NONDESTRUCTIVE ASSAY USING ACTIVE AND PASSIVE COMPUTED TOMOGRAPHY1

drums temporarily stored at nearly 40 sites within the United States. Contents of these drums must be characterized before they are transported for permanent disposal. Traditional gammaray methods used to characterize nuclear waste introduce errors that are related to non-uniform measurement responses associated with unknown radioactive source and matrix material distributions. These errors can be reduced by application of tomographic techniques, that measure these distributions. The Lawrence Livermore National Laboratory (LLNL) has developed two tomographicbased waste assay systems. They use external radioactive sources and tomography-protocol to map the attenuation within a waste drum as a function of mono-energetic gamma-ray energy in waste containers. Passive tomography is used to localize and identify specific radioactive waste contents within the same waste containers. Reconstruction of the passive data via the active images allows internal waste radioactivities in a drum to be corrected for any overlying heterogeneous materials, thus yielding an absolute assay of the waste radioactivities. Calibration of both systems requires only point source measurements and are independent of matrix materials. The first system is housed at LLNL and was developed to study and validate research concepts. The second system is being developed with BioImaging Research, Inc. (BIR) and is housed within a mobile waste characterization trailer. This system has traveled to three DOE facilities to demonstrate the active and passive computed tomography capability. Both systems have participated in and successfully passed the requirements of formal DOE-sponsored intercomparison studies. The systems have measured approximately 1 to 100 grams of plutonium within a variety of waste matrix materials. Laboratory and field results from these two systems over the past several years show that both systems are capable of a precision of 1 to 4% and an accuracy of better than 30% of the true values of known standards for all drums measured.