Surface Treatment and Corrosion
Corrosion Process of Inorganic Zinc-Rich Painted Steel Exposed to a High-Chloride Atmospheric Environment
2018 Volume 58 Issue 2 Pages 316-322
Details
2018 Volume 58 Issue 2 Pages 316-322
The corrosion protection behavior of an inorganic zinc-rich paint (IZRP) layer on steel after exposure for five years in Okinawa, relatively high-chloride atmosphere, was studied. Zinc particles were uniformly dissolved in the IZRP-coated steel, independent of the distance from the IZRP surface. From XRD, FT-IR, Raman spectroscopy and EPMA analyses, Zn5(OH)8Cl2H2O and ZnO were generated in the entire corrosion layer. Mg and Ca were observed in the outer area. We proposed the corrosion process of IZRP-coated steel from a detailed analysis of the corrosion products.
Organic and inorganic zinc-rich paints (ZRP) have been widely used as primers for coatings on bridges and ships.1) Metallic zinc particles, which are highly loaded with binder in the ZRP layer, are expected to protect steel in corrosive environments. Protection by ZRP is assumed to be achieved by two mechanisms: sealing of voluminous zinc corrosion products and cathodic protection by zinc particles.2,3)
Theiler showed that voluminous zinc corrosion products, predominantly comprising basic zinc carbonate, act as a barrier against penetration of corrosive agents and slow down the corrosion rate of ZRP-coated steel.4) Moreover, he pointed out the optimal ratio of zinc content for the protection property of ZRP.4) Feliu et al. examined ZRP coating as a function of the composition of binders, content of zinc particles and amount of conductive extenders of Fe2P.5) They exposed the steel coated with ZRP for 10 years to atmosphere in Madrid, Spain.5) They found that no iron rust was seen when the steel was covered by the epoxy silicate-based ZRP coating. When the steel was covered by the ethyl silicate-based ZRP coating, however, iron rust appeared on the ZRP-coated steel. Iron rust also appeared when the ZRP layer contained an insufficient amount of zinc and some amount of zinc was replaced by Fe2P. Romagnoli et al. tested the protection property of four zinc-ethyl silicate paints.6) Abreu et al. examined the electrochemical impedance spectroscopy of ZRP-coated steel immersed in 3% NaCl solution.7) They found from the transient of the resistance of the ZRP layer that the protection was shifted from initial cathodic protection to the barrier protection in the later stage. They suggested that the grain size of powder zinc should be uniform to extend the period of the cathodic protection stage. The study of the ZRP-coated steel has focused on evaluating the ZRP property using electrochemical techniques such as impedance and corrosion potential measurement and/or corrosion tests. There have been, however, few detailed studies on the long-term atmospheric corrosion of the ZRP coated steel and the corrosion products.
In the present study, we investigated the protection property of inorganic zinc-rich paint (IZRP) -coated steel. The IZRP layer is penetrated by water and zinc particles in the IZRP layer can induce the cathodic protection. After prolonged exposure, the IZRP layer contains large amount of corrosion products. To clarify the corrosion mechanism of IZRP, it is necessary to examine the corrosion products formed on the entire layer. In this study, the IZRP-coated steel was exposed for five years to an atmosphere near a seashore in Okinawa, Japan, which is considered to be a high-chloride environment. The zinc corrosion products in the IZRP layer were analyzed by X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FT-IR), electron probe microscopic analysis (EPMA) and Raman spectroscopy. From the distributions of Zn, Ca, and Mg compounds of the corrosion products measured by these analyses, the possible corrosion mechanism was discussed.
The chemical composition of the steel substrate used here is shown in Table 1. A steel specimen of 70 mm in width, 150 mm in length and 3 mm in thickness was cut from a plate. The surface of the steel specimen was shot-blasted by a steel sphere and IZRP was air-sprayed onto the steel specimen to achieve a 60 μm thick coat. After curing the sprayed IZRP for one week in an air-conditioned atmosphere of about 25°C temperature and about 60% relative humidity, a modified epoxy-resin coating was sealed at the rearside and corners of the specimen. NIPPE ZINKY 1000QC was used as the IZRP. This paint was a mixture of metal powder and coating solution at a weight ratio of 3:1. The metal powder comprises 90 to 95% zinc particles, 0.1 to 1.0% quartz and 0.1 to 1.0% lead. The composition of the coating solution is 1.6% ethyl benzene, 8.6% xylene, 15 to 20% propylene glycol mono-ethyl ether, 25 to 30% propylene glycol mono-ethyl ether acetate and 15 to 20% 3-methoxy-3-methylbutanol. The cathodic corrosion protection of the IZRP layer formed on steel is caused by zinc particles and water absorbing in the layer.
| C | Si | Mn | P | S | Cr | others |
|---|---|---|---|---|---|---|
| 0.02 | 0.26 | 2.61 | 0.007 | 0.001 | 5.99 | Al, Cu, Ni |
An optical microscopic photograph of a cross section of the original IZRP layer is shown in Fig. 1. The thickness of the IZRP layer was around 60 μm. Zinc particles with a diameter of several microns were distributed uniformly in binder. The average ratio of the cross-sectional area of zinc particles to the entire coating area was about 0.35. From the photograph, it was found that the electric contact among the zinc particles and between the zinc particles and steel substrate was conformed. EPMA elemental mapping of the IZRP layer is presented in Fig. 2. The binder was found to comprise Al, Si, and O and to contain small amounts of Ca and Mg.
Optical microphotograph of a cross section of the IZRP layer.
EPMA elemental mappings of a cross section of the initial IZRP layer.
The steel specimens coated with IZRP were exposed to the seashore in Okinawa from November 2008 to November 2013. The corrosion environment of this site corresponds to ISO 9223-2012 class C5.8) The test specimens were set horizontally and placed in a shelter. The test specimens faced the ground.
2.3. Analysis of the Zinc Corrosion Product LayerThe IZRP layer containing corrosion products on the steel was examined by optical microscopy, XRD, FT-IR, EPMA elemental mapping and Raman scattering spectroscopy.
The spot of the incident X-ray beam (Co-Kα) was 15 mm in width and 5–30 mm in length, which varied with the incidence angle. The scan rate was 1°/min. The IR spectra of the corrosion products on the specimen were measured by FT-IR with the attenuated total reflection (ATR) method using diamond crystal. The incidence angle was 45°, the spectral resolution was 4 cm−1 and the number of cumulative counts was 128.
The cross section of the corrosion product layer on the specimen surface was analyzed by EPMA and Raman spectroscopy as well as by optical microscopy. The exposed specimen was cut into a sample with a width of 20 mm and a length of 8 mm in a dry atmosphere and then mounted in resin. The cross section was ground by emery paper to 1500 grit using alcohol, followed by buff-polishing with alumina. EPMA elemental mappings were obtained for Fe, O, Zn, Cl, Mg, Ca, Si and Al. Na could not be measured, because the characteristic X-ray main spectrum (Kα line) of Na overlapped with Zn (Lβ line). Raman spectra were measured at the outer, middle and inner points in the cross section of the corrosion product. The spot diameter of the excitation laser beam was about 10 μm. The conditions of the Raman spectroscopic measurement were following; excitation laser light at a wavelength of 532 nm with power of about 50 mW and the accumulation with 10 s exposure multiplied by 50 times for the Raman signal detection.
2.4. Cathodic Polarization MeasurementCathodic polarization measurement for the specimens was performed in 0.01 mol L−1 Na2SO4 solution at 30°C before and after the exposure. For the measurement, specimens of 15 mm in width, 30 mm in length and 3 mm in thickness were used. The surface area of the working electrode was constrained to 0.5 cm2 by sealing with a silicon resin. The volume of the test solution was 150 mL. After bubbling with air, cathodic polarization measurement was conducted in stagnant conditions at a sweep rate of 20 mV min−1. The reference electrode was saturated silver-silver chloride (SSE) electrode and the counter electrode was a platinum wire.
Test specimen surface exposed to Okinawa for five years was covered with white rust and no iron rust appeared. The surface view of the exposed steel is shown in Fig. 3. The cross section of the surface rust layer is presented in Fig. 4, wherein the thickness of the corrosion product layer was 150 to 200 μm, which is about three to four times greater than the original 40 to 60 μm thick IZRP layer (see Fig. 1). It should be noted that zinc particles still remained in the entire corrosion products. The zinc particles became smaller and were dispersed quite uniformly. This indicates that anodic dissolution of zinc particles occurred uniformly in the entire IZRP layer.
Photograph of the IZRP-coated specimen after the exposure test.
Optical microphotograph of a cross section of the IZRP rust layer after the exposure test.
EPMA elemental maps of the cross section of corrosion products on the IZRP-coated steel exposed for five years are shown in Fig. 5. No Fe was present in the corrosion products, indicating that the corrosion of steel substrate was inhibited in the IZRP-coated steel. It should be noted that Zn and O spread throughout the corrosion products, whereas Cl is enriched inside and Ca and Mg are enriched outside.
EPMA analysis of a cross section of the IZRP rust layer after the exposure test.
XRD of the corrosion product is presented in Fig. 6. In addition to metallic Zinc, NaZn4(SO4)(OH)6Cl(H2O)6, Zn5(OH)8Cl2H2O and NaCl were identified. The presence of CaCO3 can be suggested as well. In addition, some minor evidence of the existence of CaMg(CO3)2,with one diffraction peak in XRD, was found.
XRD patterns of the IZRP rust layer after the exposure test.
The FT-IR spectra of the corrosion product are presented in Fig. 7. The powders of Mg5(CO3)4(OH)2(H2O)5 and Zn5(CO3)2(OH)6H2O were also measured as references. As shown in Fig. 7, several absorption peaks of the corrosion product corresponded to those of Mg5(CO3)4(OH)2(H2O)5, suggesting that Mg5(CO3)4(OH)2(H2O)5 was also present in the corrosion products of the IZRP layer.
FT-IR of the IZRP rust layer after the exposure test.
To clarify the local change in corrosion products, spot analysis using Raman spectroscopy was conducted. The measurement points by Raman spectroscopy are shown in Fig. 8. The measurements were conducted at three points, the outer, middle and inner area of the corrosion product layer. The results are shown in Fig. 9. For assignment of the Raman spectra, data presented by Ohtsuka and Matsuda9) and C. A. Argullo et al. were used.10) As shown in Fig. 9, the peaks at 265 and 400 cm−1 was assigned to those of Zn5(OH)8Cl2H2O and the peaks at 440 and 570 cm−1 to those of ZnO. From the peak at 1070–1100 cm−1 Zn5(CO3)2(OH)6H2O was possibly formed at the outer and middle area. Zn5(OH)8Cl2H2O and ZnO were present at all measurement points, although the peaks at 400 cm−1 of Zn5(OH)8Cl2H2O and at 440 cm−1 of ZnO were overlapped and were not clearly separated. It should be noted that the Raman peaks of ZnO changed in position and intensity with crystal orientation and crystallinity.9,10) The results of the corrosion products measured are summarized in Table 2.
Measured point circle of Raman scattering spectroscopy in the zinc corrosion product layer.
Raman scattering spectra in the zinc corrosion product layer at three points.
| 1) Analysis from the surface | ||
|---|---|---|
| Appearance | XRD | FT-IR |
| •Zinc corrosion products (No iron rust) | •NaZn4(SO4)(OH)6Cl(H2O)6 •Zn5(OH)8Cl2H2O •CaCO3 •CaMg(CO3)2 | •Mg5(CO3)4(OH)2(H2O)5 |
| 2) Cross-section analysis | ||
|---|---|---|
| Optical microphotograph | EPMA | Raman Scattering Spectroscopy |
| •Zinc corrosion product layer 150–200 μm •Zinc particles reside in the entire corrosion product | •Outer: Zn,O,Mg,Ca,Cl •Middle - Inner: Zn,O,Cl | •Outer: Zn5(OH)8Cl2H2O,ZnO,Zn5(CO3)2(OH)6H2O •Middle: Zn5(OH)8Cl2H2O,ZnO,Zn5(CO3)2(OH)6H2O •Inner: Zn5(OH)8Cl2H2O,ZnO |
To compare the identification results from XRD and IR by which the rust layer was analyzed from the surface, we must consider the detection depth from the surface. X-ray can penetrate to several 10 μm distance from the surface. Because no diffraction peaks of substrate steel appear in XRD as shown in Fig. 6, X-ray is not assumed to reach the inner area of the rust layer and thus the XRD data may represent the composition of the rust to middle area. The detection depth of IR-ATR is limited in the area in which the evanescent wave reaches and is known at about 0.5 to 2 μm. The IR-ATR results indicate a composition in the surface region of the rust.
Then we discuss the formation process of the zinc corrosion products of the IZRP layer. From Table 2, NaZn4(SO4)(OH)6Cl(H2O)6, ZnO, Zn5(OH)8Cl2H2O and Zn5(CO3)2(OH)6H2O were identified as corrosion products. Leygraf et al. presented the long-term transient change in rust compounds formed on metallic zinc exposed to a marine environment containing sulfur dioxide.11) They showed that the initial rust comprises Zn(OH)2 that then changes to ZnO, Zn5(CO3)2(OH)6H2O and/or Zn5(OH)8Cl2H2O and finally to NaZn4(SO4)(OH)6Cl(H2O)6. The corrosion products of the IZRP-coated steel possibly undergo the same changes as the corrosion products. NaZn4(SO4)(OH)6Cl(H2O)6 was identified by XRD (Fig. 6), and Zn5(CO3)2(OH)6H2O was found by IR and Raman spectroscopy at the outer area of the corrosion product layer (Figs. 7 and 9). The presence of both compounds indicates that the corrosion process of the IZRP layer is influenced by carbon dioxide and sulfur dioxide dissolved from the atmosphere. Mg and Ca compounds at the outer area of the IZRP layer were identified as Mg5(CO3)4(OH)2(H2O)5, CaCO3 and CaMg(CO3)2 by XRD and FT-IR, as shown in Figs. 6 and 7. Mg and Ca are supposed to be originated from sea salt particles. Matsumoto et al. calculated the stable region of the above compounds as a function of solution pH. They reported that CaCO3 and Mg (OH)2 tend to form in the alkaline region.12) It was assumed that oxygen reduction reaction occurred at the outer side of the IZRP layer, generating an alkaline environment, and Mg and Ca compounds were formed.
3.3. Difference in Cathodic CD between the Specimens before and after ExposureCathodic polarizations in Na2SO4 solution were performed for the IZRP-coated steel specimen before and after exposure. The results are presented in Fig. 10. Both specimens showed a corrosion potential of approximately −1000 mV (vs. SSE), corresponding to the corrosion potential of metallic zinc. This potential demonstrates that zinc cathodic protection is still working even after five years of exposure. The cathodic current density (CD) of the specimen decreased after the exposure as compared with that before exposure. Comparing the CD at −1100 mV, the CD of the exposed specimen was about one-seventh smaller than that of specimen before exposure. It has been reported that the corrosion protection of galvanized steel is owing to both the cathodic protection and the barrier effect of the corrosion products formed.13) Matsumoto et al. suggested that the cathodic CD on steel was smaller in MgCl2 solution as compared with NaCl solution.14) They assumed that the decrease of cathodic CD in MgCl2 solution is due to an insulating product of Mg(OH)2 covering the bare steel surface. We assume that the decrease in cathodic CD is due to the increase in thickness of corrosion products, the formations of zinc corrosion products and Mg and Ca compounds (CaMg(CO3)2, CaCO3, and Mg5(CO3)4(OH)2(H2O)5) on the zinc particles. These Mg and Ca compounds are insulating materials covering the surface of zinc particles. This suggests that the corrosion products formed from the corrosion reaction of the IZRP layer restrains the cathodic reaction.
Cathodic polarization of the specimen before and after exposure (pH 6.0, 30°C, 0.01 mol L−1 Na2SO4, air. sat 20 mV/min).
The corrosion process was characterized by the following three features obtained from the analysis of corrosion products of IZRP-coated steel. First, zinc particles became smaller in diameter, independent of the distance from the steel, as shown in Fig. 4. It is assumed that anodic dissolution uniformly occurs on the zinc particles in the IZRP layer. Second, Mg and Ca compounds were observed in the outer area of the corrosion product layer, as shown in Fig. 5. CaMg(CO3)2, CaCO3 and Mg5(CO3)4(OH)2(H2O)5 were observed in the outer area of the corrosion product layer. These compounds precipitate under alkaline conditions, suggesting that the cathodic reaction of oxygen reduction occurs on the zinc particle surface in the outer area of the IZRP layer. Thirdly, Zn5(OH)8Cl2H2O and ZnO were present in all corrosion product layers, as shown in Fig. 9. These zinc compounds also precipitated in an alkaline environment.15) These compounds are considered to be formed as follows. The anodic reaction of zinc dissolution takes place on the zinc particles of the entire IZRP layer due to electric contact among the zinc particles and between the particles and the steel substrate. The cathodic current of oxygen reduction can also take place on the zinc particle surface and the steel substrate. We can assume that the oxygen concentration decreases with increasing distance from the surface of the IZRP layer. Therefore, the cathodic reaction of oxygen reduction occurs on the surface of zinc particles at the initial stage. After the long exposure, the precipitate on the zinc particles decreases of oxygen reduction rate and thus oxygen reduction is capable of occurring on the steel substrate as well as on the zinc particles. Oxygen reduction produces alkaline environment in whole layer and Zn5(OH)8Cl2H2O and ZnO were possibly deposited even in the inner area of the IZRP layer.
From the above discussion, the corrosion process of the IZRP-coated steel exposed to a high-chloride environment is described in Fig. 11. In Fig. 11(a), anodic dissolution of zinc initially takes place on the zinc particles throughout the IZRP layer. The Cathodic current of oxygen reduction occurs on the zinc particles present in the outer area of the IZRP layer. Along with zinc corrosion products, Mg and Ca compounds such as Mg5(CO3)4(OH)2(H2O)5, CaMg(CO3)2 and CaCO3 are then formed on the zinc particles, inhibiting cathodic oxygen reduction. This inhibition allows the dissolved oxygen to diffuse to the steel substrates and cathodic reaction begins on the steel surface, as shown in Fig. 11(b). The cathodic reduction of oxygen taking place on both the zinc particles and the steel substrate causes an alkaline condition in the entire coating layer. Owing to the alkaline condition, Zn (OH)2 and Zn5(OH)8Cl2H2O are deposited on the steel substrate. The cathodic oxygen reduction on the steel surface is possibly decreased by the deposits, resulting in prolonged lifetime of the IZRP system in a corrosive environment. When metallic zinc particles are lost after a long period, the steel substrate may corrode, as shown in Fig. 11(c).
Estimation of the corrosion process of IZRP-coated steel.
The corrosion protection behavior of an IZRP layer on steel was studied after exposure for five years in Okinawa, whose environment is a relatively high-chloride atmosphere. To clarify the corrosion mechanism of IZRP, we examined the corrosion products locally at inner, middle and outer area of the IZRP layer.
(1) Zinc particles were uniformly dissolved in the IZRP-coated steel, independent of the distance from the IZRP surface.
(2) NaZn4(SO4)(OH)6Cl(H2O)6 was identified in the corrosion product layer. Zn5(CO3)2(OH)6H2O was found in the outer and middle area of the corrosion product layer. Zn5(OH)8Cl2H2O and ZnO were present throughout the layer.
(3) Mg and Ca were observed in the outer area. The compounds of Mg and Ca were estimated to be CaCO3, CaMg(CO3)2 and Mg5(CO3)4(OH)2(H2O)5.
(4) We offered a corrosion process of IZRP-coated steel from the distribution of the corrosion products.
