Monte Carlo modeling of the influence of strong magnetic fields on the stem-effect in plastic scintillation detectors used in radiotherapy dosimetry.

Abstract

To investigate the impact of strong magnetic fields on the stem-effect in plastic scintillation detectors (PSDs) using Monte Carlo methods. Prior to building the light guide model, the properties of the Čerenkov process in GEANT4 were investigated by simulating depth-dose and depth-Čerenkov emission profiles in water as functions of Čerenkov process input parameters. In addition, profile simulations were performed for magnetic field strengths ranging from 0T to 1.5T. A PMMA light guide was constructed in GEANT4 using data from the manufacturer and literature. Simulations were performed with the model as functions of depth and fiber-beam angle where the simulated stem-effect spectrum and the Čerenkov light ratio (CLR) were scored and compared to measured data in the literature. The light guide optical properties were iteratively adjusted until agreement between the simulated and measured data was achieved. Simulations were performed with the validated model as functions of depth and magnetic field strength and the simulated data were compared to measured data in the literature. The model was also used to evaluate the sensitivity of the CLR to the various optical properties of the light guide in different irradiation conditions. No significant changes in the depth-dose or depth-Čerenkov emission profiles were observed with step-size restrictions imposed by the Čerenkov process input parameters, which was attributed to the condensed history algorithm and transport parameters used in this work. Similar changes in the depth-dose and depth-Čerenkov emission profiles were observed with increasing magnetic field strength, which indicates the Čerenkov process is not adversely impacted by the presence of the magnetic field. Following optimization of the light guide optical properties, agreement within two standard deviations was observed between the simulated and measured optical data for all validation geometries considered. Agreement within one standard deviation was observed between the simulated and measured data for all depths and field strengths ≥0T whereas discrepancies were observed for magnetic field strengths <-0.35T. These significant differences were attributed to insufficient measurement data for this irradiation configuration during model validation. Of the light guide optical properties investigated, the fluorescence signal had the greatest impact on the CLR sensitivity to the magnetic field. No significant change in the Čerenkov emission per dose in water was observed for magnetic field strengths up to 1.5T. The nominal fiber fluorescence signal was found to have a significant impact on the CLR sensitivity to varying irradiation conditions where changes up to 11.7% were observed whereas the mirror reflectivity and fiber attenuation had a modest impact with maximum CLR changes of 2.6% and 1.2% relative to 0T, respectively. The results of this work suggest light guides with low fiber fluorescence should be used with PSDs for dosimetry measurements in magnetic fields to minimize the impact of the magnetic field on the CLR correction.