Ph.D. Defense - Lun Chan

Committee:        

Professor Robert Kelly (MSE), Committee Chair

Professor Bi-Cheng Zhou (MSE), Committee Member

Professor Ji Ma (MSE), Committee Member

Professor Christopher Paolucci (CHE), Committee Member

Professor John Scully (MSE), Advisor

Dissertation Abstract

Molten salt is becoming an integral part of tomorrow’s energy. Notably, it offers a combination of heat capacity, vapor pressure, and thermal stability that makes the development of Generation IV molten salt reactor (MSR) technology possible. The MSR reactor utilizes a molten fluoride salt between 500°C and 700°C as the medium for nuclear fission, heat storage, and transfer. Examples of molten fluorides are eutectic mixtures of LiF-NaF-KF (46.5-11.5-42 mol%), LiF-BeF₂ (66-34 mol%), and NaF-ZrF₄ (59.5–40.5 mol%) salts. These molten fluorides are prone to containing dissolved hydrofluoric acid (HF) due to moisture and/or metallic cation impurities (e.g., Fe, Cr) that drive the corrosion of candidate structural containment alloys, which include 316SSH, Hastelloy N, and Inconel 600, and may possibly form a damage morphology that is susceptible to the loss of engineering function. Corrosion in molten fluorides remains a critical challenge today. The underlying factors that regulate the corrosion process must be well understood to prevent materials failures and ensure the overall sustainability of MSR development.

A key knowledge gap in understanding molten salt corrosion is the inadequacy of mechanistic insights and interpretations that target the fundamental principles of electrode reactions, such as the identification of reaction rate factors at the electrode-salt interface or transport through metallic or ionic conductor phases when alloys are exposed to molten fluorides. Particularly for alloy systems, it is unknown whether there exists a “window of corrosion susceptibility” where select metals and alloys are immune or susceptible to a particular form of corrosion damage, e.g., bicontinuous dealloying, uniform corrosion, grain boundary attack, or a mixture of them. Current progress has been limited primarily by capsule or flow-loop types of time-based exposures followed by surface characterization after periods of exposure. This approach often lacks diagnostics, does not assess rates except by serial exposure, and makes it difficult to separate variables critical to the corrosion process, leaving questions such as: (1) What is the valence of metal dissolution? (2) What are the rate-determining steps governing corrosion rate? (3) What is the role of microstructure and salt chemistry? The real challenge remains that these experiments are often high fidelity. Therefore, investigation into the corrosion behavior of structural metals should begin with model materials (such as pure metal or high-purity binary alloys), with a focus on targeting the fundamental principles of corrosion thermodynamics and kinetics. 

The objective of this work is to develop a mechanistic understanding of the corrosion process of model metals and alloys (Cr, Ni, Ni-Cr) in molten LiF-NaF-KF (or FLiNaK) salts using the principles of electrochemical thermodynamics and kinetics. As a foundation, the spontaneity and thermodynamic driving force for metal dissolution in molten fluorides were first assessed by developing potential-activity diagrams for Cr in molten fluorides (akin to Pourbaix Diagrams), which were verified by experimental cyclic voltammetry measurements (CV) for Cr(II) and Cr(III) in molten FLiNaK at 600°C. The thermodynamic predictions were then utilized to understand the dissolution kinetics and possible rate-determining steps (RDS) for Cr dissolution using linear sweep voltammetry. Results showed that at low potentials, Cr dissolution is charge-transfer controlled and resulted in a faceted surface morphology; at high potentials, Cr dissolution is mass-transport controlled due to the formation of alkali metal-Cr-salt films leaving behind a smooth plan view-morphology. These phenomena allow for the fate of Cr—whether dissolved as ions or retained in the metal—to be tracked and used to quantify corrosion rates. This insight inspires the development of an in-situ electrochemical method to calculate Cr mass loss in real time by measuring the concentrations of Cr(II) and Cr(III) species using CV with a secondary Pt working electrode.

This framework was subsequently utilized to understand the corrosion and dealloying behavior of model Ni-20Cr alloys (wt.%) in molten LiF-NaF-KF at 600°C. Here, Cr is the less noble element and Ni is the more noble element. The RDS and corrosion morphology were examined over the electrode potentials from 1.75 VK+/K  to 2.75 V K+/K. Between 1.75 VK+/K  and 2.10 V K+/K, Cr was selectively dissolved under an initial charge-transfer controlled behavior at a significantly faster rate than Ni. This process left behind a micron-scale bicontinuous structure characterized by a network of interconnected pores and ligaments rich in Ni driven by Ni surface diffusion, which undergo further surface coarsening until 2.30 V K+/K. Cr bulk solid outward diffusion, consisting of lattice and grain boundary diffusion of Cr, was calculated to play a major role in regulating the rate of Cr dealloying in molten FLiNaK due to the high homologous temperature (Tₕ: 0.53) corresponding to the 600oC test environment. To identify the RDS controlling molten fluorides dealloying, the rates of Cr oxidation (Jelectric), Cr(III) ionic diffusion (JCr(III)), Cr solid bulk outward diffusion (Jbulk) were calculated over the duration of dealloying. It was found that over time Jbulk (consisting of both lattice and grain boundary diffusions) is insufficient to support Jelectric, suggesting that molten fluoride dealloying can be bulk diffusion limited. To validate this hypothesis cold working was performed on Ni-20Cr alloy (10%, 30%, 50%) to ameliorate this effect by providing a high density of dislocation structures, serving as short-circuit diffusion pathways to accelerate Cr bulk diffusion. Results revealed that dealloying of cold-worked Ni20Cr (wt%) was found to be controlled by charge-transfer mechanism instead. 

In summary, this work contributes to the fundamental understanding of metal and alloy corrosion, particularly the Ni-Cr systems, in molten fluorides, establishing a systematic framework that relates species electrode potential—a readily measurable parameter—to the corrosion rate and possible damage morphology of the metal and alloy. This framework is transferable for studying compositionally complex alloys considered as candidates for MSR applications. It can also guide the discovery of new alloy compositions or material structures designed to resist corrosion or catastrophic cracking. Ultimately, the successful development of modular molten salt reactors can be the basis for the democratization of relatively clean, abundant energy in both rich and poor countries

Limited quantity of pizza, coffee, fried chicken, apples, donuts are served.

All interested persons are invited to attend.