Text

Final Rept on Failure Analysis of Inconel 600 Tubes from Once Through Steam Generator a & B of TMI-1. W/Three Oversize Figures.Aperture Cards Are Available in PDR
	ML20070H537
Person / Time
Site:	Crane
Issue date:	06/30/1982
From:	Ami Agrawal, Berry W, Stiegelmeyer W; Battelle Memorial Institute, COLUMBUS LABORATORIES
To:	;
Shared Package
ML20070H501	List: ML20070H509; ML20070H518; ML20070H537;
References
	NUDOCS 8212270056
	Download: ML20070H537 (132)
	v • d • e

{{#Wiki_filter:L i [ ~ E FINAL REPORT on E l FAILURE ANALYSIS OF INCONEL 600@ [- TUBES FROM OTSG A AND 110F i 1 THREE MILE ISLAND UNIT-1 I [ to E GPU-NUCLEAR June 30, 1982 by [ Arun K. Agrawal, William N. Stiegelmeyer Warren E. Berry [ [ BATTELLE Columbus Laboratories 505 King Avenue [ Columbus, Ohio 43201 8212270056 821027 F PDR ADOCK 05000289 F _. P __ PDR j

TABLE OF CONTENTS I =

1.0 INTRODUCTION

1 2.0 APPROACH. 2 3.0 RESULTS OF NON-DESTRUCTIVE EXAMINATION. 3-1 3.1 Introduction 3-1 3.2 Radiation Level Check.... 3-1 3.3 Visual and Photographic. 3-1 3.3.1 Results 3-1 3.3.2 Significant Observations... 3-14 3.4 X-ray Radiographic Examination 3-14 3.4.1 Results 3-14 3.4.2 Significant Observations. 3-19 3.5 Eddy Current Examination 3-21 3.5.1 Results 3-21 3.5.2 Significant Observations. 3-21 4.0 RESULTS OF OTHER EXAMINATIONS 4-1 4.1 Tub e B 2 5...................... 4-1 I 4.2 Tube B-ll-23 4-18 4.3 Tubes A-23-93, A-88-ll and A-ll2-5 4-25 4.4 Tube A-71-126.. 4-31 4.5 Tube A-146-6 4-43 4.6 Tube A-146-8 4-64 l ) 5.0 RESULTS OF Y-RAY ISOTOPIC ANALYSIS. 5-1 6.0 DISCUSSION. 6-1

7.0 CONCLUSION

S 7-1 8.0 ACKNOWLEDGMENTS 8-1 1

I I LIST OF FIGURES I Page I Figure 1. Appearance of As-Received Tube B-8-25 at 0, 90, 180 and 270 Degree Positions.. 3-3 Figure 2. A Closeup View of the Dryout Marks on Tube B-8-25 at 90 Degree Position. 3-4 Figure 3. Appearance of As-Received Tube B-ll-23 at 0, 90, 180 and 270 Degree Positions. 3-5 Figure 4. Appearance of As-Received Tube A-23-93 at 0, 90, 180 and 270 Degree Positions. 3-6 i Figure 5. A Closeup View of the Dryout Marks on Tube A-23-93 at 90 Degree Position 3-7 Figure 6. Appearance of As-Received Tube A-71-126 1 Segment 1 at 0, 90, 180 and 270 Degree Positions. 3-8 Figure 7. Appearance of As-Received Tube A-71-126 Segment 2 at 0, 90, 180 and 270 Degree Positions. 3-9 l l l Figure 8. Appearance of As-Received Tube A-71-126 Segment 3 at 0, 90, 180 and 270 Degree Positions. . 3-10 Figure.9, Appearance of As-Received Tube A-71-126 Segment 4 at 0, 90, 180 and 270 Degree Positions.. 3-11 Figure 10. Appearance of As-Received Tube A-71-126 Segment 5 at 0, 90, 180 and 270 Degree Positions. . 3-12 Figure 11. Appearance of As-Received Tube A-88-ll at 0, 90, 180 and 270 Degree Positions. . 3-13 i Figure 12. A Closeup View of the Dryout Marks on Tube j A-88-11 at 0 Degree Position. . 3-15 Figure 13. Appearance of As-Received Tube A-112-5 at 0, 90, 180 and 270 Degree Positions. . 3-16 Figure 14. Appearance of As-Received Tube A-146-G at 0, 90, 180 and 270 Degree Positions. . 3-17 Figure 15. Appearance of As-Received Tube A-146-8 at 0, 90, 180 and 270 Degree Positions. . 3-18 Figure 16. Photograph of As-Cut Slices Al through A8 From Tube B-8-25 4-3 Figure 17. Photomicrographs of the Cross-Section of Bent Specimen A2 from Tube B-8-25............. 4-4 1 l L

I I LIST OF FIGURES (Continued) l Page I Figure 18. SEM Photographs of the Fractured Surface of Specimen A6 from Tube B-8-26............. 4-6 Figure 19. TEM Photomicrographs of the Replica of Fracture Surface of Specimen A4 from Tube B-8-25 After Descaling 4-8 Figure 20. SEM Photograph of the Apex of U-Bend of Specimen A7 from Tube B-8-25 4-9 Figure 21. Photograph of Specimen B from Tube B-8-25 Showing Multiple Fracture Surfaces Stacked Together i For ESCA Analysis . 4-14 Figure 22. A Tree-Like Brown Deposit on the ID Surface of Tube B-8-25 . 4-17 Figure 23. Photomicrograph of IGC in Specimen E from Tube B-ll-23 4-20 i Figure 24 Photomicrograph of IGA in Specimen E from Tube B-11-23 . 4-21 Figure 25. Photomicrograph of Specimen A from Tube B-11-23 Showing IGC and Severe IGA on Either Side of the Crack . 4-22 Figure 26. Photomicrograph of IGA on the ID Surface of Specimen A from Tube B-ll-23............ 4-23 Figure 27. Photomicrograph of the Microstructure of Specimen A from Tube B-ll-23............ 4-24 Figure 28. Photograph of the Descaled Specimen F from Tube A-71-126 Showing Heavy Scoring . 4-35 Figure 29. Photograph of ID Surfaces of Two Halves of Tube A-71-126 Between 51.0 and 60.0 Inch Before Descaling; A) 45' B) 135' C) 225' D) 315' Face. 4-36 Figure 30. Photomicrograph of the Transverse Cross Section of Specimen I from Tube A-71-126. . 4-37 i Figure 31. Photomicrograph of the Longitudinal Cross Section i of Specimen J from Tube A-71-126.. i . 4-38 Figure 32. Photographs of ID Surfaces of Two Halves of Tube l A-146-6; A) 315' B) 225' C) 135' D) 45' Face. 4-45 l l Figure 33. Photomicrographs of the Microstructure of Three Different Regions of Specimen B from Tube A-146-6 . 4-47 L I

I I I LIST OF FIGURES (Continued) I Page Figure 34. Microhardness values at 8 Different Sub-I Locations on Specimen B from Tube A-146-6.. 4-48 Figure 35. Photomicrograph of Through Wall IGC in I Specimen C from Tube A-146-6 4-49 Figure 36. SEM Photomicrograph of a Brown Spot and White Deposit on the ID Surface of Specimen D from Tube A-146-6 4-50 E Figure 37. SEM Photograph of an IGA Pit on Descaled Specimen D from Tube A-146-6 4-51 I Figure 38. Photomicrograph of the Cross Section of a Pit in Specimen D from Tube A-146-6........... 4-53 Figure 39. Back Scatter Electron Image and X-Ray Images of l I Elements Ti, S, Cr, Ni and Fe of a Pit in I Specimen G from Tube A-146-6 4-54 i Figure 40. Relative Ion Intensity Versus Sputtered Depth in SIMS. 4-63 g Figure 41. Photographs of ID Surfaces of Two Halves of Tube A-146-8; A) 45' B) 135' C) 225' D) 315' 4-66 Figure 42. Photomicrographs of the Microstructure of Two Different Regions of Specimen B from Tube A-146-8. 4-67 I Figure 43. Microhardness values at 8 Different Sub-Locations on Specimen B from Tube A-146-8........... 4-68 [ Figure 44 SEM Photograph of a Brown Decoration on the ID I Surface of Specimen E2 from Tube A-146-8 4-70 Figure 45. SEM Photograph of a Crusty Deposit in a Brown Spot en the ID Surface of Specimen El from Tube A-146-8 4-71 Figure 46. SEM Photograph of the Fracture Surface of Specimen E2 from Tube A-146-8 4-72 I Figure 47. Photomicrograph of an IGC in Specimen El from Tube A-146-8 4-73 Figure 48. SEM Photograph of a Shallow Pit on the ID Surface of Specimen I from Tube A-146-8....... 4-75 j Figure 49. SEM Photograph of a Descaled Pit on the ID Surface of Specimen I from the Tube A-146-8. 4-76 ll 1l

LIST OF FIGURES (Continued) l Page Figure 50. Photomicrograph of an IGA on Specimen G from Tube A-146-8.......... 4-77 Figure 51. Photomicrograph of an IGC in Longitudinal Cross Section of Specimen A from Tube A-146-8.... 4-78 Figure 52. Photomicrograph of an IGC in Transverse Cross Section of Specimen C from Tube A-146-8.. 4-79 ~ LIST OF TABLES a Page Table 1. Defect Indications in Radiographs of Tubes from I j TMI-l Steam Generators A and B. 3-20 Table 2. Results of EC Examination of Tubes from TMI-l Steam Generators A and B Using Differential and Pencil j Probes.. 3-22 Table 3. Examination Results of Tube B8-25 4-2 Table 4. EDAX Analysis of Fracture Surface of Specimen A6 from Tube B-8-25 4-7 Table 5. AES Analysis of Fracture Surface of Specimen A3 frcm Tube B-8-25 4-11 1 Table 6. ESCA Analysis of Fracture Surface of Specimen j A3 from Tube B-8-25 4-12 Table 7.

ESCA Analysis of Fracture Surface of Specimen B from Tube B-8-25.................. 4-15 Table 8.

ESCA Binding Energ'.es and States of Elements on Specimen B from Tube B-8-25 4-16 ( Table 9. Examination Results of Tube B-11-23 4-19 Table 10. Wall Thickness of Tube B-ll-23....... 4-26 Table 11. Examination Results of Tube A-23-93 4-27 Table 12. Examination Results of Tube A-88-ll 4-28 Table 13. Examination Results of Tube A-ll2-5 4-29 Table 14. Wall Thickness of Tube A-23-93. 4-30 Li 1 l l

I I LIST OF TABLES (Continued) Page i Table 15. Examination Results of Tube A-71-126 . 4-32 Table 16. Composition of Tube A-71-126 Alloy . 4-40 Table 17. ESCA Analysis of ID Surface of Specimen C from Tube A-71-126 . 4-41 Table 18. ESCA Binding Energies and States of Elements on Specimen C from Tube A-71-126 . 4-42. Table 19. Examination Results of Tube A-146-6......... 4-44 Table 20. ESCA Analysis of ID Surface of Specimen F2 i from Tube A-146-6............ . 4-55 Table 21. ESCA Analysis of ID Surface of Specimen Fil from Tube A-146-6.................. 4-56 Table 22. ESCA Analysis of ID Surface of Specimen E from = Tube A-146-6 4-57 Table 23. ESCA Binding Energies and States of Elements on Specimen F2 fron Tube A-146-6.......... . 4-59 Table 24. ESCA Binding Energies and States of Elements on Specimen Fil from Tube A-146-6 . 4-60 Table 25. ESCA Binding Energies and States of Elements on Specimen E from Tube A-146-6 . 4-61 Table 26. Examination Results of Tube A-146-8.... . 4-65 Table 27. Ga=ma Ray Isotopic Analysis Results. 5-2 I I I I I E

FINAL REPORT on FAILURE ANALYSIS OF INCONEL 600* TUBES FROM OTSG A AND B 0F THREE MILE ISLAND UNIT-1 to GPU-NUCLEAR from BATTELLE Columbus Laboratories June 30, 1982 1 by Arun K. Agrawal, William N. Stiegelmeyer Warren E. Berry INTRODUCTION g-Three Mile Island-Unit 1 power plant was brought to hot B functional status between August and September of 1981 af ter a long cold shutdown for about two and a half years. The plant was brought back to ( cold shutdown status and was then hydrotested in November of 1981. On I November 21, 1981 small leaks from primary side to secondary side were detected in tubes of the once-through-steam-generator (OTSG). Subsequently, leaking tubes were identified by bubble test and eddy I current examination. Eddy current examination indicated defects also in some other tubes. As a result, a few tubes with defect indications were removed, along with known leakers, from OTSG-B for deter =ining the nature of the defect (s). Two tubes from OTSG-B, identified as B-8-25 and B-11-23 were received at BCL on December 28, 1981 for failure analysis. Tube B-11-23 i l I ll t

2 was a known leaker; a quick metallographic examination of the defect in this tube established that the failure was due to intergranular stress-corrosion cracking (IGSCC), and the ICSCC had initiated on the inside surface of the tube. While detailed examinations of the two tubes were proceeding, two more shipments of tubes pulled from the OTSG-A were received at BCL for similar examinations. The first shipment trom GPU-Nuclear was received at BCL on January 21, 1981; it contained four tubes with identifications A-71-126, A-88-ll, A-112-5 and A-146-8. The second shipment contained two tubes with identifications A-23-93 and A-146-6; this shipment was received on January 27, 1982. All the tubes shipped to BCL were from the upper tubesheet (UTS) region of OTSGs A and B. Tubes B-8-25, B-11-23, A-23-93, A-88-11, A-112-5, A-146-6 and A-146-8 were "short-pulls", i.e., these were approximately 12 to 12.5 inch long segments cut from the upper tubesheet l crevice region. l Only one tube, A-71-126, was a "long-pull", i.e., it included sections of the tube from beyond the UTS crevice. This tube was rer.ived in five segments, labled A-71-126(1) for the top section and then sequentially up to A-71-126(5) for the bottom-most section. The total combined length of all the five segments was 68 inches. This report contains the results of various examinations conducted on the above tubes at Battelle. Probable cause of the attack also has been identified l l APPROACH I The detailed failure analysis program for TMI-Unit-1 tubes was I developed in close consultation with GPU-Nuclear personnel. There were two main objectives in this program: 1) to identify and characterize the nature of defect (s) in various tubes, and 2) to determine the probable cause(s) of attack which produced the defect (s). A third objective of the program was to identify, based upon the plant history data provided by GPU-N, the environmental condition (s) which may have been responsible for the attack. I

1 I 3 i The broad base failure analysis program consisted of the following examinations and tests: Nondestructive examination (NDE) Metallography and Microstructural examinations Microanalytical surface examination and Physical tests. The NDE included visual and photographic inspection, X-ray I radiography and eddy currect (EC) inspection. The EC inspection used two different types of probes: a standard probe (i.e., differential probe) and an absolute probe (i.e., pencil probe). The pencil probe was used to specifically inspect the roll transition region, i.e., the arca g between the rolled section and the unrolled section of a tube. 5 Radiography was used to locate defects that produced suf ficient l discontinuity in the tube wall for a relatively easy penetration of X-rays. For example, the technique was v.ry helpful in locating the through-wall, but not easily visible cr ek in tube B-11-23. There were a few tubes which 12d obvious " lip-cracks", i.e., i a broken off wall within 0.25 inch of t.se top end. These " lip-cracks" were photographed for documentation purposes. Metallographic and microstructural examination involved examining longitudinal and transverse cross section of specimens } removed from different locations of various tubes. Specimen: vere examined in the as-polished condition and some after etching with nital or phosphoric acid. Some specimens were examined in the scanning electron f microscope (SEM). One fracture surface was examined with the transmis-3 sion electronmicroscope (TEM). In order to determine the nature of corrodent responsible for 1 the attack, surface compo:litions of several tube specimens were investigated. Both the inside diameter (ID) surface and the attacked area, e.g., fracture surface were analyzed. Techniques used in microanalysis of surfaces were energy dispersive X-ray analysis (EDAX), Auger electron spectroscopy (AES), l electron spectroscopy for che=ical analysis (ESCA), secondary ion mass I l L

4 spectroscopy (SIMS) and ilectron microprobe analysis (EPMA) for X-ray images of elements. X-ray differaction (XRD) also was used for corrosion product identification. Bulk composition of tube material also was determined. X-ray fluorescence and other standard methods were used for tlas purpose. Physical tests of tubes or tube material included,1) tension testr 2) wall thickness measurement and 3) microhardness measurement using Knoop diamond pyramid tester. The susceptibility of a few tubes to intergranular attack (IGA) by polythionic acid was investigated using the electrochemical potentio-kinetic reactivation technique (EPR). In support of the failure analysis program, five wipe samples from OTSGs A and B were also supplied by GPU-Nuclear. These were analyzed using y-ray isotopic analysis for the determination of active isotopes present in OTSGs. 1 G

[ 3.0 RESLTLTS OF NONDESTRUCTIVE EXAMINATION l 3.1 Introduction j Tube segments from GPU-Nuclear were received packaged in 40 gallon drums. After opening the packages, each tube segment was subjected to four different nondestructive examinations (NDEs). The NDEs performed h were: (1) Radiation level check (2) Visual and photographic (3) X-ray radiographic (4) Eddy current inspection j 3.2 Radiation Level Check 1 The radiation level of each tube segment was checked to establish safe working conditions. The radiation level at contact for different tubes ranged between 20 mR/hr and 40 mR/hr. The above radiation levels ( were considered low, nonetheless, appropriate safety precautions were still required when working with these tubes, in order to protect both personnel and work facilities from radioactive contamination. s 3.3 Visual and Photographic s 3.3.1 Results. Initial visual inspection of tubes (segments) showed no obvious defects in the form of a crack or a pit on the OD surface, except for the dryout water marks and some scratch marks; the latter prob-ably were from the tube cutting and pulling operations. Tube ends in a few cases, however, were more ragged than in others, this again was con-sidered to be a result of tube removal operations. Tube defects in the form of lip-cracks were observed at the upper (top) end of tubes A-88-ll, A-ll2-5 and A-146-8. The lip-crack in each case was associated with the 0* orientation slot cut in the tube.

2 Following the initial visual inspection, each tube was photo-graphed in four different positions, i.e., 0*, 90*, 180* and 270*. An inch scale also was photograghed along side of each tube to show its length. Photographs of tube B-8-25 are shown in Figure 1; water marks are visible along the whole length of the tube and in all four positions, i.e., 0*, 90*, 180* and 270*. A closeup view of the dryout mark at 90* { is shown in Figure 2. Photographs of tube B-ll-23 are shown in Figure 3; water marks are visible along the whole length of the tube and in all four positions. The 270* quadrant, however, was relatively cleaner than the other three quadrants. k Photographs of tube A-23-93 are shown in Figure 4 with water marks visible along the whole length of the tube and in all four positions. { A closeup view of the top end at 90* position is shown in Figure 5. Some vertical scratch marks are clearly visible in the rolled section (0-1.0 in.) of the tube. The end of the rolled section is indicated by the circum-ferential white ring at 1.0 inch. Tube A-71-126 was a long-pull and it was received at BCL in five segments. These segments were marked 1 to 5, in the order in which they were removed from the OTSC, i.e., 1 for the top segment and 5 for the bottom most. Photographs of segments one through five are shown in Figures 6 through 10, respectively. Water marks on tube A-71-126 were less extensive and dense, particularly those on lower segments, than those observed on the previous three tubes. However, all four quadrants of the five segments of tube A-71-126 showed some water spots. No particular quadrant was free of water spots. Figure 7 for tube A-71-126 segme.nt 2 shows circumferential shiny ( ring pattern over most of the tube surface, particularly in the first 7 inches of the segment. These rings are obviously from the mechanical operation of tube removal, but such an extensive marking was not observed on any other tubes. Photographs of tube A-88-ll are shown in Figure 11, water marks on the 0* quadrant are relatively heavier than on the other three quadrants.

summa WillW M W W W W W W W l 1 l l u 't

180, a.

t -. s.. ..g, n., -. - ~ ~. ,]=3, j8 9 lu r i a 0 ti + Jj ' j ggj [ ; gy,! ' i 2 3j t } ,j; j':g ! lg ,7 i { { FIGURE 1. APPEARANCE OF AS-RECEIVED TUBE B-8-25 AT 0, 90, 180 AND 270 DEGREE POSITIONS

,A 4** 8 5 B B B B m i l I MPHPJ'

!M"#3'"

l

WVW umW BiEB emaw unu Wuns W W M L' - -J W l 90' Ji ' i -di i 5; i ' 6i ' 71 I l !8I I '9 i 10 i ilt i w l 1 5! 6i' 47ll 8!! '9 ! '.10_'..ii '.l. l! ! '.' 11 a) it T ~ ' 3! 4; -i 270* .l~' ~ ..., ~... FIGURE 3. APPEARANCE OF AS-RECEIVED TUBE B-11-23 AT 0, 90,180 AND 270 DEGREE POSITIONS ,.l3, i ;';'.,;( _ l _,; " ' )' . l ' ' + (' : .' ; ; ;4 ; -, ].;s 7...... ,. y . j: .g

; - y...

.n 4

WB EEN immer ums W MB M M unas M M T M M M M M o* I i; Il-l5i ' ] 4 ait I I ll l I uii.j i a; ; j ppg llaj:p+f '9ji' s t ~$", l 1 i 2 3.,,,. 1-i., I 7 ' 8. ' ti 10' 2 11! It 13 ru s.: se ~ 90* 'o l / . m.

1,

.,,,o i,i %, : l.11, 7 l 1 2 1 1 '5' 7 I 8 ' 9r 10 .11' 11 t :i 6 \\ . ~ l'80* ...,J.a:. +i -;

3 i

1 5 6 7; ' '8 9:.._. l o 11 tJ raMn m .,e 270* j)

)

'w1 1 1 5 6 7 8 9. 10 11' ' 12 13 FIGURE 4. APPEARANCE OF AS-RECElVED TUBE A-23-93 AT 0, 90, 180 AND 270 DECREE POSITIONS 1

7 l ? 4 p. i FIGURE 5. A CLOSEUP VIEW OF THE DRYOUT MARKS ON c TUBE A-23-93 AT 90 DEGREE POSITION r h ~ ) ) w_-....

e -v vm VW WWF 1552 M IEEE W W s asp.< - He .m.- .: h I 5'. 6 ' 7; ' ll 'l lit 11 i Wit 90* misa4' ,f* ~ ~""' ; ' fA f I 1 i 6 7: 11 ') ' itt ll I 180* 1 4 "** ' ' ' ' ~ 5

** ^ T

~ ,1 j n.

1. :

.a. I i 3 1 t 'l'! 3; t 6 7 f]~ 9 ),; si i; 1 ev4 270* W'l . w;,.: ~ a v -e 'i I i 6 7. 11 9 til i1 i; ,3 FIGURE 6. APPEARANCE OF AS-RECEIVED TUBE A-71-126 SECHENT 1 0, 90, 180 AND 270 DEGREE POSITIONS l l

r-m v.c - =aur = - - m unas y i., l 1, - i '3, -l'6' ' 7i ' ' ' ' ll ' I' 9,' 10 11: 12 !.i l ,o,1 1, on F, - ,.i.i 1.. -.. 3..- I 6-7,, ;pi g: i i gj i > ia'.-..jj;, 1 g j. ,3 i e Iso- ,{Lo y ; N ci. 4 1 1 . ' ' 5 6 7, 11 4 9!. 10 11-12. t l f ' _ ;' 'l l * ~ A 94*- i..J:i .. --- --. 7; ' ll, '! >i.; 7 1 5-s. 1 s. 6' 9! -..li' 1) 10 FIGURE 7. APPEARANCE OF AS-RECEIVED TUBE A-71-126 SEGMENT 2 AT 0, 90,180 AND 270 DEGREE POSITIONS ,- ' { ');., ' f, ' a. - l l 'l') f.i -l '.?: T,...

L [_l

?{,=g.l y 5 '* lf].'Jjf.i ': (l:,f l{ > ~

mmum -i 3 m ZO m hm C/) ~ 3^ ~ ;. %O W i ta2 a: 4 ~ Q W ~ Q ~ ~ Om N I O i O Z ~ ~ ~ i C co t i -_. ' g ', ~j c ~ ?' g p l e a cl H ~ i i l l _l ~ m Z ~ taJ tw N 7 N U = N y 2 M ^ N H 0 4 C: w w. b J j i <c rg i s-c::2 'O. O gp_ 1 r* be O Es2 m Es2 ~ U i c: s C/1 <C t e i C la2 Y y Z g c.

C N

C:3 cz: D U N e o .e e .o W N N

j W W W W W W W W W W W W W W W W W W W 4 k l 4 1 P'l ; ! !.'i. - :r' i, j 4 [ 2 3< t 31

6;

'7' ' '8 l9 'lO ' if.,.._!'12l . ',' l j. i ] i i '1 n: /!. i_1 - ~ _. ~ 13.. i 1 i w i 'i -) H 180' 1

iP 1

.h ll m' I :3. '6' Ti ', A :'... i 1 9f

, ~ 10

]t! '._~12' j i I 3 i '8 's 13 i s l l 1 i 270* i i '!lt:, 8 9i ' la 11:. l.:

j

< ll: El* ~ 2 1. ,'5 'r6 7 3 i l FIGURE 9. APPEARANCE OF AS-RECEIVED TUBE A-71-126 SEGMENT 4 AT 0, 90, 180 AND 270 DECREE POSITIONS l l

l1l1 0 7 2 E DN 1 A 3 1 1 4 1 0 8 1 E 1 3 3 l 3 0 i. 9 1 t i2 0 E i 1 ' '2 T A L I

ti j

5 ll !{ T l O i N 3l E + M G 0 E 0 1 1 D S E . 9 .I 6 2 [ - 1 y9 [ .,1 o9 1 j LI m 7 S ,8

8

'I s, m A 1. g E -j b p p- .8 B l r. Q L' U q T E _, a .,;7 s 7 v 2 u. J i 1 2 D E n .:7 V p n I s o. u s o E u. C v. . 6 _. o. E . '6 R i. 6 i

9. S n

. N O

5 rI S i ES r ,5 OT i I CO NP I. A ~ 1 .I RE S AE ER PG PE 2 F AD 0 1 E R U G IF t g i1 ,5 p. -J 'l 0 0 0 g 9 8 7 I 2 l llll1 t 1

i - e e me e m mmmmmW W W WmM hW^ t t ';! 83-12 -o ~ ...n._,._ .. r j- !"l.i-il m i 90* , ;,;l ;.;].f gh. Ho:;j ' 7i ' g,l p;!.jui;aj.,g U + -m ,3 i ,1 L 1 j l5: 'B' 'l 9;..... ' l 0..... _ _1 1 ; ' -._1r i 6' t l I , 'i l "1. ,,r g 180* I i i j 'il 11 'i' I '~ 270* l l l l FIGURE 11. APPEARANCE OF AS-RECEIVED TUBE A-88-ll AT 0. 90, 180 AND 270 DECREE POSITIONS .i 4

. I I 14 A closeup view of the dryout deposits in 0* quadrant is shown in Figure 12. The lip-crack, described earlier, can be seen at the slotted end in the 0* photograph. Photographs of tubes A-112-5, A-146-6 and A-146-8 are shown in Figures 13, 14 and 15 respectively. Dryout water marks are present along the whole length and in all four quadrants of these tubes. Some vertical .g scratch marks in the rolled section of tubes and circumferential water 5 ring at the end of rolled section are also visible in these photographs. Some scratch marks at other places along the tube length also can be seen in Figures 13 and 14. In Figure 13 a lip-crack can be seen extending from the slot in tube A-112-5. A similar lip-crack is visible in tube A-146-8 in Figure 15. 'g 3.3.2 Significant Observations. Visual and photographic 'W examinations of the OD surface of tubes are summarized below: (1) Dryout water marks were present practically on all tubes. These marks were present practically on all four quadrants of each tube. = (2) Vertical scratches were observed on all tubes at various locations, but the scratches were particularly - I heavy in the rolled section of each tube. (3) Some shiny circumferential ring patterns were observed on some tubes; the ring patterns were numerous on ^ segment 2 of tube A-71-126 and are probably the result of the tube pulling operation. (4) Lip-cracks were observed in tubes A-88-ll, A-ll2-5 and A-146-8; the cracks were associated with O' orientation I slot in each case. 3.4 X-rav Radiographic Examination 3.4.1 Results. Each tube segment was radiographed individ-ually in four different positions, at approximately 0*, 90*, 180* and I

ni !!! 5 o' g, WMI _. ~. -... ~. - 1 9g, r.u U > '- m. i. s (b .i 339 ,r 1 4 ( .1 i ;'!q q p;"p' q.sp m nj. qq.;1::p q g is u 5i ' I I 3 '6: _.7 :.- '8 '9

10....' _'..l l..i,...

e ...b i ( l i l 270* c f{ l f ,a I'. >, f'[ I. 3 1.._-.'5 i,-_6.-_ _ 7.4 '8 9; 1._0__ _, __1 1. .._1.t I I l l FIGURE 13. APPEARANCE OF AS-RECEIVED TUBE A-112-5 AT 0, 90,180 AND 270 DECREE POSITIONS

v s s c - usum-l'Il !: t, t 4 um.. 0*

j + i 'in j 5l'ppvidi,h.;,ir,r' ' ' 7_.

' ' i!;qlI+,;jyl!ip.r'l ' lo , m.T ;j.g ~... s'. 8!!' 9; ; '.10 ' ~ 11'. '.l_12[ ' 1 2 3! t' l I .' '6! -) i !11+ ei-O! u... 90* f .l ,,f!4I L t 5: 6' ' 7, 8' 9! ' 10 ' lli,, 4 ^ 1 1 . it I. 1 , !.+, i. : '. ' ' h,

a. * - a 180P g"

! ~] $ .4 elli ib i,!i. s

7l '.> j.ilyi ;igt;.i,iljyli.;![.{ ;, m

'j = g, ' ' f i gj.

p,n j. l ; -

g, i 2 3 __'..-.j : { }g _.___3._.- 8,p ' ' . )gj ' .~r 270* i 34 1 Si ' '6 '7i ' ' I 8 I ' 9 !'10 J P - '..~ 11. -. 1 - FIGURE 14. APPEARANCE OF AS-RECEIVED TUBE A-146-G AT 0, 90, 180 AND 270 DEGREE POSITIONS

ll g ,si ii is. i - i .~gn-mmr a o n 's r-' T I "'1 I H \\ I H " i l 4- - -- f I I a '{ .t t )Mh 8. .' ' ' - ~. .." k - l ~ -r j = w m 7 i. .....g t -m ~~ H 3~' ~"* i . ; 2 }< m y r, - I l l -~~1 o 4 g w ~: W _t ,ZI Zf Of O 8 .-. j g =. a T._ ' '.-1 o n - g Tf.7-N n O ~-(_, g I ~ 1-- A '} ' C

~

c3 - CC . =. m ~ M . 6.C3 3 C3 =y l '~:~ 7 O w-+ '.I ~ o e

- %M 1

=- = Q? D ^L ~-T' u

- ~_

g ~.. a IC ~T .i.e. t .-c ' =. - 4 + ..z.. . -- C ~ 'j ,a ..r ~ .T [ [ .h - = - - - M Z T .F r ~ ~. ~* h~ E C-gg E jig 5 ~6 ~ ih 1 ~ =-- a M .~ _ ' ' ~ ~ ~ r= Z., N.- U y ___2-T

Q,

ng W

CI: f i 6 - -l I tr) ~7l ~ ._ ~ l y ~ O

, {

k{> m.

m e

_ l M O

.c I

l 2 i l es

' s M

t e~ s _ m 4 O N W CE: D C w N s 0 e I s s n ~

B 19 270*. Then each radiograph was examined over a light viewer to observe changes in light density over the radiograph's surface. In several radio-graphs, dark hairlines were observed against the light general background. These dark hairlines were considered as indicative of defects (possibly narrow hairline cracks) in the corresponding area on the tube. The hairlines in practically all cases ran in transverse direction and covered anywhere from 1/10th to 3/4th of the tube diameter. A defect shown by a hairline was classified as a " clear" indication if the hairline l was sufficiently dark, otherwise as a " faint" indication. The location of each indication was recorded. All the indications observed on various tubes are listed in Table 1. No vertical (longitudinal) hairline indications were observed I corresponding to the OD surface scratches described earlier in Sub-t section 3.2. This implies that the scratches were superficial, only surface deposit deep, and did not scar the metal underneath, otherwise indications would have been obtained. 3.4.2 Significant Observations. Results of radiographic examination of various tubes are summarized below: (1) Defect indications for lip-cracks were obtained in tubes A-23-93, A-88-11, A-112-5 and A-146-8. The indications were faint in tube A-88-ll, but clear in the other three. (2) Defect indications in the vicinity of roll trarsition area (i.e., location 0.75 to 1.25 inch) were ob tained in tubes A-88-ll, A-112-5, A-146-6, A-146-8 and B-11-23. The indication was faint in tube A-146-8 but clear in the other four tubes. (3) There were several clear defect indications in all tube segments examined; including segment 5 of i l tube A-71-126 which was removed from below the upper tube sheet of OTSG-A. ll 1l

l M M TAllLE 1. DEFECT INDICATIONS IN RADIOGRAPitS OF TULLES FROM TMI-l STEAM CENERATORS A AND 11 Tube Number Defect Location From the Top of the Tube, Inch A-23-93 I.ip Crack 2.75-3.25 12.0 12.25 Faint Faint Clear A-71-126/1 1.75 12.25 Clear Clear A-71-126/2 1.0 1.75 6.5 8.5 9.75 i i Faint Faint Clear Faint Clear A-71-126/3 4.0 7.5 9.0 9.5 9.75 10.0 12.25 Clear Clear Clear Clear Clear Clear Clear A-71-126/4 7.25 8.5 9.5 10.0 10.75 11.25 12.25 Clear Faint Clear Clear Clear Clear Clear A-71-126/5 Numerous indications faint to clear along the whole length, 14.5 clear. U A-88-Il 0.25 1.5 2.75 3.5 7.0 9.75 11.75 Faint Clear Clear Clear V Shape Faint Clear A-ll2-5 0.25 1.25 2.0 6.0 11.5 I.ip Crack Clear Clear Clear Clear A-146-6 0.75 1.0 3.25-4.0 4.25 4.75 5.25 6.0 6.5 Clear Clear Clear Clear Clear Faint Clear Clear A-146-8 I.ip Crack 0.25 i 0 4.0 4.25 6.75 8.5 1 Clear Faint Fe!.t Faint Faint Faint B-8-25 10.25 11.5 Faint Paint 11-11-23 1.0 1.25 1.5 2.5 Faint Clear Clear Clear .__.....-..__-----...__.__-...__m_____...-.___ _ _ -._.. _ _ __. __.__ _. _ __..,._ _. m _ _ _ _.. _

p 21 3.5 Eddy Current Examination 3.5.1 Results. Two types of eddy current (EC) probes were ( used in the examination of each tube segment. These were the differential or standard probe (Model No. A520LC) and the pencil probe (also known as the absolute probe). Both probes were obtained from ZETEC Corporation cf Washington. An Eddy Scope Model EM 3300 from Automation Industries was used for displaying the signals, which were then recorded with a Polaroid camera. Test frequencies with the differential and pencil probes were 400 kHz and 350 kHz, respectively. The gain in the Eddy Scope was f set such that a 100 percent through-wall defect in the ASME Section 11 standard produced a signal which covered one-third the width of the scope The differential probe was calibrated using an ASME Section 11 screen. standard. No calibration was done for the pencil probe. Results of the EC examination of various tubes are listed in Table 2. The location of each defect indication is given with respect to ( the top of the tube examined. The circumferential position of a defect is indicated by degrees, following the standard practice (W-axis as 0*) recomended by GPU-Nuclear. 3.5.2 Significant Observations. The results of EC examination are summarized below: (1) Defect indications in the roll transition region j were obtained in tubes B-8-25, A-88-ll, A-112-5 and A-146-6. (?) Defect indications were obtained in all tube segments, including those from the long-pull from below the upper tube sheet crevice region. Location of all indications from various tubes are listed in 1 Table 2. (3) Some defects indicated by the differential probe were { j not confirmed by the" pencil proba, and vice versa. I \\ However, in most cases indication of one probe was confirmed by the other. ( I

22 TABLE 2. RESULTS OF EC EXAMINATION OF TUBES FROM TMI-1 STEAM GENERATORS A AND B USING DIFFERENTIAL AND PENCIL PROBES Defect Location

DP PP Tube Number in.

in, degree Comment Note B-8-25 1.25, 90-180 Large Signal Bracing roll transition 1.50, 320 Large Signal 2.75, 180 Large Signal 3.0 490% Wall 3.5 N90% Wall 4.0 Surface Defect 4.5 Small Signal 4.5, 180 Large Signal 5.0 Small Signal 4.75, 110 Large Signal l 5.0, 90 Large Signal' l l l 5.25, 110 Large Signal l 5.75 Surface Defect 6.0 Surface Defect 6.75, 90 Large Signal I 7.75, 60 Large Signal 9.25, 340 Small Signal 9.25 Surface Defect 9.75 i Surface Defect I L E

B I 23 TABLE 2. (Continued) I Defect Location DP PP Tube Number in. in, degree Comment Note B-11-23 0.25, 25 Large Signal Tube Stub (2.0 in) I was examined afcer 0.75, 180 Large Signal cross sectioning the tube. No I 1.0, 300 Large Signal indication in roll transition region. 10.25 Small Signal Remainder of the I tube (long piece) after removing the thru-wall crack. A-23-93 1.75 Small Signal 1.75, 20-140 Large Signal 2.5 s80% Wall 2.5, 20-140 Large Signal A-71-126 2.75, NO Small Signal 3.0 small Signal A-71-126 14.25 Possibly 0.D. Ends flared to E allow probes l 24.25 Possibly 0.D. j j A-71-126 33.25 Small Signal Ends flared to allow probes 33.25 Small Signal 36.75 Possibly 0.D. A-71-126 49.25 Possibly 0.D. Ends flared to I allow probes A-71-126 52.75 <20% Wall 54.25, s0 Small Signal l t I 1 I

2 '. TABLE 2. (Continued) I Defect Location I DP PP Tube Number in. in, degrees Comment Note A-88-11 1.25, 270 Large Signal Bracing roll transition 5.0, 90 Large Signal 5.0 s100% Wall A-112-5 1.25, 270 small Signal Bracing roll 3.25, 270 Large Signal 3.5 90% Wall 5.25, 270 Large Signal 6.25 90% Wall 6.25 90% Wall 6.50, 270 Large Signal 8.0 80% Wall 8.0, 270 Large Signal A-146-6 1.25, 5-260 Large Signal Bracing roll I transition 4.0 Possibly 0.D. 8.25 %70% Wall 8.25, 5-260 Large Signal 10.25 s70% Wall 10.25, 180-290 Large Signal l 1 1 1 1

25 TABLE 2. (Continued) Defect Location DP PP Tube Nu:nber in. in, degrees Coument Note A-146-8 2.0 Mediums Signal 2.0, 0-270 Medium Signal 3.75 490% Wall 3.5, O Large Signal 6.0 490% Wall 6.0, 350 Large Signal I

Defect Locacica refers to the top end of each tube as 0.0 inch and degrees are referenced with respect to W-axis as 0*.

l 1 I I I I I I e

I I 4.0 RESULTS OF OTHER EXAMINATIONS 4.1 Tube B-8-25 g Short-pull tube B-8-25 was examined using the following W techniques: (1) Visual examination of U-bend slices (2) Visual examination of ID surface I (3) Metallographic examination (4) SEM (Scanning electron microscope) and EDAX (energy dispersive X-ray analysis) examinations of fracture I face (5) AES and ESCA examinations of fracture surface film. Examination results for tube B-8-25 are summarized in Table 3. Also given in Table 3 are: I a) the identification number of each specimen removed from the tube for examination; b) location of each specimen with respect to the top end and O' axis of the tube; c) type of any defect indication from NDEs, and d) type of examination performed on each specimen. Details of the examination results are given below. Top end (specimen A) of tube B-8-25 containing the roll-transition area was slit into eight longitudinal slices. This section had an EC indication at 1.25 inch location. Photographs of these slices (specimens Al-A8) are shown in Figure 16. Specimen A4 broke into two pieces during slitting, and specimen A3 also showed a through-wall crack in handling. Specimens A2, A5 and A6 were bent into U-shape with ID in f tension, all three specimens showed cracks with s90 percent wall penetra-tion. The cracks in specimens A2 through A6 were located at 1.25 inch and spanned from 45' to 270*. Photomicrographs of the cross section of bent specimen A2 is shown in Figure 17. A crack with 90 percent wall penetration is clearly 1 I visible in Figure 17. The crack origin is at the ID surface of the tube. I ( All locations given herein and after use top of the as-received tube I as 0.0 inch reference, with inch fractions converted to nearest 0.25 inch.

7__ y-_, a ty i TABLE 1 EXAMINATION,RESULTS OF TUBE B8-25 Specimen location Number Inches Degrees Type of Indication Examination Result / Comment A (Al-A8) 0-2.25 EC 1.25 Slit and Bend No Crack at 0025 AI 0 2.25 0-45 OD Tension, Visual No Crack A2 0-2.25 45-90 ID Tension, Met. 125 IGC,90% Wall y Structure A3 0-2.25 90 135 SEM/EDAX 1.25 IGC,100% Wall;S 1.9% AES/ESCA Fission Products and S (0.4%), Be (72%), Ag (03%) Sulfide A4 0 !.25 135-180 SEM (top) 1.25 ICC,100% Wall n TEM (bottom) No Striations A5 0-2.25 180-225 ID Tension, Visual 1.25 IGC,90% Wall A6 0-2.25 225-270 ID Tension, 125 IGC,90% Wau,S 53% Near ID SEM/EDAX A7 0-2.25 270-315 OD Tension,SEM No Crack A8 0-2.25 315-360 OD Tension, Visual No Crack B 2.25-3.25 0-180 EC 3.0, Visual Crack AES/ESCA Iligh C (>50 afo), S" (~l a/o), C1(<1 a/o), B (~3 a/o) No SOJ C1 3.25-12.0 0-180 ECSeveral Places ID Visual Tree Decoration C2 2.25-12.0 180-360 EC Several Places ID Visual Tree Decoration, Crack Appearance at 3.0,3.5,4.5 1

[ ~ ( [ [ ,, - = x. ..... ~. p.... *' ~ ' ~ ' hu..%,u d. ' l ' h N' N.$. ( --, g O g 4 7.,. q. ~ .. m_4 g c.',, ..j._ _ d4 Y' )7N* f;r ' -b fin.%..regnap.-jr............ J M

.x. - h,

-4 4.,.ad....w .l ~ A - - - ~- ~ ' I '.~..'W' _m.- . 3 C y y[... } .L,$.e w/ 4,. .i

..,.a..

._s . - '. Yt V d ,e.s.*. g ( ._ / - ( -. n. ~ m. - - - [ = : y. e v.: o

y..e,

hY, j h -l: 3 [ 4 tn W U [ M ~ b. O [ .-6,-~.. I m .vfi. :.,~,. g ' .{ j.: r n N.,, W,+,. ,v. yl.[' :: "... W 6,.D*?,.R:f..f 'u ^ / 3. t- ..:..' ! ' ',. " '.l;((.[r. y a.. f_p* : k 9 +. y;f1, 5 c.g, y,. ' d ' ' ' O gsv..,..., : l}".h.L

r,%.

y ..-y

f...[n.,fgh4.'

. C' 4 ~.:*L' VQx-;',,f '%,; ! Y ,y

(( 'h?.~

.? g

m G%-(,;y),$.

?ktfN. 8 ? A p(.u t y,.wiym..y& g y d w $. y.*<,. n aw: w.m%.,,...t .av ,;q W.m. i .-; n... -. o&c. s -- W %.g y. 2 e cc a . ' l w y:,p L '.y r A, QAq ;,.. ..g 9. a m ,g e. %. g. ,. f 1 4 g ' b.. _ g*,.}_ ?gss.;. n ^

- ~~

,. w .h,. &,r.

y.

w V y1. , % L. --f en m k _j, -Q;j p:y S. n -----,_T.. . < pp s

/

$,f.. i {g%q. 4<tq.{%g.,;-.ggG S l;d. g-. l '^ .,.. g..ev., w.

'" g ;

c. <

M M T" 4p g.a}* gra Oh y q

i B 4 5 if _ '~ OD I 6 . r.' I e -Q ~ 4 l j l I + . 7 c 4. T I [ I.., b.,.. h l f B ~ ~ ...m ID l 6ox (a) MAIN CRACK B w 4 OD l ^- l - l _'y ,.. - y. y_ ; . ~ ~ e y, ;., y,, M,' r

- - - )S.

N - r 1W^~p;. ,*p( $ - .. y Y^~ .g ii. * > 4,,_-v ';- Q, B s. s, ';-%-) ; ():-

4

-i ?'~ l t W [- 's., _ - j .,, " 4 y 7.:- 5 ^ .q ~

., y..

l ..i n.- .rk's* -Q e: - zy .;p c, a 120X (b) INTERGRANULAR ATTACK W w FIGURE 17. PHOTOMICROGRAPHS OF THE CROSS-SECTION OF BENT SPECIMEN A2 FROM TUBE B-8-25 --ii-i

p [ 5 A second IG penetration on the ID surface %40 mils above the main crack { was also observed, Figure 17b. This IGA penetration is N50 percent through wall cnd the spread at the mid-wall is about 10 grains. SEM photographs of the fractured surface of specimen A6 are shown in Figure 18. The crack is completely intergranular and the wall penetra-tion is >90 percent. Cavities visible on the fractured surface in the mid-wall region indicate that the intergranular penetration was not limited to the transverse direction, but that some penetration also occurred in the longitudinal direction. The fractured surface of specimen A6, at high magnification showed fluffy deposits on the grain faces, Figure 18. The EDAX analysis of the fractured surface indicated the presence of sulfur. The concentra-tion of S on the surface ranged from 0.4 to 5.3 percent (as relative X-ray intensity % atomic percent), depending upon the area analyzed. Table 4 lists the EDAX results. The fracture surface of the broken specimen A4 was descaled and then replicated for TEM examination. TEM photomicrographs of the replica are shown in Figure 19. No fatigue striations were observed on the grain facets. This indicates that the failure of the cube was not due to corrosion fatigue but probably was the result of intergranular stress-corrosion cracking (IGC). Specimens A1, A7 and A8 were bent with OD in tension. None of these showed any crack on the OD or ID surface. Specimen A7 when examined with SEM showed only minor surface tears at the U-bend apex, but no deep cracks, Figure 20. The crack in specimen A at 1.25 inch was circumferential and limited to specimens A2 to A6 which covered 45' to 270* of the tube. Specimen A3 was used for AES and ESCA analyses. The surface [ of A3 that was analyzed contained intergranularly cracked area (s 80 per-cent of surface) and freshly fractured area (%20 percent of surface); the latter was produced from the mechanical removal of the uncracked { ligament from the specimen. The AES/ESCA instrument used in surface analyses was a Leybold-Heraeus LHS 10 System. The analytical chamber of the instrument operated [ [

f h I t l g g,- % rv i m. a- .b I i l 120X (aT OVERALL VIEW i. = I 'u . ;,j 4

s2f,

?!% ..g v : n *., l 'A3 pf,l f - ~ f . -n ; ~& e ., ~ ~ a og. N ps_ _ .g I (b) FLUFFY DEPOSITS i i FIGURE 18. SEM PHOTOGRAPHS OF THE FRACTURED SURFACE OF SPECIMEN l A6 FROM TUBE B-8-25 I

[ [ [ TABLE 4. EDAX ANALYSIS OF FRACTURE SURFACE { OF SPECIMEN A6 FROM TUBE B-8-25 Relative X-Ray Intensity [ { Tear Area In Fracture Close Fracture Fracture Close Remaining Ligament To Tear Center To I.D. { Cr 16.5 17.1 18.3 18.0 Fe 10.1 7.2 7.3 4.0 Ni 73.4 74.9 73.4 72.6 S 0.7 0.4 5.3 [ [ [ [ E [ rL lu w h

[ 8 ^ ; p -- r ~ , yh,,- l .N 'W ~. ~ ,~. .a .g,., s.:. 7. MMi:LY~ F55N4 ^ ~ + y

p CI

,A ._y c { W ~ <-;k' l f..ie ?. ~ ..i A's * ,.n ,,%ff -.~ . ?.:.; &: y'h; & ?. T U.h ( .w - ; :-, ;_3 .;. n. - y,l } '.. t . f g,., y i-( .6 .W L. / 4 4 ~

f..,.. f. ?

_ *5N'Q_ W~2.. 3:lig. JQ i '^n. 5100x 14gggg- { (a) GENERAL VIEW (b) GRAIN BOUNDARY ^ ~ T "r g ggr - - =,3--t;. jf..,.. ~ f a.7; .c 9 .p e [ m}, s.t i

jh

~ .S ^ ) (?:r.;f f. ' %~ ~ r ^ .k .*'f,

f..f'5

, p; );V \\

r...

. g t. g Ml,.%yd, fi* . <.%g . &f"..

,)
I

[ r - [- lg.rt: .: p s.- y ~ w e. ~.. A, . y.., f,*- Sk 12000X I (c) GRAIN BOUNDARY TRIPLE POINT l FIGURE 19. TEM PHOTOMICROGPAPHS OF THE REPLICA 0F FRACTURE SURFACE OF SPECIMEN A4 FROM TUBE B-8-25 AFTER DESCALING ^

L 9 [ [ [ [ ~~M-._ -~ m.- ^ r' -='G: Da p .,--2__q 1

>W '~ - &g- ::re --

L r FW**[ c, '~~ _ _ _..~.'.4 6* W 4. f(w ~ [ ,,.y- -Q-,.~ ~~ { W" ~..._ - E n ( 650X [ r l FIGURE 20. SEM PHOTOGRAPH OF THE APEX OF U-BEND OF SPECIMEN A7 FROM TUBE B-8-25 1 L L I F l

L [ ( 10 -10 -9 at 5 x 10 to 1 x 10 torr vacuum. Auger spectra were produced with a scanned electron beam of 5 kev, having a 5 um spot size. Scanned areas on specimens for auger analysis ranged from 0.5 x 0.5 mm to 1 x 1 mm. ( Auger electrons emitted from a specimen were analyzed with a hemispherical photomultiplier detector. The Auger spectra of specimens were matched to ( standards given in " Auger Electron Spectroscopy Reference Manual"* for the identification of elements. ESCA spectra of specimens were produced with Mg-k, 1254 eV x-rays, i from a gun operated at 12 kV and incident at s45' onto 1 cm2 area at sample location. Thus, the area of a specimen analyzed was that exposed to x-rays in the 1 cm region. If the surface area of the exposed specimen I 2 was smaller than 1 cm, the x-ray photoelectrons for ESCA were produced from the available surface, and the resulting signals were proportionately weaker. The spectra of specimens were interpreted on the bases of standard binding and kinetic energies of electrons given in " Handbook of X-Ray Photoelectron Spectroscopy."** The A3 surface was analyzed with AES in the as-received condition, and after sputtering away 600, 900, 1200 and 15001 of the surface film. The AES spectra was taken at several spots on the A3 surface, namely, in [ the IGC area at tube ID edge, center OD edge and front edge of the specimen. The freshly fractured area of A3 also was analyzed. A quantitative esti- { mation of elemental distribution over the entire A3 fractured surface was determined with ESCA after sputtering 0, 600 and 9001 of the surface film. [ The results of AES analysis are given in Table 5 and those of the ESCA analysis in Table 6. The AES detected a host of elements on the fracture surface as [ shown in Table 5. The elements detected were C, 0, Fe, Ni, Cr, S, P, C1, B, Be, Zr and various fission products of uranium. The most predominant ele-ment on the surface was C, its concentration was 90 atom percent in the top layer but decreased to 64.8 percent at 9001 depth.

G. E. McGuire, " Auger Electron Spectroscopy Reference Manual", Plenum Press, NY (1979).

l

- C. W. Wagner et al, (Eds.), " Handbook of X-Ray Photoelectron Spectroscopy" Perlain Elma Coro., Minn. (1979).

F L

( 11 I l [ ( TABLE 5. AES ANALYSIS OF FRACTURE SURFACE OF SPECIMEN A3 FROM TUBE E8-25 Depth Area on { Sputtered Fracture Surface Elements Detected None IGC Center C, O, Cs, Y, Sn, La { Fresh Fracture C 600A IGC Center C,K,B,Sb,P,Be ( IGC Edge C, K, Be, P, Y, S, B, Ag, Te ( Fresh Fracture C,Ni 900A IGC Edge C, Ag, Ca, S, P, Be, K, Ca ( 1200A IGC Edge C, Si, Be, S, Cl, Ar, Rh, Ag, O, Cr IGC OD Side C,Be,Ca,Ag ( IGC ID Side C, Si, B IGC Center C, Si, B, Ag, O Fresh Fracture C, Ni, Cu, O ( 1500A IGC Center C, S, Ru, Ag, Cs, O, Be, P IGC Edge C, Si, Be, P, Cl, Ag, Te, O, La ( ( ( (

l 12 lI i II l TABLE 6. ESCA ANALYSIS OF FRACTURE SURFACE l OF SPECDfEN A3 FROM TUBE B8-25 I Depth Sputtered, Surface Composition, atomic percent A Ni Fe Cr O C S Ag Be Cs P Ru Cu* None ~10 ~90 l 600 3.0 1.7 1.1 11.1 75.8 4.6 l 900 3.4 2.3 1.5 11.8 64.8 0.4 0.3 7.2 0.4 0.5 0.9 6.6 i l c a. ~.,,a n.u n. 1 1I 1 l lI 1 1I I l

I 13 The distribution of elements on the fracture surface was extremely nonuniform. Not only the distribution varied with the sputter-ing depth but also from one spot to the next, which might be only a few micrometers away. Worth pointing out is the heavy concentration, 7.2 I atompercentofBeobservedat9001depthinESCAanalysis. The total fracture surface area on A3 specimen was not large enough for very detailed ESCA analysis. Therefore, a second specimen, namely B, was used which consisted of multiple fracture surfaces stacked together, as shown in Figure 21. The results of ESCA analysis are summarized in Tables 7 and 8. Note that both S and Cl were detected at N1 atom percent level uptothe2300Idepthanalyzed. Boron also was detected at this depth in 3.3 atom percent concentration. Carbon as usual was high, 64.2 atom percentat30I,butlower48.9atompercentat2300Idepth. Note also theverylowconcentrationofoxygen,20.6atompercentat30I,which decreased further with depth to 14.9 atom percent at 23001. The chemical state of the elements, as determined from their respective binding energies in the ESCA spectra, are given in Table 8. Only Cr and Fe are shown to be associated with oxygen. Sulfur was in its reduced state as sulfide, S", (probably as NiS) and C is thought to be in a form similar to graphite. I The ID surfaces of specimens C1 and C2 were only visually examined. The whole surface was covered with dull-color surface film. Along the length of the specimens several brown color decorations also were observed. Each decoration had the appearance of a tree, i.e., it had a main trunk and some spreadout branches, see Figure 22. The brown color decorations gave an appearance of some corrosion attack on the metal surface, underneath the surface film. I Some dark brown circumferential lines (wider than a hair strand) were observed at locations 3.0, 3.5 and 4.0 inch on the C2 specimen. These lines gave an appearance of circumferential cracks, which may have been the cause of EC indications at these locations as reported in Table 2. The specimen was not examined any f: ..ie r. I I

I 14 I I I I I g

~

I I 4X I I FIGURE 21. PHOTOGRAPH OF SPECIMEN B FROM TUBE B-8-25 SH0tJING MULTIPLE FRACTURE SURFACES STACKED TOGETHER FOR ESCA ANALYSIS I I I I

I 15 i I 1I l j ' TABLE 7. ESCA ANALYSIS OF FRACTURE SURFACE 1 0F SPECIMEN B FROM TUBE B8-25 1 Depth Sputtered, Surface Composition, atomic percent i A Ni Fe Cr O C Cl S B I 30 8.9 2.1 2.8 20.6 64.2 0.8 0.6 150 18.1 3.2 3.6 16.8 58.3 0.6 0.7 j 400 15.4 2.8 5.6 18.4 55.6 1.5 0.7 1100 15.7 3.2 5.1 15.5 59.7 0.7 '~* '~' !I ,iI ! I l. I lI l I I i

[ 16 [ [ [ ) { TABLE 8. ESCA BINDING ENERGIES AND STATES OF ELEMEl'TS ON SPECIMEN B FROM TUBE B8-25 Depth Sputtered, A Ni Fe Cr C Cl(a) B(a) 3 30 852.0 Ni 709.0 FeO 576.0 Cr2 3 285.0 C 162.0 S-2 0 150 852.0 Ni 709.0 FeO 576.0 Cr2 3 162.0 S-2 0 400 852.0 Ni 709.0 FeO 576.0 Cr2 3 198.0 162.0 S-2 0 [ l100 190.0 { 2300 198.0 190.0 162.0 S-2 (a) Too low intensity for hish resolution. E E E l FL L

17 I m .M-e s [ I ,I

~

l FIGURE 22. A TREE-LIKE BROWN DEPOSIT ON THE ID SURFACE OF TUBE B-8-25 I I I l

I I 18 I 4.2 Tube B-ll-23 I Short-pull tube B-11-23 was examined using the following techniques: lI (1) Visual examination of U-bend slices (2) Metallographic examination I (3) AES and ESCA examinations of fracture surface film (4) Wall thickness measurements. Examination results are summarized in Table 9, using the same format as in Table 3. Details of the results are given below. Visual examination of U-bend.**. ices from specimen F (i.e., F1 to F4) showed cracks at location 1.25 inch, 180 to 360*. The specimen had an EC indication at 1.0 inch location. The crack penetration was 70 to 100 percent through wall. The matching specimen E (i.e. O to 180*) also showed 100 percent through wall crack, see micrograph in Figure 23. An intergranularly attack?d area, %20 percent through wall, was observed on the 0* face of specimen E; the IGA is shown in Figure 24. A crack in tube B-ll-23 at location 2.5 inch, 90 to 270*, was l through wall and visible to the unaided eye. A photomicrograph of specimen A containing the above crack is shown in Figure 25. The section of the crack shown is at the 225* circumferatal location on the tube. The crack is clearly intergranular, with IGA spread at least 7 grains deep on I either side of the crack. IGA also was observed on the ID surface of the tube, away from the crack, +0.25 to - 0.25 inch, and also on the wall (45* face) opposite to the crack. The IGA in some places was 3 to 4 grains l deep as shown in Figure 26. The specimen was examined after an exalic acid etch, and no IGA was found on the OD surface, either at the 45* location or the 225* crack side location. This clearly indicates that the IGC initiated on the ID side of the tube. I The microstructure of specimen A after a phosphoric acid etch is l shown in Figure 27. The grain size ranged between 5 and 50 pm. The average u size of grains, when viewed at 100X, corresponded to ASTM No. 7 or 8. Discrete r e' ~

m m W W W W W W W W W W W W W W W W M ~' TABI.E 9. EXAMINATION RESULTS OF TUBE 1111-23 Specimen 1.ocation Number Inches Degrees Type of Indication Examination Result / Comment A 2.0-2.75 45 225 Radiograpi, Visual Metallographic IGC 100% Wall,ID-lGA y Structure Discrete Carbide ppt Inter + intra Granular 11 7.0-7.5 0-360 None STEM Sent to MIT C 2.0-3.0 225-270 Radiograph, Visual AES/ESCA Crack Surface liigh C Some S D 6.5 7 0 0-360 None EPR Sensitization Specimen Mounted But Not Used E 0 2.0 0180 EC, Radiograph Wall Thickness OD348 min /0D365 max. r y Structute IGC 100% Wall 180", IGA 20% Wall 0* F (Fl-124) 0-2.0 180-360 EC, Radiograph Wall Thickness 0.036I min /0D377 max. !!cnd Visual FI 0-2.0 f S0-225 1 25,1.50,~ 100% Wall F2 0-2.0 225-270 1.25,225 250 70% Wall F3 0-2.0 270-315 1.25,270 315 70% Wall F4 0-2.0 315-360 1.25,315 360 90% Wall G 7.5-12.0 0 360 EC 10.0 Corrosiim Test Sent to ORNL

) 20 [ [ [ l O d..,T;, t. y e, w, l e.,.. e m 7.j ;t. 9 ;;.. e gg. : tm r, s.q -g -c.-', . "J. cq. :%7 h - ? , Q 2. _ lrf ]% - AG.C ' 49., fsw s ++ sC%h*b <-b)? ( ,~ ~ ' ',

- "A,

!! 0-' cf; s ; j< pk,. s.;* .f,y - 4 4.-%,l. - r.. .,3, y .l_, ks j f ',,.g t,, y si y x' s.W/,g ?....(- ~, '. -.. 'f ;, .1 -f,. e g -4. 4.+ c..m J'%',- < g [, jr.' *% ,, s,* Q'y f e L' [ fl ' 'sa' iff. ?L 1 ' $ .,' 7,. -', 'f_ P '. - [ 9. t [ g+O- . F p ',Ai j g 'Q}' g-n ,,;.,e ?, ...,.p 5_ 2c. , - y A(, e j i n. . c$ .-s +2v y.y. w y. m n '7

w..y N;; y,*,;*- M,

; -., g ;

[

m.., :. r

. -.,-; r,..., o ? w,, -. 44 y e 3 :,, a J +, ' r E[^ f+ s q, r, - e 3-- ~ f. ,~ ? -j g4 , 'h;;; -{ ? ],.ly t ,- 3.-,

f. g., ;.,%

1 [ s'+8'-'3...m h-y',. T,. --4# 1 e g -.g ; 1 -,1 1 ,'X - I1 ". >, g. I[ y. .*sr '4',t I p',* +.,.. '.[T,F4 . ", T h. ~ ' - { ' 'e ? <.,,, Y^E .. ; e. 4 *

2w G11Mtpu

=>: ';;g

~ .'), e y.. -. ?- %l ,.nw 3R ji d W,.% k.g,'~. $ f W D-- , [ h fi- ; ..x., v t w.g;A.g. h, 1-- .). ( r., p twp. : .. e. a as <~ .. sc v, i * ' '. -j ( J.' ,i$Ier '.,f',e . og= + a e.d* fy :. + %,' '- F [ 100X FIGURE 23. PHOTOMICROGRAPH OF IGC IN SPECIMEN E FROM TUBE B-11-23 ) [ [

L [ [ 21 [ [ [ . u. g. c m w* ::'. e. a* Au, ? ' . }~;j-- ~, :;.g4 n. x' j. n; ;. + OD - ~ Q [ - y .<., :,d;f5s,6 . g - v :r e }.qe ;' q. .,s - =. M. * >:d' y*4 g.

r. f

- s c, ,, ~.. %, e.'. E,t%.A % l,f.,.',. ' ,.'.x * % $ s %- Z *3 J t#

~

s , &e w e 4 - Y *f.?",. v. Y, [ ,..L.R*'..f n-t af&? ; 21' w: y ,, *w.r i ~Q/-*Y*h.Fr'.?igt M 94: 6Q+..yw. a. y?,%~h*; :: - ~crs n. - ' As -' ? *.fm * - l' V '.' T ' ' *'* s n_ J

fQ4 " ;kl.XL*

$ ]$ . Q p,4-m.4 f ** n g*a

0 p 9 ',

% v, q .*p. et a. ws e, ^* +4 ns

e

~s -.',q g. o + = ri.'*%'r, 4 ; ry l } ,? ' 4

_.G

~ 4$. gaf 'ny %. f\\f K '

v 4

L , %sg:.~;u : -:4 5 s..',} f. S w %.a*yL D.'; F *.7.~.*=; j 9: ~ . /- s* , W. f.,#O. ( ; L f. -$ p p. 4. .c $:.%,,,~:<.4 . < C,.? ; -

-*.'.Q'b g W >Y w'.,: a n.'k y w

.u..~;J V [ i .s u.. . - y e. ,sG. -r-g s -m. ., g w. _1 1.' s,* g..,M,. ~..i. ] ~ ~ [ 100X [ [ FIGURE 24. PHOTOMICROGRAPH OF IGA IN SPECIMEN E FROM TUBE B-11-23. [ [ (

l l 22 l i l a OD @.h...mi,.~ ..t i ,.,,,v .u._. a w, ,.Qp'-?.,.,7< g< ..so j.t ng

4,

s 1 qy': &;f [L.,

7 } tlf[':f]**fp;.y'*"y... -

* ;a.

-? 'f~,';N, F.,.;

P 3:-(

' P (;E j . e-...~ M Mr i c 70, g w ,.y ..; g,.. 1 1 9. e ,. W.. T.:. -- C . W-S,;;i ~ p., ./ l -X. . /~. 1 I % x,w;:x.wi. Q \\ l f.; q3.,',p y.~m )

. s ' V !,

\\ pg, gf %,4 ~ M r j- +' c.y,e v..;... r -: g

L.,. L N. "

.?., I b;;;, ;,vll-N: }m? _. 91: j,j. y g-c .. _ y. _. ? y.,

I'.;7:;:L Y Y,..

E ,7 ::, - C;: a ': 4 J ' j, %,, ?.- ~. ;, y. %, h.",.. [.,;.[i,/ j I

c. e..- :.,

og B '.. e;,. ;,., IO;'. ::: ys~

r. ;

ID I 60X E FIGURE 25. PHOTOMICROGRAPH OF SPECIMEN A FROM TUBE B-11-23 SHOWING ICC AND SEVERE IGA ON EITHER SIDE OF 4 THE CRACK I i

I ( 23 [ [ [ [ [ [ ~.,1< , ;;J .. Jb .~>:.. c. . a. 60X [ l FIGURE 26. PHOTOMICROGRAPH OF ICA ON THE ID SURFACE OF SPECIMEN A FROM TUBE B-11-23 l [ [ l ( k

l [ l, f 24 { ( ( ( -? ,lfh.').,,h.W]ilp.f,.g-. ?!Y, $& E ~=:swh)i$.m 's},.<. 2. '.;LGd. f:.U Bf$y. iSY C f. s. h),. ~.,;.;A .; y m +e w.3:. PN,ii&,.1 .. s.. s o c. ea rei:" *--.

.pgj_.g,..a;.

L, 7:.W.yi.s.1~.W....~jl$:#p,y. t3 ^?c#. ".!3.:]y, I (. $d

s... L - e,... ~

-,n. . <w - -....

a. _. : w.,.,g,. y%;,..:..:..

=;. ..w a ( W1..?.Wj;~ll?$h#..,..s..y f.:lu-[y$ $..? '?. w z. a s &?;.. .-.n.- Q 2... x. t n.. u.r ~,..; L..o_ .w Q n,.; eq~:;..,-.. .., y %My.- n _..'Z;r.y ;>...~ c... ~... . ~n, a ,y.,.. . 9.. /, %.g.%w,. ;.'~,, s.g :. .,2;y...,,. .s d.::..... v v.~,g,.,a\\ ' ; i. 1.*,.;: .... -w - a ..y. ;. J g. n q- .,v...~ ~ u.. .r [ [%+ib,.y. r +:.

1 m..

-;;,.~ . x 5.-N.[.. .- d .:w. -~.-... hD..,.. b.. $1. ,.#tg. ~ s :.. <: e....;; w N..e,. ,e...,' .+ 4 s,Ae.- n. s .. :.y. - .KN$h..- t. ?~'T ;$. 4Dl, i. f.$s.W d* j i%,~52{ s.c2 - - m m...m. A.,,_u..,.,j;_,, g;c.,;a 500X { FIGURE 27. PHOTOMICROGRAPH OF THE MICROSTRUCTURE OF SPECIMEN A FROM TUBE B-11-23 ( {

[ {. chromium carbide precipitates inside the grains and also on the grain boundaries ace visible. The grain boundaries already affected by IGA were heavily attacked on etching with phosphoric acid. ( The remainder of the above crack contained in specimen C was analyzed by AES and ESCA. Spectra of the as-cut specimen showed C, 0 [ and N1 on the fracture surface. The surface was sputtered with 2 kev argon ion beam, and after removing 20 I of the surface film, sulfur was detected at one edge of the surface. Carbon on the surface was in very large concentration in comparison to the rest of the elements. Because of the overwhelming concentration of carbon, a quantitative estimation ( of other elements was not attemptsd. i The fracture surface was further sputtered and analyzed at 50, 100 and 200 A depths. With depth, the concentration of C and 0 decreased ~ but that of Fe, Ni and Cr increased. The concentration of S increased up to50Ibutdecreasedthereafter. The analysis was not carried beyond [ 2001 depth. Wall thickness measurement results are given in Table 10 for ( specimens E and F. The maximum thickness measured was 0.0377 inch and the minimum 0.0348 inch. The thickness was generally 0.001 inch lower { in the rolled section than in the unrolled tube. 4.3 Tubes A-23-93, A-88-ll and A-112-5 Short-pull tubes A-23-93, A-88-ll and A-ll2-5 were examined [ only visual'ly for defects in the roll transition region, by slitting the ( 1 upper section (0-2.25 inch) of each tube into eight slices and bending l 1 { them into a U-shape. Wall thickness measurements, however, were taken of one tube, A-23-93. Tube A-23-93 had an EC indication at location 1.75 inch [- and tubes A-88-ll and A-ll2-5 at 1.25 inch. Results of U-bend examinations of tubes A-23-93, A-88-ll and A-112-5 are summarized, respectively, in Tables 11, 12 and 13. Wall thick-ness measurements of tube A-23-93 are given in Table 14. c L F L r-(

lI i 1 26 I I TABLE 10. WALL THICKNESS OF TUBE B11-23 I Wall Thickness. inch

Location, 45*

135* 225' 315" inches Position Position Position Position 1/8 0.0348 0.0350 0.0368 0.0365 1/4 0.0354 0.0351 0.0361 0.0368 3/8 0.0354 0.0355 0.0364 0.0367 1/2 0.0352 0.0353 0.0362 0.0363 5/8 0.0352 0.0352 0.0364 0.0362 I 3/4 0.0356 0.0351 0.0364 0.0362 7/8 0.0353 0.0350 0.0368 0.0362 1 0.0355 0.0350 0.0370 0.0365 11/8 0.0362 0.0360 0.0372 0.0370 11/4 0.0363 0.0359 0.0372 0.0372 13/8 0.0364 0.0361 0.0372 0.0371 1-1/2 0.0362 0.0361 0.0373 0.0370 I l5/8 0.0362 0.0361 0.0377 0.0368 13/4 0.0365 0.0360 0.0372 0.0370 17/8 0.0365 0.0359 0.0372 0.0370

~T R __ f T R R _R R CM-W- M ~ M' TABLE 11. EXAMINATION RESULTS OF TUBE A23-93 Specimen location Number inches Degrees Type of Indication Examination Result / Comment A(Al A8) 023 EC 135.20140 Wall Thickness 0.0357 min /0.0371 max. Radiograph Lip Crack Slit and Hend 1.5 Pits, No Crack Visible On As Cut llalves Visual AI 02.5 045 No Crack A2 0-2.5 45-90 135,6040 Crack ~ 100% Wall g A3 02.5 90-135 I35,90-135 Crack ~100% Wall, Also Pits A4 0-2.5 135 180 135,135-140 Crack ~ 95% Wall A5 0-2.5 180-225 No Crack A6 0-2.5 225-270 No Crack 2. 23 -270 Pits A7 0-2.5 270 315 0.25,270-315 Lip Crack 100% Wall A8 0-2.5 315 360 No Crack 11 2.5 12 0 360 Reserved for il&W 1 . ~.

W W W M M E E TABLE 12. EXAMINATION RESULTS OF TUBE A88-ll Specimen Location Number Inches Degrees Type of imlicalion Examina: ion Result / Comment A (Al-A8) 0-2.5 EC, Radiograple Slit and Bend Pits at Several Places Visual AI 0-2.5 045 025 Lip Ciack A2 0-2.5 45@0 2.25,6090 Crack 90% Wall A3 0-2.5 90-135 2.25,90-120 Crack 90% Wall A4 0-2.5 135-180 1.25,130180 Crack ~100% Wall A5 0-2.5 I80-225 1.25, I80-200 Crack ~100% Wall A6 0-2.5 225 270 0.25 Lip Crack A7 0-2.5 270-315 No Crack A8 0-2.5 315-360 025 Lip Crack 2.25,330-360 Crack 80% Walt B 2.5 12.5 Unused I i I t

r-- r, c-r-- c-, e-- c-- . n, n TABLE 13. EXAMINATION RESULTS OF TUBE A112-5 Specimen 12walion Number inches Degrees Type ofIndication Examination Result /Conwnent A (Al-A8) 0-2.5 EC, Radiograph Slit and Bend % nual l AI 0-2.5 045 1.25,045 Crack ~100% Wall A2 0 2.5 45-90 1.25,4540 Crack 90% Wall, Lip Crack A3 0-2.5 90-I35 w No Cracks A4 0-2.5 135-180 0.25 Lip Crack A5 02.5 I80-225 0.25 Lip Crack A6 0-2.5 225-270 ) 025 Lip Crack A7 0-2.5 270 315 0.25 Lip Crack A8 0-2.5 315-360 IJD l.25,315-360 ~100% Wall H 2.5-12.5 0-360 Reserved for BW l 1

L [. {- 30 l [ [ [ TABLE 14. WALL T11ICKNESS OF TUBE A23-93 Wall Thickness, inch

Location, 45' 135' 225' 315' inches Position Position Position Position 1/8 0.0368 0.0361 OD359 OD357 1/4 OD371 0.0360 0.0358 OD357 3/8 0.0370 0.0359 04357 OD358 1/2 0.0359 04360 0.0358 OD358 I

( 5/8 0.0359 04358 OD358 OD358 3/4 0.0360 0.0360 OD358 0.0356 { 7/8 0.0361 0.0363 0.0357 0D358 1 0.0363 0.0362 OD358 OD357 11/8 0.0371 0.0365 0.0361 04362 I 11/4 0.0369 0.0362 0.0361 OD365 13/8 04366 0.0362 0.0362 OD368 ( l 1/2 0.0362 0.0362 0.0362 OD369 15/8 0.0368 0.0362 0.0365 04368 { 13/4 0.0362 0.0362 0.0370 0.0367 17/8 0.0362 OD361 0.0368 OD366 b [ [

[ 31 Visual examination of tube A-23-93 revealed lip-cracks at ( location 0.25 inch, 270-315', and a second crack, nearly through wall, at location 1.75 inch, 60-140'. Some pits, probably due to grain dropping, { were also observed at 1.5 inch and 2.25 inch locations. Wall thickness of tube A-23-93 was much more uniform than tube B-11-23. The maximum thickness in tube A-23-93 was 0.0371 inch and the minimum was 0.0357 inch, see Table 14. Three separate cracks were observed in tube A-88-11. One was ( a lip-crack at location 0.25 inch, 315-45*, the second crack was at 1.25 inch, 130-200', and the third crack was at 2.25 inch, probably { between 330 and 120*, see Table 12. Two cracks were observed in tube A-112-5. One was a lip-crack at 0.25 inch, 135-315*. The second crack was nearly through wall at [ 1.25 inch, 315-60', see Table 13. ( 4.4 Tube A-71-126 { Long-pull tube A-71-126 was examined with the following techniques: [ (1) Visual examination of U-bend slices (2) Visual examination of ID surface af ter descaling ( (3) Meta 11ographic examination (4) Tensile test { (5) Alloy composition l (6) EPR sensitization test (7) AES and ESCA of the ID surface. Results of these examinations are summarized in Table 15 according to the format of Table 3. Details of the results are given below. Specimens for visual examination of U-bend slices were taken { from three different locations of the tube. The specit:en X (X1 to X8) was from location 0-2.0 inch, specimen Y (Y1 to Y4) from 2.0-4.0 inch, and specimen F from 50.5-58.5 inch. Thus, specimen F was from the long-L pull section that was well below the tube sheet crevice region. i

7 rm r unal e e m m m m m m m m m m m m g g TAnl.E 15. EXAMINATION RESUI.TS OF TURE A71-126 Specimen location Number inches Degrees Type of Indication fixamination Result / Comment X (XI-X8) 0-2 D Radiograph 1.75 Slit and Bend Visual Xi 0-2 D 045 0.25 Lip Crack X2 0-2.0 4560 0.25 Lip Crack,ID Pits - X3 0-2.0 9'l-135 l D Pits X4 02.0 135-l80 No Cracks X5 0-2.0 180-225 No Cracks X6 0-2.0 225-270 Possible IGA X7 02.0 270 315 0.25 Rolled Metal X8 0-2.0 315-360 0.25 Rolled Metal Y (Yl-Y4) 2.04.0 EC 3D Slit and llend Visual YI 2.04.0 090 No Cracks u 12 2.04 D 90 180 No Cracks Y3 2 D4.0 180-270 No Cracks Y4 2.04.0 270-360 No Cracks Z 4.0-13 0 0-360 Unused S 13.0 25.75 0-360 Unused T 25.75 38.25 0-360 Sent to ORNL U 38.25-51.0 0-360 Sent to ORNL A 59.0 60.0 0-180 None Alloy Composition OD34% C,15. 3% Cr B 59.5-60D 180-360 None EPR Sensitization Activation Potential I10 mV (SCE) C 52.0 52.5 180-360 Nonc ID AliS/liSCA B (3.5%), Zr (0.3%), S (I D%), SOj Top /S" BotIom D 52.5-59.5 180 3.;0 Radiograph Visual Descaled, No Obvious Crack li S I D 52.0 180-360 Visual - Gouges None Unused 1: 50.5-58.5 0-180 Radiograph ID llend, Visual No Cracks (3X Visual) l

O O_ O O O O O O M O O O O V D UM~M TAlli.E 15. (Continued) Specimen location Number inches Degrees Type ofindication Examination Result / Comment G 60.0 68.25 0 360 Radiograph Tensile Test, Slit YS 53 KSI, ITIS 101 KSi Visual No Other Cracks, Descaled 11 68.25-68.5 0-360 Visual -Gouges None Unused I 53.25-53.75 180-360 Visual - Pits Transverse Met. Mechanical Indentations J 53.75 54.25 180-360 Visual - Pits inngitudinal Met. Mechanical Indentations 1 l l N i i

1 1 34 1 In specimen X, a lip-crack was observed at 0.25 inch, 0-90*. Some pits were also observed at 1.0 inch, 45-135*. The nature of these pits, i.e., whether they were formed from mechanical damage of the tube or from grain dropping was not obvious. No other crack, beside the lip-crack was found in specimen X. The radiographic indication at 1.75 inch was not confirmed as a crack with visual examination. No cracks were found in specimen Y, which had an EC indication at 3.0 inch location. Specimen F was a 8.0 inch long slice which was bent into a C-shape over a 4-inch diameter tube and then examined. No I crack was found over its entire length. The ID surface of specimen F, however, before bending was found to be heavily scored practically all over its length. A photograph of the descaled scored surface is shown in Figure 28. The scoring marks were possibly responsible for the defect indications in radiographs, described in Section 3.3. I A general view of the ID surface of the tube section between 51.0 and 60.0 inches, before descaling is shown in Figure 29 Score marks can be seen on all quadrants of the section over its entire length. The mechanical nature of the ID surface defects, as shown in Figures 28 and 29, is fairly obvious from the photomicrographs in Figures 30 and 31. The photomicrograph is Figure 30 is a transverse cross section of specimen J, see Table 15 for their locations. The well defined trapezoidal g shape of the indentation in Figure 30, and the crushed grains on the B surface at the defect site in Figure 31 are indicative of the mechanical penetration. Specimen G, location 60.0-68.25 inch, was used for determining the mechanical properties of the tube material. Two bullets (each 3.125 inch long) were inserted in the tube, one at either end, for proper gripping and defining the gage section. Thus the gage section in the g specimen between the bullet heads was 2.0 inch. The length of the 1 original 8.25 inch specimen after testing was 9.37 inch, i.e., the total elongation in the specimen was 1.12 inch. Since no reference marks in i i

[ { 35 I ( [ 1 . :. i > yc -+. [ ,g D *' '[ d cl:. - =,..} ' 'i q. ' - [ [p. r. [ . c.' - li ;"l*. k,,f.^ f ~- i - k, y [ $.*d'.. ) i je o ..:~ .,. 3l- - h t ,( k-t 1 FIGURE 28. PHOTOGRAPH OF THE DESCALED SPECIMEN F FROM TUBE A-71-126 SHOWING HEAVY SCORING (

l O O DOCUMENT PAGE "~ PULLED O 9 ANO.w me a NO. OF PAGES, ( REASON O PAGE inEGS;.E. D wD coe< fad AT. von er ot.e 1 D del 1ER COP (REQUESTED ON _ PAGE100 LARGE10 FILM. >> * '=a " *014R _6 ~ AMMb-- O hRLMED ON APERTURE CARD NO ~ - - - - - - -. - _ _ _ _ _ _ _ _ _ _ _ _

l 36 !I I lI i j lI i

I-tl FIGURE 29.

PHOTOGRAPH OF ID SURFACES OF T'n'0 HALVES OF TUBE tu A-71-126 BETWEEN 51.0 AND 60.0 INCH BEFORE DESCALING: A) 45' B) 135" C) 225* AND, D) 315* FACE g <SEE NEx1 PACE) i II 1 i i

I

~I i !I iI

37 -.~. 7,--., p 3 +n - _ ~. ':.;;.,-hbibhNh" !,;f^ r. w ,,,,. s?.%?:k*t~ ~ u#J... 3 - N.,e**^^~, & yyyY '~

s "

y-

u.., ;;

ST M.4 e J.. r .+ '..L : ' \\.. sr ..1, .,. s v,; b r g .,,,,;,. f ~~- ?} g, or n ~ ry ,1 ,.,i 1; ! y?*~ ,-~ < j; ID 200X FIGURE 30. PHOTOMICROGRAPH OF THE TRANSVERSE CROSS SECTION OF SPECIMEN I FROM TUBE A-71-126

I 38 7 , ;.f*:w".=. o s . w( !

Q, _, ~ -: -

eA. ~ s- , d. 1. ~~~,b i, n - - - I s ' ;.; it..'. ~ < -r. .. A.u,; r.,. '.b ^ -3 .j., .3 b, t._ ...r.'* ..a so [., ,S h.C. ~ f M " E it I b I 2 y,3 ~.ViX % f l$-. M. 'e'.;i. ' '. :.. q,Rd '.,1 ID V ' /Ifr Dl rt % ['. ' ' 2 Q,. :;~.+;r :.a e ',. i <s a g l-e y:q s. . 7., ':!..Tf'f. %. -

n '.,. f.

%~g$_y d *As, M.c.,;3:.2,;.1 i ' y. '*+p 4 .n 1 p v.. M,y e.o.,A,. 3. h._,4.. -c' Ve .. '.. :.. ?; c....,, ,; s,d,,...:f:

n

.e.,..- -. 1 , t <.. g 1 .g.A. . x. ~ m,, ;, ;hqs... - : m y '.c: -: es:3 ve a:Iq n q + ;. +. a., f. e p. u,..;e } s os s V : . aa - j _ y

,n_r

{e.. ..., e : 7 .t ~ L.,V. :.-.,.n +s.. .y a -. a ~., .A- .. *,y sj '._y. ,s;yT e1'; ; t 4.g. ! r* , s s ^

y. %

s

. #.y' s .,-A > 12... - +.. ; 1,, _,,+ r. o ~., 4 t., , 3 - -,,n. ..a,. y, z s. t, . s. ~,. , 4 r-; e ;. s .. a,. 4 .c.an..a - _..acA w ,.e L 200X FIGURE 31. PHOTOMICROGRAPH OF THE LONCITUDINAL" CROSS SECTION OF SPECIMEN J FROM TUBE A-71-126 W

i ll i I 39 I the gage section were made, the true percent elongation in the ma-terial could not be calculated. The 0.2 percent offset (a s14e gana I used) yield strength of the material was determined as 52.8 ksi from 3650 lb. load at yield and the ultimate tensile strength as 101 ksi from the 6990 lb. maximum load attained. The above values are normal for fabricated Inconel 600@ tubing. Specimen A, location 59.0-60.0 inch, was used for determining alloy composition By the X-ray fluoresence technique for all elements except C and N. The C and N were determined with Leco instrument and N by a standard method. The composition of the specimen is given in Table 16. The composition range with 0.034 percent carbon, and less I than or equal to 0.01 percent S and P, is normal for an Inconel-600@ tubing. The electrochemical potentiokinetic reactivation (EPR) test was done on specimen B to detect its susceptibility to polythionic acid attack. The EPR technique used was that described by Airey et al of Westinghouse.* The activation peak potential for the alloy was found to be 110 mV (SCE). After the test, the specimen surface was found to I be severely attacked in an intergranular mode. The high peak potential of 110 mV suggests, according to the Westinghouse paper, that the tube is in a heat treated condition which makes it extremely susceptible to polythionic acid attack. The ID surface of specimen C, location 52.0-52.5 inch, was analyzed with ESCA. Results are given in Tables 17 and 18. The prime i I elements detected on the surface were Ni, Fe, Cr, 0, C, B. Zr and S. ) The concentrations of Fe, Ni and Cr, as might be expected, increased with the sputtering depth, and those of C and 0 decreased. No definite trend emerged for the minor elements B, Zr and S. It should be noted that the 0 on ID surface of the present specimens at N40 atom percent is about twice as high as that observed on the fractured surface of specimen B from tube B-8-25, and conversely C at N20 atom percent is less than half of that on the above fractured surface. l G. P. Airey, et al, Journal of Metals, 33,, 28 (1981). I lI 1

[ { 40 TABLE 16. COMPOSITION OF TUBE A-71-126 ALLOY ( Element Wc. Percent Ni balance ( Cr 15.3 Fe 9.6 { Mn 0.36 T1 0.21 Co 0.10 Cb <0.10 Mo <0.10 h Cu <0.10 I Al <0.10 { Si Not determined properly (M).2) P 0.01 S 40.01 C 0.034 { N 0.013 E [ [ [ [ [ i..

1 [ ( 41 [ [ [ TABLE 17. ESCA ANALYSIS OF ID SLRFACE OF SPECIMEN C FROM TUBE A71-126 Depth ( Sputtered, Surface Composition, atomic percent A Ni Fe Cr O C B Zr S { None 14.0 2.8 6.9 44.6 28.1 2.3 0.2 1.1 30 18.3 4.1 8.4 43.3 21.4 3.5 0.3 0.6 630 24.1 5.1 10.7 41.1 14.6 3.2 0.3 1.0 1230 25.2 6.8 10.2 39.2 12.9 4.7 0.3 0.8 3630 28.9 8.0 11.7 39.8 10.4 0.0 0.4 0.8 [ [ [ [ l [ [ [ [

[ 42 i ( TABLE 18. ESCA BINDING ENERGIES AND STATES OF ELEMENTS ) I ON SPECIMEN C FROM TUBE A71-126 [ Depth [ Sputtered, Ratio A Ni Fe Cr C S S-2 so4 = f None 856.0 Ni(OH)2 711.0 FeOOH 577.0 Cr2 3 285.0 C 169.0 SO -2 0 { 0 4 l 30 O [ 856.0 Ni(OH)2 711.0 FeOOH 577.0 Cr2 3 285.0 C 169.0 SO -2 1.0 4 162.0 S-2 [ 630 852.0 Ni 710.0 FeO 577.0 Cr2 3 285.0 C 169.0 SO -2 2.0 O 4 162.0 S-2 [ 1230 852.0 Ni 710.0 FeO 577.0 Cr2 3 169.050 -2 2.0 O 4 { 162.0 S-2 2430 852.0 Ni 710.0 FeO 577.0 Cr2 3 162.0 S-2 > 10 ( 0 855.0 NiO 3630 852.0 Ni 710.0 FeO 577.0 Cr2 3 > 10 0 855.0 NiO ( ( [

[- l [. l 43 4.5-Tube A-146-6 Short-pull tube A-146-6 was examined using the following ( techniques: (1) Visual and photographic examinations of the ID surface l (2) Microstructure and microhardness measurements ( (3) EPR sensitization test (4) Metallographic and SEM examinations i (5) Electron microprobe analysis of a pit I I (6) XRD analysis of ID surface scrapings (7) AES, ESCA and SIMS analysis of ID surface. Results of various examinations are summarized in Table 19 using the same format as in Table 3. Details of various results are presented below. The whole segment of tube A-146-6 was slit into two halves, and their ID surfaces were examined visually and were photographed. The photographs are shown in Figure 32. The ID surface was dull in appear-ance and it was covered with thin deposits. Several spots were decorated ( with brown color irregular patterns; these appear as dark patterns in Figure 32. These decorations gave the impression that corrosion (or ICA) had occurred at several points on the tube surface, and the corrosion f product had seeped around them and produced these patterns. { The patterns described here are similar to those shown in Figure 22 from tube B-8-25, and described previously as tree decoration. A minor difference between the two is that the lines forming the pattern f in Figure 22 are narrower than the corresponding lines in Figure 32. The reason for their appearance on the tube surface, however, is probably the same. The microstructure examination and microhardness measurements were done on specimen B, location 0.5-1.5 inch. The microstructure of L l fl..

l M O M M M M M M i TABLE 19. EXAMINATION RESULTS OF TUBE A146-6 Specimen location Number Inches Degrees Type of Indicatian Examination Result / Comment A 1.0 1.5 0-135 EC 1.25, Visual TEM A not Used, Substituted by B8-25 (A4) B 0.5-l.5 0180 None p Structure Discrete Carbide ppt. No Continuous Network p liardness Dril Max. 270/ Min.170 C 8.0-9.0 180 360 EC 8.25, Visual 360 Longitudinal Met. IGC 8.25,100% Wall Both Edges D 10.0-10.75 180 360 EC 10.5, Visual ID SEM, Met. 10.5 Pit + IGC,100% Wall E 8.25 8.75 0-135 EC 8.5, Visual ID AES/ESCA 3100 A S (low) 0 (46%) C (8%) B (3%) Zr (0.2%) F2,3400 A S(0.7%)O(33%)C(19%)B(4%)Zr(0.3%) rgFI-Fil) 0-5.5 225-360 Pits, Visual ID SIMS/ESCA Fi l,3500 A S (1.2%) O (41%) C (6%) B (8%) Zr (0.3%) (0.5 x 1I = 5.5) Ni, Nio, FeO, Cr2 3, S" 0 g G 5.5-6.5 180-360 Pits, Visual Microprobe ID S and Tiin Pit 11 10.75-11.25 180-360 None EPR Sensitization Activation Potential 110 mV (SCE) 1 5.25-5.5 300-360 Pits, Visual XRD Unknown Spinels, Ni3TiOS (Part of Fil) J 1.0-1.5 180-360 CD Deposit, Visual Transverse Met. No Defect Under Deposit, No IGA ID or OD (F3 Specimen) K 2.0-2.5 300-3t>0 OD Deposit, Visual Longitudinal Met. No Defect Under Deposit, No IGA ID or OD (F5 Specimen) L 1.5-8.25 0-180 Tree Decoration, Visual None Specimen Descaled M 8.75 12.25 0180 Tree Decoration, Visual None Specimen Descaled

i r L 45 { [ [ I [ [ [ [ [ FIGURE 32. PHOTC';RAPHS OF ID SURFACES OF TWO HALVES OF TUBE A-146-6; A) 315* B) 225' C) 135' { AND, D) 45' FACE (SEE NEXT PAGE) [ [ [ [ [ [ ~ r L

0 DOCU\\/1ENT PAGE PU _ LED ~ ANO.~ ~ NO. OF PAGES ( REASON D PAGE ELLEGlBl. I D HARD COPr FILED AT. PDR CF i OTHER D BETTER COP (REQUESTED ON / 1 h PAGE100 LARGE10 FILM. h MARD. COP (FILED AT: PDR ') OTHER "D DD ~O . FILMED ON APERTURE CARD NO

46 the alloy is shown in Figure 33 from three different sub-lcoations, i.e., rolled region, roll transition region, and the unrolled region. There is no noticeable difference between the three microstructures. In all three, the grain size is the same, and the carbide precipitates are discrete over grain boundaries and inside grains. No continuous network of carbide is visible in Figure 33. Microhardness values at 180' face of specimen B, taken at ap-proximately 1/8 inch intervals are shown in Figure 34. Knoop diamond pyramid hardness readings at any sublocation was taken every 4 mils starting at %2 mils from the ID edge of the tube. The highest values were in the rolled section of the tube (location numbers 1, 0.5 inch, to 5, 1.0 inch, in Figure 34) near the ID, the maximum was 270. In the roll transition region (location number 6,1.1 inch) the maximum value at ID was only 210. The maximum value below the roll transition region was <200. These values are considered normal for fabricated Inconel-6008 tubes, and the rolled region of the tube is not excessively cold worked. An EPR test was done on specimen H from location 10.75-11.25 'g inch. The activation peak potential was obtained at 110 mV (SCE), = This value is the same as that obtained for specimen B of tube A-71-126. According to the criterion used before, tube A-146-6 is extremely susceptible to IGA by polythionic acid. Specimen C, location 8.0-9.0 inch, had a visible crack and also an EC indication at 8.25 inch. Brown color patterns were present on either side of the crack. Figure 35 is a photomicrograph of the I crack, which is intergranular and 100 percent through wall. A defect associated with another brown color decoration was selected for SEM and metallographic examinations. Some white color I deposits also were present at a few places over the decoration. The area selected was specimen D from location 10.0-10.75 inch, which had an EC indication at 10.5 inch. A SEM photograph of an area including the white deposit is shown in Figure 36. The surface appears crusty around the white deposit. The specimen was then descaled and re-examined. Several areas on the surface were found to have pits and severe IGA. Pits appear to be in I 1 places where white deposits were presen't on the surface before descaling of the specimen. One such pit is shown in Figure 37.

47 [ %m.:z L ':?;gg pfegmz.7 ~. ep#d 'v4.9 m ,; - Lte: x x .[

m. m,.,.

A. a, w, . y. J.,."..:w. ,n4 .a. ,x ml R% ,v-f, n .c% ~ u;t;.; e-e- a.:. c Q n se.'{ o ln, .+6 s ~q w}:r.* p Q.l. u 2 4

3,%.e '}. 'of.).&

g%y t. o e m

'P

-.f,w;;

w..yk.-

N* m me..m:,;.h' p .x

p. m3

., y.

u. u,.,

e..

8 -,.~.2*y.4',. .w: e R..,. o.,; p' ' m: t i,m'a-; ,y

, 4;; e %. o. n-4,. *:e 3;,p@._ x.

g g.g:; 3 - (a) ROLLED REGION

5..m e.

s ,... p. :.:.5) c . W. g. e, yt.pg,.., s.3,. -a a os ,4 a. .m;%. ~ a.c. & :.. s t* *j .y'- L.t.e:f,.. w* 4 ;:,. ye.;%aw 'J [. g> y.n p + W e m .*9,5 9- .R 3.y %. :;r g.5c. m,2,. h.% ry:.W.Y V m f k *,. 'v 4,.f.w$.w.b ,$c..k*2; h. r.s

e. -e a,

. a

. f f &,+.

n., ;. tv,. -, ,,a w,, [ .n,-.

. - a,. g.s.; n <..'s..

b,.*..

n...g... ,,p : 'd. /... .... c.a w ., +... .f 2;p; : 9,;V"yy~ ",. lK.Mor ~:t h _ . : We .a, ^ 'm a g; $. A.p ( p&_.N ue.M E M aW M. m y Wxb L- +.: wr-itM:q$t f f. t :....fi ff .: W. : w 'v v. ?c W'- { fk fh m. s.ger. :. ~., v -. m; .,,l.,.. ??.D'p:.a&g. &..-,..Q ,. f6~;l W ) m$ N~h.u _.-.f 'V n _ 5~;s, p., h, qn

?.

/1 QQl' fWM. ~. - x s. P 4@m. v o. o W,.t W n f c.. y,; N,, 3,. n .*tq r: w [-k'h h '..,, p p.s ' w. ,.. f y: . is .q / W# . u, h.Y,h' y,,MO*$ '.i(' [. f. s. 2 '(b) ROLL TRANSITION j s 4,9 Q.&,,;; : g&. W D ,v, ? REGION m y; 7, . ; f'~1(~ ~ V-';, t j

d., f @Q_.,@. i 1 :{ 4./ J/ 3 $.l. W ] y.)i, h c*.

95, .4 .f* _,I I e> ~ .c y ;y, .s, .c w., )QJQ;[('Q:g ~9.g,y,,,, n. w,. &c ;,a*. - j ' ~$'% ;y w.s ' ~ p ~ ;, w ~ .],

1

~ ~ y '. y g %.7, n.l l w ~.

i e <<.+

p., %. [:$.Q..,.,. t..o.. ,,.,( )[h,&eyv.,W L ' 1< > y j

~

r .q ,py,. a m~s.s.;..,,

g. '...

,5t fm .,m n--. s,m ~

w. &en e a

m.,., - .r A,fgmp ;;&;m. 't z_ n., w gu ~. 34W. .Yf &; ~',N +fj'h & ** &< m.s,',~h ',0.? l}m4 '?. 7, .,.,. '...'k .+ %.k .- w - a. 2 3 ,7+' '* I,,.,, j.~..,..N'..h '* },**(n ,~ 7', 3 v ). .a w .?W 6,, h. dm. sv.,;.3 1 -. s :lc.-( 9 5 1 r-' C,)t g "g,.; C e.* 'N - .-(* ,Q. ,L a,,-g4 ,n . a } %.,+', y~,;7..v., hlp ~ .; y.lg. 2 :. ~ .,It- ... s. m ./ y,.O.. 3.j.,W

:.- i

. l(c) UNROLLED REGION s. I...,..,.. p/, - s,. . r-i.c...

,,r. 4 ml;>... 7_ ;,,..

to w - e. m ~;*' ?'ll .,*e ? l e N.ygf ..,% Th.;',k.: U '..h'<., ' j g . s.-L.4. %,*A. 6_.',1,- s. '. / h *.,,.'. .4 9 ,el .. ; \\ r *~ A. o' x. 4t?- s' ,,.y y.' .g, O P .. s,.\\. ~,. .4. g . i. g, ,, w. . m.. j ,. 4, 7,$ h..d'.'..'j h,.-s.'.b. '.[.. h ...~,. ,N. s - ~ ~ .t - e k*ide g/a. ,d.%. M.fg g w A e d AW,.., 6 =5. a FIGURE 33. PHOT 0MICR0 GRAPHS OF THE MICROSTRUCTURE OF THREE DIFFERENT REGIONS OF SPECIMEN B FROM TUBE A-146-6

[ 48 2e0 [ Loation 1 2g0 Loestion 5 Rolled Section Rolled Section 240 - 100 ) ISO l I 290 L'**"'" 2 Rolled section Locetion 6 260 Roll Transecon 240 r 220 L ) (

=

l 100 - j 100 1 200 f 1' 260 l***'i'" 3 Location 7 ) Rolled Section Unrolled Sectson dl 240 220 ( 200 180 160 l 290 260 Location 4 Location 8 Rolled Section Unrolled Section 240 220 ( m 180 0.004" 0.004" l gg I I 1 I I f f I I t t i e i 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 ID OD 10 ( POLTION ACROSS THE TUBE WALL TU8E A 146 6 FIGURE 34. MICR0 HARDNESS VALUES AT 8 DIFFERENT SUB-LOCATIONS ON SPECIMEN B f FROM TUBE A-146-6 i

= .llu . 1 I I 1 1 1 grgj 'P h*khO:hNd [ww#mw%ww$[j -_ s 1 C&kh.&g :G y *u *~ .h.b'N[jkes1.;f r*: 4i ' M^4 l In n., ,. ud.[e.-29. 7 - W.'u.. ~- .. s,q D M ~ l .Kyl' fy} 'CQ' ' J' 'L: *;F.3.'*y. - 'r.-s > O .d.; ' f.:, % ri,.;.- '. ..%..* c. h N kh.")s{.3l_~ h ;'.. ,' fe' 9 .n u. ,e. f :,' c:* ' M - ea. .c ;J,,, y. ,'. : +,.. 1 .LM '. w. y{ ^ '- y. m.. {,--t.. ~ 5 "","'k -[ f g,.. ,-. [ ,l.rk'EP s'

1. '@ y
g,y
- >: T.+ ' ' J: :H '... Jj p*. C4:!:Gi

<.: n .V

6 'df r,,s L L.r. M *i.:s.. f;' ', ~,%.

>t L B 100X l FIGURE 35. PHOTOMICROGRAPH OF THROUGH WALL IGC IN SPECIMEN C FROM TUBE A-146-6 I 1 1

L F L 50 F L L rl ( r { t y /. ,,+ t., i "N f. Y.. Y ( [ _'. -)- y._- s.. --ma

v.

. - 'r fe

. '..... m.

.p y1:. g 3c.J.g J . ~ ~ f..,, ^ :** % .G r. ',. W. .LQ,. ' n u :.c . x 135X FIGURE 36. SEM PHOTOMICROGRAPH OF A BROWN SPOT AND WHITE DEPOSIT ON f THE ID SURFACE OF SPECIMEN D FRr ! TUBE A-146-6 L f L I r L I

L rc 51 e !L I L r~ L b c i L% e .. #,~. r f ' ;i ',., p pp. e.=: i .s n r-w-

'},

b, ( j . ~ ~ f l h Y'5- .; =.g y p,.; 14 s,,_ .j u -

y x % y,.-s p / y y

. ? : ! J j'

%:'_AI, $fpf* J 's'hff'~:, (' .+-

,.s h'p: ' p +

g.- _,..,., -w u,. ~y,,f M' w, 140X FIGURE 37. SEM PHOTOGRAPH OF AN IGA PIT ON DESCALED SPECIMEN D FROM TUBE A-146-6 m r t k 1L

[ [. 52 The specimen was further cross sectioned very carefully (in the transverse direction with respect to the tube axis) through one of the large pits for metallographic examination. A photomicrograph of the cross sectioned pit is shown in Figure 38. A through-wall ICC running from the bottom of the IGA pit is clearly visible in Figure 38. { From the examination of specimen C and specimen D, it is fairly certain that the crusty brown deposits and the white deposits on the ID surface are corrosion products. These were either released from the IGA [ areas and solidiff ed on the surface, or they were formed af ter the re-leased cerrosion product reacted with the tube surface. A deposit similar to that shown in Figure 36 was cross sectioned and metallographically prepared for electron microprobe analysis. This { was specimen G from location 5.5-6.5 inch, which had no EC defect indication. A back-scattered electron image of the ID surface of the ( specimen is shown in Figure 39. A p1.t under the deposit is clearly visible. X-ray images of the specimen for elements Ti, S, Cr, Ni and Fe also are shcwn in Figure 38. The deposit inside the pit appears to be rich in Cr and Ti, but depleted in Ni and Fe with respect to the base metal. Sulfur is clearly present in all of the deposit. X-ray image of Cl was attempted, but no C1 was detected on the specimen. These results again indicate that some corrosion product (s), e.g., Fe released from the pit may have precipitated in the form of oxide (or hydrated oxide) giving rise to the brown color patterns seen on the ID surface. Some sur.! ace deposits, similar to that shown in Figure 36, were scraped from specimen I, location 5.25-5.5 inch, for XRD ' analysis. XRD patterns obtained indicated Fe, Ni, Cr, some unknown spinels and Ni 05 (pattern no. 30-865). No sulfur compound matched any of the 3 XRD patterns. I ESCA results for quantitative analysis of the surface deposits ( of specimens F2, Fil and E are given, respectively, in Tables 20, 21 and 22. The distribution of Fe, Ni and Cr on the three specimens at ( nearly the same sputtering depth were not significantly different. In all three cases, Fe, Ni and Cr increas,ed with the sputtering depth, as [ might be expected. L

L E L 53 [ lL, ,,m., - ? S- '- k. 'l.> ~m.1.,A' #,,,gp',,;,- ,s', +, s - b .~. )v'!M$3. %, s,;yi ) '{.*i,. /'. t, 7? ' $..,. <?i;, 51;. '. +. ' %,Ps, : V' +

c

--(. ;s,.:y~ ~b d;.::;. q'@*'

^';% " t2 h'if',4j,S'r',,.l.,*:,m '~ 9 WW, s.: :ys< * '$ 7lIy; m'%. g, -_~ LO,,c: % 4 - f ' [ y~a.y u;;:v., e,.

w..

Nf !Lfl~ OD V ID L R; m 5;& c, m..

w; ~

- A.f:?;pd",;T. v 54

f +I >.4?-84 g

100X r FIGURE 38. PHOTOMICROGRAPH OF THE CROSS SECTION OF A PIT IN SPECIMEN D FROM TUBE A-146-6

M M M M M M M M M M M M M M M M M M l i l $$g' g Qf.

.f_ ~ '

?" _?;' ^ c' 5-[.L[k].4,['-fj{l,p.,fq_, . ' i. '$ J.. M , Y ', _ g [. c.. ' - -,' y. y..p.yg. ; -.. s n; ' J , 4.> :

t. 1. u..g,.

.,=,. ?qg ? ,I IT .s~ W;..._ \\ -y. c es

s

.f.?.., \\ yl. k ~ .n... .a.-? <u. e E " d[<.a h. f Y,Wft ' ?. ] -,._ ~'h,, %. ?. y..h.,, _?, .U, r .., { % ...s - -u . = .y.- ,r, ..p v ,rg...!' z.,.. z '... f. i,.. ;.

5

- g u W .L . ""s .*a 4: qm 1.::.4..y., %r - a" ..e

~

2 .?. ;; ; ,.,.,t.. 6' ~ .X-.QfQ ,1;;y%, ,y. eASg

..,, h. *. _ N, *. -

.l'/ _ - ya, .g. .?., +.-r Ti S y . : ; ;'. ; f.. ' ' '. ' y :. '?', ' ; ' y.L[/;.n

M

?-- .';;.y M:: *i~'._#j '. ' D: q -,?_* ^l f$. ~ _ J$

s..

n +,+;;.,..,.,._n.+. ...~ .__ y 1 ~ ' ~ *

2

-.,.,.2 c, %,. ::1. - ?,,2 .v:. Y. l . % ? ?;, ;,.[ [.f (. ' l ~ Gi.. . ;;e,.y..,. ' t, eh i ',, V s. L _ iff2 'E($h d*de 6 Cr Ni Fe 700X FIGURE 39. BACK SCATTER ELECTRON IMAGE AND X-RAY IMAGES OF ELEMENTS Ti, S, Cr, Ni AND Fe OF A PIT IN SI'ECIMEN G FROM TilBE A-146-6 i l

(- 55 [ [ ( TABLE 20. ESCA ANALYSIS OF ID SURFACE OF SPECIMEN F2 FROM TUBE A146-6 Depth Sputtered, Surface Composition, atomic percent ,A Ni Fe Cr O C B Zr S None 5.5 3.7 4.1 43.3 39.5 3.8 0.2 30 7.6 6.6 5.1 39.7 36.5 4.2 0.2 530 10.9 7.6 6.3 35.3 33.9 4.9 0.2 0.9 1200 15.0 7.4 4.6 33.6 34.3 4.3 0.2 0.6 ( 2400 24.9 8.8 7.6 32.7 22.3 3.2 0.2 0.3 3400 24.3 8.1 8.4 33.4 19.9 4.9 0.3 0.7 [ [ [ i

c 56 [- rL TABLE 21 ESCA ANALYSIS OF ID SURFACE OF SPECIlfEN Fil FROM TUBE A146-6 Depth [ Sputtered. Surface Composition. atomic percent A Ni Fe Cr O C B Zr S I None 17.1 4.1 3.7 47.9 21.9 4.2 0.1 0.9 30 18.6 7.2 5.1 45.1 15.4 7.2 0.2 1.2 L 530 22.0 6.7 6.0 45.2 12.4 6.4 0.2 1.0 1130 25.2 8.5 6.7 39.8 11.5 6.6 0.2 1.5 2330 24.8 9.5 7.5 41.6 8.1 7.4 0.2 0.9 ( 3530 25.6 9.7 7.6 40.9 6.5 8.3 0.3 1.2 [ [ [ [ rL m F L a s

.I I 1 TABLE 22. ESCA ANALYSIS OF ID SURFACE OF = SPECIMEN E FROM TUBE A146-6 Depth Sputtered, Surface Composition, atomic percent A Ni Fe Cr O C B Zr S None 8.9 3.9 5.6 68.3 13.4 350 12.5 6.5 6.0 57.8 17.3 600 13.9 4.0 6.9 58.9 16.3 1100 27.7 8.2 6.4 44.6 9.9 3.9 1600 28.3 8.7 7.3 43.5 7.7 4.3 2100 26.I 8.2 7.0 44.2 8.4 6.2 3100 22.0 8.6 12.1 46.3 7.7 3.2 0.2 3 4 1 1 1 1 1 1

58 [ AES/ESCA analyses were performed on the ID surface of three different specimens, namely, F2, Fil and E. Locations of these in the above order were 0.5-1.0 inch, 5.0-5.5 inch and 8.25-8.75 inch. AES { was' used in conjunction with ESCA only to spot check the presence of various elements on the surface. Therefore no separate results were obtained with AES and none are reported here. Oxygen on specimen F2 at 30 I depth was 39.7 atom persent and marginally decreased with sputtering to 33.4 atom percent at 3400 I depth. Similarly, oxygen or specimen Fil at 30 I was 45.1 atom percent and 40.9 atom percent at 3530 I. On specimen E, oxygen was 57.8 atom percent at f 35 I and 46.3 atom percent at 31001. The order of oxygen concentration on the three specimens was the same, i.e., 30 to 60 atom percent. With sputtering, the total drop was minimal. The C concentration on three specimens decreased noticeably I with sputtering, see Tables 20 to 22. The concentration dropped from i 36.5 to 19.9 for F2, 15.4 to 6.5 for Fil and 17.3 to 7.7 atom percent for E. Specimen F2 had nearly twice as much C on the surface as on Fil or E. Boron concentration on all three specimens ranged between 3.2 and 8.3 atom percent. Similarly Zr was about 0.2 atom percent on the specimens. Sulfur on specimen F2 fluctuated between 0.9 and 0.7 atom percent, and on spacimen Fil between 1.5 and 0.9 atom percent. Sulfur on specimen E was detected by AES at the depths investigated but the concentration was not high enough (<0.1 atom percent) for detection by ESCA. I Binding energies of elenents and their chemical states as l determined by ESCA analysis are given in Tables 23 to 25 for specimens F2, Fil and E, respectively. The top layer, O to %301, of the surface in each case contained hydrated forms of oxides of Fe and Ni, i.e., Fe00H and Ni(OH)2 The most probable form of Cr on the surface was Cr 0 ' 23 l however, the possibility of Cr00H cannot be ruled out. With some sputtering, the prominent forms of Fe, Ni and Cr oxides on the surface were deter =ined to be Feo, Nio and Cr 0. Nickel 23 also was present in its elemental form. l

L [ 59 [ [ l ) p' TABLE 23. ESCA BINDING ENERGIES AND STATES OF ELEMENTS L ON SPECIMEN F2 FROM TUBE A146-6 Depth Sputtered, A Ni Fe Cr None 855.4 Ni(OH)2 710.8 FeOOH 576.4 Cr,0 or 3 CrDOH b 30 855.4 Ni(OH)2 710.8 FeOOH. 576.4 Cr2 3 0 { 530 852.2 Ni 709.9 FeO 576.6 Cr2 3 0 854.8 NiO 1200 ' 852.2 Ni 854.8 NiO 709.9 FeO 576.6 Cr2 3 0 ( i 2400 852.2 Ni l ( 854.8 NiO 709.9 FeO 576.6 Cr2 3 0 3400 852.2 Ni ( 854.8 NiO 709.9 FeO 576.6 Cr2 3 0 E [ [ E L 1

'I l 60 ) !I I TABLE 24. ESCA BINDING ENERGIES AND STATES OF ELEMENTS ON SPECIMEN Fil FROM TUBE A146-6 I Depth Sputtered, Patio A Ni Fe Cr C S S-2 so4 / None 856 Ni(OH)2 714 517 Cr2 3 285 C 168 SO -2 0 O 4 287 CH CH O 3 2 30 856 Ni(OH)2 711 FeOOH 577 Cr2 3 285 C 162 S-2 3,4 0 169 SO ~,- 4 520 852.2 Ni 710 FeO 577 Cr2 3 285 C 162 S-2 3 0 855.6 NiO 169 SO -4 4 } l130 852.3 Ni 710 FeO 577 Cr2 3 285 C 162 S-2 4 0 855.0 NiO 169 SO -2 4 2330 852.3 Ni 710 FeO 577 Cr2 3 285 C 162 S-2 > 10 0 855.0 Ni 3530 852.3 Ni 710 FeO 577 Cr2 3 162 S-2 y go 0 859.0 NiO i lI 1I ll L ~~ ~

L ( 61 ( [ l ( TABLE 25. ( ESCA BINDING ENERGIES AND STATES OF ELEMENTS ON SPECIMEN E FROM TUBE A146-6 ) ( Depth Sputtered, A Ni Fe Cr None 855.6 Ni(OH)2 711 FeOOH 576.6 Cr2 3 O { 35 852.1 Ni(OH)2 711 FeOOH 576.6 Cr2 3 0 600 852.1 Ni 854.6 NiO 710 FeO 576.7 Cr2 3 0 1100 852.1 Ni 854.6 NiO 710 FeO 576.3 Cr2 3 0 1600 852.1 Ni 854.6 NiO 710 FeO 576.1 Cr2 3 0 2100 852.1 Ni 854.6 NiO 710 FeO 576.1 Cr2 3 O 3100 852.1 Ni 854.6 NiO 710 FeO 576.1 Cr2 3 0 ( { ( {

1 I 62 The binding energy of C suggests that it is present as either graphitic carbon or bonded in long chain hydrocarbons. The S form on the top layer of specimen Fil was SO4, but immediately af ter some (301) sputtering sulfide, S", was also revealed. With further sputtering, the ratio of S" to SO " progressively increased, 4 see Table 24. The S forms on specimens F2 and E are thought to be the same as on Fil. Specimen F2 was examined by SIMS af ter the ESCA studies, i.e., I af ter ion sputtering to 3600 I depth. The SIMS data were taken while sputtering with 1p ampere argon ion beam. The ion mass spectra were investigated from two areas on the surface. The areas used were (1) an apparent shallow pit and (2) a bright metallic spot. The principal ions detected were Cr, Fe, Ni, Cr, A1, masses 69 and 70, with intermittent detection of B and S

2 Plots of ion intensity on a relative log scale versus approximate sputtering depth are shown in Figure 40 for the two areas 1

investigated. In the pitter area, there was little change in the intensities of mass 69, Fe, Ni, mass 70, Cr0 and Al up to the sputtered depth of %2pm. Sulfur, however, did go through a maximum at sl.5 pm and then decreased. In the metallic area, practically all ions showed a decreasing trend with sputtering up to Nipm. Mass 70, however, went through a minimum at NO.5pm depth bu*: then returned to the original value. I The masses 69 and 70 are ascribed to hydrocarbons with possible + species of cyclopentane C and its radical C 5 10 59' I# "" could be the source of carbon found in the ESCA spectra. All the examinations and their results described heretofore pertained to the investigation of the ID surface of tube A-146-6.

However, two specimens, I and J, were metallographically examined with special attention to the OD surface. Specimen I was from location 5.25-5.5 inch and J from location 1.0-1.5 inch. These two were selected for examination because they had deposits on the OD surface.

Metallographic examinations of the above two specimens showed no defects or ICA under the deposits or at any other locations in the cross sections examined.

[ { 63 [ V l I I I I I Pitted Area Metallic Area (Analyzed Area 1 x 1 mm) (Analyzed Area 1 x 1 mm) Cr i [ ( 103 [ M..s Fe Ni [ E Mass 69 3 Mass 70 u ( o Cr0 4, ( g 102 g ( AI I L S I 101 r ~ , si e Si L L 100 y I 0 0.5 1.0 1.5 2 0 0.5 1.0 Sputtered Depth, um e FIGURE 40. RELATIVE ION COUNT IN SIMS VERSUS APPROXIMATE SPUTTERED DEPTH ON THE ID SURFACE OF SPECIMEN F2 FROM TUBE A-146-6 m.

[' 64 [ 4.6 Tube A-146-8 Short-pull tube A-146-8 was examined using the following techniques: (1) Visual and photographic examinations ( (2) Microstructure and microhardness measurements (3) EPR sensitization test I { (4) SEM and EDAX examinations of ID surface and a fracture fact (5) Meta 11ographic examinations. Results of all examinations are summarized in Table 26 using the same { format as in Table 3. Details of these results are presented below. The whole segment of tube A-146-8 was slit into two halves. Their ID surfaces were examined visually and were photographed. The photographs are shown in Figure 41. The features described for tube A-146-6 also were observed on tube A-146-8, namely, dull color surface, deposits on surface, and brown decorations. Some scored areas also are visible in the photograph. ( Specimen B from location 0.5-1.5 inch was used for examination of the microstructure and making microhardneus measurements. Photomicrographs of the etched specimen from three different regions of the tube, namely, rolled region, roll transition region, and the unrolled region are shown in Figure 42. The grain size in all ( three regions is the same; similarly, the carbide decoration of grains and grain boundaries. No continuous networks of carbides were observed. [ The microstructure of this tube is identical to that of tubes A-71-126 (Figure 31) and A-146-6 (Figure 33 ). The microhardness measurements on specimen B were done in the same way as on tube A-146-6. Results for tube A-146-8 are shown in Figure 43. The maximum value,DPH 241, was obtained on the ID surface in L the rolled region, and the minimum value, DPH 170, in the mid wall section of the roll transition region. These values again suggest that the tube r was not excecsively cold worked in any region. L I

M M M M M M M M M M M M M M M M M M' TABLE 26. EXAMINATION RESULTS OF TUBE A146-8 l Specimen location Number Inches Degrees Type ofIndication Examination Result /Conunent A 04.5 0 180 Radiograph, Visual langitudinr.I Met. U Shape IGC B 03-1.5 0 180 Radiograph,1.0 p Structure Discrete Inter. + Intragranular Carbide ppt y liardness No IGA on ID or OD, DPil 24i Max./183 Min. C 0 4.25 I80-360 Visual U Shape Transverse Met. 3 IGC,~ 100% Walt D 0.75 1.25 180-360 None EPR Sensitization Not Used, Substituted by L El 33 4.5 90-180 EC 3.75, Tree Decoration Longitudinal Met. IGC,70% Wall, Microstructure As B No ID or OD IGA E2 3.54.5 0-90 EC 3.75, Tree Decoration SEM/EDAX IGC, Fluffy Deposit on Fracture Face S and Tl F 334.25 180-360 EC 3.75 Tree Decoration Transverse Met. No Defect Detected;No lGA ID or OD G 85-9.0 0-180 None Longitudinal Met. AGA 0.004 in. Deep /0.015 in. Wi le at ID II 85-9.0 180 360 None Transverse Met. No Defect I 6.75-7.25 0-180 Visual Pits SEM/EDAX Pats-IGA, ID Crust, S (7E%), Ti (2.7%), Ca (0.1%) 1 10.5-11.C 180-360 Visual Scratcht longitudinal Met. No Defect K 5.75-6.25 0-180 Visual Tree Decoration SIMS Not Used L 0.254.75 180 360 None EPR Sensitization Activation Potential,125 mV (SCE) M l.5-2.25 0-180 OD Deposit I.ongitudinal Met. No Defect Under Deposit, No IGA ID or OD N 3.0-3.5 0-180 OD Deposit Transverse Met. No Defect Under Deposit,No IGA ID or OD

L [ 66 l WW [ [ [ [ [ [ [ FIGURE 41. PHOTOGRAPHS OF ID SURFACES OF TWO HALVES OF TUBE A-146-8; A) 45' B) 135' C) 225* AND, D) 315' (SEE NEXT PAGE) [ [ [ [ [ [ [ c L

m =.

e

. DOCUMENT

e. T' PAGE

~ PULLED ANO.Amm3 NO. OF PAGES I REASON O PAGE 4, LEGS 2 D e cewraD A1. von er ote 3 J_ D DE1TER CCW REQUESTED ON _ t hPAGEICC LARGE10 FEM. JimcceraDe conoie @) I b\\hkDOO%-O3 l / \\ D.FEMED ON APERTURE CNtD NO \\

( 67 [ u.y Q. -.._ n...~~7,. y. ,.. c...y... ;.. _x.., y ' -. t. ' ontl '? J. g-wa

,e vm~

u. \\ lf., eJ [ n .e ,,1

n. ;,a.

.. N1.,',j

)dl L W

,y g} f p ' > > :. -L - b r y: 1 g xp y, 3 [

g. ' ' ~ -

y e;. ., [? ajfN'., l,%l **:.[r ,. [/ 1.# i. a c' y .m.;i.. ;.:.s < : n'sy > ;u ; y ~ ( ~y a, 4 ?.,v. A g..:>"5h

uk[.;.,'

j.d h 6;/ + + ~.

sW. m ,0 lj,"," JM

r r

Q 1 ] k,,"Y,O, h.\\ w..af, : +J,.Y.P f:r,i.f R,; g' ?.- a n ;;;* G. N..

y. 'p < c

,4. p,#- ,.3 w f +. p g. gn a;u,.4, ?.w. <, m,:. tm,3..y. ;.<,.;c W;; ;7;a,

s n, (a) ROLL TRANSITION REGION

[ r :..a =~ j..-@. ;. n :. s.'. :s.-lyQ.h *~ ? .'W ,.. ~. ' '

e

.u

1.,-u.a-

.s. .~ [ ~

Q@

- i.. b ' f,;, &.,c (;;*y ';f

Q

~- y

. fer *

..3g 2

%. 3 - s.a.

C-a . :N ** -e.. s..l . L ,. = W., a,..,, -. s'. - (*G,,., f. -,o i ~* r. [ y .% f . '.... E - t/@. ',i ( ~*. f>' . le j ~*

g,1

..'g* ..7

. y. g",'

m. ,;9 _ g y. n .. &,:e ? :w., ...** i

r; s.:f*:

x. . ~.... ' ., fy ;, '. I;. p* ;.;. 4.- ~ [ ID ,',e ' r'..n ' ~-

oe.'_.? A r:. :

n. n - c.x - ... x. a,';.i. L: R p - .c e... x ,=. t'i:,*. 9~; ". ;l. *: = M.. :. V. [ K.i..,4.. ' :.+ n+px m.,: 3: x.... w.: c.c. . 4 m s.,.... ~p. .c.. .O,....\\.-. '<y\\.% f[q,k.) [ ,. j 1, b s..... y.. t ~ ?

. m ;q-m ca..

a . R: - + '.,, ,_.,l..- u.:

o. "__. n.._._g._

.? . r. [ ,r. 500X H P04 Etch 3 { (b) UNROLLED REGION FIGURE 42. PHOTOMICROGRAPHS OF THE MICROSTRUCTURE OF T'*O DIFFERENT REGIONS OF SPECIMEN B FROM TUBE A-146 I

-I 68 280 l 260 Locaten 1 Locaten 5 Rolled Section Rol:ed Section 240 220 200 180 I t t t t t I t I I i t I i 160 280 I 260 Location 2 Location 6 Rolled Section Roll Transiten 1. 18 g l 160 l I I I I i e t Z s 1 h 280 5 3 260 Location 3 Location 7 h Rolled Section Unrolled Section t a 240 l 220 a= 200 180 160 f f I f f I i I I I I t 280 260 Locaten 4 Location 8 Rolled Sect on Unrolled Section I 240 - s 220 _f 200 l 180 - O.004" 0.004" e I I I I I I I I 160 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 90 ID OD 10 POSITION ACROSS THE TUBE WALL TUBE A 146 8 FIGURE 43. }!ICR0 HARDNESS VALUES AT 8 DIFFERENT SUB-LOCATIONS ON SPECIltEN B FROM TUBE A-146-8 (SEE TEXT) ~q V s amm

L [ 69 The EPR sensitization test was done on specimen L from location 0.25-0.75 inch. The activation peak potential for this tube was 125 mV(SCE), which indicates a very high susceptibility to IGA by poly- { thionic acid. This value is not significantly different from the 110 mV value obtained for other tubes. A defect with EC indication at 3.75 inch was selected for SEM/EDAX and metallographic examinations. A circumferential crack at 3.75 inch location was visible to the unaided eyes. Some brown color decorations around the crack also were present. Two specimens containing the defect were prepared from location 3.5-4.5 inch, specimen El (90-180 degree) for metallographic examination, and specimen E2 (0-90 degree) for SEM/EDAX analysis. The general appearance of the ID surface of the tube in the brown decoration area is simwa in the SEM photograph in Figure 44 The surface is rough and is covered with granular and flaky deposits. Another spot from a similar area showed crusty deposits on the surface, as shown in the SEM photograph in Figure 45. ( EDAX analysis of the spot in Figure 45 detectedonlyNi,N,' Fe, T1 and S on the surface. The S was present in about 0.2 weight { percent concentration. Approximate concentrations of other elements on the spot in atom percent were: Ni-44; Cr-25; Fe-20; and Ti-0.9. In relation to the composition of base metal, i.e., Inconel 600# the spot appears to be rich in Fe and Cr but depleted in N1. Iron in the form of oxide may have produced the brown coloration on the surface. The apparent crack in specimen E2 was mechanically pulled apart, and the fracture surface was examined in the SEM. A SEM photograph of the fractured surface in Figure 46 shows the intergranular nature of the crack. A photomicrograf of the same crack present in metallographic specimen El is shown in Figure 47. The crack is intergranular and penetration is N70 percent through wall. No IGA was noticed on the ID or OD surface of the tube in the vicinity of the crack. ( A transverse cross section of the tube, specimen F 3.5-4.25 inch, 180-360 degree, showed no IGA attack on the ID or OD when examined metallographically. m l L

l I a I I 1 3 1 l .I u l bh I w g "? I <. s..g e _g ]ff.yf, t ,s. M 1300X FIGURE 44 SEM PHOTOGRAPH OF A BROWN DECORATION ON THE ID SURFACE OF SPECIMEN E2 FROM TUBE A-146-8 I 1 l 1

71 ~~ ME R .1 - j. 1 ,S ? /* ' ? d

/ -

T$ ~. g'gf .p ,.~ }. gl x.'. . x_y r pr p j. ~. e _g * 's.. 8 '*e .*'i e sw

9; y

,rs, 9,9. ~Q.*- .,c '. I, [ y, , 1 J 9 ';.*f k s = ' 'WA, _a 's.h 1300X a FIGURE 45. SEM PHOTOGRAPH OF A CRUSTY DEFOSIT IN A BROWN SPOT ON THE ID SURFACE OF SPECIMEN El FROM TUBE A-146-8 L j lu Y N

72 i I 4 e i I 1 200X FIGURE 46. SEM PHOTOGRAPH OF THE FRACTURE SURFACE OF SPECIMEN E2 FROM TUBE A-146-8

B l n B B B B l .D: l.Q'. . -y <. ., y '* j } "^

O.,.

Q,. * * *, 1 +

- N,J w
. 4.

et, ' y ;,.>. 5' i n. ; .r

p. -

s. g-*,,-~,,r,,;*,- 2 *' h, s '.' ' m.~ . " _,, L '/ 0, s. 8 '.- Je ** y _, ~,# " I ^ f*w ;1 '.;2 7' g' '7, [ *y '.'p. 4. ~ $' ]. <

4 L.

- g.Q.?' - I '* A ..... -. K'.*c ' {S-C" '.g, + . +, N, g g, , ' $3 % *-.. ~ ^.', ' ', ' * - y 4 ,g wN.s e'r%$p ..f ,'N$Y ... - - h* %.4,,,

A

.' ', % y. h ;.Q.* f:!h,s a,, '. g 7 9

,,,s'

,"*,A,,' '~="? ; ;L ~. g.,*- y*fiy _} E

,;y ~: - % lm[..

~ ~ '

b f %, ~-{.jpZ,CgQ-R*

~' _.W <,.- m n .4. ~.' , < w.f,,-> n~y':R.? - s s- < g g,.. x, I .s. F: -v ._3,,y, f r* 4A ~ _. n

g~

t', IE ..~ {~ .',I*",,,e .G. r ; Q$. L' I#*

* ^

j. I , ~_ ' = , ? s - " i'r f* 'O, e. -a ,c "p, t e-v. e

n. ',,

g D 4 g M A a, mz .a

(.

h/ E g, ,- = .~

..s.

, / A. .(* %.m o g'.' ,e, / + a.,,. I ~-. v.-. -. > l n ?..~.+. i g P i ..s.- - L :n -W. t- ** ? A*~' ID 100X B B FIGURE 47 PHOTOMICROGRAPH OF AN IGC IN SPECIMEN El FROM TUBE A-146-8 I I E I

74 I Specimen I was taken from location 6.75-7.25 inch for another SEM and EDAX examination. There was no defect indication from NDE for this location, but a shallow pit was visible on the ID surface. ASEM photograph of the pit is shown in Figure 48. Encrustation around the pit is similar to that shown in Figure 45. EDAX analysis of the surface found, in approximate atom percent, elements: Ni-44, Cr-41, Fe-12, T1-0. 8, S-0. 8, and Si and Ca in small Another analysis of the inside surface of the pit found the amounts. deposits to be rich in Cr (76 percent) and low in Fe (-G percent) and Ni (~17 percent). Other elements T1, S, Ca and Si were also present in small amounts. At one area inside the pit very high sulfur (7.8 percent) was detected, Ca (0.1 percent) and Ti (2.7 percent) also were present in the area. The descaled pit is shown in Figure 49. Intergranular attack is clearly visible in the SEM photograph. Specimen G, location 8.5-9.0 inch, with no apparent defect on the ID surface and no NDE indication was examined metallographically. IGA was observad on the ID surface of the tube, as shown in Figure 50. The IGA is approximately 0.004 inch deep and 0.015 inch wide at the ID surface. A lip-crack from location 0-0.5 inch was metallographically examined. Two specimens A and C from the 0-0.5 inch location were prepared for tne' examination. Specimen A, 0-180 degree, was examined in longitudinal cross section, whereas specimen C, 180-360 degrees, was examined in transverse cross section with re pect to the tube axis. Micrographs of specimens A and C are shown in Figures 51 and 52, respectively. In specimen A, the tube wall separated at the main crack while preparing the metallogre.phic mount, the separated wall is visible in Figure 51. A second crack which appears to be a major branch of the main crack also is nearly through wall. The transverse section in specimen C thewed three different unconnected cracks. One such crack only is shown in Fidure 52; the other cracks were similar in nature. Two specimens H and J from locations 8.5-9.0 inch and 10.5-11.0 inch showed no IGA attack on any surface ID or OD when examined =etal-lographically. There were no NDE defect indication at the above locations. = _...

[ 75 I e L I .,. sw. T $. hf T ~ g ~ i ^ s s, 210X ^ FIGURE 48. SEM PHOTOGRAPH OF A SHALLOW PIT ON THE ID SURFACE OF SPECIMEN I FROM TUBE A-146-8 F L F ? L ~

l 'r 76 I 1 { F 'JT. . Aw; l (Aw '.\\.. 4.e&y$:gl ' _. ' _ ;. '.7 ? : ';j'. 0. ' q '. _ .y~:.' 7 '.,.

o,

..y ,i.. Wf, '? -. :J- '. - r' '.'.. j ., '..f .. Y c 4,,c p Q --= ja!g ;%,- ~;, ;, ' [ ^' , *....Q' s =. p; s j m. h .',rj.,; p j. _*,p e .4,-

f. -

" '. - 14 , 4' . p,', $/ ' %. F ' s-i r*,Y*,. I4 A ~.' J' 'a g 1 4 --...,- 9., .- u. '. '.4 4_'.l,% W, -..,... .;- / - .n' f.p,y. e ' _.j " .. '., )h,,rc ~ 'N*.,* n. l ..,."a.-, g> j ' s a n, .n. J.. y: ~ p c ,x 3 .. N;. g.;G_.?f ...q. 4 'l [.- .', q _ [ N I 2_ [.. gf i ~

t. ; g ; ~_.".;g;f
g y: .: ( e -

.,. g' _ ',. >. -: g., o, '.t.4 a f x ns - 4 _ f,. Y ..... + h _ {&.' }.,,pl,.**qif- ,- p _. ',J4 : ..~ r - n...,, s : g, .h c . R. %;...,... l 'st... :, ..* s' f.A no 4 'W '-- h. ' , " W* l 1 p. m e. , $l ~ ' ' l, ,'f l.. ~ f ' %* ' ' hf ._.b.'. I'; ~. ,.A+ -. g-9,/- N y' p y ;;. !c. 4.~ - 7....r 1 L 700X ) FIGURE 49. SEM PHOTOCRAPH OF A DESCALED PIT ON THE ID SURFACF OF SPECLMEN I FROM THE TUBE A-146-8 t 5 J s

I I 77 5 I I I m ' ' e. g .g B L 2[f I ~ g ,u g ID 100X OD I FIGURE 50. PHOTOMICROGRAPH OF AN ICA ON SPECIMEN G B FROM TUBE A-146-8 I 5 I I

78 c \\ ^ 5 1. ) s ?P5M"M" J# dies """" "" "" " [ l

4 k [ 79 ( 4_. OD EN +' r [ '3 f* 3 ~ _- . ~.%. T4, >V ' ID

Q ll by&, ; h; $ '; 'll 100X

[ } FIGURE 52. PHOTOMICROGRAPH OF AN ICC IN TRANSVERSE CROSS SECTION OF SPECIMEN C FROM TUBE A-146-8 [

1 1 80 3 Two more specimens, M and N, from locations 1.5-2.25 inch and 3.0-3.5 inch showed no defects on either the ID or CD surface in metal-lographic examinations. These specimens were particularly selected for OD examination since deposits were present on the OD surface. A similar result was obtained for tube A-146-6. It is reasonable to conclude that there are no defects on the OD side of tube A-146-6 or A-146-8. I i 1 1 I 1 1 1 1 I i 1 l

5.0 RESULTS OF Y-RAY ISOTOPIC ANALYSIS Five wipe samplas from GPU-Nuclear were received for Y-ray isotopic analysis. Results of the analysis of each wipe sample are summarized in Table 27. Fission products detected on various samples 241 57 125 l06 110 were Am g 33 , Ru , gg , Cs13', 254 60 and Co The Co l was the main source of Y-rays in these samples. "he next most prevalent ' were detected on only one I isotope was Ru Americium and Cs wipe sample which was from OTSG-B. .+ 4 s I 's 1

==

unies

=m

N TABLE 27. CAMMA RAY ISOTOPIC ANALYSIS RESULTS ISOTOPE PERCENT I SAMPLE IDENTIFICATION AU( AU( AL( BU BL R71T126 R149T15 Batch #66 R75T3 Batch #66 Isotope Sample #31 Sample #6 Sample #20 Sample #13 Sample #28 241 Am 0.1 y 57 Co ' O.2 0.3 0.2 0.2 0.2 Sb 3.1 3.5 3.5 0.4 Ru 5.9 7.4 4.4 8.4 4.9 Ag 0.2 0.7 0.3 0.8 0.5 Cs 0.2 Mn 1.3 1.3 1.6 1.0 0.9 60 Co 89.3 86.8 93.5 86.1 92.8 .r, Notes: (1) Ilhole sample analyzed; (2) %1/5th sample analyzed; (3) %1/5th sample analyzed; (4) Whole sample analyzed; (5) %1/5th sample analyzed I'

b cL FL 6.0 DISCUSSION The results of various examinations are discussed here. The aspects considered in the discussion are a) nature of defect, b) defect location vs. NDE, c) physical and chemical properties of OTSG tubes, d) surface film composition and finally,e) probable cause of attack. Nature of Defects Tubes A-146-6 and A-146-8 had typical dryout deposits on their L D surfaces. Four specimens, two from each tube, having these deposits, were metallographically examined in longitudinal and transverse cross section. The two specimens from tube A-146-6 were from just below the roll transition region, whereas those from tube A-146-8 were from the lower part of the tube segment. No IGA or other defects were cbserved on these specimens. The OD surfaces of the above two tubes also were visually examined af ter brushing off the deposits at a few places. The surface underneath the deposit had a metallic luster, but no corrosion attack or significant etching was observed. The OD surfaces of these tubes were not examined in the SEM. However, on the basis of visual and metallographic results it is assumed that no OD defects exist under dryout deposits on tubes A-146-6 and A-146-8. I Since tubes A-146-6 and A-146-8 had typical dryout marks, it is L likely that other tubes also do not have OD defects. From the extensive metallurgical and SEM examinations of various specimens from different tubes, it is f airly obvious that the main mode of attack in the tubes is IGA. Further, the attack initiated on the ID surface of tubes. This attack produced three different kinds of defect geometry in the tube walls: a) ICA-islands, with grains retained, b) ICA-pits and, c) stress assisted deep inurgranular penetrations which may be called g L intergranular stress corrosion cracks (IGSCC). Eu L

L [ 2 One prominent ICA island 0.004 inch deep and 0.015 inch wide on the ID surface of tube A-146-8 is shown in Figure 50. But some minor IGA areas were also seen in the SEM on a specimen from tube A-146-6. F L It is very likely that similar other islands are present on other tubes, particularly under ID deposits. ICA-pits ranged in size from a few grains deep and a few grains wide, as in Figure 37 to %0.014 inch deep as in Figure 38. ICA-pits are a result of grain dropping from the heavily attacked IGA-islands. Crains L from attacked areas dropped off either because the grain boundaries were heavily attacked and no cementing bond existed between the grains, or L because of the force of voluminous corrosion products senerated at the grain boundaries, or both. In some cases the voluminous corrosion product filled the entire pit; an example of this is in Figure 39. The cracks, i.e., IGSCC, were circumferential and spanned from p 1/8 to 3/4 of the circumference of a tube. The IGSCC penetration in i various tube sections examined ranged between 20 and 100 percent through wall. Some of the cracks were vide open, e.g., in tube B-ll-23, whereas others had to be opened up mechanically as in the case of B-8-25. The IGSCC was produced in areas of IGA where stresses were high [ locally. A striking example of this is shown in Figure 38, in which the origin of the crack is clearly at the bde of the pit. The pit in this s case apparently had acted as a stress concentrator in the tube, and subsequently, high stresses opened the attack grain boundaries thus forming the crack. FL Another example of IGA leading to IGSCC is the crack in tube B-11-23, Figure 25. The IGA was found three to four grains deep on the ID surface of the tube, NO.1 inch on either side of the main crack, Figure 26. Most of the. cracks examined had some branching. The IGA also was present along sides of crack walls, but the extant of IGA varied considerably from one crack to the next. Crack wall ICA was only one to two grains deep in some cases, see Figure 47, but seven to ten grains deep in others, L Figure 25. Meandering of the crack or IGA in a plane perpendicular to the fracture surface is fairly evident in Figure 19. The deep cavities visible IL e L F

B 3 in the mid wall are indicative of the spiraling. A similar conclusion can be drawn from Figure 52, which shows branched cracks present on a transverse section of tube A-146-8. The three different kinds of defect geometries, mentioned earlier, mainly reflect differsnt degrees of ICA at localized areas. Several factors related to mechanical, environmental and metallurgical I conditions, e.g., localized stress, uneven distribution of aggressive chemicals (e.g., sulfur oxyanions and oxygen) on the tube surface, localized differences in degree of sensitization, uneven oxide film, surface deposits, inclusions, and manufacturint defects, etc., may have been responsible for the different geometries. Considerable amount of experimental work will be required to pin-point the actual cause. There is, however, some indication in the literature

that the rate of IGA, in the case of polythionic acid corrosion of sensitized Type 304, is strongly stress dependent. The rate of IGA on sensitized I

Type 304 is substantially low in the cbsence of stress, but very high in the presence of applied stress. The actual races were not given by the authors, but the unstressed specimens showed only minor IGA when they were exposed for 11 days in polythionic acid; on the other hand, the specimens stressed to 95 percent of their yield failed in a few hours. If a similar condition is believed to apply to IGA in sensitized Inconel 6008 tubes, the local variation in residual stresses could have I been one principal cause of the different degree of attack observed in various parts of the same tube. But contributions from other factors, such as oxygen distribution, etc., cannot be completely ignored. The possibility of tube failure due to corrosion fatigue was ruled out from the results of the TEM examination. No fatigue striations were observed on grain feces of the fracture examined, Figure 19. Some mechanical defects also were found on the ID surface of several tubes. However, these were considered to be the results of either

I. Matsushina, " Electrochemical Characteristics of Polythionic Acid I

Corrosion Cracking," Proceedings of the 6th International Congress on Metallic Corrosion, Sydney, Australia (1975). E 8

B I the tube pulling operation or were manufacturing defects. The heavily scored ID surface of tube A-71-126 on metallographic examination showed mechanical indentations on the wall. The presence of deposits inside indentations suggest that the defects were caused by tube manufacturing I processes and the deposits formed after the exposure of tubes to aqueous environment in the OTSG. Physical and Chemical Properties'of Tubes. The chemical compo-sition, tensile properties, microhardness, grain size, and the micro-structure with respect to carbide precipitates of tubes are those of a normal cold drawn stress relieved Inconel 6008 tube. The grain size (ASTM No 7 - No. 8) was uniform at different locations examined for I No continuous network of carbides was observed in any several tubes. tube, thus, indicating no severe sensitization. However, the EPR test showed that the tubes are in a heat treated condition which is very susceptible to polythionic acid intergranular attack. The EPR peak potentials of tubes were between 110 and 125 cN (SCE). During the EPR test, tubes were, in fact, heavily attacked intergranularly. Surface Film Composition The major foreign elements detected on the fracture surface of any specimen were C, S and C1, not considering oxygen. The distribution of S on the surface was quite non-uniform, as shown by all the microanalytical techniques, i.e., EDAX, AES and ESCA. The concentration ranged from practically nothing (i.e., below detection limit) to almost 8 atom percent. Chlorine was detected only by ESCA at sl atem percent levelupto2300I depth analyzed. Sulfur in the attacked area was found all the way down to I the base metal as indicated by X-ray images of a shallow pit, see Figure 39. No chlorine was detected by this technique. The concentration of carbon on the fracture surface was very high >50 atom percent and that of oxygen comparatively low, 18 atom percent. B E I

I 5 Chemical states of the major elements were Ni as elemental Ni or tied to sulfur, Fe as Fe0 and Cr as Cr.,0. The S was P. tes reduced 3 ~2 state as S The very low level of oxygen on the fracture surface agrees with the reduced states of Ni and S. Carbon was as in graphitic carbon or long-chain hydrocarbon. Major constituents of the ID surface film on tube A-146-6 were Ni, Fe, Cr, O and C. Small amounts of B, S and Zr also were present. The same constituents also were found at location 52.0 inch on the long-pull tube A-71-126. Nominal concentrations of the elements as determined by ESCA up to a depth of 3500 I on both tubes were (in atom percent) Ni 25, Fe 8, Cr 9 B 3, S 1, and Zr 0.2. No significant difference was observed in element concentrations because of the location of specimens with respect to the tube sheet. Sulfur was nominally at 1 atom percent. SIMS analysis of one specimen showed that S on the ID surface was I persistent up to at least 1.5 um depth. The topmost layer (%1100 I) of the ID surface film was some-what oxidized (e.g., S as SO "; Ni as NiO) as indicated by ESCA. But 4 thereaf ter, reduced S as S was the prevalent form. Nickel, Fe and Cr at this level were as Ni and NiO, Fe0 and Cr 0. 23 These analyses show that there is no significant difference g between the states of elements on the fracture surface and the lower B layers of the ID surface. The one noticeable difference is that C on the fracture surface is at twice the concentration of that on the ID surface. The oxygen level is also two to three times higher on the ID surface, but this could be simply because the ID surface was exposed to ambient environment for a long time, whereas the ICA-affected areas were protected by corrosion products. The presence of high carbon in the ID surface film and in the fractured surface film is disconcerting. The SIMS analysis of ID surface, g Figure 40, showed that C is persistent down to %2.0 pm depth analyzed. g The fracture surface film was analyzed with AES/ESCA to only %2300 I (0.23 um) depth, but it is likely that C was present at greater depths. In both cases, the chemical form of C determined by ESCA was the same, 1 1

L p 6 ( the binding energy for C atoms was as in graphitic-carbon or long ( chain hydrocarbons. It is reasonable to conclude that the C was deposited on both the surfaces from the same source. 'the tube specimens for surface analyses was sectioned using a hand operated jeweler's saw (hack saw). No lubricants were used in the operation. Sectioned specimens were stored immediately in clear plastic vials. The possibility of specimen contamination with carbon during sectioning and handling operations was therefore low. ( However, minor contamination of specimens from air exposure and also inside the vacuum chamber (from residual vacuum pump oil vapors) of instruments is a definite possibility. Carbon from such contamination is often detected on specimens in AES/ESCA analyses. But this type of contamination is usually limited to the uppermost layer (50 to 100 A) of the surface film on specimens, if the film is non-porous. The contamina-tion is easily removed by argon ion sputtering. ( During the AES/ESCA analyses of OTSG tubes, the specimen holder, which was made of copper, was checked in a few instances for surface contamination. Carbon was detected on the copper holder prior to any argon ion sputterit.;;, butafterabout501ofsputtering,Cwasvirtually all removed. [ In the case of specimens with porous surface film, which probably was the case with OTSG tubes, the C contamination could have been deeper, and therefore, not easily cleaned off by argon ion sputtering. It is likely, therefore, that the total C measured on the tube surfaces had some { contribution from contamination in the instrument. The exact contribution to C analyzed is not determinable, but it is considered to be a small fraction of the total. The Preliminary Failure Analysis Report issued by GPU-N indicates that some oil may have been accidently introduced in the reactor coolant during the plant layn.p in March, 1979. It is likely that some oil migrated to the steam generators, adhered on the tube walls and permeated the surface { film. The chemical form of C in the oil would easily account for that determined by ESCA on various tube specimens. [ ~

I I 7 If the oil was not removed from the OTSG system, it is likely that the oil survived the hot functional and it was present in the system during the subsequent layup. If the cracking of tubes occurred (either before or af ter the hot functional) when the oil was present in the system, I indeed carbon could have deposited on the fracture surfaces. The fracture surfaces (Figure 21) analyzed by AES/ESCA were from a crack which had penetrated s90 percent of the tube 5:all. Therefore, it can be presumed that the crack walls were sufficiently open, under the tensile stress present during the cold layup, to allow entry of oil into the crack. I Defect Location vs NDE I Defects confirmed by destructive examination of numerous specimens from different tubes always corresponded to EC observations, except in two In tube A-146-6 at location 6.0 inch, a small pit was observed, cases. Figure 39, which was not detected by EC at Battelle, and the CPU-Nuclear EC data file also does not show any indication for this position. Similarly, no EC indications were given by either Battelle or GPU-Nuclear for the IGA defect, Figure 50, in tube A-146-8 at location 8.75 inch. The first undetected defect, i.e., the pit was rather small, 0.003 $nch in diameter and 0.001 inch deep, but the ICA was 10 percent through wall. Sizes of the above defects are almost below the detection limit of some of the EC probes. Defect indications obtained from radiographs were not always con-firmed as IGSCC or IGA. These indications were most probably from the I scoring of ID surface during tube pulling operation. Radiographs also were not able to detect very tight cracks. A majority of confirmed defects in tubes were located in the roll transition area. The main reason for the preponderance of cracks in that area appears to be high concentration of stress, which assisted in propagation of IGA as discussed earlier. I I I

[ 8 [ Probable Cause of Attack and Failure Scenario [ Considerable amount of sulfur was detected on the ID surface of tubes, inside pits, and on fracture surfaces. Sodium thiosulfate and sulfuric acid were accidently introduced several times into the reactor cooling system between 1980 and 1981 during the plant layup. Undoubtedly, ( these were the source of sulfur on the tubes. Sulfur compounds, such as thiosulfate and polythionic acids are known* to produce IGSCC in sensitized Inconel 6008 alloy in the presence of oxygen in the environment. The heat treatment of tubes was such that they were extremely susceptible to 1 IGSCC by polythionic acid, as indicated by the EPR results. Therefore, the most probable cause of ICA on the OTSG tubes appear to be sulfur compounds. ( The exact period in which the attack occurred, between 1980 and the time when the defects in tubes were first suspected, is uncertain. { The OTSGs were brought to hot-functional status in August-September,1981, and at this time no leaks were detected nor any defects were suspected. [ Leaks in tubes were detected the first time during the hydrotest in November, 1981, and subsequent EC examination indicated defects in a large number of tubes in both OTSGs A and B. The absence of any leak during the ( hot-functional operation strongly suggests that the defects in the tubes developed moscly afterwards. ( The sequence of events leading to post hot-functional IGA of tubes is postulated to be as follows: (1) The environment inside the OTSGs during the layup period prior to hot-functional was devoid of oxygen but contamina-ted with sulfur compounds (i.e., thiosulfate acd H SO ). Any trace amount 2 4 [ of oxygen in the system, if present at all, was probably consumed by some of the thiosulfate (say, Na 3 0223+HO+02 "2 04 + H SO ). Because l ~ 2 2 4 of the absence of oxygen, the Inconel 600@ tubes did not suffer the ICA ( to any significant extent during this period. (2) The sulfur detected on [

Karl Sieradzki, et al., Paper No. 224, Corrosion /82, Houston (1982)

[

k L 9 on the tubes is primarily in the form of sulfide (NiS or N1 8 ). The 23 sulfur compounds present in the OTSGs were reduced to either sulfur or sulfide at high temperature in the presence of hydrogen during the hot-functional. The reduced compounds ultimately reacted with the tube surface producing the sulfide film. (3) Following the hot-functional, the OTSGs were partly drained for plant maintenance. The water level was kept near the upper tube sheet, but fluctuated by several inches be-tween September and November of 1981. The atmosphere inside the channel head during this period was primarily air. It is believed that the nickel sulfide reacted with the oxygenated water and produced the polythionic acid (e.g., 8NiS + 2H O + 1102 = 4N1 02 3 + 2H 8 0 ), by a mechanism similar 2 246 to that proposed originally by Brophy* and later confirmed by Ahmad et al.* The polythionic acid attacked the susceptible Inconel 6003 and produced the IGA. The attack was primarily limited to those regions of tubes where the stress concentration was high and the oxygen was readily available. The above two conditions vera adequately present at the water / air interface in the upper tube sheet region, particularly at the rolled section of tubes. This explains the preponderance of defects in tubes in the upper tube sheet region. (4) Several elements, primarily fission products and carbon, ( beside sulfur were detected on the tube surface as well as on fracture sur-faces. Whether these elements played any role in the attack mechanism can not be said with certainty. However, in view of the strong attack by poly-thionic acid, the role of other elements is considered to be nominal, if any.

S. Ahmad et al., Corrosion. 38, 347 (1982).

[ rt r L

[ [

7.0 CONCLUSION

S 'Our general conclusions regarding the failure of Inconel 6008 tubes in the OTSC A and B of TMI-1, based on examination results are as follows: (1) Inconel 600 tubes failed by intergranular cracking (IGC) (2) The cracks initiated on the inside surface of tubes and propagated outward. (3) The cracks are characterized by severe inter-(' granular attack (IGA) on either side of the crack { (4) Some small areas on the inside surface have IGA % 5 mils deep but with no cracks associated with them. [ (5) The cracks and the ICA have been found in the entire length of the short-pull section of tubes. f-(6) Whether the attack (IGA or IGC) extends beyond the short-pull length has not been fully evaluated { (7) There was a preponderance of cracks in the roll transition region of tubes, high stress concentra-tion in that part of the OTSG appears to be responsible (8) The most probable species responsible for [ the intergranular attack are the derivatives of sulfur (such as polythionic acid and { thiosulfate) which was found in significant concentration on the fracture surface (9) Thcre was no defect observed on the OD side of tuhes under dryout deposits [ 7 [

t [ l 2 [. (10) Some fission products, carbon and beryllium ( also were found on the fracture surface and on the inside surface of tubes; their role, {. if any, on the tube degradation process (es) is not discernible at present. (11) The attack on tubes most probably occurred during the layup following the hot-functional. { I [ [: [ [ [ [ [ [ ~ g r I

[ i i ( 8.0 ACKNOWLEDGMENTS [ The following persons actively participated in accomplishing the work cited herein. Paul Tomlin with metallography and radiography; i A. E. Austin with Auger, ESCA and secondary ion mass spectroscopy; Gene Sands with scanning electronmicroscopy and energy dispersive X-ray analysis; Don Hayford with addy current examination; Larry Lowry with Y-ray analysis and tensile testing; and Andy Skidmore with TEM. The ( arduous task of typing this report in a hurry was done by Irene Knight. (- [' [ [. ( [ [ [ [ [ [ ( o}}

ML20070H537

Contents

Text

1.0 INTRODUCTION

7.0 CONCLUSION

==

=m

7.0 CONCLUSION

Navigation menu

ML20070H537

Text

1.0 INTRODUCTION

7.0 CONCLUSION

==

=m

7.0 CONCLUSION

Navigation menu

Search