Issue 35

G. Gobbi et alii, Frattura ed Integrità Strutturale, 35 (2016) 260-270; DOI: 10.3221/IGF-ESIS.35.30 267 most of the failure surfaces showed brittle features, as in Fig. 5. Some cracks also propagated from lateral groves. Surely, to confirm the fully brittle behavior of this steel in presence of hydrogen more data would be necessary. Figure 4 : Contours of k (Eq.6) at crack tip region (continuum elements), plane strain model. In literature, there is no analytical law to calculate the hydrogen concentration in traps C T , to be compared with this numerical result. Moreover, the diffusion coefficient of this steel should be experimentally measured, including its effective and apparent contributions ( D L and D H ). If these two diffusivity values were available, their ratio would allow an estimation of C T , according to [7]. As a final comment to our result, it is even possible that the law expressed in Eq. 4 is not completely able to describe the mechanical behavior of AISI 4130, even if it included a wide class of materials. Fig. 6 compares force displacement (F-V LL ) curves from experimental tests and numerical simulations (plane strain and plane stress). The plot shows the mechanical responses of both models in hydrogen uncharged and charged conditions. It is evident the difference between the two behaviors without and with hydrogen. The macro-mechanical response of the specimen during toughness test is deeply influenced by hydrogen. The curve is decreased not only in its load peak, but also in the maximum displacement and in the damage history, as confirmed from TSL data in Tab. 2. Figure 5 : Fracture surface of a specimen after the toughness test. The specimen was hydrogen pre-charged before the test. Typical brittle surface is evidenced by facets (black arrow), intergranular cracks (white arrow) along grain borders and secondary cracks (yellow arrow).