Citation

Lu Yanhui, Pei Chengang, Li Wenqiang, Liu Qing, Pang Huan, Yu Xu. Tailoring active sites of cerium and nitrogen Co-doped rhenium disulfide for enhanced hydrogen evolution reaction[J]. Chin. Chem. Lett., 2025, 36(12): 111646  

Yanhui Lu^a, Chengang Pei^a, Wenqiang Li^c, Qing Liu^d, Huan Pang^a,b,*, Xu Yu^a,*     

Hydrogen, recognized as a carbon-free energy carrier, is poised to occupy a pivotal role in the future energy landscape, and economic production of hydrogen fuel remains a long-standing objective in hydrogen research [1,2]. The most common electrocatalysts for hydrogen evolution reaction (HER) are noble metals-based materials (Pt, Ru, etc.) [3–5]. However, resource scarcity and high costs are still the primary issues to be addressed. Innovative and sustainable catalysts are essential for facilitating efficient hydrogen production while reducing dependence on precious metals [6].

Transition metal dichalcogenides (TMDCs) present promising candidates as HER electrocatalysts due to their relative abundance, reasonable cost, and favorable electrocatalytic activity [7]. Rhenium disulfide (ReS₂) has emerged as an efficient HER catalyst owing to its high chemical stability and distinctive layered structure, which can expose more active edge sites. This unique morphology facilitates the exposure of abundant active sites for hydrogen production. Despite extensive efforts to develop layered ReS₂-based catalysts, the low intrinsic catalytic activity of ReS₂ continues to constrain its HER performance [8,9].

Numerous effective strategies have been proposed to enhance catalytic activity, including element doping [10], defect introduction [11], construction of hierarchical structures [12–14], and regulation of surface interfaces [15]. The element dopants can adjust the electronic structures to improve the catalytic effect. Heteroatoms (N, P, and S) can balance the adsorption and desorption of intermediates by redistributing atomic charge [16,17]. Heteroatom N with high electronegativity can boost the catalytic performance by improving electron donor characteristics and conductivity [18]. The edges of the TMDC layer exhibit inherent electrocatalytic activity, and metallic doping creates sulfur vacancies on layered facets, enhancing electrical conductivity and increasing the number of active sites [19]. Furthermore, metallic doping induces unbalanced Coulomb forces, resulting in atomic reshuffling to generate active edges [20]. Thus, doping metals into TMDC layers enhances fast electron transfer and promotes HER. Although the basal plane is less active than edge sites, metal doping can activate these surfaces, significantly increasing the density of catalytically active sites, particularly through defect engineering on the basal planes [21]. Metallic Ce element as an activator can modify the electronic structure, and the accessible electronic communication of Ce³⁺/Ce⁴⁺ promotes electron transfer and significantly reduces the H₂O dissociation energy barrier [22].

Herein, the metallic Ce and heteroatom N dual-doped ReS₂ are vertically grown on carbon fiber paper (Ce, N-ReS₂) via a facile hydrothermal method. The inhibition of restacking phenomena results in more exposed active edge sites, and elemental incorporation modulates the surface electronic structure and the defective edge sites of ReS₂. Electrochemical results reveal that a middle Ce content doped Ce, N-ReS₂ (Ce (M),N-ReS₂) shows exceptional electrocatalytic HER activities of 139 mV in acidic and 130 mV in alkaline solutions at the high current density of 100 mA/cm², as well as superior catalytic stability.

During the hydrothermal process, Ce, N-ReS₂ is synthesized with the hydroxylamine hydrochloride and cerium nitrate as N and Ce doping precursors, respectively. During the hydrothermal process, the hydroxylamine hydrochloride was thermally decomposed, and the formed nitrogenous species (NH₃/NH₂^-/NO^-) can occupy sulfur sites in the ReS₂ lattice, leading to the formation of an N-doped structure [23]. The catalysts by introducing low and high amounts of cerium nitrate denoted as Ce (L),N-ReS₂ and Ce (H),N-ReS₂. X-ray diffraction (XRD) was utilized to assess the phase composition of Ce, N-ReS₂. Fig. 1a displays diffraction peaks corresponding to the ReS₂ phase (PDF #89–0341), and no observable peaks for cerium species indicate that no new crystalline phase is formed by Ce, N dual-doping. The magnified (002) facet diffraction peak of Ce, N-ReS₂ shows a negative shift compared to ReS₂, attributing to the lattice expansion resulting from the different atomic radius of Ce and N [24]. It is noteworthy that there are intense diffraction peaks at 26.5° and 54.9°, which belong to the (002) and (004) planes of the carbon fiber paper substrate [25].

Figure 1

	Download: JPG larger image
Fig. 1. (a) XRD, (b, c) SEM, (d-f) TEM and HRTEM, and (g) corresponding elemental mapping images of Ce (M),N-ReS2.

The structural characteristics of catalysts were evaluated using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Figs. 1b and c, Ce (L),N-ReS_2, and Ce (M),N-ReS₂ exhibits uniform interweaving (Figs. S1-S3 in Supporting information). Whereas Ce (H),N-ReS₂ displays the agglomeration, potentially diminishing the active surface area and deactivating the catalyst due to excessive interactions between Ce ions and ReS₂. The TEM image of Ce (M),N-ReS₂ nanosheet demonstrates an optimal architecture for mass transfer and active site exposure (Fig. 1d). As confirmed by reported lectures [26,27], the atomic doping strategy can effectively expand the interlayer spacing and lattice structure catalyst, which is favorable to expose the active sites and facilitate the electron transfer. High-resolution TEM images show lattice spacings of 0.66, 0.73, and 0.80 nm corresponding to the (002) planes of ReS₂ in Figs. 1e and f, and the dashed circles indicate lattice defects and layer spacing expansion. Curved lattice stripes further suggest that Ce and N doping induces lattice distortion, increases basal defect, and exposes electrocatalytic active sites, thus enhancing electrocatalytic performance. The defective sites and electron transfer of the catalyst were confirmed by electron paramagnetic resonance (EPR) spectra in Fig. S4 (Supporting information). The presence of sulfur vacancies is conclusively evidenced by the EPR signal at g ≈ 2.003 [28,29]. Ce (M),N-ReS₂ shows a broader peak with a higher peak intensity than that of pristine ReS₂, implying the enhanced lattice defects and more unpaired electrons by elemental dual-doping [30]. The redistribution of electrons induced by spin and electronegativity effectively mitigates the significant loss of bridging or apical sulfur atoms, collectively enhancing the HER performance [31]. Energy dispersive spectroscopy (EDS) elemental mapping of Ce (M),N-ReS₂ (Fig. 1g) reveals a homogeneous distribution of Ce, Re, S, and N, confirming the successful heteroatom doping. Moreover, the EDS spectrum also confirms the uniform distribution of Re, Ce, S, and N elements for Ce(M),N-ReS₂ (Fig. S5 in Supporting information).

The surface chemistry of Ce (M),N-ReS₂ was studied by X-ray photoelectron spectroscopy (XPS) in Fig. S6 (Supporting information), and the atomic ratio of Ce and N for Ce (M),N-ReS₂ is 3.26% and 3.16% in Table S1 (Supporting information). As further confirmed by Inductively Coupled Plasma in Table S2 (Supporting information), the ratio of Re/Ce is matched with the XPS result. For Re 4f spectra in Fig. 2a, the peaks at 44.2 and 41.8 eV in Ce (M),N-ReS₂ correspond to Re 4f_7/2 and Re 4f_5/2. In comparison, Ce (L),N-ReS₂ shows a positive shift of Re 4f_7/2 peak, indicating a reduction in electron cloud density around Re atoms. Meanwhile, a negative shift of Re 4f_7/2 peak can be found for Ce (H),N-ReS₂, attributing to valence electron transfer from the weakly electronegative Ce atom to Re. For Ce 3d spectra of Ce (L),N-ReS₂ in Fig. 2b, the deconvoluted peaks correspond to Ce³⁺/Ce⁴⁺ of 3d_3/2 and 3d_5/2 [22]. In comparison, the Ce³⁺ 3d_5/2 (887.6 eV) peak of Ce (M),N-ReS₂ shifts to a low binding energy, indicating a strong electronic interaction with Re species. A larger negative shift and a reduced proportion of Ce³⁺ 3d_5/2 for Ce (H),N-ReS₂ possibly result from the excessive Ce enhancing electron interactions to facilitate the transition to Ce⁴⁺ [32].

Figure 2

	Download: JPG larger image
Fig. 2. XPS spectra for (a) Re 4f, (b) Ce 3d, (c) S 2p, and (d) N 1s.

The divided S 2p spectra for Ce (L),N-ReS₂ consists of two main characteristic peaks of S 2p_1/2 and S 2p_3/2 in Fig. 2c, corresponding to the characteristic peaks of Re-S bonding [12]. The positive shift of S 2p_1/2 and S 2p_3/2 for Ce (M),N-ReS₂ may be ascribed to the modulated electron cloud density around S species and the formation of heterostructure interfaces. N 1s spectra are divided into two peaks of metal-N and N—H for Ce, N-ReS₂ (Fig. 2d), unveiling the successful N doping substitution of S [33]. The electronegativity of N affects the binding energies in Ce, N-ReS₂. The XPS analysis proves that the optimal Ce, N dual doping can modulate electronic interactions at the ReS₂ interface and enhance the adsorption and dissociation energies of intermediates, thereby improving the electrocatalytic activity of Ce (M),N-ReS₂ [22,32].

To explore the enhanced effects of Ce on the electrocatalytic performance of ReS₂, the catalytic HER performance of Ce, N-ReS₂ was investigated in a standard three-electrode system in 1 mol/L KOH solution. Fig. 3a and Fig. S7 (Supporting information) show the linear sweep voltammetry (LSV) curves of all catalysts after IR-correction. Ce (M),N-ReS₂ requires a low overpotential of 44 mV at 10 mA/cm², which is much lower than Ce (L),N-ReS₂ (88 mV), Ce (H),N-ReS₂ (97 mV), Ce-ReS₂ (115 mV),N-ReS₂ (121 mV), and ReS₂ (136 mV), respectively. Moreover, the Ce, N-ReS₂ exhibits a higher current response during the entire potential range compared with pristine ReS₂ (130 mV@100 mA/cm²), certifying the enhanced HER activity by Ce and N dual-doping (Fig. S8 in Supporting information). To explore the HER kinetics, Tafel plots were calculated by integrating the LSV curves in Fig. 3b. The value of Tafel slope for Ce (M),N-ReS₂ (90.6 mV/dec) is much lower than that of Ce (L),N-ReS₂ (104.3 mV/dec), Ce (H),N-ReS₂ (119.3 mV/dec) and ReS₂ (160.0 mV/dec), implying much faster HER kinetics on Ce (M),N-ReS₂ surface. The Volmer-Heyrovsky step is the dominant rate-determining step during the hydrogen desorption process.

Figure 3

Download: JPG larger image

Fig. 3. Electrocatalytic HER activity in 1 mol/L KOH. (a) LSV curves, (b) Tafel slopes and (c) EIS of all catalysts at an overpotential of 44 mV. (d) LSV curves before and after 1000 cycles (inset: Long-life durability of Ce (M),N-ReS2 at different current densities). Electrocatalytic HER activity in 0.5 mol/L H2SO4. (e) LSV curves, (f) Tafel slopes and (g) EIS of all catalysts at an overpotential of 79 mV. (h) Catalytic HER performance compared with other reported transition metal-based catalysts.

As verified by electrochemical impedance spectroscopy (EIS) in Fig. 3c and Fig. S9 (Supporting information), Ce (M),N-ReS₂ conducts the lowest charge-transfer resistance (R_ct) compared with other electrocatalysts (Table S3 in Supporting information), implying the fast reaction kinetics. The double-layer capacitance (C_dl) for Ce (M),N-ReS₂ (92.0 mF/cm²), is about two times larger than that of ReS₂ in Fig. S10 (Supporting information). Meanwhile, Ce (M),N-ReS₂ owns the largest electrochemical surface area (ECSA) in Fig. S11 (Supporting information), indicating more exposed surface active sites. LSVs of normalizing the ECSA (Fig. S12 in Supporting information) demonstrate the Ce (M),N-ReS₂ further proving a better intrinsic activity for HER. The coordinative-unsaturated S-edge sites can serve as active centers for HER, and the introduction of defects into basal planes can create active sites and improve activity [34]. The vertically aligned morphology of Ce (M),N-ReS₂ can expose abundant defects and more S-edge active sites, and the increased electronic conductivity of ReS₂ via Ce/N co-doping can provide the unsaturated electron. Moreover, the construction of Ce-S-Re interaction leads to high instinct activity and more accessible active sites of ReS₂ for HER [12,24].

Temperature-dependence of Ce (M),N-ReS₂ on the catalytic activity was evaluated in Fig. S13 (Supporting information). The apparent electrochemical activation energy (E_a) for hydrogen production was determined using the Arrhenius relationship. As shown in Fig. S14 (Supporting information), the slope of the Arrhenius plot (exchange current density, j₀ vs. temperature, 1/T) indicates that Ce (M),N-ReS₂ possesses the lowest energy barrier, reflecting a significant enhancement in its intrinsic catalytic activity. This enhancement is attributed to cerium doping to adjust the adsorption energy of transition state intermediates and reduce the kinetic energy barrier during the electrochemical process, thereby improving electrocatalytic performance. As shown in Fig. 3d, a negligible overpotential loss after 1000 CV cycles and well-remained current density after chronoamperometry test (CA) for 50 h reveal excellent catalytic durability of the Ce (M),N-ReS₂. Furthermore, the electrocatalytic selectivity is also a critical characteristic. As illustrated in Fig. S15 (Supporting information), Ce (M),N-ReS₂ achieves approximately 100% hydrogen yield in the Faradaic efficiency test, suggesting its high selectivity for the HER.

To further verify the effect of Ce doping on the improved electrocatalytic HER performance, the LSV curves were tested in 0.5 mol/L H₂SO₄ (Fig. 3e and Fig. S16 in Supporting information), and Ce (M),N-ReS₂ exhibits a superior HER activity. At 10 mA/cm², Ce (L),N-ReS₂, Ce (M),N-ReS₂, Ce (H),N-ReS₂, Ce-ReS₂, N-ReS₂, and ReS₂ catalysts show the overpotentials of 92, 79, 115, 136, 138, and 155 mV, respectively. As the current density of 50 and 100 mA/cm² shown in Fig. S17 (Supporting information), Ce (M),N-ReS₂ still shows a smaller overpotential than other control samples (139 mV@100 mA/cm²). The enhanced HER activity of Ce (M),N-ReS₂ is attributed to the co-doping of Ce and N, which efficiently boosts electron transfer, alongside its unique layered morphology that provides abundant reaction sites. Tafel slopes of all catalysts indicate a lower Tafel slope for Ce (M),N-ReS₂ (64.2 mV/dec) than Ce (L),N-ReS₂, Ce (H),N-ReS₂ and ReS₂ (Fig. 3f), suggesting that Ce and N doping can accelerate the catalytic kinetics. Ce (M),N-ReS₂ has the lowest R_ct value (Fig. 3g and Fig. S18 in Supporting information), unveiling its superior electron transfer capability. ECSA was estimated through CV test in Figs. S19 and S20 (Supporting information), and Ce (M),N-ReS₂ owns the highest C_dl value of 113.6 mF/cm², implying the largest ECSA. The ECSA-normalized curves of Ce, N-ReS₂ demonstrate its high intrinsic activity in Fig. S21 (Supporting information). As confirmed by CV for 1000 cycles and CA test for 50 h, Ce (M),N-ReS₂ exhibits exceptional catalytic HER stability in Fig. S22 (Supporting information). Additionally, Ce (M),N-ReS₂ owns exceptional HER activity compared to other reported transition metal-based catalysts in both acidic and alkaline solutions (Fig. 3h, Tables S4 and S5 in Supporting information).

To explore the change of morphology and crystal structure of the Ce (M),N-ReS₂, SEM, and TEM were conducted after the CA test in 1 mol/L KOH solution. In comparison, no apparent change of morphology can be observed in Fig. S23 (Supporting information), confirming its excellent structural stability. Fig. S24 (Supporting information) shows the elemental mapping images of Ce (M),N-ReS_2, and all elementals are uniformly distributed. The change of surface chemistry for Ce (M),N-ReS₂ after the CA test was probed by XPS in (Figs. 4a and b, Fig. S25 in Supporting information). In comparison to its initial state, no significant peak shift for Ce (M),N-ReS₂ can be observed in the high-resolution Ce 3d, N 1s, and Re 4f spectra after the catalytic stability test, and the S 2p spectra exhibit negligible oxidation following the HER test.

Figure 4

	Download: JPG larger image
Fig. 4. (a) Re 4f, (b) Ce 3d spectra of Ce (M),N-ReS2 catalyst after stability test. (c) The simulated calculation models. (d) The reaction energy for water dissociation. (e) Gibbs free energy diagrams for HER on different catalyst surfaces.

To deeply reveal the influence of Ce and/or N dopants on the HER catalytic performance of the ReS₂ catalyst, density functional theory (DFT) calculations were performed to analyze the electronic structures of ReS₂, Ce-ReS₂, N-ReS₂, and Ce, N-ReS₂, as well as the possible HER mechanism. According to experimental results, the ReS₂ (002) surface was chosen for the calculations (Fig. 4c and Fig. S26 in Supporting information). Since the catalytic process mainly occurs on the surface of the catalyst, models for the Ce-ReS₂, N-ReS₂, and Ce, N-ReS₂ systems were constructed by substituting one Re and/or S atom in the first layers of the ReS₂ (002) surface with a Ce and/or N atom.

The synergistic activation of electronic interactions and charge redistribution reduces the energy barriers by optimizing intermediate adsorption/desorption behavior. A low water dissociation energy indicates an easy disconnection of the O–H bonds in the H₂O molecule. Gibbs free energies of H* adsorption (ΔG_H*) is an important parameter for evaluating the HER activity, and a value close to 0 eV indicates a low H* adsorption/desorption energy barrier for HER [35]. As shown in Fig. 4d and Fig. S27 (Supporting information), Ce (M),N-ReS₂ owns the weaker water adsorption free energy (0.07 eV) compared to Ce-ReS₂ (−0.39 eV),N-ReS₂ (−0.45 eV), and ReS₂ (−0.52 eV), suggesting an easy activation of water molecules for Ce (M),N-ReS₂. Meanwhile, the relative energy in the Volmer step (from *H₂O to *[OH···H] is calculated to be 0.7 eV for the Ce (M),N-ReS₂, which is much lower than other control samples, indicating its small energy barrier to split the first water molecule [36,37]. These calculations illustrate that both Ce and N play a beneficial role in promoting the H₂O adsorption and H₂O dissociation step to accelerate the catalytic HER kinetics. The optimal adsorption sites and ΔG_H* diagram for intermediates were determined through calculations, and the optimized structure of hydrogen adsorption on Ce, N-ReS₂ is depicted in Fig. 4e and Fig. S28 (Supporting information), the calculated pure ReS₂ has large ΔG_H* values (−0.17 eV), which indicates strong H* adsorption that is unfavorable for H* desorption in H₂ evolution. With the doping of Ce or N, the ΔG_H* for Ce-ReS₂ (ΔG_H* = −0.12 eV), and N-ReS₂ (ΔG_H* = −0.15 eV) decreased, indicating that the introduction of Ce or N can increase the proton affinity of the substrate and promote the Heyrovsky step. Remarkably, with the Ce and N dual-doped ReS₂, the ΔG_H* value decreased to −0.1 eV, reflecting that charge redistributed Ce, N-ReS₂ weaken H* desorption energy barriers to improve HER kinetics, which aligns with experimental results.

In summary, Ce and N co-doped ReS₂ electrocatalysts are in-situ prepared using a hydrothermal approach, and the effect of Ce doping content on the electrocatalytic HER performance is compared. Ce (M),N-ReS₂ with the optimized Ce doping content shows excellent catalytic HER activity in both alkaline and acidic electrolytes. The HER values are 44 and 79 mV at 10 mA/cm², and the performance remains good at 131 and 139 mV at 100 mA/cm². This excellent catalytic activity is attributed to the synergistic effect of metallic Ce and non-metallic N co-doping. The aligned ReS₂ nanosheets guarantee the exposure of defective edge sites, and the adjusted surface chemistry can induce the lattice strain to boost the catalytic active sites. The low Tafel slope and charge transfer resistance confirm the fast reaction dynamics during the HER process, and the comparison of morphology and surface chemistry after the CA test confirms that Ce (M),N-ReS₂ shows excellent electrocatalytic stability. Experimental and DFT calculations have been used to demonstrate that the adsorption energy of intermediates containing hydrogen can be optimized by altering the electronic structure of the ReS₂ active site through the introduction of Ce and N, thereby promoting the HER process. This work provides an efficient strategy to prepare the layer-structured catalysts for advanced hydrogen production.

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.

Yanhui Lu: Writing – original draft, Investigation, Formal analysis, Data curation. Chengang Pei: Formal analysis. Wenqiang Li: Software. Qing Liu: Formal analysis. Huan Pang: Writing – review & editing. Xu Yu: Writing – review & editing, Supervision, Conceptualization.

The work is supported by the Six Talent Peaks Project of Jiangsu Province (No. XCL-103), the Postgraduate Research & Practice Innovation Program of Jiangsu Province (Yangzhou University, No. SJCX22_1727) and the 'High-End Talent Project' of Yangzhou University. We also acknowledge the technical support at the Testing Center of Yangzhou University.

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111646.