Towards the design of novel boron- and nitrogen-substituted ammonia-borane and bifunctional arene ruthenium catalysts for hydrogen storage.

Electronic-structure density functional theory calculations have been performed to construct the potential energy surface for H2 release from ammonia-borane, with a novel bifunctional cationic ruthenium catalyst based on the sterically bulky β-diketiminato ligand (Schreiber et al., ACS Catal. 2012, 2, 2505). The focus is on identifying both a suitable substitution pattern for ammonia-borane optimized for chemical hydrogen storage and allowing for low-energy dehydrogenation. The interaction of ammonia-borane, and related substituted ammonia-boranes, with a bifunctional η(6)-arene ruthenium catalyst and associated variants is investigated for dehydrogenation. Interestingly, in a number of cases, hydride-proton transfer from the substituted ammonia-borane to the catalyst undergoes a barrier-less process in the gas phase, with rapid formation of hydrogenated catalyst in the gas phase. Amongst the catalysts considered, N,N-difluoro ammonia-borane and N-phenyl ammonia-borane systems resulted in negative activation energy barriers. However, these types of ammonia-boranes are inherently thermodynamically unstable and undergo barrierless decay in the gas phase. Apart from N,N-difluoro ammonia-borane, the interaction between different types of catalyst and ammonia borane was modeled in the solvent phase, revealing free-energy barriers slightly higher than those in the gas phase. Amongst the various potential candidate Ru-complexes screened, few are found to differ in terms of efficiency for the dehydrogenation (rate-limiting) step. To model dehydrogenation more accurately, a selection of explicit protic solvent molecules was considered, with the goal of lowering energy barriers for H-H recombination. It was found that primary (1°), 2°, and 3° alcohols are the most suitable to enhance reaction rate.