| Visible spectroscopy, involving line profile and line intensity measurements, was used to study the turbulent heating of the rectangular cross-section high-beta tokamak Torus II. The spectroscopy was done in the visible wave-length region using a six channel polychrometer having 0.2 (ANGSTROM) resolution, which is capable of radial scans of the plasma. The plasma, obtained by ionizing helium, is heated by poloidal skin currents, induced by a rapid ((tau)(,R) (DBLTURN) 1.7 sec) change of the toroidal magnetic field either parallel or anti-parallel to the initial toroidal bias magnetic field, which converts a cold toroidal Z-pinch plasma into a hot tokamak plasma. In anti-parallel operation, line broadening of Helium II 4686 (ANGSTROM) and impurity lines (silicon and oxygen from the glass vacuum vessel walls) indicates that strong turbulence (E (,V) (DBLTURN) 10% and n(,e) (DBLTURN) 10('15)cm('-3), has an initially peaked radial ion temperature profile (T(,i) (DBLTURN) 180 eV at the plasma center for 25 GW anti-parallel heating power). Within (DBLTURN) 10 (mu)sec, the He II-determined ion temperature becomes nearly spatially flat (T(,i) (DBLTURN) 75 eV) and then decreases slowly in time. T(,i) and E scale roughly as the anti-parallel bias heating power. However, the frequency of the turbulence increases at lower ((DBLTURN) 12 GW) anti-parallel heating power. In parallel bias field operation, the induced skin current is lower by roughly a factor of 5 and measureable turbulence does not develop. The He II-determined ion temperature at 25 GW parallel heating power is comparable to that at 12 GW anti-parallel operation. The measured line profiles also indicate a macroscopic poloidal circulation of the plasma ions, which is charge and mass independent, and is possibly due to an E(,r) x B(,T) drift. The macroscopic radial electric field E(,r) (DBLTURN) 100 V/cm is strongest at the plasma edges corresponding to a poloidal rotation which vanishes at the plasma center. Electron temperature was determined from impurity ionization rate, line-to-continuum, and X-ray flux measurements and was found to be the same as T(,i) during the high beta equilibrium phase (after heating is completed). The impurity ionization rates also indicate (DBLTURN) 1% silicon and (DBLTURN) 2% oxygen impurity concentration; impurity line radiation accounts for most of the observed plasma cooling on a 15 (mu)sec time scale. Detals of turbulent heating derived from the analysis of ion line profiles are found to be consistent with theories describing heating from ion acoustic turbulence. The spectroscopically observed turbulence level can be used to estimate that the plasma electrical resistivity due to turbulence should be (DBLTURN) 100 times the classical Spitzer value; this is consistent with experimental values obtained with magnetic field probes. The level of ion-acoustic turbulence can be used to predict theoretically an ion heating rate which agrees with experiment. |
