Materials Chemistry
Corrosion Mechanism of Q235A under 3.5% NaCl Salt Spray
2020 Volume 61 Issue 12 Pages 2342-2347
Details
2020 Volume 61 Issue 12 Pages 2342-2347
Low-carbon steel Q235A is the main material used in marine construction, but it experiences corrosion because of salt spray and seawater exposure. In this study, we investigated the corrosion performance of Q235A in the presence of 3.5% NaCl salt spray. We characterized the microstructure of the material using electron backscattered diffraction, scanning electron microscopy, and specimen weight loss. The phases and morphologies of the corrosion products were then analyzed after salt-fog corrosion testing. The degree of coverage and compactness of surface corrosion products gradually increased from the 24 to 96 hours in the salt spray environment. Under short-term corrosive conditions, surface corrosion products were primarily composed of a loose, porous layer of flocculent γ-FeOOH. With increasing time, the corrosion products gradually changed to a lump-like and block-like dense structure of β-FeOOH and α-FeOOH. The results of electrochemical polarization tests of the corroded steel samples showed that the γ-FeOOH might accelerate corrosion, and that β-FeOOH and α-FeOOH inhibit corrosion.
Fig. 6 Corrosion micrograph of Q235A steel, the corrosion time of (a), (b), (c), (d) were 24, 48, 72 and 96 h, respectively.
Low-carbon steel Q235A, which has good plasticity, hot workability, and low cost,1) is widely used in engineering. Following hot rolling, Q235A steel plates and pipes are suitable for the manufacture of various welded structural parts and machine parts.2,3) However, the corrosion mechanism of the steel varies with the state of the service environment.4,5) Consequently, extensive research has been conducted on the microstructure, corrosion, and protection of this material. For example, Wang et al.6) examined solid-phase diffusion bonding between Q235A and austenitic stainless steel AISI304. They found that brittle compounds segregated to improve the performance of diffusion-welded joints when the welding temperature was controlled at 850°C. In another example, Hu et al.7) studied the corrosion behavior of Q235A steel in reverse-osmosis water treatment equipment. They reported that corrosion under these service conditions is controlled by the diffusion of oxygen and corrosion products, in which a conductive Fe3O4 layer forming accelerates the corrosion process. Chen et al.8) studied a series of inhibitors that effectively retard the corrosion of Q235A at high concentration HCl. The compound H-NMR (D6-DMSO) shows 95.5% inhibition efficiency at 100 mg/L HCl. Li et al.9) studied Q235A steel corrosion induced by chloride ions under stray current conditions and constructed a model based on an artificial neural network to accurately predict corrosion, as well as the corrosion potential and corrosion current density. Jianguo et al.10) studied the polarization curves, corrosion processes, and mechanisms of Q235A steel under dry and wet cycling conditions. They reported the corrosion current of a dry-wet sample was 22.99 µA/cm2, and this was much higher than that of the bare steel (9.3 µA/cm2). To improve the corrosion protection performance of Q235A steel, some researchers have treated steel surfaces with nickel-nanodiamond composite coatings or plasma-clad tungsten carbide particles or by laser cladding ultra-hard alloys.11–13)
The good welding features and mechanical properties (yield strength of 235 MPa, ultimate tensile strength of 410 MPa, and elongation of 23%) of Q235A steel make it a preferred material in the construction of shipping containers. However, in the marine environment, steel materials are constantly exposed to salt spray and seawater, making them prone to salt spray corrosion, reducing the strength and life service of the ship. Consequently, determining the corrosion of Q235A steel in a salt spray environment is important for the design and manufacture of the ship. In this study, we examined the corrosion products over time during a salt spray test and examined the effect of the corrosion products on the corrosion.
The size of the metal samples used in the salt spray experiments was 10 mm × 10 mm × 5 mm, and the individual metal samples were cleaned and weighed to the nearest 0.1 mg before testing. Test samples were cleaned with acetone to remove surface oil, polished with #1200 sandpaper, dried in the dry box, and soaked in pure ethanol.
A corrosion test was conducted based on the ISO 9227-2017 standard using a salt spray tester containing a 3.5% NaCl solution. The metal samples were subjected to salt-fog test for 24, 48, 72, and 96 hours. The pH of the salt-fog solution was 6.4, temperature of the sample chamber was at 35°C, salt spray settlement occurred at 1.0–2.0 mL/80 cm2·h, and relative humidity of the chamber was 70%. After each test, the weight of the individual sample was recorded, and the average weight loss was determined from the four-sample experiment.
2.2 Electrochemical experimentsCorroded steel samples were subjected to electrochemical experiments using the classical three-electrode electrochemical test system where the working electrode was the corroded Q235A steel sample. One side of the test sample was sealed with an epoxy coating, and the other side served as the working electrode with a 1 cm2 working surface area. The reference electrode was a saturated calomel electrode, the auxiliary electrode was a platinum electrode, and the electrolyte was 3.5% NaCl solution. The test was conducted using a CS350 electrochemical workstation (Wuhan CorrTest Instruments Corp., Ltd., Wuhan, China). The test sample was immersed in the test cell for 10 minutes, followed by potentiodynamic polarization at a scan rate of 0.001 V/s. The scan was made in potential from corrosion potential to catholic direction to −0.85 V and then to anodic direction to −0.40 V.
2.3 Weight loss test and microstructure observationFor weight loss experiments, the corrosion products of the retrieved samples were removed chemically by immersion in a specific solution (500 mL HCl + 500 mL distilled water + 3.5 g hexamethylenetetramine) that was vigorously stirred for 10 minutes at 25°C according to ISO 8407.
The microstructure of the Q235A was determined using electron backscattered diffraction (EBSD). The microstructure of the material before and after corrosion was examined using scanning electron microscopy (SEM, JEOL JSM-6510, JEOL Ltd., Tokyo, Japan). The compositions of the corrosion products on the test sample surface were determined by X-ray diffraction (XRD, Rigaku Ultima IV, Rigaku Corporation, Tokyo, Japan). The analysis was performed using Cu-Kα with an operating voltage of 40 kV, operating current of 50 mA, scanning range of 10°–90°, and scanning rate of 2°/minute.
The composition of the Q235A steel used in this study was determined using X-ray fluorescence spectroscopy (Table 1). EBSD analysis was used to determine the test material’s crystal structure. The results confirmed that the steel was mainly a body-centered cubic structure with a small amount of Fe3C. The grain size was asymmetrically distributed in the range of ∼3–35 µm, in which the larger size proportion monotonously decreased (Fig. 1(a) and 1(b)). Ferrite was the main phase of Q235A and the elemental carbon melted in α-Fe to form a second solid solution (Fig. 1(c)). Pearlite crystals were mainly distributed along the material’s grain boundary and exhibited a lamellar and short columnar structure.
Microstructure of Q235A: (a) EBSD grain orientation map; (b) Grain size; (c) Metal microstructure map; (d) Partially enlarged view of (c).
Figure 2 shows the morphology of Q235A steel after the corrosion in a salt-fog environment for different time periods. (a), (b), (c) and (d) in Fig. 2 were the morphology appearances of corrosion for 24, 48, 72, and 96 hours, respectively. After 24 hours corrosion, the surface corrosion was partially covered by corrosion products, and the rust layer was lighter in color than that of the longer exposure period. After 48 hours corrosion, the coverage of the rust layer expanded, becoming thick and darkened in color. After 72 hours corrosion, the coverage area of the rust layer further expanded, becoming dark black with substantial edge corrosion. After 96 hours corrosion, the rust layer completely covered the surface. In this case, the corrosion products were much thicker, and became bright reddish-brown. In summary, the corrosion of Q235A steel in the salt spray environment was started in a heterogeneous manner. The corrosion gradually expanded to cover the entire surface over the course of the experiment, and the degree of corrosion at the edges was pronounced.
Macroscopic morphology of Q235A steel after salt spray experiment.
The weight loss of the Q235A steel test samples subjected to the 3.5% NaCl salt spray corrosion test showed a linear relationship with time (Fig. 3). It has been reported that the rate of weight loss initially increases and then decreases.14) This observation may indicate that the surface corrosion products in the early test stages are thin, loose, and porous, thus allowing accelerated corrosion. The authors of that study suggested that as the salt-fog exposure time increased, the thickness and density of the corrosion product layer increased, hindering further corrosion and decreasing the weight loss from the sample.14)
Corrosion weight loss curve of Q235A steel.
XRD analysis indicated that the XRD peaks of metallic Fe were only found on Q235A before the exposure test (Fig. 4). The peaks of the metallic Fe were still observed on the XRD pattern after the exposure test for 24 and 48 hours, and small peaks of γ-FeOOH were also present. The γ-FeOOH may be assumed as a reducible substance.15) After 72 hours, the corrosion products were mainly composed of β-FeOOH and γ-FeOOH. After 96 hours, the relative amount of γ-FeOOH decreased and α-FeOOH began to form. The α-FeOOH formed is thermodynamically stable in structure, and can work as a protective layer toward the steel corrosion.16)
XRD pattern of surface corrosion products of Q235A.
No obvious cracks appeared on the corrosion products after the 24 hours corrosion on the SEM view in Fig. 5(a), and the thickness that was evaluated from the cross-section view of the corrosion layer was small, with an average thickness of about 4 µm (Fig. 5(b)). After 72 hours, a large number of cracks appeared on the corrosion products (Fig. 5(c)). The thickness of the corrosion layer increased to the average thickness larger than 10 µm, and the more heterogeneous corrosion was seen (Fig. 5(d)). This indicates that the corrosion products after 24 hours were relatively loose, whereas those after 72 hours were dense. The previous research indicated that α-FeOOH was dense and provided a protective property to steel corrosion, and β-FeOOH was a dense structure, whereas γ-FeOOH was property to a loose structure.15–17) Thus, with increasing corrosion time, the surface morphology changed from a loose structure to a dense structure and the corrosion layer became thicker.
Surface morphology under salt spray corrosion: (a) 24 h, (c) 72 h; Cross-section morphology under salt spray corrosion: (b) 24 h, (d) 72 h.
Corrosion micrographs of the Q235A sample after the exposure time of 24, 48, 72, and 96 hours are shown in Fig. 6(a)–(d), respectively. After 24 and 48 hours, the surface corrosion products exhibited a porous, flocculent corrosion layer (Fig. 6(a), (b), white circles). As mentioned, the XRD phase analysis indicated that the corrosion products were γ-FeOOH. For the 72 hours corrosion, the amount of initial flocculent substance was significantly reduced and a cotton-like substance newly appeared (Fig. 6(c), (d), white box). From the XRD results, we conclude that this cotton-like substance is β-FeOOH. Although cracks were formed on the surface as a result of dehydration of the corrosion products, the overall structure was dense. After 96 hours, no flocculent corrosion substance was observed. The corrosion products were primarily block-like and cotton-like substances (β-FeOOH and α-FeOOH, respectively), and the corrosion layer was very dense.
Corrosion micrograph of Q235A steel, the corrosion time of (a), (b), (c), (d) were 24, 48, 72 and 96 h, respectively.
The corrosion potential became negative during the initial stages of corrosion, and then became positive with increased exposure time (Fig. 7). The corrosion potential of the 48 hours corrosion sample was almost equal to that of 72 hours. The flat region of current density at potentials higher than −0.5 V was caused by the set-up of the electrochemical workstation. The corrosion current density increased with the exposure time to 72 hours, and the increasing ratio slowed, and then slightly decreased between 72 and 96 hours (Table 2 and Fig. 8). This indicated that the corrosion rate of the steel was initially fast and then slowed with increased exposure time. These results are consistent with those reported previously in which the rate increased firstly and then slowed with increased exposure time.18,19)
Electrochemical polarization curve of Q235A after 3.5% NaCl salt spray test.
The trend of corrosion current densities (icorr) of Q235A steel with the exposure time.
It has been assumed that the cathode reaction on bare steel was controlled by the diffusion of oxygen in the neutral pH solution, and that the anode reaction was controlled by charge transfer on the steel surface. In initial corrosion, amorphous Fe2+ and Fe3+ hydroxides, including anions, are formed; then, the amorphous compounds gradually change to mainly γ-FeOOH.20) For the 24 h salt spray test sample, because the corrosion products were partially covered, the cathode reactions are assumed to be taken by the oxygen reduction and the reduction in corrosion products from Fe3+ to Fe2+.21,22) For the 48, 72 and 96 h salt spray test samples, because of the increase in the amount of the corrosion products, the cathode reaction was dominated by the reduction reaction of the corrosion products and the corrosion current density was gradually inhibited.
3.6 Corrosion mechanismFrom the results above presented, we divided the corrosion process for this type of steel into the following steps (Fig. 9). The first step is the iron dissolution. Activation by chloride ions in the salt-fog accelerates iron dissolution to Fe2+, Fe(OH)+, and Fe(OH)2. In addition, because the solution on the surface is changed from neutral to weakly acidic, the dissolution of bare iron is increased to produce more Fe2+.23,24) The main reactions in this process are presented in eqs. (1) and (2):
| \begin{equation} \text{Fe} + \text{2H$^{+}$} \to \text{Fe$^{2+}$} + \text{H$_{2}{}\uparrow$} \end{equation} | (1) |
| \begin{equation} \text{Fe$^{2+}$} + \text{H$_{2}$O} \to \text{Fe(OH)$^{+}$} + \text{H$^{+}$} \end{equation} | (2) |
Schematic diagram of the corrosion process.
The second step is the oxidation of dissolved divalent iron ions. In a neutral or weakly acidic environment, the main corrosion product is γ-FeOOH, with a smaller amount of β-FeOOH. It has been suggested that Fe3O4 was formed in the iron corrosion process;25–27) however, we did not observe Fe3O4 in the present study. The main reaction processes are shown in eqs. (3) and (4) as
| \begin{equation} \text{2Fe$^{2+}$} + \text{1/2O$_{2}$} + \text{3H$_{2}$O} \to \text{2$\gamma$-FeOOH} + \text{4H$^{+}$} \end{equation} | (3) |
| \begin{equation} \text{Fe$_{3}$O$_{4}$} + \text{O$_{2}$} \to \text{$\gamma$-FeOOH} \end{equation} | (4) |
The third step is the conversion of iron oxyhydroxide. For the corroded Q235A, because of the presence of a rust layer, dissolved oxygen cannot be the main oxidizing agent. Rather, the corrosion product γ-FeOOH acted as an oxidizing agent, and because of its porous structure, it accelerated the corrosion.28,29) The main reaction processes are shown in eqs. (5)–(7):
| \begin{equation} \text{FeCl$_{2}$} \xrightarrow{\text{H$^{+}$}} \text{$\beta$-Fe$_{2}$(OH)$_{3}$Cl} \end{equation} | (5) |
| \begin{equation} \text{$\beta$-Fe$_{2}$(OH)$_{3}$Cl} \longrightarrow \text{$\beta$-FeOOH} \end{equation} | (6) |
| \begin{equation} \text{$\gamma$-FeOOH} \xrightarrow{\text{H$^{+}$ or O$_{2}$}} \text{$\alpha$-FeOOH} \end{equation} | (7) |
Because α-FeOOH is assumed to consist of smaller particles than γ-FeOOH and β-FeOOH, α-FeOOH works as a dense protective layer to the steel corrosion and thus improves the steel’s corrosion resistance.
In a 3.5% NaCl salt spray corrosive environment, the degree of corrosion of Q235A steel increased with exposure time. During the first 24 hours, corrosion heterogeneously occurred and the corrosion products gradually covered whole surface. After 96 hours, thick and dense corrosion products formed. With increasing salt spray exposure time, the corrosion products on the steel gradually changed from the γ-FeOOH to β-FeOOH and α-FeOOH. The γ-FeOOH formed a flocculent layer, with a loose and porous structure, whereas the β-FeOOH and α-FeOOH were block-like and cotton-like in structure, respectively, and their structures were dense. From the electrochemical polarization, γ-FeOOH produced after 24 hours of salt-fog exposure promoted corrosion of the steel. After 96 hours, β-FeOOH and α-FeOOH corrosion products inhibited corrosion.
This work is supported by the Natural Science Foundation of Jiangsu Province for Universities and Colleges (No. 19KJB430031) and the Nantong Science and Technology Project (No. GY12018032).
