EU-PWI TF meeting, Caderache, 17-19 October 2005 Progress with T-recovery techniques 2004-5 P Coad 1, GF Counsell 1, G. Dinescu 6, H. G. Esser 4, JA Ferreira.

EU-PWI TF meeting, Caderache, 17-19 October 2005

Progress with T-recovery Progress with T-recovery techniques 2004-5techniques 2004-5P Coad1, GF Counsell1, G. Dinescu6, H. G. Esser4, JA Ferreira2, MJ Forrest1, M. Freisinger4, K Gibson1, C Grisolia3, A Grosman3, H-K Hinssen4, C. Hopf5, W. Jacob5, A Kreter4, K Kuhn4, W Jacob5, A. Lyssoivan4, R Moormann4, J-M. Noterdaeme5, V Philipps4, H. Reimer4, V. Rohde5, U. Samm4, A. Semerok3, G. Sergienko4, and M. Schluter5, FL Tabarés2, E Tristrone3

1 UKAEA2 CIEMAT3 CEA4 IPP-Juelich5 IPP-Garching6 Romanian Ass.


Both retention mitigation and in situ detritiation will likely be needed in ITER

• ITER with planned PFC mix will need

– a factor 30-200 reduction in tritium retention

or

– a factor 10,000 increase in tritium removal rates

compared to model predictions and extrapolations

• Almost certainly a combination of both will be required

– even if inventory limit takes 8 months to reach, removal rates would still need to be a factor 50 improved to clean vessel in 4 month shut-down


• Many techniques and technologies under consideration (though world-wide effort is still small)

• Several issues -

– intra-shot, inter-shot, overnight, weekly

– Collateral damage to in-vessel components (inc. PFC’s)

– Compatibility with plant (pumps, heating systems)

– Degree of intervention (vessel let-up?, fields energised?, vessel entry?)

– Access (i.e. to all regions of retention)

• Existence of a Be Wall (in particular) impacts on many of these issues

Be Wall in ITER may help reduce retention but also complicates other approaches


RF assisted O2 GDC in TEXTOR

Trilateral Euregio Cluster Assoziation EURATOM-Forschungszentrum JülichInstitut für Plasmaphysik

Total ion flux: 6 A

Wall area: 35 m2

1·1014 O+/cm2 s

Wall(-)

Removal rate:neutral pressure pumping speed

RF 13.2 MHz250 W

DC 6A (feedback controlled),400-600V

Electrode (+)

Sheath potential 280-400 V (O+ impact energy)

O2 + He

10-3 mbar

O+,O

CO

C

O+

CO, CO2, O2, He

Differentially pumped

quadrupole mass

spectrometer (QMS)

• CO weakly dependent on O2

• CO2 rises with O2

• CO/CO2 independent of He

A Kreter et al

~150C


• HD molecules about 7% of removed C

H+D atoms about 28% of C (assuming H:D = 1:1), in agreement with previous observations

• C removal rate 2.3x1019 C/s (5.2 g C in total)

Removal rates by RFA-O2 GDC


GDC GDC

Integral data from QMS

CO

CO2

HD

00:00 01:00 02:00 03:00 04:000.0

0.5

1.0

1.5

2.0

2.5

rem

ova

l, 10

23 a

tom

s

time, hh:mm

0

1

2

3

4

5

rem

ove

d C

, g

A Kreter et al


ICRF conditioning in O2 in TEXTOR

• All injected O2 converted into CO, CO2

• Only 15% of C removal during ICRF low duty cycle

• CO+CO2 outgassing until 140 sec: ~4·1020 O-atoms

about half of O amount absorbed by wall, wall not saturated

Typical shot:

• Bt = 2.3 T

• ICRF 29 MHz 50 kW (absorbed) 1-6 s

• Continuous He flow

• Fast O2 injection 2.1·1020 O/s 1.1-5.1 s (8.5x1020 in total)


0 50 100 1500.0

0.1

0.2

0.3

time, s

par

tial

pre

ssu

re, 1

0-3 m

bar

mass 44 mass 28 mass 32 mass 4 mass 3

CO2

He+D2

CO

O2

#97084IC

RF

HD

A Kreter et al

C removal rate 1.8x1019 C/s



TechniqueC removal rate

Advantages DrawbacksPossible

improvements

Oxygen venting

2.5·1018 C/s for 0.3mbar, Twall=620K

•simplicity•access to all wall areas•selective removal of redeposited layers

•low removal rates•Twall > 600K needed

•higher O pressure

Glow discharge conditioning

2-3·1019 C/s•higher removal rates •applicable for low Twall

•incompatible with steady state B•non-selective carbon removal (?)•limited wall area access

•higher GD current•higher pumping rate

ICRF conditioning

1.8·1019 C/s for 1:10 duty cycle for pump out

•higher removal rates•applicable for low Twall

•compatible with steady state magnetic field

•O injection limited by pressure limit at antenna box•non-selective carbon removal (?)•limited wall area access

•higher pumping rate (also for steady state ICRF)

Summary of oxygen cleaning in TEXTORIntegral TEXTOR carbon redeposition rate ~ 2.7·1020 C/s

A Kreter et al


O2 GDC in AUG

Use of He/O2 mixture plasmas in AUG• Stability of glow discharge• Less sputtering compared to pure O2 discharge or heavier noble gas admixtures• Oxidation of W in O2 plasma saturates and is reversible in an H2 discharge• No erosion in shielded places: tile gaps, behind first wall, and in deep in the divertor

Exposure time (min)Ch. Hopf et al


Thermal Desorption spectra of H2 and CH4 from gaps

a-C:H coated Thermocoax 1mm

Al mask: 4mmDepos. in H2/CH4 GDCleaning in 5%O2/He GD

As deposited After 45’ GD

J.A. Ferreira and F.L. Tabarés, CIEMAT


• Co-deposit removed at ~0.5m/h at 185 C (prob. higher at 130 C)20m co-deposit removed in 2d also removes 0.7mm EK98 (ass. surface oxidation)

• Eroded surface becomes roughened & chemisorbtion forms stable C-O complexes (to >700C)

125C

135C130C

~ 1mm EK98

~ 4

0m

/h E

K98 • Oxidation rates of TEXTOR

EK98 for 2.3% O3 in O2

• Peaks at ~50m/h at 130C

• Decreases with burn-off


Oxidation with Ozone

H-K Hinssen et al

EFDA/04-1174


N2 seeding – impact of N ions on a-C:H

• N2+ beam irradiation of a-C:H films in

MAJESTIX

• Chemical sputtering dominates a-C:H erosion by N2

+ at low energies

• Erosion yields remarkably high: ~1 above ~50eV

W Jacob et al


• Flash-lamp assembly used for in-vessel trials transferred to BeHF

• Operated remotely – PSU and water cooling outside BeHF

• OPL and G4 tiles exposed to ~80x250J pulses at range of locations

• Tritium off-gas measured

Flash-lamp trials in JET BeHF


• Flash-lamp footprint covers ~30cm2

• Surface temperature rises to ~1200K during 250J flash on plasma-damaged tile

• Pulse half-width is short - less than 150 s is typical

• Peak power density ~180MW/m2 at target for 250J flash



Treated areas

Untreated areas

Heavily treated area with initially thick powdery deposit

• Clearly visible changes to co-deposit in treated areas



-10

0

10

20

30

40

50

60

70

1700 1750 1800 1850 1900 1950 2000 2050 2100

Channel Number

Coun

ts

JET 8360 (reference)

JET 8374 (treated)

• ~0.5 GBq of T released in ~20ms exposure to flash-lamp from ~50cm2 of G4 tile

• No observed T release from OPL

• NRA spectrum for D, shows that D is absent from the outer 0.5-1 m at the surface – but at least 7m codeposit remaining

• No C13 on exposed regions compared to unexposed control

• T analysis at FZK shows difference between exposed and unexposed regions within statistical variation

• Total T content ~5GBq on peak regions of G4 tile

• Results consistent with removal rate ~0.2m/flash at 250J – lower than expected

• Flash-lamp focus less sharp than predicted, energy density ~3J/cm2 at 250J.

• Trials will be repeated in November at 500J, now available.



(0.4-0.5) J/cm2 for co-deposited layer

(2.5±0.5) J/cm2 for graphite

First trials with high 100 ns repetition rate 250W mean power Nd-YAG laser :

Laser treatment TORE SUPRA


20 W, λ≈1 μm, 10kHz, 100ns pulse duration

h~50m

h~0m10 scanings

h~50m1 scaning, 2s

TEXTOR tile

• Ablation at 50 cm• Field depth of some centimetres• 100 W laser power 1 m2/h of 50 μm co-deposit

Laser treatment TORE SUPRA


YAG V1 IPG Fiberlaser

Wavelength (nm) 1064 1064Power (W) 200 20Repetition rate (kHz) 10 20Energy (J) 20 1Pulse length (ns) 70 120Divergence M2 35 1,6Beam diameter (mm) 1 0,25Depth of focus (mm) ±2 ±30Fluence (J/cm2) 2.5 2

TORE SUPRA

Trials of scanning Fibrelaser planned for JET in vessel or BeHF

Laser treatment


Ar torch cleaning

•Plume: ~ 2mm; length ~ 10mm •Optimal distance to surface ~ 10mm•RF power: 10 to 100 W•Water cooling system needed•Pressure: 10-1100 mbar (argon)

Detritiation assessment by exposure to a small plasma torch planned for 2005-2006 WP:

•Surface scanning•Torch diameter reduced to 20 mm•RF power increased to 200 W•Test on codeposited material•Test in magnetic Field

G Dinescu et al

TORE SUPRA


Selected other activities World-wide:

• ICRH on LHD

PICRH~8-149kW, 38.47MHz, 3 loop antennas, He 0.01-0.1 Pa, 3s on:2s offDischarge stable even at low pressure but H removal ~5-10 x higher with 20-40kW GDC at 5Pa He

• O-ventilation, O-ICR, 4:1He/O-ICR and O-GDC + O-removal explored on HT-7

Fastest removal with 1.5Pa O-GDC (~5.1x1014 C/cm2s). Factor 3-4 lower in 0.1Pa O-ICR and 6-10 lower in 4:1He/O-ICRWall temp only 400-450CHe/O-ICR reduces oxygen retention – 5.8x1012 O/cm2sHe-ICR or He-GDC both effective to remove oxygen (10-50% in 0.5-1h)

• Mitigated disruption removal: ITPA-led Proposal submitted (P Stangeby leading) throught JET TFE for trials using newly fitted fast, high throughput valve


Be (and its alloys) may act at significant trap –

O2 from leak:15mg/shotIntrinsic O in Be 1.5 - 5mg/shot - convective erosion50mg/shot - ELMs (no melt loss) 0.1% O impurity0.5g/shot - Unmitigated disruptionsTarget CFC erosion0.15g/shot (reduces from 2-5g/shot)

Conclusions - - Purity of Be a critically important issue (how would O2 de-tritiation schemes affect this?)

- Melt loss during Type I ELM interaction with limiter Be could result in large T retention

- Be might substantially reduce aC:H T retention - but still significant due to ELM erosion

Need to explore T-removal techniques for Be trapping?

aC:H is not the only T-trap

EU-PWI TF meeting, Caderache, 17-19 October 2005 Progress with T-recovery techniques 2004-5 P Coad 1, GF Counsell 1, G. Dinescu 6, H. G. Esser 4, JA Ferreira.

Documents

fast o

mbar o

half of o

o impact energy o

injected o

c slide

eupwi tf meeting

c grisolia