Bjørner, Martin Gamel4; Shapiro, Alexander1; Kontogeorgis, Georgios1
1 Center for Energy Resources Engineering, Center, Technical University of Denmark2 Department of Chemical and Biochemical Engineering, Technical University of Denmark3 CERE – Center for Energy Ressources Engineering, Department of Chemical and Biochemical Engineering, Technical University of Denmark4 Department of Chemistry, Technical University of Denmark
The applicability of the Multicomponent Potential Theory of Adsorption (MPTA) for prediction of the adsorption equilibrium of several associating binary mixtures on different industrial adsorbents is investigated. In the MPTA the adsorbates are considered to be distributed fluids subject to an external potential field emitted by the adsorbent. In this work, the theory is extended to include the Cubic-Plus-Association (CPA) equation of state (EoS), for the description of the fluid-fluid interactions of associating mixtures. The Dubinin-Radushkevich-Astakhov (DRA) potential function is utilized to describe the solid-fluid interactions. The potential is extended to include adsorbate-absorbent specific capacities rather than an adsorbent specific capacity. Correlations of pure component isotherms are generally excellent with individual capacities, although adsorption on silicas at different temperatures still poses a challenge. The quality of the correlations is usually independent on the applied EoS. Predictions for binary mixtures indicate that the MPTA+SRK is superior when adsorption occurs on non-polar or slightly polar adsorbents, while MPTA+CPA performs better for polar adsorbents, or when the binary mixtures only contain associating compounds. Predictions are typically improved by about 3% when individual capacities are employed, but improvements can in some cases be as large as 45%. When individual capacities and the best performing EoS are used, average absolute deviations of the selectivity are as low as 7-12%. Predictions of the selectivity are generally superior to predictions of the adsorbed amounts. The sensitivity of the model has also been tested, and it is concluded, that predictions are very sensitive to the adsorption energies.
Industrial and Engineering Chemistry Research, 2013, Vol 52, Issue 7, p. 2672-2684