Effects of Ultrasonic-Aided Quenching on the Corrosion Resistance of GB 35 CrMoV Steel in Seawater Environment

The effects of ultrasonic-aided quenching (USQ) on the microstructure and corrosion properties of 35CrMoV steel in a simulated seawater environment (wt % (NaCl) = 3.5) were studied. The corrosion properties of the steel were studied by electrochemical impedance spectroscopy (EIS) and potentiodynamic polarization curves. The scanning electron microscope (SEM) and energy dispersive spectrometer (EDS) were used to characterize the microstructure and corrosion products. Combined with the EIS equivalent circuit, the corrosion mechanism was investigated. The results show that the microstructure of the USQ specimen is small martensite and that there was no residual ferrite compared with an un-ultrasonic-aided quenching (UUSQ) sample. It was also found that the corrosion products of USQ sample are less than the UUSQ sample, and the corrosion cracks are lighter and thinner than that of the UUSQ sample, which indicates that the corrosion resistance of 35CrMoV in a seawater environment has been obviously improved.


Introduction
With the development of science and technology, developing and utilizing marine resources has become an integral part of the national economy, and people have gained much more interest in the use of the ocean than before.However, in the marine environment, steel has a limited service life and high maintenance costs due to corrosion [1][2][3].According to the existing data, the corrosion protection cost of metal structures in the general environment covers 2-4% of the total cost, while it accounts for 10-30% in the marine environment.For a 500,000-ton ship, the consumption of steel is as high as 40 tons a year, so protection against corrosion is significant to the development of marine resources.
People are paying more attention than before to ultrasound for its convenience and environmental friendliness.Although the mechanism of its action is not entirely clear until now, many studies show that the technology has wide application prospects.Ultrasonic nanocrystal surface modification (UNSM) has already been used as a treatment for many materials [4][5][6][7].The grain size [4], the properties of wear [5] and fatigue [6] of materials can be markedly improved through UNSM, as it produces surface hardness and compressive residual stress.Ultrasonic welding has also received much attention as an alternative joining technique [7,8].Barbosa et al. investigated that ultrasonic casting can refine grain size and reduce the porosity shrinkage of the sample [9].Kessler et al. examined that ultrasonic quenching can improve the mechanical properties of materials by destabilizing the vapor film [10].However, only a small part of the recent studies have paid attention to the corrosion behavior of steels with ultrasonic treatment.Mordyuk et al. determined that the corrosion resistance can be markedly improved by ultrasonic peening [11].
35CrMoV (Chinese standard) steel belongs to CrMo steel series, a typical mid-carbon steel, which is widely used in the exploration of oceans due to its high static strength, high fatigue limit, and good creep strength.Moreover, its price is far lower than the chromium-nickel steel with the same mechanical properties which is commonly used to replace the large section gear and shaft.
Therefore, the purpose of this paper was to investigate the effects of ultrasonic quenching on the corrosion behavior of 35CrMoV steel in the marine environment.The study will be beneficial for the quenching process of 35CrMoV steel in the seawater environment.

Materials
The composition of the 35CrMoV steel is shown in Table 1.Samples were heated up to 850 • C and maintained for 0.5 h with a high-temperature intelligent box furnace (SX2-4-10, Yuandong, Changsha, China), followed by water cooling with and without ultrasonic vibration (Figure 1).The samples were cut into 10 × 10 × 20 mm cubes with a wire cutter machine (DK7735, Guofu, Taizhou, China).The cutting samples were fixed in PC pipes and exposed to an area of 10 × 10 mm for testing.by destabilizing the vapor film [10].However, only a small part of the recent studies have paid attention to the corrosion behavior of steels with ultrasonic treatment.Mordyuk et al. determined that the corrosion resistance can be markedly improved by ultrasonic peening [11].35CrMoV (Chinese standard) steel belongs to CrMo steel series, a typical mid-carbon steel, which is widely used in the exploration of oceans due to its high static strength, high fatigue limit, and good creep strength.Moreover, its price is far lower than the chromium-nickel steel with the same mechanical properties which is commonly used to replace the large section gear and shaft.
Therefore, the purpose of this paper was to investigate the effects of ultrasonic quenching on the corrosion behavior of 35CrMoV steel in the marine environment.The study will be beneficial for the quenching process of 35CrMoV steel in the seawater environment.

Materials
The composition of the 35CrMoV steel is shown in Table 1.Samples were heated up to 850 °C and maintained for 0.5 h with a high-temperature intelligent box furnace (SX2-4-10, Yuandong, Changsha, China), followed by water cooling with and without ultrasonic vibration (Figure 1).The samples were cut into 10 × 10 × 20 mm cubes with a wire cutter machine (DK7735, Guofu, Taizhou, China).The cutting samples were fixed in PC pipes and exposed to an area of 10 × 10 mm for testing.

Microscopic Analysis
The specimens were wet ground in the sequence of 120, 320, 600, and 1000 grit emery papers and polished with 1 μm diamond suspensions, degreased in ethyl alcohol, rinsed in distilled water, and dried with a stream of hot air.They were then immersed into a nital etchant (2 mL of 70% nitric acid and 48 mL of anhydrous, denatured ethyl alcohol), treated with alcohol swapping, and dried in an air stream.The scanning electron microscope (EVO MA10, ZEISS, Jena, Germany) was used to observe the microstructure of the samples.

Electrochemical Measurements
The electrochemical experiment was investigated using the electrochemical workstation (Chi660e, Chenhua, Shanghai, China), a traditional three-electrode system (Figure 2).Prior to the tests, samples were wet-ground up to 1000 grit finish and polished with 1 μm diamond suspensions,

Microscopic Analysis
The specimens were wet ground in the sequence of 120, 320, 600, and 1000 grit emery papers and polished with 1 µm diamond suspensions, degreased in ethyl alcohol, rinsed in distilled water, and dried with a stream of hot air.They were then immersed into a nital etchant (2 mL of 70% nitric acid and 48 mL of anhydrous, denatured ethyl alcohol), treated with alcohol swapping, and dried in an air stream.The scanning electron microscope (EVO MA10, ZEISS, Jena, Germany) was used to observe the microstructure of the samples.

Electrochemical Measurements
The electrochemical experiment was investigated using the electrochemical workstation (Chi660e, Chenhua, Shanghai, China), a traditional three-electrode system (Figure 2).Prior to the tests, samples were wet-ground up to 1000 grit finish and polished with 1 µm diamond suspensions, treated with alcohol swapping, and dried in an air stream.A saturated calomel (Ag/AgCl saturated with KCl) electrode was used as a reference electrode and a platinum piece as a counter electrode.The 35CrMoV samples were used as the working electrode placed in the simulated seawater solution (wt % (NaCl) = 3.5).

Corrosion Surface Analysis
The corrosion surfaces of the samples were rinsed with deionized water and ethanol solution and then dried in an air stream.The corrosion morphology, microstructure, and composition of the corrosion products were observed by Phenom scanning electron microscope.

Microstructure
As can be seen from Figure 3, the microstructure of the two samples was quite different.The UUSQ sample consisted of a mixture of martensite and residual ferrite, while the USQ sample was all small martensite.This is because 35CrMoV steel belongs to mid-carbon steel.Therefore, if the specimens wanted to cool from the complete austenite temperature to room temperature without high-temperature transformation, the quenching medium must have a higher quenching intensity.Although the water had a certain quenching intensity, some local high-temperature changes still existed during the quenching process.When the ultrasonic wave was added, the hot steam film on the surface of the specimen was ruptured in advance due to its cavitation and sound effect [12].Therefore, the microstructure of the USQ sample was all small martensite, while the microstructure of the UUSQ sample was bulky martensite and a small amount of residual ferrite.Before commencing the experiments, the open circuit potential test was applied for 2400 s.The potentiodynamic polarization was scanned at 0.5 mV/s.The electrochemical impedance spectroscopy (EIS) measurements were carried out after 2400 s, at the open circuit potentials, with a range of frequency from 0.1 Hz to 100,000 Hz with an amplitude of 5 mV and a sampling rate of 12 points per decade.

Corrosion Surface Analysis
The corrosion surfaces of the samples were rinsed with deionized water and ethanol solution and then dried in an air stream.The corrosion morphology, microstructure, and composition of the corrosion products were observed by Phenom scanning electron microscope.

Microstructure
As can be seen from Figure 3, the microstructure of the two samples was quite different.The UUSQ sample consisted of a mixture of martensite and residual ferrite, while the USQ sample was all small martensite.This is because 35CrMoV steel belongs to mid-carbon steel.Therefore, if the specimens wanted to cool from the complete austenite temperature to room temperature without high-temperature transformation, the quenching medium must have a higher quenching intensity.Although the water had a certain quenching intensity, some local high-temperature changes still existed during the quenching process.When the ultrasonic wave was added, the hot steam film on the surface of the specimen was ruptured in advance due to its cavitation and sound effect [12].Therefore, the microstructure of the USQ sample was all small martensite, while the microstructure of the UUSQ sample was bulky martensite and a small amount of residual ferrite.high-temperature transformation, the quenching medium must have a higher quenching intensity.Although the water had a certain quenching intensity, some local high-temperature changes still existed during the quenching process.When the ultrasonic wave was added, the hot steam film on the surface of the specimen was ruptured in advance due to its cavitation and sound effect [12].Therefore, the microstructure of the USQ sample was all small martensite, while the microstructure of the UUSQ sample was bulky martensite and a small amount of residual ferrite.

Electrochemical Impedance Spectroscopy
Figure 4 shows the electrochemical impedance spectroscopy (EIS) results of 35CrMoV in 3.5% NaCl solution.It can be seen from the diagram that the electrochemical impedance spectrum of the two samples was similar in morphology, that is, only an arc of resistance was shown in the entire frequency range of the test, indicating that the electrode reaction is the interface reaction process.Additionally, the semicircular diameter of the USQ sample was significantly larger than that of the UUSQ sample.The semicircular diameter reflects the corrosion resistance of the material, and they are positively related [13].Thus, the Nyquist curves indicated that the corrosion resistance of the ultrasonic quenching sample was significantly improved.

Electrochemical Impedance Spectroscopy
Figure 4 shows the electrochemical impedance spectroscopy (EIS) results of 35CrMoV in 3.5% NaCl solution.It can be seen from the diagram that the electrochemical impedance spectrum of the two samples was similar in morphology, that is, only an arc of resistance was shown in the entire frequency range of the test, indicating that the electrode reaction is the interface reaction process.Additionally, the semicircular diameter of the USQ sample was significantly larger than that of the UUSQ sample.The semicircular diameter reflects the corrosion resistance of the material, and they are positively related [13].Thus, the Nyquist curves indicated that the corrosion resistance of the ultrasonic quenching sample was significantly improved.
The equivalent circuit model is shown in Figure 5.All the spectra can be well described by the equivalent circuit.This model consists of solution resistance (Rs), charge-transfer resistance (Rct), corrosion products resistance (Rf), and constant phase element (CPE), Q. CPE1 and CPE2 represent double-layer capacitance of the reaction interface and corrosion products capacitance, respectively.CPE is the constant phase element which contains two parameters [13][14][15] that can be expressed as: where ω = 2πf and f is the frequency in units of Hz.The value of CPE parameter, n varies from 0 to 1.
When n = 1, the CPE behaves as a pure capacitor, and when n = 0, the CPE behaves as a pure resistor.This electrochemical element is best suited to characterize the corrosion mechanism that occurs at the solution or metal interface.The fitting results are shown in Table 2.It was found that the corrosion resistance (Rp = Rf + Rct) of the USQ sample was 1513.613Ω•cm 2 , three times as much as 498.88 Ω•cm 2 of the UUSQ sample.This showed that the ultrasonic quenching could obviously improve the corrosion resistance of the sample in the seawater environment, which was consistent with the result of the impedance curve.The equivalent circuit model is shown in Figure 5.All the spectra can be well described by the equivalent circuit.This model consists of solution resistance (R s ), charge-transfer resistance (R ct ), corrosion products resistance (R f ), and constant phase element (CPE), Q. CPE1 and CPE2 represent double-layer capacitance of the reaction interface and corrosion products capacitance, respectively.CPE is the constant phase element which contains two parameters [13][14][15] that can be expressed as: where ω = 2πf and f is the frequency in units of Hz.The value of CPE parameter, n varies from 0 to 1.When n = 1, the CPE behaves as a pure capacitor, and when n = 0, the CPE behaves as a pure resistor.This electrochemical element is best suited to characterize the corrosion mechanism that occurs at the solution or metal interface.The fitting results are shown in Table 2.It was found that the corrosion resistance (R p = R f + R ct ) of the USQ sample was 1513.613Ω•cm 2 , three times as much as 498.88 Ω•cm 2 of the UUSQ sample.This showed that the ultrasonic quenching could obviously improve the corrosion resistance of the sample in the seawater environment, which was consistent with the result of the impedance curve.
resistance (Rp = Rf + Rct) of the USQ sample was 1513.613Ω•cm 2 , three times as much as 498.88 Ω•cm 2 of the UUSQ sample.This showed that the ultrasonic quenching could obviously improve the corrosion resistance of the sample in the seawater environment, which was consistent with the result of the impedance curve.Table 2. Fitting parameters calculated by an equivalent circuit.

Potentiodynamic Polarization Curve
As can be seen from Figure 6, the cathodic currents of the USQ sample and the UUSQ sample, dominantly representing oxygen reduction at -1.0 V were 0.25 mA/cm 2 and 0.61 mA/cm 2 , respectively.The reduction process can be described as [16] The polarization curves of the two samples were similar, and there were no passivation areas.This was because Cl − was adsorbed on the surface of the sample and hindered the formation of the passivation film.Combined with Table 3, the corrosion potential of the USQ sample was −651 mV, and the corrosion current density was 6.9 × 10 -6 A•cm −2 .Compared with the UUSQ sample, the corrosion potential was increased by 22 mV, and the corrosion current density decreased by several orders of magnitude.According to the electrochemical theory, the corrosion properties of the USQ sample have been improved.
This was because Cl was adsorbed on the surface of the sample and hindered the formation of the passivation film.Combined with Table 3, the corrosion potential of the USQ sample was -651 mV, and the corrosion current density was 6.9 × 10 -6 A•cm −2 .Compared with the UUSQ sample, the corrosion potential was increased by 22 mV, and the corrosion current density decreased by several orders of magnitude.According to the electrochemical theory, the corrosion properties of the USQ sample have been improved.

Corrosion Products and Corrosion Mechanism Analysis
It can be seen from Figure 7 that the cracks were all over the surface after dehydration and the corrosion products on the surface started along the grain boundary.This was typical intergranular corrosion.Because the activity of grain boundary was relatively large, so the corrosion process began at or near the grain boundary first [17].Furthermore, the corrosion products of the USQ sample were attached to the surface, while the corrosion products of the local area of the UUSQ sample had fallen off, and pitting occurred (Figure 7b,e).Additionally, the corrosion cracks of the UUSQ sample were much wider, deeper, and denser than that of the USQ sample (Figure 7c,f).This was because the grains of the carbides of steel obtained by UUSQ were big and unevenly distributed in the microstructure [18,19].While under the effect of ultrasound, the cooling speed of the sample was greater than the critical cooling rate, which can obtain more uniformly distributed, fine-grained, and regularly shaped carbides to help steel improve the corrosion resistance.In other words, the corrosion

Corrosion Products and Corrosion Mechanism Analysis
It can be seen from Figure 7 that the cracks were all over the surface after dehydration and the corrosion products on the surface started along the grain boundary.This was typical intergranular corrosion.Because the activity of grain boundary was relatively large, so the corrosion process began at or near the grain boundary first [17].Furthermore, the corrosion products of the USQ sample were attached to the surface, while the corrosion products of the local area of the UUSQ sample had fallen off, and pitting occurred (Figure 7b,e).Additionally, the corrosion cracks of the UUSQ sample were much wider, deeper, and denser than that of the USQ sample (Figure 7c,f).This was because the grains of the carbides of steel obtained by UUSQ were big and unevenly distributed in the microstructure [18,19].While under the effect of ultrasound, the cooling speed of the sample was greater than the critical cooling rate, which can obtain more uniformly distributed, fine-grained, and regularly shaped carbides to help steel improve the corrosion resistance.In other words, the corrosion resistance of the USQ sample was better than that of the UUSQ sample in seawater environment.The same conclusion can be obtained from the electrochemical test analysis.
Figure 8 shows the corroded surface microstructure and EDS analysis of 35CrMoV.It can be observed that the corrosion layer has been seriously damaged.At the same time, the agglomerated-like corrosion products can be seen in the corrosion cracking area.In combination with Figure 8 and Table 1, it is not difficult to see that the corrosion products of the samples are mainly composed of C, O, and Fe elements.We can speculate that the activation of erosion due to the Cl − adsorbed on the metal surface, and Fe 2+ dissolved from the anode, unites with Cl − to form FeCl 2 •4H 2 O, and then generate Fe(OH) 2 .However, the Fe(OH) 2 was not stable, and would gradually break down into FeO or combine with O 2 in water to form FeOOH [16,20,21].Finally, the oxygen content of the UUSQ sample is significantly increased compared with USQ sample, indicating that ultrasonic quenching can improve the corrosion resistance of steel in a seawater environment.
resistance of the USQ sample was better than that of the UUSQ sample in seawater environment.The same conclusion can be obtained from the electrochemical test analysis.Figure 8 shows the corroded surface microstructure and EDS analysis of 35CrMoV.It can be observed that the corrosion layer has been seriously damaged.At the same time, the agglomeratedlike corrosion products can be seen in the corrosion cracking area.In combination with Figure 8 and Table 1, it is not difficult to see that the corrosion products of the samples are mainly composed of C, O, and Fe elements.We can speculate that the activation of erosion due to the Cl − adsorbed on the metal surface, and Fe 2+ dissolved from the anode, unites with Cl − to form FeCl2•4H2O, and then generate Fe(OH)2.However, the Fe(OH)2 was not stable, and would gradually break down into FeO or combine with O2 in water to form FeOOH [16,20,21].Finally, the oxygen content of the UUSQ sample is significantly increased compared with USQ sample, indicating that ultrasonic quenching can improve the corrosion resistance of steel in a seawater environment.

Conclusions
The effects of USQ on 35CrMoV steel corrosion in a seawater environment were investigated by SEM, EDS, and EIS.
1. Ultrasonic waves can increase the quenching intensity of the quenching medium, and helped to reduce residual ferrite and refine the martensite in 35CrMoV steel.

Conclusions
The effects of USQ on 35CrMoV steel corrosion in a seawater environment were investigated by SEM, EDS, and EIS.

1.
Ultrasonic waves can increase the quenching intensity of the quenching medium, and helped to reduce residual ferrite and refine the martensite in 35CrMoV steel.

2.
Compared with the UUSQ sample, the corrosion potential of the USQ sample increased, and the corrosion current density decreased by several orders of magnitude.

3.
The corrosion cracks of the UUSQ sample were much wider, deeper, and denser than that of the USQ sample, verifying that the USQ sample had a better corrosion resistance in a seawater environment.

Metals 2018, 8 ,
x FOR PEER REVIEW 3 of 8 Before commencing the experiments, the open circuit potential test was applied for 2400 s.The potentiodynamic polarization was scanned at 0.5 mV/s.The electrochemical impedance spectroscopy (EIS) measurements were carried out after 2400 s, at the open circuit potentials, with a range of frequency from 0.1 Hz to 100,000 Hz with an amplitude of 5 mV and a sampling rate of 12 points per decade.

Figure 5 .
Figure 5. Equivalent circuit model used to fit the EIS data.

Figure 5 .
Figure 5. Equivalent circuit model used to fit the EIS data.Figure 5. Equivalent circuit model used to fit the EIS data.

Figure 5 .
Figure 5. Equivalent circuit model used to fit the EIS data.Figure 5. Equivalent circuit model used to fit the EIS data.

Figure 6 .
Figure 6.Potentiodynamic polarization curves of 35CrMoV in the artificial seawater solution.

Figure 6 .
Figure 6.Potentiodynamic polarization curves of 35CrMoV in the artificial seawater solution.

Table 3 .
Electrochemical test results of 35CrMoV.