Effect of Al Content on the Long-Term Corrosion Behavior of Arc-Sprayed ZnAl Alloy Coatings

: The corrosion of steel structures in aggressive marine environments is a vital issue that induces signiﬁcant degradation of their performance and lifespan. Herein, three arc-sprayed ZnAl coatings with varied Al contents of 0 wt.%, 15 wt.%, and 50 wt.% were deposited onto a hull steel substrate. The effect of Al content on the long-term corrosion protection performance of ZnAl coatings left in a chloride-containing solution for 840 h was systematically investigated. The evolutions of open-circuit potential, polarization curves, and electrochemical impedance spectroscopy of different ZnAl coatings during the long-term immersion test were examined. The morphologies and phase constitutions of the corrosion products were characterized. The results indicated that the corrosion rate of ZnAl coatings decreased as the Al content increased, and the ZnAl50 coating exhibited the most superior long-term corrosion protection performance. Moreover, for the three ZnAl coatings with an Al contents varying from 0 to 50%, their corrosion rate increased with immersion time in the initial 360 h due to the formation of the unstable and porous corrosion product ZnO; after 360 h immersion, their corrosion rate decreased with the prolonging of immersion time. This was revealed to be related to the formation of different corrosion products. ZnO and stable Al 2 O 3 were the main corrosion products for the pure Zn coating and ZnAl15 coating, respectively. Al 2 O 3 and powerful layered double hydroxide Zn 6 Al 2 (OH) 16 CO 3 · 4H 2 O were found to be the dominant corrosion products of the ZnAl50 coating, which was responsible for its remarkable long-term corrosion protection performance.


Introduction
Steel is one of the most widely employed structural materials in marine environment due to its various advantages including excellent mechanical and welding properties, cost-effectiveness, and so on.The corrosion of steel equipment and structures in marine environment is a major issue that urgently requires solving, deteriorating their service performance and lifespan significantly [1].That is, the exposure of steel structures in aggressive chloride-containing environments always leads to their severe corrosion and thus the degradation of their properties.Therefore, the development of anti-corrosion measures is of significant importance.
Numerous efforts have been made to improve the anti-corrosion performance of steel.The deposition of protective coatings on steel structures has been proved to be an effective and economical method to inhibit or mitigate their corrosion.Among them, sacrificial tion of simonkolleite, hydrozincite, and Zn-Al layered double hydroxide, which could effectively seal the defects in the coating and resist the ingress of the solution toward the substrate.Liu et al. [22] studied the corrosion mechanism of arc-sprayed Zn and ZnAl alloy coatings systematically.They concluded that the ZnAl alloy coating exhibited better corrosion protection capability than the Zn coating owing to the remarkable self-sealing effect of the corrosion products of the ZnAl coating.
The composition of the ZnAl alloy coating, i.e., the Al content, is a key factor significantly influencing the morphology and constitutions of the corrosion products, thus determining the corrosion resistance of the coating.Nevertheless, few studies have aimed to investigate the influence of Al content on the long-term corrosion behavior of thermalsprayed ZnAl alloy coatings systematically.Moreover, currents studies mainly focused on the corrosion behavior of ZnAl alloy coatings with relatively low Al content, e.g., less than 15 wt.%.In general, ZnAl alloys with higher Al content are difficult to obtain due to the formation of brittle phases during the alloying process of ZnAl alloys with high Al composition [23].Zhu et al. [24] deposited three types of arc-sprayed ZnAl alloy coatings with Al contents of 15.9, 21.1, and 30.1 wt.% to investigate the effect of Al content on their electrochemical behavior.Results revealed that the corrosion resistance of the ZnAl alloy coatings improved with the increase in Al content.However, the underlying mechanism was not demonstrated.Hu et al. [25] compared the corrosion protection performance of new, patented arc-sprayed ZnAl alloy coatings with Al fractions of 15, 30, and 50 wt.%.They concluded that the ZnAl alloy coating with 30% Al composition demonstrated the lowest corrosion rate owing to the effective protection of its compact and stable corrosion products.Nonetheless, the investigation was not that systematic and the results require further confirmation.
Therefore, in this study, arc-sprayed ZnAl coatings with Al contents of 0, 15 wt.%, and 50 wt.%were deposited onto a hull steel substrate.The effect of Al content on the microstructure, long-term corrosion performance, and the correlating corrosion mechanism of ZnAl coatings in a 3.5% NaCl aqueous solution were systematically investigated.The corrosion protection performances of different types of coatings were examined using electrochemical methods.The morphologies and chemical compositions of the corrosion products were determined using SEM and EDS.The XRD evolution of corrosion products during long-term immersion test was examined to clarify their corrosion mechanisms.

Coating Preparation
A commercial high-speed arc spray system (CMD-AS3000, Beijing Xindi Surface Technology Equipment Co., Ltd., Beijing, China) was used to deposit ZnAl coatings with Al compositions of 0, 15 wt.%, and 50 wt.%,which were hereinafter defined as pure Zn, ZnAl15, and ZnAl50 coatings, respectively.Pure Zn wires (purity: 99.9%, diameter: Φ2.8 mm), ZnAl15 alloy wires (chemical compositions: Al: 14.5 wt.%, Zn: 85.5 wt.%, diameter: Φ2.8 mm), and ZnAl50 alloy wires (chemical compositions: Al: 49.5 wt.%, Zn: 50.5 wt.%, diameter: Φ2.8 mm) were utilized as feedstock materials to fabricate the Zn coating, ZnAl15 coating, and ZnAl50 coating, respectively.The Zn and ZnAl alloy wires were provided by Jiangsu Linlong New Materials Co., Ltd., Wuxi, China.A typical hull steel, 921A, with dimensions of 20 mm × 20 mm × 2 mm was used as the substrate.The chemical composition of the 921A substrate is listed in Table 1.Before coating deposition, the steel substrates were firstly ultrasonically cleaned in acetone, followed by grit blasting.The arc-sprayed Zn, ZnAl15, and ZnAl50 coatings with thicknesses of ~300 µm were obtained a spraying voltage and current of 26 V and 180 A, respectively.The compressed air pressure and spraying distance were 0.55 MPa and 200 mm, respectively.

Coating Characterization
Scanning electron microscopy (SEM, Zeiss Auriga, Oberkochen, Germany) was used to characterize the cross-sectional microstructure of the arc-sprayed ZnAl coatings and the surface morphologies of the corrosion products.The chemical compositions of the sprayed coatings and the coatings after the immersion test were analyzed using an Energy Dispersive Spectrometer (EDS) equipped on the SEM.The phase constitutions of the arcsprayed ZnAl coatings with varied Al contents and their corrosion products were examined via X-ray diffractometer (XRD, Bruker D8 advance, Hanau, Germany) with a Cu kα X-ray.The XRD was performed at a scanning angle ranging from 10 • to 90 • with a scanning speed of 10 • /min and scanning step of 0.02 • per step.Image analysis method was conducted on the cross-sectional images of arc-sprayed ZnAl coatings to measure their apparent porosities.Ten images were analyzed for each coating to obtain the average porosity value.

Long-Term Soaking Test
A long-term soaking test was performed for the arc-sprayed Zn, ZnAl15, and ZnAl50 coatings to investigate the effect of Al content on the long-term corrosion protection behavior of ZnAl alloy coatings.The long-term soaking test was conducted in a 3.5% NaCl aqueous solution at ambient temperature.After immersion for 120 h, 360 h, 600 h, and 840 h, coating samples were taken out from the soaking solution and used for the following electrochemical test, SEM characterization, and XRD testing to reveal the influence of the Al content on the corrosion mechanism of ZnAl coatings.Prior to the soaking test, the surface of the coatings was ground and polished firstly, followed by sealing their back and side faces with epoxy resin, exposing the coating surface only.The soaking solution was changed every 2 to 3 days.

Electrochemical Test
The electrochemical test was conducted on an electrochemical workstation (CHI760, CH Instruments Co., Ltd., Shanghai, China) to reveal the corrosion behavior of ZnAl coatings with different Al contents.A three-electrode system was utilized for the electrochemical test.The working electrode, reference electrode, and counter electrode were the coating sample, a saturated calomel electrode (SCE), and a platinum plate, respectively.The exposed surface area of the coating sample used for the electrochemical test was 1 cm 2 .The open circuit potential (OCP), potentiodynamic polarization curves, and electrochemical impedance spectroscopy (EIS) were measured in sequence.Firstly, OCP versus time was measured.Then, the potentiodynamic polarization curve was tested, which started from −0.5 V, corresponding to the OCP, and terminated at a potential of +0.6 V relative to the OCP value with a potential scanning rate of 0.5 mV/s.At last, EIS was tested at the corrosion potential value with an amplitude of 10 mV.The scanning frequency of EIS ranged from 0.01 Hz to 100 kHz, and Zsimpwin software v 3.60 was used to fit the EIS data.Three samples were examined for each coating to obtain the average values.

Coating Microstructure
Figure 1 demonstrates the cross-sectional SEM images of the arc-sprayed Zn, ZnAl15, and ZnAl50 coatings.It is observed that all the three types of sprayed coatings exhibited dense microstructures with good bonding and good adherence to the steel substrate.Furthermore, the three coatings presented uniform lamellar structures, which is a typical characteristic of thermal-sprayed coatings [26,27].Some small black inclusions were found to be distributed within the coatings, which were verified to be oxides formed during the Coatings 2023, 13, 1720 5 of 17 spraying process via EDS.Moreover, different contrast regions were observed within the ZnAl50 coatings.The regions with bright-white and dark-grey contrasts were confirmed to be a Zn-rich phase and Al-rich phase, respectively, using EDS analysis, whereas no distinct contrast difference was found within the ZnAl15 coating.This will be explained in detail in the following.

Coating Microstructure
Figure 1 demonstrates the cross-sectional SEM images of the arc-sprayed Zn, ZnAl15, and ZnAl50 coatings.It is observed that all the three types of sprayed coatings exhibited dense microstructures with good bonding and good adherence to the steel substrate.Furthermore, the three coatings presented uniform lamellar structures, which is a typical characteristic of thermal-sprayed coatings [26,27].Some small black inclusions were found to be distributed within the coatings, which were verified to be oxides formed during the spraying process via EDS.Moreover, different contrast regions were observed within the ZnAl50 coatings.The regions with bright-white and dark-grey contrasts were confirmed to be a Zn-rich phase and Al-rich phase, respectively, using EDS analysis, whereas no distinct contrast difference was found within the ZnAl15 coating.This will be explained in detail in the following.Additionally, pores and inter-lamellar un-bonded interfaces were apparently observed within the arc-sprayed Zn and ZnAl coatings.These pores and un-bonded interfaces mainly originated from two aspects: one is the insufficient filling and wetting of molten droplets towards pores, and the other is due to the formation of oxides during the Additionally, pores and inter-lamellar un-bonded interfaces were apparently observed within the arc-sprayed Zn and ZnAl coatings.These pores and un-bonded interfaces mainly originated from two aspects: one is the insufficient filling and wetting of molten droplets towards pores, and the other is due to the formation of oxides during the spraying process, which hinders effective bonding between lamellae [28][29][30].The interconnection of the pores and un-bonded interfaces distributed within the coatings would promote the formation of through-thickness pores, thus favoring the penetration of corrosive media towards the substrate and impairing the barrier effect of the coating itself [31].The apparent porosities of the arc-sprayed Zn, ZnAl15, and ZnAl50 coatings were examined to evaluate their barrier effect.As listed in Table 2, apparently, the ZnAl15 coating demonstrated the lowest porosity of 3.2% and the pure Zn coating exhibited the highest porosity of 5.6%.That is to say, the Al content within the ZnAl coatings exhibited a significant influence on the coating porosity.
Coatings 2023, 13, 1720 6 of 17 In the case of the pure Zn coating, the Zn droplets melted early during the spraying process due to the metal's relatively low melting point (419.5 • C), which led to severe oxidation of Zn droplets and finally poor interface bonding and high porosity.This is further confirmed by the high oxygen content within the Zn coating, as listed in Table 2.For the ZnAl15 alloy coating, the addition of the alloying element Al into Zn resulted in an increased melting point of the ZnAl alloy and decreased oxidation during the spraying process, owing to the higher melting point of Al (660.3 • C) compared to Zn.In this way, low porosity was achieved.Nevertheless, when the Al content increased to 50%, i.e., for the ZnAl50 alloy coating, its porosity increased instead compared to that of the ZnAl15 coating.This can be explained as follows: the high Al content led to a relatively high melting point of the ZnAl50 coating, which would in turn hinder the sufficient melting of particles and thus the filling of pores.Moreover, Al is more inclined to oxidation than Zn since the Gibbs free energy for Al to form oxides is lower than that of Zn [32].The above two reasons resulted in the higher porosity of the ZnAl50 coating than that of the ZnAl15 coating.The chemical compositions and phase constitutions of the arc-sprayed Zn, ZnAl15, and ZnAl50 coatings were examined using EDS and XRD.As shown in Table 2, the chemical compositions of the three types of coatings were found to be comparable to that of the feedstock wires, indicating that there was no apparent elemental loss during the spraying process.Figure 2 presents the XRD results of the three arc-sprayed coatings.In the case of the as-sprayed pure Zn coating, only the Zn phase (PDF card No. 65-3358) was detected.For the as-sprayed ZnAl15 and ZnAl50 coatings, both Zn-rich and Al-rich phases (PDF card No. 65-2869) were identified.Moreover, the Zn-rich phase was found to be the primary phase for the ZnAl15 coating.When the Al content increased to 50%, i.e., in the case of the ZnAl50 coating, the Al-rich phase turned to be the dominant phase.Since the amount of Al-rich phase was low, distinct contrast differences would not be found in the ZnAl15 coating, which was in accordance with the SEM results shown in Figure 1c.

Long-Term Electrochemical Behavior of Zn, ZnAl15, and ZnAl50 Coatings
An electrochemical test over a long-lasting period was conducted for the arc-sprayed ZnAl coatings with Al contents varied from 0 to 50%.The evolution of OCP, polarization curves, and EIS during the long-term soaking process was examined to demonstrate the effect of Al content on the long-term protection performance for steel substrates.As can be seen from Figure 3, all the three types of ZnAl coatings exhibited

Long-Term Electrochemical Behavior of Zn, ZnAl15, and ZnAl50 Coatings
An electrochemical test over a long-lasting period was conducted for the arc-sprayed ZnAl coatings with Al contents varied from 0 to 50%.The evolution of OCP, polarization curves, and EIS during the long-term soaking process was examined to demonstrate the effect of Al content on the long-term protection performance for steel substrates.As can be seen from Figure 3, all the three types of ZnAl coatings exhibited similar OCP variation tendency with the prolonging of the immersion period.That is, for each coating, the OCP values moved to a negative direction in the first 360 h and then shifted to a positive direction after 360 h.Additionally, as shown in Figure 3 and the results listed in Table 3, after immersion for 840 h, the OCP values for the three types of coatings decreased in the following sequence: ZnAl50 coating > ZnAl15 coating > Zn coating.Moreover, it is found from Figure 4 that the corrosion potential of the three coatings during the long-term soaking test presented similar evolution tendencies to that of OCP.Since OCP and corrosion potential of a coating represent its corrosion tendency in thermodynamics, it is suggested that the corrosion tendency of ZnAl coatings decreased as the Al content increased from 0 to 50%.Furthermore, it could be inferred that the corrosion tendency of all the three coatings increased in the initial 360 h and then decreased until 840 h.Noteworthy, the OCP and corrosion potential of the three ZnAl coatings with Al contents of 0, 15%, and 50% were always much lower than the corrosion potential of −0.71V for the 921A steel substrate throughout the long-term soaking test.This indicates that the arc-sprayed Zn, ZnAl15, and ZnAl50 coatings acted as sacrificial anodes to protect the steel substrate all through the immersion test.Corrosion current density is another key factor usually used to evaluate the coating corrosion rate.In general, higher corrosion current density suggests a higher corrosion rate [33,34].It is obviously found from Figure 4 and Table 3 that, after long-term immersion testing for 840 h, the corrosion current densities of the ZnAl coatings decreased with the increase in Al content.The ZnAl50 coating demonstrated the lowest corrosion current density of 9.05 µA/cm 2 , which is approximately two orders of magnitude lower than that of the pure Zn coating.This fact indicates that the corrosion protection ability of the ZnAl coating improves as the Al content increases from 0 to 50%.In other words, a higher Al content contributes to a lower corrosion rate and greater corrosion protection performance in the 840 h soaking test.Moreover, for each coating, the corrosion current density increased in the initial 360 h soaking period; after 360 h, this value began to decrease as the soaking test continued, which is in accordance with the corrosion potential results.These results suggest that the corrosion rate of each coating increased in the first 360 h and then slowed down after 360 h.This may be due to the variation of the dominant corrosion mechanism before and after 360 h for the Zn, ZnAl15, and ZnAl50 coatings, which will be discussed in detail in the following section.Corrosion current density is another key factor usually used to evaluate the coating corrosion rate.In general, higher corrosion current density suggests a higher corrosion rate [33,34].It is obviously found from Figure 4 and Table 3 that, after long-term immersion testing for 840 h, the corrosion current densities of the ZnAl coatings decreased with Coatings 2023, 13, 1720 9 of 17 EIS was measured to further clarify the long-term corrosion behavior of different coatings.Figure 5 shows the evolution of Nyquist plots during 840 h of immersion for the three different coatings.The Nyquist plots consisted of two capacitive loops at high frequency and low frequency, respectively.In general, a larger diameter of the capacitive loop means a lower corrosion rate [33,35,36].Thus, it is indicated from Figure 5 that for each coating, at the early corrosion stage, the diameter of the capacitive loop decreased in the first 360 h.From 360 h to 840 h, the diameter of the capacitive loop increased.As observed from the phase angle Bode plots shown in Figure 6, two time constants were observed, corresponding to the two semi-circle loops in the Nyquist plots.Therefore, the generally used equivalent electrical circuit of R s (Q c (R c (Q dl R ct ))) (Figure 7) was employed to fit the EIS data and the fitted results of EIS were listed in Table 4. R s is the solution resistance; Q dl represents the constant phase element correlated to the inner double layer capacitance; Q c and R c are the constant phase element and resistance related to the corrosion product film; R ct is the charge transfer resistance.It is well known that R ct and impedance modulus value at low frequency |Z| f = 0.01 Hz are two important factors frequently used to evaluate the corrosion activity.Greater R ct and |Z| f = 0.01 Hz values usually represent lower corrosion rates [37,38].As shown in Figure 8 and Table 4, the R ct and |Z| f = 0.01 Hz values presented similar tendencies during the long-term immersion test, i.e., they tended to decrease over the initial 360 h and then underwent an increasing tendency after 360 h.These results indicate that the corrosion rate increased in the first 360 h and then decreased after 360 h, which is consistent with the corrosion current density results shown in to fit the EIS data and the fitted results of EIS were listed in Table 4. Rs is the solution resistance; Qdl represents the constant phase element correlated to the inner double layer capacitance; Qc and Rc are the constant phase element and resistance related to the corrosion product film; Rct is the charge transfer resistance.It is well known that Rct and impedance modulus value at low frequency |Z|f = 0.01 Hz are two important factors frequently used to evaluate the corrosion activity.Greater Rct and |Z|f = 0.01 Hz values usually represent lower corrosion rates [37,38].As shown in Figure 8 and Table 4, the Rct and |Z|f = 0.01 Hz values presented similar tendencies during the long-term immersion test, i.e., they tended to decrease over the initial 360 h and then underwent an increasing tendency after 360 h.These results indicate that the corrosion rate increased in the first 360 h and then decreased after 360 h, which is consistent with the corrosion current density results shown in

Effect of Al Content on the Corrosion Mechanism of Arc-Sprayed ZnAl Coatings
In order to reveal the effect of Al content on the corrosion mechanism of the arc-sprayed ZnAl coatings with Al contents varying from 0 to 50%, XRD evolutions of the corrosion products during the long-term soaking test were examined, and the results are shown in Figure 9.For the pure Zn coating, during the initial 360 h immersion, ZnO (PDF card No. 36-1451) was found to be the primary corrosion product, and the amount of ZnO increased with the prolonging of the immersion period in the first 360 h.This can be explained as follows: at this stage, the activation dissolution of Zn was the dominant corrosion mechanism.It has been reported that the corrosion protection ability of corrosion products was mainly attributed to their effective inhibition of the surface redox reaction.ZnO is usually loose and porous and could not provide effective barrier effect; more importantly, ZnO is a semi-conductor, which could hardly hinder the redox reaction [3,6].Therefore, ZnO is generally an undesirable corrosion product that demonstrates poor corrosion protection performance.In this way, the corrosion rate of the Zn coating increased gradually in the initial stage.After 360 h, it is found from Figure 9a that the amount of ZnO began to decrease.This phenomenon may be due to the transformation of ZnO to zinc salts such as hydrozincite and simonkolleite, which are usually found to be the corrosion products of Zn in chloride-containing solutions [8,9].Nevertheless, the diffraction peaks of hydrozincite or simonkolleite were not identified, which is likely due to their low amounts at the coating surface.These zinc slats are slightly soluble in aqueous solutions and present denser and more stable characteristics than that of ZnO; moreover, they could mitigate redox reactions to some extent.Thus, after 360 h, it can be inferred that, attributed to the presence of zinc salts such as hydrozincite or simonkolleite at the Zn coating surface, the corrosion protection performance would be

Effect of Al Content on the Corrosion Mechanism of Arc-Sprayed ZnAl Coatings
In order to reveal the effect of Al content on the corrosion mechanism of the arcsprayed ZnAl coatings with Al contents varying from 0 to 50%, XRD evolutions of the corrosion products during the long-term soaking test were examined, and the results are shown in Figure 9.For the pure Zn coating, during the initial 360 h immersion, ZnO (PDF card No. 36-1451) was found to be the primary corrosion product, and the amount of ZnO increased with the prolonging of the immersion period in the first 360 h.This can be explained as follows: at this stage, the activation dissolution of Zn was the dominant corrosion mechanism.It has been reported that the corrosion protection ability of corrosion products was mainly attributed to their effective inhibition of the surface redox reaction.ZnO is usually loose and porous and could not provide effective barrier effect; more importantly, ZnO is a semi-conductor, which could hardly hinder the redox reaction [3,6].Therefore, ZnO is generally an undesirable corrosion product that demonstrates poor corrosion protection performance.In this way, the corrosion rate of the Zn coating increased gradually in the initial stage.After 360 h, it is found from Figure 9a that the amount of ZnO began to decrease.This phenomenon may be due to the transformation of ZnO to zinc salts such as hydrozincite and simonkolleite, which are usually found to be the corrosion products of Zn in chloride-containing solutions [8,9].Nevertheless, the diffraction peaks of hydrozincite or simonkolleite were not identified, which is likely due to their low amounts at the coating surface.These zinc slats are slightly soluble in aqueous solutions and present denser and more stable characteristics than that of ZnO; moreover, they could mitigate redox reactions to some extent.Thus, after 360 h, it can be inferred that, attributed to the presence of zinc salts such as hydrozincite or simonkolleite at the Zn coating surface, the corrosion protection performance would be improved and its corrosion rate would be thus decreased, which is consistent with the electrochemical test results.However, from a long-term perspective, zinc salts like hydrozincite and simonkolleite are not that stable, as they might convert to ZnO in an alkaline solution and are incapable of offering long-term corrosion protection.
Coatings 2023, 13, x FOR PEER REVIEW 13 of 18 improved and its corrosion rate would be thus decreased, which is consistent with the electrochemical test results.However, from a long-term perspective, zinc salts like hydrozincite and simonkolleite are not that stable, as they might convert to ZnO in an alkaline solution and are incapable of offering long-term corrosion protection.In the case of the ZnAl15 coating, as shown in Figure 9b, for the initial 360 h, ZnO was also the dominant corrosion product, as with the Zn coating, indicating that the dissolution of Zn is the primary corrosion mechanism at this stage.As shown in Figure 2, both the Al-rich phase and Zn-rich phase existed in the as-sprayed ZnAl15 coating.Galvanic cells would be formed between the Al-rich and Zn-rich phases, promoting the preferential dissolution of the Zn-rich phase to form ZnO. Since ZnO could not provide effective corrosion protection, the corrosion rate of the ZnAl15 coating increased at this stage.During the second corrosion stage, i.e., after 360 h, as can be seen from Figure 9b, the Al-rich phase began to be the dominant phase instead of the Zn-rich phase.This could be explained as follows: after 360 h, most of the Zn-rich phase at the coating surface had dissolved, leaving the Al-rich phase to become the primary phase.This fact is in accordance with the EDS results (Table 5), which indicated that the Zn/Al mass ratio at the coating surface was 0.86 after 840 h immersion, which is significantly deviated from the original 5.7 Zn/Al ratio in the as-sprayed ZnAl15 coating.In this way, the Al-rich phase started to dissolve to form Al2O3 (PDF card No. 50-1496), as indicated in Figure 9b.The corrosion product Al2O3 is dense, stable, and an excellent insulator, facilitating superior corrosion protection for the under-layered coating to some extent [39], which resulted in a gradual decrease in the corrosion rate of the ZnAl15 coating after 360 h.In the case of the ZnAl15 coating, as shown in Figure 9b, for the initial 360 h, ZnO was also the dominant corrosion product, as with the Zn coating, indicating that the dissolution of Zn is the primary corrosion mechanism at this stage.As shown in Figure 2, both the Al-rich phase and Zn-rich phase existed in the as-sprayed ZnAl15 coating.Galvanic cells would be formed between the Al-rich and Zn-rich phases, promoting the preferential dissolution of the Zn-rich phase to form ZnO. Since ZnO could not provide effective corrosion protection, the corrosion rate of the ZnAl15 coating increased at this stage.During the second corrosion stage, i.e., after 360 h, as can be seen from Figure 9b, the Al-rich phase began to be the dominant phase instead of the Zn-rich phase.This could be explained as follows: after 360 h, most of the Zn-rich phase at the coating surface had dissolved, leaving the Al-rich phase to become the primary phase.This fact is in accordance with the EDS results (Table 5), which indicated that the Zn/Al mass ratio at the coating surface was 0.86 after 840 h immersion, which is significantly deviated from the original 5.7 Zn/Al ratio in the as-sprayed ZnAl15 coating.In this way, the Al-rich phase started to dissolve to form Al 2 O 3 (PDF card No. 50-1496), as indicated in Figure 9b.The corrosion product Al 2 O 3 is dense, stable, and an excellent insulator, facilitating superior corrosion protection for the under-layered coating to some extent [39], which resulted in a gradual decrease in the corrosion rate of the ZnAl15 coating after 360 h.
For the ZnAl50 coating, its corrosion also consisted of two stages.In the primary 360 h, similar to the ZnAl15 coating, ZnO was the primary corrosion product owing to the preferential dissolution of the Zn-rich phase.Since ZnO could not provide powerful corrosion protection, the corrosion rate thus increased with the extension of the immersion period.After 360 h, apart from the Zn-rich phase, Al-rich phase, and Al 2 O 3 phase, the diffraction peaks of Zn 6 Al 2 (OH) 16 CO 3 •4H 2 O (PDF card No. 38-0486), a type of layered double hydroxide (LDH), were also detected.Moreover, with the increase in the soaking time, the amount of LDH increased.LDH is formed due to the co-deposition of Zn ions and Al ions in the aqueous solution, and is proven to be a superior and desirable corrosion product that could effectively prevent further corrosion of the steel substrate.The remarkable corrosion protection capability of LDH is mainly derived from two aspects: on the one hand, LDH is more stable and insoluble in chloride-containing aqueous solutions compared to ZnO and zinc salts, and the redox rate of the ZnAl coating could be thus mitigated in the presence of LDH.On the other hand, LDH is generally dense and compact, acting as a superior protective barrier inhibiting further corrosion of the coating [6,21].Therefore, the presence of both Al 2 O 3 and LDH at the surface of the ZnAl50 alloy coating contributed to the drastically decreased corrosion rate and remarkable long-term corrosion protection performance for the steel substrate.This was also confirmed through the EDS results shown in Table 5, which indicate that the mass ratio of Zn/Al at the coating surface is 0.88, which is only slightly different from the Zn/Al ratio of ~1 in the as-sprayed ZnAl50 alloy coating.This fact suggests that the dissolving of Zn is effectively inhibited and the corrosion products are relatively stable in the aqueous solution.The corrosion product morphologies of the Zn coating, ZnAl15 coating, and ZnAl50 coating after soaking in 3.5% NaCl aqueous solution for 840 h were characterized via SEM to further reveal their long-term corrosion mechanisms.As shown in Figure 10, after 840 h immersion, the ZnAl coatings with Al contents of 0, 15%, and 50% presented significantly different corrosion morphologies.For the pure Zn coating, its corrosion product, ZnO (as confirmed by XRD in Figure 9a), was loose and porous as expected, exhibiting a granular and nanosheet morphology consistent with the reported morphology of ZnO [8].In the case of the ZnAl15 coating, discontinuous granular and flocculent corrosion products were formed on the coating surface, which have been confirmed to be mostly Al 2 O 3 .In addition, some pores and cracks were still observed on the coating surface.On the contrary, for the ZnAl50 coating, instead of porous corrosion products, dense and uniform corrosion products were observed and defects such as pores and cracks on the coating surface essentially disappeared.
On the basis of the above results, the long-term corrosion behavior and corrosion mechanism of arc-sprayed ZnAl coatings with Al content ranged from 0 to 50% could be summarized as follows.In general, the corrosion of the ZnAl coatings with Al contents varying from 0 to 50% was composed of two main stages.At the first stage, in approximately ten days, the activation dissolving of Zn phase was the dominant corrosion mechanism for all three ZnAl coatings.However, the loose and porous corrosion product ZnO could hardly hinder the further corrosion of the inner coating.Thus, the corrosion rates of the Zn and ZnAl coatings continuously increase with prolonging of the immersion period at the initial stage.
At the second stage, ZnO was still the primary corrosion product for the Zn coating, which could not inhibit the extensive dissolving of Zn and therefore was unable to provide long-term corrosion protection for the steel substrate.In this way, the pure Zn coating presented the worst corrosion protection behavior for the steel substrate.For the ZnAl15 and ZnAl50 coatings, new and powerful corrosion products were formed.In terms of the ZnAl15 coating, due to the addition of Al, stable and dense Al 2 O 3 turned out to be the dominant corrosion product, which could block a fraction of pores, i.e., exhibiting a selfhealing effect, and impede the dissolving of the Zn-rich phase to some extent.Nevertheless, due to the low Al content in the ZnAl15 alloy coating, the amount of the formed Al 2 O 3 layer was low and it was unable to form a continuous protective barrier layer; thus, its protection effect was limited.When the Al content increased to 50%, i.e., for the ZnAl50 coating, besides Al 2 O 3 , LDH was also formed since the high Al content favors the formation of LDH.The relatively stable, compact, and uniform corrosion product LDH could construct a continuous and powerful barrier layer on the coating surface.Moreover, the corrosion products Al 2 O 3 and LDH presented effective self-healing to block the pores within the coating.In this way, the most powerful and excellent long-term corrosion protection for the inner coating and substrate could be realized with the ZnAl50 coating.Thus, the long-term corrosion protection performance of the arc-sprayed ZnAl coatings improved with the increase in Al content.On the basis of the above results, the long-term corrosion behavior and corrosion mechanism of arc-sprayed ZnAl coatings with Al content ranged from 0 to 50% could be summarized as follows.In general, the corrosion of the ZnAl coatings with Al contents varying from 0 to 50% was composed of two main stages.At the first stage, in approximately ten days, the activation dissolving of Zn phase was the dominant corrosion mechanism for all three ZnAl coatings.However, the loose and porous corrosion product ZnO could hardly hinder the further corrosion of the inner coating.Thus, the corrosion rates of the Zn and ZnAl coatings continuously increase with prolonging of the immer-
Figure 4. Additionally, when comparing the R ct and |Z| f = 0.01 Hz values among ZnAl coatings with varied Al contents, it is obviously found that the Zn coating demonstrated the lowest R ct and |Z| f = 0.01 Hz values and the ZnAl50 coating exhibited highest R ct and |Z| f = 0.01 Hz values after the long-term immersion test.This fact further confirms that the corrosion protection ability of the ZnAl coating improved as the Al content increased from 0 to 50%.Coatings 2023, 13, x FOR PEER REVIEW 10 of 18

Figure 4 .
Additionally, when comparing the Rct and |Z|f = 0.01 Hz values among ZnAl coatings with varied Al contents, it is obviously found that the Zn coating demonstrated the lowest Rct and |Z|f = 0.01 Hz values and the ZnAl50 coating exhibited highest Rct and |Z|f = 0.01 Hz values after the long-term immersion test.This fact further confirms that the corrosion protection ability of the ZnAl coating improved as the Al content increased from 0 to 50%.

Figure 7 .
Figure 7. Equivalent electrical circuit employed for EIS data fitting.

Figure 7 .
Figure 7. Equivalent electrical circuit employed for EIS data fitting.

Figure 7 .
Figure 7. Equivalent electrical circuit employed for EIS data fitting.

Figure 10 .
Figure 10.Corrosion morphologies of different coatings after 840 h immersion.(a,b) Zn coating; (c,d) ZnAl15 coating; (e,f) ZnAl50 coating.(b,d,f) are the corresponding enlarged images of (a,c,e), respectively.The yellow arrows inserted in (a,c,e) denote the corrosion products.

Figure 10 .
Figure 10.Corrosion morphologies of different coatings after 840 h immersion.(a,b) Zn coating; (c,d) ZnAl15 coating; (e,f) ZnAl50 coating.(b,d,f) are the corresponding enlarged images of (a,c,e), respectively.The yellow arrows inserted in (a,c,e) denote the corrosion products.To sum up, the effect of Al content on the long-term corrosion mechanisms of arcsprayed ZnAl coatings could be summarized as follows.The variation of corrosion resistance among ZnAl coatings with different Al contents was principally derived from the formation of different corrosion products.The Al content within the arc-sprayed ZnAl

Table 1 .
Chemical compositions of 921A hull steel substrate.

Table 2 .
Chemical compositions and porosities of three arc-sprayed coatings.

Table 3 .
Electrochemical parameters for different coating samples.

Table 3 .
Electrochemical parameters for different coating samples.

Table 4 .
EIS data obtained from EIS plots after fitted by equivalent circuit diagram.

Table 4 .
EIS data obtained from EIS plots after fitted by equivalent circuit diagram.

Table 4 .
EIS data obtained from EIS plots after fitted by equivalent circuit diagram.

Table 5 .
Chemical compositions of the coating surface for different types of coatings after 840 h immersion.