Design Principles of Narrow and Wide Bandgap-Based Betavoltaic Batteries

Abstract

A Monte Carlo electron transport code PENELOPE was used to analyze beta particle energy deposition in semiconductors for titanium tritide and beryllium tritide. The source thickness was incorporated into the model in order to take into account the self-absorption of beta particles in the source material. Furthermore, an isotropic source was modeled with the full beta energy spectrum of tritium to make the beta particle transport method more realistic. The simulated results for a 0.4- $\mu \text{m}$ -thick titanium tritide source with silicon carbide agreed well with the experimental results. The simulated results obtained for an optimized 2.5- $\mu \text{m}$ -thick beryllium tritide source with silicon carbide was about two times higher than the power for a 0.4- $\mu \text{m}$ -thick titanium tritide source with silicon carbide. An approximately two times higher short-circuit current density was obtained for beryllium tritide with silicon compared to silicon carbide. However, the power output density was about $10\times $ higher for silicon carbide. The width of the depletion region for a p-type dopant concentration of $\textsf {1} \times \textsf {10}^{\textsf {19}}/\textsf {cm}^{\textsf {3}}$ and an n-type dopant concentration of $\textsf {5} \times \textsf {10}^{\textsf {14}}/\textsf {cm}^{\textsf {3}}$ is about 1.4 and $2.5~\mu \text{m}$ in silicon and silicon carbide, respectively. The estimated beta particle penetration depth is about 1.48 ± 0.015 and $\textsf {1.03} \pm \textsf {0.015}~\mu \text{m}$ in silicon and silicon carbide, respectively. This suggests that the proper choice of dopant concentrations will achieve greater energy deposition in the depletion region, which in turn will increase the power output.