In order to establish disposal plans for sludge, which is one of the untreated waste materials from domestic nuclear power plants, it is necessary to determine the radioactivity concentration of radioactive isotopes. In this study, we aim to evaluate the gross alpha radioactivity of sludge containing radioactive contaminants after pre-treatment, in order to assess the level of sludge waste and obtain analytical data for discussing disposal methods. Samples of sludge generated from nuclear power plants were pre-treated, solutionized, and prepared as analysis samples for evaluating the gross alpha radioactivity.

Radioactive contamination distribution in nuclear facilities is typically measured and analyzed using radiation sensors. Since generally used detection sensors have relatively high efficiency, it is difficult to apply them to a high radiation field. Therefore, shielding/collimators and small size detectors are typically used. Nevertheless, problems of pulse accumulation and dead time still remain. This can cause measurement errors and distort the energy spectrum. In this study, this problem was confirmed through experiments, and signal pile-up and dead time correction studies were performed. A detection system combining a GAGG sensor and SiPM with a size of 10 mm × 10 mm × 10 mm was used, and GAGG radiation characteristics were evaluated for each radiation dose (0.001~57 mSv/h). As a result, efficiency increased as the dose increased, but the energy spectrum tended to shift to the left. At a radiation dose intensity of 400 Ci (14.8 TBq), a collimator was additionally installed, but efficiency decreased and the spectrum was distorted. It was analyzed that signal loss occurred when more than 1 million particles were incident on the detector. In this high-radioactivity area, quantitative analysis is likely to be difficult due to spectral distortion, and this needs to be supplemented through a correction algorithm. In recent research cases, the development of correction algorithms using MCNP and AI is being actively carried out around the world, and more than 98% of the signals have been corrected and the spectrum has been restored. Nevertheless, the artificial intelligence (AI) results were based on only 2-3 overlapping pulse data and did not consider the effect of noise, so they did not solve realistic problems. Additional research is needed. In the future, we plan to conduct signal correction research using ≈10×10 mm small size detectors (GAGG, CZT etc.). Also, the performance evaluation of the measurement/analysis system is intended to be performed in an environment similar to the high radiation field of an actual nuclear facility.

For the transport of spent nuclear fuel, it is necessary to evaluate the amount of radioactivity for each assembly and the total amount of radioactivity for each cask. Currently, KHNP is evaluating the radioactivity using the Express mode of the OrigenArp program in the SCALE6.1 code. Express mode is a method to evaluate the radioactivity assuming that it has been burned with the same power per cycle, and Detail mode is a method to evaluate the actual combustion history such as power and cooling time for each cycle. For a total of 3,795 assemblies, including 1,391 assembliess for Kori Unit 1, 1,427 assemblies for Hanbit Unit 2, and 977 assemblies for Hanul Unit 3, the radioactivity was evaluated in Express mode and Detail mode, respectively, and the results were compared. As a result of the evaluation, it was confirmed that the results of the Express mode were evaluated more conservatively by 2.5~12.9% than that of the Detail mode. Accordingly, KHNP established a plan to change the evaluation method from Express mode to Detail mode in order to improve the accuracy of the radioactivity assessment results and eliminate conservatism.