Please note the following report is for informational use. It may not be cited as a scientific reference nor may it be quoted in publications without permission. Contact Jim Terry (terry@psfc.mit.edu) with comments or questions.



Quarterly Progress Report on Alcator C-Mod - July 1997

The primary activities at Alcator C-Mod during the third quarter of FY97 were resumption of tokamak plasma operation, execution of part of the 1997 run campaign, continued work on major systems upgrades (the RF transmitters and the Diagnostic Neutral Beam), continued analysis of C-Mod data from previous run campaigns, preparation for review of the C-Mod 5-year plan at the National Tokamak Workshop, and continued participation in the critical assessment of fusion science. A more detailed account of these activities follows.




Scientific Results
 With the resumption of plasma operation on Alcator C-Mod a number of new experimental studies were carried out.

The Fast Scanning Probe had been modified to include an inertial valve which injects trace impurity gases at known and controllable radii in the SOL, up to the last closed flux surface. Using N2 as the trace gas, plumes in NII and NIII emission were observed using a TV camera (the `plume' camera) viewing the resulting emission plume from above. Detailed analysis of these data is underway with our U. of Toronto collaborators (Stangeby and Lisgo) using the DIVIMP impurity transport code. The local plasma conditions at the injection point were significantly perturbed by the plume, even for the relatively small amount of N2 injected in these shots, and these effects need to be included in the interpretation. Larger amounts of N2 were injected into low-density targets ([`n]e ~ 1×1020 m-3) using the probe valve, in order to investigate screening in the SOL. For injection first at the LCFS, and then into the SOL at 4mm outside the separatrix, the intensities and time-histories of the Li-like N emission along a midplane chord, which does not view the plume, were the same. Some Doppler shift data indicative of flows may be available from these larger N2 puffs, but these injections were highly perturbing, so the data may be difficult to interpret. The gas delivery system works according to design, depositing up to 1019 molecules in the 10 ms deacceleration period during the scanning probe's end-of-stroke. Tests indicate that bypass leaks in the gas delivery capillary are small, about 1018 molecules/second for the largest plenum pressures. The `plume-camera' also recorded emission patterns in HeI and HeII when He was puffed.





Fig. 1 - The main plasma brightness of Li-like N is compared

for the cases with the cryopump on and off. The same amount

of N2 was puffed in each case. The cyropump has also been

shown to affect puffed, recycling impurities Ne and Ar.


The prototype divertor cryopump was operated for the first time to investigate its effect on impurity transport and pumping. Pumping speeds in the range of 1000-1500 l/sec were determined. In nitrogen injection experiments, pump operation reduced nitrogen concentrations in the core by about a factor of two (see Fig. 1). In the private flux region of the divertor the reduction was even greater. The cryopump also pumped the puffed recycling impurities, Ar and Ne. The core electron density was also affected by the pump, since slightly larger puff rates were required to obtain the same density during the flattop portion of the discharge when the pump was on. In addition, the density pump-out (occurring during the ramp-down phase) was faster with the cryopump on. These results were obtained with a single prototype pump, representing about 10% of the ultimate pumping speed, once the full cryopump system is installed.




Fig.2 - Results for the Ar compression experiment


Initial experiments were also carried out to determine where the injected impurities which do not go into the core plasma reside. The new Divertor RGA `Prisma' was used to measure the neutral density of the injected gas in the divertor. The core levels were determined from absolutely calibrated VUV and x-ray measurements of impurity line radiation. The amounts of gas injected are determined from calibrations of the piezo valves used. Argon was the injected impurity used in this experiment; future runs will involve other gases, including N2 which behaves as a non-recycling impurity, and will include H-mode targets. Analysis of the RGA data using the calibration numbers from the offline vacuum test stand work indicates high Ar neutral densities in the divertor (of order .1 mTorr Ar) with nAr/nD2 ~ .3% in that region. The compression of Ar into the divertor is also quite high, with nArdiv/nArcore ~ 150. Based on these numbers we can account for all of the injected Ar. Time histories of these and other quantities are shown in Fig. 2. A final online calibration must still be performed on the tokamak to confirm the values, which are high compared to those measured in similar experiments on AUG and DIIID.

A series of experiments testing the feasibility of fast disruption termination using massive D2 pellet injection was carried out. This experiment is partly motivated by a request from the ITER JCT. Earlier experiments on C-Mod and other tokamaks have shown that it is relatively easy to use high-Z `killer pellets' to radiate quickly all of the plasma thermal energy. However, theoretical work by Marshall Rosenbluth predicts that injection of high-Z pellets (such as our Ag-doped lithium pellets) will result in dangerous high energy runaway electron populations via an avalanche mechanism. Furthermore, theory predicts that the problem is lessened by injecting low-Z material rather than high-Z. ITER is so concerned about this prediction that they have proposed using fast injection of massive amounts of deuterium (or hydrogen or helium) for fast and/or preemptive termination of the discharge. The present experiments were designed to test whether this technique actually speeds up the current quench. Alcator C-Mod has a 20-barrel cryogenic deuterium pellet injector with the unique capability of having all barrels fire simultaneously. In this mode of operation, a mixture of pellet sizes having a total of 3×1021 deuterons can be launched at a somewhat reduced velocity of 800 m/s. The massive D2 pellet injection raises the density 10 to 15-fold, resulting in [`n]e @ 1.6×1021 m-3. Thomson scattering measurements of the density and temperature profiles were taken just after the injection but just prior to the disruption thermal quench. The density profile was extremely hollow, and Te dropped to the few hundred eV range. This is consistent with simple adiabatic cooling. The results conclusively showed that massive D2 pellet injection does NOT speed up the current quench in C-Mod, nor does it reduce the halo current. No sign of runaways electrons was observed. The details of how these effects scale to ITER are still not tested experimentally. The same result was obtained when we tried to kill a VDE. Simultaneous injection of D2 pellets and tungsten (from the laser blowoff) did not do any better. We were able to ``kill'' Ohmic, L-mode, ELMy H-mode, and ELM-free H-modes plasmas, with identical results.





Fig. 3 - The recombination rate per unit area determined

for different viewed areas of the inner and outer divertor plates.

The inner divertor plate is detached, while the outer is not. Also

shown are the viewing chords, the LCFS, and the strike points.


The study of volume recombination in the C-Mod divertor has continued. We had already determined that significant volume recombination of the majority ions occurs under detached conditions. The recombination regions have now been spatially resolved. The distribution of the recombination has been measured for a case where the inner divertor plate is detached, while the outer plate is not. The recombination rate in the detached region in front of the inner divertor plate is approximately a factor of two higher than the rate of ion collection at the plate. For the attached plate, the recombination occurring in front of it is still small in comparison to the ion current to the plate. For both plates the recombination is peaked near the strike points. These results are shown in Fig. 3.

In collaboration with the group at the University of Texas, initial comparisons between C-Mod data and the IFS-PPPL transport model were made. The theory does well in predicting the ion temperature profiles for H-mode plasmas, but yields substantial underestimates for L-mode. The calculated electron temperatures are larger than the measurements. We note that the collisionality range for the data is outside the limits used for interpolating the gyro-fluid simulations. Work has commenced to extend the fluid calculations to include C-Mod conditions. Gyro-kinetic simulations will be run to look for additional modes which may be affecting C-Mod results. Details can be found in the PhD thesis ``Local Transport Analysis on the Alcator C-Mod Tokamak'' by Jeff Schachter.

The theory and computational activities at C-Mod which are directly involved with interpretation and understanding of C-Mod observations include:

1)
Global modelling of halo currents during C-Mod disruptions. The TSPS-3D code has been developed by McCarrick and Freidberg as part of a Ph.D. thesis provides a new tool for studying non-axisymmetric disruptions. The original work showed a good agreement between the measurements and predictions of ``typical'' disruptions. We are continuing to exercise the code, in particular applying it to the outlying high halo current disruptions and to radial disruptions. The code continues to indicate that the forces generated by the toroidal eddy currents (which are not measured) are larger than the forces that result from the (measured) poloidal halo currents.
2)
An enhanced D-alpha H-mode which is a high confinement plasma with substantially lower reduced impurity buildup is observed on C-Mod. This mode of operation has been observed to be accompanied by both broadband density fluctuations (up to 400 KHz) and by bursting low-m magnetic fluctuations. We are hoping to obtain a theoretical understanding of the underlying physics that causes the observed behavior. One approach that is being examined is an edge localized resistive MHD mode, which would be predicted by the MARS code. A second candidate is an edge localized micro-tearing mode, driven unstable in the very high temperature and density gradients that exist in the edge region of a high quality H-mode [L.L. Lao et al., NF 30, 1035 (1990), NF 36 965 (1996).]. We have determined that the dispersion relation produces the prediction of instability in the parameter range that is measured in C-Mod and we are presently calculating the detailed eigenmodes to add to our understanding of this instability. Additionally three dimensional simulations of drift-ballooning by Rogers and Drake at the University of Maryland indicate the possibility of enhanced pressure gradient driven transport due to ballooning-type modes well below the ideal MHD limit [Rogers and J.F. Drake, to be published in PRL (1997)]. We will pursue simulations in the C-Mod parameter range in collaboration with the Maryland group.






Operations and Diagnostics

Tokamak plasma operations resumed on Alcator C-Mod on May 14. As expected, the initial plasmas were dirty and resistive. The fact that a plasma was obtained with only very minor adjustment from settings established last year indicates that control and power systems were indeed behaving nominally, and no significant perturbations were introduced during the disassembly/reassembly process. Six weeks of commissioning and plasma experiments followed. (See the results discussed above and below.) On June 26th, following a successful 1.2MA, 8T shot, the alternator, which provides AC power for the experiment, experienced an electrical fault indication. The protection and safety circuits functioned as designed, and the system shut down and was secured without incident. A ground fault on the rotor was found. General Electric, which manufactured the alternator, and carried out an extensive inspection and refurbishing of the alternator last year, has evaluated the status of the system and estimate that repairs can be accomplished in about ten weeks. Assuming the repair work is started soon, we expect to continue the run campaign in October.

A number of the runs involved commissioning and initial experiments with the 40 MHz ICRF system. In one experiment the toroidal field was ramped down first to 4T and then to 2.8T; the latter is appropriate for D(H) heating with the 40MHz system. A good ELM-free RF H-mode was obtained at the lower field with only 0.4MW of RF power, when the hydrogen fraction was sufficiently low for efficient minority heating. The new 1mm spatial resolution soft x-ray array was able to resolve clearly the steep gradient in the barrier region.

The RF ran reliably at 40 MHz at a power level of 2MW, and preliminary exploration of heating scenarios using the 40 MHz system was carried out. Because of the high H concentration (over 10% on shot 1, increasing to about 50% later in the run), D(H) minority heating or standard D(3He) minority heating scenarios were not accessible, so a D-3He-H mode conversion experiment was carried out. Toroidal field and 3He concentration were varied, at currents of 0.8MA and 1MA. It was found that high 3He concentration (about 10 torr-l) and ``high field'' (around 5T) produced the best heating for both electrons and ions. Significant neutron rates were observed, possibly indicating generation of a deuterium tail. Cynthia Phillips and Randy Wilson of PPPL participated in these RF experiments, and will be involved in the data analysis. Operation with the 40 MHz system provided an opportunity to debug the hardware and gave valuable experience with this new system, which will be the basis of current drive and heating experiments in campaigns scheduled for next year. After these experiments, the 40 MHz transmission line and tuners were disassembled, and the lines for the 80 MHz system were reinstalled. This system will be used for the bulk of the physics program for the remainder of the present campaign. Testing and conditioning the 80 MHz system was begun and was fairly successful. After boronization, ELM-free H-modes were obtained on most discharges with at least 1MW of RF power, indicating good wall conditions.

The deuterium pellet injector (DPI) is operational during this campaign. Initially two pellets were injected on each of three shots, late in the discharge. ``Snake'' phenomena were observed on several diagnostics, including the ECE, following several of the pellets. A decrease in the toroidal rotation rate, as measured by high-resolution x-ray spectroscopy (HIREX), was also observed following pellet injection. After these experiments the DPI was used in the disruption termination studies (see above).

The lithium pellet injector (LPI) system was installed and is operational. The LPI was relocated from K to C port, in anticipation of the installation of the new ICRF antenna next year.

The core Thomson scattering system, which is now using a 30 Hz NdYAG laser borrowed from PPPL, is operational.

The new inner cylinder flux loops, which detect radial fields and are sensitive to motion and alignment of the central solenoid as well as providing information on disruption dynamics, were shown to work.

The tangential two-color interferometer (TTCI), which will provide density information based on a midplane view, has been installed.

A new diagnostic, the omegatron ion mass spectrometer, recorded its first ion spectrum during operation the week of June 2. The omegatron combines a gridded energy analyzer with an ion mass spectrometer. The ion mass spectrometer exploits the local magnetic field (approximately the toroidal field on axis) and RF potentials applied to the cavity to separate ions with the resonant cyclotron frequency.

Two new diagnostics installed to the study the role that neutrals play in C-Mod were successfully commissioned. They measure radiated power, but are not normally sensitive to neutrals (in contrast to typical foil bolometers). One of these silicon diode arrays views a horizontal cross-section of the plasma, while the second views the extreme edge of the plasma with high radial resolution (2mm). Effects of sawteeth and impurity injections have been readily observed on the wide view array (core array). The edge array, on the other hand, confirmed the very steep gradient at the edge, which, according to preliminarily analysis, gets much steeper during H modes.

The Diagnostic Neutral Beam (DNB) test lab is being constructed. Access holes for the DNB transmission lines have been installed. The as-built drawing of the Faraday cage has been completed, and work on modifications needed for its installation on C-Mod has begun. The Mod/Reg tube has been installed. A preliminary design of the high voltage transmission line is complete.


C-Mod Web Sites
 When the machine or ECDC is operating, a new WEB page is available to display invessel video at:

http://www.psfc.mit.edu/cmod/inside.html

This wide angle view is updated approximately every 10 seconds, and when ECDC is operating, shows much of the invessel hardware. The discharge moves between the inner and outer wall as the toroidal field is scanned.

During tokamak operation, a live view of the control room is now provided at:

http://www.psfc.mit.edu/cmod/control_room.html

A new set of invessel survey images is available:

http://www.psfc.mit.edu/cgi/cmod_images?name=invessel/

with prefixes of ``survey''. These pictures are useful for new diagnostics and invessel installations. These images, as well as the live view of the experimental cell, are all accessible from the ``Facility On-line'' section of the C-Mod links page at

http://www.psfc.mit.edu/cmod/cmod_links.html




The National Tokamak Workshop

As part of long range planning for the US fusion program, a National Tokamak Workshop was held July 8-10, 1997 to discuss the programs of the two major US tokamaks, DIIID and Alcator C-Mod. A ``White Paper'' was written which outlines the Alcator group's proposal for what the C-Mod program should consist of during approximately the next five years. This document can be found on the WEB at:

http://cmod.pfc.mit.edu/ ~ hutch/whitepaper/overview.html

http://cmod.pfc.mit.edu/ ~ hutch/whitepaper/transport.html

http://cmod.pfc.mit.edu/ ~ hutch/whitepaper/divertor.html

http://cmod.pfc.mit.edu/ ~ hutch/whitepaper/rf_at.html

http://cmod.pfc.mit.edu/ ~ hutch/whitepaper/mhd.html

The paper and the presentations at the workshop outline the future research plans on Alcator C-Mod. The research will continue to contribute to and support the paths foreseen for the development of fusion energy. These paths are ITER, a Compact High-Field Ignition Experiment, an Advanced Tokamak Reactor, and Alternate Confinement Concepts. Ian Hutchinson, Earl Marmar, Martin Greenwald, Bob Granetz, Jim Terry, Yuichi Takase, Miklos Porkolab, Brian LaBombard, Amanda Hubbard, and Steve Wolfe presented the C-Mod plans and participated in the reviews and discussions at the workshop. The responses to the C-Mod plans were positive and encouraging. Documentation of the review and recommendations will be forthcoming.




Collaborations and Participation in the Fusion Science Community

Scattered laser light from the plasma was observed on the X-point Thomson scattering system on several C-Mod discharges. The peak in the scattering amplitude appears to occur near the X-point. The scattered signal was observed during a horizontal scan of the laser beam with respect to the viewing optics and occurred at a position which was to one side of the nominal center. This and other data indicate that the next issue that must be resolved, before the X-point Thomson scattering system becomes a routine diagnostic, is that of maintaining alignment both during the plasma shot and long term during over a run day. The required alignment system has already been partially implemented. The X-point Thomson scattering system is a collaborative effort among PPPL, MIT and Princeton Scientific Instruments.

Our collaboration with the Univ. of Maryland (Welch, Weaver, and Griem) provides high (spectral) resolution measurements of visible/UV lines at the plasma edge. This experiment is now routinely monitoring the ratio of the Ha to Da emission, which indicates the relative hydrogen concentration in the discharge. The H-to-(D+H) ratio which was ~ 40% after each of two vacuum breaks at the beginning of the run campaign decreased < 10% during the subsequent first 15 operating days. Doppler shifts indicative of plasma flows are also observed. Preliminary analysis indicates neutral deuterium flows at up to 5 km/sec and ionized carbon (C2+) flows at up to 10 km/sec. Two new tangential views have provided a clear indication of these flows. The views result from two fibers looking in opposite toroidal directions. The measured red shift of one view and the blue shift on the other view shows symmetric shifts and increases the resolution of the flow measurement by a factor of 2.

The Kodak fast framing visible camera (1000 fps with a controllable electron shutter) from Los Alamos National Labs (Maqueda and Wurden) was set up to view tangentially at the outer midplane. Without any filter present on the camera, flying ``macro'' particles were observed, together with bursts of highly emitting plasma that originated from the Fast-Scanning Probe puffs which were in the field-of-view. This highly radiating plasma then drifted toroidally/poloidally.

Using the fast visible Kodak camera with an Da filter, and in collaboration with Stewart Zweben from PPPL, edge turbulence filaments were observed near the RF antenna. These filaments, with a poloidal wavelength of approx. 2-4 cm (as observed with 10msec exposure), are probably the low frequency ( ~ 50 kHz), largest-scale part of the normal edge turbulence spectrum. No correlation from frame to frame was observed at the 1 kHz framing rate. We plan to improve the imaging of this turbulence by using a close-up view and shorter exposures to see smaller scales, and by reducing the RF pickup.

The LANL collaborators (Maqueda and Wurden) have also designed a new IR imaging system which has been installed at C-Mod. This system is based on an Amber Radiance IR camera and a periscope utilizing ZnSe optics. The IR images of the divertor region obtained during the May cleanup period (i.e. during ECDC) had spatial resolution, field depth, and field of view consistent with ray tracing simulations and bench tests. An example of one of the available, in-vessel views of C-Mod can be seen at:

http://wsx.lanl.gov/ricky/ecdc_new.gif

Despite this proof for the imaging capabilities of the system, when tokamak operations started no images were obtained. This was because the optics were damaged by the extremes of heat and cold which the re-entrant periscope experienced. Plans are underway to provide better temperature control of the periscope environment.

A Motional Stark Effect (MSE) instrument to measure q(r) using the Diagnostic Neutral Beam (DNB) is being designed by N. Bretz at PPPL in cooperation with W. Rowan and R. Bravenec (FRC, Univ. of Texas, Austin) and J. Terry and E. Marmar (PFC/MIT). Several meetings at MIT (Preliminary MSE Design Presentation, 5/13/97 and C-Mod Active Spectroscopy Design Workshop, 7/14/97) have resulted in refinements on this design particularly with respect to the MSE optical system which must view the neutral beam. A code has been written to simulate the expected signal levels from an arbitrary midplane geometry, and critical features of the design have been identified. A final system has not yet been determined, but the implementation of combined MSE and BES viewing optics with re-entrant lenses appears to be the best choice and is compatible with active CXRS measurements of rotation and ion temperature. PPPL intends to supply all of the external detectors, filters, amplifiers, and data acquisition equipment from the TFTR MSE system.

Graduate student Maxim Umansky has recently starting working with members the MIT Divertor Theory group (O. Batischev and S. Krashinninikov) to help model new divertor geometries for Alcator C-Mod. The effects of incorporating a septum and/or providing for active neutral recycling flux control are being considered. Preliminary results were presented at the recent National Tokamak Workshop.

Doug McCune and Jim Hirsch of PPPL visited C-Mod to discuss implementation of MDSplus (the C-Mod data system software) on NSTX. After this visit MDSplus was installed on a VMS cluster at PPPL, allowing evaluation of this software for NSTX use. While no final decision has been made, it seems likely that PPPL will use the software for NSTX. Similar discussions have begun with GA. DIII-D is planning a major upgrade of their data system and is considering the use of MDSplus as part of that upgrade. Following the National Tokamak Workshop, Martin Greenwald spent a day with GA staff discussing these issues. Key to success at DIII-D will be a timely port of the MDSplus server to UNIX. (The client side of MDSplus has already been ported to various flavors of UNIX and to Windows NT/95.) A demonstration system has been set up at GA which gives access to PTDATA via MDSplus and a general migration plan is being outlined. Qian Peng, an applications programmer with DIII-D, has visited MIT to discuss issues related to migrating EFIT to an MDSplus environment. A group of physicists and programmers will be visiting MIT the last week in July for further talks. The advantages of uniform access to experimental data and analysis across the whole tokamak program are obvious.

127 time-slices of 1996 C-Mod data have been added to the new ITER pedestal database. This submission is in support of work on the ITER Urgent Task on characterization of the H-mode pedestal.

Martin Greenwald, Steve Wolfe, and Joe Snipes attended a joint ITER Experts Group Meeting and Confinement Database and Modeling meeting in San Diego, 14-18 April 1997. Joe Snipes presented an analysis of the H-mode threshold linear regressions using data from 9 different tokamaks in the ITER H-mode threshold database. This is shown in Fig. 4. In addition, he has generated a fit of the edge Te at the L-H transition to other global parameters.




Fig.4 - The L-to-H-mode threshold power scaling as fit

in the ITER database `DB2'. The threshold power is shown,

scaled as a function of the plasma parameters, ne, BT, and

major radius. A threshold power for ITER of 140 MW is extrapolated.


One student, Jeff Schachter, received his Ph.D. He is now employed by the DIII-D group at GA.


Selected Domestic Travel

Bob Granetz and Paul Bonoli were at the Sherwood meeting in Madison, Wisconsin the first week in May. Bob Granetz gave an invited review talk, ``Disruptions in Tokamaks'', at the meeting.

Martin Greenwald was in Los Angeles the week of May 7th for an ESNET meeting.

Yuichi Takase was at Princeton the week of May 19th for the NSTX Physics Advisory Committee Meeting.


International Travel

Yuichi Takase, Miklos Porkolab, Ian Hutchinson, Spencer Pitcher, Joe Snipes, and Jim Terry presented papers at the 24th EPS Conference on Cont. Fusion and Plasma Physics in Berchtesgaden, Germany.

After the EPS meeting, Joe Snipes went on to visit JET for a High Performance Workshop organized by Gormezano and Parail of JET from 16 - 19 June. At this workshop, he presented work by Paul Bonoli on calculations of possible high performance RF scenarios in C-Mod.

After the EPS meeting Yuichi Takase attended the ITER Heating and Current Drive Expert Group Workshop in Garching on June 16-17, 1997. The primary objective of this workshop was to provide input to the Physics Basis Document.

Earl Marmar was in Korea the week of June 23rd as a member of the committee for the KSTAR physics validation review. The all-superconducting-magnet tokamak is proposed to be built at the Korea Basic Science Institute, Taejon, Korea.

Francesca Bombarda spent two weeks at Frascati, from Jun 9 to 20, for a periodic exchange of information between C-Mod and FTU.




Near Term Plans

Our near term plans are focussed on analysis of the physics results gained from the machine operation this quarter and on the continuation of the run campaign following the repair of the alternator. Mini-proposals for experiments are being solicited from the community. (Approved MP's can be found at http://www.psfc.mit.edu/server-java/MiniProposals.) In addition, at the urging of the ITER JCT, we plan to initiate studies of the suitability of ECDC (discharge cleaning) for wall cleaning. The interest is in studying the effects of ECDC when the ECH resonance is NOT swept to contact all of the plasma-facing vessel walls. Although the usual C-Mod operation is to sweep the resonance by varying the field, this option is not available for ITER.

New Publications


S. Coda and M. Porkolab, 24th EPS Conf. on Contrl. Fus. and Plas. Phys., Berchtesgaden, Germany, 1997,

paper P3.053, in press

I.H. Hutchinson, et al., 24th EPS Conf. on Contrl. Fus. and Plas. Phys., Berchtesgaden, Germany, 1997, P2.018, in press

ITER H-mode Threshold Database Working Group, presented by J.A. Snipes, 24th EPS Conf. on Contrl. Fus.

and Plas. Phys., Berchtesgaden, Germany, 1997, P2.018, in press

A. Yu. Pigarov, J.L. Terry and B. Lipschultz, 24th EPS Conf. on Contrl. Fus. and Plas. Phys., Berchtesgaden,

Germany, 1997, paper P2.023, in press

C.S. Pitcher, et al., 24th EPS Conf. on Contrl. Fus. and Plas. Phys., Berchtesgaden, Germany, 1997, P2.024, in press

M. Porkolab, et al., 24th EPS Conf. on Contrl. Fus. and Plas. Phys., Berchtesgaden, Germany, 1997, P2.021, in press

J.A. Snipes, et al., 24th EPS Conf. on Contrl. Fus. and Plas. Phys., Berchtesgaden, Germany, 1997, P2.020, in press

Y. Takase, et al., 24th EPS Conf. on Contrl. Fus. and Plas. Phys., Berchtesgaden, Germany, 1997, P2.019, in press

J.L. Terry, et al., 24th EPS Conf. on Contrl. Fus. and Plas. Phys., Berchtesgaden, Germany, 1997, P2.022, in press

File converted from TEX by TTH, version 0.8.