CC BY
42
UDC 621.039.58, 621.384.6
GEANT4 Code Application for Radiation Environment Prediction at the NICA Complex
G. Timoshenko*, M. Paraipan^
* Laboratory of Radiation Biology Joint Institute for Nuclear Research Joliot-Curie 6, 141980 Dubna, Moscow region, Russia ^ Institute of Space Research P.O. Box MG-23, Ro 077125, Bucharest-Magurele, Romania
The operation of a high-energy ion facility provokes secondary radiation along an accelerator ring and especially at the local sites of maximum beam losses (outlet devices, targets, and beam dumps). An essential condition for the commissioning of a relativistic heavy ion accelerator is appropriate radiation shielding for every radiation element of the complex. The shielding design is connected with two crucial problems: the estimation of the source term and the prognostication of the neutron fluence and equivalent dose distributions around the shielding.
As regards the first problem, the experimental data on the double differential cross section and secondary neutron production in thick targets for a primary uranium beam with the energy of several GeV/n are practically lacking. Few Monte Carlo multipurpose codes able to simulate the uranium ion interaction with, and transport into, the matter are now available. A comparison of FLUKA, GEAT4, and SHIELD simulations with unique experimental data on neutron production in a 1GeV/n 238U beam interaction with a thick Fe target was performed to find the most suitable code. As a result, the GEANT4 code was chosen to carry out a simulation of the NICA (Nuclotron-Based Ion Facility at JINR) complex radiation shielding. Forming the secondary radiation field inside and behind the ordinary concrete shielding was analyzed as well. Some regularities of the secondary neutron field generation in a 4.5 GeV/n uranium beam interaction with thick targets are discussed.
As concerns the second problem, it was found that the crucial point determining the NICA shielding design is that the yearly equivalent dose at the border of the Laboratory site must not exceed 1 mSv. The radiation situation at long distances from NICA will be formed by neutrons which escaped from the shielding of the NICA radiation sources and were then multiscattered in the air and ground ("skyshine" neutrons). The GEANT4 calculations of the "skyshine" neutron radial distributions around all the elements of the NICA complex were carried out, and guidelines for the shielding construction were worked out for different operation modes of the complex.
Key words and phrases: Monte Carlo code, shielding data, relativistic heavy ions, secondary radiation field, thick target, neutron yield, attenuation curve, neutron fluence and equivalent dose.
1. Introduction
The design of the NICA radiation shielding and prognostication of radiation situation around it is a rather complicated task owing to the uncommon primary source terms, proximity to the border of radiation control area, and a collection of radiation sources of various types (the booster, Nuclotron, stripping station, collider, beam transport channels, beam stoppers, and MPD) disposed over the large territory of the Veksler and Baldin Laboratory of High Energy Physics (VBLHEP). The general layout of the NICA acceleration complex with the main beam parameters is shown in Fig. 1. The NICA complex is intended to accelerate heavy ions with A ~ 200 and energy up to 4.5 GeV/n for 197Au and 238U. It is planned to build the booster inside the synchrophasotron magnet at the place of its vacuum chamber. Ions with an energy ~ 0,5 GeV/n coming from the booster will be stripped by a thin target with efficiency about 40%. The beam of non-fully stripped ions will be blocked within the station
Received 24th November, 2009.
The authors would like to thank A. Sidorin, I. Meshkov, and G. Trubnikov for their constant interest in this work.
stopper. The collider will accumulate 17 bunches x 109 ions of 197Au x 2 rings for ~ 2 min; then it will be keeping the beam during one hour; afteewards, it will unload the beam into the stopper.
Injector 2*109 ions/pulse
of 197Au or233U atenergy of6 MeV/u jr Booster
Figure 1. The principal scheme of the NICA complex
The placement of the structural elements of the complex (radiation sources) is shown in Fig. 2. The crucial point determining the NICA shielding design is meeting the indispensable requirement that the yearly equivalent dose be < 1 mSv on the border of the LHEP site [1]. The radiation situation around the NICA complex will be formed by the neutrons escaped from the shielding of the NICA radiation sources and multiscattered in the air and ground ( "skyshine" neutrons).
Nuclotron
sources
\ border of the VBLHEP site
booster
' icOTrtlS ■
.. " ' stripping target
mj, -
Old synchropftasofron building
Experimental li.ill
collider stopper
Figure 2. The placement of the NICA complex at the VBLHEP site
In order to reduce the dose level at the border of the Laboratory site, it was first proposed to arrange the ion stripping station with the stopper near the second magnet quadrant. It was also proposed to combine both collider stoppers consisting of two rings into a single bidirectional stopper with the iron core and arrange it behind the
collider (as is shown in Fig. 2). The iron of the synchrophasotron magnet will be the perfect radiation shielding of the booster ring, but the shielding has to overlap the open linear spaces of the synchrophasotron magnet. The main radiation sources contributing to the dose at the border of the Laboratory site will be the Nuclotron, collider, and beam transport channels. It is necessary to mount the thick upper shielding of the Nuclotron ring, which is placed within the former cable tunnel around the synchrophasotron. The coupled collider rings and beam transport channels also have to be surrounded by thick compact shielding owing to the immediate vicinity of the control area border.
The total annual effective dose of "skyshine" neutrons and gammas from all radiation sources at the NICA control area border must be less than 0.5 mSv with the reserve coefficient 2 [2]. The real annual dose level will depend on several parameters: beam losses at all stages of ion acceleration and storage, yearly NICA schedule for all modes of its operation, shielding efficiency, distance from each source to the control area border, and so on. The choice of the appropriate method for the source term description and calculating the internuclear cascade within the shielding matter is very important as well in order to reliably simulate neutron leakage from the source shielding.
2. Simulation
The simulation of the accelerated 238U and 197Au ions interaction with the beam guide material ¡and the internuclear cascade within the shielding were carried out with the GEAN4 code after verific ation old so me cod es withavailable exper imental d at a [3].
To test the neutron source term, considered was, first of all, secondary neutron production in the interaction of the projectiles with different atomic masses and the same energy per nucleon i4.5 GeV/n) with a thick Fe target (10x10x20 c m). A Fe target was chosen since iron is a typical material of an accelerator vacuum chamber. Neutron yield s from the target in differe nt entrgy and angular ranges are shown in Fig1 3 depending on the projectile mass (relative to neutron yields from 4.5-GeV protons). The factors in the round brackets are the neutron yields per proton required to obtain the absolute values on the neutron yields for projectiles.
Figure 3. Relative secondary neutron yields from a th ick Fe target bombarded by beam projectiles with different atomic; masses and an energy of 4.5 GeV/n. NA+Fe is the
neutron yield for the beam projectiles, Np+Fe is the neutron yield for a 4.5-GeV proton beam. The solid lines are linear approximations. The dashed lines show the
A-equivalentproton approaches
The neutron yield from a thick target in 4^ sr and in the whole energy range per one uranium ion with an energy of 4.5 GeV is about 1.5103. The forward flying neutrons spread up to energies higher than the initial projectile energy per nucleon owing to the cumulative effect. This effect is shown in Fig. 4. At large emission angles, the lower energy neutrons dominate in the yields due to the evaporation process. As a consequence, the changes in the neutron yield and the spectrum shape related to an increase in projectile energy are weaker in the lateral direction than in the forward direction.
Figure 4. Depende nces of the neutron spectra from a thick Fe target on uranium projectile energy in the forward and lateral directions
The target material dependences of the neutron spectra generated by 4.5-GeV/n uranium ions in the forward and leteral directions are shown in Fig. 5. All spectra are normalized to one neutron emitted from the target. It is clear that the hardest spectra in 1the forward (direction correspond to a light target.
Figure5. Dependences od the neucron spectra from 4.5-GeV/n uranium ions on the target material in the forward and lateral directions
The Nuclotron, collider, and surrounding equipment were simulated by an iron torus pipe with 5-cm thick walls for the calculation of the vacuum chambers of the booster. The angles of the ion loss emission from the beam used for the calculation were within the statistical sampling of 0o-1o. The secondary hadrons from the vacuum chambers irradiate the internal surfaces of the lateral concrete shielding. The detailed geometries of the shielding design and the simulation of "skyshine" neutron fields around every NICA element are given in [4].
In the real geometry of the lateral thick shielding irradiation by an extended source of secondary neutrons (a circular vacuum chambers of the accelerators and collider), the attenuation curves for the neutron fluence within the shielding don't have buildup maximums (contrary to the case of normal irradiation) and slightly change with the energy of heavy ions increasing from 1 to 4.5 GeV/n. For example, the attenuation curves for the concrete (p = 2.34g-cm-3) shielding of the collider are shown in Fig. 6 for two energies of the uranium beam. The geometry of shielding iiradiation is schematically shown in the inset in Fig. 6.
For the booster and Nuclotron shielding, simulations of near-real energy distributions of the ion beam losses were taken into account. Then the radial distributions of the "skyshine" neutron effective dose were simulated for each NICA element. The spatial dose rate distributions around the Nucloton with the upper 3-m concrete shielding are presented in Fig. 7A for different energies of lost uranium ions (normalized to one
0 12 3 4
Depth of concrete shielding, m
Figure 6. Total neutron fluerece attenuation inside ordinary cencreee shielding of the collider depending on uranium ion energy
ion lost alonn the vacuum chamb er ring per one seeon d). In Fig. 7B, "the similar fpateal distribution of eeutron dose rate eround the collider is shown.
border of the control area
U238 - 3 m concrete
—■— 4.5 GeV/rr
— □— 3.5 GeV/n
—▲— 22.55 GeV/n
—A— 1.5 GeV/n
—9— 1.0 GeV/n
—o— 0.44 GeV/n
cR
bflL
_
^ctR
80 100 120 140 160 1(30 Distance from the Nuclotron center, m
^ B
border of the contrnl area
60 70 80 90 100 Distance from the collider center, m
-9
110
Figure7. Uranium ion energy dependenues or the radial distribution of the "skyshine" neutron effective dose rate around the Nuclotron with 3-m concrete upper shielding of
the tunnel (A) and the radial distribuf ion of the "skyshine" neutron effective dose rate around the collider with 3-m compact concrete shielding surrounding; the coupled
rings (B)
Analogous spatial distributions were obtained for other NICA radiation sources both for uranium and gold ions. Different variants of the Nuclotron and collider shielding design were tested as well. For the Nuclotron tunnel upper shielding, 0.77-m thick steel was considered as one of the variants. A self-sunk design of the collider located near the Nuclotron was also examined. This variant combines the concrete and ground shielding of the rings as shown in Fig. 8A. The dependences of the neutron effective dose rate around the underground collider on the thickness of the ground layer are presented in Fig. 8B.
Estimations of radiation conditions at the border of the NICA control area were done for different designs of the Nuclotron and collider shielding. The values of the
Figure 8. A design of the collider semi-sunk; in the ground (A) and spatial distributions of the "skyshine" neutron effective dose rate depending on the ground layer under the
rings (B)
uniformly distributed beam losses were taken asfollows: the booster accounts for 5% (E > 50 MeV/n), the ion stripping station - 60% (E = 0.45 GeV/n), the Nuclotron — 5% (0.45 < E < 4.5 GeV/n), the beam transport channel — 3% (£. = 4.5 GeV/n), and the collider — 20% (E = 4.5 GeV/n) during one hour of the beam storage. The ion stripping station stopper will be holding 6-108 ion-s-1 with an energy of 0. 45 GeV/n, and the single bidirectional stopper of the collider will damp 1.12 • 1011 ions with an energy 4.5 GeV/n once in an hour. A total ol1 4000 hours of the annual operation of the complex with uranium or gold ions accelerated to 4.5 GeV/n and additionally 500 hours of tuning time were taken in consideration. The results of estimations for uranium and gold ions are presented in Tables 1 and 2.
Table 1
The total annual neutron effective dose and the partial contributions to it at the border of the NICA control area associated with the acceleration of uranium and gold ions for different designs of the Nuclotron and collider shielding (within Building 205).
238 JJ
Radiation source, 238 y acceleration Booster Stripping station stopper Nuclotron Beam transport channels Collider Collider stopper NICA Complex, total
Annual neutron dose, mSv 0.0018 0.0018 0.0166 0.0166 0.0082 (3 m concrete) 0.0604 (2 m concrete) 0.0501 0.0501 0.0347 (3 m concrete) 0.0943 (2.5 m concrete) 0.0562 0.0562 0.1676 0.2794
It is well known that the contribution of gammas to the total "skyshine" effective dose at large distances from high-energy accelerators does not exceed 10%. Thus, the proposed variants of the shielding of the NICA radiation sources will ensure that the annual equivalent dose at the border of the VBLHEP site will be less than 0,5 mSv (even taking into account a 10% contribution of gammas to the total dose). It will allow additional proton runs at the Nuclotron with a yearly duration of ^1500 h and the proton beam intensity of 1011 protons/cycle without exceeding the prescribed dose limit.
Table 2
The total annual neutron effective dose and the partial contributions to it at the border of the NICA control area associated with the acceleration of uranium and gold ions for different designs of the Nuclotron and collider shielding (within Building 205).
197Au
Radiation source Booster Stripping station stopper Nuclotron Beam transport channels Collider Collider stopper NICA Complex, total
Annual 0.0015 0.0137 0.0499 0.0414 0.0779 0.0465 0.231
neutron 0.0015 0.0137 (2 m con- 0.0414 (2.5 m 0.0465 0.185
dose, crete) con-
mSv 0.0034 crete)
(0.77 m 0.0779
steel) (2.5 m
con-
crete)
References
1. Timoshenko G., Paraipan M. Radiation Safety Standards NRB-99. — Moscow: Atomizdat, 1999.
2. Russian Ministry of Health, Moscow, 2000. — Main Sanitary Rules of Radiation Protection Guarantee for Workers and the Public OSPORB-99.
3. Verification of Monte-Carlo Transport Codes FLUKA, GEANT4 and SHIELD for Radiation Protection Purposes at Relativistic Heavy Ion Accelerators / L. Beskrov-naia, B. Florko, M. Paraipan, N. Sobolevsky // NIM B. — 2008. — Vol. 266. — Pp. 4058-4060.
4. Timoshenko G., Paraipan M. Formation of Secondary Radiation Fields at NICA // NIM B. — 2009. — Vol. 267. — Pp. 2866-2969.
УДК 621.039.58, 621.384.6
Применение программы GEANT4 для прогнозирования радиационной обстановки на комплексе NICA
Г. Н. Тимошенко*, М. М. ПарайпаН
* Лаборатория 'радиационной биологии Объединённый институт ядерных исследований ул. Жолио-Кюри д.6, Дубна, Московская область, 141980, Россия ^ Институт космических исследований P.O. Box MG-23, Ro 077125, Бухарест-Магуреле, Румыния
При работе высокоэнергетичного ускорителя тяжелых ионов генерируется вторичное излучение как по кольцу ускорителя, так и на локальных участках, где наблюдаются максимумы потерь пучка ядер (выводные устройства, мишени и ловушки пучка). Необходимым условием реализации проекта ускорителя тяжёлых ядер является создание радиационной защиты всех элементов ускорительного комплекса. Острыми проблемами при конструировании радиационной защиты такого комплекса являются проблема корректного описания источника излучения и проблема оценки флюенса и эквивалентной дозы нейтронов вокруг ускорителя.
Экспериментальные данные по двойным дифференциальным выходам нейтронов из толстых мишеней, бомбардируемых ядрами урана с энергией в несколько ГэВ/н, практически отсутствуют. В настоящее время существует несколько универсальных Монте-Карло транспортных программ, которые могут моделировать взаимодействия с веществом ядер урана. Для выбора наиболее адекватной программы было выполнено сравнение расчетов по программам FLUKA, GEANT4 и SHIELD с данными единственного
эксперимента по изучению выхода нейтронов из толстой Fe мишени, облучённой пучком ядер 238U с энергией 1 ГэВ/н. В результате сравнения для расчета радиационной защиты комплекса NICA в ОИЯИ была выбрана программа GEANT4. С её помощью был исследован механизм формирования полей вторичного излучения как внутри, так и за защитой из обычного бетона. В данной статье исследованы также некоторые особенности формирования полей вторичного нейтронного излучения при взаимодействии пучка ядер урана с энергией 4,5 ГэВ/н с толстыми мишенями.
При расчете защит ускорительного комплекса учитывалось, что годовая эффективная доза нейтронов на границе площадки Лаборатории (санитарно-защитной зоны) не должна превышать 1 мЗв. Радиационная обстановка на больших расстояниях от радиационных источников комплекса определяется нейтронами утечки из защит, многократно рассеянными в окружающей среде (воздухе и грунте). Приведены результаты расчетов радиальных распределений эффективной дозы таких нейтронов вокруг каждого элемента комплекса NICA для различных вариантов работы комплекса.
Ключевые слова: программа Монте-Карло, радиационная защита, релятивистские тяжелые ионы, поле вторичного излучения, толстая мишень, выход нейтронов, кривая ослабления, флюенс и эквивалентная доза нейтронов.
