Demonstration of a Tritiated Nitroxide Nuclear Battery.

ABSTRACT

Unattended, compact, terrestrial and space sensors require sources that have high energy and power densities to continuously operate for 3 to 99 years depending on application. Currently, chemical sources cannot fully satisfy these applications, especially in solid state form. Betavoltaic (ßV) nuclear batteries using ß--emitting radioisotopes possess energy densities 1000 times greater than conventional chemical sources. Their power density is a function of ß- flux saturation point relative to the planar (2D) configuration, ß- emission range, and the semiconductor converter, the betavoltaic (ßV) cell, properties. The figure of merit is the beta (ß-)-flux surface power density ( [Formula see text] in µWn per cm2 footprint), where an optimal portion of incident beta particles penetrates the surrounding semiconductor depletion region. Tritiated nitroxides are favorable radioisotope sources with the potential to have the highest specific activity (Am in Ci/g) and [Formula see text] for an organic compound in solid form. The goal of this research is to demonstrate a tritiated nitroxide nuclear battery using the planar (2D) coupling configuration. The reproducible tritiation procedure produced stable product with a Am of approximately 635 Ci/g, which was 70% of the theoretical Am. For the nuclear battery demonstration, the tritiated nitroxide, dissolved in methanol, was deposited on a 4H-SiC ßV and InGaP photovoltaic (PV) cell using a dispensing apparatus and micropipette. Both devices' characteristics were measured beforehand using a controlled electron beam source to approximate the surface radioactivity from the deposited radioisotope. The maximum power point (MPP) of the 4H-SiC and InGaP were 7.77 nW/cm2 and 1.63 nW/cm2 with 100 mCi and 67 mCi, respectively. The power and total efficiency were lower than expected due to partial solvent evaporation and droplet thickness. Numerical models using MCNP6 Monte Carlo code were used to simulate an optimal nuclear battery prototype. The models' accuracy was confirmed with the device calibration curves and a previous metal tritide model based on empirical results. Based on optimal model results, the tritiated nitroxide saturation layer thickness (D0.99) and [Formula see text] (D0.99) were 10 µm and 558 nW/cm2, respectively, using a 4H-SiC.