Organic Letters
Letter
halides. In cases of primary acids, we could see a good functional
group tolerance to C−C double bond (22) or triple bond (23),
chloride (24), ether linker (25), ketone (26), and also ester (27
and 28). Dihydrocinnamic acids (29−31) and the acids with
one carbon extension (33) or abridged (34) also reacted well.
To our delight, this decarboxylative thioesterification can be well
applied to the modification of naturally occurring carboxylic
acids and several drugs (35−44). The transformation of amino
acid proline gave the thioester (38) in 45% isolated yield, while
the side-chain carboxylic acid group of aspartic acid and glutamic
acid could also be modified (39 and 40). Furthermore, the
carboxylic groups of drugs such as Fenbufen (41), Gemfibrozil
(42), and Indometacin (44) could be transformed readily to the
corresponding thioesters in good to high yields. As outlined in
Scheme 1, under this tandem photoredox and copper catalytic
system, various primary, secondary, and tertiary carbon radicals,
including benzylic radicals (16, 20, and 34) can all be readily
generated from the corresponding redox-active esters and
coupled subsequently with thiobenzoic acid to afford the
desired thioester products, allowing for a facile access to
thioesters and their important derivatives2 with diverse
structures.
from the photoinduced decarboxylation was trapped by
TEMPO instead (53, Figure 2, a), suggesting a radical pathway
The past few years have witnessed a fast-growing research
activity on the study of sulfonyl fluorides as selective probes in
the context of chemical biology and molecular pharmacology.13
Thiols are important precursors for the synthesis of sulfonyl
fluorides.13a,15 As exemplified in a scaled up reaction (Scheme
2a), the free thiol can be readily obtained upon in situ hydrolysis
Scheme 2. Hydrolysis to Free Thiol and Direct Conversion of
Thioesters to Sulfonyl Fluorides
Figure 2. Mechanistic studies and a possible mechanism.
is operative. We next conducted a series of fluorescence
quenching experiments with the photocatalyst [Ru(bpy)3]Cl2·
6H2O.14 Although no apparent fluorescence decrease was
observed with either CuBr or bpy/Ph3P ligand alone, the
addition of the CuBr/bpy/Ph3P combination resulted in a
significant fluorescent quenching (Figure 2b, left), indicating the
ligand coordination could make the Cu(I) more reducing. In
fact, Cu(I) complexes have been reported as a photoredox
catalyst able to reduce the redox-active ester via a single electron
transfer (SET).7f,12d,16 Notably, in contrast to the nitrogen or
oxygen nucleophiles,12d−g the thiobenzoic acid that was found
can effectively quench the fluorescence of the excited [RuII]
photocatalyst and, thus, could also act as an electron donor to
generate the highly reducing [RuI] which is required to reduce
the NHPI ester. Based on these results and the related
mechanistic studies in literature,12d−g,17 a possible mechanism
was outlined in Figure 2c. Under light irradiation, the
photocatalyst [RuII] is first excited and then accepts an electron
from LnCu(I) species (A), which can be easily formed from
CuBr and thiobenzoic acid by anion exchange in the presence of
base,18 to afford the corresponding highly reducing [RuI] and
the key CuII species (B). A single electron transfer from [RuI] to
the redox-active ester delivers the alkyl radical (R·), which can
be trapped by CuII B to generate the key intermediate C, which
then undergoes a reductive elimination to afford the desired
thioester products, while the incoming thiobenzoic acid can
regenerate A. Alternatively, *[RuII] can be reductively quenched
by thiobenzoic acid17b to generate [RuI], and the PhCOS·
generated concurrently can be trapped by LnCu(I)X species to
form B as well (for simplicity, not shown in the catalytic cycle).
with aq. NaOH after the decarboxylative thioesterification
finished. Considering the instability and unpleasant smell of free
thiols, we became interested to develop a direct conversion of
thioester to sulfonyl fluorides. After examined several reaction
conditions, we found that Selectcfluor (chloromethyl-4-fluoro-
1,4-diazoniabicyclo [2.2.2] octane bis(tetrafluoroborate)) is a
suitable reagent for this goal (Scheme 2, b). With this protocol,
primary, secondary, and also tertiary thioesters can be readily
converted to the corresponding sulfonyl fluorides (46−52) in
one step.
To gain a better understanding about this decarboxylative
thioesterification reaction, we first conducted a radical trapping
reaction with TEMPO under the standard conditions, which
showed no thioester 3 formed. The cyclohexyl radical generated
C
Org. Lett. XXXX, XXX, XXX−XXX