Influence of Metallurgical Condition on the In-Reactor Dimensional Changes of Zircaloy Fuel Rods

Abstract

Microstructure strongly affects the in-reactor creep and growth behavior of Zircaloy. The observed influence of microstructure on the in-reactor creep behavior can be opposite to that found ex-reactor. Fully annealed Zircaloy exhibits less in-reactor growth strains than stress-relief-annealed Zircaloy. Two approaches to estimate the effect of microstructural variation on the in-reactor creep and growth behavior are discussed. These evaluations are based on diametral and axial strain data, for exposures of up to 8 × 1021 neutrons/cm2 (E > 0.82 MeV), obtained from tests conducted in the Obrigheim reactor (Germany) and in Calvert Cliffs I (United States). In the former tests the Zircaloy-4 cladding was manufactured by several vendors using different combinations of degree of cold work and final heat treatment while in the latter tests cladding of one type was used. All irradiated cladding had similar crystallographic textures. The metallurgical condition of Zircaloy tubing is controlled in part by the degree of cold work prior to the final annealing treatment and by the annealing conditions. In one approach, the net effect of annealing time and temperature is empirically defined in terms of a “normalized annealing time.” A nonlinear relationship was obtained between the “normalized annealing time” and the in-reactor diametral and axial strains. The other approach involves using the high-temperature yield strength as an effective reflection of the microstructure. The high-temperature yield strength of Zircaloy is routinely included in the Zircaloy tubing manufacturer's certifications and can be easily measured. There is a semilogarithmic relationship between the in-reactor diametral strain and the hot (673 K) yield strength of Zircaloy. The model is also useful in estimating the diametral strain variation within a lot if the within-lot yield strength variability is known. A similar correlation was obtained between the axial strain and the unirradiated yield strength. Although based on data from fuel rods of one design, the correlations are expected to be qualitatively applicable to fuel rods of different designs. Relationships between these empirical correlations and creep and growth mechanisms are discussed.