A Study on Cladding Performance Metrics for Accident Tolerant Fuels During Large Break Loss of Coolant Accident

Abstract

During Loss of Coolant Accidents (LOCAs) in nuclear power plants, rapid deterioration of fuel rods occurs due to steam oxidation of Zr-based cladding, which releases large quantities of heat and hydrogen. If the reaction is unmitigated, it can lead to core meltdown and containment explosion, as witnessed at Fukushima Daiichi nuclear power plant in 2011. To improve plant safety, Accident Tolerant Fuels (ATFs) are being developed, which feature enhanced material properties, reaction kinetics, and fission product retention. Previous research shows that prospective ATF claddings such as FeCrAl and SiC can offer enhanced thermal properties, potentially extending coping time beyond conventional Zr-based cladding and increasing the chances of successful recovery. However, these studies lack detailed investigation into which properties contribute to these outcomes and their extent. Additionally, optimal property combinations that maximize coping time remain unexplored. Therefore, this work conducts a sensitivity analysis to identify and quantify the influence of cladding thermophysical and thermochemical properties on oxidation heat released, hydrogen gas generated, and ultimately coping time during an unmitigated large-break LOCA. Hypothetical cladding materials were modeled within a MAAP-UQLab framework and sensitivity indices, including correlation and regression coefficients, were calculated to examine the relationship between the properties and the performance metrics. Results revealed that the heat capacity notably influenced coping time, while the melting temperature significantly affected oxidation heat and hydrogen production. Simultaneous enhancement of both properties improved all performance metrics. Conversely, the pre-exponential constant and heat of reaction presented minor effects, while thermal conductivity showed no impact on any of the metrics. Furthermore, a machine learning regression model was trained to estimate the performance metrics during LOCA for any given set of thermophysical and thermochemical properties defined within the imposed boundaries. Overall, this analysis aims to support the establishment of selection criteria to systematically narrow down promising candidates for ATF.

Date of Award	11 Jul 2024
Original language	American English
Supervisor	Saeed Alameri (Supervisor)