ITR: Physics-Based Modeling of Plastic Flow that Couples Atomistics of Unit Processes with Macroscopic Simulations

这是一项赠款，用于响应提交给信息技术研究（ITR）倡议的（小）提案。 这项研究将与劳伦斯利弗莫尔国家实验室（LLNL）和牛津大学（英格兰）的团队合作完成。 三维模拟越来越多地为材料加工以及支持各种技术的组件的设计和制造提供了强大的基于信息的方法。 这项研究的目的，它的重点是现象，涉及金属材料的大塑性变形，是开发一个基于物理的方法，精确建模。 在整个系统级，即，在材料处理或复杂载荷下的部件的情况下，拉格朗日和任意拉格朗日-欧拉（ALE）有限元实现是大多数软件的计算基础。 然而，在原子和分子尺度上出现的独特单元过程通常控制临界现象，这是许多模型明显不足的地方。 特别是，几乎所有的非弹性模拟代码仅限于一类狭窄的晶体材料，即那些主要包括面心立方（fcc）材料的紧密堆积。 有一个挑战，开发更广泛的工程材料的精确模型，并适应这些模型到目前的大规模有限元，流体动力学和位错动力学代码。 本研究的重点是发展多尺度模型和算法，用于精确和可验证的模拟具有复杂非平面位错核结构的金属材料的变形行为。研究的重点是三维原子结构的缺陷，他们的流动性，和宏观塑性流动之间的关系。 将开发的多尺度方法将适用于许多领域，包括纳米技术问题。 主要目标是：（1）发展一套严谨的方法，将电子、原子、介观和宏观尺度的理论联系起来;（2）发展基于物理的连续本构关系，解释非平面位错核产生的复杂现象;（3）探索这些缺陷对应变局部化和断裂等关键现象的影响。 我们将考虑一系列来自不同晶体类别的技术上重要的材料，以及在材料加工和承受机械载荷的组件中出现的条件。要开发的算法将准备安装到大规模有限元代码中，例如，Abaqus，无论是在纳米级制度和宏观成分的多晶体。 尽管本构关系的结构与目前使用的结构有显着不同，但在大规模并行代码中的实现将是简单的。 %这是一项资助，以响应提交给信息技术研究（ITR）倡议的（小）提案。 这项研究将与劳伦斯利弗莫尔国家实验室（LLNL）和牛津大学（英格兰）的团队合作完成。 研究的重点是缺陷的三维原子构型，它们的流动性和金属材料中的宏观塑性流动之间的关系。 将开发的多尺度方法将适用于许多领域，包括纳米技术问题。 主要目标是：（1）发展一套严谨的方法，将电子、原子、介观和宏观尺度的理论联系起来;（2）发展基于物理的连续本构关系，解释非平面位错核产生的复杂现象;（3）探索这些缺陷对应变局部化和断裂等关键现象的影响。 我们将考虑来自不同晶体类别的一系列技术上重要的材料，以及在材料加工和承受机械载荷的组件中出现的条件。

This is a grant funded in response to a (small) proposal submitted to the Information Technology Research (ITR) Initiative. The research will be done in collaboration with groups at Lawrence Livermore National Laboratory (LLNL) and Oxford University (England). Three-dimensional simulations are increasingly providing a powerful information-based approach to both material processing and the design and manufacturing of components that support a wide range of technologies. The objective of this research, which focuses on phenomena that involve large plastic deformations of metallic materials, is to develop a physics-based methodology for accurate modeling. At the overall system level, i.e., microstructural in the case of materials processing or components under complex loading, both Lagrangian and Arbitrary Lagrangian-Eulerian (ALE) finite element implementations are the computational basis of most software. However, unique unit processes that arise at atomic and molecular scales often control critical phenomena, and this is where many models are significantly deficient. In particular, nearly all inelastic simulation codes are limited to a narrow class of crystalline materials, namely those that are close-packed which primarily includes face-centered-cubic (fcc) materials. There is a challenge to develop accurate models for a broader range of engineering materials and to adapt these models into current large-scale finite element, hydrodynamic, and dislocation dynamics codes. The focus of this research is on the development of multiscale models and algorithms for the accurate and verifiable simulation of the deformation behavior of metallic materials possessing complex - non-planar - dislocation core structures. The research focuses on the relationship between the three-dimensional atomic configurations of defects, their mobility, and macroscopic plastic flow. The multiscale methodologies to be developed will be applicable to many areas, including problems in nanotechnology. The principal goals are: (1) to develop a rigorous methodology to link theories at scales ranging from electronic and atomic through mesoscale and macroscopic; (2) to develop physically-based continuum constitutive relations that account for complex phenmonea arising from non-planar dislocation cores; and, (3) to explore the effects of such defects on critical phenomena such as strain localization and fracture. We will consider a range of technologically important materials from different crystal classes and under conditions that arise in both material processing and in components subjected to mechanical loading.The algorithms to be developed will be ready for installation into large-scale finite element codes, e.g., Abaqus, both for polycrystals in nanoscale regimes and macroscopic components. Although the structure of the constitutive relations will be significantly different from those currently in use, implementation in massively parallel codes will be straightforward. %%%This is a grant funded in response to a (small) proposal submitted to the Information Technology Research (ITR) Initiative. The research will be done in collaboration with groups at Lawrence Livermore National Laboratory (LLNL) and Oxford University (England). The research focuses on the relationship between the three-dimensional atomic configurations of defects, their mobility, and macroscopic plastic flow in metallic materials. The multiscale methodologies to be developed will be applicable to many areas, including problems in nanotechnology. The principal goals are: (1) to develop a rigorous methodology to link theories at scales ranging from electronic and atomic through mesoscale and macroscopic; (2) to develop physically-based continuum constitutive relations that account for complex phenmonea arising from non-planar dislocation cores; and, (3) to explore the effects of such defects on critical phenomena such as strain localization and fracture. We will consider a range of technologically important materials from different crystal classes and under conditions that arise in both material processing and in components subjected to mechanical loading.***