Ab initio investigations of low Z materials under extreme conditions

Physics;Chemistry;Condensed matter;High pressure;Extreme conditions;Hydrogen;Lithium;Hydrogen-helium mixtures;Quantum Monte Carlo (QMC);Density functional theory (DFT);Benchmarking;Forces

Clay, Raymond C.

关键词: Physics;    Chemistry;    Condensed matter;    High pressure;    Extreme conditions;    Hydrogen;    Lithium;    Hydrogen-helium mixtures;    Quantum Monte Carlo (QMC);    Density functional theory (DFT);    Benchmarking;    Forces;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/95454/CLAY-DISSERTATION-2016.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Low Z elements, particularly hydrogen and helium, make up the vast majority of matter in the universe.These elements are observed to exist over the widest known ranges of pressures and temperatures, from interstellar plasmas to the cores of stars.In modeling Jovian planets or inertial confinement fusion, the pressures and temperatures of hydrogen and helium can even vary by orders of magnitude within a single system.Thus, understanding these systems requires an accurate phase diagram over a large range of thermodynamic conditions, particularly at high pressures.Attaining this level of accuracy has been an ongoing challenge for experimentalists and theorists since the phase diagrams of low Z elements exhibit surprising complexity.Experimentally, static compression experiments are frustrated by the high reactivity of lithium and hydrogen, which greatly limits the pressure ranges that can be accurately characterized.Additionally, determining the relevant crystal structures of interesting phases can be very difficult for various reasons.Theoretically, most \textit{ab initio} based methods that can treat bulk systems have to treat electron correlation and nuclear quantum effects approximately.In hydrogen and lithium, the errors introduced by these approximations are comparable to the enthalpy differences between competing phases, giving inaccurate phase boundaries.We believe that several outstanding questions in low Z phase diagrams can be resolved through a careful and systematic application of ab initio methods, particularly quantum Monte Carlo.Quantum Monte Carlo is well suited for the study of high-pressure low Z elements, as it is possible to treat all electrons quantum mechanically and with few uncontrolled approximations.This level of accuracy is suitable not only for direct applications of QMC to problems of interest, but also to benchmark and establish confidence in widely used density functional theory (DFT) calculations.The purpose of my thesis will be two-fold: to use ab initio methods like quantum Monte Carlo to shed light on the phase diagram of bulk low Z elements, and to use these systems as a test bed for new QMC methods which make use of forces.The first part of this thesis will cover the necessary theoretical background, including our work on force and stress estimators.Then we will discuss our QMC based benchmarking method in both hydrogen and helium.The results of these benchmarking studies are then used in an attempt to resolve some major issues in the phase diagrams of hydrogen and hydrogen-helium mixtures.Lastly, we will present our results for the melting line and solid phase diagram for dense lithium.