| A proper understanding of the thermodynamic and transport properties of fluids confined within nanomaterials is essential for the development of existing and emerging technologies, including but not limited to, gas separations, gas storage and the development of new classes of sensors. In this dissertation, we have used molecular simulation to study the adsorption and transport of fluids in various carbon nanostructures, namely, straight and constricted single wall carbon nanotubes, nanoporous carbon membranes and C70 fullerene crystals.; The main thrust of this research was the study of the application of single wall carbon nanotubes (SWCNs) for energy efficient separations of nitrogen or oxygen from air. Industrially, this process is usually carried out by energy-demanding cryogenic distillation. Energy efficient, and therefore economical alternatives, such as adsorption-based and membrane-based separation processes are the result of favorable differences in the equilibrium adsorption capacities and/or the transport properties of the mixture components in nanoporous media. Carbon nanotubes, with their unique properties, offer a possible addition to the existing class of nanoporous materials, such as zeolites and carbon molecular sieves, which are commonly used for such processes.; To facilitate the rational design of carbon nanotube-based separation processes, grand canonical Monte Carlo (GCMC) and equilibrium molecular dynamics (EMD) simulations are used in this thesis to predict the equilibrium adsorption-based and membrane-based transport selectivity of N2/O2 mixtures through carbon nanotubes. A hierarchical approach is followed to achieve this. GCMC simulations are used first to obtain adsorption isotherms, adsorption energies and isosteric heats of adsorption of pure N2, O2 and their mixtures in SWCNs. Equilibrium adsorption-based selectivity is directly obtained from the mixture adsorption isotherms. EMD simulations are used to obtain self-, corrected diffusivities of pure N2 and O2 and the self-diffusivities and phenomenological coefficients of N2/O2 mixtures in SWCNs. To obtain macroscopic membrane separation properties, these results are combined with the adsorption isotherms to obtain transport diffusivities, and by the use of a continuum description of mass transport to obtain single-component and mixture fluxes. From these results an understanding is obtained of the effects of nanotube diameter, pressure and temperature on adsorption and transport of N2, O 2 and their mixtures in SWCNs to achieve an optimal separation of air.; In addition to the specific analysis of N2/O2 separations using nanotubes, the approach and the simulation programs developed in this research were also extended to understand the mechanism for the higher permeation rates of oxygen relative to nitrogen in nanoporous carbon membranes (NPCs) that has been reported in experiments. The key mechanism for such large differences in the permeation rates has not been understood as nitrogen and oxygen have very similar molecular sizes and interaction energetics.; We used a crystalline periodic structure of C168 Schwarzite to represent NPCs and study the single-component transport of N2 and O2. Two types of interaction potentials are used in this research to model adsorbate-adsorbent interactions; the empirical Steele potential based on adsorption at infinite dilution on planar graphite sheets, and an ab initio-based potential that incorporates the effect of carbon curvature and the presence of nonhexagonal rings in C168 Schwarzite. The quantum mechanical potential parameters were obtained from the recently developed Hybrid Method for Interaction Energies (HM-IE), which requires considerably less computational time and computer resources than previous methods to obtain an accurate estimate of interaction energies calculated at a high level of theory with a large basis set. However, it is still computationally very demanding to obtain an accurate pote... |
