Linear scaling approach for atomistic calculation of excitonic properties of 10-million-atom nanostructures

TL;DR

This paper introduces a linear scaling computational method combining empirical tight-binding and configuration interaction to accurately calculate excitonic properties in nanostructures with over 10 million atoms, enabling atomistic modeling at unprecedented scales.

Contribution

The authors develop and benchmark a scalable atomistic approach that significantly reduces computational complexity for large nanostructures, facilitating detailed excitonic property calculations.

Findings

Method successfully benchmarks on InAs/GaAs quantum dots.

Applied to crystal phase quantum dots with over 10 million atoms.

Achieves atomistic accuracy with linear computational scaling.

Abstract

Numerical calculations of excitonic properties of novel nanostructures, such as nanowire and crystal phase quantum dots, must combine atomistic accuracy with an approachable computational complexity. The key difficulty comes from the fact that excitonic spectra details arise from atomicscale contributions that must be integrated over a large spatial domain containing a million and more of atoms. In this work we present a step-by-step solution to this problem: combined empirical tight-binding and configuration interaction scheme that unites linearly scaling computational time with the essentials of the atomistic modeling. We benchmark our method on the example of wellstudied self-assembled InAs/GaAs quantum dot. Next, we apply our atomistic approach to crystal phase quantum dots containing more than 10 million atoms.