From:
WOLFE@PSFC.MIT.EDU
Reply-To: WOLFE@PSFC.MIT.EDU
Subject: Alcator
C-Mod Weekly Highlights
Newsgroups: sci.physics.plasma
Organization:
MIT
Alcator
C-Mod Weekly Highlights
June
26, 2000
Operations
----------
Plasma operations continued
at Alcator C-Mod last week, with four run days
scheduled and completed. A
total of 75 plasmas were produced, with a
reliability of over 80%.
Experiments included a continuation of the study of
the physics of density
limits, tests of the J-port ICRF antenna with a
modified feed
configuration, a study of isotopic changeover rates, and a study
of ICRF
mode conversion physics in H-He3-D plasmas.
Following completion of
last week's runs, a fresh boronization was carried out
over the weekend.
Plasma operations are scheduled to continue this week.
Physics and
Analysis
----------------------
We continued our density limit
experiments which emphasize the role of
transport in determing the
observed empirical limit. Successful fast scanning
probe data was
collected in divertor plasmas covering a density range 0.4 -
1.0 x
limit. We now have the data to
characterize the fluctuations and
convective transport across this
range. As reported previously,
the
fluctuation levels and autocorrelation times for fluctuations at the
separatrix
increase near the density limit. We
also got good data from the
PCI.
It showed the growth of incoherent low frequency fluctuations (<100
kHz)
as the limit was approached. There was also a clear shift toward
lower
frequencies.
In order to design a particle control system
for C-Mod, we need to understand
the role of the wall in pumping and
fueling. The wall dominates particle
inventories, containing up to 1000
times the number of hydrogenic atoms as the
plasma. Isotope changeover experiments were carried
out to study these
effects. We began with a machine in which virtually all
the hydrogen isotope
had been removed. Before the changeover the ratio
R=H/(H+D) = 2.5% This rose
to 30% on the first shot with H2 puffing. After 5 shots the ratio rose to 65%.
The
total H2 puffed on each shot was approximately 50 Torr liters. The wall
inventory is estimated to be
about 300 times larger, so a puff with 0.3% of
wall inventory changes over
30% of plasma ions. The characteristic turnover
time (R/dR/dt) was 3
seconds for the first shot then slowed down gradually for
later shots,
reaching about 30 seconds after 7 shots.
Thus the changeover is
not at all exponential.
The
advantages of EDA type H-modes have been made clear previously. Our
studies emphasize the physics which
underlies this mode, with the goal of
extrapolating it to reactor like
conditions. Resistive ballooning has been
predicted to be a dominant mode
in the plasma edge. In the
resistive
ballooning dispersion relation, m_i^{1/2}/q is a stabilizing
term. This might
explain the q
dependence we see in the EDA/ELMfree boundary.
High q would
make the term smaller, destabilize resistive
ballooning and lead to EDA. If
this
hypothesis is correct, hydrogen discharges, with m_i = 1 should stay in
EDA
down to lower q. We carried out a
series of experiments looking for the
presence of EDA in ohmically heated
hydrogen H-modes. All H-mode
discharges
produced showed clear signs of EDA based on Halpha, radiation
and the presence
of the quasi-coherent mode on PCI, reflectometry and the
A side probe. During
this run, the
Hydrogen fraction never rose much above 0.6 - thus meff dropped
to
1.4. If the threshold depended on
sqrt(meff) this would have dropped from
1.41 to 1.17, roughly a change of
20%. For a nominal q threshold in the
range
3-4 for D, we would expect a drop to 2.5-3.3 for the mix in this
experiment.
On the surface, this would tend to confirm a dependence on m_i
in the proposed
direction.
However, we need to check that the profiles were otherwise
similar.
Experiments
at JET to try to access the "EDA" H-mode regime, as seen on
Alcator
C-Mod, were carried out last week. ICRF heating with over 9MW of power
was
employed. Upper and lower triangularities of about 0.4 were obtained,
with
q95 scanned from 4.2 to 5.9; operation was at relatively high density
(5e19 at
1.7MA, and 3.5e19 at 1.2MA). No EDA transitions were
obtained. With strong
gas puffing
during the H-mode in feed-forward, the background D alpha emission
was
raised somewhat, but the plasma still went into a clear Type I ELMy
regime,
although the ELMs were smaller than those seen in the usual NB case.
ICRF
mode conversion physics was investigated through detailed measurements of
the
electron power deposition profile, and observation of the mode-converted
ion
Bernstein wave (IBW) with the Phase Contrast Imaging (PCI) diagnostic.
Limiter
discharges with H-He3 plasmas were employed for this
experiment.
Mode-converted ion Bernstein waves (IBW) were observed in the
C-Mod core
plasma for the first time with the PCI sytem. The observed wave
numbers,
in the range of 4 to 6 cm^-1, are in agreement with the dispersion
relation
of an ion Bernstein wave located closely beyond the H-He3 ion-ion
hybrid
layer, where mode conversion is expected to take place. Signal at 5 to
7
times background noise level was seen for both D port and E port antennas
(the
PCI laser passes through E port top and bottom), although it was stronger
for
D port. Preliminary investigation of the GPC electron temperature
diagnostic
indicates off-axis electron heating due to the electron Landau
damping of
the ion Bernstein wave.
ICRF System
------------
External
connections to the PPPL ICRF antenna at J-Port were reconfigured over
the
weekend of 6/17-18. The center two
straps were connected as a dipole to
transmitter #3, while the outer two
straps were shorted to ground to remove
them from the circuit. This configuration is expected to reduce
possible
leakage coupling to plasma surface waves and operation in this
mode should
provide additional information on the reduced heating
efficiency of this
antenna.
Careful measurements of antenna loading as a function of power and
plasma-antenna
separation as well as heating efficiency comparisons were made
for all
three antennas. Preliminary indications are that, as a dipole, J-port
heats
approximately as well as D and E according to central Te and
neutrons.
This suggests that properly grounding the antenna box and shielding
the
feed straps should improve the 4-strap configuration. Further analysis of
these
results is in progress.
DNB System
-----------
During
the previous week, the beam was used to inject a hydrogen beam into a
deuterium
plasma. Data was acquired for the
resulting H/D ratio and the
impact of the beam on the discharge prefill. The BES system was used to
acquire data
for an estimate of the beam density to compare with the data from
the DNB
internal diagnostics. Analysis is now
being completed.
Travel and Visitors
-------------------
Earl
Marmar, Rejean Boivin, Jerry Hughes, Yijun Lin, Chris Boswell, Norton
Bretz,
Ned Eisner, Bill Rowan, and Ricky Maqueda attended the High Temperature
Plasma
Diagnostics meeting in Tucson last week, and presented papers on C-Mod
work.
Randy
Wilson came to MIT 6/21-22 to participate in the ICRF measurements
and to
help review plans for further antenna modifications during the
July
vacuum opening.
Ian Hutchinson and Joe Snipes are at JET, where they
participated in
the experiments to obtain "EDA" H-modes on that
machine.
Miklos Porkolab attended the 27th EPS Conference on Plasma
Physics and
Controlled Fusion in Budapest on June 12-16. He also
represented APS
on the EPS Board Meeting and on the Program Committee
Meeting. He then attended
the 3rd Europhysics Workshop on the "Role
of Electric Fields in Plasma
Confinement and Exhaust" on June 18. At
the EPS meeting he presented the Poster
"Initial Results from an
Upgraded ICRF System on Alcator C-Mod",
On his return from
attendance at the PSI Meeting in Rosenheim and the
International Workshop
on Electrical Probes in Magnetized Plasmas in
Berlin, David Winslow (UT-FRC) arrived at C-Mod last
week to complete
installation and begin experiments with his turbulence
probe. After
working through some
torrvac problems last week and addressing some
continuing problems with
the probe position sensor and the stepping motor
controller, David plans
to continue taking data on C-Mod this week.
Alan Lynn, a UT-FRC
graduate student, joined the on-site contingent at
C-Mod for an extended
visit. He will divide his time between
DNB and FRC
ECE.
Gary Hallock (UT-FRC) and his graduate student
Ashley Shugart visited to
assist with operation of the C-Mod PCI
system.