Nucleophilic substitution of nitro groups by [18F]fluoride in methoxy-substituted ortho-nitrobenzaldehydes-A systematic study

Article abstract of DOI:10.1016/j.jfluchem.2008.10.003

As model reactions for the introduction of [¹⁸F]fluorine into aromatic amino acids, the replacement of NO₂ by [¹⁸F]fluoride ion in mono- to tetra-methoxy-substituted ortho-nitrobenzaldehydes was systematically investigated. Unexpectedly, the highly methoxylated precursors 2,3,4-trimethoxy-6-nitrobenzaldehyde and 2,3,4,5-tetramethoxy-6-nitrobenzaldehyde showed high maximum radiochemical yields (82% and 48% respectively). When the electrophilicity of the leaving group substituted carbon atom is expressed by its ¹³C NMR chemical shift a good correlation with the reaction rate at the beginning of the reaction (first min) was found (R² = 0.89), whereas the maximum radiochemical yields correlated much poorer with this electrophilicity parameter. This may be caused by side reactions becoming influencial in the further reaction course. As possible side reactions the demethylation of methoxy groups and intramolecular redox reactions could be detected by HPLC/MS.

Full text of DOI:10.1016/j.jfluchem.2008.10.003

Journal of Fluorine Chemistry 130 (2009) 216–224
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Nucleophilic substitution of nitro groups by [18F]fluoride in
methoxy-substituted ortho-nitrobenzaldehydes—A systematic study
a
b
Bin Shen a, Dirk Lo¨ffler a, Gerald Reischl a, , Hans-Ju¨rgen Machulla , Klaus-Peter Zeller
*
a Radiopharmacy, PET Centre, Eberhard Karls University Tu¨bingen, Germany
b Institute of Organic Chemistry, Eberhard Karls University Tu¨bingen, Germany
A R T I C L E I N F O
A B S T R A C T
As model reactions for the introduction of [18F]fluorine into aromatic amino acids, the replacement of
NO2 by
18F]fluoride ion in mono- to tetra-methoxy-substituted ortho-nitrobenzaldehydes was
Article history:
Received 18 July 2008
[
Received in revised form 10 October 2008
Accepted 13 October 2008
Available online 25 October 2008
systematically investigated. Unexpectedly, the highly methoxylated precursors 2,3,4-trimethoxy-6-
nitrobenzaldehyde and 2,3,4,5-tetramethoxy-6-nitrobenzaldehyde showed high maximum radio-
chemical yields (82% and 48% respectively). When the electrophilicity of the leaving group substituted
carbon atom is expressed by its 13C NMR chemical shift a good correlation with the reaction rate at the
beginning of the reaction (first min) was found (R2 = 0.89), whereas the maximum radiochemical yields
correlated much poorer with this electrophilicity parameter. This may be caused by side reactions
becoming influencial in the further reaction course. As possible side reactions the demethylation of
methoxy groups and intramolecular redox reactions could be detected by HPLC/MS.
Keywords:
Nucleophilic substitution
[
[
18F]Aromatic amino acid
18F]Fluoride ion
Isotopic labelling
PET
ß 2008 Elsevier B.V. All rights reserved.
1. Introduction
standard and most efficient production route for production of
fluorine-18 is the 18O(p,n)18
F
nuclear reaction resulting in
Positron emission tomography (PET) is an important imaging
technique that exhibits the unique feature of following metabolic
functions in vivo. The use of radioactively labelled compounds (i.e.
tracers or PET biomarkers) allows the quantitative measurement of
the distribution of radioactivity in vivo according to the
physiological or pathophysiological processes to be investigated.
18F-labelled aromatic amino acids, like FDOPA or tyrosine
derivatives, have been of interest for many years now. Commonly,
they are synthesized by electrophilic substitution, using [18F]F2,
[
18F]fluoride ion with a high specific activity ready for nucleophilic
substitution. Today, this reaction is available at every medical
cyclotron and therefore the method of choice.
For that reason nucleophilic aromatic substitution (SNAr) is
considered to be the most desirable reaction type to introduce 18
F
into aromatic amino acids. This approach, indeed, has become a
challenging task to radiopharmaceutical groups world wide [1].
Especially the demand to make [18F]FDOPA easily available
became a major driving force in various attempts. Enantiomeric
purity of this product often is a major problem [2], which was
circumvented by e.g. the use of a chiral catalyst [3]. Still, the
synthesis of [18F]FDOPA via SNAr is complex and laborious [4].
As the high electron density in the aromatic part of the amino
acids is counter-productive to a nucleophilic attack, a particular
interest has to be paid to the choice of leaving group in the
precursors to be applied for the particular radiofluorination.
Furthermore, auxiliary groups exerting an electron-withdrawing
effect are required to enhance the reactivity for the nucleophilic
attack of [18F]fluoride ion. For the synthesis of 18F-labelled arenes
typical precursors contain a nitro substituent as leaving group
which is activated for nucleophilic substitutions by an ortho- or
para-positioned aldehydic function [5]. In addition to its role as
auxiliary group in the nucleophilic introduction of 18F, the formyl
substituent has the advantage, that in the further course of the
[
18F]acetylhypofluorite etc. as reagents. Due to the type of nuclear
processes that are used at the cyclotron, labelling results in
products with low specific activities (ratio of 18F-compound (in Bq)
to 18F- plus 19F-compound (in mol)). At least in the case of
[
18F]FDOPA this generates no problem for the PET investigation,
but nevertheless it is a limitation. In many cases high specific
activities are crucial for following up metabolic processes without
altering the biochemical processes that are on-going in vivo.
In order to obtain labelled aromatic amino acids with high
specific activities, the incorporation of 18F into the tracer is to be
carried out with no-carried-added fluorination reagents. The
* Corresponding author. Tel.: +49 7071 2987443; fax: +49 7071 295264.
E-mail address: gerald.reischl@uni-tuebingen.de (G. Reischl).
0022-1139/$ – see front matter ß 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2008.10.003

B. Shen et al. / Journal of Fluorine Chemistry 130 (2009) 216–224
217
synthesis it can be used as starting point for the synthetic addition
of the amino acid residue to the aromatic part of the molecule [6].
Therefore, o-nitrobenzaldehydes substituted by methoxy groups
(as masked hydroxyl groups) have been selected as labelling
precursors for the synthesis of 18F-amino acids such as 2-
have a methoxy group at C-4 (meta-oriented with respect to NO2)
in common. Lower, however, still satisfactory radiochemical yields
were found for the fluorination reaction of f-NO2 (69%), d-NO2
(70%), b-NO2 (57%) and k-NO2 (48%). Compound h-NO2 which is
substituted by two methoxy groups in o- and p-position to the
leaving group showed a significantly lower maximum yield of only
13%. Furthermore, the RCY of 2,3-dimethoxy-6-nitrobenzaldehyde
(e-NO2) was low (22%); the reason could be the formation of a
radioactive by-product of unknown structure accounting for a
labelling yield of 15%. Therefore, this compound cannot be
considered to contribute to the discussion about SNAr in multiple
substituted benzaldehydes. In all other cases, no radioactive by-
product was observed, neither by radio-TLC nor by radio-HPLC.
In a further set of experiments the nitro substituent was
replaced by bromine as leaving group, thus resulting in identical
18F-labelled products. These experiments were carried out in order
to allow identification of labelled compounds (see Method 3,
analytical assay, Section 4). Results similar to the nitro compounds
were observed, however, due to the poorer qualities of the bromo
substituent as leaving group, the radiochemical yields generally
were much lower (Fig. 2).
[
18F]fluorotyrosine and [18F]FDOPA.
In the case of 5-methoxy-2-nitrobenzaldehyde a radiochemical
yield (RCY) of 5% was reported for the substitution of NO2-group by
18F [7]. The poor RCY is to be understood in terms of the strong +M-
effect of the CH3O-substitutent p-oriented to the leaving group.
Compared to this result, the radiochemical yields obtained for 4,5-
dimethoxy-2-nitrobenzaldehyde (25–55%) as labelling precursor
[8,9] are considered to be unexpectedly high. That prompted us to a
systematic study about the dependence of the RCY on the number
and positions of methoxy substituents present in o-nitrobenzal-
dehydes.
2. Results and discussion
2.1. 18F-labelling conditions
The goal of this work was to study the influence of additional
methoxy groups on the reactivity of 2-nitrobenzaldehyde in the
nucleophilic aromatic fluorination. In this type of reaction, the use
of polar aprotic solvents is mandatory in order to take advantage of
the nucleophilicity of the 18Fꢀanion. In literature, DMSO is
commonly used as solvent in this type of labelling reactions.
However, we previously observed the potential of DMSO to give
rise to oxidation of the benzaldehyde precursors to benzoic acids
[10]. Therefore, DMSO was replaced by DMF through this work in
order to avoid oxidation as a side reaction. For 18F-labelling the
reaction conditions were applied as determined previously [11],
i.e. the reactions were performed in 1 mL of DMF at 140 8C up to
30 min using 10 mg of precursor. RCYs were measured at different
reaction times as presented in Table 1 and Fig. 1.
For checking possible carrier effects (see Method 2, analytical
assay, Section 4) yield curves of n.c.a. and c.a. nucleophilic
radiofluorination were determined by TLC at different time points
as presented in Fig. 2. For both reactions similar trends were
observed indicating both the n.c.a. and c.a. substitution to proceed
by principally the same substitution mechanism, but labelling
yields were partly different as it is often observed in radiochemical
reactions due to concentrations being different from those of
typical organic reactions.
The electronic influence of methoxy groups on aromatic rings is
described by a combination of the electron-donating +M-effect and
the electron-withdrawing ꢀI-effect. With respect to the ortho- and
para-positions relative to the methoxy-substituted carbon atom
the mesomeric effect dominates, and, therefore, renders nucleo-
philic attack more difficult. Indeed, both, the 3-methoxy-(d-NO2)
and the 5-methoxy-substituted derivative (b-NO2) gave lower
yields in comparison to the standard model compound a0-NO2. The
lowest yield was observed when both positions were methoxy-
lated (h-NO2). In contrast, introducing CH3O in position 4 (c-NO2)
and 6 (a-NO2) (both meta relative to NO2) did not result in a larger
change of the radiochemical yields. For meta positions the
electron-withdrawing ꢀI-effect becomes more important, which
can result in an increasing reactivity of the nitro-substituted
carbon towards nucleophilic attack. On the other hand, in this
substitution pattern the methoxy group suppresses to some
extend the activating effect of the ortho-formyl substituent by +M-
donation.
2.2. Effect of methoxy groups on nucleophilic aromatic substitution
The systematic study on the relationship between RCY and an
increasing number of methoxy substituents in o-nitrobenzalde-
hydes was started from 2-nitrobenzaldehyde (a0). In this case, a
radiochemical yield of 84% was observed. The derivatives obtained
by introduction of an increasing number of methoxy groups to the
aromatic ring of a0-NO2 can be roughly devided in three groups
with respect to the incorporation yield of fluorine-18.
Remarkably high yields close to or even above the value for a0-
NO2 were noted for compounds c-NO2 (89%), a-NO2 (79%), g-NO2
(87%), i-NO2 (86%) and j-NO2 (82%) (Table 1). These compounds
Table 1
13C NMR chemical shifts of the nitro-substituted carbon (ArC-NO2) and radiochemical yields (RCY; as mean ꢁ SD) of methoxy-substituted 2-nitrobenzaldehydes (n ꢂ 3).
Nr.
ArC-NO2 (ppm)
k* (minꢀ1
)
RCY (%)
1 min
3 min
7 min
89 ꢁ 2.4
10 min
20 min
87 ꢁ 3.7
30 min
c-NO2
a0-NO2
a-NO2
j-NO2
g-NO2
i-NO2
b-NO2
e-NO2
k-NO2
d-NO2
f-NO2
h-NO2
151.0
149.1
148.1
143.8
143.6
142.3
141.8
139.0
137.8
137.2
137.1
131.9
2.00
1.62
1.19
1.09
1.32
1.11
0.62
0.09
0.48
0.61
0.65
0.01
86.5 ꢁ 0.6
80.3 ꢁ 1.6
69.5 ꢁ 0.4
66.3 ꢁ 2.0
73.2 ꢁ 1.7
67.0 ꢁ 6.4
46.2 ꢁ 1.2
8.9 ꢁ 0.5
88.3 ꢁ 1.8
84.1 ꢁ 0.8
72.3 ꢁ 1.1
73.8 ꢁ 6.0
81.5 ꢁ 1.0
72.9 ꢁ 4.7
53.3 ꢁ 1.2
15.8 ꢁ 1.8
43.7 ꢁ 1.6
64.3 ꢁ 2.1
65.2 ꢁ 3.6
6.0 ꢁ 0.9
86.3 ꢁ 4.2
81.9 ꢁ 4.4
78.5 ꢁ 3.5
81.5 ꢁ 2.4
86.8 ꢁ 1.8
85.9 ꢁ 4.2
53.1 ꢁ 5.6
22.3 ꢁ 2.1
45.3 ꢁ 3.5
67.4 ꢁ 5.0
67.9 ꢁ 4.1
13.0 ꢁ 1.5
86.6 ꢁ 3.3
82.3 ꢁ 3.6
78.7 ꢁ 3.1
65.9 ꢁ 4.1
79.7 ꢁ 1.6
84.6 ꢁ 3.6
53.6 ꢁ 4
84.3 ꢁ 0.4
77.6 ꢁ 1.4
80.4 ꢁ 4.3
85.2 ꢁ 2.2
70.7 ꢁ 9.7
57.4 ꢁ 1.0
20.6 ꢁ 2.3
47.6 ꢁ 4.5
70.1 ꢁ 3.1
69.6 ꢁ 1.6
11.2 ꢁ 0.2
82 ꢁ 4.1
78.9 ꢁ 4.0
74.0 ꢁ 4.8
81.7 ꢁ 1.0
84.8 ꢁ 4.7
54.9 ꢁ 4.6
21.0 ꢁ 1.3
44.2 ꢁ 5.8
67.4 ꢁ 2.8
67.0 ꢁ 2.5
11.8 ꢁ 1.6
18.9 ꢁ 1.0
41 ꢁ 6.9
38.0 ꢁ 2.0
45.7 ꢁ 3.1
47.8 ꢁ 2
66.5 ꢁ 4.7
60.8 ꢁ 2.4
10.4 ꢁ 0.7
1.4 ꢁ 0.2

218
B. Shen et al. / Journal of Fluorine Chemistry 130 (2009) 216–224
Fig. 1. Maximum radiochemical yields (RCYMAX) for 18F-labelling of ortho nitro- and bromobenzaldehyde derivatives (LG: leaving group), n ꢂ 3.
The combined action of both effects may be the reason why
such derivatives give more or less the same radiochemical yield as
the reference compound a0-NO2.
The overall effect of the presence of both o/p-and m-methoxy
groups is difficult to predict, because steric effects become
superimposed on the electronic effects [7]. In particular, steric
hindrance works against the +M-effect as the orbital overlap
Only the time needed for the completion of the reaction should
vary according to the electrophilicity of the precursor. In reality,
however, side reactions and product decomposition contribute to
the outcome of a labelling reaction and decrease the yield. These
side effects should become more influencial the less reactive the
precursor is. This may explain why the maximum radiochemical
yields correlate to some extent with the 13C-chemical shift values
between the lone pair at the oxygen atom and the aromatic
p
-
(
d
) as reactivity parameter.
system becomes disturbed. The total effect of all substituents and
the electron density of the carbon atom bearing the leaving group
may be derived from the 13C-chemical shift of this carbon position.
The 13C-chemical shift of the nitro-substituted C-atom of the o-
nitrobenzaldehydes studied (Table 1) fell in the range between
At the beginning of a labelling experiment any competing side
effect lowering the yield should be less disturbing. In our
experiments the course of the radiochemical yields was deter-
mined up to 30 min (Table 1 and Fig. 2). The strongest yield
increase was generally observed within the first minute. Provided
that the 13C-chemical shifts are reliable measures of the
d
= 132 ppm (h-NO2) and 151 ppm (c-NO2). In similar reaction
series Rengan et al. [12] and Ding et al. [7] observed an
approximately linear correlation of the RCY with the deshielding
of the leaving group substituted carbon atom. An analogous
treatment of our results (Fig. 3) showed a trend but not a good fit
(R2 = 0.62). The deshielding of the nitro-substituted carbon atom is
considered as a measure of its electrophilicity. In principle, such a
reactivity parameter cannot be expected to correlate with
maximum radiochemical yields obtained after different reaction
times. In radiofluorination experiments the [precursor]/[18F]Fꢀ
ratios applied are very high. If no product decomposition and side
reactions consuming or deactivating 18Fꢀ intervene, this should
result in a complete incorporation of 18Fꢀ into the reaction product.
electrophilicities, the NMR data should give a satisfactory
correlation with the rate constants in this reaction interval.
Because only the measurements after 1 min are available, the
(pseudo)monomolecular rate constants (k*) derived from the
radiochemical yields after 1 min have to be considered with
caution. Despite of the uncertainty of the k* values a satisfactory
correlation (R2 = 0.89) resulted for the plot k* vs.
d (Fig. 4).
As outlined before, with respect to the SNAr reaction more
reactive o-nitrobenzaldehydes derivatives should have better
chances to escape unwanted side reactions and, therefore, attain
higher yields in the radiofluorination steps. The precursor
molecules 2,3,4-trimethoxy-6-nitrobenzaldehyde (j-NO2) and

B. Shen et al. / Journal of Fluorine Chemistry 130 (2009) 216–224
219
Fig. 2. Radiochemical yields (RCY) of 18F-labelling of methoxy-substituted 2-nitro and 2-bromobenzaldehydes in dependence on the reaction time. Labelling conditions:
precursor (10 mg), DMF (1 mL), Kryptofix1222 (15 mg), 3.5% K2CO3 (100
L); 140 8C; carrier KF (50 g).
m
m
2,3,4,5-tetramethoxy-6-nitrobenzaldehyde (k-NO2) are a good
illustration for that assumption. Compound j-NO2 (ArC-
NO2) = 144 ppm) was about two times as reactive as compound
k-NO2 (ArC-NO2) = 138 ppm) as the corresponding maximum
radiochemical yields were 81.5% (j-NO2) and 47.6% (k-NO2). It
should be emphasized that the yield obtained from the fully
substituted derivative k-NO2 is still acceptable if discussing from a
practical point of view. It is even higher than expected when
compared with 3,5-dimethoxy-2-nitrobenzaldehyde (h-NO2)
From the data compiled in Table 1 it can be seen that the
precursors with chemical shifts of the nitro-substituted carbon
atom near to the corresponding value found in 2-nitrobenzalde-
(
d
(
d
hyde (a0-NO2,
chemical yields, e.g. c-NO2
(ArC-NO2) = 144 ppm), 81.5%.
In these cases, because of the high reactivities of the precursors
d
(ArC-NO2) = 149 ppm) gave high maximum radio-
(d
(ArC-NO2) = 151 ppm), 89%; j-NO2
(d
towards the SNAr reaction, side reactions do not play an important
role and high yields (ꢂ80%) are found in the radiofluorination.
(13%,
d
(ArC-NO2) = 132 ppm).
However, for the less reactive precursors exhibiting d-values in the
Fig. 3. Correlation of maximum radiochemical yields (RCY) with 13C NMR chemical
shift values ( ). Compound e-NO2 was omitted because a radioactive by-product
(15%) of unknown structure was detected.
Fig. 4. Correlation between k* and 13C NMR chemical shift values (
d). Compound e-
NO2 was omitted because a radioactive by-product (15%) of unknown structure was
d
detected.

220
B. Shen et al. / Journal of Fluorine Chemistry 130 (2009) 216–224
Fig. 5. Proposed mechanism for the demethylation of the methoxy group as a possible side reaction of methoxylated o-nitrobenzaldehydes.
range of 132–137 ppm a decrease in the radiofluorination yields
(13–70%) is observed. Because the yields have a maximum within
the reaction interval of 30 min, the reduced yields cannot be
explained by insufficiently long reaction times. Obviously, with
compounds less prone to the SNAr reaction other processes
consuming 18Fꢀ become important.
In the course of our experiments with methoxylated o-
nitrobenzaldehydes we indeed obtained indication of possible
side reactions. HPLC/APCI-MS measurements revealed the forma-
tion of phenolates which could be formed in an SN2-like
demethylation by attack of Fꢀ at the methoxy group (Fig. 5).
In all cases a HPLC-peak eluting faster than the starting
compound with a mass corresponding to the [MꢀH]ꢀ ion of the
phenolic demethylation product could be detected. For 2,3,4-
trimethoxy-6-nitrobenzaldehyde (j-NO2) the phenolic product
was isolated and characterised.
3. Conclusions
For the nucleophilic aromatic fluorination of methoxylated o-
nitrobenzaldehydes by [18F]fluoride ion a correlation was found
between the downfield shift of the nitro-substituted aromatic
carbon atom in the 13C NMR spectrum and the reaction rate at the
beginning of the reaction. Side reactions and product decomposi-
tion decreased the radiochemical yields to some extent the longer
the reaction proceeded and became relevant in the case of less
reactive precursor molecules.
4-Methoxy-2-nitrobenzaldehyde (c-NO2), 4,5-dimethoxy-2-
nitrobenzaldehyde (g-NO2) and 3,4-dimethoxy-2-nitrobenzalde-
hyde (i-NO2) are potential candidates for the synthesis of 18F-
labelled tyrosine or DOPA, respectively, when the aldehydic
function is utilised for connecting the aromatic substructure with
the amino acid residue. All the three o-nitrobenzaldehyde
precursors gave high maximum radiochemical yields (>85%).
Even the threefold and fourfold methoxylated precursors j-NO2
and k-NO2 reacted in remarkably good yields to the corresponding
18F-labelled products (82% and 48%, respectively).
The proximity of a nitro and an aldehyde group in the precursor
molecules could give rise to an intramolecular redox process
resulting in o-nitrosobenzoic acids. If o-nitrosobenzoic acids are
formed as side products, they deactivate the nucleophilic power of
[
18F]fluoride ion by interacting with the acidic hydrogens
potentially ending up in strong hydrogen bond formations. Such
intramolecular redox-processes are well known in photochem-
istry [13] and electron-impact induced fragmentation (mass
spectrometric ortho-effect) [14]. Under these reaction conditions
the intramolecular process is initiated by abstraction of the
aldehydic hydrogen by an unpaired electron at a nitro oxygen
4. Experimental
4.1. General
For performing the labelling reactions, as solvent DMF (stored
over molecular sieve) was purchased from Fluka (Germany).
Acetonitrile (for DNA synthesis) and Kryptofix 222 were obtained
from Merck (Darmstadt, Germany). For the middle pressure liquid
chromatography system (MPLC, Bu¨chi, Switzerland) silica gel 60
(0.040–0.063 mm, Merck) or reverse phase material (POPLYGO-
PREP 60-50, C18, MN, Germany) was used, eluents were mixtures
of petroleum ether (60/90 8C) and ethyl acetate or water and
acetonitrile. Precursors and reference standards were charac-
terised by their melting point (Gallenkamp MPG 350, Germany,
values uncorrected), IR (Spectrum One FT-ATR-IR, PerkinElmer,
Boston, USA), MS (Finnigan-MAT TSQ 70, Bremen, Germany) and
NMR (Bruker Avance 400, Rheinstetten). For NMR measurements,
as internal standards the deuterated solvents were used (DMSO-
formed by
n
p* excitation and ionisation, respectively. An
analogous thermal process could be started by intramolecular
nucleophilic addition of a nitro oxygen atom to the neighbouring
carbonyl group (Fig. 6).
To test a thermal intramolecular redox reaction as a possible
side reaction intervening in our labelling experiments, o-nitro-
benzaldehyde was heated under conditions identical with the
labelling experiments except that no fluoride ion was present. On
heating in DMF in the presence of the K2CO3–crown ether complex
a bluish colour was built up. Subsequent HPLC-APCI-MS analysis
indicated the formation of an isomeric compound eluting faster
than the starting material. The mass spectrum (negative mode)
exhibits strong [MꢀH]ꢀ(m/z 150) and [MꢀHꢀCO2]ꢀ(m/z 106)
ions. In contrast, the negative ion mass spectrum of the
starting compound is dominated by the radical anion Mꢃꢀ at
m/z 151. The blue colour (typical for C-nitroso compounds) and
the mass spectral data are in accordance with the formation of
o-nitrosobenzoic acid.
d6:
d
d
= 2.49 in 1H and
d
= 39.5 in 13C; CDCl3:
d
= 7.25 in 1H and
= 77.0 in 13C). The chemical shift
d
in ppm was referred to the
internal standard. All radiochemical yields given in this work
represent an average of 3–5 experiments.
4.2. Precursors and reference standards
As precursors and reference standards, the following com-
pounds were of the highest purity available from either Sigma-
Aldrich, Fluka, ABCR, Alfa Aesar, AVOCADO or Fluorochem and
were used as received (authenticity of the fluoro compounds used
as standards was confirmed by NMR and MS): 3-methoxy-2-
nitrobenzaldehyde (d-NO2, >97%), 2-fluoro-3-methoxybenzalde-
hyde (d-F, 97%), 2-fluoro-4-methoxybenzaldehyde (c-F, 98%),
Fig. 6. Intramolecular redox-process in ortho-nitrobenzaldehyde by intramolecular
nucleophilic addition of a nitro oxygen atom to the neighbouring carbonyl group.

B. Shen et al. / Journal of Fluorine Chemistry 130 (2009) 216–224
221
2-fluoro-5-methoxybenzaldehyde (b-F, 98%), 2-fluoro-6-methox-
ybenzaldehyde (a-F, 98%), 2-nitrobenzaldehyde (a0-NO2, 98%), 2-
fluorobenzaldehyde (a0-F, 97%), 2-nitro-4,5-dimethoxybenzalde-
hyde (g-NO2, 96%), 2-fluoro-4,5-dimethoxybenzaldehyde (g-F,
98%).
a-NO2, b-NO2, c-NO2, e-NO2, e-Br and e-F were obtained
according to literature [10,11,15]. All other precursors in Fig. 1
were synthesized as described below.
dimethoxy-benzaldehyde) at 0 8C within 5 min and then this
solution was stirred at room temperature for 1 h. The resulting
mixture was poured into water (50 mL) and was extracted with
diethyl ether (2 ꢄ 50 mL). The combined organic layers were
washed with satd. NaHCO3 (50 mL) and brine (50 mL). Then the
organic layer was separated, dried over Na2SO4 and the solvent was
evaporated. 0.53 g f-NO2 (42%) was obtained as yellow crystals
after separation on the MPLC (silica gel, petroleum ether:ethyl
acetate, 2:1, v:v). mp 172 8C (reported [19] 171–172 8C). 1H NMR
4.2.1. 2-Bromo-5-methoxybenzaldehyde (b-Br)
(DMSO-d6, 400 MHz): d = 10.24 (s, 1H, –CHO), 7.67 (d,
3
To
a
stirred solution of 3-methoxybenzaldehyde (3.35 g,
3JH,H = 9.3 Hz, 1H, Harom), 7.45 (d, JH,H = 9.3 Hz, 1H, Harom), 3.94
24.6 mmol) in dry CHCl3 (200 mL) bromine (1.89 mL, 3.69 mmol)
was introduced slowly during 20 min. The reaction solution was
stirred at room temperature for 0.5 h. Then the resulting solution
was poured into diethyl ether (100 mL) and the mixture was
washed with satd. Na2S2O3 (2 ꢄ 50 mL) and water (50 mL). The
organic layer was separated and dried over Na2SO4. After
evaporation of the solvent, the crude product was purified on
MPLC (silica gel, petroleum ether:ethyl acetate, 1:1, v:v). 4.37 g b-
Br were obtained (82%). mp 77 8C (reported [16] 78–80 8C). 1H
(s, 3H, –OCH3), 3.85 (s, 3H, –OCH3) ppm. 13C NMR (DMSO-d6,
100 MHz):
d = 57.2, 57.3, 114.9, 116.0, 121.7, 137.0, 143.7, 155.1,
186.7 ppm. MS-EI (70 eV): m/z (%) = 211 (18) [M]ꢃ+, C9H9NO5.
4.2.5. 3,4-Dimethoxy-2-nitrobenzaldehyde (i-NO2)
Into the stirred solution of 4-formyl-2-methoxy-3-nitrophenyl
benzenesulfonate [20] (2.80 g, 8.3 mmol) in ethanol (70 mL),
aqueous NaOH solution (2 mL, 1 g/mL) was added. Then this
solution was refluxed for 0.5 h and ethanol was evaporated. To the
residue water (100 mL) was added and pH was adjusted to 6 by 2N
HCl. The resulting mixture was extracted with diethyl ether
(200 mL) and the organic layer was washed with satd. NaHCO3
(50 mL), water (50 mL) and brine (50 mL) respectively, then dried
over Na2SO4 and concentrated to afford 0.86 g 4-hydroxy-3-
methoxy-2-nitrobenzaldehyde crude product, which was used
directly in further synthetic procedure.
NMR (CDCl3, 400 MHz): d
= 10.30 (s, 1H, –CHO), 7.50 (d, 3J = 8.6 Hz,
1H, Harom), 7.39 (d, 4JH,H = 3.3 Hz, 1H, Harom), 7.02 (dd, 3JH,H = 8.7 Hz,
4JH,H = 3.1 Hz, 1H, Harom), 3.83 (s, 3H, –OCH3) ppm. 13C NMR (CDCl3,
100 MHz):
d = 55.7, 112.6, 117.9, 123.1, 133.9, 134.5, 159.2,
191.7 ppm. MS-EI (70 eV): m/z (%) = 216 (95) [M]ꢃ+, C8H781BrO2;
214 (100) [M]ꢃ+, C8H779BrO2.
4.2.2. 3,5-Dimethoxy-2-nitrobenzaldehyde (h-NO2)
4-Hydroxy-3-methoxy-2-nitrobenzaldehyde (0.86 g, 4.4 mmol),
K2CO3 (0.72 g, 5.2 mmol) and MeI (0.32 mL, 5.2 mmol) were stirred
in DMF (15 mL) at room temperature for 2 h. The reaction mixture
was poured into water and extracted with diethyl ether (3 ꢄ 50 mL).
The combined organic layers were washed with water (100 mL) and
dried over Na2SO4. After evaporation of the solvent, the residue was
purified by MPLC (silica gel, petroleum ether:ethyl acetate, 1:1, v:v)
to yield 0.81 g i-NO2 was obtained (46%, based on 4-formyl-2-
methoxy-3-nitrophenyl benzenesulfonate). mp 60–62 8C (reported
3,5-Dimethoxybenzaldehyde (2.0 g, 12 mmol) was portion-wise
addedto90%HNO3 (10 Meq.basedon3,5-dimethoxybenzaldehyde)
while stirring at 0 8C. After this addition, the reaction solution was
stirredat 0 8C and monitored byTLC. 1 h laterthe starting compound
was completely consumed. The resulting mixture was poured into
water (100 mL) and extracted with diethyl ether (100 mL). The
organic layer was washed with satd. NaHCO3 (2 ꢄ 50 mL) and brine
(50 mL).Thentheorganiclayerwasseparated,driedoverNa2SO4 and
evaporated. 0.41 gh-NO2 (17%) was obtained asyellow crystals after
separation on the MPLC (silica gel, petroleum ether:ethyl acetate,
1:1, v:v). mp 104–106 8C (reported [17] 104–106 8C). 1H NMR
[21] 58–61 8C). 1H NMR (DMSO-d6, 400 MHz):
d = 9.79 (s, 1H, –
CHO), 7.90 (d, JH,H = 8.6 Hz, 1H, Harom), 7.51 (d, JH,H = 8.6 Hz, 1H,
Harom), 4.01 (s, 3H, –OCH3), 3.83 (s, 3H, –OCH3) ppm. 13C NMR
3
3
(DMSO-d6, 400 MHz):
1H, Harom), 7.14 (d, 4JH,H = 2.3 Hz, 1H, Harom), 3.92 (s, 3H, –OCH3), 3.92
(s, 3H, –OCH3) ppm. 13C NMR (DMSO-d6, 100 MHz):
= 56.4, 57.2,
d
= 9.88 (s, 1H, –CHO), 7.18 (d, 4JH,H = 2.5 Hz,
(DMSO-d6, 100 MHz): d = 57.2, 62.0, 114.2, 119.5, 131.3, 140.2,
142.3, 158.1, 188.3 ppm. MS-EI (70 eV): m/z (%) = 211 (18) [M]ꢃ+
,
d
C9H9NO5.
104.9, 109.1, 129.2, 131.9, 152.3, 161.6, 189.4 ppm. MS-EI (70 eV):
m/z (%) = 211 (48) [M]ꢃ+, C9H9NO5.
4.2.6. 2,3,4-Trimethoxy-6-nitrobenzaldehyde (j-NO2)
The nitration of 2,3,4,-trimethoxybenzaldehyde is similar to the
nitration of 2,5 dimethoxybenzaldehyde.
4.2.3. 2-Bromo-3,5-dimethoxybenzaldehyde (h-Br)
To a stirred solution of 3,5-dimethoxybenzaldehyde (1.5 g,
9 mmol) in dry CHCl3 (90 mL) the bromine (0.6 mL, 11.7 mmol)
was introduced slowly over 20 min. This reaction solution was
stirred at room temperature for 0.5 h. Then the resulting solution
was poured into diethyl ether (100 mL) and the mixturewas washed
with satd. Na2S2O3 (2 ꢄ 50 mL) and water (50 mL). The organic layer
was separated and dried over Na2SO4. After evaporation of the
solvent, the crude product was purified on MPLC (silica gel,
petroleum ether:ethyl acetate, 5:1, v:v). 1.86 g h-Br was obtained
(84%). mp 114–115 8C (reported [18] 115–116 8C). 1H NMR (CDCl3,
2,3,4-trimethoxybenzaldehyde (2 g, 10.2 mmol) and a mixture
of 100% HNO3 and 100% AcOH (10 mL, 1:1, v:v) were stirred at 0 8C
for 10 min. After extraction with diethyl ether and purification on
the MPLC (silica gel, petroleum ether:ethyl acetate, 3:1, v:v), 1.62 g
j-NO2 was obtained as yellow crystals (66%) mp 83–84 8C (reported
[22] 80–82 8C). 1H NMR (DMSO-d6, 400 MHz):
CHO), 7.51 (s, 1H, Harom), 3.94 (s, 3H, –OCH3), 3.93 (s, 3H, –OCH3),
3.86 (s, 3H, –OCH3) ppm. 13C NMR (DMSO-d6, 100 MHz):
= 57.4,
d = 10.12 (s, 1H, –
d
60.9, 62.7, 104.2, 117.3, 143.7, 144.7, 153.8, 156.7, 187.4 ppm. MS-
EI (70 eV): m/z (%) = 241 (34) [M]ꢃ+, C10H11NO6.
250 MHz):
6.93 (d, 4JH,H = 2.7 Hz, 1H, Harom), 3.89 (s, 3H, –OCH3), 3.82 (s, 3H, –
OCH3) ppm. 13C NMR (CDCl3, 60 MHz):
= 55.8, 56.8, 104.4, 105.7,
d
= 10.24 (s, 1H, –CHO), 6.97 (d, 4JH,H = 2.8 Hz, 1H, Harom),
4.2.7. 6-Bromo-2,3,4-trimethoxybenzaldehyde (j-Br)
d
j-Br was synthesized from 5-bromo-2,3-dimethoxybenzalde-
hyde [20] by the following steps. Baeyer Villiger reaction: 30%
H2O2 (4 mL) and formic acid (11 mL) with catalytic amount of
concentrated H2SO4 were stirred at room temperature for 1 h.
The solution of 5-bromo-2,3-dimethoxybenzaldehyde (4.05 g,
16.5 mmol) in formic acid (25 mL) was added dropwise into the
solution at 0 8C, then the mixture was kept stirring at 0 8C
107.0, 134.3, 156.9, 159.7, 191.7 ppm. MS-EI (70 eV): m/z (%) = 246
(97) [M]ꢃ+, C9H981BrO3; 244 (100) [M]ꢃ+, C9H979BrO3.
4.2.4. 3,6-Dimethoxy-2-nitrobenzaldehyde (f-NO2)
To a solution of 2,5-dimethoxybenzaldehyde (1.0 g, 6 mmol) in
AcOH (5 mL) was added 70% HNO3 (1–1.5 M eq. based on 2,5-

222
B. Shen et al. / Journal of Fluorine Chemistry 130 (2009) 216–224
overnight. TLC analysis indicated complete conversion of the
starting material. The reaction was quenched with water (50 mL)
and extracted with diethyl ether (200 mL). The organic layer was
washed with satd. NaHCO3 (100 mL) and brine (50 mL). Then the
organic layer was separated, dried over Na2SO4 and evaporated to
afford 5.5 g 5-bromo-2,3-dimethoxyphenol crude product as red
oil which was used directly in the next synthesis step.
5-Bromo-2,3-dimethoxyphenol was formylated by Duff reac-
tion. 5-Bromo-2,3-dimethoxyphenol (4.0 g, 17.1 mmol) and hex-
amethylenetetramine (2.3 g, 17.2 mmol) were refluxed at 130 8C in
TFA (40 mL) for 4 h. The reaction solution was cooled down and
quenched by adding 6N HCl (120 mL). After 20 min stirring, the
resulting solution was extracted with DCM (2 ꢄ 100 mL). The
combined organic layers were washed with satd. NaHCO3 solution
(2 ꢄ 50 mL), water (50 mL) and brine (50 mL). After evaporation of
the solvent, 1.55 g crude product 6-bromo-2-hydroxy-3,4-
dimethoxybenzaldehyde was afforded and used directly in the
next synthesis step.
d = 55.9, 60.8, 61.1, 62.8, 103.7, 123.7, 146.5, 148.8, 149.6, 151.5,
188.4 ppm. MS-EI (70 eV): m/z (%) = 226 (100) [M]ꢃ+, C11H14O5.
4.2.10. 2,3,4,5-Tetramethoxy-6-nitrobenzaldehyde (k-NO2)
2,3,4,5-Tetramethoxybenzaldehyde (0.5 g, 2.2 mmol) and 70%
HNO3 (1.25 M eq. based on 2,3,4,5-tetramethoxybenzaldehyde)
were stirred in acetic acid (5 mL) at 0 8C for 10 min, then the ice
bath was removed and the reaction solution was stirred for 1 h at
room temperature. After extraction with diethyl ether and
purification on MPLC (silica gel, petroleum ether:ethyl acetate,
3:1, v:v), 0.23 g k-NO2 was obtained as yellow crystals (39%, based
on 2,3,4,5-tetramethoxybenzaldehyde) mp 67–70 8C. 1H NMR
(DMSO-d6, 400 MHz):
3.98 (s, 3H, –OCH3), 3.92 (s, 3H, –OCH3), 3.81 (s, 3H, –OCH3) ppm.
13C NMR (DMSO-d6, 100 MHz):
= 61.4, 61.5, 62.6, 63.0, 115.0,
d = 10.09 (s, 1H, –CHO), 4.03 (s, 3H, –OCH3),
d
137.7, 140.9, 148.0, 152.7, 153.1, 186.2 ppm. HRMS (EI): m/z [M]ꢃ+
calculated for C11H13NO7: 271.069167; found 271.07104 MS-EI
(70 eV): m/z (%) = 271 (31) [M]+, C11H13NO7; 254 (100) [MꢀOH]+,
6-Bromo-2-hydroxy-3,4-dimethoxybenzaldehyde (1.55 g,
6
C
11H12NO6; 224 (30) [MꢀOHꢀNO]+, C11H12O5; 211 (82), 195 (42),
mmol), K2CO3 (0.8 g, 6 mmol) and MeI (0.36 mL, 6 mmol) were
stirred in DMF (20 mL) at room temperature for 3 h. After
extraction with diethyl ether and purification on the MPLC (silica
gel, petroleum ether:ethyl acetate, 2:1, v:v), 0.41 g j-Br as yellow
oil was obtained (13%, based on 5-bromo-2,3-dimethoxybenzal-
181 (46), 167 (27), 152 (33), 137 (35), 127 (19), 125 (16), 95 (10),
n
93 (8), 65 (5) [C5H5]+, 53 (10). IR: , cmꢀ1 (T%) = 2998 (87%), 2952
(77%), 2876 (80%), 1692 (26%), 1591 (60%), 1537 (39%), 1465 (28%),
1444 (48%), 1404 (32%), 1383 (29%), 1364 (29%), 1336 (17%), 1267
(44%), 1194 (44%), 1125 (34%), 1082 (22%), 1054 (21%), 1011 (17%),
963 (25%), 943 (30%), 901 (47%), 878 (44%), 823 (50%), 787 (54%),
745 (35%), 684 (59%).
dehyde). 1H NMR (DMSO-d6, 400 MHz):
7.19 (s, 1H, Harom), 3.90 (s, 3H, –OCH3), 3.86 (s, 3H, –OCH3), 3.76 (s,
3H, –OCH3) ppm. 13C NMR (DMSO-d6, 100 MHz):
= 56.7, 60.6,
62.4, 113.9, 117.6, 120.9, 141.4, 156.5, 157.8, 188.8 ppm. MS-EI
(70 eV): m/z (%) = 276 (93) [M]ꢃ+, C10H1181BrO4; 274 (100) [M]ꢃ+
10H1179BrO4.
d = 10.09 (s, 1H, –CHO),
d
4.2.11. 2-Bromo-3,4,5,6-tetramethoxybenzaldehyde (k-Br)
,
Into a solution of 2,3,4,5-tetramethoxybenzaldehyde (0.89 g,
4 mmol) in trifluoroacetic acid (5 mL) N-bromosuccinimide (NBS,
0.84 g, 4.8 mmol) was added over 15 min at 0 8C. The reaction was
monitoredbyTLC.After0.5 hthestartingcompoundwascompletely
reacted. The resulting mixture was poured into water and extracted
withdiethylether(100 mL).Thecombinedorganiclayerwaswashed
with water (100 mL) and dried over Na2SO4. After evaporation of
solvent, the residue was purified on MPLC (silica gel, petroleum
ether:ethyl acetate, 5:1, v:v) to yield 0.91 g k-Br as yellow oil (75%).
C
4.2.8. 2-Hydroxy-3,4,5-trimethoxybenzaldehyde
30% H2O2 (15 mL) was added to a stirred solution of 2,3,4-
trimethoxybenzaldehyde (7.50 g, 38.4 mmol) in methanol (75 mL)
with a catalytic amount of concentrated H2SO4 (0.2 mL). After 2 h
TLC analysis indicated complete conversion of the starting
material. The reaction was quenched with water (50 mL) and
extracted with diethyl ether (200 mL). The organic layer was
washed with satd. NaHCO3 (50 mL) and brine (50 mL). The organic
layer was separated, dried over Na2SO4, evaporated and purified on
MPLC (silica gel, petroleum ether:ethyl acetate, 2:1, v:v) to afford
4.44 g 2,3,4-trimethoxyphenol and used directly in the next
synthesis step.
1H NMR (DMSO-d6, 400 MHz):
OCH3), 3.85 (s, 3H, –OCH3), 3.84 (s, 3H, –OCH3), 3.75 (s, 3H, –OCH3)
ppm. 13CNMR(DMSO-d6, 100 MHz):
= 60.8, 61.1, 61.3, 62.5, 111.7,
d = 10.13 (s, 1H, –CHO), 3.96 (s, 3H, –
d
123.6, 146.2, 147.1, 151.8, 152.8, 189.5 ppm. MS-EI (70 eV): m/z
(%) = 306 (95) [M]ꢃ+, C11H1381BrO5; 304 (100) [M]ꢃ+, C11H1379BrO5.
2,3,4-Trimethoxyphenol (3.40 g, 18.5 mmol) and hexamethy-
lentetramine (3.15 g, 22.5 mmol) were refluxed at 130 8C in TFA
(20 mL) for 24 h. After extraction with DCM, the crude product was
chromatographied on the MPLC (silica gel, petroleum ether:ethyl
acetate, 2:1, v:v) to yield 2.0 g 2-hydroxy-3,4,5-trimethoxyben-
zaldehyde as colorless oil (51%, based on 2,3,4-trimethoxyphenol).
4.2.12. 6-Fluoro-2,3,4-trimethoxybenzaldehyde (j-F)
The mixture of KF (0.24 g, 8.2 mmol, spray dried) and Kryptofix
222 (1.16 g, 6.3 mmol) was heated at 140 8C under Argon for
20 min, then j-NO2 0.50 g (4.2 mmol) in DMF (20 mL) was added
and the reaction solution was heated for 15 min. The reaction
solution was poured into water (25 mL) and this mixture was
extracted with diethyl ether (2 ꢄ 25 mL). The combined organic
layer was washed with 1N NaOH (25 mL) (this basic washing
solution was kept for isolation of phenolic by-product), water
(25 mL) and brine (25 mL), then was dried over Na2SO4, filtered
and evaporated to afford 0.22 g crude product. After purification on
MPLC (reverse phase material, CH3CN:water, 30:70, v:v) 25 mg
product j-F was obtained. This 25 mg product was purified again
1H NMR (DMSO-d6, 400 MHz):
Harom), 3.87 (s, 3H, –OCH3), 3.77 (s, 3H, –OCH3), 3.76 (s, 3H, –OCH3)
ppm. 13C NMR (DMSO-d6, 100 MHz):
= 55.9, 60.7, 60.9, 105.6,
d = 10.12 (s, 1H, –CHO), 7.01 (s, 1H,
d
117.1, 141.4, 146.0, 149.0, 150.0, 190.6 ppm. MS-EI (70 eV): m/z
(%) = 212 (100) [M]ꢃ+, C10H12O5.
4.2.9. 2,3,4,5-Tetramethoxybenzaldehyde
2-Hydroxy-3,4,5-trimethoxybenzaldehyde (1.70 g, 8 mmol),
K2CO3 (1.24 g, 9 mmol) and MeI (0.56 mL, 9 mmol) were stirred in
DMF (30 mL) at room temperature overnight. After extraction with
diethyl ether and purification on MPLC (silica gel, petroleum
ether:ethyl acetate, 2:1, v:v), 1.45 g 2,3,4,5-tetramethoxybenzalde-
hyde was obtained (80%, based on 2-hydroxy-3,4,5-trimethoxyben-
on preparative HPLC (Luna, phenyl-hexyl 5
m
m, 150 mm ꢄ 10 mm,
Phenomenex, USA; CH3CN:water, 3:1, v:v; 5 mL/min) to yield
12 mg pure j-F (1.3%). 1H NMR (CDCl3, 400 MHz):
d
= 10.21 (s, 1H, –
CHO), 6.45 (d, 3JH,F = 8.4 Hz, 1H, Harom), 3.99 (s, 3H, –OCH3), 3.90 (s,
3H, –OCH3), 3.82 (s, 3H, –OCH3) ppm. 13C-NMR (CDCl3, 100 MHz):
d
= 56.3 (–OCH3), 61.1 (–OCH3), 62.2 (–OCH3), 96.1 (d, J = 26.3 Hz,
zaldehyde). 1H NMR (DMSO-d6, 400 MHz):
d
= 10.17 (s, 1H, –CHO),
Carom-H), 112.0 (d, 2J = 8.8 Hz, Carom-CHO), 138.4 (d, J = 3.7 Hz,
Carom-OCH3), 156.0 (d, J = 7.3 Hz, Carom-OCH3), 159.3 (d, J = 13.9 Hz,
7.02(s, 1H, Harom),3.88(s, 3H,–CH3),3.86(s, 3H, –OCH3),3.85(s, 3H,–
OCH3), 3.80 (s, 3H, –OCH3) ppm. 13C NMR (DMSO-d6, 100 MHz):
1
Carom-OCH3), 159.9 (d, JC,F = 259.8 Hz, Carom-F), 186.3 (d,

B. Shen et al. / Journal of Fluorine Chemistry 130 (2009) 216–224
223
3J = 2.2 Hz, –CHO) ppm. 19F NMR (CDCl3, 400 MHz): 117.1 (Carom-F)
ppm. HRMS (EI): m/z [M]ꢃ+ calculated for C10H11FO4: 214.064116;
osilicate USP Type 1 glass) containing 100
m
L of 3.5% aqueous
K2CO3 and 15.0 mg Kryptofix 222. The [18F]fluoride ion solution
was dried for 20 min under a mild stream of argon (ca. 2 mL/min)
at 140 8C by azeotropic distillation with acetonitrile (2 ꢄ 1 mL).
Then, precursor (10 mg) dissolved in DMF (1.0 mL) was added into
the vial containing the [Kryptofix 222] K+18Fꢀ complex. The sealed
found 214.06256 MS-EI (70 eV): m/z (%) = 214 (100) [M]ꢃ+
,
C
10H11FO4; 199 (79) [MꢀCH3]+, C9H8FO4; 181 (66)
[MꢀCH3ꢀH2O]+, C9H6FO3; 171 (16) [MꢀCH3ꢀCO]+, C8H8FO3;
153 (30) [MꢀH2OꢀCO]+, C9H9FO2; 138 (18), 113 (8), 97 (8), 83
(9), 57 (10). IR:
n
, cmꢀ1 (T%) = 2942 (95%), 1691 (82%), 1604 (76%),
vial was kept heating at 140 8C. A sample (1–5 mL) was withdrawn
1574 (91%), 1492 (88%), 1457 (88%), 1404 (92%), 1386 (87%), 1342
(83%), 1251 (84%), 1198 (87%), 1134 (80%), 1096 (85%), 1028 (91%),
989 (88%), 925 (95%), 820 (94%), 800 (92%).
for determination of the radiochemical yield at 1, 3, 7, 10, 20 and
30 min. For a number of experiments (n = 20), the amount of
radioactivity that remained adsorbed in the reaction vial after
removing the reaction solution and washing with water (3 ꢄ 3 mL)
was determined as 12 ꢁ 8% (based on the total activity in the vial
after labelling).
4.2.13. Phenolic by-product
The basic washing solution was acidified by 2N HCl (50 mL) and
extracted with diethyl ether (2 ꢄ 50 mL). The ether layer was
washed with water (50 mL) and brine (50 mL). The organic layer
was separated, dried over Na2SO4, and evaporated to afford 0.08 g
phenol by product. The position of the hydroxyl group was not
4.4.2. c.a. (carrier-added) 18F-fluorination
The activity was introduced into a 5.0 mL sealed vial containing
KF (50
mg), 3.5% aqueous K2CO3 (100 mL) and Kryptofix 222
determined. 1H NMR (DMSO-d6, 400 MHz):
CHO), 7.23 (d, J = 8.2 Hz, 1H, Harom), 3.92 (s, 3H, –OCH3), 3.77 (s, 3H,
–OCH3) ppm. 13C NMR (DMSO-d6, 100 MHz):
= 56.9 (–OCH3),
60.6 (–OCH3), 100.9 (Carom-H), 109.9 (Carom-CHO), 138.9 (Carom
NO2), 145.5 (Carom-OCH3), 154.1 (Carom-OH), 156.6 (Carom-OCH3),
188.2 (–CHO) ppm. IR:
, cmꢀ1 (T%) = 3095 (84%), 2948 (75%), 2850
d
= 10.13 (s, 1H, –
(15.0 mg), the following step was the same as for n.c.a. 18F-
fluorination.
d
-
4.5. Analytical assay
n
The radiochemical standard protocol for checking identity and
purity of 18F-labelled products in analytical radiochemistry is done
by radio-TLC and radio-HPLC (Method 1). In this work, for
compounds f, h, i and k no reference standards were synthesized,
therefore only purity could be checked by Method 1, identity was
assured by Method 2 or Method 3 (see Table 2).
(81%), 2324 (91%), 1632 (45%), 1508 (35%), 1460 (54%), 1445 (49%),
1416 (56%), 1389 (35%), 1317 (38%), 1283 (33%), 1250 (21%), 1203
(29%), 1139 (18%), 1035 (42%), 982 (42%), 961 (29%), 896 (32%), 856
(41%), 775 (23%), 728 (32%), 686 (40%).
4.3. Production of [18F]fluoride ion
Method 1: 18F-labelled products were identified and checked for
purity by both radio-TLC and radio-HPLC using the non-
radioactive reference compounds for detection of the UV signals
in comparison to the corresponding radioactive signals. For
product purity control, an aliquot of the reaction mixture was
taken at time of maximum RCY (see Table 1) and injected onto
HPLC. After separation, the product fraction was collected and
measured by means of a gamma-counter (1480 Wallac WIZARD
300, PerkinElmer, USA). The radiochemical product yield was
calculated by relating the radioactivity of the product peak with
the amount of radioactivity injected onto HPLC.
No-carrier-added (n.c.a.) [18F]fluoride ion was produced at the
PETtrace cyclotron (General Electric Healthcare, Uppsala, Sweden)
via the 18O(p,n)18F nuclear reaction by irradiating 1.5 mL of >95%
enriched [18O]water (Rotem, Israel) with 16.5 MeV protons.
4.4. Labelling
4.4.1. n.c.a. (no-carrier-added) 18F-fluorination
The activity (50–100 MBq) was introduced into a 5.0 mL sealed
vial (Supelco, graduated screw top V-Vials1, autoclavable bor-
Table 2
Rf values and retention times (Rt) of the 18F-labelled products as analysed by TLC and HPLC.
Precursor
Rf (TLC) of 18F-labelled product
Rt on HPLC
LC/MS mass peak
[
19F]fluoro standard (min)
18F-labelled product (min)
k0
a0-NO2
a-NO2
0.69
0.44
0.68
0.68
0.61
0.59
0.52
0.52
0.54
0.54
0.26
0.45
0.45
0.30
0.61
0.61
0.40
0.40
7.49
6.44
8.82
8.82
7.94
7.75
7.49
7.49
8.02a
7.00a
3.22
2.68
3.93
9.36a
M + 1, M + 15
*,+
b-NO2
b-Br+
c-NO2
d-NO2
9.37a
8.47a
3.46
3.36
3.24
8.28a
j-NO2
8.05a
M + 1
M + 1
*,+
j-Br*,+
8.05a
k-NO2
10.15a
10.16a
9.71b
8.88b
8.89b
9.48b
17.86b
17.87b
10.74b
10.78b
*,+
k-Br*,+
*
f-NO2
M + 1, M + 15
M + 1, M + 15
*,+
g-NO2
8.59
8.59
4.92
6.16
g-Br*,+
*
i-NO2
M + 1, M + 15
+
h-NO2
h-Br+
+
e-NO2
10.45
10.45
e-Br+
a
HPLC conditions A.
b
*
HPLC conditions B (see Section 4.5.2).
Method 2.
+
Method 3 (see analytical assay; Method 1 was used for all compounds in Table 2).

224
B. Shen et al. / Journal of Fluorine Chemistry 130 (2009) 216–224
Method 2: Labelling reactions were performed as described
above but in the presence of 8.6 mol KF (spray-dried) carrier
Eluent was acetonitrile/water (30/70, v/v) containing 0.1% KF (UV
detection at 254 nm).
m
(c.a. reaction). In this case, the carrier added nucleophilic
fluorination allowed analyses by LC/MS in addition to the
detection by Radio-HPLC. Thus, by finding the mass of the
19F-product within the fraction of the eluted 18F-product
characterised by the retention time, the identification was
independently assured.
Method 3: Precursors were chosen having the same chemical
structure but a different leaving group, so that they resulted in
the same 18F-labelled product, i.e. the substitution of the Br- or
NO2-group ended in the same 18F-product, which was identified
by the corresponding retention time on HPLC. In the labelling
reactions the nitro compounds (b-NO2, e-NO2, g-NO2, h-NO2, j-
NO2 and k-NO2) and the corresponding brominated compounds
(b-Br, e-Br, g-Br, h-Br, j-Br and k-Br) were studied and the
analytical data of 18F-products are shown in Table 2. The
retention times of radioactive product peaks were found to
match exactly with each other (error < 0.05 min).
4.5.3. LC/MS analysis
LC/MS measurements were performed on a quadrupole LC/MS
(Agilent Technologies, Model 6120) with an HPLC (Agilent
Technologies, Model 1200). All solvents (methanol and water)
were of HPLC-grade purity.
For identification of the product in the c.a. 18F-labelling reaction
(Method 2): A HP ODS (100 mm ꢄ 2.1 mm, 5
mm) column was
used. Eluent was methanol/water (40/60, v/v) buffered with
ammonium acetate (144 mg/L).
For the detection of by-product in the labelling reaction: A
Phenomenex Luna (100 mm ꢄ 2.0 mm, 5
mm) C18 (2) column was
used. Eluent was a gradient from 100% water to 100% methanol
(within 30 min) buffered with ammonium acetate (144 mg/L).
References
[1] P.L. Jager, W. Vaalburg, J. Pruim, E.G.E. de Vries, K.J. Langen, A. Piers, J. Nucl. Med.
42 (2001) 432–445.
4.5.1. TLC analysis
[2] T. Tierling, Juelich-Report Juel-3952, Research Centre Juelich, 2002.
[3] R.N. Krasikova, V.V. Zaitsev, S.M. Ametamey, O.F. Kuznetsova, O.S. Federova, I.K.
Mosevich, Y.N. Belokon, S. Vyskocil, S.V. Shatik, M. Nader, P.A. Schubiger, Nucl.
Med. Biol. 31 (2004) 597–603.
[4] G. Reischl, B. Shen, D. Lo¨ffler, M. Uebele, W. Ehrlichmann, H.-J. Machulla, J. Nucl.
Med. 49 (2008) 296P.
[5] L. Cai, S. Lu, V.W. Pike, Eur. J. Org. Chem. (2008) 2853–2873.
[6] C. Lemaire, S. Gillet, S. Guillouet, A. Plenevaux, J. Aerts, A. Luxen, Eur. J. Org. Chem.
(2004) 2899–2904.
[7] Y.-S. Ding, C.-Y. Shiue, J.S. Fowler, A.P. Wolf, A. Plenevaux, J. Fluorine Chem. 48
(1990) 189–205.
An aliquot of the reaction solution on a silica gel plate
(Polygram1 Silica G/UV254, 8 cm ꢄ 4 cm, Macherey&Nagel, Ger-
many) was developed with petroleum ether/ethyl acetate (3/1, v/
v). The radioactive spots were quantitatively assessed by means of
electronic autoradiography (Instant Imager, Canberra Packard,
USA). The Rf values are presented in Table 2. The size of the TLC
plate and the location of the reference standard were marked by
radioactive spots on the plate, thus
a correlation between
radioactive labelled product and non-radioactive standards was
assured.
[8] C. Lemaire, M. Guillaume, R. Cantineau, L. Christiaens, J. Nucl. Med. 31 (1990)
1247.
[9] G.N. Reddy, M. Haeberli, H.-F. Beer, A.P. Schubiger, Appl. Radiat. Isot. 44 (1993)
645.
4.5.2. HPLC analysis
¨
[10] B. Shen, D. Lo¨ffler, K.P. Zeller, M. Ubele, G. Reischl, H.J. Machulla, J. Fluorine Chem.
HPLC was applied for identification of radiolabelled product.
HPLC was carried out by means of a Hewlett-Packard Model 1050
equipped with an UV detector and a NaI(Tl)-scintillation detector
in series. Retention times of products were in agreement with the
UV peaks of the reference compounds, Rt and k0 values are
presented in Table 2. KF was added to the eluent as carrier in order
to assure radioactive [18F]fluoride ions to be eluted as a sharp peak
(k0 = 1.0).
128 (2007) 1461–1468.
[11] A. Al-Labadi, K.P. Zeller, H.-J. Machulla, Radiochim. Acta 94 (2006) 143–146.
[12] R. Rengan, P.K. Chakraborty, M.R. Kilbourn, J. Label. Compds. Radiopharm. 33
(1993) 563–572.
[13] J. Michalak, J. Gebicki, T. Baly, J. Chem. Soc. Perkin Trans. 2 (1993) 1321.
[14] F. Benoit, J.L. Holmes, Org. Mass Spectrom. 3 (1970) 993.
[15] A. Al-Labadi, K.P. Zeller, H.-J. Machulla, J. Radioanal. Nucl. Chem. 270 (2006) 313–
318.
[16] M. Rosillo, G. Dominguez, L. Casarrubios, U. Amador, J. Perez-Castells, J. Org.
Chem. 69 (2004) 2084–2093.
[17] J.Y. Chang, M.F. Yang, C.Y. Chang, C.M. Chen, C.C. Kuo, J.P. Liou, J. Med. Chem 23
(2006) 6412–6415.
[18] A.K. Sinhababu, K. Achintya, R.T. Borchardt, J. Org. Chem. 48 (1983) 2356–2360.
[19] P. Cotelle, J.-P. Catteau, Synth. Commun. 22 (1992) 2071–2076.
[20] A. Al-Labadi, Ph.D. Thesis, University Tu¨ bingen, 2006.
[21] J.L. Neumeyer, B.R. Neustadt, K.H. Oh, K.K. Weinhardt, C.B. Boyce, F.J. Rosenberg,
D.G. Teiger, J. Med. Chem. 16 (1973) 1223–1226.
Conditions A: A C18 column (Luna, C18 5
m
m, 250 mm ꢄ
4.6 mm, Phenomenex, USA) was used with a flow rate of 1 mL/min.
Eluent was acetonitrile/water (50/50, v/v) containing 0.1% KF (UV
detection at 254 nm).
Conditions B: A C18 column (Luna, C18 5
4.6 mm, Phenomenex, USA) was used with a flow rate of 2 mL/min.
m
m, 250 mm ꢄ
[22] A.J. Lin, S.R. Pardini, B.J. Lillis, A.C. Sartorelli, J. Med. Chem. 17 (1974) 668–672.