Influence of Nanosized CoTiO3 Synthesized via a Solid-State Method on the Hydrogen Storage Behavior of MgH2

Magnesium hydride (MgH2) has received outstanding attention as a safe and efficient material to store hydrogen because of its 7.6 wt.% hydrogen content and excellent reversibility. Nevertheless, the application of MgH2 is obstructed by its unfavorable thermodynamic stability and sluggish sorption kinetic. To overcome these drawbacks, ball milling MgH2 is vital in reducing the particle size that contribute to the reduction of the decomposition temperature. However, the milling process would become inefficient in reducing particle sizes when equilibrium between cold-welding and fracturing is achieved. Therefore, to further ameliorate the performance of MgH2, nanosized cobalt titanate (CoTiO3) has been synthesized using a solid-state method and was introduced to the MgH2 system. The different weight percentages of CoTiO3 were doped to the MgH2 system, and their catalytic function on the performance of MgH2 was scrutinized in this study. The MgH2 + 10 wt.% CoTiO3 composite presents the most outstanding performance, where the initial decomposition temperature of MgH2 can be downshifted to 275 °C. Moreover, the MgH2 + 10 wt.% CoTiO3 absorbed 6.4 wt.% H2 at low temperature (200 °C) in only 10 min and rapidly releases 2.3 wt.% H2 in the first 10 min, demonstrating a 23-times-faster desorption rate than as-milled MgH2 at 300 °C. The desorption activation energy of the 10 wt.% CoTiO3-doped MgH2 sample was dramatically lowered by 30.4 kJ/mol compared to undoped MgH2. The enhanced performance of the MgH2–CoTiO3 system is believed to be due to the in situ formation of MgTiO3, CoMg2, CoTi2, and MgO during the heating process, which offer a notable impact on the behavior of MgH2.


Introduction
In recent decades, hydrogen has emerged as the most viable energy carrier, especially in the transportation sector [1]. As far as we are concerned, the usage of fossil fuels has led to a global energy crisis and environmental pollution [2]. Thus, energy transition from fossil fuel-based to renewable energy is required. As environmentally-friendly energy storage, the best solution is to convert energy from "green" sources into chemical storage [3]. Hydrogen has received outstanding interest as an energy carrier because it has a high energy content (142 MJ/kg) [4]. Hydrogen might supersede natural gas and solid fuels as an energy source by 2050. It is anticipated that it will be extensively used in the transportation, chemical, and long-term aviation and maritime sectors [5]. The use of hydrogen as an energy carrier will ultimately aid humankind in achieving a "low" or "zero" carbon future. Recently, various countries have started using hydrogen energy, particularly in transportation. For instance, the Hyundai Nexo hydrogen fuel cell vehicle (FCV) sold 1000 units in South Korea in 2019 [6], and the company is also setting up a number of hydrogen refueling stations [7]. Nevertheless, hydrogen-based energy requires a convenient and a reliable catalysts are favorable in ameliorating the hydrogen storage behavior of MgH 2 by boosting the electron transfer between Mg and H [32].
In this study, it is of interest to examine the catalytic activity of another transition metal oxide, CoTiO 3 . What is noteworthy, is that the catalytic efficiency is influenced by the dispersion of catalytic activity on the system. Nanocatalysts offer noticeably more active catalytic sites and intimate interaction because of their increased surface area. Thus, in this study, we synthesized nanosized CoTiO 3 via a solid-state method. The addition of CoTiO 3 was projected to remarkably boost the behavior of MgH 2 by lowering the initial decomposition temperature and rapid hydrogen absorption/desorption kinetics. Although there is a previous study reported on the effect of CoTiO 3 [41], there remains a lack of mechanism of interaction between MgH 2 and CoTiO 3 . Furthermore, different synthesis methods present different catalytic activities. This study broadens the way to explore the effect of CoTiO 3 and demonstrates the different catalytic activities of CoTiO 3 on the hydrogen storage behavior of MgH 2 . Hence, in this paper, the influence of different weight percentages of CoTiO 3 was systematically explored.

Materials and Methods
MgH 2 (95% pure), Co 3 O 4 (99.99% pure), anatase TiO 2 (99.95% pure), and citric acid monohydrate (>98% pure) provided by Sigma Aldrich, St. Louis, MO, United States were used without any pretreatment. The nanosized CoTiO 3 was synthesized using the solidstate method. The stoichiometric amount of Co 3 O 4 and TiO 2 were ground together for 15 min. Thereafter, citric acid monohydrate was added and continuously ground for 15 min. The powder was then calcined at 850 • C for 5 h. Thereafter, different weight percentages of CoTiO 3 (5,10,15, and 20 wt.%) were added with MgH 2 to study its catalytic activity on the hydrogen storage behavior of MgH 2 . The mixture was milled together in a planetary ball mill (NQM-0.4) at 400 rpm for 1 h.
Sievert-type pressure-composition-temperature apparatus from Advanced Materials Corporation, Pittsburgh, PA, United States which is also known as Gas Reaction Controller (GRC) apparatus, was used to conduct the temperature-programmed desorption (TPD) measurement and the hydrogen absorption and desorption experiment. The GRC provides a reliable and convenient way to evaluate desorption/absorption characteristics of materials. Temperature, capacity, and time data obtained through TPD measurements were used to determine the capacity, kinetics, and thermodynamic properties of material. The TPD measurement was conducted in a vacuum chamber heated to 450 • C (heating rate: 5 • C/min). For the isothermal tests, experiments were conducted at 200 • C, 33.0 atm hydrogen pressure (absorption kinetics), and 300 • C, 1.0 atm hydrogen pressure (desorption kinetics).
To study the thermal properties and to calculate the activation energy, differential scanning calorimetry (DSC, Mettler Toledo, Columbus, OH, United States)(DSC/TGA 1) was used. Six to eight milligrams of the sample were placed in a crucible and heated to 500 • C under an argon flow of 50 mL/min at four different heating rates (15 • C/min, 20 • C/min, 25 • C/min, and 30 • C/min). Rigaku MiniFlex, Tokyo, Japan, X-ray diffraction (XRD) with Cu Kα radiation was used to examine the phase structure. The patterns were scanned over diffraction angles from 20 • to 80 • at 2.00 • /min. The sample microstructure and morphology were examined using a scanning electron microscope (SEM; JEOL, Akishima, Tokyo, Japan) (JSM-6350LA). The particle size distributions of the samples were calculated using Image J software. The SEM images was used to calculate the distributions size of the sample, where a number of measurements were collected for the diameter of the particle. Fourier transform infrared (FTIR) spectrometry was conducted in the range of 400-2000 cm −1 using IR Tracer-100, Shimadzu, Kyoto, Japan and Raman spectra were conducted at room temperature (0.1% power laser) using Renishaw Raman spectroscopy (532 nm radiation).

Synthesis of CoTiO 3
The phase composition of the as-synthesized CoTiO 3 was investigated using XRD, as presented in Figure 1a. The diffraction peaks at 2θ of 23 (1)), as follows [42]: where L, K, λ, β, and θ are referring to the average crystallite size, Scherer constant (0.94), X-ray wavelength, full width at half maximum, and diffraction angle, respectively. The FTIR spectrum of the as-synthesized CoTiO 3 is presented in Figure 1b. The bands between 400 and 800 cm −1 are the typical peaks of CoTiO 3 that correspond to the stretching vibrations of the metal ions [43]. The FTIR spectrum exhibited bands between 640 and 450 cm −1 , which is due to Ti-O-Ti and Co-Ti-O band formation [44]. The strong band at around 431 cm −1 corresponds to the bond of Co-O [45]. Maensiri et al. [46] and Rashad et al. [47] stated that the peaks in the range of 450-600 cm −1 were ascribed to the Ti-O-Ti bond. Thus, the peaks at 473 cm −1 are assigned to the Ti-O-Ti bond [48] and the peaks at 626 cm −1 correspond to the Ti-O-O bond [46]. Moreover, the Raman spectra in Figure 1c confirm the formation of CoTiO 3 [44]. The strongest Raman modes observed around 692 cm −1 were attributed to the high frequency of the vibrational mode of the CoO 6 octahedra, known as the symmetry stretching mode, while other remaining Raman modes correspond to the lattice vibrations of the phonon modes [43]. The SEM image (Figure 1d) of the as-synthesized CoTiO 3 revealed that the as-synthesized nanosized CoTiO 3 comprises a good dispersion of eclipse-like grain shape. The particle size of the as-synthesized CoTiO 3 is measured using Image J software, with an average particle size of 0.16 µm. The distributions of the particle sizes are depicted in a histogram in Figure 1e. The results show that it is relevant to conclude that pure nanosized CoTiO 3 was successfully synthesized using the solid-state method.

Synthesis of CoTiO3
The phase composition of the as-synthesized CoTiO3 was investigated using XRD, as presented in Figure 1a. The diffraction peaks at 2θ of 23 (1)), as follows [42]: where L, K, λ, β, and θ are referring to the average crystallite size, Scherer constant (0.94), X-ray wavelength, full width at half maximum, and diffraction angle, respectively. The FTIR spectrum of the as-synthesized CoTiO3 is presented in Figure 1b. The bands between 400 and 800 cm −1 are the typical peaks of CoTiO3 that correspond to the stretching vibrations of the metal ions [43]. The FTIR spectrum exhibited bands between 640 and 450 cm −1 , which is due to Ti-O-Ti and Co-Ti-O band formation [44]. The strong band at around 431 cm −1 corresponds to the bond of Co-O [45]. Maensiri et al. [46] and Rashad et al. [47] stated that the peaks in the range of 450-600 cm −1 were ascribed to the Ti-O-Ti bond. Thus, the peaks at 473 cm −1 are assigned to the Ti-O-Ti bond [48] and the peaks at 626 cm −1 correspond to the Ti-O-O bond [46]. Moreover, the Raman spectra in Figure 1c confirm the formation of CoTiO3 [44]. The strongest Raman modes observed around 692 cm −1 were attributed to the high frequency of the vibrational mode of the CoO6 octahedra, known as the symmetry stretching mode, while other remaining Raman modes correspond to the lattice vibrations of the phonon modes [43]. The SEM image (Figure 1d) of the as-synthesized CoTiO3 revealed that the as-synthesized nanosized CoTiO3 comprises a good dispersion of eclipse-like grain shape. The particle size of the as-synthesized CoTiO3 is measured using Image J software, with an average particle size of 0.16 µm. The distributions of the particle sizes are depicted in a histogram in Figure 1e. The results show that it is relevant to conclude that pure nanosized CoTiO3 was successfully synthesized using the solid-state method.

Hydrogen Storage Properties
The catalytic activity of the as-prepared CoTiO 3 on MgH 2 was studied by TPD experiments, as depicted in Figure 2a. The performance of the modified MgH 2 with CoTiO 3 was significantly enhanced compared to undoped MgH 2 . In terms of decomposition temperature, commercial MgH 2 started decomposing at 420 • C, whereas as-milled MgH 2 started at approximately 340 • C. This outcome implies that the process of ball milling for 1 h is beneficial in lowering the decomposition temperature of MgH 2 as it decreased by 80 • C when compared with commercial MgH 2 . This result is consistent with a prior study that found that enhanced performance of metal/complex hydride is due to the milling process, which is effective in refining the particle size [49]. The milling process in essential in reducing the particle size, which contributes to the decrease of the decomposition temperature. However, the milling process would become inefficient in reducing particle sizes when the equilibrium between cold-welding and fracturing is achieved [50]. A previous study also reported that increasing the milling time causes the decomposition temperature to shift to higher temperatures due to the agglomeration of particles and reduction of their free surface area [51]. After augmenting MgH 2 with 5 wt.% of CoTiO 3 , the onset decomposition temperature downshifted to 298 • C. The reduction in the decomposition temperature was more significant when the amount of CoTiO 3 was increased to 10 wt.%. The MgH 2 + 10 wt.% CoTiO 3 sample began to decompose at 275 • C. For the 15 wt.% CoTiO 3 -doped MgH 2 , the decomposition temperature was almost similar to the MgH 2 + 10 wt.% CoTiO 3 . The influence of different wt.% on the performance of MgH 2 was further explored with 20 wt.% of CoTiO 3 . The composite sample of MgH 2 + 20 wt.% CoTiO 3 also began to release hydrogen lower than undoped MgH 2 , which started to release at 295 • C. However, the onset decomposition for the MgH 2 + 20 wt.% CoTiO 3 was slightly higher than MgH 2 + 10 wt.% CoTiO 3 . This outcome was comparable with a previous study that found that the decomposition temperature for the 15 wt.% and 20 wt.% SrFe 12 O 19 -doped MgH 2 was slightly higher than MgH 2 + 10 wt.% SrFe 12 O 19 [52]. Sulaiman et al. [53] also claimed the same condition, where the decomposition temperature for the MgH 2 + 50 wt.% Na 3 FeF 6 was slightly higher than for MgH 2 + 10 wt.% Na 3 FeF 6 and MgH 2 + 20 wt.% Na 3 FeF 6 . This condition may be due to the excessive catalyst that may block the diffusion path of hydrogen, which limits the hydrogen diffusion [54]. Even the temperature for the hydrogen release for the 20 wt.% of catalyst was higher than that for the 10 wt.%; the temperature for the MgH 2 to release hydrogen doped with a different weight percentage of CoTiO 3 was significantly reduced when compared with undoped MgH 2 . Of note is the fact that the presence of CoTiO 3 could modify the performance of MgH 2 and be beneficial in lowering the decomposition temperature of MgH 2 . The reversibility of the MgH2-CoTiO3 system was further studied via an abs kinetic experiment conducted at a low temperature (200 °C). The doped sample wa ied with a different weight percentage of CoTiO3, as demonstrated in Figure 2b. the four different wt.%, MgH2 + 10 wt.% CoTiO3 produces the fastest absorption r represents improved absorption kinetics performance than does undoped Mg shown in Figure 2b, in the first 10 min, the as-milled MgH2 absorbed 4.3 wt.% H2, w the MgH2 + 10 wt.% CoTiO3 sample absorbed 6.4 wt.% H2. Even after completing min duration, the as-milled MgH2 could not attain as high capacity as that of the doped 10 wt.% CoTiO3. In the meantime, the hydrogen absorbed for the 5, 15, and 2 of catalyst were 2.5, 4.1, and 6.0 wt.% H2, respectively, in the first 10 min. For the of CoTiO3, the MgH2 + 5 wt.% CoTiO3 shows faster kinetics than milled MgH2 in t 11s and intercept with the MgH2 at 12s and become slower than undoped MgH2. It concluded that the hydrogen absorption behavior was affected by the amount of c In the context of desorption kinetics, as depicted in Figure 2c, the MgH2-doped composite displays a rapid desorption rate compared to undoped MgH2. Simila performance of the absorption kinetics, the MgH2 + 10 wt.% CoTiO3 presents a fa sorption kinetics rate compared to the 5, 15, and 20 wt.% CoTiO3-doped MgH2 sa In the first 10 min, the undoped MgH2 roughly desorb hydrogen (<0.1 wt.%). Mea the MgH2 + 10 wt.% CoTiO3 sample desorbed at approximately 2.3 wt.% H2, which The reversibility of the MgH 2 -CoTiO 3 system was further studied via an absorption kinetic experiment conducted at a low temperature (200 • C). The doped sample was studied with a different weight percentage of CoTiO 3 , as demonstrated in Figure 2b. Among the four different wt.%, MgH 2 + 10 wt.% CoTiO 3 produces the fastest absorption rate and represents improved absorption kinetics performance than does undoped MgH 2 . As shown in Figure 2b, in the first 10 min, the as-milled MgH 2 absorbed 4.3 wt.% H 2 , whereas the MgH 2 + 10 wt.% CoTiO 3 sample absorbed 6.4 wt.% H 2 . Even after completing the 60 min duration, the as-milled MgH 2 could not attain as high capacity as that of the MgH 2 -doped 10 wt.% CoTiO 3 . In the meantime, the hydrogen absorbed for the 5, 15, and 20 wt.% of catalyst were 2.5, 4.1, and 6.0 wt.% H 2 , respectively, in the first 10 min. For the 5 wt.% of CoTiO 3 , the MgH 2 + 5 wt.% CoTiO 3 shows faster kinetics than milled MgH 2 in the first 11s and intercept with the MgH 2 at 12s and become slower than undoped MgH 2 . It can be concluded that the hydrogen absorption behavior was affected by the amount of catalyst. In the context of desorption kinetics, as depicted in Figure 2c, the MgH 2 -doped CoTiO 3 composite displays a rapid desorption rate compared to undoped MgH 2 . Similar to the performance of the absorption kinetics, the MgH 2 + 10 wt.% CoTiO 3 presents a faster desorption kinetics rate compared to the 5, 15, and 20 wt.% CoTiO 3 -doped MgH 2 samples. In the first 10 min, the undoped MgH 2 roughly desorb hydrogen (<0.1 wt.%). Meanwhile, the MgH 2 + 10 wt.% CoTiO 3 sample desorbed at approximately 2.3 wt.% H 2 , which represents a 23-times-faster desorption rate when compared with undoped MgH 2 . In the same period, the 5, 15, and 20 wt.%-doped MgH 2 desorbed 0.8, 1.8, and 1.2 wt.% H 2 , respectively, which is also faster than undoped MgH 2 . Table 1 summarizes the hydrogen storage behavior of undoped and doped MgH 2 with a difference percentage of CoTiO 3 . Notably, low catalyst content is also beneficial in ameliorating the kinetics performances of MgH 2 . Similar to the TPD result, the MgH 2 + 10 wt.% CoTiO 3 performs the fastest absorption and desorption rate. These findings are in line with a previous study that demonstrated MgH 2 + 0.5 mol Nb 2 O 5 performs superior desorption kinetics than MgH 2 doped with 0.05, 0.1, 0.2, and 1.0 mol Nb 2 O 5 [55]. Similar to a study conducted by Ranjbar et al. [54], the desorption behavior of the MgH 2 with 10 wt.% of SiC presents a faster desorption rate, which becomes slower when doping with 20 wt.% of SiC. They indicated that excessive catalysts in the composite may restrict the hydrogen diffusion to some extent, limiting the Mg-H reaction. From the findings, noticeably, the performance of MgH 2 can be affected by the amount of the catalyst. By comprehensively considering the absorption/desorption hydrogen performance, the MgH 2 with 10 wt.% of the CoTiO 3 sample demonstrates the ultimate absorption/desorption performance. Thus, in this study, the MgH 2 + 10 wt.% CoTiO 3 sample is applied to examine the catalytic activity of CoTiO 3 on the hydrogen storage behavior of MgH 2 . These results are similar to previous studies, in which 10 wt.% was an optimum amount of additive to provide a synergetic catalytic effect on the hydrogen storage performance of a metal hydride and a complex hydride [56,57]. Moreover, an excessive amount of additive may limit the diffusion of hydrogen to some extent, thereby reducing the reaction between the Mg and the hydrogen [54]. The effect of CoTiO 3 was further explored with the cycling performance of MgH 2 + 10 wt.% CoTiO 3 , as shown in Figure 3. As reported previously, the cyclability of the undoped MgH 2 is poor, where the performance would erode drastically after undergoing the first cycle [58]. Therefore, the cyclability of the undoped MgH 2 is not included in this study because it desorbs only a small amount of hydrogen for the first cycle. Figure 3a shows the absorption kinetics curve of MgH 2 + 10 wt.% CoTiO 3 over the ten cycles. Surprisingly, after completing the 10th cycle, there is only a small amount of degradation in the hydrogen capacity for the absorption kinetics performance. The MgH 2 + 10 wt.% CoTiO 3 was able to absorb 7.1 wt.% H 2 after the 10th cycle. Figure 3b shows the desorption kinetics curve of the MgH 2 + 10 wt.% CoTiO 3 for the ten cycles. The MgH 2 + 10 wt.% CoTiO 3 composite maintain good cyclability, with the ability to desorb 5.7 wt.% H 2 , even after completing the 10th cycle. It is evident that the addition of CoTiO 3 is beneficial in maintaining a superior cyclability of MgH 2 .
To explore the kinetics mechanism of the MgH 2 + 10 wt.% CoTiO 3 composite further, the sorption kinetics performance was analyzed using the kinetics model. For instance, a previous study has demonstrated the kinetic model characterization for the absorption/desorption time for the alloys materials [59,60]. A series of kinetics models have been widely explored and reviewed by Pang and Li [61] in their study. In this study, the kinetics mechanism was studied using Johnson-Mehl-Avrami (JMA) and contracting volume (CV) models, as indicated in Table 2. The JMA and CV models were considered because they are suitable for the experimental data and are precise without requiring further assumptions or approximations [61]. Moreover, other researchers have utilized this method extensively in prior studies to understand the rate-limiting step of the researched materials [62]. With kinetic equations such as CV and JMA, the rate-limiting step of the kinetics can be derived from the experimental data. In this study, the rate-limiting step is determined using the best linear plot of the experimental data with the kinetics equations. To explore the kinetics mechanism of the MgH2 + 10 wt.% CoTiO3 composite furth the sorption kinetics performance was analyzed using the kinetics model. For instance previous study has demonstrated the kinetic model characterization for the abso tion/desorption time for the alloys materials [59,60]. A series of kinetics models have be widely explored and reviewed by Pang and Li [61] in their study. In this study, the kinet mechanism was studied using Johnson-Mehl-Avrami (JMA) and contracting volum (CV) models, as indicated in Table 2. The JMA and CV models were considered becau they are suitable for the experimental data and are precise without requiring further sumptions or approximations [61]. Moreover, other researchers have utilized this meth extensively in prior studies to understand the rate-limiting step of the researched mate als [62]. With kinetic equations such as CV and JMA, the rate-limiting step of the kinet can be derived from the experimental data. In this study, the rate-limiting step is det mined using the best linear plot of the experimental data with the kinetics equations.

Kinetic Equation Rate Limiting
Step α= kt Surface controlled.

Kinetic Equation Rate Limiting
Step α = kt Surface controlled.
[-ln(1-α)]1/2 = kt JMA 2D: 2D growth of existing nuclei with constant interface velocity. Figure 4 present the kinetics curve that was calculated on the basis of the equation in Table 2. The absorption and desorption kinetics curves of the MgH 2 + 10 wt.% CoTiO 3 were calculated for reacted fractions ranging from 0% to 80% [63]. From the figure, for the absorption kinetics at 200 • C, the rate-limiting step is best represented using the 3D growth diffusion controlled with decreasing interface velocity, whereas the desorption kinetics of the MgH 2 + 10 wt.% CoTiO 3 at 300 • C was best described using the 2D growth of existing nuclei with constant interface velocity. These results suggest that hydrogen diffusion through the MgH 2 + 10 wt.% CoTiO 3 was fast. The nanosized CoTiO 3 may be beneficial in allowing a faster dissociation rate of hydrogen and hence fasten the diffusion of hydrogen, which results in faster kinetic performance.
The thermal characteristics of MgH 2 + 10 wt.% CoTiO 3 , performed using DSC (heating rate: 30 • C/min), are presented in Figure 5. The DSC trace demonstrates one endothermic peak correlated to the hydrogen release of MgH 2 . The peak of hydrogen release of MgH 2 + 10 wt.% CoTiO 3 happens at a lower temperature (382 • C) compared with as-milled MgH 2 (445 • C), indicating that the dehydrogenation kinetics of MgH 2 was enhanced by milling with CoTiO 3 . omaterials 2022, 12, x FOR PEER REVIEW 9 of beneficial in allowing a faster dissociation rate of hydrogen and hence fasten the diffusi of hydrogen, which results in faster kinetic performance. The thermal characteristics of MgH2 + 10 wt.% CoTiO3, performed using DSC (he ing rate: 30 °C/min), are presented in Figure 5. The DSC trace demonstrates one endoth mic peak correlated to the hydrogen release of MgH2. The peak of hydrogen release MgH2 + 10 wt.% CoTiO3 happens at a lower temperature (382 °C) compared with as-mil MgH2 (445 °C), indicating that the dehydrogenation kinetics of MgH2 was enhanced milling with CoTiO3. The enhancement in the dehydrogenation kinetic is associated with the energy b rier for hydrogen released from MgH2. In this study, the activation energy for hydrog  The thermal characteristics of MgH2 + 10 wt.% CoTiO3, performed using DSC ing rate: 30 °C/min), are presented in Figure 5. The DSC trace demonstrates one end mic peak correlated to the hydrogen release of MgH2. The peak of hydrogen rel MgH2 + 10 wt.% CoTiO3 happens at a lower temperature (382 °C) compared with as MgH2 (445 °C), indicating that the dehydrogenation kinetics of MgH2 was enhan milling with CoTiO3. The enhancement in the dehydrogenation kinetic is associated with the energ rier for hydrogen released from MgH2. In this study, the activation energy for hy release from the doped and undoped MgH2 was evaluated using the DSC results. 6a,b display the DSC curve of as-milled MgH2 and MgH2 + 10 wt.% CoTiO3 at four ent heating rates. The Kissinger equation (Equation (2)) is used to evaluate the act energy during decomposition. The enhancement in the dehydrogenation kinetic is associated with the energy barrier for hydrogen released from MgH 2 . In this study, the activation energy for hydrogen release from the doped and undoped MgH 2 was evaluated using the DSC results. Figure 6a,b display the DSC curve of as-milled MgH 2 and MgH 2 + 10 wt.% CoTiO 3 at four different heating rates. The Kissinger equation (Equation (2)) is used to evaluate the activation energy during decomposition.
where E A is the activation energy, R is the gas constant, T p is the endothermic peak corresponding to the decomposition temperature, β is the heating rate, and A is a linear constant. Thereafter, the activation energy is determined on the basis of the Kissinger plot of ln [β/T p 2 ] versus 1000/T p , as in Figure 6c. The activation energy of the MgH 2 + 10 wt.% CoTiO 3 and as-milled MgH 2 were calculated to be 104.6 and 135.0 kJ/mol, respectively. The activation energy of MgH 2 was decreased by 23% with the presence of CoTiO 3 compared with undoped MgH 2 . Adding CoTiO 3 into MgH 2 remarkably reduces the energy barrier, which led to reduced activation energy and enhanced kinetic performance. ln [β/Tp 2 ] = −EA/RTp + A (2) where EA is the activation energy, R is the gas constant, Tp is the endothermic peak corresponding to the decomposition temperature, β is the heating rate, and A is a linear constant. Thereafter, the activation energy is determined on the basis of the Kissinger plot of ln [β/Tp 2 ] versus 1000/Tp, as in Figure 6c. The activation energy of the MgH2 + 10 wt.% CoTiO3 and as-milled MgH2 were calculated to be 104.6 and 135.0 kJ/mol, respectively. The activation energy of MgH2 was decreased by 23% with the presence of CoTiO3 compared with undoped MgH2. Adding CoTiO3 into MgH2 remarkably reduces the energy barrier, which led to reduced activation energy and enhanced kinetic performance. Lower activation energy provided the reduction in the desorption barrier, which results in a dramatic decrease in the decomposition temperature. Low decomposition temperature indicates the weakening of the hydride stability [64]. As the bond energy of the Mg-H is the decisive factor in hydride stability, the reduction in hydride stability is attributed to the bond weakening between Mg and H. Typically, the addition of transition metal elements or their oxides helps in decreasing the binding energy between the Mg and H due to the electron exchange process between the MgH2 and additives in oxidation or reduction reactions [65]. As reported by a previous study, adding a catalyst is the most favorable approach for promoting the dissociation of the Mg-H bond by lowering its bond energy, which then offers the reduction of the MgH2 stability and decomposition temperature [66]. Thus, it is reasonable to conclude that the addition of CoTiO3 is beneficial in Lower activation energy provided the reduction in the desorption barrier, which results in a dramatic decrease in the decomposition temperature. Low decomposition temperature indicates the weakening of the hydride stability [64]. As the bond energy of the Mg-H is the decisive factor in hydride stability, the reduction in hydride stability is attributed to the bond weakening between Mg and H. Typically, the addition of transition metal elements or their oxides helps in decreasing the binding energy between the Mg and H due to the electron exchange process between the MgH 2 and additives in oxidation or reduction reactions [65]. As reported by a previous study, adding a catalyst is the most favorable approach for promoting the dissociation of the Mg-H bond by lowering its bond energy, which then offers the reduction of the MgH 2 stability and decomposition temperature [66]. Thus, it is reasonable to conclude that the addition of CoTiO 3 is beneficial in promoting the superior performance of MgH 2 by lowering the energy barrier.
The catalytic activity of CoTiO 3 on the microstructure of MgH 2 was further explored by SEM characterization. Figure 7 displays the micrograph of commercial MgH 2 , MgH 2 after milling for 1 h, and MgH 2 + 10 wt.% CoTiO 3 . Referring to Figure 7a, commercial MgH 2 are irregular in shape, with solid flake-like shaped particles. After being milled for 1 h (Figure 7c), the solid flake-like shaped particles were broken into smaller particles with some agglomerations. Before the milling process, the comparatively smooth surface of the particle is substituted by the asperities and surface defects. Because of these transformed surface structures, as-milled MgH 2 decomposes at a lower temperature than commercial MgH 2 . Meanwhile, upon the addition of CoTiO 3 (Figure 7e), the agglomeration was reduced, and the particle size was transformed to a finer shape and size. The size of the MgH 2 + 10 wt.% CoTiO 3 was much smaller than ball milled MgH 2 due to the hardness of CoTiO 3 , which is higher than MgH 2 . The hardness of the MgH 2 was 0.58 GPa [67], while the hardness of CoTiO 3 was 3.4-7.5 GPa [68]. Therefore, it is speculated that the CoTiO 3 introduced the pulverization effect and helped to reduce the particle size of MgH 2 . The particle size of MgH 2 + 10 wt.% CoTiO 3 is smaller, and they are finer in shape, even after undergoing the absorption ( Figure 7g) and desorption (Figure 7i) process. No big changes occur on the particle after undergoing the absorption and desorption process. From the morphological properties, the CoTiO 3 is able to reduce the particle size of MgH 2 and enhances the absorption/desorption performance of MgH 2 . The differences of the microstructure and morphologies of the undoped and doped MgH 2 may correspond to the presence of a catalyst that is finely dispersed on the surface of the MgH 2 particle, thus reducing the H diffusion distance and offering more reaction site and hence stimulating the faster absorption and desorption performance of MgH 2 [69]. This catalyst is embedded on the surface of MgH 2 and prevent the sample from agglomerating, indicating that CoTiO 3 is beneficial in distributing particles and constraining agglomeration and particle growth in the MgH 2 system [22].
The particle size of the commercial MgH 2 , as-milled MgH 2 , MgH 2 + 10 wt.% CoTiO 3 after milling as well as after absorption and desorption were further evaluated by Image J software. Figure 7b,d,f,h,j shows the distributions of the particle size, which are depicted in a histogram. Based on the histogram, the average particle size of commercial MgH 2 , asmilled MgH 2 , MgH 2 + 10 wt.% CoTiO 3 after milling, after absorption, and after desorption was calculated to be~60, 0.47, 0.22, 0.29, and 0.28 µm, respectively. The morphological change and drastic size reduction contribute to the high surface defects and more grain boundaries around the surface of the composite [70]. Consequently, a higher amount of reaction for the nucleation sites and better diffusion channels for the hydrogen can be achieved with amplified grain boundaries that enhance the hydrogen sorption performances of the MgH 2 -CoTiO 3 composite. Similarly, a recent study found that reducing particle size improves absorption-desorption kinetics performance substantially [71,72]. Nanomaterials 2022, 12, x FOR PEER REVIEW 12 of 18  To reveal the catalytic mechanism of CoTiO 3 for MgH 2 , XRD analysis for MgH 2 + 10 wt.% CoTiO 3 was performed, as in Figure 8. The main peaks that existed after 1 h of milling ( Figure 8a) were only MgH 2 and CoTiO 3 . No new phase was detected, indicating that no chemical reaction occurs during the milling process. Figure 8b presents the sample of the MgH 2 + CoTiO 3 after desorption at 450 • C. Peaks of Mg were dominant, indicating that MgH 2 has been totally transformed to Mg, corresponding to the complete decomposition of MgH 2 , as in (Equation (3)). Additionally, the peaks of MgO and MgTiO 3 could be detected after dehydrogenation.
Nanomaterials 2022, 12, x FOR PEER REVIEW 13 of 18 To reveal the catalytic mechanism of CoTiO3 for MgH2, XRD analysis for MgH2 + 10 wt.% CoTiO3 was performed, as in Figure 8. The main peaks that existed after 1 h of milling ( Figure 8a) were only MgH2 and CoTiO3. No new phase was detected, indicating that no chemical reaction occurs during the milling process. Figure 8b presents the sample of the MgH2 + CoTiO3 after desorption at 450 °C. Peaks of Mg were dominant, indicating that MgH2 has been totally transformed to Mg, corresponding to the complete decomposition of MgH2, as in (Equation (3)). Additionally, the peaks of MgO and MgTiO3 could be detected after dehydrogenation.
MgH2  Mg + H2 (3) To further explore the catalytic mechanism of CoTiO3 during desorption, a sample of MgH2 + 50 wt.% CoTiO3 was prepared. As displayed in Figure 9a, after 1 h of milling, only the parent materials, MgH2 and CoTiO3, were detected. No new compound was formed at this stage. After the desorption process at 450°C (Figure 9b), the peaks of MgH2 and CoTiO3 disappeared and new peaks of MgO, MgTiO3, CoMg2, and CoTi2 appeared.  To further explore the catalytic mechanism of CoTiO 3 during desorption, a sample of MgH 2 + 50 wt.% CoTiO 3 was prepared. As displayed in Figure 9a ing (Figure 8a) were only MgH2 and CoTiO3. No new phase was detected, indicatin no chemical reaction occurs during the milling process. Figure 8b presents the sam the MgH2 + CoTiO3 after desorption at 450 °C. Peaks of Mg were dominant, indicatin MgH2 has been totally transformed to Mg, corresponding to the complete decomp of MgH2, as in (Equation (3)). Additionally, the peaks of MgO and MgTiO3 could tected after dehydrogenation.
MgH2  Mg + H2 To further explore the catalytic mechanism of CoTiO3 during desorption, a sam MgH2 + 50 wt.% CoTiO3 was prepared. As displayed in Figure 9a, after 1 h of millin the parent materials, MgH2 and CoTiO3, were detected. No new compound was f at this stage. After the desorption process at 450°C (Figure 9b), the peaks of MgH CoTiO3 disappeared and new peaks of MgO, MgTiO3, CoMg2, and CoTi2 appeared    (Figure 10a), the main peaks of MgH 2 were discovered, demonstrating that the Mg was fully converted to MgH 2 , as per Equation (3). The peaks of MgO and MgTiO 3 were still detected. The appearance of the MgO and MgTiO 3 species occurred, indicating that MgH 2 reacts with CoTiO 3 throughout the heating process. For the MgH 2 + 50 wt.% CoTiO 3 sample, similar peaks are presented in the absorption sample (Figure 10b), which are the Mg peaks superseded by the MgH 2 . New peaks of MgO, MgTiO 3 , CoMg 2 , and CoTi 2 were also detected, suggesting that interaction between MgH 2 and CoTiO 3 may occur during the heating process, as indicated in (Equation (4)). The mechanism of the absorption/desorption process of the MgH 2 -CoTiO 3 system is shown in Figure 11. 11MgH 2 + 3CoTiO 3 → 6MgO + MgTiO 3 + 2CoMg 2 + CoTi 2 + 11H 2 (4) Nanomaterials 2022, 12, x FOR PEER REVIEW Figure 10 presents the MgH2 + 10 wt.% CoTiO3 and MgH2 + 50 wt.% CoTiO3 after absorption at 200 °C. For the MgH2 + 10 wt.% CoTiO3 sample (Figure 10a), t peaks of MgH2 were discovered, demonstrating that the Mg was fully converted t as per Equation (3). The peaks of MgO and MgTiO3 were still detected. The appea the MgO and MgTiO3 species occurred, indicating that MgH2 reacts with CoTiO3 t out the heating process. For the MgH2 + 50 wt.% CoTiO3 sample, similar peaks sented in the absorption sample (Figure 10b), which are the Mg peaks supersede MgH2. New peaks of MgO, MgTiO3, CoMg2, and CoTi2 were also detected, sugges interaction between MgH2 and CoTiO3 may occur during the heating process, as in in (Equation (4)). The mechanism of the absorption/desorption process of the CoTiO3 system is shown in Figure 11.
11MgH2 + 3CoTiO3 → 6MgO + MgTiO3 + 2CoMg2 + CoTi2 + 11H2    (Figure 10a), the main peaks of MgH2 were discovered, demonstrating that the Mg was fully converted to MgH2, as per Equation (3). The peaks of MgO and MgTiO3 were still detected. The appearance of the MgO and MgTiO3 species occurred, indicating that MgH2 reacts with CoTiO3 throughout the heating process. For the MgH2 + 50 wt.% CoTiO3 sample, similar peaks are presented in the absorption sample (Figure 10b), which are the Mg peaks superseded by the MgH2. New peaks of MgO, MgTiO3, CoMg2, and CoTi2 were also detected, suggesting that interaction between MgH2 and CoTiO3 may occur during the heating process, as indicated in (Equation (4)). The mechanism of the absorption/desorption process of the MgH2-CoTiO3 system is shown in Figure 11.
11MgH2 + 3CoTiO3 → 6MgO + MgTiO3 + 2CoMg2 + CoTi2 + 11H2 (4)   On the basis of the above result, we postulate that the MgH 2 -CoTiO 3 system presents a superior hydrogen storage performance of MgH 2 due to the synergetic effect of the in situ formation of MgTiO 3 , CoMg 2 , CoTi 2 , and MgO. These active species act like active sites at the surface of the MgH 2 matrix and provide a fast channel for the diffusion of H atoms in the absorption and desorption process [73,74]. For instance, a previous study stated that adding MgTiO 3 ameliorates the sorption kinetics performance of MgH 2 [75]. The MgH 2 + MgTiO 3 absorbed 5 wt.% of H 2 in 500 s at 307 • C. Gao et al. [69] also presented the notable performance of MgH 2 with the formation of Co-containing material after the desorption and absorption processes. It is believed that 3D electron orbitals in the transition metal stimulate the dissociation of H 2 molecules by serving as the antibonding of H 2 molecules. Thereafter, Ares-Fernandez and Aguey-Zinsou [50] presented faster absorption/desorption rates of MgH 2 when it was doped with MgO. In another work, they stated that MgO may take part as a process control agent that can reduce and prevent MgH 2 from agglomerating by achieving an optimal breakage rate [76]. The catalytic effect of MgTiO 3 , CoMg 2 , CoTi 2 , and MgO could work together to provide a synergetic effect and further boost the performance of MgH 2 . Nevertheless, more research is needed, such as applying X-ray photoelectron spectroscopy and high-resolution transmission electron microscopy, to elucidate the catalytic function of CoTiO 3 on the hydrogen storage properties of MgH 2 .

Conclusions
The hydrogen storage performance of MgH 2 was enhanced with the addition of nanosized CoTiO 3 , which was synthesized via the solid-state method. The different weight percentage of CoTiO 3 was added into MgH 2 , and the 10 wt.% of the CoTiO 3doped MgH 2 sample present a superior performance of hydrogen storage properties. The MgH 2 + 10 wt.% CoTiO 3 started releasing hydrogen at a temperature of 275 • C, 65 • C lower than as-milled MgH 2 (340 • C). The MgH 2 + 10 wt.% CoTiO 3 composite also presents a faster absorption and desorption rate when it can absorb 6.4 wt.% H 2 in the first 10 min and performs a 23-times-faster desorption rate when compared with as-milled MgH 2 . Moreover, the activation energy of the MgH 2 decreased from 135 kJ/mol to 104.6 kJ/mol after the addition of 10 wt.% of CoTiO 3 . Adding nanosized CoTiO 3 also results in a smaller and fine particle that promotes the favorable improvement of MgH 2 performance. The superior performance of MgH 2 with the addition of CoTiO 3 was also attributed to the in situ formation of MgTiO 3 , CoMg 2 , CoTi 2 , and MgO during the heating process, which offers a synergetic catalytic role in improving the performance of MgH 2 . These findings may be beneficial for the modification of the MgH 2 system for solid-state hydrogen storage in the future.

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