DEVELOPMENT OF A SPECIFIC ACTIVITY DISTRIBUTION ESTIMATION METHOD FOR LARGE LOW-LEVEL RADIOACTIVE WASTE USING SHAPE MEASUREMENT TECHNIQUE

When large items of low-level radioactive waste are to be transported, it is desirable to pack them into a large container instead of a drum to reduce the cost of cutting them into small pieces and also to protect workers from unnecessary radiation exposure. According to IAEA Safety Standards No. TS-R-1, this type of waste is regarded as a ‘low specific activity (LSA)-II material’, and it is possible to transport it as an industrial package (IP) when there is no extreme nonuniformity in the distribution of its activity. The simple criterion of a uniform distribution is suggested in IAEA Safety Standards TS-G-1.1 as “the differences in specific activity between portions of a factor of less than 10 would cause no concern”. In our previous study, it was clarified that a conventional nondestructive radiation measurement method is inapplicable for judging the transport requirement of a completed large waste container because the uncertainty is too large when the filling rate is above approximately 10%. Thus, the authors have developed a new specific activity distribution estimation method for large low-level radioactive waste. In this method, the specific activity estimation of a segment, which refers to a certain amount of radioactive waste placed in a large container, is repeated until a waste package is completed, by the use of gamma-ray measurement, mass measurement, shape measurement by photogrammetry and Monte Carlo calculation techniques. The specific activities of the portions of the large waste package can be estimated as a summation of the specific activities of the segments. The degree of standard uncertainty in the specific activity estimation of segments was evaluated experimentally by using standard radioactive sources and mock-metal-waste samples and varying the filling rate, the distance between the detector and the waste, and the thickness of the segment. The combined standard uncertainty of the specific activity of portions was also estimated by a Monte Carlo simulation. As a result, the applicable scope of this method was clarified in terms of the parameters of segment size and filling rate.