Low-scaling GW algorithm applied to twisted transition-metal dichalcogenide heterobilayers

TL;DR

This paper introduces a low-scaling, efficient GW algorithm that leverages locality to enable large-scale electronic structure calculations on complex twisted transition-metal dichalcogenide heterobilayers, significantly reducing computational time.

Contribution

The authors develop a periodic, low-scaling GW algorithm that drastically reduces computational cost, enabling calculations on systems with nearly 1000 atoms per unit cell.

Findings

Achieved G0W0 calculations on a 984-atom bilayer in 42 hours using 1536 cores.

The new algorithm is four orders of magnitude faster than traditional plane-wave methods.

Enables unprecedented computational studies of electronic excitations in large nanoscale systems.

Abstract

The $GW$ method is widely used for calculating the electronic band structure of materials. The high computational cost of $GW$ algorithms prohibits their application to many systems of interest. We present a periodic, low-scaling and highly efficient $GW$ algorithm that benefits from the locality of the Gaussian basis and the polarizability. The algorithm enables $G_0W_0$ calculations on a MoSe$_2$/WS$_2$ bilayer with 984 atoms per unit cell, in 42 hours using 1536 cores. This is four orders of magnitude faster than a plane-wave $G_0W_0$ algorithm, allowing for unprecedented computational studies of electronic excitations at the nanoscale.