SUMMARY (max. 125 words)

The paper describes a study performed with the large clearance monitor (LCM) simulation software SIStec presenting a novel approach to take the contribution of natural occurring radioactive material (NORM) nuclides into account when analysing measurements by a large clearance monitor.

The approach utilizes two essential features of the SIStec software: the calculation of the efficiencies of the LCM detectors, and the activity localisation via an imaging algorithm.

The combination of these features allows an accurate estimate of the NORM contribution and consequently the residual activity due to contamination or activation of the material. By performing an analysis in accordance to regulatory norms the result can be used to calculate clearance sum values.

KEYWORDS

SIStec, clearance, activity, NORM

INTRODUCTION

The dismantling of nuclear power plants entails producing a large amount of material that has to be disposed of. Generally, the material is either sent to final repositories as nuclear waste, or released (under restricted or unrestricted conditions) as conventional waste.

In order to efficiently manage the large amount of waste packages produced during the decommissioning of a nuclear power plant, the release procedure must be sufficiently short and thus the measurement of the radioactivity of each single piece of waste needs to be optimized. To this end, large clearance monitors (LCM) are used to handle that challenging task. However, one major drawback to the release procedure is the long time needed for the sample preparation before one can measure its radioactivity and release it (e.g. the surface of metal waste must be flattened and cleaned before it can be scanned with a hand-held beta-sensitive monitor).

The key to make the clearance process more efficient is to reduce the complexity of each step of the process by removing conservative assumptions and as a consequence to have a more realistic analysis.

A crucial step in the preparation of the clearance measurement is the creation of sample calibrations,

i.e. estimating the calibration factors to calculate the amount of activity on the sample from the measured count rates. SIStec Basic provides an easy-to-use interface to perform complex Monte Carlo calculations for estimating the calibration factor for each detector of the LCM. The calculations are based on a model of the LCM, of the sample carrier and of the sample which can be created with ease in SIStec. By specifying a nuclide vector, the exact energies of the emitted gamma radiation from a nuclear decay are used for the calculation and a nuclide vector specific calibration is estimated. Due to its digital

nature all parameters in the calculation can be easily varied to perform a parameter study to estimate the effect of a deviating parameter (e.g. mass, density, nuclide vector composition).

The key to make the clearance measurement itself quicker lies on the fact that each LCM has detectors distributed over the full 4π solid angle: applying state of the art statistical imaging in SIStec Pro [1] one can partition the waste into small virtual blocks whose activity can be estimated. Utilizing the Monte Carlo approach for propagating uncertainties, the characteristic limits necessary for the clearance of sample material are calculated in accordance to DIN EN ISO 11929 [2]. The approach used in SIStec Pro removes the overly conservativity which assumes that entire sample activity is located in the point of worst efficiency. This is achieved by simply localizing the activity in the sample and using the real efficiency for the calculation of the characteristic limits. In addition, the imaging of the activity removes the time-consuming work to prove a homogeneous activity distribution by manual measurements with a handheld contamination monitor by simply measuring the real activity distribution over the sample.

In the case that the sample itself contains natural occurring radioactive material (NORM) as in the case of concrete, the power of the combination of the abilities of SIStec Basic, to calculate the detector efficiencies for different gamma radiation energies, and SIStec Pro, to take the NORM nuclides into account when calculating the characteristic limits, is shown.

In this paper we discuss a novel approach to take the contribution of NORM nuclides into account when analysing a measurement with the SIStec Pro software. We start with a description of the approach in section 2 and discuss the obtained results in section 3 before we conclude in the final section.

Removing the contribution of NORM from LCM measurements

The activity in the sample is calculated using an imaging algorithm based on the maximisation of the Informational Conditional Entropy [3] (Conditional Entropy Maximisation or, from now on, CEM). For this the sample is partitioned into small virtual cells and the activity for each of the cells is estimated. The model of evaluation in use cannot be written in a simple form, as it uses an iterative approach to maximize the entropy. Nevertheless, in equation (1) a basic form of the model of evaluation is given. The only simplification is that the calculation of the activities 𝑎^𝑖 from the input parameters is shown as an arrow.