Cation Vacancies Anchored Transition Metal Dopants Based on a Few-Layer Ti₃C₂T_x Catalyst for Enhanced Hydrogen Evolution

Abstract

This study addresses the efficiency and cost challenges of hydrogen evolution reaction (HER) catalysts in the context of carbon neutrality strategies by employing a synergistic approach that combines cation vacancy anchoring and transition metal doping on two-dimensional (2D) MXenes. Using an in situ LiF/HCl etching process, the aluminum layers in Ti₃AlC₂ were precisely removed, resulting in a few-layer Ti₃C₂T_x MXene with an increased interlayer spacing of 12.3 Å. Doping with the transition metals Fe, Co, Ni, and Cu demonstrated that Fe@Ti₃C₂ provided the optimal HER performance, characterized by an overpotential (η₁₀) of 81 mV at 10 mA cm⁻², a low Tafel slope of 33.03 mV dec⁻¹, and the lowest charge transfer resistance (R_ct = 5.6 Ω cm²). Mechanistic investigations revealed that Fe’s 3d⁶ electrons induce an upward shift in the d-band center of MXene, improving hydrogen adsorption free energy and reducing lattice distortion. This research lays a solid foundation for the design of non-precious metal catalysts using MXenes and highlights future avenues in bimetallic synergy and scalability.

1. Introduction

Hydrogen energy is a core energy carrier for achieving carbon neutrality, and its large-scale production heavily relies on efficient and low-cost water electrolysis technology [1]. However, the industrialization of hydrogen production through water electrolysis faces critical bottlenecks. The sluggish kinetics of the oxygen evolution reaction (OER) and the HER lead to high energy consumption in the system, while the high cost and scarcity of precious metal catalysts (such as Pt and IrO₂) further limit their widespread application [2,3]. Current commercial catalysts suffer from insufficient active site utilization, low charge transport efficiency, and inadequate stability, necessitating solutions through material innovation and mechanistic breakthroughs [4].

Atomically thin catalysts, due to their unique 2D confinement effects and tunable electronic structures, provide new avenues to address these challenges [5,6,7]. Among these, MXenes, which are new 2D transition metal carbides and nitrides characterized by the chemical formula M_n+1X_nT_x, exhibit excellent catalytic potential owing to their high conductivity (>10⁴ S m⁻¹), tunable surface functional groups (-O, -F, etc.), and abundant active sites. Their layered structure results in an ultra-high specific surface area (>1000 m² g⁻¹), while the surface dangling bonds and vacancy defects provide an ideal platform for the precise construction of active sites [8,9]. However, as early transition metal compounds, MXenes have a limited number of non-bonding d-orbitals, which results in insufficient adsorption/desorption kinetics for reaction intermediates, necessitating defect engineering to optimize their intrinsic activity [10].

To tackle this challenge, defect control and activation strategies have emerged as key methods to enhance the catalytic performance of MXenes [11,12,13,14,15]. Specifically, transition metal doping (Fe, Co, Ni, Mo) into MXene frameworks has proven highly effective: it not only modulates the electronic structure (reducing bandgap by 0.2–0.4 eV) and enhances conductivity (>6500 S m⁻¹), but also creates synergistic active sites by coupling dopant atoms with intrinsic vacancies [16]. These dopants optimize intermediate adsorption free energies (ΔG_*H ≈ 0.05–0.12 eV) and promote charge transfer at the atomic level, significantly improving reaction kinetics [17]. Theoretical analyses indicate that the introduction of cation vacancies can significantly modulate the material’s electronic band structure (ΔE ≈ 0.3–0.5 eV), optimize the adsorption free energy of intermediate products (ΔG_*H ≈ 0.08 eV), and expose high-density active sites [18,19,20,21,22,23]. Compared to traditional oxygen vacancies or lattice defects, cation vacancies display stronger charge localization effects and coordination unsaturation, allowing for more efficient anchoring of transition metal atoms and achieving synergistic activation of active sites through interface electronic coupling [24]. However, existing research lacks a systematic understanding of the atomic-level design of vacancy configurations and their activation mechanisms, limiting the rational development of high-performance catalysts.

This study proposes a “cation vacancy anchoring–transition metal doping” synergistic strategy to precisely construct Ti vacancies (V_Ti) on the surface of few-layer Ti₃C₂T_x using an HF etching method, and to anchor transition metal clusters (including Cu, Co, Ni, and Fe), successfully synthesizing a series of high-efficiency HER catalysts. The experimental results show that the Fe@Ti₃C₂T_x exhibits remarkable catalytic performance in alkaline media, with an overpotential (η₁₀) as low as 81 mV, outperforming other transition metal-doped catalysts. In-depth analysis reveals that transition metal doping upshifts the d-band center of MXene, optimizes hydrogen adsorption kinetics (ΔG_*H close to thermoneutral value), and forms strong metal–support interactions, thereby synergistically enhancing catalytic activity and durability. This research advances the design of atomically precise defect-catalyzing materials.

2. Results and Discussion

2.1. Structural Evolution and Etching Mechanism Analysis

Figure 1 clearly demonstrates the synthesis route of different catalysts. Selective etching was initially employed to remove the Al layer from the MAX-phase precursor (Ti₃AlC₂), thereby generating few-layer Ti₃C₂T_x MXene. Subsequently, Fe, Co, Ni, or Cu ions were incorporated into the MXene via impregnation. Under high-temperature annealing, these metal ions underwent in situ reduction, resulting in atomically dispersed single-metal atoms anchored within the MXene interlayer. To clarify the structural transformation of the MAX phase precursor (Ti₃AlC₂) into MXene (Ti₃C₂T_x), we systematically characterized the crystal structure and morphology evolution during the etching process using X-ray diffraction (XRD) and scanning electron microscopy (SEM). As shown in Figure 2a and Figure S1, the diffraction pattern of Ti₃AlC₂ matches well with the standard JCPDS card (No. 52-0875), confirming the structural identity and high phase purity of the precursor material. The characteristic peak of the (002) plane of the original Ti₃AlC₂ at 2θ = 9.0° shifts significantly to the left at 7.1° after etching with LiF/HCl. Based on Bragg’s Law (nλ = 2d sinθ), this shift implies that the interlayer spacing has expanded from 9.5 Å to 12.3 Å, confirming the selective removal of Al atomic layers and the formation of layered Ti₃C₂T_x. Notably, the full width at half maximum (FWHM) of the (002) peak increased from 0.21° to 0.38°, and the complete disappearance of the closely packed structure characteristic peaks ((104) at 38.5° and (105) at 44.5°) further indicates that the etching process disrupts interlayer order and fully delaminates the Al layers. Additionally, the absence of the TiO₂ (101) peak at 27° in the spectrum demonstrates that the mild HF in situ etching strategy effectively suppressed the oxidation side reactions of titanium, ensuring the inherent structural integrity of the MXene. The Raman spectrum (Figure S2) further supports this conclusion, displaying characteristic peaks associated with Ti–C and Ti–O vibrations, which confirm the successful synthesis of Ti₃C₂. Critically, the post-etching crystalline framework remains intact upon transition metal loading (Cu, Co, Ni, Fe), as evidenced by unchanged (002) peak positions in M@Ti₃C₂ composites (Figure 2b). This demonstrates that metal decoration occurs without altering the MXene’s basal plane structure.

SEM analysis further corroborated this structural evolution: the dense bulk morphology of the original Ti₃AlC₂ (Figure 3a) transformed into a typical layered stacked structure (Figure 3b), with a significantly increased interlayer spacing, which further confirms the successful and efficient conversion of the MAX phase into few-layer Ti₃C₂ MXene after the etching process. The averaged particle size of the Ti₃C₂ MXene was approximately 15 µm (Figure S3). Additionally, BET analysis indicates a specific surface area of 10.4 m² g⁻¹ (Figure S4), reflecting a relatively large surface area conducive to surface interactions. According to previous reports, the etching process generates abundant cation vacancies on the MXene surface, providing a favorable structural basis for subsequent transition metal doping [25,26,27,28,29]. Furthermore, as shown in Figure 3c–f, the layered morphology of MXene remains well-preserved after the introduction of transition metals, indicating their excellent structural stability.

2.2. Transition Metal Doping Characterizations

To verify the successful anchoring of transition metal cations onto Ti₃C₂T_x surfaces through a vacancy doping strategy, this study systematically characterized a representative Fe-doped system. As shown in Figure 4, energy dispersion spectrum (EDS) elemental mapping demonstrates atomic-level homogeneous distribution of Fe species (cyan signals) across the Ti₃C₂T_x substrate, with no detectable nanoclusters or secondary phases. Quantitative analysis reveals an ultralow Fe loading of 0.64 at% (Table S1), further confirming that Fe atoms exclusively occupy specific interlayer sites of MXene as isolated single atoms rather than aggregated nanoparticles.

X-ray photoelectron spectroscopy (XPS) was then employed to provide further evidence for vacancy occupation by Fe dopants. The distinct Fe 2p peaks observed in the XPS spectrum (Figure 5a) further confirm the successful Fe incorporation. The 0.68 eV negative shift observed in Ti 2p binding energy (Figure 5b) originates from electronic structure reconstruction induced by precise Fe occupation of Ti vacancies: when Fe cations fill Ti vacancies, neighboring Ti atoms—previously compelled to transfer electrons outward due to vacancy defects—restore their electron supply equilibrium, thereby increasing local electron cloud density around Ti atoms. This electron density redistribution reduces the nucleus-electron binding force in Ti atoms, manifesting as negative binding energy shifts. The experimentally observed electronic evolution aligns perfectly with theoretical predictions of vacancy doping-induced charge compensation mechanisms, providing direct evidence for precise anchoring of transition metal cations at Ti vacancy sites. Additionally, the XPS spectra of Cu, Ni, and Co@Ti₃C₂ are provided in the Supporting Information (Figures S5–S7) to further confirm the presence and chemical states of these elements.

2.3. HER Catalytic Performance and Kinetic Mechanisms

The effects of transition metal doping on HER catalytic performance were investigated via a standard three-electrode configuration in purified 1.0 M KOH. Linear sweep voltammetry (LSV) was first conducted, which reveals the significant regulatory effect of transition metal doping on the HER performance of MXene-based catalysts (Figure 6a). Fe@Ti₃C₂ demonstrated the best catalytic activity, with an overpotential (η₁₀) of only 81 mV, markedly superior to that of Ni@Ti₃C₂ (165 mV), Co@Ti₃C₂ (154 mV), Cu@Ti₃C₂ (194 mV), and undoped Ti₃C₂T_x (356 mV); and a current density (η = 200 mV) of 80.5 mA cm⁻², outperforming those of Ni@Ti₃C₂ (20.0 mA cm⁻²), Co@Ti₃C₂ (17.2 mA cm⁻²), Cu@Ti₃C₂ (11.2 mA cm⁻²), and the undoped Ti₃C₂T_x (4.2 mA cm⁻²). Although Fe@Ti₃C₂ (Ƞ₁₀ = 81 mV) exhibits a higher overpotential than commercial Pt/C (Figure 6b), it performs comparably to MXenes loaded with noble metals (Table S2), therefore offering compelling advantages, such as significantly reducing material costs while maintaining competitive activity. Tafel analysis (Figure 6c) further confirms the differences in kinetic mechanisms: the Tafel slope of Fe@Ti₃C₂ is as low as 33.03 mV dec⁻¹, close to the theoretical value of 30 mV dec⁻¹ for chemically desorption-limited processes, indicating that its HER rate is limited by the rapid desorption of H*; conversely, the high Tafel slope of Cu@Ti₃C₂ (108.01 mV dec⁻¹) reveals that its kinetics are hindered by the slow H* adsorption (Volmer step), closely related to the weak electronic coupling effect caused by the filled 3d¹⁰ orbitals of Cu. This kinetic disparity is attributed to the differing electronic configurations of the transition metals and their influence on the MXene substrate’s electronic structure. Specifically, Fe’s 3d⁶ electrons induce a significant upward shift in the MXene’s d-band center (∆E ≈ 0.45 eV) via strong d-p orbital hybridization [30], thereby optimizing the hydrogen adsorption free energy and lowering the activation barrier for the Volmer step [16,17,31]. Conversely, the weak electronic coupling effect caused by Cu’s filled 3d¹⁰ orbitals hinders the adsorption step. Complementary turnover frequency (TOF) measurements at η = 200 mV (Figure 6d) quantitatively corroborate this kinetic trend: Fe@Ti₃C₂ achieves a TOF of 44,000 s⁻¹, significantly outperforming Cu@Ti₃C₂ (5311 s⁻¹), Ni@Ti₃C₂ (4517 s⁻¹), and Co@Ti₃C₂ (5855 s⁻¹). This activity hierarchy directly validates the Tafel analysis, confirming Fe@Ti3C2′s superior kinetic efficiency in H* management. Furthermore, Fe@Ti₃C₂ demonstrates exceptional long-term stability, maintaining negligible activity decay after 24 h chronopotentiometry at 10 mA cm⁻² with no observable morphological changes (Figure 6e). Combined with the tunable electronic structure and abundant active sites inherent to transition metal-doped MXenes—which provide a versatile platform for performance optimization—these attributes (high activity, favorable kinetics, and robust stability) establish Fe@Ti₃C₂ as a sustainable and economically viable alternative to Pt-based catalysts for large-scale hydrogen production.

2.4. Intrinsic Catalytic Activity and Interface Charge Transfer

An analysis using double-layer capacitance (C_dl) and electrochemical impedance spectroscopy (EIS) reveals the synergistic regulation of active site density and charge transport dynamics resulting from transition metal doping. C_dl tests indicate that Fe@Ti₃C₂ has a substantially lower C_dl value (5.71 mF cm⁻²) compared to Cu@Ti₃C₂ (6.58 mF cm⁻²), Co@Ti₃C₂ (9.16 mF cm⁻²), and Ni@Ti₃C₂ (13.82 mF cm⁻²) (Figure 7a–e), suggesting that Fe@Ti₃C₂ possesses a smaller electrochemical active surface area (ECSA) and a lower density of exposed active sites. It is noteworthy that, based on exhibiting the highest apparent activity, the intrinsic activity of Fe@Ti₃C₂ (14.1 A F⁻¹) is further enhanced compared to Cu@Ti₃C₂ (1.7 A F⁻¹), Co@Ti₃C₂ (1.9 A F⁻¹), and Ni@Ti₃C₂ (1.4 A F⁻¹) after normalized by C_dl. This phenomenon indicates that the doping of Fe elements effectively restructured the surface electronic structure of Ti₃C₂, successfully constructing new catalytic active sites. The mutual verification of electrochemical characterization and theoretical calculations confirms that the Fe@Ti₃C₂ achieves precise regulation of active sites at the atomic scale, thereby demonstrating the most outstanding intrinsic catalytic activity.

EIS results (Figure 7f) further demonstrate that the charge transfer resistance (R_ct) of Fe@Ti₃C₂ is much lower (5.6 Ω cm²) than those of Ni@Ti₃C₂ (9.6 Ω cm²), Co@Ti₃C₂ (8.8 Ω cm²), and Cu@Ti₃C₂ (18.7 Ω cm²), indicating optimal efficiency for interfacial electron transport. This difference probably arises from the upward shift in the d-band center induced by Fe doping, enhancing electron delocalization.

2.5. Structure–Performance Relationship and Catalytic Mechanisms

Based on comprehensive experimental analysis, we propose a mechanistic rationale for the enhanced intrinsic HER activity in alkaline media, as depicted in Figure 8. Firstly, Electronic Structure Regulation occurs through hybridization between the dopant’s 3d orbital electrons and the MXene matrix. This interaction upshifts the position of the d-band center, a critical electronic descriptor, leading to more favorable adsorption energy for the key hydrogen intermediate (H*), thereby lowering the activation barrier. Secondly, Geometric Effects arise from the compatibility of the dopant’s atomic radius with the host lattice. This minimizes structural distortion, promoting the creation of uniform, highly active catalytic sites and enabling mild, beneficial modulation of the surface geometry. Thirdly, enhanced Interfacial Charge Transfer is driven by strong metal–support interactions (SMSIs) between the dopant and the MXene. This effectively reduces the charge transfer resistance (R_ct), accelerating the overall reaction kinetics. Among various dopants, Fe@Ti₃C₂ emerges as the optimal catalyst. This superiority stems from the synergistic combination of its specific 3d⁶ electronic configuration, which provides exceptional control over H* adsorption strength, and its inherent high structural compatibility with the Ti₃C₂ lattice, minimizing disruptive geometric changes. This interplay between electronic and geometric factors underpins their outstanding HER activity.

3. Materials and Methods

3.1. Materials and Chemicals

All chemicals were used as received without further purification. Titanium aluminum carbide (Ti₃AlC₂, 99.8% purity) was obtained from Xinen Technology Co., Ltd. (Guangzhou, China). Copper(II) nitrate trihydrate (Cu(NO₃)₂·3H₂O, AR grade), nickel(II) nitrate hexahydrate (Ni(NO₃)₃·9H₂O, AR grade), cobalt(II) nitrate hexahydrate (Co(NO₃)₂·6H₂O, AR grade), iron(II) sulfate heptahydrate (FeSO₄·7H₂O, AR grade), lithium fluoride (LiF, AR grade), and potassium hydroxide (KOH, AR grade) were purchased from Aladdin Industrial Corporation (Shanghai, China). Hydrochloric acid (HCl, 36~38 wt.%) was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).

3.2. Preparation of MXene

Few-layer Ti₃C₂T_x MXene was synthesized via an in situ LiF/HCl etching method. The Ti₃AlC₂ precursor was mixed with LiF/HCl in a weight ratio of 1:1.5 and stirred at 500 rpm in a 45° C oil bath for 48 h, during which HF was generated in situ to selectively etch the aluminum atomic layers. The etching product underwent gradient washing using 1 mol L⁻¹ HCl to remove residual LiCl/F⁻, followed by deionized water washing until pH ≈ 6. Subsequent ultrasonic delamination for 1 h and centrifugation separated the upper few-layer Ti₃C₂T_x suspension, which was freeze-dried for later use. XRD analysis confirmed that the interlayer spacing of the (002) crystal plane expanded from 9.5 Å to 12.3 Å, with no TiO₂ impurity peaks detected at 27°, validating the structural stability of the mild etching process.

3.3. Transition Metal Doping

Transition metals (Fe³⁺, Co²⁺, Ni²⁺, Cu²⁺) were anchored onto the surface cation vacancies of MXene by mixing the few-layer Ti₃C₂T_x suspension (8 mg mL⁻¹) with 0.02 mol L⁻¹ metal nitrate solutions in a 1:1 volume ratio, followed by stirring at 600 rpm for 24 h and freeze-drying. The resulting powder was then annealed in an Ar/H₂ atmosphere, heated at a rate of 5 °C min⁻¹ to 300 °C for 2 h to form the transition metal-doped M@Ti₃C₂T_x (M = Fe, Co, Ni, Cu).

3.4. Material Characterization

SEM analysis was conducted on a ZEISS Sigma300 system equipped with an SE2 secondary electron detector. Sample morphology was characterized at an acceleration voltage of 3 kV, while energy spectrum mapping was performed at 15 kV. The powder XRD patterns were obtained on a Rigaku. Transmission electron microscopy (TEM, Talos F200X, ThermoFisher Scientific, Waltham, MA, USA) coupled with energy-dispersive X-ray spectroscopy (EDS) at an acceleration voltage of 200 kV. X-ray diffraction (XRD, NanoFlex600, Rigaku, Tokyo, Japan) with Cu-Kα radiation (λ = 0.154 nm, 40 kV, 40 mA). XPS was carried out on an Thermo Scientific K-Alpha+ instrument and all the high-resolution XPS spectra were corrected by the peak of C1s (284.8 eV). N₂ adsorption–desorption isotherm and the BET surface area measurements were measured at 77 K using Micromeritics ASAP 2460 instrument.

3.5. Electrode Preparation

The catalyst (4 mg), naphthol (10 μL), and ethanol (390 μL) were mixed ultrasonically for 30 min and uniformly drop-coated onto a 1.0 cm² carbon paper substrate. The preparation was subsequently dried under a heat lamp, creating the working electrode. Both the control and experimental groups (e.g., Fe@Ti₃C₂) were prepared using the same procedure.

3.6. Evaluation of Hydrogen Evolution Performance

All electrochemical measurements were conducted at 25.0 °C under ambient pressure using a CHI 760E electrochemical workstation (Chenhua Instruments, Shanghai, China) with a standard three-electrode configuration in 1.0 M KOH electrolyte, where catalyst-loaded carbon paper (loading mass of 4.0 mg cm²), platinum mesh counter electrode, and Hg/HgO (1.0 M KOH) reference electrode were employed. LSV was performed from −0.8 to −1.6 V (vs. Hg/HgO) at 5 mV s⁻¹ with 85% iR compensation, from which the overpotential at 10 mA cm⁻² (η₁₀) and Tafel slope were derived. Double-layer capacitance (C_dl) was determined via cyclic voltammetry in the non-Faradaic region (−0.4 to 0.0 V) across scan rates of 10–100 mV s⁻¹. Electrochemical impedance spectroscopy (EIS) measurements were set from 0.01 to 10⁵ Hz, with an amplitude of 5 mV, and the R_ct was obtained by fitting the Nyquist curve. The stability test of the catalyst was performed using chronoamperometry at a constant applied potential of −1.01 V.

3.7. Data Processing

Overpotential was converted using the formula η = E (vs. RHE) − 0.059 × pH (pH = 13.6, x = 0.098). The Tafel slope was derived from the fitting of η = a + b log|j|. EIS data were analyzed using ZSimpWin version 3.60, where the semicircular arc radius reflects the size of R_ct. The intrinsic activity of electrocatalysts was evaluated through turnover frequency (TOF), defined as the number of hydrogen molecules produced per active site per second. TOF values were calculated at specified overpotentials using the following Equation (1):

$$TOF={\displaystyle \frac{N_{H2}}{N_{Sites}}}={\displaystyle \frac{\left(\frac{j}{2F}\right)\times N_A}{ECSA\ast {\rho }_{sites}}}$$

(1)

where j is the measured current density, F is the Faraday constant (96,485 C mol⁻¹), N_A is the Avogadro constant (6.022 × 10 mol⁻¹), ECSA is the electrochemical surface area, and ρ is the surface density of active sites.

4. Conclusions

This study presents an efficient HER catalyst utilizing a 2D MXene material system through cation vacancy anchoring for transition metal doping. An in situ LiF/HCl etching technique was employed to selectively remove aluminum layers from Ti₃AlC₂, yielding few-layer Ti₃C₂T_x MXene. Surface cation vacancies served as optimal sites for transition metal doping, with Fe@Ti₃C₂ showing the highest catalytic performance, characterized by a η₁₀ of 81 mV in alkaline conditions, a Tafel slope of 33.03 mV dec⁻¹, and the lowest R_ct of 5.6 Ω cm². Mechanistic investigations reveal that Fe’s 3d⁶ configuration enhances catalytic properties, indicating a novel approach for low-cost, high-performance non-precious metal catalysts for green hydrogen technology. Future research should focus on exploring bimetallic synergistic effects and scalable production methods.