Development of Performance Criteria for High Burnup Cladding Materials, 20-R9565Printer Friendly Version
Inclusive Dates: 10/03/05 Current
Background - Cladding failure of fuel rods caused by hydride-induced embrittlement is a reliability concern for spent nuclear fuel after extended burnup. High burnup increases the thickness of the oxide layer on zirconium alloy cladding, the amount of absorbed hydrogen in cladding, and the internal fuel rod pressure. The mechanical integrity of zirconium alloy cladding degrades as the burnup increases because of a higher susceptibility to premature fracture resulting from hydride-induced embrittlement and wall thinning by oxidation. Although a significant quantity of the future spent nuclear fuel from nuclear power plants will reach high burnup (greater than 45 GWd/MTU), performance criteria for predicting radial hydride fracture of high burnup cladding have not been established.
Approach - The objective of this project is to develop a probabilistic micromechanical methodology for establishing performance criteria and assessing the probability of hydride-induced failure in zirconium alloy cladding. The methodology that is under development includes micromechanical models for describing pertinent failure mechanisms and probabilistic modules for treating uncertainties in random variables, such as fuel burnup, cladding temperature, internal gas pressure, oxide layer thickness, hydrogen content, and hydride morphology. Various computer codes are being integrated into a computational methodology for predicting the probability of cladding failure caused by hydride-induced failure. Case studies will be performed for various scenarios under transportation and dry storage conditions to assess the risk of cladding failure.
Accomplishments - A set of micromechanical models for treating oxide fracture, blister fracture, delayed hydride cracking, and cladding fracture have been developed and incorporated into a computer code called ZRCRACK. Each of the individual failure modes is coded as a subroutine, and the organization of the code is shown schematically in Figure 1. The code has been used to compute the time to failure, by delayed hydride cracking in Zr-2.5%Nb at 25 and 200 °C [77 and 392 °F], shown in Figure 2. The computed time to failure is significantly reduced when the cladding temperature is increased. ZRCRACK has been integrated into probabilistic structural analysis code NESSUS® as a damage model subroutine for treating time- and temperature-dependent cladding failure. The integration includes definition of regression models for various cladding properties and insertion of the damage model. Initial probabilistic sensitivity analyses for the set of random variables for the oxide cracking failure mode indicate that burnup is the most important variable to the probability of this failure mode.