You are here

Integrated Computational Materials Engineering in Multiphysics Software

Description:

OBJECTIVE: The objective of this research is to demonstrate a spatially-dependent calculation of detailed microstructural evolution (e.g., grain structure, texture, precipitation kinetics, phase transformations, etc.) in the modeling of processing of a structural material ultimately providing or enhancing commercial multiphysics software (e.g., general multiphysics or specialized for welding, forging, molding, casting, extrusion, thermal spraying, powder processing, etc.). This project supports the goals of the Materials Genome Initiative (MGI) in the area of Integrated Computational Materials Engineering (ICME). DESCRIPTION: During materials processing operations, the microstructure, physical properties, and mechanical properties of a material can change significantly, both globally and in localized regions of complex shapes. These changes can affect the behavior of the material during the processing operation, as well as the final properties and performance in service. Several simple examples include the evolution of spatially-varying microstructure and texture during the closed-die forging of titanium alloys and nickel-base superalloys which often involve complex strain, strain rate, and temperature histories. Despite such complexity, most research groups persist in using oversimplified phenomenological material models such as engineering stress-strain relations, Avrami-based recrystallization relations, and other related approaches. Though the bases for these models revolve around some physical concepts (e.g., nucleation and growth phenomena), the models themselves are often ad hoc or limited to a very specific processing regime over which measurements are available. The parameters that designers and analysts use to represent materials behavior are convenient for databases, and the models calculate rapidly for analysis, but predictions often match measured behavior only approximately and sometimes worse. A long-term vision for Integrated Computational Materials Engineering and the Materials Genome Initiative involves providing complete spatial descriptions of the evolution of microstructure, texture, and defects during materials processing operations and resulting service properties. This would enable the design of both the material and process for end-state properties tailored for a particular application. Current multi-physics software packages generally provide only rudimentary single-value, or temperature-dependent, material properties. Although single-parameter state variables provide useful insight for such parameters as"damage", they are not sufficient to describe completely (or even partially) the state of the material. Some codes provide the ability to estimate microstructural information, such as grain size or fraction of transformed/recrystallized phase developed during or following deformation based on phenomenological relations. In addition, mesoscale simulations such as those based on Monte-Carlo, cellular automaton, and phase-field approaches may provide spatially resolved information, but often such codes need input from global FEM type process-simulation runs. Such de-coupled methods (involving post-processing-FEM predicted field variables as input to mesoscale simulations of microstructure) are computationally simple, and provide simple system models for optimization, but may not totally account for real material non-linearity and the path dependence of microstructure evolution during multi-step processes. Furthermore, in many instances, only specialized university-developed (non-commercial) codes provide the ability to calculate microstructure and property evolution in complex solidification, deformation, and heat treatment operations. PHASE I: The successful phase I research will present a generic structure for complex data comprising of microstructure and properties. The investigators will provide a detailed plan to integrate this data structure, and the evolution of this data structure, within the existing confines of a thermomechanical-processing simulation code, that is suitable for modeling the multi-step manufacture of a metal with arbitrary boundary conditions such as finite element analysis. It is important to note that this data structure must be sufficiently flexible to allow modeling with a number of different microstructural and textural features/components. PHASE II: In the phase II effort, the investigators will carry out the plan devised in phase I and incorporate the materials microstructure data structure into their code, and demonstrate performance for pertinent microstructural features (or feature such as grain size, grain, shape, precipitate size/volume fraction) in a multi-step manufacturing process. They shall also demonstrate that they can expand the system on demand to accommodate a large number of features/states for a material, and associated local constitutive behaviors. Note that many interesting behaviors will include non-uniform deformation and large changes in local strain rate or temperature during processing. The code must accommodate this in a reasonable manner for the analyst/designer to understand and utilize. This effort must also include verification and validation of the simulation code and material models, quantification of uncertainty, transport of data between process and service-behavior simulations, and methods of describing the detailed pedigree and sharing of all material and process data for other computer-base applications. PHASE III: The design of components requires the ability to determine accurately the structure and properties that evolve during primary processing and manufacturing operations. This toolset will allow the metals suppliers to better simulate their processing operations, which will enable superior design of their processing and manufacturing operations. This is of considerable interest to the aerospace and automotive industries. This will speed the design process and result in systems optimized for superior reliability. PRIVATE SECTOR COMMERCIAL POTENTIAL/DUAL-USE APPLICATIONS: The processing of materials using non-uniform applied conditions is routine in all of processing and manufacturing. The local behavior of the material becomes extremely important to the response of the material, and the overall success in the processing operation. We anticipate that the specific code developed in this project will find immediate use in the processing industry for critical components in all sectors of the economy. We further anticipate that the approaches and methods developed in this project will find their way into other codes for process modeling and simulation. Although the methods and approaches may be too expensive computationally for proof-of-concept-level designs, we expect these approaches to become routine for the integrated detail design of materials, processes, manufacturing operations, and components.
US Flag An Official Website of the United States Government