4176
S.-G. Kim, K. H. Ahn / Tetrahedron Letters 42 (2001) 4175–4177
Table 1. Enantioselective Michael addition catalyzed by BTO–tBuOK complexes
(S,S,S)-BTO 1-3
CO2Me
+
Ph
CO2Me
CO2Me
KOtBu
Ph
CO2Me
(R)
Entry
BTO (mol%)
KOtBu (mol%)
Conditiona
Solvent
Time
Ee (%)b
Yield (%)c
1
2
3
4
5
6
7
8
1 (10)
1 (5)
10
10
20
50
20
20
20
20
20
20
20
A
A
A
A
B
A
A
A
A
B
Toluene
Toluene
Toluene
Toluene
Toluene
CH2Cl2
Wet toluene
THF
3 h
6 h
3 h
15 min
3 h
35
36
46
31
56
47
–
75
51
86
55
80
80
–
17
63
83
85
1 (10)
1 (10)
1 (10)
1 (10)
1 (10)
1 (10)
2 (10)
2 (10)
3 (10)
6 h
d
3 h
3 h
8 h
3 h
3
9
10
11
Toluene
Toluene
Toluene
64
82e
41
A
a Condition A: a mixture of tBuOK and BTO in toluene was stirred at room temperature for 1 h, then the mixture was cooled to −78°C followed
by the addition of the ester and methyl acrylate. Condition B: a mixture of tBuOK and the ester in toluene was stirred for 15 min at −78°C
followed by the addition of BTO, and the resulting mixture was further stirred for 15 min before the dropwise addition of methyl acrylate in
toluene.
b Determined from optical rotation data (Ref. 8).
c Isolated by SiO2 chromatography.
d Very slow.
e [h]2D2=−73.0 (c 1.0, EtOH).
expected, coordinating solvents such as THF or the
presence of water in toluene gave poor reactivity and
enantioselectivity. Also, the molar ratio of the base to
BTO 1 was found to affect the enantioselectivity. Thus,
when we doubled the amount of tBuOK, from 10 to 20
mol%, the enantioselectivity was raised from 35 to 46%
ee. However, further increasing the base gave worse
effects in both the yield and the enantioselectivity.
Interestingly, an apparent improvement in the enan-
tioselectivity (increment of 10% ee) was observed when
we changed the addition sequence of reactants, from A
to B. Thus, an optimized reaction condition is to use 10
mol% of the BTO and 20 mol% of tBuOK in toluene
under condition B at −78°C. It is notable that the
reaction did not proceed at all when the base was
replaced with tBuONa.
Michael reaction. This expectation was found to be the
case: A significant improvement in the enantioselectiv-
ity was achieved under the established condition using
tBu-BTO 2, and an enantioselectivity of 82% ee was
obtained, a comparable enantioselectivity to the highest
ones obtained so far (83–85% ee).3 In the case of
Ph-BTO 3, a slightly lower enantioselectivity was
observed compared to that observed with iPr-BTO 1.
The observed enantioselectivities depending on the
BTOs indicate that for better enantioselectivity more
tight molecular interactions are required between the
Michael acceptor and the potassium enolate, which is
presumably coordinated to the BTO.
In regard to the mechanism of the Michael addition, we
believe that a potassium enolate of the ester bounds to
the BTO through a tripodal coordination of the oxazo-
line nitrogens to the K+ ion. A face-selective approach
of the Michael acceptor to a p-face of the enolate
would result in the addition product enriched in one
enantiomer.‡ Thus, the geometric isomerism of the
enolate3d as well as the chiral environment around both
the enolate and the acceptor would determine the sense
and degree of the enantio-discrimination. A reliable
description is guaranteed by a further mechanistic
study. To answer an important question whether the
BTOs coordinate the K+ ion in the described tripodal
mode, not in a bidentate chelated one, we have synthe-
sized a new chiral bis(oxazoline), BBO 4,§ and evalu-
Next, applying the established reaction condition, we
evaluated two other ligands, tBu-BTO 2 and Ph-BTO
3.9 The former had to be synthesized from (2,4,6-
trimethylbenzene)-1,3,5-triacetic acid and tert-leucinol,
following the procedure used for the synthesis of the
others.† The bulky tBu groups on the oxazoline rings
are expected to provide a highly crowded environment
around the potassium ion coordinated to the oxazoline
ligands, which may increase the enantioselectivity of the
† tBu-BTO 2: overall 22% yield in a one-pot reaction; Rf=0.45 (ethyl
acetate/hexanes=3/2, by volume); mp 134–135°C; [h]2D1=−137.6 (c
1
0.50, CHCl3); H NMR (300 MHz, CDCl3): l 4.07 (dd, 3H, J=8.6,
9.9), 3.92 (dd, 3H, J=8.0, 8.6), 3.78 (dd, 3H, J=8.0, 9.9), 3.72 (s,
6H), 2.37 (s, 9H) 0.83 (s, 27H); 13C NMR (75 MHz, CDCl3): l
166.0, 136.1, 131.1, 76.0, 69.0, 34.0, 30.5, 26.3, 17.6; MS (FAB): m/z
(rel. intensity) 538 (M+1, 100), 480 (13). Elemental analysis for
‡ Possibility of enantioselective protonation of a potassium enolate as
the chiral discrimination mode is excluded by examining the optical
purity of reaction mixture at different time intervals, which gave the
same optical purity in each case.
C33H51N3O3: calcd: C, 73.70; H, 9.56; N, 7.81. Found: C, 73.53; H,
§ To be published elsewhere.
9.56; N, 7.93.