Comparative Study of the Corrosion Resistance of Air-Plasma-Sprayed Ca₂SiO₄ and Al₂O₃ Coatings in Salt Water

Abstract

In this study, Ca₂SiO₄ coating was sprayed on stainless steel substrate and the corrosion resistance of the as-sprayed coating was studied in salt water. At the same time, Al₂O₃ coatings were produced by air-plasma-sprayed technology as comparison. Immersion test was carried out to evaluate the protection performance of coatings. Potentiodynamic polarization curves and electrochemical impedance spectroscopy (EIS) plots were also analyzed. The results indicated that Ca₂SiO₄ coatings showed a better protection performance than Al₂O₃ coatings. During the immersion, various calcium carbonate crystals appeared on the surface of Ca₂SiO₄ coatings. Ca(OH)₂ was released from Ca₂SiO₄ coatings into NaCl aqueous solution, increasing the alkalinity, which is in favor of the formation of passivation film, and thus improves the corrosion resistance. Ca₂SiO₄ coatings became denser after immersion due to the fact that the pores and micro cracks were filled with hydration products i.e., hydrated calcium silicate (C–S–H) gel. On the contrary, the microstructure of Al₂O₃ coatings became loose and obvious rusty spots were observed on the surface after the immersion test.

1. Introduction

Marine resources play an important role in the development of world economics. However, sea water is a corrosive medium. Sea water contains large number of ions, such as Na⁺, Cl⁻, Mg²⁺, SO₄²⁻, K⁺ and CO₃²⁻. The salt concentration is about 3.5 wt % in sea water. Therefore, corrosion in sea water is very severe [1]. Marine engineering materials, especially metal structural materials, suffer from different levels of corrosion damage. Thus, measures must be taken to protect the metal materials served in the harsh marine environment [2]. Spraying anti-corrosion coatings on the metal material surface can effectively slow down the corrosion rate [3].

At present, the widely applied anti-corrosion technologies in the marine corrosion field include organic paint, zinc-aluminium alloy, amorphous alloy coating, oxide ceramic coatings [4], etc. Organic paint might age and then spall when it works for a long time under the exposure of sun or in sea water, so its durability needs to be improved [5]. Zinc-aluminium alloy and amorphous alloy coatings exhibit good corrosion resistance in the marine environment, but bad electrical insulation which cannot hinder the transfer of electric charge in the corrosion reaction, thus they are sacrificial anode protection [6]. The oxides such as Al₂O₃, Al₂O₃/TiO₂ and Al₂O₃/Cr₂O₃ showed good corrosion resistance [7] and in previous studies, ceramic coatings were effective to isolate the corrosive medium [8].

Cement is another inorganic material used widely in the marine environment, especially in marine building infrastructure. The main components of cement are dicalcium silicate (C₂S), tricalcium silicate (C₃S), tricalcium aluminate and tetracalcium aluminoferrite [9]. Earlier studies have revealed that tricalcium aluminate would react with sulfate ion in the sea water to generate swelling salt, causing the volume expansion of concrete and inducing crack propagation [10]. Cement is a kind of gel material with dicalcium silicate and tricalcium silicate as the main composition, which would react with water to generate hydrated calcium silicate gel. Hydrated calcium silicate gel would fill the pores and micro cracks in the structure and improve the permeability resistance. In general, the more hydrated the calcium silicate, the denser the structure. Therefore, the compressive strength of cement stone would be higher. The equal mass of C₂S would produce more hydrated calcium silicate than C₃S does in the process of hydration. It was found that after two years’ hydration process, the ultimate strength of C₂S block was 143.5 MPa but the ultimate strength of C₃S block was 113 MPa, the ultimate strength of C₂S block is higher by 26% than that of C₃S block [11]. Thus, dicalcium silicate may produce more C–S–H gel and improve the density of coating. The melting point of silicate minerals is very high. Plasma spraying uses high-temperature plasma as its heat source; the flame temperature is so high that most ceramic material and metal material can be flame-melted [12]. Plasma-sprayed coatings usually have good bond strength. X. Liu et al. mentioned that the bond strength of dicalcium silicate coatings was about 38.9 ± 3.5 MPa (the thickness of the coating was about 380 μm) [13]. However, almost no study discussed the corrosion resistance of dicalcium silicate coatings in marine environment.

In the current work, dicalcium silicate coatings and alumina coatings were produced by air-plasma-sprayed technology, the phase composition and microstructure of coatings were studied. In addition, immersion test was carried out to evaluate the protection performance of coatings. Potentiodynamic polarization curves and electrochemical impedance spectroscopy (EIS) plots were also analyzed.

2. Materials and Methods

2.1. Feedstock Power and Coating Preparation

Dicalcium silicate (Ca₂SiO₄) powders were synthesized by solid-state reaction using analytical-grade calcium carbonate and silicon dioxide with an initial CaCO₃/SiO₂ molar ratio of 2.0. The materials were wet-mixed for 10 h. The typical size of Ca₂SiO₄ powders was about 30–50 μm. Commercially available alumina powders were used. F4-MB plasma gun was applied to deposit the coatings including NiCr bond layer and oxide layer; 304 stainless steel with the dimensions of 30 mm × 15 mm × 1.25 mm was used as substrate. Prior to spraying, the stainless steel substrate was degreased ultrasonically in ethanol and then grit-blasted with corundum. The detailed plasma spraying parameters are listed in

Table 1. Plasma-spraying operating parameters.

Parameters	NiCr	Ca2SiO4	Al2O3
Arc current, A	600	650	650
Primary plasma gas (Ar), slpm	57	49	49
Secondary plasma gas (H2), slpm	8	9	9
Carrier gas, slpm	3.5	4.0	4.0
Powder feed rate, g/min	18	35	35
Spray distance, mm	120	110	110

. The thickness of bond coat is about 15–30 μm, and the Ca₂SiO₄ and Al₂O₃ coatings are both 130–150 μm.

2.2. Microstructure and Property Characterization

The crystalline phase compositions of the powders and coatings were analyzed by a powder X-ray diffraction using a Rigaku D/Max2550 diffract meter (Rigaku, Tokyo, Japan) with nickel-filtered Cu Kα radiation (λ = 0.15406 nm). The microstructures of coatings were investigated by scanning electron microscope (SEM) TM3000 (Hitachi, Tokyo, Japan) equipped with energy dispersive spectrometer (EDS). The Image-Pro Plus software (version 6.0.0.) was applied to measure the porosity of coatings according to the Image analysis method. Each data of porosity was based on 10 pieces of coating cross-section SEM images with 1000×. The bond strength is measured according to ASTM C-633 [14]; Instron-5592 tension, pressure universal testing machine (Instron, New York, USA) is used. The substrate is 1Cr 13 stainless steel, and the dimension is φ 25 mm× h 60 mm.

2.3. Corrosion Tests

The corrosion resistance of plasma-sprayed coatings was investigated by immersion test in unstirred 3.5 wt % NaCl aqueous solution at (35 ± 2) °C. After being immersed for 5, 10, 20, 30 days, the coatings were denoted as Al₂O₃-X and C₂S-X (X = 5, 10, 20, 30). Electrochemical corrosion test was also conducted. The electrochemical systems used in the tests were Ametek 1287 & 1260 (Berwyn, PA, USA). Polarization curves were recorded from −0.5 V to 0.6 V vs. open circuit voltage (Eocp) at 0.5 mV/s potential sweep rate. Electrochemical tests were conducted in a three-electrode cell using platinum mesh as counter electrode, a saturated calomel electrode (SCE) as reference electrode and the coating as working electrode. Electrolyte was 3.5 wt % NaCl aqueous solution prepared from analytic reagent sodium chloride and deionized water. Electrochemical impedance spectroscopy (EIS) plots were recorded after immersing the specimens in 3.5 wt % NaCl aqueous solution for about 30 min when the open circuit potentials became almost steady, the recorded frequency range was from 10⁶ Hz to 10⁻² Hz, the applied perturbation was 10 mV AC amplitude.

3. Results and Discussion

3.1. Phase Compositions and Microstructures

The XRD patterns of the powders and as-sprayed coatings are shown in Figure 1. Ca₂SiO₄ powders were composed of γ-Ca₂SiO₄, while the phase in as-sprayed Ca₂SiO₄ coatings was β-Ca₂SiO₄ with a part of glassy phase. In Figure 1c, the increase in the background area under the peaks indicated the glassy phase [13]. In the high-temperature plasma flame flow, γ-Ca₂SiO₄ powders were rapidly melted into droplets, and the droplets hit/impacted the substrate, then quickly deposited on the surface to form coating. β-Ca₂SiO₄ was generated during the rapid cooling process, some of the droplets did not crystallize in time and were converted into glass phase. The rate of change in heat evolution leads to the formation of different types of polymorphs (α, α_H’, α_L’, β and γ) of Ca₂SiO₄ [15], as shown in Figure 2.

γ-Ca₂SiO₄ is the stable phase at room temperature. It is generally believed that the higher the formation temperature, the higher the hydration rate, so the hydration rate rank of the different phase is α > α_H’ > α_L’ > β > γ [16]. Previous investigations indicated that γ-Ca₂SiO₄ almost cannot react with water [17], while β-Ca₂SiO₄ can react with water to generate gel of hydrated calcium silicate. The alumina powders were α-Al₂O₃, but as-sprayed coating contained α-Al₂O₃ and γ-Al₂O₃. A part of α-Al₂O₃ transformed into γ-Al₂O₃ during the plasma spraying process. The phase of Al₂O₃ coatings produced by plasma spraying mainly existed in γ-Al₂O₃ phase [18].

The surface and cross-sectional morphologies of as-sprayed coatings are shown in Figure 3. The surface morphology of coatings revealed that the coating was built up from melted droplets. The cross-sectional view of the coating showed obvious pores and micro cracks. The porosity of Ca₂SiO₄ coating and Al₂O₃ coating were 6.50% and 8.85%, respectively. It showed that Ca₂SiO₄ coating was a little denser than Al₂O₃ coating. The bond strength of Ca₂SiO₄ coatings and Al₂O₃ coating were 35.0 ± 3.5 MPa and 26.0 ± 3.5 MPa, which revealed that the bonding strength between Ca₂SiO₄ coating and metal substrate was better than that for Al₂O₃ coating.

3.2. Immersion Test

Figure 4 shows the surface morphologies of Ca₂SiO₄ coating after being immersed in still 3.5 wt % NaCl solution at (35 ± 2) °C for 5, 10, 20 and 30 days.

As shown in Figure 4, after being immersed in 3.5 wt % NaCl solution for 5 days, there were some rhombic solid particles on the surface of Ca₂SiO₄ coating. XRD patterns in Figure 5a show that the particles are rhombic calcite (CaCO₃). More and more calcium carbonate crystals were formed on the surface of Ca₂SiO₄ coating as the immersed time increased. When immersed for 20 days, chrysanthemum-like or spindle-like [19] aragonite (CaCO₃) and spherical vaterite (CaCO₃) appeared. After 30 days, calcium carbonate crystals spread all over the coating surface. It was easy for calcium carbonate crystals to fall from the coating surface.

The chemical processes that took place on the surface of Ca₂SiO₄ during immersion of specimens in 3.5% NaCl solution are as follows [20]:

Ca₂SiO₄ + nH₂O → mCaO·SiO₂·xH₂O + (2 − m)Ca(OH)₂ (n = x + 2 − m)

(1)

Ca(OH)₂ + CO₂ → CaCO₃·H₂O

(2)

or Ca(OH)₂ + CO₂ → CaCO₃ + H₂O

(3)

And the Na⁺ and Cl⁻ ions have no effect on the hydration process of Ca₂SiO₄ coatings.

Han et al. studied the factors affecting the phase and morphology of CaCO₃ prepared by a bubbling method. They found that low Ca²⁺ concentration or sufficient supply of CO₂ was in favor of the formation of spherical vaterite [21]. At the beginning of the immersion in 3.5 wt % NaCl solution, Ca₂SiO₄ coating reacted with water to release Ca(OH)₂. High Ca²⁺ concentration promoted to form rhombic calcite as shown in Figure 5a. As the slow dissolution of CO₂ gas, Ca²⁺ concentration decreased, spherical vaterite formed on the coating surface as shown in Figure 5b. In reference [21], the size of calcium carbonate particle was below 10 μm, but the particle size of calcite and vaterite formed on Ca₂SiO₄ coating were more than 50 μm. During the immersion, the release of Ca(OH)₂ and dissolution of CO₂ gas were very slow, and the supersaturation of solution remained at a low level. In addition, supersaturation played an important role on the rate of crystal nucleus formation and crystal growth, and the supersaturation influenced the size and distribution of the formed particles. When the supersaturation was low, the ratio of crystal growth rate and crystal nucleus formation rate is bigger, so the crystal is bigger with relatively complete crystal shape, but the crystallization rate was slow [22]. When the supersaturation of solution was kept low for a long time, the crystal growth was predominant in the solution. Therefore, the size of calcium carbonate particles on Ca₂SiO₄ coating was larger than that in other studies.

Figure 6 shows the surface morphologies of Al₂O₃ coating after being immersed in still 3.5 wt % NaCl solution at (35 ± 2) °C for 5, 10, 20 and 30 days. When immersed in 3.5 wt % NaCl solution for 5 days, there were more pores on the Al₂O₃ coating surface, thus the corrosion medium could permeate into the coating more easily than as-sprayed Al₂O₃ coating. Rust staining was discovered on the 10th day, and then rust staining became larger. Because of the dissolution of CO₂ gas in air, the pH value of the solution was below 7, and the swell was caused by the generation of hydrogen during the corrosion process. Figure 7 shows the surface morphologies of coatings after being immersed 30 days, the Al₂O₃ coating appeared obvious rusty spot, but there was no rusty spot on Ca₂SiO₄ coating. Figure 8 shows the EDS analysis of the region marked with a yellow rectangle in Figure 6, the rust staining contains Fe, Ni, Cr and Cl elements. It is indicated that the corrosive medium penetrated into the metal substrate and the substrate suffered from corrosion.

Figure 9 shows the cross-section images of the Ca₂SiO₄ coating after being immersed for different times. In order to present the hydration area of the Ca₂SiO₄ coating, the morphology of the yellow box in Figure 9c is shown in Figure 9f. The bottom figures, Figure 9d–f, are the same as the top ones with higher magnification. From Figure 9, the porosity of as-sprayed coating was 6.05%, C₂S-10 was 3.05%, and C₂S-30 was 1.10%. The porosity of C₂S coating was reduced after being immersed. Figure 10 also shows the change of coating surface. It was evident that the pore became smaller and the cracks became less after immersion for 30 days. Figure 10 shows that hydration products C–S–H gel formed during the immersion test, and the pores and micro cracks were filled with C–S–H gel.

In Figure 11, it is obvious that hydration phenomenon occurred at the top of 80–110 μm thickness of the coating after being immersed for 30 days. The CO₂ gas in air dissolved in NaCl solution to form CO₃²⁻ or HCO₃⁻. They reacted with Ca(OH)₂ and CaCO₃ crystal was formed at the coating surface. When the corrosion medium containing CO₃²⁻ or HCO₃⁻ penetrated into the internal of the coating, C₂S-30 showed carbon element enrichment at the top of 80–110 μm thickness.

As shown in Figure 12, during the immersion, the internal of Al₂O₃ coatings becomes looser and more porous. The porosity of as-sprayed coatings was 8.85%, which increased to 10.14% after immersion for 10 days and then to 12.36% after 30 days. Corrosion products were observed in Al₂O₃-30 coatings. The bond strength of Ca₂SiO₄ coating and Al₂O₃ coating after being immersed in 3.5 wt % NaCl solution for 12 days were 25.0 ± 3.5 MPa and 20.5 ± 3.5 MPa, respectively. The bond strength of C₂S-12 and Al₂O₃-12 were lower than that in as-sprayed Ca₂SiO₄ coatings and Al₂O₃ coating.

From Figure 9 and Figure 12, it is concluded that Ca₂SiO₄ coating becomes denser, the micro cracks and pores are filled with hydration products, such as C–S–H gel and CaCO₃. The dense internal structure is beneficial in isolating the metal substrate from the corrosive medium. Al₂O₃ coating becomes loose and porous after being immersed for 30 days. Corrosion medium attacked the metal substrate under Al₂O₃ coating.

3.3. Potentiodynamic Polarization

Figure 13 shows the potentiodynamic polarization curves of Ca₂SiO₄ coatings and Al₂O₃ coatings after being immersed in still 3.5 wt % NaCl solution at (35 ± 2) °C for 0, 5, 10, 20 and 30 days. The corrosion parameters for the tested electrodes are given in

Table 2. Corrosion parameters of Ca₂SiO₄ and Al₂O₃ coatings after being immersed in 3.5 wt % NaCl solution for different time.

Samples	Ecorr (V)	Icorr (μA/cm2)	−βc (V)	βa (V)	Rp (kΩ·cm2)
as-sprayed C2S	−0.37	1.24	0.161	0.376	39.53
C2S-5	−0.44	0.227	0.048	0.370	81.38
C2S-10	−0.62	0.739	0.071	0.797	38.36
C2S-20	−0.64	0.999	0.067	0.688	26.57
C2S-30	−0.65	1.61	0.094	0.957	23.11
as-sprayed Al2O3	−0.36	0.219	0.100	0.086	91.22
Al2O3-5	−0.38	2.06	0.150	0.403	23.08
Al2O3-10	−0.59	9.78	0.282	0.887	9.51
Al2O3-20	−0.71	25.5	0.158	1.186	2.38
Al2O3-30	−0.55	9.13	0.178	0.788	6.91

. I_corr was a key parameter to evaluate the kinetics of the corrosion reaction, which was inversely proportional to the corrosion protection. The polarization resistance (R_p) is calculated as the slope of the E vs. I graph near the E_corr. The polarization resistance (R_p) value of the electrodes was calculated according to the formula:

$$I_{\mathrm{corr}}\left(\mathrm{A}/{\mathrm{cm}}^2\right)=\frac{{\mathsf{\beta }}_{\mathrm{a}}·{\mathsf{\beta }}_{\mathrm{c}}}{2.3\times \left({\mathsf{\beta }}_{\mathrm{a}}+{\mathsf{\beta }}_{\mathrm{c}}\right)·R_{\mathrm{p}}}$$

(4)

The I_corr value of as-sprayed Ca₂SiO₄ coating was 1.24 μA/cm², then decreased to 0.227 μA/cm² after being immersed for 5 days. The I_corr value of C₂S-30 was 1.61 μA/cm². The R_p value of as-sprayed Ca₂SiO₄ coating was 39.53 kΩ·cm², after being immersed, R_p value of C₂S-5 went up to 81.38 kΩ·cm², the R_p value of C₂S-30 was 23.11 kΩ·cm². The I_corr value of as-sprayed Al₂O₃ coating was 0.219 μA/cm². The I_corr of Al₂O₃ coating went up sharply after being immersed, the I_corr value of Al₂O₃-20 was 25.5 μA/cm², the I_corr value of Al₂O₃-30 was 9.13 μA/cm², and it was larger than that of C₂S coating. The R_p value of as-sprayed Al₂O₃ coating was 91.22 kΩ·cm², after being immersed R_p value of Al₂O₃-5 went down sharply to 23.08 kΩ·cm², the R_p value of Al₂O₃-30 was 6.91 kΩ·cm².

When immersed for 30 days, the corrosion current density of Ca₂SiO₄ coatings increased slightly. For Ca₂SiO₄ coating, at the first 20 days, the corrosive medium penetrated to the metal substrate to promote the formation of passive film, which suppressed the corrosion process, but the passive film did not cover the whole metal surface. With the increasing immersion time, more solution arrived at substrate to induce pitting; therefore, the corrosion resistance of C₂S-30 reclined. Polarization curves show the shifting of corrosion potential (E_corr) towards more negative potential with increasing immersed time, indicating that immersed coatings preferentially inhibit the cathodic process.

From Figure 13a, Ca₂SiO₄ coatings show obvious passivation phenomenon. Ca₂SiO₄ coatings released a lot of Ca(OH)₂ into NaCl aqueous solution. The solution was alkaline and the pH was above 8.0, although the dissolution of CO₂ gas in air may decrease the pH value. The alkaline environment was in favor of the formation of passivation film [23]. Therefore, Ca₂SiO₄ coating promoted the formation of passivation film and passivation film could protect the metal from corrosion effectively. Only Al₂O₃-20 showed the passive behavior for Al₂O₃ coating. The passivation film might form partly. On the other hand, pitting corrosion occurred easily when the Al₂O₃ coating was loose. So, the corrosion current density was very large, the R_p was very low. For Al₂O₃-30, the corrosion product possibly blocked up the crack and pores in the coating, the corrosion rate reclined and current became lower; R_p increased to 6.36 kΩ·cm². From the potentiodynamic polarization curves and corrosion parameters of Ca₂SiO₄ and Al₂O₃ coatings, it could be concluded that Ca₂SiO₄ coating was more corrosion-resistant than Al₂O₃ coating.

3.4. Electrochemical Impedance Spectroscopy

Figure 14 shows the EIS plots of Ca₂SiO₄ coatings after immersion. There was a depressed semi-circle at the high frequencies, it indicated that a CPE was placed in parallel to a resistor. The diameter of the semi-circle increased at the initial 10 days, then decreased with the increase of immersion time. At low frequencies, there was a line.

In Figure 14, the points present the actual data recorded by the test system, and the solid lines represent fitting data. Furthermore, the lines fit the actual data very well. The equivalent circuits of Ca₂SiO₄ coating are given in Figure 15. As shown in Figure 15, R_s is the resistance of solution; R_t is the resistance of the charge transfer. The greater the R_t, the better corrosion-resistance. R_pore indicates the protective ability of coating, the higher porosity and larger pores would cause low R_pore; Q_c represents the constant phase element of coating; Q_dl is the double-layer capacitance between substrate and coating interface; Q_p represents the constant phase element of passive film; a capacitor is more suitable for the actual Ca₂SiO₄ coating at the last 30 days, and C_c is the capacitance of coating; W_s is a closed-loop diffusion-related element of Warburg impedance, it is used to represent the finite layer diffusion resistance.

For as-sprayed Ca₂SiO₄ coating, the Nyquist plot could be divided into three parts, two depressed semi-circles at high frequencies and a line at low frequencies. The first small depressed semi-circle was produced as Q_c was placed in parallel to R_pore [24], and the second large depressed semi-circle was produced as Q_dl was placed in parallel to R_t, the line is produced by W_s. The equivalent circuits for as-sprayed Ca₂SiO₄ coating are shown in Figure 15a. After immersion, passivation film was formed on the substrate, the equivalent circuit was changed, only one capacitive reactance arc occurred. For immersed Ca₂SiO₄ coating, as shown in Figure 15b, the high-frequencies capacitive reactance arc presents R_t in parallel with the Q_c; and at low frequencies, the plots show the character of passivation film [24]. In Figure 14b, the diameter of the high-frequencies capacitive reactance loop increased at the initial 10 days, then decreased with the increase of immersion time, it indicated that the R_t became larger firstly and decreased as the corrosion medium penetrated to the internal of coatings.

The corrosion parameters of Ca₂SiO₄ and Al₂O₃ coatings after being immersed in 3.5 wt % NaCl solution for different times are shown in

Table 3. Corrosion parameters of Ca₂SiO₄ and Al₂O₃ coatings after being immersed in 3.5 wt % NaCl solution for different time.

Time (Days)	Ca2SiO4		Al2O3
	Rt (Ω)	Qp-p	Ws-R (Ω)	Qc-p
0	6288	/	32.45	0.726
5	7319	0.865	253.4	0.501
10	7759	0.883	239.3	0.475
20	4550	0.885	222	0.516
30	420.6	0.809	414.0	0.411

. For Ca₂SiO₄ coating, the R_t value of C₂S was 6288 Ω, then it went up after being immersed, R_t value of C₂S-5 was 7319 Ω, R_t value of C₂S-10 was 7759 Ω, but it finally went down to 420.6 Ω. At the first 10 days, the corrosive medium penetrated to the metal substrate to promote the formation of passive film, which suppressed charge transfer, giving rise to the R_t value. The passive film did not cover the whole metal surface, with the increasing immersion time, more solution arrived at substrate to induce pitting, the value of R_t went down [25]. At the last 30 days, the value of R_t was just 420 Ω, which might mean the resistance of solution in the pit. The electrochemical impedance spectroscopies in Figure 14a are lines with different slopes at low frequencies, which represents the great diffusion resistance of corrosive medium. The larger slope means better capacitive character. The slope is presented by the value of Q_p-p. When the value of Q_p-p was closer to 1.0, the passivation film was denser and thicker. From Table 3, the value of Q_p-p went up and was closer to 1.0 after being immersed; it is concluded that the passivation film is more like a pure capacitance to hinder the pass of electronic, indicating better corrosion resistance [25]. From Figure 14c, it was shown that at 0.01 Hz, the values of impedance modulus increased after immersion, which was consistent with the lines length of EIS at low frequencies in Figure 14a. If the constant phase element is a pure capacitance, its phase angle will be 90°, and in Figure 13d, the phase angle went up and was more and more close to 90°, which presents the good capacitance of passive film. The change of phase angle was consistent with the value of Q_p-p.

The EIS results of Al₂O₃ coatings are presented in Figure 16. Figure 17 shows the equivalent circuits of Al₂O₃ coatings for different time. Q_c represents the constant phase element of coating; and W_s is a closed-loop diffusion related element of Warburg impedance. For Al₂O₃ coating, the high-frequencies capacitive reactance arc represents the resistance of W_s as shown in Figure 17. In Figure 16b, the diameter of the high-frequencies capacitive reactance loop was increased with the immersed time indicating that the resistance of W_s enlarged. The W_s value of as-sprayed Al₂O₃ coating was 32.45 Ω, and it went up after being immersed. W_s value of Al₂O₃-5 was 253.4 Ω, W_s value of Al₂O₃-30 was 414.0 Ω. It might be attributed to the fact that corrosion products blocked the diffusion channel in coatings [26]. In Figure 16b, at low frequencies, Nyquist plots of Al₂O₃ coatings also present lines indicating the diffusion resistance of corrosive medium. From Table 3, the value of Q_c-p went down, the Q_c-p value of as-sprayed Al₂O₃ was 0.726, Q_c-p value of Al₂O₃-5 was 0.506, and it finally went down to 0.411. It was concluded that Al₂O₃ coating became loose and porous. The protection performance of Al₂O₃ coating is weakened. In Figure 16d, the phase angle went down with the increasing immersed time, which was consistent with the change of the value of Q_c-p.

4. Conclusions

From the results of the above investigation, it is concluded that Ca₂SiO₄ coating shows a better corrosion resistance than Al₂O₃ coating. Ca₂SiO₄ coatings released Ca(OH)₂ in NaCl solution to promote the formation of calcium carbonate crystal and passivation film. The crystal structure and grain size of calcium carbonate might be affected by the concentration of Ca²⁺ and CO₃²⁻ in the solution, and the low supersaturation of solution made the crystal growth predominant. Therefore, the size of calcium carbonate particles on Ca₂SiO₄ coating was larger than that in other studies. Passivation film would prevent the corrosive medium from metal substrate, and EIS showed that the passive film became denser and thicker with the increase of immersion time, and the passivation film was like a pure capacitance to hinder the pass of electron, indicating better corrosion resistance. The pores and micro cracks in Ca₂SiO₄ coatings were filled with hydration products C–S–H gel to make the coating denser. The corrosion current density of Al₂O₃ coatings went up sharply after being immersed, and the microstructure became looser and porous with obvious rusty spots during the immersion test.