The diffusion of chloride ions in hardened cement paste (HCP) under steady-state conditions and accounting for the highly heterogeneous nature of the material is investigated. The HCP microstructures are obtained through segmentation of X-ray images of real samples as well as from simulations using the cement hydration model CEMHYD3D. Moreover, the physical and chemical interactions between chloride ions and HCP phases (binding), along with their effects on the diffusive process, are explicitly taken into account. The homogenized diffusivity of the HCP is then derived through a least square homogenization technique. Comparisons between numerical results and experimental data from the literature are presented.

A rate-dependent self-consistent crystal plasticity model was incorporated with the Marciniak-Kuczyński model in order to study the effects of anisotropy on the forming limits of BCC materials. The computational speed of the model was improved by a factor of 24 when running the simulations for several strain paths in parallel. This speed-up enabled a comprehensive investigation of the forming limits of various BCC textures, such as and fibers and a uniform (random) texture. These simulations demonstrate that the crystallographic texture has significant (both positive and negative) effects on the resulting forming limit diagrams. For example, the fiber texture, which is often sought through thermo-mechanical processing due to a high -value, had the highest forming limit in the balanced biaxial strain path but the lowest forming limit under the plane strain path among the textures under consideration. A systematic investigation based on the results produced by the current model, referred to as 'VPSC-FLD', suggests that the -value does not serve as a good measure of forming limit strain. However, model predictions show a degree of correlation between the -value and the forming limit stress.

In this study, an alloy phase-field model is used to simulate solidification microstructures at different locations within a solidified molten pool. The temperature gradient and the solidification velocity are obtained from a macroscopic heat transfer finite element simulation and provided as input to the phase-field model. The effects of laser beam speed and the location within the melt pool on the primary arm spacing and on the extent of Nb partitioning at the cell tips are investigated. Simulated steady-state primary spacings are compared with power law and geometrical models. Cell tip compositions are compared to a dendrite growth model. The extent of non-equilibrium interface partitioning of the phase-field model is investigated. Although the phase-field model has an anti-trapping solute flux term meant to maintain local interface equilibrium, we have found that during simulations it was insufficient at maintaining equilibrium. This is due to the fact that the additive manufacturing solidification conditions fall well outside the allowed limits of this flux term.

We study the evolution of prior columnar phase, interface phase, and phase during directional solidification of a Ti-6Al-4V melt pool. Finite element simulations estimate the solidification temperature and velocity fields in the melt pool and analyze the stress field and thermal distortions in the solidified part during the laser powder bed fusion process. A phase-field model uses the temperature and velocity fields to predict the formation of columnar prior-(Ti) phase. During the solidification of phase from an undercooled liquid, the residual liquid below the solidus temperature within the columns results in phase. The finite element simulated stress and strain fields are correlated with the length scales and volume fractions of the microstructure fields. Finally, the coalescence behavior of the (Ti) cells during solidification is illustrated. The above analyses are important as they can be used for proactive control of the subsequent modeling of the heat treatment processes.