materials Article

Effect of Mg on the Microstructure and Corrosion Resistance of the Continuously Hot-Dip Galvanizing Zn-Mg Coating Anping Dong 1,2 , Baoping Li 3, *, Yanling Lu 4, *, Guoliang Zhu 1,2, *, Hui Xing 1,2 , Da Shu 1,2 , Baode Sun 1,2,5 and Jun Wang 1,2,5 1

2 3 4 5

*

School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China; [email protected] (A.D.); [email protected] (H.X.); [email protected] (D.S.); [email protected] (B.S.); [email protected] (J.W.) Shanghai Key Lab of Advanced High-temperature Materials and Precision Forming, Shanghai 200240, China Hechi Industry and Information Committee, Hechi 547000, China Department of Nuclear Materials Science and Engineering, Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201800, China State Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, China Correspondence: [email protected] (B.L.); [email protected] (Y.L.); [email protected] (G.Z.); Tel.: +86-21-5474-5597 (G.Z.)

Received: 26 June 2017; Accepted: 31 July 2017; Published: 22 August 2017

Abstract: The microstructure of continuously hot-dip galvanizing Zn-Mg coating was investigated in order to obtain the mechanism of the effects of Mg on the corrosion resistance. In this paper, the vertical section of the Zn-0.20 wt % Al-Mg ternary phase diagram near the Al-low corner was calculated. The results indicates that the phase composition of the Zn-0.20 wt % Al-Mg ternary phase diagram near the Al-low corner is the same as Zn-Mg binary phase diagram, suggesting Al in the Zn-Mg (ZM) coatings mainly concentrates on the interfacial layer between the coating and steel substrate. The microstructure of continuously hot-dip galvanizing ZM coatings with 0.20 wt % Al containing 1.0–3.0 wt % Mg was investigated using tunneling electron microscopy (TEM). The morphology of Zn in the coating changes from bulk to strip and finally to mesh-like, and the MgZn2 changes from rod-like to mesh-like with the Mg content increasing. Al in the ZM coatings mainly segregates at the Fe2 Al5 inhibition layer and the Mg added to the Zn bath makes this inhibition layer thinner and uneven. Compared to GI coating, the time of the first red rust appears increases by more than two-fold and expansion rate of red rust reduces by more than four-fold in terms of salt spray experiment. The ZM coating containing 2.0 wt % Mg has the best corrosion resistance. The enhanced corrosion resistance of ZM coatings mainly depends on different corrosion products. Keywords: continuously hot-dip galvanizing; Zn-Mg coating; TEM; corrosion resistance

1. Introduction In recent years, hot-dip galvanizing Zn coatings containing Mg, referred to as a promising next generation of Zn-based coatings, have received considerable attention, primarily due to their superior corrosion resistance with respect to GI (Zn-0.20 wt % Al) coating [1–7] or Zn-Al coatings [8,9]. The Zn-based coatings containing Mg can be roughly classified into two types: Zn-Al-Mg coatings with high Al and Zn-Al-Mg coatings with very low Al or without Al. For Zn-Al-Mg coatings with high Al, the role of Al is to impede the oxidation of Zn liquid and enhance corrosion resistance. The conventional continuously hot-dip galvanizing technique was applied to mass-produce this type of Zn based coatings [3–5,8–12]. For Zn-Al-Mg coatings with very low Al or without Al, hereafter referred to as Zn-Mg coatings, other techniques, such as physical vapor deposition (PVD) [7,13,14], Materials 2017, 10, 980; doi:10.3390/ma10080980

www.mdpi.com/journal/materials

Materials 2017, 10, 980

2 of 14

batching galvanizing [6,15], electrodeposition (EG) [16–18], and the two-step method [2], were selected to produce these Zn-based coatings to avoid interference of the oxidation of Mg with production. For PVD and the two-step method, the oxidation of Mg can be eliminated completely. For batching galvanizing method, prior to hot-dip galvanizing, the oxidation layer of Mg can be sweep away from the surface of Zn liquid manually. With the Al in the Zn coating increasing, the welding [19] and cathodic protection performance [20] will deteriorate. In this regard, the Zn-Mg coatings with a low Al content is beneficial to expand application in different fields. Al (0.20 wt %) was added into the Zn bath in order to utilize the processing parameters of the GI coating. Prosek investigated the corrosion resistance of Zn-Mg alloys and reported that when the Mg content in the Zn-Mg alloys was less than 8.0 wt %, the corrosion resistance of Zn-Mg alloy increased with the Mg content increasing after exposure to humid air [3]. To date, the systematical investigation relevant to microstructure and corrosion resistance of the continuously hot-dip galvanizing Zn-Mg coatings with relatively high Mg contents is scarce. Most of the studies focused on high Mg and Al content addition to the zinc coating. However, it is not good for producing high-quality hot-dip galvanized automobile sheet, mainly because high Al content leads to excessive zinc dross, thus affecting the surface quality of the hot-dip galvanized automobile sheet, and increase zinc consumption. In addition, too high Mg results in severe oxidation in the galvanizing temperature. In this study, we propose the low Mg and Al content in the zinc coating to avoid the oxidation of zinc liquid and improve wettability. The Mg in hot-dip Zn coatings can inhibit the grain boundary corrosion and improve the adherence of the corrosion product layer on the coating, thereby ameliorating the corrosion resistance. In this paper, the Zn-0.20 wt % Al-Mg (ZM)-coated steel sheets with 1.0–3.0 wt % Mg were prepared through the continuously hot-dip galvanizing with the same processing parameters as GI coated steel sheets in the laboratory. The vertical section of the ternary Zn-0.20 wt % Al-Mg phase diagram near the Al-low corner was calculated using the thermodynamic data of the available literature [21] to predict the phase composition. The microstructure and corrosion behaviors of ZM coatings were investigated in our previous research [22]. In this part, we further investigate the microstructure of ZM coatings using tunneling electron microscopy (TEM). In addition, corrosion resistance of ZM coatings is also studied to optimize the coating composition. 2. Experimental 2.1. Phase Diagram Calculation To predict the phase composition of ZM coating, the vertical section of this ternary phase diagram with 0.20 wt % Al near Al-low corner at 460 ◦ C (typical hot-dip galvanizing temperature) was calculated using the ThermoCalc program (CALPHAD, Thermo-Calc , Stockholm, Sweden) in terms of the literature [21]. 2.2. Materials and Microstructure Characterization Three sets of continuously hot-dip galvanizing ZM-coated steel sheets and GI-coated steel sheets were investigated in this paper by using the continuously hot-dip galvanizing simulator in the Baosteel Group Corporation. The temperature of the steel sheet entering the zinc pot is about 780 ◦ C, the hot-dip galvanizing temperature is 460 ◦ C, and the immersion time is 3 s. The steel substrate is the IF steel sheet with a thickness of 0.5 mm and its chemical composition is listed in Table 1. Nominal compositions (wt %) of GI and ZM coatings are Zn-0.20 Al, Zn-0.20 Al-1.0 Mg, Zn-0.2 0Al-2.0 Mg, and Zn-0.2 Al-3.0 Mg, hereafter referred to as GI, ZM1, ZM2, and ZM3, respectively. The microstructure of ZM coatings and the interfacial layer between the coating and steel substrate were investigated using TEM (JEM-2000EX, JEOL, Tokyo, Japan. Accelerating voltage: 160 KV). The TEM samples less than 100 nm in thickness were prepared with a focused ion beam (FIB) system (FEI Dual Beam 820, FEI, Hillsboro, OR, USA).

Materials 2017, 10, 980

3 of 14

Table 1. Chemical composition of IF steel sheet (wt %). C

Si

Mn

P

S

N

Al

Cr

Ni

0.009

0.008

0.091

0.0095

0.0093

0.0017

0.037

0.015

0.017

2.3. Evaluation of Corrosion Resistance A salt spray test and linear polarization method were conducted to evaluate the corrosion resistance of GI- and ZM-coated steel sheets. At least three identical samples were used for every chemical composition to obtain reliable data. For the salt spray test, the samples of 40 mm × 40 mm were cut from the perfect areas of GI- and ZM- coated steel sheets. Prior to the test, the surface contaminants on samples were removed by an ultrasonic treatment in acetone solution, and then the periphery of samples was sealed with epoxy resin. Finally, the samples were introduced to a standard salt spray chamber (FQY, Yuanyao, Dongguan, China) according to DIN EN ISO 9227, detailed parameters of which are shown elsewhere [1]. For the linear polarization experiment, samples of 10 mm × 10 mm were cut from the perfect areas of GI- and ZM-coated steel sheets. The samples for linear polarization experiment were prepared as follows: first, the samples were cleaned as in the case of the salt spray test. Secondly, one surface of samples was welded with a Cu wire and then this surface of samples was sealed together with the periphery of samples using epoxy resin. A standard three-electrode system (CHI 660 C, CHI, Wildwood, NJ, USA) was used with platinum as the counter electrode, SCE (saturated calomel electrode) as the reference electrode, and samples as the working electrodes. The samples were immersed in an aerated and quiescent 5 wt % NaCl solution for 15 min to stabilize their open potentials before the linear polarization experiment. After that, linear polarization measurements were carried out using a sweep rate of 0.01 V/s and a potential range from −1.8 to 0.2 V. 2.4. Characterization of Corrosion Products The phase composition of corrosion products on GI and ZM coatings after the salt spray test were identified using XRD (Philips X’pert, PANalytical B.V., Almelo, The Netherlands) and Fourier-transformed infrared spectroscopy. For the XRD experiment, the scanning range of 2θ is from 10◦ to 70◦ . The scanning rate is 0.02◦ /s. For the Fourier-transformed infrared spectroscopy experiment, 0.70 mg of corrosion products were mixed with a spectroscopic-grade potassium bromide (KBr) powder to obtain 140 mg of a mixture. The mixture was ground in a mortar and pressed into pellets in a die. Spectra were collected by adding 16 scans at 8 cm−1 resolution in the range from 400 to 4000 cm−1 . A background spectrum was obtained from a pure KBr pellet of the same weight and size. The corrosion products on GI and ZM coatings were also investigated using SEM (JEOL, Tokyo, Japan). 3. Results and Discussion 3.1. Zn-Al-Mg Ternary Phase Diagram Figure 1 shows the vertical cross-section of the Zn-0.20 wt % Al-Mg ternary phase diagram near the Al-low corner. Compared with the Zn-Mg binary phase diagram [23], it is found that the phase composition of this ternary phase diagram is the same as in the Zn-Mg binary phase diagram near the Al-low corner, namely Zn plus Mg2 Zn11 , implying no Mg-containing intermetallic compounds form in the coating during the equilibrium solidification. In this regard, the Zn-0.20 wt % Al-Mg coating can be considered as Zn-Mg coating. The temperature of eutectic point of Zn-0.20 wt % Al-Mg ternary system is about 637 K, which is very close to that of the Zn-Mg binary system. Additionally, it is of interest to note that no Al-containing intermetallic compounds form in the coating by means of this phase diagram, suggesting Al added to the Zn bath may be still segregate at the interfacial layer between the coating and steel substrate.

Materials 2017, 10, 980

4 of 14

Materials 2017, 10, 780 Materials 2017, 10, 780

4 of 14 4 of 14

Figure the Al-low Al-low corner. corner. Figure 1. 1. Vertical Vertical section section of of Zn-0.2 Zn-0.2 wt wt % % Al-Mg Al-Mg ternary ternary phase phase diagram diagram near near the Figure 1. Vertical section of Zn-0.2 wt % Al-Mg ternary phase diagram near the Al-low corner.

3.2. Microstructure 3.2. 3.2. Microstructure Microstructure The microstructure ZM coatings was investigated using SEMSEM in our published paper The microstructureof ZM coatings was investigated using inprevious our previous published The microstructure ofofZM coatings was investigated using SEM in our previous published paper [22]. In this paper, the TEM observation was performed to investigate further the coating paper [22]. this paper, the TEM observation performedtotoinvestigate investigatefurther further the the coating [22]. In thisIn paper, the TEM observation waswas performed coating microstructure and the interfacial layer between the coating and steel substrate. substrate. Figure 22 shows the microstructure microstructure and and the the interfacial interfacial layer layer between between the the coating coating and and steel steel substrate. Figure Figure 2 shows shows the the TEM micrographs micrographs of Zn in in ZM ZM coatings. coatings. As seen in the Figure Figure 2, 2, the the Zn Zn morphology morphology changes changes with with TEM of Zn As seen in the TEM micrographs of Zn in ZM coatings. As seen in the Figure 2, the Zn morphology changes with different Mg into into theZn Zn bath.Many Many bulkgrains grains are observed ZM1 coating (Figure With the different are observed in in ZM1 coating (Figure 2a).2a). With the Mg different Mg Mg intothe the Znbath. bath. Manybulk bulk grains are observed in ZM1 coating (Figure 2a). With the Mg increasing towt 2.0 wtmany %, many Zn grains are observed in ZM2 coating (Figure increasing to 2.0 longlong stripstrip Zn grains are observed in ZM2 coating (Figure 2b). 2b). WithWith Mg Mg increasing to 2.0%, wt %, many long strip Zn grains are observed in ZM2 coating (Figure 2b). With Mg increasing to 3.0 %, mesh-like grainsare areobserved observedasasshown shown in ZM3 coating coating (Figure 2c). increasing to 3.0 wt wt %, thethe mesh-like ZnZn grains 2c). Mg increasing to 3.0 wt %, the mesh-like Zn grains are observed as shownininZM3 ZM3 coating (Figure (Figure 2c). Similarly, the morphology of MgZn 2 changes with the Mg content in the ZM coating. The rod-like Similarly, the morphology of MgZn changes with the Mg content in the ZM coating. The rod-like Similarly, the morphology of MgZn22 changes with the Mg content in the ZM coating. The rod-like MgZn 2 is observed in ZM2 coating (Figure 3). The mesh-like MgZn2 is observed in ZM3 coating MgZn is observed is observed MgZn22 is observed in in ZM2 ZM2 coating coating (Figure (Figure 3). 3). The The mesh-like mesh-like MgZn MgZn22 is observed in in ZM3 ZM3 coating coating (Figure 2c). (Figure 2c). (Figure 2c).

(a) (a)

(b) (b) Figure 2. Cont.

Materials 2017, 10, 980

5 of 14

Materials 2017, 10, 780 Materials 2017, 10, 780

5 of 14 5 of 14

(c) (c) Figure ZM coatings, (a) ZM1; (b) ZM2; and (c) (c) ZM3. Figure 2. 2. TEM TEM micrographs micrographs of of Zn Zn on on Figure 2. TEM micrographs of Zn on ZM ZM coatings, coatings, (a) (a) ZM1; ZM1; (b) (b) ZM2; ZM2; and and (c) ZM3. ZM3.

Figure 3. The TEM micrograph of rod-like MgZn2. Figure 3. The TEM micrograph of rod-like MgZn2. Figure 3. The TEM micrograph of rod-like MgZn2 .

In general, an Al-rich inhibition layer (Fe2Al5) with a thickness of 500 nm was formed between In general, an Al-rich inhibition layer (Fe2Al5) with a thickness of 500 nm was formed between the steel substrate the inhibition coating for the GI which was produced by the hotIn an and Al-rich (Fecoating a thickness of 500 nm wascontinuously formed between 2 Al5 ) with the steelgeneral, substrate and the coating forlayer the GI coating which was produced by the continuously hotdip galvanizing method To find there is still Al-rich inhibition layer between the the substrate and the[24]. coating for out the whether GI coating which wasan produced by the continuously hot-dip dip steel galvanizing method [24]. To find out whether there is still an Al-rich inhibition layer between the coating and steel substrate forfind ZM coatings, a representative linear scan inhibition is shown inlayer Figure 4. As seen galvanizing method [24]. To out whether there is still an Al-rich between the coating and steel substrate for ZM coatings, a representative linear scan is shown in Figure 4. As seen in Figure 4, steel an abrupt Al-rich peak is obviously observed,linear implying there an Al-rich coating and substrate for ZM coatings, a representative scan isthat shown in exists Figure 4. As seen in Figure 4, an abrupt Al-rich peak is obviously observed, implying that there exists an Al-rich inhibition layer. in Figure 4, an abrupt Al-rich peak is obviously observed, implying that there exists an Al-rich inhibition layer. inhibition layer.

Materials 2017, 10, 980 Materials 2017, 10, 780 Materials 2017, 10, 780

6 of 14 6 of 14 6 of 14

Figure 4. Linear scanning micrographs of the ZM3 coating. Figure 4. Linear scanning micrographs of the ZM3 coating. Figure 4. Linear scanning micrographs of the ZM3 coating.

Figure 55 shows shows the the TEM TEMmicrographs micrographsofofthe theinterfacial interfaciallayer layerononZM ZMcoatings, coatings, and Γ phase Figure and a Γa phase is Figure 5the shows the TEM micrographs of thecoating interfacial layer5a). on ZM coatings, and aΓ phase is found on interfacial situation of the ZM1 (Figure However, for ZM2 and ZM3is found on the interfacial situation of the ZM1 coating (Figure 5a). However, for ZM2 and ZM3 found onthere the exists interfacial situation of thelayer ZM1 coating (Figure 5a). However, for the ZM2 and ZM3 coatings, aa Fe Al55 interfacial interfacial (Figure 5b,c). For these these ZM coatings, coatings, thickness of coatings, there exists Fe22Al layer (Figure 5b,c). For ZM the thickness of coatings, there exists a Fe 2 Al 5 interfacial layer (Figure 5b,c). For these ZM coatings, the thickness of the interfacial layer is about 300 nm which is obviously thinner than that formed in the GI coating [24]. the interfacial layer is about 300 nm which is obviously thinner than that formed in the GI coating the interfacialitlayer is about is obviously than that formed the as GI seen coating Additionally, is found that300 thenm Fe2which Al inhibition layerthinner in some regions is very in thin, in [24]. Additionally, it is found that the Fe52Al 5 inhibition layer in some regions is very thin, as seen in [24]. Additionally, it is found that the Fe 2 Al 5 inhibition layer in some regions is very thin, as seen in the ZM3 ZM3 coating. This result result implies implies that that the the steel the coating. This steel sheet sheet surface surface may may not not be be covered covered by by the the Fe Fe22Al Al55 the ZM3 coating. This result implies that of thethe steel sheet surface may not regions be covered byZn the Fe2Al5 inhibition layer, resulting resulting in the the formation Fe-Zn compound in local local of the the coating. inhibition layer, in formation of the Fe-Zn compound in regions of Zn coating. inhibition layer, resulting in the formation of the Fe-Zn compound in local regions of the Zn coating. As aa result, to the of the the As result, adherence adherence of of the the Zn Zn coating coating to to the the steel steel substrate substrate decreases decreases due due to the formation formation of As a result, adherence of It the Znbecoating to that the steel substrate decreases dueoftothe theFeformation of the Fe-Zn compound [25,26]. can inferred Mg can affect the formation Al inhibition 2 5 Fe-Zn compound [25,26]. It can be inferred that Mg can affect the formation of the Fe2Al5 inhibition Fe-ZnTo compound [25,26]. It can be inferred that Mg can affect the formation of bath the Fe2Al5 inhibition layer. fully inhibit inhibit the formation formation of Fe-Zn Fe-Zn compounds, Al added added to the the Zn Zn Mg layer. To fully the of compounds, Al to bath containing containing Mg layer. To fully inhibit the formation of Fe-Zn compounds, Al added to the Zn bath containing Mg should increase increase accordingly. accordingly. should should increase accordingly.

(a) (a)

(b) (b) Figure 5. Cont.

Materials 2017, 10, 980

7 of 14

Materials 2017, 10, 780

7 of 14

(c) Figure thethe interfacial layer on on thethe ZMZM coatings, (a) ZM1; (b) ZM2; and Figure 5. 5. The TheTEM TEMmicrogpraphs microgpraphsofof interfacial layer coatings, (a) ZM1; (b) ZM2; (c) andZM3. (c) ZM3.

3.3. 3.3. Corrosion Corrosion Resistance Resistance The the salt salt spray spray experiment 2. The The first first white white rust rust time time for for ZM ZM The results results of of the experiment are are showed showed in in Table Table 2. and GI coatings is the same (two days). It is noteworthy that, with respect to GI, a more than two and GI coatings is the same (two days). It is noteworthy that, with respect to GI, a more than two times rust time time and and aa more more than than three three times times increase increase in severe red red rust rust times increase increase in in the the first first red red rust in the the severe time time are are observed observed for for ZM ZM coatings. coatings. The The best best corrosion corrosion resistance resistance can can be be obtained obtained for for the the ZM2 ZM2 coating. coating. The time from the first red to severe red rust increases by four times for ZM coatings compared to The time from the first red to severe red rust increases by four times for ZM coatings compared to the the GI coating. A similar was reported in a previous studythe where thecoating Zn-Mgwas coating was GI coating. A similar resultresult was reported in a previous study where Zn-Mg prepared prepared by electrodeposition and its corrosion resistance evaluated by a salt spray test [16]. From by electrodeposition and its corrosion resistance evaluated by a salt spray test [16]. From the salt spray the salt spray experiment results, the corrosion resistance GI and from ZM coatings to poor experiment results, the corrosion resistance of GI and ZMof coatings good to from poor good is as follows: is as follows: ZM2 > ZM1 >ZM2 ZM3>>ZM1 GI. > ZM3 > GI. Table 2. Corrosion resistance of GI and ZM coatings evaluated by the salt testspray in 5 wt % NaCl solution. resistance of GI and ZM coatings evaluated byspray the salt test in 5 wt % Table 2. Corrosion NaCl solution. Time to First Designation

Designation GI GIZM1 ZM2 ZM1 ZM3 ZM2 ZM3

Time to First White Rust (Day) Time to First White Rust2 (Day)

Red Rust Time to First (Day) Red Rust 8 (Day)

Time to Severe Red Rust (Day) Time to Severe Red Rust 18 (Day)

Area of Severe Red Rust (%) Area of Severe Red Rust (%) 16.8

Time to Corrosion Expansion (Day) Time to Corrosion Expansion (Day) 10

2 2 2 2 2 2 2

18 8 18 18 18 18 18

5918 5959 5959 59

16.6 16.8 15.4 16.6 17 15.4 17

4010 4040 4040 40

Figure 6 shows linear polarization curves of GI and ZM coatings after different immersion times in 5.0Figure wt % NaCl solution. the immersion is less or equal two days, the anodictimes and 6 shows linear When polarization curves oftime GI and ZMthan coatings aftertodifferent immersion cathodic parts of the linear polarization curves for GI and ZM coatings are similar, especially on the in 5.0 wt % NaCl solution. When the immersion time is less than or equal to two days, the anodic and very high potential or rather low potential regions. After 11 days of immersion, on the anodic part of cathodic parts of the linear polarization curves for GI and ZM coatings are similar, especially on the the linear polarization curve forpotential the GI regions. coating,After the second electrochemical reaction is very high potential or rather low 11 days anodic of immersion, on the anodic part of observed near −0.7 V, implying corrosion possibly occurred on the steel substrate. the linear polarization curve for the GI coating, the second anodic electrochemical reaction is observed near −0.7 V, implying corrosion possibly occurred on the steel substrate.

Materials 2017, 10, 980

8 of 14

Materials 2017, 10, 780

8 of 14

Figure 6. 6. Linear Linear polarization polarization curves curves of of GI GI and and ZM ZM coatings coatings after after different different immersion immersion times times in in 55 wt wt % % Figure NaCl solution. (a) 0 day; (b) 1 day; (c) 2 days; (d) 11 days. NaCl solution. (a) 0 day; (b) 1 day; (c) 2 days; (d) 11 days.

Free corrosion potential and free corrosion current density curves versus time, derived from Free corrosion potential and free corrosion current density curves versus time, derived from linear linear polarization curves, are shown in Figures 7 and 8, respectively. These curves clearly polarization curves, are shown in Figures 7 and 8, respectively. These curves clearly characterize the characterize the corrosion resistance. corrosion resistance. As indicated in Figure 7, it can be seen that the general variation tendency of the free corrosion As indicated in Figure 7, it can be seen that the general variation tendency of the free corrosion potential for GI and ZM coatings is the same. At the very initial stage of corrosion, the free corrosion potential for GI and ZM coatings is the same. At the very initial stage of corrosion, the free corrosion potentials for all samples were relatively high, which may be connected with the formation of a thin potentials for all samples were relatively high, which may be connected with the formation of a thin semiconductor metal oxide layer (ZnO or a mixture of ZnO and MgO) on the coating surface in semiconductor metal oxide layer (ZnO or a mixture of ZnO and MgO) on the coating surface in contact contact with atmosphere [3]. However, the free corrosion potentials for ZM coatings are lower than with atmosphere [3]. However, the free corrosion potentials for ZM coatings are lower than that of that of the GI coating because magnesium oxide has a lower potential than zinc oxide [3]. After one the GI coating because magnesium oxide has a lower potential than zinc oxide [3]. After one day of day of immersion, the free corrosion potentials for all samples reduce due to the active corrosion of immersion, the free corrosion potentials for all samples reduce due to the active corrosion of the metal the metal coatings. The free corrosion potential for the GI coating is still higher than that of the ZM coatings. The free corrosion potential for the GI coating is still higher than that of the ZM coatings. coatings. After two days of immersion, the free corrosion potentials for all the samples increase due After two days of immersion, the free corrosion potentials for all the samples increase due to the to the formation of corrosion products which can impede the infiltration of aggressive corrosion formation of corrosion− products which can impede the infiltration of aggressive corrosion mediums, mediums, such as Cl ions. It is noteworthy that the free corrosion potential of the ZM2 coating is such as Cl− ions. It is noteworthy that the free corrosion potential of the ZM2 coating is slightly higher slightly higher than that of the GI coating, suggesting its corrosion products probably differ with the than that of the GI coating, suggesting its corrosion products probably differ with the GI coating. GI coating. After 11 days of immersion, the free corrosion potentials for the ZM coatings are higher After 11 days of immersion, the free corrosion potentials for the ZM coatings are higher than the than the GI coating, in which the ZM2 coating possesses the highest free corrosion potential, GI coating, in which the ZM2 coating possesses the highest free corrosion potential, suggesting the suggesting the corrosion products formed on ZM coatings have a better protection ability to the steel corrosion products formed on ZM coatings have a better protection ability to the steel substrate with substrate with respect to the GI coating [27]. respect to the GI coating [27].

Materials 2017, 10, 980

9 of 14

Materials 2017, 10, 780

9 of 14

Figure 7. 7. Free Free corrosion corrosion potential potential curves curves of of GI GI and and ZM ZM coatings coatings versus versus time time after after different different immersion immersion Figure times in in 55 wt wt % % NaCl NaCl solution. solution. times

As indicated in Figure 8, at the very initial stage of corrosion, it is obvious that the free corrosion As indicated in Figure 8, at the very initial stage of corrosion, it is obvious that the free corrosion current density of the GI coating is lower than that of the ZM coatings, suggesting Mg added to the current density of the GI coating is lower than that of the ZM coatings, suggesting Mg added to the Zn Zn bath can accelerate the corrosion process. This result was reported in the literature where MgZn2 bath can accelerate the corrosion process. This result was reported in the literature where MgZn2 was was corroded preferentially in the Zn-Al-Mg coating [1,28]. After one day of immersion, the free corroded preferentially in the Zn-Al-Mg coating [1,28]. After one day of immersion, the free corrosion corrosion current density of the ZM2 coating is slightly lower than that of the GI coating, however, current density of the ZM2 coating is slightly lower than that of the GI coating, however, the rest of the the rest of the ZM coatings are still higher than that of the GI coating, indicating the corrosion ZM coatings are still higher than that of the GI coating, indicating the corrosion products on the ZM2 products on the ZM2 coating have the best protection ability. After two days of immersion, the free coating have the best protection ability. After two days of immersion, the free corrosion current density corrosion current density for all ZM coatings is clearly lower than that of the GI coating. The lowest for all ZM coatings is clearly lower than that of the GI coating. The lowest free corrosion current free corrosion current density is still observed for the ZM2 coating. Based on the results of the linear density is still observed for the ZM2 coating. Based on the results of the linear polarization experiment, polarization experiment, corrosion resistance of GI and ZM coatings is as follows: ZM2 > ZM1 > ZM3 corrosion resistance of GI and ZM coatings is as follows: ZM2 > ZM1 > ZM3 > GI. This is in accordance > GI. This is in accordance with the result of the salt spray experiment. The corrosion resistance is with the result of the salt spray experiment. The corrosion resistance is closely connected with the closely connected with the microstructure of Zn-Mg coatings. When 1.0 wt % Mg was added to the microstructure of Zn-Mg coatings. When 1.0 wt % Mg was added to the Zn bath, the microstructure Zn bath, the microstructure of the Zn-Mg coating is similar to the GI coating [22], its corrosion of the Zn-Mg coating is similar to the GI coating [22], its corrosion behavior is also similar to GI behavior is also similar to GI except for elimination of grain boundary corrosion [22]. With an except for elimination of grain boundary corrosion [22]. With an addition of 2.0 wt % Mg to the Zn addition of 2.0 wt % Mg to the Zn bath, a suitable amount of (Zn + MgZn2) eutectics resulted mainly bath, a suitable amount of (Zn + MgZn2 ) eutectics resulted mainly in general corrosion [22]. With an in general corrosion [22]. With an addition of 3.0 wt % Mg to the Zn bath, the coarse and mesh-like addition of 3.0 wt % Mg to the Zn bath, the coarse and mesh-like grains give rise to severe localized grains give rise to severe localized corrosion [22]. It seems that the suitable content of (Zn + MgZn2) corrosion [22]. It seems that the suitable content of (Zn + MgZn2 ) eutectics and morphology of MgZn2 eutectics and morphology of MgZn2 are very important factors to improve the corrosion resistance are very important factors to improve the corrosion resistance of Zn-Mg coatings. of Zn-Mg coatings. It was reported that when the content of Mg in the Zn-Mg alloy was less than 8.0 wt %, its corrosion It was reported that when the content of Mg in the Zn-Mg alloy was less than 8.0 wt %, its resistance increased with the Mg content increasing [3]. However, according to the present results, corrosion resistance increased with the Mg content increasing [3]. However, according to the present it is found that the optimal corrosion resistance of the Zn-Mg coating containing 0.20 wt % Al can results, it is found that the optimal corrosion resistance of the Zn-Mg coating containing 0.20 wt % Al be obtained with an addition of 2.0 wt % Mg to the Zn bath. The reasons for this difference are can be obtained with an addition of 2.0 wt % Mg to the Zn bath. The reasons for this difference are associated with the phase composition and morphology of the microstructure. In the case of Zn-Mg associated with the phase composition and morphology of the microstructure. In the case of Zn-Mg alloys, Mg2 Zn11 was found due to equilibrium solidification. However, in the case of Zn-Mg coatings, alloys, Mg2Zn11 was found due to equilibrium solidification. However, in the case of Zn-Mg coatings, the transformation of MgZn2 to Mg2 Zn11 was inhibited due to the rapid solidification rate under the the transformation of MgZn2 to Mg2Zn11was inhibited due to the rapid solidification rate under the condition of hot-dip galvanizing [23]. Obviously, the phase composition difference will lead to a condition of hot-dip galvanizing [23]. Obviously, the phase composition difference will lead to a difference of corrosion resistance. As we know, the morphology of a phase is an important factor to difference of corrosion resistance. As we know, the morphology of a phase is an important factor to determine determine the the corrosion corrosion resistance resistance [29]. [29]. The The morphologies morphologies of of Zn Zn and and Zn-Mg Zn-Mg compound compound in in the the Zn-Mg Zn-Mg coatings are different from that in the Zn-Mg alloys. Thus, this difference in morphology also coatings are different from that in the Zn-Mg alloys. Thus, this difference in morphology also results results in in aa difference difference of of corrosion corrosion resistance. resistance.

Materials 2017, 10, 980

10 of 14

Materials 2017, 10, 780

Materials 2017, 10, 780

10 of 14

10 of 14

Figure 8. Free corrosion current density curves of GI and ZM coatings versus time after different Figure 8. Free corrosion current density curves of GI and ZM coatings versus time after different immersion times in 5 wt % NaCl solution. immersion times in 5 wt % NaCl solution.

Figure 8. Free corrosion current density curves of GI and ZM coatings versus time after different immersion times 5 wt % NaCl solution. 3.4. Characterization of in Corrosion Products

3.4. Characterization of Corrosion Products morphologyofand phase Products composition of corrosion products are important factors to improve 3.4.The Characterization Corrosion morphology andofphase composition corrosion products are important factors to improve theThe corrosion resistance Zn based coatings. of Figure 9 shows XRD patterns of corrosion products on The morphology and phase composition of corrosion products are important factors to improve theGI corrosion resistance of Zn based coatings. Figure 9 shows XRD patterns of corrosion products and ZM coatings after salt spray experiment. From Figure 9, it can be seen that corrosion productson theZM corrosion resistance of Zn based coatings. Figure 9 Figure shows XRD of corrosion products on GI in and afterofsalt spray experiment. From 9, Zn it patterns can be8Cl seen corrosion products crystal coatings form consist soluble ZnCl 2, ZnO (zinc oxide), and 5(OH) 2.Hthat 2O (simonkolleite) for GI and ZM coatings after salt spray experiment. From Figure 9, it can be seen that corrosion products in crystal form of soluble ZnCl2 , ZnOIn(zinc oxide), Zn5 (OH) O (simonkolleite) GI, ZM1, andconsist ZM3 coatings, respectively. contrast, theand corrosion products the ZM2 coatingfor 8 Cl2 ·H2on in crystal form consist of soluble ZnCl2, ZnO (zinc oxide), and Zn5(OH)8Cl2.H2O (simonkolleite) for GI,consist ZM1, of and ZM3 respectively. In ZnCl contrast, corrosion products the species. ZM2 coating Znand 5(OH)coatings, 8Cl2.H2O, Zn(OH) 2 without 2 andthe ZnO. Simonkolleite theon main GI, ZM1, ZM3 coatings, respectively. In contrast, the corrosion productsison the ZM2 coating (OH) Cl · H O, Zn(OH) without ZnCl and ZnO. Simonkolleite is the main species. consist of Zn 8 82Cl2.H 2 2O, Zn(OH)22 without ZnCl2 and 2 consist of5 Zn5(OH) ZnO. Simonkolleite is the main species.

Figure 9. XRD patterns of corrosion products on GI and ZM after the salt spray experiment. Figure 9. XRD patterns of corrosion products on GI and ZM after the salt spray experiment. Figure 9. XRD patterns of corrosion products on GI and ZM after the salt spray experiment.

Figure 1010 shows of corrosion corrosionproducts productsononGIGI and ZM coatings after Figure showsinfrared infraredtransmission transmission spectra spectra of and ZM coatings after theFigure salt spray experiment. 10,ofpeak peak sites of ofinfrared infrared transmission spectra of 10spray shows infrared As transmission spectra corrosion products on GI and ZMspectra coatings the salt experiment. As seen seen in in Figure Figure 10, sites transmission of after corrosion products are almost the same, implying the corrosion products have identical phase corrosion products are almost the same, implying the corrosion products have identical phase the salt spray experiment. As seen in Figure 10, peak sites of infrared transmission spectra of corrosion composition. According totothe previous study [3], the the corrosion corrosion products canbebeidentified identified easily. composition. According the previousthe study can easily. products are almost the same, implying corrosion productsproducts have identical phase composition. −1 correspond Broad peaks with the maxima at 3500–3400 cm correspond to vibrations and rotations of water Broad peaks with the maxima at 3500–3400 to vibrations and rotations of water According to the previous study [3], the corrosion products can be identified easily. Broad peaks with −1 −1 −1 −1 molecules and theOH OH group. group. Thepeaks peaks at at 1500, 1390, and 835 cm are ascribed toto ZnZn 5(OH) 6(CO 3)2 3)2 molecules and the The 1390, and 835 cm are ascribed 5(OH) 6(CO − 1 the maxima at 3500–3400 cm correspond to vibrations and rotations of water molecules and the −1 −1 confirm Thepeaks peaksatat905 905 cm cm−1−1 and and 720 cm in in thethe confirm the thepresence presenceofofsimonkolleite simonkolleite −(hydrozincite). 1 group. TheThe OH(hydrozincite). peaks at 1500, 1390, and 835 cm−1 are ascribed to Zn5 (OH)6 (CO 3 )2 (hydrozincite). corrosion products. The results of infrared transmission spectra analyses show the corrosion products corrosion products. The results of infrared transmission spectra analyses show the corrosion products −1 and 720 cm−1 confirm the presence of simonkolleite in the corrosion products. The peaks at 905 cm and ZMcoatings coatingsconsist consistof ofhydrozincite hydrozincite and products onon thethe onon thethe GIGI and ZM and simonkolleite. simonkolleite.The Thecorrosion corrosion products The results of infrared transmission spectra analyses show the corrosion products on the GI and

Materials 2017, 10, 980

11 of 14

ZM coatings consist of hydrozincite and simonkolleite. The corrosion products on the GI and ZM Materials 2017, 10, 780 11 of 14 coatings after the salt spray experiment can be identified combined with the results of XRD. The Materials 2017, 10, 780 11 of 14 corrosion products of GI, ZM1, andspray ZM3experiment coatings consist of ZnCl2 combined , ZnO, Znwith GI and ZM coatings after the salt can be identified the8 Cl results of and 5 (OH) 2 ·H2 O, XRD. The corrosion products of GI, ZM1, and ZM3 coatings consist of ZnCl 2 , ZnO, Zn 5 (OH) 8 Cl 2 .H 2 O, Zn5 (OH) (CO ) . The corrosion products of the ZM2 coating consist of Zn (OH) Cl · H O, Zn(OH) 6 ZM 3 2coatings after the salt spray experiment can be identified combined 5 8 2the 2results of 2, GI and with and Zn 5(OH) 6(CO 3)2Zn . The corrosion products of the ZM2 coating consist of Zn 5a (OH) 8Cl2.H 2O, and Zn (OH) (CO ) . (OH) (CO ) and Zn (OH) (CO ) are believed to be dense corrosion XRD. of consist of ZnCl2, ZnO, Zn5(OH)8Cl2.H2O, 5 The 6corrosion 3 2 products 5 6 GI, ZM1, 3 2 and ZM3 5 coatings 6 3 2 Zn(OH) , respect and6(CO Znto 5(OH)6(CO3)2. Zn5(OH)6(CO3)2 and Zn5(OH)6(CO3)2 are believed to be a dense product with ZnO, which can enhance the corrosion resistance Zn coatings and Zn52(OH) 3)2.porous The corrosion products of the ZM2 coating consist of Znof 5(OH) 8Cl2.H2O, [30]. corrosion product with 6respect to porous ZnO, whichZn can enhance the corrosion resistance Zn Zn(OH) 2, and Zn 5(OH)to (CO 3a ) 2.beneficial Zn 5(OH)6(CO 3)2 andproduct 5(OH) 6(CO 3)2 are believed to be a of dense Zn(OH) is also supposed be corrosion in improvement of corrosion resistance 2 coatings [30]. Zn(OH) 2 is also supposed to be a beneficial corrosion product in improvement of corrosion product with respectthe to porous ZnO, which can enhance the corrosion resistance of Znto its for Zn coatings [16]. Therefore, enhanced corrosion resistance of ZM2 coating is related corrosion[30]. resistance for2 Zn coatings [16]. Therefore, the enhanced corrosion resistance of ZM2 coating coatings Zn(OH) is also supposed to be a beneficial corrosion product in improvement of corrosion products based onproducts the results ofon salt spray experiment. However, itHowever, seems that enhanced is related resistance to its corrosion the results of salt spray experiment. seems corrosion for Zn coatingsbased [16]. Therefore, the enhanced corrosion resistance of ZM2itcoating corrosion resistance of ZM1 and ZM3 coating cannot be ascribed to the difference in the inphase that enhanced of ZM1 coating ascribed toHowever, the difference is related to its corrosion corrosion resistance products based onand the ZM3 results of saltcannot spray be experiment. it seems composition the corrosion products. the phaseofcomposition of the corrosion products. that enhanced corrosion resistance of ZM1 and ZM3 coating cannot be ascribed to the difference in the phase composition of the corrosion products.

Figure 10. Infrared transmission spectraof ofcorrosion corrosion products and ZMZM coatings afterafter the salt Figure 10. Infrared transmission spectra productsononGIGI and coatings the salt spray experiment. sprayFigure experiment. 10. Infrared transmission spectra of corrosion products on GI and ZM coatings after the salt

spray experiment. The corrosion resistance of Zn coatings is closely related to the morphology of the corrosion

The corrosion of Zn coatings morphologies is closely related the morphology products. Figureresistance 11 represents the corrosion of theto corrosion products onofGIthe andcorrosion ZM The corrosion resistancethe of Zn coatingsmorphologies is closely related to the morphology of theon corrosion products. Figure corrosion GI and ZM coatings after 11 therepresents salt spray experiment. As seen in Figure 11,ofa the largecorrosion number ofproducts particle-like porous products. Figure 11 represents the corrosion morphologies of the corrosion products on GI and ZM corrosion products and a experiment. large number As of large on the corrosion surface ofofthe GI coating are coatings after the salt spray seen holes in Figure 11, a large number particle-like porous coatings after the salt spray experiment. As seen in Figure 11, a large number of particle-like porous observed (Figureand 11a).a large number of large holes on the corrosion surface of the GI coating are corrosion products corrosion products and a large number of large holes on the corrosion surface of the GI coating are observed (Figure 11a). observed (Figure 11a).

(a)

(b)

(a)

(b) Figure 11. Cont.

Materials 2017, 10, 980

12 of 14

Materials 2017, 10, 780

12 of 14

Materials 2017, 10, 780

12 of 14

(c)

(d)

Figure 11. SEM micrographs of corrosion products on GI and ZM coatings after salt spray experiment,

Figure 11. SEM micrographs of corrosion products on GI and ZM coatings after salt spray experiment, (a) GI; (b) ZM1; (c) ZM2; and (d) ZM3. (a) GI; (b) ZM1; (c) ZM2; and (d) ZM3. (c)

(d)

EDS analysis shows that these particles with a low Cl− content (6.53 at %) may be ZnO Figure 11. SEM micrographs of corrosion products on GI and ZM coatings after salt spray experiment, − content (Figure 12a,b). These defects reducewith the protection of the corrosion product EDS analysis shows thatremarkably these particles a low Clability (6.53 at %) maylayer be ZnO (a) GI; (b) ZM1; (c) ZM2; and (d) ZM3. to the steel substrate. For this reason, thereduce corrosion of GI coating is very inferior.product In contrast, (Figure 12a,b). These defects remarkably theresistance protection ability of the corrosion layer to a large quantity of the sheet-like corrosion products on ZM coatings are observed (Figure 11b–d). − the steel substrate. For this reason, the corrosion resistance of GI coating is very inferior. In contrast, EDS analysis shows that these particles with a low Cl content (6.53 at %) may be ZnO EDS(Figure analysis shows these sheet-like corrosion products (Figure 12c,d). 12a,b). These defects remarkably reduce the protection ability of simonkolleite theare corrosion product layer a large quantity of the sheet-like corrosion products onare ZMprobably coatings observed (Figure 11b–d). Stoichiometry and XRDFor analysis will the be corrosion conducted to test the corrosion products. Additionally, no to the steel substrate. this reason, resistance of GI coating is very inferior. In contrast, EDS analysis shows these sheet-like corrosion products are probably simonkolleite (Figure 12c,d). corrosion areofobserved on ZM coatings. In thisonsense, enhanced resistance of ZM a large holes quantity the sheet-like corrosion products ZM coatings are corrosion observed (Figure 11b–d). Stoichiometry and XRD analysis will be conducted to test the corrosion products. Additionally, coatings is partly attributed to the dense morphology of corrosion products compared with GI EDS analysis shows these sheet-like corrosion products are probably simonkolleite (Figure 12c,d). no corrosion holes ZM are observed ZM Intest thisthesense, enhanced corrosion resistance Stoichiometry and coatings, XRD analysis will becoatings. conducted to corrosion products. Additionally, no coating. Among a on few particle-like ZnO particles are still observed for ZM1 and ZM3 of ZM coatings is partly attributed to the dense morphology of corrosion products compared corrosion holes are on ZMare coatings. In this sense, enhanced ZMwith coatings. However, noobserved ZnO particles observed for the ZM2 coating.corrosion It can beresistance concludedofthat the GI coating. Among coatings, a few particle-like ZnO particles are products stillinobserved formorphology ZM1 and ZM3 is ZM partly attributed to the dense ofincorrosion compared with GI bestcoatings corrosion resistance can be acquired withmorphology 2.0 wt % Mg ZM coatings terms of the coating. Among coatings, a feware particle-like arecoating. still observed and ZM3that the coatings. However, noZM ZnO particles observedZnO forparticles the ZM2 It canforbeZM1 concluded of the corrosion products. coatings.resistance However, can no ZnO particles are observed coating. It canin beterms concluded that the best corrosion be acquired with 2.0 wtfor % the MgZM2 in ZM coatings of the morphology best corrosion resistance can be acquired with 2.0 wt % Mg in ZM coatings in terms of the morphology of the corrosion products. of the corrosion products.

(b)

(a) (a)

(c) (c)

(b)

(d) (d)

Figure 12. EDS analyses of representative corrosion products on GI and ZM3 coatings after the salt spray experiment. (a) GI; (b) EDS results of GI; (c) ZM3; and (d) EDS results of ZM3.

Materials 2017, 10, 980

13 of 14

4. Conclusions (1)

(2)

(3)

(4)

Phase composition of the ZM coating containing 0.20 wt % Al is the same as that of Zn-Mg coating based on the vertical section of the calculated Zn-0.20 wt % Al-Mg ternary phase diagram near the Al-low corner. Morphology of Zn in the coating changes from bulk to strip, and finally to mesh-like, and MgZn2 changes from rod-like to mesh-like with Mg in the coating increasing from 1.0 to 3.0 wt %. The Mg added to the Zn bath can affect the formation of the Fe2 Al5 inhibition layer. Compared to the GI coating, the corrosion resistance of ZM coatings improves by more than two-fold in terms of the time to first red rust. The expansion rate of red rust for ZM2 coating reduces by more than four-fold. ZM2 coating is the best corrosion resistant coating. The enhanced corrosion resistance of ZM1 and ZM3 coatings is related to the morphology of corrosion products and that of ZM2 is associated with not only the phase composition, but also the morphology of corrosion products.

Acknowledgments: The authors gratefully acknowledge the financial supports from National Natural Science Foundation of China (Project 51204110, 51674237). The authors also faithfully thank Xin Liu from Baosteel Group Corporation for the help of the sample preparation, Minqi Sheng from Shanghai University for the electrochemical experiments, and Jianhua Wang from Changzhou University. Author Contributions: A.D. and Y.L. conceived and designed the experiments; B.L., Y.L., and G.Z. performed the experiments; D.S. and J.W. analyzed the data; H.X. contributed reagents/materials/analysis tools; A.D. and B.S. wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

References 1.

2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12.

Schuerz, S.; Fleischanderl, M.; Luckeneder, G.H.; Preis, K.; Haunschmied, T.; Mori, G.; Kneissl, A.C. Corrosion behaviour of Zn-Al-Mg coated steel sheet in sodium chloride-containing environment. Corros. Sci. 2009, 51, 2355–2363. [CrossRef] Hosking, N.C.; Ström, M.A.; Shipway, P.H.; Rudd, C.D. Corrosion resistance of zinc-magnesium coated steel. Corros. Sci. 2007, 49, 3669–3695. [CrossRef] Prosek, T.; Nazarov, A.; Bexell, U.; Thierry, D.; Serak, J. Corrosion mechanism of model zinc-magnesium alloys in atmospheric conditions. Corros. Sci. 2008, 50, 2216–2231. [CrossRef] Shimizu, T.; Fukio, Y.; Yasushi, M.; Atsushi, A. Corrosion products of hot-dip Zn-6%Al-3%Mg coated steel sheet subjected to atmospheric exposure. Tetsu-to-Hagané 2003, 89, 166–173. [CrossRef] Prosek, T.; Larché, N.; Vlot, M.; Goodwin, F.; Thierry, D. Corrosion performance of Zn-Al-Mg coatings in open and confined zones in conditions simulating automotive applications. Mater. Corros. 2010, 61, 412–420. [CrossRef] Nishimura, K.; Hidetoshi, S.; Hiromasa, N.; Kenji, K. Corrosion resistance of hot dip Zn-Mg galvanized steel sheet. Tetsu Hagane-J. Iron Steel Inst. Jpn. 2003, 89, 174–179. [CrossRef] Volovitch, P.; Allely, C.; Ogle, K. Understanding corrosion via corrosion product characterization: I. Case study of the role of Mg alloying in Zn-Mg coating on steel. Corros. Sci. 2009, 51, 1251–1262. [CrossRef] De Bruycker, E.; Zermout, Z.; Cooman, B.C. Zn-Al-Mg coatings: Thermodynamic analysis and microstructure related properties. Mater. Sci. Forum 2007, 539, 1276–1281. [CrossRef] Suarez, L.; Leysen, F.; Masquelier, C.; Warichet, D.; Houbaert, Y. Effects of Mg additions on surface morphology and corrosion resistance of hot-dipped Zn coatings. Defect Diffus. Forum 2008, 273, 300–305. Nomura, H.; Kimata, Y.; Kanai, H. Corrosion resistance of prepainted Zn-11%Al-3%Mg-0.2%Si coated steel sheet. In Proceedings of the Galvatech 2004 Conference, Chicago, IL, USA, January 2004; pp. 761–770. Schürz, S.; Luckeneder, G.H.; Fleischanderl, M.; Mack, P.; Gsaller, H.; Kneissl, A.C.; Mori, G. Chemistry of corrosion products on Zn-Al-Mg alloy coated steel. Corros. Sci. 2010, 52, 3271–3279. [CrossRef] Jonathan, E.; Spittle, J.A.; Sullivan, J.H.; Worsley, D.A. The effect of magnesium additions on the microstructure and cut edge corrosion resistance of zinc aluminium alloy galvanised steel. Corros. Sci. 2008, 50, 1650–1658.

Materials 2017, 10, 980

13.

14. 15. 16. 17. 18. 19. 20.

21. 22.

23. 24. 25.

26.

27.

28. 29. 30.

14 of 14

Myeong-Hoon, L.; Kim, J.-D.; Oh, J.-S.; Yang, J.-H.; Baek, S.-M. Morphology and crystal orientation on corrosion resistance of Mg thin films formed by PVD method onto Zn electroplated substrate. Surf. Coat. Technol. 2008, 202, 5590–5594. Bernd, S.; Schwerdt, C.; Seyfert, U.; Zimmer, O. Innovative steel strip coatings by means of PVD in a continuous pilot line: Process technology and coating development. Surf. Coat. Technol. 2003, 163, 703–709. Monojit, D.; Halder, A.K.; Singh, S.B. Morphology and properties of hot dip Zn-Mg and Zn-Mg-Al alloy coatings on steel sheet. Surf. Coat. Technol. 2010, 205, 2578–2584. Masao, M.; Koyama, K.; Murase, M.; Mori, Y. Improvement in the corrosion resistance of zinc-plated steel by electrodeposition of magnesium from a molten salt. ISIJ Int. 1996, 36, 714–719. Masao, M.; Koyama, K.; Mori, Y. Inhibition of anodic dissolution of zinc-plated steel by electrodeposition of magnesium from a molten salt. ISIJ Int. 1997, 37, 55–58. Masao, M.; Koyama, K.; Mori, Y. Self-healing ability of zinc-plated steel coated with magnesium electrodeposited from a molten salt. Materials Transactions. JIM 1997, 38, 719–723. Matsuda, H.; Matsuda, Y.; Kabasawa, M. Effects of aluminium in the Zn coating on electrode life in welding galvanized steel sheet. Weld. Int. 1996, 10, 605–613. [CrossRef] Panossian, Z.; Mariaca, L.; Morcillo, M.; Flores, S.; Rocha, J.; Peña, J.J.; Herrera, F.; Corvo, F.; Sanchez, M.; Rincon, M.; et al. Steel cathodic protection afforded by zinc, aluminium and zinc/aluminium alloy coatings in the atmosphere. Surf. Coat. Technol. 2005, 190, 244–248. [CrossRef] Liang, H.; Chen, S.; Chang, Y. A thermodynamic description of the Al-Mg-Zn system. Metall. Mater. Trans. A 1997, 28, 1725–1734. [CrossRef] Li, B.; Dong, A.; Zhu, G.; Chu, S.J.; Qian, H.W.; Hu, C.G.; Sun, B.D.; Wang, J. Investigation of the corrosion behaviors of continuously hot-dip galvanizing Zn-Mg coating. Surf. Coat. Technol. 2012, 206, 3989–3999. [CrossRef] Okamoto, H. Comment on Mg-Zn (magnesium-zinc). J. Phase Equilib. 1994, 15, 129–130. [CrossRef] Monojit, D.; Singh, S.B. Effect of strip temperature on the formation of an Fe2 Al5 inhibition layer during hot-dip galvanizing. Scr. Mater. 2009, 60, 643–646. Song, G.-M.; Sloof, W.G. Effect of alloying element segregation on the work of adhesion of metallic coating on metallic substrate: Application to zinc coatings on steel substrates. Surf. Coat. Technol. 2011, 205, 4632–4639. [CrossRef] Song, G.M.; Vystavel, T.; Van Der Pers, N.; De Hosson, J.T.M.; Sloof, W.G. Relation between microstructure and adhesion of hot dip galvanized zinc coatings on dual phase steel. Acta Mater. 2012, 60, 2973–2981. [CrossRef] Volovitch, P.; Vu, T.N.; Allély, C.; Abdel, A.A.; Ogle, K. Understanding corrosion via corrosion product characterization: II. Role of alloying elements in improving the corrosion resistance of Zn-Al-Mg coatings on steel. Corros. Sci. 2011, 53, 2437–2445. [CrossRef] James, S.; Mehraban, S.; Elvins, J. In situ monitoring of the microstructural corrosion mechanisms of zinc-magnesium-aluminium alloys using time lapse microscopy. Corros. Sci. 2011, 53, 2208–2215. Pardo, A.; Merino, M.C.; Coy, A.E.; Arrabalb, R.; Viejob, F.; Matykinab, E. Corrosion behaviour of magnesium/aluminium alloys in 3.5 wt % NaCl. Corros. Sci. 2008, 50, 823–834. [CrossRef] Ishikawa, T.; Masato, U.; Kazuhiko, K.; Takenori, N. Air permeability of the artificially synthesized Zn-Al-Mg alloy rusts. Corros. Sci. 2007, 49, 2547–2556. [CrossRef] © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).