3D finite element and experimental study of the size requirements for measuring toughness on tempered martensitic steels

Abstract

The fracture properties of the tempered martensitic steel Eurofer97, which is among the main candidates for fusion power plant structural applications, were studied with two sizes of pre-cracked compact specimens (0.35T C(T) and 0.87T C(T)). The fracture toughness behavior was characterized within the temperature range -80 to -40 ºC. The ductile-to-brittle transition reference temperature, as defined in the ASTM standard E1921, was around T 0 ≈ -75 ºC. At -60 °C, it was found that two sets of toughness data obtained with 0.35T and 0.87T C(T) specimens are not consistent with the size adjustments recommended in the ASTM standard. It was then shown that the underlying reason of this inconsistency is an inappropriate specimen size limit of the ASTM standard for this type of steel. From published fracture toughness data on the tempered martensitic steel F82H steel, similar results were also highlighted. 3D finite elements simulations of the compact specimens were performed to compare the stresses and deformations at the onset of fracture. A local approach model based on the attainment of a critical stress and a critical volume was used to study the constraint loss phenomenon. Within the framework of this model, the strong toughness increase by reducing the specimen size could be satisfactorily explained.2

Investigation of constraint and warm prestressing effects by means of a local approach to fracture

“…It is noted that the model predicts the very weak temperature dependence on the lower shelf of fracture toughness. This is consistent with the results in [3,10]. Note that the critical stress is consistent with the Griffith fracture stress and the Weibull stress scale parameter, which are related to the micro cleavage strength of the material.…”

“…It is shown in [15,16] that a solution independent of the initial notch geometry is eventually attained provided that the opening of the notch exceeds five times the initial notch dimensions. This notch radius is consistent with that in [10]. Due to symmetry considerations, only one half (for 2D) or quarter of the specimen (for 3D) is modeled.…”

“…The r ⁄ À A ⁄ model (and its extensions) is based on the following two hypotheses [3,10]: (1) brittle fracture is triggered when a critical area A ⁄ (or volume V ⁄ ) of material encompasses a critical stress r ⁄ , and (2) the critical values r ⁄ and A ⁄ (or V ⁄ ) are temperature-independent material properties. Based on the assumptions, the model provides the unique value of cleavage fracture parameters.…”

“…Based on the assumptions, the model provides the unique value of cleavage fracture parameters. The critical stress r ⁄ is defined as the ultimate maximum principal stress r 1 [3,10]. The stressed area A ⁄ (or volume V ⁄ ) is defined as the area (or volume) of material where the maximum principal stress, r 1 , is higher than r ⁄ (r 1 > r ⁄ ).…”

“…Mueller and Spätig [10], Mueller et al [11] and Ilchuk et al [12] applied this model for the modeling of brittle fracture and WPS quantifications of fusion materials. These models are in principle able to predict both size and temperature effects on fracture toughness in the DBT region.…”