Published by Maney Publishing (c) IOM Communications Ltd Influence of plasticity on corrosion and stress corrosion cracking behaviour in near neutral pH environment X. C. Li, R. L. Eadie and J. L. Luo* This study focused on the effect of plasticity on stress corrosion cracking of pipeline steel (X-52) in near neutral pH environment. A simulated near neutral pH soil solution, referred to as C-1, was used for all the tests. The effects of plastic deformation on both corrosion current and open circuit potential were studied. Crack advance was measured in samples with various amounts of plastic deformation by slow strain rate tests and corrosion fatigue tests using potential drop to measure crack advance. Crack morphologies and the crack features on the fracture surface were observed by SEM. All the cracking tests indicated that the stress corrosion cracking rate was enhanced by the plastic deformation. The electrochemical tests showed that corrosion current increased with increasing plastic deformation. The relationship between the cracking mechan- isms and prior plastic deformation are discussed. Keywords: Near neutral pH SCC, Pipeline steel, Crack growth rate Introduction Since 1985, when the first near neutral pH stress corrosion cracking (NNPHSCC) was observed by TransCanada PipeLines Ltd (TCPL) in a tape coated pipeline, 1 tens of failures have been identified in the pipeline system in Canada. 2 Recently, the Office of Pipeline Safety in the United States released a study of SCC as it affects American pipelines. 3 They indicated that NNPHSCC is a significant threat in their pipeline system as well. The phenomenon is characterised by wide transgranular cracks in dilute carbonate bicarbo- nate solutions with pH of the ground water found under the tape coating measured to be in the range 5?5–7?5. Previously, substantial work has been carried out to investigate the effects of the three fundamental factors contributing to NNPHSCC, 2 but there is still a great deal that remains unknown or uncertain in terms of the roles played by the key factors such as stress and stress induced plastic deformation. This work studied the effect of plastic deformation on the corrosion behaviour and NNPHSCC behaviour of a typical pipeline steel X-52 that would have been installed in the early 1970s, which is the vintage principally affected by this phenomenon. From the design point of view, the pipelines are designed to try and ensure that only elastic deformation will result when they are operating under pressure. However localised plastic deformation from bending, rolling, or handling is often encountered in the manufacture and assembly of pipeline components that later are exposed to service conditions. 4 There are also many naturally occurring discontinuities such as the long seam weld that can raise the local stress level above the elastic limit. Many corrosion situations such as pitting can cause local discontinuities that will elevate the stress locally above the yield strength. Furthermore, there are several other situations causing localised microplastic deformation at stress levels below the yield stress of the steel. First, the surface layer of the pipe wall thickness can deform before the bulk of the wall thick- ness; 5 second, cyclic loading can cause steels to exhibit microplastic strains at nominal stress levels where plastic strains would not be expected. 6 Also, hydrogen assisted plasticity can occur, which may delay the exhaustion of the primary room temperature creep and increase the SCC susceptibility. 7 It is therefore desirable to know the pipeline’s response to NNPHSCC conditions after it experiences some plastic deformation. The plasticity in this study was induced either by cold rolling or tensile strain. Electrochemical tests were undertaken on the same steel to study the corrosion characteristics of deformed materials in order to under- stand the effect of plastic deformation on anodic dissolution during NNPHSCC. Experimental Materials Several joints of X-52 low carbon steel were removed from service after more than 30 years in the field by Enbridge oil pipeline. Steel from this line pipe, which had an external diameter of 34 in (864 mm) and wall thickness of about 7?1 mm, was used in this Department of Chemical and Materials Engineering, University of Alberta, Edmonton, Alberta, Canada *Corresponding author, email [email protected]ß 2008 Institute of Materials, Minerals and Mining Published by Maney on behalf of the Institute Received 21 December 2006; accepted 30 October 2007 DOI 10.1179/174327808X286220 Corrosion Engineering, Science and Technology 2008 VOL 43 NO 4 297
Transcript
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
Influence of plasticity on corrosion and stresscorrosion cracking behaviour in near neutralpH environment
X C Li R L Eadie and J L Luo
This study focused on the effect of plasticity on stress corrosion cracking of pipeline steel (X-52) in
near neutral pH environment A simulated near neutral pH soil solution referred to as C-1 was
used for all the tests The effects of plastic deformation on both corrosion current and open circuit
potential were studied Crack advance was measured in samples with various amounts of plastic
deformation by slow strain rate tests and corrosion fatigue tests using potential drop to measure
crack advance Crack morphologies and the crack features on the fracture surface were
observed by SEM All the cracking tests indicated that the stress corrosion cracking rate was
enhanced by the plastic deformation The electrochemical tests showed that corrosion current
increased with increasing plastic deformation The relationship between the cracking mechan-
isms and prior plastic deformation are discussed
Keywords Near neutral pH SCC Pipeline steel Crack growth rate
IntroductionSince 1985 when the first near neutral pH stresscorrosion cracking (NNPHSCC) was observed byTransCanada PipeLines Ltd (TCPL) in a tape coatedpipeline1 tens of failures have been identified in thepipeline system in Canada2 Recently the Office ofPipeline Safety in the United States released a study ofSCC as it affects American pipelines3 They indicatedthat NNPHSCC is a significant threat in their pipelinesystem as well The phenomenon is characterised bywide transgranular cracks in dilute carbonate bicarbo-nate solutions with pH of the ground water found underthe tape coating measured to be in the range 55ndash75Previously substantial work has been carried out toinvestigate the effects of the three fundamental factorscontributing to NNPHSCC2 but there is still a greatdeal that remains unknown or uncertain in terms of theroles played by the key factors such as stress and stressinduced plastic deformation This work studied theeffect of plastic deformation on the corrosion behaviourand NNPHSCC behaviour of a typical pipeline steelX-52 that would have been installed in the early 1970swhich is the vintage principally affected by thisphenomenon
From the design point of view the pipelines aredesigned to try and ensure that only elastic deformationwill result when they are operating under pressureHowever localised plastic deformation from bendingrolling or handling is often encountered in the
manufacture and assembly of pipeline components thatlater are exposed to service conditions4 There are alsomany naturally occurring discontinuities such as thelong seam weld that can raise the local stress level abovethe elastic limit Many corrosion situations such aspitting can cause local discontinuities that will elevatethe stress locally above the yield strength Furthermorethere are several other situations causing localisedmicroplastic deformation at stress levels below the yieldstress of the steel First the surface layer of the pipe wallthickness can deform before the bulk of the wall thick-ness5 second cyclic loading can cause steels to exhibitmicroplastic strains at nominal stress levels where plasticstrains would not be expected6 Also hydrogen assistedplasticity can occur which may delay the exhaustion ofthe primary room temperature creep and increase theSCC susceptibility7 It is therefore desirable to know thepipelinersquos response to NNPHSCC conditions after itexperiences some plastic deformation
The plasticity in this study was induced either by coldrolling or tensile strain Electrochemical tests wereundertaken on the same steel to study the corrosioncharacteristics of deformed materials in order to under-stand the effect of plastic deformation on anodicdissolution during NNPHSCC
Experimental
MaterialsSeveral joints of X-52 low carbon steel were removedfrom service after more than 30 years in the fieldby Enbridge oil pipeline Steel from this line pipewhich had an external diameter of 34 in (864 mm) andwall thickness of about 71 mm was used in this
Department of Chemical and Materials Engineering University of AlbertaEdmonton Alberta Canada
Corresponding author email jingliluoualbertaca
2008 Institute of Materials Minerals and MiningPublished by Maney on behalf of the InstituteReceived 21 December 2006 accepted 30 October 2007DOI 101179174327808X286220 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 297
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
investigation The chemical composition of the steel isFendash0261Cndash1150Mnndash0008Pndash0019Sndash0036Sindash0033Cundash0002Snndash0032Nindash0027Crndash0013Mondash0002Alndash0003Vndash0003Nbndash0002Tindash0006Co (wt-) The microstructure consistedof fine pearlite and ferrite with a grain size of 10ndash20 mmThe specified minimum yield stress for the material was3590 MPa (X-52) and the ultimate tensile strength was589 MPa
Test SolutionNS4 synthetic solution has been widely used to simulatethe soil solution in the study of NNPHSCC behaviourHowever another synthetic soil solution termed C-1with composition shown in Table 1 was designed basedon field data from pipeline failure sites In previousresearch8 this C-1 solution was shown to be moreaggressive in causing NNPHSCC than NS4 solution andwhen bubbled with 5CO2 exhibited a slightly lowerpH The C-1 solution was prepared using deionisedwater and was bubbled with 5CO2 in oxygen free N2
gas for at least 24 h before being introduced to the testcells The bubbling gas was also maintained during eachtest in order to keep the pH value of the solution at 589and also to get an anaerobic environment Anaerobicenvironments have been identified as being associatedwith NNPHSCC12
Slow strain rate testing (SSRT)Slow strain rate testing were carried out on smoothcylindrical tensile samples which are shown in Fig 1The length direction of the sample was parallel to thecircumferential direction of the pipeline steel in order toensure that the subsequent crack growth was in thelongitudinal direction of the pipe as is typically observedin the field After they were machined from the pipeusing electrical discharging machining (EDM) thesamples were prestrained in an Instron hydraulic testframe at 002 mm min21 Different amounts of plasticdeformation from 1 to 8 were obtained as determinedby an extensometer The sample surface in the gaugesection was polished to a 600 grit finish in an orientationparallel to the subsequent loading direction of the
SSRT This ensured similar surface conditions for alltests
The SSRT was conducted on samples in bothdeaerated C-1 solution at open circuit potential (OCP)and in air The strain rate for samples tested in C-1solution was normally begun at a crosshead speed of00004 mm min21 which corresponded to an initialelastic strain rate of 26261027 s21 while a crossheadspeed of 005 mm min21 was used for the samples testedin air (1256 faster) which corresponded to a strain rateof 32861025 s21 The faster rate can be used since it isknown that results in air are unaffected by the change instrain rate The load elongation behaviour was recordedAfter the test was complete the fractured sample wasimmediately removed ultrasonically cleaned usingacetone and placed in a desiccator for storage untilfuture SEM examination
Corrosion fatigue testing using potential drop tomeasure crack advanceThe samples cut from the longitudinal direction of thepipeline were cold rolled to different amounts of plasticdeformation (5 8 10) Then the compact toughness(C-T) specimens (as shown in Fig 2) were made usingEDM according to ASTM Standard E813-89 except forsample thickness B which was limited by the maximumpipe wall thickness that could be utilised A sharp notchwas introduced on the sample along the longitudinaldirection with the notch radius being 015 mm The C-Tsamples were prefatigued in air using an INSTRONmachine to produce a sharp crack from the machinednotch tip in accordance with ASTM E647 Theprefatigued crack length on both surfaces was controlledto be 2ndash3 mm long The difference in crack length on thetwo sides of each sample was less than 02 mm
EnduraTEC equipment was used for cyclic strainingduring the potential drop tests A sine waveform loadwas used with a loading frequency of 00025 Hz (216cyclesday) and R ratios of 0375 05 and 0625 Thistest frequency is known to be similar to strain rates in oilpipelines but somewhat fast for gas pipelines9 Themaximum and minimum stresses were controlled toachieve a maximum stress intensity factor K close to40 MPa m12 and a DK in the range 15 to 25 MPa m12
Table 1 Chemical components of NS4 and C-1 solution (bubbled with 5CO2zN2 balanced)
Composition g L21 MgSO47H2O CaCl2 KCl NaHCO3 CaCO3 pH
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
298 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd After the test the samples were broken open by
cooling the sample in liquid nitrogen and fracturing itwith a sudden crack opening load The crack faces werecleaned ultrasonically in acetone solution and thecleaned fracture surfaces were examined by opticalmicroscopy to measure the actual crack growth lengthwhich corresponded to the total potential changerecorded This gave the correspondence between crackgrowth length and potential change
The stress intensity factor KI could be calculatedaccording to ASTM 399-90 (Ref 10)
fa
w
~
KIBw12
P~
1a
w
32
2za
w
0886z464 aw1332 a
w
2z
1472 aw
356 a
w
4
24
35
8lt
9=
where a and B are crack length and sample thicknessrespectively w is the width of the sample and P is theapplied load
Electrochemical testsA smooth round tensile specimen as shown in Fig 1was also used for potentiodynamic polarisation scansThe polarisation curve was obtained at a scan rate of02 mV s21 starting approximately 250 mV more nega-tive than the OCP and scanning in the noble direction toa potential 650 mV more positive than OCP
Results
Increase of yield strengthFigure 3 shows the yield strength of the samples withdifferent amounts of plastic deformation The yieldstrength of X-52 steel increased with increasing plasticdeformation for samples tested in air and also for thosetested in C-1 solution There is a small reduction in yieldstrength when testing in C-1 solution compared to air
Increase of hardnessResults of the average of three hardness measurementson each sample are shown in Fig 4 the hardness ofX-52 steel increased with increasing plastic deformation
Loss of ductilityThe remaining ductility of all the prestrained samples issummarised in Fig 5 As expected the ductility of thematerial was reduced because of prior deformation Itcould be concluded that the ductility decreased with theincrease of the amount of prestrain and for the sampleswith the same amount of prestrain the ductility tested inair was always higher than that tested in C-1 solutionThe reason for this difference could be explained byconsidering the quasi-cleavage failure observed when thematerial was exposed to C-1 solution Since quasi-cleavage is associated with hydrogen influenced crack-ing the fracture was being influenced by hydrogen whena slow strain rate test was carried out in C-1 solution
Stress corrosion cracking susceptibilityassessmentA comparison of the reduction in area RASCCRAair wasused as the indicator of susceptibility to NNPHSCCThe magnitude of the failure ratio indicates the SCCseverity with a decreasing value indicating more severeSCC
The failure ratios from the SSRT versus amount ofadditional deformation are plotted in Fig 6 The samplewith the highest deformation (8) had the least
3 Yield strength of deformed samples tested in air and in
C-1 solution
4 Hardness of deformed samples (applied load 2 kg
loading time 20 s)
5 Remaining ductility (EL) of various amounts of pre-
strained samples (initial gauge length is 254 mm)
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 299
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
resistance to NNPHSCC as indicated by having thelowest failure ratios of 629 Conversely the 0prestrained sample had the greatest resistance to SCCwith a failure ratio of 669 The resistance to SCCdecreased monotonically as the prestrain increased
The variation in the severity of the SCC with avariation in the amount of prestrain indicates that SCCsusceptibility is a function of prior plastic deformationAs the prior plastic deformation increased so too didthe SCC susceptibility
The reproducibility of this SCC susceptibility assess-ment was checked by another two samples (0 and 4)immersed in C-1 solution The error bars of thereduction in area ratio showing in Fig 6 imply thatthe reproducibility of this SSRT method is reasonableand it can be used to assess SCC susceptibility ofprestrained samples
Potential drop profileA typical potential drop profile is shown in Fig 7 Asillustrated on this curve during the initial 24 h or so(stage I) the change in potential drop was extremelyfast This part of the curve was eliminated since it didnot reflect actual crack growth but rather the establish-ment of equilibrium conditions in the solution A similarlarge change is observed even in C-T samples that arenot cracking (no fatigue starter notch) ie the change in
potential is associated not with cracking but with theequilibration of the specimen in the corrosive solutionFor the stages after stage I where DKIlt25 and20 MPa m12 the curves obtained from potential droptests for all the samples were relatively linear Thisindicated that the crack growth length could be relatedto the measured potential drop directly However forthe 0 sample at the first stage where DKIlt15 MPa m12 the data were more scattered It is believedthat this problem was caused by temperature controldifficulties These temperature control difficultiesoccurred because of an air conditioning failure in thebuilding that changed the room temperature variationduring the day which influenced our ability to controlthe sample temperature to within 01uC as required forthis technique In order to eliminate this scatter morecycles were used for this stage which allowed crackgrowth measurement to be made from the linear part ofthe curve From the point of consistency more cycleswere also given to the second stage with DKIlt15 MPa m12 of the 0 sample and all the stages withDKIlt15 MPa m12 of the other three samples
Crack growth measurementsAfter the test the sample was broken open using liquidnitrogen to permit crack growth measurement by opticalmicroscopy As shown in Fig 8 the region between theprefatigue boundary and crack tip boundary was thecrack growth area due to NNPHSCC The crack growthlength could be measured by averaging three points suchas A B and C as shown in Fig 8
Crack growth rateAccording to Johnson11 and McIntyre12 within therelative crack size range (aW) of 03 to 05 when theratio of the crack length increment to the initial cracksize (Daa0) is less than 02 the crack length incrementcan be quite reasonably related to the potential drop bya linear relationship
The crack growth rate for each DKI stage wasobtained for each sample based on this linearity Thecalculated average crack growth rates are given inFig 9
In Fig 9 when DKI was larger than 14 MPa m12 thecrack growth rate increased with per cent deformationand the crack growth rate in terms of mmcycle was in
6 Stress corrosion cracking performances of various
amounts of prestrained samples in C-1 solution
7 Potential drop curves of 0 prestrained sample at dif-
ferent DKI
8 Fractograghs of SEM showing crack growth of 0 pre-
strained sample
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
300 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
the range of 4061025 to 13661023 mmcycle Thegrowth rate fell into the same range that Harle andBeavers obtained13 Their crack growth rates also basedon potential drop measurements in API X-65 line pipesteel ranged from 1025 to 1024 mmcycle It is clear thatcrack velocity increased by about a factor of 2 withdeformations around 10
Fractographic observationSlow strain rate testing
The fracture surfaces of the prestrained samples failed inC-1 solution were examined using SEM and the results
are shown in Fig 10 All of these pictures were takennear the outside edges of the fracture surfaces
Figure 10a is the fracture surface of the 0 prestrainsample showing a large proportion of quasi-cleavagefacets The size of the biggest facet in the centre of theimage was around 35 mm as shown by the arrow Alsothe 2 sample as shown in Fig 10b had a quasi-cleavage facet about 40 mm in diameter
Corrosion fatigue crack advance measured by potentialdrop
The fracture features examined by the SEM are shownin Fig 11 The fracture surfaces of all the samplesdisplayed quasi-cleavage and secondary cracks
Electrochemical characteristicsFigure 12 shows the polarisation curves of samples withdifferent amounts of plastic deformation measured in C-1 solution at room temperature The figure demon-strated that the OCP of X-52 steel in near neutral pHenvironment was around 2740 mV and there was noactive to passive transition This confirms previousobservations14ndash16 Therefore during the crack growthprocess dissolution was not localised and there was nopassivation on the crack walls to prevent lateraldissolution and retain the crack shape
Discussion
Stress corrosion cracking occurrenceFigure 13 shows the surface cracks in samples that havebeen cold worked different amounts The flat surfacecracks which are perpendicular to the stress axis are
9 Crack growth rates of prestrained samples at different
DKI conditions
a 0 sample b 2 sample10 Images (SEM) of fracture surfaces showing quasi-cleavage
a 0 b 1011 Image (SEM) showing quasi-cleavage on fracture surface of prestrained samples
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 301
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
thought to occur at the locations with the highest tensilestress rather than being caused by the shear stress whichis associated with slip This has been taken by previousworkers17 to indicate that the local fracture was causedby NNPHSCC
Also as shown in Figs 10 and 11 there are somesecondary cracks on the fracture surfaces Both quasi-cleavage and secondary cracks are indications of theoccurrence of NNPHSCC
Increased crack growth rate with increasedyield strengthFigure 14 shows the crack growth rate at a testfrequency of 00025 Hz plotted against the yieldstrength for three different DKI conditions The yieldstrength increase of the 10 sample compared to the 0sample was about 41 However when DKI was around15 20 and 25 MPa m12 the crack growth rates of the10 sample severely increased to 22 22 and 23 timesrespectively compared to those of the 0 sample
Accelerated crack growth rate by severe loadingconditionIn Fig 9 for the same sample with increasing DKI thecrack growth rate increases which means that moresevere loading condition gives much higher crack growthrates From the slopes of the trend lines of each loadingstage in Fig 14 the increase rate of crack growthvelocity under DKI lt25 MPa m12 was eight times thatunder DKIlt15 MPa m12 This implies that higher DKI
gave much greater crack growth rate increase withhigher yield strength
Effect of plastic deformation on corrosion rateand OCPThe OCP and corrosion current densities of the steelsdetermined from the polarisation curves are plottedagainst per cent prestrain in Fig 15 It is seen that theOCP appeared to be not very sensitive to plastic defor-mation (the maximum difference of OCP betweenvarious amounts of prestrained samples was just25 mV) However the corrosion current increasedsignificantly and monotonically for samples withincreasing plastic deformation The total increase of
12 Polarisation curves of samples with prestrained sam-
ples in C-1 solution
a 0 b 813 Surface cracks near failed ends of prestrained samples tested in C-1 solution
14 Crack growth rates against yield strength of pre-
strained samples at different DKI conditions in near
neutral pH environment
15 Corrosion rates and OCPs of samples with various
amounts of plastic deformation
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
302 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
corrosion current density was around 17(130 mA cm22)
Enhanced enrichment of hydrogenThe major effect of cold work is an increase indislocation density the dislocation density in a metalthat has been highly deformed may be as high as1010 mm2218 This would be expected to produce anincrease in the number of hydrogen trapping sites1920
and an increase in the hydrogen adsorption coverage onthe surface
The increase in dislocation density would produce anincrease in the yield stress which increases the hydro-static stress fields around the crack tip This means thata higher concentration of hydrogen can be attracted tothe crack tip This enrichment is caused by the fact thathydrogen is attracted to the crack tip in proportion tothe hydrostatic stress parameter p which is the averageof the principal stresses Gu et al16 also suggested thathydrostatic stress could induce more hydrogen accumu-lation On the other hand the higher yield strengthcould produce a smaller plastic zone size at the crack tipand lower ductility This combination would producemore cracking in near neutral pH environment
Accelerated anodic dissolutionWhen steels are plastically deformed some fraction ofthe deformation energy (y5) is retained internallyThe major portion of this stored energy is as strainenergy associated with dislocations18 The increase in theamount of strain energy would allow easier dissolutionon specific sites since this energy is released on dis-solution This causes an increase in the corrosion ratewith an increase in the amount of cold work
In addition when hydrogen is trapped in the defectsin steel the lattice is dilated and the interatomiccohesion is decreased It has been estimated that 1hydrogen concentration could reduce the atomic bond-ing energy by 5 for steels21 So absorbed hydrogen canaccelerate the corrosion process16 22
To date it is broadly accepted that dissolvedhydrogen and anodic dissolution are the two mainfactors contributing to NNPHSCC23ndash25 Based on all theabove analysis the SCC susceptibility of materials withmore plastic deformation would be higher since bothprocesses are enhanced by plastic deformation
Conclusions1 The OCP of X-52 pipeline steel measured in the de-
aerated C-1 solution was in the range of 2730 to2760 mV (SCE) and it appeared to be insensitive to theprior plastic deformation However the corrosioncurrent density increased with the amount of priorplastic deformation when the steel was exposed to nearneutral pH environment Ten per cent additional priordeformation increased the corrosion current density byabout 17
2 Increasing DKI from 15 to 25 MPa m12 increasedSCC velocity by nearly an order of magnitude
3 Plasticity had a significant effect on the suscept-ibility of X-52 pipeline steel to NNPHSCC the materialwith more plastic deformation had higher SCC velocity(10 deformation increased the crack velocity by morethan two times and decreased the ratio of area reductionin the solution to that in air from 67 to 63
Acknowledgement
This work was funded by Alberta Energy ResearchInstitute of Canada through COURSE program and byNSERC of Canada
References1 B S Delanty and J OrsquoBeirne Oil Gas J 1992 15 39ndash44
2 lsquoSCC on Canadian oil and gas pipelinesrsquo Report no MH-2-95
National Energy Board Calgary Alberta Canada 1996
3 M Baker Jr Inc lsquoStress corrosion cracking studyrsquo Integrity
Management Program Delivery Order DTRS56-02-D-70036
TTO8 Department of Transportation Research and Special
Programs Administration Office of Pipeline Safety January 2005
4 T M Ahmed S B Lambert R Sutherby and A Plumtree
Corrosion 1997 53 581ndash590
5 B N Leis Proc 1st Pipeline Technology Conf on lsquoUpdate on
SCC life prediction models for pipelinesrsquo Ostende Belgium
October 1990
6 CEPA Proc MH-2-95 Vol 2 Appendix D Table 1 1996
Calgary Alta the National Energy Board
7 R N Parkins Proc NACE Northern Area Eastern Conf Ottawa
Ont Canada October 1999 NACE 1ndash22
8 W Chen R L Eadie and R L Sutherby Proc 2nd Int Conf on
lsquoEnvironment-induced cracking of metalsrsquo Banff Alta Canada
September 2004 Elsevier vol 2 211
9 J Been R Eadie and R Sutherby Proc Int Pipeline Conf
Calgary Alta Canada September 2006 ASME Paper IPC2006ndash
10345
10 T L Anderson lsquoFracture mechanics fundamental and applica-
tionsrsquo 2nd edn 1995 Boca Raton FL CRC Press
11 H H Johnson Mater Res Stand 1965 5 (9) 442ndash445
12 P McIntyre J Soc Environm Eng 1974 13ndash3 3ndash7
13 B A Harle and J A Beavers Corrosion 1993 49 861ndash863
14 R N Parkins Proc 37th Conf of Metallurgist lsquoMaterials for
resource recovery and transportrsquo (ed L Collins) 35ndash49 1998
Montreal The Metallurgical Society of CIM
15 J A Beavers C L Durr and S S Shademan Proc 37th Conf of
Metallurgist lsquoMaterials for resource recovery and transportrsquo (ed
L Collins) 51ndash69 1998 Montreal The Metallurgical Society of
CIM
16 B Gu J Luo and X Mao Corrosion 1999 55 96ndash106
17 W Chen F King T R Jack and M J Wilmott Metall Mater
Trans A 2002 33A 1429ndash1436
18 R W Hertzberg lsquoDeformation and fracture mechanics of
engineering materialsrsquo 4th edn 85 1996 Hoboken NJ John
Wiley amp Sons Inc
19 S X Xie and J P Hirth Corrosion 1982 38 486ndash493
20 H Huang and W J D Shaw Corrosion 1995 51 30ndash36
21 T McMullen M J Stott and E Zaremba Physical Review B
1987 35 1076ndash1081
22 J T Bulger lsquoThe Effect of microstructure on near-neutral pH
SCCrsquo MSc thesis University of Alberta Alta Canada 2000 137
23 R N Parkins W K Blanchard Jr and B S Delanty Corrosion
1994 50 394ndash410
24 L J Qiao J L Luo and X Mao J Mater Sci Lett 1997 16
516ndash520
25 B T Lu J L Luo and B McCrady Proc Int Pipeline Conf
Calgary Alta Canada SeptemberndashOctober 2002 ASME Paper
IPC2002ndash27234
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 303
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
investigation The chemical composition of the steel isFendash0261Cndash1150Mnndash0008Pndash0019Sndash0036Sindash0033Cundash0002Snndash0032Nindash0027Crndash0013Mondash0002Alndash0003Vndash0003Nbndash0002Tindash0006Co (wt-) The microstructure consistedof fine pearlite and ferrite with a grain size of 10ndash20 mmThe specified minimum yield stress for the material was3590 MPa (X-52) and the ultimate tensile strength was589 MPa
Test SolutionNS4 synthetic solution has been widely used to simulatethe soil solution in the study of NNPHSCC behaviourHowever another synthetic soil solution termed C-1with composition shown in Table 1 was designed basedon field data from pipeline failure sites In previousresearch8 this C-1 solution was shown to be moreaggressive in causing NNPHSCC than NS4 solution andwhen bubbled with 5CO2 exhibited a slightly lowerpH The C-1 solution was prepared using deionisedwater and was bubbled with 5CO2 in oxygen free N2
gas for at least 24 h before being introduced to the testcells The bubbling gas was also maintained during eachtest in order to keep the pH value of the solution at 589and also to get an anaerobic environment Anaerobicenvironments have been identified as being associatedwith NNPHSCC12
Slow strain rate testing (SSRT)Slow strain rate testing were carried out on smoothcylindrical tensile samples which are shown in Fig 1The length direction of the sample was parallel to thecircumferential direction of the pipeline steel in order toensure that the subsequent crack growth was in thelongitudinal direction of the pipe as is typically observedin the field After they were machined from the pipeusing electrical discharging machining (EDM) thesamples were prestrained in an Instron hydraulic testframe at 002 mm min21 Different amounts of plasticdeformation from 1 to 8 were obtained as determinedby an extensometer The sample surface in the gaugesection was polished to a 600 grit finish in an orientationparallel to the subsequent loading direction of the
SSRT This ensured similar surface conditions for alltests
The SSRT was conducted on samples in bothdeaerated C-1 solution at open circuit potential (OCP)and in air The strain rate for samples tested in C-1solution was normally begun at a crosshead speed of00004 mm min21 which corresponded to an initialelastic strain rate of 26261027 s21 while a crossheadspeed of 005 mm min21 was used for the samples testedin air (1256 faster) which corresponded to a strain rateof 32861025 s21 The faster rate can be used since it isknown that results in air are unaffected by the change instrain rate The load elongation behaviour was recordedAfter the test was complete the fractured sample wasimmediately removed ultrasonically cleaned usingacetone and placed in a desiccator for storage untilfuture SEM examination
Corrosion fatigue testing using potential drop tomeasure crack advanceThe samples cut from the longitudinal direction of thepipeline were cold rolled to different amounts of plasticdeformation (5 8 10) Then the compact toughness(C-T) specimens (as shown in Fig 2) were made usingEDM according to ASTM Standard E813-89 except forsample thickness B which was limited by the maximumpipe wall thickness that could be utilised A sharp notchwas introduced on the sample along the longitudinaldirection with the notch radius being 015 mm The C-Tsamples were prefatigued in air using an INSTRONmachine to produce a sharp crack from the machinednotch tip in accordance with ASTM E647 Theprefatigued crack length on both surfaces was controlledto be 2ndash3 mm long The difference in crack length on thetwo sides of each sample was less than 02 mm
EnduraTEC equipment was used for cyclic strainingduring the potential drop tests A sine waveform loadwas used with a loading frequency of 00025 Hz (216cyclesday) and R ratios of 0375 05 and 0625 Thistest frequency is known to be similar to strain rates in oilpipelines but somewhat fast for gas pipelines9 Themaximum and minimum stresses were controlled toachieve a maximum stress intensity factor K close to40 MPa m12 and a DK in the range 15 to 25 MPa m12
Table 1 Chemical components of NS4 and C-1 solution (bubbled with 5CO2zN2 balanced)
Composition g L21 MgSO47H2O CaCl2 KCl NaHCO3 CaCO3 pH
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
298 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd After the test the samples were broken open by
cooling the sample in liquid nitrogen and fracturing itwith a sudden crack opening load The crack faces werecleaned ultrasonically in acetone solution and thecleaned fracture surfaces were examined by opticalmicroscopy to measure the actual crack growth lengthwhich corresponded to the total potential changerecorded This gave the correspondence between crackgrowth length and potential change
The stress intensity factor KI could be calculatedaccording to ASTM 399-90 (Ref 10)
fa
w
~
KIBw12
P~
1a
w
32
2za
w
0886z464 aw1332 a
w
2z
1472 aw
356 a
w
4
24
35
8lt
9=
where a and B are crack length and sample thicknessrespectively w is the width of the sample and P is theapplied load
Electrochemical testsA smooth round tensile specimen as shown in Fig 1was also used for potentiodynamic polarisation scansThe polarisation curve was obtained at a scan rate of02 mV s21 starting approximately 250 mV more nega-tive than the OCP and scanning in the noble direction toa potential 650 mV more positive than OCP
Results
Increase of yield strengthFigure 3 shows the yield strength of the samples withdifferent amounts of plastic deformation The yieldstrength of X-52 steel increased with increasing plasticdeformation for samples tested in air and also for thosetested in C-1 solution There is a small reduction in yieldstrength when testing in C-1 solution compared to air
Increase of hardnessResults of the average of three hardness measurementson each sample are shown in Fig 4 the hardness ofX-52 steel increased with increasing plastic deformation
Loss of ductilityThe remaining ductility of all the prestrained samples issummarised in Fig 5 As expected the ductility of thematerial was reduced because of prior deformation Itcould be concluded that the ductility decreased with theincrease of the amount of prestrain and for the sampleswith the same amount of prestrain the ductility tested inair was always higher than that tested in C-1 solutionThe reason for this difference could be explained byconsidering the quasi-cleavage failure observed when thematerial was exposed to C-1 solution Since quasi-cleavage is associated with hydrogen influenced crack-ing the fracture was being influenced by hydrogen whena slow strain rate test was carried out in C-1 solution
Stress corrosion cracking susceptibilityassessmentA comparison of the reduction in area RASCCRAair wasused as the indicator of susceptibility to NNPHSCCThe magnitude of the failure ratio indicates the SCCseverity with a decreasing value indicating more severeSCC
The failure ratios from the SSRT versus amount ofadditional deformation are plotted in Fig 6 The samplewith the highest deformation (8) had the least
3 Yield strength of deformed samples tested in air and in
C-1 solution
4 Hardness of deformed samples (applied load 2 kg
loading time 20 s)
5 Remaining ductility (EL) of various amounts of pre-
strained samples (initial gauge length is 254 mm)
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 299
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
resistance to NNPHSCC as indicated by having thelowest failure ratios of 629 Conversely the 0prestrained sample had the greatest resistance to SCCwith a failure ratio of 669 The resistance to SCCdecreased monotonically as the prestrain increased
The variation in the severity of the SCC with avariation in the amount of prestrain indicates that SCCsusceptibility is a function of prior plastic deformationAs the prior plastic deformation increased so too didthe SCC susceptibility
The reproducibility of this SCC susceptibility assess-ment was checked by another two samples (0 and 4)immersed in C-1 solution The error bars of thereduction in area ratio showing in Fig 6 imply thatthe reproducibility of this SSRT method is reasonableand it can be used to assess SCC susceptibility ofprestrained samples
Potential drop profileA typical potential drop profile is shown in Fig 7 Asillustrated on this curve during the initial 24 h or so(stage I) the change in potential drop was extremelyfast This part of the curve was eliminated since it didnot reflect actual crack growth but rather the establish-ment of equilibrium conditions in the solution A similarlarge change is observed even in C-T samples that arenot cracking (no fatigue starter notch) ie the change in
potential is associated not with cracking but with theequilibration of the specimen in the corrosive solutionFor the stages after stage I where DKIlt25 and20 MPa m12 the curves obtained from potential droptests for all the samples were relatively linear Thisindicated that the crack growth length could be relatedto the measured potential drop directly However forthe 0 sample at the first stage where DKIlt15 MPa m12 the data were more scattered It is believedthat this problem was caused by temperature controldifficulties These temperature control difficultiesoccurred because of an air conditioning failure in thebuilding that changed the room temperature variationduring the day which influenced our ability to controlthe sample temperature to within 01uC as required forthis technique In order to eliminate this scatter morecycles were used for this stage which allowed crackgrowth measurement to be made from the linear part ofthe curve From the point of consistency more cycleswere also given to the second stage with DKIlt15 MPa m12 of the 0 sample and all the stages withDKIlt15 MPa m12 of the other three samples
Crack growth measurementsAfter the test the sample was broken open using liquidnitrogen to permit crack growth measurement by opticalmicroscopy As shown in Fig 8 the region between theprefatigue boundary and crack tip boundary was thecrack growth area due to NNPHSCC The crack growthlength could be measured by averaging three points suchas A B and C as shown in Fig 8
Crack growth rateAccording to Johnson11 and McIntyre12 within therelative crack size range (aW) of 03 to 05 when theratio of the crack length increment to the initial cracksize (Daa0) is less than 02 the crack length incrementcan be quite reasonably related to the potential drop bya linear relationship
The crack growth rate for each DKI stage wasobtained for each sample based on this linearity Thecalculated average crack growth rates are given inFig 9
In Fig 9 when DKI was larger than 14 MPa m12 thecrack growth rate increased with per cent deformationand the crack growth rate in terms of mmcycle was in
6 Stress corrosion cracking performances of various
amounts of prestrained samples in C-1 solution
7 Potential drop curves of 0 prestrained sample at dif-
ferent DKI
8 Fractograghs of SEM showing crack growth of 0 pre-
strained sample
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
300 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
the range of 4061025 to 13661023 mmcycle Thegrowth rate fell into the same range that Harle andBeavers obtained13 Their crack growth rates also basedon potential drop measurements in API X-65 line pipesteel ranged from 1025 to 1024 mmcycle It is clear thatcrack velocity increased by about a factor of 2 withdeformations around 10
Fractographic observationSlow strain rate testing
The fracture surfaces of the prestrained samples failed inC-1 solution were examined using SEM and the results
are shown in Fig 10 All of these pictures were takennear the outside edges of the fracture surfaces
Figure 10a is the fracture surface of the 0 prestrainsample showing a large proportion of quasi-cleavagefacets The size of the biggest facet in the centre of theimage was around 35 mm as shown by the arrow Alsothe 2 sample as shown in Fig 10b had a quasi-cleavage facet about 40 mm in diameter
Corrosion fatigue crack advance measured by potentialdrop
The fracture features examined by the SEM are shownin Fig 11 The fracture surfaces of all the samplesdisplayed quasi-cleavage and secondary cracks
Electrochemical characteristicsFigure 12 shows the polarisation curves of samples withdifferent amounts of plastic deformation measured in C-1 solution at room temperature The figure demon-strated that the OCP of X-52 steel in near neutral pHenvironment was around 2740 mV and there was noactive to passive transition This confirms previousobservations14ndash16 Therefore during the crack growthprocess dissolution was not localised and there was nopassivation on the crack walls to prevent lateraldissolution and retain the crack shape
Discussion
Stress corrosion cracking occurrenceFigure 13 shows the surface cracks in samples that havebeen cold worked different amounts The flat surfacecracks which are perpendicular to the stress axis are
9 Crack growth rates of prestrained samples at different
DKI conditions
a 0 sample b 2 sample10 Images (SEM) of fracture surfaces showing quasi-cleavage
a 0 b 1011 Image (SEM) showing quasi-cleavage on fracture surface of prestrained samples
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 301
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
thought to occur at the locations with the highest tensilestress rather than being caused by the shear stress whichis associated with slip This has been taken by previousworkers17 to indicate that the local fracture was causedby NNPHSCC
Also as shown in Figs 10 and 11 there are somesecondary cracks on the fracture surfaces Both quasi-cleavage and secondary cracks are indications of theoccurrence of NNPHSCC
Increased crack growth rate with increasedyield strengthFigure 14 shows the crack growth rate at a testfrequency of 00025 Hz plotted against the yieldstrength for three different DKI conditions The yieldstrength increase of the 10 sample compared to the 0sample was about 41 However when DKI was around15 20 and 25 MPa m12 the crack growth rates of the10 sample severely increased to 22 22 and 23 timesrespectively compared to those of the 0 sample
Accelerated crack growth rate by severe loadingconditionIn Fig 9 for the same sample with increasing DKI thecrack growth rate increases which means that moresevere loading condition gives much higher crack growthrates From the slopes of the trend lines of each loadingstage in Fig 14 the increase rate of crack growthvelocity under DKI lt25 MPa m12 was eight times thatunder DKIlt15 MPa m12 This implies that higher DKI
gave much greater crack growth rate increase withhigher yield strength
Effect of plastic deformation on corrosion rateand OCPThe OCP and corrosion current densities of the steelsdetermined from the polarisation curves are plottedagainst per cent prestrain in Fig 15 It is seen that theOCP appeared to be not very sensitive to plastic defor-mation (the maximum difference of OCP betweenvarious amounts of prestrained samples was just25 mV) However the corrosion current increasedsignificantly and monotonically for samples withincreasing plastic deformation The total increase of
12 Polarisation curves of samples with prestrained sam-
ples in C-1 solution
a 0 b 813 Surface cracks near failed ends of prestrained samples tested in C-1 solution
14 Crack growth rates against yield strength of pre-
strained samples at different DKI conditions in near
neutral pH environment
15 Corrosion rates and OCPs of samples with various
amounts of plastic deformation
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
302 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
corrosion current density was around 17(130 mA cm22)
Enhanced enrichment of hydrogenThe major effect of cold work is an increase indislocation density the dislocation density in a metalthat has been highly deformed may be as high as1010 mm2218 This would be expected to produce anincrease in the number of hydrogen trapping sites1920
and an increase in the hydrogen adsorption coverage onthe surface
The increase in dislocation density would produce anincrease in the yield stress which increases the hydro-static stress fields around the crack tip This means thata higher concentration of hydrogen can be attracted tothe crack tip This enrichment is caused by the fact thathydrogen is attracted to the crack tip in proportion tothe hydrostatic stress parameter p which is the averageof the principal stresses Gu et al16 also suggested thathydrostatic stress could induce more hydrogen accumu-lation On the other hand the higher yield strengthcould produce a smaller plastic zone size at the crack tipand lower ductility This combination would producemore cracking in near neutral pH environment
Accelerated anodic dissolutionWhen steels are plastically deformed some fraction ofthe deformation energy (y5) is retained internallyThe major portion of this stored energy is as strainenergy associated with dislocations18 The increase in theamount of strain energy would allow easier dissolutionon specific sites since this energy is released on dis-solution This causes an increase in the corrosion ratewith an increase in the amount of cold work
In addition when hydrogen is trapped in the defectsin steel the lattice is dilated and the interatomiccohesion is decreased It has been estimated that 1hydrogen concentration could reduce the atomic bond-ing energy by 5 for steels21 So absorbed hydrogen canaccelerate the corrosion process16 22
To date it is broadly accepted that dissolvedhydrogen and anodic dissolution are the two mainfactors contributing to NNPHSCC23ndash25 Based on all theabove analysis the SCC susceptibility of materials withmore plastic deformation would be higher since bothprocesses are enhanced by plastic deformation
Conclusions1 The OCP of X-52 pipeline steel measured in the de-
aerated C-1 solution was in the range of 2730 to2760 mV (SCE) and it appeared to be insensitive to theprior plastic deformation However the corrosioncurrent density increased with the amount of priorplastic deformation when the steel was exposed to nearneutral pH environment Ten per cent additional priordeformation increased the corrosion current density byabout 17
2 Increasing DKI from 15 to 25 MPa m12 increasedSCC velocity by nearly an order of magnitude
3 Plasticity had a significant effect on the suscept-ibility of X-52 pipeline steel to NNPHSCC the materialwith more plastic deformation had higher SCC velocity(10 deformation increased the crack velocity by morethan two times and decreased the ratio of area reductionin the solution to that in air from 67 to 63
Acknowledgement
This work was funded by Alberta Energy ResearchInstitute of Canada through COURSE program and byNSERC of Canada
References1 B S Delanty and J OrsquoBeirne Oil Gas J 1992 15 39ndash44
2 lsquoSCC on Canadian oil and gas pipelinesrsquo Report no MH-2-95
National Energy Board Calgary Alberta Canada 1996
3 M Baker Jr Inc lsquoStress corrosion cracking studyrsquo Integrity
Management Program Delivery Order DTRS56-02-D-70036
TTO8 Department of Transportation Research and Special
Programs Administration Office of Pipeline Safety January 2005
4 T M Ahmed S B Lambert R Sutherby and A Plumtree
Corrosion 1997 53 581ndash590
5 B N Leis Proc 1st Pipeline Technology Conf on lsquoUpdate on
SCC life prediction models for pipelinesrsquo Ostende Belgium
October 1990
6 CEPA Proc MH-2-95 Vol 2 Appendix D Table 1 1996
Calgary Alta the National Energy Board
7 R N Parkins Proc NACE Northern Area Eastern Conf Ottawa
Ont Canada October 1999 NACE 1ndash22
8 W Chen R L Eadie and R L Sutherby Proc 2nd Int Conf on
lsquoEnvironment-induced cracking of metalsrsquo Banff Alta Canada
September 2004 Elsevier vol 2 211
9 J Been R Eadie and R Sutherby Proc Int Pipeline Conf
Calgary Alta Canada September 2006 ASME Paper IPC2006ndash
10345
10 T L Anderson lsquoFracture mechanics fundamental and applica-
tionsrsquo 2nd edn 1995 Boca Raton FL CRC Press
11 H H Johnson Mater Res Stand 1965 5 (9) 442ndash445
12 P McIntyre J Soc Environm Eng 1974 13ndash3 3ndash7
13 B A Harle and J A Beavers Corrosion 1993 49 861ndash863
14 R N Parkins Proc 37th Conf of Metallurgist lsquoMaterials for
resource recovery and transportrsquo (ed L Collins) 35ndash49 1998
Montreal The Metallurgical Society of CIM
15 J A Beavers C L Durr and S S Shademan Proc 37th Conf of
Metallurgist lsquoMaterials for resource recovery and transportrsquo (ed
L Collins) 51ndash69 1998 Montreal The Metallurgical Society of
CIM
16 B Gu J Luo and X Mao Corrosion 1999 55 96ndash106
17 W Chen F King T R Jack and M J Wilmott Metall Mater
Trans A 2002 33A 1429ndash1436
18 R W Hertzberg lsquoDeformation and fracture mechanics of
engineering materialsrsquo 4th edn 85 1996 Hoboken NJ John
Wiley amp Sons Inc
19 S X Xie and J P Hirth Corrosion 1982 38 486ndash493
20 H Huang and W J D Shaw Corrosion 1995 51 30ndash36
21 T McMullen M J Stott and E Zaremba Physical Review B
1987 35 1076ndash1081
22 J T Bulger lsquoThe Effect of microstructure on near-neutral pH
SCCrsquo MSc thesis University of Alberta Alta Canada 2000 137
23 R N Parkins W K Blanchard Jr and B S Delanty Corrosion
1994 50 394ndash410
24 L J Qiao J L Luo and X Mao J Mater Sci Lett 1997 16
516ndash520
25 B T Lu J L Luo and B McCrady Proc Int Pipeline Conf
Calgary Alta Canada SeptemberndashOctober 2002 ASME Paper
IPC2002ndash27234
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 303
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd After the test the samples were broken open by
cooling the sample in liquid nitrogen and fracturing itwith a sudden crack opening load The crack faces werecleaned ultrasonically in acetone solution and thecleaned fracture surfaces were examined by opticalmicroscopy to measure the actual crack growth lengthwhich corresponded to the total potential changerecorded This gave the correspondence between crackgrowth length and potential change
The stress intensity factor KI could be calculatedaccording to ASTM 399-90 (Ref 10)
fa
w
~
KIBw12
P~
1a
w
32
2za
w
0886z464 aw1332 a
w
2z
1472 aw
356 a
w
4
24
35
8lt
9=
where a and B are crack length and sample thicknessrespectively w is the width of the sample and P is theapplied load
Electrochemical testsA smooth round tensile specimen as shown in Fig 1was also used for potentiodynamic polarisation scansThe polarisation curve was obtained at a scan rate of02 mV s21 starting approximately 250 mV more nega-tive than the OCP and scanning in the noble direction toa potential 650 mV more positive than OCP
Results
Increase of yield strengthFigure 3 shows the yield strength of the samples withdifferent amounts of plastic deformation The yieldstrength of X-52 steel increased with increasing plasticdeformation for samples tested in air and also for thosetested in C-1 solution There is a small reduction in yieldstrength when testing in C-1 solution compared to air
Increase of hardnessResults of the average of three hardness measurementson each sample are shown in Fig 4 the hardness ofX-52 steel increased with increasing plastic deformation
Loss of ductilityThe remaining ductility of all the prestrained samples issummarised in Fig 5 As expected the ductility of thematerial was reduced because of prior deformation Itcould be concluded that the ductility decreased with theincrease of the amount of prestrain and for the sampleswith the same amount of prestrain the ductility tested inair was always higher than that tested in C-1 solutionThe reason for this difference could be explained byconsidering the quasi-cleavage failure observed when thematerial was exposed to C-1 solution Since quasi-cleavage is associated with hydrogen influenced crack-ing the fracture was being influenced by hydrogen whena slow strain rate test was carried out in C-1 solution
Stress corrosion cracking susceptibilityassessmentA comparison of the reduction in area RASCCRAair wasused as the indicator of susceptibility to NNPHSCCThe magnitude of the failure ratio indicates the SCCseverity with a decreasing value indicating more severeSCC
The failure ratios from the SSRT versus amount ofadditional deformation are plotted in Fig 6 The samplewith the highest deformation (8) had the least
3 Yield strength of deformed samples tested in air and in
C-1 solution
4 Hardness of deformed samples (applied load 2 kg
loading time 20 s)
5 Remaining ductility (EL) of various amounts of pre-
strained samples (initial gauge length is 254 mm)
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 299
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
resistance to NNPHSCC as indicated by having thelowest failure ratios of 629 Conversely the 0prestrained sample had the greatest resistance to SCCwith a failure ratio of 669 The resistance to SCCdecreased monotonically as the prestrain increased
The variation in the severity of the SCC with avariation in the amount of prestrain indicates that SCCsusceptibility is a function of prior plastic deformationAs the prior plastic deformation increased so too didthe SCC susceptibility
The reproducibility of this SCC susceptibility assess-ment was checked by another two samples (0 and 4)immersed in C-1 solution The error bars of thereduction in area ratio showing in Fig 6 imply thatthe reproducibility of this SSRT method is reasonableand it can be used to assess SCC susceptibility ofprestrained samples
Potential drop profileA typical potential drop profile is shown in Fig 7 Asillustrated on this curve during the initial 24 h or so(stage I) the change in potential drop was extremelyfast This part of the curve was eliminated since it didnot reflect actual crack growth but rather the establish-ment of equilibrium conditions in the solution A similarlarge change is observed even in C-T samples that arenot cracking (no fatigue starter notch) ie the change in
potential is associated not with cracking but with theequilibration of the specimen in the corrosive solutionFor the stages after stage I where DKIlt25 and20 MPa m12 the curves obtained from potential droptests for all the samples were relatively linear Thisindicated that the crack growth length could be relatedto the measured potential drop directly However forthe 0 sample at the first stage where DKIlt15 MPa m12 the data were more scattered It is believedthat this problem was caused by temperature controldifficulties These temperature control difficultiesoccurred because of an air conditioning failure in thebuilding that changed the room temperature variationduring the day which influenced our ability to controlthe sample temperature to within 01uC as required forthis technique In order to eliminate this scatter morecycles were used for this stage which allowed crackgrowth measurement to be made from the linear part ofthe curve From the point of consistency more cycleswere also given to the second stage with DKIlt15 MPa m12 of the 0 sample and all the stages withDKIlt15 MPa m12 of the other three samples
Crack growth measurementsAfter the test the sample was broken open using liquidnitrogen to permit crack growth measurement by opticalmicroscopy As shown in Fig 8 the region between theprefatigue boundary and crack tip boundary was thecrack growth area due to NNPHSCC The crack growthlength could be measured by averaging three points suchas A B and C as shown in Fig 8
Crack growth rateAccording to Johnson11 and McIntyre12 within therelative crack size range (aW) of 03 to 05 when theratio of the crack length increment to the initial cracksize (Daa0) is less than 02 the crack length incrementcan be quite reasonably related to the potential drop bya linear relationship
The crack growth rate for each DKI stage wasobtained for each sample based on this linearity Thecalculated average crack growth rates are given inFig 9
In Fig 9 when DKI was larger than 14 MPa m12 thecrack growth rate increased with per cent deformationand the crack growth rate in terms of mmcycle was in
6 Stress corrosion cracking performances of various
amounts of prestrained samples in C-1 solution
7 Potential drop curves of 0 prestrained sample at dif-
ferent DKI
8 Fractograghs of SEM showing crack growth of 0 pre-
strained sample
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
300 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
the range of 4061025 to 13661023 mmcycle Thegrowth rate fell into the same range that Harle andBeavers obtained13 Their crack growth rates also basedon potential drop measurements in API X-65 line pipesteel ranged from 1025 to 1024 mmcycle It is clear thatcrack velocity increased by about a factor of 2 withdeformations around 10
Fractographic observationSlow strain rate testing
The fracture surfaces of the prestrained samples failed inC-1 solution were examined using SEM and the results
are shown in Fig 10 All of these pictures were takennear the outside edges of the fracture surfaces
Figure 10a is the fracture surface of the 0 prestrainsample showing a large proportion of quasi-cleavagefacets The size of the biggest facet in the centre of theimage was around 35 mm as shown by the arrow Alsothe 2 sample as shown in Fig 10b had a quasi-cleavage facet about 40 mm in diameter
Corrosion fatigue crack advance measured by potentialdrop
The fracture features examined by the SEM are shownin Fig 11 The fracture surfaces of all the samplesdisplayed quasi-cleavage and secondary cracks
Electrochemical characteristicsFigure 12 shows the polarisation curves of samples withdifferent amounts of plastic deformation measured in C-1 solution at room temperature The figure demon-strated that the OCP of X-52 steel in near neutral pHenvironment was around 2740 mV and there was noactive to passive transition This confirms previousobservations14ndash16 Therefore during the crack growthprocess dissolution was not localised and there was nopassivation on the crack walls to prevent lateraldissolution and retain the crack shape
Discussion
Stress corrosion cracking occurrenceFigure 13 shows the surface cracks in samples that havebeen cold worked different amounts The flat surfacecracks which are perpendicular to the stress axis are
9 Crack growth rates of prestrained samples at different
DKI conditions
a 0 sample b 2 sample10 Images (SEM) of fracture surfaces showing quasi-cleavage
a 0 b 1011 Image (SEM) showing quasi-cleavage on fracture surface of prestrained samples
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 301
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
thought to occur at the locations with the highest tensilestress rather than being caused by the shear stress whichis associated with slip This has been taken by previousworkers17 to indicate that the local fracture was causedby NNPHSCC
Also as shown in Figs 10 and 11 there are somesecondary cracks on the fracture surfaces Both quasi-cleavage and secondary cracks are indications of theoccurrence of NNPHSCC
Increased crack growth rate with increasedyield strengthFigure 14 shows the crack growth rate at a testfrequency of 00025 Hz plotted against the yieldstrength for three different DKI conditions The yieldstrength increase of the 10 sample compared to the 0sample was about 41 However when DKI was around15 20 and 25 MPa m12 the crack growth rates of the10 sample severely increased to 22 22 and 23 timesrespectively compared to those of the 0 sample
Accelerated crack growth rate by severe loadingconditionIn Fig 9 for the same sample with increasing DKI thecrack growth rate increases which means that moresevere loading condition gives much higher crack growthrates From the slopes of the trend lines of each loadingstage in Fig 14 the increase rate of crack growthvelocity under DKI lt25 MPa m12 was eight times thatunder DKIlt15 MPa m12 This implies that higher DKI
gave much greater crack growth rate increase withhigher yield strength
Effect of plastic deformation on corrosion rateand OCPThe OCP and corrosion current densities of the steelsdetermined from the polarisation curves are plottedagainst per cent prestrain in Fig 15 It is seen that theOCP appeared to be not very sensitive to plastic defor-mation (the maximum difference of OCP betweenvarious amounts of prestrained samples was just25 mV) However the corrosion current increasedsignificantly and monotonically for samples withincreasing plastic deformation The total increase of
12 Polarisation curves of samples with prestrained sam-
ples in C-1 solution
a 0 b 813 Surface cracks near failed ends of prestrained samples tested in C-1 solution
14 Crack growth rates against yield strength of pre-
strained samples at different DKI conditions in near
neutral pH environment
15 Corrosion rates and OCPs of samples with various
amounts of plastic deformation
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
302 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
corrosion current density was around 17(130 mA cm22)
Enhanced enrichment of hydrogenThe major effect of cold work is an increase indislocation density the dislocation density in a metalthat has been highly deformed may be as high as1010 mm2218 This would be expected to produce anincrease in the number of hydrogen trapping sites1920
and an increase in the hydrogen adsorption coverage onthe surface
The increase in dislocation density would produce anincrease in the yield stress which increases the hydro-static stress fields around the crack tip This means thata higher concentration of hydrogen can be attracted tothe crack tip This enrichment is caused by the fact thathydrogen is attracted to the crack tip in proportion tothe hydrostatic stress parameter p which is the averageof the principal stresses Gu et al16 also suggested thathydrostatic stress could induce more hydrogen accumu-lation On the other hand the higher yield strengthcould produce a smaller plastic zone size at the crack tipand lower ductility This combination would producemore cracking in near neutral pH environment
Accelerated anodic dissolutionWhen steels are plastically deformed some fraction ofthe deformation energy (y5) is retained internallyThe major portion of this stored energy is as strainenergy associated with dislocations18 The increase in theamount of strain energy would allow easier dissolutionon specific sites since this energy is released on dis-solution This causes an increase in the corrosion ratewith an increase in the amount of cold work
In addition when hydrogen is trapped in the defectsin steel the lattice is dilated and the interatomiccohesion is decreased It has been estimated that 1hydrogen concentration could reduce the atomic bond-ing energy by 5 for steels21 So absorbed hydrogen canaccelerate the corrosion process16 22
To date it is broadly accepted that dissolvedhydrogen and anodic dissolution are the two mainfactors contributing to NNPHSCC23ndash25 Based on all theabove analysis the SCC susceptibility of materials withmore plastic deformation would be higher since bothprocesses are enhanced by plastic deformation
Conclusions1 The OCP of X-52 pipeline steel measured in the de-
aerated C-1 solution was in the range of 2730 to2760 mV (SCE) and it appeared to be insensitive to theprior plastic deformation However the corrosioncurrent density increased with the amount of priorplastic deformation when the steel was exposed to nearneutral pH environment Ten per cent additional priordeformation increased the corrosion current density byabout 17
2 Increasing DKI from 15 to 25 MPa m12 increasedSCC velocity by nearly an order of magnitude
3 Plasticity had a significant effect on the suscept-ibility of X-52 pipeline steel to NNPHSCC the materialwith more plastic deformation had higher SCC velocity(10 deformation increased the crack velocity by morethan two times and decreased the ratio of area reductionin the solution to that in air from 67 to 63
Acknowledgement
This work was funded by Alberta Energy ResearchInstitute of Canada through COURSE program and byNSERC of Canada
References1 B S Delanty and J OrsquoBeirne Oil Gas J 1992 15 39ndash44
2 lsquoSCC on Canadian oil and gas pipelinesrsquo Report no MH-2-95
National Energy Board Calgary Alberta Canada 1996
3 M Baker Jr Inc lsquoStress corrosion cracking studyrsquo Integrity
Management Program Delivery Order DTRS56-02-D-70036
TTO8 Department of Transportation Research and Special
Programs Administration Office of Pipeline Safety January 2005
4 T M Ahmed S B Lambert R Sutherby and A Plumtree
Corrosion 1997 53 581ndash590
5 B N Leis Proc 1st Pipeline Technology Conf on lsquoUpdate on
SCC life prediction models for pipelinesrsquo Ostende Belgium
October 1990
6 CEPA Proc MH-2-95 Vol 2 Appendix D Table 1 1996
Calgary Alta the National Energy Board
7 R N Parkins Proc NACE Northern Area Eastern Conf Ottawa
Ont Canada October 1999 NACE 1ndash22
8 W Chen R L Eadie and R L Sutherby Proc 2nd Int Conf on
lsquoEnvironment-induced cracking of metalsrsquo Banff Alta Canada
September 2004 Elsevier vol 2 211
9 J Been R Eadie and R Sutherby Proc Int Pipeline Conf
Calgary Alta Canada September 2006 ASME Paper IPC2006ndash
10345
10 T L Anderson lsquoFracture mechanics fundamental and applica-
tionsrsquo 2nd edn 1995 Boca Raton FL CRC Press
11 H H Johnson Mater Res Stand 1965 5 (9) 442ndash445
12 P McIntyre J Soc Environm Eng 1974 13ndash3 3ndash7
13 B A Harle and J A Beavers Corrosion 1993 49 861ndash863
14 R N Parkins Proc 37th Conf of Metallurgist lsquoMaterials for
resource recovery and transportrsquo (ed L Collins) 35ndash49 1998
Montreal The Metallurgical Society of CIM
15 J A Beavers C L Durr and S S Shademan Proc 37th Conf of
Metallurgist lsquoMaterials for resource recovery and transportrsquo (ed
L Collins) 51ndash69 1998 Montreal The Metallurgical Society of
CIM
16 B Gu J Luo and X Mao Corrosion 1999 55 96ndash106
17 W Chen F King T R Jack and M J Wilmott Metall Mater
Trans A 2002 33A 1429ndash1436
18 R W Hertzberg lsquoDeformation and fracture mechanics of
engineering materialsrsquo 4th edn 85 1996 Hoboken NJ John
Wiley amp Sons Inc
19 S X Xie and J P Hirth Corrosion 1982 38 486ndash493
20 H Huang and W J D Shaw Corrosion 1995 51 30ndash36
21 T McMullen M J Stott and E Zaremba Physical Review B
1987 35 1076ndash1081
22 J T Bulger lsquoThe Effect of microstructure on near-neutral pH
SCCrsquo MSc thesis University of Alberta Alta Canada 2000 137
23 R N Parkins W K Blanchard Jr and B S Delanty Corrosion
1994 50 394ndash410
24 L J Qiao J L Luo and X Mao J Mater Sci Lett 1997 16
516ndash520
25 B T Lu J L Luo and B McCrady Proc Int Pipeline Conf
Calgary Alta Canada SeptemberndashOctober 2002 ASME Paper
IPC2002ndash27234
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 303
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
resistance to NNPHSCC as indicated by having thelowest failure ratios of 629 Conversely the 0prestrained sample had the greatest resistance to SCCwith a failure ratio of 669 The resistance to SCCdecreased monotonically as the prestrain increased
The variation in the severity of the SCC with avariation in the amount of prestrain indicates that SCCsusceptibility is a function of prior plastic deformationAs the prior plastic deformation increased so too didthe SCC susceptibility
The reproducibility of this SCC susceptibility assess-ment was checked by another two samples (0 and 4)immersed in C-1 solution The error bars of thereduction in area ratio showing in Fig 6 imply thatthe reproducibility of this SSRT method is reasonableand it can be used to assess SCC susceptibility ofprestrained samples
Potential drop profileA typical potential drop profile is shown in Fig 7 Asillustrated on this curve during the initial 24 h or so(stage I) the change in potential drop was extremelyfast This part of the curve was eliminated since it didnot reflect actual crack growth but rather the establish-ment of equilibrium conditions in the solution A similarlarge change is observed even in C-T samples that arenot cracking (no fatigue starter notch) ie the change in
potential is associated not with cracking but with theequilibration of the specimen in the corrosive solutionFor the stages after stage I where DKIlt25 and20 MPa m12 the curves obtained from potential droptests for all the samples were relatively linear Thisindicated that the crack growth length could be relatedto the measured potential drop directly However forthe 0 sample at the first stage where DKIlt15 MPa m12 the data were more scattered It is believedthat this problem was caused by temperature controldifficulties These temperature control difficultiesoccurred because of an air conditioning failure in thebuilding that changed the room temperature variationduring the day which influenced our ability to controlthe sample temperature to within 01uC as required forthis technique In order to eliminate this scatter morecycles were used for this stage which allowed crackgrowth measurement to be made from the linear part ofthe curve From the point of consistency more cycleswere also given to the second stage with DKIlt15 MPa m12 of the 0 sample and all the stages withDKIlt15 MPa m12 of the other three samples
Crack growth measurementsAfter the test the sample was broken open using liquidnitrogen to permit crack growth measurement by opticalmicroscopy As shown in Fig 8 the region between theprefatigue boundary and crack tip boundary was thecrack growth area due to NNPHSCC The crack growthlength could be measured by averaging three points suchas A B and C as shown in Fig 8
Crack growth rateAccording to Johnson11 and McIntyre12 within therelative crack size range (aW) of 03 to 05 when theratio of the crack length increment to the initial cracksize (Daa0) is less than 02 the crack length incrementcan be quite reasonably related to the potential drop bya linear relationship
The crack growth rate for each DKI stage wasobtained for each sample based on this linearity Thecalculated average crack growth rates are given inFig 9
In Fig 9 when DKI was larger than 14 MPa m12 thecrack growth rate increased with per cent deformationand the crack growth rate in terms of mmcycle was in
6 Stress corrosion cracking performances of various
amounts of prestrained samples in C-1 solution
7 Potential drop curves of 0 prestrained sample at dif-
ferent DKI
8 Fractograghs of SEM showing crack growth of 0 pre-
strained sample
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
300 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
the range of 4061025 to 13661023 mmcycle Thegrowth rate fell into the same range that Harle andBeavers obtained13 Their crack growth rates also basedon potential drop measurements in API X-65 line pipesteel ranged from 1025 to 1024 mmcycle It is clear thatcrack velocity increased by about a factor of 2 withdeformations around 10
Fractographic observationSlow strain rate testing
The fracture surfaces of the prestrained samples failed inC-1 solution were examined using SEM and the results
are shown in Fig 10 All of these pictures were takennear the outside edges of the fracture surfaces
Figure 10a is the fracture surface of the 0 prestrainsample showing a large proportion of quasi-cleavagefacets The size of the biggest facet in the centre of theimage was around 35 mm as shown by the arrow Alsothe 2 sample as shown in Fig 10b had a quasi-cleavage facet about 40 mm in diameter
Corrosion fatigue crack advance measured by potentialdrop
The fracture features examined by the SEM are shownin Fig 11 The fracture surfaces of all the samplesdisplayed quasi-cleavage and secondary cracks
Electrochemical characteristicsFigure 12 shows the polarisation curves of samples withdifferent amounts of plastic deformation measured in C-1 solution at room temperature The figure demon-strated that the OCP of X-52 steel in near neutral pHenvironment was around 2740 mV and there was noactive to passive transition This confirms previousobservations14ndash16 Therefore during the crack growthprocess dissolution was not localised and there was nopassivation on the crack walls to prevent lateraldissolution and retain the crack shape
Discussion
Stress corrosion cracking occurrenceFigure 13 shows the surface cracks in samples that havebeen cold worked different amounts The flat surfacecracks which are perpendicular to the stress axis are
9 Crack growth rates of prestrained samples at different
DKI conditions
a 0 sample b 2 sample10 Images (SEM) of fracture surfaces showing quasi-cleavage
a 0 b 1011 Image (SEM) showing quasi-cleavage on fracture surface of prestrained samples
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 301
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
thought to occur at the locations with the highest tensilestress rather than being caused by the shear stress whichis associated with slip This has been taken by previousworkers17 to indicate that the local fracture was causedby NNPHSCC
Also as shown in Figs 10 and 11 there are somesecondary cracks on the fracture surfaces Both quasi-cleavage and secondary cracks are indications of theoccurrence of NNPHSCC
Increased crack growth rate with increasedyield strengthFigure 14 shows the crack growth rate at a testfrequency of 00025 Hz plotted against the yieldstrength for three different DKI conditions The yieldstrength increase of the 10 sample compared to the 0sample was about 41 However when DKI was around15 20 and 25 MPa m12 the crack growth rates of the10 sample severely increased to 22 22 and 23 timesrespectively compared to those of the 0 sample
Accelerated crack growth rate by severe loadingconditionIn Fig 9 for the same sample with increasing DKI thecrack growth rate increases which means that moresevere loading condition gives much higher crack growthrates From the slopes of the trend lines of each loadingstage in Fig 14 the increase rate of crack growthvelocity under DKI lt25 MPa m12 was eight times thatunder DKIlt15 MPa m12 This implies that higher DKI
gave much greater crack growth rate increase withhigher yield strength
Effect of plastic deformation on corrosion rateand OCPThe OCP and corrosion current densities of the steelsdetermined from the polarisation curves are plottedagainst per cent prestrain in Fig 15 It is seen that theOCP appeared to be not very sensitive to plastic defor-mation (the maximum difference of OCP betweenvarious amounts of prestrained samples was just25 mV) However the corrosion current increasedsignificantly and monotonically for samples withincreasing plastic deformation The total increase of
12 Polarisation curves of samples with prestrained sam-
ples in C-1 solution
a 0 b 813 Surface cracks near failed ends of prestrained samples tested in C-1 solution
14 Crack growth rates against yield strength of pre-
strained samples at different DKI conditions in near
neutral pH environment
15 Corrosion rates and OCPs of samples with various
amounts of plastic deformation
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
302 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
corrosion current density was around 17(130 mA cm22)
Enhanced enrichment of hydrogenThe major effect of cold work is an increase indislocation density the dislocation density in a metalthat has been highly deformed may be as high as1010 mm2218 This would be expected to produce anincrease in the number of hydrogen trapping sites1920
and an increase in the hydrogen adsorption coverage onthe surface
The increase in dislocation density would produce anincrease in the yield stress which increases the hydro-static stress fields around the crack tip This means thata higher concentration of hydrogen can be attracted tothe crack tip This enrichment is caused by the fact thathydrogen is attracted to the crack tip in proportion tothe hydrostatic stress parameter p which is the averageof the principal stresses Gu et al16 also suggested thathydrostatic stress could induce more hydrogen accumu-lation On the other hand the higher yield strengthcould produce a smaller plastic zone size at the crack tipand lower ductility This combination would producemore cracking in near neutral pH environment
Accelerated anodic dissolutionWhen steels are plastically deformed some fraction ofthe deformation energy (y5) is retained internallyThe major portion of this stored energy is as strainenergy associated with dislocations18 The increase in theamount of strain energy would allow easier dissolutionon specific sites since this energy is released on dis-solution This causes an increase in the corrosion ratewith an increase in the amount of cold work
In addition when hydrogen is trapped in the defectsin steel the lattice is dilated and the interatomiccohesion is decreased It has been estimated that 1hydrogen concentration could reduce the atomic bond-ing energy by 5 for steels21 So absorbed hydrogen canaccelerate the corrosion process16 22
To date it is broadly accepted that dissolvedhydrogen and anodic dissolution are the two mainfactors contributing to NNPHSCC23ndash25 Based on all theabove analysis the SCC susceptibility of materials withmore plastic deformation would be higher since bothprocesses are enhanced by plastic deformation
Conclusions1 The OCP of X-52 pipeline steel measured in the de-
aerated C-1 solution was in the range of 2730 to2760 mV (SCE) and it appeared to be insensitive to theprior plastic deformation However the corrosioncurrent density increased with the amount of priorplastic deformation when the steel was exposed to nearneutral pH environment Ten per cent additional priordeformation increased the corrosion current density byabout 17
2 Increasing DKI from 15 to 25 MPa m12 increasedSCC velocity by nearly an order of magnitude
3 Plasticity had a significant effect on the suscept-ibility of X-52 pipeline steel to NNPHSCC the materialwith more plastic deformation had higher SCC velocity(10 deformation increased the crack velocity by morethan two times and decreased the ratio of area reductionin the solution to that in air from 67 to 63
Acknowledgement
This work was funded by Alberta Energy ResearchInstitute of Canada through COURSE program and byNSERC of Canada
References1 B S Delanty and J OrsquoBeirne Oil Gas J 1992 15 39ndash44
2 lsquoSCC on Canadian oil and gas pipelinesrsquo Report no MH-2-95
National Energy Board Calgary Alberta Canada 1996
3 M Baker Jr Inc lsquoStress corrosion cracking studyrsquo Integrity
Management Program Delivery Order DTRS56-02-D-70036
TTO8 Department of Transportation Research and Special
Programs Administration Office of Pipeline Safety January 2005
4 T M Ahmed S B Lambert R Sutherby and A Plumtree
Corrosion 1997 53 581ndash590
5 B N Leis Proc 1st Pipeline Technology Conf on lsquoUpdate on
SCC life prediction models for pipelinesrsquo Ostende Belgium
October 1990
6 CEPA Proc MH-2-95 Vol 2 Appendix D Table 1 1996
Calgary Alta the National Energy Board
7 R N Parkins Proc NACE Northern Area Eastern Conf Ottawa
Ont Canada October 1999 NACE 1ndash22
8 W Chen R L Eadie and R L Sutherby Proc 2nd Int Conf on
lsquoEnvironment-induced cracking of metalsrsquo Banff Alta Canada
September 2004 Elsevier vol 2 211
9 J Been R Eadie and R Sutherby Proc Int Pipeline Conf
Calgary Alta Canada September 2006 ASME Paper IPC2006ndash
10345
10 T L Anderson lsquoFracture mechanics fundamental and applica-
tionsrsquo 2nd edn 1995 Boca Raton FL CRC Press
11 H H Johnson Mater Res Stand 1965 5 (9) 442ndash445
12 P McIntyre J Soc Environm Eng 1974 13ndash3 3ndash7
13 B A Harle and J A Beavers Corrosion 1993 49 861ndash863
14 R N Parkins Proc 37th Conf of Metallurgist lsquoMaterials for
resource recovery and transportrsquo (ed L Collins) 35ndash49 1998
Montreal The Metallurgical Society of CIM
15 J A Beavers C L Durr and S S Shademan Proc 37th Conf of
Metallurgist lsquoMaterials for resource recovery and transportrsquo (ed
L Collins) 51ndash69 1998 Montreal The Metallurgical Society of
CIM
16 B Gu J Luo and X Mao Corrosion 1999 55 96ndash106
17 W Chen F King T R Jack and M J Wilmott Metall Mater
Trans A 2002 33A 1429ndash1436
18 R W Hertzberg lsquoDeformation and fracture mechanics of
engineering materialsrsquo 4th edn 85 1996 Hoboken NJ John
Wiley amp Sons Inc
19 S X Xie and J P Hirth Corrosion 1982 38 486ndash493
20 H Huang and W J D Shaw Corrosion 1995 51 30ndash36
21 T McMullen M J Stott and E Zaremba Physical Review B
1987 35 1076ndash1081
22 J T Bulger lsquoThe Effect of microstructure on near-neutral pH
SCCrsquo MSc thesis University of Alberta Alta Canada 2000 137
23 R N Parkins W K Blanchard Jr and B S Delanty Corrosion
1994 50 394ndash410
24 L J Qiao J L Luo and X Mao J Mater Sci Lett 1997 16
516ndash520
25 B T Lu J L Luo and B McCrady Proc Int Pipeline Conf
Calgary Alta Canada SeptemberndashOctober 2002 ASME Paper
IPC2002ndash27234
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 303
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
the range of 4061025 to 13661023 mmcycle Thegrowth rate fell into the same range that Harle andBeavers obtained13 Their crack growth rates also basedon potential drop measurements in API X-65 line pipesteel ranged from 1025 to 1024 mmcycle It is clear thatcrack velocity increased by about a factor of 2 withdeformations around 10
Fractographic observationSlow strain rate testing
The fracture surfaces of the prestrained samples failed inC-1 solution were examined using SEM and the results
are shown in Fig 10 All of these pictures were takennear the outside edges of the fracture surfaces
Figure 10a is the fracture surface of the 0 prestrainsample showing a large proportion of quasi-cleavagefacets The size of the biggest facet in the centre of theimage was around 35 mm as shown by the arrow Alsothe 2 sample as shown in Fig 10b had a quasi-cleavage facet about 40 mm in diameter
Corrosion fatigue crack advance measured by potentialdrop
The fracture features examined by the SEM are shownin Fig 11 The fracture surfaces of all the samplesdisplayed quasi-cleavage and secondary cracks
Electrochemical characteristicsFigure 12 shows the polarisation curves of samples withdifferent amounts of plastic deformation measured in C-1 solution at room temperature The figure demon-strated that the OCP of X-52 steel in near neutral pHenvironment was around 2740 mV and there was noactive to passive transition This confirms previousobservations14ndash16 Therefore during the crack growthprocess dissolution was not localised and there was nopassivation on the crack walls to prevent lateraldissolution and retain the crack shape
Discussion
Stress corrosion cracking occurrenceFigure 13 shows the surface cracks in samples that havebeen cold worked different amounts The flat surfacecracks which are perpendicular to the stress axis are
9 Crack growth rates of prestrained samples at different
DKI conditions
a 0 sample b 2 sample10 Images (SEM) of fracture surfaces showing quasi-cleavage
a 0 b 1011 Image (SEM) showing quasi-cleavage on fracture surface of prestrained samples
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 301
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
thought to occur at the locations with the highest tensilestress rather than being caused by the shear stress whichis associated with slip This has been taken by previousworkers17 to indicate that the local fracture was causedby NNPHSCC
Also as shown in Figs 10 and 11 there are somesecondary cracks on the fracture surfaces Both quasi-cleavage and secondary cracks are indications of theoccurrence of NNPHSCC
Increased crack growth rate with increasedyield strengthFigure 14 shows the crack growth rate at a testfrequency of 00025 Hz plotted against the yieldstrength for three different DKI conditions The yieldstrength increase of the 10 sample compared to the 0sample was about 41 However when DKI was around15 20 and 25 MPa m12 the crack growth rates of the10 sample severely increased to 22 22 and 23 timesrespectively compared to those of the 0 sample
Accelerated crack growth rate by severe loadingconditionIn Fig 9 for the same sample with increasing DKI thecrack growth rate increases which means that moresevere loading condition gives much higher crack growthrates From the slopes of the trend lines of each loadingstage in Fig 14 the increase rate of crack growthvelocity under DKI lt25 MPa m12 was eight times thatunder DKIlt15 MPa m12 This implies that higher DKI
gave much greater crack growth rate increase withhigher yield strength
Effect of plastic deformation on corrosion rateand OCPThe OCP and corrosion current densities of the steelsdetermined from the polarisation curves are plottedagainst per cent prestrain in Fig 15 It is seen that theOCP appeared to be not very sensitive to plastic defor-mation (the maximum difference of OCP betweenvarious amounts of prestrained samples was just25 mV) However the corrosion current increasedsignificantly and monotonically for samples withincreasing plastic deformation The total increase of
12 Polarisation curves of samples with prestrained sam-
ples in C-1 solution
a 0 b 813 Surface cracks near failed ends of prestrained samples tested in C-1 solution
14 Crack growth rates against yield strength of pre-
strained samples at different DKI conditions in near
neutral pH environment
15 Corrosion rates and OCPs of samples with various
amounts of plastic deformation
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
302 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
corrosion current density was around 17(130 mA cm22)
Enhanced enrichment of hydrogenThe major effect of cold work is an increase indislocation density the dislocation density in a metalthat has been highly deformed may be as high as1010 mm2218 This would be expected to produce anincrease in the number of hydrogen trapping sites1920
and an increase in the hydrogen adsorption coverage onthe surface
The increase in dislocation density would produce anincrease in the yield stress which increases the hydro-static stress fields around the crack tip This means thata higher concentration of hydrogen can be attracted tothe crack tip This enrichment is caused by the fact thathydrogen is attracted to the crack tip in proportion tothe hydrostatic stress parameter p which is the averageof the principal stresses Gu et al16 also suggested thathydrostatic stress could induce more hydrogen accumu-lation On the other hand the higher yield strengthcould produce a smaller plastic zone size at the crack tipand lower ductility This combination would producemore cracking in near neutral pH environment
Accelerated anodic dissolutionWhen steels are plastically deformed some fraction ofthe deformation energy (y5) is retained internallyThe major portion of this stored energy is as strainenergy associated with dislocations18 The increase in theamount of strain energy would allow easier dissolutionon specific sites since this energy is released on dis-solution This causes an increase in the corrosion ratewith an increase in the amount of cold work
In addition when hydrogen is trapped in the defectsin steel the lattice is dilated and the interatomiccohesion is decreased It has been estimated that 1hydrogen concentration could reduce the atomic bond-ing energy by 5 for steels21 So absorbed hydrogen canaccelerate the corrosion process16 22
To date it is broadly accepted that dissolvedhydrogen and anodic dissolution are the two mainfactors contributing to NNPHSCC23ndash25 Based on all theabove analysis the SCC susceptibility of materials withmore plastic deformation would be higher since bothprocesses are enhanced by plastic deformation
Conclusions1 The OCP of X-52 pipeline steel measured in the de-
aerated C-1 solution was in the range of 2730 to2760 mV (SCE) and it appeared to be insensitive to theprior plastic deformation However the corrosioncurrent density increased with the amount of priorplastic deformation when the steel was exposed to nearneutral pH environment Ten per cent additional priordeformation increased the corrosion current density byabout 17
2 Increasing DKI from 15 to 25 MPa m12 increasedSCC velocity by nearly an order of magnitude
3 Plasticity had a significant effect on the suscept-ibility of X-52 pipeline steel to NNPHSCC the materialwith more plastic deformation had higher SCC velocity(10 deformation increased the crack velocity by morethan two times and decreased the ratio of area reductionin the solution to that in air from 67 to 63
Acknowledgement
This work was funded by Alberta Energy ResearchInstitute of Canada through COURSE program and byNSERC of Canada
References1 B S Delanty and J OrsquoBeirne Oil Gas J 1992 15 39ndash44
2 lsquoSCC on Canadian oil and gas pipelinesrsquo Report no MH-2-95
National Energy Board Calgary Alberta Canada 1996
3 M Baker Jr Inc lsquoStress corrosion cracking studyrsquo Integrity
Management Program Delivery Order DTRS56-02-D-70036
TTO8 Department of Transportation Research and Special
Programs Administration Office of Pipeline Safety January 2005
4 T M Ahmed S B Lambert R Sutherby and A Plumtree
Corrosion 1997 53 581ndash590
5 B N Leis Proc 1st Pipeline Technology Conf on lsquoUpdate on
SCC life prediction models for pipelinesrsquo Ostende Belgium
October 1990
6 CEPA Proc MH-2-95 Vol 2 Appendix D Table 1 1996
Calgary Alta the National Energy Board
7 R N Parkins Proc NACE Northern Area Eastern Conf Ottawa
Ont Canada October 1999 NACE 1ndash22
8 W Chen R L Eadie and R L Sutherby Proc 2nd Int Conf on
lsquoEnvironment-induced cracking of metalsrsquo Banff Alta Canada
September 2004 Elsevier vol 2 211
9 J Been R Eadie and R Sutherby Proc Int Pipeline Conf
Calgary Alta Canada September 2006 ASME Paper IPC2006ndash
10345
10 T L Anderson lsquoFracture mechanics fundamental and applica-
tionsrsquo 2nd edn 1995 Boca Raton FL CRC Press
11 H H Johnson Mater Res Stand 1965 5 (9) 442ndash445
12 P McIntyre J Soc Environm Eng 1974 13ndash3 3ndash7
13 B A Harle and J A Beavers Corrosion 1993 49 861ndash863
14 R N Parkins Proc 37th Conf of Metallurgist lsquoMaterials for
resource recovery and transportrsquo (ed L Collins) 35ndash49 1998
Montreal The Metallurgical Society of CIM
15 J A Beavers C L Durr and S S Shademan Proc 37th Conf of
Metallurgist lsquoMaterials for resource recovery and transportrsquo (ed
L Collins) 51ndash69 1998 Montreal The Metallurgical Society of
CIM
16 B Gu J Luo and X Mao Corrosion 1999 55 96ndash106
17 W Chen F King T R Jack and M J Wilmott Metall Mater
Trans A 2002 33A 1429ndash1436
18 R W Hertzberg lsquoDeformation and fracture mechanics of
engineering materialsrsquo 4th edn 85 1996 Hoboken NJ John
Wiley amp Sons Inc
19 S X Xie and J P Hirth Corrosion 1982 38 486ndash493
20 H Huang and W J D Shaw Corrosion 1995 51 30ndash36
21 T McMullen M J Stott and E Zaremba Physical Review B
1987 35 1076ndash1081
22 J T Bulger lsquoThe Effect of microstructure on near-neutral pH
SCCrsquo MSc thesis University of Alberta Alta Canada 2000 137
23 R N Parkins W K Blanchard Jr and B S Delanty Corrosion
1994 50 394ndash410
24 L J Qiao J L Luo and X Mao J Mater Sci Lett 1997 16
516ndash520
25 B T Lu J L Luo and B McCrady Proc Int Pipeline Conf
Calgary Alta Canada SeptemberndashOctober 2002 ASME Paper
IPC2002ndash27234
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 303
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
thought to occur at the locations with the highest tensilestress rather than being caused by the shear stress whichis associated with slip This has been taken by previousworkers17 to indicate that the local fracture was causedby NNPHSCC
Also as shown in Figs 10 and 11 there are somesecondary cracks on the fracture surfaces Both quasi-cleavage and secondary cracks are indications of theoccurrence of NNPHSCC
Increased crack growth rate with increasedyield strengthFigure 14 shows the crack growth rate at a testfrequency of 00025 Hz plotted against the yieldstrength for three different DKI conditions The yieldstrength increase of the 10 sample compared to the 0sample was about 41 However when DKI was around15 20 and 25 MPa m12 the crack growth rates of the10 sample severely increased to 22 22 and 23 timesrespectively compared to those of the 0 sample
Accelerated crack growth rate by severe loadingconditionIn Fig 9 for the same sample with increasing DKI thecrack growth rate increases which means that moresevere loading condition gives much higher crack growthrates From the slopes of the trend lines of each loadingstage in Fig 14 the increase rate of crack growthvelocity under DKI lt25 MPa m12 was eight times thatunder DKIlt15 MPa m12 This implies that higher DKI
gave much greater crack growth rate increase withhigher yield strength
Effect of plastic deformation on corrosion rateand OCPThe OCP and corrosion current densities of the steelsdetermined from the polarisation curves are plottedagainst per cent prestrain in Fig 15 It is seen that theOCP appeared to be not very sensitive to plastic defor-mation (the maximum difference of OCP betweenvarious amounts of prestrained samples was just25 mV) However the corrosion current increasedsignificantly and monotonically for samples withincreasing plastic deformation The total increase of
12 Polarisation curves of samples with prestrained sam-
ples in C-1 solution
a 0 b 813 Surface cracks near failed ends of prestrained samples tested in C-1 solution
14 Crack growth rates against yield strength of pre-
strained samples at different DKI conditions in near
neutral pH environment
15 Corrosion rates and OCPs of samples with various
amounts of plastic deformation
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
302 Corrosion Engineering Science and Technology 2008 VOL 43 NO 4
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
corrosion current density was around 17(130 mA cm22)
Enhanced enrichment of hydrogenThe major effect of cold work is an increase indislocation density the dislocation density in a metalthat has been highly deformed may be as high as1010 mm2218 This would be expected to produce anincrease in the number of hydrogen trapping sites1920
and an increase in the hydrogen adsorption coverage onthe surface
The increase in dislocation density would produce anincrease in the yield stress which increases the hydro-static stress fields around the crack tip This means thata higher concentration of hydrogen can be attracted tothe crack tip This enrichment is caused by the fact thathydrogen is attracted to the crack tip in proportion tothe hydrostatic stress parameter p which is the averageof the principal stresses Gu et al16 also suggested thathydrostatic stress could induce more hydrogen accumu-lation On the other hand the higher yield strengthcould produce a smaller plastic zone size at the crack tipand lower ductility This combination would producemore cracking in near neutral pH environment
Accelerated anodic dissolutionWhen steels are plastically deformed some fraction ofthe deformation energy (y5) is retained internallyThe major portion of this stored energy is as strainenergy associated with dislocations18 The increase in theamount of strain energy would allow easier dissolutionon specific sites since this energy is released on dis-solution This causes an increase in the corrosion ratewith an increase in the amount of cold work
In addition when hydrogen is trapped in the defectsin steel the lattice is dilated and the interatomiccohesion is decreased It has been estimated that 1hydrogen concentration could reduce the atomic bond-ing energy by 5 for steels21 So absorbed hydrogen canaccelerate the corrosion process16 22
To date it is broadly accepted that dissolvedhydrogen and anodic dissolution are the two mainfactors contributing to NNPHSCC23ndash25 Based on all theabove analysis the SCC susceptibility of materials withmore plastic deformation would be higher since bothprocesses are enhanced by plastic deformation
Conclusions1 The OCP of X-52 pipeline steel measured in the de-
aerated C-1 solution was in the range of 2730 to2760 mV (SCE) and it appeared to be insensitive to theprior plastic deformation However the corrosioncurrent density increased with the amount of priorplastic deformation when the steel was exposed to nearneutral pH environment Ten per cent additional priordeformation increased the corrosion current density byabout 17
2 Increasing DKI from 15 to 25 MPa m12 increasedSCC velocity by nearly an order of magnitude
3 Plasticity had a significant effect on the suscept-ibility of X-52 pipeline steel to NNPHSCC the materialwith more plastic deformation had higher SCC velocity(10 deformation increased the crack velocity by morethan two times and decreased the ratio of area reductionin the solution to that in air from 67 to 63
Acknowledgement
This work was funded by Alberta Energy ResearchInstitute of Canada through COURSE program and byNSERC of Canada
References1 B S Delanty and J OrsquoBeirne Oil Gas J 1992 15 39ndash44
2 lsquoSCC on Canadian oil and gas pipelinesrsquo Report no MH-2-95
National Energy Board Calgary Alberta Canada 1996
3 M Baker Jr Inc lsquoStress corrosion cracking studyrsquo Integrity
Management Program Delivery Order DTRS56-02-D-70036
TTO8 Department of Transportation Research and Special
Programs Administration Office of Pipeline Safety January 2005
4 T M Ahmed S B Lambert R Sutherby and A Plumtree
Corrosion 1997 53 581ndash590
5 B N Leis Proc 1st Pipeline Technology Conf on lsquoUpdate on
SCC life prediction models for pipelinesrsquo Ostende Belgium
October 1990
6 CEPA Proc MH-2-95 Vol 2 Appendix D Table 1 1996
Calgary Alta the National Energy Board
7 R N Parkins Proc NACE Northern Area Eastern Conf Ottawa
Ont Canada October 1999 NACE 1ndash22
8 W Chen R L Eadie and R L Sutherby Proc 2nd Int Conf on
lsquoEnvironment-induced cracking of metalsrsquo Banff Alta Canada
September 2004 Elsevier vol 2 211
9 J Been R Eadie and R Sutherby Proc Int Pipeline Conf
Calgary Alta Canada September 2006 ASME Paper IPC2006ndash
10345
10 T L Anderson lsquoFracture mechanics fundamental and applica-
tionsrsquo 2nd edn 1995 Boca Raton FL CRC Press
11 H H Johnson Mater Res Stand 1965 5 (9) 442ndash445
12 P McIntyre J Soc Environm Eng 1974 13ndash3 3ndash7
13 B A Harle and J A Beavers Corrosion 1993 49 861ndash863
14 R N Parkins Proc 37th Conf of Metallurgist lsquoMaterials for
resource recovery and transportrsquo (ed L Collins) 35ndash49 1998
Montreal The Metallurgical Society of CIM
15 J A Beavers C L Durr and S S Shademan Proc 37th Conf of
Metallurgist lsquoMaterials for resource recovery and transportrsquo (ed
L Collins) 51ndash69 1998 Montreal The Metallurgical Society of
CIM
16 B Gu J Luo and X Mao Corrosion 1999 55 96ndash106
17 W Chen F King T R Jack and M J Wilmott Metall Mater
Trans A 2002 33A 1429ndash1436
18 R W Hertzberg lsquoDeformation and fracture mechanics of
engineering materialsrsquo 4th edn 85 1996 Hoboken NJ John
Wiley amp Sons Inc
19 S X Xie and J P Hirth Corrosion 1982 38 486ndash493
20 H Huang and W J D Shaw Corrosion 1995 51 30ndash36
21 T McMullen M J Stott and E Zaremba Physical Review B
1987 35 1076ndash1081
22 J T Bulger lsquoThe Effect of microstructure on near-neutral pH
SCCrsquo MSc thesis University of Alberta Alta Canada 2000 137
23 R N Parkins W K Blanchard Jr and B S Delanty Corrosion
1994 50 394ndash410
24 L J Qiao J L Luo and X Mao J Mater Sci Lett 1997 16
516ndash520
25 B T Lu J L Luo and B McCrady Proc Int Pipeline Conf
Calgary Alta Canada SeptemberndashOctober 2002 ASME Paper
IPC2002ndash27234
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 303
Pub
lishe
d by
Man
ey P
ublis
hing
(c)
IOM
Com
mun
icat
ions
Ltd
corrosion current density was around 17(130 mA cm22)
Enhanced enrichment of hydrogenThe major effect of cold work is an increase indislocation density the dislocation density in a metalthat has been highly deformed may be as high as1010 mm2218 This would be expected to produce anincrease in the number of hydrogen trapping sites1920
and an increase in the hydrogen adsorption coverage onthe surface
The increase in dislocation density would produce anincrease in the yield stress which increases the hydro-static stress fields around the crack tip This means thata higher concentration of hydrogen can be attracted tothe crack tip This enrichment is caused by the fact thathydrogen is attracted to the crack tip in proportion tothe hydrostatic stress parameter p which is the averageof the principal stresses Gu et al16 also suggested thathydrostatic stress could induce more hydrogen accumu-lation On the other hand the higher yield strengthcould produce a smaller plastic zone size at the crack tipand lower ductility This combination would producemore cracking in near neutral pH environment
Accelerated anodic dissolutionWhen steels are plastically deformed some fraction ofthe deformation energy (y5) is retained internallyThe major portion of this stored energy is as strainenergy associated with dislocations18 The increase in theamount of strain energy would allow easier dissolutionon specific sites since this energy is released on dis-solution This causes an increase in the corrosion ratewith an increase in the amount of cold work
In addition when hydrogen is trapped in the defectsin steel the lattice is dilated and the interatomiccohesion is decreased It has been estimated that 1hydrogen concentration could reduce the atomic bond-ing energy by 5 for steels21 So absorbed hydrogen canaccelerate the corrosion process16 22
To date it is broadly accepted that dissolvedhydrogen and anodic dissolution are the two mainfactors contributing to NNPHSCC23ndash25 Based on all theabove analysis the SCC susceptibility of materials withmore plastic deformation would be higher since bothprocesses are enhanced by plastic deformation
Conclusions1 The OCP of X-52 pipeline steel measured in the de-
aerated C-1 solution was in the range of 2730 to2760 mV (SCE) and it appeared to be insensitive to theprior plastic deformation However the corrosioncurrent density increased with the amount of priorplastic deformation when the steel was exposed to nearneutral pH environment Ten per cent additional priordeformation increased the corrosion current density byabout 17
2 Increasing DKI from 15 to 25 MPa m12 increasedSCC velocity by nearly an order of magnitude
3 Plasticity had a significant effect on the suscept-ibility of X-52 pipeline steel to NNPHSCC the materialwith more plastic deformation had higher SCC velocity(10 deformation increased the crack velocity by morethan two times and decreased the ratio of area reductionin the solution to that in air from 67 to 63
Acknowledgement
This work was funded by Alberta Energy ResearchInstitute of Canada through COURSE program and byNSERC of Canada
References1 B S Delanty and J OrsquoBeirne Oil Gas J 1992 15 39ndash44
2 lsquoSCC on Canadian oil and gas pipelinesrsquo Report no MH-2-95
National Energy Board Calgary Alberta Canada 1996
3 M Baker Jr Inc lsquoStress corrosion cracking studyrsquo Integrity
Management Program Delivery Order DTRS56-02-D-70036
TTO8 Department of Transportation Research and Special
Programs Administration Office of Pipeline Safety January 2005
4 T M Ahmed S B Lambert R Sutherby and A Plumtree
Corrosion 1997 53 581ndash590
5 B N Leis Proc 1st Pipeline Technology Conf on lsquoUpdate on
SCC life prediction models for pipelinesrsquo Ostende Belgium
October 1990
6 CEPA Proc MH-2-95 Vol 2 Appendix D Table 1 1996
Calgary Alta the National Energy Board
7 R N Parkins Proc NACE Northern Area Eastern Conf Ottawa
Ont Canada October 1999 NACE 1ndash22
8 W Chen R L Eadie and R L Sutherby Proc 2nd Int Conf on
lsquoEnvironment-induced cracking of metalsrsquo Banff Alta Canada
September 2004 Elsevier vol 2 211
9 J Been R Eadie and R Sutherby Proc Int Pipeline Conf
Calgary Alta Canada September 2006 ASME Paper IPC2006ndash
10345
10 T L Anderson lsquoFracture mechanics fundamental and applica-
tionsrsquo 2nd edn 1995 Boca Raton FL CRC Press
11 H H Johnson Mater Res Stand 1965 5 (9) 442ndash445
12 P McIntyre J Soc Environm Eng 1974 13ndash3 3ndash7
13 B A Harle and J A Beavers Corrosion 1993 49 861ndash863
14 R N Parkins Proc 37th Conf of Metallurgist lsquoMaterials for
resource recovery and transportrsquo (ed L Collins) 35ndash49 1998
Montreal The Metallurgical Society of CIM
15 J A Beavers C L Durr and S S Shademan Proc 37th Conf of
Metallurgist lsquoMaterials for resource recovery and transportrsquo (ed
L Collins) 51ndash69 1998 Montreal The Metallurgical Society of
CIM
16 B Gu J Luo and X Mao Corrosion 1999 55 96ndash106
17 W Chen F King T R Jack and M J Wilmott Metall Mater
Trans A 2002 33A 1429ndash1436
18 R W Hertzberg lsquoDeformation and fracture mechanics of
engineering materialsrsquo 4th edn 85 1996 Hoboken NJ John
Wiley amp Sons Inc
19 S X Xie and J P Hirth Corrosion 1982 38 486ndash493
20 H Huang and W J D Shaw Corrosion 1995 51 30ndash36
21 T McMullen M J Stott and E Zaremba Physical Review B
1987 35 1076ndash1081
22 J T Bulger lsquoThe Effect of microstructure on near-neutral pH
SCCrsquo MSc thesis University of Alberta Alta Canada 2000 137
23 R N Parkins W K Blanchard Jr and B S Delanty Corrosion
1994 50 394ndash410
24 L J Qiao J L Luo and X Mao J Mater Sci Lett 1997 16
516ndash520
25 B T Lu J L Luo and B McCrady Proc Int Pipeline Conf
Calgary Alta Canada SeptemberndashOctober 2002 ASME Paper
IPC2002ndash27234
Li et al Influence of plasticity on corrosion and SCC behaviour in near neutral pH environment
Corrosion Engineering Science and Technology 2008 VOL 43 NO 4 303
