A Diamond Gammavoltaic Cell

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

This work presents a new design for a gammavoltaic cell, based on diamond and the surface transfer doping it exhibits. The design evolved from the observation of gammavoltism in a diamond dosimeter during previous work at the University of Bristol. The work first gives an overview of gammavoltaics in literature, highlighting the small size of the field as it currently stands, and the difficulty in comparing between previous works due to the range of approaches used both for conceiving gammavoltaic devices, and for testing them. As part of this overview, it draws attention to some other radiovoltaic work to illustrate where gammavoltaics sit in a wider context. It then discusses the processes by which high-energy photons are converted to conduction electrons within a device. Discussion is split into high-energy photon scattering processes, electron scattering processes (for hot electrons), and charge transport processes (for lower-energy electrons and their holes). As the presented gammavoltaic design is based on diamond, space is then given to the historical context of diamond research and of research into diamond devices under irradiation. Three device-specific aspects of diamond research are then covered in greater depth: the formation of high- and low- barrier electrical contacts on diamond, and the surface transfer doping effect. With this background given, the design for a gammavoltaic cell on which this work is based, is presented: a thick, insulating diamond with opposing, dissimilar contacts, covered on all surfaces with a certain coverage of hydrogen termination.

To marshal the lessons from existing gammavoltaic literature, and also to give clarity and structure to the experiments conducted for this work, five factors are introduced: comparability, accountability, capability, applicability and longevity. No gammavoltaic has yet seen a real-world deployment, to my knowledge. I assert here that each of these five factors has an important role to play in the conceptual development of the field and the progress it makes towards gammavoltaics becoming industrially useful. I define an energy range of 1 - 2000 keV, the Gamut, which I suggest contains all relevant photon energies for gammavoltaic purposes, both application and study. For similar reasons, I suggest a notation convention which lends itself to use for benchmarks in gammavoltaics, with two given directly: Co-60100𝔓, the volumetric maximum power-point density under a 100 Gy/h air KERMA dose rate of Co-60 radiation, in nW/cm3, and Co-60100𝔭, the analogous areal quantity, in nW/cm2.

A Theory chapter is included which gives a brief overview of the basic physics of traditional solar cells based on silicon pn-junctions, the equivalent circuit model which arises from this, and the use of the Lambert W function to derive an explicit I-V expression from the model. It then attempts to apply the same concepts to the diamond gammavoltaic cell presented in this work, although there are several elements which have not been included in the theoretical treatment at this time. An extended equivalent-circuit model based on opposing diodes, the opposing-diodes model, is derived from this circuit. Finally, the theory chapter discusses the fitting parameters space for both models, and the benefit of using orthogonal distance regression for radiovoltaic work, to account for the fact that uncertainty in the applied bias may be substantial as well as the measured current. There is also a Methods chapter which covers the computational aspects of the above as well as the other experimental methods employed in the work.

The first set of results presented are for the purposes of capability and accountability. They seek to prove that the design presented here is capable of working, to a reasonable degree, as a gammavoltaic, and to verify that it works in the way intended. Due to the unusual method of hydrogen termination, required by the fact electrode contacts must be deposited prior to termination, an x-ray photoelectron spectroscopy study is presented which shows the enabling hydrogen termination to have consisted of approximately 0.3 monolayers.

The second set of results here goes further down the route of accountability, using a chain of synchrotron experiments and GEANT4 simulations to first attempt to validate the modelling approach used in this work, and then attempt to probe the operation of a diamond gammavoltaic cell under high fluxes of photons as the energy is varied between 50 keV - 150 keV, the region in which Compton scattering being to dominate over photoelectric absorption for diamond. Unlike in work published on the latter experiment, conclusions drawn are fairly limited due to uncertainties about the method that have arisen since publication.

The third set of results addresses the factors of comparability, applicability (to the gamma field in a nuclear waste store), and longevity (again, in the context of deploying in a nuclear waste store). Measurements are performed under irradiation from gamma rays from both Cs-137 and Co-60. In each case, air KERMA dose-rates of tens to thousands of Grays per hour are attained. An equivalent circuit model based on two opposing diodes is fitted to each curve in an attempt to extract quasi-physical parameters and observe their change with dose rate, to mixed success. For comparability, benchmark values are found of Co-60100𝔭 = 27 nW/cm2 and Co-60100𝔓 = 179 nW/cm3. Longevity testing showed that the device design is promising for longevity. It suffered no catastrophic degradation after 800 kGy air KERMA of Cs-137, after an irradiation at 1,350 Gy/h air KERMA for over 3.5 wks. At most, degradation was around 16 \%, but this is an overestimate due to temperature and humidity effects also playing a role. The longevity status of the gammavoltaic design presented here is promising but requires more study, over longer periods of irradiation.

The summary and conclusions of the work include avenues for improvement in the device, and early work that has been undertaken to address them. The thesis culminates with details of a demonstration in which a diamond gammavoltaic device powers a Bluetooth monitoring circuit, sufficiently to transmit a temperature and humidity measurement after charging from x-rays for 10 h. This is believed to be the first time a gammavoltaic has been used to power any kind of electronic device.
Date of Award9 May 2023
Original languageEnglish
Awarding Institution
  • University of Bristol
SupervisorThomas Bligh Scott (Supervisor) & Neil A Fox (Supervisor)

Keywords

  • Diamond
  • Gammavoltaics
  • Photovoltaics
  • Nuclear Waste

Cite this

'