- Email: [email protected]
Microstructure and hydrogen storage properties of non-stoichiometric Zr–Ti–V Laves phase alloys
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/he
Microstructure and hydrogen storage properties of non-stoichiometric ZreTieV Laves phase alloys Y.L. Zhang, J.S. Li, T.B. Zhang*, R. Hu, X.Y. Xue State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi’an 710072, PR China
article info
abstract
Article history:
The non-stoichiometric C15 Laves phase alloys namely Zr0.9Ti0.1Vx (x ¼ 1.7, 1.8, 1.9, 2.1, 2.2,
Received 22 July 2013
2.3) are designed and expected to investigate the role of defect and microstructure on
Received in revised form
hydrogenation kinetics of AB2 type Zr-based alloys. The alloys are prepared by non-
4 September 2013
consumable arc melting in argon atmosphere and annealed at 1273 K for 168 h to ensure
Accepted 10 September 2013
the homogeneity. The microstructure and phase constitute of these alloys are examined by
Available online 2 October 2013
SEM, TEM and XRD. The results indicate the homogenizing can reduce the minor phases aZr and abundant V solid solution originating from the non-equilibrium solidification of as-
Keywords:
cast alloys. Twin defects with {111}<011 > orientation relationship are observed, and the
Non-stoichiometry
role of defects on hydrogenation kinetics is discussed. Hydrogen absorption PCT charac-
Microstructure
teristics and hydrogenation kinetics of Zr0.9Ti0.1Vx at 673e823 K are investigated
Hydrogen storage
by the pressure reduction method using a Sievert apparatus. The results show the hypo-
Laves phase
stoichiometric alloys preserve faster hydrogenation kinetics than the hyper-stoichio
Hydrogenation kinetics
metric ones due to the decrease of dendritic V. The excess content of Zr3V3O phase decreases the hydrogenation kinetics and the stability of hydrides. In addition, the different rate controlled mechanisms during hydrogen absorption are analyzed. The effects of nonstoichiometry on the crystal structure and hydrogen storage properties of Zr0.9Ti0.1Vx Laves alloys are discussed. Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Intermetallic compounds are widely used in hydrogen storage and the purification of hydrogen isotopes. For the excellent hydrogen absorption ability, large solubility and high diffusivity for the adsorbed gases and ease of activation, AB2 type Zr-based intermetallic compounds with C14/C15 Laves structure possess potential applications in fields of storage and separation of hydrogen and its isotopes, and have received much attention during past decades [1e5]. ZrV2 alloy has been reported to reach the hydrogen absorption capacity 4.8 H/A at 1 atm hydrogen pressure without changing crystal structure
[6], and the value can be 5.3 H/A at 12 atm hydrogen pressure [7]. Meanwhile, ZrV2 alloy preserves an ultra low equilibrium hydrogen pressure at room temperature for the potential applications as non-evaporable getter materials, which can upgrade and sustain the vacuum inside the cavity of vacuumtype devices. However, the shortage in hydrogen absorption kinetics and the hydrogen desorption hysteresis makes it difficult to meet the requirements of applications in the nuclear industry [8]. Many studies have been carried out to improve the activation, discharge capacity, absorption/desorption kinetics, cycle life and decrease the stability of metal hydrides of
* Corresponding author. Tel.: þ86 29 88491764; fax: þ86 29 88460294. E-mail address: [email protected] (T.B. Zhang). 0360-3199/$ e see front matter Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2013.09.040
14676
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Zr-based AB2 alloys by adding some alloying elements partially substituting the element in A or B site [9e11]. The work of Iwakura et al. shows that the increase of rA/rB value or decrease of unit cell volume in ZrV2xyMnxNiy Laves phase alloys results in the decrease of the interaction between constituent metals and hydrogen and the increase of the hydrogen diffusion rate. Meanwhile, the absolute value of enthalpy change reduces which indicates the reduction of hydrides stability and improvement of hydrogen desorption [12]. Doping Ti for Zr in (ZreTi)(NieMneCreV) AB2 alloys, the smaller Ti content or the larger unit cell volume, the larger reversible absorption capacity and the lower equilibrium plateau pressure can be obtained [13]. The AB2 type ZreTieV series compounds, including the Zr1xTixV2 (x ¼ 0e0.3) and the ZrTi0.2V1.8, show improved activation characteristics, desorption hysteresis and hydrogenation kinetics [14,15]. Proper Ti substitution for Zr is an efficient way to improve the hydrogenation kinetics comparing with the primary ZrV2 due to the introduction of twin defects and the V BCC solid solutions. Twin defects can be preferential diffusion paths for hydrogen and the elastic stresses generated by the twins favor the nucleation of the hydrides [16]. An autocatalytic mechanism caused by V BCC solid solutions results in the acceleration of hydrogenation kinetics [15]. But Ti substitution for Zr decreases hydrogen absorption capacity. Summarize the above mentioned, the substitutions usually induce a change of the unit cell volume, phase compositions, chemical affinity to hydrogen or a multi-phase synergy for absorption, making alloys achieve better hydrogen storage properties we need. Non-stoichiometric alloys are selected to obtain higher storage capacities, lower plateau pressures, improved cycle life and easier activation [17]. The electrochemical properties are improved and hydride stability is reduced by a small substitution of Ti at Zr site in Zr-based AB2 Laves alloys for NieMH rechargeable batteries [18]. The work of Kandavel et al. about hydrogenation properties of non-stoichiometric Zr-based AB2 alloys gives the evidence that the hydrogenation kinetics and the amount of hydrogen absorption can be improved in over-stoichiometric alloys due to the increase in the atomic proportions of hydrogen absorbing elements, the plateau pressure increases for the contraction of unit cell volume [19e21]. From the above mentioned references summary, the structure usually does not change with the introduction of non-stoichiometry, but there is a small change in lattice parameters and considerable changes in their hydrogen storage properties such as plateau pressure, plateau slope and kinetics of sorption and electrochemical discharge capacity. It’s evident that non-stoichiometry plays an important role on modification of hydrogen storage materials. Up to now, the poor kinetics is still obstacle for application of Zr-based Laves phase alloys. The methods to improve hydrogenation kinetics in hydrogen storage alloys are basically non-stoichiometry, grain refinement and alloying, owing to the introduction of lattice distortion, microdefects, larger specific surface area, the synergy of multiphase structure and better affinity to hydrogen [12,14,21]. According to this, non-stoichiometry is a promising way to further improve hydrogenation kinetics on the base of
alloying. However, the mechanism about effect of nonstoichiometry on hydrogenation properties of pseudobinary ZreTieV Laves alloys is still uncertain. It is necessary to explore the interrelation between microstructure/ phase constitute and hydrogenation kinetics/thermodynamic properties of non-stoichiometric Zr-based Laves alloys, which is beneficial for their practical application in fields of rapid storage and separation of hydrogen and its isotopes. Considering all above, the Zr0.9Ti0.1Vx alloys are designed where the Zr in A site is substituted by a small amount of Ti and the V in B site is non-stoichiometry based on our previous work about the role of defect structure on hydrogenation properties of Zr0.9Ti0.1V2 [15]. In the present work, the effects of phase content, microstructure and cell volume on hydrogenation kinetics and thermodynamic properties of nonstoichiometric Zr0.9Ti0.1Vx alloys are investigated. The defects in these alloys are observed, and the role of defects on hydrogenation kinetics is discussed. Detailed studies of hydrogenation kinetics are carried out to calculate reaction rate constant. Meanwhile, the mechanism about improvement of hydrogenation kinetics is further clarified. The possible hydrogenation mechanisms of non-stoichiometric Zr0.9Ti0.1Vx alloys are proposed based on the aforementioned experiments. The hydrogen storage properties including PCT characteristic and thermodynamic parameters from 673 to 823 K are investigated.
2.
Experimental procedures
Alloys with the compositions of Zr0.9Ti0.1Vx (x ¼ 1.7, 1.8, 1.9, 2.1, 2.2, 2.3) were prepared by non-consumable arc melting under argon atmosphere in a water-cooled copper crucible using 99.8wt% titanium, 99.4wt% zirconium and 99.5wt% vanadium. The ingots were turned over and re-melted 3 times to ensure the homogeneity, then annealed at 1273 K for 168 h in quartz tube under vacuum of 6 103 Pa followed by furnace cooling to room temperature. The phase compositions of the alloys were determined by a DX-2700 X-ray diffractometer using Cu Ka radiation. With a step scanning mode (step width 0.03 , counting time 2s), the diffraction data were collected at room temperature between 20 and 80 (2q). The microstructure of each sample was investigated by a scanning electron microscope (SEM) using secondary electron imaging and a transmission electron microscope (FEI Tecnai G2 F30, America) operating at 200 kV. The hydrogen storage characteristics including the P-t and PCT curves for each sample were obtained by using a Sievert type apparatus in temperature range of 673 Ke823 K. The limits of the three capacitance manometers range from 100 kPa to 104 Pa. Prior to the absorption runs, each sample was activated by heating the sample chamber to 723 K and pumping the chamber for 40e60 min to the set vacuum in order to create a clean surface free from the effect of surface contamination. The hydrogen absorption content was determined by calculating the pressure variation during the reaction while maintaining constant temperature of the reaction chamber.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
14677
3.
Results and discussion
Table 1 e The phase content and cell volume of each phase of the Zr0.9Ti0.1Vx alloys.
3.1.
Phase constitute and microstructure
Alloys
In order to characterize the phase compositions of the alloys, XRD analysis was performed. The XRD patterns of the as-cast and annealed Zr0.9Ti0.1Vx alloys are shown in Fig. 1. The phase content and cell volume of each phase of the Zr0.9Ti0.1Vx alloys are summarized in Table 1. The major diffraction peaks of the as-cast alloys shown in Fig. 1(a) can be identified to C15 ZrV2 Laves phase which crystallizes in the cubic (MgCu2 type) structure with space group Fd-3m. Besides the Laves phase, minor a-Zr and V-based BCC phases are also present. The structure does not change significantly with the introduction of non-stoichiometry. Meanwhile, a small amount of Zr3V3O phase is detected in Zr0.9Ti0.1V2.1 due to the oxidized raw materials or slight oxidation during the melting process, which has been observed in literature [11,22,23]. Existence of
Zr0.9Ti0.1V1.7
Zr0.9Ti0.1V1.8
Zr0.9Ti0.1V1.9
Zr0.9Ti0.1V2.1
Zr0.9Ti0.1V2.2
Zr0.9Ti0.1V2.3
Fig. 1 e XRD patterns of the Zr0.9Ti0.1Vx (x [ 1.7, 1.8, 1.9, 2.1, 2.2, 2.3) alloys, (a) as-cast, (b) annealed.
Phase content (wt. %)
Cell volume (103 nm3)
As-cast
Annealed
Annealed
C15(49) V(36) a-Zr(15) C15(45) V(38) a-Zr(17) C15(43) V(39) a-Zr(18) C15(38) V(41) a-Zr(19) Zr3V3O(2) C15(36) V(44) a-Zr(20) C15(36) V(43) a-Zr(21)
C15(54) Zr3V3O(43) V(3) C15(52) Zr3V3O(44) V(4) C15(55) Zr3V3O(40) V(5) C15(39) Zr3V3O(50) V(11)
416.11 1811.05 28.22 417.14 1818.92 28.42 414.13 1801.91 28.17 417.24 1810.23 28.30
C15(26) Zr3V3O(62) V(12) C15(40) Zr3V3O(46) V(14)
415.31 1808.89 28.27 414.74 1805.11 28.21
minor phases can be ascribed to the non-equilibrium solidification process of the arc-melted ingots and the incomplete peritectic reaction in the pseudo-binary system [24]. From Table 1, it can be noticed that the content of V-BCC solid solution phase slightly increases with increasing V content of the as-cast alloys, the C15eZrV2 phase content decreases accordingly and the a-Zr phase content hardly changed. The homogenizing treatment can eliminate a-Zr phase and generate Zr3V3O phase (see Fig. 1(b)). It can be seen from Table 1 that the C15 phase content of each annealed alloy increases comparing with the corresponding as-cast one except Zr0.9Ti0.1V2.2. The a-Zr phase and V-based BCC phase generated during the rapid solidification process have been almost eliminated by homogenizing in hypo-stoichiometric alloys. In the case of hyper-stoichiometry, the a-Zr phase can be eliminated but the V-based BCC phase can just be cut down. The content of Zr3V3O phase in annealed Zr0.9Ti0.1V2.2 alloy is higher than the others. The phase constitute has a significant influence on hydrogenation kinetics and thermodynamic properties which will be discussed later. The cell volume of each phase of the annealed alloys listed in Table 1 shows the regularity that cell volume of each phase decreases gradually with increasing V content in hyper-stoichiometric alloys. For the case of hypo-stoichiometry, the cell volume of each phase increases first then decreases to the least with increasing V content. The effects of cell volume on hydrogenation thermodynamic properties will be illustrated later. Observe Fig. 1 carefully, it can be seen that some peaks in the patterns are asymmetric. According to our analysis, asymmetric Bragg peaks in XRD patterns are probably due to axial divergence and non-ideal specimen geometry. The instrumental factor, especially axial divergence, is the main reason for the asymmetric Bragg peaks. The combined asymmetry effects result in the low angle sides of Bragg peaks being considerably broader than their high angle sides.
14678
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Fig. 2 gives the SEM images of as-cast Zr0.9Ti0.1Vx alloys with typical peritectic microstructure. Two-phase microstructure consisting of the dark dendritic V-based solid solution and ZrV2 Laves phase matrix are formed under the rapid solidification, the big thick V dendritic are quite prevalent. According to ZreV binary phase diagram [25], Zr and V have a classical peritectic reaction at 1300 C. During nonequilibrium solidification, the dendritic primary V-based solid solution precipitates firstly, then surrounded by ZrV2 Laves phase. The peritectic reaction from V-based solid solution and liquid phase to ZrV2 phase is suppressed, the residual liquid phase crystallizes to ZrV2 and b-Zr phase, then b-Zr phase transforms to a-Zr phase. The micro-segregation in microstructure of as-cast alloys is evident owing to the nonuniform composition in grain size during solidification. SEM images of annealed Zr0.9Ti0.1Vx alloys are shown in Fig. 3. After annealing at 1273 K for 168 h, the dendritic structures break up and the matrix regions increase during the annealing process. As can be seen from Fig. 3 (a) and (b), the peritectic microstructure observed in corresponding as-cast alloys have been almost eliminated, only a small quantity of V-based solid solution dispersed in the matrix phase. The metastable phase transition completes and the multistable configuration is obtained by annealing for hypo-stoichiometric alloys as has been observed in Zr0.9Ti0.1V2 alloy in Ref. [15]. But it is different in hyper-stoichiometric alloys, the dendritic V solid solution are still large and abundant in Fig. 3 (c) and (d), which results from the excess content of V element in hyper-stoichiometric alloys, these results accord well with the XRD analysis.
The TEM microstructures and corresponding selected area diffraction patterns (SADPs) of the annealed Zr0.9Ti0.1V1.7 and Zr0.9Ti0.1V2.2 alloys are shown in Fig. 4. The SADP1 and SADP2 indicates the existence of V-BCC phase and the C15eZrV2 phase respectively, the SADP3 indicates that the observed micro twins in C15eZrV2 phase are of {111}<011 > orientation relationship. There are large quantities of finely spaced line traces in C15eZrV2 matrix of both the hypo- and hyperstoichiometric alloys, which are identified as annealing twins and/or stacking faults. According to report of Zhu et al., the possible defect mechanisms in binary Laves phases are constitutional vacancy or anti-site substitution [26]. In V-rich alloys, the exhibition of twins or stacking faults could be explained as ZrCr2 Laves alloys [27]. The excess V atoms occupy the Zr sublattice sites which results in larger free volumes between two neighboring close packed {111} atomic planes, and thereby twins and stacking faults can be more easily introduced [28]. Such a result is also shown in the Vpoor compositions but the defect mechanisms may be more complicated. The defects can improve the hydrogen absorption kinetics by providing a preferential diffusion path for hydrogen and helping the nucleation of the hydride [16].
3.2.
Hydrogen absorption kinetics
The initial absorption capacities at room temperature of the experimental alloys are listed in Table 2. The hydrogen absorption capacity of Zr0.9Ti0.1V1.7 reaches 2.33 wt. % and the absorption capacity of V-poor alloys are slightly larger than
Fig. 2 e SEM secondary electron images of as-cast Zr0.9Ti0.1Vx alloys (a) x [ 1.7, (b) x [ 1.9, (c) x [ 2.1, (d) x [ 2.3.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
14679
Fig. 3 e SEM secondary electron images of annealed Zr0.9Ti0.1Vx alloys (a) x [ 1.7, (b) x [ 1.9, (c) x [ 2.1, (d) x [ 2.3.
that of the V-rich alloys. It indicates that the hydrogen storage capacity can be improved with the variety of the composition stoichiometry towards hypo-stoichiometry. The hydrogen absorption kinetics curves of annealed Zr0.9Ti0.1Vx alloys at 773 K are shown in Fig. 5. It can be seen that the hydrogen absorption concentration of the experimental alloys increases quickly with time and reaches kinetic stabilization within 50 s, the value for Zr0.9Ti0.1V1.7 is 15 s, which illustrates these alloys have a strong ability for rapid absorption hydrogen to saturation. The hydrogen absorption content at 773 K reaches 1.10 wt. % for Zr0.9Ti0.1V1.7 and Zr0.9Ti0.1V1.9, and is 0.98 wt. % for Zr0.9Ti0.1V2.3. Several mathematical models have been proposed to evaluate the hydrogenation kinetics mechanism of Zr-based AB2 alloys, such as Jander model or Chou model [29]. The experimental hydrogenation kinetics data can well fitted with the calculated results of theoretical models [30], but Jander model has its own limitation as just being applicable for small conversion rates and stable diffusion. Moreover, these methods are suitable for the case where hydrogen diffusion is the rate-controlling step both in the solid solution phase (a phase) and the hydride (b phase). In this work, the hydrogenation time of the experimental alloys are considerably short comparing with other alloys, especially for the a phase region. Understanding the diffusion theory and H diffusion rate-controlling mechanisms in other alloys [19,31], the diffusion can not be the rate-controlling step in such a short time. The possible hydrogen absorption reaction mechanism is proposed by investigating the observed
absorption rate based on Hirooka’s kinetic theory [32], which is developed on the assumption that the reaction rate to reach equilibrium is proportional to the deviation from the equilibrium. Fig. 6 displays the typical plot of ln[( pt - peq)/( p0 - peq)] vs. the reaction time t of Zr0.9Ti0.1Vx alloys at 773 K. It is evident that all the curves fit into two different linear parts, one is rapid absorbing stage 1 and the other is stable absorbing stage 2, the slope of each stage stands for relevant reaction rate constant ka1 or ka2. These values are summarized in Table 2. The ka1 values of the V-poor alloys are larger than those of the V-rich alloys, which indicates the hypo-stoichiometric alloys preserve faster hydrogenation kinetics. This could be explained by the break up and reduction of dendritic V solid solution and increase of ZrV2 matrix that enhance the flow of hydrogen in the hypo-stoichiometric alloys. Additionally, the ka1 values of the Zr0.9Ti0.1Vx alloys (except Zr0.9Ti0.1V2.2) are bigger comparing with those of the Zr(V1xFex)2 alloys at 773 K [11]. Base on the comparison of phase compositions in these alloys, the possible reason for this is the introduction of V BCC solid solutions in these alloys. It is reported the synergy of V BCC solid solutions leads to an autocatalytic mechanism to improve hydrogenation kinetics due to the increasing active surface area [15,33]. The ka1 value 0.228 for Zr0.9Ti0.1V1.7 is significantly larger than other experimental alloys and ZrTi0.2V1.8 alloy [14]. The better kinetics of Zr0.9Ti0.1V1.7 reveals the less content of V BCC solid solutions, the more positive synergy effect can be obtained. Meanwhile, the twin defects observed in Zr0.9Ti0.1V1.7 would accelerate the hydrogenation
14680
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Fig. 4 e TEM microstructures and corresponding selected area diffraction patterns (SADPs) of the annealed Zr0.9Ti0.1V1.7 (a)(d), (b) SADP1, (c) SADP2, (e) SADP3 and Zr0.9Ti0.1V2.2 (f)(g).
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
14681
Table 2 e The hydrogen absorption capacity at room temperature and hydriding reaction rate constant at 773 K of the Zr0.9Ti0.1Vx alloys. Zr0.9Ti0.1Vx x ¼ 1.7 x ¼ 1.8 x ¼ 1.9 x ¼ 2.1 x ¼ 2.2 x ¼ 2.3 wt. % ka1 ka2
2.33 0.228 0.015
2.26 0.170 0.014
2.27 0.164 0.023
2.22 0.147 0.032
2.25 0.107 0.019
2.20 0.140 0.027
kinetics. It can be concluded that the V BCC solid solutions should be restricted to a small content to achieve accelerated hydrogenation kinetics. The ka1 value 0.107 for Zr0.9Ti0.1V2.2 reveals the alloy preserves the slowest hydrogenation kinetics among the experimental alloys, which can be attributed to the large content of Zr3V3O phase hinder the rapid absorption of hydrogen. These curves involve a phase transformation and demonstrate the existence of two phase regions namely (a þ b) and b phase, which are observed by the different slopes of the two segments. The a-phase region could not be seen due to the fast hydriding reaction at relatively high pressure. The slope variation in the two stages intimates the rate-controlling mechanism changes during hydrogenation course. As proposed by Srinivas et al., the rate-controlling mechanism in a, (a þ b) and b phase regions can be assumed to the surface process, the interface process and the diffusion of hydrogen atoms through the growing hydride layer, respectively [34]. In stage 1, hydrogen atoms firstly penetrate through grain boundaries and dissolve in the bulk. Then the hydride layer can be formed at the grain circumference, the initial reaction rate is controlled by hydrogen flow through grain boundaries. As the hydrides grow, the volume of the grain expands gradually with increase of the hydrogen concentration. As a result, the reacting grains are pressed together and then narrows the transfer channel which ultimately reduces the hydrogen flow into the deeper parts of the sample [35]. When the b phase hydrides form completely, the reaction rate of stage 2 is controlled by hydrogen diffusion through the hydride phase.
Fig. 6 e ln[( pt e peq)/( p0 e peq)] vs. t plots of annealed Zr0.9Ti0.1Vx (x [ 1.7, 1.8, 1.9, 2.1, 2.2, 2.3) alloys at 773 K.
3.3.
Thermodynamic characteristics
The pressure-concentration-temperature (PCT) characteristics of annealed Zr0.9Ti0.1Vx (x ¼ 1.7, 1.8, 1.9, 2.2) alloys from 673 to 823 K are shown in Fig. 7. The absorption curves clearly indicate the presence of a, a þ b and b phase regions according to slope change at various temperatures as observed in some Zr-based AB2 alloys [20,21]. During the initial stage of absorption, physisorption and chemisorptions take place and only a solid solution phase is formed in the a-phase region, which is very short in the PCT curves of the experimental alloys. As the absorption continuing, the hydride phase nuclei starts growing and the plateau pressure is obtained owing to the formation of b hydride phase. When the b phase hydrides form completely, the hydrogen atoms get redistributing in the hydride phase, the hydrogen absorbing capacity reaches saturation. It is difficult to determine the plateau pressure from the PCT curves as the curves exhibit the characteristic of sloping plateaus. The presence of a sloping plateau is due to the formation of different hydrides. The slope of plateaus increases while the width decreases as the composition of the alloys changes from V-poor to V-rich. All the hydrogen desorption curves at 823 K of the experimental alloys are close to the absorption curves at that temperature, that is to say no significant hysteresis effect exists between absorption and desorption. From the PCT characteristics of Zr0.9Ti0.1Vx (x ¼ 1.7, 1.8, 1.9, 2.2) alloys, the relative thermodynamic properties of dissolved hydrogen could be obtained at a particular hydrogen concentration by Van’t Hoff equation [36]: InPeq ¼ DHQ =RT DSQ =R
Fig. 5 e Hydrogen absorption kinetics curves of annealed Zr0.9Ti0.1Vx (x [ 1.7, 1.8, 1.9, 2.1, 2.2, 2.3) alloys at 773 K.
(1)
The plots for lnP vs. 1000/T of the experimental alloys are displayed in Fig. 8, where P is the equilibrium pressure obtained from the absorption isotherms at several specific hydrogen concentrations in the a þ b two-phase region. The experimental data of each sample can be interpolated linearly
14682
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Fig. 7 e Hydrogen absorption/desorption PCT curves of the Zr0.9Ti0.1Vx alloys from 673 to 823 K (a) x [ 1.7, (b) x [ 1.8, (c) x [ 1.9, (d) x [ 2 [15], (e) x [ 2.2.
well and exhibit good relativity, which illustrates the relatively stable hydrides are generated in the range of H/A content. The enthalpy change (OHQ) and entropy change (OSQ) for a particular hydrogen concentration can be calculated from the slope and intercept of the least square fit for lnPeq vs. 1000/T plots, respectively. The thermodynamic parameters of the experimental alloys are listed in Table 3. It is easy to find both the -OHQ and -OSQ values are higher than that of some other Zr-based Laves phase alloys [14,37,38], the higher -OHQ values indicate the Zr0.9Ti0.1Vx alloys forming more stable hydrides. It is obvious that the -OHQ and -OSQ values of each experimental alloy increase gradually with the increasing hydrogen content. This is understandable in view of the overall process that more hydrogen transfers from the gaseous state into the solid state. The -OHQ values for Zr0.9Ti0.1V2.2 are considerable smaller than those for the Vpoor alloys which can be attribute to the low affinity of Zr3V3O phase to hydrogen, large content of Zr3V3O phase decreases the stability of hydrides. This will be beneficial to decrease the hydrogen desorption temperature. According to the works of Jain and Lundin et al. [39,40], the interstitial size increases with the expansion of ZrV2 cell volume in the alloys. As a result, the binding energy between metal and hydrogen atoms reduces with the increasing metal-hydrogen distance, and correspondingly leads to the higher -OHQ/-OSQ value and lower plateau pressure. But in the experimental hypostoichiometric alloys, the regularity between cell volume and -OHQ value is contrary to the aforementioned. The cell
volume of each phase increases first then decreases to the least with increasing V content, and the -OHQ value decreases first then increases to the maximum accordingly. The reason for this should be the multi-phase structure in these alloys, the three phases ZrV2, Zr3V3O and V BCC solid solutions all can form corresponding hydrides. The -OHQ value is the combined result of three hydrides, so the factors which affect -OHQ value are more complex. It suggests that the thermodynamic properties are more affected by the relative content of each phase and its corresponding hydride than the cell volume. The equilibrium hydrogen pressures at room temperature can be calculated by Van’t Hoff equation, these extrapolated values of the Zr0.9Ti0.1Vx alloys are all below 107Pa, which indicates these alloys have strong ability to remove traces of residual hydrogen. Even this method has its indirect measurement deviation and indeterminacy, the extrapolated equilibrium pressures at room temperature are helpful to give the data which can not be obtained directly from the experiment owing to the extreme low pressure and they can be regard as references.
4.
Conclusions
In summary, the microstructure and hydrogen absorption kinetics and thermodynamic properties of non-stoichiometric AB2 Laves alloys Zr0.9Ti0.1Vx have been investigated in this work. The multiphase structure consisting of ZrV2 Laves
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
14683
Fig. 8 e Van’t Hoff plots for the Zr0.9Ti0.1Vx alloys (a) x [ 1.7, (b) x [ 1.8, (c) x [ 1.9, (d) x [ 2.2.
phase, a-Zr and V-BCC solid solution phase are analyzed. The homogenizing treatment can eliminate the minor a-Zr phase and reduce the V BCC phase in as-cast alloys. Twin defects with {111}<011 > orientation relationship are observed, and they can accelerate hydrogenation kinetics. The hypo-
Table 3 e Thermodynamic parameters of the Zr0.9Ti0.1Vx alloys. Alloys Zr0.9Ti0.1V1.7
Zr0.9Ti0.1V1.8
Zr0.9Ti0.1V1.9
Zr0.9Ti0.1V2.2
Content of H/A
OHQ (KJ mol1)
OSQ (J mol1 K1)
0.3 0.45 0.6 0.75 0.9 0.3 0.45 0.6 0.75 0.9 0.3 0.45 0.6 0.75 0.9 0.3 0.45 0.6 0.75
80.53 90.39 99.38 105.75 107.99 81.57 84.37 89.35 91.50 94.47 85.60 96.82 108.01 115.97 122.57 60.45 63.06 69.88 78.86
115.86 133.69 150.22 163.24 170.98 118.03 128.13 140.72 149.29 158.89 122.59 141.91 160.67 175.27 187.67 75.30 83.83 98.23 116.26
stoichiometric alloys perform faster hydrogenation kinetics than the hyper-stoichiometric ones due to the break up and reduction of dendritic V enhancing the diffusion of hydrogen. The synergy induced by a small content of V BCC solid solutions can improve hydrogenation kinetics. Two stages during hydrogen absorption process could be attributed to the different rate controlled mechanism including hydrogen flow through grain boundaries and hydrogen diffusion through the hydride phase. The excess content of Zr3V3O phase decreases the hydrogenation kinetics and the stability of hydrides.
Acknowledgments This work was financially supported by State Key Laboratory of Solidification Processing (NWPU) (70-QP-2010). The Program of Introducing Talents of Discipline in the Project of Advanced Materials and their Forming Technology (B08040) is also acknowledged. Prof Shi Liu and Mr. Liang-yin Xiong from the Institute of Metal Research (Chinese Academy of Sciences, Shenyang, China) are gratefully appreciated for their help in hydrogenation measurement and helpful discussions.
references
[1] Ruiz FC, Castro EB, Peretti HA, Visintin A. Study of the different ZrxNiy phases of Zr-based AB2 materials. Int J Hydrogen Energ 2010;35:9879e87.
14684
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
[2] Jain A, Jain RK, Agarwal S, Ganesan V, Lalla NP, Phase DM, et al. Synthesis, characterization and hydrogenation of ZrFe2xNix (x ¼ 0. 2, 0.4, 0.6, 0.8) alloys. Int J Hydrogen Energ 2007;32:3965e71. [3] Young K, Nei J, Ouchi T, Fetcenko MA. Phase abundances in AB(2) metal hydride alloys and their correlations to various properties. J Alloy Compd 2011;509:2277e84. [4] Jain A, Jain RK, Agarwal G, Jain IP. Crystal structure, hydrogen absorption and thermodynamics of Zr1xCoxFe2 alloys. J Alloy Compd 2007;438:106e9. [5] Bulyk II , Basaraba YB, Trostianchyn AM. Features of the HDDR process in ZrT2 (T ¼ Cr, Mn, Fe, Co) compounds. J Alloy Compd 2004;367:283e8. [6] Pebler A, Gulbransen EA. Thermochemical and structural aspects of the reaction of hydrogen and intermetalic compounds of zirconium. Electrochem Tech 1966;4:211e5. [7] Shaltiel D, Jacob I, Davidov D. Hydrogen absorption and desorption properties of AB2 laves-phase pseudobinary compounds. J Less-Common Met 1977;53:117e31. [8] Jurczyk M, Rajewski W, Wo´jcik G, Majchrzycki W. Metal hydride electrodes prepared by mechanical alloying of ZrV2type materials. J Alloy Compd 1999;285:250e4. [9] Xu Y-h, Wang G, Chen C-p, Wang Q-d, Wang X. The structure and electrode properties of non-stoichiometric A1.2B2 type C14 Laves alloy and the effect of surface modification. Int J Hydrogen Energ 2007;32:1050e8. [10] Kim JH, Lee H, Hwang KT, Han JS. Hydriding behavior in Zrbased AB(2) alloy by gas atomization process. Int J Hydrogen Energ 2009;34:9424e30. [11] Zhang TB, Wang XF, Hu R, Li JS, Yang XW, Xue XY, et al. Hydrogen absorption properties of Zr(V1xFex)2 intermetallic compounds. Int J Hydrogen Energ 2012;37:2328e35. [12] Iwakura C, Kasuga H, Kim I, Inoue H, Matsuoka M. Effect of alloy composition on electrochemical properties of the Zrbased Laves-phase hydrogen storage alloys. Electrochim Acta 1996;41:2691e4. [13] Klein B, Simon N, Klyamkine S, Latroche M, PercheronGuegan A. Improvement of the thermodynamical and electrochemical properties of multicomponent Laves phase hydrides by thermal annealing. J Alloy Compd 1998;280:284e9. [14] Yang XW, Zhang TB, Hu R, Li JS, Xue XY, Fu HZ. Microstructure and hydrogenation thermokinetics of ZrTi0.2V1.8 alloy. Int J Hydrogen Energ 2010;35:11981e5. [15] Yang XW, Li JS, Zhang TB, Hu R, Xue XY, Fu HZ. Role of defect structure on hydrogenation properties of Zr0.9Ti0.1V2 alloy. Int J Hydrogen Energ 2011;36:9318e23. [16] Michel N, Poulat S, Priester L, Dantzer P. Microstructural analysis of the thermodynamically controlled hydride phases grown in the ZrNi-H2 system. Mat Sci Eng A-struct 2004;384:224e31. [17] Young K, Ouchi T, Yang J, Fetcenko MA. Studies of offstoichiometric AB(2) metal hydride alloy: Part 1. Structural characteristics. Int J Hydrogen Energ 2011;36:11137e45. [18] Kim D-M, Jang K-J, Lee J-Y. A review on the development of AB2-type Zr-based Laves phase hydrogen storage alloys for NieMH rechargeable batteries in the Korea Advanced Institute of Science and Technology. J Alloy Compd 1999;293e295:583e92. [19] Kandavel M, Ramaprabhu S, Jung M, Wipf H. Solubility and diffusion of hydrogen in AB(2)-type Laves phase alloys. J Alloy Compd 2005;404:265e8.
[20] Kandavel M, Ramaprabhu S. Hydriding properties of Tisubstituted non-stoichiometric AB(2) alloys. J Alloy Compd 2004;381:140e50. [21] Kandavel M, Ramaprabhu S. The effect of non-stoichiometry on the hydrogen storage properties of Ti-substituted AB(2) alloys. J Phys-condens Mat 2003;15:7501e17. [22] Gondor G, Lexcellent C. Analysis of hydrogen storage in metal hydride tanks introducing an induced phase transformation. Int J Hydrogen Energ 2009;34:5716e25. [23] Huot J, Akiba E, Ishido Y. Crystal structure of multiphase alloys (Zr, Ti)(Mn, V)2. J Alloy Compd 1995;231:85e9. [24] Smith JF. Phase diagrams of binary vanadium alloys. Alloy phase diagrams. Metals Park, Ohio: ASM; 1992. [25] Klein B, Redeker A, Zu¨chner HZ. Electrochemical measurements of hydrogen diffusion in ZrV2 and ZrV0.5Ni1.5. Z Physik Chem 1993;181:95e101. [26] Zhu JH, Liu CT. Defect structures in ZrCo2 Laves phase. Acta Mater 2000;48:2339e47. [27] Kanazawa S, Kaneno Y, Inoue H, Kim WY, Takasugi T. Microstructures and defect structures in ZrCr2 Laves phase based intermetallic compounds. Intermetallics 2002;10:783e92. [28] Takasugi T, Yoshida M, Hanada S. Deformability improvement in C15 NbCr2 intermetallics by addition of ternary elements. Acta Mater 1996;44:669e74. [29] Chou K-C, Li Q, Lin Q, Jiang L-J, Xu K-D. Kinetics of absorption and desorption of hydrogen in alloy powder. Int J Hydrogen Energ 2005;30:301e9. [30] Cui X-Y, Li Q, Chou K-C, Chen S-L, Lin G-W, Xu K-D. A comparative study on the hydriding kinetics of Zr-based AB2 hydrogen storage alloys. Intermetallics 2008;16:662e7. [31] Vojtech D, Knotek V. Study of the diffusion kinetics and mechanism of electrochemical hydriding of Mg-Ni-Mm alloys. Int J Hydrogen Energ 2011;36:6689e97. [32] Hirooka Y, Miyake M, Sano T. A study of hydrogen absorption and desorption by titanium. J Nucl Mater 1981;96:227e32. [33] Latroche M. Structural and thermodynamic properties of metallic hydrides used for energy storage. J Phys Chem Solids 2004;65:517e22. [34] Srinivas G, Sankaranarayanan V, Ramaprabhu S. Hydrogen absorption and desorption properties of Ho1x MmxCo2 alloys. Int J Hydrogen Energ 2007;32:2480e7. [35] Bloch J. The temperature-dependent changes of the kinetics and morphology of hydride formation in zirconium. J Alloy Compd 1995;216:187e95. [36] Flanagan TB, Clewley JD. Hysteresis in metal hydrides. J LessCommon Met 1982;83:127e41. [37] Nakano H, Wakao S. Substitution effect of elements in Zrbased alloys with Laves phase for nickel-hydride battery. J Alloy Compd 1995;231:587e93. [38] Jain A, Agarwal S, Jain IP. Correlation between the milling time and hydrogen-storage properties of nanostructured ZrFeNi ternary alloy. J Alloy Compd 2009;480:325e8. [39] Jain A, Jain RK, Agarwal S, Jain IP. Thermodynamics and structural aspects of hydrogen absorption in Zr1xCrxFe2 alloys. Int J Hydrogen Energ 2007;32:2445e9. [40] Lundin CE, Lynch FE, Magee CB. A correlation between the interstitial hole sizes in intermetallic compounds and the thermodynamic properties of the hydrides formed from those compounds. J Less-Common Met 1977;56:19e37.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/he
Microstructure and hydrogen storage properties of non-stoichiometric ZreTieV Laves phase alloys Y.L. Zhang, J.S. Li, T.B. Zhang*, R. Hu, X.Y. Xue State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi’an 710072, PR China
article info
abstract
Article history:
The non-stoichiometric C15 Laves phase alloys namely Zr0.9Ti0.1Vx (x ¼ 1.7, 1.8, 1.9, 2.1, 2.2,
Received 22 July 2013
2.3) are designed and expected to investigate the role of defect and microstructure on
Received in revised form
hydrogenation kinetics of AB2 type Zr-based alloys. The alloys are prepared by non-
4 September 2013
consumable arc melting in argon atmosphere and annealed at 1273 K for 168 h to ensure
Accepted 10 September 2013
the homogeneity. The microstructure and phase constitute of these alloys are examined by
Available online 2 October 2013
SEM, TEM and XRD. The results indicate the homogenizing can reduce the minor phases aZr and abundant V solid solution originating from the non-equilibrium solidification of as-
Keywords:
cast alloys. Twin defects with {111}<011 > orientation relationship are observed, and the
Non-stoichiometry
role of defects on hydrogenation kinetics is discussed. Hydrogen absorption PCT charac-
Microstructure
teristics and hydrogenation kinetics of Zr0.9Ti0.1Vx at 673e823 K are investigated
Hydrogen storage
by the pressure reduction method using a Sievert apparatus. The results show the hypo-
Laves phase
stoichiometric alloys preserve faster hydrogenation kinetics than the hyper-stoichio
Hydrogenation kinetics
metric ones due to the decrease of dendritic V. The excess content of Zr3V3O phase decreases the hydrogenation kinetics and the stability of hydrides. In addition, the different rate controlled mechanisms during hydrogen absorption are analyzed. The effects of nonstoichiometry on the crystal structure and hydrogen storage properties of Zr0.9Ti0.1Vx Laves alloys are discussed. Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Intermetallic compounds are widely used in hydrogen storage and the purification of hydrogen isotopes. For the excellent hydrogen absorption ability, large solubility and high diffusivity for the adsorbed gases and ease of activation, AB2 type Zr-based intermetallic compounds with C14/C15 Laves structure possess potential applications in fields of storage and separation of hydrogen and its isotopes, and have received much attention during past decades [1e5]. ZrV2 alloy has been reported to reach the hydrogen absorption capacity 4.8 H/A at 1 atm hydrogen pressure without changing crystal structure
[6], and the value can be 5.3 H/A at 12 atm hydrogen pressure [7]. Meanwhile, ZrV2 alloy preserves an ultra low equilibrium hydrogen pressure at room temperature for the potential applications as non-evaporable getter materials, which can upgrade and sustain the vacuum inside the cavity of vacuumtype devices. However, the shortage in hydrogen absorption kinetics and the hydrogen desorption hysteresis makes it difficult to meet the requirements of applications in the nuclear industry [8]. Many studies have been carried out to improve the activation, discharge capacity, absorption/desorption kinetics, cycle life and decrease the stability of metal hydrides of
* Corresponding author. Tel.: þ86 29 88491764; fax: þ86 29 88460294. E-mail address: [email protected] (T.B. Zhang). 0360-3199/$ e see front matter Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2013.09.040
14676
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Zr-based AB2 alloys by adding some alloying elements partially substituting the element in A or B site [9e11]. The work of Iwakura et al. shows that the increase of rA/rB value or decrease of unit cell volume in ZrV2xyMnxNiy Laves phase alloys results in the decrease of the interaction between constituent metals and hydrogen and the increase of the hydrogen diffusion rate. Meanwhile, the absolute value of enthalpy change reduces which indicates the reduction of hydrides stability and improvement of hydrogen desorption [12]. Doping Ti for Zr in (ZreTi)(NieMneCreV) AB2 alloys, the smaller Ti content or the larger unit cell volume, the larger reversible absorption capacity and the lower equilibrium plateau pressure can be obtained [13]. The AB2 type ZreTieV series compounds, including the Zr1xTixV2 (x ¼ 0e0.3) and the ZrTi0.2V1.8, show improved activation characteristics, desorption hysteresis and hydrogenation kinetics [14,15]. Proper Ti substitution for Zr is an efficient way to improve the hydrogenation kinetics comparing with the primary ZrV2 due to the introduction of twin defects and the V BCC solid solutions. Twin defects can be preferential diffusion paths for hydrogen and the elastic stresses generated by the twins favor the nucleation of the hydrides [16]. An autocatalytic mechanism caused by V BCC solid solutions results in the acceleration of hydrogenation kinetics [15]. But Ti substitution for Zr decreases hydrogen absorption capacity. Summarize the above mentioned, the substitutions usually induce a change of the unit cell volume, phase compositions, chemical affinity to hydrogen or a multi-phase synergy for absorption, making alloys achieve better hydrogen storage properties we need. Non-stoichiometric alloys are selected to obtain higher storage capacities, lower plateau pressures, improved cycle life and easier activation [17]. The electrochemical properties are improved and hydride stability is reduced by a small substitution of Ti at Zr site in Zr-based AB2 Laves alloys for NieMH rechargeable batteries [18]. The work of Kandavel et al. about hydrogenation properties of non-stoichiometric Zr-based AB2 alloys gives the evidence that the hydrogenation kinetics and the amount of hydrogen absorption can be improved in over-stoichiometric alloys due to the increase in the atomic proportions of hydrogen absorbing elements, the plateau pressure increases for the contraction of unit cell volume [19e21]. From the above mentioned references summary, the structure usually does not change with the introduction of non-stoichiometry, but there is a small change in lattice parameters and considerable changes in their hydrogen storage properties such as plateau pressure, plateau slope and kinetics of sorption and electrochemical discharge capacity. It’s evident that non-stoichiometry plays an important role on modification of hydrogen storage materials. Up to now, the poor kinetics is still obstacle for application of Zr-based Laves phase alloys. The methods to improve hydrogenation kinetics in hydrogen storage alloys are basically non-stoichiometry, grain refinement and alloying, owing to the introduction of lattice distortion, microdefects, larger specific surface area, the synergy of multiphase structure and better affinity to hydrogen [12,14,21]. According to this, non-stoichiometry is a promising way to further improve hydrogenation kinetics on the base of
alloying. However, the mechanism about effect of nonstoichiometry on hydrogenation properties of pseudobinary ZreTieV Laves alloys is still uncertain. It is necessary to explore the interrelation between microstructure/ phase constitute and hydrogenation kinetics/thermodynamic properties of non-stoichiometric Zr-based Laves alloys, which is beneficial for their practical application in fields of rapid storage and separation of hydrogen and its isotopes. Considering all above, the Zr0.9Ti0.1Vx alloys are designed where the Zr in A site is substituted by a small amount of Ti and the V in B site is non-stoichiometry based on our previous work about the role of defect structure on hydrogenation properties of Zr0.9Ti0.1V2 [15]. In the present work, the effects of phase content, microstructure and cell volume on hydrogenation kinetics and thermodynamic properties of nonstoichiometric Zr0.9Ti0.1Vx alloys are investigated. The defects in these alloys are observed, and the role of defects on hydrogenation kinetics is discussed. Detailed studies of hydrogenation kinetics are carried out to calculate reaction rate constant. Meanwhile, the mechanism about improvement of hydrogenation kinetics is further clarified. The possible hydrogenation mechanisms of non-stoichiometric Zr0.9Ti0.1Vx alloys are proposed based on the aforementioned experiments. The hydrogen storage properties including PCT characteristic and thermodynamic parameters from 673 to 823 K are investigated.
2.
Experimental procedures
Alloys with the compositions of Zr0.9Ti0.1Vx (x ¼ 1.7, 1.8, 1.9, 2.1, 2.2, 2.3) were prepared by non-consumable arc melting under argon atmosphere in a water-cooled copper crucible using 99.8wt% titanium, 99.4wt% zirconium and 99.5wt% vanadium. The ingots were turned over and re-melted 3 times to ensure the homogeneity, then annealed at 1273 K for 168 h in quartz tube under vacuum of 6 103 Pa followed by furnace cooling to room temperature. The phase compositions of the alloys were determined by a DX-2700 X-ray diffractometer using Cu Ka radiation. With a step scanning mode (step width 0.03 , counting time 2s), the diffraction data were collected at room temperature between 20 and 80 (2q). The microstructure of each sample was investigated by a scanning electron microscope (SEM) using secondary electron imaging and a transmission electron microscope (FEI Tecnai G2 F30, America) operating at 200 kV. The hydrogen storage characteristics including the P-t and PCT curves for each sample were obtained by using a Sievert type apparatus in temperature range of 673 Ke823 K. The limits of the three capacitance manometers range from 100 kPa to 104 Pa. Prior to the absorption runs, each sample was activated by heating the sample chamber to 723 K and pumping the chamber for 40e60 min to the set vacuum in order to create a clean surface free from the effect of surface contamination. The hydrogen absorption content was determined by calculating the pressure variation during the reaction while maintaining constant temperature of the reaction chamber.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
14677
3.
Results and discussion
Table 1 e The phase content and cell volume of each phase of the Zr0.9Ti0.1Vx alloys.
3.1.
Phase constitute and microstructure
Alloys
In order to characterize the phase compositions of the alloys, XRD analysis was performed. The XRD patterns of the as-cast and annealed Zr0.9Ti0.1Vx alloys are shown in Fig. 1. The phase content and cell volume of each phase of the Zr0.9Ti0.1Vx alloys are summarized in Table 1. The major diffraction peaks of the as-cast alloys shown in Fig. 1(a) can be identified to C15 ZrV2 Laves phase which crystallizes in the cubic (MgCu2 type) structure with space group Fd-3m. Besides the Laves phase, minor a-Zr and V-based BCC phases are also present. The structure does not change significantly with the introduction of non-stoichiometry. Meanwhile, a small amount of Zr3V3O phase is detected in Zr0.9Ti0.1V2.1 due to the oxidized raw materials or slight oxidation during the melting process, which has been observed in literature [11,22,23]. Existence of
Zr0.9Ti0.1V1.7
Zr0.9Ti0.1V1.8
Zr0.9Ti0.1V1.9
Zr0.9Ti0.1V2.1
Zr0.9Ti0.1V2.2
Zr0.9Ti0.1V2.3
Fig. 1 e XRD patterns of the Zr0.9Ti0.1Vx (x [ 1.7, 1.8, 1.9, 2.1, 2.2, 2.3) alloys, (a) as-cast, (b) annealed.
Phase content (wt. %)
Cell volume (103 nm3)
As-cast
Annealed
Annealed
C15(49) V(36) a-Zr(15) C15(45) V(38) a-Zr(17) C15(43) V(39) a-Zr(18) C15(38) V(41) a-Zr(19) Zr3V3O(2) C15(36) V(44) a-Zr(20) C15(36) V(43) a-Zr(21)
C15(54) Zr3V3O(43) V(3) C15(52) Zr3V3O(44) V(4) C15(55) Zr3V3O(40) V(5) C15(39) Zr3V3O(50) V(11)
416.11 1811.05 28.22 417.14 1818.92 28.42 414.13 1801.91 28.17 417.24 1810.23 28.30
C15(26) Zr3V3O(62) V(12) C15(40) Zr3V3O(46) V(14)
415.31 1808.89 28.27 414.74 1805.11 28.21
minor phases can be ascribed to the non-equilibrium solidification process of the arc-melted ingots and the incomplete peritectic reaction in the pseudo-binary system [24]. From Table 1, it can be noticed that the content of V-BCC solid solution phase slightly increases with increasing V content of the as-cast alloys, the C15eZrV2 phase content decreases accordingly and the a-Zr phase content hardly changed. The homogenizing treatment can eliminate a-Zr phase and generate Zr3V3O phase (see Fig. 1(b)). It can be seen from Table 1 that the C15 phase content of each annealed alloy increases comparing with the corresponding as-cast one except Zr0.9Ti0.1V2.2. The a-Zr phase and V-based BCC phase generated during the rapid solidification process have been almost eliminated by homogenizing in hypo-stoichiometric alloys. In the case of hyper-stoichiometry, the a-Zr phase can be eliminated but the V-based BCC phase can just be cut down. The content of Zr3V3O phase in annealed Zr0.9Ti0.1V2.2 alloy is higher than the others. The phase constitute has a significant influence on hydrogenation kinetics and thermodynamic properties which will be discussed later. The cell volume of each phase of the annealed alloys listed in Table 1 shows the regularity that cell volume of each phase decreases gradually with increasing V content in hyper-stoichiometric alloys. For the case of hypo-stoichiometry, the cell volume of each phase increases first then decreases to the least with increasing V content. The effects of cell volume on hydrogenation thermodynamic properties will be illustrated later. Observe Fig. 1 carefully, it can be seen that some peaks in the patterns are asymmetric. According to our analysis, asymmetric Bragg peaks in XRD patterns are probably due to axial divergence and non-ideal specimen geometry. The instrumental factor, especially axial divergence, is the main reason for the asymmetric Bragg peaks. The combined asymmetry effects result in the low angle sides of Bragg peaks being considerably broader than their high angle sides.
14678
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Fig. 2 gives the SEM images of as-cast Zr0.9Ti0.1Vx alloys with typical peritectic microstructure. Two-phase microstructure consisting of the dark dendritic V-based solid solution and ZrV2 Laves phase matrix are formed under the rapid solidification, the big thick V dendritic are quite prevalent. According to ZreV binary phase diagram [25], Zr and V have a classical peritectic reaction at 1300 C. During nonequilibrium solidification, the dendritic primary V-based solid solution precipitates firstly, then surrounded by ZrV2 Laves phase. The peritectic reaction from V-based solid solution and liquid phase to ZrV2 phase is suppressed, the residual liquid phase crystallizes to ZrV2 and b-Zr phase, then b-Zr phase transforms to a-Zr phase. The micro-segregation in microstructure of as-cast alloys is evident owing to the nonuniform composition in grain size during solidification. SEM images of annealed Zr0.9Ti0.1Vx alloys are shown in Fig. 3. After annealing at 1273 K for 168 h, the dendritic structures break up and the matrix regions increase during the annealing process. As can be seen from Fig. 3 (a) and (b), the peritectic microstructure observed in corresponding as-cast alloys have been almost eliminated, only a small quantity of V-based solid solution dispersed in the matrix phase. The metastable phase transition completes and the multistable configuration is obtained by annealing for hypo-stoichiometric alloys as has been observed in Zr0.9Ti0.1V2 alloy in Ref. [15]. But it is different in hyper-stoichiometric alloys, the dendritic V solid solution are still large and abundant in Fig. 3 (c) and (d), which results from the excess content of V element in hyper-stoichiometric alloys, these results accord well with the XRD analysis.
The TEM microstructures and corresponding selected area diffraction patterns (SADPs) of the annealed Zr0.9Ti0.1V1.7 and Zr0.9Ti0.1V2.2 alloys are shown in Fig. 4. The SADP1 and SADP2 indicates the existence of V-BCC phase and the C15eZrV2 phase respectively, the SADP3 indicates that the observed micro twins in C15eZrV2 phase are of {111}<011 > orientation relationship. There are large quantities of finely spaced line traces in C15eZrV2 matrix of both the hypo- and hyperstoichiometric alloys, which are identified as annealing twins and/or stacking faults. According to report of Zhu et al., the possible defect mechanisms in binary Laves phases are constitutional vacancy or anti-site substitution [26]. In V-rich alloys, the exhibition of twins or stacking faults could be explained as ZrCr2 Laves alloys [27]. The excess V atoms occupy the Zr sublattice sites which results in larger free volumes between two neighboring close packed {111} atomic planes, and thereby twins and stacking faults can be more easily introduced [28]. Such a result is also shown in the Vpoor compositions but the defect mechanisms may be more complicated. The defects can improve the hydrogen absorption kinetics by providing a preferential diffusion path for hydrogen and helping the nucleation of the hydride [16].
3.2.
Hydrogen absorption kinetics
The initial absorption capacities at room temperature of the experimental alloys are listed in Table 2. The hydrogen absorption capacity of Zr0.9Ti0.1V1.7 reaches 2.33 wt. % and the absorption capacity of V-poor alloys are slightly larger than
Fig. 2 e SEM secondary electron images of as-cast Zr0.9Ti0.1Vx alloys (a) x [ 1.7, (b) x [ 1.9, (c) x [ 2.1, (d) x [ 2.3.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
14679
Fig. 3 e SEM secondary electron images of annealed Zr0.9Ti0.1Vx alloys (a) x [ 1.7, (b) x [ 1.9, (c) x [ 2.1, (d) x [ 2.3.
that of the V-rich alloys. It indicates that the hydrogen storage capacity can be improved with the variety of the composition stoichiometry towards hypo-stoichiometry. The hydrogen absorption kinetics curves of annealed Zr0.9Ti0.1Vx alloys at 773 K are shown in Fig. 5. It can be seen that the hydrogen absorption concentration of the experimental alloys increases quickly with time and reaches kinetic stabilization within 50 s, the value for Zr0.9Ti0.1V1.7 is 15 s, which illustrates these alloys have a strong ability for rapid absorption hydrogen to saturation. The hydrogen absorption content at 773 K reaches 1.10 wt. % for Zr0.9Ti0.1V1.7 and Zr0.9Ti0.1V1.9, and is 0.98 wt. % for Zr0.9Ti0.1V2.3. Several mathematical models have been proposed to evaluate the hydrogenation kinetics mechanism of Zr-based AB2 alloys, such as Jander model or Chou model [29]. The experimental hydrogenation kinetics data can well fitted with the calculated results of theoretical models [30], but Jander model has its own limitation as just being applicable for small conversion rates and stable diffusion. Moreover, these methods are suitable for the case where hydrogen diffusion is the rate-controlling step both in the solid solution phase (a phase) and the hydride (b phase). In this work, the hydrogenation time of the experimental alloys are considerably short comparing with other alloys, especially for the a phase region. Understanding the diffusion theory and H diffusion rate-controlling mechanisms in other alloys [19,31], the diffusion can not be the rate-controlling step in such a short time. The possible hydrogen absorption reaction mechanism is proposed by investigating the observed
absorption rate based on Hirooka’s kinetic theory [32], which is developed on the assumption that the reaction rate to reach equilibrium is proportional to the deviation from the equilibrium. Fig. 6 displays the typical plot of ln[( pt - peq)/( p0 - peq)] vs. the reaction time t of Zr0.9Ti0.1Vx alloys at 773 K. It is evident that all the curves fit into two different linear parts, one is rapid absorbing stage 1 and the other is stable absorbing stage 2, the slope of each stage stands for relevant reaction rate constant ka1 or ka2. These values are summarized in Table 2. The ka1 values of the V-poor alloys are larger than those of the V-rich alloys, which indicates the hypo-stoichiometric alloys preserve faster hydrogenation kinetics. This could be explained by the break up and reduction of dendritic V solid solution and increase of ZrV2 matrix that enhance the flow of hydrogen in the hypo-stoichiometric alloys. Additionally, the ka1 values of the Zr0.9Ti0.1Vx alloys (except Zr0.9Ti0.1V2.2) are bigger comparing with those of the Zr(V1xFex)2 alloys at 773 K [11]. Base on the comparison of phase compositions in these alloys, the possible reason for this is the introduction of V BCC solid solutions in these alloys. It is reported the synergy of V BCC solid solutions leads to an autocatalytic mechanism to improve hydrogenation kinetics due to the increasing active surface area [15,33]. The ka1 value 0.228 for Zr0.9Ti0.1V1.7 is significantly larger than other experimental alloys and ZrTi0.2V1.8 alloy [14]. The better kinetics of Zr0.9Ti0.1V1.7 reveals the less content of V BCC solid solutions, the more positive synergy effect can be obtained. Meanwhile, the twin defects observed in Zr0.9Ti0.1V1.7 would accelerate the hydrogenation
14680
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Fig. 4 e TEM microstructures and corresponding selected area diffraction patterns (SADPs) of the annealed Zr0.9Ti0.1V1.7 (a)(d), (b) SADP1, (c) SADP2, (e) SADP3 and Zr0.9Ti0.1V2.2 (f)(g).
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
14681
Table 2 e The hydrogen absorption capacity at room temperature and hydriding reaction rate constant at 773 K of the Zr0.9Ti0.1Vx alloys. Zr0.9Ti0.1Vx x ¼ 1.7 x ¼ 1.8 x ¼ 1.9 x ¼ 2.1 x ¼ 2.2 x ¼ 2.3 wt. % ka1 ka2
2.33 0.228 0.015
2.26 0.170 0.014
2.27 0.164 0.023
2.22 0.147 0.032
2.25 0.107 0.019
2.20 0.140 0.027
kinetics. It can be concluded that the V BCC solid solutions should be restricted to a small content to achieve accelerated hydrogenation kinetics. The ka1 value 0.107 for Zr0.9Ti0.1V2.2 reveals the alloy preserves the slowest hydrogenation kinetics among the experimental alloys, which can be attributed to the large content of Zr3V3O phase hinder the rapid absorption of hydrogen. These curves involve a phase transformation and demonstrate the existence of two phase regions namely (a þ b) and b phase, which are observed by the different slopes of the two segments. The a-phase region could not be seen due to the fast hydriding reaction at relatively high pressure. The slope variation in the two stages intimates the rate-controlling mechanism changes during hydrogenation course. As proposed by Srinivas et al., the rate-controlling mechanism in a, (a þ b) and b phase regions can be assumed to the surface process, the interface process and the diffusion of hydrogen atoms through the growing hydride layer, respectively [34]. In stage 1, hydrogen atoms firstly penetrate through grain boundaries and dissolve in the bulk. Then the hydride layer can be formed at the grain circumference, the initial reaction rate is controlled by hydrogen flow through grain boundaries. As the hydrides grow, the volume of the grain expands gradually with increase of the hydrogen concentration. As a result, the reacting grains are pressed together and then narrows the transfer channel which ultimately reduces the hydrogen flow into the deeper parts of the sample [35]. When the b phase hydrides form completely, the reaction rate of stage 2 is controlled by hydrogen diffusion through the hydride phase.
Fig. 6 e ln[( pt e peq)/( p0 e peq)] vs. t plots of annealed Zr0.9Ti0.1Vx (x [ 1.7, 1.8, 1.9, 2.1, 2.2, 2.3) alloys at 773 K.
3.3.
Thermodynamic characteristics
The pressure-concentration-temperature (PCT) characteristics of annealed Zr0.9Ti0.1Vx (x ¼ 1.7, 1.8, 1.9, 2.2) alloys from 673 to 823 K are shown in Fig. 7. The absorption curves clearly indicate the presence of a, a þ b and b phase regions according to slope change at various temperatures as observed in some Zr-based AB2 alloys [20,21]. During the initial stage of absorption, physisorption and chemisorptions take place and only a solid solution phase is formed in the a-phase region, which is very short in the PCT curves of the experimental alloys. As the absorption continuing, the hydride phase nuclei starts growing and the plateau pressure is obtained owing to the formation of b hydride phase. When the b phase hydrides form completely, the hydrogen atoms get redistributing in the hydride phase, the hydrogen absorbing capacity reaches saturation. It is difficult to determine the plateau pressure from the PCT curves as the curves exhibit the characteristic of sloping plateaus. The presence of a sloping plateau is due to the formation of different hydrides. The slope of plateaus increases while the width decreases as the composition of the alloys changes from V-poor to V-rich. All the hydrogen desorption curves at 823 K of the experimental alloys are close to the absorption curves at that temperature, that is to say no significant hysteresis effect exists between absorption and desorption. From the PCT characteristics of Zr0.9Ti0.1Vx (x ¼ 1.7, 1.8, 1.9, 2.2) alloys, the relative thermodynamic properties of dissolved hydrogen could be obtained at a particular hydrogen concentration by Van’t Hoff equation [36]: InPeq ¼ DHQ =RT DSQ =R
Fig. 5 e Hydrogen absorption kinetics curves of annealed Zr0.9Ti0.1Vx (x [ 1.7, 1.8, 1.9, 2.1, 2.2, 2.3) alloys at 773 K.
(1)
The plots for lnP vs. 1000/T of the experimental alloys are displayed in Fig. 8, where P is the equilibrium pressure obtained from the absorption isotherms at several specific hydrogen concentrations in the a þ b two-phase region. The experimental data of each sample can be interpolated linearly
14682
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
Fig. 7 e Hydrogen absorption/desorption PCT curves of the Zr0.9Ti0.1Vx alloys from 673 to 823 K (a) x [ 1.7, (b) x [ 1.8, (c) x [ 1.9, (d) x [ 2 [15], (e) x [ 2.2.
well and exhibit good relativity, which illustrates the relatively stable hydrides are generated in the range of H/A content. The enthalpy change (OHQ) and entropy change (OSQ) for a particular hydrogen concentration can be calculated from the slope and intercept of the least square fit for lnPeq vs. 1000/T plots, respectively. The thermodynamic parameters of the experimental alloys are listed in Table 3. It is easy to find both the -OHQ and -OSQ values are higher than that of some other Zr-based Laves phase alloys [14,37,38], the higher -OHQ values indicate the Zr0.9Ti0.1Vx alloys forming more stable hydrides. It is obvious that the -OHQ and -OSQ values of each experimental alloy increase gradually with the increasing hydrogen content. This is understandable in view of the overall process that more hydrogen transfers from the gaseous state into the solid state. The -OHQ values for Zr0.9Ti0.1V2.2 are considerable smaller than those for the Vpoor alloys which can be attribute to the low affinity of Zr3V3O phase to hydrogen, large content of Zr3V3O phase decreases the stability of hydrides. This will be beneficial to decrease the hydrogen desorption temperature. According to the works of Jain and Lundin et al. [39,40], the interstitial size increases with the expansion of ZrV2 cell volume in the alloys. As a result, the binding energy between metal and hydrogen atoms reduces with the increasing metal-hydrogen distance, and correspondingly leads to the higher -OHQ/-OSQ value and lower plateau pressure. But in the experimental hypostoichiometric alloys, the regularity between cell volume and -OHQ value is contrary to the aforementioned. The cell
volume of each phase increases first then decreases to the least with increasing V content, and the -OHQ value decreases first then increases to the maximum accordingly. The reason for this should be the multi-phase structure in these alloys, the three phases ZrV2, Zr3V3O and V BCC solid solutions all can form corresponding hydrides. The -OHQ value is the combined result of three hydrides, so the factors which affect -OHQ value are more complex. It suggests that the thermodynamic properties are more affected by the relative content of each phase and its corresponding hydride than the cell volume. The equilibrium hydrogen pressures at room temperature can be calculated by Van’t Hoff equation, these extrapolated values of the Zr0.9Ti0.1Vx alloys are all below 107Pa, which indicates these alloys have strong ability to remove traces of residual hydrogen. Even this method has its indirect measurement deviation and indeterminacy, the extrapolated equilibrium pressures at room temperature are helpful to give the data which can not be obtained directly from the experiment owing to the extreme low pressure and they can be regard as references.
4.
Conclusions
In summary, the microstructure and hydrogen absorption kinetics and thermodynamic properties of non-stoichiometric AB2 Laves alloys Zr0.9Ti0.1Vx have been investigated in this work. The multiphase structure consisting of ZrV2 Laves
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
14683
Fig. 8 e Van’t Hoff plots for the Zr0.9Ti0.1Vx alloys (a) x [ 1.7, (b) x [ 1.8, (c) x [ 1.9, (d) x [ 2.2.
phase, a-Zr and V-BCC solid solution phase are analyzed. The homogenizing treatment can eliminate the minor a-Zr phase and reduce the V BCC phase in as-cast alloys. Twin defects with {111}<011 > orientation relationship are observed, and they can accelerate hydrogenation kinetics. The hypo-
Table 3 e Thermodynamic parameters of the Zr0.9Ti0.1Vx alloys. Alloys Zr0.9Ti0.1V1.7
Zr0.9Ti0.1V1.8
Zr0.9Ti0.1V1.9
Zr0.9Ti0.1V2.2
Content of H/A
OHQ (KJ mol1)
OSQ (J mol1 K1)
0.3 0.45 0.6 0.75 0.9 0.3 0.45 0.6 0.75 0.9 0.3 0.45 0.6 0.75 0.9 0.3 0.45 0.6 0.75
80.53 90.39 99.38 105.75 107.99 81.57 84.37 89.35 91.50 94.47 85.60 96.82 108.01 115.97 122.57 60.45 63.06 69.88 78.86
115.86 133.69 150.22 163.24 170.98 118.03 128.13 140.72 149.29 158.89 122.59 141.91 160.67 175.27 187.67 75.30 83.83 98.23 116.26
stoichiometric alloys perform faster hydrogenation kinetics than the hyper-stoichiometric ones due to the break up and reduction of dendritic V enhancing the diffusion of hydrogen. The synergy induced by a small content of V BCC solid solutions can improve hydrogenation kinetics. Two stages during hydrogen absorption process could be attributed to the different rate controlled mechanism including hydrogen flow through grain boundaries and hydrogen diffusion through the hydride phase. The excess content of Zr3V3O phase decreases the hydrogenation kinetics and the stability of hydrides.
Acknowledgments This work was financially supported by State Key Laboratory of Solidification Processing (NWPU) (70-QP-2010). The Program of Introducing Talents of Discipline in the Project of Advanced Materials and their Forming Technology (B08040) is also acknowledged. Prof Shi Liu and Mr. Liang-yin Xiong from the Institute of Metal Research (Chinese Academy of Sciences, Shenyang, China) are gratefully appreciated for their help in hydrogenation measurement and helpful discussions.
references
[1] Ruiz FC, Castro EB, Peretti HA, Visintin A. Study of the different ZrxNiy phases of Zr-based AB2 materials. Int J Hydrogen Energ 2010;35:9879e87.
14684
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 6 7 5 e1 4 6 8 4
[2] Jain A, Jain RK, Agarwal S, Ganesan V, Lalla NP, Phase DM, et al. Synthesis, characterization and hydrogenation of ZrFe2xNix (x ¼ 0. 2, 0.4, 0.6, 0.8) alloys. Int J Hydrogen Energ 2007;32:3965e71. [3] Young K, Nei J, Ouchi T, Fetcenko MA. Phase abundances in AB(2) metal hydride alloys and their correlations to various properties. J Alloy Compd 2011;509:2277e84. [4] Jain A, Jain RK, Agarwal G, Jain IP. Crystal structure, hydrogen absorption and thermodynamics of Zr1xCoxFe2 alloys. J Alloy Compd 2007;438:106e9. [5] Bulyk II , Basaraba YB, Trostianchyn AM. Features of the HDDR process in ZrT2 (T ¼ Cr, Mn, Fe, Co) compounds. J Alloy Compd 2004;367:283e8. [6] Pebler A, Gulbransen EA. Thermochemical and structural aspects of the reaction of hydrogen and intermetalic compounds of zirconium. Electrochem Tech 1966;4:211e5. [7] Shaltiel D, Jacob I, Davidov D. Hydrogen absorption and desorption properties of AB2 laves-phase pseudobinary compounds. J Less-Common Met 1977;53:117e31. [8] Jurczyk M, Rajewski W, Wo´jcik G, Majchrzycki W. Metal hydride electrodes prepared by mechanical alloying of ZrV2type materials. J Alloy Compd 1999;285:250e4. [9] Xu Y-h, Wang G, Chen C-p, Wang Q-d, Wang X. The structure and electrode properties of non-stoichiometric A1.2B2 type C14 Laves alloy and the effect of surface modification. Int J Hydrogen Energ 2007;32:1050e8. [10] Kim JH, Lee H, Hwang KT, Han JS. Hydriding behavior in Zrbased AB(2) alloy by gas atomization process. Int J Hydrogen Energ 2009;34:9424e30. [11] Zhang TB, Wang XF, Hu R, Li JS, Yang XW, Xue XY, et al. Hydrogen absorption properties of Zr(V1xFex)2 intermetallic compounds. Int J Hydrogen Energ 2012;37:2328e35. [12] Iwakura C, Kasuga H, Kim I, Inoue H, Matsuoka M. Effect of alloy composition on electrochemical properties of the Zrbased Laves-phase hydrogen storage alloys. Electrochim Acta 1996;41:2691e4. [13] Klein B, Simon N, Klyamkine S, Latroche M, PercheronGuegan A. Improvement of the thermodynamical and electrochemical properties of multicomponent Laves phase hydrides by thermal annealing. J Alloy Compd 1998;280:284e9. [14] Yang XW, Zhang TB, Hu R, Li JS, Xue XY, Fu HZ. Microstructure and hydrogenation thermokinetics of ZrTi0.2V1.8 alloy. Int J Hydrogen Energ 2010;35:11981e5. [15] Yang XW, Li JS, Zhang TB, Hu R, Xue XY, Fu HZ. Role of defect structure on hydrogenation properties of Zr0.9Ti0.1V2 alloy. Int J Hydrogen Energ 2011;36:9318e23. [16] Michel N, Poulat S, Priester L, Dantzer P. Microstructural analysis of the thermodynamically controlled hydride phases grown in the ZrNi-H2 system. Mat Sci Eng A-struct 2004;384:224e31. [17] Young K, Ouchi T, Yang J, Fetcenko MA. Studies of offstoichiometric AB(2) metal hydride alloy: Part 1. Structural characteristics. Int J Hydrogen Energ 2011;36:11137e45. [18] Kim D-M, Jang K-J, Lee J-Y. A review on the development of AB2-type Zr-based Laves phase hydrogen storage alloys for NieMH rechargeable batteries in the Korea Advanced Institute of Science and Technology. J Alloy Compd 1999;293e295:583e92. [19] Kandavel M, Ramaprabhu S, Jung M, Wipf H. Solubility and diffusion of hydrogen in AB(2)-type Laves phase alloys. J Alloy Compd 2005;404:265e8.
[20] Kandavel M, Ramaprabhu S. Hydriding properties of Tisubstituted non-stoichiometric AB(2) alloys. J Alloy Compd 2004;381:140e50. [21] Kandavel M, Ramaprabhu S. The effect of non-stoichiometry on the hydrogen storage properties of Ti-substituted AB(2) alloys. J Phys-condens Mat 2003;15:7501e17. [22] Gondor G, Lexcellent C. Analysis of hydrogen storage in metal hydride tanks introducing an induced phase transformation. Int J Hydrogen Energ 2009;34:5716e25. [23] Huot J, Akiba E, Ishido Y. Crystal structure of multiphase alloys (Zr, Ti)(Mn, V)2. J Alloy Compd 1995;231:85e9. [24] Smith JF. Phase diagrams of binary vanadium alloys. Alloy phase diagrams. Metals Park, Ohio: ASM; 1992. [25] Klein B, Redeker A, Zu¨chner HZ. Electrochemical measurements of hydrogen diffusion in ZrV2 and ZrV0.5Ni1.5. Z Physik Chem 1993;181:95e101. [26] Zhu JH, Liu CT. Defect structures in ZrCo2 Laves phase. Acta Mater 2000;48:2339e47. [27] Kanazawa S, Kaneno Y, Inoue H, Kim WY, Takasugi T. Microstructures and defect structures in ZrCr2 Laves phase based intermetallic compounds. Intermetallics 2002;10:783e92. [28] Takasugi T, Yoshida M, Hanada S. Deformability improvement in C15 NbCr2 intermetallics by addition of ternary elements. Acta Mater 1996;44:669e74. [29] Chou K-C, Li Q, Lin Q, Jiang L-J, Xu K-D. Kinetics of absorption and desorption of hydrogen in alloy powder. Int J Hydrogen Energ 2005;30:301e9. [30] Cui X-Y, Li Q, Chou K-C, Chen S-L, Lin G-W, Xu K-D. A comparative study on the hydriding kinetics of Zr-based AB2 hydrogen storage alloys. Intermetallics 2008;16:662e7. [31] Vojtech D, Knotek V. Study of the diffusion kinetics and mechanism of electrochemical hydriding of Mg-Ni-Mm alloys. Int J Hydrogen Energ 2011;36:6689e97. [32] Hirooka Y, Miyake M, Sano T. A study of hydrogen absorption and desorption by titanium. J Nucl Mater 1981;96:227e32. [33] Latroche M. Structural and thermodynamic properties of metallic hydrides used for energy storage. J Phys Chem Solids 2004;65:517e22. [34] Srinivas G, Sankaranarayanan V, Ramaprabhu S. Hydrogen absorption and desorption properties of Ho1x MmxCo2 alloys. Int J Hydrogen Energ 2007;32:2480e7. [35] Bloch J. The temperature-dependent changes of the kinetics and morphology of hydride formation in zirconium. J Alloy Compd 1995;216:187e95. [36] Flanagan TB, Clewley JD. Hysteresis in metal hydrides. J LessCommon Met 1982;83:127e41. [37] Nakano H, Wakao S. Substitution effect of elements in Zrbased alloys with Laves phase for nickel-hydride battery. J Alloy Compd 1995;231:587e93. [38] Jain A, Agarwal S, Jain IP. Correlation between the milling time and hydrogen-storage properties of nanostructured ZrFeNi ternary alloy. J Alloy Compd 2009;480:325e8. [39] Jain A, Jain RK, Agarwal S, Jain IP. Thermodynamics and structural aspects of hydrogen absorption in Zr1xCrxFe2 alloys. Int J Hydrogen Energ 2007;32:2445e9. [40] Lundin CE, Lynch FE, Magee CB. A correlation between the interstitial hole sizes in intermetallic compounds and the thermodynamic properties of the hydrides formed from those compounds. J Less-Common Met 1977;56:19e37.