- Email: [email protected]
Reaction products between TiCl3 catalyst and Li3AlH6 during mechanical mixing
Journal of Alloys and Compounds 419 (2006) 176–179
Reaction products between TiCl3 catalyst and Li3AlH6 during mechanical mixing Jae-Hyeok Shim, Gil-Jae Lee, Young Whan Cho ∗ Nano-Materials Research Center, Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea Received 22 June 2005; accepted 7 July 2005 Available online 15 November 2005
Abstract The reaction between TiCl3 and Li3 AlH6 was investigated to understand reaction products in Ti-catalyzed Li3 AlH6 using mechanical milling and thermodynamic calculation. As reaction products, LiCl, TiAl3 and TiH2 are observed by X-ray powder diffraction, which is in agreement with the results of thermodynamic calculation. TiAl3 forms in an L12 structure rather than an equilibrium D022 structure during milling. It is predicted that the mole fraction of TiAl3 increases and that of TiH2 decreases as temperature increases during milling. Differential scanning calorimetry confirms that TiAl3 acts as a dehydrogenation catalyst in Li3 AlH6 . © 2005 Elsevier B.V. All rights reserved. Keywords: Energy storage materials; Mechanochemical synthesis; X-ray diffraction; Thermal analysis; Hydrogen storage materials; Catalyst
1. Introduction The development of solid-state hydrogen storage at low and medium temperatures has been recognized as a key technology for hydrogen fuel cell applications. Especially for vehicular applications, it is important to find and design new reversible light-weight hydrogen materials that exhibit high hydrogen capacity [1]. Complex metal alanates (aluminohydrides) have received great attention as hydrogen storage materials owing to their inherent high theoretical hydrogen capacity [2–5]. In 1997, Bogdanovi´c and Schwickardi [2] first demonstrated that reversible hydrogen storage could be achieved under moderate conditions (temperature and pressure) with accelerated kinetics in NaAlH4 and Na3 AlH6 by doping Ti-bearing catalysts such as TiCl3 and Ti(OBu)4 through wet chemistry. After this finding, Jensen and coworkers [6,7] reported more improved kinetics of sodium alanate by dispersing Ti(OBu)4 with a dry method. Currently, mechanical ball milling is being widely adopted to disperse a small amount of catalysts effectively into sodium alanates in solid state [8–11]. In addition to sodium alanates, Chen et al. [3,12] reported that Li3 AlH6 exhibits reversible hydrogen storage when catalyzed with Ti-bearing materials by
∗
Corresponding author. Tel.: +82 2 958 5465; fax: +82 2 958 5379. E-mail address: [email protected] (Y.W. Cho).
0925-8388/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2005.07.079
mechanical milling. In spite of the outstanding performance of Ti-bearing catalysts, there is still no clear understanding on how they play a catalytic role in alanates. The first step toward the understanding of this catalytic mechanism might be to confirm what form of Ti exists in alanates. However, it is quite difficult to confirm the Ti state (metallic Ti, Ti compound or Ti substitution in alanates) in alanates using most analytical techniques, because a very small amount of Ti-bearing catalysts is usually introduced [2,3,8,11,13,14]. In this paper, we report our investigation on chemical reactions between TiCl3 and Li3 AlH6 in order to gain an insight into the Ti state in Li3 AlH6 using both experimental and theoretical methods. The catalytic effect of the reaction products is also investigated. 2. Experimental procedures 95% pure LiAlH4 , 95% pure LiH and 99% pure TiCl3 powders were purchased from Sigma–Aldrich. A mixture of LiAlH4 and LiH with a molar ratio of 1:2 was prepared and then a 5 g mixture was charged together with ten 14.0 mm and thirty 9.5 mm diameter zirconia balls into a silicon nitride vial under an Ar atmosphere in a glove box. The ball-to-powder weight ratio was approximately 37:1. The mixture was milled in a Fritsch P4 planetary mill at 350 rpm for 4 h 30 min. After milling, the formation of Li3 AlH6 by the reaction between LiAlH4 and LiH was confirmed by X-ray powder diffraction (XRD) using Cu K␣ radiation. The mechanochemically prepared Li3 AlH6 and TiCl3 were mixed with molar ratios of 6:1, 2:1 and 1:1. A 1 g mixture was charged together with
J.-H. Shim et al. / Journal of Alloys and Compounds 419 (2006) 176–179
177
seventeen 7.9 mm diameter Cr-steel balls in a tool steel vial under an Ar atmosphere. The ball-to-powder weight ratio was approximately 35:1. The mixture was milled in a SPEX-8000 mill for 2 h. The reaction products of the milled powders were characterized by XRD. Some of the milled powders were rinsed in distilled water and filtered in order to selectively remove chloride formed during milling. The reaction products between Li3 AlH6 and TiCl3 were again milled with Li3 AlH6 by a SPEX-8000 mill for 1 h at the same milling conditions. The amount of the reaction products added into Li3 AlH6 was 10–20 wt.%. To check the catalytic effect of the reaction products, differential scanning calorimetry (DSC) was performed in flowing Ar by NETZSCH DSC 204 with a heating rate of 2 ◦ C/min for Li3 AlH6 milled with the reaction products.
3. Thermodynamic calculation Thermodynamic calculation of the Li–Al–H–Ti–Cl system was carried out based on the CALPHAD method to understand what equilibrium phases are for the Li3 AlH6 and TiCl3 mixtures. The phases included in this calculation were Li3 AlH6 , LiAlH4 , LiH, TiCl3 , LiCl, Ti, Al, TiAl, TiAl3 , Ti3 Al, TiAl2 , Ti5 Al11 , TiH2 and H2 . Thermodynamic data for Li3 AlH6 and LiAlH4 were taken from Ref. [15]. The data for the Ti–Al intermetallic phases and all the other phases were from the SGTE solution and substance databases, respectively, which are incorporated into Thermo-Calc [16]. 4. Results and discussion
Fig. 2. XRD patterns of reaction products between Li3 AlH6 and TiCl3 : (a) 6:1, (b) 2:1, (c) 1:1.
Balema et al. [17] showed that Li3 AlH6 can be formed by mechanical milling of a LiAlH4 and LiH mixture according to the following reaction: LiAlH4 + 2LiH = Li3 AlH6
(1)
An XRD pattern of the mixture of LiAlH4 and LiH milled for 4 h and 30 min in the present work is presented in Fig. 1. The pattern is in good agreement with Li3 AlH6 obtained by Balema et al. [17], indicating that reaction (1) was completed during milling. XRD patterns of reaction products between LiAlH6 and TiCl3 milled for 2 h are shown in Fig. 2. For the 6:1 mixture of Li3 AlH6 and TiCl3 , LiCl and L12 –TiAl3 form as reaction products. Very small peaks of Li3 AlH6 are also observed in the pattern. In
Fig. 1. XRD pattern of Li3 AlH6 prepared by mechanochemical reaction.
the case of the 2:1 and 1:1 mixtures, LiCl and L12 –TiAl3 are still observed. However, small peaks of Li3 AlH6 disappear. It is found that there exists another reaction product in the 1:1 mixture, besides LiCl and L12 –TiAl3 (Fig. 2(c)). That phase could be indexed as ␦-TiH2 with a cubic structure, although the main peak of TiH2 overlaps those of LiCl and TiAl3 . The reaction product of the 1:1 mixture was rinsed in water to facilitate the identification of the phases by removing LiCl. An XRD pattern of the rinsed reaction products is given in Fig. 3. The presence of ␦-TiH2 is clearly shown together with L12 –TiAl3 . From these mechanochemical reactions with various mixing ratios between Li3 AlH6 and TiCl3 , it is inferred that TiCl3 transforms mainly into L12 –TiAl3 when it is introduced in Li3 AlH6 as catalyst and
Fig. 3. XRD pattern of reaction products between Li3 AlH6 and TiCl3 (1:1 mixture) after rinsing in water.
178
J.-H. Shim et al. / Journal of Alloys and Compounds 419 (2006) 176–179
starts to produce ␦-TiH2 as well as TiAl3 at high concentrations of TiCl3 . As for NaAlH4 , there are some evidences on the formation of TiAl3 when catalyzed with TiCl3 . Graetz et al. [18] reported the presence of TiAl3 in NaAlH4 containing 2–4 mol% TiCl3 through an X-ray absorption study, although the crystal structure of TiAl3 was not clarified. Majzoub and Gross [19] prepared a 3:1 mixture of NaAlH4 and TiCl3 using mechanical milling. They clearly showed that L12 –TiAl3 and NaCl form in the mixture using XRD. The formation of L12 –TiAl3 in lithium and sodium alanates during milling is quite interesting, because the equilibrium phase at 75 at.% Al in the Ti–Al phase diagram is not L12 (cubic) but D022 –TiAl3 (tetragonal). Although D022 –TiAl3 is energetically more stable than L12 by about 0.05 eV/atom [20], L12 –TiAl3 rather than D022 is very frequently observed rather than D022 particularly during processes that are far from equilibrium such as mechanical milling and thin film deposition [21]. The formation of metastable L12 –TiAl3 might be attributed to its lower kinetic nucleation barrier compared to equilibrium D022 –TiAl3 , because the L12 structure shows lower degree of order than D022 [21]. Moreover, it has been known for a long time that the third elements such as Cu, Cr, Fe, Ni and Mn much stabilize L12 –TiAl3 [22]. Therefore, Li and impurity elements such as Fe, which might have been introduced from balls and vials during milling, probably helped stabilize L12 –TiAl3 . Table 1 summarizes the results of the thermodynamic calculation for the reaction products at 25 ◦ C. In this calculation, D022 –TiAl3 was taken into account instead of L12 –TiAl3 , because thermodynamic data of L12 –TiAl3 are not available. Thermodynamics tells that LiCl and TiAl3 are equilibrium reaction products, which is in good agreement with the results of the present milling experiments. However, TiH2 always appears along with TiAl3 , which shows only partial agreement with the experiment. TiH2 is experimentally observed only in the 1:1 mixture of Li3 AlH6 and TiCl3 in the XRD patterns. In addition, although Li3 AlH6 remains after the reactions in the 2:1 mixture in the calculation, Li3 AlH6 is not observed in the experiment. Calculated mole fractions of the equilibrium phases for the 2:1 mixture between 0 and 100 ◦ C are shown in Fig. 4. Interestingly, the mole fraction of TiAl3 increases and that of TiH2 decreases with increasing temperature. TiAl3 becomes dominant over TiH2 above about 40 ◦ C. Above about 40 ◦ C, Li3 AlH6 disappears as the mole fraction of TiAl3 increases. This is probably because Ti decomposes Li3 AlH6 to obtain Al from Li3 AlH6 and form TiAl3 as TiH2 changes into TiAl3 . Temperature inside
Fig. 4. Calculated equilibrium phase fractions in 2:1 mixture of Li3 AlH6 and TiCl3 between 0 and 100 ◦ C.
a vial, surely, increases by the collisions between balls and vial walls during milling, although it is very difficult to measure or predict the exact temperature increase. The increase in temperature during milling might explain that TiH2 and Li3 AlH6 are not well observed in the 6:1 and 2:1 mixtures. It is shown in Fig. 4 that LiH forms by the decomposition of Li3 AlH6 . By chance, the peaks of LiH are almost coincident with L12 –TiAl3 in an XRD pattern. The overlap of the peaks seems to make the peaks of TiAl3 high in the XRD patterns of the 6:1 and 2:1 mixtures compared to those of the 1:1 mixture, as shown in Fig. 2(a) and (b). DSC curves of Li3 AlH6 , in which 10–20 wt.% of the reaction products from the 6:1 and 1:1 mixtures are dispersed, are given in Fig. 5. All the curves show endothermic peaks of the dehydrogenation of Li3 AlH6 . Without adding the reaction products, Li3 AlH6 decomposes around 210 ◦ C. On the other hand, it is found that the dehydrogenation temperatures come down
Table 1 Calculated equilibrium phase fractions (%) at 25 ◦ C Mixing ratios of Li3 AlH6 and TiCl3
H2 LiCl TiAl3 TiH2 Li3 AlH6
6:1
2:1
1:1
7.3 9.3 2.1 3.1 78.1
19.4 25.0 5.6 8.3 41.7
33.3 42.9 9.5 14.3 0.0
Fig. 5. DSC measurements of Li3 AlH6 catalyzed with reaction products between Li3 AlH6 and TiCl3 .
J.-H. Shim et al. / Journal of Alloys and Compounds 419 (2006) 176–179
by about 30 ◦ C when that the reaction products are dispersed. TiAl3 is likely to play a crucial catalytic role in Ti-catalyzed Li3 AlH6 , considering that the main product formed in all the mixtures is TiAl3 . Chlorides formed together with TiAl3 such as LiCl is not likely to act as catalyst, because the chloridefree reaction product, which is rinsed in water, shows no big difference compared to the chloride-containing ones. However, chlorides probably make a contribution to the in situ formation of nanosized TiAl3 , which might maximize the catalytic effect with a small amount, because it has been known that nanosized reaction products are often produced in chloride-mediated mechanochemical reactions [23,24]. 5. Conclusions Our investigation into reaction products in Ti-catalyzed Li3 AlH6 reveals that LiCl and L12 –TiAl3 form by the reaction between TiCl3 and Li3 AlH6 during mechanical milling. ␦-TiH2 is also observed in the mixture of high TiCl3 concentration. The formation of TiAl3 and TiH2 is in agreement with the results of thermodynamic calculation. However, it is predicted according to thermodynamic calculation that the mole fraction of TiAl3 increases and that of TiH2 decreases as temperature increases during milling. DSC measurements of Li3 AlH6 containing the reaction products show that TiAl3 plays a catalytic role in Li3 AlH6 during dehydrogenation when Li3 AlH6 is catalyzed with TiCl3 . Although chloride itself is not catalyst, it might make a contribution to the formation of nanosized TiAl3 that can maximize the catalytic effect. Acknowledgement This work has been financially supported by the Hydrogen Energy R&D Center under the 21st Century Frontier R&D Program of the Ministry of Science and Technology, Republic of Korea.
179
References [1] http://www.hydrogen.energy.gov. [2] B. Bogdanovi´c, M. Schwickardi, J. Alloys Compd. 253–254 (1997) 1. [3] J. Chen, N. Kuriyama, H.T. Takeshita, T. Sakai, Adv. Eng. Mater. 3 (2001) 695. [4] M. Fichtner, J. Engel, O. Fuhr, O. Kircher, O. Rubner, Mater. Sci. Eng. B108 (2004) 42. [5] H. Morioka, K. Kakizaki, S.-C. Chung, A. Yamada, J. Alloys Compd. 353 (2003) 310. [6] C.M. Jensen, R. Zidan, N. Mariels, A. Hee, C. Hagen, Int. J. Hydrogen Energy 24 (1999) 461. [7] R.A. Zidan, S. Takara, A.G. Hee, C.M. Jensen, J. Alloys Compd. 285 (1999) 119. [8] G. Sandrock, K. Gross, G. Thomas, J. Alloys Compd. 339 (2002) 299. [9] K.J. Gross, G. Sandrock, G.J. Thomas, J. Alloys Compd. 330–322 (2002) 691. [10] D. Sun, T. Kiyobayashi, H.T. Takeshita, N. Kuriyama, C.M. Jensen, J. Alloys Compd. 337 (2002) L8. [11] D.L. Anton, J. Alloys Compd. 356–357 (2003) 400. [12] J. Chen, N. Kuriyama, Q. Xu, H.T. Takeshita, T. Sakai, J. Phys. Chem. B 105 (2001) 11214. [13] D. Sun, T. Kiyobayashi, H.T. Takeshita, N. Kuriyama, C.M. Jensen, J. Alloys. Compd. 337 (2002) L8. [14] J. ´I˜niguez, T. Yildirim, T.J. Udovic, M. Sulic, C.M. Jensen, Phys. Rev. B 70 (2004) 060101. [15] J.-W. Jang, J.-H. Shim, Y.W. Cho, B.-J. Lee, Thermodynamic calculation of LiH ↔ Li3 AlH6 ↔ LiAlH4 reactions, J. Alloys Compd., in press, doi:10.1016/j.jallcom.2005.10.040. [16] http://www.thermocalc.se. [17] V.P. Balema, V.K. Pecharsky, K.W. Dennis, J. Alloys Compd. 313 (2000) 69. [18] J. Graetz, J.J. Reilly, J. Johnson, A.Yu. Ignatov, T.A. Tyson, Appl. Phys. Lett. 85 (2004) 500. [19] E.H. Majzoub, K.J. Gross, J. Alloys Compd. 356–357 (2003) 363. [20] T. Hong, T.J. Watson-Yang, A.J. Freeman, T. Oguchi, T. Xu, Phys. Rev. B 41 (1990) 12462. [21] T. Klassen, M. Oehring, R. Bormann, J. Mater. Res. 9 (1994) 47. [22] S.S. Nayak, B.S. Murty, Mater. Sci. Eng. A367 (2004) 218. [23] T. Tsuzuki, P.G. McCormick, J. Mater. Sci. 39 (2004) 5143. [24] V.P. Balema, J.W. Wiench, K.W. Dennis, M. Pruski, V.K. Pecharsky, J. Alloys Compd. 329 (2001) 108.
Reaction products between TiCl3 catalyst and Li3AlH6 during mechanical mixing Jae-Hyeok Shim, Gil-Jae Lee, Young Whan Cho ∗ Nano-Materials Research Center, Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea Received 22 June 2005; accepted 7 July 2005 Available online 15 November 2005
Abstract The reaction between TiCl3 and Li3 AlH6 was investigated to understand reaction products in Ti-catalyzed Li3 AlH6 using mechanical milling and thermodynamic calculation. As reaction products, LiCl, TiAl3 and TiH2 are observed by X-ray powder diffraction, which is in agreement with the results of thermodynamic calculation. TiAl3 forms in an L12 structure rather than an equilibrium D022 structure during milling. It is predicted that the mole fraction of TiAl3 increases and that of TiH2 decreases as temperature increases during milling. Differential scanning calorimetry confirms that TiAl3 acts as a dehydrogenation catalyst in Li3 AlH6 . © 2005 Elsevier B.V. All rights reserved. Keywords: Energy storage materials; Mechanochemical synthesis; X-ray diffraction; Thermal analysis; Hydrogen storage materials; Catalyst
1. Introduction The development of solid-state hydrogen storage at low and medium temperatures has been recognized as a key technology for hydrogen fuel cell applications. Especially for vehicular applications, it is important to find and design new reversible light-weight hydrogen materials that exhibit high hydrogen capacity [1]. Complex metal alanates (aluminohydrides) have received great attention as hydrogen storage materials owing to their inherent high theoretical hydrogen capacity [2–5]. In 1997, Bogdanovi´c and Schwickardi [2] first demonstrated that reversible hydrogen storage could be achieved under moderate conditions (temperature and pressure) with accelerated kinetics in NaAlH4 and Na3 AlH6 by doping Ti-bearing catalysts such as TiCl3 and Ti(OBu)4 through wet chemistry. After this finding, Jensen and coworkers [6,7] reported more improved kinetics of sodium alanate by dispersing Ti(OBu)4 with a dry method. Currently, mechanical ball milling is being widely adopted to disperse a small amount of catalysts effectively into sodium alanates in solid state [8–11]. In addition to sodium alanates, Chen et al. [3,12] reported that Li3 AlH6 exhibits reversible hydrogen storage when catalyzed with Ti-bearing materials by
∗
Corresponding author. Tel.: +82 2 958 5465; fax: +82 2 958 5379. E-mail address: [email protected] (Y.W. Cho).
0925-8388/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2005.07.079
mechanical milling. In spite of the outstanding performance of Ti-bearing catalysts, there is still no clear understanding on how they play a catalytic role in alanates. The first step toward the understanding of this catalytic mechanism might be to confirm what form of Ti exists in alanates. However, it is quite difficult to confirm the Ti state (metallic Ti, Ti compound or Ti substitution in alanates) in alanates using most analytical techniques, because a very small amount of Ti-bearing catalysts is usually introduced [2,3,8,11,13,14]. In this paper, we report our investigation on chemical reactions between TiCl3 and Li3 AlH6 in order to gain an insight into the Ti state in Li3 AlH6 using both experimental and theoretical methods. The catalytic effect of the reaction products is also investigated. 2. Experimental procedures 95% pure LiAlH4 , 95% pure LiH and 99% pure TiCl3 powders were purchased from Sigma–Aldrich. A mixture of LiAlH4 and LiH with a molar ratio of 1:2 was prepared and then a 5 g mixture was charged together with ten 14.0 mm and thirty 9.5 mm diameter zirconia balls into a silicon nitride vial under an Ar atmosphere in a glove box. The ball-to-powder weight ratio was approximately 37:1. The mixture was milled in a Fritsch P4 planetary mill at 350 rpm for 4 h 30 min. After milling, the formation of Li3 AlH6 by the reaction between LiAlH4 and LiH was confirmed by X-ray powder diffraction (XRD) using Cu K␣ radiation. The mechanochemically prepared Li3 AlH6 and TiCl3 were mixed with molar ratios of 6:1, 2:1 and 1:1. A 1 g mixture was charged together with
J.-H. Shim et al. / Journal of Alloys and Compounds 419 (2006) 176–179
177
seventeen 7.9 mm diameter Cr-steel balls in a tool steel vial under an Ar atmosphere. The ball-to-powder weight ratio was approximately 35:1. The mixture was milled in a SPEX-8000 mill for 2 h. The reaction products of the milled powders were characterized by XRD. Some of the milled powders were rinsed in distilled water and filtered in order to selectively remove chloride formed during milling. The reaction products between Li3 AlH6 and TiCl3 were again milled with Li3 AlH6 by a SPEX-8000 mill for 1 h at the same milling conditions. The amount of the reaction products added into Li3 AlH6 was 10–20 wt.%. To check the catalytic effect of the reaction products, differential scanning calorimetry (DSC) was performed in flowing Ar by NETZSCH DSC 204 with a heating rate of 2 ◦ C/min for Li3 AlH6 milled with the reaction products.
3. Thermodynamic calculation Thermodynamic calculation of the Li–Al–H–Ti–Cl system was carried out based on the CALPHAD method to understand what equilibrium phases are for the Li3 AlH6 and TiCl3 mixtures. The phases included in this calculation were Li3 AlH6 , LiAlH4 , LiH, TiCl3 , LiCl, Ti, Al, TiAl, TiAl3 , Ti3 Al, TiAl2 , Ti5 Al11 , TiH2 and H2 . Thermodynamic data for Li3 AlH6 and LiAlH4 were taken from Ref. [15]. The data for the Ti–Al intermetallic phases and all the other phases were from the SGTE solution and substance databases, respectively, which are incorporated into Thermo-Calc [16]. 4. Results and discussion
Fig. 2. XRD patterns of reaction products between Li3 AlH6 and TiCl3 : (a) 6:1, (b) 2:1, (c) 1:1.
Balema et al. [17] showed that Li3 AlH6 can be formed by mechanical milling of a LiAlH4 and LiH mixture according to the following reaction: LiAlH4 + 2LiH = Li3 AlH6
(1)
An XRD pattern of the mixture of LiAlH4 and LiH milled for 4 h and 30 min in the present work is presented in Fig. 1. The pattern is in good agreement with Li3 AlH6 obtained by Balema et al. [17], indicating that reaction (1) was completed during milling. XRD patterns of reaction products between LiAlH6 and TiCl3 milled for 2 h are shown in Fig. 2. For the 6:1 mixture of Li3 AlH6 and TiCl3 , LiCl and L12 –TiAl3 form as reaction products. Very small peaks of Li3 AlH6 are also observed in the pattern. In
Fig. 1. XRD pattern of Li3 AlH6 prepared by mechanochemical reaction.
the case of the 2:1 and 1:1 mixtures, LiCl and L12 –TiAl3 are still observed. However, small peaks of Li3 AlH6 disappear. It is found that there exists another reaction product in the 1:1 mixture, besides LiCl and L12 –TiAl3 (Fig. 2(c)). That phase could be indexed as ␦-TiH2 with a cubic structure, although the main peak of TiH2 overlaps those of LiCl and TiAl3 . The reaction product of the 1:1 mixture was rinsed in water to facilitate the identification of the phases by removing LiCl. An XRD pattern of the rinsed reaction products is given in Fig. 3. The presence of ␦-TiH2 is clearly shown together with L12 –TiAl3 . From these mechanochemical reactions with various mixing ratios between Li3 AlH6 and TiCl3 , it is inferred that TiCl3 transforms mainly into L12 –TiAl3 when it is introduced in Li3 AlH6 as catalyst and
Fig. 3. XRD pattern of reaction products between Li3 AlH6 and TiCl3 (1:1 mixture) after rinsing in water.
178
J.-H. Shim et al. / Journal of Alloys and Compounds 419 (2006) 176–179
starts to produce ␦-TiH2 as well as TiAl3 at high concentrations of TiCl3 . As for NaAlH4 , there are some evidences on the formation of TiAl3 when catalyzed with TiCl3 . Graetz et al. [18] reported the presence of TiAl3 in NaAlH4 containing 2–4 mol% TiCl3 through an X-ray absorption study, although the crystal structure of TiAl3 was not clarified. Majzoub and Gross [19] prepared a 3:1 mixture of NaAlH4 and TiCl3 using mechanical milling. They clearly showed that L12 –TiAl3 and NaCl form in the mixture using XRD. The formation of L12 –TiAl3 in lithium and sodium alanates during milling is quite interesting, because the equilibrium phase at 75 at.% Al in the Ti–Al phase diagram is not L12 (cubic) but D022 –TiAl3 (tetragonal). Although D022 –TiAl3 is energetically more stable than L12 by about 0.05 eV/atom [20], L12 –TiAl3 rather than D022 is very frequently observed rather than D022 particularly during processes that are far from equilibrium such as mechanical milling and thin film deposition [21]. The formation of metastable L12 –TiAl3 might be attributed to its lower kinetic nucleation barrier compared to equilibrium D022 –TiAl3 , because the L12 structure shows lower degree of order than D022 [21]. Moreover, it has been known for a long time that the third elements such as Cu, Cr, Fe, Ni and Mn much stabilize L12 –TiAl3 [22]. Therefore, Li and impurity elements such as Fe, which might have been introduced from balls and vials during milling, probably helped stabilize L12 –TiAl3 . Table 1 summarizes the results of the thermodynamic calculation for the reaction products at 25 ◦ C. In this calculation, D022 –TiAl3 was taken into account instead of L12 –TiAl3 , because thermodynamic data of L12 –TiAl3 are not available. Thermodynamics tells that LiCl and TiAl3 are equilibrium reaction products, which is in good agreement with the results of the present milling experiments. However, TiH2 always appears along with TiAl3 , which shows only partial agreement with the experiment. TiH2 is experimentally observed only in the 1:1 mixture of Li3 AlH6 and TiCl3 in the XRD patterns. In addition, although Li3 AlH6 remains after the reactions in the 2:1 mixture in the calculation, Li3 AlH6 is not observed in the experiment. Calculated mole fractions of the equilibrium phases for the 2:1 mixture between 0 and 100 ◦ C are shown in Fig. 4. Interestingly, the mole fraction of TiAl3 increases and that of TiH2 decreases with increasing temperature. TiAl3 becomes dominant over TiH2 above about 40 ◦ C. Above about 40 ◦ C, Li3 AlH6 disappears as the mole fraction of TiAl3 increases. This is probably because Ti decomposes Li3 AlH6 to obtain Al from Li3 AlH6 and form TiAl3 as TiH2 changes into TiAl3 . Temperature inside
Fig. 4. Calculated equilibrium phase fractions in 2:1 mixture of Li3 AlH6 and TiCl3 between 0 and 100 ◦ C.
a vial, surely, increases by the collisions between balls and vial walls during milling, although it is very difficult to measure or predict the exact temperature increase. The increase in temperature during milling might explain that TiH2 and Li3 AlH6 are not well observed in the 6:1 and 2:1 mixtures. It is shown in Fig. 4 that LiH forms by the decomposition of Li3 AlH6 . By chance, the peaks of LiH are almost coincident with L12 –TiAl3 in an XRD pattern. The overlap of the peaks seems to make the peaks of TiAl3 high in the XRD patterns of the 6:1 and 2:1 mixtures compared to those of the 1:1 mixture, as shown in Fig. 2(a) and (b). DSC curves of Li3 AlH6 , in which 10–20 wt.% of the reaction products from the 6:1 and 1:1 mixtures are dispersed, are given in Fig. 5. All the curves show endothermic peaks of the dehydrogenation of Li3 AlH6 . Without adding the reaction products, Li3 AlH6 decomposes around 210 ◦ C. On the other hand, it is found that the dehydrogenation temperatures come down
Table 1 Calculated equilibrium phase fractions (%) at 25 ◦ C Mixing ratios of Li3 AlH6 and TiCl3
H2 LiCl TiAl3 TiH2 Li3 AlH6
6:1
2:1
1:1
7.3 9.3 2.1 3.1 78.1
19.4 25.0 5.6 8.3 41.7
33.3 42.9 9.5 14.3 0.0
Fig. 5. DSC measurements of Li3 AlH6 catalyzed with reaction products between Li3 AlH6 and TiCl3 .
J.-H. Shim et al. / Journal of Alloys and Compounds 419 (2006) 176–179
by about 30 ◦ C when that the reaction products are dispersed. TiAl3 is likely to play a crucial catalytic role in Ti-catalyzed Li3 AlH6 , considering that the main product formed in all the mixtures is TiAl3 . Chlorides formed together with TiAl3 such as LiCl is not likely to act as catalyst, because the chloridefree reaction product, which is rinsed in water, shows no big difference compared to the chloride-containing ones. However, chlorides probably make a contribution to the in situ formation of nanosized TiAl3 , which might maximize the catalytic effect with a small amount, because it has been known that nanosized reaction products are often produced in chloride-mediated mechanochemical reactions [23,24]. 5. Conclusions Our investigation into reaction products in Ti-catalyzed Li3 AlH6 reveals that LiCl and L12 –TiAl3 form by the reaction between TiCl3 and Li3 AlH6 during mechanical milling. ␦-TiH2 is also observed in the mixture of high TiCl3 concentration. The formation of TiAl3 and TiH2 is in agreement with the results of thermodynamic calculation. However, it is predicted according to thermodynamic calculation that the mole fraction of TiAl3 increases and that of TiH2 decreases as temperature increases during milling. DSC measurements of Li3 AlH6 containing the reaction products show that TiAl3 plays a catalytic role in Li3 AlH6 during dehydrogenation when Li3 AlH6 is catalyzed with TiCl3 . Although chloride itself is not catalyst, it might make a contribution to the formation of nanosized TiAl3 that can maximize the catalytic effect. Acknowledgement This work has been financially supported by the Hydrogen Energy R&D Center under the 21st Century Frontier R&D Program of the Ministry of Science and Technology, Republic of Korea.
179
References [1] http://www.hydrogen.energy.gov. [2] B. Bogdanovi´c, M. Schwickardi, J. Alloys Compd. 253–254 (1997) 1. [3] J. Chen, N. Kuriyama, H.T. Takeshita, T. Sakai, Adv. Eng. Mater. 3 (2001) 695. [4] M. Fichtner, J. Engel, O. Fuhr, O. Kircher, O. Rubner, Mater. Sci. Eng. B108 (2004) 42. [5] H. Morioka, K. Kakizaki, S.-C. Chung, A. Yamada, J. Alloys Compd. 353 (2003) 310. [6] C.M. Jensen, R. Zidan, N. Mariels, A. Hee, C. Hagen, Int. J. Hydrogen Energy 24 (1999) 461. [7] R.A. Zidan, S. Takara, A.G. Hee, C.M. Jensen, J. Alloys Compd. 285 (1999) 119. [8] G. Sandrock, K. Gross, G. Thomas, J. Alloys Compd. 339 (2002) 299. [9] K.J. Gross, G. Sandrock, G.J. Thomas, J. Alloys Compd. 330–322 (2002) 691. [10] D. Sun, T. Kiyobayashi, H.T. Takeshita, N. Kuriyama, C.M. Jensen, J. Alloys Compd. 337 (2002) L8. [11] D.L. Anton, J. Alloys Compd. 356–357 (2003) 400. [12] J. Chen, N. Kuriyama, Q. Xu, H.T. Takeshita, T. Sakai, J. Phys. Chem. B 105 (2001) 11214. [13] D. Sun, T. Kiyobayashi, H.T. Takeshita, N. Kuriyama, C.M. Jensen, J. Alloys. Compd. 337 (2002) L8. [14] J. ´I˜niguez, T. Yildirim, T.J. Udovic, M. Sulic, C.M. Jensen, Phys. Rev. B 70 (2004) 060101. [15] J.-W. Jang, J.-H. Shim, Y.W. Cho, B.-J. Lee, Thermodynamic calculation of LiH ↔ Li3 AlH6 ↔ LiAlH4 reactions, J. Alloys Compd., in press, doi:10.1016/j.jallcom.2005.10.040. [16] http://www.thermocalc.se. [17] V.P. Balema, V.K. Pecharsky, K.W. Dennis, J. Alloys Compd. 313 (2000) 69. [18] J. Graetz, J.J. Reilly, J. Johnson, A.Yu. Ignatov, T.A. Tyson, Appl. Phys. Lett. 85 (2004) 500. [19] E.H. Majzoub, K.J. Gross, J. Alloys Compd. 356–357 (2003) 363. [20] T. Hong, T.J. Watson-Yang, A.J. Freeman, T. Oguchi, T. Xu, Phys. Rev. B 41 (1990) 12462. [21] T. Klassen, M. Oehring, R. Bormann, J. Mater. Res. 9 (1994) 47. [22] S.S. Nayak, B.S. Murty, Mater. Sci. Eng. A367 (2004) 218. [23] T. Tsuzuki, P.G. McCormick, J. Mater. Sci. 39 (2004) 5143. [24] V.P. Balema, J.W. Wiench, K.W. Dennis, M. Pruski, V.K. Pecharsky, J. Alloys Compd. 329 (2001) 108.