Issue 78

V. I. Kolesikov et alii, Fracture and Structural Integrity, 78 (2026) 220-232; DOI: 10.3221/IGF-ESIS.78.13

regime. In the plastic regime, the equiaxial 40CrNiMo phase has a somewhat lower resistance to plastic flow compared to the columnar one, as evidenced by higher vM ε 〈 〉 values at the same global load step. In Fig. 4c, the large / a c vM τ σ 〈 〉 〈 〉 values indicate that internal slip resistance is higher as compared to the macroscopic stress, implying that plastic flow is inhibited and a very strong/hardened microstructure is formed. The decrease in / a c vM τ σ 〈 〉 〈 〉 with strain indicates that vM σ grows faster than a c τ 〈 〉 , meaning the effective plastic flow resistance drops. Finally, the plateau in the / a c vM τ σ 〈 〉 〈 〉 plot implies flow at nearly constant resistance. Consequently, this mechanistic indicator reveals how the internal resistance to the plastic flow evolves with vM ε 〈 〉 . RVEs of soft/hard composites “40CrNiMo/Cu/40CrNiMo” and “Cu/40CrNiMo/Cu” Figs. 5 a-h, 6 a-h and 7 a-h display interfacial contacts between the soft and hard phases exhibiting different microstructure types. Specifically, Fig. 5 demonstrates the Cu phase with an equiaxed microstructure sandwiched between a columnar 40CrNiMo phase, while Fig. 7 shows the reverse arrangement. Comparing these to the RVE of hard/hard composite “40CrNiMo/40CrNiMo/40CrNiMo” discussed in the preceding section, one can see the significant drop in maximum P and vM σ values (see Figs. 5 b, 5 c, 6 c, 6 g, 7 b, and 7 c) due to the stress redistribution by the softer Cu phase, as evidenced by the larger deformations of the Cu layers compared to the almost non-deformed 40CrNiMo phase in Figs. 5-7.

a)

b)

d)

c)

40CrNiMo

ε vM F p

Cu

σ vM , Pa

P , Pa

40CrNiMo

f)

e)

h)

g)

γ a

F e

τ c a , Pa

Figure 5: Simulated fields for RVE of soft/hard composite “40CrNiMo/Cu/40CrNiMo” with columnar and equiaxial microstructures at t =144 s. This is particularly evident through IPF color legend changes in the Cu phase, contrasting with almost no change in the 40CrNiMo phase (see Figs. 6 a and 6 e). However, the a c τ values remain inflated due to the latent work-hardening of inactive slip systems (see Figs. 5 f, 6 f, and 7 f). As expected, plastic deformation is localized in grain colonies with easy active slip transfer (see Figs. 3 c, 3 g, 6 c and 6 g). It is noteworthy that the observed phenomenon must not correlate with a single slip system in polycrystalline materials [28]. Although the largest equivalent von Mises stress concentration should be observed at grain boundaries with the largest crystal orientation mismatch, the exact mapping could be complex due to the three-dimensional morphology of the microstructures and latent work-hardening. In principle, this can be studied from our resulting fields in more detail c τ magnitude and vM ε maps along the x -axis in Figs. 6b and 6d. While they can be removed with higher 64 × 64 × 128 RVE pixel resolution, doing so would incur a prohibitive computational cost and significantly increase file sizes for us. As demonstrated and discussed below, both these artifacts and the potential increase in resolution have a negligible effect on the plotted properties. for individual grains, but it is out of the scope of this study. Note that there are some faint periodic patterns in the a

227