Organic Letters
Letter
the aryl ring, rendering the following C−H bond cleavage
more difficult. On the basis of this analysis, a cyclic formamide
(3u) that could efficiently inhibit the twist of the aryl ring was
tested and yields of ≤84% was achieved. Other aryls (3v and
3w) and heteroaryls (3x) also worked well in the reactions,
providing yields of ≤97% yield.
Scheme 5. Mechanistic Experiments and a Plausible
Mechanism
We then proceeded to investigate the substituent effect of
alkynes (Scheme 4). Various alkyl substituents, including both
Scheme 4. Scope of Alkynes
bond along with alkyne insertion or Ni-catalyzed direct H
transfer into alkyne occurred in this step.6,12 Intra- and
intermolecular competitive experiments gave two similar kH/kD
values, 1.56 and 1.50, respectively (eqs 3 and 4), suggesting
that the second C−H activation may occur through a
concerted pathway instead of the typical SEAr mechanism.6
31P NMR tracing experiments showed that small amounts of
the SPO-bound Ni−Al complex could be formed (95.9 ppm)
when equal numbers of equivalents of L11, AlMe3, and
Information). However, the process can be accelerated by
addition of the formamide, whereas the alkyne did not have the
same influence. The result suggested that the formamide that
coordinated to Ni promoted isomerization of SPO with AlMe3.
Notably, the resulting complex was found to be able to
promote the annulation of alkyne 2a, providing product 3a in
60% yield. On the basis of these observations and previous
discussions,13 we proposed a plausible mechanism for this
reaction (Scheme 5b). First, the formed SPO-bound Ni−Al
bimetallic complex initiates formyl C−H bond cleavage
through oxidative addition (or H transfer).14 Subsequent
alkyne insertion provides the key intermediate (A), which
undergoes aryl C−H bond cleavage to produce the
intermediate (B). Final alkyne insertion and reductive
elimination led to the desired product 3a.
a
Reaction conditions: 1a (0.5 mmol), 2 (1.0 mmol), and toluene (0.2
b
mL) under N2 for 2 h. Yield for isolated products. L4 (5 mol %) was
used instead of L11.
linear alkynes (4a and 4b) and cyclic alkyne (4c), can be
tolerated well, providing the corresponding product in yields of
85−89%, whereas diphenyl alkyne was not suitable for this
annulation, leading to only undesired acrylamides. We
speculated that the insertion of diphenyl alkyne would
significantly increase the steric hindrance of the intermediate,
thus inhibiting subsequent aryl C−H activation. Therefore, in
this case, more flexible SPO ligands were then re-examined. To
our delight, L4 can promote the reaction, albeit only in 25%
yield in the current stage (4d). Although low regioselectivities
were obtained for asymmetric dialkyl alkynes (4e/4e′ and 4f/
4f′), asymmetric alkylphenyl alkynes (4g/4g′ and 4h/4h′) led
to pretty good regioselectivities (>10:1). Moreover, the phenyl
was exclusively proximate to the carbonyl group, which was
These results suggested that the second alkyne insertion of the
nickelacycle could start from the aryl−Ni bond, thus
preferentially forming a more stable benzylic nickel species.
To understand the mechanism, some mechanistic experi-
ments were performed. Tracking formyl H of the amide by the
use of deuterated formamide 1a-d under the standard
conditions showed that D was completely transferred into
the acrylamide and the alkene (Scheme 5a, eq 1). Parallel
reactions disclosed no isotope effect for the formyl C−H bond
(eq 2), excluding this C−H bond cleavage from the rate-
determining step. Either oxidative addition of Ni to the C−H
In summary, we have developed a ligand-controlled method
for nickel-catalyzed dual C−H annulation of arylformamides
and alkynes, affording various quinolin-2(1H)-ones in yields of
≤97%. This SPO-enabled bimetallic catalysis demonstrates its
unique ability to activate C−H activation and suppress
decarbonylation of arylformamides, which will improve the
future design of other types of C−H bond functionalization in
our lab.
C
Org. Lett. XXXX, XXX, XXX−XXX