Speaker
Description
IRE (Institute of radioelement in Fleurus - Belgium) produces radioisotopes for medical use. To produce them, irradiated uranium targets are dissolved through chemical processes.
These processes generate various kind of high activity waste. These wastes are conditioned in small drums (10L) placed in interim storage in order to decrease the dose rate. When the dose rate is low enough, the waste can be conditioned into 220L drums for final characterization before sending them to the Belgian national authority for waste (ONDRAF-NIRAS). Up to date the strategy is based on dose rate measurement at the arrival of each waste and a second measurement is foreseen after few months. The isotopic vector is quite variable in each waste and so the dose rate could decrease rapidly or not. The dose rate can vary from few millisievert to few sievert per hour.
The purpose of this work is to improve this process with studying the possibility to characterize the isotopic content of each waste drum through gamma spectroscopy using newly developed detection technologies in order to optimize the decreasing time and so the waste stream.
Several constraints have to be considered. Indeed, the measurement location is already defined and cannot be moved. The drum to characterize is placed in a small shielded bunker. A small removable aperture is available to look inside this bunker. The measurement system has to be placed in a restricted environment (in terms of available space) with a high radiological environment (drum waste storage room with an ambient dose rate between 5 and 30 µSv/h). The measurement time of each drum is fixed to 600s. The number of isotopes to identify and quantify is limited to 13 elements. The dose rate of the drum can vary from few mSv/h to hundred Sv/h in extreme cases.
A screening of available detectors technology on the markets has be done (classical HPGe, MicroGe™, CZT and NaI detectors). A series of preliminary tests are carried out to determine which detector is the best candidate to reach the initial aim. The identification and quantification capabilities are tested. Resolution of peak in function of dead time are also analyzed. Some typical drums are selected and measured with the different systems. The results of the preliminary tests conduct to further investigate the use of a MicroGe system.
The newly developed MicroGe detector is commercialized by Mirion. This system is relatively compact and suitable for high gamma-ray flux environments and adapted to high count rate. Two different size of crystal are available (10 x 10 mm crystal and 20 x 20 mm for increased efficiency). Tests conducted during this study are done with a crystal of 10 x 10 mm with an energy resolution around 1.7 keV at 661 keV. This detector is very compact and have a cooling down time under half an hour. The relative efficiency is around 0.04% at 1332.5 keV. Efficiency curve can be obtained by simulation. A MicroGe version optimized for higher fluxes has a saturation for 137Cs source at a dose rate up of 8 mSv/h, to be compared with 1 mSv/h for the standard version; while maintaining very good spectroscopic performance.
One of the main topics of this study is to establish the theorical measurable activity knowing the geometrical constraints of the measurement and the saturation dose rate of the detector. For this, some Monte Carlo calculations are done considering some hypothesis. Calculating the dose rate at contact of the detector for some isotopic vector and fixed activity in the drum then the dose rate around the drum is calculated. Comparisons are done with input data concerning dose rates. These results fixed a range for operation.
Additional studies are required, however, as some waste cannot be measured, given their excessive dose rate. The high variability of the dose rate is also a challenge that needs to be addressed, for instance by implementing an automated adaptive collimation/shielding process based on a dose rate or dead time measurement of the MicroGe detector.
The MCNP model also needs to be validated by experimental measurements of a reference drum.
Finally, before the implementation of a fully automated characterization system using MicroGe, tests of the entire process under real conditions must be carried out.
