Organic Letters
Letter
methylation reactions, including in a variety of tandem
annulation processes based on the installation of the carbonyl
group through this platform.12
selective N-tert-butoxycarboxylation, thus enabling to engage
generic acyclic amides in the reactivity platform.4,5 After
experimentation, we were delighted to discover that the
combination of KOt-Bu as a base and trimethylsulfoxonium
iodide as a precursor to dimethyloxosulfonium methylide
afforded the desired product at room temperature conditions
(Table 1, entry 1). As expected, the relative ratio of
trimethylsulfoxonium iodide and the base are crucial for the
reaction efficiency (Table 1, entries 1−9). Out of various bases
screened, KOt-Bu was found to be optimal (Table 1, entries
10−15). KOH and NaOtBu were less effective. Furthermore,
DMSO was identified as an effective solvent for the reaction
(Table 1, entries 16−19). Moreover, it is important to note
that the best results are obtained by nucleophile activation to
yield dimethyloxosulfonium methylide prior to the addition to
the amide (Table 1, entries 20−21). Overall, the optimized
process is highly efficient; the N-Boc deprotection deactivating
the amide bond is not observed. Furthermore, overaddition of
dimethyloxosulfonium methylide to give the epoxide7−9 as well
as direct displacement to give amino-epoxide18 (path b, Figure
1B) are not observed, consistent with the stability of the
tetrahedral intermediate and high selectivity of the process.
Likewise, overreaction product of sulfoxonium ylide with
amide has not been observed.
Having optimized conditions in hand, we next focused our
attention on evaluating the scope of this process (Scheme 1A).
As shown, the reaction displays broad scope and is successful
with a variety of amides. As such, electronically neutral (3a),
electron-donating (3b−3c), and electron-withdrawing sub-
stituents (3d−3g) on the aromatic ring are well-compatible
with the process. Of note, different fluorine substitution
important from the medicinal chemistry standpoint, such as
CF3, OCF3, F, or 3,4-F2 is well tolerated. Perhaps most notably
the method is fully compatible with sensitive functional groups,
such as bromo (3h), nitro (3i), cyano (3j), and ester (3k),
thus demonstrating significant practical advantage over metal-
catalyzed methods and providing handles for further
functionalization. Moreover, ortho-steric hindrance (3l), as
well as the presence of conjugated systems, such as biaryls
(3m) and differentially substituted naphthalenes (3n−3o), is
well accommodated. Likewise, the presence of heterocyclic
systems, such as furyl (3p) or thienyl (3q) are readily
tolerated. Finally, we were pleased to find that the method can
be extended to vinyl amides (3r) and aliphatic 1° and 2°
amides (3s−3u). At the present stage, 3° amides are not
tolerated because of the excessive steric hindrance. Unactivated
amides are recovered unchanged, consistent with resonance
destabilization as the driving force for the reaction.5,6
Aldehydes and ketones are not tolerated because of epoxide
formation.9
In this context, our laboratory has been focused on
developing new methods for amide bond activation.4a,b,5,6 In
continuation of studies, herein, we report the direct synthesis
of sulfoxonium ylides from amides by selective N−C(O)
cleavage (Figure 1B). The reaction proceeds through the
nucleophilic addition of dimethylsulfoxonium methylide to the
amide bond in acyclic twisted amides under exceedingly mild
room temperature conditions. The method advances the
growing portfolio of transition-metal-free reactions of activated
amides by tetrahedral intermediates, including amidation,13
esterification,14 thioesterification,15 and acylation,16,17 to the
synthesis of sulfoxonium ylides.
The method is notable in that the protocol provides the first
general approach to sulfoxonium ylides by amide N−C(O)
bond cleavage. The method displays a wide substrate scope
and can be applied to the late-stage derivatization of
pharmaceuticals. Furthermore, we present mechanistic studies
that outline the relative order of reactivity of amide derivatives
in this process.
Selected optimization experiments are summarized in Table
1. As shown, we selected monoactivated N-Boc/Ph amide as a
twisted amide precursor for this process (vide infra). These
amides feature decreased amidic resonance (RE = 7.2 kcal/
mol; τ = 29.1°; χN = 8.4°),6 while N-acyl-Boc-carbamates are
readily prepared from common 2° or 1° amides (vide infra) by
a
Table 1. Optimization of the Reaction Conditions
base (y
Me3SO+I− (x
yield
(%)
entry
1
2
base
equiv)
equiv)
solvent
KOt-Bu
KOt-Bu
KOt-Bu
KOt-Bu
KOt-Bu
KOt-Bu
KOt-Bu
KOt-Bu
KOt-Bu
Et3N
Et3N
Et3N
LiHMDS
LiHMDS
NaH
5.0
5.0
4.0
3.0
3.0
2.2
3.0
3.0
2.2
5.0
5.0
3.0
3.0
2.0
1.2
3.0
2.0
1.2
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
DMSO
DMSO
dioxane
toluene
THF
THF
76
65
61
95
63
62
81
61
47
<2
<2
<2
62
39
75
52
92
81
84
<5
<5
b
3
4
5
6
c
7
c
8
c
9
10
11
12
13
14
15
16
17
18
19
3.0
6.0
1: 1 vol/vol
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
Prompted by the broad scope of the process, we
demonstrated the utility of the reaction in the late-stage
derivatization of pharmaceuticals (Scheme 1B). As shown, the
reaction can be used for the preparation of the sulfoxonium
ylide from Febuxostat (antigout) (3v) and Probenecid
(antihyperuricemic) (3w), further demonstrating functional
group tolerance of the process and potential utility in
pharmaceutical settings.
NaH
KOt-Bu
KOt-Bu
KOt-Bu
KOt-Bu
KOt-Bu
d
20
e
21
We next studied the potential of various amides as
electrophilic precursors for this process (Scheme 2). As
shown, in addition to N-Boc/Ph amides, electronically tuned
N-acyl-Ts-sulfonamides (1x), atom-economic N-Ms amides
(1y), readily prepared from 1° amides N,N-Boc2 amides (1z),
a
Conditions: amide (1.0 equiv), Me3SO+I− (x equiv), base (y equiv),
solvent (0.10 M), nucleophile activation (65 °C, 2 h), then 23 °C, 15
b
c
d
h. Nucleophile activation (65 °C, 5 h). THF (0.2 M). Without
e
nucleophile activation, 23 °C. Without nucleophile activation, 65 °C.
4819
Org. Lett. 2021, 23, 4818−4822