Measurement of lower-hybrid waves with

microwave scattering on TST-2

N. Tsujii, Y. Takase, A. Ejiri, H. Furui, H. Homma, K. Nakamura,W. Takahashi, T. Takeuchi, H. Togashi, K. Toida, T. Shinya,

M. Sonehara, S. Yajima, H. Yamazaki and Y. Yoshida

The University of Tokyo, School of Frontier Sciences

18th International ST Workshop/Princeton, November 3, 2015

1/22

Outline

1 Introduction

2 Microwave scattering

3 Full-wave analysis of LH waves

4 Summary

5 Backup

2/22

A microwave scattering diagnostic is being developed tomeasure the Lower-Hybrid (LH) waves directly and testnumerical simulations

BackgroundI Non-inductive plasma start-up with LH waves has been studied

on TST-2I The current drive efficiency in the actual experiment is much

lower than what is predicted by GENRAY-CQL3D

Microwaves in the range of 12-40 GHz have wavenumber similarto LH waves in TST-2 and suited for diagnosing these waves

The wave measurements will be compared with numericalsimulations to quantitatively understand the current drive physics

Previous direct LH wave measurementsI Microwave back-scattering with a reflectometer [Baek 2014]I CO2 laser scattering [Takase 1985]

3/22

The TST-2 Machine

R = 36 cm

a = 23 cm

RF non-inductive start-up

Bt < 0.15 T(pulse length ∼80 ms)

ne < 1018 m−3

Ip < 25 kA

RF 400 kW @ 200 MHz

Ohmic (inductive) start-up

Bt < 0.3 T(pulse length ∼30 ms)

ne < 1019 m−3

Ip < 110 kA

4/22

Non-inductive plasma start-up using LH waves is studiedon TST-2

t

Ip

ne

3©2©1©

LH NB 1 Plasma initiation (ECH, LH)

2 Ip ramp-up with LH waves

3 Further ramp-up with neutralbeam injection

The LH wave has one of the highest current drive efficiencyamong rf wavesSimilar scenario tested on JT-60U [Shiraiwa 2004]The plasma density must be kept low during Ip ramp-up← LH high-density limit

5/22

LH waves @200 MHz can access 1018 m−3 for n‖ = 5-9

1014 1015 1016 1017 1018 1019 1020

ne [m-3]

0.00

0.05

0.10

0.15

0.20

0.25

0.30B

[T

]

ω = 2ωLH

accessible

5

MC

9 OH

RFP = 0

Density limit low at low field

ω > 2ωLH to avoid parametric decay instability6/22

The experimentally achieved plasma current (< 25 kA) isan order of magnitude smaller than simulated

0 5 10 15 20ne [1017m-3]

0

100

200

300

400

Ip [

kA

]

0.05 0.10 0.15 0.20Bt [T]

0

5

10

15

20

ne [

10

17m

-3]

@h

alf I

p m

ax

MC (n||=6)

ne [1017m−3]

0.077 T0.1100.1650.220

(exp)

Simulation by GENRAY (ray-tracing) and CQL3D(Fokker-Planck solver)

The experimental density limit qualitatively consistent withsimulation

7/22

Outline

1 Introduction

2 Microwave scattering

3 Full-wave analysis of LH waves

4 Summary

5 Backup

8/22

A launched microwave beam can be Bragg-scattered bydensity fluctuations of the LH waves

digitizer

power divider

voltagecontrolledoscillator

plasma

focusing mirrors

mixer

LH

swept homodyne system

9/22

Optimum scattering geometry was investigated from thesimulated typical wave trajectory on TST-2

-0.4

-0.2

0.0

0.2

0.4

Z [

m]

rays at toroidal angle=75°

-500 0 500 1000kR [m-1]

-1000-800

-600

-400

-200

0

200400

kZ

[m

-1]

10

100

fre

q [

GH

z]

parameters at phase match

-0.4

-0.2

0.0

0.2

0.4

xd [

m]

-40 -20 0 20 40incident beam angle θ [deg]

-0.15-0.10

-0.05

0.00

0.05

0.10

0.15Y

s [

m]

xdxd

1st pass2nd

3rd

10/22

LH waves propagating in a wide range of poloidalcross-section can be detected with a microwave launchedfrom the midplane

Scan the incident beam angle θ

Scan the beam frequency

Detect the scattered light at multiple poloidal locations (xd)

The source frequency swept at 1 kHz→ the whole wavenumber spectrum is measured every 1 ms

Incident beam angle needs to be scanned on a shot-to-shot basis

11/22

The O-mode microwave scattering system

Microwave bandsI Ku band (WR-62):

12.4-18.0 GHzI K band (WR-42):

18.0-26.5 GHzI Ka band (WR-28):

26.5-40.0 GHz

Incident beam from themidplane horizontal port

4-antenna arraybelow the plasma

Small toroidal tilt tocompensate wave kφ

12/22

Outline

1 Introduction

2 Microwave scattering

3 Full-wave analysis of LH waves

4 Summary

5 Backup

13/22

Full-wave simulation of LH waves was performed withAORSA

E‖ ne

ne = 5× 1017 m−3, Te = 500 eV, Maxwellian, Prf = 1 MW14/22

The expected signal level is within the detectable range

0.2 0.3 0.4 0.5 0.6R [m]

-0.4

-0.2

0.0

0.2

0.4

Z [

m]

-2 -1 0 1 2kV/m

0.2 0.3 0.4 0.5 0.6R [m]

-4 -2 0 2 41016 m-3

0.2 0.3 0.4 0.5 0.6R [m]

-4 -2 0 2 41016 m-3

nphi=-15nphi=-15 φ=-105°φ=-105°nphi=-15nphi=-15

neE‖ ne

φ ∼ reλneL

∼ 2.8× 10−15m · 0.01m · 2× 1016m−3 · 0.05m = 0.0315/22

Outline

1 Introduction

2 Microwave scattering

3 Full-wave analysis of LH waves

4 Summary

5 Backup

16/22

The microwave scattering system was designed fordirectly measuring LH waves on TST-2

The current drive efficiency on TST-2 is lower than what ispredicted by GENRAY-CQL3D

A microwave scattering diagnostics at 12-40 GHz is designedand being fabricated

The LH fluctuation level estimated with AORSA is within thedetectable range

Future workI Detect LH wavesI Investigate changes in the location of LH waves and their

wavenumber spectrum for various parametersI Test GENRAY-CQL3DI Test full-wave codes (AORSA, TORLH)

17/22

References

S. G. Baek, et al., Phys. Plasmas 21, 012506 (2014)

Y. Takase, et al., Phys. Fluids 28, 983 (1985)

S. Shiraiwa, et al., Phys. Rev. Lett. 92, 035001 (2004)

J.C. Wright, et al., Phys. Plasmas 16, 072502 (2009)

R.W. Harvey and M.G. McCoy, Proceedings of the IAEATechnical Committee Meeting on Simulation and Modeling ofThermonuclear Plasmas, Montreal, Canada, 1992

18/22

Outline

1 Introduction

2 Microwave scattering

3 Full-wave analysis of LH waves

4 Summary

5 Backup

19/22

Strong correlation of plasma current and density leads toupper limit in the driven current at a given magnetic field

0.04 0.06 0.08 0.10 0.12 0.14 0.16Bt [T]

0

5

10

15

20

25

30

Ip [

kA

]

0 2 4 6 8 10ne [1017m-3]

0

20

40

60

80

100

PLH [

kW

]

(n||=5)(n||=5)MC (n||=6)

Density limit substantially lower than expected at higher field?

20/22

LH accessibility condition

LH ↔ FW mode conversion condition:ωpi

ω= N‖Y ±

√1 + N2

‖ (Y 2 − 1),

Y 2 =ω2

ωceωci.

In terms of N‖,

N‖ =ωpi

ωY ±

√1 +

ω2pi

ω2(Y 2 − 1).

Launching the wave from the low-density side, the parameter isaccessible if

N‖ > Na =ωpi

ωY +

√1 +

ω2pi

ω2(Y 2 − 1).

21/22

Full-wave analysis will be performed

TORLH [Wright 2009]

Finite Larmor radius code(k⊥ρL < 1)

θ: spectral ansatz

ψ: finite element

Coupled with CQL3D[Harvey 1992]

→ Non-thermal electrondistribution

Weak absorption,multi-pass regime

→ Full-wave effect may beimportant

-0.3 -0.2 -0.1 0.0 0.1 0.2R-R0 [m]

-0.2

0.0

0.2

Z [m

]

0.01 0.10 1.00|E||| (a.u.)

22/22