Modelization of the $$\hbox {H}_{2}$$H2 adsorption on graphene and molecular dynamics simulation

In the search for efficient molecular dynamics simulation models both simplicity and acceptable accuracy matter. In the present study, a model of the graphene-\(\hbox {H}_2\) physisorption system is used to explore its performance and limitations under canonical NVT and microcanonical NVE simulation conditions. The model implies several simplifications that can be summarized in (a) a single ideal planar frozen graphene-like layer of C atoms, (b) rigid rotor \(\hbox {H}_2\) molecules and (c) interaction potentials written as C–H2 and \(\hbox {H}_2\)–\(\hbox {H}_2\) site–site Improved Lennard-Jones potentials parameterized to reproduce DFT calculations. This model can be used in a variety of molecular dynamics simulation conditions, both in NVT and NVE ensembles. Such simulations lead to the formation of a single layer of adsorbed \(\hbox {H}_2\) molecules in dynamically stable equilibrium with a fluid-phase region. In addition, the incipient formation of secondary layers for high-density conditions is also observed. Some properties as average pressure, temperatures and fluid-phase densities are discussed as well as possible improvements of the model.