**1. Introduction**

Currently, as fossil energy is on the verge of disappearing [1] and pollution caused by fossil fuels is becoming more serious [2], it is urgent to develop clean energy. As an important type of clean energy, hydrogen (H2) energy has the advantages of being nonpolluting, being easy to be produced, and having extremely high energy density. Compared with other clean energy sources, such as geothermal energy, wind energy, and tidal energy, it is the best choice. The utilization of H2 energy involves many aspects, such as the production, transportation, storage, and utilization of H2 energy. Among them, what restricts the use of H2 energy is the storage technology of H2 energy. The storage of H2 energy can be roughly divided into two types: physical storage methods and chemical storage methods.

Physical storage includes high-pressure compressed gas storage and ultra-low temperature storage. Compressed H2 requires high-pressure vessels to be loaded, and these vessels should be strong, light, and anti-explosive under special circumstances. Actually, the volume density of H2 increases with pressure, but the weight density decreases at the same time. Therefore, as the weight density in the compressed gas system decreases,

**Citation:** Liu, Y.; Chabane, D.; Elkedim, O. Intermetallic Compounds Synthesized by Mechanical Alloying for Solid-State Hydrogen Storage: A Review. *Energies* **2021**, *14*, 5758. https:// doi.org/10.3390/en14185758

Academic Editor: Attilio Converti

Received: 3 August 2021 Accepted: 10 September 2021 Published: 13 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

improvements in the volume storage density are sacrificed. At the same time, the future pressure vessel is expected to consist of three layers: a polymer lining, a carbon fiber composite material (which is a stress-bearing component), and an outer aramid layer that can withstand mechanical and chemical damage [3]. This also drive the cost higher. In summary, although compressed gas H2 storage technology is simple and mature, lighter, safer, and low-cost containers still need to be researched. For ultra-low temperature liquid H2 storage, liquefaction and continuous evaporation of H2 require a relatively large amount of energy, which limits the application of liquid H2 storage in mobile equipment, such as vehicles. Liquid H2 storage is suitable for H2 in a relatively short period of time. Applications include being consumed, such as in the aerospace industry [4].

Chemical storage refers to the combination of H2 and materials through van der Waals forces or chemical bonds to store H2 in the H2 storage material in the form of atoms. This H2 storage method is the most attractive method because, for chemical H2 storage, reversible H2 absorption and desorption reactions can be carried out in a mild environment, and its H2 storage density is high, and there is no safety risk. There are many types of H2 storage materials, which can be divided into solid H2 storage materials and Liquid Organic H2 Carriers (LOHCs) according to their different material states.

The theoretical H2 storage capacity of LOHCs is extremely high, but the production of these liquid materials often requires extremely complicated steps, and a wide variety of catalysts are required to assist in the reaction when the reaction is at room temperature, so it is not suitable for the engines of vehicles; liquid materials are more suitable for H2 storage at fixed sites. Solid H2 storage materials are more reliable and safer than liquid and gaseous materials in ultra-low temperature storage [5].

Solid H2 storage materials can be divided into many types, such as intermetallic compounds, complex hydrides, chemical hydrides, etc. [6]. In this review, the focus is on intermetallic compounds.

In recent years, intermetallic compounds have attracted great attention because of their wide application in the development of H2 storage alloys [7]. There are many applications for intermetallic compounds, including H2 storage systems, Nickel Metal Hydride (NiMH) battery electrodes, H2 sensors and catalysts, and cooling systems [8]. Intermetallic compounds are attractive in the development of H2 storage alloys because they can absorb large amounts of H2. In addition, they are abundant and have diverse ingredients. Generally, H2 reacts with intermetallic compounds to produce a solid solution of H2 in the respective compound or the formed hydride. The hydride produced by the intermetallic compound is called the intermetallic hydride. The general formula is *AmBnHx*. The interactions between metal atoms and interstitial H2 atoms determine its properties, so it largely depends on the crystal structure of the compound [9].

The materials used as solid-state H2 storage intermetallic compounds can be divided into A2B, AB, AB2, AB3, AB5, and BCC according to their composition. According to the different main component, it can be divided into Mg-based H2 storage alloy, Ti-based H2 storage alloy, RE-based H2 storage alloy, etc.

Mg and Mg-based alloys have long been considered potential H2 storage materials. Fifty years ago, scholars discovered that Mg metal and its alloys can be used as H2 storage materials. As a potential commercial H2 storage material, this material has some special advantages: (i) high abundance (2% of the composition of the surface of the Earth and almost unlimited in seawater), (ii) being non-toxic compared with other light elements and their hydrides that rapidly exothermic oxidation in the air, and (iii) being highly safe. In addition, the production technology of Magnesium (Mg) is very mature, and the cost of raw materials is very low.

Research on the synthesis of hydride nanoparticles of Mg and Mg-based alloys by the ball milling has been studied in depth. This method has a low cost, has a simple and convenient procedure, and is a technology that has been studied extensively by scholars. It is suitable for the synthesis of various low-particle-size powder materials and can even synthesize nanomaterials [10]. For H2 storage material synthesized by mechanical alloying, its particles have a nanometre diameter and, due to its higher specific surface area and larger number of defects, provide hydrogenation nucleation sites and shorter H2 diffusion distances. Therefore, a fine powder has better H2 reaction kinetics than big particles [11]. The effect of particle size on the absorption kinetics of Mg-based H2 storage alloys was shown in [12].

As an active metal, there is always an oxide layer on the surface of Mg. The activation process is to decompose the surface oxide layer that inhibits H2 absorption. This process is usually carried out at high temperatures. However, mechanical alloying methods can overcome these difficulties because the powder particles undergo severe plastic deformation during mechanical alloying. The oxide layer on the particle surface is then destroyed. Additionally, by synthesizing these alloys by mechanical alloying, the most significant improvement in hydrogenation/dehydration kinetics can be achieved [13]. A catalyst is usually needed to further accelerate the H2 and dehydrogenation kinetics of the alloy.

Some factors may restrict the hydrogenation kinetics of Mg and Mg alloys. These factors include (i) the formation of oxide layers on the surface that inhibits the penetration of H2 into the alloys [14], (ii) the slow dissociation rate of H2 molecules on the Mg surface, (iii) the low speed of the MgH2/Mg interface [15], and (iv) the slow diffusion of H2 through magnesium hydride [15]. Most of these factors can be solved by nanostructured Mg using mechanical alloying with the existing catalytic additives.

Titanium H2 storage alloys mainly include TiMn2 [16], Ti2Ni [17], TiFe, etc. The TiFe intermetallic compound and its doped derivatives are a potential H2 storage alloy because it is inexpensive and has a relatively high H2 storage capacity, about 1.8 wt.%; rapidly absorbs and desorbs H2; moderates H2 absorption conditions; and has abundant reserves, so it has a very high potential as a H2 storage material [18]. The most common method of preparing TiFe is to melt the element mixture of Ti and Fe at high temperatures in an inert atmosphere and then to place it at 800 to 900 ◦C for a long time. The activation of TiFe obtained in this way requires a high-temperature heat treatment (about 400 ◦C) under high vacuum conditions and high H2 pressure (about 5 MPa) [19]; at the same time, the materials are prone to poisoning. TiFe alloy also has many other shortcomings, that is, the poor absorption and desorption kinetics, the obvious lag of hydride formation and decomposition, and the sensitivity to gas impurities [20].

In the past 30 years, great efforts have been made to modify TiFe alloys, such as alloying with one or several elements to partially replace Fe or Ti; chemical surface modification; and mechanical alloying to reduce particle size or alloying with catalysts, additives and other methods. Surface modifications can catalyze the adsorption and dissociation of H2 molecules on the surface of the alloy by a certain transition metal or alloy deposited on the surface to promote the hydrogenation of the material or can use acids or alkalis to eliminate stable oxide layers. These methods also help to improve the anti-poisoning ability of TiFe [21]. The study by Edalati et al. showed that, through the process of High Pressure Torsion (HPT) [22,23] and Roll Groove (GR) [24], TiFe is activated after severe plastic deformation. TiFe samples activated by HPT and GR processes can absorb and release H2 even after being stored in the air for several months. The study on the activation mechanism of HPT-treated TiFe shows that various defects such as nanoparticle grain boundaries and micro-cracks are the pathways for H2 to pass through the oxide layer that hinders material transport on the surface. The main shortcoming of GR is that the sample is not easy to be fully activated immediately. The biggest disadvantage of the HPT process is that the sample amount is usually less than 1 g [25]. In order to expand the production of TiFe for H2 storage, it is necessary to develop another severe plastic deformation technology that improves activation. This technology must have simple procedures, low cost, and high efficiency in producing severe plastic deformation. Mechanical alloying may be the ideal technology.

RE-based H2 storage alloys can be divided into AB5 type alloys and superstructure alloys (AB3, A2B7, A5B19, AB4, etc.) according to their compositions. AB5-type H2 storage alloy is one of the first H2 storage alloys discovered, and it has also been used as a commercial H2 storage alloy for a long time in nickel–H2 batteries. The AB5 type H2 storage alloy has good H2 storage performance. Its typical alloy is LaNi5, which was first developed in 1969 by the Philips laboratory in the Netherlands.

After discovering the H2 storage capacity of AB5 alloy [26], an important research area focused on the engineering application of these compounds [27]. However, the AB5 alloys used are synthesized by the high-temperature equilibrium method [28].

At the beginning of the LaNi5 study, it was found that the capacity of this alloy decreases significantly during the cycles. After 100 charge and discharge cycles, the capacity drops by 40%, which also determines that pure LaNi5 is not suitable for battery applications. Microscopic studies have shown that, during the cycle, the alloy particles change from an initial size of 7 to 2 mm to fine powder after 25 cycles. This is caused by the repeated strain of the crystal lattice in the process of H2 absorption and desorption. The decrease in particle size leads to a longer exposure of the alloy surface to the electrolyte and increases in alloy corrosion due to the high affinity of LaNi5 for water. The needle-like crystals of La(OH)3 were detected by scanning transmission electron microscopy. During the continuous pulverization of LaNi5 particles, La atoms are thrown to the grain boundary, where La is oxidized to form La(OH)3. Therefore, the electrode is consumed by the corrosion reaction, and the H2 storage capacity is reduced [29].

Extensive research on H2 storage alloy LaNi5 is to partially replace La and Ni with other elements to reduce the volume expansion ratio of the alloy to its hydride [30]. The equilibrium pressure of the alloy can be reduced by introducing many elements: Cr, Co, Cu, Al, and Mn, etc. [31].

At the beginning of this century, RE–magnesium–nickel (RE–Mg–Ni)-based H2 storage alloys gradually emerged. This type of alloy has a unique structure with a long-period one-dimensional superstructure, in which AB5 units (CaCu5 type structure) and A2B4 units (Laves type structure) are accumulated in a rhombus at a ratio of 1:1 along the c-axis. Kadir et al. [32] showed this structure. Due to its unique structure, this type of alloy has the advantages of a high capacity and a good activation performance. Therefore, this type of alloy is expected to make up for the shortcomings of the low discharge capacity of AB5 type alloys, which has become one of the research hotspots.

Mechanical alloying is a kind of "non-equilibrium processing" technology. Since Libowitz et al. [33] first reported the synthesis of a metal alloy hydride ZrNiH3 by mechanical alloying in 1958, this method has always been a research hotspots in the field of H2 storage material synthesis. Unlike traditional melting methods, mechanical alloying has some unique advantages. First, the mechanical alloying procedure is simple, has a low cost, and uses simple equipment. Second, under the heat of positive mixing, mechanical alloying can also alloy two metals and can synthesize metastable alloys, amorphous alloys, or quasicrystals, etc. This is to expand the scope of additives and to develop new H2 storage alloys to lay a good foundation, which is difficult to achieve using melting methods. Mechanical alloying can realize the nanometerization of the particle size, which often greatly improves the H2 storage performance of the alloy. However, this method also has some drawbacks, such as powder contamination, difficulty in precise control of the composition, etc.

After the metal or alloy undergoes the mechanical alloying process, the particle size and crystal grain size are greatly reduced while the micro-strain and lattice distortion in the crystal increase. This has a great impact on the H2 storage performance of the material, mainly in terms of absorption/desorption kinetics and thermodynamics [34,35] as well as H2 storage capacity. First, the micro-strain and lattice distortion act on thermodynamics and kinetics, which can reduce the temperature of H2 desorption, etc. Second, for kinetics, grain refinement reduces the diffusion distance, and micro-strain and distortion increase diffusion and reduce the apparent activation energy. Additionally, the free energy of the material after ball milling is increased, which is released in the subsequent heating recovery process. For the H2 storage capacity, in the AB, AB5, and AB2 materials [36,37], a significant decrease in capacity was observed after ball milling, while the capacity of Mg2Ni and Mg

materials did not change significantly [38]. The reason may be the formation of amorphous or disordered structure and the intervention of impurities.

This review focuses on the application of mechanical alloying in the synthesis of intermetallic compounds as solid-state H2 storage materials.

#### **2. Application of Mechanical Alloying in Mg-Based Hydrogen Storage Alloys**

Magnesium in the pure form can absorb up to 7.6 wt.% of H2 (above 400 ◦C), but it has low stability and low H2 absorption/desorption kinetics [39]. Some known compounds, such as Mg–Ni compounds, Mg–Fe compounds, etc., seem to be possible alternatives to magnesium hydride. These materials sacrifice volume or mass H2 capacity to obtain a better balance of pressure, stability, or cost [40].

Mg–Ni series alloys can form MgNi2 alloy and Mg2Ni alloy as well as the newly emerged MgNi alloy. Among them, Mg2Ni alloy form Mg2NiH4 after H2 absorption. The H2 content is 3.6 wt.%, which is much higher than the commercial H2 storage alloy LaNi5, which has a H2 absorption of 1.4 wt.%. This alloy also has good H2 absorption and desorption kinetics at low temperatures. The theoretical electrochemical capacity of Mg2Ni can reach 1000 mAh/g. On the basis of binary alloys, researchers often add third elements such as Fe, Ti, La, etc. into Mg–Ni materials [41]. The addition of new elements changes the composition and content of the alloy phase and can also play a catalytic role, thus improving the H2 storage performance of the alloys. Due to the significant difference between the high vapor pressure of Mg and the different melting temperature of the constituent elements (melting temperature for Mg is 650 ◦C, and that for Ni is 1452 ◦C), it is difficult to produce Mg2Ni alloys through the traditional smelting method. The mechanical alloying technology can easily synthesize these alloys because it is a completely solid state reaction process. In 1995, Singh and Zaluski [42,43] used mechanical alloying to produce Mg2Ni. Mg2Ni prepared by mechanical alloying was found to be able to react with H2 at a relatively low temperature, even room temperature, while the alloy synthesized by the melting method requires a hydrogenation temperature of 250–350 ◦C and a pressure of 15–50 bar [44]. The H2 absorption and desorption properties of some Mg–Ni-based alloys are listed in [45].

N. Cui et al. [46] believe that the Ni in the Mg–Ni alloy has high electrocatalytic activity for the H2 desorption reaction of the alloy in the alkaline solution. Cold welding occurs during the mechanical alloying process. During this period, Ni as the second phase is evenly distributed. The Ni particles on the surface of the Mg alloy are the electrocatalytic active centres of the H2 evolution reaction. This is because they can reduce the charge transfer resistance, resulting in a significant increase in electrochemical capacity. At the same time, the mechanical alloying reduces the particle size and increases defect density caused by crystallization, which can improve the diffusion of H2 in the alloy. The Mg and Ni powder in the mechanical alloying process is not only a mixture of two components but also a new composite system with evenly distributed electrocatalytic active centers, large interface, small particle size, and high reactivity due to high defect density. However, he also proposed that excessive grinding damages the crystal lattice and causes a sharp drop in electrochemical capacity.

L. Zaluski et al. [43] prepared a nano-sized Mg2Ni alloy using the mechanical alloying method under the conditions of a ball-to-alloy ratio of 5 and a grinding time of 60 h. He found that the nanocrystalline Mg2Ni alloy synthesized by mechanical alloying showed better H2 adsorption performance than the alloy prepared by conventional methods and that the produced powder can easily absorb H2 without activation because the mechanically alloyed nanocrystalline produces many very active fresh surfaces in the ball milling process. Conventional polycrystalline Mg2Ni can react with H2 at a temperature higher than 250 ◦C, while nanocrystalline Mg2Ni can also absorb H2 at a lower temperature (for example, at 200 ◦C, which is below the structural transition temperature of Mg2NiH4 hydride). No activation is required. Pd can catalyze the H2 absorption kinetics of nanocrystalline Mg2Ni

at 200 ◦C. Nanocrystalline Mg2Ni with a small amount of Pd can absorb H2 even at room temperature, does not require activation, and has good kinetics.

The nanocrystalline Mg2Ni intermetallic compound formed by mechanically alloying Mg and Mg2Ni can quickly absorb H2 without activation. Mg + Mg2Ni composites need to be activated, but their H2 absorption rate is faster than Mg2Ni at 150 ◦C and 12 bar, and the capacity is 4.2 wt.% [47]. Some other additive were also investigated [48–51].

The nanocrystalline and amorphous MgNi alloy prepared by mechanical alloying is also a potential alloy as a negative electrode for NiMH batteries. For example, after 10 hours of ball milling, the MgNi alloy had an initial discharge capacity of 522 mAh/g [52]. In addition, unlike traditional AB5 and AB2 H2 storage materials, this MgNi alloy does not require any activation and absorbs H2 directly. They also have the advantages of being almost nontoxic and low cost. However, their H2 absorption and desorption kinetics still needs to be further improved, and the actual discharge capacity is difficult to reach its theoretical value. In addition, from the point of view of commercial applications, this alloy has poor cycle stability as a hydride electrode. For example, after only 20 charge and discharge cycles, the discharge capacity of the MgNi electrode decays by more than 70%. Such cycle stability prevents MgNi from being used as a battery electrode, so it must be improved by other methods. The decrease in capacity is related to the irreversible corrosion of the alloy electrode by the KOH in the battery. This reaction forms a Mg(OH)2 layer [53–55] on the surface of particles. This not only consumes the alloy itself but also greatly increases the charge transfer resistance at the alloy/electrolyte interface and may hinder the diffusion of H2 into and out of the alloy body [56]. The pulverization of the alloy during the H2 absorption and desorption cycle exacerbates this harmful phenomenon because the pulverization produces a new active surface and thus forms new additional Mg(OH)2 layers after making contact with the electrolyte.

Mustafa Anik et al. [57] studied the electrochemical properties of Mg2Ni and MgNi synthesized by Mechanical Alloying (MA). The results showed that the charge and discharge capacity of Mg2Ni alloy increased sharply with the increase in grinding time within 40 h. The capacity of the alloy for which the grinding time exceeds 40 h no longer increases. They also found that the electrochemical performance of MgNi is much better than that of Mg2Ni, and the charge–discharge reversibility of the Mg2Ni alloy is very poor. The lower initial discharge capacity and cycle stability of the Mg2Ni alloy are not only due to the blocking effect of the Mg(OH)2 layer but also maybe owing to the highly irreversible reaction of the alloy. The author believes that the existence of free electrocatalytically active Ni particles on the surface of the MgNi particles is the main factor that promotes the H2 transfer reaction on the surface of the alloy.

Chiaki Iwakura et al. [58] dissolved Ti and V into MgNi alloy by mechanical alloying. They found that the two-element solid-solution amorphous Mg0.9Ti0.06V0.04Ni alloy prepared from MA is better than the single-element solid-solution Mg0.9Ti0.1Ni or Mg0.9V0.1Ni alloys show better cycle performance. The AES depth distribution shows that, after the charge–discharge cycle in an alkaline solution, the oxide layer on the surface of the Mg0.9Ti0.06V0.04Ni alloy is thinner than the surface of the Mg0.9Ti0.1Ni or Mg0.9V0.1Ni alloy. The XRD data show that a composite oxide layer composed of Ti and V species precipitates on the surface of the alloy particles, which may be the reason for the synergistic effect of the solid solution of the two elements to promote the charge–discharge cycle performance.

Stephane Ruggeri et al. [59] synthesized a MgNiTi ternary alloy by adding Ti through ball milling on the basis of MgNi alloy. It was found that the initial discharge capacity of MgNi0.95Ti0.05 (C1 = 575 mAh/g) was significantly increased compared with the MgNi electrode (C1 = 522 mAh/g). The author believes that this promotion of initial discharge capacity may be related to the formation of a multi-phase structure. The initial discharge capacity of Mg0.5Ti0.5*Ni* is 338 mAh/g, and after 10 cycles, the capacity stability is 75%. This cycle life is better than that of the initial MgNi alloy. The improvement in cycle life seems to be attributed to the formation of TiO2 that limits the formation of Mg(OH)2 on the alloy surface. However, the corrosion resistance of the electrode still needs further enhancement.

Jian-Jun Jiang et al. [60] believed that the modification of MgNi materials is very complicated. Substituting other elements for Mg does not significantly change its performance. Therefore, they used MA to synthesize amorphous MgNi alloy electrodes. He found that amorphous MgNi alloys do not require an additional activation process, and the mass transfer reaction largely depends on the oxide on the surface of the alloy particles, not on the electrode itself. The reaction rate is controlled by the film thickness. In the initial cycle, the film thickness is very low, so the electron exchange process is relatively smooth. In further charge and discharge cycles, the thickness of the oxide film increases and begins to hinder the transport of electrons. Additionally, this layer of film cannot provide an effective anti-corrosion effect due to its open structure. For the alloy electrode covered by the oxide film, the exchange current density greatly decreases as the thickness of the oxide film increases. Therefore, if this alloy electrode is used in a strong alkaline solution, surface modification must be considered to improve performance.

O Elkedim and L Huang et al. [61] researched the substitution of Mn and Ti in the Mg2Ni phase by first principles Density Functional Theory (DFT) calculations. The results show that the doping of Mn reduces the thermodynamic stability of Mg2Ni. When Ti is doped into Mg2Ni, with the increase in Ti content, the thermodynamic stability of Mg2Ni gradually decreases. That is, the doping of Ti and Mn can promote the thermodynamics of H2 release of Mg2Ni.

Amirkhiz et al. [62] pointed out that the addition of Single Wall Carbon Nanotube (SWCNT) can be used as a grinding aid in mechanical alloying to prevent the powder from consolidating on the surface of the ball and the container. Yao et al. [63] also believed that carbon materials have high dispersibility and catalytic activity, so they can promote the mechanical alloying process of ductile metals to produce finer particles. They believed that SWCNT can penetrate the thin hydroxide layer on the surface of MgH2 and can act as a 'H2 pump' to move H2 atoms to the surface. Multiwall Carbon Nanotube (MWCNT) also has a similar function. L.W. Huang et al. [64] studied the effects of Al substitution and the addition of MWCNTs about their structure and electrochemical properties of Mg2Ni alloys. They found that the ground alloy particles showed smaller particle size, agglomeration and better dispersibility, indicating that MWCNT can act as a grinding aid. After ball milling, most MWCNTs retain their tubular structure. All ball milled alloys with additives show excellent activation properties. However, adding MWCNTs is difficult to enhance the cycle life of the electrode. On the one hand, the discharge capacity is improved by the refinement of alloy particles. On the other hand, due to the enhanced dispersibility, more alloy particle surfaces are exposed to the KOH solution, which weakens the corrosion resistance of the ball milled alloy.

Based on the high H2 capacity of Mg and certain Ti-based alloys, scholars have developed new compounds of Mg–Ti–Ni–Fe alloys. Guo et al. [65] used mechanical alloying to synthesize Mg76Ti12Fe12−*x*Ni*x*(*x* = 0, 4, 8, 12) alloy and studied the effect of ball milling time on the H2 storage performance of the alloy. The results show that increasing the grinding time up to 80 hours increases the amount of Mg amorphous phase, which reduces the H2 storage performance. They also compared the H2 storage performance of alloys of various compositions after grinding for 40 hours and found that the storage capacity of alloys doped with Ni and Fe at the same time is much higher than that of alloys doped with single elements. In addition, the H2 absorption plateau pressures of Mg76Ti12Fe8Ni4 and Mg76Ti12Fe4Ni8 are reduced. The author explains that this phenomenon is due to the simultaneous formation of Mg2Ni and NiTi phases during the mechanical alloying process. They also observed that, as the Ni content increases, the Fe content decreases and that the hysteresis between the H2 absorption and desorption curves gradually decreases.

Mechanical alloying has also been applied to the synthesis and optimization of some binary magnesium-rich intermetallic compounds, such as LaMg12, La2Mg17, etc. X. P. Gao et al. [66] used mechanical alloying to add Ni to the LaMg12 alloy and studied its electrochemical H2 storage performance. When the weight ratio of LaMg12 to Ni is 1:3, the maximum discharge capacity of the alloy can reach 1010 mAh/g. However, its cycle performance needs to be further improved. Zhang Yanghuan et al. [67] synthesized LaMg11Ni + x wt.% Ni(x = 100, 200) alloys by mechanical alloying and studied the gaseous and electrochemical H2 storage kinetics. They found that increasing the Ni content can improve the kinetics, which is attributed to the decrease in activation energy and enthalpy. The milling time also affects two kinetics; the gaseous H2 absorption kinetic and electrochemical kinetic have maximum values with the change in milling time. LI Xia et al. [68] put Ni and La2Mg17 alloy together for ball milling and obtained La2Mg17-x wt.%Ni(x = 0, 50, 100, 150, and 200) alloy. They found that the increase in Ni increases the proportion of the amorphous phase. This leads to a decrease in H2 absorption capacity and cycle stability but an increase in discharge capacity. The H2 absorption capacity of La2Mg17-50 wt.%Ni is 5.796 wt.% (3 MPa), and the maximum discharge capacity is 353.1 mAh/g.

MY Song et al. [69] believed that, under certain conditions, the nucleation process controls the hydrogenation and dehydrogenation reactions of Mg, so nucleation can be promoted by creating a large number of defects, and shorten the H2 diffusion distance by reducing the particle size of Mg particles. Therefore, the hydrogenation and dehydrogenation kinetics of magnesium can be improved by mechanical alloying. This method generates many defect nuclei on the surface and/or inside of magnesium through severe plastic deformation or can be added by adding additives act as active sites for nucleation and, at the same time, greatly reduces the particle size of magnesium to shorten the diffusion distance. They also believe that mechanical alloying changes the rate control step of the hydrogenation reaction because it promotes the diffusion of H2, so the control step becomes the gas-phase mass transfer and the chemical adsorption of H2.

Some complex additives have also been applied to improve the H2 storage performance of pure Mg. Mykhaylo Lotoskyy et al. [70] found an outstanding effect of graphite and TiH2 adding to Mg by high-energy reactive ball milling. This composite can reversely absorb 6 wt.% H2. The addition of graphite greatly increases the cycle stability of the material. The author believes that graphite not only reduces the recrystallization of particles during the cycle but also helps to further refine the particles. C Zhou et al. [71] studied the effects of TiH2, TiMn2, and VTiCr on the kinetics of Mg hydrogenation. They used ball milling to produce alloy powder. The results show that the cycle kinetics of the material with VTiCr is better. The author found that the cycle kinetics of the material is relatively good at 300 ◦C but that the kinetics deteriorates severely at low temperatures (25–150 ◦C). The author attributed it to the growth process of microscopic grains during the reaction. Pavel Rizo-Acosta et al. [72] reported the effect of Early Transition Metal (ETM) hydride (ETMH*x*) as a catalyst on the performance of cycling H2 storage of Mg. They used reactive ball milling to mix the hydrides of Sc, Y, Ti, Zr, V, and Nb with Mg and found that the presence of ETMH*<sup>x</sup>* is beneficial to the decomposition of H2 molecules. Later, due to the high diffusion coefficient, H can diffuse rapidly in ETMH*x*, resulting in The rapid nucleation of MgH2. For cycle life, ETMH*<sup>x</sup>* can limit grain growth and improve structural stability. Among all of the additives studied, TiH2 has the best performance. The reversible H2 capacity for 20 cycles is 4.8 wt.%.

#### **3. Ti-Based Hydrogen Storage Alloys**

In 2000, Chiang, C.H., et al. [73] studied the hydrogenation performances of TiFe, TiFe2, and Ti during mechanical alloying in the H2 gas atmosphere. They found that, through reaction ball milling, TiFe can directly absorb H2 without activation. Single-phase TiH is produced during the mechanical alloying. In addition, TiFe1.924 can also be hydrogenated by mechanical alloying in H2 because the alloy decomposes to form TiFeH, TiH, and Fe during this process. Based on the hydrides of Ti, TiFe and TiFe2 and the thermal stability of the product powder during the ball milling process, they proposed that the TiFe milling reaction in H2 includes four steps: (1) The particles are broken to produce a fresh surface. (2) The powder absorbs H2. (3) H2 supersaturates in the powder. (4) TiFeH decomposes into TiH and Fe.

The research results of Hoda Emami et al. [74] showed there exists a close relationship between the activation of TiFe alloy and its particle size. Annealed TiFe with a grain size of micrometers does not absorb H2. Rolled samples with submicron to micron grain sizes are partially activated. The samples processed by HPT have nano- to sub-micron grain sizes and are fully activated under 3 MPa H2 pressure. The ball-milled sample with a nanometer particle size is fully activated and absorbs H2 at a pressure as low as 1 MPa.

The results of the literature [75] also show that the grinding of TiFe alloy and the transformation of alloy particles to nanometer sizes greatly simplify the activation process. In 2012, V. Zadorozhnyi et al. [76] demonstrated the possibility of directly synthesizing nano-size TiFe by solid-state reaction method compounds from a single component Fe and Ti using a mechanical alloying method. They discovered the exothermic effect of TiFe during the mechanical alloying process and believed that it was caused by the accumulation of excessive internal energy in the form of defects during the synthesis process. This characteristic plays a crucial role in the powder compaction process after mechanical alloying. It enhances the adhesion of the powder particles, thus ensuring that they are more easily compressed into blocks. Their other experimental results [77] show that the compacted sample of TiFe powder synthesized by mechanical alloying maintains the H2 adsorption performance of powdered TiFe and still maintains a considerable capacity after 20 absorption–adsorption cycles. After the first absorption–absorption cycle, the TiFe sample that was not compacted by mechanical alloying was immediately destroyed. They proposed that the improved cycle stability of TiFe samples produced by ball milling can be attributed to the formation of a special microstructure that resembles a sponge and acts as a specific bridge between powder particles. The formation of these contact bridges may be due to the material accumulating a great amount of energy in the process of mechanical alloying.

Later, they proposed that, although the nanocrystalline state of the intermetallic compound formed by mechanical alloying plays a crucial role in the activation stage, it is not so important in the subsequent absorption–desorption cycle because of the lattice repetition caused by the reaction itself. Expansion and contraction also form nano-sized grains and introduce a large number of defects. Inui et al. [78] added the importance of lattice defects generated during the MA treatment of TiFe. This partially disordered structure leads to the expansion of the solid solution area in the pressure–composition– isotherm curve, the reduction of the plateau pressure, and it also makes it easier for H2 to enter the crystal grains.

They have conducted a lot of research on synthesizing TiFe by mechanical alloying and adding various elements for doping. They are doped with Mg and S, Co and Ni, and Al and Cr. The concentrations of Mg and S are as high as 2 and 1 at.%, respectively. These alloys TiFe + 1% S and TiFe + 2% Mg show 0.6–0.7 wt.% reversible H2 capacity. The S-containing alloy has a very simple activation procedure, namely heating to 100 ◦C in H2 and placing it for about 20 minutes [79]. The results of doping Co and Ni also showed that the content of the two elements is as high as 2 at. %, for the mechanical alloying ones, and the extension of the H solid solution region and the reduction in the (*α* + *β*) plateau have been found [80]. Their work also showed the results of doping Al and Cr. The concentrations of the two are 20 and 6 at.%. Mechanical alloying doped TiFe + 5 % Al and TiFe + 4 % Cr showed an extremely simple activation procedure and a reversible H2 capacity of 0.7 wt.%, and the alloys containing Al had a higher plateau pressure. Compared with unalloyed TiFe, the hysteresis was significantly reduced. This characteristic is related to the smaller lattice expansion when the b hydride phase is formed. They also reported the data of TiFe–Mn, TiFe–Zr, and TiFe–Cu alloys produced by ball milling [81].

Liang et al. [82] listed some of the ball milling parameters and the corresponding H2 storage performance.

G. K. Sujan et al. [83] mentioned some problems in the preparation of TiFe powder by mechanical alloying in his review. The main disadvantage is that, for mechanical alloying, the raw material must be powder and the production cost of titanium powder is very high. Reducing the adhesion of powder on the container wall caused by cold welding is very important to ensure the output and quality of the product. The literatures use process control agents, mainly cyclohexane and benzene, to effectively reduce powder adhesion. Falcao et al. [84] developed an alternative way to produce nanocrystalline TiFe, using TiH2 in the raw material instead of Ti powder, resulting in a higher yield of TiFe. Contamination during mechanical alloying (for example, carbides formed by organic grinding aids and chromium in stainless steel containers and balls) is also a serious problem.

Tohru Nobuki et al. [85] used a mechanical alloying method to quickly synthesize a TiNi alloy within 20 min. They demonstrated through the cross section of the powder sticking to the grinding ball that the formation of the alloy occurs through the inter-diffusion between thin layers of co-laminated pure elements. The hydrogenation thermodynamics and kinetics of short-term mechanical alloying TiNi are similar to TiNi obtained by melting.

Z. Zhang et al. [86] used mechanical alloying to dope Mg into TiNi alloy to produce TiMgNi*<sup>x</sup>* (*x* = 0.1, 0.5, 1, 2) alloy and studied its structure and H2 storage performance. The results show that the average discharge capacity decay of the samples is very low, less than 1.1 % per cycle. Among all of the samples, TiMgNi showed the highest discharge capacity. The author believes that this is related to the MgNi amorphous phase. They have conducted a lot of research on this ternary alloy synthesized by mechanical alloying, including adding additive [87,88] and its application in NiMH batteries [89]. They also used mechanical alloying to prepare alloy TiMgNi*<sup>x</sup>* (*x* = 0.2, 0.4, 0.6, 0.8, 1) samples with different Ni content and explored the effect of Ni content [88]. They found that, when the Ni content increases, the discharge capacity and activation performance of TiMgNi*<sup>x</sup>* alloy increase linearly. Other researchers also explored the properties of Ti–Mg–Ni alloy produced by ball milling [90,91].

B. Hosni et al. [92] synthesized Ti2Ni by mechanical alloying at room temperature, and studied its structure and H2 storage performance. The results show that the activation of the alloy is very simple, requiring only one charge and discharge cycle. As for the cycle life, with the temperature gradually increases, the electrochemical discharge capacity loss after several cycles increases. The maximum capacity increases with the increase in temperature, and at the same time, the corrosion current density decreases. The author believes that the decrease in the oxide surface layer caused this phenomenon. Additive can also be added into Ti2Ni by mechanical alloying [93].

Hailiang Chu et al. [94] used mechanical alloying to mix Ti0.9Zr0.2Mn1.5Cr0.3V0.3 alloy based on TiMn2 with LaNi3.8Mn0.3Al0.4Co0.5 (AB5) and La0.7Mg0.25Zr0.05Ni2.975Co0.525 (AB3.5) additives to study the structure and electrochemical performance. The addition of additives did not change the AB2 structure of the main body but significantly increased the electrochemical discharge capacity, reaching 310.4 mAh/g and 314.0 mAh/g, respectively. The author believes that the AB3.5 alloy can reduce the charge transfer resistance and that the AB5 alloy can improve the H2 diffusion of the AB2 alloy. Myong JinChoi et al. [95] mixed TiMn2 and TiFe by the mechanical alloying method, then added different content of Ni, and studied its electrochemical performance. The results show that the H2 absorption capacity of the TiMn0.9Fe0.55 composite alloy is about 0.9 wt.% and that the alloy containing 30 wt.% Ni has the highest discharge capacity of 110 mAh/g, and this value was maintained for 20 cycles. The author attributed the improvement in performance to Ni catalyzing the electrochemical reaction and at the same time improving the corrosion resistance of the alloy.

#### **4. RE-Based Hydrogen Storage Alloys**
