Research

Adsorption Thermodynamics: Segregation and Siting Effects

Predicting the adsorption behavior of multicomponent systems from single-component isotherms is one of the primary objectives of adsorption thermodynamics. Many theoretical models have been proposed for this purpose, but in general this is still a difficult task. Such models are essential for the design of adsorption separation processes and for the rational selection and optimization of adsorbents.

In the adsorption of mixtures, it is thought that different species may segregate into different regions of an adsorbent. Knowledge of where the molecules adsorb is the most basic starting point for any fundamental description of adsorption. Nevertheless, little is known experimentally about this phenomenon because it is difficult to observe by direct experiment. Our current focus is to find out when and why segregation occurs, as well as how it affects adsorption thermodynamics in nanoporous materials.

Using molecular simulations, we have seen that even simple spherical molecules segregate into different pore environments in zeolites like mordenite. To extend this work to additional systems, we recently developed a general method for performing grand canonical Monte Carlo simulations of molecules with internal degrees of freedom. Some systems still require too much computational time, however. To overcome this, we have developed a hierarchical approach in which short atomistic simulations provide parameters that are fed into more efficient lattice models that can then be used to predict the macroscopic properties of interest.

Critical predictions from our modeling have been tested by 129Xe NMR experiments in our laboratory.

Current work is focused on liquid-phase adsorption and adsorption of chiral molecules.

Shaji Chempath, Mo Murthi, Louis Clark, Ken Czaplewski, Amit Gupta, Yi Mao, and Jui Yang have been involved in this work, which is supported by a CAREER grant from the National Science Foundation and by an industrial sponsor, UOP.