Abstract Details
Abstracts
Author: Michael Barnes
Requested Type: Poster
Submitted: 2019-02-22 07:38:54
Co-authors: F. I. Parra, M. Landreman, J. A. Alonso, I. Calvo, J. M. Garcia-Regana, J. L. Velasco
Contact Info:
University of Oxford
Clarendon Laboratory, Parks Ro
Oxford, OX1 3PU, 00000
United Kingdom
Abstract Text:
Turbulent transport is expected to limit energy confinement in stellarators that are optimized to reduce neoclassical transport. However, there have been relatively few studies of turbulence or even of micro-instabilities in stellarators. These studies have assumed a Maxwellian velocity space distribution for the plasma equilibrium. However, at reactor-relevant temperatures the collisional mean free path should be sufficiently long that particles trapped in the non-axisymmetric magnetic field diffuse radially outward with a characteristic step size much larger than the ion gyroradius. These large radial excursions lead to significant deviations of the equilibrium velocity distribution from a Maxwellian. In the limits of large aspect ratio [1] or approximate omnigeneity [2], it is still possible to define a small parameter so that these deviations can be treated as perturbations to a Maxwellian. However, due to the sensitivity of turbulent heat fluxes to gradients, the non-Maxwellian corrections may nonetheless impact turbulent transport.
We present results for micro-stability in the low collisionality regime, with these non-Maxwellian features taken into account. The deviation from a Maxwellian is calculated using the drift kinetic code SFINCS [3]; the fluctuations are computed using a new gyrokinetic code called stella [4]. In particular, we present results on the dependence of micro-stability on collision frequency and on the ratio of gyroradius to system size. A discussion of the implications of these results for optimized stellarator core confinement will also be given.
[1] D.D.-M. Ho and R. M. Kulsrud, Phys. Fluids 30, 442 (1987).
[2] I. Calvo et al., Plasma Phys. Control. Fusion 59, 055014 (2017).
[2] Landreman et al., Phys. Plasmas 21, 042503 (2014).
[3] Barnes et al., arXiv:1806.02162v1 (2018).
Comments:
The online template requires a zip code that is purely numeric. As UK postal codes are not purely numeric (they include letters), I was forced to put in a nonsensical zip code (00000).
