Organic Letters
Letter
Table 1. Optimization of Cross-Coupling of N-
a
Acylcarbazoles
b
entry
catalyst
base
solvent
yield (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[Pd−PEPPSI−IPr]
[Pd−PEPPSI−IPr]
[Pd−PEPPSI-IPr]
[Pd−PEPPSI−IPr]
[Pd−PEPPSI−IPr]
[Pd−PEPPSI−IPr]
[Pd−PEPPSI−IPr]
[Pd−PEPPSI−IPr]
[Pd−PEPPSI−IPr]
[Pd(IPr)(cin)Cl]
[Pd(IPr)(1-t-Bu-ind)Cl]
[Pd(IPr)(allyl)Cl]
[Pd−PEPPSI−IPent]
[Pd−PEPPSI−IMes]
Pd(OAc)2/PCy3HBF4
K2CO3
K3PO4
Na2CO3
Cs2CO3
KOH
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
THF
THF
THF
THF
THF
toluene
dioxane
DCE
THF
THF
THF
THF
THF
THF
THF
93
52
<5
17
13
89
51
28
86
92
81
71
50
15
12
c
d
a
Conditions: amide (1.0 equiv), 4-Tol-B(OH)2 (3.0 equiv), catalyst
b
(3 mol %), base (4.5 equiv), solvent (0.25 M), 80 °C, 15 h. GC/1H
NMR yields. 120 °C. Pd(OAc)2 (3 mol %), PCy3HBF4 (12 mol %),
H3BO3 (2.0 equiv), 80 °C. cin = cinnamyl; ind = indenyl; PEPPSI =
3-Cl-pyridine.
c
d
Figure 1. (a) Amide bond resonance. (b) Activation of amides and
derivatives. (c) This work: Nlp to Ar conjugation switch, N-
acylcarbazoles: new class of highly reactive amides for N−C(O)
cross-coupling.
Interestingly, the screen of various Pd(II)−NHC catalysts
indicated that catalysts bearing allyl-type throw-away ligands,
such as Pd(IPr)(cin)Cl, Pd(IPr)(1-t-Bu-ind)Cl, and Pd(IPr)-
(allyl)Cl, are also effective in this reaction, delivering the
coupling product in good to high yields (entries 10−12). In
contrast, there is a significant impact of the steric demand of
the NHC ancillary ligand, with both less sterically demanding
IMes and more hindered IPent giving low yield of the cross-
coupling product (entries 13 and 14). Finally, we also tested
the use of Pd/phosphane conditions, which resulted in low
conversion (entry 15); thus, the high σ-donation of the NHC
ligand using bench-stable, well-defined Pd(II)−NHCs is highly
beneficial for the coupling.
studies provide key insight into bond destabilization of the
amide bond. Overall, we expect that bench- and hydrolytically
stable N-acylazolides that permit facile N−C(O) activation will
provide a very attractive approach to the generation of
acylmetals from amides for a variety of coupling reactions.
In agreement with our previous studies, we hypothesized
that diminution of amidic resonance in anilides might be
rendered possible by channeling the nN → π*CO resonance
into another functional group.
After significant experimentation optimizing amide bond
geometry, we identified N-acylcarbazoles as suitable substrates
for this process. The amide bond in a model N-
benzoylcarbazole (1a) is relatively planar16c,d (τ = 25.1°; χN
= 3.2°, Winkler−Dunitz parameters, N−C(O) bond length of
1.400 Å, C = O of 1.212 Å), which can be compared with a
model predominantly planar N-methylanilide (N−C(O) bond
length of 1.355 Å, CO of 1.230 Å).13
With the optimized conditions in hand, the scope of this
transformation was examined with respect to the amide
component (Scheme 1). As shown, this amide bond activation
is well-compatible with neutral (3a), electron-donating (3b,
3c), and electron-withdrawing (3d−3g) substituents on the
amide. It is particularly noteworthy that electrophilic functional
groups that would be problematic in classical addition of hard
organometallics, such as ketones (3f) and esters (3g), are
tolerated under these catalytic conditions. Furthermore, steric
hindrance (3h), aliphatic amides (3i), and heterocyclic amides
conjugated at the deactivating, electron-rich position (3j−3k)
were easily tolerated. Next, we examined the scope of the
reaction with respect to the boronic acid component (Scheme
1). As shown, electron-neutral (3b′), electron-rich (3c′), and
electron-deficient (3d′, 3f′, 3g′) arylboronic acids were
successful substrates, albeit a lower yield using electron-
withdrawing groups has been noted (vide infra). Furthermore,
sterically hindered boronic acids (3h′), polyaromatic boronic
acids (3l, 3m), and heterocyclic boronic acids (3k′, 3n, 3o)
furnished the cross-coupling products in good to high yields.
Note that the moderate yield for the thienyl product is likely a
Selected optimization studies of the Suzuki−Miyaura cross-
coupling of N-benzoylcarbazole (1a) with 4-tolylboronic acid
are presented in Table 1. The optimized conditions utilize Pd−
PEPPSI−IPr (3 mol %), K2CO3 (4.5 equiv) in THF at 80 °C
(Table 1, entry 1, see Figure SI-1 for catalyst structures). The
use of other bases, including K3PO4, Na2CO3, Cs2CO3, and
KOH, was less successful (entries 2−5). The low yields using
Na2CO3 and Cs2CO3 are likely a balance in precatalyst
activation and acylmetal stability. Several other solvents were
examined, such as toluene, dioxane, and DCE, and provided
inferior results (entries 6−8). A higher temperature was also
compatible with this reaction, consistent with the stability of
N-acylcarbazole moiety under these conditions (entry 9).
B
Org. Lett. XXXX, XXX, XXX−XXX