(ReMoV)X2 (X = S, Se) Ternary Alloy Nanosheets for Enhanced Electrocatalytic Hydrogen Evolution Reaction

Abstract

Modulating the electronic structure of 2D transition metal dichalcogenides via alloying can extend their potential applications. In this study, composition-tuned ternary alloy nanosheets of (ReMoV)X₂ (X = S and Se) are synthesized using solvothermal and colloidal reactions, respectively. Ternary alloying occurred with homogeneous atomic mixing over a wide range of compositions (x_V = 0.16-0.80). Compared to (ReV)X₂ binary alloying, ternary alloying produces a more metallic phase with less oxidation. Increasing x_V induces a phase change into a more metallic 1T phase. The (ReMoV)S₂ nanosheets demonstrate enhanced electrocatalytic activity toward the acidic hydrogen evolution reaction (HER) compared to (ReV)S₂. Density functional theory calculations predict that ternary alloying increases the metallicity of the nanosheets. In addition, the Gibbs free energy calculation for hydrogen adsorption (ΔG_H*) shows that ternary alloying effectively activates the basal S atoms for the HER, supporting the enhanced catalytic performance observed experimentally.

Keywords: ReMoV ternary alloys; first‐principles calculations; hydrogen evolution reaction; metallic nanosheets; transition metal dichalcogenide.

Scheme 1

a) Schematic diagram for the synthesis of (ReMoV)S₂ and (ReMoV)Se₂ alloy nanosheets, respectively, using solvothermal and colloidal reactions. Ternary plot for the composition (x _Re, x _Mo, x _V) of b) (ReMoV)S₂ and c) (ReMoV)Se₂ alloy nanosheets synthesized in the present work. For (ReMoV)S₂, 1S: (0.42, 0.42, 0.16), 2S: (0.25, 0.5, 0.25), 3S: (0.50, 0.25, 0.25), 4S: (0.33, 0.33, 0.33), 5S: (0.16, 0.42, 0.42), 6S: (0.42, 0.16, 0.42), 7S: (0.25, 0.25, 0.5), 8S: (0.167, 0.167, 0.67), 9S: (0.125, 0.125, 0.75), and 10S: (0.10, 0.10, 0.80). For (ReMoV)Se₂, 1Se: (0.25, 0.5, 0.25), 2Se: (0.50, 0.25, 0.25), 3Se: (0.33, 0.33, 0.33), 4Se: (0.25, 0.25, 0.5), 5Se: (0.4, 0.1, 0.5), 6Se: (0.3, 0.1, 0.6), 7Se: (0.167, 0.167, 0.67), and 8Se: (0.125,0.125, 0.75).

Figure 1

a) SEM and HRTEM images of Re_0.33Mo_0.33V_0.33S₂ (sulfide sample 4S) and Re_0.33Mo_0.33V_0.33Se₂ (selenide sample 3Se). b) HAADF STEM image and EDX elemental mapping of Re (M shell), Mo (L shell), V (K shell), and S (K shell) or Se (L shell) for 4S and 3Se. Insets: EDX spectra showing composition tuning in the 10 samples of (ReMoV)S₂ and 8 samples of (ReMoV)Se₂. For sulfides, the inset corresponds to the enlarged edge part of overlapped Mo L/S K shell, showing the composition tuning of Mo. c) Atomic‐resolution HAADF STEM images and corresponding FFT images for 1S (x _V = 0.16), 4S (x _V = 0.33), 8S (x _V = 0.67), 3Se (x _V = 0.33), and 7Se (x _V = 0.67). The [01$\overline{1}$0] spots (FFT) matched to d ₀₁₀ = 0.28 nm for sulfide and 0.29 nm for selenide. Darker regions of VS₂ become dominant as x _V increases. Line profile along the marked area, showing the random distribution of Re, Mo, and V atoms that were identified by comparing the intensity of adjacent atoms.

Figure 2

a) Left: full‐range XRD patterns of ReS₂, MoS₂, VS₂, and (ReMoV)S₂ samples, together with the reference peaks generated using the calculated lattice constants of 1T′′ ReS₂, 1T′ MoS₂, 3R‐1T VS₂, and Re_0.31Mo_0.38V_0.31S₂ (for sample 4S). Right: magnified XRD peak at 2θ = 54°–62° for (2 $\overline{4}$ 1)_1T′′ of ReS₂, ($\overline{2}$ 05)_1T′ of MoS₂, (110)_3R‐1T/(113)_3R‐1T of VS₂, and (110)_1T of 4S. b) Crystal structures (ball‐and‐stick models) of the two‐layered (4 × 4 × 1) supercell and relative energy (E_rel) for 2H and 1T‐like phase of Re_0.31Mo_0.38V_0.31S₂, projected along the basal plane (c axis, top) and b‐axis (bottom). Gray, violet, red, and yellow balls represent Re, Mo, V, and S atoms, respectively. c) Left: full‐range XRD patterns of ReSe₂, MoSe₂, VSe₂, and (ReMoV)Se₂ samples, together with the reference peaks calculated using the lattice constants of 1T′′ ReSe₂, 2H MoSe₂, and 1T VSe₂. Right: magnified XRD peak at 2θ = 52°–60°.

Figure 3

a) Fine‐scan XPS peaks of Re 4f (4f _7/2 and 4f _5/2 separated by 2.43 eV), Mo 3d (3d _5/2 and 3d _3/2 separated by 3.13 eV), V 2p (2p _3/2 and 2p _1/2 separated by 7.64 eV), and VBS for ReS₂, MoS₂, VS₂, and sulfide samples (1S–10S). Positions of neutral Re⁰ (4f _7/2 at 40.3 eV), Mo⁰ (3d _5/2 at 228.0 eV), and V⁰ (2p _3/2 at 512.2 eV) are marked by dotted vertical lines. The experimental data (open circles) are fitted by a Voigt function after a Shirley‐type baseline correction. The sum of the resolved bands is represented by a black line. In VBS, the Fermi level (E_F) is marked by a dotted vertical line. The d‐band center (ɛ_d) value is also labeled with a dotted line arrow to emphasize the redshift with increasing x _V. b) XANES and (non‐phase‐corrected) k ²‐weighted FT EXAFS above the Re L₃‐edge (10.53 keV), Mo K‐edge (20.01 keV), and V K‐edge (5.46 keV), and their corresponding WT EXAFS (k‐space vs R‐space) for sulfide samples 1S, 4S, 7S, 8S, and 10S.

Figure 4

a) LSV curves (scan rate: 2 mV s⁻¹) versus RHE for (ReMoV)S₂ samples, and Pt/C toward HER in H₂‐saturated 0.5 m H₂SO₄. The values in parenthesis correspond to η _{J = 10}. b) Tafel plots (η vs log J) derived from the LSV curves, based on the equation η = b log(J/J ₀), where b is the Tafel slope, and J ₀ is the exchange current density (extrapolated value at η = 0). Linear fit provides the b values (in parenthesis). c) CA responses of x _V = 0.67 (sample 8S) and Pt/C at η_{J = 10} (104 mV) for 5 days and d) comparison of 1st and 2000th cycled LSV curves. e) η_{J = 10} versus x _V for (ReMo)_{1‐ x} V _x S₂ (sphere symbols) and Re_{1‐ x} V _x S₂ samples (columns, ref. [20]). f) LSV curves and g) Tafel plots for (ReMoV)Se₂ samples. The corresponding η_{J = 10} and b values are listed in the parentheses. h) η_{J = 10} versus x _V for (ReMo)_{1‐ x} V _x Se₂ (sphere symbols), Re_{1‐ x} V _x Se₂ ref. [19]/Mo_{1‐ x} V _x Se₂ ref. [14] samples (columns). i) Ex situ XANES above the Re L₃‐edge (10.53 keV), Mo K‐edge (20.01 keV), and V K‐edge (5.46 keV) for x _V = 0.67 (sample 8S) after applying an overpotential (𝜂 = 0–100 mV vs RHE) under HER conditions (H₂‐saturated 0.5 m H₂SO₄).

Figure 5

a) TDOS near the Fermi level for 2H and 1T‐like phases of Re_0.31Mo_0.38V_0.31S₂, Re_0.75V_0.25S₂, Re_0.5V_0.5S₂, and Re_0.25V_0.75S₂ ref. [20]. The Fermi level (black line) is set to zero. b) Geometry of H adsorption on Re_0.31Mo_0.38V_0.31S₂. The ΔG_H* value is given for each site. Gray, violet, red, and yellow balls represent Re, Mo, V, and S atoms, respectively. The small blue ball represents the adsorbed H atom. c) Histograms representing ΔG_H* for various S sites of Re_{1‐ x} V _x S₂ ref. [20] and Re_0.31Mo_0.38V_0.31S₂ models.