Topologies of capillary condensation
Tóm tắt
Calculations are presented to illustrate the dependence of capillary adsorption upon the interactions present in model pores. The sequence of phase transitions at zero temperature is determined for a Lennard-Jones lattice gas in a pore consisting of 4 × 4 × ∞ sites. The dependence of the specific filling sequence upon the comparative strength of the gas-pore wall and the gas-gas interaction well-depths is determined. Grand canonical Monte Carlo simulations of sorption at finite temperature in the continuum version of the same model pore are also reported. Both the theory and the simulations were performed with variable gas-solid and gas-gas energy well-depths. At a temperature of 90 K, the gas-solid heterogeneity associated with atoms adsorbed in the corners, on the walls and in the interior pore volume gives rise to sequential adsorption similar to that observed in the lattice gas calculation at 0 K. A gradual approach to non-wetting behavior is observed as the gas-solid well-depth decreases. Values of the gas-solid well-depth needed to produce pore filling at saturation (i.e., “pore-wetting”) are discussed.
Tài liệu tham khảo
Allen, M.P. and D.J. Tildesley, Computer Simulation of Liquids, Oxford University Press, Oxford, 1987.
Bojan, M.J. and W.A. Steele, Langmuir, 5, 625 (1991).
Gregg, S.J. and K.S.W. Sing, Adsorption, Surface Area and Porosity, Second Edition Academic Press, London, 1982.
Nicholson, D. and N.G. Parsonage, Computer Simulation and the Statistical Mechanics of Adsorption, Academic Press, London, 1982. Norman, G.E. and V.S. Filinov, High Temperature (Teplofizika Vsokikh Temperatur), (translated from the Russian) 7, 216 (1969).
Steele, W.A., J. Phys. Chem., 82, 817 (1978). It is known that the difference between the periodic and bulk systems is small if the periodicity length exceeds the pore width. See Swift, M.R., E. Cheng, J.R. Banavar, and M.W. Cole, Phys. Rev., B48, 3124 (1993).
The finite temperature version of the calculations of Section 2 is feasible by transfer matrix methods but only if the number of sites in the transverse section is sufficiently small (≤10). See reference (Swift et al., 1993).