Enhanced hydrogen storage properties of magnesium hydride using a KH-TiO2-Nb2O5/carbon-coated nickel nanoparticles catalyst Robinson Aguirre Ocampo a,*, Julian Arias Velandia a, Julian A. Lenis a, Alejandro A. Zuleta Gil b, Sindy Bello c, Esteban Correa c, Carlos Arrieta d, Francisco J. Bolívar a, F´elix Echeverria Echeverria a a Centro de Investigaci´on, Innovaci´on y Desarrollo de Materiales – CIDEMAT, Facultad de Ingeniería, Universidad de Antioquia UdeA, Calle 70 No 52 – 21, Medellín, Colombia b Grupo de Investigaci´on de Estudios en Dise˜no - GED, Facultad de Dise˜no Industrial, Universidad Pontificia Bolivariana, Sede Medellín, Circular 1 No 70 – 01, Medellín, Colombia c Grupo de Investigaci´on Materiales con Impacto – MAT&MPAC, Facultad de Ingenierías, Universidad de Medellín UdeM, Carrera 87 No 30 – 65, Medellín, Colombia d Grupo de Investigaci´on en Energía – GRINEN, Facultad de Ingenierías, Universidad de Medellín UdeM, Carrera 87 No 30 – 65, Medellín, Colombia A R T I C L E I N F O Keywords:

Nanoparticles Nickel Hydrogen storage TiO2 Nb2O5 Magnesium hydride A B S T R A C T Magnesium hydride (MgH₂) is a promising hydrogen-storage material, but its slow sorption kinetics and high thermodynamic stability continue to limit practical implementation. In this work, we demonstrate an original and scalable approach to enhance MgH₂ performance by employing a catalyst prepared entirely from commer­ cially available powders (KH, TiO₂, and Nb₂O₅), combined with commercial carbon-coated Ni nanoparticles (C@Ni). Unlike MXene-based catalysts widely reported in the literature, our catalyst system avoids complex and hazardous synthesis steps and enables industrially scalable processing. Two incorporation routes (one-step ball milling and a two-step mixing strategy) were evaluated. The catalyst decreased the dehydrogenation onset temperature from 321 ◦C to below 236 ◦C and reduced the activation energy from 152 to 93.8 kJ⋅mol⁻¹ . The optimized material exhibited fast hydrogen desorption (6.38 wt% in 10 min at 300 ◦C) and a low-temperature absorption (5.77 wt% at 150 ◦C and 3.47 wt% at 75 ◦C), a performance not previously reported for catalyst systems derived solely from commercial precursors. Moderate cycling stability was also achieved, retaining approximately 79% capacity after 20 cycles. These results highlight the originality and practical relevance of a catalyst design strategy that enables high-performance MgH₂-based hydrogen storage without the need for laboratory-specific or non-scalable synthesis routes.

1. Introduction

Hydrogen (H2) has the potential to replace conventional fossil fuels and power an environmentally sustainable civilization. However, under standard temperature and pressure, hydrogen has a low density, which presents significant storage challenges [1–3]. Because hydrogen has a low boiling point of 20.4 K at 1 atmosphere, it is stored in high-pressure tanks or cryogenic systems [2,4,5]. Therefore, this entails higher energy costs and safety hazards associated with storage and transportation [6, 7]. To solve the drawbacks of employing H2 in its innate form, re­ searchers have focused on solid-state materials for hydrogen storage [8–12]. Among the solid materials used to store hydrogen, magnesium-based compounds are well known for their abundance, low cost, cycling stability, and high hydrogen storage capacity (approxi­ mately 7.6 wt%) [2,13–15]. However, magnesium hydride (MgH2) re­ leases hydrogen slowly and decomposes at elevated temperatures, often exceeding 360 ◦C at 0.1 MPa of pressure in an argon atmosphere, because of its remarkably high activation energy, which has been re­ ported to exceed 152 kJ/mol [16–19].

Recent work has explored solar-driven strategies that couple pho­ tothermal heating with catalytic interfaces to reduce the external energy required for reversible storage in light-weight hydrides. For MgH₂, atomic-level reconstructions that integrate photothermal absorption and catalysis have enabled stable solar-driven cycling under concentrated irradiation by co-localizing light-induced heat and active sites [1]. Complementary concepts that employ photothermal MXene-based

* Corresponding author.

E-mail address: robinson.aguirre@udea.edu.co (R.A. Ocampo). Contents lists available at ScienceDirect Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom https://doi.org/10.1016/j.jallcom.2026.188174 Received 2 March 2026; Received in revised form 17 April 2026; Accepted 20 April 2026 Journal of Alloys and Compounds 1065 (2026) 188174 Available online 25 April 2026 0925-8388/© 2026 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC license ( http://creativecommons.org/licenses/bync/4.0/ ).

layers have also demonstrated fully solar-driven cycling of MgH₂ [18]. Beyond Mg-based systems, light-enabled hydrogen storage has been reported in borohydrides where photogenerated vacancies destabilize B–H bonds at near-ambient conditions [15]. In parallel with these light-assisted approaches, investigators have pursued thermally driven strategies that rely on nanostructuring of MgH₂, the use of diverse cat­ alysts, mechanical alloying, and the creation of multiphase composites [20,21]. Numerous elements have been incorporated into MgH2 to boost its behavior in hydrogenation and dehydrogenation, including Zr, Fe, Mn, V, Cr, Ti, Nb, and Ni, among others [17]. Titanium and niobium have been utilized as catalysts in their metallic form, as well as in oxides, carbides, and other variations [22]. MXene refers to 2D transition metal carbonitrides, carbides, and nitrides. Mxenes have been utilized in various applications, including microelectronics, nanoelectronics, supercapacitors, and electrocatalytic water splitting [23,24]. Titanium and niobium MXenes have been utilized as additives to boost the hydrogen storing characteristics of MgH2 [9,25–27].

The two-dimensional structure and high surface area significantly enhance the MgH2 hydrogenation and dehydrogenation behavior and lower the dehydrogenation temperature. Titanium MXene was produced by Zhu et al. [28]; the resulting MgH2-titanium MXenes composites exhibited a hydrogen desorption of 3 wt% at 150 ◦C in 2.5 h. Similarly, niobium MXenes have also been used as additives to improve the hydrogenation and dehydrogenation behavior of MgH₂. Liang et al. [29] produced MgH2-niobium MXenes composites, which exhibited hydrogen desorp­ tion and absorption of approximately 6 and 7 wt% respectively at 200 ◦C in 6 min. Potassium hydride (KH) has recently emerged as an effective mechanochemical reductant for generating defect-rich suboxides from robust transition-metal oxides, thereby creating catalytically active in­ terfaces relevant to MgH₂ [30]. Milling TiO₂ with KH produces “black” TiO₂₋ₓ containing Ti³⁺ and oxygen-vacancy defects, and when used with MgH₂ it lowers the dehydrogenation onset to ~194 ◦C and enables rapid low-temperature cycling [31]. Fu et al. [32] developed a niobium-based bimetallic catalyst, CuNb2O6, with exceptional catalytic performance. The CuNb2O6/MgH2 composite can absorb 4.28 wt% of hydrogen in

10 min at 100 ◦C. Additionally, the composite can absorb approximately

2 wt% hydrogen within 60 min at 50 ◦C. For hydrogen desorption, the

composite can desorb 4.65 wt% of hydrogen within 20 min at 225 ◦C. After 50 cycles at 300 ◦C, the capacity retention was 87%. Wang et al. [33] developed a composite of MgH₂ and CeO₂–Ni-rGO. With the addi­ tion of 10 wt% CeO₂–Ni–rGO, the MgH₂ composite achieves a hydrogen uptake of 4.0 wt% within 60 s at 100 ◦C. According to the authors, the hydrogenation activation energy decreases to 52.45 kJ/mol, and the onset of dehydrogenation temperature decreases to 195 ◦C. The MgH₂ and 10 wt% CeO₂–Ni–rGO composite releases about 5 wt% of hydrogen in 5 min at 250 ◦C. After 30 hydrogen absorption/desorption cycles at

300 ◦C, the capacity retention was 97%. Liu et al. [34] developed a

coral-like NiO/NiFe2O4 heterojunction catalyst. The MgH2–NiO/Ni­ Fe2O4 composite can release 6.2 wt% of hydrogen at 275 ◦C in 20 min. On the other hand, the composite can absorb 4.4 wt% of hydrogen in

30 min at 100 ◦C. According to the authors, the hydrogen desorption

onset decreases to 188 ◦C, which represents a 133 ◦C difference compared to ball-milled MgH2. Besides, the activation energy decreases to 86.97 kJ/mol, representing a reduction of 62.17 kJ/mol relative to ball-milled MgH₂.

Despite significant progress, current MgH₂ catalytic systems face persistent barriers that limit translation. State-of-the-art catalysts (most notably MXenes) remain largely non-commercial and are typically produced under restricted laboratory conditions with hazardous etch­ ants and complex, laboratory-specific syntheses, which increases cost and limits scalability [35,36]. In parallel, mechanistic interpretations at metal/oxide–hydride interfaces continue to evolve, particularly regarding the roles of defect-rich suboxides and hydrogen spillover in lowering nucleation barriers and accelerating Mg–H bond dynamics [16, 26,37]. These constraints highlight a clear gap for catalyst solutions that couple strong kinetic benefits with industrial practicality, specifically, derivation from commercial precursors and processing through scalable routes.

Motivated by this gap, the present work introduces a fully com­ mercial catalyst system in which KH-modified TiO₂/Nb₂O₅ (CAT) is combined with carbon-coated Ni nanoparticles (C@Ni). We investigate how this catalyst, incorporated through two distinct routes—a one-step high-energy ball-milling method and a two-step mixing strat­ egy—affects key performance metrics, including dehydrogenation onset temperature, hydrogen absorption and desorption kinetics, apparent activation energy, low-temperature hydrogen uptake, and cycling sta­ bility. By demonstrating these effects without relying on laboratoryspecific synthesis routes, such as MXene fabrication, this study pro­ poses a scalable and practical alternative for improving MgH₂-based hydrogen storage materials.

2. Materials and methods

Magnesium hydride (≥98%), TiO₂ nanoparticles (99.0%), and Nb₂O₅ powders (99.0%) were purchased from Lead Optima Element Tech Co. Carbon‑coated nickel nanoparticles (C@Ni, ≥99.9%) were purchased from Nanostructured & Amorphous Materials. All samples were handled inside an argon-filled glovebox. (Vigor Tech USA), maintaining O2 and H2O levels less than one part per million. Potassium hydride (KH, 30 wt % dispersion in mineral oil) was purchased from Sigma-Aldrich. To remove the mineral oil, the KH was washed repeatedly in hexanes in an argon-filled glovebox. No further treatment was applied to any of the remaining powders. As-received magnesium hydride had a particle size of 8.19 ± 7.44 µm. TiO2 nanoparticles had a particle size of

23.05 ± 9.67 nm, with more than 80% measuring below 30 nm (see

Fig. S1). Nb2O5 particles had a mean size of 154.69 ± 74.30 nm, with over 70% smaller than 200 nm (see Fig. S2). The carbon-coated nickel nanoparticles had a particle size of 28.30 ± 16.10 nm, with 75% of them being smaller than 40 nm (see Fig. S3).

The catalyst (CAT) was synthesized by mixing TiO2, Nb2O5, and KH in equal molar ratios (1:1:1), following the stoichiometry reported by Dai et al. [31]. In this configuration, KH serves to pre-induce defect-rich Ti/Nb suboxides within the commercial KH–TiO₂–Nb₂O₅ precursor. This approach offers a scalable synthesis route for generating highly active oxide–metal interfaces. The mix was ball-milled in a Retsch Emax high-energy ball mill with a ball-to-powder weight ratio (BPR) of 40:1, 50% of filling vial percentage, and 3 mm ZrO2 balls in a 125 mL stainless steel jar coated with ZrO2 at 400 RPM. Ball milling was carried out for

8 h in cycles of 20 min milling followed by 5 min cooling. No additional

post-treatment was applied to the CAT after milling. The as-milled powder was collected and stored in the glovebox. The CAT and the C@Ni were incorporated to the MgH2 by two methods, as displayed in Fig. 1. The first method involves a single step of ball milling of all components (MgH2, CAT, and the C@Ni). The ball milling was per­ formed in the high-energy ball mill employing a 125 mL stainless steel jar coated with ZrO2, with 3 mm ZrO2 balls, a BPR of 80:1 and a 50% filling vial percentage. The ball milling process lasted one hour, with 10-minute intervals of ball milling, followed by 10 min of rest. The second method consisted of two steps. In the first step, the CAT and MgH2 are ball-milled under the same conditions as those explained in the first method. Following that, in the second step, we used a high-energy ball mill set at 800 RPM for one hour to combine 400 mg of the ball-milled sample with C@Ni, without grinding bodies, in a 50 mL stainless steel jar coated with ZrO2. Hereafter, BM and MIX are the ab­ breviations for the first and the second method, respectively. For com­ parison, MgH2 subjected to ball milling without CAT or C@Ni (hereafter referred to “as-milled MgH2”) was used. In our previous work [2], the MIX and BM methods were systematically evaluated using C@Ni as the additive, and the optimal additive content for each method was inde­ pendently determined. Accordingly, the C@Ni loading used in both MIX and BM routes was selected on the basis of those experimental results. Information about morphological and compositional R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

characterization, as well as hydrogen storage assays, is provided in supplementary information. More details about the self-designed Sie­ vert-type apparatus could be found in [38].

3. Results

Fig. 2 presents the SEM image and EDS elemental mapping of the catalyst. The catalyst particles are predominantly 200 and 300 nm in size and form agglomerates of approximately 1–2 µm. The EDS maps further confirm a uniform spatial distribution of Ti, K, O, and Nb throughout the catalyst, indicating effective mixing and homogenization during the milling. Elemental composition from EDX of CAT is shown in Table S1. The XPS spectra of the catalyst are shown in Fig. 3. In the C 1s–K 2p region (Fig. 3a), four characteristic peaks are identified. The signals at 285 eV and 288.45 eV correspond to C–C and C––O bonding, respectively, whereas the peaks at 292.78 eV and 295.57 eV are asso­ ciated with K–O species [39,40]. In the Nb 3d region (Fig. 3b), three fitted peaks are observed. The peaks at 206.59 eV (Nb 3d₅/₂) and 209.30 eV (Nb 3d₃/₂) are attributed to Nb⁴⁺–O bonding [41–43]. Whereas the intermediate peak at 207.56 eV is consistent with Nb⁵⁺–O species [41]. Similarly, the Ti 2p region (Fig. 3c) displays three well-resolved contributions. The peaks at 457.89 eV (Ti 2p₃/₂) and

463.79 eV (Ti 2p₁/₂) correspond to Ti³⁺ states, whereas the peak at

458.30 eV is assigned to Ti⁴⁺ (Ti 2p₃/₂) [44–48]. Regarding the O 1 s

region (Fig. 3d), three distinct contributions are identified at 532.85, 530.98, and 529.40 eV. The peak at 532.85 eV is related to surface hy­ droxyl groups [49,50]. On the other hand, the peak at 530.98 eV is attributed to non-stoichiometric oxide species (ONS), which are typically associated with the reduced Ti³⁺ and Nb⁴⁺ states present in the catalyst. In contrast, the peak at 529.40 eV corresponds to lattice oxygen (OL), originating from fully oxidized Ti⁴⁺ and Nb⁵⁺ species [43,45,47,51–53]. The relative integrated areas of these components yield an ONS/OL ratio of 2.46, indicating a highly defective oxide surface enriched in non-stoichiometric oxygen species.

Fig. 4 shows the HRTEM images of the catalyst together with their corresponding FFT patterns, and the measured d-spacings are summa­ rized in Table S2. Several observed d-spacing values exceed 0.3509 nm, which is the largest interplanar distance reported for TiO₂ polymorphs (specifically the anatase (011) plane) [54–56]. These enlarged spacings therefore indicate the presence of titanium suboxides. Additional d-spacing values match well with those expected for niobium suboxides, such as NbO and NbO₂. Furthermore, the spacings greater than 0.6 nm are consistent with complex niobium suboxides, particularly Nb₁₂O₂₉, suggesting that KH reduction promotes the generation of highly defec­ tive and sub-stoichiometric oxide phases within the catalyst. The SEM images and EDS elemental mapping of the MgH₂–3 wt% CAT–3 wt% C@Ni–MIX composite are presented in Fig. 5. As shown in Fig. 5 and Fig. S4, the incorporation of CAT and C@Ni does not produce significant changes in particle size of MgH₂, indicating that the milling and mixing steps do not induce substantial morphological changes of the hydride matrix. Moreover, the homogeneous distribution of Ti, K, Ni, and Nb observed in the EDS maps confirms a uniform dispersion of both the catalyst and the carbon‑coated Ni nanoparticles throughout the MgH₂ particles. Similar results were obtained from the SEM micrographs and EDS mapping of the MgH₂–1 wt% CAT–0.5 wt% C@Ni–BM com­ posite (See Figs. S4 and S5). Importantly, no Zr‑related signals were Fig. 1. Methods for adding the catalyst and the C@Ni to MgH2. R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

detected in the EDS spectra, despite the use of ZrO₂ grinding media and ZrO₂‑coated milling jars. This confirms that no contamination from the milling equipment was introduced into the final material. It is important to note, however, that SEM-EDS provides areal or volumetric informa­ tion rather than surface-specific measurements. It cannot distinguish nanoparticles exposed at the outer surface from those that are partially embedded within MgH₂/CAT agglomerates, and it cannot quantify the fraction of catalytically accessible Ni near the surface. In addition, the finite interaction volume and peak overlaps limit nanoscale resolution and the quantification of light elements, for example the carbon shells, and EDS cannot resolve the chemical state of nickel (e.g., Ni⁰ versus NiO). Consequently, although the maps confirm compositional homo­ geneity and rule out phase segregation or core–shell morphologies at the microscale, they do not directly establish the surface accessibility of C@Ni/CAT sites. This limitation probably differentiates the BM and MIX routes and underpins the superior low-temperature hydrogen uptake observed for MIX. Direct quantification of exposed active sites will require surface-sensitive and nanoscale analyses beyond SEM-EDS. The TPD curves obtained at a heating rate of 5 ◦C min⁻¹ are shown in Fig. 6, and the corresponding onset temperatures, DSC peak tempera­ tures, activation energies, and hydrogen capacities at 300, 275, and

260 ◦C are summarized in Table 1. Incorporating CAT and C@Ni

Fig. 2. SEM images and EDS mapping analysis of the catalyst. Oxygen is red, potassium is violet, titanium is cyan, and niobium is orange in the EDS mapping study. R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

significantly decreases the dehydrogenation onset temperature of MgH₂, reducing it from 321 ◦C for the as-milled MgH₂ to below 236 ◦C in all catalyst-containing samples (Fig. 6, Table 1). This reduction demon­ strates the strong catalytic influence of the KH-modified TiO₂/Nb₂O₅ phases and the carbon-coated Ni nanoparticles.

The DSC profiles for samples with and without CAT and C@Ni, also recorded at 5 ◦C min⁻¹ , are presented in Fig. 7. As shown in Fig. 7 and Table 1, the addition of the catalyst system leads to a decrease of at least

9 ◦C in the DSC peak temperature relative to the as-milled MgH₂, with

the most pronounced shift reaching 32 ◦C. These collective results confirm that the introduction of CAT and C@Ni effectively destabilizes Mg–H bonding and lowers the temperature required for hydrogen release.

Fig. 8 shows the isothermal hydrogen desorption and absorption kinetics of the CAT- and C@Ni-containing samples at 300, 275, and

260 ◦C. When compared with the as-milled MgH₂, the catalytic effect is

apparent. At 300 ◦C, the as-milled MgH₂ requires approximately

200 min to reach 6.5 wt% hydrogen desorption (Fig. S6), whereas the

MgH₂–3 wt% CAT–0.5 wt% C@Ni–BM composite achieves 6.45 wt% in only 20 min (Table 1). In terms of hydrogen absorption at this temper­ ature, all catalyst-containing samples exhibit rapid uptake, reaching approximately 6.1 wt% within 5 min.

At 275 ◦C, the MgH₂–1 wt% CAT–0.5 wt% C@Ni–BM sample dem­ onstrates the best dehydrogenation performance, releasing 4.72 wt% hydrogen in 40 min, while the remaining compositions desorb between

4.1 and 4.5 wt% under the same conditions. Hydrogen absorption at

275 ◦C is likewise fast for all samples, with values close to 6 wt%

achieved in approximately 8 min.

At 260 ◦C, all composites desorb around 2.7 wt% hydrogen within 60 min, except for the MgH₂–1 wt% CAT–3 wt% C@Ni–MIX sample, which reaches 2.4 wt%. Hydrogen absorption at this temperature also proceeds efficiently across the samples, reaching ≈6 wt% within

10 min, except for MgH₂–3 wt% CAT–3 wt% C@Ni–MIX, which attains

5.84 wt% under the same conditions.

Hydrogen absorption at 150 ◦C was evaluated for all samples over

50 min (Fig. 9). Under these conditions, the MgH₂–1 wt% CAT–3 wt%

C@Ni–MIX composite exhibits the highest performance, absorbing

5.77 wt% hydrogen, whereas the remaining samples reach 3.80 wt% or

less. This result highlights the superior low-temperature activity of the MIX-processed material.

Fig. 10 presents the isothermal absorption kinetics of the MgH₂–1 wt % CAT–3 wt% C@Ni–MIX sample at 150, 100, and 75 ◦C. Notably, this material maintains significant hydrogen uptake even at substantially lower temperatures, absorbing 4.28 wt% at 100 ◦C and 3.47 wt% at

75 ◦C within 120 min, demonstrating exceptional catalytic efficiency for

low-temperature hydrogenation.

Cycling stability was assessed at 300 ◦C for the MgH₂–1 wt% CAT–0.5 wt% C@Ni–BM composite (Fig. 11). After 20 cycles, the sam­ ple retains a reversible desorption capacity of 5.07 wt% (corresponding to a 78.72% retention) and a reversible absorption capacity of 5.22 wt% (78.77% retention). These results confirm that the catalyst system en­ hances hydrogen sorption kinetics without compromising cycling durability.

The JMAK and Arrhenius plots for all evaluated samples are Fig. 3. X-ray photoelectron spectroscopic spectrums of the catalyst. C 1 s – K 2p region (a), Nb 3d region (b), Ti 2p region (c), O 1 s region (d). R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

presented in Fig. 12, and the corresponding activation energies (Ea) derived from these analyses are summarized in Table 1. As shown in Fig. 12 and Table 1, every catalyst‑containing sample exhibits a sub­ stantial lower Ea than as‑milled MgH₂. In all cases, the activation energy decreases by at least 44 kJ mol⁻¹ relative to the reference material (Fig. S7), confirming the strong catalytic effect of the KH‑modified TiO₂/ Nb₂O₅ phases and the carbon‑coated Ni nanoparticles on dehydroge­ nation kinetics. Among all compositions, the MgH₂–3 wt% CAT–3 wt% C@Ni–MIX sample displays the lowest activation energy, highlighting the enhanced efficiency of the MIX incorporation method in promoting faster hydrogen sorption pathways.

Fig. 13 presents the SEM images of the MgH₂–1 wt% CAT–0.5 wt% C@Ni–BM composite after the cycling tests. As observed, the particle size distribution becomes non‑uniform after repeated hydro­ genation–dehydrogenation cycles: some particles range from 1 to 4 µm, whereas others are smaller than 1 µm. At higher magnification, the larger agglomerates are revealed to consist of nanometric sub‑particles (< 50 nm), indicating progressive fragmentation and reagglomeration during cycling. This multiscale morphology is typical of MgH₂ systems subjected to repeated volumetric expansion and contraction, and it is consistent with the moderate capacity retention observed for this sample.

4. Discussion

According to the TPD curves, the MIX-processed samples exhibit onset temperatures approximately 12 ◦C lower than their BM counter­ parts. Moreover, all compositions incorporating CAT and C@Ni show a marked reduction in the dehydrogenation onset temperature relative to the as-milled MgH₂, confirming the strong catalytic effect of the com­ bined KH-modified oxides and carbon-coated Ni nanoparticles. A similar trend is observed in the DSC measurements: samples containing CAT and C@Ni display a substantial decrease in the DSC peak temperature when compared with as-milled MgH₂, further demonstrating that the catalyst system effectively destabilizes the Mg–H bond and facilitates earlier hydrogen release.

These results are consistent with previous findings by Dai et al. [31], who synthesized black TiO₂ through KH-assisted ball milling and used it to enhance the hydrogenation–dehydrogenation behavior of MgH₂. Their material exhibited a TPD onset temperature of 194 ◦C at a heating rate of 2 ◦C min⁻¹ , representing a 96 ◦C reduction relative to their reference MgH₂ sample. The agreement between these earlier Fig. 4. High-resolution TEM ((a), (b)) and associated FFT patterns ((c), (d)) of the catalyst (CAT), highlighting lattice fringes and reciprocal-space features. R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

observations and the present results reinforces the role of KH-induced oxide reduction (including the formation of Ti³⁺ /Nb 4+ species), significantly promoting hydrogen sorption kinetics in MgH₂-based systems.

At 300, 275, and 260 ◦C, both MIX and BM samples display excellent dehydrogenation behavior, particularly in terms of their reduced acti­ vation energies and accelerated hydrogen desorption kinetics. Notably, particular attention should be given to the performance of the MgH₂–1 wt% CAT–3 wt% C@Ni–MIX sample, which exhibits remark­ able low‑temperature hydrogen absorption. This composite absorbs

5.77 wt% hydrogen in 50 min at 150 ◦C, and remains active even at

significantly lower temperatures, reaching 4.28 wt% at 100 ◦C and

3.47 wt% at 75 ◦C. These results highlight the exceptional catalytic ef­

ficiency of the MIX‑processed material under low‑temperature hydro­ genation conditions.

Table 2 compares the hydrogen storage performance obtained in this Fig. 5. SEM micrographs and EDS mapping of the MgH₂–3 wt% CAT–3 wt% C@Ni–MIX composite. Elemental distribution is represented as follows: O (red), K (violet), Mg (yellow), Ti (cyan), Nb (orange), C (white), and Ni (green). R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

work with values previously reported in the literature. Among the cited studies, MgH₂–Nb-MXene composites stand out for achieving rapid hydrogen desorption, approximately 6 wt% at 200 ◦C in 6 min. How­ ever, despite their excellent kinetics, these materials have notable drawbacks: MXenes are not commercially available, and their synthesis typically requires hydrofluoric acid, a highly corrosive and hazardous reagent [26,29]. These limitations hinder scalability and practical deployment. In contrast, most MgH₂-based hydrogen storage materials reported in the literature are evaluated at temperatures above 280 ◦C, and their catalytic additives are often produced via specialized labora­ tory synthesis routes that involve controlled atmospheres, multi-step procedures, or non-commercial precursors. The approach employed in the present study differs fundamentally from these strategies. Here, the catalyst system was synthesized exclusively from commercially avail­ able powders (KH, TiO₂ nanoparticles, Nb₂O₅ nanoparticles, and carbon-coated Ni nanoparticles) yet delivers competitive performance at lower additive loadings. For example, the MgH₂–1 wt% CAT–0.5 wt% C@Ni–BM composite achieves ~2.5 wt% hydrogen desorption at 260 ◦C in 1 h, demonstrating the catalyst effectiveness even at minimal loading. Furthermore, when comparing these results with our previous study [2], which evaluated only commercial carbon-coated Ni nanoparticles, the beneficial effect of incorporating CAT becomes clear. In the earlier work, the best-performing materials desorbed hydrogen at 275 ◦C and absor­ bed hydrogen at 150 ◦C, whereas in the present study, the optimized composites can desorb at 260 ◦C and absorb at temperatures as low as

75 ◦C. This improvement highlights the synergistic effect between

KH-modified TiO₂/Nb₂O₅ and C@Ni and underscores the novelty and practical relevance of using a fully commercial, scalable catalyst system to improve MgH₂ hydrogen storage performance.

Pure MgH₂ is well known for limited cycling stability, with capacity retention often falling below

70%

under repeated hydro­ genation–dehydrogenation cycles [67]. Other studies report slightly improved behavior, such as 82% retention after only five cycles, which still highlights the intrinsic degradation issues associated with MgH₂-based materials [68]. In contrast, the MgH₂–1 wt% CAT–0.5 wt% C@Ni–BM composite developed in this work exhibits an enhanced Fig. 6. Temperature-programmed desorption (TPD) curves of the evaluated samples, recorded at a heating rate of 5 ◦C min⁻¹ . Table 1 Onset temperature from TPD, lowest peak temperature from DSC, and Activation Energy of all samples. Sample Onset dehydrogenation temperature from TPD (◦C) Temperature at DSC Peak (◦C) Activation Energy (Ea) (kJ/mol) Hydrogen content at 300 ◦C. Dehydrogenation/ Hydrogenation (wt%) Hydrogen content at 275 ◦C.

Dehydrogenation/ Hydrogenation (wt%) Hydrogen content at 260 ◦C.

Dehydrogenation/ Hydrogenation (wt%) MgH2-3 wt

% CAT-

0.5 wt%

C@Ni-BM

234.24

326.90

97.9 ± 2.7

6.45/6.22 4.16/6.16 2.63/6.22 MgH2-1 wt

% CAT-

0.5 wt%

C@Ni-BM

235.06

317.92

108.1 ± 6.6

6.44/6.13 4.72/6.08 2.77/6.04 MgH2-3 wt

% CAT-

3 wt%

C@Ni-

MIX

218.13

329.93

100.5 ± 6.3

5.73/6.22 4.56/6.29 2.71/6.01 MgH2-1 wt

% CAT-

3 wt%

C@Ni-

MIX

220.60

341.20

93.8 ± 10.0

5.62/6.12 4.14/6.11 2.36/5.84 Ball-milled MgH2

321.59

350.40

152.5

± 15.29 6.5/6.5 ——————————— ——————————— Fig. 7. Differential scanning calorimetry (DSC) curves recorded at 5 ◦C min⁻¹ . R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

Fig. 8. Isothermal kinetics curves for hydrogen desorption at 300 ◦C (a), 275 ◦C (c), and 260 ◦C (d), and hydrogen absorption at 300 ◦C (b), 275 ◦C (e), and 260 ◦C (f).

R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

cyclability, maintaining a dehydrogenation capacity of 78.72% after 20 cycles. This performance is comparable to previously reported catalyst-modified systems [26,37,69,70], demonstrating that the com­ bination of KH-modified oxides and carbon-coated Ni nanoparticles mitigates structural and kinetic degradation during long-term cycling. The decrease in hydrogen capacity during both desorption and absorp­ tion may be attributed to particle coarsening and agglomeration, which reduce the available surface area and thereby limit hydrogen diffusion (see Fig. 13) [71]. Another possible cause is the redistribution of cata­ lytically active species like C@Ni or defect-rich Ti/Nb suboxides, due to particle coarsening and agglomeration, which reduces catalytic effi­ ciency [71,72].

From the XPS analysis (Fig. 3), signals corresponding to Nb⁴⁺, Ti³⁺, and K–O bonding are clearly resolved. The presence of K–O species in­ dicates the chemical transformation of KH during the milling process, consistent with the formation of non-stoichiometric oxides and oxygenvacancy–rich phases (see Equations S1 and S2). According to the liter­ ature, such mixed-oxide systems often enhance hydrogen storage performance by providing active sites that facilitate both hydrogen ab­ sorption and desorption [30,31]. Furthermore, the reduction of TiO₂ and Nb₂O₅ by KH promotes the formation of titanium and niobium sub­ oxides, as confirmed by both XPS and HRTEM analyses. These defects play a crucial role in modifying the surface electronic structure, improving charge transport, and enhancing catalytic activity of the resulting oxide phases [73]. Titanium and niobium suboxides are also known to facilitate H₂ dissociation, hydrogen spillover, and the subse­ quent release and uptake of hydrogen in MgH₂, thereby accelerating both hydrogenation and dehydrogenation kinetics by acting as highly reactive catalytic sites [74–76]. In addition, the catalytic effect of carbon-coated Ni nanoparticles (C@Ni) on MgH₂ sorption kinetics has been extensively detailed in our previous work [2], where Ni-assisted hydrogen activation and spillover mechanisms were shown to enhance MgH₂ performance significantly. The present results further reveal a synergistic interaction between the KH-modified oxide catalyst (CAT) and C@Ni, enabling hydrogen desorption below 275 ◦C and hydrogen absorption at temperatures as low as 75 ◦C. This synergy demonstrates the combined benefits of defect-rich oxide phases and Ni-mediated hydrogen activation, reinforcing the effectiveness of this fully com­ mercial and scalable catalyst system.

SEM/EDS data indicate compositional homogeneity for both routes (Figs. 5 and S5), yet the microstructural scenarios they create differ. In BM, simultaneous high-energy milling of MgH₂, CAT, and C@Ni can partially embed C@Ni within MgH₂/CAT agglomerates and promote local coalescence, which reduces the fraction of Ni accessible at the surface. In MIX, the second mixing step, performed without grinding media, preserves the integrity of C@Ni and favors its deposition at the surface near defect-rich Ti/Nb suboxides, as supported by XPS of CAT and HRTEM d-spacings data. The resulting microstructure increases the density of exposed dissociation and spillover sites while shortening hydrogen diffusion paths. This picture consistently explains the observed kinetic trends. MIX composites show lower TPD onsets (218.1–220.6 ◦C) than their BM counterparts (234.2–235.1 ◦C, Fig. 6) and reach the lowest apparent activation energy (Eₐ = 93.8 ±

10.0 kJ mol⁻¹, Fig. 12), while both routes decrease the DSC peak tem­

perature relative to as-milled MgH₂ (Fig. 7). Consistent with enhanced near-surface exposure of Ni, MIX delivers superior low-temperature hydrogenation, reaching 5.77 wt% at 150 ◦C in 50 min and remaining active down to 100 ◦C with 4.28 wt% in 120 min and 75 ◦C with 3.47 wt % in 120 min (Figs. 9–10). Taken together, these results establish a Fig. 9. Isothermal hydrogen absorption profiles of the samples at 150 ◦C. Fig. 10. Isothermal absorption profiles of the MgH₂–1 wt% CAT–3 wt% C@Ni–MIX sample at 150, 100, and 75 ◦C.

Fig. 11. Hydrogen desorption and absorption cycle curves at 300 ◦C for the MgH₂–1 wt% CAT–0.5 wt% C@Ni–BM material.

R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

Fig. 12. JMAK plots of MgH2-3wt% CAT- 0.5 wt% C@Ni-BM (a), MgH2-1wt% CAT- 0.5 wt% C@Ni-BM (c), MgH2-3wt% CAT- 3 wt% C@Ni-MIX (e), MgH2-3wt% CAT- 3 wt% C@Ni-MIX(g). Arrhenius plots of MgH2-3wt% CAT- 0.5 wt% C@Ni-BM (b), MgH2-1wt% CAT- 0.5 wt% C@Ni-BM (d), MgH2-3wt% CAT- 3 wt% C@Ni- MIX (f), MgH2-3wt% CAT- 3 wt% C@Ni-MIX(h).

R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

direct process–structure–property link. The incorporation route controls the near-surface availability of C@Ni adjacent to defect-rich (Ti/Nb)Oₓ domains, which in turn governs nucleation barriers, diffusion lengths, and the measured onset temperatures, activation energies, and lowtemperature uptake.

Solar-driven platforms address the energy-input limitation of MgH₂ by integrating broad-band absorbers and catalytic constructs that deliver localized heating and facilitate H₂ activation and Mg–H bond scission. Representative studies report solar-driven reversible storage via Mg₂Ni(Cu) architectures and Cu@MXene photothermal layers. In parallel, light-enabled destabilization mechanisms in borohydrides leverage photogenerated vacancies to achieve hydrogen release and uptake under mild conditions. Our catalyst design is orthogonal and complementary to these approaches. It uses only commercial powders and high-energy milling or mixing to generate defect-rich Ti/Nb sub­ oxides and Ni-based sites, which already lower onset temperatures and activation energies under conventional thermal operation. Because these defect-rich oxides and Ni interfaces are compatible with photo­ thermal scaffolds, future integration of our optimized MIX composites with light-harvesting layers could further reduce energy input while preserving the scalability of our processing route. Beyond MXenes, several catalysts assembled from commercial ox­ ides, salts, or carbons have demonstrated strong kinetics in MgH₂, including TiO₂ quantum dots [57], Nb₂O₅ hollow spheres [58], Ni or Fig. 13. SEM micrographs of MgH₂–1 wt% CAT–0.5 wt% C@Ni–BM after 20 hydrogenation–dehydrogenation cycles. Table 2 Comparative summary of hydrogen-storage metrics from this study and previously reported MgH₂-based composites. Additives Activation energy for hydrogen desorption (kJ/mol) Lower dehydrogenation temperature, time, and hydrogen content Hydrogenation temperature, time, and hydrogen content Reference MgH2-1 wt% CAT- 3 wt% C@Ni-MIX

93.8

260 ◦C, 60 min, 2.36 wt%

150 ◦C, 50 min, 5.77 wt%

This work MgH2-1 wt% CAT- 0.5 wt% C@Ni-BM

108.1

260 ◦C, 60 min, 2.77 wt%

150 ◦C, 50 min, 3.80 wt%

This work TiO2 Quantum Dots ————————————

300 ◦C, 4.5 min, 6 wt%

100 ◦C, 30 min, 5 wt%

[57] Nb2O5 hollow spheres

101

250 ◦C, 20 min, 4 wt%

200 ◦C,20 s, 5 wt%

[58] Black TiO2 (K-TiO2-X)

69.1

260 ◦C, 60 min, 3 wt%

Room temperature,6 h, 5 wt% [31] CuMoO4

110.2

325 ◦C,10 min,6.03 wt%

200 ◦C,60 min, 5.7 wt%

[59] Ni nanoparticles dispersed on porous hollow carbon nanospheres

98

250 ◦C, 30 min, 6.2 wt%

150 ◦C, 250 s, 5.7 wt%

[60] flower-shaped niobium nitride (NbN) and 2D titanium carbide (Ti3C2) (MXene)

44.8

300 ◦C, 180 s, 5.9 wt%

100 ◦C, 60 min, 5.8 wt%

[61] Nano-Carbon supported CeO2 nanoparticles

101.9

320 ◦C, 50 min, 5.9 wt%

320 ◦C, 2 min, 5.6 wt%

[62] Ni-MOF@Pd

34.6

250 ◦C, 50 min, 1 wt%

150 ◦C, 300 s, 6.1 wt%

[63] Hydrangea-like NiO@KNbO3

71.8

285 ◦C,50 min, 4.2 wt%

150 ◦C, 5 min, 4.95 wt%

[64] KNbO3

104

300 ◦C,10 min, 3.9 wt%

300 ◦C,5 min, 5 wt%

[65] NiTiO3 nanoparticles ————————

275 ◦C,30 min, 6 wt%

150 ◦C, 60 min, 6 wt%

[66] Nb2C/Ni@MgH2 (MXene) ————————

200 ◦C,6 min, 6 wt%

200 ◦C,6 min, 7 wt%

[29] R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

NiO–niobate composites [77], and CeO₂@carbon systems [78]. These studies achieve rapid de/absorption under thermal operation, e.g., ~5.5 wt% at 300 ◦C in 5 min with Nb₂O₅ hollow spheres, or 5.5 wt% at

300 ◦C in 100 min for Co/CeO₂@C, but commonly require multi-step

fabrication of tailored morphologies or heterostructures. In contrast, our catalyst is produced solely from commercial powders by ball milling or mixing, yet it lowers the dehydrogenation onset to ≤236 ◦C, reduces Eₐ by ≥44 kJ mol⁻¹ , and enables low-temperature uptake down to

75 ◦C. We further note that high-performing MXene catalysts, although

valuable benchmarks, rely on HF-based etching and remain non-commercial at scale, which complicates deployment relative to the present, milling-only route based on commodity oxides and KH.

5. Conclusions

Magnesium hydride (MgH₂) remains limited by slow sorption ki­ netics and elevated operating temperatures. Here, we demonstrate a scalable catalytic strategy based entirely on commercial powders employing a KH-modified TiO₂/Nb₂O₅ catalyst (CAT) combined with carbon-coated Ni nanoparticles (C@Ni)—incorporated into MgH₂ via ball milling (BM) or a two-step mixing route (MIX). This catalyst system consistently lowers the dehydrogenation onset temperature from 321 ◦C (as-milled MgH₂) to ≤236 ◦C and reduces the activation energy by ≥44 kJ mol⁻¹ , to a minimum of 93.8 kJ mol⁻¹ , depending on compo­ sition and processing route.

The optimized composites also deliver fast kinetics at 300 ◦C (~6.4–6.5 wt% H₂ desorbed in ~20 min) and meaningful performance at lower temperatures (e.g., ~2.7 wt% in 60 min at 260 ◦C). Notably, the MIX route enables exceptional low-temperature hydrogen uptake, with the MgH₂–1 wt% CAT–3 wt% C@Ni–MIX sample absorbing

5.77 wt% at 150 ◦C (50 min) and maintaining activity down to 100 ◦C

(4.28 wt% in 120 min) and 75 ◦C (3.47 wt% in 120 min). Cycling tests show approximately 79% capacity retention after 20 cycles at 300 ◦C, indicating moderate durability with no evidence of contamination from ZrO₂ milling media.

Mechanistically, XPS and HRTEM reveal defect-rich reduced oxides (Ti³⁺/Nb⁴⁺) with abundant non-stoichiometric oxygen species, which, combined with Ni-assisted H₂ activation and spillover, rationalize the lower onset temperatures, reduced Ea, and enhanced low-temperature absorption.

Unlike many MgH₂ catalysts that rely on laboratory-specific or noncommercial materials, our methodology uses only commercial pre­ cursors and straightforward processing, yet achieves state-of-practice kinetic and thermal metrics. The combination of performance and scalability (together with the processing insight that the MIX route en­ hances low temperature uptake) positions this approach as a practical route toward deployable MgH₂-based hydrogen storage. CRediT authorship contribution statement Francisco J. Bolívar: Supervision, Project administration. Carlos Arrieta: Supervision, Project administration. F´elix Echeverria Eche­ verria: Writing – review & editing, Supervision, Project administration, Funding acquisition. Julian Arias Velandia: Software, Resources, Formal analysis. Robinson Aguirre Ocampo: Writing – review & editing, Writing – original draft, Investigation, Formal analysis. Ale­ jandro A. Zuleta Gil: Supervision, Project administration. Julian A. Lenis: Visualization, Data curation, Conceptualization. Esteban Cor­ rea: Writing – review & editing, Supervision, Project administration. Sindy Bello: Visualization, Data curation, Conceptualization. Declaration of Competing Interest 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.

Acknowledgements The authors are pleased to acknowledge the financial assistance of the "Sistema General de Regalías - SGR" through the project BPIN 2022000100089, Centro de Investigaci´on para el Desarrollo de la Innovaci´on (CIDI) from Universidad Pontificia Bolivariana (Rad: 822C- 06/23–35), Centro de Investigaci´on en Ingenierías (CEIN) from Uni­ versidad de Medellín and "Estrategia de Sostenibilidad de la Universidad de Antioquia". RAO and JAL were supported by COLCIENCIAS (currently Minciencias) PhD grants.

Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jallcom.2026.188174. Data availability Data will be made available on request References [1] X. Zhang, S. Ju, C. Li, J. Hao, Y. Sun, X. Hu, W. Chen, J. Chen, L. He, G. Xia, F. Fang, D. Sun, X. Yu, Atomic reconstruction for realizing stable solar-driven reversible hydrogen storage of magnesium hydride, Nat. Commun. 15 (2024), https://doi.org/10.1038/s41467-024-47077-y.

[2] R. Aguirre Ocampo, J. Arias-velandia, J.A. Lenis, A.A. Zuleta, S. Bello, E. Correa,

C. Arrieta, F. Bolívar, Francisco J. Echeverría Echeverría, Catalytic effect of

commercial carbon-coated nickel nanoparticles on the hydrogen storage performance of magnesium hydride, Int. J. Hydrog. Energy 129 (2025) 91–100, https://doi.org/10.1016/j.ijhydene.2025.04.226.

[3] G. Zhang, B. Sun, K. Xue, S. Ni, Y. Chen, X. Li, Research on the effect of NiCo/CNT addition on hydrogen storage properties of MgH2, J. Mater. Sci. 61 (2026) 11935–11946, https://doi.org/10.1007/s10853-026-12520-6. [4] J. Zhang, Y. Li, X. Ji, X. Li, Q. Xu, X. Zhou, X. Lu, P. Peng, Dual-phase FeCoNi1.4Mox high-entropy alloys with FCC/μ phases for enhanced hydrogen storage of MgH2, J. Alloy. Compd. 1057 (2026), https://doi.org/10.1016/j. jallcom.2026.186897.

[5] A. Razouk, M. EL Kassaoui, M. Boubkri, H. Abu-Farsakh, M. Bououdina, O. Mounkachi, Efficient hydrogen storage through thermodynamic and kinetic optimization of MgH2 (110) using Nb2O5 and V2O5 catalysts, J. Alloy. Compd. 1063 (2026) 187662, https://doi.org/10.1016/j.jallcom.2026.187662. [6] K. Reddad, H. Labrim, R.El Bouayadi, Predictive modeling of temperature programmed desorption (TPD) in magnesium hydride MgH2, Fuel 403 (2026) 136152, https://doi.org/10.1016/j.fuel.2025.136152. [7] X. Zhang, Y. Liu, Z. Ren, X. Zhang, J. Hu, Z. Huang, Y. Lu, M. Gao, H. Pan, Realizing

6.7 wt% reversible storage of hydrogen at ambient temperature with non-confined

ultrafine magnesium hydrides, Energy Environ. Sci. 14 (2021) 2302–2313, https:// doi.org/10.1039/d0ee03160g.

[8] H. Lu, J. Li, K. Nie, R. Zhang, Y. Chen, F. Pan, Halogen-terminated Ti3C2T MXenes as efficient catalysts for enhanced hydrogen storage kinetics of magnesium hydride, Sep. Purif. Technol. 377 (2025) 134126, https://doi.org/10.1016/j. seppur.2025.134126.

[9] C. Li, Y. Ding, X. Zhang, X. Hu, X. Yu, D. Sun, G. Xia, Hydrogen Storage in Magnesium Hydride at Room Temperature Enabled by Graphene-Stabilized Multivalent Niobium Oxides, Adv. Mater. 37 (2025) e11759, https://doi.org/ 10.1002/adma.202511759.

[10] D. Feng, Y. Wang, J. Zhang, H. Li, X. Jiang, L.W. Zhang, H. Li, Y. Zhang, Synergistic effect of ball milling duration on microstructural evolution and hydrogen storage kinetics of MgH2–VMnFeCoNi high-entropy alloy composites, Intermetallics 192 (2026) 109258, https://doi.org/10.1016/j.intermet.2026.109258. [11] Q. Zhang, Y. Li, F. Sun, X. Lin, Z. Hu, H. Yang, J. Zou, Boosting hydrogen storage performances of MgH2 by using a Ni-MOF derived Ni/NiO@C composite containing Ni/NiO nanoheterojunctions, Chem. Eng. J. 525 (2025) 169631, https://doi.org/10.1016/j.cej.2025.169631.

[12] C. Wang, X. Hou, Q. Cao, D. Zhao, D. Li, X. Xie, P. Zhu, X. Ye, G. Suo, G. Yang, G. Xu, Multi-component synergistic catalytic effect to enhance hydrogen storage kinetic of non-activated MgH2-Ni/Ti3C2Tx composites, J. Energy Storage 132 (2025), https://doi.org/10.1016/j.est.2025.117811. [13] J.S. Cortínez, A. G´omez, A.A. Zuleta, J.A. Tamayo, E. Correa, F.J. Bolívar, F. Echeverría, Improved hydrogen storage in magnesium thin flakes via nickel and titanium/titanium carbide additives, Int. J. Hydrog. Energy 153 (2025), https:// doi.org/10.1016/j.ijhydene.2025.150264.

[14] J.S. Cortinez, A. G´omez, A.A. Zuleta Gil, J.A. Tamayo, E. Correa, F.J. Bolívar, F. E. Echeverria, Production of Mg thin flakes with enhanced hydrogen storage performance, Int. J. Hydrog. Energy 71 (2024) 1191–1200, https://doi.org/ 10.1016/j.ijhydene.2024.05.254.

[15] X. Zhang, C. Li, J. Ye, X. Hu, W. Chen, F. Fang, D. Sun, Y. Liu, X. Yu, G. Xia, Light- Enabled Reversible Hydrogen Storage of Borohydrides Activated by R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

Photogenerated Vacancies, J. Am. Chem. Soc. 147 (2025) 2786–2796, https://doi. org/10.1021/jacs.4c15744.

[16] C. Liu, Z. Yuan, X. Li, Y. Sun, T. Zhai, Z. Han, L. Zhang, T. Li, Review on improved hydrogen storage properties of MgH2 by adding new catalyst, J. Energy Storage 97 (2024) 112786, https://doi.org/10.1016/j.est.2024.112786. [17] C. Lang, X. Yao, Enhancing hydrogen storage performance of magnesium-based materials: A review on nanostructuring and catalytic modification, J. Magnes. Alloy 13 (2025) 510–538, https://doi.org/10.1016/j.jma.2025.01.015. [18] X. Zhang, Y. Sun, S. Ju, J. Ye, X. Hu, W. Chen, L. Yao, G. Xia, F. Fang, D. Sun, X. Yu, Solar-Driven Reversible Hydrogen Storage, Adv. Mater. 35 (2023) 2206946, https://doi.org/10.1002/adma.202206946.

[19] J. Yang, J. Zheng, C. Wu, H. Shen, D. Tang, C. Su, P. Zhou, Nano-sized TiVH catalysts unlocking high-performance hydrogen storage in magnesium hydride, Fuel 423 (2026) 139308, https://doi.org/10.1016/j.fuel.2026.139308. [20] R. Aguirre Ocampo, J. Velandia-Arias, J.A. Lenis, A.A. Zuleta, S. Bello, E. Correa,

C. Arrieta, F.J. Bolívar, F. Echeverria Echeverria, Microwave ‑ assisted synthesis of

- MgH2 nanoparticles for hydrogen storage applications, J. Nanopart. Res. (2025)

1–14, https://doi.org/10.1007/s11051-025-06217-1.

[21] F. Liu, J. Wang, W. Chen, X. Zhang, S. Li, J. Li, Y.F. Song, J. Chu, G. Liu, In-situ formed Ti/TiH2 from exfoliated few-layered Ti3C2Tx as hydrogen pump enhances the hydrogen storage properties of MgH2, J. Colloid Interface Sci. 700 (2025) 138335, https://doi.org/10.1016/j.jcis.2025.138335.

[22] L.R. Xiao, C. Chen, S. Wang, Z.Y. Dai, H. Kimura, C. Ni, C.X. Hou, X.Q. Sun, S. J. Guo, W. Du, X.B. Xie, Effects of Ti- and Nb-based transition element from single to multiple compound oxides and carbon-based composite additives on Mg-MgH2 hydrogen storage material, Tungsten 7 (2025) 32–49, https://doi.org/10.1007/ s42864-024-00287-9.

[23] G.B. Pour, L.F. Aval, H. Nazarpour-Fard, Effectiveness of Nitrogen-Doping in MXene-Based Composites for Supercapacitor Electrodes, Results Chem. 13 (2025) 101965, https://doi.org/10.1016/j.rechem.2024.101965.

[24] G. Behzadi pour, L. Fekri aval, Recent advances in supercapacitors based on MXene surface modification: A review of symmetric and asymmetric electrodes material, Results Eng. 24 (2024) 103045, https://doi.org/10.1016/j.rineng.2024.103045. [25] Y. Zhao, B. Wang, L. Ren, Y. Li, X. Lin, Q. Zhang, Z. Hu, J. Zou, Nanostructured MXene-based materials for boosting hydrogen sorption properties of Mg/MgH2, Mater. Rep. Energy 4 (2024) 100255, https://doi.org/10.1016/j. matre.2024.100255.

[26] J. Deng, Y. Li, H. Ning, P. Qing, X. Huang, H. Luo, L. Zhang, G. Li, C. Huang, Z. Lan, W. Zhou, J. Guo, X. Wang, H. Liu, MXenes as catalysts for lightweight hydrogen storage materials: A review, Mater. Today Catal. 7 (2024) 100073, https://doi.org/ 10.1016/j.mtcata.2024.100073.

[27] Z. Yan, W. Zheng, G. Hao, Y. Wei, M. Luo, Recent advances in MXene catalyst towards enhanced hydrogen storage of Mg/MgH2: a review, Nanotechnology 36 (2025) 162001, https://doi.org/10.1088/1361-6528/adbb72. [28] W. Zhu, L. Ren, C. Lu, H. Xu, F. Sun, Z. Ma, J. Zou, Nanoconfined and in Situ Catalyzed MgH2 Self-Assembled on 3D Ti3C2 MXene Folded Nanosheets with Enhanced Hydrogen Sorption Performances, ACS Nano 15 (2021) 18494–18504, https://doi.org/10.1021/acsnano.1c08343.

[29] H. Liang, P.F. Zhang, Y. Wang, J. Zheng, Enhancing hydrogen storage performance of magnesium hydride with Ni cluster-loaded MXene-Nb2CTx (Tx = F, O), Rare Met (2025), https://doi.org/10.1007/s12598-025-03255-5.

[30] Q. Li, N. Miao, J. Zhong, C. Wu, H. Leng, Different effect between KH and KOH additives on the hydrogen storage properties of MgH2 doped with graphene, J. Alloy. Compd. 940 (2023) 168795, https://doi.org/10.1016/j. jallcom.2023.168795.

[31] M. Dai, J.A. Bolarin, G. Lei, Z. Li, T. He, H. Cao, P. Chen, Potassium hydride reduced black TiO2−x for boosting the hydrogenation of magnesium at room temperature, J. Alloy. Compd. 897 (2022) 162750, https://doi.org/10.1016/j. jallcom.2021.162750.

[32] H. Fu, S. Long, J. Hu, J. Li, H. Xue, Y. Chen, F. Pan, Situ Generated Cu/Nb Catalytic Interfaces for Enhancing MgH2 Hydrogen Storage, ACS Nano 20 (2026) 3904–3916, https://doi.org/10.1021/acsnano.5c21374.

[33] H. Wang, J. Sun, H. Shang, Breaking kinetic and thermodynamic barriers: CeO2-NirGO synergistic catalysis for comprehensively enhanced hydrogen storage performance of MgH2, J. Alloy. Compd. 1052 (2026) 186167, https://doi.org/ 10.1016/j.jallcom.2026.186167.

[34] Y. Liu, W. Sun, Z. Yang, R. Zhang, B. Li, L. Sun, F. Xu, W. Guo, C. Shi, T. Yu, Z. Liang, Q. He, The doping of coral-ball-shaped heterojunction NiO/NiFe2O4 as a catalyst for MgH2: improvement of hydrogen storage performance, Int. J. Hydrog. Energy 201 (2026) 152959, https://doi.org/10.1016/j.ijhydene.2025.152959. [35] T. Kopac, Exploring the impact of MXene-based materials on hydrogen storage performance: Experimental insights, applications, and challenges, Mater. Today Sustain 31 (2025) 101173, https://doi.org/10.1016/j.mtsust.2025.101173. [36] C.E. Shuck, Y. Gogotsi, Taking MXenes from the lab to commercial products, Chem. Eng. J. 401 (2020) 125786, https://doi.org/10.1016/j.cej.2020.125786. [37] Y. Yang, X. Zhang, L. Zhang, W. Zhang, H. Liu, Z. Huang, L. Yang, C. Gu, W. Sun, M. Gao, Y. Liu, H. Pan, Recent advances in catalyst-modified Mg-based hydrogen storage materials, J. Mater. Sci. Technol. 163 (2023) 182–211, https://doi.org/ 10.1016/j.jmst.2023.03.063.

[38] J. Arias, R. Aguirre-Ocampo, J.S. Cortinez, C. Arrieta, F. Bolivar, A. Zuleta, J. A. Tamayo, A.F. Vargas, E. Correa, F. Echeverria, Development of a sieverts-type apparatus with enhanced automation using a dual-valve pressure control device for solid state hydrogen storage applications, Int. J. Hydrog. Energy 220 (2026) 154042, https://doi.org/10.1016/j.ijhydene.2026.154042.

[39] A. Miyakoshi, A. Ueno, M. Ichikawa, XPS and TPD characterization of manganesesubstituted iron-potassium oxide catalysts which are selective for dehydrogenation of ethylbenzene into styrene, Appl. Catal. A Gen. 219 (2001) 249–258, https://doi. org/10.1016/S0926-860X(01)00697-4.

[40] R. Sawyer, H.W. Nesbitt, R.A. Secco, High resolution X-ray Photoelectron Spectroscopy (XPS) study of K 2O-SiO2 glasses: Evidence for three types of O and at least two types of Si, J. Non Cryst. Solids 358 (2012) 290–302, https://doi.org/ 10.1016/j.jnoncrysol.2011.09.027.

[41] C. Zhou, Y. Zhao, L. Shang, Y. Cao, L.Z. Wu, C.H. Tung, T. Zhang, Facile preparation of black Nb4+ self-doped K4Nb6O17 microspheres with high solar absorption and enhanced photocatalytic activity, Chem. Commun. 50 (2014) 9554–9556, https://doi.org/10.1039/c4cc04432k.

[42] M.V. Nogueira, G.M.M.M. Lustosa, Y. Kobayakawa, W. Kogler, M. Ruiz, E. S. Monteiro Filho, M.A. Zaghete, L.A. Perazolli, Nb-Doped TiO2 Photocatalysts Used to Reduction of CO2 to Methanol, Adv. Mater. Sci. Eng. 2018 (2018), https:// doi.org/10.1155/2018/7326240.

[43] Y. Wang, S. Aghamohammadi, D. Li, K. Li, R. Farrauto, Structure dependence of Nb2O5-X supported manganese oxide for catalytic oxidation of propane: Enhanced oxidation activity for MnOx on a low surface area Nb2O5-X, Appl. Catal. B Environ. 244 (2019) 438–447, https://doi.org/10.1016/j.apcatb.2018.11.066. [44] Z. Li, H. Li, S. Wang, F. Yang, W. Zhou, Mesoporous black TiO2/MoS2/Cu2S hierarchical tandem heterojunctions toward optimized photothermalphotocatalytic fuel production, Chem. Eng. J. 427 (2022), https://doi.org/ 10.1016/j.cej.2021.131830.

[45] M.C. Biesinger, L.W.M. Lau, A.R. Gerson, R.S.C. Smart, Resolving surface chemical states in XPS analysis of first row transition metals, oxides and hydroxides: Sc, Ti, V, Cu and Zn, Appl. Surf. Sci. 257 (2010) 887–898, https://doi.org/10.1016/j. apsusc.2010.07.086.

[46] B. Sun, W. Zhou, H. Li, L. Ren, P. Qiao, W. Li, H. Fu, Synthesis of Particulate Hierarchical Tandem Heterojunctions toward Optimized Photocatalytic Hydrogen Production, Adv. Mater. 30 (2018) 1–8, https://doi.org/10.1002/ adma.201804282.

[47] L.Bin Xiong, J.L. Li, B. Yang, Y. Yu, Ti 3+ in the surface of titanium dioxide: Generation, properties and photocatalytic application, J. Nanomater 2012 (2012), https://doi.org/10.1155/2012/831524.

[48] J. Wang, X. Wang, J. Yan, Q. Tan, G. Liang, S. Qu, Z. Zhong, Enhanced photoelectrochemical properties of Ti3+ self-doped branched TiO2 nanorod arrays with visible light absorption, Mater. (Basel) 11 (2018), https://doi.org/10.3390/ ma11101791.

[49] V.A. Online, H. Yap, N.S. Mustafa, M. Ismail, RSC Adv. (2014), https://doi.org/ 10.1039/C4RA12487A.

[50] A.F. Zedan, S. Gaber, A.S. Aljaber, CO Oxid. -Ambient Temp. TiO 2 -Support. Pd-Cu Catal. Promot. Eff. Pd-Cu Nanointerface TiO 2 Morphol. (2021). [51] G.I. Supelano, F. Mesa, C.A.P. Vargas, J.A.M. G´omez, A. Dussan, Assessment of surface and electrical properties of the TiO2@zeolite hybrid materials, Sci. Rep. 13 (2023) 1–15, https://doi.org/10.1038/s41598-023-30529-8. [52] A. Trenczek-Zajac, M. Synowiec, K. Zakrzewska, K. Zazakowny, K. Kowalski, A. Dziedzic, M. Radecka, Scavenger-Supported Photocatalytic Evidence of an Extended Type I Electronic Structure of the TiO2@Fe2O3Interface, ACS Appl. Mater. Interfaces 14 (2022) 38255–38269, https://doi.org/10.1021/ acsami.2c06404.

[53] R. Sultana, K. Islam, S. Chakraborty, Tuning Optical and Electrochemical Properties of Nb2O5 Thin Films via WO3 Doping, Trans. Electr. Electron. Mater. 26 (2025) 48–59, https://doi.org/10.1007/s42341-024-00572-x. [54] M.E. Bowden, G.V. White, I.W.M. Brown, M.J. Ryan, G.J. Gainsford, Improved powder diffraction patterns for the titanium suboxides, 11 (1996) 60–68. [55] R. Aguirre O, F. Echeverría E, Effects of fluoride source on the characteristics of titanium dioxide nanotubes, Appl. Surf. Sci. 445 (2018) 308–319, https://doi.org/ 10.1016/j.apsusc.2018.03.139.

[56] R.A. Ocampo, M. Echeverry-Rend´on, I. DeAlba-Montero, S. Robledo, F. Ruiz, F. Echeverría Echeverría, Effect of surface characteristics on the antibacterial properties of TiO2 nanotubes produced in aqueous electrolytes with carboxymethyl cellulose, J. Biomed. Mater. Res. Part A 109 (2021) 104–121, https://doi.org/10.1002/jbm.a.37010.

[57] S.K. Pandey, A. Bhatnagar, V. Shukla, R. Kesarwani, U. Deshpandey, T.P. Yadav, Catalytic mechanism of TiO2 quantum dots on the de/re-hydrogenation characteristics of magnesium hydride, Int. J. Hydrog. Energy 46 (2021) 37340–37350, https://doi.org/10.1016/j.ijhydene.2021.09.006. [58] X. Zhang, K. Wang, X. Zhang, J. Hu, M. Gao, H. Pan, Y. Liu, Synthesis process and catalytic activity of Nb2O5 hollow spheres for reversible hydrogen storage of MgH2, Int. J. Energy Res. 45 (2021) 3129–3141, https://doi.org/10.1002/er.6006. [59] Y. Pan, X. Yang, X. Lu, J. Su, J. Kong, D. Huo, W. Zhang, Q. Hou, Enhancement catalysis of layered CuMoO4 on hydrogen storage performance of MgH2, J. Alloy. Compd. 1017 (2025), https://doi.org/10.1016/j.jallcom.2025.179129. [60] S. Wang, M. Gao, Z. Yao, K. Xian, M. Wu, Y. Liu, W. Sun, H. Pan, High-loading, ultrafine Ni nanoparticles dispersed on porous hollow carbon nanospheres for fast (de)hydrogenation kinetics of MgH2, J. Magnes. Alloy 10 (2022) 3354–3366, https://doi.org/10.1016/j.jma.2021.05.004.

[61] Z. Yang, J. Wu, Y. Wang, S. Wang, Y. Zou, C. Xiang, F. Xu, L. Sun, Y.S. Chua, Improved hydrogen storage performance of magnesium hydride catalyzed by two dimensional Ti3C2-coated NbN, J. Alloy. Compd. 1029 (2025) 180752, https:// doi.org/10.1016/j.jallcom.2025.180752.

[62] C. Liu, Z. Yuan, W. Li, T. Zhai, Z. Han, X. Zhang, Improved hydrogen storage performance of MgH2 by adding CeO2@NC with high catalytic activity, J. Mater. Sci. Technol. (2025), https://doi.org/10.1016/j.jmst.2025.02.066. [63] Z.Y. Li, L.X. Sun, F. Xu, Y.M. Luo, Y.P. Xia, S. Wei, C.C. Zhang, R.G. Cheng, C.F. Ye, M.Y. Liu, J.L. Zeng, Z. Cao, H.G. Pan, Modulated noble metal/2D MOF R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174

heterostructures for improved hydrogen storage of MgH2, Rare Met 43 (2024) 1672–1685, https://doi.org/10.1007/s12598-023-02496-6.

[64] J. Wu, H. Zhang, Y. Wang, Y. Zou, B. Li, C. Xiang, L. Sun, F. Xu, T. Yu, Hydrangealike NiO@KNbO3 as catalyst doping for MgH2: Boosting hydrogen storage performance, J. Alloy. Compd. 961 (2023) 171160, https://doi.org/10.1016/j. jallcom.2023.171160.

[65] M.H. Abdul Rahman, M.A. Shamsudin, A. Klimkowicz, S. Uematsu, A. Takasaki, Effects of KNbO3 catalyst on hydrogen sorption kinetics of MgH2, Int. J. Hydrog. Energy 44 (2019) 29196–29202, https://doi.org/10.1016/j.ijhydene.2019.02.186. [66] N. Yan, X. Lu, Z. Lu, H. Yu, F. Wu, J. Zheng, X. Wang, L. Zhang, Enhanced hydrogen storage properties of Mg by the synergistic effect of grain refinement and NiTiO3 nanoparticles, J. Magnes. Alloy 10 (2022) 3542–3552, https://doi.org/10.1016/j. jma.2021.03.014.

[67] C. Peng, Y. Li, Q. Zhang, Enhanced hydrogen desorption properties of MgH2 by highly dispersed Ni: The role of in-situ hydrogenolysis of nickelocene in ball milling process, J. Alloy. Compd. 900 (2022) 163547, https://doi.org/10.1016/j. jallcom.2021.163547.

[68] D. Tan, C. Peng, Q. Zhang, Microstructural characteristics and hydrogen storage properties of the Mg–Ni–TiS2 nanocomposite prepared by a solution-based method, Int. J. Hydrog. Energy (2023) 1–13, https://doi.org/10.1016/j. ijhydene.2023.01.185.

[69] C. Yin, S. Qiu, Y. Xia, F. Xu, L. Sun, H. Chu, Ni-based catalyst assisted by MnO to boost the hydrogen storage performance of magnesium hydride, Int. J. Hydrog. Energy (2023) 1–9, https://doi.org/10.1016/j.ijhydene.2023.11.213. [70] C. Yin, S. Qiu, Y. Wang, Q. Wei, Z. Peng, Y. Xia, Y. Zou, F. Xu, L. Sun, H. Chu, Promoted hydrogen storage properties of MgH2 by Ti3+ self-doped defectmediated TiO2, J. Alloy. Compd. 966 (2023) 1–10, https://doi.org/10.1016/j. jallcom.2023.171610.

[71] C. Zheng, B. Liu, D. Zhou, Z. Hou, D. Zhao, A systematic review on nanostructure engineering of Mg/MgH2 for enhanced hydrogen storage performance, Int. J. Hydrog. Energy 160 (2025) 150629, https://doi.org/10.1016/j. ijhydene.2025.150629.

[72] D. Zhou, H. Sun, S. Guo, D. Zhao, J. Li, Y. Zhang, Hydrogen storage properties of Mg-based alloys modified with metal-organic frameworks and carbon-based porous materials: A review and summary, Int. J. Hydrog. Energy 57 (2024) 1373–1388, https://doi.org/10.1016/j.ijhydene.2024.01.127.

[73] C. Yin, S. Qiu, Y. Wang, Q. Wei, Z. Peng, Y. Xia, Y. Zou, F. Xu, L. Sun, H. Chu, Promoted hydrogen storage properties of MgH2 by Ti3+ self-doped defectmediated TiO2, J. Alloy. Compd. 966 (2023) 1–10, https://doi.org/10.1016/j. jallcom.2023.171610.

[74] L. Ren, W. Zhu, Y. Li, X. Lin, H. Xu, F. Sun, C. Lu, J. Zou, Oxygen Vacancy-Rich 2D TiO2 Nanosheets: A Bridge Toward High Stability and Rapid Hydrogen Storage Kinetics of Nano-Confined MgH2, Nano-Micro Lett. 14 (2022) 1–16, https://doi. org/10.1007/s40820-022-00891-9.

[75] L. Ren, Y. Li, Z. Li, X. Lin, C. Lu, W. Ding, J. Zou, Boosting Hydrogen Storage Performance of MgH2 by Oxygen Vacancy-Rich H-V2O5 Nanosheet as an Excited H-Pump, Nano-Micro Lett. 16 (2024), https://doi.org/10.1007/s40820-024-

01375-8.

[76] H. Zhang, Q. Kong, S. Hu, D. Zhang, H. Chen, C.C. Xu, B. Li, Y. Fan, B. Liu, Engineering the Oxygen Vacancies in Na2Ti3O7 for Boosting Its Catalytic Performance in MgH2 Hydrogen Storage, ACS Sustain. Chem. Eng. 10 (2022) 363–371, https://doi.org/10.1021/acssuschemeng.1c06444. [77] M.C. Brum, M.O.T. Da Conceiç˜ao, P.M. Jardim, D.S. Dos Santos, The use of onedimensional Niobate to improve MgH2 hydrogen sorption, J. Alloy. Compd. 615 (2015) S698–S701, https://doi.org/10.1016/j.jallcom.2013.12.148. [78] H. Yu, Z. Yuan, Q. Han, J. Li, T. Li, Effect of high-activity Co/CeO₂@C on hydrogen storage kinetics of MgH2, J. Alloy. Compd. 1053 (2026), https://doi.org/10.1016/ j.jallcom.2026.186245.

R.A. Ocampo et al.

Journal of Alloys and Compounds 1065 (2026) 188174