The bulk of this thesis is concerned with the calculation of molecular wavefunctions - calculated within a Different Orbitals for Different Spins framework. Initially other important methods of constructing molecular wavefunctions are considered, and the DODS method is discussed in relation to them. Two different DODS methods are considered in detail, namely the Alternant Molecular Orbital method and the Non-Paired Spatial Orbital method. Other workers have previously used these methods and obtained satisfactory wavefunctions. Both methods make good allowance for electron correlation, but for benzene, and several other hydrocarbons, the NPSO method proves superior. Pauncz has stated that, (67), because the NPSO calculations are cumbersome,"one cannot expect to apply this method to larger systems." In this thesis this NPSO method is extended to larger molecules, namely azulene, anthracene and phenanthrene. In all cases the NPSO method proves superior to the AMO method • With azulene, for example, a better ground state energy is obtained using a one-parameter NPSO wavefunction, than with a five-parameter optimised AMO wavefunction. Having found satisfactory ground state wavefunctions for these molecules, and observed that most endorse the suggestions that k - mln 0.25 , the NPSO method was applied to open shell systems, such as benzyl radical, and butadiene negative ion. The results for the benzyl radical are more satisfactory than for other open shell systems. ~he spin density distribution of the benzyl radical was calculated by various approximations, but with limited success. In this context we developed a fully spin projected NPSO method.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:595018 |
| Date | January 1973 |
| Creators | Linington, Mary Elizabeth |
| Publisher | University of Warwick |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://wrap.warwick.ac.uk/72805/ |
Page generated in 0.0015 seconds