Impact of Stoichiometry on the Hydrogen Storage Properties of LiNH2-LiBH4-MgH2 Ternary Composites

We recently reported (Yang, J.; et al. Angew. Chem., Int. Ed. 2008, 47, 882) a novel hydrogen storage composite involving a 2:1:1 LiNH2:LiBH4:MgH2 ratio. On the basis of in-depth experimental and computational analysis, this composite was found to release hydrogen via a complex multistep reaction cascade, which seeded the products of a subsequent reversible hydrogen storage reaction. This so-called autocatalytic reaction sequence was found to result in favorable kinetics, ammonia attenuation, and partial low-temperature reversibility. Here, we extend our original study by examining the effects of reactant stoichiometry on the ensuing hydrogen storage desorption pathway and properties. In particular, we examine four (LiNH2)X-(LiBH4)Y-(MgH2)Z composites, where X:Y:Z ) 2:1:2, 1:1:1, 2:0.5:1, and 2:1:1 (original stoichiometry). For each sample, we characterize the postmilled mixtures using powder X-ray diffraction (PXRD) and infrared spectroscopy (IR) analyses and observe differences in the relative extent of two spontaneous milling-induced reactions. Variabletemperature hydrogen desorption data subsequently reveal that all composites exhibit a hydrogen release event at rather low temperature, liberating between 2.3 (1:1:1) and 3.6 (2:0.5:1) wt % by 200 °C. At higher temperatures (200-370 °C), the hydrogen release profiles differ considerably between composites and release a total of 5.7 (1:1:1) to 8.6 (2:0.5:1) wt %. Utilizing variable-temperature IR and PXRD data coupled with first-principles calculations, we propose a reaction pathway that is consistent with the observed phase progression and hydrogen desorption properties. From these data, we conclude that premilled reactant stoichiometry has a profound impact on reaction kinetics and high-temperature reaction evolution because of reactant availability. From this enhanced understanding of the desorption process, we recommend and test a stoichiometrically optimal ratio (3:1:1.5) which releases a total of 9.1 wt % hydrogen. Finally, we assess the reversibility (at 180 °C) of the four primary composites over two desorption cycles and find that only the 2:1:1 and 2:0.5:1 are reversible (3.5 wt % for 2:0.5:1).