The Current Status of Hydrogen Storage Alloy Development for Electrochemical Applications

In this review article, the fundamentals of electrochemical reactions involving metal hydrides are explained, followed by a report of recent progress in hydrogen storage alloys for electrochemical applications. The status of various alloy systems, including AB5, AB2, A2B7-type, Ti-Ni-based, Mg-Ni-based, BCC, and Zr-Ni-based metal hydride alloys, for their most important electrochemical application, the nickel metal hydride battery, is summarized. Other electrochemical applications, such as Ni-hydrogen, fuel cell, Li-ion battery, air-metal hydride, and hybrid battery systems, also have been mentioned.


Introduction
Hydrogen storage alloys are important for a few electrochemical applications, especially in the energy storage area. The basic of electrochemical use of the hydrogen storage alloy can be described as follows: when hydrogen enters the lattice of most transition metals, interstitial metal hydride (MH) is formed. The electrons accompanying the hydrogen atoms form a metal-hydrogen band right below the Fermi level, which indicates that the interstitial MH is metallic in nature. While protons in the interstitial MH hop between neighboring occupation sites by quantum mechanical tunneling, the electrons remain within a short distance (3-10 angstroms) of the protons to maintain local charge neutrality. Under the influence of an electric field, electrons and protons will move in opposite OPEN ACCESS directions. In an electrochemical environment, a voltage is applied to cause electrons to flow, and the charges are balanced out by moving conductive ions through a highly alkaline aqueous electrolyte with good ionic conductivity. During charge, a negative voltage (with respect to the counter electrode) is applied to the metal/metal hydride electrode current collector, and electrons enter the metal through the current collector to neutralize the protons from the splitting of water that occurs at the metal/electrolyte interface (Figure 1a). This electrochemical charging process is characterized by the half reaction: The standard potential of this redox half reaction depends on the chosen MH and is usually as low as possible to maximize the amount of stored energy without exceeding the hydrogen evolution potential (−0.83 V versus standard hydrogen electrode). Zn is an exception. With a complete 3d shell, Zn is a natural prohibitor for hydrogen evolution and thus a more negative voltage is possible, which increases the operation voltage of Ni-Zn battery.
The most important electrochemical application for MH is the negative electrode material for nickel metal hydride (NiMH) batteries. Together with a counter electrode from the Ni(OH) 2 /NiOOH system, which has been used in NiCd and NiFe batteries as early as 1901 by Thomas Edison, the NiMH battery was first demonstrated by researchers in Battelle in 1967 with a mixed TiNi + Ti 2 Ni alloy as the negative electrode [1]. Commercialization of the NiMH battery was independently realized by Ovonic Battery Company, Sanyo, and Matsushita in 1989 with AB 2 and AB 5 MH alloys. NiMH battery development started from small cylindrical cells (0.7 to 5 Ah) for portable electronic devices and progressed to 100 Ah prismatic cells for electric vehicle applications. The first commercially available electric vehicle in the modern era was the EV1 produced by General Motors in 1999. It was powered by a 26.4 kWh NiMH battery pack. Since then, NiMH batteries have powered more than 5 million hybrid electric vehicles made by Toyota, Honda, Ford, and other automakers, demonstrating the robustness and longevity of the NiMH battery. Recently, the NiMH battery has ventured into the stationary application market with advantages in long service life, a wide temperature range, low costs averaged over the service life, abuse immunity, and environmental friendliness. Several reviews on the topic of MH used in NiMH batteries are available [2][3][4][5][6][7][8][9][10]. In this report, we present the recent progress since the last review made in 2010 [10].
Besides NiMH batteries, MH (most commonly the misch metal-based AB 5 MH alloy) can also be used in other electrochemical applications such as lithium-ion based batteries and metal-air batteries. Metal hydride electrodes have a potential window of 0.1 to 0.5 V versus Li+/Li and the lowest polarization among conversion electrodes. These MH electrodes have shown the capability for greater capacity and can be used as anode electrodes in lithium-ion battery [10][11][12]. An air-MH battery that utilizes a misch metal-based AB 5 alloy in conjunction with a perovskite oxide-based cathode has been demonstrated by several research groups [13][14][15]. New types of V-flow/NiMH [16,17] and lead-acid/NiMH hybrid batteries [18] have been developed at the University of Hong Kong. Pd-treated LaNi 4.7 Al 0.3 has been used in a Ni-hydrogen battery [19]. Another application of LaNi 5 is the use as a cathode in a photo-electrochemical cell for water decomposition [20].

Hydrogen Storage Alloys for NiMH Battery Negative Electrodes
The use of MH in NiMH batteries started with research conducted by Schmitt and Beccu at Battelle Memorial Institute (TiNi-based) [1,21] and William and Buschow at Philips Research Laboratories (LaNi 5 -based) [2,22]. Almost 50 years of research on this subject have been conducted. In 2001, nine requirements were established for a suitable MH alloy for NiMH batteries: high capacity, good electrochemical catalysis, easy formation, excellent corrosion resistance, suitable hydrogen equilibrium, good kinetics and efficiency, long cycle life, small pressure-concentration-temperature (PCT) hysteresis, and low cost [6]. Today, the main implementation of NiMH batteries has shifted from consumer portable devices to hybrid electric vehicle and stationary applications. Accordingly, new criteria for suitable MH alloy design requirements, such as low self-discharge, good kinetics at low temperature, fast proton diffusion in the bulk, low pulverization rate during service life, and endurance of high temperature storage, are required for these applications. The amount of metallic inclusions in the MH surface oxide after activation and the high-rate dischargeability (HRD) are characteristics that are essential in meeting many of these new requirements. While the former can be quantified by magnetic susceptibility, the latter is studied by analyzing the surface reaction current density and bulk diffusion constant. Typical values from several alloy systems are compared and listed in Table 1 for reference. Recent progress of MH alloys for NiMH battery applications is reviewed in the following sub-sections categorized by alloy system. Table 1. Properties comparison of several metal hydride (MH) alloys. Saturated magnetic susceptibility (M S ) is proportional to the total amount of metallic nickel in the surface after activation. Applied magnetic field corresponding to half of the saturated magnetic susceptibility (H 1/2 ) is inversely proportional to the average number of Ni atoms in a cluster. Surface exchange current (I 0 ) and diffusion constant (D) are qualitative measurements of the catalytic nature of the surface reaction and the proton transportation in the bulk of the alloy, respectively.

Rare Earth-Based AB 5 Alloys
Vucht et al. first reported the hydrogen storage capability of rare earth-based AB 5 intermetallic alloy in 1970 [27]. Since then, this alloy system has become the most widely used intermetallic alloy in all metal hydride applications. After over 40 years of research, many compositions, structures, processes, and electrode fabrication modifications have been performed. Recent efforts have focused on (1) cost reduction by introducing Fe and Cu into the alloy formula to reduce/eliminate expensive Co and (2) improvement in HRD and low-temperature performance. The effects of substituting Ni by Cu, Fe, and Mo on charge-transfer resistance at −40 °C are summarized in Figure 2. The unit of Ω·g for charge-transfer resistance is used instead of the traditionally used Ω to rid the contribution from amount of electrode material; therefore, the values can be compared fairly among various MH alloys. The effect of Fe on charge-transfer resistance performance is interesting and can be explained by the evolutions of two factors: surface reaction area and catalytic ability. While surface catalytic ability is increased monotonically as the level of Fe-addition increases, surface reaction area is increased by 1% of Fe but decreased by further addition [28]. Consequently, charge-transfer resistance decreases with 1% Fe due to the improvements in both surface reaction area and catalytic ability, increases with a little higher Fe-content as the surface area diminishes, and finally decreases (but not to the level achieved with 1% Fe) with high amount of Fe when the surface catalytic ability takes on a more dominating role. Other research works on AB 5 alloys are summarized in Table 2.   Figure 2. Plot of charge-transfer resistance measured at −40 °C by AC impedance as a function of Fe-, Mo-, or Cu-substitution in AB 5 MH alloy [28][29][30]. All three additives at the lowest substitution level contribute positively in lowering the resistance.

Laves Phase-Based AB 2 Alloys
AB 2 alloys for NiMH battery applications are composed of main phases belonging to a family of materials known as Laves phases: hexagonal C14 phase and face-center-cubic C15 phase. C36 phase may also be present but is difficult to distinguish from C14 in XRD analysis. The main controlling factor for the C14/C15 ratio is the average electron density (e/a). Nei et al. reported that chemical potential can be used to fine-tune the C14/C15 threshold when Ti, Zr, and Hf, which have the same number of outer-shell electrons, are used together [68]. The common minor phases are Zr 7 Ni 10 , Zr 9 Ni 11 , ZrNi, and TiNi. The microstructures [69][70][71][72] and the contributions [73][74][75][76][77] of these phases were studied extensively in recent years.
While many studies on AB 2 alloys were reported, most of them focused on the influences of A-or B-site substitutions on electrochemical properties. These results are summarized in Table 3. The effect of partial substitution of Ni by various modifiers on −40 °C charge transfer resistance is summarized in Figure 3.
As shown in Figure 3, both La and Si are very effective in improving the low-temperature performance. Additionally, the influence of stoichiometry (B/A ratio) on electrochemical properties was reported [78,79]. PCT hysteresis was correlated to the pulverization rate in AB 2 alloys [80][81][82]. Gas atomization and hydrogen annealing were introduced to produce AB 2 metal powder directly [83,84]. V-free AB 2 alloys were also developed to reduce raw material costs [85,86]. Table 3. Summary of recent research on Laves phase-based AB 2 MH alloys.

Base alloy Substitution Major effects Reference
C14-domintaed Al Al improves bulk diffusion and surface reactivity. Al and Co together improves all electrochemical performances [87,88] C14-domintaed B B improves HRD and low-temperature performance but decreases charge retention, capacity, and cycle life [85] C14-domintaed C C increases HRD and charge retention but decreases low-temperature, capacity and cycle life [85] C14-domintaed Co Co provides easy activation, improves/decreases capacity, better cycle life and charge retention, but impedes HRD [87,89,90]  C14-domintaed Cu Cu increases capacity, facilitates activation, but decreases HRD. [92] C14-domintaed Fe Fe facilitates activation, increases total electrochemical capacity and effective surface reaction area, decreases HRD and bulk diffusion, and deteriorates low-temperature performance [87,93] C14-domintaed Gd Gd improves low-temperature performance, but decreases charge retention, HRD, capacity, and cycle life [85] C14-domintaed La La improves capacity, HRD, and low-temperature performance with a trade-off of inferior cycle stability [94] C14  La-and Si-modified alloys demonstrate the lowest resistance.

Superlattice A 2 B 7 -Type (A 2 B 4 -AB 5 -Hybrid-Type, such as AB 3 , A 2 B 7 , A 5 B 19 , and AB 4 ) Alloys
Mg-containing superlattice alloys have been used extensively in Japanese-made NiMH batteries for retail market since the introduction by Sanyo (eneloop) [103][104][105]. Batteries with superlattice A 2 B 7 -type alloys as negative electrode exhibit much lower self-discharge compared to ones with traditionally used AB 5 alloys [105]. Therefore, NiMH battery's storing and ready-to-use-out-of-the-pack capabilities can be much improved by the use of superlattice alloy, which enable NiMH battery to grow much more competitive compared to primary battery. Furthermore, superlattice alloys have been reported to have superior capacity, cycle life, and high-rate performances compared to the conventional alloys and are becoming a dominating force in both consumer and hybrid electric vehicle markets. The multi-phase MH system, with an overall B/A ratio between 3 and 4, is composed of a number of AB 5 slabs placed between two A 2 B 4 slabs ( Figure 4). Mg is needed to lower the average metal-hydrogen bond strength in order to obtain the appropriate heat of hydride formation suitable for NiMH battery applications (≈ −39 kJ· mol −1 H 2 at room temperature and 1 atm), and its replacement amount for rare earth is about 30% and 15% for La and Nd, respectively. Mg mainly replaces rare earth elements on the A 2 B 4 slab. The distribution of Mg (which has a high vapor pressure during melting) in the alloy is particularly important because Mg-lean regions have a tendency to form AB 5 phase. Lin et al. presented a brief review of different types of A 2 B 7 alloys [106]. The superlattice MH alloys can be classified into three groups: La-only, La-Pr-Nd, and Nd-only. The addition of Ce promotes AB 5 phase and therefore is rarely used. The La-only group has the highest capacity but the shortest cycle life due to the easy oxidation of La. The Nd-only group has the best charge retention and cycle life performance but the lowest capacity. Properties of the La-Pr-Nd group fall between the properties for La-and Nd-only alloys. Annealing is the most convenient method to manipulate phase distribution, and its effect was reported on La-only [107][108][109][110][111][112][113], Nd-only [24], La-Gd [114], La-Nd [115], La-Pr-Nd [116,117], and La-Ce-Pr-Nd [118] alloys. While Nd-only and La-Pr-Nd superlattice alloys are used in low self-discharge and high capacity NiMH batteries, respectively, La-only superlattice alloy is not used in commercial products due to its low cycle life from pulverization [119]. However, La-only alloy is studied extensively in the research society [120] due to easy sample preparation. The recent progress in the NiMH battery applications of superlattice alloys is summarized in Table 4.

Ti-Ni-Based Alloys
The Ti-Ni-based system was the first MH alloy used in NiMH batteries in the early 1970 s. Although its development was interrupted by the fast growth of AB 5 alloy, the Ti-Ni system remains a popular research topic due to its low cost, high hydrogen storage capacity, and fast activation. Two types of Ti-Ni binary alloys, TiNi and Ti 2 Ni, are able to absorb large amounts of hydrogen and are candidates as negative electrode material in NiMH batteries. TiNi alloy exhibits polymorphism, which is the basis of its outstanding shape memory property that is fully utilized in many industrial applications. Upon cooling, TiNi transforms from a B2 cubic structure (austenite) to a B19' monoclinic structure (martensite). Early electrochemical studies on TiNi alloy demonstrated great activation performance and an electrochemical discharge capacity of 210 to 250 mAh· g −1 [159,160]. Compared to TiNi, Ti 2 Ni MH alloy had higher hydrogen storage capacity due to its higher content of hydride forming Ti in the formulation, but its electrochemical capacity was only 170 mAh· g −1 with its stronger metal-hydrogen bond strength, an indicator of dehydriding difficulty [160]. By combining TiNi and Ti 2 Ni phases, the electrochemical capacity increases up to 320 mAh· g −1 [160,161] with the assistance of synergetic effect between the two phases. During hydrogen desorption, TiNi, which has better desorption kinetics, first dehydrides and contributes to the overall electrochemical capacity; then, the hydrogen stored in Ti 2 Ni phase is transferred internally into the dehydrided portion of TiNi and discharged through TiNi. Without the assistance of TiNi phase, the hydrogen stored in Ti 2 Ni phase cannot be released due to the stronger meta-hydrogen bond strength. However, the cycle stability of such a biphasic system suffers due to the corrosion and oxidation of Ti 2 Ni [159,162].
Most research efforts on TiNi alloys focused on elemental modifications on both A-and B-sites. Different material manufacturing procedures, such as mechanical alloying, annealing, and different quench rates, were introduced to study the effect of structure on overall electrochemical properties. Emami et al. investigated the influence of Pd by partially replacing Ni [163,164]. While the unit cell of TiNi is enlarged by the larger-sized Pd, the electrochemical capacity is reduced since the Pd-substitution increases the stability of TiNi intermetallic alloy and decreases the stability of their hydrides according to the electronic calculations. Recently, a study on substituting Ni in TiNi with various modifiers revealed that Fe, Co, and Cr increase the electrochemical capacity up to 400 mAh· g −1 with good activation performance [165]. Zhao et al. fabricated amorphous Ti 2 Ni alloy by solid-state sintering and ball milling [166], and the resulting material shows improved cycle stability but also lower capacity compared to crystalline Ti 2 Ni. Annealing treatment was then proposed in addition to sintering and ball milling, which results in a thicker oxide layer on the surface of amorphous Ti 2 Ni and improves both capacity and charge retention of the alloy [167]. In order to increase the capacity, Zhao et al. developed another fabrication method for Ti 2 Ni: induction melting, ball milling, and annealing [162]. The product is amorphous and nanocrystalline in nature and demonstrates reasonable capacity and good cycle stability; however, the highest reported capacity of 363 mAh· g −1 for Ti 2 Ni MH alloy is obtained at higher temperatures while cycle stability suffers. Zr-substitution for Ti was used as another way to potentially increase capacity and improve cycle stability of amorphous Ti 2 Ni [168]. Zr was reported to destabilize amorphous Ti 2 Ni phase and promote TiNi phase when partially substituting Ti in ball-milled Ti 2 Ni from elemental powders [169]. An amorphous Ti 3 Ni 2 alloy was prepared to combine advantages from TiNi and Ti 2 Ni, and although its capacity was lower than the crystalline form, cycle stability was much improved [170].
Ti-based icosahedral quasicrystal MH alloy, which has a crystallographically disallowed five-fold rotational symmetry, has only recently been studied for NiMH battery applications. The icosahedral phase is believed to contain much higher densities of tetrahedral interstitial sites compared to normal crystals [171]. Since hydrogen atoms enter favorably into tetrahedral sites, the icosahedral phase can absorb a large amount of hydrogen. Two recent studies, with and without milling after melt-spin, were performed varying the Zr/Ni ratio in Ti 45 Zr 38 Ni 17 [172,173]. All alloys consisted of 100% or close to 100% icosahedral phase and demonstrated that higher levels of Ni increase capacity up to 86 to 88 mAh· g −1 ; however, this value is dramatically lower than the theoretical capacity [172,173]. By changing the Zr/Ti ratio in Ti 45 Zr 30 Ni 25 , it was shown that higher Zr deteriorates capacity. Also with annealing, an alloy that has been arc-melted and mechanically alloyed has higher capacity (up to 130 mAh· g −1 ) compared to an unannealed amorphous alloy [174]. Hu et al. improved capacity up to 278 mAh· g −1 with decent cycle stability by adding Mn in a melt-spun TiVNi-based alloy [175]. The effect of Sc-addition was reported by the same group in TiVNi-based quasicrystal [176], with extended cycle life.

Mg-Ni-Based Alloys
Due to Mg's abundance, low cost, light weight, and high hydrogen storage capacity (2200 mAh· g −1 theoretical electrochemical capacity [9]), Mg-based MH alloy continues to be an interesting research topic and a strong candidate for the negative electrode material of NiMH batteries. In order to be applicable for room temperature battery operation, a stoichiometry of Mg:Ni close to 1:1 is required. However, the MgNi intermetallic compound does not exist on the Mg-Ni binary phase diagram. Therefore, non-equilibrium fabrication methods, such as mechanical alloying, RF sputtering, laser ablation, and melt-spin, are often used to prepare MgNi alloys. Such material usually has a mixed structure of amorphous and nanocrystalline character. Despite the high capacity that MgNi alloy offers, its sluggish hydriding/dehydriding kinetics and poor corrosion resistance in alkaline media prevent MgNi from use in practical application. Various types of modifications to the MgNi system were studied to improve the overall electrochemical performance, such as replacements of A-(by rare earth, transition metals, or others) and B-sites (by transition metals), combinations of different fabrication procedures, and surface treatments [177]. Table 5 illustrates recent elemental substitution efforts on MgNi-based alloy. By varying the ball milling time, Anik et al. found that 15 h was sufficient to obtain the amorphous/nanocrystalline state [178,179]; however, 25 h was required to incorporate all Ni into the main MgNi phase. Electrochemical discharge capacity is also influenced by the milling period. The same report demonstrated that capacity increased sharply with up to 15 h milling time, stabilized with up to 25 h milling time, and decreased with further increase in milling time. Based on the role of each constituent element, a series of systematic elemental substitutions was conducted [178][179][180][181]. Anik et al. reported an improvement in electrochemical performance with Mg 0.80 Ti 0.15 Al 0.05 Zr 0.05 Ni 0.95 alloy (420 mAh· g −1 and 90% capacity retaining rate at the 20th cycle) compared to the base MgNi alloy (495 mAh· g −1 and 35% capacity retaining rate at the 20th cycle) [178,181]. The pulverization mechanism of MgNi was investigated by in-situ monitoring of hydride-dehydride cycles using acoustic emission technique coupled with electrochemical measurement [182,183]. Unlike most cases where pulverization is caused by volume expansion and contraction with repeated hydrogenation, mechanically alloyed MgNi consists of porous agglomerates made up of many particles cold welded together, which are likely to be easily broken down by the mechanical action of hydrogen bubbles during hydrogen evolution. Hydrogen combustion synthesis was employed as another MgNi fabrication method; and with the help of subsequent ball milling and surface protection from NAFION coating, a 400% increase in capacity could be achieved [184].  [180] Mg 2 Ni is another Mg-Ni system that has been extensively studied for its potential applications in hydrogen storage and NiMH battery. With the higher content of hydride-former in the Mg 2 Ni formulation, it is capable of storing a greater amount of hydrogen compared to MgNi; however, Mg 2 Ni's discharge kinetics and corrosion resistance are worse. Nanocrystalline/amorphous structure was shown to improve hydriding/dehydriding kinetics [189]; such structure can be achieved by mechanical alloying and melt-spin fabrication techniques or elemental substitution. Although the exact effect of melt-spin cooling rates depends on the type of elemental substitution, a higher cooling rate was frequently found to be beneficial in enhancing electrochemical properties and cycle stability [190][191][192][193][194][195][196][197]. Furthermore, the length of milling time plays an important role in varying the performances of mechanically alloyed materials [198,199]. Melt-spin in magnetic field has been reported to greatly improve both capacity and stability [200]. Zhu et al. provided a comparison among the influences of various processing methods on electrochemical characteristics [201]. Table 6 summarizes the recent elemental substitution efforts on the Mg 2 Ni system. It should be noted that besides the Mg-Ni MH systems, other Mg-based alloys, such as MgTi [202] and MgAl [203] systems doped with Ni, were also investigated for their electrochemical properties.  [190,191,194,195] Mn MS Mg 2 Ni 1−x Mn x 0 to 0.4 up up then down up Promotes amorphous phase [192,194,195] Cu MS Mg 2 Ni 1−x Cu x 0 to 0.4 up up then down up - [193][194][195]

Laves Phase-Related BCC Solid Solutions
"Laves phase-related body-centered-cubic (BCC) solid solution" [208] is an interesting MH alloy with a general formula of AB x , where A is from Group 4A (mainly Ti), B is from Group 5A, 6A, and 7A (mainly V), and x is between 1 and 6. It has a unique two-phase microstructure composed of a BCC phase and a Laves phase (mostly C14). Microstructure evolution as a function of C14/BCC ratio is constructed based on literature review and presented in Figure 5. During solidification, the high V BCC phase solidifies first and form a 3D framework and the rest of the liquid turns into C14 phase, which is also a complementary 3D framework, and such phenomena produces the following evolution in microstructure with the increase in C14/BCC ratio: C14 starts to appear at the grain boundary ( Figure 5a) → sections of C14 phase start to connect and form a 3D network (Figure 5b) → BCC phase forms 3D framework and the cross-section is composed of isolated islands embedded in C14 matrix (Figure 5c) → BCC phase forms fish-bone type of inclusions (Figure 5d). The high density of phase boundaries contributes to the hydrogen storage properties in two ways: the promotion of a synergetic effect between BCC (high hydrogen storage capability) and C14 phases (better absorption kinetics and easy formation due to its brittleness), and the formation of a coherent and catalytic interface in between.  [209], Gué guen et al. showed that only a small amount of C14 is needed as a catalyst for the gaseous phase storage [210]. In order to study the correlation between the ratio and electrochemical properties of the alloy, we fabricated a series of Laves/BCC alloys with the Laves phase abundance from 10.5 to 91.6 wt %; their bulk diffusion coefficients and crystallite sizes are plotted in Figure 6. When one phase dominates, the crystallite size is smaller and the diffusion coefficient is larger. Therefore, a small but sufficient amount of secondary phase is preferable in the multi-phase MH alloy system. In addition, a BCC/C14 multi-phase alloy, reported by Wang and Ning, shows an electrochemical capacity of 450 mAh· g −1 , but degradation through cycling is severe [211]. The BCC/C14 alloy was remelted with 10 wt % LaNi 3 [212] and up to 10 wt % LaNi 5 [213,214] by arc melting, and the activation behavior, HRD, low-temperature performance, and cycle life are all improved due to the synergetic effect between main and secondary phases. BCC/Ti-Ni-based multi-phase systems were also fabricated previously [117,215]. The existence of the TiNi secondary phase enhances HRD and cycle life and shows the highest capacity of 470 mAh· g −1 [215]. Recently, an anneal-and-quench method [216] has been implemented to prepare BCC/C14 alloys. Furthermore, quasicrystal-included alloys have also been investigated by Liu et al. [217][218][219][220][221], and capacity, HRD, and cycle stability at the 30th cycle up to 422 mAh· g −1 , 88%, and 81%, respectively, were reported. Figure 6. Plots of diffusion constant and crystallite size of the C14 phase determined by FWHM from XRD analysis as functions of C14 phase abundance for a series of Laves phase-related BCC solid solution alloys. The bulk transport is enhanced when the grain size is small and much of the boundary interface is available to contribute to the synergetic effect.

Zr-Ni-Based Alloys
Since the AB 2 intermetallic compound does not exist on the Zr-Ni binary phase diagram, one or more secondary non-Laves Zr x Ni y phases composed of neighboring phases of AB 2 are often observed when fabricating a Zr-Ni-based MH alloy with an AB 2 composition. Between the working main phases and the catalytic secondary phases, a synergetic effect arises and contributes positively to the overall electrochemical performance. Although these minor phases do not have an appropriate metal-hydrogen bond strength when compared to the AB 2 stoichiometry, it is essential to evaluate various Zr x Ni y alloys in order to gain better understanding of their roles in AB 2 MH alloys and potentially develop alternative rare earth-free MH alloys for NiMH battery through composition modification. A systematic examination of Zr 8 Ni 21 , Zr 7 Ni 10 , Zr 9 Ni 11 , and ZrNi alloys correlating composition, structure, gaseous phase hydrogen storage, and electrochemical properties was recently reported by Nei et al. [77]. Annealing is detrimental to the electrochemical capacity of Zr x Ni y alloys due to the elimination of minor phases, similar to the annealing effect on AB 2 alloy. Furthermore, the electrochemical capacity maximizes at Zr 7 Ni 10 , which demonstrates that both hydrogen desorption/discharge rate and theoretical maximum hydrogen storage determined by the Zr/Ni ratio influence the electrochemical discharge capacity. Among all, Zr 8 Ni 21 shows the highest HRD and easiest activation. Zr 7 Ni 10 and Zr 8 Ni 21 were selected for further composition modification to improve their electrochemical properties [76,[222][223][224][225][226], and the results are summarized in Table 7. One of the drawbacks of AB 2 alloy compared to AB 5 alloy is its lower HRD. For the purpose of increasing HRD in the Zr-based alloys, alloys with higher B/A ratios, such as ZrNi 5 [26] and Zr 2 Ni 7 [25], become more attractive.
Performances of V-modified ZrNi 5 and Zr 2 Ni 7 are summarized in Table 7. Although their capacities are too low for practical application (≤177 mAh· g −1 ), their bulk hydrogen diffusion properties are superior to those of existing AB 5 , AB 2 , and A 2 B 7 alloys. With further modification, electrochemical performances of ZrNi 5 and Zr 2 Ni 7 alloys are expected to improve.
Researchers have ventured out of well-understood MH alloy systems in order to develop a novel material that fulfills all requirements of electrochemical applications. R 6 T 23 systems (R = Gd, Ho, T = Mn, Fe) were recently studied. He et al. performed B-site substitution using Co on Ho 6 Fe 23 and found that the alloys have Th 6 Mn 23 -type structure [234]. As the level of Co increases, activation, capacity (443 mAh· g −1 at 150 mA· g −1 ), cycle stability, and HRD all improve. Electrochemical results were also reported on the Gd-Co-Mn system, and a promising capacity of 377 mAh· g −1 at 150 mA· g −1 was obtained [235]. The HRD of this type of alloy, however, is relatively low and in the range of 50% to 60%.

Conclusions
Hydrogen storage alloys for electrochemical application have been extensively studied for many years. We have presented a review of recent research activities on metal hydride alloys for nickel metal hydride battery and also provided an overview of the use of metal hydrides in other electrochemical applications. AB 5 and AB 2 alloys are very well established systems. In order to potentially dominate the future electric vehicle and stationary applications, self-discharge, low-temperature performance, and cycle stability become more important to study, and the trend of recent research reflects the efforts on improving the aforementioned properties. Superlattice A 2 B 7 -type alloy, which possesses the advantages of both AB 5 and AB 2 and low self-discharge capability, is likely to be the next generation of metal hydride alloy used as the negative electrode material in nickel metal hydride batteries and has attracted much attention. Although the Ti-Ni alloy system is difficult to process and has poorer high-rate performance, its much lower raw material cost makes the system one that merits further studies for improvement. The Mg-Ni alloy system holds great promise in achieving very high capacity, and recent research efforts have concentrated on improving its kinetics and cycle capability for the purpose of practical implementation. Laves phase-related BCC solid solution has high capacity; enhancing its stability is currently the most essential topic. The incorporation of quasicrystals by various fabrication methods remains an interesting subject. Zr-Ni alloy systems were systematically investigated in the last few years. Although their performance might not be satisfactory for electrochemical applications at the present time, further elemental modifications or use as a composite modifier can assist in realizing their potential.