Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.729748
Title: Computational predictions of structures, inclusion behaviour and properties of organic molecular crystals
Author: Nyman, Jonas
ISNI:       0000 0004 6497 2132
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Thesis embargoed until 29 Jun 2020
Access from Institution:
Abstract:
This thesis is about the computational prediction of crystal structures and their properties. Polymorphism, where the same molecule crystallizes in more than one structure is investigated. Several structures of polymorphs and clathrates, porous crystals that adsorb gases, are predicted without prior experimental data. Rigid-molecule lattice dynamics in an anisotropic force field is used to calculate temperature-dependent properties of large sets of crystals. Using the lattice-vibrational free energy as scoring function incrystal structure prediction is discussed. Brillouin zone sampling and convergence difficulties of lattice dynamics calculations are addressed, with a kernel density estimationmethod offered as a solution. The advantage of computationally affordable force field methods over electronic structure methods will be described. I demonstrate that multipole-based force fields canbe comparable in accuracy to dispersion-corrected generalized gradient approximation density functional theory and that such force fields are very suitable for crystal structure prediction as they allow the calculation of realistic free energies for hundreds ofstructures at a relatively small computational cost. For the prediction of clathrate structures, I show that a combination of the free energy and the guest-to-cavity volume ratio can be a suitable scoring function. I suggest thatporous structures in predicted crystal energy landscapes should be interpreted as possible solvates, clathrates and hydrates, and that these should be carefully considered in theanalysis of prediction results.
Supervisor: Day, Graeme Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.729748  DOI: Not available
Share: