The use of neutron-scattering methods in atomic, and molecular level character­ization of virtually all types of potential hydrogen storage materials has provided remarkably fine detail of their structures and the associated hydrogen dynamics which would be impossible to obtain by any other single experimental technique. Hydrogen storage materials are, of course, ideal subjects for neutron scattering studies on account of the neutron’s sensitivity to hydrogen, the fact that both diffraction and spectroscopic investigations can readily performed, and that this can be done under a wide range of experimental conditions (pressure, temperature, hydrogen content) with relative ease.

In the case of complex and chemical hydrides it is the knowledge of hydrogen positions in the various crystal structures, identification of the nature of the H-species, in part by INS studies together with computation, nature of diffusive processes, either self-diffusion or aided by reorientations, which have helped to develop some tentative picture of possible discharge-processes in a number of important systems, and in some cases regeneration paths as well. A large number of important questions remain to be answered in detail, most notably perhaps, the role of the catalyst where it is necessary for the function of the material.

Neutron scattering studies of sorption-based storage systems have provided a wealth of detail of the mainly weak interactions of hydrogen molecules with the various host systems. Perhaps the most noteworthy aspect of these investigations is the remarkable sensitivity of the rotational tunnelling transition of the adsorbed H2 molecules, that can readily be observed by INS, and are found to range over more than two orders of magnitude. While we typically observe a higher barrier to rotation for a more strongly-bound molecule, we note that these barriers are not directly related to the centre-of-mass binding energies of the hydrogen molecule, but to angle-dependent interactions with the host. Nonetheless, these can be coupled by computational studies, which include the development of multi-dimensional potential energy surfaces. The effort to obtain such PESs and to understand hydrogen sorption in MOFs and other porous media by computational analysis has therefore spawned an extensive theoretical work to improve treatment of non­bonded, or mainly dispersive, interactions in DFT methods, which have long been overdue in the study of many other weakly-interacting systems.

Much of the detailed interpretation and computational analysis of some of the results described above depend on the fact the vast majority of the materials under investigation are crystalline so that the positions of the atoms can be known. These must be known, for example, as a starting point for most theoretical modelling studies. Difficulties do arise from the presence of disorder or various types, and the very large zero-point and thermal motion of the weakly bound hydrogen molecules in sorption-based systems. This problem becomes less and less tractable for storage systems that are amorphous, or for the liquid carriers, including those, which have chemical hydrides in suspension. Structural studies of amorphous and liquid sys­tems by the well-established pair distribution function (PDF) approach are therefore likely to become more common, as these approaches to hydrogen storage gain in importance. Structural results from these techniques (preferably a combination of neutron PDF for H and other light atoms, and X-ray PDF for heavier atoms) are probabilistic in nature rather than giving accurate positional parameters, but do bear a direct relationship to molecular dynamics simulations by way of comparisons of observed and calculated pair-correlation function. Spectroscopic studies by INS will suffer from inhomogeneous broadening of the lines from the disordered structures, and the fact that it is usually best done at low temperatures. Quasielastic neutron scattering may become more important in these cases as well, but may depend even more critically on molecular-dynamics simulations.

We can certainly expect that neutron scattering methods in conjunction with computational analyses will continue to play the pivotal role in the ultimate development and design of a practical, materials-based hydrogen storage system for mobile applications, which at this time is viewed as a long-range solution to this problem. In the near term, however, highly compressed hydrogen gas at room temperature is the storage medium of choice for most of the experimental, and commercial (starting in 2015) hydrogen powered fuel-cell cars, including battery hybrids, on the road.