From: WOLFE@CMOD.PFC.MIT.EDU
Date: Mon, 9 Feb 1998 14:31:06 -0500
Reply-To: WOLFE@PSFC.MIT.EDU
Message-ID: <980209143106.254000b7@CMOD.PFC.MIT.EDU>
Subject: Alcator C-Mod Weekly Highlights
Newsgroups: sci.physics.plasma
Organization: MIT Plasma Fusion Center

Alcator C-Mod Weekly Highlights
Feb. 9, 1998

Plasma operations resumed on Alcator C-Mod last week. Four run-days were
scheduled and completed. 115 plasma discharges were successfully produced. On
Friday, the toroidal field was increased to 7.9T for the first time in the
present campaign.

Tuesday's run was devoted to H-mode confinement scaling (MP#155). A scan of
plasma current from 0.4 to 1.4MA was carried out, at a toroidal field of 5.6T
with the density near the boundary between Elm-free and Enhanced D-alpha (EDA)
operation. Stored energy scaled roughly linearly with Ip, though it was about
30% below our best results. The impurity particle confinement time, tauI,
also seems to scale with current. The soft xray pedestal width scaled with
1/Ip, though with some ambiguous shots. We scanned Bt to resolve edge
electron temperature profiles with the ECE grating polychromator (GPC). These
didn't show much Ip scaling, suggesting that either the GPC and the edge xray
array are looking at quantities with different widths, or that the GPC
resolution is too large to see these differences. Fluctuations were seen with
Bp probes and on the reflectometer - more analysis needs to be done. Some
scanning probe data was obtained at the lower currents. Disruptions limited
operation at the highest currents, with q95< 3.0. We got much stronger EDA's
at low currents and ELMfree discharges at high currents. This suggests some
sort of q dependence on the as yet unknown EDA mechanism.

Confinement studies continued on Wednesday with a scan of triangularity
(MP143B). We ran cases at 3 upper triangularities, [.16,.4,.62], and
3 lower triangularities, [.4,.5,.6]. Density scans were performed
at each. At the end of the day continuous, i.e. single shot, dynamic
triangularity scans were performed at fixed density. In the first, delta_l =
.6->.4 and delta_u = .35->.17; for the second only delta_u was scanned =
.5->.2 A wide variety of H-mode types were observed. There is apparently a
great deal of operating space in between pure ELMfree and pure steady EDA.
Using the appearance of a steady EDA as a measure, we find that it requires
the highest densities and divertor pressures when the triangularity is low,
and less for the moderate and higher triangularities. Xray pedestal widths
seem to be a good metric for h-mode phase. In all cases it is narrower for
ELMfree, wider for the mixed modes, and widest for EDA. The dynamic scans
showed that the width also responds to triangularity independent of
confinement mode and in the same sense. As the triangularity scanned down, the
width decreases, all in EDA - later, as the confinement mode switches to
ELMfree, the width continues to narrow.

On Thursday, we revisited the question of impurity pumping by the prototype
cryo-pump (MP#177). The previous experiment had produced ambiguous results,
and it was suspected that the vessel walls were pumping more than usual due to
the recent boronization. Ohmic plasmas with medium density and density ramps
to divertor detachment and EDA H-mode plasmas were investigated.
'Steady-state' and reproducible conditions were targeted in order to
investigate the utility of the cryopump. Argon or krypton and helium-3 was
puffed into these types of plasmas with the cryopump on or off. The core
impurity ion densities were monitored with x-ray and VUV spectrometers and the
divertor impurity neutral densities with an RGA. We got good reproducible
data with the pump cold and warm, during both H and L mode discharges.The
first part of the run was devoted to quantifying the effects the cryopump had
on both core and divertor parameters. Generally there are clear differences
in the warm and cold cases. The divertor pressures and h-alpha signals are
lower, source gas required higher, and densities measurably lower when the
cryopump is cold. The divertor parameters are dramatically different in the
warm and cold cases. With the pump cold the density and radiated power are
down a factor or two or so and the electron temperature is up a similar
amount. Unlike our last attempt running mp177, there is no doubt that the pump
can have a large effect on divertor operation.

In Ohmic plasmas with constant medium density, the core argon and krypton are
"pumping out", i.e., the time history of the signal is decaying away. With the
cryopump operational, the core argon density is slighlty lower than when the
pump is warm. Krypton line brightnesses also show a slight decrease in the
equilibrium value with the cryopump operational. When the density in these
Ohmic plasmas is ramped to divertor detachment, the krypton line brightness
does not change. The penetration factor for puffed krypton entering the core
is lower after detachment with the cryopump on than it is after detachment with
the cryopump off. This is similar to the results seen for argon in run 980121.
Measurements of the neutral particle density in the divertor indicate that the
krypton density is not much different with either the pump cold or warm.
In EDA H-modes, the "pump out" effect is still evident on the core
argon density and there is no difference between plasmas with and without the
cryopump operational. Core krypton brightness measurements show no "pump out"
behavior. There is a slight decrease in the core krypton brightness with the
cryopump operational. There is a slight decrease, <15%, in the divertor
neutral krypton with the cryopump operational. Also, there is less krypton in
the divertor with an H-Mode plasma than an L-mode plasma. The analysis of
the helium-3 data is ongoing.

Friday was our first run of this campaign to call for toroidal field in the 8
tesla range. Voltage regulation on the alternator has been improved, resulting
in higher voltage from the TF supplies, and a consequent reduction in the time
required to reach high field, and in the magnet heating and between-shot
cool-down times. Power systems behaved well during the high-field
operation. The physics run on Friday was in support of MP#150A, on D(He3)
heating. In the process of raising the toroidal field up to ~8T shot-to-shot,
H-modes were obtained for D(H) minority heating with the H cyclotron resonance
as far as 16.7 cm off-axis. The plasma current was also increased such that
the scan was done at constant q. The H-factor (calculated assuming 90%
absorbed RF power injected) progressively decreased as the resonance was
moved. This result differs from JET's experience and reconfirms earlier C-Mod
results. Furthermore, the inverse sawteeth phenomena observed on the inner
wall b-dot probes vanished approximately when the H cyclotron resonance passed
beyond the sawtooth inversion radius. The speculation is that the sawteeth
redistibute the minority ions, which decreases the absorption. Another result
is the bursting neutron phenomena often observed after a H->L transition. The
H and 2nd harmonic D resonance is well outside the inversion radius, so 2nd
harmonic D absorption cannot be the cause of the increased neutron rate.

For fields above 7.1 T, where the H cyclotron resonance is approximately at
the antenna position, He3 was added to systematically investigate the minority
and mode conversion regimes. All of these high-field plasmas were in L-mode.
The physics results are encouraging. A scan of He3 concentration was performed
at Bt~7.9T. The RF was square-wave modulated at frequencies from 55Hz up to
200 Hz to permit spatial localization of the electron (mode-conversion) heating
as the He3 concentration was varied. At low concentrations of He3, the bulk
ion temperature increases by ~1.2 keV while the electron temperature increases
by ~0.7 keV. For higher He3 concentrations, significant mode conversion was
observed. The electron heating was observed to move further off-axis as the He3
fraction increased, in agreement with theoretical expectations. In a single
shot, as the He3 concentration increases, the RF goes from predominantly
minority to mode conversion (ion heater to electron heater).

Over the weekend, the tuning of FMIT#3 was investigated. A faulty variac was
found and is to be repaired and remounted. Retuning of FMIT#3 to 78 MHz will
continue this week, time permitting.

Ken Takase and Paul Bonoli have successfully ported the TORIC code (written by
M. Brambilla) to a local workstation. This will allow analysis of C-Mod
discharges on a more routine basis.

Progress continued on the power supplies for the DNB. Refurbishment of the
control system for the Accel supply continued. The protection crowbar for
the Mod/Reg tube was installed. The cables and transformer oil for the
Mod/Reg filament supply were installed in prepartion for oil conditioning
using the techniques developed for the Accel supply.

Miklos Porkolab attended Advanced Tokamak Workshop at GA last week, and
presented the talk: "Alcator C-Mod Advanced Tokamak Research". He also
attended the DIII-D Advisory Committee meeting Tuesday morning, to hear the
DIII-D Program Overviews.

This week will be a plasma operations week at Alcator C-Mod. A fresh
boronization is planned prior to the scheduled physics runs.