Ni0.6Zn0.4O Synthesised via a Solid-State Method for Promoting Hydrogen Sorption from MgH2

Magnesium hydrides (MgH2) have drawn a lot of interest as a promising hydrogen storage material option due to their good reversibility and high hydrogen storage capacity (7.60 wt.%). However, the high hydrogen desorption temperature (more than 400 °C) and slow sorption kinetics of MgH2 are the main obstacles to its practical use. In this research, nickel zinc oxide (Ni0.6Zn0.4O) was synthesized via the solid-state method and doped into MgH2 to overcome the drawbacks of MgH2. The onset desorption temperature of the MgH2–10 wt.% Ni0.6Zn0.4O sample was reduced to 285 °C, 133 °C, and 56 °C lower than that of pure MgH2 and milled MgH2, respectively. Furthermore, at 250 °C, the MgH2–10 wt.% Ni0.6Zn0.4O sample could absorb 6.50 wt.% of H2 and desorbed 2.20 wt.% of H2 at 300 °C within 1 h. With the addition of 10 wt.% of Ni0.6Zn0.4O, the activation energy of MgH2 dropped from 133 kJ/mol to 97 kJ/mol. The morphology of the samples also demonstrated that the particle size is smaller compared with undoped samples. It is believed that in situ forms of NiO, ZnO, and MgO had good catalytic effects on MgH2, significantly reducing the activation energy and onset desorption temperature while improving the sorption kinetics of MgH2.


Introduction
Due to its enormous energy density (142 MJ/kg), abundance, and completely clean combustion, hydrogen is gaining more attention as an alternative energy carrier [1,2]. However, the limited availability of effective storage solutions has prevented the widespread use of hydrogen. The three conventional systems for storing hydrogen are cryogenic liquid storage (5-10 bar, 253 • C), compressed gas storage (350-700 bar at ambient temperature), and solid-state storage [3,4]. For solid-state storage, hydrogen can be stored in a chemical hydride such as ammonia borane [5,6] or in metal hydrides such as MgH 2 , LiAlH 4 , NaAlH 4 , and other materials and is expected to have a high hydrogen capacity [7][8][9]. Nevertheless, MgH 2 is appealing because of its abundance of resources, cheapness, and high gravimetric capacity (7.60 wt.%) [10][11][12]. The practical applications for MgH 2 were still lacking because of sluggish kinetics, a high temperature (more than 400 • C), and a high dissociation enthalpy (∆H = −74.5 kJ/mol) [13,14]. Several attempts have been conducted to overcome the drawbacks of MgH 2 , such as using the ball milling technique (to create smaller particles size) and doping with additives/catalysts including transition metals (Cu, Nb, Ti, Zn, Ni, and Co) and their compounds (likes carbides, fluorides, oxides and hydrides), nonmetallic materials (such as carbon nanotubes, graphene, carbon, and graphite), and intermetallics [15][16][17][18].
Ni is one of the effective catalysts for the MgH 2 system. A previous study revealed that by synthesizing the Ni@rGO catalyst, the desorption temperature of MgH 2 + 10 wt.% Ni 4 @rGO 6 samples decreased by 61 • C [19]. Besides that, the MgH 2 + 10 wt.% Ni 4 @rGO 6 samples are capable of absorbing 5.00 wt.% H 2 in 20 min at 100 • C and desorbing 6.10 wt.% of

Materials and Methods
For the first part, Ni 0.6 Zn 0.4 O was synthesized by the solid-state method by using Ni (≥99% pure; Sigma Aldrich, St. Louis, MO, USA), citric acid (≥98% pure; Sigma Aldrich), and sinc oxide (<100 nm; Sigma Aldrich). All of these materials were ground together for 15 min using the following amounts: 0.1195 g of Ni, 0.1521 g of citric acid and 0.0326 g of zinc oxide. The sample was then calcined for 1 h at 1000 • C.
Next, Ni 0.6 Zn 0.4 O was used as an additive in order to improve the hydrogen storage performance of MgH 2 . For this step, all the handling processes, including weighing, were completed in a glove box (MBRAUN UNIlab) with a pure argon atmosphere to prevent oxidation. The different weight percentages of Ni 0.6 Zn 0.4 O samples were milled together by using a planetary ball mill (NQM-0.4) to produce MgH 2 -X wt.% Ni 0.6 Zn 0.4 O samples (where X = 5, 10, 15, and 20). In this experiment, commercial MgH 2 was acquired from Sigma Aldrich (≥95% pure). The sample was milled at 400 rpm for 1 h (15 min of milling time, 2 min of resting time, and 3 cycles) at room temperature. Each milling consists of four balls made of steel, and the ball-to-powder weight ratio is equal to 40:1.
To analyze the onset desorption temperature and absorption/desorption kinetics of the samples, Sievert-type pressure composition temperature (Advanced Materials Corporation, Pittsburgh, PA, USA) was used. The samples were heated up to 450 • C from room temperature. Meanwhile, 33.0 atm and 1.0 atm of pressure were used for the absorption/desorption kinetics process, which was carried out at 250 • C and 300 • C, respectively. The differential scanning calorimetry (DSC) was examined using a Mettler Toledo (Columbus, OH, USA) TG/DSC 1 in an Argon gas flow at 50 mL/min with various heating rates (15,20,25 and 30 • C/min) applied. About 3-5 mg of samples were loaded into an alumina crucible and heated from room temperature to 450 • C.
Structural characterization was performed with the help of the X-ray diffraction (XRD; Rigaku Miniflex, Tokyo, Japan) technique with a Cu-kα radiation range of 20-80 • and a speed of 2.00 • /min. The morphology of the composite was characterized by using scanning electron microscopy (SEM; JEOL JSM-6360LA) and energy dispersive X-ray spectroscopy (EDS; JEOL JSM-6360LA). The bonding of the samples was investigated using Fourier transform infrared spectroscopy (IR Shimadzu Tracer-100, Kyoto, Japan). Each FTIR data were obtained by averaging 40 scans from 400 to 2700 cm −1 , and Renishaw Raman spectroscopy was conducted at room temperature with a 0.1% power laser measurement. Pure MgH 2 and milled MgH 2 were also characterized by using all the instruments to compare the results between the samples.

Results and Discussion
The XRD spectra of Ni 0.6 Zn 0.4 O samples prepared by the solid-state method are presented in Figure 1a. Referring to the reported standard Ni 0.6 Zn 0.4 O sample (JCPDF 75-0273), all reported diffraction peaks match it perfectly. Zinc oxide and nickel oxide, two potential impurity phases, were not found in the XRD spectra. The diffraction peaks at 36 [37]. The crystallite sizes (L) are estimated at 11.24 nm through the Scherrer formula as in Equation (1) below: where λ is the X-ray used (0.154 nm), β (physical broadening) is the full width at half the maximum, θ is the angle of Bragg's diffraction, and shape factor K = 0.94 constant. Figure 1b depicts the FTIR spectra of Ni 0.6 Zn 0.4 O, and the peaks at 418 cm −1 correspond to the Ni-O bond as suggested in the previous study [38,39]. Meanwhile, the peaks at 502 cm −1 are attributed to the Zn-O peaks as indicated by Raja et al. [40] and Handore et al. [41]. Furthermore, Raman spectra of Ni 0.6 Zn 0.4 O samples were present as in Figure 1c, clearly showing 3 distinct Raman bands. The peaks at 351 cm −1 and 401 cm −1 were attributed to the Zn-O peak, as proven by Bhunia et al. [42] and Marinho et al. [43]. Meanwhile, the peak at 481 cm −1 matches the Ni-O peak as exposed by Bose et al. [44]. O bond as suggested in the previous study [38,39]. Meanwhile, the peaks at 502 cm −1 are attributed to the Zn-O peaks as indicated by Raja et al. [40] and Handore et al. [41]. Furthermore, Raman spectra of Ni0.6Zn0.4O samples were present as in Figure 1c, clearly showing 3 distinct Raman bands. The peaks at 351 cm −1 and 401 cm −1 were attributed to the Zn-O peak, as proven by Bhunia et al. [42] and Marinho et al. [43]. Meanwhile, the peak at 481 cm −1 matches the Ni-O peak as exposed by Bose et al. [44]. EDS characterization for Ni0.6Zn0.4O was conducted as shown in Figure 2 in order to recognize the existence and distribution of Ni0.6Zn0.4O. To evaluate the element distribution over a broad region, the magnification was set at 500×. The EDS mapping below shows the distribution of different elements. Figure 2a displays the Ni0.6Zn0.4O samples; Figure 2b,c show the Ni and Zn elements, respectively. However, Figure 2d illustrates the O element. From the result obtained, it is shown that elements Ni, Zn, and O are uniformly distributed. Table 1 below proved the element of the Ni0.6Zn0.4O. From the results of XRD, FTIR, Raman, and EDS mapping, it is proved that pure Ni0.6Zn0.4O was successfully synthesized by the solid-state method. The SEM images as presented in Figure 2e indicated that Ni0.6Zn0.4O has a spherical morphology and the particles were agglomerate. Meanwhile, Figure 2f shows the results of the particle size distribution (PSD) analysis. The PSD was analyzed using Image J. Most of the particles are concentrated at a size of approximately 98.6 µm.  Figure 2f shows the results of the particle size distribution (PSD) analysis. The PSD was analyzed using Image J. Most of the particles are concentrated at a size of approximately 98.6 µm. Figure 3a exhibited the temperature-programmed desorption curves of pure MgH 2 , milled MgH 2 , and MgH 2 doped with different weight percentages (5, 10, 15, and 20) [45] stated that the total desorption capacity of MgH 2 -TiF 2 is less than MgH 2 because TiH 2 somehow does not evolve H 2 and acts as a dead weight for the MgH 2 -TiH 2 system.   Figure 3a exhibited the temperature-programmed desorption curves of pure MgH2, milled MgH2, and MgH2 doped with different weight percentages (5, 10, 15, and 20) of Ni0.6Zn0.4O. The pure MgH2 decomposes at 418 °C with an approximate 7.10 wt.% total dehydrogenation capacity. Milling MgH2 for 1 h lowered the onset desorption temperature to 341 °C, 77 °C lower than pure MgH2. Remarkably, the onset desorption temperature reduces after the Ni0.6Zn0.4O additive is added. MgH2-5 wt.% Ni0.6Zn0.4O samples decompose at 280 °C with a 6.80 wt.% total H2 release. After the addition of 10 wt.%, 15 wt.%, and 20 wt.% of Ni0.6Zn0.4O as an additive with MgH2, the initial desorption temperatures were lowered to 285 °C, 305 °C, and 293 °C, respectively. In addition, as the amount of Ni0.6Zn0.4O increased to 10 wt.%, 15 wt.% and 20 wt.%, the total hydrogen release declined to 6.80 wt.%. 6.50 wt.% and 6.30 wt.%, respectively. Numerous studies have shown that this trend was caused by the dead weight of the Ni0.6Zn0.4O. Research conducted by Bhatnagar et al. [45] stated that the total desorption capacity of MgH2-TiF2 is less than MgH2 because TiH2 somehow does not evolve H2 and acts as a dead weight for the MgH2-TiH2 system.
The absorption kinetics was conducted at 250 °C for 1 h, and the result also shows that the addition of the Ni0.6Zn0.4O additive enhances the performance of MgH2, as proved in Figure 3b. The results showed that within only 5 min, milled MgH2 was able to absorb 4.80 wt.% of H2. The addition of 10 wt.%, and 15 wt.% of Ni0.6Zn0.4O with MgH2 increased their absorption capacity to 6.50 wt.% of H2 in the same amount of time. A slight increment in the absorption capacity for MgH2-20 wt.% Ni0.6Zn0.4O samples can be observed, which  The absorption kinetics was conducted at 250 • C for 1 h, and the result also shows that the addition of the Ni 0.6 Zn 0.4 O additive enhances the performance of MgH 2, as proved in Figure 3b. The results showed that within only 5 min, milled MgH 2 was able to absorb 4.80 wt.% of H 2 . The addition of 10 wt.%, and 15 wt.% of Ni 0.6 Zn 0.4 O with MgH 2 increased their absorption capacity to 6.50 wt.% of H 2 in the same amount of time. A slight increment in the absorption capacity for MgH 2 -20 wt.% Ni 0.6 Zn 0.4 O samples can be observed, which is 5.40 wt.%. However, MgH 2 -5 wt.% Ni 0.6 Zn 0.4 O samples showed the lowest amount of absorption capacity, which is 4.10 wt.% under the same circumstances. The absorption kinetics of MgH 2 with another catalyst were also included for comparison purposes, as shown in Table 2   Using kinetic models to represent the behavior of absorption and desorption is a great idea to gain a better understanding of the kinetic mechanism in MgH2-10 wt.% Ni0.6Zn0.4O samples. In 2010, Luo et al. [50] investigated two kinds of kinetic models, which are the Jander model and the Chou model, on the hydriding kinetics of Mg-Ni based alloys. For instance, Cheng et al. [51] proposed a kinetic model based on the characteristics of desorption time for TiVNbCr alloy using the Jander diffusion model, the Ginstling-Brounshtein model, and the Johnson-Mehl-Avrami-Kolmogorov (JMA) equation. Furthermore, JMA plots of the Mg90Ce5Y5 alloy with various catalysts such as MoO3, MoO2, and Mo were explored by Wang and co-workers [52]. In this study, the kinetic models of JMA and Contracting Volume (CV) were analyzed as can be seen in Table 4 [53]. According to Pang and Li [54], these models were chosen because they accurately fit the experimental data and did not require any additional approximations or assumptions. Additionally, these models have been used by other researchers to comprehend the ratelimiting steps of the material.   [25], the Mg-H bond was significantly stretched by the Ni catalyst action, which is more favorable for H separation and can speed up the desorption rate of MgH 2 . It is clearly apparent that introducing Ni 0.6 Zn 0.4 O as an additive will significantly reduce the onset desorption temperature and enhance the absorption/desorption kinetics of MgH 2 as summarized in Table 3 below. Pure MgH 2 and milled MgH 2 were also included for comparison. Considering the influence of the Ni 0.6 Zn 0.4 O as an additive on the onset desorption temperature and sorption kinetics, MgH 2 -10 wt.% Ni 0.6 Zn 0.4 O samples as an additive were selected for further study. Using kinetic models to represent the behavior of absorption and desorption is a great idea to gain a better understanding of the kinetic mechanism in MgH 2 -10 wt.% Ni 0. 6 [52]. In this study, the kinetic models of JMA and Contracting Volume (CV) were analyzed as can be seen in Table 4 [53]. According to Pang and Li [54], these models were chosen because they accurately fit the experimental data and did not require any additional approximations or assumptions. Additionally, these models have been used by other researchers to comprehend the rate-limiting steps of the material. Table 4. Equation for kinetic models used for absorption and desorption kinetics of this study.

Integrated Equation Model
A = kt Surface-controlled (chemisorption)  Table 4 determined the ratelimiting steps as shown in Figure 4a,b, respectively. The kinetic curves for the samples were measured for the reacted fraction in the range of 0 to 80%. As shown in the following figure, the CV 3D decrease surface can best explain the absorption and desorption kinetics at 250 • C and 300 • C, respectively.
The DSC curves for milled MgH 2 and MgH 2 -10 wt.% Ni 0.6 Zn 0.4 O samples were evaluated at different heating rates, as represented in Figure 5a and 5b, respectively. One endothermic peak is visible in both samples, indicating the decomposition of MgH 2 to Mg. Increasing the heating rates resulted in an increase in the temperature of the samples. 1 − (2α/3) − (1 − α) 2/3 = kt diffusion controlled with decreasing interface velocity Where t is time, k is a reaction rate constant and α is reacted fraction.
In this context, the best linear plot of the absorption and desorption kinetics of MgH2-10 wt.% Ni0.6Zn0.4O samples with the kinetic equations in Table 4 determined the ratelimiting steps as shown in Figure 4a,b, respectively. The kinetic curves for the samples were measured for the reacted fraction in the range of 0 to 80%. As shown in the following figure, the CV 3D decrease surface can best explain the absorption and desorption kinetics at 250 °C and 300 °C, respectively.  Table 4 for (a) absorption kinetics at 250 °C and (b) desorption kinetics at 300 °C.
The DSC curves for milled MgH2 and MgH2-10 wt.% Ni0.6Zn0.4O samples were evaluated at different heating rates, as represented in Figure 5a and 5b, respectively. One endothermic peak is visible in both samples, indicating the decomposition of MgH2 to Mg. Increasing the heating rates resulted in an increase in the temperature of the samples. For comparison, DSC traces at 20 °C/min for milled MgH2 and MgH2-10 wt.% Ni0.6Zn0.4O samples were examined as in Figure 5c. From the result obtained, the temperature for milled MgH2 was 428 °C, while the MgH2-10 wt.% Ni0.6Zn0.4O samples were 397 °C. It is noticeable that the Ni0.6Zn0.4O additive affected the endothermic peak of hydrogen desorption to shift remarkably to a lower temperature. Besides, it was observed that the inclusion of Mg(Nb)O resulted in a reduction in the endothermic peak of MgH2, which is due to the weakening of Mg-H bonds caused by Mg(Nb)O [55].  Table 4 for (a) absorption kinetics at 250 • C and (b) desorption kinetics at 300 • C.  The remarkable effect of the Ni0.6Zn0.4O additive on the desorption kinetic properties of MgH2 was further examined by calculating the apparent activation energy (EA) using the Kissinger equation below (Equation (2)): where Tp is the peak temperature in the DSC curve, β is the heating rate of the samples, R is the gas constant, and A is a linear constant. Figure 6 revealed the Kissinger plots of the The remarkable effect of the Ni 0.6 Zn 0.4 O additive on the desorption kinetic properties of MgH 2 was further examined by calculating the apparent activation energy (E A ) using the Kissinger equation below (Equation (2)): where T p is the peak temperature in the DSC curve, β is the heating rate of the samples, R is the gas constant, and A is a linear constant. Figure 6 Figure 7a. A similar outcome was discovered by Mahsa et al. [60]. They exposed that the morphology of pure MgH2 has irregular shapes with larger particles. It should be noted that smaller particle sizes can be observed after MgH2 is milled for 1 h, as presented in Figure 7b. This proved that the performance of MgH2 was also directly affected by the milling process. Next, changes in the morphological parameters of the powder can also be detected by Czujko et al. [61]. According to Shahi et al. [62], the onset desorption temperature of pure MgH2 decreased from 422 °C to 367 °C. It may be pointed out that the milling process of MgH2 for 25 h reduces the particle size of MgH2, thereby lowering the desorption temperature of MgH2. As expected, MgH2-10 wt.% Ni0.6Zn0.4O samples exhibited a smaller particle size as compared with milled MgH2 (as can be seen in Figure 7c). Ali et al. [56] introduced CoTiO3 to MgH2 and showed that the particle size of the composite changed to a finer and smaller size. According to the research results of Somo et al. [63], smaller particle sizes allow quick   Figure 7a. A similar outcome was discovered by Mahsa et al. [60]. They exposed that the morphology of pure MgH 2 has irregular shapes with larger particles. It should be noted that smaller particle sizes can be observed after MgH 2 is milled for 1 h, as presented in Figure 7b. This proved that the performance of MgH 2 was also directly affected by the milling process. Next, changes in the morphological parameters of the powder can also be detected by Czujko et al. [61]. According to Shahi et al. [62], the onset desorption temperature of pure MgH 2 decreased from 422 • C to 367 • C. It may be pointed out that the milling process of MgH 2 for 25 h reduces the particle size of MgH 2 , thereby lowering the desorption temperature of MgH 2 . As expected, MgH 2 -10 wt.% Ni 0.6 Zn 0.4 O samples exhibited a smaller particle size as compared with milled MgH 2 (as can be seen in Figure 7c). Ali et al. [56] introduced CoTiO 3 to MgH 2 and showed that the particle size of the composite changed to a finer and smaller size. According to the research results of Somo et al. [63], smaller particle sizes allow quick dissociation into the surface of materials. Besides that, the addition of Nb to MgH 2 creates a large number of hydrogen diffusion channels and speeds up hydrogen flow along the MgH 2 /Mg interfaces, continuing to improve the sorption kinetics of MgH 2 [64]. In light of this, it is obvious that adding Ni 0.6 Zn 0.4 O causes the particle size to be greatly decreased, which is useful for improving the performance of MgH 2 . The PSD of pure MgH2, milled MgH2, and MgH2-10 wt.% Ni0.6Zn0.4O samples were analyzed using Image J (version 2022). As shown in Figure 8a, the PSD calculated for pure MgH2 was 84.8 µm. The calculated PSD for milled MgH2 decreased to 0.29 µm as shown in Figure 8b. A study led by Maddah et al. [65] exposed that the average particle size of MgH2 decreased from 30 µm to 2.2 µm. Furthermore, as the milling time is extended up to 30 h, no discernible difference is seen. However, in this study, the PSD was decreased to 0.13 µm when 10 wt.% of Ni0.6Zn0.4O was added to MgH2, as shown in Figure 8c. This demonstrated how significantly MgH2's size was reduced after the addition of Ni0.6Zn0.4O as an additive. Moreover, Xiao and colleagues [66] stated that the particle size of milled MgH2 decreased to a range of 80 to 80 nm and lowered to 50 to 400 nm after LiCl was added. The PSD of pure MgH 2 , milled MgH 2 , and MgH 2 -10 wt.% Ni 0.6 Zn 0.4 O samples were analyzed using Image J (version 2022). As shown in Figure 8a, the PSD calculated for pure MgH 2 was 84.8 µm. The calculated PSD for milled MgH 2 decreased to 0.29 µm as shown in Figure 8b. A study led by Maddah et al. [65] exposed that the average particle size of MgH 2 decreased from 30 µm to 2.2 µm. Furthermore, as the milling time is extended up to 30 h, no discernible difference is seen. However, in this study, the PSD was decreased to 0.13 µm when 10 wt.% of Ni 0.6 Zn 0.4 O was added to MgH 2 , as shown in Figure 8c. This demonstrated how significantly MgH 2 's size was reduced after the addition of Ni 0.6 Zn 0.4 O as an additive. Moreover, Xiao and colleagues [66] stated that the particle size of milled MgH 2 decreased to a range of 80 to 80 nm and lowered to 50 to 400 nm after LiCl was added.
The effect of Ni 0.6 Zn 0.4 O addition on the MgH 2 bonding was investigated by using FTIR, as shown in Figure 9. All the samples exhibited two bands: (i) 400-800 cm −1 , corresponding to Mg-H bending bands, and (ii) 800-1400 cm −1 , attributed to the Mg-H stretching bands as previously shown by Zhang et al. [67]. For milled MgH 2 , an obvious peak around 515 cm −1 is attributed to Mg-H bending bands. This peak indicated that the milled MgH 2 was stable during the milling process. In our study, the bending and stretching bands were at about 772 cm −1 and 1380 cm −1 , respectively. No new peak was detected due to the low amount of Ni 0.6 Zn 0.4 O as an additive. However, after the addition of 10 wt.% Ni 0.6 Zn 0.4 O as an additive, the peaks were shifted to a low wavenumber, which indicates the weakness of the Mg-H bond. Furthermore, Ismail et al. [68] also agreed with these findings. The effect of Ni0.6Zn0.4O addition on the MgH2 bonding was investigated by using FTIR, as shown in Figure 9. All the samples exhibited two bands: (i) 400-800 cm −1 , corresponding to Mg-H bending bands, and (ii) 800-1400 cm −1 , attributed to the Mg-H stretching bands as previously shown by Zhang et al. [67]. For milled MgH2, an obvious peak around 515 cm −1 is attributed to Mg-H bending bands. This peak indicated that the milled MgH2 was stable during the milling process. In our study, the bending and stretching bands were at about 772 cm −1 and 1380 cm −1 , respectively. No new peak was detected due to the low amount of Ni0.6Zn0.4O as an additive. However, after the addition of 10 wt.% Ni0.6Zn0.4O as an additive, the peaks were shifted to a low wavenumber, which indicates the weakness of the Mg-H bond. Furthermore, Ismail et al. [68] also agreed with these findings.  The effect of Ni0.6Zn0.4O addition on the MgH2 bonding was investigated by using FTIR, as shown in Figure 9. All the samples exhibited two bands: (i) 400-800 cm −1 , corresponding to Mg-H bending bands, and (ii) 800-1400 cm −1 , attributed to the Mg-H stretching bands as previously shown by Zhang et al. [67]. For milled MgH2, an obvious peak around 515 cm −1 is attributed to Mg-H bending bands. This peak indicated that the milled MgH2 was stable during the milling process. In our study, the bending and stretching bands were at about 772 cm −1 and 1380 cm −1 , respectively. No new peak was detected due to the low amount of Ni0.6Zn0.4O as an additive. However, after the addition of 10 wt.% Ni0.6Zn0.4O as an additive, the peaks were shifted to a low wavenumber, which indicates the weakness of the Mg-H bond. Furthermore, Ismail et al. [68] also agreed with these findings. The XRD pattern of the MgH 2 -10 wt.% Ni 0.6 Zn 0.4 O samples after milling for 1 h, after desorption at 450 • C, and after absorption at 250 • C at the 1st cycle is exhibited in Figure 10a. As shown in Figure 10 NiO, and MgO remained unchanged even after the 10th cycle. Another peak of the XRD spectra for absorption at 10th cycles was also reported in Figure 10b below, labeled 10th absorption. The peaks of MgH2 were found, which revealed the Mg peaks were transformed into MgH2. Nevertheless, the in situ forms of ZnO, NiO, and MgO still appeared and remain unchanged. Based on the result obtained, the in situ formation may also provide a significant effect that will help boost the hydrogen sorption performance of MgH2. A previous work discovered that the performance of hydrogen storage MgH2 is significantly improved by the inclusion of metal oxide as a catalyst or additive [69]. According to a study by Zou et al. [70], the polarization might weaken the Ti-O bonds and Mg-H bonds, which make MgH2 decompose quickly after the addition of TiO. New peaks of ZnO, NiO, and MgO could also be seen as the samples were heated up. However, the peaks of Mg were completely transformed into MgH 2 during the absorption process at 250 • C, while the peaks of ZnO, NiO, and MgO remained unaltered (labeled absorption).
The XRD pattern for MgH 2 -10 wt.% Ni 0.6 Zn 0.4 O samples after the 10th cycle of desorption and absorption was analyzed and illustrated as in Figure 10b. Obviously, the Mg peak dominates even at the 10th cycle, and no peak of MgH 2 was found, as demonstrated in the figure below (labeled 10th desorption). However, the peaks of ZnO, NiO, and MgO remained unchanged even after the 10th cycle. Another peak of the XRD spectra for absorption at 10th cycles was also reported in Figure 10b below, labeled 10th absorption. The peaks of MgH 2 were found, which revealed the Mg peaks were transformed into MgH 2 . Nevertheless, the in situ forms of ZnO, NiO, and MgO still appeared and remain unchanged. Based on the result obtained, the in situ formation may also provide a significant effect that will help boost the hydrogen sorption performance of MgH 2 .
A previous work discovered that the performance of hydrogen storage MgH 2 is significantly improved by the inclusion of metal oxide as a catalyst or additive [69]. According to a study by Zou et al. [70], the polarization might weaken the Ti-O bonds and Mg-H bonds, which make MgH 2 decompose quickly after the addition of TiO. Furthermore, Huang et al. [71] discovered that faster absorption/desorption kinetics of MgH 2 can be observed after the addition of Sc 2 O 3 and TiO 2 . Further findings indicate that the surface defects and grain boundaries created by the milling process after the addition of Sc 2 O 3 and TiO 2 provide a significant number of diffusion channels and active sites that greatly enhance the kinetics of MgH 2 .
In this study, the in situ formation of MgO, NiO, and ZnO was observed during the heating process of MgH 2 -10 wt.% Ni 0.6 Zn 0.4 O samples. The formation of MgO after the addition of additive/catalysts has well agreed with previous research. Aguey-Zinsou et al. [72] indicated that the role of MgO is rationalized in the concept of a "Process Control Agent". On top of that, MgO has dispersed properties and good lubricant thus preventing MgH 2 from clumping together. Additionally, Shan et al. [73] also revealed that one of the final reaction products of CoFe 2 O 4 and MgH 2 is MgO, which may help reduce the onset desorption temperature from 440 • C for as-received MgH 2 to 160 • C after doping with 7 mol% of CoFe 2 O 4 . In order to tailor MgH 2 performance, Ali et al. [74] introduced 10 wt.% of MgNiO 2 to MgH 2, and the results show that MgH 2 -10 wt.% MgNiO 2 samples can desorb roughly 5.10 wt.% of H 2 within 10 min at 320 • C and begin to decompose at 258 • C. Surprisingly, at 200 • C, MgH 2 -10 wt.% MgNiO 2 samples continue to absorb 6.10 wt.% of H 2 in just 10 min. The performance of MgH 2 as a hydrogen storage material is boosted by the formation of new MgO and NiO compounds.
A previous study reported that adding a Co 2 NiO catalyst can lower the desorption temperature by 117 • C (pure MgH 2 ) and 70 • C (milled MgH 2 ) and decrease the activation energy by 65 kJ/mol and 15 kJ/mol for pure MgH 2 and milled MgH 2 , respectively [48]. According to a study by Zhang et al. [75], the bond between Mg and H is weaker than the bond between transition metals such as Ni. The release of the H atom and H 2 recombination from the MgH 2 surface is encouraged by the weakening of the bond between H and Mg caused by the strong bonding between Ni and H. Besides, Patah et al. [76] also exposed the fact that adding ZnO to MgH 2 reduces the onset desorption peak of the DSC curves from 375 • C to 360 • C. Along this line, it is valuable to conclude that the addition of Ni 0.6 Zn 0.4 O as an additive significantly enhances the sorption properties of MgH 2 . A study on the catalytic mechanism revealed that in situ formations of metal oxides such as MgO, ZnO, and NiO during the heating process may help in improving the hydrogen storage performance of MgH 2 .

Conclusions
In this work, Ni 0.6 Zn 0.4 O samples were successfully synthesized via the solid-state method, and the catalytic effects of Ni 0.6 Zn 0.4 O on the hydrogen storage performance of MgH 2 were systematically studied for the first time. Different weight percentages (5, 10, 15, and 20 wt.%) of Ni 0.6 Zn 0.4 O were milled together for 1 h, and the onset desorption temperature was reduced to a range of 280 • C to 305 • C, which is lower than pure MgH 2 (418 • C) and milled MgH 2 (341 • C). The absorption and desorption kinetics of MgH 2 could be largely enhanced by the addition of 10 wt.% of Ni 0. 6  Smaller particles size provided more grain boundaries and larger surface area which benefited the diffusion path for hydrogen during the absorption and release process. From these results, it can be concluded that the reduction in particle size and the in situ generated (ZnO, NiO, and MgO) during the heating process played synergistic catalytic effects that boosted the hydrogen storage performance of MgH 2 .

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.