The continuous design of cubic lattice architecture materials provides a wide range of mechanical properties. It makes possible to control the stress magnitude and the local maxima in the structure. This study reveals some architectures specifically designed to reach a good compromise between mass reduction and mechanical strength. Decreased local stress concentration prevents the early occurrence of localized plasticity or damage, and promotes the fatigue resistance. The high performance of cubic architectures is reported extensively, and structures with the best damage resistance are identified. The fatigue resistance and S–N curves (stress magnitude versus lifetime curves) can be estimated successfully, based on the investigation of the stress concentration. The output data are represented in two-dimensional (2D) color maps to help mechanical engineers in selecting the suitable architecture with the desired stress concentration factor, and eventually with the correct fatigue lifetime.

The grain size evolution of cold-rolled L-605 cobalt-base superalloy during ultra-rapid annealing is investigated in this paper. Cold-worked specimens undergo static recrystallization, leading to grain refinement or grain coarsening depending on the annealing time and temperature. The kinetics of grain growth is found to be independent of the initial deformation. The evolution of grain size can be simply described by a grain growth model for high temperatures and long annealing times, and the mobility of interfaces is estimated by modelling. Fast annealing treatment process is a very promising technique to customize grain size and enhance mechanical strength. In particular, the reduction of annealing time is an efficient method to produce a refined microstructure through static recrystallization.

A new criterion for nucleation in the case of dynamic recrystallization is proposed in order to include the contribution of the grain boundary energy stored in the microstructure in the energy balance. Due to the nucleation events, the total surface area of pre-existing grain boundaries decreases, leading to a nucleus size smaller than expected by conventional nucleation criteria. The new model provides a better prediction of the nucleus size during recrystallization of pure copper compared with the conventional nucleation criterion.