Organic Letters
Letter
functional groups such as methyl, methoxyl, fluoride, chloride,
and bromide were introduced to the phenyl ring of the
arylsulfone moiety in 2-vinyl anilines, and the desired products
were obtained in good to excellent yields (Scheme 2, c16−
c21). Further experiments found that, besides arylsulfone
moieties, the substrates equipped with quinoline sulfonyl,
naphthalene sulfonyl, and trifluoroacetyl as the N-protected
group were also compatible with the reaction to give yields of
86, 71, and 73%, respectively (Scheme 2, c22−c24).
Regrettably, when N-Boc 2-vinyl aniline was used, no desired
product was observed (Scheme 2, c25). It is noteworthy that
the structure of c23 has been confirmed by single-crystal X-ray
diffraction analysis.21
detected by high-resolution mass spectrometry (HRMS),
revealing that a thiyl radical is involved in the reaction
(Scheme 4a). It is commonly known that once a sulfur radical
Scheme 4. Reaction Mechanism Investigations
We further explored the variety of thiols. It is known that the
formation of thiyl radicals from aryl thiols is faster than that of
alkyl thiols, leading to the addition of aryl thiols to styrene
preferentially affording anti-Markovnikov-type adducts via a
radical pathway. Notably, in this photoredox/cobalt-catalytic
system, the anti-Markovnikov-type adducts were completely
suppressed, and the Markovnikov-type adducts were exclu-
sively obtained with excellent regioselectivity, no matter
whether the phenyl ring of the aryl thiols was substituted
with electron-withdrawing groups or electron-donating groups
(Scheme 3, c26−c30). Moreover, various commercially
is generated, it is preferential to access the anti-Markovnikov-
type product via a radical pathway. However, in this catalytic
system, Markovnikov-type products were exclusively obtained.
We assume that this is attributed to the fact that the rate of the
addition of the thiyl radical to the cobaloxime catalyst is faster
than that of alkene, which could suppress the generation of an
anti-Markovnikov-type adduct. The deuterium labeling experi-
ment indicated that the hydrogen source in the product c1
mainly originated from the thiol substrate b1 (Scheme 4b).
Recently, Chen, Xiao, and coworkers22 reported a nitrogen-
radical-mediated difunctionalization of 2-vinyl anilines. To
confirm the possibility of a nitrogen radical’s involvement in
this reaction, we designed several control experiments. 3-Vinyl
aniline a40 and 4-vinyl aniline a41 were exposed to the
standard conditions, and no reaction occurred (Scheme 4c,d).
Furthermore, when the N−H moiety of a1 was methylated, the
reaction still proceeded smoothly in 70% yield (Scheme 4e).
These results suggest that a nitrogen-radical-mediated pathway
could be excluded in this reaction. Nevertheless, when the N−
H moiety was protected by bulky groups such as benzyl (a43),
isopropyl (a44), and tert-butyl (a45), the reaction could not
happen (Scheme S2). We speculate that the steric hindrance
would affect the coordination ability of nitrogen.
a
Scheme 3. Scope of Thiols
a
Reaction conditions: a1 (0.22 mmol), b (0.20 mmol), PC (5 mol
%), Co(dmgH)2PyCl (20 mol %), CH2Cl2 (3.0 mL), under N2, room
b
temperature, 450 nm LED irradiation for 24 h. Isolated yields.
available alkyl thiols were examined. The results revealed
that primary thiols such as 2-phenyl ethyl mercaptan (b31),
ethyl thioglycolate (b32), 2-methy butyl mercaptan (b33), n-
octyl mercaptan (b34), and furan-2-yl methanethiol (b35)
reacted well with a1 to generate the corresponding
Markovnikov-type products in good yields (Scheme 3, c31−
c35). Regrettably, secondary and tertiary thiols were ineffective
in this transformation, probably due to the steric hindrance
effect (Scheme 3, c36−c39).
On the basis of the above observations, a reasonable
mechanism is proposed in Scheme 5. The protected amine
group as a directing group for activating the alkene is
coordinated to a cobalt catalyst to generate the species I-1.
Scheme 5. Proposed Reaction Pathway
To gain more insights into the reaction mechanism, a series
of control experiments were carried out. Quenching experi-
ments showed that the excited photocatalyst *PC could be
effectively quenched by thiol b1, indicating an electron-transfer
process between b1 and *PC (Figure S1). Furthermore, when
a radical inhibitor (2,2,6,6-tetramethyl-1-piperidinyloxy
(TEMPO)) was added, the desired product was completely
inhibited, and the TEMPO-trapped thiyl radical adduct was
3606
Org. Lett. 2021, 23, 3604−3609