Abstract

I present the ab initio random structure searching method (AIRSS), developed with co-workers at the University of Cambridge, UK and University College London, UK. AIRSS uses random numbers to suggest different structural configurations of atoms within a material. The AIRSS method has been incredibly successful at predicting the stable phases of materials at high pressures, such as those in the so-called “gas giant” planets. I have been using AIRSS in more “every day” conditions to predict the structures of materials relevant to technology[1,2]. I will present the AIRSS method and show how it has been applied to these problems. The storage of electricity in batteries is a major technological challenge. To fully understand how each new battery works we must understand the many different structures formed within the battery as it charges and discharges. AIRSS is particularly good for this. I illustrate the approach, via the use of silicon, which is a potential candidate to replace graphite in the next generation of lithium batteries. We show how we can use AIRSS to understand what happens as this material is charged and discharged. We can predict not only the structural changes as silicon is lithiated, but also the defects formed and the voltages of the structures on charging [3].

[1] Andrew J. Morris, Chris J. Pickard and R. J. Needs, Phys. Rev. B 78 , 184102 (2008). [2] Andrew J. Morris, Chris J. Pickard and R. J. Needs, Phys. Rev. B 80 , 144112 (2009). [3] Andrew J. Morris, R. J. Needs, C. P. Grey and C. J. Pickard, in preparation.