Text

TMI-2 Vessel Investigation Project (VIP) Metallurgical Program.Progress Report,October 1989 - June 1990
	ML20028H286
Person / Time
Site:	Crane
Issue date:	11/30/1990
From:	Diercks D; Argonne National Lab (ANL)
To:	; Office of Nuclear Regulatory Research
References
	CON-FIN-L-1005 ANL-90-34, NUREG-CR-5524-V02, NUDOCS 9011290028
	Download: ML20028H286 (79)
	v • d • e

- _ _.

NUREG/CR-5524 ANL--90/34 Vol. 2 TMI-2 Vesse: Investigation Project (VIP) Metallurgical Program Progress Report October 1989-June 1990

- l'rcluted by D. R. Diercks 1

- Argonne National Laboratory Prepared for U.S. Nuclear Regulatory Commisalon 9011290020 901130 PDR ADOCK 0500 go

I i

0 4

AVAILADluTY NOTICE Awdtabiltty of Reference Matores Cited in NRC Pubhcations Most documents cited in NRC publications will be avaliable from one of the following sources:

1.

The NRC Public Document Room, 2120 L Street, NW, Lower Level. Washington, DC 205%

2.

The Superintendent of Documents, U.S, Government Printing Office, P.O. Box 37082, Washington, DC 20013-7002 3.

The National Techrincal Information Servke, Springftcid, VA 22101 Although the listing that follows represents the majority of documents cited in NRC publications, it is not intended to be exhaustive.

Ref erenced documents available for inspection and copying for a fee from the NRC Public Document Room include NRC correspondence and Internal NRC memoranda; NRC Office of Inspection and Enforcement bulletins, circular 6, Information notices, inspection and investigation notices; Licensee Event Reports; von <

dor reports and correspondence; Commission papers; and applicant and licensee documents and corre-spondence.

The following documents in the NUttEO series are evallable for purchase from the OPO Sales Program; formal NRC staff and contractor reports, NRC-sponsored conference proceedings, and NRC booklets and brochures. Also available are Re0ulatory Guides, NRC regulations in the Code of federal Regulations, and NucIcar Regulatory Commission Issuances.

Docurnents available from the National Technical Information Service include NVREO series reports and ter hnical reDorts prepared by other federal agencies and reports prepared by the Atornic Energy Commis-sion, forerunner agency to the Nuclear Regulatory Cornmission, Documents available from public and special technical libraries Irve.lude all open literature itomt, such as books, journal and periodical articles, and transactions, feceral Register notices, federal and state leglsta-tion, and congressional reports can usually be obtained frotn these libraries.

Documents such as theses, dissertations, foreign reports and translations, and non-NRC conference pro-ceedings are available for purchase from the organization sponsoring the publication cited.

Single copics of NRC draft reports are avaliabic free, to the extent of supply, upon written request to b3 Office of information Resources Management, Distribution Section, U.S. Nuclear Regulatory Commission, Washington DC 20555.

Coples of industry codes and standards used in a substantive manner in the NRC regutatory process are maintained at the NRC Library,7920 Norfolk Avenue, Dothesda, Maryland, and are available there for refer-ence we by the public. Codes and standards are usually copyrighted and may be purchased from the originating organ 12ation or, if they are American National Standards, from the American National Standards Institute,1430 Broadway, New York, NY 10018.

DISCLAIMER NOTICE This report was prepared as an account of work sponsored by an agency of the Unitod States Government.

Nolthor the United States Govemment nor any agency thoroof, or any of their employcos, makes any warranty, exprosed or impliod, or assumos any logal hability of respons!Dility for any third party's use, or the results of such uso, of any Intormation, apparatus, product or process disclosed 6n this Ioport, or roptosents that its use by such third party would not infringo privatoly owned rights.

NUREG/CR-5524 ANLc90/34 Vol. 2

't T MI-2 Vessel Investigation Project (VIP) Metallurgical Program Progress Report October 1989-June 1990 Manuscript Completed: Septernber 1990 Date Published: November 1990 Prepared by D. R. Dicrcks Argonne National latboratory 9700 South Cass Avenue Argonne,IL 60439 Prepared for Division of Engineering Omce of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555 NRC FIN L1005

7-TMI 2 Vessel Investigation Project (VIP) Metallurgical Program by D. It Diercks Abstract This report summartzes the work performed by Argonne National Laboratory (ANL) on the Thtee Mile Island Unit 2 (TM12) Vessel Investigation Project (VIP) j Metallurgical Program during the nine month period from October 1989 through June 1990. During the reporting period, a series of heat treatment experiments on archive material from the lower head of the Midland ncelcar reactor has been completed, the resulting microstructures have been examined, and hardness values have been determined. The results have been compared with the predictions of published continuous cooling transformation diagrams for A533 Grade B steel.

Round robin microstmetural characterizauons and mechanical tests on the archive material have also been completed by the participaung Organisation for Economic Co operation and Development (OECD) partner laboratories. Good agreement was gencially obtained in these evaluations and tests, although one laboratory obtained somewhat different results with circular and rectangular cross section test specimens in room temperature tensile tests and in 600*C stress rupture tests.

Decontamination of samples from the nil 2 lower head is underway at ANL; the procedure being utilized is described in this report. Detailed microstructural and scanning electron microscope examinations of Specimen E 6 have been carried out in an attempt to determine the extent and cause of cracking observed in the clad surface. Metallographic examination revealed that the cracks extended through the stainless steel cladding, but continued for only =3mm into the underlying ferritic steel base metal. Extensive secondary cracking of the cladding, not connected to the surface, was observed in one metallographic section. Iron oxides and resolidilled control assembly constituents were observed on the crack faces, but only a minimal amount of fuel, in the form of small fragments, was present.

The results of microstructural observations, hardness measurements, and tensile -

tests indicate that the base metal in the vicinity of the crack attained a maximum temperature of -1000 to 1100'C during the accident. The molten fuel apparently did not penetrate into the cracks and interact with the base metal, it was tentatively concluded that the cracking of the cladding in Sample E 6 probably occurred during the early stages of cool down after the accident, when the still hot cladding layer was placed into tension because of thermal contraction. This latter process is analogous to hot tearing during welding.

111

___ ____.. ~ _ _ _ _ _ _ _ _ _ _ _ _. _. _ _. _ _. _.. _

Contents Ex e c u tiv e S u m ma ry............................................................................................................................ I 1

I n t r o d u e t i o n..............................................................................

...............................5 2

Dackgmund..................................................................................................................................5 3

11 eat 'IYeatment Experirnents on Archive Matcrials................................................... 7 3.1 Su mmary o f 1Ieat Treatm ent Experimen1s....................................................... 7 3.2 Results of Microstructural Examinations and I lartiness Measunsnents......................................................................................... 1 0 4

Evalua tion of Archive M a t crial........................................................................................... 2 2

- 4.1 Micrmtmetural Characte&ation......................................................................... 2 2 4.2 Summary of Mechanical Testing Program....................................................... 2 7 4.3 Results of Tensile Tests Conducted at ANI..................................................... 2 9 f

4.4.

Results of Round Robin Tenstle Testing Progmm........................................ 32 4.5 Results of Round Robin Stress Rupture Testing Program........................ 32 5

Decontamination of Samples Recovered from the TMI 2 Lower llead............. 40 5.1

' Sample Description................................................................................................... 4 0

- 5.2 L Sample Decontsmination....................................=

=..................................42 6

' Examination of'IMI.2 lower IIcad Sample E-6......................................................... 4 3 1

6.I'

' I n i t i a l Exa mi na t i o n.................................................................................................... 4 3 6.2

. Macmhaninms Mmsurements............................................................................. 4 6 6.3 Metallographic Examination and Microhardness Measurements.......... 50

' 6.4 Elemental Analysis...................................e

.....................................................50

- 6.5 '

Examination by Scanning Electron Microscopy.........................,.................. 56 0.6 Te n s il e Te s i s............................................................................................................. 6 2 '

6.7 Discussion...

=....................................................................................................62-References...........................................................................................................................................65 i

Y 1

4-

_, _~- -...,

,A..,

J

l Figures 1

1. Generalized Representation of Isothermal Heat Treatment for Archive f

Matcrial........................................................................................................................................8 l

2. Generalized Representation of Transient Heat Treatment with Normal Heating Rate of ~40 C/ min, and Three Cooling Rates for Archive Matertal.........................................................................................................................................8

3. Generalized Representation of Transient Heat Treatment with Rapid i I ea t ing Ra t e for Archive M a terial...................................................................................... 9

4. Continuous Cooling Transformation Diagram for A533 Grade B Steel.

Dased On Data of Amano et al. and Pelli et al............................................................... I 1

5. CCT Diagram.h Saarstah) 20MnMoNi55 Austenttized at 875'C........................12

6. Microstructure of As Received Archive Material in Base Metal near Inner Surface Weld Clad...................................................................................................

7. Microstructure of As Received Archive Material near Center of Plate..............14

8. Microstructure of Archive Material after Isothermal IIcat Treatment at 800'C............................................................................................................................................15

9. Microstructure of Archive Material after Isothermal !! cat Treatment at900'C.......................................................................................................................................15

10. Microstructure of Archive Material after isothermal Heat Treatment at100TC..................................................................................................................................16

11. Microstructure of Archive Material after Isothermal Heat Treatment t

l l

at1107C....................................................................................................................................16 l

12. Microstructure of Archive Material after Transient llent Treatment at 800^C and Cooling Rate of l'C / min................................................................................. 18

13. Microstructure of Archive Material after Transient Heat Treatment at 900'C and Cooling Rat e of 1 *C/ min................................................................................. 18

14. Microstructure of Archive Material after Transient lleat Treatment at 1000 C and Cooling Ra t e o f l 'C / min............................................................................. 19

15. Microstructure of Archive Material after Transient Heat Treatment at 1 100 C and Coolin g Ra t e o f 1 *C / min.............................................................................. 19

16. Microstructure of Archive Material after Transient Heat Treatment at 800 C a nd C ooling Ra t e of 10'C / m1n............................................................................. 2 0 v1

17. Microstructure of Archive Material after Transient Heat Treatment at 9 00'C and Cooling Ra te of 10'C / min............................................................................. 2 0

18. Microsti ucture of Archive Material after Transient Heat Treatment at 1000'C and Cooling Ra t e of 10'C / min........................................................................... 21

19. Microstructure of Archive Material after Transient licat Treatment at 1 100* C and Cooling Ra te o f 10'C / min........................................................................... 21

20. Microstructure of Archive Material after Transient licat Treatment at 800'C and Coolin g Ra t e of 100'C / min........................................................................... 2 3

21. Microstructure of Archive Material after Transient IIcat Treatment at 9 00'C a nd Cooling Ra t e of 100*C/ min........................................................................... 2 3

22. Microstructure of Archive Material after Transient IIcat Treatment at 1000*C and Cooling Rate of 100 C/m1n......................................................................... 2 4

23. Microstructure of Archive Material after Transient lleat Treatment at 1 100 C and Cooling Rate of 100 C/ min......................................................................... 2 4

24. Microstructure of Archive Material after Transient Ilent Treatment at 1000*C with licating Rate of-100'C/s and Cooling Rate of 10'C/ min............. 25

25. Microstructure of Archive Material after Transient licat Treatment at 1100*C with licating Rate of =100*C/s and Cooling Rate of 10*C/ min............. 25

26. Design of Flat Tensile and Stress Rupture Test Specimen for Round-Robin Testing Program on Arthive Material................................................................ 2 8

27. Design of Round Tensile and Stress Rupture Test Specimen for Round-Robin Testing Program on Archive Matcrial................................................................ 2 8

28. Diagram Showing Locations from Which Tensile Specimens Were Taken Relative to th e Cen ter of Archive Pla t e 4 2.................................................................. 31

29. Comparison of Stress-Rupture Data Obtained by.'

nd the OECD Partner laboratories for A533 Grade B Steel Arct, Material with Extrapolated NRIM Data for Average Behavior of Five Ileats of Materta'.......... 40

30. Map of the TMI 2 Lower llead. Showing Extent of Damage Observed.

Positions ofInstrumentation Nozzles, and Locations of Samples Rem ove d for Exa m i n a t i o n.................................................................................................. 4 1

31. Configuration and Approximate Dimensions of a Typical Sample Removed from the 'IMI 2 lower ilead........................................................................... 4 2 l

l ykk w

g

32. Top View of TMI 2 Sample E 6 As Received. Showing Crack in Clad t

I ns id e Su rfa c e o f Lowe r H e a d.......................................................................................... 4 4

33. End View of TM1-2 Sample E-6. Showing Metal Disintegration-Machined Face fn Which Cracking Had Been Tentatively identified.'................ 45

34. Locations of Two Metallographic Sections Taken through the Thickness of Sample E 6 Pam11cl to the MDM Face....................................................................... 4 6

35. Low Magnificatior View of First Metallographic Section through SampleE-6................................................................................................................................47

36. Low Magnillcation View of Unetched Surface from Second Me tallographic Section through Sample E 6............................................................. 4 8

/

37. Higher Magnification View of Portion of Cladding in Fig. 36 After Etching Showing Interdendriuc Nature of Secondary Cracking......................... 49

38. Diagram Sh'owing Mac?ahardness Readings Obtained from TM12 Lower Head Sample E 6 and Sectioning Diagram of the Sample for Metallographic Examination, Elemental Analysis, and Heat Treatment.......... 50

39. Microstructure of As Received E-6 Base Metal at a Depth of ~50 mm below the inside Surface of the Lower H ea d...............................................................'. 51

40. Highur Magnification View of Microstructure of Fig. 39........................................,51

41. Microstructure of As Received E-6 Base Metal at a Depth of =35 mm below the Inside S urface of the Lower H cad............................................................... 53

42. Microstructure of As Received E 6 Base Metal at a Depth of =20 mm

-below the inside So rface e f die Lower Head............................................................... 53

43. Variation of Microhardness Readings in Banded Microstructure of Base Metal from TMI-2 lower Head Sample E 6................................................................. 54

44. Microstructure of Sample E 6 Base Metal after Heat Treatment i

AccorCing to Original Fabrication Procedure.............................................................. 5 5

45. Microstructure of As Received Sample K 13 Base Metal....................................... 55 4

46'.. Optical Micrograph of Oxidized Region in the A533 Grade B Ferritic f

Steel Base Metal of the TMI-2 Lower Head at the Bottom of th e C1a d d in g C ra ek......................................................................................................... 5 8

47. Higher-Magnillcation View of a Portion of the Lighter Gray Oxide Phase of Fig. 46. Showing In-Sn Pnase Filling Some of the Cracks in the Oxide......,59 vill

1

48. SEM Back-Scattered Electron Image of the Region near the Top of the Oxide Wedge Seen in Fig, 46.............................................................................................. 6 0

49. Low Magnificati a SEM Image of the Multilayered Material Deposited on the Exposed Crack Faces of the Cladd1ng...,.................................................................. 60

50. Higher Magnification View of the Region in Fig. 49 Indicated bytheAttow..............................................................................................................................61

51. Tensile Curves at Room Temperature for Material Cut from TMI 2 Lower Head Sample E-6 and As Received Archive Material.................................. 63

52. Ten % ",urve at 600'C for Material Cut from TMI 2 Lower Head Sample E-6 a n d As Receivcd Archive M aterial........................................................................... 6 3 4

Y n

i Tables

1. Experimental Heat Treatment Matrix for A533 Grade B Steel Amhive Materia 1....................................................................................................................... 1 0

2. Observed Macrohardness Values for A533 Grade B " tel Archive Material after the Experimental Heat Treatments Describ' Table 1.......................... 13

3. Tcst Specimen Configurations Utilized for Round Robin Testing of Amhive Ma terial...................................................................................................................... 2 9

4. Summary of Room-Temperature Tensile Data Obtained by ANL for A533 Grade B Steel Archite Material from Test Specimens Oriented Parallel.

4 5'. and 9 0' to the Rollin g Direc t1o n............................................................................. 3 0

5. Summary of 600'O Tensile Data Obtained by ANL for A533 Grade B Steel Archive Material from Test Specimens Oriented Parallel, 45'. and 90' to t h e Roll i n g Dire e t10 n........................................................................................................... 3 3

6. Summary of Tensile Data Obtained by the Studiecentrum voor Kernenergle/ Centre d' Etude de l'Energic Nucleaire of Belgium for A533 Grade B Steel Archive Material. Flat Specimens........................................................ 34

7. Summary of Tensile Data Obtained by the Centre d' Etudes Nucleaires de Saclay of France for A533 Grade B Steel Archive Material. Flat Specimens. 34

8. Summary of Tensile Data Obtained by the Staatliche Materialprofungs-anstalt of the FRG for A533 Grade B Steel Archive Material.

Fla t Specimens........................................................................................................................ 3 5

9. Summary of Tensile Data Obtained by the Staatliche Materialprofungs-anstalt of the FRG for A533 Grade B Steel Archive Material.

- Round Specimens..............

................................................................................................35

10. Summary of Tensile D.

otained by the Comitato Nazionale per la 4

ricerca e per lo sviluppo dell'Energia Nucleare e delle Energie Alternative ofItaly for A533 Grade B Steel Archive Material. Flat Specimens...................... 36 11; Summary of Tensile Data Obtained by Equipos Nucleares, S.A. and Centro de Inspecci6n y Asistencis Tecnica. S.A. of Spain for A533 Grade B Steel Archive Ma terial.....................................................-................................................................ 3 6

12. Comparison of Average Values of Room-Temperature Tensile Properties Obtained by Laboratories Participating in the Round Robin Tests on A533 G ra d e B S t e el Archive M a t erial........................................................................................... 3 7 x

l

13. Comparison of Average Values of 600*C Tensile Properties Obtained by Laboratories Participating in the Round Robin Tests on A533 Grade B Steel Arthive Material............................................................................................................. 3 8

14. Summary of Stress Rupture Data Obtained by OECD Laboratories on A533 G rad e B St eel Archive Ma t ertal........................................................................................... 3 9

15. Tentative Classification of TMI 2 Lower Head Samples in Terms of Expected Damage from Core Mcltdown Accident...................................................... 4 3

16. Elemental Composition of Material from TMI 2 Lower Head Sample E 6 Compared with Vendor's TMI 2 Heat Analysis and Analyses o f Ar c h i v e M a t e rt al.................................................................................................................. 5 7

17. Comparison of Tensile Properties of As Received TMI 2 E 6 Material with Average Values for Archive Material at Room Temperature and 600 C.............. 62 XI

d Executive Summary his report summarizes the work performed by Argonne National Laboratory (ANL);on the Three Mile Island Unit 2 (TMI 2) Vessel InvestigMton Project (VIP)

Metallurgical Program during the nine months from October 1989 through June 1990. This program is a part of the international TMI-2 Vessel Investigation Project being conducted jointly by the U.S. Nuclear Regulatory Commission and the Organisation for Economic Co-operation and Development (OECD). We overall project consists of three phases, namely, (1) recovery of material samples from the lower head of the nil 2 reactor: (2) examination and analysis of the lower head samples and preparation and testing of archive material subjected to a similar thermal history; and (3) procurement, examination, and analysis of companion cor'e material located adjacent to or near the lower head material.

The specific objectives of the ANL Metallurgical Program, which makes up a major portion of Phase 2, are to prepare metallographic and mechanical-test specimen blanks from the.TMI-2 lower head material, prepare similar test specimen blanks from suitable chive material subjected to the appropriate thermal processing, determine me mechanical properties. of the lower vessel head and archive materials under the core-meltdown accident conditions, and assess lower head integrity and margin to-failure during the accident. The ANL work consists of three tasks: (1) fabrication of metallurgical and mechanical test specimens from TMI-2 pressure vessel samples, (2) archive materials program, and (3) mechanical property characterization of TMI-2 lower pressure vessel head and archive material.

- A series of experimental heat treatments on archive material from the Midland

)

nuclear reactor lower head has oeen completed, and the resulting samples have been examined metallographically. One group of specimens was subjected to a series ofisothermal heat treatments in which they were heated to a prescribed maximum temperature at a rate of 40 C/ min, held at that temperature for 2 h, and cooled at at rate of 1*C/ min. These specimens were generally found to have microstructures and hardnesses quite similar to specimens given a similar l

transient heat treatment in which the 2-h hold time at maximum temperature was eliminated. For a given maximum temperature, the microstructure became progressively finer and the hardness greater as the cooling rate was increased in the transient heat treatment experiments. Increasing the heating rate to

-100'C/sec for a few selected heat treatments appeared to have no effect on microstructure or hardness. Response to the heat treatment was, in general, consistent with pt6hhed continuous cooling transformation diagrams for A533

- Grade B ead similar steels.

+

w w

e r

2 A series of round robin mechanical tests and microstructural studies on Ole as-received archive material was carried out by the participating partner laboratories.

The laboratories all reported a quenched and tempered microstructure in the base metal typical of that expected for an A533 Grade B ferritic steel pressure vessel material. Banding was observed along the rolling direction in the steel, along with associated variations in microhardness. The bands were found to consist of alternating regions of upper and lower bainite, and a variation in the relative abundances of the two phases through the thickness of the plate was observed.

Analysis of the cr.rbides present with carbon extraction replicas and electron difTraction unslynis revealed only cementite (Fe3C).

Round robin tensile tests were conducted on the archive material at room temperature and 600*C, and stress rupture tests were conducted at 600'C, Both circular-and rectangular-cross-section test specimens (hereafter identified as round and flat specimens, respectively) were used in these tests. In general, good agreement was obtained in the tensile tests, although the Staatliche Material-prOfungsanstalt (MPA) in the Federal Republic of Germany found the yield strengths of the flat specimens to be significantly lower than those of the round specimens at room temperature. No r"nh variations were seen at 600'C. In the stress rupture tests, the MPA data inJeated that the flat specimens had somewhat shorter lives than the round specimens, an effect apparently associated with a tendency of the flat specimens to fail at the extensometer notches. Otherwise, the round robin stress-rupture data were in good agreement.

Fifteen samples have been recovered from the TMI 2 lower head, including 11 samples without and 4 with nozzles. These samples have been received at ANL and are being photographed and decontaminated. In the decontamination process, the specimens are first immersed in a dilute hcl solution to dissolve a thin surface layer containing most of the radioactive contamination. A mechanical milling process is then used to remove an additional thin surface layer, After this two-step treatment, the level of radioactMty on the surface of the specimens is typically on the order of 10 mR/h or less, with no loose actMty.

A detailed examination of Sample E-6, which was the first sample removed from the TM12 lower head, has been carried out at ANL. This sample was of particular interest, because it contained a portion of a surface cladding crack that largely encircled an adjacent instrument penetration nozzle. Metallographic sections through the sample revealed that the cracks extended completely through the Type 308L stainless steel cladding but for a distance of only -3 mm into the underlying A533 Grade B ferritic steel base metal. Considerable secondary cracking of the cladding. not connected to the cladding surface, was seen in one of the metallographic sections.

3 An initial elemental analysis of base metal material from Sample E-6 revealed its carbon content to be 0.28 wt.%, which is above the maximum specifled level of 0.25 wt % for this alley, Subsequent analyses of three additional speck ens indicated carbon levels of 0.25 wt.%.

Scanning electron microscope / energy dispersive X ray analyses of the principal phases in the region of the cracks showed that a thick layer of predominantly Fe oxides was present on the exposed surface of the ferritic steel base metal, and a much thinner layer of Fe-Cr-Ni oxide was present on the stainless steel cladding. The base metal oxide was extensively cracked, and an in Sn phase filled some of these cracks. This In Sn phase was apparently control assembly material from the reactor core that flowed into the cracks in the molten state and resolidilled in place. Similar control assembly material apparently flowed onto and interacted with the stainless steel crack surfaces to produce a variety of phanes containing Fe, Cr, Ni, In, Sn, Ag, and Cd. Spherical beadc of Ag-Cd control-assembly material that were deposited as a liquid and solidified in place were also observed on the crack surfaces. Only minute quantitles of Um Zr O fuel, in the n

form of angular porous shards, were seen in the vicinity of the cracks. These particles do not appear to have melted, but, instead, probably fell onto this surface after cladding cracking and oxidation of the underlying base metal had occurred.

Microstructural examinations of the A533 Grade B base metalin the vicinity of the surface cracks in Sample E-6 indicated that the base metal reached a maximum temperature on the order of 1000 to 1100 C during the accident. This conclusion was supported by hardness measurements and by tensile test results at room temperature and 600 C, both of which showed that the base metal had been hardened compared to the as fabricated condition, in contrast, microstructural and hardness results from the base metal of TMI-2 Sample K-13 indicated that this material did not exceed the critical ferrite to austentte transformation temperature of 727'O during the accident. This latter sample was taken from a region away from the core relocation site.

It was tentatively concluded that cladding cracking in Sample E-6 probably occurred during the early stages of cool down after the accident, when the still-hot cladding layer was placed into tension because of thermal contraction. This latter process is analogous to hot tearing during welding.

1 I

5 1

Introduction Extensive studies have been conducted during the past five years to determine the extent of damage suffered by the Three Mile Island Unit 2 (TMI-2) nuclear reactor as a result of the loss of coolant accident in March 1979. These studies have focused primarily on the end-state core configuration, and they have confirmed the occurrence of significant melting and relocation of core material to the lower plenum region of the reactor vessel. It is estimated that approximately 20 metric tons of molten core debris fell onto the bottom head of the reactor.1 and evidence of thermal damage to instrument structures in the lower plenum and around flow holes in the flow distributor has been observed. The objectives of the present program are to (1) determine a scenario for, and deduce the temperatures of, the steel in the lower vessel head during the accident: (2) determine the mechanical properties of steel from the lower head under the core meltdown accident conditions: and (3) assess the integrity of the TMI-2 vessel.

2 Background

This program is a part of an international TMI-2 Vessel Investigation Project (VIP) being conducted jointly by the U.S. Nuclear Regulatory Commission (NRC) and the Organisation for Economic Co-operation and Development (OECD).

Participants in the international project include the U.S.,' Japan, the Federal Republic of Germany (FRG), Finland, France, Italy, Spain, Sweden, Switzerland, and the United Kingdom (U.K.). Government agencies from these countries are contributing to the cost of the project; in return, they will share the results obtained. Several of these countries are actively participating in the research, and this participation constitutes a part of their financial contribution.

The first phase of the TMI-2 Vessel Investigation Project (VIP) is the recovery of material samples from the lower head. A contract for this work has been placed with MPR Associates, Inc., and extraction tools and procedures have been developed. The first samples from the TMI-2 lower head were scheduled to be shipped to Argonne National Laboratory (ANL) in about January 1990. Eight to 20 prism shaped samples, each approximately 152 to 178 mm (6 to 7 in.) long. 64 to 89 mm (2.5 to 3.5 in.) wide, and 64 to 76 mm (2-1/2 to 3 in.) deep, have been cut from the inner surface of the lower head. The locations from which the specimens were taken were (1) near the area of impact by the primary jet of molten material on the lower head; (2) toward the radial center of the lower head underneath the maximum thickness of debris: (3) in the quadrant of the lower head where a " wall" of consolidated 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

f 6

(5) locations with one or more instrument penetrations, particularly where surface cracks had been observed visually.

The second phase of the TMI-2 VIP is examination and analysis of the lower head samples and preparation and testing of archive material that had been subjected to a similar thermal history. This second phase is being carried out by ANL (in the present program) and by the Idaho National Engineering Laboratory (INEL). The lower head samples recovered from TMI-2 will be documented, examined by opbcal metallography and other techniques to determine the maximum temperhture reached, decontaminated, machined into mechanical test specimens, and tested. Because the supply of lower head material is limited, further mechanical tests will be conducted on archive material of similar chemistry and fabrication history that has been subjected to heat treatments simulating those to which various regions of the lower head were subjected during the accident.

Some of these testing and evaluation activities will be carried out by other members of the NRC/OECD, joint program, and all of the results obtained by the various participating laboratories will be integrated into a comprehensive final report.

The third phase of the TMI-2 VIP, to be carried out by INEL, is the procurement, examination, and analysis of companion core material located adjacent to or near the lower head material that is to be studied in Phase 2. This core material will be analyzed to determine, in conjunction with the information obtained in Phase 2, the maximum temperature attained and the nature of the attack that occurred at the lower head inner surface during the accident. The temperature distribution obtained will be used to calculate a stress distribution in the lower head. Based on this information and the mechanical properties of the lower head and archive material determined in Phase 2, the margin-to-failure of the. lower head will be assessed.

The spec 1De objectives of the ANL portion of the project are to prepare metallographic and mechanical-test specimen blanks from the TMI-2 lower head material, prepara similar test specimen blanks from suitable archive material subjected to.1ppropriate thermal processing, determine the mechanical properties of the lower vessel head and archive materials under core-meltdown accident conditions, and assess the lower head integrity and margin to failure during the accident. INEL is responsible for the microstructural characterization of the TMI 2 lower head material, the integration of all experimental and analytical data and studies performed on the TMI-2 lower head material, and the issuance of a nnal report on the work conducted in Phase 2.

The ANL program consists of three tasks: (1) fabrication of metallurgical and mechanical test specimens from the TMI-2 pressure vessel samples, (2) archive materials program, and (3) mechanical property characterizat:on of TMI-2 lower

l 7

pressure vessel head and archive material. The progress to date on these tasks 2 s summarized here.

since the previous progress report i

3 Heat Treatment Experiments on Archive Materials 3.1 Summary of Heat Treatment Experiments One purpose of the archive materials program is to produce a set of standard microstructures under controlled heat treatment conditions. Comparison of these standard specimens with the microstructures observed in the material from TMI 2 will permit more accurate determination of the thermal history of the TMI-2 lower head dtuing the accident.

Archive specimens of A533 Grade B steel have been subjected to three general types of heat treatment. The first, and simplest, is the isothermal treatment shown schematically in Fig.1. The sample is brought up to a specified maximum tensperature at a heating rate of 40'C/ min, held at the maximum temperature for 2 h, and then slow cooled at l'C/ min to below 300*C before being air cooled to room temperature. This heat treatment simulates the thermal cycle at the mid-and outer-wall locations in the lower head with no quench. In this scenario, the massive vessel and its contents slowly cool to ambient temperature after the accident, and ess.entially equilibrium microstructures would be expected.

Maximum temperatures for these isothermal heat treatments range from 500 to 1300'C in 50'C increments or, above 900'C,100 C increments, for a total of 13 heat treatments.

The second type of heat treatment used is the transient heat treatment shown in Fig. 2. The heating rate to maximum temperature is 40*C/ min, but specimen cooling begins immediately upon reaching the maximum temperature, with no hold time. Three cooling rates,1,10, and 100 C/ min, are used down to 300 C, below which the specimens are air cooled to room temperature. Again, this heat treatment again simulates the thermal cycle at the mid-wall and outer-wall locations, but now with the added effects of a quench. Maximum temperatures range from 750 to 900 C in 50 C increments and from 1000 to 1300'C in 100'C increments, for a total of eight temperatures. Because three different cooling rates are employed for each of these eight maximum temperatures, a total of 24 temperature-cooling rate combinations are being studied. Maximum temperatures below 750'C are not being investigated in these heat treatments because they are below the eutectoid transformation temperature, and no microstructural effect of the varied cooling rates would be expected. (Lower temperatures are included in j

8

- - - T max 2

I l

a i

i E

l 1

l

=40 C/ min l

E i

i Cooling rate =

l l

1 C/ min

'e 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> +'

A l

l Time Fig.1. Generalized Representation ofisothermal Heat Treatmentfor Archive Material

- - - -. T max -

_ 2-3-

Three cooling 8

- rates

b b

. e E'

r-0-

-40 C/ min Time Ag. 2.

Generalized Representation of Transient Heat Treatment with Normal Heating Rate of =40*C/ min and Three Cooling Rates (1,10, and 100*C/ min)for Archive Material i

I l

9

- - - - T max g

a Cooling rate =

2 10 C/ min 8.

E f

i

% =100 C/sec Time Fig. 3.

Generalized Representation of hansient Heat heatment with Rapid Heating Ratefor Archtuc Material the 1 cthermal heat treatments, because the relatively long time at-temperature might produce some tempering of the bainite phase,)

The final type of heat treatment that was investigated, the transient heat treatment with rapid heating, is represented in Fig. 3. 'Ihis treatment is similar to those described above except that a very rapid heating rate, on the order of 100*C/sec, to the maximum temperature was employed. This heat treatment simulates the rapid thermal ramp that probably occurred at the lower head inner wall when the molten core material was deposited on it. A single cooling rate of

-10*C/ min was used. Four maximum temperatures, namely, 1000, 1100, 1200, and 1300 C, were studied.

A test matrix outlining the series of heat treatment experiments described above is shown in Table 1. The 1200 and 1300*C heat treatments indicated in this table have not yet been carried out, because they are to be performed only if microstructural analyses of the TMI 2 lower head material indicated that it actually reached temperatures this high. Thirty-one specimens, excluding the 10 for the 1200 and 1300 C exposures, were required to carry out the thermal treatments shown in this table for each location through the thickness of the plate. Specimens for heat treatment were taken from two locations in the archive plate, namely, near the inner surface weld clad and near the center of the plate. This means that the

l 10.

Th ble 1.

Dcperimental Heat 7Yeatment Matrixfor AS33 Grade B Steel Archive Material.

Maximum Isothermal Transient Temperature Heating rate ('C/ min):

40 40 40 40 6000a

('C)

Cooling rate ('C/ min):

1 1

10 100 10 500 X

550 X

600 X

650 X

700 X

750 X

X X

X 800 X

X X

X 850 X

X X

X 900 X

X X

X 1000 X

X X

1100 X

X X

1200b X

X X

1300b X

X X

a6000*C/ min = 100*C/sec, bThe 1200 and 1300 C heat treatments will be carried out only if the metallographic studies indicate that these temperatures were attained in the TMI-2 lower head material during the accident.

total number of archive specimens that were heat treated and examined is 62 (or 82 If the 1200 and 1300'C treatments are subsequently carried out).

3.2 Results of Microstructural Examinations and Hardness Measurements The range of microstructures resulting from the heat treatment experiments described above will be discussed briefly with photomicrographs of representative specimens and the appropnate continuous cooling transformation (CCT) diagrams.

One such diagram for A533 Grade B steel, adapted from Ref. 3, is shown in Fig. 4.

Superimposed upon this diagram are cooling curves from a 900'C austenitizing temperature corresponding to cooling rates of 1,10, and 100 C/ min. Some variability in the position of the curves, particularly for the onset of ferrite formation, is apparent in this diagram, depending upon the data source anc the austenitizing conditions. 'lhus, the predictions obtained from this diagram should be considered to be only approximate.

11 Time (h) 10 2 10'1 10 10' 0

i ii i

ii i.ui i

i iiiioi Austenitized 950 C,10 min., Amano et al.

Austenitized 900 C,120 min., Pelli et al.

1000 Cooling rate from 900 :

l _

10d / min

~10 / min 1/ min

-o 800

~Austenite t

i I

forfifi"

]

600 "Qc- =

-l h

-[g,

- Balnite -

400

_g w

............7-200 0

1 2

3 4

6 10 10 10 10 10

-Time (s)

Fig. 4.

Continuous Cooling TransformaUon (CCT) Diagramf~~ A533 Grade B Steel, Based On Data qf Amano et al. (Ref 4) and Pelli at al. (Ref. 3). Cooling curves corresponding to cooling rates of.',10, anl 100*C/minfrom an austenttizing temperature are superimposed nrt d1e diagram. (Adaptedfrom Ref. 3).

A second, more detailed, CCT diagram is shovm in Fig. 5 in this cace, for the German steel Saarstahl 20MnMoN155. This steel is almost identical in composition to A533 Grade B, and the general features of the this diagram agree reasonaoly well with those of Fig. 4. The CCT diagram of Fig. 5 incluces information on the percentages of the various phases expected to be formed from the austenite for selected cooling rates. Cooling curves for rates greater than 20'C/ min are labelled using cooling parameters, which are the times in minutes required to cool between 800 and 500'C.

Hardness measurements were also made on the heat-treated archive specimens described above, These macrohardness measurements were obtained as

12 Time (h) 10'3 10 2 10'I 10 10 0

I 1200 i

ini i ji i i i n ii i ji i i i n ii i

i i i i nii

.... i o n ii i

- Austenttizing temperature = 875*C --

-Ac 3 = 850' -

I Cooling rate =

Ac 3 =6/ min) ~"-'

715

-1000 Cooling parameter 20 / min (0.4

/

410*/ min :

ek 6.9 -

E-0 f

800 u

p

\\.)

0.40

._... I -

10 kSM8k. Ferrite + Pearlite...

s a

600

--- kOS

\\

5 4

5 L % transformed N--w x

g g

+

k g\\..

h.. 4....

\\

4

.-. {..

6 _ \\ 89 2.5*/ min 400

_. Dalnqe_.

s 97 88 4 61

'4 m&.transk Martensite 200

\\......

0 0

1 2

3 4

5 10 10 10 10 10 10 Time (s)

Fig. 5.

CCT Diagram for Saarstahl-20MnMONt55 (0.17 0.23 C, 0.15 0.35 St,1.20-1.50 Mn, 0.45 0.60 Mo, 0.50-0.80 NO Austenttized at 875*C, The cooling parameter is the cooling time in minutes between 800 and 500*C.

-i Vickers hardness numbers (VHN) with a diamond pyramid indenter and a 30 kg load, The resulting indentations were large relative to the fine banding present in many of the microstructures, and average macrohardness (rather than local microhardness values) were thus obtained. The results of the hardness determinations are summarized in Table 2 for the archive base metal specimens, taken from near the clad surface.

Figure 6 shows the as-received microstructure of the archive material from near the weld clad at the inside surface. The microstructure consists primarily of fine tempered bainite, along with possible regions of tempered martensite. Near the center of the archive plate, where cooling from the initial austenitizing treatment was somewhat slower, the microstructure consists almost entirely of tempered bainite (Fig. 7), and the prior austenite grain boundaries are visible.

Specimens with these two as-received microstructures were used in the heat treatment experiments.

13 Table 2.

Observed Macrohardness Values (VHN)for AS33 Grade B Steel Archive Mater (aga q/ter the Dcpertmental Heat Treatments Descr(bed in Table j,b Maximum Isothermal Transient Temperature Heating rate ('C/ min):

40 40 40 40 6000C

('C)

Cooling rate ('C/ min):

1 1

10 100 10 500 195 550 192 600 189 650 191 187 700 180 750 194 180 222 211 800 188 189 219 241 850 185 188 221 270 900 188 176 215 254 1000 154 154 231 265 231 1100 130 174 233 262 234 aHardness of as received archive material = 188.

bThe data represent the average of four hardness readings taken on each specimen.

ceooo*C/ min = 100'C/sec.

For specimens subjected to maximum temperatures of 700'C or less in the isothermal heat treatments, essentially no change in microstructure occurred, as was discussed in a previous progress report.2 In conformance with this observation, the hardness of these specimens (for specimens taken from near the inside surface, Tatle 2) was also virtually unchanged from the as received condition These results reflect the fact that no transformation of the as received microstructure to austenite occurred at these temperatures, and the times at 500 to 700'C were too short to produce any significant tempering effects.

For isothermal heat treatments at temperatures above 700'C, partial or complete transformation of the as received microstructure did occur, and this austentte phase retransformed upon cooling. The CCT diagrams shown in Figs. 4 and 5 indicate that the austenite would transform largely to the relatively soft ferrite and pearlite phases for the slow isothermal cooling rate of 1*C/ min. This N confirmed by Figs,811, which show the microstructures of the specimen of Fig. 6 after isothermal heat treatments at 800, 900,1000, and 1100'C, respectively.

With increasing maximum temperature, the microstructure coarsens, and the primary microconstituents present are ferrite and pearlite, along with some moderately fine bainite. The hardness data in Table 2 indicate no hardening over

14, I

.~

t

'j

....s

..;:l i. l_ f; s.

1 l'

~

?,. ;

~,z, ;..

(,

'. /...

[

' [ g%

' I.; ~ '.,,. ' '

w

_. ;,, lu

.. ;. :* 'T <4_ 6.;

,,,v,...

. : n l.

Fig. 6.

Micmstructure of As-Received Archive Material in Base Metal near inner-Surface Weld Clad.

i I

', 11 m

L i

l 7, 6125 m ' i, kws.Swa *,rgeretrd Fig. 7.

Microstructure ofSVR'eceived Archive Material near Center of Plate.

l 15

i..

.i o

?shh f125 pmW,j j'" -3.. L -)

R g. 8.

Microstructure of Arthtve Matettal qRer Isothermal Heat heatment at 800*C.

.$7 ig,.

il* ;

s

;\\,l ',,.,.--

! F 9

fby

-[

p M4,,

tr pN j

y gt:125 lim l c di a

.s

'l

'.'37.k

'i-U.,u, A g. 9.

Microstructure of Archive Material qfter Isothermal Heat heatment at 900*C.

16 y - n y [e q ',

m y v:

e-3,]'..

gy;gp -

3 ;

y

\\

3 n

c x.-

f y Nj;'

f g hi l

h l

r h

.s 3r 4;

s 6

o4

,s g..

-- Q, r &g jy

,). 3 v.-.

e

[

~3..,9 fj _

i [(;.,l Wf;zy,,hg[Q: :ay;f

/?

c

./

y' gy. -

Q w:u l

{g..:.cfs y. 4'y &

~-

+

V

<,{., J.Hg

'1

+ '

, +. c.

o s -

3

.m 3w

.g

\\

i

/

'e

&c.;. n 9l,l;.d j-1:

k\\

j'i _

~q '..' '

/

l. c.

n( - y i;(

.r.%

M,:

x.

QWj b,%n.w?

- A sk. a

.m a l-M'

j, 125 pm,j Q.

l :: Q.

O. a...

i i ;

. & w. +aw.

u

+...

Rg. 10. Microstructure of Archive Material qfter Isothermal Heat 7Yeatment at 1000*C.

w mm mencm:: ga y.w~,~;n'm ry w&p,n%&ml9ll, yll$pyfyx;h;y,[f, pl!f ' (s.\\ W f

N ll Lj p

3 y g,

99.

'My

,ili-: %;; p;+ ~ ~w x~q du b?Q Q;:+An L.

4,

' C'

.' 5+{ b,NO S,c dS 6;Q?,-O La Wm q FNM Hc 9n( :.

q : m,w

.v

+. 3].

e w: m,. ; ;;;m( m g,, % fTt% g/ in g g% j,c, F -R J i

.7y

. p;, e* r

< ww a c+d 1 a

=

yu.

m.

o yz n

,f, a

it;E yr '

as - - T r ;.- ef 9.,

9 tg. + -

i m.t -

i I

q l

.g

. D My y;w m

~

= y? y;hrj W y l

s MlMf 3j.

tRy W N:mv*~, '-

.J:f d

W y

wp:g ;@j. m jm%

3;1 h m.

eq y i

c n;,.. : ' ' ( i

'*, ), R (: 3 1

p m

f

). _

r u -.

L y *.

J g;p h,g,... M'N

%,, W.dl dW.;s mg /W. R,,

a s m:

Agg m,

.8 &Q';M, 4. g/+/.,w%jpW: q, g: y m;p g) 3 ;f

{ igg

f.

L Anm< py G

-n, 6 '> : ' s n

b-s,..

y ; p3 w

v m~ w;.

. Wik q&

q 30 t@w;qq[;2;;

, 3pgy l 9 V.

,y i lx w < ? #[ U $, f

,-<125 ma,].

J$ '; M:[.

.d WA !iWI e

u a \\;x

. yk l ;

l,> S ' &.lf s.

-. \\.x. ;,, [, ' ~:'; {r 4,. h,

?'

j

.~,n aw.~

<a Rg. 11. Microstructure of Archive Material qfter Isothermal Heat Treatment at 1100*C.

17 that of the as received condition for temperatures up to 900 C, and significant softening at 1000 and 1100'C. This softening effect occurs because the as, fabricated archive material was austenttized at 871899'C Heat treating this material to higher temperatures during the isothermal heat treatments produces an increase in the austenite grain size and a general coarsening of the transformed microstructure, resulting in decreased hardness.

The archive material specimens subjected to the transient heat treatments with a cooling rate of l'C/ min exhibit microstructures very similar to the corresponding specimens from the isothermal heat treatments. This is illustrated by the microstructures of the specimens heated to maximum temperatures of 800, 900,1000 and 1100 C, shown in Figs 12-15, which may be compared with the corresponding isothermally treated specimens shown in Figs. 8-11. The hardnesses of these specimens are also essentially the same, at least up to 1000'C, as seen in Table 2. These similarities are to be expected, because the two heat treatments differ only in the absence of a two hour hold time at the peak temperature in the transient heat treatment.

The specimen subjected to the above transient heat treatment at 1100'C is an exception, in that its hardness is signif!cantly higher than that of the specimen isothermally heat treated at the same maximum temperature. This specimen represents a repeat of this heat treatment, and it was prepared after the first specimen heat treated in this manner was found to have an average hardness of 190. No explanation can be ofTered for the observed difference in hardnesses between the isothermally and transient-heat-treated specimens at 1100 C.

The specimens subjected to the transient heat treatment and cooled at a rate of 10'C/ min are shown in Figs 16-19 for maximum temperatures of 800,900, 1000, and 1100'C, respectively, These specimens display somewhat finer microstructures with less ferrite and pearlite and more bainite than the corresponding specimens cooled at 1 C/ min. The hardnesses (Table 2) are likewise somewhat greater, These observations are again consistent with the CCT diagrams of Figs. 4 and 5, which indicate that the 10 C/ min cooling curve intersects only a portion of the ferrite nose, with the majority of the austenite transforming to bainite.

An isothermal specimen with this cooling rate but heated to a maximum temperature of only 700 C was also inadvertently prepared, even though it is not in the experimental matrtx. The hardness of this specimen was found to be no greater than that of the isothermal specimens heated to maximum temperatures of 700 C or less. This observation again reflects the fact that no transformation of the as-received microstructure to austenite occurred at 700 C, and so there was no opportunity for hardening upon cooling.

18

6 $ 7*4 @@,%g d.

' 'DN 576

>< t 1

igg,g j

f

[kf' p

go,

n. ~ 8,x a,1g ze. cl;,

t.

$k 4

%.gt:;,/*$$ffi?G,f,YM)k;((;h.ch, sgy} q*,.;

gEq$

Qf$.gdi 3,14'M 3y;L f

.93

. : :l,y c,y:

1 ry

(,,

IfffQN1f T..*.(#f 4",

k.23g' 5!;.

rgf3 (q.pg, 125 pm i

p V

tYadN*$4'?...n, d3V.1G':

e Fig. 12. Microstructure of Archive Material qfter Transient Heat Treatment at 800*C and Cooling Rate of1*C/ min.

f(R W $ M Q F M:

k

(

hw#8ing) $" 44 1

b brie IN, N mv=E Fig. 13. Microstructure of Archive Material qfter Transient Heat TYeatment at 900*C and Cooling Rate of 1 *C/ min.

. -.-..-.=.- - - =.

Jl 19 l

Y l'

%7.'gf!iRQ ^

d

., n.: ' [%&b[, J '

a e c% ] h.;,.i $/ {A.' v _ _ '2

,-.125 m h'

Nh kh j

f y

W 4tw Rg.14.. Microstructure of Archive Material qRer hansient Heat Treatment at 1000*C and Cooling Rate of 1*C/ min.

Y%lh@f..:.

M

, ;i m

cy[

7y

[

A j.)

h Sh 4ffrys?

)

<y, M,.. -

m

.g f

e.

i v ' "" Aa m

.af

', s k

. Oki N l:, i;fjbj s

' d;Y

~_

,g

. '"3 p'

4pp.

9

, a

.:n 3;m t

.,n V\\

k i jf'.

p

.gf 3

l 1

A P

40 t'

ge

~

f f;b[k.1 7, 2125 pm:

1, :

,I'[

n,

. i.-

g, E

ac.wtr m rva -

- v Ag. ' 15. Microstructure of Archive Material qfter hansient Heat heatment at 1100*C and Cooling Rate of 1 *C/ min.

1 20 t

kaY. h&f

. d a:JI mr.r u 4 ro e.

Fig. 16. Microstructure of Archive Material qfler hansient Heat Treatment at 800*C and Cooling Rate of10*C/ min.

P a

Fig. 17. Microstnicture of Archive Material after Transient Heat Treatment at 900*C

\\

and Cooling Rate of10*C/ min.

21

$y fj j

<igang$}g.[gg*[y' h8

'[-r l

ppasgg%w?aphy$n'[j thesee ha; hggu heep((gg(j d4NNN[ghgjugj'j p*E'd 3 IQ$$y t,

e'

y r EWi@uand$naM ig@wne w

f irt

? b '.

'I; I :

a Ag. 18. Microstructure of Archive Material qfter Transient Heat Treatment at 1000*C and Cooling Rate of10*C/ min.

l

'?

s.~;- e.{

5; f

,$), qq:c Va'f^. ' gr. lf~ :^% -

k(

':{ ~

g

)$) jfd "A

'i ~

+

h1

Q:

.i i?

,2 ig(f4p,%lQ j (y':'g?4$p

.g Yd; Y

~ ;}

'y k1

l/

jy

...y n>

' %;.4#l' g.,, ?j,,

ygg '

}._

d5 N

Jf;

, - 125 pm p

.i i:

em iM.

'M.n:i3:..

u,..sa~n J

Ag. 19. Microstructure of Archive Material qfter Transient Heat Treatment at 1100*C and Cooling Rate of 10*C/ min.

2,2 The CCT diagrams (Figs. 4 and 5) predict that, for a cooling rate of 100*C/ min from the austenitizing temperature (l.c., a cooling parameter of 3 in Fig. 5),

transient heat-treated specimens would form no ferrite or pearlite The microstructure would consist almost entirely of bainite, with perhaps a small amount of martensite also present. This prediction is borne out by the microstructures seen in Figs. 20-23. The hardnesses (Table 2) are also greater, because (1) the all bainite microstructure is harder than a mixed bainite ferrite-pearlhe microstructure, and (2) the bainite formed at a cooling rate of 100'C/ min is finer and harder than that formed at cooling rate of 10*C/ min.

Finally, the specimens treated by very rapid heating (-100'C/s) to the peak temperature and cooling at 10*C/ min have the same microstructures (Figs. 24 and

25) and hardnesses (Table 2) as those heated at 40 C/ min to the same maximum temperature and cooled at the same rate. This behavior is also to be expected, since the material is completely austenttized in both heat treatments, independent of the heating rate, and the transfonnation products would be the same because the cooling rates a're identical.

4 Evaluation of Archive Material 4.1 Microstructural Characterization Round-robin mechanical tests and microstructural studies on the as received archive material were developed to better characterize this material and to determine the level of variability in mechanical-test data obtained by the participating laboratories. The OECD partner laboratories participating in this test i

program are in Belgium, France, the FRG, Italy, and Spain. Finland and the U.K.

participated in the metallographic studies, but did not conduct any mechanical

. tests.

The participating laboratories received two pieces of archive material for microstructural studies, each =25 mm (1 in.) on a side and =145 mm (5.7 in.) long.

4 for these studies. The length of the pieces corresponded to the full thickness of Archive Plate 4 2, from which the material was taken. Further details on specimen material location are given in the previous progress report.2 The participating laboratories all reported a quenched and tempered microstructure in the base metal typical of that e::pected for A533 Grade B ferritic nteel pressure vessel material. The microstructure showed considerable banding along the rolling direction, and a noticeable variation in features through the thickness of the plate was generally observed. A region of base metal =7-10 mm

23 C

W 9

F gr i.?

5.

if m W

. I>,)p? *. '

y;

.i? g

' %j. ;

K' f>-

n

'I '

. s ~ jf, '

V Y,

Rh;f&

~ 'w) r m-s I

^

M ds a r.. (-

py k

a i h.1

?

125 pm;-

.3

!gr i

e' andma:.e.:>. m.w n.'.

r, Fig 20. Microstructure of Archive Material qfter kansient Heat Ret tmer.c ~ ' 800*C and Cooling Rate of 100*C/ min.

f,3.?

I

h{

b d'

g.

m 7

.i

(

, g5 1

"ju i

ug

'wM' y.

L P

'x 1 ";

/A '* '

Mk'h,

,f p3{8'7 t#[

7,

{ Ehy,,

'u.

edt.dlebY,taark-h#

'Is FP# lie

" h,jk'i ikman-w.ml Fig. 21. Microstructure of Archive Material qller hansient Heat Treatment at 900*C and Cooling Rate of100*C/ min.

24 ha' '

( $

p, 4 27, a %' @u, hh,d$y t

mp94 f?kb?$hf kf t?w$$e$acpqh3a. bag,~

w c

o

$kh 5

f g p, tr(.E@%gy:,.,s?$;q$a b)

4" ;!h b n MA 3

t

,'*V$g gsr w

p p

3Q, s qpRpy,)yO lhM gggi4Q bin "i &y p[b,Q2h?%s,ydg%iQ).,g,4 9 hwyp:nf%a

s. g <, v.v

.s.v&

yy e 4

ht k%hAf Mh m4%W$t&A MMb%49 tim 5 46 MWf

[Ihd*hdkhkff "M$2dy~+W

$'"QMi*h;:y'f wp 4%

0 s hbNIhh h

j) 5

'f$f' 1

3 EAhMD4Mfkkk$k'ed'*

E Y $/% h@ M t6 fl,fhhh

{$3/4 ;s a M!a&p(dyjj${jsPMdl,125 pm g

- 't h

' Fig. 22. Microstructure ofArchive Material qller Transient Heat Treatment at 1000*C and Cooling Rate of 100*C/ min.

e,a2,,#

b ' f t

}'Q}xMtNQg$h'm.

N hh h

4 %jid[Gpl9 4 i$i,,

Lw,.;p iVoyp:

i

$4M~t@b$M.

~

p3

?Y h4 Y!

fk -

manak. f N [ ke 125pm_1 ewe i

Fig. 23. Microstructure of Archloc Material after 'Wansient Heat Treatment at 1100*C and Cooling Rate of 100*C/ min.

q 25 s.

a5jw

.%/

4t N,.

7 gre se 97 g-

-- h"h #, [$ vw6

, w. Tl> &g 2s Aw;I \\}

'"4 o

g e,3 4

r.y k

e; y

=!

Fig. 2K ' Microstructure of Archive Material qfter Transient Heat Treatment at 1000*C

\\

with Heating Rate qf =100*C/s (6000*C/ min) and Cooling Rate qf 10*C/ min.

)

s*w* dEbr/h ushh[w.- hI qu Ib 0

s 4

d&, C %m),d yb e fM@p%yNj)[6 1

E gs9&i A3 s.

d{E[ "y'M m

YII' y Y

O T

&{,* ? $Y,?*#pi$hnb$$%Q ti hihh M

125 itm? j 4:

i Y d.d iYL k

$ $ 5 4t h $ b b;l. N $ U n i n w.. m h b.

3 Rg. 25. - Microstructure of Archive Material qfter Transtent Heat Treatmen at 1100*C with Heating Rate of #100*C/s (6000*C/ min) and Cooling Rate of 10*C/ min.

e

'i) *,

g-

20 thick immediately below the stainless steel cladding had reaustenttized during the weld cladding process, and a gradation of microstructures was present in this heat-affected zone. A thin (<100 pm) decarburized layer had fonned in the base metal w;nedWely below the cladd!ng b"ause of diffusion of carbon from the reaustentttzed base metal in the weld cladding. 'Ihe portion of this layer in contact with the cladding was cooled quickly enough after cladding depo

tion by the quenching effect of the mass of cold underlying plate material to produce a thin

(-20 pm) layer of martensite at the base metal / cladding interface. The remainder of the reaustenitized layer consisted of tempered bainite. with a gradation of grain siz:s from relatively coarse (ASTM 4) near the cladding to fine (ASTM 7) at the interface with the non reavstenttized base metal. The delta ferrite content of the weld cladding was found to be 4 to 5 % by investigators at Equipos Nucleares, S. A.

In Spain by metallographic techniques, and a value of -3 % was obtained by the Centre d' Etudes Nucleaires de Saclay in France by means of magnetic measurements.

1 Within the non reautenttized base metal that makes up the bulk of the sample thickness, the microstructure consisted of relatively fine grained tempered bainite in a banded conflguration. A detailed examination of the banding by researchers at liarwell in the U.K. determined that the banding renected the twc, forms of bainite that were present. The darker-etching bands contained a fairly uniform d.'stribution of second phase carbide particles and were identined as lower bainite, in the lighter etching bands, the carbides tended to be more aligned in rows delineating lath and lath packet boundaries, and these regions were identitled as upper bainite. A ^;.y small regions containing essentially no carbides, possibly islands of primr y ferrite, were also occasionally observed. The !!arwell investigators determined that the base metalJust below the clad surface consisted of essentially 100% lower bainite, but this fraction fell to about 20% near the center of the plate, with the balance being upper bainite. At the outer (non clad) surface of the spec; men, the distribution of phases was about 60% lower bainite and 40% upper bainite.

The nature of the carbide phase present was also investigated at Harwell by means of electron microscopy and electron diffraction performed on carbon extraction replicas taken from the sample. The only phase detected was cementite (Fe3 ), and X ray diffraction analysis determined that a few percent of the Fe in C

this phase had been replaced by Mn and Mo. No evidence of M23CG, Mo2C, or AIN precipitates was seen, although some of these phases may have been present in small quantitles. The Mo2C phase has been observed in a heat of a conventionally quenched and tempered A533 Grade B steel by researchers at the Valtion Teknillinen Tutkimuskeskus (Technical Research Centre of Finland, or VTr).3 L

1

27 liardness proflies along the length of the through thickness specimen were determined by several participating laboratories. A peak macrohardness of =225 VilN was typically observed in the base metal immediately below the cladding. The hardness values throughout the remainder of the specimen were typically on the order of 185 to 200 VilN, with a tendency for the region near the center of the specimen to be slightly softer than the regions nearer the surfaces, Microhardness determinations made at liarwell detected a variation in hardness across the banded microstmeture, with the darker etching lower bainite phase having an average microhardness of 310 VilN and the lighter etching upper bainite phase having an average hardness of 277 VIIN. The fact that the microhardness values are greater than the macrohardnea values from the same materialis not unusual, because hardness values tend to decrease with increasing applied load for loads greater than about 1 kg. The liarwell microhardness measurements were determined with a 25 g load, and the macronardness measurements were detennined with a 30 kg had.

4,2 Summary of Mechanical Testing Program Six tensile tests (three duplicate tests at room temperature and three at GOO'C, all at strain rates of 4 x 10"1/s) were to be conducted by the OECD partner laboratories participating in the round robin mechanical testing program. In addition, three stress rupture tests were to be conducted at 600'C, The suggested stress levels for these tests were 215,155, and 110 MPa, which should result in stress rupture lives on the order of 1,10, and 100 h, respectively, A more extensive stress rupture testing program was conducted by ANL to generate a stress rupture curve for the archive material from <1 h to =500 h to failure, The round robin tensile and stress rupture tests described above were to be conducted on rectangular and/or circular cross section specimens (referred to as flat or round specimens, respectively), depending on the preference of the individuallaboratory, The design adopted for the flat specimen is shown in Fig. 26, and that for the round specimen is shown in Fig. 27. These designs were chosen to satisfy AS'lhi Standards E8 and E139, as well as applicable standards of the Deutsches Institut for Normung (DIN). The test specimen conflgurations requested by the various participants in the testing program are summarized in Table 3.

The flat specimens were prepared by ANL and distributed to the requesting laboratories, lloles in each end (for grippe i the specimens during testing) were smitted so that the various laboratories might drill holes with diameters appropriate for their testing equipment, in the case of the round specimens (Fig.

27), grips and extensometry vary so greatly from one laboratory to the next that it was not nossible to r. gree on a standard design. Figure 27, therefore, indicates the

28 DIA = 6.3 to 6.5

+0.1, 0.0

12.0 **- 15.0 -- : :

24.0 --*

+ 4.00 t 0.08 +

v 8-

-(+s a ;;;

r v

~

R = 7.0 c

78.0 i 1.0

> 3.0 -*

r Fig. 26. Design ofFlat Tcnstle and Stress Rupture Specimensfor Round Robin

\\

Testing Program on Archive Material. All dimensions are in mm.

i a

- 24 min. -*

+

v I

s s

4.D0 i 0.08 8.0 min.

l.).

f

+

h 95 min, e

=

R = 4 min.

Fig. 27. Design qfRound Tensile and Stress Rupture Specimensfor Round Robin Testing Program on Archive Material. All dimensions are in mm: those.

labelled min, are minimum dimensions.

l 29 j

Table 3.

Test Specimen Cortflgurations Utilizedfor Round Robin Testing of Archive Material.

Tensile St ess Rupture Participant Specimens

'ipecimens Belgium Flat Flat France Round & Dat Round & flat l

ITO Round & Dat Round & Dat Italy Flat Flat Spain Round Round Finland (metallography only)

U.K.

(metallography only) suggested configuration for the specimen gage section: details of the grips and the possible presence of extensometer shoulders were left to the individual laboratories. Specimen blanks, rather than machined specimens, were sent for the round specimens. All of the mechanical test specimen ' material came from Archive Plate 4 2, and further details on the material location are given in a previous progress report.2 4.3 Results of Tenslie Tests Condlicted at ANL Tensile tests on the archive material in the as received condition were conducted by ANL at both room temp..'ature and 600'C. The tests were conducted on flat specimens (shown in Fig. 26) in conformance with ASTM Standard E8, and the room temperature yield strength was determined by the 0.2WoiTset strain method. The strain rate used in the tests was 4 x 10 4 s*l.

'Ihe results of the room temperature tensile tests conducted at ANL are summartzed in Table 4. Tests were conducted on specimens cut with their tensile axis parallel to the apparent rolling direction of Archive Plate 4-2, as well as on specimens oriented at 45 and 90' with respect to the rolling direction. The tensile specimen locations relative to the thickness of Plate 4 2 is indicated in Fig. 28. The first set of tensile Epecimens oriented parallel to the rolling direction (tests 1442,1443, and 1444 in Tible 4) was cut from Pieces 1 and 2 (Fig. 28).

The gage section of these specimens was -13 mm (0.5 in.) immediately above or below the center of the plate. The second set of tensile specimens parallel to the

Summary of Room-Temperature Tensile Data Obtained by N1 for A533 Grade B Steel Archive Material

Table 4..

from Test Specimens Oriented Parallel 45'. and 90* to the Rolling Direction (RD).

Paralk1 to RDb^

Parallel to RDC 45' to RDC 90* to RDc Test -Test Test Test Test Test Test Test Test Test Test Test Property 1442 1443 1444.1456 1457 1458 1452 1454 1455 1448 1450 1451 Tensile Strength (MPa) 589.

601~

587 607.614 612 608 611 608 598 607 606 Yield Strength (MPa) 443-422 441 454 464 460-456 460 458 446 454 454 Uniform Elongation (%)

12.7 - 14.2 14.5 13.9 13.7 13.5 12.9 13.5 13.6 13.9 13.5 14.2 Total Elongation (%) _

20.5 20.3 19.5 19.0 19.1 19.4 20.8 19.2 18.4 19.3 16.8 20.3 Reduction of Area (%)

68.1 65.3 68.0 69.6 68.6 70.3 68.4 68.6 68.7 70.1 70.3 70.5 g

4s-l. 'Ihe fabricatkm heat

~

Ihe material was tested as formed plate in the as-fabricated andation at a strain rate d4 x 10 treatment consisted of hot pa4. auster*M at 871-899 C for 5.5 h and brine-quencheng. tempering at 649 to 677'C for 5.5 h and air-cooling, and stress-relieving at 607*C for 50 h.

bSpe cut frtun Archive Plate 4-2 so that the gage 1erigths were within 13 mm (0.5 inJ of the center of the plate (Pleces 1 and 2 in Fig. 28).

cSph cut from Archive Plate 4-2 so that the gage lengths were -38 mm (1.5 inJ from the center of the pinte toward the inner (clad) surface (Plec: 3 in Fig. '!8).

3

+

l 31 l

Clad surface i

u

,, : ' j,.............

.a mm

'b

',l; n

e 25 mm (Center of

,,,,,,,,,,,P---------- ~.g ~

25 mm plate) 9 Place

.<,*'l"""~"" "./

A s

thickness 9

(-140 mm) g '.

gg

/

u i

Fig. 28. Diagram Showing incationsfmm Which Tensile Specimens Were 7hken Relative to the Center of Archive Plate 4 2. Tensile specimens parallel to 1

the mlling direction were cut from Pieces 1 and 2, and spccimens parallcl, 45*, and 90* to the rolling direction were cutfrom Piccc 3.

l

- rolling direction, as well as the specimens oriented at 45 and 90* to the rolling direction, were cut from Piece 3 (Fig. 28). This places the gage section of these 3

specimens =30 mm (1.5 in.) above the center of the plate, toward the inner (clad) i surface.-

A comparison of the tensile properties of the material from Piece 3 for all three orientations with respect to the rolling direction indicates no apparent

-l l

variations in either strength or ductility. This finding is important, because it l-indicates that sample orientation will not have a strong influence on the mechanical properties of specimens cut from the lower head of TMI-2.

i On the other hand, a slight variation in strengths is observed between the specimens parallel to the rolling direction cut from near the center of Plate 4-2 (Pieces 1 and 2) and those cut from nearer the clad surface (Piece 3). The average j

tensile strength of the former specimens is 592 MPa whereas that for the latter it is 611 MPa, a difference of about 3%. The corresponding average yield strength values are 435 and 459 MPa, for a difference of about 5% This slight strengthening effect nearer the plate surface is probably associated with the i

somewhat faster cooling during heat treatments, and it may also be influenced by i

I 32 the weld cladding process. it should be noted that Piece 3 is not sufficiently near the clad surface for its microstructure to be significantly altered relative to Pieces 1 and 2.

The tensile data obtained at 600'C are summarized in Table 5. The specimens were sectioned from Archive Plate 4 2 as described above for the room-temperature tests. Again, no apparent variations in either strength or ductility are observed as a function of orientation with respect to the rolling direction. In addition, the slight variation in tensile properties observed at room temperature between specimens cut from pieces 2 and 3 is not observed at 600'C, possibly because the slight strengthening effect present in specimens nearer the plate surface has annealed out at this test temperature.

4,4 Results of Round Robin Tensile Testing Program The data obtained by the OECD partner laboratories in the round robin tensile testing program on the as received archive material are presented in Tables 611.

Spain tested round specimens: Delgium, France, and Italy tested flat specimens, and tw FRG tested both. Additional tests by France on round specimens have not yet been completed.

The tensile data from the OECD partner laboratories and ANL are summarized in Tables 12 and 13. Also included in Tables 12 and 13 are average tensile properties obtained by the National Research Institute for Metals (NRIM) in Japan for five heats of A533 Grade D steel, not including the archive material 5 and data for a single heat of A533 Grade B steel subjected to a similar heat treatment obtained by the VIT in Finland.3,6 Good agreement is generally seen, although the room temperature data from the FRG on flat specimens indicates unusually low values for the yleid strength. Some scatter in the total elongation values obtained by the various laboratories at 600 C is also observed. However, the results indicate I

that the overall reproducibility in tensile data among the partner laboratories is quite good.

4,5 Results of Round Robin Stress Rupture Testing Program The data obta!ned by the OECD partner laboratories in the round robin stress-rupture testing of the A533 Grade B steel archive material are presented in Table 14 and plotted in Figure 29. The specimen configurations were the same as those L

used in the tensile tests described above. The tests to be conducted at the CEA in France have not yet been completed.

Table 5.

Stusuicry of 600T Tensue Data Obtamed by ANLfor A533 Grade B Steel Ardtive MaiertaPfrom Test Specimens Oriented Parallel. 45*, and 90* to the Rolling Direction (RD).

Parallel to RDb Parallel to RDC 45' to RDc 90' to RDC Test Test Test Test Test Test Test Test Test Property 1442 1443 1444 1467 1469 1464 14fi5 1460 1463 3

Tensile Strength (MPa) 296 298 301 295 295 297 296 295 293 Yield Strength (MPa) 282 283 280 284 285 290 263 279 276 Uniform Elongation (%)

3.8 3.8 3.4 -

3.0 3.1 3.0 3.3 3.3 3.2 Total Elongation (%)

29.2 29.6 28.2 29.7 30.9 27.9 30.8 34.1 33.9 Reduction of Area (%)

90.9 87.8 90.3 93.1 90.6 91.2 90.4 90.7 90.1 g

i

}

aThe material was tested as formed plate in the as-fabricated condillon at a strain rate of 4 x 104 s-1. 'Ilie fabrication heat j

treatment conststed of hot-pressing. austenitizing at 871-899'C for 5.5 h and brine. quenching. tcu+.ciing at 649 to 677'C for 5.5 h and air-cooling. and stress-relieving at 607'C for 50 h.

bSpcGucas cut from Archive Plate 4-2 so that the gage lengths were within 13 rmn (0.5 in.) of the center of the plate (Pieces I and 2 l

in Fig. 28).

cSpeGucus cut from Archive Plate 4-2 so that the gage lengths were ~38 mm (I.5 in.) from the center of the plate. toward the inner (clad) surface (Piece 3 in Fig. 28).

i i

i 4

[

i l

34, Table 6.

Summarts of Tensile Data Obtained by the Studtecentrum voor Kernenergte/ Centre d' Etude de (Energie Nucleatre (SCK/CEN) of Belgium for A533 Grade B Steel Archive Material. Flat Specimens.a Room Temperature 600'C Spec.

Spec.

Property A42 13 A42 14 B42 21 A42 12 B42 20 B42 24 Tensile Strength (MPa) 623 611 616 328 320 322 Yield Strength (MPa) 447 451 454 308 306 311 Uniform Elongation (%)

13.3 14.6 13.6 1.5 1.4 1.4 Total Elongation (%)

20.1 21.4 22.4 28.3 31.7 29.1 Reduction of Area (%)

65.5 59.5 59.7 80 82 82 aThe material was tested as flat tensile specimens machined from formed plate in the as-fabricated condition. The strain rate for the tests is not specifled; the yield strength values reported at 600*C were obtained with a 0.2% offset strain. The fabrication heat treatment consisted of hot pressing, austenittzing at 871899'C for 5.5 h and brine quenching, tempering at 649 to 677'C for 5.5 h and air-cooling, and stress-relieving at 607'C for 50 h.

Table 7.

Summary of Trnstle Data Obtained by the Centre d' Etudes Nucleatres de Saclay (CEA) of Francefor A533 Grade B Steel Archive Material. Fiat Specimens.a Room Temperature 600'C Spec.

Spec.

Property B986 B987 B988 B983 B984 B985 Tensile Strength (MPa) 608 611 607 303 303 302

^

Yield Strength-(MPa) 448 451 447 280 281 285 Uniform Elongation (%)

10.8 9.5 10.8 1.4 1.4 1.0 Total Elongation (%)

23.0 19.4 21.8 34.8 32.8 38.9 Reduction of Area (%)

66 61 65 81 82 84 alhe material was tested as flat tensile specimens machined from formed plate in the as-fabricated condition. The strain rate for the tests is not specified; the yield strength values -

reported at both room temperature and 600*C were obtained with a 0.2% offset strain. The fabrication heat treatment is described in the footnote to Table 6.

35 Table 8.

Summar1) of Tensile Data Obtained by the Staatliche biaterialpnifungs-anstalt (biPA) of the FRGfor A533 Grade U Steel Archive hiaterial, Plat Specimens.a Room Temperature 600 C Spec.

Spec.

Property A42.23 A42.24 A42.25 A42.22 A42.15 A42.16 Tensile Strength (MPa) 624 604 613 322 323 318 Yield Strength (MPa) 336 386 446 288 284 289 Uniform Elongation (%)

Total Elongation (%)

25 23 24 34 33 29 Reduction of Area (%)

60 59 64 80 81 80 GThe material was tested in conformance with Procedure DIN 50145 as flat tensile specimens machined from formed plate in the as fabricated condition. The strain rate for the tests is not spectfled; the yield strength values reported at both room temperature and 600'C were obtained with a 0.2% offset strain. The fabrication heat treatment is described in the footnote to Table 6.

Tb ble G.

Summary of Tensile Data Obtained by the Staatliche hiatcrial-pnVungsanstalt (MPA) of the FRGfor A533 Grade B Steel Archive Matcrial, Round Specimens.a Room Temperature 600'C Spec.

Spec.

Property D1 D2 D8 D3 D4 Tensile Strength (MPa) 627 624 594 321 313 Yield Strength (MPa) 456 458 433 259 269 Uniform Elongation (%)

Total Elongation (%)

24 24 20 44 35 Reduction of Area (%)

70 69 65 88 88 aTest conditions are identical to those described in the footnote to Table 8, except that round tenstle specimens were used,

36 Table 10. Summary qf Tensile Data Obtained by the Comitato Nazionale per la ricerca e per lo sviluppo dell'Energia Nuclcare e delle Energic Alternative (ENEA) ofItalyfor A533 Grade D Steel Archive biaterial. Mat Specimens.a Room Temperature 600*C Spec.

Spec.

Property 142 143 144 145 146 147 Tensile Strength (MPa) 612 578 587 316 304 205 Yield Strength (MPa) 457 438 437 290 278 277 Uniform Elongation (%)

10.2 10.8 12.0 1.7 2.0 1.8 Total Elongation (%)

20.5 21.5 22.~7 30,7 34.8 30.0 Reduction of Area (%)

67.8 69.3 63.9 88.9 86.5 85.4 aThe material was tested as flat tensile specimens machined from (grmed plate in the as-fabricated egndition. The strain rate for the tests was -5 x 10 5 3 3 in the clastic range and

=3 x 10 4 s 8 in the plastic range; the yield strength values reported at both room temperature and 600*C were obtained with a 0.2% oliset strain. The fabrication heat treatment is described in the footnote to Table 6.

Table 11. Summary of Tensile Data Obtained by Equipos Nuclearcs, S.A. (ENSA) and Centro dc inspecct6n y Asistencia Tecnica. S.A. (CIAT) of Spainfor A533 Grade B Steel Archive biatcrial.a Room Temperature 600 C CIAT ENSA CIAT Spec.

Spec.

Spec.'

Spec.

Property D4 D5 D-6 D-7 D8 D9 Tensile Strength (MPa) 595

.b 597.8 301 303 319 431.2 283 286 307 l

Yield Strength (MPa) 440 Uniform Elongation (%)

Total Elongation (%)

24.0 22.7 25 23 24 Neduction of Area (%)

69.7 84,8 85.9 86.7 I

aThe material was tested as round tenstle specimens machined from formed plate in the as-fabricated condition in conformance with Procedure ASTM A 370. The strain rate for the tests is not specified; the yield strength values reported at 600'C were obtained with a 0.2%

' offset strain. The fabrication heat treatment is described in the footnote to Table 6.

bSpecimen damaged during fabrication, i

i t

Table 12. Comparison ofAverage Values ofRoorn-Teuw uiure Tensile Properties Obtained by Laboratones Participating in the Round-Robin Tests on A533 Grade B Steel Archive hfaterial Belgium France FRG Italv Spain USA NRIMb vg re Property Flat Flat Rounda Flat Round Flat Round Flat Tensile Strength (MPa) 617 609 614 615 592 596 604 586 626 Yield Strength (MPa) 451 449 389 449 444 436 451 435 483 Uniform Elongation (%)

13.8 10.4 11.0 13.7 Total Elongation (%)

21.3 21.4 24.0 22.7 21.6 23.4 19.4 26 25 Reduction of Area (%)

61.6 64.0 61.0 68.0 67.0 69.7 68.9 71 69

Ihe tensile testing of round sp ; - m at the CEA in France has not yet been completed.

bNRIM = National Re.M Institute for Metals in Japan. Data are average values for five heats of material (not including the archive material). as reported in Ref 5. The material was tested as a plate in the as-fabricated condition; the strain rate was not specified. 'Ihe fabrication heat treatment consisted of hot-rolling. austenttizing at 920*C for 1.5 or 9 h and air-cooling, tempering at 650"C for 3 h or 680'C for 12 h and air-cooling, and stress-relieving at 625'C for 10 h and air-cooling.

CVIT = Valtion Teknittinen Tutkimuskeskus (Technical Rmid Centre of Dnland). Data are for a heat of A533 Grade B steel containing 0.21 wt. % C (heat F). as reported in Refs. 3 and 6. Prior to testing, the material was normalized for 4 h at 925T and cooled at 3*C/ min, austenitized at 900 C for 4 h and quenched at 50*C/ min. tuupu ui at 665'C for 4 h and cooled at 50'C per hour, and flnaDy stress-relieved at 620*C for33 h and cooled at 50T/h.

Table 13. Comparison ofAverage Values of 600*C TensGe Properties Obtained bu laboratories Pdrticipating in the Round-Robin Tests ort A533 Grade B Steel Archive Matenal Property Belgtum France FFG Iialv Spain USA NRIMb Flat Flat Rounda Flat Round Flat Round Flat Tensile Strength (MPa) 323 303 321 317 305 307 295 306 l

Yield Strength (MPa) 308 282 287 264 282 292 282 244 Uniform Elongation (%)

1.4 1.1 1.8 3.7 Total Elongation (%)

29.7 35.5 32.0 39.5 31.8 24.0 29.0 36 Reduction of Area (%)

81.3 83.3 80.3 88 86.9 85.8 89.7 86

Ihe tensile testing of round W.Luun at the CEA in France has not yet been completed.

Data are average values for five heats of material (not including the archive material), as reported in Ref. 5. The material was b

tested as plate in the as-fabricated corxittion; the strain rate was not specified. The fabrication heat twebiwit consisted of hot-rolling, austenttidng at 920'C for 1.5 or 9 h and air-cooling, tempering at 650 C for 3 h or 680*C for 12 h and alrW and stress-relieving at 625'C for 10 h and air-cooling 7

39 Table N.

Stunmar}t of Stress Rupture Data Obtained by OECD Laboratories on A533 Grade B Steel Archive Material Data Stress Time to Elongation at Reduction Source (MPa)

Fall (h)

Failure (%)-

of Area (%)

SCK/CEN 110 176.7 18.2 21 (Belgium) 155 32.2 25.6 37 215 2.9 42.1 76 MPA, flat 110a 108.9 23 15 (FRG) 155 21.6 24-28 215a 1.05 12 76 MPA round

-110 177.7 19 22 j

(PRO) 155 25.4 20 51 215 2.4 44 82 ENEA 110 179.2 3.5 (Italy) 155-32.8 9.4 215 2.5 13.2

-81 TECNATOM 110 214.5 20.0 5.0.

(Spain) 155 46.9 36.0 32.0 215 3.75 52.0 69.8 j

ANL 89.6 355.2

-17 l

(USA) 119.9 112.5 15.8 18-l 137.9 56.2

.18.0

-28 175.1 14.7 34.0 42 l

188.4 8.3 45.5' 74 239,3 0.9 43.2 89 I

I aFalled at the extensometer notches.

l Again, the agreement among the laboratories is generally quite good.. However.

l two of the three flat _ stress rupture specimens tested by MPA failed at.the.

1 extensometer notches and had slightly lower stress rupture lives than were observed in comparable tests at the other laboratories. The curve obtained by l

NRIM which was extrapolated from lower temperature data.using a Manson-flaferd time temperature correlation.5 represents the average behavior of five heats of A533 Grade B steel, not including the archive material, t

I

40 103 Stress Ru sture of

~

A533B Stee at 600*C

-D D O

^ 10 2 m

g V SCK/CEN(Belgium)

"g g

A MPA flat (FRG)

A MPA round (FRG)

O ENEA(Italy) 10 1

+ TECNATOM (Spain) i D AHL (USA)

- NRIM extrapolated (Japan)

,,,, n i ii

,,,,n,i

,,,, i i.,1

,,,,of

,,i i

0 2

8 4

10 10' 10 10 10 Time to Failure (h) ny. 29.

Comparison of Stress Rupture Data Obtained bt1 ANL and the OECD Partner Laboratoricsfor A533 Grade D Steel Archtuc Material with Extrapolated NRIM Data for Average Dchautot of Rue Heats of Matettal.

5 5

Decontamination of Samples Recovered from the TMI 2 Lower Head 5.1 Sample Description A map of the TMI 2 lower head, showing the extent of damage observed, the positions of the instrumentation nonles, and the locations of the samples removed for examination is shown in Fig. 30, Fifteen samples have been recovered! 11 samples without nonles and 4 with nonles. Figure 31 shows the configuration and approximate dimensions of a typical sample, in this case without a nonle. The 65 mm depth of the samples is slightly more than halfway through the thicimess of the lower head.

The samples removed from the TMI 2 lower head have been tentatively classified into categories oflittle damage, moderate damage, and severe damage, based upon the observed damage to in core instrumentation, nonles, and guide tubes, as well as available infonnation concerning the accident. This classification is summarized in Table 15.

41 A

B C

D E

F G

H K

L M

N O

P R

NORTH r

y w

/

N

/

\\

y

j:.' : ::.:

tt:

N is

[

\\

14

F:

+ :

/

M::'

'IIf h M :

~

idt22+fll 13

/

_ :m$:

, :h$

12

[

[h L

. [.k,f 11

[:

h ki::$::

.:f::: $.E"

- C$$:

bf::

10

f" i fH2 I

g f

~ g4g ;-

"4"

p"' ;

g ;

fp k $;

i

$:' I

O "E:: )4

!::NIE 8

x-

-2 7

{

"f;:

... f:..

' IA

.h. 31D

.E

\\

3: 4 Nt e Mc :.

g:. :4 s

(

it a :

p W-3; :p $:,

' j:

s

+

\\

TIE f.::

p :

f:

4

\\

p

+; t

4 :k t

a

\\

:N:::

2

\\

1 f::

POSITIONWITH TE~

MINOR DAM AGE TO IN-CORE f

INCORE INSTRUMENT GUIDE TUDE (IIGT)

INSTRUMENTATION (DAMAGE TO M 6' THIN SECTION)

~

NEW THERMOCOUPLE

~ ~]!

MINOR DAM AGE TO IN CORE JUNCTION FORMED DELOW

.4 NOZZLE (MORE THAT HALF OF E.

LOWER GRID PLATE NOZZLE STILLINTACT)

~ T

"^^* ^" ^ 5 ' '"

"5 '"$' au"'"'

"' 5

~fb.

INSTRUMENT GUIDE TUBE (IIGT)

I TUDE IN COOD CONDITION (DAMAGE TO THICK SECTION OF llGT)

~

MAJOR DAM AGE TO IN. CORE IT IN-CORE NOZZLE IN GOOD NOZZLE MORE THAN HALF OF IS CONDITION

-I NOZZLE GONE)

~~

II CRACK INDICATIONS IN VESSEL LOCATION OF SAMPLES REMOVED CLADDING FOR EXAMINATION 1

ng. 30.

Map qf TMI 2 Lower Head Showing Edent of Damage Observed, Positions ofInstrumentation Nozzles, and Locations ofSamples Removed for Etamination.

e i

~

42 r

Stainless Steel Clad i

inner Sudace of

,A65@* /

Lower Head

/

\\

- =70 mm

/

\\

h

\\

~65 mm o

ng. 31.

Cot \\flguration and Approximate Dimensions of a 711pical Sample Removed from the TMI 2 Lower Head.

5.2 - Sample Decontamination The as received lower head samples typically display associated activity levels i

of the order of 1 R/h of combined beta gamma radiation. In addition, considerable loose alpha activity is t/pically present on the sample surfaces. Thesp activity levels require that the samples be handled remotely and that they be decontaminated before any hands on operations are performed. The remote handling is carried out

~ in ANL's Alpha Gamma 110t Cell Facility.

As received samples are first photographed from several angles at a

- rngnification'of 1x, and any unusual surface features are additionally photographed

)

at higher magnifications. A two step decontamination procedure is then used to.

clean these samples. In the first step of this procedure, the samples are immersed

- in_ a mixture of equal parts of 37% 11C1 and water for -90 min to chemically remove.

a thin surface layer. This step reduces the overall radioacth'ity to -100 mR/h of beta gamma radiation and the loose alpha activity on surface smears to -1000 disintegrations / min. The samples are then clean enough to be transferred from the Alpha Gamma liot Cell Facility to a machine shop area set up to handle radioactive materials.11ere, a thin layer is removed from all of the original surfaces of the sample by mechanical ~ milling in the second stage of the decontamination --

process. This typically reduces the radiation level of the samples to less than e.,.

~,-w,,4,--

_m..,

.~.mo,n.,._,..,

,-,,.,_,,m

.r-v,-

43 Table 15. Tentative Classillcation of TMI 2 louer ficad Samples lident(fled by the Coordinates in ng. 30) in Terms of bpected Damagefrom Core Meltdown Accident.

Little Damage Moderate Damage Severe Damage 11 4 F 10 E6 K 13 K7 E8 M 11 L9 F5 M8 G8 D 10 (nozzle) 118 (nozzle)

E Il (nozzle) 115 (nozzle) i 10 mil /h with no loose activity. The samples can then be handled directly for sectioning, metallography, and the preparation of mechanical test specimens.

6 Examination of TMI-2 Lower Head Sample E-6 6.1 Initial Examination The first sample to be removed from the lower head of the 'rMi 2 reactor by MPR Associates and sent to ANL was Sample E 6. This sample contained a portion of a surface cladding crack that largely encircled an adjacent instrument penetration nozzle at coordinates E 7 in Fig. 30. (As indicated in Fig. 30, Sample E 6 actually cuts across the corner of coordinates E 6, E 7, F-6, and F 7.) A cursory visual examination of Sample E 6 by personnel at TM12 immediately after removal suggested that the surface crack penetrated essentially through the entire depth of the specimen, i.e., more than 6 cm (2.5 in.) into the lower head. The purpose of the ANL examination was to characterize the extent of the cracking and to determine its cause.

The sample was unpacked, surveyed, and placed into the clean transfer area of the ANL Alpha Ganuna llot Cell Facility for more careful examination. The sample was photographed and measured dimensionally, and samples of loose material were scraped from the cladding surface for subsequent analysis by scanning electron microscopy (SEM). The surface crack that penetrated the stainless steel cladding was readily apparent, but because of the roughness of the cut surfaces, the depth of cracking into the sample could not be determined visually Figure 32 shows the clad face of the sample, corresponding to the inside surface of the lower head, and E

44 g>

1 F.E gar: t;

.+.,

=.,c;_=.m w m ew

..e

.y i G Q'i b8:if

y F

pgk

[ c, I -N,.,,M gjQgd34.hw

" Argonno National

. A

,4..A Laborotory IMg. 32. Top View of'Ihfl 2 Sample E-6 As Received, Sholving Crack in Clad Inside-Surface ofIntver ilead.

the crack in the cladding is clearly visible. Figure 33 is an end view of the sample, showing the as cut face obtained by metal disintegration machining (MDM) on which a crack extending into the depth of the sample had been tentatively identif!cd at the plant. The cladding surface is to the left in this photograph.

Initial sectioning and metallography on Sample E-6 were carried out without prior decontamination to avoid removing any material from the crack faces, Two metallographic sections were taken through the thickness of the sample, parallel to the MDM cut face, as shown in Fig. 34. The first metallographic section, taken el em (0.4 in.) from the MDM cut face, is shown in Fig 35 in the region of the crack. The stainless steel cladding, which is unctched here, lies to either side of

,he crack, and the etched A533 Grade B steel base metal lies below the clad surface. The crack may be seen to extend only a short distance (-3 mm) into the base metal. The thick gray region in the base metal immediately below the crack is mostly iron oxide, but it, as well as the side walls of the crack, appear to be covered by a thin surface layer of resolidified material from the reactor core. The identity the various phases present will be discussed in more detail below.

The second metallographic section indicated in Fig. 34, located -3.2 cm (1.25 in.) from the section described above, is shown in Fig. 36 in the unetched condition. The main crack is not as wide here, and it extends only =2 mm into the

)

w ww M

i

,a y{p } g

' -$ek s* % *'.

.'Q t **

fN

Yf f

hhkN h,h{,k$ hff!Nkhh$

' q*^ p Q

-y m

.F kf@p1%xm%e%nw#a$

%wek.s.;4 gG 9%

. 7 %'g,1 k'Iyeyg.g w wW y~

y

.C y y~..

.. gy +IN*%[..

7,,

i*e

.k O'

'y

[ -

p' fl fjgp glM Q'l E9

'(TM@w$iywse,fiy}$px(Q%

d m

W E

tm n

r%n*.Vf$Cif$ay,h:g'nl.g@qwy,q y.

, Am+ygy;pp RR i

x be

.pw

s.,f,.n,4,.,g, sa p,p,s g

46 p

N-kk

,., y{U [

n s g ? + m@ 6, h {is

s..-, *,./ t *,

,g i;

Yf

'ha S$

!'h - KJ4' jf4-1 ef,q v

.Q,

.so Fggi vf(!!pe S g !$

3Y.v!h.a'$hlh:e.:Q..f.

'bfk 3

5f A

,1 %,y o 4d'T*,T $qQJi.d ff[ht*;"%'*

f v

fW/

'dty.%.y,a.,;, d.c r ;d,r.4.g.,.

f5p

.,r e

p

? '*N'ob ',ry O

, h,)j' ;

ct i, ['d $'3 A)s,-

. h*1]jg"l% s s(n

,,w vna

-v,v.. -e c ~d1/

10mm

',,9 h[,f, H

.m.

-u

),

nu :n 1:rul new v.! TMI : senvie 1:+ stunvirui ucuannev<;rcaum yachiru-a nice m W1uch t 'rnckuni lh;;! liern Tenunu'clu hientp!vd

46 i

p*v,

,, - m.

,,.,.m_,,,mq,g.

{

'y % fif t.t fnetall>

r, f-Py r. q, a, a.w pp,W.w~

w y4 4,y

-. m' S graphic section p

~

ys-

'2 h f,4.,_ m o%",3 Ap f&.),

f' s

../,.

b i

,.~

.7 y

9 m l W. %l,j :,. ]N'.

u, w

} f ',

f* ' * *,,,

b :b

/y f5

~h

.v.

r

,,

ly 7

+

y 6,bllvs M

M%

4

- ~,~ m uun u.x ~ n.w winub;at t

+s

~m.

Seconct metatio-25 mm grarnic tection

}

l togonno National r

Laboratory Ffg. 34.

Locations of Tipo MetallographIc Sectfons Takeri throup,1 the Thickness of Sarnple C G Parallel to the MDM Face.

base metal. However, considerably rnore secondary cracking in the cladding is evident. Figure 37 shows a higher-inagnincation view of this secondary cracking in an etched specirnen, and the interdendritic nature of the cracking is evident.

6.2 Macrohardness Measurements Following the initial tuetallography, the top surface of Satuple E G. Including all of the stainless steel cladding, was removed to a depth of -16 ruin to preserve the region in the vicinity of the surface crack. The retnaining portion of the sample was decontaininated as deserthed above, and the sarnple was sectioned as shown in Figure 38 to produce specimens for clean (uncontaminated)

Inetallographic examinations, hardness incasurements, and elemental analysis. A tlice -10 nun thick was then cut from the 38 nun-high lower portion parallel to the plane of the Ogure, and macrohardness readings were taken along the length of that nitee with a diamond pyramid indenter.

The resulting Vickers hardness values at intervals along the specimen are shown in Figure 38. The hardness values range from 251 to 298, compared with a

l l

47 i,

J

\\

n "g

s 4

\\

'i/1 h,

de

~

. s;_ _ e r.>

i>

lA o,F t

'yc

' ~,,.5 f,

.., y,.

e 4:.:?

...i.

s.

w

,,? y

., d,. _ '. ;?a

,e t

, : 6l< ?, i n...*

y:Q: b '..~.pp 6.w, m..w e t:a

, s'.!,3,

3 k

.bg -f v ['

v,, x,' y ;4. p.y.

y 6,

u go.g. p (y

...x,,. e

.g

.wp,...

. p. :.a k ;$

. y;> : b 6, h $ 5 p\\ 4, :";. a.9 J.,m (qh.s lN,k

!Y f,0.~.h?"Q'~ *l,:4yW& ;,0.l?' t R.

l,

+

I.

, y, yyy t;/f s

' ~; :s f r

s

.y l.i,Y f.[M.Y;rk,chl i"$f,,,kf g' [~[h

!Y. ' l;' h % h h:., b,Q !h h$k5 l ',

k($y y,

f S N

f y k N(*

h[Ek-[;5g[,A d

. A,y g.v s,,,1.s

, t; t -

,e r y[f 4 'it-{;[y; ;,,(,t = g' N/p,,.r. p -;'d g,'l *fi'gn I<,Q c.

. -s

,s

57. 4;

/ -m.n i

s.f.

yp&

2mm

,t 7.,

, ~, s

-t-c,Jr

,;M$y'5fl ! (,g$$f($fy

.r. Q" '. y,.1 ' "':I fE(-[d[N / :

f 1

t.[f f

atx> tory

}

Hg. 35.

Exno-blagn(0 cation Victo qf nrst hietallograpitic Section through Santple &6.

48

~

l

$5 e

kik 0

=

-- R f

.e 3

9 h

k n

- s y

.. / 'N 8

M s,

=a'

}u

,'~

f E

- #,, -,.. ~

?

,. ~.

g c

s k

3 1

3 N

49 Rhdd$@

1l Nf[%F,

p[,bitQh ')@.

,) ' )

%,, gg', r :V

^

.,2s nm Ag. 3 7.

Higher Magnification Vicw of Ebrtion of Cladding in Ag. 36 After Etching Showing Interdendritic Nature of Secondary Cracking, value of -188 for the as-fabricated archive nmterial. These observations indicate that the base metal from Sample E 6 was either in a harder state as fabricated or that it was hardened as a result of the accident. It is not clear why the two hardness values near the bottom of the specimen are greater than the remaining values.

50.

4 a

i Weld Clad i

Weld Clad 16 mm U

__q 7---

l l

M255 l

Eejsjs' I

)l( 262 l

Trea ent ni Samplo l

I Sample 255 38 mm X 251 Met.

Mot.

kl271 Sample Sample (ANL) j (B&W)

>l( 298 U

l i

i IMg, 38. Diagram Showing hiacmhardness Readings Obtainedfrom 7MI 2 inwer Head Sampic E 6 and Sectioning Diagram of the Samplefor hfetallographic Examlatation, Elemental Analysis, and Heat Treatment.

6.3-Metallographic Examination and Microhardness Measurements Following the hardness readings, the slice was cut into the four pieces indicated in Fig. 38. One of the two metallurgical samples shown was sent to Dabcock and Wilcox (B&W) for examination, and the second sample was examined i

at ANL.

The nilcrostructure of the as received E 6 base metal at a d pth of -50 tam-

-(2 in.) below the inner surface of the pressure vessel lower head is shown in Fig. 39 at a magnification of 200x. 'Ihe microstructure consists of fine bainite and possibly some martensite, and these microconstituents are segregated into alternating bands of coarser and finer etching features. 'ntis difference is illustrated more clearly 'n the higher-magnification rr..crograph of Fig. 40, which shows the interface between a coarse-etching region (top) and a finer-etching region (bottom). At a depth of =35 mm below the inside surface, the microstructure of

51

_g vr x

p

' 3,, '

o 3

~..

V

.h'

??<

y

.h

+

!* j* l hfjlpf r b j

,. p n _1 A

t pcy

-).

.,4!!% -

ig h,

+

l,,.

" %ln ;> gy:.:,'s;.(

&;,~}

i Q

s..

i.

n. -

ydhn..

hh-1 A

d! F)

[ ~125 m l

u-V%!'k

,e.. m na e 4}

=

ng. 30. hitcrostructure of As Received IC 6 Base bietal at a Depth of ~50 mm (2 in.)

below the inside Sarface of the Exwcr IIcad.

h$

fNhh 5l4f?&hh.h5 f jl?l i

,'/.'e)h[' W w)hbh'( b k

A

.?

g e w)W( E y a;#.ch;;,;g;y s' m @;:r.r q

W W e m

es(Wm w

hh, en ma%

s> ey, >,p,.g~n.;m > p% f:Ny&pfe av y b - 2

^:n ci

~

Qg

.a.

ua

,((tk,y(%aqw'g2;p:fy W-W'![k. M b'Y p yQ r.,yanA,7*'N Qt'** f h

$/ M s**> ~

?

b t(h :a,:g,,r:19 ~

g,,

W a<

w;kg*o @% w

.%< chm,.A w, rt s

r

!&A%-@3%l4' trpMip@w,ydd. ')hy A *%*(,pg47

W. 9.u a O S ^T

h. Pi U pf't./:?

YFRElk

g. -

y

% p %l;,$ yQ),ty;W Mi f W~ d e

9

Qde.

MUni% Vl[&Q& Qt:p>W:

W Q

q iOi0. W :Shh;Q Mi L {g :h l :t6

E{!Gu'?i$y: ; V 4lr]$n;9y# %f?chlYll'l??b?

5

.p. 4 nf 4M 1' 7 jQQ QW.

?

W, lY '

m

[b/W)jD7 Liq.W.@j%{.E5($M.%NAC6{

c, M-S sP B 1pMMP fF 1,a.jp,hJifS ;:,g.;'rg g; kg;pg M 4,

d 50 m, ;y

?vgNg y

yqw n;

a a.h it. dn, m y w.s

.w A;.r mx;w m;am ng. 40, Higher hfagnlllcation View of hitcrostructur& of ng. 39.

1

53 the E 6 base metal is somewhat coarser and die prior aui enite Ica 2 size somewhat larger (Fig. 41), susge.3ttog that material was austenittzed at a higher temperature. Further coarsentrag is evident in the microstructure of an area at a distance of =20 mm from the inalde surface of the vessel IFig 42).

Microhardness readtngs taken in the bands IFig. 43) revealed that the darker-etching btmds had hardnesses on the order of 260 DPN, and the lighter etching bands had hardnesses cf 450 to 360. 'niese findings were confirmed by L&W, which also found, from microprobe analyses, that the banding correlated with variations in Mo distribution in the specunen.

The fabrication heat treatment of the TMI 2 vesse* was virtually identical to that of the Midland vessel (the source of the archive material), and one would expect the base metals from the two reactor lower heads to have very 91milar microstructures in the as fabricated condition. Ilowever, the microstructure of the as received E-6 material at a depth of 50 mm below the vessci inride (Fig. 39) is dis' netly different from that of the as received archive material (Figs. 6 and 7). In fact, the microstructure of the archive material that compares most favorably with Fig. 39 is that r.cen in the archive material subjected to a transient heat treatment at 1000*C and cooled at a rate of 100'C/ min (Fig. 22). At a d! stance of 35 mm below the vessel inside surface, the E 6 base metal has a microstructure similar to that of the archive material auntenitized at 1100"C and cooled at 100'C per minute (Fig. 23). These observations suggest that the TMI 2 vessel at location E 6 was heated to these temperatures during the accident and cooled relatively rapidly, To check this hypdhesis, the campic for heat treatment from Sample E 6 (Fig. 38) was subjected to the original fabrication heat treatment prescribed for the TMI 2 lower head This fabricailon heat treatment consisted of auatenttizing for 5.5 h at 871 to 899'C and brine quenching, ternpering for 5.5 h at 649 to 677'C and air cooling, and finally stress relieving for 50 h at 607'C The small heat treatment sample from Sample E 6 was given the same heat treatment, except that it was alt cooled rather than brine-ouenched from the austenttizing temperature to simulate the approximate cooling trete a the interloc of the 137 mm (5.375 in.)-

thick original lower head piece. The resulting micrastructure, shown in Fig. 44, is virtually identical to the microstructure of the aweceived archive material shown in Fig. 7 and is distincuy different from that of the as received E 6 base metal (Figs.

39 42).

In addition, macrohardness values for the re' neat treated TMI 2 E 6 material was found to bc ~195 VHN, based upon the average of four readings. This compares favorably with the average hardness of the as received archive material

(-188) and is distinctly lower than the hardnesa of the E 6 bise metal in the as-received condition (251298). These results support the conclusion that the

53

['

k

^

t'

,, 1.

L f,,

kij$

h.

i' 1

yn

. e gpu

.."3 T

gp V

3 r

=

{$f y'.

4,g:a ij

..?,

(

h f

i 125 nm 1

h I

r a,a; rw saw. x:_._9 Ag. 41. Micrc~itna.We qf As Received &G Base Metal at a Depth of ~35 tw (2 in.)

belov> heins& Surface of the Lowcr ficad.

{t u 9.pp i 'w'g ut i

ht,

k f$ 4

^

g

$ '@N $ Id..,Q g:l:p?fy 7

.Q6 iflADy Q

J A;,

t g&y

):

Y 3 O,f5 k 4

E{. p I $ % $ b; z w % +3r1 W ;'

mM.

y

a h

k *c egg!f:;.?:gg w n

q q g#Mkk N U $$"*

Il g

Nkl

$3x<L Rg. 42. Microstructure qi As Received s : 6 Dase Metal at a 3 epth of ~20 mm (2 in.)

t clow the Inside surfu ~ of tha 1.ouer llead.

t

' 54, J

i

~

g,

}-

q e

i' h

6,'

~

I*g b.J..

~

Q>

f'

.y' i)[{.

ag g

\\.

,)

,.'j.-

f t.

g 4

'?

~

.y

~

- Fig. 43. ' Variation of Microhardness Readings in Banded Microstructure qf Baec Metal from,7Mh2Lotver Head Sample E 6. The microhardness values in the darker etching bands (indentation on left) were on the order of 260 DPN,

. and those in the lighter etching bands (indentation on right) were ~350 to 360.

i soriginal condition of the E 6 material was similar to that of the as received archive

- material and:that it was altered by exposure to temperatures on the order of 1000-1100'C during the accident..

Further support'for this conclusion was obtained from the metallographic

' examination of a second TMI-2 lower head sample, Sample K-13. This is a so-o called " hillside" sample taken from a region that was apparently not seriously i

damaged.by the core relocation (Fig. 30).. The microstructure present in Sample K-13 in the as received condition is shown in Fig. 45) -This microstructure is again very similar to that ol'the as fabricated archive material (Fig.17) and-to that of the.

..c Sample 1E 6 bas.e metal after it was subjected to the original fabrication heat

treatment ~ (Fig.?44). Furthermore, a series of hardness measurements on Sample K-13 indicated an average VHN of =184, This value 111tewise compares. favorably with the hardness of the as fabricated archive material and the reheat-treated E-6

material and is again lower than the hardness of the as received E 6 base metal.

Based on these observations, it is concluded that TMI.2 Sample K 13, unlike.

e

55

'4 pA t.p y,qr $ If'.g(..qQ. I +*...NNp@r. -

'gOge(;,-#: ph yi*

p-+<

N 5

d %8" Q s b

.!'$@sh,,gye.Qfp$jgyg+jj/g,Ag;h 2 6 # N' b (( W t

f f, d.'p.a,g.y j Ne.'

r Q

2 p.,

"g p> 3 wt pv.

m 4

t.

,r.1 4

h[.y((evkUMid9 yf[A>c[.ff,hw@pT m e

d' M.$

k.q.bv M nme

-er 6

n-Abc 9/

,kli 125 pm i) i 4

[h kN b Argonne National S L h.d.a.d*!. /f N2tyM 20(w 1.oboratory 9

+

'?Vh lm M'fe eh.ca.1 w.

.m

,.,,s Fig. 44. Microstructure of Sample &G Base Metal qller Heat Trcatment According to Original Fabrication Procedure, h

ls'h

~

J 1.

$$$h

~

s hk$$.!xR f 'whh!Qb h

i-N % ?$NYhhdh N$k.% ~

Y YIEg

.h"-n p.wf

'x

?*

g.;fg$Qy;;3gL hW gyb't

.Wp$"bq#y" M tp 9

AU p%qMk?

ff. f%

h1%W'hkhh#

Ebikd j

Nhf G

Wh.f;jfg 9;Ij lm

}[ g$1 ix:kg s

8f ~,%g$

g a.g y

y,ugg g A

1

, at.

< fp g%etg :yiq yr-k n:%.,an;4y s g

a n

j p. 4

'7 -

a u swO c m. vv wg f

Ek hM"[ d pg, (b

{$$$@N.)!Nf$.N% @I$i M d b W 5Td @$

4

$ T d f fM kbM f%he[3}&&sMalf EiLO N!

' 5 "* ' :

.E6$$g$P*y

$14 j p A'gonno National (d r

Fig. 45. Microstructure of As Received Sample K 13 Base Metal 1

56 Sample E 6, did not exceed the critical ferrite-to austentte transfonnation temperature of 727'C during the accident.

6.4 Elemental Analysis Determinations of the elemental composition of the E-6 meterial were made on the piece indicated in Fig. 38, and the results (Analysis 1) are compared in Table 16 with the vendor's heat analysis of the TM12 lower head material and the vendor's heat and average sample analyses ice the archive material. Analysis 1 Indicated a carbon content of 0.28 wt.%. This is appreciably higher than the 0.22 wt.% average sample analysis obtained for the archive material and is also above the 0.25 wt,% maximum specified for A533 Grade B steel. In addition, the silicon analysis for the E 6 material (0.11 wt,%) was below the 0.15 wt % minimum specification for this alloy. For this reason, three identical additional pieces were cut from Sample E 6 and analyzed. All three analyses indicated carbon levels of 0.25 wt.% and silicon levels of from 0.16 0.20 wt,% (Analyses 2. 3, and 4 in Table 16),

6,5 Examination by Scanning Electron Microscopy The principal phases present in the cladding crack of Sample E 6 were identifled by SEM/ energy dispersive X ray (EDAX) analysis techniques. The identity and probable origins of these phases are discussed here.

Figure 46 shows an optical micrograph of an oxidized region of the A533 Grade B ferritic steel base metal at the bottom of the cladding crack (lower left-hand portion of crack seen in Fig. 35), At least two oxide phases are present: a lighter gray phase makes up a majority of the oxide, and a thin, darker gray phase lines the top and central core of the oxide wedge, EDAX analyses indmate that while both phases are predominantly Fe oxides, Mn is also detected in the lighter phase but not in the darker. At a higher magnification (Fig. 47), some of the cracks and grain boundaries in the lighter oxide are found to be filled with an In Sn phase, The latter control assembly material from the reactor core would be expected to have a relatively low melting temperature, and it apparently was present as a liquid phase when the oxide formed.

Other particulate materials were occasionally detected as small particles in the axide or, more commonly, at the oxide surface. Figure 48 shows an SEM back-scattered electron image of the region near the top of the oxide wedge shown in Figure 46. (Because of the optics of the SEM versus those of the conventionallight metallograph used to photograph Fig. 46, the two images are reversed.) The

57 9l 3

o N N8855E$$$85 j j dueeeeeeeee w,

w

]>

x k'

e R e$

0$5 88iN3 i

ex easeee ee

,b v*

e -

8 nuaune j

o-eo eoooeoo 3

+

01 1

a" e o aaa8sasssss a

$ 7<m

=

6"d6 6666d66 be 39Y e

f n

x e

6 s El 9nE8532888s g*r j g eaeeeeeeeee 7

l b

gw i

wE R as:sim8885s i

pk p

easeeeeeeee e

e I"]a o

e

, e f

es N82s53,ss,,

re em e: eeee ee e

c j

e ma f

ussemEas6a6 8

i B

5 3

58,

1 1

b 4

g

ga i,.

{-

4 t.d.

i e

p.

w.

f C

J t

h?-

(w.

(,

.)'... k k

jg.T

\\;.'!'.,:

W s

j.,

.... =

,,[} $ y :' 3, y - 500 m i.

,u,

,i 7:, :,

1 I

~~

Ftg. 46.

Optical Micrograph of Oxidized Region in the AS33 Orade B Ferritic Steel Base Metal of the TMI 2 Lower Head at the Bottom of the Cladding Crack.

angular porous particles are shards of Um Zr -O fuel. These particles do not n

appear to have melted, but instead probably fell onto this surface after cracking and oxidation of the cladding hr.d occurred, Also present are spherical beads of Ag-Cd control assembly material that were deposited as a liquid and solidified in place, as well as Fe heads that were produced by the MDM process used to cut Specimen E-6 from the lower head.

l 59 y-fi:

f 1[;

hhk,ijb

![i4m$a;;M.x,n Q

"Nks O phNhh f

'74{m k

.-? I ;

7 we.

n 73 p

"h i.

g7p.

k' JIM M)

(

.n f

9125 pm sed w m Fig. 47. Higher Magnification View of a Portion qf the Ughter Gray Oxide Phase of Fig. 46, Showing In-Sn Phase Filling Some of the Cracks in the Oxide.

A multilayered material was found on the exposed crack faces of the stainless steel cladding and atop the ferritic oxide. Figure 49 shows a low magnification

- SEM image of such a region: Fig. 50 shows a higher-magn 111 cation view of a portion of Fig. 49. An Fe-Ni Cr rich layer containing In and Sn was present immediately adjacent.to the stainless steel cladding (top left in Fig. 46). The composition of this layer suggests that it was produced by chemical interaction of molten control l

assembly material with the stainless steel. Immediately adjacent to this is a blocky

60 aJ(.$ k f [

d'

(

s a(

w 1s, y.

Lv.,4

.s z. k(; y

_. gg7 I? y 8

$y0 t pcv;;..

y,. (,

9 s

_ N'y, k

, c,yg. -

=

,.d

q. t; ;.

~

,s e

C'i<.

500 pm i I

.I Fig. 48. SEM Back-Scattered Electron Image of the Region near the Top of the Oxide Wedge Seen in Fig. 46.

mrywwnynmp;pyyyr

,e q:. c w.A

...o

s. -

y dlW p]){i)fr3-f. }m;??

.t.c

/

%p

J: ' e? '$+}i.M 0 a x g

- n ' fy

%; ; g, y f,gn.

9 ;

3.+

y g?

9" '

G I[

p..

t

.i iW I

i' 200 nm ' i i.

i

.1 Fig. 49. Low Magn (f1 cation SEM image of the Multilayered Material Deposited on the Exposed Crack Faces of the Cladding. Arrow indicates the region magn (fled in Fig. 50.

l 61 qqr l

^U

.,k gi 1@%Ed p

3

%;.,,y

e:

q w

g, y

j k 4

S kk iM

%jjF 3$ x@!W Q q

yrd

,P w.s

.nw i

Rg, 50. Higher Magnification View of the Region in Ag, 49 Indicated by the Arrow.

Fe rich phase containing Ag, In, Cd, and Sn, along with In Sn-rich needles (central portion of Fig. 50), The thin mottled phase adjacent to th' contains all of the L

above elements plus U and Zr, From all indications, this multilayered material was deposited on the crack surfaces in the liquid form and solidified in place, and the various layers separated out during the solidification process.

No stainless steel oxide was observed on the top surface of the cladding to either side of the crack. However, occasional small cracks penetrated into the cladding from this top surface, and these cracks typically contained Ag Cd In and Fe-Sn In phases. Needles of a high Cr phase were also observed in some cracks, and shards of deposited fuel were occasionally found on the top surface of the cracks. However, these fuel particles appear to have been deposited after the crack formed and are incidental to the cracking process. - Closed tears in the cladding were likewise found to sometimes contain solidified stainless steel constituents from the core as well as trace amounts of fuel and control-rod cor_stituents.

62.

6.6 Tensile Tests Tensile tests at room temperature and 600*C were conducted on specimens machined from the as received E 6 material. The results are compared with the tensile properties previously determined for the as received archive materialin Table 17 and Figs. 51 and 52. The higher strengths and lower ductilities of the E 6 material support the earlier conclusion that the A533 Grade B material was transformed during the accident and cooled fairly rapidly to produce a hardened microstructure.

6.7 Discussion Examination of TMI 2 lower head Sample E-6 suggests that the cladding failed by a process similar to hot tearing, causing extensive cracking along interdendritic boundaries. The precise nature of the loading that produced the cracks is not clear, but it apparently resulted from thermal stresses imposed as a result of the accident. Microstructural examination of the underlying base metal and the results of_ tensile and hardness tests indicate that Sample E-6 reached temperatures of approximately 1000 to 1100'C during the accident and then cooled rapidly.

Temperatures of this magnitude would be expected to impose significant thermal stresses on the cladding and base metal during a transient event; they would Table 17.; Comparison of Tcnstle Properties of As Received TMI 2 E 6 Material with Average Values for Archluc Material at Room (R.T.) Temperature and 600*C.

, TMI 2 E-6 Material Archive Materiala Property -

R. T.

~ 600*C R. T.

600'C Tensile Strength;(MPa) 773_

382 604 295 Yield Strength (MPa) 650 344 451-282 Uniform Elongation (%)

9.0 4.0 13.7-3.7 -

Total Elongation (%)

16.3 39.8 27.8 43.8 Reduction of Area (%)

1.7 73.9 68.9 90.6-aRoom temperature values are averages of 12 tests: 600 C values are averages of_9 tests.-

. ~.. -..

l I-63 iiiiiiiiiiiiiiiiiiii ii 800 TMI 2 E-6 Material, As Received

~

Archive Material, ~

~

As Received 600 (Test 1439)

~

g o.

y 500 g 400 CO 300 200 100 0

0 2

4 6

8 10 12 14 16 18 20 22 24 Strain (%)

Mg. 50.

Tensile Curves at Room Temperaturefor Material Cutfrom nil-2 Lower Head Sample E 6 and As-Received Archive Material.

600 i

i i

i TMI 2 E 6 Material, 500 As Received

$400 3

300 H

e D 200 Archive Material,

[

O As Received 100 (Test 1443)

F 0

0 5

10 15 20 25 30 35 40 Strain (%)

Ag. 51.

Tensile Curves at 600*Cfor Material Cutfrom TMI-2 Lower Head

\\

1 Sample E-6 and As-Received Archive Material

64 reduce the resistance of the Type 308L weld cladding alloy to hot tearing and would cause recrystallization of the base metal, thereby erasing any evidence of deformation.

The juxtaposition oflayers observed on the crack surfaces may provide some clues to the sequence of events that took place during the accident. An Fe based oxide layer is present immediately adjacent to both the ferritic steel base metal and the stainless steel cladding. The depth of this layer, thick on the ferritic steel but much thinner on the stainless steel cladding, reflects the relative resistance of these two alloys to oxidation. Immediately above the oxide layer is a layer of molten control materials (Ag-Cd In Sn). Any small shards of fuel containing material lie above this layer of molten control materials.

These observations suggest that the first material from the core to reach the lower head was molten control-assembly constituents. Rapid oxidation of the ferritic vessel steel alloy occurred in the presence of this material and possibly superheated_ water vapor, The molten fuel that subsequently flowed over the lower head did not penetrate the cracks in significant quantitles, either because it was not sufliciently fluid or because the intervening layer of control-assembly materials prevented it from doing so. The observed clad cracking may have been produced by thermal shock during the initial contact with the molten core materials.

However, it appears 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. This latter process is analogous to hot tearing during welding. Because the cooling rate from the maximum temperature was relatively high (on the order of 1006C/ min), significant thermal stresses would be expected.

No evidence has been found to indicate that 11guld metal embrittlement contributed to the clad cracking, even though the embrittlement of Fe-base alloys by liquid Ag Cd In, and Sn has been reported in the literature.7 The fact that much of the clad cracking was interrm1, i.e. not connected to the outside surface of the clad layer (see Fig. 36), would appear to preclude this mechanism. However, it is apparent that further examination of Specimen E-6, as well as other melmens from-the TMI 2 lower head, will be required to satisfactorily determine (lic sequence of events that occurred during the accident and define the factors responsible for the clad cracking.

l 65 References 1.

R. L. Moore and C L. Tolman, Estimated 7M!-2 Ve.sse! TherTnal Response Based on the lower Meturm Debris Cort /lguration, EGO M 01288, Idaho National Engineering Laboratory (July 1988),

2.

D. R. Dierchs, TMI 2 Vessel Investigation Prq)cct (VIP) Metallurgical Program, Progress Report January September 1989, NUREGlCR 5524, Vol.1. ANL 90/2, Argonne National Laboratory (March 1990),

3.

R. Pelli, P. Nenonen M. Kemppainen, and K. Torronen, Reactor Vessel Stects ASME SA533B and SA508 C1.2, Microstructuralinvestigations. Research Report 219 Valtion Teknillinen Tutkimuskeskus (Technical Research Centre of Finland), Espoo Finland (1983),

4.

M. Amano, V. Sliga, and T. Naiki, Fabrication and Testing of1%ll Size Pressure Vessel Model, pp. 929 950 in Pressure Vessel Technology, Part II, Proc, First Int, Conf, on Pressure Vessel Technology, Delft, Netherlands, (Sept. 29-Oct. 2, 1969).

5.

Data Shects on the Elevated Temperature Properties of 1.3 Mn 0.5 Mo-0.5 Ni Stect Platesfor Dollers and Other Pressure Vessels (SBV 2), NRIM Creep Data Sheet No,18B, National Research Institute for Metals Tokyo (1987),

6, R. Pelli and K. Torronert, Reactor Vessel Steels ASME SA533B and SA508 CI.2, Summary Report of VTT Research, Research Report 340, Valtion Teknillinen Tutkimuskeskus (Technical Research Centre of Finland), Espoo, Finland (1985),

7.

M. G. Nicholas, A Survey ofliterature on Liquid Metal Embrittlement of Metals and Alloys, pp. 27-50 in Embrittlement by Liquid Metals, M, H, Kamdar, ed.,

The Metallurgical Society of AIME, New York (1983).

l 66 Distribution for NUREG/CR-5524 Vol. II (ANI-90/341 Internal 1 D. R. Diercks (50)

TIS File (3)

ANL Patent File ANL Contract File External:

Manager, Chicago Operations Office, DOE ANL Libraries (2)

Materials and Components Technology Didston Review Committee:

P. Alexander, Lord Corporation, Erie, PA M. S. Dresselhaus, Massachusettr Institute of Technology, Cambridge, MA S. Green, Electric Power Research Institute, Palo Alto, CA R. A. Greenkorn, Purdue University. West Lafayette, IN L. J. Jardine, Lawrence Livermore National Laboratory, Livermore, CA C.-Y, Lt. Cornell University, Ithaca, NY R. E, Scholl, Counter Quake Corporation Redwood City, CA P. G. Shewmon Ohio State University. Columbus, OH R. E Smith, EPRI NDE Center, Charlotte, NC D. W. Akers, Idaho National Engineering Laboratory, EG&G Idaho, Inc., Idaho Falls, ID Dr. Banaschik, Gesellschaft for Reaktorsicherheit, Zentralstelle Forschungsbetreuung, Koln 1. Federal Republic of Germany E, Beckjord, OITice of Nuclear Regulatory Research, U.S Nuclear Regulatory Commission, Washington, DC G. A. Berna, Idaho National Engineering Laboratory, EG&G Idaho, Inc., Idaho Falls, ID J. Bros, TECNATOM S.A., Components Integrity Group, Madrid, Spain S. Chakraborty, Swiss Federal Nuclear Safety inspectorate, WQrenlingen, Switzerland N. Cole, MPR Associates, Washington, DC F Corsi. ENEA/ VEL-MEP, Rome, Italy J. Cortez, U.S. Nuclear Regulatory Commission, Washington, DC F. Costanzi, U.S. Nuclear Regulatory Commission Washington, DC P. DeJonghe, Study Centre for Nuclear Energy SCK/CEN, Bruxelles, Belgium

' J. Duco, Department d' Analyse de Soret 6, CEN/FAR, Cedex, France J. M. Figueras, Consejo de Seguridad Nuclear, Subdireccion de Analysis y Evaluacion, Madrid, Spain D. W. Golden. Idaho National Engineering Laboratory, EG&G Idaho, Inc., Idaho Falls, ID.

W. Gomolinski, IPSN/OSSN, CEN/FAR. Cedex, France J. A. Hudson, B388 Harwell Laboratory, UKAEA, Oxfordshire, United IUngdom S. Kawasaki, Department of Fuel Safety Research, Japan Atomic Energy Research Institute, Ibaraki ken, Japan

i 67 S. Kinnersly. Technical Area, Severe Accident Analysis, UKAEA, Dorset, United Kingdom G Korth, Idaho National Engineering Laboratory EG&G Idaho, Inc., Idaho Falls, ID R. Landry, U.S. Nuclear Regulatory Commission, Washington, DC S. Levin, TMI 2, GPU Nuclear, Middletown, PA Mr. C Marlechiolo, ENEA/ DISP, Division of Mechanical Analysis & Technology, Rome, Italy M. Mayfield, Office of Nuclear Regulatory Research, Materials Engineering Branch, U.S. Nuclear Regulatory Commission, Washington, DC R. K. McCardell, Idaho National Engineering Laboratory, EG&G Idaho, Inc., Idaho Falls, ID N, R. Mcdonald, ANSTO, Lucas Heights Research Laboratories, Lucas Heights, Australia P. Milella. ENEA/ DISP, Division of Mechanical Analysis & Technology Rome, Italy R. C, Monroy, Planning Department Nuclear R&D Projects, UNIDAD Electrica, S.A.,

Madrid, Spain L

H. Njo, Swiss Federal Nuclear Safety inspectorate, W0renlingen, Switzerland C. Ottoson, Finnish Centre for Radiation and Nuclear Safety. Helsinki, Finland W. F. Pasedag, U.S. Department of Energy, Office of LWR Safety and Technology, Washington, DC R. Pelli, Technical Research Centre of Finland, Espoo, Finland G. Petrangeli, ENEA/ DISP, Sector for Develcpment and Research, Rome, Italy K. Pettersson, Department of Structural Integrity, Swedish Nuclear Power Inspectorate Stockholm, Sweden G. Saponaro, ENEA DISP, Regulatory Research Commitment, Rome, Italy H. Schulz, Gesellschaft for Reaktorsicherheit, Zentralstelle Forschungsbetreuung, K51n 1, Federal Republic of Germany C. Z. Serpan, Office of Nuclear Regulatory Research, Materials Engineering Branch, U.S. Nuclear Regulatory Commission, Washington, DC L. C. Shao, Division of Engineering, RES, U.S. Nuclear Regulatory Commission, Washington, DC D. Sheron U.S. Nuclear Regulatory Commission, Washington, DC M. Shiba, Japan Atomic Energy Research Institute, Ibaraki ken, Japan P. Soulat. Service de Recherches Metallurgiques Appliquees, CEN Saclay, Cedex, France T, Spels, U.S.' Nuclear Regulatory Commission, Washington, DC K, B. Stadie OECD, Agence pour l'Energie Nucleaire, Paris, France J. Strosnider, Nuclear Safety Division, OECD, / ' nee pour l'$nergie Nucleaire, Paris, France D. Sturm, Staatliche MaterialprQfungsanstalt, Universitat Stuttgart, Stuttgart, Federal Republic of Germany W Vandermeulen, Study Centre for Nuclear Energy. SCK/CEN, Bruxelles, Belgium P. Veron, Equipos Nucleares S.A., Maliano, Cantabria, Spain F. Weehuizen, Swiss Federal Nuclear Safety inspectorate, WQrenlingen, Switzerland 1

NHCFOHMXs U. 8. WCLEAR ftothioRY CoMMGSON WEKET NUMHER (14%

(Auenessymc Adou.sa,nav.

EYE-BIBLIOGRAPHIC DATA SHEET NtiREO 54 1

tsos omrucwu m u reven)

,yg,99,94 2 mLE ANDSUBnRE yo], 2 -

l Ddi-2 Vessel Investigation Project (VIP).

s-car voit Puntmto Metallurgical Program Mmfu l

YEAR November 1990 Progress Report.

I October 1989 June 1990 L1005 s Aufwhiss -

a iet or Huon L D. R. Diercks.

Technical:- Progress -

?. PL Hloo CoVEHLD (mc4mue Deras)

October 1989, June 1990:

s PEnronMINo oRoANEATON-NAMI:AND ADDRESS (# NHC, prove then, ome, or Aspm US NVek Repuleewy Commasm eruf ma&y adfowJmntecer,.

svovw name wuf mahy adAuss J Argonne National Laboratory 9700 South Cass Avenue

Argonne, !L 60439 e $PoNSOHNo oRoAN2 ANON - NAME AND ADDRESS (# NRC, inw 'Same as abovet #cantrocar, provide NHC Dmmm oMce or Mapm U.8, Nudear Hagulakry

, Commmam arwf madry ediane ) -

Division of Engineering -

OfDce of Nuclear Regulatory Research -

U. S. Nuclear Regulatory Commission Washington, DC 20555

10. SUPREMENTAHY NOTES

11. ABST11ACT (200 wortts or less)

During the period from October 1989 through June 1990, a series of heat treatment experiments-

on archive matertal from the lower head of the Midland nuclear reactor was completed, the f resulting microstructures were examined, and hardness values were determined. Round-robin microstructural characterizations and mechanical tests on the archive material were also completed iby the participating Organisation for Economic Cooperation and Development (OECD) partner =

laboratories. Good agreement was generally obtained in these evaluations and testai 'Decontami-D nation of samples from the TMI-2 lower head is underway at ANL, and detailed microstructural and.

scanning electron microscope examinations of Specimen E-0 were carried out.', Metallographic -

. examination revealed that the surface cracks present extended through the stainless steel cladding, i

t but continued for only =3mm into the underiytng ferritic steel base metal,1 Extensive secondary cracking of the cladding, not connected to the surface, was observed.' The eva% ion indicated thai

the base metal in the vicinity of the crack attained a maximum temperature of -1000 to 1100*C.

a during the accident,- The molten fuel apparently did not penetrate into the cracks and interact with the base metal,

12. KEY WoHDS,oESCRIP TOHS (last wordt or phrases Nt war seselrasaarchers n bcalmp res reprt) -

13. AVAILA01ulY 6(AIEMENI; f

Three Mile Island-2 Reactor Unlimited

{

Midland-1. Reactor

14. SECURRY CLASSFCAf 0N -

f Steel-ASTM-A533 s (ram espa;

}

I Metallography.

Unclassified >

Mechanical Properties (rn.nore :

" Unclassified 15.NUMBulOF PAGES :

16 PHCE N@ F OHM 3M2-4M 1 1

I

\\

+

THIS DOCUMENT WAS PRINTED USING RECYCLED PAPER.

n gi

~

4

!a t ;,-

J

..; J UNITE 9 STATES mca rovm eess n:m s

L LCLEAR REGULATERY C :MMISSl:N d"iaQ'5 " :

WASHINL ON, D.C. 20555

f RMit fin C.af

. i 0FFICIAL BUSINESS -1.

.Ptt4ALTY FOR PAIVATE USE, $300 :

l I

M' c5 N'C - 0A D M i20 % 5139531 3 1AN

IV FOIA r. PubllCAI!ONS SVC5 TPS PDR-NUREG c-2?\\

WASHINGTON CC 20555 1

m i -

' l.i.

'.k '

f s

j^,6 j '

IJ-j

.j s,-,

i* )-

I w i j - !. >

-)

.4

.i,

.g*

I h,

,,n..

,.a

i

' ^I t

5

('J sf ?;-1 4

1 1

i 4

8

..I

-[, l -

1 tji s

L 1

-?fI{.

g--

i 3

e d.- j i

1

' ~ "\\

_?

-.~[-~

X,.

P ! _' _ i l'

i n

I 1 4 j

a t

i

. i.-

, :s T

.2,3.

2 i

s (y

t i'

(

,;.]

y

(

)

!], y

^;6 E p

.o p,-6

Q q y. t i,, j p'

'(

s

\\s

~).)),'

^,;

/ < 'I 5

i; ?, '

1 31,

[-d i, (f5 0-.-3 k

I 4

t F

l.

n s

.y l

il 1

n; q

?i i.,

5

ML20028H286

Text

2 Background

Navigation menu

Search