Hydrothermal In-Situ Synthesis and Anti-Corrosion Performance of Zinc Oxide Hydroxyapatite Nanocomposite Anti-Corrosive Pigment

: With the rapid development of the Chinese marine economy, the anti-corrosion of ships and marine engineering facilities has become urgent to be solved. In this paper, a stable zinc-hydroxyapatite (ZnO-HAP) nanocomposite anti-corrosive pigment was prepared by using the hydrothermal in situ synthesis technique, which supported nano-ZnO onto the hydroxyapatite (HAP) surface. The phase composition, microstructure, and performance of the ZnO-HAP nanocomposite were investigated by X-ray diffraction (XRD), Fourier transform infrared (FT-IR), electron probe microanalysis (EPMA), and electrochemical impedance spectroscopy (EIS), and the anti-corrosion mechanism of the ZnO-HAP nanocomposite was discussed. The results show that the corrosion resistance of the ZnO-HAP/epoxy is better than that of the pure epoxy resin coating. The opti-mum condition to fabricate ZnO-HAP/epoxy with suitable corrosion resistance was found to be a ZnO/HAP ratio of 0.65/0.35. The synergistic complementation mechanism of ZnO and HAP enriches the metallic anti-corrosion theory and provides a new idea for the synthesis of novel and promising anti-corrosive pigments. To explore the inﬂuence of the addition of ZnO-HAP nanocomposites on the epoxy resin coating, ZnO-HAP nanocomposites with a ratio of 65 wt.% ZnO and 35 wt.% HAP (a), epoxy resin coating (b), and ZnO-HAP/epoxy resin coating (c) were tested, respectively. The test results are shown in Figure 3. Figure 3a shows the characteristic absorption peaks of the Zn–O bond at 467 and 415 cm − 1 , the peaks at 1089 and 1031 cm − 1 are due to the asymmetric stretching of the P-O bond of phosphate groups, the sharp peaks at 604 and 563 cm − 1 are attributed to the bending modes of the O–P–O bonds in phosphate groups, and the stretching vibration of –OH caused the peaks at 3434 and 604 cm − 1 . The FT-IR spectra of ZnO-HAP nanocomposites further show that the anti-corrosive pigment is composed of ZnO and HAP, consistent with the XRD results. In Figure 3b, the stretching vibration absorption peaks of the N–H group and –OH were at 3380 cm − 1 [17]. In Figure 3c, the polymer peaks of ZnO-HAP/epoxy resin show that the characteristic peaks of ZnO and HAP increased, compared with the peaks of pure epoxy resin. In contrast, the position of the peaks remained unchanged, indicating that ZnO-HAP nanocomposites can still exist stably after being added into epoxy resin.


Introduction
Metal corrosion is widely found in daily life and industrial production, causing severe economic loss [1]. It is imperative to explore new anti-corrosive materials to improve the service life of metals. At present, the most widely used method is to coat the organic coating on the surface of metals [2,3]. Epoxy resin is widely used among many organic coatings because of its excellent metal adhesion ability. Nevertheless, epoxy resin is susceptible to aging under long-term ultraviolet radiation, which affects its anti-corrosion performance. In addition, electrolyte diffusion through the pores and defects of the epoxy resin can also lead to corrosion; thus, the barrier performance of the coating is seriously decreased. Therefore, how to improve the anti-corrosion performance of epoxy resin coatings is an urgent problem.
In recent years, researchers have found that nanomaterials' unique physical, chemical, and mechanical performance play an essential role in enhancing the anti-corrosive performance of materials [4]. Zinc oxide (ZnO) is a nanometer particle commonly used in coatings. It is considered to be a multifunctional material because of its unique physical and chemical performance, including non-toxicity, chemical stability, and low cost [5,6]. Compared with other nanoparticles such as TiO 2 and Al 2 O 3 , ZnO presents a stable surface, easy dispersion, anti-bacterial, aging resistance, and UV shielding [7,8]. ZnO can also fill the pores in the polymer matrix, so it has been widely used in anti-corrosive coatings. However, ZnO has poor compatibility with epoxy resin [9], resulting in easy agglomeration of powder in epoxy resin. Adding ZnO into epoxy resin will reduce the mechanical performance of the coating and improve the anti-corrosion effect. Therefore, ZnO must be modified to make up for this defect. Hydroxyapatite ((Ca 10 (PO 4 ) 6 (OH) 2 ), HAP) is an environmentally friendly, non-toxic functional material [10,11], with stable chemical performance [12], good metal adhesion [13], significant adsorption capacity, and low water solubility [14]. The compatibility of the anti-corrosive pigment and epoxy resin can be improved by combining HAP and ZnO. In addition, Xue et al. [15] proved that the rich hydroxyl in hydroxyapatite could replace free Cl − in the coating to form more stable chlorine apatite and reduce metal damage.
The composite of ZnO and HAP improves the compatibility of the anti-corrosive pigment and metal, reducing the coating porosity and solidifying Cl − . In this paper, nano-ZnO powders were synthesized on the surface of HAP sheets by the hydrothermal in situ synthesis technique. In other words, after adding zinc acetate dihydrate, Zn 2+ first interacted with the active groups and the surface of HAP to form zinc hydroxide in situ, which was thermally decomposed into ZnO under alkaline conditions, then formed the ZnO-HAP nanocomposite. The anti-corrosive pigment was added to the epoxy resin to prepare the ZnO-HAP/epoxy composite coating. The anti-corrosive pigments and coatings were characterized by XRD, FT-IR, XPS, and SEM. The anti-corrosive performance of ZnO-HAP nanocomposites on mild steels was studied by the electrochemical method.

Fabrication of HAP Pigment
First, 4.29 g of Na 2 HPO 4 ·12H 2 O was prepared and dissolved in 50 mL of deionized water, and 2.22 g of CaCl 2 was dissolved in 20 mL of deionized water. Then, the solution of CaCl 2 was slowly dropped into a solution of Na 2 HPO 4 , and 3 g of urea as a pH regulator was dissolved into the above solution. The mixture was continuously stirred for 30 min and then poured into hydrothermal reaction kettles. The reactions occurred at 180 • C for 9 h (Equation (1)). After the reactions, the synthetic sample was washed. HAP was obtained after washing and drying at 80 • C for 12 h.

Fabrication of ZnO-HAP Nanocomposites Pigment
HAP and (CH 3 COO) 2 Zn·2H 2 O were weighed according to the mass ratio of ZnO:HAP (80:20 wt.%, 65:35 wt.%, and 50:50 wt.%). The weighed starting materials were fully dissolved in deionized water to obtain the mixed solutions, then NaOH was added to the mixed solutions to adjust the pH to 14. The mixtures were transferred to Teflon reactors and reacted at the heating temperature of 180 • C for 6 h in an oven (Equations (2) and (3)). In the end, the ZnO-HAP precipitates were washed with deionized water and dried at 80 • C for 12 h in the oven.

Preparation of Anti-Corrosive Coatings
First, ZnO, HAP, and ZnO-HAP (80:20 wt.%, 65:35 wt.%, 50:50 wt.%) pigments were respectively added to 10 g of epoxy resin (51 wt.%), and the pigments in the epoxy resin were whisked for 30 min to ensure an even spread. Then, 8 g of polyurethane (PU) was added to the above mixtures and stirred for 30 min. After achieving homogenous deconcentration, the commixtures were brushed onto the mild steel panels burnished by emery papers and cleaned by absolute ethyl alcohol beforehand. A TC100 coating thickness gauge was used to measure the thickness of the coatings and control them to 50 ± 5 µm (instrument error: ±(3% + 2 µm)). This thickness was experimentally determined as the most optimal for obtaining the maximum protective properties of the coatings. Finally, the coated mild steel panels were placed in a drying oven at 60 • C and cured for one week. The preparation process of the coating is shown in Figure 1.

Preparation of Anti-Corrosive Coatings
First, ZnO, HAP, and ZnO-HAP (80:20 wt.%, 65:35 wt.%, 50:50 wt.%) pigments were respectively added to 10 g of epoxy resin (51 wt.%), and the pigments in the epoxy resin were whisked for 30 min to ensure an even spread. Then, 8 g of polyurethane (PU) was added to the above mixtures and stirred for 30 min. After achieving homogenous deconcentration, the commixtures were brushed onto the mild steel panels burnished by emery papers and cleaned by absolute ethyl alcohol beforehand. A TC100 coating thickness gauge was used to measure the thickness of the coatings and control them to 50 ± 5 µm (instrument error: ±(3% + 2 µm)). This thickness was experimentally determined as the most optimal for obtaining the maximum protective properties of the coatings. Finally, the coated mild steel panels were placed in a drying oven at 60 °C and cured for one week. The preparation process of the coating is shown in Figure 1.

Characterizations
The coating thickness gauge (TC100, Shenzhen Red Dragon Instrument, Shanghai, China) was used to measure the coating thickness. The crystal structures and chemical composition were investigated by X-ray diffractometer (XRD) (Model D/max-IIIA, Rigaku, JEOL, Tokyo, Japan) and infrared spectroscopy (FT-IR) (Nicolet6700, TMO, Waltham, MA, USA). A scanning electron microscope (SEM) (JSM-IT300, JEOL, Tokyo, Japan) was used to analyze the anti-corrosive pigments' morphology and the coatings' section morphology. The dispersibility of the prepared anti-corrosive pigments in the coating was studied by electron probe microanalysis (EPMA) (JSM-IT300, JEOL, Tokyo, Japan) using an optical drop shape analyzer (DropMeter-A-100, Haishu Maishi Scientific Test Co. Ningbo, China). The corrosion resistance of the coatings was tested by an electrochemical workstation (CHI660E, SCE, Shanghai, China). A three-electrode system was used in the electrochemical tests. The mild steel panels served as working electrodes, the saturated calomel electrode as the reference electrode, and the platinum sheet (20 mm × 20 mm × 0.1 mm) as the counter electrode. Polarization curve tests were employed in the frequency range of 10 −2~1 0 5 Hz with amplitude perturbation of 20 mV. The polarization curves were tested with a scanning rate of 1 mV/s.

Characterization of Pigments
To understand the crystal phase composition of anti-corrosive pigments, XRD was used. Figure 2a shows the XRD pattern of HAP. The pattern of HAP was consistent with the standard peaks (PDF card No. 74-0565). The XRD pattern of ZnO is shown in Figure  2b, which was also consistent with the standard peaks (PDF card No. 79-2205). The crystal structures of HAP and ZnO are hexagonal [11,16]. As shown in Figure 2c-e, the peaks of ZnO and HAP were found in the XRD diffraction pattern, and the peaks were consistent

Characterizations
The coating thickness gauge (TC100, Shenzhen Red Dragon Instrument, Shanghai, China) was used to measure the coating thickness. The crystal structures and chemical composition were investigated by X-ray diffractometer (XRD) (Model D/max-IIIA, Rigaku, JEOL, Tokyo, Japan) and infrared spectroscopy (FT-IR) (Nicolet6700, TMO, Waltham, MA, USA). A scanning electron microscope (SEM) (JSM-IT300, JEOL, Tokyo, Japan) was used to analyze the anti-corrosive pigments' morphology and the coatings' section morphology. The dispersibility of the prepared anti-corrosive pigments in the coating was studied by electron probe microanalysis (EPMA) (JSM-IT300, JEOL, Tokyo, Japan) using an optical drop shape analyzer (DropMeter-A-100, Haishu Maishi Scientific Test Co. Ningbo, China). The corrosion resistance of the coatings was tested by an electrochemical workstation (CHI660E, SCE, Shanghai, China). A three-electrode system was used in the electrochemical tests. The mild steel panels served as working electrodes, the saturated calomel electrode as the reference electrode, and the platinum sheet (20 mm × 20 mm × 0.1 mm) as the counter electrode. Polarization curve tests were employed in the frequency range of 10 −2~1 0 5 Hz with amplitude perturbation of 20 mV. The polarization curves were tested with a scanning rate of 1 mV/s.

Characterization of Pigments
To understand the crystal phase composition of anti-corrosive pigments, XRD was used. Figure 2a shows the XRD pattern of HAP. The pattern of HAP was consistent with the standard peaks (PDF card No. 74-0565). The XRD pattern of ZnO is shown in Figure 2b, which was also consistent with the standard peaks (PDF card No. 79-2205). The crystal structures of HAP and ZnO are hexagonal [11,16]. As shown in Figure 2c-e, the peaks of ZnO and HAP were found in the XRD diffraction pattern, and the peaks were consistent with those of spectral ZnO and HAP, which indicates that the synthesized powders are ZnO-HAP nanocomposites. To explore the influence of the addition of ZnO-HAP nanocomposites on the epoxy resin coating, ZnO-HAP nanocomposites with a ratio of 65 wt.% ZnO and 35 wt.% HAP (a), epoxy resin coating (b), and ZnO-HAP/epoxy resin coating (c) were tested, respectively. The test results are shown in Figure 3. Figure 3a shows the characteristic absorption peaks of the Zn-O bond at 467 and 415 cm −1 , the peaks at 1089 and 1031 cm −1 are due to the asymmetric stretching of the P-O bond of phosphate groups, the sharp peaks at 604 and 563 cm −1 are attributed to the bending modes of the O-P-O bonds in phosphate groups, and the stretching vibration of -OH caused the peaks at 3434 and 604 cm −1 . The FT-IR spectra of ZnO-HAP nanocomposites further show that the anti-corrosive pigment is composed of ZnO and HAP, consistent with the XRD results. In Figure 3b, the stretching vibration absorption peaks of the N-H group and -OH were at 3380 cm −1 [17]. In Figure 3c, the polymer peaks of ZnO-HAP/epoxy resin show that the characteristic peaks of ZnO and HAP increased, compared with the peaks of pure epoxy resin. In contrast, the position of the peaks remained unchanged, indicating that ZnO-HAP nanocomposites can still exist stably after being added into epoxy resin.
with those of spectral ZnO and HAP, which indicates that the synthesized ZnO-HAP nanocomposites.
To explore the influence of the addition of ZnO-HAP nanocomposites resin coating, ZnO-HAP nanocomposites with a ratio of 65 wt.% ZnO and (a), epoxy resin coating (b), and ZnO-HAP/epoxy resin coating (c) were tively. The test results are shown in Figure 3. Figure 3a shows the characteri peaks of the Zn-O bond at 467 and 415 cm −1 , the peaks at 1089 and 1031 c the asymmetric stretching of the P-O bond of phosphate groups, the shar and 563 cm −1 are attributed to the bending modes of the O-P-O bonds groups, and the stretching vibration of -OH caused the peaks at 3434 and FT-IR spectra of ZnO-HAP nanocomposites further show that the anti-corr is composed of ZnO and HAP, consistent with the XRD results. In Figure 3b vibration absorption peaks of the N-H group and -OH were at 3380 cm −1 3c, the polymer peaks of ZnO-HAP/epoxy resin show that the characteristic and HAP increased, compared with the peaks of pure epoxy resin. In contra of the peaks remained unchanged, indicating that ZnO-HAP nanocomposi ist stably after being added into epoxy resin.    Figure 4a shows ZnO prepared under pH and hydrothermal conditions. The diameter of the ZnO crystal is about 150-200 nm. In Figure 4b, the morphology of ZnO attached to HAP remained unchanged after HAP modification. The element distribution is shown in Figure 5, and the surface elemental composition of ZnO-HAP nanocomposites is analyzed. The EDS patterns of the area scan and three selected spots are exhibited in Figure  6, and the EDS semi-quantitative results are listed in Table 1. The presence of Ca, P, Zn, and O in the images and the distribution of each element proved that ZnO and HAP were successfully combined. ZnO distributed uniformly on the HAP surface, which significantly reduced the accumulation of ZnO nanoparticles. The EDS semi-quantitative results proved that the tiny grains attached to HAP are ZnO.   Figure 4a shows ZnO prepared under pH and hydrothermal conditions. The diameter of the ZnO crystal is about 150-200 nm. In Figure 4b, the morphology of ZnO attached to HAP remained unchanged after HAP modification. The element distribution is shown in Figure 5, and the surface elemental composition of ZnO-HAP nanocomposites is analyzed. The EDS patterns of the area scan and three selected spots are exhibited in Figure 6, and the EDS semi-quantitative results are listed in Table 1. The presence of Ca, P, Zn, and O in the images and the distribution of each element proved that ZnO and HAP were successfully combined. ZnO distributed uniformly on the HAP surface, which significantly reduced the accumulation of ZnO nanoparticles. The EDS semi-quantitative results proved that the tiny grains attached to HAP are ZnO.   Figure 4a shows ZnO prepared under pH and hydrothermal conditions. The diameter of the ZnO crystal is about 150-200 nm. In Figure 4b, the morphology of ZnO attached to HAP remained unchanged after HAP modification. The element distribution is shown in Figure 5, and the surface elemental composition of ZnO-HAP nanocomposites is analyzed. The EDS patterns of the area scan and three selected spots are exhibited in Figure  6, and the EDS semi-quantitative results are listed in Table 1. The presence of Ca, P, Zn, and O in the images and the distribution of each element proved that ZnO and HAP were successfully combined. ZnO distributed uniformly on the HAP surface, which significantly reduced the accumulation of ZnO nanoparticles. The EDS semi-quantitative results proved that the tiny grains attached to HAP are ZnO.     To study the internal structure of the coating, SEM was used to observe the crosssection of the epoxy resin coating after adding the anti-corrosive pigment. Figure 7a shows the fracture surface of the pure epoxy coating, while Figure 7b shows the cross-section of   To study the internal structure of the coating, SEM was used to observe the crosssection of the epoxy resin coating after adding the anti-corrosive pigment. Figure 7a shows the fracture surface of the pure epoxy coating, while Figure 7b shows the cross-section of  To study the internal structure of the coating, SEM was used to observe the crosssection of the epoxy resin coating after adding the anti-corrosive pigment. Figure 7a shows the fracture surface of the pure epoxy coating, while Figure 7b shows the cross-section of HAP/epoxy resin. From the cross-section of the coating, HAP appears to agglomerate in the epoxy resin, which will affect the anti-corrosion effect. Figure 7c shows the cross-section of zinc oxide/epoxy resin. No significant accumulation of zinc oxide particles was found. Figure 7d shows the cross-section of ZnO-HAP/epoxy resin with 65% ZnO content. As zinc oxide was attached to the HAP surface, HAP agglomeration improved. ZnO-HAP was evenly distributed in the epoxy resin coating, which increases the density of the coating and makes the diffusion channel of the electrolyte more complicated. HAP/epoxy resin. From the cross-section of the coating, HAP appears to agglomerate in the epoxy resin, which will affect the anti-corrosion effect. Figure 7c shows the cross-section of zinc oxide/epoxy resin. No significant accumulation of zinc oxide particles was found. Figure 7d shows the cross-section of ZnO-HAP/epoxy resin with 65% ZnO content. As zinc oxide was attached to the HAP surface, HAP agglomeration improved. ZnO-HAP was evenly distributed in the epoxy resin coating, which increases the density of the coating and makes the diffusion channel of the electrolyte more complicated. To study the dispersion of ZnO-HAP nanocomposites in coatings, EDS was used to analyze the ZnO-HAP/epoxy resin. The element distribution is shown in Figure 8, and the surface elemental composition of ZnO-HAP nanocomposites is analyzed. The EDS patterns of the three selected spots are exhibited in Figure 9, and the EDS semi-quantitative results are listed in Table 2. Ca, P, and Zn elements were evenly distributed in the scanning surface of the coating, which proves that ZnO-HAP nanocomposites have good dispersion in an epoxy resin coating.  To study the dispersion of ZnO-HAP nanocomposites in coatings, EDS was used to analyze the ZnO-HAP/epoxy resin. The element distribution is shown in Figure 8, and the surface elemental composition of ZnO-HAP nanocomposites is analyzed. The EDS patterns of the three selected spots are exhibited in Figure 9, and the EDS semi-quantitative results are listed in Table 2. Ca, P, and Zn elements were evenly distributed in the scanning surface of the coating, which proves that ZnO-HAP nanocomposites have good dispersion in an epoxy resin coating. HAP/epoxy resin. From the cross-section of the coating, HAP appears to agglomerate in the epoxy resin, which will affect the anti-corrosion effect. Figure 7c shows the cross-section of zinc oxide/epoxy resin. No significant accumulation of zinc oxide particles was found. Figure 7d shows the cross-section of ZnO-HAP/epoxy resin with 65% ZnO content. As zinc oxide was attached to the HAP surface, HAP agglomeration improved. ZnO-HAP was evenly distributed in the epoxy resin coating, which increases the density of the coating and makes the diffusion channel of the electrolyte more complicated. To study the dispersion of ZnO-HAP nanocomposites in coatings, EDS was used to analyze the ZnO-HAP/epoxy resin. The element distribution is shown in Figure 8, and the surface elemental composition of ZnO-HAP nanocomposites is analyzed. The EDS patterns of the three selected spots are exhibited in Figure 9, and the EDS semi-quantitative results are listed in Table 2. Ca, P, and Zn elements were evenly distributed in the scanning surface of the coating, which proves that ZnO-HAP nanocomposites have good dispersion in an epoxy resin coating.

Mechanical Property Analysis
To study the mechanical performance of different coatings, the hardness and adhesion of the coatings were investigated according to the GB/T6739-2006 pencil method to determine the hardness of paint film and the GB/T9286-1998 color paint and paint film grid test, respectively. The grade description of the adhesion test is shown in Table 3, where the higher the proportion of unshed lattice is, the higher the adhesion level of the coating. Table 4 shows the mechanical performance test results of different coatings. The hardness test results showed that the coating hardness of the ZnO-HAP composite pigment with 80 wt.% ZnO content was the lowest. From the adhesion test, the adhesion between the ZnO/epoxy resin and metal was the worst. With the proportion of HAP in pigment gradually increasing, the adhesion of the coating also steadily increased.  Figure 9. EDS analysis of area scan and three selected spots in Figure 7. Table 2. EDS analysis results of the area scan and three selected spots shown in Figure 9.

Mechanical Property Analysis
To study the mechanical performance of different coatings, the hardness and adhesion of the coatings were investigated according to the GB/T6739-2006 pencil method to determine the hardness of paint film and the GB/T9286-1998 color paint and paint film grid test, respectively. The grade description of the adhesion test is shown in Table 3, where the higher the proportion of unshed lattice is, the higher the adhesion level of the coating. Table 4 shows the mechanical performance test results of different coatings. The hardness test results showed that the coating hardness of the ZnO-HAP composite pigment with 80 wt.% ZnO content was the lowest. From the adhesion test, the adhesion between the ZnO/epoxy resin and metal was the worst. With the proportion of HAP in pigment gradually increasing, the adhesion of the coating also steadily increased.   Figure 10 depicts the Nyquist plots of the mild steels coated with different coatings soaked in 3.5 wt.% NaCl solution for 24 and 72 h. In Nyquist plots, the resistance of the coating increased with the radius of the arc [18]. As shown in Figure 10, the arcs of the pure epoxy resin coating, HAP/epoxy resin, and ZnO/epoxy resin were small, indicating that the corrosion resistance of these three coatings is not excellent and mild steel electrodes are prone to damage in 3.5 wt.% NaCl solution. The Nyquist plots' ZnO-HAP/epoxy system arc after soaking for 24 and 72 h was always higher than the others. The steels coated with ZnO-HAP nanocomposites always exhibited more excellent corrosion resistance. These results indicate that adding the ZnO-HAP composite anti-corrosive pigment can significantly improve the anti-corrosive effect of the epoxy resin coating. Among the three kinds of ZnO-HAP coatings with different proportions, the ZnO-HAP coating with the ratio of 65 wt.% ZnO + 35 wt.% HAP had the best anti-corrosion performance. When the HAP content was too high (50 wt.% ZnO + 50 wt.% HAP), due to the influence of the hydrophilic group -OH on the HAP surface, the hydrophobicity of the coating surface decreased, and the anti-corrosion performance of the coating also decreased. pigment can significantly improve the anti-corrosive effect of the epoxy resin coating. Among the three kinds of ZnO-HAP coatings with different proportions, the ZnO-HAP coating with the ratio of 65 wt.% ZnO + 35 wt.% HAP had the best anti-corrosion performance. When the HAP content was too high (50 wt.% ZnO + 50 wt.% HAP), due to the influence of the hydrophilic group -OH on the HAP surface, the hydrophobicity of the coating surface decreased, and the anti-corrosion performance of the coating also decreased. To further explore the anti-corrosive performance of the anti-corrosive pigment, the Bode plots of the coatings were analyzed. Figures 11 and 12 are Bode plots of different coatings soaked for 24 and 72 h, respectively, where (a) is the impedance modulus of each coating at different frequencies, and (b) is the phase angle plots of each coating at different frequencies. As shown in Figure 11, the coatings were still in the initial stage after soaking for 24 h. As can be seen from the plots, these phase angle curves have only one time constant, indicating that the organic coatings at this time are equivalent to an isolation layer with a significant resistance value and a small capacitance. At the initial immersion stage, as the electrolyte solution penetrated the organic coating, the impedance value in the curve gradually decreased with the increase of the horizontal coordinate. The electrolyte solution gradually penetrated the coatings, increasing the capacitance and decreasing the resistance.

EIS Experiments
As shown in Figure 12, the Bode plots after soaking for 72 h presented two characteristics of time constants. The one at higher frequency implies the penetration of electrolytes through the coating, and the other at middle frequency relates to the corrosion reaction. A region with a low slope can be observed from the middle frequency band in Figure 12a, which presents Warburg impedance characteristics. Due to the blocking effect of the additive particles in the organic coating, the electrolyte solution occupies the space between the particles and sinuous infiltration, and the mass transfer direction of the reaction particle is not parallel to the direction of the concentration gradient, which is called "tangential diffusion" [19]. From the impedance spectra in Figures 11 and 12, it can be seen that the resistance of the coating of 65 wt.% ZnO and 35 wt.% HAP anti-corrosive pigments was the largest, indicating the best anti-corrosive performance.
Bode plots of different electrodes soaked for 24h (a) impedance modulus spectrum, (b) phase angle spectrum Email address: Xiaohong Xu: xuxh@whut.edu.cn; Jianfeng Wu: wujf@whut.edu.cn   The EIS results were fitted using the equivalent circuit shown in Figure 13. Figure 13a-c show the equivalent circuits of different coatings soaked for 24 and 72 h. In the circuit model, resistors R1, R2, and R3 correspond to solution resistance, coating resistance, and corrosion reaction linear polarization resistance, respectively [20,21]. Equivalent element CPE represents non-ideal capacitance, where CPE1 represents coating capacitance and CPE2 represents electric double-layer capacitance. The W element represents the Warburg component [22]. The EIS results were fitted using the equivalent circuit shown in Figure 13. Figure  13a-c show the equivalent circuits of different coatings soaked for 24 and 72 h. In the circuit model, resistors R1, R2, and R3 correspond to solution resistance, coating resistance, and corrosion reaction linear polarization resistance, respectively [20,21]. Equivalent element CPE represents non-ideal capacitance, where CPE1 represents coating capacitance and CPE2 represents electric double-layer capacitance. The W element represents the Warburg component [22]. As shown in Figure 13a, ZnO-HAP/epoxy coatings and HAP/epoxy coatings have sound anti-corrosion effects. The electrolyte did not penetrate the coating/base material interface after immersion for 24 h [23]. At this point, the coating is equivalent to an insulating layer with a considerable resistance value and a small capacitance value. Figure 13b shows the equivalent circuit of the pure epoxy resin coating. After soaking the pure epoxy resin coating for 24 and 72 h, respectively, the electrolyte solution entering the coating reached saturation, and the electrolyte solution reached the coating/metal interface, resulting in corrosion and damage to the metal surface. Since no anti-corrosive pigment was added to the pure epoxy resin coating, the diffusion process of the electrolyte solution was not affected by the mass transfer process. Equivalent circuits of ZnO/epoxy coatings As shown in Figure 13a, ZnO-HAP/epoxy coatings and HAP/epoxy coatings have sound anti-corrosion effects. The electrolyte did not penetrate the coating/base material interface after immersion for 24 h [23]. At this point, the coating is equivalent to an insulating layer with a considerable resistance value and a small capacitance value. Figure 13b shows the equivalent circuit of the pure epoxy resin coating. After soaking the pure epoxy resin coating for 24 and 72 h, respectively, the electrolyte solution entering the coating reached saturation, and the electrolyte solution reached the coating/metal interface, resulting in corrosion and damage to the metal surface. Since no anti-corrosive pigment was added to the pure epoxy resin coating, the diffusion process of the electrolyte solution was not affected by the mass transfer process. Equivalent circuits of ZnO/epoxy coatings soaked for 24 and 72 h, HAP/epoxy coatings soaked for 72 h, and ZnO-HAP/epoxy coatings soaked for 72 h are shown in Figure 13c. Since the charge and discharge process of coating and the corrosion reaction of metal are affected by the mass transfer process, the Warburg element was added to the equivalent circuit.
The fitting parameters of the circuit are shown in Table 6. The resistance (R2) of ZnO-HAP/epoxy resin was significantly higher than that of the pure epoxy resin system, ZnO/epoxy system, and HAP/epoxy system. Significant resistance indicates that corrosive substances are difficult to enter the coating to corrode the mild steel. The high W value of ZnO-HAP/epoxy resin indicates that the ZnO-HAP nanocomposite can inhibit the diffusion of corrosive substances and thus slow down the corrosion reaction. Note: "-" means there is no such data in the circuit fitting result.
The deviation from pure capacity reflected by CPE indicates the roughness and energy dissipation of the electrode surface [24,25]. The following formula expresses the impedance of a CPE element: where ω is the angular frequency (rad/s), and n is the power number, defined as n = α (π/2) (α is the constant phase angle of CPE (rad)), which indicates the deviation of the interface from the ideal capacitor, representing resistive, Warburg, and capacitive behaviors for n = 0, 0.5, and 1, respectively [26,27]. Y 0 (admittance) is a CPE constant (s n ·Ω −1 ·Cm −2 ), directly proportional to the active surface area exposed to the electrolyte. Both Y 0 and n are independent of frequency [28,29].
The parameter values estimated by impedance spectrum fitting analysis are summarized in Table 5. As described in Table 6, the polarization resistance corresponding to R2 is the highest, indicating that the epoxy coating has better corrosion resistance due to ZnO-HAP. Y 1 is related to the active surface area involved in electrochemical reactions. With the addition of anti-corrosive pigments to the coating, this value decreased (Y 1 value was highest for pure epoxy coating), which can be attributed to the role of anti-corrosive pigments as a barrier during corrosion, which reduces the porosity of the epoxy coating and makes it more difficult for the metal surface to contact the electrolyte [30,31]. An increase in n 1 means a decrease in the inhomogeneity and roughness of the coating surface, indicating a reduction in the corrosion sensitivity of the coating relative to the substrate. A higher n value means a better surface performance of the coatings [32]. In Table 6, the n 1 values of HAP/epoxy coatings were the largest, followed by ZnO-HAP (65% ZnO, 35% HAP) coatings, and the ZnO/epoxy coating had the smallest n 1 value. The results show that the ZnO-HAP/epoxy system has better surface performance. With the extension of the immersion time, the Y 1 value of the ZnO-HAP/epoxy coating decreased, and the n 1 value increased due to the increased density of the coating due to the refractory chlorine apatite solidified by HAP after the chloride ion entered the coating.

Polarization Curve Measurements
The corrosion rates of different mild steels were investigated by potentiodynamic polarization curves (Figure 14). Polarization curves are composed of anode and cathode polarization curves: Tafel anode slope (b a ) and Tafel cathode slope (b c ) can be fitted, respectively, and corrosion potential (E corr ) and corrosion current density (I corr ) can also be obtained by the Tafel extrapolation method [33]. These Tafel parameters can estimate the anti-corrosion performance of anti-corrosive pigments. The E corr value is proportional to the corrosion resistance of the coating, and the I corr value is proportional to the corrosion inhibition performance of the coating. The fitting parameters are shown in Table 7.

Polarization Curve Measurements
The corrosion rates of different mild steels were investigated by potentiodynamic polarization curves (Figure 14). Polarization curves are composed of anode and cathode polarization curves: Tafel anode slope (ba) and Tafel cathode slope (bc) can be fitted, respectively, and corrosion potential (Ecorr) and corrosion current density (Icorr) can also be obtained by the Tafel extrapolation method [33]. These Tafel parameters can estimate the anti-corrosion performance of anti-corrosive pigments. The Ecorr value is proportional to the corrosion resistance of the coating, and the Icorr value is proportional to the corrosion inhibition performance of the coating. The fitting parameters are shown in Table 7.   Table 7, the Ecorr value of pure epoxy resin was −0.828 V and the Icorr value was 34.410 µA/cm 2 , while the Ecorr value of ZnO/epoxy resin was −0.869 V and the Icorr value was 71.923 µA/cm 2 . The reason for this result is that the adhesion between the ZnO/epoxy resin and metal was the worst: the high polarization destabilizes this type of coating, and it peels off the substrate and cracks, and this creates accessibility for the flow of the corrosive medium to the steel. Thus, it shows a slightly worse result than pure epoxy. The Ecorr value of ZnO-HAP/epoxy was the highest when the ZnO to HAP ratio was 65:35 wt.% and its Icorr value was only 1.2 µA/cm 2 . High Ecorr and low Icorr values confirmed that ZnO-HAP nanocomposites could effectively inhibit the corrosion reaction. The results show that ZnO and HAP can protect mild steel. Figure 15 illustrates the anti-corrosion mechanism of the ZnO-HAP nanocomposites. As shown in Figure 15a, for the coating of pure epoxy resin with no anti-corrosion   Table 7, the E corr value of pure epoxy resin was −0.828 V and the I corr value was 34.410 µA/cm 2 , while the E corr value of ZnO/epoxy resin was −0.869 V and the I corr value was 71.923 µA/cm 2 . The reason for this result is that the adhesion between the ZnO/epoxy resin and metal was the worst: the high polarization destabilizes this type of coating, and it peels off the substrate and cracks, and this creates accessibility for the flow of the corrosive medium to the steel. Thus, it shows a slightly worse result than pure epoxy. The E corr value of ZnO-HAP/epoxy was the highest when the ZnO to HAP ratio was 65:35 wt.% and its I corr value was only 1.2 µA/cm 2 . High E corr and low I corr values confirmed that ZnO-HAP nanocomposites could effectively inhibit the corrosion reaction. The results show that ZnO and HAP can protect mild steel. Figure 15 illustrates the anti-corrosion mechanism of the ZnO-HAP nanocomposites. As shown in Figure 15a, for the coating of pure epoxy resin with no anti-corrosion pigment, electrolyte solution can penetrate directly through the coating to the steel substrate. Cl − in the electrolyte easily causes local corrosion, such as pitting corrosion and gap corrosion, and destroys metal passivation. As shown in Figure 15b, ZnO adheres to the surface of the HAP sheets to form a complex staggered structure, filling the pores of the epoxy resin and complicating the permeating path of the aggressive species. Meanwhile, the substitution reaction between Cl − in solution and -OH on HAP produced stable chlorapatite [34], which improved the corrosion resistance of the coating. Therefore, the synergistic anti-corrosion effect of ZnO and HAP can provide adequate protection for mild steel. pigment, electrolyte solution can penetrate directly through the coating to the steel substrate. Cl − in the electrolyte easily causes local corrosion, such as pitting corrosion and gap corrosion, and destroys metal passivation. As shown in Figure 15b, ZnO adheres to the surface of the HAP sheets to form a complex staggered structure, filling the pores of the epoxy resin and complicating the permeating path of the aggressive species. Meanwhile, the substitution reaction between Cl − in solution and -OH on HAP produced stable chlorapatite [34], which improved the corrosion resistance of the coating. Therefore, the synergistic anti-corrosion effect of ZnO and HAP can provide adequate protection for mild steel.

Conclusions
The ZnO-HAP nanocomposite anti-corrosive pigment was successfully prepared using the hydrothermal in situ synthesis technique, loading ZnO onto the HAP sheets. Electrochemical analysis showed that the anti-corrosive effect was the best when the ratio of ZnO and HAP in the ZnO-HAP anti-corrosive pigment was 65:35 wt.%. By combining zinc oxide with HAP, the adverse effect of zinc oxide on the mechanical properties (adhesion) of the epoxy resin coating was effectively improved, and the adhesion of the coating improved from grade 2 to grade 0. Similarly, zinc oxide was attached to the HAP surface, which improved the hydrophobicity of HAP. The complementary mechanism of zinc oxide and HAP significantly improved the coating's anti-corrosion effect, reduced the porosity of the epoxy coating, and complicated corroded materials' penetration path. At the same time, Cl − penetrated the coating and exchanged with hydroxyl groups in the HAP crystal to form insoluble chlorine apatite, which was fixed in the coating to further decrease the pores of the coating and increase the service life.
Author Contributions: Writing-review and editing, resources, methodology, X.X.; writing-review and editing, methodology, investigation, formal analysis, H.W.; resources, project administration, funding acquisition, conceptualization, J.W.; formal analysis, data curation, Z.C.; investigation, X.Z.; formal analysis, M.L. All authors have read and agreed to the published version of the manuscript.
Funding: There is no funding for our project.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest.

Conclusions
The ZnO-HAP nanocomposite anti-corrosive pigment was successfully prepared using the hydrothermal in situ synthesis technique, loading ZnO onto the HAP sheets. Electrochemical analysis showed that the anti-corrosive effect was the best when the ratio of ZnO and HAP in the ZnO-HAP anti-corrosive pigment was 65:35 wt.%. By combining zinc oxide with HAP, the adverse effect of zinc oxide on the mechanical properties (adhesion) of the epoxy resin coating was effectively improved, and the adhesion of the coating improved from grade 2 to grade 0. Similarly, zinc oxide was attached to the HAP surface, which improved the hydrophobicity of HAP. The complementary mechanism of zinc oxide and HAP significantly improved the coating's anti-corrosion effect, reduced the porosity of the epoxy coating, and complicated corroded materials' penetration path. At the same time, Cl − penetrated the coating and exchanged with hydroxyl groups in the HAP crystal to form insoluble chlorine apatite, which was fixed in the coating to further decrease the pores of the coating and increase the service life.
Author Contributions: Writing-review and editing, resources, methodology, X.X.; writing-review and editing, methodology, investigation, formal analysis, H.W.; resources, project administration, funding acquisition, conceptualization, J.W.; formal analysis, data curation, Z.C.; investigation, X.Z.; formal analysis, M.L. All authors have read and agreed to the published version of the manuscript.
Funding: There is no funding for our project.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest.