A Scalable Strategy for Enhancing MgH₂ Hydrogen Storage: Pre-Hydrogenation and Catalyst Integration

Abstract

Magnesium has significant potential for hydrogen storage in the solid state because its capacity is about 7.6 wt%. However, the high stability of magnesium hydride requires operating temperatures superior to 380 °C for hydrogen release. It is well known that Ni could catalyze the hydrogen absorption and desorption in magnesium. In this study, carbon-coated nickel nanoparticles were employed as catalysts to enhance the hydrogen absorption and desorption kinetics of pre-hydrogenated magnesium particles. The carbon-coated nickel nanoparticles were uniformly dispersed across the surface of the pre-hydrogenated magnesium particles. In dehydrogenation at 375 °C and 350 °C, the best sample desorbs 4.90 and 4.1 wt%, respectively, in 10 min. After 45 cycles at 375 °C, the hydrogen desorption capacity is 4.91 wt%, indicating a retention capacity of 100%. Our results demonstrate that carbon-coated nickel nanoparticles can be effectively incorporated into pre-hydrogenated magnesium without the need for ball milling, significantly enhancing hydrogen absorption and desorption performance.

1. Introduction

The global energy transition toward a sustainable future has positioned hydrogen as a key energy vector. Its zero carbon emissions at the point of use and high gravimetric energy density make it an ideal substitute for fossil fuels [1]. However, large-scale adoption, particularly in transportation or mobile applications, faces a critical bottleneck: the development of safe, compact, and reversible storage methods that operate under ambient temperature and pressure conditions [2]. Among emerging technologies, solid-state storage using metal hydrides offers a promising solution due to their high volumetric densities. Magnesium-based hydrides have been extensively studied for their high theoretical storage capacity (up to 7.6 wt%), abundance, and low cost. Nevertheless, practical use of magnesium is hindered by two fundamental limitations: slow hydrogenation/dehydrogenation kinetics and the high stability of magnesium hydride (MgH₂), which requires operating temperatures above 350 °C for hydrogen release [3,4]. To overcome these challenges, research has focused on two main strategies: nanostructuring and catalyst addition. In this regard, our research group has demonstrated the positive impact of reducing Mg particle size (e.g., through the production of thin flakes) to improve kinetics, as well as the success of microwave-assisted synthesis in obtaining nanometric magnesium hydrides [5,6].

Magnesium nickel alloys are compounds attractive for hydrogen storage; these compounds are generally produced by melt casting and high-energy ball milling (HEBM). However, melt casting of Mg-Ni alloys has several disadvantages due to the differences in melting points between Mg and Ni, including the formation of heterogeneous microstructures and undesired phases resulting from the uneven distribution of alloying elements [7]. On the other hand, the HEBM could solve the issues associated with melt casting, producing Mg-Ni alloys with a high surface area due to the lower grain size [7,8]. However, it is well known that HEBM has issues with scaling up, mainly due to batch size limitations, because HEBM is only effective when used with small sample amounts [9].

Catalyst addition is a more direct route to modulate kinetic barriers and lower desorption temperatures. Nickel is a well-known additive that reacts with Mg to form the intermetallic phase Mg₂Ni, which acts as a diffusion pathway for H₂ [10,11,12]. Carbon, on the other hand, not only serves as a structural support that prevents grain growth but also promotes the hydrogen spillover effect [13,14]. Previous work by our team has explored the catalytic effect of carbon-coated nickel nanoparticles (Ni&C) on the storage performance of magnesium hydride. While the effectiveness of this additive has been confirmed, the final properties of the material are highly dependent on the incorporation method and the initial state of the magnesium [15]. The remaining challenge lies in optimizing the catalyst-substrate interface to maximize activity.

Considering the above, this study focuses on an optimized synthesis strategy that combines the pre-hydrogenation of Mg particles under high pressure and temperature conditions (80 bar, 375 °C), followed by high-energy mixing with Ni&C nanoparticles. This methodology aims to enhance the homogeneity and dispersion of the catalyst within the already formed MgH₂ matrix, thereby improving its kinetics. Therefore, the main objective of this work is to investigate the synergistic impact of pre-hydrogenation followed by high-energy mechanical mixing between MgH₂ and commercial Ni&C nanoparticles on the hydrogen absorption and desorption properties, with the goal of establishing a scalable and highly efficient preparation protocol for magnesium hydrides.

2. Materials and Methods

The magnesium powder (96% purity) used in this study was sourced from Tangshan Weihao Magnesium Powder Co., Ltd., Qian’an, China, while the carbon-coated nickel nanoparticles (Ni&C, ≥99.9% purity) were obtained from Nanostructured & Amorphous Materials, Katy, TX, USA, Los Alamos. The magnesium particles were hydrided in a custom-built reactor vessel connected to a manifold system, under a pressure of approximately 80 bar and a temperature of 375 °C. The process consisted of five cycles, each involving one hour of hydrogenation at high pressure followed by one hour of vacuum treatment at 0.025 bar. The mixing of pre-hydrogenated magnesium with Ni&C nanoparticles was carried out using a Retsch Emax high-energy ball mill, Haan, Germany, equipped with a stainless-steel jar coated with ZrO₂, operating at 800 RPM for 1 h without grinding media. All samples were manipulated in a glovebox filled with Ar (O₂ and H₂O levels were maintained under 1 ppm).

ThermoFisher Scientific Apreo 2 field emission scanning electron (FESEM) microscope, Waltham, MA, USA, equipped with an energy-dispersive X-ray (EDX) microprobe, was used for morphological characterization. X-ray diffraction (XRD) studies were performed using a PANanalytical EMPYREAM model, Malvern, UK, with Cu Ka radiation and a 2θ sweep from 5 to 90°. A polycarbonate domed sample holder for air-sensitive samples was used to measure the samples in the XRD apparatus. The polycarbonate dome in the sample holder for air-sensitive samples is associated with the wide peak around 17°. Using the XRD data, the following compounds were identified: MgH₂ (ICSD 98–016–8831), Mg (98-005-2260), and Mg₂NiH₄ (ICSD 98–016–2413).

Hydrogen absorption and desorption essays were developed in a self-constructed Sievert-type apparatus. The hydrogen absorption experiments were carried out at a hydrogen pressure of 2.5 MPa. The hydrogen desorption tests were conducted at a vacuum pressure of 0.003 MPa. The temperature-programmed desorption (TPD) profile was developed in the self-constructed Sievert-type apparatus at a vacuum pressure of 0.003 MPa at 5 °C/min from 100 to 400 °C.

The activation energy (E_a) can be estimated using the Arrhenius equation:

$$\ln (k)=-{\displaystyle \frac{E_a}{RT}}+\mathrm{l}\mathrm{n}(k_0)$$

(1)

R represents the universal gas constant, T is the temperature in Kelvin, and k₀ is a constant. The reaction rate constant (k) is calculated through Johnson-Mehl-Avrami-Kolmogorov (JMAK) equation:

$$\ln (-\ln (1-\alpha ))=n \ln (k)+n \mathrm{l}\mathrm{n}(t)$$

(2)

n is the Avrami exponent and α is the reaction rate fraction.

3. Results and Discussion

Figure 1 shows SEM images of samples M1 and M2. In sample M2, visible agglomerations of Ni&C nanoparticles are observed, in contrast to sample M1, where the Ni&C particles appear uniformly distributed across the surface of the pre-hydrogenated magnesium.

Table 1. Sample production conditions.

Sample	Amount of Hydriding Mg	Amount of Ni&C/wt.%	Post-Hydriding Process After Mixing with Ni&C
M1	400 mg	12 mg/3	No
M2	400 mg	20 mg/5	No
M3	5 g	0.15 g/3	No
M4	5 g	0.15 g/3	Yes

shows the production conditions for all samples. Figure 2 presents the hydrogen absorption and desorption curves at 375 °C for samples M1 and M2. The data indicate that increasing the amount of Ni&C nanoparticles does not lead to improved hydrogen storage capacity or enhanced kinetics during absorption and desorption. Sample M1 exhibited superior performance, desorbing 4.65 wt% of hydrogen in just 10 min, compared to 3.43 wt% for sample M2. Based on these results, along with the morphological observations from Figure 1, we selected the synthesis conditions of sample M1 for process scaling.

Figure 3 presents SEM micrographs of sample M3, revealing a uniform distribution of Ni&C nanoparticles across the surface of the pre-hydrogenated magnesium, with no significant agglomeration issues observed during process scale-up. This observation is supported by EDS analysis (Figure S1), which confirms a fine dispersion of nickel within the magnesium matrix. Figure 4 displays the hydrogen absorption and desorption curves for sample M3 at 375, 350, 300, and 275 °C. As the temperature decreases, a corresponding reduction in hydrogen storage capacity is observed. Notably, the sample retains the ability to desorb approximately 3 wt% at 300 °C and around 2 wt% at 275 °C. In previous reports, MgH₂ materials with various catalysts, including nickel compounds, achieve at least a 5 wt% hydrogen desorption at temperatures of 300 °C or higher [16,17,18]. At lower temperatures (250–300 °C), the capacity may decrease to 4 wt% or to lower values [16,17,18]. It is important to note that these results are obtained after several hours of ball milling [18]; however, we emphasize that the material developed in this work can desorb at 300 °C and 275 °C without ball milling.

To assess the effect of a second hydrogenation step on storage performance, an additional hydriding process was carried out after mixing pre-hydrogenated magnesium with Ni&C nanoparticles, using the same conditions as the initial treatment. Figure 5 displays the hydrogen absorption and desorption curves at 350 °C for samples M3 and M4. The desorption kinetics show only minor differences between the two samples, and their hydrogen storage capacities are nearly equivalent. Likewise, no significant variation was observed in the absorption behavior. These findings suggest that a second hydrogenation cycle does not provide any substantial improvement in hydrogen absorption or desorption performance.

Figure 6 displays the XRD spectra of samples M1, M3 (after Sieverts analysis), and M4. Prior to the Sieverts test, sample M1 exhibits diffraction peaks corresponding exclusively to MgH₂ and metallic magnesium, with no additional phases detected. In contrast, sample M3 shows new diffraction signals after Sieverts analysis, which are attributed to the formation of Mg₂NiH₄. This phase is also present in sample M4. The emergence of Mg₂NiH₄ is noteworthy, as it has been referred to by several authors as a ‘hydrogen pump’ due to its ability to facilitate hydrogen dissociation and enhance diffusion within the MgH₂ matrix [19,20,21,22].

Figure 7 presents the temperature-programmed desorption (TPD) profile for sample M3. As shown, hydrogen desorption begins at approximately 250 °C, although the initial release is minimal. A more pronounced desorption event occurs around 300 °C. These results are consistent with the kinetic data, confirming the temperature-dependent behavior of hydrogen release in the material.

Figure 8 illustrates the hydrogen absorption and desorption cycles of sample M3 at 375 °C. In the first cycle, the sample desorbs 4.90 wt% hydrogen, and after 45 cycles, the desorption capacity remains 4.91 wt%, corresponding to a retention rate of 100%. In contrast, the absorption capacity shows an increase, rising from 4.32 wt% to 5.10 wt% over the same number of cycles. Figure S2 presents SEM micrographs of sample M3 post-cycling. As observed, the sample undergoes morphological changes such as pulverization and segregation into small particles; nevertheless, the original shape and dimensions of the Mg particles remain discernible. Pulverization resulting from hydrogen absorption and desorption in solid materials for hydrogen storage has been previously reported by Okumura et al. [23]. According to the authors, the pulverization could increase after several cycles of hydrogenation and dehydrogenation. Pulverization could benefit hydrogen absorption and desorption by promoting the formation of small magnesium particles. Figure S3 presents the EDX analysis of sample M3 post-cycling. The results indicate that nickel remains uniformly distributed across the Mg particles following the hydrogen cycling process.

Figure 9 presents the JMAK and Arrhenius plots used to calculate the activation energy for hydrogen desorption. Based on the linear fit, the activation energy was determined to be 132.9 ± 20.9 kJ/mol. According to the literature, activation energies for magnesium dehydrogenation typically range between 150 and 240 kJ/mol [15,24,25]. These results indicate that the catalyzed magnesium in this study exhibits a significantly lower activation energy compared to untreated magnesium, confirming the beneficial effect of Ni&C nanoparticles on the desorption kinetics. Typically, values for MgH₂ catalyzed with nickel compounds (including nanoparticles) range from 115 to 66 kJ/mol [15]. However, our results show a moderate decrease in the activation energy relative to values reported in scientific literature. Regarding the Avrami exponent (slope of the JMAK equation), at all temperatures, these values remain below 2, indicating one-dimensional nucleation and growth of the magnesium phase [26].

Compared to previously reported ball-milled MgH₂ systems combined with nickel nanoparticles [20,27,28,29], the material developed in this study shows competitive hydrogen desorption kinetics; however, there is a reduction in hydrogen capacity and absorption kinetics. This could be related to particle size and incomplete hydrogenation. This reduction in capacity may be attributed to the presence of unreacted metallic Mg in the larger-scale samples, as evidenced by the residual Mg peaks observed in Figure 6. During the pre-hydrogenation step, the increased batch mass (5 g vs. 400 mg) limits hydrogen diffusion into the particle core, leading to incomplete conversion to MgH₂. Notably, this synthesis approach eliminates the need for ball milling, simplifying the preparation process, reducing energy consumption and cost, but more importantly, it preserves the original high-activity structure of the Ni&C nanoparticles, preventing potential agglomeration. An alternative strategy for catalyzing magnesium involves the formation of Mg–Ni alloys to enhance hydrogenation and dehydrogenation kinetics. However, due to the significant difference in melting points between Mg and Ni, conventional casting methods are not feasible [30]. Consequently, researchers have explored techniques such as high-energy ball milling (HEBM) and hydriding combustion synthesis (HCS). While HEBM requires prolonged milling times to promote the Mg–Ni reaction [30], HCS demands temperatures above 600 °C and high hydrogen pressures [31,32,33]. Regarding the hydrogen absorption/desorption behavior of the Mg-Ni alloys, it is essential to highlight that the maximum gravimetric storage capacity is 3.6 wt% at operating temperatures above or higher than 300 °C [30,32,34]. In contrast, the method proposed in this work offers a simpler and more energy-efficient alternative, achieving comparable or even superior hydrogen storage performance.

4. Conclusions

In this work, we demonstrate that carbon-coated nickel nanoparticles can be uniformly incorporated into pre-hydrogenated magnesium without ball milling, producing a catalyst–hydride interface that delivers competitive or superior performance compared with existing Ni- or carbon-modified MgH₂ systems prepared by traditional high-energy milling routes.

The developed material achieves 4.90 wt% H₂ desorption in 10 min at 375 °C, which is comparable to the fastest Ni-based catalysts reported in the literature that require several hours of ball milling. More importantly, our material desorbs ~3 wt% at 300 °C and ~2 wt% at 275 °C, whereas most Ni-catalyzed MgH₂ composites exhibit significant performance degradation below 300 °C despite extensive mechanical activation.

The material retains 100% of its desorption capacity after 45 cycles, whereas Ni-based MgH₂ systems commonly report 10–20% capacity loss over 20–50 cycles due to catalyst agglomeration or structural coarsening. SEM/EDS analyses confirm that the carbon shell prevents Ni particle migration, enabling the exceptional stability observed.

Our process demonstrates the ability to scale from 400 mg to 5 g while preserving catalyst dispersion and kinetics, something not achievable using high-energy ball milling because of its batch-size and energy-efficiency limitations.

The technological readiness level of the developed material is TRL 3–4, corresponding to laboratory scale validation. The demonstrated desorption behavior (300–375 °C), fast kinetics, excellent cycling stability, and scalable preparation suggest strong potential for medium temperature applications such as waste heat assisted hydrogen reactors, stationary storage, and solar thermal systems.