quantification of the ααα(o) and prior-ββββ phase ... · contents of ααα(o) and...
TRANSCRIPT
1J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Quantification of the αααα(O) and Prior-ββββ phase fractions
and their oxygen contents
in high temperature (HT) oxidised Zr alloys (Zy-4, M5™)
Outlines : to be able to quantify the local « ECR » of HT oxidised cladding
tubes (i.e. to characterize locally the resultant clad microstructure after a
LOCA test leading to a non homogeneous oxidation of the cladding tube,
for examples : due to azimuthal and/or axial thermal gradients along the
tube, due to secondary oxidation in the vicinity of a failure, etc…)
• illustration of application of the methodology to study the influence of :
(1) Conventional Furnace Heating (FH) vs. Induction Heating (IH)
(2) Single Face (SF) vs. Double Face (DF) oxidation mode
on M5™ oxidised at 1000°C (before/after «break-away» occurrence)
J.C. Brachet, L. Portier, D. Hamon, Ph. Trouslard, S. Urvoy, V. Rabeau
CEA - Saclay (France)
+ A. Lesbros (EDF-Septen) and J.P. Mardon (FRA-ANP)
2J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Parameters influencing the αααα(O) and Prior-ββββ
phases thickness and morphology of HT oxidised
clads and their local oxygen contents :
(1) Temperature transient conditions : that is,
heating/cooling rate, time & temperature of the HT
oxidation (as an example, the increase of the
oxidation temperature => increase of the oxygen
solubility in Prior-β phase) ;
(2) Alloy types (Zircaloy’s, Niobium containing,…);
(3) Hydriding (due to nominal in-service oxidation)
(4) Other…
3J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
The experimental methodology has been developed and applied to
HT oxidised Zy-4 and M5™ cladding tubes :
(1) Quantification of the α(O) and Prior-β phases thickness by
Scanning Electron Microscopy (SEM) in back-scattered
electron imaging mode coupled with image analysis
procedure
=> illustration on M5™ oxidised at 1000°C
– influence of the heating mode (furnace vs. induction)
and Single-Face vs. Double-Face oxidation modes
(2) Measurements of the local oxygen content by Electron Probe
Micro Analysis (EPMA) and thanks to the use of Nuclear
Reaction Analysis (NRA) to take into account the influence
of « surface contamination » (nanometric ZrO2 layer formed
on the sample surfaces during the preparation)
+ Vickers microhardness (50-100 g) measurements
4J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
M5 après essai d’oxydation
à 1200°C pendant 1422s
dans Dézirox
Micrographie optique Micrographie électronique (rétro-diffusés)
Image de départ en niveau de gris obtenue en microscopie
électronique Image binaire obtenue après traitement de la micrographie
MEB (seuillage, érosion, dilation, fermeture, ouverture, ...)
Ex-ββββ
αααα-Zr(O) αααα-Zr(O)
Ex-ββββ
Ex-ββββ
αααα-Zr(O) après
binarisation
(1) Experimental procedure for the quantification of the α(O) and
Prior-β phase thickness – illustration on the M5TM alloy
Optical micrograph
(polarised light)
Electron (SEM) micrograph
Electron (SEM) micrograph
(backscattered electron mode)
Binarisation (Image Analysis)
M5TM
SF oxidation
for 1422 s.
at 1200°C
α(O) after image
binarisation
Prior-β
Prior-βPrior-β
α(O) α(O)
5J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
N°3
0
10
20
30
40
50
60
70
80
90
100
110
0 50 100 150 200 250 300 350 400 450 500
epaisseur en µm
% phase
Clad thickness (µm)
α(O) %
Image analysis measurement of
the α(O) fraction evolution
within the thickness of a SF
oxidised M5™ clad
ZrO2/αααα(O)
interface
Internal clad
surface
Continuous
αααα(O) layer
αααα(O) +
Prior-ββββContinuous
Prior-ββββ layer
8 fields per
tube section
are considered
6J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Evolution of the αααα(O) and Prior-ββββ phase phase thickness
as a function of the oxidation time at 1000°C of M5TM
{Single Face (SF) – Furnace Heating (FH)} - « Dezirox » facility
vs. {Double Face (DF) – Induction Heating (IH)} – « Cinog » Facility
0 20 40 60 80 100 1200
5
10
15
20
25Measurements performed
for 4 typical oxidation times
M5™
SF oxidised
at 1000°C
7J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
0 20 40 60 80 100 120 1400
50
100
150 DF - IH (Ext. + Int.)
SF - FH (Ext.)
Parabolic growth kinetics (diffusion controlled)
5-6 min
21-24 min
55 min
3h45min-4hbeginning of "breakaway" (5000s)
1/2
M5™ SF & DF
oxidised at 1000°C
8J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90 100
Relative distance (%) / clad metal thickness
% of Alpha(O)
SF/FH - 3270s
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90 100
Relative distance (%) / clad metal thickness
% of Alpha(O)
DF/IH - 3270s
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90 100
Relative distance (%) / clad metal thickness
% of Alpha(O
)
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90 100
Relative distance (%) / clad metal thickness
% of Alpha(O)
Direct comparison of the Alpha(O) phase fraction spatial distributions
within the thickness of M5™ oxidised tubes for the both oxidation modes
9J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Direct comparison of the Alpha(O) phase fraction spatial distributions
within the thickness of M5™ oxidised tubes for the both oxidation modes
SF/FH - 14550s
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90 100
Relative distance (%) / clad metal thickness
% of Alpha(O
)
DF/IH - 13500s
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90 100
Relative distance (%) / clad metal thickness
% of Alpha(O)
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90 100
Relative distance (%) / clad metal thickness
% of Alpha(O
)
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90 100
Relative distance (%) / clad metal thickness
% of Alpha(O)
10J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
No significant influence on M5TM oxidised at 1000°C :
of the heating mode (FH vs. IH)
nor of the oxidation mode (SF vs. DF)
on the αααα(O) thickness evolution vs. oxidation time
- and thus, no significant influence on the « Breakaway » (this last
point has been also confirmed by post-test hydrogen analysis and by
the observation of the ZrO2 morphology evolution, which gave similar
results for the two heating-oxidation conditions applied in CEA)
- the α(O) growth kinetics seems to be parabolic even after
breakaway occurrence (up to 4 hours of oxidation at 1000°C)
Partial conclusions
11J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
(2) Measurements of the local oxygen content by :
• Electron Probe Micro Analysis (EPMA)
=> CAMECA « Microbeam », 15kV, ~10 nA
=> Volume analysed ~1µm3
• Nuclear Reaction Analysis (NRA) :
⇒ HVEE AN-2000 Van de Graaff
(2A & 2F) positive in accelerator [*]
=> nuclear reaction 16O(d,p1)17O
=> 900 keV deuteron beam
=> incident current = 150 nA (take angle = 150°)
=> spot diameter ~ 0.5 mm - resolution in depth ~ 1µm
[*] L. Beck, A. de Château-Thierry, J.-P. Frontier, S. Pellegrino, Ph.Trouslard,
Twenty five years of IBA teaching experience at the National Inst. For
Nuclear Science and Technology, France, Nucl. Instr. and Meth. B (in press)
12J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Problem with EPMA measurements :
Overestimation of the oxygen content due to surface
contamination (contribution of the surface oxide layer)
Incident
electron
beam
X-Ray (WDS analysis of oxygen)
~1µm3
«Native» nanometric oxyde
layer on the sample surface
Polished sample
13J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Solutions to the EPMA oxygen overestimation problem :
(1) Systematic use of calibration samples consisting of a
Zy-4 or a Zr-1%Nb matrix with a very low oxygen
content (100-200 wt-ppm)
=> assumption that the surface nanometric oxide layer has
the same contribution on the global oxygen WDS analysis
on both low oxygen and highly oxidised Zr alloys (?)
(2) Use of NRA measurements which allows to
deconvolute the contribution of the oxygem atoms
near the sample surface from that of the oxygen atoms
in the bulk
14J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Sample preparation for NRA measurements of the oxygen
content within the Prior-Beta phase internal layer of High
Temperature SF oxidised cladding tubes :
before polish after polish
Internal cladding
tube surface
Prior-ββββ
αααα (O)
ZrO2
NRA spot
~ 200µm2
15J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division16O(d,p1)
17O
0
20
40
60
80
100
120
140
160
180
200
0 200 400 600 800 1000 1200 1400
Energie [keV]
Contenu
Expérimental
Simulé
Oxygène en profondeur
Oxygène en surface
Condutions expérimentales
Ed 900 keV
Ξη��&��� 150°
i 250 nA
q 400 µC12 µm Mylar
O in the bulk
O in the
sample surface
Typical NRA oxygen
spectrum obtained within the
Prior-ββββ phase internal layer
of a HT SF oxidised sample
=> It is possible to deconvolute the contribution of the oxygen atoms at the sample surface
c
o
u
n
t
s
Energy (keV)
experimental
simulation
16J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Comparison between oxygen content measurements performed
by EPMA and NRA on the internal Prior-β phase layer
0 0,2 0,4 0,6 0,8 1 1,20
0,2
0,4
0,6
0,8
1
1,2
17J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Microhardness evolution of the internal Prior-Beta phase layer
vs. oxygen content measured by NRA
(SF oxidations performed for different times at 1000-1200°C)
⇒Scattering of the µhardness
values (especially for Zy-4
alloy) - probably due to the
coarse grain sizes and local
crystallographic texture effects
+ α(O) phase penetrations within
the internal Prior-β phase layer
⇒ µhardness measurements
can only give semi-quantitative
information on the local
oxygen concentration ; they
should depend also on the
alloy, on the quenching rate, …
0 0,2 0,4 0,6 0,8 1 1,2200
250
300
350
400
450
500
18J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
CONCLUSIONS
• Because of the strong influence on quenching behavior and on the post-quench
mechanical properties of HT oxidised Zr base cladding tubes, a methodology has
been developped and applied to quantify the fractions and the local oxygen
contents of αααα(O) and prior-ββββ phases in the thickness of HT oxidised tubes.
• The methodology is able to quantify locally the ECR and thus can be applied on
cladding tubes wich have been subjected to a non homogeneous HT oxydation
=> validation of the methodology by comparing (re)calculated ECR v.s. «realistic
ECR» derived from weight measurements on homogeneously HT oxidised tubes is
in progress
• The methodology can be transposed to irradiated samples (thanks to the avaibility
of SEM, EPMA & NRA in CEA-Saclay hot cells)
• An application of the methodology to M5™ alloy oxidised at 1000°C has shown
no significant influences of the heating (FH vs. IH) and of the oxidation (SF vs.
DF) modes on the αααα(o)/prior-ββββ thickness evolution and thus on the breakaway
occurrence => check the consistency between the two main HT-oxidation
facilities used in CEA on non-irradiated clads (“Dezirox” & “Cinog” facilities)
1J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Zy-4:
(3)
(2)
(1)
(1) Brittle failure of α(O)
(2) Incursions of brittle
α(O) in ex-β zone
(3) Ex-β zone with flatdimples indicating a quite
low residual ductility
No ZrO2 (delamination)EPMA X-ray maps of oxygen (+ Fe, Cr) and
typical fractographs of an impact tested Zy-4
specimen failed at R. T. (Single Face oxidation
for ~3000s at 1100°C)
(from Halden IAEA meeting, J.C. Brachet, 2001);
Zy-4:
O-Kα
Fe-Kα Cr-Kα
ZrO2 αααα(O) Ex-ββββ
=> Illustration of the strong influence
of the α (O) and Prior-β phase
characteristics (morphology : i.e.
spatial repartition and thickness,
oxygen contents…) on the post-
quench mechanical properties
2J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
Electron Probe X-ray maps -ECR=17%
Zy-4:
M5TM:
O-Kα
Fe-Kα Cr-Kα
Nb-KαO-Kα
ZrO2 αααα(O) Ex-ββββ
Illustration of the influence of alloy type on the α/Prior-β phases :
EPMA X-ray maps of Zy-4 and M5TM oxidised up to ~17%-ECR at 1100°C
(from Halden IAEA meeting, J.C. Brachet et al., 2001)
1J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy Division
M5TM –– 1000°C
323 sec. 360 sec.
M5 Dé zyrox Essa i 190
0
20
40
60
80
100
120
0 20 40 60 80 100 120
Dis tanc e en % épais s eur gaine
% phase a form
é
Moyenne M5 DC 3
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
100.00
0.00 20.00 40.00 60.00 80.00 100.00
Distance en % epaisseur gaine
% phase a form
é
SF + FH DF + IH% α(O)
Relative distance (%) / clad metal thickness Relative distance (%) / clad metal thickness
% α(O)
2J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy DivisionMicrographies MEB-rétrodiffusé comparatives (épaisseur métal.)
M5TM – Dezirox/Cinog –1000°C 1290 sec. 1450 sec.
M5 Dézyrox Essai 159
0
20
40
60
80
100
120
0 20 40 60 80 100
Distance en % épaisseur gaine
% de phase a form
é
M5 DC 9
0
20
40
60
80
100
120
0 20 40 60 80 100
Distance en % épaisseur
% phase a formé
M5TM –– 1000°C
% α(O)SF + FH
% α(O)DF + IH
Relative distance (%) / clad metal thickness Relative distance (%) / clad metal thickness
3J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy DivisionMicrographies MEB-rétrodiffusé comparatives (épaisseur métal.)
M5TM – Dezirox/Cinog –1000°C 3270 sec. 3270 sec.
M5 Dézyrox Essa i 192
0
20
40
60
80
100
120
0 20 40 60 80 100
Dis tance en % épaisseur gaine
% phase a form
é
M5 Cinog DC20
0
20
40
60
80
100
120
0 20 40 60 80 100
Dis tance en % épaisseur gaine
% phase a form
é
% α(O)
M5TM –– 1000°C
SF + FH % α(O) DF + IH
Relative distance (%) / clad metal thickness Relative distance (%) / clad metal thickness
4J.C. Brachet et al., CEA-Saclay, DEN/DMN/SRMA, 91191 Gif-Sur-Yvette, France
e-mail : [email protected]
Nuclear Energy DivisionMicrographies MEB-rétrodiffusé comparatives (épaisseur métal.)
M5TM – Dezirox/Cinog –1000°C 14551 sec. 13500 sec.
M5 Cinog essai NT3
0
20
40
60
80
100
120
0 20 40 60 80 100Distance en % épaisseur gaine
% phase form
é
M5 Dézirox Essai 152
0
20
40
60
80
100
120
0 20 40 60 80 100Distance en % épaisseur gaine
% phase a form
é
Ep.métal ~530µm Ep.métal ~520µm
% α(O)
M5TM –– 1000°C
SF + FH % α(O)DF + IH
Relative distance (%) / clad metal thickness Relative distance (%) / clad metal thickness