参考文献(15)阻抗谱分析

The application of AC impedance to study the performance of lacquered aluminium specimens in acetic acid solutionJ.D.Scantlebury a,K.Galic´b,*a Corrosion and Protection Centre,University of Manchester Institute of Science and Technology,P.O.Box88,Manchester,M601QD,UKb Faculty of Food Technology and Biotechnology,Laboratory of Physical Chemistry and Corrosion,University of Zagreb,Pierottijeva6,HR-10000Zagreb,CroatiaReceived4March1996;revised version received15November1996;accepted23January1997AbstractAluminium tubes with single and double coat solvent based and water based lacquers,based on epoxy-phenolic resins,were analysed.To determine the electrochemical parameters to correlate with the actual behaviour of a collapsible tube,impedance spectroscopy was used. The measurements were performed in3%(v/v)acetic acid,at room temperature.After impedance measurements specimens surface were analysed by scanning electron microscopy(SEM).As the coating degraded from penetration of electrolyte a rapid increase of coating capacitance,C p,(from10−10to10−8F cm−2)and a rapid decrease in pore resistance,R p,(from108to106Q cm2)occurred after192h of immersion.Analysis of data from impedance measurements,such as R p,C p,and delaminated area(A del),enabled determination of coating performance exposed to an aggressive media.©1997Elsevier Science S.A.Keywords:Acetic acid;AC impedance;Aluminium;Delamination;Epoxy/phenolic resins;Immersion test1.IntroductionFood manufacturers are requiring better performance from food can lacquers especially those aimed as aggressive foodstuff packaging.At the same time there has been a significant change from organic solvents to water based lacquer systems which requires better methods for analysing lacquered aluminium corrosion behaviour.Since interac-tions between the food products and the cans are essentially electrochemical in nature,electrochemical tests are highly suitable for rapid testing of the behaviour of metal cans (tubes),particularly lacquered ones.The processes occurring at the metal–electrolyte inter-face are numerous and different in nature.The presence of an organic coating makes the system even more complex,as the electrical and electrochemical properties of the coating are introduced.As explained by Tait[1]traditional DC measurements cannot fully describe the situation.A system like this may be analysed and measured by impedance tech-niques.The applications of AC impedance to food packa-ging materials has shown the importance of electrochemical parameters,enabling both lacquer quality and corrosion evaluation in the presence of aggressive food products,to be analysed[2–4].The aim of this work was to investigate the corrosion performance of lacquered aluminium tubes used for tomato puree´packaging.In order to simulate aggressive conditions found in such a food product3%acetic acid was used as an electrolyte.2.ExperimentalAluminium(purity=99.7%)tubes((3.5×18.5cm,P= 2.03dm2)lacquered with epoxy-phenolic resins were obtained directly from a producer(‘TOP’,Zagreb).Two sets of specimens were obtained:1.aluminium tubes with solvent based and2.water based lacquer.Both sets of specimens were available as a single(5m m) and double coated(10m m)aluminium tubes.In order to verify coating thickness,ten measurements were made along a length of the test specimens,before and after removal of lacquers using‘cellosolve’as reagent. The thickness specified for the measured specimensrepre-Progress in Organic Coatings31(1997)201–2070300-9440/97/$17.00©1997Elsevier Science S.A.All rights reservedPII S0300-9440(97)00005-2*Corresponding author.sents an average thickness of coating on duplicate samples measured by micrometer.Specimens:1.Al/single-coat solvent based lacquer(S1x)2.Al/double-coat solvent based lacquer(S2x)3.Al/single-coat water based lacquer(W1x)4.Al/double-coat water based lacquer(W2x)The working electrodes were prepared by cutting the tubes into sheets,masking the edges so thefinal active sur-face area was35cm2.The test solution was3%v/v acetic acid(pH=2.8)in distilled water which is,according to EEC rules[5]speci-fied as appropriate to simulate an aggressive product such as tomato paste.Measurements were performed at room tem-perature.A two identical electrode system,galvanostatically con-trolled(the applied potential was set to0volts,assuming that the potential of the working electrode is its corrosion potential)which is recommended for use in cases of high resistance systems[6],was used for impedance measure-ments(Four Channel Frequency Response Analyser,Solar-tron Schlumberger1254,plus potentiostat ACM P325VI, combined with a computerised system for data analysis and storage).Impedance measurements were carried out by applying a sinusoidal signal of15mV in amplitude in a frequency range of10kHz–10mHz.In order to evaluate the impedance changes of the system with time,specimens were left immersed in3%acetic acid (v=500ml)and analysed at regular intervals.The obtained impedance data represent average of duplicate samples. Scanning electron microscopy(SEM),coupled to X-ray aluminium mapping was performed after making the impe-dance measurements.From the impedance data,the capacitance and resistance of the paintfilm were determined byfitting a simple RC circuit to the impedance data using the appropriate software. In order to evaluate the impedance changes of an electro-chemical cell with frequency,it is reasonable to consider a hypothetical equivalent circuit which is a combination of elements of the system under corrosion.If the circuit con-sists of a metal coated with an organicfilm,the equivalent circuit can be represented as in Figs.1a,b,where R o repre-sents the resistance of the electrolyte,R p is pore resistance due to electrolyte penetration,C p is the capacitance of the intact coating layer,R ct represents the charge transfer resis-tance and C dl is the capacity of the faradaic reaction at the metal/electrolyte interface.Modeling of impedance data for coating delamination was performed by Haruyama et al.[7]who suggested that the decrease of R p and R ct and the increase of C dl with exposure time were caused by an increase of delaminated area(A del)according to:R p=R o pA del(1) R ct=R o ctA del(2)C dl=C o dl A del(3) whereR o ct=(r d(Q cm2)(4)R ct(Q cm2)and C dl o(m F cm−2)were characteristic values for the corrosion reaction at the coating/metal interface,which was assumed not to change with coating degradation;d was coating thickness;r was the coating resistivity.Using a value for C dl o=20m F cm−2[8]and experimen-tally measured C dl(F)as a function of exposure time,the delaminated area,as well as delamination ratio(D)were calculated for all investigated samples.The delamination ratio,D,was calculated as:D=A delA(5)where A was the total exposure area of the polymer coated metal.3.Results and discussionAs can be seen from Fig.2there was a difference between the samples after2h of exposure in3%acetic acid.On immersion the pore resistance of the specimens,having two coats of epoxy/phenolic lacquer(Fig.3)exceeded value of1.4×108Q cm2.These samples,S2x andW2x, Fig.1.Equivalent circuit for(a)dry organic coating and(b)defective organic coating.202J.D.Scantlebury,K.Galic/Progress in Organic Coatings31(1997)201–207showed lower C p values (Fig.4)compared to samples which had thinner coating layer.A specimen with a single coat solvent based lacquer,S1x (Fig.2),showed high frequency semicircle which can be attributed to paint film properties as it was associated with a capacitance of 1×10−9F cm −2(Fig.4)while R p value was 6×107Q cm 2(Fig.3).Low fre-quency semicircle showed C dl value of 1.5×10−8F cm −2(Fig.5)and is related to the degree of delamination and/or corrosion [7–10].After 24h of immersion (Fig.6)the semicircle diameter decreased for all specimens.It is widely believed that degradation and loss of adhesionis indicated by a rapid increasing C p and a rapid decrease in R p [11,12].According to many authors [13,14]the systems retain their corrosion protection while the coating resistance remains high (108–109Q cm 2),but fail when the resistance is low (below 107Q cm 2).Poor coatings are associated with measurements of 106Q cm 2.Mansfeld et al.[11]suggested that the observed decrease of pore resistance with exposure time is due to damage of the coating and the formation of conductive paths.The results obtained for all investigated samples,with a long immersion time (t imm Ͼ96h),showed a distinct separation between the semicircles (Fig.7,as all samples showed similar behavior,S2x sample is used as representa-tive),where the low-frequency semicircle decreased indi-cating corrosion of the substrate through the blistered paint film which was confirmed by SEM analysis (Figs.8and 9).Similar behaviour has been observed by many authors [15–17].James [18]provided evidence that blistering of the coat-ings when exposed to water is caused by an osmotic mechanism and that slowly evaporating,hydrophilic sol-vents retained in the coatings could be responsible.It was shown by Funke [19]that high-boiling hydrophilic,water-miscible solvents,such as glycol ethers,tend to produce this kind of blister formation.In our previous paper [20]it was shown that the amountFig.2.Nyquist impedance plots of an epoxy/phenolic–coated aluminum in 3%acetic acid after 2h of immersion.Fig.3.R p values as a function of exposure time.203J.D.Scantlebury,K.Galic /Progress in Organic Coatings 31(1997)201–207Fig.4.C p values as a function of exposuretime.Fig.5.C dI values as a function of exposuretime.Fig.6.Nyquist impedance plots of an epoxy/phenolic–coated aluminium in 3%aceic acid after 24h of immersion.204J.D.Scantlebury,K.Galic /Progress in Organic Coatings 31(1997)201–207of residual solvents was higher for the solvent based speci-mens (1.877mg m −2)compared with water based specimens (0.904mg m −2)with ethylene glycol–monoethyl ether con-centration being higher for the solvent based specimens.This could be one of the possible explanation for the larger number of blisters formed on the surface of the solvent based lacquer (Fig.8)compared with water based lacquer (Fig.9)having two coats of epoxy/phenolic lacquer.Kojima and Watanabe [21]found that water based and solvent based epoxy/phenolic coatings show different mechanism of adhesion to the substrate although both coat-ings possess excellent adhesion performance.Samples with single coat of epoxy/phenolic lacquer showed no blisters formed after 528h of exposure which could be due to better adhesion of the thinner coating to the substrate.As our earlier investigations showed [20]both sets of specimens showed certain porosity,expressed either in number of pores having different sizes (chemical method)or in current density values (electrochemical method).The consequence of external defects on the coatings was a rapid decrease of pore resistance (Fig.3)due to entry of acetic acid solution into the capillaries.The laws of permeation suggest that when a penetrant molecule passes through a polymer,three steps [22]are involved:the adsorption on and dissolution in the polymer,followed by diffusion through the polymer and finally des-orption at the other side.The first step depends on polarity,i.e.a polar penetrant molecule dissolves best in a polar polymer.Molecular movement,the second step,depends on size and shape of the penetrant and the concentration gradient.In this case water permeated through the polymer together with the carboxylic acid and interfered with the adhesive forces between the two layers,producing H 3O +.At the low pH (2.5)the protective aluminium oxide layer is dissolved under reaction with acetic acid [23]which induce adhesion losses for the coating and electrochemical under-film corrosion.According to Sotoudeh et al.[24]the corro-sion rate of aluminium alloy in acetic solution is low but appreciable.The result of water uptake by the coating [11,25,26]was the increase of C p (Fig.4).For epoxy coating,an increaseinFig.7.Nyquist impedance plots of alumnium with double-coat solvent based lacquer in 3%aceticacid.Fig.8.Scanning electron micrograph of an aluminium sample having double coat solvent based lacquer,after impedance measurements.Expo-sure time:528h in 3%aceticacid.Fig.9.Scanning electron micrograph of an aluminium sample having double coat water based lacquer,after impedance measurements.Exposure time:528h in 3%acetic acid.205J.D.Scantlebury,K.Galic /Progress in Organic Coatings 31(1997)201–207C p values,after immersion in an electrolyte,reaching a constant value after some hours,was found by many authors [27–29].At the end of exposure,C p was lowest for S1x followed by W2x while S2x and W1x showed identical and the highest values.A continuous increase of the double layer capacitance,C dl (Fig.5)was evidence that the area at which delamination occurred increased (Fig.10).This increase although occurred first for S1x sample (Fig.10),after 192h remained unchanged and the lowest,while at the same time R p value increased (Fig.3).This increase was probably due to alu-minium acetate formation,which is slightly soluble,[24]causing ‘blockage’of exposed substrate through the pores in the coating.In addition aluminum acetate forms hydrates and binds 0.5–2.5molecules of water per molecule [30].At this point,some type of adsorbed intermediate could be responsible for the inductive loops appearance [31],which can be seen in Fig.7.Therefore,the magnitude of change in C p as a function of time is a qualitative measure of water uptake.A large water uptake often is an indicator that a coating is porous and has a high probability for early deterioration.Further studies would be necessary to establish the mechanism of adhesion to the substrate and species of lac-quer composition which are likely to enhance corrosion.4.ConclusionsImpedance spectroscopy can be applied to characterise coatings on aluminium in an acidic solution.Pore resistance values decreased during immersion,indi-cating constant reduction of the protective properties.The change of coating capacitance and pore resistance with exposure time,to a corrosive medium,gives information on electrolyte uptake of different coating systems.Based on the highest R p and the lowest C p values as well as the lowest delamination ratio,single coat solvent based,S1x,sample provide the best corrosion protection properties of an epoxy/phenolic coated aluminium samples in an acetic acid.AcknowledgementsK.Galic ´would like to thank The Royal Society of Chem-istry for provision of a J.W.T.Jones Travelling Fellowship.The same author would also like to thank the Corrosion and Protection Centre,UMIST for provision of research facil-ities as well as to colleagues who provided invaluable advice and assistance with the analysis of the results.References[1]W.S.Tait,Poly.Mater.Sci.Eng.,58(1988)322.[2]A.Montanari,A.Pezzani,A.Cassara `,anese and G.Barbieri,Industria Conserve,67(1992)186.[3]A.Montanari,Industria Conserve,61(1986)129.[4]Z.Klenowicz and J.Rozwadowska-Lelinska,Proc.Eurocorr.’91,GTE,Budapest 1991,p.640.[5]Council Directive,85/572/EEC,Oj.No.L 372,31.12.85.pp.14–21.[6]S.H.Zhang and S.B.Lyon,Colloquium on Electrochemical Mea-surements,London,1994.[7]S.Haruyama,M.Asari,T.Tsuru,Proc.Symp.Corrosion Protectionby Organic Coatings,87–2.Pennington,NJ:Electrochemical Society,1987,p.197.[8]F.Mansfeld,C.H.Tsai and H.Shih,in Proc.Symp.Corrosion Pro-tection by Organic Coatings,89–13.Pennington,NJ:Electrochemi-cal Society,1989,pp 228–240.[9]B.J.Hepburn,K.R.Gowers and J.D.Scantlebury,Br.Corros.J.,21105(1986).[10]B.J.Hepburn,L.M.Callow and J.D.Scantlebury,J.Oil ColourChem.Assoc.,67(1984)193.[11]F.Mansfeld,M.W.Kendig and S.Tsai,Corrosion,38(1982)478.[12]W.S.Tait,J.Coat.Technol.,46(1989)768.[13]J.E.O.Mayne and ls,J.Oil Colour Chem.Assoc.,58(1975)155.[14]H.Leidheiser,Jr.,.Coat.,7(1979)79.[15]G.W.Walter,J.Electroanal.Chem.,118(1981)259.Fig.10.Delamination ratio as a function of exposure time.206J.D.Scantlebury,K.Galic /Progress in Organic Coatings 31(1997)201–207[16]N.Pebere,Th.Picaud,M.Duprat and F.Dabosi,Corros.Sci.,29(1989)1073.[17]J.F.McIntyre and L.Hang,Corros.Sci.,50(1994)27.[18]D.M.James,J.Oil Colour Chem.Assoc.,43(1960)653.[19]W.Funke,.Coat.,9(1981)29.[20]Z.Mesic´,K.Galic´and N.Cikovic´,in Proc.Symp.Corrosion Protec-tion by Organic Coatings,95–13,The Electrochemical Society,Pen-nington,NJ:1995,pp.40–50.[21]S.Kojima and Y.Watanabe,Polym.Eng.Sci.,33(1993)265.[22]R.M.Barrer,Diffusion in and Through Solid Materials,CambridgeUniversity Press,1941.[23]B.Alwitt,The aluminium-water system,in J.W.Diggle and A.K.Vijh(Eds.),Oxides and Oxide Films,Ch.3,Vol.IV.Marcel Dekker, New York,1976,p.169.[24]K.Sotoudeh,T.H.Nguyed,R.T.Foley and B.F.Brown,Corrosion,37(1981)358.[25]U.Rammelt and G.Reinhard,.Coat.,21(1992)205.[26]G.W.Walter,Corros.Sci.,32(1991)1041.[27]E.P.M.Van Westing,G.M.Ferrari and J.H.W.de Wit,Electrochim.Acta,39(1994)899.[28]E.P.M.Van Westing,G.M.Ferrari,E.M.Geenen,J.H.W.de Wit,Electrochim.Acta,23(1993)89.[29]F.Deflorian,V.B.Miskovic-Stankovic,P.L.Bonora and L.Fedrizzi,Corrosion,50(1994)438.[30]Deutschen Chemische Gesellschaft,Verlag Chemie,Gmelin Hand-buch der Inorganische Chemie,Berlin35(A),1935,pp.295–300.[31]W.J.Lorenz and F.Mansfeld,Corros.Sci.,21(1981)647.207J.D.Scantlebury,K.Galic/Progress in Organic Coatings31(1997)201–207。