Synthesis of two-dimensional N-terminated molybdenum carbides using an alloying strategy in molten salt†

Weiyan Jiang‡ ^abc, Zihan Gao‡^abc, Miao Shen*^abc, Rui Tang^abc, Jing Zhou*^d, Chuanqiang Wu^e, Linjuan Zhang^abc and Jian-Qiang Wang*^abc
aShanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201800, P.R. China. E-mail: shenmiao@sinap.ac.cn; wangjianqiang@sinap.ac.cn
^bUniversity of Chinese Academy of Sciences, Beijing 100049, P.R. China
^cKey Laboratory of Interfacial Physics and Technology, Chinese Academy of Sciences, Shanghai 201800, P.R. China
^dZhejiang Institute of Photoelectronics & Zhejiang Institute for Advanced Light Source, Zhejiang Normal University, Jinhua, Zhejiang 321004, P.R. China. E-mail: zhoujing5457@zjnu.edu.cn
^eInformation Materials and Intelligent Sensing Laboratory of Anhui Province, Key Laboratory of Structure and Functional Regulation of Hybrid Materials of Ministry of Education, Institutes of Physical Science and Information Technology, Anhui University, Hefei 230601, P.R. China

First published on 6th August 2024

---

---

1 Introduction

MXenes are two-dimensional materials like transition metal carbides, nitrides, and carbonitrides.^1–5 Owing to their distinctive structures, electronic properties, and mechanical characteristics, they have garnered considerable attention in various fields.^6–8 Ti₃C₂T_x, the most widely used MXene, is used as a supercapacitor.^9–13 Its preparation generally involves a redox method wherein the Al element is etched from the ternary Ti₃AlC₂ precursor using hydrofluoric acid (HF) or a Lewis acid molten salt as the oxidation agent.^8,14–16

Unlike the conventional MXene, which is widely used as a supercapacitor, Mo₂CT_x has an electronic structure similar to that of platinum, thereby making it suitable for various catalytic applications.^17–19 However, unlike the efficient Al etching from the Ti₃AlC₂ precursor, rapid and efficient Ga etching is a considerable challenge owing to the higher exfoliation energy of the Mo₂Ga₂C precursor.⁸ Synthesis of Mo₂CT_x reportedly required nearly a week of Ga etching in highly oxidative HF or LiF-HCl solutions.^20–22 To date, Mo₂CT_x with surface functional groups other than –F, –Cl, and –OH has not been synthesized in a single step, despite the potential to chemically modify surface terminations (T_x) in MXenes by adjusting the etching agent or through post-treatment.²³

Our study was inspired by the observation that interactions between the external metals and MAX phases can induce the diffusion of A-site atoms, thereby forming new phases. For example, Au reacts with Ti₃SiC₂ to form Ti₃AuC₂ and Ti₃Au₂C₂,²⁴ while Cu reacts with Ti₃AlC₂ to form Ti₃(Al_1−δCu_δ)C₂.²⁵ By leveraging atomic diffusion to reduce the reaction energy barrier, we can synthesize new MAX phases that are typically challenging to obtain using conventional solid sintering methods.

In this study, we propose a novel approach for efficient Ga etching from a Mo₂Ga₂C precursor to produce Mo₂CT_x, which involves using an iron triad element (Fe, Co, or Ni) and Li₃N in a KCl–LiCl molten salt. The iron triad element forms an alloy with Ga, thereby lowering the reaction energy barrier and facilitating Ga etching. Simultaneously, adding Li₃N to the molten salt introduces new nitrogen functional groups and promotes the formation of Mo₂CN₂ nanoflakes. The synthesized N-terminated Mo₂CT_x exhibits higher electrocatalytic oxygen evolution reaction activity than Mo₂CT_x prepared using conventional HF solution and commercial IrO₂ in a 1 M KOH solution. This new strategy not only enables the one-step preparation of novel MXenes, but also presents new possibilities for their application in electrocatalysis.

2 Experimental section

2.1 Chemicals and materials

Molybdenum gallium carbon (Mo₂Ga₂C, 99 wt%) was purchased from 11 technology company (China). Lithium nitride (Li₃N, 99.4 wt%), potassium chloride (KCl, 99.8 wt%), lithium chloride (LiCl, 99.99 wt%), nickel powder (Ni, 99.5 wt%), cobalt powder (Co, 99.9 wt%), iron powder (Fe, 99.9 wt%), Nafion (5 wt%), isopropyl alcohol (99.7 wt%), potassium hydroxide (KOH, ≥85.0 wt%), iridium oxide (IrO₂, 99 wt%), acetylene black (99.9 wt%), hydrochloric acid (HCl, 37 wt%), and hydrogen fluoride (HF, 40 wt%) were purchased from Sinopharm Chemical Reagent Co., Ltd.

2.2 Synthesis of A-MS-Mo₂CT_x

Mo₂Ga₂C (100 mg), metal powder (Ni or Co or Fe, 100 mg), and Li₃N (200 mg) were thoroughly mixed and pressed into a tablet at 30 MPa under an air atmosphere. The prepared tablet was placed in a glassy carbon crucible and transferred to a resistance furnace within an argon-protected glovebox. The reaction temperature was increased from room temperature to 700 °C at a rate of 3 °C min⁻¹ and maintained at 700 °C for 24 h. After naturally cooling to room temperature, the product was washed with deionized water and HCl to remove residual salt and metal impurities, respectively. Finally, A-MS-Mo₂CT_x was obtained by drying in a vacuum oven for 12 h.

2.3 Synthesis of HF-Mo₂CT_x

LiF (1.0 g) was added to 15 mL of 12 M HCl and stirred continuously in a Teflon reactor at 90 °C for 1 h. After gradually adding 0.5 g of Mo₂Ga₂C and 5 mL of HF solution, the Teflon reactor was kept in a muffle furnace at 150 °C for 72 h and naturally cooled to room temperature.²⁶ The product was centrifuged at 5000 rpm for 5 min, and rinsed using deionized water, and this was repeated several times until the pH of the supernatant reached 6–7. Finally, the product was dried at 80 °C for 24 h in a vacuum oven to obtain HF-Mo₂CT_x.

2.4 Material characterization

X-ray diffraction (XRD) analyses were performed using a Bruker D8 Advance X-ray diffractometer with Cu-Kα radiation (λ = 1.54 Å). Morphological analyses were conducted using transmission electron microscopy (TEM, JEM-2100) coupled with energy-dispersive X-ray spectroscopy (EDX) and scanning electron microscopy (SEM, FEI Apreo S) coupled with energy-dispersive spectroscopy (EDS). X-ray photoelectron spectroscopy (XPS) was performed on an ESCALAB 250Xi (Thermo Fisher) spectrometer using an Al-Kα radiation (λ = 8.34 Å) power source. Peak fitting was performed using Casa XPS. Raman spectroscopy data were collected on a Lab RAMHR800 instrument (Jobin Yvon, France) equipped with an air-cooled CCD array detector in the backscattering configuration. The 473 nm line of an argon ion laser (100 mW) was used to excite the sample. The atomic structure was analyzed using transmission electron microscopy (STEM) mode on a double Cs corrected FEI Titan Themis Z60-300 system equipped with a Super-X EDS, operated at 300 kV. The Mo K-edge X-ray absorption spectroscopy (XAS) data were collected at the 14W1 beamline in the Shanghai Synchrotron Radiation Facility (SSRF). The X-rays were monochromatized using a double-crystal Si (111) monochromator, and the energy was calibrated using molybdenum metal foil for the Mo K-edge. A gas chromatograph (GC2060) equipped with three thermal conductivity detectors was used to determine the gas composition during the reaction process.

2.5 Evaluation of catalytic performance

Electrochemical tests were conducted using an Autolab electrochemical workstation (PGSTAT 302 N, Metrohm Auto Lab Co., Ltd) with a rotating-disk three-electrode system (RDE, Pine Research Instrumentation). Prior to each test, 1 M KOH was saturated with O₂ for 30 min at room temperature (approximately 25 °C). A mixture of isopropyl alcohol and deionized water was prepared in a 7:3 volume ratio. Furthermore, 10 mg of catalyst, 5 mg of acetylene black, and 80 μL of Nafion (Sigma-Aldrich) were added and ultrasonicated for 0.5 h to form a uniform catalyst ink. The prepared ink was later applied to a glassy carbon (GC) electrode with an area of 0.196 cm² to create a working electrode. A graphite rod and Hg/HgO (1.0 M KOH) electrodes served as the counter and reference electrodes, respectively. The potentials versus Hg/HgO were converted to the reversible hydrogen electrode (RHE) as expressed in eqn (1) as follows:^27,28

ERHE = EHg/HgO + 0.0591 × pH + 0.097	(1)

LSV polarization curves were measured in a 1.0 M KOH aqueous electrolyte using a rotating-disk electrode at 1600 rpm with a scan rate of 5 mV s⁻¹. Tafel slopes were calculated using eqn (2) by fitting the linear portions of the LSV curve at a low overpotential as follows:²⁹

η = blogj + a	(2)

where b is the Tafel slope, η the overpotential, and j the current density.

Electrochemical impedance spectroscopy (EIS) analysis was conducted over a frequency range from 100 kHz to 100 mHz at a current density of 10 mA cm⁻². Cyclic voltammetry (CV) curves were recorded within the potential range of 0.1–0.3 V vs. RHE at various scan rates. The C_dl was calculated by plotting the current differences at a potential of 0.225 V vs. RHE against the scan rates as expressed in eqn (3) as follows:³⁰

Cdl = Ic/v	(3)

where I_c represents the charging current and v the scan rate.

2.6 Computation details

Density functional theory (DFT) calculations⁵ were conducted using a frozen-core projector-augmented wave (PAW) method³¹ with the Vienna ab initio simulation package (VASP).³² The Perdue–Burke–Ernzerhof (PBE) version of the generalized gradient approximation (GGA) was used to treat the electron–electron exchange and correlation interactions.³³ Transition-state calculations were conducted using a CI-NEB method.³⁴ DFT was used to analyze the process of etching Ga from Mo₂Ga₂C and to assess the stability of the surface functional groups. The transition states for Ga etching from Mo₂Ga₂C via alloying and oxidation were simulated using the nudged elastic band (NEB) method. The K-point was set to 3 × 6 × 1, while the vacuum layer and plane wave cut-off energy were 15 Å and 500 eV, respectively. Owing to the limitations of the super-soft pseudopotential in reciprocal space, which affects exchange-correlation and underestimates the band gap for GGA functions, a GGA + U method was employed to accurately describe the d orbitals of transition metal elements. The U_eff values of Mo, Ga, and Ni are 4.0,³⁵ 3,³⁶ and 2.95 eV,³⁷ respectively. The atomic positions were optimized until the total energy change was less than 1 × 10⁻⁶ eV per atom and maximum force on each atom was less than 0.01 eV Å⁻¹. As regards the calculation of adsorption energy of functional groups with different valence states, varying numbers of Li atoms were placed and fixed approximately 4 Å above the surface of the Mo₂Ga₂C supercell to adjust the valence of the terminal groups. The adsorption energy on the surface was expressed using eqn (4) as follows:³⁸

Eads = ET/slab − Eslab − ET	(4)

where E_T/slab and E_slab represent the energies with and without the termination adsorbate, respectively, and E_T the energy of the termination adsorbate in an empty box. Here, the classical d-band models can be used as effective descriptors to evaluate electrocatalytic reactions.³⁹ The band center level (b) is expressed using eqn (5) as follows:⁴⁰

	(5)

where D (E) represents the density of states (DOS) of Mo d-orbitals. Here, the model identifies the position of the orbital center and filling degree of the orbitals. Furthermore, the d-band center level shifts down deeper than the Fermi level, thereby resulting in higher d-electron occupation in σ* states.⁴¹

3 Results and discussion

3.1 Structure of Mo₂CT_x etched using the alloying method in molten salt

Mo₂CT_x was synthesized using Mo₂Ga₂C as a precursor via a sequence of alloying and etching steps, as shown in Fig. 1a. The Mo₂Ga₂C was mixed with Ni (Fe or Co) powder, Li₃N, and a eutectic KCl–LiCl molten salt, and was later pressed into a pellet shape (Fig. S1†). The pellet was heated at 700 °C for 24 h in an inert atmosphere. The product was washed using deionized water and hydrochloride to remove residual molten salts and GaNi₃ alloy byproducts (Fig. S2†). This product was labeled as A-MS-Mo₂CT_x. As shown in the XRD pattern in Fig. 1b, the characteristic peaks of Mo₂Ga₂C were observed at 9.93°, 34.19°, 37.65°, 39.96°, 42.85°, 49.61°, 53.46°, and 61.17°, which correspond to the (002), (100), (103), (008), (105), (107), (108), and (110) planes, respectively. These peaks correspond with those reported in the literature.⁴² After alloying and etching in the molten salt, the distinct peaks corresponding to Mo₂Ga₂C considerably reduced, while five new characteristic peaks emerged at 6.46°, 12.87°, 19.40°, 25.81°, and 32.44°, thereby indicating a notable increase in the c-lattice parameter (c-LP) from 17.86 to 26.60 Å. Contrary to the Mo₂Ga₂C precursor, which had a high Ga content and closely stacked layers (Fig. S3†), A-MS-Mo₂CT_x showed a structure comprising ultra-thin sheets arranged in a flower-like pattern (Fig. 1c). The Ga content in the product was considerably reduced from 13.06 to 1.00% (Fig. S3 and S4†). Adjusting the reaction temperature and time further decreased the Ga content (Fig. S5 and S6†). The uniform distribution of Mo, C, and N on the surface (Fig. 1d) indicates that –N functional groups were successfully introduced into A-MS-Mo₂CT_x. Transmission electron microscopy (TEM) images showed that A-MS-Mo₂CT_x had an ultra-thin nanosheet layer morphology (Fig. 1e) with an interlayer spacing of 1.37 nm (Fig. 1f), which corresponds to the XRD analysis. STEM images confirmed the lamellar microstructure of A-MS-Mo₂CT_x (Fig. S7†). Furthermore, electron energy loss spectroscopy (EELS) detected an N K-edge at 400 eV (Fig. 1g), thereby indicating the incorporation of N into A-MS-Mo₂CT_x. EDS and EDX analyses determined a Mo/N ratio of approximately 1 (Fig. S4 and S8†), thereby indicating the formation of Mo₂CN₂ following its etching in the Ni–Li₃N–KCl–LiCl system. Additionally, the content of the N groups on the surface can be adjusted by varying the ratio of Mo₂Ga₂C to Li₃N (Fig. S9†).

To further investigate the surface structure of A-MS-Mo₂CT_x, numerous techniques were used, including X-ray photoelectron spectroscopy (XPS), Raman spectroscopy, X-ray absorption near-edge structure (XANES), and extended X-ray absorption fine structure (EXAFS). The distinctive peaks of Ga, including Ga 2p (binding energies of 1144.7 and 1117.9 eV),^18,43 Ga 3s, Ga 3p, and Ga 3d (centered at 160.5, 105.5, and 19.2 eV, respectively) were considerably reduced after Mo₂Ga₂C was etched by the alloying process in molten salt (Fig. 2a). As presented in Table S1†, the Ga 2p content decreased significantly from 17.56% in Mo₂Ga₂C to 1.63% in A-MS-Mo₂CT_x, thereby indicating that a considerable amount of the Ga had been eliminated.⁴²

The high-resolution XPS spectra of Mo₂Ga₂C at the Mo 3d core level exhibited three doublet peaks at 227.8 (230.9 eV), 228.8 (232.0 eV), and 232.2 eV (235.4 eV), corresponding to Mo–C, Mo⁴⁺, and Mo⁵⁺, respectively (Fig. 2b and Table S2†). The Mo valence states (+4 and +5) were attributed to their surface oxides.^42,44 Conversely, the Mo 3d XPS spectra for A-MS-Mo₂CT_x exhibited only two doublet peaks at 229.2 (232.4) eV and 232.0 (235.1) eV, corresponding to N–Mo–C and Mo⁵⁺,^18,44 respectively (Fig. 2b and Table S2†). This shift toward higher valence states of Mo in A-MS-Mo₂CT_x can be attributed to the removal of Ga and introduction of N-termination. The C 1s spectrum of Mo₂Ga₂C exhibited four peaks at 283.3, 284.8, 285.7, and 288.7 eV (Fig. 2c and Table S3†), which correspond to the Mo–C, C–C, C–O, and CO bonds, respectively.^18,42 Contrarily, the binding energy of the Mo–C bond shifted toward a higher value in the A-MS-Mo₂CT_x spectrum. Additionally, a new peak at 286.5 eV was identified, thereby corresponding to the C–N bond (Fig. 2c).⁴⁴ The N 1s high-resolution spectrum further showed that the main peak at 397.2 eV corresponded to the Mo–N bond, while the peaks at 398.0, 399.2 and 400.3 eV were attributed to pyridine N, pyrrole N, and graphitic N, respectively (Fig. 2d and Table S4†).^44–46 The ratio of Mo/N determined using XPS was approximately 1 (Table S1†), which corresponds to the findings from STEM and EDS analyses, further supporting the formation of Mo₂CN₂. Raman spectroscopy was used to analyze Mo₂CT_x with various functional groups. As shown in Fig. 2e, Mo₂CT_x synthesized using HF (HF-Mo₂CT_x, Fig. S10†) displayed two peaks at 91 and 285 cm⁻¹, which correspond to the E_g and A_g modes of the MXene flakes dominated by Mo atoms, respectively.⁴⁷ The presence of F and O terminated groups derived from the HF solution, which were randomly distributed at different sites on the surface of the Mo atoms, resulted in complex Raman peaks of HF-Mo₂CT_x (labeled as *). However, the characteristic peak of Mo–C in A-MS-Mo₂CT_x was enhanced at 88 cm⁻¹ owing to the introduction of the N-terminations on the surface, thereby indicating that A-MS-Mo₂CT_x had a more ordered Mo–C framework and better crystallinity, which is in agreement with the XRD results. The in-plane A_1g band at 660 cm⁻¹ represented the vibrations of the Mo atoms on the surface, with its position and width influenced by the atoms attached to Mo.^47,48 The peak positions differed between HF-Mo₂CT_x (664 cm⁻¹) and A-MS-Mo₂CT_x (632 cm⁻¹) owing to the different surface functional groups. Further investigation into the local atomic coordination and electronic structure of Mo was conducted using XAS. The absorption edge positions of 20006.7, 20012.9, and 20016.0 eV corresponded to various Mo valence states in Mo (0), MoO₂ (+4), and MoO₃ (+6), respectively (Fig. 2f). The absorption edges of Mo in Mo₂Ga₂C and A-MS-Mo₂CT_x were measured at 20009.6 and 20012.1 eV, respectively. The XANES results indicated that the Mo valence states in Mo₂Ga₂C and A-MS-Mo₂CT_x were approximately +2 and +3.5, respectively (Fig. 2g), thereby indicating that the removal of Ga atoms and introduction of N-terminations increased the average Mo valence state in A-MS-Mo₂CT_x, which corresponded to the XPS analysis. The Fourier-transform EXAFS spectra revealed peaks at 1.63 and 2.63 (2.23) Å relative to the coordination of Mo–N/C (Mo–T_x) and the Mo–Mo shell for A-MS-Mo₂CT_x in R-space, respectively (Fig. 2h). The fitting bond length of Mo–N/C (Mo–T_x/C) was 2.11 Å, while the two Mo–Mo bond lengths were 2.79 and 3.10 Å, thereby corresponding to the Mo–Mo distance between different layers and within a similar layer, respectively (Fig. S11 and Table S5†). The first shell scattering coordination number for Mo–N/C (Mo–T_x/C) was 5.9, which is considerably higher than that of Mo₂Ga₂C (3.0 as presented in Table S5†). Wavelet transform (WT) analysis of the Mo K-edge EXAFS oscillations showed that the maximum WT for A-MS-Mo₂CT_x at 5.0 and 10.0 Å⁻¹ was attributed to the Mo–C/T_x and Mo–Mo bonds (Fig. 2i), respectively, which is different from those of Mo₂Ga₂C (4.0 and 8.0 Å⁻¹, Fig. 2i). These XAS results further confirmed the removal of Ga and introduction of N surface terminations.

3.2 Experimental and density-functional theory (DFT) analyses of the Mo₂Ga₂C etching mechanism

The formation of MXenes involve the etching of element A from the MAX precursor and adsorption of functional groups.^13,49,50 The etching process generally alters the valence state of element A via conventional redox reactions.^8,51,52 However, in the etching product GaNi₃ alloy (Fig. S2 and S12†), the valence state of Ga remains unchanged compared to its state in the Mo₂Ga₂C precursor †. Hence, two possible reaction pathways may result in the formation of GaNi₃ alloys. One pathway is the conventional redox process, wherein Ga in Mo₂Ga₂C is initially oxidized to Ga₂O₃ in molten salt (Fig. S14†), which is further reduced to Ga by Li₃N, and finally alloyed with Ga to form GaNi₃. Another etching pathway involves the direct alloying of Ga in Mo₂Ga₂C with Ni, thereby forming GaNi₃ alloy particles without any redox reaction. To further investigate the Ga etching pathways, DFT calculations were performed. The etched products of Ga₂O₃ and GaNi₃ were modeled above the (110) interface of Mo₂Ga₂C, both featuring Ga defect sites (Fig. S15 and S16†). As shown in Fig. 3, Ga migrates from the Mo₂Ga₂C interlayer to the defect sites, thereby forming Ga₂O₃ via oxidation or GaNi₃ via alloying. The Gibbs free energy of Ga₂O₃ formation is −3.4 eV, which is considerably lower than −0.28 eV for GaNi₃ alloy formation, thereby indicating that Ga is thermodynamically more likely to be etched via oxidation. However, transition state searches using the climbing image nudged elastic band (CI-NEB) method (Fig. S17 and S18†) revealed that the kinetic energy barrier for Ga oxidation is 1.5 eV, which is higher than the 1.18 eV for Ga alloying. Hence, a direct alloying pathway is more favorable for dynamic Ga etching from the Mo₂Ga₂C interlayer. Furthermore, in the presence of Ni and absence of Li₃N in the reaction system, Ga atoms can also be etched via alloying to form GaNi₃ (Fig. S19†), thereby confirming that Ga etching from Mo₂Ga₂C occurs via a direct alloying path.

Although Ga atoms can be etched by alloying with Ni, the etched Mo₂C* structure is unstable without Li₃N in the reaction system and will transform into β-Mo₂C at high temperatures (Fig. S19†). Hence, the adsorption of anions from molten salt to form functional groups is crucial for maintaining a stable layered structure.⁵³ Considering that Li₃N easily hydrolyzes to LiOH,^54,55 the adsorption energies of different anions, such as N, Cl, OH, and O as surface termination were calculated. The adsorption energy of N generated by the LiN cluster as the functional groups is −3.67 eV (Fig. 3), which is considerably lower than those of cluster LiCl, Li₂O, and LiOH. Additionally, as the Ga atoms are etched and number of functional groups increases, the adsorption energy of the LiN cluster on the Mo₂C* surface decreases (Fig. S20–S22†). Hence, more N anions produced by the LiN cluster can be easily adsorbed on the Mo surface, thereby forming a more stable phase (Mo₂CN₂) than that the other anions owing to the lower average adsorption energy (Fig. S23†). Furthermore, H₂ was detected using gas chromatography when Mo₂Ga₂C was etched in the Ni–Li₃N–KCl–LiCl system (Fig. S24†), thereby indicating that LiOH produced by Li₃N hydrolysis was involved in the reaction. Hence, the etching reaction mechanism can be expressed using eqn (6). Herein, the Ga atoms in Mo₂Ga₂C are etched by forming GaNi₃. Simultaneously, the specific N ions produced by the reaction of Li₃N and LiOH are adsorbed on the surface of Mo₂C* as a stable functional group, expressed as follows:

Mo2Ga2C + 6Ni + 2Li3N + 6LiOH → Mo2CN2 + 2GaNi3 + 6Li2O + 3H2↑	(6)

As a synthetic medium, an appropriate amount of LiCl–KCl can improve the homogeneity of the reaction and promote the conversion of Mo₂Ga₂C to Mo₂CN₂ (Fig. S25†). Under similar conditions, Mo₂Ga₂C can be etched by alloying with Fe or Co in a KCl–LiCl molten salt containing Li₃N, thereby resulting in the formation of Mo₂CN₂ (Fig. S26 and S27†).

3.3 Improvement of electrochemical properties by N-termination

Incorporating non-metal heteroatoms (N, O, and S) into MXenes is effective in regulating the surface chemistry, thereby facilitating electron/ion movement and enhancing the electrocatalytic activity.⁵⁶ Tang et al.⁵⁷ found that Ti₃C_1.6N_0.4 and Ti₃C_1.8N_0.2, with a higher N content and pyridinic N species, exhibited higher OER activity than pristine Ti₃C₂T_x. This implies that Mo–N species can promote charge transfer and improve the catalytic performance. To evaluate the OER activity, linear sweep voltammetry (LSV) was conducted on A-MS-Mo₂CT_x, HF-Mo₂CT_x, and commercial IrO₂ in 1 M KOH (Fig. 4a). As shown in Fig. 4b, the overpotential at 10 mA cm⁻² for A-MS-Mo₂CT_x was 320 mV, which is considerably lower than that of conventional F-terminated HF-Mo₂CT_x (570 mV) and commercial IrO₂ (410 mV). Furthermore, A-MS-Mo₂CT_x exhibited a Tafel slope of 53.7 mV dec⁻¹, which surpasses that of HF-Mo₂CT_x (82.8 mV dec⁻¹) and IrO₂ (73.3 mV dec⁻¹), thereby indicating enhanced OER kinetics and a faster reaction rate (Fig. 4c). EIS was used to investigate the charge and mass transfer characteristics of the materials, and the results are shown in Fig. 4d and Table S6.† A-MS-Mo₂CT_x exhibited a smaller semicircle (8.6 Ω) than IrO₂ (58.2 Ω) and HF-Mo₂CT_x (101 Ω) (Fig. S28 and Table S6†), thereby indicating its lower charge transfer impedance. This reaction impedance can be attributed to the N surface termination in Mo-based MXene, which enhances the migration of electrons and ions. To calculate the electrochemically active surface area (ECSA), the double-layer capacitance (C_dl) was measured using CV at different sweep rates (Fig. S29†). The C_dl of A-MS-Mo₂CT_x was 18.6 mF cm⁻² (Fig. S30†), which corresponds to an estimated active site area of approximately 465 cm_ECSA² cm_geo⁻².³⁰ Compared to the C_dl of IrO₂ (1.9 mF cm⁻²), the ECSA of A-MS-Mo₂CT_x was approximately ten times larger than that of IrO₂. The ECSA-normalized oxygen evolution current (j) was used to highlight the intrinsic catalytic activity (Fig. S31†). The results indicate that the superior OER performance of N-terminated Mo₂CT_x, relative to IrO₂, is principally attributed to its enhanced ECSA, which is an advantage among two-dimensional materials. Furthermore, the electrocatalytic durability of A-MS-Mo₂CT_x has been tested at a current density of 10 mA cm⁻², wherein it maintained 97% of its potential after 60 h (Fig. 4e). Even at a high current density of 500 mA cm⁻², A-MS-Mo₂CT_x maintained relative stability (Fig. S32†). Consequently, A-MS-Mo₂CT_x exhibits outstanding OER performance, thereby surpassing most MXene composites (Fig. S33†) and rivaling precious-metal benchmark catalysts.

To further understand the OER mechanism, the structures of the samples were characterized using ex situ XRD after different reaction times. As regards HF-Mo₂CT_x, the characteristic peak at 8.2° gradually disappeared after the OER (Fig. S34†), thereby indicating the destruction of its two-dimensional layer structure using high ECSA, which correspondingly reduced its activity. However, the characteristic peaks at 6.46° and 12.87° for A-MS-Mo₂CT_x persisted throughout the OER process (Fig. S35†), thereby indicating that the stable two-dimensional layer structure was maintained, resulting in good durability over long reactions. Furthermore, the stability was confirmed by the SEM-EDS results, thereby showing that the morphology of the layered structure remained unchanged, and Mo, C, N and O elements were uniformly distributed (Fig. S36†).

Furthermore, A-MS-Mo₂CT_x was characterized using XAS and XPS after the OER to explore the origins of its high activity. The Mo K-edge XANES analysis indicated that the valence of Mo increased from +3.5 to +4.0 after the OER (Fig. S37†), which is beneficial for OER activity.²³ Considering that A-MS-Mo₂CT_x maintains a stable 2D structure during the OER, it is speculated that the increase in the Mo valence state is owing to the change in the surface functional group during the adsorption/desorption of intermediate products. Surface electronic information is obtained using XPS, which has a detection depth of 2–5 nm.⁵⁸ Compared to the bulk, the Mo valence on the surface after the OER considerably increased, with the content of Mo⁶⁺ reaching 38.2% (Fig. S38 and Table S7†), which is likely influenced by the change in the surface functional groups. The change in Mo valence in HF-Mo₂CT_x was also investigated using XPS. The results showed that the T_x–Mo–C bond disappeared, thereby indicating a corresponding decrease in the activity (Fig. S38 and Table S7†). To further understand the difference in Mo valence between A-MS-Mo₂CT_x and HF-Mo₂CT_x, DFT calculations were performed. The Mo-based MXene models with –F, –N, and –O surface terminations were constructed to simulate the effects of different functional groups (Fig. S39 and Table S8†). The calculated Bader charges for Mo were found to be +1.55, +1.61, and +1.78, for –F, –N, and –O-termination, respectively, thereby corresponding to experimental observations of the valence states. Beyond the charge distribution at Mo sites, surface functional groups also influence the d-band centers, which are key descriptors for OER activity.^39,59 The d-band center of Mo with an N-termination is approximately −2.52 eV, which is lower than that with an O-termination (−1.96 eV) and F-termination (−1.88 eV) (Fig. S40†). This downshift results in a more stable bonding state and stronger covalent interaction,^60,61 thereby enhancing the OH⁻ absorption and enabling the OER. Electron localization function (ELF) analysis revealed that N-terminated Mo₂C frameworks exhibit weakened ionization around Mo atoms, forming stable N–Mo–C bonds and improving the stability of the 2D structure (Fig. S41†). Hence, the N-terminated Mo₂CT_x optimizes the electronic structure of surface Mo by inducing high valence and strong covalency, thereby promoting the OER catalytic performance.

4 Conclusions

A novel and highly efficient alloying strategy was developed in this study for synthesizing N-terminated Mo₂CT_x. Ga atoms in Mo₂Ga₂C were selectively etched into alloys using an iron triad element (Fe, Co, or Ni) in LiCl–KCl molten salt with Li₃N. The N clusters generated by Li₃N in the molten salt were adsorbed on the Mo surface as functional groups, thereby forming Mo₂CN₂ nanoflakes. The direct alloying process reduced the reaction energy barrier and enhanced the etching kinetics. The synthesized N-terminated Mo₂CT_x exhibited higher electrocatalytic OER activity than conventional HF-Mo₂CT_x and commercial IrO₂ under alkaline conditions. This novel direct alloying method shows great potential for synthesizing a wide variety of 2D materials and paves the way for new explorations of their applications in electrocatalysis.

Data availability

The data that support the findings of this study are available from the corresponding authors, M. Shen & J. Zhou & J.-Q. Wang, upon reasonable request.

Author contributions

Weiyan Jiang: writing – original draft, validation, investigation, formal analysis, data curation. Zihan Gao: data curation, formal analysis, writing – review & editing. Miao Shen: conceptualization, investigation, validation, writing – review & editing, funding acquisition. Jing Zhou: methodology, investigation, formal analysis, writing – review & editing. Rui Tang: validation, software, methodology. Chuangqiang Wu: methodology, formal analysis. Linjuan Zhan: resources & formal analysis. Jian-Qiang Wang: funding acquisition, resources, supervision.

Conflicts of interest

The authors declare that they have no conflict of interest.

Acknowledgements

This study was financially supported in part by the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA0400000), Natural Science Foundation of Shanghai (Grant No. 21ZR1476200), Zhejiang Provincial Natural Science Foundation of China (LR24A050001), and “Transformational Technologies for Clean Energy and Demonstration” Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA21000000). The authors also appreciate the Shanghai Synchrotron Radiation Facility (14W1, SSRF) for their support in material characterization.

References

1. H. Ding, Y. Li, M. Li, K. Chen, K. Liang, G. Chen, J. Lu, J. Palisaitis, P. O. Å. Persson, P. Eklund, L. Hultman, S. Du, Z. Chai, Y. Gogotsi and Q. Huang, Science, 2023, 379, 1130–1135 CrossRef CAS PubMed .
2. D. Wang, C. Zhou, A. S. Filatov, W. Cho, F. Lagunas, M. Wang, S. Vaikuntanathan, C. Liu, R. F. Klie and D. V. Talapin, Science, 2023, 379, 1242–1247 CrossRef CAS .
3. M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi and M. W. Barsoum, Adv. Mater., 2011, 23, 4248–4253 CrossRef CAS .
4. M. Naguib, M. W. Barsoum and Y. Gogotsi, Adv. Mater., 2021, 33, 2103393 CrossRef CAS .
5. W. Kohn and L. J. Sham, Phys. Rev., 1965, 140, A1133 CrossRef .
6. A. VahidMohammadi, J. Rosen and Y. Gogotsi, Science, 2021, 372, eabf1581 CrossRef CAS PubMed .
7. X. Li, Z. Huang, C. E. Shuck, G. Liang, Y. Gogotsi and C. Zhi, Nat. Rev. Chem, 2022, 6, 389–404 CrossRef .
8. K. R. G. Lim, M. Shekhirev, B. C. Wyatt, B. Anasori, Y. Gogotsi and Z. W. Seh, Nat. Synth., 2022, 1, 601–614 CrossRef .
9. X. Wang, T. S. Mathis, K. Li, Z. Lin, L. Vlcek, T. Torita, N. C. Osti, C. Hatter, P. Urbankowski, A. Sarycheva, M. Tyagi, E. Mamontov, P. Simon and Y. Gogotsi, Nat. Energy, 2019, 4, 241–248 CrossRef CAS .
10. M. Shen, W. Jiang, K. Liang, S. Zhao, R. Tang, L. Zhang and J.-Q. Wang, Angew. Chem., Int. Ed., 2021, 60, 27013–27018 CrossRef CAS PubMed .
11. C. Liu, Y. Bai, W. Li, F. Yang, G. Zhang and H. Pang, Angew. Chem., Int. Ed., 2022, 61, e202116282 CrossRef CAS PubMed .
12. M. K. Aslam, Y. Niu and M. Xu, Adv. Energy Mater., 2021, 11, 2000681 CrossRef CAS .
13. M. R. Lukatskaya, O. Mashtalir, C. E. Ren, Y. Dall'Agnese, P. Rozier, P. L. Taberna, M. Naguib, P. Simon, M. W. Barsoum and Y. Gogotsi, Science, 2013, 341, 1502–1505 CrossRef CAS .
14. Y. Li, H. Shao, Z. Lin, J. Lu, L. Liu, B. Duployer, P. O. Å. Persson, P. Eklund, L. Hultman, M. Li, K. Chen, X.-H. Zha, S. Du, P. Rozier, Z. Chai, E. Raymundo-Piñero, P.-L. Taberna, P. Simon and Q. Huang, Nat. Mater., 2020, 19, 894–899 CrossRef CAS .
15. L. Liu, M. Orbay, S. Luo, S. Duluard, H. Shao, J. Harmel, P. Rozier, P.-L. Taberna and P. Simon, ACS Nano, 2022, 16, 111–118 CrossRef CAS .
16. D. Gandla, Z. Zhuang, V. V. Jadhav and D. Q. Tan, Energy Storage Mater., 2023, 63, 102977 CrossRef .
17. H. Ang, H. Wang, B. Li, Y. Zong, X. Wang and Q. Yan, Small, 2016, 12, 2859–2865 CrossRef CAS .
18. H. Zhou, Z. Chen, E. Kountoupi, A. Tsoukalou, P. M. Abdala, P. Florian, A. Fedorov and C. R. Müller, Nat. Commun., 2021, 12, 5510 CrossRef CAS PubMed .
19. R. B. Levy and M. Boudart, Science, 1973, 181, 547–549 CrossRef CAS PubMed .
20. J. Halim, S. Kota, M. R. Lukatskaya, M. Naguib, M.-Q. Zhao, E. J. Moon, J. Pitock, J. Nanda, S. J. May, Y. Gogotsi and M. W. Barsoum, Adv. Funct. Mater., 2016, 26, 3118–3127 CrossRef CAS .
21. Y. Guo, S. Jin, L. Wang, P. He, Q. Hu, L.-Z. Fan and A. Zhou, Ceram. Int., 2020, 46, 19550–19556 CrossRef CAS .
22. A. Byeon, C. B. Hatter, J. H. Park, C. W. Ahn, Y. Gogotsi and J. W. Lee, Electrochim. Acta, 2017, 258, 979–987 CrossRef CAS .
23. Y. Wang, Y. Nian, A. N. Biswas, W. Li, Y. Han and J. G. Chen, Adv. Energy Mater., 2021, 11, 2002967 CrossRef CAS .
24. H. Fashandi, M. Dahlqvist, J. Lu, J. Palisaitis, S. I. Simak, I. A. Abrikosov, J. Rosen, L. Hultman, M. Andersson, A. Lloyd Spetz and P. Eklund, Nat. Mater., 2017, 16, 814–818 CrossRef CAS PubMed .
25. M. Nechiche, T. Cabioc'h, E. N. Caspi, O. Rivin, A. Hoser, V. Gauthier-Brunet, P. Chartier and S. Dubois, Inorg. Chem., 2017, 56, 14388–14395 CrossRef CAS PubMed .
26. W. Jiang, H. Shi, M. Shen, R. Tang, Z. Tang and J.-Q. Wang, ACS Appl. Mater. Interfaces, 2022, 14, 14482–14491 CrossRef CAS PubMed .
27. Y. Hu, J. Zhou, L. Li, Z. Hu, T. Yuan, C. Jing, R. Liu, S. Xi, H. Jiang and J.-Q. Wang, J. Mater. Chem. A, 2022, 10, 602–610 RSC .
28. H. Jo, Y. Yang, A. Seong, D. Jeong, J. Kim, S. H. Joo, Y. J. Kim, L. Zhang, Z. Liu and J.-Q. Wang, J. Mater. Chem. A, 2022, 10, 2271–2279 RSC .
29. T. Shinagawa, A. T. Garcia-Esparza and K. Takanabe, Sci. Rep., 2015, 5, 13801 CrossRef PubMed .
30. Z. W. Seh, K. D. Fredrickson, B. Anasori, J. Kibsgaard, A. L. Strickler, M. R. Lukatskaya, Y. Gogotsi, T. F. Jaramillo and A. Vojvodic, ACS Energy Lett., 2016, 1, 589–594 CrossRef CAS .
31. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169 CrossRef CAS PubMed .
32. J. Hafner and J. Compu, Chem, 2008, 29, 2044–2078 CAS .
33. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865 CrossRef CAS .
34. G. Henkelman, B. P. Uberuaga and H. Jónsson, J. Chem. Phys., 2000, 113, 9901–9904 CrossRef CAS .
35. M. Samanian and M. H. Ghatee, J. Mol. Model., 2021, 27, 268 CrossRef CAS .
36. J. Kaczkowski, Acta Phys. Pol., A, 2012, 121, 1142–1144 CrossRef CAS .
37. Y. Xie and J. A. Blackman, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 69, 172407 CrossRef .
38. L. Dai, F. Yao, L. Yu, C. Fang, J. Li, L. Xue, S. Zhang, P. Xiong, Y. Fu and J. Sun, Adv. Energy Mater., 2022, 12, 2200974 CrossRef CAS .
39. S. Jiao, X. Fu and H. Huang, Adv. Funct. Mater., 2022, 32, 2107651 CrossRef CAS .
40. B. Wei, Z. Fu, D. Legut, T. C. Germann, Q. Zhang, S. Du, H. Zhang, J. S. Francisco and R. Zhang, J. Phys. Chem. C, 2021, 125, 4477–4488 CrossRef CAS .
41. J. Yuan, Y. Liu, T. Bo and W. Zhou, Int. J. Hydrogen Energy, 2020, 45, 2681–2688 CrossRef CAS .
42. C. Wang, H. Shou, S. Chen, S. Wei, Y. Lin, P. Zhang, Z. Liu, K. Zhu, X. Guo, X. Wu, P. M. Ajayan and L. Song, Adv. Mater., 2021, 33, 2101015 CrossRef CAS .
43. F. Scharmann, G. Cherkashinin, V. Breternitz, C. Knedlik, G. Hartung, T. Weber and J. A. Schaefer, Surf. Interface Anal., 2004, 36, 981–985 CrossRef CAS .
44. L. Dai, F. Yao, L. Yu, C. Fang, J. Li, L. Xue, S. Zhang, P. Xiong, Y. Fu, J. Sun and J. Zhu, Adv. Energy Mater., 2022, 12, 2200974 CrossRef CAS .
45. J. Jia, T. Xiong, L. Zhao, F. Wang, H. Liu, R. Hu, J. Zhou, W. Zhou and S. Chen, ACS Nano, 2017, 11, 12509–12518 CrossRef CAS PubMed .
46. Y.-Y. Chen, Y. Zhang, W.-J. Jiang, X. Zhang, Z. Dai, L.-J. Wan and J.-S. Hu, ACS Nano, 2016, 10, 8851–8860 CrossRef CAS .
47. U. Yorulmaz, A. Özden, N. K. Perkgöz, F. Ay and C. Sevik, Nanotechnology, 2016, 27, 335702 CrossRef PubMed .
48. H. Kim, B. Anasori, Y. Gogotsi and H. N. Alshareef, Chem. Mater., 2017, 29, 6472–6479 CrossRef CAS .
49. V. Kamysbayev, A. S. Filatov, H. Hu, X. Rui, F. Lagunas, D. Wang, R. F. Klie and D. V. Talapin, Science, 2020, 369, 979–983 CrossRef CAS PubMed .
50. V. Natu and M. W. Barsoum, J. Phys. Chem. C, 2023, 127, 20197–20206 CrossRef CAS .
51. M. Anayee, C. E. Shuck, M. Shekhirev, A. Goad, R. Wang and Y. Gogotsi, Chem. Mater., 2022, 34, 9589–9600 CrossRef CAS .
52. S. Wei, P. Zhang, W. Xu, S. Chen, Y. Xia, Y. Cao, K. Zhu, Q. Cui, W. Wen, C. Wu, C. Wang and L. Song, J. Am. Chem. Soc., 2023, 145, 10681–10690 CrossRef CAS .
53. H. Ding, Y. Li, M. Li, K. Chen, K. Liang, G. Chen, J. Lu, J. Palisaitis, P. O. Persson and P. Eklund, Science, 2023, 379, 1130–1135 CrossRef CAS PubMed .
54. J. M. McEnaney, A. R. Singh, J. A. Schwalbe, J. Kibsgaard, J. C. Lin, M. Cargnello, T. F. Jaramillo and J. K. Nørskov, Energy Environ. Sci., 2017, 10, 1621–1630 RSC .
55. F. Meng, J. Qin, X. Xiong, H. Zhang, M. Zhu and R. Hu, Cell Rep. Phys. Sci., 2023, 4, 101307 CrossRef CAS .
56. Y. Tang, C. Yang, X. Xu, Y. Kang, J. Henzie, W. Que and Y. Yamauchi, Adv. Energy Mater., 2022, 12, 2103867 CrossRef CAS .
57. Y. Tang, C. Yang, Y. Tian, Y. Luo, X. Yin and W. Que, Nanoscale Adv., 2020, 2, 1187–1194 RSC .
58. S. Oswald, Encyclopedia of Analytical Chemistry: Applications, Theory and Instrumentation, 2006 Search PubMed .
59. F. Ando, T. Gunji, T. Tanabe, I. Fukano, H. D. Abruna, J. Wu, T. Ohsaka and F. Matsumoto, ACS Catal., 2021, 11, 9317–9332 CrossRef CAS .
60. J. Suntivich, K. J. May, H. A. Gasteiger, J. B. Goodenough and Y. Shao-Horn, Science, 2011, 334, 1383–1385 CrossRef CAS PubMed .
61. V. Giulimondi, S. Mitchell and J. Pérez-Ramírez, ACS Catal., 2023, 13, 2981–2997 CrossRef CAS PubMed .

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta03999h

‡ These authors contributed equally to the work.

This journal is © The Royal Society of Chemistry 2024