RFX-mod programme workshop 2011 February 7 - 9, 2011
Wall conditioning techniques: status and
perspectives
S.Dal Bello, A.Canton, L.Grando, P.Innocente, S.Munaretto
RFX-mod programme workshop 2011 - February 7- 9
Outline
Why wall conditioning is needed/desired
First Wall conditioning in RFX-mod
Effects of present wall conditioning
techinques on plasma properties in RFX-
mod and in other Experiments
Open points and activities aimed at
improving the present wall conditionig
effectiveness
Conclusions
RFX-mod programme workshop 2011 - February 7- 9
FW cleaning (after vessel venting)
graphite is a good absorber of humidity(after the production, tiles were degassed at 2000°C)
even the vessel is continuously vented by dry N2 (overpressure of hundreds of mbar), during the shutdown some air retro diffuses through the open porthole
Why FW conditioning
RFX-mod programme workshop 2011 - February 7- 9
Why FW conditioning (2)
Impurities control (performance)
the graphite wall prevents the presence of high-Z impurities
main impurities in RFX-mod: C, O2, B or Li (when conditioning procedures applied)
Zeff in the range 2-4
RFX-mod programme workshop 2011 - February 7- 9
Wall desaturation
Why FW conditioning (3)
at a certain time, reached the “saturation level” (that is not a parameter, yet) the removal of Hydrogen from the wall is necessary to operate the machine at “allowed” density values
RFX-mod programme workshop 2011 - February 7- 9
First Wall reset
Why FW conditioning (4)
Between to (very) different experimetal campaigns the status of the wall is considered needy of a reset.
The resetting wall treatment should remove the memory of different PW interactions
RFX-mod programme workshop 2011 - February 7- 9
Possible control of recycling – density (performance)
Why FW conditioning (5)
We would try to modify the behaviour of the wall (recycling factor 1) increasing the operational range we have to modify/obtain the “desired” density, hopefully at different plasma currents.
RFX-mod programme workshop 2011 - February 7- 9
FW conditioning in RFX-mod
BakingGlow Discharge Cleaning
Plasma discharges (i.e. He)
Boronization
Lithization
What are the possibilities we have:
RT Pellet injector (solid Li)
LLL used as limiter or evaporator (liquid Li)
(performed by Glow Discharge @170°C)
RFX-mod programme workshop 2011 - February 7- 9
170 °C
Tav
time (hours)
10 20 30 40 50 60 70 80 90
H2GDC
He
GDC
25 °C
100
- Temperature rised to 170°C and kept for the whole sequence- Baking (4h) for outgassing pressure stabilization- H2 GDC (30-40h) to extract O2, N2 (CH4 also formed and
removed)- He GDC (6-10h) to remove H2
- Baking to favorite the removal of He (15h)
RF-assisted Glow Discharges in continuous flow
Executed at Room Temperature (during experimental sessions as desaturation technique) or at high temperature
RFX-mod programme workshop 2011 - February 7- 9
170 °C
Tav
time (hours)
10 20 30 40 50 60 70 80 90
H2GDC
He
GDC
25 °C
100
B2H6
+
HeGDC
precursor gas: B2H6 (Diborane) 10% - He 90%
RF-assisted Glow Discharges in continuous flow
In 2010 (from Jan to Oct) we performed 4+1 boronizations: 30’, 120’, 240’, 240’ + 360’ (to possibly reduce the Li2CO3
formation/effects)
Boronization @170°C by
RFX-mod programme workshop 2011 - February 7- 9
driver gas
sabot loader
pumping gate
optical detectors
bumper and recovery box
sabot
pellet
Pellet feature:
V = 8.8 mm3
4x1020 atoms
∆ne~1020 m-3
In both the cases, we deposited in or on (?) the wall approximately 2g of Lithium
Lithization: during pulses (pellets and LLL) or between pulses (LLL used as evaporator)
RFX-mod programme workshop 2011 - February 7- 9
Why wall conditioning is needed/desired
First Wall conditioning in RFX-mod
Effects of present wall conditioning
techinques on plasma properties in RFX-
mod and in other Experiments
Open points and activities aimed at
improving the present wall conditionig
effectiveness
Conclusions
RFX-mod programme workshop 2011 - February 7- 9
... on impurities ...
Impurity reduction
• The main sources of impurity at RFX-mod are C and O
• For both after Lithization the influx is lower than with clean graphite
• The main sources of impurity at RFX-mod are C and O
• For both after Lithization the influx is lower than with clean graphite
0
20
40
60
80
100C wall
Li wallB wall
0 1 2 3 4 5 6 7
<ne>*1019 (m-3)
0
1
2
3
4
5
0 1 2 3 4 5 6 7
C wall
Li wall
B wall
<ne>*1019 (m-3)
RFX-mod programme workshop 2011 - February 7- 9
Graphite WallLi WallB Wall
0
1
2
3
4
5
6
7
8
0.0 10.0 20.0 30.0 40.0 50.0
Filled Particles (*1020)
0.0
1.0
2.0
3.0
4.0
-2.0 0.0 2.0 4.0 6.0
P (
*10
-3 m
Bar
)
t (s)
#27710
H discharge post boronization
Filled+Puffed atoms = 8.2*1020
Desorbed atoms = 6.9*1019
Flat top electron density versus total filled particles between two He-GDC performed to remove H from the wall
After Lithization a greater quantity of filled particles does not lead losing density control
Flat top electron density versus total filled particles between two He-GDC performed to remove H from the wall
After Lithization a greater quantity of filled particles does not lead losing density control
Increased wall gas adsorbing capacity
Increased wall gas adsorbing capacity
... on H2 recycling...
RFX-mod programme workshop 2011 - February 7- 9
• Different density radial profile
• Statistically it can be verified by the Peaking Factor:
PF = n(0)/<n>
• On RFX-mod PF is a function of plasma current and average density
• After the Lithization it is clearly higher than with a clean graphite first wall
0.6
0.8
1.0
1.2
1.4
0.5 1.0 1.5 2.0
CLi
Fit_Pn=1.5I
p
0.29n-0.36
1.0
1.5
2.0
2.5
3.0
3.5
0.0 0.2 0.4 0.6 0.8 1.0
C wallLi wall
r/a
... on density radial profile modification...
RFX-mod programme workshop 2011 - February 7- 9
•Measured cut-off layer position is averaged over similar density discharges
•In Li discharges cut-off layer position moves inward by about 1-2 cm
•Edge density with Thermal Helium Beam diagnostic
•After Lithization edge density is lower than on clean graphite
•It has been confirmed by the cut off layer position measurements of a fixed frequency reflectometer
reference
post Li
... on density at the edge ...
RFX-mod programme workshop 2011 - February 7- 9
Gas Puffing Imaging diagnostic raw signals: measurement of edge density fluctuations
After Li
Reference
• larger edge blobs
• less plasma volume occupied
• different turbulent particles transport at the edge
• larger edge blobs
• less plasma volume occupied
• different turbulent particles transport at the edge
to quantify the differencesto quantify the differencesAutocorrelation time Packing
fractionC
Li
C
Li
C
Li
C
Li
... on edge transport ...
RFX-mod programme workshop 2011 - February 7- 9
No difference on the energy confinement time with respect to the Graphite wall
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0.0 1.0 2.0 3.0 4.0 5.0
C wallLi wallB wall
<ne>*1019 (m-3)
Lower Hα emission at the edge at the same density with a Li
wall
Improvement of the particles confinement
time
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0.0 1.0 2.0 3.0 4.0 5.0
C wallLi wallB wall
<ne>*1019 (m-3)
... on global properties ...
RFX-mod programme workshop 2011 - February 7- 9
Why wall conditioning is needed/desired
First Wall conditioning in RFX-mod
Effects of present wall conditioning
techinques on plasma properties in RFX-
mod and in other Experiments
Open points and activities aimed at
improving the present wall conditionig
effectiveness
Conclusions
RFX-mod programme workshop 2011 - February 7- 9
Suppression of Oxygen contamiantion of plasma
this is the main effect on all experiments (TEXTOR, FTU, TORE SUPRA, DIII-D...)
Boron is a getter for Oxygen !
Suppression of contamination by wall material
In the case of high-Z wall materials, the advantage is the reduction of Zeff (Alcator C-Mod, Asdex-U with W divertor...)
In the case of carbon wall, boronization was seen to reduce the chemical sputtering of carbon due to Oxygen impurities (via
formation of carbon-dioxide)[Hino et al. Thin Solid Films 253 (1994) 518]
Effect of BORONIZATION
RFX-mod programme workshop 2011 - February 7- 9
Reduction in Hydrogen recycling
DIII-D: They observe that particle fueling from plasma facing surfaces is reduced after boronization. They speculate that B binds H more effectively than C
[Jackson et al., J. Nucl. Mat. 196-198 (1992), 236]
TEXTOR: yes[Winter et al., J. Nucl. Mat. 162-164 (1989),
713]
TFTR: no significant change in Hydrogenic recycling properties[Dylla, J. Nucl. Mat. 176-177 (1990), 337]
Effect of BORONIZATION
RFX-mod programme workshop 2011 - February 7- 9
TFTR Enhanced energy confinement time Reduction of the H recycling Peaking of plasma parameters Suppression of impurities
[Mansfield et al. 1996 Phys. Plasmas 3 1892]
NSTX Enhanced energy confinement time Increase in the total power radiated from the
plasma Reduction in the edge electron density
[Bell et al. 2009 PPCF 51 124054]
Effect of LITHIZATION
RFX-mod programme workshop 2011 - February 7- 9
FTU Similar to Boronization but with better results Peaking of electron density radial profile Strong pumping effect of Lithium modificates
the SOL Te
[Apicella et al. 2007 Journal of Nuclear Materials 363-365 1346]
TJ-II Improved density control by external puffing Suppression of impurities
[Tabarès et al. 2008 PPCF 50 124051]
Effect of LITHIZATION
RFX-mod programme workshop 2011 - February 7- 9
Why wall conditioning is needed/desired
First Wall conditioning in RFX-mod
Effects of present wall conditioning
techinques on plasma properties in RFX-
mod and in other Experiments
Open points and activities aimed at
improving the present wall conditionig
effectiveness
Conclusions
RFX-mod programme workshop 2011 - February 7- 9
DISUNIFORMITY
(non uniform deposition/interaction with the FW during treatment)
and
QUANTITY ?
(thickness of Boron and Lithium layer
deposited on the wall)
Are the (small) effects of the wall conditioning we see strongly affected by
RFX-mod programme workshop 2011 - February 7- 9
2D
9D
Glow Discharge disuniformity in toroidal direction
from ISIS measurements
Confirmed also by analyses of samples (SIMS at IENI institute)
RFX-mod programme workshop 2011 - February 7- 9
By adding GDC anodes we can improve the uniformity, or better, the number of peaks but a good uniform density can only be reached with an number of GDC anodes that cannot be proposed !
Tests with a toroidal field (in the order of some mT) applied during the Glow Discharge process will be done in the near future (next P17 campaign)
We could (hopefully) discover that only one anode is necessary to have a very good and uniform glow discharge ... In this case “Thanks toroidal filed ! ”
RFX-mod programme workshop 2011 - February 7- 9
If the GDC plant will be able to sustain the discharge with the toroidal field necessary to obtain a good uniformity the test will be done also during the boronizazion
RFX-mod programme workshop 2011 - February 7- 9
If possible,we could try to modify the position of our LLL (as evaporator or limiter in equatorial ports) or install new evaporator(s)
RFX-mod programme workshop 2011 - February 7- 9
CDU Proto-Liter evaporator
•100-200 mg/min evaporated lithium
•Cos Lithium distribution
•Horizontal set-up
... just for example ...
RFX-mod programme workshop 2011 - February 7- 9
An idea could be the centrifugal Lithium multi-pellet injector ...
RFX-mod programme workshop 2011 - February 7- 9
Just questions ...
Can we obtain the same plasma performances we obtained so far “simply” performing 5’He GDC shot by shot (or every two shots)?
What about to try with daily i.e. 20’H2 + 40’He to maintain
impurities at low values and a good wall absorbing capability far from boronization?
RFX-mod programme workshop 2011 - February 7- 9
if, at the end, to change the wall conditioning technique will not lead to a good improvement of the First Wall behaviour …
… try to change the First Wall !!