Monte Carlo modelling of a N-type coaxial high purity germanium detector

Abstract

High-purity germanium detectors are widely used in the field of nuclear safeguards because of their high efficiency and very good energy resolution. In particular, they are used for determining the 235U enrichment with non-destructive assay (NDA) methods. In this case, calibration with reference standards is required when applying the enrichment meter principle. Since the objects to be measured in the field can be very different and since the enrichment calibration constants depend on many parameters such as collimators, container sizes and chemical composition, MCNP modelling may represent an alternative to experimental determination of the calibration constants when uranium reference standards are not available.The MCNP transport code represents one of the most used programs to simulate gamma detector response. By way of examples, a few studies on this topic are listed in [1] to [14], but all except [12] concern P-type detectors. In most of these studies, substantial discrepancies between calculated and experimental data are observed. The efficiency calculated with the Monte Carlo method is typically 10–50% higher than what is found experimentally.This deficiency in the observed detector efficiency is commonly attributed to an underestimation of the thickness of the dead layer on the front area of the detector which is caused by the N+ contact of P-type detectors. Hence, this thickness is often adjusted in the model to match Monte Carlo calculated efficiencies with experimental efficiencies. However, in the case of N-type detectors, this fitting possibility is not availabe since the dead layer at the outer P+ contact is extremely thin (0.3μm). An alternative explanation for the efficiency discrepancy, especially in the case of N-type detectors, are uncertainties of the physical dimensions of the germanium crystal based on supplied specifications.The purpose of this work is to study the response of a N-type HP Ge detector and to deduce the cause of the efficiency deficit found.