30 September

Computational schemes for defects energetics

Friday 30 September 2022, 11:00am

ICN2 Seminar Room, ICN2 Building, UAB - Hybrid Event


PhD: Arsalan Akhtar

DirectorsProf. Pablo Ordejón, ICN2 Director and Theory and Simulation Group Leader and Dr Miguel A. Pruneda, CSIC Tenured Scientist at the same group.

Abstract:Addressing energetics of defects is helpful for understanding and identifying them in materials, a challenge that cannot be accomplished solely with experimental techniques. The objective of this thesis is to facilitate the computational study of defect properties ab initio, in particular from Density Functional Theory (DFT). Among the diverse classes of imperfections in materials, we focus on the simplest one, known as point defects: a single missing atom or a single atom impurity in the pristine crystal for example. The objective is to compute the energetics of these defects. Although the number of defects, structures, and configuration combinations pose a tremendous challenge for these computational approaches, the recent development of high-throughput computational materials offers a framework that, together, enables a route to address the problem: state of the art high-performance facilities, platforms for submitting and monitoring thousands of calculations, storing the information in searchable databases, and workflows that automatize the complex combinations of DFT calculations required to solve the electronic structure. With this in mind, I developed all the required Siesta machinery for computing defects´energetics using three particular workflow packages in AiiDA, a powerful platform that enables high-throughput computations in materials. We focus on two specific energies for point defects: Formation Energies and Migration Barrier Energies. The formation energy is the energy required to create the defect, which is related to the probability of having a certain concentration of that defect within the material. In the case of a charged defect, extra careful considerations are needed due to the electrostatic interaction of the charge with its periodic images. The migration barrier energy is the minimum energy required to activate the movement of diffusion of the defect in a particular direction in the system. This energy is related to ionic conductivity, and thus is of great interest for industrial applications such as batteries, solid oxide fuel cells (SOFC), or nanoionics supercapacitors. In the last part of the thesis, we focus on the effect of surfaces and interfaces on the properties of point defects. These boundaries break the translational symmetry of the crystal and are prone to the accumulation of larger concentrations of defects. From the computational point of view, they also require special tools, starting from the very first step of constructing adequate structural models. In the thesis we developed these three workflows (“siesta-defects”, “siesta-barriers”, and “siesta-surfaces”), discuss the pros and cons of different approaches to calculate the charged defect formation energies, discuss the problematics of the calculation of vacancy migrations with the NEB method when we use a strictly localized basis-set, and illustrate the combination of the different tools developed to study the effect of defects on the electronic properties of HfO2/Graphene heterostructures.