Carnegie Mellon University
Browse

Understanding Grain Growth in 3D: Microstructural Evolution of Nickel during Annealing using Three-dimensional X-ray Microscopy

Download (9.69 MB)
thesis
posted on 2021-06-10, 20:56 authored by Aditi BhattacharyaAditi Bhattacharya
This thesis characterizes and quantifies microstructural evolution in nickel. The orientation maps of nickel were collected using a non-destructive, synchrotron based 3D characterization technique called Near-field High Energy X-ray Diffraction Microscopy (nf-HEDM) by the Suter group at Carnegie Mellon University. Six snapshots of the nickel sample were collected after being annealed at 800°C in 30 minute intervals. The temporal changes in crystallographic orientation, grain size and misorientation in the microstructure were measured over the different anneal states. The observed microstructure was also modified to represent a more isotropic system by merging twin related domains into a single grain. Subsequently the microstructures were also meshed and smoothed to determine grain boundary properties as a function of the five grain boundary parameters. Various morphological and topological features such as grain size, number of neighbors, difference between nearest and average of second nearest neighbors, grain curvature and integral mean grain curvature were analyzed with respect to grain growth. It was observed that a grain’s volume change is best predicted by the difference between the number of neighbors a grain has and the average number of neighbors the neighboring grains have as well as curvature. However, the correlation between neighborhood topology and curvature becomes significant only when adjacent twin related domains are merged. A study of the influence of grain boundary curvature on grain boundary velocity revealed that they did not show a positive linear correlation. Grain boundary velocity and curvature also varied with crystallographic parameters. An isotropic simulation used to predict grain growth of the merged twin microstructure did not reproduce the observed microstructure.
A statistical data multivariate model called Canonical Correlation Analysis was used to examine grain growth in both twinned and twin merged microstructures. The evolution of the twin merged microstructure was better predicted than the evolution of the twinned microstructure. Input parameter sensitivity and interrelationship study revealed that grain curvature, grain radius and neighborhood topology were the most important factors that correlated to grain volume change.

History

Date

2021-02-26

Degree Type

  • Dissertation

Department

  • Materials Science and Engineering

Degree Name

  • Doctor of Philosophy (PhD)

Advisor(s)

Greg Rohrer

Usage metrics

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC