Semester

Spring

Date of Graduation

2022

Document Type

Dissertation

Degree Type

PhD

College

Eberly College of Arts and Sciences

Department

Physics and Astronomy

Committee Chair

Aldo H. Romero

Committee Co-Chair

Matthew Johnson

Committee Member

Matthew Johnson

Committee Member

Hyowon Park

Committee Member

Tudor Stanescu

Abstract

The electronic correlations in materials drive a variety of fascinating phenomena from magnetism to metal-to-insulator transitions (MIT), which are due to the coupling between electron spin, charge, ionic displacements, and orbital ordering. Although Density Functional Theory (DFT) successfully describes the electronic structure of weakly interacting material systems, being a static mean-field approach, it fails to predict the properties of Strongly Correlated Materials (SCM) that include transition and rare earth metals where there is a prominent electron localization as in the case of d and f orbitals due to the nature of their spatial confinement.

Dynamical Mean Field Theory (DMFT) is a Green’s function based method that has shown success in treating SCM. This dissertation focuses on the development of a user-friendly, open-source Python/Fortran framework, “DMFTwDFT” combining DFT and DMFT to characterize properties of both weakly and strongly correlated materials. The DFT Kohn- Sham orbitals are projected onto Maximally Localized Wannier Functions (MLWF) which essentially maps the Hubbard model to a local impurity model which we solve numerically using quantum Monte Carlo methods to capture both itinerant and localized nature of electrons. Additionally, we provide a library mode for computing the DMFT density matrix which can be linked and internally called from any DFT package allowing developers of other DFT codes to interface with our package and achieve full charge-self-consistency within DFT+DMFT.

We then study the stability and diffusion of oxygen vacancies in the correlated material LaNiO3. By treating Ni-d as correlated orbitals along with a Ni-O hybridization manifold, we show that certain configurations undergo a MIT based on the environment of their vacancies. We also compute the transition path energy of a single oxygen vacancy through means of the nudged elastic band (NEB) method. We show that the diffusion energy profile calculated through DFT+U differs from that of DMFT, due to correlation effects that are not quite well captured with static mean-field theories.

Additionally, DMFTwDFT was utilized to study strongly correlated alloys and materials useful for neuromorphic computing applications.

Download

Share

COinS