(109a) Molecular Modeling of Nanoporous Carbons: Atomistic Models and Simulated Adsorption

Palmer, J. C. - Presenter, Princeton University
Brennan, J. K. - Presenter, U.S. Army Research Laboratory
Gubbins, K. E. - Presenter, North Carolina State University

Disordered nanoporous carbons (DNCs) are highly heterogeneous porous materials that have found use in many industrial processes. In addition to their nanoscale porous features and high specific surface areas, which often lead to superior performance in many separations and storage-based applications, large scale implementations of these materials benefit from their low cost of production since they are synthesized from abundant precursors, such as woods, coconut shells, coals and crystalline carbides. While properties such as the pore size distribution (PSD) and specific surface area (SSA) may be modulated through selection of the precursor material and synthesis conditions, complete optimization of DNCs for specific applications remains a formidable challenge due to difficulty in fully characterizing their structural features and the inherently microscopic nature of the phenomena (e.g., adsorption, diffusion and reaction) that take place within their confining porous features. These challenges are compounded in many real-world applications where multiple phenomena occur in concert.

Molecular simulations methods can provide complementary information to experimental studies in the ongoing effort to accurately characterize the structural properties of DNCs and understand the impact that these features have on the thermodynamic and dynamic behavior of confined guest phases. We discuss recent advances in molecular simulation methods designed to model these complex materials, with particular emphasis placed on the those that are based in statistical mechanics (e.g., Palmer et al., Carbon, 47, 2904 (2009); Carbon, 48, 1116 (2010)). New methods for examining the adsorptive behavior in materials with tortuous porous features will also be demonstrated. We show that these methods give insight into the underlying adsorption mechanisms of pure components in model DNCs and that the application of these methods can be extended to mixtures and used to develop design criteria for new materials with improved separations capabilities. Finally, we identify and discuss the remaining challenges that must be overcome in order to develop models which are fully capable of providing quantitative descriptions of the adsorptive characteristics of DNCs, and show preliminary results obtained from improved methods designed to address these issues.