(665a) Computer Simulation of Adsorption and Molecular Recognition Phenomena in Imprinted Polymers

Polymerization in the presence of templates, followed by their consequent removal, leads to structures with cavities capable of molecular recognition. This molecular imprinting technology has been employed to create porous polymers with tailored selectivity for adsorption, chromatographic separations, sensing and other applications. Performance of these materials crucially depends on the availability of highly selective binding sites. This parameter is a function of a large number of processing conditions and is difficult to control. Furthermore, the nature of molecular recognition processes in these materials is poorly understood to allow a more systematic design.

In this talk we present a range of models of molecularly imprinted polymers mimicking the actual process of their formation. We start with a simple model where individual species are represented as hard spheres or rigid clusters of hard spheres, some of them also featuring small association sites. We demonstrate that a range of molecular recognition effects emerge in this model and that they are qualitatively consistent with the experimental observations. The model also provides a wealth of information on how binding sites form and function in the imprinted structures.

Quantitative prediction of adsorption in imprinted polymers requires a more detailed model. In the next part of the presentation, we show how limited information about the structure of the real imprinted polymers from physical adsorption experiments can be used to construct atomistically detailed models, based on realistic intermolecular potentials. We apply this approach to investigate adsorption of aromatic species, such as pyridine, benzene and toluene, in model imprinted polymers and discuss the implications of the results.