Quantum dynamics simulation of electrons in materials on high-performance computers

Andre Schleife, Erik W. Draeger, Victor Anisimov, Alfredo A. Correa, Yosuke Kanai

Research output: Contribution to journalArticlepeer-review

Abstract

Advancement in high-performance computing allows us to calculate properties of increasingly complex materials with unprecedented accuracy. At the same time, to take full advantage of modern leadership-class supercomputers, the calculations need to scale well on hundreds of thousands of processing cores. We demonstrate such high scalability of our recently developed implementation of Ehrenfest non-adiabatic electron-ion dynamics up to 1 million floating-point processing units on two different leadership-class computing architectures. As a representative example of material properties that derive from quantum dynamics of electrons, we demonstrate the accurate calculation of electronic stopping power, which characterizes the rate of energy transfer from a high-energy particle to electrons in materials. We discuss the specific case of crystalline gold with a hydrogen atom as the high-energy particle, and we illustrate detailed scientific insights that can be obtained from the quantum dynamics simulation at the electronic structure level. Please note that two animation videos of the time evolution for Figure 3 are available as Web extras at http://youtu.be/WxiMZ2DVBbM and http://youtu.be/bAcaxF9ARzM.

Original languageEnglish (US)
Article number6924632
Pages (from-to)54-60
Number of pages7
JournalComputing in Science and Engineering
Volume16
Issue number5
DOIs
StatePublished - Sep 1 2014

Keywords

  • HPC
  • computational materials science
  • high-performance computing
  • quantum electron dynamics
  • scientific computing

ASJC Scopus subject areas

  • Computer Science(all)
  • Engineering(all)

Fingerprint

Dive into the research topics of 'Quantum dynamics simulation of electrons in materials on high-performance computers'. Together they form a unique fingerprint.

Cite this