Computational studies of silica.
Doctoral thesis, UCL (University College London).
There are three areas of research in this thesis. The �first is concerned with the silica polymorph, tridymite, with simulations carried out using three computational methods: free energy minimisation, molecular dynamics and Density Functional Theory. A number of tridymite structures with di�fferent atomic configurations have been found in nature. The simulations explore various properties of these di�fferent forms of tridymite and investigate whether it is possible to distinguish between them using the three computational techniques. It was found that the interatomic potential and simulation technique used, rather than the simulation temperature, were the main factors a�ffecting the resulting structure. There are a number of possible explanations for this result: The techniques may not be sensitive enough to deal with an energy landscape as at as in the case of tridymite. Another reason is that the potentials have been parameterised to distinguish between structures which have reconstructive transitions (where bonds are broken and formed) and may not be able to deal with displacive transitions (where only angles between atoms change) as with tridymite. The �final possible explanation is that a number of the known structures may be meta-stable and/or poorly characterised. For the second research area molecular dynamics simulations using a rigid ion two body potential were carried out in order to investigate the properties of silica melts and glasses. A number of different silica crystals were melted to see whether the melts are all similar or whether their properties can be di�fferentiated according to the original crystal structure. At sufficiently high temperatures the starting structure did not a�ffect the properties of the melt. Several properties of silica melts and glasses were investigated: mean square displacement, autocorrelation functions, pair distribution functions, the extent to which silicon and oxygen atoms move together, Arrhenius plots, coordination number, bond lengths and angles. Investigations were also carried out as to whether it is possible to use a shell model to simulate a silica melt. Various properties were calculated and it was found that agreement with experiment was not as accurate as when using the rigid ion model. The third research area is an exploration of the properties of amorphous silica at elevated pressures and a range of temperatures, using molecular dynamics with a rigid ion two body potential. Calculations show that, at low temperatures, the distortion of the tetrahedra is not recovered upon decompression whereas experimental results �find complete recovery of the tetrahedra. There is little available experimental data on the behaviour of silica at both high pressures and temperatures. Calculations show that at high temperatures all properties of the initial structure before compression are recovered.
|Title:||Computational studies of silica|
|Open access status:||An open access version is available from UCL Discovery|
|UCL classification:||UCL > School of BEAMS > Faculty of Maths and Physical Sciences > Chemistry|
Archive Staff Only