- Email: [email protected]
Structural and electrochemical hydrogen storage properties of MgTiNix (x = 0.1, 0.5, 1, 2) alloys prepared by ball milling
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 x x x ( 2 0 1 5 ) 1 e6
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling Z. Zhang a,*, O. Elkedim a, M. Balcerzak b, M. Jurczyk b FEMTO-ST, MN2S, Universite de Technologie de Belfort-Montbeliard, Site de Sevenans, 90010 Belfort Cedex, France Institute of Materials Science and Engineering, Poznan University of Technology, Sklodowska-Curie 5 Sq., 60-965 Poznan, Poland
a
b
article info
abstract
Article history:
A series of MgeTieNi alloys with different Ni content was prepared by mechanical alloying
Received 28 September 2015
using a planetary high-energy ball mill. The structural transformation was characterized
Received in revised form
by XRD and SEM. It indicated that the addition of 10 at% Ni did not hinder the amorph-
26 November 2015
ization of MgTi BCC phase. After 40 h of milling, the alloys of MgTiNi and MgTiNi2 mainly
Accepted 30 November 2015
consisted of the mixture of TiNi þ MgNi þ Mg2Ni and TiNi þ TiNi3 þ MgNix (x > 1),
Available online xxx
respectively. Their discharge capacities were investigated by electrochemical measurements at galvanostatic conditions. It was shown that all of the studied samples possessed
Keywords:
good cycling performance. Among them, MgTiNi showed the highest discharge capacity. In
MgeTieNi alloys
the ternary alloys with the same atom ratio of Ti and Mg, the number of activation cycles
Mechanical alloying
decreased with increasing the Ni content.
Phase structure
Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Electrochemical hydrogen storage properties
Introduction As energy crisis and environmental pollution are becoming more and more serious, hydrogen is widely regarded as the most promising energy resource to replace the traditional fossil fuel. An efficient and safe hydrogen storage material with small volume and low cost is the key issue for the utilization of hydrogen as an energy carrier. Besides, another notable application area for hydrogen storage material is in Ni-MH batteries. In recent years, a lot of high performance hydrogen storage alloys have been developed. Among them, magnesium based and titanium based alloys have attracted
much attention due to their extremely high discharge capacity and low cost [1e5]. Theoretically, the hydrogen storage abilities of MgH2 and TiH2 are 7.6 wt.% and 4.0 wt.% respectively. However, due to the irreversible hydrogen absorb reaction of TiH2 at room temperature and the low decompositionformation kinetics of MgH2, neither of them can be practically applied. In order to improve their hydrogen storage properties, transition metal elements as electrocatalysts have been introduced into the system. For example, amorphous MgNi which was measured in the electrochemical reaction possessed a high electrochemical capacity reaching 500 mAh/ g, but with a bad cycling performance [6,7]. The crystalline Mg2Ni can reversibly absorb and desorb hydrogen at high
* Corresponding author. Tel.: þ33 753341514; fax: þ33 384583000. E-mail addresses: [email protected], [email protected] (Z. Zhang). http://dx.doi.org/10.1016/j.ijhydene.2015.11.168 0360-3199/Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
2
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 x x x ( 2 0 1 5 ) 1 e6
temperature. However, the poor kinetics and bad antioxidation of Mg2Ni can not be ignored [8e10]. TiNi phase was found to possess a capacity of 230 mAh/g and good corrosion resistance in alkaline solution [11e13]. Unfortunately, the poor absorption/desorption kinetics and complex activation procedure become the obstacles for the practical application of TiNi alloys as hydrogen storage material. In order to investigate the synergistic effect of different phases on the enhancement of electrochemical performance, a lot of researchers have made efforts to optimize the composition for exploring the balance of discharge capacity and cycling performance [11,14e19]. In this work, the influence of Ni content on the structure and electrochemical performance of TieMgeNi alloys has been investigated. The obtained results are expected to provide more data about structural transformation and electrochemical properties of TieMgeNi ternary system, and to contribute to the design of new hydride compounds.
Experimental methodology The samples were prepared by Mechanical Alloying (MA) from the pure elemental powders of Mg (purity 99.8%, particle size 50 mm, GoodFellow), Ti (purity 99.5%, particle size 150 mm, GoodFellow), Ni (purity 99.5%, particle size 250 mm, GoodFellow) according to the following stoichiometries: MgTiNix (x ¼ 0.1, 0.5, 1, 2). The nominal composition of mixtures were introduced in cylindrical stainless steel vials (capacity 50 ml) with two stainless steel balls (diameter 20 mm) under a high pure argon atmosphere. The MA was performed for 20 h and 40 h at room temperature using a planetary high-energy ball mill (Retsch PM 400) at the speed of 400 RPM. The ball-to-powder weight ratio (BPR) was 10:1. In order to dissipate the heat and reduce the cold welding of powder particles during high-energy milling, the milling was interrupted for 30 min after every 1 h of operation. And after every 10 h of milling, the ball milling was stopped for scraping the powder adhered to the milling chamber walls and grinding balls. All of these operations were carried out under protective atmosphere (Ar). To investigate the transformation of phase structure, the samples with different compositions after different milling time were characterized by X-ray diffraction with Co Ka radiation (l ¼ 1.789 A). The morphology of the samples were determined using a scanning electron microscope (SEM, JEOL, SM-5800LV). The electrochemical charge and discharge measurements were performed using Muti-channel Battery Interface ATLAS 0461. The negative electrodes (working electrodes) materials consisted of as-milled powder and carbonated nickel with the ratio of 90:10. The mixtures were compressed between two nickel nets into a small pellet. NiOOH/Ni(OH)2 and Hg/HgO were employed as counter electrode and reference electrode respectively. All of the electrodes were soaked in 6 M KOH solution which was prepared from KOH flakes and distilled water. Before testing, in order to eliminate adverse oxidative effects (activation process), the electrodes were soaked in the electrolyte for 24 h at room temperature and 1 h at 100 C. The current density for charging and discharging were 40 mA/g.
The cut-off potential was 0.7 V vs. the reference electrode. Cycle stabilities were evaluated by capacity retaining rate after 18th cycle: Rh ¼ (C18/Cmax) 100%, where C18 and Cmax were discharge capacities at the 18th cycle and maximum discharge capacity, respectively.
Results and discussion Microstructure analysis A series of overlapped X-ray diffraction patterns of different composition samples with fixed milling times of 40 h are shown in Fig. 1. Apparently, it can be found that, accompanying with the increasing of Ni element content (from bottom to top), the main peaks of the XRD pattern moved towards the higher angles gradually. In the XRD pattern of MgTiNi0.1 (Fig. 1(a)), a broad peak was centered at 2q ¼ 44 , which is typical for amorphous phase. Substantially, in the case of MgTi alloys, a BCC phase at around 38 (characterized by XRD with Cu Ka radiation) has been reported several times [20e22]. To the best of our knowledge, this is the first time when a BCC phase was observed in Ni doped MgTi system [23]. In our previous work, MgTi BCC phase had been obtained in MgTi binary alloys after 60 h of milling (not shown) which means, after adding slight amount of Ni, the milling duration of MgTi amorphous is reduced to 40 h. This fact implies that the addition of 10 at% Ni did not seem to hinder the amorphization of MgTi alloys but may rather promote the grinding efficiency. In Fig. 1(b), the XRD pattern of MgTiNi0.5 after 40 h milling is shown. In order to analyze the structural transformation, the XRD analysis of this composition with respect to the different milling time is shown in Fig. 2. After 20 h of milling, the peaks of Ti, Mg and Ni can still be identified as elemental states, despite the fact that the latter two peaks inclined slightly to the Ti peak centered direction (2q ¼ 47 ). After further milling for 20 h, the diffraction peaks of Ni disappeared whereas the peaks of Mg were present. The main broad peak with two shoulders was observed which was assumed to be the mixture
Fig. 1 e Overlapped XRD patterns of samples milled for 40 h: (a) MgTiNi0.1, (b) MgTiNi0.5, (c) MgTiNi and (d) MgTiNi2.
Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
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 x x x ( 2 0 1 5 ) 1 e6
3
Fig. 2 e XRD patterns of MgTiNi0.5 mixture after different milling time: (a) 20 h, (b) 40 h.
Fig. 3 e XRD patterns of MgTiNi mixture after different milling time: (a) 20 h, (b) 40 h.
of TiNi-based phase and Ti. After comparing to Fig. 2(b), the tendency of amorphization of Ti and Mg became more apparent indicating that a stationary state of this composition might be attained by prolonging milling. It is certain that the elemental Mg and Ti can be detected after 40 h of milling based on the visualized comparison in Fig. 2. The diffraction peak of Mg maintained relatively sharp while that of Ti significantly broadened. The same phenomenon had also been reported, which is related to the investigation of milling production of pure Ti and Pure Mg [24]. According to the index results of phase, the broad feature of Ti centered peak is attributed to the presence of the overlapped diffraction peaks of Ti2Ni at 47.8 . It is notable that both of Ti and Mg were abundant in this composition, but only TiNibased phase was detected after milling. The formation of Ti2Ni instead of Mg2Ni may be due to two following reasons. Firstly, the atomic radii of Mg and Ti are 0.160 nm and 0.147 nm, respectively, which means Ti possesses a relatively higher diffusivity than Mg. Secondly, the enthalpy of formation of Mg2Ni and Ti2Ni are 51.9 kJ/mol [25] and 71.4 kJ/mol [26] respectively. All these evidences indicate that, the combination of Ti and Ni in atomic scale is easier than that of Mg and Ni. Besides, according to the phase diagram of TiNi system, Ti abundant is more favorable for the formation of Ti2Ni phase than that of other phases. As a result, the phase structure of MgTiNi0.5 after 40 h milling consisted of Ti2Ni, Mg and Ti. The XRD patterns of MgTiNi with different milling time (40 h and 20 h) are presented in Fig. 3. The typical amorphous phase at 49 was found in both of Fig. 3(a) and (b). Particularly, the diffraction pattern shown in Fig. 3(b) had been observed in other references [16,27,28] concerning with the products of the MgNi by mechanical alloying. The broad peak belonged to the MgNi amorphous phase, and the remaining sharp peak was related to the nickel residue. It is indicated that, in this case, 20 h of milling is not enough for Ni to neither completely dissolve into the MgNi amorphous phase nor form TiNi crystalline phase. Therefore, the diffraction peaks of titanium are
considered to be superimposed in the broad feature. After 40 h of milling, the diffraction peaks of MgNi and TiNi were detected. The main characteristic peak increased in intensity and became sharper while the peaks of Ni residue disappeared. All of these changes suggest that crystallization was promoted in this milling duration. Additionally, Mg2Ni was supposed to be formed during 40 h of milling which achieved by following analysis. As seen in Fig. 3(b), there was no peak of MgTi phase detected which means only the TieNi and MgeNi binary phase can be synthesized. While the observed phases, so far, were mainly binary phase with equal atom ratio. According to the stoichiometric proportion, these may lead to two possible assumptions. One is a high degree of excess dispersion of magnesium or/and titanium, while the other one is the existence of another Mg-rich or/and Ti-rich phase. As no protruded elemental peak was observed in the XRD pattern, the formation of nanocrystals Mg2Ni and Ti2Ni cannot be excluded. In fact, the recrystallization of MgNi amorphous phase during prolongation milling was report elsewhere [7,29]. The crystallizations of MgNi alloy into a mixture of nanocrystals Mg2Ni and MgNi2 took place after a certain time of milling which may differ from the various mill mode. In n et al. [30] sugaddition, an observation reported by Guzma gested that Mg2Ni and MgNi2 nanocrystals detected by highresolution microscopy bedded on amorphous MgNi may also lead to the absence of MgeNi phase in the diffraction pattern. And it is considered that the formation of Mg2Ni is reasonable compared with that of MgNi2 in such a superstoichiometric magnesium environment. Fig. 4 shows the XRD patterns of MgTiNi2 with respect to different milling time. It is obvious to see the phase transformation in this duration. The broad peak centered at 49.8 was observed in both of patterns which also appeared in Ref. [17]. Authors of this work obtained a broad peak which consisted of TiNi and MgNi amorphous phases after milling MgTiNi2 alloys. However, our results may differ from them. In our case, after comparing with the XRD of Fig. 3, the peak of MgNi amorphous phase in MgTiNi and MgTiNi2 were not the
Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
4
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 x x x ( 2 0 1 5 ) 1 e6
was characterized clearly which was coexisted with TiNi and MgNix (x > 1) amorphous in the 40 h as-milled sample. The SEM images of different samples after 40 h of milling are shown in Fig. 5. It can be found that, all of the morphology of samples consisted of micrometric particles. In comparison with the different micrographs, with increasing the Ni content, the particles were getting more irregular which had a tendency to agglomerate. Accompanying the increasing of the Ni content, more fine particles smaller than 5 mm and agglomerated particles bigger than 20 mm were found. This reflects that, repeated fracturing and the cold-welding were occurring together during the ball milling process. Ni as a positive factor which can improve the milling efficiency of the alloys which is consistent with our previous analysis of XRD.
Electrochemical tests Fig. 4 e XRD patterns of MgTiNi2 mixture after different milling time: (a) 20 h, (b) 40 h.
same which were centered in 49 and 49.8 , respectively. Guo et al. [31] investigated the effect of Ni content on the structural transformation of MgNi amorphous phase. In our observations, the peak of MgNi amorphous phase slightly shifted to higher angle with adding Ni, which is pronounced for Guo's result. This fact implies that the MgNix (x > 1) amorphous phase was formed in MgTiNi2. Moreover, peaks of TiNi were found in both of patterns. The shoulder peak obtained in Fig. 4(a) which was corresponding to the dissolving Ni disappeared after continually 20 h of milling. Instead, TiNi3 phase
The discharge capacities of as-milled samples according to the cycle numbers are shown in Fig. 5 respectively. Most of the samples achieved their maximum capacities after several cycles of activation except MgTiNi0.1, whose hydrogen discharge capacity was very low, even lower than 5 mAh/g. It was reported that the hydrogen discharge capacity of ballmilled binary MgTi alloys was close to zero [20]. By adding 10 at% of Ni, the electrochemical activity of MgTi alloy was not improved. The dispersion of Ni in MgTi BCC phase can not motivate the chargeetransfer reaction. The capacity retaining rate (Rh) of different samples is shown in Table 1. Among all of the electrochemical studied materials, MgTiNi possessed the best discharge capacity of 154 mAh/g in the second charge/discharge cycle. The good
Fig. 5 e SEM micrographs of samples milled for 40 h: (a) MgTiNi0.1, (b) MgTiNi0.5, (c) MgTiNi, and (d) MgTiNi2. Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
5
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 x x x ( 2 0 1 5 ) 1 e6
Table 1 e Phase structure and discharge capacities of 40 h as-milled samples. Composition Main phase Cmax (mAh/g) C18 (mAh/g) Retaining rate (R18)
MgTiNi0.5
MgTiNi
MgTiNi2
Ti2Ni, Ti and Mg 81 67 82%
MgNi, TiNi, Mg2Ni 154 123 80%
TiNi3, MgNix (x > 1), TiNi 93 86 92%
electrochemical performance is considered to attribute to the formation of MgNi amorphous phase. MgNi amorphous phase possessed a high discharge capacity but also decreased rapidly with cycling [32]. Unlike that, the average capacity decay of MgTiNi was slow as 1.1%, which is due to the coexisted TiNi and Mg2Ni. TiNi possesses good corrosion resistance in alkaline solution. Meanwhile, the synergistic effects of TiNi and Mg2Ni MA alloys on the electrochemical performance was also reported by Huang et al. [33]. In contrast to MgTiNi alloy, as found in Table 1, MgTiNi2 alloy which also consisted of TiNi and MgNi amorphous phases possessed a lower discharge capacity but better cycling performance comparing with MgTiNi. This is due to the increasing content of Ni. Firstly, TiNi3, as one of the main phase achieved in the as-milled powder is an inactive phase which can not store hydrogen [17]. Secondly, the features of MgNix (x > 1) amorphous and MgNi amorphous phases are different. Zhang et al. [10] reported the characteristics of amorphous MgNi1.5 and MgNi2 whose discharge capacity was getting lower but the circulation property was getting better gradually, accompanying with the increasing of the Ni addition. This is in agreement with our result. Therefore, it can be concluded that, the phases of MgNix (x > 1) and TiNi3 can be formed by adding Ni, resulting in the improvement of the cycling performance at the expense of reduction of the discharge capacity (see Fig. 6). In Table 1, MgTiNi0.5 possessed the worst discharge capacity which is not surprising considering the existence of big amount of unreacted Ti and Mg. The hydrogen storage phase in this phase is Ti2Ni. It is certain that the electrochemical property of MgTiNi0.5 could be improved by prolongation of milling time. Additionally, it is noted that, there was a large range of difference between the first two discharge capacities
of MgTiNi0.5, which means it needs several cycles of activation. This phenomenon is a typical characteristic of Ti2Ni alloys which was reported elsewhere [34,35]. In comparison of other samples, it is found that the number of activation cycles decreased with increasing the Ni content in the ternary MgeTieNi alloys, which is in agreement with that obtained in Ref. [17].
Conclusion In this study, MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys was prepared by mechanical alloying using a planetary high-energy ball mill. It is found that: - Adding 10 at% of Ni in MgTi alloys did not hinder the amorphization of MgTi BCC phase, nor improve the discharge capacity of MgTi alloys. - Based on the XRD analysis, the alloys of MgTiNi and MgTiNi2 mainly consisted of the mixture of TiNi þ MgNi þ Mg2Ni and TiNi þ TiNi3 þ MgNix (x > 1), respectively. The average discharge capacity decay of studied samples were all less than 1.1% per cycle. Among them, MgTiNi showed the highest discharge capacity which is attributed to the MgNi amorphous phase. - In the ternary alloys with the same atom ratio of Ti and Mg, the number of activation cycles decreased with increasing the Ni content.
Acknowledgments This research is supported by China Scholarship Council (CSC) and UTBM in the framework of UT-INSA project. We would like to thank M. Bruno Nicolas and M. Yangzhou Ma (LERMPS, UTBM, France) for their assistance with the SEM and XRD measurements.
references
Fig. 6 e The discharge capacities of 40 h as-milled samples.
[1] Sakintuna B, Lamari-Darkrim F, Hirscher M. Metal hydride materials for solid hydrogen storage: a review. Int J Hydrogen Energy 2007;32:1121e40. [2] Schlapbach L, Zu¨ttel A. Hydrogen-storage materials for mobile applications. Nature 2001;414:353e8. [3] Jain I, Lal C, Jain A. Hydrogen storage in Mg: a most promising material. Int J Hydrogen Energy 2010;35:5133e44. [4] Tliha M, Khaldi C, Boussami S, Fenineche N, El-Kedim O, Mathlouthi H, et al. Kinetic and thermodynamic studies of hydrogen storage alloys as negative electrode materials for Ni/MH batteries: a review. J Solid State Electrochem 2014;18:577e93.
Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
6
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 x x x ( 2 0 1 5 ) 1 e6
[5] Shang C, Bououdina M, Song Y, Guo Z. Mechanical alloying and electronic simulations of (MgH2þM) systems (M ¼ Al, Ti, Fe, Ni, Cu and Nb) for hydrogen storage. Int J Hydrogen Energy 2004;29:73e80. [6] Iwakura C, Nohara S, Inoue H, Fukumoto Y. Surface modification of MgNi alloy with graphite by ball-milling for use in nickel-metal hydride batteries. Chem Commun 1996:1831e2. [7] Ruggeri S, Lenain C, Roue L, Liang G, Huot J, Schulz R. Mechanically driven crystallization of amorphous MgNi alloy during prolonged milling: applications in Ni-MH batteries. J Alloys Compd 2002;339:195e201. [8] Inoue H, Ueda T, Nohara S, Fujita N, Iwakura C. Effect of ballmilling on electrochemical and physicochemical characteristics of crystalline Mg2Ni alloy. Electrochim acta 1998;43:2215e9. [9] Ueda TT, Tsukahara M, Kamiya Y, Kikuchi S. Preparation and hydrogen storage properties of MgeNieMg2Ni laminate composites. J Alloys Compd 2005;386:253e7. [10] Zhang SG, Hara Y, Morikawa T, Inoue H, Iwakura C. Electrochemical and structural characteristics of amorphous MgNix (x 1) alloys prepared by mechanical alloying. J Alloys Compd 1999;293:552e5. [11] Li X, Elkedim O, Nowak M, Jurczyk M. Characterization and first principle study of ball milled TieNi with Mg doping as hydrogen storage alloy. Int J Hydrogen Energy 2014;39:9735e43. [12] Luan B, Cut N, Liu H, Zhao H, Dou S. Low-temperature surface micro-encapsulation of Ti2Ni hydrogen-storage alloy powders. J Power Sources 1994;52:295e9. [13] Balcerzak M, Nowak M, Jakubowicz J, Jurczyk M. Electrochemical behavior of nanocrystalline TiNi doped by MWCNTs and Pd. Renew Energy 2014;62:432e8. [14] Zheng Q, Pivak Y, Mooij LPA, van der Eerden AMJ, Schreuders H, de Jongh PE, et al. EXAFS investigation of the destabilization of the MgeNieTi (H) system. Int J Hydrogen Energy 2012;37:4161e9. lis L. Mechanically alloyed MgeNieTi [15] Meyer M, Mendoza-Ze and MgeFeeTi powders as hydrogen storage materials. Int J Hydrogen Energy 2012;37:14864e9. L. Comparative study on the [16] Rousselot S, Guay D, Roue structure and electrochemical hydriding properties of MgTi, Mg0.5Ni0.5Ti and MgTi0.5Ni0.5 alloys prepared by high energy ball milling. J Power Sources 2011;196:1561e8. L, Huot J, Schulz R, Aymard L, Tarascon J-M. [17] Ruggeri S, Roue Properties of mechanically alloyed MgeNieTi ternary hydrogen storage alloys for Ni-MH batteries. J Power Sources 2002;112:547e56. [18] Anik M. Effect of titanium additive element on the discharging behavior of MgNi alloy electrode. Int J Hydrogen Energy 2011;36:15075e80. [19] Zhang Y, Zhang S-K, Chen L-X, Lei Y-Q, Wang Q-D. The study on the electrochemical performance of mechanically alloyed MgeTieNi-based ternary and quaternary hydrogen storage electrode alloys. Int J Hydrogen Energy 2001;26:801e6. L. Structure and [20] Rousselot S, Bichat M-P, Guay D, Roue electrochemical behaviour of metastable Mg50Ti50 alloy prepared by ball milling. J Power Sources 2008;175:621e4.
[21] Hida M, Asai K, Takemoto Y, Sakakibara A. Solid solubility and transformation in mechanically alloyed TieMg powders. Materials Transactions, JIM 1996;37:1679e83. [22] Asano K, Enoki H, Akiba E. Synthesis of HCP, FCC and BCC structure alloys in the MgeTi binary system by means of ball milling. J Alloys Compd 2009;480:558e63. [23] Kalisvaart W, Wondergem H, Bakker F, Notten P. MgeTi based materials for electrochemical hydrogen storage. J Mater Res 2007;22:1640e9. [24] Asano K, Enoki H, Akiba E. Synthesis process of MgeTi BCC alloys by means of ball milling. J Alloys Compd 2009;486:115e23. [25] Kubaschewski O, Alcock CB. Metallurgical thermochemistry. In: International series on materials science and technology. 5th ed.vol. 24. Oxford: Pergamon Press; 1979. revised and enlarged. [26] Takeshita HT, Tanaka H, Kiyobayashi T, Takeichi N, Kuriyama N. Hydrogenation characteristics of Ti2Ni and Ti4Ni2X (X ¼ O, N, C). J Alloys Compd 2002;330e332:517e21. [27] Anik M. Electrochemical hydrogen storage capacities of Mg2Ni and MgNi alloys synthesized by mechanical alloying. J Alloys Compd 2010;491:565e70. [28] Orimo S-i, Fujii H. Hydriding properties of nanostructured Mg1ex at.% Ni (x ¼ 33-50) with a different amount of amorphous MgNi. Int J Hydrogen Energy 1999;24:933e7. [29] Vijay R, Sundaresan R, Maiya M, Murthy SS. Comparative evaluation of MgeNi hydrogen absorbing materials prepared by mechanical alloying. Int J Hydrogen Energy 2005;30:501e8. n D, Ordon ~ ez S, Serafini D, Rojas PA, Aguilar C, [30] Guzma Santander M. Thermal stability of amorphous Mg50Ni50 alloy produced by mechanical alloying. J Non-Cryst Solids 2010;356:120e3. [31] Guo ZP, Huang ZG, Konstantinov K, Liu HK, Dou SX. Electrochemical hydrogen storage properties of nonstoichiometric amorphous -carbon composites (-0.3). Int J Hydrogen Energy 2006;31:2032e9. [32] Ma T, Hatano Y, Abe T, Watanabe K. Effects of bulk modification by Pd on electrochemical properties of MgNi. J Alloys Compd 2005;391:313e7. [33] Huang L, Elkedim O, Nowak M, Chassagnon R, Jurczyk M. Mg2-xTixNi (x ¼ 0, 0.5) alloys prepared by mechanical alloying for electrochemical hydrogen storage: experiments and firstprinciples calculations. Int J Hydrogen Energy 2012;37:14248e56. [34] Balcerzak M, Jakubowicz J, Kachlicki T, Jurczyk M. Effect of multi-walled carbon nanotubes and palladium addition on the microstructural and electrochemical properties of the nanocrystalline Ti2Ni alloy. Int J Hydrogen Energy 2015;40:3288e99. [35] Li X, Elkedim O, Nowak M, Jurczyk M, Chassagnon R. Structural characterization and electrochemical hydrogen storage properties of Ti2-xZrxNi (x ¼ 0, 0.1, 0.2) alloys prepared by mechanical alloying. Int J Hydrogen Energy 2013;38:12126e32.
Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling Z. Zhang a,*, O. Elkedim a, M. Balcerzak b, M. Jurczyk b FEMTO-ST, MN2S, Universite de Technologie de Belfort-Montbeliard, Site de Sevenans, 90010 Belfort Cedex, France Institute of Materials Science and Engineering, Poznan University of Technology, Sklodowska-Curie 5 Sq., 60-965 Poznan, Poland
a
b
article info
abstract
Article history:
A series of MgeTieNi alloys with different Ni content was prepared by mechanical alloying
Received 28 September 2015
using a planetary high-energy ball mill. The structural transformation was characterized
Received in revised form
by XRD and SEM. It indicated that the addition of 10 at% Ni did not hinder the amorph-
26 November 2015
ization of MgTi BCC phase. After 40 h of milling, the alloys of MgTiNi and MgTiNi2 mainly
Accepted 30 November 2015
consisted of the mixture of TiNi þ MgNi þ Mg2Ni and TiNi þ TiNi3 þ MgNix (x > 1),
Available online xxx
respectively. Their discharge capacities were investigated by electrochemical measurements at galvanostatic conditions. It was shown that all of the studied samples possessed
Keywords:
good cycling performance. Among them, MgTiNi showed the highest discharge capacity. In
MgeTieNi alloys
the ternary alloys with the same atom ratio of Ti and Mg, the number of activation cycles
Mechanical alloying
decreased with increasing the Ni content.
Phase structure
Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Electrochemical hydrogen storage properties
Introduction As energy crisis and environmental pollution are becoming more and more serious, hydrogen is widely regarded as the most promising energy resource to replace the traditional fossil fuel. An efficient and safe hydrogen storage material with small volume and low cost is the key issue for the utilization of hydrogen as an energy carrier. Besides, another notable application area for hydrogen storage material is in Ni-MH batteries. In recent years, a lot of high performance hydrogen storage alloys have been developed. Among them, magnesium based and titanium based alloys have attracted
much attention due to their extremely high discharge capacity and low cost [1e5]. Theoretically, the hydrogen storage abilities of MgH2 and TiH2 are 7.6 wt.% and 4.0 wt.% respectively. However, due to the irreversible hydrogen absorb reaction of TiH2 at room temperature and the low decompositionformation kinetics of MgH2, neither of them can be practically applied. In order to improve their hydrogen storage properties, transition metal elements as electrocatalysts have been introduced into the system. For example, amorphous MgNi which was measured in the electrochemical reaction possessed a high electrochemical capacity reaching 500 mAh/ g, but with a bad cycling performance [6,7]. The crystalline Mg2Ni can reversibly absorb and desorb hydrogen at high
* Corresponding author. Tel.: þ33 753341514; fax: þ33 384583000. E-mail addresses: [email protected], [email protected] (Z. Zhang). http://dx.doi.org/10.1016/j.ijhydene.2015.11.168 0360-3199/Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
2
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 x x x ( 2 0 1 5 ) 1 e6
temperature. However, the poor kinetics and bad antioxidation of Mg2Ni can not be ignored [8e10]. TiNi phase was found to possess a capacity of 230 mAh/g and good corrosion resistance in alkaline solution [11e13]. Unfortunately, the poor absorption/desorption kinetics and complex activation procedure become the obstacles for the practical application of TiNi alloys as hydrogen storage material. In order to investigate the synergistic effect of different phases on the enhancement of electrochemical performance, a lot of researchers have made efforts to optimize the composition for exploring the balance of discharge capacity and cycling performance [11,14e19]. In this work, the influence of Ni content on the structure and electrochemical performance of TieMgeNi alloys has been investigated. The obtained results are expected to provide more data about structural transformation and electrochemical properties of TieMgeNi ternary system, and to contribute to the design of new hydride compounds.
Experimental methodology The samples were prepared by Mechanical Alloying (MA) from the pure elemental powders of Mg (purity 99.8%, particle size 50 mm, GoodFellow), Ti (purity 99.5%, particle size 150 mm, GoodFellow), Ni (purity 99.5%, particle size 250 mm, GoodFellow) according to the following stoichiometries: MgTiNix (x ¼ 0.1, 0.5, 1, 2). The nominal composition of mixtures were introduced in cylindrical stainless steel vials (capacity 50 ml) with two stainless steel balls (diameter 20 mm) under a high pure argon atmosphere. The MA was performed for 20 h and 40 h at room temperature using a planetary high-energy ball mill (Retsch PM 400) at the speed of 400 RPM. The ball-to-powder weight ratio (BPR) was 10:1. In order to dissipate the heat and reduce the cold welding of powder particles during high-energy milling, the milling was interrupted for 30 min after every 1 h of operation. And after every 10 h of milling, the ball milling was stopped for scraping the powder adhered to the milling chamber walls and grinding balls. All of these operations were carried out under protective atmosphere (Ar). To investigate the transformation of phase structure, the samples with different compositions after different milling time were characterized by X-ray diffraction with Co Ka radiation (l ¼ 1.789 A). The morphology of the samples were determined using a scanning electron microscope (SEM, JEOL, SM-5800LV). The electrochemical charge and discharge measurements were performed using Muti-channel Battery Interface ATLAS 0461. The negative electrodes (working electrodes) materials consisted of as-milled powder and carbonated nickel with the ratio of 90:10. The mixtures were compressed between two nickel nets into a small pellet. NiOOH/Ni(OH)2 and Hg/HgO were employed as counter electrode and reference electrode respectively. All of the electrodes were soaked in 6 M KOH solution which was prepared from KOH flakes and distilled water. Before testing, in order to eliminate adverse oxidative effects (activation process), the electrodes were soaked in the electrolyte for 24 h at room temperature and 1 h at 100 C. The current density for charging and discharging were 40 mA/g.
The cut-off potential was 0.7 V vs. the reference electrode. Cycle stabilities were evaluated by capacity retaining rate after 18th cycle: Rh ¼ (C18/Cmax) 100%, where C18 and Cmax were discharge capacities at the 18th cycle and maximum discharge capacity, respectively.
Results and discussion Microstructure analysis A series of overlapped X-ray diffraction patterns of different composition samples with fixed milling times of 40 h are shown in Fig. 1. Apparently, it can be found that, accompanying with the increasing of Ni element content (from bottom to top), the main peaks of the XRD pattern moved towards the higher angles gradually. In the XRD pattern of MgTiNi0.1 (Fig. 1(a)), a broad peak was centered at 2q ¼ 44 , which is typical for amorphous phase. Substantially, in the case of MgTi alloys, a BCC phase at around 38 (characterized by XRD with Cu Ka radiation) has been reported several times [20e22]. To the best of our knowledge, this is the first time when a BCC phase was observed in Ni doped MgTi system [23]. In our previous work, MgTi BCC phase had been obtained in MgTi binary alloys after 60 h of milling (not shown) which means, after adding slight amount of Ni, the milling duration of MgTi amorphous is reduced to 40 h. This fact implies that the addition of 10 at% Ni did not seem to hinder the amorphization of MgTi alloys but may rather promote the grinding efficiency. In Fig. 1(b), the XRD pattern of MgTiNi0.5 after 40 h milling is shown. In order to analyze the structural transformation, the XRD analysis of this composition with respect to the different milling time is shown in Fig. 2. After 20 h of milling, the peaks of Ti, Mg and Ni can still be identified as elemental states, despite the fact that the latter two peaks inclined slightly to the Ti peak centered direction (2q ¼ 47 ). After further milling for 20 h, the diffraction peaks of Ni disappeared whereas the peaks of Mg were present. The main broad peak with two shoulders was observed which was assumed to be the mixture
Fig. 1 e Overlapped XRD patterns of samples milled for 40 h: (a) MgTiNi0.1, (b) MgTiNi0.5, (c) MgTiNi and (d) MgTiNi2.
Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
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 x x x ( 2 0 1 5 ) 1 e6
3
Fig. 2 e XRD patterns of MgTiNi0.5 mixture after different milling time: (a) 20 h, (b) 40 h.
Fig. 3 e XRD patterns of MgTiNi mixture after different milling time: (a) 20 h, (b) 40 h.
of TiNi-based phase and Ti. After comparing to Fig. 2(b), the tendency of amorphization of Ti and Mg became more apparent indicating that a stationary state of this composition might be attained by prolonging milling. It is certain that the elemental Mg and Ti can be detected after 40 h of milling based on the visualized comparison in Fig. 2. The diffraction peak of Mg maintained relatively sharp while that of Ti significantly broadened. The same phenomenon had also been reported, which is related to the investigation of milling production of pure Ti and Pure Mg [24]. According to the index results of phase, the broad feature of Ti centered peak is attributed to the presence of the overlapped diffraction peaks of Ti2Ni at 47.8 . It is notable that both of Ti and Mg were abundant in this composition, but only TiNibased phase was detected after milling. The formation of Ti2Ni instead of Mg2Ni may be due to two following reasons. Firstly, the atomic radii of Mg and Ti are 0.160 nm and 0.147 nm, respectively, which means Ti possesses a relatively higher diffusivity than Mg. Secondly, the enthalpy of formation of Mg2Ni and Ti2Ni are 51.9 kJ/mol [25] and 71.4 kJ/mol [26] respectively. All these evidences indicate that, the combination of Ti and Ni in atomic scale is easier than that of Mg and Ni. Besides, according to the phase diagram of TiNi system, Ti abundant is more favorable for the formation of Ti2Ni phase than that of other phases. As a result, the phase structure of MgTiNi0.5 after 40 h milling consisted of Ti2Ni, Mg and Ti. The XRD patterns of MgTiNi with different milling time (40 h and 20 h) are presented in Fig. 3. The typical amorphous phase at 49 was found in both of Fig. 3(a) and (b). Particularly, the diffraction pattern shown in Fig. 3(b) had been observed in other references [16,27,28] concerning with the products of the MgNi by mechanical alloying. The broad peak belonged to the MgNi amorphous phase, and the remaining sharp peak was related to the nickel residue. It is indicated that, in this case, 20 h of milling is not enough for Ni to neither completely dissolve into the MgNi amorphous phase nor form TiNi crystalline phase. Therefore, the diffraction peaks of titanium are
considered to be superimposed in the broad feature. After 40 h of milling, the diffraction peaks of MgNi and TiNi were detected. The main characteristic peak increased in intensity and became sharper while the peaks of Ni residue disappeared. All of these changes suggest that crystallization was promoted in this milling duration. Additionally, Mg2Ni was supposed to be formed during 40 h of milling which achieved by following analysis. As seen in Fig. 3(b), there was no peak of MgTi phase detected which means only the TieNi and MgeNi binary phase can be synthesized. While the observed phases, so far, were mainly binary phase with equal atom ratio. According to the stoichiometric proportion, these may lead to two possible assumptions. One is a high degree of excess dispersion of magnesium or/and titanium, while the other one is the existence of another Mg-rich or/and Ti-rich phase. As no protruded elemental peak was observed in the XRD pattern, the formation of nanocrystals Mg2Ni and Ti2Ni cannot be excluded. In fact, the recrystallization of MgNi amorphous phase during prolongation milling was report elsewhere [7,29]. The crystallizations of MgNi alloy into a mixture of nanocrystals Mg2Ni and MgNi2 took place after a certain time of milling which may differ from the various mill mode. In n et al. [30] sugaddition, an observation reported by Guzma gested that Mg2Ni and MgNi2 nanocrystals detected by highresolution microscopy bedded on amorphous MgNi may also lead to the absence of MgeNi phase in the diffraction pattern. And it is considered that the formation of Mg2Ni is reasonable compared with that of MgNi2 in such a superstoichiometric magnesium environment. Fig. 4 shows the XRD patterns of MgTiNi2 with respect to different milling time. It is obvious to see the phase transformation in this duration. The broad peak centered at 49.8 was observed in both of patterns which also appeared in Ref. [17]. Authors of this work obtained a broad peak which consisted of TiNi and MgNi amorphous phases after milling MgTiNi2 alloys. However, our results may differ from them. In our case, after comparing with the XRD of Fig. 3, the peak of MgNi amorphous phase in MgTiNi and MgTiNi2 were not the
Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
4
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 x x x ( 2 0 1 5 ) 1 e6
was characterized clearly which was coexisted with TiNi and MgNix (x > 1) amorphous in the 40 h as-milled sample. The SEM images of different samples after 40 h of milling are shown in Fig. 5. It can be found that, all of the morphology of samples consisted of micrometric particles. In comparison with the different micrographs, with increasing the Ni content, the particles were getting more irregular which had a tendency to agglomerate. Accompanying the increasing of the Ni content, more fine particles smaller than 5 mm and agglomerated particles bigger than 20 mm were found. This reflects that, repeated fracturing and the cold-welding were occurring together during the ball milling process. Ni as a positive factor which can improve the milling efficiency of the alloys which is consistent with our previous analysis of XRD.
Electrochemical tests Fig. 4 e XRD patterns of MgTiNi2 mixture after different milling time: (a) 20 h, (b) 40 h.
same which were centered in 49 and 49.8 , respectively. Guo et al. [31] investigated the effect of Ni content on the structural transformation of MgNi amorphous phase. In our observations, the peak of MgNi amorphous phase slightly shifted to higher angle with adding Ni, which is pronounced for Guo's result. This fact implies that the MgNix (x > 1) amorphous phase was formed in MgTiNi2. Moreover, peaks of TiNi were found in both of patterns. The shoulder peak obtained in Fig. 4(a) which was corresponding to the dissolving Ni disappeared after continually 20 h of milling. Instead, TiNi3 phase
The discharge capacities of as-milled samples according to the cycle numbers are shown in Fig. 5 respectively. Most of the samples achieved their maximum capacities after several cycles of activation except MgTiNi0.1, whose hydrogen discharge capacity was very low, even lower than 5 mAh/g. It was reported that the hydrogen discharge capacity of ballmilled binary MgTi alloys was close to zero [20]. By adding 10 at% of Ni, the electrochemical activity of MgTi alloy was not improved. The dispersion of Ni in MgTi BCC phase can not motivate the chargeetransfer reaction. The capacity retaining rate (Rh) of different samples is shown in Table 1. Among all of the electrochemical studied materials, MgTiNi possessed the best discharge capacity of 154 mAh/g in the second charge/discharge cycle. The good
Fig. 5 e SEM micrographs of samples milled for 40 h: (a) MgTiNi0.1, (b) MgTiNi0.5, (c) MgTiNi, and (d) MgTiNi2. Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
5
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 x x x ( 2 0 1 5 ) 1 e6
Table 1 e Phase structure and discharge capacities of 40 h as-milled samples. Composition Main phase Cmax (mAh/g) C18 (mAh/g) Retaining rate (R18)
MgTiNi0.5
MgTiNi
MgTiNi2
Ti2Ni, Ti and Mg 81 67 82%
MgNi, TiNi, Mg2Ni 154 123 80%
TiNi3, MgNix (x > 1), TiNi 93 86 92%
electrochemical performance is considered to attribute to the formation of MgNi amorphous phase. MgNi amorphous phase possessed a high discharge capacity but also decreased rapidly with cycling [32]. Unlike that, the average capacity decay of MgTiNi was slow as 1.1%, which is due to the coexisted TiNi and Mg2Ni. TiNi possesses good corrosion resistance in alkaline solution. Meanwhile, the synergistic effects of TiNi and Mg2Ni MA alloys on the electrochemical performance was also reported by Huang et al. [33]. In contrast to MgTiNi alloy, as found in Table 1, MgTiNi2 alloy which also consisted of TiNi and MgNi amorphous phases possessed a lower discharge capacity but better cycling performance comparing with MgTiNi. This is due to the increasing content of Ni. Firstly, TiNi3, as one of the main phase achieved in the as-milled powder is an inactive phase which can not store hydrogen [17]. Secondly, the features of MgNix (x > 1) amorphous and MgNi amorphous phases are different. Zhang et al. [10] reported the characteristics of amorphous MgNi1.5 and MgNi2 whose discharge capacity was getting lower but the circulation property was getting better gradually, accompanying with the increasing of the Ni addition. This is in agreement with our result. Therefore, it can be concluded that, the phases of MgNix (x > 1) and TiNi3 can be formed by adding Ni, resulting in the improvement of the cycling performance at the expense of reduction of the discharge capacity (see Fig. 6). In Table 1, MgTiNi0.5 possessed the worst discharge capacity which is not surprising considering the existence of big amount of unreacted Ti and Mg. The hydrogen storage phase in this phase is Ti2Ni. It is certain that the electrochemical property of MgTiNi0.5 could be improved by prolongation of milling time. Additionally, it is noted that, there was a large range of difference between the first two discharge capacities
of MgTiNi0.5, which means it needs several cycles of activation. This phenomenon is a typical characteristic of Ti2Ni alloys which was reported elsewhere [34,35]. In comparison of other samples, it is found that the number of activation cycles decreased with increasing the Ni content in the ternary MgeTieNi alloys, which is in agreement with that obtained in Ref. [17].
Conclusion In this study, MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys was prepared by mechanical alloying using a planetary high-energy ball mill. It is found that: - Adding 10 at% of Ni in MgTi alloys did not hinder the amorphization of MgTi BCC phase, nor improve the discharge capacity of MgTi alloys. - Based on the XRD analysis, the alloys of MgTiNi and MgTiNi2 mainly consisted of the mixture of TiNi þ MgNi þ Mg2Ni and TiNi þ TiNi3 þ MgNix (x > 1), respectively. The average discharge capacity decay of studied samples were all less than 1.1% per cycle. Among them, MgTiNi showed the highest discharge capacity which is attributed to the MgNi amorphous phase. - In the ternary alloys with the same atom ratio of Ti and Mg, the number of activation cycles decreased with increasing the Ni content.
Acknowledgments This research is supported by China Scholarship Council (CSC) and UTBM in the framework of UT-INSA project. We would like to thank M. Bruno Nicolas and M. Yangzhou Ma (LERMPS, UTBM, France) for their assistance with the SEM and XRD measurements.
references
Fig. 6 e The discharge capacities of 40 h as-milled samples.
[1] Sakintuna B, Lamari-Darkrim F, Hirscher M. Metal hydride materials for solid hydrogen storage: a review. Int J Hydrogen Energy 2007;32:1121e40. [2] Schlapbach L, Zu¨ttel A. Hydrogen-storage materials for mobile applications. Nature 2001;414:353e8. [3] Jain I, Lal C, Jain A. Hydrogen storage in Mg: a most promising material. Int J Hydrogen Energy 2010;35:5133e44. [4] Tliha M, Khaldi C, Boussami S, Fenineche N, El-Kedim O, Mathlouthi H, et al. Kinetic and thermodynamic studies of hydrogen storage alloys as negative electrode materials for Ni/MH batteries: a review. J Solid State Electrochem 2014;18:577e93.
Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168
6
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 x x x ( 2 0 1 5 ) 1 e6
[5] Shang C, Bououdina M, Song Y, Guo Z. Mechanical alloying and electronic simulations of (MgH2þM) systems (M ¼ Al, Ti, Fe, Ni, Cu and Nb) for hydrogen storage. Int J Hydrogen Energy 2004;29:73e80. [6] Iwakura C, Nohara S, Inoue H, Fukumoto Y. Surface modification of MgNi alloy with graphite by ball-milling for use in nickel-metal hydride batteries. Chem Commun 1996:1831e2. [7] Ruggeri S, Lenain C, Roue L, Liang G, Huot J, Schulz R. Mechanically driven crystallization of amorphous MgNi alloy during prolonged milling: applications in Ni-MH batteries. J Alloys Compd 2002;339:195e201. [8] Inoue H, Ueda T, Nohara S, Fujita N, Iwakura C. Effect of ballmilling on electrochemical and physicochemical characteristics of crystalline Mg2Ni alloy. Electrochim acta 1998;43:2215e9. [9] Ueda TT, Tsukahara M, Kamiya Y, Kikuchi S. Preparation and hydrogen storage properties of MgeNieMg2Ni laminate composites. J Alloys Compd 2005;386:253e7. [10] Zhang SG, Hara Y, Morikawa T, Inoue H, Iwakura C. Electrochemical and structural characteristics of amorphous MgNix (x 1) alloys prepared by mechanical alloying. J Alloys Compd 1999;293:552e5. [11] Li X, Elkedim O, Nowak M, Jurczyk M. Characterization and first principle study of ball milled TieNi with Mg doping as hydrogen storage alloy. Int J Hydrogen Energy 2014;39:9735e43. [12] Luan B, Cut N, Liu H, Zhao H, Dou S. Low-temperature surface micro-encapsulation of Ti2Ni hydrogen-storage alloy powders. J Power Sources 1994;52:295e9. [13] Balcerzak M, Nowak M, Jakubowicz J, Jurczyk M. Electrochemical behavior of nanocrystalline TiNi doped by MWCNTs and Pd. Renew Energy 2014;62:432e8. [14] Zheng Q, Pivak Y, Mooij LPA, van der Eerden AMJ, Schreuders H, de Jongh PE, et al. EXAFS investigation of the destabilization of the MgeNieTi (H) system. Int J Hydrogen Energy 2012;37:4161e9. lis L. Mechanically alloyed MgeNieTi [15] Meyer M, Mendoza-Ze and MgeFeeTi powders as hydrogen storage materials. Int J Hydrogen Energy 2012;37:14864e9. L. Comparative study on the [16] Rousselot S, Guay D, Roue structure and electrochemical hydriding properties of MgTi, Mg0.5Ni0.5Ti and MgTi0.5Ni0.5 alloys prepared by high energy ball milling. J Power Sources 2011;196:1561e8. L, Huot J, Schulz R, Aymard L, Tarascon J-M. [17] Ruggeri S, Roue Properties of mechanically alloyed MgeNieTi ternary hydrogen storage alloys for Ni-MH batteries. J Power Sources 2002;112:547e56. [18] Anik M. Effect of titanium additive element on the discharging behavior of MgNi alloy electrode. Int J Hydrogen Energy 2011;36:15075e80. [19] Zhang Y, Zhang S-K, Chen L-X, Lei Y-Q, Wang Q-D. The study on the electrochemical performance of mechanically alloyed MgeTieNi-based ternary and quaternary hydrogen storage electrode alloys. Int J Hydrogen Energy 2001;26:801e6. L. Structure and [20] Rousselot S, Bichat M-P, Guay D, Roue electrochemical behaviour of metastable Mg50Ti50 alloy prepared by ball milling. J Power Sources 2008;175:621e4.
[21] Hida M, Asai K, Takemoto Y, Sakakibara A. Solid solubility and transformation in mechanically alloyed TieMg powders. Materials Transactions, JIM 1996;37:1679e83. [22] Asano K, Enoki H, Akiba E. Synthesis of HCP, FCC and BCC structure alloys in the MgeTi binary system by means of ball milling. J Alloys Compd 2009;480:558e63. [23] Kalisvaart W, Wondergem H, Bakker F, Notten P. MgeTi based materials for electrochemical hydrogen storage. J Mater Res 2007;22:1640e9. [24] Asano K, Enoki H, Akiba E. Synthesis process of MgeTi BCC alloys by means of ball milling. J Alloys Compd 2009;486:115e23. [25] Kubaschewski O, Alcock CB. Metallurgical thermochemistry. In: International series on materials science and technology. 5th ed.vol. 24. Oxford: Pergamon Press; 1979. revised and enlarged. [26] Takeshita HT, Tanaka H, Kiyobayashi T, Takeichi N, Kuriyama N. Hydrogenation characteristics of Ti2Ni and Ti4Ni2X (X ¼ O, N, C). J Alloys Compd 2002;330e332:517e21. [27] Anik M. Electrochemical hydrogen storage capacities of Mg2Ni and MgNi alloys synthesized by mechanical alloying. J Alloys Compd 2010;491:565e70. [28] Orimo S-i, Fujii H. Hydriding properties of nanostructured Mg1ex at.% Ni (x ¼ 33-50) with a different amount of amorphous MgNi. Int J Hydrogen Energy 1999;24:933e7. [29] Vijay R, Sundaresan R, Maiya M, Murthy SS. Comparative evaluation of MgeNi hydrogen absorbing materials prepared by mechanical alloying. Int J Hydrogen Energy 2005;30:501e8. n D, Ordon ~ ez S, Serafini D, Rojas PA, Aguilar C, [30] Guzma Santander M. Thermal stability of amorphous Mg50Ni50 alloy produced by mechanical alloying. J Non-Cryst Solids 2010;356:120e3. [31] Guo ZP, Huang ZG, Konstantinov K, Liu HK, Dou SX. Electrochemical hydrogen storage properties of nonstoichiometric amorphous -carbon composites (-0.3). Int J Hydrogen Energy 2006;31:2032e9. [32] Ma T, Hatano Y, Abe T, Watanabe K. Effects of bulk modification by Pd on electrochemical properties of MgNi. J Alloys Compd 2005;391:313e7. [33] Huang L, Elkedim O, Nowak M, Chassagnon R, Jurczyk M. Mg2-xTixNi (x ¼ 0, 0.5) alloys prepared by mechanical alloying for electrochemical hydrogen storage: experiments and firstprinciples calculations. Int J Hydrogen Energy 2012;37:14248e56. [34] Balcerzak M, Jakubowicz J, Kachlicki T, Jurczyk M. Effect of multi-walled carbon nanotubes and palladium addition on the microstructural and electrochemical properties of the nanocrystalline Ti2Ni alloy. Int J Hydrogen Energy 2015;40:3288e99. [35] Li X, Elkedim O, Nowak M, Jurczyk M, Chassagnon R. Structural characterization and electrochemical hydrogen storage properties of Ti2-xZrxNi (x ¼ 0, 0.1, 0.2) alloys prepared by mechanical alloying. Int J Hydrogen Energy 2013;38:12126e32.
Please cite this article in press as: Zhang Z, et al., Structural and electrochemical hydrogen storage properties of MgTiNix (x ¼ 0.1, 0.5, 1, 2) alloys prepared by ball milling, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.11.168