Characterization of a lacquer film formulated with phosphating reagents for corrosion protection of galvanized substrates.

This paper analyzes the chemical composition of the interface resulting from the application of an organic coating (lacquer) containing phosphating reagents on galvanized steel, galvanneal, and galfan substrates and its stability after short periods of exposure to condensing humidity and UV light (UVCON test). X-ray photoelectron spectroscopy (XPS) has shown that the lacquer drying process gives rise to a number of discontinuities on the lacquer surface (pores), exposing the phosphate layer formed on the original metallic coating surface. It is interesting to note the detection of fluoride and nitrite ions and phosphoric acid not combined with zinc (perhaps as HP[O.sub.4.sup.2-] and [H.sub.2]P[O.sub.4.sup.-]) on the lacquered surfaces before testing, which suggests a tendency of these species to concentrate on the outer surface of the phosphate layer or at the lacquer-phosphate layer interface (in zones covered by the lacquer). After one day of exposure to the UVCON test, XPS reveals the disappearance of the fluoride and nitrite ions and of the free phosphoric acid. After 15 days of exposure to the UVCON test, the carbon content is seen to have decreased considerably, while the zinc, phosphorus, and titanium contents have risen. The low atomic percentages of carbon (only moderately higher than those obtained with the coatings in bare state) and Zn/P atomic ratios close to 1.5 suggest the removal of a very substantial percentage of the lacquer, leaving the zinc phosphate formed on the surface of the different metallic coatings exposed. This quick and significant drop in the lacquer content barely seems to have a repercussion on the degradation of the metallic substrate during the UVCON test, since its visual aspect remains unaltered.

Keywords: Lacquer, phosphating reagents, interface, X-ray photoelectron spectroscopy (XPS), fluorides, nitrites

**********

In a humid atmosphere, white corrosion products, mainly zinc oxides and hydroxides, form quite quickly on the surface of galvanized steel. (1,2) To avoid this problem, some treatments applied after galvanizing afford additional protection to the galvanized steel. The presence of chromium in inhibitor formulations can substantially prolong the white corrosion resistance of zinc, compared with the effect of other inhibitor compounds that do not contain chromium. However, since chromates contaminate the environment, a search is currently under way for nontoxic and chromate-free formulations that develop similar protective properties. (3-5)

The application of organic films also supplements the protection of galvanized steel. (6-8) Lacquer films protect the galvanized steel sheet during its transportation and storage and prevent the effect of fingerprints on the processing and utilization of finished parts. A further advantage of organic films is that they act as a dry lubricant, making it unnecessary to apply lubricating and anticorrosive protection oils which need to be removed before final painting. (9)

In recent years, attention has been drawn to a new technique which combines the application of phosphating reagents and a thin organic film in one simple step and on the same production line, avoiding the multiple operations of other processes. (10-13)

Many researchers have studied the chemical reactions that take place between the surface of metallic substrates and the phosphating bath, (14) as well as the effect of the presence of accelerating agents (fluorides, nitrites, and titanium phosphates) on the phosphate coating. (9,15-23) However, fewer researchers have focused on the characteristics of the interface formed on metallic coatings when phosphating reagents and lacquer are simultaneously applied as well as its subsequent corrosion resistance. Lin et al. (10-13) suggest a chemical and physical reaction of phosphating reagents with the metallic surface to produce the phosphate layer and, at the same time, the formation of P-O-C (phosphorus-oxygen-carbon) type covalent bonds with the organic coating. This treatment improves the coating's adhesion to the surface, and also reduces the corrosion of the substrate without using deleterious hexavalent chromium ([Cr.sup.6+]) compounds. (13) These studies refer to organic films of a thickness of several tens of microns. (10-13) The technological interest (24) of combinations of phosphating reagents and lacquer films of the order of 1 [micro]m as additional protection of galvanized steel has encouraged us to consider them in this study.

The UVCON test, an accelerated test that uses alternative cycles of humidity and radiation, is recently coming to be seen as an efficient instrument for studying painted metals. Notable advantages of this test are its lower aggressivity than the salt fog test and the inclusion of exposure to UV radiation. Verstappen et al. (25) note that many consider this to be the best test for painted hot-dip galvanized steel. Its main advantage is that it is a slow test. The special technological interest aroused by the degradation of water-based acrylic coatings in exposure to UV radiation (26-28) and the possible high deterioration rate of the lacquer film used in this study (thickness less than 1 [micro]m) justify the decision to use this test.

[FIGURE 1 OMITTED]

With the assistance of X-ray photoelectron spectroscopy (XPS), this work attempts to establish possible relations between the phosphate layer formed by applying a lacquer with phosphating reagents and the presence or absence of segregated elements on the surface of the metallic coating used as substrate. It has also been intended to obtain information on the chemical composition of the phosphate layer-lacquer interface resulting from the simultaneous application of phosphating reagents and lacquer. Finally, by means of the UVCON test and the use of a similar lacquer formulated with chromating reagents, an attempt has been made to highlight the relation between the rapid loss of lacquer adhesion and the presence of ionic species at the phosphate layer-lacquer interface in the specific case of lacquers with phosphating reagents.

EXPERIMENTAL

Materials Tested

SUBSTRATES -- The following three types of zinc-based coatings, obtained by different companies by means of the hot-dip galvanizing of steel sheet during its manufacturing process, have been tested as substrates:

--pure zinc with coating masses of 100 and 275 g/[m.sup.2], referred to as coatings Z100 and Z275, respectively.

--zinc-based alloy with approximately 5% Al (galfan) with coating masses of 100 and 250 g/[m.sup.2], referred to as coatings ZA100 and ZA250, respectively.

--zinc-based alloy with approximately 10% Fe (galvanneal) with a coating mass of 110 g/[m.sup.2], referred to as coating ZF110.

The specimens were cut from galvanized steel sheets provided by Aceralia (ZF110 coating), VASL (Z100 coating), CRM (Z275 coating), Thyssen (ZA100 coating), and Sollac (ZA250 coating).

LACQUER -- A thin acrylic film of approximately 1 [micro]m thickness incorporating phosphate or chromate type inhibitor compounds, was applied to the five types of hot-dip galvanized coatings: (Z100, Z275, ZF110, ZA100, and ZA250). The films were formed by the action of an aqueous dispersion of polymer particles of colloidal dimensions and phosphorus or chromium compounds on the metallic surface. Subsequent evaporation of the water left a polymer film (polyacrylate type) adhered to the metallic surface. Since these are industrial processes currently under development, the company supplying the lacquer (Henkel) has preferred not to reveal detailed information on the composition of the products used and the procedure followed in the treatment of the galvanized sheets.

[FIGURE 2 OMITTED]

XPS Analysis

Photoelectron spectra were recorded using a Fisons MT500 spectrometer equipped with a hemispherical electron analyzer (Clam 2) and an Mg K[alpha] X-ray source operated at 120 W. The specimens were mechanically fixed on small flat discs supported on an XYZ manipulator placed in the analysis chamber. The residual pressure in this ion-pumped analysis chamber was maintained below 5 X [10.sup.-7] Pa during data acquisition. Spectra were collected for 20-90 min, depending on the peak intensities, at a pass energy of 20 eV, which is typical of high-resolution conditions. Intensities were estimated by calculating the area under each peak after smoothing and fitting the experimental curve to a mix of Lorentzian and Gaussian lines of variable proportion. Although specimen charging was observed, accurate binding energies (BE) were determined by referencing to the adventitious C1s peak at 285.0 eV. Atomic ratios were computed from peak intensity ratios and reported atomic sensitivity factors. (29) The high resolution peaks were fitted using the XPSPEAK program, version 4.1 (Raimund W. Kwok, University of Hong Kong).

[FIGURE 3 OMITTED]

UVCON Test

The UVCON test was performed according to ASTM G 53-77 standard (30) using a period of four hours of UV radiation followed by four hours of darkness with water condensation, with a period of a half-hour between them, and with temperatures in the black panel (aluminum referenced panel coated with a black paint used to measure the temperature in the specimen rack) of 65[degrees] and 45[degrees]C, respectively. The test was performed in a QUV Accelerated Weathering Tester--Model QUV/SE equipment with FS-40, 313/280 nm lamps.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

EIS Measurements

The possibility of protecting galvanized coatings using lacquers with phosphating reagents has been studied by means of the UVCON test and subsequent immersion in a 3% NaCl solution for one day.

An NaCl solution was used as corrosive medium for the EIS test, since distilled water possesses low conductivity, which is an impediment to electrochemical measurements. One day was needed to reach a certain stability in the corrosion potential which allows EIS measurements to be made.

The electrochemical cell was prepared by fixing a glass cylinder onto the lacquered metallic sheet previously exposed to the UVCON test. The cell was formed by a saturated calomel reference electrode (SCE), a platinized titanium counter electrode, and the working electrode was comprised by the surface (9 [cm.sup.2]) of the coated sheet delimited by the inner wall of the cylinder.

Impedance data was collected at the [E.sub.corr] potential, with a Solartron 1250 frequency response analyzer connected to an electrochemical interface. The analyzed frequency interval went from 65 kHz to 1 mHz, with frequency values spaced logarithmically (five per decade). The width of the sinusoidal voltage signal applied to the system was 10 mV rms (root-mean-square).

RESULTS

XPS Analysis of the Surface of Unexposed Lacquered Coatings

Table 1 shows the elemental composition obtained by XPS on the surface of galvanized steel (Z100 and Z275), galfan (ZA100 and ZA250), and galvanneal (ZF110) coatings after applying the lacquer with phosphating reagents. An important carbon content was observed, which may be related to the presence of the acrylic lacquer film. Significant amounts of zinc, phosphorus, titanium, fluorine, and nitrogen are also detected.

Figure 1 compares the C1s high resolution XPS spectra obtained on the outer surface of the unexposed lacquered galvanized steel, galvanneal, and galfan (Figures 1a-c) with those obtained after one day of exposure to the UVCON test (Figures 1d-f). The spectra may be fitted to three components with different intensities: the first and most intense, at 285.0 eV, may be attributed to the presence of C-C/C-H groups; a second less intense component, at 286.5 eV, may be associated with the presence of C-O groups; and a third component at the highest binding energies, 288.5 eV, is associated with the presence of carboxyl type O-C=O groups. (31)

Figure 2 compares the O1s high resolution spectrum obtained on the surface of unexposed lacquered galvanized steel (Figure 2a) with that observed after one day of exposure to the UVCON test (Figure 2c) and the Zn2[p.sub.3/2] high resolution spectra obtained on these same surfaces (Figures 2b and 2d). These spectra are representative of the O1s and Zn2[p.sub.3/2] spectra observed on the surfaces of the other lacquered substrates (ZA100 and ZF110). The O1s spectrum shows the most intense component at a binding energy of 531.8 eV, associated with the presence of C=O groups (acrylate type)/P[O.sub.4] (3-32) and two other less intense components that appear at binding energies of 530.0 and 533.5 eV which may be attributed to the presence of oxygen in the form of oxide and C-O groups (acrylic lacquer)/[H.sub.2]O (Figure 2c). The Zn2[p.sub.3/2] high resolution spectra contain one single component with a binding energy of 1022.5 eV (Figures 2b and 2d). This binding energy is typical of zinc in its form [Zn.sup.2+]. (33)

Figure 3 shows the P2p, Ti2[p.sub.3/2], N1s, and F1s high resolution XPS spectra obtained on the surface of the lacquered Z100 coating in unexposed state (Figures 3a-d) and after one day of exposure to the UVCON test (Figures 3e-h). These spectra are representative of the P2p, Ti2[p.sub.3/2], N1s, and F1s spectra observed on the surfaces of the other lacquered substrates (ZA100 and ZF110). In the P2p spectrum (Figures 3a and 3e), one component is seen with a binding energy of 134.2 eV, characteristic of phosphorus in an ionic state (of the type P[O.sub.4.sup.3-]). (32) In the Ti2[p.sub.3/2] spectrum (Figures 3b and 3f), one component is observed at 458.5 eV which may be attributed to the presence of Ti as [Ti.sup.4+]. (34) In the N1s spectrum (Figure 3c), two components are observed at binding energies of 400.2 and 402.5 eV, respectively. The component that appears at the lower binding energy may be associated with the presence of nitrogen in N-H covalent bonds (e.g., as amine or amide), while the component at higher binding energies may be associated with the presence of nitrogen in an ionic state (as nitrites). (35) Finally, the F1s spectrum (Figure 3d) may be fitted to a component with a binding energy of 686.0 eV associated with the presence of fluoride ions ([F.sup.-]). (36)

XPS Analysis of the Surface of Lacquered Coatings After One Day of Exposure to the UVCON Test

The elemental percentages determined on the surfaces of the lacquered Z100, ZA100, and ZF110 coatings after one day of exposure to the UVCON test are shown in Table 1. No great differences were observed in the carbon and oxygen contents corresponding to the three types of lacquered metallic coatings compared with the unexposed surfaces, though an increase is seen in the zinc content. On the other hand, it is interesting to note the absence of fluorine and the reduction by approximately half of the nitrogen content compared with the unexposed surfaces.

No significant changes are observed in either the shape or the position of the C1s, O1s, Zn2p, P2p, and Ti2[p.sub.3/2] high resolution XPS spectra obtained after one day of exposure to the UVCON test, compared with those obtained on the unexposed lacquered surfaces (Figures 1-3). Figure 3g shows the N1s high resolution XPS spectrum obtained on the surface of lacquered galvanized steel after one day of exposure. In contrast to the unexposed lacquered coating (Figure 3c), the N1s high resolution XPS spectrum reveals the absence of the component at higher binding energies associated with the presence of nitrites. Furthermore, the intensity of the F1s peak decreased considerably after one day of exposure (Figure 3h).

XPS Analysis of the Surface of Lacquered Coatings After 15 Days of Exposure to the UVCON Test

The elemental composition obtained by XPS on the surfaces of the different lacquered metallic coatings (Z100, ZF110, and ZA100) after 15 days of exposure to the UVCON test are also shown in Table 1. A considerable reduction in the carbon content and an increase in the oxygen, zinc, phosphorus, and titanium contents are detected on these surfaces. Attention is drawn to the absence of significant amounts of F and N. It is also interesting to see that the Zn/P atomic ratios obtained by XPS on the lacquered Z100 and ZF110 coatings reached the value of 1.5 corresponding to (P[O.sub.4])[.sub.2][Zn.sub.3].

[FIGURE 6 OMITTED]

EIS and SEM Characterization of the Lacquered Coatings After Exposure to the UVCON Test

Figure 4 shows the evolution of the Nyquist diagrams obtained for the lacquered metallic coatings after exposure to the UVCON test in a quiescent and aerated 3% NaCl solution. With the galvanized and galfan coatings, an arc (or flattened semicircle) is seen at HF-MF, followed by a second arc or ill-defined tail (Figures 4a and 4c). As exposure time to the UVCON test increased, the arc or tail at LF tends to become more patent (Figures 4d and 4f). At the start of the test the first semicircle is due to the organic film, but after longer testing times (15 days) it may be attributed to charge transfer and the effect of ionic double layer capacitance. The LF arc or tail may indicate a diffusion process in finite thickness layer, related mainly to the reduction of oxygen. The diagrams referring to the galvanneal coating (Figures 4b and 4e) differ from those above in that they apparently present one single arc, possibly as a result of the superposition of several effects, including an inductive effect at the lowest frequencies. (37)

[FIGURE 7 OMITTED]

Figure 5 shows the evolution of the [R.sub.t] values deduced from the HF-MF arcs with increasing exposure time to the UVCON test. Taking the bare coating as reference, the lacquer films with phosphating reagents tended to increase the size of the HF-MF arc in the Nyquist diagram and, therefore, the value of [R.sub.t] associated with this arc. However, this effect was lost after only 15 days of exposure to the UVCON test of the galvanized, galfan, and galvanneal coatings protected with lacquer with phosphating reagents, again reaching similar [R.sub.t] values to those of the bare coating (Figure 5).

Figure 6 compares the micrographs obtained on the surface of the lacquered ZF110 coating after one day of exposure with those obtained after 15 days. No marked differences are observed during the test. Since the lacquer was transparent, and furthermore of very low thickness (approximately 1 [micro]m), it has not been possible to observe changes in it.

Effect of Exposure to the UVCON Test on the Surface of the Lacquer with Chromium

Figure 7 compares the variation in the atomic percentages of carbon on the outer surface of a lacquer with phosphating reagents (continuous line) and on a lacquer with chromium (dotted line) as a function of exposure time to the UVCON test. On the different metallic substrates coated with the lacquer with phosphating reagents a considerable reduction is seen in the carbon contents after 15 days of exposure, so that these contents (Figure 7, continuous line) differed relatively little from those of the bare metallic coating (Figure 7, dashed line). In contrast, with the lacquer with chromium, the carbon contents (Figure 7, dotted line) remain high throughout the 15 days and similar to those obtained with the unexposed lacquered coatings.

[FIGURE 8 OMITTED]

DISCUSSION

It is of fundamental interest to know the transformations that take place on the surface of the metallic substrate during the drying of the lacquer with phosphating reagents, since they determine the initial adhesion of the lacquer and ultimately the corrosion resistance of the metallic substrate. Given that fast and significant detachment of a lacquer film incorporating phosphating reagents has been observed during the UVCON test (Figures 5 and 7), the discussion will focus mainly on the segregation of fluoride and nitrite type ionic species towards the interface, weakening the possible bonds that are established between the lacquer and the phosphating layer. In view of the low thickness of the lacquer (less than 1 [micro]m) and its porous nature, (37) the relative changes in the chemical composition of the surface of the material obtained by XPS provide valuable information on the interface resulting from the application of the lacquer and its possible evolution as a result of exposure to the UVCON test.

CHEMICAL COMPOSITION OF THE INTERFACE RESULTING FROM DRYING OF THE LACQUER: In the XPS analyses performed on the outer surface of the specimens after the lacquer drying process, attention is drawn to the detection of zinc and phosphorus (Table 1), which indicated the presence of a series of discontinuities or pores in the lacquer film that leave the phosphate layer formed on the original metallic coating surface exposed. These results are similar to those obtained by Barranco et al., (37) using EIS measurements of interfacial capacitance, which indicated a high initial porosity of the lacquer film. On the other hand, the absence of aluminum and iron suggests that these alloying elements are not incorporated in a significant way in the phosphate layer.

In Table 1, comparison of the atomic percentages of zinc obtained on the unexposed lacquer surfaces revealed that Zn contents are two or even three times higher on the lacquered galvanized steels (Z100 and Z275) than on the coatings that incorporate alloying elements (ZA100, ZA250, and ZF110). These important differences cannot be attributed exclusively to the replacement of part of the zinc content by iron or aluminum in the chemical composition of the metallic coatings (maximum of 10% iron in the galvanneal substrate and 5% aluminum in the galfan substrate). Figure 8 shows the Zn, O, and Al contents obtained on the outer surfaces of the bare Z100, ZA100, and ZF110 metallic coatings and their evolution with sputtering time. The zinc content on the surface and sub-surface of the galvanized steel coating (Z100) is seen to be twice that of the galfan (ZA100) and galvanneal (ZF110) coatings (Figure 8a). It might be imagined that the segregation towards the surface of significant aluminum contents during the obtainment of galfan coatings (Figure 8c) and the presence of oxygen-rich impurities (Figure 8b) from the alkaline cleaning process during the obtainment of the galvanneal coating (38) may have caused a certain inhibition of the surface reaction of dissolution of zinc and its subsequent precipitation in the form of phosphate.

[FIGURE 9 OMITTED]

In the XPS analyses carried out on the surfaces of the different lacquered substrates (Table 1 and Figures 3c and 3d) the presence of an important level of fluoride and nitrite type ionic compounds is initially observed. The Zn/P ratios in Table 1 are well below the theoretic value of 1.5 corresponding to (P[O.sub.4])[.sub.2][Zn.sub.3], which may be due to the presence of a significant amount of phosphoric acid not combined with zinc (perhaps as HP[O.sub.4.sup.2-] and [H.sub.2]P[O.sub.4.sup.-]). (16) In the literature that has been consulted, only Puomi et al. (39) have noted the presence of a high fluoride content on the surface of galvanized steels and galfan treated with a commercial zirconic acid-based pretreatment without the final stage of rinsing with water. In our case, it seems logical to attribute the presence of these compounds to the absence of a rinsing-type process in the combined application of the lacquer and phosphating reagents.

It is important to note the detection by XPS of significant amounts of titanium on the outer surface of the phosphate layer (Table 1 and Figures 3b and 3f). Figure 9 shows the variation in the titanium content detected on the surface of the lacquered ZA100 coating as a function of the phosphorus content after 1, 15, 30, and 60 days of exposure to the UVCON test. A tendency of the atomic percentage of Ti to rise as the phosphorus content increased is clearly observed. The literature mentions difficulties in the detection of titanium on the surface when it acts as a nucleation agent for phosphate crystals. (9,18,21,22,34) In particular, Van Roy et al., (34) using titane phosphate nucleation agents on aluminum substrates, did not find evidence of these species by SEM-EDX, Auger, or XPS, and only by AFM observed that these phosphates had a thickness of 7-8 nm and a length of 0.1-0.3 [micro]m. In this study, the presence of titanium may suggest that this element comes to form part of the chemical composition of the outer surface of the phosphate crystals.

[FIGURE 10 OMITTED]

EFFECT OF ONE DAY OF EXPOSURE TO THE UVCON TEST ON THE CHEMICAL COMPOSITION OF THE SURFACE: It is interesting to note the absence of nitrite or fluoride compounds and the increase in the zinc content on the surface of the specimens after one day of exposure to the UVCON test (Table 1 and Figures 3g and 3h) compared to the unexposed specimens (Table 1 and Figures 3c and 3d). Only one day of exposure in high humidity conditions probably causes the dissolution of the nitrite and fluoride ions and phosphoric acid. This rapid dissolution would be favored by the porous character of the initial lacquer film and because these compounds are loosely bonded to the phosphate layer. Probably for this same reason, Puomi et al. (39) also observed rapid dissolution of the fluorine present on the surface of a galvanized steel previously treated with a commercial zirconic acid-based pretreatment during the final stage of rinsing with water.

It is of fundamental interest to know whether this high ionic species content resulting from the lacquer drying process is preferentially located at the interface between the phosphate layer and the lacquer or on the outer surface of the lacquer. If these ionic species were to be homogeneously distributed on the outer surface of the metallic coating-lacquer system, their removal by humidity should lead to an increase in the carbon content (characteristic element of the lacquer) in the coated zones, and, at the same time, an increase in the zinc concentration (specific element of the phosphate layer) in the discontinuities in the lacquer film that leave the phosphate layer exposed. After one day of exposure to the UVCON test similar C contents are obtained to those found on the surface of the unexposed lacquered substrate (Table 1). This absence of significant changes in the C content suggests that the soluble ionic species are not mainly concentrated on the outer surface of the lacquer, but rather on the phosphate layer exposed in its discontinuities. In consonance with this interpretation, however, a notable increase in the Zn content has been determined, probably because the surface of the exposed phosphate layer has been left free of soluble species.

EFFECT OF 15 DAYS OF EXPOSURE TO THE UVCON TEST ON THE LACQUERED METALLIC COATING SURFACES: Since the lacquer is transparent and very thin (thicknesses of less than 1 [micro]m), SEM observation has not revealed appreciable changes on the outer surface with the UVCON test (Figure 6). However, it has been possible to establish a substantial rise in the porosity of the lacquer after 15 days of exposure to the UVCON test by means of electrochemical impedance spectroscopy (EIS) and surface analysis (XPS) techniques.

In Figures 5a-c it is possible to chart the progress of [R.sub.t] values with UVCON exposure time on the Z100, ZA100, and ZF110 coatings protected by the lacquer. Irrespective of the alloying elements content in the original metallic coating, the [R.sub.t] values of the lacquered coatings are considerably lower after 15 days, becoming similar to those observed on the bare metallic coatings (Figures 5a-c). The important reduction in [R.sub.t] values observed on these lacquered substrates suggests a significant increase in the size of the lacquer pores, which caused the protective effect of the lacquer to disappear.

As occurs with the [R.sub.t] values, the carbon contents observed by XPS on the lacquer with phosphating reagents are considerably lower after 15 days than those observed at the start and after one day of exposure, reaching similar values to the same metallic coatings in bare state (Figures 7a-c). After 15 days, the important reduction in the carbon contents and the rise in the Zn/P atomic ratio (Table 1) seemed to be related with a considerable deterioration of the lacquer layer and its replacement by a layer of zinc phosphate ((P[O.sub.4])[.sub.2][Zn.sub.3]) as the major component of the surface.

This important reduction in the carbon content (degradation of the lacquer film) observed by XPS after 15 days of UVCON exposure could be related in some way with the loss of adhesion of the lacquer film due to the presence of the ionic species at the interface between the phosphate layer and the lacquer. It may also possibly be caused by the combined effect of humidity and UV radiation in the UVCON test on the outer surface of the organic film. However, the latter seems to be unlikely in view of the absence of significant changes when the same lacquer film is tested accompanied by chromating reagents instead of phosphating reagents. Therefore, the first explanation is the most plausible.

Possible processes that might occur during the drying of the lacquer with phosphating reagents and in subsequent exposure to the UVCON test are sketched in Figure 10. As a result of the application of the lacquer, the segregation of ionic species (phosphate layer growth accelerators) takes place towards the interface between the phosphate layer and the lacquer (Figure 10a). In exposure to the humidity of the UVCON test, practically from the start there is a possibility that water molecules may come into contact with these ionic species, directly via macropores or microscopic defects or pores in the thin acrylate film (Figure 10b). Since the ionic concentration at the interface is greater than in the water condensed during the test, moisture will be drawn through the lacquer film, which will be locally lifted off the metal substrate (osmotic blistering, Figure 10c).

CONCLUSIONS

The main conclusions obtained by XPS analysis of the outer surfaces of the lacquered coatings (galvanized, galvanneal, and galfan) after the combined application of lacquer and phosphating reagents and their exposure to the UVCON test are as follows:

(1) Due to the higher zinc content present on the outer surface of the galvanized coating, zinc dissolution reactions and subsequent precipitation in the form of zinc phosphate were favored compared with the other two coatings.

(2) In the galfan and galvanneal coatings, the segregation of an important aluminum content towards the outer surface of the galfan coating, and the presence of oxygen-rich impurities in the galvanneal coating, seemed to inhibit the phosphating reaction.

(3) On the surface of all the coatings an important level of fluoride and nitrite type species has been observed as a result of the combined application of lacquer and phosphating reagents. These species are concentrated at the lacquer/phosphating layer interface.

(4) The exposure of the lacquered coatings to the UVCON test has in all cases caused the rapid disappearance (dissolution) of the ionic species initially present at the lacquer/phosphating layer interface (one day has been sufficient). It has been seen that this process is closely associated with the deterioration of the lacquer layer.

(5) The complementary EIS tests confirmed the conclusions of the XPS tests on the rapid degradation of the lacquer film due to the effect of exposure of the lacquered coatings to the UVCON test.

Table 1 -- Atomic Percentages Observed by XPS on the Outer Surface of
the Different Unexposed Lacquered Metallic Coatings, After One Day and
15 Days of UVCON Exposure

Specimen             %C  %O  %Zn  %P   %Ti  %F  %N   %Al  %Zn/%P

Unexposed (0 Days)
Z100                 63  28  1    3.5  1    1   2.5  0    0.3
Z275                 64  27  1.5  3    0.5  1   3    0    0.5
ZA100                70  25  0.5  1    0.5  1   2    0    0.5
ZA250                69  25  0.5  1    1    1   2.5  0    0.5
ZF110                73  22  0.5  1    0.5  1   2    0    0.5

1 Day of Exposure
Z100                 65  28  2    3    1    0   1    0    0.7
ZA100                68  27  1    1.5  0.5  0   1    1    0.7
ZF110                69  27  2    1    0.5  0   0.5  0    2

15 Days of Exposure
Z100                 43  41  8    5    3    0   0    0    1.6
ZA100                50  38  4    3.5  2    0   0    2    1.1
ZF110                56  34  5    3.5  1.5  0   0    0    1.4

ACKNOWLEDGMENTS

The research was supported by ECSC project 7210-PR 121. The authors acknowledge Thyssen Krupp Stahl AG, Centre de Recherches Metallurgiques, Aceralia Corporacion Siderurgica, Sollac and Voest-Alpine Stahl Linz for providing the coated steel sheets for the performance of the research. Henkel KGaA is acknowledged for providing the lacquer system used in this study.

References

(1) Graedel, T.E., "Corrosion Mechanisms for Zinc Exposed to the Atmosphere," J. Electrochem. Soc., 130, 193C (1989).

(2) Zhu, F., Persson, D., Thierry, D., and Taxen, C., "The Influence of C[O.sub.2] and NaCl on the Atmospheric Corrosion of Zinc," Corrosion, 56, 1256 (2000).

(3) Treacy, G.M., Wilcox, G.D., and Richardson, M.O.W., "Behavior of Molybdate-Passivated Zinc Coated Steel Exposed to Corrosive Chloride Environments," J. Appl. Electrochem., 29, 647 (1999).

(4) Manov, S., Lamazonere, A.M., and Aries, L., "Electrochemical Study of the Corrosion Behavior of Zinc Treated with a New Organic Chelating Inhibitor," Corros. Sci., 42, 1235 (2000).

(5) Aramaki, K., "The Inhibition Effects of Chromate-Free Anion Inhibitors on Corrosion of Zinc in Aerated 0.5 M NaCl," Corros. Sci., 43, 591 (2001).

(6) McMurray, H.N., Powell, S.M., and Worsley, D.A., "Mechanistics Changes in Cut Edge Corrosion Induced by Variation of Inhibitor Pigmentation in Organically Coated Galvanized Steel," British Corrosion Journal, 36, 42 (2001).

(7) Garlatti, G. and Leonardi, F., "I Revestimenti Organici Nelle Linee di Zincatura per Mastri," La Metallurgie Italiana, No. 9, 29 (2001).

(8) Mertens, S.F., Xhoffer, C., DeCooman, B.C., and Temmerman, E., "Short-Term Deterioration of Polymer-Coated 55% Al-Zn-Part 1: Behavior of Thin Polymer Films," Corrosion, 53, 381 (1997).

(9) Wolpers, M. and Angeli, J., "Activation of Galvanized Steel Surfaces Before Zinc Phosphating-XPS and GDOES Investigations," Appl. Surf. Sci., 179, 281 (2001).

(10) Wang, C., Chuang, Y.Y., and Lin-C-T., "In-Situ Phosphatizing Coatings I: An Air-Dried Lacquer System," JOURNAL OF COATINGS TECHNOLOGY, 71, No. 892, 61 (1999).

(11) Yu, T., Chuang, Y.Y., and Lin, C-T., "In-Situ Phosphatizing Coatings II: A High-Solids Polyester Baking Enamel," JOURNAL OF COATINGS TECHNOLOGY, 71, No. 892, 69 (1999).

(12) Yu, T., Chuang, Y.Y., and Lin, C-T., "In-Situ Phosphatizing Coatings III: A Water Reducible Alkyd Baking Enamel," JOURNAL OF COATINGS TECHNOLOGY, 71, No. 895, 87 (1999).

(13) Neuder, H., Chuang, Y.Y., and Lin, C-T., "Chrome-Free Single Step In-Situ Phosphatizing Coatings on a Ti-6Al-4V Titanium Alloy," JOURNAL OF COATINGS TECHNOLOGY, 74, No. 930, 37 (2002).

(14) Rausch, W., The Phosphating of Metals, ASM International, Metal Park and Finishing Publications, Teddington, Middlesex, England, 1990.

(15) Heung, W.F., Yang, Y.P., Wong, P.C., Mitchell, K.A.R., and Foster, T., "XPS and Corrosion Studies on Zinc Phosphate Coated 7075-T6 Aluminum Alloy," J. Mater. Sci., 29, 1368 (1994).

(16) Heung, W.F., Yang, Y.P., Zhou, M.Y., Wong, P.C., Mitchell, K.A.R., and Foster, T., "X-Ray Photoelectron Spectroscopy and Corrosion Studies of Zinc Phosphate Coated on 7075-T6 Aluminum Alloy. Part II: Effect of Coating pH," J. Mater. Sci., 29, 3653 (1994).

(17) Heung, W.F., Wong, P.C., Mitchell, K.A.R., and Foster, T., "Studies of Zinc Phosphate Coatings on 7075-T6 Aluminum Alloy," J. Mater. Sci. Lett., 14, 1461 (1995).

(18) Ying, F., Zhou, M.Y., Flinn, B.J., Wong, P.C., Mitchell, K.A.R., and Foster, T., "The Effect of Ti-Colloid Surface Conditioning on the Phosphating of 7075-T6 Aluminum Alloy," J. Mater. Sci., 31, 565 (1996).

(19) Fang, J., Wong, P.C., Mitchell, K.A.R., and Foster, T., "Observations Related to the Phosphating of Aluminum Alloy 7075-T6 Using a Spraying Technique," J. Mater. Sci., 33, 1541 (1998).

(20) Kok, W.H., Sun, X, Shi, L., Wong, K.C., Mitchell, K.A.R., and Foster, T., "Formation of Zinc Phosphate Coatings on AA6061 Aluminum Alloy," J. Mater. Sci., 36, 3941 (2001).

(21) Tegehall, P.E., "Colloidal Titanium Phosphate, the Chemical Activator in Surface Conditioning before Zinc Phosphating," Colloids Surf., 42, 155 (1989).

(22) Tegehall, P.E., "The Mechanism of Chemical Activation with Titanium Phosphate Colloids in the Formation of Zinc Phosphate Conversion Coatings," Colloids Surf., 49, 373 (1990).

(23) Van Roy, I., Terryn, H., and Goeminne, G., "Study of the Phosphating Treatment of Aluminum Surfaces: Role of the Activating Process," Colloids Surf. A: Physicochemical and Engineering Aspects, 136, 89 (1998).

(24) Wormuth, R., ESCS Project 7219/PR/121, Final Report, 2002.

(25) Verstappen, H.G.J.M., Van Susteren, S.J., and Van Ooij, W.J., "Durability of Powder-Coated Hot-Dip Galvanized Steel," JOURNAL OF COATINGS TECHNOLOGY, 74, 43 (2002).

(26) Vink, P., Koster, T.P.M., Fontijn, H.F.N., and Mackor, A., "UV Stability of Water-Borne Acrylic Coatings," Polym. Degrad. Stab., 48, 155 (1995).

(27) Allen, N.S., Parker, M.J., Regan, C.J., McIntyre, R.B., and Dunk, W.A.E., "The Durability of Water-Borne Acrylic Coatings," Polym. Degrad. Stab, 47, 117 (1995).

(28) Allen, N.S., Regan, C.J., McIntyre, R., Johnson B., and Dunk, W.A.E., "The Photostabilization of Water-Borne Acrylic Coating Systems," Polym. Degrad. Stab., 52, 67 (1996).

(29) Wagner, C.D., Davis, L.E., Zelleer, M.V., Taylor, J.A., Raymond, R.H., and Gale, L., "Empirical Atomic Sensitivity Factors of Quantitative Analysis by Electron Spectroscopy for Chemical Analysis," Surf. Interface Anal., 3, 211 (1981).

(30) ASTM Standard G 53-77, Practice for Operating Light- and Water-Exposure Apparatus (Fluorescent UV-Condensation Type) for Exposure of Nonmetallic Materials."

(31) Pijpers, A.P. and Donners, W.A., "Quantitative Determination of the Surface Composition of Acrylate Copolymer Latex Film by XPS (ESCA)," J. Polym. Sci., Polym. Chem. Ed., 23, 453 (1985).

(32) Brow, R.K., "An XPS Study of Oxygen Bonding in Zinc Phosphate and Zinc Borophosphate Glasses," J. Non-Cryst. Solids, 194, 267 (1996).

(33) Maeda, S. and Yamamoto, M., "The Role of Chromate Treatment After Phosphating in Paint Adhesion," Prog. Org. Coat., 33, 89 (1998).

(34) Day, M.E. and Defino, M., J. Electrochem. Soc., 143, 264 (1996).

(35) Hersh, L.S., Onyiriuka, E.C., and Hertl, W., "Amine-Reactive Surface Chemistry of Zinc Phosphate Glasses," J. Mater. Res., 10, 2120 (1995).

(36) Wolan, J.T. and Hoflund, G.B., "Surface Characterization Study of AgF and Ag[F.sub.2] Powders Using XPS and ISS," Appl. Surf. Sci., 125, 251 (1998).

(37) Barranco, V., Feliu, S. Jr., and Feliu, S., "Trabajo de Impedancia, Part 2," Corros. Sci., submitted for publication.

(38) Vazquez, A.J., "El Desengrase de las Superficies Metaalicas en la Galvanizacion por Inmersion en Caliente," Revista de Metalurgia Madrid, 10, 193 (1974).

(39) Puomi, P., Fagerholm, H.M., Rosenholm, J.B., and Jyrkas, K., "Optimization of Commercial Zirconic Acid Based Pretreatment on Hot-Dip Galvanized and Galfan Coated Steel," Surf. Coat. Technol., 115, 70 (1999).

S. Feliu, Jr. ([dagger]) and V. Barranco -- Centro Nacional de Investigaciones Metalurgicas*

*Avda. Gregorio del Amo, 8, 28040-Madrid, Spain.

([dagger]) Author to whom correspondence should be addressed: fax: [phone]; email: [e-mail].