(A fee may be required to access this article.)  Samples of siliceous solids up to 1 g were dissolved completely by successive fusions with potassium fluoride and pyrosulfate in platinum dishes in the presence of appropriate tracers. Samples which contaminate or react deleteriously with platinum were treated extensively with hydrofluoric and perchloric acids before fusion with pyrosulfate in glass. The pyrosulfate cake was dissolved in dilute hydrochloric acid and the actinides were precipitated isomorphously with barium sulfate. The barium sulfate was dissolved in alkaline ethylenediaminetetraacetic acid (EDTA), and the actinides were precipitated as the hydroxides, filtered on a membrane filter and analyzed in an alpha spectrometer. When necessary, the hydroxides were converted to fluorides in the presence of various oxidizing agents to obtain improved resolution and additional separations. Isotopes of all elements from thorium through curium can be determined in one sequential procedure. The procedure was simpler and faster than others available, with higher yields and greater reliability. Resolution of the subsequent alpha spectra was comparable to that obtained by electrodeposition. The results were reproduced with a relative standard deviation of 1 to 2%. The accuracy of the determination on standards was generally as good as the statistical uncertainty of the measurement permits. Some results of actinides in pond sediments, incinerator ash, baghouse dusts, and ion exchange resins are given.

Simultaneous determination of all alpha activity has always been and continues to be a highly desirable option in radiochemical analysis. However, the usual practice of gross alpha counting samples of raw water evaporated to dryness in planchets, air dusts collected on filters, planchets full of soils, etc., and then reporting the results quantitatively is neither accurate nor reliable, since it is virtually impossible to determine the correct counting efficiency with which the count was actually made. Attempts to calibrate empirically for a given set of conditions are also of marginal value because of wide variability in the deposition patterns. Direct gross alpha counting of samples without chemical preparation is useful when the activity present is substantially higher than that from the natural activity in its surroundings, or when a rapid order-of-magnitude indication of alpha activity is desired. All too often the results of these screening methods are reported with precision and accuracy that they cannot possibly have. Even is excess alpha activity is indicated, a gross alpha count cannot distinguish between a man-made transuranium (TRU) radionuclide and a naturally occurring one. The current procedure provides a true total alpha while eliminating the problems just discussed.