|Fig. 1: Schematic of a semiconductor p-n junction-based betavoltaic device.|
Over the last fifty years, the driving force behind the development of betavoltaic devices has been the need for reliable, long-lived, high energy density power sources for operating electrical systems in hostile and inaccessible environments. It is well established that conventional electrochemical batteries, despite their widespread use in electronic devices, have limited longevity and a strong tendency to degrade under extreme environmental conditions.  For situations where battery replacement is inconvenient, such as in remote sensing applications in space or aquatic environments, or potentially life threatening, such as in the case of implantable biomedical prosthetic devices, alternatives to electrochemical battery technologies are desired.  Betavoltaic power sources are one such alternative that can be operated continuously, for years, in harsh environments to generate electricity for low power applications. 
Betavoltaic devices are self contained power sources that convert high energy beta (β) particles emitted from the decay of radioactive isotopes into electrical current. As shown in Fig. 1, a typical betavoltaic device, in its simplest form, consists of a layer of beta-emitting material placed adjacent to a semiconductor p-n junction or Schottky diode. A convenient way to understand the fundamental operation of a betavoltaic device is to consider it as the "nuclear" analog to the familiar solar cell, where, in place of the sun, a beta-emitting isotope provides the source of ionizing radiation. When the semiconductor material is bombarded by high energy beta particles, electron-hole pairs are generated by impact ionization (see Fig. 1). Since the average kinetic energy of typical beta particles used for betavoltaic devices is in the kiloelectron volt (keV) range, a single beta particle can be responsible for generating multiple electron-hole pairs.  According to the Klein formula, the average kinetic energy required to beta-generate an electron-hole pair of energy equal to the semiconductor band gap (Eg) is 2.8 Eg + 0.5 eV.  In addition, during the conversion process, 1.8 Eg eV and 0.5 eV are lost by emission of acoustic and optical phonons, respectively. 
Similar to photovoltaics, electron-hole pairs that are beta-generated inside of or within a minority carrier diffusion length of the depletion region are separated by the built-in electric field and drifted apart (see Fig. 2). The accumulation of separated electron-hole pairs in the quasi-neutral regions of the semiconductor, electrons on the n-side and holes on the p-side, results in the junction becoming forward biased and current flowing through an externally connected load. Despite their operational similarities with photovoltaic devices, betavoltaic devices are usually strictly limited to low power applications.  This is directly related to the fact that the typical flux of beta particles emitted from a beta source is a minute fraction of the photon flux emitted by the sun.  As a result, betavoltaic devices typically generate currents on the order of nano- to micro-amperes, which are several orders of magnitude smaller than currents generated by similarly sized photovoltaic devices. [3,7]
|Fig. 2: Energy-band diagram of the semiconductor p-n junction portion of a betavoltaic device under beta particle radiation.|
When selecting a beta source for a betavoltaic device, fluence rates and isotope half-lifetimes are important aspects that must be considered. Obviously, utilizing long half-lifetime isotopes that can generate sufficient beta particle fluxes is critical to the design of long-lasting betavoltaic power sources. However, the effects of radiation damage in the semiconductor material must also be taken into account. Ideally, the maximum kinetic energy (Emax) of the beta particles emitted from the beta source should be smaller than the radiation damage threshold of the material (Eth).  Otherwise, the emitted beta particles would have sufficient energy to displace atoms in the semiconductor lattice. Radiation induced defects in the semiconductor material can result in shortened minority carrier diffusion lengths, increased leakage currents, and overall device performance degradation.
The importance of selecting an appropriate beta source was quickly identified early on from the rapid degradation of output power observed in early betavoltaic devices that coupled 50 mCi Sr90-Y90 beta sources with silicon p-n junctions.  Despite the 20 year half-lifetime of Sr90-Y90, radiation damage caused by high energy beta particles with maximum kinetic energies up to 2 MeV limited the device lifetime to 14 hours. [4,9] As a result, in order to keep radiation damage at tolerable levels, feasible beta sources for betavoltaics are typically limited to H3 (Emax = 18 keV), Ni63 (Emax = 67 keV), and Pm147 (Emax = 230 keV).  The properties of candidate radioisotopes for betavoltaics are summarized in Table 1. 
For the semiconductor portion of the device, radiation hardness, long minority carrier diffusion lengths, and low leakage currents are important considerations for the design of the semiconductor junction. Wide band gap materials, such as SiC, are often sought out as junction materials because they have higher radiation damage resistance and, as a result of the band gap dependence of electron-hole pair generation, the potential to achieve greater betavoltaic conversion efficiencies when compared with smaller band gap materials.  However, wide band gap materials are typically characterized by lower mobilities and carrier lifetimes, which adversely impacts diffusion lengths.  In addition, the growth of high quality junctions with defect densities sufficiently low enough for use in betavoltaic devices has been a major challenge for some wider band gap materials.  The growth of high quality low defect density junctions is required in order to minimize leakage currents. As a result, the majority of betavoltaic devices have been fabricated using Si (Eth ~ 200 keV), GaAs (Eth ~ 225 keV) or Ge (Eth ~ 350 keV) junctions because of the maturity of their crystal growth technologies. [1,10]
|Table 1: Characteristics of feasible beta sources for betavoltaic devices.|
Over the years, research and development into betavoltaic devices has been sporadic. While the prospect of utilizing radioactive decay products as primary sources of energy for specialized applications has generated interest in betavoltaic devices, high costs, limited availability, and regulatory concerns associated with suitable beta-emitting radioisotopes have been a major challenge for this energy storage technology. The most notable use of betavoltaic devices has been in cardiac pacemakers. During the 1970s, researchers from Donald W. Douglas Laboratories developed a Pm147-Si betavoltaic power source that was implanted in pacemakers in over 100 test patients.  The Pm147-Si battery achieved a conversion efficiency of ~ 4% and had an expected lifetime of ~ 10 years.  However, high costs and concerns over gamma radiation emitted from the contaminate isotope Pm146 contributed to the demise of this power source.  More recently, interest in betavoltaics has been renewed with the demonstration of a 6% power conversion efficiency Ni63- 4H SiC betavoltaic cell and the first regulatory general license for manufacturing and distribution issued to City Labs for their Nano-Tritium betavoltaic device. [3,12] Efforts to improve betavoltaic conversion efficiencies are currently on-going as researchers experiment with three-dimensional device architectures, optimize semiconductor material growth techniques, and investigate utilizing tritium stored in different phases of matter. [3,13] With recent regulatory advances and the prospect that technological improvements could enable betavoltaics to approach their theoretical conversion efficiencies, this decades-old technology may finally be adopted for standard use in certain niche low power applications.
© Sara E. Harrison. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 F. K. Manasse et al., "Schottky-Barrier Betavoltaic Battery," IEEE Trans. Nucl. Sci. 23, 860 (1976).
 R. Bao et al., "Betavoltaic Performance of Radiation-Hardened High-Efficiency Si Space Solar Cells," IEEE Trans. Electron Devices 59, 1286 (2012).
 L. C. Olsen, P. Cabauy and B. J. Elkind, "Betavoltaic Power Sources," Physics Today 65, No. 12, 35 (December 2012).
 L. C. Olsen, "Betavoltaic Energy Conversion," Energy Conversion 13, 117 (1973).
 C. A. Klein, "Bandgap Dependence and Related Features of Radiation Ionization Energies in Semiconductors," J. Appl. Phys. 39, 2029 (1968).
 L. C. Olsen, "Review of Betavoltaic Energy Conversion," in Proc. 12th Space Photovolt. Res. Technol. Conf, 1993, p. 256.
 C. J. Eiting et al., "Demonstration of a Radiation Resistant, High Efficiency SiC Betavoltaic," Appl. Phys. Lett. 88, 064101 (2006).
 P. Rappaport, "The Electron-Voltaic Effect in p-n Junctions Induced by Beta-Particle Bombardment," Phys. Rev. 93, 246 (1954).
 W. G. Pfann and W. Van Roosbroeck, "Radioactive and Photoelectric p-n Junction Power Sources," J. Appl. Phys. 25, 1422 (1954).
 H. Flicker, J. J. Loferski and T. S. Elleman, "Construction of Promethium-147 Atomic Battery," IEEE Trans. Electron Devices 11, 2 (1964)
 P. Bhattacharya, Semiconductor Optoelectronic Devices, 2nd Ed. (Prentice-Hall Inc., 1996).
 M. V. S. Chandrashekhar et al., "Demonstration of a 4H SiC Betavoltaic Cell," Appl. Phys. Lett. 88, 033506 (2006).
 B. Liu et al., "Power-Scaling Performance of a Three-Dimensional Tritium Betavoltaic Diode," Appl. Phys. Lett. 95, 233112 (2009).