Theoretical astrophysics and plasma physics at RPC

Abstract:

The ITER Neutral Beam Injection (NBI) system is based on negative ions, produced in an RF-driven plasma source. The ITER NBI lines must deliver a current density of 230 A/m2 of negative hydrogen ions, accelerated to 870 keV, or a current density of 200 A/m2 of negative deuterium ions accelerated to 1 MeV. NBI systems, based on negative ions, are compromised by a process known as stripping losses, in which negative ions are neutralized in the grid system before achieving full energy. For a source filling pressure of p fill = 0.3 Pa, 29% of the extracted H -(D -) ions are predicted to be lost by stripping in the ITER full-scale NBIs system (7 grid acceleration system). To compensate for these stripping losses, a larger amount of negative ions has to be extracted from the source (329 A/m2 in hydrogen and 286 A/m2 in deuterium). The ELISE test facility is based on a 1/2-size ITER source. It extracts H -(D -) ions using a 3-grid acceleration system, with a maximum extraction voltage of 10 kV and acceleration voltage of 50 kV is achieved. In a 3-grid acceleration system, 10% of stripping losses is predicted for both isotopes. This contribution focuses on experimental measurements of stripping losses at ELISE. Experimentally, stripping losses are monitored using Beam Emission Spectroscopy (BES), which analyzes the Doppler-shifted spectrum of the Balmer Hα (Dα ). To not underestimate the number of stripping losses the full area between the unshifted Peak background (H 2 dissociation and excitation) and the Doppler Peak (fully-accelerated beam particles excitation) needs to be considered. However, the influence of BES background and signal-to-noise ratio (SNR) can affect the calculation of stripping losses, mainly for hydrogen measurements at low filling pressures (< 0.4 Pa). To accurately predict the value of the stripping losses, only signals with high-enought SNR should be used. When this effect is considered, no differences between hydrogen and deuterium are found in terms of stripping losses. For a filling pressure of 0.3 Pa, a stripping fraction of 6.0±0.8% was found for hydrogen and 6.2±0.7% for deuterium. A systematic comparison of the stripping losses between hydrogen and deuterium under various experimental conditions is presented.

Abstract:

A set of analytic solutions for the plunging region thermodynamics has been developed recently under the assumption that the fluid undergoes a gravity-dominated geodesic plunge into the black hole. We test this model against a dedicated 3D global general relativistic magnetohydrodynamics simulation of a thin accretion disc around a Schwarzschild black hole using the code athenak . Provided that we include the effects of non-adiabatic heating (plausibly from grid-scale magnetic dissipation), we find excellent agreement between the analytic model and the simulated quantities. These results are particularly important for existing and future electromagnetic black hole spin measurements, many of which do not include the plunging fluid in their emission modelling. This exclusion typically stems from the assumption of a zero-stress boundary condition at the innermost stable circular orbit (ISCO), forcing all thermodynamic quantities to vanish. Instead, we find a non-zero drop in the angular momentum over the plunging region, which is consistent with both prior simulations and observations. We demonstrate that this stress is small enough for the dynamics of the fluid in the plunging region to be well-described by geodesic trajectories, yet large enough to cause measurable dissipation near to the ISCO – keeping thermodynamic quantities from vanishing. In the plunging region, constant -disc models are a physically inappropriate framework.

Abstract:

Curvature-driven instabilities are ubiquitous in magnetised fusion plasmas. By analysing the conservation laws of the gyrokinetic system of equations, we demonstrate that the well-known spatial localisation of these instabilities to regions of ‘bad magnetic curvature’ can be explained using the conservation law for a sign-indefinite quadratic quantity that we call the gyrokinetic field invariant. Its evolution equation allows us to define the local effective magnetic curvature whose sign demarcates the regions of ‘good’ and ‘bad’ curvature, which, under some additional simplifying assumptions, can be shown to correspond to the inboard (high-field) and outboard (low-field) sides of a tokamak plasma, respectively. We find that, given some reasonable assumptions, electrostatic curvature-driven modes are always localised to the regions of bad magnetic curvature, regardless of the specific character of the instability. More importantly, we also deduce that any mode that is unstable in the region of good magnetic curvature must be electromagnetic in nature. As a concrete example, we present the magnetic-drift mode, a novel good-curvature electromagnetic instability, and compare its properties with the well-known electron-temperature-gradient instability. Finally, we discuss the relevance of the magnetic drift mode for high- fusion plasmas, and in particular its relationship with microtearing modes.