High Harmonic Fast Wave Experiments on TST2

1
High Harmonic Fast Wave Experiments on TST-2

• Y. Takase, A. Ejiri, S. Kainaga, H. Kasahara1),
  R. Kumazawa1), T. Masuda,
• H. Nuga, T. Oosako, M. Sasaki, Y. Shimada, F.
  Shimpo1), J. Sugiyama,
• N. Sumitomo, H. Tojo, Y. Torii, N. Tsujii, J.
  Tsujimura, T. Yamada2)

University of Tokyo, Kashiwa, 277-8561
Japan 1)National Institute for Fusion Science,
Toki, 509-5292 Japan 2)Kyushu University, Kasuga,
816-8580 Japan
12th International Workshop on Spherical Torus
2006 Chengdu 11-13 October 2006
2
Outline

• TST-2 spherical tokamak and RF system
• HHFW experiment
• Electron heating experiment
• Wave diagnostics
• RF magnetic probes
• Reflectometry
• Wave measurements
• parametric decay
• scattering
• TORIC full-wave analysis
• EC start-up experiment
• Plans
• 200MHz experiments on TST-2
• RF sustainment of high b plasmas in UTST

3
TST-2 Spherical Tokamak

• R / a 0.38 / 0.25 m (A 1.5)
• Bt 0.3 T / Ip 0.1 MA

ECH 2.45GHz (lt 5 kW) HHFW 21MHz (lt 200 kW x 2)
4
21 MHz Matching/Transmission System
5
Variable k Two-Strap Antenna
Mo limiters

• RF power
• 400 kW
• Frequency
• f 13, 21, 30 MHz
• (w/WH 7 at BT 0.2 T, f 21 MHz)
• Toroidal wavenumber
• kf nf/R0 11, 16, 26 m-1
• (nf 4.3, 6, 10)
• varied by changing the strap spacing
• Faraday shield angle
• 30

current straps (0, p)
Faraday shield
6
Single-pass Absorption Calculation

• Single-pass absorption is greater for
  double-strap excitation
• Single-pass absorption
• increases with ne
• increases with Te
• decreases with Bt
• (increases with be )

double-strap
7
Single-Pass Absorption Improves with be
Imag ( k?)
ELD TTD
ELD TTD CROSS
ELD
ELD CROSS
Bt 0.15 T ne 1.01019 m-3 Te 100 eV nf
10
single-pass absorption 0.18
8
TORIC Full-Wave Calculations

• Analysis of HHFW heating scenarios used on TST-2
• is being carried out using the TORIC full-wave
  code.
• Bt 0.2 T, f 21 MHz, n 10, ne0 2 ? 1019
  m-3, Te0 0.2 keV

Electron absorption 100
9
Electron Heating by HHFW
R0.19m
High field side
R0.26m
360kW
R0.38m
180kW
R0
Center
no HHFW
R0.43m
PS noise

• Soft X-ray increased, but density and radiated
  power did not change
• ? electron heating
• Strongest response near plasma center

R0.54m
Low field side
10
Single-Strap vs. Double-Strap Excitation

• Increases in stored energy and visible-SX
  emission are greater for double-strap excitation
• Consistent with single-pass absorption calculation

PNET 120 kW
PNET 120 kW
11
Wave diagnostics on TST-2

• RF magnetic probes
• Sensitive to electromagnetic component
• Plasma edge only
• Reflectometry
• Sensitive to electrostatic component
• Can probe the plasma interior
• Both parametric decay instability (PDI) and
  frequency broadening due to scattering by density
  fluctuations were observed.
• These processes can alter the wavenumber
  spectrum, and affect both wave propagation and
  absorption.

12
RF Magnetic Probes at 14 Toroidal Locations
Bz
Bf
Bf
f-60
f-55
f-65
f-30
f-115
f-9
f-120
f0
1-turn loop
f9
f-125
f30
2cm
f55
toroidal direction
f60
Semi-rigid cable
f65
f155
f150
Core (insulator)
Direction of B field to be measured
f145
Slit
Top view
S.S. enclosure
13
TST-2 Reflectometer System
Gunn 25.85 or 27.44 GHz
VCO 6-10GHz
scalar horn
coaxial
waveguide
X5
X4
Ep x Bt
5-20mW
24-40GHz 100mW
D.C. -3dB
cos(fpwtfRF)
I
X10
RF
LO
Q
F.G.
Aeiwt
DC-500MHz
Aeiwtip
sin(fpwtfRF)
Digitizer or Oscilloscope lt250MHz sampling
RF 21MHz eiwt

• One of three sources is used
• Frequency sweepable VCO for profile measurements
• Fixed Gunn Osc. (25.85 or 27.44 GHz) for RF
  measurements

14
Microwave Reflectometer
Mirror
Window f200
TST-2 V.V.
Mirror
TF Coil
Horn Antennas
15
Parametric Decay FW ? IBW ICQM
RF probe
?
probe ?
PDI
Magnetic field dependence
?
Parametric Decay Instability (PDI)
Most probable decay process High Harmonic Fast
Wave (HHFW) ? Ion Bernstein Wave (IBW)
Ion-Cyclotron Quasi-Mode (ICQM)
WH at outboard edge
Threshold power 20 kW
16
Comparison of RF Probe and Reflectometer Spectra
HHFW 250 kW
Reflectometer 25.85 GHz (cosine)
1.8 MHz
Reflectometer 25.85 GHz (sine)
QM
IBW
RF Probe (dB/dt) Antenna Limiter, P12
Time ms
Frequency MHz
17
Comparison of RF Probe and Reflectometer Spectra
0
Reflectometer (cos)
-20
-40
P (dB)
-60
-80
0
Reflectometer (sin)
-20
QM
IBW
-40
P (dB)
-60
noise (Al reflector)
-80
0
RF Magnetic Probes
-20
FW ?
-40
P (dB)
IBW
-60
-80
15
25
10
20
0
5
f (MHz)
18
Dependence on Plasma Position
PDI becomes stronger as the plasma outer boundary
approaches the antenna
19
Outboard vs. Inboard Comparison
RF probes

• Inboard spectrum similar to outboard, but weaker

?
?
?
?
20
Inboard Side Spectra
port10
RF probes
Z 150mm
?

• Broadened spectrum is only weakly dependent on
  vertical position

?
Z 0mm
?
?
Z -150mm
?
?
midplane Bf
?
?
midplane Bz
21
Frequency Broadening

• Frequency broadening of the pump wave
• by the plasma is observed.
• Possible processes
• Parametric decay
• Scattering by density fluctuations

22
Density Dependence Varies with Probe Position
Preliminary
Pump wave power
Df (HHFW)
Lower sideband power
DSX / SX

• PDI is generally reduced at high density
• Only weak effect on heating

nel
nel
23
Summary (HHFW Heatng)

• A k variable antenna was installed in TST-2,
  and the RF power capability was increased to
  400kW.
• Dependence on k spectrum (same spectral shape
  but different k) will be studied.
• Single-pass absorption is expected to change from
  10 to 35 when ne0 3.0?1019 m-3 and Bt 0.3T.
• In electron heating experiments, soft X-ray
  emission increased with RF power.
• Stored energy increase was larger for
  double-strap excitation.
• More direct measurement of Te is necessary (TS in
  preparation).
• Analysis of HHFW scenarios used on TST-2 is being
  carried out using TORIC.

24
Summary (RF Measurements)

• PDI and frequency broadening due to scattering
  were observed by RF magnetic probes.
• The strength of PDI increased as the outer
  boundary of the plasma approached the antenna.
• Density dependence varies with RF probe location.
• Parametric decay became weaker at high densities
  where single-pass absorption is predicted to
  become stronger.
• The effect of parametric decay on plasma heating
  is not clear.
• Initial results of RF wave detection inside the
  plasma by microwave reflectometry were obtained.
• PDI spectrum clearly observed
• Differentiation of ES and EM components may be
  made.

25
EBW Heating on TST-2_at_K (2003)
D(dW/dt) indicates Pabs/Pin gt 50 when ?ne in
front of antenna is steep enough
lt 200 kW _at_ 8.2 GHz
Thursday S. Shiraiwa, et al., Study of EBW
Heating on TST-2
26
Typical EC Start-up Discharge
Previously achieved 1kA/1kW (2.45GHz)
4kA/100kW (8.2GHz)
(a) Bt decreases gradually. (RECH decreases
gradually.) (b) IPF is kept constant. (c) PEC
is kept constant. (d) Ip increases with time,
but disappears when the w We layer moves out of
the vacuum vessel. (up to 0.5 kA produced by 4
kW) (e) ne is almost constant near the cutoff
density.
27
Dependence on power and resonance position
Ip increases with PEC, whereas ne saturates
around the cutoff density 7 ? 1016 m-3 NL
(5-6) ? 1016 m-3.
time
inboard limiter
Ip depends on the w We resonance position
(RECH).
28
Dependence on PF Strength and Decay Index

• Ip maximizes at a certain field strength.
• Highest Ip is achieved with PF2 (medium decay
  index).

29
Effect of HHFW Injection
3 cases are compared ? ECH only ? ECH
HHFW ? ECH turned off during HHFW

• Ip responds quickly to HHFW
• ne and Prad increase during HHFW
• After ECH turn-off, Ip decays and HHFW
  reflection increases.

30
Single Particle Orbit Analysis

• Phase space for confined orbits is large for low
  A devices.
• Orbit-averaged toroidal precession is
  co-directed for all confined orbits.

Confined region in phase space (inside outermost
blue boundary)
Electron orbits starting from R 0.38m / Z 0m
1
(c) ctr trapped
(b) co trapped
(a) circulating
Velocity is normalized by V0RsWp (Wp is the
cyclotron frequency corresponding to Bp)
-1
0.1
0.8 m
A. Ejiri, et al., to be published.
31
Driven Current Based on Single Particle Analysis
Under a low Te (or high BZ) approximation (VTe ltlt
V0), j is given by
This current has the same parameter dependence
and the same order of magnitude as pressure
driven currents
The generated field is
where DR is the thickness of the EC deposition
layer
32
Comparison with Experiment

• Bz dependence agrees qualitatively with
  experiment.
• Dependence on PF curvature is different.
• Predicted current for PF3 is negligibly small.

Assuming correspondence of Ipmax for PF2
Further assuming
estimate for driven Ip is
33
Conclusions (EC Start-up)

• Ip of up to 0.5 kA was obtained by ECH start-up
  (2.45 GHz / 4 kW) .
• Ip increases with the decrease in Bt.
• Highest Ip was achieved with PF configuration
  with medium index and appropriate field strength.
• Ip and ne increase with PEC, but ne saturates
  around the cutoff density.
• Ip increased by up to 0.4 kA by addition of HHFW
  power.
• Analytic expressions for generated current were
  derived based on single particle orbits
• Generated current has a similar form to pressure
  driven currents.
• Self-field generation becomes siginificant at
  high bp.
• The proposed model and experimental results are
  not inconsistent, but further theoretical and
  experimental studies are needed.

34
Preparation in Progress for 200 MHz Experiments
TST-2
200 MHz transmitters (from JFT-2M)
Full-wave calculation by TASK/WM
Eq
Combline antenna (GA)
Bt 0.3 T, f 200 MHz, n 10, ne 2 ? 1018
m-3, Te 0.3 keV
35
A New Ultra-High b ST Based on Plasma Merging
UTST
Tokyo
TS-3 / TS-4
Formation of ultra-high b ST plasma using plasma
merging
UTST
Merging scenario
TST-2
Sustainment using innovative RF methods

• HHFW ( 20 MHz)
• LHFW ( 200 MHz)

Friday Y. Ono, et al., Initial operation of
UTST High-Beta Spherical Tokamak Merging Device