Special Issue on Aluminium and Its Alloys for Zero Carbon Society, ICAA 18
Self-Healing Coatings with Double-Layered Structure for Corrosion Protection of Aluminum Alloys
2023 年 64 巻 2 号 p. 473-478
詳細
2023 年 64 巻 2 号 p. 473-478
Surface treatments of Al alloys are generally performed to protect them from corrosion before they are used as many kinds of industrial product. However, corrosion protection abilities of the surface layers formed by the surface treatment are lost by unexpected physical damages because the metal substrate is exposed to corrosive environments at the damaged area. From the viewpoint described above, surface films with self-healing properties should be developed to keep corrosion protection abilities high for long periods without any maintenance. In this study, double-layered films with self-healing abilities were formed on Al alloys by the combination of anodizing and organic coating. Corrosion tests in Cl−/Cu2+ solutions after physical damaging were carried out to compare corrosion protection abilities of the double-layered films formed by three processes: formation of film including 1) no healing agent in the outer and inner layers, 2) healing agents only in the inner layer, and 3) healing agents both in the outer and inner layers. The double-layered film including healing agents in both layers showed much higher self-healing ability than other coatings.
Improvement of corrosion protection of Al and Al alloys is a quite important issue because they are used as fundamental materials, including automobiles, airplanes, electronic devices, and so on. This is due to their low densities, low costs and high heat conductivities.1–3) Surface treatments of Al alloys include anodizing, chemical treatment, organic coating, metal plating, and thermal spraying. They are selected and sometimes combined each other for the purpose of the corrosion protection of industrial products employed for different periods and in different environments. However, all the surface treatments have a weak point that the corrosion protection ability of surface films formed by the treatments is easily lost by unexpected mechanical damages of the film. Corrosion may proceed severely, when the substrate is exposed to corrosive environments by the physical damage. Therefore, maintenance is carried out at intervals to keep the corrosion protection ability of the film high. Thus, “self-healing coatings” that can be healed spontaneously with no maintenance after damaging, are significantly useful in many industrial fields.4,5)
In previous studies6–12) we developed two types of self-healing coating for corrosion protection of Al and Al alloys. The first one is the formation of organic coatings with micro-capsules containing healing agents6–10) on electropolished specimens. The coatings can be formed by the following procedure. At first, the solution of prepolymer, as a precursor of polyurethane shell of micro-capsules, synthesized from diisocyanate and polyole, is dripped to the polyole and surfactant solution under agitation. During the process, the prepolymer solution forms spherical micelles in polyole, and then the spherical micelles become smaller to form an emulsion of polyole solution and prepolymer solution. Finally, polyurethane capsules including healing agents are formed (Fig. 1), and dried at room temperature after collecting on filtering sheet. The detail of the reaction mechanism on the formation of the micro-capsules has been described elsewhere.6–10) The capsules are dispersed in prepolymer/ethylene glycol solution and spread on the surface of electropolished specimens (Fig. 2(a)).
Formula of reaction between prepolymer and glycerol to form polyurethane shell.
Schematic model of (a) polyurethane coating with microcapsules, (b) physical damage of coating and IPDI flow into damaged area, (c) formation of polymer, covering the damaged area, (d) filling pores of porous anodic oxide film with IPDI, (e) physical damage of pore wall and IPDI flow into damaged area, and (f) formation of polymer, covering the damaged area.
When the organic coating is physically damaged, capsules are destructured simultaneously, and healing agents flow from capsules into the damaged area of coating (Fig. 2(b)). The healing agent flowing reacts with moisture in air to form polyurethane-like polymer at the damaged area of coating, leading to the coverage of the substrate exposed by mechanical damaging (Fig. 2(c)).
The second self-healing coating technique is the filling in nanopores of a porous type of anodic oxide film on Al alloys with healing agents11,12) (Fig. 2(d)). Initially, Al alloys are anodized in oxalic acid solutions to form a porous type anodic oxide film13) and immersed in phosphoric acid solutions to make the pore diameter larger. The specimen after anodizing and pore widening treatment (P.W.-treatment) is immersed in Isophorone-Diisocyanate (IPDI) solution under ultrasonic vibration for 100 min to fill pores with the solution (Fig. 2(d)). By forming a thin polyurethane coating on the surface, IPDI remaining unreacted in the pores is covered. When the specimen filled with IPDI in nanopores of anodic oxide films is physically damaged, pore walls are destructured, and healing agents flow into the damaged area of coating (Fig. 2(e)). Thus, the damaged area of coating is healed with polyurethane-like polymer formed by the reaction with moisture in air (Fig. 2(f)).
Both previous studies6–12) showed appreciable positive effects on the self-healing for corrosion protection, suggesting the mechanism described above. However, self-healing properties were not high enough, when damages were employed severely on the coating. This is because the amount of flowing IPDI is not large enough to cover the whole area damaged.
In the present study, a novel process is attempted to form surface films with self-healing abilities much higher than those obtained in previous studies.6–12) The process is the combination of the formation of organic coating with micro-capsules and the filling of nanopores with IPDI on Al alloys covered with a porous type anodic oxide film. This is obtained by combining the first procedure developed in a previous study with the second one and called “Self-healing double-layered coating” (Fig. 3). The self-healing double-layered coating consists of two layers: outer layer and inner layer. The outer layer is organic films with micro-capsules including healing agents, and the inner layer is porous anodic oxide films including healing agents in nanopores.
Schematic model of double-layered coating with (a) outer polyurethane layer and inner porous anodic oxide layer, prepared by Process-I, (b) outer polyurethane layer and inner porous anodic oxide layer containing IPDI, prepared by Process-II, and (c) outer polyurethane layer with microcapsules and inner porous anodic oxide layer containing IPDI, prepared by Process-III.
In this study, self-healing abilities of three types of double-layered coating are compared, by corrosion tests in a CuSO4/KCl solution. The first one includes procedures that organic film without healing agents is coated on anodic oxide film without healing agents (Process-I, Fig. 3(a)). The second one includes procedures that organic film without healing agents is coated on anodic oxide film with healing agents (Process-II, Fig. 3(b)). The third one includes procedures that organic film with micro-capsules containing healing agents is coated on anodic oxide film with healing agents (Process-III, Fig. 3(c)).
1050-Al alloy plates (Fe: 0.25, Si: 0.15, Cu: 0.05, Al: 99.5 mass%) were cut into 20 mm × 20 mm × 1.5 mmt and electropolished in 78 vol% - acetic acid/22 vol% - perchloric acid solution with a constant voltage of 30 V for 30 s, as a pretreatment. Pretreated specimens were anodically oxidized in 2 mass% - oxalic acid solution at 40°C for 60 min with a current density of 200 Am−2 to form a porous type of oxide film on the surface in Processes-I, -II, and -III. Some specimens were subjected to pore widening treatment by immersing in 5 mass% phosphoric acid solution for 30 min at room temperature after anodizing.14) Under this condition, the thickness of porous oxide films formed on the substrate is about 30 µm11) and the diameter of pores of porous oxide films is about 100 nm.9)
2.2 Pore filling with healing agentsIn Processes-II and -III, anodized specimens with/without pore widening treatment were immersed in IPDI, as a healing agent of coating, for 100 min under ultrasonic agitation. According to Hirasawa et al.,11) 0.84 mg cm−2 of IPDI can be contained in pores of porous film with pore-widening under these conditions.
2.3 Formation of capsules including healing agentsIn Process-III, micro-capsules containing healing agents were synthesized by the following procedure (Fig. 4). Firstly, tolylene diisocyanate (TDI) was reacted with glycerol in cyclohexanone for 24 h under 600 rpm at 75°C to form prepolymer of polyurethane shell. Here, only for the first 1 h, N2 gas was blown into the mixture to remove water dissolved. Then, healing agents, IPDI, and xylene were mixed to the prepolymer solution obtained above. The mixed solution was dripped to a 3 mass%-sodium dodecyl sulfate (SDS)/5 mass%-glycerol solution under 1150 rpm of agitation at 90°C for 30 min. During the process, prepolymer reacts with glycerol to form polyurethane capsules containing healing agents of IPDI, cyclohexanone and xylene.5,6) In this solution, the amounts of prepolymer, IPDI, cyclohexanone, and xylene were 10.0 g, 9.5 g, 6.5 g, and 9.9 g, respectively. The polyurethane micro-capsules were collected on a filtering sheet and dried at room temperature.
Flow diagram of the formation of polyurethane micro-capsules containing healing agents, IPDI.
In Processes-I, and II, outer polyurethane layer without micro-capsules was coated on the specimens prepared in 2.1, and 2.2 by spreading prepolymer/ethylene glycol (mass ratio: 75:10) The spreading in Process-1 was carried out just after anodizing/pore-widening by dropping prepolymer/ethylene glycol on the specimen and tilting the specimen in different directions to obtain uniform thickness of 30 µm. In Process-II, the spreading was carried out just after pore-filling with healing agents, as described above.
In Process-III, micro-capsules were dispersed in prepolymer/ethylene glycol at the mixing mass ratio of 1:75:10 and spread on the specimen prepared in 2.2. Some specimens in Process-III were prepared by dispersing micro-capsules to prepolymer/ethylene glycol at the mixing mass ratio of 5:75:10, to examine the effect of amount of healing agents on the self-healing property of a coating. In the Process-III, spreading of the outer polyurethane layer with micro-capsules was carried out just after pore-filling with healing-agents in the inner layer, as described above. After coating of the outer layer, all the specimens prepared by Processes I, -II, and -III were kept for 48 h in air atmosphere at room temperature to age the polymer in the outer layer.
2.5 Corrosion test after physical damaging of coatingIn order to evaluate the self-healing abilities of specimens prepared by Processes-I, -II, and -III, all the specimens were damaged with a cutter blade by employing loads of 7 and 18 N in the length of 12–17 mm, and then kept for 24–48 h in air atmosphere at room temperature to allow the damaged coating to be cured. Curing of polymer proceeds fast during the initial 1–2 h, and 24–48 h is long enough for stable state. The specimens after curing were immersed in 1.57 × 10−3 M - CuSO4/0.57 M - KCl solution for 24 h at room temperature. After corrosion tests, organic coating, anodic oxide film, corrosion products and Cu particles deposited were removed from the Al alloy substrate by immersing in a commercially available coating-remover and a 10 mass%-phosphoric acid/4 mass%-chromic acid solution, sequentially. Surfaces of specimens before/after corrosion tests and also after removing films and corrosion products were observed by scanning electron microscopy (SEM).
Figure 5 shows SEM images of specimens covered (a) with outer polyurethane layer and inner porous anodic oxide layer, prepared by Process-I, (b) with outer polyurethane layer and inner oxide layer with IPDI, prepared by Process-II, and (c) and (d) with the outer polyurethane layer with micro-capsules and inner oxide layer with IPDI, prepared by Process-III. All the images were taken at areas damaged with a load of 7 N. Both Fig. 5(c) and (d) are taken to examine the reproducibility of self-healing abilities on the specimens prepared by Process-III. In Fig. 5(a), there is a crack with a width of 40 µm at the center of the image, and there appears the metal substrate with a width of 5 µm at the bottom of the crack continuously. This suggests that the specimen prepared by Process-I has no self-healing ability for corrosion protection. In Fig. 5(b), the metal substrate at the bottom of the crack is covered discontinuously with some substances. The substance covering the substrate may be polyurethane-like polymer that has been formed by the reaction of IPDI with moisture in air after IPDI-flowing from destroyed porous anodic oxide into the crack. This suggests that the specimen prepared by Process-II has appreciable self-healing abilities for corrosion protection. In Figs. 5(c), and (d), there appears cracks and no metal substrate at the bottom of the crack. This strongly suggests that the self-healing ability of the specimen prepared by Process-III is much higher than that of the specimen prepared by Process-II. This is because the specimen prepared by Process-III can supply large amounts of IPDI into the damaged area from both outer and inner layers.
SEM images of specimen covered (a) with outer polyurethane layer and inner porous anodic oxide layer, prepared by Process-I, (b) with outer polyurethane layer and inner oxide layer with IPDI, prepared by Process-II, and both (c) and (d) with outer polyurethane layer with micro-capsules and inner oxide layer with IPDI, prepared by Process-III. All the images were taken at areas damaged with a load of 7 N.
Figure 6 shows SEM images of specimens prepared by (a) Process-I, (b) -II, and (c) -III, obtained after physical damaging with 7 N and corrosion tests in CuSO4/KCl solution. In Fig. 6(a), large amounts of corrosion products appear at the damaged area, filling the whole volume of the crack produced by physical damaging. In Fig. 6(b), there are Cu particles deposited at damaged area, and the diameter of the largest particle is about 50 µm. In Fig. 6(c), there are zigzag-shaped cracks and a small number of Cu particles with 26 µm diameter in maximum.
SEM images of specimens prepared by (a) Process-I, (b) -II, and (c) -III, and subjected to physical damaging at 7 N and corrosion tests in CuSO4/KCl solution.
When aluminum is immersed in CuSO4/KCl solution, the following reactions occur.
| \begin{equation} \text{Al} + \text{3H$_{2}$O}\to \text{Al(OH)$_{3}$} + \text{(3/2)H$_{2}$} \end{equation} | (1) |
| \begin{equation} \text{2Al} + \text{3Cu$^{2+}$}\to \text{2Al$^{3+}$} + \text{3Cu$\downarrow$} \end{equation} | (2) |
Based on the theory described above, one can evaluate the progress of corrosion from the amounts of Al(OH)3 and Cu particles deposited on the specimen. Amounts of Al(OH)3 and Cu deposited on specimens after corrosion tests are in the order of the specimen prepared by
| \begin{equation*} \text{Process-I} > \text{Process-II} > \text{Process-III} \end{equation*} |
Figure 7 shows SEM images of specimens prepared by (a) Process-I, (b) Process-II, and (c) Process-III, obtained after damaging by a load of 18 N. On all the specimens, there are cracks with a width of about 40 µm. They are as wide as those produced by 7 N of load (see Fig. 5). This may be because the width of cracks can be determined by that of blade used. However, it is considered that the depth of cracks is deeper as larger load is employed.
SEM images of specimens prepared by (a) Process-I, (b) Process-II, and (c) Process-III, and subjected to damaging at a load of 18 N.
In Figs. 7(a) and (b), the Al alloy substrate is exposed continuously at the whole area of bottom of cracks, but in Fig. 7(c) the substrate is covered discontinuously with polymer.
It can be seen from Fig. 7(c) that the amount of polymer supplied into damaged area is not large enough, even on the specimen prepared by Process-III, for covering the whole area of the substrate exposed in the case of large load such as 18 N.
Figure 8 shows SEM images of specimens prepared by (a) Process-1, (b) Process-II, and (c) Process-III, and subjected to immersion in commercially available coating-remover and phosphoric acid/chromic acid solution after damaging at 18 N and corrosion test in CuSO4/KCl solution. The immersion in the coating-remover and phosphoric acid/chromic acid solution leads to the removal of organic compounds, anodic oxide film, corrosion products, and Cu particles deposited, leaving only the metal substrate. In Fig. 8(a), there appears an irregular-shaped pit with a length of 150 µm and a width of 30–50 µm, and in Fig. 8(b), there appear irregular-shaped shallow concaves. In Fig. 8(c), there is a circular pit with a diameter of 60 µm on rough surface. From Fig. 8, one can understand that corrosion rate of the substrate is higher in the order of the specimen prepared by Process-I > Process-II > Process-III, and that Process-III can’t completely protect the corrosion of the substrate after damaging at 18 N of load.
SEM images of specimens prepared by (a) Process-1, (b) Process-II, and (c) Process-III, and subjected to immersion in commercially available coating-remover and phosphoric acid/chromic acid solution after damaging at 18 N and corrosion test in CuSO4/KCl solution.
In order to improve the self-healing ability of the specimen prepared by Process-III, we attempt to increase the amount of IPDI in the outer and inner layers by three strategies. The first strategy is to increase the pore volume of porous anodic oxide films by P.W. treatment. By the treatment under the condition described in 2.2, the diameter of each pore becomes 1.5 times, and thus the pore volume becomes 2.25 times.14) Conclusively, one can pack much larger amounts of IPDI in pores with P. W. treatment than without P. W. treatment. The second strategy is to increase the number of capsules dispersed in the outer layer. As described in 2.4, the number of capsules in the outer layer is 5 times as large as that in regular procedure. The third strategy is the combination of the first and second strategies.
Figure 9(a) shows SEM image of the specimen prepared by Process-III, in which 2.25 times the amount of IPDI is contained in the inner layer by P. W. treatment and subjected to damaging at 18 N of load. Figure 9(b) shows SEM image of the specimen shown in Fig. 9(a), after corrosion tests in CuSO4/KCl solution and immersion in commercially available coating-remover and in phosphoric acid/chromic acid solution. The substrate remains exposed discontinuously at the bottom of cracks after damaging (Fig. 9(a)), and a circular pit with a diameter of 60 µm is formed by corrosion of the substrate (Fig. 9(b)). It is clear from Fig. 9 that P.W. treatment under the condition in this study can’t protect the corrosion completely after damaging at a load of 18 N.
SEM images of (a) the specimen prepared by Process-III, in which 2.25 times the amount of IPDI is contained in the inner layer by P. W. treatment, and subjected to damaging at 18 N, (b) the specimen shown in Fig. 9(a), after corrosion tests in CuSO4/KCl solution and immersion in commercially available coating-remover and in phosphoric acid/chromic acid solution.
Figure 10(a) shows SEM image of the specimen prepared by Process-III, in which micro-capsules are dispersed in the outer layer by 5 times as large as in that of Figs. 5–9, and subjected to damaging at a load of 18 N, and Fig. 10(b) shows SEM image of the specimen shown in Fig. 10(a), after corrosion tests in CuSO4/KCl solution and immersion in commercially available coating-remover and in phosphoric acid and chromic acid solution. It can be seen from Figs. 10(a), and (b) that the increase in the number of capsules in the outer layer enables polyurethane-like polymer to cover almost the whole of the damaged area, but small pits are formed by corrosion during corrosion tests at uncovered areas.
SEM images of (a) the specimen prepared by Process-III, in which micro-capsules are dispersed in the outer layer 5 times as large as in that of Figs. 5–9, and subjected to damaging at 18 N, and (b) the specimen shown in Fig. 10(a), after corrosion tests in CuSO4/KCl solution and immersion in commercially available coating-remover and in phosphoric acid/chromic acid solution.
Finally, Fig. 11(a) shows SEM image of the specimen prepared by Process-III, in which five times the number of micro-capsules is dispersed in the outer layer, and 2.25 times the amount of IPDI are contained in the inner layer and subjected to damaging at a load of 18 N. Figure 11(b) shows SEM image of the specimen shown in Fig. 11(a), after corrosion tests in CuSO4/KCl solution and immersion in commercially available coating-remover and in phosphoric acid/chromic acid solution. It seems that the whole area damaged is covered thoroughly with polyurethane-like polymer (Fig. 11(a)), but a circular pit with 33 µm of diameter is formed by corrosion near the damaged area (Fig. 11(b)).
SEM images of (a) the specimen prepared by Process-III, in which 5 times the number of micro-capsules is dispersed in the outer layer, and 2.25 times the amount of IPDI are contained in the inner layer, and subjected to damaging at 18 N, and (b) the specimen shown in Fig. 11(a), after corrosion tests in CuSO4/KCl solution and immersion in commercially available coating-remover and in phosphoric acid/chromic acid solution.
Conclusively, three attempts to increase in the amount of IPDI in the inner and/or outer layer can improve the self-healing ability of the specimen prepared by Process-III, but further investigation is needed to achieve the complete corrosion protection after severe physical damage such as a load of 18 N.
In the present study, a novel technique for forming double-layered film with self-healing ability for the corrosion protection of Al alloys were studied, and the following conclusions are drawn.
This research was financially supported by JKA (Keirin). Specimens used in this study were supplied by UACJ, Inc.
