MURI Center for Material Failure Prediction through Peridynamics

1. Introduction - Identification of the research and issues: Although important discoveries and advancements have been made toward understanding the microstructure-to-material-property relationships and the influence that processing conditions have on both the material chemistry and the material microstructure, the fundamental challenge is still to predict the complex nonlinear relations between microstructure, material properties and processes while including failure initiation and progression. Such predictive capability will enable the exploration of the relationships between spatial arrangement of material constituents at the appropriate length scales and their bulk properties and performance, and will permit the description of material properties and failure behavior in terms of the micro-architectural parameters. Currently, the multiscale material models that account for structure variability at different scales pose significant mathematical and computational challenges when applied to predicting damage and failure from the micro-scale up and being able to design new materials with a desired performance at the macroscale. Our research team has extensive expertise in advancing PeriDynamics (PD) theory for coupled multi-physics problems and dynamic materials failure, multiscale materials modeling, material characterization at multiple length scales, stochastic and statistical modeling, mathematical analysis and convergence properties of PD models, and high performance computing. Therefore, we are in a unique position to develop a predictive capability based on the PD theory for multiscale and multi-physics modeling that will lead to the design of new materials and enable the exploration of micro-architectural features to manipulate properties suitable for the desired performance. This capability, based on the integration of materials science, experimental characterization, and computational continuum mechanics of failure and damage in heterogeneous systems, will enable rapid characterization of material structure and properties at different length scales. 2. Proposed research: PD theory permits the integration of micro-structural architecture and property relations while including the effect of stochastic and statistical variability of properties and the calibration against experimental measurements at various length scales. This includes the parameterization of selected properties as a function of micro-architectural features and dimensions, extension to bulk property length scales; and parameter calibration. To achieve these objectives, we address the following topics in an integrated manner: (1) Multiscale and multi-physics material fracture and failure modeling with PD, (2) Experimental characterization of material properties and damage behavior at multiple length scales, (3) Stochastic and statistical modeling, and homogenization methods for evolution of failure (4) Mathematical analysis and convergence properties of nonlinear PD models, (5) High performance computing (HPC) in PD, and (6) Validation and verification of the PD predictive tool.