The simplified local density / Peng-Robinson model (SLD-PR) was modified to improve its predictive capability when dealing with near-critical and supercritical adsorption systems of the type encountered in coalbed methane recovery and CO 2 sequestration. The goal was to develop efficient equation-of-state (EOS) computational frameworks for representing adsorption behavior, as well as to improve our understanding of the phenomenon. The ability of the modified SLD-PR to correlate accurately data for supercritical adsorption systems is demonstrated using adsorption measurements on activated carbon, Illinois [6 coal, Fruitland coal, and Lower Basin Fruitland coal. The results indicate that the modified SLD-PR model, which incorporates a modified repulsive parameter ''b'' for the PR EOS, is capable of modeling the adsorption of pure methane, nitrogen, and CO at coalbed conditions. Inclusion of a slit geometry in the adsorbent matrix yields results 2 superior to our previous two-dimensional EOS models for the adsorbates considered. The results also indicate that accounting for the adsorption surface structure within the SLD-EOS framework is effective in improving modeling capability for high-pressure adsorption phenomena. An explanation is offered as to why the adsorbed-phase densities are close to the EOS reciprocal co-volumes. Further, the model (a) generates direct estimates for the adsorbed-phase densities (which facilitate reliable prediction of absolute gas adsorption) and (b) readily describes the observed maximum in Gibbs-adsorption isotherms of CO at the temperatures and pressures encountered in coalbeds. 2