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

net/publication/259811814

Predicting Induced Activity in the Havar Foils of the 18F Production Targets
of a PET Cyclotron and Derived Radiological Risk

Article  in  Health Physics · August 2014

DOI: 10.1097/HP.0000000000000064

CITATIONS READS

7 1,498

2 authors, including:

Javier Martinez-Serrano
University of Malaga
6 PUBLICATIONS   54 CITATIONS   

SEE PROFILE

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

Modeling dynamics of plasma produced by laser ablation of nanoparticles of different compositions View project

All content following this page was uploaded by Javier Martinez-Serrano on 27 May 2016.

The user has requested enhancement of the downloaded file.

 Paper
PREDICTING INDUCED ACTIVITY IN THE HAVAR FOILS OF THE
18
F PRODUCTION TARGETS OF A PET CYCLOTRON AND
DERIVED RADIOLOGICAL RISK
J. Javier Martinez-Serrano* and Antonio Diez de los Rios†
Abstract—The PET cyclotron at the Centre of Molecular Imag-
ing of the Universidad de Malaga (CIMES) is a 16.5 MeV GE
PETtrace cyclotron working at dual beam (40 μA beam−1).
The cyclotron is dedicated mainly to 18F production. The 18F tar-
get has two thin circular foils composed of a metal alloy (Havar),
that are highly activated by the proton beam and secondary
neutrons. The main purpose of this study is to assess induced ac-
tivity radiological risk derived from the Havar foils activation. In-
duced activity in Havar foils was estimated by two procedures.
One consisted in estimating neutron and proton fluxes with
MCNPX and using them as inputs in the activation code ACAB.
Alternatively, given the regular periodicity of the irradiation
cycles, an analytical expression was derived to estimate activity
concentrations of activation products using production rates
calculated with MCNPX. Large differences were found in the
induced foil activities predicted by the two procedures. There-
fore, an irradiated vacuum foil was measured with a Ge detec-
tor to analyze activity levels. Cobalt-58 (58Co) and 56Co activities
calculated with ACAB match well with measurements. Cobalt-60
(60Co) activity estimated with the alternative method agrees ac-
ceptably with the measured activity, and 58Co activity is slightly
overestimated. Cobalt-57 (57Co) is the activation product of con-
cern in the long term. The vacuum and window foils will be
exempted in 3.3 y and 5.5 y, respectively, after replacement. Cal-
culated effective dose with MCNPX and ICRP reference HML
phantoms in the foils replacement operation is 0.34 mSv, and
annual effective dose would be 2.1 mSv, which is below the an-
nual limits.
Health Phys. 107(2):103–110; 2014
Key words: cyclotron; dose assessment; Monte Carlo; protons
INTRODUCTION
THE CENTRE of Molecular Imaging of the Universidad de
Malaga (CIMES) has a PETtrace cyclotron for radionuclide
*Departamento de Física Aplicada I, Facultad de Ciencias, Boulevar
Louis Pasteur s/n, Universidad de Malaga, 29010, Malaga, Spain;
†Departamento de Radiología y Medicina Física, Facultad de Medicina,
Boulevar Louis Pasteur 32, Universidad de Malaga, 29010, Malaga, Spain.
The authors declare no conflicts of interest.
For correspondence contact: J. Javier Martínez Serrano, Departamento de
Física Aplicada, Facultad de Ciencias, Boulevar Louis Pasteur s/n, Universidad
de Malaga, 29010, Malaga, Spain, or email at javiermserrano@uma.es.
(Manuscript accepted 29 October 2013)
0017-9078/14/0
Copyright © 2014 Health Physics Society
DOI: 10.1097/HP.0000000000000064
www.health-physics.com
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
production. Maximum energy of the proton beam is
16.5 MeV, and proton beam intensity is 40 μA. The cyclo-
tron produces mainly 18F, which is produced in a silver
cell in the target (known as the silver body) containing
water enriched 98% in 18O. In its pass through the target,
the proton beam encounters the vacuum foil (a disk of
26 mm diameter and 25 μm thick), the refrigeration cham-
ber (a cylindrical cell of 2 cm length and 15 mm diameter
filled with He at a pressure of 3 atm and 20°C), and a sec-
ond foil (window foil) of 28 mm diameter and 50 μm
thick, finally reaching the silver body (cell cavity has ap-
proximate dimensions of 23
15
8 mm). The vacuum
foil, also known as the isolation foil, isolates the target
from the cyclotron acceleration chamber. The window
foil seals the silver body cell. A scheme of the 18F produc-
tion target is shown in Fig. 1. The foils are made of Havar,
a metal alloy of high cobalt content (elemental composi-
tion shown in Table 1). The foils are highly activated by
the cyclotron proton beam and to a lesser extent are acti-
vated by the secondary neutrons generated in the reaction:
H2 18 O þ p → H2 18 F þ n: (1)
The PET cyclotron at CIMES runs at double beam
(80 μA), producing mainly 18F in two targets. The 18F pro-
duction targets are irradiated approximately 4 h daily from
Monday to Friday. Following the manufacturer’s recom-
mendations, the foils are changed when a workload of
5,000 μA h−1 is reached on each of them. This corresponds
to a total irradiation time of 125 h. Considering the cyclo-
tron beam use, the Havar foils of the 18F production targets
are replaced every 2 mo.
The aims of this research are to assess the activity in-
duced by the proton beam and secondary neutrons in the
Havar foils of the 18F production targets and to assess
the radiological risk associated with the induced activity
in the foils in terms of dose.
Several authors have studied the activation of the
Havar foils. O’Donnell et al. (2004) and Manickam et al.
(2009) measured the activity induced in the Havar foils of
103
104 Health Physics
Fig. 1. Drawing of the target showing the silver body cell (1), the window foil (2), and the vacuum foil (3).
a PETtrace cyclotron. Mochizuki et al. (2006) also mea-
sured the activity in the Havar foils of a 16.5 MeV,
20 μA PET cyclotron. On the other hand, Calandrino et al.
(2006) calculated the activity induced in the compo-
nents of two cyclotrons, a 11 MeV Siemens-CTI RD
Eclipse (Siemens Medical Solutions USA, Inc. 51 Valley
Steram Parkway. Malvern, PA 19355 US) and a 18 MeV
IBA Cyclone 18/9 (IBA-Radiopharma Solutions. Chemin
du Cyclotron, 3. 1348 Louvain-la Neuve, Belgium), in-
cluding the Havar foils. The standard way to estimate
induced activities in a material subjected to neutron
and proton fluxes is to use an activation code [ORIGEN
(Croff 1983), ACAB (Sanz et al. 2008)]. A novel method
is proposed in this paper to predict induced activities in
the Havar foils of a PET cyclotron that makes no use of
an activation code. The activation products concentration
balance equation is solved analytically for the specific ir-
radiation conditions of the targets.
METHODS
Activity induced in the foils
The main activation products present in the foils, nu-
clear reactions originating them, and half-lives are shown
Table 1. Elemental composition of Havar alloy.
Element Atomic fraction (%)
Be 0.3
C 1.0
Cr 22.2
Mn 1.7
Fe 18.1
Co 41.6
Ni 12.8
Mo 1.4
W 0.9
www.health-physics.com
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
August 2014, Volume 107, Number 2
in Table 2. Two different approaches were followed to
assess induced activity in the Havar foils. In a first approach,
proton- and neutron-induced activities were assessed with
the activation code ACAB (Sanz et al. 2008). EAF2007
(Forrest 2007; Forrest et al. 2007) activation libraries
were used in the calculations. ACAB can be used to per-
form space-dependent inventory calculations, allowing
for a very flexible geometry and neutron flux description.
The code solves the general nuclear transmutation chains
for multidimensional neutron flux distributions. ACAB
needs as input proton and neutron fluences as a function
of energy. Neutron and proton fluences for the ACAB in-
put were obtained with the Monte Carlo radiation trans-
port code MCNPX v2.6.0 (MCNPX 2008). As a second
approach, the next method was followed (Martínez-
Serrano and Díez de los Ríos 2010): atomic concentrations
of activation products per proton (Cp) and per neutron (Cn)
were assessed with MCNPX using the F4 and FM cards
(MCNPX 2008). The F4 is a tally of average fluence in a
volume. The FM card is used to calculate atomic concentra-
tion of activation products from average proton/neutron
fluence and activation cross sections. The standard distribu-
tion of MCNPX does not include proton activation libraries.
TENDL‐2010 (Koning and Rochman 2010) proton activa-
tion libraries were used.
Production rates due to proton and neutron activation
were calculated using the following equations:
I0
γ0;p ¼ C p
(2)
e
I0
γ0;n ¼ C n

N; (3)
e
where I0 is the proton beam intensity, e is electron charge,
and N is the average number of secondary neutrons gener-
ated by a proton of the beam impacting on the silver body.
 PET and radiological risk c J. J. MARTINEZ-SERRANO AND A. DIEZ DE LOS RIOS 105

Table 2. Main radionuclides present in irradiated Havar foils and easily solved as the addends are terms of geometric suc-
nuclear reactions generating them. cessions, resulting for wp,n(t):
Radionuclide Nuclear reaction T1/2 (d)
58 58
Ni (n,p) 58Co γ0;p;n e5λ −1  λΔt  
Co wp;n ðt Þ ¼ e −1 1−e−λt ; (9)
59
Co (n,2n) 58Co λ ðe −1Þðe −1Þ
7λ λ
59
Co (p,pn) 58Co 70.86
58
Fe (p,n) 58Co where t is the time period (in d) from start to end of irradia-
59
Fe 59
Co (n,p) 59Fe tion, Δt is the time (in d) that the target is irradiated (in d−1)
58
Fe (n,γ) 59Fe 44.495 at present irradiation conditions, and λ is the decay constant
54
Mn 54
Fe (n, γ) 54Mn of the induced radionuclide (in d−1 units).
55
Mn (n, 2n) 54Mn 312.12 Proton- and neutron-induced activity concentration
54
Cr (p,n) 54Mn [Ap(t) and An(t)] are then given by:
57 58
Co Ni (n,2n) 57Co  λΔt  −λt 
γ0;p;n e5λ −1
57
Fe (p,n) 57Co 271.74 Ap;n ðtÞ ≡ λwp;n =ρ ¼ e −1 1−e ; ð10Þ
ρ ðe −1Þðeλ −1Þ
7λ
60 59
Co Co (n,γ) 60Co
60
Ni (n,p) 60Co 1925.28 where ρ is the density of Havar foils.
60
Fe (p,n) 60Co Radiological risk derived from Havar foils activation
56 56
Co Fe (p,n) 56Co 77.233 An important issue from the radiation protection
52 52
Mn Cr (p,n) 52Mn 5.591 point of view is to estimate the radiological risk associated
with Havar foils activation. Two locations in the cyclotron
room have been considered to estimate radiation dose. The
As the cyclotron does not run continuously, production operation in the cyclotron room that involves the highest ex-
rates are functions of time: posure to Havar foils is their replacement. Room location
T is 10 cm from one of the 18F production targets (the one
I ðt Þ closest to the room floor), in the forward direction of the
γp ðt Þ ¼ C p
(4)
e beam. Room location S is 10 cm to the left of the shielding
wall. This location is frequently occupied by cyclotron per-
sonnel in maintenance operations. Both room positions
I ðt Þ
γn ðt Þ ¼ C n

N; (5) are shown in Fig. 2. First, the ambient equivalent dose‡ rate
e [H˙ * (10)], an estimator of effective dose, was calculated at
where γp(t) and γn(t) are periodic pulse functions de- both room locations. For a given activation product, the am-
fined by: bient equivalent dose per photon [H*p(10)] was obtained by
numerical simulation with MCNPX in a 5 cm
5 cm
5 cm

f
γ 0; p;n 4 h per day ðirradiationÞ cubic cell centered at both locations using a tally of aver-
Monday to Friday age photon fluence in a volume (tally F4) and converting
γp;n ðtÞ ¼ 0 20 h per day ðcooling Þ
to ambient equivalent dose with ICRP 74 (1997) fluence
Saturday and Sunday ðcooling Þ 0 ð6Þ
to dose conversion factors implemented in a DF card
(MCNPX 2008).
The volume concentrations of neutron- and proton- H˙ *(10) may differ significantly from effective dose
induced activation products (wp and wn respectively) at rate. A precise estimation of effective dose requires the use
time t follow this differential equation: of a voxel phantom. Equivalent organ dose rate estimations
were carried out with the ICRP reference HML§ (male and
dwp;n ðt Þ female) voxel phantoms (Capello et al. 2012) at room
¼ γ p;n ðtÞ−λwp;n ðt Þ: (7)
dt positions T and S. The male HML phantom is contained in
a hexahedral lattice with dimensions 54.3 cm
27.1 cm

The solution to this equation is: 177.6 cm (voxel dimension is 2.14 mm
2.14 mm

Z  8 mm). The female HML phantom fills a hexahedral lattice
t
wp;n ðt Þ ¼ γp;n ðuÞe du e−λt :
λu
(8)
0 ‡
Ambient dose equivalent H*(10) at the point of interest in the ac-
tual radiation field is the dose equivalent that would be generated in
the associated oriented and expanded radiation field at a depth of 10
The integral in brackets can be solved analytically, as mm on the radius of the ICRU sphere, which is oriented opposite to
the direction of incident radiation.
γp,n(u) is a pulse function. By reordering terms, it can be §
Human Monitoring Laboratory, National Internal Radiation Assess-
expressed as the product of several sums. These sums are ment Section, Radiation Protection Bureau, Ottawa, Ontario, Canada.
www.health-physics.com

Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
106 Health Physics August 2014, Volume 107, Number 2

Fig. 2. Scheme of the cyclotron room showing positions T and S. Length dimensions in m.

with size 53.1 cm
24.3 cm
168.4 cm and voxel dimen-
sion 1.77 mm
1.77 mm
4.84 mm (in the direction of X,
Y, and Z axes). Following the recommendations of ICRP
103 (2007), the effective dose at both locations was esti-
mated by averaging effective doses obtained with the male
and female voxel phantoms. The lattice containing the
phantoms is arranged symmetrically in the X-direction rel-
ative to the target. Distance from the isolation foil to the
phantom lattice is 13.8 cm.
RESULTS AND DISCUSSION
Induced activity in the Havar foils of the 18F
production targets
Havar foils of the 18F production target reach their
maximum activity at the end of their use period, after
125 h of irradiation (2 mo after their replacement). The
activity induced by the direct proton beam and by second-
ary neutrons was estimated at the moment of foils re-
placement with the activation code ACAB (Sanz et al.
2008). Activity was also estimated applying eqn (10)
and the induced radionuclide production rates obtained
with MCNPX. The results are shown in Table 3 for the ac-
tivation products in the foils with major contributions to
activity. At the moment of their replacement, the foil
activity is mostly due to 58Co, 56Co, 52Mn, and 57Co. Ac-
tivities predicted using both procedures for the vacuum
foil present large differences. The best agreement is found
for the radionuclide that gives the most important contri-
bution to foils activity, 58Co. For the window foil, the two
procedures match in the order of magnitude of 60Co and

Table 3. Induced activity (in Bq) in Havar foils of 18F production targets estimated by activation code ACAB and by
MCNPX and eqn (10) (second and fourth columns).
Vacuum foil Window foil
Radionuclide ACAB Bq MCNPX Bq ACAB Bq MCNPX Bq
60 4 3 4
Co (2.29 ± 0.08)
10 (1.9 ± 0.2)
10 (5.4 ± 0.2)
10 (3.5 ± 0.4)
104
56
Co (1.22 ± 0.12)
108 (2.4 ± 0.3)
107 (2.4 ± 0.2)
108 (3.9 ± 0.5)
107
57
Co (8.7 ± 0.5)
107 (2.3 ± 0.3)
105 (1.74 ± 0.10)
108 (3.9 ± 0.5)
105
58
Co (2.58 ± 0.14)
108 (4.1 ± 0.5)
108 (5.2 ± 0.3)
108 (6.6 ± 0.8)
108
52
Mn (1.2 ± 0.7)
108 (3.7 ± 0.5)
107 (2.35 ± 0.14)
108 (6.0 ± 0.7)
107
54
Mn (1.92 ± 0.10)
107 (1.6 ± 0.2)
105 (3.8 ± 0.2)
107 (2.6 ± 0.3)
105
59
Fe (2.56 ± 0.09)
104 (1.5 ± 0.2)
103 (5.13 ± 0.16)
104 (2.3 ± 0.3)
104
Total activity (6.0 ± 0.4)
108 (4.7 ± 0.6)
108 (1.21 ± 0.08)
109 (7.6 ± 0.9)
108

www.health-physics.com

Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
 PET and radiological risk c J. J. MARTINEZ-SERRANO AND A. DIEZ DE LOS RIOS 107

Table 4. Measured and estimated induced activity concentrations (in Bq g−1) in a vacuum foil irradiated for about 125 h in
a 2-mo period.

measured ð%Þ measured ð%Þ

ACAB MCNPX
Measured ACAB MCNPX
56
Co (1.26 ± 0.06)
109 (1.11 ± 0.09)
109 0.88 (2.2 ± 0.3)
108 0.17
57
Co (1.97 ± 0.20)
108 (7.9 ± 0.5)
108 5.93 (2.1 ± 0.3)
106 0.011
58
Co (2.33 ± 0.11)
109 (2.34 ± 0.13)
109 1.004 (3.7 ± 0.5)
109 1.61
60
Co (1.45 ± 0.06)
104 (2.08 ± 0.09)
105 14.34 (1.7 ± 0.2)
104 1.17
54
Mn (3.34 ± 0.17)
107 (1.75 ± 0.09)
108 5.24 (1.4 ± 0.2)
106 0.04

59
Fe, and there is a relatively good agreement for 58Co. activity for 60Co, a factor 14.3 larger than measured activity.
There is a large discrepancy in the estimated activities for In general, ACAB overestimates the activity in the foil with
54
Mn and 57Co using the two activity prediction procedures the exception of 56Co. The alternative procedure gives an
in both vacuum and window foils. acceptable estimation of 60Co as compared with the mea-
In order to understand this discrepancy, production rates sured value, only a factor 1.17 greater than the measured ac-
were calculated using MCNPX with the same proton/neutron tivity and a slight overestimate of 58Co activity. For the rest
activation cross section libraries as ACAB (EAF2007). of the activation products, the procedure based on eqn (10)
The resulting activities obtained were not much closer underestimates activity, making the poorest activity predic-
to ACAB-predicted values. Therefore, to contrast the two tion for 57Co, a factor of 0.01 lower than measured activity.
predicted activities, a vacuum foil sample, stored for 1,177 d
after its replacement, was analyzed by gamma spectrom- Dose due to Havar foils activation
etry. The foil sample was measured with a Canberra Ambient equivalent dose rates were estimated in room
HPGe GR‐6022 reverse-electrode detector (Canberra In- positions T and S at the moment of their replacement, when
dustries, Inc., 800 Research Pkwy., Meriden, CT, USA). Havar foils reach maximum activity. For this calculation,
The detector has a relative efficiency of 60% and a resolu- measured activity values of 60Co, 58Co, 56Co, 57Co and
54
tion of 2.2 keV for the 1,332.5 keV 60Co γ-rays. The HPGe Mn were used for dose estimation due to the vacuum foil.
is housed in a 10‐cm-thick high-purity lead shield. The elec- For the dose contribution of the window foil, activity
tronic setup consisted of a spectroscopy amplifier module concentrations estimated with ACAB were used except for
60
(Canberra 2026) with a shaping time constant of 4 μs and Co, which used the activity predicted with eqn (10). The
an ultra-fast and fixed dead time Analog-to-Digital Con- predicted results are shown in Table 5. Estimated H˙ * (10)
verter (Canberra 8715 ADC) suitable for high count-rate at 10 cm from the 18F production target is 8.7 ± 0.8 mSv
nuclear spectroscopy. The electronic circuitry allowed h−1. At the position most often occupied in the cyclotron
quantitative gamma analysis with minimal dependence room, the estimated ambient equivalent dose rate is 38 ±
on system count rate due to a Pile Up Rejector/Live 4 μSv h−1. The main contribution to dose at position T,
Time Corrector (PUR/LTC), minimizing the spectral near the target, is due to 58Co, followed by 52Mn and 56Co,
distortion caused by pile up at high count rates. The foil which have a similar contribution to dose. At room location
sample was measured at 10 cm from the detector window, S, 56Co gives the highest contribution to dose, followed by
52
reducing the pile up and random coincidences during the Mn and 58Co. In this case, gamma radiation emitted
measurement. Additionally, the pile up corrections were by 58Co is more attenuated by the shielding wall than that
obtained according to the Wyttenbach formula. Finally, emitted by 52Mn and 56Co, and its contribution to dose is
the coincidence-summing corrections for each radionu-
Table 5. Predicted ambient equivalent dose rate simulation results
clide emission were calculated using GEANT4 code in cyclotron room locations S and T due to 18F target foil activation.
(Agostinelli et al. 2003). Table 4 shows the measured ac-
Radionucleid T μSv/h S μSv/h
tivity concentration in the Havar foil sample (decay corrected
to the foils replacement date). In the second and fourth
60
Co 0.38 ± 0.05 (2.8 ± 0.3)
10−3
56
columns, estimated activity concentrations for the vacuum Co 1896 ± 160 20.1 ± 1.9
foil were included to compare easily with the measurements.
57
Co 587 ± 60 (2.6 ± 0.3)
10−2
58
Co 3486 ± 420 6.0 ± 0.7
No 52Mn and 59Fe activity levels were detected, as these 52
Mn 2151 ± 130 11.6 ± 1.0
radionuclides have very short half-lives. 54
Mn 603 ± 30 0.29 ± 0.02
By inspection of Table 4, it can be inferred that ACAB 59
Fe 0.76 ± 0.03 (4.2 ± 0.2)
10−3
results in good estimates for the main contributors to the foil
Total 8725 ± 800 38 ± 4
activity, 58Co and 56Co. ACAB gives the poorest predicted
www.health-physics.com

Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
108 Health Physics August 2014, Volume 107, Number 2

Table 6. Equivalent organ dose rates and effective dose rates (in μSv h−1) in cyclotron room location T due to Havar
foil activation.
56 58 52 54 57
Organ Co Co Mn Mn Co Total

marrow 409 ± 25 430 ± 35 216 ± 13 16.2 ± 0.9 2.29 ± 0.14 1074 ± 75

adrenals 498 ± 31 624 ± 50 282 ± 18 22.6 ± 1.2 3.06 ± 0.19 1429 ± 100
brain 271 ± 16 236 ± 19 151 ± 9 10.5 ± 0.5 1.48 ± 0.09 669 ± 46
colon 888 ± 54 1312 ± 101 511 ± 31 42.8 ± 2.3 6.8 ± 0.4 2761 ± 189
pharynx 618 ± 38 626 ± 52 322 ± 20 23.6 ± 1.2 4.6 ± 0.3 1595 ± 112
trachea 806 ± 49 1052 ± 83 452 ± 28 36.8 ± 2.0 5.9 ± 0.4 2352 ± 162
gall bladder 678 ± 41 775 ± 63 354 ± 22 27.3 ± 1.4 5.2 ± 0.3 1839 ± 128
heart 534 ± 32 708 ± 55 304 ± 19 24.7 ± 1.3 3.5 ± 0.2 1575 ± 108
kidney 734 ± 44 922 ± 73 405 ± 25 32.5 ± 1.7 5.5 ± 0.3 2099 ± 144
liver 550 ± 33 605 ± 49 281 ± 17 21.4 ± 1.1 3.9 ± 0.2 1461 ± 101
lungs 400 ± 24 519 ± 40 224 ± 14 17.8 ± 0.9 2.8 ± 0.2 1164 ± 79
muscle 534 ± 33 554 ± 45 280 ± 17 21.1 ± 1.1 3.7 ± 0.2 1393 ± 97
esophagus 865 ± 53 1196 ± 94 495 ± 31 40.5 ± 2.2 6.6 ± 0.4 2603 ± 179
pancreas 509 ± 32 582 ± 41 251 ± 16 22.1 ± 1.1 3.5 ± 0.2 1368 ± 90
prostate 308 ± 19 280 ± 23 172 ± 11 12.2 ± 0.6 3.7 ± 0.2 776 ± 53
skull 504 ± 31 641 ± 50 285 ± 18 22.8 ± 1.2 5.7 ± 0.3 1459 ± 100
spine 533 ± 32 608 ± 49 289 ± 18 22.3 ± 1.2 6.1 ± 0.4 1458 ± 100
thorax-sternum 585 ± 35 857 ± 65 349 ± 21 29.2 ± 1.6 7.5 ± 0.4 1827 ± 124
pelvis 419 ± 25 541 ± 42 237 ± 15 18.6 ± 1.0 3.0 ± 0.2 1219 ± 83
skin 877 ± 53 1276 ± 97 512 ± 32 42.7 ± 2.3 6.7 ± 0.4 2715 ± 185
small intestine 420 ± 26 491 ± 39 226 ± 14 17.9 ± 1.0 2.6 ± 0.2 1158 ± 80
spleen 828 ± 50 1068 ± 84 452 ± 28 36.6 ± 1.9 6.3 ± 0.4 2392 ± 165
stomach 537 ± 33 630 ± 44 271 ± 17 23.0 ± 1.1 4.4 ± 0.3 1466 ± 96
testes 715 ± 44 728 ± 61 369 ± 23 27.3 ± 1.4 5.4 ± 0.3 1844 ± 129
thyroid 667 ± 41 679 ± 57 354 ± 22 25.8 ± 1.4 5.2 ± 0.3 1731 ± 122
bladder 766 ± 47 1169 ± 88 469 ± 29 39.3 ± 2.1 6.2 ± 0.4 2449 ± 166
breast 775 ± 48 1054 ± 85 335 ± 21 26.6 1.4 5.6 0.3 2196 155
ovaries 516 ± 31 916 ± 73 354 ± 22 28.3 1.6 3.6 0.2 1818 128
uterus 1105 ± 67 1299 ± 103 845 ± 52 63.3 3.4 7.9 0.5 3320 226
Ė (ICRP-103) 650 ± 40 840 ± 66 350 ± 22 28.1 ± 1.5 4.8 ± 0.3 1880 ± 130
Ė (ICRP-60) 640 ± 39 830 ± 65 350 ± 22 28.2 ± 1.5 4.9 ± 0.3 1860 ± 130

lower than at position T. Macroscopic cross sections of
photons emitted by 58Co in the shielding material (Pb) are
larger than those of 52Mn and 56Co.**
The ambient equivalent dose may differ substantially
from effective dose. Thus, organ equivalent doses were
calculated at room locations T and S with the ICRP refer-
ence HML male and female voxel phantoms. Estimations
of organ equivalent dose were carried out for the activa-
tion products in the Havar foils with a significant contri-
bution to dose. Equivalent organ doses at room positions
T and S are shown in Table 6 and Table 7, respectively.
During foil replacement, the dose at this location is mainly
due to 58Co, 56Co, and 52Mn.
**Linear attenuation coefficient μ was estimated by MC simula-
tion with MCNPX for radiation emitted by 58Co, 56Co, and 52Mn.
For photons emitted by 58Co, estimated effective μ is 1.02 ± 0.2 cm−1. Cal-
culated effective μ for 56Co and 52Mn are 0.603 ± 0.010 cm−1 and
0.723 ± 0.010 cm−1, respectively.
www.health-physics.com
The effective dose rate at positions T and S was calcu-
lated with ICRP 103 (ICRP 2007) and ICRP 60 (ICRP
1990) tissue weighting factors. The estimated effective
dose rate close to the 18F production target using ICRP
103 tissue weighting factors is 1.88 ± 0.13 mSv h−1 and
1.86 ± 0.13 mSv h−1 using the older ICRP 60 tissue
weighting factors. Assuming that the Havar foils replacement
operation takes about 5 min and their transport for storage
1 min (O’Donnell et al., 2004), the estimated effective dose
in the foils replacement operation of one target is 0.19 mSv.
The estimated effective dose in the replacement and transport
of Havar foils of the two 18F production targets is 0.34 mSv
and annual effective dose 2.1 mSv.
The effective dose rate at room position S is 17.1 ± 1.2
μSv h−1 (ICRP 103 tissue weighting factors) and 16.9 ± 1.2
μSv h−1 if ICRP 60 tissue weighting factors are used. The
ambient equivalent dose overestimates effective dose by
a factor of 4.6 at room position T and by a factor of 2.2
at room position S.

Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
 PET and radiological risk c J. J. MARTINEZ-SERRANO AND A. DIEZ
CONCLUSION
Activity induced in the Havar foils of the 18F produc-
tion target of a GE PETtrace cyclotron has been deter-
mined by two different procedures, one consisting in
using the activation code ACAB with the proton and neu-
tron fluxes estimated with the radiation transport program
MCNPX and the other based in solving the balance differ-
ential equation for each induced radionuclide production
using production rates estimated with MCNPX. The two
procedures gave different results. To validate the foil ac-
tivity predictions, an irradiated vacuum foil was measured
by gamma spectrometry. ACAB activity estimations for
the activation products (56Co and 58Co) that give the most
important contribution to activity in the foil matched well
with the measured values. The alternative procedure using
eqn (10) gives a good estimation of 58Co activity of the
vacuum foil and an acceptable estimation of 60Co (thus
this procedure complements the poorest prediction of ACAB
for this radionuclide activity), underestimating the activity
for the rest of the activation products in the foils.
During foil replacement, activity in the foils is due
to 58Co and 56Co. In the long term, the foils activity is
due to 57Co; therefore, the activity level of this radionuclide
determines Havar foils waste management. In 3.3 y, 57Co
activity in vacuum foil decays below the exemption limit††
(1 MBq), and this foil can be handled as conventional mate-
rial. After 5.5 y of storage, the window foil will be exempted.
Ambient equivalent dose rates and organ equivalent
dose rates with the arms-down Zubal phantoms, calculated
by numerical simulation close to the 18F production target
and at the point of higher occupation frequency, showed that
ambient equivalent dose overestimates effective dose. The
maintenance operation inside the cyclotron room that in-
volves the highest radiological risk is the foil replacement.
The estimated annual effective dose, 2.1 mSv, does not sur-
pass the annual limits for radiation-exposed professionals.
Acknowledgments—The authors want to thank Rafael Larrosa (SCBI,
Universidad de Malaga, Spain) for MPI installation and setup of MCNPX
on the PICASSO cluster at SCBI and Antonio M. Roldan (CIEMAT,
Madrid, Spain) for allowing us to run MCNPX in the EULER cluster at
CIEMAT. We thank J.A. Ruiz (CIMES, Málaga, Spain) and Clemente Ruiz
(GE Healthcare, Madrid, Spain) for providing useful information on the
PETtrace cyclotron and 18F production targets. We thank Kevin Capello
(HML, Ottawa Canada) for providing the ICRP reference HML male and fe-
male voxel phantoms. We acknowledge Santiago Hurtado (Universidad de
Sevilla, Spain) for foil activity analysis and Daniel Lopez (UNED, Madrid,
Spain) for support with ACAB inputs.
††
Exemption limits of radioactive material adopted by Spanish leg-
islation are those published in Council Directive 96/29/EURATOM
(Council Directive 1996).
www.health-physics.com
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
DE LOS RIOS 109
REFERENCES
Agostinelli S, Allison J, Amako K, Apostolakis J, Araujo H,
Arce P, Asai M, Axen D, Banerjee S, Barrand G, Behner F,
Bellagamba L, Boudreau J, Broglia L, Brunengo A, Burkhardt
H, Chauvie S, Chuma J, Chytracek R, Cooperman G, Cosmo
G, Degtyarenko P, Dell’Acqua A, Depaola G, Dietrich D,
Enami R, Feliciello A, Ferguson C, Fesefeldt H, Folger G,
Foppiano F, Forti A, Garelli S, Giani S, Giannitrapani R, Gibin
D, Gómez Cadenas J J, González I, Gracia Abril G, Greeniaus
G, Greiner, Grichine V, Grossheim A, Guatelli S, Gumplinger
P, Hamatsu R, Hashimoto K, Hasui H, Heikkinen A, Howard
A, Ivanchenko V, Johnson A, Jones F, Kallenbach J, Kanaya
N, Kawabata M, Kawabata Y, Kawaguti M, Kelner S, Kent P,
Kimura A, Kodama T, Kokoulin R, Kossov M, Kurashige H,
Lamanna E, Lampén T, Lara V, Lefebure V, Lei F., Liendl M,
Lockman W, Longo F, Magni S, Maire M, Medernach E,
Minamimoto K, Mora de Freitas P, Morita Y, Murakami K,
Nagamatu M, Nartallo R, Nieminen P, Nishimura T, Ohtsubo
K, Okamura M, O’Neale S, Oohata Y, Paech K, Perl J, Pfeiffer
A, Pia MG, Ranjard F, Rybin A, Sadilov S, Di Salvo E, Santin
G, Sasaki T, Savvas N, Sawada Y, Scherer S, Sei S, Sirotenko V,
Smith D, Starkov N, Stoecker H, Sulkimo J, Takahata M,
Tanaka S, Tcherniaev E, Safai Tehrani E, Tropeano M, Truscott
P, Uno H, Urban L, Urban P, Verderi M, Walkden A, Wander W,
Weber H, Wellisch JP, Wenaus T, Williams DC, Wright D,
Yamada T, Yoshida H, Zschiesche D. GEANT4: a simulation
toolkit. Nucl Instr Methods Phys Research A 506; 2003.
Calandrino R, del Vecchio A, Savi A, Todde S, Griffoni V,
Brambilla S, Parisi R, Simone G, Fazio F. Decommissioning
procedures for an 11 MeV self-shielded medical cyclotron af-
ter 16 years of working time. Health Phys 90:588–596; 2006.
Capello K, Kedzior S, Kramer GH. Voxel phantoms: the new ICRP
computational phantoms: how do they compare? Health Phys
102:626–630; 2012
Council Directive 96/29/EURATOM. Laying down basic safety
standards for the protection of the health of workers and the
general public against the dangers arising from ionizing radi-
ation. Brussels, Belgium: EURATOM; Official Journal of
the European Communities; 1996.
Croff AG. ORIGEN 2: a versatile computer code for calculating
the nuclide characteristics of nuclear material. 1983.
Forrest RA. The European Activation File: EAF‐2007 deuteron
and proton induced cross section library. EASY documenta-
tion series. Abingdon, Oxfordshire, UK: EURATOM/UKAEA
Fusion Association; UKAEA FUS 536; 2007.
Forrest RA, Kopecky J, Sublet JCh. The European Activa-
tion File: EAF‐2007 neutron-induced cross section library.
EASY documentation series. Abingdon, Oxfordshire, UK:
EURATOM/UKAEA Fusion Association; UKAEA FUS 535
2007; 2007.
ICRP. The 2007 recommendations of the International Commis-
sion on Radiological Protection. Publication 103, Annals of
the ICRP, Volume 37; 2007.
ICRP. Recommendations of the ICRP. Publication 60, Annals of
the ICRP 21 No. 1–3; 1990.
ICRP. Conversion coefficients for use in radiological protection
against external radiation. Publication 74, Annals of the ICRP
Volume 26/3; 1997.
Koning AJ, Rochman D. TENDL‐2010: TALYS-based evalu-
ated nuclear data library. Petten, The Netherlands: Nuclear
Research and Consultancy Group; 2010.
Manickam V, Brey RR, Jenkins PA, Christian PE. Measurements
of activation products associated with Havar Foils from a
GE PETtrace Medical Cyclotrón using high resolution gamma
spectroscopy. Radiat Safety J 96(Suppl 1):S37–S42; 2009.
110 Health Physics August 2014, Volume 107, Number 2

Martínez-Serrano JJ, Díez de los Ríos A. Prediction of neutron
induced radioactivity in the concrete walls of a PET cyclo-
tron room with MCNPX. Med Phys 37:6015–6021; 2010.
MCNPX v 2.6.0 user’s manual. Pelowitz DB, ed. Los Alamos,
NM: Los Alamos National Laboratory; LANL Report No.
LA-CP‐07‐1473; 2008.
Mochizuki S, Ogata Y, Hatano K, Abe J, Ito K, Ito Y, Nishino
M, Miyahara H, Ishigure N. Measurements of the in-
duced radionuclides in production of radiopharmaceuticals
for positron emission tomography (PET). J Nucl Science
43:348–353; 2006.
O’Donnell RG, León Vintró L, Duffy GJ, Mitchell PI. Measure-
ment of the residual radioactivity induced in the front foil of
a target assembly in a modern medical cyclotron. Applied
Radiat Isotopes 60:539–542; 2004.
Sanz J, Cabellos O, Herranz NG. ACAB. Inventory code for nu-
clear applications: user’s manual V. Madrid, Spain; NEA; 2008.
■■

www.health-physics.com

View publication stats Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.