the Pt–Ni metallic heterojunction were well controlled, which was useful for the follow-up study of the relationship between the oxygen-containing microenvironment and the electrocatalysis activity.

2.2 Electrocatalytic performances for the alkaline HOR

The electrochemical alkaline HOR performance of Pt–Ni/CNT-p was examined in a H₂-saturated 0.1 M KOH solution using a rotating disk electrode (RDE) with a standard three-electrode system. At the same time, a commercial 20 wt% Pt/C benchmark and control samples were tested for comparison. The steady-state linear sweep voltammetry (LSV) polarization curves for the HOR on the electrocatalysts are shown in Fig. 3a. At a potential greater than 0 V (all the potentials were calibrated based on the reversible hydrogen electrode (RHE)), the anodic current of Pt–Ni/CNT-p significantly increased with the potential, while a negligible anodic current was observed in the N₂-saturated electrolytes (Fig. S12†), indicating its remarkable activity toward the HOR. Notably, at an overpotential (η) of 50 mV, Pt–Ni/CNT-p exhibited the highest geometrical area-specific current density (j) of 1.91 mA cm⁻², and this value was even better than that of the Pt/C benchmark (1.42 mA cm⁻²). It was also noteworthy that the limiting current density (j_L) platform was consistent with the theoretical value, indicating that the anode current was generated by the oxidation of molecular hydrogen, rather than the false-positive current caused by double-layer capacitor charging or Ni metal autoxidation.²²

Fig. 3 Alkaline HOR performance. (a) HOR polarization curves of different electrocatalysts in a H₂-saturated 0.1 M KOH solution at a rotating speed of 1600 rpm with a scan rate of 5 mV s⁻¹. (b) HOR polarization curves recorded at various rotation speeds (inset shows the K–L plot at η = 50 mV). (c) Tafel plots with Butler–Volmer fitting curves. (d) Micro-polarization region. (e) Comparison of the electrochemical parameters of Pt–Ni/CNT-p (orange), Pt–Ni/CNT-o (lilac), Pt–Ni/CNT-r (brown) and Pt/C (carmine). (f) Comparison of the mass-specific kinetic current (j_k,m) at an overpotential of 50 mV and the exchange current (j₀) with those reported in previous studies. Detailed data are provided in Table S5.† (g) Chronoamperometry response of Pt–Ni/CNT-p and commercial Pt/C in a H₂-saturated aqueous solution of 0.1 M KOH at η = 50 mV. (h) Poisoning experiments of Pt–Ni/CNT-p and commercial Pt/C electrocatalysts with 1000 ppm CO/H₂.

The HOR polarization curves recorded at different electrode rotation speeds are shown in Fig. 3b, where it can be seen that the anodic j increased with the rotation rate due to the more rapid mass transport.^20,21 More reliably, we confirmed and extracted the kinetic current (j_k) for the HOR by the extrapolation intercept method of Koutecky–Levich (K–L), instead of directly applying the equation to an algebraic calculated j_k value, which would be more likely to mix false-positive currents and cause the j_k value to be erroneously large. A linear correlation was obtained by fitting j⁻¹ at η = 50 mV as a function of the square root of the reciprocal rotation speed (ω^−1/2), whereby the fitting result afforded a slope of 4.86 cm² mA⁻¹ s^−1/2 (see inset of Fig. 3b), which was extremely similar to the theoretical value (4.87 cm² mA⁻¹ s^−1/2).^20,22,23 This result revealed that the HOR catalytic behaviour on Pt–Ni/CNT-p was identical to the ideal/theoretical two-electron pathway. The intercept of the extrapolated line corresponded to the inverse of the purely kinetic current density, and the j_k for Pt–Ni/CNT-p was calculated to be 3.2 mA cm⁻² at η = 50 mV (see the inset in Fig. 3b). The calculated mass-normalized kinetic current density (j_m,k) at an overpotential of 50 mV for Pt–Ni/CNT-p was 881.1 A g_Pt⁻¹, while that of the Pt/C benchmark was 665.0 A g_Pt⁻¹, Pt–Ni/CNT-o was 774.3 A g_Pt⁻¹ and Pt–Ni/CNT-r was 293.7 A g_Pt⁻¹ (Fig. S13–15†). We found that the mass activity of Pt–Ni/CNT-p was higher than those of the typically reported Pt- and Ni-based HOR electrocatalysts, presenting significant performance benefits.^24,25 The geometrical area-exchange current densities (j₀) of the electrocatalysts were further extracted via fitting j_k according to the Butler–Volmer equation in the Tafel regions (Fig. 3c). It could be seen that the fitted curve of the Butler–Volmer equation was in good agreement with the experimental values, reconfirming that no false-positive current was introduced into the electrochemical parameters for the alkaline HOR.^22,26 Of note, the j₀ for Pt–Ni/CNT-p (2.12 mA cm⁻²) was higher compared with the commercial Pt/C counterpart (1.09 mA cm⁻²) Pt–Ni/CNT-o (1.9 mA cm⁻²) and Pt–Ni/CNT-r (0.7 mA cm⁻²) samples, displaying its strong intrinsic activity for the HOR. More importantly, the j₀ values were also estimated from linear fitting of the micro-polarization regions (Fig. 3d and Table S4†), and the two obtained results were almost consistent with each other.^26–28 Another important parameter is the symmetry factor (α), and the values were estimated according to the Bulter–Volmer equation for Pt–Ni/CNT-p and Pt/C as 0.6 and 0.5, respectively (Table S4†), both of which were very close to the theoretical value of 0.5, indicating their symmetrical HOR and HER electrocatalytic behaviours.^29–31Fig. 3e comprehensively summarizes the HOR performance parameters of the Pt–Ni/CNT-x (x = o, p, r) and Pt/C electrocatalysts. Overall, the performance of Pt–Ni/CNT-p was superior to those of the other electrocatalysts, including the Pt/C benchmark. Next, electrochemical impedance spectroscopy (EIS) was carried out to analyse the interfacial charge-transfer process for the alkaline HOR. Compared to the other reference catalysts, Pt–Ni/CNT-p displayed the shortest semicircle diameter in the Nyquist plots for the EIS (Fig. S16†), indicating the lowest electron-transfer resistance between the active site of Pt–Ni/CNT-p and the electrolyte during the alkaline HOR. From this, we could conclude that the change in internal interfacial charge-transfer resistance between the solution and the catalyst affected the external HOR properties. To sum up, the HOR performance of Pt–Ni/CNT-p was superior to most other reported electrocatalysts, including the Pt/C benchmark (Fig. 3f and Table S5†).

The durability of an electrocatalyst is particularly important for the alkaline HOR. Therefore, chronoamperometry was carried out to evaluate the stability of Pt–Ni/CNT-p (Fig. 3g). After continuous operation at an overpotential of 50 mV for 20 h, the Pt–Ni/CNT-p electrocatalyst showed an inconspicuous degradation and retained 91.7% of its initial activity (current retention rate). EIS measurements were also conducted to evaluate the difference in the electrochemical conductivity of Pt–Ni/CNT-p before and after the stability test (Fig. S17†). The two Nyquist plots show similar curves, which consisted of one semicircle in the high-frequency region, demonstrating the good stability of Pt–Ni/CNT-p. By sharp contrast, the catalytic activity of the Pt/C benchmark degraded to 10% in under 7 h, indicating the superior durability of the Pt–Ni/CNT-p electrocatalyst. More realistically, H₂ fuels are usually derived from industrial reforming gases and inevitably contain trace amounts of CO impurities (ranging from 100–30 000 ppm).³² The commercial Pt/C benchmark is susceptible to CO poisoning even at very low 10 ppm levels, resulting in decreased durability for AEMFCs.³³ Therefore, there is a crucial need to design a CO-resistant HOR electrocatalyst. The CO tolerance of each electrocatalyst was evaluated by recording time–current curves at η = 50 mV and with a 1000 ppm CO/H₂ gas mixture. As clearly presented in Fig. 3h, the current density of the Pt/C benchmark sharply dropped to 1% after ∼600 s continuous operation, indicating that it would not be functional under such conditions. Impressively, Pt–Ni/CNT-p retained 90.2% of its initial current density after 3000 s continuous service test, suggesting that it could sustain a high-level HOR activity under the 1000 ppm CO atmosphere.

2.3 HOR mechanism investigation and DFT calculations

Next, we conducted (semi)quantitative analyses of the Pt–Ni/CNT-x samples with different percentages of Ni²⁺ and Pt⁰ with different degrees of oxidation in order to establish a correlation with the HOR performances. As shown in Fig. 4a and S18,† the intrinsic activities of j₀ decreased along with the increase in the Ni²⁺ percentage, while there was no correlation with the percentage of Pt⁰. Furthermore, the hydrogen underpotential deposition (HUPD) with Pt species characteristics showed no change during the catalyst modification process (Fig. S19†). The experimental results confirmed that the HOR activity of the Pt–Ni/CNT-x samples in alkaline conditions should be predominantly determined by the degree of Ni oxidation. For the oxyphilic elements, the oxidation degree of the metal surface can significantly affect the alkaline HOR properties of the electrocatalyst via affecting the hydroxyl adsorption strength at the catalytic sites.^34,35 For surfaces with a single type of active site, the excessive adsorption of hydroxyl groups hinders the reaction of H*. In contrast, for dual-active-site surfaces, noncompeting OH* can react with the adjacent adsorbed H* to generate water, thus accelerating the overall HOR.³⁶ Next, O₂-TPD analysis was used to characterize the adsorption type and strength of the oxygen-containing intermediates on the surface Ni-related sites of the prepared electrocatalysts.³⁷ Fig. S20† shows that the desorption temperature of the main peak was the lowest for the Pt–Ni/CNT-p samples, indicating that it had a weaker *OH adsorption strength compared with the other Pt–Ni/CNT-o and Pt–Ni/CNT-r samples. The difference in adsorption strength at the Ni-related sites stemmed from oxidation degree of the Pt–Ni interface in the various samples. It could be concluded that the adsorption strength of *OH at the Ni-related sites decreased with the oxidation degree of the initial Pt–Ni interface until it reached the optimal state (Pt–Ni/CNT-p); whereby, as the Pt–Ni interface continued to be oxidized across the optimal state, its adsorption strength increased in a reverse trend, forming a “volcanic” diagram.^38,39

Fig. 4 DFT calculations. (a) Correlation between the percentage of Ni²⁺ 2p/Pt 4f and j₀. (b and c) Calculated *OH binding energies (△E_*OH) in the Pt–Ni-zero and Pt–NiO_x-middle and -excess models. (d) Gibbs free energy diagram of the HOR process in the pure Pt–Ni-zero and Pt–NiO_x-middle and -excess models. (e) Schematic illustration of the promotion mechanism of HOR in the Pt–NiO_x-middle model.

In order to better simulate and analyse the above experimental results, as shown in Fig. 4b, we prepared reference models with a concentration gradient of O atoms embedded into Pt–Ni interface (denoted as Pt–Ni-zero and Pt–NiO_x-middle, -excess, representing models with O atoms from zero, middle to excess, respectively). The *OH binding energy (ΔE_*OH) was then calculated for the different models and adsorption sites (Fig. S21–25†). The optimized ΔE_*OH for each model is presented in Fig. 4c. Notably, the Ni site on the Pt–NiO_x-middle showed an increased ΔE_*OH of 0.7 eV compared to 0.64 eV for Pt–Ni-zero, implying that the hydroxyl adsorption was weakened; then, as the oxidation degree continued to increase, ΔE_*OH decreased, indicating that the hydroxyl adsorption was strengthened, which was consistent with the results in Fig. S20.† To further investigate the oxidation effect on the HOR process, the Gibbs free energies of the reaction intermediates in the three models were calculated. As shown in Fig. 4d and e, the potential-determining step (PDS) for Pt–NiO_x-middle was different from that of the two Pt–Ni-zero and Pt–NiO_x-excess electrocatalysts, and we noticed that the free energy changes of PDS over Pt–NiO_x-middle was 0.55 eV (1/2H₂ + * → *H), which was significantly lower than those for Pt–NiO_x-excess (1.08 eV, * + OH⁻ → *OH) and Pt–Ni-zero (0.87 eV, * + OH⁻ → *OH) (Table S6†), thus Pt–NiO_x-middle exhibited superior HOR catalytic activity from a thermodynamics perspective.^40,41 Combining the above ΔE_*OH results in Fig. 4c, Pt–Ni-zero displayed strong *OH adsorption at the active sites, which could hardly react with *H in the Volmer step, thus hindering the overall reaction. In contrast, *OH adsorption at the Pt–NiO_x-middle interface was weakened, resulting in a decrease in the energy barrier of the Volmer step, then leading to a PDS shift and better HOR catalytic activity. However, for Pt–NiO_x-excess with overoxidation, its *OH adsorption was strengthened again compared with Pt–NiO_x-middle, leading the Volmer step to become the PDS again. Based on the above characterization results, we propose that adjusting the micro-oxidation environment of the catalytic interface is beneficial for improving the performance of the alkaline HOR.

3 Conclusions

In summary, a Pt–Ni/CNT-p electrocatalyst with an appropriate oxygen-containing microenvironment obtained through a hydrothermal synthesis-annealing reduction process was rationally developed, and the resulting electrocatalyst exhibited outstanding HOR activity in alkaline media, with a mass activity and exchange current density of 881.1 A g_Pt⁻¹ and 2.12 mA cm⁻², respectively. Experiments combined with DFT calculations revealed that the change in oxygen content on the Ni substrate was caused by the annealing reduction treatment, which could optimize the adsorption strength of OH* intermediates, and thus reduce the energy barrier of the overall reaction. This study not only promotes the development of HOR anode electrocatalysts for alkaline exchange membrane fuel cells but also provides a new way to explore interfacial micro-oxidation regulation to improve catalytic performance.

Data availability

The data supporting this article have been included as part of the main text and ESI.†

Author contributions

Jiantao Fu: synthesis of material; characterizations of catalysts; electrochemical testing; manuscript preparation. Wuyi Feng: analysis of electrochemical data; mechanism diagram drawing. Xinye Zheng: DFT calculations; mechanism exploration. Yingzheng Zhang: HR-TEM measurement. Di Zhao: design of research plan; review and editing of the manuscript; funding. Jiatao Zhang: review and editing of the manuscript; funding.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 22309011, 52272186, 22105116) and Beijing Institute of Technology Research Fund Program for Young Scholars. We thank the Analysis & Testing center, Beijing Institute of Technology.

Notes and references

1. T. Wang, M. Wang, H. Yang, M. Xu, C. Zuo, K. Feng, M. Xie, J. Deng, J. Zhong, W. Zhou, T. Cheng and Y. Li, Energy Environ. Sci., 2019, 12, 3522–3529.
2. L. Ban, D. Huang, Y. Liu, P. Liu, X. Bian, K. Wang, Y. Liu, X. Liu and J. He, Nanoscale, 2025, 17, 864–876.
3. Y. Wang, G. Wang, G. Li, B. Huang, J. Pan, Q. Liu, J. Han, L. Xiao, J. Lu and L. Zhuang, Energy Environ. Sci., 2015, 8, 177–181.
4. F. Gao and M. Gao, Acc. Chem. Res., 2023, 56, 1445–1457.
5. H. Liu, J. Fu, H. Li, J. Sun, X. Liu, Y. Qiu, X. Peng, Y. Liu, H. Bao, L. Zhuo, R. Cao, S. Zhang and J. Luo, Appl. Catal., B, 2022, 306, 121029.
6. Z. Zhuang, S. Giles, J. Zheng, G. Jenness, S. Caratzoulas, D. Vlachos and Y. Yan, Nat. Commun., 2016, 7, 10141.
7. T. Tang, X. Liu, X. Luo, Z. Xue, H. Pan, J. Fu, Z. Yao, Z. Jiang, Z. Lyu, L. Zheng, D. Su, J. Zhang, L. Zhang and J. Hu, J. Am. Chem. Soc., 2023, 145, 13805.
8. Y. Zhou, W. Yuan, M. Li, Z. Xie, X. Song, Y. Yang, J. Wang, L. Li, W. Ding, W.-F. Lin and Z. Wei, Nat. Energy, 2024, 9, 1297–1309.
9. Q. Xu, J. Zhang, H. Zhang, L. Zhang, L. Chen, Y. Hu, H. Jiang and C. Li, Energy Environ. Sci., 2021, 14, 5228–5259.
10. B. Chen, S. Sui, F. He, C. He, H. Cheng, S. Qiao, W. Hu and N. Zhao, Chem. Soc. Rev., 2023, 52, 7802–7847.
11. Y. Yang, X. Sun, G. Han, X. Liu, X. Zhang, Y. Sun, M. Zhang, Z. Cao and Y. Sun, Angew. Chem., Int. Ed., 2019, 58, 10644–10649.
12. F. Yang, X. Bao, P. Li, X. Wang, G. Cheng, S. Chen and W. Luo, Angew. Chem., Int. Ed., 2019, 58, 14179–14183.
13. L. Su, D. Gong, N. Yao, Y. Li, Z. Li and W. Luo, Adv. Funct. Mater., 2021, 31, 2106156.
14. F. Song, W. Li, J. Yang, G. Han, P. Liao and Y. Sun, Nat. Commun., 2018, 9, 4531.
15. F. Yang, P. Han, N. Yao, G. Cheng, S. Chen and W. Luo, Chem. Sci., 2020, 11, 12118–12123.
16. W. Ji, C. Zhan, D. Li, Y. Xu, Y. Zhang, L. Wang, L. Liu, Y. Wang, W. Chen, H. Geng and X. Huang, J. Mater. Chem. A, 2021, 9, 26323–26329.
17. J. Ren, Y. Wang, Y. Song, L. Chen and Z. Yuan, Appl. Catal., B, 2022, 306, 121279.
18. Y. Tan, Z. Zhang, W. Wu, S. Chen, W. Chen and N. Cheng, Appl. Catal., B, 2023, 331, 122684.
19. H. Shi, T. Dai, X. Sun, Z. Zhou, S. Zeng, T. Wang, G. Han, Z. Wen, Q. Fang, X. Lang and Q. Jiang, Adv. Mater., 2024, 36, 2406711.
20. J. Ding, Y. Ji, Y. Li and G. Hong, Nano Lett., 2021, 21, 9381–9387.
21. H. Yuan, D. Jiang, Z. Li, X. Liu, Z. Tang, X. Zhang, L. Zhao, M. Huang, H. Liu, K. Song and W. Zhou, Adv. Mater., 2024, 36, 2305375.
22. F. Gao, Y. Wang, Y. Yang, J. Liao, J. Mu, X. Zhang, Z. Niu, P. Yang and M. Gao, Nano Res., 2023, 16, 10787–10795.
23. G. Meng, H. Cao, T. Wei, Q. Liu, J. Fu, S. Zhang, J. Luo and X. Liu, Chem. Commun., 2022, 58, 11839.
24. J. Liu, B. Zhang, Y. Fo, W. Yu, J. Gao, X. Cui, X. Zhou and L. Jiang, Chem. Eng. J., 2023, 464, 142692.
25. X. Zhang, Z. Li, X. Sun, L. Wei, H. Niu, S. Chen, Q. Chen, C. Wang and F. Zheng, ACS Mater. Lett., 2022, 4, 2097–2105.
26. S. Lu and Z. Zhuang, J. Am. Chem. Soc., 2017, 139, 5156–5163.
27. L. Wu, L. Su, Q. Liang, W. Zhang, Y. Men and W. Luo, ACS Catal., 2023, 13, 4127–4133.
28. J. Wu, X. Gao, G. Liu, X. Qiu, Q. Xia, X. Wang, W. Zhu, T. He, Y. Zhou, K. Feng, J. Wang, H. Huang, Y. Liu, M. Shao, Z. Kang and X. Zhang, J. Am. Chem. Soc., 2024, 146, 20323–20332.
29. J. Fang, H. Wang, Q. Dang, H. Wang, X. Wang, J. Pei, Z. Xu, C. Chen, W. Zhu, H. Li, Y. Yan and Z. Zhuang, Nat. Commun., 2024, 15, 4236.
30. Y. Men, X. Su, P. Li, Y. Tan, C. Ge, S. Jia, L. Li, J. Wang, G. Cheng, L. Zhuang, S. Chen and W. Luo, J. Am. Chem. Soc., 2022, 144, 12661–12672.
31. W. Sheng, H. Gasteiger and S. Yang, J. Electrochem. Soc., 2010, 157, B1529–B1536.
32. K. Wei, X. Wang and J. Ge, Chem. Soc. Rev., 2024, 53, 8903–8948.
33. Y. Yang, F. Gao, X. Zhang, S. Qin, L. Zheng, Y. Wang, J. Liao, Q. Yang and M. Gao, Angew. Chem., Int. Ed., 2022, 61, e202208040.
34. Q. Li, H. Peng, Y. Wang, L. Xiao, J. Lu and L. Zhuang, Angew. Chem., Int. Ed., 2019, 58, 1442–1446.
35. Y. Fang, C. Wei, Z. Bian, X. Yin, B. Liu, Z. Liu, P. Chi, J. Xiao, W. Song, S. Niu, C. Tang, J. Liu, X. Ge, T. Xu and G. Wang, Nat. Commun., 2024, 15, 1614.
36. Q. Wang, H. Wang, H. Cao, C. Tung, W. Liu, S. Hung, W. Wang, C. Zhu, Z. Zhang, W. Cai, Y. Cheng, H. Tao, H. Chen, Y. Wang, Y. Li, H. Yang, Y. Huang, J. Li and B. Liu, Nat. Catal., 2023, 6, 916–926.
37. H. Huang, K. Liu, F. Yang, J. Cai, S. Wang, W. Chen, Q. Wang, L. Fu, Z. Xie and S. Xie, Angew. Chem., Int. Ed., 2023, 62, e202315752.
38. S. Alia and B. Pivovar, J. Electrochem. Soc., 2018, 165, F441–F455.
39. C. Wei, R. R. Rao, J. Peng, B. Huang, I. E. L. Stephens, M. Risch, Z. J. Xu and Y. Shao-Horn, Adv. Mater., 2019, 31, 1806296.
40. Y. Duan, Z. Yu, L. Yang, L. Zheng, C. Zhang, X. Yang, F. Gao, X. Zhang, X. Yu, R. Liu, H. Ding, C. Gu, X. Zheng, L. Shi, J. Jiang, J. Zhu, M. Gao and S. Yu, Nat. Commun., 2020, 11, 4789.
41. M. Kundu, R. Mishra, T. Bhowmik and S. Barman, J. Mater. Chem. A, 2018, 6, 23531–23541.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta00678c

---