The detection of alpha-emitting radioactive materials using UV radioluminescence methods has been subject to increasing attention in recent years. By detecting the UV photons emitted when alpha particles interact with nitrogen and other molecules in air, this method provides the opportunity to detect alpha-emitting radionuclides at distances beyond the small path length of alpha particles in air.
To date – while detectors have been shown to detect alpha-emitting materials at large distances, in low quantities and in short duration – the capability of such detectors to produce accurate, repeatable, quantitative measurements of alpha-emitting sources has not been comprehensively explored. This lack means that comparison of the technique with others employed by industry has not been possible and little work has considered how devices’ analytical performance might be assessed.
This thesis presents methods for the calibration of alpha-induced radioluminescence detectors against traceable sources of known activity to enable the quantitative estimation of sources of unknown activity.
The design of a novel, portable detector capable of such measurements over a range of distances is presented, along with findings about the ideal design of detectors for quantitative measurement. This detector was shown to perform effectively in a wide variety of operating environments, ranging from light tight boxes to operational lighting in one of Sellafield’s active laboratories.
The detector was used to develop four calibration methods – manual calibration, microprocessor-controlled confinement of a source, conveyor-based scanning and imaging by varying the pan/tilt angle of the detector. Using these methods, it was shown that the detector’s response scales linearly with the activity of americium-241 sources measured.
Through the analysis of performance of two methods, it was shown that calibrations should only be applied to measurements obtained using the same acquisition process. This method of performance analysis provides a framework the for comparison of both methodological and detector performance.