Program Information
A Monte Carlo Investigation of a Novel Detector Arrangement for the Energy Spectrum Measurement of a 6MV Linear Accelerator
S Taneja*, L Bartol , W Culberson , L DeWerd , School of Medicine and Public Health, University of Wisconsin-Madison, Madison, WI
Presentations
TH-CD-201-2 (Thursday, August 4, 2016) 10:00 AM - 12:00 PM Room: 201
Purpose: Direct measurement of the energy spectrum of a 6MV linear accelerator has not been successful due to the high fluence rate, high energy nature of these photon beams. Previous work used a Compton Scattering (CS) spectrometry setup with a shielded spectrometer for spectrum measurements. Despite substantial lead shielding, excessive pulse pile-up was seen. MCNP6 transport code was used to investigate the feasibility and effectiveness of performing measurements using a novel detector setup.
Methods: Simulations were performed with a shielded high-purity germanium (HPGe) semiconductor detector placed in the accelerator vault’s maze, with a 2 cm diameter collimator through a 92 cm thick concrete wall. The detector was positioned 660 cm from a scattering rod (placed at isocenter) at an angle of 45° relative to the central axis. This setup was compared with the shielded detector positioned in the room, 200 cm from the scattering rod at the same CS angle. Simulations were used to determine fluence contributions from three sources: (1) CS photons traveling through the collimator aperture, the intended signal, (2) CS scatter photons penetrating the detector shield, and (3) room-scattered photons penetrating the detector shield. Variance reduction techniques including weight windows, DXTRAN spheres, forced collisions, and energy cutoffs were used.
Results: Simulations showed that the number of pulses per starting particle from an F8 detector tally for the intended signal decreased by a factor of 10² when moving the detector out of the vault. This reduction in signal was amplified for the unwanted scatter signal which decreased by up to a factor of 10⁹.
Conclusion: This work used MCNP6 to show that using a vault wall to shield unwanted scatter and increasing isocenter-to-detector distance reduces unwanted fluence to the detector. This study aimed to provide motivation for future experimental work using the proposed setup.
Contact Email: