Dr Sebastian Galer

An important requirement across a range of sensitive detectors is to determine accurately the energy deposited by the impact of a particle in a small volume. The particle may be anything from a visible photon through to an X-ray or massive charged particle. We have been developing nanobridge Josephson junctions based SQUIDs and nanoSQUID devices covering the entire range of particle detection energies from 1eV to MeV. In this paper we discuss some developments in nanobridge Josephson junctions fabrication using focussed ion beam (FIB) and how these developments impact future applications. We focus on tuning of the transition temperature of a superconducting thin-film absorber, with the aim to match the absorber Tc to the working temperature range of the SQUID and also on using a new Xe FIB to improve Josephson junction and superconducting film quality.

The increase in the number of therapeutic proton and ion beam centres worldwide has prompted renewed interest in measuring and simulating microdosimetric spectra in order to help understand the complexity underlying the Relative Biological Effectiveness (RBE) of these treatment modalities. In this context we have studied the capability of the Geant4 toolkit to simulate microdosimetric spectra measured with a Tissue Equivalent Proportional Counter (TEPC) in a clinical carbon ion beam. The simulated spectra were compared with published experimental data obtained along the depth dose curve of a 194 MeV/u carbon beam at the GSI, Darmstadt (Gerlach et al., 2002). The initial beam energy and energy spread employed in the simulation were tuned to match the calculated and measured depth dose distributions. A good agreement was found at all depths after a shift of 4.025 mm was taken into account with agreement for the microdosimetric derived RBE values to within 0.4% and 11.9% for depths 40 and 66 mm in PMMA (Perspex). This work demonstrates that the Geant4 toolkit can accurately reproduce experimental microdosimetric data and can thus be used for independent calculation of lineal energy spectra from which RBE estimates can be derived using the equation of Pihet et al. (1990). The work highlights the difficulty in using experimental work to benchmark Monte Carlo simulations and the need for detailed descriptions of experimental setups used.

The high uncertainty in the Relative Biological Effectiveness (RBE) values of particle therapy beam, which are used in combination with the quantity absorbed dose in radiotherapy, together with the increase in the number of particle therapy centres worldwide necessitate a better understating of the biological effect of such modalities. The present novel study is part of performance testing and development of a micro-calorimeter based on Superconducting QUantum Interference Devices (SQUIDs). Unlike other microdosimetric detectors that are used for investigating the energy distribution, this detector provides a direct measurement of energy deposition at the micrometre scale, that can be used to improve our understanding of biological effects in particle therapy application, radiation protection and environmental dosimetry. Temperature rises of less than 1μK are detectable and when combined with the low specific heat capacity of the absorber at cryogenic temperature, extremely high energy deposition sensitivity of approximately 0.4 eV can be achieved. The detector consists of 3 layers: tissue equivalent (TE) absorber, superconducting (SC) absorber and silicon substrate. Ideally all energy would be absorbed in the TE absorber and heat rise in the superconducting layer would arise due to heat conduction from the TE layer. However, in practice direct particle absorption occurs in all 3 layers and must be corrected for. To investigate the thermal behaviour within the detector, and quantify any possible correction, particle tracks were simulated employing Geant4 (v9.6) Monte Carlo simulations. The track information was then passed to the COMSOL Multiphysics (Finite Element Method) software. The 3D heat transfer within each layer was then evaluated in a time-dependent model. For a statistically reliable outcome, the simulations had to be repeated for a large number of particles. An automated system has been developed that couples Geant4 Monte Carlo output to COMSOL for determining the expected distribution of proton tracks and their thermal contribution within the detector. The correction factor for a 3.8 MeV proton pencil beam was determined and applied to the expected spectra. The corrected microdosimetric spectra was shown to have a good agreement with the ideal spectra.