2021 年 61 巻 4 号 p. 1099-1103
The influence of scratch size on the hydrogen permeation behavior of Zn coated steel during wet and dry cycle corrosion tests was investigated electrochemically. A size controlled scratch was formed on Zn coated steels with a pulsed YAG laser machining technique. The hydrogen permeation current was observed independent of formed scratch size. The periodic changes of the hydrogen permeation currents following the cyclic changes of the relative humidity were observed. After the tests, the red rust and zinc-hydroxychloride (simonkolleite) were observed at initially placement of NaCl solution. The total amount of permeated current did not proportionally increase with increasing the formed scratch area, indicating that the corrosion products played an important role in the hydrogen permeation behavior during wet and dry corrosion.
A lot of researches for developing high strength steels was made, because improvement of safety at collisions and reducing energy consumption of vehicles.1,2,3) It is well known that the high strength steels have high susceptibility to hydrogen embrittlement (delayed fracture).4,5) A reduction in susceptibility to hydrogen embrittlement of high strength steels is an important issue to increase reliability and safety in the practical use of the steels. Therefore, hydrogen embrittlement has attracted much attention recently.6,7,8,9,10,11,12,13,14,15,16)
Atmospheric (wet and dry cycle) corrosion plays an important role in hydrogen generation for the practical use of steels including high strength steels. Therefore, a research group “Comprehensive understanding of hydrogen-passive surface on steels for prevention of hydrogen embrittlement” to investigate the hydrogen entry process focused on properties of surfaces in The Iron and Steel Institute of Japan was worked,17) and the results of the research group were published in special issue of ISIJ International.18) The hydrogen permeation behavior during wet and dry cycles19,20) and site exposure tests21) have been reported.
The corrosion related hydrogen generation can be reduced by increasing corrosion resistance of steels. Shinohara reported that the atmospheric corrosion rate of Zn was more than twenty times smaller than that of steels.22) Because of this slow corrosion rate in atmospheric environment and galvanic behavior of coated layer, Zn and Zn alloy coatings are widely used to increase the corrosion resistance of steels.23,24,25)
These protective coatings are liable to be scratched during use. When steel is exposed to the environment, resulting in the formation of galvanic couples of the coated layer and steel substrate. Such galvanic corrosion is more serious than the corrosion of the Zn coating or steel substrate alone, making it necessary to investigate the effect of scratches formed on Zn coated steels and their role in hydrogen generation and permeation behavior during atmospheric corrosion. The hydrogen permeation behavior of coated steels with scratch during wet and dry corrosion tests reported.26,27,28,29,30) It is still not fully elucidated the effect of scratch size on hydrogen permeation behavior of the coated steels. The purpose of this study is to investigate effect of size of formed scratches on the hydrogen permeation behavior of coated steels during wet and dry cycle corrosion tests.
Steel sheets (15 mm in width, 30 mm in length, 0.5 mm in thickness) with about 17 μm thickness of Zn coating were used as specimens. To remove formed Zn layer, one side (hydrogen detection side) of the specimens was mechanically ground under running water by SiC paper up to #1500. After that the specimens were cleaned for 300 s each in highly purified water and ethanol ultrasonic baths, and following this the side for the hydrogen detection was plated electrochemically with Ni in 0.312 kmol m−3 NiSO4/0.781 kmol m−3 H3BO3 solution at 4 mA cm−2 for 720 s.
2.2. Formation of Size Controlled ScratchThe Ni plated specimen was subject to scratch formation. A contamination free and area controlled scratch formation technique is essential to conduct this study, a laser machining technique30,31) was applied to form the scratches on the Ni plated specimens. The used laser was a focused Nd-YAG laser (wave length 532 nm, pulse duration 5 ns, pulse frequency 10 s−1, and power 20 mW at the front of the lens). The size of formed scratches, S, was controlled to be from 2.0 and 2.6 mm2 by scanning of the laser beam. Figure 1 shows surface optical image of S = 2.6 mm2 specimen, and a dotted line in the figures indicates formed scratch. From the figure, used laser machining technique makes it possible to form size and shape controlled scratch.
Surface optical image of scratch formed specimen.
Previously reported electrochemical hydrogen detection system27,28,29,30) was used in this experiment. Platinum wires were used as the counter and the reference electrodes for the hydrogen detection cell. The specimens for the hydrogen detection (the nickel plated) side was placed on hydrogen detection cell filled with 1 kmol m−3 NaOH. The cell was placed in a relative humidity (RH) and temperature controlled chamber. The specimens were polarized at −30 mV vs. Pt to measure the hydrogen generated by the corrosion and permeating through the specimen by electrochemical process. The applied anodic potential in this experiment was sufficient to ionize any hydrogen reaching at the surface after permeating through the specimen. After the current of the hydrogen detection cell reached a sufficiently low and steady value (about 86.4 ks after the potential applied), 10 μL of 0.1 kmol m−3 NaCl solution was placed on the center of the scratch of the specimens to initiate the corrosion. After placement of the solution, hydrogen permeation current, RH, and temperature were recorded. A wet and dry cycle,27) which consists 6 h for one cycle, is as follows: wetting for 1 h (90% RH, 303 K), wet to dry transition for 1 h, drying for 3 h, and dry to re-wet transition for 1 h. After the relative humidity RH and temperature in the chamber reached the initial values, total 11 wet and dry cycles were carried out. The RH change induced the cycle of wet and dry at the area that initially NaCl solution was placed. The temperature change during wet and dry tests induced the cyclic change of the residual current of detection cell.33,34) The temperature dependent current was measured using masked specimen, and the measured current was subtracted from the currents that obtained using scratch formed specimen during wet and dry corrosion tests. In this paper, the current which subtracted non corrosion related current is named as the hydrogen permeation current.
2.4. Surface Observation and AnalysisThe specimen surfaces after the wet and dry corrosion tests were observed by optical microscope and a scanning electron microscope, SEM, equipped with an energy dispersive X-ray spectroscope, EDS. The chemical composition of corrosion products, which were formed during wet and dry corrosion tests, were investigated by an microscopic Raman spectroscopy.
Figure 2 shows changes in hydrogen permeation currents measured from scratch formed specimens during wet and dry corrosion tests. As a reference of hydrogen permeation behavior of coated layer itself, results of scratch free specimen (S = 0 mm2) is shown in the figure. Independent of the size of scratch, periodic changes of the hydrogen permeation current is observed, and the maximum value of the current changed with cycle number. The corrosion of Zn coating occurred under the relatively thick solution layer (droplet) at the first cycle, consequently, the shape of the current at first cycle in every specimen is deferent from other cycles. It is clearly seen that the magnitude of current detected from scratch free specimen is about one decade smaller than that of scratch formed specimens. This result is due to high corrosion resistance of Zn in atmospheric environment22) and low hydrogen entry efficiency of Zn coating.35)
Changes in hydrogen permeation currents detected from different size scratch formed specimens during wet and dry cyclic corrosion tests.
To investigate in detail change of hydrogen permeation current, an enlargement of the current changes in Fig. 2 of 1st and 10th cycle are shown in Fig. 3. Plots of the temperature and RH during the cycle of the tests are added in the figures. The changes of RH are synchronized with those of the temperature, while the changes in the hydrogen permeation current are not synchronized with the changes in temperature and RH. Omura et al. reported that the similar behavior between hydrogen permeation current and RH.36) They explained the behavior based on opposite humidity dependency between the corrosion rate and the chloride ion concentration in thin solution layer. During heating (298 K to 303 K), solution layer thickness decreases. This causes condensation of the Cl− in the solution layer, and accelerates the corrosion and hydrogen generation rate. To clarify the detailed mechanism of this hydrogen permeation behavior, further in depth study is required.
Hydrogen permeation current, temperature and humidity at 1st and 10th cycle as a function of time.
The hydrogen permeation charge at one cycle as a function of cycle number is shown in Fig. 4. Based on small difference of scratched area in the present study, the effect of scratch size on permeation charge at each cycle does not show clear scratch area dependency. However, the charge at each cycle shown in Fig. 4 is larger than that previously reported S < 1.0 mm2 specimens results.30) This result suggests that the hydrogen permeation behavior at same cycle number is affected by formed scratch area. The charge at first cycle shows largest value compared with other cycle. One of the possible reason is difference of solution layer thickness and coverage of corrosion products. According to the experimental procedure, the corrosion was progressed under the relatively thick solution layer at the first cycle, while it was progressed under very thin solution layer that was formed by wetting. Moreover, no Zn corrosion product was covered on the scratch surface at the first cycle, while part of the surface may covered by the corrosion products after the second cycle. The thin solution layer and formed corrosion products may restrict the cathodic reaction and hydrogen permeation.
Hydrogen permeation charge as a function of wet and dry cycle number.
To investigate effect of scratch area on hydrogen permeation behavior, the total amount of hydrogen permeation charge of specimens during tests (total charge) are also compared. In previous paper,30) the total charge of specimens with S < 1.0 mm2 was examined, and the extrapolated value of the total charge based on previous study is about 5 mC for S = 2.5 mm2, respectively. The total charge obtained from S = 2.0 and 2.6 mm2 specimens in this study is 9 mC and 7 mC. The values obtained in this study are larger than that expected from previously reported S < 1.0 mm2 specimen’s results. To investigate the reason for this unexpected detection of hydrogen permeation charge and the variation of hydrogen permeated charge with wet and dry cycle number, surface observation and EDS elemental analysis were carried out. Figure 5 shows optical image of specimen surfaces after the test. Dotted lines in the figures indicate formed scratches. White corrosion products at placement of solution are observed on both specimens. Red corrosion products were observed at the formed scratches by optical microscope. Figure 6 shows the EDS elemental mapping of S = 2.6 mm2 specimen. Cl is detected outside the formed scratch, while O and Zn are observed at the placement of solution. In previously reported results of S < 1.0 mm2 specimens,30) the scratch area was covered with white corrosion products and no red rust was observed. The presence of oxygen and red corrosion products obtained in this study indicate that corrosion reaction (dissolution of iron) takes place on the scratch, and suggest that corrosion reaction on the scratch area could not be totally suppressed by galvanic reaction of Zn. Therefore, not fully cover of Zn corrosion products on the scratch surface is possible reason for unexpected hydrogen permeation charge was observed from S = 2.0 and 2.6 mm2 specimens.
Optical images of specimens with scratch after the wet and dry corrosion tests.
EDS elemental mapping results of specimens after wet and dry corrosion tests. (Online version in color.)
Several different Zn corrosion products have been reported.37,38) The observation of microscopic Raman spectrometry of the white corrosion product was carried out to investigate the chemical structure of white corrosion products. Figure 7 shows Raman spectra obtained from white corrosion products formed on specimen with S = 2.6 mm2 in Fig. 5. The circles indicated in the Fig. 7 are Zn hydroxychloride (simonkolleite), ZnCl2[Zn(OH)2]4H2O, related peaks.37) From this result, formed white corrosion product is attributed to simonkolleite. Formation of simonkolleite is indicated to reduce the total amount of Cl− in the thin water layer which formed on the samples during wetting period in wet and dry cyclic corrosion tests. This decline in Cl− concentration in the layer is also reason for reducing the corrosion rate of the specimens during wet and dry cycle tests.
Raman spectra of formed corrosion products in Fig. 5, S = 2.6 mm2 specimen.
The size controlled artificial scratch was formed on Zn coated steels by laser machining, and the effect of size of scratches on the hydrogen permeation behavior of Zn coated steels during wet and dry cycle corrosion tests was investigated. Following conclusions may be drawn.
(1) From scratch formed specimens, periodic changes of the hydrogen permeation current is observed and the maximum value of the current at each cycle changes with cycle number.
(2) The total hydrogen permeated charge obtained from S = 2.0 and 2.6 mm2 specimens is larger than that expected from previous results obtained from S < 1.0 mm2 specimens.
(3) Zinc-hydroxychloride forms mainly around the scratch, and red rust forms at the scratch.
The SEM-EDS observation of this study was conducted at Laboratory of XPS analysis, Hokkaido University, supported by “Material Analysis and Structure Analysis Open Unit (MASAOU)”.
