The environmental pollution and energy crisis are common challenges that human beings are facing. Moreover, the greenhouse effect resulting from the use of fossil fuels is an important problem in the development of society. Creating technology that converts carbon dioxide into valuable products is an attractive way to reduce the greenhouse effect [1,2]. More importantly, it is necessary to develop renewable energy to entirely solve the energy problem. Hydrogen energy is recognized as one of the ideal secondary energy sources due to its non-toxicity, non-polluting nature, high energy density, and renewability [3]. However, efficient and safe hydrogen storage technology still sets a strict limit on the development of the utilization of hydrogen energy [4]. Magnesium and its hydride are recognized as the optimal solid hydrogen storage material due to their high storage density of up to 7.6 wt%, good reversibility, and low cost [5]. However, the high decomposition temperature and slow reaction kinetics of Mg/MgH₂ significantly limit its practical applications [6].

To address these challenges, various methods have been explored to improve the hydrogen storage performance of magnesium-based materials, such as alloying with other elements [7,8,9], doping catalysts [10,11,12], nanosizing [13,14,15], and so on. In materials science, it is known that symmetry plays a key role in hydrogen absorption and desorption in alloys, as it affects structural stability, atomic arrangements, and phase transitions. The alloying method consists of adding other elements to magnesium by alloy smelting to obtain alloys, thus enhancing the hydrogen storage performance of magnesium. Xie et al. [16] prepared Mg-Ce alloys with various Ce contents, and the phase compositions, micro-structures, and hydrogen storage performance of the alloys were investigated. They found that the addition of Ce significantly improves the hydrogen absorption kinetic properties of the alloy, and the dehydrogenation temperature can be obviously reduced with the increase of Ce content. Yang et al. [17] prepared several Mg₂₄Y₃M (M = Ni, Co, Cu, Al) ternary alloys, and the research results indicate that the alloy with different chemical components showed significantly different features in its microstructure and hydrogen storage properties. Mg₂₄Y₃Ni had the best hydrogen absorption and desorption kinetic performance. Doping catalysts is also an effective method to improve the hydrogen storage kinetic properties of Mg/MgH₂ [18]. Currently, the catalysts used for hydrogen storage materials mainly include metals and alloys [19,20], oxides [21,22], halides [23,24], carbide [11], and carbonaceous materials [25]. The addition of transition metal elements or their inorganic compounds has proven to be an effective method for enhancing the hydrogen storage performance of magnesium-based materials. Oelerich et al. [26] doped different transition metallic oxides (Sc₂O₃, TiO₂, V₂O₅, Cr₂O₃, Mn₂O₃, Fe₃O₄, CuO/Al₂O₃, and SiO₂) into MgH₂ by a simple ball-milling method, and the hydrogen storage measurement results indicate that all these oxide catalysts can improve the hydrogen absorption and desorption performance of MgH₂. Li et al. [27] synthesized a nanocrystalline Ni@C composite using a self-templating method and then doped it into MgH₂ to investigate its catalytic impact on the hydrogen storage properties of MgH₂. The research results demonstrate that the dehydrogenation temperature of the composite was lowered by more than 74 °C compared to pure MgH₂.

In recent years, metal sulfides have aroused more attention due to their outstanding catalytic properties the in water–electrolytic hydrogen evolution and photocatalytic fields [28,29]. Moreover, metal sulfides also have unique catalytic effects in improving the hydrogen storage performance of magnesium-based materials. Yuan et al. [30,31,32] compared the catalytic effects of CoS₂ and MoS₂ nano-particles on the hydrogen storage properties of Sm–Mg binary alloy. They found that both of the sulfides could obviously improve the hydrogenation and dehydrogenation kinetics of Sm–Mg alloy, and the composite catalyzed by MoS₂ exhibits better hydrogen storage properties. Wang et al. [33] also synthesized NiS₂ particles by a one-step hydrothermal method, and various MgH₂–NiS₂ composites with different NiS₂ contents were prepared by mechanical ball-milling. The study indicates that the NiS₂ decomposed and transformed into MgS and Mg₂NiH₄ phases during the hydrogenation process, and the catalytic phases formed in situ significantly improved the hydrogenation and dehydrogenation kinetic properties of MgH₂. Based on previous studies, a Mg₉₅Ce₅ alloy was prepared by using the vacuum induction melting method in this study. Then TiS₂ was selected as a catalyst to dope into Mg₉₅Ce₅ alloy by using the mechanical ball-milling method to prepare a series of Mg₉₅Ce₅–x wt% TiS₂ (x = 0, 3, 5 and 10) composites with different TiS₂ contents. The effects of TiS₂ content on microstructure and hydrogen storage properties of Mg₉₅Ce₅–TiS₂ composites were studied, and the catalytic mechanism of the TiS₂ on hydrogen absorption and desorption reactions of Mg₉₅Ce₅ alloy was investigated in detail.

The phase compositions and microtopography features of the TiS₂ catalyst and Mg₉₅Ce₅ alloy were characterized using XRD and SEM analysis, and the results are illustrated in Figure 1. It can be observed in Figure 1a that the TiS₂ sample exhibits distinct diffraction peaks, indicating it is a typical crystalline structure. Figure 1b presents the SEM image of TiS₂ powder. It can be seen that the TiS₂ particles exhibit a polygonal flake-like structure and particle size of 5~20 μm. The XRD analysis results in Figure 1a indicate that the Mg₉₅Ce₅ alloy is composed of CeMg₁₂ and Mg phases, and the sharp diffraction peaks imply the alloy has good crystallinity. Figure 1c shows the SEM image (backscattered electron image) of the Mg₉₅Ce₅ alloy ingot. It is evident that the alloy mainly consists of two structures: the one is a coarse and gray-white blocky phase (marked A), and the other is a eutectic structure composed of a striped or globular gray-white phase and a black phase (marked B). In general, elements with bigger atomic numbers appear brighter in backscattered electron images. Combined with XRD analysis and a Mg–Ce binary phase diagram [34], we can conclude that the gray-white phase in areas A and B is CeMg₁₂, and the black phase in area B is Mg.

Mg₉₅Ce₅–TiS₂ composites with different TiS₂ content were prepared by the mechanical ball-milling method, and the XRD patterns of the composites are shown in Figure 2. It can be observed that the ball-milling process has almost no influence on the phase compositions of the alloy, and the ball-milled Mg₉₅Ce₅ alloy still consists of CeMg₁₂ and Mg phases. However, the diffraction peaks of the ball-milled Mg₉₅Ce₅ alloy are broadened and weakened, which can be attribute to the lattice distortion and grain refinement resulted from the ball-milling process [35]. Moreover, Figure 2 also illustrates that the TiS₂ content has little effect on the phase composition of Mg₉₅Ce₅–TiS₂ composites. Meanwhile, the diffraction peaks of TiS₂ cannot be observed in the Mg₉₅Ce₅–TiS₂ composites, even when the TiS₂ content reaches 10 wt%. This may be due to the fact that the crystal structure of TiS₂ is destroyed under the high-impact energy in the ball-milling process [36,37].

Figure 3 shows the SEM images of the ball-milled Mg₉₅Ce₅–TiS₂ composites with different TiS₂ contents. The figure illustrates that these ball-milled composites contain irregular particles with a particle size of 20~50 μm. Increasing TiS₂ content appears slightly reduce the particle size of the alloy powders. This phenomenon may be ascribed to the added grinding effect of the TiS₂.

To explore the phase transformations of the Mg₉₅Ce₅–TiS₂ composites during hydrogenation processes, XRD tests were conducted on the hydrogenated and dehydrogenated composite materials. It can be seen from Figure 4a that the hydrogenated Mg₉₅Ce₅ sample mainly contained MgH₂ and CeH_2.73. Clearly, the CeMg₁₂ and Mg phases in the alloy were completely transformed into MgH₂ and CeH_2.73 in the hydrogen absorption process. This result is consistent with the hydrogenation process of Mg₃RE (RE represents rare earths) intermetallic compounds [38]. Moreover, the diffraction peaks of CeS₂ and TiH_1.5 could be detected in the hydrogenated Mg₉₅Ce₅–TiS₂ samples. Therefore, it can be asserted that the TiS₂ decomposed and transformed in situ into CeS₂ and TiH_1.5 during hydrogenation. The appearance of CeS₂ suggests that sulfur is more likely to combine with cerium rather than magnesium to form a CeS₂ compound. Figure 4b shows the XRD patterns of the dehydrogenated Mg₉₅Ce₅–TiS₂ composites. The figure shows that the diffraction peaks of MgH₂ disappeared to be replaced by the diffraction peaks of Mg. Obviously, the MgH₂ in hydrogenated samples can be completely decomposed and release hydrogen during dehydrogenation process. Moreover, there is no chemical reaction occurs in the CeH_2.73, CeS₂ and TiH_1.5 phases during hydrogen absorption and desorption cycles. The CeH_2.73 and TiH_1.5 has good chemical stability and there is decomposition reaction only at higher temperatures [39,40]. Despite of this, the in-situ formed CeH_2.73, CeS₂ and TiH_1.5 can dramatically catalyze the hydrogen absorption and desorption reactions.

After five hydrogen absorption and desorption cycles, the Mg₉₅Ce₅–TiS₂ composites could achieve a stable reaction kinetics. Then the TPD curves, isothermal hydrogen absorption/desorption kinetic curves, and p–c–T curves could be measured. Figure 5 exhibits the TPD curves of the composite samples at a heating rate of 3 °C min⁻¹. The dehydrogenation peak temperature of the Mg₉₅Ce₅ alloy was approximately 389.5 °C, and the entire desorption reaction was completed when the temperature exceeded 450 °C. Notably, the dehydrogenation temperature of the Mg₉₅Ce₅–TiS₂ composite material was much lower than that of the Mg₉₅Ce₅ alloy sample, and the dehydrogenation peak temperature of the materials gradually decreased to 329.7 °C when the TiS₂ content increased to 10 wt%. However, the addition of TiS₂ increased the total weight of the Mg₉₅Ce₅–TiS₂ composites and inevitably reduced its effective hydrogen storage capacity.

Figure 6 presents the hydrogen absorption and desorption curves of the Mg₉₅Ce₅–TiS₂ composites at various temperatures. From the hydrogen absorption kinetic curves shown in Figure 6a–c, we can observe that all the samples exhibited good hydrogen absorption kinetics at 380 °C. That is, the TiS₂ catalyst did not have a significant impact on the hydrogen absorption kinetics of the composite at higher temperatures. With the decreases of operating temperature, the hydrogen absorption rate of the pure alloy material began to be limited. For example, the Mg₉₅Ce₅–10 TiS₂ composite reached full hydrogenation saturation within 75 min at 250 °C, whereas the Mg₉₅Ce₅ only reached full saturation at around 300 min. From the hydrogen desorption kinetic curves shown in Figure 6d–f, it can be seen that the hydrogen desorption kinetics of the Mg₉₅Ce₅–TiS₂ composites were also much faster than those of the Mg₉₅Ce₅ alloy. More importantly, increasing TiS₂ content significantly increased the hydrogen desorption rate. However, the maximal hydrogen absorption and desorption capacities of the composites gradually decreased with the increase of TiS₂ content, which coincided with the results of the TPD test.

Generally, the dehydrogenation reaction of magnesium hydride follows the nucleation growth model [41]. The Johnson–Mehl–Avrami–Kolmogorov (JMAK) formula can be used to fit the dehydrogenation kinetic curves of the Mg₉₅Ce₅–TiS₂ composites for a better understanding of the dehydrogenation mechanism. The JMAK formula can be expressed as [42]:

where α represents the extent of the dehydrogenation reaction at moment t, n denotes the Avrami exponent, and k is the reaction rate constant, which depends on temperature. The kinetic curve of hydrogen desorption at various temperatures was fitted with a linear JMAK equation by plotting ln[−ln(1 − α)] against lnt. In this plot, n corresponds to the fitting slope, while nlnk is the intercept. The JMAK fitting results are illustrated in Figure 7a–d, where a clear linear relationship is observed in each plot, indicating that the dehydrogenation reaction of the Mg₉₅Ce₅–TiS₂ composite material conforms to the JMAK model. The calculated values of n and lnk are presented in Table 1. The values of n for the ball-milled Mg₉₅Ce₅ alloy range from 3.305 to 3.696, suggesting that three-dimensional interfacial growth determines the dehydrogenation process [19]. After doping with TiS₂, the Avrami coefficients of the samples varied from 1.341 to 1.788, implying that the dehydrogenation process of the TiS₂-catalyzed Mg₉₅Ce₅–TiS₂ composites could be described using a zero-nucleation rate mechanism and the reaction was limited by the one-dimensional diffusion of the elements [27]. It is obvious that the CeH_2.73, CeS₂, and TiH_1.5 formed in situ play the role of nucleating agents during the dehydrogenation process of the Mg₉₅Ce₅–TiS₂ composites and significantly accelerate the hydrogen desorption kinetic performance of MgH₂ in the composites.

The dehydrogenation kinetic performance is typically characterized by the apparent dehydrogenation activation energy (E_a), which can be calculated using the Arrhenius equation [43]:

where lnk can be determined from Equation (1), E_a represents the apparent dehydrogenation activation energy (kJ mol⁻¹), R denotes the molar gas constant (8.314 J mol⁻¹ K⁻¹), T is the thermodynamic temperature in Kelvin, and k₀ is a constant. Figure 7e illustrates the Arrhenius fitting curves of lnk plotted against 1000/T. The E_a values can be derived from the slope of the fitted lines, and these obtained values are also compared in Figure 7f. We notice that the E_a value for dehydrogenation reactions decreased from 117.4 kJ mol⁻¹ to 75.0 kJ mol⁻¹ with the increase of TiS₂ content from 0 to 10 wt%. The E_a value of the Mg₉₅Ce₅–TiS₂ sample in this study wass much lower than that of the MgH₂/TiO₂ composites (106.7 kJ mol⁻¹) [44] and MgH₂–Ni/TCN material (82.6 kJ mol⁻¹) [45] reported before. Therefore, we can conclude that adding TiS₂ to Mg₉₅Ce₅ alloy can obviously decrease the dehydrogenation activation energy and therefore significantly improves the hydrogen desorption kinetic performance of the alloy.

To further investigate the impact of TiS₂ on the hydrogen storage thermodynamics of the Mg₉₅Ce₅ alloy, the p–c–T curves of the samples at different temperatures were measured, and the results are illustrated in Figure 8a–d. It is evident that all the p–c–T curves of the samples exhibited only one flat desorption plateau at each temperature, indicating that only one phase, namely Mg/MgH₂, is involved in the reversible hydrogenation cycling. The changes in enthalpy (ΔH) and entropy (ΔS) related to the hydrogen absorption and desorption reactions of the composites can be calculated using the van’t Hoff equation [46]:

Here p_eq is the dehydrogenation equilibrium pressure (MPa), p₀ is the standard atmospheric pressure (0.101325 MPa), R represents the gas constant (8.314 J mol⁻¹ K⁻¹), and T represents the experimental temperature (K). The van’t Hoff plots for the hydrogenation and dehydrogenation reactions of the Mg₉₅Ce₅–TiS₂ composites are shown in Figure 8e,f, and the calculated ΔH and ΔS values are listed in Table 2. Notably, there is no significant difference in ΔH and ΔS values, indicating that the TiS₂ had little effect on the hydrogen absorption and desorption thermodynamic properties of Mg₉₅Ce₅ alloy. This is mainly because the de/re-hydrogenation reactions in the Mg₉₅Ce₅–TiS₂ composites were Mg + H₂ ↔ MgH₂, and the catalyst could not markedly affect the hydrogen absorption and desorption reaction process.