Formation and characteristics of Al-Zn hydrotalcite coatings on galvanized steel.

In this study, a process for depositing hydrotalcite (HT) coatings on galvanized steel was developed and the resulting coatings were characterized. Results showed that coatings formed spontaneously on galvanized surfaces upon exposure to ambient temperature alkaline aluminate solutions. Anodic polarization and electrochemical impedance spectroscopy experiments showed clear evidence of surface passivation. Scanning electron microscopy showed the formation of a continuous and conformal surface film comprised on a compact mass of crystallites. X-ray diffraction confirmed that the coating contained an Al-Zn hydrotalcite compound. Coating formation was enhanced by oxidizer and ammonium salt additions. Coatings formed by using best practices were deposited in less than 10 minutes and demonstrated good surface coverage and good organic coating adhesion. HT coatings formed by using best practices showed excellent organic coating adhesion compared to zinc phosphate control coatings. In salt spray testing, the presence of a hydrotalcite conversion coating under an epoxy neat resin was found to delay the onset of red rusting compared to control samples that were epoxy coated, but not conversion coated.

Keywords: Electrochemical impedance, corrosion, corrosion protection, accelerated testing, adhesion, corrosion, automotive--OEM, pollution, conversion coating, phosphate, iron, cast iron, steel

**********

Phosphate conversion coating involves the formation of a surface layer of hydrated metal phosphate crystals whose coverage approaches 100%. (1,2) Coatings form spontaneously via precipitation during simple immersion in a properly formulated acidic bath. Iron, zinc, and manganese phosphate processes are regularly used, depending on the application and the substrate. Zinc phosphate coatings are used mainly on galvanized steel. (3) The main crystalline components of zinc phosphate layers are hopeite ([Zn.sub.3](P[O.sub.4])2*4[H.sub.2]O and phosphophyllite ([Zn.sub.2]Fe(P[O.sub.4])[.sub.2]*4[H.sub.2]O). An important application of phosphate coatings is the pretreatment of galvanized steel surfaces prior to electrophoretic painting. This process is used widely in the automotive industry.

Over the years, phosphate bath chemistries have evolved considerably, mainly driven by the desire to accelerate coating formation, and to enhance coating properties. For example, nickel additions accelerate coating formation and promote corrosion resistance. Accelerators that increase the rate of coating formation and suppress hydrogen evolution include nitrites, nitrates, chlorates, and peroxides. Surfaces can be activated by immersion in a colloidal Ti-bearing solution prior to immersion in the phosphating bath. Post-coating sealing in chromate or transition metal fluoanion solutions increases resistance to alkaline environments and promote corrosion resistance.

Environmental issues are forcing changes in phosphate conversion technology. There is a demand to reduce the use of heavy metals. [Ni.sup.2+]-based sealing solutions are being replaced with more environmentally friendly chemistries such as those based on [Cu.sup.2+]. Widely used chromate sealing processes are being eliminated due to the hazards of [Cr.sup.6+] to workers and the environment. Additionally, there is a desire to reduce the number of surface finishing processes used in manufacturing. As a result, there is a premium placed on surface finishing methods that work well on components comprised of different metals. In this regard, processes that form good coatings on aluminum alloys, steels (galvanized and ungalvanized), and magnesium alloys are highly desired.

A range of coating processes based on the formation of hydrotalcite (HT) surface layers has been developed for aluminum alloys. The foundation of these processes is the spontaneous precipitation of hydrotalcite crystals on an alloy surface during exposure to an alkaline solution containing coating forming species. (4-6) These processes yield a continuous and conformal coating comprised of a compact mass of hydrotalcite crystallites. The coating forms in a matter of minutes and does not require the use of externally applied voltages or currents. Modifications to the bath chemistry are made to suppress gas evolution, accelerate coating formation, or form coatings on Cu-rich alloys. (7) Process modifications and post-treatments have been developed to remove copper, (8) increase corrosion resistance, and impart low electrical contact resistance. (9) Active corrosion protection can be developed in HT coatings by sealing them with cerium. (10) To date, HT coatings have been developed exclusively for use with aluminum and its alloys.

The objectives of this study are to develop and demonstrate environmentally friendly hydrotalcite coating processes for galvanized steel substrates that are procedurally similar to phosphate immersion coatings, and to characterize the resulting coating structure, morphology, and properties including corrosion protection and paint adhesion.

EXPERIMENTAL PROCEDURES

Materials

Coating formation experiments were carried out on two different types of galvanized steel substrates. The first type of material was hot-dipped galvanized (HDG) cold-rolled steel (CRS) that conformed to the Society of Automotive Engineers' specification SAE1008 and the American Society for Testing and Materials' specification ASTM D 609. These materials were donated, and the manufacturer is unknown. From metallographic cross sections, the thickness of the galvanized layer was found to range from 22-25 [micro]m. The second type of material was coated with a Zn-45% Al coating supplied under the trade name Galvalume[TM] and manufactured by U.S. Steel. The thickness of this layer was not determined.

[FIGURE 1 OMITTED]

Commercially prepared phosphate-coated samples were used as control samples. Samples were obtained from ACT Laboratories, Inc., Hillsdale, MI. Samples consisted of hot-dipped galvanized CRS. Substrates were supplied with a Bonderite 958 zinc phosphate treatment from Henkel Surface Technologies, Inc., Southfield, MI. Coatings were supplied in an unsealed condition. From metallographic cross sections, it was found that the thickness of the galvanized layer on this material ranged from 15-19 [micro]m.

Reagents and Coating Processes

All in-lab cleaning and coating processes were carried out using reagent grade chemicals and deionized (DI) make-up water. Processing was carried out in four-liter glass beakers under mechanical agitation. Coatings were formed on either 50 X 25 mm, or 200 X 300 mm sized substrates. No special precautions were taken to mask the steel substrate exposed at cut edges.

[FIGURE 2 OMITTED]

Several different substrate precleaning procedures were used throughout this study. Best practice cleaning involved a single step dip in 3% NaOH at room temperature to remove bulk greases and oils. This procedure was used in all the experiments, unless otherwise noted. A large component of this study involved exploring variations in bath chemistry. Such variations explored included bath pH, coating time, [Al.sub.2][Na.sub.2][O.sub.4] concentration, pH, and supplemental ingredients. The range of these variations and their effect on coating formation is presented in the Results and Discussion section. During coating, samples were rinsed in overflowing DI water between each step. After coating, the samples were rinsed and allowed to air-dry for 24 hr before any further handling or processing.

Epoxy Coating

Epoxy neat resins were applied to HT-coated substrates to gain an indication of hydrotalcite layer compatibility with organic overlayers. An MIL-P-53022-compliant bisphenol A-epichlorohydrin epoxy resin EPON 1001-X-75, produced by Shell Epoxy Resins, LLC, was used in all the experiments. The curing agent for this resin was a polyamide supplied under the tradename EPI-CURE 3115-X-70 (Shell). Coatings were formulated using an indexing ratio of 2:1 by weight resin to curing agent. This mixture was then diluted with eight parts by volume methylethyl ketone to one part by volume of the resin hardener mixture. The coatings were applied by dipping and were allowed to dry and cure in air for at least one week under ambient lab conditions before any testing. Such coatings were used to test the paintability and adhesion on hydrotalcite-coated HDG substrates.

Electrochemical Testing

Anodic polarization was used to screen for effective bath chemistries. Anodic polarization was carried out using a Princeton Applied Research Model 273A potentiostat operated under computer control. A three-electrode configuration was used in which the counter electrode was a graphite rod and the reference electrode was a saturated calomel electrode (SCE). A range of different solution chemistries was explored. These are detailed in the Results and Discussion section.

EIS was used to assess prospective coating chemistries, corrosion resistance of as-coated substrates, and epoxy-coated samples. EIS experiments were conducted using a three-electrode cell configuration operated under potentiostatic control using a 10 mV sinusoidal voltage perturbation. The experiments were controlled using a PAR Model 273A potentiostat and a Solartron Model 1255 frequency response analyzer. Experiments were coordinated and data were logged by Scribner and Associates Zplot[TM] EIS software.

Salt Spray Exposure Testing

Coated samples were subject to neutral salt spray exposure conditions. A salt fog was generated by atomizing a 5 wt% NaCl solution and injecting it into an enclosed chamber. The fog deposition rate ranged from 1-2 mL per hour. The temperature of the chamber was held at 95 [+ or -] 3[degrees]F. (11) The pH of the solution in the chamber was not measured during the experiment. Coated samples were scribed through to the steel substrate prior to exposure. Once testing commenced, the samples were inspected visually on a periodic basis. The extent of red rusting after 1000 hr of salt spray exposure was evaluated according to standard methods. (12)

[FIGURE 3 OMITTED]

Scanning Electron Microscopy and X-Ray Diffraction

Scanning electron microscopy (SEM) was used to characterize the morphology of hydrotalcite coatings formed on substrate surfaces. SEM was carried out using a Philips XL-30 environmental SEM in normal mode. Prior to examination, sample surfaces were carbon-coated to eliminate charging effects and to enhance image resolution. A range of imaging conditions (accelerating voltage and beam current) was used depending on the needed resolution and sample conductivity. The chemistry of some samples was characterized by energy dispersive spectroscopy (EDS).

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Hydrotalcite and phosphate coatings are comprised of crystalline compounds that can be detected by X-ray diffraction (XRD). XRD patterns were collected in 0-20 mode using a Scintag PAD-V diffractometer. This system uses a Cu anode, 2 kW sealed tube X-ray source ([lambda] = 1.544[Angstrom]) with a fully interlocked cabinet for X-ray safety.

Paint Adhesion and Cathodic Disbonding

A cathodic disbonding analysis based on the Rapid Electrochemical Assessment of Paint (REAP) method was used to compare the relative adhesion of epoxy neat resins to hydrotalcite and zinc phosphate coated surfaces. (13) According to this methodology, coating barrier properties are assessed by EIS measurement immediately upon exposure and after 24 hr of exposure to aerated 0.5 M NaCl solution. Susceptibility to cathodic disbonding was assessed by scribing samples using a sharp edge of a knife to expose bare metal. Two 2-cm scribes at right angles were made on each sample. Scribed samples were then exposed to aerated 0.5 M NaCl solution and cathodically polarized at -1.05 [V.sub.sce] for 24 hr at ambient lab temperatures. Following cathodic polarization, samples were removed from the cell and adhesive tape was applied over the scribe then pulled off to remove any delaminated or loosely adherent epoxy. The samples were then dipped in a CuS[O.sub.4] solution for 10 min. Cu plated out on exposed regions and gave a clear indication of the extent of coating removal. Such samples clearly revealed differences in resistance the cathodic disbondment.

RESULTS AND DISCUSSION

Phosphate Coatings on HDG Steel

Commercially obtained Zn phosphate HDG steel samples were evaluated to develop benchmarks against which Al-Zn HT coatings could be compared. Figure 1 shows scanning electron micrographs of a Zn phosphate coating on HDG steel. The coating is comprised of a compact mass of fine crystallites that cover the galvanized surface uniformly. There are no visible gaps in coverage and the coating follows the contours of the underlying substrate suggesting that it is not overly thick. This surface is rough and somewhat porous. It does not provide high intrinsic corrosion resistance, but does provide excellent adhesion for subsequently applied paints and adhesives. (2)

Figure 2 shows the evolution of the EIS response over a 24-hr exposure to aerated 0.5 M NaCl solution. Overall, the level of corrosion protection is not high. The maximum total coating resistance at any time during the experiment is only on the order of 15,000 [ohm]*[cm.sup.2]. For comparison, the typical [R.sub.c] values measured for chromate conversion coating on the aluminum alloys is in the range of [10.sup.6] [ohm]*[cm.sup.2]. (14) At early times, the data indicate a two-time constant response consistent with charge transfer (at higher frequencies) and diffusion (at low frequencies). This is consistent with other observations of the EIS response from unsealed phosphate coatings. (15) The inductive response at the lowest frequencies is likely due to drift in the corrosion potential. At longer exposure times, the spectra suggest a purely diffusion-controlled response, probably associated with electrochemical activity within pores in the coating.

Passivation on Zn in Alkaline Aluminate Solutions

A methodology used previously to develop an HT coating process for aluminum alloys was used here to develop Al-Zn HT coatings for Zn. (4,5) In the present study, the intent was to form a hydrotalcite compound based on coprecipitation of Al and Zn. Oxides of both these metals are amphoteric. (16,17) Under alkaline conditions, Al will dissolve to give the aluminate ion, Al[O.sub.2.sup.-] (or, equivalently Al(OH)[.sub.4.sup.-]), and Zn will dissolve to form zincate, Zn[O.sup.2-] (or, equivalently Zn(OH)[.sub.4.sup.2-]) and bizincate, HZn[O.sub.2.sup.-]. By analogy to Li-Al HT formation, Al-Zn formation should occur during anodic polarization of Zn when exposed to an alkaline Al[O.sub.2.sup.-] solution with the proper pH and Al[O.sub.2.sup.-] concentration. Figure 3 shows the anodic polarization curve for HDG steel in a 0.1 M NaCl pH 11.0 with and without the addition of 0.03 M [Na.sub.2][Al.sub.2][O.sub.4]. When Al[O.sub.2.sup.-] is present in solution, the electrode exhibits an active passive transition commensurate with that observed for Al electrodes immersed in alkaline Li solutions. (4) This is taken as an indication of the tendency for passive film formation that can be exploited as the basis for a coating process. It should be noted that the current density at the nose of the active peak is in excess of 1 mA/[cm.sup.2], prior to passivation. This current density represents the intensity of Zn dissolution, which supplies Zn(OH)[.sub.4.sup.2-] anions necessary for hydrotalcite formation. Interestingly, these surfaces repassivate very readily on reverse scanning. Although the "passive" dissolution rates shown in Figure 3 are above 100 [micro]A/[cm.sup.2], they immediately fall as the direction of the potential scan direction is reversed. This behavior is not unexpected. The steel substrate should readily passivate under the alkaline condition present in this solution if the galvanized layer is perforated. The Zn layer should also passivate due to the formation of Al-Zn-HT in one case, and the formation of a zinc carbonate, ZnC[O.sub.3], corrosion product layer in the Al[O.sup.2-]-free solution. (17)

Figure 4 shows the evolution in the EIS response of HDG steel samples exposed to a 0.05 M [Na.sub.2][Al.sub.2][O.sub.4] solution under open circuit conditions. Over the span of 4.5 hr a steady increase in the charge transfer resistance is observed. This shows that coating formation is spontaneous and an applied voltage or current is not needed to induce passivation.

Figure 5 shows anodic polarization curves of Galvalume samples in solutions with and without Al[O.sub.2.sup.-]. In this case, clear evidence of an active passive transition is absent, despite the wide potential sweep range used. The electrochemical behavior of Galvalume electrodes is considerably different than that of HDG surfaces in that they appear to be spontaneously passive. These samples exhibit passive current densities ranging from about 20-50 [micro]A/[cm.sup.2], independent of the presence of Al[O.sub.2.sup.-] in solution. These samples also exhibit a well-articulated pitting potential that is shifted to slightly more positive potentials in the presence of Al[O.sub.2.sup.-]. Nonetheless, the high fraction of Al in the metallic layer appears to lower the electrochemical activity of the surface resulting in a lower Zn dissolution rate, which inhibits the formation of HT and its attendant passivation. These surfaces also demonstrate repassivation potentials that exceed their pitting potentials. This indicates that these coatings also readily repassivate. SEM characterization of exposed Galvalume surfaces did not indicate the formation of a hydrotalcite coating, and passivation phenomena were attributed to the stabilization of native oxides. While it is possible that HT coatings can be developed for Galvalume surfaces, they do not represent a useful substrate for early-stage coating development work. On the basis of these results, it was concluded that further coating formation experiments would be carried out on HDG samples only.

Defining the Bath Chemistry Foundation

Conversion coatings that form via precipitation of a crystalline compound are usually formed under oxidizing alkaline conditions where the bath is "seeded" with cations of the substrate on which the coating is precipitated (e.g., [Zn.sup.2+]). (2,18) The seeding process often has the effect of shortening the time required to passivate the surface, which decreases the extent of etching. Therefore, the primary variables in the bath chemistry foundation are the concentrations of the film forming agent Al[O.sub.2.sup.-], an oxidizer, potassium nitrate KN[O.sub.3], in this case, [Zn.sup.2+] added as Zn(N[O.sub.3])[.sub.2], solution pH, immersion time, and bath temperature.

[FIGURE 7 OMITTED]

Figure 6 shows the evolution in the EIS response for an Al-Zn HT-coated HDG steel sample exposed to an aerated 0.5 M NaCl solution to illustrate the nature of the EIS spectra obtained in coating evaluation experiments. At short times, a single time constant charge transfer process is detected. This gradually evolves into a two-time constant response as a diffusional component becomes apparent. This transition represents the penetration of electrolyte into the pores of the coating. The diffusional impedance increases with time, indicating pore plugging and a form of self-healing. This phenomenon has been observed with HT coatings on Al alloys. (7)

Table 1 shows the different permutations of coating ingredient concentrations, coating times and temperatures, and bath pH values that were used in exploratory trials. The response variables used to determine coating effectiveness was the total coating resistance, [R.sub.c], determined by an EIS measurement conducted after 24 hr of exposure to an aerated 0.1 M NaCl solution, and the appearance of the coated surface before and after exposure to the chloride solution. The [R.sub.c] value is essentially the total resistance of the coated surface demonstrated at the lowest measured frequencies. These values are relatively low, particularly in comparison to chromate conversion coatings. (14) However, phosphate coatings and the HT coatings being developed here are not necessarily intended to provide high intrinsic corrosion resistance themselves, but rather make the surface receptive to organic coatings and help suppress undercoat corrosion and blistering. (2) Although all the samples were pitted after exposure to solution, the [R.sub.c] value is thought to give at least an indirect indication of surface coverage, which is important to organic coating adhesion. [R.sub.c] values for the various HT coatings range from about 240 to 3675 [ohm]*[cm.sup.2]. The [R.sub.c] value for an uncoated surface was measured to be about 1700 [ohm]*[cm.sup.2].

[FIGURE 8 OMITTED]

[R.sub.c] values were consistently high for baths containing 5-10 mL/L 45% [Na.sub.2][Al.sub.2][O.sub.4], and 0.5 g/L Zn(N[O.sub.3])[.sub.2]*6[H.sub.2]O and a pH of 11 or 12. There is no strong dependence on the KN[O.sub.3] concentration over the range it was studied. Elevating the bath temperature was generally found to lower the [R.sub.c] values.

X-ray diffraction was performed on coated samples to verify HT compound formation. Figure 7 shows XRD patterns collected on samples exposed for different lengths of time. The pattern contains reflections from the Zn alloy substrate and the Al-Zn HT compound. Arrows in the plot indicate the reflections due to the hydrotalcite compound. (19) The formation of both AlZnC[O.sub.3]-HT and AlZnN[O.sub.3]-HT compounds are possible in the coating baths used here. Carbonate anions are present in the bath due to the uptake of atmospheric C[O.sub.2] into the alkaline solution. Dissolved C[O.sub.2] then speciates to the C[O.sub.3.sup.2-] ion. Due to the similarities in XRD patterns, AlZnC[O.sub.3]-HT cannot be easily distinguished from AlZnN[O.sub.3]-HT. Therefore, the proportions of the two HT phases present in the coating is not known. In any case, as immersion time increases, the intensity of the hydrotalcite peaks increase, suggesting that the film continues to thicken over 15 hr of immersion. Figure 8 shows SEM micrographs of Al-Zn HT-coated surfaces illustrating the distinctive morphology of polycrystalline hydrotalcite coatings. (4)

[FIGURE 9 OMITTED]

The results shown in Figure 7 and Table 1 indicate that long times are required to form coatings with detectable HT and high [R.sub.c] values. These times are too long for practical conversion coating treatments. Chemistry variations that successfully reduced the coating processing time are described in the following.

Effect of pH and Aluminate Concentration

[Na.sub.2][Al.sub.2][O.sub.4] is a strong base, and small additions to solution will raise the pH significantly. A set of coating trials were carried out in which [Na.sub.2][Al.sub.2][O.sub.4] concentrations and pH were systematically varied to characterize their effect on Al-Zn HT coating formation and properties. In these trials, prior to immersion in the coating bath, HDG steel coupons were cleaned by immersion in 3% NaOH for one minute at room temperature then rinsed in DI water. Five different coating baths were used. [Na.sub.2][Al.sub.2][O.sub.4] was added to 0.01 M Zn(N[O.sub.3])[.sup.2] solution in concentrations of 0.005, 0.010, 0.025, 0.050, and 0.085 M resulting in solution pH values of 11.5, 12.1, 12.5, 12.7, and 13.0, respectively. Coupons were immersed for either 30 min or 15 hr. Samples were analyzed for coating surface coverage by SEM and for corrosion resistance by EIS. Figure 9 shows the variation in [R.sub.c] as a function of solution pH. [R.sub.c] is small and invariant for coatings formed by a 30-min immersion over the pH/[Na.sub.2][Al.sub.2][O.sub.4] range studied. For coatings formed by a 15-hr immersion, [R.sub.c] is relatively large for the smallest [Na.sub.2][Al.sub.2][O.sub.4] concentration used, but is invariant otherwise. SEM showed that HT coating surface coverage and substrate etching increased with increasing pH and [Na.sub.2][Al.sub.2][O.sub.4] concentration. Despite the high [R.sub.c] value measured, no HT crystallite formation was detected in the 15-hr coating formed at the lowest [Na.sub.2][Al.sub.2][O.sub.4] concentration. Based on surface coverage, coating uniformity, and etching response, baths with a nominal pH of 12.2 to 12.5 and a [Na.sub.2][Al.sub.2][O.sub.4] concentration of 0.25 M were selected for best practice coating experiments.

[FIGURE 10 OMITTED]

Effect of Supplemental Ingredients

Thirteen salts that are either hydrotalcite formers (LiN[O.sub.3], [Na.sub.2]C[O.sub.3]), oxidizers (Zn(N[O.sub.3])[.sub.2], KN[O.sub.3], N[H.sub.4]Cl, N[H.sub.4]N[O.sub.3], [Na.sub.2][S.sub.2][O.sub.8], [Na.sub.2]S[O.sub.4]), phosphate coating bath supplements ([Na.sub.2][B.sub.4][O.sub.7], ZnS[O.sub.4], NiS[O.sub.4]), or inhibitors (NaV[O.sub.3], [Li.sub.2]Mo[O.sub.4]) were tested by the addition to the coating bath in small amounts to determine if they led to enhanced coating formation or properties. The compounds were added at 0.01 M concentrations separately into 0.05 M [Al.sub.2][Na.sub.2][O.sub.4] coating baths whose pH was about 11.5 to 12.0 and temperature was about 23[degrees]C. Prior to coating, the HDG steel panels were cleaned in Alconox[TM] detergent. After coating, the samples were dried and analyzed by SEM to check for coating formation and coverage, EDS for chemical analysis, and EIS for corrosion protectiveness and surface coverage.

[FIGURE 11 OMITTED]

[FIGURE 12 OMITTED]

Table 2 is a summary of the results from these examinations. The different bath chemistries are distinguished by the supplemental reagent added. "No addition" refers to a bath that contained 0.05 M [Na.sub.2][Al.sub.2][O.sub.4] only and "Bare HDG" refers to an uncoated sample. Immersion times other than 15 hr were used for the NiS[O.sub.4] and NaV[O.sub.3] baths. These times are noted in the table.

[FIGURE 13 OMITTED]

Measured [R.sub.c] values ranged considerably suggesting a significant effect due to the various supplemental ingredients examined. The highest values recorded were typically associated with the oxidizing compounds. High [R.sub.c] values were only observed for certain inhibitors, hydrotalcite formers, or phosphate supplements depending on whether or not the chemical possessed some intrinsic oxidizing character.

The EIS spectra exhibited common features for all 14 baths. Representative examples of these spectra are shown in Figure 10. At high frequencies, the EIS response was consistent with a charge transfer resistance operating in parallel to an interfacial capacitance. This led to the occurrence of a depressed arc in the complex plane. At lower frequencies, a diffusional impedance was evident. Two different forms of diffusion were observed. (20) The first type is characterized in equivalent circuit modeling as a transmission line with a transmissive boundary. This is an R-C transmission line terminated by a resistance. When this type of response is detected, the applied voltage perturbs the base of a pore to stimulate a charge transfer process, i.e., corrosion. On the complex plane, this type of a diffusional process results in a misshapen arc that this flattened on its high frequency side (left side) compared to its low frequency side. At the lowest measured frequency, the arc reaches or almost reaches the real axis. Figure 10b is an example of this type of behavior. In Table 2, samples responding in this fashion are termed "transmissive" with respect to their diffusional character. The second type of behavior is also characterized in equivalent circuit modeling with a transmission line. However, in this case, the transmission line is terminated by a capacitance. This situation arises when pore bottoms are passivated or when the applied voltage perturbation does not affect the base of an unprotected pore. This latter situation can arise when pores are very deep, or when they have a very large aspect ratio. On the complex plane, this type of response is characterized by data that trends away from the origin and does not appear to arc back to the real axis. Figure 10c is an example of this type of behavior. In several cases, the diffusional character could not be determined by inspection. Such is the case in Figure 10a. This is probably due to nonstationary behavior of the electrode and such situations are referred to as "indeterminate." The interpretation taken here is that reflective boundary transmission line behavior is an indication of thicker surface films whose pore bases could not be perturbed by the voltage excitation. Reflective behavior was demonstrated predominantly by coatings formed from baths modified with supplemental oxidizers.

EDS data also sheds light on the relative thicknesses of the Al-Zn HT layers formed in these experiments. Table 2 shows the ratio of Al to Zn on an atomic percentage basis. In EDS, as it was performed here, characteristic X-rays were generated from the coating and from the substrate. The coating contained Al and Zn and the substrate contained predominantly Zn (the underlying steel substrate was not detectable). The thicker the coating the less the underlying metallic Zn layer was excited and the greater the observed Al:Zn ratio. The largest Al:Zn ratios are observed for coatings formed in the presence of oxidizing supplemental ingredients. Notably, low ratios are detected for inhibitor additions, suggesting that these ingredients may actually be inhibiting coating formation. Inspection of Table 2 shows that reflective boundary diffusional character is closely associated with high Al:Zn ratios. In fact, the Al:Zn ratios range from 1.20-1.77 for reflective behavior compared to a range of 0.10-1.17 for transmissive behavior. Overall, this comparison suggests that small additions of oxidizer strongly enhance Al-Zn HT film formation.

[FIGURE 14 OMITTED]

Examination of coated surfaces by scanning electron microscopy indicates that HT-like crystals formed on HDG surfaces from all of the baths after three hours of immersion, with an exception of the bath containing NaV[O.sub.3]. Surface coverage was generally quite good with some nonuniformity in the [Na.sub.2][Al.sub.2][O.sub.4]-only bath, and some shrinkage cracking in the coating formed from the KN[O.sub.3]-bearing bath. A variety of crystallite morphologies were observed, again indicating that supplemental ingredients can profoundly affect the formation of Al-Zn HT films. Selected micrographs illustrating these variations are shown in Figure 11.

[Zn.sup.2+] Additions

The practice of "seeding" a coating bath with metal cations from the substrate being coated is a common practice used with precipitated coatings. (2,18) Al-Zn HT film formation depends on the production of [Zn.sup.2+] cations by substrate dissolution. These cations subsequently hydrolyze to form Zn(OH)[.sub.4.sup.2-] in the high pH bath. Zn additions must be made with care because they may inadvertently induce substrate passivation that suppresses HT coating formation by inducing the precipitation of nonprotective zinc hydroxides or by increasing the equilibrium reversible potential for Zn oxidation such that Zn dissolution is inhibited. Experimentally, additions of [Zn.sup.2+] (through additions of (Zn(NO)[.sub.3]) were found to promote uniform film formation, particularly when they were used in conjunction with ammonium salt additions.

N[H.sub.3]/N[H.sub.4.sup.+] Additions

Ammonium hydroxide N[H.sub.4](OH) (equivalent to ammonia N[H.sub.3]*[H.sub.2]O) and N[H.sub.4]N[O.sub.3] additions were found to promote the formation of thick Al-Zn HT layers of dense crystallites with excellent surface coverage. Ammonia/ammonium additions also promoted rapid coating formation. Ammonia is a good complexing agent for [Zn.sup.2+], and a thermodynamic analysis shows how additions of N[H.sub.3] might facilitate the Zn dissolution rates necessary for HT formation, as suggested by Figure 3.

To understand the effect of [Zn.sup.2+], N[H.sub.3] and pH on the thermodynamics of the Zn dissolution, it is necessary to consider the reaction equilibria governing the dissolution of Zn under alkaline conditions. Pourbaix states the equilibrium between metallic Zn dissolving to HZn[O.sub.2.sup.-] as (17):

HZn[O.sub.2.sup.-] + 3[H.sup.+] + 2[e.sup.-] [left and right arrow] Zn + [H.sub.2]O [E.sub.o] = +0.054 [V.sub.she] (1)

The equilibrium potential for this reaction depends on pH and on the HZn[O.sub.2.sup.-] concentration:

E = [E.sup.o] - 0.0886pH + 0.0295log[HZn[O.sub.2.sup.-]] (2)

At pH 12.0 and an [HZn[O.sub.2.sup.-]] equal to [10.sup.-2] M (concentration of Zn added to the bath), a reversible potential of -1.07 [V.sub.she] (-1.31 [V.sub.sce]) is predicted.

Tromans writes the dissolution equilibrium reaction between zinc and zincate under alkaline conditions in terms of the hydroxylated zincate ion (21):

Zn(OH)[.sub.4.sup.2-] + 4[H.sup.+] + 2[e.sup.-] [left and right arrow] Zn + 4[H.sub.2]O [E.sup.o] = +0.369 [V.sub.she] (3)

[FIGURE 15 OMITTED]

Compared to equation (1), the equilibrium potential of this reaction has a different dependence on pH, but an identical dependence on ion concentration, which is given by:

E = [E.sup.o] - 0.118pH + 0.0295log[Zn(OH)[.sub.4.sup.2-]] (4)

At a pH of 12.0 and [Zn(OH)[.sub.4.sup.2-]] = [10.sup.-2] M, the reversible potential for this reaction is -1.11 [V.sub.she] (-1.35 [V.sub.sce]).

When ammonia is added to solution, it is necessary to consider a third equilibrium reaction, namely the equilibrium of Zn with Zn(N[H.sub.3])[.sub.4.sup.2+]:

Zn(N[H.sub.3])[.sub.4.sup.2+] + 2[e.sup.-] [left and right arrow] Zn + 4N[H.sub.3] [E.sup.o] = -1.04 [V.sub.she] (5)

This equilibrium is not explicitly dependent on pH, but does depend on [N[H.sub.3]] and [Zn(N[H.sub.3])[.sub.4.sup.2+]]:

E = [E.sup.o] - 0.118log[N[H.sub.3]] + 0.0295log[Zn(N[H.sub.3])[.sub.4.sup.2+]] (6)

Implicitly, the equilibrium potential for this reaction does depend on pH via the equilibrium between ammonia and the ammonium ion. (22)

At pH 12.0 and at [Zn(OH)[.sub.4.sup.2-]] = [10.sup.-2] M and [N[H.sub.3]] = [10.sup.-2] M, the [Zn(N[H.sub.3])[.sub.4.sup.2+]] will be approximately [10.sup.-10] M. The reversible potential for this reaction then becomes -1.10 [V.sub.she] (-1.34 [V.sub.sce]).

In alkaline, ammoniacal solutions, chemical equilibria between zincate, Zn(OH)[.sub.4.sup.2-], and the zinc tetraamine cation, Zn(N[H.sub.3])[.sub.4.sup.2+], is possible (21):

Zn(N[H.sub.3])[.sub.4.sup.2+] + 4[H.sub.2]O [left and right arrow] Zn(OH)[.sub.4.sup.2-] + 4N[H.sub.3] + 4[H.sup.+] [K.sub.eq] = 2.2 X [10.sup.-48] (7)

For the bath chemistries used in this study, the equilibrium reversible potential for equation (5) is not substantially different than the reversible potentials for the reactions in equations (1) or (3), and it is not immediately obvious why complexation of [Zn.sup.2+] might facilitate Zn dissolution. However, as the soluble Zn concentration increases, in either the solution near the HDG steel surface or in the bulk bath solution with prolonged usage, the reversible potentials for the reactions in equations (1) and (3) will increase, thereby increasing the tendency for substrate passivation without hydrotalcite formation. However, as long as the N[H.sub.3] remains high, the equilibria in equations (5) and (7) are shifted to the left preserving a low reversible potential reaction pathway for Zn oxidation that does not depend on Zn(OH)[.sub.4.sup.2-] concentration. Additionally, a higher N[H.sub.3] concentration promotes the formation of the Zn(N[H.sub.3])[.sub.4.sup.2+] complex at the expense of Zn(OH)[.sub.4.sup.2-] [equation (7)]. It might be expected that Zn(N[H.sub.3])[.sub.4.sup.2+] formation might compete with HT coating deposition. However, experimental evidence suggests that this effect is not significant if it is occurring. Figure 12 shows scanning electron micrographs of Al-Zn HT-coated HDG steel surfaces after 10 min of immersion in an 0.01 M Zn(N[O.sub.3])[.sub.2], 0.25 M [Al.sub.2][Na.sub.2][O.sub.4], pH 12.2-12.5 at 23[degrees]C. Micrographs a and c show surfaces coated in a bath from which N[H.sub.4]N[O.sub.3] was withheld. Micrographs b and d show surfaces coated in a bath to which 0.01 M N[H.sub.4]N[O.sub.3] was added. The addition of N[H.sub.4]N[O.sub.3] provides a source of N[H.sub.3], which facilitates Zn dissolution that promotes uniform, rapid coating formation. When ammonium ion is added to the bath, a coating that is continuous and conformal is formed. Without the ammonium addition, coating formation is very sparse and indistinct. It is concluded that minor additions of [Zn.sup.2+] and N[H.sub.3] (via N[H.sub.4]N[O.sub.3]) to the coating bath are essential for activating galvanized surfaces to Al-Zn HT coating formation and for shortening the coating formation process from hours to minutes.

[FIGURE 16 OMITTED]

Impedance spectra show the increase in corrosion resistance due to adding N[H.sub.4]N[O.sub.3] to coating baths. Figure 13 compares the EIS response of Al-Zn HT coatings formed with and without N[H.sub.4]N[O.sub.3] additions after 24 hr of exposure to aerated 0.5 M NaCl solution. When N[H.sub.4]N[O.sub.3] is added [R.sub.c] increases and the diffusional response is no longer evident, suggesting the level of porosity in the coating has been altered so that diffusion no longer contributes to the measured spectrum.

Paint Adhesion-REAP

Figure 14 shows the extent of epoxy coating delamination after cathodic disbonding, as specified in the REAP method. (13) In these images, dark areas (highlighted by electroless deposited Cu) are areas where the epoxy coating was removed. Epoxy coatings were aged for 10 or 20 days in ambient temperature air prior to the cathodic disbonding treatment. For coatings aged for 10 days, significant coating removal occurred on all the samples. There is no particular association of the delamination with the scribe marks. Samples aged for 20 days in air gave a different delamination response. For the HDG surface that was immersed in 3% NaOH only and the Zn phosphate coated surface, there was total epoxy delamination within the region exposed to solution during the cathodic disbonding treatment. No coating removal was detected from the Al-Zn HT-coated sample, even in the vicinity of the scribe marks.

Corrosion Resistance of Epoxy-Coated Samples By Salt Spray

Figures 15 and 16 show optical photographs from replicate epoxy-coated and scribed samples exposed to salt spray for 1000 hr. Figure 15 shows scribed HDG steel samples that were Al-Zn HT conversion-coated per best practice--the dip-coated with an epoxy neat resin. All samples showed white rusting from corrosion of the galvanized layer and localized red rusting where the galvanized layer was perforated. There is some tendency for preferential corrosion in corrosion product tails. In a minority of cases, there is preferential corrosion in the scribes. Figure 16 shows a set of replicate panels that was degreased in 3% NaOH and epoxy-coated with application of an intermediate conversion coating. These samples showed generalized white rusting and localized red rusting. The extent of red rusting is greater in this sample set, as shown in Figure 15. The comparison suggests that the presence of Al-Zn HT coating slows the progression of corrosion damage under organically coated galvanized surfaces. A comparison of red rust formation on the HT-coated samples to Zn phosphate coated samples was not possible because of the difference in the galvanized layer thicknesses between the two sets of samples.

CONCLUSIONS

* Hydrotalcite formation on galvanized steel can be made to occur via immersion in alkaline aluminate solutions.

* A range of process chemistries and procedures was explored. The following relevant findings were made:

-- Immersion coating at ambient lab temperatures (23[degrees]C) in a solution comprised of 0.01 M Zn(N[O.sub.3])[.sub.2], 0.01M N[H.sub.4]N[O.sub.3], 0.25 M [Al.sub.2][Na.sub.2][O.sub.4], at a pH of 12.2-12.5 led to the best overall HT coating properties.

-- The addition of an ammonium/ammonia source enhanced coating formation. This is attributed to the formation of a soluble zinc tetramine cation Zn(N[H.sub.3])[.sub.4.sup.2+] and action of dissolved N[H.sub.3].

-- The effect of bath pH was explored over the range of 11.5-12.5. On the basis of EIS and SEM tests, the best coatings formed in solutions with pH values ranging from 12.2-12.5.

-- Coating time and bath chemistry were found to be closely coupled, but uniform coatings with good properties were formed in 10 min using the current best practices.

-- The effect of coating temperature was explored over the range of 23[degrees]-95[degrees]C. It was found that there was no advantage in either coating processing or properties associated with elevating bath temperature. In fact, coatings formed at 23[degrees]C showed the best corrosion resistance.

* Corrosion resistance has been measured by EIS. Overall, the corrosion resistances of Zn phosphate and Al-Zn HT conversion coatings is low. The corrosion resistance of Al-Zn HT coatings formed by best practices is increased over unpretreated samples, but is not as high as that measured for Zn phosphate control coatings.

* Adhesion of dip-coated epoxy neat resins has been studied. Cathodic disbonding and salt spray exposure results suggest that Al-Zn HT layers promote the adhesion of subsequently applied organic coatings.

Table 1 -- Basic Bath Chemistry Variations Explored

Zn(N[O.sub.3])[.sub.2]*  KN[O.sub.3]  45% [Na.sub.2][Al.sub.2][O.sub.4]
6[H.sub.2]O (g/L)           (g/L)                  (mL/L)

0.0                         0                        0
0.0                        30.3                      0.2
0.0                        30.3                      0.5
0.0                        30.3                      5.0
0.0                        30.3                     25.0
0.0                        30.3                     70.0
0.5                        10.1                      5.0
0.5                        30.3                      5.0
0.5                        50.5                      5.0
0.5                       101.1                      5.0
0.5                         0.0                      5.0
0.5                        10.1                      5.0
0.5                        30.3                      5.0
0.5                        50.5                      5.0
0.5                       101.1                      5.0
0.5                         0.0                      5.0
0.0                        10.1                     10.0
0.0                        10.1                     10.0
0.0                        10.1                     10.0
0.0                        30.3                     10.0
0.0                        30.3                     10.0
0.0                        30.3                     10.0
0.0                        50.5                     10.0
0.0                        50.5                     10.0
0.0                        50.5                     10.0
0.0                       101.1                     10.0
0.0                       101.1                     10.0
0.0                       101.1                     10.0
0.0                         0.0                     10.0
0.0                         0.0                     10.0
0.0                         0.0                     10.0

Zn(N[O.sub.3])[.sub.2]*  KN[O.sub.3]    Temp & Time
6[H.sub.2]O (g/L)           (g/L)     ([degrees]C)/hr)          pH

0.0                         0            Bare HDGa
0.0                        30.3            23/15                9.0
0.0                        30.3            23/15               10.0
0.0                        30.3            23/15               11.0
0.0                        30.3            23/15               12.0
0.0                        30.3            23/15               12.6
0.5                        10.1            23/15               11.0
0.5                        30.3            23/15               11.0
0.5                        50.5            23/15               11.0
0.5                       101.1            23/15               11.0
0.5                         0.0            23/15               11.0
0.5                        10.1            23/15               12.0
0.5                        30.3            23/15               12.0

0.5                        50.5            23/15               12.0
0.5                       101.1            23/15               12.0
0.5                         0.0            23/15               12.0
0.0                        10.1            70/0.5              11.5
0.0                        10.1           100/0.16             11.5
0.0                        10.1            23/15               11.5
0.0                        30.3            70/0.5              11.5
0.0                        30.3           100/0.16             11.5
0.0                        30.3            23/15               11.5
0.0                        50.5            70/0.5              11.5
0.0                        50.5           100/0.16             11.5
0.0                        50.5            23/15               11.5
0.0                       101.1            70/0.5              11.5
0.0                       101.1           100/0.16             11.5
0.0                       101.1            23/15               11.5
0.0                         0.0            70/0.5              11.5
0.0                         0.0           100/0.16             11.5
0.0                         0.0            23/15               11.5

Zn(N[O.sub.3])[.sub.2]*  KN[O.sub.3]  [R.sub.c] ([ohm]*[cm.sup.2])
6[H.sub.2]O (g/L)           (g/L)          0.1 M NaCl at 24 hr

0.0                         0                      1704
0.0                        30.3                    1287
0.0                        30.3                    1257
0.0                        30.3                    1091
0.0                        30.3                    2979
0.0                        30.3                     241.6
0.5                        10.1                    1531
0.5                        30.3                    3174
0.5                        50.5                    1659
0.5                       101.1                    2634
0.5                         0.0                    1176
0.5                        10.1                    3674
0.5                        30.3                    1573
0.5                        50.5                    2681
0.5                       101.1                    1949
0.5                         0.0                    1729
0.0                        10.1                     825.6
0.0                        10.1                     985.3
0.0                        10.1                    2372
0.0                        30.3                     883.8
0.0                        30.3                     725.2
0.0                        30.3                    2625
0.0                        50.5                     744.3
0.0                        50.5                     884.2
0.0                        50.5                    2088
0.0                       101.1                     629.8
0.0                       101.1                     821.8
0.0                       101.1                    1868
0.0                         0.0                    1520
0.0                         0.0                     805.1
0.0                         0.0                    1031

Zn(N[O.sub.3])[.sub.2]*  KN[O.sub.3]             Sample Surface
6[H.sub.2]O (g/L)           (g/L)                After Exposure

0.0                         0               Pits
0.0                        30.3             Dissolved, pits
0.0                        30.3             Dissolved, pits
0.0                        30.3             Light gray coating, pits
0.0                        30.3             Light gray coating, pits
0.0                        30.3             White precipitate, pits
0.5                        10.1             Light gray coating, pits
0.5                        30.3             Light gray coating, pits
0.5                        50.5             Light gray coating, pits
0.5                       101.1             Light gray coating, pits
0.5                         0.0             Light gray coating, pits
0.5                        10.1             Gray coating, pits
0.5                        30.3             Gray coating, pits
0.5                        50.5             Gray coating, pits
0.5                       101.1             Gray coating, pits
0.5                         0.0             Gray coating, pits
0.0                        10.1             Light gray coating, pits
0.0                        10.1             Light gray coating, pits
0.0                        10.1             Dark gray coating pits
0.0                        30.3             Light gray coating, pits
0.0                        30.3             Light gray coating, pits
0.0                        30.3             Dark gray coating pits
0.0                        50.5             Light gray coating, pits
0.0                        50.5             Light gray coating, pits
0.0                        50.5             Dark gray coating pits
0.0                       101.1             Light gray coating, pits
0.0                       101.1             Dark gray coating pits
0.0                       101.1             Dark gray coating pits
0.0                         0.0             Dark gray coating pits
0.0                         0.0             Dark gray coating pits
0.0                         0.0             Dark gray coating pits

(a) All the HDG panels from U.S. Steel, Inc. were degreased using
detergent and then deoxidized using Sanchem 1000[TM] at 23[degrees]C.

Table 2 -- Bath Supplemental Ingredients Explored

Bath Chemistry                                   [R.sub.c]
(a) [Na.sub.2][Al.sub.2][O.sub.4] 0.05 M, pH   ohm.[cm.sup.2]
11.5, RT, 15 hr                                0.5MNaCl 30min

No addition                                         393
NiS[O.sub.4]                  0.01 M  0.5 hr        225
NaV[O.sub.3]                  0.01 M  3 hr          807
[Na.sub.2]C[O.sub.3]          0.01 M                827
ZnS[O.sub.4]                  0.01 M                257
[Na.sub.2]S[O.sub.4]          0.01 M               1110
[Na.sub.2][S.sub.2][O.sub.8]  0.01 M                989
[Li.sub.2]Mo[O.sub.4]         0.01 M                989
[Na.sub.2][B.sub.4][O.sub.7]  0.01 M                340
N[H.sub.4]OH                  0.01 M                427
LiN[O.sub.3]                  0.01 M                466
N[H.sub.4]N[O.sub.3]          0.01 M                846
Zn(N[O.sub.3])[.sub.2]        0.01 M               3470
KN[O.sub.3]                   0.01 M                468
Bare HDG                                            356

Bath Chemistry                                Al:Zn Ratio  Diffusional
(a) [Na.sub.2][Al.sub.2][O.sub.4] 0.05 M, pH    by EDS     Character in
11.5, RT, 15 hr                                 (at.%)         EIS

No addition                                      0.56      Indeterminate
NiS[O.sub.4]                  0.01 M  0.5 hr     0.32      Transmissive
NaV[O.sub.3]                  0.01 M  3 hr       0.10      Transmissive
[Na.sub.2]C[O.sub.3]          0.01 M             0.62      Transmissive
ZnS[O.sub.4]                  0.01 M             0.80      Transmissive
[Na.sub.2]S[O.sub.4]          0.01 M             1.20      Reflective
[Na.sub.2][S.sub.2][O.sub.8]  0.01 M             1.37      Reflective
[Li.sub.2]Mo[O.sub.4]         0.01 M             1.17      Transmissive
[Na.sub.2][B.sub.4][O.sub.7]  0.01 M             1.16      Transmissive
N[H.sub.4]OH                  0.01 M             1.89      Reflective
LiN[O.sub.3]                  0.01 M             1.68      Indeterminate
N[H.sub.4]N[O.sub.3]          0.01 M             1.77      Reflective
Zn(N[O.sub.3])[.sub.2]        0.01 M             1.68      Reflective
KN[O.sub.3]                   0.01 M             0.73      Transmissive
Bare HDG

(a) Basic bath chemistry common to all baths.
Sample pretreatment: Cleaned by Alconox[TM] and rinsed using DI
[H.sub.2]O.

ACKNOWLEDGMENTS

This work was supported by Concurrent Technologies Corporation, Largo, FL under contract no. 030100247, Mr. L. Gintert, program manager.

References

(1) Donofrio, J., Metal Finishing, 98, 57 (2000).

(2) Ogle, K. and Buchheit, R.G., "Conversion Coatings," p. 460 in Encyclopedia of Electrochemistry, Wiley-VCH, Weinheim, Germany, 2003.

(3) Amirudin, A. and Thierry, D., Prog. Org. Coat., 28, 59 (1996).

(4) Buchheit, R.G., Bode, M.D., and Stoner, G.E., Corrosion, 50, 205 (1994).

(5) Drewien, C.A., Eatough, M.O., Tallant, D.R., Hills, C.R., and Buchheit, R.G., J. Materials Res., 11, 1507 (1996).

(6) Leggat, R.B., Zhang, W., Buchheit, R.G., and Taylor, S.R., Corrosion, 58, 322-328 (2002).

(7) Zhang, W. and Buchheit, R.G., Corrosion, 58, 591 (2002).

(8) Buchheit, R.G., J. Appl. Electrochem., 28, 503 (1998).

(9) Buchheit, R.G., Martinez, M.A., Montes, L.P., Cella, N., Stoner, G.E., and Taylor, S.R., "Inorganic Cr-Free Conversion Coatings for High Corrosion Resistance and Low Electrical Contact Resistance," Paper No. 98212 in Corrosion/98, NACE, Houston, TX, 1998.

(10) Buchheit, R.G., Mamidipally, S.B., Schmutz, P., and Guan, H., Corrosion, 58, 3-14 (2002).

(11) "Standard Practice for Salt Spray (Fog) Apparatus," 1, ASTM B117, ASTM International, West Conshohocken, PA, 1999.

(12) "Standard Test Method for Evaluating Degree of Red Rusting on Painted Steel Surfaces," p. 14, D 610-01, ASTM International, West Conshohocken, PA, 2001.

(13) Kendig, M., Jeanjaquet, S., Brown, R., and Thomas, F., "Rapid Electrochemical Assessment of Paint," JOURNAL OF COATINGS TECHNOLOGY, 68, No. 863, 39 (1996).

(14) Buchheit, R.G., Kendig, M.W., Martinez, M.A., Cunningham, M., and Jensen, H., Corrosion, 54, 61 (1998).

(15) Weng, D., Jokiel, P., Uebleis, A., and Boehni, H., Surf. Coat. Tech., 88, 147 (1996).

(16) Pourbaix, M., Atlas of Electrochemical Equilibria in Aqueous Solutions, p. 168, NACE-Cebelcor, Houston, TX, 1974.

(17) Pourbaix, M., Atlas of Electrochemical Equilibria in Aqueous Solutions, p. 406, NACE-Cebelcor, Houston, TX, 1974.

(18) Wernick, S., Pinner, R., and Sheasby, P.G., The Surface Treatment and Finishing of Aluminium and Its Alloys, 5th ed., p. 158 ASM International, Metals Park, OH, 1987.

(19) Velu, S., Ramkumar, V., Narayan, A., and Swamy, C.S., J. Mat. Sci., 32, 957 (1997).

(20) Raistrick, I.D., "Theory," p. 56 in Impedance Spectroscopy: Emphasizing Solid Materials and Systems, John Wiley & Sons, New York, NY, 1987.

(21) Tromans, D., Corros. Sci., 39, 1307 (1997).

(22) Pourbaix, M., Atlas of Electrochemical Equilibria in Aqueous Solutions, p. 493, NACE-Cebelcor, Houston, TX, 1974.

R.G. Buchheit and H. Guan--Ohio State University*

*Dept. of Materials Science and Engineering, Fontana Corrosion Center, 477 Watts Hall, 2042 College Rd., Columbus, OH 43210.