IUC07 Beyond 3D: Tools for tracking spatiotemporal microstructure evolution

← Back to Participant Projects and Use Cases

Main Task Area: TA-WSD
Other related Task Areas: TA-OMS, TA-CI
Possible connections within NFDI: NFDI4Ing, FAIRmat
Material/Data: Spatiotemporal point, triangle cloud/mesh, and tensorial field data from computer simulations of microstructure evolution at the microscopic and atomic scale
Main Success Scenario: The user can characterize the geometry (and topology) of the crystal defects and describe their motion and interaction in time by flexibly exchanging spatiotemporal data from representative-volume-element (RVE) models, dislocation dynamics codes and force-field/MD codes.
Added value for the MatWerk community: A computational geometry description for crystal defects (point, line, and surface patches) is connected with a corresponding ontology. This enables users to identify crystal defects agnostic of a specific implementation in a simulation code, enables a seamless encoding between the atomic and the microscopic/macroscopic scales and provides access to the study of the time evolution of defects.

Main requirements

  • Workflows to store/extract e.g. atomic positions, atom types, material point data (continuum, voxel data) including metadata and contextualization
  • Workflows to execute existing modeling tools for microstructure evolution (plasticity, recrystallization, grain growth, precipitation, dislocation dynamics, and force-field codes)
  • Workflows how to use existent MatWerk post-processing tools including metadata and contextualization
  • Development of ontology for defects
  • Tools for 3D visualization
  • Integration in institutional education

Related Participant Projects

Description

Spatiotemporal microstructure data are the key results of continuum full-field and atomic-scale computer simulations, as well as cutting-edge diffraction microscopy experiments. Typically represented as voxel, point, line, or finite-element meshes, these data cover a variety of materials information, such as the spatial distribution of crystal defects and their embedding into three-dimensional field quantities like temperature, mechanical stress and strain, or concentration fields of chemical species. Especially when many such time-dependent snapshots are taken, the volume and acquisition velocity can quickly become impractical to work with (GB, TB, or even larger). In addition, the large variety of computer codes within the MSE community, limits interoperability and repurposeability when hand-shaking between different microstructure characterization tools. Here, we aim for infrastructure work that describes crystal defects and their motion embedded in fields across different length and time scales in a code-agnostic representation that can complement existent code-specific outputs. Proof-of-concept workflows are developed, which exemplify how such description and ontology connection can be exported from existent MSE simulation tools, covering continuum-scale crystal plasticity, RX, GG, precipitation, dislocation dynamics, and atomistics.