Pengyang Zhao

An integrated full-field model of concurrent plastic deformation and microstructure evolution: Application to 3D simulation of dynamic recrystallization in polycrystalline copper

Abstract Many time-dependent deformation processes at elevated temperatures produce significant concurrent microstructure changes that can alter the mechanical properties in a profound manner. Such microstructure evolution is usually absent in mesoscale deformation models and simulations. Here we present an integrated full-field modeling scheme that couples the mechanical response with the underlying microstructure evolution. As a first demonstration, we integrate a fast Fourier transform-based elasto-viscoplastic (FFT-EVP) model with a phase-field (PF) recrystallization model, and carry out three-dimensional simulations of dynamic recrystallization (DRX) in polycrystalline copper. A physics-based coupling between FFT-EVP and PF is achieved by (1) adopting a dislocation-based constitutive model in FFT-EVP, which allows the predicted dislocation density distribution to be converted to a stored energy distribution and passed to PF, and (2) implementing a stochastic nucleation model for DRX. Calibrated with the experimental DRX stress–strain curves, the integrated model is able to deliver full-field mechanical and microstructural information, from which quantitative description and analysis of DRX can be achieved. It is suggested that the initiation of DRX occurs significantly earlier than previous predictions, due to heterogeneous deformation. DRX grains are revealed to form at both grain boundaries and junctions (e.g., quadruple junctions) and tend to grow in a wedge-like fashion to maintain a triple line (not necessarily in equilibrium) with old grains. The resulting stress redistribution due to strain compatibility is found to have a profound influence on the subsequent dislocation evolution and softening.

Effect of nonlinear and noncollinear transformation strain pathways in phase-field modeling of nucleation and growth during martensite transformation

The phase-field microelasticity theory has exhibited great capacities in studying elasticity and its effects on microstructure evolution due to various structural and chemical non-uniformities (impurities and defects) in solids. However, the usually adopted linear and/or collinear coupling between eigen transformation strain tensors and order parameters in phase-field microelasticity have excluded many nonlinear transformation pathways that have been revealed in many atomistic calculations. Here we extend phase-field microelasticity by adopting general nonlinear and noncollinear eigen transformation strain paths, which allows for the incorporation of complex transformation pathways and provides a multiscale modeling scheme linking atomistic mechanisms with overall kinetics to better describe solid-state phase transformations. Our case study on a generic cubic to tetragonal martensitic transformation shows that nonlinear transformation pathways can significantly alter the nucleation and growth rates, as well as the configuration and activation energy of the critical nuclei. It is also found that for a pure-shear martensitic transformation, depending on the actual transformation pathway, the nuclei and austenite/martensite interfaces can have nonzero far-field hydrostatic stress and may thus interact with other crystalline defects such as point defects and/or background tension/compression field in a more profound way than what is expected from a linear transformation pathway. Further significance is discussed on the implication of vacancy clustering at austenite/martensite interfaces and segregation at coherent precipitate/matrix interfaces.Structural transformation: a less linear approachA method for modeling complex changes in the crystal structures of solids is developed by researchers in the USA. Yunzhi Wang from the Ohio State University and his colleagues’ model provides a more accurate description of crystal structure rearrangement during a phase change known as martensitic transformation. Even though this structural evolution has be modeled successfully using the phase-field microelasticity theory, the existing models do not incorporate some the more complex nonlinear transformation pathways that have been seen when using atomistic simulations. Wang and co-workers now extended phase-field microelasticity theory to include these complex pathways. They show that configuration and activation energies of a critical nucleus of the martensitic phase differ significantly when such nonlinear coupling is considered. This model has applications to understanding structural transformations in metals and ceramics.

New Insights into Deformation of Metallic Glasses by Combining Mesoscale Simulation and Fluctuation Electron Microscopy

We present a new mesoscale deformation model of metallic glasses (MGs) that incorporates nanoscale atomic ordering information from fluctuation electron microscopy (FEM). Understanding the deformation mechanism of MGs, including shear banding at room temperature, is crucial to overcome their limited ductility and early failure, which have been a major obstacle to widespread applications of MGs. The current understanding attributes shear band initiation to collective interaction among shear transformation zones (STZs) where shear deformation occurs among a group of atoms in a nanoscale volume [1]. We previously showed that mesoscale simulation involving diverse STZ types can provide useful insights into shear banding and overall deformation characteristics of MGs beyond the time and length scale of atomistic simulations [2]. Here we demonstrate that more realistic deformation simulations can be achieved by incorporating the experimentally measured medium range ordering (MRO) information using FEM directly into the heterogeneously randomized STZ environment in our model. Our approach is based on the assumption that each MRO type, which resides at the same length scale of STZ, will have a different set of parameters that characterize the STZ events, which has also been indicated by others [see, e.g., 3].

Direct Determination of Structural Heterogeneity in Metallic Glasses Using Four-Dimensional Scanning Transmission Electron Microscopy

We report the first direct quantification of the structural heterogeneity in metallic glasses using intensity variance and angular correlation analyses of the 4-dimensional (4-D) scanning transmission electron microscopy (STEM) data. We demonstrate that the real-space reconstruction and analyses of the 4-D nanodiffraction data acquired using a pixelated fast STEM detector enables quantitative determination of the details of local structural heterogeneity, including the type, size, volume fraction and spatial distribution of local ordering at the nano- to meso-scale, beyond the limits of the previous measurements using conventional detectors. We show that different types of local ordering are present in Zr55Co25Al20 glass, leading to a high degree of structural heterogeneity, with the total volume of locally ordered regions making up to ∼14% of the entire volume. These findings are significant, as the structure-property relationship in metallic glasses and other amorphous materials has been difficult to establish because of the lack of detailed structural information from experiments.