PSI - Issue 68

Cainã Bemfica et al. / Procedia Structural Integrity 68 (2025) 1188 – 1195 Ludovic Vincent et al. / Structural Integrity Procedia 00 (2025) 000–000

1194

7

the Eqs. (1)-(4). For each loading step, the values of K Jc were calculated using the J-integral and associated with a value of 2 # calculated using Eq. (9). The surface creation energy parameter + is the only parameter of the MIBF model that needs to be identified on fracture toughness results. A choice is made to tune its value in order to describe at best the experimental results of the MM material (Figure 5, left). The corresponding value is + = 11.3 J/m². Then, the same value is applied to compute the isoprobabilities of the two other materials, HMM and IM.

IM

MM

HMM

Figure 5 – Comparison between experimental fracture toughness results (symbols) and fracture isoprobabilities (lines) obtained by the MIBF model for the three materials with an identical surface creation energy, + = 11.3 J/m². One can observe from the comparisons between experimental and simulated results plotted in Figure 5, that the model predicts fairly well the experimental results of HMM and IM material. Especially, the increase of the DBTT of HMM material is quite well captured by the model. The principal reason for the good performances of the model is the integration of the grain size effect proposed by Petch (Petch, 1986) on the critical size of carbides that should be taken into account to compute fracture probability. Also, the slightly softer mechanical behaviour of the IM material compensates the slightly higher probability to find large carbides in this material, compared to the MM material, resulting eventually in similar fracture toughness properties. 6. Conclusions In the present work, the influence of the carbide population and the ferritic grain size on the fracture toughness of three forged low alloy steels was modelled using the MIBF model. For such, three materials with similar chemical compositions, but significantly different thermomechanical histories, were investigated: a model material (MM), a homogenized model material (HMM) and an industrial material (IM). The only tuneable parameter of the model, the effective surface energy γ f , was identified on the MM results and subsequently used to estimate the DBTT of the HMM and the IM. Very good estimates were obtained. This result suggests that, despite the significantly different thermomechanical histories of the low-alloyed steels investigated in the present work and their subsequent dissimilar fracture toughness properties, the modelling approach proposed in this work succeeds to describe these fracture toughness distributions. These fracture properties were found to be driven mainly by the yield stress, the carbide population, and the ferritic grain size. 7. Acknowledgements The authors thank Gaëlle Leopold (EdF), Antoine Andrieu (EdF) and Daniel Brimbal (Framatome) for the results of the mechanical tests, and Amélie Gangloff (CEA) for her help in microstructural characterization. 8. References

Chekhonin, P., Das, A., Bergner, F., Altstadt, E., 2023. Microstructural characterisation of brittle fracture initiation sites in reactor pressure vessel