5 Nov. 2019, IAEA Headquarters, Vienna, Austria

Atomic Daestimate Da [atom]

F (n)

ETe

RTeBn->2rec

Bn->2exc

Measured Da

ne

Bn->2

Da [D2,D2+,D-]

- ETe

Da [D2]

Da [D2+, D-]

EstimateDamax D2

ne

nD₂/neΔL

RTe

Db [D2+, D-]

SeparateDa [D2+, D-]

neRTe

TD₂

EstimateDamax D2

ne

nD₂/neΔL

RTe

Spectroscopic analysis

Previously, atomic particle/powersources & sinks measured [1,2]

Particle balance

Power balance

Dα measurements

TCV density ramp: 3 phases

1) Attached & I (ion current)t I (ion source)i

rise linearly with ne

>> Precl Pion

2) Detachment start flattens -> deviates from Ii It

linear increase with ne

~ 2 Precl Pion

3) roll-overIt

rolls overI (ion source)i

lesser roleI (recombination)r ~ Precl Pion

At detachment onset, Dα rises above estimationsatomic

+D

D-

D

D2+

D2

+D3

Dα

Mol. Dα too low, but in+(D , D) SOLPS2

agreement after post-processing

• -> reactions with Mol. Dα + -D D2

• Underestimated by SOLPS-> Radiation (power losses)-> Particle gains/losses (MAR/MAI) ‘Post-processed’ SOLPS: AMJUEL (vibr. states)

Preliminary results

9

Investigating the in�uence of molecules on power/particle/momentum balancein the detached TCV divertor

Dα > expected (atomic)

Conclusions

Measurement details (spectroscopy)

Third IAEA Technical Meeting on Divertor Concepts

a b,aa,b b c b a c d,a a c K. Verhaegh , B. Lipschultz , B.P. Duval , B. Dudson , A. Fil , O. Février , D.S. Gahle , J.R. Harrison , L. Martinelli , D. Moulton , O. Myatra , e c c * **A. Perek , C. Theiler , M. Wensing and the TCV and EUROfusion MST1 teams

PII 37

Measured Estimated (atomic)

SOLPS (syn. diag) Total (AMJUEL Atomic

Molecular (AMJUEL Molecular (SOLPS)

0

1

2

0.2 0.3 0.4 0.5

D

α (

10

21 p

h/s

)

Core Greenwald fraction

0

5

10

15

0

0

50

100

150

200

10

21 io

ns/

s

It

Ii

Ir

Pow

er (k

W)

Precl

Pion

Pdiv

vibr. states)

vibr. states) • Increased (mostly ) hydrogenic radiation +D2

• Increased ion losses ( ) (due to & ) - MAR +D2

-D

larger than ( )electron-ion recombination EIR

- - ( + ) < Ionisation MAR EIR ion target �ux

• Onset of near ‘power limitation’+D MAR2

/detachment, followed by andD MAR EIR

• Increase ion source ( & D₂) negligibleMAI +D2

0.4 0.5

Cause? ? Moleculesreaction -> excited atom -> emission

0

10

15

4

7

10

30

0.2 0.3 0.4 0.5

Core Greenwald fraction

+D

D-

D

D2+D2

+D3

Dα

Position wrt target (m)0.1 0.30.2

target x-point

Spatial profiles (n / n ~ 0.48)e GW

Density ramp

Particle balance2110 ions/s

Particle losses21

10 ions/s

Ion target currentIonisationRecombination

+MAR H2

MAR -H

Hydrogenic radiation(kW)

1

2

Fulcher (600-616 nm) 4

emission (x 2 10 )Break-down of Dα

21 2(10 ph/s m )

0

0

0

Ion sinks /sources21 2(10 part. / m s)

33

66

4

8

12

5

0

0

#56567

#56567

Preliminary conclusion / DEMO implications :

• Signi�cant contributions-> Balmer lines

• (eg ) near cold target ->Mol. Dα +D2

- Dassociated MAR/rad. loss near target;+D2 ,

- D

below ; wider than ion. region EIR region

• [5] atFulcher (600-616 nm) emission different location than ion Dα emiss

+D2 - D

No extra Dα with N seeding2

assuming

• Dα emission & anti-correlation Dα & I cannot be explained with atoms t+ - • Additional Dα could be due to D (and/or D ) - only high with CX & vibr. states2

• Additional Dα does not appear during N seeding 2

• Particle/power losses plasma-mol. interactions have been analysed experimentally

[3]

[1]: K. Verhaegh, et al. NF, 2019[2]: K. Verhaegh, et al. PPCF, 2019[3]: K. Verhaegh, Thesis, 2019[4]: D. Wünderlich, U. Fantz; Atoms; 2016[5]: U. Fantz, et al. 2001, JNM[6]: M. Groth, et al. 2019, NME

a Culham Centre for Fusion Energy, Culham, UKb York Plasma Institute, University of York, UK

c Swiss Plasma Center, EPFL, Lausanne, Switzerlandd University of Strathclyde, Glasgow, UK

e DIFFER, Eindhoven, The Netherlands* See author list of ‘B. Labit, et al 2019, Nucl. Fusion 59 086020'** See author list of ‘S. Coda, et al 2019, Nucl. Fusion accepted'

Detachment introduction

• Detachment: Simultaneous reduction , T t It

Requires:• Particle losses (ion sink/power limitation• Power losses• Momentum losses

kevin.verhaegh@ukaea.uk

Analysis: Balmer n=5,6 atomic only

rec

RTe

Existing atomic particle/power sink source analysis [2]

Recombination rate

Ionisation rateIonisation rateIonisation rateIonisation rateIonisation rate

+Da [D ]2

Balmerline shape

(n>6)

Stark broadening

ne

Balmerline ratio

Frec(n)

Frec(n)

Bn->2

Bn->2exc

Bn->2rec

Recombination rate

ne ΔL

neno/ne

IL

Prad,Lexc

RL

ΔL n/no e

Prad,Lrec

Fexc(n)

Analysis steps InputOutput

ETe

Molecular particle/power sink source analysis; with Yacora [4]

MAI per Dα+[D₂ D₂ ] (AMJUEL)

Rad. per Dα+ [D₂ D₂ D ] (YACORA)

Atomic Dαestimate Dα [atom]

ETe

RTeBn->2rec

Bn->2exc

Measured Dα

ne

Bn->2

- ETe

[D2]

+ -Dα [D₂ D₂ D ] Dα [D₂] SOLPS estim

Dα [D₂]

+ -Dα [D₂ D ]

Separate(with Dα/Dβ)

+ -Dβ [D₂ D ]

+Dα [D₂ ]-Dα [D ]

MAR per Dα+ [D₂ D ] (AMJUEL) MAI

MAR

Rad. power

Obtain : molecular Dαmolecular = -total atomic

Separate mol. Dαin parts+ D₂ D₂ D

Dα ( ) X + D₂ D₂ D rad./reac per Dα photon ratios

Molecular = -Total AtomicAssume all ‘missing’ Dα -> plasma-molecule

Veri�ed against syn. diagnostic (SOLPS)

Estimation mol. contributions other Balmer lines

Iterative scheme

Balmer n=3,4,5,6 - atomic & molecular

Density ramp

• Signi�cant contributions ocation emission D different to 2+

D2

- D

emission, & regionradiation MAR

Lyβ(~40%)

Lyα (~40%)

Emission /absorption

[SOLPS simulations MAST-U post-processed]

-> Different location , Lyβ Lyαemission -> opacity of both signi�cant

Total

15

Contribution plasma-mol.Balmer lines (%)

n=3n=4n=5n=6

• Power losses: signi�cant plasma-mol. rad in hydrogen spectra during detachment (similar to atomic excitation radiation); losses mol. bands small [6]; • Particle losses/gains: signi�cant MAR larger than EIR• Plasma-mol. interaction can in�uence Balmer & Lyman series lines

• The inclusion of those losses in plasma-modeling codes may be limited-> important for extrapolating to DEMO

Plasma-mol.-> implications for opacity