SVET method for characterizing anti-corrosion performance of metal-rich coatingsMaocheng Yan *,Victoria J.Gelling **,Brian R.Hinderliter,Dante Battocchi,Dennis E.Tallman,Gordon P.BierwagenDepartment of Coatings and Polymeric Materials,North Dakota State University,Fargo,ND 58108,USAa r t i c l e i n f o Article history:Received 17January 2010Accepted 12April 2010Available online 28April 2010Keywords:A.Metal-rich coating A.SteelB.SVETB.Microelectrode techniqueC.Galvanic interactiona b s t r a c tThe galvanic interaction between a metal-rich coating and the underlying metal substrate was character-ized by a new analysis method based on the scanning vibrating electrode technique (SVET).The total ano-dic current at various immersion periods was evaluated by integrating the anodic current density on SVET maps.Zinc-rich paints (ZRPs)coated on a steel panel were used to demonstrate the experimental approach.The anti-corrosion performance of the ZRP was analyzed based on the integrated anodic cur-rent and the experimental E OC –i Int diagram.Closely correlative behaviour was found between the inte-grated anodic current and the open-circuit potential.Published by Elsevier Ltd.1.IntroductionAs one of the most cost effective methods for corrosion protec-tion of metallic objects,organic coatings provide corrosion protec-tion mainly by four ways:a barrier effect,sacrificial cathodic protection,corrosion inhibitor release,and anodic protection.Metal-rich coatings (MRCs)[1,2]are a class of corrosion protection coatings containing sacrificial metal pigments that are more elec-trochemically reactive than the underlying metal substrate,which inhibit corrosion by providing sacrificial/cathodic protection to the metal substrate.MRCs are generally designed with high volume fraction of metal pigment (near critical pigment volume concentra-tion,CPVC)dispersed in non-conductive polymer or inorganic matrix.The most effective and commonly used MRCs for steels are Zn-rich primer (ZRP)coatings [3–5].Most recently,Mg-rich coatings have been developed and found to provide similar protec-tion to aerospace Al alloys [6–8].Various electrochemical methods have been employed to assess the anti-corrosion performance of metal-rich coatings,such as cor-rosion potential measurements,electrochemical impedance spec-troscopy (EIS)[3,9–12],electrochemical noise methods (ENM)and galvanic coupling measurement [3,13].Murray [9,14]reviewed electrochemical methods used for evaluating organic anti-corrosion coatings.Sekine [15]gave a review on characteristics of various electrochemical measurement methods and even their correlations.The development of microelectrode techniques and scanning elec-trode techniques has made it possible to measure electrochemical processes on a local scale,which has yielded new types of informa-tion relevant to the local electrochemical processes on corroding surfaces and advanced the investigations of localized corrosion.Among all the techniques are the Scanning Vibrating Electrode Tech-nique (SVET),the Scanning Reference Electrode Technique (SRET),Local Electrochemical Impedance Spectroscopy (LEIS),and the Scan-ning Kelvin Probe (SKP).SVET was originally devised for detecting the extra-cellular cur-rent near living cells in the 1970s [16].It was firstly developed to study localized corrosion processes by Isaacs in the 1980s [17,18].The electrochemical process of corrosion contains an ionic current flow in the electrolyte balanced by the electron flow through the metal.The ionic current flow causes a potential gradi-ent to exist in the solution at the electrochemically active site.SVET was designed to detect the potential gradient via a movable vibrating microelectrode.The electrode potential difference between the two extreme points of its vibration,r /,is recorded at the extremes of the vibration amplitude,generating a sinusoidal AC signal.The AC signal is then converted to the ion current density (i )by a calibration procedure [10,19].The local current is related to r /and the electrolyte conductivity k by Ohm’s law [20]i ¼Àj r /ð1ÞSVET systems are designed to oscillate the probe in a Lissajous mode so that both parallel component i x and perpendicular compo-nent i z of the current can be obtained by partial differentiation of (1)with respect to x or z .0010-938X/$-see front matter Published by Elsevier Ltd.doi:10.1016/j.corsci.2010.04.012*Corresponding author.Tel.:+17012318027;fax:+17012318439.**Corresponding author.E-mail addresses:Maocheng.Yan@ (M.Yan),V.J.Gelling@ (V.J.Gelling).Corrosion Science 52(2010)2636–2642Contents lists available at ScienceDirectCorrosion Sciencej o u r n a l h o m e p a g e :w w w.e l s e v i e r.c o m /l o c a t e /c o r s ci1.1.SVET for characterizing anti-corrosion performance of coatingsSVET has enjoyed wide acceptance as a powerful electrochemi-cal technique for evaluation of corrosion inhibitor,detection of cor-rosion activity and quantification of corrosion defects in coatings. SVET has been used in the research of various types of corrosion, such as pitting[21],cut-edge corrosion[22–24],galvanic corrosion [18],microbiologically influenced corrosion(MIC)[25],weld corro-sion,and stress corrosion cracking(SCC)[26].In the case of corro-sion of a coated metal,SVET is able to give detailed insights into the electrochemical interactions between a coating and its sub-strate at a defect,which has been provided valuable information on the anti-corrosion mechanism by a coating,including the gener-ation and development of defects,and the influence of pigments/ inhibitors on corrosion of substrate at a defect[27,28].In previous studies from this laboratory,series of coatings for Al alloy(AA)2024-T3has been characterized by SVET.To monitor both the corrosion activity of the substrate and the possible gal-vanic interaction between the coating and the substrate,the mea-surement was conducted in the vicinity of a scratch exposing the underlying substrate.The SVET results for polypyrrole deposited on AA2024-T3showed that a large anodic current occurred at the defect due to anodic dissolution of the alloy and that the catho-dic current was rather uniformly distributed over the polymer sur-face,which implies that a p-doped CP would promote active dissolution of the AA2024-T3substrate at the defect if the passiv-ation could not be obtained[29,30].Most recently,interesting interactions between neutral or n-doped poly(2,3-dihexylthieno-[3,4-b]pyrazine)and AA2024-T3has been demonstrated by SVET [31].The n-doped conjugated polymer exhibited the ability to sac-rificially protect the exposed Al alloy in a defect.For a redox inactive barrier coating,such as a plain epoxy coat-ing on steel or AA2024-T3,the SVET showed that both anodic cur-rent and cathodic current were located at the scribe[32].Due to the high impedance/low-conductivity of the intact barrier coating, a complete corrosion cell,if any,would be established within the defect,and no current was distributed on the coating.In the case of metal-rich coatings,such as Mg-rich primer coated on Al alloys, the SVET results exhibited a well-defined cathodic current peak above the scratch.The anodic current(related to the anodic disso-lution of the sacrificial pigment)distributed on the primer,which demonstrates that the sacrificial pigment functions by the cathodic protection mechanism[1,33,34].In this work,a SVET method is provided for further understand-ing the galvanic interaction between a metal-rich coating and the metal substrate,and evaluating the anti-corrosion performance of the metal-rich coating.A zinc-rich primer(ZRP)coated on steel was examined to demonstrate the efficacy of the method.Several characteristic indexes are obtained from the SVET current density map to characterize the galvanic interaction between the ZRP and its substrate.The total anodic current(and hence the corrosion rate)over the scan area is evaluated by integration of the overall anodic current density on the SVET maps.The variations of the to-tal anodic current and that of open-circuit potential(E OC)were analyzed as a function of the immersion time.Additionally,the SVET current indexes were compared with the galvanic current ob-tained by zero resistance amperometry(ZRA).2.Experimental2.1.Materials and electrode preparationAn epoxy resin(Epon828,from Hexion)and a modified poly-amide(Epikure3175,from Hexion)curing agent were mixed in a 1:1.1stoichiometric ratio.This ratio results in the optimal barrier properties and hardness of the primer by giving near to the max-imum amount of crosslinking.Methyl isobutyl ketone(MIBK) was used as solvent.Then,zinc powder was added to the solu-tion and was stirred to form a thick mortar-like mixture.A steel panel(R-35,from Q-Panel)was pretreated by grinding with400-and600-grit SiC sandpaper,followed by degreasing with hexane. The coatings were applied using a drawn down bar at a wet thickness of200l m.The coated panels were placed in a convec-tion oven at70°C for24h afterflashing off for approximately 30min.For the primers of pigment volume concentration(PVC) lower than25%,the zinc pigment wasfirstly dispersed in MIBK for full dispersion.The dryfilm thicknesses were in the range of140–190l m.2.2.SVET measurement and data analysisThe current distribution over the interface of solution/ZRP(67% PVC,subsequently referred to as ZRP67)was measured using a SVET system from Applicable Electronics(USA).The Pt–Ir microelectrode (Microprobe Inc.)with a10l m diameter tip which was platinized to a20l m diameter sphere.The microprobe was vibrated$200l m above the samples with the amplitude20l m along the X and Y directions.A pair of platinized platinum wires was used as both the reference and bath ground electrodes.The probe made20Â20 measurements in each scan($600s),generating a400-point mesh across the surface.Scans were initiated5min after immersion and repeated every60min.The ZRP67sample(1Â1cm2)was masked by a polyester tape,exposing an open area of3Â3mm2as the scan-ning area.An artificial scratch was introduced in the center of the scanning area.For comparison,the same SVET measurement was also conducted on an unscratched ZRP67.All SVET measurements were performed at the free-corrosion condition in a cell containing $5mL3.5wt.%NaCl aqueous solution.The SVET current density mapping and the statistical analysis of the data were performed with Origin software.The current densi-ties were displayed in three-dimensional(3D)maps,showing the spatial distribution of the current density as a function of the(x, y)position in the scan region on ZRP.The current values in the SVET map are positive for anodic currents and negative for catho-dic currents.The contour map of the current densities is at the bot-tom of the3D map.The SVET current density vector images superimposing the measured current vector onto an optical image of the sample showed images of the sample surface as well as the locations of anodic/cathodic area.Based on the SVET current density map,the anodic current den-sity peak(i A,max),the cathodic current density peak(i C,max),the average current density(i Ave)and the integrated anodic current (I Int)were used to characterize the anti-corrosion properties of the coating.I Int was evaluated by integration of the overall anodic current(I A)on a SVET current density map,which is theoretically equal to the total cathodic current(I C)over the ZRP surface.Split-ting the scan area(S,3Â3mm2)into20Â20small squares,we calculate the anodic or cathodic current on each square,and sum all the resulting currents to obtain I Int(l A)on the scan area,as shown byI Int¼SPi An¼ÀSPi Cnð2Þwhere i A is the anodic current density(i A P0),i C the cathodic cur-rent density(i C<0)and n the number of measurement points in each scan(n=400).2.3.Galvanic coupling measurementGalvanic coupling(both the mixed potential,E Mix,and the cou-pling current)between the coated metal and the bare substrateM.Yan et al./Corrosion Science52(2010)2636–26422637was measured using a Gamry PC4/300potentiostat in a zero resis-tance ammeter(ZRA)mode.The experiments were carried out in a two-compartment enclosed cell described in our previous work [29].The two-compartment cell permitted careful control of the atmospheric conditions in each compartment.The working elec-trode in one compartment was ZRP67-coated on the steel(subse-quently referred to as the ZRP-compartment)and the working electrode in the other compartment was the bare steel(subse-quently referred to as the steel-compartment).The exposed area of ZRP67was 1.0cm2and the bare steel was in a pinhole of 0.04cm2(simulating a coating defect),yielding an area ratio (ZRP67to steel)of ca.25.To simulate the condition for a top-coated sample,where a topcoat would protect the primer from di-rect oxygen access,the solution in the coating-compartment was purged with N2,while the solution in the alloy-compartment was purged with air.3.Results and discussion3.1.SVET current density mapsThe SVET current density maps for the bare steel in3.5%NaCl solution,as presented in Fig.1,showed several anodic current peaks that appear after several minutes,due to possible pitting nucleation.After30min,these anodic current peaks combined into one broad anodic peak,where dark corrosion products began to ap-pear on the steel surface.Fig.2displays SVET current density maps above the defect on the scratched ZRP67(67%PVC)after various immersion periods in3.5%NaCl solution.The cathodic current was mainly located at the scratch where a well-defined cathodic peak existed throughout the5-day immersion period.The anodic area appeared at different sites on ZRP67during the immersion.At the beginning of the immersion from0.1to4h,anodic areas were found to initiate only at the corners of the scratch,as shown in Fig.2a and b.After5h of the immersion,anodic areas were scattered around the scratch (Fig.2c).After10h of the immersion,significant changes occurred both in current distribution and in value,as presented in Fig.2d. The anodic activity moved from one area to another near the scratch;the cathodic area included to almost all the scratch and a well-defined cathodic peak was observed.After20h of the immersion(Fig.2e and f),the cathodic current decreased with time,and the anodic current was evenly distributed over the sur-face of the primer.The variations of the anodic current peak(i A,max),the cathodic current peak(i C,max)and the average current density(i Ave)in SVET maps are shown in Fig.3.The i A,max decreased over the immersion time on the whole,except for two peaks appearing at3and10h. The SVET experiment was conducted under free-corrosion condi-tion(without external polarization applied)where the anodic cur-rents and cathodic currents above the ZRP67are balanced and the net current should be zero.It should be note that,in a scanning plane above the free-corro-sion surface,the integrated anodic current I A should be theoreti-cally equal to the integrated cathodic current I C in the absolute value and hence the average current density(i Ave)should be zero. But deviations between I A and I C are usually obtained by SVET, which causes i Ave to deviate from zero.The deviations may be attributed to the fact that the current density on a SVET map was not taken at the same time.The corrosion behaviour and current distribution on the scan area are changing during scanning(one scan takes$10min).3.2.Integrated anodic current of the ZRP obtained by SVETThe ZRP paints under study here are heterogeneous systems with pores and zinc particles distributed randomly in the binder. The probable reactions on the primer in an electrolyte are as fol-lows:Zinc dissolution to the oxide(ZnO)/polymeric-binderfilm, electrochemical dissolution of the active zinc particles,and oxygen reduction on zinc particles or on the substrate through pores[35].A key aspect of the above mentioned mechanisms is the galvanic interaction between ZRP and the metal substrate.The galvanic interaction between ZRP and substrate may be influenced by any or all of the following factors:the electrochemical state of zinc par-ticles at ZRP/solution interface,reactive Zn/Fe area ratio(S Zn/Fe) interface,as well as the diffusion process through the coating and the deposit of Zn corrosion products[3,12].The electrochem-ical behaviour and cathodic protection performance of ZRPs have been well studied by corrosion potential monitoring[3],EIS [3,12,36],conductive atomic force microscopy(AFM)[2],as well as the scanning electron microscopy(SEM)[4,12].In this work, the electrochemical and corrosion performance of the ZRP was characterized by the SVET method.The integrated anodic current I Int above the ZRP67obtained by the SVET method is presented in Fig.4as a function of the immer-sion time,together with E OC measured under the same conditions. For the scratched ZRP67,closely correlative behaviour was found between E OC and the total anodic current.Most obviously,two sig-nificant anodic current peaks appeared during the immersion ex-actly before and after the E OC peak occurred.For comparison,I Int and E OC for the unscratched ZRP67are also presented in Fig.4. The trend of I Int of the unscratched ZRP67was similar to that of the scratched ZRP67but with much lower amplitude.The most2638M.Yan et al./Corrosion Science52(2010)2636–2642positive potential of ZRPs(unscratched)reached at$3h immersion and this potential was observed to depend significantly on the PVC, as shown in Fig.5.The most positive potentials for ZRPs with PVC5%,15%,45%and67.5%were0.166,0.00,À0.60andÀ0.91V,respec-tively.The following several stages were clearly recognized from both I Int and E OC shown in Fig.4for the scratched ZRP67.distributions over an artificial line scratch on ZRP(67%superimposed with current vectors(right of a and f)M.Yan et al./Corrosion Science52(2010)2636–264226393.2.1.The activating stageThe E OC(ca.À0.9V initially)shifted gradually towards negative direction during thefirst1h immersion as a result of the activation of Zn particles through dissolving ZnOfilm or soaking the binder film,approximatingÀ1.0V(E OC of the zinc particles)at$1h.In the following3h,the E OC shifted in the positive direction,implying that the steel substrate began to be wetted by the electrolyte through pores(decreasing the ratio of S Zn/Fe).The I Int($1.2Â10À2l A initially)slightly decreased in thefirst 1h immersion.Then,it gradually increased as a result of the disso-lution of ZnO/zinc particles at the ZRP surface.From3h immer-sion,I Int decreased sharply accompanying with the rapid increase of E OC due to the wetting process of the steel substrate until E OC reaching a peak(À0.71V)at5h,where the steel substrate was ex-pected to be totally wetted.A current valley(7.3l A)exactly corre-sponded to the E OC peak at5h.Once the substrate was wetted, galvanic interaction between zinc particles and steel substrate would be expected to occur.The period of the beginning$5h immersion can be referred to as the activating stage,where the main process was the dissolution of the ZnOfilm and then the zinc particles reaction with the electrolyte.In the activating stage,the electrochemical reaction process mainly occurred on the ZRP/solu-tion interface.3.2.2.The sacrificial protection stageEven after the steel surface became completely wetted at$5h in Fig.4,some zinc particles were still covered by thick oxides or by binderfilms.Zinc particles continued to be activated by the dis-solution of the ZnOfilm and/or galvanically coupling to the sub-strate.At the beginning of the galvanic stage,corrosion product was formed in pores in the primer,which tended to seal the pores, reducing the number and size of pores[3].This process shifted E OC toward negative directions and increased I Int sharply until attain-ing2.3Â10À2l A at$11h.Thefluctuation of E OC in the range ofÀ0.78toÀ0.87V in the vicinity of9h and the significant decrease in I Int at$10h might be attributed to the accumulation of Zn corrosion product which improved the barrier property of the coating.For the unscratched ZRP67,the E OCfluctuation at$8h immersion disappeared,which further implied that the E OCfluctuation might be related to the de-posit of the zinc corrosion product in the defect and its adhesion to the surface.3.2.3.The barrier stageThe E OC continued to decay,with somefluctuations,beginning from$11h(Fig.4).I Int decayed to a low value of1.2Â10À2l A, where it remained for the following duration of the experiment. The distinct decrease in I Int may be attributed to the zinc depleting and/or the improved barrier property of the primer.The barrier properties of the primer would improve by deposit of the corrosion product on the primer.The zinc corrosion product also trends to improve barrier effect by sealing pores in the primer,as has been proposed and observed by other authors[3,4,36].3.2.4.E OC–i Int diagramThe evolution of the E OC–i Int results for the scribed ZRP67is shown in Fig.6,with the numbers indicating the immersion time (in hour)and the dashed lines showing the shifting direction of the E OC–i Int points.Several stages(the wetting stage,the cathodic protection stage and the barrier stage)were demonstrated in the E OC–i Int evolution for the scratched ZRP67during the immersion period.Almost all the conventional electrochemical measurements re-quire an externally imposed polarization which would possibly have an unfavorable effect on the specimen.By contrast,SVET al-lows to measure integrated corrosion current at the free-corrosion condition(without external polarization applied)which is benefi-cial,especially for the highly active metal-rich coatings.2640M.Yan et al./Corrosion Science52(2010)2636–26423.3.Galvanic coupling between ZRP and the bare steelThe coupling current and mixed potential for ZRP67(1.0cm2,N2 purged)coupled with bare steel(0.04cm2,air purged)in3.5%NaCl solution are shown in Fig.7.The coupling current started at 4Â10À3l A and rapidly increased to12Â10À3l A at$25min.After immersion,the zinc particles were gradually activated through dis-solution of the oxide(ZnO)surface and water penetration through polymeric-binderfilm.The zinc particles were fully activated at the end of1h,where the mixed potential attained the lowest value and the coupling current was highest.Then the mixed potential in-creased and the coupling current declined steadily.At$26h,the E Mix attainedÀ0.85V,and the coupling current was6.9Â10À3l A.In the case of the galvanic coupling measurement in the two-compartment cell,the bare steel pinhole(simulating coating de-fect)is separated in another cell,where the bare steel was not af-fected by the deposit of the zinc corrosion product in the defect. The depletion of the zinc particles and the lack of the corrosion product deposit may lead to the steady decrease of the coupling current,which was a fundamental difference from the case of the SVET measurement,where the barrier stage was clearly observed as the result of a barrier effect of the Zn products deposit.4.ConclusionsBy integrating the overall anodic current density measured by scanning vibrating electrode technique(SVET),the total anodic current(and therefore the corrosion rate)over the coating surface was obtained and used to evaluate the galvanic interaction be-tween coatings and substrates.This SVET method allows the mea-surement of corrosion current at the free-corrosion condition (without external polarization applied),which is highly beneficial, especially for active metal-rich coatings.For the zinc-rich primer, several stages during immersion were distinctly recognized by the SVET method:the activating stage,the sacrificial protection stage and the barrier stage.Closely correlative behaviour was found between the total anodic current and the open-circuit poten-tial.This SVET analysis method may provide a new insight for the corrosion protection performance and even the service life of me-tal-rich coatings.AcknowledgmentThe authors would like to thank the US Army Research Labora-tory(Contract#W911NF-04-2-0029)for sponsoring this research. References[1]G.Bierwagen,D.Battocchi,A.Simoes,A.Stamness,D.Tallman,The use ofmultiple electrochemical techniques to characterize Mg-rich primers for A1 alloys,.Coat.59(2007)172–178.[2]G.Bierwagen,K.Allahar,B.Hinderliter,H.Jung,Zn-rich coatings revisited,in:Tri-Service Corrosion Conference,Denver,Co.,2007.[3]C.M.Abreu,M.Izquierdo,M.Keddam,X.R.Novoa,H.Takenouti,Electrochemical behaviour of zinc-rich epoxy paints in3%NaCl solution, Electrochim.Acta41(1996)2405–2415.[4]M.Morcillo,R.Barajas,S.Feliu,J.M.Bastidas,A SEM study on the galvanicprotection of zinc-rich paints,J.Mater.Sci.25(1990)2441–2446.[5]C.Hare,Corrosion control of steel by organic coatings,in:R.W.Revie(Ed.),Uhlig’s Corrosion Handbook,John Wiley&Sons,New York,2000,pp.1023–1038.[6]M.E.Nanna,G.P.Bierwagen,Mg-rich coatings:a new paradigm for Cr-freecorrosion protection of Al aerospace alloys,JCT Res.1(2004)69–80.[7]D.Battocchi, A.M.Simoes, D.E.Tallman,G.P.Bierwagen,Electrochemicalbehaviour of a Mg-rich primer in the protection of Al alloys,Corros.Sci.48 (2006)1292–1306.[8]G.Bierwagen,R.Brown,D.Battocchi,S.Hayes,Observations on the Testing ofMg-rich Primers for Totally Chromate-free Corrosion Protection of Aerospace Alloys,Department of Defense Corrosion Conference,Gaylord National, Washington DC,2009.[9]J.N.Murray,Electrochemical test methods for evaluating organic coatings onmetals:an update.Part III:multiple test parameter measurements,.Coat.31(1997)375–391.[10]G.Grundmeier,W.Schmidt,M.Stratmann,Corrosion protection by organiccoatings:electrochemical mechanism and novel methods of investigation, Electrochim.Acta45(2000)2515–2533.[11]F.Mansfeld,Use of electrochemical impedance spectroscopy for the study ofcorrosion protection by polymer-coatings,J.Appl.Electrochem.25(1995) 187–202.[12]J.R.Vilche,E.C.Bucharsky,C.A.Giudice,Application of EIS and SEM to evaluatethe influence of pigment shape and content in ZRP formulations on the corrosion prevention of naval steel,Corros.Sci.44(2002)1287–1309. [13]S.E.Faidi,J.D.Scantlebury,P.Bullivant,N.T.Whittle,R.Savin,Anelectrochemical study of zinc-containing epoxy coatings on mild steel, Corros.Sci.35(1993)1319–1328.[14]J.N.Murray,Electrochemical test methods for evaluating organic coatings onmetals:an update.Part II:single test parameter measurements,.Coat.31(1997)255–264.[15]I.Sekine,Recent evaluation of corrosion protective paintfilms byelectrochemical methods,.Coat.31(1997)73–80.[16]L.F.Jaffe,R.Nuccitelli,An ultrasensitive vibrating probe for measuring steadyextracellular currents,J.Cell Biol.63(1974)614–628.[17]H.S.Isaacs,Initiation of stress corrosion cracking of sensitized type304stainless steel in dilute thiosulfate solution,J.Electrochem.Soc.135(1988) 2180–2183.[18]H.S.Isaacs,The measurement of the galvanic corrosion of soldered copperusing the scanning vibrating electrode technique,Corros.Sci.28(1988)547–558.[19]J.Elvins,J.H.Sullivan,J.A.Spittle,D.A.Worsley,Short term predictive testingfor cut edge corrosion resistance in zinc–aluminium alloy galvanised steels, Corros.Eng.Sci.Tech.40(2005)43–50.[20]J.He,Applications of the Scanning Vibrating Electrode Technique to the Studyof Corrosion Protection by Conductive Polymers,Ph.D.Thesis,North Dakota State University,2002.[21]H.Krawiec,V.Vignal,R.Oltra,Use of the electrochemical microcell techniqueand the SVET for monitoring pitting corrosion at MnS inclusions,Electrochem.Commun.6(2004)655–660.M.Yan et al./Corrosion Science52(2010)2636–26422641。
