Dehydrogenation Mechanism of Monoammoniated Lithium Amidoborane [Li(NH 3 )NH 2 BH 3 ] S. Bhattacharya, †,# Zhitao Xiong, ‡ Guotao Wu, ‡ Ping Chen, ‡ Y. P. Feng, § C. Majumder, ∥ and G. P. Das* ,† † Department of Materials Science, Indian Association for the Cultivation of Science, Kolkata 700 032, India ‡ Dalian Institute of Chemical Physics, Dalian, China § Department of Physics, National University of Singapore, Singapore ∥ Chemistry Division, Bhabha Atomic Research Centre, Mumbai 400 085, India * S Supporting Information ABSTRACT: Monoammoniated lithium amidoborane has been experimentally synthesized. When this is heated to a temperature of 40−55 °C, this releases 9−11 wt % hydrogen. First-principles density functional calculations have been carried out to understand the underlying mechanism of dehydrogenation. Theoretical results predict that the reaction is a three-step process; each step consists of 3.7, 3.9, and 4.0 wt % H 2 uptake with an altogether capacity of 12 wt % dehydrogenation. Whereas the first dehydrogenation is a direct interaction between lithium amidoborane and NH 3 monomers, the subsequent reaction steps lead to further dehydrogenation, provided that the activation barrier falls within reasonable limits, and this has been achieved by forming higher-order nanoclusters of [Li(NH 2 )- NH 2 BH 3 ] n . 1. INTRODUCTION Among the light metal complex hydrides, ammonia borane (NH 3 BH 3 ) [AB] is considered to be one of the most promising hydrogen storage materials because of its high hydrogen content (19.6 wt %) capacity. 1 The pristine AB molecule contains both hydridic B−H and protic N−H bonds and a strong B−N bond so that hydrogen release from solid AB is more favorable than dissociation to ammonia and diborane under most conditions. 1 However, subsequent release of hydrogen with the increase of temperature leads to the generation of volatile toxic species, such as borazine, 2 which can poison the fuel cells. This results in a poor H-release kinetics and dehydrogenation mechanism from pristine AB, which needs further improvement for its effective practical implementations in on-board H-storage devices. One way to improve the performance of AB is substituting one H atom in the [NH 3 ] unit by metals, such as Li, Na, etc., to form lithium amidoborane (LiNH 2 BH 3 )[LiAB] and sodium amidoborane (NaNH 2 BH 3 )[NaAB] with a gravimetric efficiency of 10.9 and 7.5 wt %, respectively. 3 These materials have also been highlighted as some of the best potential hydrogen-storage materials in the 2008 DOE hydrogen program annual progress report. 4 With more electrons being donated from metal to [NH 2 BH 3 ] − ions, the hydridic B−H bond of [NH 2 BH 3 ] − ions is elongated, which enhances its activity as compared with those in pure AB. Therefore, the reaction barrier between [NH 2 BH 3 ] − ions would be lower than that between two neutral NH 3 BH 3 molecules. In addition, the charged [NH 2 BH 3 ] − ion creates more polar surroundings compared with the symmetric NH 3 BH 3 complex. The obvious con- sequence of this is a much faster reaction kinetics in metal− amidoboranes, compared with the same for pristine AB. 5 It has been reported that a lower dehydrogenation temperature (≈90 °C) can be achieved in LiAB compared with pristine AB (≈110 °C). 3 To improve the operating properties of these materials, such as rapid H 2 release near room temperature, it is vital to understand the underlying mechanism for the release of H 2 . What is known is that the [NH 2 BH 3 ] − unit attracts the metal cation Li + , thereby enhancing the reactivity of the hydritic B−H bond. This results in a BH−NH interaction between the adjacent units of LiAB. The obvious consequence of this is the reduction of the lowered dehydrogenation temperature. 5 Recently, Xia et al. have experimentally synthesized mono- ammoniated LiAB [Li(NH 3 )NH 2 BH 3 ], which shows more favorable dehydrogenation characteristics in the temperature range of 40−70 °C. 6 However, there is still no acknowledged model calculation to explain the detail decomposition mechanism following this experimental finding. To theoretically model this reaction mechanism, one has to find reaction pathways, where the activation barrier lies in the range of 20− 25 kcal/mol, which is acceptable for gas-phase calculations. Received: June 15, 2011 Revised: March 15, 2012 Published: March 16, 2012 Article pubs.acs.org/JPCC © 2012 American Chemical Society 8859 dx.doi.org/10.1021/jp210315u | J. Phys. Chem. C 2012, 116, 8859−8864