Methods for mass transport coefficient calculation in crystalline alloys used in irradiated environments

Chattopadhyay, Soham

Permalink

https://hdl.handle.net/2142/122120 Copy

Description

Title

Methods for mass transport coefficient calculation in crystalline alloys used in irradiated environments

Author(s)

Chattopadhyay, Soham

Issue Date

2023-11-27

Director of Research (if dissertation) or Advisor (if thesis)

Trinkle, Dallas R

Doctoral Committee Chair(s)

Trinkle, Dallas R

Committee Member(s)

• Bellon, Pascal
• Ertekin, Elif
• Schleife, Andre

Department of Study

Materials Science & Engineerng

Discipline

Materials Science & Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• diffusion
• Green's function
• interstitial dumbbells
• deep learning
• high entropy alloys
• variational principle for mass transport
• kinetic monte carlo

Abstract

Diffusion of atoms leads to microstructural phenomena that affect the properties as well as the processing conditions of a wide variety of materials such as steels, superalloys, semiconductors and those used in alternative energy devices. Such transport phenomena are also important in alloys for applications in extreme irradiated environments such as nuclear reactors, where excess mobile point defects such as vacancies and interstitials are responsible for often debilitating segregation and/or precipitation of solute elements. Accurate knowledge of the transport properties of solutes in alloys under irradiation is thus crucial for effective precautions and design against radiation damage. This dissertation presents some methodologies for transport coefficient calculations, building upon existing theories, that are efficient, and quantitatively reliable. In the first part of the dissertation, interstitial dumbbell-mediated diffusion of solutes in dilute alloys has been considered. Interstitial dumbbells are not important in conventional alloy applications due to their high formation energies, but under irradiation, they form in significant amounts due to impinging high-energy particles, and have been known to influence the microstructure and properties of irradiated components. Accordingly, the Green’s function method, originally developed for vacancy-mediated diffusion, was extended to compute transport properties due to interstitial dumbbell-mediated diffusion. This methodology neglects no correlations between atomic jumps, is fast enough to produce accurate transport coefficients in a few seconds at a given temperature on a single computer processor, and is suitable for coupling with high-throughput applications. A drawback of the Green’s function methodology is that it is practically feasible for only the dilute limit, where a single solute and a single defect are considered in an infinite lattice. Due to this, although physically relevant insights regarding diffusion mechanisms can be inferred, transport phenomena in concentrated systems cannot be studied with the Green’s function method. Moreover, recently, highly concentrated systems such as High Entropy Alloys (HEAs) have been gaining popularity due to their impressive resistance to radiation damage. For these systems, Kinetic Monte Carlo (KMC) based methods are most commonly used, aside from some recent semi-analytical models. However, the simplifying assumptions in such methods, as well as the high computational cost of KMC present challenges towards obtaining reliable estimates of transport properties. To address this problem, the second part of the dissertation establishes a methodology based on deep learning, in combination with the variational principle of mass transport, to compute reliable transport coefficients using data for just two KMC steps at the most, with vacancy-mediated diffusion of solutes in HEAs as the working example. It is shown that this method can lead to significant savings in migration barrier calculations compared to standard KMC. It is also shown that through statistical analysis, insights regarding mechanisms affecting vacancy diffusion can be obtained, along with empirical laws that enable quick prediction of transport coefficients across wide temperature ranges.

Graduation Semester

2023-12

Type of Resource

Thesis

Copyright and License Information

Copyright 2023, Soham Chattopadhyay

Owning Collections