Search

A review of TFTR plasma transport studies is presented. Parallel transport and the confinement of suprathermal ions are found to be relatively well described by theory. Cross-field transport of the thermal plasma, however, is anomalous with the momentum diffusivity being comparable to the ion thermal diffusivity and larger than the electron thermal diffusivity in neutral beam heated discharges. Perturbative experiments have studied nonlinear dependencies in the transport coefficients and examined the role of possible nonlocal phenomena. The underlying turbulence has been studied using microwave scattering, beam emission spectroscopy and microwave reflectometry over a much broader range in k perpendicular to than previously possible. Results indicate the existence of large-wavelength fluctuations correlated with enhanced transport.

Recent operation of the Tokamak Fusion Test Reactor TFTR, has produced plasma equilibria with values of {Lambda} {triple bond} {beta}{sub p eq} + l{sub i}/2 as large as 7, {epsilon}{beta}{sub p dia} {triple bond} 2{mu}{sub 0}{epsilon}/{much lt}B{sub p}{much gt}{sup 2} as large as 1.6, and Troyon normalized diamagnetic beta, {beta}{sub N dia} {triple bond} 10{sup 8} aB{sub 0}/I{sub p} as large as 4.7. When {epsilon}{beta}{sub p dia} {approx gt} 1.25, a separatrix entered the vacuum chamber, producing a naturally diverted discharge which was sustained for many energy confinement times, {tau}{sub E}. The largest values of {epsilon}{beta}{sub p} and plasma stored energy were obtained when the plasma current was ramped down prior to neutral beam injection. The measured peak ion and electron temperatures were as large as 24 keV and 8.5 keV, respectively. Plasma stored energy in excess of 2.5 MJ and {tau}{sub E} greater than 130 msec were obtained. Confinement times of greater than 3 times that expected from L-mode predictions have been achieved. The fusion power gain. Q{sub DD}, reached a values of 1.3 {times} 10{sup {minus}3} in a discharge with I{sub p} = 1 MA and {epsilon}{beta}{sub p dia} = 0.85. A large, sustained negative loop voltage duringmore » the steady state portion of the discharge indicates that a substantial non-inductive component of I{sub p} exists in these plasmas. Transport code analysis indicates that the bootstrap current constitutes up to 65% of I{sup p}. Magnetohydrodynamic (MHD) ballooning stability analysis shows that while these plasmas are near, or at the {beta}{sub p} limit, the pressure gradient in the plasma core is in the first region of stability to high-n modes. 24 refs., 10 figs.« less

Recent experiments in the DIII-D tokamak [J. L. Luxon, Nucl. Fusion 42, 614 (2002)] show that the resistive wall mode (RWM) can be stabilized by smaller values of plasma rotation than previously reported. Stable discharges have been observed with beta up to 1.4 times the no-wall kink stability limit and ion rotation velocity (measured from CVI emission) less than 0.3% of the Alfven speed at all integer rational surfaces, in contrast with previous DIII-D experiments that indicated critical values of 0.7%–2.5% of the local Alfven speed. Preliminary stability calculations for these discharges, using ideal magnetohydrodynamics with a drift-kinetic dissipation model, are consistent with the new experimental results. A key feature of these experiments is that slow plasma rotation was achieved by reducing the neutral beam torque. Earlier experiments with strong neutral beam torque used “magnetic braking” by applied magnetic perturbations to slow the rotation, and resonant effects of these perturbations may have ...

In an effort to understand the role of neutral beam orientation and plasma geometry in stabilizing fishbone activity in PBX, comparisons were made between near-perpendicular and tangential injection, and discharges with and without significant (>10%) indentation. Periodic spikes in the fast neutral flux and drops in the neutron production, reminiscent of sawtooth oscillations observed during tangential injection in PLT, were observed during tangential injection in PBX. Changes in the fast neutral signal and neutron emissivity were also measured in association with high frequency (120-210 kHz) instabilities in these discharges. Although neither phenomenon has been correlated yet with a mechanism like fishbones for expelling beam ions from the plasma, their depletion from the center of the discharge suggests the need for controlling them to ensure efficient heating of the plasma center with tangential beams.

The dissipative trapped ion instability (DTII) is observed as an m=1 mode localized to the mirror cell of the Columbia Linear Machine (CLM) [Plasma Phys. 24, 185 (1982)]. A gated feedback diagnostic allows the direct measurement of the DTII linear growth rate. The DTII growth rate is measured as a function of trapped fraction, keeping all other parameters fixed, and clearly displays the scaling with trapped fraction predicted by the radially local linear dispersion relation, including ion Landau damping. The saturated DTII mode amplitude is measured concurrently with the growth rate. The saturated amplitude displays a nearly linear dependence on the measured growth rate, normalized to the DTII real frequency, and is nearly proportional to the square root of the measured growth rate. A striped particle collector is used to measure the ambipolar radial plasma flux as the trapped fraction is varied. Radial diffusion coefficients are calculated from the measured radial particle fluxes and display a scaling in...

Density fluctuations, δn n ≲ 20%, were observed to have no effect on the classical diffusion of a collisional helium plasma across a poloidal octupole field with an additional weak toroidal field. The implication of these results on tokamak scattering experiments is discussed.

The Columbia Linear Machine has been redesigned to enter the parameter regime of the collisionless, curvature-driven, trapped-particle instability. With the new machine configuration, a strong ($\frac{\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{n}}{N}\ensuremath{\gtrsim}60%$), coherent mode is observed which bears all the characteristics of this instability. These include strong localization to the mirror cell, broad radial-mode structure, and MHD-like body motion of the plasma column. Furthermore, the mode exhibits all the correct parametric dependences of increasing mode amplitude with increasing mirror ratio and rf power, and decreasing mirror-cell length.

The operational characteristics and performance of the two channel 10 Megawatt MHD feedback control system as installed by Los Alamos National Laboratory on the Columbia University HBT-EP tokamak are described. In the present configuration, driving independent 300 uH saddle coil sets, each channel can deliver 1100 Amperes and 16 kV peak to peak. Full power bandwidth is about 12 kHz, with capabilities at reduced power to 30 MHz. The present system topology is designed to suppress magnetohydrodynamic activity with m=2, n=1 symmetry. Application of either static (single phase) or rotating (twin phased) magnetic perturbations shows the ability to spin up or slow down the plasma, and also prevent (or cause) so-called "mode-locking". Open loop and active feedback experiments using a digital signal processor (DSP) have been performed on the HBT-EP tokamak and initial results show the ability to manipulate the plasma MHD mode frequency.

The Columbia Linear Machine [Plasma Phys. 24, 185 (1982)] has been redesigned to enter the parameter regime of the collisionless, curvature driven, trapped particle instability. The changes that were made reduced the neutral collisionality by over a factor of 2 and the Coulomb collisionality by an order of magnitude, and increased the curvature drive by nearly an order of magnitude. With the new machine configuration, a strong (n/N≳60%), coherent, m=1 fluctuation is observed which is identified as a collisionless, curvature and rotationally driven, trapped particle instability. The mode is localized strongly to the region of trapping. The radial structure of the mode is broad and the mode exhibits magnetohydrodynamiclike body motion of the plasma column. Furthermore, the mode displays the expected parametric dependences of increasing mode amplitude with increasing mirror ratio and rf power and decreasing mirror cell length. A theory whose predictions are in fair agreement with the experiment has been dev...

External n = 1 magnetic fields are applied in RFX-mod and DIII-D low safety factor Tokamak plasmas to investigate their interaction with the internal MHD dynamics and in particular with the sawtooth instability. In these experiments the applied magnetic fields cause a reduction of both the sawtooth amplitude and period, leading to an overall stabilizing effect on the oscillations. In RFX-mod sawteeth eventually disappear and are replaced by a stationary m = 1, n = 1 helical equilibrium without an increase in disruptivity. However toroidal rotation is significantly reduced in these plasmas, thus it is likely that the sawtooth mitigation in these experiments is due to the combination of the helically deformed core and the reduced rotation. The former effect is qualitatively well reproduced by nonlinear MHD simulations performed with the PIXIE3D code. The results obtained in these RFX-mod experiments motivated similar ones in DIII-D L-mode diverted Tokamak plasmas at low q 95. These experiments succeeded in reproducing the sawtooth mitigation with the approach developed in RFX-mod. In DIII-D this effect is correlated with a clear increase of the n = 1 plasma response, that indicates an enhancement of the coupling to the marginally stable n = 1 external kink, as simulations with the linear MHD code IPEC suggest. A significant rotation braking in the plasma core is also observed in DIII-D. Numerical calculations of the neoclassical toroidal viscosity (NTV) carried out with PENT identify this torque as a possible contributor for this effect.