T. Nakano Japan Atomic Energy Agency

W transport studies in JT-60U

3Sep2013ADAS Workshop

Badhonnef, GE

Tungsten:suitable for plasma facing components

for reactors

High melting pointLow fuel retentionLow sputtering yield (long life time)

UnsuitableHighly radiativeNarrow operation window as PFCs

( TDBTT< T <Trecrystalliation)Neutron damage ( transformation, etc )

Tungsten:suitable for plasma facing components

for reactors

High melting pointLow fuel retentionLow sputtering yield (long life time)

UnsuitableHighly radiativeNarrow operation window as PFCs

( TDBTT< T <Trecrystalliation)Neutron damage ( transformation, etc )

Present study: Suppression of W accumulationPresent study: Suppression of W accumulation

T~104 eVn~1020 m-3

Wq+

(q~40-60)Transport

W divertor plates

Tungsten: a candidate for PFCs in reactors

W divertor plates in JT-60UW coated CFC tiles:

50 m with Re multi-layer

11 tiles (1/21 toroidal length )

W tiles

W tilesDome

(C)

Dome(C)

Inner Div.(C)

Inner Div.(C)

OuterDiv.(C)

OuterDiv.(C)

Standard configuration

W tileW tile

W tileW tile

W exp. configuration

Diagnostics Short-wavelength VUV spectrometer

( 0.5- 40 nm ) – On-axis : W XLVI intensity (core)

Long-wavelength VUV spectrometer( 20 – 120 nm )

Off-axis: sensitivity calibration Visible spectrometer

– sensitivity calibration

PIN Soft X-ray (>3keV) CXRS Toroidal rotation TMS Te, ne

FIR, CO2 line density

*) M.F.Gu, Can. J. Phys. 86 (2008) 675. http://sprg.ssl.berkeley.edu/~mfgu/fac/

Identification of VUV spectrum (on-axis)

•W41+ - W52+ were identified•Isolated W45+ line (W XLVI) at 6.2 nm is used for W density•W41+ - W52+ were identified•Isolated W45+ line (W XLVI) at 6.2 nm is used for W density

1. Wq+ spectrum <= FAC*2. Adjust Fractional

Abundance (FA)3. Wq+ spectrum x FA 4. Sum-up5. Comparison with

observed spectrum

Steps of spectral analysis:

0.5

keV

4x10

19 m

-3

* ¼ picsFWHM

*) M.F.Gu et al., Astrophys. J. 582 (2003) 1241. http://sprg.ssl.berkeley.edu/~mfgu/fac/

Identification of VUV spectrum (on-axis)

•W41+ - W52+ were identified•Isolated W45+ line (W XLVI) at 6.2 nm is used for W density•W41+ - W52+ were identified•Isolated W45+ line (W XLVI) at 6.2 nm is used for W density

1. Wq+ spectrum <= FAC*2. Adjust Fractional

Abundance (FA)3. Wq+ spectrum x FA 4. Sum-up5. Comparison with

observed spectrum

Steps of spectral analysis:

I W45+(6.2 nm): 4s 2S1/2 - 4p 2P3/2 =

Excitation rate

I W44+(6.1 nm): 4s4s 1S0 - 4s4p 1P1

Close excitation energy (199 ev and 204 eV)Similar energy dependence of Ce

Close excitation energy (199 ev and 204 eV)Similar energy dependence of Ce

~ 0.44 (Ioniz.rate)

(Recomb.rate)

Ioniz. Equi.

Calculation

Measurement

Evaluation of W44+ ionization / W45+ recombination rate

Waveform of Negative Shear discharge with EC injection

Negative shear discharge-W accumulation occurs

Te decrease from 10 keV to 5 keV

During Te decrease, IW45+ and IW44+ increases, and then decreases

Te -scan data for W45+ / W44+

Comparison with ionization equilibrium

Te -scan data for W45+ / W44+

Comparison with ionization equilibrium

ECNB

IP

Te

ne

FAC calculation reproduced measured W45+/W44+

Accuracy of ionization/recombination rates calculated with FAC were evaluated in JT-60U experimental data

Accuracy of ionization/recombination rates calculated with FAC were evaluated in JT-60U experimental data

Cal:nW

45+ / nW44+

= S44+ / 45+

Exp:nW

45+ / nW44+

= I45+ / I44+ / 0.44

Waveform of W accumulation shot

Switch Co. to Ctr NBs.With decreasing VT, W XLVI increases, while W I is constant. W accumulationThe same phase between W XLVI and SX(5)W XLVI is a measure inside the Sawtooth layer

Systematic experiments on W accumulation against VT were performed

Systematic experiments on W accumulation against VT were performed

Ctr

Co

Neutral Beam

Plasma rotation and central heating effective in avoiding W accumulation

T. Nakano and the JT-60 team, J. Nucl. Mater. S327 (2011) 415.

3%

Radiation collapse

Radiative power ( line radiation ) is highest between 2 – 4 keVDominant charge states change at Te ~ 4 keV from highly raditive n=4-shell to lowly radiative n=3-shellDecrease of Lw

Radiative power ( line radiation ) is highest between 2 – 4 keVDominant charge states change at Te ~ 4 keV from highly raditive n=4-shell to lowly radiative n=3-shellDecrease of Lw

Radiative power rates calculated with FAC

4f4f

*T Putterich et al Nucl. Fusion 50 (2010) 025012

Comparison of calculated radiative power rate with NLTE5 workshop results**

FAC calculation is in agreement with the NLTE5 resultsFAC calculation is in agreement with the NLTE5 results

**Y Ralchenko et al AIP Proceedings 1161 (2009) 242*T Putterich et al Nucl. Fusion 50 (2010) 025012

Evaluated radiative power in agreement with bolometoric measurement

PBOL = Pbefore – Pafter

PNB = 15 MWPrad

core ~ 4 MW(Te ~ 5 – 6 keV )

Negative Feed-Back seems to result in radiation collapse:W accumulation => Radiation increase => Te decrease => Lw increase => Radiation increase => …

Negative Feed-Back seems to result in radiation collapse:W accumulation => Radiation increase => Te decrease => Lw increase => Radiation increase => …

Radiation collapse

Summary and Conclusions

W XLVI ( 6.2 nm ) intensity was measured with absolutely calibrated VUV

spectrometers.

Validity of Ioniz./Recomb. rate calculated with FAC was confirmed from W45+/W44+ density ratio under ionization equilibrium with coronal model.

Quantitative measurement of -W density: ~ 10-3 in W accumulation cases.

>> ITER allowable level (10-5). -W radiative power: agrees with bolometoric measurement

Thank you!

W63+(3s-3p,3p-3d) at 2 nm identified in JT-60U*

Calculated by FAC12 keV, 4x1019 m-3

JT-60U

Wavelength ( nm )* J. Yanagibayashi, T. Nakano et al., accepted to J. Phys. B**Y. Ralchenko et al J. Phys. B 41 (2008) 021003

EBIT(NIST)** 3s-3p lines at 7-8 nm identified in EBIT** were reproduced by the FAC calculation. 3s-3p at 2.3 nm 3p-3d at 2 nm

3s-3p lines at 7-8 nm identified in EBIT** were reproduced by the FAC calculation. 3s-3p at 2.3 nm 3p-3d at 2 nm

The W63+ line at 2.3 nm will be a good diagnostic line for ITER high temperature plasma.

The W63+ line at 2.3 nm will be a good diagnostic line for ITER high temperature plasma.

2 co-tang. PNBs (~4.5MW)21

Neutral Beam injectors• 11 positive-ion-based NBs (PNBs~85keV) • 2 co-tangential NB, 2ctr-tangential NBs, and 7 perp. NBs. Combination of tangential and perpendicular NBs leads to

wide range of toroidal rotation.

7 perp. PNBs(~15.75MW)

2 ctr-tang. PNBs(~4.5MW)

3d104s

3d10

Radiative = 1 / 4.4x1011 = 2.3x10-12 sExcitation = 1 / 7.8x10-16 4x1019 = 3.2x10-5 sIonization = 1 / 1.2x10-17 4x1019 = 2.1x10-3 s

Radiative << Excitation < Ionization

Radiative = 1 / 4.4x1011 = 2.3x10-12 sExcitation = 1 / 7.8x10-16 4x1019 = 3.2x10-5 sIonization = 1 / 1.2x10-17 4x1019 = 2.1x10-3 s

Radiative << Excitation < Ionization

Comparison of time scales of atomic process:Colonal model is valid

4p

n=5

4d4f

Exc

itatio

n

Ioni

zatio

nR

adia

tive

tran

sitio

n 4.

4x10

11 s

-1

1.2x10-17

7.8x10-16

5 keVW45+

W46+

nene

Deexcitation is dominated by radiative transitionDeexcitation is dominated by radiative transition

W generation

W sputtering yield against D ~ 0.25% ( too high )Possible W sputtering mechanisms

• by impurity ( C )• by high energy particles during ELM

W sputtering yield against D ~ 0.25% ( too high )Possible W sputtering mechanisms

• by impurity ( C )• by high energy particles during ELM

Te~ 20 eV

High energy particles seem a key for W sputtering

Te~ 20 eV With decreasing VT, Yphys.decreases while Te increases Opposite trend Needs ELM-resolved

data With decreasing VT, ELM frequency becomes high and Wdia decreases*Similar trend between Yphys. and Wdia

Time average ~ 1 s

W sputtering is possibly due to high energy particles expelled during ELM

W sputtering is possibly due to high energy particles expelled during ELM

*) K.Kamiya et al., Plasma Phys. Control. Fusion 48 (2006) A131.

Tungsten as a plasma-facing componentMerit : high melting point => compatible with high temperature fusion plasma

: low hydrogen (T) retention => safety, economy: low sputtering yield => long lifetime: low dust production

Demerit : high Z (74) highly radiative ( allowable nW/ne < 10-5) accumulation in the core plasma

Issues of W transport studyUnderstanding of

Transport in core plasma* => accumulation mechanism in core plasmaLocal transport in divertor, global migration,,,

Control ofW generation, W penetration, W accumulation,,,

Preparation of diagnostics at high Te ~ 15 keV ( ~ Wq+ : q > 60)Evaluation of W density, W ion distribution*, radiative power,,,

Tungsten in Fusion Research

Cross section of ITER

W Plasma

Divertor

*present study

*) M.F.Gu et al., Astrophys. J. 582 (2003) 1241. http://sprg.ssl.berkeley.edu/~mfgu/fac/

Requirement for W atomic data=>calculation with an atomic structure code,FAC*

① 二電子性再結合断面積の計算 ② JT-60U, LHD スペクトルの解析

Significant difference in Ionization equilibrium

Te ( eV )Te ( eV )103 104103 104

FLYCHK code

LLNL code

*T Putterich et al Plasma Phys. Control. Fusion 50 (2008) 085016

Atomic data ( Ioniz./Recomb. rates ) are still to be checked Atomic code calculation with FAC Experimental validation in JT-60U plasmas

Atomic data ( Ioniz./Recomb. rates ) are still to be checked Atomic code calculation with FAC Experimental validation in JT-60U plasmas

44+45+

46+

AUG*

Ionization equilibrium:Difference between AUG* and FAC calculation

*T Putterich et al Plasma Phys. Control. Fusion 50 (2008) 085016

Still different:Shift to lower Te

in AUG calculation

1

0.1

0.01

44+

45+ 46+

Fra

ctio

nal A

bund

ance

AUG*

FAC

Ionization equilibrium:Sq+=>(q+1)+ ･ nW

q+ = (q+1)+=>q+ ･ nW

(q+1)+

S = Sdirect + Sexcit.autoioniz.

= radiative + die-electronic

*present study

*S Loch et al., Phys. Rev. A 72 (2005) 052716 **T Putterich et al., Plasma Phys. Control. Fusion 50 (2008) 085016

Accurate recombination rates required=> Calculated with FAC

Present Ref**

Ionization FAC (DW) Loch code* (DW)

Dielectronic RecombinationW44+-46+ : FACthe others: ADPACK mod. ADPACK mod.

( x 0.39 )Radiative Recombination FAC

Te ( eV )

W confinement ti

me:

~

0.5 s inside sawtooth layer

Present work: nWtotal = I WXLVI / Cexcite / ne / FFA(45+) / r

ST ( m-3 )W = S/XB * I WI ( 1/s )

W = nWtotal * Vp

ST / W ( s )

Present work: nWtotal = I WXLVI / Cexcite / ne / FFA(45+) / r

ST ( m-3 )W = S/XB * I WI ( 1/s )

W = nWtotal * Vp

ST / W ( s )

*) T. Nakano et al., Nucl. Fusion 49 (2009) 115024.

Significant W accumulation at negative toroidal rotation*

Previous work*: W accumulation was evaluated in A.U.Previous work*: W accumulation was evaluated in A.U.

I W X

LVI /

I W

I n e

(0)

( a.

u.)

Calculation model: Example for W 15+

Electron configuration:

4d10 4f11 5s2

4d10 4f11 5s1 5*1;5s=0

4d10 4f12 5s1

4d10 4f11 5s1 6*1

4d9 4f12 5s2

Coronal model

Excitatio

n

Excitatio

n

Radiative

transitio

nRadiative

transitio

nAtomic structurecalculation

Energy level:

Excitation rate:Radiative transition rate:

Calculation model: Example for W 15+

Term Energy ( eV )Pop

ulat

ion

norm

aliz

ed a

t th

e gr

ound

leve

l

4d9 4f12 5s2

4d10 4f11 5s2 (Ground state)

Electron configuration:

4d10 4f11 5s2

4d10 4f11 5s1 5*1;5s=0

4d10 4f12 5s1

4d10 4f11 5s1 6*1

4d9 4f12 5s2ExcitationRadiative transitionCoronal model

Calculated W spectra

JT-60U peripheral plasma: two peaks needed

*) T. Nakano et al., Nucl. Fusion 49 (2009) 115024.

Contents

Introduction Experimental set-up/Diagnostics

- Absolute calibration of VUV spectrometers Results

- Evaluation of Ionization equilibrium- Quantitative evaluation of W confinement time, density, radiative power- W generation

Conclusions

Long-VUV

Visible

Sensitivity Calibration of VUV spectrometers:“ Triple” Branching ratio method

Short-VUV

Short-VUV

Sensitivity Calibration of VUV spectrometers:

Absolute sensitivity ~ 6.2 nm was obtainedW XLVI is used for W density measurementAbsolute sensitivity ~ 6.2 nm was obtainedW XLVI is used for W density measurement