Search

The article reviews the physics of fusion alpha particles and energetic neutral beam ions studied in the final phase of TFTR operation, with an emphasis on observations in reversed magnetic shear (RS) and enhanced reversed shear (ERS) DT plasmas. Energy resolved measurements of the radial profiles of confined, trapped alphas in RS plasmas exhibit reduced core alpha density with increasing alpha energy, in contrast to plasmas with normal monotonic shear. The measured profiles are consistent with predictions of increased alpha loss due to stochastic ripple diffusion and increased first orbit loss in RS plasmas. In experiments in which a short tritium beam pulse is injected into a deuterium RS plasma, the measured DT neutron emission is lower than standard predictions assuming first orbit loss and stochastic ripple diffusion of the beam ions. A microwave reflectometer measured the spatial localization of low toroidal mode number (n), alpha driven toroidal Alfven eigenmodes (TAEs) in DT RS discharges. Although the observed ballooning character of the n = 4 mode is consistent with predictions of a kinetic MHD stability code, the observed antiballooning nature of the n = 2 mode is not. Furthermore, the modelling does not show the observed strong dependence of mode frequency on n. These alpha driven TAEs do not cause measurable alpha loss in TFTR. Other Alfven frequency modes with n = 2-4 seen in both DT and DD ERS and RS discharges are localized to the weak magnetic shear region near qmin. In 10-20% of DT discharges, normal low n MHD activity causes alpha loss at levels above the first orbit loss rate.

The Pellet Charge Exchange (PCX) diagnostic on the Tokamak Fusion Test Reactor (TFTR) presently measures trapped alpha distribution functions with very small pitch angle (v {parallel} /v {approximately} 0.05) at the midplane. The measured PCX alpha signal exhibits a depletion region near the outboard region. Results of the alpha energy spectra and radial profile suggest stochastic ripple diffusion is the cause of the depletion. Comparison of the ripple stochastization boundary with Goldston-White-Boozer theory also shows the correct functional dependence on alpha energy and q-profile.

Radially-resolved energy and density distributions of the energetic confined alpha particles in D-T experiments on TFTR are being measured by active neutral particle analysis using low-Z impurity pellet injection. When injected into a high temperature plasma, an impurity pellet (e.g. Lithium or Boron) rapidly ablates forming an elongated cloud which is aligned with the magnetic field and moves with the pellet. This ablation cloud provides a dense target with which the alpha particles produced in D-T fusion reactions can charge exchange. A small fraction of the alpha particles incident on the pellet ablation cloud will be converted to helium neutrals whose energy is essentially unchanged by the charge transfer process. By measuring the resultant helium neutrals escaping from the plasma using a mass and energy resolving charge exchange analyzer, this technique offers a direct measurement of the energy distribution of the incident high-energy alpha particles. Other energetic ion species can be detected as well, such as tritons generated in D-D plasmas and H or He{sup 3} RF-driven minority ion tails. The diagnostic technique and its application on TFTR are described in detail.

Doublet plasma have been produced in Doublet IIA using an active field-shaping coil system. Significant increases in the electron density and energy confinement time have been observed for the doublet configuration relative to elliptic and circular plasmas in Doublet IIA.

Doublet II is a noncircular cross-section toroidal device with three magnetic axes. Theoretical considerations show that confinement of plasma in a doublet device might be similar to that of a tokamak with a much stronger toroidal magnetic field. Experiments carried out in Doublet II verify this in that the plasmas are found to have densities, temperatures, $q$ values, and confinement properties close to those of comparable circular-cross-section tokamak plasmas.

Using a neutral-beam injection power of 3.4 M W, volume-averaged toroidal betas of up to βT = 4.5% have been obtained in low-toroidal-field, low-qψ, vertically elongated discharges in the Doublet III tokamak. This level of βT is above the minimum level required for a tokamak reactor, thus demonstrating that reactor level values of βT are possible in a tokamak device. The observed enhancement of βT with vertical elongation lends confidence in the design of future devices which rely on vertical elongation.

Before a complete set of impurity control techniques were implemented in the Doublet III tokamak, two distinct discharge types were obtained, with nearly equal probability at high power densities with the same external control parameters. In the normal discharge (type S), sawtooth activity develops quickly and no impurity peaking on axis occurs thereafter. The other type of discharge (type O) initiates with more plasma/wall interaction, proceeds through the current-ramp phase with no central MHD activity, and impurity density profiles are found to be peaked on axis. Later in the current flat-top phase, type-O discharges experience the sudden onset of a large oscillation with m = 1, n = 1, and a concurrent decrease of central impurity densities. The (sinusoidal) oscillation eventually evolves into sawteeth, and type-O discharges attain a final state identical to type-S discharges. Measurements of absolute impurity densities show a direct correlation of impurity confinement with the presence or absence of tearing-mode activity, which suggests that the impurity transport is modified by the influence of the m = 1 magnetic island.

Analysis of the expanded boundary divertor in Doublet III reveals a medium density operating regime ( n e ≲ 4 × 10 13 cm -3 ), in which (1) the divertor density ne(Div) rises with the cube of the main plasma line average density, n e , and (2) the neutral pressure at the exhaust end of the divertor has a quadratic dependence on n e (with divertor plasma temperatures ≲ 15 eV). In this regime, injection of argon into the discharge results in a steady state concentration of argon, which is two orders of magnitude higher in the divertor plasma than in the main discharge. This trapping effect suggests that the argon concentration in the divertor region may be useful as an externally controllable parameter; that is, injected argon effectively cools the plasma boundary, yet is not deleterious to the plasma as a whole. A model which successfully describes the behavior of the edge and divertor plasmas in ohmically heated discharges suggests that these desirable trapping and radiative features may be maintained in the high powered neutral beam heating phase.

This report describes the experimental measurements and data analysis techniques used to evaluate the energy confinement in noncircular plasmas produced in Doublet III. Major aspects of the confinement measurements and analysis techniques are summarized. Machine parameters, diagnostic systems and discharge parameters relavent to the confinement measurements are given. Magnetic analysis techniques used to determine the plasma shape are reviewed. Scaling of the on-axis values of electron temperature, confinement time and Z/sub eff/ with plasma density is presented. Comparison with scaling results from other circular tokamaks is discussed. Numerical and analytic techniques developed for calculating the plasma energy confinement time and self-consistent profiles of density, temperature, current, and flux in non-circular geometries are described. These techniques are applied to the data and used to determine the central and global electron energy confinement time for a typical doublet plasma. Additional aspects of the confinement such as the radial dependence of the electron thermal conductivity and the estimated ion temperature are explored with the aid of a non-circular transport simulation code. The results of the confinement measurements are summarized and discussed. A brief summary of the theoretically expected effects of noncircularity on plasma confinement is included for reference as Appendix I.

Summary The first Neutral Injection System Beamline became operational on the Doublet III tokamak in September 1981. Power delivery tests and system debugging proceeded through December and initial experiments on plasma heating were started in January 1982. This first beamline has been running well and as of February 1982 is routinely delivering about 2.0 MW of neutral hydrogen power to various circular and elongated dee-shaped deuterium plasmas. The second beamline is now being brought on-line and we expect to have 5–6 MW of heating power available for definitive β studies in mid 1982. We find that the change in the total plasma stored energy caused by the beams is essentially independent of toroidal field and plasma current, but increases weakly with density. The total stored energy increases with current and density, but is independent of toroidal field. A volume-average βT= 2.2% is obtained in low toroidal field discharges.