Premium
Non‐Additivity of Attractive Potentials in Modeling of N 2 and Ar Adsorption Isotherms on Graphitized Carbon Black and Porous Carbon by Means of Density Functional Theory
Author(s) -
Ustinov Eugene A.,
Do Duong D.
Publication year - 2004
Publication title -
particle and particle systems characterization
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.877
H-Index - 56
eISSN - 1521-4117
pISSN - 0934-0866
DOI - 10.1002/ppsc.200400924
Subject(s) - adsorption , argon , thermodynamics , carbon black , density functional theory , additive function , activated carbon , intermolecular force , monolayer , chemistry , carbon fibers , nitrogen , porosity , materials science , computational chemistry , organic chemistry , molecule , physics , composite material , mathematical analysis , biochemistry , natural rubber , mathematics , composite number
We present a new approach accounting for the non‐additivity of attractive parts of solid–fluid and fluid–fluid potentials to improve the quality of the description of nitrogen and argon adsorption isotherms on graphitized carbon black in the framework of non‐local density functional theory. We show that the strong solid–fluid interaction in the first monolayer decreases the fluid–fluid interaction, which prevents the two‐dimensional phase transition to occur. This results in smoother isotherm, which agrees much better with experimental data. In the region of multi‐layer coverage the conventional non‐local density functional theory and grand canonical Monte Carlo simulations are known to over‐predict the amount adsorbed against experimental isotherms. Accounting for the non‐additivity factor decreases the solid–fluid interaction with the increase of intermolecular interactions in the dense adsorbed fluid, preventing the over‐prediction of loading in the region of multi‐layer adsorption. Such an improvement of the non‐local density functional theory allows us to describe experimental nitrogen and argon isotherms on carbon black quite accurately with mean error of 2.5 to 5.8% instead of 17 to 26% in the conventional technique. With this approach, the local isotherms of model pores can be derived, and consequently a more reliable pore size distribution can be obtained. We illustrate this by applying our theory against nitrogen and argon isotherms on a number of activated carbons. The fitting between our model and the data is much better than the conventional NLDFT, suggesting the more reliable PSD obtained with our approach.