Markus Ziegner1, Tobias Schmitz2, Rustam Khan3, Matthias Blaickner4, Hugo Palmans5, Peter Sharpe6, Gabriele Hampel2, Helmuth Böck7. 1. AIT Austrian Institute of Technology GmbH, Vienna A-1220, Austria and Institute of Atomic and Subatomic Physics, Vienna University of Technology, Vienna A-1020, Austria. 2. Institut für Kernchemie, Johannes Gutenberg-Universität, Mainz DE-55128, Germany. 3. Pakistan Institute of Engineering and Applied Sciences (PIEAS), Islamabad PK-44000, Pakistan. 4. AIT Austrian Institute of Technology GmbH, Vienna A-1220, Austria. 5. Acoustics and Ionising Radiation Division, National Physical Laboratory, Teddington TW11 0LW, United Kingdom and Medical Physics Group, EBG MedAustron GmbH, Wiener Neustadt A-2700, Austria. 6. Acoustics and Ionising Radiation Division, National Physical Laboratory, Teddington TW11 0LW, United Kingdom. 7. Institute of Atomic and Subatomic Physics, Vienna University of Technology, Vienna A-1020, Austria.
Abstract
PURPOSE: In order to build up a reliable dose monitoring system for boron neutron capture therapy (BNCT) applications at the TRIGA reactor in Mainz, a computer model for the entire reactor was established, simulating the radiation field by means of the Monte Carlo method. The impact of different source definition techniques was compared and the model was validated by experimental fluence and dose determinations. METHODS: The depletion calculation code origen2 was used to compute the burn-up and relevant material composition of each burned fuel element from the day of first reactor operation to its current core. The material composition of the current core was used in a mcnp5 model of the initial core developed earlier. To perform calculations for the region outside the reactor core, the model was expanded to include the thermal column and compared with the previously established attila model. Subsequently, the computational model is simplified in order to reduce the calculation time. Both simulation models are validated by experiments with different setups using alanine dosimetry and gold activation measurements with two different types of phantoms. RESULTS: The mcnp5 simulated neutron spectrum and source strength are found to be in good agreement with the previous attila model whereas the photon production is much lower. Both mcnp5 simulation models predict all experimental dose values with an accuracy of about 5%. The simulations reveal that a Teflon environment favorably reduces the gamma dose component as compared to a polymethyl methacrylate phantom. CONCLUSIONS: A computer model for BNCT dosimetry was established, allowing the prediction of dosimetric quantities without further calibration and within a reasonable computation time for clinical applications. The good agreement between the mcnp5 simulations and experiments demonstrates that the attila model overestimates the gamma dose contribution. The detailed model can be used for the planning of structural modifications in the thermal column irradiation channel or the use of different irradiation sites than the thermal column, e.g., the beam tubes.
PURPOSE: In order to build up a reliable dose monitoring system for boron neutron capture therapy (BNCT) applications at the TRIGA reactor in Mainz, a computer model for the entire reactor was established, simulating the radiation field by means of the Monte Carlo method. The impact of different source definition techniques was compared and the model was validated by experimental fluence and dose determinations. METHODS: The depletion calculation code origen2 was used to compute the burn-up and relevant material composition of each burned fuel element from the day of first reactor operation to its current core. The material composition of the current core was used in a mcnp5 model of the initial core developed earlier. To perform calculations for the region outside the reactor core, the model was expanded to include the thermal column and compared with the previously established attila model. Subsequently, the computational model is simplified in order to reduce the calculation time. Both simulation models are validated by experiments with different setups using alanine dosimetry and gold activation measurements with two different types of phantoms. RESULTS: The mcnp5 simulated neutron spectrum and source strength are found to be in good agreement with the previous attila model whereas the photon production is much lower. Both mcnp5 simulation models predict all experimental dose values with an accuracy of about 5%. The simulations reveal that a Teflon environment favorably reduces the gamma dose component as compared to a polymethyl methacrylate phantom. CONCLUSIONS: A computer model for BNCT dosimetry was established, allowing the prediction of dosimetric quantities without further calibration and within a reasonable computation time for clinical applications. The good agreement between the mcnp5 simulations and experiments demonstrates that the attila model overestimates the gamma dose contribution. The detailed model can be used for the planning of structural modifications in the thermal column irradiation channel or the use of different irradiation sites than the thermal column, e.g., the beam tubes.