A METHOD FOR PREDICTING THE DIFFUSION COEFFICIENT BASED ON THE PHYSICAL CHARACTERISATION OF POROUS CARBONS BY ADSORPTION

摘要

Diffusion is important in many applications of nanoporous carbons, including in gas separation, catalysis, gas storage, controlled release and electrochemical energy storage. It is, therefore, important to be able to model diffusion of fluids in nanoporous carbons when seeking to design them and associated processes in these and other contexts. The long standing approach to modelling diffusion in porous solids involves averaging a pore-level transport model over the pore size distribution (PSD) and then using the so-called tortuosity~1 to account for pore system topology and other features omitted from the averaging (see, for example, ref 2). Diffusion experiments are the main means used to determine tortuosities. This has a number of disadvantages – there is currently no commercially available apparatus, the experiments are challenging and potentially time consuming, and a sufficient quantity of the carbon must exist. By combining molecular simulation at the single-pore level with a pore network (PN) model whose physical characteristics are determined using equilibrium experiments, these issues can be largely circumvented. This paper briefly describes such a method along with its initial assessment.