Progress towards steady state on the National Spherical Torus Experiment NSTX

1
Progress towards steady state on the National
Spherical Torus Experiment (NSTX)
College WM Columbia U Comp-X General
Atomics INEL Johns Hopkins U LANL LLNL Lodestar MI
T Nova Photonics New York U Old Dominion
U ORNL PPPL PSI Princeton U SNL Think Tank,
Inc. UC Davis UC Irvine UCLA UCSD U Colorado U
Maryland U Rochester U Washington U Wisconsin
Presented by David A. Gates for the NSTX
Group Princeton Plasma Physics Laboratory,
Princeton, NJ at Columbia University Feb. 2,
2007 New York, NY
Culham Sci Ctr U St. Andrews York U Chubu U Fukui
U Hiroshima U Hyogo U Kyoto U Kyushu U Kyushu
Tokai U NIFS Niigata U U Tokyo JAERI Ioffe
Inst RRC Kurchatov Inst TRINITI KBSI KAIST ENEA,
Frascati CEA, Cadarache IPP, Jülich IPP,
Garching ASCR, Czech Rep U Quebec
2
Outline

• Introduction
• Progress towards steady state operation
• MHD stability
• Transport
• Non-inductive startup
• Wave physics
• Boundary physics

3
The Spherical Torus (ST) is a Low Aspect Ratio
Tokamak
ST maximizes the field line length in the good
curvature (stable) region - strong toroidicity
- Benefit to stability and confinement
Strong natural shaping - Enhances stability
and bootstrap current
4
NSTX Addresses Key Issues for Fusion Energy
Development, ITER Physics and Plasma Science

• Enable attractive CTF for success of DEMO
  addresses unique ST development issues, synthesis
  of ST and tokamak results
• Support USBPO and ITPA activities can address
  physics in ITER physics regimes
• Complement and extend toroidal confinement
  physics operating regimes unique to ST, overlap
  with that of conventional aspect ratio tokamaks

Major Radius R0 0.85 m Aspect Ratio
A 1.3 Elongation k 2.8 Triangularity
d 0.8 Plasma Current Ip 1.5 MA Toroidal Field
BT 0.55 T Pulse Length 1.5 s NB Heating (100
keV) 7 MW bT,tot up to 40
5
STs Can Lead to Attractive Fusion Systems

• Component Test Facility (CTF) will be needed
  after ITER to carry out integrated DEMO power
  testing and development
• ST enables highly compact CTF with full remote
  maintenance and high duty factor, and it provides
  potentially attractive reactor configuration
• Issue Plasma formation and current ramp-up

Peng et al, PPCF 47, B263 (2005)
6
AT/ST requires high bootstrap fraction, fbs,
simultaneous with high ?t

• High fbs and high ?t competing requirements (at
  fixed shape and ?N)
• Progress for ST and advanced tokamak given by the
  sustainable ?t ?sus ? fbs?t S?N2, where S
  q95(Ip/(aBt)) is the shaping factor
• If ?Nmax CTroyon, then only shape improves ?sus
• ?sus increases linearly with increasing S S
  q95(Ip/(aBt))
• Component Test Facility requires ?t 20 and fbs
  50, ?sus 10
• ARIES-ST requires ?t 40 and fbs 90
• NSTX has optimized shaping with new PF coils for
  high triangularity and elongation
• NSTX has achieved record values of elongation and
  shape factor
• Leads directly to record values of the ?sus for
  the ST
• For NSTX 100 non-inductive operation with ?N 7
  only with strong shaping

7
Continuous shaping enhancements have enabled
progress towards steady state

• Shaping factor has nearly doubled since 2001 -
  record ? 3
• Major upgrades include
• Power supply modifications
• Control system upgrades
• PF Coil modifications
• Real time equilibrium reconstruction for shape
  control

? 1.8, ? 0.6, S 22
? 2.0, ? 0.8, S 23
? 2.75, ? 0.8, S 37
? 2.3, ? 0.6, S 27
? 3, ? 0.8, S 41
2004
2006
2002-3
2005
2001
8
Reduction in control system latency increases
elongation (2004)

• Vertical instability determines upper bound on
  boundary elongation ? b/a
• Control system latency (time between an event and
  the mitigating control action) major factor in
  vertical controllability
• Faster control system leads to better plasmas

Plasma elongation (?) plotted versus normalized
internal inductance (li)
Each point in the plot represents one EFIT
equilibrium in the NSTX database
9
Upgrade of PF1A coil has enabled simultaneous
achievement of high ? and ?

• Triangularity (Rx-R0)/a increases edge q for
  given Ip, Bt by increasing length of the plasma
  boundary on high field side
• Location of X-point (Rx,Zx) determined on NSTX by
  PF1A coil
• Coil was modified to enable simultaneous
  achievement of high ? and ?

Old PF1A Coil
Modified PF1A Coil
10
Implementation of rtEFIT improves shape control
reproducibility

• rtEFIT used for boundary control
• Define locations where boundary should be,
  control on difference between flux at the control
  points and flux at the x-point
• Collaboration with General Atomics - very
  successful
• Has lead to marked improvements in reliability
  of shape control

Boundary overlay time window
3cm includes MHD perturbations
Control points
Gates, PoP 2006
11
Shape control modifications substantial increase
in NSTX operating space
Pulsed average ?1/2?p?t vs S (All years)
Pulse averaged ?t versus pulse length Data are
sorted by year and by S

• Strong correlation between improved shaping and
  the achieved plasma performance
• Improvements in 2006 are due to improved
  reliability at high shaping (rtEFIT)

12
High Performance Can Be Sustained For Several
Current Redistribution Times at High
Non-Inductive Current Fraction

• ?p and NBI current drive provide up to 65 of
  plasma current ?
• Relative to earlier results, High bN ?
  H89P now sustained 2 ? longer

TRANSP non-inductive current fractions
116313G12
D. Gates, PoP 13, 056122 (2006)
13
MHD-Induced Redistribution of NBI Current Drive
Contributes to NSTX Hybrid-Like Scenario as
Proposed for ITER

• qmingt1 for entire discharge, increases during
  late n1 activity

• Fast ion transport converts peaked JNBI to flat
  or hollow profile
• Redistribution of NBICD makes predictions
  consistent with MSE

n1 mode onset
n1 mode onset

• High anomalous fast ion transport needed to
  explain neutron rate discrepancy during n1

J. Menard, PRL 97, 095002 (2006)
14
Integrated Modeling Points to Importance of
Shaping, Reduced ne, and Increased Te/tE for
Higher fNI and High bN

• n20(0)0.85,
• ?2.2
• H981.1
• ?N 5.6
• q(0) 1.15
• n20(0)0.36,
• ?2.2
• H981.1
• ?N 5.6
• q(0) 1 _at_ 0.8 s
• n20(0)0.75,
• ?2.55
• H981.35
• bN 6.6
• q(0) 1.4

n(0)0.75e20
15
Fully Non-Inductive Scenario at Higher bN
Requires Higher Confinement, Higher q, Strong
Plasma Shaping

• Higher k for higher q, bP, fBS
• High d for improved kink stability

k 2.3, dX-L 0.75 dRSEP -1cm
k 2.6, dX-L 0.85 dRSEP -2mm

• Need 60 higher T, 25 lower ne
• Lithium?
• higher q0 ? qmin ? 2.4 (higher with-wall limit bN
  lt 7.2)

16
Dynamic Error Field Correction (DEFC) Extended
Pulse
Six EF/RWM Coils powered by SPA supplies with
up to 6 kA-turn currents at 1 kHz Supported by
unique RWM sensors (24 BR and 24 BZ internal
coils over 150 other magnetic sensors) with 51
ch.
Plasma rotation sustained after correction of
intrinsic error fields
J. Menard et al., IAEA 2006
17
RWM Actively Stabilized at Low, ITER-Relevant
Rotation

• First demonstration in an ST
• Plasma rotation reduced by non-resonant n3
  magnetic braking
• No-wall b-limit computed by DCON

• Optimize RWM control
• Fully understand stabilization
• physics

Sabbagh et al., PRL 97 (3006) 04500
18
Observed Rotation Follows Neoclassical Toroidal
Viscosity (NTV)Theory

• First quantitative agreement with NTV theory
• Due to plasma flow through non-axisymmetric field
• Trapped particle, 3-D field spectrum important
• Computed using experimental equilibria
• Viable physics for simulations of rotation
  dynamics in future devices (ITER, CTF)

Magnetic braking due to applied n3 field
Zhu et al., PRL 96 (2006) 225002
19
Both Internal and External Modes Can be
b-Limiting in NSTX
Resistive Wall Modes can limit bT at
low-q (Sabbagh et al., NF, 44 2004 560)
Discharge (in black) collapses as rotation
flattens and decreases
Non-linear M3D results consistent with experiment
bT ()
bT ? 31
bT ? 23
ff(0) (kHz)
Mode Bq (Gauss)
q0 (w/o MSE)

• Maintaining high rotation is a key to stabilizing
  both modes

20
NSTX Addresses Transport Turbulence Issues
Critical to Both Basic Toroidal Confinement and
Future Devices

• NSTX offers a novel view into plasma TT
  properties
• NSTX operates in a unique part of dimensionless
  parameter space R/a, bT, (r, n)
• Dominant electron heating with NBI relevant to
  a-heating in ITER

• Excellent laboratory in which to study electron
  transport electron transport anomalous, ions
  close to neoclassical
• Large range of bT spanning e-s to e-m turbulence
  regimes
• Strong rotational shear that can influence
  transport
• Localized electron-scale turbulence measurable
  (re 0.1 mm)

21
Dedicated H-mode Confinement Scaling Experiments
Have Isolated the BT and Ip Dependences
Scans carried out at constant density,
injected power (4 MW)
0.50 s
0.50 s
22
Dedicated H-mode Confinement Scaling
ExperimentsHave Revealed Some Surprises
Strong dependence of tE on BT
Weaker dependence on Ip
H98y,2 0.9 ? 1.1 ? 1.4
H98y,2 1.4 ? 1.3 ? 1.1
4 MW
4 MW
tE,98y,2 BT0.15
tE,98y,2 Ip0.93
NSTX tE exhibits strong scaling at fixed q
tEIp1.3-1.5 at fixed q
tE,98y,2Ip1.1 at fixed q
23
Variation of Electron Transport Primarily
Responsible for BT Scaling
Broadening of Te reduction in ce outside
r/a0.5 with increasing BT
Ions near neoclassical
Neoclassical
24
Ion Transport Primarily Governs Ip Scaling- Ions
Near Neoclassical Level -
GTC-Neo neoclassical includes finite banana
width effects (non-local)
ci,GTC-NEO (r/a0.5-0.8)
25
Turbulence Measurements Gyrokinetic
Calculations Have Helped Identify Possible
Sources of Transport
Ion and electron transport change going from L-
to H-modes
Electron transport reduced, but remains anomalous
Ion transport during H-phase is neoclassical
- Localized measurement (axis to edge) -
Excellent radial resolution (6 cm)
26
Strongly Reversed Magnetic Shear L-mode Plasmas
Achieve Higher ?Te and Reduced Transport
Linear GS2 calculations indicate reduced region
of mtearing instability for RS plasma
GS2 calcs also indicate ETG stabilized by RS
F. Levinton, APS 2006
27
CHI Is Being Used for Solenoid-Free Startup
Transient CHI Axisymmetric reconnection leads to
the formation of closed flux surfaces
28
160 kA of Closed Flux Current Produced in NSTX
by Transient CHI
Ip decays after ICHI?0 ?High Ip flux closure
EFIT reconstruction

• 2006 discharges operated at high RF and injector
  flux
• Magnetic sensors and flux loops used in
  reconstruction
• Plan to optimize high current CHI discharges,
  couple to OH, HHFW

R. Raman et al., PRL 97 (2006) 175002
29
EBW/HHFW Coupling Studies Being Carried Out
B-X-O mode coupling understood in L-mode at fce
coupling in H-mode low
HHFW htg efficiency improved at high BT and k
L-mode
30
NSTX Accesses ITER-Relevant Fast-Ion Phase-Space
Regime

• ITER will operate in new, small r regime for
  fast ion transport
• kr  1 means "short" wavelength Alfvén modes
• Fast ion transport expected from interaction of
  many modes
• NSTX can access multi-mode regime via high bfast
  / btotal and vfast / vAlfven

NSTX observes that multi-mode TAE bursts induce
larger fast-ion losses than single-mode bursts
1 neutron rate decrease
5 neutron rate decrease
E. Fredrickson, Phys. Plasmas 13, 056109 (2006)
31
Alfvén Cascades (RSAE) Observed at Low be on NSTX
(also on MAST)

• Frequency chirp indicates evolution of qmin
• Use for q-reconstruction, MSE verification

32
Angelfish Identified as Form of Hole-Clumps,
Consistent with Theory

• Mode satisfies Doppler-shifted resonance
  condition for TRANSP calculated fast ion
  distribution
• Growth rate estimates from theory is 0.04 from
  observation is 0.053
• Engineering of fast-ion phase space can suppress
  deleterious instabilities

• Berk et al., (PoP, 99)

33
Peak Heat Flux Can Be Reduced By Plasma Shaping

• Flux expansion decreases peak heat flux despite
  reduced major radius
• Compare single-null double-null configurations
  with triangularity ? 0.4 at X-point and high
  triangularity ?0.8 double-null plasmas
• Measure heat flux with IR thermography of carbon
  divertor tiles
• Peak heat flux decreases as 1 0.5 0.2
• ELM character changes Type I ? Mixed ? Type V

34
Peak Heat Flux Can Be Reduced With No Loss of
Confinement
35
Lithium Evaporator (LITER) Produced Particle
Pumping and Improved Energy Confinement in H-mode
Plasmas
Wtot 20 higher in post-Li H98y,21.1?1.3 post-Li
36
Edge Imaging Has Been Key to Studying Edge
Turbulence Phenomena (Blobs, ELMs)
ELM dynamics and rotation have been measured
Excellent blob measurements have allowed
connection to theory
37
Summary

• NSTX normalized performance approaching ST-CTF
  level
• Only ST in world with advanced mode stabilization
  tools and diagnostics
• Unique tools for understanding core and edge
  transport and turbulence
• Uniquely able to mimic ITER fast-ion instability
  drive with full diagnostics
• Improved understanding of HHFW and EBW
  coupling/heating efficiency
• Demonstrated 160kA closed-flux plasma formation
  in NSTX using CHI
• Developing understanding and unique tools for
  heat flux and particle control