Influence of Ultrasonic Excitation Sealing on the Corrosion Resistance of HVOF-Sprayed Nanostructured WC-CoCr Coatings under Different Corrosive Environments

The corrosion behavior of unsealed and sealed high-velocity oxygen-fuel (HVOF)-sprayed nanostructured WC-CoCr cermet coatings under different corrosive environments was investigated using scanning electron microscopy (SEM), open circuit potential (OCP), potentiodynamic polarization and electrochemical impedance spectroscopy (EIS). Ultrasonic excitation sealing with aluminum phosphate was performed in an external ultrasonic bath with the frequency of 40 kHz at atmospheric pressure and room temperature. SEM micrographs revealed that the exposed area of the coating was effectively reduced by the coverage of aluminum phosphate sealant on the majority of pores. Electrochemical measurements demonstrated that the sealant with the help of ultrasonic energy could shift the corrosion potential to a more noble direction, reduce the corrosion current density, increase the resistance of charge transfer, and effectively improve the corrosion resistance of the coating in both 3.5 wt % NaCl and 1 mol·L−1 HCl solutions.


Introduction
Corrosion is common in different industry segments such as the aerospace, water, biomedical, resources, power, and maritime sectors. The detriments in the mechanical properties or the waste of raw materials are among the main consequences of the corrosion. During the last decades, enormous effort has been made to understand corrosion phenomena and mechanisms, and to elucidate the leading cause that influences the service lifetime of materials, particularly metals [1][2][3]. The properties of metal materials in aggressive environments are critical for a sustainable society and result not only from microstructures and physico-chemical characteristics, but also from the performance of their surfaces [4,5]. On this basis, there is significant ongoing research in this area seeking to improve the surface performance, ranging from development of anti-corrosion materials to selection of surface treatment technologies [6][7][8].
Cermets are a new class of ceramic materials with a relevant scientific and applicative interest as they have a favorable combination of high-temperature stability, exceptional mechanical properties, tolerance to damage under cyclic loadings, and high wear and corrosion resistance in different environments. The anti-corrosion properties of cermets depend upon chemical composition and microstructure, which are the most influential parameters [9][10][11][12]. Thermal spraying technologies are widely adopted industrially to deposit different kinds of coatings (i.e., metallic, intermetallic, ceramic, and cermet) for protecting substrate, which is exposed to extremely difficult operating conditions related with high temperatures, aggressive environments, and mechanical load [13]. Among different preparation methods, high-velocity oxygen-fuel (HVOF) spraying is one of the best methods for preparing carbide cermet coatings due to the suppression of decarburization and dissolution of the carbide grains [14]. Nonetheless, additional improvement in quality of HVOF-sprayed coatings is incessantly sought in order to further enhance the durability and performance, which can be achieved through post-treatments such as sealing, annealing and remelting [15][16][17]. Recently, it has been accepted that the anti-corrosion properties of the coatings can be improved by inorganic sealants with the help of different sealing techniques [18][19][20][21]. Aluminum phosphate sealant has been used not only to prevent the penetration of corrosive media, but also to enhance the cohesion strength of the coating by combining conventional impregnation sealing treatment with ultrasonic excitation [22]. Hence, it is expected that a good corrosion resistance of HVOF-sprayed cermet coatings can be obtained by ultrasonic excitation sealing with aluminum phosphate.
Despite numerous works concerning the conventional impregnation sealing with aluminum phosphate, limited attempts have been made in ultrasonic excitation sealing. To our knowledge, only a few papers reported the effect of aluminum phosphate sealant on the corrosion resistance of HVOF-sprayed coatings [22,23]. However, the investigation of nanostructured cermet coatings is still lacking. Till now, no literature is available concerning the effect of aluminum phosphate sealant on the corrosion behavior of nanostructured WC-CoCr cermet coating under different corrosive environments.
Therefore, the aim of the present work is to implement the aluminum phosphate sealant by ultrasonic excitation sealing to further improve the corrosion resistance of HVOF-sprayed nanostructured WC-CoCr cermet coating. Our previous work [24] reported that the nanostructured WC-CoCr coating exhibited higher corrosion resistance than the hard chromium coating in 3.5 wt % NaCl solution. Here, the effect of aluminum phosphate sealant using ultrasonic excitation sealing on surface morphology and corrosion behavior of the coating in neutral and acid environments was evaluated. The ultrasonic excitation sealing treatment can enhance the corrosion resistance of WC-CoCr cermet coatings in HCl solution, thereby verifying the effectiveness of aluminum phosphate sealant in HCl solution and expanding the application range of the coatings.

Experimental Procedure
A commercial nanostructured WC-CoCr cermet powder (Infralloy-7410, Inframat Advanced Materials Corp., Farmington, CT, USA) with composition of 4.0 wt % Cr, 10.0 wt % Co, 5.3 wt % C, balance W, was used for HVOF spraying in this study. The cermet coating was deposited on AISI 1045 steel substrates with the thickness of approximately 400 µm by using an HVOF spray system (Praxair Tafa-JP8000, Danbury, CT, USA). Table 1 describes the values of the deposition parameters of the HVOF spraying process. Details of the preparation process and microstructures of the cermet coatings have been reported previously [24]. The sealing agent was aluminum phosphate, which was based on the mixture of aluminum hydroxide (Al(OH) 3 ) and orthophosphoric acid (85 wt % H 3 PO 4 ) with the mass ratio of 1:4.2. The mixed solution was dissolved into 20 wt % distilled water for dilution and complete reaction, and then slightly Coatings 2019, 9, 724 3 of 9 heated to 70 • C under magnetic stirring for 1 h. The aluminum phosphate sealant was obtained, of which the molar ratio of Al to P was about 1:3 [19]. Figure 1 shows the schematic diagram of ultrasonic excitation sealing apparatus. The as-sprayed coatings were immersed in the sealant under ultrasonic excitation with the frequency of 40 kHz at atmospheric pressure and room temperature. After immersion for 4 h, the sealed coatings were taken out, heat-treated from room temperature to 100 • C for 2 h, 200 • C for 2 h, and 250 • C for 1 h in sequence, and then cooled down to room temperature slowly inside the furnace [21]. Subsequently, the coating specimens were ground and polished in order to remove excess sealant and ensure surface roughness of the sealed coating for electrochemical corrosion testing.
Coatings 2019, 9,724 3 of 10 then slightly heated to 70 °C under magnetic stirring for 1 h. The aluminum phosphate sealant was obtained, of which the molar ratio of Al to P was about 1:3 [19]. Figure 1 shows the schematic diagram of ultrasonic excitation sealing apparatus. The as-sprayed coatings were immersed in the sealant under ultrasonic excitation with the frequency of 40 kHz at atmospheric pressure and room temperature. After immersion for 4 h, the sealed coatings were taken out, heat-treated from room temperature to 100 °C for 2 h, 200 °C for 2 h, and 250 °C for 1 h in sequence, and then cooled down to room temperature slowly inside the furnace [21]. Subsequently, the coating specimens were ground and polished in order to remove excess sealant and ensure surface roughness of the sealed coating for electrochemical corrosion testing. The surface morphologies and microstructures of the coating specimens with and without ultrasonic excitation sealing treatment were observed by a scanning electron microscope (SEM, Hitachi S-3400N, Tokyo, Japan) with an energy dispersive spectroscopy (EDS, EX250).
The corrosion behavior of the coatings with and without sealing treatment was evaluated by electrochemical measurements including open circuit potential (OCP), potentiodynamic polarization curves, and electrochemical impedance spectroscopy (EIS), with which the kinetics and mechanism of the electrode processes, the changes in corrosion current and corrosion potential, as well as the film formation and destruction can be revealed, respectively [25][26][27]. Electrochemical tests were conducted by Parstat 2273 advanced electrochemical system in a three-electrode cell, where coating specimen, platinum wire and saturated calomel electrode were used as working electrode, counter electrode and reference electrode, respectively. Electrolytes used in the present work were 3.5 wt % NaCl aqueous solution and 1 mol•L −1 HCl solution. The detailed description of preparation of the coating specimens can be found in the previous investigation [24]. EIS tests were performed at OCP with frequency interval from 100 kHz to 10 mHz. The width of the signal amplitude perturbation applied to the system was 10 mV. The EIS data were fitted and interpreted based on equivalent circuits by using the software ZsimpWin version 3.21. Potentiodynamic polarization curves were recorded at 1 mV•s −1 as the OCP became steady. The corrosion current density (icorr) and corrosion potential (Ecorr) were determined by the Tafel extrapolation method based on the intersection point fitting to the anodic and cathodic polarization curves. All the experiments were repeated at least twice to ensure good reliability of the results. Figure 2 shows the surface morphologies of HVOF-sprayed nanostructured WC-CoCr coatings before and after ultrasonic excitation sealing with aluminum phosphate. It can be seen from Figure  2a that there are some open or semi-closed pores on the rough surface of unsealed coating, which is attributed to the successive deposition and solidification process of molten cermet droplets involving incomplete contact between semi-melted or un-melted particles [28]. After ultrasonic excitation sealing, small filamentous substances are dispersed in the open pores on the coating surface, as shown in Figure 2b. The filamentous substance (shown in the inset of Figure 2b) is primarily sealant with the chemical composition of Al6P25O53C3Cr1Co2W10 (wt %). It is illustrated that the exposed area of the coating is effectively reduced by the coverage of aluminum phosphate sealant on the majority The surface morphologies and microstructures of the coating specimens with and without ultrasonic excitation sealing treatment were observed by a scanning electron microscope (SEM, Hitachi S-3400N, Tokyo, Japan) with an energy dispersive spectroscopy (EDS, EX250).

Morphologies of Nanostructured WC-CoCr Cermet Coatings
The corrosion behavior of the coatings with and without sealing treatment was evaluated by electrochemical measurements including open circuit potential (OCP), potentiodynamic polarization curves, and electrochemical impedance spectroscopy (EIS), with which the kinetics and mechanism of the electrode processes, the changes in corrosion current and corrosion potential, as well as the film formation and destruction can be revealed, respectively [25][26][27]. Electrochemical tests were conducted by Parstat 2273 advanced electrochemical system in a three-electrode cell, where coating specimen, platinum wire and saturated calomel electrode were used as working electrode, counter electrode and reference electrode, respectively. Electrolytes used in the present work were 3.5 wt % NaCl aqueous solution and 1 mol·L −1 HCl solution. The detailed description of preparation of the coating specimens can be found in the previous investigation [24]. EIS tests were performed at OCP with frequency interval from 100 kHz to 10 mHz. The width of the signal amplitude perturbation applied to the system was 10 mV. The EIS data were fitted and interpreted based on equivalent circuits by using the software ZsimpWin version 3.21. Potentiodynamic polarization curves were recorded at 1 mV·s −1 as the OCP became steady. The corrosion current density (i corr ) and corrosion potential (E corr ) were determined by the Tafel extrapolation method based on the intersection point fitting to the anodic and cathodic polarization curves. All the experiments were repeated at least twice to ensure good reliability of the results. Figure 2 shows the surface morphologies of HVOF-sprayed nanostructured WC-CoCr coatings before and after ultrasonic excitation sealing with aluminum phosphate. It can be seen from Figure 2a that there are some open or semi-closed pores on the rough surface of unsealed coating, which is attributed to the successive deposition and solidification process of molten cermet droplets involving incomplete contact between semi-melted or un-melted particles [28]. After ultrasonic excitation sealing, small filamentous substances are dispersed in the open pores on the coating surface, as shown in Figure 2b. The filamentous substance (shown in the inset of Figure 2b) is primarily sealant with the chemical composition of Al 6 P 25 O 53 C 3 Cr 1 Co 2 W 10 (wt %). It is illustrated that the exposed area of the coating is effectively reduced by the coverage of aluminum phosphate sealant on the majority of pores, which is expected to enhance the corrosion resistance of as-sprayed coating. Liu et al. [22] proposed that ultrasonic excitation sealing would promote the penetration of the sealant and block the microdefects within the coating. Hence, the ultrasonic energy is not only beneficial for the escape of air within the pores and the sealant, but it will also guarantee a degree of compatibility between the sealant and the coating. of pores, which is expected to enhance the corrosion resistance of as-sprayed coating. Liu et al. [22] proposed that ultrasonic excitation sealing would promote the penetration of the sealant and block the microdefects within the coating. Hence, the ultrasonic energy is not only beneficial for the escape of air within the pores and the sealant, but it will also guarantee a degree of compatibility between the sealant and the coating.  Figure 3 presents the OCP curves of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol•L −1 HCl solutions. From the changes in the potential values of both unsealed and sealed coatings, it can be seen that there is an obvious reduction at the first few minutes, suggesting that the penetration of the electrolyte and the dissolution of the oxides on the coating surface may occur in different degrees [29]. After 1 h, the OCP values of unsealed and sealed WC-CoCr coatings in 3.5 wt % NaCl solution are −438 and −249 mV, respectively. In contrast, unsealed and sealed WC-CoCr coatings have OCP values of −376 and −12.9 mV in 1 mol•L −1 HCl solution, respectively. The lower corrosion tendency of the sealed coatings is considered to be mainly determined by the character of forming a compact microstructure without microdefects that are sensitive to electrolyte absorption [30]. Moreover, there is significant difference in the response of unsealed and sealed WC-CoCr coatings to the corrosion medium, where the increase of the OCP value in 1 mol•L −1 HCl solution (Figure 3b) is greater than that in 3.5 wt % NaCl solution (Figure 3a). This reflects that the aluminum phosphate sealant with the help of ultrasonic energy primarily prevents the evolution of corrosion of WC-CoCr coating and provides superior protection in acid solution.   Figure 3 presents the OCP curves of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol·L −1 HCl solutions. From the changes in the potential values of both unsealed and sealed coatings, it can be seen that there is an obvious reduction at the first few minutes, suggesting that the penetration of the electrolyte and the dissolution of the oxides on the coating surface may occur in different degrees [29]. After 1 h, the OCP values of unsealed and sealed WC-CoCr coatings in 3.5 wt % NaCl solution are −438 and −249 mV, respectively. In contrast, unsealed and sealed WC-CoCr coatings have OCP values of −376 and −12.9 mV in 1 mol·L −1 HCl solution, respectively. The lower corrosion tendency of the sealed coatings is considered to be mainly determined by the character of forming a compact microstructure without microdefects that are sensitive to electrolyte absorption [30]. Moreover, there is significant difference in the response of unsealed and sealed WC-CoCr coatings to the corrosion medium, where the increase of the OCP value in 1 mol·L −1 HCl solution (Figure 3b) is greater than that in 3.5 wt % NaCl solution (Figure 3a). This reflects that the aluminum phosphate sealant with the help of ultrasonic energy primarily prevents the evolution of corrosion of WC-CoCr coating and provides superior protection in acid solution.

Open Circuit Potential (OCP)
Coatings 2019, 9,724 4 of 10 of pores, which is expected to enhance the corrosion resistance of as-sprayed coating. Liu et al. [22] proposed that ultrasonic excitation sealing would promote the penetration of the sealant and block the microdefects within the coating. Hence, the ultrasonic energy is not only beneficial for the escape of air within the pores and the sealant, but it will also guarantee a degree of compatibility between the sealant and the coating.  Figure 3 presents the OCP curves of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol•L −1 HCl solutions. From the changes in the potential values of both unsealed and sealed coatings, it can be seen that there is an obvious reduction at the first few minutes, suggesting that the penetration of the electrolyte and the dissolution of the oxides on the coating surface may occur in different degrees [29]. After 1 h, the OCP values of unsealed and sealed WC-CoCr coatings in 3.5 wt % NaCl solution are −438 and −249 mV, respectively. In contrast, unsealed and sealed WC-CoCr coatings have OCP values of −376 and −12.9 mV in 1 mol•L −1 HCl solution, respectively. The lower corrosion tendency of the sealed coatings is considered to be mainly determined by the character of forming a compact microstructure without microdefects that are sensitive to electrolyte absorption [30]. Moreover, there is significant difference in the response of unsealed and sealed WC-CoCr coatings to the corrosion medium, where the increase of the OCP value in 1 mol•L −1 HCl solution (Figure 3b) is greater than that in 3.5 wt % NaCl solution (Figure 3a). This reflects that the aluminum phosphate sealant with the help of ultrasonic energy primarily prevents the evolution of corrosion of WC-CoCr coating and provides superior protection in acid solution.   corrosion parameters derived from the curves, including corrosion potential (E corr ) and corrosion current density (i corr ) are listed in Table 2. For unsealed coatings, the E corr values in 3.5 wt % NaCl and 1 mol·L −1 HCl solutions are found to be −312 and −395 mV, respectively. In the cases of sealed coatings, the E corr values are shifted to −247 and −52 mV, respectively. In general, E corr mainly describes the thermodynamic property and a small value of E corr correlates to high corrosion tendency [31]. It is clear that all sealed coatings have higher E corr values than unsealed coatings, which is due to the effective barrier performance of aluminum phosphate sealant on the coating surface leading to the decrease of electrolyte permeation towards the steel substrate. Likewise, ultrasonic excitation sealing with aluminum phosphate lowers the i corr values of unsealed coatings in both 3.5 wt % NaCl and 1 mol·L −1 HCl solutions. However, ultrasonic excitation sealing shows a much higher impact on the corrosion resistance of the coatings in 1 mol·L −1 HCl solution (Figure 4b) compared to that in 3.5 wt % NaCl solution (Figure 4a). For 1 mol·L −1 HCl solution, the i corr value decreases from 43.75 µA·cm −2 for unsealed coating to 4.12 µA·cm −2 for sealed coating and the reduction is by more than one order of magnitude, indicating an obvious enhancement of kinetic resistance to corrosion. This can be attributed to the superior protection behavior and the stability of aluminum phosphate sealant in HCl solution, although HCl solution is particularly aggressive as a result of the continuous attack of chloride ions and the removal of oxide films [21,32]. As it can be seen, ultrasonic excitation sealing has changed both the anodic and cathodic slopes of the polarization curves and, in turn, the corrosion rate of unsealed coating. Furthermore, it can be found that there are obvious passivation phenomena for both unsealed and sealed coatings in 1 mol·L −1 HCl solution, whilst there is no passivation process in 3.5 wt % NaCl solution. It can be speculated that pitting corrosion plays an important role in 1 mol·L −1 HCl solution while uniform corrosion dominates the corrosion process in 3.5 wt % NaCl solution. Unsealed coating has a passivation current density of 312.5 µA·cm −2 , a passivation interval width of 146 mV and a pitting potential of about −123 mV. As regards sealed coating, the passivation current density decreases to 96.1 µA·cm −2 , and the pitting potential approaches 252 mV, which is related to the fewer microdefects on the surface of sealed coating, as indicated in Figure 1. These results indicate that an appropriate sealing treatment would increase the corrosion resistance of the coatings.  Figure 4 displays the potentiodynamic polarization curves of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol•L −1 HCl solutions. The corresponding electrochemical corrosion parameters derived from the curves, including corrosion potential (Ecorr) and corrosion current density (icorr) are listed in Table 2. For unsealed coatings, the Ecorr values in 3.5 wt % NaCl and 1 mol•L −1 HCl solutions are found to be −312 and −395 mV, respectively. In the cases of sealed coatings, the Ecorr values are shifted to −247 and −52 mV, respectively. In general, Ecorr mainly describes the thermodynamic property and a small value of Ecorr correlates to high corrosion tendency [31]. It is clear that all sealed coatings have higher Ecorr values than unsealed coatings, which is due to the effective barrier performance of aluminum phosphate sealant on the coating surface leading to the decrease of electrolyte permeation towards the steel substrate. Likewise, ultrasonic excitation sealing with aluminum phosphate lowers the icorr values of unsealed coatings in both 3.5 wt % NaCl and 1 mol•L −1 HCl solutions. However, ultrasonic excitation sealing shows a much higher impact on the corrosion resistance of the coatings in 1 mol•L −1 HCl solution (Figure 4b) compared to that in 3.5 wt % NaCl solution (Figure 4a). For 1 mol•L −1 HCl solution, the icorr value decreases from 43.75 μA•cm −2 for unsealed coating to 4.12 μA•cm −2 for sealed coating and the reduction is by more than one order of magnitude, indicating an obvious enhancement of kinetic resistance to corrosion. This can be attributed to the superior protection behavior and the stability of aluminum phosphate sealant in HCl solution, although HCl solution is particularly aggressive as a result of the continuous attack of chloride ions and the removal of oxide films [21,32]. As it can be seen, ultrasonic excitation sealing has changed both the anodic and cathodic slopes of the polarization curves and, in turn, the corrosion rate of unsealed coating. Furthermore, it can be found that there are obvious passivation phenomena for both unsealed and sealed coatings in 1 mol•L −1 HCl solution, whilst there is no passivation process in 3.5 wt % NaCl solution. It can be speculated that pitting corrosion plays an important role in 1 mol•L −1 HCl solution while uniform corrosion dominates the corrosion process in 3.5 wt % NaCl solution. Unsealed coating has a passivation current density of 312.5 μA•cm −2 , a passivation interval width of 146 mV and a pitting potential of about −123 mV. As regards sealed coating, the passivation current density decreases to 96.1 μA•cm −2 , and the pitting potential approaches 252 mV, which is related to the fewer microdefects on the surface of sealed coating, as indicated in Figure 1. These results indicate that an appropriate sealing treatment would increase the corrosion resistance of the coatings.

Electrochemical Impedance Spectroscopy (EIS)
EIS studies were performed in order to characterize the complex electrochemical processes involved. Figure 5 exhibits the Nyquist plots and the corresponding Bode plots of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol·L −1 HCl solutions. As shown in Figure 5a,c, the capacitive arc for unsealed coatings appears to be a semicircle, while the arc shape of capacitive reactance for sealed coatings is a flattened arc, suggesting different features of the protective films. In addition, the radius of the capacitive arc for sealed coatings is larger than that for unsealed coatings in both 3.5 wt % NaCl and 1 mol·L −1 HCl solutions. It is well recognized that the larger the radius of capacitive reactance arc is, the greater the resistance of the coating is [33]. Thus, ultrasonic excitation sealing with aluminum phosphate can effectively protect the unsealed coating and the substrate from corrosion damage. From Figure 5b, it can be found that the impedance modulus values of the sealed coating are higher than those of the unsealed coating over the full frequency range in 3.5 wt % NaCl solution, although the impedance magnitude of both coatings is the same. While as for 1 mol·L −1 HCl solution is concerned, the impedance modulus value of sealed coating at low frequency is significantly higher than that of unsealed coating by one order of magnitude, as shown in Figure 5d. From the frequency-phase-angle diagrams, it can be noticed that there are higher phase angle values for sealed coating in medium frequencies as compared to those of unsealed coating, confirming the presence of more protective film under the treatment of ultrasonic excitation sealing. The above factors demonstrate that the sealants are relatively stable in HCl solution and are able to provide an effective corrosion protection for unsealed coating with the help of ultrasonic energy, which is consistent with the results of the potentiodynamic polarization tests. Especially, the i corr value of the sealed coating is 4.12 µA·cm −2 in 1 mol·L −1 HCl solution, which is close to value of 5 µA·cm −2 for 316L stainless steel previously reported in [34]. This also indicates that sealed coating exhibits a comparable corrosion resistance in HCl solution as compared to 316L stainless steel, a well-known material with high corrosion resistance. Table 2. Corrosion current densities (icorr) and potentials (Ecorr) obtained from the potentiodynamic polarization curves of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol•L −1 HCl.

Electrochemical Impedance Spectroscopy (EIS)
EIS studies were performed in order to characterize the complex electrochemical processes involved. Figure 5 exhibits the Nyquist plots and the corresponding Bode plots of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol•L −1 HCl solutions. As shown in Figure 5a,c, the capacitive arc for unsealed coatings appears to be a semicircle, while the arc shape of capacitive reactance for sealed coatings is a flattened arc, suggesting different features of the protective films. In addition, the radius of the capacitive arc for sealed coatings is larger than that for unsealed coatings in both 3.5 wt % NaCl and 1 mol•L −1 HCl solutions. It is well recognized that the larger the radius of capacitive reactance arc is, the greater the resistance of the coating is [33]. Thus, ultrasonic excitation sealing with aluminum phosphate can effectively protect the unsealed coating and the substrate from corrosion damage. From Figure 5b, it can be found that the impedance modulus values of the sealed coating are higher than those of the unsealed coating over the full frequency range in 3.5 wt % NaCl solution, although the impedance magnitude of both coatings is the same. While as for 1 mol•L −1 HCl solution is concerned, the impedance modulus value of sealed coating at low frequency is significantly higher than that of unsealed coating by one order of magnitude, as shown in Figure 5d. From the frequency-phase-angle diagrams, it can be noticed that there are higher phase angle values for sealed coating in medium frequencies as compared to those of unsealed coating, confirming the presence of more protective film under the treatment of ultrasonic excitation sealing. The above factors demonstrate that the sealants are relatively stable in HCl solution and are able to provide an effective corrosion protection for unsealed coating with the help of ultrasonic energy, which is consistent with the results of the potentiodynamic polarization tests. Especially, the icorr value of the sealed coating is 4.12 μA•cm −2 in 1 mol•L −1 HCl solution, which is close to value of 5 μA•cm −2 for 316L stainless steel previously reported in [34]. This also indicates that sealed coating exhibits a comparable corrosion resistance in HCl solution as compared to 316L stainless steel, a well-known material with high corrosion resistance. In order to simulate the EIS data, the electrochemical processes of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol•L −1 HCl solutions are modeled by equivalent circuits utilizing ZSimpWin commercial software (USA) as shown in Figure 6, where the chi-squared (χ 2 ) value reaches the level of 10 −3 . According to the characteristics of Nyquist plots of the unsealed coating in both 3.5 wt % NaCl and 1 mol•L −1 HCl solutions and the sealed coating in 3.5 wt % NaCl solution, the equivalent circuit with two time constants (Figure 6a) is used to fit the EIS data, where Rs, Rp, Rct, Qc, and Qdl represent the resistance of electrolyte solution, the resistance of In order to simulate the EIS data, the electrochemical processes of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol·L −1 HCl solutions are modeled Coatings 2019, 9, 724 7 of 9 by equivalent circuits utilizing ZSimpWin commercial software (USA) as shown in Figure 6, where the chi-squared (χ 2 ) value reaches the level of 10 −3 . According to the characteristics of Nyquist plots of the unsealed coating in both 3.5 wt % NaCl and 1 mol·L −1 HCl solutions and the sealed coating in 3.5 wt % NaCl solution, the equivalent circuit with two time constants (Figure 6a) is used to fit the EIS data, where R s , R p , R ct , Q c , and Q dl represent the resistance of electrolyte solution, the resistance of pores, the resistance of charge transfer, the capacitance of the coating, and the capacitance of the double layer, respectively. The constant phase element (CPE or Q) instead of the capacitance is related to the nonuniformity of surface and diffusion factors [35], which can be described as Z Q = Y 0 −1 (jω) −n (0 < n < 1). As for sealed coating in 1 mol·L −1 HCl solution, an additional element, Warburg impedance (Z w ), appears in the equivalent circuit as presented in Figure 6b, which corresponds to the accumulation of corrosive products in pores and the diffusion-controlled mechanism [36,37]. The EIS fitting results are summarized in Table 3. It can be seen that R p and R ct values of sealed coatings are higher than those of unsealed coatings, while the opposite trend is observed for Q dl values of the coatings. Especially, the Q dl values of sealed coatings are one order of magnitude lower than those of unsealed coatings. Higher R p and R ct values refer to sealed coatings with less porosity and weaker charge exchange, respectively. It can be explained that the sealant covers the coating well and reduces open pores and cracks on the surface of the coatings. The lower the Q dl value is, the thicker the protective film will be [38]. It is proved further that the content of protective films on sealed coating is maintained at a relatively high level in both 3.5 wt % NaCl and 1 mol·L −1 HCl solutions, demonstrating that the protective effect on steel substrate and the corrosion resistance of the coating are improved effectively by ultrasonic excitation sealing. In order to simulate the EIS data, the electrochemical processes of unsealed and sealed nanostructured WC-CoCr coatings in 3.5 wt % NaCl and 1 mol•L −1 HCl solutions are modeled by equivalent circuits utilizing ZSimpWin commercial software (USA) as shown in Figure 6, where the chi-squared (χ 2 ) value reaches the level of 10 −3 . According to the characteristics of Nyquist plots of the unsealed coating in both 3.5 wt % NaCl and 1 mol•L −1 HCl solutions and the sealed coating in 3.5 wt % NaCl solution, the equivalent circuit with two time constants (Figure 6a) is used to fit the EIS data, where Rs, Rp, Rct, Qc, and Qdl represent the resistance of electrolyte solution, the resistance of pores, the resistance of charge transfer, the capacitance of the coating, and the capacitance of the double layer, respectively. The constant phase element (CPE or Q) instead of the capacitance is related to the nonuniformity of surface and diffusion factors [35], which can be described as ZQ = Y0 −1 (jω) −n (0 < n < 1). As for sealed coating in 1 mol•L −1 HCl solution, an additional element, Warburg impedance (Zw), appears in the equivalent circuit as presented in Figure 6b, which corresponds to the accumulation of corrosive products in pores and the diffusion-controlled mechanism [36,37]. The EIS fitting results are summarized in Table 3. It can be seen that Rp and Rct values of sealed coatings are higher than those of unsealed coatings, while the opposite trend is observed for Qdl values of the coatings. Especially, the Qdl values of sealed coatings are one order of magnitude lower than those of unsealed coatings. Higher Rp and Rct values refer to sealed coatings with less porosity and weaker charge exchange, respectively. It can be explained that the sealant covers the coating well and reduces open pores and cracks on the surface of the coatings. The lower the Qdl value is, the thicker the protective film will be [38]. It is proved further that the content of protective films on sealed coating is maintained at a relatively high level in both 3.5 wt % NaCl and 1 mol•L −1 HCl solutions, demonstrating that the protective effect on steel substrate and the corrosion resistance of the coating are improved effectively by ultrasonic excitation sealing.

Conclusions
The corrosion behavior of HVOF-sprayed nanostructured WC-CoCr cermet coating was studied in different corrosive environments before and after ultrasonic excitation sealing with aluminum phosphate. Compared with unsealed coating, sealed coating exhibited fewer microdefects, lower corrosion current density, higher corrosion potential and resistance of charge transfer, and then provided notable improvement of corrosion resistance in both 3.5 wt % NaCl and 1 mol·L −1 HCl solutions. Aluminum phosphate sealant was found to decrease the corrosion rate of WC-CoCr coating by more than one order of magnitude in 1 mol·L −1 HCl solution with the help of ultrasonic energy, demonstrating that the sealants are relatively stable in HCl solution and can become a candidate for WC-CoCr cermet coating used in an acidic environment.