(198c) Rational Design Of Shape Selective Separation And Catalysis: Entropic Contributions And Portal Flexibility

The use of zeolites as molecular sieves and catalysts has today been well established in a wide variety of processes. However, almost the totality of applications involves a very small number of nearly circular structures, like Linde Type A, Faujasite and ZSM-5. These structures are usually modified to meet the specific needs of each process. Modification techniques, such as ion exchange or coke deposition, usually result in a distribution of pore sizes and shapes, something that retards the ability of the zeolite to be highly selective. On the other hand, there is a great variety of natural and synthetic zeolites that have been developed, but no significant effort is being made to find potential catalysis and separation applications for them. There can very well be existing structures that are highly selective in their unmodified state, or requiring a small amount of modification, just because their windows happen to be of the proper size and shape. The principal aim of this work is to develop a systematic computational framework that can identify such zeolite structures and provide researchers with a rigorous way to determine the best candidate portals for the process of their interest.

In previous work [1,2,3], we introduced a method that has an energetic basis and is based on the concept of Strain Index, which is a measure of the distortion needed for a given molecule to penetrate through a given portal. According to this approach, a molecule and a portal are sets of soft spheres that can be squeezed so as penetration to take place. An optimization framework was developed that robustly calculates the host / guest conformation that exhibits the least distortion. Given that different molecules require different amounts of distortion (expressed as activation energies), one can use the strain index results to calculate selectivities between sets of molecules, and identify the most potential structures to be used in separation or catalysis applications.

The optimization procedure that was followed aimed at locating the globally optimal solution and the framework was based on the inherent assumption that only global minimum energy conformations do lead to actual penetration. In the improved approach, we relax this assumption by taking into account all identified locally optimal conformations. In order to do this, we conduct a sufficiently thorough local search that allows us to map the energetic landscape of the interactions. By assuming uniform distribution of initial conformations, as the molecule approaches the portal, we can link the frequency of occurrence of a local optimum with the actual size of the corresponding basin of attraction. Thus, we can quantify the entropy contribution due to multiple orientations of the penetrating conformers and calculate expectations of quantities that are of interest, such as Strain Indices, equilibrium concentrations or selectivities between two different molecules [4,5]. Since the equilibrium sorption concentration of a molecule inside the zeolite is given by the free energy of adsorption, which involves the entropy, the new results are expected to improve the accuracy of the predictions.

In a second improvement to our framework, we also addressed the potential flexibility of the portal structure. This was achieved by augmenting the model with additional variables to represent the, formerly fixed, O and T atom coordinates and allowing some variation around their nominal values. The softness of the O-O and T-O bonds was described by a standard quadratic potential that was added to the objective function.

[1] C. E. Gounaris, C. A. Floudas, and J. Wei, Rational Design of Shape Selective Separation and Catalysis: I. Concepts and Analysis. Chemical Engineering Science, 61, 7933-7948 (2006).

[2] C. E. Gounaris, J. Wei, and C. A. Floudas, Rational Design of Shape Selective Separation and Catalysis: II. Mathematical Model and Computational Studies, Chemical Engineering Science, 61, 7949-7962 (2006).

[3] J. Wei, C. A. Floudas, and C. E. Gounaris, Frontiers of Shape Selective Separations, Proceedings of the Eleventh International Conference on Properties and Phase Equilibria for Product and Process Design (PPEPPD 2007), To appear (2007).

[4] J.L. Klepeis and C.A. Floudas, Free Energy Calculations for Peptides via Deterministic Global Optimization, Journal of Chemical Physics, 110, 7491-7512 (1999).

[5] C. E. Gounaris, J. Wei, and C. A. Floudas, Calculating the Effect of Locally Optimal Conformations in Shape Selective Separation and Catalysis, In preparation (2007).