Collaborative Research: NSF-EU Materials Collaboration:Multiscale Modelling of Recrystallization in Metals Based on a Digital Material Framework

该奖项是为了响应在材料世界网下提交的NSF和EC在计算材料方面合作研究的提案而颁发的。卡内基梅隆大学（CMU）和普林斯顿大学提议与欧洲的三个机构合作，解决建模再结晶的多尺度挑战。该项目的美国方面将解决材料的三维表示问题，包括微观结构和状态的相关方面，导致成核的非均匀亚晶粗化，恢复过程中的位错动力学以及与微观结构演变相关的晶界特性。例如，水平集方法的最新发展开辟了新的可能性，使我们能够定量预测在恢复过程中发生的复杂错位运动，包括爬升和滑动。在较粗的长度尺度上，亚晶粒粗化的建模和理论已经发展到可以应用于变形材料中观察到的复杂和可变结构的地步。因此，该项目将对材料的基础研究产生具体影响，通过定量预测模型开发对微观结构演变过程的新理解。通过定期进行国际交流对学生进行教育，并通过开发可轻易转移到工业中模拟商品材料加工的工具，它也将产生广泛的影响。合作的总体目标是开发数字微观结构方法，该方法将开发计算工具，通过边界运动所需的原子尺度的数值表征和模拟，通过变形的位错动力学，以及在细胞和亚晶粒尺度上的恢复，解决非均匀塑性变形与恢复和再结晶过程的组合问题。以晶粒尺度方面的塑性变形，加上边界的远距离运动（如再结晶前沿）。该团队的美国成员将专注于关键方面，如晶界的动力学特性（迁移率），恢复中的位错动力学和再结晶过程的介观模拟，包括所有相关特征，如晶粒结构和结晶取向。将通过比较计算微观结构的统计数据与实验测量的微观结构（基于先前工作的可用数据）来验证这组计算工具的预测能力。该提案概述了整个合作项目，但只有在美国工作的那些方面得到了全面描述。确定了将在美国实施的项目的要素

This award is being made in response to a proposal submitted under the Materials World Net for collaborative research on computational materials between the NSF and EC. Carnegie Mellon University (CMU) and Princeton University propose to team with three institutions in Europe to address the multiscale challenges of modeling recrystallization. The US side of the project will address issues of three-dimensional representation of materials that include the relevant aspects of microstructure and state, heterogeneous subgrain coarsening leading to nucleation, dislocation dynamics during recovery and grain boundary properties relevant to microstructural evolution. Recent developments in level set methods, for example, have opened up new possibilities and will allow us to make quantitative predictions of the complex dislocation motions that occur during recovery that include climb as well as glide. At a somewhat coarser length scale, modeling and theory of subgrain coarsening has progressed to the point where it should be applied to the complex and variable structures observed in deformed materials. Thus the project will have specific impact on fundamental research on materials in terms of developing new understanding of the processes of microstructural evolution via quantitative, predictive models. It will also have broad impact through the education of students with regular international exchanges and through the development of tools that can be readily transferred to industry for modeling the processing of commodity materials. The overall thrust of the collaboration is to develop the Digital Microstructures approach that will develop computational tools that address the combination of heterogeneous plastic deformation with recovery and recrystallization processes through numerical characterization and simulation from the atomistic scale required for boundary motion, through dislocation dynamics of deformation followed by recovery at the cell and subgrain scale, to the grain-scale aspects of plastic deformation, coupled with long-range motion of boundaries (as recrystallization fronts). The US members of the team will concentrate on key aspects such as kinetic properties of grain boundaries (mobility), dislocation dynamics in recovery and mesoscopic simulation of the recrystallization process including all relevant features such as grain structure and crystallographic orientation. Verification of the predictive capability of the set of computational tools will be conducted by comparing the statistics of the computed microstructures to microstructures measured experimentally (based on data available from prior work). This proposal outlines the entire collaborative project but only those aspects to be worked on in the US are described in full. The elements of the program to be conducted in the US are identified.***