High-harmonic fast wave coupling and heating experiments in the CDX-U spherical tokamak

摘要

Next generation low-aspect-ratio (spherical) tokamaks will require a variety of auxiliary heating and current drive tools to achieve high $beta$. Fast magnetosonic waves with frequencies well above the fundamental ion-cyclotron frequency but below the lower-hybrid frequency are predicted to damp strongly on electrons in the high-$beta$ plasma of the spherical tokamak. These waves may provide a means of both heating plasma electrons and driving toroidal plasma current. The magnetic topology of the spherical tokamak is unique among tokamaks in that field lines at the outboard side of the plasma are strongly tilted off the equatorial mid-plane. The size and potential variability of this tilt during a plasma discharge could make efficient excitation of the fast wave difficult.;To better characterize the fast wave coupling efficiency in a low-aspect-ratio tokamak geometry, a two-strap antenna with arbitrary strap phasing was installed in the CDX-U spherical tokamak. The novel feature of this antenna is that it is manually rotatable between plasma discharges allowing fast wave coupling, propagation, and electron heating to be studied as a function of strap angle. A simplified cold-plasma fast wave coupling model was derived for this thesis and good agreement is found between predicted and measured coupling efficiency as a function of strap angle and strap phasing. Through the use of insulating antenna limiters, nearly all coupled antenna power at most strap angles can be attributed to radiation of fast waves at sufficiently high power levels. Far-forward microwave scattering measurements confirm the presence of the fast wave in the plasma core, and the core wave energy density inferred from the scattered signal has the same dependence on strap angle as the theoretically calculated fast wave radiation resistance. Using Thomson scattering, Langmuir probe, and bolometric diagnostics, increases in electron temperature and radiated power can account for 50-80% of the RF input power during fast wave heating experiments. However, the measured heating profile is much broader than fast wave ray-tracing theory predicts. Explaining this finding and more accurately modeling the microwave scattering results are important topics of future research.