systematicmv222.com

MaterialsChemistryandPhysics88(2004)221–226ThemetastablepittingofmildsteelinbicarbonatesolutionsYumingTang,YuZuoFacultyofMaterialsScienceandEngineering,BeijingUniversityofChemicalTechnology,No.
15EastRoadBeisanhuan,Beijing100029,PRChinaReceived9May2004;receivedinrevisedform1July2004;accepted12July2004AbstractThemetastablepittingbehaviorofmildsteelin0.
5MNaHCO3solutionscontainingdifferentconcentrationsofClwasstudiedwithpotentiodynamicandpotentiostaticpolarizationmethods.
Currentuctuationsoccurredbeforestablepitting,withthecharacteristicofquickincreaseandslowdecrease.
TheoccurringpotentialformetastablepittingEmobeysnormaldistribution,anddecreaseswiththeincreaseofchlorideconcentration.
Potentialhasnoapparenteffectonthefrequencyofuctuations.
Inpotentiostatictest,whenpotentialisaboveEmandfarbelowthepittingbreakdownpotentialEb,thecurrentuctuationsremainforaperiodthendisappear,resultinginpitswithdiametersofmicrons.
WhenpotentialisclosetoEb,thepolarizationcurrentmaynallyrisescontinuouslyafteracertaintimeofuctuations,andmetastablepittingisreplacedbystablepitting.
2004ElsevierB.
V.
Allrightsreserved.
Keywords:Mildsteel;Metastablepitting;Currentuctuations;Bicarbonate1.
IntroductionDuringtheearlystageofpittingofsomepassivatedmet-alsandalloys,smallcurrentuctuationsareobserved,whicharetheresultsofverysmalllocalcorrosionandrepassiva-tionandaredenedasmetastablepitting.
Manyresearchers[1–3]havestudiedthecharacteristicsofmetastablepittingandtherelationshipbetweenmetastablepittingandstablepitting,tryingtopredictpittingtendencyandpropagatingratebymetastablepittingbehaviors.
However,mostofthepreviousstudieshavefocusedonmetastablebehaviorsofstainlesssteelandaluminumalloys.
Forcarbonsteelsthepublishedstudiesinthisareaarerelativelyscarce.
Chengandco-workers[4,5]andDongetal.
[6]studiedcurrentandpo-tentialuctuationsforA516-70carbonsteeland16Mnsteelin0.
5MNaHCO3+0.
1MNaClsolutionsbyelectrochemicalnoisemeasurementseparately.
Itwassuggestedthatthequickriseandslowrecoveryofthecurrentuctuationsreecttheinitiation,growthandrepassivationofmetastablepits,whileCorrespondingauthor.
Tel.
:+86-1064434818;fax:+86-1064750181.
E-mailaddress:zuoy@mail.
buct.
edu.
cn(Y.
Zuo).
potentialuctuationsreecttheresponseoftheelectrodeca-pacitancetothepitgrowthcharge.
Jiangetal.
[7]reportedcurrentuctuationsofmildsteelinphosphate–boratebuffersolutionwithchlorideions.
Theparametersofcurrentuc-tuations,suchasfrequency,peakcurrentanddeclinerateofthecurrent,followcertainstatisticaldistributions.
Duetothecomplexityandrandomnessoftheactualsystem,inordertounderstandmetastablepittingbehaviorsofcarbonsteelandtheirrelationshiptostablepitting,moresystematicstudiesareneeded.
Inthispaper,potentiodynamicandpotentiostaticpolarizationmethodswereusedtomeasurethecurrentuc-tuationofmildsteelin0.
5MNaHCO3solutionscontainingvariousconcentrationsofchlorideions,andtheuctuationbehaviorandtherelatedcorrosiondamagesonthesurfacewerestudied.
2.
ExperimentalmethodsThetestmaterialwasmildsteelwiththechemicalcom-position(wt.
%):C0.
19;Si0.
22;Mn0.
56;P0.
0086;and0254-0584/$–seefrontmatter2004ElsevierB.
V.
Allrightsreserved.
doi:10.
1016/j.
matchemphys.
2004.
07.
014http://www.
paper.
edu.
cn222Y.
Tang,Y.
Zuo/MaterialsChemistryandPhysics88(2004)221–226S0.
022.
Specimensweremanuallynishedupto800gritSiCabrasivepapers,degreasedinalcoholandrinsedwithde-ionizedwater.
Thespecimenswerecoatedwithepoxyresin,leavinganareaof0.
04cm2exposedtosolution.
Allthesolutionsweremadefromanalyticalgradereagentsandde-ionizedwater.
Thetestswerecarriedoutatambienttemperature.
Potentiodynamicandpotentiostaticpolarizationtestswerecarriedouttostudythecurrentuctuationsduringmetastablepittingofthesteelin0.
5MNaHCO3solutionscontainingvariousconcentrationsofchlorideions(pH=8.
8).
Thecurrentwasrecordedagainsttime.
Inpotentiody-namictest,specimenswererstcathodicallypolarizedfor3minatpotential400mVbelowopencircuit,andthenan-odicpolarizationwascarriedoutinapotentialscanningrateof0.
1mVs1untilstablepittingoccurred.
Inpotentiostatictest,thesamepolarizationprocessaspotentiodynamictestwasusedrst,butwhentherstcurrentuctuationoccurred,thepotentialwaskeptconstant,andthecurrentwasrecordedagainsttimeuntilcurrentuctuationsdisappearedorstablepittingoccurred.
PolarizationtestswerecontrolledwithanHDV-7-typepotentiostat.
Thelowerlimitofcurrentdetectioninthestudywas0.
02A.
Becauseofthestochasticpropertiesofmetastablepitting,undereachexperimentalconditionvetestswererun.
Afterexperiments,thespecimenswereexam-inedusingSEM.
AllthepotentialsreportedwererelativetoFig.
1.
(a)Thecurrentuctuationsofcarbonsteelin0.
5MNaHCO3+0.
1MNaClsolutionunderpotentiodynamiccontrol;and(b)Observedpitsonthesamesampleasshownin(a)afterpolarization.
asaturatedcalomelelectrode,andthecounterelectrodewasplatinum.
3.
ResultsanddiscussionFig.
1(a)showsthecurrentuctuationsofthesteeltestedin0.
5MNaHCO3+0.
1MNaClsolutionunderpotentio-dynamiccontrol.
Currentuctuationsbegantooccuratthepotentialofabout85mVwiththecharacteristicofquickincreaseandslowdecrease.
Herewedenotethepotentialasmetastablepittingpotential,Em.
Mostoftheuctuationshaveverysmallpeakcurrents,lessthan0.
3A.
Thepeakcurrentsincreasedwiththeincreaseofpotential.
Thecurrentrisescontinuouslyat43mV,indicatingthebeginningofsta-blepitting,andthepotentialisdenedaspittingbreakdownpotentialEb.
Aftertheexperiment,pitswereobservedonthesamplesurfaceunder50*microscope.
Fig.
1(b)showstheobservedpitsonthesamesamplewithSEM.
Thepitsshowrelativelylargeandirregularshapes,whichsuggeststhatthepitshavebeenformedbyaccumulatedlocaldisso-lutionprocesses.
Becauseofthestochasticcharacteristicofpittingprocess,fortheveparalleltests,themetastablepit-tingpotentialEmandthepittingbreakdownpotentialEbofthesteelin0.
5MNaHCO3+0.
1MNaClsolutionsvariedincertainranges—Emvariedfrom166to26mVwithY.
Tang,Y.
Zuo/MaterialsChemistryandPhysics88(2004)221–226223Fig.
2.
(a)Thecurrentuctuationsofcarbonsteelin0.
5MNaHCO3+0.
1MNaClsolutionunderpotentiostaticcontrolat100mV;and(b)Observedpitsonthesamesampleasshownin(a)afterpolarization.
Fig.
3.
(a)Thecurrentuctuationsofcarbonsteelin0.
5MNaHCO3+0.
1MNaClsolutionunderpotentiostaticcontrolat26mV;and(b)Observedpitsonthesamesampleasshownin(a)afterpolarization.
224Y.
Tang,Y.
Zuo/MaterialsChemistryandPhysics88(2004)221–226anaverageof118mVandEbvariedfrom56to+24mVwithanaverageof15mV.
Inpotentiostaticpolarizationtest,itdependsonthepoten-tialwhetherstablepittinghappensafteraperiodofcurrentuctuations.
Fig.
2(a)showsthecurrent–timecurverecordedataconstantpotentialof100mV,whichisaboveEmandfarbelowEbforthesystem.
Manycurrentuctuationsareseen,butaftercertaintimethepeakcurrentsoftheuctuationsdeclinedgraduallyandnallytheuctuationsdisappeared.
Afterpolarization,thesamplesurfacewasexaminedwithSEM,andsomesmallpitswithdiametersofafewmicronswereobserved,asshowninFig.
2(b).
Severalmicro-pitstendtogathertogether,suggestingthatonceametastablepitrepas-sivated,thesurroundingareasarestillactiveareasforfollow-ingmetastablepitstonucleate.
Thisresultrevealsthatforthestudiedmaterial/solutionsystem,thesmallcurrentuctua-tionsmayleadtolocaldissolutionandaccumulatedpittingdamagesonthesamplesurface,eventhoughstablepittingdoesnothappen.
IfthesampleiscontrolledataconstantpotentialclosetoEb,thecurrentmaynallyrisesafteracer-taintimeofuctuationsandstablepittinghappens.
Fig.
3(a)showsthesituationwherethesamplewaspotentiostaticallypolarizedat26mVin0.
5MNaHCO3+0.
1MNaClsolu-tion.
Aftertheexperiment,relativelylargepitswereobserved,asshowninFig.
3(b).
Hence,boththecurrent–timetran-sientsandthemorphologyofpitsobtainedinpotentiostati-calpolarizationareverysimilartothoseinpotentiodynamicalpolarization.
Fig.
4showsseveraltypicalcurrentuctuationsselectedfromFig.
3(a).
Thecurrentpeaksdisplaythetypicalshapeofquickcurrentrisefollowedbygradualdecrease,whichisattributedtofastgrowthofthemetastablepitandrelativelyslowerrepairofthepassivelm.
Theareaundereachcurrentpeakisrelatedtothedissolvedvolumeofthecorrespondingmetastablepit.
Asrstapproximation,supposethepeakcur-rentis0.
2A,thelifetimeofthepeakis1sandthepithasahemisphericshape,theestimatedpitradiusisabout1.
2Awhichisconsistentwiththeobservedvalues.
Itisobviousthatpitsmaybeformedfromtheaccumulatedlocaldissolutionprocesses.
Fig.
4.
Typicalcurrentuctuationsofcarbonsteelin0.
5MNaHCO3+0.
1MNaClsolution.
Fig.
5.
ProbabilitydistributionofEm.
TheaboveresultsshowthatwhenthesamplepotentialwasbetweenmetastablepotentialEm,andpittingbreakdownpo-tentialEb,thecurrentuctuationsmaystillcauselocaldis-solutionsandaccumulatedpittingdamages.
Therefore,thepotentialEmisalsoanimportantparametertocharacterpit-tingresistanceofthesteel.
ThemeasuredEmvaluesvaried,whichreectsthestochasticcharacteristicofmetastablepit-ting.
Fig.
5showstheprobabilitydistributionofmeasuredEmvaluesformildsteelstudiedin0.
5MNaHCO3+0.
05MNaClsolution.
TheresultshowsthatEmobeysnormaldis-tribution,similartothatofpittingbreakdownpotentialEbreportedelsewhere[8].
TheinuencesofchlorideconcentrationonEmandEbofmildsteelin0.
5MNaHCO3+NaClsolutionsareshowninFig.
6.
EachpointinFig.
6isanaveragevalueofveex-periments.
Withtheincreaseofchlorideconcentration,bothEmandEbdecreaseandfollowsimilartendencies.
Similarphenomenawerereported[9]for316LstainlesssteelinNaClsolutions.
Thisresultsuggeststhatthenucleationprocessesformetastablepittingandstablepittingaremoreorlesssim-ilar.
Fig.
6.
TheinuenceofchlorideconcentrationonEmandEb.
Y.
Tang,Y.
Zuo/MaterialsChemistryandPhysics88(2004)221–226225Fig.
7.
Theinuenceofpotentialonfrequencyfuctuationsofcarbonsteelinofcarbonsteelin0.
5MNaHCO3+0.
05MNaClsolution.
Eachcurrentpeakoncurrent–timecurvesrepresentstheinitiation,growthandrepassivationofametastablepit.
So,thefrequencyofcurrentuctuationsisdirectlyassociatedwiththerateofmetastablepitinitiation.
Fig.
7showstheinuenceofpotentialonuctuationfrequencyofmildsteelin0.
5MNaHCO3+0.
05MNaClsolutionunderpotentiody-namiccontrol.
Theinitialuctuationfrequencyisrelativelyhigh,andtheuctuationsshowanegativedependenceonthepotential.
Fig.
8showsthetimedependenceofuctu-ationfrequencyofmildsteelin0.
5MNaHCO3+0.
05MNaClsolutionataconstantpotentialof90mV,atwhichthecurrentuctuationsnallydisappearedandnostablepit-tinghappened.
Theuctuationfrequencyalsodecreaseswithtime,similartotheresultobtainedunderpotentiodynamicalcondition.
TheresultsshowninFigs.
7and8suggestthatnomatterstablepittinghappensornot,thenucleationfrequencyofmetastablepitsdecreaseswithtime.
Itiswellaccepted[10]thatamongvariousfactorsinu-encingpittingbehaviorofcarbonsteel,thekind,size,shapeanddensityofinclusionsplayveryimportantroles.
Espe-Fig.
8.
Timedependenceoffrequencyofuctuationsofcarbonsteelinofcarbonsteelin0.
5MNaHCO3+0.
05MNaClsolutionataconstantpotentialof90mV.
ciallysuldeinclusionsarethemostsusceptiblesitesforpitstonucleate.
Atagivenpotential,thecurrentuctuationfre-quencydecreaseswithtimeduetothedecreaseofactivesites.
Withtheincreaseofpotential,moresitesmaybeactivated.
BursteinandPistorius[11]reportedthatthemetastablepitsinitiationratefor304stainlesssteelbecomesgreaterwithincreasingpotentialandreachesamaximumatacertainpo-tential,thendecreaseswithpotentialduetoexhaustionofac-tivesites.
ChengandLuo'sstudyonA516-70carbonsteelinNaHCO3–NaClsolution[4]showsthatthereisatransitionalpotential(Etr).
BelowtheEtr,potentialhasapositiveroleonrateofpitoccurrence,andabovetheEtr,anegativeroleisobserved.
However,thepresentresultshowsthatthedecayofactivesitesplaysagreaterrolethanpotential,whichleadtothedecreaseofuctuationfrequencyinpotentiodynamicpolarization.
FromFig.
1(a,b)therearemuchmorecurrentuctuationsthanobservedpitsonthesurface,henceeachpitmustbeanaccumulatedresultbymanysmallerdissolu-tions.
Onceametastablepittingeventhappens,thefollowingmetastablepitsmaytendtooccuraroundthedamagedarea,leadingtoaccumulatedlocaldissolution.
Whileforstainlesssteelaformalmetastablepitcannotbecomeapreferentialsiteforfollowingpitstonucleate[12].
Theweakerpassivityofmildsteelcomparedtostainlesssteelshouldexplainthedifference.
Therefore,formildsteel,theinuenceofinclu-sionsonnucleationofmetastablepitsisstilltobestudiedfurther.
4.
Conclusions1.
Metastablepittingbehaviorisobservedformildsteelin0.
5MNaHCO3+NaClsolutions,whichisrepresentedbycurrentuctuationswiththecharacteristicofquickincreaseandslowdecrease.
2.
TheoccurringpotentialformetastablepitsEmobeysnor-maldistribution,anddecreasedwiththeincreaseofchlo-rideconcentration,indicatingthatClpromotesnucle-ationofmetastablepits.
Fluctuationfrequencydecreaseswithtime.
Potentialhasnoapparenteffectonthefre-quencyofuctuations.
3.
Inpotentiostatictest,whenpotentialisaboveEmandfarbelowEb,thecurrentuctuationsremainforaperiodthendisappear,resultinginsmallpitswithdiametersofmi-crons.
WhenpotentialisclosetoEb,thecurrentmay-nallyrisescontinuouslyafteracertaintimeofuctuations,andmetastablepittingisreplacedbystablepitting.
AcknowledgementsTheauthorsaregratefultoNationalNaturalFoundationofChina(Contract50271005)andtheStateKeyPlanforBasicResearch(G19990650)fornancialsupportstothiswork.
226Y.
Tang,Y.
Zuo/MaterialsChemistryandPhysics88(2004)221–226References[1]G.
S.
Frankel,L.
Stockert,F.
Hunkeler,H.
Boehni,Corrosion43(1987)429.
[2]P.
C.
Pistorius,G.
T.
Burstein,Corros.
Sci.
33(1992)1885.
[3]P.
C.
Pistorious,G.
T.
Burstein,Corros.
Sci.
36(3)(1994)255.
[4]Y.
F.
Cheng,J.
L.
Luo,J.
Electrochem.
Soc.
146(3)(1999)970.
[5]Y.
F.
Cheng,M.
Wilmott,J.
L.
Luo,Corros.
Sci.
41(1999)1245.
[6]Z.
-H.
Dong,X.
-P.
Guo,J.
-X.
Zheng,L.
-M.
Xu,Corros.
Sci.
Prot.
Tech.
13(4)(2001)195.
[7]T.
Jiang,Y.
Zuo,J.
-P.
Xiong,J.
Chin.
Soc.
Corros.
Prot.
21(6)(2001)357.
[8]T.
Shibata,T.
Takeyama,Corrosion33(1977)243.
[9]Y.
Zuo,H.
-T.
Wang,J.
-M.
Zhao,J.
-P.
Xiong,Corros.
Sci.
44(2002)13.
[10]L.
J.
Gainer,G.
R.
Wallwork,Corrosion35(1979)435.
[11]G.
T.
Burstein,P.
C.
Pistorius,Corrosion51(1995)380.
[12]D.
E.
Williams,J.
Stewart,H.
Balkwill,Corros.
Sci.
36(1994)1213.