The Kinetics of the Zirconium-Uranium Dioxide Reaction

Abstract

The kinetics of the zirconium'UOn reaction was studied using crystal-bar zirconium and fused UOn or UOn high'fired in hydrogen. The reaction was studied with (1) solid zirconium cylinders and UOy powder, (2) compacted mixtures of zirconium and UOn powders, and (3) sandwich' type elements. Experiments performed with sandwich-type elements yielded the most useful data both from a practical and a fundamental point of view. Exploratory experiments designed to give a wider range of experimental materials were unsuccessful in producing substoichiometric UO^. Results obtained from sandwich-type experiments showed that slow reactions at PSO and 1100 F formed reaction zones less than 0.0002 in, thick in 65 days, A measurable reaction took place at 1300 F, while at 2000 F, the reaction was very rapid. The UOn seems to be reduced to its elements, uranium and oxygen, which diffuse simultaneously, but not with equal speed, into the zirconium. Reactants heated at 1300 c and slowly cooled to room temperature, were found to have formed alpha zirconium and alpha uranium adjacent to the UOn, Deeper in the original zirconium region the epsilon uraniumzirconium intermetallic phase was found dispersed in a zirconium-oxygen solid solution. Elements slowly cooled from 2000 F contained alpha zirconium and alpha uranium at the UOn-metal interface. A continuous stained band having the alpha uranium lattice structure and containing some zirconium was observed beneath the two-phase region. Oxygen had diffused beyond the uranium-containing region, forming the hard zirconium-oxygen solid solution. On the basis of limited data obtained, it has been estimated that penetration of oxygen through the 0.030-in.-thick zirconium jacket walls probably would cause their destructive embrittlement in about 300 days at 1300'*F or 2-1/2 days at 2000^. The reaction between Zircaloy 2 and UOn at 1300 F appears to be somewhat less rapid than that between unalloyed zirconium and UOn. The phases formed by this reaction have not been identified.

On the origins and the evolution of the fuel-cladding bonding phenomenon in PWR fuel rods

“…9 presents an optical micrograph of a Zy-4jUO 2.00 interface, for a model system annealed for 250 h at 400 C, under a purified Ar atmosphere and a compressive load of about 10 MPa. Here, no sign of Zr oxidation can be seen, in agreement with Mallett et al [18] results obtained on a Zr-UO 2.00 stacked diffusion couple annealed for 13 872 h at 399 C under an isostatic pressure of 345 bar of He. Fig.…”

“…The five dark areas observed at the Zy-4jUO 2.10 interface (Fig. 10), similar to ZrO 2 islets in irradiated fuels, also appears to be similar to those observed as a continuous layer by Mallett et al [18] on a Zr/hyper-stoichiometric UO 2.10 diffusion couple, composed of a zirconium cylinder buried in a uranium dioxide powder and annealed for 18 h at 825 C in Ar. It is important to mention that in our case, like in theirs, the analogous Zr/stoichiometric UO 2.00 sample did not exhibit any precipitate at the ZrjUO 2 interface.…”

“…Our results are fully in agreement with this hypothesis, considering that the formation of ZrO 2 on a Zr substrate only occurs in presence of hyper-stoichiometric UO 2þx . As oxidation kinetics are known to be thermally activated, the annealing temperature in the case of Mallett et al [18], twice higher than ours would explain the formation of a continuous ZrO 2 layer in their case and only ZrO 2 islets in our case.…”

“…Hyper-stoichiometric uranium dioxide (UO 2.10 in our case) was used to verify the effect of excess of oxygen in UO 2 upon the solidstate oxidation of zirconium. Indeed, based on Mallett et al results [18], the excess of oxygen in UO 2 was highly suspected to be responsible for oxidation of the cladding inner side in PWR fuel rods.…”

“…It is important to mention that in our case, like in theirs, the analogous Zr/stoichiometric UO 2.00 sample did not exhibit any precipitate at the ZrjUO 2 interface. Based on these observations, Mallett et al [18] hypothesized this continuous layer to be a zirconium oxide layer whose formation would be linked to the excess of oxygen in the hyper-stoichiometric UO 2.10 . Our results are fully in agreement with this hypothesis, considering that the formation of ZrO 2 on a Zr substrate only occurs in presence of hyper-stoichiometric UO 2þx .…”