Determination of detection efficiency on HPGe detector for point-like and volumetric samples based on Geant4 simulations

Abstract

Production of medical isotopes has recently been developed at the Institute of Modern Physics (IMP) in China and many aqueous samples containing produced radioisotopes at wide range of volume were generated. Evaluation of radionuclidic purity and quantification of radioactivity for these samples are of significant importance, especially for medical purpose. High purity germanium (HPGe) detectors and gamma spectrometry are widely used in radionuclidic purity evaluation due to the high energy resolution. However, radionuclidic purity assessment and determination of radioactivity cannot be accurately conducted by using the HPGe detector installed at Laboratory of Nuclear Chemistry in IMP due to the lack of competent calibration standard sources. Therefore, an efficient method based on Geant4 simulation was proposed to calculate the detection efficiency curve for samples with different geometry, composition and source-to-detector distance. A set of efficiency data measured with a point-like standard source containing multi-radionuclides was used to determine the dimension of germanium crystal and thickness of dead layers. Afterwards, a point-like and 3 volumetric standard sources (5 ml, 10 ml and 20 ml, respectively) containing 152Eu were measured at 5–50 cm source-to-detector distances. The EFFTRAN code was applied to calculate the correction factor of coincidence summing for all measurements. The full energy peak (FEP) efficiencies with energy range of 121–1408 keV were calculated and compared with the values obtained from simulations. The simulated FEP efficiency values were in good agreement with experimental results, which illustrated a satisfactory precision of the Geant4 simulations. The feasibility of obtaining detection efficiency for point-like and volumetric samples by means of Geant4 simulations was confirmed. This method can be applied to address the efficiency problem caused by sample geometry and composition.