Organic Letters
Letter
their powerful ability to convert alkanes to alkenes.6 From the
mechanistic point of view, copper(II) species should
coordinate to a carbonyl substrate during the desaturation of
the dienophile precursor.6e,i However, electron-rich dienes
generally used in the DA reaction usually do not contain a
carbonyl group. As a result, a catalytic oxidant system, which
has been developed in those dehydrogenative approaches, may
be not applicable to the DDDA reaction. Thus, to realize our
hypothesis, a new catalytic system for double-dehydrogenation
of dienophile and diene precursors should be established.
Herein, we describe a copper(I)/DDQ-mediated DDDA
reaction of butenes with 1,4-diketones and indolones via
activation of fourfold inert C(sp3)−H bonds (Scheme 1, eq 3).
We initially investigated the DDDA reaction of prenyl
benzene 1a wth 1,4-bis(4-chlorophenyl)butane-1,4-dione 2a in
the presence of CuI as a catalyst, DDQ as an oxidant, and
several common N-containing bidentate ligands (Table 1,
entries 1−4). The target product 3a was obtained in the
presence of CuI (15 mol %), L1 ligand (30 mol %), and DDQ
(2.6 equiv) in chlorobenzene at 115 °C for 96 h, which is
constant with our hypothesis. Of the ligands examined (entries
1−4), 4,4′-di-tert-butyl-2,2′-bipyridine (L4) gave the best result
(entry 4) for the copper-catalyzed DDDA reaction. Di-tert-
butyl peroxide, generally used in many reported metal-
catalyzed oxidative dehydrogenation reactions, returned only
a trace amount of the desired product (entry 6). Subsequently,
a variety of copper catalysts and solvents were evaluated
sequentially (entries 7−15). When the DDDA reaction was
performed by employing DDQ as the oxidant, CuI as the
catalyst, and 4,4′-di-tert-butyl-2,2′-bipyridine (L4) as the ligand
in trifluoromethylbenzene, the expected product 3a could be
isolated in 91% yield (entry 15). Reducing the loading of
catalyst and ligand to 10 mol % and 20 mol % led to a lower
yield (entry 16). Surprisingly, in the absence of a copper
catalyst, product 3a was obtained in 19% yield (entry 17),
while no desired product was detected when the reaction was
conducted in the absence of L4 (entry 18). These results
suggest that the ligand plays a critical role in the oxidation of
1,4-diketone 2a to a 1,4-enedione intermediate. As expected, in
the absence of oxidant no reaction resulted (entry 19). These
control experiments revealed that all reaction components in
the DDDA reaction were indispensable. Notably, although
poor diastereoselectivities of 3a were observed, both of them
are endo-selectivity, resulting from the stereospecificity of the
DA reaction.
With the optimal catalytic conditions determined, we then
evaluated the substrate scope in terms of substituted but-2-
enes by employing 1,4-bis(4-chlorophenyl)butane-1,4-dione
2a as the dienophile precursor. As shown in Scheme 2, the
reaction could be amenable to a wide range of butene 1 with
diverse substituents, furnishing cyclohexene derivatives 3 in
good to excellent yields and excellent endo selectivities. Prenyl
benzene substrates bearing varieties of functional groups such
as halogens, alkoxy, and alkyl at the ortho-, meta-, and para-
position of the phenyl ring were subjected to this new DDDA
reaction, and the corresponding products, in most cases, were
obtained in excellent yields (3a−m, 83−91%; 3n−o, 73% and
62%). Excellent yields were also gained with either thienyl- or
naphthalenyl-substituted prenyl substrates (3p−q, 88% and
81%). Moreover, diphenyl substituted but-2-ene 1r returned
the expected product 3r in 90% yield. In addition, several 1,4-
diketones were further explored, and the reactions proceeded
smoothly to generate the desired products in excellent yields
(3am−ac, 82−93%).
Subsequently, the substrate scope with respect to diaryl-
substituted 1,4-diketones was investigated with the use of
diphenyl substituted butenes. Notably, but-1-ene-1,4-diyldi-
benzene (1s) and but-2-ene-1,4-diyldibenzene (1t) were all
compatible and afforded the same desired product 4a in 81%
and 85% yields, respectively. As summarized in Scheme 2, a
series of diaryl-substituted 1,4-diketones bearing electronically
diverse functionalities at any position on the phenyl ring
smoothly underwent the DDDA reaction with also exclusive
endo selectivity (4b−m, 51−92%, from 1t). Nevertheless, 1,4-
diketones with different substituents at the ortho-positon of
the phenyl ring exhibited lower reactivity probably due to
steric hindrance, and only moderate yields were furnished (4j−
l, 51−67%). Additionally, thienyl- and naphthyl-substituted
1,4-diketones were well-tolerated, and high yields were
observed (4n−p, 73−84%, from 1t). The structure of product
4p was further assigned by X-ray analysis (CCDC 1961346).
To further enlarge the generality of the DDDA reaction, we
performed this protocol with benzoylmethyl indolones. As
shown in Scheme 2, an array of indolones derived from N-
methyl isatins reacted effectively with prenyl benzene 1a under
the optimal catalytic conditions, leading to the desired
a
Table 1. Optimization of the Reaction Conditions
b
c
entry
catalyst
CuI
CuI
CuI
CuI
solvent
ligand oxidant yield (%)
dr
1
2
3
4
5
6
7
8
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
L1
L2
L3
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
−
DDQ
DDQ
DDQ
DDQ
26
73
37
89
n.d.
trace
77
80
24
1:1.2
1:1.1
1:1.2
1:1.3
−
d
CuI
CuI
BQ
e
DTBP
DDQ
DDQ
DDQ
DDQ
DDQ
DDQ
DDQ
DDQ
DDQ
DDQ
DDQ
DDQ
−
−
CuBr
CuCl
CuCl2
CuBr2
CuSO4
Cu(OAc)2
CuI
CuI
CuI
CuI
−
1:1.3
1:1.3
1:1.2
1:1.1
−
9
10
11
12
13
14
15
21
trace
trace
75
72
91
78
19
n.d.
n.d.
−
PhBr
1:1.2
1:1.3
1:1.3
1:1.2
1:1.1
−
PhCH3
PhCF3
PhCF3
PhCF3
PhCF3
PhCF3
f
16
17
18
19
CuI
CuI
L4
−
a
Reactions were carried out with 1a (0.20 mmol), 2a (0.10 mmol),
DDQ (0.26 mmol), CuI (15 mol %), L (30 mol %) at 115 °C in
b
c
solvent (2.0 mL) under N2 for 96 h. Isolated yield. The dr value was
d
e
1
determined by H NMR. BQ = p-benzoquinone. DTBP = di-tert-
butyl peroxide. CuI (10 mol %) and L4 (20 mol %) were used.
f
B
Org. Lett. XXXX, XXX, XXX−XXX