Research on the Pitting Mechanism of Dispersion-Strengthened High-Strength Steel Under Wet-Dry Cycling in a Marine Environment

LI Ping, GENG Yanming, and DU Min

Research on the Pitting Mechanism of Dispersion-Strengthened High-Strength Steel Under Wet-Dry Cycling in a Marine Environment

LI Ping, GENG Yanming, and DU Min*

The Key Laboratory of Marine Chemistry Theory and Technology, Ministry of Education, College of Chemistry and Chemical Engineering, Ocean University of China, Qingdao 266100, China

Due to the wet-dry cycling in the ocean tidal zone, the supply of dissolved oxygen and salt-containing particles were sufficient, so the corrosion was serious. Pitting corrosion was a common form of localized corrosion. This paper studied the pitting corrosion mechanism of dispersion-strengthened high-strength steel under different wet-dry ratio environments. Electrochemical Impedance Spectroscopy was used to study the changes of corrosion rate and electric double layer structure of the dispersion- strengthened high-strength steel. Scanning electron microscope, Raman spectroscopy, electron probe microanalysis and laser confocal scanning microscope were used to observe the corrosion product morphologies, analyze the corrosion product compositions, analyze the secondary distribution of alloy elements in the corrosion products and analyze the pitting information of the steel after the corrosion products were removed. The results showed that the degree of the steel corrosion was slight when the wetting time was longer, the size and depth of the corrosion pits on the surface were smaller. This was because the environment with longer wetting time made the corrosion products denser. In this environment, the conversion of γ-FeOOH to Fe3O4was promoted. In addition, it could also promote the alloying elements to be more concentrated in the rust layer. The above phenomena hindered the further corrosion of the matrix by dissolved oxygen and Cl?in the seawater.

marine; dispersion-strengthened high-strength steel; wet-dry cycling; pitting; rust

1 Introduction

The ocean tidal zone had sufficient supply of dissolved oxygen and salt-containing particles due to the wet-dry cycling conditions. Therefore, the corrosion rate of marine steel structures in the tidal zone was greater than that in the seawater full immersion zone. Generally, the average corrosion rate in the tidal zone of the spray was 0.3–0.5mmyr?1, and the local corrosion were relatively more serious (Yu., 2018). Pitting corrosion was one of the most common forms of local corrosion in marine atmospheric environment (Park., 2015; Wang., 2015). In the drying stage of wet-dry cycling environment, with the decrease of humidity, Cl?in the liquid film was concentrated. When Cl?exceeded the critical concentration, pitting would occur on the surface of high-strength steel. Pitting corrosion was difficult to detect at the beginning of its occurrence. The pitting corrosion that occurred in the equipment could cause perforation of steel materials, which greatly brings about the construction safety hazards and economic losses. The rust layer form- ed on the surface of steel structure was also very important for corrosion in the marine tidal zone (Evans and Taylor, 1972; Stratmann., 1989). Due to the sufficient supply of dissolved oxygen during the dry wet alternation process, the oxidizing effect of the rust layer itself increased the cathodic current, and the nature of the rust layer would affect the pitting corrosion of steel (Wang., 2019; Zhang and Ma, 2019).

There were many factors that affected the induction of pitting corrosion, such as inclusions, alloy elements, rust layers and so on (Soltis, 2015). This paper studied a new type of high-strength steel that was dispersion-streng- thened (a strengthening method by adding hard particles to a homogeneous material). Shi. (2019) used transmission electron microscope, high resolution transmission electron microscope, three-dimensional atom probe, ato- mic force microscope and combined with first-principles calculation method to study the dispersion-strengthened high-strengthened steel nano-precipitated phase. The experimental results showed that a large number of dispersed nano-precipitates were distributed in the matrix of dispersion-strengthened high-strength steel. Its size was a few nanometers to tens of nanometers and the main components were C, N, Ti, Nb, V,. Its crystal structure was FCC face centered cubic structure. The pitting corrosion resistance of this high-strength steel is not yet known. This kind of steel added a variety of corrosion-resistant alloying elements Cr, Ni, Mo,., as shown in Table 1. The relationship between the corrosion resistance of steel structure materials and alloying elements in marine environments varied with the environment (Díaz., 2018). The alloying elements in low-alloy steel have a certain impact on the corrosion resistance of steel structures in the marine tidal zone. The influence of the same alloying elements on steel structures (such as steel pile wharves, offshore oil platforms,.) starting from the marine atmosphere zone, passing through the marine wave splash zone, the marine tidal range, the seawater full immersion zone and the submarine mud zone were also completely different (Nishimura., 2001). Chen (2005) believed that certain alloying elements could improve the corrosion resistance of steel materials in the marine tidal zone, but the effect on the seawater full immersion area was not obvious, and may increase the corrosion rate.

In this paper, the influence of different dry and wet ratios on the rust layer was studied by simulating the wet- dry cycling environment of the marine. And from the per- spective of the rust layer, the pitting mechanism of the dispersion-strengthened high-strength steel was discuss- ed.

2 Experiment

2.1 Experimental Materials and Medium

The research object was dispersion-strengthened high- strength steel, and its chemical composition was shown in Table 1. The steel samples were cut into 10mm×10 mm×3mm and polished with 800#, 1000# and 2000# water matte paper in order to light. Wash the surface with distilled water and absolute ethanol respectively.

The experimental medium was the natural seawater of Qingdao Wheat Island.

Table 1 Compositions of dispersion-strengthened high-strength steel

2.2 Wet-Dry Cycling Settings

The pitting corrosion behavior of steel under two different dry wet cycles was studied. Taking 12h as a cycle period, one was drying for 10h and wetting for 2h, the other was drying for 2h and wetting for 10h. The dry-wet ratios were 5:1 and 1:5 respectively. Hang the samples in the experiment box with a glass rod, so that the samples were in a wet-dry cycling environment (Table 2).

Table 2 Schedule of dry and wet cycles

2.3 Experimental Methods

Electrochemical impedance spectroscopy (EIS) tests were carried out in a three-electrode cell through a Germany Reference 600 electrochemical measurement system. Dispersion-strengthened high-strength steel was a working electrode that was embedded in an epoxy resin PVC holder in a square form (10mm×10mm×3mm) with a working area of 1cm2. A platinum electrode and a saturated calomel electrode (SCE) were used, respectively, as auxiliary and reference electrodes. Before the electrochemical test, took the electrode out of the experimental device and put it in a jar filled with seawater and let it stand for 30min to stabilize the open circuit potential (OCP) of the steel. The test frequency range was 100kHz –10mHz, and the amplitude of the sine wave AC signal was 10mV with respect to OCP.

JSM-6700F scanning electron microscope (SEM) was used to observe the morphology of the inner and outer rust layers of dispersion-strengthened high-strength steel.

Raman spectroscopy was used to determine the compositions of the rust layers. A DXR Microscope model of Raman spectrometer was used, the selected light source was 532nm, and the scanning range was 0–1500cm?1.

JXA-8230 electron probe microanalysis (EPMA) was used to test the distribution of alloying elements in the rust layer. First took the sample out of the experimental device and rinsed it with distilled water. After it dries, it was sealed with epoxy resin. The section of the rust layer was exposed, and then polished to smooth with 800# and 1000# water matte paper respectively. Then put it into the electron probe microanalyzer, and selected a rectangular area containing the inner and outer rust layers, and detected the element distribution in this area.

VK-X200 laser confocal scanning microscope (LCSM) was used to observe the pitting on sample surface after removing corrosion products. In each test, four different areas on the same straight line were selected for observation. After obtaining the pitting corrosion data, used data analysis software to build a 3D model for analysis. The pitting corrosion information listed in this paper were the average of 4 data in this environment.

3 Results and Discussion

3.1 Electrochemical Analysis

Fig.1 was the EIS plots of the dispersion-strengthened high-strength steel under two dry and wet environments. When the drying time was longer (cycle I), it could be seen in the Nyquist plot (Fig.1(a)) that it was composed of a single capacitive loop during the first 5d of corrosion. After 10d of corrosion, it consisted of a semicircle in the high frequency area and a diagonal line in the low frequency area. The diagonal line in the low frequency area indicated the presence of Warburg. Warburg was related to the propagation and diffusion process of materials near the interface. This indicated that the corrosion products generated on the electrode surface at this time had an impact on the electrochemical process. At this time, the elec- trochemical process on the electrode surface was controlled by the mixing of the charge transfer process and the diffusion process. However, when the wetting time was longer (cycle II), the Warburg did not appear until 30d after corrosion.

The modulus of the impedance in the intermediate frequency zone had a linear relationship with frequency and both were less than ?1. When the slope was close to ?1, the surface state of the electrode was completer and more compact. It meant that the electrode was closer to the complete capacitor (Porcayo., 2017). After the same corrosion time, the slope of the intermediate frequency region of the modulus curves (Figs.1(c) and (d)) of the dispersion-strengthened high-strength steel with longer wetting time (cycle II) were closer to ?1.

When the drying time was longer (cycle I), it could be seen in the phase angle plot (Fig.1(e)) that with the extension of the corrosion time, the phase angle peak gradually shifted to the low frequency region. This showed that although the corrosion products of the dispersion-streng- thened high-strength steel have increased, its compactness did not increase; However, when the wetting time was longer (cycle II), in the first 10d of corrosion, the phase angle peaks (Fig.1(f)) moved to the low frequency area, and then the phase angle peaks position did not change much. The above showed that in wet-dry cycling environ- ments, an environment with longer wetting time was more conducive to the generation of dense corrosion products.

Fig.1 EIS plots of dispersion-strengthened high-strength steel under two wet-dry cycling environments. (a, c, e) Nyquist plot, Modulus plot, Phase angle plot under Cycle I; (b, d, f) Nyquist plot, Modulus plot, Phase angle plot under Cycle II.

In the environment of longer drying time (cycle I), the corrosion products on the surface were less after 5d of corrosion. The EIS plots were fitted by the equivalent circuit diagram shown in Fig.2(a). After 10d of corrosion, Warburg impedance could be seen from the Nyquist diagram (Fig.1(a)), and the rust layer was stratified. Therefore, the equivalent circuit diagram shown in Fig.2 (c) was used for fitting, and the fitting results were listed in Table 3. In the environment of longer wetting time (cycle II), the EIS plots of the first 10d of corrosion were fitted by the equivalent circuit diagram shown in Fig.2(a). With the increase of corrosion products on the surface, the rust layer began to be divided into inner and outer layers. The equivalent circuit diagram shown in Fig.2(b) was used for fitting when the corrosion lasts for 15 to 25d. Warburg impedance appeared after 30d of corrosion (Fig.1(b)), so the equivalent circuit diagram shown in Fig.2(c) was used for fitting. Fitting results were listed in Table 4.

Fig.2 Equivalent circuits used to fit the EIS measurement.

Table 3 The electrochemical parameters fitted from the EIS data of dispersion-strengthened high-strength steel at different corrosion time under cycle I

Table 4 The electrochemical parameters fitted from the EIS data of dispersion-strengthened high-strength steel at different corrosion time under cycle II

The impedance of a CPE is a function of the frequency and it is defined as:

whereCPEwas the magnitude of CPE, Ω?1scm?2;was the square root of ?1;was the angular frequency, rads?1; and the exponent(?1≤≤ 1) denoted the distribution of time constant.

It could be seen from the fitting results that thectof the dispersion-strengthened high-strength steel were larger under the environment of cycle II, which indicated that the corrosion rate of the high strength steel was slow when the wetting time was longer. The change of surface rust resistance of dispersion-strengthened high-strength steel in two environments were shown in Fig.3. Fig.3(a) showed that the resistance of inner and outer rust layer decreased with the increase of corrosion time under longer drying time environment (cycle I). It showed that the protection of rust layer decreased with the increase of corrosion time. However, the resistance of inner and outer rust layers showed different tendency in the environment of longer wetting time (cycle II) (Fig.3(b)). With the increase of corrosion time, the resistance of outer rust layer decreased and the resistance of inner rust layer increased. The results showed that a longer wetting time promoted the formation of dense inner rust layer, which played a protective role for the dispersion-strengthened high- strength steel in wet-dry alternate environment.

Fig.3 Change of inner and outer rust layer resistance. (a), cycle I; (b), cycle II.

3.2 The Effect of Wet-Dry Cycling on Pitting Corrosion

In order to observe the pitting corrosion on the surface of dispersion-strengthened high-strength steel, the corroded steel was derusted and observed by LCSM. Fig.4 showed the 3D morphologies of the dispersion-streng- thened high-strength steel after 15d and 30d of corrosion in two environments. The blue areas were the corrosion pits and the red areas were the platforms. Under the condition of a longer drying time (cycle I), there were many large and deep corrosion pits on the surface of dispersion-strengthened high-strength steel. As the corrosion time increased, the depth of the corrosion pits increased. The surfaces of high-strength steel were relatively smooth under the condition of a longer wetting time (cycle II), and the size and depth of corrosion pits were significantly smaller. The corrosion pits information of dispersion- strengthened high-strength steel under different environments were summarized in Table 5.

Fig.4 3D images of dispersion-strengthened high-strength steel under two wet-dry cycling environments. (a), 15d under cycle I; (b), 30d under cycle I; (c), 15d under cycle II; (d), 30d under cycle II.

Through Table 5 we could intuitively understand the corrosion pits information of the dispersion-strengthened high-strength steel under the two wet-dry cycling environments. As the corrosion time increased, the pitting of the dispersion-strengthened high-strength steel under the two environments had a development trend, and the depth and the size were expanding. However, after corrosion for the same time, the maximum pit depth and the equivalent diameter of the pits were significantly larger in the environment with a longer drying time (cycle I). The maximum pit depth of steel in the environment of longer drying time (cycle I) was about 10μm larger than that in the environment of longer wetting time (cycle II). The equivalent diameter of the pits in the longer drying time environment (cycle I) was 2–3 times of that in the other environment (cycle II). When the drying time was longer (cycle I), the maximum depth of the pits and the diameter of the corrosion pits were larger than when the wetting time was longer (cycle II), but the density of pitting corrosion was lower. It might be that multiple corrosion pits gradually developed and connected after drying for a longer time. As a result, the number of pits decreased.

Table 5 Information on pits of dispersion-strengthened high-strength steel

3.3 The Effect of Wet-Dry Cycling on the Rust Layer

3.3.1 Analysis of rust layer morphologies

The process of wet-dry cycling was essentially a process of changing the thickness of the thin liquid film on the metal surface. The reduction reaction of dissolved oxygen in the thin liquid film at different stages also leaded to a step change in the corrosion potential. The rust layer also had a great influence on the corrosion process. As the corrosion progressed the rust layer became thicker, and the seawater in the rust layer took longer to evaporate. Therefore, the outer rust layer directly affected the wetting time of the metal surface.

Fig.5 Corrosion product morphologies of dispersion-strengthened high-strength steel under two wet-dry cycling environments. (a–d) 15d inner layer, 15d outer layer, 30d inner layer, 30d outer layer under cycle I; (e–h) 15d inner layer, 15d outer layer, 30d inner layer, 30d outer layer under cycle II.

Fig.5 was the morphologies of the inner and outer layers of the corrosion products of the dispersion-streng- thened high-strength steel corroded for 15 and 30d under two wet-dry cycling environments. When the drying time was longer (cycle I), after 15d of corrosion, the inner and outer rust layers were loose. The shape of the rust layer was granular and had many gaps and holes. The existence of these gaps and holes provided a powerful transmission channel for dissolved oxygen and corrosive ions in seawater. The looseness of the rust layer increased its water retention, so that the surface of the dispersion-streng- thened high-strength steel remained wet for a long time. Therefore, when the drying time was longer (cycle I), the loose rust layer promoted corrosion after 15d of corrosion. When the wetting time was longer (cycle II), the inner rust layer after 15d was lumpy and granular, and the outer rust layer was granular and piled up in clumps. The rust layer was densely packed and had few holes, which hindered the progress of corrosion to a certain extent. After 30d of corrosion, the density of the inner rust layer had increased under the environment of a longer drying time (cycle I), but it could be seen that the outer rust layer was still relatively loose and could store more seawater. However, the inner and outer rust layers were denser under the environment of longer wetting time (cycle II) which could protect the steel to a certain extent. Therefore, from the perspective of the protection of the rust layer, when the wetting time was longer (cycle II), the rust layer was denser, with fewer gaps and holes and better protection. Therefore, it could prevent the corrosive ions and dissolved oxygen in the seawater from penetrating the rust layer to the steel surface, preventing the further occurrence of pitting corrosion of the dispersion-streng- thened high-strength steel. That was consistent with the results of EIS.

3.3.2 Analysis of rust layer compositions

Stratmann. (1990) pointed out that the protective properties of the rust layer depended on its chemical composition, adhesion, compactness and moisture absorption capacity. In the wet-dry cycling environment, rust layer acted as a strong oxidant, which could promote electrochemical reactions under wet or humid stages (Thee., 2014). Raman spectroscopy was used to analyze the compositions of the rust of the dispersion-streng- thened high-strengthened steel in two wet-dry cycling en- vironments for different times. The results were shown in Figs.6 and 7. It could be seen that the corrosion products of dispersion-strengthened high-strength steel in wet-dry cycling environment were mainly Fe3O4and γ-FeOOH. The corrosion product compositions under different conditions were summarized in Table 6. After 15d of corrosion, a small amount of β-FeOOH appeared in the corrosion products in both environments. When the drying time was longer (cycle I), β-FeOOH was in the inner rust layer. However, when the wetting time was longer (cycle II), β-FeOOH only existed in the outer rust layer. β-FeOOH was formed in the presence of chloride ions and under low pH environment (Nomura., 1988). Combined with the corrosion product morphology diagrams in Fig.5, when the drying time was longer (cycle I), the outer rust layer was very loose. Cl?could penetrate the loose outer rust layer to the inside of the rust layer, and reacted with the inner rust layer to form β-FeOOH. The crystal structure of β-FeOOH had holes like tunnels, and Cl?could be filled in the holes to form basic hydrox- ides with chlorine (Deliyanni., 2001). Nishimura. (1995) believed that β-FeOOH was loose and had low adhesion. β-FeOOH could turn into a green layer in the dry period in a wet-dry cycling environment containing Cl?. Therefore, β-FeOOH would destroy the protective of the rust layer and accelerate the corrosion rate of steel.

Table 6 Summary of corrosion products compositions under two wet-dry cycling environments

Fig.6 Raman diagrams of rust layercompositions of dispersion-strengthened high-strength steel under cycle I environment. (a), 15d; (b), 30d.

Fig.7 Raman diagrams of rust layer compositions of dispersion-strengthened high-strength steel under cycle II environment. (a), 15d; (b), 30d.

The anodic reaction on the surface of dispersion- strengthened high-strength steel at the initial stage of corrosion was the dissolution of iron and the depolarization reaction of oxygen at the cathode, as shown in formulas 2 and 3.

With the progress of the corrosion electrochemical reaction, the rust layer on the surface of the steel gradually grew and increased, and the rust layer also began to participate in the reaction. The corrosion process of the steel surface under the wet-dry cycling marine environment could be divided into the following three processes (Yadav., 2004; Morcillo., 2014):

1) Surface wetting process

During the transition from the dry state to the wet state, the anode reaction was the dissolution of iron, and the cathodic reaction was the reduction of γ-FeOOH in the rust layer (Evans and Taylor, 1972).

2) Complete surface wetting stage

When in a wet state, after the γ-FeOOH that could be reduced was consumed, the reduction of dissolved oxygen became the main reaction of the cathode. In this process, the rust layer was filled with electrolyte, which was not conducive to the diffusion of O2, so the corrosion rate was slow. The dissolution rate of the metal depended on the limiting diffusion current density of the O2reduction reaction in the pore electrolyte of the rust layer. The cathode reaction was as follows:

3) Surface drying process

In the process of transition from the wet state to the dry state, the thin liquid film on the surface of the rust layer was getting thinner and thinner. A large amount of O2was supplied through the thin liquid film, and the diffusion of O2was accelerated, and the corrosion rate was accelerated. O2reduction reaction occurred during the cathode process, and γ-Fe·OH·OH would be oxidized to γ-FeOOH. The reaction formula was as follows:

In the wet-dry cycling environment, the components of the rust layer were mutually transformed. The reduction ability of the components in the rust layer was β-FeOOH>γ-FeOOH>α-FeOOH (Lair., 2006). The electrochemical reactions involved in the conversion of the rust layer were mainly γ-FeOOH and Fe3O4. When the drying time was longer (cycle I), the γ-FeOOH in the rust layer was more than another environment and Fe3O4showed the opposite trend. The reason was that in the wet stage of the wet-dry cycling environment, the process of converting γ-FeOOH to Fe3O4mainly occurred, and vice versa in the dry stage (Matsushima, 2004).

After 30d of corrosion, the types of corrosion products generated in the two environments were the same. When the wetting time was longer (cycle II), the α-FeOOH in the rust layer was only detected in the inner rust layer. α-FeOOH had a stable structure and was difficult to undergo electrochemical reactions, and to a certain extent, so it could protect the dispersion-strengthened high- strength steel from corrosion.

3.4 The Effect of Wet-Dry Cycling on the Secondary Distribution of Alloying Elements

The alloying elements added to the high-strength steel played an important role in its corrosion resistance. The alloying elements had different solubility in the steel and the rust layer (Zhou., 2013). As the corrosion progressed, the alloying elements would undergo secondary distribution in the corrosion products. EMPA was used to observe the distributions of alloying elements in the rust layer, as shown in Fig.8. The left side of each figure was a cross-sectional view of the rust layer, and the right side was the element distribution in the red frame of the rust layer section. It could be seen from the cross section of the rust layer that the longer the drying time (cycle I), the thicker the rust layer. It could also indicate that the corrosion of dispersion-strengthened high-strength steel was more serious when the drying time was longer (cycle I). After 15d of corrosion with longer drying time, Cr, Ni, Mo and Cl elements were concentrated in the boundary area between the inner and outer rust layers. After 30d of corrosion with longer drying time, Cr, Ni and Mo elements ‘diffused’ from the inner rust to the outer rust, and the distribution areas increased. On the other hand, in the environment with longer wetting time (cycle II), enrichment zones of Cr, Mg and Mo appeared at the junction of inner and outer rust layer at 15d. Ni were distributed throughout the rust layer, but mainly concentrated in the inner rust layer. The enrichment of alloy elements at the junction of inner and outer rust layer could hinder the diffusion of the external corrosive medium to the steel, hinder the further growth of the rust layer grains and make the rust layer grew dense. At 30d of corrosion with longer wetting time, the enrichment zones disappeared. Cr and Ni elements were mainly concentrated in the internal rust, and Mo elements were distributed throughout the rust layer. At this time, the Cl element detected in the rust layer was very little. Therefore, in a longer wetting time environment (cycle II), during the secondary distribution of alloying elements, the alloying elements were easily concentrated in the internal rust layer.

It could be seen from the Fig.6 that the distribution of Ni element and Cl element were opposite to each other. Nishimura. (2003)believed that the alloying element Ni easily entered the rust layer during the corrosion process and formed a spinel oxide NiFe2O4with the same structure as Fe3O4. Compared with Fe3O4, NiFe2O4had higher thermodynamic and electrochemical stability, which was beneficial to reduce the corrosion rate. In addition, studies had shown that NiFe2O4had cation selective permeability, which could prevent Cl?in the marine passing through the rust layer to the surface of the steel. This could prevent Cl?from inducing pitting corrosion of the dispersion-strengthened high-strength steel. It showed that under the environment with a longer wetting time (cycle II), the alloying elements concentrated inside the rust layer during the secondary distribution, and increased the protective of the rust layer. Cr element was more likely to produce a certain degree of element segregation in the rust layer, which could repair defects such as pores and cracks in the rust layer and enhance the compactness of the rust layer (Chai., 2016; Jung., 2018). The Cr element was mainly distributed in the inner rust layer and the interface between the inner and outer rust layers. The enrichment of the Cr element in the inner rust layer would reduce the degree of crystallization of corrosion products, refine the rust layer and promote the formation of α-FeOOH. In this process, the secondary distribution of Cr promoted the increase of α-FeOOH and Fe3O4in the inner rust layer. So the inner rust layer contained less γ-FeOOH and more Fe3O4.

Fig.8 EMPA diagrams of cross-section elementdistributions of rust layer of dispersion-strengthened high-strength steel under two wet-dry cycling environments. (a), 15d under cycle I; (b), 30d under cycle I; (c), 15d under cycle II; (d), 30d under cycle II.

4 Conclusions

The control steps of the electrochemical reaction were different in the two marine wet-dry cycling environments. After 10d of corrosion with a longer drying time (cycle I), the electrochemical process was jointly controlled by charge transfer process and diffusion process. The electrochemical process was controlled by charge transfer and diffusion after 30d of corrosion when the wetting time was longer (cycle II). Before that, the electrochemical process was only controlled by charge transfer process. The pitting corrosion of dispersion-strengthened high- strength steel was more serious when the drying time was longer (cycle I). However, when the wetting time was longer (cycle II), the high-strength steel corroded slightly, and the size and depth of the corrosion pits on the surface were small. This was because in an environment with a longer wetting time (cycle II), the dispersion-strengthened high-strength steel could be promoted to generate denser corrosion products. It could promote the conversion of γ-FeOOH to Fe3O4in the rust layer and the formation of α-FeOOH in the inner rust layer. In addition, it also promoted the concentration of Cr, Ni, Mo alloying elements in the inner rust layer. Due to the above reasons, the dissolved oxygen and Cl?were prevented from penetrating the rust layer into the surface of the dispersion-streng- thened high-strength steel, which reduced the further occurrence of pitting corrosion.

Acknowledgement

The study is supported by the National Natural Science Foundation of China (No. U1706221).

Chai, F., Jiang, S., and Yang, C. F., 2016. Effect of Cr on characteristic of rust layer formed on low alloy steel in flow-accelerated corrosion environment., 23 (6): 602-607.

Chen, Y. Y., Tzeng, H. J., Wei, L. I., Oung, J. C., and Shin, H. C., 2005. Corrosion resistance and mechanical properties of low-alloy steels under atmospheric conditions., 47 (4): 1001-1021.

Deliyanni, E. A., Bakoyannakis, D. N., Zouboulis, A. I., Matis, K. A., and Nalbandian, L., 2001. Akaganeite-type β-FeOOH nanocrystals: Preparation and characterization., 42 (1): 49.

Díaz, I., Cano, H., Lopesino, P., de la Fuente, D., Chico, B., Jiménez, J. A.,., 2018. Five-year atmospheric corrosion of Cu, Cr and Ni weathering steels in a wide range of environments., 141: 146-157.

Evans, U., and Taylor, C., 1972. Mechanism of atmospheric rusting., 12 (3): 227-246.

Jung, K. M., Ahn, S. H., Kim, Y. J., Oh, S. K., Ryu, W. H., and Kwon, H. S., 2018. Alloy design employing high Cr concentrations for Mo-free stainless steels with enhanced corrosion resistance., 140: 61-72.

Lair, V., Antony, H., Legrand, L., and Chausse, A., 2006. Electrochemical reduction of ferric corrosion products and evaluation of galvanic coupling with iron., 48 (8): 2050-2063.

Matsushima, R., translated by Jing, Y. K., 2004.. Metallurgical Industry Press, Beijing,230pp.

Morcillo, M., Díaz, I., Chico, B., Cano, H., and Fuente, D., 2014. Weathering steels: From empirical development to scientific design, a review., 83 (7): 6-31.

Nishimura, T., and Kodama, T., 2003. Clarification of chemical state for alloying elements in iron rust using a binary-phase potential-pH diagram and physical analyses., 45 (5): 1073-1084．

Nishimura, T., Noda, K., and Kodama, T., 2001. Corrosion behavior of tungsten-bearing steel in a wet/dry environment containing chloride ions., 57 (9): 753-758.

Nishimura, T., Tanaka, K., and Shimizu, Y., 1995. Effect of NaCl on rusting of steel in wet and dry corrosion cycle., 81 (11): 1079-1084.

Nomura, K., Tasaka, M., and Ujihira, Y., 1988. Conversion electron Mossbauer spectrometric study of corrosion products of iron immersed in sodium chloride solution., 44 (3): 131-135.

Park, I. J., Lee, S. M., Kang, M. W., Lee, S. J., and Lee, Y. K., 2015. Pitting corrosion behavior in advanced high strength steels., 619: 205-210.

Porcayo, C., Rivera, M. J., Peza-Ledesma, E. M., Casales-Diaz, C., and Martinez-Gomez, M. L., 2017. Sustainable development of palm oil: Synthesis and electrochemical performance of corrosion inhibitor., 8: 133.

Soltis, J., 2015. Passivity breakdown, pit initiation and propagation of pits in metallic materials – Review., 90: 5-22.

Stratmann, M., Bohnenkamp, K., and Engell, H. J., 1989. An electrochemical study of phase-transitions in rust layers., 23 (9): 969-985.

Stratmann, M., and Streckel, H., 1990. On the atmospheric corrosion of metals which are covered with thin electrolyte layers-I. Verification of the experimental technique., 30 (6-7): 681-696.

Thee, C., Hao, L., Dong, J. H., Mu, X., Wei, X., Li, X. F., and Ke, W., 2014. Atmospheric corrosion monitoring of a weathering steel under an electrolyte film in cyclic wet-dry condition., 78: 130-137.

Wang, L. W., Xin, J. C., Cheng, L. J., Zhao, K., Sun, B. Z., Li, J. R.,., 2019. Influence of inclusions on initiation of pitting corrosion and stress corrosion cracking of X70 steel in near-neutral pH environment., 147: 108- 127.

Wang, Y. F., Cheng, G. X., Wu, W., and Qiao, Q., 2015. Effect of pH and chloride on the micro-mechanism of pitting corrosion for high strength pipeline steel in aerated NaCl solutions., 349: 746-756.

Yadav, A. P., Nishikata, A., and Tsuru, T., 2004. Electrochemical impedance study on galvanized steel corrosion under cyclic wet-dry conditions-influence of time of wetness., 46 (1): 169-181.

Yu, J. X., Wang, H. K., Yu, Y., Luo, Z., Liu, W. D., and Wang, C. M., 2018. Corrosion behavior of X65 pipeline steel: Comparison of wet-dry cycle and full immersion., 133: 276-287.

Zhang, B., and Ma, X. L., 2019.A review–Pitting corrosion initiation investigated by TEM., 35: 1455-1465.

Zhou, Y. L., Chen, J., Xu, Y., and Liu, Z. Y., 2013. Effects of Cr, Ni and Cu on the corrosion behavior of low carbon microalloying steel in a Cl?containing environment., 29 (2): 168-174.

November 16, 2020;

April 17, 2021;

June 8, 2021

? Ocean University of China, Science Press and Springer-Verlag GmbH Germany 2021

. Tel: 0086-532-66781637

E-mail: ssdm99@ouc.edu.cn

(Edited by Ji Dechun)