Abstract
The passivity degradation of carbon steel can be characterised by its low resistance to breakdown, high passive current density and retarded passivation. It can be caused by tensile stress well below the yield strength of steel. This effect is significantly enhanced when dissolved hydrogen is present in steel. The passivity degradation is found to be related to formation of a passive film containing more defects.
Introduction
Owing to the protection of passive films, many metallic materials are applicable in aggressive environments. Compared with the passive films on stainless steels, few investigations have been carried out on the passive film on carbon steels,1,2 despite the fact that carbon steels are the most commonly used metallic structural materials. The passivity of steel is affected by various factors related to environmental and material aspects. Hydrogen generated from electrochemical reactions is a primary contributor to many kinds of corrosion related failure. Recent research indicated that hydrogen produced can enter the passive film. The ingress of hydrogen leads to passivity degradation,3–5 which has been well demonstrated by low pitting resistance, high passive current density and retarded passivation. The hydrogen induced passivity degradation is related to certain changes in film composition and structure. A hydrogen doped passive film will have a high defective film structure and more M–OH bonds, which are the weak points to be attacked by chloride or other harmful species.5,6
During service, engineering structures will be subjected to loads of various kinds. The maximum designed stress is normally well below the yield strength of material. Besides, residual stresses may be introduced in the fabrication processes. An experimental research 7 showed that tensile stress below the yield strength might reduce the time required for pit initiation in a type 316L stainless steel immersed in a near neutral 0.5M NaCl solution at 50°C. Increasing tensile stress would reduce the critical chloride concentration for the breakdown of a passive film on 304 stainless steel. 8 Furthermore, the passivity degradation could be related to the formation of a passive film containing more defects and would be significantly enhanced when hydrogen and tensile stress are present simultaneously. 7
Although the chemical compositions of materials and passive films are very different for carbon and stainless steels, the phenomena and mechanisms for the H induced passivity degradation are quite similar for the both steels.3,6–8 Until now, no experimental investigation has been reported on the passivity degradation in carbon steel caused by tensile stresses and combinations of hydrogen/tensile stresses, despite the fact that carbon steels are the most commonly used metallic structural materials. This is the topic of the current investigation.
Experimental
The test material was the type A516-70 steel sheet under hot rolled condition with chemical composition of (in wt-%) ∼0.27C, 0.85-1.2Mn, 0.15-0.4Si, < 0.035P and < 0.035S. The mechanical properties of the steel were σy = 374 MPa, σU = 548 MPa and EL = 32%. The tensile specimens were machined from the steel sheet as per ASTM E8M-13A. The width and thickness in the gauge length of specimen were 10 and 1.8 mm respectively.
The electrolyte used in the electrochemical measurements was 0.5M NaHCO3+0.01M NaCl. The solutions were prepared with analytical grade reagents and distilled water. A traditional three-electrode corrosion cell was employed, wherein a saturated calomel electrode (SCE) was used as the reference electrode and a platinum wire as the counter electrode. The tensile specimen was used as the working electrode. Before electrochemical tests, the test surface of working electrode was wet ground with SiC papers up to 600 grit, then cleaned by rinsing with distilled water and degreasing with acetone and finally dried up by air gun.
The specimens were divided into four groups:
stress free and without H charge
H charged but free of applied stress
non-charged but subjected to tensile stress of different levels
H charged and subjected to the tensile stress.
A homemade tensile facility was employed to apply tensile load onto the specimens. The details of the tensile facility have been described elsewhere.7,8 The tensile stresses applied in the electrochemical measurements were respectively 170 MPa (45% σy) and 240 MPa (64% σy). Before the electrochemical measurements, the non-charged specimens were cathodically polarised at − 0.9V(SCE) in the test solution for ∼0.25 h to remove any air formed surface oxide film and then passivated at − 0.1 V(SCE) for 1.5 h under either stress free or stressed condition. All the electrochemical measurements were performed in ambient air at room temperature. The H charged specimens were prepared with the following procedures: the specimens were first cathodically charged in 0.1M NaOH+150 ppm As2O3 aqueous solution at 80 A m− 2 for 1 h, then rinsed quickly using distilled water. The H charged specimen was then immediately transferred into the test solution to perform the electrochemical measurements following the same procedures used for the non-charged specimens. The electrochemical measurements included the following:
the response of open circuit potential to the application of tensile stress
the transient current response under potentiostatic control [ − 0.1V(SCE)] to the application of tensile stress
the current decay due to passivation [E = − 0.1 V(SCE)]
Mott–Schottky measurement.
Mott–Schottky measurements were conducted at a constant frequency of 1000 Hz. The potential range for the Mott–Schottky plots was from − 0.4 to 0.1 V(SCE), wherein the polarisation was applied in the anodic direction at the rate of 0.002 V s− 1. The peak to peak amplitude of the ac signal was 0.01 V. Each test was conducted in duplicate under various conditions.
Results and discussion
Passive film breakdown caused by tensile load
The influence of the applied tensile stresses on the open circuit potential E corr of A516-70 steel is shown in Fig. 1. At the beginning, the open circuit potential increased with the immersion time slowly, owing to the formation of a deposit layer or film on the surface of the electrode. After ∼1530 s, E corr dropped suddenly upon the application of the tensile stress of 240 MPa and thereafter recovered slowly. The maximum magnitude of E corr drop was about 53 ± 5 mV when applied stress was 240 MPa, and was 33 ± 4 mV when the applied stress was 170 MPa. The potential response to the action of tensile load is slimier to those observed in the electrochemical noise signal produce by pit initiation. 4 Therefore, the drop of open circuit potential was likely related to the localised breakdown of passive film caused by the elastic deformation of working electrode, and the potential recovery was possibly a result of the repassivation.
Open circuit potentials versus time for A516 steel specimens in 0.5M NaHCO3+0.01M NaCl solution
Effect of hydrogen on passive film breakdown induced by tensile stress
Effect of passive breakdown was also demonstrated by the current responses, measured under potentiostatic control (E = 0.1 V(SCE)], to the action of tensile stress, as shown in Fig. 2. A sharp current peak caused by passive film breakdown was observed immediately after the tensile stress was applied, followed by a current decay due to repassivation. If the amplitude of applied tensile stress was raised from 170 to 240 MPa, the peak current value would increase by about 40-100%. The H charge would increase the current response by almost an order of magnitude when the amplitude of applied stress was same.
Current response under potentiostatic control [E = − 0.1 V(SCE)] to application of tensile stress of a non-charged and b H charged A516 steel (H charging condition: 80 A m− 2 h− 1)
Thermodynamics analysis and experimental investigation indicated that the effects of hydrogen charge and applied tensile stress, as well as their combinations, on the kinetics of anodic dissolution of carbon steel with a film free surface were very limited. 9 Experimental evidence also indicated that applied tensile stress and hydrogen charged into carbon steel could significantly reduce the resistance of passive film to breakdown. 10 Therefore, the high current peak was more likely to result from, at least partially, a large bare surface area generated by passive film breakdown, implying that H charge could reduce the rupture ductility of passive film.
Effects of hydrogen and tensile stress on passivation
The effect of tensile stress and H charge on the passivation of a film free surface is depicted in Fig. 3. The specimens were cathodically reduced at − 0.9 V(SCE) in the test solution for ∼0.25 h to remove any air formed surface oxide film and then anodic polarisation [ − 0.1 V(SCE)] was imposed. The anodic current density over the electrode surface decayed with time due to passivation. The retarded passivation due to tensile stress was insignificant but still observable (Fig. 3a). The effect of H charge was more remarkable than that of tensile stress (Fig. 3b). A steady state of passivation was observable in the non-charged specimens after an exposure of 1 h, while the steady state was not yet fully reached in the H charged specimen after it was anodically polarised for 1.5 h. In accord with the measurements in Fig. 3, H charge would result in a remarkable increase in passive current density, indicating the passivity degradation caused by hydrogen. This is consistent with the previous findings7,8 and will be discussed in the following section.
Effects of hydrogen charge and tensile stress on passivation of A516 steel [E = − 0.1 V(SCE), 0.5M NaHCO3+0.01M NaCl; in b, current response of non-charged electrode without applied stress is presented for comparison]
Furthermore, H charge could also enhance the effect of tensile stress on the passivation, suggesting that the combined effect of H charge and tensile stress cannot be formulated with a linear superposition model, i.e. certain synergistic effect may exist.
Effects of hydrogen and tensile stress on structure of passive film
The Mott–Schottky analysis was performed for the non-charged specimens without and with the application of tensile stress at different levels. The Mott–Schottky plots (the plots of C − 2 versus E in Fig. 4) exhibited characteristics of n type semiconductor, i.e. the positive slopes within the passive potential range of steel [from − 0.2 to 0.1 V(SCE)]. Hydrogen and tensile stress would not change the nature of semiconductor film. The slope of Mott–Schottky plots of an n type semiconductor is formulated by:
Mott–Schottky plots of passive films on a non-charged and b H charged A516 steel electrode subjected tensile stress of different levels
Effects of hydrogen and tensile stress on donor density in passive films (error bars represent difference of duplicate measurements)
Possible mechanisms for passivity degradation induced by tensile stress and hydrogen
Experimental evidence12–14 indicated that the tensile stress higher than the yield strength of steel would enhance the pitting susceptibility markedly, as a result of mechanical rupture of the passive film caused by the slip step emergence. The experimental observations in this work showed that even a tensile stress well below the yield strength (∼45% σy) was still able to degrade the passivity. The passive films on carbon steel in the test solution mainly comprise ferrous and ferric oxide or hydroxide. 15 The crystalline oxides in passive film contain less defects than amorphous hydroxides.2,3 The Mott–Schottky and X-ray photoelectron spectroscopy measurements 16 indicated that hydrogen incorporation could retard the formation of oxide film and led to the formation of a passive film containing more hydroxides. In accord with the point defect model developed by Chao et al., 17 the dissolution rate of passivated metal would be governed by diffusion of point defects within the passive film, suggesting an increase in passive current density with increasing donor density. Higher passive current densities of H charged electrodes were observed by other researchers.7,8 It was attributed, at least partially, to the summation of the oxidation of hydrogen atoms migrating out of the passive film and the accelerated diffusion and dissolution of cations (Fe2+ and Fe3+) and anions (O2 − ) caused by H charge.
Tensile stress has a similar, but less powerful, impact on structure of passive film (Fig. 5). Actually, the anodic dissolution during the ‘uniform’ corrosion is non-uniform in microscale and prefers to occur at some active locations, such as the areas around surface defects. The hydrogen dissolved in steel is likely to segregate around the crystalline defects near the surface region, which can cooperate with external tensile stress to make these defective spots more unstable.18–20 The influence of stress and hydrogen can be ascribed to the following factors. First, a hydrogen containing film may be prone to the damage under tensile load due to its higher defect density. It was observed that passive films containing more hydroxides and/or defects had lower rupture ductility. 21 Second, thermodynamic analysis has indicated that the self-diffusion rates for cations/anions and/or vacancies decrease under compressive stress, but increase with tensile stress.22,23 Thus, as indicated by Fig. 3, the synergistic effect of hydrogen and tensile stress on the passivity may result from the combination of a passive film containing more point defects formed on an H charged electrode surface9,24,25 and the accelerated diffusion of the point defect under the tensile load. 26 Third, the tensile stress may promote the hydrogen reduction in corrosive medium. 23 Finally, hydrogen atoms tend to concentrate at the zones where tensile stress is present, 9 thereby further increasing the likelihood for the hydrogen incorporation into the passive films and reducing the stability of passive films.
Furthermore, the applied stress may increase the activity and/or amount of active sites near the metal surface region. 27 Even under the tensile stress lower than the yield point of a bulk material, some weak points on the surface layer may experience microplastic deformation. As a result, more anodic active sites will form on the metal surface, which enhance the anodic dissolution eventually. If the tensile stress is applied after the formation of passive film, the localised breakdown of the surface oxide might occur because of the mismatch deformation between the metal and its surface oxide film. It is known that, during passivation film formation, a compressive stresses shall build up in the oxide due to the volume expansion and induce a preferential oxide orientation and structure to reduce the accumulated stress. Under an external tensile load, the orientation and structure of passive film would be changed to match new stress level, possibly leading to the localised oxide breakdown and repassivation, especially at some weak spots.
Implications in engineering application
The experimental findings imply a potential role of hydrogen in stress corrosion cracking that is dominated by film rupture mechanism. Under certain cases, e.g. exposure to concentrated aqueous solution of carbonate–bicarbonate or concentrated NaOH, stress corrosion cracking of carbon steel is dominated by the film rupture mechanism. As has been observed in stainless steel, 14 hydrogen induced passivity degradation may accelerate cracking via increasing the frequency of film rupture events due to lower film rupture ductility and high dissolution rate during repassivation.
Residual stresses can be introduced into engineering components during manufacture processes. It has been demonstrated by a STEM observation that the resistance of passive film that is subjected to tensile stress is lower than that subjected to compressive stress. 28 Hydrogen dissolved in steel tends to concentrate at the zones where tensile stresses are present. As such, the passivity degradation in the area where tensile residual stresses are present is likely to be more noticeable.
Conclusions
Similar to the hydrogen charged into steel, the application of tensile stress with a level well below the yield strength of steel can degrade the passivity of carbon steel, which is characterised by a film structure containing more defects, the low resistance to breakdown and the retarded passivation. The passivity degradation can be markedly enhanced when hydrogen and tensile stress are presented simultaneously, due to the combined effects between the two factors.
Acknowledgements
This project was supported by a Strategic Project grant of the Natural Sciences and Engineering Research Council of Canada.