WTTC9 Proceedings 2002 Comor COSTIS Solid Target p59

Proceedings of the Ninth International
Workshop on Targetry and Target Chemistry

Turku, Finland, May 23-25, 2002
Edited by Sven-Johan Heselius, Jan-Olof Lill and Johan Rajander
Turku PET Centre, Accelerator Laboratory, Åbo Akademi University,

Porthansg. 3, FIN-20500 Turku, Finland
Proceedings of the Ninth International
Workshop on Targetry and Target Chemistry
Report ÅA Acc-1 (2004)

Turku, June 2004
Edited by Sven-Johan Heselius, Jan-Olof Lill and Johan Rajander
The authors are responsible for the contents of the contributions submitted to these
Proceedings. The contributions have been only slightly edited.
ISBN 952-12-1383-3
Printed by Gillot Oy Painotalo-Turku, Finland, 2004
Proceedings of the Ninth International Workshop on
Targetry and Target Chemistry
Workshop organized by the Turku PET Centre
Organizing Committee International Advisory Committee

Sven-Johan Heselius, Chairman John Clark, Cambridge, UK
Jörgen Bergman Sven-Johan Heselius, Turku, Finland
Kjell Någren Didier Le Bars, Lyon, France
Hannu Sipilä Jeanne Link, Seattle, USA
Olof Solin Thomas J. Ruth, Vancouver, Canada
Mika Teräs Stefan Zeisler, Vancouver, Canada
Acknowledgements
The Organizing Committee gratefully acknowledges the financial support of
St. Louis Meetings 1999
and the following companies in alphabetical order:
Amersham Health
Bioscan Europe Ltd.
Comecer Srl
CTI Cyclotron Systems
GE Medical Systems
Hidex Oy
Ion Beam Applications, S.A.
Marshall Isotopes Ltd.
Metorex International Oy
PerkinElmer Life Sciences
Rados Technology Oy
Veenstra Instruments
Von Gahlen Netherland B.V.
I
Preface
The Ninth International Workshop on Targetry and Target Chemistry (WTTC9) was held at
the Turku PET Centre, Turku, Finland, May 23-25, 2002, prior to the IX Turku PET
Symposium. The Workshop was attended by 181 scientists and engineers representing 23
countries. Besides the traditional targetry workshop programme a Training Course for
students and young scientists was introduced into the WTTC9. The members of the
International Advisory Committee stimulated us to arrange this Training Course. The
programme of the course was planned in co-operation with members of the Advisory
Committee.
The Training Course was highly appreciated by the attendees. This is reflected in the
feedback received. Especially younger scientists have informed me they learned a lot from the
course. The cyclotron suppliers CTI, EBCO, GE and IBA are acknowledged for their open
willingness to share their experiences with the targets and the radiochemistry units they
brought to the Training Course.
I extend my thanks to the members of the Organizing Committee of the WTTC9 as well as to
the members of the International Advisory Committee for excellent co-operation, and to all
who helped in one way or another with the preparations for the Workshop and in the technical
assistance during the Workshop. Special thanks go to the staff of the Congress Office of the
University of Turku for their excellent work and for creating a stimulating atmosphere for the
meeting. I also thank all the scientists and engineers who participated in the Workshop and
the Training Course for their input and enthusiasm. The input of the attendees is always the
source for the success of a meeting.
I am grateful to the session coordinators for participating in the organizing of the scientific
programme of the Workshop and for moderating the sessions, and to the lecturers of the
Training Course for well prepared and covering review lectures. The WTTC luminaries who
guided the students of the Training Course are acknowledged for their enthusiasm showed to
the students.
My co-editors, Mr. Jan-Olof Lill and Mr. Johan Rajander, deserve thanks for their time-
consuming work with these Proceedings.
Finally, I express my best wishes to Prof. Jerry Nickles and the Organizing Committee of the
WTTC10 for success with the Tenth Workshop on Targetry and Target Chemistry, which will
be arranged at the University of Wisconsin, Madison, August 13-15, 2004.
Sven-Johan Heselius
Chairman of the WTTC9
Organizing Committee
II
NINTH INTERNATIONAL WORKSHOP ON TARGETRY AND TARGET CHEMISTRY
Turku PET Centre, Turku, Finland, May 23-25, 2002
Targetry Workshop Venue: Educarium Building, University of Turku, Assistentinkatu 5
PROGRAMME
Thursday, May 23rd
Training Course in cyclotron targetry
08.30-09.15 Technology of cyclotrons for radionuclide production, Kenneth Buckley, TRIUMF.
09.15-10.00 Cyclotron targetry for radionuclide production, David Schlyer, Brookhaven National
Laboratory.
10.00-10.20 Coffee Break

15
10.20-12.20 Practical Training. IBA Cyclone 3: Cyclotron service and production of O-gases
15
and O-water. Note: This part of the Training Course takes place at the Turku PET
Centre. Coordinated by Hannu Sipilä, John Clark and Bruce Mackay.
12.20-13.20 Lunch
13.20-14.00 Oral presentations by cyclotron suppliers (CTI, EBCO, GE and IBA).
14.00-17.40 Practical Training (CTI, EBCO, GE and IBA). The trainees will be divided into 4
groups of approx. 8 persons. Each group will visit 4 venues according to the timetable
below. The group will carry out their work under the supervision of representatives of
the suppliers. The groups will also be accompanied by some of the following WTTC
luminaries: Kenneth Buckley, John Clark, Lewis Carroll, Bob Dahl, Mikael Jensen,
Jeanne Link, Didier Le Bars, Jerry Nickles, Bruce Mackay, Andy Roberts, David
Schlyer and Tom Ruth together with the local team. This part of the Training Course
takes place at the Educarium Building and is composed of:
- Hands-on training in handling and mounting of different types of targets,
- Training in cyclotron operation, and
- Presentation of chemistry modules.
Group CTI EBCO GE IBA

A 14:00-14:50 14:50-15:40 16:00-16:50 16:50-17:40
B 14:50-15:40 14:00-14:50 16:50-17:40 16:00-16:50
C 16:00-16:50 16:50-17:40 14:00-14:50 14:50-15:40
D 16:50-17:40 16:00-16:50 14:50-15:40 14:00-14:50
18.00-20.00 Welcome Reception at Educarium Building.
20.00-21.00 Evening Lecture: "Commissioning and Decommissioning of Cyclotrons"

(45 minutes plus discussions), Lewis Carroll, Carroll Ramsey Associates.
III
Friday, May 24th
08.30-08.45 Opening of the Workshop

Sven-Johan Heselius, Turku PET Centre
08.45-10.15 Water targets and high beam currents - Specific radioactivity of F-18
Coordinators:
Andrew D. Roberts, University of Wisconsin
Olof Solin, Turku PET Centre
10.45-12.15 High-power enriched gas targets

Coordinators:
Volker Bechtold, Karlsruhe
Bruce Mackay, Hammersmith Hospital
Thomas J. Ruth, TRIUMF
12.15-13.30 Lunch
13.30-15.00 Solid targets at low-energy irradiations. Transport of targets.

Coordinators:
Ron D. Finn, Sloan Kettering
Robert J. Nickles, University of Wisconsin
Michael J. Welch, Washington University
15.30-17.00 High-energy irradiations of solid and gas targets

Coordinators:
Syed M. Qaim, Forschungszentrum Jülich GmbH, Germany
Nigel R. Stevenson, Theragenics Corporation, USA
18.30-18.45 Group Photo in front of the Aboa Vetus & Ars Nova Museums by the Aura river.
19.00 - 22.00 Evening Party at the Aboa Vetus Museum, Address: Itäinen Rantakatu 4-6.
IV
Saturday, May 25th
08.30-10.00 Poster session and Coffee
10.00-11.30 Oral presentations of selected poster

Coordinators :
Rich Ferrieri, Brookhaven National Laboratory
Jeanne Link, University of Washington
11.30-12.45 Lunch
12.45-14.45 Visit to facilities (2 groups, 15 visitors/group):

- Accelerator Laboratory and Radiochemistry Laboratory, Gadolinia, Åbo Akademi
University.
- Turku PET Centre, Turku University Central Hospital.
15.15-16.30 Reports from the laboratories

Coordinators:
Didier Le Bars, University of Lyon
Steve McQuarrie, University of Alberta
16.30-17.00 Summary of the Workshop and Closing Remarks

17.00-17.10 Closing of the Workshop

Sven-Johan Heselius, Turku PET Centre
18.00-20.00 WTTC9 Farewell Party /Turku PET Symposium, Get-together Party,

Educarium Building
V
Contents
Acknowledgements I
Preface II
Targetry Workshop Programme III
Water Targets and High Beam Currents. Specific Radioactivity of F-18 1

Operational Experience with a Niobium Target Body for the Irradiation of O-18-H2O
K.R. Buckley, J. Wilson, S. Jivan, M.J. Adam, P. Piccioni, E.T. Hurtado
and T.J. Ruth 2
Production Test Results from the Pulsar Linac Targets
Robert W. Hamm and G.D. Robinson, Jr. 3
F-18-Fluoride Production in Recirculating O-18 Target
M. Kiselev and J. Armbruster 8
Production of F-18-Fluoride in the Nuclear Interface Titanium Target on a
MC-17 Cyclotron
J. Medema, P.H. Elsinga, M. Bruessermann, J. van Essen and A.M.J. Paans 9
Recirculating Low-Pressure Target for F-18 and N-13 Production
B.H. Mock and L.A. Corbin 11
Production of Multi Curie [18F]Fluoride using a Niobium Target Chamber
at Small PET Cyclotrons
F. Schmitz, M. Monclus, J. Van Naemen, D. Ekelmans, S. Goldman, R .Verbruggen,
F. Vamecq, Y. Jongen, F. Van Langevelde, P.S. Kruijer, J. Van Leuffen, R. Mooij
and P.J. Van der Jagt 12
Effect of the Beam Profile on the F-18-Fluoride Target Saturation Yield
D. Soloviev, C. Tamburella, M. Ghyoot and G.-J. Beyer 17
A Convenient Method for the Purification of O-18 Water
K. Weber, H. Marx and M. Eisenhut 18
Self-Regulating Thermosyphon Water Target for Production of F-18-Fluoride
at Proton Beam Power of One kW and Beyond
B. Wieland, C. Illan, M. Doster, A. Roberts, R. Runkle, C. Rowland and J. Bida 19
Regenerative Turbine Pump Recirculating Water Target for Producing
F-18-Fluoride Ion with Several kW Proton Beams
B. Wieland and B. Wright 21
High Power Enriched Gas Targets 23

Target Foil Grid Supports for PET Isotope Production
G.T. Bida, B.W. Wieland, J.W. Lenz and C.W. Alvord 24
Further Progress on Targetry for the Production of Ultra-High Quantities
of F-18-Fluoride
K.R. Buckley, D.W. Becker, R. Dahl, S. Jivan and T.J. Ruth 30
Development of Target System for [F-18] Fluorine Production via the
O-18(p,n)F-18 Nuclear Reaction
K.S. Chun, S.D. Yang, S.H. Ahn, S.W. Kim, M.G. Hur, B.S. Moon and S.W. Hong 31
Production of an Intense O-15 Radioactive Ion Beam using Low Energy Protons
S. Lapi, T.J. Ruth, A. Zyuzin, S. Jivan and J.M. D’Auria 32
VI
Solid Targets at Low-Energy Irradiations. Transport of Targets 33
The Electron Beam Test Facility for Cyclotron Isotope Targets Examination
A. Arzumanov, A. Borissenko, S. Ilmatov, V. Koptev and S. Lyssukhin 34
High Current Production of 11C, 13N, 15O, 17F, and 18F for PET using a 3 MV
Electrostatic Tandem Accelerator
Todd E. Barnhart, Alexander K. Converse, Kevin A. Dabbs, R.J. Nickles and
A.D. Roberts 38
COSTIS: Compact Solid Target Irradiation System
J.J. Čomor, G.-J. Beyer, U. Ketzscher, M. Brüssermann, M. Rajčević and
D. Košutić 46
A Recoil-Escape Fiber Target using 94Mo(p,n)94mTc to Produce 94TcO4-
Precursor for Radiolabeled Compounds Useful in Positron Emission Tomography
C.D. Illan and B. W. Wieland 48
Gravity Can Offer Help: Stable Production of C-10 by Bombarding a Molten Sea of
Boron Oxide with a Vertical Proton Beam
M. Jensen, G. Wolber, J. Nickles and A. Runz 58
Optimisation Studies on the Production of High-Purity I-124 and I-120g at
a Small-Sized Cyclotron
S.M. Qaim, A. Hohn, G. Blessing, Th. Bastian, K.M. El-Azoney, S. Spellerberg,
B. Scholten and H.H. Coenen 60
The Present Status of a High Current Inclined Target and Solid Target
Changer at Washington University
D.J. Rowland, R. Laforest, T.J. McCarthy, M.J. Welch, B.J. Hughey,
R.E. Klinkowstein and R.E. Shefer 62
The Production of I-121 at the Vertical Beam-Line of the MC32NI Cyclotron
at the DKFZ Heidelberg
Armin Runz, Gerd Wolber, Michael Eisenhut and Wolfhard Semmler 64
Subliming Activity from White Hot Targets. A General Method, Tested and
Rationalized by 10C Production
M.J. Schueller, R.J. Nickles, A.D. Roberts, T.E. Barnhart and M. Jensen 67
Some Aspects on Tellurium Targets for Iodine-123 Production
L.M. Solin, B.K. Kudelin and V.A. Jakovlev 71
Remote and Automated Solvent Extraction of Gallium-66
L. Tang, D.E. Reichert, D.W. McCarthy and M.J. Welch 74
Preparation of Carbon-13 Target Material from C-13-Benzene
Elizabeth M. Zippi and Amy Pilcher 78
High-energy Irradiations of Solid and Gas Targets 83

A New Method for Cu-64 Radioisotope Production Theoretical and
Experimental Results
K. Abbas, C. Birattari, M. Bonardi, F. Groppi, H.S. Mainardi, E. Menapace,
F. Simonelli, U. Holzwarth and M.F. Stroosnijder 84
Excitation Functions of Proton Induced Reactions on Enriched Rb-85 up
to 100 MeV: Yields of Sr-82 and Sr-83
S. Kastleiner, S.M. Qaim, G. Blessing, H.H. Coenen, F.M. Nortier and
T.N. van der Walt 86
Production Plans for the New 100 MeV Isotope Production Facility at Lanl
F.M. Nortier, R.C. Heaton, E.J. Peterson and D.R. Phillips 88
67
Cu Production by Proton Induced Reactions on Enriched 68Zn
R. Schwarzbach, J. Jegge, K. Zimmermann, R. Schibli and P.A. Schubiger 89
VII
Reports from the Laboratories 91
Status of Cyclotron based Radioisotopes Production in Kazakhstan.
Report from the Laboratory
A. Arzumanov, N. Berdinova, A. Borissenko, G. Chumikov, N. Gorodisskaya,
A. Knyazev, V. Koptev, Yu. Popov, G. Sychikov and D. Zheltov 92
Performance of a Flow-Through Target for the Production of [11C]CH4
T.E. Barnhart, K.R. Buckley, S. Jivan, A.D. Roberts and T.J. Ruth 98
Qualitative Evaluation of Helium-Cooling Windows
D.W. Becker, K.R. Buckley, J. Lenz, B. Hagen, P. Piccioni and T.J. Ruth 103
The Turku PET Centre Medical Compact Cyclotron CC-18/9
P.V. Bogdanov, O.A. Kovalchuk, V.A. Kislov, Yu.P. Severgin, Yu.I. Stogov,
I.N. Vasilchenko, S.-J. Heselius and S. Johansson 104
Website for Proceedings of the Workshops on Targetry and Target Chemistry
K.R. Buckley and T.J. Ruth 107
GE PETtrace MeI Microlab: Six Years' Experience with System
C.S. Dence, W. Margenau, H. Lee, C. Bognar, S.M. Moerlein, M. Al-Qahtani
and M.J. Welch 108
The Installation of a Multi-Radiopharmaceutical P.E.T. Facility at Washington University
in St.Louis, Mo, USA
G. Gaehle, S. Schwarz, M. Mueller and M.J. Welch 110
A Research Resource in Radionuclide Research
D.W. McCarthy, H.M. Bigott, D.J.R. Rowland, T.A. Perkins, T.F. Voller, R. Laforest,
D. Reichert, M.R. Lewis and M.J. Welch 111
Modified Module for the Production of C-11-Methyl Iodide from C-11-Methane,
Produced in Target
J. Medema, S.P. Tilkema, J. van Essen, H. Keizer, W. Vaalburg and P.H. Elsinga 113
A Cryo-Buncher Based on a Ten-Port Switching Valve
Jon A. Nye, David W. Dick, Micheal J. Schueller, Mikael Jensen and Jerry Nickles 115
Targetry and Chemistry for the Production of [O-14]: Application to
Radioactive Ion Beam Experiments
J.P. O’Neil, C.A. Ramsey, J. Powell and J. Cerny 118
A Versatile and Ready-to-Automate Set-Up for Research and Development of
Radiopharmaceuticals
J. Sambre and J. Courtyn 121
Ten Years Oxygen-15 Production at the Turku University Central Hospital
H.T. Sipilä and M. Teräs 123
CS-30 Cyclotron Operation at the Duke University PET Facility:
Report from the Labs
B. Wieland, C. Illan, M. Dailey, S. Murphy and R. Coleman 125
Radiochemistry and Automation 127

OPC Based Control Software
K.R. Buckley 128
Radiochemical Tl-201 Production Process
G.A. Casale, G.J. Maslat and J.L. Lago 129
Synthesis of C-11-Labelled Acamprosate for Pharmacodynamic Studies by PET
J. Courtyn and J. Sambre 131
Rapid Screening of PET Volatile Organic Infusions Generated from
2-F-18-FDG Synthesis by Solid Phase Microextraction Coupled with GC/MS
A. Kanu, M. Dixon, Y. Yin Zu, J.M. Gillies, J. Bailey, P. Thomas and J. Zweit 133
VIII
Direct PC Automation of Radiochemical Tl-201 Process
G.J. Maslat and G.A. Casale 135
Experiences with the Amercare Syringe Fill Station
Y. van der Knaap, H. Duiker, G. Wilson, J. Medema, N.H. Hendrikse and P.H. Elsinga 137
Radiopharmaceutical Production Quality Control Considerations
S.W. Schwarz, K. Lechner, M.J. Mueller, G. Gaehle and M.J. Welch 139
Synthesis of Fluorine-18 Labeled Recombinant Annexin-V Derivative,
for Identification and Quantification of Apoptotic Cells with PET
S. Zijlstra, J. Gunawan and W. Burchert 140
Radiation Sources and Radiodetectors 141

Compact, Solid-State Radiation Detectors for use in PET Isotope and
Radio-Chemistry Laboratories
L.R. Carroll and M.A. Jackson 142
Estimating the Radiation Source Term for PET Isotope Targets
L.R. Carroll 147
Potential of TlBr Semiconductor Detector for Nuclear Medicine
H.J. Sipilä, M.A. Leskelä and V.I. Kozlov 152
List of Participants 153
Author Index 172
IX
Water Targets and High Beam Currents.
Specific Radioactivity of F-18
Coordinators:
Andrew D. Roberts, University of Wisconsin
Olof Solin, Turku PET Centre
Operational Experience with a Niobium Target Body for the
Irradiation of O-18-H2O
K.R. Buckley1, J. Wilson2, S. Jivan1, M.J. Adam1, P. Piccioni1, E.T Hurtado1 and T.J. Ruth1
1
UBC/TRIUMF PET Program, 4004 Wesbrook Mall, Vancouver, BC, Canada
2
Cross Cancer Centre, University of Alberta, Edmonton, AB
Aim
To assess the performance of a water target constructed of niobium metal.
Methods
We have operated a niobium bodied water target for the production of F-18-fluoride for
approximately 1 year. During this time we have observed the operational characteristics of
this target and had occasion to compare the synthesis of F-18-FDG made from this F-18-
fluoride with that made from F-18-fluoride produced in a silver bodied water target.
Results
Typically the saturation yield for a 2-hour irradiation at 15-25 microAmperes is 130
mCi/microAmpere.
Conclusions
We will present the operational parameters, trends in yield data, and synthesis yields.
2
Production Test Results from the Pulsar Linac Targets
Robert W. Hamm1 and G. D. Robinson, Jr.2
1
AccSys Technology, Inc., 1177A Quarry Lane, Pleasanton, CA 94566, USA
2
Sandia National Laboratory, Livermore, CA 94550, USA
Introduction
Several types of linear accelerators (linacs) have been built and tested during the past
decade for the production of PET radionuclides1,2,3,4 but these systems are not yet in
general use. To achieve wide acceptance for this application, the targets for any linac must
perform reliably with radionuclide yields comparable to compact cyclotron targets. The
purpose of the present study was to demonstrate that the targets developed for use on the
AccSys PULSAR™ proton linac system5 can produce the quantities of F-18 and C-11
required by PET users currently employing compact cyclotrons. These tests were also
completed to obtain initial data for the development of linac targets to produce O-15 and
N-13. Initial results with this target at lower beam currents have been previously
presented.6
The 7 MeV proton beam from the PULSAR™ Model PL-7 linac has a pulsed peak current
variable from 5 to 15 mA at a duty factor up to 1.3 % with a repetition rate of 100 Hz.
This linac accelerates either H+ or H-, is 4.2 m long, weighs 1200 kg, and requires 25 kW
of AC power. The Model PL-7 PET radionuclide production targets are similar to
cyclotron targets, but incorporate grid supported foil windows that allow use of a thinner
window, provide more cooling and provide an expanded irradiation area7.
Target description
This patented target*, developed by Brookhaven National Laboratory in a collaboration

with AccSys, produces fluorine-18 via the 18O(p,n)18F reaction. The target body, either
silver or niobium, is held in an aluminum holder for cooling and support. A 55 %
transparent grid of copper and stainless steel supports a 25-mm diameter, 25 micron thick
titanium foil in front of a 0.3 mm thick O-18 water target volume. The volume of O-18
water in the target is 0.3 ml. The prototype target used to conduct the 18F production tests
presented in this paper is shown in figure 1.
The entrance foil and O-18 water thickness were selected to degrade the proton energy to
less than 3.5 MeV as the beam enters the back of the target. A finite element thermal
analysis was conducted for the target and the results of this analysis are shown in figure 2.
The key to the performance of this target is that the temperature of the target water is
independent of the beam size and only depends on the beam current density in a single
aperture of the target not exceeding a limit of 35 µA/cm2. Hence, the expected output of
this target has been calculated as a function of the beam irradiation diameter and is shown
in figure 3.
*U.S. Patent No. 5,915,874
3
Fig. 1. Prototype water target for production of 18F-.
C a lc u la te d O -1 8 Wa te r T a r g e t T e m p e r a tu re
300
250
Target temperature ( deg C )
Boiling point at 300 p s ig

200
150
100
50
0
0 5 10 15 20 25 30 35 40 45 50
2
Target C urrent D ensity (µA/cm )
Fig. 2. Thermal analysis results for gridded target.
2
E xp ected F -18 Yield & B eam C u rr en t @ 35 µA/cm
2.50 215
Target B eam current (µA )
2.00 172
F-18 Yield (C i/hr)
1.50 129
1.00 86
0.50 43
0.00 0
0 0.5 1 1.5 2 2.5 3
B eam diameter (cm)
Fig. 3. Calculated yield for gridded target.
4
Experimental results
The F-18 target designed for the PULSAR™ linac was irradiated under argon
overpressure of 300 psig with beam currents in the range of 7-35 µA and irradiation times
from 15 to 120 minutes. The grid-supported foil window was tested with both silver and
niobium bodied O-18 water targets in more than 50 test runs. Low enrichments of O-18
water were used to reduce the activity being handled during this testing. The data from all
runs are summarized in figure 4, normalized to the equivalent F-18 production per hour
with 95 % enriched O-18 water. The theoretical data plotted for comparison were
calculated using the saturated yields for F-18 recently published by Hess et. al8.
These tests demonstrated that the supported foil target configuration is robust enough to be
used for F-18 production with a proton beam current density up to 35 µA/cm2 at a pressure
of 300 psig. F-18 production yields varied from 60 to 95 % of theoretical, with the lowest
yields always achieved on the first run with a "fresh" target. Extended runs have shown
that the yield from the target is 1000 mCi/hr for a target current of less than 60 µA. In
addition, the yields from the niobium target were found to be statistically identical to the
yields from the silver target.
Experimental F-18 Yields vs. Target Current

1000
900
Experimental data
800
80% theoretical yield
F-18 Yield (mCi/hr)
700
600
500
400
300
200
100
5 10 15 20 25 30 35 40 45 50 55
Target Current (uA)
Fig. 4. Experimental results for gridded water target.
The C-11 gas production target, which can also be used to produce O-15 from N-15, is
made of aluminum with a 100 mm long tapered bore (25 to 38 mm) and uses the same foil
support system as the F-18 target. The 14N(p,α)11C reaction was used for carbon-11
production. The C-11 target performed as expected during 10 minute irradiations at beam
currents of 6-12 µA. Production of large quantities of C-11 were demonstrated. The
pressure rise ratio (PRR) method was used to optimize the target fill pressure during these
limited test runs. These results are summarized in figure 5. For both F-18 and C-11, target
5
yields were measured in a calibrated ionization chamber and radionuclidic purity was
estimated from the measured decay curves.
C -11 G as T arg et P relim ina ry D ata
0.70
6.25 µA
0.60
12.5 µA
7.7 µA
Pressure rise ratio
0.50
0.40
0.30
0.20
0.10
0.00
50 60 70 80 90 100 110
Initial target pressure (p sig)
Fig. 5. Experimental results for gridded gas target.
Conclusion
These experimental measurements have proven that the linac targets using a grid
supported window are robust and reliable for F-18 or C-11 production with the 7 MeV
PULSAR™ linac. Target performance was predicted through thermal analysis to be
dependent on incident beam current density, not total current, and experimental
measurements have proven this to be the case. F-18 production was routinely
demonstrated at an average of 80 % of theoretical yield over a beam current range of 7 to
35 µA. The performance of the PULSAR™ 7 using the grid supported F-18 target meets
its current production specification of 1 Ci /hour. The goal is to soon demonstrate routine
production of 2 Ci /hour of F-18 by operating the linac at a higher output current with a
larger beam spot on the target. The grid supported window has also been experimentally
proven for use with the PULSAR™ gas target. In addition, gas target performance has
been shown to not be affected by the pulsed linac beam.
The experimental results reported are being used to guide fabrication of the production
targets for the PULSAR™ to be shipped to Japan in 2003. The gas target for C-11 and O-
15 production will also be completed and tested on that system. Development of the N-13
production target for the PULSAR™ is now in the early design stages.
Acknowledgement
The authors gratefully acknowledge the assistance of Dr. David Schlyer of Brookhaven
National Laboratory and Dr. Henry Van Brocklin of Lawrence Berkeley National
6
Laboratory in the experimental measurements described in this paper. In addition, the
dedicated staff at AccSys is acknowledged for their support during these tests.
References
1. A.E. Dabiri, et. al., “A 3He++ Radio-Frequency Quadrupole Accelerator for Positron
Emission Tomography”. Nuclear Technol. 92:127-133 (1990).
2. R.E. Shefer, et. al., “Production of PET Radionuclides with a High Current
Electrostatic Accelerator”. Proc. Fourth Intl. Workshop on Targetry and Target
Chemistry, PSI, Villigen, Switzerland, September 1991, 4-10.
3. R.J. Adler and R.J. Richter-Sand, “High voltage, high power nested high voltage
accelerator,” Proc. 1991 Part. Accel. Conf., IEEE Conf. Record 91 CH 3038-7, 3201
(1991).
4. R.W. Hamm, “RF Linacs for Radioisotope Production”, Proc. Fifth Intl. Workshop on
Targetry and Target Chemistry, BNL, Upton, NY, September 1993, 12-19.
5. L. Picardi, C. Ronsivalle, S. Frullani & R. Hamm, "The TOP Project Status", Proc.
2001 Part. Accel. Conf., IEEE Conf. Record 01CH 37268, 2491 (2001).
6. G.D. Robinson, Jr. and R.W. Hamm, “Fluorine-18 Production via the 18O(p,n)18F
Reaction Using the PULSAR™ PL-7 RF Proton Linac: Results of Initial Target
Tests”, Proc. Seventh Intl. Workshop on Targetry and Target Chemistry, DKFZ,
Heidelburg, Germany, June 1997, 20-22.
7. D. Schlyer and M.L. Firouzbakht, "Correlation of Hole Size in Support Windows with
Calculated Yield Strengths", Proc. Sixth Intl. Workshop on Targetry and Target
Chemistry, TRIUMF, Vancouver, Canada, August 1995, 142-143.
8. E. Hess et. al., "Excitation function of the 18O(p,n)18F nuclear reaction from threshold
up to 30 MeV", Radiochim. Acta, 89, 357-362 (2001).
7
F-18-Fluoride Production in Recirculating O-18 Target
M. Kiselev and J. Armbruster
IBA Eastern Isotopes, Inc. PO BOX 1096 Ashburn VA 20146 , Vincent Tadino, IBA Radio-
Isotopes France 59, Boulevard Pinel, 69003 Lyon, France
Aims
To investigate the possibility of recirculating O-18 water through the cyclotron target
during irradiation which allows extraction and delivery of F-18 fluoride without
interrupting irradiation and/or emptying the target.
Methods
Approximately 10 ml of O-18 water was loaded in an apparatus containing a 5 ml storage

vessel, pump, 1.8 or 2.0 ml silver target attached to 16.5 or 18 MeV cyclotron, 10 micron
filter, backpressure regulator, conductivity meter, several valves and ion exchange
cartridges. The water was continuously pumped through the target during proton
bombardment at a rate 5 ml/min. Every 1-2 hour the water flow was directed through an
ion exchange cartridge to remove accumulated F-18 fluoride. The later was recovered by
rinsing the cartridge with bicarbonate solution and delivered to the remote hot cell. Target
material was also passed through deionising cartridges to reduce electrical conductivity.
Water pressure, conductivity and radioactivity levels in the storage vessel and ion
exchange cartridge were continuously monitored by a computer.
Results
7 hour continuous irradiation with beam current ranging from 30 to 40 µA (total integrated
beam current was 219 µAh), resulted in total of 486 GBq of F-18 delivered in 6 portions
averaging 2.2 GBq/µAh. Total of 2 ml of O-18 water was lost, presumably due to
evaporative loses and leaks, averaging over 200 GBq of F-18 per ml of consumed water.
Conductivity of target material at the end of the last production cycle was lower than at
the beginning of the first irradiation. It was also shown that the ion exchange cartridges
could be reused for over 20 production cycles allowing for weekly maintenance schedule
suitable for a production facility.
Conclusion
The results indicate that recirculating of the target water can significantly increase
production of F-18 in relation to consumed O-18 water material. It can also increase
productivity by eliminating unnecessary idle periods for re-filling the target. A
backpressure regulator can reliably control target pressure. This method also allows for
continuous monitoring of the target material conductivity and accumulated radioactivity as
well as target pressure improving system reliability and predictability of production
results.
8
Production of F-18-Fluoride in the Nuclear Interface Titanium
Target on a MC-17 Cyclotron
J. Medema1, P.H. Elsinga1, M. Bruessermann2, J. van Essen1 and A.M.J. Paans1
1
PET-Center, Groningen University Hospital, P.O. Box 30001, 9700 RB Groningen, The
Netherlands
2
Nuclear Interface, Dorotheenstrasse 26a, D-48145 Münster, Germany
Aims
All the MC-17 cyclotrons (Scanditronix AB) were installed with a water target for F-18
production that was made from silver with a Havar entrance foil. The silver target has been
used for more than seven years and produced sufficient amounts of fluoride (26 GBq after
1 hour irradiation). The beam intensity put on this target was limited to 12 µA to have a
reliable daily production. Under these conditions no cleaning or other corrective
maintenance was required. The preventive maintenance includes drying of the targets and
the teflon lines with helium after each production for at least 30 minutes and changing the
teflon lines every year. In order to increase the production capacity to 60 GBq after a 1
hour irradiation (25 µA), we developed an interface for the Nuclear Interface high pressure
titanium water target to make the target suitable for the MC-17 cyclotron.
Methods
Window for entrance of beam and water cooling is a 50 µm titanium foil, 950 µl target
volume. The entrance of 19 mm of the collimator results in a beam transmission of 50 %
calculated from the Faraday cup, due the fact that the MC-17 is installed without external
beam line and focusing optics. Based on the excitation function and the power dissipated
energy (GBq/W), the optimal beam energy (from 17 MeV to 14.5 MeV) was obtained by
inserting a 0.5 mm thick alumina entrance foil. This also decreased the problem of
activation of the titanium target leading to the production of V-48 (production rate in
irradiated water: < 1.5 kBq/µAh). During the use of the first version of the target the main
problem was that the maximal allowable beam current was 15 µA. Higher currents resulted
in an unacceptable target pressure (>2700 kPa). Therefore we tested the following
parameters: enhanced helium cooling, water cooling on both the collimator and the target,
water cooling flow, external either internal expansion volume (100 and 200 µl), different
thicknesses of the titanium entrance or back foil (25-75 µm).
Results
In all these cases there was no significant improvement. Increasing the beam current to its
maximal allowable pressure lead to malfunctioning due to a pinhole exactly on the
borderline of the electronic welded foil in the insert, mostly on the water cooling side of
the target. For the daily production we used a beam current of 15 µA corresponding to a
target pressure of 1430 +/- 430 kPa. The target has been used for nearly three years for
9
daily routine production. When the target beam current is limited to 15 µA, the mean life
time of the titanium insert is about 1500 uAh. The saturation yield of the the total 796 runs
is 7.8+/- 2.4 GBq/µA. To increase the life time of the insert work is in process to develop
an insert with a front foil of 75 µm and the back foil of 300 µm.
Conclusions
The target has consistenly been able to produce yields of fluoride up to 40 GBq after 1
hour irradiation giving high radiochemical yields of products made with this fluoride
(FDG: 50-70 % EOB) and made the system useful for the routine production of
[18F]compounds.
10
Recirculating Low-Pressure Target for F-18 and N-13 Production
B.H. Mock and L.A. Corbin
Department of Radiology, Indiana University School of Medicine, Indianapolis, Indiana

46202, U.S.A.
AIMS: The increasing clinical demand for F-18 Fludeoxyglucose requires a greater yield
of F-18 fluoride along with conservation of the highly enriched O-18 target water.
Research PET facilities also need to maintain N-13 capabilities. We describe a simple
redesign of the low-pressure silver-body water target of the RDS-112 cyclotron that meets
all three requirements.
METHODS: By providing 3-way valves at both the load and unload ports, the target was
easily re-configured for O-18 water recirculation and F-18 production, as well as for bolus
loading of O-16 water for the occasional N-13 production. In-target pressure was
controlled either by the flow rate of an HPLC pump during O-18 water circulation, or
regulated by adjustable pressure relief valves during N-13 production. Continuous inert
gas purging of the target chamber and common valves between production runs helped
minimize cross-contamination and dilution of the enriched O-18 target water.
RESULTS: Our conventional low-pressure F-18 target (170 kPa) typically produced only
2.83 GBq/µA when irradiated for 90 min with 22 µA of ~10.2 MeV protons. When
operated in the new recirculation mode at only 4 mL/min and 690 kPa, a 30 µA run for 90
min on 10 % O-18 water yielded 0.43 GBq/µA of recovered F-18 fluoride ion. At 95 %
enrichment, such yield would reach 4.04 GBq/µA. Furthermore, under constant flow
conditions of 4-5 mL/min, the redesigned target has tolerated 45 µA irradiations while
producing an F-18 yield only 10 % lower than that obtained at 30 µA. When hi-purity O-
16 water was irradiated under no-flow conditions and 275 kPa pressure, N-13 production
reached 0.27 GBq/µA with a 10 µA beam. However, at 20 µA, the yield dropped to 0.22
GBq/µA and at 30 µA, to only 0.17 GBq/µA. While slightly lower than that obtained with
the high-pressure RDS-112 N-13 target, especially at higher beam currents, the N-13 yield
from this dual-purpose target was more than sufficient to produce unit doses of N-13
ammonia for cardiac perfusion PET-imaging.
Following irradiation, the O-18 target water flow was diverted through a small anion
exchange column (~20 mg MP-1:HCO3- form) to collect the F-18 fluoride ions. Dry
nitrogen gas was then routed through the collection resin to sweep away and recover the
trapped O-18 water, prior to fluoride elution with 2 mg K2CO3. The mass of enriched O-
18 water loss per irradiation and recovery was periodically measured and found to be
<100 mg.
CONCLUSIONS: With minor plumbing changes, and the mere implementation of

continuous target water recirculation during irradiation, we have more than doubled the F-
18 fluoride output of a low-pressure silver target. The chemical reactivity of the recovered
fluoride for FDG production has not been adversely affected by this modification.
Automatic loading and unloading routines have been developed to facilitate rapid
changeover whenever N-13 production is required.
11
Production of Multi Curie [18F]Fluoride using a Niobium Target
Chamber at Small PET Cyclotrons
F. Schmitz1, M. Monclus1, J. Van Naemen1, D. Ekelmans1, S. Goldman1, R. Verbruggen2,

F. Vamecq2, Y. Jongen2, F. Van Langevelde3, P.S. Kruijer3, P.J. Van Leuffen3, R. Mooij3
and P.J. Van der Jagt3
1
PET/Biomedical Cyclotron Unit, ULB-Hospital Erasme, Route de Lennik,
808, B-1070, Brussels, Belgium
2
Ion Beam Applications, Chemin du Cyclotron 3, B-1348, Louvain-la-Neuve, Belgium
3
BV Cyclotron VU, Vrije Universiteit, De Boelelaan 1085, 1081 HV Amsterdam,
the Netherlands
Irradiation of 18O-enriched water is the mostly used method to produce [18F]Fluoride on a

multi curie scale. Based on a target chamber design optimised in respect of experimental
thick target yields the performance and reliability of a target system is mainly determined
by selection of foils, seals and target chamber cooling.
In particular the target chamber materials have to be carefully chosen. Material properties
to be taken into account are (see Table 1):
- chemical resistance;
- thermal conductivity;
- activation of the target chamber;
- reactivity of the [18F]Fluoride produced;
- contaminants in the [18F]Fluoride produced.
Until now Silver and Titanium have found to be the most favourable target chamber
materials.1,2,3,4 Recently, Zeisler et al. described the successful use of Niobium as target
material5. Their results as such are very promising. However, the construction of the target
chamber, based on two Niobium hemispheres drawn from circular Niobium foils, is
unsuited for application on a low energy cyclotron, because energy loss in the cooling
water in front of the entrance foil is unacceptable when 10 or 18 MeV cyclotrons are used.
For this reason, based on our experience with Silver and Titanium target chambers, a
Niobium insert is designed and machined that fits in a standard IBA target support (Figure
1). The Niobium target was tested both at an IBA 10/5 and an IBA 18/9 PET cyclotron. In
a standard target system the inner cavity is completely closed and self pressurised under
bombardment. The target window system consists of 25 µm and 50 µm Havar foils for
respectively the IBA 10/5 and IBA 18/9 PET cyclotron. The target windows are cooled
with a stream of Helium and the target chamber is cooled with water.
It is demonstrated that a Niobium insert can be successfully used for the production of
multi Curie amounts of [18F]Fluoride using a standard system. Therefore the results
obtained by the Niobium insert could directly be compared to results obtained by target
inserts constructed from Silver and Titanium. The results are summarised in Table 2 and
3. The drawback of Silver (in particular the formation of colloids leading to a decrease of
yield of [18F]Fluoride and to obstruction of tubings and valves) and Titanium (the high
12
activation of the target chamber) can be avoided by using Niobium as target chamber
material.
As could be expected there is no significant difference in experimental thick target yields

for the different materials by the same effective volume of the insert. The amount of
produced [18F]Fluoride is directly related to the maximum operational beam current. It is
demonstrated that the beam intensity influences the internal pressure of the target. The
cooling efficiency of the target chamber is determined by the thermal conductivity of the
insert material and the insert wall thickness.
It is demonstrated that, related to differences in thermal conductivity, higher beam

currents can be used in target chambers constructed of Niobium instead of Titanium,
leading to higher amounts of produced [18F]Fluoride.
We could not detect any radionuclidic impurities originated from nuclear reactions on
Niobium, in the resulting [18F]Fluoride solution. The only detected radionuclidic
impurities were identified as nuclides formed under proton irradiation of Havar. Therefore
in future research the possibility of exchanging Havar foils for Niobium foils will be
investigated.
Fig. 1. Front and back view of the target chamber.
13
Table 1. Properties of target chamber materials.
Material properties Ag Ti Nb
(+ = high; 0 = average ; - = low)
chemical resistance - + +
thermal conductivity + - 0
radiation safety of target chamber1 + -2 +3
reactivity of 18F produced 0 + +
contaminants in 18F produced - - +
target pressure + - 04
1
Plus sign means relatively low activation of the target body.
2
see Figure 4
3
see Figure 3
4
see Figure 2
Table 2. Yields obtained with IBA 10/5 after 120 minutes of 10 MeV proton irradiation.
18
Target F Yield Thick target yield
Beam intensity
insert @ EOB @ sat
(µA)
material (Ci) (mCi/µA)
Ag 70 4.2 ± 0.5 (n=23) 116
Ti 55 3.1 ± 0.3 (n=63) 106
Nb 65* 3.5* 111*
*extrapolated value
14
Table 3. Yields obtained with IBA 18/9 after 120 minutes of 18 MeV proton irradiation.
18
Target F Yield Thick target yield
Beam intensity
insert @ EOB @ sat
(µA)
material (Ci) (mCi/µA)
Ag 40 5.1 ± 0.7 (n=95) 237
Ti 30 3.9 ± 0.3 (n=51) 245
Nb 35 4.4 ± 0.4 (n=95) 237
100
80 Ag
Ti
60
Nb
Pressure (bar)
40
20
0
0 10 20 30 40 50 60
B e a m in te n s ity (µ A )
Fig. 2. Evolution of the pressure v.s. beam intensity.
20
15
Dose rate (Sv/hr)
10
0
0 20 40 60
T im e (h o u rs )
Fig. 3. Evolution of the dose rate at 30 cm from the Nb target set up on a Cyclone-18/9.
15
20
18
16
Dose rate (mSv/h)
14
12
10
8
6
4
2
0
0 20 40 60 80 100 120
Production number
Fig. 4. Evolution of the dose rate at 30 cm from the Ti target set up on a Cyclone 18/9.
References
1. Bormans G., Mishani E., Ghyoot M., Schmitz F., Verbruggen R., Boeyen P.E.,
Seventh International Workshop on Targetry and Target Chemistry, Heidelberg, 8-11
June, 1997.
2. Ghyoot M., Verbruggen R., Schmitz F., Ohkoshi M., Vamecq F., Jongen Y., 38th
Meeting of the Japanese Society of Nuclear Medicine, 5-9 October, 1998.
3. F. Schmitz, M. Ghyoot, Patent n° : 09900072, Publication n° : 1011263A6.
4. Van der Jagt P., Van Leuffen P., Van Langevelde F., Kruijer P., Ghyoot M., Schmitz
F., Verbruggen R., Vamecq F., Destexhe C., Jongen Y., 8th International Workshop on
Targetry and Target Chemistry, St.-Louis, 24-26 June, 1999.
5. Zeisler S.K., Becker D.W., Pavan R.A., Moschel R., Rühle H., Int. J. Appl. Radiat.
Isot., 53, 449-453, 2000.
16
Effect of the Beam Profile on the F-18-Fluoride Target Saturation
Yield.
D. Soloviev, C. Tamburella, M. Ghyoot and G.-J. Beyer
Cyclotron Unit HUG, Geneva, Switzerland; and IBA, Leuvain-la-Neuve, Belgium
AIMS: Beam quality parameters play an increasing role in terms of efficacy of target
irradiation at PET cyclotrons. In order to quantify the effect of the quality of the beam,
namely the particle beam density and the shape of the beam, on the productivity of certain
targets we performed a series of measurements. Two different approaches were used to
influence on the beam parameters. Increasing the distance between the beam exit and the
target makes use of beam natural divergence, this is achieved by installation of a spacer
tube before the vacuum target foil (called "beam line " by the manufacturer). Installation of
additional metal foils (called “degrader”) on the way of the beam degrades the particle
energy and adds dispersion of the beam.
METHODS: Irradiations by 18-16 MeV proton beams of 5-30 µA intensities were

performed on CYCLON 18/9, IBA cyclotron. Beam line was produced by IBA and
represented a spacer tube of aluminium with the length of 30 cm installed before the
vacuum foil of the target. Degrader was placed prior to the vacuum foil of the target at the
end of the beam line and consisted of the stack of 20 Al foils having thickness of 0.02 mm
each. The beam spot geometry was assayed by autoradiography of the target foil using
Cyclone Storage Phosphor Screen (Packard) or the Linear Analyzer (Berthold ). Efficiency
of the irradiation was estimated by measuring the saturation yield of F-18-fluoride from
the standard IBA small volume silver target (Ag body, 0.6 ml cavity for O-18-enriched
water).
RESULTS: Installation of the beam-line enlarged the beam spot only by 10 %.

Nonetheless 15 % increase of the F-18-fluoride saturation yield was obtained. Installation
of degrader had a major influence on the homogeneity of the beam profile. The "hot spot"
measured as a Full Width at Half Maximum of the radioactivity profile on the irradiated
target foil (FWHM) was increased by 62 %. Installation of the degrader induced a loss of
energy of the proton beam which caused significant decrease of the F-18-fluoride
saturation yield at 15 µA (measured 5.7 GBq/µA for combination of degrader with the
beam-line vs 6.7 GBq/µA for beam-line alone). However, the energy loss could be
compensated by the better homogeneity of the beam profile, as visualised by
autoradiography. This allowed to run the F-18-fluoride target at higher currents (at 20 µA
saturation yields were 4.9 vs 4.0 GBq/µA correspondingly).
CONCLUSIONS: The quality of the beam profile is important for the reliable operation of
the targets at higher intensities. The distance between the stripper and the target has a
lower influence on the beam homogeneity than beam dispersion on the metal degrader
foils. Installation of the 30 cm beam line between the stripper and the target in
combination with the degrader allowed to improve significantly the homogeneity of the
beam. This fact is important for irradiation of sensitive solid materials as for instance Te-
124-tellurium oxide for I-124-iodide production.
17
A Convenient Method for the Purification of O-18 Water
K. Weber, H. Marx and M. Eisenhut
Deutsches Krebsforschungszentrum, Abt. E0301, Im Neuenheimer Feld 280, D-69120

Heidelberg, GERMANY
Aims
During the purification of O-18-water by conventional distillation certain contaminations,

like acetonitrile will not be removed entirely because they form an azeotropic mixture.
Moreover the enrichement of O-18-oxygen is reduced by an exchange with O-16-oxygen
from e.g. MnO4-. The extremely small amounts of O-18-water used per batch, force the
user to collect several batches to receive sufficient material for the distillation process. A
satisfying yield of F-18-fluoride from the irradiation of O-18-water, however, can only be
achieved if there are no organic contaminations and if the isotopic enrichement is
sufficient.
Methods
Used O-18-water is cleaved into its constituents O-18-oxygen and hydrogen by

electrolysis, thereafter the gases are carried via different tubes into a fuel cell where it is
catalytically oxydized back into water. Since electrolysis proceeds quite slowly, the supply
of incoming water is regulated by a remote controlled module. Automation of the process
provides continual operation of the system.
Results
Used O-18-water can continually be purified in a simple manner from contamination

originating from the separation of F-18-fluoride. The volume of the electrolyte chamber
can be kept quite low at 4 to 6 ml. Over 99,9 % of the contamination could be removed in
a single pass.
Conclusion
Electrolysis combined with fuel cell technology is suggested as a convenient alternative

for O-18-water purification.
18
Self-Regulating Thermosyphon Water Target for Production
of F-18-Fluoride at Proton Beam Power of One kW and Beyond
B. Wieland, C. Illan, M. Doster, A. Roberts, R. Runkle, C. Rowland and J. Bida
PET Facility, Duke University Medical Center, PO Box 3808, Durham, NC 27710, USA
Aims
PET cyclotrons can deliver proton beam currents above the limit tolerated by current
generation enriched O-18-water targets producing F-18-fluoride ion. We seek to
demonstrate that much more of this beam power can be utilized by implementing bottom-
pressured thermosyphon targets, which more effectively optimize the evaporation/
condensation mode of heat transfer than top-vented or top-pressured reflux targets.
Methods
Thermosyphon geometry appropriate for enriched water targets uses the beam strike as the
boiler section and a volume above it as a condenser section. The device is completely filled
with water, closed off at the top, and high pressure helium is applied on a liquid-filled exit
port at the bottom. Beam heating causes thermal expansion to force water out of the target
bottom until boiling begins, and then additional liquid is moved out due to beam strike
bubbles and condenser vapor volume, where bubbles leave the liquid surface and condense
at the top to run back down the sides. This condenser vapor volume increases as necessary
to provide an evaporation/condensation space adequate to remove increasing beam heat,
and decreases if beam power is reduced. This self-regulation results in a full liquid fill
again if all beam power is removed. Transport of pure water vapor is thus not impeded by
the presence of foreign gas molecules. Experiments at 22 MeV and 30 atm were conducted
on three prototype targets (TS-1, TS-2, TS-3), all with a 10 mm diameter by 10 mm deep
boiler volume, and with condenser volumes of 0.3, 0.5, and 1.0 times the boiler volume.
Calorimetry on side and back wall cooling partitioned heat removal at about 4 to 1
respectively, and detected no beam reaching the back. Pressure is monitored with a Kulite
transducer in the target base. Natural water exeriments used a vertical glass sight tube (3
mm ID) with water extending continuously from inside the target up into the tube.
Dynamic movements of the water in and out of the targets as a function of beam current
were observed via TV by watching the tube level. LabVIEW control software coupled to
field point modules loaded and delivered target water via valves and pumps, and displayed
and recorded pressures and temperature differentials on line.
Results
TS-1 and -2 reached heat transfer limits at 22 and 28 microamps (464 and 616 watts)
respectively, when vapor began intruding the beam strike. TS-3 operated comfortably to 46
microamps (1012 watts), at which point the cyclotron's preseptum sustained damage. The
TS-3 data extrapolate to 60-70 microamps, but experimental measurement requires a
19
negative ion cyclotron with a safe extracted beam power of at least 1500 watts. One hour
runs on enriched water with TS-2 below its power limit produced F-18 saturation yields of
12.2 GBq per microamp, in good agreement with data obtained from E. Hess and S. Qaim.
TS-3 yields at 40 microamps are planned at both a bottom and a top pressure configuration
for comparison.
Conclusion
Our data indicate a self-regulating bottom-pressured thermosyphon target may offer a way
to significantly increase yields of F-18 at existing and future PET cyclotron facilities. We
plan to pursue beta testing at appropriate cyclotron sites to establish performance limits,
reliability, and ability to synthesize FDG.
20
Regenerative Turbine Pump Recirculating Water Target for
Producing F-18-Fluoride Ion with Several kW Proton Beams
B. Wieland and B. Wright
PET Facility, Duke University Medical Center, PO Box 3808, Durham, NC 21110, USA
Aims
Some PET cyclotrons can deliver proton beam current at power levels above 1.5 kW. We
are not aware of demonstrated or previously published proposed O-18(p,n)F-18 target
systems capable of dissipating this level of power while maintaining yield-thick water in
the beam strike. We are pursuing the goal of designing and building a system which will
operate reliably at several kW. This can be done if target water can be passed through the
beam strike with a transit time so short that the boiling point will not be reached, or that a
minimum amount of Bragg peak vapor bubbles will be produced and those surviving will
be quickly swept away and condensed. The target must also recirculate the enriched target
material through heat exchangers capable of removing the beam heat. The cost of the
recirculating water inventory must also be held to an acceptable value. The key to the
success of such a system is a minature pump that will operate cleanly and reliably at high
flow rate and high pressure. After considerable study, we have selected the regenerative
turbine as the pump of choice. This pump has the added advantage of tolerating pumped
fluid containing entrained vapor without sustaining cavitation damage.
Methods
Our recirculating target system is being designed as an integrated interconnected assembly,

with all surfaces contacting target water actively cooled except the turbine rotor (pump
housing, intercoolers, and target body). The initial emphasis has been on sizing the
preliminary pump design by characterizing the flow characteristics with mechanical
prototypes. Following this, details of the intercoolers and target body can be developed, as
well as pump seals and bearings appropriate for extensive beam testing. A magnetically-
coupled motor drive is under consideration.
Results
A 0.86 ml prototype pump with 2 mm thick 25 mm diameter rotor has been tested from
1500-3500 rpm up to 1 atm pressure differential and up 0-30 atm overpressure. Flow
varied from 1.3-2.7 liters/min at maximum and minimum pressures respectively. This
pump has been modified to a 2.41 ml prototype with 3 mm thick 28 mm diameter rotor,
and is being prepared for testing with a larger drive motor.
21
Conclusions
Feasibility of acceptable performance and operating cost has been studied for a high
power recirculating water target system based on a minature regenerative turbine pump.
This system has the potential to operate at proton beam currents of several kW. The first
pump prototype test has been successful. Preliminary design of the intercoolers and target
body for an integrated system is in process. These will be tailored to the pump geometry
developed from ongoing mechanical testing. We hope to fabricate an integrated system
suitable for beam testing by late 2002, to allow the start of beam testing at 1 kW on the
Duke CS-30 cyclotron by the end of the year. We speculate that considerable evolution of
the target body configuration will occur during thorough 1 kW beam testing, as well as
significant modification of the pump and intercoolers. Beta testing at an accelerator site
that can provide several kW of proton beam power is anticipated following successful
completion of Duke beam testing.
22
High Power Enriched Gas Targets
Coordinators:
Volker Bechtold, Karlsruhe
Bruce Mackay, Hammersmith Hospital
Target Foil Grid Supports for PET Isotope Production
G.T. Bida1, B.W. Wieland2, J.W. Lenz3 and C.W. Alvord4

1
Biomedical Research Foundation, Shreveport LA
2
Duke University PET Facility, Durham NC
3
JW Lenz & Associates, Waxahachie TX
4
CTI Cyclotron Systems, Knoxville TN
Introduction
Proton cyclotrons remain the accelerator of choice for production of radioisotopes for PET
applications; however, alternative approaches have been explored over the years. One
such alternative was a linear RFQ device that accelerated helium-3 [1]. The particle
choice, energy and accelerated beam dimensions presented formidable challenges to
targetry development. The peak power densities of pulsed rf beams are normally handled
by the use of high aspect ratio beam strike geometries in the targets. The downside of the
approach is the loss in target foil rupture pressure. The problem is exacerbated with He-3
due to the particle's high rate of energy loss in materials, thus demanding the use of very
thin windows. Others have encountered similar target design difficulties [2, 3].
The use of mechanically supported target foils has been previously reported [4]; however,
to address some of the issues at hand, a novel target foil grid support was conceived: the
foil would be 1) supported by an array of hexagonally-shaped holes separated by a thin
septum, 2) fabricated from a high thermal conductivity metal(aluminum or copper) and 3)
water-cooled via channels machined into the structure's perimeter. A honeycomb pattern
affords a greater fraction of beam transmission than round holes and efficient heat removal
can result from high water coolant velocities. The grid support's potential was extended by
further suggesting that a second, upstream grid could be used as a heat shield or
"sacrificial" structure.
Before fabrication of any such supports was initiated, the merits of the design concept
were tested via Computational Fluid Dynamics/Finite Element Analysis(CFD/FEA)
modeling. This analysis helped to identify the key parameters that would most likely
affect the grid's performance in the anticipated environment of the He-3 RFQ. Analyses
were extended to include the structure's utility for dc proton accelerator targetry.
Results presented herein for this grid support concept include:
• CFD/FEA analyses for beams of 12 MeV protons and 10.5 MeV helium-3;
• Experimental failure pressures for several typical target foil materials supported by a
hexagonal grid. Some failure pressures are compared to those generated in the presence
of a 11 MeV proton beam;
• Calorimetry measurements intended to evaluate the grid's heat removal capabilities
under proton irradiation.
24
Materials and methods
Grid supports were fabricated from 6061-T6 aluminum or OFHC copper and comprised
an array of 3.175 mm hex holes within a 10 mm diameter, separated by septa of varying
dimensions(0.254-0.381 mm) plus integral coolant channels at the structure's perimeter.
Physically bonded and non-bonded foil/grid combinations were used in this investigation.
Several common target foil materials were pressure tested using the hex hole grid support
and compared to unsupported foils of the same diameter. Foils were placed in a special
rupture fixture with pressure supplied by either regulated helium gas or an hydrostatic
pump.
Selected grid-supported and unsupported foils were hydrostatically ruptured under proton
irradiation conditions using the Duke CS-30 cyclotron. The target and associated
apparatus are diagrammed below. The test target was aluminum, 10 mm ID, and variable
depth. Water cooled grids were either aluminum or Hycon 3 Cu/Be. A 9 mm water-cooled
collimator/degrader dropped the proton energy to 11MeV.
DI water from the lab

Expansion volume
Test gauge
28 MeV proton beam
Degrader from 28 MeV to 11 MeV protons

Hydrostatic pump
Grid or standard window support flange,
with drain port for rupture material
80 ft. of 1/16" o.d. x .030" i.d. PEEK tubing
through vault wall
Target body
Calorimetry experiments were performed via 1) 18 MeV H+ irradiation of a modified CTI

RDS112 ammonia target fitted with an edge-cooled aluminum grid connected to a closed
loop cooling system decoupled from the accelerator cooling and 2) 22 MeV H+ irradiation
of a test target fitted with an upstream aluminum collimator/degrader/sacrificial grid
assembly followed by an aluminum target body support grid. Conventional accelerator
system cooling was used. Grid, collimator/degrader and target body inlet/outlet water
temperatures were monitored via T- or K-type thermocouples.
Results
CFD/FEA Analyses
To minimize initial computational complexities, the following design parameters and

operating conditions were assumed:
25
• the structure would be a copper disc with 3.175 mm cell size honeycomb holes and 0.508
mm septum to which is soldered a 0.00635 mm Havar foil. A heat shield would be
ignored at first;
• edge cooling would be provided by 13oC water, 80 psig pressure drop, through three
0.508 wide by 2.54 mm deep channels;
• uniformly distributed dc beams of 12 MeV x 75 µA protons (1 cm diam) and 10.5 MeV
x 100 µA helium-3 (1 cm diam and 2 x 10 cm) were assumed;
• two water target scenarios were investigated: i) submerged forced-jet cooling of the
Havar and ii) a 350 psig pressurized natural convection target. A separate CFD analysis
was required to determine a heat transfer coefficient for the natural convection case.
The most important results derived from these initial calculations included:
• Havar temperature profile is predominately due to target water cooling heat transfer
coefficient vs heat conduction to the copper structure;
• for a 1 cm beam diameter, forced jet cooling is required in the case of helium-3;
• the heat shield may not be necessary.
Additional calculations were suggested by making obvious changes in some parameters,

to include beam dimensions, target foil and target plus heat transfer scenarios. The results
of these calculations and the various scenarios are summarized in Figure 1. What was
learned from these additional calculations included:
• thermal conductivity of the foil material makes a profound difference in the foil's
calculated beam strike temperature;
• for uniform beam heating of an edge-cooled window, the maximum temperature of the
foil is independent of the foil thickness. This result is of more value to a pulsed proton
beam than helium-3;
• with the larger beam dimension, a natural convection target appears workable with a dc
beam.
Foil failure tests
Results of rupture test for various grid supported and unsupported foil materials in the
presence and absence of a proton beam are summarized in Tables 1 and 2.
Calorimetry: Closed-loop experiments
All experiments were done at 250 psig with water flowing through the beam strike at 1
cc/min. Assuming the water enters the target at 20°C and leaves at 208°C (boiling point at
250 psig), 13.1 watts are removed through this path. Initial scouting experiments
determined optimum grid coolant volume and general behavior of the system, especially
any other heat loss. The slope of the temperature drop of the reservoir after beam is turned
off gives a heat loss from the warm grid coolant to the colder cyclotron coolant of about
55 watts.
Irradiations were performed at 20, 30 and 40 µA (20 min) for the grid only cooled via the
3 liter water supply. The power that goes with the temperature rise rate observed was
26
calculated, then the 55 watt loss was added to get the actual heat removed. In all cases,
this was greater than the heat expected to be dissipated solely in the foil and grid. The
cooling scheme was reversed, running the 3 liter supply through the target cooling path,
and putting regular cyclotron chilled water across the grid. The loss in conduction between
the two stages was expected to be the same but in the opposite direction, so in the last two
calculations, 55 watts is still added. Additionally, 13.1 watts lost from the temperature rise
of the beam strike water is added. Nevertheless, we cannot account for as much heat from
the target body as one would expect from the beam that makes it through the foil and grid.
In all cases there is an apparent, and not insignificant, cooling of the target water by the
grid/foil combination.
Max. Havar Foil Temp.(deg C)

2037
2000
H+/1
1750 He-3/1
He-3/2x10
1500
1250 1182 1129

Cu/Be
1000
817
750
554
500
186 219
250 154
99
0
Nat Forced No No HS; N2 Gas
Conv Jet Target Target NTC Cool
Cool Cool Cool Cool
Fig. 1. Parametric variation of Havar foil temperature for proton and helium-3 DC beam
environments (HS = heat shield; NTC = no target cooling).
Table 1. Beam-on foil rupture data.
Foil, thickness(mm) Proton current(µA) Rupture(psig)¶
Cu/2 % Be, 0.0508 20 740

Cu/2 % Be, 0.0508 20 880
Cu/2 % Be, 0.0508 30 880
Cu/2 % Be, 0.0508 40 820
Havar, 0.0127 20 880

Havar, 0.0127 30 880
Havar, 0.0127 40 840
¶ Havar was grid supported, Cu/Be was not.
Calorimetry: Thermosyphon target experiments
A target configuration based on a self-regulating thermosyphon operating principle has

proved invaluable in demonstrating the heat removal benefits of an edge-cooled grid up to
nearly a kW of beam power. The target components that include combinations of
27
upstream sacrificial and target body grids plus a sight tube for beam strike target water
expansion measurements have allowed for careful heat balance and partition
measurements among the components used.
The reader is directed to these Proceedings' paper by B. Wieland et al. on "Self-Regulating

Thermosyphon Water Targets for Production of F-18-Fluoride at Proton Beam Powers of
One kW and Beyond" for results and discussion of grid performance at relatively high
beam powers.
Table 2. Static rupture pressures of various foils.
3.175 mm hex/
Thickness 10 mm diam 0.254 mm septum Temper
Foil (mm) Vendor rupture(psig) rupture(psig) Hex/Thru tested
Al 0.0254 unknown 140 370/430 2.86 unknown

Au 0.0254 Reactor Expts. 30/25 150 5.45 unknown
Arnavar 0.00762 Arnold 105 635/615 5.95 unknown
Havar 0.00635 Hamilton 145 520/500 3.52 cold rolled
Havar 0.01016 Hamilton 275 800 2.91 cold rolled
Havar 0.0254 Hamilton 1000 n/a n/a unknown
Havar 0.03048 Hamilton 1200 2800/2900-3000 2.4 cold rolled
Havar 0.03048 Hamilton 1700 4500 2.65 annealed
Ta 0.0254 Alfa Aesar 440 1200 2.73 unannealed
Ta 0.0508 Goodfellow 975 n/a n/a annealed
Cu/2 % Be 0.008255 Hamilton 90 320 3.56 hard
Cu/2 % Be 0.0254 Goodfellow 200 750-800 3.75-4.00 as rolled
Cu/2 % Be 0.0508 Goodfellow 925 2700 2.92 half hard
Ti 0.0254 unknown 860 >1600 >1.86 unknown
Data in italic are hydrostatic rupture pressures.
Table 3. Heat balance data from BRF.
Observed power Apparent heat

Watts into Total power in (corrected for heat removed from
µA x time grid (target) Watts into foil (calculated) loss) target water
20 x 20 73.8 12.1 85.9 119.8 33.9

30 x 20 111 18.1 129.1 162.7 33.6
40 x 20 148 24.2 172.2 198.3 26.1
20 x 20 274 --- 274 183.2 91
30 x 20 411 --- 411 261.6 146.4
28
Summary
Resource constraints during the 3He RFQ project demanded that minimal time be devoted
to accelerator cleanup resulting from target foil failures. This requirement was nicely
satisfied by the high aspect ratio hexagonal pattern foil grid support. Edge cooling of the
device provides the other primary feature, the benefits of which are being exploited in the
ongoing development of high power, high pressure proton targetry for F-18-fluoride ion
production. Grids have been easily fabricated from high thermal conductivity materials
(e.g., aluminum, copper, copper/beryllium and graphite) via conventional CNC and wire
EDM technologies. The pressure integrity, heat removal and high transmission of the hex
grids allow for single foil target operation, eliminating the need for helium window
cooling.
Our computational and experimental results reported herein lead us to the following
conclusions regarding the foil grid support:
• grid supported foil rupture pressures were found to be 2.5 to 6 times greater than
unsupported foil pressures. The ability to operate water targets at higher boiling points
allows higher temperature differentials at the internal surfaces of the target;
• thermal contact between foil and support is adequately provided by the target internal
pressure without having to resort to bonding agents;
• thermal conductivity of the target foil profoundly impacts its beam strike temperature;
• when backed by target water, foil supported and unsupported beam-on failure pressures
dropped insignificantly or not at all up to 40 µA of 11 Mev protons;
• we estimate that 10-25 % of the heat deposited in a proton-irradiated water target is
actually removed by the foil/grid composite.
References
1. Link JM, Krohn KA, Bida JT, DeHaas R, Larson DJ, Johanning J, Young PE,
Pasquinelli RJ. Production of PET Radionuclides using a 10.5 MeV 3He RFQ
Accelerator. Proceedings of the Fifteenth International Conference on Applications of
Accelerators in Research and Industry 1999; The American Institute of Physics, pp.
1010-1013.
2. Hughey BJ, Shefer RE, Klinkowstein RE, Welch MJ. Design Considerations for Foil
Windows for PET Radioisotope Targets. Proceedings of the IVth International
Workshop on Targetry and Target Chemistry 1992; PSI-Proceedings 92-01, pp. 11-18.
3. Schlyer DJ, Firouzbakht ML, Garcia I, Ferrieri RA. Correlation of Hole Size in Support
Windows with Calculated Yield Strengths. Proceedings of the Fourteenth International
Conference on Applications of Accelerators in Research and Industry 1997; The
American Institute of Physics, pp. 1363-1365.
4. Nickles RJ. A High Pressure Gas Target for Cyclotrons. Nucl. Instr. Meth.
177:593(1980).
5. Alvord CW. Computer Simulation Tools as Applied to Radioisotope Production Target

Design. Accepted for publication in J. Label. Cmpds. Radpharms.
29
Further Progress on Targetry for the Production of Ultra-High
Quantities of F-18-Fluoride
K.R. Buckley1, D.W. Becker2, R. Dahl1, S. Jivan1 and T.J. Ruth1
1
2
Now at ARGOS Zyklotron Betriebs-Ges.m.b.H., LKH Klagenfurt, PET-Zentrum NMSE,
St. Veiter Strasse 47, A - 9020 Klagenfurt
Aims
To build a target capable of producing multi-curie quantities of F-18-fluoride while being

able to recycle the target material.
Methods:
We have reported our work on a gas target system to produce ultra-high quantities of F-18-
fluoride at the 8th Workshop on Targetry and Target Chemistry and in the literature[1]
using a SS target body. While these results demonstrated proof-of-principle concepts there
were improvements to be made around the efficiency of F-18 extraction from the target
walls and the chemical reactivity of the F-18-fluoride. Since that time we have designed
another gas/liquid handling system and other target bodies.
Results
We will report on the performance of the modified system with respect to yields and
chemical reactivity of the obtained F-18-fluoride.
Conclusions
A gas target for producing F-18-fluoride has been demonstrated to operate at 50

microAmperes with the possibility of even higher beam currents which would result in
extremely high production rates.
References
1. Applied Radiation and Isotopes 55 (2001) 457-461.
30
Development of Target System for [F-18] Fluorine Production via
the O-18(p,n)F-18 Nuclear Reaction
K.S. Chun, S.D. Yang, S.H. Ahn, S.W. Kim, M.G. Hur, B.S. Moon and S.W. Hong
Lab. of cyclotron application, Korea Cancer Center Hospital, 215-4, Gongneung-dong,

Nowon-ku, Seoul, Korea
AIMS:The number of PET scanners installed in Korea has been sharply increasing for a
few years. F-18-fluorine has been gaining increasing importance in various F-18
radiopharmaceuticals production because its electrophilic reaction can afford many
versatile pathways in F-18 compounds synthesis.[1] We have experience of success in
making the enriched O-18 water target for ourselves for the routine F-18-FDG production
with nucleophilic substitution reaction. In order to produce new F-18 compounds,
promising agents in oncology study, we decided to make the enriched O2-18 gas target
system to produce F-18 fluorine.
METHODS: Our target system for F-18 fluorine production consisted of target chamber,
enriched O2-18 gas loading/cryogenic recovery system, collimator/energy degrader[2], gas
and vacuum control system. Conical shaped target chambers were fabricated of aluminum
and the target chambers were mounted on the multi-target irradiation station made by
ourselves. Also, 4 sectors carbon collimator (dia.=1.5 mm) was installed to control the
proton beam on the target. All valves in the gas and vacuum control system were the VCR
type of Swagelok product to maintain high pressure and vacuum. .
RESULTS: Two-step irradiation [3] protocol was adopted to recover the F-18 produced
with the first irradiation on the enriched O-18 gas and sticked into the chamber wall. 15
bar of the O-18 gas was intoduced to the chamber for the first irradiation and the beam
current could be increased up to 40 µA. In the second irradiation, after crygenic recovery
of O-18 gas, Ar/1 % F2 mixture was introduced and a short proton irradiation was done for
recovery of the F-18 fluorine from the target chamber.
CONCLUSION: Our target system is very useful for routine production of F-18 fluorine in
saturation yields of 59 mCi/µA at EOB. The loss of O-18 gas after production has been
lower than 1 % in 12 runs without any problem .
REFERENCES:
1. Roberts A.D., Oakes T.R. and Nickles R.J. Appl. Radiat. Isot. 46: 87 (1995).
2. Williams C.F., Boujot J.P., and Picard J. Tables of Range and Stopping Power of
Chemical Elements for Charged Particles of Energy 0.5 to 500 MeV, Rapport CAE-R
3042 (1966).
3. Bishop A., Satyamurthy N., Bida G., Hendry G., Phelps M., and Barrio J.R. Nucl. Med.
Biol. 23: 189 (1996).
31
Production of an Intense O-15 Radioactive Ion Beam using Low
Energy Protons
S. Lapi, T.J. Ruth, A. Zyuzin, S. Jivan and J.M. D’Auria
Simon Fraser University, Burnaby, B.C., Canada

TRIUMF, Vancouver, B.C., Canada
The production of adequate quantities of 15O for astrophysical applications has been a
source of concern at TRIUMF and ISAC for some time. The production of a 15O beam is
needed for two experiments (15O(α,γ)19Ne) and 15O(6Li,d)19Ne) at ISAC. Difficulties arise
due to the fact that oxygen is very reactive chemically and thus is difficult to extract from a
thick spallation target. The beam flux required for these experiments is extremely high,
between 109 and 1011 15O/s for the (α,γ) reaction) and thus high efficiencies at all steps in
the process will be required.
These studies involve increasing our understanding of the process of elemental synthesis in
stars, and the mechanism of the cataclysmic events of stars such as novae, x-ray bursts and
supernovae, which lead to the release of these elements into the universe. A series of fast
(seconds) network of nuclear reactions occur in explosive events involving short-lived
radioactive nuclei. Observations of gamma rays using observational telescopes confirm our
general understanding of these events, but further research is needed to truly understand
them. A key reaction in the ignition of novae, x-ray bursts and possibly supernovae is the
reaction of the isotope, 15O with helium nuclei in the very hot stellar environment. The rate
of this reaction at stellar temperatures is presently unknown. At TRIUMF with the new
radioactive beam facility, ISAC, combined with the new DRAGON reaction measurement
facility, the rate of this reaction could be measured and shed light on these ignition
processes. Unfortunately, oxygen is a very reactive element and it is difficult to produce an
intense beam of the radioisotope, 15O, (half-life = 122.2 sec).
The possibility of using one of the small cyclotrons on site (TR13, CP42 or TR30) for the
production of this isotope (15O) has been discussed. This new production approach will
involve the use of low energy protons to interact with a nitrogen gas target via the
15
N(p,n)15O reaction, which is accessible with attainable particle energies using these
cyclotrons. The BEARS project in Berkeley has had considerable success (1.0 x 108
particles/sec) with producing 14O beams with a similar method. Preliminary results using
the 14N(p,n)14O reaction as a model for the 15N(p,n)15O reaction have been obtained and
will be presented, describing the appropriate on-line chemistry to separate, trap, purify, and
transfer the 15O. The preferred chemical form of 15O, (C15O ) would be available for
introduction into an RF type of ion source for the production of the high intensity 15O ion
beam at ISAC.
32
Solid Targets at Low-Energy Irradiations.
Transport of Targets
Coordinators:
Ron D. Finn, Sloan Kettering
Robert J. Nickles, University of Wisconsin
Michael J. Welch, Washington University
The Electron Beam Test Facility for Cyclotron Isotope Targets
Examination
A. Arzumanov, A. Borissenko, S. Ilmatov, V. Koptev and S. Lyssukhin
Institute of Nuclear Physics, 480082, Almaty, KAZAKHSTAN
Abstract
Aims: Use of internal cyclotron beam for radioisotopes production requires reliable solid
targets, stable under high power radiation and heat flows. The main goal of the present
work was to design and construct experimental equipment allowing to generate the heat
flow with the same or higher power density than under cyclotron irradiation.
Methods: Electron gun was selected as a heater. Electron beam test facility equipped with
solenoid lens and electromagnetic beam scanning system allows to form homogeneous
irradiation area of different size. The target system used for tests is exactly the same as for
radioisotopes production at the cyclotron.
Results: Electron beam stand with maximal energy of electrons of 25 keV and beam
current up to 200 mA was designed and constructed. A number of experiments with
different targets was performed. The results of electron beam tests with high and known
beam-power density was used for calculation of maximal cyclotron beam current for
specific target materials. The technology of Zn and Tl high current target preparation was
successfully developed.
Conclusion: The electron beam test facility is routinely used for cyclotron targets
examination before radioisotope production. Such quality control gives a reliable
guarantee of safety of cyclotron irradiation. This technique also provides wide possibilities
for research and development of new target construction and target preparation
technologies.
Introduction
In the Institute of Nuclear Physics NNC RK on the basis of isochronous cyclotron facility
and radiochemical laboratory works are carried on for production of a set of radioactive
isotopes for applications in medicine, ecology-related investigations, industry and other
fields. These are radiopharmaceutical preparations “Thallium-201 chloride”, “Gallium-67
citrate”, sources 57Co and 109Cd, tracers 237Pu, 88Y, 85Sr. Nuclear reactions on protons with
energies up to 30 MeV and alpha-particles with energies up to 50 MeV are used for their
production. The irradiations are performed at the isochronous cyclotron U-150 with
internal and external beams.
Target material is coated on copper substrate by the method of electrolytical deposition.

At irradiation of a target of 8 cm2 with proton beam of energy 30 MeV and current
intensity 300 µA the maximal density of heat flow is 1800 W/cm2. Quality of the target
material adhesion, high density of the heat flow can cause overheating of a target and its
destruction.
34
The electron beam test facility was created for preliminary heat load targets examination
before their irradiation at the cyclotron. This facility also enables to expedite development
of technologies for isotope target production.
The electron-beam stand
The stand (fig.1) is represented by a vacuum chamber with an electron gun and a target
unit situated at the end-walls with base of 600 mm. An electron tube was chosen as a
heater due to several reasons [1]. It is possible to obtain high density of heat flow by
means of an electron beam. At that, it is easy to adjust rate of the heat flow with time.
Using beam forming systems and an external scanning device it is easy to change size and
location of a heating zone.
Fig. 1. The general view of the stand.
Vacuum pumps provide vacuum in the working chamber of 10-4 Pa. A cathode of the tube
is of lanthanum-boride with indirect heating. The tube is equipped with solenoid lens that
makes it possible to get the minimal beam size at a target to be 1.5 mm. Maximal beam
current is 200 mA at the electron energy up to 25 keV. Electronic power supply units are
protected from emergency conditions.
Electromagnetic control system of the electron beam assures its moving over the whole
target surface (20 x 40 mm2) in manual and automatic modes. The beam scanning
frequencies over a target (50 and 1200 Hz across and down, correspondingly) were chosen
to provide uniform heating.
The target unit of the stand is identical with the same of the cyclotron. Their cooling
systems are identical as well. Water pressure in the cooling system is 20 bar what provides
water speed in the cooling channel up to 30 m/s.
35
The stand is equipped with TV system enabling to control the beam location on a target
and control the irradiation process.
Electron beam density distribution influences considerably the results of the target thermal
resistance estimations. Measurements performed by wire probes and by the method of
small hole have shown that the best for the performed tasks is the beam of 4-5 mm in
diameter. At that the current density distribution is close to the Gaussian one.
It should be pointed out that irradiation at the stand almost completely reproduces the
target conditions at the cyclotron. At cyclotron irradiation the power deposition occurs at
deeper layers. Cooling water is subjected to radiolysis with formation of active oxygen
resulted in intensive oxidation of the cooling substrate surface.
Stand examinations make it possible to solve a set of practical tasks:

1. to perform regular preliminary examinations of targets prepared for irradiation at
the cyclotron.
2. to test targets when develop technology of their production.
There were performed tests of Ni, Zn and Ag targets. Electrodeposition modes for
target materials were optimized using the examination results. A zinc target
production technology for targets endured heat flows of ~38 W/mm2 was
developed as a result of investigations of zinc target with various intermediate
layers.
3. to determine maximal possible heat flow rate in a target with further recalculation
of this value for the cyclotron beam.
At irradiation with proton beam the power liberation occurs in the target bulk. The
stopping range of 20 keV electrons in metals is about 1 µm. That is why at irradiations
with electron beam of materials with low thermal conductivity the melting temperature at
the target surface is reached at lower heat flows. For example, thermal conductivity of
thallium is 44 W/m×K. Melting temperature at the surface of thallium target (160 µm
thick) deposited on 2 mm copper substrate is obtained at irradiation with electrons of heat
flow 1750 W/cm2 when at irradiation with protons this value comprises 2300 W/cm2.
Temperature distribution in a target for both cases is presented in fig. 2.
Fig. 2. The calculated temperature distribution in the target induced by

irradiation with protons (1) and with electrons (2). Maximal heat flow density
in the first case is equal to 2300 W/cm2, in the second case 1750 W/cm2.
36
Conclusion
The electron beam stand is routinely used for cyclotron target tests before radioisotopes
production. Such quality provides safe operation. This technique also provides wide
possibilities for research and development of new target design and target preparation
technologies.
References
1. Peletskiy, V.E., “The investigations of thermophysical properties of matter at the

electron heating conditions”, Nauka, 1983 (in Russian).
37
High Current Production of 11C, 13N, 15O, 17F, and 18F for PET using
a 3 MV Electrostatic Tandem Accelerator
Todd E. Barnhart1, Alexander K. Converse1, Kevin A. Dabbs1, R.J. Nickles1 and
A.D. Roberts1,2
1
Medical Physics, University of Wisconsin, Madison, Wisconsin
2
Psychiatry Departments, University of Wisconsin, Madison, Wisconsin
Target systems for the production of positron emitting radioisotopes used for medical
research with positron emission tomography (PET) have been developed for a 3 MV
electrostatic tandem accelerator (NEC 9SDH-2). While low current radioisotope
production has been previously examined with this accelerator (Roberts et al., 1998,
Ohlsson et al., 1996), high power target systems are presented for the production of [15O]
H2O (yield at saturation 9.8 mCi/µA), [17F]F2 (25 mCi/µA), [18F]fluoride (11.1
mCi/µA),[18F]F2 (15 mCi/µA), in target [13N] NH3 from carbon (4.0 mCi/µA), [11C]CO2
(8.4 mCi/µA), and [11C]CH4 (5.3 mCi/µA). Water-cooled support grids for the entrance
windows allow for beam currents in excess of 100 µA on gas, liquid, or solid targets.
Methods
Accelerator
The National Electrostatics Corporation 9SDH-2 Pelletron was designed to provide 100
µA of 6 MeV protons or deuterons within a maximum 10 mm diameter beam spot. The
actual performance regularly exceeds these specifications. The Torvis multi-cusp negative
ion source typically achieves more than 150 µA. The two charging chains are rated at 150
µA each, and accelerated beam currents in excess of 115 µA have been demonstrated. The
dome voltage of 2.97 MV required for 6 MeV single charge beams (with a 50 keV ion
source voltage) is conservative, and the accelerator has been run up to 3.67 MV (for 7.4
MeV beam). The beam tuning components include low energy steering, and high energy
quad focusing and steering magnets. The tuning capabilities coupled with an in-line
rotating wire beam profile monitor allow for fine, continuous control of the beam shape
and position. Independent adjustment of beam width in two dimensions is typically from 6
to 10 mm, with a full practical range of 2 to 20 mm FWHM.
Water cooled support grid
The support grid pattern consisted of circular holes arranged in a hexagonal pattern
(Barnhart et al., 2002, Nickles, 1980; Schlyer et al., 1997). Trials were done with differing
size holes to maximize the allowable beam current. Circular holes were made with 3.0 mm
or 1.7 mm diameters, of which the smaller diameter holes withstood higher pressures and
beam current. The aluminum between the holes was machined to a minimum of 0.18 mm
while the grid thickness was approximately 12.5 mm along the beam direction. Thicker
38
grids have been used with no change in performance. The gas and solid targets for the
tandem accelerator had the grid holes arranged inside a 1.9 cm diameter circle to minimize
the heating on the grid. The water target grid was done inside 1.25 cm diameter circle, for
the minimum volume of the enriched water. The grids are water cooled via two straight
channels in opposite sides of the support grid. Chilled water at 18 c flows at 2.3 l/min
through the cooling channels. The nec 9sdh-2 has a kf-40 quick connect flange designed to
connect to the end of the beamline. Figure 1 shows the basic design of this grid with 80
1.7 mm diameter holes. Single entrance window foils varied from 12.7 to 25.4 µm
aluminum and 2.5 to 12.7 µm Havar.
Fig. 1. Basic design of water-cooled support grid.
Gas targets
Flow-through gas target systems are used for the production of [15O]H2O, [17F]F2, [18F]F2,
[18F]F-, [11C]CO2 and [11C]CH4 (figure 2). Each product requires a mixture of gasses,
summarized in Table 1. Target gasses are supplied to either a dual flow tube gas
proportioner (Air Products) or a stainless steel manifold and feed the target via 1/8” O.D.
stainless tube with a short PTFE section for electrical isolation. The gas enters the target at
the downstream end though a stainless steel 1/8-NPT fitting, and exits through a 1/16-NPT
fitting within 1 cm of the upstream end. The radioactive gas is then delivered to the
analysis station though a short 1/16” PTFE tube followed by 30 m of HPLC grade 0.8 mm
I.D. stainless steel tube or 0.8 mm I.D. PTFE tube. The target pressure is monitored with a
corrosion resistant capacitance manometer (Entran) mounted on the gas line near the target
feed. The target pressure is controlled by the gas flow (and supply pressure) and the
impedance set by a needle valve on the outlet of the gas target. The water-cooled target
body is 6061-T1 aluminum with a 12.7 cm by 1.9 cm I.D. Gas pressures of the targets
typically operate at 15-50 psig O2 or N2 gas, although pressures to 450 psi are acceptable.
39
Table 1. Target Summary
Isotope/ Flowrate
Form Target composition Reaction Pressure (psi) (ml/min)
11 14
[ C] CO2 N2/1 % O2 N(p,α)11C 40 150
[11C] CO2 B2O3/ He 11
B(p,n)11C 20 80
[11C] CH4 N2/5 % H2 14
N(p,a)11C 50 200
[13N] NH3 CH4/ 20 % H2 12
C(d,n)13N 40-250 150-1000
[13N] NH3 Carbon/ H2 12
C(d,n)13N 10 40
15 14 15
[ O] H2O N2/ 1 % H2 N(d,n) O 40 150
[17F] F2 O2/ 10 % He/ .5 % F2 16
O(d,n)17F 150 200
[18F] F2 Ne/ 10 % He /< .5 % F2 20
Ne(d,α)18F 150 200
[18F] F- Ne 20
Ne(d,α)18F 150 50
18 18 18 18
[ F] F- O-H2O O(p,n) F 0 0
The gas target for the production of [18F]fluoride via the 20Ne(d,α)18F reaction varies from
the other gas targets. The aluminum target body has been nickel-plated and a Niobium
tube (15.7 mm I.D.) is inserted in the bore of the target. Holes drilled the niobium tube
align with the gas ports at front and rear, and with a third port centered on the bottom of
the target, to drain the water wash. Water is preheated to 85ºC, and kept hot through the
washing process with heat tape wrapped around the target. Total water volume is
approximately 20 ml per wash. The [18F]fluoride target runs with a beam profile of 6 mm
FWHM due to the smaller bore diameter whereas the beam profile for irradiating on the
standard large bore gas targets is typically held from 8 to 10 mm FWHM.
Fig. 2. Basic design of gas target.
Liquid target
A liquid target system is also used for production of [18F]fluoride from 18O(p,n)18F on
enriched water (figure 3). The water delivery lines are PTFE or HPLC grade stainless
tubing (0.8 mm I.D.). The switch valves at the target are manual stainless steel ball valves
with PTFE packing (Whitey, 40 series). The target body is Ag with stainless steel tubing
top and bottom for water loading. The tubing is close fit to the silver, then silver soldered
in place. The cylindrical beamstrike volume is 12.7 mm O.D. by 3 mm deep, sufficient to
stop 6 MeV protons in water. The beamstrike volume is 380 µl, with a total of 500 µl
required to fill between the switch valves. The target is completely filled, then sealed at
atmospheric pressure before irradiation. Target pressure during irradiation is not
monitored. The target mounts to the water-cooled support grid design used for the gas
target, with the window foil compressed against a PTFE O-ring. A water jet at the back of
the body cools the target. 6.0 MeV protons irradiated the water through Havar foil of
40
thickness 10.1 to 12.7 µm. Beam profile is kept to a maximum of 6 mm FWHM on the
water target.
Fig. 3. Basic design of [18O] water target.
Solid targets
Two solid target systems are used for the production of [11C]CO2 and [13N]NH3. The first
target (figure 4) is the same basic design as the gas target system, with the addition of a
concave brass ramp cut to a 15-degree angle. The ramp is filled with B203 for [11C]CO2
(Clark and Buckingham, 1975) or with a form of carbon for [13N]NH3 production. 6.0
MeV protons or deuterons irradiated the target through single aluminum entrance window
foils of 12.7 to 25.4 µm thick. Beam profile is kept to 8 to 10 mm FWHM while
irradiating on B203 target to preserve integrity and reduced to 6 mm FWHM or less on
carbon targets to increase the radiation induced diffusion.
Fig. 4. Water-cooled ramp target design for 11C production.
The second target system was based on using a 19 mm o.d. quarts tube as the main target
body so that the solid target (carbon) could be heated (figure 5). The quartz tube connected
the same water cooled support grid using ¾” cajon ultra-torr fitting into a water cooled
aluminum flange. The aluminum flange was water cooled via two straight channels in
opposite sides of the bore. Sweep gas entered into the front aluminum flange through a
1/8-npt fitting. Gasses exited the target through another ¾” cajon fitting which was
reduced down to 1/8-npt with stainless steel connectors. The quartz tube was heated to 950
c with a carbolite tube furnace mounted concentrically around the target material.
41
Fig. 5. Heated solid target design for 13N production.
Results
Gas targets
14
N(p,α)11C, for[11C]CO2
11
[ C]CO2 (t1/2=20.3 min) is produced by proton irradiation of natural N2 plus 1 % O2. This
method is commonly used on most higher energy accelerators, but is attractive for higher
current low energy accelerators because of target’s simplicity(e.g. Heselius et al., 1987).
With 6.88 MeV incident proton beam at 102 µA, the [11C]CO2 is trapped in soda-lime at a
saturation yield of 8.4 mCi/µA providing 850 mCi 11C for practical use.
14
N(p,α)11C, for [11C]CH4
Where most 11C is created in target as [11C]CO2 and then converted to [11C]CH4 for
labeling as [11C]CH3I, flow through direct production of [11C]CH4 bypasses this step.
[11C]CH4 is produced by proton irradiation of natural N2 plus 5 % H2 (Buckley et al.,
2000). Although the yield is lower at 5.2 mCi/µA at 6.02 MeV, conversion efficiency and
time favor its practical use with possible improvements in specific activity.
14
N(d,n)15O, for in target[15O]H2O
Water labeled with 15O (t1/2=122 s) is formed in target by deuteron irradiation of natural
N2 plus 1 % H2 (J.C. Clark 1995, Dahl et al. 1998). The total gas flow rate was maintained
at 200 sccm. No flow control valve was used on the Teflon line out of the target as
trapping occurs resulting in lower yields. Under these conditions with a 30 m, 0.8 mm i.d.
delivery line target pressure is 40 psig. The gas is bubbled through 10 ml water in a
calibrated dosimeter. The yield at 50 µA gave 9.8 mCi/µA and produced 491 mCi of
activity. Higher beam current test have not been done since this yield far surpasses the
need in blood flow imaging.
16
O(d,n)17F, for[17F]F2
[17F]F2 (t1/2=65 s) is produced for the production of [17F]CH3F by the deuteron irradiation
of natural oxygen gas with 10 % of a mixed gas containing 5 % F2 in He (Mulholland et
al, 1987, Roberts et al. 1998). This results in a nominal 0.5 % F2 mix in 150 psig oxygen
flowing at 200 sccm. The thick target yield for 100 µA of deuterons is 25 mCi/µA,
totaling 2.5 Ci of 17F.
42
20
Ne(d,α)18F, for [18F]F2
Production of [18F]F2 is similar to that of [17F]F2, where the same 10 % by volume mixed
gas of helium/fluorine except that instead of oxygen, a 90 % neon is used as the main
target gas. Using this gas mixture throughout the irradiation, 15 mCi/µA was the yield at
90 µA. Further trials will be performed to minimize the amount of carrier F2 required to
maintain a sufficiently high specific activity so that the [18F]F2 is useful as a precursor and
comparable to [18O] gas target systems (Nickles et al., 1984; Roberts et al., 1995b).
20
Ne(d,α)18F, for aqueous [18F]fluoride
[ F]fluoride (t1/2=110 min) is used in the production of [18F]FDG, the most commonly
18
used PET tracer. Using 20Ne has both cost advantages as well as advantages in targetry
consideration over [18O]H2O targetry (Helus et al., 1994). Operating gas pressures of 150
psig used here are much lower than sealed or over pressurized targets. Low gas flows
using a needle valve stabilize target pressure during the irradiation. Water washes of the
target have given yields of 11.1 mCi/µA at 85 µA.
Liquid targets
18
O(p,n)18F, for aqueous [18F]fluoride
The 18F is produced by 18O(p,n)18F using highly enriched [18O]water as the target material
(Roberts et al., 1995a). While the reaction yield at <6 MeV is significantly less than that
e.g. at 11 MeV, amounts sufficient for on site PET use can be produced without difficulty.
The saturation yield for 18F at 100 µA is 8.9 mCi/µA on fully enriched 18O water, lower
than the reported 44 mCi/µA at 10 µA (Ohlsson et al., 1996).
Solid targets
12
C(d,n)13N for [13N]NH3
13
N (t1/2=10 min) is produced by deuteron irradiation of a solid carbon target. This a highly
efficient means for 13N production for PET, even at incident energies <1.5 MeV (Dence et
al., 1994; Ferrieri et al., 1983;Shefer et al., 1994). Several forms of carbon have been
tested, with the best so far being coconut charcoal in a quartz tube heated to 950ºC gives a
saturation yield of >20 mCi/µA with 4 mCi/µA collected as in-target [13N] NH3 at 60 µA.
11
B(p,n)11C for[11C]CO2
The 11B(p,n)11C theoretical yield of 11C from protons on 11B is nearly twice that from
14
N(p,α)11C in the 6 MeV range, but we have been able to only coax slightly more than
half of 11C yield from it. At this time, no more than 42 µA of protons have been
successfully irradiated the B2O3 on the ramp with a yield of 5.3 mCi/µA. 11C is still more
effectively produced in the gas target. A water-cooled ramp target is in production, so that
variable cooling may be added to optimize the extraction of 11C from the B2O3 melt.
The developments in target systems have already improved production of currently

desired isotopes. Table 2 below summarizes where the current status of the accelerator
targets for the 9SDH-2 with theoretical yields shown for 6 MeV protons or deuterons.
43
Table 2. Measured Target Beam Currents and Yields.
Isotope/ Form Target Type Max Beam Current (µA) Yield (mCi/ µA)
[11C] CO2 Gas 108 8.4(@6.88 MeV)
[11C] CO2 Solid 42 5.3
[11C] CH4 Gas 50 5.2
[13N] NH3 Solid 60 4
[15O] H2O Gas 50 9.8
17
[ F] F2 Gas 100 25
[18F] F2 Gas 100 15
[18F] F- Gas 85 15
[18F] F- liquid 100 8.9
Conclusion
The NEC 9SDH tandem accelerator is ideal for reliable production of 17F and 15O
precursors for blood flow tracers, and the high beam current allows production of a
number of other useful radioisotopes for PET. Developments of targets capable of
withstanding the 100 µA of 6 MeV deuteron and proton beams demonstrate practical
production of lower yielding isotopes such as 11C, 13N, and 18F. The target hot atom
chemistry operating inside the target simplifies post irradiation production of [15O]H2O,
[11C]CH4 and [13N]NH3.
References
Barnhart, TE, A K Converse, K Dabbs, R J Nickles, K R Buckley, S Jivan, T J Ruth, A D

Roberts. Water-Cooled Grid Support System for High Power Irradiation with Thin Target
Windows. Applied Radiation and Isotopes In Press 2002.
Buckley, KR, J Huser, S Jivan, K S Chun, T J Ruth, 2000, C-11-methane production in

small volume, high pressure gas targets: Radiochimica Acta, v. 88, p. 201-205.
Clark, JC and Buckingham PD, Short-Lived Radioactive Gases for Clinical Use.
Butterworth & Co., England 1975.
Clark, JC,1995 Private Communication.
Dahl, JR, TC Chaley, RA Matacchieri, A Yee, V Dhawam, S Horowitz, R Jespersen, D

Margonleff, D Eidelberg, 1998. An Apparatus for the Preparation of [15O]-H2O for Rapid
Repetitive PET Studies. Applications of Accelerators in Research in Industry, AIP
Conference Preceedings 475. p.1019-1022.
Dence, CS, M J Welch, B J Hughey, R E Shefer, R E Klinkowstein, 1994, Production of

[N-13] Ammonia Applicable to Low-Energy Accelerators: Nuclear Medicine and
Biology, v. 21, p. 987-996.
44
Ferrieri RA, DJ Schlyer, BW Wieland, AP Wolf, 1983, Online Production of Nitrogen-N-
13 from a Solid Enriched Target-C-13 and its Applications to Ammonia-N-13 synthesis
Using Microwave-Radiation: International Journal of Applied Radiation and Isotopes, V.
34 (6), p. 897-900
Heselius, S.J., P Malmborg, O Solin, B Langstrom, 1987, Studies of Proton-Beam

Penetration in Nitrogen-Gas Targets with Respect to Production and Specific
Radioactivity of C-11: Applied Radiation and Isotopes, v. 38, p. 49-57.
Helus, F, V Uhlir, G Wolber, H Gasper, W Maier-Borst, 1994, Contribution to Cyclotron

Targetry II. Testing of the Target Construction Materials for the 18F Production via
20Ne(d,α)18F. Recovery of the 18F from various Metal Surfaces. J. Radioanalytical and
Nuclear Chemistry Articles. v. 182(2). p. 445-450.
Mulholland GK, Hutchins GD, Toorongian SA, Jewett DM 1987.Entry into F-17 (T 1/2 =
65 sec.) Radiopharmaceuticals - On Line Electrophilic Synthesis of the blood-Flow Agent
[F-17] Fluoromethane Journal of Nuclear Medicine v. 28(6) p. 1082.
Nickles, RJ. 1980. A High Pressure Gas Target for Cyclotrons. Nuclear Instruments &
Methods in Physics Research Section A 177, 593-594.
Nickles, RJ, M E Daube, T J Ruth, 1984, AN O-18(2) Target for the Production of [F-
18]F-2: International Journal of Applied Radiation and Isotopes, v. 35, p. 117-122.
Ohlsson, T, A Sandell, R Hellborg, K Hakansson, C Nilsson, S E Strand, 1996, Clinical

useful quantities of [F-18]fluoride produced by 6 MeV proton irradiation of a (H2O)-O-18
target: Nuclear Instruments & Methods in Physics Research Section A- Accelerators
Spectrometers Detectors and Associated Equipment, v. 379, p. 341-342.
Roberts, AD, RJ Davidson, R J Nickles, 1998, Production of 17F, 15O and other
Radioisotopes for PET Using a 3 MV Electrostatic Tandem Accelerator. Applications of
Accelerators in Research in Industry, AIP Conference Preceedings 475. P.1006-1009.
Roberts, AD, L C Daniel, R J Nickles, 1995a, A High-Power-Target for the Production of

[F-18] Fluoride: Nuclear Instruments & Methods in Physics Research Section B- Beam
Interactions with Materials and Atoms, v. 99, p. 797-799.
Roberts, AD, T R Oakes, R J Nickles, 1995b, Development of an Improved Target for [F-
18] F2 Production: Applied Radiation and Isotopes, v. 46, p. 87-91.
Schlyer, DJ., Firouzbakht, ML., Garcia, I., and Ferrieri, R A 1997. Correlation of Hole
Size in Support Windows with Calculated Yield Strengths.. Application of Accelerators in
Research and Industry, 1363-1365. New York, AIP Press.
Shefer, RE, B J Hughey, R E Klinkowstein, M J Welch, C S Dence, 1994, A Windowless

N-13 Production Target for use with Low-Energy Deuteron Accelerators: Nuclear
Medicine and Biology, v. 21, p. 977-986.
45
COSTIS: Compact Solid Target Irradiation System
J.J. Čomor¹, G.-J. Beyer², U. Ketzscher³, M. Brüssermann³, M. Rajčević¹ and D. Košutić¹
¹Vinča Institute of Nuclear Sciences, Laboratory of Physics, P.O. Box 522, 11001
Belgrade, Yugoslavia
²University Hospital of Geneva, Department of Radiology, Cyclotron Unit, CH-1211
Geneva 14, Switzerland
³Nuclear Interface GmbH, Dorotheenstr. 26a, 48145 Münster, Germany
Aim
Dedicated PET cyclotrons have a great potential for the production of interesting positron
emitting radionuclides of medium Z via (p,n) reaction. The aim of this work is the
development of a dedicated target unit for irradiating TeO2 for the production of
radioiodine via the Te(p,n)I reaction. Most interesting in this concern is 124I, but 123I or 120I
would be accessible using the same technology.
Method
We selected the experiences of the Rossendorf Cyclotron Group in the eighties [1]. The
target is a circular Pt-disk that carries the TeO2 melted into a circular grove in the center of
the disk. This target coin is mechanically very stable and easy to handle in both phases:
irradiation and radiochemical processing. The target coin is inserted into the station, fixed
pneumatically in the irradiation position, deblocked remotely after irradiation and falls
down driven by gravity into the transport container. The irradiated target disk is then
inserted into a dedicated furnace of the separation module developed by Nuclear Interface
and the radioiodine is released by applying a special temperature program, transported in
an air stream, and trapped in a pH=8 buffer solution. The target disk can be re-used for the
next irradiation without further treatment.
Results
The target unit has five main components:

1. Connecting flange, which fits to the standard IBA target connection. This flange can be
customized to fit virtually any other cyclotron.
2. A water-cooled graphite collimator that is electrically isolated, in order to monitor the
beam spill.
3. The target assembly, including the exchangeable window foil which can be used as a
degrader, isolating the closed He cooling loop from the cyclotron vacuum.The He
stream is cooling the window foil on the one side and the TeO2 surface on the other side
in the same time. The target plate is electrically isolated for current monitoring.
4. The water-cooling jet directed to the rear of the target disk.
46
5. A remotely controlled pneumatic locking mechanism that holds the target disk in place
during irradiation and releases the target disk so that it can fall down into a transport
container. The complete target unit has been manufactured and tested.
Conclusion
This compact target unit for irradiation of sensitive materials, with He cooling that is
common in PET target technology, can be easily adapted to the target ports of any PET
cyclotron. Earlier experiences illustrate that one can apply a proton beam of around 15
MeV up to 20 µA/cm2 density to such a target. The target area depends on the local beam
parameters; most suitable is 1 cm2, which can carry 250–300 mg of the enriched 124TeO2.
Batches of 30–50 mCi 124I are accessible under optimal conditions.
References
1. R. Brückner and H. Gouratzsch: “Irradiation system for target materials with low heat
transmission”, in R. Fülle (ed.) ZfK-594, Rossendorf 1986, pp. 16-18.
47
A Recoil-Escape Fiber Target using 94Mo(p,n)94mTc to Produce
94
TcO4- Precursor for Radiolabeled Compounds Useful in
Positron Emission Tomography
C.D. Illan and B. W. Wieland
Department of Radiology Duke University Medical Center, NC, USA
Introduction
The purpose of this project is the design a cyclotron target system for producing
Technetium-94m (94mTc), to be used as a radiochemical labeling agent for Positron
Emission Tomography (PET) imaging.
In target approaches prior to this work, 94mTc was recovered by chemical separation from
the oxide target material, which involves a high exposure dose to the production chemist.
The proposed fiber target system to produce 94mTc will not only reduce the radiation dose to
the chemist, it will also make pertechnetate (TcO4-) readily available after end of
bombardment (EOB), without chemical processing.
PET agents have been successfully produced with a slurry target and a fiber target, using the
principle of recoil escape. This method consists of maintaining the target atoms in a
permanent chemical and geometric form, while recovering the product radionuclide ion by
stopping it in a liquid or gas. The product radionuclide is then transported out of the target
while leaving the target atoms behind to be used in subsequent production runs We have
used this approach to design the novel 94mTc target system which is the focus of this thesis.
The positron-emitter 94mTc can be used for PET in numerous radiochemical forms that have
been developed for 99mTc, avoiding the usual extensive animal and metabolic studies that
must precede the use of any radionuclide application. The advantages of PET include high
resolution and quantitative tissue concentrations. The subsequent availability of kinetic data
for compartmental modeling is extremely useful in studies to understand the mechanism of
drug action.
The proposed target is designed with thin molybdenum fibers distributed in a container of
gas (steam or a noble gas). The advantage of steam is that the final product is pertechnetate
(TcO4-), which is the labeling agent used in the pharmaceutical kits widely available and
used by almost every nuclear medicine department.
Nuclear reaction cross section data are needed for optimizing the production of a
radioisotope, especially for calculating production yield and impurities as well as for target
design and chemical processing. Reaction cross-section data are obtained from published
work. Ranges of protons in the target mixtures and stopping power of the mixtures are
48
calculated using TRIM, a program that allows a quick calculation of parameters for any type
of mixture.
Information about the mechanism of a nuclear reaction is often obtained from the
distribution of recoil range of its product nuclide. This distribution is frequently required in
charged particle activation analysis and radionuclide production. These measurements have
been done using catcher foil technique methodology, such as results published by Iwamoto
et al [17]. They will be used in this work to analyze the energy of the 94mTc nuclides
recoiling from the molybdenum fibers. From the conservation of energy and momentum, the
maximum and minimum recoil energy can be calculated for a reaction that proceeds via the
compound nucleus (it is assumed that the recoil nucleus is formed solely by the decay of the
compound nucleus).
Target Nuclide + Incident Particle

↓
Compound Nucleus
↓
Product Nuclide + Emitted Particle
94
Mo + p+ -> 94m/gTc + n°
This concept takes the following mathematical form:
Epmax =Mc -2 {[Me Mi Ei ] 0.5 + [Me (Mc Q+Mi Ei )]0.5 } 2
Epmin =Mc -2 {[Me Mi Ei ] 0.5 - [Me (Mc Q+Mi Ei )]0.5 } 2
Epmax and Epmin are the maximum and minimum recoil energies of the product nucleus
(94mTc) in this case. A negative sign will indicate that the recoil is backward (does not
occur in our case). Following the same criteria the most probable recoil energy is the
kinetic energy of 94mTc and is expressed as
<Ep>=Mp Mi Ei Mc –2
Me = 939.56 MeV (mass of the emitted particle)

Mc = 174948.37 MeV (mass of the compound nucleus)
Mi = 938.27 MeV (mass of the incident particle)
Mp = 87476.31 MeV (mass of the product nuclide)
Mt = 87472.06 MeV (mass of the target nuclide)
Ei = 6 to 13 MeV (energy of the incident particle)
Production of 94mTc recoil energies for 6 to 13 MeV protons are tabulated below
49
Tab 1. Energy of the recoil ion in function of incident proton energy.
EI Epmin <Ep> Epmax

6 6 46E-05 0 063 0 244
7 7.91E-05 0.073 0.264
8 2.19E-05 0.084 0.283
9 4.14E-04 0.094 0.302
10 6.55E-04 0.105 0.321
11 9.36E-04 0.115 0.339
12 1.25E-04 0.126 0.357
13 1.59E-03 0.136 0.375
These values will be used to calculate the range of 94Tc in Mo. These ranges require the
use of fibers with diameters smaller than one micron. Using this data we evaluated the
thick target yield and a tentative geometry for the system.
Since the target material is expensive and not readily available, a totally experimental
approach is not feasible. Thus the system was simulated using FIBTAR (a computer code,
based on the Monte Carlo Method, that has been written specifically for fiber targets). The
code provides the total yield of the system, and the recovered yield (from recoil escape
and subsequent fluid region capture) of the radionuclide of interest. It predicted
successfully the behavior of previous fiber targets, with experimental results in good
agreement with the simulations. FIBTAR results will be used to design a prototype target
geometry to be experimentally tested. If the fibers contain any material other than the
target atoms, or if the target atoms are coated onto a support fiber, the influence of this
material on the target behavior must be investigated.
Technetium-94m produced by the 94Mo(p,n)94mTc reaction on natural molybdenum is

accompanied by several radionuclidic contaminants. Enriched 94Mo is available from
vendors in the USA and Russia at prices from $4 to $6 per mg. It is available with
different grades of enrichment, and as metal or oxide. The trioxide (94MoO3) can be
chemically dissolved and reoxidized for reuse as target material.
The impurities produced in the target influence toxicity of the product. Different types of
radioactive emissions will contribute to the dosimetry of the agent while not providing any
image information. Gamma emissions will contribute little to patient radiation dose
compared to β- and α emissions which lose all their kinetic energy in tissue. High energy
β+ will blur the image degrading the quality of the study.
The thermal characteristics of a target system are influenced by the chemical form of the
isotope to be used as the target material, its physical form, and the material used for
windows and target containers. Metallic Mo has one of the highest melting points of the
naturally occurring elements. Window materials may require special cooling provisions.
Excessive heat will affect the physical and chemical properties of the target.
Water and steam were first considered as stopping fluids. Heat deposited in the system
may cause water to be present in more than one physical phase, which affects the length of
the target. Water vapor behaves as an ideal gas between 480 and 1000 K. Helium or argon
(also candidate fluids) have no chemical influence on the 94mTc ion, but must be evaluated
for radionuclidic impurities produced by proton irradiation in the selected energy range.
50
The advantage of the noble gases is that they stay in the gaseous state, but they must be
mixed with enough water vapor to produce the pertechnetate ion. The chemical form of
the recovered product will determine if separation of undesired byproducts is necessary.
Technetium in the form of pertechnetate is chemically the most desirable form, since it is
the labeling agent used in pharmaceutical kits. The monographs of the US Pharmacopoeia
(USP) are the standards that establish the limits of radionuclidic contaminants, and the
amount of carrier allowed for human use applications of radiochemicals.
One published target system employed molten Mo, extracting the technetium ion by
sublimation. This method may be useful for fast recovery of the technetium ion, but the
target must be irradiated using a vertical beam (not usually available), and the efficiency
of the product recovery is problematic.
Results and conclusions
A target (porosity = 0.963) was modeled with the Monte Carlo code FIBTAR for 0.2
micrometer diameter Mo fibers in steam at a pressure of 6.8 atm and a temperature of 800
K. A Gaussian proton beam was collimated to a diameter of 7 mm before entering a target
body with inside diameter of 14 mm. Figure 3-1 shows the 94mTc yield as a function of
incoming proton beam energy up to 14 MeV (maximum energy for practical small
cyclotron applications). The 14 MeV total thick target yield at saturation is 74 mCi /µA as
compared to 117 mCi/µA for a pure Mo target as published by Rosch, et al (see reference
8, Chapter 2). This reduction in yield to 63 % is a reasonable concession to obtain the
recoil escape into steam necessary to recover the radionuclide product. The yield
recovered in the fluid is about 20 % of the total fiber target yield, which is reasonable.
Based on this promising result, an extensive heat transfer analysis was performed on fiber
target geometries.
Following the completion of the heat transfer analysis and preliminary design, it was
decided to extend the fiber/steam yields of Figure 1 to the maximum extent of cross
section data at 18 MeV (for completeness). As shown in Figure 2, the cumulative yield
showed a decrease above 14 MeV (not physically possible unless the target has an
excessive exit energy). In order to resolve this discrepancy, a much simpler analytical
model based on solid Mo compared to homogenous mixtures of Mo/fluid was undertaken.
A homogenous option of FIBTAR was also used for modeling the same cases. The
resulting comparison is shown in Figure 3, showing that the Monte Carlo modeling gave
good results for a solid target, but gave homogenous Mo/fluid results that were about ten
times too low, and also turned down above 14 MeV. Figure 4 shows an expanded version
of the homogenous Monte Carlo data, from which another discrepancy is obvious. The
various fluids vary greatly in magnitude, where the analytical model fluids are tightly
grouped as expected.
At this point, attention was directed back to the Monte Carlo Mo-fiber/steam modeling.
Volumetric heating as a function of distance into the target was calculated from FIBTAR
results as shown in Figure 5. Since stopping power increases as proton energy decreases,
we expected an increasing heating rate as proton energies decreased with penetration.
Since this was not the result, the same case was treated analytically for a homogeneous
51
target as shown in Figure 6. This shows heat generation increasing with penetration as
expected, and a target length for 14 MeV down to 2 MeV proton energy of about 45 cm.
The magnitude of volumetric heating is much less than the Monte Carlo fiber/fluid target,
which had a corresponding length of only 4.3 cm (Figure 5). Figure 7 shows the FIBTAR-
based heat generation result for a homogenous target, which indicates yet another
inappropriate combination of shape, magnitude, and depth. Figure 8 compares the widely
varying results of Figures 5 to 7. FIBTAR was reluctantly temporarily abandoned at this
point, with the conclusion that the submicrometer fibers and very high porosities required
by this design (not to mention the billions of histories required) will require extensive
code modifications that are beyond the scope of this thesis. FIBTAR had worked in the
past where fiber dimensions were in the several micrometer range for the nuclear reactions
and target materials involved.
At this point, the Monte Carlo modeling was replaced with and analytical model shown in
Figure 9, based on a unit cell with a 0.022 micrometer Mo layer followed by a 44.5
micrometer steam layer at 10 atm and 800 K (porosity = 0.9776). Target length is 45 cm.
This configuration was used to recalculate the heat transfer model, resulting in the radial
temperature profile shown in Figure 10. The 25ºC temperature rise from bulk cooling
water to the target centerline for a 1 µA beam is based on averaging the volumetric heat
generation rate over the length of the target. The target will easily tolerate a 10 µA beam,
which is more than adequate to produce enough 94mTc-pertechnetate for
radiopharmaceutical labeling. It is possible that the target pressure could be increased to
30 atmospheres, reducing target length to 15 cm. Fiber temperatures would be high
enough to vaporize liquid water pumped into the target through small orifices in the walls
and/or back. Using a HPLC pump (high pressure capability), water could be metered in at
a controlled rate, expanded into steam, and removed through a small bore teflon tube
(approximately 0.25 mm). The tube would provide an appropriate pressure drop to
maintain the target at high pressure while providing a recovery delivery flow of a few
hundred cm3/min without the use of a metering valve. Vaporization of the water by the hot
fibers would also result in considerable evaporative cooling via the change in state. If non-
enriched Mo or MoO3 fibers become available, this design approach can be evaluated
experimentally prior to beam testing.
In summary, the feasibility of an enriched 94Mo(p,n)94mTc target to deliver 94mTc-

pertechnetate on-line has been established. The detailed choice of fiber size for a prototype
target must await the revision of the FIBTAR Monte Carlo code to accurately model the
high-porosity submicron fiber/steam configuration needed to construct an efficient
prototype target for beam testing. As a parallel effort, the identification of an appropriate
fiber material fabrication process must be accomplished. These steps must precede the
construction and beam testing of a prototype target system.
52
Fig. 1. FIBTAR Monte Carlo model of fiber target to 14 MeV.
Fig. 2. FIBTAR Monte Carlo model of fiber target to 18 MeV.
Fig. 3. Comparison of solid Mo and homogenized Mo/fluid targets anlytically

modeled with TRIM, and Monte Carlo modeled with FIBTAR.
53
Fig. 4. Homogenized Mo/fluid targets, Monte Carlo modeled with FIBTAR (not
valid).
Fig. 5. Volumetric heat generation in a Mo-fiber/STEAM target Monte Carlo

modeled with FIBTAR (not valid).
Fig. 6. Expected volumetric heat generation in a homogenized target.
54
Fig. 7. Volumetric heat generation in a homogenized Mo/steam target Monte
Carlo modeled with FIBTAR (not valid).
Fig. 8. Comparison of volumetric heat generation in a Mo/steam target modeled

three ways.
Fig. 9. Thick target yields of solid Mo and homogenized Mo/fluid targets

analytically modeled with TRIM.
55
Fig. 10. Radial temperature profile for a Mo/STEAM target for 1 µA of 14 MeV
protons.
References
1. R.J. Nickles et al.94mTc Flow Agents: Bridging PET and SPECT. Proceedings of the
38th Annual Meeting. Society of Nuclear Medicine. Journal of Nuclear Medicine Vol
32. Number 5. May 1991.
2. R.J. Nickles et al. Quantitating Technetium Pharmacokinetics With PET, SPECT and
Beta Spectroscopy.
3. Wieland B.W. et al. In-target Production of [13N] Ammonia via Proton Irradiation of
Dilute Aqueous Ethanol and Acetic Acid Mixtures. Applied Radiation and Isotopes Vol
42 No 11, pp 1095-1098, 1991.
4. Rosch et al. Excitation Functions of α-particles induced nuclear reactions on

highlyenriched 92Mo: Comparative Evaluation Routes For 94mTc. Radiochemica Acta
68 13-20 (1995).
5. J.J.Hogan. The 96Mo (p,n) 96Tc Reaction from 10 to 65MeV. Journal of Inorganic and
Nuclear Chemistry 1973 Vol 35 pp1429-1433.
6. E.A. Skakum et al. Excitation Functions and Isomeric Ratios for the Interaction of
Protons of less than 9 Mev with Zr and Mo Isotopes. Sov Journal of Nuclear Physics 46
(1), July 1987.
7. R.J. Nickles et al. Technetium-94m Teboroxime Synthesis, Dosimetry, and Initial PET
Imaging Studies The Journal of Nuclear Medicine Vol. 34 No 7 (1993).
8. Rosch et al. Nuclear Data Relevant to the Production of the Positron Emitting
Technetium Isotope 94mTc via the 94Mo(p,n)-reaction. Radiochemica Acta 62 115-121
(1993).
56
9. Rosch et al. Thermo-chromatographic Separation of 94mTc From Enriched
Molybdenum Targets and Its Large Scale Production for Nuclear Medical
Applications. Radiochemica Acta 64 113-120 (1994).
10. B. T.Christian. Improving the Radionuclidic Purity of 94mTc for PET Imaging. Applied
Radiation and Isotopes VOL 46 No 2 pp. 69-73 (1995).
11. C.K.Stone et al. Technetium 94m-labeled Methoxyisobutyl Isonitrile: Dosimetry and

Resting Cardiac Imaging with Positron Emission Tomography Journal of Nuclear
Cardiology September/October 1994, Part I.
12. Dunn WL et al. Monte Carlo Analysis of Microfiber Targets for Production of 15O-
ozone. 3rd Topical Meeting on Industrial and Radioisotpe Measurements and
Application, Raleigh NC October 6-9 1996, abstract book p.109.
13. Ziegler JF. TRIM, the TRansport of Ions in Matter (version 95.xx, Jan 9, 1995),
distributed by JF Ziegler, IBM Research 28-0, Yorktown, NY 10598.
14. A.G.Jones.(1997). Technetium in Nuclear Medicine, Radiochemica Acta 70/71 289-

297
15. R.J.Nickles et al. A Vertical Beam Facility for The Irradiation of Molten Targets.
Proceedings of the 38th Annual Meeting. Society of Nuclear Medicine. Journal of
Nuclear Medicine Vol. 32. Number 5. May 1991.
16. R.D.Evans. The Atomic Nucleus. Mc Graw Hill New York 1955.
17. M. Iwamoto et al. Distribution of Forward Recoil Range of 18F Formed by Various
Nuclear Reactions. Radiochemica Acta 1982 30, pp. 73-77.
18. R.Gautreau and W. Savin. Modern Physics Mc Graw Hill. 1995.
57
Gravity Can Offer Help: Stable Production of C-10 by Bombarding
a Molten Sea of Boron Oxide with a Vertical Proton Beam.
M. Jensen1, G. Wolber2, J. Nickles3 and A. Runz2
1
Royal Veterinary and Agricultural University, Frederiksberg, DK
2
Deutsches Krebsforschungszentrum, Heidelberg, Germany
3
University of Wisconsin, Madison, WI, USA
Why C-10
C-10-CO2 has shown improved performance over 15O labelled water in the rendering of
brain and kidney perfusion (I. Law et al, J. Cereb.Blood Flow Metab, 21, 1003-1012, M.
Jensen et al, J Nucl Med, 41, 1363) but the actual use has been hampered by the long term
instability of the target. The molten glass of enriched boron oxide slowly creeps out of the
beam strike area under the action of gravity when using standard horizontal beam
geometry. We have set out to investigate the use of a vertical beam on a molten horizontal
target for C-10-CO2 production.
How to make C-10
The production method and cross section are described by Alves (JARI 52,899ff). The
method mimics old fashioned boron 11C production, but in our case target temperature and
release conditions have to be optimised because of the much shorter half life. Titanium
was chosen for the crucible because of the high melting point and stability of the oxide
(Common experience to anyone having tried to hard solder Ti, even using B2O3 as flux).
Methods
300 mg of B2O3 enriched to 99.5 % in 10B was melted as a homogeneous glassy disc into
at Titanium crucible with an internal diameter of 14.9 mm, giving a target thickness of 172
mg/cm2. The crucible was positioned in a stainless steel target canister with an internal
volume of 90 cm3. The target was flushed with 99.996 % pure Helium with a flow of 200
cm3/min through a CuO converter CO-->CO2 at 400°C and an NOX trap (T. Tewson,
JARI 40,765ff), carrying the C-10-CO2 through 10 meters of 1/16” Teflon tubing to an
Ascarite trap mounted in a dose calibrator. Vertical proton beams at 15, 21 and 28 MeV
were incident on the Boron Oxide surface. The beam diameter and profile was initially
inspected with a beam viewer showing aprox. 10 mm FWHM diameter. This beam was
collimated down to 8 mm directly in front of the target, yielding an almost uniform
circular beam about 8 mm diameter.
58
Power balance
Only the target entrance window flange was water cooled. The target housing was allowed
to heat up during bombardment until radiative and convective heat loss from the target
canister surface (300°C?) balanced beam power input. A deliberately poor thermal contact
between target crucible and target body allowed the radiative stabilisation of target surface
temperature. Assuming black body cooling conditions the crucible reached 1500°C at our
maximally tested beam power of 420 W (20 microA @21MeV).
10-CO2 Steady state
3000
21.5 MeV
15.2 MeV
2500 28.2 MeV
Steady state activity (MBq)
2000
1500
1000
500
0
0 5 10 15 20 25
Beam current (uA)
Yield
Release of C-10-CO2 is a very steep function of beam power. Little variation was seen in
yield across the tested energy range, consistent with the data of Alves, JARI 52, 899ff and
the fact that the present B2O3 target is "thin" to the vertical beam (incident 21 MeV
protons leaves at 17 MeV). The observed Ascarite trap yields were all decay corrected for
a total of 70 sec delay (30 sec hold up in the target, 40 sec delay in converter, trap and
tubing).
The best performance was obtained with 21.5 MeV @12 microA, where we found 46 %
of the theoretical yield calculated from the Alves data. Both target and CuO converter
temperatures were probably on the low side of optimum, and more improvement could be
expected. The target was inspected after about 2 hours bombardment at currents between
10 and 20 microA. The Boron glass had lost less than 2 mg of mass. The glass still formed
an integral disc in the crucible.
59
Optimisation Studies on the Production of High-Purity I-124 and
I-120g at a Small-Sized Cyclotron
S.M. Qaim, A. Hohn, G. Blessing, Th. Bastian, K.M. El-Azoney, S. Spellerberg,
B. Scholten and H.H. Coenen.
Institut für Nuklearchemie, Forschungszentrum Jülich GmbH, D-52425 Jülich,

GERMANY
Aims
The radionuclides 124I (T½ = 4.18 d) and 120gI (T½ = 1.35 h) are interesting positron
emitting analogues of the therapeutic radionuclide 131I (T½ = 8.0 d) and the single photon
emitting diagnostic radionuclide 123I (T½ = 13.2 h), respectively. It has been shown that
both 124I and 120gI can be produced at a small-sized cyclotron. However, several production
parameters needed detailed investigations. The aim of this study was to optimise those
parameters to be able to produce the two radionuclides in high purity.
Methods
The TeO2-target technology and dry distillation method of radioiodine separation were
studied. A target system involving He cooling in front and water cooling in the back was
developed for medium scale production. The removal of radioiodine was studied
differentially as a function of time. The integral loss of TeO2 from the target was
investigated as a function of oven temperature and time of distillation. A determination of
the positron emission intensity in the decay of 124I was done using a very pure source and
comparing the intensities of the annihilation and X-ray radiation. Production of 124I was
carried out using 200 mg targets of 99.8 % enriched Te-124-O2 on Pt-backing, 16 MeV
proton beam intensities of 10 µA, and irradiation times of about 8 h. For the production of
120g
I only a thin 30 mg target of 99.9 % enriched Te-120-O2 was available. Irradiations
were done with 16 MeV protons for 80 min at beam currents of 5 µA. The radiochemical
purity of both 124I and 120gI was checked via high performance liquid chromatography
(HPLC), thin layer chromatography (TLC) and a few known iodinating reactions.
Regarding the chemical purity, the Te content in radioiodine was determined
spectrophotometrically.
Results
The TeO2-target withstands 16 MeV proton beams up to 15 µA. A distillation time of 15

min at 750°C was found to be ideal both from the viewpoint of radioiodine removal and
TeO2 loss. The positron emission branching in 124I was exactly determined as 22.0 +/- 0.5
%. The batch yield of 124I at EOB was 550 MBq (15 mCi). At the time of application the
radionuclidic impurity 123I (T½ = 13.2 h) was < 1 %. The levels of other impurities were
negligible (126I < 0.0001 %; 125I = 0.01 %). Special care was taken to determine the 125I
impurity via X-ray spectrometry. The 120gI batch yield achieved at EOB was 700 MBq (19
60
mCi), and the only impurity detected was the isomeric state 120mI (T½ = 53.0 min) at a
level of 4.0 %. The radioiodine collected in 0.02 M NaOH solution existed > 98 % as
iodide. The amount of Te in all the radioiodine samples was found to be < 1 µg.
Conclusions
High purity 124I and 120gI can be advantageously produced on a medium scale using the
low-energy (p,n) reaction at a small-sized cyclotron. The yield of 120gI can be considerably
increased by using a thicker taget.
61
The Present Status of a High Current Inclined Target and Solid
Target Changer at Washington University
D.J. Rowland1, R. Laforest1, T.J. McCarthy1, M.J. Welch1, B.J. Hughey2, R.E.
Klinkowstein2 and R.E. Shefer2
1
Washington University in St. Louis, Mallinckrodt Institute of Radiology, 510 S.
Kingshighway Blvd., St. Louis, MO 63110, USA
2
Newton Scientific, Inc., 245 Bent Street, Cambridge, MA, 02141, USA
Aims
Solid targets have gained great importance in nuclear medicine with the production of
novel isotopes such as 76Br, 77Br, 124I, 94mTc, 45Ti and various copper isotopes. With
limited beam time available on the CS-15 cyclotron to produce these isotopes, an
automatic solid target changer for the JSW cyclotron has been designed and fabricated to
be used in conjunction with a recently installed conveyor system for remote target
retrieval.
An inclined target design has been described previously for use in the production of
halogen isotopes (1). In-beam tests have been performed with this target to study the target
material properties such that the maximum beam current could be determined. Present
disk targets for the production of Br and I allow for irradiation up to a maximum of 5
microAmps of beam current.
Methods
The target changer comprises a target cassette that can hold up to six disk targets, or two
disks combined with the inclined target described above. Target loading and unloading is
under computer control, comprising a user-initiated predetermined sequence of steps. The
control software includes multiple safety interlocks to ensure reliable target loading,
irradiation, and unloading onto the conveyor. We anticipate testing the target changer on
the JSW cyclotron in the next several months.
The inclined target (20 degrees) was prepared with [76Se]Cu2Se material by placing 70.8
mg of material into a Pt-coated depression of a tungsten target. The material was heated in
an induction furnace to melt and adhere it to the platinum. The target was irradiated on
four occasions at 2, 6, 10 and 13 microAmps for 10 minutes. The activity was monitored
over several days to determine the amount of 76Br produced.
Results
The production rate measured from the irradiations is 1.1+/-0.2 mCi/microAmp hour for
76
Br. This production rate is about 35 % of the predicted value from published cross-
section data assuming a uniform spread of the target material over the inclined surface.
62
Additionally, the measured yield decreased monotonically with current. Experiments are
planned to investigate the source of these discrepancies.
Conclusions
The results of the target irradiation show that with this target design, higher beam currents
will be permissible compared to the conventional disk targets that are currently being used
for the halogen isotopes. Further irradiation will be performed in order to determine if the
beam current can be increased further.
This research was supported by NIH grant (R24CA86307), and by Department of Energy
grants (DE-FG-02-84ER60218) and (DE-FG-02-97ER82442).
References
1. Rowland, D.J. et al., Appl. Radiat. Isot. 44: S1059 (2001).
63
The Production of I-121 at the Vertical Beam-Line of the MC32NI
Cyclotron at the DKFZ Heidelberg
Armin Runz1, Gerd Wolber2, Michael Eisenhut1 and Wolfhard Semmler2
1
Department of Radiochemistry and Radiopharmacology
2
Department of Biophysics and Medical Radiation Physics, German Cancer Research
Center (DKFZ), D – 69120 Heidelberg, GERMANY
Introduction
Radionuclides of iodine have been widely used in nuclear medicine as tracers in dia-
gnostics, therapy and research. Positron emitting iodine isotopes, i. e. 124I (T1/2 = 4.15 d)
and 121I (T1/2 = 2.12 h) have been requested by our PET group for labelling of specific
carriers used in investigations of molecular mechanisms with short- and long-term
dynamics. First work on 121I was done on enriched 122TeO2 in our institute using the
former AEG compact cyclotron (22 MeV p, 11 MeV d) many years ago (Helus et al.
1979). This compound as well as metallic Te which we are using at present is prone to
melt and may run off a perpendicular target under proton bombardment. Hence, the yield
may vary substantially. We built a vertical beam-line at our Scanditronix MC32NI
He
He + I
Thermo couple Heater
Fig. 1. Design of the vertical beam-line (upper

left) with 90° deflecting magnet (Scanditronix
Magnet AB); sketch and picture of target body
with water cooled entrance flange (right);
64
cyclotron (32 MeV p, 16 MeV d) where meltable samples can be irradiated in horizontal
position without the risk to escape from the beam spot (Fig. 1).
Aim
This feasibility study aimed at collecting data and experience in optimizing the target and
procedures. Ultimate goal is to achieve reproducible yields over long irradiation times at
elevated beam currents.
Fig. 2. Decay curves taken with dose rate

1000
decay monitor of the 17.7 and 21.5 MeV protons
on 128Te/122Te mixture (left).
Temperature dynamics in the target body
doserate / [µSv/h]
1 7 ,7 4 M e V
(lower left).
100
Target crucible showing beam spot with
covering net aside. This carries some small
2 1 ,5 6 M e V
crystals of sublimated Te (lower right)
10
60
0
120
180
240
300
360
420
480
540
600
660
720
780
840
900
960
1020
1080
target heating
500
450
temperature / °celcius
400
350
300 Bottom: platinum crucible
250
200
150
100
50 Top helium inlet
0
0 10 20 30 40 50 60 70 80 90
time / minutes
Target design: The target housing is made from stainless steel (Fig. 1). It contains a
platinum crucible filled with metallic Te covered by a platinum net to prevent loss of
material by sputtering (Fig. 2). Te has a lower melting point (452°C) and higher atomic
density compared to TeO2 (m.p. 760°C). The target bottom is warmed up to 400°C by a
temperature-controlled heater and the beam current. At operating temperature the upper
part of the target assembly acquires about 220°C, just above the sublimation point of
iodine (185°C).
Radiochemistry: The volatile iodine is flushed to a NaOH receiver or a cold trap by warm
helium, which prevents oxidation of Te during bombardment. To save cost, we used a 50
mg/1000 mg mixture of 99.6 % enriched 122Te/128Te (euriso-top) in preliminary
experiments and recalculated the yield to 100 % 122Te. 121I is probably produced most
effectively via 122Te(p,2n) in analogy to the 125Te(p,2n)124I reaction investigated recently
by Hohn et al., 2001. The proton energies on target were set to 17.7, 21.5 and 26 MeV
around the expected cross-section maximum.
65
Results
• Variation of the He flux showed a distinct maximum at 40 ml/min.

• At 17.7 MeV on target , 1 µA the yield at EOB was approximately 3.2 mCi/µAh;
at 21.5 MeV, 1 µA the yield rose to 9.4 mCi/µAh inline with earlier results (F.
Helus 1979)
• The platinum net covering the crucible turned out to be very essential to minimize
losses by evaporation of Te (Fig. 2).
Outlook
Next steps will focus on optimizing the energy window with respect to yield and
radionuclidic purity and the estimation of the radiation burden from the daughter nuclide
121
Te (EC, 16.8 d).
References
F. Helus et. al., Radiochem. Radioanal. Letters 39/1 (1979), 9 – 18.
A. Hohn, F.M. Nortier, B. Scholten, T.N. van der Walt, H.H. Coenen, S.M. Quaim,
Applied Radiation and Isotopes 55 (2001) 149 – 156.
66
Subliming Activity from White Hot Targets. A General Method,
Tested and Rationalized by 10C Production
M.J. Schueller1, R.J. Nickles2, A.D. Roberts2, T.E. Barnhart2 and M. Jensen3
1
Brookhaven National Lab 1, Upton NY
2
University of Wisconsin, Madison, WI
3
Royal Veterinary and Agricultural University, Frederiksburg, DK
Introduction
Release of product radioactivity during bombardment is often regarded as one of the many
perils threatening the target chemist. In fortunate cases, this release can be turned in to
good use, not only providing steady state supply of ultra-short lived radionuclides beyond
the reach of batch processing, but also offering elegant and maintenance-lean possibilities
of in-target separation. Among the mechanisms used for release, heat is by far the most
popular method. From early internal target probes with activity detection on the cyclotron
fore-pump outlet to the state of the art targetry for on-line isotope-separators, the method
is perhaps more used than it is understood.
Theory
Transport and release of a tracer formed in the interior of a solid target matrix follows an
Arrhenius temperature relationship, dominated by diffusion as characterized by D = Doexp
(-Ea/kT). Charged particle reactions deposit the initial radioactive product distribution into
the depth of the solid target matrix. This profile follows the energy dependence of the
excitation function σ(E(x)), from the surface to a Q-dependent depth where the cross
section goes to zero, generally a few hundred microns in. This diffusion path length to the
surface, xo, represents a diffusion time t of the order of xo2/D, greatly favoring shallow
grazing angles or finely divided powder targets and incandescent temperatures. The decay
losses during this diffusion voyage results in the log-log relationship between the activity
arriving at the surface A and the initial activity Ao deposited at depth, xo such that:
(1) ln [Do/λxo2] + ln ln (Ao/A) = Ea/kT.
At the surface, a sweep gas can entrain the sublimed activity for tuned pipeline transport
to the distant collection sites. The initial activity Ao is proportional to the neutron counting
rate, an easily observable measure of the true target nuclear yield. The recovered activity
A becomes a steep function of target temperature, often tempting the designer to use the
beam for target heating. Typical target performances for the known examples of heat-
release (p,n) reactions are listed in Table 1 below, with expected yield Ao at EOSB on 100
% enriched elemental material.
Correspondence to RJ Nickles at nickles@petrus.medphysics.wisc.edu
67
Tab. 1.
Reaction t1/2 Target mp (oC) Ao(EOSB)

10
B(p,n)10C 19 sec 10
B2O3 473 10 mCi/µA @ 11 MeV
11
B(p,n)11C 20 min nat
B2O3 473 93 mCi/µA @ 11 MeV
12
C(d,n)13N 10 min C 3400 (subl) 25 mCi/µA @ 7 MeV
76
Se(p,n)76Br 16 hr 76
Se 217 37 mCi/µA @ 11MeV
124
Te(p,n)124I 4.2 d 124
TeO2 733 36 mCi/µA @ 11MeV.
We have observed the above derived relationship (1) for thermally driven target release in
the case of ultra short lived (t1/2 = 19 sec) 10C formed in molten boron oxide glass. The
target temperature in the beam strike was monitored by the use of an IR thermocouple
(Exergen IRtc.3AMF), with germanium optics (helium-swept target window, lenses) to
pass the necessary 8-12 µm wavelengths. The extraction efficiency e = A/Ao is
proportional to the ratio of the trapped activity (eg. 10CO2 in soda lime) to the measured
neutron counting rate. If ln ln 1/ε is plotted against 1/T , the proposed relationship should
yield a straight line, as shown in Fig. 1 below.
In spite of the scatter due to crowbars and fluidic irregularities in the boiling 10B2O3 melt,
the log-log relationship is evident, with a steep slope indicating a diffusion enthalpy that
suggests bulk convection effects, such as noted by Beyer and Pimentel (Radiochim Acta
88, 175 (2000)).
Harvesting radionuclides from incandescent solid or molten targets poses a set of serious
challenges to the target chemist. Their successful resolution will provide new feedstock
for applications ranging from PET imaging to radioactive ion beams for nuclear
68
astrophysics. A theory allowing a first-principle calculation of the necessary operating
point is the now available.
Methods
Carbon-10 dioxide is used as an example of fast sublimation because of its potential

importance in the PET imaging of regional blood flow (I Law, et al: J Cereb Blood Flow
Metab 21, 1003-1012 (2001)). The yield from the 10B(p,n)10C reaction (Alves et al: Int J
Appl Rad Isotopes 54, 899-903 (2000)) favors high proton energy and thick targets of
elemental 10B, while successful sublimation from the target substrate strongly favors the
use of the molten boron oxide. To maintain the molten glassy target material in a
horizontal beam (see Jensen et al: Gravity can offer help: Stable production of 10C by
bombarding a molten sea of boron oxide with a vertical proton beam; these Proceedings).
the enriched boron oxide is sintered onto a woven mesh angled at 30o to the horizontal.
Mesh materials include rhenium, molybdenum, Monel and 304 stainless steel, ordered in
preference, based on their refractory nature and inertness to the melt. The mesh, spot
welded to a stainless support tube, is thermally cantilevered away from contact with the
water-cooled target housing, shown in Figure 2, in order to allow its surface temperature
to reach white heat with the dissipation of several hundred watts of beam power. The
helium gas flow enters upstream of the beam strike, serving to clear any condensate from
a germanium window, coated to pass the 8-12 micron infra red on to a focussed IR
thermocouple that monitors the surface temperature of the melt. The IRtc was calibrated
by simultaneous measurement against a conventional thermocouple in the beam. The
helium stream then passes, or around, through the molten boron oxide and leaves the
target through a 1/16" SS tube (0.020" ID) for a 10 second tranport to the traps and
furnaces of the chemical processing station, shown in Figure 3.
Fig. 2. Target chamber for the production of 10CO2.
69
Figure 3. Schematic of the flow-through chemical processing of 10CO2.
First the contaminating 13NOx is removed by an indicating trap of silica gel, treated with
CrO3 + CuSO4 (T Tewson et al: Int J Appl Rad Isotopes 40, 765-768 (1989)). The
approximately even split between 10CO2 and 10CO is evident in gas chromatography as
well as the trapping efficiency of the activity in soda lime. For this reason, the CuO
furnace at elevated temperature insures that the label will be in the form of 10CO2 for
patient administration. Recent work has succeeded in cryotrapping the 10CO2 for pulsed
inhalation, as well as trapping the majority of the activity in 0.155 mM NaOH across a
dialysis membrane for subsequent neutralization to injectable saline. This work presages
studies to determine whether a spike input may be preferable to the steady state
administration of 10CO2 in cerebral activation studies.
As an independent investigation of the sublimation of labeled CO2 from boron compounds

at high temperatures, ballistic thermochromatography was performed on proton-irradiated
natural boron compounds. After bombardment, the elution of 11CO2 into a helium stream
was followed, as the temperature was ramped linearly from room temperature to 1100o C
in a muffle furnace. The release of 11C-activity was dramatic for B2O3 and its hydrated
form H3BO3 at 280oC, but barely perceptible for the refractory elemental boron, BN, B4C3
and AlB2. This behavior was fit to a STELLA (High Performance Software,Lyme, NH)
model searching for the diffusion constants, but confounded by imperfect knowledge of
the grain size distribution.
Conclusions
Nearly eighty years ago, the pioneering cyclotron chemists (Calvin et al: Isotopic Carbon,
John Wiley and Sons, New York, (1949)) were boiling 11CO2 away from B2O3 cyclotron
targets. Today, new radioisotopes enter new applications, ranging from the basic to the life
sciences. Nuclear engineers study radiation enhanced diffusion that embrittles fusion
reactor walls. Nuclear astrophysicists coax short-lived radionuclides out of primary targets
become radioactive ion beams, mimicking the stellar interior of supernovae. For the
biological scientist, Table 1 lists just a few of those candidates for which ready access
from incandescent targets would be welcome.
70
Some Aspects on Tellurium Targets for Iodine-123 Production
L.M. Solin, B.K. Kudelin and V.A. Jakovlev
V.G. Khlopin Radium Institute, 2nd Murinsky Av. 28, 194021 St. Petersburg, Russia
It is well known that 123I is one of the best radionuclides for medicine. V.G. Khlopin
Radium Institute cyclotron MGC-20 gives protons with the energy of 2-18 MeV. At these
conditions we can use only proton irradiation of 123Te to produce iodine-123. Tellurium-
123 dioxide used to prepare the target is not a convenient material. It is not firm under
proton bombardment, it takes special measures to attach to the platinum background, it
remains radioactive a long time after irradiation; besides, it should be of high enrichment
on 123Te. For more than 10 years we have tried to investigate some questions, part of
which may be of interest for iodine-124 production too.
Target preparation
Tellurium target is prepared by a well known method of tellurium dioxide melting on a

platinum plate. The key point here is attachment of tellurium dioxide to platinum. It is
very important to have a high chemical purity of the starting material. Also very important
to make a background surface rough enough. Some authors scratches the surface, others
makes special profile of the plate bottom. We are preparing background surface by acid
(hydrochloric and sulphuric) pickling, and we try to use not new platinum foil for
background.
Target regeneration and purification
Enriched tellurium-123 is rather expensive material, so it is important to collect all losses

of tellurium (from extraction device, after accidents with the target) and to recover
tellurium-123. Tellurium regeneration and purification method is based on TeO2 volatility
at high temperature. Technology includes stages of evaporation and dissolving [1]. The
procedure of purification allows to obtain high quality TeO2 suitable for iodine-123
production. Tellurium losses during purification are 3-5 %.
Radionuclide purity dependence on tellurium-123 enrichment
It is well known that the highest radionuclidic purity for iodine-123 may be achieved by
proton irradiation of Xenon-124 as a target. We have made efforts to improve radionuclide
purity for iodine 123 produced via proton irradiation of tellurium-123. We have started
our activity with tellurium-123 enrichment >95 %. Then we employed starting material
more high quality – 98.5 % and 99.3 % enrichment. For tellurium-123 enrichment of more
than 95 % there is only impurity of iodine-124. On the face of it, there is no limitation to
get radionuclidic purity (RNP) close to 100 % with the increasing tellurium-123
71
enrichment. Actually, it is easy to show that there is a relation between RNP and content
of tellurium-124 in the target:
A123 A
RNP = × 100% ≈ (1 - 124 ) × 100%
A123 + A124 A123
where A124 is iodine-124 impurity activity and A123 is iodine-123 activity.
In the case of tellurium target irradiation during t hours at the proton energy of Ep we have
at the end of bombardment (EOB):
A124 C124 (1 − exp(− λ124 × t))

= × × F(E p )
A123 C123 (1 − exp(− λ123 × t))
λ124, λ123 – are constants of decay for iodine-124 and iodine-123,

F(Ep) – is a function of proton energy
For t=1 hr
A124 C124
= × 0.136 × F(E p )
A123 C123
And for Ep constant F(Ep) = F
A124 C124
= × 0.136 × F
A123 C123
Fitting to experimental data gives F ≈1.0. This formula shows: diminution of tellurium-
124 content should lead to the decrease of iodine-124 impurity. But it appeared that the
decrease of the tellurium-124 content in 50 times decrease the impurity only in 15 times.
We can propose that there is another reaction for iodine-124 production - 123Te(p,γ)124I [2].
An estimation of the RNP limitation reached by diminution of tellurium-124 content gives
us 99.983 (for Ep = 15 MeV, irradiation time of 1 hr and tellurium-123 enrichment of 99.3
%). Thus it is not necessary to get tellurium-123 with very high enrichment for iodine-123
production.
72
Target activitation
It is important to investigate target activation under particle beam bombardment. It is

especially important in the case a manual target operation (without manipulators) to
reduce occupation dose. There are three stages of iodine-123 technology where target
activity is essential.
First Iodine-123 production technology demands manual positioning of the irradiated

target into the extraction tube.
Second The same target is usually used for iodine-123 production for many times. So it
will accumulate activity in spite of iodine extraction and manual target
positioning into the target device becomes dangerous.
Third Telurium-123 dioxide is an expensive material. During proton irradiation its
purity may by deteriorated by ion implantation for instance (aluminum atoms and
clusters knocked out by beam from window before target) and target may have
mechanical damage. Moreover small amount of target material is evaporated into
the iodine extraction device every extraction. That material may by gathered from
the device and used for target preparation again. It is preferable that target
material regeneration and purification, as well as following target preparation are
realized at low activity.
We have investigated activities of targets used to produce iodine-123 via proton

irradiation of tellurium-123 dioxide. The main results are:
1. Target activity after iodine extraction is mainly conditioned by gamma-radiation of

aurum isotopes and of tellurium isomers. Aurum isotopes Au-194, 195, 196, 198
are produced through natPt(p,n) reaction; tellurium-123m is a result of inelastic
proton scattering; tellurium-121m is obliged to 122Te(p,d) reaction.
2. Right after irradiation backing activity is obliged to 194Au.
3. Right before target positioning in the target device target activity is determined by
decay of isotopes produced in the target during previous irradiations. At periodic
irradiations there are different contributions to the total activity from tellurium
dioxide and from separate backing isotopes depending on frequency of irradiations.
4. When the target is "cooling" for more than 6 weeks its activity is being determined
by 195Au and 123mTe independently from the "history".
References
1. Kudelin BK, Gromova EA, Gavrilina LV and Solin LM Use of vacuum evaporation for
tellurium dioxide purification in production of iodine-123. In Proc. 3rd International
Conference on Isotopes, Vancouver, Canada, 6-10 September, 1999, Vancouver,
Canada, 321.
2. Solin LM, Jakovlev VA, Kaliteevsky AK, Godisov ON, Mjazin LP and Shepelev PK
(1998) High purity iodine-123 produced via p,n reaction. In: Cyclotron and their
Applications 1998, Proc. 15th International Conference, Caen, France, 70-73.
73
Remote and Automated Solvent Extraction of Gallium-66
L. Tang, D.E. Reichert, D.W. McCarthy and M.J. Welch
Washington University Medical School, St.Louis, MO 63110, USA
We are now producing a number of non-conventional positron emitting nuclides for

potential use in nuclear medicine at Washington University. The positron-emitting
radionuclide 66Ga (t½ = 9.49 h; ß+ 56.5 %; EC 43.5 %) can be used as a substitute for the
photon emitter 67Ga in a large number of labeled compounds. Gallium-66 has potential for
radiopharmaceutical labeling in PET imaging and for targeted radiotherapy of cancer. The
production of large quantities of 66Ga requires the development of a remote/automated
system for the routine processing of large quantities of this radionuclide.
A solid target holder designed and built for the cyclotron production of 64Cu [1] is used to
hold a thin zinc foil. This foil is mounted on a 1-cm diameter threaded platform in the
center of an aluminium target disk and held in place with a 1-cm diameter aluminum
screw cap with a 6-mm diameter window in the center. A depression in the back of the
disk provides efficient cooling during irradiation with water flowing through it. 66Ga is
produced by the 66Zn(p,n)66Ga nuclear reaction, by irradiating this thin zinc foil with
14.5MeV protons using a Cyclotron Corporation CS-15 cyclotron.
Fig. 1. Solid Target Holder and Foil Target Holder
Gallium-66 is purified by an ether extraction method. Briefly, the irradiated zinc target
material is dissolved in 12 M HCl and subsequently diluted to 7 M HCl. The diluted
sample is extracted three times into diisopropyl ether. The organic phases are combined
and washed once with a small amount of titanium chloride to reduce the Fe3+ ion
contaminants. A second wash with 7 M HCl removes any traces of Zn and Cl ions.
Finally, 66Ga is recovered by three back-extractions into milliQ water. The final product is
evaporated to dryness and resolubilized into 150 µL 0.05 M HCl. In the work by Lewis et
al. [3] this purification procedure was shown to be superior to purification by cation-
exchange chromatography in terms of processing time, concentration of stable
contaminants and radiopharmaceutical labeling yields. We have now designed a remote
system to perform these purification steps.
The inherent differences in the electrical conductivity of each layer of the biphasic system
are the basis for our automated extraction system. Two electrodes immersed in the
biphasic solution are capable of detecting the presence or absence of the highly conductive
74
aqueous medium. Two gold wires extending to the bottom of a conical-bottom plastic tube
are used as the electrodes. Using a digital voltmeter as a preliminary setup, the transition
between the two-phase aqueous/ether mixture to a single-phase ether solution can be
detected during the course of multiple back-extractions. Preliminary data collected, using
a 9V battery, has shown that the potential difference between the aqueous and organic
layers is sufficiently large to detect the end point of extraction. Mosaic Industries’
(Newark, CA) components (“Quick Easy Design” QED package, QED boards and stepper
motors) have been selected to implement the remote extraction of the different layers. A
touchscreen front panel system, the Panel-Touch Controller™, with a QED-Flash board
has been chosen to interface with four small stepper motors mounted on two printed
circuit boards. This powerful real-time controller includes a graphical user interface, 48
analog and digital I/O lines, eight 12-bit A/D inputs. Two digital I/O boards control the
motors. These four stepper motors activate the syringes for liquid displacement. The QED
board and two small resistors are set to read the voltage. Built-in software is being used to
specify the different parameters during extraction.
Panel-Touch Controller™
QED-Flash Board
Fig. 2. Schematic of the Automated System. Fig. 3. Mosaic Industries Components.
• the first 7 M HCl extract: between 8.59 to 9.47 V;

• the titanium chloride extract: between 8.65 to 9.48 V;
• the 7 M HCl wash extract: between 8.70 to 9.18 V;
• the milliQ extract: between 8.80 to 9.31 V;
• the organic layer: between 0.46 to 6.76 V.
We are currently making progress on the

electrolytic deposition of Zn material on a
Fig. 4. Preliminary Data – Experimental Potentials.
75
solid target. This will permit the recovery of this target material. The electrolyte for Zn
plating is a ZnCl2 solution at pH = 5-6, which has a small amount of hydrazine hydrate
(NH2NH2) added as a depolarizer. Final Zn concentration of the electrolyte is ZnCl2 = 0.46
M. Approximately 7-8 mL of electrolyte is needed to plate an area of φ10mm on a gold
target and approximately 2-3 mL of electrolyte is needed for an area of φ5mm. Following
a method from P. Van den Winkel et al.[2] constant current electrolysis technique is used
to deposit the target material at a constant rate. A single bipolar chopped sine wave is the
desired waveform for the zinc electrodeposition. Preliminary plating experiments using
natural Zn target material on large and small solid gold disks showed promising results:
the electroplated material was evenly distributed with no obvious dendrite formation, of a
shiny metallic coloration and appeared to be well adhered to the substrate.
Control Box (Newton Scientific Inc.,

Cambridge, MA, USA):
Risetime
Gain (V → I converter)
Function Generator
(BK Precision 4011 A):
Bipolar Sine-wave
fSW = 100 Hz
Sweep / Function generator

(BK Precision 4040 A):
Square-wave
fCH = 1000 Hz
Scopemeter Plating cell
(Fluke 123)
Fig. 5. Electroplating Setup.
anode
• Bipolar chopped sawtooth (fst = 100
Hz)
• Chopper frequency (fch = 1000 Hz)
cathode • Net cathodic constant current
(160.2 mg/cm2, 4 h plating)
Fig. 6. Chopped Sine wave. Fig. 7. Plated Zn target.
76
To test the adhesion between the gold disk and the zinc layer, the above plated Zn target
(figure 7) was irradiated for 5 minutes at 10 µA. Its appearance remained unchanged.
However at a higher proton beam current of 15 µA during a 5 minutes irradiation, the
plated materiel volatilized. The porosity of the electroplated Zn material needs to be
reviewed for higher beam current.
In conclusion, preliminary results show that using the ionic strength of aqueous layers for
the solvent extraction is a dependable way to identify a good separation of phases. The
combination of voltage level, with sufficient duration between measurements, provides a
reliable method to delimit the overall layer endpoint during the extraction process. Careful
implementation of remote and automated processes will reduce the amount of mechanical
and quality control variation and decrease the radiation dose to personnel.
Acknowledgments
The authors would like to acknowledge the assistance of Xiankai Sun, PhD. This work was
supported by grants from Research Resource NIH/NCI R24 CA86307 and DOE Grant No.
DE-FG02-87ER60512.
References
1. McCarthy, DW et al., Nucl. Med. Biol. 24, 35-43 (1997).
2. Van den Winkel, P. et al., Personal Communication, Vrije Universiteit Brussel,

Cyclotron Department, Brussels, Belgium, Description of the procedures for the
electroplating of 68Zn targets, (1995).
3. Lewis, MR et al., Proceedings of the 8th Workshop on Targetry & Target Chemistry,
120-123, (1999).
77
Preparation of Carbon-13 Target Material from C-13-Benzene
Elizabeth M. Zippi and Amy Pilcher
Department of Chemistry and Physics, Louisiana State University in Shreveport,

Shreveport, LA 71115, U.S.A.
Summary
In an effort to prepare an improved carbon-rich target material for the accelerator

production of N-13-labelled ammonia, pyrolyzed C-13-enriched sulfonated
poly(styrene/divinylbenzene) has been under investigation. The preparation of this target
material using naturally abundant carbon-12 was investigated in order to optimize
conditions for the preparation of the analogous carbon-13-target material which may
provide a cost-effective method for producing nitrogen-13 via proton irradiation. Porous C-
13-enriched poly(styrene/divinylbenzene) was synthesized in four steps starting from C-
13-benzene. Sulfonation and pyrolysis of the copolymer has been accomplished to afford a
carbon-13-rich target material.
Introduction
For the purpose of improving nitrogen-13 production for use in medical imaging,
pyrolyzed C-13- enriched sulfonated poly(styrene/divinylbenzene) has been under
investigation. In order to prepare the C-13-enriched copolymer, the corresponding styrene
monomer was synthesized in three steps starting from C-13-benzene in approximately 50
% yield1 and polymerized by emulsion polymerization using divinylbenzene (DVB) as a
cross-linking agent2 (Fig. 1).
n
* *
* * 4 steps * *
* * * *
* *
Fig. 1.
78
Cross-linking with DVB, and the addition of sulfonic acid groups on the styryl ring3
impart strength and thermal stability to the polymer for the pyrolysis stage which involves
heating the material from room temperature to 800oC at a rate of 0.5oC/min. to afford a
carbon-rich matrix (Fig. 2).
n n
* *
* * * * Porous
Carbon-13-
* * * * Matrix
* *
SO3 H
Fig. 2
Initially, styrene, prepared using naturally abundant carbon-12-reagents, was polymerized

by suspension polymerization to afford small, spherical, glassy beads that were
functionalized and pyrolyzed. Carbonized beads ranging from approximately 1-300 µm in
size were evaluated with deuteron irradiation.4 As anticipated for beads of these sizes, the
recovered nitrogen-13 yields at end of bombardment (EOB) were low suggesting that a
high fraction of the produced nitrogen-13 remained in the carbon particles. The problem
of low recoil fraction should now be alleviated through the use of emulsion
polymerization which is used to create macroreticular polymers. The porous structure may
allow for escape of the produced nitrogen-13, thus increasing the yield at EOB.
Experimental
General
When required, glassware and cannula were flame-dried prior to use. NMR spectra were
recorded on a Bruker AC 250 MHz FT-NMR spectrometer. GC analyses were performed
on a Gow-Mac Gas Chromatograph Series 350 equipped with a Thermal Conductivity
Detector using a 20 % DC 200 on Chrom.-P 80/100 mesh column. All yields reported in
the C-13-styrene synthesis were determined by GC and corrected for purity.
[13C6]Bromobenzene1
To a 250-mL pear-shaped flask equipped with magnetic stirbar and condenser with drying
tube attached were added [13C6]benzene (4.94 g, 58.8 mmol), freshly distilled
dichloromethane (70 mL), triflic acid (99+ %, 5.03 g, 33.5 mmol), and 1,3-dibromo-5,5-
dimethylhydantoin (DBMH) (9.73 g, 34.0 mmol). During the addition of the DBMH, the
79
flask was covered with aluminum foil. The cloudy, yellow mixture was allowed to reflux
in the dark for 2 h. A saturated solution of sodium bisulfite (35 mL) was then added to the
product mixture until the amber organic layer was decolorized, and the mixture was
neutralized with aqueous sodium carbonate (60 mL, 2 M). The organic layer was
separated, and the aqueous layer extracted with dichloromethane. The combined organic
extracts were dried over magnesium sulfate and distilled to afford 7.47 g of
[13C6]bromobenzene (77.9 % yield, 96.7 % GC purity); 1H NMR: δ 7.50 (m, 2H, ortho
H's), 7.25 (m, 3H, meta and para H's).
[13C6]sec-Phenethyl Alcohol1
Magnesium turnings (3.30 g, 138 mmol), a crystal of iodine, and a magnetic stirbar were
placed in a flame-dried 100-mL three-necked round-bottomed flask equipped with
thermometer, reflux condenser, and addition funnel. The apparatus was then carefully
flame-dried again, and kept under an argon atmosphere. Ether (21 mL) was introduced to
the flask through the addition funnel by cannula. [13C6]bromobenzene (4.45 mL of
mixture, 4.30 mL, 40.8 mmol) was transferred to the funnel via cannula and carefully
added to the mixture at room temperature. The beginning of the reaction was evidenced by
the disappearance of the iodine color which changed from light yellow to gray, then
finally to dark brown. The flask was placed in a warm water bath to induce reflux. After
30 min. of continuous reflux, the mixture was placed in an ice bath and allowed to cool to
0oC. To the addition funnel was added a mixture of ether (7 mL) and acetaldehyde (5.0
mL, 89 mmol) by cannula. The acetaldehyde solution was slowly added to the mixing
Grignard reagent over a 63 min. period with the temperature held below 15oC. The
reaction mixture then became milky yellow and viscous. Ice water was added to quench
the reaction, and the magnesium salts were dissolved with 15 % sulfuric acid. The organic
layer was separated, and the aqueous layer extracted with diethyl ether. The combined
organic layers were dried over magnesium sulfate and concentrated in vacuo to yield 4.31
g of [13C6]sec-phenethyl alcohol (82.6 % yield, 89.8 % GC purity); 1H NMR: δ 7.30 (m,
5H, Ar H's), 4.85 (q, 1H, methine H), 2.10 (s, 1H, OH), 1.40 (d, 3H, CH3).
[13C6]Styrene1
[13C6]sec-Phenethyl alcohol (4.55 g mixture, 4.08 g, 31.9 mmol), potassium bisulfate (0.28
g), copper metal (0.08 g), and hydroquinone (spatula tip) were added to a 15-mL pear-
shaped flask equipped with short-path distillation apparatus with thermometer, and
magnetic stirbar. The flask was wrapped with aluminum foil and placed in an oil bath.
With a vapor temperature <150oC and oil bath temperature >200oC, the distillate was
collected. The organic layer was separated, and the aqueous layer extracted with diethyl
ether. The combined organic phases were dried over magnesium sulfate and concentrated
to give 2.54 g of [13C6]styrene (72.4 % yield, 81.7 % GC purity); 1H NMR: δ 7.69 (m, 2H,
ortho H's), 7.53 (m, 3H, meta and para H's), 6.99 (dd, 1H, CH), 6.03 (dd, 1H, H cis to
ring), 5.15 (dd, 1H, H trans to ring).
[13C6]Poly(Styrene/38 % DVB)2
Freshly distilled [13C6]styrene (0.80 g, 7.3 mmol), divinylbenzene (1.05 g containing 45 %
ethylvinylbenzene, 4.25 mmol DVB), and sorbitan monooleate (0.45 g) were added to a 1-
L plastic commercial reagent bottle equipped with an overhead stirrer. The bottle was
covered with plastic wrap. Potassium persulfate (0.15 g) was dissolved in deionized water
(77 mL), and the solution was added dropwise to the reagent bottle with stirring. A white,
80
creamy emulsion formed. The stirrer was removed, and the bottle was loosely capped and
placed in a 50oC oil bath overnight. After heating, the creamy emulsion had hardened into
a porous, white solid. The solid was chipped out of the bottle in 1 cm3 blocks, oven-dried
at 40oC overnight, then washed with isopropanol (250 mL) using a Soxhlet extractor and
oven-dried at 40oC overnight to afford [13C6]poly(styrene/38 % DVB) (0.64 g, 35 %
yield).
[13C6]Poly(Styrene Sulfonic Acid/38 % DVB)3

[13C6]Poly(styrene/38 % DVB) (0.59 g) and a catalytic amount of silver sulfate were
added to a three-necked 500-mL round-bottomed flask equipped with thermometer,
addition funnel containing chlorosulfonic acid (15 mL), and rubber stopper with exit
valve. The acid was added to the contents of the flask with stirring, and the reagents were
allowed to stir for 3 h at 100oC. The mixture was then allowed to cool with stirring
overnight. The copolymer was washed with deionized water until a neutral pH was
attained, then dried overnight at 100oC. In order to determine the extent of sulfonation, the
exchange capacity was measured by titration of hydrochloric acid, generated by potassium
exchange with the sulfonic acid proton.5 [13C6]Poly(styrene sulfonic acid/38 % DVB)
(0.021 g) was washed with aqueous potassium chloride (4.3 mL, 1.0 M) for 1 h, and the
solid was collected by vacuum filtration. Sodium hydroxide (0.95 mL, 0.050 M) was used
to titrate the filtrate, and the exchange capacity was calculated (2.2 mol/kg, approx. 41 %
sulfonated).
Pyrolysis of [13C6]Poly(Styrene Sulfonic Acid/38 % DVB)6

[13C6]Poly(styrene sulfonic acid/38 % DVB) (0.36 g) was weighed into an alumina boat
and placed in a quartz tube furnace. The furnace was purged with argon to maintain a
slight positive pressure, and the argon was released into a hypochlorous acid trap. The
copolymer was heated from room temperature to 800oC at a rate of 0.5oC/min., held at
800oC for 1 h, and then allowed to cool to room temperature under argon. This afforded
0.16 g (approx. 60 % yield) of carbonaceous material.
Conclusion
C-13-enriched poly(styrene/38 % DVB) has been synthesized in four steps starting from
C-13-benzene. The copolymer has been sulfonated and pyrolyzed to afford a porous
carbon-13-target material. The material’s packing density (0.30 kg/L) was determined by
placing a weighed sample into a stainless steel cylindrical device with a hollow center,
closed on one end, gently inserting a stainless steel plug into the opening, and measuring
the volume with a vernier caliper. As anticipated, the packing density of porous
poly(styrene/DVB) made by emulsion polymerization is significantly lower than that of
glassy poly(styrene/DVB) made by suspension polymerization.7 It is anticipated that the
porous C-13-matrix will allow for greater yields of N-13-labelled ammonia at EOB due to
a higher recoil ion escape.8
81
Acknowledgments
The authors acknowledge Cambridge Isotope Laboratories, Inc. for supplying the C-13-
benzene by way of a competitive grant award, the Biomedical Research Foundation of
Northwest Louisiana, the Louisiana Board of Regents Support Fund [LEQSF(1998-01)-
RD-B-13], the Louisiana Affiliate of the American Heart Association (LA-97-UG-02),
and Louisiana State University in Shreveport for support of this research.
References
1. Zippi E.M. (2001) Synthesis of phenyl-13C6-ethene. J. Label. Comp. Radiopharm. 44,

757.
2. Zippi E.M. and Kamperman E. Synthesis of [13C6]poly(styrene) for PET. Proceedings
of the Louisiana Academy of Sciences, in press.
3. Zippi E.M., May R., Hoffmeyer M., Landry P., Stone H. and Thompson J. Sulfonation
of porous poly(styrene) for use in PET. Louisiana Academy of Sciences Meeting,
Monroe, LA, February 1999.
4. Zippi E., Kabalka G., Bida G., Satyamurthy N. and Wieland B. (1992) Deuteron
irradiation of carbonized poly(styrene/divinylbenzene) resins for the in-target
production of [13N]ammonia. Appl. Radiat. Isotopes 43, 1363.
5. Gregor H.P., Bergman J.I., Gutoff F., Broadley R.D., Baldwin D.E. and Overberger
C.G. (1951) Studies on ion-exchange resins. Capacity of sulfonic acid cation-exchange
resins. Journal of Colloid Science 6, 20.
6. Zippi E.M. and Kabalka G.W. (1996) Synthesis, characterization and pyrolysis of
poly(styrene/divinylbenzene) derivatives. Carbon 34, 1539.
7. Zippi E.M., Hoffmeyer M. and Jones G. Differential Scanning Calorimetry of styrene-
based polymers. Journal of Undergraduate Chemistry, submitted.
8. This target material is now available for cyclotron evaluation. Interested collaborators
should contact the corresponding author by e-mail (ezippi@pilot.lsus.edu).
82
High-energy Irradiations of Solid and Gas Targets
Coordinators:
Syed M. Qaim, Forschungzentrum Jülich GmbH, Germany
Nigel R. Stevenson, Theragenics Corporation, USA
A New Method for Cu-64 Radioisotope Production Theoretical and
Experimental Results
1 2 2 2 2 3
K. Abbas , C. Birattari , M. Bonardi , F. Groppi , H.S. Mainardi ,E. Menapace ,
1 1 1
F. Simonelli , U. Holzwarth and M.F. Stroosnijder
1
Institute for Health and Consumer Protection (IHCP), JRC-Ispra, European Commission,
via E. Fermi, I-21020 Ispra (VA), Italy
2
Radiochemistry Laboratory, LASA, University and INFN, via F.lli Cervi 201, I-20090
Segrate (MI), Italy
3
ENEA, Applied Physics Division, via Don Fiammelli 2, I-40128 Bologna, Italy
Aims
64
Cu is a radioisotope of great interest for PET and radiotherapy. It decays by the emission
of β+ (17.4 %), which allows PET imaging, and by the emission of β- of 0.579 MeV (39
%) making it especially suited for targeted radiotherapy. Moreover, its half-life of 12.7 h
matches the requirements for a reasonable uptake of larger tumour targeting molecules like
monoclonal antibodies or antibody fragments. The therapeutic potential of Cu-64 has been
demonstrated in a colorectal tumour-bearing hamster model where complete tumour
regression could be achieved with therapeutic doses of 64Cu-labelled mAb 1A3 without
observed toxicity. In clinical PET using Cu-64-TETA-Octeotide the compound showed
high sensitivity in lesion detection of neuroendocrine tumours with favourable dosimetry
and pharmacokinetics.
Methods
Several production methods have been investigated for the production of this radioisotope,
such as those using neutron irradiation of Cu and Zn (natural or enriched targets) and also
proton beam irradiation of an enriched 64Ni target. The latter method yields a high specific
activity product, however 64Ni is obviously very expensive. Irradiation with neutrons is
nowadays problematic since the nuclear reactors are shutting down one after the other. The
new method for N.C.A 64Cu production (where also 67Ga radioisotope of medical interest
is produced) based on natural Zn irradiated with deuterons (19 MeV) is promising and
preliminary results have been already presented elsewhere.
Results
In this paper we compared experimental thick target yield of the 64Cu and other produced
radioisotopes with the calculated results obtained using Monte Carlo simulation of the
excitation functions and the stopping power. The table below shows a good agreement
between the experimental and the theoretical results.
84
Conclusion
This paper presented experimental and theoretical results of a new method for the
production of N.C.A 64Cu radioisotope. Using the same theoretical approach, an
investigation for deuteron irradiation of enriched Zn is planned for further improvement of
the purity of the 64Cu radiochemical product.
Radioisotope Half life Nuclear Reaction Theoretical Experimental

(h) Yield Yield (MBq/µAh)
(MBq/µAh)
64Cu 12.7 64Zn(d,2p), 66Zn(d,2p2n), 24,171 30,821
67Zn(d,2p3n), 68Zn(d,2p4n)
61Cu 3.4 64Zn(d, 2p3n) 163,733 104,007
67Cu 61.9 67Zn(d,2p), 68Zn(d,2pn), 0 Not det.
65Zn 5863.2 64Zn(d,p), 66Zn(d,p2n), 0,276 0,481
64Zn(d,n) and b+ decay
69mZn 13.8 68Zn(d,p) 26,557 14,504
66Ga 9.4 66Zn(d,2n), 67Zn(d,3n), 101,426 138,38
67Ga 78.3 66Zn(d,n), 67Zn(d,2n), 19,010 16,28
68Zn(d,3n)
85
Excitation Functions of Proton Induced Reactions on Enriched
Rb-85 up to 100 MeV: Yields of Sr-82 and Sr-83
S. Kastleiner1, S.M. Qaim1, G. Blessing1, H.H. Coenen1, F.M. Nortier2 and T.N. van der
Walt2
1
Institut für Nuklearchemie, Forschungszentrum Jülich GmbH, D-52425 Jülich, Germany
2
Radionuclide Production Group, National Accelerator Centre, P.O.Bax 72, Faure 7131,
South Africa
Aims
The radionuclide 83Sr (T½ = 32.4 h) is a positron emitting analogue of the pure electron
emitting therapeutic radionuclide 89Sr (T½ = 50.5 d). The radionuclide 82Sr (T½ = 25.5 d) is
widely used for preparing the 82Sr/82Rb positron emitting generator system. The cross
section data base for the formation of both 83Sr and 85Sr was rather weak. The aim of this
work was to determine the excitation functions of all the (p,xn) reactions on 85Rb up to
100 MeV and to calculate therefrom the integral yields of 83Sr and 82Sr as well as the level
of the long-lived impurity 85gSr (T½ = 64.9 d).
Methods
Excitation functions were measured by the coventional stacked-foil technique. Thin

samples of 99.4 % enriched 85Rb-chloride were prepared by sedimentation on copper foils.
Irradiations with protons up to 45 MeV were done at the CV28 and injector of COSY at
Jülich, up to 71 MeV at PSI (Villigen), and up to 100 MeV at the cyclotron of NAC
(Faure). The beam current was measured via monitor reactions. The radioactivity of the
products was determined using high-resolution gamma-ray spectrometry. In the case of
82
Sr, a high-performance liquid chromatographic separation and X-ray spectrometry were
also applied. The cross sections were calculated using the well-known activation equation.
From the measured excitation functions the integral yields of 83Sr, 84Sr and 85Sr were
calculated.
Results
The data obtained using the four cyclotrons have maximum uncertainties of about 15 %
and are consistent. The transition from the low-energy region to the high-energy region is
smooth. The excitation function for the formation of 85Sr above 30 MeV has been studied
for the first time. The data for 83Sr agree with the literature results up to 70 MeV, but the
cross sections for 82Sr are appreciably lower than the published values. For the energy
range 30 to 37 MeV the 83Sr yield amounts to 160 MBq/µAh and the level of the long-
lived 85gSr and 82Sr impurities to < 0.25 %. The yield of 82Sr over the proton energy 40 to
60 MeV amounts to 10.5 MBq/µAh and the level of the 85gSr impurity to < 10 %.
86
Conclusion
The radionuclide 83Sr can be advantageously produced via the 85Rb(p,3n)-reaction,

provided a 40 MeV cyclotron is available. The yield of 82Sr via the 85Rb(p,4n)-reaction is
about 30 % lower than the literature value.
87
Production Plans for the New 100 Mev Isotope Production
Facility at Lanl
F.M. Nortier, R.C. Heaton, E.J. Peterson and D.R. Phillips
Los Alamos National Laboratory, MS J514, Los Alamos, New Mexico 87545, USA
To ensure that U.S. researchers have a steady supply of medical isotopes, the U.S.
Department of Energy's Los Alamos National Laboratory is building a new Isotope
Production Facility (IPF) to replace an existing facility. Once operational, the new IPF
will support eight months of isotope production annually. Combining its output with
similar isotope production capabilities at Brookhaven National Laboratory in New York
will ensure doctors and researchers an adequate, year-round supply of accelerator-
produced medical and reseach isotopes. Construction of the new IPF began in February
2000 and the first production runs are planned for October 2003.
Isotopes such as 82Sr and 68Ge have been produced at LANL for more than 20 years using
800 MeV protons. The new facility will use a 100 MeV proton beam with beam currents
of up to 250 µA extracted from the existing LANSCE accelerator by means of a pulsed
kicker magnet. This lower proton energy is much more suitable for radioisotope
production and will lead to products with substantially higher purity. The facility will
irradiate a wide range of materials to produce a variety of radioisotopes of value to the
DOE’s Office of Isotopes including 82Sr, 68Ge, 67Cu, 22Na, 32Si, 48V, 88Zr, 73As, 186Re, etc.
Up to three targets can be irradiated simultaneously utilizing the production energy

windows 90-70 MeV, 65-45 MeV and 30-10 MeV. Targetry and target preparation
facilities and techniques are being developed in order to ensure that the facility output
meets the quantity and variety demanded by researchers and commercial suppliers
worldwide. Production schedules are designed to minimize overall production cost by
maximizing the occupation of the three energy slots in such a way that a good mix of
long- and short-lived isotopes are made available on a regular basis. As an example
expected production quantities for continuous operation over an eight months period is
given below for one such proposed production schedule.
Isotope Half-life Target Quantity produced

82
Sr 25.5 d RbCl 2.5 Ci per month
68
Ge 270 d Ga 1.8 Ci per month
67
Cu 2.6 d Zn 500 mCi twice a month
73
As 80.3 d Ge 1 Ci per year
22
Na 2.7 y Mg 1.4 Ci per month
32
Si 172 y NaCl 160 µCi per year
48
V 16 d Cr 6 Ci per month
88
Zr 83.4 d Nb 2.7 Ci per month
186 186
Re 3.2 d W 250 mCi twice a month
88
67
Cu Production by Proton Induced Reactions on Enriched 68Zn
R. Schwarzbach, J. Jegge, K. Zimmermann, R. Schibli and P.A. Schubiger
Paul Scherrer Institute, Center for Radiopharmaceutical Science, CH-5232 Villigen,

Switzerland
Aims
At present we produce up to 3.7 GBq of 67Cu in the 72 MeV accelerator at Paul Scherrer
Institute by irradiating natZn. We tested the use of 68Zn in order to scale up the 67Cu
production.
Methods
Cross sections were measured using the reactions 65Cu(p,n)65Zn of monitor foils,
nat
Zn(p,2pxn)67Cu, and 68Zn(p,2p)67Cu in the proton-energy range 67 to 45 MeV.
Preparation of thin Zn-targets: 5 to 10 mg/cm2 natZn or 68Zn were electroplated on discs of
Al at 3.6 V and 20 mA/cm2 for 5-10 min. Thick Zn-targets: 30 mL 1.5 M 68ZnSO4, 0.1 M
H2SO4 solution was electrolysed at 4.5 V and 125 mA/cm2, 892 mg/cm2. 68Zn was
electroplated on disc Ag. A rotary Pt-rod (1/s) served as anode. Electrolysis was stopped
after 400 min. The pH of the solution at the start of electrolysis was 1.6, the final pH was
0.8. Current efficiency was 85 %.
Irradiations and proton flux: The entrance proton energy was 67.7 MeV after degradation
in the cooling water. Beam intensity was determined via monitor reaction and by Farady
cup. Proton beam: 1-4 µA, 5-60 min and strong 70 µA, 2 h. Radioactivity was measured
after 2-5 days and measurements were repeated up to 23 days, to allow decay of
interfering short-lived isotopes, especially of 61Cu.
Radiochemical separations: The thin 68Zn-targets were dissolved in 0.5 ml 65 % HNO3.

The Zn-solution was washed off the surface, evaporated to dryness, re-suspended in 1 mL
7 M HCl and dried. The residue was taken up in 3 mL 7 M HCl and applied (0.25
mL/min) to a Dowex 50W X 8 (2.5 mL H-form). The Zn/Cu/Co/Ni-isotopes were eluted
to a total volume of 10 mL of 7 M HCl. Under these conditions the Ga-isotopes remain on
the cation exchange resin.
Work up of irradiated production-targets: Work up and separation of 67Cu was

accomplished (total mass per target = 2.8 g 68Zn) over a combination of ion exchange
catridges (Bio-Rad AG 50W-X8 /Chelex-100 /Bio-Rad AG1-X8). Recovery of 68Zn: The
67
Cu /68Zn solution was evaporated to dryness. The residue was then taken up in 20 mL
2 M HCl and washed over Dowex 1X8 with 180 mL 1 M HCl. Co/Ni/Fe pass the column,
whereas 68Zn eludes with 80 mL 0.01 M HCl. The electrolysis was performed in sulfuric
acid solution. Thus, chloride in the 68Zn-solution was exchanged by sulfate via a Dowex
50W X 8 ion exchange column using 400 mL H2O followed by 0.5 M H2SO4 to wash off
the 68Zn. The solution was then evaporated. The residue was dissolved in 5 mL H2O and
89
used for further electrolysis. More than 90 % of 68Zn can be recuperated by this method.
Contain of 65Zn: 65Zn accumulates in the targets, when the targets are recovered. The use
of the enriched 68Zn results in a two-fold increase in the yield of 65Zn compared to natZn.
Yield of 65Zn decreases with higher proton-energy up to a maximum of 30 MeV. At above
energies the amount of 65Zn declines, whereas the production of 67Cu increases. The
formation of 67Ga reaches a maximum at 30 MeV and decreases at higher proton energy as
well. Compared to natZn, the formation 67Ga was higher by a factor at 4.1.
Conclusion
67
Cu production: in the 67 to 45 MeV energy range the yield of 67Cu from the enriched
68
Zn was found to have 4.5 fold increase compared with the natZn. The specific activity of
67
Cu should increase by means of using Chelex-purified Zinc. Currently the specific
activity of 67Cu is 3.7 x 10E+8 Bq /10E-6 g.
1.00E+05
1.00E+04
Activity ( Bq/uAh )
1.00E+03
1.00E+02
65Zn/mgnatZn
67Ga/mgnatZn
1.00E+01 67Cu/mgnatZn
65Zn/mg68Zn
67Ga/mg68Zn
67Cu/mg68Zn
1.00E+00
10 20 30 40 50 60 70
p-Energy ( MeV )
Fig. 1: 68Zn -target after electrolysis Fig. 2: 67Cu depending on the p-energy and
nat
Zn /68Zn-isotopes
References
[1] R. Schwarzbach, K. Zimmermann, P. Bläuenstein, Alan Smith and P. A. Schubiger,

Appl. Radiat. Isot. Vol. 46, No5, pp. 329-336 ( 1995 ).
90
Reports from the Laboratories
Coordinators:
Didier Le Bars, University of Lyon
Steve McQuarrie, University of Alberta
Status of Cyclotron based Radioisotopes Production in
Kazakhstan. Report from the Laboratory
A. Arzumanov, N. Berdinova, A. Borissenko, G. Chumikov, N. Gorodisskaya, A. Knyazev,
V. Koptev, Yu. Popov, G. Sychikov and D. Zheltov
Institute of Nuclear Physics, 480082 Almaty, Kazakhstan
Introduction
The variable-energy isochronous cyclotron U-150 M has been in routine operation in

Almaty since June 1967. It was designed and manufactured by the D.V. Efremov Institute
in Leningrad as “classical” machine with a single beam line and fixed energy per nucleon
10 MeV/nucleon for protons, deuterons and alphas. In 1971-72 it was converted to
variable-energy isochronous mode of operation [1]. Now it is a compact pole, three spiral
sector, positive ion machine with K=50 MeV. The cyclotron generates beams of protons 6
– 30 MeV, deuterons 12,5 – 25 MeV, helium-3 ions 18,5 – 62 MeV, helium-4 ions 25 – 50
MeV.
Radioisotope 201Tl has been manufactured in Kazakhstan since 1986 [2]. The short-lived
radoioisotope and its labeled compound was supplied to hospitals and research institutes
in Almaty and Medical Centers in nearby countries for use as tracers for diagnostic and
research purposes.
Now also radionuclides such as 67Ga, 57Co, 109Cd, 237Pu, 88Y, 85Sr are produced upon
request. Wet chemical procedures have been developed for the recovery and purification
of various radionuclides from the irradiated targets.
Target preparation
At the present time for isotopes production only solid targets are used at the internal and
external beam of the INP isochronous cyclotron. The schematic view of internal target
head is shown in Fig.1. Metallic targets prepared by electrodeposition of corresponding
metal on a copper substrate of special design are used for production of 201Tl, 67Ga, 109Cd,
57
Co.
92
Fig. 1. Schematic view of the internal target.
Choice of the method for electrolytic deposition depends on the type of metal to be
deposited. When a nickel one (57Co) is prepared the galvanostatic mode of deposition with
insoluble platinum anode is used. For getting silver (109Cd) and zinc (67Ga) layers
electrolysis with soluble anode and reverse current is used. This enables to control
simultaneously thickness of the deposited layers and spent anode material.
A technology developed by P. Van den Winkel (VUB) and handled over within IAEA
Technical Co-operation Project is used for the production of thallium targets.
Below, in Table 1, one can find detailed data on metallic targets. Powder targets produced
by pressing at ~2500 N/cm2 of RbCl, SrO and UO2 into a metallic capsule are used for
production of the nuclides 85Sr, 88Y and 237Pu, correspondingly. In all cases the thickness
of the pressed powder is 1-1.5 mm.
Tab. 1. Summary data on the internal targets prepared by electrolysis.
Radionuclide Target Metal layer Presence Electrolyte type Current mode

material thickness, µm of sublayer
201 203
Tl Tl 160 - Alkaline EDTA chopped,
sawtooth
67 nat
Ga Zn 350 - Subacid chloride Direct with reverse
109 nat
Cd Ag 290 Ag, 2 µm Thiocyanatous Direct with reverse
57 nat
Co Ni 240 Au, 10 µm Sulphate-chloride direct
For assessment of adhesion due to presence of a diffusion layer a set of experiments was
carried out for investigations of the diffusion rate for metal electrically deposited on a
copper substrate matrix at different temperatures. Profile of the coated layer prior and after
the annealing is determined at the Electrostatic tandem with 1.5 MeV protons by the PIXE
method [3]. The investigations have shown that thickness of the diffusion layer comprises
93
0.5-1 µm at target annealing temperature of 250-300°C. Therefore, to remove bulk
stresses in metal and to increase the diffusion layer thickness the targets are annealed in
vacuum at temperature 180-250°C during 3-5 hours before the irradiation.
Target irradiation
For radionuclides production both external and internal target irradiation are used. Beam
intensity on the external target now is equal to 30 µA for protons and deuterons and
limited by beam power dissipation in the electrostatic deflector. In some cases during
target irradiation the beam is swept across the target surface in a circular fashion by means
of a stator of electro-motor in order to ensure an even power distribution.
Beam power distribution for the extracted beam spot and for the circularly scanned beam
are shown in Fig. 2. When irradiating targets with high intensity internal beam the target is
inclined with respect to the beam axis at the angle of 18° in order to reduce beam power
density on the target. Thin layer 0,5 mm thick of cooling water, flowing at a velocity of
approximately 30 m/s along the target surface is employed to ensure necessary cooling of
the target. Cooling water pressure is equal to 20 bar which enables to use irradiation with
internal 30 MeV proton beam and intensity up to 500 µA. Summary of the produced
radioisotopes is presented in the Table 2.
Tab. 2. Summary of the produced radioisotopes.
Radioisotope Production route Energy of Beam Average Yield of activity

incident intensity, heat flow, (for nat),
2
ions, MeV µA kW/cm mCi/µA*h
201
Tl 203
Tl (p,3n) 201Pb → 201Tl 28,5 125 0.7 ≥ 0.2
(72,9 h)
67
Ga 68
Zn(p,2n)67Ga 30 125 0.8 ≥ 1.3
(78,2 h)
109 109
Cd- Ag (p, n) 109Cd
22 250 1.1 ≥ 2.6e-3
(463 d)
57 58
Co Ni ( p, 2p) 57Co
58 22 300 1.3 ≥ 0.3
(270 d) Ni (p, pn) 57Ni → 57Co
237
Pu 235
U(α,2n)237Pu 48 (α) 10 - ≥1.7e-10
(45,2 d)
88
Y 88
Sr (p,n) 88Y 22 10 0.044 ≥ 0.103
(106,6 d)
85
Sr 85
Rb(p,n)85Sr 22 10 0.044 ≥0.076
(64,8 d)
94
a) b)
Fig. 2. External beam spot (a). Circularly swept beam spot (b).
Radiochemical extraction of radionuclides:
Classical methods of ion exchange and extraction chromatography are used at

radiochemical processing of the irradiated targets. Developed on their basis techniques
enable to obtain high rate of purification of a desired product both from macrobacking, the
irradiated target material, and from accompanying radioactive impurities.
I. Thallium-201:
1. Extraction of 201Pb (T1/2 = 9.3 hours) by co-depositing on SrSO4 collector and optimal
exposure of the extracted 201Pb during 26 hours for 201Tl accumulation.
2. Separation of the accumulated 201Tl from the parent radionuclide 201Pb by the
extraction method.
3. Thallium-201 transfer to isotonic solution, a base for production of the
radiopharmaceutical preparation (RPP).
II. Gallium-67 [4]:
1. Chemical dilution of a zinc layer at the substrate in 12N HCl.
2. Separation of 67Ga from macroamounts of Zn and microelements Fe and Cu by the
method of extraction chromatography.
3. Gallium-67 transfer to isotonic solution in form of citrate complex, a base for
production of RPP.
III. Cobalt-57:
1. Anode dilution of a nickel layer at the substrate in 9N HCl.
2. Separation of 57Со from Ni, Fe, 65Zn by the method of ion exchange chromatography.
3. Cleansing of Cu residue by electrolysis.
IV. Cadmium-109:
1. Chemical dilution of a silver layer at the substrate in 7N HNO3.
2. Separation of 109Сd from silver macroamounts by the method of cementation on
copper.
3. Separation of 109Сd from 65Zn, Fe and macroamounts of Сu by the method of ion
exchange chromatography.
95
V. Yttrium-88 [5]:
1. Dilution of the irradiated powder in weak HNO3
2. Separation of 88Y from 65Zn, Fe, Cu and macroamounts of Sr by the method of
extraction chromatography.
VI. Strontium-85 [5]:
1. Dilution of the irradiated powder in weak KOH and filtration.
2. Separation of 85Sr from 65Zn and macroamounts of Rb by the method of ion
exchange chromatography at activated charcoal.
VII. Plutonium-237 [5]:
1. Dilution of the irradiated powder in concentrated HNO3.
2. Pu (III) oxidation to Pu (IV).
3. Extraction of Pu by the method of extraction chromatography.
Recovery
Recovery of the target material occurs in two cases only:

1. Recuperation of enriched 203Tl in metal form to the Tl target production cycle from the
spent solutions and the electrolyte. The procedure implies electrolysis with controlled
cathode potential (CCPE). The recovery rate is ≥99.5 %.
2. Recuperation natAg in metal form to the Ag target production cycle from the worked
out for ~80 A.h/l thiocyanate electrolyte. The reductioned silver is used as dissolved
anodes. The recovery is also carried out by CCPE with the recovery rate ≥99 %.
Electron beam stand
The electron beam test facility was created for preliminary heat load targets examination
before their irradiation at the cyclotron.
This facility also enables to expedite development of technologies for isotope target
production.
Electron blocks of beam focussing and deflection in vertical and horizontal directions
provide obtaining of beam dimensions of diam. 1.5 mm and its scanning against the target
surface. Scanning frequencies are the horizontal one – 50 Hz and vertical one – 1200 Hz
that provides homogenous target heating. There is an opportunity of fixed beam spot
moving at any point of the target surface. Electron beam has following parameters:
electrons energy up to 25 keV, beam intensity up to 200 mA. The water pressure in
cooling system is equal to 20 bar that provides water velocity in the cooling channel equal
to 30 m/s.
Target tests at the stand enable to determine maximal allowed heat flows which are fixed
in melting point of target material. Energy loss caused by reflected and second electrons
are accounted on basis of known reference data.
96
Conclusion
Cyclotron internal beam intensity is increased up to 500 µA for 30 MeV protons. Internal
target design enables to irradiate it for some materials with beam current of 350 µA.
Sweeping system is inserted in the beam line to produce homogeneous irradiation of the
external target. Many radioisotopes to be used in medicine, industry etc. can be produced
in Kazakhstan using K = 50 MeV cyclotron to meet need of the Republic.
Acknowledgement
The authors would like to express their gratitude to IAEA for support of this work.
References
[1] Arzumanov, A.A., Nemenov, L.M., Nucl. Instr. And Meth. 106, №2, 201-202 (1973).
[2] Malinin, A.B., Kozlova, M.D. et al., Int. J. Appl. Radiat. Isotopes. 685 (1984).
[3] Arzumanov, A.A., Borisenko, A.N. et al., “Application of electrostatic tandem in

Almaty for analysis of beryllium foils and radioactive particulates”, 6-th EPAC, 1998,
2433 - 2435.
[4] Arzumanov, A.A., Batischev, V.N., et al., “ Development of cyclotron based high
beam current technique for Ga-67 production”, 7-th EPAC, 2000, 2498 - 2499.
[5] Arzumanov, A.A., Batischev, V.N., et al.,”Production of Plutonium, Yttrium and

Strontium tracers for using in environmental research”, 16-th Conf. on Cycl. and their
Appl., 34 – 36.
97
Performance of a Flow-Through Target for the Production of
[11C]CH4
T.E. Barnhart2, K.R. Buckley1, S. Jivan1, A.D. Roberts2 and T.J. Ruth1
1
2
University of Wisconsin, Madison, WI, USA
Previous work at TRIUMF (Buckley et al.) on static irradiations of a H2/N2 target gas for
the production of methane indicated that recoverable yields at 13 MeV plateaued at about
750 mCi at EOB. The relatively poor yields were attributed in part to hot atom interactions
with the target chamber walls. In order to try and overcome this limitation an aluminum
bodied flow-through gas target was implemented and tested for the production of
[11C]CH4 on the TR13 cyclotron at TRIUMF. Nitrogen gas with varying quantities of
hydrogen has been used to investigate the optimum ratio for methane production.
Comparison of the flow through target with the static-target irradiations for production of
methane are given.
Methods
Aluminum body gas targets were designed with cylindrical bores of 12.7 mm ID and 120
to 127 mm lengths. A water-cooled support grid shown in figure 1, with 34 X 1.59 mm
holes arranged in a hexagonal pattern with 1.78 mm between centers supported aluminum
target window foils 25.4 micrometre thick (Barnhart et al.). Transmission through the grid
was calculated to be approximately 70 percent. Beam transmission studies indicated 64 %
transparency. Mixtures of N2 with up to 12 percent H2, pressurized up to 3.1 MPa were
used to make [11C]CH4. For comparison, [11C]CO2 was produced in the same target
system with a mixture of N2 and 0.5 percent O2, pressurized at 0.69 to 2.28 MPa. Gas flow
in each case was 200 to 600 ml/min. Irradiations of up to 32 µA lasted up to 60 minutes.
The radiolytic production of ammonia could be a concern for post production processing,
e.g. [11C]CH3I synthesis. To measure the ammonia this system produced, we passed the
target gas through a series of HCl traps and back-titrated to a phenolphthalein end point.
98
Fig. 1. Schematic of Grid Window Support.
Results
Testing of the flow-through target gave a saturation yield of up to 2.89 GBq/µA (79
mCi/µA) [11C]CH4 at 32 µA with a correction for the grid transmission (both beam current
on target gas and saturation yields). For comparison, [11C]CO2 was produced with a
saturation yield of 2.66 GBq/µA (73 mCi/µA) at 27 µA with correction for grid losses.
Yields improved to approximately 12 % H2 as is shown by the colleted activity in figure 2.
Higher concentrations of H2 may increase the saturation yield of [11C]CH4. The maximum
[11C]CH4 produced after a 35 minute run at 30 µA was 57.3 GBq (1.55 Ci), shown in
figure 3. Comparison of the flow through target with the static-target irradiations for
production of methane are given in figure 4. Results for 5 % hydrogen in the flow-through
target show a marked reduction in methane production over the 10 % hydrogen gas mix.
99
400
350
300
Activity (mCi)
250
200
150
100
50
0
0 2 4 6 8 10 12
% H2
Fig. 2. Collected activity vs. percent hydrogen.
1800 100
1600 95
1400 90
mCi/µA @ Sat.
mCi @ Time
1200 85
80
1000
75
800
70
600 65
400 60
Flow Target Yield
200 55
Saturation Yields
0 50
0 5 10 15 20 25 30 35 40
Time (min.)
Fig. 3. Flow-through target total activity and saturation yields vs. time.
100
125
100
mCi/µA @Sat
75
50
Static Target (10% H2)
Flow - 10% H2
25 Flow 5%
Flow -10% H2
0
0 5 10 15 20 25 30 35
On Target Beam Current
Fig. 4. Comparison of 11C-CH4 yields in static and flowing target systems with varying
hydrogen percentages. On-target beam current is the beam current corrected by
transmission losses of the grid.
Preliminary measurements of the specific activity (SA) via the synthesis of [11C]
raclopride indicate that the EOS SA is approximately 13.5 GBq/µmole (500 mCi/µmole).
Further conditioning of the target system should improve the specific activity. In the static
target we found that ammonia production reached equilibrium quickly resulting in a
production of 0.13 mmoles. The flow through target generated 8 mmoles of ammonia for a
50 µA X 20 minute irradiation, however the NH3 should be quanitatively removed with
sufficient HCl traps in the gas stream.
Conclusion
The flow-through target system for the production of [11C]CH4 provides increased
radioactivity over single shot low volume, high pressure gas targets. The [11C]CH4
saturation yield is within experimental errors to that for [11C]CO2 under similar conditions
and is 72 % of calculated theoretical maximum yield, allowing for the support grid
transmission. The cause for the relatively low specific activity is under investigation. The
high quantity of ammonia co-produced will adversely affect the production of [11C]CH3I
and must be removed for a useful system.
Acknowledgements
This work was supported through a TRIUMF Life Science project. TRIUMF is funded
through a contribution from the National Research Council of Canada.
101
References
Barnhart, TE, A K Converse, K Dabbs, R J Nickles, K R Buckley, S Jivan, T J Ruth, A D

Roberts. Water-cooled Grid Support System for High Power Irradiation with Thin Target
windows. Applied Radiation and Isotopes. In Press 2002.
Buckley, KR, J Huser, S Jivan, K S Chun, T J Ruth, 2000, C-11-methane production in

small volume, high pressure gas targets: Radiochimica Acta, v. 88, p. 201-205 and
references therein.
102
Qualitative Evaluation of Helium-Cooling Windows
D.W. Becker2, K.R. Buckley1, J. Lenz3, B. Hagen1, P. Piccioni1 and T.J. Ruth1
1
2
Now at ARGOS Zyklotron Betriebs-Ges.m.b.H., LKH Klagenfurt, PET-Zentrum NMSE,
St. Veiter Strasse 47, A - 9020 Klagenfurt
3
John Lenz & Associates, 412 Muskingum Rd., Waxahachie, TX, USA
Aim
To support the quest for higher beam current deposition on gas or liquid targets we set out
to evaluate the effectiveness of our helium cooling windows.
Methods
The parameters explored included flow rate, pressure and direction of the jets. In order to
rank the different designs of helium flow paths across the target window we built a simple
jig to deposit a reproducible amount of heat into a plate coupled to the He cooling
chamber and observed the temperature change to the plate as a function of differing
helium cooling window designs and conditions. In addition to this work a computational
fluid dynamics (CFD) analysis of helium flow and heat transfer was performed to
investigate our “standard” helium cooling window and supposed near "ideal" helium
window geometry.
Results
We will present the results of the mathematical analysis and the qualitative ranking of the
different designs.
Conclusions
Modelling of the He window can lead to improved designs for more efficient cooling.
103
The Turku PET Centre Medical Compact Cyclotron CC-18/9
P.V. Bogdanov1, O.A. Kovalchuk1, V.A. Kislov1, Yu.P. Severgin1, Yu.I. Stogov1, I.N.
Vasilchenko1, S.-J. Heselius2 and S. Johansson2
1
D.V. Efremov Scientific Research Institute of Electrophysical Apparatus (NIIEFA),
Metallostroy Road 3, St. Petersburg 196641, RUSSIA
2
Turku PET Centre, Accelerator Laboratory, Åbo Akademi University, Porthansgatan 3,
FIN-20500 Turku, FINLAND
Introduction
A medical compact cyclotron CC-18/9 has been designed and manufactured by the D.V.
Efremov Scientific Research Institute of Electrophysical Apparatus (NIIEFA). The
cyclotron has been designed for the production of radionuclides applied in positron-
emission tomography (PET) as well as for the production of other radionuclides used in
nuclear medicine. The cyclotron will be installed at Turku PET Centre, Finland, in 2003-
2004. Negative ions of hydrogen and deuterium (H- and D-) are accelerated in the
cyclotron. A multi-cusp ion source will be used for external injection of the negative ions
into the central part of the cyclotron. The particle beams of protons and deuterons are
produced by stripping the negative H- and D- ions in carbon foils in the cyclotron vacuum
tank. Efficiency of the extraction is approximately 100 %. Specified beam currents of
protons and deuterons are 100 µA and 50 µA, respectively.
General description
The cyclotron accelerates negative ions of hydrogen (H-) to 18 MeV energy and deuterons
(D-) to 9 MeV energy. The magnet of the cyclotron is of shell type with a vertically
located median plane. Four sectors per pole form an isochronous magnetic field in the
working area of the cyclotron. The gaps in hills and valleys are 27 mm and 118 mm. The
shape of the sectors maintains the isochronous acceleration of the H- beam. The
isochronous acceleration of the D- beam is achieved by movable ferromagnetic shims
which are installed in special recesses inside the hills. The cyclotron has no trim coils.
The cyclotron vacuum tank consists of a casing and two lids. The tank casing is a hollow
thick-walled cylinder made of carbon steel that also is part of the magnet yoke. The tank
lids are pole tips of the magnet with flanges of stainless (non-magnetic) steel welded to
them. The accelerating double-dee resonant system is located inside the vacuum chamber
and fixed to the side of the tank casing.
The particle beams of protons and deuterons are produced by stripping the H- and D- ions
in thin carbon foils. There are three stripping foil mechanisms for extraction of H+ and D+
beams out of the vacuum tank, two of which are for targets mounted on the cyclotron, the
third one is for the beam extraction and beam delivery to four distant target stations. The
location of the stripping mechanisms and extraction foils in the tank casing provides
besides single target irradiation, also the possibility for irradiating simultaneously one of
104
the near and one distant target, or alternatively two near targets. The near targets are
mounted on the gate valves 13 and 14 (Fig. 1). The layout of beam extraction and
equipment in the cyclotron tank is shown in Fig 1.
The beam-transport system delivers the particle beam to four distant target stations (Fig.
2). The system consists of an ion guide, a doublet of quadrupole lenses, a switching
magnet and after that four ion guides diverging at angles of ±12º and ±36º with reference
to the initial beam direction. Doublet quadrupole lenses mounted on all the four ion guides
focus the beam to the target stations. A beam-positioning magnet (11 in Fig. 1, and 2 in
Fig. 2) is installed at the beam outlet of the cyclotron to correct for the beam direction.
The layout of beam transport system with appropriate equipment is shown in Fig. 2.
The beam diagnostics in the cyclotron vacuum tank is done by using a probe, and in the
delivery system by using scanners and Faraday cups. To gain access to equipment inside
the cyclotron vacuum tank half of the magnet yoke together with the tank lid can be
moved away from the tank casing 800 mm by an electrical gear.
The operating pressure in the cyclotron tank is ensured by two cryogenic pumps with a
capacity of 3200 liter per second each and by a rough pump. The performance of the
cyclotron and beam delivery system are controlled from the control panels and by
computer.
The technical data of the CC-18/9 cyclotron are summarized in Table 1.
Table 1. Technical data of the CC-18/9 cyclotron.
Accelerated particles H-, D-

Extracted particles H+, D+
Energy of particles H/D, MeV 18 / 9
Extracted beam currents for H+/D+, µA 100 / 50
Electromagnet:
Pole diameter, cm 115
Average induction at the final radius, T 1.25
Power consumption, kW 7
Weight of magnet, ton 22
Resonant system:
Frequency, MHz 38.2
Number of dees 2
Dee voltage, kV 35
RF-generator power, kW 30
General power consumption

of the whole cyclotron complex, kW 90
105
Fig. 1. Equipment layout inside the vacuum chamber of the cyclotron: 1-vacuum chamber;
2-RF system; 3-power input; 4-trimmer; 5, 6, 7- carbon stripper units; 8-probe; 9, 10-
baffle-boards; 11-steering magnet; 12, 13, 14- valves; 15-cryogenic pump; 16-RF probe.
Fig. 2. Equipment of the cyclotron. 1- cyclotron; 2- steering magnet; 3- magnetic

corrector; 4- scanner; 5- Faraday cup; 6- doublet quadrupole lenses 7LK; 7- switching
magnet; 8- doublets quadrupole lenses 4LK;9- target stations; 10- rough vacuum pump;
11- pneumatic station; 12- cryogenic pump power supply; 13- external injection system.
106
Website for Proceedings of the Workshops on Targetry and
Target Chemistry
K.R. Buckley and T.J. Ruth
Aim
To make all past WTTC Proceedings available on line to the target community and
beyond.
Methods
TRIUMF hosts a website for the proceedings of the WTTC. The website is at
http://www.triumf.ca/wttc. Proceedings of past workshops that were not available in
electronic format have been scanned and converted to Adobe portable document format
(pdf) files. Proceedings that were available in electronic format have been converted
directly to pdf files. Suggestions on further content for the website (such as useful links)
are welcome and may be directed to the Advisory Committee or webmaster as obtained
from the website.
Results
All Proceedings are now available for viewing and downloading.
Conclusions
Electronic access to the previous Targetry Workshops makes the historical record on the
progress of target development available for the wider community.
107
GE PETtrace MeI Microlab: Six Years' Experience with System*
C.S. Dence, W. Margenau, H. Lee, C. Bognar, S.M. Moerlein, M. Al-Qahtani and
M.J. Welch
Dept of Radiology, Washington University School of Medicine, St. Louis, MO 63110, USA
Aims
We wish to describe our six years' experience with the GE PETtrace MeI Microlab (the
box) since its installation in March, 1996. We are particularly interested in describing the
problems encountered after receiving the unit and our approach to solve them. We will
detail the use of the box in our current radiosyntheses.
Overview
The GE PETtrace MeI Microlab is a fast and reliable system to synthesize C-11 MeI by
free radical formation from C-11 methane in the presence of iodine at 720ºC. The box was
acquired in March of 1996, one of the early units on the market. Soon after receiving the
unit we started to have problems with the Process Unit because of clogging of the needle
piercing the lower septum of the iodination column. This problem was traced to
condensation of iodine around the needle at the end of synthesis during the cooling phase.
This problem caused considerable downtime while cleaning the faulty piece and
reconditioning the column. A permanent solution involved the elimination of the
septa/needles in favor of O-rings located around the Teflon piece, where the quartz
column (containing the iodine) is inserted by gentle pressure from above. This simple
solution later appeared in a newer model of the box by GE Medical System (GEMS), as
described in their manual Rev.1D November 30, 1998. We have also adapted an
inexpensive clear plastic tube to act as a holder for refilling the quartz iodination tube.
This plastic tube is marked permanently with the distances to be refilled with quartz wool,
ascarite and iodine, thus making the preparation of a new iodination tube a quick and
accurate operation from run to run. Another change in our system has been the elimination
of water from the ascarite on top of the quartz. Wet ascarite was a cause for plugging of
the iodination tube, and slow recirculation during C-11 MeI production. At the present
time, we consistently use the iodination column for 10-12 runs before a new column is
installed. We do not remove the iodination column from the system during longer periods
of non-use (such as weekends), thus increasing the chemical safety for the operator.
The box is located in a hot cell at a distance of about 35 feet from the methylation gantry
currently in use. The C-11 MeI is transported with a polyetherether-ketone (PEEK) line
0.010" ID x 1/16" OD, the flow from the box is modulated with a flow controller to 8-10
mL/min and the total tranfer time in the box is reset to 350 seconds. The radioactivity is
first passed through a glass tube 15 cm long X 1.2 cm ID filled with NaOH pellets. From
May 2000 to the present, we have used this methylation system to produce over 400
batches of MeI with a yield of approximately 355 Ci. We routinely run a short
bombardment (30 sec or less) to check the box after a new column is installed. Some of
108
the applications of this device have been for the labeling (n=number of preparations) of
raclopride (114), lactic acid (29), omega fatty acids (41), flumazenil (45), NMB (19), and
a number of other tracers. Specific activities at EOS for clinical batches have exceeded
92.5 GBq (2.5 Ci)/umol for raclopride and 55.5 GBq (1.5 Ci)/umol for flumazenil.
* This work supported in part by NIH Grant No. 5 PO1 HL 13851.
109
The Installation of a Multi-Radiopharmaceutical P.E.T. Facility at
Washington University in St.Louis, Mo, USA.
G. Gaehle, S. Schwarz, M. Mueller and M.J. Welch
Washington University Medical School, Mallinckrodt Institute of Radiology, 510 South

Kingshighway, Box 8225, St.Louis, MO 63110, USA
A new compact PET pharmacy has been installed at Washington University to meet
impending FDA regulations and increased demand for radiopharmaceuticals used in
clinical PET. The new pharmacy was installed at the site of the tandem cascade accelerator
that was decommissioned in April of 2001. The 1880 square foot pharmacy consists of a
production lab, quality control lab, reagent preparation lab, shipping and storage area and
an office. A CTI RDS111 cyclotron installed in the fall of 2001 supplies precursors in the
following form for radiopharmaceutical compounding and use, 18F fluoride, 18F fluorine,
11
C carbon dioxide, 11C carbon monoxide, 13N ammonia, 15O oxygen, 15O carbon dioxide
and 15O carbon monoxide.
The pharmaceutical production facility consist of eight minicells BBS2 and one general
hotcell MIP-1 from Comecer. Radiopharmaceutical production is accomplished using
chemistry modules from CTI. F-18 FDG is produced using the CTI QuadRx module. C-11
acetate and C-11 Methyliodide are produced using the CTI grinard module. We also
installed the CTI hydrogen cyanide and methylation module. This leaves us with three
minicells for further expansion.
The quality control lab has a general purpose fume hood, and a laminar flow hot cell
MIP1-1P LAF 1250 from Comecer used for dispensing. The quality control lab is
equipped with a Bioscan AR2000 TLC, an Agilent 6890 gas chromatograph and a Agilent
1100 high performance liquid chromatograph for chemical and radiochemical purity
analysis. A Capintec CRC-15W and a Aptec-NRC multi channel analyzer are available for
dose measurement and radionuclidic purity testing. Finally a Fisher Scientific isotemp
incubator is used for pyrogen and sterility testing.
The reagent lab houses a general fume hood and a class II laminar flow hood from
Labconco. An isotemp explosion proof refrigerator/freezer is used for reagent storage. The
equipment used in the reagent preparation lab consists of a dry oven from Fisher Scientific
and a Napco 9000D autoclave.
The facility began processing FDG for patient use in February 2002. 15O oxygen carbon
monoxide and water can be shipped to three different PET suites. The furthest PET suite
from the RDS111 cyclotron is 2500 feet away. 11C acetate and 13N ammonia are both
proceeding through the approval process for patient use.
110
A Research Resource in Radionuclide Research
D. W. McCarthy, H. M. Bigott, D.J. R. Rowland, T.A. Perkins, T. F. Voller, R. LaForest,
D. Reichert, M. R. Lewis and M. J. Welch
Mallinckrodt Institute of Radiology, Washington University School of Medicine, 510 S.

Kingshighway Blvd., St. Louis, MO 63110, USA
Aims
Since September 1999, Washington University has been funded as a Research Resource in
Radionuclide Research by the National Cancer Institute. The goals of the Resource
include providing novel radionuclides for collaborative research and other applications. In
addition, imaging facilities at Washington University, including the microPET (Concorde
Microsystems, Knoxville, TN), the first commercial device based on the design of Cherry
and colleagues (1), are being made available to collaborators to assist in the evaluation of
radiotracers for both diagnosis and therapy.
Introduction
64
Cu is a medium half-lived positron emitter, whose positron energy is similar to that of
18
F. This makes 64Cu a very attractive isotope for imaging studies utilizing the microPET.
Currently, 64Cu is being made available to researchers at Washington University and to 14
other institutions. Since the start of the Resource, 56 Curies of 64Cu have been produced
for studies at Washington University and for shipping to investigators across the country.
We are also making progress on the production of 76Br, 77Br, 124I, 86Y and 66Ga. We have
shipped small quantities of 76Br and plan to ship 124I and 77Br in the near future. In
addition, we are making available several short lived isotopes for collaborative studies
carried out at Washington University: 94mTc, 60Cu and 61Cu.
Due to the high energy positrons, some of these non-standard isotopes are less than ideal
for imaging. The imaging characteristics of these isotopes is being evaluated using a mini
Derenzo phantom and other phantoms (2, 3).
Conclusions
The National Cancer Institute’s funded Research Resource in Radionuclide Research has
provided nonstandard isotopes to investigators at Washington University as well as
investigators at other institutions. The resource has also made available imaging facilities,
such as the microPET, to collaborators. In addition, progress has been made in the
production of 76Br, 77Br, 124I, 86Y and 66Ga. These isotopes are being evaluated for their
imaging properties on the microPET.
111
This research was supported by NIH grants (R24CA86307) and (R24CA83060), and by
Department of Energy grants (DE-FG-0287ER60512) and (DE-FG-02-84ER60218).
References
(1) Cherry, S. R., et al., IEEE Transactions on Nuclear Science 44, 1151-1166 (1997).
(2) Laforest, R. et al., J. Label. Compound. Radiopharm., 44, Suppl.1 (2001).
(3) Laforest, R et al., High Resolution Imaging in Small Animals: Instrumentation,

Applications and Animal Handling, Rockville, Maryland, Sept 9-11, 2001.
112
Modified Module for the Production of C-11-Methyl Iodide from C-
11-Methane, Produced in Target
J. Medema, S.P. Tilkema, J. van Essen, H. Keizer, W. Vaalburg and P.H. Elsinga
PET-center, Groningen University Hospital, P.O. Box 30001, 9700 RB Groningen, The
Netherlands
Aims
Carbon-11 labelled methyliodide is a frequently used labelling agent for the production of
PET-tracers. Besides the classical ‘wet’ production method, i.e. reduction of C-11-CO2 to
C-11-methanol followed by hydrolysis to C-11-methyl iodide, the so-called ‘gasphase’
production method is being applied. Several groups published data on this production
method, in which C-11-methane is iodinated to form C-11-methyl iodide. The
commercially available GE-module is being widely applied for routine production. This
module uses C-11-CO2 as starting agent, which is being converted to C-11-methane.
The presently reported home-built module uses C-11-methane as input being produced in-
target in order to achieve maximal specific activity. The target is a Scanditronix
aluminium C-11-target, filled with N2 + 5 % H2. The ‘gasphase’ production of C-11-
methyl iodide from C-11-methane produced in target has been published earlier by Larsen
et al., but our module has been adapted to be optimal compatible with the MC-17
cyclotron environment.
Methods
C-11-Methane produced in target is trapped on a liquid nitrogen cooled Porapak N trap

located in the cyclotron vault. The target and the trap were evacuated by a vacuum pump
to remove the hydrogen gas. The trap was warmed to ambient temperature. The C-11-
methane-in-target production has been implemented in the STEP-5 software. The C-11-
methane was transferred to the helium preconditioned module in the hot cell over a
distance of 20 m using a helium flow of 500 ml/min. A carbosphere column equipped with
two radioactivity detectors at both ends was used to collect the C-11-methane. When all
C-11-methane was present on the carbosphere, the recirculation process was started by
activation of the membrane pump. The following parameters were used. Temperature of
empty tube: 675ºC, temperature I2 85-90ºC. Recirculation flow 400 ml/min. The formed
C-11-methyl iodide is being trapped at the top-end of the same carbosphere column.
Accumulation of C-11-methyl iodide was followed by the top-end radioactivity detector.
The radioactivity profile is a result of a gradually increasing C-11-methyl iodide
accumulation and still unreacted C-11-methane. The latter component shows an oscillating
pattern. C-11-Methyl iodide is released from the carbosphere by heating the carbosphere
to >300ºC and using a helium flow of 20-30 ml/min in the opposite direction. Heating of
carbosphere is achieved by moving the column into a preheated furnace pneumatically.
After the production the column is moved outwards to ensure fast cooling down and short
preparation time for the next run. The module was controlled by Labview software.
113
Results
Using this system, C-11-methyl iodide is produced with a radiochemical yield of 50-60 %
(EOB) with a synthesis time of 15 min and specific activities of >20,000 GBq/mmol
(EOB) were achieved. The system is ready for the next run within 10 min.
Conclusions
A fully automated 'gas phase' methyl iodide system has been developed that uses C-11-
methane as input, producing C-11-methyl iodide with high specific activities.
114
A Cryo-Buncher Based on a Ten-Port Switching Valve
Jon A. Nye1*, David W. Dick1, Micheal J. Schueller2, Mikael Jensen3 and Jerry Nickles1
1
Department of Medical Physics, Madison, WI,
2
Brookhaven National Lab, Upton, NY,
3
Royal Veterinary and Agriculture University, Fredericksburg, DK
Introduction
A device was constructed based on a 10-port switching valve developed for the
administration of gaseous C14O2 in concentrated form to the lung for cerebral activation
studies. The oxygen-14 tracer has desirable characteristics (t1/2 = 71 sec, β+, 2314 keV γ),
which allow for repeated studies at an increased throughput. The 10-port switching valve
is plumbed in the “column regeneration” mode where a steady stream of the C14O2 tracer
is trapped at LO2 temperatures while the other loop is subsequently heated and eluted into
a stream of He gas. The trap and release mechanism permits repetitive patient studies with
a controllable delivery time between each administered dose. The delay time between
trapping and subsequent elution of the columns allows for 1-2 minute acquisition periods
followed by several minutes of delay time without any loss in activity for the column in
the “trapping” phase of the cycle.
Methods
A target gas mixture of natural nitrogen with one part per hundred oxygen is irradiated
with protons and monitored by a ruthenium-based optical probe. The irradiation is done on
a CTI RDS 112 negative-ion cyclotron at fixed proton energy of 11 MeV. Oxygen-14 is
produced by the following reaction with natural nitrogen, 14N(p,n)14O, with a measurable
thick target yield of 2 mCi/µA. Several contaminants are present in the output stream and
are summarized in the table below.
Reaction Q(MeV) A(ESOB) Chem Form Action

14
N(p,α)11C -2.92 60 mCi/µA CO2, CO Trap for later
use CO2, "
14
N(p.n)14O -5.93 2 O2 Pass on
14
N(p,pn)13N -10.55 0.1 N2 Can't trap,
innocuous
NOx Trap out
15 15
N(p,n) O -3.54 0.2 @ 0.3 % O2 Can't separate
from 14O2
*Correspondence to Jon Nye at jnye@students.wisc.edu
115
The target output stream is dominated by 11CO2 formed directly in the target chamber and
by oxidation of 11CO with the first CuO furnace. Any 11CO2 formed in the target chamber
is immediately trapped in soda lime coupled to the output stream
before reaching any furnaces. The remaining activity is piped into a
hood, stripped of 13NOx (Tewson et al), and sent though a CuO
furnace to convert the remaining 11CO to 11CO2, which is then
trapped in soda lime or LO2 for later use. The residual activity
stream, consisting of 13N2, 15O2 and 14O2, is then sent to a charcoal
furnace to convert the oxygen species to carbon monoxide and
dioxide products. Finally, the activity streams are passed through
another CuO to convert any last remaining carbon monoxide
compound to carbon dioxide. The final activity stream is then
Fig 1. Plumbing of sampled and evaluated on a GC system (PORPAK Q) showing an
10-port switching approximate 10:1 mixture with 13N2. Upon evaluation of the final
valve
product, the activity stream is passed through the cryo-concentrator
based on to the 10-port switching valve shown in Figure 1. The
trapping column is submerged in LO2 while the other loop is resistively heated to elute the
trapped C14O2 into a He stream.
Results
Production of C14O2 followed flow-through chemistry where the target gas was piped
through a series of traps and furnaces as discussed in the previous section. The trapped
activity was monitored with a heavily shielded BGO crystal coupled to a photomultiplier
tube. Energy windows were set to monitor the 511 keV annihilation photon and 2314 keV
coincident gamma. Figure 2 shows a drawing of the inline chemistry and graph of the
activity trapped over time. An approximate 6-fold increase in the concentration of activity
is achievable compared to a steady state delivery. It was observed that 2.5 mCi could be
trapped in a volume of 60 ml with a trapping time of 6.5 half-lives (rise time seen on
graph) and elution of the trapped activity is on the order of several seconds. This is an
ideal time period for a patient scan of 2-3 min and an interscan delay of several minutes
for decay and clearance of the remaining activity.
C14O2 that enters the lung is rapidly transported across the alveolar membrane and into the
blood stream. The transport of CO2 out of the lung and into the blood stream depends on
which element in the compound is labeled. Oxygen-14 labeled CO2 transfers its label to
the water pool, forming carbonic acid, labeled bicarbonate, HCO3-, and labeled water,
H214O on a time scale of milliseconds. Figure 3 compares some various inert diffusible
tracers detailing their physical characteristics and lung clearance that dictate the shape of
the arterial input function. The dominant characteristics when choosing to administer the
activity by pulsing rather than steady state is the clearance time of the tracer in the lung.
The behavior of the C14O2 distribution in the lung, brain, and expired breath are shown in
Figure 4. The fast clearance kinetics and delta function input is evident by the initial
decrease in activity in the lung and expired breath tracers.
116
Fig. 2. Schematic of flow through Fig. 3. Reasons why giving a pulsed input
chemistry for the production and trapping of C14O2 is something desirable.
of C14O2.
Fig. 4. Clearance of C14O2 from the lung, Fig. 5. Flow sensitivity of CO2 analogue
brain, and expired breath. tracers used for cerebral blood flow
activation.
Steady state delivery of isotopes used for cerebral brain mapping have suffered in their
ability to show sufficient contrast in activated regions. The sensitivity in steady state of
several CO2 analogues is shown in Figure 5. A peak in the sensitivity occurs with C14O2
where the physical decay rate λp=ln(2)/71 sec is perfectly matched to the gray matter flow
rate f=60 ml/min-dl. The short half-life, fast lung clearance, and ability to cryogenically
trap C14O2 allows delivery of an instantaneous gas pulse to a patient's lungs. This is ideally
suited for dynamic imaging of the impulse response function of regional cerebral blood
flow. The breath of gas is then held for several seconds and cleared from the lung by
normal breathing with a half-life of 4 sec. The bolus then travels through the heart
undergoing turbulent mixing and on to the blood brain barrier.
Conclusions
Cyrogenically trapping C14O2 for bolus injection to the lung takes advantage of the short
lung half-life and flow-through chemistry. In addition, delivery of the activity can be
automated for quick and repeated brain mapping studies. Oxygen-14, either as CO2 or
water, promises faster study cycling and more efficient utilization of the PET scanner.
These issues are becoming critical in today's clinical setting.
117
Targetry and Chemistry for the Production of [O-14]: Application
to Radioactive Ion Beam Experiments
J. P. O’Neil, C. A. Ramsey, J. Powell and J. Cerny
Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA, U.S.A
A number of interesting problems in nuclear astrophysics as well as nuclear structure are

best studied with radioactive ion beams. BEARS is an ongoing project to provide light-ion
radioactive beam capability to the 88-inch Cyclotron at LBNL. To date work has focussed
on the production of accelerated Carbon-11 with beams of more than 108 ions/sec being
used for various experiments.
Development of Oxygen-14 as the second BEARS beam has presented considerable
technical challenges due to the short half-life of 71 seconds, the in-target radiation
chemistry of oxygen as well as the chemical processing of the target gas. The standard
techniques developed for the medical uses of Oxygen-15 involve the addition of
significant amounts of carrier oxygen not compatible with the high vacuum needs of the
88-Inch Cyclotron’s ion source.
We have developed a method for the production of Oxygen-14 as carrier-free water,
followed by a two-step chemical conversion to carbon dioxide, a gaseous carrier of the
Oxygen-14 compatible with the BEARS transport system.
Targets and chemical processing equipment have been incorporated into the CTI RDS-111
medical cyclotron and initial tests of producing and accelerating an Oxygen-14 beam have
been performed.
In addition to the uses of Oxygen-14 for nuclear science research, this short-lived, positron
emitting isotope shows potential in nuclear medicine imaging. For example: Fast PET
scanning for brain activation studies with [O-14]water. (Sajjad et al. J. Nucl. Med. (2000)
5, 244.)
Development of capabilities
Involvement of the Department of Functional Imaging in the BEARS project includes but
is not limited to:
Target design for the BIF CTI RDS-111 cyclotron
Target chemistry and post irradiation chemistry
Radioactive species production
In order to provide the capability to run continuous and automated production sequences
for days, modifications to the RDS-111 cyclotron and support facility have been made. As
a result, enhanced performance and reliability have been achieved.
The original BEARS system has been modified to allow the production of Oxygen-14
radioactive ion beams. National Instruments LabVIEW software and FieldPoint modules
are used to control the post irradiation chemistry as well as the gas tranfer between
118
cyclotrons. In addition the LabVIEW software monitors environmental safety interlocks
and is interfaced with the RDS control system to coordinate unload timing.
Initially, a Carbon-11 beam was developed as previously reported (J. Powell et al., Nucl.
Instr. and Meth. A 455 (2000) 452). The most recent Carbon–11 experiment ran
continuously for 60 hours and delivered over 100 Curies of [Carbon-11]CO2 to the ion
source of the 88” cyclotron over 300 meters away.
Development of an oxygen-14 radioactive ion beam
Oxygen-14 presents the challenges of a 71 second halflife and high reactivity.

To realize a working beam, a target and target material are needed which provide Oxygen-
14 that is:
1) Easily extracted from the target.
2) Quickly and efficiently converted to a chemical form suitable for rapid gas-
phase transportation to the 88” cyclotron.
3) Amenable to efficient trap and release with low carrier gas loading on the ECR
ion source.
Due to intense radiolysis within the target (>50 eV deposited per molecule in 40mA, 90
sec run), initial attempts at exchange reactions with additives such as CO failed.
Carbon monoxide radiolysis: 4CO ¤ CO2 + C3O2.

In addition, radical species, especially oxygen, have strong interactions with the target
walls and thus we have not been able to extract appreciable amounts of O-14 from
aluminum targets.
Adding even small amounts of stable oxygen as a carrier, as is routinely done in the case
of 15O production for medical, was not feasible since dilution with stable oxygen exceeds
pressure limitations of the high vacuum ECR ion source.
However, after additional experimentation, a mixture of 1-5 % hydrogen in nitrogen
irradiated in a gold plated target body produced H214O in good recoverable yield.
Conversion: H214O ¤ C14O ¤ [14O]CO2

14 14
O labeled carbon dioxide can be made from target produced O labeled water as
follows:
A) Separate H214O from other target gases including ammonia, hydrogen and
11
CH4 through momentary freezing at -40°C followed by resistive heating of
the trap.
B) Convert H214O to carbon monoxide by flowing over 1000°C graphite in a
helium stream.
C) Oxidize C14O to carbon dioxide over platinum at 180°C (add some O2 for this
step).
119
Fig. 1. Diagram of the apparatus used to produce O-14 labeled cabon dioxide
Comparison of O-14 and C-11 Production

Efficiencies: Oxygen-14 Carbon-11
Thick-target production 5 x 109/s 1 x 1011/s
Recovery from target ~50 % ~50 %
H214O ¤ [14O]CO2 ~60 % n/a
Activity decay ~30 % ~80 %
Ionization 3.6 % (6+) 11 % (4+)
Other efficiencies similar
From known 11C4+ beam on target => 1-2 x 108 ions/s --

estimate 14O6+ beam => 5 x 105 ions/s
Early test runs with poor beam diagnostics (tuned on 14N)

provided 14O6+ beam => ~ 1 x 105 ions/s
Acknowledgements
Due to the hard work of both the 88” Cyclotron staff and the BIF staff as well as support
from the U.S. Department of Energy (DE-AC03-76SF00098), a means of producing
Oxygen-14 RIBs has been developed and tested at Lawrence Berkeley National
Laboratory.
120
A Versatile and Ready-to-Automate Set-Up for Research and
Development of Radiopharmaceuticals
J. Sambre and J. Courtyn
PET Centre, Ghent University Hospital and Laboratory for Analytical Chemistry, Ghent
University, Institute for Nuclear Sciences, Proeftuinstraat 86, B-9000 Gent, Belgium
Aims
Automated systems for the production of radiopharmaceuticals are usually ‘dedicated’ and
not very flexible. We designed a versatile set-up to perform research and development
(R&D) of a wide range of labelled molecules. The set-up allows easy monitoring during
R&D and step-by-step full automation.
Methods
The main functions of the set-up are:

- transport of gases and liquids by a series of 7 multi-port rotary valves (Hamilton), using
Ar-pressure (liquids) and sometimes by peristaltic pumps,
- moving the reaction vessels by 2 hydraulic units for heating and cooling, driven by tap-
water-pressure,
- purification by high pressure liquid chromatography with automated injection,
- monitoring and recording temperature, gas flow, valve position and radioactivity via
PLC-(Programmable Logic Controller) to personal computer,
- process control by PLC.
Results
The set-up has already been applied for 11C labelled molecules: 11C-thymidine, 11C-
flumazenil, and 11C-acamprosate, produced from 11C-carbondioxyde via 11C-iodomethane
or 11C-acetyl chloride, as previously described (1, 2, 3). The optimisation of the process
parameters and the automation went side by side. The reliability was outstanding.
Conclusion
The design of the set-up - the combination of the multifunctional valve-system and PLC-
control - proved to be very suitable for the fast development of a series of carbon-11-based
radiopharmaceuticals.
The main innovation of the set-up is the use of a series of multi-port rotary valves. The
combination of these valves avoids the use of a very large number of separate 1- and 2-
way valves, tees and tube-fittings. This reduces significantly the dead volumes. Our
121
experience proved they are very reliable and the feedback assures safe operation. The
flexibility of the system allows easy switching between different types of synthesis.
References
(1) Goethals P. et al. A remotely controlled production system for routine preparation of
[Methyl-C-11]Thymidine APPL RADIAT ISOTOPES 43: (7) 952-954 JUL 1992.
(2) Maziere M. Synthesis of Ethyl 8-Fluoro-5,6-dihydro-5-[C-11]methyl-6-oxo-4H-
imidazo[1,5-a][1,4]benzodiazepine-3-carboxylate (RO 15.1788-C-11): A specific
radioligand for the in vivo study of central benzodiazepine receptors by positron
emission tomography INT J APPL RADIAT ISOT 35: (10) 973-976 1984.
(3) Courtyn J. et al. Synthesis of C-11-labelled acamprosate for PET studies.
122
Ten Years Oxygen-15 Production at the Turku University Central
Hospital
H.T. Sipilä and M. Teräs
Turku PET Centre, Turku University Central Hospital, PO Box 52, 20521 Turku, Finland
In 1990 a decision was made to install an oxygen generator at the Turku University
Central Hospital. The Turku PET Centre has its PET-isotope production and
radiochemistry facilities at Åbo Akademi University which is located about 800 m from
the hospital. A low-energy (3.5 MeV) deuteron accelerator Cyclone 3 (IBA sa., Belgium)
was our solution for oxygen-15 production at the hospital.
Installation of the Cyclone 3 started in the beginning of January 1992. First beam was
found in the end of April and the machine was preliminary accepted in the end of June
1992. However, the performance of the machine was not satisfying because the magnet
poles and sectors were still experimental. The magnet had to be rebuilt. After rebuilding
the magnet the performance of the machine was improved considerably. The final
acceptance tests were performed in August 1993. During the first year, 1992-1993, after
the preliminary acceptance, oxygen-15 production and oxygen-15 chemistry was
established. First patient study was done 28th of April in 1993.
The Cyclone 3 was installed in a vault about 60 meters form the hotlab and the PET
scanners. This installation was in use about six years until the scanners were moved into
the new PET building in March 1999. Cyclone 3 is still operated from the old vault and
the present distance between Cyclone 3 and hotlab is about 180 m. Target gas flow rate
can be maintained at 500 ml/min in the three times longer FEP tube (o.d. 1/4", i.d. 2 mm)
with increased target gas preussure of 2.1 bar againts the 7 µm thick titanium target
window. Because of the long distance between the hotlab and cyclotron the data
communication of the cyclotron control system had to be improved with a RS-232
converter (Westermo, Access MA-44).
When the distance was 60 m a target gas pressure of 1.9 bar was used and the oxygen-15
saturation yield was about 260 MBq/µA. The oxygen-15 saturation yield in the new
laboratory is about 190 MBq/µA which has been satisfying for PET studies. Naturally the
main oxygen-15 product has been oxygen-15 labelled water for blood flow studies. During
the last few years we have developed a new radiowater production system (1) which is
more safe and easier to use for different kinds of blood flow studies.
Table 1. Oxygen-15 studies during the years 1993 - 2001
Brain Heart Muscle Oncology Other Total

H2O 3578 2207 469 39 46 6339
CO 83 879 106 19 8 1095
O2 74 232 261 7 574
Total 3735 3318 836 58 61 8008
123
Rerference
(1) H.T. Sipila, J.C. Clark, O. Peltola and M. Teras, An automatic O-15-H2O production-
system for heart and brain studies, Journal Labelled Compounds and Radio-
pharmaceuticals, 2001, Vol 44. Suppl. 1 S1-S1082.
124
CS-30 Cyclotron Operation at the Duke University PET Facility:
Report from the Labs
B. Wieland, C. Illan, M. Dailey, S. Murphy and R. Coleman
PET Facility, Duke University Medical Center, PO Box 3808, Durham, NC 27710
The Duke CS-30 cyclotron is a positive-ion four-particle machine accelerating protons to

27 MeV, deuterons to 15 MeV, 3He to 39 MeV and alphas to 30 MeV. The external beam
is magnetically switched to seven different target stations, which are currently outfitted
with two proton 18F targets, a deuteron 15O target, a proton 13N ammonia target, a proton
11
C target, a tandem proton 15O water / 13N ammonia target, and a beam port for
experimental targets. We have previously produced 124I from proton bombardment of
enriched tellurium targets, and have bombarded a porous carbon target with 3He. We
operate an instrumented internal beam target station that has been used for six years for the
high level production of 211At (monoclonal antibody label for glioblastoma therapy) by
bombarding bismuth targets with alpha particles.
The cyclotron is currently operated approximately1000 hours per year, the majority of
which is used to produce 18F for clinical PET. F-18 FDG is typically synthesized three
times a day Tuesday through Friday to support a PET scanner throughput of about 3000
patients per year. 15O water and CO, and 13N ammonia are produced for 250 research
studies per year. 211At production averages about an hour per week for chemistry studies,
with a four-hour run for patient therapy dose when required. Two new 18F target systems
are under development using the experimental beam line: a thermosyphon batch target and
a regenerative turbine recirculating target. The purpose of these targets is to significantly
increase the beam current that a target can tolerate, thus increasing the 18F produced and
making more effective use of the high beam power available from current and future
generation cyclotrons.
The cyclotron staff consists of Bruce Wieland, PhD/PE, Claudio Illan, MSEE/MSNE, and
operators Shawn Murphy, BSNE, and Michael Dailey, AE Mechanical/Electronics. The
cyclotron is typically staffed from 4 AM to 6 PM. Two part-time operators (physics
graduate students) are also available for late night runs as required.
125
126
Radiochemistry and Automation
OPC Based Control Software
K.R. Buckley
Aim
To develop a method of implementing chemistry module control systems that allow simple
and rapid development of a graphical user interface and supports sequencing for
automation tasks.
Methods
I reported at the previous WTTC meeting (http://www.triumf.ca/wttc) on our use of

Lookout (National Instruments http://www.ni.com) control software for our chemistry
control systems. Lookout is a commercial software package that allows the rapid
development of graphical user interfaces to control hardware. Lookout lacks a simple
environment for implementing scripting or sequence execution necessary to effectively
implement automated synthesis. A software model known as component object modelling
implemented by Microsoft (http://www.microsoft.com) as object linking and embedding
(OLE) allows for the communication of multiple software processes running on one host
or distributed hosts. This model has been adopted by the process control community and
implemented as OLE for Process Control (OPC) (OPC Foundation
http://www.opcfoundation.org). The practical effect of this is to provide a software layer
(the OPC server) that interacts with the control hardware and interacts with any other
software program (the OPC client) that implements the OPC protocol. In this way multiple
software programs (on the same or different computers) can communicate with a single
hardware system via the OPC server.
Results
I have implemented this model on our chemistry synthesis systems to partition the GUI
tasks to Lookout and the automation tasks to Visual Basic (Microsoft Corporation
http://www.microsoft.com) programs. In this way the strengths of each environment are
utilized without the weaknesses.
Conclusions
OPC provides an add-in for computer control systems which allows implementing
automation tasks outside commercial software packages which may not support such tasks.
128
Radiochemical Tl-201 Production Process
G.A. Casale, G.J. Maslat and J.L. Lago
Cyclotron Facility, Ezeiza Atomic Centre. Comision Nacional de Energía Atómica. Av

Libertador 8250 Bs As , ARGENTINA
Aims
201
Tl, in spite of the previsions, continuous being the most frequently used cyclotron
produced radionuclide in SPECT nuclear medicine; in the present work was sought to
obtain a productive chromatographic process, that permit to obviate the classic lead
precipitation and extraction step.
Methods
The distribution KD constants, were determined for Tl+ ,Tl3+ and Pb2+ in cationic exchange
resins of different crosslinked, particle sized and different concentrations of HCl and
HNO3. Thallium was labelled with 201Tl and correlated by atomic emission at 3446 Å
(sensibility 1 ppm ). Lead was labelled with 210Pb and by atomic absorption at 2170 Å
sensibility 0.01 ppm. Chromatographs were developed using a peristaltic pump, a fraction
collector coupled to a NaI(Tl) detector with an integrative and registering system.
Chromatographic Development
HNO .1M HCL .1M HCL 2M
The simulated cyclotron target used for 150
development was prepared with 700 mg
of TlNO3, 1 mg of PbNO3 and labels of 100
Tl Pb
201
Tl and 210Pb. With a low pressure
chromatographic equipment, these targets 50
were eluted using different types of
columns, resin bulks, gradients and
0
solvent flows. This development was 0 50 100 150 200 250 300 350 400 450
based on optimising Pb-Tl separation,
seeking the greater number of theoretical
[ml]
plates getting the following conditions:
Resin type :AG 50W-X12 200-400
Solvent : HNO3 0.1 M
Resin bulk : 120 g
Flow : 1.3 ml/ min.
Temperature : 25°C
129
These conditions give the best relationship for KD Pb2+ / KD Tl+ in HNO3 0.1 M. More
than 99.9 % of Tl+ was eluted and all Pb2+ remained in the column. The production
process is the charge of the dissolved irradiated 203Tl in the column, washed the column
with HNO3 0.1 M, 32 Hs decay time, elution of 201Tl produced in column with HCl 0.1 M,
dryness, and physiological solution.
Results
This method was tested in the laboratories of Ezeiza Atomic Centre Cyclotron, Buenos
Aires and is working from -97 up to now in 201Tl production. Earlier test were performed
with natural Tl as target. In these conditions large quantities of 203Pb (which was used to
evaluate withholding of Pb into the column), were produced.
Yields obtained: Withholding of Pb2+ into the column: 99,8 %; Radionuclidic purity :
201
Tl: 100 % 200Tl: 1 % 200Tl: 0,4 %; Ratio 201Tl exp./ theor.: 87-95 %; Recovery of 203Tl:
99,9 %
Conclusions
The radiochromatographic method has great advantages, avoids precipitation and

extraction, is simple, easy to automate and low maintenance. The high number of
theoretical plates of the column produces high Tl quality. In the other contribution
submitted (G.J. Maslat and G.A. Casale) we are presenting an automated system for this
method.
130
Synthesis of C-11-Labelled Acamprosate for Pharmacodynamic
Studies by PET
J. Courtyn and J. Sambre
PET-Centre, Laboratory for Analytical Chemistry and Ghent University Hospital, Ghent
University, Institute for Nuclear Sciences, Proeftuinstraat 86, B-9000 Gent, Belgium
Aims
Although acamprosate is used therapeutically in Europe for several years to reduce relapse
in weaned alcoholics, the mechanisms by which acamprosate decreases alcohol ingestion
and relapse rates of alcoholics are still not clear. It is suggested that acamprosate acts by
reducing neuronal hyperexcitability, responsible for the acute alcohol withdrawal
syndrome, on the level of both excitatory glutamate type and inhibitory GABA (gamma-
aminobutyric acid) type neurotransmitter pathways. (1)
An interesting method of investigation of the pharmacodynamic properties of acamprosate

is to look for the drug action site. Since PET has made it possible to examine the
biodistribution of certain compounds in vivo, this imaging technique might be a powerful
instrument for the characterisation of these binding sites in the human brain. For this
application it is necessary to label the molecule of interest with a short-lived, positron-
emitting radionuclide as carbon-11 (T1/2=20.4 min). Therefore, this study aimed to develop
a synthesis for C-11-labelled acamprosate. From the structure of acamprosate or calcium-
N-acetyl homotaurinate, the acetyl group is considered to be the most accessible position.
Methods
Carbon-11 radionuclide was produced by irradiation of pure nitrogen with 18 MeV protons
accelerated in a cyclotron for 20 min at 15 µA beam intensity. During irradiation carbon-
11 reacts with traces oxygen to form [C-11]carbon dioxide. After irradiation [C-11]carbon
dioxide was collected in an automated set-up and transformed into [C-11]acetyl chloride,
according to the method described by Luthra et al. (2) This labelling agent finally reacted
with homotaurinate to form [C-11]N-acetyl homotaurinate.
The [C-11]N-acetyl homotaurinate was separated from the reaction mixture on a semi-
preparative reversed phase C18 column. Finally, the collected fraction was purified by SPE
(solid phase extraction), sent through a 0.22 µm sterile filter and formulated for
intravenous administration.
Results
The radiosynthesis lasted 45-50 min from the end of the radionuclide production to
provide [C-11]N-acetyl homotaurinate ready for intravenous injection. The radiochemical
yield was about 1.10 GBq. This corresponds to 12 % of initial [C-11]carbon dioxide
radioactivity, decay-corrected. The specific radioactivity at the end of synthesis was more
131
then 8 GBq/µmol. For the assessment of radiochemical purity and chemical purity
analytical high pressure liquid chromatography (HPLC) and gas chromatography were
used.
Conclusion
The ability to produce radiochemically pure [C-11]N-acetyl homotaurinate with a

sufficient specific radioactivity enables a study of the distribution of acamprosate in the
brain with PET.
References
(1) Dahchour A. and De Witte P. Prog. Neurobiol. 60: 343-362 (2000).

(2) Luthra S.K., Pike V.W. and Brady F. Appl. Radiat. Isot. 41: 471-476 (1990).
132
Rapid Screening of PET Volatile Organic Infusions Generated
from 2-F-18-FDG Synthesis by Solid Phase Microextraction
Coupled with GC/MS
A. Kanu, M. Dixon, Y. Yin Zu, J.M. Gillies, J. Bailey, P. Thomas and J. Zweit
Volatile Organic Compounds Group, DIAS, UMISTP O Box 88, Manchester M60 1QD,
UK
RTI, Paterson Institute for Cancer Research, Christie Hospital NHS Trust, Wilmslow
Road, Manchester, M20 4BX
Aims
2-F-18-Fluoro-2-Deoxy-D-glucose (2-FDG) is the most important commonly used

radiopharmaceutical in PET imaging. The F-18-FDG synthesis involves the conversion of
nucleophilic 18F- into F-18-FDG. The mannose triflate precursor, dissolved in acetonitrile,
is introduced in the reactor. A nucleophilic substitution reaction occurs, in which the
trifluoromethane sulfonate group is replaced by the F-18 atom, and results in 2-F-18-
fluoro-1,3,4,6-tetra-O-acetyl-D-glucose. During this time the reverse phase cartridges are
conditioned with ethanol and water. This study aims to develop an experimental method
using headspace solid phase microextraction (HS-SPME) coupled with GS/MS to
quantitatively determine the amounts of volatile organic solvent (VOC) that may be
present in clinical preparations of aqueous 2-FDG manufactured at the Paterson.
Methods
Pre-conditioned fused silica fibers (1 cm long, 75 µm diameter) coated with a

PDMS/Carboxen polymeric stationary phase were bonded with a high temperature epoxy
resin, to a stainless steel plunger that was installed in a syringe type holder. The syringe
type device allow penetration of the F-18-FDG sample vial septa in order to sample
directly from the vial head space. Organic analytes then absorbed/adsorbed onto the fiber
coating from the head space. Once removed from the head space the fiber was then
transferred to an Optic 2 injector fitted to a GC, where the analytes were thermally
desorbed from the coating and entered the GC column for separation and analysis.
Results
In order to develop a quantifiable experimental method for the analysis of VOC's

contained in F-18-FDG samples, it was necessary to optimise the SPME parameters such
as extraction time, extraction temperature and phase-volume ratio (volume of sample to
headspace). This was done in order to establish which of these parameters gave the
greatest extraction efficiency for the VOC's. It was found that an extraction time of 10
minutes, extraction temperature of 60°C and a pahse-volume ratio of 1.00 gave the greatest
extraction efficiency. Both the calibration curves generated for the acetonitrile and ethanol
samples and for low acetonitrile concentrations were found to be linear with correlation
coefficients calculated to be 0.9963, 0.9913 and 0.8227 respectively. The detection limits
133
for the experimental method developed were in the order of 2.52 ppm for ethanol and 0.49
ppm for acetonitrile.
Conclusion
This study has established that this experimental method is fast, inexpensive and solvent
free. It has also demonstrated an ability to analyse ppm residues of both acetonitrile and
ethanol contained within clinical preparations of F-18-FDG.
134
Direct PC Automation of Radiochemical Tl-201 Process
G.J. Maslat and G.A. Casale
Cyclotron Facility, Ezeiza Atomic Centre. Comision Nacional de Energía Atómica. Av

Libertador 8250 Bs As , ARGENTINA
Aims
In order to continue our chromatographic radiochemical method for cyclotron irradiated

Thalium targets, an automated system was developed. The philosophy was an equipment
with both possibilities: auto/manual, commanded by an open software, with a
radiochemical friendly script language. We called it “Direct PC” because all the systems
are directly interfaced to a common PC without PLC or dedicated electronics systems. The
PC microprocessor has all the information of the radiochemical process.
Methods
Two main programs drive the module: “Main

Control Program“ and “Radio-chemical Editor
Program”. Both developed in Visual Basic 3.0
for 16 bits machine or 6.0 for 32 bits. The
radiochemical module has an active part in a hot
cell (150mm lead) and an inactive one outside
the hot cell with different reagents pumps and
manifold selection valves without risk of
radiation damage. The active module consists
only of a dissolver, two Rheodyne PEEK valves,
exchange column and an evaporator plumbed by
tygon/teflon lined tubes. These valves are
excellent in high radiation fields.
All the components are interfaced directly to the

PC via In/Out ports made with the well known
8255PPI directly interfaced to the PC bus,
drivers and relays for the electrovalves action.
The analogs ports with the MAX147, (12bit, 8
channels) measures the temperatures of target,
Tl+/Tl0 potential during dissolution, valves
position and bubble detector. The pump
GILSON 402 Dilutor, is interfaced via RS232.
135
Results
The equipment was tested in cold and hot chemistry, all the radiochemical parameters were
repeated obtaining the same laboratory results:
Pb2+ Column retention: >99.5 % 201

Tl elution after decay time: >95 %
203
Tl recovery: >99.9 % Tl+ total in batch: < 2 ppm
203 200
Pb in batch(with postcolumn):0.003 % Tl/202Tl in batch: 1-2 % / 0.1 % (depends
of irradiation parameters)
Processing Times: First Process: Dissolution, load, wash column: 1.45 h

Second Process: Tl elution, dryness, batch extraction: 3 h
This equipment has been working since 1997 up to now. The average of activity has been
55.5 GBq/batch (1.5 Ci/batch) (80-90 µA, 10-12 h irradiation).
Conclusions
The direct PC automation has great advantages: is simple, not too much electronic
involved, the microprocessor has all the parameters with total control. The graphical
interfaces are easily operated, the chemist can change differents radiochemical parameters,
like flows, volumes, temperatures, reagents, etc. The accuracy, repeatability, minimum
sizes, low personal dose, low maintenance (only a column by process and o-ring dissolver
each 6 process) during these years, make the whole system very successful.
136
Experiences with the Amercare Syringe Fill Station
Y. van der Knaap1, H. Duiker2, G.Wilson3, J. Medema1, N.H. Hendrikse1 and
P.H. Elsinga1
1
PET-center, Groningen University Hospital, P.O. Box 30001, 9700 RB Groningen, The
Netherlands
2
Von Gahlen BV, P.O. Box 25, 6940 BA Didam, The Netherlands
3
Amercare Ltd Thame park road, Thame Oxfordshire, OX9 3UH UK
Aims
Because of the increasing demand for FDG and the concomitant manual dispensing of
patient doses, the finger radiation burden has become a serious issue that had to be
adressed. To reduce the finger radiation dose, the Amercare dispensing unit had been
selected for several reasons: 1) patient doses are being delivered in a shielded 10 ml
syringe, 2) fast preparation of a bolus (< 3 min), 3) application to other PET-tracers, 4) use
of sterile disposables only, 5) accuracy of radioactivity measurement within 10 %, 6)
GMP compliance. Here we present our experience with the dispensing unit over the last
year of full operation.
Methods
The fill station has been developed to completely replicate the manual syringe filling
process. It has a central syringe and multidose vial position and is capable of handling
shielding up to 20 mm of tungsten. The calculated volume of product is entered by the
operating console (FILL-action). We use the option to fill the syringe to 8 ml with saline
(ADD-action). To measure the activity (MBq/ml) a PIN silicon photodiode is integrated,
which signal is translated to absolute radioactivity on a LCD display. The photodiode is
linear in the range of 50-1000 MBq and is located in such a way that air bubbles > 0.5 ml
are detected by appearance of discrepancy between measured and calculated amount (>15
% deviation). The shielded syringe can easily be taken out of the system and put in a lead
shielded transport system. At the PET-camera the shielded syringe is connected to the
injection cannula. After installation of the fill station, several optimation actions have been
carried out to increase accuracy and repeatability. During the last year the following
parameters were documented: accuracy of the FILL-action, correlation between
radioactivity as measured by a dosecalibrator and by the photodiode after the ADD-action.
Results
The overall accuracy of the FILL-action is 95 +/- 6 % (n=224). This number fell below 90
% at fill volumes < 0.5 ml. On average, at volumes of 0.3-0.5 ml 86 % of the calculated
volume was dispensed, whereas at volumes < 0.3 ml this number was 73 %. The relatively
low accuracy below 0.5 ml forced us to split the FDG-multidose vial in two portions when
the concentration of FDG was > 400 MBq/ml. This splitting action is also performed by
the Amercare syringe fill station. On a few occasions (3 % of the total), a substantial
137
amount of air (1-3 ml) was observed in the syringe. In this case, a discrepancy between the
LCD-readout and the desired dose was observed. The correlation coefficient between the
radioactivity in the syringe as measured by the dose calibrator and the photodiode is 0.94
+/- 0.06 (n=55). Because the radiation dose is 15-20 µSv/h at 1 m distance from the
tungsten shielded syringe, containing 500 MBq FDG, the fill station has been put in a
lead-shielded laminar air flow hood to create a safe working environment from radiation
protection point of view.
Conclusions
We successfully implemented the syringe fill station, which plays a crucial role in the
daily manipulations with FDG. The finger radiation dose has been reduced to almost zero
during dispensing and injection into the patient.
138
Radiopharmaceutical Production Quality Control Considerations
S.W. Schwarz, K. Lechner, M.J. Mueller, G. Gaehle and M.J. Welch
Mallinckrodt Institute of Radiology (MIR), Washington University School of Medicine,

510 S. Kingshighway, St.Louis, MO 63110, USA
A large number of radiopharmaceuticals are being utilized for routine clinical, clinical
research and development. At Washington University in St. Louis we operate three
cyclotrons for patient/animal imaging on six PET scanners in four separate buildings.
They are a Cyclotron Corporation CS15 cyclotron which currently can only accelerate
protons, a Japan Steel Works 16/8 cyclotron which can accelerate protons and deuterons
and a RDS 111 cyclotron which produce only protons. These cyclotrons produce
radiopharmaceuticals for two clinical PET scanners, three research human. Currently 18
radiopharmaceuticals are produced routinely for human use. Quality control often
becomes the limiting factor in scheduling. Our QC division has two HPLC systems, a
Waters and an Agilent®, one Varian® GC, a Waters LC/MS, an ion chromatography
system, a dose calibrator, and a germanium detector. BET and sterility testing are also
performed on all injectable human radiopharmaceuticals.
Table 1 - Examples of Daily Scheduling
DATE/TIME LOCATION RaPh QC SYSTEM

3/20/02 7 MIR FDG (8 patients) LC
: 0800 Clinical PET GC-residual solvents
ECAT and C-PET Dose Calibrator
: 0830 East Bldg FDOPA HPLC System 1
Siemens 943 Dose Calibrator
: 0900 7 MIR O-15 Water GC
Research HR+ Dose Calibrator
: 0900 4CSRB MicroPET FDG dose NA
: 0915 7 MIR C-11 Acetate HPLC System 2
Research HR+ GC residual solvents
Dose Calibrator
: 0915-1000 4CSRB Research Tc-94m target
------------
: 1300 7 MIR C-11 Glucose HPLC System 2
: 1300 7 MIR F-18 FES HPLC System 2
: 1400 10 NICU C-11 FMZ HPLC System 1
ECAT Dose Calibrator
An example of a typical production schedule is shown in Table 1 above. Given the

number of radiopharmaceuticals produced on a daily basis, it is necessary to schedule
production of isotope, synthesis of the compound, and the quality control to make sure
that all can be accomplished pre-release as required.
139
Synthesis of Fluorine-18 Labeled Recombinant Annexin-V
Derivative, for Identification and Quantification of Apoptotic
Cells with PET
S. Zijlstra, J. Gunawan and W. Burchert
Herz- und Diabeteszentrum Nordrhein-Westfalen, Universitätsklinik der Ruhr-

Universität, Bochum, Bad Oeynhausen, Germany
Apoptosis has been recognized as an CF3SO3-

+
important process for the cardiovascular (CH 3)3N CO2C2H5
system. Acute cardiovascular incidences
(myocardial infarction, myocarditis and
18F-
heart failure) as well as other
cardiovascular diseases (atherosclerosis,
vascular - and cardiac allograft rejection 18F CO2C2H5
and diabetes) are characterized
pathologically by apoptosis. Increasing 1) NaOH
knowledge of the apoptotic process in 2) HCl
relevance to the cardiovascular system,
may develop new strategies for the 18F CO2H
therapy of cardiovascular diseases.
Apoptosis is a form of programmed cell
death, which is gene-regulated. During O-[N-succinimidyl]-N,N,N,N-
tetramethyluronium tetrafluoroborate
apoptotis, phosphatidyl-serine O
externalizes from the inner leaflet of the
plasma membrane to the outer leaflet. 18F COO N
An endogenous human protein that has
high affinity for binding to phosphatidyl- O
serine is Annexin V. In this report, we
describe the synthesis of fluorine-18 Annexin-V
labeled recombinant Annexin-V derivate
(4-[18F]-FBA). O
18F C NH A
In a four step procedure (figure 1), 4-
[18F]-FBA is synthesized in a 4-[ 18F]fluorobenzoyl-Annexin-V
microcomputer controlled, automated Figure 1. Synthesis of 4-[18F]-FBA
module (figure 2) within 90 minutes. The
radiochemical yield is in the range of 15-
20 % (corrected for decay) with a specific
activity of more than 3.5 GBq/µmol. The
radiochemical purity is higher as 95 %.
Batches of 4-[18F]-FBA of up to 2000
MBq were prepared from 25 GBq [18F]-
fluoride.
140
Radiation Sources and Radiodetectors
Compact, Solid-State Radiation Detectors for use in PET Isotope
and Radio-Chemistry Laboratories
L.R. Carroll1 and M.A. Jackson2

1
Carroll & Ramsey Associates, Berkeley, CA, USA
2
GE Medical Systems, Waukesha,WI, USA
Abstract We present a collection of compact and novel solid-state radiation detector systems
used to monitor the quantity, position, and flow of activity in radioactive fluids and gasses in
radio-chemistry laboratory processes, apparatus, and applications. The detectors described
below include:
1) Multichannel systems incorporating relatively low-sensitivity, subminiature Si diode
probes for use in tight spaces inside process-automation apparatus;
2) High-sensitivity systems with compact probes incorporating small (1 cu cm) scintillating
crystals coupled to Silicon PIN diodes for analytic high-performance liquid
chromatography (HPLC) and gas chromatography (GC);
3) Very high-sensitivity probes incorporating large (50 mm x 50 mm x 25 mm) CsI(Tl)
crystals for use as stack-effluent monitors; and
4) A compact bench-top scanning detector system for thin-layer chromatography (TLC)
applications. The TLC scanner incorporates a multi-mode Si diode detector is designated
OMNI-RADtm and is intended for use with any of the commonly-used beta-emitting,
positron-emitting and low-energy gamma-emitting radio-nuclides.
Multi-Channel Radiation Detector The model 101-HDC3 is a radiation detector system

intended for process monitoring applications in laboratory hot cells. The system comprises a
set of up to three miniature detector probes, plus a three-channel trans-resistance (current-to-
voltage converter) amplifier. The maximum gain of the amplifier in each detector channel is
nominally 5 x l09 ohms, that is, a diode current of one nano-ampere produces a voltage of 5
volts at the output of the amplifier. A multi-turn trim potentiometer on the output of each
amplifier chain allows the user to adjust the overall gain of the system over a range of
approximately 10:1.
The active element in the detector probe is a silicon PIN diode which is enclosed in a
small metal capsule to shield it against light and stray electro-magnetic fields. Each probe
is connected to its amplifier through a length of small-diameter cable. The probes operate
in "DC" mode and are intended to be used in close proximity to concentrated sources of
gamma-emitting radio-nuclides. The output of each detector channel may be monitored by
any high-input-impedance recording device: voltmeter, chart recorder, computer data
acquisition system, etc.
142
High-Sensitivity Radiation Detector The model 105-S is a high-sensitivity radiation
detector system used for HPLC, flow monitoring in radiochemical synthesis, monitoring
of stack effluents, and similar applications.
In HPLC and flow-monitoring applications, the system employs a compact detector probe (2
cm x 2 cm x 3 cm) in conjunction with a bench-top amplifier / console unit (7.75"W x 7.5"D
x 2.5"H). The sensitive element of the detector probe is a 1 cm3 CsI(Tl) scintillating crystal,
optically coupled to a 1 cm2 silicon PIN diode which, in turn, is connected to a charge-
integrating1 preamplifier.
143
Stable, reliable operation at low photon fluxes is enhanced by operating the detector in
AC-coupled pulse-mode. In this mode of operation, individual gamma ray photon
interactions in the scintillating crystal are converted to flashes of light which are then
converted in the diode / preamplifier to discrete pulses of current. The pulses of current are
amplified, thresholded, and integrated to produce a "DC" signal which is proportional to
the count rate of photons which exceed threshold.
1
Covered by one or more of the following US Patents: 5,990,745; 6,054,705.
For stack-monitor applications, the same console unit is used with a larger, more sensitive
probe: a 50 mm x 50 mm x 25 mm CsI(Tl) scintillating crystal which is optically coupled
to a 10mm x 20mm silicon PIN diode plus charge-integrating preamplifier. The probe is
epoxy-cast for protection against moisture incursion, and sealed into a compact 65 mm x
65 mm x 35 mm aluminum enclosure for convenient shielding.
OMNI-RADtm Detector for thin-layer chromatography This versatile, PIN-diode-based

module is used for scanning TLC plates which have been ‘spotted’ with compounds
labeled with commonly-used medical and research isotopes, including low-energy gamma
emitters, positron-emitters, or beta (-) emitters.
Beta Detection: In beta (or positron) mode the detector module utilizes anti-parallel
(opposed-polarity), back-to-back, 1 sq. cm. PIN-diodes mounted behind a 3mm wide slit
in a 1/4" thick lead plate centered over the scanning bed. The diode which is closest to the
TLC plate responds to both beta’s and gamma’s emitted from the sample; the second
diode, which is “shadowed” by the first diode, responds only to gamma’s. By
electronically subtracting the second (gamma) signal from the first (beta plus gamma)
signal, we have a composite detector which, in effect, responds to beta’s only, providing
excellent spatial resolution for beta + (positron) or beta (-) emitters, but without the need for
bulky lead shielding -- even in the presence of an intense gamma background.
144
The effectiveness of the concept is illustrated in the ‘Mock” TLC scan, where the sample
under test (56Co) emits both positrons (Emax = 1.459 MeV) as well as intensely penetrating
gamma rays (846, 1240, 1760, 2600 keV, 3260 keV and others).
Gamma Detection: In gamma detection mode, the signals from the two detector diodes are
summed instead of subtracted. Spatial resolution for low-energy gamma-emitters (~140
keV max.) is provided by a 3 mm wide slot cut in the 1/4" thick lead bottom plate.
EZ-SCAN tm is a compact, bench-top TLC scanner / detector system that occupies a

relatively small foot-print on scarce lab-bench real-estate. The system is designed for
routine use in a busy lab environment. Using EZ-SCAN with the Omni-Rad detector
module, one can ‘spot’ a small sample drawn directly from a production run, without
requiring any dilution. The annotated thumb-nail image at the right is reproduced from a
time-record from PeakSimpletm – a computer-based Chromatography data acquisition
system. This was a ‘mock’ scan using a TLC plate on which small drops of 18F-labeled
145
fluoride were placed by hand. The plate was later cut up and the pieces assayed in order to
ascertain the amount of activity represented in each peak. Total width of the plot
represents 5 minutes time duration for a 10 cm scan.
146
Estimating the Radiation Source Term for PET Isotope Targets
L. R. Carroll
Carroll & Ramsey Associates, Berkeley, CA 94710, USA
ABSTRACT An estimate of the source term - the neutron fluence and spectra for nuclear
reactions commonly used for producing PET isotopes - is required to validate the design
of shielding, and to estimate the potential for activation of materials and components in
and around the cyclotron.
We present data on neutron fluence and spectra for the most commonly used (p,n) PET
isotope nuclear reactions, including 18O(p,n)18F for proton energies of 11 MeV and 17
MeV. Data were obtained from our own neutron measurements on bare and partially-
shielded targets at 11 MeV. At the higher proton energy we conducted computer
simulations utilizing the program ALICE-91.
Neutron spectra at 11 MeV The Energy spectrum of neutrons from un-shielded RDS-
112 Targets (CTI, Inc., Koxville, TN, USA) were measured using an NE-213 Proton-
recoil Spectrometer*. Pulse-shape discrimination produced separate and distinct energy
spectra for gamma-ray events versus neutron-induced proton-recoil events. The raw
detector data were 'unfolded' using our own algorithm which incorporates a correction
*
Beam energy on target -- accounting for energy loss in the foils – was 10.4 MeV.
147
detector data were 'unfolded' using our own algorithm which incorporates a correction for
the intrinsic non-linearity of response of the scintillator at low recoil energies, a correction
for the energy-dependent fall-off in detector efficiency and, finally, channel-by-channel
discrete differentiation to extract spectral information from the (more-or-less) featureless
continuum of the raw data. The algorithm was validated by comparing an 'unfolded'
AmBe neutron spectrum against examples from the literature1.
Angular distribution of neutrons This was measured by fast-neutron activation of small

material samples placed at several angles around (and close to) bare RDS -112 targets:
Small pellets of P2O5 were activated via the reaction 31P(n,p)31Si to record neutrons above
2 MeV; small iron bolts were activated via the reaction 56Fe(n,p)56Mn to record neutrons
above 6 MeV. In the figures below, data for respective threshold reactions are normalized
– not according to relative intensity (which is apparent from the above figures) -- but
rather for ease of visualization of angular distribution.
148
Spectrum and Fluence for Ep = 17 MeV For bombarding energies greater than Ep = 12
MeV, a number of additional neutron-producing reaction channels open up. However, to
the best of our knowledge, direct experimental data on total neutron fluence and spectrum
for reaction energies higher than ~12 MeV is not available at this time.
149
Discussion: The reaction 18O(p,n)18 F is today -- by far -- the most widely used reaction
for PET isotope production. Below Ep = ~12 MeV the rate of neutron production is
virtually identical to the 18F saturation yield, i.e., the rate of production of ‘new’ 18F
atoms2.
Above Ep = 11 MeV a number of additional 18O(p,x) neutron-emitting reaction channels
become energetically possible3, but experimental cross-section data are not readily
available to help us determine rates of ‘excess’ neutron emission. We turn, therefore, to
computer codes, such as ALICE 91 to help us make these estimates.
150
The plots above show that below Ep = ~8 MeV ALICE 91 underestimates 18F saturation
yield and– by inference – also underestimates neutron production per uA on an 18O water
target.
Above Ep = 8 MeV, ALICE 91 begins to overestimate both 18F saturation yield and
neutron yield.
At 17 MeV ALICE 91 plausibly predicts approximately 1.5 times as many neutrons per
second relative to its own internal model calculation for 18F saturation yield, but this is
almost 4 times as many neutrons per second relative to experimentally-measured -- and
widely accepted -- 18F yield data4 scaled for a 100 % enriched water target.
Thus, between 11 MeV and 17 MeV ALICE 91 provides a credible – albeit highly
conservative – upper bound for neutron yield per uA from this reaction, offering a wide
safety margin relative to enriched water targets as well as higher-performance 18O2 gas
targets for use in shielding and activation calculations.
References
1. Carroll L.R., Pekrul E.:"Radiation Measurements in the Design and Evaluation of a

Self-Shielded Accelerator for PET". Proceedings of the Twentieth Midyear Topical
Meeting of the Health Physics Society. Reno, Nevada, Feb., 1987.
2. Bair, Miller, and Wieland; IJARI, Vol. 32, pp. 389-395, 1981.
3. Threshold and ‘Q-value’ data base at <http://t2.lanl.gov>
4. T. Ruth, et al; EXFOR charged-particle nuclear reaction data base at
www.nndc.bnl.gov
151
Potential of TlBr Semiconductor Detector for Nuclear Medicine
H.J. Sipilä1, M.A. Leskelä2 and V.I. Kozlov2
1
Metorex International, P.O.Box 85, FIN-02631 Espoo, FINLAND
2
Department of Chemistry, P.O.Box 55, FIN-00014 University of Helsinki, FINLAND
Aims
During the last years space organizations in Europe and USA have actively developed new
high density semiconductor detectors for the energy range 30-200 keV. Both position and
energy resolution are important. All those specifications are important in nuclear medical
applications as well.
TlBr is one of the material developed. Because TlBr has a high mean atomic number and a
density of 7,6 g/cm3, its radiation absorption properties are comparable with BGO. The
material is still in the development phase but very promising spectroscopic properties has
been demonstrated. Main effort is put to the purification and growing of detector grade
single crystals.
Methods
The purity of the crystal and a low density of defects are critical properties for the detector
grade crystals. Purification is done using several steps. The raw material is usually
purified by the multiple distillation and crystallization method. Liquid distillation is
studied. Trace element concentrations from different process steps are analyzed. The
crystals are chracterized by several methods including X-ray diffraction and IR reflection
spectroscopy. Electron collection properties are measured.
Results
Several detectors are produced from TlBr single crystals. X-ray spectra are measured
using 55Fe, 109Cd and 241Am sources. Energy resolutions 800 eV at 22 keV and 3 keV at 60
keV have been measured.
Conclusions
According to our studies TlBr is a potential detector material for nuclear medicine. It is a
wide band semiconductor which can operate at room temperature. Because TlBr has a
high atomic number and a high density its properties match well to the medical
applications. Good energy resolution can be used for the background suppression in
imaging applications. Next step is to refine the detector grade crystal growing process.
152
Ninth International Workshop on Targetry and
Target Chemistry, Turku, Finland, May 23-25, 2002
List of Participants
Name & Address E-mail Address
Jorma Aaltonen jorma.o.aaltonen@helsinki.fi

Laboratory of Radiochemistry
Department of Chemistry
University of Helsinki
P.O. Box 55
FIN-00014 UNIVERSITY OF HELSINKI
FINLAND
Kamel Abbas kamel.abbas@jrc.it

Joint Research Centre
European Comission
Cyclotron TP 500, Via Fermi
21020 ISPRA, VA
ITALY
Brian Abeysekera babeyseker@mds.nordion.com

MDS Nordion
4004 Wesbrook Mall
V6T 2A3 VANCOUVER
CANADA
Larry Adriano adriano@theragenics.com

Theragenics Corporation
5203 Bristol Industrial Way
30518 BUFORD, GEORGIA
USA
Margit Ahman-Kantola marahm@utu.fi

Turku PET Centre
Radiopharmaceutical Chem. Lab.
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
Bill Alvord bill.alvord@cti-pet.com

CTI Cyclotron Systems
810 Innovation Dr.
KNOXVILLE, TN 37932
USA
Artyom Arzumanov arzumanov@inp.kz

Institute of Nuclear Physics
1 Ibragimov Str.
480082 ALMATY
KAZAKHSTAN
Todd Barnhart tebarnhart@students.wisc.edu

University of Wisconsin, Madison
1500 N. Highland Ave
T-121 Waisman Center
53705 MADISON, WI
USA
153
Colin Beaslev colin.beasley@amersham.com
Amersham Plc
The Grove Centre (AL22)
White Lion Road, Amersham, Bucks
HP7 9LL AMERSHAM
UK
Volker Bechtold bechtold@hzy.fzk.de

Forschungszentrum Karlsruhe
P.O. Box 3640
76021 KARLSRUHE
GERMANY
Dirk Becker dirk.becker@argos.co.at

PET-Zentrum
Argos Zyklotron
St. Veiter Str. 47
9026 KLAGENFURT
AUSTRIA
Paolo Bedeschi comecer@comecer.com

Comecer S.R.L.
Via Emilia Ponente N. 390
48014 CASTELBOLOGNESE
ITALY
Jörgen Bergman jorgen.bergman@abo.fi

Turku PET Centre
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
Jan Olof Bergström janolof.bergstrom@med.ge.com

GE Medical Systems
Husbyborg
SE-752 29 UPPSALA
SWEDEN
Gerd-Jürgen Beyer gerd.beyer@cern.ch

Cyclotron Unit
University Hospital
24, Rue Micheli du Crest
1211 GENOVA
SWITZERLAND
Anna Bogni bogni@istitutotumori.mi.it

National Cancer Institute
Via Venezian, 1
20133 MILAN
ITALY
Albina Borissenko a.borisenko@inp.kz

Institute of Nuclear Physics
Ibragimov str. 1
480082 ALMATY
KAZAKHSTAN
Michael Brüssermann michael.brussermann@med.ge.com

GE Medical Systems (Nuclear Interface)
Dorotheenstrasse 26 A
48145 MÜNSTER
GERMANY
154
Ken Buckley buckley@triumf.ca
TRIUMF
4004 Westbrook Mall
V6T 2A3 VANCOUVER
CANADA
Stefano Buono stefano.buono@cern.ch

SL Division
CERN
CH 1211 GENEVA 23
SWITZERLAND
Edward Carroll daniel@kfshrc.edu.sa

King Faisal Specialist Hospital & ResearchCentre
MBC-03
P.O. Box 3354
11211 RIYADH
SAUDI ARABIA
Lewis Carroll cra@carroll-ramsey.com

Carroll & Ramsey Associates
950 Gilman St.
94710 BERKELEY, CA
USA
Rosangela Casati
20122 MILANO
ITALY
Raman Chirakal chiraklr@mcmaster.ca

Dept. of Nuclear Medicine
Hamilton Health Sciences
McMaster Site, 1200 Main St. West
L8N 3Z5 HAMILTON, ONTARIO
CANADA
Jaroslaw Choinski jch@nov.slcj.uw.edu.pl

Heavy Ion Laboratory
Warsaw University
Ul. Pasteura 5 A
02-093 WARSAW
POLAND
John Clark jcc24@wbic.com.ac.uk

Wolfson Brain Imaging Centre
University of Cambridge
P.O. Box 65
Addenbrooke's Hospital
CB2 2QQ CAMBRIDGE
UK
Josef Comor jcomor@vin.bg.ac.yu

Laboratory of Physics
Vinca Institute of Nuclear Sciences
P.O. Box 522
11001 BELGRADE
YUGOSLAVIA
Matthieu Conjat mconjat@free.fr

Universite de Nice
19 Bis,
Rue du Soleil
06100 NICE
FRANCE
155
Jan Courtyn jan.courtyn@rug.ac.be
Ghent University Hospital
Proeftuinstraat 86
9000 GENT
BELGIUM
Robert Dahl robertdgreat@mindspring.com

24 Pearce Place
11021 GREAT NECK, NY
USA
Philippe Damhaut pdamhaut@mds.nordin.be

MDS Nordion SA
Zoning Industriel
6220 FLEURUS
BELGIUM
Carmen Dence DenceC@mir.wustl.edu

School of Medicine
Washington University
510 S. Kingshighway
63110 ST LOUIS, MO
USA
David Dick dwdick@petrus.medphysics.wisc.edu

University of Wisconsin, Madison
1530 MSC, 1300 University Avenue
53706 MADISON, WI
USA
Henk Duiker info@vongahlen.nl

Von Gahlen Nederland BV
Postbus 25
6940 BA DIDAM
THE NETHERLANDS
Michael Epstein marshal@themail.com

Marshall-Isotopes Ltd
P.O.Box 21154
61211 TEL-AVIV
ISRAEL
Per-Olof Eriksson per-olof.eriksson@abo.fi

Accelerator laboratory
Turku PET Centre
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
Olli Eskola oesko@utu.fi

Turku PET Centre
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
Richard Ferrieri rferrieri@bnl.gov

Chemistry Department
Brookhaven National Laboratory
Bldg. 555
11973-5000 UPTON
USA
156
Ronald Finn finnr@mskcc.org
Memorial Sloan-Kettering Cancer Center
Cyclotron Facility
1275 York Avenue
10021 NEW YORK
USA
Toshimitsu Fukumura t_fukumu@nirs.go.jp

National Institute of Radiological Sciences
4-9-1 Anagawa, Inage-ku Chiba
263-8555 CHINA
Greg Gaehle gaehle@mir.wustl.edu

Research Institute
510 S. Kingshighway
63110 ST. LOUIS
USA
Nic Gillings nic@pet.rh.dk

Rigshospital
Copenhagen University Hospital
Afsnit 3982
Bleedamsvej 9
2100 COPENHAGEN
DENMARK
Hans Jürgen Gils hans.juergen.gils@ikfzk.de

Forschungszentrum Karsruhe
P.O. Box 3640
76021 KARLSRUHE
GERMANY
Alexandre Giusto alex.giusto@free.fr

AIMA
Porte de l'Anrénas C, 455 Promenade des
Anglais
06200 NICE
FRANCE
Ville Haaslahti ville.haaslahti@hidex.com

Hidex Oy
Mustionkatu 2
FIN-20750 TURKU
FINLAND
Michael Haka mike@nucmed.buffalo.edu

University of Buffalo
105 Parker Hall
3435 Main St.
14214 BUFFALO
USA
Ulf Hällsten ulf.hallsten@abo.fi

Department of Physics
Abo Akademi University
Porthansgatan 3
FIN-20500 TURKU
FINLAND
Robert W. Hamm rhamm@linacs.com

CEO
AccSys Technology, Inc.
1177 A Quorry Lane
94566 PLEASANTON, CA
USA
157
Richard Hammond richard.hammond@csc.mrc.ac.uk
Imaging Research Solutions Limited (RSL)
Hammersmith Hospital
Cyclotron Building
Du Cane Road
W12 0NN LONDON
UK
Chunlei Han chinlei.han@pet.utu.fi

Turku PET Centre
P.O. Box 52
FIN-20510 TURKU
FINLAND
Terry Hanson thanson@bioscan.com

Nuclear Medicine & PET
Bioscan Europe Ltd.
15 West Werberside
EDINBURGH EH4 1SZ
UK
Kerttuli Helariutta kerttuli.helariutta@helsinki.fi

P.O. Box 55
FIN-00014 HELSINKI
FINLAND
Semi Helin semi.helin@abo.fi

Turku PET Centre
Porthaninkatu 3
FIN-20520 TURKU
FINLAND
Sven-Johan Heselius sven-johan.heselius@abo.fi

Turku PET Centre
Accelerator Laboratory
Porthansgatan 3
FIN-20500 TURKU
FINLAND
Richard D. Hichwa richard-hichwa@uiowa.edu

PET Center
University of Iowa Hospitals & Clinics
200 Hawkins Drive, 0911 JPP
52242 IOWA CITY, IA
USA
Andrew Horti ahorti@intra.nida.nih.go

Division of Intramural
NIDA
P.O. Box 5180
5500 Nathan Shock Drive
21224 BALTIMORE, MD
USA
Eric Hostetler eric_hostetler@merck.com

Merck
WP44C-Z
P.O. Box 4
19486 WEST POINT, PA
USA
158
Tamara Hurtado magic@triumf.ca
PET Programme
Triumf
4004 Wesbrook Mall
V6T 2A3 VANCOUVER, BC
CANADA
Mehmet Husnu husnum@syncor.com

Syncor Advanced Isotopes
5601 N. Powerline Road
33309 FORT LAUDERDALE, FL
USA
Claudio Illan claudio.illan@duke.edu

Radiology
Duke University Medical Center
P.O. Box 3949-27710
Room 0402 Yellow South
27710 DURHAM
USA
Teruo Inoue iteruo@drl.co.jp

Daiichi Radioisotope Lab. Ltd
17-10, Kyobasahi
1-chome chuo-ku
104-0031 KYOBASHI
JAPAN
Joerg Jegge joerg.jegge@psi.ch

Paul Scherrer Institute
Center for Radiopharmaceutical Science
CH-5232 VILLIGEN PSI
SWITZERLAND
Holger Jensen holger@pet.rh.dk

PET & Cyclotron Unit
KF 3982
Blegdamsvej 9
DK-2100 COPENHAGEN
DENMARK
Mikael Jensen mikaelj@kvl.dk

Royal Veterinary and Agricultural University
IMF, Thorvaldsensvej 40
DK-1871 FREDERIKSBERG
DENMARK
Svend Borup Jensen svend@pet.auh.dk.

Århus PET-Center
Århus Kommunehospital
Nørrebrogade 44
DK-8000 ÅRHUS
DENMARK
Salma Jivan salma@triumf.ca

TRIUMF
University of British Columbia
4004 Wesbrook Mall
V6T 2A3 VANCOUVER
CANADA
159
Stefan Johansson sjohanss@abo.fi
Turku PET Centre
Porthansgatan 3
FIN-20500 TURKU
FINLAND
Richard Johnson dickj@attglobal.net

9271 Gormond. Rd
V7E IN6 RICHMOND
CANADA
Abu Kanu abu.kamu@stud.umist.ac.uk

UMIST
Dais, Umist
P.O. Box 88, Sackville Street
M60 1QD MANCHESTER
UK
Lars Kervefelt lars.kervefelt@med.ge.com

GE Medical Systems Sverige AB
P.O. Box 314
SE-171 75 STOCKHOLM
SWEDEN
Maxim Kiselev mkiselev@ibamed.com

IBA Eastern Isotopes
13 Crescent Ct
20164 STERLING, VA
USA
Laurent Köhler kohler@iba.be

Ion Beam Applications SA (IBA)
Chemin Du Cyclotron No:3
B-1348 LOUVAIN_LA_NEUVE
BELGIUM
Nikolai Konyakhin
Cyclotron Co., Ltd.
1 Bondarenko Sq.
249033 OBNINSK
RUSSIA
Michael S. Kovacs mkovacs@ebcotech.com

EBCO Technologies
7851 Alderbridge Way
V6X 2A4 RICHMOND
CANADA
Jacek Koziorowski jacek.koziorowski@studsvik.se

Studsvik Nuclear AB
SE-611 82 NYKÖPING
SWEDEN
Thomas Krauss
CTI
Am Schiewenhuegel 8
D-48366 LAER
GERMANY
160
Jaroslav Kroulik
Nuclear Physics Insitute, CAS
Rez U Prahy
25068 PRAGUE
CZECH REPUBLIC
Jaana Kumpulainen jaana.kumpulainen@mapmedical.fi

MAP Medical Technologies Oy
Elementtitie 27
FIN-41160 TIKKAKOSKI
FINLAND
Richard Laforest laforest@mir.wustl.edu

Medical School
510 S. Kingshighway
Campus Box 8225
63110 ST. LOUIS, MO
USA
Desideriu Laky dlaky@hadassah.org.il

Nuclear Medicine Department
Hadassah Medical Organization
JERUSALEM
ISRAEL
Suzanne Lapi slapi@sfu.ca

SFU
102-3760 Albert St.
V5C 548 BORNABY
CANADA
Nina Laurén nina.lauren@utu.fi

Turku PET Centre
Radiopharmaceutical Chem. Lab
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
Efraim Lavie efi@soreq.gov.il

Soreq, NRC
81800 YAVNE
ISRAEL
Didier Le Bars lebars@univ-lyon.fr

CERMEP
59 Bd Pinel
69003 LYON
FRANCE
Tom Lewellen
Division of Nuclear Medicine, Medical
Center
University of Washington
NW040, BOX 356004
Pacific Way NE
98195 SEATTLE, WA
USA
Dewi Lewis dewi.lewis@amersham.co

Amersham
The Grove Centre
White Lion Road
HPI6 9DG AMERSHAM, BUCKS
UK
161
Sten Lindholm sten.lindholm@abo.fi
Turku PET Centre
Porthansgatan 3
FIN-20500 TURKU
FINLAND
Jeanne M. Link jeanne@u.washington.edu

UW Medical Center
University of Washington
1959 NE Pacific St.
Rm. NW041, Box 356004
98195-6004 SEATTLE, WA
USA
Donald Lyster dmlyster@interchange.ubc.ca

Vancouver Hospital
VANCOUVER
CANADA
Bruce Mackay bruce.mackay@csc.mtc.ac.uk

Imaging Research Solutions Limited (RSL)
Hammersmith Hospital
Cyclotron Building
Ducane Road
W12 0NN LONDON
UK
Jacob Madsen j.madsen@rh.dk

Blegdamsvej 9
Afd. KF 4011
DK-2100 COPENHAGEN
DENMARK
Pierre Mandrillon pierre.mandrillon@aima.fr

Director
AIMA
Porte de l'Arenas C
455 Promenade des Anglais
06200 NICE
FRANCE
Mario Matarrese marion@mednuc.hsr.it

CNR-INB; Institute H San Raffaele
Via Olgettina, 60
I-20132 MILANO
ITALY
Deborah McCarthy mccarthy_d@mir.wustl.edu

Washington University Medical School
P.O. Box 8225
510 S. Kingshighway Blvd.
63110 ST. LOUIS, MO
USA
Neil McPhail nmcphail@mds.nordion.com

MDS Nordion
4004 Wesbrook Mall
V6T 2A3 VANCOUVER
CANADA
162
Steve McQuarrie stevemcq@cancerboard.ab.ca
Faculty of Pharmacy
University of Alberta
DP Centre
T6G 2N8 EDMONTON, AB
CANADA
Jitze Medema j.medema@pet.azg.nl

Groningen University Hospital
Hanzeplein 1
P.O. Box 30.001
NL-9700 RB GRONINGEN
THE NETHERLANDS
Luca Menichetti
Institute of Clinical Physiology
CNR - National Research Council
Via Moruzzi 1
I-56100 PISA
ITALY
John Mercer john.mercer@ualberta.ca

University of Alberta
3118 Dentistry Pharmacy Building
T6G-2N8 EDMONTON
CANADA
Salman Miliebari salman@kfshre.edu.sa

Cyclotron Section
King Falsal Specialist Hosp.
P.O. Box 3354
11211 RIYADH
SAUDI ARABIA
Eyal Mishani mishani@urda.huji.ac.il

Department of Nuclear Medicine
Hadassah Hospital
9100 JERUSALEM
ISRAEL
Bruce H. Mock bmock@iupui.edu

Assoc. Prof.
Indiana University School of Medicine
4020 Rommel Drive
46228-3242 INDIANAPOLIS, IN
USA
Benjamin Mommens mommens@iba.be

Ion Beam Applications Sa (IBA)
BELGIUM
Roel Mooij
BV Cyclotron
Vrije Universiteit
De Boelelaan 1085
NL-1081 HV AMSTERDAM
THE NETHERLANDS
Jean-Luc Morelle jeanluc.morelle@med.ge.com

GE Medical Systems (Coincidence Tech.)
Parc Scientifique du Sart Tilman
Avenue Pré-Aily (c/o Socran)
B-4031 LIEGE
BELGIUM
163
Michael Mueller muellermi@mir.wustl.edu
510 S. Kingshighway
63110 ST. LOUIS
USA
Michael Nader michael.nader@argos.co.at

ARGOS Zyklotron GmbH
Seilerstätte 4
4010 LINZ
AUSTRIA
Kjell Någren kjell.nagren@utu.fi

Turku PET Centre
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
Paliek Napiorkowski pjn@slcj.uw.edu.pl

Heavy Ion Laboratory
Warsaw University
Ul. Pasteura 5 A
02-093 WARSAW
POLAND
Jerry Nickles niddes@petrus.medphysics.wisc.edu

Medical Physics Dept.
University of Wisconsin
1530 Medical Science Center
1300 University Ave
53706 MADISON, WI
USA
Outi Nihtilä outi.nihtila@mapmedical.fi

MAP Medical Techonogies Oy
P.O. Box 9
FIN-00251 HELSINKI
FINLAND
Petri Numminen petri.numminen@pet.tyks.fi

Turku PET Centre
P.O. Box 52
FIN- 20521 TURKU
FINLAND
Mikako Ogawa mogawa@ri.ncvc.go.jp

National Cardiovascular Center
5-7-1 Fujishirodai, Suita
565-8565 OSAKA
JAPAN
Tomas Ohlsson Tomas.Ohlsson@skane.se

Radiation Physics
University Hospital
SE-22185 LUND
SWEDEN
Matti Olsen
GE Medical Systems
P.O. Box 185
FIN-00701 HELSINKI
FINLAND
164
James O'Neil jponeil@lbl.gov
Biomedical Isotope Facility
Lawrence Berkeley National Lab
1 Cyclotron Road
Mail stop 55-121
94720 BERKELEY
USA
Barak Ostraich ostr@menix.bgu.ac.il

Rotem Industries
P.O. Box 9046
84190 BEERSHEVA
ISRAEL
Claudio Pascali pascali@istitutotumori.mi.it

National Cancer Institute
V. Venezian, 1
I-20133 MILAN
ITALY
Harri Pättikangas
GE Medical Systems
P.O. Box 185
FIN-00701 HELSINKI
FINLAND
Martin Plánicka planicka@ujf.cas.cz

Nuclear Physics Insitute, CAS
Rez U Prahy
25068 PRAGUE
CZECH REPUBLIC
Ilan Pri-Bar
Applied Chemistry
Rotem Industries
P.O. Box 9046
84190 BEERSHEVA
ISRAEL
Syed M. Qaim s.m.qaim@fz-juelich.de

Institut für Nuklearchemie
Forschungszentrum Juelich
D-52425 JUELICH
GERMANY
Anatoli Razbash razbash_isotop@obninsk.com

Cyclotron Co., Ltd.
1 BondarenKo Sq.
249033 OBNINSK
RUSSIA
Andrew Roberts arobert5@facstaff.wisc.edu

1500 N. Highland Ave
T-135 Waisman Center
53705 MADISON, WI
USA
Sharon Roded sharon@marshall-isotopes.com

Marshall-Isotopes Ltd
P.O.Box 21154
61211 TEL-AVIV
ISRAEL
165
Douglas Rowland rowland@mir.wustl.edu
Institute of Radiology
Washington University in St. Louis
510 S. Kingshighway
63110 ST. LOUIS, MO
USA
Armin Runz a.runz@dkfz.de

German Cancer Research Center
Im Neuenheimer Feld 280
D-69120 HEIDELBERG
GERMANY
Thomas Ruth truth@triumf.ca

UBC
TRIUMF PET Program
4004 Wesbrook Mall
VGT 2A3 VANCOUVER
CANADA
Aino Ruutu aino.ruutu@helsinki.fi

P.O. Box 55
FIN-00014 UNIVERSITY OF HELSINKI
FINLAND
Johan Sambre johan.sambre@rug.ac.be

Ghent University Hospital
Proeftuinstr 86
B-9000 GENT
BELGIUM
Anders Sandell Anders.Sandell@radfys.lu.se

Radsatron Physics
University Hospital
SE-22185 LUND
SWEDEN
Nina Savisto nijosa@utu.fi

Turku PET Centre
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
David Schlyer schlyer@bnl.gov

Brookhaven National Laboratory
Bldg 901
57 Cornell Ave
UPTON, NY 11973-5000
USA
Frederic Schmitz verbruggen@iba.be

ION Beam Applications SA (IBA)
Chemin du Cyclotron no. 3
B-1348 LOUVAIN LE NEUVE
BELGIUM
Rolf Schwarzbach rolf.schwarzbach@psi.ch

Paul Scherrer Institut
CH-5232 VILLIGEN PSI
SWITZERLAND
166
Yosef Shohet shohet@soreq.gov.il
Soreq Nrl
81800 YAVNE
ISRAEL
Paul D. Shrever pshrever@umich.edu

University of Michigan
B1 6505 University Hospital
Box 0038
48109-0078 ANN ARBOR,MI
USA
Jukka Siikki jukka.siikki@mapmedical.fi

MAP Medical Technologies Oy
Elementtitie 27
FIN-41160 TIKKAKOSKI
FINLAND
Norman Simpson nrs17@columbia.edu

College of Physicians & Surgeons
Columbia University
177 Ft Washington Ave MHB
BI-004
10032 NEW YORK
USA
Hannu Sipilä hannu.sipila@tyks.fi

Turku PET Centre
P.O. Box 52
FIN-20521 TURKU
FINLAND
Heikki Sipilä heikki.sipila@metorex.fi

Metorex International
P.O. Box 85
Nihtisillankuja 5
FIN-02631 ESPOO
FINLAND
Leonid Solin solin@mail.convey.ru

V.G. Khlopin Radium Institute
Veteranov av. 104, 57
198260 ST. PETERSBURG
RUSSIA
Olof Solin olof.solin@abo.fi

Turku PET Centre
Porthansgatan 3
FIN-20500 TURKU
FINLAND
Dmitri Soloviev dmitri.soloviev@hcuge.ch

Cyclotron Unit
University Hospital
CH-1211 GENEVA
SWITZERLAND
Raymond Steichen steichen.raymond@mayo.edu

Mayo Clinic-Rochester
200 First Street Southwest
55906 ROCHESTER, MN
USA
167
Joerg Steinbach j.steinbach@iif-leipzig.de
Institut fuer Interdisziplinaere Isotopen
Forschung
D-04318 LEIPZIG
GERMANY
Erkki Stenvall
Turku PET Centre
Porthansgatan 3
FIN-20500 TURKU
FINLAND
Nigel Stevenson stevenson@theragenics.com

Theragenics Corporation
5203 Bristol Industrial Way
30518 BUFORD, GA
USA
Sharon Stone-Elander sharon.stone@ks.se

Karolinska Institute, Karolinska Pharmacy
Karolinska Hospital
SE-171 76 STOCKHOLM
SWEDEN
Kazutoshi Suzuki kazutosi@nirs.go.jp

4-9-1 Anagawa, Inage-ku
263-8555 CHINA
Vincent Tadino vtadino@bioscan.com

Bioscan
4590 Mc Arthur Blvd N.W.
20007 WASHINGTON DC.
USA
Claire Tamburella claire.tamburella@hcuge.ch

Cyclotron Unit
University Hospital
1211 GENOVA
SWITZERLAND
Lucie Tang tangle@mir.wustl.edu

School of Medicine
510 S. Kingshighway
Campus Box 8225
63110 ST. LOUIS, MO
USA
Anthony Tascione tascionea@syncor.com

PET Regional Mgr
Syncor International Corp.
5601 N. Powerline Rd
Suite 108
33309 FT. LAUDERDALE
USA
Mika Teräs mika.teras@tyks.fi

Turku PET Centre
P.O. Box 52
FIN-20521 TURKU
FINLAND
168
Tim Tewson timothy-tewson@uiowa.edu
University of Iowa
PET Imaging Center 0917 JPP
200 Hawkins Drive
52242-1077 IOWA CITY, IOWA
USA
Jan-Olov Thorell jan-olov.thorell@ks.se

Karolinska Pharmacy
Karolinska Hospital
SE-117 76 STOCKHOLM
SWEDEN
Sergio Todde sergiot@mednuc.hsr.it

University of Milano-Bicocca
Via Olgettina 60
I-20132 MILANO
ITALY
Phong Truong phong.truong@ks.se

Psykiatri Centrum
Karolinska Sjukhuset
Fouu-sektionen
Hus R5: U1
SE-171 76 STOCKHOLM
SWEDEN
Marian Turiceanu turiceanu@yahoo.com

Turku PET Centre
P.O. Box 52
FIN-20521 TURKU
FINLAND
Peter van Leuffen p.van.leuffen@cyclotron.nl

BV Cyclotron
Vrije Universiteit
De Boelelaan 1085
NL-1081 HV AMSTERDAM
THE NETHERLANDS
Joan Quarles van Ufford joang@veenstra-instr.nl

Veenstra Instruments BV
P.O. Box 115
NL-8500 AC JOURE
THE NETHERLANDS
Marleen Vandecapelle marleen.vandecapelle@rug.ac.be

Ghent University
Harelbekestraat 72
B-9000 GHENT
BELGIUM
Chris Vascoe cvascoe@biomed.org

Biomedical Research Foundation
P.O. Box 38050
71133-8050 SHREVEPORT, LA
USA
Simo Vauhkala simo.vauhkala@utu.fi

Turku PET Centre
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
169
Rudy Verbruggen verbruggen@iba.be
Ion Beam Applications SA (IBA)
BELGIUM
Christiaan Vermeulen etienne@tlabs.ac.za

IThemba LABS
P.O. Box 722
7129 SOMERSET WEST
SOUTH AFRICA
Tapio Viljanen taavi@utu.fi

Turku PET Centre
Porthaninkatu 3
FIN-20500 TURKU
FINLAND
G. Leonard Watkins len-watkins@uiowa.edu

PET Center
University of Iowa Hospitals & Clinics
200 Hawkins Drive, 0911 JPP
52242 IOWA CITY, IA
USA
Klaus Weber K.Weber@DKFZ.de

Abl. E0301 Deutsches
Krebsforschungszentrum
D-69120 HEIDELBERG
GERMANY
Juergen Wendel wendel@hzy-fzk.de

Forschungszentrum Karlsruhe
Hardtstrasse. 3
D-76689 KARLSDORF-NEUTHARD
GERMANY
Tom Wickström twickstr@abo.fi

Turku PET Centre
Porthansgatan 3
FIN-20500 TURKU
FINLAND
Bruce Wieland bruce.wieland@duke.edu

Duke University Medical Center
Room 0402 Yellow Zone, Duke Hospital
South
Box 3808
27710 DURHAM
USA
John Wilson john.wilson@ualberta.ca

Cross Cancer Institute
11102-76 Ave
T6G OJ8 EDMONTON
CANADA
Gerd Wolber g.wolber@dktz.de

Dept. of Radiology
German Cancer Res.Ctr.
D-69120 HEIDELBERG
GERMANY
170
Bruce Wright bowryght@aol.com
Bruce Technologies, Inc.
3420 11th Avenue A
61265-2410 MOLINE IL
USA
Frank Wuest f.wuest@fz-rossendorf.de

Forschungszentrum Rossendorf
PF 510119
D-01328 DRESDEN
GERMANY
Nikki Young
1530 Medical Science Center
1300 University Ave
53706 MADISON, WI
USA
Alessia Zanelli comecer@comecer.com

Comecer S.R.L.
Via Emilia Ponente N. 390
I-48014 CASTELBOLOGNESE
ITALY
Ming Rong Zhang zhang@nirs.go.jp

SHI Accelerator Service Co. Ltd.
4-9-1 Anagawa, Inage-ku
263-8555 CHINA
S'yts'e Zijlstra szijlstra@hdzmrw

Herz- und diabeteszentrum Bad
Oeynhausen
Georgstrasse 11
D-32545 BAD OEYNHAUSEN
GERMANY
171
Author Index
Abbas, K. 84 Gillies, J.M. 133

Adam, M.J. 2 Goldman, S. 12
Ahn, S.H. 31 Gorodisskaya, N. 92
Al-Qahtani, M. 108 Groppi, F. 84
Alvord, C.W. 24 Gunawan, J. 140
Armbruster, J. 8 Hagen, B. 103
Arzumanov, A. 34, 92 Hamm, R.W. 3
Bailey, J. 133 Heaton, R.C. 88
Barnhart, T.E. 38, 67, 98 Hendrikse, N.H. 137
Bastian, Th. 60 Heselius, S.-J. 104
Becker, D.W. 30, 103 Hohn, A. 60
Berdinova, N. 92 Holzwarth, U. 84
Beyer, G.-J. 17, 46 Hong, S.W. 31
Bida, G.T. 24 Hughey, B.J. 62
Bida, J. 19 Hur, M.G. 31
Bigott, H.M. 111 Hurtado, E.T. 2
Birattari, C. 84 Illan, C. 19, 48, 125
Blessing, G. 60, 86 Ilmatov, S. 34
Bogdanov, P.V. 104 Jackson, M.A. 142
Bognar, C. 108 Jakovlev, V.A. 71
Bonardi, M. 84 Jegge, J. 89
Borissenko, A. 34, 92 Jensen, M. 58, 67, 115
Brüssermann, M. 9, 46 Jivan, S. 2, 30, 32, 98
Buckley, K.R. 2, 30, 98, 103, 107, 128 Johansson, S. 104
Burchert, W. 140 Jongen, Y. 12
Carroll, L.R. 142, 147 Kanu, A. 133
Casale, G.A. 129, 135 Kastleiner, S. 86
Cerny, J. 118 Keizer, H. 113
Chumikov, G. 92 Ketzscher, U. 46
Chun, K.S. 31 Kim, S.W. 31
Coenen, H.H. 60, 86 Kiselev, M. 8
Coleman, R. 125 Kislov, V.A. 104
Čomor, J.J. 46 Klinkowstein, R.E. 62
Converse, A.K. 38 Knyazev, A. 92
Corbin, L.A. 11 Koptev, V. 34, 92
Courtyn, J. 121, 131 Košutić, D. 46
D’Auria, J.M. 32 Kovalchuk, O.A. 104
Dabbs, K.A. 38 Kozlov, V.I. 152
Dahl, R. 30 Kruijer, P.S. 12
Dailey, M. 125 Kudelin, B.K. 71
Dence, C.S. 108 Laforest, R. 62, 111
Dick, D.W. 115 Lago, J.L. 129
Dixon, M. 133 Lapi, S. 32
Doster, M. 19 Lechner, K. 139
Duiker, H. 137 Lee, H. 108
Eisenhut, M. 18, 64 Lenz, J.W. 24, 103
Ekelmans, D. 12 Leskelä, M.A. 152
El-Azoney, K.M. 60 Lewis, M.R. 111
Elsinga, P.H. 9, 113, 137 Lyssukhin, S. 34
Gaehle, G. 110, 139 Mainardi, H.S. 84
Ghyoot, M. 17 Margenau, W. 108
172
Marx, H. 18 Stroosnijder, M.F. 84
Maslat, G.J. 129, 135 Sychikov, G. 92
McCarthy, D.W. 74, 111 Tamburella, C. 17
McCarthy, T.J. 62 Tang, L. 74
Medema, J. 9, 113, 137 Teräs, M. 123
Menapace, E. 84 Thomas, P. 133
Mock, B.H. 11 Tilkema, S.P. 113
Moerlein, S.M. 108 Vaalburg, W. 113
Monclus, M. 12 Vamecq, F. 12
Mooij, R. 12 Van der Jagt, P.J. 12
Moon, B.S. 31 van der Knaap, Y. 137
Mueller, M.J. 110, 139 van der Walt, T.N. 86
Murphy, S. 125 van Essen, J. 9, 113
Nickles, R.J. 38, 58, 67, 115 Van Langevelde, F. 12
Nortier, F.M. 86, 88 Van Leuffen, P.J. 12
Nye, J.A. 115 Van Naemen, J. 12
O’Neil, J.P. 118 Vasilchenko, I.N. 104
Paans, A.M.J. 9 Weber, K. 18
Perkins, T.A. 111 Welch, M.J. 62, 74, 108, 110, 111, 139
Peterson, E.J. 88 Verbruggen, R. 12
Phillips, D.R. 88 Wieland, B.W. 19, 21, 24, 48, 125
Piccioni, P. 2, 103 Wilson, G. 137
Pilcher, A. 78 Wilson, J. 2
Popov, Yu. 92 Wolber, G. 58, 64
Powell, J. 118 Voller, T.F. 111
Qaim, S.M. 60, 86 Wright, B. 21
Rajčević, M. 46 Yang, S.D. 31
Ramsey, C.A. 118 Yin Zu, Y. 133
Reichert, D.E. 74, 111 Zheltov, D. 92
Roberts, A. 19 Zijlstra, S. 140
Roberts, A.D. 38, 67, 98 Zimmermann, K. 89
Robinson, Jr., G.D. 3 Zippi, E.M. 78
Rowland, C. 19 Zweit, J. 133
Rowland, D.J. 62, 111 Zyuzin, A. 32
Runkle, R. 19
Runz, A. 58, 64
Ruth, T.J. 2, 30, 32, 98, 103, 107
Sambre, J. 121, 131
Schibli, R. 89
Schmitz, F. 12
Scholten, B. 60
Schubiger, P.A. 89
Schueller, M.J. 67, 115
Schwarz, S.W. 110, 139
Schwarzbach, R. 89
Semmler, W. 64
Severgin, Yu.P. 104
Shefer, R.E. 62
Simonelli, F. 84
Sipilä, H.J. 152
Sipilä, H.T. 123
Solin, L.M. 71
Soloviev, D. 17
Spellerberg, S. 60
Stogov, Yu.I. 104
173

WTTC9 Proceedings 2002 Comor COSTIS Solid Target p59

Uploaded by

Copyright:

Available Formats

WTTC9 Proceedings 2002 Comor COSTIS Solid Target p59

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

WTTC9 Proceedings 2002 Comor COSTIS Solid Target p59

Uploaded by

Copyright:

Available Formats

Proceedings of the Ninth International

Workshop on Targetry and Target Chemistry

Edited by Sven-Johan Heselius, Jan-Olof Lill and Johan Rajander

Turku PET Centre, Accelerator Laboratory, Åbo Akademi University,

Report ÅA Acc-1 (2004)

Edited by Sven-Johan Heselius, Jan-Olof Lill and Johan Rajander

Workshop organized by the Turku PET Centre

Organizing Committee International Advisory Committee

St. Louis Meetings 1999

and the following companies in alphabetical order:

Targetry Workshop Venue: Educarium Building, University of Turku, Assistentinkatu 5

Thursday, May 23rd

Training Course in cyclotron targetry

08.30-09.15 Technology of cyclotrons for radionuclide production, Kenneth Buckley, TRIUMF.

10.00-10.20 Coffee Break

13.20-14.00 Oral presentations by cyclotron suppliers (CTI, EBCO, GE and IBA).

Group CTI EBCO GE IBA

15.40-16.00 Coffee Break

18.00-20.00 Welcome Reception at Educarium Building.

20.00-21.00 Evening Lecture: "Commissioning and Decommissioning of Cyclotrons"

08.30-08.45 Opening of the Workshop

10.15-10.45 Coffee Break

10.45-12.15 High-power enriched gas targets

13.30-15.00 Solid targets at low-energy irradiations. Transport of targets.

15.00-15.30 Coffee Break

15.30-17.00 High-energy irradiations of solid and gas targets

08.30-10.00 Poster session and Coffee

10.00-11.30 Oral presentations of selected poster

12.45-14.45 Visit to facilities (2 groups, 15 visitors/group):

14.45-15.15 Coffee Break

15.15-16.30 Reports from the laboratories

16.30-17.00 Summary of the Workshop and Closing Remarks

17.00-17.10 Closing of the Workshop

18.00-20.00 WTTC9 Farewell Party /Turku PET Symposium, Get-together Party,

Targetry Workshop Programme III

Water Targets and High Beam Currents. Specific Radioactivity of F-18 1

High Power Enriched Gas Targets 23

High-energy Irradiations of Solid and Gas Targets 83

Radiochemistry and Automation 127

Radiation Sources and Radiodetectors 141

List of Participants 153

Author Index 172

To assess the performance of a water target constructed of niobium metal.

This patented target*, developed by Brookhaven National Laboratory in a collaboration

Boiling point at 300 p s ig

Fig. 2. Thermal analysis results for gridded target.

B eam diameter (cm)

Fig. 3. Calculated yield for gridded target.

Experimental F-18 Yields vs. Target Current

Fig. 4. Experimental results for gridded water target.

C -11 G as T arg et P relim ina ry D ata

Initial target pressure (p sig)

Fig. 5. Experimental results for gridded gas target.

Approximately 10 ml of O-18 water was loaded in an apparatus containing a 5 ml storage

Department of Radiology, Indiana University School of Medicine, Indianapolis, Indiana

CONCLUSIONS: With minor plumbing changes, and the mere implementation of

F. Schmitz1, M. Monclus1, J. Van Naemen1, D. Ekelmans1, S. Goldman1, R. Verbruggen2,

Irradiation of 18O-enriched water is the mostly used method to produce [18F]Fluoride on a

As could be expected there is no significant difference in experimental thick target yields