Applying Baghdadite / PCL / Chitosan Nanocomposite Coating on AZ91 Magnesium Alloy to Improve Corrosion Behavior, Bioactivity, and Biodegradability

: Magnesiumalloyshavereceivedagreatamountofattentionregardingbeingusedinbiomedical applications; however, they show high degradability, poor bioactivity, and biocompatibility. To improve these properties, surface modification and various types of coatings have been applied. In this study, an anodized AZ91 alloy was coated with a polymer matrix composite made of polycaprolactone / chitosan (PCL / Ch) with different percentages of baghdadite to improve its resistance to corrosion, bioactivity, and biocompatibility. The effects of different percentages of baghdadite (0 wt %, 1 wt %, 3 wt %, and 5 wt %) on the surface microstructure, corrosion resistance, roughness, and wettability were evaluated. The results indicated that the applied nano-polymer-ceramic coating including 3 wt % baghdadite was hydrophobic, which consequently increased the corrosion resistance and decreased the corrosion current density of the anodized AZ91 alloy. Coating with 3 wt % baghdadite increased the roughness of AZ91 from 0.329 ± 0.02 to 7.026 ± 0.31 µ m. After applying the polymer-ceramic coating on the surface of anodized AZ91, the corrosion products changed into calcium–phosphate compounds instead of Mg(OH) 2 , which is more stable in a physiological environment.


Introduction
Some metals have been used as implants for biomedical applications due to their suitable properties [1,2]. Most of the metals used in vivo are bioinert and have a high stability as well as low degradation rate [3,4]. These, in turn, lead to the emergence of some problems such as the stress shielding phenomenon, failure to stimulate the surrounding tissue to grow, release of metal ions into the environment, and inflammation in the body [4,5]. In the meantime, since metal implants are heavy [6][7][8], medical science has progressed towards using lightweight materials with a controllable degradation rate to allow tissue ingrowth [7]. Magnesium alloys have recently attracted the attention of many scientists due to their high strength-to-weight ratio and degradability [8,9]. Magnesium is one of the most important and essential elements in living organisms, which indirectly affects the mineral metabolism [10,11]. In addition, stress shielding phenomenon does not occur for magnesium alloys due to having density and Young's modulus close to that of natural bone. The density of magnesium alloys is between 1.7 and 2.0 g/cm 3 , which is very close to that of bone (1.8-2.1 g/cm 3 ) [12]. Also, the Young's moduli of magnesium and natural bone are 45 GPa and 10-30 GPa, respectively [13].

Micro-Arc oxidation (MAO) Coating Process
In order to prepare the electrolyte of MAO process, 100 g of NaOH and 100 g of Na 2 SiO 3 were mixed in 1 L of distilled water [36]. Samples of an AZ91 magnesium alloy and a stainless-steel plate were placed in the solution as the anode and cathode, respectively. MAO was carried out at a voltage of 60 V for 30 minutes. Next, the samples were washed with distilled water, and then ultrasonically cleaned in an acetone solution. After that, they were again rinsed with distilled water and dried under the blowing of warm air.

Preparation of Baghdadite Nanopowder
The sol-gel method was used to produce baghdadite nanopowder [30]. In this method, zirconium oxynitrate, calcium nitrate, and tetraethyl orthosilicate (TEOS) (all from Sigma-Aldrich, St. Louis, MI, USA) were used as raw materials with a stoichiometric ratio of 1:3:2. First, TEOS, ethanol, and nitric acid, with a molar ratio of 1:8:0.16 were mixed together using a magnetic stirrer for 30 minutes. Then, the raw materials were added to the solution and the obtained slurry was stirred at room temperature for five hours. After that, the clear sol was dried at 60 • C temperature for one day and then at 100 • C for two days. The dried gel was then calcined at 1150 • C for three hours followed by 2 h ball milling.

Ceramic/Polymer Coating Preparation
In order to create a polymer coating on the anodized samples of AZ91, 90 wt % of polycaprolactone (PCL) (with a molecular weight of 60,000, Sigma-Aldrich) and 10 wt % of Chitosan (Ch) (with a molecular weight of 50,000, Sigma-Aldrich) were dissolved in a mixture of acetic acid and formic acid with a weight ratio of 3 to 1 [35]. Next, different percentages of baghdadite nanopowder (0 wt %, 1 wt %, 3 wt %, and 5 wt %) were added to the polymer solutions and the solutions were stirred for 3 h. Table 2 shows the designation of different samples. The coating operation was performed through the immersion method at the entry and exit speeds of 6 mm/min, and five times for each sample. The coated samples were dried in an air atmosphere for 48 h. In order to investigate the corrosion behavior of different samples, electrochemical tests including a polarization corrosion test and electrochemical impedance spectroscopy (EIS) were carried out in a simulated body fluid (SBF) using a potentiostat (Model: PARATAT 2273, AMETEK, Oak Ridge, TN, USA). To this end, an electrochemical cell containing three electrodes with the following configuration was established: The samples were placed at the position of the working electrode, the platinum electrode at the position of the counter electrode, and the saturated calomel electrode (SCE) at the position of the reference electrode. All corrosion tests were performed after reaching a steady state within 60 min. A potentiodynamic polarization test was performed at a scanning speed of 1 mV/s, and the frequency range for the EIS test was chosen between 100 kHz and 10 mHz. A wettability test was carried out using an angle finder in order to investigate the effect of baghdadite nanoparticles on the hydrophilic or hydrophobic behavior of the applied polymer coating on the AZ91 samples. In each of the tests, a droplet of 2 µL was laid on the surface of the samples using a DragonLab StepMate stepper (Model: JV62782, DragonLab, Beijing, China). The average wettability angle was reported using ImageJ software version 1.52q. In order to examine the surface roughness before and after coating, a Mitutoyo roughness meter (Model: SJ-210, Grainger, Chicago, IL, USA) was used, and the mean surface roughness (R a ) was reported. To ensure the accuracy of the results, all the tests were repeated three times. Phase analysis of different samples was performed using X-ray diffraction (XRD) technique by Philips (Modle:PW1730, Philips, Amsterdam, The Netherlands) with Cu Kα radiation with λ = 0.154 nm, 20 kV, and 30 mA, in the 2θ range of 20 • -70 • (step size 0.05 • and time per step 1 s). Scanning electron microscopy (SEM) (Philips XL30, acceleration voltage of 30 kV, Amsterdam, The Netherlands) and energy-dispersive spectroscope (EDS) (Philips XL30, Amsterdam, The Netherlands) were used to investigate the surface morphology and elemental analyses of the surface of modified and unmodified samples. The powder morphology and crystallite size were investigated by a transmission electron microscope (TEM) (TEM, Philips, EM208S, Amsterdam, The Netherlands) with the applied voltage of 100 kV. ImageJ software was used to measure the baghdadite crystallite size.

Results and Discussions
TEM micrograph of baghdadite powder fabricated by sol-gel method is shown in Figure 1. The baghdadite powder had crystals mostly with a spherical shape, which were highly agglomerated. The crystal size of baghdadite nanopowder was in the range of 20-70 nm.

Results and Discussions
TEM micrograph of baghdadite powder fabricated by sol-gel method is shown in Figure 1. The baghdadite powder had crystals mostly with a spherical shape, which were highly agglomerated. The crystal size of baghdadite nanopowder was in the range of 20-70 nm.       [37]. Besides increasing the corrosion resistance of AZ91 substrate, the presence of uneven porous layer on the surface improves the adhesion of the polymer coating to the substrate [38]. As can be seen in Figure 3c, applying the polymer coating formed a smooth and uniform layer on the surface. The addition of 1 wt % baghdadite nanoparticles roughed the surface of the specimens (Figure 3d,e). With increasing the percentage of baghdadite nanoparticles up to 5 wt %, the surface roughness increased. The formation of baghdadite agglomerates can be seen on the surface of M4 sample (shown by arrows in Figure 3h,i). Nonetheless, no crack or hole was observed due to the addition of baghdadite nanoparticles.   The obtained results are in a good agreement with the studies conducted by other researchers on the anodized AZ91 alloy [37]. Besides increasing the corrosion resistance of AZ91 substrate, the presence of uneven porous layer on the surface improves the adhesion of the polymer coating to the substrate [38]. As can be seen in Figure 3c, applying the polymer coating formed a smooth and uniform layer on the surface. The addition of 1 wt % baghdadite nanoparticles roughed the surface of the specimens (Figure 3d,e). With increasing the percentage of baghdadite nanoparticles up to 5 wt %, the surface roughness increased. The formation of baghdadite agglomerates can be seen on the surface of M4 sample (shown by arrows in Figure 3h,i). Nonetheless, no crack or hole was observed due to the addition of baghdadite nanoparticles.   [37]. Besides increasing the corrosion resistance of AZ91 substrate, the presence of uneven porous layer on the surface improves the adhesion of the polymer coating to the substrate [38]. As can be seen in Figure 3c, applying the polymer coating formed a smooth and uniform layer on the surface. The addition of 1 wt % baghdadite nanoparticles roughed the surface of the specimens (Figure 3d,e). With increasing the percentage of baghdadite nanoparticles up to 5 wt %, the surface roughness increased. The formation of baghdadite agglomerates can be seen on the surface of M4 sample (shown by arrows in Figure 3h,i). Nonetheless, no crack or hole was observed due to the addition of baghdadite nanoparticles.    Figure 4 shows a schematic illustration of forsterite and MgO formation in an MAO process. When magnesium is dissolved from the outer layer of the AZ91 substrate (due to its contact with the electrolyte solution and its reaction with the O 2− ion), an MgO layer is formed according to the following reaction [37]: According to the literature, the presence of SiO 3 2− in the electrolyte solution is due to the dissolution of Na 2 SiO 3 in water, which occurs according to Reaction 2. It is converted into Si(OH) 4 as a result of being hydrolyzed by water according to Reaction 3, and eventually converted into SiO 2 as a result of applying a strong electric field and anodizing procedure at a high temperature (Reaction 4) [37,39]: Coatings 2019, 9, x FOR PEER REVIEW 6 of 14 Figure 4 shows a schematic illustration of forsterite and MgO formation in an MAO process. When magnesium is dissolved from the outer layer of the AZ91 substrate (due to its contact with the electrolyte solution and its reaction with the O 2− ion), an MgO layer is formed according to the following reaction [37]: According to the literature, the presence of SiO3 2− in the electrolyte solution is due to the dissolution of Na2SiO3 in water, which occurs according to Reaction 2. It is converted into Si(OH)4 as a result of being hydrolyzed by water according to Reaction 3, and eventually converted into SiO2 as a result of applying a strong electric field and anodizing procedure at a high temperature (Reaction 4) [37,39]: During the process of plasma anodizing, MgO and SiO2 melt locally and are cooled in the aqueous medium of the electrolyte, which result in the formation of forsterite according to the following reaction [17,37].
The results obtained from the roughness test (Table 3) suggested that the anodizing operation on the substrate increased the surface roughness from 0.329 ± 0.02 μm to 11.356 ± 0.45 μm. The formation of polymer coating reduced the surface roughness down to 4.743 ± 0.23 μm, whereas the addition of baghdadite nanoparticles (1 wt % to 5 wt %) to the coating increased the surface roughness from 4.743 ± 0.23 μm (M3) to 7.792 ± 0.34 μm (M4), 7.026 ± 0.31 μm (M5), and 10.610 ± 0.21 μm (M6). One of the important benefits of applying a coating on the metal implants, is to facilitate the ingrowth and penetration of surrounding tissue into the coating and subsequently provide appropriate bonds between the implant and the surrounding tissue. This will be possible if an appropriate surface roughness is developed on the surface so that the tissue can adhere to the coating [38]. As can be seen, applying the polymer coating reduced the surface roughness, whereas addition of baghdadite nanoparticles increased the surface roughness, and created an uneven and porous surface, which will be appropriate for biostability, initial adhesion, and the growth of tissue. During the process of plasma anodizing, MgO and SiO 2 melt locally and are cooled in the aqueous medium of the electrolyte, which result in the formation of forsterite according to the following reaction [17,37].
The results obtained from the roughness test (Table 3) suggested that the anodizing operation on the substrate increased the surface roughness from 0.329 ± 0.02 µm to 11.356 ± 0.45 µm. The formation of polymer coating reduced the surface roughness down to 4.743 ± 0.23 µm, whereas the addition of baghdadite nanoparticles (1 wt % to 5 wt %) to the coating increased the surface roughness from 4.743 ± 0.23 µm (M3) to 7.792 ± 0.34 µm (M4), 7.026 ± 0.31 µm (M5), and 10.610 ± 0.21 µm (M6). One of the important benefits of applying a coating on the metal implants, is to facilitate the ingrowth and penetration of surrounding tissue into the coating and subsequently provide appropriate bonds between the implant and the surrounding tissue. This will be possible if an appropriate surface roughness is developed on the surface so that the tissue can adhere to the coating [38]. As can be seen, applying the polymer coating reduced the surface roughness, whereas addition of baghdadite nanoparticles increased the surface roughness, and created an uneven and porous surface, which will be appropriate for biostability, initial adhesion, and the growth of tissue. A water contact angle test was used in order to examine the hydrophobicity or hydrophilicity of the applied coatings on the anodized AZ91 alloy. The results showed that applying the polymer coating to the anodized AZ91 alloy reduced the wettability angle from 86 • ± 1 • to 33 • ± 1.3 • , whereas adding baghdadite nanoparticles increased the wettability angle to 66 • ± 1.3 • in the presence of 5 wt % baghdadite ( Figure 5 and Table 3). Considering the fact that wetting angles between 0 • to 90 • show a hydrophilic surface [25], it can be concluded that all the specimens have a hydrophilic surface. Although the addition of baghdadite nanoparticles increased the wetting angle, the coating maintained its hydrophilic state in all samples.  A water contact angle test was used in order to examine the hydrophobicity or hydrophilicity of the applied coatings on the anodized AZ91 alloy. The results showed that applying the polymer coating to the anodized AZ91 alloy reduced the wettability angle from 86° ± 1° to 33° ± 1.3°, whereas adding baghdadite nanoparticles increased the wettability angle to 66° ± 1.3 ° in the presence of 5 wt % baghdadite ( Figure 5 and Table 3). Considering the fact that wetting angles between 0 to 90 show a hydrophilic surface [25], it can be concluded that all the specimens have a hydrophilic surface. Although the addition of baghdadite nanoparticles increased the wetting angle, the coating maintained its hydrophilic state in all samples. A potentio-dynamic polarization test and electrochemical impedance spectroscopy (EIS) were carried out in a simulated body fluid (SBF, pH = 7.4) in order to examine the corrosion rate and corrosion resistance of the specimens. The potentio-dynamic polarization curves in Figure 6 were used to evaluate the corrosion rate of the specimens before and after first and second coating. In general, lower corrosion current density shows higher corrosion potential and greater corrosion resistance. From Figure 6 and the results in Table 4 (Icorr (corrosion current density), Ecorr (corrosion voltage)), it can be seen that the corrosion current density for the uncoated AZ91 sample was 2 × 10 −4 (M), which decreased to 0.5 × 10 −6 due to the performing of the anodizing process (M1). Applying the A potentio-dynamic polarization test and electrochemical impedance spectroscopy (EIS) were carried out in a simulated body fluid (SBF, pH = 7.4) in order to examine the corrosion rate and corrosion resistance of the specimens. The potentio-dynamic polarization curves in Figure 6 were used to evaluate the corrosion rate of the specimens before and after first and second coating. In general, lower corrosion current density shows higher corrosion potential and greater corrosion resistance.
From Figure 6 and the results in Table 4 (I corr (corrosion current density), E corr (corrosion voltage)), it can be seen that the corrosion current density for the uncoated AZ91 sample was 2 × 10 −4 (M), which decreased to 0.5 × 10 −6 due to the performing of the anodizing process (M1). Applying the polymer coating and adding baghdadite nanoparticles to the coating up to 3 wt % caused the decreasing trend to be continued, but with increasing the percentage of baghdadite to 5 wt %, I corr increased, which can be attributed to the increased surface roughness (10.61 ± 0.23 µm) and hydrophilicity (66 • ± 1.3 • ) of the coating. In addition, the corrosion potential for the AZ91 is equal to −1.71 V, whereas by performing MAO process and applying the baghdadite/polymer composite coating up to 3 wt % baghdadite, an upward trend was observed in the corrosion potential. On the other hand, adding more baghdadite to the coating (up to 5 wt %) decreased the corrosion potential. According to the results of the potentio-dynamic polarization test, it can be concluded that the addition of baghdadite nanoparticles to the coating decreased the formation of cracks in the coating. As a result, an improvement was detected in the corrosion resistance of these specimens compared to that of the uncoated AZ91 sample. Furthermore, M5 specimen exhibited the best corrosion resistance among the coated specimens. The results of some studies on a pure magnesium alloy with a PCL/PLA polymer coating demonstrated a 10-fold decrease in the corrosion rate of the magnesium alloy in an SBF solution, whereas in the present study, we observed a 104-fold decrease by applying a ceramic/polymer composite coating [40]. polymer coating and adding baghdadite nanoparticles to the coating up to 3 wt % caused the decreasing trend to be continued, but with increasing the percentage of baghdadite to 5 wt %, Icorr increased, which can be attributed to the increased surface roughness (10.61 ± 0.23 μm) and hydrophilicity (66° ± 1.3 °) of the coating. In addition, the corrosion potential for the AZ91 is equal to −1.71 V, whereas by performing MAO process and applying the baghdadite/polymer composite coating up to 3 wt % baghdadite, an upward trend was observed in the corrosion potential. On the other hand, adding more baghdadite to the coating (up to 5 wt %) decreased the corrosion potential.
According to the results of the potentio-dynamic polarization test, it can be concluded that the addition of baghdadite nanoparticles to the coating decreased the formation of cracks in the coating. As a result, an improvement was detected in the corrosion resistance of these specimens compared to that of the uncoated AZ91 sample. Furthermore, M5 specimen exhibited the best corrosion resistance among the coated specimens. The results of some studies on a pure magnesium alloy with a PCL/PLA polymer coating demonstrated a 10-fold decrease in the corrosion rate of the magnesium alloy in an SBF solution, whereas in the present study, we observed a 104-fold decrease by applying a ceramic/polymer composite coating [40].  According to the EIS curves shown in Figure 7a (Nyquist) and Figure 7b (Bode), the application of the MAO, MAO/polymer, and MAO/nano-baghdadite/polymer coatings on the AZ91 substrate made dramatic changes in the corrosion resistance of this alloy. The Nyquist curve of the uncoated AZ91 sample showed a capacity quasi-loop (from high to moderate frequencies) and an inductive quasi-loop (at a low frequency), which can be attributed to the load transfer and the anion absorption/desorption process on the surface of the uncoated magnesium sample, respectively. The reason for observing the capacity loop from high to moderate frequencies in the AZ91 sample was the presence of an oxide layer, which could somewhat protect the metal from corrosion, although this oxide layer is not enough. Based on the EIS results for the coated specimens (Figure 7a), the MAO/polymer-coated specimen exhibited a better corrosion resistance behavior compared to the MAO-coated specimen; (i.e., it has a larger capacity loop diameter than the MAO-coated specimen). According to the EIS curves shown in Figure 7a (Nyquist) and Figure 7b (Bode), the application of the MAO, MAO/polymer, and MAO/nano-baghdadite/polymer coatings on the AZ91 substrate made dramatic changes in the corrosion resistance of this alloy. The Nyquist curve of the uncoated AZ91 sample showed a capacity quasi-loop (from high to moderate frequencies) and an inductive quasi-loop (at a low frequency), which can be attributed to the load transfer and the anion absorption/desorption process on the surface of the uncoated magnesium sample, respectively. The reason for observing the capacity loop from high to moderate frequencies in the AZ91 sample was the presence of an oxide layer, Coatings 2019, 9, 789 9 of 14 which could somewhat protect the metal from corrosion, although this oxide layer is not enough. Based on the EIS results for the coated specimens (Figure 7a), the MAO/polymer-coated specimen exhibited a better corrosion resistance behavior compared to the MAO-coated specimen; (i.e., it has a larger capacity loop diameter than the MAO-coated specimen). Addition of one percent baghdadite nanoparticles improved the corrosion behavior and caused a bigger capacity loop. Moreover, the single-loop Nyquist curve of the specimen containing 3 wt % baghdadite nanoparticles indicated that the electrolyte failed to pass through the MAO/polymer coating to reach the substrate. As a result, a better corrosion behavior was observed. Adding 5 wt % baghdadite to the polymer coating decreased the corrosion resistance (a decrease in the radius of the capacity loop) due to an increase in the roughness and defects in the coating. Figure 7b shows the Bode curves of the coated and uncoated specimens. Also, Table 5 shows the magnitude of impedance for different specimens at low frequencies. As can be seen, the magnitude of impedance for the uncoated specimen, the MAO-coated specimen, and the specimen with a polymer coating containing 3 wt % baghdadite increased from 2100 to 4500 and 76000 Ω·cm 2 in the low frequency zone, respectively. According to the literature [17,19], increasing the impedance in the lower frequency zone increases the corrosion resistance of the material. In order to further investigate the details of the EIS curves, equivalent electrical circuits simulated by Zview software version 3.4 were used for the uncoated specimen, the MAO-coated specimen, and the specimen with a ceramic/polymer composite coating (Figure 8 and Table 6). In these models, Rs, Ra, and Rcoat are the resistance of solution, the charge-transfer resistance, and the coating resistance, respectively. CPEdl and CPEcoat are constant phase elements corresponding to the double-layer capacitive behavior and the capacitive capacitance of the coating, respectively. As  In order to further investigate the details of the EIS curves, equivalent electrical circuits simulated by Zview software version 3.4 were used for the uncoated specimen, the MAO-coated specimen, and the specimen with a ceramic/polymer composite coating ( Figure 8 and Table 6). In these models, R s , R a , and R coat are the resistance of solution, the charge-transfer resistance, and the coating resistance, respectively. CPE dl and CPE coat are constant phase elements corresponding to the double-layer capacitive behavior and the capacitive capacitance of the coating, respectively. As shown in Table 6, the total resistance of the specimens (R t = R a + R coat ) increased as a result of applying the MAO/nano-baghdadite/polymer coating compared to those of the uncoated specimens and the specimens only coated with MAO. Showing a better capacitive behavior due to the application of this coating compared to that of the uncoated specimen was also proof of an improvement in the corrosion behavior. Among all the specimens, M5 exhibited the highest corrosion resistance, the lowest capacitive capacitance, and the lowest corrosion rate.    Figure 9a-l shows the SEM images of modified and unmodified specimens after corrosion in two different magnifications. As seen, the uncoated specimen (M) was affected by pitting corrosion on its entire surface. Higher magnifications revealed that corrosion products with spherical morphology exist inside the cavities. Based on the previous studies, corrosion of biodegradable magnesium alloys occurs in SBF solution containing Cl − ions according to the following reactions [33,38]: Subsequently, after immersing the AZ91 alloy in the SBF solution, Cl − ions penetrated into the substrate, pitting corrosion occurred, and the corrosion product Mg(OH) 2 was deposited on the surface. However, with applying the polymer coating on the surface, the number of cracks and cavities on the surface decreased, and as a result, the corrosion rate reduced [35]. In fact, coating acted as a protective layer between the substrate and the electrolyte. As can be seen in the SEM images of M1 and M3, pitting corrosion occurred; however, it was not observed in specimens with ceramic/polymer coating. Only some cracks can be seen in some parts of the coating after 1 h of corrosion testing. In these specimens, the calcium, silicon, and zirconium ions in the surface of coating were combined with the phosphate group in the electrolyte to produce calcium-phosphate compounds, which was deposited on the surface and improved the corrosion resistance [33]. Furthermore, the formation of these compounds on the surface improved the bioactivity of samples. M5 exhibited the highest number of calcium-phosphate compounds on its surface, and as a result showed a higher corrosion resistance. The EDS results M5 (Figure 9n) showed that the white particles on the surface of this sample contained calcium and phosphor elements, which showed the apatite formation and bioactivity of these specimens. The obtained results were in a good agreement with those of polarization. The AZ91 alloy, coated with the MAO/nano-baghdadite/polymer composite, can be an appropriate candidate for medical implants; however, further bio-clinical studies are required. M5 exhibited the highest number of calcium-phosphate compounds on its surface, and as a result showed a higher corrosion resistance. The EDS results M5 (Figure 9n) showed that the white particles on the surface of this sample contained calcium and phosphor elements, which showed the apatite formation and bioactivity of these specimens. The obtained results were in a good agreement with those of polarization. The AZ91 alloy, coated with the MAO/nano-baghdadite/polymer composite, can be an appropriate candidate for medical implants; however, further bio-clinical studies are required.

Conclusions
In this study, a nanocomposite coating made of PCL/Ch with different percentages of baghdadite was successfully applied to an anodized AZ91 alloy through an economical immersion method. The microstructure evaluation of the applied ceramic/polymer coatings showed the formation of a dense coating without any cracks and porosities on the AZ91 substrate, which improved corrosion resistance and decreased corrosion rate of this alloy. The results showed that the coating containing 3 wt % baghdadite exhibited the best corrosion resistance, the most appropriate roughness, and the lowest corrosion rate. Furthermore, adding baghdadite nanoparticles prevented pitting corrosion in the coating, and instead of the formation of unstable magnesium hydroxide in the medium containing Cl − , for these specimens, the corrosion products changed into calcium-

Conclusions
In this study, a nanocomposite coating made of PCL/Ch with different percentages of baghdadite was successfully applied to an anodized AZ91 alloy through an economical immersion method. The microstructure evaluation of the applied ceramic/polymer coatings showed the formation of a dense coating without any cracks and porosities on the AZ91 substrate, which improved corrosion resistance and decreased corrosion rate of this alloy. The results showed that the coating containing 3 wt % baghdadite exhibited the best corrosion resistance, the most appropriate roughness, and the lowest corrosion rate. Furthermore, adding baghdadite nanoparticles prevented pitting corrosion in the coating, and instead of the formation of unstable magnesium hydroxide in the medium containing Cl − , for these specimens, the corrosion products changed into calcium-phosphate compounds, which were more stable and improved the corrosion resistance of the coating.