Hisatake Itoga

Pressure Cycle Testing of Cr–Mo Steel Pressure Vessels Subjected to Gaseous Hydrogen

Pressure cycle tests were performed on two types of Cr–Mo steel pressure vessels with notches machined on their inside under hydrogen-gas pressures, between 0.6 and 45 MPa at room temperature. Fatigue crack growth (FCG) and fracture toughness tests of the Cr–Mo steels samples from the vessels were also carried out in gaseous hydrogen. The Cr–Mo steels showed accelerated FCG rates in gaseous hydrogen compared to ambient air. The fracture toughness of the Cr–Mo steels in gaseous hydrogen was significantly smaller than that in ambient air. Four pressure vessels were tested with gaseous hydrogen. All pressure vessels failed by leak-before-break (LBB). The LBB failure of one pressure vessel could not be estimated by using the fracture toughness in gaseous hydrogen KIC,H; accordingly, the LBB assessment based on KIC,H is conservative and there is a possibility that KIC,H does not provide a reasonable assessment of LBB. In contrast, the fatigue lives of all pressure vessels could be estimated by using the accelerated FCG rates in gaseous hydrogen.

SSRT and fatigue crack growth properties of high-strength austenitic stainless steels in high-pressure hydrogen gas

Slow strain rate tests (SSRTs) were performed with two types of high-strength austenitic stainless steels, Types AH and BX, as well as with two types of conventional austenitic stainless steels, Types 304 and 316L. The tests used the following combinations of specimen types and test atmospheres: (i) non-charged specimens tested in air, (ii) hydrogen-charged specimens tested in air (tests for internal hydrogen), and (iii) non-charged specimens tested in hydrogen gas at pressures of 78 ∼ 115 MPa (tests for external hydrogen). Type 304 exhibited a marked reduction of ductility in the tests for both internal hydrogen and external hydrogen, whereas Types AH, BX and 316L exhibited little or no degradation. In addition, fatigue crack growth (FCG) tests for the four types of steels were also carried out in air and hydrogen gas at pressures of 100 ∼ 115 MPa. In Type 304, FCG in hydrogen gas was more than 10 times as fast as that in air, whereas the acceleration rate remained within 1.5 ∼ 3 times in Types AH, BX and 316L. It was presumed that, in Types AH and BX, a small amount of additive elements, e.g. nitrogen and niobium, increased the strength as well as the stability of the austenitic phase, which thereby led to the excellent resistance against hydrogen.Copyright

Fatigue-life and leak-before-break assessments of CR-MO steel pressure vessels with high-pressure gaseous hydrogen

Pressure cycle tests were performed on two types of Cr-Mo steel pressure vessels with inner diameters of 306 mm and 210 mm and notches machined on their inside under hydrogen-gas pressures, varied between 0.6 and 45 MPa at room temperature. One of the Cr-Mo steels had a fine microstructure with tensile strength of 828 MPa, while the other had a coarse microstructure with tensile strength of 947 MPa. Fatigue-crack growth (FCG) and fracture-toughness tests of the Cr-Mo steels were also carried out in gaseous hydrogen. The Cr-Mo steels showed accelerated FCG rates in gaseous hydrogen compared to ambient air with an upper bound corresponding to an approximately 30-times higher FCG rate. Furthermore, in gaseous hydrogen, the fracture toughness of the Cr-Mo steel with coarse microstructure was significantly smaller than that of the steel with fine microstructure. Four pressure vessels were tested; then, all of the pressure vessels failed by leak-before-break (LBB). Based on the fracture-mechanics approach, the LBB failure of one pressure vessel could not be estimated by using the fracture toughness in gaseous hydrogen. The fatigue lives could be estimated by using the upper bound of the accelerated FCG rates in gaseous hydrogen.Copyright

Hydrogen-assisted cracking of Cr-Mo steel in slow strain rate tensile test with high-pressure gaseous hydrogen

Slow strain rate tensile (SSRT) tests were performed using smooth specimens of quenched and tempered JIS-SCM435 steels with three different tensile strengths (TS), which are ranged from 824 to 1127 MPa. The tests were carried out in 115 MPa hydrogen gas and reference gases (air or 115 MPa nitrogen gas) at three different temperatures; 233 K, room temperature (RT) and 393 K. In the reference gases, the specimens exhibited the so-called cup-and-cone fracture at every temperature. On the other hand, in hydrogen gas, a number of cracks initiated at specimen surface and grew, which led to a marked reduction in ductility at every temperature. The crack growth curves were obtained as a function of true strain by observing the specimen surface of the fractured specimens. The true strain at which the hydrogen-assisted cracking starts was strongly dependent on the microstructure, strength level and test temperature. However, in all the materials tested at RT, the hydrogen-assisted cracking did not occur during the uniform deformation, but occurred in the necking process. Even at 233 K and 393 K, the material with a moderate strength did not exhibit the hydrogen-enhanced cracking before reaching the TS. The result ensured that the Cr-Mo steel with a moderate strength can maintain the TS even in 115 MPa hydrogen from the viewpoint of fracture mechanism.Copyright

Tensile- and Fatigue-Properties of Low Alloy Steel JIS-SCM435 and Carbon Steel JIS-SM490B in 115 MPA Hydrogen Gas

Slow strain rate tests using smooth specimens of two types of steels, low alloy steel JIS-SCM435 and carbon steel JIS-SM490B, were carried out in nitrogen gas and hydrogen gas under a pressure of 115 MPa at three different temperatures: 233 K, room temperature and 393 K. In nitrogen gas, these steels exhibited the so-called cup-and-cone fracture at every temperature. On the other hand, in hydrogen gas, in both steels a number of cracks initiated on the specimen surface and coalesced with each other at every temperature, which led to a marked reduction in ductility. Nonetheless, even in hydrogen gas, JIS-SCM435 exhibited a certain reduction of area after the stress-displacement curve reached the tensile strength (TS), whereas JIS-SM490B exhibited little, if any, necking in hydrogen gas. In addition, tension-compression fatigue testing at room temperature revealed that in both steels there was no noticeable difference between the fatigue strengths in air and 115MPa hydrogen gas, especially in a relatively long life regime. Considering that there was little or no hydrogen-induced degradation in either TS or fatigue strength in JIS-SCM435, it is suggested that JIS-SCM435 is eligible for fatigue limit design on the basis of a safety factor (i.e. TS divided by the allowable design stress) for mechanical components used in hydrogen gas up to 115 MPa.Copyright

Measurement of Fracture Properties for Ferritic Steel in High-Pressure Hydrogen Gas

Recently, the measurement of threshold stress intensity factors for various low alloy ferritic steels in high-pressure hydrogen gas of 103 MPa was performed, and it was revealed that the subcritical cracking threshold under rising displacement was lower than the subcritical cracking threshold for crack arrest under constant displacement. These experimental results demonstrate the importance of the testing method for evaluating the fracture properties in high-pressure hydrogen gas. We measured the subcritical cracking threshold under rising displacement for ASME SA-372 Grade J ferritic steels in high-pressure hydrogen gas at pressure up to 115MPa. In contrast to other reported procedures where the applied displacement was increased continuously, in this study crack length was determined using an unloading elastic compliance method. The values of the subcritical cracking threshold measured by the unloading elastic compliance method are consistent with previous measurements in which the applied displacement continuously increased. These results suggest the possibility that subcritical cracking thresholds do not depend on the applied displacement path, i.e., periodic unloading vs. continuously rising displacement.Copyright

Effects of External and Internal Hydrogen on Tensile Properties of Austenitic Stainless Steels Containing Additive Elements

Effect of hydrogen on the slow strain rate tensile (SSRT) properties of five types of austenitic stainless steels, which contain small amounts of additive elements (e.g., nitrogen, niobium, vanadium and titanium), was studied. Some specimens were charged by exposing them to 100 MPa hydrogen gas at 543 K for 200 hours. The SSRT tests were carried out under various combinations of specimens and test atmospheres as follows: (i) non-charged specimens tested in air at room temperature (RT), (ii) non-charged specimens tested in 0.1 MPa nitrogen gas at 193 K, (iii) hydrogen-charged specimens tested in air at RT, (iv) hydrogen-charged specimens tested in 0.1 MPa nitrogen gas at 193 K, and (v) non-charged specimens tested in 115 MPa hydrogen gas at RT. In the tests without hydrogen (i.e., cases (i) and (ii)), the reduction of area (RA) was nearly constant in all the materials, regardless of test temperature. In contrast, in the tests of internal hydrogen (cases (iii) and (iv)), RA was much smaller at 193 K than at RT in all the materials. It was revealed that the susceptibility of the materials to hydrogen embrittlement (HE) can successfully be estimated in terms of the nickel equivalent, which represents the stability of austenite phase. The result suggested that the nickel equivalent can be used for evaluating the material compatibility of austenitic stainless steels for hydrogen service.Copyright

Fracture Properties of a Cr-Mo Ferritic Steel in High-Pressure Gaseous Hydrogen

Recently, the measurement of threshold stress intensity factors for various low alloy ferritic steels was performed in high-pressure hydrogen gas, and it was revealed that the threshold for subcritical crack extension under rising displacement was lower than that for crack arrest under constant displacement. We previously examined the threshold for subcritical crack extension of ASME SA-372 Grade J ferritic steels using an unloading elastic compliance method in gaseous hydrogen at pressure up to 115 MPa, and reported that the cracking thresholds under the unloading elastic compliance method are consistent with the cracking threshold under the continuously rising displacement method. In the current study, the threshold stress intensity factor of JIS SCM 435 steel with yield strength of 700 MPa was measured using the unloading compliance method in high-pressure hydrogen gas. The roles of environmental (gas pressure) and testing (displacement rate) parameters were evaluated. The hydrogen gas pressure ranged from 10 to 115 MPa, and the displacement rate was varied between 2×10−3 and 2×10−5 mm/s. The measured values of the threshold for subcritical crack extension, KJIC,H, decreased with either increasing hydrogen gas pressure or decreasing displacement rate. When the hydrogen pressure was 115 MPa, however, the cracking threshold showed a modest dependence on displacement rate, varying from 63 to 43 MPa m1/2. These results demonstrate that the measurement conditions, such as the hydrogen pressure and the displacement rate, affect values of the subcritical cracking threshold under rising displacement.Copyright

Effect of Hydrogen Gas on the Growth of Small Fatigue Crack in JIS-SCM435

The fatigue crack growth (FCG) from a small hole in a low alloy steel JIS-SCM435 round bar was investigated using tension-compression fatigue tests in 0.7 MPa hydrogen gas and ambient air. In the higher FCG rate regime (e.g. da/dN > 108 m/cycle), FCG was accelerated in hydrogen gas compared to in air. On the other hand, in the lower FCG rate regime (e.g. da/dN < 108 m/cycle), FCG in hydrogen was rather slower than that in air. There was no noticeable difference in fatigue limits between these two atmospheres. The FCG in the respective atmospheres showed a typical small crack behavior, i.e. the da/dN for small cracks were much greater than those for large cracks obtained by compact tension (CT) specimen when they were compared at the same ΔK level. In order to unify such a discrepancy of FCG behavior between small crack and large crack, the strain intensity factor range ΔKε was adopted. As a result, the da/dN data for various crack sizes was gathered in a narrow band, i.e. the small crack effect was successfully evaluated with the strain intensity. Moreover, the crack growth life was predicted based on the da/dN-ΔKε relation. The reproduced S-N curve showed a conservative agreement with the fatigue life obtained by experiments.