Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
Me
mb
er
of th
e H
elm
ho
ltz A
sso
cia
tio
n
D.Borodin | Institute of Energy Research – Plasma Physics | Association EURATOM – FZJ
Modelling of Beryllium Erosion/Deposition and
Local Transport at ITER
First Wall Blanket Modules Using the ERO code
D.Borodin1, A.Kirschner1, S.Carpentier-Chouchana2, R.Pitts2, C. Björkas3,
P.C.Stangeby4, J.D. Elder4, A.Galonska1, D.Matveev1, V.Philipps1, U.Samm1
1Institute of Energy and Climate Research - Plasma Physics, Forschungszentrum Jülich GmbH,
Association EURATOM-FZJ, Partner In the Trilateral Euregio Cluster, Jülich, Germany 2ITER organization, Science Division, Route de Vinon sur Verdon – 13115 St Paul Lez Durance – France
3EURATOM-Tekes, Department of Physics, P.O. Box 43, FI-00014 University of Helsinki, Finland 4University of Toronto Institute Aerospace Studies, Ontario, Canada M3H 5T6
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 2
ITER Be wall – BM erosion
• Blanket module (BM) shapes optimized for
heat loads (P.C.Stangeby)
Aim – predictive modelling of
ITER, including first wall life time
BM 15
BM11
FLFS close to 2nd separatrix =>
First PFC life time estimates assuming
limiter-like contact on outboard BM11
Be
+ low Z
- high erosion
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 3
(mmBe/h)
peak erosion
redeposition
toroidal
dist (m)
poloidal
dist (m)
(mmBe/h)peak erosion
toroidal
dist (m)
poloidal
dist (m)
High density case Low density case
<Yeff> ~ 7%, ~50% particles locally redeposited
Net peak erosion ~ 0.06 mm/h
PFC lifetime ~ 1500 shots
T-retention* ~ 0.083 gT/h for one module
~ 3 gT/h for 36 BM11-18
Limit ~ 1920 shots (assuming: 50:50 D:T plasma, maximum safety limit ~640g)
<Yeff> ~ 6%, ~10% particles locally redeposited
Net peak erosion ~0.0025 mm/h
PFC lifetime ~ 36 000 shots
T-retention* < 1.3 mgT/h for 36 BM11-18
* 2D estimation of (D+T)/Be = f(Tsurf, Eimp, ГD/ГBe) [PICSES-B scaling law, G. De Temmerman, R. Doerner]
2D Net erosion-redeposition patterns on BM11
LIM code results [S.Carpentier et.al, PSI-19,
J.Nucl.Mater. (2010), in press]
Important issue for ITER:
benchmark with ERO focusing on
life time
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 4
Background Plasma
EB
C, Be, …
CD4
Cx+,Bex+,…
CDz0,+
re-eroded/
reflected particles
PFC (substrate Be, C, W, Mo, …)
Input:
ne, Te,i,
geometry
ionisation, dissociation
friction (Fokker-Planck), thermal force
Lorentz force
cross-field diffusion
physical sputtering/reflection
chemical erosion (CD4)
(re-)erosion and (re-)deposition
HMM and SDTrimSP surface models
Local transport: Plasma-surface interaction:
3D MC impurity transport code ERO
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 5
ITER Be wall – plasma parameters
Aim – predictive modelling of
ITER first wall life time
Complications:
• Complex geometry e.g. leading
to shadowing
• Uncertainty in atomic and
surface data for Be
• Other uncertainties: enhanced
re-erosion, carbide and alloy
formation, Be-D molecules, etc.
high density case (HDC) BM11
LIM predicted life time due to transient
events is not a limiting factor – we
concentrate on steady state
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 6
Plasma parameters at limiter surface
Te = 10eV = const
Ti = 20eV = const
Te = 7.1eV = const
Ti = 18.6eV = const
High density case
(HDC)
Low density case
(LDC)
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 7
Shadowing patterns (LIM and ERO)
BM11, steady state BM15, start-up
shadowed
wetted shadowed
wetted
In shadowed areas we assume no BG erosion and re-deposition of intrinsic Be impurity
Modelling is in
progress, out of
the scope!
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 8
LIM and ERO overview
Code ERO LIM
Type Monte-Carlo impurity tracing (BG plasma import)
BM implementation Shape, shadowing, plasma parameter, etc. - SIMILAR
Geometry 3D 2D
Test-particle tracing resolving gyro-motion guiding centre
Intrinsic Be impurity concentration in D+ flux possible
Collisions with surface resolved angle and energy sheath potential
Multiple BM tiles periodic boundary particle "mirrors"
Many routines in ERO for ITER BM are “imported” from the LIM code.
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 9
Poloidal positions of toroidal profiles
Profile y=5mm
Profile y=-187mm
Profile y=-387mm
Maximal erosion
Background (BG) plasma erosion; BM11, HDC
For LDC the maximal erosion point is elsewhere . . .
[Be/cm-2]
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 10
-600 -400 -200 0 200 400 600
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
layer
evolu
tio
n [
mm
/h]
x (toroidal) [mm]
HDC, y=-187mm
LIM depos.
LIM BG eros.
LIM Be s-sputt.
LIM net eros.
ERO Be s-sputt.
ERO BG eros.
ERO net eros.
ERO depos.
Erosion/deposition profiles
Very good agreement between ERO and LIM
Deposition
Erosion by
background (BG)
plasma Layer
evolu
tion [
mm
/h]
BM11, „HDC‟: profile at y=-187mm
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 11
-600 -400 -200 0 200 400 600
-0.09
-0.08
-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0
layer
evolu
tio
n [
mm
/h]
x (toroidal) [mm]
HDC, y=-187mm
LIM total eros.
LIM Be s-sputt.
ERO total eros.
ERO Be s-sputt.
Erosion profiles
Self-sputtering in ERO is larger and more concentrated near the ridge
Be self-sputtering
Total erosion
Small influence on total erosion . . .
Layer
evolu
tion [
mm
/h]
BM11, „HDC‟: profile at y=-187mm
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 12
-600 -400 -200 0 200 400 600
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
layer
evolu
tio
n [
mm
/h]
x (toroidal) [mm]
HDC, y=-187mm
LIM depos.
LIM BG eros.
LIM Be s-sputt.
LIM net eros.
ERO Be s-sputt.
ERO BG eros.
ERO net eros.
ERO depos.
The net erosion in LIM and ERO is in a very good agreement
Life time limiting erosion:
1[cm]/0.05[mm/h] = 200h
Net erosion
Net erosion profiles Layer
evolu
tion [
mm
/h]
BM11, „HDC‟: profile at y=-187mm
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 13
-600 -400 -200 0 200 400 600
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
layer
evolu
tio
n [
mm
/h]
x (toroidal) [mm]
HDC, y=-187mm
LIM depos.
LIM BG eros.
LIM Be s-sputt.
LIM net eros.
ERO Be s-sputt.
ERO BG eros.
ERO net eros.
ERO depos.
ERO uses in this case angle averaged W.Eckstein 2002 data for
sputtering yields from LIM and ADAS ‟93‟ Be ionization . . .
Life time limiting erosion:
1[cm]/0.05[mm/h] = 200h
Erosion/deposition profiles Layer
evolu
tion [
mm
/h]
BM11, „HDC‟: profiles at y=-187mm
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 14
Deposition in LDC is very low (in diffrence to HDC)
Life time limiting erosion:
1[cm]/1.5*10-4[cm/h] = 6600h
Maximal
erosion for
BM11, LDC is
at y=-367mm!..
Erosion profiles
BM11, „LDC‟: profiles at y=-187mm Layer
evolu
tion [
mm
/h]
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 15
Deposition: low vs. high density plasma
Be is ionized close to surface . . . Large redeposition.
Be is ionized far away from surface . . . Small redeposition.
Density of Be+[cm-2] – High density case (HDC)
Density of Be+[cm-2] – low density case (LDC)
In both LIM and ERO deposition dependence on plasma parameters is feasible!
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 16
ERO vs. LIM benchmark summary
• ERO in agreement with LIM . . .
a) Shadowing patterns, plasma parameters, etc.
implemented in a similar way.
b) Erosion, deposition and finally net erosion profiles are
in good agreement for BM11 (steady state).
c) Plasma conditions dependence is in agreement.
• ERO vs. LIM disagreements
a) Self-sputtering in ERO is larger and more concentrated
near the ridge. May indicate some difference in Be
transport assumptions . . .
Benchmark with existing experiments
would be useful!
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 17
Be sputtering yields – extreme assumptions
Normal incidence! Angle dependence should be taken into the account!
Only the ‘calculated’ data are included!
1) “maximum” – static TRIM + MD
2) “minimum” – SDTrimSP with 50% of D
(reasonable limit)
Experimental data too much scattered!
1) Large deviations: no sense to analyse
shape of curves
2) Various effects are difficult to separate
Be by D+ sputtering Be by D+ sputtering
„ERO-min“
„ERO-max“
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 18
Be sputtering yields – self sputtering
Estimation based on calculated data as for
Be by D+ sputtering Angular part is essential!
For following BM simulations ERO uses Eckstein 2007 fit.
Normal incidence Angular dependence
„ERO-max“ „ERO-min“
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 19
Sputtering by BG plasma (and intrinsic Be)
- “integration” produces effective sputter yields -
η (“B-angle”)
B=4.8T
Calculation of sputtering yield according to Eckstein’s fit 2007 for Y(E, α), using
angle and energy distribution as calculated by ERO (including gyration and sheath)
Injection of D+ or Be+ with Maxwell
energy around Ti and uniform initial
angle distribution
stagnation
α( angle of incidence)
ERO pre-calculation
normal
0 10 20 30 40 50 60 70 80 900
50
100
150
Angle of incidence [o]
En
erg
y [e
V]
D+ ions
"B-angle" = 1o
"B-angle" = 5o
"B-angle" = 20o
"B-angle" = 35o
0 10 20 30 40 50 60 70 80 900
50
100
150
200
250
300
Angle of incidence [o]
En
erg
y [e
V]
Be+, Be2+ ions
"B-angle" = 1o
"B-angle" = 5o
"B-angle" = 20o
"B-angle" = 35o
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 20
Be by D+ sputtering yields – assumptions
LIM data is ‘Eck2002’ averaged by incidence angle assuming uniform distribution
Incidence energy dependence Integrated yields
• HDC conditions (Te=10eV, Ti = 20eV)
BM11,HDC X
ELIM = 3Te+2Ti =70eV
Averaged by angle!
Normal
incidence
„Eck2002‟ is
[W.Eckstein,
IPP-report 2002]
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 21
Be self-sputtering yields – assumptions
Averaged by angle! BM11,HDC X
ELIM (Be+)= Z*3Te+2Ti =70eV
Normal
incidence
LIM data is ‘Eck2002’ averaged by incidence angle assuming uniform distribution
„Eck2002‟ is
[W.Eckstein,
IPP-report 2002]
Incidence energy dependence Integrated yields
• HDC conditions (Te=10eV, Ti = 20eV)
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 22
-600 -400 -200 0 200 400 600
-0.08
-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0
0.01
layer
evolu
tio
n [
mm
/h]
x (toroidal) [mm]
ERO-max (0.081mm/h)
ERO-min (0.021mm/h)
Eck2002(LIM) (0.062mm/h)
Sputter yields assumptions and net erosion
BM11, „HDC‟: net erosion (deposition) profile at y=-187mm
In most pessimistic case life time about 30% less than in earlier LIM predictions
LIM predictions
4200 ITER shots
1500 ITER shots
1100 ITER shots
LIM predictions Layer
evolu
tion [
mm
/h]
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 23
-600 -400 -200 0 200 400 600
-0.08
-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0
0.01
0.02
layer
evolu
tio
n [
mm
/h]
x (toroidal) [mm]
ERO-max (0.081mm/h)
ERO-max, 1% Be (BG) (0.074mm/h)
ERO-max, 3% Be (BG) (0.063mm/h)
ERO-max, 5% Be (BG) (0.051mm/h)
Intrinsic Be impurity influence on net erosion
BM11, „HDC‟: net erosion (deposition) profile at y=-187mm
Deposition of Be impurity from plasma dominates over additional Be self-sputtering
0 1 2 3 4 50
0.5
1
1.5
2
2.5
Intrinsic Be concentration
Su
rfa
ce e
volu
tio
n [
a.u
.]
BM11. HDC. 'ERO-max'
Total erosion
Deposition
Net erosion
Additional Be
self-sputtering
Deposition from
background
La
yer
evolu
tion
[m
m/h
]
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 24
Intrinsic Be impurity and sputter yields effect
BM11, „HDC‟: net erosion (deposition) profile at y=-187mm
Influence of enhanced Be re-erosion (typical ERO assumption) of is not yet studied!
LIM predictions
Sputering yield and intrinsic Be assumptions determine the outcome to a large extent!
Deposition (ERO-min +3%Be)
Layer
evolu
tion [
mm
/h]
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 25
Summary
1) ITER Organization requested a benchmark of LIM erosion-re-
deposition simulations made for steady state plasma exposure
on shaped, Be-armoured FW panels
ERO code
2) Using same input plasma parameters, shadowing geometry,
Be sputtering yields, ERO (3D) in excellent agreement with LIM
(2D)
LIM low limit for FW panel erosion lifetime reproduced
by ERO (~1500 ITER reference QDT = 10 discharges)
3) Large range of erosion lifetime dominated by uncertainties in
input plasma parameters and Be sputtering yields
inclusion of different sputtering models in ERO yields
variation of lower limit from 1100 4200 reference discharges
for BM11, HDC. This variation can also be influenced by other
assumptions e.g. intrinsic Be impurity.
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 26
To make progress now we require:
A) Experimental benchmarking of the impurity transport
codes in ITER-relevant geometry
IO planning dedicated benchmark on EAST (MAPES
manipulator with shaped tiles similar to ITER FW panel
profile
JET ILW the ideal location for a dedicated experiment
looking at erosion of Be tiles (e.g. upper part of the
vacuum vessel in the secondary X-point region)
Outlook – A
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 27
PISCES-B
Perfect for Be sputtering yields
benchmark
1. Spectroscopy
2. Target weight loss
3. Witness plate
The ERO was earlier applied for
modelling of PISCES-B
Outlook – B
B) Improvement in sputtering
yield uncertainty
model testing in PISCES-B
D.Borodin | PFMC-13, Rosenheim | Forschungszentrum Jülich 12.05.2011 No 28
The End