Work presented in this thesis details the development of new applications for molecular crystalline systems using first-principles simulation. In particular work has focused on the most important type of intermediate interactions-the hydrogen bond. A new computational procedure to more accurately mimic the crystalline environment has been developed and applied to two systems: the test system ammonia and the more unusual dihydrogen bonded system BH₃NH₃. Both generated surprising results, which challenged the conventional view of bonding in the solid state. Work has also focused on the dynamics of the hydrogen bond, resulting in the implementation of a constraint molecular dynamics (MD) algorithm for the popular simulation package, CASTEP. This code development allows molecular systems to be treated as rigid or semi-rigid bodies, thus allowing appreciable increase in the first-principles MD time step. It also allows interesting chemistry to be explored at the ab-initio level, which would be inaccessible by any other route. The method has been applied to the phase I structure of ammonia and a full vibrational analysis is reported.