Angewandte
Communications
Chemie
(2y), ketone (2ee), phthalimide (2u, 2v), nitro (2p), cyano
(2bb), and phosphoryl (2x) moieties, were well tolerated
under the reaction conditions, and might be attributed to the
mild reaction conditions employed. Notably, 1,1-disubstituted
terminal alkenes are also suitable for the trifluoromethyl-
fluorosulfonylation reaction, thus providing moderate yields
of the corresponding trifluoromethylfluorosulfonylation
products (2aa, 2bb). It is interesting that 1,2-disubstituted
cyclic internal alkenes can also be subjected to the trans-
formation, thus generating the desired products in acceptable
yields (2cc, 2dd). However, styrene-type alkenes and elec-
tron-poor alkenes are not good candidates for the trifluor-
omethylfluorosulfonylation reaction, thus suggesting a pro-
found effect of electronic properties of the alkenes (2gg,
2hh). In addition, terminal alkenes derived from relatively
complex molecules 4-methylumbelliferone and estrone were
compatible with the reaction conditions and yielded the
desired products in satisfactory yields, which demonstrates
the potential of the trifluoromethylfluorosulfonylation reac-
tion in late-stage functionalization (2ee, 2 ff). Moreover, the
trifluoromethylfluorosulfonylation reaction of the alkene 1v
on gram scale was carried out to study the possible scalability
of the reactions. It was found that the reaction proceeded
smoothly and yielded the desired product in 81% yield (2v),
thus exhibiting excellent scalability of the trifluoromethyl-
fluorosulfonylation reaction. In addition, further transforma-
tion of the trifluoromethylfluorosulfonylation products was
preliminarily investigated. As demonstrated in Equa-
tions (1)–(3), the reactions of the trifluoromethylfluorosulfo-
nylation product 2v with either MeONa, piperidine, or
NH3·H2O conveniently afforded the corresponding and
potentially useful sulfonate 3 and sulfonamides 4 and 5 in
good yields. And thus conveniently install two important
functional groups into the original lead compounds or drugs.
d = À87.8 ppm, which were considered to represent AgCF3
and AgSO2CF3,[12] respectively, and is consistent with the
literature.[13] After 3 hours, the signal of AgCF3 almost
disappeared while the intensity of the signal of AgSO2CF3
was enhanced, and observation that might be attributed to the
reaction of AgCF3 with SO2 generated in situ from
Ag(O2CCF2SO2F). Notably, the addition of NFSI into the
resulting reaction mixture led to the formation of CF3SO2F
because of the facile fluorination of sulfinate AgSO2CF3 by
NFSI.[14] Next, we investigated the stepwise reaction of
Ag(O2CCF2SO2F) in acetonitrile with 1a and NFSI
[Eq. (5)]. It was found that the reaction of 1a and NFSI
with the 15 minute decomposition reaction mixture of
Ag(O2CCF2SO2F) in acetonitrile afforded 49% yield of the
desired product 2a together with 21% yield of CF3SO2F,
while the 3 hour decomposition reaction mixture provided
a trace amount of 2a, and 41% yield of CF3SO2F was
obtained as the main product. These experimental results
suggest that AgCF3 generated in situ from Ag(O2CCF2SO2F)
may be the key CF3 radical source. Finally, addition of the new
alkyl radical, formed by addition of the CF3 radical to the
alkene, to sulfur dioxide, and subsequent reaction with NFSI
generate SO2F groups in the desired products.[14,10b] It should
be mentioned that Reiser et al. recently reported an elegant
visible-light-mediated trifluoromethylchlorosulfonylation of
unactivated alkenes with CF3SO2Cl in the presence of
a copper catalyst, and it was considered to proceed by an
inner-sphere radical mechanism,[5a] which is different from
ours.
In conclusion, we have developed, for the first time,
a novel intermolecular radical trifluoromethylfluorosulfony-
lation reaction of unactivated alkenes with the readily
available Ag(O2CCF2SO2F) and NFSI under mild reaction
conditions with good functional-group compatibility. The
reaction mechanism involves the addition of a CF3 radical,
generated from a AgCF3 species, to the alkene substrate to
generate an alkyl radical intermediate, sulfur dioxide inser-
tion, and subsequent fluorination reaction with NFSI. Nota-
bly, both CF3 and sulfonyl groups originate from
Ag(O2CCF2SO2F). Further investigation on the development
of other fluoroalkylfluorosulfonylation, or direct fluorosulfo-
nylation reactions is in progress.
To gain more insights into the reaction mechanism
(Scheme 2), several control experiments were conducted on
1a. First, several radical inhibition and radical-clock exper-
imental results demonstrate that the reaction may proceed
through a radical pathway and possibly involve a CF3 radical
species (see the Supporting Information for details). Second,
the decomposition of Ag(O2CCF2SO2F) itself in acetonitrile
and its stepwise reactions with either 1a or NFSI were
monitored and investigated by 19F NMR analysis [Eqs. (4)
and (5); see the Supporting Information for details]. 19F NMR
spectra of the reaction mixture of Ag(O2CCF2SO2F) in
acetonitrile at room temperature under an Ar atmosphere
for 15 minutes showed signals at d = À22.2, À24.7 ppm, and
Acknowledgements
The authors gratefully acknowledge the financial support
from the National Natural Science Foundation of China (Nos.
21421002, 21302207, 21672239, 21737004), the National Basic
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!