This paper summarized our recent studies on the test specimen size (TSS) effects on Jc of a material in the ductile-to-brittle transition temperature (DBTT) region. The validity of the deterministic approach to transfer the fracture toughness Jc obtained with different thickness specimens is demonstrated in these works. Based on the detailed finite element analysis results, it was found that the crack-tip stresses were different at the identical J in the test specimen thickness (TST) effect on Jc observed with both the non-proportional and proportional specimens. And adjusting loads to make the stress level equivalent showed increment in J that was equivalent to the Jc difference due to TST effect on Jc. This was similar with the past result obtained for the planar size effect on Jc (the difference in Jc due to the planar specimen configuration including crack length difference for the same specimen type or the specimen type difference). Thus, it was concluded that all of the TSS effects on Jc could be explained as due to J’s disability to characterize the crack-tip stress field accurately, or in a more general explanation, due to the finite size effect. In addition, the (4δt, σ22c) failure criterion (Dodds et al., 1991) was verified to transfer Jcs obtained for different specimen thicknesses and planar configurations. The critical value σ22c varied for only a few percent. The fact that these critical values were always reached at the specimen mid-plane and the fact that cleavage always initiated at the specimen mid-plane supported the validity of the deterministic approach. Because the (4δt, σ22c) failure criterion requires only “single” set of test data for Jc transfer and because σ22c shows only a few percent scatter, it seems to have a possibility to replace what Weibull stress is expected to do.
