SiC Detector Thickness Optimization for Enhanced Response Variability

Abstract

Neutron spectroscopy is a crucial point in several nuclear applications. Accurately measuring fast neutron energy distributions in high-flux conditions reveals a significant technology gap, hindering the acquisition of precise energy fluence distributions. This project investigates the potential of machine learning to bridge this gap, focusing on neutron energies from 100 keV to 20 MeV and fluence rates from 1010 n/cm2s to 1012 n/cm2s using solid detectors such as Silicon Carbide (SiC) and Chemical Vapor Deposition (CVD) diamonds. This paper details the simulation design phase of our project, emphasizing the exploration of optimal SiC solid detector thickness to introduce crucial variability for machine learning training.