E. D. Fredrickson, D. Darrow,
N. N. Gorelenkov, S. Medley, J. Menard, L. Roquemore
45th Annual Meeting of Division of Plasma PhysicsAmerican Physical Society
October 27 – 31, 2003Albuquerque, New Mexico
MHD-induced Fast Ion Losses on NSTX
Supported by
Columbia UComp-X
General AtomicsINEL
Johns Hopkins ULANLLLNL
LodestarMIT
Nova PhotonicsNYU
ORNLPPPL
PSISNL
UC DavisUC Irvine
UCLAUCSD
U MarylandU New Mexico
U RochesterU Washington
U WisconsinCulham Sci Ctr
Hiroshima UHIST
Kyushu Tokai UNiigata U
Tsukuba UU Tokyo
JAERIIoffe Inst
TRINITIKBSI
KAISTENEA, Frascati
CEA, CadaracheIPP, Jülich
IPP, GarchingU Quebec
Will fast ion driven instabilities be important in the NSST, CTF?
• Fast ion instabilities important for fusion reactors;– Fast ion losses raise ignition threshold, first wall damage
– Potentially beneficial (alpha-channeling)?
• ST's, with intrinsic low field, are particularly susceptible to fast ion driven instabilities.– Vfast ion / Valfvén>1; fast ion energy available to drive modes
– fi/aplasma; sets mode structure scale, fast ion transport
– Also affected by q and density profiles, thermal ion/electron .
• Need to understand instability drive, loss mechanisms.– Result important for conventional aspect ratio reactors (ITER).
10
100
1000
0.1 0.2 0.3TIME (s)
FR
EQ
UE
NC
Y (
kHz)
“fishbones”
TAE
CAE/GAE
108170
A broad spectra of kinetic instabilities are seen on NSTX
“mostly harmless”
Usually do not result in large fast ion losses.
However, in some cases…
Gorelenkov - GI2.001, Taylor - QP1.048, Fu - QP1.122
Neutron rate drops, D bursts imply fast ion loss in some shots
• Neutrons are beam-target; drops mean fast ion loss– S nfi
• D bursts probably due to fast ions hitting wall or divertor plates– Can be confused with ELMs
• In L-mode, sometimes correlated with D drops.
• Loss also seen in iFLIP, redistribution in NPA.
H-alpha (a.u.)
Plasma Current (MW)
Pnbi (MW)
0.8
0.4
0.0
1.0
0.01.0
0.06
0
-6
shot 108530
4
00.2 0.30.0 0.1 0.4
TIME (s)0.20 0.25 0.30
Neutrons (1014/s)
Mirnov Coil (a.u.)
TAE bursts responsible for most of fast ion loss here
• Neutron drops ≈ 10-15 %.• Period is ≈ 10 ms.• In steady-state, predicted
reduction in fast ion beta of 40 %.
• TAE have strong bursting character with multiple modes present.
• Fishbones have small neutron drops
black n=1 red n=2 green n=3 blue n=4 yellow n=5
FR
EQ
UE
NC
Y (
kHz)
200
150
100
50
0
108530
fTAE= VA(0)/4qR0 + n “frot”
0.20 0.300.22 0.24 0.26 0.28
1.0
0.0
Neutrons (1014/s)
TIME (s)
Large fast ion losses also with " fishbones”
• Up to 20% drops in the neutron rate, with f.b. periods of ≈ 10 ms; steady-state reduction in fast ion population of ≈50%.
• Why do some fishbones cause neutron drops, others no drop?
• Even no-drop f.b.'s affect CAE.• Do CAE trigger f.b.’s?• Correlation of (large) fishbones
with H-mode transitions?
Ip = 0.8 MA, Pb = 3.4 MW, tor = 16%)
CAE (0.5-1.2MHz)
FR
EQ
UE
NC
Y (
kHz)
TIME (s)0.15 0.20 0.25 0.30 0.35
1.41.0
0.00.4
(x10
14/s
)
0.80.6
0.0
0.40.2
D-alpha
150
100
50
0
“TAE”
“Fishbone”
108170
Neutron Rate
3
2
1
050
0
fishbone (10-90 kHz)
(mG
)(G
)
Strong frequency chirping correlated
with fast ion loss
• In this case, multiple modes also seem to co-exist.
• Lack of neutron drop may mean loss of lower energy ions; but not according to NPA.
• It would be very nice to have good internal data.
S. Medley - NPAIp = 0.8 MA, Pb = 6 MW, tor = 18%)
0.3 0.35 0.40 0.45
0
20
40
60
80109022
TIME (s)
5
04
2
032
10
5
0
n = 1 n = 2 n = 3 n = 4
FR
EQ
UE
NC
Y (
kHz)
SN
PA
(10
5 /s)
(10
13/s
)(G
)
NEUTRON RATE
ED = 50 keV
ED = 10 keV
5 - 50 kHz, 0.45 - 1.2 MHz (x100)
TAE, f.b. and CAE/GAE interact
• TAE bursts precede fishbones.
• TAE-induced fast-ion losses don't affect CAE.
• Fishbone-induced bursts cause drops in CAE amplitude, small effect on neutron rate.
H-alpha (a.u.)
Plasma Current (MW)
Pnbi (MW)
0.8
0.4
0.0
1.0
0.01.0
0.0
shot 108530
4
00.2 0.30.0 0.1 0.4
TIME (s)0.20 0.25 0.30
Neutrons (1014/s)
~rm
s(B
)
103
102
101
103
102
101
100
102
101
100
10-1CAE 500 - 1500 kHz
TAE 70 - 150 kHz
f.b. 10- 40 kHz
Ip = 0.65 MA, Pb = 3.6 MW, tor = 10%)
"Not your typical fishbones…"
• Retain strong frequency chirping and periodic bursting character.
• Toroidal mode numbers up to n=5 have been seen.
• Often with q(0) > 1 and m >1• Onset frequency well above
precession drift frequency.– Probably bounce resonance*
– May be relevant for ITER…
0.1 0.2 0.3
0
50
100
150108794
TIME (s)
FR
EQ
UE
NC
Y (
kHz)
2
1
0
Neutron Rate (x1013/s)
black n=1 red n=2 green n=3 blue n=4 yellow n=5
*Fredrickson, Chen, White, Nucl. Fusion
(Ip =1.0 MA, Pb = 1.7 MW, tor = 11%)
Summary
• "Collective" fast ion losses seen on NSTX– "fishbone" induced losses of up to 20% and new
fishbone physics– TAE induced losses of 10-15% per burst– Coupled TAE/fishbone induced losses
• Losses are significant, either because they may:– Raising ignition threshold– Increase first wall power loading
• Need physics basis to extrapolate to NSST or ITER
Bounce-resonance, as well as
precession, can drive f.b.’s
• For effective drive, need large average bounce angle…
• …and bounce frequency in reasonable range.
• For most conventional, beam heated tokamaks, this is not the case.
• However, ITER may be a different story.
Bou
nce
Ang
le
0
1
2
3
0 20 40 60 800
20
40
60
80
Energy (keV)
Fre
quen
cy (
kHz)
mode frequency range
108794_0.145
Principal diagnostics of mode structure are
Mirnov arrays, soft x-ray cameras
• Mirnov coils have limited poloidal coverage; plasma-coil separation varies significantly.
• Soft x-ray data dependent on plasma conditions; H-modes are bad. 2.01.00.0
0.0
2.0
-2.0
1.0
1.0
MAJOR RADIUS (m)
ELE
VA
TIO
N (
m)
108794_0.15s
Mirnov Coils
Mode amplitude typically largest near core, data limited.
• Data acquisition rate ≈ 200 kHz.
• Horizontally viewing camera.
• High frequency f.b.s also seen; much weaker.
• No obvious phase inversion; no island?
• Signal low near edge.Soft x-ray data from JHU cameras.
108794(a) (b)
0 200 400
0
-10
-20
-30
-40
-50
-60
0.140 0.145 0.150
TIME (s)
‘MIN
OR
RA
DIU
S’ (
cm)
Chord Integrated soft x-ray
(a.u.)
3
0
-3Gau
ss B~
(c)
Fishbones appear to be strongly ballooning?
• Interpretation a little tricky due to varying plasma-coil separation.
• Outboard poloidal wavelength reasonably long.
• Inboard wavelength irrelevant?
360
270
180
90
0
0 90 180 270 360
POLOIDAL ANGLE
RE
LAT
IVE
PH
AS
E
108794_0.145
0.8
0.6
0.4
0.2
0.0
AM
PLI
TU
DE
(G
)
TAE/fishbones synergistic?
150
100
50
0
0.20 0.22 0.24 0.26
1.5
1.0
0.5
0.0
109022F
RE
QU
EN
CY
(kH
z)S
(10
14/s
)
TIME (s)
(b) Neutron Emission
TAE (n = 1, 2, 3, 4)
5 - 50 kHz 70 - 110 kHz
10.0
1.0
0.1
0.01
Brm
s(G
)(a) Mirnov Spectrum
(c)
• Correlation of f.b. and TAE bursts suggests coupling.
• Attribution of cause of fast ion losses difficult to make.
• TAE bursts cause initial, fast drop, fishbones later, slower drop.
• Neutron drops related to loss of fastest ions; slower ions may also be lost.
“Small” losses can add up
• Fast ion population is modeled as constant source and exponential decay, S(t) = S0(1-et/).
• 20% drops every 18 ms with s ≈ 36 ms means average drop of 30%.
• Lots swept under rug, but coincidently the prediction is pretty good… (tor = 16%)
TIME (s)0.0 0.2 0.4 0.6 0.8
0
2
4
6
80
1
2
3
4
5
Pbe
am (
MW
)N
EU
TR
ON
RA
TE
(10
14/s
) 108736z01
TRANSP
Measured
30% reduction
Many modes are present simultaneously during losses
• Largest amplitude (on Mirnov) at lower n, but higher m’s should have faster fall-off from plasma edge.
0 100 20050 150 250
n=2 n=3n=4
n=5
0.4
0.2
0.0
Am
plitu
de (
a.u.
)
108530_0.2847s
Frequency (kHz)
n=6
Fast ion loss parameters
• Vfast ion / VAlfvén;
– How much, if any, of energy in fast ion distribution available to drive waves.
– Fusion ’s are all at full energy, only fraction of beam power at full energy.
• fi/aplasma;
– Measure of mode structure, effectiveness at causing fast ion losses.
• But high q-profile may ameliorate,
MHD-induced Fast Ion Losses on NSTX
E. D. Fredrickson, D. Darrow N. N. Gorelenkov,
S. Medley, J. Menard, L. Roquemore
Princeton Plasma Physics Laboratory, Princeton, New Jersey 08543
45th Annual Meeting of APS-DPPAlbuquerque, New Mexico
Oct. 27-31, 2003