Journal of Alloys and Compounds 481 (2009) 761–763
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jallcom
Study on hydrogen storage properties of LiAlH4 Zheng Xueping ∗ , Liu Shenglin School of Materials Science and Engineering, Chang’an University, Xi’an 710061, China
a r t i c l e
i n f o
Article history: Received 9 January 2009 Received in revised form 5 March 2009 Accepted 14 March 2009 Available online 25 March 2009 Keywords: LiAlH4 Additives Hydrogen release
a b s t r a c t LiAlH4 doped with 5 mol% Ti, Ni, Ce(SO4 )2 and LaCl3 additives has been studied by pressure–composition– temperature (PCT) experiment and X-ray diffraction. Doping with Ce(SO4 )2 induced the largest decrease in the temperature of hydrogen release about 38 ◦ C, signiﬁcantly, the amount of hydrogen release had only a slight increase. Comparatively, doping with Ti and LaCl3 decreased the temperature of hydrogen release, and markedly decreased the amount of hydrogen release. The study on rehydrogenation showed that doping with Ti caused the most amount of rehydrogenation, however, doping with Ce(SO4 )2 induced the least adsorption amount. The result obtained by XRD spectra for LiAlH4 doped with 5 mol% Ti heated at 270 ◦ C and adsorped at 180 ◦ C under ∼8 MPa hydrogen pressure indicated slow composition of LiH, Al and H2 into Li3 AlH6 . © 2009 Elsevier B.V. All rights reserved.
1. Introduction The introduction of hydrogen as a major global energy carrier depends on solving a number of fundamental and technical problems. Perhaps, the largest challenge today is to ﬁnd hydrogen storage solutions that at the same time are safe, compact, light, reversible, and cheap . Hydrogen storage in metal materials is considered as one of the most attractive methods, however, their reversible gravimetric hydrogen capacity is limited to 2.5 wt.%, signiﬁcantly below the goal of, for example, ≥5 wt.% hydrogen set recently by the International Energy Agency . A breakthrough for the possible use of alanates as a hydrogen storage material was achieved by the successful catalysis of NaAlH4 with Ti-compounds giving reversible hydrogenation, enhanced kinetics and lowered desorption temperatures . This ﬁnding has led to a considerable attention to complex hydrides based on Al as promising reversible hydrogen storage media. Comparatively, the gravimetric storage capacity for LiAlH4 is higher than for NaAlH4 , and the desorption of hydrogen is at lower temperatures for the ﬁrst two reactions [4,5]. 3LiAlH4 → Li3 AlH6 + 2Al + 3H2 (5.3 wt.%H2 )
Li3 AlH6 → 3LiH + Al + 3/2H2 (2.6 wt.%H2 )
3LiH + 3Al → 3LiAl + 3/2H2 (2.6 wt.%H2 )
The decomposition temperatures are reported to be between 150–175 ◦ C (Eq. (1)) and 180–220 ◦ C (Eq. (2)) [6,7]. In recent studies, the focus has been on LiAlH4 with ball-milling process and adding catalysts. Research has established that adding catalysts can
∗ Corresponding author. Tel.: +86 29 82337348. E-mail address: [email protected]
(Z. Xueping). 0925-8388/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2009.03.089
have a distinct effect on the decomposition of LiAlH4 . The principal catalysts studied were elemental titanium, TiCl4 , TiCl3 , AlCl3 , FeCl3 , elemental iron, elemental nickel, vanadium and carbon black [8–10]. Resan  reported that the addition of TiCl3 and TiCl4 to LiAlH4 eliminated the ﬁrst step of hydrogen evolution and signiﬁcantly lowered the decomposition temperature of the second step. Doping with elemental iron caused only a slight decrease in the amount of hydrogen released and did not eliminate the ﬁrst step of hydrogen evolution. The study by Blanchard indicated that ball-milling of LiAlD4 and VCl3 or TiCl3 ·1/3AlCl3 reduced the thermal decomposition temperatures for the ﬁrst step by 50–60 ◦ C [2,11]. The current paper focuses mainly on LiAlH4 with Ti, Ni, Ce(SO4 )2 and LaCl3 . Its purpose is to show their effects on the capacity of hydrogen release in the two-step decomposition reaction and the rehydrogenation properties. 2. Experimental LiAlH4 (96 wt.% pure) was purchased from Tianjin Beidouxing Fine Chemical Co., Ltd. The nickel powder, Ti, Ce(SO4 )2 ·4H2 O and LaCl3 ·7H2 O were obtained from the Central Iron and Steel Research Institute. In this study, LiAlH4 , Ti and Ni were used as-received with no additional puriﬁcation. Due to a mass of crystal water in Ce(SO4 )2 ·4H2 O and LaCl3 ·7H2 O, dehydration was carried out before LiAlH4 was mixed with Ce(SO4 )2 and LaCl3 in order to prevent the effect of the crystal water on LiAlH4 . Ce(SO4 )2 ·4H2 O and LaCl3 ·7H2 O were heated to 160 ◦ C and 150 ◦ C and then kept at this temperatures for 4 h and 3 h, respectively. All operations on the samples were done under dry argon atmosphere in a glove box to prevent reaction with moisture and oxygen in the air. LiAlH4 , usually 2 g, was mixed with 1 mol% Ni, Ti, Ce(SO4 )2 and LaCl3 by ball-milling for 20 min at a gyration rate of 200 rpm using a Spex mill. In addition, LiAlH4 was also milled for 20 min in order to calculate and allow for the effects of ball-milling on the results. Three hardened steel vials sealed under argon with 14 steel balls (1 g each) were used. Air-cooling of the vials was employed to prevent them from heating during the ball-milling process. The ballmilled samples were then transferred to 3-ml glass bottles in a glove box under dry argon atmosphere.
Z. Xueping, L. Shenglin / Journal of Alloys and Compounds 481 (2009) 761–763
Fig. 2. PCT curves of LiAlH4 and samples doped with 5 mol% Ce(SO4 )2 and LaCl3 . Fig. 1. Dehydrogenation curves of LiAlH4 and samples doped with 5 mol% Ni and Ti in the ﬁrst and second steps.
Hydrogen desorption experiments were carried out in a pressure–composition– temperature (PCT) apparatus. This can be operated up to 10 MPa and at 400 ◦ C. The pressure of hydrogen released in relation to volume was displayed by a pressure transducer. The experimental studies were done by a reactor. This consisted of two parts: heater and sample vessel. The former was used to connect with the pressure transducer and thermocouple. It had a 2.2 cm outside diameter (OD), 0.5 cm wall and 20 cm internal length. It was loaded with the sample vessel (1 cm OD, 0.1 cm wall and 5 cm internal length). The sample vessel was loaded with about 0.1 g of NaAlH4 . The reactor was heated with an air furnace. During heating, the hydrogen released overﬂowed from the sample vessel ﬁrstly into the heater and then into the transit pressure transducer. The value of hydrogen pressure can be clearly read. According to the computing formula, the weight percentage of release hydrogen is obtained. The samples doped with Ni, Ti, Ce(SO4 )2 and LaCl3 were heated under a vacuum atmosphere at a heating rate of 2 ◦ C/min to 250 ◦ C. During the heating process, all pressure and temperature data were acquired and then the ﬁrst and second step curves were drawn by special software. Rehydrogenation studies were carried out with LiAlH4 doped with 5 mol% LaCl3 , Ce(SO4 )2 , Ti and Ni. After the ﬁrst complete dehydrogenation (ﬁrst two reactions), the samples were kept at 180 ◦ C under ∼8 MPa hydrogen pressure for 2 h. The uptake of hydrogen for the samples is evident by the pressure decreasing with time in this closed system. The change of hydrogen pressure was recorded by the pressure transducer. The phase structure was determined by a MXP21VAHF X-ray diffractometer (XRD with Cu K␣ radiation, a tube voltage of 40 kV and a tube current of 200 mA). A small amount of sample was spread evenly in the sample holder. The samples were placed in a sealed container which was ﬁlled with argon before they were measured. The X-ray intensity was measured over diffraction angles from 10◦ to 90◦ with a velocity of 0.02◦ per step. XRD was used to analyze phase composition and to obtain the precise lattice parameters of samples.
3. Results and discussion In this study, the hydrogen release capacity of LiAlH4 doped with 5 mol% Ti, Ni, Ce(SO4 )2 and LaCl3 was studied ﬁrst by PCT experiments. Fig. 1 shows the thermal desorption results for LiAlH4 , LiAlH4 doped with 5 mol% Ti and Ni in the ﬁrst and second steps. It is very marked that doping signiﬁcantly decreased the amount of hydrogen release in the ﬁrst and second steps, especially in the ﬁrst step. However, the effects of dopants on the dehydrogenation temTable 1 Dehydrogenation amount (wt.%) and temperature (◦ C) of doped samples in the ﬁrst (R1) and second (R2) steps. Samples
R1 (◦ C)
R2 (◦ C)
LiAlH4 5 mol% Ce(SO4 )2 5 mol% LaCl3 5 mol% Ni 5 mol% Ti
4.0 4.2 3.6 3.3 3.5
2.0 1.9 1.5 1.8 1.6
148 110 125 145 135
176 175 170 184 175
6.0 6.1 5.1 5.1 5.1
perature showed signiﬁcant difference. Doping with Ti obviously decreased the temperature of hydrogen release, comparatively, doping with Ni caused only a slight decrease in the start decomposition temperature. According to Table 1, LiAlH4 began to decompose at 148 ◦ C and continued up to 176 ◦ C. In the ﬁrst step, about 4.0 wt.% hydrogen was released and in the second step about 2.0 wt.% hydrogen, so a total of about 6.0 wt.% hydrogen was released from the LiAlH4 . Doping with 5 mol% Ti markedly decreases the decomposition temperature. The decomposition reaction started at about 135 ◦ C and was completed at about 175 ◦ C. The amount of hydrogen release was 3.5 and 1.6 wt.% in the ﬁrst and second steps, respectively, a total of about 5.1 wt.% hydrogen. Doping with 5 mol% Ni only slightly decreased the start decomposition temperature in the ﬁrst step. Compared to LiAlH4 , the ﬁrst-step dehydrogenation temperature was only lower by about 3 ◦ C. However, the second dehydrogenation temperature was higher by about 5 ◦ C. The total amount was lower by about 0.9 wt.%. In addition, it is not difﬁcult to ﬁnd from the slopes of curves that doping presents signiﬁcant effects on the rate of hydrogen release. Compared to LiAlH4 in the as-received condition, doping decreased markedly the rate of hydrogen release in the ﬁrst step. Thus, the reason that doping with Ni caused increase in the second temperature step might be that the lower rate made hydrogen release time prolonged. Fig. 2 shows the thermal desorption results for LiAlH4 , LiAlH4 doped with 5 mol% Ce(SO4 )2 and LaCl3 . It is easy to be seen that doping with Ce(SO4 )2 induced a larger decrease in the temperature than doping with LaCl3 . Compared to LiAlH4 , the dehydrogenation temperature of doping with Ce(SO4 )2 was substantially lower by about 38 ◦ C in the ﬁrst step (Table 1). Whereas, the amount of hydrogen release had no signiﬁcant decrease, contrarily, it had a slight increase. Comparatively, doping with LaCl3 decreased the temperature of hydrogen release, simultaneously, the amount of hydrogen release presented an obvious decrease. It is same to doping with Ni and Ti that doping with Ce(SO4 )2 and LaCl3 induced an obvious decrease in the rate of hydrogen release. Fig. 3 compares the adsorption hydrogen of the samples doped with 5 mol% Ti, Ni, Ce(SO4 )2 and LaCl3 during the ﬁrst rehydrogenation cycle carried out in PCT at 180 ◦ C under 8 MPa after being discharged of hydrogen at 270 ◦ C. LiAlH4 doped with Ti presents Table 2 Adsorption amount of samples doped under 8 MPa and at 180 ◦ C. Samples
5 mol% LaCl3 5 mol% Ce(SO4 )2
5 mol% Ni 5 mol% Ti
Z. Xueping, L. Shenglin / Journal of Alloys and Compounds 481 (2009) 761–763
type apparatus. Directly after ball-milling there were no signs of any Ti-containing phases, and the unit-cell of LiAlD4 and Al gave no indication of any solid solutions. Hence it was concluded that the Ti was in an amorphous state directly after ball-milling. In addition, the study on NaAlH4 doped with TiF3 and TiCl3 additives found that there were no indications from diffraction studies of any solid solution of Ti in bulk, on the other hand, no Ti or Ti-compound was detected after ball-milling [12,13]. After some cycles, however, a crystalline phase interpreted as an Al1-xTix solid solution with x < 1/4 was observed, and TiF3 and TiCl3 additives reacted apparently differently. The result indicated that no crystalline NaF was found but broad reﬂection of NaCl was seen by powder X-ray diffraction (PXD) . Therefore, it is very important to study further the state of the additives in LiAlH4 and the reaction principle of the additives with LiAlH4 . Fig. 3. Adsorption curves of LiAlH4 doped with 5 mol% Ti, Ni, Ce(SO4 )2 and LaCl3 .
4. Conclusions Doping LiAlH4 with Ti, Ni and LaCl3 each resulted in a greatly decreased amount of desorbed hydrogen. Differently, doping with Ce(SO4 )2 caused a slight increase in the ﬁrst dehydrogenation. There was also a dramatic decrease in the temperature associated with the ﬁrst hydrogen release step. Doping with Ce(SO4 )2 induced the largest decrease in the temperature than doping with Ti, Ni and LaCl3 . The slopes of curves indicated that doping caused an obvious decrease in the rate of hydrogen release. In addition, the result obtained by rehydrogenation curves for LiAlH4 doped with 5 mol% Ti, Ni, LaCl3 and Ce(SO4 )2 indicated that the sample doped with Ti presents the most amount of hydrogen release. The XRD pattern of the sample rehydrogenated shows slow composition of LiH, Al and H2 into Li3 AlH6 . Fig. 4. XRD data for dehydrogenation (a) and rehydrogenation (b) of LiAlH4 doped with 5 mol% Ti.
the most absorption of hydrogen. According to Table 2, the most amount of rehydrogenation was 0.83 wt.%, the absorption amount of LiAlH4 doped with LaCl3 was 0.62 wt.%, while the absorption amount of LiAlH4 doped with Ce(SO4 )2 only had 0.12 wt.%. In the next study, X-ray diffraction (XRD) measurements were carried out in order to verify the dehydrogenation/rehydrogenation characteristics of 5 mol% Ti-doped LiAlH4 . Fig. 4 (curve a) shows XRD spectra for LiAlH4 doped with Ti heated at 270 ◦ C. It can be seen that the major phases presented in the XRD pattern of the decomposed sample were LiH and Al, with no noticeable LiAlH4 and Li3 AlH6 phases. This result indicated that the ﬁrst and second reactions went to completion. Fig. 4 (curve b) shows the XRD pattern of the sample rehydrogenated. The additional peaks of micro-crystalline Li3 AlH6 indicated slow composition of LiH, Al and H2 into Li3 AlH6 according to the following reaction: 3LiH + Al + 3/2H2 → Li3 AlH6
This result proved that the Li3 AlH6 was reversible. In addition, no Ti or the secondary Ti-containing phases were found from the XRD spectra for LiAlH4 doped with Ti. This result is similar to the samples with TiF3 additives reported by Brinks et al. . LiAlD4 samples with TiF3 additives have been investigated by synchrotron X-ray diffraction, neutron diffraction and a Sieverts-
Acknowledgement This research has received ﬁnancial support from the Project ﬁnancially supported by National Natural Science Foundation funded projects (50806007). References  O.M. Løvvik, J. Alloys Compd. 373 (2004) 28–32.  D. Blanchard, H.W. Brinks, B.C. Hauback, P. Norby, Mater. Sci. Eng. B 108 (2004) 54–59. ´ M. Schwickardi, J. Alloys Compd. 253–254 (1997) 1–9.  B. Bogdanovic,  H.W. Brinks, A. Fossdal, J.E. Fonneløp, B.C. Hauback, J. Alloys Compd. 397 (2005) 291–295.  H.W. Brinks, B.C. Hauback, P. Norby, H. Fjellvåg, J. Alloys Compd. 351 (2003) 222–227.  C. Iwakura, T. Oura, H. Inoue, M. Matsuoka, Y. Yamamoto, J. Electroanal. Chem. 398 (1995) 37–41.  K. Ikeda, S. Orimo, A. Züttel, L. Schlapbach, H. Fujii, J. Alloys Compd. 280 (1998) 279–283.  M. Resan, M.D. Hampton, J.K. Lomness, D.K. Slattery, Int. J. Hydrogen Energy 30 (2005) 1413–1416.  M. Fichtner, J. Engel, O. Fuhr, O. Kircher, O. Rubner, Mater. Sci. Eng. B 108 (2004) 42–47.  V.P. Balema, J.W. Wiench, K.W. Dennis, M. Pruski, V.K. Pecharsky, J. Alloys Compd. 329 (2001) 108–114.  D. Blanchard, A.I. Lem, S. Øvergaard, H.W. Brinks, B.C. Hauback, J. Alloys Compd. 458 (2008) 467–473.  H.W. Brinks, C.M. Jensen, S.S. Srinivasan, B.C. Hauback, D. Blanchard, K. Murphy, J. Alloys Compd. 376 (2004) 215–221.  C. Weidenthaler, A. Pommerin, M. Felderhoff, B. Bogdanovic, F. Schüth, Phys. Chem. Chem. Phys. 5 (2003) 5149–5153.