Radiochemistry provides a promising aporeach to the study of particle formation. The mode of association of activity with the particles is reflected in the relationship between activity (per particle or per unit mass) and particle size. Furthermore, since the different fission products have different chemical properties, and some have relatively long-lived gaseous precursors, they may become associated with the particles in different ways (fractionation). As a result different nuclides may exhibit, on radiochemical analysis, different relationships with particle size. Other features such as transport end deposition of fallout ine volve meteorological conditions, which are beyond the scope of this project. 2.2 BACKGECUND Radiochemical fractionation has been defined as any variation in the relative fission product nuclide abundance.i/ The phenomenon was discovered at any early stage of atomic weapons testing when it was observed that certain fission product ratios vary with the type and locetion of sample.2,3/ During the course of routine calculations of cascade impector data from Operation GREENHCUSE,'/ the authors discovered that gross fission product decay was related exponentially to particle size according to the equation, NMD = kefN, where NMDsis the number median diameter, nis the slope of the gross decey curve, and k and f are constants. It therefore appeared likely that radiochemical fractionation wes primarily a function of particle size distribution in a sample. Operetion JANGL® afforded an opvortunity to study fractionation in greater detail. Size-graded fallout from the underground shot was investigated radiochemically.5/ The spetific activity of several muclides was found to vary regularly with perticle size, in the range from 50 to 70 p. Different nuclides were found to be distributed differently in the particles, e.g., sr39 on the surface and Zr95 in the body; of the rarticle. gaseous precursors, by Adams et 21.7/ A theory was proposed based on the existence of Similar ideas have been discussed by Cadle6/ and : . A similar investigation was undertaken at Operation IVY (Bouton et al8/) for particles in the range from 70 to 220 p for the study of a greater number of nuclides than at JANGL®. The analysis was seriously hampered by the presence of a large quantity of water in the fallout samples. Certain nuclides were selectively dissolved in significent amounts and contaminated the inactive particles, As a result no reliance cculd be placcd on figures for the ver cent active particles, which are necessary to correct the specific activities. For this reason the rediochemical data hsve thus far defied theoretical inter- pretation. They did sugg>st some strong possibilities, however, It appeared likely at the tire thet 099, like Sr99, was deposited on the surface of particles, possible as a result of the volatility of Mo"3 under the high temperatures associated with thermonuclear experiments, 15 SECRET — RESTRICTED DATA