ML20058F663

Results of Mechanical Tests & Supplementary Microstructural Exams of TMI-2 Lower Head Samples

ML20058F663
Person / Time
Site:	Crane
Issue date:	06/30/1993
From:	Dierks D, Neimark L ARGONNE NATIONAL LABORATORY
To:	NRC
Shared Package
ML20058F640	List: ML20058F639 ML20058F643 ML20058F648 ML20058F656 ML20058F663 ML20058F672 ML20058F679 ML20058F691
References
TMIV(93)AL02, TMIV(93)AL2, NUDOCS 9312080220
Download: ML20058F663 (75)

Text

-

s TMIV(93)ALO2 U

i P=4 Jeo{ OECD-NEA-TMI-2

, Vessel Invest.igat. ion i

Project m __

_v l

2 cc Results of Mechanical Tests and Supplementary Micro-l-

structural Examinations of the TMI-2 Lower Head Samples

-g-D. R. Diercks and L. A. Neimark

/

j j

June 1993 l

l 1

NOTICE This report is for the benefit of OECD TMl participants and their designees only This report has taren prepared pursuant to Of CD agreements on the TMl-2 Vessel investigation Project. It is the policy that the information contained in this report be used only for the benefit of the participants and the participants designees.

The contents of this report should not be disclosed to others or reproduced wholly or partially unless authorlied in accordance j

with the laws, regulations, policies, or written permission of j

the appropriate project participant.

j 9312080220 931126 d

PDR ADOCK 05000320 (l-r ! 7 0,90,). jl' l

P PDR a

I Vessel Investigation Project.The contents of this report were developed as part of This project is jointly sponsored by eleven countries Cooperation and Development.under the auspices of the Nuclear Energy Agency The sponsoring organisations are:

The Centre d' Etudes d'Energic Nucl6aires of Belgium, The Sateflyturvakeskus of Finland.

The commi33arial s l'Energie Atomique of France.

The Gesellschaft f0r Reaktorsicherheit mbH of Germany.

The Comitato Nazionale per La Ricerca e per Lo Sviluppo Dell'Energia Nucleare e Delle Energie Alternative of Italy, The Japan Atomic Energy Research Institute, The Consejo de Seguridad Nuclear of Spain, The Statens Kurnkraftinspektion of Sweden, The Offlee Fbderal de l'Energie of Switzerland, AEA Technology of the United Kingdom.

The United States Nuclear Regulatory Conunission, and The Electric Power Research Institute.

The primary objectives of the Nuclear Energy Agency (NEA) are to promote nuclear development, and on assessing the future role of nucle contributor to economic progress.

This is achieved by:

encouraging harmonisaton of governments' regulatory policies and practices in the nuclear field, with particular reference to the safety of nuclear installations protection of man against fontsing radiation and presenation of the environment, radioactive waste management, and nuclear third party liabliity and insurance:

- keeping under review the technical and economic characteristics of nuclear for the different phases of the nuclear fuel cycle and the potent contribution of nuclear power to overall energy demand:

- developing exchanges of scentific and technical information on nuclear energy particularly through participation in common senices:

- setting up internation research and development programmes amd undertakings jointly organised and operated by OECD countries.

Atomic Energy Agency in Vienna, with which it has concluded Agreement, as well with other international organisations in the nuclear field,

TM1V(93)ALO2 Results of Mechanical Tests and Supplementary Microstructural Examinations of the TMI-2 Lower Head Samples by D. R. Diercks and L. A. Ne; mark June 1993 Energy Technology Division Argonne National Laboratory Argonne, IL 60439 Prepared for the OECD-NEA-TMI-2 Vessel investigation Project Work Sponsored by the U.S. Nuclear Regulatory Commission under an Agreement with the U.S. Department of Energy Contract W-31-109-Eng-38

W.4 44 m. We a @h adMF yes f J,W G@re hhNAE sq.h a eA&4.a-',4u+a 5,N-CM.-5.3 4--a-.Fga-.p.-sd4wnm4J+.4-ymou-4 Las a A.ur._J aw 4 an.-m ag*dr 46.4 d.4.a mm 44A.a4L,.&.44 4.dh.4 lame-4w h4p4,,g,.3 4 m.Aa__e4iw44>e,4A-.

')

1.

1 e

a I

I t

' t l

l I

l 1

e f

I l

)

l i

i i

l I. -

e I

t i

h

' Il b

e g'!

2 6

I P

r l

l e

j

. i i

3 1

1 l

P 1

i l

d 5

l s

l i

l l.

I t

I 1

I.

.. j i

'. m 1

5 l

t i

t e

l I

)

i-1

- j l

J l

i i

4 i

t i

s F

i s

].

I 11 i

e t

1 1

I J

e

.~.

___..__. _ _..... _ ~,.. _.. _

1 Results of Mechanical Tests and Supplementary Microstructural Examinations of the TMI-2 Lower Head Samples by i

D. R. Diercks and L. A. Neimark l

l Abstract

?

t Metallographic examinations and mechanical tests have been completed on specimens from 15 prism-shaped samples cut from the lower head of the TMI-2 l

pressure vessel as a part of the TMI-2 Vessel Investigation Project (VIP). The results of these examinations and tests are summadzed here.

The metallographic results were in general agreement with earlier INEL observations. Four samples were found to have attained temperatures as high as 1100*C during the accident, with an estimated cooling rate from the maximum temperature of from 10-100oC/ min. Portions of two adjacent samples also exceeded 727'C. and one laboratory found that a region near the surface of another sample apparently also exceeded 727"C. even though this sample was not near the hot spot. The remaining samples appanntly did not exceed 727'C but four samples probably approached this temperature.

Tensile tests were conducted on the lower head material at room temperature and at temperatures af 600-1200*C. A strong dependence of yield and tensile strengths on temperature was observed, and the data generally matched well with literature data on i

A533. Grade B stael. However, the observed strengths of material from the hot spot in the as-received condition lay well above the remaining data, reflecting the heat treatment received during the accident.

Creep tests were conducted on the lower head material over the temperature range of 600-1200"C at stress levels resulting in failure times of 1-100 h. The data from the lower head material compared well with similar data obtained earlier on archive material from the Midland reactor 600*C. However, at higher temperatures.

the TMI-2 lower head data fell increasingly above data from the Midland material. The TMI-2 data were fit using both Larson-Miller and Manson-Haferd time-temperature parameters.

Charpy V-notch impact tests were conducted on four groups of test specimens.

Specimens from the hot spot showed signillcantly lower upper shelf energies and higher transition temperatures than specimens from regions that did not exceed 727 C during the accident.

I Cracks were found in the stainless steel cladding of boat samples from the hot spot.

]

The cracks appeared to be the result of hot-tearing. probably assisted by intergranular penetration ofliquid Ag-Cd. Crack propagation into the A533 vessel steel was a maxirnum of -6 mm. Materials in the cracks suggest the presence of control-assembly debris on the lower head before the massive fuel flow arrived.

111

ni J.-

p.

.h a

J k

-+-

6 2

e i

i 1

' h i

h t

t 9

a I

f i

i p

' i

+

5 i

e 1

b I

I b

4 h

I L

I f

5 i

1 t

+

' t

. r D

n-I a

i f

5 r

b t

I 4

i 5

?

!'.l

" I 4

1 h

h i

1; I

. i b

9 I

' E

^

i

.Ir b

k iv

[

i 4

e--

- +-..-

h Contents i

t i

I Execuuve Summary......................................................................._........ 1 r

1 I n trod u e ti on.......................

........ 3

[

1 5.

Preparation of Mechanical-Test Specimens.....

4

[

e 3

Examinat1on of Lower Head Cladding......

11 t

3.1 Physical Condition...................

I1

)

3.2 Scanning Electron Microscopy Examinations........................

12 4

Results and Discussion...........

........ 21 l

.i 4.1 Metallogmphic E:mminations..................................................... 2 1

~

4.2 Tensile Tests -

36' 4.3 C reep Test s.............

.....38 4

4.4 Impaei Tests....

45 i

a a

i 4.5 - Cladding Craeks.......,.

..................,..46 i

5 Summary and Conclusions.=

_.........................................51 t

Refe ren ces....

54 Appendix A: Sectioning Diagrams for TMI-2 Lower Head Samples...

57 Appendix B: Strain-vs.-Time Curves for Creep Tests Condut:ted on TMI-2 Lower Head Material--

91 1

1 I

l l

i d

1 W

4

-t l

rf.-,y

-. -,n-

..w-

,,--c--

-,-m-,-,--

r-

..., ~.-..-

2

List of Figures 4

1. Configuration and approximate dimensions of a typical sample from TMI-2 pressure vessel lower head

..............................4

2. Map of the the lower head of the TMI-2 pressure vessel showing locations from which samples were taken.................................

5

3. Test specimen used for tensile and creep tests of TMI-2 lower head material _

6

4. Charpy V-notch test specimen used for impact testing of TMI-2 lower head matedal..........

6 L

5. Surface of Sample E-6 showing end of one leg of crack around Nozzle E-7......... I 1 i

6. Cross section through principal cmck in Sample E-6............................................... 13

7. Surface of Sample G-8 showing two cracks in cladding.

.... 14

3. Cross section of larger crack in Sample G-8......

. 15

9. ~ Bottom" end of large cmck shown in Fig. 8. showing fuel debris in an Fe-oxide matrix

...._. 1 6

10. Cross sectbn thmugh small cmck in G-8....................._........................................ 17 i

I1. Cleaned and etched surface of Sample F-10 showing small interdendritic

.f craeks in a weld pass......

......... 1 8 l

12. SEM-BSE images of multi-layere.1 material on crack surfaces of Sample E-6 19 l 3. Internal tears in the cladding of Sample G-8...................................................... 2 0 i

j

14. Metallographic specimen from lower head sample E-6 showing absence of feathery carbide precipitate layer at cladding / base-metal interface....

25

15. Metallographic specimen from lower head Sample K-13 showing presence of feathery carbide precipitate layer at cladding / base-metal interface.......

... 26

16. Metallographic specimen from lower head Sample E-6 showing spheroidization of delta ferrite phase in Type 304L weld cladding layer....... 27

17. Metallographic specimen from lower head Sample E-8 showing spheroidization of delta ferrite phase in cladding....

... 28

18. Metallographic specimen from lower head Sample E-8 showing austenite l

grain growth in base metal =

.... _.. 2 9

19. Metallographic specimen from lower head Sample E-8 showing absence of carbide layer at cladding / base-metal interface.................................................-. 3 0

)

\\

20. Metallographic specimen from lower head Sample E-8.....

.... 31

21. Metallographic specimen from lower head Sample F-10 showing absence 4

of carbide layer at cladding / base-metal interface........................................................ 32 1

vi 1

b l

i i

[

22. Metallographic specime,n from lower head Sample F-5 showing evidence

[

of partial reaustenitization of base metal to a depth of -15 mm below the

[

cladding / base-metal interface..

...................34 j

23. Metallographic specimen from lower head Sample M-11 showing evidence of partial reaustenitization near cladding / base-metal interface...

37

24. Tensile and yield strengths of TMI-2 lower head material compared with Japanese National Research Institute for Metals data for other heats of f

A5 33, G ra d e B s t eel...

.......................... 4 0

25. Stress vs. time to rupture data from creep tests conducted on TMI-2 lower i

hend material with estimated best-fit curves........................ -

....... 4 1

26. Best-fit curves to data of Fig. 25 plotted vs. data previously obtained for Midland an-hive material in OECD round-robin tests..........

.................................44

(

27. Plot of log (o) vs. Larson-Miller parameter (C = 12.5) for TMI-2 lower head matertal creep data...

45 i

28. Stress vs. time to rupture dafa from creep tests conducted on TMI-2 lower f

head material compared with best-flt curves from the Larson-Miller time-temperature correlation -

.. 4 6 i

i

29. Plot of log (o) vs. Manson-Haferd parameter (ta = 7.57. Ta = 520) for TMI-2 lower head material creep data....-

... 4 7 i

30. Stress vs. time to rupture data from creep tests conducted on TMI-2 lower i

head material compared with best-fit curves from the Manson-Haferd time-I temperature correlation..

..._. 4 8 i

a

.1 l

31. Absorbed impact energy vs. test-temperature data from Charpy V-notch l

Impact tests on specimens from TMI-2 lower head...

........ 5 0 i

Al. Dimensions and initial sections from TMI-2 lower head Sample D-10..

.. 5 9 l

t A2. locations of mechanical test specimens cut from Sample D-10.......................... 60 A3. Dirnensions and initial sections from TMI-2 lower head Sample E-6.

...... 61 l

4 i

A4. Dimensions and initial sections from TMI-2 lower head Sample E-8

.62-i 6

A5. Locations of mechanical test specimens cut from Sample E-8..

......... 63

{

AG. Dimensions and initial sections from TMI-2 lower head Sample E-11...

... 64 f

A7. Locations of mechanical test specimens cut from Sample E-11............................. 65 i

i A8. Dimensions and initial sections from TMI-2 lower head Sample F-5.................. 66 L

A9. Locations of mechanical test specimens cut from Sample F-5..

... -. 6 7 2

l A10. Dimensions and initial sections from TMI-2 lower head Sample F-10......... 68 1

1

)

i i

vil

~_,

~

i Al1. Locations of mechanical test specimens kl through k6 cut from l

Sample F-10...

- 69 l

l Al2. IAcations of mechanical test specimens k7 through k12 cut from i

Sample F-10...................................

........................70 A13. - Dimensions and initial sections from Bil-2 lower head Sample G-8. _.....

71 l'

A14.14 cations of mechanical test specimens cut from Sample G-8......

72 A15. Dimensions and initial sections from TMI-2 lower head Sample H-4................... 73 i

i A16.14 cations of mechanical test specimens kl through k6 cut from-Sampie H-4 _..

. 7 4

(

i A17. Locations of mechanical test specimens k7 through k12 cut from j

Sample H -4.........._.......

_. 7 5 l

A18. Dimensions and initial sections from TMI-2 lower head Sample H-5.................. 76

~j i

A19. Locations of mechanical test specimens cut from Sample H-5

.. 7 7.

-l t

A20. Dimensions and initial sections from TMI-2 lower head Sample H-8.................. 78 A21. Locations of mechanical test specimens cut from Sample H-8.

79 l

A22. Dimensions and initial sections from TMI-2 lower head Sample K-7

_. 8 0 l

1 A23. Locations of mechanical test specimens cut from Sample K-7

.81 l

A24. Dimensions and initial sections from TMI-2 lower head Sample K 82 A25. Locations of mechanical test specimens cut from Sample K-13........................... 83

[

t A26. Dimensions and initial sections from TMI-2 lower head Sample L-9................... 84

{

A27. Locations of mechanical test specimens cut from Sample L-9.....

.85 j

l A28. Dimensions and initial sections from TMI-2 lower head Sample M-8...._... 86 j

i A29. Locations of mechanical test specimens cut from Sample M-8.....

87 l

A30. Dimensions and initial sections from TMI-2 lower head Sample M-11....

88 l

l A31. Locations of mechanical test specimens cut from Sample M-11......

....................89 1

l L

F r

vli!

I i

- ~

..i l

l i

i List of Tables e

i 1.

Summary of mechanical testing matrix for Bil-2 lower head material-

..... 7 2.

Number of metallographic and mechanical test specimens obtained from TMI-2 lower head samples -

............8 l

t 3.

Distribution of TMI-2 lower head mechanical test specimens to OECD partner laboratories -

............._.............9 i

4.

Distribution of TMI-2 lower head base metal metallographic specimens to OECD partner laboratories..... =

1 0 i

5.

Summary of results from examinations of metallographic samples from TMI-2 lower head--

=...........

........._.. 2 2 l

6.

Summary of tensile data obtained from base-metal specimens of the j

TMI-2 lower head -

... 39 7.

Summary of tensile data obtained from cladding specimens of the TMI-2 l

lowe r h ea d............................ -....

_. 40 i

8.

Summary of creep data obtained on specimens from TMI-2 lower head...

42 9.

Summary of Charpy V-notch impact data on specimens from TMI-2 lower i

i hea d.... --

.... 4 9 1

1 l

i

.i k

1 4

9 t

I

/

i a

i 0

k l

I 1

1 ix

_ -_~.,_ _

Results of Mechanical Tests and Supplementary Microstructural l

Examinations of the TMI-2 Lower Head Samples by 9

D. R. Diercks and L. A. Neirnark Executive Summary Metallographic examinations and mechanical tests have been completed on specimens from 15 prism-shaped samples cut from the lower head of the TMI-2 pressure vessel. These tests were conducted as a part of the TMI-2 Vessel Investigation Project (VIP), an international program conducted jointly by the U.S.

Nuclear Regulatory Commission (NRC) and the Organisation for Economic Co-operation and Development / Nuclear Energy Agency (OECD/NEA). The results of these examinations and tests, which were conducted jointly by Argonne National Laboratory (ANL) and the European partner laboratories, are summanzed and compared with die metallographic results reported earlier by the Idaho National Engineering Laboratory (INEL).

The metallographic results were in general agreement with INEL observations.

Specimens from Samples E-6. E-8. F-10. and G-8 which comprised a so-called " hot spot" near the bottom of the lower head, were found to have attained temperatures as high as 1100 C during the accident. The cooling rate from the maximum temperature was generally estimated to have been from 10-100 C/ min. The end of Sample H-8 adjacent to the hot spot was also found to have attained temperatures in excess of 727*C, as did portions of nearby Sample F-5. One laboratory also found that a region near the surface of Sample M-11 apparendy also exceeded 727 C. even though this sample was not near the hot spot. The remaining samples apparendy did not exceed 727*C. but tempering of the bainite, which was observed by one laboratory in Samples 11-4. H-5. M-8. and L-9. suggested that these four samples probably approached this temperature.

Tensile tests were conducted on the lower head material at room temperature and at temperatures of 600-1200 C. A strong dependence of yield and tensile strengths on temperature was observed; the room-temperature values were reduced by more than a factor of 2 at 600*C and by more than a factor of 10 at 900*C. The data generally matched well with data earlier obtained by the Japanese National Research Institute for Metals (NRIM) for five other heats of A533. Grade B steel. However, the observed strengths of material from Samples E-6 and E-8 in the as-received condition lay well above the remaining data, reflecting the austenitizing heat treatment and relatively rapid cooling that this matedal was exposed to during the accident.

Creep tests were conducted on the lower head material over the tempelature range of 600-1200*C at stress levels resulting in failure times of 1-100 h. No significant effect of prior thermal history on stress-rupture life was observed, although no samples for which the maximum temperature had significantly exceeded 727"C were tested. The data from the lower head material compared well with similar data l

J

~.-

. =--

__ _-~

2 i

i obtained earlier on archive matedal from the Midland reactor 600 C. However, at l

higher temperatures, the TMI-2 lower head data fell increasingly above data from the j

Midland matedal. The TMI-2 data were f1t using both Larson-Miller and Manson-Haferd time-temperature parameters. Of the two correlations. the Manson-Haferd analysis produced the better 111.

i Charpy V-notch impact tests were conducted on four groups of test specimens.

}

Specimens from Samples D-10. H-4, and E-11, for which the maximum temperature did not exceed 727"C, showed similar behavior, with an upper-shelf energy of -170 J and a transition temperature on the order of 20*C. However, specimens from Sample -

F-10. for which the maximum temperature was as high as -1050*C, had an upper-shelf energy of-120 J and a transition temperature of-70 C.

i Cracks were found in the stainless steel cladding of boat samples from the so-called " hot spot" (E-6. G-8. and F-10). The cracks appeared to be the result of hot-teadng probably assisted by intergranular penetration ofliquid Ag-Cd. Crack l

propagation into the A533 vessel steel was a maximum of -6 mm. Materials in the i

cracks suggest the presence of control-assembly debris on the lower head before the massive fuel flow ardved.

l 1

1 I

i t

j a

I t

1 2

i s

I 1

1 2

l i

n

..--~--. -

n c

, ~-

v

3-1 Introduction f

The TMI-2 Vessel Investigation Project (VIP) is an international program being l

conducted jointly by the U.S. Nuclear Regulatory Commission (NRC) and the l

Organisation for Economic Co-operation and Development / Nuclear Energy Agency

[

(OECD/NEA). Participants in the international project include the U.S., Japan,

_l Belgium, the Federal Republic of Germany (FRG), Finland, France, Italy, Spain, Sweden, Switzerland, and the United Kingdom (U.K.).

f During the first phase of the project,15 samples were recovered from the lower j

head by MPR Associates Inc. The samples are prism-shaped, each -152-178 mm (6-7 l

in.) long, 64-89 mm (2.S -3.5 in.) wide, and 64-76 mm (2 1/2-3 in.) deep, as shown in Fig.1. The samples were cut from the inner surface of the lower head and typically extend through approximately half the lower head thickness. The specimens were i

taken from (1) near the area of impact by the primary stream of molten material on the

{

lower head; (2) toward the radial center of the lower head underneath the maximum I

thickness of debris: (3) in the quadrant of the lower head where a " wall" of consolidated l

debris similar to a lava front had developed: (4) in a location of the lower head not contacted by the molten material (to act as a control sample); and (5) locations with one or more instrument penetrations, particularly where surface cracks had been observed visually. The locations from which the lower head samples were taken are i

shown in Fig. 2.

Cladding cracks were observed in three of the lower head samples, namely E-6.

G-8, and F-10, during initial examinations conducted at Argonne National Laboratory (ANL). Cladding cracks were also detected at location G-6 in the TMI lower head, but i

no sample was removed at this location. Metallographic and scanning electron microscopy (SEM) examinations were conducted on Samples E-6 and G-8 in some detail, to characterize the nature and extent of the cracking. The results of these examinations are reported below.

Following the initial examinations, metallographic specimens were cut from the l

lower head samples, decontaminated, and sent to the Idaho National Engineering l

laboratory (INEL). These specimens were subjected to detailed characterization by optical metallography and hardness measurements to determine the maximum

{

temperature attained at various lower head locations during the accident.1 Supplemental examinations, the results of which are summarized in this report, have been conducted by ANL and participating OECD partner laboratories. Based in part' 1

upon the results of the ANL examinations, a mechanical-testing matrix was developed to determine the tensile and creep properties of the lower head material under -

conditions relevant to the accident scenario. The tests were conducted by ANL and participating OECD partner laboratories, and the results are summarized here. These -

results have been used by analysts at INEL to assess the integrity of the lower head and

~!

its margin-to-failure during the accident.

.l r

i

1 I

4 Stainless-Steel-Clad inner Surface of

/

Lower Head 7

Ar 5 **

/

f o

y l

+ =70 mm i

t l

g

\\

j

\\

\\

\\

=65 mm i

U Fig.1. Configuration and approximn._ Hrr tsions of a typical sample from TMI-2 pressure vessel lower head.

2 Preparation of Mechanical-Test Specimens The tensile and stress-rupture tests that are described below were conducted on specimens of rectangular-cross-section, as shown in Fig. 3. This design was chosen to i

satisfy ASTM Standards E8 and E139 and applicable standards of the Deutsches Institut fQr Normung (DIN). Specimens of both flat and circular cross section were used in earlier round-robin creep tests on archive material obtained from the lower head of the Midland reactor.2 and no significant differences were observed in the results obtained from the two specimen designs. The flat design was used exclusively in the case of the TMI-2 lower head samples to conserve material. The specimen design used for the impact tests was the conventional Charpy V-notch test specimen (ASTM E23) shown in Fig.4.

The mechanical-testing matrix developed by ANL and the participating OECD laboratories for specimens from the lower head material is summarized in Table 1.

Input was obtained from the analysts at INEL responstble for assessing the lower head integrity to ensure that this test matrix included all of the properties and test conditions needed for these analyses. Tensile tests were conducted at room temperature for purpose of comparison with data in the literature. All other tensile and creep tests were conducted at a minimum test temperature of 600 C. It was judged that little or no damage would have occurred to those portions of the lower head for which the maximum temperature did not exceed this value and that failure was unlikely at these locations. The maximum temperature of 1200 C for these tests lies slightly above the maximum lower head temperature believed to have been attained during the accident.

l l

l l

4 5

l l

A B

C D

E F

G H

K L

M N

O P

R Q

I I

NORTH m

/

N

[

j l

l K

15

\\

[.

O!

l O

14

/

oi o oj o o

13 7 ;

o:j

,o o'

t 12 j

[

oi oo i

11

/

.b... b.._.

M......

0 I

!. 0 i..... /

i. '

0 S

_ (f.!

M.. i.._[_..

l 8

...b.. 2.

I 1._

l

_ 0i

...s..

\\,a i

o o:

i e-io o i

e

\\

gg oj o

o!

5

\\

....._.....i..

.U

!.._. 2

\\

o!

o!o 3

(

2 l

yo,.. I - oJ.

\\

u.._.2 N

Locations of Boat Samples r

o Nozzle Positions Nozzles Examined at ANL Nozzles with Associated Cladding Cracks o

s'tg. 2. Map of the the lower head of the TMI-2 pressure vessel showing locations from which samples were taken.

9 8

6 i

i i

i DIA = 6.3 to 6.5

+0.1, -0.0 l

• 12.0-15.0 =

24.0 ~

=

+ 4.032 s.08 +

a

~

(S

(

~

o

)

)

h n

~

I I

R = 7.0 78.0 1.0 3.0 +

~

=

=

Fig. 3. Test specimen used for tensile and creep tests of Th'-2 lower head material.

All dimensions are in mm.

t c

55.0 I

Y V

n A

h o

a:i o

o 4

l Thickness = 10.0 mm; V-notch angle = 45 with 0.25 mm radius at tip.

t Fig. 4. Charpy V-notch test specimen used for impact testing of TMI-2 lower head material. All dimensions are in mm.

i 1

l

7 Table 1. Summary of mechanical testing matrix for TMI-2 lower head material.

Temperature Damage

("C)

Level Tensile Tests Creep Tests Impact Tests a

R. T.b' Low Belgium Italy France U.S.c Moderate Spain Italy b. S.

Belgiume Severe Belgium Italy Spain 600 Low Belgium Belgium France Moderate Spain Spain U.S.

700 Low Belgium France France Moderate Spain U.S.

U.S.

800 Severe France Belgium U.S.

900 Severe Belgium U.S.

Spain 1000 Severe France Spain U.S.

I100 Severe Belgium France Spain Belgiume 1200 Severe France U.S.

U.S.

aEach series of creep tests consists of four tests with stress-rupture lives of =1,5, 20, and 100 h.

bR.T. = room temperature.

cTest on specimen from cladding.

The damage levels listed in Table I refer to the level of damage believed to have be n sustained by the lower head samples during the accident, based upon preliminary metallographic and hardness information available at the time the test matrix was prepared. Low damage refers to that sustained at a maximum temperature of <727'C during the accident, moderate damage to that sustained at a maximum temperature of

-727-900*C, and severe damage to that sustained at a maximum temperature >900 C.

Subsequent examinations revealed that estimates ofinitial damage were not accurate for some lower head samples, but specimens with various damage levels were nonetheless tested.

.m r

i 8

Table 2.

Number of metallographic and mechanical test specimens obtained from TMI-2 lower head samples.

l Sample Tensile and Creen Soecimens impact Metallography l

Number Base Metal Claddinga Specimens Specimens j

D-10 0

0 6

5 E-G 1

0 0

4 E-8 8

2 0

9 E-11 0

0 5

5 I

F-5 15 3

0 8

F-10 0

0 10 4

G-8 11 0

0 3

{

11-4 0

0 12 6

H-5 14 2

0 5

H-8 17 3

0 4

K-7 14 3

0 8

K-13 18 3

0 7

L-9 17 3

0 8

M-8 14 3

0 5

M-11 17 3

0 8

Totals 146 25 33 89 l

aSome cladding specimens were not completely decontaminated and were therefore not tested.

1 Because the number specimens with severe-damage-levels was limited, it was necessary, in some cases, to heat treat low-damage specimens before testing to produce the microstructure associated with a severe level of damage. This heat treatment consisted of heating the specimen to 1000*C, holding it at this temperature for 2 h, and then cooling it to room temperature at 50 C per min. For specimens to be tested at 1000 C or above. this prior heat treatment was omitted, because its effects would be negated by the thermal treatment imposed during testing.

Detailed diagrams that show how the lower head samples were sectioned to provide the test specimens for the matrix of Table 1 are presented in Appendix A. The number of specimens of each type obtained from each of the lower head samples is summarized in Table 2. These specimens were distnbuted to the laboratories participating in the mechanical-testing program as indicated in Tables 3 and 4.

i 9

Table 3.

Distdbution of TMI-2 lower head mechanical test specimens to OECD l

partner laboratories.a TMI-2 Lower Number Test Specimen Ilead Sample OECD Partner of Test Type of Identification l

Number Laboratory Specimens Specimen Numbers l

D-10 Italy 6

Impact kl-k6 E-8 Belgium 1

Tensile 18 i

Spain 1

Tensile t7 E-11 Italy 3

Impact kl-k3 F-5 Belgium 9

Tensile ll.12b 17-t l 3 i

Spain 5

Tensile 114-118 j

F-10 Italy 9

Impact kl-k3 k5, k7, k8. kl0-k12 11-4 Italy 9

Impact kl, k2, k4. k5-k8, kl0,k12 K-7 Spain 8

rensile 17-114 K-13 Belgium 8

Tensile 17-114 L-9 France 8

Tensile 17-t14 Spain 2

Tensile tl5. tl6 M-11 France 8

Tensile 17-t14 aThe " tensile" specimens are used for both tensile and creep tests. All specimens are from the lower head base metal except as noted.

bSpecimens 11 ann.2 from Sample F-5 are cladding specimens.

o

j 10

i Table 4.

Distribution of TMI-2 lower head base metal metallographic specimens to-l OECD partner laboratories.

6 TMI-2 Specimen -

Lower Head OECD Partner Number of Identification Sample Number Laboratory Specimens Numbers j

i D-10 Italy 1

m1 E-8 Delgium 1

m5 f

Finland 2

m4 and ml0 France 1

m7 FRG 1

m9 Spain 1

m6 F-5 Belgium 1

m5 FRG 1

m6 l

U.K.

I m4 l

F-10 Italy 1

m5 l

U.K.

I m4 l

H-4 FRG 1

m5 Italy 1

m4 l

U.K.

I m6 l

H-5 FRG 1

m5 '

U.K.

I m1 K-7 FRG 1

m6 Spain 1

m5 U.K 1

m4 K-13 Belgium 1

m4 FRG 1

m5 U.K.

I m9 L-9 France 1

m10 FRG 1

m9 U.K.

I m7 M-8 FRG 1

m4 U.K.

I m1 M-11 France 1

m7 FRG 1

m5 SpMn 1

m6 U.K.

I m4

i 11 i

7 e

'I

.t t

. ;p; #

g 4.a p

Ut

'i j ;./

.e t 4 i*

@ <ij f

^f-m,. t

~-

u

.n y--q,,.

i n

+

2 %-

l

y

} '1 Yg 6 jR lllQ,. Mit"if y

. :p I

9_t m.#

.$, *t

,jQ) ' ; @7_,;;;c' l;&; t.%, j!.

s..

j s

y.,

e

- Qi

_g;.,, v.

4,

.g

(

.es

- ):-

'. gM e-tA.

1-4., f:

i - ;

- {4(

5

+

s mg ',,. s + yg3l g y.1

{.

~ % : }. ', ;y g

., j..,. w ', ;

s.

v

..:gr,;

x 6,

4 : <q s :h? y c

L W (t:: 3i

. &, e 4.

r-..'..,"

'.. p; h, y

..-v

. ha Q: e df f. sb ln

.u 2

s,y wl%g:;y

'

• 41, ff Q,3.. & ofy @M e:,. m.o % y ;. ;

r zy g,a.5

%m. %, x.

?.

9'-

m 2

L - %e*$f **ly:41%v p*}aa.

. \\., : uvg.3;W M';s% :q,.

7

• 13li

' grtg u

&m ;w y d p e"hN ' MW*

b f;s h:u N.A y p

<y. A

,+.:n d y

% p;.

c.

y

.,t

.m 25 mm

L u

m Fig. 5. Surface of Sample FA showing end of one leg of crack around Nonle E-7.

MCT 276674 3.

Examination of Lower Head Cladding 3.1 Physical Condition After removal of the hard layer of core debris. the lower head was visually examined by video camera. It was found that a significant U-shaped crack in the surface encircled the E-7 noule. The E-G boat sample that was taken encompassed j

one leg of this emck. as shown in Fig. 5. A cursory visual inspection of Sample E-6 by personnel at TMI-2 immediately after removal suggested that the crack penetrated j

1 i

i 12 i

essentially through the entire depth of the sample, i.e., significantly more than 6 cm into the lower head. However, a closer examination at ANL revealed that what was

{

thought to be the penetrating crack on the end of the sample was actually the intersec-tion of the two cutting planes made by the metal disintegration machining process (MDM). Metallographic cross sections through the crack, Fig. 6, confirmed that the i

crack penetrated the A533 vessel steel only superficially, ~3 nun.

f The appearance of the crack in Fig. 6 strongly suggests that the Type 308L i

cladding failed along interdendritic boundaries by a hot-tearing process that apparently l

was the result of thermal stress when this location was cooled rapidly at the rate of 10-100'C per min.

l I

Inspection of two other boat samples at ANL, G-8 and F-10, indicated that they, too, had cracks in the cladding. The cracks in the surface of G-8 are shown in Fig. 7, Whereas the activity of the E-6 metallographic sample was suinciently low not to require preparation in a hot-cell, the G-8 sample was atypically very radioactive, Indicating the presence of fuel /11ssion products in the cracks. Cross sections through

[

both the large and small cracks, Figs. 8. 9. and 10, indeed, showed fuel particles trapped in an fron (oxide) binder. Both cracks show the same. evidence of hot tearing as Sample Fe6, with graphic evidence of the elevated-temperature ' ductility of the Type 308L weldment. Penetration of the A533 vessel steel was somewhat greater than at E-6. i.e. -6 mm.

_j After surface debris had been removed by chemical means, the surface of Sample

[

F-10 was determined to be cracked. A portion of the etched cladding surface is shown in Fig. I 1.

Light cracking can be seen in the longitudinal interdendritic boundaries in the weld passes. This cracking could have occurred either during fabrication or at the same time as the formation of the cracks at E-6 and G-8; the F-10 sample was on the periphey of the oval-shaped hot spot in the vessel wall. The cracking at the G-8, E-6, and F-10 locations provides additional evidence for this hot spot 3.2 Scanning Electron Microscopy Examinations The debris contained in the cracks of the E-6 and G-8 samples and surface -

l scrapings from other boat samples were analyzed by scanning electron microscopy and I

energy-dispersive X-ray analysis (SEM-EDX) in an attempt to better understand the conditions on the lower head when the cracks were formed. Thi crack surfaces in the E-6 sample were coated with adherent and conforming layers of non-metallic debris.

[

apparently oxides of debris constituents, that appear to have been molten and present at the time of, or shortly after, crack formation. The principal constituents of these layers, some of which are shown in Fig.12, were Fe, Cr. and Ni with Sn In, Ag, and Cd i

in combinations as second phases or discrete particles. The structure within these i

layers indicates that the constituents were once in a molten state and not formed

{

simply as oxidation products of the base material. In particular, there were trapped, rounded nodules of Ag-Cd and needles ofin-Sn in a matrix of principally Fe-oxide.

The material surrounding the stainless steel cladding fragment at the base of the crack in Fig. 6 indicates that the fragment, like the surfaces of the crack, was being dissolved l

by a liquid phase that contained Fe, Ni. and Cr as the major constituents, with Mn. In, i

and Sn as minor constituents.

t i

'f

ll l

i l

13 i

j i

I

/

1..

i s

j

{\\

sa

'h i T}

-(< : p...

m hl

..,g ~

j

\\

(i a

c l

I

/

t l

If l

^

t

+

1 j

,,i

. ;C;

$b ',

i 3_:

g, -

.l.

,7..a~*

'[

[

f s :'i r,

.q -.

.,s y

.e..

gst-

+

i a.

2mm Fig. 6. Cross section through principal crack in Sample F,-G. Arrows identjfy larger Ag-Cd nod u'es. Etched.

MCT 27G818

e e

14

~~

}Y W" i

r ** * ' ? m'

^

, *.. - g..w. p 4 3.,

s

,e,,

gy t

O

' NT I

N

,f e

lr i #,;

.?

g W,

s *

~

e,.

.'.,,8...

4

-9+

, 3 *

, e 4, *^ e? '

,,%. 3,

,5 '".,t-

/

p* ?

t 4.

-j<N. 4.

s

p. + & E

.nA..

W.

g.Y y. "

., 4s g

ar

,s r

s c,,,: g "'e, t

.jp Eu 4* e "

q.f fl

.{

g -

l

^

0 s

'*f, '

~

f. $ ;.,%.. Qi-[r

',WJ %r l

i fp.7 ",-.

4

[_a#)-

• t o

^

m'.Jy ;;j,$e #g, c[~,I S

)-

'?

' )

n$

.?

~~

,'Qf-M p,

e m s.

p g

9

[+

- 'i

• yv

',%T

.,4 4 s,,, ';. ",

-5 y;, 7' +, M 'vPg i

• E

'q

,g

. 7. '

' 5 il

4.

7 A

~

d' _gw A,

3.3 s.

ga m;,

c.:. p-w

,4c+zVh %,1J sg.

3 f.*.eaa^

wa ;

+,.

S.

9.,. g g y

'N. # $ e 4..

f y y *, 4 y%gll.' e ty4,.'q % $.,'

Af w r

g

. w

'?. A;, j,

/,4

,, g.(je s,py( ? j,;.,

, g.

. ig"

,i

. +

., s pg..A.g %

i A g' ao eg

.,,e'--

u 4

.x -

v i~-

[ f

,.4 g + a a

', -f f u

}.

f..

, " a;.

s.4m

/

• s%,.

s-t

_ t.

g >.

g

,d'

. ki-

[

-?

E y ',, '_ }

t?

u e

s-6+

"1 p.

,,,. '.. $ V ',.

• 4 i

%*F y

l_

5

-f

+,e

. s,

[

4 <

.4 r

gis

, e.A >

.5

)

+

e

". y

. e t

7,,

y, g

[

+

.p 33'

.g

?

e, f ;'

e

• b _' -

l M.f4

• R ) s ;c 3 + l,,**

s gg' ^+gs %,

x '; &.,

^

e

. t pto gc c,,.

. e w#

g t$ pk.*. -

L %.

ao I, f r' s,

w-y ^,

4 i gg 3

p.

g b-3f?, '

f~s

i..Q

. h ",' N 'p

A ~

' ~

?

g x

y g

e t

>ej d

4; 1

rm I

I 4

r s

t l

a

'.i

)

.f

~.

r

,1,.

s.~

A t

4 I

- i 6

s

- eV

\\i 10 mm l _

i 8 -

I I

r 4

g 4a 4

i Fig.I1. Cleaned and etched surface of Sample F-10 showing sinall interdendritic cracks in a weld pass.

MCT 278090 s

19 l

,j f, y) j!

-k t

,,: {,tl 1

t ?-

y l

..-g-

..(

e_

)-

,{>~/g,,t- !y?'(,

l, f.

~y,,X p.;j,,b[ ' gp% ni y

l*

. ~. l -

7.,,,, u

..y A a,c ~%

~*

s

,l%.'lrMQ(%;hj

. i-

?,?);',.,.V/ m

?. '.

-'si; V y.'%

3y

[Y.$,

g i;~

y,W,

,) ;-

.y i

y y',, p'

'/ d,4,l>1.'*? %c. 3, ~. y]

,;. a A. N (,[4.,.

. u.1

/

5

[: ys: - i-

<r,.

c 9

4t,,

' y:;. ;, s.,' ; -

+- ' k s 5%,9 t

3 (a)

(b)

mmr,

y <

p Ik

. p.;

3,.

f; h.

g i, V fj,'I k hh i'.\\.,,,,(?. j'. A-plJ

'J.c' L. ;flv'?S

.' l %

sf):l"syf.+f.F

.,,B a 1 n

.. n,

'.. ' ' A )r. l E M..f y

p<o..

g.

,g Y

s+V.M i

,, {m fs. r mma a, i

f 4 l

.-),

n 4',,4 l; N )m.. c

'4 3

" y;4?dg, 4 2

y g

(

rs

Qy,y,,.i%n(.4,Q, '

, A

' v v,

' (h

,e

.,J J.,

4_

(c) i l

Fig.12. SEM-BSE images of multi-layered material on crack sm f aces of Sample E-6.

i (a) Fuel particles (white) on oxide surface (arrow): (b) area outlined in (a); and (c) In-Sn needles in Fe matrix. AR-Cd spheroids (A), and fuel particle (D).

20

/

):ej (I )

/ c.e,

W-1 s a..n, wem

. - Jo 4

) -

A.hh,h

~

i

, [ 7%w~fs J W, 1

'1

~

' i[

1 J-

3;

'WWa

1

,e

',i[m t

7m.

.h" jy l..

I Fig.13. Internal tears in the cladding of Sample G-8. Solidified Ag-Cd Masses are shown (arrow) ~4 mm beneath the surface. (SEM-BSE image) i of the boat sample cladding near the crack contained several small surface penetrations that contained Ag that appeared to have penetrated intergranularly as a liquid.

)

Solidilled masses of Ag-In-Cd were also found within some tears -4 mm below the surface of the cladding, as shown in Fig.13.

1 l

The materials in the crack in Sample G-8 were essentially the same as those found in the E-6 crack, except more fuel shards were present. These fragments were in a j

matrix of Fe oxide. Solidified masses of Ag-In-Cd and some fuel particles were found in the Fe-rich matrix within the crack extension into the vessel steel. The surface of the i

j stainless steel cladding within the crack exhibited surface reaction layers similar to l

those in the E-6 crack. The upper cladding surface, however, was more ragged than that at E-6 and the intrusion of Ag-Cd stdngers beneath the surface was more prevalent and obvious. The crack contained numerous pure-Fe spheroids within a thin oxide

]

coating that apparently were from the MDM cutting operations. These spheroids were clearly independent of the core debris in the crack. Copper stringers were found in the cladding next to the crack, suggesting a reason for the hot tearing of the cladding.

l The Cu was occasionally combined in the stringers with Ag and In.

l Small quantities of the surface debris on each boat sample were scraped from the surface for SEM-EDX analysis. However, only the scrapings from Samples F,-6, E-8, 1

-~ - - -.. - =,

i 21 E-11, and F-10 were analyzed. On Sample F-10 the fragmented parueles ranged in size from -10 m to a few millimeters and consisted principally of Fe and Cr-oxides i

with a few flecks of U-Zr. The structure was generally inhomogeneous, and Zr. In, and Ag were also found. The particles collected from E-8 were generally angular and.

basically, fragments of fuel containing U. Zr, Fe N1, and Cr in widely varying l

concentrations. On E-11, the particles were agglomerates made up of small particles j

from <10 to -300

m. The small particles were U and Zr fuel of varying compositions, i

and the agglomemte matrix was essentially Fe-oxide. The scrapings from the E-6 sample consisted of spherical and angular particles on the order of 100 pm and less.

The spheroids were Fe and the angular particles resembled the inhomogeneous scrap-ings from the Sample F-10.

In summary, the scrapings appeared to be both material laid down during the accident and adventitious material (Fe spheres) that arrived later during sample re-moval. The collective inhomogeneity in composition of the fuel particles on the j

paruculate scale contmsts with the apparent gross homogeneity of the mass of "compcalon" material that had lain on the lower head.3 It is not possible to determine when during the accident these fuel fragments arrived on the lower head, i.e., before or j

during the massive fuel relocation to the lower head.

i 4

Results and Discussion 4,1 Metallographic Examinations i

Results from the metallographic examinations of specimens from the lower head are summarized in Table 5. With a few exceptions, the the estimates of maximum temperature by the participating laboratories are in good agreement. Samples E-6. E-8, F-10 and G-8 attained the h!ghest temperatures (up to -1100 C) during the accident, I

and these samples, along with one end of nearby Sample H-8, compnse the so-called l

" hot spot" that had been identifled in preliminary work at ANL4 and confirmed by more i

detailed examinations conducted at INEL.I 11 also appears that portions of Sample F-5.

which was near the hot spot, the maximum temperature exceeded 727 C, and a small portion of sample M-11 also may have reached or slightly exceeded 727 C. The j

supponing metallographic observations for these samples are summarized below.

Sample E-6.

Metallographic specimens from Sample E-6 were examined at ANL and INEL. The estimates of maximum temperature obtained by INELI were based upon three general microstructural features, namely (1) the dissolution (which begins after

-10 minutes at 900*C) of a thin feathery carbide layer at the cladding / base metal j

interface; (2) prior austenite grain size in the ferritic steel base metal, where grain j

growth is observed at ~900*C, with significant growth at temperatures in excess of

{

-1000"C; and (3) spheroidization of the delta ferrite islands in the austenitic weld cladding layer, which begins to occur at 1000-1100 C. The extent of carbon diffusion j

into the stainless steel from the base metal that is obsened at the interface was also used as an indicator, as were the measured hardnesses of the base metal and interface regions. Standards were prepared by subjecting TMI 2 lower head material to carefully controlled heat treatments and comparing the resulting microstructural features in

)

these standards with those observed in the metallographic samples.

{

i l

l

22 Table 5.

Summary of results from examinations of metallographic samples from TMI-2 lower head.

Sample No.,

Specimen Maximum Temperature Attained Laboratory Number During Accident ( C)

Sample D-10 Italy m1

<727 ANL m3

<727 INEL m2

<727 Sample E-6 ANL 402A-1 and 4 1000-1100 l

INEL m1 1075-1100 l

Sample E-8 Belgium m5

>727 i

Finland m4 and ml0 1100 in cladding:

950 at 34 mm below clad interface France m7 1000-1100 FRG m9

>850: probably >1000 Spain m6

>1000 f

ANL m2 1000-1100 INEL m3 1075-1100 Sample E-Il ANL m2

<727

{

INEL m3

<727 Sample F-5 Belgium m5

>727 to 20-30 mm below clad (7) l FRG m6 730-850 to -15 mm below clad UK m4

>727 to 15 mm below clad:

)

-727 to 40 mm below clad ANL m2

<727 INEL m3

<727 Sample F-10

)

l UK m4

>727 ANL m2 900-1000 INEL m3 1040-1060

t I

23 Table 5.'

Summary of results from examinations of metallographic samples from TMI l]

lower. head (cont'd.).

)

l

' Sample No.,

Specimen Maximum Temperature Attained Laboratory Number During Accident ('C)

-l I

Sample G-8 ANL 408P-2.3, and 4 1000-1100 l

INEL m1 1040-1060 l

l Sample 1-I-4

{

\\

FRG m5

<727 Italy m4

<727 UK m6 less than but possibly near 727 l

ANL m2

<727 INEL m3

<727 Sample 11-5 l

FRG m5

<727 UK mI less than but possibly near 727

-l ANL m3

<727

)

INEL m2

<727 Sample H-8 ANL m3

<727 INEL m2

>727 at one end Sample K-7 FRG m6

<727 Spain m5

<727 UK m4

<727 ANL m2

<727 INEL m3

<727 Sample K-13 Belgium m4

<727 FRG m5

<727 UK m9

<727 ANL m2

<727 INEL m3

<727 4

1

24-i l

Table 5.

Summary of results from examinations of metallographic samples from TMI-2 I

lower head (cont'd.).

I l

Sample No.,

Specimen Maximum Temperature Attained Laboratory Number During Accident ( C)

Sample L-9 France m10

<727 FRG m9

<727 UK m7 less than but possibly near 727 i

ANL m2

<727' INEL m3

<727 Sample M-8

{

FRG m4

<727 UK mI less than but possibly near 727 ANL m2

<727 INEL m3

<727 Sample M-11 France m7

<727 FRG m5

<727 Spain m6

<727 UK m4 2727 to a few mm below clad; i

<727 in remainder

.j ANL m2

<727 INEL m3

<727 The cladding / base-metal interface region of Sample E-6 was not examined at INEL because their metallographic laboratory can only work with nonradioactive material and they were unable to completely decontaminate the interface sample provided by ANL.

t Instead, a second metallographic sample from E-6, which included only the completely decontaminated base metal, was examined. This means that the interface carbide layer and the delta ferrite phase in the cladding were not examined directly at INEL.

However, from their examinations of the base-metal sample and from hot-cell photomicrographs of the interface provided by ANL, INEL personnel concluded from their established criteria that Sample E-6 had reached a maximum temperature of i

1075-1100 C near the surface. INEL personnel also inferred a cooling rate of 10-50*C/ min from the hardness values.

i The maximum temperature of between 1000 and 1100 C estimated by ANL for Sample E-6 was based upon several observations. First, a heat-affected zone produced by the weld cladding process is normally present in the base metal to a depth of -8 mm below the cladding / base-metal interface. The absence of this heat-affected zone

}

25 j

i

/

s I

I

. f p

l

m'A g

i07 n-[ '

fh y

?Q$d[f%U2:h.,.q$Mh 14-1 $M

(*-

y'

~

T NY W ~

u ?

~

^

y }hj,,;bf((y& : W Ql;{& 'iTP!;;?,y U {$3 r*

J i

E%

j'W f*>

QW@: i 8'p p.phy 8

' l 100 pm, g' %V.-}g ::

~cau og u

Shg-pj e

.,ca~/..,ak.,;g y m,eey g p ;

e, suum.nu, vm2.

xsr m.:

mam

=

Fig.14. Metallographic specimen from lower head sample E-G showing absence of feathery carbide precipitate layer at cladding / base-metal interface. Note interdendritic cracking of cladding layer in upper half of micrograph.

l t

9 indicates that the base metal reaustenitized during the accident. Since reaustenitization

]

upon heating begins at 727 C and is complete at -830*C for A533, Grade B steel, this j

observation indicates a maximum temperature in excess of 727 C and probably in excess of 830*C. The absence of a feathery carbide layer at the cladding / base-metal interface suggested a maximum temperature in excess of 900'C (Fig.14). For comparison, the Intact carbide layer at the cladding / base metal interface is shown in Fig.15 for Sampic K-13, which did not exceed the ferrite-to-austenite transformation temperature of 727"C during the accident.5 In addition, the prior austen!te grain size in the bainitic microstructure of Sample E-G near the interface corresponded to that produced in the i

Midland archive material by a 2-h isothermal heat treatment at 1000-1100 C. The spheroidization and partial redissolution of the delta ferrite phase in the weld cladding indicated similar maximum temperatures (Fig.16). At 50 mm below the interface, the grain size corresponded to that produced by a similar heat treatment at 900-1000 C.

The observed hardness of 250-2G0 VliN in the base metal suggested a cooling rate of between 10 and 100'C/ min from the austenitizing temperature.

i i

26 N

l De:

p.'

jh hL

?%

m g~) w ;,Q; W :3 %x 2

- n my)! ':" yt>.

QA W

~f np 2 j+g, }n

%y e qa.

,f a e.

W.

W S.IW?ip,&

pf,N[hfeig

?&W{ ?..

t C '*'JP A ' M M M,kf, 4 b

C M ; & PN" Q h:tidbt I$

- ~ & h,.

l

%. j Q,, gy@ 3,

m v

!g y; t:4 *y. 2 ',

& n

.z?! x } Y k b jfQg

• A bh, ;

@f 4},y,u.t e.c y e6 ;; yW ; ;9 ', J J y

in i y*$(l%ig7 O-a 100 pm i Tw i

A#

&:W.D y '

d'

'LJ dmuiann)3 tddiW3M Fig.15. Metallographic specimen from lower head Sample K-13 showing presence of feathery carbide precipitate layer at cladding / base-metal interface. The maximum temperature did not exceed 727*C at this location during the accident. (From Ref. 5).

Sample E-8. Lower licad Sample E-8 was also examined at INEL and ANL. In addition, this sample was examined at the Study Centre for Nuclear Energy (SCK/CEN) l In Belgium,5 the Technical Research Centre (VfT) of Finland,6 the Centre d' Etudes l

Nucleaires de Saclay (CEN) in France,7 the Staatliche Materialpr0fungsanstalt (MPA) in l

the Federal Republic of Germany (FRG),8 and Equiptos Nucleares S. A. (ENSA) in Spain.9 Based upon comparisons with standard microstructures produced in both heat-f treated Midland archive material and samples of lower head material. INEL researchers concluded that the maximum temperature attained in Sample E-8 was between 1075 and 1100'C at the interface, assuming a time-at-temperature of 30 min. They further estimated that the maximum temperature at 45 mm below the interface was ~50-150*C l

lower than the peak interface temperature. A cooling rate of 10-50 C/ min was again inferred. The examination at ANL indicated a maximum temperature of 1000-1100 C for this sample. based upon prior austenite grain size in the bainite and the

{

spheroidization of tin delta ferrite phase in the weld cladding. The cooling rate was again estimated to be between 10 and 100 C/s.

The examination conducted at the SCK/CEN in Belgium revealed the absence of a heat-affected zone in the base metal, spheroidization and partial dissolution of the delta ferdte phase in the cladding and austenite grain growth near the interface (Figs.17 4

and 18). These observations, coupled an the observed increase in hardness throughout the base metal, led to the conclusion that the maximum temperature of this specimen during the accident was substantially above 727"C.

f 27 s

7A ts s

1 c,

e v

1

,J

• g v

g p

t j

e'

[

s 0

j f.

4

?

N 3

e l.

f e

s-

\\

O'.

J 0

0

([

\\

g i

e-C.,, r t

g, f

i k

(

L i

o

~,-d

'1

,j, "3

o c.

e g

,'r h

n s-4 y

6 t.

'y v-

)yn cu ' t.i i

~

7 tr 4.-

s dr.

(

• ' '3 g

},,,f y

+

~ ~'~-

b i

Sg*

O

^,.

C w.

.:4 4

y 4;

e w

c.

'\\>

e,

~

y

,c

ty p

n g f 25 um i

_j i

,, s I

8 c

s,g t.

3 w

s.,

I Fig.1 G.

Metallographic specimen from lower head Sample E-G showing spheroidization of delta ferrite phase in 'lype 304L weld cladding layer.

The V1T of Finland conducted detailed optical metallographic and SEM ex:uninations of two specimens from Sample E-8. They noted the same microstructural features as had been seen in the SCK/CEN investigation and estimated a maximum temperature of 1100'C in the cladding (Fig. 19),1050*C at 2 mm below the interface.

1000"C at 21 mm below the interface. and 950'C at 34 mm below the interface. No cooling rate was estimated, but the study revealed that the cooling was sufficiently fast to produce full hardening through the specimen thickness but slow enough to permit some carbide precipitation and austempering of the bainite.

The examination conducted by the CEN in France similarly deduced maximum base-metal temperatures of 1000-1100*C, based upon the absence of a heat-affected 1

zone the observed austenite grain size, and measured hardnesses. 'Diey estimated the i

cooling rate to be much faster than the 1*C/ min that they used in their simulation experiments, and probably of the order of 50-100"C/ min.

Personnel at the MPA in FRG also observed the absence of a heat-affected zone in the base metal of Sampic E-8. the dissolution of carbides at the interface, and i

hardnesses characteristic of complete austenitization during the accident. They estimated the maximum temperature to have exceeded 850*C and probably 1000'C.

j i

.. ~

i b.

4 l

28 I

i l

l

^

.(,,

h a5 yy e

i

~

c. ;

u i

+

j r

j 7

3-5 l

r;.' [ t 'ngi i

j

- i y

l

c. S.,7...

i s

i i

l t

• t

.c q

1

- t.7 i

i

~>;

_c.

, f[ i w,

l 300 um j

i

,3 I

f l

• at' r

Fig.17. Metallographic specimen from lower head Sample E-8 showing spheroidization of delta ferrite phase in cladding. (From Ref. 5) l l

l The examination conducted at ENSA in Spain likewise observed the absence of a j

heat-affected zone in the base metal. the elimination of the thin carbide layer at the l

Interface, the larger prior austenite grain size in the base metal, and a reduction in the j

amount of delta ferrite in the weld overlay (Fig. 20). From this evidence, they deduced a j

maximum temperature >1000 C.

i i

Sample F-10. Sample F-10 was examined at INEL. ANL. and Harwell Laboratory in f

the U.K.10 Using examination techniques similar to those described above for Sample E-8, INEL personnel estimated the maximum temperature of this sample to have been j

between -1040 and 1060"C at the interface and the cooling rate was again placed at i

between 10 and 50'C/ min. The examination at ANL suggested a slightly lower l

maximum temperature of between 900 and 1000*C. based primarily on the somewhat j

smaller prior austenite grain size as compared with Samples E-6 and E-8. The Harwell i

examination indicated that the maximum temperature had been " considerably above the l

Al" (727'C). based upon both microstructural evidence and observed hardness values l

(Fig. 21).

1 1

?.

7 i

29 l

i

wun y;g;-y

4 m(.w m[m-~.ey r epx.pss, y Qf n%%:,:C-e* lQ', }&]}G W"::f$

l wx.~4tge ';h m$h,($75@%f d W.6 9,fM$ !.' q: - i % g3 M T J.e,[ v.4.. % 4(, e &.k.Y b *I i W,,40 %fW M,R[h~Wh.,& -' g:-- i D@h l a fb T W f,psfb7(Ayfy;9ffgf !54'&' l

in A& ig fsFdW2f5 hSNS$g$<@w$g,?hfE5$$k;/

i ?.d .d t -

p. rm i::L 4-mp2g q K710-n,mu,yp?.4a.h:

ne a u o.- m c]; Pf ;zrsj/.i et 100 pm i g - Q(M .Sc S ,,ag.s J .4., y ) e i T' PzdIM.Jejig. i A, d M { L%e;;r-;2 /$ .h thatbkWfMSbn.wagrg..r i c i l 4 t (a) l i

i i

g,c,W ;cu.g-p-qm c . N,y n;. _ V.r yy cye

w. : p c

.'*em ' '.4 e% p(Wp5+ie l 4%Q:; s.fe \\_. lG,.. y, O .w. y. .y; ;r 7.y,::gm, yypy. , par,..3 '. _ :., z:., y; m,e mwggg 99, :m. 9 g r n 3.s. L. :.C ? y.; a; y,n.. g..N;y.nt fym, dw er (r;; ~.. ~ .g 2 u } ;;p 3. w,y g,7,y;Q g,g,.Q&,f..v;nll%;., - 4; wwp;; ,qn 3s ,} -2, e s ; -., ,,,p y u. Q s ,.;. 7.%r f; y ,,,s a r, g.c, h.o.., w_y,. ,y c __ ,u. ,,e.. 4 p

v :.o c Ka-.Ary.hg;m-Qif 3

c. s, v. ?. x s..T.,1+-,2,, *,v e r; w .., (.1, d)a ': t s,-g o 4, -W h-l. RN n,'g. %,, - q .., J :. 8 / ';,.,g:,.;.M '!. yd pt gh.,,y.g)&s p *N L w% = cl 4.;y*.;f5~z*3l, gy p I,w. gi,,;' . &&l4yf;W) . ~ '. '.

• &,fWf%

ti:Ki

3.,. v.y,.g.l3 Q' gw$

h ,. pr,;..~ 1... w y,., pgA w4 an x. t an

s. m

,.,, rm.y,,y.;yy}.::g. f;%r 9ypy

y m rm.

..a y;a.g#qu ,: yyn

• R.. cW-. g:.;

s: ,# pw F g -; y,h; +ypg., p . g.2 Cyw .p. +. n.c..%,%;y.).;.f' g n. 'ka J

a ;., H'

= s

r 100 pm i n

) 2.j,N'.&,p. > @q,#.&m:W;.syyqpw;,h v: s w ..a.. 3, i io h (b) Fig,18. Metallographic specimen from lower head Sample E-8 showing austenite grain ) growth in base metal (a) I mm below cladding / base-metal interface and (b) 3 mm below interface. (From Ref. 5) 6 g ,-e y- - - - 4 ,r-3,,- -w

l j 30 l i f i a _; t -w, ,7 - n l .t s w; - _+.. .f .'r .[*[.. '. ' L.l h!f . 5 / )I 5,c.

• .<..,,,, ;, p.-_.,

,y.

2~:

l ./. c e.. n; ' /\\%; p/, ',,.;.

• -L,A g ? ' ;

7, +2- , w._.' N. ,w { 3 ', ;,;, '. 1. +

• :-= q ; ;

.9 a ~.-. U,4, r y ^ ":/, g:',y.;qf i[_ u: } m' a., .g ;

, p.:3;. _,

l

.f i ; _,

q b & Q L,b' y 7.,' f., .t .f ?. - " J ',- y, , v, - - 3. * ' :;j ; g K,s p - ;. .f, r-7 x. j p ., +.J,' ' S, g P l ' n' , -;., m - l f. c c, e ' ;..,, u' q L n.s v j' t 2 w c,,100 um, c. t 3,. .'x ,.4..r ; .,..,.u-a ..- r 1 Fig.19. Metallographic specimen from lower head Sample E-8 showing absence of i carbide layer at cladding / base-metal interface. (From Ref. 6) P Sample G-8. Sample G-8 was the last of the four lower head samples that were entirely within the lower head hot spot. Metallographic specimens from this sample were examined at INEL and ANL. Because of extensive cladding cracking and the penetration of these cracks by core material. It was again not possible to obtain for INEL a completely decontaminated metallographic specimen that included the cladding and interface regions. Instead, worker at INEL inferred the maximum temperature of this sample from prior austenite grain size in the base metal. Because the microstructure i was similar to that present in Sample F-10, they estimated the maximum temperature j to be between 1040 and 1060 C. The examination at ANL placed the maximum temperature in the range of 1000-1100 C. The cooling rate was again estimated to be between 10 and 50'C/ min by INEL and between 10 and 100*C/ min by ANL. l Sample F-5. Sample F-5 was adjacent to one end of the "het spot" identified near the bottom of the TMI-2 lower head. While ANL and INEL found that their metallographic specimens had not exceeded 727'C, observations at Belgium. FRG and = the U.K. on adjacent metallographic specimens (see Appendix Al suggested a maximum temperature slightly in excess of 727"C. although the results from Belgium were somewhat ambiguous. b f

31 L r L t + t + -J-I \\._ t. 4..-p .o a t s 3: l' .,..'*y',_ { t N.. ), y _ - -

e.

4 f. ,},, ,q, w ., y u' . - r: 2 i o, .w. .m 5 "J [ -a .,a.- .. c ..,c .., w _. ,i i .,, ' j. e ' ~, r r ~.u a: ~ 0;1mm; i ~ < . E L I h t-(a) i -m,-- , g.._ ...e wf p; f ., i '. b -.. ~<_ , ^,.+c- .p:;p y ,,r .' 3 :ac% 'y ?m'* 5 ;'

• y e..

~y,] Nb [~,,-;'((,(.I.6 ; ] Q, .* q ?,,. . ~ m' *,v :,. [3 ,.,...;v..- QJ. 'n']q. v .,. x;:,. s

s.. a j:.c

..e,g. u. .,s...... '. ~.,; <, 1. a /.h _.,v ..s.,.. d :; r:. +- = 1 ,e ~ q r,, a.,. n. .,.r --t,s , w t: .,n. r.. t. .., o s g (' i < a ~...N,-,,..<,f.._.. .): 4.: -f.- /,. e, i ,... r e ,.e _ p. .. -...,. ts.,

• s,&,,,

. j. -.- g. c4 s, 4. fetg ,s fif ' -~j~ p ~ :42 %y,. ,4 , u. 4 u.- c . i,,, -.x.3 ^ _ eg,4 :f,{ r...* u. e , ~,,,,... - 1,; ,4 , f,..v,. c t + e e . -, mm; O.., e,)M' .e .a 7, '4 ,,.. e,. .,..o w ..= s 1 (b) Fig. 20. Metallographic specimen from lower head Sample E-8 showing fa) absence of carbide layer at cladding / base-metal interface and (b) austenite grain growth in base metal 4 mm below interface. (From Ref. 9) i i l j

3 #.). \\ h. ,6 s ..s s v s ~. ...e,.. ' .s

~.

t .s "<3, i 9 > 7,,,.... -. ; gy.., g, i s,;,i :,, s s

.u

~ .,4. 4.,- n ,a s c,. .~ r ,c;,- ,.. ~ ,. :.,, a,,: r

,+

g, r. < 9.. a (a)

t _ s I

k . v 4. -/ 1 ^ ) .p. A ~. ~ ^ y ~ - i k P .L. a,,. n 1 i (b) i ,.n a v'.u w : ;j /m, ~ y, -,,2 4~ -: t. m p; <... w -~ yg<( ,;y, y. ., w e, n :.~ n,{.;. '. h{,!E*,$'-fi?i,.,' 4 4.-].'v-C' , -- ), .n ~. 7 ,g-p>,.s:,7 g #y;; t.-. y a t w* ty ,4 .e = s..- :, 7 v v *q[:. - Ag., og$. 6.*

s. xg*. A <w~,7 4

--M ~ ~ er ,.. un' g' -s i .>,-r,.,.., m,, - -., gz,-- .v.. v....... p,. .1 v w, ; s j. n.A,,' [ ' ....t." ,f jd. ' / <; ) M y * ,.. ~.,,, ,,,c.. . n.. ,i'* a ( ,.f, .'M+ f,' y. 40c.~. V4^, 4 , v ' ff., g

y

~l- .,,..s.. -,.,..,,,

~

, c,.,. ~. ,.a,,_ ,,.~ .@j.ymi. '. a, e % Y e.- 3 sse e

1. F n..',

, M. :. - ' ey ( s - r,. 2. s, .*~ =

• rw g,r.,

m ' s, g D.m'

• N j..

.y

n t

h;- ~.. ._. *L ; ^ fj'Y :lt --; r. Q-n,, .s .y. >,.y, >-w. pr ;. 9.,: y pg ; y, z.. . -,, s,. e t y. .gr. .r ~M A 1 se e l ..e e [c) 0.05mm l I Fig. 21. Metallographic specimen from lower head Sample F-10 showing absence of carbide layer at cladding / base-metal interhee (a) and austenite grain growth in base metal 1 mm (b) and 5 mm (c) below interface. (From Ref.10) 1 i I i e I

33 The conclusion at INEL and ANL that the maximum temperature of Sample F-5 was (727*C was based primarily upon hardness measurements and the observation that the heat-affected zone in the base metal produced by the weld cladding was still present. Af ter the European laboratories reported maximum temperatures somewhat in excess of 727 C hardness measurements were repeated at INEL on the reverse side of their specimen, thinking that perhaps a temperature gradient might be present through the thickness, liowever, these measurements again indicated a maximum temperature of <727 C. At SCK/CEN in Belgium, a significant increase in hardness of the base metal to a depth of -30 mm below the cladding / base-metal interface was noted. This hardness increase corresponded roughly to a region of coarser, larger grained bainite that was 1 present to a depth of -20 mm below the interface, suggesting that the temperature of the base metal had exceeded 727'C to a depth of 20-30 mm during the accident. 4 Ilowever, the heat-affected zone in the L: 'e metal produced by the weld-cladding process was still clearly visible, as noted in the previous paragraph. As an alternate explanation, the Belgian researchers speculated that the transformation evidence observed in the first 20-30 mm of the base metal may have been produced by some unspecified local overheating after the vessel was heat treated but before the weld cladding was applied. Investigators at the MPA in FRG noted an increase in hardness of the base metal to a depth of -15 mm below the interface. They found that the microstructure in this region corresponded to that produced by a partial reaustenitization (Fig. 22),' and therefore concluded that the maximum temperature was between -730 and 850 C (the two-phase ferrite plus austenite region) during the accident. The metallographic study conducted at Harwell Laboratory in the U.K. also revealed microstructural evidence of partial transformation to a depth of -15 mm below the interface. In addition. increased hardness was observed to a depth of -30 mm. Based upon these observations. the investigators concluded that the base metal temperature had exceed 727"C (but probably not 8502C) to a depth of -15 mm below the clad and had approached 727"C to a depth of -40 nun. It should be noted that the hardness profiles determined for Sample F-5 by the SCK/CEN. the MPA, and Harwell all showed peak hardness of the order of 230-250 VHN extending for distances of 15-30 mm below the cladding / base-metal interface. This contrasts with the hardness profile obtained at INEL, where peak hardness of -210 V11N extended for only -5 mm below the interface. Thus, it appears that the observed transformations were quite localized in this sample, in keeping with its location near the perimeter of the lower head hot spot. Sample H-8. Limited material for metallographic specimens was also available from Sample H-8 and examinations were conducted at INEL and ANL. As can be seen in the initial sectioning diagram for Sample H-8 in Appendix A, the metallographic specimens m2, examined to INEL, and m3 examined by ANL were adjacent to each other at one end of the boat sample. This was the end most distant from the hot spot, j and both samples were found not b have exceeded 727*C during the accident. Metallographic specimen m5 from the opposite end of Sample H-8 was subsequently

t ) l i i 34 i 4 _o I l I i i ~10 ~' ;+ ^- ' N[. t. . M4MD5k$

R.1;.@.p M-*~,t..,%.,. 5,.%v M.. e$@p/

l V s y m .c -.+ n y g phN= ^ a[. _ y

,1 s,y.

,. ~ ;~ \\ -20 -ae.

m. w I

%M.N '.". %.u ps. t d/mm 4w,w..... %- i 'g e;- r u.:6 * ..,.3. s* H W d.. 2 J Ws;- a L. nr =d.<~'?

r ":.

n l ~.s r I polished etched j Icm l % g'. Q g % Q G 2 i g. ~ '* 4h' 'WaJ p SS , M A lit h ~y y?.y. p}~M.py 4.[.y mz . s A Q}.S,a,y# - > r ..,-~,.4.x e. -...4..,.... 1.s,. u_- a

,yw. p+.3 1

.. n. e. .s' s y. u i s ..,.s m r %-: = M p c! 4' m.- ::. h:. b4 , Z~' ',y;;C:g ".,-l*,h. ~-t>kq'-ey

4.1-;5,6,'f[97%,p

i~:.%..

,py cd j D, ;. s &,, _m,. <,,..g. > w ;u. c'.. n%: y + s...j.,,.,, .. g. ~,.- ., p f,F ; 3. ~,..s e .3, i

g

., > w' v.m.st .e., g, y y. crp ,p fr. 3 x-s ..fi , -.'...) m . c. ' c.,", - ,"4 . :f,, l W.pw r Y,w 'r X '

,xp*;,

m' 9-e. wL,y~ut *L k,-:A e e, 'N. r ,':'y.- ~ e.g. s.ynumg g ,x,;*3.,:.a. ',. : ;..;. :?.i, [ '.,..;g3?g'Q. 3g,jWp, re ~ g-r ,. z,. r '.a

.un'X ".C *c

,v~, s. t 9)<.;nQ i ..6, v.,.L .q...,s. 3, < . ;;i w. g.,.j. ? Qp. *,Y ~,.. .py* y 4p s, u, y.*% ;g;,hj.*y .-: h,' L^; : ~7,1C. y%,, ~. ' c...,,. d, ,-z ( *

• g p,,,r; : ) [h ) s v;-!! S.E'
• .f,L;Q d

,$M?fifj,,A l ~ - f.}.,s &y.f.,e $ h *,. kSO'54.;h>1.;-,W.*, fk.Yk

:r...,. flj l

r ~..s-n.:r;, :e... :-W.. n ,r , w. ,,.. e,,.y.. ;e.n o u. m t. d=0mm d=2. Sam t Fig. 22. Metallographic specimen from lower head Sample F-5 showing evidence of l partial reaustenitization of base metal to a Gpth of -15 mm below the t l cladding / base-metal interface. (From Ref. 8) f !l I

a; I 35 t I i i ~.. :,..4 w.>: ;..,, e . x;

m y,4 ~

k'$ I drb ' ? .~ s,,?c.c,&. w w. y.a m,,e,,.a.. +m,.~ m. e.,c A,&sq %re.: s 1 .a. w. .. +, n.e.., m +,. y:..-v :, ,~n.,. +..,.,s. m.:. a.s .,,, p. a ..e w:m.. n y,~a.,;[ +.mxy 4(<...t.,. ;,. gy c .>s cw. .s. v. q,,..x.v. w ; ~ n > v.,.;,y-c nj.. .w~ r

A. v. q, p.<.:,_ p,.. s. g.., w.. ma ? x. e.yfj ( Wl**'.-+

";.:f'm #. w,yg^ we r 9 s. g*[,, y,. .s, m .,w,.., a. :h>.a*- h *oy'd..s.k ,g.: g. %*di .A y 9 g.% L 6: r . ;L%y. w...+n,e ?:=t - s ,... + M;x .x. .: 4. ~ y s e .N- " % Ac ',M.s. h m: w o b.W c l n,os_fNw4..,w~,.5ge:. M A. n m,,;s w: n. * .r. . u...,.L,. ,.,w <, .w. .J v-t

• %g).,O..

r v r a. t -.e w y & R [i Q p y ' Q Q fl f.} f. h. f., t,g,,fhR, ' hd$ .v e

• m,*r -:'."

f,,. .a 4. ?y Wl A.cR' jj.h?.R @ 'Q~ s __ % r ~~pl, f.-. a .~-w. .p Sh. ,4 s Q j + l^ f'@ k -, M.;'.,l. cl.y, f a'..; N.-+ f. - 4 w - h,.wp.y

q. %t. s. [.-4 k y 9.>.,z pWy.y*;*y%, ;g 3.,..~....:q. :-+ v. : g. ~ -

n;. >. -

b V. Q '\\. " [g.p *;3. 3;p'g y j - s ll '. *;m..:% y". - z [' Q N ^ :. :,,a ':. 4 ~. ,.:> y g. .y I E f ] d-7mm 100 pm 100 m d-11mm j ,p c.m :r :, ^ ~:vu- . r - rp-m e ~.- c2m h h ?.bh % L1. w-~ k khe'r .2 c.)g,. p.+A %.,. ~.o,.~... w'.

v.. v.

r < W9* &,+, t c: M$hMNN$;;2'%p.r. >.M, r. N., L.$., M..ff^U @ ..e .s <~.. u.*.M in gG v ' : k 49% ;f ;M ',2 ONO i L &,.:,,m%.' e. y<? %..'., 4,, W... pR,j.,..4.. y /u. g.N

3.. gh,....,

7

c.. o g

+ .w.

5..

.- w .. ~w,, g. g.. ' P..m..g, yv. v-f,. m.; w.<w.,.m y 't c+p,m -.. - ~. a >- m ., p..,. y - 4 i m' y.%;.e.y, : 2-

.c N

k p.. ..s..Qg M...s; . a. . y; O,s,2[g 4 ,t. .. * +q. ~ y. c... +. -, *~t ~ r;' ' g h y; p h ; x4 . Ly, & +. s y

u. ~.

.r..

e. c.

na .t g -:." m e e,~e gp, %:;[4 @W. w rm;m y'g:)[4 , yy s

r n n n w -.,, F..

..C: 4 _sqr s a U l EhN [,[-iz'.; M./;$ M 2G*f ,,]&.~ w(.gigfayg;g;- fna q n gl n Y M .l ~ 4'$.%- h. Q.s

.K' n.2-r,f. MSQQ-&Q

,9. h,,,;;fq% g, :,., L4,y..s. 2 (_f.; '. jj ..L. . fgs :. 3. y.. t. , p.y 2.; - p .. fF ~/ yjg. y q q ,.7 c.3 w.+ g.p,h%,yog,.-.. 9.;u ~- q j. , m .n .orel' x.h. ~. sQ. ~*: ' 9,,:f. 4*. ff~f.U,, W@.:,ly. ~ . Q.o >-) >- j$. y 2 ...l, q.? t Q ;.* ,., _ o f.. ' ' -% .h }l g g 4 g;. Q s . y, 100 pm d-20mm l d-20mm 100 m 6 I

• 1.

L 4 F!g. 22. (Cont'd.) A t e i . ~.

1 j f 3G l i i sent to INEL for enamination to see if a temperature gradient existed along the sample length. Unfortunately, this specimen contained embedded radioactive contamination that INEL was unable to remove, so it could not be examined. l Longitudinal strips remaining after the tensile specimens were cut from Sample 11-8 were then sent to INEL for metallographic and hardness studies. These strips are ~ indicated in the second sectioning diagram for Sample H-8 (Appendix Al as pieces x3, x4, x5, x9, x10, and x15. Ilardness measure 4nents indicated that three of the strips I had exceeded a temperature of 727 C near the end adjacent to Sample G-8 and the hot spot. The observed distances from the end nearest G-8 over which transformation had I occurred were ~15 mm for specimens x15 and x10 and -25 mm for specimen x4. Specimen x15 was from the bottom of the sample and x10 and x4 were from the same l side. As expected, the transformation distance was greatest for specimen x4, which - j was nearest the surface. The orientation of Sample H-8 relative to the hot spot suggests l that the side containing specimens x3 x5, and x9 should have been slightly closer to the high-temperature region. Ilowever, none of these three samples was found to have j exceeded 727'C over any portion ofits length. In any case, it seems clear that the end i of Sample H-8 nearest the hot spot did exceed 727'C during the accident. f Sample M-11. Metallographic specimens from Sample M-11 were examined at -[ INEL, ANL, CEN in France MPA in FRG, ENSA in Spain, and Harwell in the U.it I Results obtained at the 11rst flve laboratories indicated that the maximum temperature { at this location had not exceeded 727'C. However, the examination conducted by liarwell in the U.K. revealed subtle microstructural evidence near the interface that j i suggested that the base metal had attained or slightly exceeded the Al transformation i temperature of 727 C for a distance of 5 mm or less below the interface with tla cladding (Fig. 23). Because Sample M-1I was located -1.5 m from the center of the hot spot, this finding suggests that portions of the lower head away from the hot spot still reached rather high temperatures during the accident and that, locally, these temperatures may have approached or even slightly exceeded 727"C near the interface. i Other Samples. Metallographic and hardness results from the remaining lower i head samples, inclu6 'q results from examinations performed in Italy that were not described above.Il i-cated that none of the samples exceeded 727'C during the l accident. Ilowever, researchers at the Harwell laboratories noted significant tempering of the bainite microstructure in the base metal of Samples 11-4, H-5, M-8, and L-9, suggesting that the temperature in these samples probably approached 727 C, at 1 cast j near the surface. e i 4.2 Tensile Tests l The results of the tensile tests conducted on the lower head specimens are I presented in Table 6 for the base-metal specimens and Table 7 for the cladding l specimens. These tests, carried out at ANL as well as in Belgium.5 France,7 and i Spain,12 were conducted in general accordance with ASTM Standards E8 and E8M, and all elevated-temperature tests were conducted in an Ar or He environment. The l l strain rate for the clastic portion of the loading was 55 x 10-4 s-1, and the strain rate i during plastic loading was 4 x 10-4 s'l il x 10-4 s-1 Tue reported yield strength values i

7_. 37 1 e i a r t Ms 2, s44 . _ m l>,., - ,w V ( [ ,. J ~ + .j ~ y t n w -r> 4 s ;,. (a) -w g g s. a 1 4 + i s' 'g -_ I-I f ,s i -. v' s .. j. e,.... ?- . c s. + -e (b) i .,n 2. 2-1 i "Y d s ;, .. -. ~. ~. .3.,..., . n... 3. i . m.. ....-w.. a. s-1 .... _ ~.. ,;s... c.:..: ,,..:..,.....,...,..., p ~, ,u ,t 4,. g..,., ~,.. .* w;.K. a ; ..g. i o. , ; y;aci,7, c J

,

:, a 1, >.{i;;'.* @-s -.';s.._

...r.,.,. 'e- . a,^ 4 a.o., ~ '.. c

.;;., n,-(,

3.,,- t m;,., _ ;

' y n. .. 'y%p;, ::4 >.,., ?;. pa..+ +- s:;-j ~ :.m;p . *, y.,.0 v., %, Q. * ;,.;.* Cl Y'*,*,' Er ..., % -y c _' ' ' y~'jh,.. ..;% ) *l', '2,.,.

s..-. c, y :,.. 1,N p "f. ; y y :,,7, :-

,i.t)... e .r. 9, . c~, w.,, c.1.w..v.s y,.,. g .,a_ 4 . 3- .,_ n.... ~ gg) 0.05 m m Fig. 23. Microstructure of lower head Sample M-11 at cladding / base metal interface I (a), 2 mm below interface (b), and 6 mm below interface (c) showing evidence ) of partial reausteniUzation. (From Ref.10) I i 1 1 l i 1 J

~ - e 38 i I i were obtained by the 0.2% offset method, except where discontinuous yielding i occurred; in these cases, the observed upper yield strength is reported. ) The base-metal tensile and yield strength data of Table 6 are plotted in Fig. 24 i together with average values reported by the the Japanese National Research Institute j j for Metals (NRIM) for five other heats of A533, Grade B steel.13 The NRIM data were obtained at a strain rate of 5 x 10 5 s-1 up to yield and 1.25 x 10-3 s 1 for the remainder f of the test. The NRIM tensile strength data suggest a strain-aging effect between 100 l and 300*C resulting in a local tensile. strength minimum at -150 C. Both the tensile [ and yield strengths of this alloy are strongly temperature dependent; the room-a temperature values are reduced by more than a factor of 2 at 600*C and by more than a factor of 10 at 900'C. l The data for specimens taken from lower head samples E-6 and E-8 are plotted. l separately in Fig. 24, and these data lie significandy above the best-fit curve to the remaining data. Both of these samples were heated to maximum temperatures of I -1000-1100oC during the accident, followed by a relatively rapidly cooling. The l resulting hardening has produced significant increases in strength at both room j temperature and 600 C. On the other hand, no perceptible strengthening is seen in I specimens from sample M-11, which came from a location where the maximum l temperature may have approached or slighuy exceeded 727 C near the surface. l 4 I i Limited tenstle data obtained on the stainless steel cladding material are reported j in Table 7. Additional tests were not perfonned because it was detennined tnat die f analysts were not planning to include any structural contribution by the cladding layer to l the mechanical behavior of the lower head. d 4.3 Creep Tests l t ) The creep test results are summarized in Table 8 and the stress-vs.-time-to-failure data are plotted in Fig. 25. These tests were carried out at ANL and in Belgium 5 { France,7 and Spain,14 and they were conducted in general accordance wiut ASTM i j Standard E139. The tests were conducted in an Ar or IIe em-ironment except those j j conducted by the SCK/CEN in Belgium. All but one of the Belgian tests was conducted j in vacuum, as indicated in table; a single test at 800*C and 30 MPa was conducted in an Ar environment. Strain-vs.-time curves from the creep tests of Table 8 are { presented in Appendix B. Ilowever, no curves are available for the three creep tests conducted in France at 1100 C. I i Materials with slightly different thennal histories were tested at both 600 and 700^C. - At 600 C, tests were conducted on specimens from Sample K-13, for which the f maximum temperature during the accident did not exceed 727'C, as well as on specimens from Sample F-5, for which the maximum temperature was apparently ' ) somewhat >727eC over a portion of the sample. No significant difference in time to failure is observed in Fig. 25. This lack of an effect may be attributed to the fact that the { maximum temperature probably did n" significandy exceed the transformation temperature of 727'C in F-5. particularly in the bottom half of the sample from which i the creep test specimens were taken. Similarly at 700*C, specimens from Sample { M 11. for which the maximum temperature may have approached or slightly exceeded i

39 Table 6. Summary of tensile data obtained from base-metal specimens of the TM1-2 lower head. i Test Max. Spect-Tensile YleId Uniform Total Reduct. Temp.; Sample Temp. men Strength Strength Elong. Elong. of Area j Country No. (*C) No. (MPa) (MPa) (%) (%) (%) i Room Temperature i Belgium K-13 <727 17 594 414 11 24 72 f France M-11 ~727 17 581 408 11 22 65 Spain K-7 <727 t7 600 426 13 29 63 U.S. L-9 <727 t18 592 423 15 24 67 E-6 -1050 773 650 9.0 16 62 Belgium E-8 -1100 18 778 653 4.5 14 50 Spain E-8 -1100 17 769 633 9.2 18 51 r 600 C I Belgium K-13 <727 t11 257 253 0.8 7.5 72 France M-11 -727 18 239 224 1.2 53 75 Spain K-7 <727 18 247 238 3.2 4d 81 U.S. L-9 <727 15 256 231 1.6 44 91 i E-6 -1050 382 344 4.0 40 74 4 i 700 C i Belgium K-13 <727 tl2 120 106 1.7 77 90 l France M-11 -727 19 146 136 1.6 42 66 3 Spain K-7 <727 19 110 89 4.8 83 87 j ) U.S. 11-8 -727 (4 137 126 2.8 50 86 800^C France L-9 1000a 17 79 44 18 64 43 U.S. G-8 -1050 15 77 52 15 65 900*C j l Belgium F-5 1000a 11 0 49 38 13 43 31 i j Spain L-9 1000a 115 40 29 13 36 27 1000*C 3 i U.S. 11-8 -727 15 30 20 14 42 35 ] France L-9 <727 112 32 21 9 23 23 j 1100 C Belgiu m F-5 -727 17 20 14 13 124 97 Spainh L-9 <727 t16 19 'l1 13 (110) 1200 C U.S. II 8 -727 19 12.0 7.6 12 93 99 Francec L-9 <727 til 18 13 7 >40 99 aSpectmen heat treated by holding at 1100#C for 2 h and cooling to room temperature i at 10-50 C/ min to simulate severed damage. ITest conducted at 1070*C because of experimental difficulties. cTest conducted at Il50"C. J l =

40 Table 7. Summary of tensile data obtained from cladding specimens of the TMI-2 lower head. Test Max. Speci-Tensile Yield Uniform Total Reduct. Temp.; Sample Temp. men Strength Strength Elong. Elong. of Area Count 2y No. ( C) No. (MPa) (MPa) (%) (%) (%) Room Temperature Belgium F-5 -727 t2 553 330 37 40 30 U. S. F-5 ~727 t3 551 322 28 30 34 I100*C Belgium F-5 -727 t1 30 29 0.8 16 14 800 > >. 3. i > i

• i i e i i i i

i i.. g UTS YS 700 e a Tuisamples(<727t) c3 1 O O Tuisamples E-6 and E-8 n5 600 oO NR:M Data (not TMi matenal) m = E 500 o 400 h o---- e ~ O o 300 0-E E 200 m g H 100 N 0 0 200 400 600 800 1000 1200 Test Temperature ( C) Fig. 24. Tensile and yield strengths of TMI-2 lower head material compared with Japanese National Research Institute for Metals (NRIM) data for other heats of A533. Grade B steel.

F e b 41 i e 1000 ..i mm-m m. - a p. m 9.. t~. M m-4..~ .m 1 --

5.. p..

+-- , a ;; q.. +.+ji 4 ...w.. ........y4... 1-- .-o-.. 'E _m.. ... 4...j..$..>.e.j.. _f. . i!!

• -*"".".*."..."_4"

'i = i!' i ..._..i.+..;....b.,.1._...

t. rH..i.1-Mitt:-

=+-H n 100 3 g.:t,C~~:.... 600 CM: ^- r3 b ........a..:.... ..E ',. .a ~+- ,f*'***" } ...........L ......~.-~...~..+--ru.e - - ~ ~ - .700 ~~ ~f - ~~a - + -= ...s,...... u.a..a + .....u._ y '~ 800 C en 10

- 7 gg :,,

-g:=;g

=:ii:_.

900 C!.5

:

• 600*C (Belgium)

~ "T ~ ((" a 600*C (Spain) ' ~ " ' A 700*C (U. S.) 50D~"~' ~- _.- A 700 C (France) w- + 800*C (Belgium) 1 900*C (U. S.) E D 1000 C (Spain) ..E ED ._...:d f "iENE + 1100 C (France) ::T ~~ 'i']; ? !! " " ~ ~

L 0 1200*C (U. S.)

. :1,_ _

..J:7"~~~~t:11::....:: .l::ME" ,,,,ut ,, ;. it , ir i i.u. ,,,,,,,i 4 0 1 2 3 10 10 10 10 10 Time to Failure (h) Fig. 25. Stress vs. time to rupture data from creep tests conducted on TMI-2 lower head material with estimated best-f1t curves, i l 727*C, show no difference in behavior when compared with specimens from Sample 11-8. for which the maximum temperature remained below 727 C. The best-fit curves to the creep data from the lower-head specimens of Fig. 25 are ) replotted in Fig. 26 together with data previously obtained from the Midland archive material in OECD round-robin tests. The agreement between the 600 C data and that obtained in the round-robin tests is reasonably good. However, at higher tempemtures, the best-flt cunes to the lower-head data fall noticeably above time-to-failure data from tests on the archive material. These differences may be caused in part by differences in the microstructure and prior thermal-processing history of the two materials but it seems unlikely that prior thermal history would have any effect in tests conducted at temperatures of 1000*C and greater. Two time-temperature correlations were explored in an attempt to fit the creep data of Table 8. The first of these was the larson-Miller parameter L15 -l i j

42 3 Table 8. Summary of creep data obtained on specimens from TMI-2 lower head. Test Max. Spect-Time to Elong. at Reduct. l Temp.: Sample Temp. men Stress Failure Failure of Area. Country No. ('C) No. (MPa) (h) (%) (%)' l 600 C Belgium K-13 <727 t8a 240 0.2 36 81 i (13a 225 1.0 43 84 110a 155 23.1 37 33 19a 115 39 32 71 114a 115 128 23 22 Spain F-5 -727 114 232 2.47 42 73 t18 221 4.14 57 76 115 194 9.47 47 65 l 116 157 17.75 51 73 tl7 114 92.8 51 41 700*C France M-11 -727 t10 60 13.5 54.4 19 55 20 72.9 111 40 43 41.6 1 U.S. H-8 ~727 117 95.1 1.34 34 85 t16 80.0 3.27 33 82 113 52.1 27.6 73 89 114 41.6 46.0 77 93 11 5 34.5 S 1.6 96 90 800 C Belgium F-5 1000b tl3a 70 0.95 67 44 2 tila 50 5.4 46 31 19a 40 15.5 45 29 18a 30 27 39 23 112a 23.7 111 43. 23 900*C U.S. H-5 1000 t16 35.0 1.09 41 30 t15 26.0 4.55 36 30 114 19.0 18.1 39 45 til 14.8 42.3 40 30 112 9.51 159.5 33 30 1 l 3 r m- - - -

t l 43 l \\ Table 8. Summary of creep data obtained on specimens from TMI-2 lower head { (cont'd.). l Test Max. Spect-Time to Elong. at Reduct. f Temp.: Sample Temp. men Stress Failure Failure ofArea f Country No. (*C) No. (MPa) (h) (%) (%) t 1000 C Spain K-7 <727 111 16.9 1.90 38 48 l t10 11.5 7.54 32 66 l tl2 8.7 29.64 22 73 tl3 6.3 152.8 30 40 1100 C l France L-9 <727 t12 15.0 0.17 9.6 l t11 13.0 3.3 24.1 t10 8.0 4.33 3.3 i 1200'C ~ U.S. M-8 <727 16 9.0 0.98 96 99 15 6.0 7.26 115 99 l 14 4.0 48.2 99 99 i 17 3.4 55.1 81 99 l aTests conducted in vacuum; remaining tests were conducted in an Ar or He environment. i bSpecimen heat treated by holdia_g a 1100 C for 2 h and cooling to R.T. at l l 10-50'C/ min to simulate severe damage. 1 i L = TIC + logio (tr)l. where T is temperature in Kehin, tr is time to failure in hours. and C is a 111 ting constant. A least squares analysis determined that the optimum value of C for the present data base was 12.5. and stress a was related to the Larson-Miller pararieter by the relation logio[o) = 4.3406 - 0.00018767 L. (1) where the applied stress o is in MPa. Figure 27 shows the present data plotted in the form of log (o) vs. the calculated Larson-Miller parameter. assuming C = 12.5. The calculated coefIlcient of correlation r for this fit is 0.98277. The creep data of Table 8 are replotted in Fig. 28 in the usual format. along with the Larson-Miller best-flt curves obtained as described above. The fit is only fair. with the straight-line fits inherent in the larson-Miller correlation deviating noticeably from the actual data particularly at the lower temperatures. i l

44 1000 ._..1 ; o j. .. 1 .......__.4..._...+...;7 round toenn) -.-_._.600 C".~ ^ ,4- -r e 627t(NEL) ~~' ' O 727t (NEL) n .......A....i......E.U...... ..d .... 4 4.+ 4... - + - -.. ~ - -. - +... + = do 700 C 100 =:==== _..:..:.. --,_ = : &.4 _ + - E 32EE.1800oC;. _,... ......:.L +- C c. n. ,a g ...---,..a.+...e 900 C: i O i- ' + Wg ...~~~1000 C' ~~~~ ~~~~ " ^ ~ - ~ ~ ~ ~ ~ ~ ~ G + 4 10 1200cC. m ....e ...i..-... ,7 + O 877"C(NEL) $ 977%(NEL) --.---.L . - - - + - - 4 1100T (INEL) d 'd 1 10~l 10 10 10 10 0 1 2 3 Time to Failure (h) Fig. 26. Best-Dt curves to data of Fig. 25 plotted vs. data previously obtained for Midland archive material in OECD round-robin tests. The Manson-Haferd time-temperature correlation 16 was also evaluated in an attempt to obtain a better Dt to the data. The Manson-Haferd parameter M has the form M = loglo(td - ta ~ T-Ta where tr is time to failure in h. T is test temperature in Kehin, and ta and Ta are Dtting i constants. A least squares analysis was again carded out, and the optimum values for ta and Ta were found to be 7.57 and 520 respectively, The plot of log (o) vs. M for the present data is shown in Fig. 29, and the best fit curve is the second order polynomial. loglo(c) = -0.80467 - 261.41 M - 5291.25 M2 (2) The calculated coefficient of correlation for this fit is 0.99347. A comparison of the resulting best fit curves with the actual a vs. tr data in Fig. 30 shows an improved Dt when compared to the Larson-Miller correlation.

However,

45 2.5 >>i '>i i i>>>i.' i ocboj ~ ~ 1 O C 1 + 2 _ - ~ ~ ~ ' = j y o:

E 1.5 t

y g = c u) O.. g ~ g _g_ _0, o-o 0.5 ~ ' ~ ' - log (c) = 4.3406 - 0.00018767 L 0 8 10 12 14 16 18 20 22 Larson-Miller Parameter, L x 10'3 Fig. 27. Plot of log (o) vs. Larson-Miller parameter (C = 12.5) for TM1-2 lower head material creep data. l systematic departures of the best-fit cunes from the actual data are noted in the 700-900"C region. This problem may be associated. In part, with the ferrite-to-austenite 1 phase transformation that occurs over the temperature regime from 727 to -850*C. 4.4 Impact Tests The Charpy V-notch impact data obtained in It11yl7 on specimens from the lower head are sumnnrized in Table 9, and the absorbed impact energy is plotted as a function of test temperature in Fig. 31. The three groups of test specimens for which the maximum temperature did not exceed 727 C show similar b;havior, with an upper shelf energy of -170 J and a transition temperature of the order of 20*C. However, the i data from specimens of Sample F-10. for which the maximum temperature was as high as -1050*C, stand in marked contrast. The F-10 material shows a significantly higher ductile-to-brittle' transition temperature of -70 C, as well as a lower upper-shelf energy of -120 J. These differences reflect the reduced ductility and impact resistance of that s produced in this material by the high temperatures and relatively rapid cooling associated with the accident. 1

4 46 1000 m. m 3 C OTE2......[..l.. 27 ..l.~22...' ' '.l2. 3... _.......E....'.'[.22...L. .!.2E 2?.p. _a.. m... m, .............m r .T^~' -I2 ......_....-..L.. 4 4.. i........-..... ..........7., ..._....i..;..#... E. l g.. s.. ..i._...;..n.........-..... l. 88 i ~

~m i!

100 7-r r::- s- . v-r ...... +,,, ;,, -60 0"C.M....-; m O. .._..._........4i.4.. .. g...; i...... g _..2_.....:.......g..._...... 1...+... 3.._. ~~700 CtO-m 800*Ci " g f + en .:4 --+ ro. 900*C1,_ 10 r = -~ ~- N-- ~ : r..... _ .~1000CN

1. 600'C (BeIgium)

- _. ~.. -1100 CJ- "-~ 2-a 600 C (Spain) A 700 C (U. S.) ~- - ~ ' - "1200 CE~ a 700*C (France) --e-1-- + 800 C (Belgium) 0 1 900*C (U. S.) E O 1000 C (Spain) + 1100'C (France) 0 1200 C (U. S.) - - - - ~ ~ - - - H- ,,,,1 ,,,,,1 ,,,,,,I ,,,,,,i, i 10'I 10 10 10 10 0 1 2 3 Time to Failure (h) Fig. 28. Stress vs. time to rupture data from creep tests conducted on TMI-2 lower head material compared with best-fit curves from the I;1rson-Miller time-temperature correlation. 4.5 Cladding Cracks Examination of Samples E-6 and G-8 suggests that the cladding failed by a process similar to hot tearing, causing extensive cracking alosig interdendritic boundaries. ' Die precise nature of the loading that produced the cracks is not clear, but it apparently was caused by thermal stresses imposed as a result of the accident, probably during j cooling. Microstructural examination of the underlying base metal and the results of 1 tensile and hardness tests indicate that both samples reached temperatures of -1000-1100 C during the accident and then cooled rapidly. Temperatures of this magnitude would be expected to impose signiflcant thermal stresses on the cladding and base metal during a transient event, reduce the resistance cf Type 308L weld cladding alloy to hot tearing. and cause recrystallizatiori of the base metal, thereby erasing any 4

47 ? 2.5 i i o o o ~ o ooi 2 o o n as o O. ~ o b 2 O o o 8 1.5 4 g o O 3 _o o 1 2/ ~ log (o) =-0.80467 - 261.41 M - 529125 M I 0.5 -0.025 -0.02 -0.015 -0.01 -0.005 l Manson-Haferd Parameter, M Fig. 29. Plot of log (o) vs. Manson-Haferd parameter (ta = 7.57. Ta = 520) for TMI-2 lower head material creep data. evidence of deformation. Although the observed cladding cracking may have been produced by thermal shock during initial contact with the hot core materials. It is more likely that it occurred during the early stages of cooldown when the still-hot cladding layer was placed into tension because of thermal contraction of the underlying base metal. This latter process is analogous to hot tearing during welding, and produces cracks similar to those found in the F-10 sample. Because the cooling rate from the maximum temperature was relatively high (on the order of 10 to 100*C per min), significant thennal stresses would be expected. The cracking at the F-10 location was apparently less severe because the location was only at the periphery of what is believed to have been the hot spot in the vessel wall. The composition and superposition of the reaction layers on the crack surfaces provide some clues to the sequence of events that took place during the accident. The crack surfaces in the stainless steel are covered by previously molten Fe-Cr-rich oxide layers that also contain In and Sn. and Ag-Cd precipitates. The molten In-Sn phase in the grain boundaries of the oxide in the ferritic steel also indicates that there was a j source of the molten material when the high-temperature oxide was formed. probably very shortly after crack fornmtion. Fuel particles were present only on top of these 2 I I

48 -e ......a.j {.w. I._4. 44-..... -.~a-dj- - i ........._.u....... g.... +- ,_.._.i s.- e mi l-i ._.....9..f..... ....;.._q. 4.;.i - L. 4 4

i 100 ru.: -

- n.i.:. Au..... -M g - + - m+- - ::600 C:r : ~ e41. 1:.::A., _~ .n Q. .... _..ry ... 4....; i.. a.. ec ~.._t. 4._.... i;.e. ..a

4......_.;.a. '...

s ............-.....q.5... ...a... .t.;+- 4.a +;..... ..._......A .-700 C---- 7... u> !+ g .. g..... t + 800"C cn + 10 = _.. - .. _ ~........ _. = ,, _ - _..._... :: 9 00 oC=.:=: - u

== r

• 600 C (Belgium)

~ ~...-..- 1.1000 CI: ..-y... s 600 C (Spain) ~ - - + " ~ 7~ ~~"1100 CI" A 700"C (U. S.) " ~ ~ ~ - ~ ~~ T- . 1200 C. _ A 700 C (France) _y } + 800 C (Belgium) 0 1 900 C (U.S.) E D 1000*C (Spain) L._.:2,....__... ._.._.. h b;. ..;.' i-6

. + 1100*C (France)

..,......::l l;... _ :...... _...-...-.1: l 0 1200*C (U. S.) --- - --+ r - - - - T i e ...i,I i *,.i,i e i i + ;, il i i,, i, i, ~1 0 1 9 3 10 10 10 10" 10 l t Time to Failure (h) i Fig. 30. Stress vs. time to rupture data from creep tests conducted on TMI-2 lower head material compared with best-fit curves from the Manson-Haferd time-temperature correlation. I solidified oxides or as very minor constituents in them. The absence of significant quantities of fuel that solidified in situ and reacted with the molten oxides on the stainless steel indicates that the massive flow of fuel to the lower head was not the source of these liquid matedals. However, this massive flow could have been the source of the solidilled-fuel shards found in the cracks atop the solidifled oxides. This fuel would have solidilled when it came into contact with the lower head, if not earlier in its movement from the core region. The flnding of numerous surface cracks and internal tears in the cladding that contained solidified Ag-Cd masses indicates that a molten source of these materials was on the lower head when the cracks formed. It is also quite likely that penetration of these liquid materials interdendritically into the cladding contributed to the hot-tearing of the cladding.

aaa.L L 4L -.m-4---.3- . + - - e-h w,- 49 i ? Table 9. Summary of Charpy V-notch impact data on specimens from TMI-2 lower head. Test Lateral Specimen Temperature Energy Ductility Expansion Hardness i Number ( C) (J) (%) (mm) (HV) Sample D-lO (maximum temperature <727'C) I k3 -20 31.35 10 0.38 179 k4 0 99.82 30 0.75 181 I k1 10 108.26 40 0.76 183 k2 22 117.69 55 0.80 183 i k6 100 170.29 100 1.1 186 k5 200 185.84 100 1.02 180 Sample H-4 (maximum temperature <727 C) k10 -20 66.83 10 0.36 190 k4 0 85.37 25 0.70 184 k8 10 118.82 55 0.70 199 k1 22 127.69 50 0.87 189 ~ k7 35 130.67 90 0.85 198 k2 50 167.83 100 0.98 183 k5 100 173.67 100 1.04 190 k6 200 171.24 100 1.09 192 k12 300 160.26 100 1.04 187 t Sample E-11 (maximum temperature <727 C) k2 22 100.68 55 0.75 186 k1 35 143.38 90 1.02 211 k3 50 135.47 95 0.98 182 f i Sample F-10 (maximum temperature ~1050*C) k3 0 28.00 0 0.23 246 k1 22 53.61 30 0.44 246 i k5 22 40.72 20 0.41 242 k7 35 43.79 20 0.42 250 k2 50 48.88 50 0.52 247 k11 75 99.51 80 0.70 245 k8 100 112.53 100 0.86 246 j k10 200 123.24 100 0.94 243 k12 300 110.43 100 0.90 248 .j ) i

~ 50 I 200 _'l.'\\.'I's's'\\.'e.', 0 U 0 O-l i 9 150 - - ~ o ". o9

• 9 g

^ 8 luo I w

o A

A N 100 0-0 ^ a U E ~ O i 5 ~ A F o D-10 50 ~A .c O s O H-4 .'i A o o E-11 3 A F-10 l 'i'1'l'f'i'i'i* 0 -50 0 50 100 150 200 250 300 350 } Temperature ( C) l Fig. 31. Absorbed impact energy vs. test-temperature data from Charpy V-notch l 3 impact tests on specimens from TMI-2 lower head. l f i These observations on core materials suggest that the first material from the core to reach the lower head was from control assemblies that failed early in the accident. l When the massive fuel flow reached the lower head, this layer of control-assembly j 3 materials would have re-melted and then penetrated the cladding in the vessel hot-l 4 spot region. After the cracks fonned. instantaneous oxidation of the crack surfaces and j j the ferritic vessel probably occurred in the presence of trapped superheated water- { vapor, which also would have oxidized constituents in the liquid metal, leaving the ' j more noble Ag-Cd unoxidized in the liquid oxide. The reacted crack surfaces and the - l Jagged surface of the G-8 sample suggest that this liquid oxide was corrosive to the j stainless steel. The fact that there is essentially no registry of the mating surfaces j across the cracks suggests loss of material, perhaps by a dissolution process. It may be noted in Fig. 8 that the bent-over ligament of cladding in the crack would extend above the cladding surface if it were set upright. This indicates that the cladding surface in { the vicinity of the crack was dissolved away to some extent in a manner similar to that of the seemingly lost material in the crack. The cladding surfaces of the F-10 and FA i i, samples did not exhibit a similarjagged appearance, suggesting that such erosion was l very dependent on local conditions. The fact that the cracks in these samples were not i filled with a solidilled liquid suggests that the liquid was held out by either gas pressure i or surface tension. l ~ f f

51 l The composition of the scrapings taken from the surface of the boat samples j offers no real clue to the nature of the surface material that was originally in contact ) with the vessel. These materials were a heterogeneous mix of fuel and control-assembly constituents, generally in an Fe-oxide matrix. This morphology is similar to ) that found for the scale on a number of the instrument nozzles.18 It could be concluded that the original control assembly debris was consumed into the larger mass of fuel debris that arrived later. 5 Summary and Conclusions I Microstructural characterizations and mechanical tests have been conducted by ANL and the OECD partner laboratories on material from 15 locations in the lower head of the pressure vessel of the TMI-2 nuclear reactor. The microstructural characterizations were conducted by conventional optical metallography, hardness measurements, scanning electron microscopy (SEM) on etched specimens and surface replicas, and analytical transmission electron microscopy on thin foils and carbon extraction replicas. The mechanical tests consisted of tensile tests at room temperature, tensile and creep tests at 600-1200 C, and Charpy impact tests at 300*C. The. specimens tested were taken from locations where the maximum temperature had not exceeded 727 C during the accident and from locations where the maximum temperature had been as high as 1100*C. The results of nese investigations lead to the following conclusions: 1. Metallographic specimens from Samples E-6, E-8, F-10. and G-8 were all j found to have reached maximum temperatures in the range of 1000-1100*C during the accident. These specimens were all from the so-called " hot spot" in the lower head that had been identified earlier. The cooling rate from the peak temperature was estimated to be 100 C/ sed for these specimens. m na 1 2. Metallographic specimens from Sample F-5, which was near the hot spot, were found by investigators in Belgium, FRG, and the U.K. to have slightly exceeded 727 C near the surface. However, examinations at ANL and INEL on adjacent i s specimens did not detect any evidence of transformation, which would have occurred at temperatures >727 C. j 3. Metallographic specimens from Sample H-8 were found to have exceeded 727*C near the end of this sample closest to the hot spot, but the remainder of the sample remained below this temperature. j 4. Subtle evidence of maximum temperatures slightly in excess of 727 C near the j surface in a specimen from Sample M-11 was observed by investigators in the U.K. Five other laboratories examining adjacent specimens did not report i indications of a phase transformation. I a' ) i 5. Researchers at the llanvell laboratories in the U.K. noted signillcant tempering of the bainite microstructure in the base metal of Samples H-4. H-5, M-8. and L-9. Indicating that the temperature in these samples probably approached 727*C, at least near the surface. i a y,-r, c ,w3 e.- - --.,, ._.m-., .w,

52 ? I i 6. The results of tensile tests conducted on base metal specimens for which the maximum temperature during the accident (Tmax) did not exceed 727'C agree well with with literature data for A533B steel and show a dramatic drop in strength at temperatures above 600 C. l 7. Tensile specimens from samples for which Tmax exceeded 727 C showed significantly higher strengths at room temperature and 600 C when compared i to specimens for which the temperature did not exceed 727 C. l 8. Creep tests at 600 and 700 C indicated no significant Iference in behavior between base-metal specimens for which Tmax was of.ite order of 727 C and i those for which it was well below this value. Fifty-hour stress-mpture stresses l were found to be ~8 MPa at 1100 C and <4 MPa at 1200 C. i 9. The stress-rupture data for the lower head material was found to be in good agreement with data previously obtained on archive material from the Midland r reactor at 600 C. However, the lower head material was found to be substantially stronger in creep than the Midland material at higher temperatures. l j

10. The stress-rupture data obtained from base-metal specimens could be more l

accurately fit with a Manson-Haferd time-temperature parameter than a t Larson-Miller parameter. i i

11. Charpy V-notch impact tests conducted on lower head base-metal material

-j noted a substantial difference between specimens from Sample F-10, for which Tmax was as high as -1050 C, as compared with specimens from samples for 1 which Tmax was <727 C. The F-10 material showed a significantly higher [ ductile-to-brittle transition temperature as well as a lower upper-shelf energy i value. I

12. Cracks through the stainless steel cladding of Samples E-6 and G-8 appear to have been hot tearing phenomena, probably assisted by interdendritic l

penetration of liquid Ag-Cd.

13. Materials in the cladding cracks suggest the prese1ce of control-assembly i

1 debris on the lower heat before the massive flew of fuel arrived. t q-l t s e I ? c 2 - - - -.

l 53 i Acknowledgment J 1 The authors gratefully acknowledge the support and direction provided for this work by C. Z. Serpan. E. Hackett A. Rubin, and M. Mayfield of the NRC. The financial support and significant technical contributions made by the OECD partner laboratories participating in the TMI-2 Vessel Investigation Project Metallurgical Program are also gratefully acknowledged. The following persons at ANL contributed to the completion i of this work: T. L. Shearer. D. O. Pushis. F. M. Basso, and S. L. Phillips (sample inspection. decontamination and preparation); J. A. Zic. W. Kettman, and F. Pausche (metallography): J. E. Sanecki and A. G. Hins (SEM examinations); and W. F. Burke and W. A. Moll (mechanical testing). i l t i l I i d. I .l 4 i i 'l il a n. ,--~

= 54 l References 1. G. E. Korth, Metallographic and Hardness Emminations of TMI-2 Mwer Pressure Vessel l Samples. TMI V(92) EG01, Idaho National Engineering Laboratory (January 1992). j 2. D. R. Dicrcks TML-2 Vessel Investigation Project (VIP) Metallurgical Program, Progress l Report October 1969, June 1990. NUREG/CR-5524. Vol. 2 ANL-90-34, Argonne National Laboratory (November 1990). 3. D. Akers, S. M. Jensen, and B. K. Schuetz. Companion Sample Emminations, EGG. l OECD-9810, (April 1992). l 4. D. R. Diercks Decontamination and Examinaticn of TMI-2 Lower Head Samples. TMIV(90) ALO5. presented at TMI-2 VIP Program Review Meeting, Jackson Hole, j WY, November 27,1990. l l i. 5. W. Vandermeulen and W. Hendrix. Examination Report of the Samples of the DTI-2 RPV Received by SCK/CEN (Belgium). SCK/CEN. Mol (March 1992). l 6. Reijo Pelli, Metallographic Examinations of TMI-2 RPV Lower Head Sample E-8 and the Archfue Material of Midland Reactor. TMIV(92) SF01, VIT Technical Research Centre of Finland, Espoo (April 1992). l l 7. F. Le Naour, CEA Contribution to the TMI-2 Vessel MatenalInvestigation Project, N.T. d SRMA 92-1956, Centre d' Etudes de Saclay (May 1992). 8. H. Ruoff, K.-H. Katerbau, and D. Sturm, Metallographic Examination of Dil-2 lower Pressure Vessel Head Samples, TMIV(91) D001, Staatliche Materialprofungsanstalt, j Stuttgart (September 1991). l t 9. L. Pedrero and P. Veron, Metallographic investigation of TMI-2 Lower Pressure Vessel .i l Head Samples. TMIV(92) E002 Equiptos Nucleares S. A., Milia60 (April 1992). l i

10. J. M. Titchmarsh and R. Cooke AEA-Technology Emminations of TMI-VIP Lower Head Samples TMIV(91) UK2, AEA Technology, Harwell (September '991).

11. A. Masperont and P. P. Mile 11a, Metallographic Emminations ofArchive Material (Midland Reactor) and TMI-2 RPV Lower Head Samples D-10 and H-4, TMIV(93) 101, ENEA. Rome (April 1993).

12. M. J. Callejas Cano, OECD Bff-2 VIP Program Tensile Test, TMIV(92) E003, CIAT, Maddd (February 1992).

l i

13. Data Sheets on the Elevated-Temperature Propentes of 1.3 Mn-0.5 MO-0.5 Ni Steel i

Platesfor Boilers and Other Pressure Vessels (SBV 21, NRIM Creep Data Sheet No. 18B. National Research Instjtute for Metals, Tokyo (1987).

14. A. Ballesteros Dff-2 VesselInvestigation Project Creep Tests. TMIV(92) E004 Tecnatom, S. A., Maddd (May 1992).

15. F. R. Larson and J. Miller, Trans. ASME, vol. 74, pp. 765-771 (1952).

16. S. S. Manson and A. M. Haferd. NACA Tech. Note 2890 (March 1953).

., ~, a ,.e,. ~ -

55 l7. P. P. Mllella and F. Digagit. Charpy V Testing ofSpecimens of the Dil 2 VesselInwer Head. TMIV(92) 101. ENEA. Rome (May 1992). t

18. L. A. Nelmark. et al.. TMI-2 Instrument Nozzle Examinations at Argonne National i

IAboratonJ. TMI(93)ALOI. February 1993. 'I E I o 1 ? i i 't l l l i l I ) .I i l r ] l 1 i

I,. e e 1 56 I L I ( e i e B s J W 'l y i I l 4 i a, I l I i 1

P 57 5 I I Appendix A: 5 Sectioning Diagrams for TMI-2 Lower Head Samples i t i P i f i 4

--~ --.-.----- -...--....-, -... - .....--...,nn,.~. =.a. ,,s ... ~... a u.a., .-awe I l 8 i 4 i I l 1' 58 I 4 i f i ? l h 1 4 1 4 9 a t i I a 1 i i I j 1 I i 4 I 1 4 5 l l I i + e 1 l l 1 i !t k i . i l 1 I l i 4 l i l l I i 1 5 i e i 1 1 i, l 1 1 I ? 1 F a ..y _,_.r---,_m- .,_.,.,.y,.m,_,. -...,,.m,.. %.....,_,.y- -,,w-wen-y- - -y.. 4

3 P ? og Sample No. D-10 h + 1.2.7.7 i.2 .8 s.o r I l 1 1 A 8 l L = 15.8 cm Iil i I 3.o I w .;3_ V 8 W = 6.5 cm I l l I I a l H = 5.6 cm I i l i I I E l l l l l 1.S cm DIA l 8 t= 0.3 cm n I i i l l I E 5 L r m g W r = a 5 d 7 9 l I d 1 I i i h_ \\ l I I I i 12121 i I Il \\ \\ l 0 \\ \\ I i 1,1,1 I I l n I \\ \\ i I H lilli i I l imp cispecimens n e a g g E. 1 I15lyl I l I g g 54 W 1 11 I I g gI V i iI i I s si i X in 6.2 O 4 7.9 5 P Specimen identification numbers stamped on this end

60 Weld Clad Layer 1.7 mm Saw Cut 3.9mm \\ / =4 / 10.4mm x.3 o io a., o.30 g2 o-io k3 x-3 (hah) (ray) (hah) 10.7mm x-3 D-10 k-d D 20 k-5

• -4 (haV)

(RaV) After first base metal slice, dimensions 10.7mm x-5 o-io k4 x-6 include a 0.28 mm (Ray) electronic wire cut i x-7 ! 4. A2. Locations of mechanical test specimens cut from Sample D-10. i b e

4 3 9 d. 9 9 1.C 2.2 1.0 1.t

• 10.2 9

Sample No. E-6 E S d i.y i lyi E l L = 16.4 cm si ? I!si ei g g g (cladding cracks) W= 7.1 cm g l > l.. I ^ 402A-1 # .._ l E H= 6.7 c. -i i i i w n n r e / 'lN 'N3$8d O t= o.e c~: ? I?l ?! ? 8 402A-IC m (to INEL) m C cn a m W. r L ~ =

i 9

V 1I w ( l i i i archive 1.0 d l l g l n

r

~ i l l l tensile specimens for I6 preHulin y tests at ANL H l I I L_ T-0.4 p l l l p 9 1.0 i i I L---------- eE Note allow an Y l l l heat-treatment specimens 0.7 m additional =0.35 6 ent material loss ,/ d 4- 0.9 ~ per sawcut m-1 (metallography to INEL)

3 Ta .? 9 9 9 Sample No. E-8 + 1.5 .7 7!" 8.57 5 1 3 18 * $.o P k l I I L = 15.1 cm i i i i i i liI dk S W = c.o-4.6 cm IiI E W e.o Iiiiii 4.s E H = b 2-4.0 cm IIIIIi Iill l l V e t= 0.3 Cm II 7 1 m W l m 0 L W r m = d9 y M g i i i i.1 I I 1, ( k N il j e-l3lin9ls$l $l b E $l E l l o e 4.0

y T nsii specimens g

t H s.2 f a g l gl I I ' ' ' '- 9 3 E j Specimen identification numbers stamped on this end Metallographic sample cut k from m-1 and examined at ANL -,,w wa e, ,w,

==. - - < - - -

63 Weld Clad Layer 2.0 mm Saw Cut 5.0mm \\ x-1 lE8 t-1 l E-8 t-2lx-2[ 3.4mm "-3 E-8 t3 E -8 t -4 4 3.5mm 3.Ea t-5 E-B tE/, 3.5mm ...-s... t.7..x 4 E.8 ..=. 8 i 3.5mm =-7 E8 t.s 3.5mm "9 9 x 10 EB l 3.6mm , E.8 i.io, ,.ii A i ma2 Af ter first base metal 15.8mm slice, dimensions l include a 0.28mm electronic wire cut y Drawing shown looking down on stamped ends of test specimens 7 Fig. A5. Locations of mechanical test specimens cut from Sample E-8. j I l I 1

i h 8; l c j s Sample No. E-11 1.5 l*-3.8M 1.1l.h &l + 1.2.7.7 6.5 g l l t i II g L = 15.5 cm i ii l ii I+ I I W = 6.3 cm III ili l i I g g H = 5.5 cm iii l i i I II I t= 0.5 Cm 1.5 cm DIA 1 II I l l 8 w

r W

x t r en o = 5 a 1 N 7 [ _ 1 I d i 1 _1 I / / i.- I di i 12121 I I / / lil { l 4 I I I-E III I / / II I [ l H IilSi i / / I I impact Specimens 1 IYidi I / / Il w / i liI l Il g / E 'I I II l II a Specimen identification numbers stamped on this end 9 e, ,e --r-m .m., -.av- . - + - m,-_.m-e., m--,m-. --__--_ _..-mm .c- -..-

65 Weld Clad Layer 1.7 mm Saw Cut 6.3mm ,4 10Amm r.1 E-11 k-1 E 11 k2 E 11 k3 x-2 thav) (haw (haly) 10.7mm x-3 E-11 k4 E-11 k-5 x-4 After first base metal slice, dimensions 10.7mm x-5 E-11 k-6 x-6 include a 0.28mm electronic wire cut x-7 Fig. A7. LocatJons of meenanical test specimens cut from Sample E-11.

9 r = u j s s r n e e b m m ) C ic nu T 2 n e n 8 i s it d p o Ao I a n Dp ic e t a f mt i s t s ci a ni h at

7. n i n d

o 0( n o ed me 7 1 hll i ' i I I I I I I I p i m l l i i L 6 i I I I I I Ss cem pa t 4 i ~ 1 li 7 . l l i ' I 1 l I I Ig I I i i 1l l 1 I I l I I Ia 1 I 8 !s5 . i i Ll l l I I l I I 14a1 i l 6 l l l I + i 2 I !g$ l~l jI *: i 1 l Gl l l i I 7 igl l l 1 I l l 7 l il I l ' i 0 2 2 i 1 i I l l I m _s 1 3 r 1 J H 7 5 m m m F c c m c C W .o 8 7 7 5 'y AJg N s 6 5 0 r e = lp = = = m L WH i a S ? 5. 95g g* S E E k g m @ 0 d :: M $n",$ mm@ '

67 Weld Clad Layer ] 2.0 mm Saw Cut ) F-5 t-1 F-s 12 F-s t3 x.2 6.6mm I I. 7 isg 3.5mm

c..

i4 F.s i.s F.s 3.5mm x3 Fs t-7 F5 14 3.5mm x5 F-s t -9 F-5 t10 x6 3.5mm F-5 t11 F5 t-12 I 3.5mm ,F 3 i-33 F.s

i. i 4, 3.5mm

\\ ',, _Fs i.is _ _x;8,, .x- '.o 3.5mm x x-9 F.s i.is 3.5mm x-11 x-2 2 After first base metal i F-s

i. i 7 3.5mm F.s i.is include a 0.28mm I

b electronic wire cut x-13 16.9mm Drawing shown looking down on y stamped ends of test specimens i i Fig. A9. Locations of mechanical test specimens cut from Sample F-5. l i

3 P h 9 0g Sample No. F-10 + 1.1.7.7.6 m 5.72 5.72 5-.8 0 iii l i a 8 L = 15.7 cm l Il l 1 l z I E W = 0.1 cm I III I i l H= 5.3 cm i i ! I i i E III I I l l s t= 0.3 Cm a lil l l l Er il if W s L = = = 5 d V { A __ 1 1 Li i l __ A y i12121il i 12 e iil l l l j g i i mpact specimens i impact specimens i i g H liit S l!!$l l I l i lkIthll l lh ~ 7 I ll i I I T. O m J Specimen identdcation numbers P stamped on this end I .g f

69 i Weld Clad Layer 1.7 mm Saw Cut 6.4mm x4 10.4mm x.1 F-10 k-1 F 10 k-2 F 10 k-3 x-2 thav) (hav) (nay) 10.7mm x-3 F-10 k4 F-10 k5 x-4 (reject) (hay) After first base metal slice, dimensions l include a 0.28mm l 10.7mm x.s F-30 k4 x.s electronic wire cut (,,g) x7 i Fig. Al l. Locations of mechanical test specimens kl through k6 cut from Sample F 10. l

70 Weld Clad Layer f 1.7 mm Saw Cut 4.3mm \\ / x4 2 10.4mm x-1 p.30 g.7 p.10 k-8 F-10 k-9 x-2 (hah) (hah) (ANL) 5 -4 10.7mm =3 F 10 k 10 F-10 k 11 (hay) (hay) i After first base metal Slice, dimensions 10.7mm vS F-10 k 12 x-6 include a 0.28mm (k^Y) electronic ylire cut x-7 i Fig.A12. Locations of mechanical test specimens k7 through k12 cut from Sample i F-10. ) i l 1 j i 1

3 9 h P og Sample No. G-8 +3 8* 1.9 l6

• 89 o

? la la l j.o 13 il I 5 L = 16.8 cm m ~ a n ga. l' n W-cm o l l (cladding cracks) a E H= cm I i i E ? 9 9 7 a t = =0.6 cm 8 8 = i c.8 ; ! -r8 c. i c. c. i-- g er . -l<-r 'I r l \\ E ' 408P-1-D2 r L r 5 5' W 5 g II 1f p r ( l 1 i ardllve N I I I tensile specimens d M, i l lI -0.7 t i g H I I I S. l l lI I T Note allow an l l 9 additional -0.35 I Il E cui material loss per sawcut m-1 (nietallography to EG&G + % 0,4 o> Speamen identificaton nurnbers stamped on this end

..... -.. ~.. 72 i r 3.5mm oe ii c4 i2 { 3.5mm c.s i.a c4 ia 3.5mm c4 is c4 is i 3.5mm os i.7 ce i.e 3.5mm oa i.o 3.5mm ,ci i.io 3.5mm c4

i. i t AL n-1 12.9mm.

After first base metal { j slice, dimensions i include a 0.28mm Y electronic wire cut Drawing shown looking down on stamped ends of test specimens i t Fig. A14. Locations of mechanical test specimens cut from Sample G-8. i i b h [ I 1 7

3 P 3 P Ol Sample No. 11-4 s.72 4 s.72 r f.e + .5.e.7.7 - h I i i i y g i L = 15.6 cm ,1Ii I I, 'I W = 6.1 cm i I I, g ,III E H= 5.2 cm ilI i I Il E. 1 I I I s t= 0.5 Cm I l 3 'III I I, 8 un 0 i L r W e e b V 5 1 Ii 1 i i l _ L. a jlylyl}l l l7ly 27 si I II ll. I.I a t is H I til y impact specimens I impact specimens I i 5-143l3 I II-INE $ I l$1b l 5E 7 il ii l iI a 7 Specimen ident:ficaton numbers .O stamped on this end -~ . _ ~,. - -

74 8 Weld Clad Layer 1.7 mm Saw Cut \\ [ 5.0mm x4 x-3 H4 k1 H4 k -2 Ha k-3 (!la9) (haly) (ANL) 10.7mm x.3 g4 x., g4 x.s x_4 (!!a!y) (Italy) 10.7mm x5 H4 k4 x.6 (!!aly) x-7 After first base metal slice, dimensions inc:ude a 0.28mm electronic wire cut Fig. AIG. Locations of mechanical test specimens kl through k6 cut from Sample H-4. i

75 Weld Clad Layer 1.7 mm Saw Cut 6.4mm x-8 10.4mm x-i s4 57 H4 wo H4 w.o x-2 (Ray) thav) (ANL) s 1 10.7mm x-3 H4 k 10 H4 k 11

• • 4 ghaV)

(ANL) 10.7mm x-S H4 k-12 x6 (tay) Af ter first base metal ,.7 slice, dimensions ) include a 0.28mm electronic wire cut i i i i Fig A17. Locations of mechanical test specimens k7 through k12 cut from Sample 11-4. i l A

h ? 5" l Sample No. 11-5 + .6.6.7.7.7 2.1 .9 M 8.6 5 0 lI i' l A (_ = 1 5. 5 Cm ll l l l 3,3 I llly4 W =._6.2 cm lk E H = 5.6 cm iilli I cE I ll 11I t= 0.4 Cm 1.5 cm u, I R IIIII dia. E 3 W L = g g r- = 5 g 1r y l l , ( J LLLJ l ll-- I A I ilil212111 I II O E I I II III I II 4 i ss

r I

i H I I 81&l i I I I Tensile Specimens I i 4gl313III I I IIIEI I il l I 7 II I I l I I 5" h Specimen identification numbers P stamped on this end G m w a m -- wa.m

77 l Weld Layer 1.1 mm Saw Cut t-2 f " (yk;t o contaminated i Hs -i s.s 4.2mm 3.4mm x-3 ss ta s.s is _,x a, 3.6mm x-s g.3 ,.7 x.s g5 3.6mm x7 ss 14 s5 t-9 x-8 3.6mm s-s t-t o s.s

i. i t 3.6mm

.( s-s t-12 H5 t-13 /,l ...-i o.. 3.6mm x-9 ss t.i 4 x 3.6mm x-" s-s t-1 s x.12 x-13 "~ ~'" 3.5mm s-s t.3 s - x-14 3.5mm s-s t.17 l After first base metal x 1s slice, dimensions include a 0.28mm 16.1 mm electronic wire cut i I Drawing shown looking down on stamped ends of test specimens 1 Fig.A19. Locations of mechanical test specimens cut from Sample H-5. 4 l i

s P h9 C ,s Sample No. 11-8 + 1.2 .7.7 1.s 1.016 4 8.8 .e 3 l l 1.i I 5 L = 15.1 cm III i i i fD 1. w lll gll l-i. W = 6.7 cm [s I,h i I H= 5.9 cm i i i ll I dia. II I ~ 8 t= 0.5 Cm a l II II i 5 3 a if L = = W r = a

j u

9 - 1. I L 1 i L.1 1.11 i -- to r ~ v e g I i d E l El l/ / [ El ls 4 I II( / II I l H Il111 h / II TensHe Specimens l 15lll / II le E. I l 'l*l. / II l' k 7 I Ii iI I l Specimen identification numbers / -g stamped on this end i i ...ma .__..___m._._m___...m..-_..,_-_-_______________________._-_______.m___-.__-__-___.___m___s___u --,_--am_ e- _ _., --. -v 4-w

{ 79 Weld Layer 2.5mm Saw Cut 3.8mm \\* I "* t1lH4 t2-l H4 t-3 l x-2! 3.4mm kH8 [ t -4 H4 t5 He t4 -3 ...*..'d.. 3.5mm .."..... He t-7 H8 t-8 3.5mm x5 He 19 - He t-10 84 3.Smm x-7 He i-ii se i.12 x8 r 3.5mm He i is H8 i-14 3.5mm hHe 11f/ t-15 H8 3.5mm .=.... He (-17 x-10 9 1 3.6mm ";,,', H-8 t-18 ,*l,12 3.5mm x-13 H4 t-19 x.14 3.6mm H ,t e i.20j After first base metal t j' slice, dimensions " S include a 0.28mm 14.5mm electronic wire cut y Drawing shown looking down on stamped ends of test specimens Fig. A21. Locations of mechanical test specimens cut from Sample H-8. i f a

3 i P i$ P o s Sample No. K-7 + i.i.7.7.7 w 8.26 .6 .9 .4 1.5 M a I i i i l I 8 L = 15.8 cm i III I i i i VJ m I Iii i l i I h IIII l l l 1 H=G3 cm I l l i i l l I E I III I I i 1 M t= 0.4 Cm a i lIi i i l I ti s .7 1.2 F = W = 4 L r 5 d I y A ( _i il I i i l 1 9 m 9 N ca g. it?lMI Isi el si s 1 1.I I ii l i t n g H liljil Tensiie specimens ii l i E. 13131I I il @ l I eel $1 l l l l l I $v V iiIi l i I I Specimens identificaton numbers M stamped on this end 9 w w --,,--r,-w-- ~ - - < ~

---

.--.-. ----------...---------.--- a-

1 I 81 i Weld Clad Layer j 2.4 mm Saw Cut \\x1 x7 t-i ! x-7 t-2 x7 t.3 5.3mm x-2 kk7 t-* l k-7 'S lK7 24l[ 4.5mm 5.0mm 2.8mm / 3.2mm 1.9mm \\ s 3.3mm 3.4mm x3 x-7 i-7 x-7 ts x4 1.9mm 3.6mm x' K-7 t9 K-7 t 10 x6 3.5mm \\x-7 t-ii x.7 t-123 3.6mm x7 K-7 t-13 x-8 3.5mm x-9 x7 t-14 x-10 3.5mm x-11 K-7 t15 x 12 3.5mm =13 x-7 t-is 3.5mm , K-7 t-17; x-14 After first base metal 1 slice, dimensions

1. 1-5 include 0.28mm" 16.2mm elecuonic wire at Drawing shown looking down on 1r stamped ends of test specimens Fig. A23. Locations of mechanical test specimens cut from Sample K-7.

5 J b

I n5 i$" cl Sample No. K-13 + 1.2 .7.7

• 8.26

.6 5 .9 .9 4 I i l I iI h L = 15.6 Cm I I I I II l I'I W = 6.7 cm 1I I I II I -a E H = 6.0 cm i i i lil I u liI II i 1 ~ t= 0.5 cm a l l I i l i I 8 m + 4.7 4 E E L r = W r = 5 N 7 ~ d k d 1171 71 71 11111 I s Ii11 II I h H IJilil Tensile Specimens I,,1 l S 14 3131 I _d $' l l IEl 14 1 ? E 9 II II I I l 'l l E Specimen idonnficacon numbers stamped on this end. n ~, = - - - - - ,-e, n -~ +.,

t e 83 i Weld Clad Layer 1.1 mm Saw Cut ' ',I K '3 i.i l x.is

-2 l x.is 5.2mm i.3 x-2 3.4mm

\\ / x.13 .4 K.is -s K.is ts 3.5mm hx.ia i 9/,.' I 7 x.is is x ta 3.5mm x-3 x ia -io x.is i.ii .x 4 3.5mm "5 K.ia i.22 x-13

i. i s

1. 4 3.6mm f

x 13 -14 x ia i.is i.i7[ 3.6mm x.13 iis 3.5mm .8 g.i3

i., 8 x -9 3.5mm

". '.o. t.19..x. 21 x.i3 3.6mm ',2

• - ' 3 x.is i.20 e M ban mal 3.6mm x.is i 2i slice, dimensions b

include a 0.28mm x.14 electronic wire cut 16.8mm y Drawing shown looking down on stamped end of test specimens Fig. A25. Locations of mechamcal test specimens cut from Sample K-13.

s P ise ol Sample No. L-9 _.- i.s .7.7 i.s ..e.5.e

• 82s r

I l l ii i i C 5 L = 15.8 cm I ii 1111I W = 6.8 cm i II IIIII g E H= 5.9 cm iIi 11iii w t= 0.5 Cm a a ll I liliI 8m 1.3 .5.6 I W r = L r g N V 5 A i i i i i i I,1 A [ i 1712Ii171217131 4 t I.I. I IIIII ? H lilli iiil l Tensile specimens o em 1313 1 1219 El eel lII 21 l I l E 7 I I I i! IIi 1.0.8 l specimen identification numbers 9 stamped on this end. 1 9 4 m 4 .-,c, m -i-% ...e- ,.-, -....i--.- + y-m--.r-w-. .u.-

3 N } // g IMAGE EVALUATION / O o g \\///' / 5 TEST TARGET (MT-3) A f ff a y

• %V4gg

,jo ,sg, l I l l.0 n n. ^~ l 1.1 ll _ ;8 I

• l.25

dIA

, 1.6 waa I i 150mm p Ir4 6 l 1 y s, +p,A css o ,e 4f q\\ // . -gs \\ -U g O- {,L'- -t / q 0' '~' M ,}

c,- !O f4 e ,t IMAGE EVALUATION [Q/// TEST TARGET (MT-3) c '3 I/ eg)f* g ' 4(P 2g ^ 4g,,/yy a k' 6 x: s.e s I.O 22 1 l 1.1 - 20 = l $u;m.8-1 l lm,=I.25 IWi 1.4 il 91.6 = um e um l -- - - 150 mm - - - - - - - - - - - - - - - - - - - I f' 'N 6 rf, 9t 7 3 ,,s8 v %,hy&'y f;;/4 n% / c, / L + +-

J ,, n s 7O -7 IMAGE EVALUATION O 'q b ?s, ///p TEST TARGET (MT-3) %g %g,a p- & $+ 1 1.0 n ( 20 R I.8 1 - 71 1 l E l!u!l.25 ILIA j"l! i.6 ,== u see ;;w. l l l i s o m,,,. _ - -. -. 4 3 .w l %g 1;p,,,A s_ \\ . w, x gy qs ' x //dx ,7 4 sp r;y m

85 l l Weld Clad Layer i 1 1.9 mmSaw Cut 1 \\x-1 L-9 t-1 L-9 t-2 L-9 t-3 x-2 5.0mm 3.4mm \\ L-9 t-4 L-9 t-5 L-9 %l l kx-3 3.8mm L.9 t.7 t.9 t-s_,,,* f,, 3.6mm x-5 t-9 t-9 L-9 t-10 x-6 3.6mm x-7 L-9 t-11 L-9 t-12.x-8 3.5mm L-9 t.13 t.9 i.34 3.5mm .< L.9 t.15 L-9 t-16j 3.5mm x-9 t.9 t17 x-10 ~ 3.5mm x-11 t.9 t.18.x-12 3.5mm x-13 t-9 t.19 x.3 4 3.6mm L-9 t-20 After first base metal h slice, dimensions include a 0.28mm ,,3 3 electronic wire cut i 16.0mm F Drawing shown looking down on stamped ends of test specimens Fig. A27. Locations of mechanical test specimens cut from Sample L-9. 6 -y I i 1 L

m5sk 50 O j g Sample No. M-8 + .7.7.7.7 4 8.28 r m 4.4s - a a U - l i l L = 15.7 _cm I lIi i w a W = 0.5 cm o-1 III l E H= 5.6 cm I iii I E I I' I I I t= 0.5 Cm L i m 2 I III I C 4 o + aw = W r = L = g g 5 Gl V 2 = ( ii.I i i i y d ~ .- y cu n ilelsIsi I -O IIil I t a i H I tilD Tensile specimens i uncuiMateriai i e-2 o m a 8.

• l cl "I J l

3 tt z l wI l< El I m 7 II I l-I [ Specimen identificaton numbers P stamped on this end i e e .-.v,--... o, -,v. -r- ,-2,+.-..mv.+.-wv.m.-~ -w w s,..-++w<-,~,...m-v-,, ,--wm 1.,.--- ,-._w,rw ..w,r, .-w,,%4w, em-..-*-.--n-, -,-.,'.w..wm w,e -+w.,-w,-w-,v.ww.--..w,+,m-1,,,,.-.,,--,w.,---y

87 Weld Layer 1 l 1.0 mm Saw Cut I u4 n l u4 t.2 l ua t-3 7.4mm 7 i 3.4mm x-3 u4 r4 u.8 i-s

1. 4 3.6mm x-5 ua i4 u4 i.7 x6 3.5mm x7 M8 t-8 M4 29 x-8 l

3.5mm ua no u4 ni 3.5mm Au8 n2 us u s/. 3.6mm x9 x-io ua 2 24 3.6mm x 11 us os x.12 x-' 3.6mm us os ,.3 4 1"* "#2 After first base metal b slice, dimensions x-' S include a 0 28mm 14.8mm electronic wire cut ly Drcwing shown looi ing down on stamped ends of test specimens Fig.Id9. Locations of mechanical test specimens cut from Sample M-8 l 1

m5 DP O g Sample No. M-11 + 1.2.7 .7.e

• a.2e r

s.e.e 1.4 + [D I I i iti !i L = 15.6 cm l III II II m E W = 6.7 cm - i o-l l l l 11 II E H = e.o cm I i lI il iI cE I III li Il g t= M cm

g i, C

bm 8 W = L r s' 9 J.7 g y 9 d __ l i I I II II d i lElElil lilI121 i 1Il 1 ll ll t H Ijijii Tensile specimens Ii iI g 8. I 21 al l lIl!l a .s a.

• a 151$1s1 l l FI M

7 IIII l-1 Il X c y .( [ Specimen identifcation numbers stamped on this end. 4 ~, -, ,,m, m- ,w .. -. ~. <e -,..~-- w-r v-- ,..-.1-- ,..,.~.+ - - w..%.....,.,..r..r,.,,-.-,,.,-v. .,w

89 Weld Layer 2.0 mm Saw Cut i 2 l u-u -3 ,.2 5.5mm 1 un n-un 3.4mm lu.n it u_ n t-s u.n i4[ 3.5mm \\x.3 _ff u., i .7 u.i t

-s 3.5mm f2

,,[ un i.9 M-11 t-10 3.5mm x-7 x;8 u.n i.u u.n i.12 3.5mm u.n i.33 Tu n. ',\\u.u os/j 3.5mm ns u.u 3.6mm x-9 uti t 37 x io 3.5mm X" .x-i2 un os 3.5mm X-13 M 1i -19 x-14 l 2 ' After first base metal )L slice, dime;tsions include a 0.28mm i 15.7mm electronic wire cut J t V Drawing shown looking down on stamped ends of test specimens d Fig. A31. Locations of mechanical test speci~. ens cut from Sample M-11. l 1 l b f 4 I i

g_ _, a w 2 a., _.. m.m u,_am. 24. .A2. ) I e e 90 L 1 - i .. { b I t t i ' I + t t hi 6 + f ? f i f f e L x L h h t f 4 i I t

• f f

I a o 1 ' Li l ? e i a ) i. r a, t l t b. -{ k s ' i r i o . j

n z..

c& -l O O 5 _E i o o3 o i r1

• 9 i

@w t t i' t i ? o O W i i I.i. o 4 o i m i t .t i i i n oO v= ,w i 2 t 4 o o ,i l t 4 . ~. .e. i o 4.g....g o. m o. a o 4 t Oo i i I oo o d d do Q, n a n o _a o i o 2.= (%) N011V0N073 j e oN .. v - oN O 2 a 5 to u e u= i C.

• T, t

= W Eb OV o .j bo-i M =E C v o C..= 1-- < F i i l t h

.,de ,a.--. ~ 6 J l 94 i - i w 4

a. O =-

U d j . Z.. I cD i oo 3 h ao O O. t e Wd i ? I I O O-n { Q g h O ^ ~ D. ~ _ i O ~ N. 1 I e O y g e

• Z i

j j v i O W ~ La. ':- E ~ f l O i -- p y* i ) O i n-i 4 eio n. -i e-I N 2 --I O pw-g 7 N gg N N

==- > i W. : 4 Q h o O-o.. O i i 6 i i I C i 9 es O O O O O o1 T M N @ 2.: (%) N011V0N073 l CO l N j vM 6 i I ..d

• mw i

- C 0 =: o m - C l C. W h. A. .UD O HU" =G Gv l crjU C. t j

b

. v i O-

n l

ze6 i e o t r .e E t aC O l E9 m ma i 1 1 i, _c N i 6 i o W e a i s t 4 i n ,,, = i. -. m w. t 3 - __ o i i. i O.cm !s _- c >= * : v,,i.. n. == > : e x< .o iO' i j o. I r 5 O i i i a a i i o o o o o o o e n v n n oo c-c2~ (%) N01.1.VONO 13 o 1 g 1 ..Dn q 9 h 8"A 4 6m t ed eo Q Y Q N o E =O [ 1

c. W h.

- c a + h 'O v C o ** a =E E v Wo C. I"* < b i J I i J e i t i t !l I I ; gm 9 J 1 3.m 1 Ou KTig l I. e

.VB oN:

mt y nr eo im co 0 eb 4 p ;i S1 S 1 3 j' L ~ 0 !t 3 .I !)' 5 r 2 ) 0 (H t 2 E M I T 5 r i e

iII tiI 1

0 ,!I 1 s 9 s T e_ 3 s 1 K il 5 y jl -lt!kl$ I 1 0 C 0 5 0 5 0 5 0 5 0 a 4 3 3 2 2 1 1

• 0 P 0M RV$s<Eod 6

h 5 9 1 3 e1 r u:

e t

sr asu r el e r i pt a mSF ed o Tet i l e tspm epi TAT 1 4 t 4 lij i' f A b s,Fm .1 Ee- =-,2-. --si '=..6A-a,,- -+hd 5 46 d.-h+S-- Ann,.- -s-- A-2 A + 6 9 l 97 j f; 4 i esf l 7d5 xpn d T> .e6>0 Z.. 1 CD oo c= w t -L R O N ^- a e - e M e g t i O CN t O i I C -O I O i w O n "Ccc i, a d -= O n i 7 a v e i "OW f '"* ~~-? g2 f I i N i O ~* g 4 _A a. -~ C v f C 5

w g

--+ E i .e

• j 2 e

w } >-

• i en A..d

.l..~. O ~- CN I M ad i:U o i 10 i i Io 7.. O s -w E m-t j I 'l t O i t i 6 O e o i 3 1 0 0 O O O i oy= N N c 7 .. h3 m (%) N011 YON 073 i be m ao ed ee y N bas E$ =

c. e =c u

" W k. c I UO O bo~ u =O=% v I WU C. : b<b s 5 t -t ~ Test Temperature: 600 C Specimen No.: F-5, (14 Applied Stress: 232 MPa laboratory: teenatoin, s.a. Tirne to Fallure: 2.47 h Spain STRAIN (%) 40 .35 o o 30 0 0 8 25 o 0 0 0 g 20 9 0 9 0 o 15 0 0 0 0 10 o 0 0 O 5 o O O 1 L O 0 0,5 1 1,5 2 2,5 ELAPSED TIME (h) l l ..-~.._,........-.--,_-,---,.-,,,......,---.-,.-,.~.-.-w. ..----..4 .... -. - ~. _, -._,,.r._ ...m.....-,

---

..c

t Test Temperature: 600 C Specimen No.: F-5. (18 Applied Stress: 221 MPa Laboratory: teenatorn, s.a. Tirne to Failure: 4.14 h Spain STRAIN (%) 60 0 50 o o 40 0 0 0 e 30 e 9 o 20 O o c, o ,0 0 10 o (' o oV 0 0 O 1 2 3 4 5 ELAPSED TIME (h) m

m m Test Temperature: 600 C Specimen No.: F-5 t 15 Applied Stress: 194 MPa I;1boratory: teenatom, s.a. Time to Failure: 9.4 7 li Spain STRAIN (%) o 40 8 0 30 o O 0 0 0 20 9 o o nO g.. 0 10 o 0 0 o g 0 0 0 L t 0 2 4 6 8 jo ELAPSED TIME (h) 6. 4 m m.

m.. m am

_m, m--...w ..w-u e-e. ,_ww.....r.m..--_s ._.r,.--.-t---- w.m wem-...% .-4.wsmet-wn-.~,.......,. .,...,.e.,-w

.t Test Temperature: 600*C Specimen No.: F-5, llG ~t Applied Stress: 157 MPa Laboratory: teenatom, s.a. Time to Failure: 17.75 h Spain STRAIN (%) 60 50 0 0 o i O 40 o o o o o 30 o' o o 0 0 o 20 go o0 o ',' ' ,, o 0 10 n 0 o n o c, 1 --t J 0 0 5 iO 15 20 ELAPSED TIME (h)

Test Temperature: 600*C Specimen No.: F-5, t 17 Applied Stress; 114 MPa laboratory: teenatom, s.a. Time to Failure: 92.8 11 Spain i STRAIN (%) 60 0 50 0 0 40 - 0 0 0 l3 30 9 n'~ o-20 oV-O ,O U 0 10 0 'n 0 o 1 0 n 0 1 i 0 1 0 20 _ 40 60 80 10 0 ELAPSED TIME (h) t ____...________.__._______.______.__1.______________._________m.____z_____.___________________=_________.-__, _m____._._.___-..m_..-_._.,_~___m_..-..-..-

e F Test Temperature: 700 C Specimen No.: M-11, (10 Applied Stress: 60 MPa I;1boratory: - CEA Time to Failure: 13.5 h France a l 55 i i i 50-4 5-- 40-- e ~~ 5~~ ~ t O l to 1 u 3 0-- a 9 e 2 5-- t / 20-- + t i 15-- 10-l 5-0 l l l l j 0 2 4 6 8 10 12 14 'dur6e (heures) .2..

d i Test Temperature: 700*C Specimen No.: M-11,19 Applied Stress: 55 MPa I;1boratory: CEA Time to Failure: 20 h France 80 i 70-- 4 60-C f 50-o- 1 5 u a 40-- 9 e 30- / r i 20-- t 10-0 l l l l l l l l t O 2 4 6 8 10 12 14 16 18 0 22 dur6e (heures) mm '___.,-mm ...,,. --v u ..s -..--e-<- -.m

• w,

.,v n.---,w r - = -. .---ew .-=u-- -n' w ,__u---__---2--

Test Temperature: 700*C Specimen No.: M-11. t11 Applied Stress: 40 MPa Laboratory: CEA Time to Failure: 43 h France 45 i i i i i i i i 40-l 35-- f 30-- f ~ 1 o-m u 2 5-a 9 e 2 0-- 15--- 10-- i 5-0 l l l l l l l l 0 5 10 15 20 25 30 35 40 45 dur6e (heures) l

106 m b 3 cs ,'t _ s. = ze 'y < :D Z b. =ua N jj

....g

...g...>;. .i y_2 c m _< to. _c-v O O, O 0 ~ O 0:O oOq., N. E F oa wg O CD. _m., O o uj o O O . O... ?o i,,,,i.,,,i,,,,i.....t c, pp o o o o o O in v co N r-C= O sc ~~* n_,.n uu-6 ji ; y %'uie.11s Supeau! ug ee2 o e =e

c. -

C D 2. Ut o Hv-E O e m C. C U C. b H<H

l l: oN s e G t I a ( t S 8 d 1 e Lt 1 i Nn

oAU N:

y nr eo mta i r ..~ 5 co eb pm SI 4 ) h ( o O o e oO m 3 i T o o d e s p o -o a 2 l o E o o e' i; i oi 1 0 ( Ca 0 0 0 0 0 0 P 5 4 3 2 1 00Mh 7 7

0. 2 0

1. _Em3aGCC@ s.OCU 3

e8 r

c L u:

e t

sr a su r el e r i pt a mSF ed o Tet i t l e spn eph TAT

Test Temperature: 700 C Specimen No.: 11 8, (13 Applied Stress: 52.1 MPa laboratory: ANL Time to Failure: 27.6 h United States 70 ....i.... ....i ....i.... o 2 60 L j g g v i 2 ,d 50 m> ca 40 a i C

• C 30 m

m c 20 G C w 10 O O 0 00 0 00 0 5 "" * ' ' ' O 0 0 5 10 15 20 25 30 Elapsed Time (h) I I m. ___.________.__,_.m. _.,.m_, ,,,,,..L_,.. ,,.,,,,.,__.,_._....r.4

2 se t 4 a 1 t S ( 8. d e - Lt 1 i 1 Nn .AU oN: y nr 0 . : 2:.- 5 eo

4.*2:.i :~

i:-

mta l r eo eb d ) pm SI 0 i 4 0 ) i-3 h ( e o m iT o 0 i' hi i ,l 6' 2 O' O' O' 9 0 4 i: i 1 o .: F.*

? :-

F:.r::d0 0

0 0 0 0 0 0 0 0 8 7 6 5 4 3 2 1 Ca

• P 00Mh 0

7 66 C.$ m cCC$C.mCU

1 4

c e4 L ru:

e t

sr a su r el e ri pt a nSF r ed o Tet i e l tspm epi TAT

4 o s m e t 5 a 1 t c S ( 8 de Li 1 t 1 m Nn .AU o N: y nr eo mt 0 a ir 0 co eb 1 pA SI i. . ?s 0 8 ) h ( e 0 m i:l l !ii-6 i T d e sp 0 a c e l 4 E i-i' l .ji 0 2 o o0 m Ca 0" P 0 0 0 0 0 O 0Mh 0 8 6 4 2 7 G. 1

5. 1 4

8

e3 r

1. c,mu5ect@c5co m

u:

e t

sr t a su r el e ri pt a mSF ed o Teti e l tspm e pi TAT

.. ~.. F i .t i ~111 i L r M '~ C s ,.l.' -.Q : 8 na CC ~ ~~ v I .. > O O. E..D e M2 U ~C v~ o u-O'- O. 2 y) i 4 I i i O 4 g l i i O q s 6 O 3 O ~ c. fi 1 / 3 + i n oI -gv y l 3 H i O -.i g 3 I .j O l ..i_ i i g. .t O . g-._.! g d - n. -., ;o >= W' ~.. L w>

u 1

.1 O

O

.o !. m j i O t O, i 4 a a a i a i 4 oc-O O O O O O O O oC~ h LD C v O N e

• 2 E!

(7) N011YONO~i3 l .. o e .t u> 6 m y 7 IA L C@ O O O. *w"" h.C I EO .t 4 O O b 'O o *~ -5E W X o .C H<H i f

Test Temperature: 800 C Specimen No.: F-5, il 1 Applied Stress: 50 MPa labomtory: V.I.T.O. Time to Failure: 5.4 h Belgium l -['j[g-AO-i 800 C 50 MPA 30-7 v z 940 20_ O f 10-- -t 0-r i .0 1.0 20 3.0 4.0 S,0 6.0 TIME (H) e 1

i [ l 113 t O .. = '? O E

n. p n

-i D

• > =

c z.. =b l oo C* .= 2 u8 o oe. a m.q 1 1 e r l ~

• t u

e - i t N A i os w 2 i i. N -o i-i -e i i i4 A L- ~v

7..i i. o

' a :o i M N j --+ i= .I i o l i o o o o o o o o o v n m i o - ~- i o i '1 (%) N011 YON 073 1 o e. e ..o~ Ov W==.... p "E g> =u V o C. ~m C E * '* H v ~o Ot s9u mWc v-F<H

Test Tempeniture: 600 C Specimen No.: F-5, 18 Applied Stress: 30 MPa bibomtory: V.I.T.O. Time to Failure: 27 li 13c1giu m 40 i FS< 35-us[T8 es3 i

• M C 3Q EPA 30--

I i 25-C Gi a v 2 0 -- p 407 j$_ _._..._._......,p. C- _J taJ j Q.. ..9......_..-_. S-0- i g i -} i- - 0 5 10 15 20 25 30 TIME (H) g ,~..,,.v,__.y ..-[%.. c.,_,,+%.g..,_,, r,,,,_w-.--,.4.' ..,,,....w,., .m .,_.--m,,,, .-,,.>,.c e.,~.7 ,#.,_m ,,-.w-w..... ,.f.-_.,_,,._,-~._..,-m. ,c ..,m. ...,,,.e

Test Temperature: 800*C Specimen No.: F-5, (12 Applied Stress: 23.7 MPa laboratory: V.I.T.O. Time to Failure: 111 h Delgium 40 F5-TI2 35-n2jos '9'**'*" 30- + n " 25 - v z 20-- l-l+ E /-- 9 15-w 10-- - + - - / n-- 5-y / 0--- 1 1. I i i i i I r 1 0 10 20 30 40 50 60 70 80 90 100 110 timo (H) l l l l

Test Temperature: 900 C Specimen No.: 11-5. (1G Applied Stress: 35.0 MPa Imboratory: ANL Time to Failure: 1.09 la United States 50 i 'u i 8 40 8 ^ ^ 8 .C g 30 ~ m g C ~ c: e 20 + q) t .E ~ ~ c) go c o n w 10 o o O'''IL ''f O 0.2 0.4 0.6 0.8 1 1.2 Elapsed Time (h) m b= m m _e-, . -, - - + - - -.., i.,.,,--.--.-.. ,-w + ,2-. +. -. -, 3

s se 5 ta 1 t 1 5. S d e 1 Li t 1 Nn

.AU oN:y nr e o mta i rco eb 5

pa SL ~- 0 i 00-4 0 0 O ) h ( o ,1 l 0,i 3 e O m iT O o d e O s 9 0 2 p i' 0 a lE 0 0 0 i ' 0 1 10 0 O O' 8O CaP 0 0 0 0 0 0 00Mh 5 4 3 2 1 9 5

0. 5 6

e2 4

) g C N$ OCCeDc.OCJ ru: U e t sr a su r el e ri pt a mSF ed o Tet i l e tspm epi TAT

a

m s

e m 4 ta l t l 5, S d e 1 Lt 1 i Nn

.AU 3

u oN:y nr eo mta i rco 0 eb _ ~ pa 2 SL O m 6 1 ) h ( n l 2 e V m 1 G T i n o d 0 e 0 s o-8 p n I o a O - lE o -O0 0 '9 4 i 4 00 0 0 0' 0' e _ ~ ~*' 50 Ca 0 0 0 0 0 0 P 00Mh 5 4 3 2 1 9 1 08 9 od 5.mbn gCcmmC-QCW 1 e1 r c u:

e t

sr a su r el e ri pt a mSF ed o Tet i l e tspm epi TAT 4,

Test Ternperature: 900*C ""I Applied Stress: 14.8 MPa ^ T2ne to Failure: 42.3 h United States 50_....i....!.... ,....i.... _ g 40 i cc b 30 u) ~. i C c e 20 o o e o o o C o:o 00; j g 9 o w 10 oo s*# 8: '''''''''''''''''' 0 O 10 20 30 40 50 Elapsed Time (h)

Test Temperature: 900*C Specimen No.: 11-5, 112 Applied Stress: 9.51 MPa Laboratory-ANL Time to Failure: 159.5 h United States 50 i 1 1 g 40 ..c G 30 ~ 4 w o" g i c 'c m 20 e .c w C tu 10 t ~ f 0 0 40 80 120 160 200 Elapsed Time (h) a r-,--_.,r.w.._, ,.,.,_.--..m.- .+v,,,.m,, ..,.,mm r.n_...,- ,...i-,..-w..-,, +, -... _ - -. ,y _.,,W

6 Test Temperature: 1000'C Specimen No.: K-7, t i 1 Applied Stress: 16.9 MPa Imboratory: teenatorn, s.a. Tirne to Failure: 1.90 h Spain STRAIN (%) 40 6 o O 30 o 0 o') 0 G ~ O O 20 O o c> h. ,.3 10 O ,,y b l------- 0 0 0,5 1 1,5 2 l ELAPSED TIME (h)

Test Ternperature: 1000'C Specimen No.: K-7, (10 Applied Stress: 11.5 MPa laboratory: teenatom, s.a. Tirne to Failure: 7.54 h Spain STRAIN (%) 35 30 - ,o e. t) O 25 o 0~ 0 G o o 20 n es - 0 15 0 9 10 O ,.s 0 0 2 4 6 8 ELAPSED TIME (h) b m - - -. - -, ~.. -. - -., -. -.... - - - -.. -..,. - -,,. .. _, _..,.. - -... -..., ~, -.. ~... .-..-r . - ~.

Test Temperature: 1000 C Specimen No.: K-7.112 Applied Stress: 8.7 MPa laboratory: teenatom. s.a. Time to Failure: 29.64 h Spain STRAIN (%) 25 ~' o s. 20 0 o (N O O 15 0 o o O O 10 O c. 4 4. 5 o O O t 1-------- i 0 5 10 15 20 25 30 ELAPSED TIME (h)

Test Temperature: 1000*C Specimen No.: K-7,113 Applied Stress: G.3 MPa Laboratory: teenatom, s.a. Time to Failure: 152.8 h Spain STRAIN (%) 0 -30 0 0 G 0 25 o 0 0 2 0 I 20 9 0 0 0 o 15

c. 0 0'

10 ,O o (, ~

y 5

0 9 0 0 0 i i i i i i 0 0 20 40 60 80 100 120 14 0 160 b .n <-rn. --..-,-.--w~, n-,-~-,~-, -,-~,,-,---.---------------.*-+--+x - - - ~,.

---

.----w

5 Test Temperature: 1100*C Specimen No.: L-9, t 10 Applied Stress: 8 MPa Laboratory: CEA Time to Failure: 4.33 h France 4 i 4 ftuage de l'ecler A 533 grade a 4 1100*C sous 8 MPs 1 3.5 -- = 3 -- a 2.5 -- L tJ 1 G 1 o 2 - e 0 m e " 1.5 -- t 1 -- f f 0.5 --- 0 l l } l l l 0.000 0.500 1.000 1.500 2.000 2.500 3.000-i tenps ( heures ) _ _ _ _ _ _ _ _ _ _.. _ _. _ _ __ _. _ _ _ _ _ _. _ _., _. _. ~. _ _ _. _ _......... _......

l i 4 "wC D se t 6 a 1 t 8, S d e M Li t Nn

.AU oN:

y nr eo nt 2 ia t r 1 co eb t pa SL e a I 1 r= ,:!! ij!!!!i;i OO OoO. a ? g e 8 h ,i!!' i] 0 I e e m a iT g e d 6 e l 0 s 4 p a la E s 4 ..ii=

!l

,1 +!' 0 a 0 i 0 e 2 ,8!> !lii!' 0 a a 0 0 0 0 0 0 0 Q C 2 0 8 6 4 2 0 1 1 0aPh n M8 $.c@"o gemDEccu 2 C 1 t c 9 00

e9 ru:

e t

sr a r su el e ri pt a mSF ed o Tet li e tspm epi TAT 4

Test Temperature: 1200 C Specimen No.: M-8.15 Applied Stress: 6.0 MPa Laboratory: ANL Time to Failure: 7.26 h United States 120 _,,,,r,,,,,,,,, ,,,,,,,,,i ,,,,,,,,,,,O i i t -O O 100 -+ o +o-g .E 80 8 -(f) g e.* g 60 q s .c m o .E 40 3 cn cw 65 00, 6o 20 4O OO O 0' O 00 ~ O O 0 +- ' ' 9 i O'I'I'L 'I'I'''' -0 1 2 3 4 5 6 7 8 Elapsed Time, h l m

Test Temperature: 1200 C Specimen No.: M.8, 14 Applied Stress: 4.0 MP Laboratory ANL Time to Failure: 48.2 h United States 120 ,,,, c ,,,,n 1 ~ 100 0-f 9- -~l- .c 80 -L L y ~ cn w Cn 60 m c .c m e A .c 40 jsj9 g c 0eV w 00 p(p1 20 Og;30 00 OOOO oOOOWgpO 4 ' ' 0 ** fol O 10 20 30 40 50 Elapsed Time, h o a

e Test Temperature: 1200'C Specimen No.: M-8,17 Applied Stress: 3.4 MPa laboratory: ANL Time to Failure: 55.1 h United States 100 i,,,,,,,,,,,,,,,,,,,i i ~ .o-80 4- [ g g g

5 as 60

. - ~ w cn CD C c . k.. k m... ...-4.. ..~ ....q. .E en C [),J i 1 20 i o #p -- - l c ~ h co # l ~ 0 $ # PI'l' 'I'I' O 10 20 30 40 50 60 Elapsed Time, h ..}}