Microwave propagation and interaction in fusion plasma

Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2022
Thesis identifier
  • T16306
Person Identifier (Local)
  • 201660864
Qualification Level
Qualification Name
Department, School or Faculty
  • The interactions between electromagnetic waves and magnetic fusion plasmas are fundamental to a broad range of technologies considered vital in achieving an efficient, energy producing tokamak. Sophisticated diagnostic instruments utilise the wave-plasma scattering mechanism to make non-invasive plasma measurements while powerful microwaves can be coupled to the plasma to drive heat or currents in the plasma. This research uses endeavours to deepen our understanding of these plasma-wave interactions by utilising a variety of numerical tools and performing novel modelling. The full-wave three-dimensional (3D), finite difference time domain (FDTD) code EMIT-3D is developed and utilised in conjunction with the Hermes fluid code using BOUT++ as a framework to study cross-polarisation Doppler-backscattering, in which a low-amplitude extraordinary (X) mode wave launched into the plasma is back-scattered into an ordinary (O) mode wave on magnetic field fluctuations, and the O mode is recorded and analysed. Benchmarking of the code was achieved by evaluating the simulated scaling relationships between the back-scattered signal strength against perturbation strength, and comparing this to theory. Excellent agreement was found. Further modelling investigated non-Wentzel-Kramers-Brillouin (WKB) effects and their pertinence towards back-scattering measurements and found some significant influences at experimentally relevant density length-scales. Furthermore, the forward scattered original-polar signals are identified to have major contributions towards the perceived back-scattered intensity under certain conditions, and an asymmetry in the back-scattered profile was also identified. In addition, for high-power microwaves having X mode polarisation, electron cyclotron current drive (ECCD) was modelled using the Torbeam ray-tracing code. However, ECCD was shown to be an ineffective method of driving plasma current in high beta and high density plasmas where the fundamental cyclotron harmonics are cutoff. A promising alternative to ECCD at high plasma densities are electron Bernstein wave (EBW) heating and current drive, in which an O mode is converted into a slow X mode wave that in turn is converted into an EBW that can propagate into the dense plasma. The O-X mode conversion efficiency was studied with EMIT-3D, and excellent agreement was found between this 3D code and several other 2D codes in simple plasma geometries. However, for a few test cases the 3D code predicted lacklustre mode conversion efficiencies when a more complicated high beta MAST-U equilibrium was used in the modelling.
Advisor / supervisor
  • Eliasson, Bengt
Resource Type