# Fileset

[ECS Research Manuscript final.pdf](https://mdr.nims.go.jp/filesets/423c419a-858f-4ff7-b1ad-0282749f33ac/download)

## Creator

[Sachiko Hiromoto](https://orcid.org/0000-0003-4666-6708), Kasumi Fukuzawa, [Kotaro Doi](https://orcid.org/0000-0002-5204-1088), Makoto Chiba, Naofumi Ohtsu

## Rights

This is an author-created, un-copyedited version of an article accepted for publication/published in Journal of The Electrochemical Society. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at https://doi.org/10.1149/1945-7111/ad13db.[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[Corrosion Protection Behavior of Aluminate Ion-Loaded Layered Double Hydroxide Coating on AZ31 Magnesium Alloy](https://mdr.nims.go.jp/datasets/4f7b51ab-f98b-4f29-a61e-f0850e7dc0cc)

## Fulltext

ECS Research Manuscript final.docxCorrosion Protection Behavior of Aluminate Ion-Loaded Layered Double Hydroxide Coating on AZ31 Magnesium Alloy   Sachiko Hiromoto1, z, Kasumi Fukuzawa2, Kotaro Doi1, Makoto Chiba1, 2 and Naofumi Ohtsu3   1 Research Center for Structural Materials, National Institute for Materials Science, 305-0047 Tsukuba, Japan 2 Department of Materials Chemistry, National Institute of Technology, Asahikawa College, 071-8142 Asahikawa, Japan 3 Faculty of Engineering, Kitami Institute of Technology, 090-0015 Kitami, Japan  z Corresponding author: S. Hiromoto E-mail Address [hiromoto.sachiko@nims.go.jp]   Abstract Effect of aluminate ion loading on the corrosion behavior of layered double hydroxide (LDH) coating of Mg-3Al-1Zn (AZ31) alloy was investigated. The corrosion inhibition performance of NaAlO2 for AZ31 was examined by immersion and polarization tests. Then, aluminate ion-loaded LDH (LDHAlO2) was synthesized from hydrotalcite (LDHCO3) and co-deposited with magnesium and aluminum double hydroxide on AZ31 by electrophoretic co-deposition. Polarization, electrochemical impedance (EI) and wet-dry cyclic corrosion tests were conducted on the LDHAlO2- and LDHCO3-coated specimens. Adding 10-50 mmol/L NaAlO2 to 0.1 mol/L NaCl solution induced a clear passive region on the polarization curves of AZ31 and shifted the breakdown potential over -1.0 V (Ag/AgCl), indicating the corrosion inhibition property of NaAlO2. In the polarization and EI tests, the aluminate ion loading did not noticeably enhance the corrosion protection ability of the LDH coating as shown by higher quasi-passive current density and lower Rp than without loading. However, in the wet-dry corrosion tests, the LDHAlO2-coated AZ31 demonstrated less weight gain and fewer clusters of shallow micro-pits, while the LDHCO3-coated AZ31 showed numerous deeper pits on the entire surface. It was revealed that loading aluminate ions to the LDH coatings is promising for enhancing their corrosion protection ability in atmospheric environments.   Introduction Magnesium alloys are attractive light-weight materials for automobiles, aircrafts and so on 1, 2. However, their practical use is limited by the low corrosion resistance. To overcome this issue, various anti-corrosion coatings such as phosphate-based and fluoride-based coatings have been developed by various coating methods such as chemical conversion, anodization and plating 3, 4. As an anticorrosion coating for Mg and Al alloys, layered double hydroxide (LDH) coatings are one of the promising coatings. LDH consists of a stacking of basic layers of double hydroxides of divalent and trivalent metal ions, with various anions in the interlayer and shows an anion exchange property 5. Due to the anion exchange property, LDH is expected to be a reservoir of corrosion inhibitors and/or an absorber of chloride ions 6. Then, corrosion inhibitor-loaded LDH coatings have been developed for Mg and Al alloys and steels to enhance the corrosion protection ability and to add self-healing property 6-15. The corrosion inhibitors such as MoO42-, VO3-, WO42-, phenyl phosphonic acid (PPA), 2-mercaptobenzothiazole (MBT) and 8-hydroxyquinoline (8HQ) were loaded to the LDH coatings 10-15. Meanwhile, suitable anionic corrosion inhibitors for loading to LDH coatings are still in exploration. As a corrosion inhibitor for Mg alloys, Na2MoO4 Na2SnO3 and NaVO3 have been proposed 16-21. In addition to these, we expect NaAlO2 and KAlO2 to act as a corrosion inhibitor because the oxide film containing more than 4% Al shows high corrosion resistance for AZ series Mg alloys 22-24 and KAlO2 in the electrolyte deposits to form MgAl2O4 in the anodic oxide film of Mg alloy 25. To form the corrosion inhibitor-loaded LDH coatings, generally, LDH coatings with NO3- or CO3- ions in the interlayer (LDHNO3 or LDHCO3) are formed on the alloys by steam treatment, autoclaving, chemical conversion and electrochemical deposition, and then the interlayer NO3- or CO3- ions are exchanged with anionic inhibitors in aqueous solutions 7-15. The electrophoretic co-deposition (EP-co-D) method enables the LDH particles, which are pre-loaded with various corrosion inhibitors, to deposit on a variety of metal surfaces together with metallic hydroxide 26-28. We therefore employed the EP-co-D method and developed composite coatings of LDHCO3 particles and magnesium and aluminum double hydroxide (Mg-Al-DH) on Mg-3Al-1Zn (AZ31) alloy 29. Another finding of this previous study was that LDHNO3 formed the bottom layer in the EP-co-D coating. Another finding was that submicron to micron pores were formed in the EP-co-D layer by H2 gas generated by the electrolysis of water in the 80vol% ethanol-20vol% water mixed electrolyte 29. Recently, the water content of the electrolyte was decreased to 2.4vol% to reduce the pore formation 30. The purpose of this study was to develop a corrosion inhibitor-loaded LDH coating on Mg alloys. In this study, first, the effect of NaAlO2 on the corrosion of AZ31 was examined as a potential corrosion inhibitor by immersion and polarization tests. Then, aluminate ion-loaded LDH (LDHAlO2) was synthesized. The LDHAlO2 particles were electrodeposited on AZ31 together with Mg-Al-DH by EP-co-D. The corrosion protection behavior of the coatings was examined by polarization, electrochemical impedance (EI) and wet-dry cyclic corrosion tests because the wet-dry cyclic environment is similar to the service environment of structural materials.  Experimental Immersion tests of AZ31 in NaCl solution with NaAlO2 AZ31 (Al 2.9mass%, Zn 1.05mass%, Mn 0.41mass%, Si 0.01mass%, Fe 0.001mass%, Cu <0.01mass%, Ni <0.001mass%, Mg Bal.) magnesium alloy plates of 15×15 mm and 2 mm thick were used as a specimen. The surface was ground with SiC papers (PSI, USA) up to #1200 and ultrasonically rinsed in 2-propanol. The specimen surface was covered with TeflonTM tape, leaving a test surface of 1 cm2. For immersion tests, 0.1 mol/L NaCl (Fujifilm-Wako) with and without addition of 1–50 mmol/L NaAlO2 (Fujifilm-Wako) were prepared. The as-polished AZ31 were immersed in 200 mL of the respective solutions at room temperature for 24 h. The pH of the respective solutions and the weight of the specimen were measured before and after the immersion. The appearance of the immersed specimen was observed using digital camera. The specimen immersed with NaAlO2 was characterized using scanning electron microscope (SEM: Miniscope TM3030Plus, Hitachi) and X-ray photoelectron spectroscopy (XPS: PHI 5000 VersaProbe, ULVAC-PHI). In the XPS analysis, monochromatized Al Kα radiation (h=1486.6 eV) was adopted as the X-ray source, and the pass energy of the spectrometer was set to 23.5 eV. The photoelectron take-off angle was 45° and the diameter of the X-ray probe was about 100 m. The binding energies of all the spectra were referenced to the C 1s peak at 285 eV. The background subtraction was carried out using Shirley’s method. Spectral deconvolution was carried out mathematically using a Gaussian-Lorentzian function.  Polarization tests of AZ31 in NaCl solution with NaAlO2 The surface of AZ31 plates of 15×15 mm was ground and covered with TeflonTM tape in the same manner as the specimen for the immersion tests. 200 mL of 0.1 mol/L NaCl with and without 10 or 50 mmol/L NaAlO2 solutions at room temperature was used as electrolyte. The counter and reference electrodes were a platinum wire and a Ag/AgCl electrode, respectively. The specimen was immersed in the electrolyte for 0.5 h to stabilize the open circuit potential (OCP). Subsequently the potential of the specimen was swept from a 50 mV lower or higher potential than the stabilized OCP in anodic or cathodic direction, respectively, at a sweep rate of 1 mV/s, using potentiostat (Interface 1010, Gamry).  Synthesis of aluminate ion-loaded LDH The loading of aluminate ion (AlO2- ion) to LDH was carried out by the reconstruction process 31, 32. 5 g of LDHCO3 particles (Mg6Al2(OH)16CO3・4H2O, hydrotalcite, Sigma Aldrich) with a diameter of about 1 m was heated at 500°C for 1 h and about 3.2 g of the thermal decomposed LDH particles was obtained. Subsequently, the entire thermal-decomposed LDH powder were suspended in 200 mL of 0.24 mol/L NaAlO2 solution for 24 h at room temperature with stirring in a sealed glass vessel. After the 24 h rehydration process, the suspension was filtered, rinsed with deionized water and dried at 100°C for about 1 h, and about 4.8 g of LDHAlO2 powder was obtained. The as-synthesized LDHAlO2 powder were characterized using X-ray diffraction (XRD: D2 Phaser, Bruker) with a tube voltage of 30 kV, a tube current of 10 mA, and an increment of 0.02 degrees. To examine the intercalation of aluminate ions into LDH, the interlayer water was reduced by heat treatment at 140°C for 10 h for the as-synthesized LDHAlO2 and as-received LDHCO3 powders, and then XRD measurements were performed. The as-synthesized LDHAlO2 and as-received LDHCO3 powders were also characterized using diffuse reflectance Fourier transform infrared spectroscopy (FTIR: IRTracer-100, Shimadzu) with a resolution of 4 cm-1, confocal Raman microscope (inVia Reflex, Renishaw) with a laser of 532 nm wavelength irradiated through a 50 times objective lens, and SEM (Miniscope TM3030Plus, Hitachi) equipped with energy dispersive X-ray spectroscopy (EDS: X-stream-2, Oxford) with an accelerating voltage of 15 kV, and the results are shown in the supplementary material. Aluminate ion release from LDHAlO2 particles in air-exposed pure water was examined using aluminium ion test paper (Aluminium Check, ADVANTEC) and the results are shown in the supplementary material.  Composite coating of LDHAlO2 or LDHCO3 particles and Mg-Al DH by EP-co-D AZ31 plates of 25×30 mm and 2 mm thick were used as a substrate. The surface was ground with SiC papers up to #1200 and ultrasonically rinsed in 2-propanol. Immediately before the EP-co-D treatment, the substrate was immersed in 0.25 mol/L C10H14N2Na2O8·2H2O (FUJIFILM Wako) solution for 10–20 sec to remove the air formed oxide film. An electrolyte of 60 mmol/L Mg(NO3)2·6H2O (FUJIFILM Wako) and 15 mmol/L Al(NO3)3·9H2O (FUJIFILM Wako) in 2.4vol% water-2-Propanol was prepared. Subsequently, LDHAlO2 or LDHCO3 particles were suspended in the electrolyte at a concentration of 2wt%. The suspensions were stirred for several minutes and then ultrasonicated for longer than 0.5 h. AZ31 and carbon plates were mounted in an electrochemical cell with a distance of 10 mm as the deposition (working) and counter electrodes, respectively. Immediately after immersing the electrodes in the suspension, a constant voltage of 100 V was applied between the electrodes for 60 s. The EP-co-D specimen was dried at 100°C for 1 h, and then ultrasonicated in 2-propanol for 60 s to remove weakly glued particles. The weight of the specimens was measured before and after the EP-co-D and ultrasonication. The appearance of the specimens was observed using digital camera and the surface and cross section were observed using SEM in backscattered electron mode. The surface was characterized using XRD and confocal Raman microscope. The AZ31 coated with LDHAlO2 and LDHCO3 particles was denoted as LDHAlO2-coated and LDHCO3-coated AZ31 specimens, respectively. The as-polished AZ31 was denoted as uncoated AZ31 in contrast to the coated AZ31.  Polarization and EI tests of LDHAlO2- and LDHCO3-coated AZ31 Anodic polarization and EI tests were carried out for the LDHAlO2-coated, LDHCO3-coated and uncoated AZ31 specimens using potentiostat (SP-50e, Biologic). The counter and reference electrodes were a platinum wire and a Ag/AgCl electrode, respectively. The specimen was exposed the test surface of 1 cm2 and the surrounding area was coated with epoxy. 200 mL of 0.1 mol/L NaCl solution at room temperature was used as electrolyte. For the polarization tests, the specimen was immersed in the electrolyte for 0.5 h to stabilize OCP, and subsequently the potential of the specimen was swept from a 50 mV lower potential than the stabilized OCP in anodic direction at a sweep rate of 1 mV/s. For the EI tests, after the static immersion for 0.5 h, potential perturbation was applied with an amplitude of 5 mV at frequency from 10 MHz to 50 mHz and 5 points per decade. Equivalent electric circuits were assumed and curve-fitting for the measured EI spectra was performed to obtain the values of parameters in the circuits using ZView. Three EI tests were carried out for each specimen.  Wet-dry cyclic corrosion tests of LDHAlO2- and LDHCO3-coated AZ31 Wet-dry cyclic corrosion tests were carried out for the LDHAlO2-coated, LDHCO3-coated and uncoated AZ31 specimens. 0.3wt% NaCl-75% ethanol solution was dropped onto the 20×30 mm test surface to place 1 g/m2 NaCl. Then, the relative humidity around the specimens was controlled with one cycle of 8 h at 30%RH, 8 h at 95%RH and 8 h at 30%RH again at room temperature, and 14 cycles were repeated. The weight of the specimens before and after the corrosion tests was measured for three samples from each specimen. The appearance of the specimens was observed using digital camera and the surface and cross section was observed using SEM. The surface was characterized using XRD. The coatings in the apparently uncorroded area were analyzed using laser Raman spectroscopy. The corrosion products and the coatings were chemically removed from one corroded sample from each specimen using chromate-AgNO3 solution and the surface topography and morphology were observed using one-shot 3D microscope (VR-3100, Keyence) and SEM.  Results Immersion tests of AZ31 in NaCl with NaAlO2 Figure 1 shows the appearance of the as-polished AZ31 surface immersed in the 0.1 mol/L NaCl solution with and without NaAlO2, and a back-scattered electron SEM image of the surface immersed with 50 mmol/L NaAlO2. The weight gain of the specimens and the pH of the solutions before and after immersion are shown in Fig. 2. The weight of Mg alloys generally increases by corrosion due to the formation of corrosion products such as Mg(OH)2. As the change in surface morphology was evident and the effect of NaAlO2 was clear in the following polarization tests, weight measurements after removal of the corrosion products were not carried out. In the NaCl-only solution, about half of the surface area shows black filiform corrosion, and the entire surface is covered by a white corrosion product (Fig. 1(a)). The weight of the specimen increases (Fig. 2), and the solution pH increases from 6.1 to 9.2. The weight gain in the NaCl-only solution shows large variation, which is due to the loss of some of the corrosion products during rinsing and drying. The addition of 1 mmol/L NaAlO2 decreases the filiform corrosion area (Fig. 1(b)). No obvious corrosion is observed with 10 and 50 mmol/L NaAlO2 (Fig. 1(c) and (d)). The SEM image with 50 mmol/L NaAlO2 shows no obvious corrosion except for dark contrast spots of 5-30 m mainly around Al-Mn inclusions (bright contrast grains) (Fig. 1(e)). The weight gain decreases with an increase of NaAlO2 addition (Fig. 2). The pH of the NaCl solution with 1-50 mmol/L NaAlO2 is 9.6, 10.9 and 11.7, respectively, which does not significantly change after the immersion. Figure 3(a) shows the XPS survey spectra of the AZ31 specimen as-polished and immersed with 10 and 50 mmol/L NaAlO2. The photoelectron peaks from magnesium (Mg), aluminum (Al), silicon (Si), chlorine (Cl), oxygen (O), and carbon (C) are observed. Cl is derived from contaminated or residual NaCl.  Figure 3(b)-(e) shows the XPS narrow spectra of the C 1s, O 1s, Al 2p, Mg 2p and Zn 2p3 regions. The C 1s spectra (Fig. 3(b)) are decomposed into two peaks originating from C-C bond and carbonate. The carbonate peak is derived from magnesium carbonate formed from atmospheric CO2. The O 1s spectra (Fig. 3(c)) are decomposed into two peaks from MgO and Al2O3+Mg(OH)2. The MgO peak position is shifted by about 0.5 eV after immersion compared to that of the as-polished AZ31, which is attributed to the effect of adsorbed water. The Al 2p spectra (Fig. 3(d)) of the AZ31 immersed with NaAlO2 show a peak from Al2O3, and that of the as-polished AZ31 is decomposed into peaks from metallic Al and Al2O3. The intensity of Al2O3 peak of the specimens with NaAlO2 is significantly higher than that of the as-polished AZ31, indicating that an Al2O3 film is formed in the solution with NaAlO2. The Mg 2p spectrum (Fig. 3(e)) of the as-polished AZ31 is decomposed into two peaks from metallic Mg and MgO and/or Mg(OH)2, while that with NaAlO2 shows a peak from MgO and/or Mg(OH)2. The intensity of the Mg 2p spectra of the specimens with NaAlO2 is lower than that of the as-polished AZ31, indicating a decrease in the magnesium fraction in the surface oxide film. Although the C 1s spectra indicates the presence of slight amount of MgCO3, no distinct peak from MgCO3 is observed in Mg 2p spectra. The Zn 2p3 spectra (Fig. 3(f)) of the specimens with NaAlO2 show a peak derived from zinc oxide, while that of the as-polished AZ31 does not show any peak. These results indicate that after immersing in the NaCl solution with NaAlO2, a relatively thick Al2O3 film with a small fraction of MgO/Mg(OH)2 was formed.  Polarization tests of AZ31 in NaCl with NaAlO2 Figure 4 shows the anodic and cathodic polarization curves of AZ31 in the 0.1 mol/L NaCl solution with and without NaAlO2. The anodic polarization curves show a constant current density and a breakdown potential. The constant current density decreases by 100 times to be a passive current density of 3×10-7 Aꞏcm-2 with 10 and 50 mmol/L NaAlO2. The breakdown potential drastically increases from -1.35 V (vs. Ag/AgCl) to -1.15 V with 10 mmol/L NaAlO2. No breakdown potential appears in the potential region to -1.0 V with 50 mmol/L NaAlO2. The cathodic current density decreases by 100 times with NaAlO2. It is demonstrated that NaAlO2 suppresses the cathodic and anodic reactions on AZ31.  Synthesis of aluminate ion-loaded LDH Figure 5 shows the wide-range XRD patterns of the as-synthesized LDHAlO2 and as-received LDHCO3 powders and the 140°C-treated LDHAlO2 and LDHCO3 powders, and the magnified diffraction peaks from LDH 003 plane ((003)LDH). The as-received and 140°C-treated LDHCO3 powders show a typical XRD pattern of LDH structure of hydrotalcite (JCPDS card 22–700). The as-synthesized and 140°C-treated LDHAlO2 powders show broad diffraction peaks from LDH structure with relatively low intensity, small diffraction peaks from Al(OH)3 (boehmite) 33 and a slight peak from thermally decomposed LDH. Al(OH)3 is insoluble in aqueous solutions at room temperature. Thus, the as-synthesized LDHAlO2 powder was used for the EP-co-D. The position of the (003)LDH peak of the as-synthesized LDHAlO2 is 11.52 degrees which is similar to 11.56 degrees of LDHCO3 (Fig. 5(b)). The intercalation of aluminate ions to LDHAlO2 could not be determined from this peak position. The 140°C-treatment shifts the (003)LDH peak of the LDHAlO2 to the higher angle side of 12.25 degrees, whereas that does not change the (003)LDH peak position of the LDHCO3.  Composite coating of LDHAlO2 or LDHCO3 particles and Mg-Al-DH Figure 6 shows the optical and SEM images of the surface of the LDHAlO2- and LDHCO3-coated AZ31 specimens after ultrasonication. The LDHAlO2 and LDHCO3 coatings cover the surface uniformly (Fig. 6(a) and (b)) but show cracking under SEM observation (Fig. 6(c) and (d)). The weight of the LDHAlO2 coating was 6.8±0.9 mg that was smaller than 8.2±1.4 mg of the LDHCO3 coating. The magnified SEM images of the coatings show that LDH particles are glued with gel-like dark contrast substance, which was attributed to Mg-Al-DH by the following XRD analysis (Fig. 6(e) and (f)). A tiny amount of Al(OH)3 needle-shaped particles is deposited in the LDHAlO2 coating. The magnified SEM images of the substrate exposed inside coating cracks show that the gel-like Mg-Al-DH and LDH particles densely cover the substrate surface (Fig. 6(g) and (h)).  Figure 6(i) and (j) shows the cross section of the LDHAlO2- and LDHCO3-coated AZ31 specimens after ultrasonication. The thickness of the coating layer was not uniform, and the coating in Fig. 6(i) appears thicker than in (j), but this does not necessarily the case for the entire coating. The cross-sectional SEM images show that the LDHAlO2 coating is coarser than the LDHCO3 coating (Fig. 6(i) and (j)). The LDHCO3 coating shows that the gel-like Mg-Al-DH deposits preferentially at the bottom of the coating to form a dense inner layer (Fig. 6 (j)) as previously reported 29. Such a preferential deposition of the gel-like Mg-Al-DH at the bottom of the LDHAlO2 coating is not significant on the cross-sectional image, but the formation of the thin dense inner layer is suggested by the surface observation (Fig. 6(i) and (g)). Figure 7(a) shows the wide-range XRD patterns of the LDHAlO2- and LDHCO3-coated AZ31 specimens after ultrasonication. The diffraction peaks from LDH and substrate AZ31 are observed for both specimens, and a very small peak from -Al(OH)3 is observed for the LDHAlO2-coated AZ31. Additionally, broad peaks are observed at around 18‒19 degrees and 32‒40 degrees which correspond to 18.649 degrees and 38.0154 degrees of brucite Mg(OH)2 (JCPDS card No. 00-44-1482), respectively. The magnified (003)LDH peaks (Fig. 7(b)) are asymmetrical owing to a shoulder peak on the lower angle side which is attributed to LDHNO3 as reported in our previous work 29. Formation of LDHNO3 is also suggested by the result that it was formed in the Mg(NO3)2 and Al(NO3)3 electrolyte without LDH particles by electrodeposition as shown in Fig. S3. Thus, the gel-like substance with dark contrast in the SEM images (Fig. 6) was attributed to Mg-Al-DH with brucite structure, which should have been a precursor to LDHNO3.  Polarization and EI behavior of LDHAlO2- and LDHCO3-coated AZ31 Figure 8(a) shows the anodic polarization curves of the LDH-coated and uncoated AZ31 specimens in the 0.1 mol/L NaCl solution. The LDHAlO2- and LDHCO3-coated AZ31 specimens show a quasi-passive current density of about 1×10-5 A cm-2 and 5×10-6 A cm-2, respectively. These are less than half of the constant current density of 3×10-5 A cm-2 of the uncoated AZ31. The breakdown of the LDHAlO2- and LDHCO3-coated AZ31 surfaces does not occur in this potential range up to -1 V. Figure 8(b) and (c) shows the Nyquist and Bode plots of EI spectra of the LDH-coated and uncoated AZ31 specimens in the 0.1 mol/L NaCl solution. The LDH-coated AZ31 specimens apparently show one capacitive semi-circle on the Nyquist plots and one peak at around 40 Hz on the phase shift curves, so an equivalent electric circuit shown in Fig. 8(d) was assumed. Rs represents the solution resistance. Rcoat and CPEcoat represent the resistance and constant phase element (CPE) of the LDH coatings, respectively. The CPE was used instead of the capacitance due to the porous structure of the LDH coatings. The uncoated AZ31 shows two semi-circles on the Nyquist plot and two peaks at around 80 Hz and 2×10-1 Hz on the phase shift curves, so an equivalent circuit shown in Fig. 8(e) was assumed. The magnified 2×10-1 Hz peak is shown as an inset in Fig. 8(c). Rfilm and CPEfilm represent the resistance and of the surface oxide film, respectively. Rct and CPEdl represent the charge transfer resistance and CPE of electric double layer, respectively. Curve fitting was performed, and the fitted curves are shown in Fig. 8(b) and (c). The obtained corrosion parameters are summarized in Table 1. The polarization resistance (Rp) of the uncoated AZ31 was estimated using the Eq. (1) and shown in Table 1 and Fig. 8(f).   (1)  The Rp of the LDH-coated AZ31 specimens is more than twice as high as that of the uncoated AZ31. The Rp of the LDHAlO2-coated AZ31 is slightly lower than that of the LDHCO3-coated AZ31. The Rcoat values of the LDH-coated specimens are higher than the Rfilm of the uncoated specimen. The CPE-T values, roughly corresponding to capacity of the surface layer, are comparable for the LDH-coated and uncoated AZ31 specimens and the CPE-P values are similar between the specimens.  Wet-dry cyclic corrosion behavior of LDHAlO2- and LDHCO3-coated AZ31 Figure 9 shows the weight gain of the LDH-coated and uncoated AZ31 specimens after the wet-dry cyclic corrosion tests. The weight measurement after removal of the coating and corrosion products was not performed because the difference in the weight gain and corrosion morphology between specimens was evident. The weight gain of the LDH-coated AZ31 specimens is significantly smaller than the uncoated AZ31, and that of the LDHAlO2-coated AZ31 is smaller than that of the LDHCO3-coated AZ31. Figure 10(a)-(f) shows the optical images of the specimens after the corrosion tests. Figure 10(g) and (h) shows the SEM images of the white dots arrowed in the optical images and reveals that the dots are corrosion sites or pits. Almost the entire surface of the uncoated AZ31 is colored in brown and covered with micrometer-sized corrosion pits, and some of them grow to millimeter-sized pits (Fig. 10(f)). The LDHAlO2-coated AZ31 shows small number of micrometer- to millimeter-sized corrosion sites (Fig. 9(a) and (d)), and the LDHCO3-coated AZ31 shows a lot of micrometer-sized corrosion pits on the entire surface (Fig. 9(b) and (e)). Figure 10(i) and (j) shows the magnified coating layer in apparently uncorroded area. Figure 10(k) and (l) shows the cross-sectional SEM images. The LDHAlO2 coating in the apparently uncorroded area remains almost original morphology. Both the LDH coatings show amorphous substance between particles, and the amount of amorphous substance deposited in the LDH AlO2 coating appears to be smaller than that in the LDHCO3 coating. Figure 11 shows the wide-range XRD patterns and magnified (003)LDH peak of the specimens after the corrosion tests. The corroded LDHAlO2-coated AZ31 shows diffraction peaks from LDH and substrate AZ31, and no obvious peaks from corrosion products are observed. The corroded LDHCO3-coated AZ31 shows peaks from LDH, substrate AZ31 and a corrosion product of dypingite (Mg5(CO3)4(OH)2·5H2O). The corroded uncoated AZ31 shows broad peaks from LDH and dypingite in addition to those from substrate AZ31. The position of (003)LDH peak of the corroded LDH-coated specimens is 11.60 degrees which does not apparently change from that of the as-coated specimens (Fig. 7). To examine the change of the interlayer anions of LDH due to corrosion, the coatings before and after the corrosion tests were analyzed using Raman spectrometer. The vibrational region of interlayer anions from 1000 to 1150 cm-1 34-37 is shown in Fig. 12. For the specimens after the corrosion tests, the coatings in a visually uncorroded area were analyzed. The area without needle-shaped particles was defined as the analysis area for the LDHAlO2 coating. The LDHAlO2 coating after the corrosion test shows a peak at 1051 cm-1, a tiny peak at 1067 cm-1 and a broad peak at 1101 cm-1, while the as-prepared LDHAlO2 coating shows a relatively large peak at 1045 cm-1 and broad shoulder peaks at around 1058 cm-1 and small peaks at 1088 and 1109 cm-1. The peak at 1045 cm-1 is generally attributed to interlayer NO3- ions of LDH 38. The peak at around 1060 cm-1 is presumably attributed to interlayer AlO2- and CO32- ions, as the peak at around 1060 cm-1 is attributed to the symmetric stretching vibrations of interlayer CO32- ions of Mg-Al-LDHCO3 34-37 and the peaks near 1056 and 1068 cm-1 are attributed to Al-O bending mode of -Al(OH)3 39, 40. The peaks at 1087 and around 1100 cm-1 are attributed to free CO3- ions 37 and CO32- symmetric stretching vibrations in magnesite (MgCO3) 37, respectively. The relative intensity of the 1045 and 1060 cm-1 peaks to the 1085 and 1100 cm-1 peaks decreases after the corrosion test, indicating that the interlayer NO3- and AlO2- ions were exchanged with CO32- ions. The LDHCO3 coating after the corrosion shows the peaks at 1050, 1063, 1090 and 1122 cm-1. The as-prepared LDHCO3 coating shows the peaks at 1045, 1061 and 1086 cm-1. The relative intensity of the 1045 cm-1 peak to the 1060 cm-1 peak decreases after the corrosion, indicating the anion exchange between interlayer NO3- and CO32- ions. The 1123 cm-1 peak is attributed to CO32- stretching vibrations in dypingite 36, 41. Figure 13 shows the optical, SEM and topography images and the corresponding line profiles after removing the coatings and corrosion products. The corrosion site and pit arrowed in Fig. 13(a) and (b) corresponds to those in the SEM images of Fig. 13(c) and (d), respectively and to the white dots arrowed in Fig. 10(a) and (b), respectively. The LDHAlO2-coated AZ31 in Fig. 10(a) shows several corrosion sites with a diameter of 0.5-1.5 mm where a lot of shallow micro-pits with a depth of a few micrometers are clustered as shown in Fig. 13(a), (c), (e) and (g). Outside of the micro-pits clustering areas, almost no corrosion is observed as shown in Fig. 13(e). The LDHCO3-coated AZ31 in Fig. 10(b) shows many micro-pits on the entire surface and sometimes shows larger pits with a depth of about 10 m as shown in Fig. 13(b), (d), (f), (h) and (i).  Discussion Synthesis of LDHAlO2 by reconstruction process The addition of NaAlO2 in NaCl solution led to the formation of a relatively thick Al2O3 film on AZ31 and the significant improvement of the corrosion resistance of AZ31 as shown by the decrease in weight gain in the immersion tests and the clear passive region on the polarization curves (Figs. 1-4). These facts demonstrated the corrosion inhibition property of NaAlO2. Therefore, aluminate ions were loaded to LDH as a corrosion inhibitor. The as-synthesized LDHAlO2 showed the (003)LDH diffraction peak at almost the same position as the as-received LDHCO3 (Fig. 5). However, the intercalation of aluminate ions into the synthesized LDHAlO2 cannot be denied solely by the XRD measurements because the (003)LDH peak position depends not only on the type of the interlayer anions but also on the orientation of the interlayer anions 42. Other anions that could be present at the interlayer of LDH besides aluminate ions are CO32-  and OH- ions. The presence of CO32- ions was suggested by the Raman results, so that the as-synthesized LDHAlO2 powder could contain LDHCO3. In the case of LDH hydroxide with interlayer OH- ions, the (003)LDH peak was shifted by 250°C-treatment but recovered its original position in a few minutes in air 43. On the other hand, the (003)LDH peak position of the as-synthesized LDHAlO2 irreversibly shifted by the 140°C-treatment, indicating that the interlayer anions were not OH- ions. As mentioned above, the position shift of the (003)LDH diffraction peak of LDHAlO2 to the higher angle after the 140°C-treatment (Fig. 5) indicated that anions other than CO32- and OH- ions were intercalated into the as-synthesized LDHAlO2 and these are circumstantially aluminate ions. It was reported that the interlayer water in the Mg-Al- and Mg-Fe-LDH desorbed over 130°C 44; however, the loss of interlayer water did not lead to the change in XRD pattern of LDH 45. Thus, the position shift of the (003)LDH peak was not attributed to the desorption of interlayer water. The interlayer spacing of the LDHAlO2 decreased by the 140°C-treatment. Then, during the heat treatment at 140°C, it is supposed that the reactions between aluminate ions (AlO2-) and interlayer water described in Eq. (2) and (3) occur at the interlayer.  𝐴𝑙𝑂 2 𝐻 𝑂 → 𝐴𝑙 𝑂𝐻 𝑂𝐻  (2) 𝐴𝑙𝑂 𝐻 𝑂 → 𝐴𝑙𝑂 𝑂𝐻 𝑂𝐻  (3)  According to the EDS analysis of the as-synthesized LDHAlO2 and as-received LDHCO3 powders, the relative atomic concentration ratio of Mg to Al of the LDHAlO2 was 1.7±0.0 which was lower than 2.1±0.0 of the LDHCO3 (Fig. S1 and Table S1 in supplementary material). The Raman spectra of the as-synthesized LDHAlO2 and the LDHAlO2 coating showed the peak being attributed to Al-O bending mode although this peak was also attributed to CO32- ions (Fig. S3 and Fig. 12(a)). These results indicate that aluminate ions were loaded to LDH either at the interlayer and/or outer surface. From the whole, it was presumed that the synthesized LDHAlO2 powder was a mixture of LDH with aluminate ions at the interlayer and outer surface, LDHCO3 and the thermally decomposed LDH.  Formation of LDHAlO2 and Mg-Al-DH composite coating The composite coating of LDHAlO2 or LDHCO3 particles and Mg-Al-DH was formed by the EP-co-D in the similar manner to the previous work 29. The LDHAlO2 coating was coarser than the LDHCO3 coating as shown in Fig. 6(i) and (j). This is presumably because the LDHAlO2 particles agglomerated during the rehydration procedure which disturbed the dense deposition of LDH particles. However, the reasons for the difference in the coating morphology depending on the composition of LDH are a future subject. The deposition of Mg-Al-DH with low crystallized brucite structure was shown by the broad diffraction peaks around 18 and 35 degrees (Fig. 7) corresponding to brucite. The amount of Mg-Al-DH appeared to be smaller in the LDHAlO2 coating than in the LDHCO3 coating, which is indicated by the cross-sectional observation and the smaller coating weight. However, the Mg-Al-DH covered the substrate surface with LDH particles as shown at the crack bottom of the coatings (Fig. 6), and the coatings could show the corrosion protection performance. A part of the Mg-Al-DH with brucite structure should have been transformed to LDHNO3 as indicated by the shoulder peak at lower angle side of the (003)LDH diffraction peak, the Raman peak at 1045 cm-1 38 (Fig. 7 and 12), and the LDHNO3 formation in the electrolyte without LDH particles (Fig. S8). The LDHNO3 could be formed at the bottom of the coating, since the LDHNO3 was present at the bottom of the composite coating formed in the suspension with higher portion of water 29.  Corrosion protection ability of LDHAlO2 and Mg-Al-DH composite coating in aqueous solution In the polarization and EI tests in the aqueous solution (Fig. 8), the LDH coatings exhibited the barrier property as shown by the decrease in the constant current density, the suppression of the surface breakdown in the potential range up to -1 V and the higher Rp, in comparison to the uncoated AZ31. The Nyquist and Bode plots of the LDH-coated AZ31 specimens showed the presence of one time constant (Fig. 8(b) and (c)). The LDH-coated and uncoated specimens showed the comparable CPE-T values, but the Rcoat values of the LDH-coated specimens were higher than the Rfilm of the uncoated specimen. These results indicate that the layer that exhibited capacitive property in the LDH coatings was qualitatively different from that on the uncoated surface. Here, the LDH coatings showed a dense inner layer consisting of mainly Mg-Al-DH, LDHNO3, and LDH particles (Figs. 6 and 7), which presumably exhibited capacitive property. The LDHAlO2 coating with the thinner inner layer showed the higher quasi-passive current density and the lower Rp than the LDHCO3 coating with the thicker inner layer (Figs. 6 and 8). These facts indicate that the barrier property of the LDH coatings depended on the dense Mg-Al-DH inner layer. Also, it is suggested that the composite coatings can be thinned to show corrosion protection. The suppression of the LDH coating breakdown can be attributed to the Cl- absorption of LDHAlO2 and LDHNO3 by anion-exchange in addition to the barrier property of the coatings. The anion exchange between Cl- and NO3- ions of LDHNO3 can be expected because the order of anion selectivity of LDH is determined approximately by the magnitude of the charge density of the anions and the ion-exchange equilibrium constant of Cl- ion is higher than that of NO3- ion 46. Here, the quasi-passive current density of the LDH-coated AZ31specimens was higher than that of the as-polished AZ31 in NaAlO2-added NaCl solution (Fig. 4 and 8(a)), indicating that the barrier property of the LDH coatings was lower than the Al2O3 film formed by AlO2- ion adsorption. The Al2O3 film was thinner but less defective than the LDH coatings. It was found that the coating with less defects can exhibit higher barrier ability than the defective coating in short duration tests like the polarization test.  Corrosion protection ability of LDHAlO2 and Mg-Al-DH composite coating in atmospheric corrosion environment In the wet-dry cyclic corrosion tests, the LDH coatings significantly suppressed the corrosion of AZ31 as can be seen from the remarkable decrease in the weight gain and the much fewer corrosion sites comparing to the uncoated AZ31 (Figs. 9, 10 and 13). Furthermore, the higher corrosion protection ability of the LDHAlO2 coating than the LDHCO3 coating was demonstrated by the smaller weight gain and the smaller number of corrosion sites with the shallower corrosion pits (Figs. 9, 10 and 13). No corrosion was observed outside of the corrosion sites on the LDHAlO2-coated AZ31 (Figs. 10 and 13). The smaller number of corrosion sites of the LDHAlO2-coated AZ31 suggests the anion exchange between Cl- and aluminate ions of LDHAlO2 5, 46 suppressed the corrosion initiation and/or promptly repaired the initiating corrosion pits. The anion exchange between Cl- and NO3- ions of LDHNO3 should also have contributed to the suppression of the corrosion initiation, which was common to the LDHAlO2 and LDHCO3 coatings. Therefore, the corrosion mechanisms shown in Fig 14 are assumed for the LDHAlO2- and LDHCO3-AZ31 specimens under the wet-dry cyclic corrosion tests. The details of the role of the anion exchange property of LDHAlO2 in repairing corrosion sites is a matter for further study. When the specimens were wet, corrosion initiated by Cl- ions at defects in the LDH coatings and corrosion products precipitated. The corrosion initiation of the LDHAlO2-coated AZ31 was suppressed because a part of Cl- ions were absorbed by LDHAlO2 and LDHNO3 by the anion exchange. Then, the released aluminate ions could precipitate with the corrosion products and repaired the initiating corrosion pits. On the LDHCO3-coated AZ31, corrosion initiated more frequently and the initiated corrosion pits could grow. When the specimens were dry, the corrosion products with and without aluminate ions were matured to show a certain corrosion protectiveness.  Conclusions The effect of aluminate ion loading on the corrosion protection property of the LDH coating was investigated. The corrosion inhibition performance of NaAlO2 on AZ31 was investigated as a potential corrosion inhibitor by the immersion and polarization tests and XPS analysis. Aluminate-ion loaded LDH (LDHAlO2) was synthesized from LDHCO3 by the reconstraction process and co-deposited with Mg-Al-DH on AZ31 by the EP-co-D method. The corrosion behavior of the LDHAlO2- and LDHCO3-coated AZ31 specimens was examined by the polarization, EI and wet-dry cyclic corrosion tests. The following findings were derived. 1) NaAlO2 showed corrosion inhibition property for AZ31 by forming the Al2O3 film at the addition of 10 mmol/L or more in the 0.1 mol/L NaCl solution. The addition of 50 mmol/L NaAlO2 decreased the constant current density by a factor of 100 and significantly shifted the breakdown potential from -1.35 V to over -1.0 V.  2) The composite coatings of LDHAlO2 or LDHCO3 particles and Mg-Al-DH showed a dense inner layer consisting of mainly Mg-Al-DH, LDH particles and LDHNO3. The dense inner layer of the LDHAlO2 coating was thinner than that of the LDHCO3 coating. 3) The LDHAlO2 and LDHCO3 coatings showed a barrier property due to the dense inner layer in the polarization and EI tests in the 0.1 mol/L NaCl solution. The LDH-coated AZ31 specimens exhibited the constant current density less than half of the uncoated AZ31 and more than double the Rp. The LDH-coated AZ31 specimens did not show breakdown on the polarization curves in the potential region up to -1.0 V. 4) The LDHAlO2-coated AZ31 showed higher corrosion resistance than the LDHCO3-coated AZ31 in the wet-dry cyclic corrosion tests. The former showed the weight gain less than half of the latter, and the corrosion pits were fewer and shallower. The aluminate ion loading to LDH coating was effective to enhance the corrosion protection ability of the LDH coatings in the atmospheric environment.  Acknowledgments This work was partially supported by the NIMS internship program and by the NIMS Molecule & Material Synthesis Platform in the Nanotechnology Platform Project of National Institute for Materials Science (NIMS). Ms. Nishikawa and Ms. Inokuma are acknowledged for their sincere support during the experiments. The authors are grateful to Dr. Uchikoshi at NIMS for his valuable advice on electrophoretic deposition method.  References 1. J. Hillis, in Magnesium Technology, Metallurgy, Design Data, Applications, H. E. Friedrich and B. L. Mordike, eds., p. 469-498, Springer-Verlag, Berlin, (2006). 2. E. Sato, in Handbook of Advanced Magnesium Technology, The Japan Magnesium Association, ed., p. 71-104, Kallos, Tokyo, (2000). 3. M. Hino, Journal of The Surafce Finishing Society of Japan, 71 (3), 205-211 (2020). 4. S. Namba, J. Surf. Finish. Soc. Jpn., 71 (3), 219-223 (2020). 5. T. Hibino, Journal of The Clay Science Society of Japan, 45 (2), 102-109 (2006). 6. K. Nakamura, M. Tsunakawa, Y. Shimada, A. Serizawa, and T. Ishizaki, Surface and Coatings Technology, 328 436-443 (2017). 7. T. Ishizaki, S. P. Cho, and N. Saito, Crystengcomm, 11 (11), 2338-2343 (2009). 8. N. Kamiyama and T. Ishizaki, Journal of The Japan Institute of Light Metals, 64 (12), 638-642 (2014). 9. F. G. Wu, J. Liang, Z. J. Peng, and B. X. Liu, Applied Surface Science, 313 834-840 (2014). 10. K. A. Yasakau, A. Kuznetsova, S. Kallip, M. Starykevich, J. Tedim, M. G. S. Ferreira, and M. L. Zheludkevich, Corrosion Science, 143 299-313 (2018). 11. T. T. Wen, R. Yan, N. Wang, Y. X. Li, T. Chen, and H. Y. Ma, Surface and Coatings Technology, 383 125255 (2020). 12. T. Hu, Y. J. Ouyang, Z. H. Xie, and L. Wu, Journal of Materials Science & Technology, 92 225-235 (2021). 13. L. Wu, X. X. Ding, Z. Zheng, A. Tang, G. Zhang, A. Andrej Atrens, and F. S. Pan, Journal of Materials Scienc & Technology, 61 66-72 (2021). 14. J. K. E. Tan, N. Birbilis, S. Choudhary, S. Thomas, and P. Balan, Corrosion Science, 205 110444 (2022). 15. B. Pillado, B. Mingo, R. del Olmo, E. Matykina, A. M. Kooijman, Y. Gonzalez-Garcia, R. Arrabal, and M. Mohedano, Journal of Magnesium and Alloys, 11 (1), 201-216 (2023). 16. S. V. Lamaka, B. Vaghefinazari, D. Mei, R. P. Petrauskas, D. Hoche, and M. L. Zheludkevich, Corrosion Science, 128 224-240 (2017). 17. M. A. Osipenko, D. A. Kharytonau, A. A. Kasach, J. Ryl, J. Adamiec, and I. I. Kuriklo, Electrochimca Acta, 414 140175 (2022). 18. D. S. Kharitonov, M. Zimowska, J. Ryl, A. Zielinski, M. A. Osipenko, J. Adamiec, A. Wrzesinska, P. M. Claesson, and I. I. Kurilo, Corrosion Science, 190 109664 (2021). 19. Z. Feng, J. Li, Z. Yang, and R. Buchheit, Materials (Basel), 13 (6), 1325 (2020). 20. L. Chen, X. Ma, Z. Ma, D. Lu, and B. Hou, Royal Society of Chemistry, 46 2105-2127 (2022). 21. B. Vaghefinazari, E. Wierzbicka, P. Visser, R. Posner, R. Arrabal, E. Matykina, M. Mohedano, C. Blawert, M. L. Zheludkevich, and S. V. Lamaka, Materials, 15 (23), 8489 (2022). 22. S. Ono, Journal of the Surface Finishing Society of Japan. 62 198-203 (2011). 23. J. H. Nordlien, K. Nisancioglu, S. Ono, and N. Masuko, Journal of the Electrochem Society., 144 (2), 461-466 (1997). 24. J. H. Nordlien, K. Nisancioglu, S. Ono, and N. Masuko, Journal of the Electrochem Society, 143 (8), 2564-2572 (1996). 25. S. Ono, H. Kijima, and N. Masuko, Materials Transactions, 44 (4), 539-545 (2003). 26. T. Uchikoshi, H. Takahashi, N. Shirahata, and Y. Sakka, Journal of the Ceramic Society of Japan, 115 (1348), 818-820 (2007). 27. T. Uchikoshi, Journal of the Society of Inorganic Materials, Japan, 8 478-483 (2001). 28. S. Heise, M. Hohlinger, Y. T. Hernandez, J. J. P. Palacio, J. A. R. Ortiz, V. Wagener, S. Virtanen, and A. R. Boccaccini, Electrochimica Acta, 232 456-464 (2017). 29. S. Hiromoto and K. Doi, Materials Transactions, 64 (5), 1011-1019 (2023). 30. S. Hiromoto and K. Doi, in "Annual Meeting of Japan Institute of Metals and Materials", Vol. 170, Online, 2022. 31. T. Yamagishi, Y. Oyanagi, and E. Narita, Nippon Kagaku Kaishi, 1993 (4), 329-334 (1993). 32. M. Iida, K. Yoshikawa, M. Togo, and A. Nakahira, Journal of the Japan Society of Powder Powder Metallurgy, 64 (11), 595-600 (2017). 33. T. Sato, Journal of the Mineralogical Society of Japan, 19 (1), 21-41 (1989). 34. S. J. Palmer, L. M. Grand, and R. L. Frost, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 79 (1), 156-160 (2011). 35. R. L. Frost, R. Scholz, A. Lopez, and F. L. Theiss, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 118 187-191 (2014). 36. M. Mora, C. Jimenez-Sanchidrian, and J. R. Ruiz, Materials Letters, 120 193-195 (2014). 37. J. T. Kloprogge, D. Wharton, L. Hickey, and R. L. Frost, American Mineralogist, 87 623-629 (2002). 38. R. Sasai, T. Fujimura, H. Sato, E. Nii, M. Sugata, Y. Nakayashiki, H. Hoashi, C. Moriyoshi, E. Oishi, Y. Fujii, S. Kawaguchi, and H. Tanaka, Bulletin of the Chemical Society of Japan., 95 (5), 802-812 (2022). 39. L. Jinlong, L. Hongyun, and X. Jinpeng, Applied Surface Science, 273 192-198 (2013). 40. N. J. Cherepy, T. H. Shen, A. P. Esposito, and T. M. Tillotson, Journal of Colloid and Interface Science, 282 (1), 80-86 (2005). 41. R. L. Frost, S. Bahfenne, and J. Graham, Journal of Raman Spectroscopy, 40 (8), 855-860 (2009). 42. Z. P. Xu and H. C. Zeng, Journal of Physical Chemistry B, 105 (9), 1743-1749 (2001). 43. V. R. L. Constantino and T. J. Pinnavaia, Inorg. Chem., 34, 883-892 (1995). 44. H. Hirahara, Y. Sawai, S. Aisawa, S. Takahashi, Y. Umetsu, and E. Narita, Journal of the Clay Science Society of Japan, 42 (2), 70-76 (2002). 45. F. Cavani, F. Trifiro, and A. Vaccari, Catalysis Today, 11 (2), 173-301 (1991). 46. S. Miyata, Clays and Clay Minerals, 31 (4), 305-311 (1983).   Table and Figure Captions   Table 1. Corrosion parameters obtained by curve fitting using equivalent electric circuits shown in Fig. 8(c) and (d).   Figure 1. (a)-(d) Optical images of AZ31 immersed in 0.1 mol/L NaCl solution supplemented with (a) 0 mmol/L, (b) 1 mmol/L, (c) 10 mmol/L, and (d) 50 mmol/L NaAlO2. (e) SEM image of AZ31 immersed with 50 mmol/L NaAlO2.  Figure 2. (a) Weight gain of AZ31 after immersion in 0.1 mol/L NaCl solution supplemented with and without NaAlO2 and (b) pH of the solution before and after immersion of AZ31.  Figure 3. (a) XPS survey spectra obtained from surface of AZ31 immersed in 0.1 mol/L NaCl solution with 10 and 50 mmol/L NaAlO2 and as-polished AZ31. XPS narrow spectra of (b) C 1s, (c) O 1s, (d) Al 2p, (e) Mg 2p and (f) Zn 2p3 regions of (i) AZ31 immersed with 50 mmol/L NaAlO2, and (ii) AZ31 immersed with 10 mmol/L NaAlO2 and (iii) as-polished AZ31.  Figure 4. (a) Cathodic and (b) anodic polarization curves of AZ31 in 0.1 mol/L NaCl solution supplemented with (i) 0mmol/L, (ii) 10 mmol/L and (iii) 50 mmol/L NaAlO2.  Figure 5. (a) Wide range XRD patterns and (b) magnified (003)LDH peak of (i) as-synthesized and (ii) 140°C-treated LDHAlO2 particles and (iii) as-received and (iv) 140°C-treated LDHCO3 particles.   Figure 6. (a) and (b) Optical and (c)-(j) SEM images of (a), (c), (e), (g) and (i) LDHAlO2-coated and (b), (d), (f), (h) and (j) LDHCO3-coated AZ31 formed by EP-co-D followed by ultrasonication. (c) and (d) low magnification surface images, high magnification surface images of (e) and (f) coating and (g) and (h) substrate exposed in a crack in coating, and (i) and (j) cross-section images.  Figure 7. (a) Wide range XRD patterns and (b) magnified (003)LDH peak of (i) LDHAlO2-coated and (ii) LDHCO3-coated AZ31 formed by EP-co-D followed by ultrasonication.  Figure 8. (a) Anodic polarization curves and (b) Nyquist and (c) Bode plots of EI spectra of LDHAlO2-coated, LDHCO3-coated and uncoated AZ31 in 0.1 mol/L NaCl solution. Assumed equivalent electric circuits for (d) LDHAlO2- and LDHCO3-coated AZ31 and (e) uncoated AZ, and (f) Rp values obtained by curve fitting.  Figure 9. Weight gain of LDHAlO2-coated, LDHCO3-coated and uncoated AZ31 after wet-dry cyclic corrosion tests.  Figure 10. (a)-(f) Optical (g)-(l) SEM images of LDHAlO2-coated, LDHCO3-coated and uncoated AZ31 after wet-dry corrosion tests. (a), (d), (g), (i) and (k) LDHAlO2-coated, (b), (e), (h), (j) and (l) LDHCO3-coated and (c) and (f) uncoated AZ31. (b), (e) and (f) are magnified images of arrowed area on (a), (b) and (c) images, respectively. (g) and (h) magnified images of a corrosion area. (i) and (j) magnified images of coating in apparently uncorroded area. (k) and (l) cross-section images.  Figure 11. (a) Wide range XRD patterns and (b) magnified (003)LDH peak of (i) LDHAlO2-coated, (ii) LDHCO3-coated and (iii) uncoated AZ31 after wet-dry cyclic corrosion tests.  Figure 12. Raman spectra of (a) LDHAlO2 and (b) LDHCO3 coatings (i) before and (ii) after wet-dry cyclic corrosion tests.  Figure 13. (a) and (b) Optical images, (c) and (d) SEM images of corroded areas indicated by an arrow on the optical images, and (e) and (f) topography images corresponding to (a) and (b) images. (g)-(i) Profiles of line 1)-3) on images (e) and (f). (a), (c), (e) and (g) LDHAlO2-coated AZ31 and (b), (d), (f), (h) and (i) LDHCO3-coated AZ31 after wet-dry cyclic corrosion tests, after removing corrosion product and coating layer.  Figure 14. Assumed corrosion mechanisms of (a) LDHAlO2- and (b) LDHCO3-coated AZ31 in wet-dry corrosion environment.       Table 1. Corrosion parameters obtained by curve fitting using equivalent electric circuits shown in Fig. 8(c) and (d).  Rs  / ꞏcm2 Rcoat  or Rfilm  / kꞏcm2 CPEcoat-T or CPEfilm-T / F sP-1ꞏcm2 CPEcoat-P or CPEfilm-P / - Rct  / kꞏcm2 CPEdl-T  / mF sP-1ꞏcm2 CPEdl-P  / - Rp  / kꞏcm2 LDHAlO2-coated AZ31 53±2.6 8.6±0.5 8.1±0.9 0.94±0.00 - - - 8.6±0.5 LDHCO3-coated AZ31 41±2.1 11±1.4 12±4.1 0.91±0.04 - - - 11.0±1.4 Uncoated AZ31 47±3.8 2.1±0.4 9.6±1.0 0.94±0.00 0.98±0.17 1.1±0.2 0.80±0.01 3.0±0.6        Figure 1. (a)-(d) Optical images of AZ31 immersed in 0.1 mol/L NaCl solution supplemented with (a) 0 mmol/L, (b) 1 mmol/L, (c) 10 mmol/L, and (d) 50 mmol/L NaAlO2. (e) SEM image of AZ31 immersed with 50 mmol/L NaAlO2.       Figure 2. (a) Weight gain of AZ31 after immersion in 0.1 mol/L NaCl solution supplemented with and without NaAlO2 and (b) pH of the solution before and after immersion of AZ31.      Figure 3. (a) XPS survey spectra obtained from surface of AZ31 immersed in 0.1 mol/L NaCl solution with 10 and 50 mmol/L NaAlO2 and as-polished AZ31. XPS narrow spectra of (b) C 1s, (c) O 1s, (d) Al 2p, (e) Mg 2p and (f) Zn 2p3 regions of (i) AZ31 immersed with 50 mmol/L NaAlO2, and (ii) AZ31 immersed with 10 mmol/L NaAlO2 and (iii) as-polished AZ31.      Figure 4. (a) Cathodic and (b) anodic polarization curves of AZ31 in 0.1 mol/L NaCl solution supplemented with (i) 0 mmol/L, (ii) 10 mmol/L and (iii) 50 mmol/L NaAlO2.       Figure 5. (a) Wide range XRD patterns and (b) magnified (003)LDH peak of (i) as-synthesized and (ii) 140°C-treated LDHAlO2 particles and (iii) as-received and (iv) 140°C-treated LDHCO3 particles.      Figure 6. (a) and (b) Optical and (c)-(j) SEM images of (a), (c), (e), (g) and (i) LDHAlO2-coated and (b), (d), (f), (h) and (j) LDHCO3-coated AZ31 formed by EP-co-D followed by ultrasonication. (c) and (d) low magnification surface images, high magnification surface images of (e) and (f) coating and (g) and (h) substrate exposed in a crack in coating, and (i) and (j) cross-section images.       Figure 7. (a) Wide range XRD patterns and (b) magnified (003)LDH peak of (i) LDHAlO2-coated and (ii) LDHCO3-coated AZ31 formed by EP-co-D followed by ultrasonication.      Figure 8. (a) Anodic polarization curves and (b) Nyquist and (c) Bode plots of EI spectra of LDHAlO2-coated, LDHCO3-coated and uncoated AZ31 in 0.1 mol/L NaCl solution. Assumed equivalent electric circuits for (d) LDHAlO2- and LDHCO3-coated AZ31 and (e) uncoated AZ, and (f) Rp values obtained by curve fitting.       Figure 9. Weight gain of LDHAlO2-coated, LDHCO3-coated and uncoated AZ31 after wet-dry cyclic corrosion tests.     Figure 10. (a)-(f) Optical (g)-(l) SEM images of LDHAlO2-coated, LDHCO3-coated and uncoated AZ31 after wet-dry corrosion tests. (a), (d), (g), (i) and (k) LDHAlO2-coated, (b), (e), (h), (j) and (l) LDHCO3-coated and (c) and (f) uncoated AZ31. (b), (e) and (f) are magnified images of arrowed area on (a), (b) and (c) images, respectively. (g) and (h) magnified images of a corrosion area. (i) and (j) magnified images of coating in apparently uncorroded area. (k) and (l) cross-section images.       Figure 11. (a) Wide range XRD patterns and (b) magnified (003)LDH peak of (i) LDHAlO2-coated, (ii) LDHCO3-coated and (iii) uncoated AZ31 after wet-dry cyclic corrosion tests.       Figure 12. Raman spectra of (a) LDHAlO2 and (b) LDHCO3 coatings (i) before and (ii) after wet-dry cyclic corrosion tests.      Figure 13. (a) and (b) Optical images, (c) and (d) SEM images of corroded areas indicated by an arrow on the optical images, and (e) and (f) topography images corresponding to (a) and (b) images. (g)-(i) Profiles of line 1)-3) on images (e) and (f). (a), (c), (e) and (g) LDHAlO2-coated AZ31 and (b), (d), (f), (h) and (i) LDHCO3-coated AZ31 after wet-dry cyclic corrosion tests, after removing corrosion product and coating layer.       Figure 14. Assumed corrosion mechanisms of (a) LDHAlO2- and (b) LDHCO3-coated AZ31 in wet-dry corrosion environment.