Computer Modeling of Nanoporous Materials: An ab initio Novel Approach for Silicon and Carbon

Abstract

Carbon and silicon have been consistently proposed as elements useful in the generation of porous materials. Carbon has been insistently postulated as a promising material to store hydrogen, and crystalline silicogermanate zeolites have recently been synthesized and are being considered in catalytic processes. In the present work we report an approach to generating porous materials, in particular porous carbon and silicon, which leads to the existence of nanopores within the bulk. The method consists in constructing a crystalline diamond-like supercell with 216 atoms with a density (volume) close to the real value, then halving the density by doubling the volume (50% porosity), and subjecting the resulting supercell to an ab initio molecular dynamics process at 300 K for Si, and 1000 K for carbon, followed by geometry relaxation. The resulting samples are essentially amorphous and display pores along some of the "crystallographic" directions. We report their radial distribution functions and the pore structure where prominent.

Ab initio Computationally Generated Nanoporous Carbon and its Comparison to Experiment

“…As in our previous work [1] we report the generation of porous structures using FASTSTRUCTURE [9], a density functional code based on the Harris functional that allows simulated annealing/molecular dynamics studies with quantum force calculations. The LDA parameterization of Vosko, Wilk and Nussair has been used in our simulations [10].…”

A New Approach to the Computer Modeling of Amorphous Nanoporous Structures of Semiconducting and Metallic Materials: A Review

“…[23]. Next we report simulations for other values of the porosity guided by the experimental data of Petkov et al [29] to discern what parameters are important in the simulations presented here.…”

“…There are three well defined peaks, and a minimum between the first and the second peaks as in typical amorphous materials [23]. …”

“…[23] where more detailed information can be found. Next we report simulations carried out with the Tersoff parameterized interatomic potential also for a 50% porous sample to compare with the first principles simulation.…”

Computational alternatives to generate amorphous nanoporous structures using ab initio molecular dynamics

“…The two methods to generate these precursor structures are the expanding lattice (EL) [5] approach and the dealloying (D) approach, named after the well known experimental process to generate porous metals, although there are some important differences between this computational method and the experimental procedure.…”