Copyright © 2014
SCK•CEN
Interaction of high flux plasma with RAFM steels:
experimental and computational assessment
(Eurofer 97 and advanced grades improved by
thermomechanical treatment)
Dmitry Terentyev, Lorenzo Malerba, Athina Puype,
Anastasia Bakaeva
SCK-CEN, Belgian Nuclear Research Centre
Copyright © 2014
SCK•CEN
SCK•CEN in short
Cradle of Belgian nuclear research, applications and energy
development in Belgium
Major international player in the field of nuclear R&D
~700 staff, >50% with academic degree + 70 PhD students
Annual turnover: 140 M€
45% government support
55% contract work
Copyright © 2014
SCK•CEN
Outline
Part I
Non-equilibrium D trapping under high flux plasma exposure:
lessons learned by studying Tungsten
Computational model to consider Non-equilibrium trapping
and its connection to diffusion limited trapping
Essential information from atomistic scale
Part II
Standard and improved high-Cr grades
Available improved 9-Cr batches
Preliminary tests including high flux plasma exposure
Work plan
Copyright © 2014
SCK•CEN
Part I
Part I
Non-equilibrium D trapping under high flux plasma exposure:
lessons learned by studying Tungsten
Computational model to consider Non-equilibrium trapping and its
connection to diffusion limited trapping
Essential information from atomistic scale
Copyright © 2014
SCK•CEN
Experimental data
Typical penetration depth ~
several µm
Intensive blistering with
pronounced by-modal size
distribution
TDS spectra: 3 release stages
Intensity of the high T stage
correlates with blister pattern
400 600 800 1000 12000
1
2
3
4
D2 r
ele
ase
, 10
13 s
-1
Temperature, K
W
5*1025
m-2
(1 shot)
400 600 800 1000 12000
1
2
3
4
D2 r
ele
ase
, 10
13 s
-1
Temperature, K
W
1027
m-2
(20 shots)
[1] M. Balden, S. Lindig, A. Manhard, J.-H. You, Journal of Nuclear Materials 414 (2011) 69-72.
[1]
Copyright © 2014
SCK•CEN
Depth distribution & sub-surface trapping
Tanabe et.al.
Copyright © 2014
SCK•CEN
Depth distribution
Three zones defining distribution
~nm
~ µm
~ mm
Tanabe et.al.
Copyright © 2014
SCK•CEN
High heat flux plasma exposures
PILOT-PSI (NL, DIFFER)
Stationary Loads & ELM-like pulses
B field = 0.4-1.6 T
Water cooling 20 C
(passive control of T surface)
Flux 1023-1024 D/m2
Exposure time (sec)
Temperature at the centre of sample
Copyright © 2014
SCK•CEN
High flux plasma changes sub-surface
Compressive stresses in sub-surface region generate:
Dense dislocation network
Limited within 10-15 µm
Poly-crystalline (ITER-specification)
Copyright © 2014
SCK•CEN
High flux plasma changes sub-surface
Compressive stresses in sub-surface region generate:
Dense dislocation network
Limited within 10-15 µm
Poly-crystalline (ITER-specification)
Recrystallized at 1600C (1h)
Copyright © 2014
SCK•CEN
High flux plasma changes sub-surface
Compressive stresses in sub-surface region generate:
Dense dislocation network
Limited within 10-15 µm
Single crystal (110 orientation)
Copyright © 2014
SCK•CEN
Experimental data
SEM edge on image
TEM image: 100-200 nm of top surface
Dislocation density 1012-1014 m-2
Random grain
boundaries
100
µm
RLAGB ~ µm
~ µ
m
R ~ 100 nм
Low angle grain
boundaries (LAGB
Dislocations
TEM sample
NRA range
Mean spacing 100±20 nm
Copyright © 2014
SCK•CEN
Experimental data
Select large grain (pre-characterized) material:
Recrystallized W
Perform controlled deformation:
Tensile test on macro sample
Sub-threshold plasma exposure:
PILOT PSI, T=470K, flux ~1025 D2/m2/sec
Exposure flux: 10, 70, 500 sec i.e. up to 5×1027
PIE sequence: SEM; TEM; TDS; SEM
0 5 10 15 20 25 30 35 400
50
100
150
200
250
P YS =1.93 (kN)
P TS =4.40 (kN)
P FS =4.32 (kN)
delta u=16.91 (mm)
delta t =17.04 (mm)
sigma YS=111 (MPa)
sigma TS=253 (MPa)
sigma FS=310 (MPa)
epsilon u=34 (%)
epsilon t =34 (%)S
tre
ss (
MP
a)
Strain (%)
Sample 1, tested on 20140127,
600°C (air atmosphere), 0.2 mm/min
0 1 2 3 4 5 6 7 8 9 100
100
200
300
400
500
0
100
200
300
400
500
Dis
loca
tio
n d
en
sity 1
01
2 m
-2
Str
ess (
MP
a)
Strain %
Density
Stress
L=0.2 µm ; k=1
6 samples 10 mm
( ) ( )d d p pM µb
dρ/dεp = M( 1/bL − kρ)
Strain hardening model by
Meckin and Cocks [1]
[1] H. Mecking, U.F. Kocks, Acta Metal. 29 (1981) 1865; [2] H. Sheng, G. Van Oost, et.al., J. Nucl. Mat. 444 (2014) 214.
Copyright © 2014
SCK•CEN
deformed recrystallized
deformed recrystallized
Exposure 70 sec
8/15
Copyright © 2014
SCK•CEN
Experimental data
Reference
Deformed
300 400 500 600 700 800 900 10000.0
3.0x1012
6.0x1012
9.0x1012
1.2x1013
1.5x1013
1.8x1013
D2 r
ele
ase
(s
-1)
Temperature (K)
Dose: 1E26
1/m2
Deformed
Reference
Absence of the high temperature peak !
… suppression or burst of blisters ?
Possible reasons:
- Shallower nucleation depth (burst)
- Internal stress (suppression of bubble
growth by loop punching)
To be associated with dislocations ?
Copyright © 2014
SCK•CEN
Experimental data
0
4
8
12
REF-10SEC
P1
P2
P2
P1
P3
400 600 800 1000 1200
0
4
8
12
D2
rele
ase
, 1
0-1
2, s
-1D
2 re
lea
se
, 1
0-1
2, s
-1
P3
D2
rele
ase
, 1
0-1
2, s
-1
Temperature (K)
REF-490SEC
0
3
6
9
12
15
REF-70SEC
P1
P2
P3
0
4
8
12
P3
P3
P3
P2
P2
P2
P1
P1
D2
rele
ase
, 1
01
2, s
-1
DEF-10SEC
P1
400 600 800 1000 1200
0
4
8
12
D2 r
ele
ase
, 1
01
2, s
-1
Temperature, K
DEF-490SEC
0
3
6
9
12
15
D
2 r
ele
ase
, 1
01
2, s
-1
DEF-70SEC
Copyright © 2014
SCK•CEN
Experimental data
Legend:
10 100 10000.0
2.5
5.0
7.5
10.0
0.0
2.5
5.0
7.5
10.0
Pea
k inte
nsity, 10
12, s
-1
Exposure time (sec)
Reference
Deformed
Stage I Stage II Stage III
I II III
reference material
plastically deformed material
Plastic deformation promotes nucleation of H bubbles
Copyright © 2014
SCK•CEN
Experimental data
Our understanding of sub-surface trapping:
Establishment of steady-state D profile
Compensation of compressive stresses by
plastic deformation
Dislocation network acts as trapping &
nucleation source
Nano-metric bubbles get to steady-state
Pathway to validate/rationalize the
hypothesis
Characterize trapping & nucleation on
dislocations: atomistic sim.
Pass atomistic data to Mean Field (kinetic rate
theory) for non-equilibrium nucleation
Combine with classical MF (diffusion limited)
Ld
surface
D Su
rface (
nm
)
Bu
lk (
mm
)
Copyright © 2014
SCK•CEN
Hypothesis about trapping on dislocations
3/15
dislocations
H
H
H
?
Trapping at dislocations
and low angle GBs
H clustering and migration along
dislocations, trapping at random
GBs
H-dislocation interaction will
determine retention at low
temperature exposure
Experimental and theoretical ‘facts’:
H-H do not form clusters: no self-trapping like for He
Binding (H-H)=0.02; (H2-H)=0.01, (H3-H)=0.02; (H4-H)=0.03 eV [1]
Migration energy = 0.4 eV
H atom covers 30 µm within 1 ms: all permeated H immediately get
trapped
Typical µ-structure of commercial W grades:
Dislocations, sub-grains, random grains
Porosity, Impurity
Possible mechanisms of bubble nucleation without vacancies:
[1] K. Heinola, T. Ahlgren et.al., Physical Review B 82 (2010) 094102.
Copyright © 2014
SCK•CEN
Interaction of Hydrogen with screw dislocation
4/15
<111>
<110>
<112>
<111> view: Differential displacements map <111>
add H
SD
SD
H
H ΔE = -0.6 eV
ΔE ?
W W
W
a b c
Migrations:
a – in core
b – in-out
c – out core
[1] D. Terentyev et.al. Nuclear Fusion Lett. 54 (2014) 042002.
Charge density map
Copyright © 2014
SCK•CEN
Clustering of Hydrogen on screw dislocation
5/15
0.186
0.891
0.163
1.615
0.152
1.637
0.146
1.681
a)
b)
c)
d)
Size of HN
cluster
SD
length=b
H + SD -0.53
H + (H-SD) -0.59
H + (H2-SD) -0.62
H + (H3-SD) -0.39
H + (H4-SD) -0.36
Screw dislocation can trap several hydrogen atoms simultaneously, but incremental binding
energy decreases with number of trapped H atoms (from 0.6 – down to 0.25 eV).
Charge density maps [e-/Å3] Binding energy [eV]
[1] D. Terentyev et.al. Nuclear Fusion Lett. 54 (2014) 042002.
Pure W One H added
Two H added Three H added
Copyright © 2014
SCK•CEN
Nucleation of stable Hydrogen cluster on dislocation
6/15
A: Dislocation + cluster HN
B: dislocation + (Jog—HN) + Jog+
SD
SD
SD
SD
J-
J+
W
W
W
W
+
+
0 2 4 6 8 10-2
0
2
4
6
8
10
(A-B
), e
V
Number of hydrogen atoms
Z = 1B
Z = 2B
Loop punching: growth of bubbles due to excess pressure via the emission of dislocation loops
Jog punching: formation of vacancy jog to accommodate HN cluster + emission of interstitial jog
No attraction of H in
bulk – no way to form
critical concentration.
Clustering at
dislocation core leads
to jog emission at
NH=8
Classical loop punching mechanism
Copyright © 2014
SCK•CEN
Description of the model
H source
defined from
steady state
solution
H im
pla
nta
tio
n
λd
H … …
~ 10-100 μm; time ~ 1000 seconds
Fu
ll-s
cale
P
hysi
cs
of
mo
del
Copyright © 2014
SCK•CEN
Computational model results & prototype tool
Estimation of experimental conditions for H bubble
formation and surface blistering
1E22 1E23 1E24 1E25 1E26
400
500
600
700
800
900
1000
1100
1200
Tem
pera
ture
(K
)
Dislocation-mediated retention
Dislocation density = 0.5x1012
Incident flux (m-2s
-1)
Loop punching (blistering)
Xu, H.Y. JNM 2014. 447(1–3): p. 22-27.
Zayachuk,Y. et al Nucl. Fus. 2013. 53(1): p. 013013
hydrogen atom
interstitial jog
dislocation line
vacancy jog
Steady state solution for
diffusion in a field of
traps
𝐶 𝑥 = 𝐶0 exp −𝑥𝑘𝐷
𝑘𝐷 = 𝜌 for dislocations
𝑘𝐺𝐵~𝑑 for grain
boundaries
Copyright © 2014
SCK•CEN
Model of H retention
Two zoned retention
Dynamic retention in sub
surface region
Trapping on natural defects
in the bulk area
Depth
Co
nce
ntr
ati
on
C s
ub
su
rface
C bulk
10 μm
10 - 1000 μm
Copyright © 2014
SCK•CEN
Release of trapped H
Release from bulk and
diffusion to the surface
Concentration of H is
defined by trapping and
growth at low angle
grain boundaries
Binding energy of 2.0 eV
Depth
Co
nce
ntr
ati
on
CS =
1.1
*10
-5
CB = 3.92*10-7
Copyright © 2014
SCK•CEN
Part II
Part II
Standard and improved high-Cr grades
Available improved 9-Cr batches
Preliminary tests including high flux plasma exposure
Copyright © 2014
SCK•CEN
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
0,1 1 10 100 1000
Radiation effects in structural materials
Dose received (dpa)
Ho
mo
log
ical
Tem
pera
ture
(T
/TM)
> 0.1 dpa, <0.35 TM
Radiation hardening and embrittlement
>> 10 dpa, T>0. 5 TM
If He>100appm
He embrittlement at GB
(intergranular fracture)
>10 dpa, 0.3TM<T<0.6TM
Phase instabilities from
radiation-induced
segregation and
precipitation
Volumetric void swelling
(dimensional instability)
>10 dpa, 0.35TM<T<0.45TM
Irradiation creep
Copyright © 2014
SCK•CEN
Standard and improved high-Cr grades for Fusion Appl.
One principle problem of 9-Cr steels: low temperature
embrittlement
Second one: creep, limiting application up to 650C
Copyright © 2014
SCK•CEN
Standard and improved high-Cr grades for Fusion Appl.
Plastic flow instability
Dynamic absorption of defects
Drastic softening of matrix
Copyright © 2014
SCK•CEN
Development of new improved Eurofer grades
Composition C Mn Si Cr W N B Zr V Ta
Target 1 0.020 0.400 0.030 8.500 2.500 0.070 0.500 0.500
Target 2 0.030 0.400 0.030 9.000 1.000 0.070 0.500 0.400
Target 3 0.030 0.400 0.030 9.000 1.000 0.070 0.250 0.150
Target 4 0.050 0.400 0.030 9.000 1.000 0.020 0.300 0.200
Target 5 0.050 0.400 0.030 9.000 1.000 0.020 0.005 0.300 0.200
Target 6 0.050 0.400 0.030 9.000 1.000 0.040 0.005 0.300 0.200
Target 7 0.070 0.400 0.030 9.000 2.000 0.070 0.300 0.200
Target 8 0.030 0.400 0.030 9.000 2.000 0.070 0.300 0.200
Target 9 0.050 0.400 0.030 9.000 2.000 0.070 0.300 0.200
Target 10 0.020 9.000 1.000 0.020 0.250 0.150
Target 11 0.020 8.000 1.000 0.020 0.250 0.150
Selection of compositions (casting done by OCAS)
Comp. for investigation on Low Temperature EUROFER
Copyright © 2014
SCK•CEN
How to reduce DBTT and enhance creep resistance
Modify precipitate distribution
Change ratio MC to MX
Modify H-treatment
Modify Composition
V and Nb:
Stabilize grains
Reduce grain coarsening
Increase ductility
9Cr remains optimum
Less Cr – poor corrosion resist
More Cr – hardening (α’) and
δ-ferrite
Copyright © 2014
SCK•CEN
Advanced Thermo-Mechanical Treatment: TMT
Wt% C Cr Mn N2 Ta V W
0.09Ta 0.11 9.13 0.40 0.011 0.09 0.23 1.04
0.11Ta 0.11 9.24 0.39 0.011 0.11 0.23 1.07
Experimental Procedure Composition of the 9wt% Cr RAFM steel batches 2014
Copyright © 2014
SCK•CEN
Advanced Thermo-Mechanical Treatment: TMT
1
2
0.09Ta
0.11Ta
Copyright © 2014
SCK•CEN
Effect of Heat Treatment
0
5
10
15
20
25
30
35
40
45
50
0.09wt%Ta_LFRT 0.11wt%Ta_LFRT
DA
V,E
q [
mic
ron
]
Effect of Ta content on grain size
TMT+HT
TMT
PAG
Block
Substructure
PAG
Block
Substructure
M23C6
0.5 µm M23C6
0.5 µm
Copyright © 2014
SCK•CEN
Precipitation of Vanadium and Tantalum
-10
10
30
50
70
90
110
130
150
170
02468
1012141618202224
850 900 950 1000 1050 1100
DA
V,
Eq
[n
m]
DA
V,
Eq
[m
icro
ns]
Annealing temperature [°C]
Martensite structure and carbide
measurements
Carbides
Block size
00.010.020.030.040.050.060.070.080.09
0.10.11
TMT Anneal Q&T
wt%V & wt%Ta precipitated
Wt% Cr Ta V
0.09Ta 9.3 0.086 0.22
Fraction of elements that precipitate
at different stages of Q&T treatment
Copyright © 2014
SCK•CEN
Mechanical properties of improved grades
300
350
400
450
500
550
600
650
700
-100 0 100 200 300 400 500 600
σU
TS
[M
Pa]
Temperature [°C]
Ultimate tensile strength
880°C
920°C
980°C
1050°C
0
2
4
6
8
10
-200 -160 -120 -80 -40 0 40 80
Dia
l en
erg
y [
J]
Temperature [°C]
I196C - 880°C - KLST specimens
DBTT
Q&T: 880°C tested at -140°C
Inte
rmed
iate
level
Q&T: 880°C tested at -120°C
Up
per
level
Copyright © 2014
SCK•CEN
Preliminary results
400 500 600 700 800
1E10
1E11
1E12
1E13
1E14
D2
re
lea
se
(1
/s/c
m2)
Temperature (K)
Exp 430-450 K
T91 (T1) ref
Eurofer 97(e3) ref
E97 non-exposed
High Flux D plasma exposure at Pilot PSI (NL)
Surface temperature: 430-450K
Flux ~1024 D/m2
Fluence ~5×1025 D/m2
Reference measurements
T91; E97
Copyright © 2014
SCK•CEN
Preliminary results
Exposure at 430-450K: Blisters and individual grains
NRA/TEM analysis to be done to confirm sub-surface retention & µ-structure
modification
Copyright © 2014
SCK•CEN
Preliminary results
Exposure at 430-450K: reference vs. advanced grade
Chemical analysis to be done to confirm observations of Carbides at surface
Copyright © 2014
SCK•CEN
Preliminary results
400 600 800 1000 1200
0.0
2.0x1011
4.0x1011
6.0x1011
8.0x1011
1.0x1012
1.2x1012
1.4x1012
1.6x1012
D2 r
ete
ntion (
1/s
/m2)
Temperature (K)
E97 exp. 960K
E97 non exposed
High Flux D plasma exposure at Pilot PSI (NL)
Surface temperature: 950-970 K
Flux ~1024 D/m2
Fluence ~5×1025 D/m2
Reference measurements
E97;