See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/310843718

Radiation Protection Studies for Medical Particle Accelerators using Fluka

Monte Carlo Code

Article  in  Radiation Protection Dosimetry · November 2016

DOI: 10.1093/rpd/ncw302

CITATIONS READS

4 929

8 authors, including:

Angelo Infantino Giulia Lucconi

CERN Policlinico S.Orsola-Malpighi
19 PUBLICATIONS   201 CITATIONS    11 PUBLICATIONS   60 CITATIONS   

SEE PROFILE SEE PROFILE

Davide Pancaldi Sara Vichi

Azienda Unità Sanitaria Locale Ferrara University of Bologna
35 PUBLICATIONS   296 CITATIONS    24 PUBLICATIONS   129 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Effects of ionizing radiation on tumour cells View project

Kinetic Theory Applications View project

All content following this page was uploaded by Domiziano Mostacci on 19 January 2017.

The user has requested enhancement of the downloaded file.

 Radiation Protection Dosimetry Advance Access published November 24, 2016

Radiation Protection Dosimetry (2016), pp. 1–7 doi:10.1093/rpd/ncw302

RADIATION PROTECTION STUDIES FOR MEDICAL PARTICLE

ACCELERATORS USING FLUKA MONTE CARLO CODE
Angelo Infantino1,*, Gianfranco Cicoria2, Giulia Lucconi2, Davide Pancaldi2, Sara Vichi2, Federico Zagni2,
Domiziano Mostacci1 and Mario Marengo2
1
Department of Industrial Engineering, Laboratory of Montecuccolino, University of Bologna, Via dei Colli
16, 40136 Bologna, Italy
2
Medical Physics Department, S. Orsola-Malpighi University Hospital, Via Massarenti 9, 40138 Bologna,
Italy

*Corresponding author: [email protected]

Downloaded from http://rpd.oxfordjournals.org/ at CERN Library on November 29, 2016

Radiation protection (RP) in the use of medical cyclotrons involves many aspects both in the routine use and for the decom-
missioning of a site. Guidelines for site planning and installation, as well as for RP assessment, are given in international
documents; however, the latter typically offer analytic methods of calculation of shielding and materials activation, in
approximate or idealised geometry set-ups. The availability of Monte Carlo (MC) codes with accurate up-to-date libraries
for transport and interaction of neutrons and charged particles at energies below 250 MeV, together with the continuously
increasing power of modern computers, makes the systematic use of simulations with realistic geometries possible, yielding
equipment and site-specific evaluation of the source terms, shielding requirements and all quantities relevant to RP at the
same time. In this work, the well-known FLUKA MC code was used to simulate different aspects of RP in the use of biomed-
ical accelerators, particularly for the production of medical radioisotopes. In the context of the Young Professionals Award,
held at the IRPA 14 conference, only a part of the complete work is presented. In particular, the simulation of the GE
PETtrace cyclotron (16.5 MeV) installed at S. Orsola-Malpighi University Hospital evaluated the effective dose distribution
around the equipment; the effective number of neutrons produced per incident proton and their spectral distribution; the acti-
vation of the structure of the cyclotron and the vault walls; the activation of the ambient air, in particular the production of
41
Ar. The simulations were validated, in terms of physical and transport parameters to be used at the energy range of interest,
through an extensive measurement campaign of the neutron environmental dose equivalent using a rem-counter and TLD
dosemeters. The validated model was then used in the design and the licensing request of a new Positron Emission
Tomography facility.

INTRODUCTION
geometry set-ups. Moreover, these guidelines consider
Cyclotrons are widely established in research sites the problems ‘one at a time’, without a comprehensive
and in the biomedical field. In the latter, cyclotrons overview. In other cases, published results are general-
support both diagnostic and therapeutic applica- ised from meta-analysis on existing facilities, as
tions: in the diagnostic field, they are employed for reported by Tesch(3), or are based on general hypoth-
the production of radionuclides for clinical use, in esis on the source term(4, 5). Experimental measure-
particular tracers for Positron Emission Tomography ments of the neutron dose field around PET
(PET); while in hadron therapy cyclotrons are used cyclotrons(6) and the neutron energy spectrum(7–10)
to accelerate protons or heavy ions. can be useful sources of information, but inevitable
In the planning stage of a facility, the aspects differences in equipment, configuration and layout
related to radiation safety are key factors: an accurate should be carefully considered when planning new
‘a priori’ knowledge of the radiation field around the facilities, since generalisation of experimental results
accelerator is essential for the design of shielding (in is not straightforward.
particular as far as ‘bad geometry’ conditions are con- Given the complexity of the physical phenomena
cerned, like mazes, ducts and wall penetrations), the involved in the transport of radiations, the availabil-
selection of appropriate engineered controls and ity of Monte Carlo (MC) codes with accurate and
safety systems, the classification of areas and the pro- up-to-date libraries for radiation transport and inter-
tection of workers, the public and the environment (1). action of neutrons and charged particles at energies
Guidelines for site planning and installation, as well below 250 MeV, together with the continuously
as for radiation protection (RP) assessment, are given increasing power of modern computers, allows the
in a number of international documents; however, systematic use of simulations with realistic geom-
these well-established guides typically discuss analyt- etries. This provides the evaluation of the source
ical methods of calculation of both shielding and term specific for the site and the type of equipment,
materials activation(2), in approximate or idealised the assessment of shielding requirements, the

© The Author 2016. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
 A. INFANTINO ET AL.
perspective estimation of activation of the acceler- pressure transducers and liquid target fillers are
ator itself and the building materials and the assess- located at the Target Panel. The Target Panel is
ment of all the quantities relevant to RP at the same mounted in front of the cyclotron, close to the tar-
time(11). FLUKA(12, 13) is a general purpose code to gets and includes a slab of 5 cm thick lead and 10 cm
model particle transport and interaction with matter; thick polyethylene, both enclosed in a structure of
it covers an extended range of applications spanning iron. The cyclotron, not provided with self-shielding,
from proton and electron accelerators shielding to cal- is installed in a bunker with inner dimensions of
orimetry, dosimetry, detector design, radiotherapy and 650 cm by 535 cm and a height of 350 cm, while the
more. Even if the high-energy physics remains its main concrete walls are 200 cm thick. Data on dimensions
field of application, the number of publications in and features of the cyclotron, its main components
which FLUKA has been used in the medical field has and the bunker were taken from technical data sheets
recently increased, covering the production of medical of the manufacturer and the project drawings of the
radioisotopes(14–16), as well as proton and ion therapy site.
applications(17, 18) and radiation safety design(11, 19, 20)
at this low energy range. Given the main energy range

Downloaded from http://rpd.oxfordjournals.org/ at CERN Library on November 29, 2016

ACSI TR19 cyclotron
of application of FLUKA code, an accurate validation
in terms of physical and transport parameters to be A new PET facility completed in 2014 at Sacro
used at low energy range is mandatory for a correct Cuore-Don Calabria Hospital in Negrar (Verona,
prediction of the quantities of radiological interest in Italy) was modelled. The ACSI TR19 installed in
the medical field. this facility is a compact cyclotron with vertical
In this work, presented in the context of the acceleration plane able to accelerate H− ions, pro-
Young Professionals Award (14th Congress of the duced by an external multicusp ion source, in the
International Radiation Protection Association, 13–19 MeV energy range, with a nominal beam cur-
Cape Town, South Africa, 9–13 May 2016), it is dis- rent of 150 µA. It is provided with two target selec-
cussed only a part of a more comprehensive study tors hosting up to four target assemblies each and
conducted on advanced aspects of RP in the use of allowing a dual beam irradiation, and two local
particle accelerators in the medical field. The shielding made of a proprietary composite material
FLUKA MC code was used to simulate two H− designed by Advanced Cyclotron Systems, Inc.(21) to
cyclotrons (16.5 and 19 MeV, respectively) for the significantly reduce the secondary gamma and neu-
production of PET radionuclides, including their tron field produced during irradiation. The layout of
targetry, the cyclotron vault and ducts. Simulations the cyclotron vault was accurately reproduced from
yielded estimates of various relevant quantities, original technical drawings and optimisation of the
including the effective dose distribution around the ducts was performed using MC simulation
equipment, the fluence of secondary neutrons and (Figure 1b). The inner dimensions of the bunker are
their spectral distribution, and the evaluation of the 460 cm by 540 cm with a height of 380 and 200 cm
long-term activation of components and shielding thick concrete walls.
for the planning of future decommissioning
strategies.
FLUKA MC set-up
In both cases, the proton beam was simulated with a
MATERIALS AND METHODS
Gaussian distribution in energy and divergence and
FLUKA MC models of cyclotrons an elliptical shape in the plane perpendicular to the
beam direction, with a Gaussian distribution on
GE PETtrace cyclotron
both x and y axes. The walls of the bunker were
The cyclotron considered in this study is a PETtrace simulated of standard concrete Portland. Radiation
(GE Medical System), a compact cyclotron with ver- decay was activated in analogue mode and all the
tical acceleration plane, capable of accelerating nega- isotopes were produced in their ground state. A
tive hydrogen and deuterium ions up to 16.5 and default set of physical mechanisms called NEW-
8.4 MeV, respectively. The cyclotron is routinely used DEFA was used and the proton transport threshold
at S. Orsola-Malpighi Hospital for the production of was set to 1 MeV. An irradiation profile of 1 hour at
PET radionuclides. A detailed MC model of the 1 μA of extracted proton current was simulated.
cyclotron and the bunker was developed using Different types of estimators were used to obtain
FLUKA. The model includes an accurate descrip- residual nuclei, neutron spectra, neutron emission
tion of a typical GE target system, comprised of a and equivalent dose. The RESNUCLE estimator
silver chamber filled with [18 O]-water to produce allowed scoring residual nuclei produced in inelastic
Fluorine-18 by the (p,n) reaction. The front of the interactions on a region basis; all activities were cal-
target body is sealed with a 25 µm thick Havar™ culated at the end of bombardment (EOB). The
foil. All external connections of the targets, target USRBIN estimator was used to score the

2
 RP FLUKA STUDIES FOR MEDICAL CYCLOTRONS

Figure 1. (a) Section of the 3D FLUKA MC model of the PETtrace cyclotron and the cyclotron vault at S. Orsola-

Downloaded from http://rpd.oxfordjournals.org/ at CERN Library on November 29, 2016

Malpighi Hospital, showing a detail of the ducts through the vault walls; red numbers indicate the locations in which the
assessment of the neutron ambient dose equivalent was performed. (b) 3D FLUKA MC model of TR19 and the cyclotron
vault, including ducts, installed at Sacro Cuore-Don Calabria Hospital.

distribution of the dose equivalent in a regular spa- RESULTS

tial structure, called mesh, independent from the
Assessment of H*(10) around a PETtrace cyclotron
geometry. The AUXSCORE card was used to assess
the neutron dose field by convoluting the neutron Figure 2 shows the neutron ambient dose equivalent
fluence with the fluence-to-dose equivalent conver- distribution assessed using FLUKA and represented
sion coefficients. USRTRACK score was used to cal- through a Cartesian mesh: for all the positions data
culate the differential neutron fluence distribution in were taken at the same coordinates as the real dose-
an energy-binning basis. To reach statistical uncer- meters. The results of MC simulations are reported
tainty inferior to 1%, 109 primary particles were in Table 1, along with the experimental measure-
simulated using FLUKA version 2011.2b.6; the ments performed in this work. Data obtained using
simulations were performed on a laptop provided the neutron rem-counter were corrected for the dead
with 4-phycal cores Intel® Core™ i7-3630QM in time of the instrument. To compare different irradi-
~48–72 hours. ation conditions, results were normalised to the total
charge accumulated on the target in each irradiation
test, expressed in μAh.
Experimental measurements
As shown in Table 1, in most of the locations there
A careful validation of the simulated dose field is a good agreement between FLUKA results and
around the PETtrace cyclotron was obtained experimental measurements, particularly with the
through a measurement campaign of the neutron neutron rem-counter; small discrepancies might be
ambient dose equivalent. The experimental set-up due to small differences in the modelling of the cyclo-
adopted in this work refers to a previous work con- tron components or in the positioning of the dose-
ducted by this group and published by Gallerani meters. As reported in the technical specifications of
et al.(6) where measurements of the ambient dose the manufacturer, the uncertainty associated with the
equivalent H*(10) were performed within the cyclo- CR39 dosemeters is a function of the measured dose,
tron vault. in the range of 54–72%. This is substantially due to
Measurements were conducted in 12 points located the significant dependence on the energy response of
along 8 directions, at the same height as the target the detector in an intense and complex radiation
used for the production of 18F (Figure 1a). Measure- field, such as the one inside the cyclotron vault dur-
ments were performed during irradiations using a neu- ing an irradiation. Besides, the spectrum used for the
tron rem-counter FHT-752 (Thermo Scientific), calibration differed from the one the dosemeters were
provided with a BF3 proportional-counter and a PE- exposed to. Finally, CR39 dosemeters allow meas-
moderator and directly calibrated in H*(10) and a set uring only neutrons with an energy above 200 keV;
of 12 CR39 fast neutron dosemeters(22). To obtain taking into account the contribution of thermal neu-
dose rates in the measurement range of the detector trons, the agreement between FLUKA and the
and therefore limit the effects of dead time, irradiation experimental measurements would therefore improve.
tests were conducted with an integrated current On the other hand, an electronic instrument like the
between 0.005 and 0.0016 μAh. Results were com- rem-counter allows for a better counting statistics
pared with MC simulations. due to the greater dimensions of the probe, compared

3
 A. INFANTINO ET AL.

Downloaded from http://rpd.oxfordjournals.org/ at CERN Library on November 29, 2016

Figure 2. FLUKA assessment of the neutron ambient dose equivalent H*(10) for the PETtrace cyclotron over the whole
vault with a Cartesian mesh (5 cm × 5 cm × 5 cm).

Table 1. Comparison of neutron ambient dose equivalent and small differences in the modelling of the cycl-
H*(10) obtained from MC simulations and experimental otron components, FLUKA simulation provided an
measurements.
excellent agreement with the experimental measu-
rements and allowed an accurate estimation of the
Position Dose ± uncertainty (mSv μAh−1) neutron dose field.
FHT-752 CR39 FLUKA

1 330 ± 130 380 ± 200 300 ± 5 Design of new PET facilities

2 430 ± 170 300 ± 160 459 ± 6
3 90 ± 40 60 ± 30 96 ± 3
The previously described MC model, validated in
4 420 ± 70 340 ± 180 475 ± 6 terms of physical and transport parameters, was
5 330 ± 130 340 ± 180 311 ± 5 used in the design and the licensing request of a new
6 27 ± 11 6±5 18.4 ± 1.0 PET facility. The details about the shielding calcula-
7 32 ± 13 19 ± 14 34.7 ± 1.6 tions can be found in previous publication(23). The
8 23 ± 9 6±5 18.5 ± 1.0 TR19 cyclotron and the first design of the cyclotron
9 55 ± 22 31 ± 19 54.1 ± 2.0 vault, including ducts for the ventilation system, the
10 45 ± 18 38 ± 22 42.6 ± 1.6 RF cables and water cooling supply, were modelled
11 190 ± 80 130 ± 70 156 ± 3 to assess the neutron ambient dose equivalent
12 22 ± 9 50 ± 29 25.6 ± 1.3
H*(10) during a dual beam irradiation.
Since the composition and the density of the local
Uncertainty is reported as 1σ.
shield were not completely known, the efficacy of the
local shield was verified by evaluating its attenuation
with the TLDs; however, its positioning is more likely factor. The cyclotron was simulated with and with-
affected by uncertainty. out the local shield and the neutron ambient dose
The ratios between FLUKA and the measurements equivalent H*(10) was scored in two points, both
obtained using the FHT-752 and the CR39 were calcu- located 1 m from the target, at 0° (beam direction)
lated for each position; then for each device an average and 90°, respectively. An average attenuation factor
over all the locations was obtained by weighting the of 146 ± 8 was found. Considering the variability in
single uncertainties. An average ratio H*(10)FLUKA/H* the composition of the mixture within the local
(10)experimental of 0.99 ± 0.12 and 1.13 ± 0.21 was found shield, this result is in good agreement with the aver-
for the neutron rem-counter and the CR39 dosemeters, age value of 100 provided by the manufacturer, thus
respectively. Considering all the factors listed above validating the modelled local shield.

4
 RP FLUKA STUDIES FOR MEDICAL CYCLOTRONS
A validated MC model is particularly useful, com- account a realistic workload of the accelerator. The
pared with analytical methods, in cases of ‘bad long-activation of the two local shielding is reported
geometry’ conditions. In Figure 3, the transmission in Figure 4. The radionuclidic inventory was assessed
of dose through the duct of the air exhaust is shown at the EOB and after 4 weeks to take into account a
as a representative example of these situations. representative waiting time before the beginning of
Alternative solutions were then studied and imple- the decommissioning of the site. The quantification
mented in the MC model: the position and the orien- of long half-life radionuclides is important not only
tation of the pipe through the wall were changed to for the RP of workers involved in the decommission-
satisfy planning and RP requirements. Figure 3 also ing, but also to assess the volume of radioactive wa-
shows the final configuration with no significant ste. According to the Italian legislation for RP, a
transmission through the duct. The case reported mixture of radionuclides has to be treated as radio-
here is a typical example of critical situations in active waste if the sum of the ratios between the
which analytical methods fail given the complexity activity concentration, expressed in Bq g−1, of a given
of the geometry and hypothesis made on the source radionuclide and the limit activity concentration is
term of radiation. Using a validated MC model, it is >1 or if radionuclides with a half-life >75 d are pre-

Downloaded from http://rpd.oxfordjournals.org/ at CERN Library on November 29, 2016

possible to accurately reproduce both the real geom- sent in the mixture. Taking into account the mass of
etry and the source of radiations without any the local shields (~3.8 t each), the radionuclides
hypothesis. Moreover, the identification of these crit- found to be critical resulted: 51Cr (27.7 d) 6.48 Bq g−1,
icalities during the design stage allows a significant 54
Mn (312.3 d) 139 Bq g−1, 55Fe (2.73 y) 160 Bq g−1
reduction of the costs of the mitigations actions to and 59Fe (44.5 d) 25 Bq g−1.
carry out once the structure is already built.
Long-term activation of local shields and vault
CONCLUSION
walls was studied to estimate the activity after 10 y
in operation and to plan strategies for decommis- The availability of MC codes makes it possible
sioning. It is particularly interesting to assess the to revise the traditional approach in the assessment
order of magnitude of the long half-life radionuclides of RP problems, in particular in the case of ‘bad
that might be managed by workers involved in the geometry’ conditions, like mazes, ducts and wall pe-
decommissioning. Generally, the cyclotron works netrations. In the past, a variety of analytical and
some hours per day, depending on the workload of semi-empirical models were proposed, trying to gen-
the facility, but between irradiations long half-life eralise the available experimental results by fitting
radionuclides do not decay significantly and the data or introducing correction factors of doubtful
effect during 1 y in operation is an accumulation of physical meaning. Nowadays, MC codes are more
the activity produced; the order of magnitude of the easily accessible and can be routinely applied to the
activity of long half-life radionuclides is therefore study of particles transport from high energy down
better reproduced by an average irradiation profile. to a few keV. Yet notwithstanding the availability of
A 10-y irradiation was thus simulated using an aver- increasingly sophisticated codes and routines model-
age proton current of 2.3 μA, calculated taking into ling complex 3D geometries, one should not lose

Figure 3. (a) First version of the planning of the TR19 cyclotron vault: a transmission of the dose through the pipe of the
out-coming ventilation was identified. (b) Optimisation of the position and the orientation of the pipe through the wall
allowed avoiding any significant transmission of dose.

5
 A. INFANTINO ET AL.

Downloaded from http://rpd.oxfordjournals.org/ at CERN Library on November 29, 2016

Figure 4. Long-term activation of local shields assessed at EOB and 4 weeks later. The two local shields are called ‘North’
and ‘South’ according to the notation used by the civil engineering in the original drawings of the facility.

sight of the need for a careful choice of the physics of RP and the study of feasibility and efficiency of
and transport parameters adopted, as well as an new radionuclides production methodologies.
accurate validation of the set-up simulated by com-
parison with experimental measurements.
In this work it is discussed the application of MC
simulation to the energy range of particle accelerators REFERENCES
of medical interest. This work shows that it is pos- 1. ICRP, 2007. The 2007 Recommendations of the
sible to accurately predict the quantities of radio- International Commission on Radiological Protection.
logical interest in the energy range of biomedical ICRP Publication 103. Ann. ICRP 2007; 37(2–4).
cyclotrons, in a unitary and comprehensive approach, 2. NCRP, 2003. Radiation Protection for Particle
more satisfactory compared with analytical methods. Accelerator Facilities: Recommendations of the National
The availability of an experimentally validated MC Council on Radiation Protection and Measurements.
model, in terms of physical and transport parameters NCRP report; no. 144, (Bethesda, MD).
3. Tesch, K. A simple estimation of the lateral shielding for
in the energy range of medical applications, makes it proton accelerators in the energy range 50 to 1000 MeV.
possible to revise the traditional approach to the Radiat. Prot. Dosimetry 11(3), 165–172 (1985).
assessment of RP problems. Currently, MC simula- 4. Birattari, C., Bonardi, M., Ferrari, A. and Silari, M.
tion can therefore be considered the ‘state of art’ in Neutron activation of air by a biomedical cyclotron and
the planning of new biomedical cyclotron installa- an assessment of dose to neighbourhood populations.
tions, being a powerful tool for both the optimisation Radiat. Prot. Dosim. 14(4), 311–319 (1985).

6
 RP FLUKA STUDIES FOR MEDICAL CYCLOTRONS
5. Birattari, C., Ferrari, A., Parnell, C.J. and Silari, M. The 89Zr production using the Monte Carlo code FLUKA.
environmental release of radioactive gases produced by App. Radiat. Isot. 69, 1134–1137 (2011).
neutron activation of air with the new Hammersmith cyclo- 15. Infantino, A., Oehlke, E., Mostacci, D., Schaffer, P.,
tron. Radiat. Prot. Dosimetry 19(3), 183–186 (1987). Trinczek, M. and Hoehr, C. Assessment of the produc-
6. Gallerani, R., Cicoria, G., Fantuzzi, E., Marengo, M. tion of medical isotopes using the Monte Carlo code
and Mostacci, D. Neutron production in the operation FLUKA: simulations against experimental measure-
of a 16.5 MeV PETtrace cyclotron. Pro. Nucl. Energy ments. Nuclear instruments and methods in physics.
50, 939–943 (2008). Nucl. Instrum. Methods Phys. Res. B. 366, 117–123
7. Silari, M., Agosteo, S., Beck, P., Bedogni, R., Cale, E., (2016).
Caresana, M., Domingo, C., Donadille, L., Dubourg, N., 16. Oranj, L.M., Kakavand, T., Sadeghi, M. and Rovias,
Esposito, A. et al. Intercomparison of radiation protection M.A. Monte Carlo FLUKA code simulation for study
devices in high-energy stray neutron field. Part III: of 68Ga production by direct proton-induced reaction.
Instrument response. Radiat. Meas. 44, 673–691 (2009). Nuclear Instruments and Methods in Physics. Nucl.
8. Guimarães, A.M., Lacerda, M.A.S., Santos, J.A.L., Instrum. Methods Phys. Res. A 677, 22–24 (2012).
Maletta, P.G.M., Rodrigues, S.L.M., Andrade, R.S., 17. Battistoni, G., Cappucci, F., Bertolino, N., Brambilla,
Vilela, E.C. and da Silva, T.A. Use of a TLD-based M.G., Mainardi, H.S. and Torresin, A. FLUKA Monte

Downloaded from http://rpd.oxfordjournals.org/ at CERN Library on November 29, 2016

multisphere spectrometry system to measure the neutron Carlo simulation for the Leksell Gamma Knife Perfex-
spectra around a not-self-shielded PET cyclotron: pre- ion radiosurgery system: homogeneous media. Phys.
liminary results. App. Radiat. Isot. 71, 92–95 (2012). Med. 29, 656–661 (2013).
9. Wiegel, B., Agosteo, S., Bedogni, R., Caresana, M., 18. Parodi, K., Ferrari, A., Sommerer, F. and Paganetti,
Esposito, A., Fehrenbacher, G., Ferrarini, M., H. Clinical CT-based calculations of dose and positron
Hohmann, E., Hranitzky, C., Kasper, A. et al. emitter distributions in proton therapy using the
Intercomparison of radiation protection devices in a FLUKA Monte Carlo code. Phys. Med. Biol. 52,
high-energy stray neutron field, Part II: Bonner sphere 3369–3387 (2007).
spectrometry. Radiat. Meas. 44, 660–672 (2009). 19. Infantino, A., Valtieri, L., Cicoria, G., Pancaldi, D.,
10. Vega-Carrillo, H.R. Neutron energy spectra inside a Mostacci, D. and Marengo, M. Experimental mea-
PET cyclotron vault room. Nuclear instruments and surement and Monte Carlo assessment of Argon-41 pr-
methods in physics. Nucl. Instrum. Methods Phys. Res. oduction in a PET cyclotron facility. Phys. Med.
A. 463, 375–386 (2001). (2015) doi:http://dx.doi.org/10.1016/j.ejmp.2015.07.
11. Infantino, A., Marengo, M., Baschetti, S., Cicoria, G., 146, In Press,.
Longo Vaschetto, V., Lucconi, G., Massucci, P., Vichi, 20. Patil, B.J., Chavan, S.T., Pethe, S.N., Krishnan, R.,
S., Zagni, F. and Mostacci, D. Accurate Monte Carlo Bhoraskar, V.N. and Dhole, S.D. Estimation of neutron
modeling of cyclotrons for optimization of shielding and production from accelerator head assembly of 15 MV
activation calculations in the biomedical field. Radiat. medical LINAC using FLUKA simulations. Nuclear
Phys. Chem. 116, 231–236 (2015). Instruments and Methods in Physics. Nucl. Instrum.
12. Böhlen, T.T., Cerutti, F., Chin, M.P.W., Fassò, A., Methods Phys. Res. B 269, 3261–3265 (2011).
Ferrari, A., Ortega, P.G., Mairani, A., Sala, P.R., 21. Advanced Cyclotron System Inc. (ACSI), 2014. TR19
Smirnov, G. and Vlachoudis, V. The FLUKA code: PET Cyclotron. Operating and Maintenance Manual
developments and challenges for high energy and medical Rev. 5.0.‘, Richmond, BC, Canada.
applications. Nuclear Data Sheets. 120, 211–214 (2014). 22. ENEA. Dosimetro a corpo intero per neutroni veloci (nv).
13. Ferrari, A., Sala, P.R., Fassò, A. and Ranft, J. Available at: http://www.irp.enea.it/it/servizi/documenti/a5-
FLUKA: a multi-particle transport code. In: CERN- dosimetro-neutroni-veloci; 2013.
2005-10, INFN/TC_05/11, SLAC-R-773 (Geneva: 23. Infantino, A.2015Advanced aspects of radiation protec-
CERN) pp. 1–467 (2005). tion in the use of particle accelerators in the medical
14. Infantino, A., Cicoria, G., Pancaldi, D., Boschi, S., field. PhD Thesis, University of Bologna, Italy.
Fanti, S., Marengo, M. and Mostacci, D. Prediction of Available at: http://amsdottorato.unibo.it/6807/

View publication stats