Phase behaviour of core-softened particles

• Main Author: Magee, James
• Published: University of Edinburgh 2003
• Subjects: 530.41
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.657189

In this thesis, we have made an extended study of a particular core-softened system in two dimensions (originally studied in Sadr-Lahijany <i>et al. </i>Physical Review Letters 81, p.4895 [1]). Whereas most studies have only examined single facets of the phase behaviour for a given model, we have systematically studied both liquid and solid state phase behaviour, using a combination of analytic and state of the art simulation methods. Our aim has been to gain a greater understanding of the key features behind the unusual phase behaviour associated with core softened systems. After an introduction, we review previous work on the phase behaviour of core-softened systems, in which we see that these apparently “simple” potentials can give rise to a range of unusual and exotic behaviours. We then briefly study the behaviour of a simple one-dimensional core softened model. This demonstrates liquid state anomalies and a zero temperature transition point, as well as illuminating the dangers of generalising between potentials and across dimensionality. We then move onto the two dimensional system. We study the solid state using a combination of the harmonic approximation, Lattice Switch Monte Carlo (a recently developed simulation technique which allows Monte Carlo moves between structures), and Gibbs-Duhem integration. We find two triangular lattice phases, separated by a region of stable square lattice phase at intermediate pressure; we also find evidence for an isostructural transition (with possible critical point) between the triangular phases at densities metastable with respect to the square lattice phase. To study the liquid state, we use Lennard-Jones Devonshire cell theory and extensive Monte Carlo simulation. We reevaluate the cell theory model, and find strong evidence that Lennard-Jones and Devonshire’s original results were misinterpreted, as were the cell theory results for the system at hand [1]. Our liquid state simulation results confirm the presence of reentrant melting and liquid-state anomalies in the system. We find that, contrary to previous suggestions, these anomalies do not originate from a liquid-liquid transition, but instead from a continuous or near-continuous melting transition, a phenomenon only possible in two dimensions.