Organic Letters
Letter
substituted phenyl boronic acids also exhibited good
reactivities to give the desired products (3f−3i) in moderate
to good yields. Di- and trisubstituted boronic acids on the aryl
ring were also compatible as coupling partners. It should be
emphasized that a high functional-group tolerance was
observed, including methyl ester (3j), ketone (3k), cyano
(3l), and halides (Br, Cl, and F; 3m−3o, respectively), which
subsequently could provide potential points for further
chemical modulation. Nevertheless, a low yield (36%) was
obtained when the electron-withdrawing substituent p-NO2
was flanked on the aromatic moiety (3p). Interestingly,
boronic acid bearing the heterocycle 1q was well tolerated
and converted to the corresponding product 3q in a 52% yield.
To further expand the scope of our methodology, the Cu-
catalyzed allylic C−H arylation of other cyclic and acyclic
alkene derivatives was investigated, as illustrated in Scheme
3.22 By varying the size of rings of the cyclic alkenes during our
acids, giving the corresponding arylated cyclohexene deriva-
tives 15a and 15j in 60% and 53% yields, respectively.
Allylbenzene derivatives 8 and 9 proved to be effective
substrates for the allylic C−H arylation reaction with both
electron-rich and electron-poor boronic acids, producing the
corresponding linear products (E)-16 and (E)-17 with (E)-
stereochemistry in moderate yields (Scheme 3). It is
noteworthy that the reaction with a noncyclic internal olefin,
such as 2-methyl-1-phenylpropene 10, worked well to give the
corresponding products (E)-18 and (Z)-18 in a 78% yield with
a ratio of 10:3, respectively. Finally, when 1-methylcyclohexene
11 was reacted with 4-methoxyphenylboronic acid, a mixture
of three products 19−21 was isolated; however, the mixture
was inseparable by flash chromatography. Among those, the
major coupling compound 19 was obtained via the reaction of
the less sterically hindered allylic C−H bond. It appears that
the regioselectivity of this allylic C−H arylation is under steric
control, thus explaining the formation of the minor arylated
exocyclic product 21 while the abstraction of this primary
allylic hydrogen is unfavored.
a
Scheme 3. Substrate Scope of Alkenes
To obtain insight into the mechanism of allylic C−H
arylation, a stoichiometric amount of 2,2,6,6-tetramethyl-1-
piperidine (TEMPO) was used as a radical scavenger. The
arylation was completely inhibited, and the methylated
TEMPO adduct was observed.17 This result suggests that the
allylic C−H arylation proceed by a radical pathway involving
the presence of methyl radicals. Based on previous mechanism
studies of Karasch−Sosnovsky-type reactions, three mecha-
nism pathways can be considered for this transformation, as
depicted in Scheme 4. Our first mechanistic hypothesis,
Scheme 4. Proposed Mechanisms
a
Reaction conditions are as follows: 1a−q (0.5 mmol), 4−10 (10
equiv), Cu2O (10 mol %), L1 (10 mol %), DTBP (2 equiv), and
DMSO (1.5 mL) under N2 at 80 °C for 48 h.
studies, slightly lower yields with both electron-rich and
electron-deficient boronic acids were observed in the presence
of cyclooctene 4 and cycloheptene 5 compared to those
obtained with cyclohexene. Products 12 and 13 were produced
in 45% to 77% yields. In contrast, a drop in the yield was
observed with cyclopentene 6, since the efficiency of the
arylation is dependent on the electronic properties of the
boronic acids. Indeed, while electron-rich boronic acids 1a and
1c led to the desired products 14a and 14c in 40% and 20%
yields, respectively, allylic C−H arylation reactions with
electron-deficient boronic acids 1k, 1n, and 1i were
unsuccessful. It is noteworthy that a substrate featuring an
exocyclic double bond at the cyclohexyl ring could also be
arylated from electron-rich 1a and electron-deficient 1j boronic
pathway A, begins with the decomposition of DTBP initiated
by LCuI via a single-electron transfer (SET) reaction to
produce a tert-butoxy radical and the oxidized copper(II)
complex I, LCuII-OtBu.23 The formed tert-butoxy radical can
either decompose by β-scission into acetone and a methyl
radical, which can be trapped by TEMPO,24 or be involved in
hydrogen atom abstraction (HAA) step from an alkene to form
an allyl radical.25 In parallel, the copper(II) complex II is
generated by the transmetelation reaction of aryl boronic acid
with LCuII-OtBu (I).21 Taking into account the known
propensity of CuII complexes to react with an organic
3132
Org. Lett. 2021, 23, 3130−3135