ww.sciencedirect.com

Nu c l E n g T e c h n o l 4 7 ( 2 0 1 5 ) 9 3 4e9 3 8

Available online at w

ScienceDirect

journal homepage: ht tp: / /www.journals .e lsevier .com/nuclear-engineer ing-and-technology/

Technical Note

CONSIDERATIONS FOR METALLOGRAPHIC OBSERVATION OFINTERGRANULAR ATTACK IN ALLOY 600 STEAM GENERATORTUBES

DO HAENG HUR*, MYUNG SIK CHOI, DEOK HYUN LEE, and JUNG HO HAN

Korea Atomic Energy Research Institute, 150 Deokjin-dong, Yuseong-gu, Daejeon, 305-353, Republic of Korea

a r t i c l e i n f o

Article history:

Received 6 April 2015

Received in revised form

13 July 2015

Accepted 9 August 2015

Available online 19 October 2015

Keywords:

Crazing

Deformation

Intergranular attack

Steam generator tube

Stress corrosion crack

* Corresponding author.E-mail address: dhhur@kaeri.re.kr (D.H. H

This is an Open Access article distributecreativecommons.org/licenses/by-nc/3.0) whdium, provided the original work is properlyhttp://dx.doi.org/10.1016/j.net.2015.09.0031738-5733/Copyright © 2015, Published by El

a b s t r a c t

This technical note provides some considerations for the metallographic observation of

intergranular attack (IGA) in Alloy 600 steam generator tubes. The IGA region was crazed

along the grain boundaries through a deformation by an applied stress. The direction and

extent of the crazing depended on those of the applied stress. It was found that an IGA

defect can bemisevaluated as a stress corrosion crack. Therefore, special caution should be

taken during the destructive examination of the pulled-out tubes from operating steam

generators.

Copyright © 2015, Published by Elsevier Korea LLC on behalf of Korean Nuclear Society.

1. Introduction

Intergranular corrosion of nuclear steam generator tubes can

be divided into at least three forms: intergranular stress

corrosion cracking (IGSCC), intergranular attack (IGA), and

intergranular penetration (IGP) [1]. In the case of IGSCC, the

corrosion morphology consists of single or multiple major

cracks with minor to moderate amounts of branching. The

morphology of IGA is characterized by a relatively uniform

attack of numerous grain boundaries to a uniform depth over

ur).

d under the terms of theich permits unrestrictedcited.

sevier Korea LLC on beha

the surface of themetallicmaterials. This is because corrosion

is localized at and adjacent to grain boundaries with relatively

little corrosion of the grains. Finally, IGP can be described as a

mixture between the other two forms.

IGA has been one of the major corrosion degradation

modes in steam generator tubes. It has been observed

mainly on the outer diameter (OD) side of the tubes in the

sludge piles on top of the tubesheet or in the deposits

adjacent the tube support structures [2e5]. However, acci-

dental ingress of thiosulfate into the primary water led to

Creative Commons Attribution Non-Commercial License (http://non-commercial use, distribution, and reproduction in any me-

lf of Korean Nuclear Society.

Nu c l E n g T e c h n o l 4 7 ( 2 0 1 5 ) 9 3 4e9 3 8 935

extensive IGA on the inner diameter (ID) side of the sensi-

tized tubes [6].

We have found that steam generator tubes with IGA are

easily crazed along the grain boundaries when under plastic

deformation. Corroded grain boundaries would lose fracture

toughness and become brittle. Therefore, a region with IGA/

IGP defects is expected to be susceptible to cracking by an

external stress. This article provides the metallographic

characteristics of IGA in Alloy 600 steam generator tubes. In

addition, the effect of the applied stress on the morphology

change of the IGA region is discussed.

2. Materials and methods

The Alloy 600 tubing material used in this study was supplied

by a commercial vendor. The tubes were mill-annealed in a

temperature range of 1,024~1,070�C for 3 minutes and then

cooled down to 500�C within 7 minutes. The nominal OD of

the tube was 19.05 mm and the nominal wall thickness was

1.07 mm. The chemical composition is listed in Table 1. To

accelerate intergranular corrosion, the tubes were addition-

ally sensitized at 590�C for 10 hours in a vacuum furnace

under about 5 � 10�6 torr.

Samples were prepared by cutting the tube circum-

ferentially into 6-cm-long pieces. For manufacturing IGA on

the inner side of the tubes, one end of each tube specimenwas

plugged with a Teflon cap so that the solution inside the tube

did not leak out. Next, the tube specimen was filled with an

oxidized solution of 0.1M sodium tetrathionate (Na2S4O6).

Solutions containing sulfur oxyanions has been known to

accelerate the corrosion of nickel-based alloys and stainless

steels along the grain boundaries [7,8]. By contrast, to produce

IGA on the OD side of the tube, both ends of each tube spec-

imen were plugged with Teflon caps. Next, the tube speci-

mens were immersed in 0.1M Na2S4O6 solution. In this way,

IGA was grown on the ID or OD side of the tube at room

temperature for 5 days.

The IGA tubes were deformed by applying several types

of stress, such as hoop stress, three-axes stress, hard roll-

ing, and indentation. If necessary, the tubes with IGA were

cut into pieces of appropriate size. The subsequent

morphology changes of the IGA area were observed using

scanning electron microscopy. The detailed information

about how stress or deformation was applied to specimens

and where the morphology was observed in the specimen is

described in the results and discussion section. Because this

work is focused on the morphology change of the IGA

specimen by an applied stress, the magnitude of the applied

stress and the corresponding deformation extent are not

quantified.

Table 1 e Chemical composition of Ally 600 tube (wt %).

C Cr Fe Ni Si Mn Ti Al S

0.025 15.52 9.30 Bal. 0.19 0.21 0.29 0.22 <0.001

Al, aluminum; C, carbon; Cr, chromium; Fe, iron; Mn, manganese;

Ni, nickel, S, sulfur; Ti, titanium.

3. Results and Discussion

Fig. 1 shows themorphology change of the IGA tube surface by

an applied stress. No defects were observed on the ID surface

of the tube without any applied stress conditions, as shown in

Fig. 1A. However, some crazing occurred along the tube axial

direction when applying hoop stress by bending the specimen

along the circumferential direction of the tube, as shown in

Fig. 1B. The arrows indicate the same location before and after

deformation. They just look like axial stress corrosion cracks

(SCCs). Similarly, some crazing occurred along the circum-

ferential direction of the tube by applying tensile stress.

Therefore, they seem to be circumferential SCCs.When three-

axes stress was applied to the IGA tube specimen, the surface

was crazed into a radial crack-like morphology (Fig. 1C).

Finally, the numerous attacked grains were apparently

revealed through a distorted deformation (Fig. 1D). These re-

sults indicate that an IGA can be misunderstood as a SCC by a

directional deformation. Similar behaviors were also observed

on the tubes with IGA on the OD side. In the IGA region, the

attacked grain boundaries become brittle, although they are

extremely tight in nature. Therefore, the corroded grain

boundaries are easily opened through a deformation by an

externally applied stress.

Fig. 2A shows a circumferential cross section of the IGA

tube. There was no evidence of IGA on the as-polished metal-

lographic sample. However, when the same tube was

expanded outward by hard rolling, IGAwas clearly revealed by

a crazing of the IGA region, as shown in Fig. 2B. The scratch and

arrow indicate the same location before and after deformation.

Fig. 3A shows a circumferential cross-section of the IGA

tube. There is no doubt that the feature of the defect type is a

single SCC. In this case, it is reasonable to call this flaw a

primary water stress corrosion crack (PWSCC) because it was

initiated from the ID surface of the tube. However, when the

same area was forced by a Vickers hardness indenter (Mitu-

toyo, model HM-122, Japan) at a load of 1 kg, abundant crazing

along the grain boundaries occurred, as shown in Fig. 3B. The

white arrows indicate the same location before and after

indentation. Consequently, this result clearly indicates that

this defect is IGA, not PWSCC.

Fig. 4 shows the fracture surface of the laboratory-grown

IGA tube and PWSCC in a tube pulled from an operating

plant. The fracture surface of the IGA tube specimen showed

the same appearance as that of typical intergranular SCCs.

Therefore, the intergranular nature of the fracture surface

cannot be proof of SCCs. Among somemechanisms of PWSCC,

the internal oxidation mechanism is related to oxygen pene-

tration at the grain boundary, resulting in the formation of a

brittle intergranular oxide [9]. The model predicts a strong

dependence on the potential. When the potential is too low,

oxidation is not possible; when it is too high, a compact oxide

grows and prevents further oxygen diffusion and oxidation

[9]. However, intergranular crazing observed in this work de-

pends on the degree of intergranular corrosion before

deformation.

The degree of IGA depends on not only the depth andwidth

of the chromium depletion along the grain boundaries [10e12]

but also the corrosive environmental factors [13,14]. Therefore,

Fig. 1 eMorphology change of intergranular attack (IGA) region: (A) before deformation, after deformation (B) by hoop stress,

(C) by three-axes stress, and (D) by distortion.

Fig. 2 e Morphology change of intergranular attack region on the transverse cross section of the tube by a hard roll

expansion: (A) before expansion and (B) after expansion.

Nu c l E n g T e c h n o l 4 7 ( 2 0 1 5 ) 9 3 4e9 3 8936

grain drops are not always identified on the IGA surface.

Similarly, IGA can sometimes be observed on an as polished

cross-section [5], whereas in some cases it can be viewed only

using proper etching methods [15]. Therefore, special caution

should be taken to evaluate the degradation type of an IGA-

affected sample.

The results obtained in this work clearly show that the

presence of IGA can be revealed by an applied deformation to

the tube sample. A region with IGA can easily be crazed along

the grain boundaries by an externally applied stress. The

extent and direction of the crazing depend on the type and

direction of the applied stress to the corroded tube. Such a

stress could be applied to the tubes during the tube-pulling

process in a steam generator for destructive examination.

Depending on the direction and extent of the stress, the

invisible IGA can be apparently revealed, whereas the IGA can

also be misevaluated as an SCC. In addition, etching tech-

niques may fail to reveal the IGA defect. Therefore, it seems

that the best way to evaluate the IGA is to expand the tube by

internal pressurization. Thereby, some of the errors described

can effectively be eliminated. It is recommended that a sec-

tion of a pulled tube be hydraulically or pneumatically

Fig. 3 e Morphology change of intergranular attack region on the transverse cross section of tube by forcing with a Vickers

hardness indenter. (A) Before indentation and (B) after indentation.

Fig. 4 e Fracture surface. (A) Intergranular attack and (B) primary water stress corrosion crack.

Nu c l E n g T e c h n o l 4 7 ( 2 0 1 5 ) 9 3 4e9 3 8 937

expanded before sectioning to the axial direction for metal-

lographic examination.

4. Conclusions

It was found that an IGA tube was crazed along the grain

boundaries into various types and directions through a

deformation from an applied stress. The direction and extent

of the crazing depended on those of the applied stress. It was

clearly shown that an IGA could be misevaluated as an SCC.

Therefore, it is recommended that a section of pulled tube be

hydraulically or pneumatically expanded for an exact evalu-

ation of the IGA defect in steam generator tubes.

Conflicts of interest

All authors have no conflicts of interest to declare.

Acknowledgments

Thisworkwas supportedby theNational Research Foundation

of Korea (NRF) grant funded by the Korea government (MSIP).

r e f e r e n c e s

[1] J.A. Gorman, J.P.N. Paine, M.J. Partridge, D.B. Lowenstein,Secondary Side Intergranular Attack and Stress CorrosionCracking, in: J.P.N.Paine (Ed.), SteamGeneratorReferenceBook,Rev.1, Vol. 2, Electric Power Research Institute, Palo Alto, CA,1994, pp. 12.1e12.75. TR-103824.

[2] D.W. Diercks, W.J. Shack, J. Muscara, Overview of steamgenerator tube degradation and integrity issues, Nucl. Eng.Design 194 (1999) 19e30.

[3] F. Cattant, Lessons learned from the examination of tubespulled from Electricite de France steam generators, Nucl.Eng. Design 168 (1997) 241e253.

[4] A. Stutzmann, F. Nordmann, Studies of IGA intergranularattack/stress corrosion cracking related parameters andremedies, Nucl. Eng. Design 162 (1996) 167e174.

[5] D.H. Hur, D.H. Lee, M.S. Choi, M.H. Song, J.H. Han, Rootcauses of intergranular attack in an operating nuclear steamgenerator tube, J. Nucl. Mater. 375 (2008) 382e387.

[6] P.M. Scott, P. Combrade, Corrosion in PressurizedWaterReactors, in:Corrosion:EnvironmentsandIndustries,Vol. 13C,ASM International, Materials Park, OH, 2006, pp. 362e385.

[7] G.S. Was, V.B. Rajan, The mechanism of intergranularcracking of Ni-Cr-Fe alloys in sodium tetrathionate,Metallurg. Trans. A 18A (1987) 1,313e1,323.

[8] Z. Fang, R.W. Staehle, Effects of the valence of sulfur onpassivation of Alloy 600, 690, and 800 at 25�C and 95�C,Corrosion 55 (1999) 355e379.

Nu c l E n g T e c h n o l 4 7 ( 2 0 1 5 ) 9 3 4e9 3 8938

[9] P. Scott, M. Le Calvar, Some possible mechanisms ofintergranular stress corrosion cracking of Alloy 600 in PWRprimary water, in: Proceedings of the 6th InternationalSymposium on Environmental Degradation of Materials inNuclear Power Systems e Water Reactors, The Minerals,Metals & Materials Society, San Diego, CA, 1993, pp.657e668.

[10] G.S. Was, H.H. Tischner, R.M. Latanision, The influence ofthermal treatment on the chemistry and structure of grainboundaries of Inconel 600, Metallurg. Trans. A 12A (1981)1,397e1,408.

[11] G.P. Airey, Microstructural characterization of Inconel Alloy600 annealed in the 700 to 1,200℉ temperature range,Corrosion 41 (1985) 2e10.

[12] C.L. Briant, C.S. O'Toole, E.L. Hall, The effect ofmicrostructure on the corrosion and stress corrosioncracking of Alloy 600 in acidic and neutral environments,,Corrosion 42 (1986) 15e27.

[13] P.J. Millett, PWR Molar Ratio Control Application Guidelines,Vol. 1, Electric Power Research Institute, Palo Alto, CA, 1995.TR-104811eV1.

[14] A. McIlree, Oxide-film Compositions on Alloy 600 Tubes fromSteam Generators, TR-104066, Electric Power ResearchInstitute, Palo Alto, CA, 1998.

[15] G.P. Airey, F.W. Pement, A Comparison of IntergranularAttack in Inconel Alloy 600 Observed in the Laboratory and inOperating Steam Generators, Corrosion. 82, Paper No. 205,NACE International, Houston, TX, 1982.