Prediction and inference of dynamics in quantum plasmas

Thesis

Quantum plasmas arise in a range of physical contexts, from planetary interiors to Inertial Confinement Fusion experiments. We require sophisticated numerical approaches to find the structural and dynamic properties of these complex, many-body quantum systems; diverse in accuracy and applicability, these computational approaches each have their own strengths and (often ignored) weaknesses. By combining existing techniques with newly developed algorithms, this thesis addresses some weaknesses that are common to many state-of-the-art calculation methods for quantum plasmas, and opens up paths to calculating properties that were not previously accessible. The focus of the thesis is a method based on Bohmian mechanics for dynamic structure calculations; alongside this, stochastic gradient Markov Chain Monte Carlo methods are developed for inferring plasma properties from experimental data.

Authors: Larder, B

• DOI: 10.5287/ora-2ab2embpx
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Markov Chain Monte Carlo; Stochastic Calculus; Warm Dense Matter; Quantum Plasmas; Bohmian Mechanics; Plasma Physics
• Subjects: Plasma Physics; Warm Dense Matter; Materials Science; Quantum Physics