2023 Volume 64 Issue 11 Pages 2629-2636
The effect of corrosion status of zinc coating on the hydrogen absorption behavior was investigated. Dry-wet corrosion test was conducted after applying NaCl solution to the zinc coated steel sheets. The permeating hydrogen during the corrosion test was measured by electrochemical technique. It was found that the hydrogen permeation was not observed at early stage of corrosion while significant hydrogen permeation was observed at middle stage of corrosion. Then, the hydrogen permeation decreased gradually at later stage of corrosion. In order to clarify the mechanism of this behavior, surface morphologies of corroded specimens were analyzed by Electron Probe Micro Analyzer. It was found that the hydrogen permeation was observed when the steel substrate was partially appeared, indicating that galvanic corrosion is related to the hydrogen permeation.
This Paper was Originally Published in Japanese in Zairyo-to-Kankyo 71 (2022) 218–224. The description of the second eq. (1) is slightly modified.
In response to the tightening of CO2 emission regulations and increasing need for collision safety in automobiles, high-strength steels are being developed in order to achieve both weight reduction and high rigidity.1,2) However, new problems of the increase of the delayed fracture susceptibility in addition to the moldability such as the decrease in the life of the mold and the increase in the amount of spring back have become apparent with the increase strength of steel.3) Delayed fracture is a phenomenon in which a material under a static load suddenly fractures by hydrogen entering the material, so-called hydrogen embrittlement.4) It has been reported that hydrogen embrittlement is more likely to occur as the amount of hydrogen increases and when the amount of hydrogen exceeds a certain critical value.5,6) Hydrogen absorption into steel is caused by a corrosion reaction on the surface in atmospheric corrosion environment.7) In the past, delayed fracture due to hydrogen embrittlement of high-strength bolts used in bridges was reported in a number of cases.8,9) Since the strength of automotive steel sheets is now approaching that of high-strength bolts and hydrogen embrittlement due to corrosion during use is a concern, it is necessary to evaluate the hydrogen absorption behavior of high-strength steel sheets under corrosive environments.
It has been reported that the amount of absorbed hydrogen into a steel sheet during corrosion varies depending on corrosion factors such as temperature, humidity and chloride deposition.10) Ootsuka et al.11,12) investigated hydrogen absorption into cold-rolled steel sheets in an automotive driving environment using a hydrogen absorption monitoring system13,14) which can measure the amount of diffusive hydrogen being absorbed into steel sheets electrochemically.
Galvanized steel sheets, such as hot dip galvanized (GI) steel sheets, galvannealed (GA) steel sheets and electrogalvanized (EG) steel sheets, are mainly used in automotive parts of under body that are exposed to atmospheric corrosion environments. It has been reported that the corrosion protection mechanism of each steel sheet is the same regardless of the type of plating, and the coating weight of the plating greatly contributes to corrosion resistance.15,16) Fujita et al.17) reported that corrosion of galvanized steel sheets proceeds in four processes, τ1 to τ4, and in the corrosion process τ2, in which the substrate steel sheet is partially exposed due to the progress of corrosion or coating defects, the zinc coating is sacrificially corroded, thereby protecting the substrate steel. In the sacrificial corrosion process in τ2, a cathodic reaction occurs, in which oxygen and water are reduced on the substrate steel sheet simultaneously with an anodic reaction in which the zinc in the coating is dissolved. Ootsuka et al.18) examined the effect of the zinc coating on the amount of absorbed hydrogen into the substrate steel in an EG steel in which a part of the substrate steel was exposed beforehand in order to simulate the τ2 corrosion process, and reported that the amount of absorbed hydrogen increases compared with that in cold-rolled steel sheets because the cathodic reaction of water, which is a hydrogen evolution reaction, is activated at the corrosion potential of the galvanized steel. However, the behavior of hydrogen absorption into galvanized steel by corrosion also changes because the zinc coating gradually disappears as corrosion proceeds. Thus, in order to understand the relationship between the hydrogen absorption behavior and corrosion reaction in galvanized steel, it is necessary to clarify the relationship between the change of the surface condition and the amount of absorbed hydrogen.
To clarify the effect of the corrosion process on the hydrogen absorption behavior in EG steel sheets under a dry-wet cyclic corrosion environment, in this study, the amount of absorbed hydrogen into EG steel sheets was measured by an electrochemical method during a dry-wet cyclic corrosion test. In addition, the influence of the change in the exposure of the substrate steel on hydrogen absorption behavior was investigated by changing the exposed area of the substrate steel due to corrosion using EG steel with different coating weights.
Cold-rolled steel sheets with a thickness of 0.75 mm were used. The chemical composition of the test materials is shown in Table 1. For the hydrogen absorption measurement, the specimens were prepared as follows. First, both sides of the sample were mechanically polished and then immersed in a solution containing hydrofluoric acid and hydrogen peroxide at a volume ratio of 6:94. By immersing the specimens and dissolving more than 50 µm on each side, the final thickness of the specimens was reduced to 0.5 mm. Then, a zinc coating was formed on one side of the specimens under the electroplating conditions shown in Table 2. The coating weights of the zinc coating on the specimens were controlled to 20 g/m2, 40 g/m2 and 80 g/m2 by adjusting the electrolysis time, and the corresponding coating thicknesses were 3.0 µm, 6.3 µm and 12 µm. The specimens used for hydrogen penetration measurements were plated with Pd about 400 nm thick on the opposite side of the electrogalvanized surface using a commercial Pd plating solution (K-Pure Pd, Kojima Chemical Co., Ltd.).19)
The hydrogen absorption monitoring system shown in Fig. 1 was used to measure the amount of absorbed hydrogen into the steel sheet. As shown in the figure, the hydrogen absorption monitoring system consists of four three-electrode electrochemical cells composed of Pt wires as counter electrodes and Ir/IrOx as reference electrodes. The treated sample was attached to the hydrogen absorption monitoring system such that the Pd-coated surface faced the cell on the hydrogen detection side, where 0.1 mol/L NaOH was injected. The NaOH was deaerated with nitrogen gas in advance for more than 24 hours. The specimen surface of the hydrogen detection side cell was left at +0.2 V vs. Ir/IrOx, and the anodic current flowing in each cell was continuously measured by potentiostatic polarization. As shown in Fig. 1, part of the steel sheet surface attached to the hydrogen absorption monitoring system was covered with a sealant to block hydrogen absorption, and a cell was installed to measure only the residual current.14) The net hydrogen permeation current was determined from the difference between the anode current measured in the cell without the sealant coating and the residual current measured in the reference cell.
Hydrogen absorption monitoring system.
The specimens for corrosion analysis and hydrogen absorption monitoring systems were subjected to the dry-wet cyclic corrosion test (DWCT). This test consists of two processes: the saltwater spray process and the dry-wet process. First, in the saltwater spray process, a 9 mass% aqueous NaCl solution was sprayed onto the specimen surface to deposit 3 g/m2. Figure 2 shows the change in the relative humidity in the dry-wet process of DWCT. This cycle was repeated for 21 days. The temperature during the DWCT was kept constant at 50°C. To maintain a constant amount of salt deposition during the test period, the surface of the samples was washed with pure water immediately after the start of the dry period twice a week, followed by saltwater spray process.
Dry-wet cyclic corrosion condition.
The specimens for the corrosion analysis with the 20 g/m2, 40 g/m2 and 80 g/m2 zinc coatings were taken from the test chamber after an arbitrary period. To determine the amount and distribution of the residual zinc coating, the specimens were immersed for 1 min in an aqueous ammonium dichromate solution heated to 80°C for to remove the corrosion products of zinc that had formed on the surface. The amount of the residual zinc coating was then calculated by measuring the Zn intensity on the surface with a fluorescence X-ray analyzer (ZSX101e; Rigaku Corp.). An intensity mapping of Fe was also prepared by area analysis of the specimen surface using an electron probe microanalyzer (EPMA, JXA-8100; JEOL Ltd.). The acceleration voltage was 15 kV, and the beam diameter was 10 µm. The area fraction where the substrate steel was exposed was calculated by binarizing the value at half of the maximum value of the Fe intensity as a threshold.
2.5 Hydrogen permeation measurement during cathode chargingHydrogen permeation during cathode charging was measured using a Devanathan-Stachurski cell,13) as shown in Fig. 3. The specimens used in this measurement were those that were Pd-plated as described in 2.1 after chemical polishing. The specimens were sandwiched between the cells in two vessels, the entire cell was immersed in a water bath, and the temperature was set at 50°C. The vessel facing the Pd-plated surface was injected with a 0.1 mol/L NaOH aqueous solution deaerated with nitrogen gas, and the surface was left at +0.2 V vs. Ir/IrOx until the anode current remained constant. (Here, this anode current is called the residual current.) The other vessel was injected with a 9 mass% NaCl aqueous solution, and the sample surface was left at −850 to −1200 mV vs. Ag/AgCl to generate hydrogen. The hydrogen permeation current under cathodic charging was obtained from the difference between the residual current and the stable value of the anode current measured on the side of the hydrogen detection surface.
Devanathan–Stachurski type electrochemical hydrogen permeation cell.
Figure 4 presents photographs of the appearance of the corrosion analysis specimens removed from the test chamber 24, 72, 168, 336 and 504 hours after the start of the corrosion test. Red rust was observed after the formation of white rust regardless of the coating weight, and the period until the occurrence of red rust increased as the zinc coating weight increased.
Temporal changes in corrosion appearances of electrogalvanized steel; (a) 20 g/m2, (b) 40 g/m2, (c) 80 g/m2.
The corrosion analysis specimens were collected from the test chamber 8 hours after the start of DWCT. Figure 5 shows the Fe intensity mapping of the specimens after removing the zinc-based corrosion product. In the specimens with coating weights of (a) 20 g/m2 and (b) 40 g/m2, islands or lines of strong Fe intensity were observed, indicating that the zinc coating was dissolved by corrosion and the steel substrate was partially exposed. No peak of Fe was detected in the specimen with the coating weight of (c) 80 g/m2, indicating that the entire surface was covered with the zinc coating. Figure 6 shows the time variations of the exposed surface area ratio measured from the Fe intensity mapping. The larger coating weights show a smaller time variation of the exposed surface area ratio.
EPMA mapping of Fe on electrogalvanized steel corroded for 8 hours; (a) 20 g/m2, (b) 40 g/m2, (c) 80 g/m2.
Relationship between exposed surface area ratio and test duration.
Figure 7 shows the time variations of the hydrogen permeation current into each steel sheet during the entire corrosion test period. The hydrogen permeation current of all the specimens started to be detected at each time after the start of the test and reached a maximum value at a certain time while repeating the increase and decrease in every dry-wet cycle. Thereafter, the hydrogen permeation current showed a tendency to decay while continuing to repeat the increase and decrease. The decay of the hydrogen permeation current moderated as corrosion progressed to some extent and increased and decreased repeatedly in the range of 100 nA/cm2 and 700 nA/cm2 in the later stages of the test. The specimens with a heavier coating weight displayed a longer time tmax to reach the maximum hydrogen permeation current. The tmax of the respective samples was 20.1 hours, 76.2 hours and 172 hours, corresponding to the increasing coating weights.
Time variations of hydrogen permeation current in electrogalvanized steel sheet in DWCT; (a) 20 g/m2, (b) 40 g/m2, (c) 80 g/m2.
Figure 8 shows the time variations of the hydrogen permeation current in the cycle around tmax and one cycle before and after it. Regardless of the coating weight, the hydrogen permeation current began to increase in the middle of the humidity increase period and reached a maximum just after the beginning of the wet period.
Time variations of relative humidity and hydrogen permeation currents around time to reach maximum current value.
Figure 9 shows the time variations of the hydrogen permeation current from the beginning of the corrosion test to 32 hours. The hydrogen permeation current was not detected for a time after the beginning of the test, but then began to be detected at a certain time. The detection times td of the hydrogen permeation currents of the 20 g/m2, 40 g/m2 and 80 g/m2 specimens were 6.2 hours, 11.7 hours and 27.8 hours, respectively. The td tended to increase with the increase in the coating weight.
Time required for detection of hydrogen permeation current in electrogalvanized steel sheet.
Using the measured hydrogen permeation current, the surface hydrogen concentration Cab of the steel sheets was calculated by eq. (1).
| \begin{equation} C_{\text{ab}} = i_{\text{p}}L/DFd \end{equation} | (1) |
Here, ip is the hydrogen permeation current (µA/cm2), L is the sheet thickness (cm), D is the hydrogen diffusion coefficient in the sample (cm2/s), F is the Faraday constant (96485 C/mol), and d is the density of steel (7.85 g/cm3). The maximum surface hydrogen concentration Cab-max was calculated by substituting the maximum value of the hydrogen permeation current into eq. (1). Figure 10 presents the relationship between Cab-max and the coating weight of the zinc coating. The dependence of Cab-max on the coating weight was small, and Cab-max was 0.014 ± 0.001 ppm.
Relationship between maximum amount of absorbed hydrogen and zinc coating weight.
As shown in Fig. 6, this experiment revealed that the exposed surface area on the specimen increased due to corrosion, and the time variation of this ratio decreased with increasing coating weight. The hydrogen permeation current of the EG steel tended to increase and decrease as corrosion progressed. In addition, it was also found that td and tmax were prolonged by increasing the coating weight. Here, the reason for these changes in the hydrogen permeation behavior will be discussed in relation to the corrosion behavior of galvanized steel sheets proposed by Fujita et al.17)
The corrosion process immediately after the start of the test is the corrosion process of the zinc coating τ1. This process occurs on the surface of the zinc coating, and consists of the dissolution reaction of zinc as an anodic reaction, as expressed by eq. (1)', and the hydrogen evolution reaction as a part of the cathodic reaction, as expressed by eq. (2):
| \begin{equation} \text{Zn} \rightarrow \text{Zn$^{2+}$} + \text{2e$^{-}$} \end{equation} | (1)' |
| \begin{equation} \text{H$_{2}$O} + \text{e$^{-}$} \rightarrow \text{H} + \text{OH$^{-}$} \end{equation} | (2) |
The hydrogen diffusion coefficient in zinc20) is on the order of 10−11 cm2/s, which is 6 orders of magnitude smaller than the hydrogen diffusion coefficient in the mild steel sheet used in this study (D = 1.1 * 10−5 cm2/s). Therefore, it takes a great deal of time for hydrogen to permeate through the galvanized coating. It is considered that hydrogen absorption into the zinc coating was not detected as a hydrogen permeation current during the τ1 corrosion process for this reason. In the results shown in Fig. 7 and Fig. 8, exposure of the substrate and detection of the hydrogen permeation current in the specimen with the coating weight of 80 g/m2 were not observed until about 24 hours after the start of the corrosion test. In other words, it is found that hydrogen absorption into the steel does not occur during the τ1 process.
When corrosion causes the dissolution of the zinc coating to proceed and the steel substrate is partially exposed on the surface, the process that indicates sacrificial corrosion protection (τ2) proceeds. Due to this sacrificial corrosion protection, hydrogen evolves (eq. (2)) on the exposed surface of the substrate, and part of this hydrogen is absorbed into the substrate. In this study, a hydrogen permeation current was detected after exposure of the substrate was confirmed, as shown in Fig. 6 and Fig. 9. In other words, td is considered to correspond to the time until the corrosion process changes from τ1 to τ2, and td is considered to be prolonged due to the increase in the coating thickness with the heavier coating weights. It has been reported that the hydrogen permeation current increases with an increase in the defect area of the zinc coating of EG steel.21) Thus, it is presumed that the increase in the hydrogen permeation current observed after td in Fig. 9 is due to an increase in the area where hydrogen evolves accompanying the dissolution of the zinc coating.
The corrosion products were removed from the corroded EG specimens in the corrosion tests in which red rust did not occur, and the amount of the residual zinc coating was measured by XRF, as shown in Fig. 11. It was found that the amount of the residual zinc coating tended to decrease in proportion to the test time, as shown in Fig. 11. The time dependence of the residual zinc coating was obtained by the least-squares method. The time required to consume the total zinc coating te was estimated by extrapolating the straight line obtained by the least-squares calculation. The te values for the specimens with the coating weights of 20 g/m2, 40 g/m2 and 80 g/m2 were 31.0 h, 88.1 h and 275 h, respectively. The τ3 process in which the steel substrate is protected by corrosion products containing zinc starts after te. Sacrificial corrosion protection does not occur in τ3 because all the metallic zinc has been consumed.
Relationship between residual zinc coating weight and test duration.
Figure 12 shows the zinc coating weight dependencies of td, tmax and te. It became clear that the period from td to te, that is, the corrosion period during the τ2 process, becomes longer as the coating weight increases. This also corresponds to the change in the appearance of the specimens shown in Fig. 4. tmax was larger than td and smaller than te, regardless of the coating weight. The hydrogen permeation current showed its maximum value during the τ2 process, and it was also found in this study that hydrogen permeation was promoted by the sacrificial corrosion process.
Zinc coating weight dependencies of time to detecting permeation current, time to appearing maximum permeation current and time to ending point of τ2.
After tmax, the peak value of the hydrogen permeation current decreased and partial generation of red rust was observed on the surface of the EG mild steel. Sacrificial corrosion protection by the zinc coating is not significant at the spot where red rust occurs, suggesting that corrosion of the substrate steel sheet is progressing. Because the corrosion potential14) of steel is approximately −0.60 V vs. Ag/AgCl and is noble compared to the corrosion potential of zinc14) (about −0.95 V vs. Ag/AgCl), the hydrogen permeation current in the τ3 process is considered to decrease compared to the hydrogen permeation current in the τ2 process.
As shown in Fig. 7, the peak value of the hydrogen permeation current stabilized at about 100 nA/cm2 to 700 nA/cm2 after te, that is, during the τ3 process. When converted to the surface hydrogen concentration by eq. (1), this is 0.001 ppm to 0.007 ppm, which is smaller than the maximum surface hydrogen concentration of 0.095 ppm in cold-rolled mild steel sheets measured by Ootsuka et al.22) in a corrosion test under the same conditions as in this study. Due to the suppression of the corrosion reaction by the corrosion products of zinc, the hydrogen permeation in the τ3 process is suppressed compared with that in cold-rolled steel sheets. It is known that the pH of the surface liquid film has a large influence on the hydrogen absorption into cold-rolled steel sheets under an atmospheric corrosion environment. Omura et al.23) reported that hydrogen absorption is promoted because the surface pH decreases with the acceleration of the hydrolysis reaction of iron ions by the concentration of chloride ions. On the other hand, in the corrosion of galvanized steel, the corrosion products of zinc have been reported to have the effect of suppressing the anodic dissolution of the substrate steel.24,25) It is thought to suppress the supply of iron ions, which is a factor in the pH decrease.
Based on the above results, the hydrogen absorption behavior associated with the transition in the corrosion process of EG steel is shown as a schematic diagram in Fig. 13. In the τ1 process, the hydrogen that evolves at the zinc surface is not absorbed into the steel substrate because the hydrogen permeation rate in the zinc coating is slow, as shown Fig. 13(I). The corrosion process shifts to the τ2 process when the zinc coating dissolves further, exposing the steel substrate. A large amount of hydrogen is generated on the substrate by sacrificial corrosion protection, some of which begins to be absorbed into the substrate, as shown Fig. 13(II). As the exposed surface area by corrosion increases, the area where a large amount of hydrogen is generated by sacrificial corrosion protection also increases, so the amount of absorbed hydrogen into the substrate increases, as shown in Fig. 13(III). The amount of absorbed hydrogen reaches its maximum value when the exposed surface area ratio of the substrate is about 35% to 55%, but if corrosion progresses further and the remaining amount of the zinc coating continues to decrease, sacrificial corrosion protection becomes ineffective and corrosion of the steel substrate begins to occur. As a result, the amount of evolved hydrogen on the substrate decreases, so the amount of absorbed hydrogen into the substrate also decreases, as shown in Fig. 13(IV). When the zinc coating has been completely consumed and the corrosion process shifts to τ3, the amount of absorbed hydrogen is smaller than that of ordinary cold-rolled steel sheets, as shown in Fig. 13(V) because the anodic dissolution of the steel substrate is suppressed by corrosion products of the zinc, and this also reduces the amount of evolved hydrogen.
Schematic for hydrogen absorption behavior in galvanized steel with transition of corrosion process.
In the corrosion of galvanized steel sheets, the corrosion potential is expected to change depending on the surface condition of the galvanized steel sheet. Since the reaction rate of hydrogen evolution shown in eq. (2) varies greatly depending on the potential, it is considered that the amount of absorbed hydrogen into the steel substrate also depends on the potential. Therefore, the effect of potential on hydrogen absorption into steel was measured by the cathode charge method. In this study, the relationship between the corrosion process and the hydrogen permeation behavior of EG steel sheets was verified, and it was found that the maximum amount of absorbed hydrogen into the steel sheets occurs in the τ2 corrosion process. However, as shown in Fig. 5, the corrosion of EG steel sheets does not proceed uniformly. In the τ2 process, the exposed surface area of the substrate and the areas protected by the zinc coating are intermixed. Because it is considered that hydrogen evolves and is absorbed at the exposed surface area of substrate, the surface hydrogen concentration per the exposed surface area of the steel sheet was calculated. From Fig. 6, the exposed surface area ratio of the substrate at tmax for EG mild steel sheets with coating weights of 20 g/m2, 40 g/m2 and 80 g/m2 were estimated to be 52%, 57% and 34%, respectively. This suggests that the amount of absorbed hydrogen reaches the maximum value when the exposed surface area ratio is approximately 35% to 55%. As shown in Fig. 10, the maximum surface hydrogen concentration of EG steel sheets was about 0.013 ppm to 0.015 ppm in the dry-wet cyclic corrosion environment. When the maximum surface hydrogen concentration obtained is converted to a value per the exposed surface area of the substrate at tmax, it is calculated to be 0.031 ppm to 0.037 ppm.
Figure 14 represents the relationship between the maximum surface hydrogen concentration obtained by the cathode charge method and the set potential. If the set potential becomes less noble, the surface hydrogen concentration becomes higher. The surface hydrogen concentration at the potential of −0.95 V vs. Ag/AgCl, which is the corrosion potential of EG steel, was estimated to be about 0.02 to 0.04 ppm. These values are consistent with the above-mentioned calculated value of the maximum surface hydrogen concentration in the corrosion test shown in Fig. 10. This result suggests that the corrosion potential is a dominant factor for hydrogen absorption into galvanized steel sheets under an atmospheric corrosion environment.
Relationship between electro potential and amount of absorbed hydrogen in cold rolled steel sheet.
The hydrogen absorption behavior of electrogalvanized steel sheets during continuous corrosion in a dry-wet cyclic environment was investigated using electrochemical techniques. The following results were obtained.
