Rotation scalings and momentum confinement in neutral-beam-injected ISX-B plasmas

Abstract

Scalings of the central rotation in non-gettered, co-injected ISX-B discharges have been measured as a function of beam power, electron density and plasma current. Extensive studies are made possible by exploiting charge-exchange excitation (CXE) of 0 VIII lines to measure Doppler shifts. The rotation velocity, vϕ(0), tends to saturate at (1.0 − 1.2) × l07 cm·s−1 when Pb≅0.5 MW, showing little further increase up to the maximum input of 2 MW; vϕ(0) is independent of ne and Ip. Momentum confinement times in quasi-steady plasmas are 10–16 ms for n̄e = 4.5 × 1013 cm−3. Counter-injection discharges always disrupt, but before this event vϕ(0) is the same as for co-injection plasmas. The addition of a third beam line, permitting injection of up to 2 MW of balanced neutral-beam power, has allowed comparisons of the energy and particle confinement in rotating and non-rotating plasmas with the same total neutral-beam input. In those cases where impurity buildup can be avoided, it is found that the ISX-B empirical scaling of energy confinement time is reproduced with balanced injection. Thus, the unfavourable dependence of is not the result of rotation. Studies of impurity behaviour under differing injection conditions have been extended to include fully stripped low-Z ions. The results are consistent with previous investigations of metallic elements which revealed strong dependences on the sense (co versus counter) of injection. The potentials calculated from momentum balance, using measured rotation profiles and typical plasma density and temperature profiles, are in qualitative agreement with the potentials measured directly for various combinations of co- and counter-injection.

Interpretation of measurements of the global momentum and energy confinement times in strongly rotating tokamak plasmas

“…(33) predicts that T decay > TJ. Decay constants are substantially larger than momentum confinement times in experiments where both are measured [3,6], which is qualitatively consistent with Eq. 33and a rapidly decreasing density.…”

“…An extensive set of central rotation velocity measurements, including a number of parameter dependences, was made on ISX-B [3]. The analysis, based on gyroviscous theory for the transfer of input torque, was able to explain these data [12].…”

“…When unbalanced neutral beam injection (NBI) is used to heat a tokamak plasma, the resulting torque causes the plasma to rotate. Rotation speeds comparable to the thermal speeds of impurity ions present in the plasma and well within a factor of ten of the thermal speed of the main plasma ion species are routinely achieved (see, for example, Refs [1][2][3][4][5][6][7]). In rotating plasmas there are new transport effects associated with the viscous radial transfer of momentum and with the plasma inertia which are not found in non-rotating plasmas.…”

Scaling of rotation and momentum confinement in JET plasmas

“…Previous analysis had shown a relationship between energy and momentum confinement [3,4,5,6,7]. It is often reported that these quantities have comparable magnitudes and have similar dependencies with individual plasma parameters.…”

X-ray observations of central toroidal rotation in ohmic Alcator C-Mod plasmas

“…Typical values of the rotation velocity with neutral beam input are in the range of a few x 10' cm/s. Some toroidal rotation during purely ohmic plasmas has been observed [1,5,7,14], and there have been some ancillary measurements of toroidal rotation during the ohmic phase of neutral beam plasmas [2,3,4,6,10]. The ohmic results are summarized in Table I.…”