Keywords

1 Introduction

In the field of bridge construction, high strength bolts, as the key connecting fasteners of bridge trusses, play a significant role in the safety connection between steel structure trusses. Thus, the safety property of the bolt will determine the life of the whole steel structure. With the continuous development of society, the construction of more and more bridge structures are towards the direction of high strength and large span, which will require better properties of high strength fasteners.

Corrosion will change the mechanical properties of steel structure, which is an important disease of steel structure bridge. According to the data of Japanese statistics, the strength of the steel structure under stress will be decreased by 10 ~ 15% when 1% of the steel is corroded; and the structure will be scrapped when 5% of each side for the steel is corroded [1]. For the steel bridges, the failure of connecting nodes is one of the dominant breaking modes. For one steel bridge along one railway in China, the annual rupture rate of high strength bolt because of corrosion is about 0.2% [2]. And the breakage rate of high strength bolts in the steel structure of Chongqing Chaotianmen Yangtze river bridge is about 0.0025% [3]. At present, many large bridges are built at home and abroad, which are all steel bridges, in the coastal cities or along the river, in the environment of high humidity, high salt or serious air pollution. For the Nanpanjiang Bridge constructed in 1998, it was detected in 2008 that the joints on the main truss were corroded, with 80% moderate corrosion (rust pit depth 0.6–1.2 mm) [4]. The precipitation and adsorption of harmful materials onto the bridge is one predominant reason for the corrosion. Besides, the exposure time, relative humidity, temperature, wind speed, corrosive medium and other factors can also influence the corrosion behavior of steel [5,6,7,8]. These investigations illustrate that the high strength bolts at the steel joints are prone to corrosion by the industrial and Marine atmosphere and undergo rupture, which will endanger the safety of the whole steel bridge.

Therefore, in this paper, based on previous research and through the investigation on atmospheric environments of Chongqing City in recent years, corrosion solution was designed to simulate the humid climate of Chongqing, and the electrochemical corrosion experiments of high strength bolts were carried out under simulated conditions of humid climate in Chongqing as functions of corrosion time and pH. The results will provide scientific fundamentals and research guidance for the maintenance and protection of the bridges.

2 Experimental Methods

2.1 Design and Formulation of Simulated Corrosion Solution

According to the Ecological Environmental Bulletins from 2010 to 2018 released by Chongqing Ecology and Environment Bureau (Chongqing Ecology and Environment Monitoring Center), the atmospheric pollution in Chongqing has transformed from coal-burning pollution to complex pollution by vehicle engine exhaust and soot. Although acid rain is effectively controlled, the pollution from nitrogen oxides, ozone and dust particles is on the rise. The pH value of precipitation has risen from about 4.5 in the past to about 5.5 in recent years.

Based on the precipitation components in Chongqing from the above Chongqing’s Bulletins, and the reports [9,10,11,12,13] by Liu and Zhang, et al., the indoor accelerated corrosion solution compositions were designed by improving the content of sulfate radical in 2017 annual precipitation in Chongqing to simulate the humid climate in Chongqing as shown in Table 1.

Table 1 Corrosion solution compositions simulating the annual precipitation in Chongqing (mg/L)
Full size table

According to the method of ion concentration equilibrium, the corrosion solution was prepared by analytical reagent and distilled water, and the pH value of the corrosion solution was adjusted by using precise pH meter, nitric acid solution and sodium hydroxide solution.

2.2 Preparation of Electrochemical Corrosion Samples

Electrochemical corrosion samples of high strength bolts with corrosion area of 10 × 10 mm were prepared. Then the samples were encapsulated with epoxy resin. One side was considered as the working surface for electrochemical testing and microscopic analysis, the gap of which was embedded with resin, while the other side was connected with electrode wires. The working surface was polished with 800#, 1200# and 2000# metallographic sandpaper, respectively, and then polished with 2.5 μm diamond spray until the surface was finished. The polished surface was washed with acetone to remove the oil, then washed with deionized water, cooled with cold air, and placed in a dryer for later use. According to relevant reports and standards [14, 15], two factors, corrosion time and the pH of the corrosion solution, were mainly considered in the electrochemical corrosion experiments.

2.3 Analysis and Test Methods

According to the design of electrochemical corrosion experiment, the samples were immersed in 200 times concentration of corrosion simulation solution with different pH values. After reaching the target corrosion time, the samples were taken out, and the electrochemical study, surface morphology and corrosion products of the corrosion surface were analyzed.

The polarization curves and electrochemical impedance spectra (EIS) of high strength bolts were analyzed by using the UTOLAB comprehensive electrochemical test system. The chemical composition and microstructure morphology of the corrosion products were analyzed by scanning electron microscope (SEM), energy dispersive spectrum (EDS) and X-ray diffraction (XRD).

3 Electrochemical Performance Analysis of 20MnTiB High Strength Bolt

3.1 Polarization Curve

Figure 1 shows the electrochemical polarization curves of the samples after immersion in the 200 times concentration of corrosion simulation solution for 0 h, 8 h, 36 h, 96 h and 168 h at different pH values.

Fig. 1
Five line graphs, a to e, plot E over V versus log of modulus I over A time centimeter superscript negative 2. The lines are plotted for p H 7.5, p H 5.5, and p H 3.5. All the graphs remain stable initially and branch out in upward and downward directions thereafter.

The electrochemical polarization curves of the samples after immersion at different pH values for different time a 0 h b 8 h c 36 h d 96 h e 168 h

Full size image

In order to further quantitatively analyze the corrosion mechanism, the polarization curves were fitted by Tafel method. The results are shown in Table 2, and the fitting parameters included the anode Tafel slope (ba), cathode Tafel slope (bc), self-corrosion potential (E), corrosion current (I), corrosion rate (Vc) and polarization resistance (Rp).

Table 2 Fitting results of the electrochemical polarization curves
Full size table

Figure 1 demonstrates that the polarization curves of the samples varied little over time when immersed in the corrosion simulation solutions with different pH values for 0–96 h, and the polarization curves at pH of 3.5 and 5.5 were similar. While after immersion for 168 h in the corrosion solution, the polarization curve was significantly changed, as Fig. 1e reveals. The corrosion potential was increased obviously with the increase of pH values. At pH 5.5 and 7.5, the anodic polarization curve showed obvious inflection point, and at the same polarization potential, the polarization curve shifted to the left with the increase of pH, indicating that anodic passivation of electrochemical corrosion is increased significantly and the anodic reaction is inhibited, thus enhancing the corrosion resistance of the materials. Simultaneously, at pH of 3.5, the anodic polarization curve of the sample still did not show an obvious inflection point. This may be mainly because the presence of a large amount of hydrogen ions and chloride ions in the strong acid corrosion simulation solution, which will result in the failure of the formation of stable corrosion products and lead to a remarkable corrosion tendency of the sample.

Table 2 demonstrates that under the same corrosion time, the corrosion current of the samples was decreased with the increase of pH. And under the same pH of 3.5 and 5.5 in the acid environment, the corrosion current was decreased with the increase of the corrosion time. When the pH was 3.5 without immersion, the corrosion current reached the peak and the corrosion rate reached 0.73839 mm/a, the corrosion resistance was the worst. While under the neutral and slightly alkaline pH 7.5, the corrosion current reached the maximum at 8 h corrosion time (only slightly higher than that at 96 h corrosion time in the acid environment), and then was decreased with the increase of corrosion time. This can be explained that when the sample is immersed for a short time (within 8 h), complete corrosion products are not formed on the sample surface, and the surface metal directly contacts with the corrosion simulation solution, which will result in a direct reaction between the sample and the erosion ions in the corrosion simulation solution. Thus, the corrosion tendency of the samples will be relatively high.

In summary, after immersion at pH of 7.5 for 168 h, the corrosion current density reached the minimum value of 1.7624 × 10–5 A/cm, while the corrosion potential and polarization resistance reached the maximum, and the corrosion resistance was the best.

3.2 Eis

According to the literature [16, 17] and the actual EIS of the samples in the corrosion simulation solution, Fig. 2a was selected as the equivalent circuit after immersion for 0 h and 8 h, and Fig. 2b was considered as that after immersion for 36 h, 96 h and 168 h.

Fig. 2
Two circuit diagrams, a and b, present the layout of circuit models with components such as R s, R t, R L, L, and Q, and R s, R t, R a d and C d l.

The equivalent circuit models a R(Q(R(RL)), b R(C(R(QR))

Full size image

In Fig. 2, Rs is the solution resistance; Rt is the charge transfer resistance; Q represents the non-ideal capacitance of the corrosion product film. Due to the uneven corrosion product film, constant phase angle element Q is used to replace the ideal capacitor C. L is inductive reactance; RL is the inductive resistance; Rad is the adsorption resistance of the electrode surface. Cdl is the double-layer capacitance on the interface; Rp represents the polarization resistance, and the corresponding value reflects the corrosion degree of the metal [16], which is defined as Rp = Rt when at the initial stage of immersion (0 h, 8 h), otherwise Rp = Rt + Rad.

Figure 3 shows the EIS of the samples at different pH values for 0 h, 8 h, 36 h, 96 h and 168 h, respectively, and EIS were fitted by Zsimdemo software according to the equivalent circuit models of R(Q(R(RL)) and R(C(R(QR)) in Fig. 2. The results are listed in Table 3.

Fig. 3
Five line graphs from a to e plot Z double dash versus Z dash. The lines are plotted for p H 7.5, p H 5.5, and p H 3.5. All the graphs depict an initial increasing trend, which is followed by a downward trend.

The EIS of the samples after immersion at different pH values for different time a 0 h b 8 h c 36 h d 96 h e 168 h

Full size image
Table 3 The fitting results of the EIS
Full size table

Figure 3a and b show that EIS was mainly composed of capacitive arc at high frequency region and inductive arc of contraction at low frequency region when the samples were immersed at different pH for 0 h and 8 h. And the contraction of the real part at low frequency region indicated the adsorption phenomenon on the surface of electrode corrosion product film [18, 19]. Figure 3c to e show that the EIS exhibited characteristics of mono-capacitive arc resistance, and no induced arc resistance with obvious shrinkage low-frequency area, when the sample is immersed in different pH corrosion simulation solution for 36 h, 96 h and 168 h, respectively. This may be because the redox reaction between metal Fe and hydroxide ions in the solution in the early stage of corrosion will initially form γ-FeOOH with a greater activity. γ-FeOOH can be considered as a cathode to absorb electrons and react with water to form Fe·OH·OH as an intermediate, which is the main reason for the inductive reactance in the EIS [17, 18]. However, this intermediate is prone to react with oxygen to obtain FeOOH and Fe3O4, which will lead to the gradual disappearance of the inductive reactance arc in the low-frequency region with the increase of immersion time [19].

Within the immersion time of 168 h, the capacitive reactance arc radius of the impedance spectrum gradually increased with the increase of pH value. Table 3 shows that Rp value was also increased with the increase of pH value. Above results indicated that the corrosion resistance of the sample in the neutral environment is significantly greater than that in the acid environment.

Table 3 also shows that at the same pH, with the increase of corrosion time, the polarization resistance of the sample first illustrated a decrease and then overall increase trend. This may be because, in the corrosion process, the corrosion product film is gradually formed on the sample surface and will completely cover the sample surface to play a protective effect on the substrate. Therefore, with the increase of immersion time, the corrosion resistance for the initial immersion 36 h in the acid corrosion simulation solution was relatively small, while that was gradually increased after 36 h. However, in the neutral environment, the corrosion rate of the sample is relatively low, and until 96 h for immersion, the formed corrosion products start to act a protective role on the sample, and the corrosion resistance will be increased accordingly. And in the conditions of immersion for 168 h and pH of 7.5, the polarization resistance reached the maximum of 742.62 Ω, which is consistent with the above polarization test results

In summary, with the increase of immersion time, the corrosion resistance of the sample was increased to some extent, which is associated with the formation of corrosion product film on the sample surface. However, with the decrease of pH, the corrosion resistance was decreased significantly, which is mainly because that in the acid conditions, the presence of large amounts of hydrogen ions and erosive chloride ions will result in stability decline of the corrosion product film. Thus, the protection of base material performance is limited.

3.3 Microstructure of Electrochemical Corrosion Specimen

The electrochemical corrosion sample was immersed in the corrosion simulation solution at different pH values (200 times the concentration of the original corrosion simulation solution) for 168 h, and the microscopic morphology was observed by SEM as shown in Fig. 4.

Fig. 4
Three micrographs, a to c, display the surface structure of the samples, which are immersed in solutions with p H 7.5, p H 5.5, and p H 3.5, respectively.

The SEM of the samples after immersion at different pH values for 168 h

Full size image

Figure 4a shows that after immersion for 168 h at pH of 7.5, the sample surface was covered with a dense passivation film, which has a certain protective effect on the substrate material. There were some corrosion products on the surface of passivation film, which were distributed in clusters or flocculent state, as an indicator of α-FeOOH [20]. Related studies [21] have shown that stable α-FeOOH can effectively prevent water, oxygen and other corrosive substances from reaching the matrix-rust interface, thus reducing the corrosion rate of the matrix material. Figure 4b shows that there were white flocculent and minor needle-like materials distributed on the sample surface after immersion in the corrosion solution with pH value of 5.5 for 168 h, as an indicator [20, 22] of major α-FeOOH and minor γ-FeOOH. Compared with Fig. 4a, the corrosion products formed on the sample surface were more than that at pH of 7.5 after immersion for 168 h, which indicated that in the acid environment, the corrosion reaction between the sample and the corrosion simulation solution was more severe.

Figure 4c reveals that the corrosion products were in the main form of needle-like states and minor form of flocculent state for the samples immersed in the corrosion simulation solution at pH of 3.5 for 168 h. It suggested that the corrosion products are mainly γ-FeOOH and minor α-FeOOH with cross distribution [20, 22, 23], and there was tiny crack on the passive film.

3.4 XRD and EDS Analyses of Corrosion Products

The the surface corrosion products were analyzed by XRD and EDS, and the results are shown in Fig. 5 and Table 4.

Fig. 5
A line graph plots intensity versus two theta. The lines are plotted for p H 7.5, p H 5.5, and p H 3.5. All the lines depict a fluctuating pattern with sharp spikes at frequent intervals.

The XRD of the corrosion products of samples after immersion at different pH values for 168 h

Full size image
Table 4 The EDS results of the corrosion products of samples after immersion at different pH values for 168 h(Wt%)
Full size table

Figure 5 shows that, after immersion at different pH values for 168 h, the main corrosion products on the sample surface were Fe2O3, Fe3O4 and FeOOH, with tiny Fe peaks. And with the change of pH value, the phase of the corrosion products did not change significantly. The presence of Fe peaks in the rust layer may be because of the thin rust layer, which will enable X-rays to penetrate the rust layer and reach the inner matrix. Combined with the microscopic morphology of corrosion products in Fig. 4, it was speculated that there are mainly two kinds of corrosion products, namely, γ-FeOOH and α-FeOOH.

Table 4 demonstrates that the corrosion products were mainly Fe and O elements, with minor S elements, which is consistent with the XRD result of iron oxides and iron hydroxide in Fig. 5. With the decrease of the pH, the O element content in the corrosion products gradually declined. This may be because the sample had a relatively faster corrosion rate in acid environment than that in neutral environment. Therefore, the faster formation of a layer of corrosion product film on the sample surface will help isolated from the corrosion solution, thus to prevent oxygen to enter.

4 Conclusion

The electrochemical corrosion performance of 20MnTiB high strength bolt was studied by simulating the humid climate in Chongqing. The results were shown that:

Under the same corrosion time, the corrosion current was decreased with the increase of pH, the arc radius of EIS capacitive reactance showed gradually increasing trend, and the polarization resistance value Rp was also increased with the increase of pH value.

At the same pH, the corrosion current was decreased with the increase of corrosion time, and polarization resistance presented first decrease and then the trend of overall increase.

The main corrosion products are mainly consist of elements of Fe and O, followed by C and S, and other elements that existing in the corrosion simulation solution. The phase of the corrosion products did not change significantly with the change of pH value, while with the increase of the corrosion time, the corrosion products transformed from the original FeOOH to the more stable Fe2O3 and Fe3O4.

All the results demonstrated that after immersion at different pH values for 168 h, the sample corrosion resistance was improved, and was significantly increased with the increase of pH value, which is closely related to the corrosion product film on the sample surface.