Coupled cluster theory is currently under active development because of its potential to be a benchmark method for solids. In coupled cluster, finite size effects cause errors in the energy and these errors are particularly severe when modeling metals with a finite simulation cell. Here, I will describe a development we have called “structure factor twist averaging”, which is a twist angle selection method. Twist angles cause changes in orbital occupations, which causes each twist to produce a different energy. Averaging over these twists reduces finite size effects but requires a large number of calculations to be run at different twists, and this can be very expensive. Our method finds the one twists that has an energy close to the average, and consequently, the smallest amount of finite size effects. This selection in our scheme is made using the transition structure factor due to its relationship with the correlation energy. I will show the effectiveness of this method across system sizes and basis set sizes using a model system before showing an application to the transition pressure in silicon.

Twist angle selection using the transition structure factor speeds up finite size convergence in periodic coupled cluster theory

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.