Antiproton annihilation physics in the Monte Carlo particle transport code SHIELD-HIT12A Vicki Trier Taasti a , Helge Knudsen a , Michael H. Holzscheiter a,b , Nikolai Sobolevsky c,d , Bjarne Thomsen a , Niels Bassler a,⇑ a Dept. of Physics and Astronomy, Aarhus University, Denmark b Dept. of Physics and Astronomy, University of New Mexico, USA c Institute for Nuclear Research of the Russian Academy of Sciences (INR), Moscow, Russia d Moscow Institute of Physics and Technology (MIPT), Dolgoprudny, Russia article info Article history: Received 11 November 2014 Received in revised form 30 January 2015 Accepted 1 February 2015 Available online 16 February 2015 Keywords: Antiprotons Monte Carlo simulations SHIELD-HIT12A AD-4/ACE Annihilation physics abstract The Monte Carlo particle transport code SHIELD-HIT12A is designed to simulate therapeutic beams for cancer radiotherapy with fast ions. SHIELD-HIT12A allows creation of antiproton beam kernels for the treatment planning system TRiP98, but first it must be benchmarked against experimental data. An experimental depth dose curve obtained by the AD-4/ACE collaboration was compared with an earlier version of SHIELD-HIT, but since then inelastic annihilation cross sections for antiprotons have been updated and a more detailed geometric model of the AD-4/ACE experiment was applied. Furthermore, the Fermi–Teller Z-law, which is implemented by default in SHIELD-HIT12A has been shown not to be a good approximation for the capture probability of negative projectiles by nuclei. We investigate other theories which have been developed, and give a better agreement with experimental findings. The consequence of these updates is tested by comparing simulated data with the antiproton depth dose curve in water. It is found that the implementation of these new capture probabilities results in an overestimation of the depth dose curve in the Bragg peak. This can be mitigated by scaling the antiproton collision cross sections, which restores the agreement, but some small deviations still remain. Best agreement is achieved by using the most recent antiproton collision cross sections and the Fermi–Teller Z-law, even if experimental data conclude that the Z-law is inadequately describing annihilation on compounds. We conclude that more experimental cross section data are needed in the lower energy range in order to resolve this contradiction, ideally combined with more rigorous models for annihilation on compounds. Ó 2015 Elsevier B.V. All rights reserved. 1. Introduction Since 2002 the AD-4/ACE research collaboration at CERN has investigated whether antiprotons provide benefits for radiation therapy compared to protons [1–4]. As described in [5] the antipro- ton depth dose curve is quite similar to the one for protons, but an extra dose is deposited in the Bragg peak due to the antiproton annihilation products. When an antiproton comes to rest, it is cap- tured by a nucleus and here it annihilates on either a proton or a neutron, whereby an energy of 1.88 GeV (two times the rest mass of the proton) is released. The annihilation energy is typically con- verted into the rest mass of 4–5 pions – a mixture of p þ , p  and p 0 to satisfy charge conservation – and most of the surplus annihila- tion energy is carried away by the pions as kinetic energy. A minor fraction of the annihilation energy,  30 MeV , is deposited locally and causes a doubling of the dose in the Bragg peak compared to a proton beam [6]. This local dose deposition comes from short-range ions created by nuclear fragmentation of the target atom, since on average 1–2 of the created pions will pen- etrate the remaining nucleus causing it to break up. Even if no pion is absorbed by the target nucleus, the instantaneous removal of a nucleon causes the target nucleus to recoil with a kinetic energy of several MeV. Besides the extra physical dose for antiprotons in the Bragg peak compared with protons, a higher RBE (Relative Biological Effectiveness) is also expected here, since the annihilation frag- ments are assumed to have a higher LET (Linear Energy Transfer). http://dx.doi.org/10.1016/j.nimb.2015.02.002 0168-583X/Ó 2015 Elsevier B.V. All rights reserved. ⇑ Corresponding author. E-mail address: bassler@phys.au.dk (N. Bassler). Nuclear Instruments and Methods in Physics Research B 347 (2015) 65–71 Contents lists available at ScienceDirect Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb