Materials Science Forum Vols. 558-559

Abstract: In the present work computer simulations using a 3-D Potts Monte Carlo model are demonstrated and discussed as a tool to study the effects of a number of parameters related to the deformation conditions as well as process and material parameters related to the nucleation and growth conditions of recrystallisation, such as e.g. second phase particles, which may influence the kinetics and texture of recrystallisation. The MC simulations have been combined with a deformation texture model to provide the deformation structure from which the recrystallisation nucleates and models which provide the relative proportions of certain nucleation mechanisms and their orientation spectrum. All together this gives a simulation tool which allows for a multitude of numerical experiments and the possibility to study parameter relationships which are often not easily available from experiments. The potential of such a simulation tool is discussed in terms of a few generic examples.

Abstract: The current study aims to improve our fundamental understanding of solute segregation and solute drag on migrating grain boundaries (GB) in three dimensions. Computer simulation combines finite difference and finite element methods. An exemplary case study is reported, in which a spherical grain is embedded inside a cubic grain and shrinks as a result of motion by curvature, as a preliminary to modeling grain growth in single phase materials. The results agree qualitatively with literature studies in 1-D.

Abstract: Recrystallization is governed by the migration of high angle grain boundaries traveling through a deformed material driven by the excess energy located primarily in dislocation structures. A method for investigating the interaction between a migrating grain boundary and dislocation boundaries using molecular dynamics (MD) was recently developed. During simulations migrating high angle grain boundaries interact with dislocation boundaries, and individual dislocations from the dislocation boundaries are absorbed into the grain boundaries. Results obtained previously, using a simple Lennard-Jones (LJ) potential, showed surprisingly irregular grain boundary migration compared to simulations of grain boundary migration applying other types of driving forces. Inhomogeneous boundary-dislocation interactions were also observed in which the grain boundaries locally acquired significant cusps during dislocation absorption events. The study presented here makes comparisons between simulations performed using a LJ- and an embedded atom method (EAM) aluminum potential. The results show similarities which indicate that it is the crystallographic features rather than the atomic interactions that determine the details of the migration process.

Abstract: The kinetics of grain growth in real systems is influenced by several unknown factors, making a theoretical treatment very difficult. Idealized grain growth, assuming all grain boundaries to have the same energy and mobility (mobility M = k/ρ, where k is a constant and ρ is grain boundary curvature) can be treated theoretically, but the results obtained can only be compared to numerical grain growth simulations, as ideal grain growth scarcely exists in nature. The validity of the simulation techniques thus becomes of great importance. In the present investigation computer simulations of grain growth in two dimensions using Monte Carlo simulations and the grain boundary tracking technique have been investigated and compared in small grain systems, making it possible to follow the evolution of each grain in the system.

Abstract: Abnormal grain growth (AGG) proceeds in case that normal grain growth is inhibited. It has long been known that the inhibition involves finely dispersed particles and/or the development of specific textures. There is another strong obstacle against the grain boundary (GB) motion; the solute atoms can reduce their energy by moving from the bulk into a GB. Resultant interaction between the solute atoms and a GB makes the GB motion more difficult. However the role of the GB segregation effect on AGG has not been clarified. In this study we simulate the 2D and 3D grain growth accompanying boundary segregation of solute atoms by using a phase-field model. It is shown that the segregation plays an important role on the occurrence of AGG. The boundary-segregation-induced AGG can take place when the average driving force of grain growth approaches a critical condition for pinning-depinning transition in solute-drag atmosphere.

Abstract: Phase-field model (PFM) in multiple orientation fields was used to simulate the grain growth in three-dimensions (3-D) for isotropic and anisotropic grain boundary energy. In the simulation, the polycrystalline microstructure was described by a set of non-conserved order parameters and each order parameter describes each orientation of grains. For isotropic grain boundary energy, the simulation showed the microstructure evolution of normal grain growth. For anisotropic grain boundary energy, however, the simulation showed that certain grains which share a high fraction of low energy grain boundaries with other grains have a high probability to grow by wetting along triple junctions and can grow abnormally with a growth advantage of solid-state wetting. The PFM simulation shows the realistic microstructural evolution of island and peninsular grains during abnormal grain growth by solid-state wetting.

Abstract: The article deals with FEM analysis, with mathematical modeling of structure evolution and with prediction of mechanical properties during hot rolling of seamless steel tubes. With use of DEFORM 3D software the present state of stretch-reducing mill of Železiarne Podbrezová, Inc was described. The next part of article presents mathematical model describing the tube production for low carbon steels grade St52 on analyzed calibration sequence. Values of temperature, strain and strain rate from FEM simulation were substituted into the mathematical models of recrystallization, grain growth, transformation and prediction of mechanical properties. The calculated values were compared with experimental ones resulting from mechanical tests of industry rolled tubes of various chemical compositions. Good correlation of structural and mechanical properties was achieved for all chemical compositions and treatment conditions.

Abstract: Coarsening of particles during liquid phase sintering is known to be an example of Ostwald ripening. This coarsening process, in a fully wetting system, is simulated in three dimensions with a kinetic Monte Carlo model. The results from the simulation for microstructures, kinetics and particle size distributions are compared to available experimental findings. It is found that the average particle volume increases linearly with time and that the particle size distributions are consistent with those obtained experimentally, as in the W-Ni-Fe and Sn-Pb systems.

Abstract: In this study the strain-induced grain growth was simulated on an aluminum bicrystal by using channel-die compression. After compression of the bicrystal up to 5% deformation the strain mapping were characterized by using digital image correlation (DIC) technique and the 2D strain filed provided data to simulate grain growth using a modified Monte Carlo method. The strain-induced grain growth on grain boundary was simulated and compared with experiment after annealing at 450°C for 4 hours. The relation between the deformation heterogeneity and the grain growth was discussed in this work.

Abstract: A physically based model is used to describe the microstructural evolution of Nb microalloyed steels during hot rolling. The model is based on a physical description of dislocation density evolution, where the generation and recovery of dislocations determines the flow stress and also the driving force for recrystallization. In the model, abnormally growing subgrains are assumed to be the nuclei of recrystallized grains and recrystallization starts when the subgrains reach a critical size and configuration. The model is used to predict the flow stress during rolling in SSAB Tunnplåt’s hot strip mill. The predicted flow stress in each stand was compared to the stresses calculated by a friction-hill roll-force model. Good fit is obtained between the predicted values by the microstructure model and the measured mill data, with an agreement generally within the interval ±15%.