Abstract
The effects of ultrasound on the initiation, growth and repassivation behaviours of pitting corrosion of SUS 304 stainless steel in NaCl aqueous solution were investigated by applying constant current density, constant potential and polarisation. Lower potentials at constant current and higher currents at constant potential responded to ultrasound applied in the passive state. Higher pitting potential was observed when ultrasound was applied from the passive state to the noble direction of potential. Ultrasound also promoted the repassivation of pits.
Highlights
The ability of ultrasound to inhibit the initiation, growth and repassivation of stainless steel in NaCl solution was confirmed.
The mechanism underlying the suppressive effect of ultrasound on the corrosion and the enhancement effect of ultrasound on the corrosion resistance of stainless steel was investigated.
Ultrasound was shown to improve the pitting potential of stainless steel in solution.
Introduction
Pitting corrosion occurs at the surface of stainless steel in chloride solutions. This type of corrosion is generally initiated from weak points in the passive film of the steel, near carbides, sulphides, oxides and other inclusions.1–3 After the initiation of pits, the corrosion reaction is accelerated by hydrolysis reactions,4,5 accompanied by the formation of metallic covers6,7 and corrosion products covers 8 on the corrosion pits. Various approaches have been utilised to enhance the corrosion resistance of stainless steel. For example, quality of the passive film can be improved by the addition of certain elements to the substrate, including Cr, Ni, Mo and N, the production of artificial films on the surface or the removal of the harmful inclusions from the passive film.9–13
Alternatively, in situ assessment has shown that pitting corrosion of stainless steel was largely suppressed when the corrosion products covering the pits were removed by a microprobe connected to an atomic force microscope. 8 In addition, the growth of both the pitting corrosion and the crevice corrosion of stainless steel could be suppressed by applying ultrasound (US) in NaCl aqueous solution.14–19 These findings were likely due to the removal of corrosion products on or within the pits and crevices by the microprobe or ultrasonic cavitation, which can diminish the enrichment of Cl− and H+ ions in the growing pits or crevices. In the case of pitting corrosion, US was found to remove the metallic covers.6,7,19 The ability of US to suppress the growth rate of corrosion pits of SUS 304 stainless steel increased as the cavitation power at a specific frequency increased; this cavitation power was determined by either the output power to the US vibrator or the distance from the vibrator to the steel surface. 18 However, the effect was diminished when the cavitation power was over a specific value.18,19
Although the ability of US to suppress the growth of pitting corrosion of stainless steel is likely associated with the repassivation of the pits, this has not been directly investigated. Furthermore, the effect of US on the initiation behaviour of pitting corrosion has not been assessed. This study therefore assessed the initiation behaviour of pitting corrosion of stainless steel in the absence and presence of US by measuring the pitting potential and applying constant potential or constant current in both the passive zone and the pitting initiation stage. In addition, the repassivation behaviour of pits was investigated in the growth stage. The results show that US affects the pitting corrosion of stainless steel.
Experimental
This study utilised a commercially obtained JIS SUS 304 stainless steel plate (solution treated; thickness, 2 mm) with a chemical composition of Fe–0.05C–0.50Si–1.11Mn–0.026P–0.003S–18.07Cr–8.04Ni (mass-%). The plate was cut into 20 × 20 mm squares, one side of which was polished with no. 800 emery paper. After ultrasonic cleaning in acetone, a central area of 10 × 10 mm on the polished side was used as the test area, and the remained area was sealed with solidified liquid silicone sealant.
Corrosion tests were performed in a corrosion cell with an air saturated 3.5 mass-%NaCl aqueous solution, using a potentiostat/galvanostat (HAB-151, Hokuto. Co.) and a US vibrator (Kaijo Co. 4292C) at a frequency f of 19.5 kHz (Fig. 1). The US vibrator was immersed in a water tank under the corrosion cell. A platinum plate was used as the counter electrode and a saturated calomel electrode (SCE) as the reference electrode. Output power from a controller (Kaijo Co., TA-4021) to the US vibrator was set at P US = 8 (corresponding to 160 W). The specimen was immersed in NaCl aqueous solution in the corrosion cell face down to the US vibrator, and temperature was maintained at 30 ± 2°C with a thermostat. Since the cooling function of the thermostat was not used, the temperature of the solution increased slightly during the application of US; this increase in temperature can be used to calculate the actual cavitation power near the specimen in solution.17,18 The distance from the vibrator to the specimen d was set at 3λ/4 = 57 mm, where λ is the wavelength of sound in water, which was calculated from the wave velocity as 1480 m s− 1. In addition, distance D from the vibrator to the solution surface was set as 3λ/2 (D = 114 mm). Previous studies have shown maximum suppression at these US conditions.18,19 Polarisation was started from E SCE = − 600 mV at the cathode side and then swept to the anode side at a constant increase in potential of 0.33 mV s− 1 under control of the potentiostat.
Apparatus for polarisation and applying ultrasound (US)
To clarify the effects of US on the initiation of pitting corrosion, the anodic current density in solution i c = 0.05 A m− 2 was kept constant from the open circuit state by the galvanostat function, and US was applied midway. The potentials were recorded to investigate the mechanism of metal dissolution. When the potential was swept to E 1 = − 200 mV(SCE), E 2 = 0 mV(SCE) or E 3 = 250 mV(SCE) in the passive zone, US was applied with acontinuous sweep of potential to the pitting corrosion zone (Fig. 2). To test the effects of US on the pitting potential E p when US was applied in the passive zone, US was applied from 0 mV, and from 250 mV, the potentials were also kept constant for 600 s or 1.2 ks, and the sweep restarted to the pitting corrosion zone. When the anodic current density reached i 0 = 0.1 or 20 A m− 2 in the pitting zone, the potential corresponding to the current density was kept constant for 600 s and US was simultaneously applied (Fig. 3). The current densities during this period were recorded and compared without and with US. Moreover, when the anodic current density reached i 0 = 20 A m− 2 or the potential reached 500 mV in the pitting zone, the potential was swept back to the passive zone to determine the protection potential E prot., without or with simultaneously applying US.
Procedures for applying potential holding (PH) and ultrasound (US) from potentials E 1 = − 200 mV(SCE), E 2 = 0 mV and E 3 = 250 mV in passive zone during polarisation
Procedures for applying PH (E 0) and US from different anodic current densities i 0 of 0.1 or 20 A m− 2 in pitting zone during polarisation
Results and discussion
Variations of potential when applying constant current
Figure 4 shows several typical variations of potential of the SUS 304 stainless steel over time when a constant current density i c was applied directly from the open circuit state and maintained for 1.2 ks in 3.5%NaCl aqueous solution. At i c = 0.05 A m− 2 (Fig. 4a), the corresponding potential started from ∼300 mV, followed by a slow increase to ∼400 mV accompanied by potential fluctuations of ∼100 mV. During this period, the passive oxide film grew as initial potential increased, and activation/passivation randomly occurred. At i c = 0.1 A m− 2, the potential during the early half stage, until t = 700 s, was similar to that at i c = 0.05 A m− 2, but thereafter sharply decreased to values lower than 0 mV. This finding suggests that the passive oxide film grew, but then broke to initiate the pitting corrosion, and that after the potential quickly decreased to the active dissolution region, the growth of the corrosion pit stopped. At i c = 0.2 and 1.0 A m− 2, the potential dropped to < 0 mV at a much earlier stage than that at i c = 0.1 A m− 2, which indicates that active dissolution occurred earlier with the stop of the growth of the corrosion pit.
Variation of potential over time by application of constant current densities i c of a 0.05, b 0.1, c 0.2 and d 1.0 A m− 2 from open circuit state, in 3.5 mass-%NaCl solution
Figure 5a and b shows the variation of potential at a constant current density of i c = 0.05 A m− 2 when the US was applied at t = 600 s. This current density was close to the passive current density in the polarisation curve (see Fig. 7). Light reduction in potential occurred in both cases after US was applied. Although this drop in potential was much smaller than that observed in Fig. 4b–d, it is likely due to either partial damage to the passive film or general active dissolution of the film caused by US cavitation. There were no changes, however, at preset current density >0.05 A m− 2 when US was applied (Fig. 5c). This finding was likely related to the early initiation of film damage before applying US.
Variation of potential over time by application of constant current density i c of a, b 0.05 A m− 2 and c 0.1 A m− 2 from open circuit state and application of US at t = 600 s, in 3.5%NaCl solution
Variations of current density at constant potential in passive zone
As described in Fig. 2, when the potential reached 0 or 250 mV(SCE) in dynamic polarisation measurements, the potential was maintained for 600 s without or with simultaneous application of US. Figure 6 shows the variations in current density during this 600 s period in 3.5%NaCl solution. At E C2 = 0 mV, the passive film was stable with fluctuations within 10− 4–10− 1 A m− 2, regardless of the application of US. In contrast, secondary noises of current different from those fluctuations within 10− 4–10− 1 A m− 2, including two instant high values, appeared on the curve when US was applied during this period. This suggests that the passive film was influenced by US cavitation, but this effect was too small to change the passive state. At E C3 = 250 mV, the potential seemed to be within the passive zone, but partial dissolution should have occurred from weak points near inclusions because it is very near the pitting potential (E p = 301 mV; see Fig. 7). In this case, when US was applied, a sudden increase in current occurred initially, followed with a quick decrease to the same level as in the absence of US for ∼100 s and then increase to a higher and stable level. This finding indicated that US promoted the dissolution of metals within this zone.
Variation of current density over time by application of PH (E c) at a 0 mV(SCE) and b 250 mV(SCE) for 600 s during polarisation, without and with application of US
Typical polarisation curves by application of US from b E 1 = − 200 mV, c E 2 = 0 mV and d E 3 = 250 mV to the end of polarisation, compared with one a without application of US
Variations of pitting potential when applying US
Figures 7 and 8 show several typical dynamic polarisation curves at different potentials in the passive zone in the presence and absence of US. The pitting corrosion E p in the absence of US, corresponding to a current density of 1.0 A m− 2, was obtained at 301 mV (Figs. 7a and Fig. 9a, O). All E p measurements were the averages of at least five tests, with careful check of the non-occurrence of crevice corrosion (Fig. 9).
a PH and US at 0 mV for 600 s; b PH and US at 0 mV for 1.2 ks; c PH and US at 0 mV for 600 s, then US to end of polarisation; d PH and US at 0 mV for 1.2 ks, then US to end of polarisation
a applying US (without PH); b applying PH or US from 0 mV
When US was applied from E 1 = − 200 mV to the end of the polarisation, the passive current density increased slightly with much fluctuation (Fig. 7b), but the value of E p showed little change (Fig. 9a, A). The relatively higher current density with greater fluctuation indicates the random partial dissolution and repassivation of the steel. It may be difficult to generate strong and homogeneous film due to the early and continuous US disturbance. In contrast, the current density after applying US from E 2 = 0 mV was much higher and more stable (Fig. 7c), which indicates a much greater steel dissolution rate. Note that this high and stable current density could not be obtained in all tests when applying US from E 2 = 0 mV to the end of the polarisation. It is considered that the selective dissolution of iron and the removal of several inclusions by US cavitation enhanced the formation of a higher quality passive film, which resulted in a higher value of E p than in the absence of US (Fig. 9a, B). According to the sudden increase and decrease in current density when US was applied from E 3 = 250 mV (Fig. 7d), the high E p value obtained was likely due to repassivation of the formed pits rather than selective dissolution of iron or removal of inclusions in the passive zone (Fig. 9a, C). With the consideration of the balance of the damage on the passive film and the repassivation of the damaged sites at different durations of US, the highest value of E p was obtained at E 2 = 0 mV (Fig. 9a, B).
Figure 8 shows several polarisation curves when potential holding (PH) or US from a potential of E 2 = 0 mV was applied. E p did not increase when only PH was applied at E 2 = 0 mV for 600 s (Fig. 9b, B-1). However, when both PH and US were applied at E 2 = 0 mV for 600 s or 1.2 ks (Fig. 8a and b), the current density increased, and then recovered to previous values after discontinuation of PH and US. The selective dissolution of iron and the removal of inclusions during this period should have resulted in a higher E p (Fig. 9, B-2 and B-3) in the presence than in the absence of PH and US. Especially at longer times of PH/US (1.2 ks) (Fig. 8b), the increase in current density resulted in lower passive current values before pitting corrosion, which indicates the formation of high quality passive film. This result also corresponds well to the relatively higher value of pitting potential than at 600 s PH/US. In contrast, the E p value when both PH and US were applied at E 2 = 0 mV for 600 s (Fig. 9b, B-3) was lower than when US was applied from E 2 = 0 mV to the end of polarisation (Fig. 9a (B) and b (B-4)), which should be associated, at least in part, to the shorter time or the lack of accelerated dissolution of metal.
When PH/US was applied at E 2 = 0 mV for 600 s, followed by US to the end of polarisation, a high passive current density with severe fluctuation appeared in the passive zone (Fig. 8c). After the current density increased to a pitting corrosion value of ∼7.0 A m− 2, a sudden drop occurred for a short time, followed by an increase. The decrease in current density implies the repassivation of formed pits, but this repassivation disappeared again due to the continuous increase in potential. In this case, both the accelerated dissolution of metal in the passive zone and the repassivation of pits should have helped enhance the pitting potential of E p (Fig. 9b, B-5). However, when the time of PH/US at E 2 = 0 mV was prolonged to 1.2 ks, followed by US to the end of polarisation, except for the repassivation of pits in the pitting zone, lower current density appeared in the passive zone (Fig. 8d). In addition, severe fluctuation was observed in other tests when US was applied in the passive zone. This phenomenon indicates the formation of compact film, which should result in the highest value of 360 mV (Fig. 9b, B-6). The highest E p should be observed at continuous US with effective dissolution of iron and the removal of inclusions in the passive zone and the repassivation of the pits in the pitting zone.
Figure 10 shows several corrosion pits observed by laser microscopy (Olympus Co., LEXT OLS4000). The pits were produced by dynamic polarisation when the potential was swept from 0 to 250 mV (less than the pitting potential) in the absence and presence of US. Shallow pits were produced even when the potential was lower than the pitting potential. Most pits that formed in the absence of US had metallic covers (Fig. 10a), under which the inclusions are considered to be located. In the presence of US, however, those pits opened (Fig. 10b), suggesting that the US had removed the metallic cover. This may help in the formation of passive films in the pits after reductions in the concentrations of H+ and Cl− . This may explain why application of US from E 2 = 0 mV to the end of the polarisation enhanced the pitting corrosion potential from 301 to >330 mV. Additional measurements and analyses are needed to clarify this issue.
Corrosion pits formed after sweeping potential from 0 to 250 mV a without and b with US during polarisation, observed by laser microscopy
Figure 11 shows the left pit in Fig. 10b observed by SEM, where energy dispersive spectroscopy analysis was carried out on spots A–C. The carbon content (relative value) in spot B at the bottom of the pit was 3.07%, almost the same as that in spot A, where pitting corrosion did not occur (2.84%). In contrast, the carbon content in spot C with a remained metallic cover was much higher (6.70%), which indicates that the inclusions, including most of the carbides at the bottom of the pit, had been removed by US. Although corrosion pits are generally initiated from spots near MnS inclusions, high sulphur content was not detected in either of the pits.
Morphology (SEM) of left pit shown in Fig. 10b and relative carbon contents analysed by energy dispersive spectroscopy
Variations of current density at pitting initiation and growth stage with application of US
Figure 12 shows polarisation curves at constant potential and current densities of i 0 = 0.1 and 20 A m− 2 in the absence and presence of simultaneous US. The current density of i 0 = 0.1 A m− 2 was low enough to simulate the initiation stage of pitting corrosion. In the absence of US, the current density increased, but the density was lower in the presence of US. This indicates that US suppressed the initiation of pitting corrosion, similar to that observed during the growth stage of pitting corrosion.14–19 This result can be more clearly seen in the measurement of accumulated electric charge during the PH, with the value of 2.8 kC m− 2 in the absence of US at i 0 = 0.1 A m− 2 being reduced to 1.4 kC m− 2. The accumulated electric charge is an indicator of the dissolution of metal ions into solution during this period.
Variation of current density over time by application of PH (E 0), corresponding to anodic current densities i 0 of a, a’ 0.1 A m−2 or b, b‘ 20 A m−2, with simultaneous US in pitting zone during polarisation
The suppressive effect of US on pitting growth at current densities higher than 0.1 A m− 2 was the same as in former reports.14,16,17 At i 0 = 20 A m− 2, the accumulated charge changed from 54 to 14 kC m− 2 (Fig. 12b and b‘). Ultrasound cavitation removed the corrosion products on or in the pits, as well as sometimes the removal of the metallic covers, the dilution of condensed Cl− and H+ in the pits and the retardation of the growth of corrosion.
Repassivation of pits with US
Figures 13 and 14 show the repassivation behaviour of the specimen when the potential was swept back from the potential corresponding to the current density of i 0 = 20 A m− 2 or the potential of 500 mV in the polarisation at different US output powers of P US = 2 (40 W), 4 (80 W), 6 (120 W) and 8 (160 W). Similar results were obtained under these conditions. In the absence of US (Fig. 13a), the repassivation potential (protect potential E prot.) appeared at − 115 mV with a necessary accumulated charge of 10 kC m− 2, while the potential increased with increase in output power. At P US = 8, the repassivation potential became 317 mV with a much smaller charge than 0.7 kC m− 2. The accumulated charges and protection potentials are shown in Fig. 15. It is clear that US had promoted the repassivation of the steel. Furthermore, higher US power resulted in a higher repassivation potential and lower accumulated charge. This was due to the removal of corrosion products or metallic covers, which promoted the repassivation of the pit inner surface by reducing the concentrations of Cl− and H+ within the pits.
Polarisation curves when sweeping potential back from potential E 0, corresponding to anodic current density i 0 of 20 A m−2, with simultaneous US at different powers P US
Polarisation curves when sweeping potential back from 500 mV(SCE) with simultaneous US at different powers (P US)
a variation of accumulated electric charge q during repassivation and b protect potential E prot. with simultaneous US at different powers P US
The above results indicate that US affects the formation of passive films and pits in three ways. First, it promotes the general or selective dissolution of passive films. General dissolution results in thinner or more damaged films with lower pitting potential of E p or a potential drop during maintenance of constant current, whereas selective dissolution results in a more compact passive film with higher value of E p, with a greater effect observed in the presence of dissolved iron than chromium. Second, US promotes the removal of inclusions and the repassivation of weak points with inclusions in passive films, which increased the value of E p. Third, US removes the metal covers and corrosion products on pits, followed by the removal of inclusions in pits, which promoted the re-formation of passive films in pits14,16,17 and increased the value of E p. The first and second mechanisms are applicable at a relative early stage of passivation and corrosion, while the third occurs during the growth stage of pitting corrosion, which resulted in higher pitting potentials of E p and higher protective potentials of E prot. The ability of US to effectively alter corrosion conditions may make it possible to develop a useful method to enhance the pitting potential of stainless steels.
Conclusions
This study evaluated the effects of US (19.5 kHz) on the initiation, growth and repassivation behaviours of pitting corrosion of SUS 304 stainless steel in 3.5%NaCl aqueous solution at constant current density, constant potential and polarisation. The obtained results can be summarised as follows.
Ultrasound resulted in a drop in potential of SUS 304 stainless steel at constant current of 0.05 A m− 2, which indicates the dissolution or damage at the surface. Application of US in the passive zone resulted in a higher current density on the SUS 304 steel, and application of US from the passive state to the end of polarisation resulted in a higher pitting potential. Application of US in the pitting corrosion zone increased the repassivation potential, with higher US power resulting in a higher repassivation potential. These findings confirm that US suppresses the initiation and the growth behaviours of pitting corrosion of SUS 304 steel. It is considered that the former one is attributed to the formation of a condense film and the latter one is due to the promotion of the repassivation of pits.
Acknowledgements
The author is very grateful to Mr T. Abe, Mr K. Nakagama, Mr Y. Shibatani, Mr S. Fujii, Mr R. Hori and Mr K. Yamada for their assistance of the experiments.