One-Part Self-Healing Anticorrosive Coatings
Design Strategy and Examples
Jinglei Yang; Mingxing Huang School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore
Abstract
Corrosion control is one of the most important issues in the fields of materials design and application, and the concept of self-healing has been recently proposed as a novel approach to achieve corrosion protection function. Although extensive investigation has been made to self-healing materials, one-part self-healing anticorrosive coatings is a relatively new topic. In the present work, the background of one-part self-healing anticorrosive coatings is presented and the strategy for their design discussed. The one-part self-healing anticorrosive coating is further demonstrated by two examples, that is, diisocyanated-based and organic silane-based coatings. One-part self-healing anticorrosive coatings are fabricated through the microencapsulation of different healing species, and good self-healing anticorrosive performance of the prepared coatings is characterized by different methods.
14.1 Introduction
Corrosion of metal is one of the most destructive processes that results in material failure, and it causes huge economic loss annually. Corrosion is a destructive attack of a metal by chemical or electrochemical reaction with the environment, and it is an electrochemical process that involves anodic reaction and cathodic reaction.1,2 Reduction reactions and oxidation reaction occur at the cathodic area and anodic area, respectively, while the corrosion of metal usually occurs at the anodic area.
As an example, the corrosion of iron is demonstrated in Figure 14.1. It can be seen that, when an iron panel is immersed in corrosive electrolyte solution, the cathodic reaction consists of the reduction of dissolved oxygen, hydrogen ion, and water, while the oxidation reaction at the anodic area dissolves iron to produce Fe2 +, leaving voids in the iron panel. Usually Fe2 + will react with OH− in the solution to yield Fe(OH)2, and after a serial of further reactions, finally Fe2O3 will be produced, and this is the rust that one usually sees on a corroded iron.
Corrosion control is one of the most important issues when people use metals. To date many strategies have been developed to alleviate corrosion damage, and traditional approaches include material selection, cathodic protection, addition of corrosion inhibitor, engineering design, and so forth.2–4 Nevertheless, among the variety of approaches for corrosion control, applying coating on an object is the most straightforward one to provide corrosion protection function to metal substrates. Coating provides a direct barrier layer to separate the underneath metal from external corrosive environment. In addition, corrosion inhibitor can be loaded into the coating layer to provide further corrosion suppression function. However, the degradation of coating is always a problematic issue. If coating degradation and failure are not detected in a timely manner, the underlying metal will be directly exposed to the corrosive environment and may suffer from severe and rapid corrosion. In this case, if the coating can autonomously respond to the degradation and failure to recover its integrity or other function, its service life will be dramatically extended, and the corrosion protection function will be prolonged as well. Materials that possess the autonomous recovery ability are self-healing materials.
Self-healing behavior is a ubiquitous phenomenon in nature. For example, animals are able to self-heal via a “bleeding” mechanism when their skin experiences injury or scratches.5 In such a process, the biological system responds to external stimuli, injury for example, and transfers healing species to the injured location to heal the injury. Inspired by nature, modern self-healing materials are also able to sense damage and release healing agents to realize self-healing function. Self-healing is advantageous over traditional techniques including welding, patching, and in situ curing of resin for repairing materials because it is much more time and cost efficient.6
Due to their great importance, self-healing materials have been the focus of researchers since the last century, and many approaches have been developed for their manufacturing. However, most of early self-healing materials require manual intervention such as heating and irradiation. For example, Jud et al.7 reported the self-healing performance of a thermoplastic polymer, but the healing was realized at elevated temperature that allowed for molecular interdiffusion. Chung et al.8 reported the self-healing behavior of a poly(methoxy methylacrylate) coating, but light irradiation was necessary to trigger the cyclo-addition reaction of cinnamoyl groups to achieve the healing purpose.
In an ideal self-healing system, the healing behavior should be completely autonomous without any manual intervention. In this sense, the first self-healing material was developed by White et al.9 in 2001. Dicyclopentadiene (DCPD) monomer was stored in poly(urea-formaldehyde) (PUF) microcapsules, which were evenly distributed in a polymer matrix. The self-healing function was realized via the polymerization of DCPD when the monomer contacted predispersed Grubbs’ catalyst particles. The whole process did not require any detection or manual intervention to initiate the repairing behavior and thus was deemed as self-healing. To date, a number of self-healing materials have appeared, and the progress on this topic can be found on many review papers.5,10–13
The concept of self-healing has been recently proposed for anticorrosion applications, and this novel approach has become a promising direction to develop anticorrosive coatings. Usually degradation of coating starts from the formation of microcracks. Microcracks will rupture embedded reservoir where healing agents are stored, and the healing agents flow into the crack plane, undergoing healing reaction to rebond the cracks of the self-healing coating. In principle, capsules, hollow tubes and microvascular network can be used as reservoirs to store the healant in the development of self-healing coatings, but considering the restriction of the small thickness of coating, only microcapsules can be employed in a coating system. For example, Sauvant-Moynot et al.14 fabricated a self-healing anticorrosive coating-based on the microencapsulation of a water-soluble and self-curable epoxy eletrodepositable adduct. When the adduct came into contact with water, it would cure and deposit in the cracks to exhibit self-healing capabilities. When an electrochemical impedance spectroscopy (EIS) test was performed on a steel panel treated with such a coating, it was revealed that the self-healing coating greatly improved its anticorrosion ability. Suryanarayana et al.,15 Samadzadeh et al.,16 and García et al.17 also developed different self-healing anticorrosive coatings with different chemistry.
When microcapsules are incorporated into polymer to create self-healing anticorrosion function, the capsules should fulfill some strict requirement. For example, the capsules should be mechanically robust enough to remain intact during processing and transportation, but on the other hand, they should be able to be ruptured on demand.18 In addition, they should be stable and compatible with the healing species, host coating matrix, and practical environment where the final coating will be exposed to.
14.2 Design Strategies of One-Part Self-Healing Anticorrosive Coatings
14.2.1 Preparation of conventional self-healing materials
Self-healing materials can be generally classified into intrinsic and extrinsic systems based on their self-healing mechanism. Intrinsic self-healing materials usually take use of the reversible reaction of some functional groups of the materials. However, for intrinsic systems, usually external stimulus such as heat or light irradiation is required to trigger the healing reaction. For example, an oxetane-substituted chitosan precursor incorporating polyurethane (PU) was reported to exhibit good self-healing performance,19 but UV radiation was necessary to trigger the healing reaction. Extrinsic self-healing materials, on the other hand, realize the self-healing function by prestored healing agents. Healing agents are stored in reservoir that is distributed in the host polymer, and they will be released in response to external stimuli. Based on the reservoirs used, self-healing materials can be generally divided into three groups: microcapsule-based,20,21 hollow tube-based,4 and microvascular-based.22,23
Preparation of microcapsules-based self-healing materials involves microencapsulation, a technique of encapsulating core material into microcapsules. Microencapsulation has been applied in various fields such as agriculture, pharmaceuticals, foods, textiles, coatings, flavoring and fragrances, and adhesives for a long time,24 and it has also become one of the most important methods to design self-healing materials because it can be used for both bulk polymer composites and thin polymer coatings and the amount of healing agent can be monitored by tuning the microcapsules diameter. The healing function is usually realized by the polymerization reaction of a healing agent that is stored in microcapsules, and in most cases, a catalyst is necessary for triggering such a reaction. In a typical microcapsule-based system, the microcapsules containing healing agents are ruptured when cracks form in the host polymer matrix, and the encapsulated healing agents are hence released. Upon contact with catalyst, the healing agents undergo polymerization, forming a film to heal the cracks, as illustrated in Figure 14.2.10 The self-healing polymer may be based on a monocapsules basis, in which the healing agent is encapsulated and the catalyst is directly distributed in a host matrix. Alternatively, self-healing polymer may be based on a dual capsules basis, in which the healing agent and catalyst are both encapsulated in discrete capsules and the capsules are then randomly distributed in the host polymer matrix. Samadzadeh et al. reviewed the recent progress of microcapsules-based self-healing materials.25
Microencapsulation can be reached via a number of methods such as coacervation26,27 and interfacial polymerization, spray drying.28,29 Among the numerous methods for microencapsulation, interfacial polymerization and in situ polymerization are most common for self-healing materials development.
Interfacial polymerization usually takes place in a normal oil-in-water emulsion system, which is prepared by dispersing a water immiscible organic liquid (oil phase) into an aqueous continuous phase with the assistance of surfactant. In a typical interfacial polymerization reaction, one reactant of the polymerization and target core materials are contained in the oil phase, while the second reactant is in the aqueous phase. Polymerization reaction between the two reactants from different phases occurs to generate a polymer wall at the interface between the aqueous and organic phases, with core material being encapsulated.25,30–32 For example, Yang et al.33 reported the microencapsulation of isophorone diisocyanate (IPDI) by PU microcapsules via an interfacial polymerization. In this study, a toluene diisocyanate (TDI)-based PU prepolymer was prepared at first, which was mixed well with IPDI, the target core material, to form the oil phase. After the generation of oil-in-water emulsion system by adding the oil phase into aqueous continuous phase, addition of chain extender 1,4-butanediol initiated its polymerization reaction with the more reactive TDI prepolymer from the oil phase, while the less reactive IPDI was encapsulated as core material.
The IPDI-filled PU microcapsules had smooth surface as shown in Figure 14.3a. From Figure 14.3b, it is seen that the diameter and shell thickness of the obtained microcapsules were highly dependent on agitation rate during the reaction. When the agitation rate was 500 RPM, the prepared capsules had a diameter of about 413 μm. This is the first successful encapsulation of liquid-state reactive isocyanate monomer, whose high reactivity with water shows the potential for the development of one-part and catalyst-free self-healing system.
In situ polymerization has been one of the most popular methods to encapsulate active agents for self-healing application. Similar to interfacial polymerization, in situ polymerization also takes place in an emulsion system, but the reactants for the polymerization are from the same phase. The most well-known example of in situ polymerization for self-healing materials development is the synthesis of PUF microcapsules.
PUF is an excellent material that has been widely used in self-healing material. Microcapsules made from PUF are robust enough to withstand processing for self-healing materials manufacturing, and they can also be ruptured easily by propagating microcracks.34 Yuan et al.35,36 reported the preparation of PUF microcapsules containing epoxy resin through a two-step reaction. In the first step, urea-formaldehyde prepolymer was synthesized in aqueous phase, and in the second step, epoxy resin, the target core materials, was added into the obtained prepolymer solution to create an oil-in-water emulsion. The in situ polymerization of the urea-formaldehyde prepolymer yielded PUF microcapsules with epoxy resin encapsulated inside.
Brown et al.20 reported a single-step reaction for the preparation of PUF microcapsules through in situ polymerization. In the synthesis, urea, ammonium chloride, and resorcinol were added to ethylene maleic anhydride surfactant solution. After dissolution of these chemicals, the pH value of the solution was adjusted to 3.5 by adding sodium hydroxide and hydrochloric acid. DCPD liquid was added into the aqueous solution under stirring to create an oil-in-water emulsion system. A 37 wt.% formaldehyde solution was then introduced into the system. The in situ polymerization between urea and formaldehyde in the aqueous phase produced PUF microcapsules surrounding the DCPD oil droplets.
SEM images of the shell structure of the prepared PUF microcapsules as shown in Figure 14.4 revealed that the capsule shell was comprised of a smooth inner surface and a rough outer surface. The thickness of the inner surface was between 160 and 220 nm, and it was largely independent on synthesis parameters. The outer surface was an agglomeration of PUF nanoparticles. The microcapsules as prepared contained 83-92 wt.% DCPD, and the capsules diameter ranged from 10 to 1000 μm depending on agitation rate.
The method as described above has been used as a standard method to prepare PUF microcapsules. For example, Suryanarayana et al.15 reported the microencapsulation of linseed oil by PUF microcapsules via an in situ polymerization following a very similar procedure, and García et al.17 also similarly encapsulated a water reactive silyl ester for self-healing purposes.
Poly(melamine-formaldehyde) (PMF) microcapsules are also extensively used for self-healing application, and they are advantageous over PUF microcapsules in their mechanical properties, stability, and chemical resistance.37–39 PMF microcapsules can also be prepared via in situ polymerization following a procedure similar to that for PUF microcapsules.37,38
Interfacial and in situ polymerization reactions have been extensively explored to synthesize single-component capsules. However, when microencapsulation is applied to develop self-healing materials, microcapsules made from single component may not meet the complicated and strict requirement in practical application.40 For example, PU microcapsules possess excellent mechanical and thermal property, but they are likely to agglomerate due to the cross-linking of residual isocyanate functional groups, and hence the distribution into host matrix may be problematic.41 PUF microcapsules also have good mechanical property, and they can be easily distributed into host polymer, but they are unable to withstand high temperature, which is usually required in materials processing. To overcome these limitations, researchers developed techniques to prepare multilayered microcapsules to encapsulate active agents for self-healing applications. Microcapsules composed of two or more materials possess the combined advantage while avoiding the disadvantage of each single material, and hence they are superior.
Multilayered microcapsules can be prepared via a procedure combining interfacial polymerization and in situ polymerization. For example, Caruso et al.42 demonstrated the microencapsulation of ethyl phenylacetate by PU-PUF double-layered microcapsules by combining interfacial and in situ polymerization in a single batch process. Compared with the normal in situ polymerization approach as described elsewhere for making PUF capsules,20 herein after the addition of urea, ammonium chloride, and resorcinol into the surfactant solution and pH adjustment, a mixture of the core material and a PU prepolymer, rather than the core material alone, was added into the above aqueous solution to generate an oil-in-water emulsion system. The rest of the operation including the further addition of formaldehyde solution, temperature control, and product collection were all carried out the same as those for normal in situ polymerization. As shown in Figure 14.5, two distinct phases were observed on the atomic force microscopy images of the prepared microcapsules, and the shell consists of a PU inner shell and a PUF outer shell. As a control, a single-layered PUF capsule was also examined for comparison. Thermogravimetric analysis (TGA) of the prepared capsules revealed that the PU-UF capsules exhibited much better protection to the core materials compared with single-layered PUF capsules.
Aside from the microcapsules-based system as discussed above, a hollow tube-based system is another important type of self-healing material. Healing agents are stored in hollow tubes, which are embedded into polymeric host matrix to induce the desired self-healing function.43–49 For self-healing polymer composite based on hollow tubes, the embedded tubes serve both as the reservoir for healing agent storage and as reinforcement fillers.50
Compared with microcapsules, hollow tubes can store more healing agents and hence the healing performance will be better.12 The mechanism of self-healing behavior of hollow tube-based self-healing materials can be illustrated in Figure 14.6. Once cracks form in a polymer composite, the embedded hollow tubes will be ruptured to release the stored healing agents into the cracks site. The polymerization reaction of the healing agents will rebond the cracks and therefore realize the self-healing function.
Glass fibers are the most important material used as hollow tubes for self-healing materials. It is primarily due to the fact that glass fibers are inert to most healing agents and host polymer matrix. In addition, they can be broken easily by cracks formed in the host matrix to release the containing healing agents.51 On the contrary, it may be difficult to sense the crack in tubes made from other materials such as copper or aluminum, and therefore the healing behavior will not be triggered.6 Compared with microcapsules-based self-healing system, a hollow tube-based system requires special attention to the compatibility of the healing agents with the hollow tubes, and hence various properties of the healing agents, such as their surface wettability, viscosity, chemical reactivity, and so on, should be comprehensively considered.11 For example, healing agents that are too viscous may be unable to be introduced into the tubes, and even after the self-healing system is eventually developed, the healing agents are less likely to flow into the cracking sites where healing behavior is desired. In addition, the wettability of the healing species in the tubes is also a critical concern.
The research on microcapsule and hollow tube-based self-healing materials has achieved great advancement. Nevertheless, neither of these materials is able to repair the same location more than once.12 For example, in a microcapsule-based materials, when the healing process is triggered due to the capsules rupture, the healing agent will be depleted, leaving voids in the host matrix. Then when another healing is desired at the same location, no healing agent will be available to meet the purpose. The same result will happen to a hollow tube-based self-healing system, because the healing agents stored are still finite even though they are in larger volume compared with those in microcapsules. Moreover, in most cases the second fracture will tend to occur at the same location as the initial one, because it is hard for the materials to reach a complete recovery in either the material integrity or the mechanical property. To overcome this problem, microvascular was proposed as an alternative approach for self-healing materials development.52–55
In a microvascular-based self-healing system, a three-dimensional vascular network is embedded into the polymer host matrix, and healing agents are stored in the network. Once damage or cracks in the host matrix rupture the vascular, the healing agents will be transported to the damaged site via the vascular network. Figure 14.7 illustrates the mechanism of the healing process in skin and in a microvascular-based self-healing system. It can be seen that in such a self-healing system, the healing process is a better imitation to the nature.5 The healing agents can be supplied from other places via the interconnected network, and their supply will not be exhausted at one location, hence the microvascular-based system can achieve multiple healing.56
Toohey et al.23 demonstrated the self-healing performance of a microvascular-based self-healing material. In such as system, a rectangular three-dimensional microvascular network was integrated throughout a polymer substrate via a direct-write assembly of a fugitive organic ink,57 and an epoxy layer modified by different percentage of Grubbs’ catalyst particles was then deposited on the substrate. The interconnected microvascular channels (d = 200 μm) were filled with DCPD as healing agents. Once crack formed on the top epoxy layer, the underneath channels were broken, and the stored DCPD was released from the vertical channels to the crack sites. Upon contact with the embedded Grubbs’ catalyst, the polymerization of DCPD formed a film to repair the crack. In the healing process, the horizontal channels severed the passage through which DCPD could be transported throughout the whole network to ensure the system was always filled with healing agents, and therefore enabled the multiple healing. The healing efficiency of the manufactured system was assessed, and the test showed as high as 70% recovery of the fracture toughness after damage in a four-point bending protocol.
14.2.2 Design of one-part self-healing anticorrosive coatings
In most existing self-healing materials, the healing reaction is based on a two-part reaction. For example, Toohey et al.54 reported a self-healing material that is based on the two-part reaction between epoxy resin and hardener. Otherwise, the healing reaction is based on the polymerization reaction of a single component, but it requires the presence of a catalyst9,23 or other external intervention such as heat or UV irradiation.19 From the mechanism of corrosion as discussed above, it is seen that the corrosion of metal always involves moisture or water, and this is a feature that can be employed to develop one-part self-healing anticorrosive coatings. In this case, the prestored healant is able to react with moisture to realize the self-healing anticorrosive function.
In principle, the approaches for the self-healing materials development as discussed above are also valid for the preparation of one-part self-healing anticorrosive coatings. Nevertheless, if the constraint of the small thickness of the coating is taken into consideration, microencapsulation is the most suitable method. For the development of one-part self-healing anticorrosive coatings, selection of the healing agent is a crucial issue. First of all, the healing agent should be curable under the circumstance that the materials are exposed to when the healing behavior is expected. Secondly, the healing agent should have good flowability so that it is able to easily flow into the cracks once microcracks are formed in a coating.
Isocyanate monomer, for example, is a good candidate for one-part self-healing anticorrosive coatings. Isocyanates are reactive with water or moisture to produce carbamic acid, which is unstable and tends to decompose to an amine which in turn reacts with isocyanate to form urea or biuret or other polymeric products, as demonstrated in Figure 14.8.
Organic silane is another example that has potential as a one-part self-healing anticorrosive coating. Upon contact with moisture, organic silane will undergo polycondensation to form a silane film as shown in Figure 14.9.
One-part self-healing anticorrosive coatings are advantageous over conventional self-healing systems. The preparation of the one-part coatings is much simpler and more time efficient. Meanwhile, only one set of microcapsules is synthesized and incorporated to develop the coatings, so the impact of microcapsules to the host polymer matrix is minimized. In addition, the use of catalyst becomes unnecessary due to the self-curable property of the healant in such coatings, which is of considerable economic importance. In the next section, the design and advantage of one-part self-healing corrosive coatings will be illustrated by two examples.
14.3 Examples of One-Part Self-Healing Anticorrosive Coatings
14.3.1 Diisocyanate-based one-part self-healing anticorrosive coating
14.3.1.1 Microencapsulation of diisocyanate via interfacial polymerization
Isocyanates are potential candidates for the development of one-part self-healing anticorrosive coatings because of their self-curable properties. Nevertheless, the high reactivity of isocyanates brings difficulty for their processing and encapsulation. Previous research on the encapsulation of isocyanate has been mainly restricted to its solid state or blocked form. For example, Yang et al. and Rawlins et al. reported the encapsulation of a blocked form of polyisocyanate by polystyrene nanocapsules via emulsion polymerization,58–60 and Walther et al. demonstrated that the microencapsulation of a solid-state naphthylene-1,5-diisocyanate by protective materials such as polystyrene, chlorinated rubber, and polyvinyl butyl ether that were inert to the encapsulated isocyanate.61 To date, very few public documents have appeared related to the encapsulation of liquid-state isocyanate monomers.
Yang et al.33 for the first time reported the microencapsulation of a liquid-state isocyanate monomer. The IPDI monomer was encapsulated by PU microcapsules via an interfacial polymerization approach in an oil-in-water emulsion system. As illustrated in Figure 14.10, the interfacial polymerization reaction between the TDI prepolymer and 1,4-butanediol produced the PU shell structure of the final microcapsules. However, the synthesis was based on a self-prepared urethane prepolymer, and thus the preparation would be quite tedious. In addition, the reactivity of IPDI is not very high, which is unfavorable for the self-healing coating because it is desirable that the healing reaction occurs in a short period of time to ensure the instant healing function.
Following a similar method, a more reactive liquid-state diisocyanate monomer, hexamethylene diisocyanate (HDI), is also successfully encapsulated into PU microcapsules through an interfacial polymerization process shown in Figure 14.11. In this synthesis, a commercial MDI prepolymer was employed as the reactant to construct the shell.
In the synthesis, the MDI prepolymer Suprasec 2644 was dissolved in HDI to yield an oil phase, which was dispersed into gum arabic surfactant solution to create an oil-in-water emulsion. When 1,4-butanediol was introduced, the reaction between the hydroxyl functional group from the aqueous phase and the isocyanate functional group from the oil phase occurred to create a PU membrane surrounding the oil droplets. The reaction mechanism of PU formation is very similar to that shown in Figure 14.10, while the only difference is that the TDI prepolymer is replaced by an MDI prepolymer. The diol in the aqueous phase diffused across the initial membrane to further react with the isocyanates and resulted in the membrane increment.62 The MDI prepolymer was much more reactive than HDI, and hence the primary reaction was between 1,4-butanediol and the MDI prepolymer to form the shell structure, while the relatively less reactive HDI liquid was encapsulated as core material to produce the final microcapsules. All of the materials used in the synthesis were commercially available, so it was seen that the microencapsulation of HDI was realized in a much more simplified procedure compared with the encapsulation of IPDI, as reported before.33 In a typical run of synthesis, the oil phase was composed of 2.9 g MDI prepolymer and 8.0 g HDI, the concentration of the gum arabic surfactant solution was 3 wt.%, the agitation rate was 500 RPM, and the temperature of reaction was 40 ºC. At such a condition, the yield of the capsules was about 70%, and the resultant microcapsules had an average diameter of 86.5 μm and a shell thickness of about 6.5 μm.
Given that the synthesis of the capsule shell wall was not a strict stoichiometric reaction, excess 1,4-butanediol was used to ensure the Suprasec 2644 was completely consumed. The yield of the synthesis is calculated simply as below:
where Wcap is the mass of the collected microcapsules after drying, and Wpre-p, Wdiol, WHDI are the masses of the MDI prepolymer, 1,4-butanediol, and HDI, respectively. This is a rough method for estimating the yield of the synthesis.
Based on the calculation described above, at a 500-RPM agitation rate, the typical yield was calculated to be 65-74% and slightly varied with reaction parameters. It was found that the yield was lower at higher agitation rates, and it declined to ~ 54% when the agitation rate was raised to 2000 RPM.
The formation of microcapsules is quite fast in the synthesis due to the high reactivity of the MDI prepolymer. Optical microscopy examination during the reaction course showed that microcapsules formed within 15 min after the addition of 1,4-butanediol. However, a minimum reaction time (MRT) was necessary to ensure the shell of the capsules was mechanically strong enough to withstand the following filtration and further processing. Finding the MRT of the synthesis is very important both for saving time and for quality control of the microcapsules shell. To determine the MRT, products were sampled from the emulsion solution at 10-min intervals until dispersed dry microcapsules could be collected by filtration. Although microcapsules were observed at quite an early stage, the capsules formed before MRT were found to collapse to yield viscous bulk polymer during the filtration. As shown in the Table 14.1, the required MRT reduced steadily with the increase of reaction temperature. It meant that the capsule shell growth was more rapid at a higher temperature, and this is consistent with the fact that the rate of polymerization reaction for PU formation is positively dependent on temperature.
14.3.1.2 Chemical constituent of microcapsules
The chemical constituent of the resultant microcapsules was characterized by FTIR. As a comparison, complete capsules together with pure grades of HDI, MDI prepolymer, shell, and core materials were investigated. As illustrated in Figure 14.12, the characteristic signals at 2267.6 cm− 1 (
NCO stretch) and 2929.5 cm− 1 (CH2 stretch) from the full microcapsule reveal that the resultant capsules contained large amounts of HDI. Meanwhile, the NCO signal (2267.6 cm− 1) is not observed from the spectrum of capsule shell. It means that the HDI was not absorbed in the microcapsule shell. In addition, it is seen that the spectrum of core materials is almost the same as that of pure HDI, indicating that the core material was mainly composed of pure HDI. Therefore, it can be concluded that the HDI was successfully encapsulated into the PU microcapsules. In addition, the absence of the NCO signal in the spectrum of the pure shell indicates that the prepolymer chains were extended to form the PU bulk shell. Based on these observations, it is logical to conclude that polymerization of the MDI prepolymer produced the PU microcapsules shell, and HDI was encapsulated as a core material in the synthesis.
14.3.1.3 Morphology and diameter of microcapsules
The synthesized microcapsules have a spherical shape with a smooth surface as shown in Figure 14.13. It was found that the inner surface of the capsules was not as smooth as the outer surface. Meanwhile, the coarse cross-section of the capsules indicates that the capsules shell may contain some voids. The average shell thickness of capsules was in the range of 1.1-12.48 μm when the agitation rate varied in the range of 300-2000 RPM.
In the development of self-healing materials through microencapsulation, control of capsule diameter is critical because the diameter greatly influences the self-healing performance,63 and in some conditions, only the capsules with a certain range of diameter are suitable. The microcapsules diameter is influenced by a combination of several factors, including the geometry of the mixing device, viscosity of the reaction media, surfactant concentration, agitation rate, temperature, and so on, while the agitation rate is the most important one. As illustrated in Figure 14.14, when all other variables remained the same during microcapsules synthesis, higher agitation speed resulted in smaller microcapsule size, and the average diameter possessed a linear relationship with the agitation rate in double-logarithm coordinates. This result is also found in previous studies.20,33
14.3.1.4 Thermal property and core fraction of microcapsules
The TGA weight loss curves of microcapsules synthesized at 500 RPM along with pure HDI and capsule shell materials as a function of temperature are shown in Figure 14.15. It can be observed that microcapsules experienced significant weight loss by 180 °C. This weight loss is in good agreement with that of HDI and reveals the successful encapsulation of HDI within the microcapsules. The decomposition of shell materials started from about 240 °C. Figure 14.15 also illustrated the derivative of the weight loss curve of microcapsule. The evaporation process of HDI in the first peak and decomposition process of shell in the peaks after 240 °C are apparent on the curve of the full microcapsules.
The HDI content or the core fraction in the prepared microcapsules was determined from the peak width of the derivative curve of full microcapsules. From Figure 14.15 it is seen that the HDI content of the microcapsules was about 62 wt.% at 500 RPM based on the weight loss of HDI.
14.3.1.5 Preparation of HDI-based self-healing anticorrosive coating
Diisocyanate-based one-part self-healing anticorrosive epoxy coating was prepared by dispersing a certain weight percentage of the synthesized HDI microcapsules into epoxy resin EPOLAM 5015 at ambient temperature, followed by mixing hardener EPOLAM 5014. The mixture was degassed for 20 min and then applied on the pretreated substrates by an adjustable film applicator. The coating was cured at ambient temperature in open air for 24 h.
14.3.1.6 Accelerated salt immersion corrosion test
An accelerated salt immersion test was performed to the HDI microcapsules-based epoxy coating. In this test, the average diameter of the microcapsules was about 100 μm, and 10 wt.% of microcapsule were mixed into the epoxy resin to yield the final coating. The prepared HDI-based coating was manually scribed following the standard method of ASTM D1654 and then exposed to 10 wt.% NaCl aqueous solution for 48 h. As shown in Figure 14.16, when the scribed specimen coated with the neat epoxy coating was immersed in salt solution, severe corrosion was observed at the scribed area. On the contrary, the specimen coated with the self-healing coating was free of corrosion after immersion. Such a difference clearly demonstrates that the epoxy coating modified by HDI-filled microcapsules exhibited excellent anticorrosion property.
In order to reveal the mechanism of the anticorrosion function of the microcapsules-based coating, the scribed area of the coating was examined by SEM after immersion. Figure 14.17 illustrates the scribed locations of the HDI-based coating and the blank epoxy coating when the coatings were immersed in salt solution for two days. After the application of scratches on the coatings, the underlying steel substrate was exposed to corrosive salt solution and would be corroded. Comparing Figure 14.17a with b, it is clearly seen that some new materials were created in the scribes of the HDI-based epoxy coating after immersion. It means that the crack on this coating was healed autonomously. This is self-healing because such a healing process did not involve any manual intervention. Due to such a self-healing activity, the scribes were sealed and the underneath steel was separated again from external corrosive environment to display corrosion protection function. The materials generated in the crack should be the product between HDI, released from ruptured microcapsules, and water from the environment. As a comparison, it could be seen from Figure 14.17c and d that the crack of the control specimen was not sealed, and therefore the corrosive salt solution can still directly attack the steel.
From the discussion for the HDI-based anticorrosive coatings as described above, it is seen that the excellent corrosion protection function of the microcapsules- based coating was realized through a self-healing/sealing mechanism. When the microcapsules modified coatings were scratched, the healing species HDI would flow from the ruptured microcapsules into the cracks. Upon contact with water or moisture, the healing species would then undergo a healing reaction in the form of polymerization, crosslink, or condensation to create a film within the scribes, separating the metal substrate from the corrosive external environment. As a result, the corrosion of the substrate was hindered. On the contrary, for the control sample, the underlying metal substrate would be severely corroded because it was directly exposed to corrosive salt solution once the coating was scratched. The healing reactions of the HDI-based self-healing anticorrosive coatings are demonstrated in Figure 14.18. HDI monomer will react with water to produce an amine, which further reacts with HDI to form a polyurea film. However, it is noteworthy that in a real situation, the actual healing reaction and the chemical constituent of the newly formed film may be much more complicated than those proposed here.
14.3.1.7 Salt spray test
In order to better assess the anticorrosive performance of the HDI microcapsules incorporated epoxy coating, a comprehensive salt spray test was performed following the standard test method of ASTM B117.
The HDI microcapsules-based epoxy coatings were manually scribed following the standard method of ASTM D1654 and placed in a salt spray chamber, while the corrosion behavior of the specimens was monitored. The influences of three parameters, the average diameter of HDI microcapsules, weight fraction of microcapsules in coating, and the thickness of final coating, to the corrosion protection ability of the coatings were investigated. For each parameter three values were taken, so overall 27 formulations of coatings were prepared for the test. In addition, three blank epoxy coatings were prepared at different coating thicknesses as control. All the formulations of the coatings for the salt spray test are summarized in Table 14.2, and they are labeled in the format of SP-D-Ф-H, while the three control samples were labeled as SP-Blk-H.
Table 14.2
Formulations of HDI Microcapsules-Based Epoxy Coating
| D (μm) | Ф (wt.%) | H (μm) | Name of Samples |
| 100 ± 31.6 | 10 | 400, 300, 200 | SP-100-10-400; SP-100-10-300; SP-100-10-200 |
| 5 | SP-100-5-400; SP-100-5-300; SP-100-5-200 | ||
| 2 | SP-100-2-400; SP-100-2-300; SP-100-2-200 | ||
| 50 ± 17.2 | 10 | SP-50-10-400; SP-50-10-300; SP-50-10-200 | |
| 5 | SP-50-5-400; SP-50-5-300; SP-50-5-200 | ||
| 2 | SP-50-2-400; SP-50-2-300; SP-50-2-200 | ||
| 30 ± 9.4 | 10 | SP-30-10-400; SP-30-10-300; SP-30-10-200 | |
| 5 | SP-30-5-400; SP-30-5-300; SP-30-5-200 | ||
| 2 | SP-30-2-400; SP-30-2-300; SP-30-2-200 | ||
| Control (pure epoxy) | 0 | SP-Blk-400; SP-Blk-300; SP-Blk-200 |
Note: Naming policy of samples is used as SP-D-Ф-H where SP means salt spray test, D is average diameter of microcapsules, Ф is weight fraction of microcapsules in coating, and H is thickness of coating.
The specimens were exposed to salt fog for two months in a monitored salt spray chamber. As shown in Figure 14.19, it is seen that after exposure, the blank epoxy coatings (SP-Blk-400, SP-Blk-300, SP-Blk-200) with three different thicknesses developed corrosion along the scribes. Among the other 27 microcapsules modified coatings, coatings SP-100-10-400 and SP-100-10-300 showed no corrosion, indicating their excellent corrosion protection property. In the meantime, slight corrosion was also observed on coatings SP-100-10-200, SP-100-5-400, SP-100-5-300, and SP-100-5-200. It means that these four formulations could provide some corrosion protection, but the performance was slightly worse compared with formulations SP-100-10-400 and SP-100-10-300. For the other formulations, severe corrosion was observed along the scribes, indicating that the protection of these coatings to steel substrates was quite poor. The results reveal that the HDI microcapsules-based epoxy coatings exhibit good corrosion protection function to metal substrate if they are properly formulated.
It was found from the salt spray test that the growth of corrosion was related to exposure time. Figure 14.20 demonstrated the process that corrosion developed on three coatings (SP-100-10-400, SP-50-10-400, and SP-30-10-400) when the salt spray test proceeded. It is seen that corrosion started to appear on SP-50-10-400 and SP-30-10-400 from the first week, and more and more rust was observed when the exposure time of the specimens in salt fog increased. This is a reasonable result because corrosion is an electrochemical process, and it proceeds with time.
Figure 14.20 also revealed the influence of microcapsules size on the corrosion protection ability of the microcapsules-based coating. The three formulations, SP-100-10-400, SP-50-10-400, and SP-30-10-400, differed only in the average diameter of the HDI microcapsules, while the other two parameters of self-healing coatings remained the same. It is seen that specimen SP-100-10-400, in which the diameter of HDI microcapsules was 100 μm, was almost free from rust, while specimen SP-30-10-400, in which the diameter of microcapsules was 30 μm, showed most serious corrosion. The extent of corrosion of specimen SP-50-10-400 was between that of these two specimens. Actually the comparison of other specimens also showed a similar trend, in which the coatings exhibited best corrosion resistance performance when the diameter of incorporated microcapsules was 100 μm, followed by the 50 μm specimens and the 30 μm specimens showed worst corrosion protection. This results indicate that the HDI microcapsules-based epoxy coating afforded better corrosion protection towards a steel substrate when the diameter of microcapsules was bigger, given other parameters of the coating remained. This result is in good agreement with that reported in another publication.63
The corrosion protection performance of the HDI-based coating was also affected by the weight fraction of microcapsules in the coating and the coating thickness. Figure 14.21 shows nine specimens after exposed to salt fog for two months. The diameter of the HDI microcapsules was at 100 μm for the nine specimens, while the weight fraction of microcapsules and coating thickness varied for each formulation. If we compare the specimens in each row, in which the weight fraction of microcapsules was constant, it is seen that, generally, more rust was observed on the specimens with smaller coating thicknesses. If the specimens in each column, in which the coating thickness was the same, were compared, it could be found that the corrosion was more severe when the weight fraction of capsules declined. A similar trend can also be found from other specimens. The results imply that higher microcapsule content in the coating as well as larger coating thickness is favorable for the HDI-based self-healing anticorrosive coating. It should be pointed out that although the edge and backside of the substrate was protected by water-resistant tape during the salt spray test, some corrosion still occurred at the noncoated part. That is why much of yellowish rust was seen on the specimen although the real corrosion was much less as for specimen SP-100-5-200 in Figure 14.21.
14.3.1.8 Influence of parameters on anticorrosive performance
The salt spray test of the HDI-based self-healing anticorrosive coating reveals that the microcapsules size, weight fraction of microcapsules in coating, and the coating thickness all significantly influenced the anticorrosion performance of the prepared coating. Generally, larger microcapsules size, higher microcapsules content, and thicker coating would afford better corrosion protection. For microcapsules-based self-healing polymer materials, the self-healing performance is influenced by factors such as the content of microcapsules in the polymer and diameter of microcapsules.64 Rule et al. pointed out that for a scratched self-healing material, the self-healing performance was directly related to the amount of healing agents that were available for delivery per unit scratch area.63 Based on a simplified model, an equation was proposed to explain the proportional effect of the microcapsules weight fraction and diameter on the self-healing performance of a microcapsules-based self-healing material. Mookhoek et al.65 also developed a model to predict influences of capsule diameter, aspect ratio, and concentration, as well as the crack opening distance on the self-healing performance. It has been revealed that the anticorrosive function of the scribed HDI-based coating was mainly realized through a self-sealing mechanism. Hence, the influences of these three factors on the anticorrosive property of the coating are discussed in terms of the self-healing behavior.
The discussion below is based on several assumptions: (1) the microcapsules with uniform diameter are evenly distributed in the coating matrix; (2) the fill content of each microcapsule is the same; (3) the shell of the microcapsules is negligible; (4) when a scratch forms in the coating, all of the microcapsules located at the scratch plane are ruptured; (5) all of the encapsulated healing agents of ruptured microcapsules will freely flow into the scratch; and (6) the healing species will spread within the scratch.
Consider a rectangle microcapsules-based coating is penetrated by a planar scratch as shown in Figure 14.22. Microcapsules with uniform diameter (d) are randomly distributed in the coating matrix. When a planar scratch penetrates the coating, all of the microcapsules lying in the scratch will be ruptured to release the healing species.
The number of microcapsules that are ruptured (n) is:
where N is the total number of microcapsules in the coating; P is the probability that the center of a microcapsule lies within the rupture zone of the scratch plane.
Because the microcapsules were assumed to be distributed evenly in the coating, the probability is:
where A is the area of the scratch plane; d is the diameter of the microcapsules; M is the mass of the coating; ρ is the density of the coating.
The total number of microcapsules in the coating can be calculated as:
and the area of the scratch plane is:
where Ф is the weight fraction of microcapsules in the coating; m is the mass of one microcapsule; H is the thickness of coating; L is the length of the scratch.
Combine Equations (14.2)–(14.5), the number of microcapsules ruptured by the scratch is calculated as:
In a typical self-healing material, the amount of the delivered healing agents has to be normalized by the area of the scratch plane because it is expected that the entire scratch plane is reconnected and rebonded.63 Nevertheless, for corrosion protection purpose, the basic requirement is that the exposed substrate is covered by the healing agents in the whole length of the scratch with a certain width, while the agents do not have to completely fill the entire scratch depth. Therefore, the amount of the delivered healing agents is normalized by the scratch projection area as below:
where t is the width of the scratch.
For a given microcapsules-based coating, the density of the coating (ρ) is basically determined by the coating matrix itself if the fraction of microcapsules is not too high, and hence it can be considered as a constant. From Equation (14.7), it is seen that the amount of healing species available at the scratched sites is proportional to the weight fraction of microcapsules (Ф), diameter of microcapsules (d), and thickness of coating (H). In addition, according to the discussion in last section, the anticorrosion function is realized through a self-healing mechanism, and better self-healing property indicates better anticorrosion property. Therefore, the corrosion protection performance is accordingly also positively related to the microcapsules diameter, microcapsules weight fraction, and coating thickness. This conclusion is consistent with the results as observed in the salt spray test. Furthermore, although the effect of the scratch width to the corrosion protection function of the coating was not investigated in our study, it is seen from Equation (14.7) that the corrosion protection function of the microcapsules-based coating should be inversely related to the width of the scratch (t). It suggests that a wider scratch is more difficult to be self-repaired by the coating itself.
14.3.2 Organic silane-based one-part self-healing anticorrosive coating
14.3.2.1 Microencapsulation of perfluorooctyl triethoxysilane via in situ polymerization
Inorganic silicates have been used for self-healing and corrosion protection applications for many years.66–68 For example, Aramaki et al. used sodium silicate in the preparation of a highly protective and self-healing film, which displayed good corrosion suppression effect to the zinc surface.67 The inorganic silicate-based corrosion-resistant coating mainly makes use of the deposition of silicates to realize the corrosion suppression function. But inorganic silicates are usually directly mixed into the coating matrix, and hence they may be susceptible to interaction with environment leading to problems in corrosion protection during long-term service. Organic silane molecules tend to hydrolyze in wet environments and crosslink to form a solid film, and such a property indicates their potential as a novel healing agent for a one-part and catalyst-free self-healing additive to corrosion-resistant coatings. To date, the research on organic silanes for self-healing materials remains largely unexplored, and only a few publications have appeared.17,31,32 Braun and coworkers reported the polydimethylsiloxane-based microcapsules applied for self-healing coatings.31,32 A mixture of hydroxyl end-functionalized polydimethylsiloxane and polydiethoxysiloxane was directly phase-separated or encapsulated and then dispersed in epoxy matrix, and the yielding coating exhibited good self-healing ability for corrosion protection. However, organo-tin catalyst was necessary for such self-healing systems. Another study examined microencapsulation of self-synthesized silyl ester for self-healing coatings.17 Octyldimethylsilyloleate, a silyl ester, was synthesized and encapsulated into PUF microcapsules, which were then incorporated into epoxy to produce a self-healing coating, and the excellent corrosion-resistant property of the prepared coating was demonstrated via self-healing. But the synthesis of silyl ester increased the complexity and cost to the procedure.
A novel one-part self-healing anticorrosive polymer coating is fabricated based on the microencapsulated an organic silane, that is, 1H,1H′,2H,2H′-perfluorooctyl triethoxysilane (POTS). There are several advantages of this coating system over the existing self-healing materials: (1) POTS is able to hydrolyze in wet environment to form a silane-based film and, as mentioned, such ability indicates that the use of catalyst can be avoided in the end self-healing product. (2) The newly formed film from hydrolysis and polycondensation of POTS and water is hydrophobic.69 Such a special wetting property will serve to repel aqueous electrolyte solution away from metal and hence provides further corrosion protection to metal substrate.17 (3) POTS is commercially available and hence the preparation of self-healing coating will be more convenient and time efficient for mass production.
PUF microcapsules containing POTS as core materials were synthesized through an in situ polymerization reaction in an oil-in-water emulsion system.20 At ambient temperature, 50 ml of deionized water and 12.5 ml of 2.5 wt.% aqueous solution of ethyl maleic anhydride copolymer were mixed in a 500 ml beaker. Under mechanical agitation, 1.25 g urea, 0.125 g ammonium chloride and 0.125 g resorcinol were dissolved in the solution. The pH of solution was raised from ~ 2.60 to 3.50 by dropwise addition of 1 M sodium hydroxide solution. One drop of 1-octanol was added to eliminate surface bubbles. 10 g of POTS liquid was dissolved in 5 g of toluene to form the oil phase, which was then slowly added into the above aqueous solution to generate emulsion. After stabilization for 10 min, 3.17 g of 37 wt.% aqueous solution of formaldehyde was added. The emulsion was covered and heated to 55 °C at a heating rate of 1 °C/min. After 4 h of continuous agitation, the stirrer and hot plate were switched off. The resultant microcapsules were filtered and washed with distilled water for several times. Microcapsules were collected for air-drying at ambient temperature for 48 h before further analysis. A typical procedure for the synthesis is outlined in Figure 14.23.
In the synthesis, ethylene maleic anhydride copolymer was surfactant, urea and formaldehyde were the monomers contributing to the microcapsule shell, and POTS was the encapsulated core material. Ammonium chloride served as hardeners while resorcinol was a further brancher involved in the polymerization reaction with urea and formaldehyde.70 Resorcinol also behaved to improve the resistance of PUF bonds to the influence of water in the synthesis.71 When the oil phase comprised of POTS and toluene was added into the aqueous solution containing surfactant, urea, ammonium chloride, and resorcinol, an oil-in-water emulsion was created. After the addition of formaldehyde solution, polymerization reaction between urea and formaldehyde occurred in the acidic aqueous phase to form a PUF membrane surrounding the oil phase, as illustrated in Figure 14.24. Further polymerization reaction produced PUF nanoparticles depositing on the initial membrane, leading to the shell increment to form the final microcapsules shell. The encapsulation was achieved at elevated temperature, and therefore most of toluene will evaporate away during the synthesis, leaving POTS as the main component of core materials in the final microcapsules, which will be illustrated by FTIR analysis in a later section.
The synthesis of the capsule was not a strict stoichiometric reaction, and most of the toluene would evaporate during the heated reaction process. The yield of the synthesis is therefore simply calculated by the ratio of the mass of collected microcapsules to the total mass of POTS, urea, ammonium chloride, and resorcinol, while the mass of toluene was ignored. Based on this calculation, the yield was around 67% in a typical synthesis, in which the agitation rate was 800 RPM. The yield of the synthesis is also dependent on the agitation rate during the synthesis, and it reduced to about 52% when the agitation rate was increased to 1500 RPM.
14.3.2.2 Chemical constituent of microcapsules
The chemical constituent of prepared microcapsules was determined from FTIR analysis. The FTIR spectra of complete capsules together with pure grades of POTS, shell, and core materials were shown in Figure 14.25. It is seen that the full capsules contained signals at 3330 cm− 1 (OH and NH stretching), 1620 cm− 1 (C
O stretching), and 1540 cm− 1 (NH bending),36,72 indicating the formation of PUF from the polymerization reaction between urea and formaldehyde. In addition, the signals at 1240 cm− 1 and 1190 cm− 1 (CF stretching) as well as the signals at 1100 cm− 1 and 1080 cm− 1 (SiOC stretching) implie the presence of POTS within the synthesized microcapsules. A simple comparison between the spectrum of full microcapsule and capsule shell clear shows that the POTS was absent in the shell part. In the meantime, it can be seen that the spectrum of capsule core is identical to that of pure POTS. Therefore, it is logical to conclude that POTS is not absorbed on the PUF microcapsules; instead, it is indeed encapsulated into the microcapsules as core materials. In other words, the synthesis successfully produced PUF microcapsules containing POTS liquid as the core material.
14.3.2.3 Morphology and diameter of microcapsules
The synthesized microcapsules were analyzed under SEM. As shown in Figure 14.26a, the produced microcapsules have a spherical shape. In addition, it is seen from Figure 14.26b that the microcapsule is composed of a rough outer surface and a relatively smooth inner surface. The inner surface is an impervious PUF shell, and the outer surface is the deposition of a large number of PUF nanoparticles.20 The cross-section of the capsule wall reveals that the thickness of the smooth inner shell is around 218 nm. The shell structure of the synthesized microcapsules is in good agreement with those described in other publications.16,20
The average diameter of the POTS microcapsule is dependent on the agitation rate during the synthesis. As illustrated in Figure 14.27, the average diameter of microcapsule would reduce from about 400 to about 40 μm when the agitation rate was raised from 300 to 1500 RPM. This observation demonstrates that the increase of agitation rate will result in smaller microcapsules. The main reason for such a correlation is that at higher agitation rate, finer oil droplets would form in the emulsion system due to the stronger shear force. Because the diameter of the final microcapsules was highly dependent on the oil droplet size in an in situ polymerization,73 in this case, the final microcapsules were accordingly smaller. Actually, as shown in Figure 14.27, the mean diameter of resultant microcapsules exhibits a linear relationship to the agitation rate in double-logarithm coordinates, and this result is consistent with other research.20,33
14.3.2.4 Thermal property and core fraction of microcapsules
The thermal property and core fraction of microcapsules were characterized by TGA analysis. The TGA weight loss curves of microcapsules synthesized at 800 RPM along with pure POTS and capsule shell material as a function of temperature are shown in Figure 14.28. It can be seen that microcapsules experienced significant weight loss from ~ 100 °C, which is in good agreement with that of pure POTS, revealing the successful encapsulation of POTS within the microcapsules. The decomposition of shell materials started from about 200 °C.
The derivative of the weight loss curve of microcapsules was also plotted in Figure 14.28. It clearly shows the evaporation process of POTS in the first peak and decomposition process of capsules shell in the peaks after 200 °C. From the peak width of the derivative curve, the core fraction of the microcapsules was determined to be around 60 wt.% at 800 RPM. The core fraction of the produced microcapsules could be increased by raising the content of POTS in the oil phase, but as will be discussed in a later section, the quality of microcapsules would be compromised.
14.3.2.5 Preparation of POTS-based self-healing anticorrosive coating
POTS-based self-healing anticorrosive epoxy coatings were prepared by dispersing a certain amount of synthesized POTS-filled PUF microcapsules into epoxy resin. During the synthesis, the agitation rate was carefully tuned so that the produced microcapsules had expected diameter. The synthesized microcapsules were mixed into epoxy resin EPOLAM 5015 at ambient temperature, followed by mixing hardener EPOLAM 5014. The mixture was degassed for 20 min and then applied on the pretreated substrates by an applicator. The epoxy resin was cured at ambient temperature in open air for 24 h.
Self-healing anticorrosive silicon elastomer was also prepared in a similar manner by dispersing a certain amount of the synthesized microcapsules into silicon resin Sylgard 184 at ambient temperature, followed by mixing hardener. The mixture was then degassed for 20 min and then applied on a pretreated substrate. The silicone elastomer was cured at ambient temperature in open air for 24 h.
14.3.2.6 Accelerated salt immersion corrosion test
An accelerated salt immersion test was performed to the POTS microcapsules-based epoxy coating to demonstrate the self-healing anticorrosive property. The prepared coating was manually scratched and then immersed into 10 wt.% NaCl aqueous solution to evaluate the anticorrosion performance. It can be seen from Figure 14.29 that the scratched area of the steel panel coated with self-healing coating was nearly fully free of corrosion after 48 h immersion in salt solution, while severe corrosion was seen in the control specimen. This considerable difference clearly demonstrates the excellent corrosion protection of the prepared coating towards steel panel.
The scratched areas of both the POTS-based epoxy coating and the control coating were inspected with SEM after the accelerated salt immersion corrosion test was completed, as shown in Figure 14.30. Comparing Figure 14.30a and b, it is obvious that the crack of the POTS-based coating was filled with newly formed materials after the immersion. The crack was in this way resealed to retard the diffusion of salt solution and thus protected the substrate from corrosion. On the contrary, from Figure 14.30c and d, it is seen that the crack of the control specimen was still open. Therefore, it could be concluded that the anticorrosion function of the coating originated from its sealing/self-healing property. However, as shown in Figure 14.30b, it is also observed that the sealing material of the POTS-based coating is not very dense and even, which may influence the corrosion protection performance of the coating, especially when a long-term effect is considered.
14.3.2.7 Corrosion protection performance of intact coating in HCl solution
The anticorrosion property of an intact POTS-based coating was demonstrated by a long-time corrosion test. POTS microcapsules-based silicone elastomer was coated on pretreated steel substrate and then exposed to 1 M hydrochloric acid solution for one month. As shown in Figure 14.31, apparently the POTS-based anticorrosive silicon elastomer shows much less corrosion than the control specimen after exposure.
When the coating was peeled off, the underlying steel substrate was examined under SEM, as shown in Figure 14.32. It is seen that prior to exposure to the HCl solution, both steel substrates have a smooth surface (Figure 14.32a and c). Nevertheless, after exposure, the surface of the anticorrosive specimen (Figure 14.32b) was much smoother than the control sample (Figure 14.32d). Given that the corrosion of metal will create rust and destroy the smooth substrate surface, this considerable difference indicates that the steel substrate coated with POTS-based anticorrosive coating was much less corroded during the exposure to corrosive HCl acid solution.
The corrosion-resistant property of the POTS-based coating was further demonstrated by EDX analysis. It is known that corrosion of steel will convert iron (Fe) into rust (Fe2O3). Hence, for a steel panel, within a given area, the ratio of element O to Fe (O/Fe) will increase when corrosion occurs, and a higher O/Fe value in turn implies more corrosion. To compare the extent of corrosion on the steel substrate that was coated with POTS-based anticorrosive elastomer or control elastomer during exposure to 1 M HCl solution, the steel substrates were analyzed by EDX to determine the O/Fe ratio of the substrate. As shown in Figure 14.33, prior to exposure to HCl solution, the steel substrates of both anticorrosive specimen and control specimen afforded only element Fe. It means that the substrates were nearly totally free of corrosion. After exposure, elements O, Fe, and Si were detected on both specimens, but it can be seen that the O/Fe value was dramatically different between two specimens. The O/Fe of the substrate coated with anticorrosive elastomer is 0.425 while that of the control sample is 1.148. It is admitted that the O/Fe difference between these two specimens alone is not very conclusive because the corrosion product may be quite complicated, and in the meantime, element O may come from other sources such as open air and silicone elastomer. Nevertheless, if the influence of other factors is assumed to be at the same level for the two specimens in the test, which is a reasonable assumption, the apparently smaller O/Fe value of the anticorrosive specimen from another aspect supports the conclusion above that the corrosion of substrate coated with the POTS-based anticorrosive coating is less compared with that of the control specimen. From the SEM and EDX analysis, it can be concluded that the integration of POTS microcapsules enhanced the corrosion protection performance of the silicon elastomer coating.
The anticorrosive property of the unscratched microcapsules-based one-part self-healing coating can also be explained by the healing reaction of the healing agents, that is, POTS that were prestored in the coating systems. Corrosion is an electrochemical process that requires the involvement of water. When metal is protected by a coating layer, although intact coating is able to effectively separate the substrate from a corrosive environment, corrosion is still inevitable in terms of long-term effect because water may penetrate the coating via diffusion to corrode the substrate. For the microcapsules-based coating, the prestored healing agents were reactive with water, so the water molecules may be trapped by healing species that are stored in the capsules during the diffusion process, as illustrated in Figure 14.34. It means that the rate that the water passing though the coating layer will be retarded, and as a result, the corrosion of the underlying metal substrate will accordingly be retarded.
14.4 Concluding Remarks and Perspectives
The past decade has witnessed great progress in the field of self-healing materials, and the concept of self-healing has been employed for corrosion control. Nevertheless, most existing self-healing anticorrosive coatings are two-part systems, or require catalyst or external intervention in order to trigger the self-healing process. One-part self-healing anticorrosive coatings can be prepared by incorporating one set of microcapsules containing suitable healing agents into a host coating matrix. Once the coating is scratched to rupture the incorporated microcapsules, the encapsulated healing agents will flow to the scratches. Upon contact with environment, the healing agents will undergo an autonomous self-healing reaction without any manual intervention. In this system, only one type of microcapsules was integrated into the coating matrix, making the fabrication of the self-healing coating much more time efficient. In addition, the self-healing function was realized from the reaction of the encapsulated healing species with its environment. Therefore, the use of a catalyst becomes unnecessary, which is of considerable economic importance.
The design of one-part self-healing anticorrosive is further demonstrated by two examples. A diisocyanate monomer and an organic silane were used as healants to develop separate one-part self-healing anticorrosive coatings, respectively. The excellent corrosion protection function of the created coatings was demonstrated by an accelerated salt immersion corrosion test and salt spray test. The anticorrosion function was realized through a self-healing mechanism, and the good self-healing performance of the coatings was quantitatively demonstrated by both DC electrochemical test and EIS measurements. The self-healing anticorrosive performance of the prepared coating was found to be related to the microcapsules diameter, microcapsules weight fraction in the coating and the coating thickness. Generally, larger microcapsules size, higher microcapsule weight fraction, and larger coating thickness are favorable to achieve better self-healing anticorrosive performance. The influences of the coating variables on the anticorrosive performance of the coating were explained based on a simplified model. It is revealed that the self-sealing behavior of the coating does not require manual intervention. In addition, in both self-sealing coatings only one healing agent is used without catalyst, so they are one-part and catalyst-free self-healing coating systems.
Although tremendous endeavors have been invested in the research of one-part self-healing anticorrosive coatings, and some products have been successfully fabricated, there are still a number of challenges to be addressed. For example, some healing agents such as diisocyanate monomers have shown great potential for the development of one-part self-healing anticorrosive coatings, but they may impose some environmental hazard, especially when long-term effect is considered. In addition, the one-part self-healing anticorrosive coating makes use of the reaction of the prestored healing species with its environment, but the high reactivity of the healant may cause some problem in the stability of the microcapsules as well as of the prepared coating.
The techniques for the development of one-part self-healing anticorrosive are not sufficiently mature for practical applications, and the research on this topic has drawn considerable attention. Here we propose our own perspectives on the research of one-part self-healing anticorrosive coatings, but the whole future picture of this field will definitely not be restrained in these aspects.
1. Microencapsulation of new healing species for self-healing anticorrosive applications. Finding new healing species with higher efficiency but lower environmental toxicity is always a very important part in the field of self-healing materials.
2. To improve the durability of the one-part self-healing anticorrosive coating. Durability is one of the most important requirements when a coating is developed for practical applications. In order to achieve high durability, the microcapsules have to be both chemically and mechanically stable. In addition, they should be able to provide good protection to the active healing agents so that the reactivity of the healing agents is maintained, and the capsules should also be compatible with the host coating matrix. Unfortunately, the durability of some one-part self-healing anticorrosive coatings is yet to be improved. For example, although the HDI-based coating exhibits excellent self-healing anticorrosive properties, it is found that the prepared microcapsule has a high permeability, and the stability of the microcapsules in organic solvents is still far from satisfactory. Hence, more efforts should be made to improve the barrier property of the microcapsules. PUF microcapsules have shown much better barrier property, so we may try to synthesize two-layer microcapsules to store the healing species. The synthesis may be realized through a two-step procedure.74 HDI-filled PU microcapsules are prepared first, which are then used as the template to synthesize the second layer of PUF microcapsules. In addition, it has been reported that the barrier property of microcapsules could be improved by adding nanoclays in the wall.75 We also can try to modify the current PU microcapsules with some nanoclays or other fillers to improve the stability of the microcapsules for filed applications. In addition, the permeability of the PU microcapsules is also determined by the chemical composition of the shell materials themselves. For example, it has been reported that the PU microcapsules made from MDI are more permeable than those made from TDI.76 Hence, we can try to redesign the synthesis by using different starting materials in order to improve the barrier property and stability of the microcapsules.
3. Different self-healing and corrosion tests have been performed to evaluate the quality of the prepared self-healing coating in the present study. But some of the test methods are only qualitative, while more quantitative results may be necessary to characterize the self-healing anticorrosive performance of the developed coating. In addition, the conditions under which these tests were conducted are still different from the real environment that the coatings may be exposed to. Hence, larger scale field tests are necessary before the coatings are considered for practical applications.