DERTERMINATION OF THE PLUTONIUM AND AMERICIUM CONTENT OF THE MOLTEN SALT EXTRACTION SCRAPS. USE OF MCNP 4C2 TO VALIDATE THE METHOD.

Americium was extracted from molten Plutonium metal into a molten salt during an electrolysis process. The scraps (spent salt, cathode, and crucible) contain extracted americium and a part of plutonium. The plutonium content must be determined with a high accuracy to respect the demand of the nuclear material management. We use the gamma spectrometry to evaluate the plutonium and the americium contents of the molten salt extraction scraps. The method, which is called the “infinite energy extrapolation method”, is based on the analysis of the gamma rays emitted by the plutonium 239 isotope between 100 and 500 keV. For several full energy peaks, apparent masses are determined from the net areas and a graph of these apparent masses versus the inverse of the energy is plotted. An extrapolation to the infinite energy (Y-axis crossing) gives the right mass because the photon self-absorption in the nuclear material is negligible. The measurement precision is influenced by the experimental set-up (the device geometry, instrumentation, screens between the sample and the detector), the counting statistics and the matrix attenuation. The self-absorption is specifically important in the spent salt. The purpose of this study is to validate the method used to determine the plutonium content in the spent salt and possibly to detect a bias. This paper presents the methodology used to estimate the effect of the different parameters, on the result of the measurement. The numerical simulations are performed with the Monte-Carlo transport code MCNP, version 4C2. The calculations are divided into two stages to speed up the statistical convergence. The first one consists in calculating the photon flux at a point located in front of the germanium detector entrance window (point detector or “F5 tally” in MCNP). The second stage consists in simulating the detector response to this incoming photon flux (energy distribution of the pulses or “F8” tally). Numerical parametrical studies are carried out to identify the key parameters, among which is the salt matrix attenuation. We suggest a correction to take this effect into account and to increase the accuracy of the results. Then we estimate the total uncertainty of the method, which is roughly ±40% on the plutonium mass.