K.R. Jayakrishnan et al. / Tetrahedron Letters xxx (xxxx) xxx
3
and also lucrative. On the basis of this hypothesis, the use of vari-
ous imidazolium pre-catalysts (4 to 7) together with KOtBu
(20 mol%) afforded the anticipated amination product 3aa in mod-
erate yields (entries 1–4, Table 1). Furthermore, NHC pre-catalysts
such as imidazolinium (8), thiazolium (9 and 10) and triazolium
(11 to 15) were less efficient and delivered 3aa in considerable
yields (entries 5–12). Satisfyingly, the aminated 2-naphthol 3aa
was obtained in 80% yield when the triazolium pre-catalyst 16
was engaged as the catalyst (entry 13), which is presumably due
to the increased acidity of triazolium precatalysts than other pre-
catalysts, wherein the in situ produced conjugate acid of the cat-
alytic species could act as proton source (see Scheme 2) [3l].
Several bases were screened for the amination of 2-naphthol
derivatives in the presence of 16, among them KOtBu furnished
the desired product in excellent yield (see ESI for the base screen-
ings). Remarkably, the amination reaction proceeded well when
CHCl3 was used as a solvent in the presence of pre-catalyst 16
and furnished the requisite product in excellent yield (96% yield,
entry 19). To ensure the complete utilization of base during
in situ-generation of active carbene catalyst, 10 mol% of pre-cata-
lyst 16 and 8 mol% of KOtBu was employed, and the expected pro-
duct 3aa was obtained in 75% yield (entry 20) [9].
With the optimized reaction conditions in hand, we went on to
scrutinize the scope of various naphthol derivatives and dialkyl
azodicarboxylates. A series of di-alkyl azodicarboxylates such as
di-tert-butyl azodicarboxylate (DtBuAD, 2a) diethyl azodicarboxy-
late (DEAD, 2b), diisopropyl azodicarboxylate (DIAD, 2c) and
dibenzyl azodicarboxylate (2d) reacted with 1a to afford the corre-
sponding aminated products (entries 3ab-ad, Table 2) in good
yields. Similarly, excellent yields were obtained in the case of 2-
naphthol bearing an aryl substituent (1b) (Table 2, entries 3ba-
bd). Whereas, an electron- donating group (such as a methoxy
group) on the 2-naphthol (1c) ring resulted in the formation of
analogous aminated product in appreciable yields (entries 3ca-
cd). Besides, the direct amination reaction proceeded well for the
substrate bearing bromo-substituent at 3- and 6-position (1d and
1e) and offered the concomitant products in better yields (entries
3da and 3ea-ed, Table 2). Electron-withdrawing groups such as
cyano, ester, amide, and ketone at the various position of the naph-
thol ring (1f-1j) reacted smoothly and delivered the requisite
products in moderate to excellent yields (entries 3fa-3ja), how-
ever, 2-naphthol bearing nitro substituent at the peri-position
failed to undergo amination reaction presumably due to strong
electron-withdrawing nature of nitro group which makes 8-
nitro-2-naphthol less nucleophilic towards amination reaction.
Additionally, 2-naphthol holding an allylic ether functionality at
7-position handed the desired product in moderate yield (56%,
entry 3ka). Interestingly, substituents at the peri-position of 2-
naphthol (1l and 1m) with DtBuAD (2a) furnished the desired
products 3la and 3ma in 74% and 68% yields, respectively. Next,
we explored 2-naphthol derivative encompassing a hetero-atom
on its ring (1n) for the amination reaction, and the corresponding
target products were obtained in moderate to good yields (entries
3na-3nd). Ultimately, the efficacy of the reaction was tested with
hydroxycoumarin (1o) with 2a as an aminating source, and the
corresponding aminated product was attained in high yield (93%,
entry 3oa).
Herein, we propose a plausible mechanism for an NHC-pro-
moted Friedel-Crafts type amination of 2-naphthol derivatives
based on the literature precedence (Scheme 2) [4]. At the outset,
the hydrogen bonding interaction between free-carbene catalyst
(I) and 2-naphthol (1a) activates the 1a for nucleophilic attack
(as shown in step II), which then, undergoes Friedel-Crafts type
amination reaction with 2a to form the intermediate III. Next,
intramolecular-proton exchange, followed by protonation of resul-
tant oxy-anion by triazolium precatalyst produces the target pro-
duct 3aa and regenerates the active free-carbene (I) in the
catalytic cycle (step IV).
NHC-promoted alcoholysis of glutaric anhydrides
Then, we envisaged an alcoholysis of glutaric anhydrides using
alcohols as pronucleophile by embracing Brønsted basic activation
of alcohol by a carbene catalyst demonstrated by Scheidt’s group
[4e]. The ensuing hemiesters (having chemically two distinctive
carbonyl groups) are important synthetic intermediate for several
bioactive molecules. Various metal catalysts and cinchona-
alkaloid based catalysts were previously known to mediate
asymmetric desymmetrization of meso-anhydrides [8].
To validate our speculation, a model reaction was performed in
the presence of a commercially available free-carbene catalyst
such as A (IMes, 10 mol%) using 3-phenylglutaric anhydride (4a)
as a precursor, MeOH (5a, 10 equiv) as the pronucleophile in
toluene. Gratifyingly, the desired hemiester (6aa) was obtained
in good yield (Scheme 3a). Next, we endeavoured the same reac-
tion by exercising carbene catalyst which is in situ generated from
NHC precatalyst (4) and DBU, and as expected the desired hemie-
ster was obtained in excellent yields (Scheme 3b) [10].
With these findings in hand, further alcoholysis experiments
were performed using various glutaric anhydrides. As summarized
in Table 3, methanolytic desymmetrization of phenyl glutaric
anhydride (4a) furnished the desired hemiester in excellent yield
(entry 6aa, Table 3). Increasing the carbon chain of the alcohol
scaffold such EtOH (5b), 2-propanol (5c), allylic alcohol (5d), and
propargyl alcohol (5e) resulted in slight deterioration of the pro-
duct yields, surmising that steric factor and acidity of the alcohols
may be the possible reasons (entries 6ab-6ae, Table 3) [4b]. Then,
we investigated the alcoholysis of glutaric anhydride bearing
substituent at the para position of the phenyl ring, for instance,
3-(4-fluorophenyl)glutaric anhydride (4b) on methanolysis
afforded the corresponding hemiester in good yield (Table 3,
entry 6ba). Next, we explored a methanolysis of unsymmetrical
cyclic anhydride like phenylsuccinic anhydride (4c), and in this
case, the hemiester was isolated as the sole product, presumably
due to steric hindrance exerted by the phenyl group (entry 6ca,
Table 3). In addition, preliminary investigations on asymmetric
version of this reaction were performed in the presence of
X
N
N
R
R
X=N or CH
base
X
N
HO
HO
tBuO2C
N
R
R
N
t
N
H
CO2 Bu
3aa
I
1a
R
N
R
N
H
H
O
O
II
X
X
N
R
N
R
t
CO2 Bu
tBuO2C
IV
N
N
t
N
H
CO2 Bu
N
2a
tBuO2C
O
H
t
N
N
CO2 Bu
tBuO2C
III
Scheme 2. Plausible mechanistic pathway.
Please cite this article as: K. R. Jayakrishnan, M. Tamilarasu, K. V. Jincy et al., N-Heterocyclic carbene as a Brønsted base catalyst for the amination of naph-