The goal of this research was to develop a methodology to collect inventory estimates for the analysis and characterization of used fuel in the United States. To accomplish this, the Spent Fuel Database (SFD) was created. Data was collected for the database from publicly available information on the 103 operating reactors in January 2012. Using this data, plant models were developed using ORIGEN-ARP, a point-depletion tool. The output for each reactor model included current inventory estimates for used fuel taken out of the reactor 0, 1, 3, 5, 10, and 20 years ago. To determine the applicability of the database, a methodology was developed to analyze and compare the SFD with mass values produced using knowledge of past fuel assembly designs for general reactor classes. The methodology was centered around the idea of the "applicability range" (AR) of the database, which was defined as the degree to which a correct estimate can be made quantitatively. Pressurized Water Reactors (PWRs) were shown to have a much higher AR than Boiling Water Reactors (BWRs), and older assembly classes were shown to have a lower AR than newer classes. The fission products in the database were shown to consistently have a high AR. Berkelium and californium had low AR for all of the assembly classes, curium had low AR for BWR classes and mixed AR for PWR classes, and americium and some plutonium isotopes had low AR for BWR classes. An assessment of the inventory estimates considered the potential radiotoxicity and heat load from these masses. The radiotoxicity by ingestion decreased by about a factor of 10 from the newest used fuel to the oldest, and the radiotoxicity by inhalation decreased by a factor of 2. While one person could never eat or inhale a spent fuel assembly, radiotoxicity was used as a metric for the upper limit of possible harm. The heat load decreased by more than a factor of 100 over the same range of fuel assemblies. On a per assembly basis, the radiotoxicity and heat load showed similar trends, with newer PWR assemblies being the highest and BWR assemblies being the lowest in both categories. Considering these results, at a potential interim storage facility, priority should be given to the oldest BWR assemblies to reduce the radiotoxic risk and heating requirements. Also, reprocessing and transmuting is highly encouraged to reduce the radiotoxicity and heat of the waste entering storage. Finally, to continue improving the SFD, future work should seek to quantify the magnitude of the impact of variations in AR for curium and for BWR classes. Moreover, future work should incorporate the used fuel from all the shutdown reactors into the database. Even in its current form, though, the SFD is a useful reference tool.
