Simulation of betavoltaic batteries with geant4

Abstract

This thesis has a simulation study with the Geant4 toolkit, which uses the Monte Carlo method in its background. The materials used were decided as a result of literature research. The simulation study compares the deposited energy on the battery body using different semiconductors and radioactive sources. Silicon (Si), silicon carbide (SiC), gallium nitride (GaN), gallium arsenide (GaAs), and aluminium gallium arsenide (AlGaAs) are used as semiconductor transducers, and Nickel-63 (Ni-63) and Tritium (H-3) are used as beta sources. The deposited energy value is in descending order in the materials SiC, Si, GaN, AlGaAs and GaAs with %90.78, %88.18, %81.90, %79.45 and %77.15 energy deposition efficiencies when using a tritium source. It is observed that using SiC semiconductor and tritium source has higher energy deposition efficiency. The deposited energy value is in descending order in the materials GaN, SiC, GaAs, Si and AlGaAs with %80.53, %76.30, %74.73, %66.85 and %61.48 energy deposition efficiencies when using a Nickel-63 source. It is observed that using GaN semiconductor and Nickel-63 source has higher energy deposition efficiency. The semiconductor converter designed for this simulation has the shape of a rectangular prism with a thickness of 10 μm and 1x1 mm2 surface area. 10 μm total thickness is divided into 0.2 μm layers for a total of 50 slices to determine the range of peak energy deposition in semiconductors. Beta particles emitted from the Tritium source lose most of their energy in the first 1 μm thickness, while beta particles emitted from the Nickel-63 source consume their energy by releasing their energy to the semiconductor at the first 5 μm thickness.