Low scaling algorithm for the random phase approximation

Description

The computationally most expensive step in conventional RPA implementations is the calculation of the independent particle polarizability χ. We present an RPA algorithm that calculates χ using the Green function G in real space and imaginary time. The systematic construction of optimized time and frequency grids for G is obtained by means of solving a fitting problem. Furthermore a non-uniform discrete Fourier transform between the two grids is introduced, which converges exponentially. We show that the usage of the Green function approach in combination with the optimized grids can be used for the calculation of the RPA correlation energy for large systems. Co-Presenters: Jiri Klimes, Georg Kresse

Period	1 Apr 2014
Event title	DPG Frühjahrstagung der Sektion Kondensierte Materie (SKM), Dresden 2014
Event type	Conference
Location	Dresden, GermanyShow on map