In the calibration of the connting equipment to yield disintegrae tion rates, calculations were pep farmed involving the flux of a thermal neutron irradiation of U°32 and the fission yields of per- tinent nuclides from the resultant fission of the U235, If the neutron flux and fission ylelds are each assigned a precision of 10 per cent, the microcurie values have an additional error of about lk per cent * independent of the analytical procedure, All but two of the Zr? 95 analyses performed on the size-~sepzrated moe pepper ties between the carrier and tracer, haracteristic This character olf has also bezn reported by Jacobs and Jordan, 1 The BalhO contamination of the sodium analyses hag never been RONES Pry sanples at early times after shot had to be repeated because of wide variations in results, Better reproducibility was obtained when the analyses were made & mnths later, There appeared to be poor exchange between carrier and tracer in the early analyses because almost identical chemical yields among the aliquots produced significantly different activities, Similar anomalous behavior of tracer zirconium has been observed in these Jaboratories during other investigations, The poor spreads in the [a analyses can also be attributed to poor exchange resolved, Repeated attempts to simulate the test shot samples have hever reproduced this barium contamination in the sodium fraction of the cation elution. No satisfactory explanaticn can be offered for this anomalous behavior of barium which has been observed only in these shot samples, 5.2 EVALUATICN OF SIZE SEPARATION: AXD AUTORADIOGRAPH TECHNIQUES There are certain inherent difficulties in any method of particle Size separation. Standard sieves are quite satisfactory for discrete, spherical particles, although the size fractions obtained show a considerable distribution of sizes on account of variations in the spacing of the screens, If the particles are irregular in shape, they may pass through the screens in varicus orientations so as to produce . very broad distributions. Frangible or agglomerated particles may be broken up by the agitation necessary to carry out the separation, These factors are greatly influenced by the length of time the sieves ne omy wh iv SECRET — RESTRICTED DATA w 37 Ew ee are shaken, This time was kept to a minimum in order to minirize these effects, Nevertheless the size distributions were quite broad and did not compare well with the sieve sizes, as may be seen in Table 5.2.