Lower hybrid heating in the Alcator A tokamak

The results and interpretation of the modest-power (∼90 kW) lower-hybrid-heating experiment on Alcator A are presented. The expected results from linear waveguide-plasma coupling theory are outlined, and the possible effects of parametric instabilities, scattering from density fluctuations, and imperfect energetic ion confinement are addressed. It is found experimentally that good coupling and the absence of RF breakdown are achieved with a double waveguide array at available RF power densities PRF ≤ 4.5 kW.cm·−2, the waveguide vacuum windows being outside the toroidal field magnets; a waveguide array having vacuum windows near the waveguide mouth so that the ω = ωce layer can be pressurized shows no breakdown at PRP > 8 kW/cm2 when a single waveguide is energized. Energetic ion production and a factor-of-50 increase in the fusion neutron rate are observed to take place at well defined values of central plasma density; below these densities electron heating occurs. The ion tail production is found to be independent of the relative phase of the double waveguide array employed. This ion heating occurs at a lower density than theoretically expected; together with the electron heating this indicates waves having n|| ∼5 being absorbed near the plasma centre. Probes at the plasma edge observe a frequency-down-shifted and broadened RF pump signal that is strongly attenuated as the plasma density increases through the neutron production band. These anomalous heating results and probe signals can be explained by a parametric decay of the pump wave into higher n|| lower hybrid waves near the plasma edge. An alternate qualitative explanation would be the poloidal scattering of the lower hybrid waves at the plasma periphery due to density fluctuations; the n|| of the scattered lower hybrid waves would then increase as they propagated inward because of magnetic shear. The neutron rate decay times imply that the RF creates ion tails having a substantial fraction of their energy above 50 keV. The neutron decay times and rates strongly depend on plasma current and indicate the expected influence of ion confinement on RF heating efficiencies. Finally, the RF heating efficiencies are assessed.