Modeling Adsorption in Complex Structures: Use of Finely-Discretized Lattice-Gas DFT to Study the Effects of Pore Length, Shape, and Surface Roughness On Adsorption of Simple Gases

Monday, November 9, 2009: 9:12 AM
Delta Ballroom B (Gaylord Opryland Hotel)

Daniel W. Siderius, Department of Chemistry, Washington University in St. Louis, St. Louis, MO
Lev D. Gelb, Department of Chemistry, Washington University in St. Louis, St. Louis, MO

It has been long understood that pore surface geometry, pore length, and adsorbent material structure play a large role in determining the characteristics of gas adsorption and capillary phenomena in porous materials. Computational models of gas adsorption often involve idealizations that fail to accurately capture experimentally-observed behavior. Density functional theory (DFT) is often used to model adsorption of simple fluids in porous materials, but practically is limited to idealized pores. Conventional DFT requires a very fine discretization of the fluid density distribution to be both accurate and numerically stable[1]. The resultant computational cost precludes the use of DFT in systems with multiple directions of asymmetry. Consequently, conventional DFT is not a practical tool for studying finite-length pores or materials with explicitly rough surfaces.

We have developed a version of nonlocal DFT applicable to finely-discretized lattice fluids that retains elements of both the coarse-grained DFT of Kierlik et al.[2-4] and off-lattice conventional DFT[5]. This DFT is well-suited to the study of fluids in asymmetric materials and small-scaled confinement that cannot be practically modeled with either conventional DFT or coarse-grained lattice-gas DFT. Additionally, it can quantitatively model light gas adsorption in atomistic model of molecular adsorbents and properly describe hysteresis in finite-length pores.

We apply this finely-discretized lattice DFT to study the effect of various pore geometries on adsorption in porous materials. We begin by revisiting the problem of open-ended finite-length pores and discuss desorption hysteresis, pore neck effects, and the oft-asked question of how long a pore need be to be treated as "infinite." Following, we study adsorption in pores with features not present in idealized slits and cylinders, such as closed ends, variable-diameters, and beveled openings. Finally, we go on to study adsorption in pores with rough surfaces and examine the effect of surface roughness on monolayer formation and structure.

[1] Frink and Salinger, J. Comp. Phys., 159, 407 (2000)

[2] Kierlik et al., Phys. Rev. Lett., 74, 4361 (1995)

[3] Kierlik et al., Mol. Phys., 95, 341 (1998)

[4] Kierlik et al., Phys. Rev. Lett., 87, 055701 (2001)

[5] Siderius and Gelb, Langmuir, 25, 1296 (2009)

Extended Abstract: File Not Uploaded