The ignition of uranium

Abstract

Experimental studies are reported of the ignition of pure uranium. Most of the experiments were performed by placing the metal samples in a flowing oxidizing atmosphere (either air or oxygen) within a furnace whose temperature was increasing at the rate of 10° C/min. The ignition temperature of the sample was determined graphically from the sample temperature-time record as the point of intersection between extensions of the pre-ignition heating rate and the post-ignition self-heating rate. The ignition temperature of 8.5-mm cubes of uranium from three sources (in air and in oxygen) was very close to 600° C; however, characteristic differences in reaction were apparent in the 400–500° C range. Metal in an “as-cast” condition self-heated to a greater extent in the 400–500° C region than did beta-quenched metal. Nitrogen-oxygen mixtures also increased selfheating in this region to a greater extent than did helium-oxygen mixtures. Ignition in air or oxygen was followed by spontaneous thermocycling. Peak burning temperatures in air were limited to about 1500° C, but in pure oxygen temperatures greater than 2150° C were recorded. Ignition temperature decreased with sample size. Studies with a series of cubes, wires and foils, with specific areas from 0.3 to 120 cm 2/g, yielded an irregular decrease of ignition temperature from 640 to 315° C in oxygen. The irregularity occurred with ignition in the temperature region 400 to 500° C and was shown to be associated with the transition of the oxide from an autocatalytic form to a protective form. A simple mathematical model based upon isothermal reaction rate laws and simple heat transfer laws yielded an ignition temperature-specific area relationship which agreed well with the experimental results and which predicted the irregularity. Experiments with aggregate samples, i.e., bundles of wires and stacks of foils, showed that ignition temperature decreased with increasing aggregation as expected from simple heat transfer considerations. It was concluded that uranium ignitions in air and oxygen are thermal ignitions which can be explained in terms of a balance of heat losses and heat generated by the oxidation reaction.