An empirical valence bond potential for aqueous hydrochloric acid systems

After the discovery of the stratospheric ozone hole, reactions involving hydrochloric acid (HCl) adsorbed on the surface of ice crystals were identified as playing an important role in ozone depletion. The first step of these reactions is believed to be ionization of the acid molecule, so there has been considerable interest in developing a detailed understanding of HCl/ice interactions. Any attempt at examining these interactions by computer simulation requires a model potential which allows for solvation and dissociation of the HCl.; In this dissertation I present a new empirical valence bond (EVB) model potential capable of describing bond stretching, breaking and formation on ice particle surfaces, and in HCl·(H2O)n clusters. The model is parameterized using an existing excess proton in water EVB model, and accurate ab initio calculations on the HCl·H 2O dimer. I demonstrate that the new potential is capable of accurately representing trends in minimum energy geometry, HCl vibrational frequency shifts, and relative energetics with water cluster size for a set of small water clusters that have recently been explored by various workers in high level ab initio studies. The model is also capable of reliably describing relative energetics of thermally accessible local minima observed in these ab initio studies. I also show that the potential can capture changes to the HCl caused by differing hydrogen bonding arrangements similar to those found on the surface of ice crystallites, and thus should be a useful tool for understanding experimental results. In addition, the potential can be computed efficiently, allowing for simulation of large realistic systems.; The temperature dependence of HCl bond length and dissociation in different settings---small clusters of 1--7 water molecules and larger ice balls containing 48 water molecules---is studied using both classical and path integral Monte Carlo calculations. The importance of quantum nuclear tunneling, dispersion, and zero-point energy effects on molecular ionization in these low temperature proton transfer systems is explored.; Finally, I address a way to generalize the potential---adding the ability to examine concentration effects by implementing a new improvement to the EVB method which allows for the treatment of clusters with multiple HCl molecules.