K. Dahl et al.
from Zn granulate surface did not have any positive effect. expect this radiosynthetic strategy to be widely adapted in PET
Dispute these efforts, we were not able to regenerate usable radiopharmaceutical research and development.
levels of 11CO. On the other hand, the performance of Mo-filled
column after replacement was reproducible.
Experimental
General experimental information
Prototype performance
Unless otherwise stated, all reagents and solvents were obtained from
A series of 11CO2 productions were conducted using the same
beam current and time, 10 μA for 1 min, to estimate the recovery
Sigma-Aldrich (Sweden) and used without further purification. No-
carrier-added [11C]carbon dioxide production was performed using a
GEMS PETtrace cyclotron (GE, Uppsala, Sweden). The 14N(p, α)11
C
from each individual step. The average radioactivity produced
from three consecutive 11CO2 productions was used as a starting reaction was employed in a pressurized gas target containing nitrogen
(nitrogen 6.0) and 0.5% oxygen (oxygen 4.8) by bombardment with
10-μA proton beam (16.5 MeV) for 1 min. High-pressure HPLC was
performed using a gradient pump (L-6200, Hitachi, Tokyo, Japan) and a
variable wavelength UV-detector (L-4000, Hitachi, Tokyo, Japan) in a series
with a Bioscan β+-flow detector. Analytical HPLC was performed using a
reverse phase column (Agilent Eclipse XDB-C18, 5 μm, 4.6 × 150 mm) eluted
with a gradient between acetonitrile (A) and 10-mM H3PO4 or 100-mM
HCO2NH4 (B). The gradient was linear between 10% and 90% (B) over 5 min
at a flow rate of 3 ml/min. Identification of all radioactive products was
confirmed by co-elution with the corresponding non-radioactive compound.
value (Table 2, Entry 1). The radioactivity received after each
individual event was measured separately and repeated three
times. Depending if 11CO2 or 11CO was to be measured, an
ascarite trap or leak-tight gas bag was used to confine the
radioactivity and was further measured in a calibrated β-
irradiation chamber. The different geometry factors of measuring
activity of a small trap or a larger bag were not corrected. The
results are summarized in Table 2. The molecular sieve trap
was identified as the step with the highest loss of radioactivity,
approximately 30%. The identity or chemical form of the
radioactivity irreversibly trapped on the molecular sieve trap
was not established. The decay-corrected overall yield of 11CO
for the reduction–carbonylation sequence, when using Zn as
the reducing agent, was 53% over five events.
General procedure for 11C-carbonylation at high pressure
A schematic of the 11CO-synthesizer prototype (GEMS PET Systems AB
(GE Healthcare, Uppsala, Sweden)) is shown in Figure 1. At the end of
bombardment, the target content was delivered to the
11CO-synthesizer prototype, where the 11CO2 was trapped on
a
Radiochemistry
molecular sieve column (0.6 g packed in a 1/4″ SS tube, mesh 80-100,
Grace) at room temperature. Two 3-port, two-way valves (V1 and V2,
Parker, P/N 009-0269-900) and 1/8″ SS tubing were used to direct
helium flow or target gas flow onto the molecular sieve column. The
accumulated 11CO2 was released into a controlled stream (10 ml/min,
In a typical experiment, the coupling reagents (aryl halide or aryl
azide, Pd-source or Rh-source, and ligand and nucleophile
dissolved in anhydrous THF) were prepared in a standard glass
(4 ml) vial and loaded into the storage loop of the apparatus mass flow controller, Bronkhorst, Ruurlo, Netherlands) of helium (helium
6.0) while heating the molecular sieve column to 360 °C. The purified
11CO2 was further concentrated on a silica gel (10 mg, 60 Å, 60–
100 mesh) trap immersed in liquid nitrogen, see article by Eriksson et
al. for silica gel trap preparation details.16 The silica gel traps, S1 and
S2, were attached on a two-position vertically movable lift to enable
rapid cooling (liquid N2, lower position) and heating (halogen lamp,
upper position). 11CO2 was reduced online to 11CO using a pre-heated
(Carbolite oven, 400 °C for Zn and 850 °C for Mo) quartz glass column
(6 × 4 × 180 mm: outer diameter × inner diameter × length) charged
with either Zn granulate (2 g, 15–50 mesh, Merck, NY, USA) or Mo
approximately 5 min prior to the start of synthesis. The reaction
mixture was transferred and pressurized (350 Bar) by anhydrous
THF to the micro-autoclave, which was pre-charged with 11CO
in helium gas. The micro-autoclave was then heated to
appropriate temperature for 5 min before releasing the crude
product into an evacuated 4-ml vial. The radioactivity was
measured before and after the vial was purged with nitrogen.
A number of carbonyl containing functional groups were
successfully synthesized using this novel prototype. The nine
radiolabeled target compounds are presented in Table 3. All powder (1.5 g, <150 μm, 99.99% trace metals basis, Sigma-Aldrich,
Sweden). Unreacted 11CO2 was subsequently removed by a sodium
hydroxide-coated silica (0.2 g, Ascarite II, 20–30 mesh) trap (30 mm
1/8″ SS tube), and the 11CO was concentrated on a silica gel (10 mg,
60 Å, 60–100 mesh) trap immersed in liquid nitrogen. After completed
entrapment, the trap was heated to release the 11CO into a closed
and empty 250-μl SS micro-autoclave reactor. It is important to note,
while heating the silica gel trap (S2) to room temperature, the He flow
was turned off, this to enable a complete 11CO transfer to the micro-
autoclave reactor. The coupling reagents (aryl halide/azide, Pd or Rh
source, and ligand and nucleophile dissolved in anhydrous THF) were
injected to the apparatus through a storage loop (250 μl, SS tube)
attached to a six-port, two-way valve (V7, Valco, P/N C2-2006D). The
coupling reagents were further transferred to the micro-autoclave
reactor using a pressure of 350 Bar generated by an HPLC pump
(2 ml/min, Smartline Pump 100, Knauer, Berlin, Germany). The micro-
autoclave was heated at the desired temperature for 5 min using a
pre-heated oil bath, after which the content of the micro-autoclave
was emptied by releasing the high pressure through V5 (five-port,
four-way valve, Valco, P/N C5-2004D) into a sealed 4-ml vial. The
micro-autoclave reactor was attached using two 1/32″ SS with a
0.005″ i.d, while all other tubing used in the 11CO-syntheziser were 1/
reactions showed high TE of 11CO (>80%) and were produced
in an RCY range of 37–88% (Table 3). For example, the well-
characterized [11C]N-benzylbenzamide ([11C]1) was produced in
an RCY of 80%, which is in accordance with earlier published
data using the high-pressure micro-autoclave technology.19 In
a
typical reaction, from approximately 1900 MBq 11CO2,
600–800 MBq of radioactivity was transferred to the collection
vial. The overall radiosynthesis time was about 15 min measured
from the end of bombardment.
Conclusion
A fully automated high-pressure synthesizer using [11C]carbon
monoxide as the labeling precursor has been evaluated. Nine
different functional groups have been successfully radiolabeled
in good to excellent RCYs. This method is technically
straightforward and represents an immediate path to
11C-labeled products. This technology has the potential to allow
for 11C-carbonylation reactions to be performed in a routinely
fashion with similar simplicity as 11C-methylations, and we 16″ SS with a 0.010″ i.d. Four 4-port, two-way valves (V3, V4, V6, and
J. Label Compd. Radiopharm 2015, 58 220–225
Copyright © 2015 John Wiley & Sons, Ltd.