2016 Volume 56 Issue 3 Pages 459-464
Hydrogen absorption into steel under an atmospheric corrosion condition is considered to occur because of the H+ ion is reduced to hydrogen atom (Hads) receiving electron formed by rust that forms by the corrosion reaction on steels. Thus, we investigated whether it is possible to reduce the amount of hydrogen absorption into steels by adding elements that improve corrosion resistance and control the formation of rust to reduce corrosion reaction. It was discovered that the amount of hydrogen absorption into steels can be reduced by adding elements that improve corrosion resistance and control the formation of rust on the surface of steels.
Improving the collision safety of vehicles and fuel efficiency by reducing the vehicle’s body weight have been suggested, and the application of high strength steels exceeding 1180 MPa is in progress. However, there is a concern of “delayed fracture (hydrogen embrittlement),1)” which is considered to absorb hydrogen into steels while using these steels in applications that require high strength steels exceeding 1180 MPa.
Corrosion of steels under an atmospheric corrosion condition which is assumed to be the environment in which vehicles are used, is considered to be progress in accordance with the Evans model.2) Under an atmospheric corrosion condition the surface of steels alternates between wet and dry consecutively because of uncertain changes in temperature, humidity, precipitations, etc. When the surface of steels are wet, an anode reaction (dissolution of Fe: Fe → Fe2+ + 2e−) and a cathode reaction (reduction of O2: 1/2O2 + H2O + 2e− → 2OH−) occur, and these reactions formed rust (Fe oxide and hydroxide). In addition, when the surface of steels dried up, the rust formed undergoes an oxidation reaction followed by hydrolysis in which Fe hydroxide and hydrogen ions are generated and accumulated (Fe2++2H2O → Fe(OH)2+2H+).3,4) H+ ion is reduced to hydrogen atom (Hads) receiving electron (H++e−→Had). In this manner, hydrogen is absorbed into steels.5)
Corrosion prevention measures such as maintenance and painting are implemented for bridges and steels structures in order to secure long-term durability, but these corrosion protection measures are costly. Therefore, the increased demand for minimum cost for the maintenance of steel structures, i.e., to make weathering steel,6,7) has been gathering attention. Weathering steel was developed in the U.S. during 1930s. By adding small amounts of elements such as Cu, Cr, Ni, and P, a dense and adhesive rust layer is formed under an atmospheric corrosion condition, reducing the corrosion of steels.8) In addition, on the basis of subsequent studies,9,10) Addition of Cr in steels has bilateral character. Densification of rust is not occurring under existing of Cl− ion conditions by Nakayama’s experiment.11) Equilibrium pH at hydrolysis of Cr drop to a lower value than Fe12) from a standpoint of chemical equilibration.13) The following findings were obtained: Cr promotes corrosion because it lowers pH maintain electroneutrality in the areas where corrosion reaction occur under a chloride corrosion condition, adding a small amount of Ti is effective in reducing the formation of β-FeOOH, which is said to have a negative effect on corrosion resistance because it lacks stability, and Cu and Ni promote the formation of amorphous rust with high protectiveness.
In the environment where hydrogen generates and absorbs into steel sheets for vehicles, it is likely that temperature and humidity generally change and there is chloride ion. Thus, in this study, we investigated whether the hydrogen absorption into steels can be reduced by controlling the characteristics of rust formed on the surface of the steels using outdoor exposure testing. In addition, to determine whether delayed fractures occur in steels when using them in the abovementioned environment, it has been proposed that delayed fracture occurs when the amount of hydrogen absorbing from the environment (HE) exceeds the amount of hydrogen at which fractures form (HC).14,15) Thus, in steel sheets with high strength (1180–1470 MPa class), we examined the effect of strength level and the structure of the delayed fracture.
Transformation Induced Plasticity (TRIP) steels and Dual Phase (DP) steels are commonly used because it can easily provide strength, which can exceed 980 MPa, and ductility in high mechanical strength steels.16) The parent phase of Trip-aided Bainitic Ferrite (TBF) steel17,18,19) is a lath-shaped bainitic ferrite structure with a high dislocation density; this structure does not contain carbide. Furthermore, it is a steel structure that combines high ductility and hydrogen trapping ability because of having stable fine retained austenite (γR) between the lath-shaped bainitic ferrites. Dual Phase (DP) steel20) is a steel sheet that contains martensite as the parent phase and ferrite as the secondary phase, Steel A (1470TBF), Steel B (1470DP), and Steel C (1180DP) were used this experiment.
Table 1 shows the chemical composition of Steel A to C. Each steel ingot was prepared by vacuum melting. After hot rolling and cold rolling, the level of strength was adjusted for Steel A by austempering17,18,19) and for Steels B and C with quenching and tempering. The test pieces were shaped by a bending process considering the actual use of that steels in automobile parts. Figure 1 shows the preparation steps for U-bend test pieces. From the test material, test piece strips with the rolling direction as the longitudinal direction and dimensions of 150 mm × 30 mm × 1.2 mm (thickness) were sampled. The edge of the longitudinal surface of the test piece was milled to remove the effect of processing strain formed at the time of the shearing process. A strip piece was cut from this thin plate, and a bending processing was applied to the center in the longitudinal direction of the strip. A strain gauge was adhered to the bent part in which load was applied by tightening a bolt and a nut. For examining the amount of corrosion, a test piece with rolling direction as the longitudinal direction and with dimensions of 150 mm × 70 mm × 1.2 mm (thickness) was sampled.
Preparation procedure of U-bend test pieces.
For the corrosion test, the U-bend and corrosion test pieces were directly exposed to an atmospheric corrosion condition.21) For the outdoor exposure test, the difference in the corrosivity for each test area was considered21) therefore, three locations were chosen—Choshi, Kakogawa, and Miyakojima coasts. To test the amount of HE, U-bend test pieces were prepared. For these pieces, the bending radius of Steel A to C was 10 mm and the stress applied to the bent portions was 0.4 TS (1470 MPa class: 590 MPa and 1180 MPa class: 470 MPa). The test period was 48 months in Choshi and 12 months for Kakogawa and Miyakojima coasts. During the test period, samples were collected regularly. Approximately a 10 mm × 10 mm steel piece was cut from the head area of the U-bend test piece to measure the amount of hydrogen, and the rust on the test piece surface was removed by shot blasting before measuring the amount of hydrogen.
The amount of hydrogen was measured using the atmospheric pressure ionization mass spectrometer (API-MS).22) Determination precision of the hydrogen content of API-MS is within 0.01 wt. ppm. Temperature was increased from room temperature to 300°C at the rate of 12°C/min. The integrated value of hydrogen emitted during this temperature rise was defined as the “the amount of diffusible hydrogen,” which is considered to effect delayed fracture. To confirm the long-term behavior of the fracture that occurred, U-bend test pieces (Steels A to C with the bending radius of 5–20 mm and 500–2000 MPa stress applied to the bent part) were exposed for 48 months in Choshi and the presence or absence of fractures were regularly observed. Load stress applied for this observation was a strain equivalent to the value of the strain multiplied by the Young’s modulus, which was measured by using a strain gauge at the bent portion and tightening the bolt and the nut. Figure 2 shows an image of head area of U-bend test piece and the collection point of hydrogen analysis sample. The degree of corrosion was obtained from the change in weight from before to after the 12-months exposure test in Choshi. In addition, to examine the characteristics of rust on the surface of steels material, the fraction of amorphous rust was deducted of crystalline rust (α-, β-, γ-FeOOH and Fe3O4) using X-ray diffraction for all amount of rust.23)
Image of head area of the U-bend test piece and the collection point of hydrogen analysis sample.
The amount of hydrogen at which fractures occur in the U-bend test piece was evaluated as follows. The amount of hydrogen absorption was changed by changing the density of the applied current to the bent part of the U-bend test piece in the solution of 0.5M-H2SO4+0.01M-KSCN. Observation at head area of U-bend test piece was carried out while an electric current had been applied. The amount of diffusible hydrogen at the time of occurring fracture was evaluated as Hc (ppm). When a fracture occurred on the U-bend test piece, the test piece was removed from the solution and a steel piece of approximately 10 mm × 10 mm from was cut out from the area containing the fracture for measuring diffusible hydrogen content.
Figure 3 shows the weight loss of specimens under atmospheric corrosion condition in Choshi for 12 months. The amount of corrosion was the highest in Steel C (for which elements that improve corrosion resistance (Cu and Ni) were not added) followed by Steel A, and then B (for which Cu, Ni were added).
Weight loss of specimens under atmospheric corrosion condition in Choshi for 12 months.
Figure 4 shows the fraction of amorphous rust in the rust formed in steels A to C that were exposed in Choshi for 12 months. The values were higher in steels A and B, for which Cu, Ni were added, compared with Steel C (which did not contain those elements).
Fraction of amorphous rust in the rust formed in steels A to C that were exposed in Choshi for 12 months.
Figure 5 shows the temporal change in the amount of diffusible hydrogen in the head area of a U-bent test piece (Steel A with a bending radius of 10 mm) for which 590 MPa stress was applied under an atmospheric corrosion condition. In all three exposure-test sites (Choshi, Miyakojima, and Kakogawa), the values reached 0.10–0.16 ppm within 6 months from the beginning of the test. However, there was no increasing trend in the amount of hydrogen beyond 6 months, exhibiting mostly a constant value. The maximum value of hydrogen in steels during the 48-months exposure test in Choshi was 0.12 ppm and the mean value was 0.09 ppm. Figure 6 shows the temporal change in the amount of diffusible hydrogen in the head area of a U-bent test piece (Steel B with the bending radius of 10 mm), for which 590 MPa stress was applied under an atmospheric corrosion condition. In all three exposure-test sites (Choshi, Miyakojima, and Kakogawa), the values reached 0.06–0.18 ppm within 6 months from the beginning of the test. However, there was no increasing trend in the amount of hydrogen beyond 6 months: the values remained close to 0.1 ppm. The maximum value of hydrogen in steels in the 48-months exposure test in Choshi was 0.18 ppm and the mean value was 0.11 ppm. Figure 7 shows the temporal change in the amount of diffusible hydrogen in the head area of a U-bent test piece (Steel C with bending radius of 10 mm) for which 470 MPa stress was applied under an atmospheric corrosion condition. In all three exposure-test sites (Choshi, Miyakojima, and Kakogawa), the values reached 0.10–0.27 ppm within 9 months from the beginning of the test. The value of hydrogen did not increase significantly after 9 months. However, compared with Steels A and B, for which Cu, Ni were added, the amount of hydrogen in Steel C remained at a higher level. In addition, the maximum value of hydrogen in the steels in 48-months exposure test in Choshi was 0.27 ppm and the mean was 0.13 ppm.
The temporal change in the amount of diffusible hydrogen content in the head area of a U-bend test piece (R10 mm - 590 MPa) of steel A under atmospheric corrosion condition.
The temporal change in the amount of diffusible hydrogen content in the head area of a U-bend test piece (R10 mm - 590 MPa) of steel B under atmospheric corrosion condition.
The temporal change in the amount of diffusible hydrogen content in the head area of a U-bend test piece (R10 mm - 470 MPa) of steel C under atmospheric corrosion condition.
Figure 8 shows the presence or absence of a fracture in the U-bend test piece after the 48-months exposure test in Choshi. The bending radius was 5–20 mm for Steels A to C, and the stress load on the bent portion was 500–2000 MPa. A fracture was only noted in a part of Steel B (stress load of 1500 MPa or more), and no fractures were noted in other test pieces.
Evaluation result of delayed fracture of U-bend test piece under atmospheric corrosion condition at Choshi 48 months.
Table 2 shows the critical hydrogen content for delayed fracture (HC: ppm) of each shape U-bend test piece steel A to C measured in laboratory. Measurements were performed based on the condition in which a fracture occurred in the outdoor exposure test shown in Fig. 8. It was shown that when the bending radius is smaller and the stress load on the bent portion is higher (i.e., the bending processing condition is severe) a fracture occurs at lower hydrogen values. The results for the steels of 1470 MPa class were similar to those reported for the thin steel sheet with a high mechanical strength of 1180 MPa class24) and 1670 MPa class.25)
The amount of hydrogen under the atmospheric corrosion condition was the highest in Steel C; this is likely because Cu and Ni (elements that improve corrosion resistance) were not added.9,10) The amount of corrosion was lower compared with the amount of corrosion seen on the typical carbon steels (SM400B) exposed for one year in Choshi (300 g/m2).26) Especially for Steels A and B with added Cu, Ni, and Ti, the amount of corrosion was reduced by 20%–30% compared with typical carbon steels. Figure 9 shows the relationship between outdoor exposure test time and the amount of diffusible hydrogen content of steel A to C. Steels A and B with a higher ratio of amorphous rust had a lower level of hydrogen in the long term compared with Steel C. This is likely because corrosion factors (Cl− and H2O) are prevented from reaching the steels surface because dense amorphous rust is formed by the addition of elements that improve corrosion resistance such as Cu and Ni,9,10) improving corrosion resistance. Nagumo et al.27) reported on the relationship between the amount of supplied electricity and the amount of absorbed hydrogen when hydrogen is added to cathode of high mechanical strength steels. They reported that as the amount of supplied electricity increased, the amount of absorbed hydrogen increased. The amount of corrosion in steels is believed to correlate with the amount of electricity supplied. By adding Cu and Ni, which improve the corrosion resistance of steels, which makes the formed rust denser, the corrosion reaction is reduced, thus reducing the amount of H+ ions generated. In addition, because all types of steel were exposed to the same environment, there should not be difference in the condition of elements that cause catalyst poisoning, which, in turn, impacts the atomization of H+ ions. Combing these findings by controlling the rust formation on the surface of steels due to corrosion reaction reduces the corrosion reaction, leading to the reduction in H+ ions; this leads to a reduction in the amount of hydrogen absorption and long-term reduction in the amount of hydrogen level. Less than other points of diffusible hydrogen at 12 months in Kakogawa, because of the mild environment at Setouchi region, the rust of the surface of steels become stabilized and decreased in corrosion rate, these lead to a reduction in amount of hydrogen at lower level.
Relationship between outdoor exposure test time and the amount of diffusible hydrogen content of steel A to C at Choshi.
In the fracture test for U-bend test piece of Steels A to C, fractures were only observed in Steel B in which the level of hydrogen at which a fracture occurs is low. HC (ppm) shown in Table 2 does not exceed HE (ppm). Future’s consideration is needed, the difference between HC and HE is small. While the amount of hydrogen evaluated in this study was collected over a specific time period, hydrogen on the surface of steels changed during the day in response to the change in temperature and humidity.28) The amount of hydrogen of the surface of the potential to become exceed momentarily high. Moreover, along with the corrosion reaction of steels, the surface becomes uneven.29) As the stress and hydrogen are concentrated on the concave portion, hydrogen embrittlement resistance are decreased,30) fractures may occur when the level of hydrogen is lower than the HC shown in Table 2. Thus, the relationship momentarily becomes HE>HC, creating the fractures.
• The amount of hydrogen absorption into the bent portion of the steels with a high mechanical strength of 1180 to 1470 MPa, which was a corrosive atmospheric environment, was approximately 0.2 ppm. The addition of elements that improve corrosion resistance and the prevention of rust formation on the steels surface allowed for the reduction in the amount of hydrogen absorption into the steels.
• TBF structure is better delayed fracture properties than DP structure just as 1470 MPa grade.