Increase in the Hydrogen-Sorption Rates and the Hydrogen-Storage Capacity of MgH 2 by Adding a Small Proportion of Zn(BH 4 ) 2

Transcription

1 Increase in the Hydrogen-Sorption Rates and the Hydrogen-Storage Capacity of MgH 2 by Adding a Small Proportion of Zn(BH 4 ) 2 Hye Ryoung Park 1, Young Jun Kwak 2, and Myoung Youp Song 3,* 1 School of Chemical Engineering, Chonnam National University, Gwangju 61186, Rep ublic of Korea 2 Department of Materials Engineering, Graduate School, Hydrogen & Fuel Cell Research Center, Engineering Research Institute, Chonbuk National University, Jeonju 54896, Re public of Korea 3 Division of Advanced Materials Engineering, Hydrogen & Fuel Cell Research Center, Engineering Research Institute, Chonbuk National University, Jeonju 54896, Republic o f Korea Abstract In this work, Zn(BH 4 ) 2 was added to improve the hydrogen-storage properties of MgH 2. A 99 wt% MgH 2 +1 wt% Zn(BH 4 ) 2 sample was prepared by milling in a planetary ball mill in a hydrogen atmosphere. The proportion of the added Zn(BH 4 ) 2 was small (1 wt%) in order to increase hydriding and dehydriding rates without reducing the hydrogen-storage capacity too much. Changes in the released hydrogen quantity, H d, with temperature, T, for as-milled 99MgH 2 +1Zn(BH 4 ) 2 was obtained by heating the sample from room temperature to 683 K with a heating rate of 5 K/min under 1.0 bar of gas. Activation of the sample was not required. 99MgH 2 +1Zn(BH 4 ) 2 had an effective hydrogen-storage capacity (the quantity of hydrogen 1

2 absorbed in 60 min) of 4.83 wt% in the first cycle. The as-purchased MgH 2 absorbed hydrogen slowly, absorbing 0.04 wt% H in 60 min. On the other hand, in the first cycle, 99MgH 2 + 1Zn(BH 4 ) 2 absorbed 3.97 wt% H in 5 min, 4.49 wt% H in10 min, and 4.83 wt% H in 60 min at 593 K under 12 bar H 2. (Received February 20, 2017; Accepted May 29, 2017) Keywords: hydrogen absorbing materials, mechanical milling, hydrogen, thermal analysis, MgH 2 -based alloy. Corresponding author. Tel.: address: songmy@jbnu.ac.kr (Myoung Youp Song) 1. INTRODUCTION Magnesium hydride has attracted attention as a hydrogen-storage material due to the high gravimetric hydrogen-storage capacity of magnesium hydride and relative cheapness and large deposits in the earth s crust of magnesium [1]. However, its hydriding and dehydriding rates are very low and its decomposition temperature is high. Many researchers have tried to increase the hydriding and dehydriding rates of magnesium by alloying with it certain metals [2-8] and by synthesizing compounds of magnesium [9,10]. Metal borohydrides [M(BH 4 ) n ] are prospective candidates for developed hydrogen storage materials since they can store a large amount of hydrogen per unit weight [11-20]. The complex metal hydride Zn(BH 4 ) 2, one of the metal borohydrides, has also attracted attention due to high gravimetric hydrogen-storage capacity (8.4 wt%) [21], low decomposition temperature ( K), and easy preparation [22-24]. Nakagawa et al. [21] synthesized 2

3 Zn(BH 4 ) 2, which formed together NaCl, by milling ZnCl 2 and NaBH 4. Zn(BH 4 ) 2 was selected as an additive to improve the hydriding and dehydriding rates of the magnesium in this work. In this work, the Zn(BH 4 ) 2 prepared in our previous work [25] was added to MgH 2 to improve its hydrogen storage properties. A 99 wt% MgH 2 +1 wt% Zn(BH 4 ) 2 sample (designated 99MgH 2 +1Zn(BH 4 ) 2 ) was prepared by milling in a planetary ball mill in a hydrogen atmosphere. The proportion of the additive was small (1 wt%) so that hydriding and dehydriding rates could be increased without decreasing the hydrogen-storage capacity too much. The hydrogen absorption and release properties of the prepared samples were investigated. 2. EXPERIMENTAL DETAILS MgH 2 (Magnesium hydride, Aldrich, hydrogen-storage grade) and Zn(BH 4 ) 2 prepared in our previous work [25] were used as the starting materials. Mechanical grinding was performed with a planetary ball mill (Planetary Mono Mill; Pulverisette 6, Fritsch), as described in our previous study [26]. A mixture with the composition of 99 wt% MgH 2 +1 wt% Zn(BH 4 ) 2 was ground in a hermetically-sealed stainless steel container (with a volume of 250 ml) filled with high purity hydrogen gas ( 12 bar). The total weight of the mixture was 8 g and the weight of the 105 hardened steel balls used was 360 g, thus corresponding to a sample to ball weight ratio of 1/45. Samples were handled in a glove box under Ar to keep from oxidizing. The disc revolution speed was 400 rpm and milling time was 2 h. The quantities of hydrogen absorbed and released by the samples as a function of time were measured by a volumetric method under nearly constant hydrogen pressures, using a Sievert s type hydriding and dehydriding apparatus, as described previously [27]. 3

4 3. RESULTS The quantity of released hydrogen, H d, is defined as the percentage of released hydrogen in relation to sample weight. Figure 1 shows the variation in the released hydrogen quantity, H d, with temperature, T, for as-milled 99MgH 2 +1Zn(BH 4 ) 2. The sample was heated from room temperature to 683 K with a heating rate of 5 K/min under 1.0 bar of gas. The value of H d increases slowly from room temperature to about 603 K and rapidly from about 635 K to 683 K. This shows that the MgH 2 phase in 99MgH 2 +1Zn(BH 4 ) 2 begins to decompose at about 635 K. The values of H d are 0.28 wt% at 523 K, 0.76 wt% at 633 K, 2.47 wt% at 653 K, and 4.34 wt% at 683 K. At different temperatures, the rates of H d with temperature T, dh d /dt, were obtained. Change with temperature in the rate of H d with temperature T, dh d /dt, for as-milled 99MgH 2 +1Zn(BH 4 ) 2 is shown in Fig. 2. An absorption peak appears at about 655 K. It is believed that MgH 2 begins to decompose at about 625 K and the rate of H d with temperature T, dh d /dt, reaches its maximum at about 655 K. The amount of absorbed hydrogen, H a, is also expressed with respect to sample weight. Figure 3 shows the variation in the H a vs. t curve of 99MgH 2 +1Zn(BH 4 ) 2 at 593 K under 12 bar H 2 with the number of cycles, n. The initial hydriding rate is quite high and the quantity of hydrogen absorbed for 60 min is quite large at n=1. The sample absorbs quite rapidly from the start to about 10 min and very slowly after about 20 min. At n=1, the sample absorbs 3.02 wt% H for 2.5 min, 3.97 wt% H for 5 min, 4.49 wt% H for 10 min, and 4.83 wt% H for 60 min. The quantity of hydrogen absorbed in 60 min is defined as the effective hydrogenstorage capacity. 99MgH 2 +1Zn(BH 4 ) 2 has an effective hydrogen-storage capacity of 4.83 wt% 4

5 at n=1. The initial hydriding rate and the effective hydrogen-storage capacity decrease in general as the number of cycles increases from n=1 to n=6. Activation is a process in which hydriding-dehydriding cycling is performed so that the sample will have maximum hydriding and dehydriding rates. During activation the particles of the sample becomes smaller, clean surfaces form, and defects are created due to expansion (by the hydriding reaction) and contraction (by the dehydriding reaction). Figure 3 shows that the initial hydriding rate and the effective hydrogen-storage capacity are the highest at n=1 and decrease as n increases. This means that the particles of the prepared sample are highly reactive and hydridingdehydriding cycling decreases the initial hydriding rate and the effective hydrogen-storage capacity probably due to coalescence of particles. Based on these results, it can be said that that the activation of 99MgH 2 +1Zn(BH 4 ) 2 is not required. At n=6, the sample absorbs 2.47 wt% H for 2.5 min, 3.38 wt% H for 5 min, 4.01 wt% H for 10 min, and 4.63 wt% H for 60 min. Table 1 shows the changes in H a with t at 593 K under 12 bar H 2 at n=1 and n=6 for 99MgH 2 +1Zn(BH 4 ) 2 The variation in the H d vs. t curve of 99MgH 2 +1Zn(BH 4 ) 2 at 623 K under 1.0 bar H 2 with the number of cycles, n, is shown in Fig. 4. At n=1, the dehydriding rate is quite high from the start to about 2.5 min, becomes higher than the initial dehydriding rate after about 2.5 min, and then becomes very low after about 20 min. The reason that the dehydrding rate becomes higher than the initial dehydriding rate after about 2.5 min is due to the initiation of the dehydriding reaction in some particles in which the nucleation of Mg-H solid solution (the main dehydriding phase is MgH 2 ) is completed. The very low dehydriding rate after about 20 min is because the limit of reactive range of particles is almost approached and there is a decrease in the unreacted core. The quantity of hydrogen released in 60 min is also quite large 5

6 at n=1. At n=1, the sample releases 0.56 wt% H in 2.5 min, 1.40 wt% H in 5 min, 3.02 wt% H in 10 min, and 4.75 wt% H in 60 min. The initial dehydriding rate and the quantity of hydrogen released in 60 min generally decrease as the number of cycles increases from n=1 to n=6. At n=6, the sample releases 0.44 wt% H in 2.5 min, 1.13 wt% H in 5 min, 2.37 wt% H in 10 min, and 4.60 wt% H in 60 min. Table 2 shows the changes in H d with t at 623 K under 1.0 bar H 2 at n=1 and n=6 for 99MgH 2 +1Zn(BH 4 ) 2. Figure 5 presents the XRD pattern of 99MgH 2 +1Zn(BH 4 ) 2 after milling in a planetary ball mill in a hydrogen atmosphere. The background is slightly high and the peaks are broad, indicating that the particles have micro-strain, resulting from severe plastic deformation, after milling in the hydrogen atmosphere. The sample contains β-mgh 2 (JCPDS diffraction data card # ), γ-mgh 2 ( ), and MgO ( ). This shows that γ-mgh 2 was formed during milling in a hydrogen atmosphere even under the low hydrogen pressure of about 12 bar. The phases related to Zn(BH 4 ) 2 were not detected. This is believed to be since only a small quantity of Zn(BH 4 ) 2 is contained in the sample and because the phases may appear at the diffraction angles similar to those of other phases. The slightly high back ground also makes it difficult to detect the weak diffraction lines. Nakagawa et al. [20] reported that Zn(BH 4 ) 2 releases hydrogen with toxic diborane (B 2 H 6 ), after melting with increasing temperature. The XRD pattern of 78.3 wt% Zn(BH 4 ) wt% MgH 2 after being heated up to 643 K showed that the sample contained NaCl, Zn, and MgH 2 [25, 28, 29]. It is thought that, during milling in a planetary ball mill in a hydrogen atmosphere and/or while heating the sample from room temperature to 683 K to measure the H d of the as-milled sample as a function of temperature, the NaCl remains un-reacted, and Zn(BH 4 ) 2 produces Zn, B 2 H 6, and H 2. 6

7 The XRD pattern of 99MgH 2 +1Zn(BH 4 ) 2 dehydrided in the 6 th cycle is presented in Fig. 6. The sample appears to contain Mg (JCPDS diffraction data card # ), β-mgh 2 ( ), and MgO ( ). In addition, the γ-mgh 2 phase ( ) cannot be observed. It is believed that, during hydriding-dehydriding cycling, MgO, NaCl, and Zn remain un-reacted while Mg absorbs and releases hydrogen. Figure 7 represents the SEM micrographs of 99MgH 2 +1Zn(BH 4 ) 2 after milling in a planetary ball mill in a hydrogen atmosphere at different magnifications. Particles are agglomerated probably due to the ductility of Mg and the low decomposition temperature ( K) of Zn(BH 4 ) 2. The agglomeration of particles may reduce the rate of hydrogen transfer to and from the surfaces of particles. The sample has both small particles and large particles. Some large particles have flat surfaces and sharp edges. The initial hydriding rate and the quantity of hydrogen absorbed in 60 min decrease in general as the number of cycles increases from n=1 to n=6, as shown in Fig. 3. Figure 4 shows that the initial dehydriding rate and the quantity of hydrogen released for 60 min generally decrease as the number of cycles increases from n=1 to n=6. The SEM micrographs of the 99MgH 2 +1Zn(BH 4 ) 2 dehydrided in the 6 th hydriding-dehydriding cycle at different magnifications are shown in Fig. 8. The particles of the agglomerates, which are observed in the as-milled sample, are now separated. The sample has small particles and large agglomerates consisting of small particles. During the dehydriding reaction, the volume of particles decreases. The contraction of the magnesium particles due to the dehydriding reaction to obtain the H d vs. T curve of the as-milled sample (Fig.1) is considered to separate the particles of the agglomerates, making the sample have the highest initial hydriding rate and the largest quantity of hydrogen absorbed in 60 min at n=1. The particles are believed to coalesce from n=1 to n=6, 7

8 leading to the disappearance of hydrogen paths like cracks inside the particles, because the sample was cycled at the relatively high temperatures between 593 K and 623 K. The disappearance of hydrogen paths such as cracks inside the particles is believed to decrease the initial hydriding and dehydriding rates and the quantities of hydrogen absorbed and released in 60 min as the number of cycles increases from n=1 to n=6. Figure shows H a vs. t curves at 593K under 12 bar H 2 for as-purchased MgH 2 [30], and 99MgH 2 +1Zn(BH 4 ) 2 at n=1. The as-purchased MgH 2 absorbs hydrogen slowly, absorbing 0.04 wt% H for 60 min. The 99MgH 2 +1Zn(BH 4 ) 2 exhibits much higher initial hydriding rates and much larger values of H a (60 min) than the as-purchased MgH 2, showing that the addition of Zn(BH 4 ) 2 by milling under hydrogen increases the initial hydriding rate and the value of H a after 60 min. High-energy ball milling in hydrogen of MgH 2 with Zn(BH 4 ) 2 is believed to create defects, produce clean surfaces, and reduce the particle size, leading to the increases in the hydriding and dehydriding rates and the hydrogen-storage capacity. Notably, activation of the sample was not required. The initial hydriding and dehydriding rates and the quantities of hydrogen absorbed and released in 60 min decrease in general as the number of cycles increases from n=1 to n=6. This is believed due to the coalescence of particles, leading to the disappearance of hydrogen paths like cracks inside the particles, because the sample was cycled at the relatively high temperatures between 593 K and 623 K. The particles of 99MgH 2 +1Zn(BH 4 ) 2 after milling in the planetary ball mill in a hydrogen atmosphere are agglomerated. The sample dehydrided in the 6 th hydriding-dehydriding cycle has separated particles. These results indicate that the contraction of particles due to dehydriding reaction for the measurement of the H d vs. T curve (Fig. 1) separates the 8

9 agglomerated particles. 5. CONCLUSIONS A 99 wt% MgH 2 +1 wt% Zn(BH 4 ) 2 sample was prepared by milling in a planetary ball mill in a hydrogen atmosphere. The activation of the sample was not required. Milling in hydrogen of MgH 2 with Zn(BH 4 ) 2 is believed to create defects, produce clean surfaces, and decrease the particle size, leading to the increases in the hydriding and dehydriding rates and the hydrogen-storage capacity. It is thought that, during heating the as-milled samples and/or during heating the sample from room temperature to 683 K to measure the H d of the as-milled sample as a function of temperature, NaCl remains un-reacted and Zn(BH 4 ) 2 produces Zn, B 2 H 6, and H 2. It is also believed that, during heating the as-milled samples to measure the H d vs. T curve, the agglomerated particles were separated. 99MgH 2 +1Zn(BH 4 ) 2 had an effective hydrogen-storage capacity (the quantity of hydrogen absorbed in 60 min) of 4.83 wt%. The as-purchased MgH 2 absorbed hydrogen slowly, absorbing 0.04 wt% H in 60 min. The effective hydrogen-storage capacity decreased as n increased, probably due to the coalescence of particles because the sample was cycled at the relatively high temperatures. On the other hand, in the first cycle, 99MgH 2 +1Zn(BH 4 ) 2 absorbed 3.97 wt% H in 5 min, 4.49 wt% H in 10 min, and 4.83 wt% H in 60 min at 593 K under 12 bar H 2. At n=1, the sample released 1.40 wt% H in 5 min, 3.02 wt% H in 10 min, and 4.75 wt% H in 60 min at 623 K under 1.0 bar H 2. REFERENCES 1. M. Y. Song, Y. J. Kwak, S. H. Lee, and H. R. Park, Korean J. Met. Mater. 51, 119 (2013). 9

10 2. K. I. Kim and T. W. Hong, Korean J. Met. Mater. 49, 264 (2011). 3. M. Y. Song, Y. J. Kwak, S. H. Lee, and H. R. Park, Met. Mater. Int. 19, 879 (2013). 4. J. J. Reilly and R. H. Wiswall, Inorg. Chem. 6, 2220 (1967). 5. J. J. Reilly and R. H. Wiswall Jr, Inorg. Chem. 7, 2254 (1968). 6. E. Akiba, K. Nomura, S. Ono, and S. Suda, Int. J. Hydrogen Energy 7, 787 (1982). 7. J.-L. Bobet, E. Akiba, and B. Darriet, Int. J. Hydrogen Energy 26, 493 (2001). 8. Z. Li, X. Liu, L. Jiang, and S. Wang, Int. J. Hydrogen Energy 32, 1869 (2007). 9. J. M. Boulet and N. Gerard, J. Less-Common Met. 89, 151 (1983). 10. D. K. Slattery, Int. J. Hydrogen Energy 20, 971 (1995). 11. A. Züttel, S. Rentsch, P. Fisher, P. Wenger, P. Sudan, Ph. Mauron, and Ch. Emmenegger, J. Alloys Compd , 515 (2003). 12. S. Orimo, Y. Nakamori, and A. Züttel, Mater. Sci. Eng. B 108, 51 (2004). 13. H. Hagemann, S. Gomes, G. Renaudin, and K. Yvon, J. Alloy. Compd. 363, 129 (2004). 14. G. Renaudin, S. Gomes, H. Hagemann, L. Keller, and K. Yvon, J. Alloy. Compd. 375, 98 (2004). 15. M. Yoshino, K. Komiya, Y. Takahashi, Y. Shinzato, H. Yukawa, and M. Morinaga, J. Alloy. Compd , 185 (2005). 16. S. Orimo, Y. Nakamori, G. Kitahara, K. Miwa, N. Ohba, S. Towata, and A. Züttel, J. Alloy. Compd , 427 (2005). 17. J. K. Kang, S. Y. Kim, Y. S. Han, R. P. Muller, and W. A. Goddard III, Appl. Phys. Lett. 87, (2005). 18. R.S. Kumar and A.L. Cornelius, Appl. Phys. Lett. 87, (2005). 19. Y. Nakamori, K. Miwa, A. Ninomiya, H.-W. Li, N. Ohba, S. Towata, A. Züttel, and S. 10

11 Orimo, Phys. Rev. B 74, (2006). 20. Y. Nakamori, H.-W. Li, K. Miwa, S. Towata, and S. Orimo, Mater. Trans. 47, 1898 (2006). 21. T. Nakagawa, T. Ichikawa, Y. Kojima, and H. Fujii, Mater. Trans. 48, 556 (2007). 22. V. I. Mikheeva, N. N. Naltseva, and L.S. Alekseeva, Zh. Neorg. Khim. 13, 1301 (1968). 23. E. Jeon and Y. W. Cho, J. Alloy. Compd. 422, 273 (2006). 24. E. Jeon and Y. W. Cho, Transactions of the Korean Hydrogen and New Energy Society 16, 262 (2005). 25. Y. J. Kwak, S. N. Kwon, S. H. Lee, I. W. Park, and M. Y. Song, Korean J. Met. Mater. 53, 500 (2015). 26. Y. J. Kwak, S. H. Lee, H. R. Park, and M. Y. Song, Korean J. Met. Mater. 51, 607 (2013). 27. M. Y. Song, S. H. Baek, J.-L. Bobet, J. Song, and S. H. Hong, Int. J. Hydrogen Energy 35, (2010). 28. Y. J. Kwak, S. N. Kwon, and M. Y. Song, Met.Mater. Int. 21, 971 (2015). 29. Y. J. Kwak, S. N. Kwon, and M. Y. Song, Korean J. Met. Mater. 53, 808 (2015). 30. M. Y. Song, Y. J. Kwak, S. H. Lee, H. R. Park, Korean J. Met. Mater. 52, 689 (2014). 11

12 Table 1 Changes in H a with t at 593 K under 12 bar H 2 at n=1 and n=6 for 99MgH 2 +1Zn(BH 4 ) 2. H a (wt% H) 2.5 min 5 min 10 min 30 min 60 min n = n = Table 2 Changes in H d with t at 623 K under 1.0 bar H 2 at n=1 and n=6 for 99MgH 2 +1Zn(BH 4 ) 2 H d (wt% H) 2.5 min 5 min 10 min 30 min 60 min n = n =

13 Figure captions Fig. 1. Variation in the released hydrogen quantity, H d, with temperature, T, for as-milled 99MgH 2 +1Zn(BH 4 ) 2. The sample was heated from room temperature to 683 K with a heating rate of 5 K/min under 1.0 bar gas. Fig. 2. Change with temperature in the rate of H d with temperature T, dh d /dt, for as-milled 99MgH 2 +1Zn(BH 4 ) 2. Fig. 3. Variation in the H a vs. t curve of 99MgH 2 +1Zn(BH 4 ) 2 at 593 K under 12 bar H 2 with the number of cycles, n. Fig. 4. Variation in the H d vs. t curve of 99MgH 2 +1Zn(BH 4 ) 2 at 623 K under 1.0 bar H 2 with the number of cycles, n. Fig. 5. XRD pattern of 99MgH 2 +1Zn(BH 4 ) 2 after milling in a planetary ball mill in a hydrogen atmosphere. Fig. 6. XRD pattern of 99MgH 2 +1Zn(BH 4 ) 2 dehydrided in the 6 th hydriding-dehydriding cycle. Fig. 7. SEM micrographs of 99MgH 2 +1Zn(BH 4 ) 2 after milling in a planetary ball mill in a hydrogen atmosphere at different magnifications. Fig. 8. SEM micrographs of 99MgH 2 +1Zn(BH 4 ) 2 dehydrided in the 6 th hydridingdehydriding cycle at different magnifications. Fig. 9. H a vs. t curves at 593K under 12 bar H 2 for MgH 2 and 99MgH 2 +1Zn(BH 4 ) 2 at n=1. 13

14 Fig. 1. Variation in the released hydrogen quantity, H d, with temperature, T, for as-milled 99MgH 2 +1Zn(BH 4 ) 2. The sample was heated from room temperature to 683 K with a heating rate of 5 K/min under 1.0 bar gas. 14

15 dh d /dt (wt% H/K) Temperature (K) Fig. 2. Change with temperature in the rate of H d with temperature T, dh d /dt, for as-milled 99MgH 2 +1Zn(BH 4 ) 2. 15

16 H a (wt% H) t (min) n=1 n=2 n=3 n=4 n=5 n=6 Fig. 3. Variation in the H a vs. t curve of 99MgH 2 +1Zn(BH 4 ) 2 at 593 K under 12 bar H 2 with the number of cycles, n. 16

17 H d (wt% H) n=1 n=2 n=3 n=4 n=5 n= t (min) Fig. 4. Variation in the H d vs. t curve of 99MgH 2 +1Zn(BH 4 ) 2 at 623 K under 1.0 bar H 2 with the number of cycles, n. 17

18 Fig. 5. XRD pattern of 99MgH 2 +1Zn(BH 4 ) 2 after milling in a planetary ball mill in a hydrogen atmosphere. 18

19 Fig. 6. XRD pattern of 99MgH 2 +1Zn(BH 4 ) 2 dehydrided in the 6 th hydriding-dehydriding cycle. 19

20 Fig. 7. SEM micrographs of 99MgH 2 +1Zn(BH 4 ) 2 after milling in a planetary ball mill in a hydrogen atmosphere at different magnifications. 20

21 Fig. 8. SEM micrographs of 99MgH 2 +1Zn(BH 4 ) 2 dehydrided in the 6 th hydridingdehydriding cycle at different magnifications. 21

22 H a (wt% H) MgH 2 MgH 2-1Zn(BH 4 ) t (min) Fig. 9. H a vs. t curves at 593K under 12 bar H 2 for MgH 2 and 99MgH 2 +1Zn(BH 4 ) 2 at n=1. 22