Fundamental atomic data and prototype techniques for a generalised collisional-radiative model of medium-weight elements in fusion and astrophysical plasma

Student thesis: Doctoral Thesis

Abstract

Quantifying the influence of heavy impurities upon plasma power balance, while seizing the opportunities they offer for passive spectroscopy, requires generalised collisional-radiative (GCR) population modelling to produce high-quality ionisation balances and cooling curves. ADAS provides a well-established framework of codes and data for the GCR modelling of light species and has been applied extensively to these scenarios. The extension of GCR modelling to medium and heavyweight elements within the ADAS framework imposes a number of updates and modifications. First, a lift of the ADAS baseline atomic structure and collision data is proposed using autostructure with the distorted-wave approximation, configuration sets selected by optimising on radiated power, and a novel, algorithmic strategy for optimising the radial scaling parameters. The truncation error of the configuration sets is bounded between an order of magnitude and 10%, while three figures of merit prove that the scaling parameter optimisation has eliminated the 20–30% structure error relative to the Cowan code. Second, fully relativistic, partially radiation-damped, Dirac R-matrix calculations of the W44+ ion are performed to showcase the challenges of generating fundamental data for heavy species. The calculations use a configuration interaction and close-coupling expansion that opens up the 3d-subshell, yielding previously unexplored transition arrays, [3d104s2–3d94s24f] and [3d104s2–3d94s4p4d], which contribute 50% of the total radiated line power coefficient (PLT) near the temperature of peak abundance. Third, collisional excitation by ion projectiles, not just electrons, must now be considered. A broad baseline of ion-impact excitation data is fulfilled by the restoration of a code, a2iratbt, that uses semi-classical, first-order perturbative equations with a limiting function, to prevent transition probability overestimates at intermediate energies, and a radial cutoffwhich ensures the xiv infinite-energy Born limit is approached at high energies. The majority of the error in this baseline comes from the neglect of close coupling, accounting for ~ 20% in triplets and <5% in doublets. Fourth, and most importantly, the resolution of GCR modelling must be moved to intermediate coupling. A prototype is built upon the LS-resolved analogue, predominantly by statistically splitting relevant quantities onto the intermediate-coupling manifold. Comparison to the unresolved fractional abundances in the literature reveal density effects of over an order of magnitude for the near neutrals, decreasing gradually towards complete agreement for the highly ionised stages. The total radiated power function and PLT s showed better agreement, generally within 50% for the higher quality sources. Investigations into the effects of ion-impact excitation and resolution upon the GCR results are performed, showing that neither can be ignored. Also, a true set of metastable terms and levels is established. In the final analysis, the real impact of this new model can only be completely assessed by applying its results in subsequent plasma transport modelling.
Date of Award1 Apr 2017
LanguageEnglish
Awarding Institution
  • University Of Strathclyde
SponsorsUniversity of Strathclyde
SupervisorMartin O'Mullane (Supervisor) & Hugh Summers (Supervisor)

Cite this