Open Archive Toulouse Archive Ouverte (OATAO) OATAO is an open access repository that collects the work of Toulouse researchers and makes it freely available over the web where possible.

This is an author-deposited version published in: http://oatao.univ-toulouse.fr/ Eprints ID : 2300

To link to this article : URL : http://dx.doi.org/10.1016/j.corsci.2007.07.014

To cite this version : Lima - Neto , Pedro de and Farias , Jesualdo P. and Herculano , Luis Flávio G. and Miranda, Hélio C. de and Araújo, Walney S. and Jorcin, Jean-Baptiste and Pébère, Nadine ( 2008) Determination of the sensitized zone extension in welded AISI 304 stainless steel using non-destructive electrochemical techniques. Corrosion Science, vol. 50 (n° 4). pp. 1149-1155. ISSN 0010-938X

Any correspondence concerning this service should be sent to the repository

administrator: staff-oatao@inp-toulouse.fr

DeterminationofthesensitizedzoneextensioninweldedAISI304stainlesssteelusingnon-destructiveelectrochemicaltechniques

PedrodeLima-Netoa,*,JesualdoP.Fariasb,LuisFlávioG.Herculanoa,b,HélioC.deMirandab,WalneyS.Araújob,Jean-BaptisteJorcinc,NadinePébèrec

a De­partame­n­to de­ Quí­mi­ca An­alí­ti­ca e­ Fí­s­i­co-Quí­mi­ca, Un­i­ve­rs­i­dade­ Fe­de­ral do Ce­ará, Campus­ do Pi­ci­, Bloco 940, 60455-970, Fortale­za, CE, Brazi­lb De­partame­n­to de­ En­ge­n­hari­a Me­talúrgi­ca e­ Mate­ri­ai­s­, Un­i­ve­rs­i­dade­ Fe­de­ral do Ce­ará, Campus­ do Pi­ci­, Bloco 714, 60455-970, Fortale­za, CE, Brazi­l

c CIR­IMAT – UMR­ CNR­S 5085, ENSIACET, 118 route­ de­ Narbon­n­e­, 31077 Toulous­e­ ce­de­x 04, Fran­ce­

Abstract

Extensionofsensitizedzone(SZ)inweldedAISI304stainlesssteelwasdeterminedbytwonon-destructiveelectrochemicaltests:doubleloopelectrochemicalpotentiokineticreactivationtechnique(DLEPR)andlocalelectrochemicalimpedancespectroscopy(LEIS).Weldingwascarriedoutusingtheshieldedmetalarcwithtwoselectedweldingenergies:thefirstone(0.7kJmm¡1)doesnotpromotethesensitiza-tionofthe304steelanditconstitutesthereferencesampleandthesecondone(2.2kJmm¡1)whichleadstotheprecipitationofchromiumcarbidesinthegrainboundariesaftertheweldingprocess.Thenon-destructiveDLEPRandLEIStestsallowedthelengthoftheSZtobedeterminedandagoodagreementbetweenthetwotechniquesandthemicrostructureofthetwoweldedsampleswasshown.Thepresenceofaninductivelooponthelocalimpedancediagramsseemstoreflectagalvaniccouplingbetweentheweldstring(anode)andtheweldedstainlesssteelplates(cathode)whichwillbeveryprejudicialtoagoodcorrosionresistanceoftheweldedsystem.Theresultsshowedthatthetwoelectrochemicaltestscouldbeappliedinpracticalcasesinindustrialfield.

Ke­ywords­: A.Stainlesssteel;B.EIS;B.SEM;C.Welding;C.Intergranularcorrosion

1. Intro­ductio­n

Itiswidelyknownthatweldedausteniticstainlesssteelscanpresentasensitizedzone(SZ)intheheataVectedzone(HAZ).This isawell-knownphenomenonandconsistsofcarbide precipitation at grain boundaries and chromiumdepletioninadjacentregions,makingthematerialsuscep-tible to intergranular corrosion.The intensityof thisphe-nomenonisaVectedby,amongotherfactors,theamountofcarboninthebasematerialandintheweldedmaterial,thewelding process, the welding energy, the exposure time ofthematerialtothetemperatureinthatsegregationphenom-enatakeplaceandthecarbidemakerelements[1].

doi:10.1016/j.corsci.2007.07.014

* Correspondingauthor.Tel.:+558533669956;fax:+558533669982.E-mai­l addre­s­s­: pln@ufc.br(P.deLima-Neto).

Usually,thesensitizedregionisdeterminedbymethodolo-giesmentionedinASTMA262StandardPractice.However,thisstandarddoesnotquantifythesensitizationdegree,thetestsareofdiYcultpreparation,exclusive foruse in labo-ratoryand theyaredestructive.Foran industrialpracticepoint of view, it is necessary to develop non-destructivemethodsandtestsfortheassessmentofsensitizedzoneinaweldedstainlesssteelonitssiteofoperation.

Doubleloopelectrochemicalpotentiokineticreactivationtest(DLEPR)isapowerfultoolusedrecentlyinourlabo-ratory to evaluate the sensitization toquantify thedegreeofsensitizationoffiveausteniticstainlesssteels(304,304L,316L,321and347)heattreatedintherangeof400°C–600°Cforuntil100hours [2,3], tostudythe lossof thecorrosionresistanceoftheUNSS31803duplexstainlesssteelagedatlow temperatures (350–550°C) [4] and finally, to assessthe eVect of the low-temperatureaging in the corrosion

resistanceofAISI444stainlesssteel[5].Theadvantagesofthistechniqueare:itisquick,non-destructive,couldbeusedfori­n­ s­i­tumeasurementsandtheresultsareindependentofthesurfacefinishing.

Localelectrochemicalimpedancespectroscopy(LEIS)isalsoanothernon-destructiveelectrochemicaltechniquethathasbeenusedinrecentyearstostudylocalizedcorrosiononbaremetalsurfaceandoncoatedalloys[6,7].Additionally,sensitizationisa localizedphenomenonwhichturns inter-estingtoapplytheLEIStechniquetolocalizethesensitizedregioninaweldingstainlesssteel.

Thus,theaimofthepresentworkistoevaluatethepoten-tialityoftheDLEPRandLEIStechniquestodeterminetheextensionofthesensitizedareainaweldedAISI304stainlesssteel.Thestudywascarriedoutusingtwoselectedweldingenergies:0.7kJmm¡1,whichdoesnotpromotethesensitiza-tionofthe304steelandusedasreference,and2.2kJmm¡1to obtain the precipitation of chromium carbides in thegrainboundariesaftertheweldingprocess.

2. Experimental

2.1. Mate­ri­al an­d we­ldi­n­g proce­s­s­

AISI 304 sheet, obtained from hot lamination process,wassuppliedbyACESITA(Brazil).Thechemicalcompo-sition was given by the manufacturer and is presented inTable1.ThecompositionisinagreementwiththestandardlimitsfortheAISI304stainlesssteel.Thesteelswereweldedusingshieldedmetalarcwelding (SMAW)techniquewiththeARCHESAWSE–308L–16electrodehaving3.25mmdiameter.TheweldingparametersarepresentedinTable2.

2.2. Me­tallographi­c e­tchi­n­gs­

MetallographicetchingaccordingtoASTMA-262wasperformed.MicrographswereacquiredusingaPhilipsXL-30scanningelectronmicroscope(SEM).Theobtainedmicro-structureswereclassifiedintothreetypes:“step”structurewithnoditchesatgrainboundaries;“dual”structure,withsome ditches at grain boundaries; and “ditch” structure,withoneormoregrainscompletelysurroundedbyditches.

Table1ChemicalcompositionofAISI304(%wt)

Element C Mn Si P Cr Ni Mo

wt% 0.02 1.39 0.46 0.03 16.82 10.06 2.03Element V Nb Sn Ti N W Cowt% 0.04 0.02 0.01 0.01 0.03 0.03 0.05

Table2Weldingparameters

Current/A Voltage/V Travelspeed/mmmin¡1

Weldingenergy/kJmm¡1

92 24 200 0.7132 28 100 2.2

SEM micrographs showing the microstructures, classifiedaccordingASTMA-262,canbefoundelsewhere[3].

2.3. DLEPR­ te­s­ts­

TheelectrochemicalcellwaspositionedontheHAZofweldedsteeltoanalysetheextensionofthesensitizedregion.Fig.1showsadrawofthesystemusedfortheDLEPRmea-surementsandanOringwasusedtomaintainthesolutioninthecellexposinganareaof1mm2.Fig.2ashowsaschemaofthediVerentpositionsoftheplatewheretheDLEPRmea-surementswereperformed.

Priortoeachexperiment,theweldedAISI304steelsur-face was polished with 400 grit emery paper, degreasedwith ethanol and cleaned in water. The working solutionwas0.5MH2SO4+0.01MKSCN.TheDLEPRtestswereconducted using a Pt foil as the auxiliary electrode and asaturatedcalomelelectrode(SCE)asthereferenceone.Theexperiments were started after nearly steady-state opencircuit potential (Eoc) had been reached (about 10min)followedbythepotentialsweep in theanodicdirectionat1mVs¡1untilthepotentialof0.3V/SCEwasreached,thenthescanwasreverseduntiltheEoc.Apotentiostat/galvano-statAUTOLABPGSTAT30,linkedwithaPCmicrocom-puterandcontrolledby theGPESsoftware,wasused foracquisitionoftheelectrochemicaldata.

2.4. LEIS me­as­ure­me­n­ts­

LEIStechniquewasusedtocorrelatethelocalelectrochem-icalbehaviourwiththemodificationofthemicrostructurein

Fig.1.DrawoftheelectrochemicalcellusedtoevaluatetheextensionofthesensitizedregioninweldedAISI304SS(DLEPRtests).

Fig.2.Schematicrepresentationsoftheweldedplate.Forthetwoelectrochemicaltechniques,thediVerentpositionswherethemeasurementswerecarriedoutareindicatedonthescheme.

Fig.3.SEMimagesshowingatypicalmicrographofthe304SSweldingwith 0.7kJmm¡1. The micrograph was observed at 4mm from the weldstring.

theHAZofthestudiedweldedmaterial.LEISmeasurementswerecarriedoutwithaSolartron1275system.Thismethodusedafive-electrodeconfigurationanddetailscanbefoundelsewhere[6,7].Asinsuchaconfigurationthelocalmeasuredcurrentdependsontheconductivityoftheelectrolyte.Thus,theexperimentswerecarriedoutin0.001MNa2SO4inorderto improve the resolutionof themethod.The local imped-ancediagramswereobtainedeach0.5mmperpendicularlytotheweldstring(Fig.2b)andwererecordedoverafrequencyrangeof10kHztoaround100MHz.Withtheusedexperi-mental set-up, only the normal component of the currentwasmeasuredandthespatialresolutionwasestimatedtobeabout1mm2foreachmeasurement.

3. Results and discussio­n

3.1. Mi­cros­tructure­ characte­ri­zati­on­

Fig.3showsaSEMimageofthe304SSsurface,weldedwith 0.7kJmm¡2, and obtained at 4mm from the weldstring.Itcanbeobservedtheabsenceofditchesonthegrainboundaries.TheobservedmicrostructurewasrepresentativeofthatobservedatdiVerentplacesintheHAZofthewelded304SS.Thissurfacemorphology,accordingtoASTM262-

A, is classified as step and associated to a non-sensitizedstainlesssteel.

The modification of the microstructure of the 304 SS,welded with 2.2kJmm¡2, with the distance from the weldstring is shown inFig.4.Themicrostructureat2mmfrom

Fig.4.SEMimagesshowingthemicrostructureatdiVerentdistancesfromtheweldstingforthe304SSweldingwith2.2kJmm¡1:(a)2mm;(b)3mm;(c)6mm;(d)9mmand(e)10mm.

theweld string (Fig. 4a) is similar to thatobserved for the304SSweldingwith0.7kJmm¡2andit isclassifiedasstep.Thisisanindicativethatfromtheweldstringtothisdistance,thetemperatureishigherthan800°C.Fromtheliterature,itisknownthattemperatureshigherthan800°Cpromotethesolubilisationofthechromiumcarbidesinthegrainmatrix,avoidingthesensitization[1].Ontheotherhand,themicro-structureat3mmfromtheweldstring(Fig.4b)presentssomeditchesaroundthegrainscharacterizingadualmicrostruc-ture.ThisindicatesthatthebeginningofthesensitizedzoneintheHAZisatabout3mmfromtheweldstring.At6mmand9mmfromtheweldstring(Fig.4candd),thegrainsare

entirelysurroundedbyditchesandthemicrostructureclassi-fiedasditch,indicatingthattheseregionsofHAZaresensi-tized.TheimagesFig.4banddsuggestthatthetemperatureofthisHAZregionisbetween800°Cand400°C,whichleadsto the precipitation of the chromium carbide in the grainboundaries.Finally,theSEMimageobtainedat10mmfromthe weld string (Fig. 4e) shows a microstructure similar tothatobservedat2mmfromtheweldstring,indicatinganon-sensitizedregioninHAZ,suggestingthatthetemperatureofthisHAZregionislowerthan400°CHAZandthattheendofsensitizedzoneisaround9mmfromtheweldstring.Thesensitizedzonehasanextensionof6mm.

3.2. DLEPR­ te­s­t

0 2 4 6 8 100.0

0.1

0.2

0.3

0.4

Ir /

Ia

Distance from weld string / mm

0.7 2.2

Welding energy / kJ mm-1

BSZ E

SZ

Fig.6.DependenceofIr/Iawiththedistancefromthefusionlineforthetwoweldingenergies.Thebeginning(Bsz)andtheend(Esz)ofthesensitizedregionareindicated.

TypicalDLEPRcurvesobtainedinthenon-sensitizedandsensitizedregionsareshowninFig.5.Ascanbeobserved,thediVerencebetweenplots is that thecurvesobtained inthesensitizedregionshowsawelldefinedreactivationpeakinthereversescan(Fig.5b).Thepeakthatappears inthereversescanisrelatedtothepreferentialbreakdownofthepassivefilmcoveringthechromium-depletedregionofthesteel.Inthistest,thedegreeofsensitizationismeasuredbydeterminingtheratioIr/Ia,whereIristhemaximumcurrentofthereversescanandIaisthemaximumcurrentinthefor-ward(anodic)scan[1].

Thevariationofthegradeofsensitization(Ir/Ia)withthedistancefromtheweldstringforthetwoweldingenergiesis shown in Fig. 6. The curve shows that the steel weldedwith theweldingenergyof0.7kJmm¡1presents the lowerIr/Iavalueswhichremainapproximatelyconstantatabout0.004,independentlyofthedistance.Ontheotherhand,thevalueofIr/Iaforthesteelweldedwith2.2kJmm¡1initiallyisabout0.005until30mmfromtheweldstring,followedbyanincreaseoftheIr/Iavalueswiththedistancereachingamaximumvalueofabout0.4at7mm,andfinally,decreases

-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3

0.000

0.003

0.006

0.009

0.012

0.015Ia

Ir

I / m

A

E / (V vs. SCE)

E / (V vs. SCE)

-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3

0.0

0.1

0.2

0.3

0.4

0.5

0.6Ia

Ir

I / m

A

a

b

Fig.5.DLEPRcurvesobtainedinthenon-sensitizedregionat2mmfromtheweldstring(a)andinthesensitizedregionat6mmfromtheweldstring(b)ofthe304stainlesssteelweldingwith2.2kJmm¡1.

forhigherdistancevaluesuntiltoreachtheminimumvalueof0.005at9mmfromweldstring.FromFig.6itwascon-firmedthat:(a)theweldingenergyof0.7kJmm¡1wasnotenoughtopromotethechromiumcarbideprecipitationintheHAZ;(b)thesteelweldedwiththeenergyof2.2kJmm¡1presents precipitation of chromium carbide in the HAZ.Theseresultsare incloseagreementwiththosepreviouslypublishedbyMajidietal.[8].Thus,thebeginning(Bsz)andtheend(Esz)ofthesensitizedzonecanbedeterminedandareindicatedinFig.6.Theextensionofthesensitizedregioninthesteelweldedwith2.2kJmm¡1isabout6mm,whichisingoodagreementwiththeSEManalysis.

3.3. LEIS te­s­t

Fig.7showsexamplesofsomelocalimpedancediagramsobtained for the AISI 304 SS welded with both weldingenergy. The three diagrams in each plot were obtained atdiVerentpositionsfromtheweldstring.Forbothsystems,thediagramsarecharacterizedinthehighfrequency(HF)rangebyacapacitivearc(notclearlydefined)followedbyan inductive loop, and in the lower frequency range (LF)byanothercapacitivepart.Theoriginofthefirstcapacitiveloopisnotyetexplained.Thepresenceofaninductivelooponthelocalimpedancediagramwasrecentlyexplainedbythe influenceofcurrentandpotentialdistributionsassoci-atedwiththegeometryoftheelectrode[9–11].Inaddition,inarecentwork,inductiveloopsintheHFpartofthelocalimpedancediagramswerealsoobservedforthecouplingofpurealuminiumwithpurecopperwhentheprobeanalyzesthecopperbehaviour[12].Inthiscase,thepresenceoftheinductiveloopwasconnectedtotheexistenceofagalvaniccoupling between the two materials, with the Cu playingtheroleofcathode.ThecouplingbetweenCuandAlwouldleadtoparticularcurrentandpotentialdistributionsassoci-atedwiththisconfiguration(electriceVect).Byanalogywiththiswork,itisprobablethatinthepresentstudyagalvaniccouplingphenomenonbetweentheweldstringandthestain-

Fig.7.Localimpedancediagramsobtainedonthewelded304SSsurfaceatdiVerentdistancesfromtheweldstring:(a)weldingwith0.7kJmm¡1and(b)weldingwith2.2kJmm¡1.

Fig. 8. Comparison of the normalized impedance modulus measured at0.5Hzwiththedistancefromtheweldstring.

lesssteelisalsooccurring.Thisassumptionwasreinforcedbythevisualobservationoftheplatesaftertheelectrochemi-caltests.Corrosionproductswereclearlyvisibleclosetotheweldstring,suggestingthatthestainlesssteelplatecouldbeactingascathodeinthecorrosionprocess.

Independentlyoftheweldingenergy,thediagramsplot-ted close to the welding (d=0mm) are relatively similar.ThisisexplainedbythefactthatthiszoneisaVectedinthesamewaybytheweldingprocess.InFig.7a,itcanbeseena purely capacitive behaviour (low frequency part of theimpedancediagrams)whentheprobemovesawayfromtheweldstringandadditionally,thediagramsarerelativelysim-ilarforthetwopositionsoftheprobe.Onthecontrary,fortheAISI304SSweldedwith2.2kJmm¡1,afrequenciesshiftisobservedontheLFcapacitivepartofthediagramswhenthepositionoftheprobeischanged(Fig.7b).

Tocomparemoreeasilytheresultsobtainedforthetwosamples,thefrequencyof0.5Hz,correspondingtothecapac-itivepartinLFrange,waschosen.Fig.8presentsthevaria-tionofthemodulusofthenormalizedimpedanceat0.5Hzalong a line perpendicular to the weld string for the twoweldingenergies.Measurementscarriedoutontheweldedplatewith0.7kJmm¡1constitutedthereference.Theimped-ancemodulus for the reference increasesgraduallyon thefirsttwomillimetresstartingfromtheweldstring,thenitsvalue remains relatively constant. For the welding energyof2.2kJmm¡1,thevariationoftheimpedancemoduluscanbe clearly connected to the localization of the sensitizedregion, since it is observed a reduction in the impedancemodulusforthesensitizedregion,comparedwiththeSEMimage(ingreyedonthefigure).ItisobservedinFig.3cthattheweldingmicrostructureofthestainlesssteelisthemostsensitized at 6mm. This point corresponds to the lowestvalueoftheimpedancemodulusonFig.8.At10mmfromtheweldstring,themicrostructureistypicalforanon-sen-sitizedmaterial (Fig.3e).Thispointcorresponds inFig.8toahighervalueofthemodulus.Theincreaseinthemod-ulus inthevicinityoftheweldstringis inagreementwiththemicrostructureobservedontheFig.3a,whichshowedthatinthisHAZregion,thechromiumcarbideprecipitatesweredissolvedinthegrainmatrix.Fortheweldingenergyof 2.2kJmm¡1, the values of the modulus are lower thanthoseobtainedfortheweldingenergyof0.7kJmm¡1.ThisresultcanbeexplainedbysomediVerencesinthepropertiesoftheoxidesformedonthestainlesssteelsurface.ThesameconclusionsfromFig.8canbemadebyfollowingthevaria-tionoftheimpedancephasealongthelineperpendiculartothewelding,showninFig9.IntheHAZ,phasevaluesarestronglydiVerentiatedcomparedtothoseobtainedforthesystemwhichisnotsensitized.

Thus,theanalysisofthelocalimpedancediagramshowsthat the sensitized zone in the HAZ was detected by theLEIS technique at a frequency of 0.5Hz. To visualize the

Fig.9.Comparisonof the impedancephasemeasuredat0.5Hzwith thedistancefromtheweldstring.

HAZ, it is necessary to work in a well defined frequencyrange (about the Hertz), which corresponds to interfacialphenomena.

4. Co­nclusio­ns

Thenon-destructiveDLEPRandLEIStestsallowedthelengthoftheSZtobedeterminedandaverygoodcorrela-tionbetweenthetwotechniquesandtheASTM262Stan-dard Practice was observed. The presence of an inductivelooponthelocalimpedancediagramsprobablyreflectagal-vaniccouplingbetweentheweldstring(anode)andplatesoftheweldedstainlesssteel(cathode)whichwillbeverydetri-mentalforagoodcorrosionresistanceoftheweldedsystem.

Inthefuture, itwillbenecessarytoobtainmoreinforma-tionfromlocalimpedancespectraparticularlyregardingthehigh frequencypartof thediagrams.Theresultsobtainedshow the feasibility of developing the DLEPR and LEIStestsinindustrialfield.

Ackno­wledg­ments

TheauthorsthankCNPq,ANP,FINEPandFUNCAP,Brazil,forfinancialassistance.WalneyS.AraújoalsothanksCNPq/CNRSproject490105/2004-1.

Ref­erences

[1] A.JohnSedriks,CorrosionofStainlessSteels,seconded.,J.Wiley&Sons,1996.

[2] S.S.M.Tavares,P.deLima-Neto,M.P.C.Fonseca,A.Maia,J.Mater.Sci.38(2003)3527.

[3] A.S. Lima, A.M. Nascimento, H.F.G. Abreu, P. de Lima-Neto, J.Mater.Sci.40(2005)139.

[4] S.S.M.Tavares,V.F.Terra,P.deLima-Neto,D.E.Matos,J.Mater.Sci40(2005)4025.

[5] J.A.Souza,H.F.G.Abreu,A.M.Nascimento,J.A.C.dePaiva,P.deLima-Neto,S.S.M.Tavares,J.Mater.Eng.Perform.14(2005)367.

[6] G.Baril,C.Blanc,M.Keddam,N.Pébère,J.Electrochem.Soc.150(2003)B488–B493.

[7] J.-B.Jorcin,E.Aragon,C.Merlatti,N.Pèbère,Corros.Sci.48(2006)1779andreferencescitedtherein.

[8] A.Z.Majidi,M.A.Streicher,Corrosion40(1984)584.[9] V.M.Huang,V.Vivier,M.E.Orazem,N.Pébère,B.Tribollet,J.Elect-

rochem.Soc.154(2007)C81–C88.[10] V.M.Huang,V.Vivier,I.Frateur,M.E.Orazem,B.Tribollet,J.Elect-

rochem.Soc.154(2007)C89–C98.[11] V.M.Huang,V.Vivier,M.E.Orazem,N.Pébère,B.Tribollet,J.Elect-

rochem.Soc.154(2007)C99–C107.[12] J.-B.Jorcin,PhDThesis,INPToulouse(2007).