Organic Letters
Letter
−SO2F groups were generated from readily available metal
fluorides rather than from cumbersome gaseous fluorosulfuryl
sources. Unlike the fluoride−chloride exchange using sulfonyl
chloride, the nucleophilic fluorination of chlorosulfate with a
metal fluoride such as KF has been relatively unexplored.14
Herein, we report the first direct synthesis of 18F-labeled aryl
fluorosulfates (Figure 1b) from both phenols and isolated aryl
imidazole sulfonates (imidazylates). We devised two strategies
for 18F-radiochemistry adaptation. The first involved the one-
pot radiofluorosulfurylation of phenols using 1,1′-sulfonyldii-
midazole (SDI) and 18F−. The second strategy focused on the
radiofluorination of isolated imidazylate precursors by
incorporating 18F− into their sulfonyl moieties. At the outset,
our protocols offered several distinct advantages over
previously reported methods. First, (radio)fluorination could
be achieved with a readily available nucleophilic fluoride
source. Second, conventional synthetic approaches could be
adapted without the need for specialized equipment or
additional safety precautions. Finally, our protocols are readily
transferrable to the efficient radiosynthesis of 18F-labeled aryl
fluorosulfates. Thus, our approach is expected to enable the
rapid screening of pharmaceutical candidates along with
exquisite PET imaging.
At the beginning of our studies, we focused on non-
radioactive fluorosulfurylation to assess the practicality of our
methods. We evaluated the formation of −OSO2F group
directly from a phenolic substrate. Initially, we attempted to
generate 2-naphthyl fluorosulfate (1b) by reacting 2-naphthol,
SDI, and silver fluoride (AgF) in a one-pot reaction under
neutral conditions (see Table S1, Section III in the Supporting
Information (SI) for the detailed procedures). This process
provided 2-naphthyl fluorosulfate (1b) with 54% yield. We
carefully characterized the reaction mixture and identified an
imidazylate derivative of 2-naphthol. Assuming that the
−OSO2F group formed via such an imidazylate intermediate,
we speculated that the imidazylate itself could be a viable
labeling precursor. Next, we evaluated a series of common
metal fluorides to identify the ideal fluoride source for
imidazylate radiofluorination (see Table S2, Section III in
the SI). Silver fluoride afforded the highest yield of 2-naphthyl
fluorosulfate (1b). Nucleophilic fluorination of naphthol
imidazylate (1a-Im) generated 2-naphthyl fluorosulfate (1b)
in up to 90% isolated yield. The substrate scope of these two
methods was evaluated, and the corresponding aryl fluoro-
sulfates were formed in moderate to high yields (see Sections
intermediate during radiofluorination was inevitable in a one-
pot reaction. Similar to the direct fluorination of the
imidazylate using the nonradioactive protocol, the aryl
[18F]fluorosulfate could be obtained from the bench-stable,
isolated imidazylate precursor.15
Then, using these optimized radiochemical conditions, we
focused our efforts on broadening the substrate scope (Scheme
1). We systematically compared two different radiofluorination
modes: the direct one-pot radiofluorosulfurylation of phenolic
precursors (Mode 1) and the radiofluorination of isolated
imidazylates (Mode 2). We hypothesized that Mode 2 would
afford higher RCYs while diminishing the formation of
nonradioactive byproducts. As expected, [18F]fluoro-
sulfurylated products were achieved with both methods,
although the RCYs differed between Modes 1 and 2. With
both methods, the radiofluorosulfurylation of simple phenols
bearing electron-withdrawing (4-F, 4-CF3, 3-I, and CN) and
electron-donating (MeO) substituents afforded [18F]3b−[18F]
11b in 15% to 80% RCYs, regardless of the position of the
substituent on the ring. Interestingly, the electronic nature of
the ring did not significantly influence the overall outcome of
radiofluorination using either Mode 1 or 2. Moderate to high
RCYs were obtained with other electron-rich phenolic
substrates. These included naphthyls (2-naphthyl [18F]1b
and 1-naphthyl [18F]2b), biphenyls [18F]12b−[18F]14b,
xylenols (2,6-xylenol [18F]15b and 3,5-xylenol [18F]16b),
mesityl [18F]17b, and benzylic ether [18F]18b. No discernible
substituent effects occurred with respect to the position and
number of methyl substituents in the xylenolic systems.
Heteroaromatic systems produced their corresponding fluo-
rosulfated heteroarenes [18F]19b−[18F]21b. Halogen-substi-
tuted pyridine [18F]21b was obtained in 38% RCY from the
corresponding imidazylate 21a-Im. This is particularly note-
worthy as [18F]21b provides a reactive synthetic handle for
potential coupling operations.
Other functionalized phenols also produced aryl fluorosul-
fates. Ketone-bearing phenols [18F]22b−[18F]24b with enoliz-
able α-protons were radiofluorinated via their imidazylates in
58−68% RCYs without the formation of radioactive
tautomeric isomers. Azide-functionalized aryl fluorosulfates
[18F]25b and [18F]26b were obtained from imidazylates 25a-
Im and 26a-Im in 68% and 57% RCYs, respectively.
Remarkably, 3-hydroxybenzyl fluorosulfate [18F]27b was
obtained from 3-hydroxybenzyl alcohol. The sulfurylative
activation of benzyl alcohol with SDI led to the selective
functionalization of the phenolic hydroxy group, leaving the
alkyl hydroxy intact. Radiofluorination of imidazylate 27a-Im
also afforded the corresponding fluorosulfate [18F]27b with
higher RCY under basic labeling conditions (18F/K2CO3/18-
crown-6). No competing radiofluorination was observed at the
benzylic position. A hydroquinone with two imidazyl sites
(28a-Im) was prepared stoichiometrically. Unsurprisingly, the
radiofluorination of this substrate occurred at only one of the
imidazyl sites to produce [18F]28b′ (see Section IX in the SI
for details). This was consistent with 18F− being a limiting
agent in 18F-radiochemistry. Unlike the previous results, [18F]
28b and [18F]28b′ were produced from radiofluorination of
28b′ in 8% and 65% RCYs, respectively. The formation of the
major radiolabeled product [18F]28b′ might presumably stem
from 19F/18F isotopic exchange, which has been recently
reported.4,16 Aryl fluorosulfates with aldehydic functionalities
([18F]29b−[18F]31b) were obtained in good RCYs from both
phenols and their imidazylate derivatives. As prosthetic
Next, we applied these −SO2F transfer reactions in 18F-
radiochemistry systems. While aryl fluorosulfate production
was efficient in the presence of AgF, the implementation of
Ag18F for routine 18F-radiochemistry was problematic. In
addition to fairly low radiochemical yields (RCYs), the Ag18F
complex generated from Ag2CO3 adhered to the wall of the
reaction vial, compromising the overall labeling efficiency.
Therefore, we used the 18F/K2CO3/18-crown-6 complex as a
phase-transfer agent during radiofluorosulfurylation. This
provided the highest RCY of radiosulfurylated 4-fluorophenol
[18F]3b (see Section VIII in the SI). Using a one-pot method,
we successfully radiolabeled phenol with 18F− to afford 4-
fluorophenyl [18F]fluorosulfate ([18F]3b). A close examination
of the HPLC-UV chromatogram revealed the presence of the
corresponding imidazylate intermediate that was previously
observed in the nonradioactive reaction mixture (see Section X
in the SI). We postulated that the formation of this
C
Org. Lett. XXXX, XXX, XXX−XXX
Kwon, Young-Do
