PSI - Issue 68

A.F Perez et al. / Procedia Structural Integrity 68 (2025) 439–445

444

6

A. F. Perez et al. / Structural Integrity Procedia 00 (2025) 000–000

under the applied stress, leading to an asymmetrical deformation pattern and the visible "shear banding" effect. The anisotropic nature of this deformation could be exacerbated by inhomogeneities within the material’s microstructure, potentially arising from the manufacturing process or because of thermal exposure during the high temperature test (Poirier, 1985). • Secondly, the crack extension was full of discontinuities, and instead of propagating in a straight line, it propagated in a parabolic way, an effect known as the tunnelling effect. This effect occurs because the material experiences different stress states across its thickness: the central region is typically under plane strain conditions, while the outer regions near the surfaces are under plane stress (James & Newman, 2003). This non-uniform crack front can lead to inaccurate toughness measurements, as standard fracture toughness calculations often assume a uniform crack front across the specimen. In addition, it is worth pointing out that the specimens had no side grooves. This choice was made to maintain consistency with the tests carried out by Calvet (2024). Side grooves prevent the crack from deviating out of plane, ensuring it remains in Mode 1 loading conditions. They also eliminate the plane stress region at the sample’s surface and remove the shear lips, promoting plane strain conditions throughout the sample's thickness (if thick enough). As a result, side grooves encourage plane strain conditions, leading to a lower-bound fracture toughness value. The use of side grooves can remedy these major differences in values, since standard methods do not take account of the tunnel effect and the potential impact of the thickness of the material.

Fig. 5. (a) J-R curves – matching final crack extension; (b) Power law trend curve. 4.2. !" values – Comparison with miniature CT samples The toughness fracture values found by Calvet (2024) were in the range [85 – 95] kPa·m and the samples were 4 mm thick and 8 mm wide, for an initial crack length of 4 mm. For the same experimental conditions (tests at 600°C, DCPD method used), the values are much lower than those found in this study. Fracture properties are known to be affected by the size of the sample. For example, conditions of plane strain or plane stress at the crack tip of a specimen are dependent on its thickness and influence the fracture behaviour of steels. In the case of miniature fracture geometries, the limited thickness results in only a minimal crack tip constraint, making it difficult to achieve plane strain fracture properties. However, achieving plane strain fracture properties is crucial for irradiated steels, where irradiation-induced embrittlement increases yield stress and might facilitate plane strain conditions. Despite their limited thickness, miniature specimens do not provide significant constraints. Even in standard fracture geometries, most ductile materials, especially at high temperatures, do not meet the minimum thickness requirement for dominant plane strain conditions. For example, in P91 steel at 600°C, the minimum thickness should be about 20 to 50 mm. Therefore, in miniature CT specimens, plane stress conditions are more likely to be relevant