Date of Award

Winter 2009

Document Type


Degree Name

Doctor of Philosophy (PhD)


Micro and Nanoscale Systems

First Advisor

Daniela S. Mainardi


One of the major impediments in the way of the realization of hydrogen economy is the storage of hydrogen gas. This involves both the storage for stationary applications as well as that of storage onboard vehicles for transportation applications. For obvious reasons, the system targets for the automotive applications are more stringent. There are many approaches which are still being researched for the storage of hydrogen for vehicular applications. Among them are the high pressure storage of hydrogen gas and the storing of liquid hydrogen in super insulated cryogenic cylinders. While both of them have been demonstrated practically, the high stakes of their respective shortcomings is hindering the wide spread application of these methods. Thus different solid state storage materials are being looked upon as promising solutions. Metal hydrides are a class of solid state hydrogen storage materials which are formed by the reaction of metals or their alloys with hydrogen. These materials have very good gravimetric storage densities, but are very stable thermodynamically to desorp hydrogen at room temperatures. Research is going on to improve the thermodynamics and the reaction kinetics of different metal hydrides.

This dissertation tries to address the problem of high thermodynamic stability of the existing metal hydrides in two ways. First, a novel carbon based lithium material is proposed as a viable storage option based on its promising thermodynamic heat of formation. Pure beryllium (Be) clusters and the carbon-beryllium (C-Be) clusters are studied in detail using the Density Functional Theory (DFT) computational methods. Their interactions with hydrogen molecule are further studied. The results of these calculations indicate that hydrogen is more strongly physisorbed to the beryllium atom in the C-Be cluster, rather than to a carbon atom. After these initial studies, we calculated the geometries and the energies of more than 100 different carbon based lithium materials with varying amounts of hydrogen. A detailed analysis of the heats of reactions of these materials using different reaction schemes is performed and based on the promising thermodynamic and gravimetric storage density, LiC4Be2H5 is divulged as a promising novel carbon based lithium material.

In the later part, this dissertation performs a detailed study on the effect of carbon when it is used as a dopant in four different well known complex hydrides, lithium beryllium hydride (Li2BeH4), lithium borohydride (LiBH4), lithium aluminum hydride (LiAlH 4) and sodium borohydride (NaBH4). Initially, the unit cells of the crystal structure are fully resolved using the plane-wave pseudopotential implementation of DFT. The supercells of each of these are then constructed and optimized. Varying amounts of carbon is introduced as impurity in these crystals in different sites such as the top, subsurface and the bulk of the crystal lattice. Using the electronic structure calculations, it is established that (i) C-Be-H, C-B-H or C-Al-H compounds are formed respectively in the cases of Li2BeH4, LiBH4 and LiAlH4 when carbon is doped in them; (ii) and carbon dopant causes a decrease in the bond strengths of Be-H, B-H and Al-H in respective cases. This reduction in the bond strengths combined with the fact that there is a decrease in the ionic interaction between the cation and the anionic hydride units of these complex hydrides causes a destabilization effect.