Atomistic Simulations of Hydrogen Storage in Carbon-based Materials

Abstract

Hydrogen technology presents a possible strategy in the present transformation towards sustainable and responsible research, development, and production. One source of hydrogen is pyrolysis, which yields large amounts of carbon as a by-product. In the present contribution, we will use computational tools to investigate the H-storage capability of carbon-based nanostructures. Firstly, we will present a strategy for constructing a realistic 3D atomistic model of nanoporous carbon (np-C) based on experimental measurements. Subsequently, we employ this model to investigate H sorption as a function of temperature and H2 gas pressure using molecular dynamics. To further enhance the storage capacity of carbon, we employ the MatlantisTM platform, together with their neural network interatomic potential, to scan through the entire periodic table and quantify how a single dopant attached to graphene locally modifies the adsorption properties of hydrogen. We probe various dopant sites (on the graphene or placed as substitutional atoms within the graphene structure) and different carbon geometries (graphene or carbon nanotubes). This approach enables the identification of dopants that significantly increase hydrogen adsorption. Our work has the ambition to steer future knowledge-based experiments on the modification of carbon for hydrogen storage.