Tension-compression fatigue tests using smooth specimens of low carbon steel JIS-SM490B were carried out in air and hydrogen gas environment under the pressure of 0.7 and 115 MPa at room temperature. In 0.7 MPa hydrogen gas, fatigue life curve was nearly equivalent to that in air. On the other hand, in 115 MPa hydrogen gas, fatigue life was significantly degraded in the relatively short fatigue life regime (e.g. Nf < 105). To clarify the effect of hydrogen environment on fracture process, fracture surfaces of these specimens were observed. In general, fatigue fracture process of steels with low or moderate strength is macroscopically divided into 3 stages. In the first stage (stage I), fatigue cracks initiate in some crystalline grains. In the second stage (stage II), the cracks propagate stably on a cycle-by-cycle basis. In the final stage (stage III), a tilted fracture surface, shear-lip, is formed by ductile tearing. In SM490B steel, this general fracture process was confirmed in air and 0.7 MPa hydrogen gas. In contrast, in 115 MPa hydrogen gas, there was no tilted portion in the stage III region, and the fracture surface was totally flat. Observation with scanning electron microscope revealed that dimples were formed by ductile tearing in the tilted fracture region in air and 0.7 MPa hydrogen gas. On the other hand, only a quasi-cleavage fracture surface existed in the final fracture region in 115 MPa hydrogen gas. To understand the cause of this peculiar fracture morphology, we conducted elasto-plastic fracture toughness tests in each environment, and investigated the fracture morphology.

As a result of fracture toughness tests, crack growth rate in air and 0.7 MPa hydrogen gas was approximately equal to each other, and both the fracture surfaces were covered by dimples. This fracture morphology was in accordance with that of stage III morphology in fatigue specimen tested in air and 0.7 MPa hydrogen gas. However, in 115 MPa hydrogen gas, the crack growth was significantly accelerated, and the whole fracture surface was covered by quasi-cleavage.

In this paper, firstly, the similarity of fracture surface between two test methods, i.e. fatigue test and fracture toughness test, is investigated. And then, the formation mechanism of the flat fracture surface is discussed by paying attention to the crack-growth acceleration in high-pressure hydrogen gas.

This content is only available via PDF.
You do not currently have access to this content.