Abstract

 

 

 

 

A STUDY OF MOLECULAR PROCESSES IN CALCIUM CLUSTERS, SARIN, AND CALCIUM CARBONATE CLUSTERS

 

Jeffrey W. Mirick, Ph.D.

George Mason University, 2002

Dissertation Director:  Professor Estela Blaisten-Barojas

 

Extensive quantum mechanical calculations have been performed on systems of atoms and molecules to include the study of calcium clusters, the degradation of sarin with OH, and the study of calcium carbonate clusters. 

The electronic structure of calcium clusters containing up to 13 atoms was studied within the General Gradient Approximation (GGA) of the density-functional formalism. For the calcium dimer it is found that the exchange functional in GGA overestimates the binding energy, while a hybrid approach including Hartree-Fock exchange gives a better agreement with the experimental results. Binding energies, optimized geometries, vibrational frequencies, and thermodynamic properties have been calculated for several isomers at each cluster size. Various structures corresponding to saddle points of the energy curve are reported, along with the isomerization reaction path for Ca₅, Ca₆, and Ca₇. It was found that Ca₁₂ undergoes a structural transition as a function of temperature, changing structure at T=318 K. A comparison of the minimum energy isomer geometry and binding energy obtained for each cluster size with those obtained from the Murrell-Mottram empirical potential shows that this potential overestimates the binding energies and does not adequately predict the optimized structures for several cluster sizes.

The breakdown of Sarin (GB), isopropyl methylphosphonofluoridate, by H₂O and OH has been investigated using a molecular orbital approach, an extended basis set, and within the Hartree-Fock approximation.  It was found that when Sarin is exposed to water, the bond between the fluorine and phosphorus atoms breaks and the water molecule dissociates into OH and atomic hydrogen.  Two new bonds are formed between the OH radical and the phosphorus atom and between the freed fluorine and hydrogen atoms giving rise to isopropylmethylphosphonic acid and HF.  When the GB degradation is attempted with OH, there are two alternative reaction paths. In one, HF and the radical CH₃PO[OCH(CH₃)₂]O are formed.  In the other reaction an intermediate radical is formed in which the hydroxyl sits in the former site of the fluorine atom and the latter moves to a position such that the P-F bond is almost perpendicular to the plane of the three oxygen atoms.  Once this intermediate radical is formed, it proceeds by unimolecular decomposition to give either HF and CH₃PO[OCH(CH₃)₂]O, or the abstraction of fluorine accompanied by the formation of isopropylmethylphosphonic acid.  The reaction paths of these four reactions are analyzed in detail and the transition barriers and geometry of the transition compounds are reported.  IR active frequencies of reactants and products are also report.

This ab-initio study was extended to the investigation of molecular clusters of calcium carbonate.  The growth of calcium carbonate crystals structure depends largely on the electronic interactions at the molecular level.  It is however very difficult (or impossible) with our present computer ability to perform ab-initio all electron calculations for clusters with many molecular units.  It is therefore interesting to describe the molecular interactions via a model potential.  Calcium carbonate is a molecular crystal with interactions of the ionic type.  The Rigid Ion Model (RIM) potential might therefore be a realistic model to describe this molecular crystal.  However, it is important that the parameters of such a model contain enough information about the cluster formation to adequately model the growth of calcium carbonate microcrystals.  The strategy then is to perform extensive all-electron calculations on small clusters containing up to four molecular units and then fit the parameters of the RIM to the set of energies and bond lengths calculated quantum mechanically.  Once the RIM potential is well parameterized, I studied the conformation of molecular clusters containing up to 100 molecular units of CaCO₃.  Structure, sequence of growth, and mean square displacement (MSD) are some of the quantities calculated.  It is found that the molecular clusters do not show local point symmetry consistent with the calcite lattice.  Some of the cluster sizes have peculiar symmetries.  Concerning the MSD dependence with temperature, it is clear that diffusion of CaCO₃ takes place only at temperatures above 1000 K.  Molecular dynamic simulations were carried out to calculate the MSD taking into consideration good angular momentum conservation.  Further studies will allow us to detect ways in which these CaCO₃ clusters transform into crystalline structures.