Search for content and authors
 

Searching for low energy configurations of disordered materials using genetic algorithms

Chris E. Mohn 

University of Oslo, Physics Department and Centre for Materials Science and Nanotechnology, N-0316 Oslo, Blindern, Oslo N-0316, Norway

Abstract

We show how a genetic algorithm together with local energy minimisation can be used to find low energy configurations on the potential energy surface of disordered periodic systems. Examples include test cases (one and two dimensional periodic Ising models) alloys (Au-Cu) and mineral solid solution (CaCO3-MgCO3). We pay particular attention to challenges in maintaining the diversity in the population due to constraints associated with the fixed composition and synchronisation. We discuss briefly scaling with system size and the tuning of the parameters of the genetic algorithm. The genetic algorithm presented allows the study of disordered material in cases where a few configurations are thermally accessible and opens the way to efficient material design involving systems with a disordered structure.

 

Legal notice
  • Legal notice:
 

Presentation: Oral at E-MRS Fall Meeting 2007, Symposium G, by Chris E. Mohn
See On-line Journal of E-MRS Fall Meeting 2007

Submitted: 2007-05-09 16:55
Revised:   2009-06-07 00:44