UCLAUCLA UCLA

Investigation of Plasma Instabilities by Fast ProfileReflectometer Measurements in DIII-D

Lei Zeng

G. Wang, E.J. Doyle, T.L. Rhodes, W.A. Peebles,

G.R. McKee1, R.Fonck1, K. H. Burrell2 , J. Boedo3, R. Moyer3

Dept. of Electrical Engineering and PSTI, University of California,Los Angeles, CA 90095

1Univ. of Wisconsin-Madison2General Atomics, San Diego, California

3Univ. of California at San Diego

UCLAUCLA UCLA

Outline

• Motivation

• Investigation of Edge Localized Mode (ELM) by fastprofile reflectometer measurements

• Investigation of Edge Harmonic Oscillation (EHO ) by fastprofile reflectometer measurements

• Summary

UCLAUCLA UCLA

• The improved fast profile reflectometer (Q-band, 33-51 GHz)has:– high temporal resolution (10 µs)

– high spatial resolution ( ≥ 2 mm)

– The measured density range from 0 to 3.2x1019 m-3 for high time resolutionin dual-polarization (X- and O- modes) received simultaneously

• The profile measurement can provide profile modificationassociated with plasma instabilities– Edge localized mode ( ELM ) in H-mode plasma

– Edge harmonic oscillation (EHO ) in quiescent double barriers (QDB) plasma

Motivation

UCLAUCLA UCLA

Some Key Improvements of Q-band Systemfor Instability Investigation

• The detail improvements for our system presented by Dr. GuidingWang

• Hardware– Optimization of polarization match, reducing the spurious reflections and

increasing the SNR

– Currently, dual-polarization simultaneous operation, increasing densitycoverage up to 3.2x1019 m-3 in high performance measurement

• Software– Implementation of robust method for automated identification of the zero

density start position, reducing uncertainties associated with automaticanalysis

– Implementation of robust and automatic profile analysis program to copewith large volume of reflectometer data, up to 60 MB per shot

UCLAUCLA UCLA

Fast Profile Reflectometer Can Track SOL DensityProfile Evolution for Type I ELMs in Long Time Record

• SOL density rise up duringevery ELM

• Robust Analysis of 6000density profilesdemonstrates the goodcapability of our fast profilereflectometer to study theedge density modification byELMs in long time record.

• Reflectometer : X- mode

• Plasma condition:– Ip = 1.4 MA

– <ne> =4.5 x1013 cm-3

– Lower single nodeconfiguration

2500 2600 2700 2800 2900Time (ms)

02

4

6

810

shot 100399Dα (a.u.)

2400 3000

2.26

2.28

2.30

2.32

2.34

2.36

2.38

2.40

R (m

)

9x1018 0.0ne (m-3)

separatrix

UCLAUCLA UCLA

Reflectometer Profiles Are in Good Agreementwith Thomson Scattering Data During ELMs

0.90 0.95 1.00 1.05 1.10 1.150

1x1019

2x1019

3x1019

4x1019n e

(m-3

)

ρ

pre-ELMpost-ELM

TS Data

Reflectometer (in X mode)

Reflectometer Profiles Compared to Thomson Data pre- and post- ELM

110399

UCLAUCLA UCLA

SOL Density Profile Modification by TypeI ELM-Density Expands to Wall

• Profile timeresolution is 25 µs

• Profile isexpanded to thewall position atthe onset of TypeI ELM.

• Calculation of Vr

of density layerwill show next

7.2 0.0ne (1018 m-3)

2.25

2.30

2.35

R (m

)

2779 ms2779.5 ms2780.2 ms2782 ms2783 ms

Major Radius (m)

shot 111279

0.005

0.01

0.015

0.02

0.0250.03

2776 2778 2780 2782 2784Time (ms)

Dα (a.u.)

n e (

m

-3)

1018

2.25 2.30 2.35 2.400

2

4

6

8

Separatrix

Tile Position

UCLAUCLA UCLA

Measured Density Profile Expansion Velocity isClose to EθθθθxB Drift Velocity During ELM

• At onset of ELM, Langmuirprobe measured Eθ , and Vr=EθxB/B2 similar to theradial velocity measured byfast reflectometer.

• BES result : Vr ~ 500 - 1000m/s for other shot, close toour measurement

• The measurement isconsistent with the ELMfilaments expand radially atEθxB drift velocity

2.262.282.302.322.342.362.38

R (m

)

7.2 0.0ne (1018 m-3)

111279

Radi

al Vo

locit

y (m

/s)Computed from reflectometer data @5x1018 m-3Computed from reflectometer data @2x1018 m-3

Time (ms)

-200

0

200

400

600

0.005

0.01

0.015

0.02

0.025

0.03

2779 2779.5 2780 2780.5

Dα (a.u.)

Vr measured by Langmuir probe

UCLAUCLA UCLA

SOL Density Evolution Can Be Tacked During ShortDuration and Varied Repetition Type III ELMs

• Fast profilereflectometermeasurement cantrack the densitymodifications byType III ELMs withshort duration (1 ms)and varied repetition.

• Profile timeresolution is 25 µs

Time (ms)

0.0

0.1

0.2

0.3

0.4

0.5Dα (a.u.)

111149

1300 1320 1340 1360 1380

2.26

2.28

2.30

2.32

2.34

2.36

2.38

R (m

)

8x1018 0.0ne (m-3)

separatrix

UCLAUCLA UCLA

SOL Density Profile Modification by Type IIIELM

• Profile expansiondistance duringType III ELM issmaller than duringType I ELM

• Profile temporalresolution is 25 µs

2.28 2.30 2.32 2.34 2.36 2.38 2.400

2x1018

4x1018

6x1018

8x10181307.4 ms1307.5 ms

1307.9 ms1308.13 ms

1307.6 ms

Major Radius (m)

n e (m-3

)

0.10

0.15

0.20

0.25Dα (a.u.)

111149

1307.2 1307.4 1307.6 1307.8 1308.0 1308.21307Time (ms)

2.30

2.32

2.34

2.36

2.38

R (m

)

8x1018 0.0ne (m-3)

separatrix

Tile Position

UCLAUCLA UCLA

Dual-Polarization Profile Measurement Can TrackDensity Modification Inside Separatrix During ELMs

• The measureddensity range from 0to 3.2x1019 m-3 withtime resolution is 25µs

• Density increaseoutside theseparatrix , butdecrease inside theseparatrix duringELMs

2.1

2.2

2.3

2.4

840 850 860 870 880 890 9000

5

10

15

200

3.2

n e (1019

m-3

)

113679

Maj

or ra

dius

(m)

Dα (a.u.)

Time (ms)

Sepa

ratri

x

UCLAUCLA UCLA

Edge Profile Including Pedestal DensityEvolution During ELM

• Time slices show , density expanded to the wall first,followed by density pedestal collapses, then pedestalgradually re-builds up.

2.20 2.25 2.30 2.35 2.400

1

2

3

4

Major Radius (m)

n e (1019

m-3

)

113679

1 234

51 892 ms 2 892.3 ms 3 892.5 ms 4 893 ms 5 895.6 ms

time

sequ

ence

separatrix before ELM

Tile Position

UCLAUCLA UCLA

Time History of Typical QDB Plasma

• QDB, Quiescent DoubleBarriers —highperformance steady stateregime

• EHO, Edge HarmonicOscillation— a continuousmagnetic and electrostaticoscillation, provides theedge particle transportnecessary for QDB regime.

-1.2-0.8-0.40.0

024

024

6

0 2000 4000 60000

1020

Dα

B amplitude (a.u.)~

Line-averaged Density (1019 m-3)

Plasma Current (MA)

ELM-free , QH-mode Edge

Time (ms)

(a.u.)6

110899(a)

(b)

(c)

(d)

UCLAUCLA UCLA

SOL Profile Modulation by the EdgeHarmonic Oscillations in QDB Plasma

•Time resolution: 10 µs•Total 200 profiles in 2ms•SOL density profile ismodulated at the EHOfundamental frequency(about 6 kHz)•Corresponding well totime history ofassociated with EHO

˜̇B

Time (ms)

2.29

2.28

2.27

2.26

2.25

2.24

Majo

r Rad

ius (

m)

2974 2974.5 2975 2975.5 2976

-30

-20

-10

0

10

20

ne (1018 m-3)4.5 0.0

Separatrix

110899B (a.u.)~.

UCLAUCLA UCLA

No Profile Modulation When EHO isAbsent

• The same dataacquisition analysisparameters as in EHOcase

• The residualfluctuations are about2 mm, 5 timessmaller than theprofile modulationgenerated by EHO

2.29

2.28

2.27

2.26

Maj

or R

adus

(m)

2691 2692 2693 2694

-30

-20

-10

0

10

20

ne (1018 m-3)4.5 0.0

Separatrix

Time (ms)

110899B (a.u.)~.

2.25

2.24

UCLAUCLA UCLA

SOL Density Fluctuation Profile Associated withEHO

• RMS of density profileperturbation associatedwith EHO fundamentalfrequency (5-9 kHz)

• The peak of fluctuationsis at the large densitygradient regime

• Consistent with theLangmuir probemeasurement at 5 cmoutside the separatrixposition in shot 1038180

0.1

0.2

0.3

0.4

0.5

-1 0 1 2 3 4 5 6 7

0

0.01

0.02

0.03

0.04

0.05

0

1

2

3

2 2.1 2.2 2.3

Thomsonreflectometer

Separatrix

n e (1

013

cm-3

)δn

e (1

013 c

m-3

)

R-Rsep (cm)

Separatrix

Separatrix

SOLdensityprofile

Major Radius (m)

Density fluctuation associated with EHO

(5 - 9 kHz)

(a)

(b)

Langmuir probe result

110399

UCLAUCLA UCLA

Double Peaks Structure in FluctuationProfile Is Observed By BES Measurement

• BES measures notonly SOL densityfluctuation profilebut also inside theseparatrix, so willdual-polarizationreflectometer try tomeasure in future

• Double peakslocalized in twolarge densitygradient regimes

2.5

2.0

1.5

1.0

0.5

0.0

EHO

Am

plitu

de (

/n, %

)

230225220215

Major Radius (cm)

Separatrix

0.0

0.5

1.0

1.5

2.0

2.5

Separatrix

Edge

Den

sity

Pro

file

(101

9 m

-3)

Thomson scattering dataSpline fit of Thomson dataReflectometer profile

Shot: 110899 time: 2814 ms

BES data

UCLAUCLA UCLA

Double-Peak Structure in Fluctuation Profile Could BeDerived from Assumption of Edge Profile Rigid Movement

• When edge density profile rigid movement atEHO frequency is assumed , the fluctuationprofile could be double-peak structure

• The simulation indicates that EHO actuallycould locate inside or within the separatrixand rigidly modulate the edge density profile.

• There is discrepancy in fluctuation amplitudebetween reflectometer and BESmeasurements

• Further more investigation is neededseparatrix

0.0

0.5

1.0

1.5

2.0

220 222 224 226 228 2300

20

40

60

Reflectometer profile

Perturbation due to profile shift 1 mm

n n~A

mpl

itude

of

(%)

Major radius (cm)

Assumed Profile shifted Rigidly

UCLAUCLA UCLA

Summary

• The improved fast profile reflectometer has been successfully applied to studythe edge density profile modifications due to edge localized mode (ELM) andedge harmonic oscillation (EHO) in DIII-D plasma.

• The radial density velocity of in Type I ELM is about 500 m/s , which isconsistent with Langmuir probes and BES measurements, also consistent withELM radial motion velocity close to EθxB drift velocity.

• SOL Density profile is expanded at onset of ELM, indicating the enhancedradial particle and energy transports to wall during ELM.

• Inside separatrix density profile evolution , especially pedestal densityevolution, during ELM is observed by reflectometer with dual-polarizationreceived simultaneously.

UCLAUCLA UCLA

Summary (continued)

• SOL density profile is modulated at the fundamental frequency of theedge harmonic oscillation (EHO)

• Both BES and fast reflectometer data shows that the fluctuation profileassociated with EHO peaks in the large density gradient regimes. Thefluctuation amplitude measured by Langmuir probe at far away fromthe separatrix is consistent with reflectometer measurement.

• Assumption of rigid edge profile modulation by EHO fundamentalfrequency could also produce double-peak structure in fluctuationprofile.