ML18059B160

Palisades Nuclear Plant Safe-End to Pipe Cracking.

ML18059B160
Person / Time
Site:	Palisades
Issue date:	01/31/1994
From:	CZAJKOWSKI C J BROOKHAVEN NATIONAL LABORATORY
To:	NRC
Shared Package
ML18059B159	List: ML18059B160
References
MCE-E2089-02, MCE-E2089-2, NUDOCS 9408260259
Download: ML18059B160 (20)

Text

• MCE-E2089-02 PALISADES NUCLEAR PLANT SAFE-END.TO PIPE CRACKING C. J. Czajkowski January 1994 Department of Advanced Technology Brookhaven National Laboratory Upton, NY 11973 1 ! i ' \

. , .....

• TABLE OF CONTENTS Page LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v1 1. INTRODUCTION

............................................

1 2. SEM/EDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

3. MICROHARDNESS

..........................................

2 4. DYE PENETRA'IT TESTING ...................................

3 5. OPTICAL MICROSCOPY

......................................

3 6. DISCUSSION

...............................................

4 REFERENCES

...................................................

5 v Figure 1 Figure 2 1 Figure 2a Figure 2b Figure 3 Figure 4 Figure 4a Figure 5 Figure Sa Figure 6 Figure 6a Figure 7 Figure 7a Figure S Figure Sa Figure Sb Figure Sc Figure Sd Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14

• LIST OF FIGURES Page Low magnification SEM fractograph . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Low magnification fractograph of area A .......................

7 lntergranular nature of the cracking -high magnification

...........

7 EDS scan of area A for contaminants . . . . . . . . . . . . . . . . . . . . . . . . . 7 Intergranular facets -low magnification fractograph . . . . . . . . . . . . . . .

S SEM photograph of area A .................................

9 EDS scan of area A for contaminants . . . . . . . . . . . . . . . . . . . . . . . . . 9 Intergranular cracking is clearly seen in area B ...................

9 EDS scan of area B for constituents

...........................

9 lnterdendritic (intercolumnar) cracking of the weld . . . . . . . . . . . . . . .

10 EDS scan of area C for contaminants

.........................

10 Large intergranular facet was seen in area D . . . . . . . . . . . . . . . . . . .

10 EDS scan of area D for constituents . . . . . . . . . . . . . . . . . . . . . . . . .

10 EDS scan of the Inconel 600 base metal . . . . . . . . . . . . . . . . . . . . . . .

11 The second base metal showed a scan typical of 316 SS ............

11 The. HAZ on the weld/inconel interface . . . . . . . . . . . . . . . . . . . . . . .

12 The HAZ on the SS/weld interface . . . . . . . . . . . . . . . . . . . . . . . . . .

12 The EDS scan of the weld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

13 Optical photomicrograph of first cross section ...................

14 Second cross section examined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

15 Possible carbide segregation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

16 A circular area of lack of penetration/porosity . . . . . . . . . . . . . . . . . .

16 A large area of lack of fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

16 The intergranular crack was located on the Inconel side HAZ . . . . . . .

17 VI

• 1. INTRODUCTION On September 16, 1993, through wall cracking was discovered in the pressurizer power operated relief valve (PORV) nozzle safe end of the Palisades Nuclear Plant. This cracking was detected early in the outage and preliminary analysis determined the cause of the cracking to be primary water stress corrosion of the inconel safe end welded to stainless steel piping. The U.S. Nuclear Regulatory Commission (NRC) contracted with Brookhaven National Laboratory (BNL) to confirm the licensee's failure analysis of the root cause of the On September 29, 1993, several contaminated and activated samples of the aforementioned piping were received at BNL for evaluation.

The evaluation was to consist of: 1. Scanning electron microscopy ( SEM) and energy dispersive spectroscopy (EDS) 2. Dye penetrant examination

3. Microhardness measurements

4. Optical microscopy

2. SEM/EDS Varioas lettered sections of the affected pipe were received at BNL. Of the sections received two had half of the fracture face from the "throu?h wall" crack. Figure 1 shows that the crack was entirely intergranular in appearance with an extremely coarsened grain size (ASTM 0-00). Figure 2 and 2a are fractographs of an area where energy dispersive spectroscopy was performed.

Note: EDS is an analytical technique, capable of performing elemental analysis of microvolumes, typically on the order of a few cubic microns in bulk samples and considerably less in thinner sections.

Analysis of X-rays emitted from a sample. is accomplished by crystal spectrometers which use energy dispersive techniques which permit analysis by discriminating among X-ray energies.

The feature of electron beam microanalysis that best describes this technique is its mass sensitivity.

For example, it is often possible to detect less than 10*16 grams of an element present in a specific microvolume of a sample. The minimum detectable quantity of a given element or its detectability limit varies with many factors, and in most cases is less than 10*16 grams/microvolume. (Note: EDS will only discern elements with atomic numbers greater than 11 so certain light elements will not be detected.)

• 1

• EDS scans of Figure 2 showed traces of Mg, Si, Ca, and Ti with Cre, Fe, and Ni. These scans (Cr, Fe, Ni) were more typical of stainless steel rather than Inconel 600 (Figure 2b ). Figure 3 is a low magnification SEM photograph, again showing intergranular cracking of this section of the crack's fracture face. The circled areas on the photograph are the areas examined by EDS. EDS scans (Figure 4 through Figure 7a) showed traces of Na, Al, Si, P, Cl, Ca, K, Ti, and V with Mn, Cr, Fe, and Ni also present. These scans characteristically bad much higher concentrations of Mn than would be expect-:-':!

Inconel 182 or Inconel 600. Although some scans showed P, S, or Cl, none of the contaminants can be confirmed in sufficient concentration to have contributed to the cracking.

Qualitative analysis (EDS) of a polished crass section was also performed.

Figures 8 through 8d are the graphical representations of these studies. The analysis did indicate that the materials of construction were as represented by the utility (lnconel 600, Inconel 182, and 316 SS). 3. MICROHARD1'1ESS Microhardness measurements were performed on two cross-sections cut from sample A (utility designation).

One of the cross-sections was taken from an area of apparent weld repau. Five readings were taken. in each area and averaged.

Sample with repair: Base metal -Inconel Weld metal -crown Weld metal -root Base metal -stainless Sample without repair: Base metal -Inconel Weld metal -crown Weld metal -root Base metal -stainless RB 90.5 90.5 99.0 (21.5 Re) 93.4 85-90 RB 94.0 90 91.5 89.5 Equiv. Tensile Str. 90 ksi 90 ksi 113 ksi 97 ksi 80-90 ksi Equiv. Tensile Str. 100 ksi 89 ksi 92 ksi 88 ksi These readings indicate that the material is in a hardened condition which would make it more susceptible to a sec phenomenon.

2

• 4. DYE PENETRANT TESTING Dye penetrant testing was performed on the inside surface and cut sides of the specimens received.

The materials used for the dye penetrant were: Magnaflux

-Penetrant SKL-HF/S, Developer SKD-S, and Cleaner SKC-NF/ZC-7.

The procedure used for the dye penetrant inspection of the specimens follows: 1. All dye penetrant examinations were performed indoors at approximately room temperature.

2. Cleaner -The specimen to be tested was wiped clean and then allowed to dry for approximately ten minutes. 3. Penetrant

-After the penetrant was applied it was allowed to stand for 10 minutes. 4. Cleaner -The penetrant was wiped off and the area allowed to dry for ten minutes. 5. Developer

-The developer was applied and the surface examined.

A possible indication was noted on sample E (utility designation) and a possible "end of crackn was seen on an extremely rough cut area of sample B. Additional sections were removed from these areas for cross-section/optical microscopy.

5. OPTICAL MICROSCOPY Three sections of the PORV recei\*:j at BNL were cut, ground, metallurgically polished, and etched. The etchant used was composed of phosphoric acid and water and electrolytically applied. This etch revealed carbides on application.

Figure 9 is an optical photomicrograph of the first section examined.

\\'ith the exception of a rather large weld root gap, no abnormalities were noted. Figure 10 is a second cross section examined.

One area of lack of fusion (Figure 13) was seen at the fusion line between weld passes. A second area of lack of penetration/porosity was also noted (Figure 12) in the weld. Figure 11 is a higher magnification photomicrograph of possible carbide segregation seen after etching . . A small dye penetrant indication was originally seen on section E (utility designation)..

This section was mounted and etched. Figure 14 is the optical photomicrograph of the resultant cross section. The crack on the inconel side of the weld metal deposit was confined to the heat affected zone. There was an unusually large weld 3

• pass deposit (-2 passes) at the weld root. The size and directionality of these passes .indicate a probable weld repair which was made from the inside surface of the pipe. These observations coupled with the visual observations of (pipe inside surface) possible weld repairs/bad fit up, indicate that the welded joint was perhaps substantially reworked.

This is partially substantiated by the extremely large grain sizes seen on the fracture surfaces.

This amount of rework/repair is also a good indication that relatively high tensile stresses (up to and possibly over yield) were present in this weld configuration.

6. DISCUSSION For stress corrosion cracking to occur, these components/conditions must be pre.;ent:

1. susceptible matl:ial 2. corrosive environment

3. tensile stress (applied or residual)

In the case of this event, the susceptible material is the high hardness Inconel 600 material [1]. Higher strength (high hardness) material has been shown to be susceptible to IGSCC [2]. The corrosive environment at work is primary water. This corrosive environment is made more aggressive if oxygen and/or hydrogen are present. This is a distinct possibility in the area of these nozzles. The final component needed for sec to occur is tensile stress (residual or applied) [3]. The areas or rework/repair on the inside surface of the pipe are a clear indication that this weldment was under considerable stress. Measurements made on austenitic welds have shown stresses up to yield. This stress can further be increased by repairs. It appears that the large grain sizes n0ted on the fracture surfaces are not necessarily the result of repair welding, but may be the result of overheating the forging. The Metals

• Handbook [4] defines burning as occurring when ' ... a metal is grossly overheated and permanent irreversible damage to the metal occurs as a result of the intergranular penetration of oxidizing gas, or occurs by incipient melting.'*

Additionally, the Handbook states 'In steels burning may manifest itself with the formation of extremely large grains and incipient melting at the grain boundaries

.. .' It is evident that burning is a more advanced or detrimental form of overheating.

Overheating is evidenced by a large grain size, an intergranular fracture surface and fine oxide particles dispersed throughout the grains. These large grains are not conducive to a crack-free weldment.

This being the case, care should be taken to carefully evaluate the continued installation of this possibly overheated forging material in critical areas of the PORV. 4

• Sufficient information has been developed to provide conclusions regarding the failure.:

1. The cracking was intergranular with some interdentritic cracking where the crack entered the weld metal. 2. There appeared to be areas of poor fit-up\weld repair which would subject the weld to thermal cycles and relatively high stresses.

3. Traces of possible detrimental elements were found (S, Cl). Their contribution to the cracking is not clear. The corrosive species is considered to be primary water. 4. The large grain sizes seen are indicative of an overheating phenomenon.

This coupled with the relatively high hardness of the material would make the alloy 600 material potentially susceptible to IGSCC. . 5. The potentially susceptible material coupled with high residual stresses in a primary water environment lead to the conclusion that the cracking is a result of PWSCC. . REFERENCES

1. Metals Handbook, Vol. 11, ASM, 9th Edition, pp 660-661. 2. Airey, G.P., "The Stress Corrosion Cracking (SCC) Performance of Inconel 600 in Pure and Primary Water Environments." Proceedings of the International Symposium

()Il Environmental Degradatio:i of Materials in Nuclear Power Systems -Water Reactors, NACE{fMS/ANS, August 22-25, 1983, pp 462-478. 3. Metals Handbook, Vol. 13, ASM, 9th Edition, p 941. 4. Metals Handbook, Vol. 10, ASM, 8th Edition, pp 305-306.

f... *-.. . "' . .,,.. .. -**.

• Figure 1. Low magnification SEM fractograph of one half of the "through wall" crack.

Figure 2. Low magnification fractograph of area A (Figure 1 ). Figure 2b. Figure 2a. Intergranular nature of the cracking seen in a higher magnification photo . EDS scan of area A for conta.minants.

• Figure 3. Intergranular facets are seen on this low magnification fractograph in addition to interdendritic cracking (in area where crack entered the weld -area C). Note the large grains throughout the fracture face not just associated with the possible repair areas on the pipe inside surface. Circled areas are locations of EDS scans. Pipe Outside Surface

• Figure 4. Figure 5. SEM photograph of area A Intergranular cracking is clearly seen in area B. Figure 4a. Figure 5a.

• EDS scan of area A for contaminants.

• EDS scan of area B for constituents.

Figure 6. Figure 7. Interdendritic (intercolumnar) cracking of the weld is seen in area C. Large intergranular facet was seen in area D. Figure 6a. EDS scan of area C for contaminants.

Figure 7a. EDS scan of area D for constituents.

• ; : ........................

t ........................

........................

l ........................

........................ , ........................

........................

........................

.. ************

.. ***"*":******"''""_*****1 i l I I I r i I 11 u i I I ! I i :, I I i I ...................

• • ........................

........................

........................ , ........................

........................

........................

........ ! ............

........................ , ........................ , ..... .

• i j i I i I !\ i '" i ! I ****;**

1 L.LLL .

60 Figure 8. Figure 8a. EDS scan of the Inconel 600 base metal showing typical Ni, Cr, and Fe. The second base metal showed a scan typical of 316 SS. 11 . L.

60

• . . ................... ........................

........................ ........................

........................

• . . . . . . . . . . . . . . . . . . .. . ...........................................

........................ ........................ ... . ..................

1..............................

...............

!****************************************

1 ***************

.**********************.************************1*****************************************************************'***********************i****, I : : : : ****** *****************:

• • • • • • • • • • '************************[*************************************************:************************:********

• • • • • • • • ******r****

• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 1************************;*****

.-. F E ***********F>*'

..... __ * ..l * ...l l _ _,""""'*'-'"*'"**..-,...

..

... ,..,.:...

...

\i... ... ,r:.. .. _f \ ... -.. .i-._..i ... * .r1"'l\ : , 1*1 UIZI'*" *.'!-'.....*

= 18<::4 11Zl.:::'."!U 60 Figure 8b .. The HAZ on the weld/inconel interface showed nonnal Cr, Fe, and Ni ratios. . F

.. ***********

.. ****:*********

.. ******* .. ******************

.. ***** ..........................

• • • • • .. . ************************l************************=************************;************************i************************:************************:********

.............

! ........................

........................

t .......................

.... . . . : : .............

........................

........................

........................

.... . . . * * . . . . . . . . .

....................*** ........................

j ............................. . ************(*********

.. **********1******************""*-r***

.. **************"*'***** . .

.. :************************:-****

E I \ \ ! : : : :

.. ************[***********************+-***

\) \.,,_ ..

._]. .. . l ...

• q ... ,_ -c:: .f . -* ¥I _1 ...

HJ.:.::'.4U 60 Figure 8c. The HAZ on the SS/weld interface was also normal. 12

• . *!***** .................

.. ***** .. ****** .. ******** :N* ...................... ...................

................................................ , ...... . . ...;-., .. "

._Jl\..,.":J,., ...... ;,.L .. /.,*--,.,,..

.... -w ... L ... , ... ,,.'1:4<.',.l..J

... J.* ... . . I . .

I . .) l : : ** -,;... ..... * *-..L... ** :

• I.: . :::: . l:1 60 Figure 8d. The EDS scan of the weld was consistent with inconel 82/182.

-** ... , . .. -. .... _ : ...... , -* ------* . *.

Figure 9. .. :_ --:: -.-' ---... :.-.... -.. . ' Optical photomicrograph examined from Palisades. . . of cross section removed from first section Note the large weld gap at the weld root. 14 Figure 10. .. ::.:.-*' Second cross section examined.

Note Jack of fusion (area) and possible Jack of penetration/porosity (area B). Crown of we]d also had significant carbide segregation.

15 I

-0\ Figure 11. ----.:_ .... -:-:*-

.--; --... . . ---*.r.":

o--* ' 0.8 MM -:1-;;-S I. ' . ' *'. ,;... ;:.) ,.. .""'" .,. 1; Possible carbide segregation (striping) is noted on the cross section. Figure 12. A circµlar area of lack of penetration/porosity is seen in area B (Figure 10). Figure 13. A large area of lack of fusion is seen between weld passes in area A (Figure 10). *,

Figure 14. / .* .*-:'!. ,_.'.-. :-: ,,

. -:.: .. * '.-*-* .. : The intergranular crack was loc<. :ed on th..: inconel side HAZ. A large area of weld repair is seen at the weld root. 17 ..