This study aims to investigate the role of aluminium (Al) in marine environment and the corrosion mechanism of galvalume coatings by conducting accelerated experiments and data analysis.
Samples were subjected to accelerated corrosion for 136 days via salt spray tests to simulate the natural conditions of marine environment and consequently accelerate the experiments. Subsequently, the samples were examined using various test methods, such as EDS, scanning electron microscopy (SEM), X-ray diffraction (XRD) and electrochemical impedance spectroscopy (EIS), and the obtained data were analysed.
Galvalume coatings comprised interdigitated zinc (Zn)-rich and dendritic Al-rich phases. Corrosion was observed to begin with a Zn-rich phase. The primary components of the corrosion product film were Al2O3 and Zn5(OH)8Cl2·H2O. It was confirmed that the role of Al was to form a dense protective film, thereby successfully blocking the entry of corrosive media and protecting the iron substrate.
This study provides a clearer understanding of the corrosion mechanism and kinetics of galvalume coatings in a simulated marine environment. In addition, the role of Al, which is rarely mentioned in the literature, was investigated.
1. Introduction
The role of zinc (Zn) in different coatings has been extensively studied (Kalendová, 2002). The principle of Zn protection for steel substrates is implemented through the use of sacrificial anodes (Marder, 2000). Alloys made of Zn and other metals have been used as coatings and have demonstrated excellent properties (Kwiatkowski et al., 1996). Aluminium (Al) alloys exhibit great corrosion resistance, and in recent years, research on Al alloys has resulted in new breakthroughs in its usage in different environments, such as seawater and air (Deng et al., 2023; Ji et al., 2023; Wang et al., 2024).
Zn-Al-based coating systems have shown great potential. Moreover, more complex alloy coatings, such as Zn-Al and Zn-Al-Mg coatings, have also been obtained (Sarkar et al., 2018; Stein et al., 2022; Pradhan et al., 2022; Wang et al., 2023a, 2023b). Another newly discovered coating is the galvalume coating. This is a type of steel that is coated on both sides of a surface with an Al-Zn alloy layer. Galvalume coatings exhibit unique dual-phase dendritic crystallisation that combines the surface protection and high durability of Al with the electrochemical protection properties of Zn (Moreira et al., 2006; Jiang et al., 2014). Typically, it is two to four times more durable and resistant to corrosion under atmospheric conditions than galvanised coatings (Vishwanatha and Panda, 2022; Hoang et al., 2022; Morcillo et al., 1994; Yasakau, et al., 2016). Galvalume coatings have several applications such as household appliances and automobile exhaust systems (Guo, et al., 2018; Jarwali and Nakamura, 2016).
Palma et al. (1998) demonstrated that coatings were subjected to localised erosion that started in the interdigitated areas of the material. Qiu et al. (2012) showed that a slightly polished galvalume surface exhibited dendritic Al-rich areas with a higher Volta potential than dendritic Zn-rich areas. Experiments have been conducted to compare the performance of Al-Zn coatings with different processes (dos Santos et al., 2015; Edavan and Kopinski, 2009; Alvarenga and Lins, 2016). The corrosion resistance of three-component alloy coatings has also been investigated (Lee et al., 2019; Hu et al., 2023). Zhang et al. (2018) demonstrated that the behaviour of the galvalume was similar as a bare Al sheet. Xiao et al. (2023) showed that galvalume and its corrosion products both affected substrate protection.
Despite significant research on galvalume coatings, there are currently few indoor accelerated salt-spray studies on galvalume coating materials in simulated marine environments. In addition, there is no clear explanation regarding the role of Al in galvalume coatings. This study investigated the role of Al and its corrosion mechanism in galvalume coatings in marine environment.
2. Experiments
2.1 Materials
S550GD+AZ was chosen as the substrate material, with the galvalume coating wrapped on the surface. The specific ingredients of the substrate materials are listed in Table 1. The coating was composed of 55% Al, 43.4% Zn and 1.6% silicon (Si) by weight (Qiu et al., 2012). The sample thickness was 1 mm, and the coating weight was 150 g/m2. The sample dimensions were 150 mm × 75 mm. The density of galvalume coating was 3.8 g/cm3. The coating thickness was in the range of 20–25 µm.
2.2 Experimental methodologies
Salt-spray acceleration test was conducted referring the standard Corrosion of metals and alloys–Accelerated cyclic tests with exposure to acidified salt spray, “dry” and “wet” conditions (ISO 16151:2005), as presented in Table 2.
Salt-spray test procedure and related parameters
| No. | Experiment process | Temperature (°C) | pH | Solution composition | Ambient relative humidity (%) | Duration (h) |
|---|---|---|---|---|---|---|
| 1 | Acid salt-spray condition | 35 ± 1 | 3.5 ± 0.1 | (50 ± 5) g/L NaCl | – | 2 |
| 2 | “Dry” condition | (60 ± 1) | – | – | <30 | 4 |
| 3 | “Wet” condition | (50 ± 1) | – | – | >95 | 2 |
| No. | Experiment process | Temperature (°C) | pH | Solution composition | Ambient relative humidity (%) | Duration (h) |
|---|---|---|---|---|---|---|
| 1 | Acid salt-spray condition | 35 ± 1 | 3.5 ± 0.1 | (50 ± 5) g/L NaCl | – | 2 |
| 2 | “Dry” condition | (60 ± 1) | – | – | <30 | 4 |
| 3 | “Wet” condition | (50 ± 1) | – | – | >95 | 2 |
Source:
The change time between different environments should be less than 30 min. Further, the switching time between dry and wet conditions should not exceed 15 min. Each cycle was 8-h long. After corrosion, the rust was removed.
2.3 Morphological observations
For macroscopic profiling, a Nikon digital camera was used to observe the macroscopic morphology after 0, 24, 56, 104 and 136 days.
To obtain micromorphological observations, an FEI Quanta250 environmental scanning electron microscope (SEM) was used to observe the surface and cross-sectional morphologies.
2.4 Observation of corrosion products
An FEI Quanta250 environmental SEM was used to observe the surface and cross-sectional morphologies of the corroded samples. To characterise the cross-section morphology, the corroded sample was encapsulated in an epoxy resin, then polished to 2000# using metallographic sandpaper, and finally polished using diamond polishing paste with particle size of 1.5 μm until the surface was non-scratched.
2.5 Characterisation of corrosion products
X-ray diffraction (XRD) analysis was conducted using A Brooke D8 ADVANCE rotating anode X-ray diffractometer to analyse the corrosion products generated on the surface of the block samples. (Wang et al., 2023a, 2023b). The test conditions were as follows: the Kα of copper (Cu) target was used as the radiation source, the tube voltage was 40 kV, the tube current was 150 mA, the scanning range was 2θ = 10°–90°, the step width was 0.02° and the scanning rate was 10°/min.
2.6 Electrochemical characterisation
A VersaSTAT3F electrochemical workstation with a three-electrode system was used to conduct the electrochemical tests. The reference electrode was a saturated calomel electrode, and the counter electrode was a Pt electrode. The exposed area was 0.785 cm2.
Electrochemical impedance spectroscopy (EIS) and linear polarisation curves are sequentially tested after the open circuit potential (OCP) EIS was stabilized for 30 min. The scanning frequency range of EIS was 105–10−2 Hz, and the AC excitation signal amplitude was 10 mV. The ZSimpWin software was used for impedance data processing to analyse the structure of the equivalent circuit and the parameters of each component. The scanning rate measured by the polarization curve was 0.5 mV/s and the scanning interval is −0.5 to +1 V (relative to OCP). According to the polarisation curve, corrosion potential Ecorr, corrosion current density icorr, anode Tafel slope βa and cathode Tafel slope βc of the sample were obtained by using the Tafel extrapolation method (Mansfeld, 2005).
The aforementioned electrochemical tests were conducted at room temperature (25°C), and each set of tests was repeated at least three times. The concentration of the solution was (50 ± 5) g/L by mass of NaCl.
3. Results and analysis
3.1 Corrosion kinetics
After 136 days of testing, the galvalume coating thickness was reduced to 5.620 μm, which was considerably less than the coating thickness of 20–25 μm, as presented in Table 3.
Corrosion weight loss data for galvalume coatings under simulated marine atmosphere test
| Time (days) | 24 | 56 | 104 | 136 |
|---|---|---|---|---|
| Weight loss per unit area (g/m2) | 11.142 | 14.224 | 20.108 | 21.356 |
| Average corrosion depth (μm) | 2.932 | 3.743 | 5.292 | 5.620 |
| Average corrosion rate (μm/a) | 44.595 | 27.800 | 18.571 | 15.094 |
| Time (days) | 24 | 56 | 104 | 136 |
|---|---|---|---|---|
| Weight loss per unit area (g/m2) | 11.142 | 14.224 | 20.108 | 21.356 |
| Average corrosion depth (μm) | 2.932 | 3.743 | 5.292 | 5.620 |
| Average corrosion rate (μm/a) | 44.595 | 27.800 | 18.571 | 15.094 |
Source:
According to Figure 1, the corrosion rate tended to decrease with increasing test time. The average corrosion rate at the end of the test was lower than that at the beginning of the test owing to the formation of a dense layer of corrosion products. The correlation coefficient for the fitted curve was R2 = 0.978. This value was greater than 0.9, indicating that the corrosion kinetics of the layer could be well fitted to the model in the equation. The value of n = 0.4043 < 1 confirmed that the corrosion of galvalume coating in such an environment followed a deceleration process.
Corrosion rain dynamics correlation curve of galvalume coating in a simulated marine environment
Corrosion rain dynamics correlation curve of galvalume coating in a simulated marine environment
3.2 Corrosion topography
Figure 2 shows the macroscopic corrosion profile in a simulated marine atmosphere. For the non-scratched coating, after 24 days, the surface of the sample was intact, with scattered Zn blobs retained on the surface of the plating and the metallic lustre appearing as greyish white. After 56 days, the metal surface dimmed in lustre, the Zn blobs were smaller and the colour whitened. This was attributed to the generation of a thin-film-like oxide film on the surface of the metal in a corrosive environment for a long period of time to prevent corrosive ions from penetrating and thus playing a protective role. When the experiment reached 104–136 days, the surface of the sample lost its metallic lustre completely, the colour turned white and the Zn bloom gradually disappeared. However, the surface of the sample did not exhibit red rust. For the scratched coating, after 24 days, there was no significant change in the width of the scratch; however, a small amount of black and white rust was visible internally, and the gloss of the sample was reduced. After 56 days, there was a slight increase in the width of the scratches and an increase in the corrosion products, but there was still no red rusting. This indicates that the corrosion products around the scratch were well insulated from the corrosive ions. For 104–136 days, the change in the corrosion width remained insignificant, and there was no red rust at the scratches.
Macroscopic corrosion profiles of galvalume coatings after different cycles of accelerated corrosion tests in a simulated marine atmospheres
Macroscopic corrosion profiles of galvalume coatings after different cycles of accelerated corrosion tests in a simulated marine atmospheres
The initial cross-sectional analysis before the test and the results of the cross-sectional analysis after the different corrosion test periods are shown in Figures 3 and 4, respectively. The galvalume coating comprised two layers. The outer layer was an Al-Zn alloy layer. The oxygen (O) and chlorine (Cl) elemental surface distributions indicated that both the outer and inner layers of the plate were corroded.
Backscattered profile for surface/section and surface scans of galvalume coating sections before corrosion experiment
Backscattered profile for surface/section and surface scans of galvalume coating sections before corrosion experiment
Backscattered profiles and surface scans of galvalume coating sections after different test cycles in a simulated marine atmosphere
Backscattered profiles and surface scans of galvalume coating sections after different test cycles in a simulated marine atmosphere
3.3 Corrosion products
Figure 5 shows the XRD patterns of the corrosion products of the galvalume coatings. The corrosion products on the surface of the plating were mainly ZnO, Al2O3, Zn5(OH)6(CO3)2 and Zn5(OH)8Cl2·H2O. The ZnO flakes easily form a powder; therefore, they did not form a protective film, whereas Al2O3 formed a dense oxide film with a protective effect.
XRD of galvalume coatings after different periods of accelerated corrosion test in a simulated marine environment
XRD of galvalume coatings after different periods of accelerated corrosion test in a simulated marine environment
3.4 Electrochemical regularities
Figure 6(a) shows the Nyquist plots. During the corrosion period, two constant values were observed at low, medium and high frequencies. The first time constant at high frequencies may be related to the corrosion–product layer of galvalume coating. The second time constant at low and medium frequencies may be related to the charge–transfer behaviour of the electrode interface. An equivalent circuit was selected to fit the electrochemical impedance spectrum, and the fitting results are presented in Table 4. Rs is the electrolyte resistance, CPEr and Rr are used to explain the corrosion product layers and CPEdl and Rct are used to describe the charge transfer processes at the interfaces.
EIS of galvalume coatings after different test cycles and the equivalent circuit of galvalume coatings in acidic (50 ± 5) g/L NaCl solution
EIS of galvalume coatings after different test cycles and the equivalent circuit of galvalume coatings in acidic (50 ± 5) g/L NaCl solution
Equivalent circuit element fitting values for galvalume coatings in acidic (50 ± 5) g/L NaCl solution
| Time (days) | Rs/(Ω·cm2) | Q1/(F·cm−2) | n1 | Rr/(Ω·cm2) | Q2/(F·cm−2) | n2 | Rct/(Ω·cm2) | χ2 |
|---|---|---|---|---|---|---|---|---|
| 24 | 19.9 | 4.6 × 10–7 | 0.7 | 1.7 × 104 | 8.9 × 10–7 | 0.5 | 1.2 × 105 | 3.2 × 10–4 |
| 56 | 8.7 × 10–3 | 1.8 × 10–6 | 0.6 | 6.4 × 103 | 1.2 × 10–8 | 1.0 | 6.3 × 104 | 1.6 × 10–3 |
| 120 | 2.4 × 10–5 | 1.3 × 10–6 | 0.6 | 2.7 × 104 | 4.2 × 10–6 | 0.7 | 2.9 × 105 | 1.6 × 10–3 |
| 136 | 9 × 10–6 | 3.0 × 10–7 | 0.7 | 1.4 × 104 | 2.9 × 10–7 | 0.5 | 2.3 × 105 | 1.6 × 10–4 |
| Time (days) | Rs/(Ω·cm2) | Q1/(F·cm−2) | n1 | Rr/(Ω·cm2) | Q2/(F·cm−2) | n2 | Rct/(Ω·cm2) | χ2 |
|---|---|---|---|---|---|---|---|---|
| 24 | 19.9 | 4.6 × 10–7 | 0.7 | 1.7 × 104 | 8.9 × 10–7 | 0.5 | 1.2 × 105 | 3.2 × 10–4 |
| 56 | 8.7 × 10–3 | 1.8 × 10–6 | 0.6 | 6.4 × 103 | 1.2 × 10–8 | 1.0 | 6.3 × 104 | 1.6 × 10–3 |
| 120 | 2.4 × 10–5 | 1.3 × 10–6 | 0.6 | 2.7 × 104 | 4.2 × 10–6 | 0.7 | 2.9 × 105 | 1.6 × 10–3 |
| 136 | 9 × 10–6 | 3.0 × 10–7 | 0.7 | 1.4 × 104 | 2.9 × 10–7 | 0.5 | 2.3 × 105 | 1.6 × 10–4 |
Source:
The experimentally drawn dephased diagrams are shown in Figure 6(b) and (c). There was no significant change in the impedance modulus or phase angle, and no obvious sine-function features were observed. On Day 136, no significant phase angle peaks were observed.
A high impedance was obtained because of the presence of a passivated layer on the surface of the plating before the experiment. However, as the experiment progressed, the impedance decreased significantly. With the generation of the insoluble corrosion products (Zn5(OH)8Cl2·H2O), the resistance of the corrosion product layer became larger and the interfacial reaction rate decreased.
The kinetic potential polarisation curves are shown in Figure 7. The cathodic and anodic polarisation behaviours of the galvalume-coated samples were relatively similar across cycles. The cathode was mainly a hydrogen precipitation reaction, and there was also a reduction reaction of oxygen, whereas the anode experienced an active dissolution process of the Zn layer. With an increase in test time, the corrosion voltage first increased and then decreased, and finally increased again. This indicated that the corrosion rate increased and decreased.
Kinetic potential polarization curves of galvalume coatings in acidic (50 ± 5) g/L NaCl solution after different test cycles
Kinetic potential polarization curves of galvalume coatings in acidic (50 ± 5) g/L NaCl solution after different test cycles
The results of the polarisation curve fitting are presented in Table 5. The corrosion voltage first increased, then decreased and finally increased again, indicating a change in the corrosion rate.
4. Discussion
4.1 Evolution of the corrosion products layer
Figure 8 is a schematic of the corrosion mechanism. The specific corrosion steps are as follows:
Schematic of the mechanism of corrosion product formation of the Zn-Al coating simulating the marine atmosphere
Schematic of the mechanism of corrosion product formation of the Zn-Al coating simulating the marine atmosphere
First, Al lost electrons for the oxygen and water in the environment:
The Al3+ then combined with the OH− to produce Al(OH)3:
Eventually, in a dry environment, the AI(OH)3 was dehydrated to Al2O3:
Meanwhile, some of the AI(OH)3 can undergo transformation when exposed to marine environment:
The Zn-rich phase was first reformed into an ionic state by an anodic reaction and the cathodic region by both oxygen and hydrogen reduction reactions:
Zn2+ generated in the anode region reacted with OH− in the cathode region to form Zn(OH)2:
Zn(OH)2 was susceptible to dehydration to ZnO under dry conditions:
When CO32− and Cl− are present, Zn5(OH)8Cl2·H2O and Zn5(OH)6(CO3)2 are produced:
4.2 Role of adding Al to plating in corrosion processes
The corrosion rate is related to corrosion electrochemistry. As shown in Tables 3 and 4, at the beginning, Al had not yet been oxidized to Al2O3, and thus, the reaction occurred rapidly. The rate continued to decrease owing to the presence of Al2O3, which isolated the substrate from the external corrosive environment.
When the coating did not contain Al, the corrosion products would fall off within a relatively short period (Xiao et al., 2023).
According to the backscattered morphology shown in Figure 4, there was no significant loss of Zn or Al after 104 days. It was not until Day 136 that a slight thinning of Al and Zn was observed. It was observed that the iron matrix in the inner layer did not suffer much damage.
According to Table 2, the resistance of the corrosion–product layer occurred as a large floating: rising–falling–rising–falling. First, Al2O3 was generated on the surface, which protected the interior from corrosion. As the time progressed, the resistance decreased. This is because the chloride ions damaged the passivation film, the oxidation of Zn was not prevented and the impedance decreased. As Al2O3 was still retained, this phenomenon was not obvious. Subsequently, the resistance increased because insoluble corrosion products continued to be generated, and the protective film reappeared. The thicknesses of Al and Zn finally decreased, and the resistance declined because the surface was completely formed. Finally, the oxide layer was completely peeled off. This implied that the new uncorroded Zn and Al layers became the surfaces. This process was repeated.
According to the oxygen distribution shown in Figure 4, Zn reacted earlier. This was owing to the dense structure of Al2O3. The dense oxide film generated by the reaction of Al can effectively prevent the flow of corrosion products of Zn, when Zn5(OH)8Cl2·H2O are not easy to peel off owing to their large volume. Moreover, the Al2O3 network + large volume of Zn can effectively block the contact between the corrosion medium and the iron substrate, which renders the iron less susceptible to corrosion and well protected.
5. Conclusions
In the marine atmosphere, the corrosion product film was mainly Zn5(OH)8Cl2·H2O + Al2O3. The Al2O3 oxide layer exhibited good protective properties for the plating.
The addition of Al was meant to protect the corrosion products of Zn such that the iron substrate could be well protected. Thus, the corrosion products produced by the Al-rich phase can play a role in fixing loose Zn corrosion products such that the corrosion products of Zn can hinder the diffusion of the media.
This work was supported by the National Corrosion and Protection Data Center.
