The effect of doping NiCl2 on the dehydrogenation properties of LiAlH4

The effect of doping NiCl2 on the dehydrogenation properties of LiAlH4

international journal of hydrogen energy 33 (2008) 6216–6221 Available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/he The ef...

504KB Sizes 3 Downloads 38 Views

international journal of hydrogen energy 33 (2008) 6216–6221

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

The effect of doping NiCl2 on the dehydrogenation properties of LiAlH4 T. Suna, C.K. Huanga, H. Wanga, L.X. Sunb, M. Zhua,* a

School of Material Science and Engineering, South China University of Technology, Guangzhou 510640, PR China Materials and Thermochemistry Laboratory, Dalian Institute of Chemical Physics, Chinese Academy of Science, Dalian 116023, China

b

article info

abstract

Article history:

To investigate the effect of NiCl2 dopant and doping processes on the dehydrogenation

Received 27 May 2008

behavior of LiAlH4, powder mixtures of LiAlH4 and NiCl2 dopant were prepared by designed

Received in revised form

mixing and milling processes. The microstructures of the powder mixtures were charac-

28 July 2008

terized by scanning electron microscopy (SEM) and X-ray diffraction (XRD). The onset

Accepted 2 August 2008

decomposition temperature and isothermal dehydrogenation kinetics were measured to

Available online 23 September 2008

identify the catalyzing effect of NiCl2 on the dehydrogenation properties of LiAlH4. It has

Keywords:

enhance the dehydriding kinetic performance of LiAlH4. The experimental result showed

been shown that NiCl2 could greatly reduce the onset decomposition temperature and Hydrogen storage

that milling LiAlH4 and NiCl2 together for 0.5 h can obtain a homogeneous mixture of

Complex hydrides

LiAlH4, which desorbed 4.2 wt% H2 in 3 h at only 100  C, and the onset decomposition

LiAlH4

temperature of LiAlH4 is decreased by 50  C. However, higher amount of dopant and

NiCl2

unfavorable distribution of NiCl2 in LiAlH4 may cause a reduction of the catalytic efficiency. ª 2008 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

Light metal complex hydrides have drawn intensive research interest due to their high hydrogen storage density and moderate working temperature, which possibly meet the target established for the on-board hydrogen source for fuel cell. Since the finding of effective way to enhance the reversibility of NaAlH4 by Bodanovic and Schwickardi [1], complex hydrides and imides, such as NaAlH4, LiAlH4, LiBH4 and LiNH2, attract wide researches in many aspects [2–6]. Unfortunately, hydrogen absorption and desorption kinetics of these systems are generally poor and great effort was being made to overcome this drawback. In recent years, substantial progress has been achieved in improving the kinetic properties by doping more effective catalysts, such as Ti-based and V-based, to complex metal hydrides [7–11]. Li–Al–H system consists of a series of alkali metal aluminum hydrides such as LiAlH4 (alanate), AxByAlH6

(elpasolite) and Li3AlH6 (cryolite). Among these hydrides, LiAlH4 seems to be the most suitable carrier for hydrogen due to its relatively lower decomposition temperature and higher theoretical hydrogen storage capacity (10.6 wt%). In general consideration, hydrogen can be released from three decomposition stages of LiAlH4, as given in Eqs. (1)–(3). Theoretically, each reaction can release 5.3 wt%, 2.6 wt% and 2.6 wt% amount of hydrogen, respectively. 150w175  C

3LiAlH4 ƒƒƒƒƒƒ! Li3 AlH6 þ 2Al þ 3H2 ð5:3 wt% H2 Þ

(1)

3 180w220  C Li3 AlH6 ƒƒƒƒƒƒ! 3LiH þ Al þ H2 ð2:6 wt% H2 Þ 2

(2)

3 350w400  C 3LiH þ 3Al ƒƒƒƒƒƒ! 3LiAl þ H2 ð2:6 wt% H2 Þ 2

(3)

Recently, many researches on exploring new kind of catalysts or modifying the microstructure to enhance the dehydrogenation kinetics and reversibility of LiAlH4 have been

* Corresponding author. Tel.: þ86 020 87113924. E-mail address: [email protected] (M. Zhu). 0360-3199/$ – see front matter ª 2008 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2008.08.027

6217

international journal of hydrogen energy 33 (2008) 6216–6221

carried out. Many reports show that transition-metal chlorides (TiCl4, TiCl3, AlCl3, and FeCl3), elementals (Fe, Ni, and V) and carbon black are effective catalytic additives [8,12,13]. With the doping of titanium in þ4 or þ3 state, the temperature of the second decomposition reaction (Eq. (2)) of LiAlH4 decreases, but with a loss of storage capacity. The addition of elemental titanium, iron, nickel and vanadium caused a slight decrease in the amount of hydrogen desorption, but did not eliminate the first step of hydrogen releasing during ballmilling process [8]. Balema et al. attributed the high catalytic activity of TiCl4 to the microcrystalline intermetallic Al3Ti that rapidly formed in situ from TiCl4 and LiAlH4 during mechanical milling [13]. Zheng et al. [14] reported that doping Ce(SO4)2 and LaCl3 both shortened the second-stage decomposition reaction (Eq. (2)) and decreased the decomposition temperature of LiAlH4 by about 30  C. In contrast, those dopants prolonged the reaction time of the first and third decomposition stages (Eqs. (1) and (3)), while doping with Ni did not show similar effect. Like other transition-metal ions with unfilled 3d electron shell, Ni2þ cation exhibits broad absorption bands in the visible and infrared spectral range. Nickel-based catalyst was found to exhibit satisfactory activity and selectivity for the carbonylation of methanol [15]. Kojima et al. [16] doped LiAlH4 with 5 wt% of nano-Ni by ball milling, and reported that the catalytic activity of Ni was improved with decreasing the Ni particle size. Ni is also a very important alloying element for improving the hydriding/dehydring properties of metal hydrides such as ZrV2, LaMg3 [17]. In this paper the influence of the microstructure, including distribution of catalysts, and grain size of NiCl2 doped, on LiAlH4 dehydrogenation was investigated. To do that, LiAlH4 was doped with different amounts of NiCl2 by different doping processes to obtain different combinations of catalyst distribution and matrix microstructure.

2.

Experimental details

LiAlH4 (with purity of 97 mass %) from Aldrich Co. Ltd and NiCl2 (with purity of 99 mass %) from Alfa Co. Ltd were used. Six samples with various ratios of LiAlH4 and NiCl2 were prepared and the mixing/milling conditions are listed in Table 1. Samples were put in a sealed stainless steel vial together with hardened stainless steel balls and the ratio of the weights of balls to powder was 40:1. The ball milling (BM) was performed in a vibratory mill with a rotation speed of

1200 rpm. The milling was interrupted for 20 min after 60 min of milling to avoid temperature rising during milling. The whole process of sample handling was performed in a glove box filled with high pure argon atmosphere with the content of water and oxygen below 3 ppm. To obtain different distributions of catalyst and crystallite size of LiAlH4, different preparing processes were adopted. Sample S1 was the raw LiAlH4. In samples S2–S5, the pre-milled NiCl2 was mixed with LiAlH4 by milling them just without balls. Raw LiAlH4 and NiCl2 were ball-milled together for only 0.5 h in sample S6. X-ray diffraction (XRD) analysis was made on a Philips X’Pert MPD Pro X-ray (Cu-Ka radiation) diffractometer. The powder samples were wrapped with liquid paraffin to prevent oxidation during XRD measurement. A LEO 1530VP scanning electron microscope (SEM) was employed to observe the microstructure of the sample treated by different milling processes. The as-prepared samples were dispersed on a whole carbon film supported by a copper grid and transferred into the SEM chamber using a seal box. Desorption temperature measurement and isothermal dehydrogenation test were performed in a gas reaction controller made by Advance Materials Corporation. All samples were heated in a vacuum chamber and the desorbed hydrogen amount was measured to determine their lowest decomposition temperature. Temperature was elevated by 5  C every 15 min if no more hydrogen was detected.

3.

Results and discussion

Fig. 1 shows the back-scattering SEM image of samples treated with different preparation processes. For sample S2, it can be seen from Fig. 1(a) that NiCl2 particles were unevenly distributed on the surface of the large original LiAlH4 particles. For the sample S4, as shown in Fig. 1(b), more NiCl2 particles agglomerated on the surface of LiAlH4 particles. Fig. 1(c) shows that the particle size of LiAlH4 was reduced to less than 3 mm, and NiCl2 particles were also uniformly distributed in sample S5. The LiAlH4 matrix and the dopant are in ‘‘loose contact’’ (note: a mixing process could only make the materials maintain a ‘‘physical touch’’ state, and is rather different from the ‘‘boundary combination’’ of each material in the system, which was prepared by milling) in samples S2, S3 (the structure was similar to S2), S4 and S5. For the sample S6, as shown in Fig. 1(d), the boundary of NiCl2 particles is not so clear compared with that in S5. This means that nano-sized NiCl2 particles were homogeneously embedded into the LiAlH4

Table 1 – Samples’ preparation condition ID

S1 S2 S3 S4 S5 S6

Component

LiAlH4 LiAlH4 þ (2 mol%)NiCl2 LiAlH4 þ (6 mol%)NiCl2 LiAlH4 þ (10 mol%)NiCl2 LiAlH4 þ (2 mol%)NiCl2 LiAlH4 þ (2 mol%)NiCl2

Pre-milled time (h) LiAlH4

NiCl2

– – – – 3 –

– 3 3 3 3 –

Mixing time (h)

Ball-milling time (h)

– 1 1 1 0.5 –

– – – – – 0.5

6218

international journal of hydrogen energy 33 (2008) 6216–6221

Fig. 1 – Back-scattering SEM image of (a) S2 sample, (b) S4 sample, (c) S5 sample, and (d) S6 sample. Ball milling LiAlH4 with NiCl2 together can obtain finer particles of powder mixture and more homogeneous microstructure. Detailed preparation process for those samples is given in Table 1.

matrix. It showed that ball milling LiAlH4 with NiCl2 together can obtain finer particles of powder mixture and more homogeneous microstructure. It seems that the boundary division of LiAlH4 becomes more and more inconspicuous in the order: S2 < S4 < S5 < S6. One possible explanation is that Ni chloride makes contribution to the refinement of particles of LiAlH4 when LiAlH4 and NiCl2 were milled together. These fine LiAlH4 particles bound with each other and formed compact matrix with nano-sized NiCl2 embedded inside. Fig. 2 gives the XRD patterns of as-prepared samples S1–S6. After milling (samples S5 and S6), the diffraction peaks of LiAlH4 became broader and the intensity decreased, which suggested the reduction of the average crystallite size of LiAlH4 caused by milling. It is noted that the NiCl2-doped LiAlH4 did not decompose during mixing (samples S2 and S4) or milling (S5). The appearance of the peaks of Li3AlH6 and LiCl in diffractogram of sample S6 shows obviously that substantial decomposition of LiAlH4 happened during milling. The fine particle and crystallite size might improve the contact between LiAlH4 and NiCl2 and thus facilitate the reaction between them. The following reaction might happen during milling and caused a loss of hydrogen capacity of these samples in the dehydrogenation process:

results are shown in Figs. 3 and 4. As seen in Figs. 3 and 4, the abrupt increase of hydrogen release content suggests the decomposition reaction starts. Two plateau regions correspond to the decomposition reaction given in Eqs. (1) and (2), respectively. In Fig. 3, we compared the effect of the amount of dopant on decomposition temperature by using samples

ball milling

5LiAlH4 þ NiCl2 ƒƒƒƒƒƒƒ! 2LiCl þ Li3 AlH6 þ 4Al þ 8H2 þ Ni Unfortunately, the diffraction peaks of Ni were not observed in the XRD results. This may be due to the relatively low doping amount and the strong absorption of Cu-Ka X-ray by NiCl2. Thermal desorption measurement for all samples was carried out to test the minimum desorption temperature. The

Fig. 2 – XRD patterns of S1, S2, S4, S5 and S6 samples. Diffraction peaks of LiAlH4 became broader and the intensity decreased only after milling. Substantial decomposition of LiAlH4 happened during milling in S6. Detailed preparation process for those samples is given in Table 1.

international journal of hydrogen energy 33 (2008) 6216–6221

6219

Fig. 3 – Thermal desorption process for S1, S3, S4, and S5. Two plateau regions correspond to the decomposition reaction given in Eqs. (1) and (2), respectively. Detailed preparation process for those samples is given in Table 1.

Fig. 5 – Isothermal dehydrogenation of S1, S2, S4, S5 and S6 at 100 8C. The extent of improvement depends strongly on doping processes and amount of NiCl2. Detailed preparation process for those samples is given in Table 1.

S2–S4 which had the same preparing condition but different amounts of NiCl2 as shown in Table 1. All NiCl2-doped LiAlH4 system (samples S2–S4) decomposed at lower temperature (101  C, 106  C and 119  C respectively) than undoped LiAlH4 sample S1 (135  C). The second stage of decomposition of LiAlH4 of sample S1 started at 160  C, while those of samples S2–S4 started at 125  C, 135  C and 142  C respectively. This result shows that doping of NiCl2 plays an important role as a catalyst in enhancing the decomposition of LiAlH4 by just mixing. It has also been shown that the lowering of onset decomposition temperature of both stages of LiAlH4 reduced with the increase of dopant amount. A possible explanation to this phenomena is that the agglomeration of NiCl2 becomes

severer in the LiAlH4 matrix with the increasing amount of NiCl2 (seen in Fig. 1(a) and (b)), and the heterogeneous distribution of additive weakens the catalysis effect of NiCl2 in reducing the onset temperature of the first decomposition of LiAlH4. In Fig. 4, we compared the effect of the milling process on decomposition temperature by using samples S2, S5 and S6 which had the same amount of NiCl2 dopant but different preparing conditions as shown in Table 1. As shown in Fig. 4, the lowering of decomposition temperatures of LiAlH4 decreases in the order: S6 (85  C) > S5 (90  C) > S2 (100  C), which shows that preparing process has obvious influence on the catalyzing effect of NiCl2. The LiAlH4 milled together with

Fig. 4 – Thermal desorption process of S3, S6, and S7. The lowering of decomposition temperatures of LiAlH4 decreases in the order: S6 (85 8C) > S5 (90 8C) > S2 (100 8C), which shows that preparing process has obvious influence on the catalyzing effect of NiCl2. Detailed preparation process for those samples is given in Table 1.

Fig. 6 – XRD pattern of S1, S2, S4, S5 and S6 samples after isothermal dehydrogenation at 100 8C. The clear characterized peaks of Li3AlH6 mean that LiAlH4 was completely decomposed. Detailed preparation process for those samples is given in Table 1.

6220

international journal of hydrogen energy 33 (2008) 6216–6221

Fig. 7 – Results of EDX measurement for the sample S5 after isothermal dehydrogenation at 100 8C. Detailed preparation process for those samples is given in Table 1.

NiCl2 (S6) has the lowest onset decomposition temperature. The SEM image given in Fig. 1(d) shows that fine NiCl2 particles are homogenously distributed in the LiAlH4 matrix, while the crystallite of matrix has also been refined greatly. The apparent broadening of diffraction peaks of sample S6 shown in Fig. 2 also proves that. Isothermal dehydring kinetics of S1, S2, S4, S5 and S6 were measured at around 100  C, which was considered to be an appropriate temperature for practical application. As shown in Fig. 5, the dehydrogenation kinetics of all samples doped with NiCl2 were improved. Furthermore, the extent of improvement depends strongly on doping processes and amount of NiCl2. The dehydrogenation rate of LiAlH4 becomes slower with the increasing doping amount. The diffractogram of S2 and S4 in Fig. 6 suggested that most part of LiAlH4 of sample S2 decomposed during the isothermal dehydrogenation process, while only a small amount of LiAlH4 of S4 decomposed at the same time under the same reacting condition. This result is consistent with the thermal desorption measurement of samples S2 and S4 in Fig. 3, in which the onset decomposition temperature of S4 was higher than that of S2. The decomposition of LiAlH4 in sample S6 was the fastest and achieved a maximal desorbed amount of about 4.2 wt% in 3.5 h, while the decomposition of LiAlH4 happened in S1, S2, S4 and S5 was still incomplete. The initial rate of dehydrogenation for samples S5 and S6 seems to deviate somewhat from samples S2 and S4. An obvious ‘‘induction period’’ in the decomposition of LiAlH4 was observed in the first part (up to around 60 min) of the kinetic curve of S6, which possibly characterized a slow production rate of nuclei of the new phase. The second part with an acceleration of the decomposition is named as bulk decomposition [18]. From Fig. 6, the characterized peaks of Li3AlH6 can be identified in samples S2, S5 and S6. This means that the second stage of decomposition did not happen at 100  C. The above results indicate that milling LiAlH4 together with NiCl2 can get a highly homogeneous mixture, with which NiCl2 embedded into the LiAlH4 particles. This structure enhances the efficiency of H released from LiAlH4 and finally improves the dehydrogenation kinetics of the whole system. The reduction of onset decomposing temperature of LiAlH4 from 135  C to

85  C (seen in Fig. 4, S6) further lowers the reacting thermal barrier and helps in advancing the decomposing kinetics either. We also noticed that an increase in the amount of dopant not only raises the onset decomposition temperature, but also lowers the decomposition kinetics (seen in Figs. 3 and 5, samples S2 and S4). Higher amount of NiCl2 may cause more severe coverage of the surface of LiAlH4, which will greatly lower the heat transfer efficiency and prevent the hydrogen releasing from the system. Unfortunately, the trace of NiCl2 could not be identified in our research even the doping amount reached 10 mol% of the whole system, although we did found a weak peak, which fits the standard characterized peak of NiCl2, at 88.632 (2q). A qualitative analysis of sample S4 (seen in Fig. 7, the particle marked by A1 is a LiAlH4 particle with NiCl2 distributed on the surface) shows that the particles consist of Al, Ni, Cl and O.

4.

Conclusion

The catalytic effect of NiCl2 on the onset decomposition and the isothermal dehydrogenation kinetics of LiAlH4 was investigated. It has been shown that doping a small amount of NiCl2 could reduce the onset decomposition temperature and enhance the dehydriding kinetic performance of LiAlH4. Results show that a sample with NiCl2 dispersed in the LiAlH4 and had intimate contact with LiAlH4, which was achieved by ball milling, can have much better dehydriding kinetic properties than those samples with NiCl2 distributed in LiAlH4 by only simply mixing. The sample, of which the LiAlH4 and NiCl2 were milled together for only 30 min, desorbed 4.2 wt% in 3.5 h at 100  C. This temperature is close to the need of practical application. Consequently, NiCl2 is an effective catalyst for the LiAlH4 complex hydride, especially for the first-stage decomposition reaction of LiAlH4 (LiAlH4 / Li3AlH6). It is believed that the Ni chloride makes a contribution to the in situ formation of LiAlH4 and active Ni species, which plays a crucial catalytic role in the LiAlH4–NiCl2 system. It has been proposed that the reaction between Cl and Liþ may release binding energy and promote the decomposition of LiAlH4.

international journal of hydrogen energy 33 (2008) 6216–6221

Acknowledgements [9]

This work was financially supported by the National Natural Science Foundation of China under grant nos. 50501009 and 50631020, 863 programs of China (no. 2006AA05Z126 and no. 2007AA05Z115).

references

[1] Bogdanovic B, Schwickardi M. Ti-doped alkali metal aluminum hydrides as potential novel reversible hydrogen storage materials. Journal of Alloys and Compounds 1997; 253:1–9. [2] Bogdanovic B, Felderhoff M, Kaskel S, Pommerin A, Schlichte K, Schuth F. Improved hydrogen storage properties of Ti-doped sodium alanate using titanium nanoparticles as doping agents. Advanced Materials 2003;15(12):1012. [3] Chen J, Kuriyama N, Xu Q, Takeshita HT, Sakai T. Reversible hydrogen storage via titanium-catalyzed LiAlH4 and Li3AlH6. Journal of Physical Chemistry B 2001;105(45):11214–20. [4] Zuttel A, Wenger P, Rentsch S, Sudan P, Mauron P, Emmenegger C. LiBH4 a new hydrogen storage material. Journal of Power Sources 2003;118(1–2):1–7. [5] Luo WF. (LiNH2–MgH2): a viable hydrogen storage system (vol 381, pg 284, 2004). Journal of Alloys and Compounds 2004; 385(1–2):316. [6] Stearns JE, Matthews MA, Reger DL, Collins JE. The thermal characterization of novel complex hydrides. International Journal of Hydrogen Energy 1998;23:469–74. [7] Wang P, Jensen CM. Method for preparing Ti-doped NaAlH4 using Ti powder: observation of an unusual reversible dehydrogenation behavior. Journal of Alloys and Compounds 2004;379(1–2):99–102. [8] Resan M, Hampton MD, Lomness JK, Slattery DK. Effects of various catalysts on hydrogen release and uptake

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

6221

characteristics of LiAlH4. International Journal of Hydrogen Energy 2005;30(13–14):1413–6. Hu YH, Ruckenstein E. Ultrafast reaction between Li3N and LiNH2 to prepare the effective hydrogen storage material Li2NH. Industrial & Engineering Chemistry Research 2006; 45(14):4993–8. Orimo S, Nakamori Y, Kitahara G, Miwa K, Ohba N, Towata S, Zuttel A. Dehydriding and rehydriding reactions of LiBH4. Journal of Alloys and Compounds 2005;404–406:427–30. Resan M, Hampton MD, Lomness JK, Slattery DK. Effect of TixAly catalysts on hydrogen storage properties of LiAlH4 and NaAlH4. International Journal of Hydrogen Energy 2005;30: 1417–21. Blanchard D, Brinks HW, Hauback BC, Norby P. Desorption of LiAlH4 with Ti- and V-based additives. Materials Science and Engineering B – Solid State Materials for Advanced Technology 2004;108(1–2):54–9. Balema VP, Wiench JW, Dennis KW, Pruski M, Pecharsky VK. Titanium catalyzed solid-state transformations in LiAlH4 during high-energy ball-milling. Journal of Alloys and Compounds 2001;329(1–2):108–14. Zheng X, Qu X, Humail IS, Li P, Wang G. Effects of various catalysts and heating rates on hydrogen release from lithium alanate. International Journal of Hydrogen Energy 2007;32(9): 1141–4. Fujimoto K, Bischoff S, Omata K, Yagita H. Hydrogen effects on nickel-catalyzed vapor-phase methanol carbonylation. Journal of Catalysis 1992;133(2):370–82. Kojima Y, Kawai Y, Matsumoto M, Haga T. Hydrogen release of catalyzed lithium aluminum hydride by a mechanochemical reaction. Journal of Alloys and Compounds 462 (2008) 275–8. Chu H-L, Qiu S-J, Sun L-X, Zhang Y, Xu F, Zhu M. Electrochemical hydrogen storage properties of La0.7Mg0. 3Ni3.5–Ti0.17Zr0.08V0.35Cr0.1Ni0.3 composites. International Journal of Hydrogen Energy 2008;33(2):755–61. Blanchard D, Brinks HW, Hauback BC. Isothermal decomposition of LiAlD4. Journal of Alloys and Compounds 2006;416(1–2):72–9.