A Glass Neutron Detector with Machine Learning Capabilities

The handheld neutron detectors have large application in homeland security. Primarily for such application a B₁₀ and Li₆ enriched, scintillating glass neutron detector was designed. The model is compact enough to be used as handheld detector and it is equipped with machine learning capabilities to determine the location of the source as well as discriminating a neutron from a gamma. Lithium Borosilicate glass samples, with up to 70% Li₆ and B₁₀ content, doped with Tb and Eu were engineered to optimize performance of the detector. The scintillation properties and neutron/gamma detection capabilities of the glass samples were tested. The model detector’s performance was simulated in Geant4 and the data was utilized for machine learning that can predict the location of the source with an Artificial Neural Network. The reported detector can achieve over 99% accuracy in neutron/gamma discrimination, and source distance estimates, and better than 4% error in radial and azimuthal angle estimates.