The corrosion susceptibility of Mg and its alloys in humid and aqueous environments limits their widespread application. The performance of several conversion coating technologies is reviewed herein. In addition to the assessment of performance based on the literature to date, thermodynamic analysis has been used to compare coating processes. Ongoing research reveals that the search for appropriate conversion coatings to replace toxic Cr-based coatings continues. It is seen that the variability between coatings of the same technology was greater than the averages between coating technologies. Additionally, coating pretreatment also appears to be more significant than the choice of coating technology itself.

Mg-based alloys present a unique corrosion protection challenge. Unlike other active light metals, such as Al and Ti, Mg alloys do not form a naturally passivating oxide film. Upon atmospheric exposure, Mg rapidly develops oxide/hydroxide/carbonate films.1–2 These films are porous, poorly bonded and inhomogeneous, and unable to provide satisfactory protection to the underlying metal against corrosion.3 

One of the most effective ways to improve the corrosion resistance of Mg alloys is to form a coating on the surface to isolate the base material from the environment, either by forming or adding some type of functional barrier layer. The coating may also form a good base for subsequent organic coatings.4 Several surface coating treatment techniques, including electrochemical plating (electroplating), conversion coatings, anodizing, hybrid coatings, and vapor-phase processes, have been developed.5 Of those, chemical conversion treatment is an effective, comparatively low-cost, and easily implemented method,6 which has been widely adopted in industrial processes.

In the conversion coating process, the substrate to be protected is immersed in a solution that reacts with the surface, altering the metal ion concentration and the pH at the metal-solution interface. The localized change in composition causes precipitation from the solution onto the surface of the substrate, forming the coating. Many factors influence the quality of conversion coatings on Mg alloys:

  • —composition of the base Mg alloy

  • —pretreatment processes

  • —composition of the conversion formulations

  • —posttreatments

  • —operational parameters, such as temperature, pH,7 immersion time, and degree of agitation8 

These can all influence the structure, composition, and performance of conversion coatings. As a result, a nonuniform coating with pores, cracks, or other defects may develop. The protective role of defective coatings will be compromised because the corrosive medium may penetrate easily to the metal surface through such defects. Therefore, one of the biggest challenges in the field of conversion coatings is to produce crack-free coatings with uniform coverage. This review is structured such that the classes of conversion coatings are presented and reviewed individually (this appears to be the most logical approach since the field has studied a number of different alloys assessed by widely varying means), followed by a discussion of relevant pretreatments, and finally a general discussion. In solution, inhibitors and sol-gel coatings are not covered. This review is intended to update the field studies since the prior review by Gray and Luan5 in 2002, reflecting the authors experience, with the intent to present a coherent review of what is a non-incremental area of research with mixed quality publications.

Chromate Coatings

Chromate coatings are an effective means of protecting reactive metals such as Mg, Al, and Zn. The chromating solution contains hexavalent Cr (from say, HCrO4) and is highly acidic. The substrate (M) is oxidized, and hexavalent Cr is reduced to trivalent Cr as follows:

formula

The trivalent Cr precipitates to form a coating of hydrated Cr and substrate oxides. Some hexavalent Cr becomes incorporated into the coating giving the coating a self-healing capability.5  Figure 1 shows the predominant phases(1) for Cr in the hexavalent (Figure 1[a]) and trivalent (Figure 1[b]) states. In the hexavalent state, Cr is soluble over the full range of concentrations and pH levels shown. When reduced to the trivalent state, Cr becomes insoluble except at low pH levels.

FIGURE 1.

Predominance diagrams for Cr in the (a) hexavalent and (b) trivalent oxidation states. These diagrams, and all others herein, were calculated using the MEDUSA software package.

FIGURE 1.

Predominance diagrams for Cr in the (a) hexavalent and (b) trivalent oxidation states. These diagrams, and all others herein, were calculated using the MEDUSA software package.

Close modal

The principle drawbacks of the chromating process are environmental issues. Hexavalent Cr is one of six substances listed in the European Union's RoHS (restriction of hazardous substances).(2) Consequently, products requiring RoHS certification cannot be used in the chromating process and, as such, are not discussed in greater detail here.

Phosphate Coatings

Metal phosphates are generally insoluble in neutral pH solutions but increasingly soluble in acidified solutions. The vast majority of Mg conversion coatings reported upon in the past decade has been for phosphate coatings; hence, this section is the largest of the reviews. Figure 2 illustrates how the predominance of Mg, Mn, Ca, and Zn in the presence of phosphate varies with pH and metal ion concentrations.(1) These diagrams were calculated for a solution containing 0.2 M total phosphate (equivalent to 20 g/L phosphoric acid [H3PO4]). For each of the metals shown, there is a boundary between an insoluble phosphate and soluble metal ions that extends diagonally from low pH, high metal ion concentration to high pH, low metal ion concentration. The location of this boundary varies according to metal species. Placing a metallic (M) item in an appropriately acidified phosphate-containing solution causes the following reaction at the interface:

formula
FIGURE 2.

Predominance area diagrams for solutions containing 0.2 M phosphate at concentrations of Mg, Mn, Ca, and Zn ions and pH.

FIGURE 2.

Predominance area diagrams for solutions containing 0.2 M phosphate at concentrations of Mg, Mn, Ca, and Zn ions and pH.

Close modal

This results in a local rise in pH and metal ion concentration, leading to the precipitation of metal phosphate on the surface. In the case of Mg, the phosphate species (magnesium hydrogen phosphate [MgHPO4·3H2O]) is poorly protective because it is insoluble only over a narrow range of conditions (Figure 2[a]).

To form a suitable phosphate-based conversion coating on Mg, other species must be present. These species serve several roles:

  • —enable the formation of coatings with lower solubilities in acid (and alkaline) solutions

  • —act as oxidizing agents to prevent or limit hydrogen evolution

  • —modify the coating formation process to create dense uniform coatings

Several different species have been reported in the literature as additions to phosphate-based coatings for Mg and its alloys. These include species containing Zn, Mn, and Ca.

Figure 2(b) shows that Mn-based phosphate coatings are stable over a wide range of conditions. A phosphate conversion coating was reported on AZ91D Mg alloy plate by Zhou, et al., based on a bath of manganese dihydrophosphate (Mn[H2PO4]2).9 The 10-μm-thick amorphous film contained not only Mn and phosphate but also Mg and Al. However, wide and narrow crack networks in the coating were attributed to the microstructure of, namely, the presence of the β-phase precipitates. Hydrogen evolution at the β-phase increased the pH in bath by consuming H+ ions and drove the hydrolyzing equilibrium for Mn(H2PO4)2 to generate insoluble phosphate precipitation. The main concern regarding such coatings was that some deep trans-cracks were observed and, as such, compromised any protective property of the coating.

Kouisni, et al.,10 added Zn2+ ions into a phosphate bath and achieved a conditionally insoluble and protective zinc phosphate (Zn3[PO4]2) layer on Mg alloy AM60. Figure 2(d) shows that Zn produces a more insoluble phosphate than Mg, but not as good as Mn. The effect of Zn2+ and nitrate ions on the phosphating process and mechanism was also examined. The Zn3(PO4)2 coating formed after 10 min in the bath provided the optimum protection against corrosion by significantly slowing down the metal dissolution process in a basic borate buffer solution at pH 9.2,11 though the resistance value ultimately obtained was not satisfactory enough for the sake of corrosion protection. Zn3(PO4)2 coatings can dissolve in mildly acidic or alkaline solutions. Even in neutral solution, such as sodium chloride (NaCl), the bulk pH can be increased substantially by Mg/magnesium dihydroxide (Mg[OH]2) dissolution. Thus, measurement of the reported corrosion resistance of Zn3(PO4)2 coating was conducted in a buffer solution with constant pH.11 

Niu, et al.,12 obtained a phosphate film of Zn3(PO4)2 on AZ91D Mg alloy in a bath based on phosphoric acid (H3PO4), zinc oxide (ZnO), and sodium fluoride (NaF). The fine crystalline Zn particles embraced phosphate crystals filling the interstices of the insoluble phosphate. Salt spray tests indicated the corrosion resistance of the coating was improved by the presence of crystalline phases in the coating. In addition, the Zn3(PO4)2 coating was microporous and adsorbed paint molecules tightly, creating a better bonding of paint film than that of paint on Cr-based coating.12 The presence of fluoride means that the coating might be a mixture of phosphate and fluoride, and the performance of fluoride coatings are discussed in the next section.

Corrosion resistance of phosphate coatings can be enhanced by adding polyvalent cations, like Mn, Co, and Ni as layer-forming cations, into a Zn3(PO4)2 conversion bath.13 Conversion baths containing Ni contents greater than 6 g/L can produce desirable phosphate coatings with smaller crystals than conventional baths. However, coatings with a dark color are not satisfactory from the viewpoint of industrial requirements since high reflectance in coatings is often necessary. Consequently, Co and Mn are selected to replace Ni partially or totally for comparative corrosion resistance while retaining a lighter coating color.13 

Complexing agents, such as sodium tartrate (C 4H4Na2O6) and film formation assistant metavanadate (VO3), also have been explored extensively in Zn phosphate conversion coating. Li, et al., proposed a low-cost phosphate coating method to produce a stable, adhesive, and anti-corrosive coating on Mg alloys by adding Zn ions and complexing agent, sodium tartrate, into the coating bath.14 The complexing agent played a critical role in stabilizing the conversion coating solution and preventing the coating surface from dusting. Coating baths containing metavanadate (ammonium metavanadate [NH4VO3], sodium metavanadate [NaVO3], or potassium metavanadate [KVO3]) can improve the coating process and produce a dense, adhesive, and protective coating.15 Wei, et al., proposed a P-Ca-V composite-phosphating solution to convert the surface of Mg alloys.16 With the aid of NH4VO3, a compact and crack-free coating was produced on Mg alloys to protect them from corrosion. The absence of water in the chemical conversion coating prevents cracks during dehydration. Ca2+ and ions in the phosphating solution had a synergistic function as corrosion inhibitors. The corroded surface area of the P-Ca-V composite conversion coating formed by this invention on AZ91D Mg alloy were reported as less than 1%, after 48 h salt spray testing.16 Permanganate (MnO4) salt, a strong oxidizing agent, analogous to chromate, may be used as an accelerator for forming phosphate conversion coatings. A suitable accelerator for phosphate treatment of Mg must not contain the anions that can be reduced to an elemental metallic state by Mg (which would induce galvanic corrosion on the surface). MnO4 ions are an ideal choice because, like chromate, when reduced, form low valency insoluble oxides (based on Mn3+ or Mn4+), which can coprecipitate with the phosphate conversion coating layer on Mg substrates, rather than being reduced to metallic Mn as:17–18 

formula
formula
formula

Vanadate and chromate also share these advantages. Ions of more noble metals such as Co and Ni may be detrimental if reduction to the metallic state occurs. Table 1 shows a comparison of oxidation-reduction potentials of the metals discussed in this section.

TABLE 1

Oxidation-Reduction Potentials of Reactions Considered for Mg Conversion Coatings

Oxidation-Reduction Potentials of Reactions Considered for Mg Conversion Coatings
Oxidation-Reduction Potentials of Reactions Considered for Mg Conversion Coatings

Zhang, et al.,19 developed a phosphate-permanganate conversion coating upon a Mg-Li alloy and revealed some protection for the substrate against corrosion. A network of cracks is, however, a general feature of permanganate-phosphate coatings,17,20–21 ascribed to hydrogen evolution or conversion coating dehydration effects.17,22 

Adding HF to a permanganate-phosphate bath can produce a surface layer that is much thinner and has fewer cracks than that produced by a nitric acid (HNO3) bath,21 because Mg reacts with F ions in the solution to give rise to insoluble magnesium fluoride (MgF2) that acts as a passive surface layer and efficiently retards Mg matrix dissolution. Composition and adherence of the phosphate-permanganate conversion coating are closely related to the composition of the Mg alloy and the treatment bath. Chong and Shih17 found that a discontinuous conversion coating was formed on the surface of the AZ61A Mg alloy in a phosphate-permanganate bath, caused from the β phase. A continuous film could be produced only if the Mg17Al12 precipitates partially dissolved together with the α-Mg matrix. Most researchers report that phosphate-permanganate treatment may produce a coating at least equivalent or more anti-corrosive than chromate treatment does. In most phosphate-permanganate conversion coating processes,20 the metal surface is pretreated by H3PO4 pickling and activated by hydrofluoric (HF) acid prior to coating.

To simplify the operation, Zhao, et al., developed a phosphate-permanganate conversion coating process without any acid pickling.23 The influence of solution composition, pH value, treating temperature, and duration time on properties of the final coatings was investigated, producing coatings performing comparative to that of the chromates. In terms of the corrosion resistance after painting, the former was better than the latter, which may be attributed to the good adhesion via the porous phosphate coating structure. In a separate study, posttreatment was also proposed for optimizing the corrosion resistance of conversion coating by sealing the open micropores on the chemical phosphate conversion coating on Mg alloys.24 

While molybdate conversion coatings have been applied extensively on steel and Zn metals for corrosion protection for years, they do not appear to be a good choice for Mg alloys because of lower oxidizing capability and ready reducibility to metallic Mo by active Mg.25 Thus, composite coatings of molybdate and phosphate were produced on AM60 Mg alloy for corrosion protection.26 A bath with a suitable molar ratio of molybdate and phosphate, namely 1:2, was reported to produce an optimum crack-free coating.

Divalent calcium cations (Ca2+) are another choice to form corrosion-resistant phosphate conversion coatings on Mg since they cannot be reduced by active Mg (Table 1). This property is postulated to make Ca2+ cations superior to other divalent cation competitors, like Co2+, Ni2+, and Zn2+. Ca and Zr combined phosphate conversion coating has been applied to Mg substrates under weak acidic conditions (pH 2.6 to 3.1).27 By adding Ca2+ ions, the resultant coating was hydrophilic and corrosion-resistant. Since Ca compounds elute preferentially to metals under a corrosive environment, they presumably induce a complex-forming reaction with H3PO4 without triggering the elution of plating metal, thus forming a dense and slightly soluble protective coating to suppress the corrosion reactions. A novel, simple, and pure calcium phosphate conversion film was developed on Mg-8.8Li. Both the cathodic hydrogen evolution and anodic corrosion reaction were restrained at the same time by the coating, thereby slowing the dissolution rate.28 

Fluoride Conversion Coatings

Alkaline earth elements (except Be) readily form insoluble fluoride species that are stable over a wide range of conditions. This is in contrast to other common active metals such as Zn and Al that are attacked by fluoride. Figure 3 compares the predominant phases of Mg, Ca, Al, and Zn in the presence of 0.2 M F solutions (equivalent to 3.8 g/L F). Figures 3(a) and (b) show magnesium fluoride (MgF2) and calcium fluoride (CaF2) to be stable over a wide range, while Figures 3(c) and (d) show that for Al and Zn the most common phases are the ionic species AlF4 and ZnF+.

FIGURE 3.

Predominance area diagrams for Mg, Ca, Al, and Zn in solutions containing 0.2 M.

FIGURE 3.

Predominance area diagrams for Mg, Ca, Al, and Zn in solutions containing 0.2 M.

Close modal

The fluoride passivation layer formed on Mg has been reported to be a Mg(OH)2−xFx film,29 which becomes close to MgF2 with increasing F concentration in the film.30 Its promising protective property attracts many researchers to coat Mg-based alloys in F solutions rather than regarding F solutions as only a means of surface activation. The more stable the fluoride, the better anticorrosive performance it has. The anticorrosive effect in aqueous solutions seems to depend on solution pH and F concentration.29 

Corrosion-inhibiting performance of F conversion coating formed in both basic29 and acidic30 baths has been investigated, respectively. Gulbrandsen, et al.,29 found that the native, porous surface Mg(OH)2 film on Mg became continuous, more compact, and thicker with F ions incorporated from an alkaline F bath. By increasing F concentration to a certain level in the bath (about 0.5 M), the resultant surface film transformed from hydroxide to a fluoride phase and had an extraordinary corrosive protection performance (corrosion current density [icorr] = 10−9 A/cm2 to 10−8 A/cm2), though the composition of this film could not be identified. Surface chemistry and electrochemistry of AM60 Mg alloy in an acidic HF solution was investigated by Verdier, et al.30 The authors observed that upon exposure to HF solution, the native Mg(OH)2 film on AM60 Mg alloy dissolved and hydrogen gas evolved. A few seconds later, a more stable and passivating coating was deposited on AM60, resulting in an elevated corrosion potential. Based on the x-ray photoelectron spectroscopy (XPS) analysis, it could not be confirmed whether it was a mixture of MgF2 and Mg(OH)2, or a MgF2 structure where fluoride ions were partially substituted by hydroxyl ions, or a Mg(OH)2 structure with a partial substitution of hydroxyl ions by F.

A fluoride conversion coating was observed on pure Mg in HF.31 After 24 h immersion in 48% HF, a 1.5-μm-thick, dense, uniform, and protective MgF2 film was formed successfully. With the aid of atomic force microscopy, it was noted the averaged mean surface roughness of the conversion-coated sample was similar to that of the untreated one. The high concentration of F ions, low temperature, and long treatment time all favored the formation of a uniform, dense, and smooth barrier coating. Results revealed that the corrosion resistance and current density of the coated Mg were improved 30 to 40 times, compared to those of the uncoated Mg. A big concern arising from the application of this research, however, is the large volumes of HF required.

The group IVB metal, F-based conversion coatings, such as fluorozirconate and fluorotitanate,32–33 are also promising options for corrosion protection of Mg alloys because they can form three-dimensional polymeric metal or metalloid-oxide matrices, from aqueous solutions as Cr3+ does,5 but are environmentally innocuous. The commercial availability and lower cost characteristics nowadays make Zr popular in many applications.34 Zr-based coatings may contain the dioxide form (ZrO2), hydroxyl oxide or hydroxyl fluoride, depending on solution composition used, while the dioxide TiO2 was the only present chemical state in the Ti-based films. F is selected to maintain the group IVB and other metals in working solution as complex fluorides and keep dissolved substrate metal ions in solution.35 The local increase in pH resulting from the reduction of water at the surface of the Mg alloy led the Zr or Ti complexes precipitating from solution.36 The film formation process was determined by the pH, F, and Zr or Ti ion concentration in the conversion solution.37 Increasing the pH from 2.2 to 5.3 gave rise to a more passive film with higher corrosion potential. Higher F ion concentration increased coating roughness with lower surface coverage and had a subsequent detrimental effect on protective film formation.

Stannate Conversion Coatings

Stannate-based conversion coatings have been proposed as environmentally acceptable alternatives to Cr coatings. Generally, resistance to pitting and crevice corrosion is said to be enhanced by such coatings from the formation of a continuous protective Mg(OH)2 layer doped with tin oxide, which act as a barrier to oxygen diffusion to the metal surface. A layer of crystalline magnesium stannate trihydrate (MgSnO3·3H2O),38 magnesium stannate hydrate (MgSnO3·H2O),39 Mg-Sn oxides,7 or mixed MgSn(OH)6 hydroxides40 have also been identified. The process of stannate conversion coating developed on AZ61 Mg alloy and the effect of the composition of the bath solution and the immersion parameters on coating property have been inspected systematically by Lin, et al.7 An incubation time was detected prior to the noticeable growth of nuclei on the metal plate in the stannate bath. The conversion coating contained a porous under-layer, contacted intimately to the metal substrate, and a hemispherical particle layer with some discontinuous sites deposited above the porous layer. Moreover, the influence of bath pH, stannate ion content, and temperature was also determined.7 Finer particles, which preferably formed in less alkaline conditions with higher stannate ion concentration, led to fewer and thinner defects and better anticorrosion characteristics. Elevating bath temperature resulted in a higher rate of the increase in the weight gain of the converted AZ61 alloy, though the bath temperature had little influence on the maximum weight gain. The final coating also contained some metallic Sn ions potentially reduced from stannate ions via Mg, which could be detrimental to the corrosion protection of the coating.

Hamdy41 investigated the influence of the concentration of Sn ions on the corrosion protection performance of AZ91D, where alkaline and/or acid etching was used as a surface pretreatment prior to the conversion coating process. Increasing stannate concentration weakened the corrosion resistance of the resultant coatings. The samples converted in the most dilute potassium stannate (K2SnO3) solution (25 g/L) showed the highest corrosion resistance. The surface treatment, namely, alkaline followed by acid etching, also provided a positive effect on improving the corrosion resistance. A review of pretreatments is given in the following section.

Elsentriecy, et al.,42 proposed a new HF pickling pretreatment for AZ91D Mg alloy and explored the effect of this pickling pretreatment on the corrosion protection ability of the stannate conversion coating. The optimal condition reported to remove the surface oxide layer without severe damage to the surface was 0.25 wt% HCl + 0.25 wt% HF for 20 s. After 1 h immersion in a stannate bath, the coating on the pickled sample was denser than that on the nonpickled sample, ascribed to the higher Mg2+ ion concentration on the pickled sample during the coating process. Hence, the pickled sample dissolved immediately upon exposure to the coating solution, and continuous metal dissolution provided a sufficient Mg2+ ion concentration for the stannate coating film precipitation; presumably, the native oxide film on the nonpickled sample lowered the dissolution rate of the metal, causing a longer coating initiation and decreased coating density. Lee and Lin43 pickled die-cast AZ91D with the die-chill skin by HF and proposed such a surface-favored MgSnO3·3H2O nucleation in subsequent stannate conversion coating.

A uniform and highly corrosion-resistant stannate conversion coating on AZ91D Mg alloy using the potentiostatic technique was developed by Elsentriecy, et al., for the first time;6,44 however, we only will deal with coatings not requiring an electrical signal herein.

Meanwhile, some researchers attempted to enhance the corrosion resistance of Mg alloys with a combination of stannate conversion coating and electroless metal plating techniques. A 3-μm to 5-μm-thick stannate conversion film with a porous structure was produced on AZ91D Mg alloy in an alkaline stannate solution for 60 min at 90°C39 followed by electroless Ni plating, forming a hybrid coating for protection of the substrate. The porous structure provided for good adherence of the electro-less Ni plating layer, which formed without cavities or crevices; additionally, the potential difference between the Ni film and substrate was decreased by the conversion coating. No hydrogen evolution was observed during 100 h immersion in 3.5 wt% NaCl solution.

Rare Earth Conversion Coatings

Rare earth (RE) salts, especially Ce and La conversion films, are approved environment-friendly corrosion inhibitors on steel and Al alloys. Recently, several researchers45–48 also successfully have exploited acidic aqueous solution of RE salts on pure Mg and its alloys to deliver conversion coatings. Protection is attributed to a hydrated RE oxide-blocking film over the cathodic sites on the metal surface. We note that weak acidic, neutral, or alkaline baths are not very suitable for Ce salts because of their high hydrolysis tendency (for instance, when [Ce3+] is 0.05 M, pH must be lower than 3.6).49 Dabala, et al.,46 produced a Ce conversion layer that can thicken rapidly within a few seconds and remain constant thereafter upon AZ63 in a CeCl3/H2O2 (cerium trichloride/hydrogen peroxide) bath. The strong oxidizing agent, H2O2, favored Ce4+ hydroxide/oxide precipitating on cathodic sites as a result of the hydrogen evolution, oxygen reduction, and pH increase. However, the unsatisfactory long-term stability of such RE coatings upon Mg substrates is still one of the largest concerns.49 Selecting an appropriate pretreatment prior to the RE conversion coating process is also critical. Acid pre-treatment is critical to enhance adherence and corrosion resistance of the Ce-based conversion coating on Mg alloys.45 HCl pretreatment induced a thicker, more adherent, more homogeneously distributed and higher corrosion-resistant RE conversion coating with a higher content of Ce than the coating on untreated surface, attributable to the higher amount of cathodic sites, higher Al content, and higher surface roughness induced by HCl pretreatment. Another method developed by Wang, et al., to improve corrosion resistance and adherence to AZ91D Mg alloy is performing the RE coating process in ethanol solution with citric acid (C 6H8O7).50 α-Mg dissolved slightly and formed insoluble magnesium citrate (Mg3Cit2) in ethanol (C2H6O). Thus, the Ce conversion coating was formed mainly on α phase, different from the coatings formed in the aqueous conversion solution. Posttreatment in sodium phosphate (Na3PO4) solution further enhanced the corrosion resistance by narrowing the microcracks in the coating by forming insoluble cerium phosphate (CePO4) (Ksp = 10−23). Neither the 120-min treating time in the conversion solution nor using ethanol as solvent is practical for industrial processes, however.

To improve corrosion resistance and paint adhesion further, other stable compounds, such as Al,51 Nb, and Zr oxides, were added into the cerium dioxide (CeO2) conversion film on AZ91D and AM50.52–54 The corrosion current density of the Ce-Zr-Nb coating was reported as 2 decades lower than that of pure Ce or Ce-Zr coating and DOW-22 chromate coatings. The conversion treatment time also plays an important role in the corrosion protective behavior of the Ce-Zr-Nb coating. A 24-h-long treatment is necessary to seal the open pores in the initially formed film to avoid the electrolyte approaching the alloy surface. The Ce-Zr-Nb coating on the AZ91 alloy was 100 times more corrosion protective than that on the AM50, though the reason is not clear.53 A Mg-Ce hydrotalcite conversion film precipitated in a bath containing cerium nitrate (Ce[NO3]3), sodium carbonate (Na2CO3), and H2O2 was seen to ennoble effectively a Mg-Gd-Y-Zr alloy. The less-soluble CeO2 (cf. Mg compounds) was found in the film and such hydrated CeO2 retarded the cathodic reaction. The cracks in the Mg-Ce hydro-talcite conversion film were less than those in the Ce conversion coating formed in the bath without Na2CO3. The high affinity of carbonate in Mg hydrotalcites presumably prevented attack by Cl, which contributed to the better corrosion resistance of the Mg-Ce hydrotalcite film.

Although different theories for the formation of Ce and La conversion films on Fe, Zn,55 Ni,56 Cu,57 and Al alloys58–59 have been reported, little is really known about the coating mechanism on Mg alloys to date. Some observations suggest that a phenomenology is:

  • —Upon immersion in an acidic Ce(NO3)3 bath, where pH is about 5.2 at 30°C, air-formed Mg(OH)2 surface film on AZ31 dissolves first, followed by metal dissolution.

  • —Al3+ ion will precipitate back to the surface as Al(OH)3 or Al2O3 because of its low solubility when pH is above 4.0.

  • —Then Mg2+ and OH ions precipitate back on the surface as a porous layer when pH is raised higher than 8.5.

  • —Meanwhile, Ce3+ ions, sufficient in the bath, also deposited on the top of the Mg(OH)2 layer, and a compact layer is formed.60 

Montemor, et al.,61 investigated how the Ce conversion films grew and developed a mixed oxide structure. They postulated a three-step mechanism:

  • —dissolution of air-formed oxide, formation of ions, and, hence, pH increase

  • —growth of an initial mixed layer of Ce(OH)4 and Ce(OH)3 upon immersion

  • —thickening of the surface film and milder pH changes with preferential deposition of Ce(OH)4 and conversion into CeO2, forming a top layer, sufficient in Ce4+ species

The corrosion behavior of AZ31 Mg alloy converted in the baths with different Ce salts, i.e., CeCl3, Ce(NO3)3, cerium(III) sulfate anhydrous (Ce2[SO4]3), and CePO4, was investigated by the same authors.62 Results revealed that the anion existing in the coating bath played a critical role in the thickness and corrosion protection efficiency on the conversion coating. The CeCl3 bath generated the thickest coating while the thinnest one was formed in the CePO4 solution. Nitrate ions led to faster oxidation of Mg and Al and stimulated formation of a layer of Mg/Al hydroxide/ oxide along with the deposition of the conversion film. Initially, the coating composition was also determined by anion ions, but finally, identical Ce4+ oxides films were produced regardless of the bath composition.

In terms of La-based conversion coating, lanthanum nitrate (La[NO3]3) was exploited as a La source to coat highly active Mg-8.8Li alloys with a homogeneous, thick (about 10 μm), and aciculate lanthanum hydroxide (La[OH]3) film for corrosion resistance.63 The La(OH)3 coating displayed passivation in a wide potential region and the corrosion current density (icorr) decreased about two orders of magnitude compared to that of the bare substrate as a result of its stability and cathodic-inhibiting characteristics. Therefore, the authors claimed that the corrosion resistance of Mg-8.8Li had been improved through the La conversion treatment. On the contrary, after converting a series of Mg alloys, AZ31, AZ61, AZ91, and AM60, in a magnesium nitrate (Mg[NO3]2) and/or La(NO3)3 bath, Takenaka, et al.,64–65 pointed out that the conversion coating formed in either Mg(NO3)2 or La(NO3)3 solution alone could not significantly improve the corrosion resistance of all the alloys; however, good protection was observed on coatings formed using both Mg(NO3)2 and La(NO3)3. Again, a drawback of such coatings is the minimum 1-h immersion time required to produce a protective coating.

Other Conversion Coatings

Hydrotalcites — Lin, et al., proposed a Mg,Al-hydrotalcite (Mg6Al2[OH]16CO3·4H2O) conversion coating on die-cast AZ91D Mg alloy.66 Although it is difficult to develop a hydrotalcite-like layer on a Mg alloy in alkaline solution, the authors succeeded by using carbonic acid (H2CO3) solution. The initial long treatment time, at least 12 h, was shortened dramatically to 2 h by drop-wise additions of 1.25 M sodium hydroxide (NaOH) solution.67 The crystalline Mg,Al-hydrotalcite conversion coating that formed was reported to decrease the corrosion rate of the substrate and only a few corroded spots appeared after 72 h salt spray.67 Deep cracks in the coating, however, though not penetrating to the substrate, may still be detrimental to the practical anti-corrosion performance in outdoor applications.

Ionic Liquid Conversion Coatings — The bis(trifluoromehtanesulfonyl) amide (TFSA)-based ionic liquid (IL) was explored by Birbilis, et al.,68 to coat pure Mg with an anticorrosive thin film (10 nm to 100 nm thick). The IL coating time is essential to determine the properties of the resulting coating on Mg. A continuous, adherent, and defect-free film, owing to the optimum corrosion protective performance, could be produced on Mg when the IL treatment time is less than 2 h. He, et al.,69 made an attempt to deposit a dense and protective coating on AZ91D alloy in an aluminum chloride (AlCl3)-NaCl molten salt. The hypothesis behind this work is that corrosion rate of Mg-Al alloys could be reduced significantly by enhancing the volume fraction of the β phase (Mg17Al12) in the skin to a high level. There was a linear correlation between the coating thickness and the treatment temperature. When the treatment temperature was higher than 300°C, cracks were observed because of the different thermal expansion coefficiency between the coating and the AZ91D substrate. The corrosion resistance of the coating was compromised by the cracks.

Multi-Elements Complex Coating — Zhao, et al.,70 claimed a newly environmentally acceptable process to develop a “multi-elements complex coating” (MECC) on AZ91D. Surface activation process by HF pickling was waived for the sake of cost and pollution control. Ca0.965Mg2Al16O27, Mn5.64P3, ZnAl2O4, and (Mg0.66Al0.34) (Al0.83Mg0.17)2O4 were the major components in the MECC. The <10-min coating time is promising for industrial uptake. Although networks of tiny cracks in the MECC were created by the hydrogen released during the coating process, they did not penetrate to the metal surface. The MECC coating formation process was divided by three stages according to the coating thickness, weight gain, and phase and element composition.71 In the first 3 min, i.e., the thickness of a compact and amorphous coating steadily increased but weight loss was positive as a result of the metal matrix corrosion. During 3 min to 5 min, the meta-phase stage, the value of the thickness decreased distinctly and weight loss was still positive. But, the coating was composed of both amorphous and crystalline phases. In the last 2 min, the final stage, the coating grew thicker, smoother, and more compact. No amorphous phase was detected. Results revealed that the MECC could provide effective protection better than that of the conventional DOW No. 1 coating.

Vanadate Coatings — Although vanadate solution is a good corrosion inhibitor widely used in the paint or pigment systems, there has been little work done to convert Mg alloys with V-based coatings. One of the reasons might be the hazardous effect of vanadate ions in the waste solution. Vanadium oxide (V2O5), NaVO3, and sodium vanadate (Na2V2O4)72–73 are used nominally as the vanadate source in conversion coating systems. The influence of coating conditions, such as vanadate concentration, immersion time, and bath temperature, on corrosion resistance was investigated by Yang, et al.74 The thickness of the coating grew with increasing immersion time and temperature. When the coating thickness was higher than 0.8 μm, cracks were induced in the coating, which deteriorated the corrosion resistance. The optimum coating was realized in the conversion bath with 30 g/L NaVO3 at 80°C for 10 min.

Organic Coatings — Although organic conversion coatings are rarely a choice for corrosion protection for Mg alloys, some organic compounds, such as tannic acid (C76H52O46)75 and organic phytic acid (C6O18O24P6),76 are capable of chelating with metal ions to make stable complexes from their structural uniqueness. A tannic acid-based conversion coating was performed on AZ91 Mg alloy.75 Pentahydroxy benzamide-Mg complex, aluminum oxide (Al2O3), and MgF2 were the major components in the coating. Longer treatment times, >600 s, induced a coarse and uneven coating surface, and resulted in the deterioration of corrosion resistance. The optimum anticorrosive coating was obtained when the samples were subjected to the treatment in the conversion solution for 300 s to 600 s. Besides the ability to form a complex with metal ions, organic phytic acids are also nontoxic. Therefore, phytic acid conversion solution was proposed as a potential replacement for Cr solution by forming a corrosion protective layer made up of complex compounds on the Mg surface.76 Polarization test results revealed that the corrosion tendency of the phytic acid conversion coating was comparable with the conventional chromate coating.

Poly (ether imides) (PEI) that have been found to be stable in the alkaline environment existing at aqueous Mg(OH)2 surfaces (such alkalinity is the reason why most other polymers are unsuitable) were chosen by Scharnagl, et al., for corrosion control of Mg alloys.77 Both microporous and nonporous coatings could be produced under specific conditions. But, the coating with micropores showed a lower degradation tendency because of the fact that the coating and the interface were intact during hydrogen evolution process, compared to the nonporous coating.

The field of conversion coating technology for Mg is significantly less advanced than that of Al—principally owing to Al being the staple of the aerospace industry. Aerospace-related research has allowed technological advances in Al coatings, and concomitantly, the fact that Mg presently is not used routinely in high-value applications means that aspects such as surface pretreatment have not received the dedicated attention they deserve. The review herein has revealed that for the same conversion coating, the pretreatment can have a major (if not the major) impact on the ultimate coating efficacy. Consequently, a brief summary of the scope and role of pretreatment is given herein—and is considered very important to the topic.

Using AZ91D as the principal example, the typical conversion coating process, including pretreatment, can be described/depicted according to Figure 4.

FIGURE 4.

Conventional conversion coating process of AZ91D Mg alloy.

FIGURE 4.

Conventional conversion coating process of AZ91D Mg alloy.

Close modal

The flow chart in Figure 4 reveals rather obviously that the conversion coating is as simple as immersion in appropriate chemicals, and can be preceded by any number of pretreatment steps to aid in the optimization of the coating (final step). The pre-treatments therefore must necessarily impart some sort of functionality to, by modification of, the surface of the alloy. The macro-and microstructural impact from the most common pretreatment procedures is shown in Figure 5 for AZ91D. It is clear that pre-treatments change the macroscopic color of the surface (Figure 5 [top]), and the concomitant changes to the underlying microstructure are captured in Figure 5 (bottom). What is seen is that following mechanical polishing, the AZ91D has a surface consisting of the Mg matrix populated with β-phase intermetallic particles. “Activator” pretreatment (which is the common term given to acid treatment) is shown to attack the Mg matrix and leave the β-phase essentially unaffected. In the case of “conditioner” pretreatment (which is the common term given to alkaline treatment), the opposite is true. The β-phase is preferentially removed (since the Al-containing phase dissolves at high pH), and the matrix is essentially unaffected, (since Mg is passive in alkaline conditions). As a result, the careful and selective choice, or even combination of, activator and conditioner treatments allows one to clean, degrease, functionalize, and homogenize the surface of an alloy prior to ultimate coating. This is demonstrated phenomenologically for the case of AZ91D in Figure 6.

FIGURE 5.

Photographs (top) and SEM images (bottom) of surface of AZ91D plates after different treatments.

FIGURE 5.

Photographs (top) and SEM images (bottom) of surface of AZ91D plates after different treatments.

Close modal
FIGURE 6.

Schematic representation of surface following the various pretreatment processes: (a) as-polished, (b) activator treated, (c) conditioner treated, and (d) activator + conditioner treated.

FIGURE 6.

Schematic representation of surface following the various pretreatment processes: (a) as-polished, (b) activator treated, (c) conditioner treated, and (d) activator + conditioner treated.

Close modal

Figure 6 reveals the ability of pretreatments to create a surface of uniform composition (viz. electrochemically homogenous), and the reader could foresee how it may be possible to customize the pretreatment process for either alloys of different composition and microstructure or the required performance of the conversion coating. One thing that is not appreciated in a large portion, if not the majority, of papers reviewed herein is that surface pretreatment is as important, if not more important, than the ultimate “coating” step (since the latter is dependent on the quality and preparedness of the incoming surface).

Detailed information about the conversion coating procedures from the papers reviewed herein and the protection afforded by the conversion coatings was extracted and has been summarized in Table 2.

TABLE 2

Summary of Coating Procedures and Corrosion Performance of the Resulting Coatings

Summary of Coating Procedures and Corrosion Performance of the Resulting Coatings
Summary of Coating Procedures and Corrosion Performance of the Resulting Coatings
Summary of Coating Procedures and Corrosion Performance of the Resulting Coatings
Summary of Coating Procedures and Corrosion Performance of the Resulting Coatings

Table 2 lists the key features of the coatings, including any pretreatment, the coating chemicals, the coating process (times, temperatures, etc.), and the ultimate corrosion performance. It is clear that there is no common method of assessment being used, with performance metrics varying widely from paper to paper with respect to techniques used (i.e., immersion tests, electrochemical impedance spectroscopy [EIS], salt spray, etc.). For the details about the corrosion measuring techniques listed in Table 2, please refer to the literature.80–83 

To further abridge the useful information in Table 2, the authors' assessment of the anti-corrosive performance of the various conversion coatings is given in Figure 7. By interpreting and weighing the relative merits of the coatings, such an assessment could be made; however, as the reader notes, rather than a detailed scrutiny, the performance has been categorized into groups. In interpreting performance, conventional chromate-based conversion coatings were used as a benchmark to evaluate the coatings.

FIGURE 7.

Authors' assessment of the anti-corrosion performance of different conversion coatings. (A: best, …, E: worst).

FIGURE 7.

Authors' assessment of the anti-corrosion performance of different conversion coatings. (A: best, …, E: worst).

Close modal

The classifications in Figure 7 are given as:

  • —Grade A means the resulting coating can provide a better corrosion protection than that of chromate coatings.

  • —The coatings with grade B either performed comparatively to that of chromate coatings or reduced the icorr dramatically by inducing a form of passivation.

  • —A coating was classified as C grade when it decreased the icorr value within one order of magnitude.

  • —Grade D provided little or minimal protection (in some cases barely enough to justify the coating cost or effort).

  • —Grade E had no sufficient corrosion protective ability that could persist for in-service application.

A brief overview relating to the top four performing coatings is given below.

Figure 7 suggests that phosphate/phosphate-permanganate conversion coating technology produces protective films on Mg alloys. With respect to case number 2, the performance observed appears to be excellent, which is in line with emerging manuscripts covering phosphate coatings. The quality of coating 2 in particular appears to arise from the combination of using a conditioner pretreatment (presumably homogenizing the surface) and the coating formulation that includes several functional components, including NaF. The NaF presumably decomposes to have free F in solution (of low pH) and allowing high rates of Mg reaction and coating formation.

The fluoride coating (coating 10) performed well; however, the use of 48% HF is not feasible in an industrial setting, with the Occupational Health and Safety (OHS)(3) restrictions in most nations preventing industrialization of this process. Additionally, the lengthy time required for coating is not entirely practical. Nonetheless, the improvement in corrosion resistance observed is not surprising if the surface could be coated with MgF2. The MgF2 would be insoluble in most aqueous media and form a suitable barrier coating if defect-free.

Coating 13 showed good performance, and emphasis was given in the role of pretreatment with this technology, indicating that the alkaline and acid pretreatments play an important role in delivering a good final coating. Again, the long coating times make this coating a little impractical; however, the ultimate performance of the coating would be presumed to be high if indeed MgF2 and non-Mg oxides could be well bonded to the Mg surface.

Coating 18 also showed some promise. The pre-treatment in this case included alkaline degreasing, and the chemicals used were considered to be OHS-friendly compared with F-containing solutions. The coating time was short, and performance in a concentrated chloride seems excellent. The characterization of surface products makes the mechanism of protection elusive, but again points to future work.

This review has shown that a large number of conversion coating processes can be used for the corrosion protection of Mg alloys. Of the best performing coatings, pretreatment was used in all cases (with the exception of the very concentrated HF immersion example). To be a real substitute for chromate coatings, there is still significant research to be done, which requires at a minimum, strategic deployment of pretreatment processes and the minimization of coating time down to a couple of minutes for the sake of practical application. Each technique reviewed has its own merits and limitations. In regard to the novel coatings reviewed, ionic liquid and molten salt conversion coatings are not applicable for practical use at this stage (and may never be, owing to technological and cost issues); vanadium-based coatings still have toxicity concerns; stearic acid coatings require high operational temperatures. There is no doubt, however, that researchers are inspired by the urgent industrial demand to develop better, simpler, cleaner, safer, and more cost-effective conversion coatings to realize a growth in application of Mg alloys.

❖ Chemical conversion coatings are a cost-effective method to assist in the corrosion protection of Mg and its alloys. Generally speaking, the protective performance of a conversion coating can be determined by several coating characteristics, such as thickness, number and type of defects, adherence to the metallic substrate, resistance to the chemical attack of the test solutions, etc. The thicker conversion coatings seemed to be more effective in retarding corrosion activity, but only when no trans-cracks were present in the coating. Chemical composition of the conversion coating also play a critical role in their anti-corrosion performance; however, it was found that the coating performance also heavily relies on appropriate pretreatments to functionalize the surface.

❖ Although a large number of Cr-free coating technologies are available for protecting Mg and its alloys, the widespread use of Mg in the automotive industry is still limited by the lack of appropriate protective coatings that can resist harsh service conditions. Every coating system brought up in this review has its own relative merits and limitations. Coating formation mechanisms, from a fundamental science point of view, are still lacking in the literature, most likely as workers seek to secure technology rights or patents for their coatings. Furthermore, standard methods to evaluate quantitatively and qualitatively real corrosion resistance of conversion-coated Mg and its alloys is also lacking, with some papers using immersion tests, salt spray, impedance, potentiodynamic polarization, etc., making cross correlations and comparisons of efficiency very subjective. Developing better, simpler, cheaper, and environmental friendly coating technologies is and will remain a hot topic, which needs further investigation so the advantages of the lower weight and excellent mechanical properties of Mg material can be fully exploited in practical applications. In summary, it appears most likely that a “family” of coatings, each of which is application-specific, will need to be developed for effective replacement of chromates.

The CAST Co-operative Research Centre was established under, and is funded in part by, the Australian Governments Co-operative Research Centres Scheme.

1.
M.
Santamaria
,
F.D.
Quarto
,
S.
Zanna
,
P.
Marcus
,
Electrochim. Acta
53
,
3
(
2007
):
p
.
1314
.
2.
G.
Song
,
A.
Atrens
,
D.S.
John
,
L.
Zheng
,
Magnesium Alloys and Their Applications
(
Weinheim, Germany
:
Wiley-VCH
,
2000
).
3.
R.
Arrabal
,
A.
Pardo
,
M.C.
Merino
,
S.
Merino
,
P.
Casajús
,
P.
Rodrigo
,
Corrosion
62
,
12
(
2009
):
p
.
817
.
4.
M.M.
Avedesian
,
H.
Baker
,
“Magnesium and Magnesium Alloys,”
in
ASM Specialty Handbook
(
Materials Park, OH
:
ASM International
,
1999
).
5.
J.E.
Gray
,
B.
Luan
,
J. Alloys Compd
.
336
,
1–2
(
2002
):
p
.
88
.
6.
H.H.
Elsentriecy
,
K.
Azumi
,
H.
Konno
,
Electrochim. Acta
53
,
2
(
2007
):
p
.
1006
.
7.
C.S.
Lin
,
H.C.
Lin
,
K.M.
Lin
,
W.C.
Lai
,
Corros. Sci
.
48
,
1
(
2006
):
p
.
93
.
8.
Y.
Song
,
D.
Shan
,
R.
Chen
,
F.
Zhang
,
E.-H.
Han
,
Corros. Sci
.
51
,
1
(
2009
):
p
.
62
.
9.
W.Q.
Zhou
,
D.Y.
Shan
,
E.H.
Han
,
W.
Ke
,
Corros. Sci
.
50
,
2
(
2008
):
p
.
329
.
10.
L.
Kouisni
,
M.
Azzi
,
M.
Zertoubi
,
F.
Dalard
,
S.
Maximovitch
,
Surf. Coat. Technol
.
185
,
1
(
2004
):
p
.
58
.
11.
L.
Kouisni
,
M.
Azzi
,
M.
Zertoubi
,
F.
Dalard
,
S.
Maximovitch
,
Surf. Coat. Technol
.
192
,
2–3
(
2005
):
p
.
239
.
12.
L.Y.
Niu
,
Z.H.
Jiang
,
G.Y.
Li
,
C.D.
Gu
,
J.S.
Lian
,
Corros. Sci
.
200
,
9
(
2006
):
p
.
3021
.
13.
T.W.
Cape
,
“High Manganese Cobalt-Modified Zinc Phosphate Conversion Coating,”
U.S. Patent Application 12/135,520, filed June 9, 2008, and issued Dec. 25, 2008
.
14.
G.Y.
Li
,
L.Y.
Niu
,
J.
Lian
,
Z.H.
Jiang
,
J.W.
Chen
,
“Magnesium Alloy Phosphorization Solution and its Phosphorized Technology,”
China Patent Application CN1598055A, issued March 23, 2005
.
15.
M.Z.
Chang
,
L.
Chen
,
Q.
Gong
,
“Method for Processing Mg Alloy Surface,”
China Patent Application CN1966766A, issued 2005
.
16.
Z.L.
Wei
,
Y.
Shen
,
Q.R.
Chen
,
C.M.
Li
,
Y.W.
Huang
,
“P-Ca-V Composite Phosphating Solution on Magnesium Alloy Surface and Chemical Conversion Processing Method,”
China Patent Application CN101418441A, issued 2009
.
17.
K.Z.
Chong
,
T.S.
Shih
,
Mater. Chem. Phys
.
80
,
1
(
2003
):
p
.
191
.
18.
H.
Umehara
,
M.
Takaya
,
Y.
Kojima
,
Mater. Trans
.
42
,
8
(
2001
):
p
.
1691
.
19.
H.
Zhang
,
G.C.
Yao
,
S.L.
Wang
,
Y.H.
Liu
,
H.J.
Luo
,
Surf. Coat. Technol
.
202
,
8
(
2008
):
p
.
1825
.
20.
F.
Zucchi
,
A.
Frignani
,
V.
Grassi
,
G.
Trabanelli
,
C.
Monticelli
,
Corros. Sci
.
49
,
12
(
2007
):
p
.
4542
.
21.
H.
Umehara
,
M.
Takaya
,
S.
Terauchi
,
Surf. Coat. Technol
.
169–170
(
2003
):
p
.
666
.
22.
C.S.
Lin
,
C.Y.
Lee
,
W.C.
Li
,
Y.S.
Chen
,
G.N.
Fang
,
J. Electrochem. Soc
.
153
,
3
(
2006
):
p
.
B90
.
23.
M.
Zhao
,
S.S.
Wu
,
J.R.
Luo
,
Y.
Fukuda
,
H.
Nakae
,
Surf. Coat. Technol
.
200
,
18
(
2006
):
p
.
5407
.
24.
W.Q.
Zhou
,
D.Y.
Shan
,
E.H.
Han
,
W.
Ke
,
“Preparation Method of Magnesium Alloy Chromeless Chemical Conversion Film and Its Used Film Forming Solution,”
China Application CN1475602A, issued Feb. 18, 2004
.
25.
D.
Hawke
,
D.L.
Albright
,
Met. Finish
.
93
,
10
(
1995
):
p
.
34
.
26.
Z.Y.
Yong
,
J.
Zhun
,
C.
Qiu
,
Y.L.
Liu
,
Appl. Surf. Sci
.
255
,
5
(
2008
):
p
.
1672
.
27.
C.E.
Tomlinson
,
“Conversion Coatings for Metal Surfaces,”
U.S. Patent no. 5380374, filed Oct. 15, 1993, and issued Jan. 10, 1995
.
28.
Y.
Song
,
D.
Shan
,
R.
Chen
,
F.
Zhang
,
E.-H.
Han
,
Surf. Coat. Technol
.
203
,
9
(
2009
):
p
.
1107
.
29.
E.
Gulbrandsen
,
J.
Taftø
,
A.
Olsen
,
Corros. Sci
.
34
,
9
(
1993
):
p
.
1423
.
30.
S.
Verdier
,
N.v.d.
Laak
,
S.
Delalande
,
F.
Dalard
,
Appl. Surf. Sci
.
235
,
4
(
2004
):
p
.
513
.
31.
K.Y.
Chiu
,
M.H.
Wong
,
F.T.
Cheng
,
H.C.
Man
,
Surf. Coat. Technol
.
202
,
3
(
2007
):
p
.
590
.
32.
J.A.
Greene
,
D.R.
Vonk
,
“Conversion Coatings Including Alkaline Earth Metal Fluoride Complexes,”
U.S. Patent 7,402,214, filed April 28, 2003, and issued July 22, 2008
.
33.
S.E.
Dolan
,
L.R.
Carlson
,
“Process for Coating Metal Surfaces,”
U.S. Patent 7,175,882, Oct. 1, 2001, and issued Feb. 13, 2007
.
34.
C.E.
Tomlinson
,
“Hydrophilic Coatings for Aluminum,”
U.S. Patent 5441580, filed Oct. 15, 1993, and issued Aug. 15, 1995
.
35.
G.A.
Reghi
,
“Coating Composition and Method,”
U.S. Patent 4338140, filed July 14, 1980, and issued July 06, 1982
.
36.
S.
Verdier
,
S.
Delalande
,
N.v.d.
Laak
,
J.
Metson
,
F.
Dalard
,
Surf. Interface Anal
.
37
,
5
(
2005
):
p
.
509
.
37.
S.
Verdier
,
N.v.d.
Laak
,
F.
Dalard
,
S.
Delalande
,
Surf. Coat. Technol
.
200
,
9
(
2006
):
p
.
2955
.
38.
M.A.
Gonzalez-Nunez
,
P.
Skeldon
,
G.E.
Thompson
,
H.
Karimzadeh
,
Corrosion
55
,
12
(
1999
):
p
.
1136
.
39.
H.
Huo
,
Y.
Li
,
F.
Wang
,
Corros. Sci
.
46
,
6
(
2004
):
p
.
1467
.
40.
L.
Anicai
,
R.
Masi
,
M.
Santamaria
,
F.D.
Quarto
,
Corros. Sci
.
47
,
12
(
2005
):
p
.
2883
.
41.
A.S.
Hamdy
,
Surf. Coat. Technol
.
203
,
3
(
2008
):
p
.
240
.
42.
H.H.
Elsentriecy
,
K.
Azumi
,
H.
Konno
,
Surf. Coat. Technol
.
202
,
3
(
2007
):
p
.
532
.
43.
Y.L.
Lee
,
C.S.
Lin
,
J. Electrochem. Soc
.
157
,
5
(
2010
):
p
.
C187
.
44.
H.H.
Elsentriecy
,
K.
Azumi
,
H.
Konno
,
Electrochim. Acta
53
,
12
(
2008
):
p
.
4267
.
45.
K.
Brunelli
,
M.
Dabalà
,
I.
Calliari
,
M.
Magrini
,
Corros. Sci
.
47
,
4
(
2005
):
p
.
989
.
46.
M.
Dabala
,
K.
Brunelli
,
E.
Napolitani
,
M.
Magrini
,
Surf. Coat. Technol
.
172
,
2–3
(
2003
):
p
.
227
.
47.
E.L.
Morris
,
“Corrosion Resistant Conversion Coatings,”
U.S. Patent 7,452,427, filed Dec. 1, 2004, and issued Nov. 18, 2008
.
48.
A.W.
Phelps
,
J.A.
Sturgill
,
J.T.
Swartzbaugh
,
“Non-Toxic Corrosion-Protection Conversion Coats Based on Rare Earth Elements,”
U.S. Patent 7,407,711, filed July 23, 2003, and issued Aug. 5, 2008
.
49.
A.L.
Rudd
,
C.B.
Breslin
,
F.
Mansfeld
,
Corros. Sci
.
42
,
2
(
2000
):
p
.
275
.
50.
C.
Wang
,
S.L.
Zhu
,
F.
Jiang
,
F.H.
Wang
,
Corros. Sci
.
51
,
12
(
2009
):
p
.
2916
.
51.
H.
Ardelean
,
C.
Fiaud
,
P.
Marcus
,
Mater. Corros
.
52
,
12
(
2001
):
p
.
889
.
52.
H.
Ardelean
,
P.
Marcus
,
“Composition and Method for Treating Magnesium Alloys,”
U.S. Patent 7156905, filed May 31, 2002, and issued Jan. 2, 2007
.
53.
H.
Ardelean
,
I.
Frateur
,
P.
Marcus
,
Corros. Sci
.
50
,
7
(
2008
):
p
.
1907
.
54.
H.
Ardelean
,
P.
Marcus
,
“Composition and Method for Treating Magnesium Alloys,”
U.S. Patent Application 20040217330A1, filed May 31, 2002, and issued Nov. 4, 2004
.
55.
S.
Böhm
,
R.
Greef
,
H.N.
McMurray
,
S.M.
Powell
,
D.A.
Worsley
,
J. Electrochem. Soc
.
147
,
9
(
2000
):
p
.
3286
.
56.
Y.
Kobayashi
,
Y.
Fujiwara
,
J. Alloys Compd
.
408–412
(
2006
):
p
.
1157
.
57.
J.A.J.
Aldykiewicz
,
A.J.
Davenport
,
H.S.
Isaacs
,
J. Electrochem. Soc
.
147
,
1
(
1996
):
p
.
147
.
58.
X.W.
Yu
,
G.Q.
Li
,
J. Alloys Compd
.
364
,
1–2
(
2004
):
p
.
193
.
59.
P.
Campestrini
,
H.
Terryn
,
A.
Hovestad
,
J.H.W.d.
Wit
,
Surf. Coat. Technol
.
176
,
3
(
2004
):
p
.
365
.
60.
C.S.
Lin
,
S.K.
Fang
,
J. Electrochem. Soc
.
152
,
2
(
2006
):
p
.
B54
.
61.
M.F.
Montemor
,
A.M.
Simões
,
M.J.
Carmezim
,
Appl. Surf. Sci
.
253
,
16
(
2007
):
p
.
6922
.
62.
M.F.
Montemor
,
A.M.
Simões
,
M.G.S.
Ferreira
,
M.J.
Carmezim
,
Appl. Surf. Sci
.
254
,
6
(
2008
):
p
.
1806
.
63.
L.
Yang
,
J.
Li
,
X.
Yu
,
M.
Zhang
,
X.
Huang
,
Appl. Surf. Sci
.
255
,
1
(
2008
):
p
.
2338
.
64.
T.
Takenaka
,
Y.
Narazaki
,
N.
Uesaka
,
M.
Kawakami
,
Mater. Trans
.
49
,
5
(
2008
):
p
.
1071
.
65.
T.
Takenaka
,
T.
Ono
,
Y.
Narazaki
,
Y.
Naka
,
M.
Kawakami
,
Electrochim. Acta
53
,
1
(
2007
):
p
.
117
.
66.
J.K.
Lin
,
C.L.
Hsia
,
J.Y.
Uan
,
Scr. Mater
.
56
,
11
(
2007
):
p
.
927
.
67.
J.K.
Lin
,
J.Y.
Uan
,
Corros. Sci
.
51
(
2009
):
p
.
1181
.
68.
N.
Birbilis
,
P.C.
Howlett
,
D.R.
MacFarlane
,
M.
Forsyth
,
Surf. Coat. Technol
.
201
,
9
(
2007
):
p
.
4496
.
69.
M.F.
He
,
L.
Liu
,
Y.T.
Wu
,
Z.X.
Tang
,
W.B.
Hu
,
Corros. Sci
.
50
,
12
(
2008
):
p
.
3267
.
70.
M.
Zhao
,
S.S.
Wu
,
P.
An
,
J.R.
Luo
,
H.
Nakae
,
Mater. Chem. Phys
.
99
,
1
(
2006
):
p
.
54
.
71.
M.
Zhao
,
S.S.
Wu
,
P.
An
,
Y.
Fukuda
,
H.
Nakae
,
J. Alloys Compd
.
427
,
1–2
(
2007
):
p
.
310
.
72.
M.-D.
Ger
,
K.-H.
Yang
,
Y.
Sung
,
W.-H.
Hwu
,
Y.-C.
Liu
,
“Method for Treating Magnesium Alloy by Chemical Conversion,”
U.S. Patent 6,755,918, filed June 13, 2002, and issued June 29, 2004
.
73.
R.G.
Buchheit
,
H.
Guan
,
V.N.
Laget
,
“Corrosion Resistant Coating with Self-Healing Characteristics,”
U.S. Patent 7,135,075, filed Jan. 20, 2004, and issued Nov. 14, 2006
.
74.
K.H.
Yang
,
M.D.
Ger
,
W.H.
Hwu
,
Y.
Sung
,
Y.C.
Liu
,
Mater. Chem. Phys
.
101
,
2
(
2007
):
p
.
480
.
75.
X.M.
Chen
,
G.Y.
Li
,
J.S.
Lian
,
Q.
Jiang
,
Surf. Coat. Technol
.
204
,
5
(
2009
):
p
.
736
.
76.
J.R.
Liu
,
Y.N.
Guo
,
W.D.
Huang
,
Surf. Coat. Technol
.
201
,
3–4
(
2006
):
p
.
1536
.
77.
N.
Scharnagl
,
C.
Blawert
,
W.
Dietzel
,
Surf. Coat. Technol
.
203
,
10–11
(
2009
):
p
.
1423
.
78.
X.L.
Liu
,
T.
Zhang
,
Y.W.
Shao
,
G.Z.
Meng
,
F.H.
Wang
,
Corros. Sci
.
51
,
11
(
2009
):
p
.
2685
.
79.
W.F.
Ng
,
M.H.
Wong
,
F.T.
Cheng
,
Surf. Coat. Technol
.
204
,
11
(
2010
):
p
1823
.
80.
S.D.
Cramer
,
B.S.
Covino
, Jr.
,
“Corrosion: Fundamentals, Testing, and Protection,”
in
ASM Handbook
,
vol. 13A
(
Materials Park, OH
:
ASM International
,
2003
).
81.
R.G.
Kelly
,
J.R.
Scully
,
D.
Shoesmith
,
R.G.
Buccheit
,
Electrochemical Techniques in Corrosion Science and Engineering (Corrosion Technology)
(
New York, NY
:
Marcel Dekker
,
2003
),
p
.
125
.
82.
G.C.
Moran
,
P.
Labine
,
eds.
,
Corrosion Monitoring in Industrial Plants Using Nondestructive Testing and Electrochemical Methods
,
ASTM STP 908
(
West Conshohocken, PA
:
ASTM International
,
1986
),
p
.
197
.
83.
M.C.H.
McKubre
,
J.R.
Macdonald
,
“Measuring Techniques and Data Analysis,”
in
Impedance Spectroscopy: Theory, Experiment, and Applications
,
eds.
E.
Barsoukov
,
J.R.
Macdonald
(
Hoboken, NJ
:
Wiley-Interscience
,
2005
),
p
.
199
.

(1) Thermodynamic diagrams were calculated using MEDUSA software.

(2) RoHS Enforcement Team, NMO, Stanton Ave., Teddington, Middlesex TW11 0JZ, United Kingdom.

(3) Workplace Safety Australia Pty Ltd., Tower 2, Suite 1303, Level 13, Westfield Plaza, 500 Oxford St., Bondi Junction NSW 2022, Australia.

Trade name.

Author notes

* CAST Co-operative Research Centre, Department of Materials Engineering, Monash University, VIC. 3800, Australia.

** Advanced Magnesium Technologies, Sydney, NSW. 2000, Australia.