Organic Letters
Letter
acylisothiouronium intermediates (Scheme 1b).16 If successful,
it would provide a feasible avenue to axially chiral anilides.
Herein, we wish to disclose our studies toward this end. A
highly atroposelective N-acylation of readily available aniline-
derived sulfonamides was realized under mild conditions by
employing homobenzotetramisole (HBTM) as the catalyst.
Various atropoisomeric sulfonyl substituted anilide products
were afforded in good yields with high to excellent
enantioselectivities.
In light of the fact that some enantioenriched sulfonamides
are pharmaceutically attractive compounds,2a,f,h and inspired
by Matsubara’s work,17 sulfonamide 1a with α,β-unsaturated
carbonic anhydride 2a were chosen as the model reaction
substrates to optimize the reaction conditions (Table 1). First,
solvents with low polarity. Toluene was the best choice, and
the corresponding chiral anilide 3a was obtained in 80% yield
with 81% ee (entry 6). Subsequent investigation of the base
(entries 6−9) implied that coordinating cations deteriorated
the ee considerably and neutral base DIPEA gave better results
than ionic bases (entry 9, 82% yield, 87% ee). When 3 Å M.S.
was used as the additive, the yield increased slightly to 91%
without influencing the enantiomeric excess (entry 10).20 To
improve the stereoselectivity, we next turned our attention to
evaluate the substituents at sulfonamide. Less bulky mesyl-
substituted substrate 1b gave a remarkable increase in the
enantioselectivity, but with a lower yield (entry 11, 56% yield,
95% ee). Careful analysis of reaction mixture by TLC indicated
that a side product was generated concurrently, which was
responsible for low isolated yield. This side product, arising
from the nucleophilic attach of anilide anion to isopropox-
a
Table 1. Optimization of the Reaction Conditions
1
ycarbonyl unit, was assigned to be compound 4b by H and
Further optimization studies revealed that a higher yield (88%)
with maintained enantioselectivity (entry 12, 95% ee) was
delivered when the reaction was performed with 1b (0.1
mmol), 2a (1.5 equiv), HBTM C3 (10 mol %), and 3 Å M.S.
(15 mg). It should be noted that when N-heterocycle carbene
was used as acyl transfer catalyst with cinnamaldehyde under
oxidative conditions no desired product was detected.21
With the optimal reaction conditions in hand, a variety of
substituted α,β-unsaturated carbonic anhydrides 2 were
examined with sulfonamide 1b (Scheme 2). Overall, regardless
of the substituent pattern and the electronic property of the
aryl moiety, the corresponding axially chiral anilides (3b−m)
were afforded in good to excellent yields with high
enantioselectivities (93−97% ee). Generally, electron-deficient
substrates (2j−l) provided a higher yield than that of electron-
rich ones (2b−d). Notably, 1-naphthyl-substituted anhydride
2m as well as anhydrides 2n and 2o derived from heteroaryl-
substituted acrylic acids were compatible in current systems,
furnishing the desired products 3n−p in 74−86% yields with
uniformly high enantioselectivities (95−96% ee). Moreover,
the reactions of anhydrides with aliphatic substitution took
place as well, and the enantioenriched products (3q and 3r)
were obtained in good yield with high enantiomeric purity.
The absolute configuration of 3l was determined to be R by X-
ray diffraction analysis and the configurations of all other
examples were assigned by comparing their CD spectra with
that of 3l.
b
c
entry
cat.
sol
base
yield (%)
ee (%)
1
2
3
4
5
6
7
8
9
C1
C2
C3
C3
C3
C3
C3
C3
C3
C3
C3
C3
DCM
DCM
DCM
THF
Et2O
tol
tol
tol
tol
tol
KOtBu
KOtBu
KOtBu
KOtBu
KOtBu
KOtBu
NaOtBu
LiOtBu
DIPEA
DIPEA
DIPEA
DIPEA
trace (3a)
NR (3a)
96 (3a)
72 (3a)
77 (3a)
67
35
57
81
60
0
87
87
95
95
80 (3a)
>95% (3a)
>95% (3a)
82 (3a)
91 (3a)
56 (3b)
88 (3b)
d
10
11
12
de
,
tol
tol
ef
,
a
Unless otherwise noted, all reactions were performed with catalyst
(20 mol %), 1a (0.10 mmol), 2a (3 equiv), and base (2 equiv) in
solvent (1.0 mL) at 35 °C. Yield of isolated product. Determined by
HPLC analysis on a chiral stationary phase. Conducted with 3 Å
M.S. (10 mg). Conducted with 1b (0.10 mmol). Carried out with 2a
(0.15 mmol), catalyst (10 mol %) and 3 Å M.S. (15 mg).
b
c
d
e
f
Encouraged by above results, the generality of sulfonamides
1 was further investigated with anhydride 2a (Scheme 3).
Installing other substituents at the para position in o-tert-butyl-
NH-anilide did not show any obvious influence on the
enantioselectivity of the present transformation (3s−v).
However, switching the o-tert-butyl group to smaller
substituents, such as methyl, bromo, or iodo groups in anilide,
led to an apparent drop in chiral control (3w−y, 81−85%
yields, 73−81% ee). Next, several sulfonamide substrates were
surveyed. Similar to p-tosyl amide 1a, the reaction with
benzenesulfonamide gave the expected N-acylation products
3aa with moderate yield (48%) and good enantioselectivity
(87% ee). The yield of products 3aa increased to 67% when
the reaction was performed with 3 equiv of anhydride 2a.
Under such conditions, benzenemethanesulfonamide 1k and
cyclohexyl sulfonamide 1l were tested, and the corresponding
products 3ab and 3ac were obtained in satisfactory outcomes
(3ab, 80 yield, 95% ee; 3ac, 73 yield, 91% ee, respectively).
representative chiral isothiourea catalysts with different
skeletons were screened with KOtBu as the base in
dichloromethane (DCM) at 35 °C. It was disappointing to
find that the reaction did not occur in the presence of either
benzotetramisole C1 or tetramisole C2 (entries 1 and 2). To
our delight, chiral homobenzotetramisole (HBTM) C3, first
developed by Birman’s group,18 could promote the N-acylation
reaction smoothly, affording the desired acyl sulfonyl
substituted anilides 3a in high yield (96%) with a promising
enantioselectivity (entry 3, 67% ee).18,19 These proof-of
principle results validated that the control of axial chirality of
anilides by isothiourea-catalyzed atroposelective N-acylation
reaction was feasible. Then other reaction parameters such as
the nature of solvent, base, additive, and catalyst loading were
examined. It was observed that the solvent displayed a
significant effect on the reactivity and enantioselectivity of
the process (entries 3−6). Generally, highly polar solvents, for
example THF, led to more inferior results than those of
B
Org. Lett. XXXX, XXX, XXX−XXX