Organic Letters
Letter
a
sterically demanding NHCs, such as N8, N9, and N10, were
less effective for promoting the catalytic Truce−Smiles
rearrangement of 1a (entries 8−10). Finally, using N7 with a
prolonged reaction time permitted the complete consumption
of 1a with the quantitative generation of 2a (entry 11). These
conditions were applicable to the reaction on a gram scale
(entry 11).
Table 1. Optimization of the Catalyst
entry
NHC
GC yield of 2a (%)
With the optimized reaction conditions in hand, we next
examined the scope of substrates (Scheme 2). A range of
functional groups, including cyano (1b), iodo (1c), methoxy
(1d), and ester (1e) groups, were tolerated with the
corresponding rearranged products being successfully pro-
duced. When para-substituted anilides were used, this
organocatalytic rearrangement proceeded efficiently with
substrates bearing a CH2Mes group (i.e., 1f−1i), possibly
because this bulkier N-protecting group suppressed the
undesired C(acyl)−N bond cleavage and the associated side
reactions. Notably, since this method does not require the use
of a strong, nucleophilic base, substrates bearing an enolizable
ketone moiety, as in 1i, were applicable. Sterically demanding
ortho-substituted anilides 1j−1m also participated in this
organocatalytic C−N cleavage reaction. Interestingly, ortho-
chloro and ortho-bromo substituents did not undergo
nucleophilic aromatic substitution by the postulated ylide
intermediate11,13 under these conditions, but rather C−N
bond substitution occurred exclusively to form the correspond-
ing cinnamamides 2j and 2k, respectively. One of the
advantages of this organocatalytic method over transition
metal and photoredox catalysis is the tolerance of halogen
groups, as evidenced in the reactions of 1c, 1j, and 1k, which
should serve as a synthetic handle for further functionaliza-
tion.15 The heteroaromatic compound 1n and π-extended
arenes 1o also participated in this catalytic reaction. The
electronic properties of the benzyl protecting groups in anilides
had no apparent effect on the yield of the products, as
evidenced by the reactions of anilides bearing 4-methoxybenzyl
(1p) and 4-trifluoromethylbenzyl (1q) groups.
1
2
3
4
5
6
7
8
N1
N2
N3
N4
N5
N6
N7
N8
N9
N10
N7
0
0
0
0
16
9
26
0
0
0
9
10
11
b
c
>95 (79)
This organocatalytic C−N bond cleavage is applicable to the
transformation of biologically active molecules containing an
aniline moiety (Scheme 3, top). For example, the C(aryl)−N
bonds in the antimicrobial sulfamethazine 1r and anesthetic
dimethocaine 1s can be transformed into C(aryl)−C bonds via
this NHC-catalyzed rearrangement to provide new amide
derivatives 2r and 2s, respectively. As mentioned above, an
ortho bromo group is tolerated in the N7-catalyzed Truce−
Smiles rearrangement of 1k to afford a cinnamamide 2k
exclusively. Interestingly, the use of N9, instead of N7, as a
catalyst led to the formation of a cyclized product 3 as a major
product, which was produced by catalytic nucleophilic
aromatic substution (Scheme 3, bottom).11,13 These results
demonstrate that a catalyst-controlled selectivity between C−
N and C−Br bond cleavage is possible.
To gain additional insights into the reaction mechanism, we
next investigated the reaction pathway by DFT calculations
using N-methyl-N-(naphthalene-1-yl)methacrylamide as a
model substrate. The energy changes at the M06-2X/6-
311+G*//M06-2X/6-31+G* level of theory [SCRF (pcm,
solvent = toluene)] are shown in kcal/mol (Scheme 4).
Because the route from an α,β-unsaturated ester and an NHC
catalyst to the ylide intermediate similar to INT1 was
previously calculated,16 our calculations focused on the critical
C−N bond cleavage process via intramolecular nucleophilic
substitution. The calculations revealed that this catalytic C−N
We previously reported on the NHC-catalyzed intra-
molecular concerted nucleophilic aromatic substitution
(CSNAr) of aryl halides bearing an α,β-unsaturated amide
moiety,11 in which a highly nucleophilic ylide species is
involved (Scheme 1D, attack a).12 In these reactions, oxygen-
based poor leaving groups, such as a OPh and even an a OMe
group, can also participate,13 although C(aryl)−O bonds are
generally assumed to be inert.14 The broad scope of leaving
groups led us to envision that α,β-unsaturated amides without
any leaving groups at the ortho position would direct the attack
of the ylide nucleophile to the most electrophilic ipso carbon of
the anilide substrate to undergo the Truce−Smiles rearrange-
ment (Scheme 1D, attack b). The key challenges to realizing
the nucleophile-initiated Truce−Smiles rearrangement are (i)
the low electrophilicity of anilides compared to aryl halides and
(ii) difficulties associated with suppressing the competing
cleavage of a more reactive C(acyl)−N bond in anilides.
With these challenges in mind, we commenced our study by
evaluating various NHC catalysts for the conversion of the
anilide 1a into the rearranged product 2a (Table 1). Although
common triazole-based (N1) and imidazole-based NHCs (N2,
N3, and N4) did not produce the desired 2a (entries 1−4), the
introduction of methyl groups at the 4,5-positions of the
imidazole ring (i.e., N5, N6, and N7) permitted the desired
C−N bond cleavage to form 2a (entries 5−7) with the
methoxy-substituted N7 being among the most effective. Less
1573
Org. Lett. 2021, 23, 1572−1576