1 Introduction

In recent decades, the energy crisis and global warming have promoted a growing demand for renewable clean energy [1, 2, 3]. As a clean and sustainable energy resource, hydrogen (H2) has been hailed as a future fuel that holds great promise in replacing ever-being-exhausted fossil fuels and aiding the transition to net-zero emissions [4, 5]. Hydrogen is the most abundant element in the universe and its combustion produces only water [6, 7, 8]. However, many obstacles need to be overcome to realize the so-called “hydrogen economy”. The usage of hydrogen energy includes hydrogen production, hydrogen storage, hydrogen transportation, and hydrogen utilization. As known, the main challenge for the applications of hydrogen energy is the development of suitable approaches for hydrogen storage [9].

Economic, efficient, and safe hydrogen storage methods play a crucial role in exploiting hydrogen energy, reducing carbon emissions, and improving the utilization efficiency of renewable clean energies [10]. Compressed gaseous hydrogen storage technologies have long taken a dominant position in this field due to their simple, convenient, low energy consumption, and efficient advantages [11]. However, high-pressure gaseous hydrogen storage technology has the disadvantage of low volumetric hydrogen storage density and the safety risk of leakage and explosion. Cryogenic liquid hydrogen storage is limited by high cost and energy consumption [29, 30, 31], and catalyzing [32, 33, 34, 35, 36]. Alloying is a simple and mature method to modify Mg/MgH2. By adding alloying elements to the Mg/MgH2 system to change its hydrogen ab/de-sorption reaction paths, the thermodynamic properties of MgH2 can be effectively improved. In 1968, Reilly et al. [37] first discovered that the intermetallic compound Mg2Ni, formed by introducing the alloying element Ni into Mg/MgH2 system, presented excellent hydrogen ab/de-sorption thermodynamic performances. Subsequently, Fe, Co, Si, Cu, and other alloying elements were also introduced into Mg/MgH2 system to investigate the hydrogen storage properties of their corresponding alloy compounds. However, the biggest drawback of alloying is that the introduction of alloying elements will lead to capacity loss, and some Mg-based hydrogen storage alloys exhibit an irreversible hydrogen ab/de-sorption process. Additionally, it is worth emphasizing that MgH2 is a semiconductor with no available electronic states at the Fermi level [38], having almost no catalytic activity toward H2 surface reactions. In contrast, The Fermi level of transition metals is located around s-type orbitals, which is a necessary condition to promote H2 surface reactions [39]. Hence, catalyzing has been a widely studied and efficient method to improve the hydrogen storage kinetics of MgH2 since the 1990s. The scope of these catalysts extends from transition metals to transition metal oxides, halides, and carbides. In addition to the catalyzing, the preparation of hydrogen storage composites by compounding complex hydrides with MgH2 is also a hot research topic in recent years [40, 41]. Although the alloying and catalyzing could indeed ameliorate the performances of MgH2 from the extrinsic perspective, the improvement is restricted by the inevitable agglomeration and growth of the additives during cycling, as well as the low density of exposed active sites. Hence, from an intrinsic point of view, nano-sized MgH2 has been widely studied through various nanotechnologies giving rise to improved hydrogen storage performances of MgH2 in recent years.

Based on the above discussions, nano-structuring has been regarded as one of the most efficient methods to destabilize MgH2 and minimize the decomposition enthalpy among various modification strategies [42, 43]. Nano-sized materials have peculiar properties that could not be expected in the bulk phase. The surface layer atoms of nano-sized materials in the sub-stable state with high surface energy are highly susceptible to combining with other atoms and converting to the stable state, due to the presence of a large number of unsaturated bonds [44, 45]. It is worth emphasizing that the essence of the hydrogen ab/de-sorption process of Mg-based hydrogen storage composites is the bonding and breaking of Mg and H atoms, the nanosizing of MgH2 has promising applications in the improvement of thermodynamics and kinetics of Mg-based hydrogen storage materials. Furthermore, it is important to distinguish between grain size, particle size, and crystal size in order to accurately describe the effects of each factor on the kinetics and thermodynamics of MgH2. The particle size refers to the size of individual particles of a material. The particle size of MgH2 can affect its thermodynamics and kinetics by altering the surface area-to-volume ratio. In general, smaller particles have a larger surface area-to-volume ratio than that of larger particles, which can enhance the reactivity and reaction rate of MgH2 by increasing the number of active sites available for de/re-hydrogenation. The crystal size of MgH2 refers to the size of individual crystals in a material. When MgH2 is in a crystalline state, the size of the crystals can impact the nucleation and growth of new phases, as well as the overall crystal structure of the material, which can affect the reaction kinetics and thermodynamics. In general, smaller crystal sizes can promote faster nucleation and growth of new phases, as well as increase the number of defects and grain boundaries that can serve as reaction sites. Furthermore, the downsizing of the MgH2/Mg crystal leads to a change in the lattice, which will result in a change in desorption energy via the changed lattice energies, thereby destabilize MgH2. The grain size of MgH2 is related to the crystal size, but refers specifically to the size of the individual grains in a polycrystalline material. In the case of MgH2, smaller grains can provide a greater surface area, more nucleation sites, and shorter diffusion path for hydrogen absorption and desorption, similar to smaller crystal sizes. Thus, the main reasons for the amelioration of the hydrogen storage properties of Mg-based materials by nanosizing are as follows:

  1. (a)

    With the increase of the specific surface area of nano-sized MgH2, the contact area between the matrix and hydrogen is increased, accelerating the diffusion rate of hydrogen to the matrix. Besides, the contact area between the MgH2/Mg and catalysts is also increased, improving catalytic efficiency.

  2. (b)

    With the nanosizing of MgH2, the surface energy increases, facilitating the adsorption of H2 on the surface of Mg particles and the destabilization of MgH2.

  3. (c)

    The high density of grain boundaries among nanoparticles provides more diffusion paths for hydrogen atoms, improving the hydrogen storage kinetics of MgH2.

  4. (d)

    Reducing the size of MgH2 effectively shortens the diffusion path of hydrogen, enhancing the kinetic properties of MgH2/Mg.

The preparation methods of nano-sized Mg-based hydrogen storage materials include ball-milling, vapor deposition method, plasma metal reaction, chemical reduction of Mg precursors, and nanoconfinement. Many high-quality reviews have been published in the last decade covering the thermal destabilization and catalytic tuning of the kinetics of Mg-based hydrogen storage materials [14, 20, 46, 47]. However, a specific focus on nanostructure engineering of Mg-based hydrogen storage materials is still missing and necessary for their applications in future energy storage.

Herein, we make great efforts to present a comprehensive overview of fundamental theory and synthetic methodologies for intricate nanostructured Mg-based hydrogen storage composites, as depicted in Scheme 1. The fundamental theories of hydrogen storage regarding MgH2 are highlighted with special emphasis on thermodynamics, kinetics, and cycling stability. This review paper summarizes the latest trends in the design of nanostructured Mg-based hydrogen storage materials, important breakthroughs in the field, and the challenges for Mg-based composites applied in the commercial energy conversion and storage devices.

Scheme 1
scheme 1

Schematic illustrations of fundamental theories and synthetic strategies for nano-engineering of Mg-based hydrogen storage materials

2 Fundamental Theories of Hydrogen Storage in MgH2

2.1 Thermodynamics and Destabilization of the Mg/MgH2 System

The hydrogen ab/de-sorption process of hydrides is a dynamic equilibrium of three phases: hydrogen, metal, and the corresponding hydride (Fig. 1). As shown in Fig. 1a, hydrogen pressure, composition, and temperature are the crucial factors determining the phase equilibrium [48]. During the process of isothermal hydrogenation, a solid solution (α-phase) is first formed. With the increase of hydrogen pressure, the solid solution starts to transform into hydride (β-phase), the β-phase nucleates and grows, and the hydrogen pressure remains unchanged as the phase transformation proceeds. Until the α-phase completely transforms to the β-phase, the hydrogenation reaction completes. By calibrating the pressure–composition–temperature equilibrium point in the process of hydrogen ab/de-sorption, the PCT curve can be obtained (Fig. 1b). The relationship between plateau pressure (Peq) and temperature (T) in the PCT curve can be described by the van't Hoff equation:

$$\ln \left( {\frac{{p_{eq} }}{{P_{0} }}} \right) = \frac{\Delta H}{{RT}} - \frac{\Delta S}{R}$$
(2-1)
Fig. 1
figure 1

a Pressure–composition isotherm plot of \({\text{Mg }} + {\text{H}}_{{2}} \rightleftarrows {\text{MgH}}_{{2}}\) transition. b van't Hoff plot related to the phase transition of \({\text{Mg }} + {\text{H}}_{{2}} \rightleftarrows {\text{MgH}}_{{2}}\). The enthalpy and entropy of hydrogenation and dehydrogenation could be obtained from the slope and intercept, respectively. Schematic representations of the α-phase (left) and β-phase (right) of metal hydride are also presented

In this formula, P0 is the atmospheric pressure (1.01 × 105 Pa); ∆H and ∆S are the enthalpy and entropy of the hydrogen ab/de-sorption, respectively; and T is the absolute temperature; R is the gas constant (R = 8.314 J mol−1 K−1). According to the linear fitting between lnP and 1000/T, ∆H and ∆S can be calculated. Notably, the value of the re/de-hydrogenation enthalpy (∆H) is an important indicator to measure the strength of the Mg–H bond. The larger the absolute value of ∆H is, the stronger the Mg–H bond will be.

The thermodynamic stability of Mg/MgH2 system is mainly determined by the feature of the Mg–H bond. The Mg–H bond in the Mg/MgH2 system is covalent–ionic mixed [49], having a relatively high bonding energy of around 3.35 eV. The nature of Mg–H bond results in a high enthalpy for the decomposition of MgH2, which is around 74.7 kJ for releasing 1 mol of H2 [50], leading to a high temperature for hydrogen desorption. Under atmospheric pressure, MgH2 starts to release H2 at a temperature of 280 °C, which is far from the requirements of practical applications [51]. Many efforts have been made to thermodynamically destabilize the Mg/MgH2 systems, such as alloying of Mg with other elements, inducing the formation of metastable γ-MgH2 phase, and nano-structuring [52]. Cheung et al. [53] simulated the relationship between grain size reduction and decreased structural stability. They concluded that significant changes in the thermodynamics of hydrogen desorption are observed only if the grain size is reduced to below 2 nm.

2.2 Kinetics of the Mg/MgH2 System

The thermodynamic parameters of the hydrogen storage materials characterize the driving force of phase transformation during hydrogen absorption and desorption. While hydrogen storage performances of the hydrides are also related to the re/de-hydrogenation rate, i.e., the kinetic properties.

The hydrogenation process of Mg includes four main steps: H2 physisorption, H2 dissociation, H chemisorption, and H atom diffusion [55]. Each step needs to overcome an energy barrier, called the reaction activation energy (Fig. 2). The relevant experiments have revealed that the rate-determining step is the nucleation-growth of MgH2, which is determined by the hydrogen diffusion rate in Mg lattice [56]. However, the diffusivity of hydrogen in Mg lattice is quite low, which is measured to be 10−20 m2 s in bulk Mg and 10−18 m2 s along grain boundaries [57, 58]. It is worth emphasizing that the firstly formed MgH2 will act as a barrier to further diffusion of hydrogen into bulk Mg, limiting the reaction rate of the hydrogenation process [59]. The hydrogen desorption of MgH2 is mainly determined by the breakage of Mg–H bonds, diffusion of hydrogen atoms in the bulk phase, and recombination of hydrogen atoms. Accordingly, it is demonstrated that the sluggish kinetics (160 kJ mol−1 H2) severely limits the wide application of Mg/MgH2 system. Particularly, in the dehydriding stage of coarse particles, successively formed fresh Mg at the surface layer may function as the diffusion barrier to the hydrogen esca** from MgH2. However, in nano-sized MgH2 particles, the Mg phase may form simultaneously all through the material and the entire process of desorption is then governed by fast hydrogen diffusion rather than slow Mg–MgH2 boundary movement. In other words, particle size of MgH2 has a profound effect on its hydrogen storage kinetic properties [60, 61].

Fig. 2
figure 2

Reproduced with permission from Ref. [54]. Copyright 2021 Elsevier

Schematic diagram describing continuous energy barriers in the process of hydrogen absorption (up) and illustration of the kinetic steps in the hydrogen storage process (down).

To understand the mechanism of reaction kinetics, it is necessary to analyze kinetic models in the process of hydrogen ab/de-sorption. The kinetic models of hydrogen storage materials in the process of de/re-hydrogenation can be summarized in Table 1 [62, 63, 64].

Table 1 Kinetic models for hydrogen absorption and desorption process

Among them, the hydrogen ab/de-sorption process of most hydrogen storage materials can be described by the nucleation and growth mechanism (JMAK model), and the equation is as follows [65]:

$$\ln \left[ { - \ln \left( {1 - \alpha } \right)} \right] = n{\text{ln}}k + n{\text{ln}}t$$
(2-2)

In this formula, k is the rate constant of hydrogen ab/de-sorption; α is the reaction fraction corresponding to t; n is the dimension determining the abstract model. By establishing the linear relationship between \(\ln \left[ { - \ln \left( {1 - \alpha } \right)} \right]\) and \({\text{ln}}t\), the value of n can be obtained from the slope, and the reaction rate constant of k can be calculated by the intercept.

Furthermore, the activation energy (Ea) of hydrogen storage materials in the process of hydrogen ab/de-sorption can be obtained through the Arrhenius formula [66]:

$$k = Ae^{{ - \frac{{E_{a} }}{Rt}}}$$
(2-3)
$${\text{ln}}k = - \frac{{E_{a} }}{RT} + {\text{ln}}A$$
(2-4)

In which k is the rate constant, A is the pre-exponential factor, and Ea is the activation energy. Thus, the activation energy can be obtained by fitting lnk verses 1/T [the reaction rate constants of k at different temperatures can be calculated from Eq. (2-2)]. The intrinsic mechanism of the hydrogen ab/de-sorption kinetic properties of materials can then be explained from the perspective of activation energy.

Moreover, Kissinger’s method is also usually used to calculate the dehydrogenation activation energy of hydride, typically using differential scanning calorimetry (DSC) experiments, following Kissinger’s equation [66, 67]:

$$\ln \left( {\frac{\beta }{{T_{p}^{2} }}} \right) = - \frac{{E_{a} }}{{RT_{p} }} + C$$
(2-5)

In this formula, β is the heating rate; Tp is the peak temperature; R is the gas constant (R = 8.314 J mol−1 K−1); C is a constant. By fitting the linear relationship between \(\ln \left( {\frac{\beta }{{T_{p}^{2} }}} \right)\) and 1000/Tp, the activation energy (Ea) can be obtained using DSC results at different heating rates. Kissinger’s method has advantages of being simple and requiring few tests. However, only a single rate-limiting step is assumed, the application of the equation is limited.

Dong et al. [19] studied the sequential MgH2 dehydrogenation mechanism by analyzing the kinetic and structural changes during the layer-by-layer hydrogen desorption process through spin-polarized density functional theory calculations with van der Waals corrections (DFT-D3). The results showed that initial dehydrogenation barriers (2.52 and 2.53 eV) were much higher than the subsequent reaction barriers (0.12–1.51 eV) (Fig. 3a). Moreover, after the desorption of all surface atomic H, the degree of electron localization in this region dropped sharply, resulting in a burst effect (Fig. 3b, c). The results of AIMD simulations indicated that after the loss of all the surface atomic H, the atomic H of MgH2 was inclined to diffuse, and therefore, the dehydrogenation kinetics could be significantly improved (Fig. 3d), which inspired us the importance of promoting the initial dehydrogenation by structural engineering (such as nanostructure engineering) to facilitating the hydrogen desorption of MgH2. It was also demonstrated that the desorption energy of MgH2 decreased as the cluster size was reduced to below 19 Mg atoms [68]. For the smallest possible cluster, the desorption energy of MgH2 dropped even to negative values, which means that the nanostructure engineering makes MgH2 unstable (Fig. 3e).

Fig. 3
figure 3

Reproduced with permission from Ref. [19]. Copyright 2022 Royal Society of Chemistry. e Calculated desorption energies for MgH2 clusters with both the HF method and DFT method. Reproduced with permission from Ref. [68]. Copyright 2005 American Chemical Society. (Colour figure online)

a Comparison of the three-layer atomic H migration and dehydrogenation energy barriers. The calculated ELFs of MgH2 before the dehydrogenation b in processes 1 and 2 (the first layer) and c in processes 3 and 4 (the second layer). d AIMD simulations on MgH2 (110) before and after surface H loss. Yellow dots are the selected points of MgH2 before (0%) and after (100%) surface H loss at 500 K. White and orange spheres represent H and Mg, respectively.

Compared with the improvement in thermodynamics, the kinetic properties of Mg/MgH2 can be adjusted more easily and effectively. Numerous strategies are effective in improving kinetic performance, such as alloying, catalyzing, nano-structuring, etc., which will be discussed in the following sections.

2.3 Cycling Stability of the Mg/MgH2 System

Long-term cycling stability also plays a decisive role in the application of Mg-based hydrogen storage composites. The experimental results show that the long-term re/de-hydrogenation cycle at high temperatures will deteriorate the hydrogen storage capacity and the hydrogen ab/de-sorption rate [43, 69, 70, 71]. The reasons for the degradation of the system can be attributed to the following two factors. On one hand, the passivation of the Mg/MgH2 interface resulted from the reaction between Mg/MgH2 and contamination gas in H2, causing the loss of capacity [72]. On the other hand, the agglomeration and growth of nanoparticles driven by the interfacial energies increase the diffusion pathway for hydrogen and deteriorate the kinetics of Mg/MgH2 systems.

To improve the cyclic stability, apart from using purified H2 to prevent capacity loss, the nanoconfinement and encapsulation strategy are effective methods to restrict the agglomeration and growth of nanoparticles. The research of nanoconfinement will be discussed in detail in the following section.

3 Tuning the Properties of the Mg/MgH2 System Through Nanotechnology

The destabilization of MgH2 by nano-structuring has been widely investigated both theoretically and experimentally. As the size of particles reduces to nanoscale, the surface energy cannot be ignored. Due to the extra interfacial free energy stored at the boundary, the hydrogen ab/de-sorption temperature will be decreased, indicating that Mg-based hydrogen storage materials can be substantially destabilized by inducing nanocrystalline structure [73]. The equilibrium pressure of nano-sized MgH2 particles and corresponding bulk MgH2 during the hydrogen desorption process obeys the following relationship [51]:

$${\text{ln}}\frac{{P_{nano}^{eq} }}{{P_{bulk}^{eq} }} = \frac{1}{RT}\left( {\frac{{3V_{{{\text{Mg}}}} \gamma_{{{\text{Mg}}}} }}{{r_{{{\text{Mg}}}} }} - \frac{{3V_{{{\text{MgH}}_{2} }} \gamma_{{{\text{MgH}}_{2} }} }}{{r_{{{\text{MgH}}_{2} }} }}} \right)$$
(3-1)

In this formula, \(P_{nano}^{eq}\) and \(P_{bulk}^{eq}\) are equilibrium pressures of nanoparticles and corresponding bulk, respectively; V is molar volume; γ is the surface energy density; r is radius of the spherical particle. It can be deduced that part of the formation enthalpy will be stored as excessive surface energy when the contribution of the size effect becomes sufficiently large, resulting in the destabilization of the MgH2 nanoparticles. Particularly, as a typical phase transformation, the hydrogen ab/de-sorption process first takes place along the interface region, thus interfacial energy plays a crucial role in tuning the hydrogen storage performance. The formation of nanocrystalline with high density of interfaces will induce high extra energy stored in the interfacial region, which can reduce the energy barrier of de/re-hydrogenation [73, 74].

Thus, nano-structuring can significantly reduce the hydrogen ab/de-sorption temperature and increase the rate of re/de-hydrogenation of MgH2, due to the introduction of defects, shortening of hydrogen diffusion paths, increasing of nucleation sites, and destabilization of Mg–H bonding. However, due to the high surface energy, nanoparticles are susceptible to agglomeration and growth. It is essential for exploring an appropriate strategy to prepare Mg-based nanostructured composites. In the following sections, we will review the synthesis methods of nanostructured Mg-based hydrogen storage materials in detail.

3.1 Synthesis of Free-Standing Nano-sized Mg/MgH2

Wagemans et al. [68] theoretically investigated the influence of crystal grain size on the thermodynamic stability of Mg/MgH2. The results showed that both MgH2 and Mg become less stable with the decreasing of cluster size and the absolute value of enthalpy reduced dramatically when crystallite sizes were decreased down to less than 1.3 nm. Particularly, a lower decomposition enthalpy of 63 kJ mol−1 H2, corresponding to a desorption temperature of only 200 °C at 1 bar hydrogen pressure, may be obtained for 0.9 nm sized MgH2 crystallites. They indicated that the downsizing of the MgH2/Mg caused a change in the lattice, thus resulting in the reduction of desorption energy via the changed lattice energies. Impressively, they found that nano-sized Mg possessed the potential to uptake a few additional percent of hydrogen above the stoichiometric MgH2 through a stepwise calculation on the hydrogen sorption processes. The extra hydrogen (10-15%) was not dissociated and adsorbed to the surface of MgH2 as hydrogen molecules, depending on the specific surface area of nano-sized MgH2, which was rare in bulk systems. It is worth emphasizing that “excess” hydrogen is less strongly bound to the MgH2 structure and can therefore be desorbed at lower temperatures. Their quantum chemical study inspires us that it is possible to prepare MgH2 clusters that can adsorb extra hydrogen molecules on the surface due to its unique nano-structure, which will release hydrogen at lower temperatures, and is suitable for the operation of proton exchange membrane fuel cell (PEMFC). While the waste heat released by the PEMFC can be used for the dehydrogenation of stoichiometric MgH2, thereby achieving maximum energy utilization. However, these assumptions are simply not possible in bulk materials.

Experimentally, preparation of Mg/MgH2 nanoparticles seems to be difficult, due to the high reactivity of Mg. In 2008, Aguey-Zinsou et al. [75] successfully prepared surfactant-stabilized Mg nanoparticles with an average diameter of 5 nm through electrochemical synthesis, which exhibited unique hydrogen storage properties (Fig. 4a). Almost all hydrogen could desorb from colloidal MgH2 at a low temperature of 85 °C. This was the first time that hydrogen desorption near room temperature was achieved by nanosized MgH2. Except for MgH2/Mg nanoparticles, Chen et al. innovatively reported the non-confined Mg nanoparticles with different morphologies (nanowires, nanoflakes, nanorods, and sea-urchin-like shapes) via a vapor-transport method, which displayed enhanced hydrogen-sorption kinetics (Fig. 4b) [76, 77]. However, unsatisfactorily, the minimum diameter of these nano-sized Mg was larger than 30 nm. Mg nanocrystals of controllable sizes smaller than 30 nm were successfully synthesized in gram quantities by Norberg et al. [44] through chemical reduction of magnesocene using a reducing solution of potassium (Fig. 4c). The prepared Mg nanocrystals with smaller diameters exhibited dramatically faster hydrogen sorption kinetics, attributed to the reduction of particle size and increase of the defect density.

Fig. 4
figure 4

Reproduced with permission from Ref. [75]. Copyright 2008 American Chemical Society; b1b3 SEM images of Mg nanowires with a diameter of 30–50 nm, 80–100 nm and 150–170 nm, respectively. b4 TEM and HRTEM images of Mg nanowires with a diameter of 30–50 nm. b5 Hydrogen absorption and b6 desorption curves of the Mg nanowires with different diameters (30–50 nm, triangle; 80–100 nm, circle; 150–170 nm, square). Reproduced with permission from Ref. [77]. Copyright 2007 American Chemical Society; c1 TEM images of Mg nanocrystal samples (scale bar = 100 nm). c2 Hydrogen absorption and c3 desorption curves of the Mg nanocrystal samples at different temperatures. Reproduced with permission from Ref. [44]. Copyright 2011 American Chemical Society; d1 TEM image and d2 HRTEM images of non-confined ultrafine MgH2. d3 TGA curves of bulk MgH2 and non-confined ultrafine MgH2. d4 Hydrogenation profile with temperature of bulk MgH2 and non-confined ultrafine MgH2, and isothermal hydrogenation curves of non-confined ultrafine MgH2. d5 Comparison of the energy barriers for the hydrogen absorption and desorption of bulk MgH2 and non-confined ultrafine MgH2. Reproduced with permission from Ref. [78]. Copyright 2021 Royal Society of Chemistry

Synthesis of the free-standing nano-sized Mg/MgH2. a1 TEM image of the Mg colloid synthesized by electrochemical method. a2 Mass spectrometry of H2 desorption for the MgH2 (hydrided state) and the Mg colloids (non-hydrided state) at 85 °C. a3, a4 TG-DSC signals of the Mg colloid after H2 absorption (hydrided sate) and desorption (non-hydrided state).

However, unfortunately, it is difficult to synthesize free-standing ultrafine MgH2 nanoparticles (< 10 nm) due to their high surface energy, strong reduction trend, and high water–oxygen sensitivity. The particle size of MgH2 synthesized by the methods mentioned above is generally large (> 30 nm) and the improvements in hydrogen storage performance are limited, unable to meet the requirements of practical applications. Recently, Zhang et al. [78] synthesized non-confined ultrafine MgH2 nanoparticles (4–5 nm) by the metathesis process of liquid–solid phase driven by ultrasound (Fig. 4d). The ultrasound was used to provide the driving force for the formation of MgH2, while combining the mechanical oscillation generated by ultrasound to inhibit the agglomeration of particles. A reversible hydrogen storage capacity of 6.7 wt% at 30 °C was achieved, bringing MgH2 a step closer to practical applications. Furthermore, the results of DFT calculations revealed that the reaction barrier for the decomposition of nano-sized MgH2 was remarkably lower than that of bulk MgH2 (Fig. 5), indicating that nanostructure engineering of MgH2 is thermodynamically and kinetically favorable to the enhanced performance.

Fig. 5
figure 5

Reproduced with permission from Ref. [78]. Copyright 2021 Royal Society of Chemistry. (IS: initial state, TS: transition state, FS: final state. Mg: green. H: white). (Colour figure online)

a Computational structure models for bulk and nanosized Mg and MgH2. b Hydrogen uptake by Mg (001) slab and Mg cluster. c Diffusion energy barrier of H atoms in the Mg (001) slab and Mg cluster. d Hydrogen release from MgH2 (110) slab and MgH2 cluster.

3.2 Preparation of Nanostructured Mg-Based Composites via Mechanical Milling

Mechanical milling plays a significant role in the field of energy storage and conversation, such as batteries, catalysis and hydrogen energy. Particularly, ball-milling has been proven to be a facile method to prepare Mg-based composites in hydrogen storage fields. It has been found that the MgH2 and catalysts with various morphologies and compositions can be mixed evenly to form nanostructured Mg-based composites through the simple ball-milling method. Impressively, plentiful defects and nanocrystalline are ingeniously generated during the process of high-energy ball milling. The introduction of defects and reduction of particle size could provide more active sites for de/re-hydrogenation and shorten hydrogen diffusion paths, thereby improving the hydrogen ab/de-sorption kinetics.

As early as the late 1990s, Huot et al. [79] first prepared MgH2 nanocrystals by using ball milling method. They found that the specific surface area of MgH2 increased from 1.2 to 9.9 m2 g−1 after ball milling, due to the reduction of particle size. However, the simple ball milling showed no obvious improvement in the performances of MgH2. Usually, de/re-hydrogenation of MgH2 is a catalytic reaction process depending on the catalytic activity of catalysts. Therefore, most of the research is focused on the introduction of catalysts into MgH2 to fabricate nanostructured Mg-based composites via the ball-milling method. The combination of the disordered MgH2 structure induced by mechanical milling with catalysts gives rise to synergetic effects and excellent ab/de-sorption properties.

Numerous catalysts can be introduced into MgH2 system, including transition metals [33, 34, 70, 80, 81, 82, 83, 84], oxides [85, 84, Full size image

Notably, the bimetallic oxides (e.g., Ni3(VO4)2 [98], TiNb2O7 [103], carbon nanotube loaded multi-valence Co [34, Full size image