Journal of the American Chemical Society
Communication
aroyl fluoride component was varied using 4-methylstyrene as
the coupling partner. Electron-rich (3ab and 3ah) and
electron-poor (3ac, 3ai, and 3aj) aroyl fluorides reacted well,
and the product ketones were obtained in 51−75% yield.
Halogen substituents were tolerated (3ad−ag), and aroyl
fluorides bearing an extended π system or heteroarenes like
thiophene and furan reacted to afford ketones 3ak−am in 52−
70% yield. Of note, no product was identified with benzoyl
chloride or benzoyl bromide instead of benzoyl fluoride. The
potential of the process was further documented by late-stage
functionalization of more complex activated alkenes. For
example, alkenes derived from D-phenylalanine, hormone,
natural product, and marketed drugs bearing an indole ring or
functional groups such as ketone, amide, halide, and ester
could all be aroylated at the α-position (3da−ha). Thus, our
process offers a good platform to access more complex α-
substituted vinyl ketones.
To illustrate the synthetic value of these α-substituted vinyl
ketones, a gram-scale reaction of 1m was conducted with
reduced amounts of the catalysts, and a comparable yield was
obtained (Scheme 3a). Furthermore, [3 + 2] annulation82 of
vinyl ketone 3a with an N-acylhydrazone provided dihydro-
1H-pyrazole 5 in 75% yield (Scheme 3b). Photoredox-
catalyzed hydroacylation of 3al with α-oxocarboxylic acid
gave 1,4-diketone 6 (Scheme 3c),83 and epoxidation of enone
3z afforded oxirane 7 in 71% yield (Scheme 3d).
To support the mechanism suggested in Scheme 1,
additional experiments were conducted. With acyl azolium
ion 8 as the substrate in the absence of the NHC, 4-
methylstyrene reacted in 51% yield to give vinyl ketone 3a,
indicating that acyl azoliums B (see Scheme 1) are competent
intermediates (Scheme 4a). The reaction between α-
methylstyrene and benzoyl fluoride with 1 equiv of p-
chlorophenyl sulfinate furnished the three-component coupling
product 9 in 48% yield (Scheme 4b). The absence of an acidic
proton at the position β to the sulfone moiety prevents the
sulfinate elimination, showing that three-component products
of type D are intermediates in these cascades.
cycles are interwoven. Along with the carbene, a photoredox
catalyst and a sulfinate catalyst are used for α-acylation of
alkenes with acyl fluorides to access α-substituted vinyl
ketones. The cascade exhibits high functional group tolerance.
Successful late-stage modification shows the potential of the
method, and useful follow-up chemistry on the product
ketones further documents the value of the process. Notably,
existing methods for acylation of aryl alkenes usually afford the
β-acylation products. Mechanistic studies indicate that the
reaction proceeds through a radical addition/coupling/
elimination cascade. We are confident that multiple catalysis
proceeding through NHC-catalyzed radical transformations
will enable the discovery of other novel transformations in the
future.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
■
sı
Experimental procedures, characterization data, and
1
copies of H and 13C NMR spectra (PDF)
AUTHOR INFORMATION
Corresponding Author
■
Armido Studer − Organisch-Chemisches Institut, Westfälische
Wilhelms-Universität, 48149 Munster, Germany;
̈
Author
Kun Liu − Organisch-Chemisches Institut, Westfälische
Wilhelms-Universität, 48149 Munster, Germany
̈
Complete contact information is available at:
Notes
The authors declare no competing financial interest.
Furthermore, to investigate the sulfinate elimination from D,
a parallel kinetic isotope effect (KIE) experiment was carried
out using 1z and [D]-1z as substrates (Scheme 4c). The two
reactions were stopped after 40 min, and on the basis of the
individual conversions, a KIE value of 2.8 was calculated. A
competition KIE experiment using equal amounts of 1z and
[D]-1z (0.5 equiv each) was stopped after 40 min, and analysis
of the unreacted starting alkene revealed a KIE value of 2.2
(Scheme 4d). These results indicate that the deprotonation
process might be involved in the rate-determining step.
Moreover, when the model reaction was conducted in the
presence of 2 equiv of 2,2,6,6-tetramethylpiperidin-1-oxyl
(TEMPO), the α-acylation was not observed. Instead,
benzoyl-TEMPO adduct 10 was isolated in 19% yield,
suggesting ketyl radical C as an intermediate (Scheme 4e).
On the other hand, the formation of adduct radical A was
supported by a radical probe experiment using styrene 11 as
the acceptor to give ring-opening product 12 (44% yield;
Scheme 4f). Finally, fluorescence quenching experiments
revealed that only sodium sulfinates quench the excited state
of Ru*(II) (see Figure S2 for details), supporting the reductive
quenching pathway. Thus, all of these experiments are in line
with our suggested mechanism.
ACKNOWLEDGMENTS
We thank the European Research Council (Advanced Grant
Agreement 692640) for supporting this work.
■
REFERENCES
■
3522.
ates. Chin. J. Chem. 2020, 38, 1167−1202.
702.
(4) De Sarkar, S.; Biswas, A.; Samanta, R. C.; Studer, A. Catalysis
Eur. J. 2013, 19, 4664−4678.
(6) Flanigan, D. M.; Romanov-Michailidis, F.; White, N. A.; Rovis,
Chem. Rev. 2015, 115, 9307−9387.
In conclusion, we have developed the first example of triple
catalysis involving a carbene catalyst in which all three catalysis
(7) Izquierdo, J.; Hutson, G. E.; Cohen, D. T.; Scheidt, K. A. A
4907
J. Am. Chem. Soc. 2021, 143, 4903−4909