This research project was conducted at McMaster University under the supervision of Dr. Troy Farncombe.
Abstract
The precise localization and quantification of radioactivity in an emergency-response scenario is critical for public safety and informed remediation efforts. The proposed drone-based radioactive contamination localization system is designed with the intention of addressing the shortcomings of conventional radiation surveying techniques. A unique 3x3 detector geometry implementing high-performing scintillator crystals and solid-state silicon-based detectors is proposed which allows for reconstruction of the spatial distribution of radioactivity. A model of the system is simulated using GATE Monte-Carlo methods in order to assess spatial response to radioactivity in a representative real-world environment. The system response is compared for simulated photon energies of 140keV and 662keV, with the full-width at half maximum (FWHM) of response profiles not seen to vary significantly with energy. However, the 662keV photons are seen to result in a greater number of net counts registered, potentially attributed to more extensive penetration of the lead collimators by photons originating outside of the field of view, compared to lower energy photons which may be attenuated. Similarly, the 662keV photons incident on the interior of the collimators may undergo further scattering relative to the 140keV simulation, resulting in normally adjacent photons becoming incident on the detector. This result indicates potential degradation of spatial resolution as a function of photon scatter and penetration of the lead collimators. An alternative analytical model based only on system geometry is compared to the Monte-Carlo results with the intention of allowing for efficient optimization of design parameters. Based on gaussian fits of response profiles, the Monte-Carlo data consistently represents a wider FWHM when compared to the analytical model, potentially attributed to statistical photon interactions permitted in the Monte-Carlo simulation. The results of the Monte-Carlo simulation aid in refinement and validation of the analytical model as representative of an idealized system not accounting for statistical processes involved in photon interactions. Ultimately, the simulation data is directly applicable in the design of a physical prototype, while the acquired spatial response data is necessary for the intended reconstruction of radioactivity.