Program - Brookhaven National Laboratory

In recent years, radiation effects have been recognized as one of serious reliability issues for modern microelectronic
devices operating under various cosmic-ray environments. In the terrestrial environment,
secondary cosmic-ray neutrons are known to cause soft errors due to single-event effects. Physics-based
simulation of the soft error phenomena is an essential tool in understanding the phenomena as well as
designing radiation-tolerant devices. To simulate the soft error phenomenon accurately, it is necessary to
employ highly reliable models describing various physical processes involved in the soft errors. In particular,
the modeling of nuclear reactions is of great importance in terrestrial neutron induced soft errors
initiated by nuclear interaction with constituent materials. Prediction of generated secondary ions has an
impact on simulation of the subsequent physical processes, i.e., charge deposition and collection processes.
Recently, we have developed a multi-scale Monte Carlo simulation framework for neutron induced soft
errors by linking a particle transport code PHITS and a 3-D TCAD simulator HyENEXSS[1,2], and have
applied it to the analyses of terrestrial neutron-induced soft errors in MOSFETs from a 65 nm to a 25 nm
design rule. From the validation of nuclear reaction models used in PHITS[1,3], we have recommended the
combined use of the ”event generator mode(e-mode)” with JENDL-3.3 or JENDL-4.0 below 20 MeV and
the modified Quantum Molecular Dynamics(MQMD) model[4] plus Generalized Evaporation Model(GEM)
above 20 MeV in the soft error simulation with PHITS. Using the simulation code system with PHITS and
HyENEXS, we have investigated major secondary ions causing terrestrial neutron induced soft errors and
the dependence on incident energy. The result has clarified that the secondary H and He ions generated
by neutrons up to several hundreds of MeV play a crucial role in terrestrial neutron-induced soft error
simulation. In addition, our recent study of multi-cell upsets (MCUs) has revealed that its dependence on
incident angle is caused by forward angular distribution of heavy recoils generated by nuclear fragmentation.
This will stimulate further improvement in the modeling of nuclear fragmentation in the future.
[1] S. Abe, Y. Watanabe, N. Shibano, N. Sano, H. Furuta, M. Tsutsui, T. Uemura and T. Arakawa,
”Multi-Scale Monte Carlo Simulation of Soft Errors using PHITS-HyENEXSS code system”, IEEE Trans.
on Nucl. Sci., vol.59, no.4 (2012) in press. [2] S. Abe, Y. Watanabe, N. Shibano, N. Sano, H. Furuta, M.
Tsutsui, T. Uemura and T. Arakawa, ”Neutron-Induced Soft Error Analysis in MOSFETs from a 65nm
to a 25nm Design Rule using Multi-Scale Monte Carlo Simulation Method”, Proc. of 2012 IEEE Int. Rel.
Phys. Symp., April 15-19, 2012, Anaheim CA USA. [3] S. Abe, S. Hirayama, Y. Watanabe, N. Sano, Y.
Tosaka, M. Tsutsui, H. Furuta and T. Imamura, ”Applicability of nuclear reaction models implemented
in PHITS to simulations on single-event effects”, J. of Korean Phys. Soc., 59, 1443-1446 (2011). [4] Y.
Watanabe and D.N. Kadrev, ”Extension of quantum molecular dynamics for production of light complex
particles in nucleon-induced reactions”, Proc. of the Int. Conf. on Nuclear Data for Sci. and Technol.,
April 22-27, 2007, Niece, France, EDP Sciences, 1121-1124 (2008).
JD 3 11:20 AM
Benchmark Experiment of Dose Rate Distribution Around Gamma Knife Medical
Apparatus
Koji Oishi, Kazuaki Kosako, Takashi Nakamura
Institute of Technology, Shimizu Corporation
Gamma Knife is one of the most popular radiation oncology apparatus by using Cobalt 60 radioisotopes.
Since the inventory of Cobalt 60 is very high and the shielding is very complicated, the shielding calculation
for the apparatus is not easy by simple calculation methods. In this study, the detailed shielding calculation
by using three-dimensional calculation method was performed and verified by the measurements.
The dose rate measurements were performed by using the passive detectors and ionization chamber as
an active detector. The measured positions were varied around the Gamma Knife apparatus. For the
143