THE SCHOOL OF MATERIALS SCIENCE AND ENGINEERING 

GEORGIA INSTITUTE OF TECHNOLOGY   

Under the provisions of the regulations for the degree
 

DOCTOR OF PHILOSOPHY
 

on Friday, November 12, 2021

12:00 PM

via

BlueJeans Video Conferencing

https://bluejeans.com/465223379/8391 

will be held the

DISSERTATION PROPOSAL DEFENSE

for

Kevin Chu

"Atomistic and Coarse-Grained Atomistic Modeling of Solute Ordering Effects on Dislocation Migration"

Committee Members:

Prof. David L. McDowell, Advisor, ME/MSE

Prof. Ting Zhu, ME/MSE

Prof. Naresh Thadhani, MSE

Prof. Richard Neu, ME/MSE

Edwin Antillon, Ph.D., U.S. Naval Research Laboratory

Abstract:

Computationally assisted design of novel alloy systems requires predictive models of dislocation interaction with heterogenous spatial distributions of solute atoms and their associated ordering. There is a pervasive tradeoff between accuracy and computational cost in modeling phenomena at multiple length scales thus motivating the use of scale-bridging approaches, or more broadly, multiscale modeling. One such example is the calculation of dislocation mobility functions using molecular dynamics as input to discrete dislocation dynamics models in a sequential multiscale approach. This approach extends the accessible model length scales but neglects the details of dislocation core effects. Determining such parameters and bridging inputs for a wide range of temperatures and compositions becomes computationally expensive for problem domains concerning compositionally complex alloys, which exhibit expansive parameter spaces. Alternatively, a combined approach is proposed leveraging existing coarse-grained concurrent atomistic-continuum (CAC) capabilities while employing average-atom (A-atom) interatomic potential formulations in order to reduce the spatial degrees of freedom in atomistic systems. This approach preserves the heterogeneity of random solutes at full atomistic resolution in critical regions of interest, such as the dislocation core, while accurately representing bulk alloy elastic behavior and long-range stress fields with finite elements in the continuum domain. We leverage this method to quantify the effect of solute strengthening in both fully random alloys and alloys exhibiting short-range ordering, using 316L austenitic stainless steel as a model system. Then, nudged elastic band methods will be implemented into CAC to support analysis of critical energetic pathways for bypass of obstacles such as solute clusters and second phase precipitates.

Complementary atomistic and coarse-grained atomistic approaches will be employed to probe dislocation strengthening behavior in random alloys to support ICME-based decision-making frameworks for alloy design. Key outcomes of this work include quantification of composition and temperature effects on dislocation mobility in 316L austenitic stainless steels to inform DDD simulations, development and validation of the CAC/A-atom approach for random alloys, application of CAC/A-atom simulations to study the effect of chemical short-range order on yield strength as relevant to novel additively manufactured alloys, and implementation of a coarse-grained atomistic nudged elastic band method to assess critical energy pathways for extended length scale dislocation bypass unit processes.  Associated computational workflows will be developed to support screening of various alloy design parameters.