Transport Properties of Divertor Edge Plasmas Measured with Multi-Spectral Imaging

Abstract

The transport of heat and particles in the boundary of a tokamak is not sufficiently understood for the purposes of constructing a pilot nuclear reactor. Improving numerical and theoretical understanding is inhibited by traditional boundary diagnostics that provide sparse and inflexible spatial coverage. In this thesis, multi-spectral imaging of helium line ratios (HeMSI) was used to create 2D poloidal maps of Tₑ and nₑ in the TCV divertor. These are the first plasma boundary measurements to provide continuous 2D coverage of Tₑ and nₑ for arbitrary magnetic geometries. These measurements were validated against co-local Thomson scattering measurements in diverted plasmas. HeMSI showed good agreement with Thomson scattering in the common flux region(CFR) of ionizing plasma for both majority helium and majority deuterium plasmas. Having validated this powerful new tool, HeMSI was used to investigate the effects of flux expansion in the TCV divertor for plasmas in the conduction limited regime. Increasing poloidal flux expansion is expected to lower the temperature of the divertor target by increasing the plasma volume and connection length of the magnetic field line between the core and target. These benefits are observed in the conduction limited regime but not in the partially detached regime. The 2D poloidal maps of Tₑ and nₑ, in concert with other measurements, were used to calculate the ionization rate of He and D, the E × B drift velocity, Spitzer heat conduction, and parallel flow in 2D. This allowed for heat transport to be locally resolved into conduction, parallel convection, and drift convection components. Similarly, particle transport was categorized into drift and parallel components. These calculations demonstrate that in relatively cool plasmas (Tₑ < 30eV), drifts compose a significant amount of the heat and particle transport. This violates the assumptions of simple two-point modeling and demonstrates the importance of accounting for drifts in modeling. Drifts may explain the boundary’s lack of sensitivity to poloidal flux expansion in the partially detached regime. Lastly, the anomalous heat and particle transport coefficients, χ⊥ and D⊥, were calculated by enforcing local power and particle balance. Values of χ⊥ close to the separatrix (ρ < 1.005), and values of D⊥ were consistent with standard modeling practices. However,χ⊥ measurements sufficiently far into the CFR(ρ > 1.005) exceeded typical modeling assumptions by two orders of magnitude. This implies that boundary codes will underestimate the radial temperature falloff length. This is shown to be true in a comparison of Tₑ measurement to simulations performed with the SOLPS-ITER code. This brings into question the validity of the assumption of diffusive heat transport in the far CFR.