D.S. Darrow

Attainment of high confinement in neutral beam heated divertor discharges in the PDX tokamak

Abstract The PDX divertor configuration has recently been converted from an open to a closed geometry to inhibit the return of neutral gas from the divertor region to the main chamber. Since then, operation in a regime with high energy confinement in neutral beam heated discharges (ASDEX H-mode) has been routine over a wide range of operating conditions. These H-mode discharges are characterized by a sudden drop in divertor density and H α emission and a spontaneous rise in main chamber plasma density during neutral beam injection. The confinement time is found to scale nearly linearly with plasma current, but can be degraded due either to the presence of edge instabilities or heavy gas puffing. Detailed Thomson scattering temperature profiles show high values of T c near the plasma edge (∼ 450 eV) with sharp radial gradients (∼ 400 eV/cm) near the separatrix. Density profiles are broad and also exhibit steep gradients close to the separatrix.

Measurement of loss of DT fusion products using scintillator detectors in TFTR

A poloidal array of MeV ion loss probes previously used to measure DD fusion product loss has been upgraded to measure the loss of alpha particles from DT plasmas in TFTR. The following improvements to the system have been made in preparation for the use of tritium in TFTR: (1) relocation of detectors to a neutronshielded enclosure in the basement to reduce neutron-induced background signals; (2) replacement of ZnS:Cu (P31) scintillators in the probes with the Y{sub 3}Al{sub 5}0{sub 12}:Ce(P46) variety to minimize damage and assure linearity at the fluxes anticipated from DT plasmas; and (3) shielding of the fiber optic bundles which carry the fight from the probes to the detectors to reduce neutron- and gamma-induced light within them. In addition to the above preparations, the probes have been absolutely calibrated for alpha particles by using the Van de Graaf accelerator at Los Alamos National Laboratory. Alpha particle losses from DT plasmas have been observed, and losses at the detector 901 below the midplane are consistent with first orbit loss.

Beta-induced Alfvén-acoustic eigenmodes in National Spherical Torus Experiment and DIII-D driven by beam ions

Kinetic theory and experimental observations of a special class of energetic particle driven instabilities called here beta-induced Alfven-acoustic eigenmodes (BAAEs) are reported confirming, previous results [N. N. Gorelenkov et al., Plasma Phys. Controlled Fusion 49, B371 (2007)]. The kinetic theory is based on the ballooning dispersion relation where the drift frequency effects are retained. BAAE gaps are recovered in kinetic theory. It is shown that the observed certain low-frequency instabilities on DIII-D [J. L. Luxon, Nucl. Fusion 42, 614 (2002)] and National Spherical Torus Experiment [M. Ono, S. M. Kaye, Y.-K. M. Peng et al., Nucl. Fusion 40, 557 (2000)] are consistent with their identification as BAAEs. BAAEs deteriorate the fast ion confinement in DIII-D and can have a similar effect in next-step fusion plasmas, especially if excited together with multiple global toroidicity-induced shear Alfven eigenmode instabilities. BAAEs can also be used to diagnose safety factor profiles, a technique know...

Alpha particle loss in the TFTR DT experiments

Alpha particle loss was measured during the TFTR DT experiments with a scintillator detector located at the vessel bottom in the ion Del B drift direction. The DT alpha particle loss to this detector was consistent with the calculated first orbit loss over the whole range of plasma current I=0.6-2.7 MA. In particular, the alpha particle loss rate per DT neutron at a given plasma current did not increase significantly with fusion power up to 10.7 MW, indicating the absence of any new collective alpha particle loss processes in these experiments

Initial results from the scoop limiter experiment in PDX

Abstract A particle scoop limiter with a graphite face backed by a 50 liter volume for collecting particles was used in PDX. Experiments were performed to test its particle control and power handling capabilities with up to 5 MW of D° power injected into D+ plasmas. Line average plasma densities of up to 8 × 1013 cm−3 and currents up to 450 kA were obtained. Plasma densities in the scoop channels greater than 2 × 1013 cm−3 and neutral densities in the scoop volume greater than 5 × 1014 cm−3 were observed. There is evidence that recycling may have occurred in the scoop channels for several discharges with large line-averaged plasma density. At beam powers up to 2.5 MW, energy confinement times above 40 ms were deduced from magnetics measurements and from transport analysis. Pressures in the vacuum vessel were in the 10 −5 Torr range, and recycling source neutral densities in the central plasma were low.

Modelling TF ripple loss of alpha particles in TFTR DT experiments

Modelling of TF ripple loss of alphas in DT experiments on TFTR now includes neoclassical calculations of first orbit loss, stochastic ripple diffusion, ripple trapping and collisional effects. A rapid way to simulate experiment has been developed which uses a simple stochastic domain model for TF ripple loss within the TRANSP analysis code, with the ripple diffusion threshold evaluated by comparison with more accurate but computationally expensive Hamiltonian co-ordinate guiding centre code simulations. Typical TF collisional ripple loss predictions are 6-10% loss of alphas for TFTR DT experiments at Ip=1.0-2.0 MA and R=2.52 m

Alfven frequency modes at the edge of TFTR plasmas

An Alfven frequency mode (AFM) is very often seen in TFTR neutral beam heated plasmas as well as in ohmic plasmas. This quasi-coherent mode has so far only been seen on magnetic fluctuation diagnostics (Mirnov coils). A close correlation between the plasma edge density and the mode activity (frequency and amplitude) has been observed, which indicates that the AFM is an edge localized mode with r/a>0.85. No direct impact of this mode on the plasma global performance or on fast ion loss (e.g., the alpha particles in DT experiments) has been observed. This mode is not the conventional TAE (toroidicity induced Alfven eigenmode). The present TAE theory cannot explain this observation. Other possible explanations are discussed

Anomalous delayed loss of trapped D-D fusion products in TFTR

A new anomalous delayed loss of D-D fusion products has been measured at the bottom of the TFTR vessel. This loss is delayed by about 0.2 s with respect to the usual prompt first orbit loss, and has a correspondingly lower energy, i.e. about half the fusion product birth energy. This loss process dominates the total fusion product loss measured 90° below the midplane for plasma currents I ≥ 1.8 MA and major radii near R = 2.45 m, e.g. for recent TFTR supershots. This delayed feature can occur without large coherent MHD activity, although it can be strongly modulated by such activity. Several possible causes for this phenomenon are discussed, but no clear explanation for this delayed loss has yet been found

ICRF heating of TFTR deuterium supershot plasmas in the 3He minority regime

The increased core electron temperature produced by ICRF heating of TFTR, D-T neutral-beam-heated supershot plasmas is expected to extend the alpha-particle slowing down time and hence enhance the central alpha-particle pressure. In preparation for the TFTR D-T operational phase, which started in late 1993, a series of experiments were conducted on TFTR to explore the effect of ICRF heating on the performance and stability of low-recycling, deuterium supershot plasmas in the 3He minority heating regime. The coupling of up to 7.4 MW of 47 MHz ICRF power to full size (R approximately 2.62 m, a approximately 0.96 m), 3He minority, deuterium supershots heated with up to 30 MW of deuterium neutral beam injection has resulted in a significant increase in core electron temperature ( Delta Te=3-4 keV). Simulations of equivalent D-T supershots predict that such ICRF heating should result in approximately a 60% increase in the alpha-particle slowing down time and an enhancement of about 30% in the central alpha pressure. Future experiments to be conducted at ICRF powers up to 12.5 MW during the upcoming TFTR D-T campaign may result in even greater enhancements in core alpha parameters. This paper presents results from experiments performed at an axial toroidal magnetic field of about 4.8 T, where the minority resonance was within 0.1-0.15 m of the plasma core. Combined ICRF and neutral beam heating powers in these experiments reached TFTR record levels of over 37 MW, which allowed an exploration of the power loading limits on the carbon limiter tiles. The plasma current was operated at 1.85 and 2.2 MA and sawtooth suppression was observed at the higher plasma current.

Measurements of DT alpha particle loss near the outer midplane of TFTR

Measurements of DT alpha particle loss to the outer midplane region of TFTR have been made using a radially movable scintillator detector. The conclusion from this data is that mechanisms determining the DT alpha loss to the outer midplane are not substantially different from those for DD fusion products. Some of these results are compared with a simplified theoretical model for TF ripple induced alpha loss, which is expected to be the dominant classical alpha loss mechanism near the outer midplane. An example is given of plasma driven MHD induced alpha particle loss to the outer midplane, but no sign of any `collective` alpha instability induced alpha particle loss has yet been observed at this location