Issues related to the potential for a large-scale sodium vapor explosion in a carbide-fueled liquid-metal fast breeder reactor accident were addressed in the AX1 test in the Transient Reactor Test Facility. Test design and operating conditions were selected to meet the spontaneous nucleation temperature criterion for an energetic, explosive molten fuel-sodium interaction. Although that criterion appears to have been achieved, thermal and mechanical analyses of the test data indicate that the interaction was not especially energetic. Comparison to similar tests on oxide fuel indicates that, under the particular test conditions employed, the conversion of thermal energy to mechanical work is similar for the two fuel types.Transient bulk fuel motion was extensive, with axial fuel motion generally coinciding in time and space with the ejection of the coolant from the original fuel region.Posttest examination of the hardware revealed that nearly all of the fuel had mixed on a microscopic scale with the stainless steel cladding. Relative proportions of iron, chromium, and nickel in the mixture varied widely. The melting point of the mixture was apparently much lower than that of stainless steel.

In a transient overpower accident, the fuel element failure threshold is a function of the rate of reactivity increase and the fuel microstructure. Test E8 simulated a hypothetical $3/s TOP accident in a liquid-metal fast breeder reactor using seven (Pu,U)O2 fuel elements of the fast test reactor (FTR) type. The test elements were pre-irradiated at 30 kW/m in the Experimental Breeder Reactor II to 5 at.% burnup, leading to a low-to-moderate power micro structure typical of FTR fuel.Data from test vehicle sensors, hodoscope, and post-test examinations were used to deduce the sequence of events occurring within the test zone. The initial fuel failure event occurred abruptly at ∼29 times the nominal power level at an estimated average enthalpy of ∼925 kJ/kg relative to 20°C, with 50% of the fuel cross-sectional area above the solidus at the suspected failure site. After the initial failure, ∼2% of the fuel was ejected above the top of the active fuel region. Sodium voiding occurred rapidly. An upper blockage was formed that apparently prevented further fuel dispersal. Inherent test vehicle limitations, loss of flow tube geometry, and nontypical power generation after fuel element failure may have caused a departure from the fuel motion predicted for the FTR conditions. No violent fuel-coolant interactions were observed in the test.

Test data from an in-pile failure experiment of high-power LMFBR-type fuel pins in a simulated $3/s transient-overpower (TOP) accident are reported and analyzed. Major conclusions are that (1) a series of cladding ruptures during the 100-ms period preceding fuel release injected small bursts of fission gas into the flow stream; (2) gas release influenced subsequent cladding melting and fuel release (there were no measurable FCI's (fuel-coolant interactions), and all fuel motion observed by the hodoscope was very slow); (3) the predominant postfailure fuel motion appears to be radial swelling that left a spongy fuel crust on the holder wall; (4) less than 4 to 6 percent of the fuel moved axially out of the original fuel zone, and most of this froze within a 10-cm region above the original top of the fuel zone to form the outlet blockage. An inlet blockage approximately 1 cm long was formed and consisted of large interconnected void regions. Both blockages began just beyond the ends of the fuel pellets.

Results of two failure experiments using LMFBR-type fuel during simulated unprotected transient overpower accidents are reported and analyzed. In both experiments, a single fresh fuel pin in a Mark0IIA loop was subjected to a temperature-limited, step-reactivity irradiation in the TREAT reactor. Total energy was 490 MJ in Test H2 and in Test E4.

Two stainless-steel-clad mixed-oxide fuel elements have been examined after irradiation in EBR-II to a burnup of 11 at.%. The comparison of fuel of low oxygen-to-metal (O/M) ratio (1.94 to 1.97) with fuel of high O/M ratio (1.99 to 2.00) indicated significant differences in fuel and fission-product behavior. The lower oxidizing potential of the low O/M ratio fuel generated no attack of the Type 316L cladding. This appeared to be related to the lack of oxidation and migration of fission-product molybdenum to the cladding surface. Carbide precipitation in the cladding also appeared to be a factor in the type of attack that occurred. Cesium, however, was mobile in both fuels. The separation of the fuel and cladding at high burnup is thought to be related to the deposition of Cs-Mo-O at the interface. In the low O/M ratio fuel, in which the molybdenum did not migrate to the interface, the fuel and cladding remained in contact. The diametral change in the low O/M ratio element, however, was less than in the high O/M ratio element, 3.55% compared with 4.60%. This lower deformation is attributed to greater internal swelling accommodation by the more plastic low O/M ratio fuel. The axial migration of cesium in the low O/M ratio element resulted in an apparently nondetrimental reaction with the UO2 blanket and insulator pellets.

Segments of Type-304L stainless-steel cladding from irradiated EBR-II fuel elements have been used for burst tests from room temperature to 1000°C. The cladding had accumulated exposures of (0.5 to 1.4) × 1022 n/cm2 (> 0.1 MeV) at temperatures between 370 and 500°C. In burst tests at ≤ 700°C, the greater strength was on the lower half of the irradiated cladding where the irradiation temperature was < 475°C. Tests at 800, 900, and 1000°C each showed uniform strength along the lengths of the cladding. The strength of the irradiated specimens decreased most rapidly with temperatures between 400 and 700°C. Unirradiated specimens were weaker than irradiated ones below 700°C; above 700°C, the unirradiated specimens were slightly stronger. The irradiated and the unirradiated specimens had low uniform strain (∼1 and ∼10%, respectively) at temperature of ∼ 400 to 500°C.