MIDAS, Microstructural Imaging using Diffraction Analysis Software

 

The MIDAS software package allows users to non-destructively image in three dimensions the microstructure, e.g. the interrelationships between the grains, making up crystalline materials.It takes as input x-ray diffraction data from area detectors, collected as a function of rotation about a single axis (omega).The output is a grain-by-grain map of crystallographic parameters in the irradiated volume including lattice parameters, orientations, positions and volumes and, in certain cases, grain shape.Currently only a few synchrotron facilities are available worldwide which can generate this type of diffraction data. The closest non-x-ray analogue is electron-backscatter diffraction, available much more widely, which provides 2D (surface) information with higher spatial resolution but poorer resolution of crystallographic parameters.

 

Distribution & Impact

MIDAS V1 is installed on the APS cluster for 1-ID users and installation on the Argonne internal (LCRC) cluster is being pursued. Distribution of code will be considered once a stable release is complete and with more experience is gained in use of the software.

MIDAS has been used successfully for data analysis by 1-ID users from General Electric Global Research (Yan Gao); ANL-Nuclear Engineering (Meimei Li);Colorado School of Mines (Aaron Stebner) and the University of Illinois (Armand Beaudoin). 

First scientific paper from MIDAS use: “High-Energy Synchrotron X-ray Techniques for Studying Irradiated Materials”, J.-S. Park et al, Journal of Materials Research (in press).

 

Funding Source

Development for different parts of MIDAS have been supported by General Electric Global Research & the Air Force Research Laboratory as well as operational funding from the APS (contract DE-AC02-06CH11357).

HPC porting supported as part of Argonne LDRD "Integrating Simulation and Observation: Discovery Engines for Big Data," Ian T. Foster, PI.

 

Please cite

A publication describing MIDAS is currently in preparation.
Related Publications
The MIDAS workflow is described in: “Big Data Staging with MPI-IO for Interactive X-ray Science”, J.M. Wozniak, H. Sharma et al, BDC-2014 (accepted).

Background work leading to MIDAS are described in:

H. Sharma, R.M. Huizenga and S.E. Offerman: A fast methodology to determine the characteristics of thousands of grains using three-dimensional X-ray diffraction. I. Overlapping diffraction peaks and parameters of the experimental setup Journal of Applied Crystallography. 45, 693 (2012).

H. Sharma, R.M. Huizenga and S.E. Offerman: A fast methodology to determine the characteristics of thousands of grains using three-dimensional X-ray diffraction. II. Volume, centre-of-mass position, crystallographic orientation and strain state of grains Journal of Applied Crystallography. 45, 705 (2012).

The first scientific paper to employ MIDAS results: “High-Energy Synchrotron X-ray Techniques for Studying Irradiated Materials”, J.-S. Park et al, Journal of Materials Research (in press).

 

Additional Development Goals

Over the next year or so, the following enhancements are planned:

  • Build a simplistic user interface
  • Track sample and processing metadata
  • Include all spacegroups using SGInfo.In the current implementation for FF-HEDM, only cubic materials are treated; this development will allow any crystallographic structure to be analysed.
  • Improved visualization.This will include developing interfaces between HEDM analysis output and both Dream3D and Paraview. 
  • Implement Hydra configuration for Far Field-HEDM analysis.Currently only a single, nominally centered Far Field detector is treated in analysis.The Hydra configuration, for which 4 GE detectors are used to simultaneously, allows the possibility of greater Q-resolution and Q-coverage.Implementing this configuration in analysis will also have the benefit of generalizing the analysis for other non-standard detector arrays / configurations.
  • Seed Near Field analysis with Far Field results.This offers potential for a drastic speed-up in the computationally-intensive Near Field analysis by limiting the search space.
  • Include intensity fitting in Near Field anaysis.In all existing near field analysis codes including MIDAS, diffraction data is thresholded to a single level to simplify and speed up computations, but important information is lost in this process.Including the full intensity information will provide greater analysis fidelity, aiding peak overlap & in so doing increasing the number of measured grains and/or higher levels of plastic deformation to be studied.
  • Implement stitching in Far Field analysis.Currently the field-of-view is limited to the beam size perpendicular to the sample rotation axis; which is horizontal and ~2mm for APS experiments.We will develop the ability to combine, or stitch, FF scans taken at different horizontal positions, which will allow the field of view to be arbitrarily large.At the outset, limited stitching (2-3x) will be pursued.