Author: Bismayan Chakrabarti

Publisher:

Published: 2017

Total Pages: 105

ISBN-13:

DOWNLOAD EBOOK

The study of strongly correlated materials is currently perhaps one of the most active areas of research in condensed matter physics. Strongly correlated materials contain localized electronic states which are often hybridized with more itinerant electrons. This interplay between localized and delocalized degrees of freedom means that these compounds have highly complex phase diagrams which makes these compounds very challenging to understand from a theoretical standpoint. Computer simulations have proved to be an invaluable tool in this regard with state of the art abinitio simulation techniques harnessing the ever-increasing power of modern computers to produce highly accurate descriptions of a variety of strongly correlated materials. One of the most powerful simulation techniques currently in existence is Dynamical Mean Field Theory (DMFT). This thesis describes this powerful simulation technique and its applications to various material systems, as well as addressing some theoretical questions concerning particular implementations of DMFT. This thesis is divided into two parts. In part 1, we describe the theory behind DMFT and its amalgamation with Density Functional Theory (DFT+DMFT). In chapters 2 and 3, we provide the basic theory theory behind DFT and DMFT respectively. In chapter 4, we describe how these two methods are merged to give us the computational framework that is used in this thesis, namely DFT+DMFT. Finally, we round o part 1 of the thesis in chapter 5, which provides a description of the Continuous Time Quantum Monte Carlo (CTQMC) impurity solver, which is at the heart of the DFT+DMFT algorithm and is used extensively throughout this thesis. In part two of the thesis, we apply the DFT+DMFT framework to address some important problems in condensed matter physics. In chapter 6, we study the Magnetic Spectral Function of strongly correlated f-shell materials to understand two important problems in condensed matter physics, namely the volume collapse transition in Cerium and the valence uctuating state ground state of -Pu. In chapter 7, we study the contribution of lattice parameters and electronic entropy to study the decades-old problem of understanding the spin state transition observed in LaCoO3, where we show how lattice expansion, octahedral rotations and electronic entropy are all essential in stabilizing the high-spin state at high temperature. In chapter 8, we switch to studying a more theoretical problem by looking at the problems with using the highly popular constrained Random Phase Approximation (cRPA) method to estimate the screening of local inter-electronic repulsion in strongly correlated systems. We show that cRPA systematically underestimates screening in such systems which makes it an unsuitable method for estimating the repulsion parameter (U) used in impurity solvers. We then develop an alternate method to estimate the screening using the full local polarization which overcomes many of these limitations. Chapter 9 contains all the conclusions obtained in this thesis, followed by references and appendices.

Download or read book Study of Two-particle Response and Phase Changes in Strongly Correlated Systems Using Dynamical Mean Field Theory written by Bismayan Chakrabarti and published by . This book was released on 2017 with total page 105 pages. Available in PDF, EPUB and Kindle. Book excerpt: The study of strongly correlated materials is currently perhaps one of the most active areas of research in condensed matter physics. Strongly correlated materials contain localized electronic states which are often hybridized with more itinerant electrons. This interplay between localized and delocalized degrees of freedom means that these compounds have highly complex phase diagrams which makes these compounds very challenging to understand from a theoretical standpoint. Computer simulations have proved to be an invaluable tool in this regard with state of the art abinitio simulation techniques harnessing the ever-increasing power of modern computers to produce highly accurate descriptions of a variety of strongly correlated materials. One of the most powerful simulation techniques currently in existence is Dynamical Mean Field Theory (DMFT). This thesis describes this powerful simulation technique and its applications to various material systems, as well as addressing some theoretical questions concerning particular implementations of DMFT. This thesis is divided into two parts. In part 1, we describe the theory behind DMFT and its amalgamation with Density Functional Theory (DFT+DMFT). In chapters 2 and 3, we provide the basic theory theory behind DFT and DMFT respectively. In chapter 4, we describe how these two methods are merged to give us the computational framework that is used in this thesis, namely DFT+DMFT. Finally, we round o part 1 of the thesis in chapter 5, which provides a description of the Continuous Time Quantum Monte Carlo (CTQMC) impurity solver, which is at the heart of the DFT+DMFT algorithm and is used extensively throughout this thesis. In part two of the thesis, we apply the DFT+DMFT framework to address some important problems in condensed matter physics. In chapter 6, we study the Magnetic Spectral Function of strongly correlated f-shell materials to understand two important problems in condensed matter physics, namely the volume collapse transition in Cerium and the valence uctuating state ground state of -Pu. In chapter 7, we study the contribution of lattice parameters and electronic entropy to study the decades-old problem of understanding the spin state transition observed in LaCoO3, where we show how lattice expansion, octahedral rotations and electronic entropy are all essential in stabilizing the high-spin state at high temperature. In chapter 8, we switch to studying a more theoretical problem by looking at the problems with using the highly popular constrained Random Phase Approximation (cRPA) method to estimate the screening of local inter-electronic repulsion in strongly correlated systems. We show that cRPA systematically underestimates screening in such systems which makes it an unsuitable method for estimating the repulsion parameter (U) used in impurity solvers. We then develop an alternate method to estimate the screening using the full local polarization which overcomes many of these limitations. Chapter 9 contains all the conclusions obtained in this thesis, followed by references and appendices.