Hydride Nucleation-Growth-Dissolution (HNGD) Model: Improvements and Implementation
Open Access
- Author:
- Passelaigue, Florian
- Graduate Program:
- Nuclear Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- October 04, 2021
- Committee Members:
- Jean Paul Allain, Program Head/Chair
Xing Wang, Major Field Member
Arthur Motta, Chair & Dissertation Advisor
Bruce Kammenzind, Special Member
Mia Jin, Major Field Member
Long-Qing Chen, Outside Unit & Field Member - Keywords:
- Zircaloy
hydrides
HNGD
Bison - Abstract:
- In nuclear reactors, waterside corrosion of the Zircaloy nuclear fuel cladding tube causes hydrogen pick-up into the material. This hydrogen can cause zirconium hydrides to precipitate within the cladding. Since these hydrides are usually more brittle than the alloy, they can decrease the ductility of the cladding. Previous efforts made to describe hydrogen behavior and hydride precipitation resulted in the Hydride Nucleation-Growth-Dissolution (HNGD) model. This model can predict the distribution of hydrogen and the partition between solid solution and hydrides in Zircaloy samples that are subjected to a thermal treatment. However, in some cases the HNGD model gives unphysical results. Notably, if the system is close to steady-state, or if the initial hydrogen distribution is significantly heterogeneous the model calculations diverge from experimental results, predicting either no hydride precipitation, or hydride precipitation in a single node of the simulation mesh. The study presented in this dissertation describes how the HNGD model was improved upon to address these shortcomings. This was done using two hypotheses described below. Chapter 1 introduces the issue of hydrogen in Zircaloy cladding, and the experimental data used to validate the HNGD model. In Chapter 2, we review the phenomena described by the HNGD model, the equations used, and associated parameters. We also expose the causes for the unphysical results previously mentioned. Chapter 3 focuses on the two hypotheses. The first hypothesis is based on the assumption that, given enough time, the hydrogen atoms will find the most favorable nucleation spots in the Zircaloy (matrix defects, dislocations, etc), resulting in a decrease of the nucleation barrier. This is translated with a decrease of $TSS_P$ during temperature holds. The second hypothesis postulates that the hydride particles deform the matrix in a way that impacts the hydrogen solubility. These two hypotheses together allow for hydride precipitation to be triggered more easily, and for the hydrides to stay stable. A complete analytical solution (i.e. for the hydrogen in solid solution and in hydrides) was derived for the steady state of the system. Using this tool and the large experimental data set from Kammenzind, the impact of the newly introduced parameters is studied. An extensive validation of the modified HNGD model is performed using Kammenzind's experiments, as well as the benchmark and validation cases used during the initial development of the HNGD model. We show that the modified HNGD is able to predict the thickness of the hydride peak at steady state, which is a significant improvement compared to the initial model. Finally, the implementation of the modified HNGD model into the nuclear fuel performance code Bison is described in Chapter 4. We describe how the quality of the code is ensured when implementing an update in Bison. This modified HNGD model yields physical results when modeling experiments that mimic reactor conditions in terms of hydrogen pick-up, and does not degrade the simulations of experiments that were accurately modeled using the initial HNGD model. This improved HNGD model represents an improvement of the capability to predict the hydrogen behavior in cladding tubes during operation and in spent nuclear fuel during storage and transport.