EFFECTS OF FAST-NEUTRON IRRADIATION AND HIGH TEMPERATURE ON BERYLLIUM OXIDE

Specimens of 0.428 to 1.18 in. dia. BeO were irradiated at 1.4 x 10/sup 20/ to 2.8 x 10/sup 21/ fast neutrons/cm/ (>1 Mev), with irradiation times of 2.03 x 105 to 5.60 x 106 sec, at 110 to 1025 C, and gamma heat generation rates of 18 to 25 w/g. The greatest damage appeared in bodies irradiated for high neutron doses at low temperatures. Greater doses could be tolerated at higher temperatures. Specimens irradiated at 110 deg C powdered at an accumulated dose of 1.5 x 10/sup 21/ 1.1 x 10/sup 21/ neutrons/cm/sup 2/ and at 737 deg C at a dose of 1.6 x 10 /sup 21/ neutrons/cm/sup 2/ remained intact. Specimens irradiated at 827 to 950 deg C at a dose of 2.1 to 2.6 x 10/sup 21/ neutrons/cm/sup 2/ fractured. Changes in linear dimensions up to 4% were observed. X-ray diffraction measurements revealed changes in the c lattice parameter up to 7 x 10-3 DELTA c/c0 for material irradiated at 110 deg C but only 1 to 2 x 10-3 DELTA c/co at 444 to 1025 deg C. The DELTA c/c0 value appeared to saturate at a dose of 3 x 10/sup 20/ neutrons/cm/sup 2/ in the latter temperature range. All temperatures above 444 deg C were equally effective in suppressing the DELTA c/c0 values. Changes in the a parameter were smaller by a factor of 7. He was retained even at irradiation temperatures of 1025 deg C. Gases could be removed only by dissolving the BeO in fused salt. The He content of the BeO varied linearly with fast-neutron dose up to 1.6 x 10/sup 21/ neutrons/ cm/sup 2/, with a slope of 205 standard mu l per g of BeO per 10 neutrons/cm/sup 2/. The tritium content of BeO irradiated at 110 deg C followed a (neutrons/cm) curve up to 6.4 ppm at 2.1 x 10 neutrons/cm/sup 2/. Specimens irradiated above 444 deg C contained much less tritium. Voids 0.1 to 0.3 mu in diameter which were not in the original material were found by electron microscopy in the interior of grains and at grain boundaries. No evdence of chemical decomposition was observed. Radial temperature differences during irradiation increased by 50 to 100%, indicating a decrease in thermal conductivity. It was concluded that the observed changes could be caused either by gas generation or atomic displacements. The hypothesis that thermal stress alone produced failure was rejected. Results indicate that great caution must be exercised in designing BeO moderators for nuclear reactors. Under some conditions physical deterioration occurs to such a degree as to render the material unsuitable for use. For greatest life the moderator should be kept above 400 deg C, and probably 800 to 1000 deg C is more desirable. The design of the irradiation assembly is described. Temperatures were controlled by varying the composition of a helium- argon mixture in an annulus. Sixtythree specimens were irradiated in six assemblies. (auth