Cationic imidazolium polymer monoliths for efficient solvent exchange, activation and fluorination on a continuous flow system

Article abstract of DOI:10.1039/c4ra04064c

Polystyrene-imidazolium (PS-Im⁺Cl^-) monolith was synthesized within a flow-through microfluidic chip and Teflon tubing for activating [¹⁸F]fluoride ions. The [¹⁸F]fluoride ions were trapped on the PS-Im⁺

Full text of DOI:10.1039/c4ra04064c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Cationic imidazolium polymer monoliths for
efficient solvent exchange, activation and
fluorination on a continuous flow system†
Cite this: RSC Adv., 2014, 4, 25348
ab
ab
ab
Rehana Ismail, Jonathan Irribaren, Muhammad Rashed Javed,
ab
ab
ab
Ariella Machness, R. Michael van Dam and Pei Yuin Keng*
+
ꢀ
Polystyrene-imidazolium (PS-Im Cl ) monolith was synthesized within a flow-through microfluidic chip
18
18
+
and Teflon tubing for activating [ F]fluoride ions. The [ F]fluoride ions were trapped on the PS-Im
monolith and were subsequently released with various phase-transfer catalysts (PTC) and carbonate or
18
bicarbonate bases in microliter volumes of organic solvents containing 0.5% of water. The activated [ F]
+
fluoride complex released from the PS-Im monolith was used to fluorinate various known PET probe
precursors with diverse reactivity in a subsequent flow-through microfluidic chip without performing
additional azeotropic distillation. The fluorination yields under the optimized condition for the protected
18
18
18
18
[
F]FDG, protected [ F]FLT, 4-[ F]fluoroethylbenzoate, and [ F]fallypride were 93%, 96%, 77% and 73%,
Received 3rd May 2014
Accepted 27th May 2014
+
respectively. This method, utilizing the PS-Im monolith on a flow through microfluidic platform, enables
the entire fluorine-18 radiochemistry to be performed on a flow-through microfluidic device within a
shorter synthesis time and with fluorination efficiency that is comparable to or higher than conventional
means.
DOI: 10.1039/c4ra04064c
www.rsc.org/advances
could allow researchers to produce desirable PET tracers
themselves at the imaging center, with the potential to increase
Introduction
Flow-through microuidic devices have been utilized in a the availability and diversity of the number of probes used in
2,6
myriad of chemical syntheses, with enhanced reaction kinetics, preclinical research. The microliter volumes in microuidics
product purity and selectivity. The small reaction volume, are also commensurate with the nanograms of product needed
2,7,8
intrinsic to microuidic devices, allows for fast heat and mass for PET imaging.
Batch microuidic platforms that can
transfer, which is difficult to achieve in conventional macro- perform multistep synthesis of F-18 labelled compounds have
1
scale chemical synthesis. In addition, microuidic devices been demonstrated, including the capability for standard
allow for the reduction in reagent consumption and waste operations such as solvent exchange, evaporation, and heat-
9,10
production. The use of small quantities of materials enables the ing.
Though ow-through microuidic devices have many
use of hazardous and sensitive substances in a safer environ- advantages over batch devices, including faster reaction rates
ment. In medical diagnostics, specically in the synthesis of and greater simplicity, many of these standard operations are
1
1,12
short-lived (F-18 t1/2 ¼ 109 min) radiotracers for positron difficult to perform on ow through microdevices.
The
emission tomography (PET), the small dimensions of micro- majority of the chemical syntheses demonstrated on a ow
13

uidics have an added advantage of dramatically reducing the through microuidic is limited to a single step, multiple
14
2
–4
amount of radiation shielding needed to protect the operator.
reaction steps without intermediate separation or multiple
15
PET is a highly sensitive and non-invasive real-time visuali- reaction steps with solid phase supported reagents. Whereas
zation technique for biochemical and physiological processes solvent exchange is routinely performed in batch microuidics,
5
in living organism. Recently, there has been much progress in there are only few examples of an efficient solvent exchange
1
3,16,17
developing microuidic radiosynthesizers to produce PET processes within ow-through microuidic systems.
In
1
8
ꢀ
probes. The compact size of a microuidic radiosynthesizer F-18-radiochemistry, which uses [ F]F ions for labeling
molecular probes, an efficient ow-through solvent exchange
1
8
ꢀ
system for concentrating and activating [ F]F ions is neces-
sary to develop a fully integrated ow-through synthesizer
system suitable for multistep syntheses.
a
Department of Molecular and Medical Pharmacology, University of California, Los
Angeles, Los Angeles, CA 90095, USA. E-mail: pkeng@mednet.ucla.edu
b
Crump Institute for Molecular Imaging, David Geffen School of Medicine, University of
18
ꢀ
There are several reports on concentration of [ F]F ions on
California, Los Angeles, Los Angeles, CA 90095, USA

ow-through microuidic devices using quaternary ammonium
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c4ra04064c
9,10,18
ion exchange resins.
However, these methods do not
1
25348 | RSC Adv., 2014, 4, 25348–25356
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
address the issue of solvent exchange on a microuidic chip systems, packing micron-sized polymer beads within a micro-
and still rely on a conventional evaporative drying process to channel in ow-through chip can be challenging. Hence, we
activate uoride ions aer release, which undermines the chose to polymerize functional polymers monolith directly
advantages of using microuidic approaches for the production within the microchannel (dimension ¼ 150 ꢂ 150 ꢂ 52000 mm)
1
8
of PET probes. The commercially available ow-through and within a ETFE tubing (ID ¼ 400 mm) for [ F]uoride
microuidic systems Nano-Tek (Advion, Ithaca, NY) and the concentration and activation. The in situ monolith approach
FlowExpert (Future Chemistry) also uses quaternary ammo- addressed the difficulties associated with packing sufficient
1
8
26
nium ion exchange resins for [ F]uoride ion concentration, amount of beads within the microuidic chips.
and utilize a conventional reactor vessel to perform azeotropic
1
8
19
distillation to activate the [ F]uoride ions. Such an approach
not only increases the overall footprint of the microuidic
+
ꢀ
Polystyrene-imidazolium-chloride (PS-Im Cl ) monolith
platform, but also suffers from signicant loss of radioactivity A polymer monolith of styrenic backbone is suitable in
1
8
when transferring the dried [ F]uoride complex from the synthetic chemistry applications due to its high mechanical and
27
macrovolume reactor vial into the ow-through (capillary) chemical stability. In this report, we chose to synthesize a
1
8
ꢀ
reactor. Recently, electrochemical methods for [ F]F concen- polymer monolith containing imidazolium cations because
tration and subsequent solvent exchange were developed in a multiple imidazolium cations in close proximity are known to
2
0–23
18
ꢀ

ow-through microuidic system.
The [ F]F complex be selective in trapping uoride ions, due to the strong
28–31
obtained from the electrochemical ow through cell was interaction of the (C–H)+/X with uoride ions.
First, a
directly used in the radiouorination reaction of various poly(vinylbenzylchloride-co-divinylbenzene) monolith was
substrates. The authors reported high uorination efficiencies synthesized on a vinylized glass microuidic chip to covalently
1
8
without performing additional drying steps of the [ F]uoride anchor the polymer monolith onto the glass surface. The photo-
23
ions. However, the glassy carbon electrode used in this report polymerization reaction between vinyl benzylchloride (mono-
21
degrades with use and resulted in poor synthesis reliability.
mer) and divinylbenzene (crosslinker) was performed according
The van Dam group later reported the use of a brass-platinum to a modied procedure described for the synthesis of poly-
32
electrodes but the electrochemical ow cell utilized a large (styrene-co-divinylbenzene) monolith. Upon a series of opti-
1
8
volume of PTC mixture (ꢁ0.8 mL) to release the [ F]uoride mization studies, we found that a 7 : 3 ratio (v/v) of vinylbenzyl
2
2
ions.
chloride to divinylbenzene and a lower concentration of the
1
8
Here we report a simple methodology for activating [ F] S-camphore quinone (initiator) compared to the literature

uoride ions in a ow-through microuidic chip (or within report yielded a monolithic structure with moderate hydrody-
ethylene tetrauoroethylene (ETFE) tubing) using an imidazo- namic ow and mechanical stability. Alternatively, the poly-
lium ion polymer monolith. This new method for activating (vinylbenzylchloride-co-divinylbenzene) monolith can also be
1
8
[
F]uoride within a compact platform has the potential to directly polymerized within a pre-graed ETFE tubing. The
reduce the overall time, space and additional evaporation vessel graing of covalent moieties onto the inner walls of the ETFE
for synthesizing PET probes. Additionally, this system can also tubing was performed according to a modied literature report
1
8
33
enable concentration of [ F]uoride ions into a small volume (details in ESI†). ETFE tubing is a cheaper alternative to glass
of solvent, which can be easily integrated with both ow- microuidic chip, specically during the developmental period
through and batch microuidic devices.
where the size and length of the tubing can be easily tailored.
For example, the one-time use glass microuidic chip cost
ꢁ
US$43, while the 5 cm ETFE tubing only cost ꢁUS$0.73 per
Results and discussions
experiment. We found that the optimal feed ratio of monomer
1
8
ꢀ
Concentration of [ F]F ions are usually carried out on to crosslinker for the polymerization within the ETFE tubing
quaternary ammonium resins, which consist of 50–200 micron was 1 : 1 (v/v). The concentration of S-camphore quinone used
sized polymer beads. Luxen et al., have used macroporous in the polymerization reaction was the same for the corre-
copolymer of N-vinyl lactame and divinylbenzene infused with sponding styrenic polymer monolith reported in the literature.
long alkyl chain quaternary ammonium carbonate to carry out The irradiation time of the polymerization reaction was also
32
1
8
the solvent exchange and activation of [ F]uoride ions. The optimized to 180 minutes. Shorter irradiation time resulted in a
1
8
quaternary ammonium [ F]uoride salts are then owed out of polymer monolith that exhibited poor resistance to pressure as
the polymer with acetonitrile and subsequently used in the the monolith burst out from the ETFE tubing upon the initial
radiouorination of various aromatic and aliphatic compounds washing step. Longer duration of photo-polymerization reac-
1
8
ꢀ
24
without further activation of the [ F]F ions. Wester et al. tion resulted in too high of a back pressure, with no ow of
1
8
have also eluted [ F]uoride ions from quaternary ammonium liquid through the polymer monolith. When the same poly-
resins using a pre-dried KOH/cryptand solution in acetonitrile merization condition was performed within non-functionalized
and performed radiouorination of various PET probes without ETFE tubing, the monolith was washed off from the ETFE
performing azeotropic distillation. Both of these approaches tubing as there was no covalent anchoring site to hold the
used beads trapped within a cartridge and required large monolith within the tubing.
25
volumes (ꢁ0.6 mL to 1 mL) of phase transfer catalyst solution to
Upon the polymerization, a solution of N-methyl imidazole
achieve >90% releasing efficiency. However, in microuidic in acetonitrile was owed slowly through both the microuidic
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 25348–25356 | 25349

View Article Online
RSC Advances
Paper
+
ꢀ
chip and ETFE tubing containing the poly(vinylbenzylchloride- PS-Im Cl monolith was synthesized in ETFE tubing instead of
ꢃ
ꢃ
co-divinylbenzene) monolith at 80 C and 100 C, respectively, to the microuidic chip. Due to the larger inner diameter of the
form the imidazolium cation. The optimal reaction temperature ETFE tubing (400 mm) than the inner diameter of the micro-
1
8
for functionalization of poly(vinylbenzylchloride-co-divinylben- uidic chip (150 mm), [ F]uoride ion from the cyclotron in
+
ꢀ
zene) monolith with N-methyl imidazole on ETFE tubing was 100% water was able to ow through PS-Im Cl monoliths in
higher than in microuidic chip presumably due to the lower ETFE tubing. The maximum ow rate that can be achieved in
ꢀ
1
heat conductivity of ETFE polymer to that of borosilicate glass the ETFE tubing was about 250 mL min before leaking was
1
8
ꢀ
of the microuidic chip and the poorer heat and mass transfer observed. Under this condition, we were able to trap [ F]F
of the larger inner diameter of the ETFE tubing. We also found ions with 97% ꢄ 4 (n ¼ 39) efficiency. In these experiments low
that a continuous ow of fresh N-methyl imidazole solution is radioactivity, in the range of 1.5 MBq–7.4 MBq, were used to
required for complete functionalization of the poly- minimize exposure to harmful radiation. To show that the
(
vinylbenzylchloride-co-divinylbenzene) monolith. In our capacity of the monolith is sufficient to trap more than 37GBq
1
8
attempt to incubate the same concentration solution and heat of radioactivity, we added KF to the [ F]uoride solution.
for the same amount of time and at the same temperature, we Assuming it is desirable to trap [ F]F from the entire cyclotron
found that the functionalization was incomplete and resulted in target, we have also determined the minimum length of the PS-
1
8
ꢀ
1
8
poor [ F]uoride trapping efficiency.
Im monolith needed to quantitatively trap the high level of
radioactivity produced in the cyclotron by adding an equivalent
amount of uoride (in the form of KF) present in 74 GBq of
1
8
+
ꢀ
[
F]Fluoride trapping on PS-Im Cl monolith in microuidic
18
ꢀ
radioactivity (based on the theoretical specic activity of [ F]F
chip and ETFE tubing
ꢀ1
ion of 62 TBq mmol ). We found that the majority of the
1
8
[
F]Fluoride ions from the cyclotron, produced by proton radioactivity was trapped within the rst 2 cm of the polymer
1
8
+
ꢀ
bombardment of [ O]H O, were owed through the PS-Im Cl
monolith. To be conservative, we used a 5 cm polymer monolith
2
monolith in the microuidic chip using a syringe pump. The for the rest of our study because the 5 cm monolith is the
+
ꢀ
back pressure of the PS-Im Cl monolith increased signicantly shortest length that can accommodate a standard size tting on
1
8
when [ F]uoride solution in 100% water was owed through both ends of the tubing. With further engineering, one could
+
ꢀ
the PS-Im Cl monolith in comparison to when organic reduce this to a 2 cm monolith to achieve lower bed volume and
solvents were used to ow through the same monolith. The thus a lower elution volume.
+
ꢀ
polystyrenic backbone in the PS-Im Cl monolith can aggregate
or collapse in the presence of bad solvents, thus creating a high
backpressure in the microuidic chip. A moderate ow of liquid
was obtained with 1 : 1 (v/v) mixture of water and tetrahydro-
furan (THF). Increasing the ow rate of [ F]uoride solution in Various PTC, bases, concentrations and solvent compositions
the THF and water mixture also resulted in increased back were investigated to achieve the highest recovery of the [ F]
1
8
+ 18
ꢀ
[
F]Fluoride releasing from PS-Im [ F]F monolith in
microuidic chip and ETFE tubing
1
8
1
8
+
ꢀ
pressure, which resulted in leaks through the low pressure uoride ion in minimal volume from the PS-Im Cl monolith.
connectors and ttings. The percentage of radioactivity released using 7.5 mM concen-
It is highly desirable to ow the [ F]F ions in [ O]H
directly from the cyclotron, without the addition of other 80%, while higher concentration of NEthyl HCO (100 mM)
1
8
ꢀ
18
+
ꢀ
3
2
O
tration of NEthyl
4
HCO
in acetonitrile with 5% water was
+
4
ꢀ
3
solvents, through the polymer monolith at the highest ow rate. with 0.1% of water only released 56% of the radioactivity. These
1
8
ꢀ
To overcome the difficulties associated with the back pressure results indicated that water may help in solvating the [ F]F
in the microuidic chip using 100% water solution, the ions into the organic phase from the solid PS-Im cations. We
18
ꢀ + 18 ꢀ
a
Table 1
[ F]F ions releasing efficiency from PS-Im [ F]F monolith
Entry
Base
Concentration (mM)
Solvent
Volume (mL)
% Releasing eff.
1
2
3
4
5
6
7
8
9
CaCO
3
50
33
50/200
25/100
100
7.5
25
25
25
25/50
25/50
H
H
2
O
O
100
100
200
200
200
200
200
200
200
200
200
94 ꢄ 6, n ¼ 2
93 ꢄ 6, n ¼ 2
88 ꢄ 6, n ¼ 2
77 ꢄ 9, n ¼ 4
53 ꢄ 11, n ¼ 2
80
91 ꢄ 5, n ¼ 3
93 ꢄ 8, n ¼ 3
68 ꢄ 28, n ¼ 3
82 ꢄ 15, n ¼ 3
85 ꢄ 3, n ¼ 3
K
2
K
2
K
2
CO
CO
CO
3
2
3
3
/K2.2.2
/K2.2.2
2
MeCN/1% H O
MeCN/0.5% H
MeCN/0.1% H
MeCN 5% H
MeCN 1% H
MeCN 0.5% H
MeCN
2
O
O
+
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
NEthyl
NEthyl
NEthyl
NEthyl
NEthyl
4
4
4
4
4
HCO
HCO
HCO
HCO
HCO
3
3
3
3
3
2
+
+
+
+
2
O
O
2
2
O
O
1
1
0
1
KHCO
KHCO
3
/K2.2.2
/K2.2.2
MeCN 1% H
MeCN 0.5% H
2
O
3
2
a
The trapping and releasing experiments for entries 1–6 were performed on microuidic chips. Entries 7–11 were performed within 5 cm ETFE
tubings.
25350 | RSC Adv., 2014, 4, 25348–25356
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
1
8
found that the [ F]uoride ion releasing efficiency using in acetonitrile/water (99.5 : 0.5 v/v) (25 mM; 200 mL) to the
various PTC in 25 mM concentration of PTC in MeCN with 0.5% mannose triate (solid) and owed through the preheated PS-
+
18
ꢀ
ꢀ1
and 1% water in acetonitrile were similar and were within Im [ F]F monolith at a ow rate of 10 mL min . This condi-
experimental error (Table 1, entry 7,8,10,11). We observed lower tion resulted in no radiouorination product but 76% of the
1
8
releasing efficiency (68 ꢄ 28% (n ¼ 3)) and higher standard trapped [ F]uoride ion was released from the monolith.
+
ꢀ
deviation of ꢁ20% when 25 mM of NEthyl
4
HCO
3
in 100% Under similar reaction conditions, 4-trimethylamonnium ethyl
MeCN was used as the PTC solution (Table 1; entry 9). There- benzoate (FB-precursors), also gave no uorination product, but
+
18
fore, we chose to use 25 mM of different phase transfer catalyst 87 ꢄ 7% (n ¼ 5) of radioactivity was recovered from PS-Im [ F]
ꢀ
solutions in MeCN with 0.5% water for the rest of the studies
described in this report.
F
monolith upon owing the mixture of PTC and precursor
+
18
ꢀ
solution through the preheated PS-Im [ F]F monolith at 120
C. We speculated that the PS-Im Cl monolith may have
ꢃ
+
ꢀ
Imidazolium cations in ionic liquids can act as a catalyst in
nucleophilic uorination reaction, hence we rst attempted to affected the uorination reaction, resulting in no radio-
1
8

uorinate mannose triate (the precursor for 2-[ F]uoro-2- uorination product.
1
8
deoxy-D-glucose, [ F]FDG) by owing the precursor in anhy-
drous MeCN through the microuidic chip with the PS-Im [ F] support in the radiouorination reaction, we rst released the
F monolith at 120 C (Fig. 1a). The radiouorination efficiency radioactivity from the PS-Im F monolith with 200 mL of 25 mM
was measured to be 100%. However, the amount of radioactivity CO /K2.2.2 in 99.5 : 0.5 v/v MeCN/water into an Eppendorf
recovered from the monolith was less than 2%. This prelimi- tube containing the FB-precursor (solid) (experimental setup
nary experiment suggested that the [ F]uoride ion and the not shown). The mixture was subsequently owed through a
PS-Im ionic interaction is too strong and needs a stronger preheated (120 C) empty microuidic chip at 10 mL min
counter ion to displace the uoride ion from the polymer using a syringe pump. Radiouorination of the FB-precursor to
To eliminate the negative microscale effect of the polymer
+
18
ꢀ
ꢃ
+
ꢀ
K
2
3
1
8
+
ꢃ
ꢀ1
1
8
support. Based on this hypothesis we chose to add K
2
CO
3
/K2.2.2 the corresponding 4-[ F]uoro-ethylbenzoate was obtained
with 90% efficiency. Radiouorination reaction of mannose
triate with the same PTC solution gave varying radio-
ꢀ
1

uorination efficiency (33–77%) at 10 mL min . The formation
of uorinated products from mannose triate and FB-precur-
ꢀ
1
sors at 10 mL min ow rate also indicated that hydrolysis of
mannose triate and FB-precursor is unlikely, in presence of
small amount of water. To eliminate the need for an interme-
diate macroscopic vessel, we postulated that the base and
precursor solutions can be owed together through the PS-
+
18
ꢀ
Im [ F]F monolith directly into a preheated microuidic chip
to conduct the uorination reaction (Fig. 1b) in a continuous

ow fashion. Furthermore the one-step, continuous ow
process overcomes the large loss of radioactivity during the
transfer step and eliminate the complexity of owing the
1
8
precursor solution and the [ F]uoride ion complex into the
next microuidic chip for the subsequent uorination reaction.
1
8
In the new reaction setup (Fig. 1b), the [ F]uoride ion was
+
ꢀ

rst trapped on the PS-Im Cl monolith, and subsequently a
mixture of precursor and PTC solution was owed through the
+
18
ꢀ
PS-Im [ F]F monolith into a preheated empty microuidic
1
8
ꢀ
chip. To demonstrate the applicability of the activated [ F]F
+
complex released from the PS-Im monolith, radiouorination
of precursors with diverse leaving group using various base and
PTC in different solvent systems were investigated. The radio-

uorination efficiency of common PET probe precursors (Fig. 2)
ꢀ1
was optimized at a ow rate of 10 mL min with a residence
time of 78 s at 120 C (Table 2).
ꢃ
The radiouorination efficiency of mannose triate was
optimized with approximately 1 : 1.9 ratios of base and
Fig. 1 Experimental set-ups for the flow through radiofluorination
18
reaction using the activated [ F]fluoride complex released from the precursor. Radiouorination of mannose triate is typically
monolith: (a) the precursor and PTC solution was flowed through the
34
carried out with K
uorination of mannose triate with this PTC resulted in low
radiouorination efficiency, although the recovery of the
2 3
CO and K2.2.2 but in our study, the radio-
+
ꢀ
same monolith containing the PS-Im F on a pre-heated heating

block, (b) a mixture of PTC and precursor was flowed through both the
+
ꢀ
activated PS-Im F monolith and an empty glass microfluidic chip on
a preheated block for the radiofluorination reaction. The blue arrow
represents the direction of the liquid flow.
+
ꢀ
radioactivity from PS-Im Cl monolith was high. To improve
the radiouorination efficiency of mannose triate, we
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 25348–25356 | 25351

View Article Online
RSC Advances
Paper
radiochemical yield dropped from 64 ꢄ 17% (n ¼ 3) to 15 ꢄ 2%
(
n ¼ 2) (Table 2, entry 6).
The radiouorination of FLT-precursor, 3-N-tert-butox-
0
0
0
0
ycarbonyl-(5 -O-(4,4 -dimethoxytriphenylmethyl)-2 -deoxy-3 -O-
4-nitrobenzenesulfonyl)-b-D-threopentofuranosyl)thymine, was
carried out with approximately 1 : 2 ratio base to precursor
concentration in acetonitrile. The use of KHCO gave the
(
3
highest radiouorination efficiency of 93% ꢄ 4 (n ¼ 3) in a 3 : 1
ratio of thexyl alcohol/acetonitrile solution at a ow rate of 10 mL
ꢀ
1
18
min (Table 2, entry 8). However we found that the [ F]uo-
+
ꢀ
ride recovery from the PS-Im F monolith was poor when the
cation of the PTC was changed from K /K
Table 2, Entry 8 and 9). Furthermore, the NEthyl
+
+
to NEthyl4
2
.2.2
+
ꢀ
(
4
HCO
3
PTC
resulted in the formation of a red precipitate during the uo-
rination reaction, which clogged the microuidic channel.
Similarly to the radiouorination of the FB-precursor, the
radiouorination of FLT-precursor was lower at a ow rate of
ꢀ
1
1
00 mL min with residence time of 7.8 seconds than at 10 mL
ꢀ
1
min (Table 3, entry 3).
The optimized radiouorination condition of tosyl-fallypride
was carried out with approximately 1 : 1.6 ratio of K CO to
2 3
precursor concentration in MeCN and 0.5% water to achieve a
1
8
Fig. 2
[ F]Fluorination of different PET probe precursors using the
18
+
activated
monolith.
[ F]fluoride ion complex released from the PS-Im
radiochemical yield of 73 ꢄ 4% (n ¼ 3) (Table 2, Entry 10). No
3
radiouorination product was observed when KHCO /K2.2.2 or
+
ꢀ
NEthyl HCO were used as the PTC. The use of thexyl alcohol
4 3
attempted to use thexyl alcohol as a co-solvent as reported by
Kim et al.; however, the radiouorination efficiency was not
also resulted in no radiouorination product. Similarly to our
35
observation with FLT and FB precursors, the radiochemical
improved. We found that switching the PTC to 25 mM
1
8
ꢀ1
yield of [ F]fallypride at a lower ow rate of 10 mL min
resident time of 78 second) was higher than at a faster ow rate
+
ꢀ
NEthyl HCO
3
solution in MeCN with 0.5% water (Table 2,
(
ꢀ
1
entry 4) at a ow rate of 10 mL min (residence time of 78
seconds) resulted in 78 ꢄ 7% (n ¼ 3) decay-corrected radio-
chemical yield. We also obtained similar radiochemical yield of
ꢀ1
of 100 mL min (resident time 7.8 seconds).
Specic activity studies were conducted using FLT-precusor
1
8
using the set-up shown in Fig. 1b. First [ F]uoride from
8
7
6 ꢄ 7%, n ¼ 4 (Table 3, entry 2) with shorter residence time of
+
cyclotron was owed through the PS-Im monolith inside ETFE
ꢀ1
.8 seconds (ow rate 100 mL min ).
tubing without addition of KF. This was followed by a solution
of FLT precursor and KHCO3 (2 : 1 molar ratio) in acetonitrile,
which was directly owed into the preheated glass microuidic
chip. The products was collected in a eppendroff tube and
Conventionally, radiouorination of FB-precursor is carried
36
out in DMSO, but due to the high viscosity of DMSO (1.996
mPa s) and the small dimension of the ETFE tubing and the
microuidic chip, the radiouorination of FB-precursor was
carried out in MeCN (viscosity 0.343 mPa s). The radio-
uorination of FB-precursor was initially carried out with 1 : 2
ratio of K CO to FB-precursor with a ow rate of 10 mL min
ꢃ
hydrolysed with 200uL of 1 M HCl at 100 C. The specic activity
1
8
of [ F]FLT using this set-up was determined to be 21 GBq/umol
n.d.c). It is known that the specic activity of a radioisotope

(
ꢀ1
,
2
3
depends on the initial activity used in the experiment and
decreases exponentially with time as shown in equation below,
resulting in an overall radiochemical yield of 41 ꢄ 21% (n ¼ 4).
The radiouorination efficiency was high (79 ꢄ 8%, n ¼ 4), but
+
ꢀ
ðꢀln2t=T1=2
Þ
the recovery of the radioactivity from the PS-Im F monolith
was poor (Table 2, Entry 5). Furthermore, we observed the
formation of a precipitate upon dissolving the FB-precursor
AðtÞe
SAðtÞ ¼
m
with the K
2 3
CO /K2.2.2 solution, which might have resulted in the
A is the amount of radioactivity (Bq), t is time (s) T is half
1
8
1/2
lower releasing efficiency of the [ F]uoride ions from the PS-
18
37
life of a radioisotope(s) for F and m is mass in mols. In our
experiment, we use low activity in the range of 74 MBq to
measure the specic activity, hence a lower specic activity was
obtained. Upon increasing the initially activity, the specic
activity can be increased using this system.
We have demonstrated comparable radiochemical yield of
the uorinated intermediate to conventional methods using
F]uoride ions that were concentrated and activated on the
imidazolium monolith in a simple ow-through system without
+
ꢀ
Im F monolith. To improve the overall radiochemical yield, we
investigated KHCO as the base for the radiouorination of the
FB precursor. The use of KHCO /K
3
and FB precursor mixture
2.2.2
3
1
8
did not form any precipitate. The radiochemical yield of 4-[ F]
uoro-ethyl benzoate was improved from 41% to 64% (Table 2,
entry 5 versus entry 6) switching to a PTC consisting of a 2 : 1
ratio of KHCO to FB-precursor. The optimal ow rate and
residence time were found to be 10 mL min and 78 second. At

3
18
[
ꢀ
1
ꢀ
1
higher ow rate, of 100 mL min (7.8 s residence time), the
performing an additional azeotropic distillation step. The
25352 | RSC Adv., 2014, 4, 25348–25356
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
ꢀ1
Table 2 Fluorination of various PET probe precursors using different PTC at a flow rate of 10 mL min
Conc. precursor
(mM)
% Fluorination
Eff.
% Releasing
Eff.
a
Entry
1
Precursor
Base/PTC (mM)
Solvent
% R.C.Y
2
K
5 K
2
CO
3
/100
/100
MeCN, 0.5%
25 ꢄ 14,
n ¼ 3
88 ꢄ 14,
n ¼ 3
45
46
34 ꢄ 21, n ¼ 3
38 ꢄ 32, n ¼ 4
2.2.2
H
2
O
9
: 1
25 K
2
CO
3
MeCN : Thexyl
alcohol 0.5%
H O
2
34 ꢄ 29,
n ¼ 4
96 ꢄ 0.6,
n ¼ 4
2
K
2.2.2
2
K
5 KHCO
3
/50
MeCN, 0.125
H O
2
22 ꢄ 11,
n ¼ 3
81 ꢄ 13,
n ¼ 3
3
4
42
48
41 ꢄ 13, n ¼ 3
93 ꢄ 4, n ¼ 3
2.2.2
25 N-Ethyl
4
MeCN, 0.5%
H O
2
78 ꢄ 7,
84 ꢄ 7,
HCO
3
n ¼ 3
n ¼ 3
2
K
5 K
2
CO
3
/100
MeCN, 0.5%
H O
2
41 ꢄ 21,
n ¼ 4
39 ꢄ 9,
n ¼ 4
5
6
48
47
79 ꢄ 8, n ¼ 4
96 ꢄ 5, n ¼ 3
2.2.2
50 KHCO
3
/100
MeCN, 0.5%
64 ꢄ 17,
n ¼ 3
89 ꢄ 5,
n ¼ 3
K
2.2.2
H
2
O
9
: 1
1
K
2 K
2
CO
3
/48
MeCN : Thexyl
alcohol 0.5%
7
8
9
21
54
51
18
8
94
2.2.2
H
1
2
O
: 3
25 N-Ethyl
4
MeCN : Thexyl
alcohol 0.5%
44 ꢄ 5,
n ¼ 3
50 ꢄ 4,
n ¼ 3
93 ꢄ 4, n ¼ 3
77 ꢄ 3, n ¼ 3
HCO
3
H
2
O
1
: 3
25 KHCO
3
/50
MeCN : Thexyl
alcohol 0.5%
H O
2
60 ꢄ 3,
n ¼ 3
96 ꢄ 2,
n ¼ 3
K
2.2.2
2
K
5 K
2
CO
3
/100
MeCN, 0.5%
H O
2
65 ꢄ 7,
n ¼ 3
91 ꢄ 10,
n ¼ 3
1
0
39
73 ꢄ 4, n ¼ 3
2.2.2
a
R.C.Y is calculated by multiplying the radiouorination efficiency (based on radio-TLC) and the percent ratio of collected radioactivity aer
uorination/starting radioactivity trapped on the PS-Im Cl monolith.
+
ꢀ

leaving group reactivity of the precursors dictated the extent of ltered with basic alumina oxide to remove the stabilizer. Tosyl-
0
0
radiouorination of the precursors as observed at a higher ow fallypride, mannose triate, 3-N-Boc-5 -O-dimethoxytrityl-3 -O-
ꢀ1
rate of 100 mL min (residence time 7.8 second) (Table 3). Our nosyl-thymidine, 4-(ethoxycarbonyl)-N,N,N-trimethylbenzena-
observation is consistent with literature, in which mannose minium triate were purchased from ABX (Advanced
triate is the most reactive precursors and the FB-precursor is Biochemical Compounds; Radeberg, Germany) and used
the least reactive precursor that were investigated in this study. without further purication. Ethylenetetrauoroethylene
Although the nosylate group is known to be more reactive than (ETFE) (OD: 0.159 cm) tubing with 400 mm inner diameter was
the tosyl group, our result with the FLT and fallypride precursor purchased from McMaster-Carr (Santa Fe Springs, CA, USA). A
yielded comparable yield (within experimental error). This 470 nm LED array (RLT-MIL470-12B-30) was purchased from
observation may be attributable to a steric factor, in which the Roithner LASER Technik (Vienna, Austria) consisted of 12 LEDs
FLT precursor is a secondary substrate while the tosyl-fallypride emitting at 470 nm with 20 mW of power each. Microprocessor-
is a primary substrate.
Controlled UV-crosslinker Spectrolinker TM XL-1000 was
bought from Spectronics Corporation (Westbury, NY, USA).
Microuidic chip (150 mm
FC_R150.676.2_PACK with chip holder was purchased from
Micronit Microuidics (The Netherlands). Analytical HPLC was
performed on a Knauer Smartline HPLC system (1 mL min
ꢂ
150 mm
ꢂ
5000 mm),
Materials
All chemicals and solvents were purchased from Sigma-Aldrich
(Milwaukee, WI, USA). Ethylene dimethacrylate (EDMA),
ꢀ1
)
divinylbenzene (DVB) and 4-vinylbenzyl chloride (VBC) were
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 25348–25356 | 25353

View Article Online
RSC Advances
Paper
ꢀ1
Table 3 Radiofluorination of various PET probe precursors using different PTC at a flowrate of 100 mL min
Conc. precursor
(mM)
% Fluorination
eff.
% Releasing
eff.
a
Entry
Precursor
Base
Solvent
% R.C.Y
2
K
5 K
2
CO
3
/100
MeCN, 0.5%
H O
2
30 ꢄ 15,
n ¼ 2
92 ꢄ 11,
n ¼ 2
1
2
39
48
34 ꢄ 10, n ¼ 2
92 ꢄ 2, n ¼ 4
2.2.2
25 N-Ethyl
4
MeCN, 0.5%
86 ꢄ 7,
n ¼ 4
90 ꢄ 2
n ¼ 3
HCO
3
H
2
O
1
: 3
2
K
5 KHCO
3
/50
MeCN : Thexyl
alcohol 0.5%
H O
2
24 ꢄ 10,
n ¼ 3
59 ꢄ 15,
n ¼ 3
3
49
50
41 ꢄ 8, n ¼ 3
25 ꢄ 7, n ¼ 2
2.2.2
50 KHCO
3
/100
MeCN, 0.5%
H O
2
15 ꢄ 2,
n ¼ 2
67 ꢄ 28,
n ¼ 2
4
K
2.2.2
a
R.C.Y is calculated by multiplying the radiouorination efficiency (based on radio-TLC) and the percent ratio of (collected radioactivity aer
uorination/starting radioactivity trapped on the PS-Im Cl monolith).
+
ꢀ

with a Phenomenex reverse-phase Luna column (5 mm, 4.6 mm divinylbenzene) monolith was carried out following a published
32
ꢂ 250 mm) (California, U.S.A) with inline Knauer UV detector procedure with slight modication. The initiator solution was
(254 nm) and a gamma-radiation coincidence detector and prepared by dissolving S-(+)-camphorquinone (6.4 mg, 0.0386
counter (B-FC-4100 and B-FC-1000) (Bioscan, Washington D.C., mmol), ethyl-4-dimethylamino benzoate (32 mg, 0.165 mmol)
U.S.A). Radio-TLC plates were scanned with miniGitastar and N-methoxyphenylpyridinium tetrauoroborate (32 mg,
(
Raytest, Wilmington, NC, USA).
0.172 mmol) in 250 mL of acetonitrile, 500 mL of 1-propanol and
1
8
No-carrier-added [ F]uoride was produced by the (p,n) 600 mL. Vinylbenzyl chloride (350 mL, 2.49 mmol) and divinyl-
1
8
reaction of recycled [ O]H
2
O (85% isotopic purity, IBA) in a benzene (350 mL, 2.46 mmol) were mixed with the initiator
RDS-112 cyclotron (Siemens; Knoxville, TN, USA) at 11 MeV solution and sonicated for 20 min. EDMA functionalized ETFE
using a 300 mL or 1 mL tantalum target with havar foil.
tubing (detail in ESI†) with 400 mm I.D was lled with this
solution using a syringe and sealed. The sealed tubing was
placed between two arrays of 470 nm LED light source (12 ꢂ 20
mW at 4.2 A, 3.7 V) at 1 cm apart in a closed box for two and a
Experimental
Synthesis of poly(vinylbenzylchloride-co-divinylbenzene) half hours. White solid was formed in the tubing, which was
monolith microuidic chip. A solution containing photo- washed thoroughly with copious amount of acetonitrile. A
initiators was prepared by dissolving S-camphorquinone typical length of ETFE tubing used for the polymerization
(
2.4 mg; 0.14 mmol), ethyl-N-dimethylbenzoate (32 mg; 0.16 reaction was 30 cm. Aer the polymerization, the 30 cm tubing
1
8
mmol), N-methyl-pyridinium tetrauoroborate (32 mg; 0.18 was then cut into several 5 cm segments for subsequent [ F]
mmol) in a mixture of acetonitrile (237 mL), isopropanol (500 uoride ion concentration, activation and uorination on the
mL) and 1-decanol (600 mL). Vinylbenzyl chloride (500 mL, 3 ow-through platform.
mmol) monomer and divinylbenzene (250 mL, 1.8 mmols)
Functionalization of poly(vinylbenzylchloride-co-divinylben-
crosslinker were mixed thoroughly with the photoinitiator zene) monolith with N-methylimidazole. The poly(vinylbenzyl-
solution and sonicated for 20 min. The monomer and cross- chloride-co-divinylbenzene) in ETFE tubing and microuidic
linker solution was owed into a vinylized glass microuidic chip was functionalized by owing N-methyl imidazole (2 mL,
ꢃ
chip (in ESI†). The chip was placed between two panels of light- 0.2 M) solution in MeCN through the microchannels at 100 C
ꢃ
emitting diodes (LEDs) (450 nm; 12 ꢂ 20 mW at 4.2 A, 3.7 V). and 80 C, respectively, at a owrate of 0.1 mL per hour. The
For consistency, the LED lights were turned on for 1 hour before microuidic chip was heated on a metal block on a hotplate,
the polymerization reaction was carried out to equilibrate the while the ETFE tubing was wrapped with an aluminum foil to
heat released from the LED lights. The polymerization mixture ensure homogeneous heating of the entire ETFE tubing.
1
8
+
ꢀ
was irradiated 35–40 minutes. The unreacted monomers and
Trapping of [ F]uoride ions on PS-Im Cl in microuidic
1
8
cross-linkers were washed thoroughly by owing copious chip. [ F]Fluoride from the cyclotron was rst diluted with a
amount of THF through the microuidic chip using a syringe mixture of 1 : 1 ratio of THF and water with potassium uoride
pump.
(0.3 mM). 200 mL of this solution (corresponds to the amount of
Synthesis of poly(vinylbenzylchloride-co-divinylbenzene) uoride ions present in 74 GBq of activity) was owed through
monolith in a pre-graed ETFE tubing. The polymerization the imidazolium monolith in microuidic chip using a syringe
ꢀ
1
reaction to generate the poly(vinylbenzylchloride-co- pump at ow rate of 250 mL min
.
25354 | RSC Adv., 2014, 4, 25348–25356
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
1
8
+
ꢀ
Trapping of [ F]Fluoride ions on PS-Im Cl in ETFE tubing. directly used to uorinate various PET probe precursors with
F]Fluoride from the cyclotron was rst diluted with water diverse reactivity in a continuous ow microuidic chip. This
1
8
[
doped with potassium uoride (0.0017 g; 60 mmol) to corre- new ow-through platform enables the concentration and
1
8
ꢀ
spond to the amount of F-18 ions present in 74 GBq of radio- activation of [ F]F ions with high radioactivity recovery, which
activity in 200 mL (0.3 mM). This solution was then owed at can be easily integrated with any ow through microuidic
ꢀ
1
1
00–250 mL min through the ETFE tubing (5 cm) containing devices and yet maintained a small overall footprint of the
+
ꢀ
the PS-Im Cl monolith. The trapping efficiency was calculated microuidic radiosynthesizer. Based on this new method, the
by subtracting the amount of activity released. To determine the entire radiouorination process, starting with 1 mL of [ F]F /
minimum length of PS-Im Cl monolith needed to quantita- [ O]H O from the cyclotron target to produce the uorinated
2
1
8
ꢀ
+
ꢀ
18
tively trap ꢁ74 GBq equivalent of uoride ion, the ETFE tubing product on a ow-through platform takes about 24 minutes. For
was cut into 1 cm intervals aer the trapping step. The amount the radiouorination of the more reactive mannose triate
1
8
of radioactivity trapped within each 1 cm piece was measured (precursor for the synthesis of [ F]FDG), the entire process can
using a dose calibrator. Nearly 100% of the radioactivity was be shortened to about 6 minutes. We anticipate this method has
found in the rst two 1 cm segments, suggested that a 2 cm the potential to expand the usefulness of ow-through micro-
length is sufficient to trap all the activity.
uidics for radiosynthesis by providing a means to perform the
Releasing [ F]Fluoride ions from the PS-Im [ F]F mono- critical solvent-exchange step entirely within microuidics
1
8
+ 18
ꢀ
1
8
ꢀ
+
lith. The [ F]F trapped on PS-Im monolith within the rather than requiring macroscale components for synthesizing
microuidic chip and ETFE tubing was released by owing a diverse arrays of PET probes on-demand.
various concentrations bicarbonate and carbonates with
different phase transfer catalyst (PTC) at a ow rate of 250 mL
References
ꢀ
1
18
ꢀ
min using a syringe pump. The releasing efficiency of [ F]F
+
ꢀ
ions from the PS-Im Cl monolith in the microuidic chip and
1 J. R. Goodell, J. P. McMullen, N. Zaborenko, J. R. Maloney,
C.-X. Ho, K. F. Jensen, J. A. Porco and A. B. Beeler, J. Org.
Chem., 2009, 74, 6169–6180.
2 A. M. Elizarov, Lab Chip, 2009, 9, 1326.
3 G. Pascali, P. Watts and P. A. Salvadori, Nucl. Med. Biol., 2013,
6, 776–787.
4 P. Y. Keng, M. Ester and R. M. van Dam, in Positron Emission
Tomography-Current Clinical and Research Aspect, InTech,
2012.
5 M. E. Phelps, Proc. Natl. Acad. Sci. U. S. A., 2000, 97, 9226–
9233.
ETFE was measured by subtracting the amount of activity
+
ꢀ
remained on the PS-Im Cl monolith from the amount of
+
ꢀ
activity trapped on the PS-Im Cl monolith aer the releasing
step.
General procedure for radiouorination of different PET
1
8
probe precursors. [ F]Fluoride from the cyclotron was rst
diluted with water doped with potassium uoride (0.3 mM). 200
1
8
ꢀ
mL of this water containing [ F]F ions was owed through
5
+
ꢀ
cm of PS-Im Cl on ETFE tubing connected to a downstream
microuidic chip with serpentine channel (Fig. 1b) using a
ꢀ
1
syringe pump at a ow rate of 100–250 mL min followed by
00 mL of acetonitrile and dried with air. Mixtures of precursors
with different PTC in various solvent system (Table 2) were
6 P. Y. Keng, S. Chen, H. Ding, S. Sadeghi, G. J. Shah,
2
A.
Dooraghi,
M.
E.
Phelps,
N.
Satyamurthy,
A. F. Chatziioannou, C.-J. Kim and R. M. van Dam, Proc.
Natl. Acad. Sci. U. S. A., 2011, 17, 690–695.
ꢀ1
18

owed at 10–100 mL min through the PS-Im-[ F]F monolith
and the empty microuidic chip, which was preheated to
7 C.-C. Lee, G. Sui, A. Elizarov, C. J. Shu, Y.-S. Shin,
A. N. Dooley, J. Huang, A. Daridon, P. Wyatt, D. Stout,
H. C. Kolb, O. N. Witte, N. Satyamurthy, J. R. Heath,
M. E. Phelps, S. R. Quake and H.-R. Tseng, Science, 2005,
310, 1793–1796.
8 V. Arima, G. Pascali, O. Lade, H. R. Kretschmer, I. Bernsdorf,
V. Hammond, P. Watts, F. D. Leonardis, M. D. Tarn,
N. Pamme, B. Z. Cvetkovic, P. S. Dittrich, N. Vasovic,
R. Duane, A. Jaksic, A. Zacheo, A. Zizzari, L. Marra,
E. Perrone, P. A. Salvadori and R. Rinaldi, Lab Chip, 2013,
ꢃ
120 C. Aer the uorination reaction of mannose triate, FB-
precursor and FLT-precursor, 200 mL of acetonitrile was owed
ꢀ1
at 100 mL min to wash the majority of the product out from
the chip. For the tosyl-fallypride reaction, 200 mL of methanol
was used to wash the product off the microuidic chip. The
product was collected and the amount of activity released was
determined with a dose calibrator. The radiouorination effi-
ciency was determined by radio-TLC. The radiochemical yield
(RCY) was calculated based on the following formula
1
3, 2328–2336.
Radioactivity of collected in vial ꢂ TLC conversion
RCY ¼
ꢂ100
9 A. Lebedev, R. Miraghaie, K. Kotta, C. E. Ball, J. Zhang,
M. S. Buchsbaum, H. C. Kolb and A. Elizarov, Lab Chip,
Radioactivity trapped on PS ꢀ ImCl
2012, 13, 136–145.
1
0 A. M. Elizarov, R. M. van Dam, Y. S. Shin, H. C. Kolb,
H. C. Padgett, D. Stout, J. Shu, J. Huang, A. Daridon and
J. R. Heath, J. Nucl. Med. Off. Publ. Soc. Nucl. Med., 2010,
51, 282–287.
Conclusions
We have developed a new method for concentrating and acti-
1
8
vating [ F]uoride ions in microliter volumes on a ow- 11 J. M. Gillies, C. Prenant, G. N. Chimon, G. J. Smethurst,
through microuidic chip and ETFE tubing that were func-
tionalized with PS-Im Cl monolith. The [ F]uoride ions were
B. A. Dekker and J. Zweit, Appl. Radiat. Isot., 2006, 64, 333–
336.
+
ꢀ
18
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 25348–25356 | 25355

View Article Online
RSC Advances
Paper
1
1
1
1
2 C. J. Steel, A. T. O'Brien, S. K. Luthra and F. Brady, J. Label. 25 S. H. Wessmann, G. Henriksen and H. Wester,
Compd. Radiopharm., 2007, 50, 308–311. Nuklearmedizin, 2012, 51, 1–8.
3 W. H. Grover, R. H. C. Ivester, E. C. Jensen and R. A. Mathies, 26 S. Frantisek, J. Chromatogr. A, 2010, 1217, 902–924.
Lab Chip, 2006, 6, 623–631.
27 C. W. Huck and G. K. Bonn, Chem. Eng. Technol., 2005, 28,
4 H. Usutani, Y. Tomida, A. Nagaki, H. Okamoto, T. Nokami
and J. Yoshida, J. Am. Chem. Soc., 2007, 129, 3046–3047.
5 I. R. Baxendale, J. Deeley, C. M. Griffiths-Jones, S. V. Ley,
1457–1472.
28 K. Sato, S. Arai and T. Yamagishi, Tetrahedron Lett., 1999, 40,
5219–5222.
S. Saaby and G. K. Tranmer, Chem. Commun., 2006, 2566– 29 E. Alcalde, I. Dinar `e s, A. Ib ´a n˜ ez and N. Mesquida, Chem.
568. Commun., 2011, 47, 3266–3268.
6 A. Hibara, K. Toshin, T. Tsukahara, K. Mawatari and 30 I. Dinar `e s, C. G. de Miguel, A. Ib ´a n˜ ez, N. Mesquida and
T. Kitamori, Chem. Lett., 2008, 37, 1064–1065. E. Alcalde, Green Chem., 2009, 11, 1507–1510.
7 B. Z. Cvetkovi ´c , O. Lade, L. Marra, V. Arima, R. Rinaldi and 31 J. Yoon, S. K. Kim, N. J. Singh and K. S. Kim, Chem. Soc. Rev.,
P. S. Dittrich, RSC Adv., 2012, 2, 11117–11122. 2006, 35, 355–360.
8 F. De Leonardis, G. Pascali, P. A. Salvadori, P. Watts and 32 Z. Walsh, P. A. Levkin, B. Paull, F. Svec and M. Macka, J. Sep.
N. Pamme, J. Chromatogr. A, 2011, 1218, 4714–4719. Sci., 2011, 33, 61–66.
9 J.-H. Chun, S. Lu, Y.-S. Lee and V. W. Pike, J. Org. Chem., 33 D. Connolly and B. Paull, J. Sep. Sci., 2009, 32, 2653–2658.
2
1
1
1
1
2
2
2010, 75, 3332–3338.
34 C. Lemaire, P. Damhaut, B. Lauricella, C. Mosdzianowski,
J.-L. Morelle, M. Monclus, J. Van Naemen, E. Mulleneers,
J. Aerts, A. Plenevaux, C. Brihaye and A. Luxen, J. Label.
Compd. Radiopharm., 2002, 45, 435–447.
0 K. Hamacher, T. Hirschfelder and H. Coenen, Appl. Radiat.
Isot., 2002, 56, 519–523.
1 H. Saiki, R. Iwata, H. Nakanishi, R. Wong, Y. Ishikawa,
S. Furumoto, R. Yamahara, K. Sakamoto and E. Ozeki, 35 Y.-H. Oh, D.-S. Ahn, S.-Y. Chung, J.-H. Jeon, S.-W. Park,
Appl. Radiat. Isot., 2010, 68, 1703–1708.
2 S. Sadeghi, V. Liang, S. Cheung, S. Woo, C. Wu, J. Ly, Y. Deng,
S. J. Oh, D. W. Kim, H. S. Kil, D. Y. Chi and S. Lee, J. Phys.
Chem., 2007, 111, 10152–10161.
2
M. Eddings and R. M. van Dam, Appl. Radiat. Isot., 2013, 75, 36 J. Becaud, L. Mu, M. Karramkam, P. A. Schubiger,
8
5–94.
S. M. Ametamey, K. Graham, T. Stellfeld, L. Lehmann,
S. Borkowski, D. Berndorff, L. Dinkelborg, A. Srinivasan,
R. Smits and B. Koksch, Bioconjugate Chem., 2009, 20,
2254–2261.
2
2
3 R. Wong, R. Iwata, H. Saiki, S. Furumoto, Y. Ishikawa and
E. Ozeki, Appl. Radiat. Isot., 2012, 70, 193–199.
4 J. Aerts, S. Voccia, C. Lemaire, F. Giacomelli, D. Goblet,
D. Thonon, A. Plenevaux, G. Warnock and A. Luxen, 37 J. Bergman, O. Eskola, P. Lehikoinen and O. Solin, Appl.
Tetrahedron Lett., 2010, 51, 64–66. Radiat. Isot., 2001, 54, 927–933.
25356 | RSC Adv., 2014, 4, 25348–25356
This journal is © The Royal Society of Chemistry 2014