oxb-converted e bws in tokamaks, spherical tokamaks, stellarators and rfps

NSTX Phys. Meeting, June 1, 2009 “OXB-converted EBWs” (F. Volpe) 1

OXB-converted OXB-converted EEBWs in BWs in Tokamaks, Spherical Tokamaks, Tokamaks, Spherical Tokamaks,

Stellarators and RFPsStellarators and RFPs

Francesco Volpe

Engineering Physics Department

University of Wisconsin, Madison

NSTX Physics Meeting at PPPL, June 1, 2009


OutlineOutline

Motivation

Principle of EBWs and Mode Conversions

EBWs in

Summary & Conclusions

This Talk Madison World

Stellarator W7-AS HSX CHS, Heliotron, LHD, WEGA

Sph.Tokamak MAST Pegasus Mini-RT, NSTX

Tokamak TCV DIII-D?

RFP RFX-mod MST


Motivation for Electron Bernstein WavesMotivation for Electron Bernstein Waves

Several MCF plasmas are too dense or have too low magnetic fieldfor conventional Electron Cyclotron Resonant Heating (ECRH)

EBWs do not suffer from upper density limits


Electron Bernstein wavesElectron Bernstein waves


Marine waves are another example of Marine waves are another example of gyrophase organizationgyrophase organization


OXB-SchemeOXB-Scheme: : special view makes modes degenerate special view makes modes degenerate and not evanescentand not evanescent

p

0

1

z

B

L

Slow X

OL

O-wave cut-off

x

In the real space In the phase space

x

sin2=N2z,opt=Y/(Y+1), Y=c/


OXB-SchemeOXB-Scheme: : O-X conversionO-X conversion

Merging of O and SX waves

when their cutoffs overlap:

SXOAiry

even

odd


SX-B mode conversionSX-B mode conversion

Finite Larmor radius

6th order correction to Booker quartic

Deflection from UHR and

Propagation at X>1 (overdense), until

Cyclotron damping

Conversion efficiency can be limited by:

- Collisional damping (if edge cold)

- Back-tunneling SX-FX (if barrier thin)

- Parametric decay (at high power)


EBW Emission, Heating & CDin the W7-AS Stellarator

(with H.P. Laqua)


AntennaAntenna

Max Transmission

Min Doppler effects

Pure O-mode detection


Oblique view makes polarization elliptical. Oblique view makes polarization elliptical. Broadband phase-shifter corrects it.Broadband phase-shifter corrects it.

polarization rotator/4 phase shifter


Ray tracingRay tracing

/

r/N

/

N/r

D

D

d

d

D

D

d

d

5th order Runge-Kutta

OX: 1) optimization of injection angle,

2) symmetries in narrow evanescent layer

XB: Ridder‘s method to correct N

Full stellarator geometryPoloidal component

Off-midplane launch

Toroidal curvature

Broken axisymmetry

Parallel computing on 16 workstation cluster

D=Det(ij+NiNj-N2ij), =hot diel.tensor


EBEEBE Spectrum and ProfileSpectrum and Profile

ne from TS and Li-Beam

B from equilibrium code TRANS} mapping f r


Heat Waves: EB-diagnostic during EB-heatingHeat Waves: EB-diagnostic during EB-heating

EBH

Overlapped thermal spectrum cannot be measured at the same harmonic

Non-thermal peaks fpumpnfLH (n=1,2,...)

Parametric decay

Problem:

EBH 2nd harm. & EBE 1st harm.

Plasma has to be O2-overdense

(at least ne >2.4·1020 m-3,

recently ne 4 ·1020 m-3)

Solution:


Edge-localized ModesEdge-localized Modes

Cross-correlation with H:


Due to oblique injection, OXB heating also drives current !Due to oblique injection, OXB heating also drives current !


Angle is fixed by OXB, but co/ctr-CD can be selected and Angle is fixed by OXB, but co/ctr-CD can be selected and NN|||| adjusted by varying the magnetic configuration adjusted by varying the magnetic configuration


EBW Emission & Heatingin the MAST Spherical Tokamak

(with V. Shevchenko)


Emission in a spherical tokamak is “banded” as a Emission in a spherical tokamak is “banded” as a consequence of low aspect ratioconsequence of low aspect ratio


EBE evidenced fast electrons at sawtooth crashesEBE evidenced fast electrons at sawtooth crashes

D

Soft-X Rays

EBE 60.5GHz

62.5GHZ

63.5GZ

66.5GHz


2D angular scans of EBW emission indicated optimal 2D angular scans of EBW emission indicated optimal direction for heating at MAST spherical tokamakdirection for heating at MAST spherical tokamak

-7 -6 -5 -40.0

0.2

0.4

0.6

0.8

1.0

f=60.5GHz, = -7deg

BX

O-E

BW

Sig

nal (

a.u.

)

Poloidal angle (deg)

-11 -10 -9 -8 -7 -6 -5

0.0

0.2

0.4

0.6

0.8

1.0f=60.5GHz, = -5.6deg

B

XO

-EB

W S

igna

l (a.

u.)

Toroidal angle (deg)

2.5o FWHM 1.2o FWHM

In spite of non-optimal frequency (60GHz) and alignment issues

Time, sec

60 kJ (~3 MW)

7 kJ (~ 0.25 MW)

250 kW

EC

RH

Po

we

rT

ota

l E

ne

rgy

Emission

Heating


Btotal

ne

PlasmaO-mode

O-mode

2D angular scans confirmed that emission is anisotropic 2D angular scans confirmed that emission is anisotropic New q-profile diagnostic conceptNew q-profile diagnostic concept

Inclination of conversion contours at various f givesInclination of field line at various radial positions


3.3585 tonnes

3.3585 tonnes

Self-balanced design spins at 12,000rpm

First measurements carried out. Analysis under way.

Special design allows tilted mirror to spin at 12,000rpm and Special design allows tilted mirror to spin at 12,000rpm and scan EBW emission within the same dischargescan EBW emission within the same discharge

EBE from

plas

ma

to ra

diom

eter


EBW Heatingin the TCV Tokamak

(with A. Mϋck, L. Curchod, A. Pochelon et al.)


Angular Scan of EBW injection shows min stray Angular Scan of EBW injection shows min stray (max conversion) close to calculated optimum (max conversion) close to calculated optimum


EBWH at EBWH at ~0.4resulted in clear, ~0.4resulted in clear, bigger-than-sawtooth Tbigger-than-sawtooth Te e increaseincrease

Te builds over e~50ms


Modeling of EBW Current Drivein the RFX-mod RFP

(with R. Bilato and A. Köhn)


Peculiarities of EBWs in RFPsPeculiarities of EBWs in RFPs

1. |B| almost poloidally symmetric.

Any launch is Low Field Side (LFS).

No High Field Side (HFS) schemes, e.g. no SX-B.

2. low B low ce low long vacuum wavelength λ cutoffs and resonances “close” to each other tunneling

3. ne fluctuations cutoff corrugated and moves.

B fluctuations optimal direction, Nz,opt

2 = ce/(ce+), fluctuates


28GHz (228GHz (2ndnd harmonic) considered for EBWCD at field harmonic) considered for EBWCD at field reversal. Expected 58% conversion efficiency reversal. Expected 58% conversion efficiency

Limiting factors to higher efficiency:curvature mismatch (included)fluctuations (not included)


Multiple reflections by graphite wall re-inject some power Multiple reflections by graphite wall re-inject some power and increase overall OXB efficiency to 67%and increase overall OXB efficiency to 67%

R or FX cutoffUH resonance

P or O cutoffL or SX cutoff


Summary and ConclusionsSummary and Conclusions

• Stellarators need EBWs

- overdense

+ Mode-converted EBWs benefit from Stellarator environment

- steep gradients, external control of configuration, tolerable ne and B fluctuations

= A marriage!

• EBW heating, emission and CD all proved in W7-AS stellarator

• In spite of 60GHz being too high, EBW heating succeded also in MAST,

- after scenario optimization (steep gradients)

- and guidance from EBW emission (optimal direction).

• “Spinning mirror” provided angular scans of OXB conversion efficiency every 10ms possible q-profile diagnostic.

• EBW heating also proved in TCV tokamak

• EBW CD at edge of RFX-mod RFP plasma might trigger tranport barrier


Long-term (power-plant and CTF) applications of EBWsLong-term (power-plant and CTF) applications of EBWs

Stellarators: compensation for undesired bootstrap current at high

Sph.Tokamaks: Non-inductive start-up and growth of a target for NBI.

Current profile tailoring.

Tokamaks: Te diagnosis of ITER divertor region? Not overdense, but Upper Hybrid resonant layer (thus, EBWs) exists.

RFPs: Improved confinement by suppressing MHD and triggering transport barriers at reversal layer.

oxb-converted e bws in tokamaks, spherical tokamaks, stellarators and rfps

Documents