SNIC SUPR
High-temperature thermodynamics of complex systems from first principles
Dnr:

SNIC 2019/1-15

Type:

SNIC Large Compute

Principal Investigator:

Andrei Ruban

Affiliation:

Kungliga Tekniska högskolan

Start Date:

2019-07-01

End Date:

2020-07-01

Primary Classification:

10304: Condensed Matter Physics

Secondary Classification:

10403: Materials Chemistry

Webpage:

Allocation

Abstract

Large computer resources are required to perform extremely computationally demanding ab initio molecular dynamics simulations of such complex systems as surfaces of (Ti-Al)(C-N-O) with defects and concentrated multicomponent 3d-metal based alloys (also known as high-entropy or equimolar) in order to investigate their thermodynamic and kinetic properties. These tasks are parts of the following ongoing projects: 1) SSF grant RMA15-0048, "CVD 2.0 - En ny generation hårda beläggningar" and 2) VR project project 2015-05538 "First-principles-based models of complex systems with non-trivial magnetic, chemical and strain-induced interactions at finite temperature". Investigated systems require the use of large models on atomic scale (about 200-500 atoms) due to their complexity and inhomogeneity . The latter is also the reason why molecular dynamics simulations at high temperature are needed: as is established, the usual quasiharmonic approximation breaks down at elevated temperatures (relevant for processing these materials) due to highly asymmetric potential energy surface of atoms next to defects or very different in the atomic size another alloy component. In order to speed up the free energy calculations, advanced thermodynamic integration techniques will be used, which are testing in cooperation with their authors at Max Planck institute in Dusseldorf right now. That kind of computational approach which should allow one to get accurate and reliable description of the thermodynamic properties of complex inhomogeneous systems has never been used before. If the progress is successful, it will result in the creation of computation environment, which can be used by other scientific groups worldwide.