English

• Electrocatalytic water splitting has garnered significant interest as a sustainable technology to produce "green" hydrogen without carbon emissions [1,2]. This process comprises two half-reactions on the electrodes which are the hydrogen evolution reaction (HER) and the oxygen evolution reaction (OER), respectively [3,4]. In particular, OER is more energy-intensive and considered to be the primary bottleneck for efficient processes due to its slow kinetics [5]. Consequently, developing electrocatalysts with low overpotential for OER is imperative. Transition metal chalcogenides (TMCs) have emerged as promising OER catalysts owing to their high activity, cost-effectiveness, controllable composition, and morphology, offering potential solutions to challenges in the OER process [6]. Compared to sulfur, selenium (Se) exhibits a larger atomic radius, lower ionization energy, and abundant d-electrons that facilitate charge transfer [7]. Additionally, the presence of Se atoms in metal selenides enhances the electrophilicity of metal atoms and augments the affinity of selenides for hydroxyl species [8].

  However, recent studies indicate that under high anodic potentials, transition metal selenides (TMSe) undergo surface oxidation and reconstruction, forming amorphous transition-metal oxyhydroxides (TMOOH), which are the actual active species instead of TMSe [9]. This reconstruction complicates the study of the electrochemical catalytic mechanism and leads to the dissolution of active metal species, resulting in poor catalyst stability [10]. Preventing TMSe surface reconstruction is thus critical for understanding structure-activity relationships and improving catalytic durability. Nonetheless, inhibiting this reconstruction during OER remains a significant challenge. Additionally, most reported TMSe catalysts exhibit high activity only at lower current densities (<100 mA/cm²), falling short of industrial-scale requirements due to high overpotentials associated with Gibbs free energies and rate-determining steps in *OOH intermediate formation [11]. A promising approach for enhancing catalyst durability is to develop the 2D cover-confined nanomaterials, also referred to as chainmail catalysts, which effectively shield enclosed nanometals from damage during rigorous reaction processes [12-14]. This innovative strategy not only improves catalytic activity but also ensures the stability of inner nanocatalysts, earning it the moniker ‘chainmail’ for catalyst [15].

  Among TMSe, cobalt-based selenides such as Co_0.85Se, CoSe, CoSe₂, and Co₃Se₄ have shown potential. Non-stoichiometric Co_0.85Se, in particular, is expected to exhibit superior electrochemical performance due to the interaction between Co 3d and Se 4p spin electrons, enhancing conductivity [16,17]. MXenes, a category of 2D transition metal carbides, have attracted significant interest for their remarkable electrical conductivity, mechanical rigidity, large surface area, and outstanding electrochemical properties [18]. Synthesized by chemically etching the MAX phase (M_n+1AX_n, in which M represents an early transition metal, A stands for a group Ⅲ/ ⅣA element, and X denotes either carbon or nitrogen), MXenes feature van der Waals gaps with layer spacing of 0.6–1.2 nm, providing an ideal confined platform for nanocatalysts [19].

  In this work, Co_0.85Se@Ti₃C₂ was synthesized by first etching Ti₃AlC₂ to obtain Ti₃C₂−MXene, followed by solvothermal growth of Co_0.85Se nanoparticles within its interlayer space. The interlayer confinement stabilizes Co_0.85Se by preventing surface reconstruction and generates an internal electric field that accelerates charge transfer rate at the electrode interface. The material exhibits superhydrophilicity (water contact angle of 9.1°) and underwater superaerophobicity (bubble contact angle of 151.1°). The catalyst demonstrates low overpotentials of 230 mV at 100 mA/cm², 376 mV at 1000 mA/cm², and 417 mV at 1500 mA/cm², with a Tafel slope of 59 mV/dec in 1 mol/L KOH, maintaining excellent stability over 200 h at 200 mA/cm². Density functional theory (DFT) is introduced, which revealed that the Co_0.85Se@Ti₃C₂ interaction reduces the energy barrier for *OOH formation from *O, enabling exceptional performance in high-current OER processes. This work provides a foundation for designing non-reconstructed OER catalysts with high activity at elevated current densities through confined structures with internal electric fields.

  Fig. 1a presents a schematic diagram illustrating the preparation process of the Co_0.85Se@Ti₃C₂ catalyst. Fig. 1b displays the XRD patterns of both raw materials and synthesized products. The diffraction peaks for the raw Ti₃AlC₂ align with its standard card JCPDS No. 52–0875. Followed by the etching of Ti₃AlC₂ using a concentrated hydrochloric acid and lithium fluoride mixture, the (002) peak shifts to lower angles of 7.2°, indicating an increase in interlayer distance due to the dissolution of aluminum layers within the crystal structure, ultimately leading to the formation of two-dimensional layered Ti₃C₂ material. Furthermore, the characteristic peaks for synthesized Co_0.85Se and Co_0.85Se@Ti₃C₂ in Fig. 1b closely match well with the standard card for Co_0.85Se (JCPDS No. 52–1008), with no additional impurity peaks [20,21]. Notably, the Ti₃C₂ characteristic peak in Co_0.85Se@Ti₃C₂ is shifted further left to 5.8° compared to that of pure Ti₃C₂ (the partial XRD within the red dashed box was magnified and is presented in Fig. 1b, resulting in an increase in interlayer distance from 1.21 nm to 1.52 nm, which manifests the successful incorporation of Co_0.85Se nanoparticles into the interlayer space of Ti₃C₂, thereby causing further expansion.

  Figure 1

  

  Figure 1.  (a) Schematic illustration of the process for the catalyst Co_0.85Se@Ti₃C₂ preparation. (b) XRD pattern of the catalysts of Ti₃AlC₂, Ti₃C₂, Co_0.85Se and Co_0.85Se@Ti₃C₂ and the corresponding local enlarged view. (c) SEM image of Ti₃C₂. (d) SEM image of Co_0.85Se@Ti₃C₂. (e) Distribution of C, Ti, Co, and Se in Co_0.85Se@Ti₃C₂ catalyst. (f) TEM image and SAED pattern of the Co_0.85Se@Ti₃C₂. (g) HRTEM image of the Co_0.85Se@Ti₃C₂.

  DownLoad: Full-Size Img PowerPoint

  Fig. 1c depicts the SEM image of Ti₃C₂, revealing a two-dimensional accordion-like multilayer structure with a smooth surface. In Fig. 1d, the SEM image of the Co_0.85Se@Ti₃C₂ composite material illustrates a rough surface due to the distribution of the Co_0.85Se nanoparticles on the surface and layer space of Ti₃C₂. Fig. S1 (Supporting information) reveals that pure Co_0.85Se demonstrates obvious agglomeration. The EDS mapping (Fig. 1e) of the Co_0.85Se@Ti₃C₂ catalyst shows uniform element distribution, indicating the uniformly dispersed Co_0.85Se nanoparticles on the surface of layered Ti₃C₂. Fig. 1f presents TEM images and electron diffraction patterns for the ultrasonically agitated Co_0.85Se@Ti₃C₂ sample. The TEM images reveal uniformly sized Co_0.85Se particles with ultrafine dimensions, while electron diffraction rings correspond to (101), (102), and (110) crystal planes of Co_0.85Se and (110) crystal plane of Ti₃C₂. The HRTEM image (Fig. 1g) demonstrates that the Co_0.85Se@Ti₃C₂ material possesses a heterostructure with varying interlayer spacings in two distinct regions: one layer spacing at 0.27 nm belonging to (101) plane of Co_0.85Se and the another layer spacing at 1.52 nm belonging to (002) plane of Ti₃C₂.

  The full XPS spectrum of the composite (Fig. S2 in Supporting information) confirms the presence of Co, Se, Ti, C, and O. The high-resolution XPS spectrum of C 1s (Fig. 2a) for Co_0.85Se@Ti₃C₂ shows peaks at 281.6, 283.8, 284.8, and 287.8 eV, corresponding to C-Ti, C-Ti-O, C—C, and C=O, respectively [22]. The O 1s spectrum (Fig. 2b) features a prominent peak at 530.5 eV and a weaker peak at 532.1 eV, attributed to Ti-O and Ti-OH groups, indicating oxygen primarily as a terminating group in Ti₃C₂ [23]. The Co 2p spectrum (Fig. 2c) displays peaks at 778.3 and 780.9 eV for the Co 2p_3/2, and at 793.3 and 796.9 eV for Co 2p_1/2, with satellite peaks at 784.6 and 802.6 eV [24]. Fig. 2d presents the detailed spectrum of Se 3d in Co_0.85Se@Ti₃C₂ material, with binding energies of 54.2 and 55.1 eV corresponding to the Se 3d_5/2 and Se 3d_3/2 peaks, respectively. The peaks observed at 58.86 and 60.2 eV are attributed to SeO_x, suggesting that the surface of the material has undergone oxidation in ambient air [25].

  Figure 2

  

  Figure 2.  (a, b) The XPS spectra of C 1s, O 1s. (c, d) The XPS spectra of Co 2p, Se 3d for the catalyst Co_0.85Se and Co_0.85Se@Ti₃C₂. (e, f) The UPS spectra of Ti₃C₂, Co_0.85Se. (g) Schematic diagram of the formation of built-in electric field between Ti₃C₂ and Co_0.85Se. E_Vac: vacuum level, E_F: fermi level, E_V: valence band edge, E_C: conduction band edge.

  DownLoad: Full-Size Img PowerPoint

  Compared to pure Co_0.85Se, the Co and Se peaks in Co_0.85Se@Ti₃C₂ shift to lower binding energies due to enhanced electron screening from charge density redistribution at the interface [26-28], suggesting charge transfer from Ti₃C₂ to Co_0.85Se. Furthermore, the comparative investigation of the peak shifts of Ti 2p between Ti₃C₂ and Co_0.85Se@Ti₃C₂ further substantiated the aforementioned discourse (Fig. S3 in Supporting information). The semiconductor type was determined using Mott-Schottky (MS) curves, revealing p-type characteristics for Co_0.85Se (Fig. S4). Ultraviolet photoelectron spectroscopy (UPS) measured the work functions (Φ), with Φ(Ti₃C₂) calculated as 2.79 eV and Φ(Co_0.85Se) as 3.62 eV (Figs. 2e and f). The energy level diagram (E_V vs. vacuum level) and the conduction band diagram (E_C vs. vacuum level) are presented in the schematic diagram of the energy band structure in Fig. 2g, visualizing the effect of the built-in electric field formed between Co_0.85Se and Ti₃C₂. It can be observed that once Co_0.85Se comes into contact with Ti₃C₂, due to the corresponding disparity in Fermi levels, a Schottky junction with built-in electric field will be established, and electrons will spontaneously flow from Ti₃C₂ to p-type semiconductor Co_0.85Se unobstructedly. This unidirectional electron transfer is conducive to enriching electrons in the Co_0.85Se region and depleting electrons in the Ti₃C₂ region, thereby further accelerating the electrocatalytic process. Consequently, the BEF structure of Co_0.85Se@Ti₃C₂ is beneficial for providing optimized active sites, thereby facilitating the adsorption of oxygen intermediates [29-32].

  The synthesized samples were subsequently investigated for their electrocatalytic performance in the OER under alkaline conditions with iR-compensation. As shown in Fig. 3a, the LSV curves of Co_0.85Se, Ti₃C₂, Co_0.85Se@Ti₃C₂, and RuO₂ were obtained for electrochemical OER analysis. It is evident that Co_0.85Se@Ti₃C₂ demonstrated superior electrocatalytic activity of the lowest onset potential. Fig. 3b was the comparison of overpotentials, in which Co_0.85Se@Ti₃C₂ exhibited the lowest overpotential for OER at high current densities of 100, 1000 and 1500 mA/cm², requiring only 230, 376 and 417 mV overpotentials, respectively, while the pure Co_0.85Se displayed significantly higher overpotentials of 415, 603 and 729 mV at the respective current densities. Furthermore, Ti₃C₂ and RuO₂ attained overpotentials of 600 and 490 mV respectively at 100 mA/cm², but failed to reach the high current densities mentioned above. Table S1 (Supporting information) highlights the superior OER performance of Co_0.85Se@Ti₃C₂ compared to other catalysts.

  Figure 3

  

  Figure 3.  (a) LSV measurements of different samples for OER in 1.0 mol/L KOH electrolyte. (b) The comparison of the overpotential to Ti₃C₂, RuO₂, Co_0.85Se and Co_0.85Se@Ti₃C₂. (c) Tafel slope derived from (a). (d) EIS Nyquist plots of different samples. Inset: the equivalent circuit for the EIS data fitting. (e) Capacitive currents at 1.17 V (vs. RHE) as a function of scan rate for Ti₃C₂, Co_0.85Se, and Co_0.85Se@Ti₃C₂. (f) The TOF to Ti₃C₂, Co_0.85Se, Co_0.85Se@Ti₃C₂. (g) Multi-current testing of Co_0.85Se@Ti₃C₂ catalysts in the range of 200–1000 mA/cm². (h) Chronopotential curves for Co_0.85Se@Ti₃C₂ at 200 mA/cm².

  DownLoad: Full-Size Img PowerPoint

  In Fig. 3c, Co_0.85Se@Ti₃C₂ exhibits the smallest Tafel slope (59 mV/dec), which is significantly surpasses the Co_0.85Se (94 mV/dec), Ti₃C₂ (111 mV/dec), and RuO₂ (102 mV/dec), suggesting its more rapid reaction kinetics. Moreover, the Tafel plot indicates that Co_0.85Se@Ti₃C₂ possesses the highest exchange current density (j₀) of 6.67 mA/cm², as detailed in Table S2 (Supporting information). Additionally, the obtained Tafel slope is in alignment with the slope derived from electrochemical impedance spectroscopy (EIS) (Fig. 3d). The charge transfer resistance (R_ct) was determined by fitting the Nyquist plots with resistance-capacitance circuits (inset in Fig. 3d, solution resistance (R_s), constant phase element (CPE)). Co_0.85Se@Ti₃C₂ exhibits an R_ct of 4.52 Ω, which is significantly lower than that of Ti₃C₂ (R_ct = 20.67 Ω), and Co_0.85Se (R_ct = 6.12 Ω), indicating a superior electron transfer efficiency and reduced charge transfer resistance than other electrocatalysts.

  To acquire the actual alkaline OER activities of the fabricated electrodes, the electrochemical surface active area (ECSA) was ascertained through double-layer capacitance (C_dl) [33]. The C_dl were obtained from cyclic voltammetry (CV) tests conducted at a non-faradic potential region (Fig. S5 in Supporting information), which represent the ECSA of the catalysts. The current density exhibits a linear correlation with the scan rates, and the value of C_dl is calculated based on the slope of this linear relationship.

  As depicted in Fig. 3e, the Co_0.85Se@Ti₃C₂ composite exhibits a significantly higher C_dl of 28.98 mF/cm² compared to the individual components, Co_0.85Se (9.73 mF/cm²) and Ti₃C₂ (4.25 mF/cm²). This elevated C_dl corresponds to the highest electrochemically active surface area (ECSA) of 724.5 cm² for Co_0.85Se@Ti₃C₂, outperforming both Co_0.85Se (243.3 cm²) and Ti₃C₂ (106.3 cm²), as well as other materials reported in prior studies [34-36]. These findings suggest that a well-designed composite structure can enhance the adsorption of surface species, resulting in a dense Helmholtz double layer between the interface of electrolyte and catalyst with high surface capacitance, which subsequently accelerates the OER process. Furthermore, the BEF at the Co_0.85Se@Ti₃C₂ interface, arising from electron transfer from MXene to Co_0.85Se, promotes electron transfer rate and enhances the charge transfer capacity during OER. This is corroborated by the electrochemical impedance spectroscopy (EIS) results shown in Fig. 3d. In Fig. 3f, Co_0.85Se@Ti₃C₂ attains significantly higher turnover frequency (TOF) values at diverse overpotentials compared with those of Ti₃C₂ and Co_0.85Se, which further substantiates the remarkable electrocatalytic performance of Co_0.85Se@Ti₃C₂.

  The Faradaic efficiency (FE) of Co_0.85Se@Ti₃C₂ for OER was determined by measuring the oxygen volume obtained at a constant current of 0.05 A using a water drainage method (Fig. S6 in Supporting information) [37]. The measured O₂ vol-time curve closely aligns with the theoretical value (Fig. S7 in Supporting information), indicating a high FE (>98%) for the OER process on Co_0.85Se@Ti₃C₂. Long-term stability is a critical factor in evaluating the OER properties of a catalyst during practical application. As illustrated in Fig. 3g, a chronoamperometry curve with multiple current steps was performed, and the immediate response demonstrated outstanding mass transport capability. Furthermore, there is minimal change in potential even at a relatively high current density of 1000 mA/cm², suggesting exceptional durability for OER. Additionally, the chronopotentiometric curves show that the material can sustain 200 mA/cm² for at least 200 h with only decay by 4% (Fig. 3h), indicating its potential for industrial applications. Despite significant gas evolution during extended electrocatalysis on Co_0.85Se@Ti₃C₂, oxygen bubbles quickly dissipate into the electrolyte solution without accumulating on the material surfaces due to its excellent gas release performance, which will be further discussed later.

  High oxidation potentials often result in significant deterioration to the physical phase and structure of the catalyst during OER, particularly metal selenides. In order to investigate the structural reconstruction of Co_0.85Se@Ti₃C₂, in-situ XRD and Raman were conducted. The schematic diagrams for in-situ XRD and Raman measurements are depicted in Figs. 4a and b. During the XRD detection, continuous CV scanning was performed from 1.13 V to 1.53 V (vs. RHE). As shown in Fig. 4c, the in-situ XRD showed no significant change in characteristic peaks for the lattice plane of (311) and (222) of Co_0.85Se (JCPDS No. 52–1008), suggesting that there was no structural reconstruction during OER. Additionally, no diffraction peaks corresponding to CoOOH and Co(OH)₂ were observed in the XRD patterns, further confirming that the restructuring was hindered [38-40]. The in-situ Raman spectra of the Co_0.85Se@Ti₃C₂ catalyst at various potentials were given in Fig. 4d. Surprisingly, even at a high voltage of 1.8 V, there is no discernible characteristic peak of CoOOH appeared (range: 400–750 cm^-1) in the spectrum. In contrast, the pure Co_0.85Se sample without Ti₃C₂ displayed distinct surface reconstruction peaks at a voltage of 1.23 V during in-situ Raman, which were identified by the generation of CoOOH and Co(OH)₂ in Fig. 4e [41-44]. Furthermore, the ex-situ XPS spectrum (Figs. S8a and b in Supporting information) of Co_0.85Se@Ti₃C₂ after the OER test still presents the intense characteristic peaks of cobalt selenide, encompassing distinct Co 2p_1/2, Co 2p_3/2, Se 3d_3/2, and Se 3d_5/2 peaks. Meanwhile, the SeO_x peak does not exhibit a significant increase. This is strikingly dissimilar from the situation where the Se 3d feature peaks of other metal selenides weaken or even vanish after the OER test and reconstruction [45-47]. Additionally, the O 1s (Fig. S8c in Supporting information) spectrum mainly corresponds to the oxygen-terminated groups on the surface of Ti₃C₂ [48], which is almost the same as that of the sample before the OER test.

  Figure 4

  

  Figure 4.  (a) Schematic diagram of the principle of in-situ X-ray diffraction. (b) Schematic diagram of the in-situ Raman device. (c) In-situ XRD plot of the catalyst Co_0.85Se@Ti₃C₂ CV cycling at 1.098–1.498 (V vs. RHE) potential. (d, e) In-situ Raman plot of catalyst Co_0.85Se@Ti₃C₂ and Co_0.85Se performed at OCP-1.73 V (V vs. RHE) potential. OCP: circuit potential.

  DownLoad: Full-Size Img PowerPoint

  The wettability of the catalyst significant influence the exposure of surface active sites to the electrolyte [49]. Moreover, the accumulation of bubbles on the catalyst surface would lead to a "bubble shielding effect", which impeded the contact between the electrolyte and catalyst [50]. Therefore, constructing catalysts with exceptional hydrophilicity and aerophobicity properties would enhance OER processes. To analyze the wettability properties of various catalysts, contact angle (CA) measurements are performed (Fig. 5a). The results indicate that Co_0.85Se@Ti₃C₂ exhibits the smallest water CA at 9.1°, demonstrating remarkable hydrophilicity compared to Co_0.85Se (26.7°) and Ti₃C₂ (32.6°). Furthermore, Co_0.85Se@Ti₃C₂ demonstrates superior aerophobicity with a larger bubble CA (151.1°) compared to Co_0.85Se (121.8°) and Ti₃C₂ (96.3°) (Fig. 5b). The above results demonstrate that the promoted efficient mass transfer and gas bubble release on the surface of Co_0.85Se@Ti₃C₂. The enhanced capability of Co_0.85Se@Ti₃C₂ to attract water is primarily attributed to its composite structure, which facilitates electron redistribution, amplifies the material’s polarity, and consequently enhances its affinity for polar water molecules. Additionally, in accordance with the solid-liquid-gas interface theory, the uniformly distributed of Co_0.85Se particles on the surface of Ti₃C₂ will also enhance the roughness of the surface, which will give rise to the formation of a three-phase interface with discontinuous contact, leading to an extremely small contact area between the bubble and the electrode surface and thereby facilitating the desorption of gaseous products [51]. Consequently, it can be inferred that the composite structure promotes the adsorption of H₂O molecules nevertheless, and improves the desorption capacity of O₂ molecules on the surface of catalysts, resulting in heightened reactivity at active sites and improved performance in OER.

  Figure 5

  

  Figure 5.  The water droplet (a) and air bubble (b) contact angles measurement of Ti₃C₂, Co_0.85Se, Co_0.85Se@Ti₃C₂. The densities of states of Ti₃C₂ (c), Co_0.85Se (d), and Co_0.85Se@Ti₃C₂ (e). (f) Adsorption energy of H₂O molecules. (g) The four step processes of OER. (h) Step diagram of Co_0.85Se and Co_0.85Se@Ti₃C₂ OER reaction process.

  DownLoad: Full-Size Img PowerPoint

  The density functional theory (DFT) approach was employed to further clarify the mechanism accountable for the improved OER performance. The calculation models of Ti₃C₂, Co_0.85Se, and Co_0.85Se@Ti₃C₂ heterojunction are shown in Fig. S9 (Supporting information). The density of states (DOS) was computed (Figs. 5c-e), suggesting that the Co_0.85Se@Ti₃C₂ heterojunction offers a higher density of electronic cloud close to the Fermi level, which is beneficial for lowering the reaction energy barrier of water splitting [52] and thereby enhancing the catalytic activity. Further d-band center computations reveal that the d-band centers of Ti₃C₂, Co_0.85Se, and Co_0.85Se@Ti₃C₂ are −0.89, −1.08, and −0.40 eV respectively. This indicates that, in line with the d-band theory, Co_0.85Se@Ti₃C₂ has higher valence states and more antibonding states [53]. Therefore, Co_0.85Se@Ti₃C₂ is beneficial for adsorption, thus enhancing the electrocatalytic activity. Fig. 5f shows that the adsorption energy of H₂O molecules on Co_0.85Se@Ti₃C₂ is −0.72 eV, which is lower than that of Co_0.85Se (−0.25 eV) and Ti₃C₂ (−0.057 eV). Consequently, it can be inferred that the heterostructure promotes the adsorption of H₂O molecules on the surface of the catalysts, thereby increasing the reactivity at active sites and improving the OER performance. Fig. 5g discloses that the OER process involves four consecutive proton-coupled electron transfer steps, including the intermediates of *OH, *O, and *OOH [54]. As presented in Fig. 5h, the transfer of electrons from *O to *OOH (Step Ⅲ) has the most significant energy barrier, suggesting that it is the rate-determining step governing the rate of the OER process for Co_0.85Se@Ti₃C₂, which has a lower energy barrier of 2.50 eV compared to Co_0.85Se (3.49 eV). Therefore, the DFT analysis verifies that the Co_0.85Se@Ti₃C₂ heterostructure exhibits superior OER activity, which is in accordance with the experimental results.

  In summary, a novel electrocatalyst (Co_0.85Se@Ti₃C₂) featuring a confined interlayer structure was proposed for alkaline OER, which demonstrates remarkable catalytic activity (376 mV@1000 mA/cm², and 417 mV@1500 mA/cm²) and exceptional long-term durability up to 200 h at a high current density of 200 mA/cm². Both experimental and theoretical studies validated that the confined structure, the built-in electric field, and the superhydrophilic/oleophobic properties synergistically optimize the spatial environment and electronic configuration of Co_0.85Se, thereby preventing the reconstruction of Co_0.85Se and reducing the energy barrier for the OER process and enhancing the intrinsic activity. This investigation establishes the foundation for designing efficient and non-reconstructive transition metal chalcogenide electrocatalysts and promotes progress in the OER field.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Yuanhua Xiao: Writing – review & editing. Jinhui Shou: Writing – original draft. Shiwei Zhang: Software. Ya Shen: Software. Junwei Liu: Resources. Dangcheng Su: Software. Yang Kong: Resources. Xiaodong Jia: Resources. Qingxiang Yang: Resources. Shaoming Fang: Resources. Xuezhao Wang: Resources.

  Acknowledgments

  The authors are grateful to the financial support from the National Natural Science Foundation of China (NSFC, Nos. U23A20579, U1904190, 52272243), the Natural Science Foundation of Henan Province (No. 242300421467), the Program for Science and Technology Innovation Talents in Universities of Henan Province (No. 22HASTIT005), Key Scientific Research Projects in Higher Education Institutions of Henan Province (Nos. 24A430047, 24A430029), China Postdoctoral Science Foundation (No. 2023M741083), Research Foundation for Talented Scholars (No. 21010744).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111441.

  
  
• 1. [1]

     N. Zhao, L. Mu, W. Liao, et al., J. Colloid Interface Sci. 682 (2025) 148–156. doi: 10.1016/j.jcis.2024.11.193

  2. [2]

     Y. Xiao, S. Zhang, Y. Shen, et al., J. Colloid Interface Sci. 664 (2024) 329–337. doi: 10.1016/j.jcis.2024.03.041

  3. [3]

     L. Mu, S. Qiu, G. Zhao, et al., J. Mater. Chem. A 12 (2024) 1714–1724. doi: 10.1039/d3ta06701g

  4. [4]

     Y. Xiao, Y. Shen, D. Su, et al., J. Mater. Sci. Technol. 154 (2023) 1–8. doi: 10.1016/j.jmst.2022.12.042

  5. [5]

     G. Zhao, L. Mu, L. Chen, et al., Int. J. Hydrog. Energy 88 (2024) 432–440. doi: 10.1016/j.ijhydene.2024.09.114

  6. [6]

     M. Li, H. Li, H. Fan, et al., Nat. Commun. 15 (2024) 6154. doi: 10.1038/s41467-024-50535-2

  7. [7]

     A. Du, Y. Zhao, Z. Zhang, et al., Energy Storag. Mater. 26 (2020) 23–31. doi: 10.1016/j.ensm.2019.12.030

  8. [8]

     Y. Cheng, L. Zhu, Y. Gong, Int. J. Hydrog. Energy 92 (2024) 1049–1058. doi: 10.1016/j.ijhydene.2024.09.267

  9. [9]

     J. Zhao, F. Wang, X. Lu, et al., J. Colloid Interface Sci. 642 (2023) 532–539. doi: 10.3390/agronomy13020532

  10. [10]

      Q. Wang, Y. Cheng, H.B. Tao, et al., Angew. Chem. Int. Ed. 62 (2023) e202216645. doi: 10.1002/anie.202216645

  11. [11]

      X. Wang, S. Xi, P. Huang, et al., Nature 611 (2022) 702–708. doi: 10.1038/s41586-022-05296-7

  12. [12]

      Y. Xiao, H. Jiang, K. Zhang, et al., Chem. Eng. J. 492 (2024) 152274. doi: 10.1016/j.cej.2024.152274

  13. [13]

      Y. Xiao, Y. Kong, X. Wang, et al., J. Colloid Interface Sci. 677 (2025) 577–586. doi: 10.1016/j.jcis.2024.07.245

  14. [14]

      Y. Xiao, Q. Le, Y. Kong, et al., Chem. Eng. J. 505 (2025) 159268. doi: 10.1016/j.cej.2025.159268

  15. [15]

      L. Tang, X. Meng, D. Deng, et al., Adv. Mater. 31 (2019) 1901996. doi: 10.1002/adma.201901996

  16. [16]

      L. Li, S. Wang, J. Peng, et al., Molecules 29 (2024) 3083. doi: 10.3390/molecules29133083

  17. [17]

      Z. Jiao, Y. Chen, M. Demir, et al., J. Energy Storag. 52 (2022) 104929. doi: 10.1016/j.est.2022.104929

  18. [18]

      A. Hanan, M. Nazim, R. Walvekar, et al., Chem. Eng. J. 483 (2024) 149107. doi: 10.1016/j.cej.2024.149107

  19. [19]

      F. Bu, M.M. Zagho, Y. Ibrahim, et al., Nano. Today 30 (2020) 100803. doi: 10.1016/j.nantod.2019.100803

  20. [20]

      X. Sun, N. Habibul, H. Du, Chin. J. Catal. 42 (2021) 235–243. doi: 10.1016/S1872-2067(20)63632-4

  21. [21]

      C. Atangana, H. Huang, H. Hong, et al., Energ. Storag. Mater. 24 (2020) 167–176. doi: 10.1016/j.ensm.2019.08.022

  22. [22]

      R. Zhao, H. Di, X. Hui, et al., Energ. Environ. Sci. 14 (2021) 5096 -5096. doi: 10.1039/d1ee90043a

  23. [23]

      C. Sun, X. Shi, Y. Zhang, et al., ACS Nano 14 (2020) 1176–1184. doi: 10.1021/acsnano.9b09541

  24. [24]

      J. Liang, S. Li, F. Li, et al., J. Colloid Interface Sci. 655 (2024) 296–306. doi: 10.1016/j.jcis.2023.11.010

  25. [25]

      K. Li, R. Cheng, Q. Xue, et al., Chem. Eng. J. 450 (2022) 137991. doi: 10.1016/j.cej.2022.137991

  26. [26]

      L. Jin, H. Xu, K. Wang, et al., Chem. Sci. 16 (2025) 329–337. doi: 10.1039/d4sc06673a

  27. [27]

      L. Mu, G. Zhao, B. Zhang, et al., J. Colloid Interface Sci. 677 (2025) 68–78.

  28. [28]

      H. Xu, L. Jin, K. Wang, et al., Fuel 369 (2024) 131716. doi: 10.1016/j.fuel.2024.131716

  29. [29]

      K. Wang, H. Xu, H. Xing, et al., Chem. Eng. J. 500 (2024) 157316. doi: 10.1016/j.cej.2024.157316

  30. [30]

      B. Zhang, N. Zhang, G. Zhao, et al., J. Colloid Interface Sci. 665 (2024) 1054–1064. doi: 10.1016/j.jcis.2024.04.002

  31. [31]

      H. Xu, L. Yang, Y. Liu, et al., Fuel 377 (2024) 132796. doi: 10.1016/j.fuel.2024.132796

  32. [32]

      X. Luan, Z. Zheng, S. Zhao, et al., Adv. Funct. Mater. 32 (2022) 2202843. doi: 10.1002/adfm.202202843

  33. [33]

      X. Zhou, Y. Mo, F. Yu, et al., Adv. Funct. Mater. 33 (2023) 2209465. doi: 10.1002/adfm.202209465

  34. [34]

      B. Lu, Z. Li, J. Yin, et al., Appl. Catal. B: Environ. 350 (2024) 123940. doi: 10.1016/j.apcatb.2024.123940

  35. [35]

      Y. Yu, Y. Wu, W. Luo, et al., Chem. Eng. J. 477 (2023) 147076. doi: 10.1016/j.cej.2023.147076

  36. [36]

      L. He, N. Wang, M. Xiang, et al., Appl. Catal. B 345 (2024) 123686. doi: 10.1016/j.apcatb.2023.123686

  37. [37]

      Y. Song, M. Sun, S. Zhang, et al., Adv. Funct. Mater. 33 (2023) 2214081. doi: 10.1002/adfm.202214081

  38. [38]

      Q. Ji, Y. Kong, H. Tan, et al., ACS Catal. 12 (2022) 4318–4326. doi: 10.1021/acscatal.2c01090

  39. [39]

      S. Ye, J. Wang, J. Hu, et al., ACS Catal. 11 (2021) 6104–6112. doi: 10.1021/acscatal.1c01300

  40. [40]

      F. Reikowski, F. Maroun, I. Pacheco, et al., ACS Catal. 9 (2019) 3811–3821. doi: 10.1021/acscatal.8b04823

  41. [41]

      K. Wang, Y. Li, J. Hu, et al., Chem. Eng. J. 447 (2022) 137540. doi: 10.1016/j.cej.2022.137540

  42. [42]

      M. Chen, D. Liu, J. Feng, et al., Chem. Eng. J. 443 (2022) 136432. doi: 10.1016/j.cej.2022.136432

  43. [43]

      Y. Zhi, Z. Li, Y. Tang, et al., Nano Energ. 134 (2025) 110565. doi: 10.1016/j.nanoen.2024.110565

  44. [44]

      Z. Ji, W. Yuan, S. Zhao, et al., Chem. Catal. 3 (2023) 100501.

  45. [45]

      Y. Huang, L. Zhang, L.W. Jiang, et al., Small 19 (2023) 2302970. doi: 10.1002/smll.202302970

  46. [46]

      S. Feng, P. Rao, Y. Yu, et al., Electrochim. Acta 463 (2023) 142870. doi: 10.1016/j.electacta.2023.142870

  47. [47]

      M. Chen, Y. Zhang, R. Wang, et al., J. Energy Chem. 84 (2023) 173–180. doi: 10.1016/j.jechem.2023.05.009

  48. [48]

      T. Hou, Q. Luo, Q. Li, et al., Nat. Commun. 11 (2020) 4251. doi: 10.1038/s41467-020-18091-7

  49. [49]

      X. Xu, H. Liao, L. Huang, et al., Appl. Catal. B 341 (2024) 123312. doi: 10.1016/j.apcatb.2023.123312

  50. [50]

      J.W. Yu, K.D. Hu, Z.Y. Zhang, et al., Energy Environ. Sci. 16 (2023) 2068–2079. doi: 10.1039/d2ee04143j

  51. [51]

      X. Shan, J. Liu, H. Mu, et al., Angew. Chem. Int. Ed. 59 (2020) 1659–1665. doi: 10.1002/anie.201911617

  52. [52]

      H. Li, Q. Yu, X. Zhu, et al., Chem. Eng. J. 493 (2024) 152860. doi: 10.1016/j.cej.2024.152860

  53. [53]

      Z. Shi, C. Mao, L. Zhong, et al., Appl. Catal. B: Environ. 339 (2023) 123123. doi: 10.1016/j.apcatb.2023.123123

  54. [54]

      X. Shen, H. Li, Y. Zhang, et al., Appl. Catal. B: Environ. 319 (2022) 121917. doi: 10.1016/j.apcatb.2022.121917

• 

  Figure 1  (a) Schematic illustration of the process for the catalyst Co_0.85Se@Ti₃C₂ preparation. (b) XRD pattern of the catalysts of Ti₃AlC₂, Ti₃C₂, Co_0.85Se and Co_0.85Se@Ti₃C₂ and the corresponding local enlarged view. (c) SEM image of Ti₃C₂. (d) SEM image of Co_0.85Se@Ti₃C₂. (e) Distribution of C, Ti, Co, and Se in Co_0.85Se@Ti₃C₂ catalyst. (f) TEM image and SAED pattern of the Co_0.85Se@Ti₃C₂. (g) HRTEM image of the Co_0.85Se@Ti₃C₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a, b) The XPS spectra of C 1s, O 1s. (c, d) The XPS spectra of Co 2p, Se 3d for the catalyst Co_0.85Se and Co_0.85Se@Ti₃C₂. (e, f) The UPS spectra of Ti₃C₂, Co_0.85Se. (g) Schematic diagram of the formation of built-in electric field between Ti₃C₂ and Co_0.85Se. E_Vac: vacuum level, E_F: fermi level, E_V: valence band edge, E_C: conduction band edge.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) LSV measurements of different samples for OER in 1.0 mol/L KOH electrolyte. (b) The comparison of the overpotential to Ti₃C₂, RuO₂, Co_0.85Se and Co_0.85Se@Ti₃C₂. (c) Tafel slope derived from (a). (d) EIS Nyquist plots of different samples. Inset: the equivalent circuit for the EIS data fitting. (e) Capacitive currents at 1.17 V (vs. RHE) as a function of scan rate for Ti₃C₂, Co_0.85Se, and Co_0.85Se@Ti₃C₂. (f) The TOF to Ti₃C₂, Co_0.85Se, Co_0.85Se@Ti₃C₂. (g) Multi-current testing of Co_0.85Se@Ti₃C₂ catalysts in the range of 200–1000 mA/cm². (h) Chronopotential curves for Co_0.85Se@Ti₃C₂ at 200 mA/cm².

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Schematic diagram of the principle of in-situ X-ray diffraction. (b) Schematic diagram of the in-situ Raman device. (c) In-situ XRD plot of the catalyst Co_0.85Se@Ti₃C₂ CV cycling at 1.098–1.498 (V vs. RHE) potential. (d, e) In-situ Raman plot of catalyst Co_0.85Se@Ti₃C₂ and Co_0.85Se performed at OCP-1.73 V (V vs. RHE) potential. OCP: circuit potential.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  The water droplet (a) and air bubble (b) contact angles measurement of Ti₃C₂, Co_0.85Se, Co_0.85Se@Ti₃C₂. The densities of states of Ti₃C₂ (c), Co_0.85Se (d), and Co_0.85Se@Ti₃C₂ (e). (f) Adsorption energy of H₂O molecules. (g) The four step processes of OER. (h) Step diagram of Co_0.85Se and Co_0.85Se@Ti₃C₂ OER reaction process.

  下载: 全尺寸图片 幻灯片
•