Effect of Li3N additive on the hydrogen storage properties ... - CiteSeerX

Effect of Li3N additive on the hydrogen storage properties of Li-Mg-N-H system Lai-Peng Ma, Zhan-Zhao Fang, Hong-Bin Dai, Xiang-Dong Kang, Yan Liang, Pei-Jun Wang, Ping Wang,a) and Hui-Ming Cheng Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, People’s Republic of China (Received 30 September 2008; accepted 29 December 2008)

The effect of Li3N additive on the Li-Mg-N-H system was examined with respect to the reversible dehydrogenation performance. Screening study with varying Li3N additions (5, 10, 20, and 30 mol%) demonstrates that all are effective for improving the hydrogen desorption capacity. Optimally, incorporation of 10 mol% Li3N improves the practical capacity from 3.9 wt% to approximately 4.7 wt% hydrogen at 200 C, which drives the dehydrogenation reaction toward completion. Moreover, the capacity enhancement persists well over 10 de-/rehydrogenation cycles. Systematic x-ray diffraction examinations indicate that Li3N additive transforms into LiNH2 and LiH phases and remains during hydrogen cycling. Combined structure/property investigations suggest that the LiNH2 “seeding” should be responsible for the capacity enhancement, which reduces the kinetic barrier associated with the nucleation of intermediate LiNH2. In addition, the concurrent incorporation of LiH is effective for mitigating the ammonia release.

I. INTRODUCTION

A major technical obstacle to the widespread use of hydrogen-powered vehicles is the lack of a safe and efficient system for onboard hydrogen storage. Hydrogen storage in solid hydrides is a promising option because it offers a volumetric density greater than that of either compressed or liquefied hydrogen.1 However, conventional metal hydrides are greatly limited by their low gravimetric density or high operating temperature.2,3 Considerable recent efforts have been directed at developing light-metal complex hydrides (alanate, amide, or borohydride) and their hybrid systems as potential H-storage media.4–6 In 2002, Chen et al.7 reported the reversible storage of over 10 wt% hydrogen by Li3N. The reversible hydrogenation/dehydrogenation of Li3N proceeds via a twostep process involving the reactions as follows: Li3 N þ H2 ⇆ Li2 NH þ LiH 5:7 wt% Li2 NH þ H2 ⇆ LiNH2 þ LiH 6:5 wt%

;

ð1Þ :

ð2Þ

This revolutionary discovery immediately attracted worldwide interest in Li-N-H and related metal-N-H systems.6 Among them, Li-Mg-N-H system receives the most extensive attention due to its combined advantages a)

Address all correspondence to this author. e-mail: [email protected] DOI: 10.1557/JMR.2009.0248

1936

J. Mater. Res., Vol. 24, No. 6, Jun 2009

of favorable thermodynamics, good reversibility, and moderate hydrogen capacity.8–15 Starting from a 2LiNH2 + MgH2 mixture, the reversible de-/rehydrogenation reactions can be described by following reactions8,9,12,13: 2LiNH2 þ MgH2 ! Li2 MgðNHÞ2 þ 2H2 ⇆ MgðNH2 Þ2 þ 2LiH 5:5 wt% : ð3Þ Compared with the Li-N-H system (e.g., DH = 66 kJ/ mol H2 for reaction 2), Li-Mg-N-H possesses a much lower dehydrogenation reaction enthalpy, just around 44 kJ/mol H2.16–18 Thermodynamically, it predicts a dissociation pressure of 0.1 MPa at 90 C.17 However, two barriers still remain for the potential application of this system.19 On the one hand, the dehydrogenation reaction is kinetically limited. For example, even at an operating temperature of 200 C, the practical capacity is typically lower than 4.2 wt%, far below the theoretical value of 5.5 wt%.9,13,20,21 On the other hand, elimination of ammonia byproduct is a great challenge for most metal-N-H systems, which not only imperils the polymer electrolyte membrane fuel cells but also contributes to cycling capacity loss.6,19 In this regard, identification of effective additives to improve the hydrogen storage properties is an important approach for developing the Li-Mg-N-H system as practical hydrogen storage medium.19 More recently, intensive efforts have focused on exploring compounds as additives to improve the reversible dehydrogenation performance.20–24 It is noteworthy that addition of LiBH4 leads not only to enhanced © 2009 Materials Research Society

L-P. Ma et al.: Effect of Li3N additive on the hydrogen storage properties of Li-Mg-N-H system

low-temperature kinetics but also suppressed ammonia release. This favorable modification was primarily attributed to the formation of solid solution Li4BN3H10 and the product seeding of Li2Mg(NH)2.23,24 Fundamentally, this finding emphasizes the understanding of reaction pathway in facilitating the exploration of effective additives for the Li-Mg-N-H system. In the present study, we demonstrate that Li3N additive is effective for improving the reversible dehydrogenation performance of the Li-Mg-N-H system. X-ray diffraction (XRD) examination indicates that Li3N is converted into stable LiNH2 and LiH phases during hydrogen cycling. Combined structure/property investigations suggest that LiNH2 “seeding” enhances the practical hydrogen capacity, whereas the excess LiH mitigates the ammonia release. We herein report these experimental findings. II. EXPERIMENTAL

The starting material LiNH2 (95%) and Li3N (99.4%) were purchased from Sigma-Aldrich Corporation (USA) and used as received. MgH2 was prepared by milling Mg powder (>99.9%) under 1 MPa hydrogen, followed by hydrogenation at 400 C under 8 MPa hydrogen. This process was repeated three times to achieve a hydrogenation ratio of 95%, as determined by volumetric measurements. The Li-Mg-N-H [Mg(NH2)2 + 2LiH] sample was prepared by hydrogenating a milled 2LiNH2/MgH2 mixture at 200 C under 5 MPa hydrogen pressure.21 The Li-Mg-N-H samples containing Li3N additive were prepared by milling the 2LiNH2/MgH2 mixture with varying amounts of Li3N, followed by hydrogenation at 200 C under 5 MPa hydrogen pressure. To minimize the sample sintering resulting from the significant exothermic effect upon hydrogenating Li3N additive, we adopted a two-step process to start the hydrogenation of the Li-Mg-N-H samples containing Li3N additive: hydrogen was first loaded into the sample cell under room temperature; then, the sample cell was heated to 200 C at a ramping rate of 2 C/min using a temperature-programmed heating unit. The mechanical milling was performed using a Fritsch P7 planetary mill (Idar-Oberstein, Germany) at 300 rpm for 5 h under Ar atmosphere. The sample weighing was performed using an electronic balance with a precision of 0.1 mg. All of the sample handling was performed in an Ar-filled glove-box equipped with a circulative purification system, in which the typical H2O/O2 levels were below 0.1 ppm. The hydrogen storage properties of the sample, with a typical amount of approximately 200 mg, were measured using a custom-made sievert-type apparatus. Isothermal desorption kinetics was measured at desired temperatures under an initial hydrogen pressure of 0.1 MPa, with a typical ending pressure of 0.14 MPa.

The measurement was started from the moment of pushing the reactor into the furnace held at desired temperatures, which resulted in a 30-min delay before reaching equilibrium. Given that no significant dehydrogenation was observed during this stage, only the isothermal curve was presented for clarity. For the as-prepared Mg(NH2)2 + 2LiH sample, an activation process was required, which diminished upon cycling. To allow for a reliable evaluation of the dehydrogenation property, we selected the desorption curves starting from the 2nd cycle for comparison. The hydrogen supply (initial purity: 99.999%) was further purified using a hydrogen storage alloy system to minimize H2O/O2 contamination. Other than specified, we calculated the H-capacity using the weight of the sample containing additives to allow for an evaluation of the practical hydrogen storage property. Hydrogen (m/e = 2) and ammonia (m/e = 17, 16, 15) release of the samples was monitored using mass spectroscopy (MS; Netzsch QMS 403C, Selb, Germany) attached to a synchronous thermal analysis (TG/DSC; Netzsch STA 449C). Approximately 20 mg of sample powder was loaded into a custom-made sample pan in the Ar-filled glovebox. Special caution was taken to minimize the H2O/O2 contamination during subsequent sample transfer from the glove-box into the chamber. Sample measurements were performed under Ar flow (purity: 99.999%) at a ramping rate of 5 C/min. The flow rate of Ar was maintained at 30 mL/min in the entire holding and heating processes. The samples were characterized using XRD (Rigaku D/MAX-2500, Tokyo, Japan; Cu Ka radiation) and Fourier transform infrared spectrometer (FTIR; Bruker TENSOR 27, MCT detector, Karlsruhe, Germany) equipped with an in situ cell. FTIR spectra were collected in the diffuse reflectance infrared Fourier transform (DRIFT) mode at a resolution of 4 cm1. The FTIR examinations were performed after the Li-Mg-N-H [Mg(NH2)2+2LiH] samples were dehydrogenated to different stages at 200 C and 0.1 MPa. The sample loading into the in situ cell was performed in the Ar-filled glove-box. III. RESULTS

Figure 1 compares the hydrogen desorption kinetics of the additive-free Li-Mg-N-H sample and those with varying amounts of Li3N additive (5, 10, 20, and 30 mol%). Overall, Li3N additive is effective for improving the hydrogen desorption capacity. In the present study, the amounts of hydrogen desorbed at 200 C over 120 min are used to compare the practical capacity, after which only a slight increase in capacity is observed for all samples. For example, the additive-free sample only desorbs 3.9 wt% hydrogen at 200 C. Upon adding 5 mol% Li3N, the practical capacity increases to near 4.4 wt%. With increasing additive amount, the capacity further increases and reaches the maximum value


1937


FIG. 1. Effect of Li3N additive on the hydrogen desorption performance of the Li-Mg-N-H sample at 0.1 MPa and 200 C in the 2nd cycle. The inset highlights the evolution of hydrogen capacity as a function of the Li3N addition, with the deviation of 0.05 wt% in capacity values.

(approximately 4.7 wt% H2) at 10 mol% Li3N addition. However, addition of Li3N above this amount results in a progressively reduced capacity to less than 4.2 wt%. It is noteworthy that for the 10 mol% Li3N sample, 20% increment in capacity is achieved relative to the additive-free sample. This improvement is significant because it drives the dehydrogenation toward completion (which is close to the theoretical capacity of 4.9 wt% hydrogen while we consider the impurities and Li3N additive). Thus, 10 mol% Li3N is deemed the optimal adding amount and is selected for further investigation. The property enhancement obtained in the sample with 10 mol% Li3N additive is well preserved even at lowered temperatures. As shown in Fig. 2, the additivefree sample desorbs 3.6 wt% hydrogen over 300 min at 180 C. For the 10 mol% Li3N-added sample, near 4.1 wt% hydrogen is released over the same time interval. At 160 C, the additive-free sample desorbs only 1 wt% hydrogen over 300 min, whereas the Li3N-added sample releases 1.2 wt% hydrogen, showing a 20% increment in capacity. Further investigation indicates that the enhancement arising upon adding Li3N is highly stable during the de-/rehydrogenation cycling. As shown in Fig. 3, the dehydrogenation curves of the 10 mol% Li3N sample at the 2nd and the 10th cycles are nearly identical at both 200 and 180 C. Figure 4 presents the XRD patterns of the as-prepared Li-Mg-N-H samples with varying amounts of Li3N additive. For comparison, the profile of the additive-free sample is also included. For the Li3N-added samples, no diffraction peaks from Li3N can be identified, whereas the characteristic diffraction peaks from LiNH2 and LiH phases appear. Compared with the additive-free sample, the intensity of LiNH2 peaks at 2y = 19.5, 1938

FIG. 2. Hydrogen desorption performance of the 10 mol% Li3N added Li-Mg-N-H sample under 0.1 MPa at desired temperatures in the 2nd cycle: (a) 180 C and (b) 160 C.

FIG. 3. Hydrogen desorption performance of the 10 mol% Li3N added Li-Mg-N-H sample under 0.1 MPa at desired temperatures in the 2nd and 10th cycles: (a) 200 C and (b) 180 C. Rehydrogenation was performed at 200 C under 5 MPa hydrogen pressure.

30.55, and 35.6 increases successively with increasing Li3N addition. Likewise, increased LiH amount is evidenced by the intensity increase of its characteristic peaks at 2y = 38.2 and 44.4 . Because no Li3N or other newly formed phases are detected, these results suggest that the hydrogenation treatment leads to the conversion of Li3N into LiNH2 and LiH following Eqs. (1) and (2). Upon dehydrogenation, the diffraction peaks of host Mg(NH2)2 disappear and the dehydrogenated product Li2Mg(NH)2 can be identified for all the samples, as shown in Fig. 5. The LiNH2 and LiH phases generated from the hydrogenation of Li3N remain stable in the dehydrogenation process, as indicated by the corresponding characteristic peaks. This is understandable from the thermodynamic limit of LiNH2/LiH system.



FIG. 4. XRD patterns of the as-prepared (hydrogenated) Li-Mg-N-H samples with varying amounts of Li3N additive: (a) additive-free, (b) 5 mol%, (c) 10 mol%, (d) 20 mol%, and (e) 30 mol%. The inset presents the magnified patterns of LiNH2 (112) peaks, highlighting the intensity evolution upon increasing the Li3N addition. A small amount of MgO was detected, which should primarily originate from the air contamination during XRD measurements.

FIG. 5. XRD patterns of the dehydrogenated Li-Mg-N-H samples with varying amounts of Li3N additive: (a) additive-free, (b) 5 mol%, (c) 10 mol%, (d) 20 mol%, and (e) 30 mol%. A small amount of MgO was detected, which should primarily originate from the air contamination during XRD measurements.

In the present study, dehydrogenation was carried out at 200 C under a hydrogen pressure of 0.10.14 MPa. This pressure is higher than the dissociation pressure of LiNH2/LiH system at 200 C and thus restrains its dehydrogenation reaction.7 IV. DISCUSSION

It is clear that Li3N addition is effective in improving the practical capacity of Li-Mg-N-H system.

The combined structure/property investigations suggest a correlation between the in situ-formed LiNH2/LiH phases and the capacity variation. Considering the related reactions involved in the present system, three possibilities may be responsible for the observed capacity enhancement: the combined involvement of LiNH2/ LiH into dehydrogenation reaction, the separate contribution from LiH, or LiNH2. We first evaluate the combined effect of (LiNH2 + LiH). For the Li-Mg-N-H system, the dehydrogenated product Li2Mg(NH)2 is characterized by considerable structural flexibility, suggesting that off-stoichiometric imides [e.g., Li2 + 2xMg1-x(NH)2, x = 01] might be formed via Li+ substitution.12 For example, dehydrogenation of host Mg(NH2)2/LiH with excess (LiNH2 + LiH) may yield such imides, as shown in Table I. However, due to the close structural similarity of LiNH2 to Li2Mg(NH)2,12 it is difficult to obtain the accurate lattice parameters of Li2Mg(NH)2 to directly clarify this point for the Li3N-added Li-Mg-N-H samples. In this regard, the relationship between the desorption property and the amount of excess LiNH2/LiH additives provides a valuable hint. Following the proposed formulas in Table I, a linear increase in H-capacity can be expected with increasing Li+ substitution [or (LiNH2 + LiH) addition]. However, the observed capacity dependence on Li3N addition actually demonstrates an optimal additive amount (10 mol% Li3N), addition above which results in a progressively reduced capacity. This result is contrary to the involvement of (LiNH2 + LiH) into the Li2Mg (NH)2 structure, suggesting that the capacity enhancement should not be attributed to the Li+ substitution. We also note that relative to the Mg(NH2)2 + 2LiH composition, incorporating excess LiH results in increased H-capacity by varying the dehydrogenation pathway and product.6,10,11,25 For example, increasing the LiH/Mg(NH2)2 ratio from 2:1 to 2.67:1 proceeds the dehydrogenation reaction to the Mg3N2/Li2NH stage under He flow, which improves the desorption capacity to 6.9 wt%.10 However, the 0.1 MPa hydrogen pressure applied here restrains further dehydrogenation reactions,25 which agrees well with the XRD results (see Fig. 5). Therefore, the capacity enhancement should not be ascribed to the contribution from excess LiH. It is clear that the above results suggest that the in situ-formed LiNH2 should be closely related to the capacity enhancement. In this regard, careful examination of the dehydrogenation reaction by using FTIR provides a valuable hint. Here, we select the hydrogenated Li-MgN-H matrix [i.e., Mg(NH2)2 + 2LiH] and examine the evolution of FTIR spectra at selected points of the dehydrogenation process. As shown in Fig. 6, with proceeding the dehydrogenation reaction, the FTIR spectra are characterized by the emergence and increase of Li2Mg (NH)2 and the simultaneous decrease of Mg(NH2)2.


1939


TABLE I. Dependence of the hydrogen capacity on Li+ substitution into the proposed dehydrogenation product Li2 + 2xMg1-x(NH)2 (x = 01). x 0 0.33 0.5 1

Li2 + 2xMg1-x(NH)2 (x = 01) Li2Mg(NH)2 Li2.66Mg0.67(NH)2 Li3Mg0.5(NH)2 Li2NH

Proposed reactions for the generation of imide products Li2 + 2xMg1-x(NH)2 Mg(NH2)2 + 2LiH ⇆ Li2Mg(NH)2 + 2H2 Mg(NH2)2 + 2LiH + (LiNH2 + LiH) ⇆ 1.5Li2.66Mg0.67(NH)2 + 3H2 Mg(NH2)2 + 2LiH + 2(LiNH2 + LiH) ⇆ 2Li3Mg0.5(NH)2 + 4H2 LiNH2 + LiH ⇆ Li2NH + 2H2

FIG. 6. Evolution of FTIR spectra (N-H vibration region) for the additive-free Li-Mg-N-H [Mg(NH2)2+2LiH] sample in different states: (a) hydrogenated, (b)–(d) partially dehydrogenated. The spectra were collected at selected points of the dehydrogenation process (under 0.1 MPa at 200 C, in the 2nd cycle). The signal intensities of dehydrogenated samples were normalized to that of the hydrogenated sample.

This effect is consistent with the continuous conversion of reactant [Mg(NH2)2 + 2LiH] into product Li2Mg (NH)2 upon dehydrogenation. Of particular interest, LiNH2 was observed through the dehydrogenation process, which shows an initial intensity increase followed by intensity decrease. The structural similarity of LiNH2 to Li2Mg(NH)2 suggests that the identified LiNH2 should act as an important reaction intermediate. Similar speculation was also proposed by Hu et al.26 The proposed LiNH2 mediation (that is, LiNH2 formation is an elementary reaction step) suggests that incorporating foreign LiNH2 may enhance the inherent reaction. To verify the favorable effect of LiNH2, we prepared and examined the Li-Mg-N-H samples containing respective LiNH2 and Li3N additive with regard to dehydrogenation performance. As shown in Fig. 7, it is obvious that the two samples exhibit similar desorption behavior. This result further supports the fact that LiNH2 should be responsible for the capacity enhancement obtained by Li3N addition. In fact, mass spectroscopy analysis indicates that the advantage of Li3N addition lies not only in enhanced hydrogen capacity but also in suppressed ammonia release. As shown in Fig. 8, we compare the ammonia release from the samples with 1940

Hydrogen capacity (wt%) 5.5 5.8 6.0 6.5

FIG. 7. Hydrogen desorption performance of the Li-Mg-N-H samples containing respective LiNH2 and Li3N additive under 0.1 MPa at desired temperatures in the 2nd cycle: (a) 200 C and (b) 180 C.

FIG. 8. Mass spectroscopy results of the gas released from the LiMg-N-H samples containing different additives: (a) 10 mol% LiNH2 and (b) 10 mol% Li3N. Hydrogen (m/e = 2, open square) and ammonia (m/e = 17, closed circle) were monitored using MS. Ammonia signals were multiplied by 20 for clarity.

LiNH2 and Li3N additive using MS. In terms of the relative amount of ammonia released, less ammonia is observed for the sample with Li3N additive. According to previous results, the improved hydrogen purity should be primarily attributed to the presence of excess LiH,



which was incorporated by Li3N to serve as an effective ammonia trapping agent.27,28 It has been clarified that the kinetic barriers in the LiMg-N-H system can be ascribed to the following: (i) the interface reaction of reactants and (ii) mass transport through the product layer.14 Li2Mg(NH)2 seeding has been reported to be effective for lowering the kinetic barrier associated with the product nucleation by incorporating foreign Li2Mg(NH)2 seed.21,23 In this work, the in situ-incorporated LiNH2 may serve as well dispersed nuclei sites for the growth of intermediate LiNH2. As a result of the lowered kinetic barrier associated with the intermediate nucleation, the LiNH2 seeding drives the dehydrogenation toward completion. Following this mechanistic understanding, the effect of Li3N additive on the property enhancement of present system is understandable. This mechanism agrees well with the capacity dependence on the Li3N addition. On the one hand, the majority of the sample is effectively activated by the seeding effect of incorporated LiNH2. On the other hand, the stable LiNH2 composition inherently lowers the theoretical capacity of Li-Mg-N-H matrix. The combination of the two inverse effects is that adding LiNH2 first improves the practical capacity, but further addition counteracts its capacity contribution. In addition, the stable enhancement upon cycling can be understood from the stability of LiNH2 composition in the de-/ hydrogenation reactions. V. CONCLUSIONS

Our screening study demonstrates that Li3N additive is effective in enhancing the practical H-capacity of LiMg-N-H system. Optimally, incorporation of 10 mol% Li3N additive improves the capacity from 3.9 wt% to approximately 4.7 wt%, leading to a 20% increment. Moreover, the capacity enhancement is highly stable during hydrogen cycling. XRD examinations indicate that Li3N additive transforms into LiNH2 and LiH phases and remains in the de-/rehydrogenation cycles. Combined structure/property investigations suggest that the in situ-formed LiNH2 should be responsible for the capacity enhancement. It is presumed that the incorporated LiNH2 acts as stable nuclei sites for promoting the growth of intermediate LiNH2. In addition, MS analysis of the effluent gas reveals that the concurrent incorporation of LiH is effective for reducing the ammonia byproduct. Our finding yields interesting clues to the reaction pathway of the Li-Mg-N-H system and emphasizes the guidance of mechanistic understanding in selecting appropriate additives. ACKNOWLEDGMENTS

The financial support for this research from the Hundred Talents Project of Chinese Academy of Sciences,

the National Natural Science Foundation of China (Grant Nos. 50571099, 50671107, and 50771094), and the National High-Tech Research and Development Program of China (863 Program; Grant No. 2006AA05Z104) is gratefully acknowledged. REFERENCES 1. L. Schlapbach and A. Zu¨ttel: Hydrogen-storage materials for mobile applications. Nature 414, 353 (2001). 2. K.H.J. Buschow: Handbook on the Physics and Chemistry of Rare Earths, Vol. 6 (North-Holland, New York, 1984), Ch. 47. 3. R.B. Schwarz: Hydrogen storage in magnesium-based alloys. MRS Bull. 24, 40 (1999). 4. W. Grochala and P.P. Edwards: Thermal decomposition of the non-interstitial hydrides for the storage and production of hydrogen. Chem. Rev. 104, 1283 (2004). 5. F. Schu¨th, B. Bogdanovic´, and M. Felderhoff: Light metal hydrides and complex hydrides for hydrogen storage. Chem. Commun. 2249 (2004). 6. S. Orimo, Y. Nakamori, J.R. Eliseo, A. Zu¨ttel, and C.M. Jensen: Complex hydrides for hydrogen storage. Chem. Rev. 107, 4111 (2007). 7. P. Chen, Z.T. Xiong, J.Z. Luo, J.Y. Lin, and K.L. Tan: Interaction of hydrogen with metal nitrides and imides. Nature 420, 302 (2002). 8. Z.T. Xiong, G.T. Wu, J.J. Hu, and P. Chen: Ternary imides for hydrogen storage. Adv. Mater. 16, 1522 (2004). 9. W.F. Luo: (LiNH2-MgH2): A viable hydrogen storage system. J. Alloys Compd. 381, 284 (2004). 10. H.Y. Leng, T. Ichikawa, S. Hino, N. Hanada, S. Isobe, and H. Fujii: New metal-N-H system composed of Mg(NH2)2 and LiH for hydrogen storage. J. Phys. Chem. B 108, 8763 (2004). 11. Y. Nakamori, G. Kitahara, K. Miwa, S. Towata, and S. Orimo: Reversible hydrogen-storage functions for mixtures of Li3N and Mg3N2. Appl. Phys. A 80, 1 (2005). 12. J. Rijssenbeek, Y. Gao, J. Hanson, Q. Huang, C. Jones, and B. Toby: Crystal structure determination and reaction pathway of amide-hydride mixtures. J. Alloys Compd. 454, 233 (2008). 13. Y. Chen, C.Z. Wu, P. Wang, and H.M. Cheng: Structure and hydrogen storage property of ball-milled LiNH2/MgH2 mixture. Int. J. Hydrogen Energy 31, 1236 (2006). 14. P. Chen, Z.T. Xiong, L. Yang, G. Wu, and W. Luo: Mechanistic investigations on the heterogeneous solid-state reaction of magnesium amides and lithium hydrides. J. Phys. Chem. B 110, 14221 (2006). 15. J. Wang: DOE Program Annual Review. Available at: http:// www.hydrogen.energy.gov/annualreview05storage.html#metal (2005). 16. Y. Kojima and Y. Kawai: IR characterizations of lithium imide and amide. J. Alloys Compd. 395, 236 (2005). 17. Z.T. Xiong, J.J. Hu, G.T. Wu, P. Chen, W.F. Luo, K. Gross, and J. Wang: Thermodynamic and kinetic investigations of the hydrogen storage in the Li-Mg-N-H system. J. Alloys Compd. 398, 235 (2005). 18. J. Yang, A. Sudik, and C. Wolverton: Activation of hydrogenstorage materials in the Li-Mg-N-H system: Effect on storage properties. J. Alloys Compd. 430, 334 (2007). 19. P. Chen, Z. Xiong, G. Wu, Y. Liu, J. Hu, and W. Luo: Metal-N-H systems for the hydrogen storage. Scr. Mater. 56, 817 (2007). 20. S. Luo, T.B. Flanagan, and W. Luo: The effect of exposure of the H-storage system (LiNH2+MgH2) to water-saturated air. J. Alloys Compd. 440, L13 (2007).


1941


21. A. Sudik, J. Yang, D. Halliday, and C. Wolverton: Kinetic improvement in the Mg(NH2)2-LiH storage system by product seeding. J. Phys. Chem. C 111, 6568 (2007). 22. Y. Liu, J. Hu, Z. Xiong, and G. Wu: Improvement of the hydrogen-storage performances of Li-Mg-N-H system. J. Mater. Res. 22, 1339 (2007). 23. J. Yang, A. Sudik, D.J. Siegel, D. Halliday, A. Drews, R.O. Carter, C. Wolverton, G.J. Lewis, J.W.A. Sachtler, J.J. Low, S.A. Faheem, D.A. Lesch, and V. Ozolins: A self-catalyzing hydrogen-storage material. Angew. Chem. Int. Ed. 47, 882 (2008). 24. J. Hu, Y. Liu, G. Wu, Z. Xiong, Y.S. Chua, and P. Chen: Improvement of hydrogen storage properties of the Li-Mg-N-H system by addition of LiBH4. Chem. Mater. 20, 4398 (2008).

1942

25. Y. Nakamura, S. Hino, T. Ichikawa, H. Fujii, H.W. Brinks, and B.C. Hauback: Dehydrogenation reaction of Li-Mg-N-H systems studied by in situ synchrotron powder x-ray diffraction and powder neutron diffraction. J. Alloys Compd. 457, 362 (2008). 26. J. Hu, Y. Liu, G. Wu, Z. Xiong, and P. Chen: Structural and compositional changes during hydrogenation/dehydrogenation of the Li-Mg-N-H system. J. Phys. Chem. C 111, 18439 (2007). 27. Y. Hu and E. Ruckenstein: Ultrafast reaction between LiH and NH3 during H2 storage in Li3N. J. Phys. Chem. A 107, 9737 (2003). 28. T. Ichikawa, N. Hanada, S. Isobe, H. Leng, and H. Fujii: Mechanism of novel reaction from LiNH2 and LiH to Li2NH and H2 as a promising hydrogen storage system. J. Phys. Chem. B 108, 7887 (2004).
