Flow Accelerated Corrosion and Erosion-Corrosion Behavior of Marine Carbon Steel in Natural Seawater

Setup and test hardware

The submerged impact jet system used for FAC and erosion-corrosion testing is shown in Fig. 11a. The test setup consists of a centrifugal pump, a pressure gauge, a flow meter and a water tank. The configuration of the unducted centrifugal impeller in the pump could allow the passage of sand particles without clogging the pump. The inside view of the water tank is shown in Figure 11b. The water reservoir contains a nozzle, a sample holder, a working electrode (WE), a saturated calomel electrode (SCE) and two counter electrodes (CE). The SCE was used as the reference electrode (RE). The working electrode was EH 32 carbon steel which is commonly used to build ships and ocean rigs. The chemical compositions (% by weight) of EH 32 steel are C 0.12, Si 0.3, Mn 1.32, Cr 0.03, Ni 0.01, Cu 0.02, P 0.016, S 0.003 and balance Fe. A copper wire was glued to the back of the WE using copper foil tape to make the electrical connection. The WE was mounted in an epoxy resin with an exposed surface of 10 × 10 mm². The microstructure of steel is mainly composed of ferrite and pearlite (as shown in the optical micrograph of Fig. 11c). The inner diameter of the nozzle is 10mm, and the distance between the nozzle outlet and the WE is 20mm. The bottom of the tank is made of four angled steel plates, ensuring that the sand particles could be recirculated back into the flow loop after ejection from the nozzle outlet.

Natural seawater collected from Bohai Bay (China) was used as the test solution. The main ion content (g L⁻¹) of natural sea water are Cl⁻ 17.09, Na⁺ 9:45 a.m. SW₄²⁻ 2.20mg²⁺ 1.06, Ca²⁺ 0.03K⁺ 0.26, HCO₃⁻ 0.13, Br⁻ 0.03. The pH of natural seawater has been measured at around 8.21. The test system was placed in an air-conditioned room, which could maintain the solution temperature at 18 ± 5°C for all tests. Silica sand particles (Fig. 11c) which were sieved through a 40 mesh sieve were used for pure erosion and erosion-corrosion tests. The average diameter of sand particles is about 430 μm.

FCC testing

FAC tests were performed in pure natural seawater. The WEs were tested at four controlled flow rates of 4.7, 14.1, 23.6 and 37.7 L min⁻¹, respectively. The average flow velocities (you) of the fluid at the outlet of the nozzle could still be calculated as follows:

where Q is the flow, D is the inside diameter of the nozzle (0.01 m). The calculated average flow velocities are 1, 3, 5 and 8 ms⁻¹ at each flow. The test duration at each flow rate was 12 h. During the FAC test, the WE, RE and CE were connected to a CS 350 electrochemical workstation (CorrTest, Wuhan China) for corrosion rate measurement. The WE corrosion rate was measured every 0.5 h using the Linear Polarization Resistance (LPR) method. The potential sweep range of the LPR measurement was -10 to +10 mV relative to the open circuit potential. The sweep rate was 0.1 mV s⁻¹. Subsequently, the instantaneous corrosion rate of the WE could be calculated as follows:

where I_correct is the corrosion current, B is the Stern-Geary coefficient which is adopted as 26 mV according to ref. ^29.33, R_p is the adjusted linear polarization resistance, CR is the instantaneous corrosion rate (mm y⁻¹), M is the molecular weight of iron, A is the area of ​​the WE, not is the number of electrons transferred, F is the faradic constant and ρ is the density of steel. Fluid shear stress induced steel loss is reported to be negligible in natural seawater without sand particles.². Accordingly, the average value of CR during 12 h of FAC test was taken as the general rate of pure corrosion ((dot W_{{{mathrm{c}}}}0})) in this work. FAC tests at each flow velocity were repeated three times to ensure repeatability.

After the 12 h FAC test, the WE was removed from the sample holder and immediately photographed by an EOS digital camera (Canon, Tokyo Japan). Then, the compositions and morphologies of the rust layer formed at different flow velocities were characterized by inVia Raman spectroscopy (Renishaw, Leeds UK) and EM-20AX Plus SEM (Coxem, Daejeon Korean). The local compositions of the rust layer were further analyzed using EDS analysis. After the characterization of the rust layer, the corrosion products on the WEs were cleaned using the solution suggested in ASTM G1-03. The local 3D profiles of the corroded areas were then examined by an OLS 5000 infinity microscope (Olympus, Tokyo Japan).

Pure erosion tests

Pure erosion tests were performed in natural seawater containing 1% sand particles (by weight). The WEs were tested at four different fluid flow velocities (1, 3, 5, and 8 ms⁻¹) and the duration of the test of each speed was 12 h. During the pure erosion test, a cathodic potential of -850 mV (vs SCE) was applied to WE to prevent corrosion from occurring. Pure erosion tests at each flow velocity were repeated three times to ensure repeatability.

After the pure erosion test was completed, the surface morphology of the WE was examined by the digital camera and characterized by SEM. Local 3D profiles and surface roughness of WEs were also measured by the infinity microscope. The WE weight loss induced by pure erosion was measured by an AUW 320 balance (Shimadzu, Kyoto Japan) with an accuracy of 0.01 mg. The general rate of pure erosion can be calculated as follows:

where (point W_{{{mathrm{e}}}}0}) is the pure erosion rate (mm y⁻¹), ∆m is the measured weight loss and Δyou is the duration of the test (12 h).

Erosion−corrosion tests

The conditions used for the erosion−corrosion tests were totally the same as those used for the pure erosion tests. The WEs were also tested in natural seawater containing 1% by weight sand particles at four different flow velocities (1, 3, 5 and 8 ms⁻¹) for 12 h. However, no external cathodic potential was applied to the WE during the erosion-corrosion tests. The WEs were simultaneously subjected to erosion and corrosion. As with the FAC tests, the instantaneous corrosion rates of the WEs were measured every 0.5 h using the LPR method. The general corrosion rate ((point W_{{{mathrm{c}}}})) could also be calculated from the mean value of the CR. Erosion-corrosion tests at each flow velocity were repeated three times to ensure repeatability.

After the erosion-corrosion tests, the WEs were examined by the digital camera, and the localized erosion-corrosion morphology was characterized by SEM. Corrosion products generated at different flow velocities were analyzed by Raman spectroscopy and EDS analysis. The local 3D profiles of the WEs were obtained after elimination of the corrosion products according to the ASTM G1-03 standard. Subsequently, WE weight losses were measured by balance and total steel loss rates ((point W_{{{mathrm{t}}}})) at different flow velocities were calculated according to Equation. 5.

It is known that the total loss of steel induced by erosion-corrosion can be expressed as follows:

where (point W_{{{mathrm{c}}}}) and (point W_{{{mathrm{e}}}}) are respectively the corrosion component and the erosion component. (dot W_{{{mathrm{c}}}}0}) and (point W_{{{mathrm{e}}}}0}) are the pure corrosion rate and the pure erosion rate, respectively, which could be obtained from the FAC and pure erosion tests. (point W_{{mathrm{c}}}}^{{{mathrm{e}}}}) and (point W_{{mathrm{e}}}}^{{{mathrm{c}}}}) are the additional steel losses induced by erosion-enhanced corrosion and corrosion-enhanced erosion, respectively. The (point W_{{mathrm{c}}}}^{{{mathrm{e}}}}) and (point W_{{mathrm{e}}}}^{{{mathrm{c}}}}) could be calculated and the interaction between erosion and corrosion could be further investigated according to the following equations.

Since three identical tests were carried out for each condition, all the components of the loss of steel by erosion−corrosion ((dot W_{{{mathrm{c}}}}0}), (point W_{{{mathrm{e}}}}0}), (point W_{{{mathrm{c}}}}), (point W_{{{mathrm{e}}}}), (point W_{{mathrm{c}}}}^{{{mathrm{e}}}}), (point W_{{mathrm{e}}}}^{{{mathrm{c}}}})) presented in this work are the mean values ​​of the three repeated tests.