Current Organic Chemistry, 2000, 4, 305-342
305
Chiral Lewis Acid Catalyzed Ene-Reactions
Luiz Carlos Dias*
Instituto de Química - Universidade Estadual de Campinas - UNICAMP, CEP: 13083970 - C.P. 6154 - Campinas - SP - Brazil
Abstract: This review covers recent progress in the use of chiral Lewis acid
catalysts in ene-reactions which involve carbonyl and imine compounds as
enophiles. Chiral Lewis acid catalysts containing aluminum, titanium,
ytterbium and copper are critically reviewed. Synthetic applications of recent
systems are specifically discussed.
Dedicated to the Brazilian Chemical Society
General Introduction
Most of the molecules in the world are chiral, and a
wide range of biological and physical functions are
generated through precise molecular recognition that
requires strict matching of chirality [1]. Chirality is not a
prerequisite for bioactivity but in bioactive molecules
where a stereogenic center is present great
differences in biological activities are usually observed
for both enantiomers as well as for the racemic mixture.
At a molecular level, asymmetry dominates biological
processes and a variety of functions responsible for
metabolism and numerous biological responses occur
because enzymes, receptors and other natural binding
sites recognize substrates with specific chirality [1].
Asymmetric synthesis is an important means by
which enantiopure chiral molecules may be obtained
for biological studies and sale, and the synthesis of
biologically relevant natural and unnatural organic
molecules in optically pure form is of fundamental
importance in medicinal chemistry and related
disciplines [1,2,3]. Of particular importance is the
development of asymmetric catalysts for the carboncarbon bond forming reactions [4,5,6]. The use of
chiral catalysts is one of the most attractive methods for
performing asymmetric reactions, because compared
to the stoichiometric use of chiral auxiliaries, a smaller
amount of a readily available chiral material is required.
Therefore, a large quantity of the naturally and
nonnaturally occurring chiral materials are directly
obtained with no need for further manipulation or
removal and recovery of the chiral auxiliary [7,8].
Because stoichiometric reagents are costly to use on a
*Address correspondence to this author at the Instituto de Química Universidade Estadual de Campinas - UNICAMP, CEP: 13083-970 C.P. 6154 - Campinas - SP - Brazil; FAX: (019)-788-3023 - e-mail:
[email protected]
1385-2728/00 $19.00+.00
mole-for-mole basis, many chemists are seeking new
asymmetric catalysts, individual molecules each of
which mediate thousands of enantioselective
conversions [4,9,10]. The strategy is to employ a
reagent that under normal circumstances does not
react with the substrate, but undergoes a selective
reaction under the influence of catalytic amounts of a
chiral compound [11,12,13].
One important criterion when considering an
asymmetric
synthesis
is
the
degree
of
enantioselectivity of a reaction. The pharmaceutical
industry requires chiral products of greater than 99%
ee, with less than 0.1% of the undesired enantiomer.
The design and development of efficient chiral
catalysts for enantioselective synthesis has become
one of the most intense, dynamic and rapidly growing
areas of organic chemical research [13,14]. Remarkable
progress in the development of catalytic asymmetric
reactions has enabled the synthesis of various optically
active compounds with high optical purity [4].
This review presents a comprehensive survey of
some modern and highly selective methods for the
enantioselective ene-reaction, in which asymmetric
induction is derived from the catalyst complex. Recent
advances in this area are turning chemist’s dreams into
reality at both academic and industrial levels.
Introduction to Ene-reaction
The ene-reaction is one of the most powerful
methods for carbon-carbon bond construction in
synthetic organic chemistry and it was first recognized
in 1943 by Alder and classified in his Nobel Lecture as
an “indirect substitution addition” or “ene synthesis” in
1950 [15,16]. The thermal and Lewis acid glyoxylate
ene-reaction was introduced more than 30 years ago
© 2000 Bentham Science Publishers B.V.
306 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
by Klimova and Arbuzov et al. and has been studied by
several groups [17-22]. The ene-reaction occurs
between an alkene having an allylic hydrogen (an
"ene") and a compound containing an electrondeficient double bond (an "enophile") to form a σ-bond
with migration of the ene double bond and a 1,5hydrogen shift (Scheme 1). It requires the energy for
activating the σ C-H and π X=Y bonds, while energy is
gained on forming the σ Y-H and σ X-C bonds. The
thermal pericyclic ene-reaction can proceed through a
concerted six-electron pathway with a suprafacial three
component orbital interaction between the HOMO of
the ene and the LUMO of the enophile (Scheme 1)
[23-24].
bond
formed
3
3
2
2
X
X
+
Y
1
allylic
transposition
H
ene
1
Y
H
enophile
ene: alkene, alkyne, allene, arene,
carbon-heteroatom bond
X=Y: C=C, C=O, C=N, C=S, O=O, N=N
bond
formed
Ene-reaction
Molecular Orbital interaction
C
C
(a)
type
I:
reaction
between
all-carba-ene
components with hetero-enophiles;
(b)
type II: reaction between hetero ene component
and all-carba-enophile;
(c)
type III: reaction occurring between hetero-ene
components and hetero enophiles.
The ene-reaction is mechanistically related to the
Diels-Alder cycloaddition, with enophiles oriented
either exo or endo with respect to the ene component,
but requires a greater activation energy and higher
temperatures. In the ene–reaction involving a
suprafacial orbital interaction, the two electrons of the
allylic σ C-H bond replace the two π electrons of the
diene in the Diels-Alder. The ene process is favored by
electron withdrawing substituents on the enophiles, by
strain in the ene component, and by geometrical
alignments that direct the components in favorable
relative positions. Whether the mechanism is
concerted or stepwise, positive charge is developed to
some extent at the ene component in Lewis acid
promoted reactions [23,24]. At this point, the Lewis
acid promoted ene reactions differ from thermal ene
reactions where steric accessibility of the double bond
and allylic hydrogen is the primary concern.
High demand within the pharmaceutical and fine
industries for efficient and economical methodologies
for the asymmetric synthesis of both simple and
complex molecules has resulted in new developments
through the ene-reaction. An increasing emphasis has
been placed on developing catalyzed asymmetric ene
bond construction as a means of addressing the issues
of cost and operational simplicity inherent in industrial
chemistry [25-28].
HOMO
C
H
H
C
O
LUMO
R
Scheme 1.
According to the nature of the reactants, the enereaction has been divided into two categories:
(1)
all-carbon ene-reactions: takes place between an
olefin bearing an allylic hydrogen atom (the
carba-ene) and an activated alkene or alkyne (the
carba-enophile) and,
(2)
the hetero-ene-reaction: describes a reaction
between an ene or enophile, either of which
contains at least one heteroatom.
The use of Lewis acid catalysts has led to
remarkable progress in the areas of both inter- and
intramolecular ene reactions. When compared to the
intermolecular reactions, intramolecular ene-reactions
are usually more facile, since these reactions take
advantage of less negative activation entropy [2932,37,42].
The carbonyl ene-reaction, particularly promoted by
a stoichiometric to catalytic amount of Lewis acid has
currently emerged as a useful method for the
asymmetric synthesis of acyclic molecules [25-28]. This
asymmetric reaction is a powerful tool for the synthesis
of enantiomerically pure, complex molecules, although
asymmetric catalysis of carbonyl-ene-reactions is
difficult to realize because the Lewis acid seems to be
relatively far from the site of formation of the new chiral
center. The selection of the central metals and the
design of the chiral ligands are particularly important.
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
One of the major problems associated with the
carbonyl-ene strategy is the limitation of the carbonyl
enophiles thus far developed. The synthetic potential
of the ene product heavily depends on the
functionality of the carbonyl enophile employed. The
carbonyl ene-reaction is severely substrate limited to
highly
activated
carbonyl
components
(e.g.
glyoxylates, formaldehyde or chloral) or to highly
activated ene components (e.g. 3-methylene-2,3dihydrofuran). Usually, stoichiometric amounts of
powerful Lewis acids are required owing to the low
nucleophilicity of the olefin and tight binding of the
homoallylic alcohol product to the catalyst. Another
problem is that ene-reactions often suffer a serious
drawback in terms of regiochemistry, for which steric
accessibility of hydrogen is an important factor.
Reactions with unsymmetrical olefins usually give a
mixture of regioisomers.
307
induced by catalytic amounts of chiral Lewis acid
complexes in the presence of molecular sieves 4A (MS
4A) [36,37]. The authors described the first example of
an asymmetric ene-reaction between prochiral,
halogenated aldehydes and alkenes catalyzed by chiral
binaphthol-derived aluminum complex (R)-5 (Scheme
3, Table 1).
RCHO
catalyst
(20 mol%)
+
Me
OH
R1
R
4
absolute configuration
established by conversion
to the known
L-2-hydroxyisocaproic acid
CH 2Cl 2
-78 oC
MS 4A
R1
The catalyst:
SiPh3
Many natural products can be prepared at an early
stage of the synthetic scheme by taking advantage of
an asymmetric ene-reaction. Of special synthetic value
among many carbonyl-ene variants is the glyoxylate
ene-reaction, which provides α-hydroxyesters of
biological and synthetic importance.
O
AlMe
O
(R)-5
SiPh3
Asymmetric Ene-Reactions
The first example of an asymmetric glyoxylate enereaction was described by Whitesell et al.in 1982 and
consisted of the use of a chiral glyoxylate and
stoichiometric amounts of a Lewis acid [33,34].
Addition of methylenecyclohexane 1 as the ene
component to the chiral glyoxylate of 8-phenylmenthol
2 in the presence of equivalent amounts of SnCl4
afforded the corresponding ene product 3 with >99%
diastereoselectivity in 94% yield (Scheme 2) [35].
OH
SnCl4
(100 mol%)
1
O +
-78
oC,
ORc*
94%
3
>99 de
ORc*
O
Table 1.
Addition
of
Alkenes
Halogenated Aldehydes
to
Prochiral
entry
R
R1
catalyst (mol%)
yield (%)
ee (%)a
1
C6F5
Ph
(R)-5 (20)
35
78
2
Cl 3C
Ph
(R)-5 (20)
43
73
3
C6F5
SPh
(R)-5 (20)
88
88
4
C6F5
SPh
(R)-5 (10)
67
78
5
Cl 3C
SPh
(R)-5 (20)
50
53
6
Cl 3C
Me
(R)-5 (20)
79
78
a. Determined by HPLC analysis after conversion to the (-)-MTPA esters
H
2
Scheme 3.
Me
O
Me
Ph
O
Rc*O- =
Me
Scheme 2.
Chiral Aluminum Lewis Acid
In 1988, Yamamoto and coworkers provided the first
indication that asymmetry in ene-reactions could be
The authors observed that the hindered 3,3'-bis(triphenylsilyl)-substituent in chiral catalyst (R)-5 is
essential to achieve good enantioselectivities since the
use of a catalyst derived from Me3Al and 3,3'diphenylbinaphthol led to the racemic product 4 in low
yields. The best result in terms of yields and
enantioselectivities was obtained in the ene-reaction
between
pentafluorobenzaldehyde
and
2(phenylthio)propene in the presence of 20 mol% of
catalyst (R)-5 and MS 4A affording the corresponding
308 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
product 4 in 88% yield and 88% ee (entry 3). With less
reactive methylene compounds, the observed yields
and enantioselectivities in the catalyzed reaction were
modest
and
the
stoichiometric
reaction
is
recommended.
Despite the good levels of asymmetric induction
obtained with this Al(III)-BINOL Lewis acid-based
complex, no Al(III)-derived Lewis acid catalyst system
that affords higher levels of enantioselectivity in the
ene-reaction has been reported to date.
( iPrO)2 TiCl2
chiral diol
1
OH
(10 mol% each)
+
Oi Pr
MS 4A, -30 o C
O
7
O
Oi Pr
H
6
O
Chiral Titanium Lewis Acids
OH
Titanium-based Lewis acid complexes have
undergone considerable development as vehicles for
effecting asymmetric catalytic ene-reactions.
Me
Both (R)- and (S)-BINOL 1 1 are commercially
available in optically pure forms. It is important to point
out that the catalyst derived from (S)-BINOL 11 affords
the (S)-alcohol and the catalyst derived from (R)-BINOL
1 1 consistently affords the corresponding (R)-alcohol.
The high levels of enantiocontrol and rate
acceleration observed with the BINOL catalysts are
apparently due to the favorable influence of inherent
C2-symmetry and the higher acidity of BINOL
compared to those of aliphatic diols. Remarkable
enantioselectivities were observed with the use of
methylglyoxylate 1 3 instead of isopropylglyoxylate 6
(up to 99%, Scheme 5). The ene-reaction of
Ph
Ph
Ph
8
Mikami and coworkers have reported an extremely
efficient asymmetric glyoxylate ene-reaction catalyzed
by the titanium complexes (R)-1 2 a / b , prepared in situ
from diisopropoxytitanium dihalide and optically active
binaphthol (BINOL) 1 1 in the presence of MS 4A
(Schemes 4 and 5) [38]. Ene-reaction between
methylenecyclohexane 1 and isopropylglyoxylate 6 in
the presence of 10 mol% of a chiral catalyst derived
from a chiral diol and TiCl2(i-PrO)2 afforded the
corresponding ene-products (Scheme 4). The best
result was obtained with the use of a chiral catalyst
derived from (R)-BINOL 1 1 and TiCl 2(i-PrO)2 (86% ee
and 82% yield). The discovery of the BINOL-derived
chiral catalyst was made after the screening of various
chiral catalysts derived from optically active diols [38].
Tetra(alcoxy)titanium complexes provided for the
expedient synthesis of chiral catalyst structures by
exploiting the facile substitution of monodentate
alkoxide ligands for chelating optically active diols.
Among the numerous chiral Ti(IV) complexes that are
available from optically active chelating diols, Lewis
acids derived from Binaphthol 1 1 (BINOL) have
undergone the largest degree of development as
catalysts for the ene-reaction (Scheme 4).
OH
Ph
9
OH
34% ee (72%)
4% ee (60%)
Ph
O
Ph
OH
Ph
OH
OH
OH
Me
OH
O
10
Ph
44% ee (70%)
11
Ph
86% ee (82%)
ee's determined by lanthanide induced shift NMR
measurement with (+)-Eu(DPPM) 3 after conversion
to the α-methoxy ester
Scheme 4.
methylenecyclohexane 1 with methylglyoxylate 1 3 in
the presence of 10 mol% of catalyst (R)-1 2 a afforded
the ene-product 1 4 in 83% yield and 97% ee. The use
of 5 mol% of catalyst (R)-1 2 b afforded the same
product in 89% yield and 98% ee. Especially
noteworthy is the difference in asymmetric catalysis
between the dibromo and dichloro catalysts. The
dibromide catalyst is superior to the dichloride in both
the reactivity and enantioselectivity for the glyoxylate
ene-reaction involving a methylene hydrogen shift in
particular (Scheme 5).
On the other hand, the dichloride catalyst is lower in
reactivity but superior in enantioselectivity for certain
glyoxylate ene-reactions involving a methyl hydrogen
shift (Scheme 6).
However, due to limiting reactivity of the catalystglyoxylate complex, only nucleophilic 1,1-disubstituted
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
309
O
(R)-12
+
OH
OMe
H
CH 2Cl 2
13
O
OMe
MS 4A
14
1
X = Cl (10 mol%), 82% (97%ee)
O
X = Br (5 mol%), 89% (98% ee)
(R)-12
O
OH
(10 mol%)
Me
Me
+
13
O
Me
16
(R)-12
OMe
OMe
Ph
CH 2Cl 2
13
O
X = Cl, 72% (95% ee)
X = Br, 87% (94% ee)
OH
(1 mol%)
H
17
O
MS 4A
O
Ph
Me
CH 2Cl 2
15
+
OMe
OMe
H
18
O
MS 4A
X = Cl, 97% (97% ee)
X = Br, 98% (95% ee)
Me
Et
Et
19
+
OMe
H
O
OH
(R)-12
OMe
+
Et
OMe
Et
CH 2Cl 2
13
Me
OH
O
20
O
MS 4A
21
O
X = Cl (10 mol%) 89 (94%ee) : 11 (>90%ee) 68%
X = Br (5 mol%)
91 (98% ee) : 9 (>90%ee) 73%
The catalyst:
X
O
Ti
O
X
(R)-12a, X = Cl
(R)-12b, X = Br
Scheme 5.
olefins can be employed. In the reaction of mono- and
1,2-disubstituted olefins, no ene product was
obtained. This limitation has been overcome by the use
of vinylic sulfides and selenides instead of mono- and
1,2-disubstituted olefins to afford the ene products
with
excellent
enantioselectivity
and
diastereoselectivity [39]. (E)-phenylsulfides (>98% E)
2 9 were used in the glyoxylate ene-reaction catalyzed
by (R)-1 2 a and provided the anti-diastereoisomers of
β-alkyl-α-hydroxy
esters
30
with
excellent
enantioselectivities (Scheme 7, Table 2). The anti
selectivity increases with the steric demand of the
vinylic substituent R (entries 1 and 4). It should be
noted that the anti-isomers are obtained in higher ee's
when compared to syn-isomers. In this same work the
authors reported that the use of (Z)-phenylsulfides
(>98% Z) affords good levels of diastereoselectivity for
the syn-isomers (entry 5).
310 Current Organic Chemistry, 2000, Vol. 4, No. 3
Et
Me
Me
(R)-1
22
+
O
OH
OMe
23
OH
OMe
+
Et
MS 4A
Me
OH
OMe
CH 2Cl 2
H
13
Luiz Carlos Dias
+
Me
24
O
OMe
Me
25
O
O
O
catalyst
23 %ee 24 %ee 25 %ee
(R)-12b, X = Br (5 mol%) 39 91
(R)-12a, X = Cl (10 mol%) 42 54
57 >98
4
4
Yield (%)
>90
93
68
Me
Me
26
Me
+
O
(R)-1
OH
Me
CH 2Cl 2
MS 4A
OMe
Me
OMe
Me
27
Me
+
O
OMe
Me
H
13
OH
28
O
O
catalyst
27 %ee
(R)-12, X = Br (5 mol%)
(R)-12, X = Cl (10 mol%)
83
87
92
28 %ee
17
13
>98
Yield (%)
91
73
Scheme 6.
H
R
Table 2.
OH
( E )-Phenylsulfides in Glyoxylate
reactions
Ene-
OMe
PhS
PhS
Me
30
(R)-12a
(E)-29
+
R
O
anti
(10 mol%)
+
O
CH 2Cl 2
OMe
H
13
O
OH
MS 4A
-30 oC
OMe
PhS
31
R
O
syn
R
yield (%)
anti
% ee
syn
% eea
1
Me
91
45
> 99
55
78
2
Et
88
81
> 99
19
84
3
nBu
90
91
> 99
9
> 90
4
iBu
94
95
> 99
5
> 90
5b
iBu
93
12
> 90
88
69
a. ee’s determined by conversion to the corresponding MTPA esters
b. (Z)-phenylsulfide was used as the ene component
The catalyst:
O
O
Cl
Ti
Cl
(R)-12a
Scheme 7.
entry
This protocol was used as the initial step in a catalytic
enantioselective synthesis of (R)-(-)-ipsdienol 3 7 , an
aggregation pheromone of bark beetles (Scheme 8)
[39].
Ene-reaction of prop-1-en-2-yl-phenylsulfide 3 2
and selenide 3 3 in the presence of 0.5 mol% of chiral
catalyst (R)-1 2 a afforded (R)-hydroxyvinylsulfide 3 4
and selenide 3 5 , respectively, in good yields and
excellent enantioselectivities. After a number of steps,
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
PhY
Me
(R)-12a
32, Y = S
33, Y = Se
OH
(0.5 mol%)
OMe
PhY
+
CH 2Cl 2
O
O
MS 4A
-30 oC
OMe
H
34, Y = S, 94% (> 99% ee)
35, Y = Se, 95% (> 99% ee)
13
O
TBS
OH
OMe
PhSe
35
O
1. TBSOTf
2,6-lutidine
O
Me
PhSe
2. DIBALH
3. Ph 3P=CMe2
57% 3 steps
Me
36
OH
1. CH2 =CHMgBr-NiCl2 (dmpe)
(3 mol%), 78%
Me
Me
2. TBAF, 94%
(R)-(-)-ipsdienol
The catalyst:
37
O
O
Cl
Ti
Cl
(R)-12a
Scheme 8.
the ene-product 3 5 was converted to (R)-(-)-ipsdienol
37.
OH
type 2,4
CHO
Mikami and coworkers have also described the first
example of catalysis of carbonyl-ene cyclizations using
the modified BINOL-Ti complexes (R)-BINOL-TiX2 (X =
ClO4 or OTf), easily prepared by the addition of silver
perchlorate or silver triflate to the corresponding
BINOL-TiCl2 complex (R)-12a [40]. The ene cyclization
of the α-alcoxy-aldehydes 3 8 in the presence of 20
mol% of catalysts (R)-1 2 afforded the product 3 9 in
moderate yields and good
enantioselectivities
(Scheme 9, Table 3) [41,42].
cat.*
O
O
Me
R
R
MS 4A
R
R
38
39
absolute configuration
deduced by the modified
Mosher method
The catalysts:
X
O
The catalyst (R)-1 2 c prepared from BINOL-TiCl2
with 2.0 equivalents of AgClO4 was observed to give a
higher level of enantioselectivity when compared to
that obtained with the dichloride (R)-12a [42] (entries
1 and 3, and 4 and 5). The BINOL-derived titanium
triflate (R)-1 2 d is shown to give comparably high levels
of enantiomeric excess (92% ee, entry 2).
Ti
O
X
(R)-12a, X = Cl
(R)-12c, X =ClO 4
(R)-12d, X = OTf
Scheme 9.
311
312 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
Ene Cyclization of α-Alcoxyaldehydes
Table 3.
entry
R
catalyst
% eea
yield (%)
1
H
(R)-12c
91
43
2
H
(R)-12d
92
40
3
H
(R)-12a
88
64
4
Me
(R)-12c
82
40
5
Me
(R)-12a
67
27
a. ee's determined by 1H NMR analysis of MTPA esters
The 6-membered cyclization of the homoallylic ether
4 0 afforded the trans-alcohol 4 1 as the major isomer in
moderate enantiomeric excess (Scheme 10). The use
of an α-alkoxy-aldehyde affords much better results in
terms of yields and enantioselectivities, when
compared to the same reaction with the carbonanalogue [42,43].
axial orientation is used to explain the trans/cis
selectivity (Scheme 11). The equatorial orientation,
which depends on the steric bulk of the chiral Lewis
acid complexes, leads to the trans alcohol 4 1 , while the
axial orientation of the complexed carbonyl group leads
to the cis alcohol 4 2 .
The Mikami group has also described the
asymmetric catalytic fluoral-ene-reaction [44]. The chiral
titanium catalyzed ene-type reaction with fluoral
provides an efficient approach for the asymmetric
synthesis of biologically important fluorine containing
compounds in high enantiomeric excess (Scheme 12).
This reaction provides the homoallylic alcohol 4 4 (major
isomer), and the allylic alcohol 4 5 as a by-product in
high ee’s independent of the solvent and halide ligand
of the BINOL-Ti catalyst (Table 4, entries 1-3).
OH
type 3,4
+
Me
CHO
X
Me
cat.*
20 mol%
X
40
Me
44
(10-20 mol%)
1
H
solvent
MS 4A
OH
CY3
CY3
45
absolute configuration
determined by the
Mosher method
43
OH
Me
X = CH2
X
> 95% ee
+
O
trans-41
+
MS 4A
CY3
(R)-12
OH
cis-42
The catalyst:
(R)-12c, 69 (55% ee) 31 (64% ee) (66%)
(R)-12a, 59 (21% ee) 41 (41% ee) (54%)
X=O
(R)-12c, 80 (84% ee) 20 (74% ee) (50%)
O
(R)-12a, 47 (70% ee) 53 (79% ee) (73%)
O
Scheme 10.
O
Me MLn*
H
O
trans-alcohol 41
H
MLn*
O
Me
O
H
cis-alcohol 42
Scheme 11.
X
(R)-12a
(R)-12b
A bicyclic transition state model in which the
complexed carbonyl group adopts an equatorial or an
H
X
Ti
Scheme 12.
It is interesting to note that when chloral is used as
the enophile, lower ee’s and more allylic alcohols are
observed (entries 4-6). The authors estimated the ene
reactivity of trihaloacetaldehydes in terms of the
balance of the LUMO energy level versus atomic
charges of reaction sites of the enophile components.
Based on semiempirical and ab-initio molecular orbital
calculations of trihaloacetaldehyde/H+ complexes, the
authors suggest that the lower LUMO energy is
responsible for the higher reactivity of fluoral/H + , giving
the homoallylic alcohols 4 4 . The chloral/H+ complex
has a higher LUMO energy but bears the greater
Chiral Lewis Acid
Table 4.
Current Organic Chemistry, 2000, Vol. 4, No. 3
313
Asymmetric Fluoral Ene-reaction
X
solvent
44 : 45a
yield (%)
Br
CH 2Cl 2
79 (>95% ee) : 21 (>95% ee)
95
Cl
CH 2Cl 2
76 (>95% ee) : 24 (>95% ee)
93
3
Cl
toluene
79 (>95% ee) : 21 (>95% ee)
82
4
Br
CH 2Cl 2
48 (45% ee) : 52 (80% ee)
40
Cl
CH 2Cl 2
52 (34% ee) : 48 (66% ee)
49
Cl
toluene
63 (11% ee) : 37 (66% ee)
35
entry
Y
1
2
F
5
Cl
6
a. ee's determined by 1H NMR analysis of (S)-(-)- and (R)-(+)-MTPA esters
positive charge at the carbonyl carbon and is therefore
more reactive in terms of the cationic (Friedel-Crafts
type) reaction affording the allylic alcohols 4 5 [44].
Later, the same authors described a new type of
asymmetric catalyst for carbonyl ene-reactions with
methylglyoxylate 1 3 (Scheme 13) [45]. While trying to
isolate BINOL-TICl 2 catalyst (R)-1 2 a , the binaphthol
(BINOL)-chiral titanium µ-oxo complex (R)-4 6 was
obtained accidentally upon azeotropic removal of
isopropanol with toluene from a solution of BINOL 1 1
and TiCl2(OiPr)2 after filtration of MS 4A. Glyoxylate
ene-reaction with α-methylstyrene 1 7 in the presence
of 0.2 mol% of catalyst (R)-4 6 afforded α-hydroxyester
1 8 in 88% isolated yield and 98.7% ee (Scheme 13).
The same reaction with methylenecyclohexane 1 in
the presence of 1 mol% of catalyst (R)-4 6 afforded the
corresponding ene-product 1 4 in > 99% ee and 68%
isolated yield.
The dimeric nature of µ-oxo complex 4 6 in solution
was confirmed by vapor pressure osmometry molecular
mass measurements, NMR and infrared spectroscopy,
O
+
Ph
Me
A highly selective asymmetric ene-reaction was
reported in 1995 by Nakai and collaborators [46], that
described the preparation and use of an µ-oxo catalyst
similar to that used previously by Mikami and Terada
[45]. This dimeric catalyst was prepared by mixing (R)OH
catalyst 46
OMe
H
OMe
Ph
CH 2Cl 2, -30 o C
13
17
as well as by mass spectrometry [45]. They found that
this catalyst exhibits a positive non-linear effect in the
glyoxylate ene-reaction. The homochiral catalyst,
prepared from pure (R)-BINOL 11 possessed catalytic
activity nine times superior to that of the heterochiral
catalyst, prepared from racemic BINOL. The molecular
mass of the µ-oxo complex prepared from
enantiomerically pure (R)-(R)-BINOL was shown to be
concentration-dependent, ranging from 682 (1.5g/L
solution) to 786 in 12g/L solution [calc. 696.4 for
(C20H12O2TiO)2] while with racemic (R)-(S)-BINOL it
remained unchanged upon dilution, which indicates a
greater stability of the meso µ-oxo dimer prepared from
racemic BINOL. These results led to the conclusion
that the heterochiral complex is relatively unstable in
solution and dissociates to the monomeric form that is,
probably, the species that is responsible for catalysis.
O
88%
O
98.7% ee
The catalyst:
OH
OH
(R)-11
Scheme 13.
1. Ti(Oi Pr) 2Cl 2
MS 4A
2. azeotropic
removal
of i PrOH
3. filtration of
MS 4A
O
O
Ti
O
O
Ti
O
(R)-46
O
18
314 Current Organic Chemistry, 2000, Vol. 4, No. 3
Ph
17
O
Luiz Carlos Dias
Me
OH
catalyst 46
OMe
+
solvent, -30
Ph
oC
OMe
18
O
H
13
O
OH
Ti(O iPr) 4
Oi Pr
O
Ti
OH
O
toluene
Oi Pr
(R)-47
(R)-11
The catalyst:
1. H2O, ∆
O
O
Ti
2. azeotropic
removal
of iPrOH
O
O
Ti
O
O
(R)-46
dark brown solid
Scheme 14.
BINOL 1 1 and Ti(OiPr)4, followed by hydrolysis and
complete azeotropic removal of isopropanol. The
structure of this complex 4 6 was confirmed by 1H-NMR
spectrum and molecular weight measurements. Using
the µ-oxo complex 4 6 , the ene-reaction of αmethylstyrene 17 with methylglyoxylate 1 3 was
studied (Scheme 14, Table 5). Solvent also plays a very
important role, since reaction in dichloromethane
afforded higher enantioselectivities and chemical yields
(entries 1 and 2) than the same reaction in toluene
(entry 3).
Table 5.
Glyoxylate Ene-reaction Catalyzed by
(R)-46
entry
catalyst (mol%)
solvent
yield (%)
% ee (conf.)a
1
5
CH 2Cl 2
93
98 (R)
2
20
CH 2Cl 2
92
97 (R)
3
20
toluene
64
95 (R)
a. ee's determined by 1H NMR analysis of (+)- and (-)-MTPA esters
The authors observed a positive non-linear effect,
with the maximum % ee for the product (98% ee) being
obtained using BINOL 11 with only 55-60% ee. This
result is consistent with the results described earlier by
Mikami and Terada [45]. This suggests that the source
of asymmetric amplification involves both mutual
enantiomer recognition, the predominant formation of
the heterochiral (meso) dimeric species over the
homochiral dimeric species, and a higher catalytic
activity of the latter species [46].
Mikami and colleagues designed a new class of
chiral titanium catalysts from binaphthyls with larger
dihedral angles derived from 6-Br-BINOL (Scheme 15,
Table 6) [47]. It is believed that the compression of the
internal bond angle X-Ti-X (φ´) would lead to higher
levels of enantioselectivity based on a greater shielding
effect over the enantioface of the glyoxylate by the
halide ligands. Glyoxylate ene-reaction with αmethylstyrene 1 7 in the presence of 0.05 mol% of
catalyst (R)-48a (X=Cl) in toluene as solvent gave the
corresponding ene product with 99% ee in 99% yield
(entry 4). The same reaction in CH2Cl2 afforded the
ene-adduct in 94% yield and 97.5% ee (entry 3) The
catalyst (R)-4 8 b (X=Br) affords the glyoxylate ene
product of methylenecyclohexane 1 in CH2Cl2 as
solvent with more than 99% ee in 82% yield (entry 5).
The use of titanium catalyst (R)-4 9 , derived from
binaphthyl ligand and a bulky trifylamine moiety
afforded no enantioselectivity (entry 1) [47]. As can be
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
seen from the example in entry 2, a certain level of
positive non-linear effect is observed in these
reactions. The 6-Br-BINOL-derived titanium complex
(R)-4 8 a with 70.4% ee affords the glyoxylate-ene
product in 85% ee.
diisopropoxytitanium dihalides (Scheme 16, Table 7)
[48,49].
OH
Me
OMe
Me
1 mol% (R)-48a/b
+
Me
+
OH
catalyst
OMe
CH 2Cl 2
OMe -30 o C
H
13
O
O
syn-50
+
0 o C, toluene
O
315
OH
MS 4A, 2h
OMe
OMe
H
O
13
O
O
Me
O
anti-51
The catalysts:
- diastereoselectivity determined by 1H NMR analysis
Br
- absolute stereochemistry determined by the Mosher method
The catalyst:
O
X
Br
Ti
O
X
(R)-48a, X=Cl
(R)-48b, X=Br
Br
O
TiX2
O
Tf
X
N
Scheme 16.
φ
Ti
N
X
Tf
(R)-49
Scheme 15.
Table 6.
Glyoxylate ene-reaction
48 and 49
(R)-48a, X=Cl
(R)-48b, X=Br
Br
catalyzed
by
ene component
This reaction affords syn-α-hydroxy-β-methyl esters
5 0 in good enantioselectivities (entries 1-5). The same
reaction catalyzed by (R)-1 2 (X = Cl, Br) provides only a
modest level of enantioselectivity, although with good
syn-diastereoselectivity (entries 6-8). As can be
observed, the sense of asymmetric induction is the
same as the reactions catalyzed by (R)-1 2 a , with (R)catalysts providing (2R)-2-hydroxy esters.
Table 7.
Glyoxylate Ene-reaction
tuted Olefins
of
Trisubsti-
ene component
Ph
1
Me
Me
Me
17
entry
ene
catalyst
yield (%)
% eeb
solvent
1
17
49
69
0
CH 2Cl 2
2a
17
48a
92
85
CH 2Cl 2
3
17
48a
94
97.5
CH 2Cl 2
4
17
48a
99
99
toluene
5
1
48b
82
> 99
CH 2Cl 2
a. 70.4% ee of catalyst was used
b. enantiomeric excess determined by chiral HPLC analysis
The Mikami group has also described a diastereoand enantioselective carbonyl ene-reaction of
methylglyoxylate 1 3 with trisubstituted
olefins
catalyzed by a chiral titanium complex (R)-4 8 a / b (1
mol%)
derived
from
6-Br-BINOL
and
Me
Me
53
52
54
Me
syn (% ee)a : anti
entry
(R)-cat.*
ene
solvent yield (%)
1
48a
52
CH 2Cl 2
44
93 (81) : 7
2
48a
52
toluene
60
93 (88) : 7
3
48b
52
toluene
84
94 (89) : 6
4
48b
53
toluene
89
97 (87) : 3
5
48b
54
toluene
63
94 (61) : 6
6
12a
52
toluene
61
93 (69) : 7
7
12b
53
toluene
80
96 (60) : 4
8
12b
54
toluene
56
95 (2) : 5
a. ee determined by 1H NMR comparison of (S)- and (R)-MTPA derivatives
316 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
OH
R3 Si
OMe
Me
56
R3 Si
Me
(10 mol%)
O
OMe
59
Me
57
O
+
-30 oC, 2h
OMe
Me
R3 Si
OMe
58
OTMS
R3 Si
OTMS
O
O
+
H
13
R3 Si
OMe
CH 2Cl 2
+
OH
OH
(R)-12a
55a, R =Ph
55b, R=Me
O
+
OMe
60
O
O
The catalyst:
O
O
Cl
Ti
Cl
(R)-12a
Scheme 17.
Interestingly, maximum enantioselectivity and better
yields were achieved with the use of a less polar
solvent such as toluene (entries 2-4). This solvent
effect is not observed for catalyst (R)-1 2 .
Table 8. Ene-reactions with Allylsilanes
entry
1
2
R
Ph
Me
product
yield (%)
% eea
56
45
95
57
0
-
58
0
-
59
0
-
60
0
-
56
45
94
57
21
86
58
17
94
59
10
92
60
7
92
a. ee's and absolute configuration determined by MTPA esters
Mikami and Matsukawa described that chiral catalyst
(R)-BINOL-TiCl2 1 2 a catalyzed the reaction of
glyoxylate esters with methallylsilanes 5 5 to afford enetype products 5 6 (allylic silanes) as the major products,
instead of the expected Sakurai-Hosomi derived
products 5 7 (Scheme 17, Table 8) [50].
Reaction of methallyl(triphenyl)silane 5 5 a (less
reactive) or methallyl(trimethyl)silane 55b
with
methylglyoxylate 1 3 either in toluene or CH2Cl2 as
solvents afforded the ene-type allylic silane 5 6 as the
major product (entries 1 and 2).
The products obtained in the reaction of
methallyl(trimethyl)silane 5 5 b are converted to the
“usual”
Sakurai-Hosomi
product
57
after
protodesilylation in high enantiomeric excess (92%
ee). Desilylation of the crude mixture of 5 6 -6 0 with 3N
HCl in MeOH afforded product 5 7 in 75% yield and
92% ee (Scheme 18). In this same work, the authors
reported that allyltrimethylsilane affords the "usual"
product instead of the ene product, that may reflect the
less-ene reactivity of monosubstituted olefins [50].
In a very interesting paper, Mikami and coworkers
described the first example of an asymmetric catalytic
formaldehyde-ene-reaction with symmetrical prochiral
bicyclic olefins [51]. The best result was obtained in the
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
317
OH
Me3 Si
OMe
OH
56
O
H
H
+
Me
61
OTBS
+
OH
Me3 Si
OMe
59
(R)-12b
(20 mol %)
9α
H
H
CH2 Cl2 , MS 3A
-30 oC, 61%
11
76% ee
O
H
+
(11S)-∆6(9a) -63
H
62
+
Me
57
OTBS
O
OH
OMe
6
O
HO
+
Me
OTMS
Me3 Si
6
7
OMe
H
60
H
O
11
OTMS
Me3 Si
OTBS
OMe
(11R)-64
58
O
88 : 12
ee determined by 1 H NMR
analysis of MTPA esters
3N HCl
The catalyst:
MeOH
OH
OMe
O
Me
57
Br
Ti
O
O
75% yield
92% ee
Br
(R)-12b
Scheme 18.
reaction of the symmetrical olefin 6 1 with formaldehyde
in the presence of (R)-BINOL-TiBr2 1 2 b (20 mol%) and
MS 3A in CH2Cl2 at -30 oC, affording adduct (11S)∆ 6(9α) -6 3 (prostaglandin numbering) as the major
isomer in 61% yield and 76% ee (Scheme 19).
The analogous reaction with chiral bicyclic olefin 6 5
in the presence of (R)-BINOL-TiBr2 catalyst 1 2 b ,
formally completes the total synthesis of a potent
analogue of (3-oxa)-isocarbacyclin 6 8 , a therapeutic
agent with promising physiological activities (Scheme
20). Treatment of chiral bicyclic olefin 6 5 with
formaldehyde 6 2 in the presence of (R)-BINOL-TiBr2
1 2 b and MS 3A in CH2Cl2 at -30 oC led to the formation
of formaldehyde-ene adduct 6 7 in 90% ∆ 6(9α)
regioselectivity (Table 9). After a sequence involving
few steps, the ene adduct 6 7 was converted to (3oxa)-isocarbacyclin 6 8 .
Scheme 19.
The corresponding (S)-BINOL-Ti catalyst afforded
the opposite ∆ 6-regioisomer 6 6 in 80% regioselectivity
(entry 2) while the use of Me2AlCl (entry 1) led to a
much lower regioselectivity in poor chemical yield
(Table 9).
Table 9.
Formaldehyde Ene-reaction
Entry
Lewis acid
ratio (66 : 67)
yield (%)
1
Me2AlCl
49 : 51
24
2
(S)-cat.
80 : 20
65
3
(R)-cat.
10 : 90
64
318 Current Organic Chemistry, 2000, Vol. 4, No. 3
H
Luiz Carlos Dias
O
H
Me
+
(R)-12b
(20 mol %)
H
H
CH 2Cl 2, MS 3A
62
OTBS
TBSO
Me
-30 oC, 64%
Me
65
OH
OH
6
9α
H
H
H
H
Me
Me
+
OTBS
TBSO
Me
Me
66
TBSO
OTBS
Me
Me
67
∆6(9a)
O
CO2 H
(3-oxa)-isocarbacyclin
H
H
Me
OH
HO
Me
Me
68
Scheme 20.
The same methodology has been applied before by
the same group in the synthesis of potential
intermediates for the synthesis of isocarbacyclin
analogues (Scheme 21) [52]. Reaction of the bicyclic
ene substrate 6 9 with methyl glyoxylate 1 3 in the
presence of 10 mol% of catalyst (R)-1 2 b (X=Br) and
MS 4A proceeds with high diastereo-(96% de) and
enantioselectivities (98% ee).
carbonyl ene-reaction between conjugated ynal 7 1
and formylacrylate 7 2 with methylenecyclohexane 1
and methylenecyclopentane 7 3 catalyzed by (R)BINOL-TiCl2 1 2 a afforded the corresponding ene
products 7 4 and 7 5 , respectively, in good yields and
good enantioselectivities (Scheme 22, Table 10) [53].
O
H
G
+
OH
catalyst
OH
Me
Me
(R)-12b
(10 mol %)
69
Me2 HSiO
+
O
MS 4A
-30 oC, 100%
OMe
Me
OMe
H
13
O
CH 2Cl 2
MS 4A
rt, 1h
O
Me
n
1, n = 1
73, n = 0
n
CO2 Me
74, n = 0
75, n = 1
The catalyst:
70
96% de
(98% ee)
Ti
O
Scheme 21.
The Mikami group described an extension of the
glyoxylate ene-reaction to the use of 3-formylpropiolate
7 1 and (E)-3-formylacrylate 72 [53]. Asymmetric
Cl
O
OSiHMe2
(R)-12a
Scheme 22.
Cl
Chiral Lewis Acid
Table 10.
Current Organic Chemistry, 2000, Vol. 4, No. 3
Asymmetric
Carbonyl
Catalyzed by (R)-12a
entry
enophile
Ene-reactions
n
yield (%)
% eea
0
85
87
1
70
94
0
80
72
1
60
86
In this same work, the authors reported that the
double asymmetric induction with chiral bicyclic olefin
6 5 catalyzed by (R)-1 2 a serves as a key step for the
total synthesis of new potent analogues of
isocarbacyclin. Reaction of 6 5 with aldehyde 7 1 in the
presence of 20 mol% of catalyst (R)-1 2 a afforded the
corresponding ene product 7 8 in 99% ∆ 6(9α)
regioselectivity and 96% ee (Scheme 24). The eneadduct 7 8 was converted to α-allenyl isocarbacyclin
derivative 8 0 after a sequence involving few steps.
O
1
H
2
CO2 Me
71
In 1996, Faller and Liu reported an interesting
Chloral ene-reaction catalyzed by a Ti(OiPr)2Cl2/racemic
BINOL poisoned with an inactive enantiopure catalyst,
diisopropyl-D-tartrate/Ti(OiPr)2Cl2 [54]. This strategy is
based on the selective deactivation of a racemic
catalyst by a chiral molecule (Scheme 25).
O
3
H
CO2 Me
4
72
a.ee's determined by analysis of the corresponding (S)- and (R)-MTPA
esters
Reaction between isobutylene 1 5 and chloral 8 1 in
the presence of Ti(OiPr)2Cl2/(S)-BINOL as catalyst,
afforded the homoallylic alcohol 8 2 and the allylic
alcohol 8 3 in a 2:1 ratio in 24% ee and 66% ee,
respectively (Scheme 26, Table 11). The ee for allylic
alcohol 8 3 can be improved to 88% after a single
recrystallization step from pentane.
It is interesting to see that formylpropiolate 7 1
affords similar levels of enantioselectivity to that
observed with methylglyoxylate 1 3 (entry 2). The
authors reported also the asymmetric desymmetrization
of prochiral olefin 6 1 with aldehyde 7 1 as a model
system for the synthesis of isocarbacyclin analogues
(Scheme 23). The product ∆ 6(9α) -7 6 was obtained with
high regio- and enantioselectivity in the presence of
(R)-1 2 a at room temperature [53].
H
H
61
OTBS
O
(R)-12a
(20 mol%)
CH 2Cl 2,-30 o C
MS 4A
Catalysts
prepared
in
situ
by
mixing
Ti(OiPr)2Cl2/racemic
BINOL
and
Ti(OiPr)2Cl2/
diisopropyl-(D)-tartrate poison were used to afford
homoallylic alcohol 8 2 as the major product, although
in low ee. The best ratio Ti(OiPr)2Cl2/diisopropyl-(D)OH
OH
4
4
6
6
CO2 Me
9α
+
H +
H
H
H
81%
OTBS
OTBS
71
CO2 Me
∆6(9α) -76
92 (89%ee)
:
The catalyst:
Cl
O
Ti
O
(R)-12a
Scheme 23.
CO2 Me
7
+
H
319
Cl
∆6 -77
8
320 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
O
H
H
+
(R)-12a
(20 mol%)
H
Me
71
TBSO
OTBS
65
Me
Me
78%
OH
OH
4
4
6
6
CO2 Me
H +
+
Me
TBSO
OTBS
CO2 Me
7
9α
H
CH 2Cl 2, 0 o C
MS 4A
CO2 Me
Me
78
H
H
Me
Me
TBSO
OTBS
Me
79
Me
∆6(9 α)-78
: ∆6 -79
99 (96% ee) :
1
C
H
H
CO2 Me
- stereoselectivity at C4 determined by
LIS analysis using (+)-Eu(hfc)3
isocarbacyclin analogue
Me
HO
OH
Me
80
- ∆6(9 α) regioselectivity determined
by 1H MMR analysis
Me
Scheme 24.
tartrate was found to be 1:3, respectively (entries 2-4).
The observed enantioselectivities as well as the
regioselectivities are greater than those with catalyst
prepared from enantiomerically pure BINOL 11 (entry
1). Both the homoallylic alcohol 8 2 and allylic alcohol
8 3 are formed with modest enantioselelectivities.
These results are consistent with the presence of a
heterodimer as the active complex and it is believed
that Ti/(R)-BINOL has been effectively deactivated
upon forming a Ti2/(R)-BINOL/(D)-DIPT complex [54].
A catalytic enantioselective carbonyl cyclization has
been used by the Mikami group in the synthesis of the
A ring of a hybrid of Vitamin D 19-nor-22-oxa D3
analogue 8 8 , which shows significant transactivation
activity, as shown by its great ability to transactivate a rat
25-hydroxy vitamin D3-24-hydroxylase gene [55]. The
Concept of asymmetric activation
(R)-Cat*---Deact*
(R)-Cat*
+
Deact*
deactivated
less reactive
ene
(S)-Cat*
chiral
deactivator
(S)-Cat*
more reactive
species
Scheme 25.
enophile
ene-product
X% ee
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
OH
Me
Me
Me
catalyst
(R)-12a
15
+
O
cyclohexanodiols 85 and 86 with high enantiomeric
purity (R,R stereochemistry as a geometrical mixture
[(trans,Z):(trans,E):(cis,Z/E) = 75:23:2]) in 65% yield
(Scheme 27). It is interesting to note that the use of (S)BINOL-TiCl2 catalyst gave the four possible isomers
(trans,Z):(trans,E):(cis,Z):(cis,E) in a ratio of 32:8:32:28,
in an apparently mismatched reaction.
CCl3
82
CH 2Cl 2
-20 o C
+
Me
321
OH
MS 4A
H
CCl3
Me
Although obtained as a geometrical mixture, the
intermediates 85 and 86 were transformed to the
same single component 8 7 after removal of the MPM
group. Further transformation led to the hybrid
analogue of 19-nor-22-oxa-1α,25(OH) 2D3 8 8 [56].
CCl3
81
83
Scheme 26.
intramolecular 6-(2,4)-carbonyl-ene-reaction of (R)MPMoxy(benzyloxyethyl)hexenal 8 4 catalyzed by (R)BINOL-Ti catalyst (5 mol%) at room temperature
provided
the
desired
pseudo
C2-symmetric
Table 11.
Mikami's mechanistic rationale to explain the source
of asymmetric induction in these Lewis acid catalyzed
carbonyl-ene cyclizations involves a 6-memberedtransition state to afford the (trans,Z)-8 5 and (trans,E)-
Chloral Ene-reaction
entry
mol% catalyst
Ti(OR) 2Cl2/BINOL
poison (mmol)
Ti(OR) 2Cl2/(D)-DIPT
yield (%)
ratio
82 : 83
% eea
82 83
1
0.10 (S)
0.0/0.0
87
67 : 33
24 66
2
0.20 (rac)
0.1/0.2
53
88 : 12
30 20
3
0.20 (rac)
0.1/0.3
58
94 : 06
48 25
4
0.10 (S)
0.05/0.15
40
90 :10
33 9
a. ee determined by GC using a cyclodex-β chiral column
OH
OH
11
OBn
BnO
OBn
Cl2Ti(OiPr)2 (5mol%)
OHC
OMPM
MS 4A, CH2 Cl2
rt, 38 h, 65%
HO
(R)-84
OMPM
(trans,Z)-85
Me
O
HO
OMPM
(trans,E)-86
(75 : 23 : 2)
OH
Me
+ (cis, Z,E)
+
CAN
CH 3CN/H 2O
rt, 89%
Me
Me
OBn
BnO
OH
HO
88
HO
HO
Scheme 27.
OH
87
OH
87
322 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
8 6 depending on the balance of acyclic allylic 1,2-strain
and repulsion between the bulkly BINOL-Ti-catalyst
and benzyloxymethyl group (Scheme 28).
*LnTi
O
these reactions, including product structure, the olefin
geometry of the silyl enol ether products, and the
insensitivity of the observed diastereoselectivity to
enolsilane geometry (entries 3 and 4).
Table 12.
H
Ene-reactions with Enolsilanes
OBn
H
entry
OMPM
enolsilane
yield (%)
syn:anti
Z:E
ee (%)
75
-
-
> 99
67
-
95:05
> 99
58
98:02
94:06
99
54
98:02
94:06
99
OTBS
(trans,Z)-85
1
Me
OTMS
*LnTi
O
2
H
Prn
BnO
H
OMPM
OTMS
3
(trans,E)-86
Me
Me
Scheme 28.
OTMS
Addition of enolsilanes 8 9 to methylglyoxylate 1 3
in the presence of 5 mol% of catalyst (R)-4 7 affords the
homoallylic alcohols 9 0 in good yields and >99% ee
(Scheme 29, Table 12) [57].
OTMS
R1
H
(R)-47
(5 mol%)
+
CH2Cl 2
O
TMSO
OH
R1
90
H
R2
O
O
The catalyst:
O
Me
OMe
OMe
13
Me
Terminal enolsilanes afford the corresponding eneproducts in nearly perfect ee's (entries 1 and 2). As can
be seen from the results showed in Table 11, high
selectivity for the formation of Z-olefin is observed
(entries 2-4) and (E)- and (Z)-enolsilanes afford the
corresponding products with high syn-selectivity and
excellent ee's (entries 3 and 4).
R2
89
4
Oi Pr
Ti
O
Oi Pr
(R)-47
Scheme 29.
The Z-isomer is formed in high selectivity
(Z:E>94:06) and although not expected, an enereaction pathway nicely explains the characteristics of
The formation of syn-products from (E)-enolsilanes
is explained by invoking the transition-state assembly
9 2 (Scheme 30). This approach avoids the developing
1,3-transanular interaction between the pseudoaxial
alkyl substituent (R2) and TiL* in the alternative
transition state 9 4 . With (Z)-enolsilane, the closed
transition-state 9 3 avoids the developing gauche
interaction between the ester residue and the
enolsilane substituent (R2) in transition state 9 5 [57].
In 1997, Mikami and Matsunaga described a very
interesting strategy for asymmetric catalysis based on a
selective activation of one enantiomer of a racemic
catalyst by addition of a chiral activator (Scheme 31)
[58].
High enantioselectivities (up to 89.8% ee) were
obtained for the glyoxylate ene-reaction between αmethylstyrene 1 7 and n-butylglyoxylate 9 6 in the
presence of racemic BINOL-Ti(OiPr)2 4 7 (10 mol%)
when (R2)-BINOL 1 1 and (R2)-5-Cl-BIPOL were used
as chiral activators (Scheme 32, Table 13).
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
(E)-enolsilanes
R1
H
(Z)-enolsilanes
OSiR3
O
R3 SiO
OH
OMe
R2
R2
(Z)
R1
R3
R1
H
OMe
O
*LTi
TMSO
H
O
H
90
O
H
R2
H
TiL*
92
93
1,2-syn
R3 SiO
R3
H
R1
TMSO
H
R1
H
OMe
R2
OSiR3
H
R2
O
O
OMe
H
94
R1
OH
TiL*
R2
(Z)
1,3-Transanular
interaction
(R 2/TiL*)
O
91
*LTi
O
gauche interaction
(R2 /OMe)
1,2-anti
Scheme 30.
Concept of asymmetric activation
ene
(R)-Cat*
(R)-Cat*---Act*
+
activated
more reactive
Act*
(S)-Cat*
ene-product
X% ee
enophile
(S)-Cat*
Scheme 31.
Oi Pr
O
Ti
O
Ph
Me
OH
(10 mol%)
O +
+
On Bu
On Bu
Ph
activator
(R)-97
toluene, -30 o C
H
96
Oi Pr
(+/-)-47
17
(R2 )-BINOL
(R)-5-Cl-BIPOL
mol (%)
Yield (%) %ee
5.0
2.5
52
35
89.8
80.0
5.0
38
80.8
OH
OH
HO
Me
Me
OH
Cl
(R2 )-BINOL
Scheme 32.
O
O
activator
H
Me
Me
(R)-5-Cl-BIPOL
Cl
95
323
324 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
Oi Pr
O
Ti
O
Ph
Me
(R1 )-47
17
O
OH
(10 mol%)
+
On Bu
On Bu
+
H
96
Oi Pr
O
Ph
activator
(10 mol%)
(R)-97
O
toluene, -30 o C
The catalyst:
OiPr
H
O
O
Ti
O
98
O
OiPr
H
Scheme 33.
The authors propose a hexacoordinated and
monomeric (R)-BINOLate-Ti(OiPr)2/(R)-BINOL complex
9 8 as the active catalyst, based on 1H and 13C NMR
experiments. They confirmed the activation of (R1)BINOLate-Ti catalyst by addition of 1 mole equivalent of
(R2)-BINOL (Scheme 33, Table 13).
Table 13.
Glyoxylate
Ene-reaction
in
Presence of a Chiral Activator
the
activates the (R1)-BINOLate-Ti(IV) catalyst less than
(R2)-BINOL, affording the ene-product in 48% yield and
86% ee (entry 4). The use of (R2)-5-Cl-BIPOL affords
66% yield and 97.2% ee (entry 2). Even the addition of
racemic BINOL to the (R1)-BINOLate-Ti(IV) catalyst
afforded better levels of enantioselectivity (95.7% ee)
when compared to the use of catalyst (R1)-47 alone
(entries 1 and 5, Table 13) [58].
OH
entry
activator
time (min)
yield (%)
% eea
Ph
1
None
60
2
(R 2)-5-Cl-BIPOL
60
20
66
94.5
O
97.2
3
(R 2)-BINOL
60
82
96.8
4
(S2)-BINOL
60
48
86.0
5
(+/-)-BINOL
60
69
95.7
a. ee's determined by HPLC (Daicel Chiral AS column)
Me
17
R1*
+
OH
R2*
OH
OH
On Bu
+
H
Using only
catalyst, the
reaction afforded the glyoxylate-ene product in 20%
yield and 94.5% ee (entry 1). When (R2)-BINOL was
added, the ene-product was obtained in 82% yield and
96.8% ee (entry 3). These results demonstrated that
racemic BINOLate-Ti(IV) and a half-mole equivalent of
(R2)-BINOL are converted to the (R1)-BINOLateTi(IV)/(R2)-BINOL complex 9 8 . The use of (S2)-BINOL
(R)-97
toluene, 0 o C
O
Me
The catalyst:
Cl
H
(R1)-BINOLate-Ti(IV)
Ph
Ti(O iPr) 4
On Bu
96
OH
Me
OiPr
H
O
O
Ti
O
Me
H
Cl
Me
Scheme 34.
O
OiPr
H
O
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
In this same line, also in 1997, Mikami and coworkers
described a "smart" self-assembly of
highly
enantioselective catalyst from pre-catalyst and neutral
ligands. These catalysts (10 mol%) were evaluated in
the glyoxylate-ene reaction with α-methylstyrene 1 7
(Scheme 34, Table 14) [59].
Table 14.
R1*(OH) 2
Ph
1
Me
Me
2
The combination of (R)-BINOL and the less acidic
(R)-TADDOL with Ti(OiPr)4 in a molar ratio of 1:1:1 forms
a new catalytic system that affords the ene-product 9 7
in 50% yield and 91% ee (entry 1). Combination of (R)BINOL with the more acidic (R)-5-Cl-BIPOL and
Ti(OiPr)4 in a 1:1:1 ratio leads to a new catalyst system
Self Assembly of Enantioselective Catalysts
entry
Me
Me
O
R2*(OH) 2
Yield (%)
% eea
50
91
0
-
66
97
-
13
75
-
20
95
Ph
OH
OH
OH
OH
O
Ph
Ph
Ph
Ph
O
OH
-
OH
O
Ph
Me
Ph
Cl
3
Me
OH
OH
Me
OH
OH
Cl
Me
Me
Cl
4
Me
OH
Me
OH
Cl
Me
5
OH
OH
a. ee's determined by HPLC (Daicel CHIRALPAK AS column)
325
326 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
that affords the (R)-ene-product 97 in 66% yield and
97% ee (entry 3). The use of (R)-TADDOL/Ti(OiPr)4
affords no product (entry 2), the (R)-5-ClBIPOL/Ti(O iPr)4 affords the (R)-ene-product 9 7 in 13%
yield and 75% ee (entry 4) and (R)-BINOL/Ti(OiPr)4
affords the ene-product 9 7 in 20% yield and 95% ee
(entry 5) [59].
In a related paper, the ene-reaction of αmethylstyrene 17 with n-butylglyoxylate 96 was
catalyzed by complexes 1 0 0 a -d prepared from (R)BINOL-Ti(OiPr)2 4 7 and several conformationally
flexible ortho-substituted 2,2’-biphenols (BIPOL) 9 9 a d or (R)-BINOL 1 1 [60] (Scheme 35, Table 15).
0.1 eq.
cat.
Ph
OH
toluene
0 oC
Me
17
+
97
O
O
R
Oi Pr
O
R
HO
toluene
+
Ti
Entry
Second Ligand
Time (h)
% eea
1
none
7
93.2
2
(R)-BINOL 11
2
91.6
3
BIPOL 99a
2
95.4
4
BIPOL 99a
7
94.8
5
Cl 4BIPOL 99b
2
96.7
6
Br4BIPOL 99c
2
96.3
7
tBu BIPOL 99d
4
2
97.3
a. ee's determined by HPLC analysis (DAICEL Chiralpak AD)
absolute configuration
based on literature data
On Bu
O
with
Ph
H
96
Ene-reaction of α-Methylstyrene
n-Butyl Glyoxylate
On Bu
18-33%
O
Table 15.
HO
Oi Pr
0 oC
The authors also observed that the addition of
cheap and readily available non-chiral biphenols to (R)BINOL-Ti(OiPr)2 complex formed a new catalytic
species. Based on NMR experiments the authors
proposed that BIPOL ligand 9 9 a -d reacts with the
optically active BINOL-Ti(OiPr)2 pre-catalyst leading to
complexes 1 0 0 a -d as the active enantioselective
catalyst.
Non-substituted BIPOL 99a (entries 3 and 4), as
well as substitution in positions 3,3’ increased the
enantioselectivities [60].
(R)-47
R
R
99a-d
R
Me
+
The catalyst:
OH
(R)-catalyst
Me
O
R
R
R
Me
L
O
O
101
Ti
O
L
100a, R = H
100b, R = Cl
OBn
OBn
O
R
L=
(R)-102
H
The catalyst:
TiCl2 (OiPr) 2
R
iPrOH
active catalyst
(R)-12a
(R)-BINOL
MS 4A
100c, R = Br
100d, R = tBu
TiLn*
Scheme 35.
Products 9 7 were isolated in low yields (18-33%),
but with very good enantioselectivities (up to 97.3 ee).
The poor yields are attributed to the polymerization of
n-butylglyoxylate. As can be seen in Table 15, an
acceleration of the reaction in entries 2 to 7 is
observed, when compared to entry 1.
O
Me
H
H
BnO
Felkin addition
Scheme 36.
Nu
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
In 1997, Mikami and coworkers described an
interesting
double-stereodifferentiating
study
(Scheme 36). Carbonyl ene-reactions of (S)benzyloxypropanal 1 0 1 catalyzed by (R)-BINOL-Ti
complex (R)-12a (matched pair) affords the antidiastereoisomer 1 0 2 with good selectivities (entries 1,
3, 4, Table 16) [61].
The (S)-catalyst (mismatched pair) provides the synisomer although in lower chemical yields and
selectivities (entries 2 and 5). The use of achiral
titanium catalysts, like TiCl2(iPrO)2 and TiCl(iPrO)3 leads
to the corresponding anti-isomer in good yields (entries
6 and 7).
Table 16.
Diastereofacial
Selectivity
Carbonyl-ene-reactions
in
ene component
Me
1
73
Me
15
entry
ene
catalyst
yield (%)
syn : antia
1
1
R
38
<1 : >99
2
1
S
10
84 : 16
3
73
R
55
<1 : >99
4
15
R
24
<1 : >99
5
15
S
7
84 : 16
6
15
TiCl 2( iPrO)2
60
87 : 13
7
15
TiCl(iPrO)3
48
91 : 09
a. Isomeric ratio determined by GC and/or HPLC analysis
The authors suggest that this reaction proceed
through an open transition state with Felkin addition as
depicted in Scheme 36 [61].
Table 17.
Role of Molecular Sieves (MS 4A)
entry
MS 4A (mg)
R = Me
% yield % ee
R = Ph
% yield % ee
1a
500
72 95
100 97
2b
none
79 7
81 10
3c
500 - none
76 95
96 97
a. in situ preparation of the chiral catalyst;
b. absence of MS 4A;
c. absence of MS 4A after filtering of MS used for preparing the chiral
catalyst
the presence of MS 4A (entries 1 and 3). These results
showed that MS 4A is essential for the formation of the
chiral catalyst but do not play an important role in the
ene-reaction. Analysis of 13C NMR spectra showed the
hydroxy-carbon signal of BINOL at δ 153 ppm (s). No
change was observed by mixing BINOL with
Ti(OiPr)2Cl2 in the absence of MS 4A. The addition of
MS 4A to a solution of BINOL and Ti(OiPr)2Cl2 lead to a
downfield shift of the hydroxy-carbon signal (m, 160163 ppm) indicating the formation of the BINOLattached chiral catalyst [62]. It appears that MS 4A
facilitates the alcoxy-ligand exchange reactions in the in
situ preparation step of the chiral catalyst BINOL-TiCl2.
catalyst
R
Me
O +
OMe
H
OH
(0.1 mol%)
CH 2Cl 2
0 oC
OMe
R
O
13
O
(1 mmol)
The catalyst:
O
The Role of Molecular Sieves
The use of molecular sieves is essential to obtain
high levels of enantioselectivity in the asymmetric
catalytic glyoxylate-ene-reaction [38]. The authors
observed no significant difference in rate and chemical
yield in these catalytic ene-reactions in the absence
and in the presence of MS 4A, but observed a low
optical yield when the catalyst solution was prepared in
the absence of MS 4A (Scheme 37, Table 17).
The authors observed that higher levels of
enantioselectivity are obtained using a catalyst solution
prepared in the presence of MS 4A. The use of a
catalyst solution obtained by removal of the MS 4A by
filtration afforded the same levels of high
enantioselectivity to that obtained for the reaction in
327
O
Cl
Ti
Cl
(R)-12a
Scheme 37.
In 1997, Mikami and coworkers further elucidated
the role of molecular sieves and proved that the
dichlorotitanium complex (R)-1 2 a is not the active
titanium catalyst based on 17O NMR and elemental
analysis [63]. The authors proposed a BINOL-Ti catalyst
composed of a µ3-oxo (Ti3O) as the active catalyst. The
dichlorotitanium complex (R)-1 2 a , prepared from
BINOL dilithium salt 103 and TiCl4 in CH2Cl2 was used
as catalyst for the ene-reaction between αmethylstyrene 17 and n-butylglyoxylate 96 (Scheme
328 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
38). The best result was obtained in the presence of
MS 4A, affording the ene product in 74% yield and
89.6% ee. In the absence of MS 4A, the ene product is
obtained in only 31% yield and 53.1% ee. These
results are consistent with the participation of MS 4A,
converting the pre-catalysts 1 0 3 into the reactive
BINOL-Ti-catalyst 1 2 a .
H2O) afforded the ene product in 94.2% ee and only
46% yield. Catalyst prepared in the presence of
commercially available and unactivated MS 4A (5.3%
m/m H2O) afforded the ene product in 97.2% ee and
96% yield. These results clearly demonstrate the
importance of MS 4A as a H 2O donor for the formation
of the active BINOL-Ti catalyst [63].
Positive Non-Linear Effect
Ph
(R)-12a
(10 mol%)
Me
17
O
OH
CH2 Cl 2
+
On Bu
On Bu
-30 o C, 1h
Ph
97
H
96
O
with MS 4A 89.6% ee (74%)
no MS 4A 53.1% ee (31%)
O
A very strong (+)-NLE (Positive Non-Linear-Effect)
was observed in the glyoxylate ene-reaction with the
titanium catalyst BINOL-TiBr2 1 2 b [64]. The authors
observed that the optical yield (% ee) for the ene
product obtained in the reaction of α-methylstyrene 1 7
with methylglyoxylate 1 3 in the presence of chiral
BINOL-TiBr2 complex was significantly higher than the
enantiomeric purity of chiral BINOL ligand (Scheme 40,
Table 18).
1. TiCl4
CH2 Cl2
-78 o C, rt, 1h
OLi
Ph
OLi
2. -2LiCl
filtration
Me
17
cat. (R)-12b
(66.8% ee)
1 mol%
O +
103
OMe
OMe
H
13
The catalyst:
OH
CH 2Cl 2
-30 o C
Ph
18
O
96%
(94.4% ee)
MS 4A
O
(1 mmol)
The catalyst:
Cl
O
Ti
O
Cl
Ti
O
It is interesting to observe the important role of MS
4A in the preparation of the active catalyst (Scheme
39). Catalyst prepared from BINOL and Ti(Cl) 2(OiPr)2 in
the presence of highly activated MS 4A (0.2% m/m
active
BINOL-Ti
catalyst
(10 mol%)
Me
17
OH
O +
On Bu
H
96
O
CH2 Cl2
-30 oC, 1h
On Bu
Ph
- activated MS 4A
(<0.2% m/m H2 O)
97
O
Scheme 40.
Table 18.
Positive Non-linear Effect
entry
BINOL 12b (% ee)
18 ( % ee)
yield (%)
1
13
59.9
94
2
33
91.4
92
3
46.8
92.9
88
4
66.8
94.4
96
5
100
94.6
98
94.2% ee (46%)
- commercially available MS 4A
(5.3% m/m H 2O) 97.2% ee (96%)
Scheme 39.
Br
(R)-12b
Scheme 38.
Ph
Br
O
(R)-12a
The reaction using a BINOL-Ti complex of 33% ee
affords the corresponding ene product with 91.4% ee
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
(entry 2). The use of a BINOL-Ti complex of 66.8% ee
affords product 1 8 with 94.4% ee, which is close to
that obtained with enantiomerically pure BINOL (entries
4 and 5). The non-linear relationship that exists
between catalyst optical purity and reaction
enantioselection
documented
by Mikami and
coworkers is indicative of catalyst aggregation in
solution. The BINOL 1 1 reacts first with Ti(OiPr)2Br2 to
form, by alkoxy exchange, both enantiomeric
monomers of BINOL-TiBr2, (R)- and (S )-1 2 . These
monomers are in equilibrium with the corresponding
dimers (R)● (R)-1 0 4 , (S )● (S )-1 0 5 and (S )-(R)-1 0 6 ,
with the dimeric nature of these titanium complexes
having been confirmed by vapor pressure osmometry
measurements (Scheme 41).
A closer
look
at
the
three-dimensional
representations of these dimers is presented in these
references and explains the stability differences
observed in solution. It is observed that steric
interactions are much more important in the C2symmetric homochiral dimers (R)● (R)-1 0 4 and
(S )● (S )-1 0 5 , since the binaphhyl moieties are synperiplanar (distorted Ti2O2 4-membered-ring). The
heterochiral complex (S )● (R)-1 0 6 which possesses
Ci-symmetry with a coplanar Ti2O2 4-membered-ring
and anti-periplanar orientation of the ligands seems to
be more stable [64] (Scheme 41).
Transition-State
reaction
Model
for
the
OH
(+/-)-BINOL 11
TiBr 2(OiPr) 2
X
O
X
Ti
O
Ti
X
(R)-12a
X
O
(S)-12a
X = Br
X
X
O
X
O
O
O
Ti
X
O
X
O
O
Ti
X
(R).(R)-104
X
X
O
Ti
O
O
O
Ti
X
X
(S).(S)-105
Scheme 41.
X
Ti
Ti
O
Mikami
Ene-
Very recently, Corey published three very
interesting papers describing experimental X-ray
crystallographic evidences for formyl CH--O and formyl
C--F hydrogen bonds [65] Fig. (1 ).
OH
O
329
X
(S).(R)-106
330 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
negative charge on boron
increases basicity of oxygen
Cl
BINOL
A
Cl
Ti
BINOL
Cl
O
Ti
B
BINOL
Ti
OH
Ti
BINOL
O
C
Scheme 42.
F
X
F
B
B
R
O
F
H
Y
O
O
H
R
R
coordination enhances the
positive charge at formyl hydrogen
Fig. (1). Formyl CH--F and CH--O hydrogen bonds.
In these papers, Corey describes the use of formyl
CH--O hydrogen bond as an additional factor that
contributes to the high degree of enantioselectivity
that is observed in several enantioselective Lewis acid
catalyzed reactions. In the last paper of this series,
Corey describes applications of this new kind of
hydrogen bond in determining
transition-state
geometry in chiral Lewis-acid catalyzed aldol, carbonyl
allylation and Diels-Alder reactions [4]. The preference
for this coplanar/eclipsed conformer derives from an
attractive interaction between the formyl hydrogen
(acidified by coordination of oxygen to the boron) and
the coplanar fluorine (more electron rich because of the
negative charge on boron).
An alternative explanation for this same fact comes
from an interaction between the HOMO (oxygen lone
pair) and LUMO (σ* B-F). This type of stabilization
cannot be ruled out although the energy of the HOMO
(oxygen lone pair) is considerably lowered because of
the positive charge on oxygen, and the energy of the
LUMO (σ* B-F) is increased because of the negative
charge on boron Fig. (2 ).
F
σ* B-F
F
Based on the fact that both methylglyoxylate 1 3 (2point-binding) and 3-methoxycarbonylpropynal 7 1 (1point-binding) afford similar levels of enantioselectivity,
the authors propose that bidentate coordination of
both carbonyl groups of the glyoxylic esters is not
essential. They propose that the aldehyde is activated
by complexation with the chiral catalyst (R)-BINOL-TiX2
via the formyl lone electron pair syn to the formyl
hydrogen to form a pentacoordinated titanium structure
Fig. (3 )
A trigonal bipyramidal geometry with the activated
aldehyde and one of the electronegative ligands in the
apical position is proposed for this pentacoordinated
complex. The authors also propose formyl CH---O
hydrogen bonding with the closer and more accessible
oxygen lone pair of the BINOL generating structure
1 0 7 Fig. (3 ).
aldehyde in apical
position
X
X
Ti
O
O
O
OCH3
O
H
CH--O hydrogen
bond
107
Fig. (3). Complexation of aldehyde to BINOL-Ti catalyst.
Approach of the nucleophile from the top (re face)
of the aldehyde is much more accessible than
approach from the si face, shielded by the close
naphthol ring [66] Fig. (4 ).
B
O
H
R
H
F
X
R
X
Fig. (2). Molecular orbital interactions.
Also in 1997, Corey et al. presented a transitionstate model for the Mikami enantioselelective enereaction based on the same arguments presented
above [66]. Although the exact structure of the
effective Mikami catalyst is unknown, the authors
believe that any of the following species may function
as effective catalytic species (Scheme 42).
Ti
O
O
R
H
O
H
OCH3
O
si face
shielded
by the
naphthol ring
Fig. (40). Transition State for the Mikami ene-reaction.
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
331
Ytterbium Lewis Acids
Copper Lewis Acids
Limited degrees of success have been achieved in
developing optically active chiral catalysts derived from
lanthanides for the ene-reaction [67]. Chiral ytterbium
complex 108 generated from Yb(OTf)3 and (S)-6,6'dibromo-binaphthol, produces a very
modest
asymmetric induction in the ene-reaction of
methylglyoxylate 13
with α-methylstyrene 1 7
(Scheme 43, Table 19). The best result in terms of
enantioselectivity (38% ee) for the derived αhydroesters
18
was
obtained
using
this
dibromoderivative (entry 2) [67].
Copper as a Lewis acid is a moderately oxophilic
metal with a high propensity for 4-coordinacy [68]. A
bidentate ligand can occupy 2 free coordination sites
and 2-point-substrate-binding is possible. Cationic
Cu(II) complexes are among the rare examples of late
transition metal based Lewis acid catalysts that have
been successfully applied to organic reaction
methodology.
C2-symmetric
bis-oxazoline-Cu
complexes are structurally related to the C2-symmetric
semicorrins developed by Pfaltz et al. [69,70] Fig. (5 ).
2
Me
Ph
chiral
ytterbium
catalyst
(20 mol%)
Me
17
+
O
OMe
H
13
O
OH
N
OMe
CH3 CN
0 oC, 24h
Me
O
(S)-18
Cu
But
Ph
N
t Bu
O
O
R
Fig. (5). Bis-oxazoline copper (II) complex.
OH
OH
In 1995, Jorgensen and coworkers described that
copper(II) bisoxazolines are very useful catalysts for the
reaction of glyoxylate esters with dienes, affording
highly valuable hetero Diels-Alder and ene products
[71]. The authors observed that the hetero Diels-Alder
product:ene product ratio is in the 1.0:0.6 to 1.0:1.8
Yb(OTf)3
+
CH 2Cl 2, 2h
R
R1
The catalyst:
O
R1
chiral ytterbium
catalyst
OEt
R2
110
R2
(S)-(-)-108
a, R = H
b, R = Br
c, R = Ph
d, R = -CCTMS
O
O
catalyst
+
+
OH
OEt
OEt
H
109
O
111
O
Me
Scheme 43.
Table 19.
The catalysts:
Glyoxylate Ene-reaction Catalyzed by
a Chiral Ytterbium Catalyst
Me
Me
entry (S)-(-)-catalyst 108
R
yield (%)
ee % (conf.)a
1
a
H
82
12 (S)
2
b
Br
78
38 (S)
3
c
Ph
87
25 (S)
4
d
-CCTMS
83
29 (S)
a. optical yields and absolute configuration determined by comparison with
literature data
Me
O
O
N
N
Cu
R
TfO
(S)-112, R = tBu
(R)-113, R = Ph
(S)-113, R = Ph
Scheme 44.
O
N
R
OTf
Me
O
N
Mg
But
I
(S)-114
t Bu
I
332 Current Organic Chemistry, 2000, Vol. 4, No. 3
Table 20.
Luiz Carlos Dias
Glyoxylate Ene-reactions with Dienes
DA product (110)
Ene product (111)
ratio
entry
R1
R2
catalyst
yield (%)
ee % (conf.)
yield (%)
ee % a
110:111
1
Me
Me
(S)-112
20
85 (S)
36
83
1.0:1.8
2
Me
Me
(R)-113
31
83 (S)
50
88
1.0:1.6
3
Me
Me
(S)-114
10
5 (S)
20
10
1.0:2.0
4
Me
H
(R)-113
33
80a
34
91
1.0:1.0
a. absolute configuration not assigned
range and is dependent on the chiral ligand attached to
the metal, the glyoxylate ester, and the reaction
temperature (Scheme 44, Table 20). These copper(II)
bisoxazoline catalysts give much better hetero DielsAlder selectivity when compared with the chiral BINOL
and titanium complexes.
As can be seen from the results in Table 19, the
copper catalysts (R)-1 1 3 and (S )-1 1 2 (entries 1 and 2)
are much better catalysts than the corresponding
magnesium (II) iodide bisoxazoline complex (S )-1 1 4
(entry 3) , both in terms of yields and
enantioselectivities.
Using isoprene as the substrate and (R)-1 1 3 as the
catalyst afforded a hetero Diels-Alder product:ene
product ratio of 1.0:1.0. In the same reaction, the chiral
BINOL titanium complex gives a 1.0:4.0 ratio [72]. The
authors observed that using the tert-butyl-substituted
bisoxazoline ligand catalyst, the methyl glyoxylate
esters gives the highest ee in the hetero Diels-Alder
and the ene-reaction. They observed also that the
absolute stereochemistry in Diels-Alder products is
dependent on the catalyst applied. The use of a
bisoxazoline ligand with a tert-butyl substituent at the
chiral center gives the opposite stereochemistry
compared with a bisoxazoline ligand having a phenyl
substituent at the chiral center (entries 1 and 2). They
propose that this result is explained by a geometrical
change at the copper atom, with a planar complex as
the intermediate when R=tBu, and a tetrahedral
CO2 Me
115
NHCbz
CO2 Me
SPh
CO2 Me
NHCbz
116
+
MeO 2C
catalyst
(R)-113
unstable
product
42%
O
SPh
NHCbz
117
94:06 diastereomeric
excess determined at
a later stage
OMe
H
13
OH
O
The catalyst:
Me
Me
O
O
N
N
TfO
Ph
OTf
(R)-113
Scheme 45.
S
NH 2
NH 2
meso-DAP 118
Cu
Ph
R
CO2 H
HO2C
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
arrangement at the metal when the bisoxazoline ligand
is phenyl. The authors attributed the turnover to a
change in metal geometry from square planar to
tetrahedral.
A very interesting synthetic application of this
methodology was used by Vederas and coworkers who
reported the use of bis(oxazoline)copper-(II) complex
(R)-1 1 3 as a chiral catalyst for the ene-reaction of the
N-Cbz-derivative
of
methyl
(S)-4-(phenylthio)
allylglycinate 1 1 6 and methylglyoxylate 1 3 affording
1 1 7 in modest 42% yield and 88% de (S) (Scheme 45)
[73]. This reaction was employed as a key step in the
synthesis of meso-diaminopimelic acid (meso-DAP)
1 1 8 , a key constituent of bacterial peptidoglycan, and
a potential target for development of new antibiotics.
The alcohol 1 1 7 was transformed into the desired acid
1 1 8 after a sequence involving few steps.
Attempts
to
use
N-(benzyloxy-carbonyl)-Lallylglycine methyl ester 1 1 5 in this reaction, in the
presence of several chiral catalysts failed because of
the unreactive terminal olefin. To circumvent this
problem, a sulfur substituent was temporally introduced
at the terminus of the allylglycine residue. It is important
to point out that the use of chiral binaphthol-titanium
complexes as Lewis acids also failed and lead only to
recovered starting materials [73].
An extremely useful and highly enantioselective
ene-reaction was reported recently by Evans and
coworkers [74]. They described that the utilization of
the bidentate bis(oxazolinyl)(box)-Cu(II) complexes
The catalysts
Me
Me
O
N
Ph
Me
2
O
N
Ph
N
Cu
But
(S)-119
(S)-113
2
O
O
N
Cu
Me
t Bu
1 1 3 , 119 and 120 are highly selective and effective
enantioselective catalysts for glyoxylate ene-reactions
(Scheme 46). These bis(oxazoline)-copper complexes
were initially used for asymmetric Diels-Alder reactions
and were reported to produce undesired ene side
products. These C2-symmetric copper-(II) complexes
provided excellent yields and enantioselectivitites in
the addition of a variety of olefins (including less
nucleophilic olefins) to glyoxylate esters (Table 21).
Ene-reaction of methylenecyclohexane 1 with
ethylglyoxylate 1 0 9 afforded (S )-1 2 1 in 97% ee (97%
yield) in the presence of 10 mol% of (S)-catalyst 1 1 9
(Table 21). The use of bis(aqua) catalyst (S )-1 2 0 (less
reactive) led essentially to the same result (entry 1). It is
interesting to note that the [Cu((S,S)-Ph-box)](OTf)2
complex 1 1 3 affords the absolute stereochemistry
(87% ee, 97% yield) of the resulting product opposite
to that produced by (S,S)-tBu-box catalysts 1 1 9 and
1 2 0 , in perfect accordance with the results previously
reported by the Jorgensen group and with those
reported by Vederas et al. [71,73].
The optimized Cu-(II)-bis-oxazoline catalyst system
has been successfully applied to the ene-reaction of
unsymmetrical 1,1-disubstituted olefins (entries 3 and
6), as well as with less nucleophilic monossubstituted
olefins (entry 4). The (S)-Ph-box-derived catalyst 1 1 3
(10 mol%) mediates the addition of 2-methyl-1heptene 1 2 8 to ethylglyoxylate 109 to afford (R)1 2 9 in 91% ee, and 90:10 regioselectivity, the best ee
obtained so far for this type of enophile (entry 6). This
result clearly demonstrate that this catalytic system can
discriminate between methyl and
methylene
hydrogens. The use of catalyst (S )-120 affords (S )1 2 9 in 96% ee but with only 74:26 regioselectivity
[74]. It is interesting to observe that bis(aqua) complex
(R)-1 2 0 , readily prepared as a bench-stable solid, can
be used as catalyst with loadings as low as 0.1 mol%
(entries 2, 5 and 6).
The sense of asymmetric induction can be
rationalized by assuming that the reaction proceeds via
the intermediacy of the square planar catalyst-
2 SbF6
2 OTf
Me
Me
Me
2
Me
N
N
N
Cu
But
H2 O
(S)-120
t Bu
N
Cu
But
O
t Bu
O
OH2
blue
solid
130
Nu
H
OEt
re face shielded by
the bulk t butyl group
2 SbF6
Scheme 46.
2
O
O
O
O
333
Scheme 47.
334 Current Organic Chemistry, 2000, Vol. 4, No. 3
Table 21.
Luiz Carlos Dias
Glyoxylate Ene-reactions Catalyzed by Copper (II) Complexes
entry
olefin
product
cat. (mol %)
T (oC)
yield (%)
% ee (conf.)a
120 (10)
0
97
97 (S)
119 (10)
0
97
97 (S)
113 (10)
0
99
87 (R)
120 (1)
0
83
96 (S)
113 (10)
0
92
92 (R)
120 (10)
25
62
98 (S)
113 (2)
25
88
92 (R)
119 (10)
25
96
98 (S)
120 (1)
0
97
93 (S)
113 (10)
0
99
89 (R)
120 (1)
25
89
96 (S)
113 (10)
25
81
91 (R)
O
OEt
1
OH
121
1
O
2
Me
Me
Me
OEt
15
OH
122
OBn
3
O
BnO
Me
OEt
123
OH
124
O
4
C3 H7
OEt
125
OH
C3 H7
126
O
5
Ph
Ph
Me
OEt
17
OH
127
C4H9
6
O
C4 H9
Me
OEt
128
OH
129
a. absolute configuration assigned by conversion to MTPA esters
b. ee's determined by GLC (Cyclodex-β) column or HPLC (Chiralcel OD-H column)
glyoxylate complex 130 (Scheme 47). The re face of
the coordinated aldehyde is blocked by the tert-butyl
substituent and the approach of olefins occurs from the
accessible aldehyde si face.
More recently, Evans and coworkers at Harvard
suggested, based on structural and mechanistic
studies, that a change in geometry at the metal center
is not necessarily responsible for the reversal in
enantioselectivity observed in glyoxylate ene-
reactions, as proposed earlier by Jorgensen et al. [75].
According to the Evans group, Cu(II)-bis(oxazoline)
complexes (S )-1 1 3 and (S )-1 1 9 catalyze the enereaction
of
methylenecyclohexane
1
and
ethylglyoxylate 1 0 9 with the enantiomeric excess of
the product dependent on the oxazoline ring
substituent (Scheme 48). α-hydroxy-ester 1 2 1 is
obtained in 97% ee employing catalyst (S,S)-1 1 3 , and
the corresponding enantiomer is obtained in 87% ee
using complex (S )-1 1 9 . The use of complex (S )-1 3 1
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
10 mol%
(S)-113
OH
OEt
OH
OEt
+
CH 2Cl 2, 0 o C
121
10 mol%
(S)-119
1
CH 2Cl 2, 0 o C
O
121
O
O
OEt
H
(S):(R) = 98.5:1.5 (97% ee)
335
(S):(R) = 6.5:93.5 (87% ee)
109
O
The catalysts
Me
Me
Me
2
O
O
N
Cu
Ph
But
Ph
(S)-113
N
a: X = OTf
Me
2X
t Bu
N
Pri
(S)-119
2
O
O
N
Cu
Me
2
O
O
2X
N
Me
2X
N
Cu
iPr
(S)-131
b: X = SbF6
Scheme 48.
afforded the corresponding ene product with only 36%
ee (S). The authors believe that the intermediacy of a
distorted square planar bis(oxazoline) Cu-(II)-substrate
complex is responsible for this reversal in
enantioselectivity.
The
Evans
group
employed
double
stereodifferentiating experiments, EPR spectroscopy,
Me
Me
2
O
O
N
N
2SbF6
Cu
But
H2 O
t Bu
OH2
(S)-120
semiempirical calculations
and
crystallographic
techniques to investigate this phenomenon (Scheme
49). X-Ray crystal structures of the Cu[(S,S)-iPrbis(oxazoline)](H2O)2(SbF 6)2
and
Cu[(S,S)- tBubis(oxazoline)](H2O)2(SbF 6)2 complexes are also
presented.
In bis-aquo-complex 1 2 0 , the Cu-(II) center
presents a geometry distortion from square planarity,
with the ligated water molecules distorted +33.3o away
from oxazoline substituents (Scheme 49). In the case
of phenyl-substituted complex 1 3 2 , the water
molecules tilt toward the oxazoline substituents by
–9.3 o. The authors observed also that these distortions
are independent of the nature of the counterion and
the absence of nonlinear effects in these glyoxylate
ene-reactions [75].
O1 -Cu-N1 -C1 dihedral < +30.2o
Ene-reactions
Esters
O2 -Cu-N2 -C2 dihedral < +35.9o
Me
Me
2
O
O
N
N
Cu
Ph
H2 O
2SbF6
Ph
OH2
(S)-132
O1 -Cu-N1 -C1 dihedral < - 11.3o
O2 -Cu-N2 -C2 dihedral < - 7.2o
Scheme 49.
with
α-Tosyl-imino
Another highly enantioselective ene-reaction was
reported by the Jorgensen group in 1998 [76]. The
authors described a highly enantioselective enereaction of tosyl- α-iminoesters with alkenes catalyzed
by 0.1 mol % of chiral CuPF6-BINAP and CuClO4BINAP complexes. This reaction afforded chiral αaminoesters 1 3 4 , that can be used to prepare both
optically active and biologically important natural and
non-natural α-aminoacids. The chiral phosphine
ligands (R)-BINAP and (R)-tol-BINAP in combination
with copper (I) salts have been found to be the best in
336 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
The catalyst
R1
+
NHTos
catalyst
(10 mol%)
R2
Tos
N
EtO
Ar
OEt
Ar
R1
CH 2Cl 2
or THF
25 o C
P
(S)-134
O
tosyl α-imino ester
si face approach
H
CuClO4
P
Ar
Ar
O 133
(R)-135a, R = Ph
(R)-135b, R = tol
Scheme 50.
terms of chemical yields and enantioselectivities for the
ene-reaction of α-methylstyrene 1 7 with tosyl-αiminoester 1 3 3 (Scheme 50). As can be seen from the
results
showed
in
Table
22,
the
best
enantioselectivities are obtained with PF6 and ClO4
(entries 1-4) as the anions showing that this reaction is
counterion dependent. The use of complexes BINAPAgOTf and BINAP-CuOTf led to lower yields and
enantioselectivities (entries 5-7).
Table
22.
Good results also are obtained for the ene-reaction
of various alkenes with the tosyl-α-iminoester 1 3 3 in
the presence of (R)-tol-BINAP 1 3 5 -CuX as the catalyst
(Scheme 50, Table 23).
R
R'
(2 equiv.)
Tos
+
catalyst
(5 mol%)
R''
BTF
R'
NHTos
OEt
N
Ene-reaction of α-Methylstyrene
with Tosyl-α-iminoester 133
(S)-134
17
25 o C
EtO
H
O
tosyl α-imino ester
si face approach
O 133
entry
R1
R2
ligand-metal salt
yield (%)
ee (%)
1
Ph
H
135a-CuClO 4
73
93
2
Ph
H
135b-CuClO 4
75
95
Ar
3
Ph
H
135a-CuPF6
77
93
P
4
Ph
H
135b-CuPF6
80
95
P
5
Ph
H
135a-CuOTf
58
76
Ar
6
Ph
H
135b-CuOTf
67
80
7
Ph
H
The catalyst:
135b-AgOTf
75
CuClO4
Ar
(R)-135
Ar = 4-MeC6 H5
73
* all reactions were run in THF as solvent
Table 23.
Ar
Scheme 51.
Ene-reaction Between Various Alkenes and α-Tosylimino Ester 133
entrya
R1
R2
ligand-metal salt
load (%)
yield (%)
ee (%)b
1
Ph
H
135b-CuClO 4
10
85
95
2
Ph
H
135b-CuPF6
1.0
80
99
3
Ph
H
135b-CuPF6
0.5
82
98
4
Ph
H
135b-CuPF6
0.1
71
95
5
p-OMe-C6H4
H
135b-CuClO 4
0.1
80
91
6
(CH 2) 3
(CH 2) 3
135b-CuPF6
0.5
74
92
7
Me
H
135b-CuPF6
1.0
62
78
a. all reactions were run in CH2Cl 2
b. ee's determined by chiral HPLC using a Chiralcel OJ or OD column
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
methylstyrene 17 and α-iminoester 133 in the
presence of 5 mol% of (R)-Tol-BINAP-CuClO4.2CH3CN
and have found that the best conditions involves the
use of benzotrifluoride (BTF) as solvent and 2.0 equiv.
of alkene (Table 24, entry 1). Under these conditions,
the corresponding ene-product 134 was obtained in
92% yield and 99% ee after 18h at room temperature.
The authors observed that reducing the catalyst
loading from 1.0 to 0.1 mol% caused no significant
reduction in yields and enantioselectivities (entries 2, 3
and 4).
Although the catalytic enantioselective ene-reaction
of carbonyl compounds is very well studied and there
are many highly efficient variants, the same is not
observed for the related ene-reaction with imines. As
pointed out by Jorgensen, one possible problem is the
fact that the imine probably competes with the chiral
ligand by coordinating to the Lewis acid which
suppresses the chiral information from the ligand
[76,77].
Several other alkenes, including heteroatomcontaining ene substrates were used, affording good
yields and excellent enantioselectivities for the
corresponding ene products (Table 24).
Removal of the tosyl group was carried out by
treatment of product 1 3 4 a with HBr/phenol, providing
the α-aminoacid 1 4 0 in 75% yield [78] (Scheme 52).
In this same year, Lectka and coworkers described a
very useful catalytic enantioselective imino enereaction of tosyl-α-imino ester 1 3 3 with alkenes in the
presence of copper-complex 1 3 5 (Scheme 51) [78].
The authors studied the reaction between αTable 24.
entry
Very recently, Rich and Elder reported a very
interesting synthetic application of this methodology to
the synthesis of the 16- and 17-membered DEF ring
Ene-reaction with α-Iminoester 133
alkene
product
yield (%)
ee (%)a
92
99
94
99
85
95
85
98
90
85
85
89
NHTos
1
Me
CO2 Et
13
134a
NHTos
2
CO2 Et
134b
136
NHTos
3
CO2 Et
134c
1
S
NHTos
4
Me
S
137
CO2 Et
134d
Tos
N
5
NHTos
N
138
CO2 Et
134e
Tos
NHTos
6
O
CO2 Et
O 139
a.ee's determined by NMR in the presence of a chiral shift reagent
337
134f
338 Current Organic Chemistry, 2000, Vol. 4, No. 3
NHTos
1. PhOH
HBr/AcOH
OEt
Ph
2. H 2O
(S)-134a
O
NH 2
Luiz Carlos Dias
Compound 1 4 2 was converted to the 17membered DEF ring system 1 4 3 of complestatin after
a few steps (Scheme 54). Acid catalyzed ring
contraction of this 17-membered ring (TFA, 50 oC)
generated the 16-membered DEF ring system of
chloropeptin.
OH
Ph
Conclusions
(S)-140
O
Scheme 52.
systems of chloropeptin and complestatin, two very
potent biologically active macrocyclic polypeptides
[79]. Ene-reaction of tosyl α-imino ester 1 3 3 with 3methyleneindoline 1 4 1 in the presence of (S)-tolBINAP-CuClO4-2CH3CN in BTF as solvent afforded the
fully protected 6-bromo-D-tryptophan 1 4 2 in 76% yield
and 94% ee (Scheme 53).
N
Br
141
catalyst
(5 mol%)
Tos
+
BTF
25 o C, 4h
Tos
N
EtO
H
O 133
Tos
N
NHTos
CO2 Et
Br
(R)-142
76%, 94% ee
ee determined by Mosher ester
analysis of the primary alchol
obtained after ester reduction
The catalyst:
Ar
Ar
P
P
Ar
CuClO4
Ar
(S)-135
Ar = 4-MeC6 H5
Scheme 53.
The progress in the catalytic asymmetric enereaction has been outstanding and impressive results
through the use of titanium Lewis acids have been
obtained by Mikami et al. and Nakai et al. In particular,
the asymmetric induction achieved by using chiral
glyoxylate esters and the catalysis with BINOL-Lewis
acid complexes are major achievements.
The use of copper Lewis acids derived from bisoxazolines as described by Evans affords also fantastic
results in terms of yields and enantioselectivities for the
glyoxylate ene-reaction.
With regard to the imino ene-reaction, the results
described recently by Jorgensen and Lectka are also
very promising.
Despite these impressive recent advances, many
unsolved problems still remain and there are some
other features of the catalytic ene-process that remain
to be improved. These include limitations with regard to
scope and frequent practical problems associated with
catalyst preparation and use, especially on large scale.
Appropriate structural design of catalyst complexes to
avoid the formation of oligomeric aggregation is also a
worthwhile goal, since monomeric structures of
catalysts would lead to enhanced catalytic activity.
Reaction enantioselection is highly sensitive to minor
variations in catalyst preparation and, presumably, the
solution-state structure that is derived therefrom,
resulting in nearly identical catalyst systems providing
different results, ranging from low to nearly perfect
asymmetric induction. The control of the regio- and
stereochemistry of the ene-reaction is still far from
developed. It is therefore not surprising that
considerable attention has been focused on the
development of metal catalyzed asymmetric variants of
this reaction.
We predict that discoveries of even more practical
chiral Lewis acid catalysts, displaying substrate
tolerance and requiring lower catalyst loading, will
continue to be a challenge for synthetic organic
chemists. We expect to see more developments in the
direction of ene-reactions involving both rich- and poor
enophile partners, which will need a new generation of
asymmetric catalysts. Further work on the design of
more practical chiral Lewis acid catalysts and new
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
339
OH
Tos
F
N
NHTos
D
Br
(R)-142
N
H H
O
CO2 Et
TBDPSO
H
N
N
H
NHTos
O
17-membered DEF
ring of complestatin
143
E
TFA, 50 oC
OH
F
D
HN
O
H
H
HO
N
N
H
NHTos
O
144
16-membered DEF
ring of chloropeptin
E
Scheme 54.
technological developments will provide further
possibilities to enlarge the scope and applications of
catalytic asymmetric ene-reactions in academics and
industry.
(Fundação de Amparo à Pesquisa do Estado de São
Paulo)
and
CNPq
(Conselho
Nacional
de
Desenvolvimento Científico e Tecnológico).
It is essential for asymmetric synthesis to
understand the factors that control π-facial selectivity in
enantioselective reactions. Using the words of Prof.
Kagan in his excellent review about Nonlinear Effects,
“the development of asymmetric synthesis gave the
opportunity to elaborate methods, reagents, and
reactions in order to prepare enantiomerically pure
compounds. But this journey is far from being over, and
there are still a lot of surprises left. No chemist would
have thought that an enantiomerically impure chiral
auxiliary or ligand could give a stereoselection higher
than its own and even equivalent to the pure one” [80].
References
[1]
For very interesting papers about chiral drugs, see: (a) Thall, E.
J. Chem. Ed., 1996, 3 (6), 481. (b) Stinson, S.C. Chemical &
Engineering News, 1994, 38.
[2]
For a very interesting review paper dealing with the
requirements for using enantioselective catalysis in the
synthesis of fine chemicals, see: Blaser, H.U.; Studer, M.
Chirality, 1999, 11(5-6), 459.
[3]
For excellent books dealing with the principles of asymmetric
synthesis, see: (a) Seyden-Penne, J. Chiral Auxiliaries and
Ligands in Asymmetric Synthesis, John Wiley & Sons, New
York, Chichester, Brishane, Toronto, Singapore, 1995. (b) Beller,
M.; Bolm, C. Eds., Transition Metals for Fine Chemicals and
Organic Synthesis, Wiley-Interscience: Chichester, UK, 1998.
[4]
For a review about chiral Lewis acid catalyzed Diels-Alder
reactions, see: Dias, L.C. J. Braz. Chem. Soc., 1997, 8, 289.
[5]
(a) For a comprehensive review of catalyzed enantioselective
aldol reactions, see: Nelson, S.G. Tetrahedron:Asymmetry,
1998, 9, 357. (b) For a review about chiral Lewis acid promoted
asymmetric aldol reaction using oxazaborolidinones, see:
Kiyooka, S. Rev. Heteroatom Chem., 1997, 17, 245.
Acknowledgments
I am grateful to both Luciana G. de Oliveira and Prof.
Roy Edward Bruns from UNICAMP, for helpful
comments and useful suggestions about English
grammar and style, during the preparation of this
manuscript. Support has been provided by FAPESP
340 Current Organic Chemistry, 2000, Vol. 4, No. 3
[6]
For an extensive review on selective reactions using allylic
metals, see: Yamamoto, Y.; Asao, N. Chem. Rev., 1993, 93,
2207.
[7]
(a) For an excellent review on asymmetric synthesis with chiral
Lewis acid catalysts, see: Ishihara, K.; Yamamoto, H. Cat.
Tech., 1997, 51. (b) For an excellent review on chiral Lewis
acids in catalytic asymmetric reactions, see: Narasaka, K.
Synthesis, 1991, 1.
(a) For a very interesting paper about enantiomeric impurities in
chiral catalysts, auxiliaries and synthons used in
enantioselective synthesis, see: Armstrong, D.W.; Lee, J.T.;
Chang, L.W. Tetrahedron:Asymmetry, 1998, 9, 143. (b) Review
article about the use of chiral metal(salen) complexes in
asymmetric catalysis: Sherrington, D.C. Chem. Soc. Rev.,
1999, 28 (2), 85.
[9]
[10]
[8]
[11]
[12]
[13]
[14]
[15]
[16]
Luiz Carlos Dias
[21]
Whitesell, J.K.; Chen, H.-H.; Lawrence, R.M. J. Org. Chem.,
1985, 50, 4663.
[22]
Whitesell, J.K.; Lawrence, R.M; Chen, H.-H. J. Org. Chem.,
1986, 51, 4779.
[23]
For mechanistic aspects of the ene-reaction, see: Paderes, G.D.;
Jorgensen, W.L. J. Org. Chem., 1992, 57, 1904.
[24]
For a paper about the mechanism of the carbonyl-ene reaction,
see; Achmatowicz, O.; Florjanczyk, E.B. Tetrahedron, 1996, 52,
8827.
[25]
For excellent reviews about asymmetric ene-reactions in
organic synthesis, see: (a) Mikami, K.; Shimizu, M. Chem.
Rev., 1992, 92, 1021. (b) Borzilleri, R.M.; Weinreb, S.M.
Synthesis, 1995, 347. (c) Mikami, K. Pure & Appl. Chem., 1996,
68, 639.
For an excellent review about stoichiometric asymmetric
processes, see: Regan, A.C. J. Chem. Soc., Perkin Trans. 1,
1999, 357.
[26]
(a) For an excellent review about catalytic asymmetric process,
see: Wills, M. J. Chem. Soc. Perkin Trans. 1, 1998, 18, 3101. (b)
enantioselective desymmetrization: Willis, M.C. J. Chem. Soc.,
Perkin Trans. 1, 1999, 1765.
For reviews about general aspects of asymmetric carbonyl-enereactions, see: (a) Mikami, K.; Terada, M.; Narisawa, S.; Nakai,
T. Synlett, 1992, 255. (b) Mikami, K. Advances in Asymmetric
Synthesis, Vol. 1, pp. 1-44, JAI Press, Inc., 1995.
[27]
For a short review on catalytic asymmetric carbonyl enereactions, see: Berrisford, D.J.; Bolm, C. Angew. Chem., Int. Ed.
Engl., 1995, 34, 1717.
[28]
For mechanistic studies of the metallo-ene-reaction, see:
Gomez-Benjoa, E.; Cuerva, J.M.; Echavarren, A.M.; Martorell,
G. Angew. Chem., Int. Ed. Engl., 1997, 36, 767.
[29]
For a recent review about thermal and Lewis acid catalyzed
intramolecular ene-reactions of allenylsilanes, see: Weinreb,
S.M.; Smith, D.T.; Jin, J. Synthesis, 1998, 509.
[30]
For a review dealing with recent developments in asymmetric
catalysis using synthetic polymers with main chain chirality,
see: Pu, L. Tetrahedron:Asymmetry, 1998, 9, 1457.
In 1997, Jenner described the effect of high-pressure kinetics of
Lewis acid catalyzed cycloaddition and ene-reactions: Jenner, G.
New J. Chem., 1997, 21, 1085.
[31]
For an excellent review article about catalytic enantioselective
construction of molecules with quaternarycarbonstereocenters,
see: Corey, E.J.; Guzman-Perez, A. Angew. Chem., Int. Ed.
Engl., 1998, 37, 388.
For solvent effects in asymmetric hetero Diels-Alder and enereactions, see: Johannsen, M.; Jorgensen, K.A. Tetrahedron,
1996, 52, 7321.
[32]
Recent review article about catalytic asymmetric addition
reactions of carbonyls: Jorgensen, K.A.; Johannsen, M.; Yao, S.;
Audrain, H.; Thorhauge, J. Acc. Chem. Res., 1999, 32, 605.
Excellent and comprehensive reviews of intramolecular enereactions: (a) Conia, J.M.; Perchec, P. Le Synthesis, 1975, 1. (b)
Taber, D.F. Intramolecular Diels-Alder and Alder-Ene
Reactions; Springer Verlag: Berlim, 1984.
[33]
Whitesell, J.K.; Bhattacharya, A.; Aguilar, D.A; Henke, K. J.
Chem. Soc. Chem. Commun., 1982, 989.
[34]
Whitesell, J.K. Acc. Chem. Res., 1985, 18, 280
[35]
Whitesell, J.K.; Bhattacharya, A.; Buchanan, C.M. Chen, H.H.;
Deyo, D.; James, D.; Liu, C.-L.; Minton, M.A. Tetrahedron, 1986,
42, 2993.
[36]
Maruoka, K.; Hoshino, Y.; Shirasaka, T.; Yamamoto, H.
Tetrahedron Lett., 1988, 29, 3967.
[37]
For intramolecular ene-reactions of unsaturated aldehydes
catalyzed by a chiral Lewis acid, see: (a) Sakane, S.; Maruoka,
K.; Yamamoto, H. Tetrahedron Lett., 1985, 26, 5535. (b) Sakane,
S.; Maruoka, K.; Yamamoto, H. Tetrahedron, 1986, 42, 2203. (c)
another example of an intramolecular ene-reaction: Cossy, J.;
Bouzide, A.; Pfau, M. J. Org. Chem., 1997, 62, 7106.
[38]
(a) Mikami, K.; Terada, M.; Nakai, T. J. Am. Chem. Soc., 1989,
111, 1940. (b) Mikami, K.; Terada, M.; Nakai, T. J. Am. Chem.
Soc., 1990, 112, 3949.
(a) For a very interesting review on transition-metal-based
Lewis acid catalysts, see: Bosnich, B. Aldrichimica Acta, 1998,
31, 76. (b) For the first synthesis and structural characterization
of an enantiomerically pure planar-chiral Lewis acid complex,
see: Tweddell, J.; Hoic, D.A.; Fu, G.C. J. Org. Chem., 1997, 62,
8286. (b) For a very interesting review that highlights the
progress that has been made in the field of Lewis acid catalysis
of carbon-carbon-bond forming reactions in aqueous solution,
see: Engberts, J.B.F.N.; Feringa, B.L.; Keller, E.; Otto, S. Recl.
Trav. Chim. Pays-Bas, 1996, 115, 457.
(a) Treibs, W.; Schmidt, H. Ber. Dtsch. Chem. Ges., 1927, 60,
2335. (b) Grignard, V.; Doeuvre, J.C.R. Acad. Sci., 1930, 190,
1164. (c) Ikeda, T.; Wakatsuki, K. J. Chem. Soc. Jpn., 1936, 57,
425.
(a) Alder, K.; Pascher, F.; Schmitz, A. Ber. Dtsch. Chem. Ges.,
1943, 76, 27. (b) Nobel Lectures-Chemistry, 1942-1962;
Elsevier: Amsterdam, pp. 253-305, 1964.
[17]
(a) Klimova, E.I.; Arbuzov, Y.A. Dokl. Akad. Nauk. SSSR, 1965,
167, 419. (b) Klimova, E.I.; Arbuzov, Y.A. Dokl. Akad. Nauk.
SSSR, 1967, 173, 1332. (c) Klimova, E.I.; Arbuzov, Y.A. J. Org.
Chem. USSR Engl., 1968, 4, 1726.
[18]
Achmatowicz, J.O.; Szechner, B. J. Org. Chem., 1972, 37, 964.
[19]
Snider, B.B.; Straten, J.W.V. J. Org. Chem., 1979, 44, 3567.
[20]
Snider, B.B. Acc. Chem. Res., 1980, 13, 426.
Chiral Lewis Acid
Current Organic Chemistry, 2000, Vol. 4, No. 3
341
[39]
Terada, M.; Matsukawa, S.; Mikami, K. Chem. Commun., 1993,
327.
[60]
Chavarot, M.; Byrne, J.J.; Chavant, P.Y.; Pardillos-Guindet, J.;
Vallée, Y. Tetrahedron:Asymmetry, 1998, 9, 3889.
[40]
Mikami, K.; Terada, M.; Sawa, E.; Nakai, T. Tetrahedron Lett.,
1991, 32, 6571.
[61]
Mikami, K.; Matsukawa, S.; Sawa, E.; Harada, A.; Koga, N.
Tetrahedron Lett., 1997, 38, 1951.
[41]
In the ene-cyclizations, the carbon members where the tether
connects the [1,5]-hydrogen shift system, are expressed in (m,n)
fashion.
[62]
(a) Dyer, A. An Introduction to Zeolite Molecular Sieves, Wiley,
Chichester, 1988. (b) Davis, M.E. Acc. Chem. Res., 1993, 26,
111.
[42]
Mikami, K.; Sawa, E.; Terada, M. Tetahedron:Asymmetry, 1991,
2, 1403.
[63]
Terada, M.; Matsumoto, Y.; Nakamura, Y.; Mikami, K. Chem.
Commun., 1997, 281.
[43]
In 1990, Ziegler described an interesting synthesis of
trichothecene anguidine, in which one of the steps was an
asymmetric ene-reaction conducted in the presence of chiral
catalysts to give relatively high diastereoselectivity but only low
levels of enantioselectivity. Ziegler, F.E.; Sobolov, S.B. J. Am.
Chem. Soc., 1990, 112, 2749.
[64]
(a) Terada, M.; Mikami, K.; Nakai, T. J. Chem. Soc., Chem.
Commun., 1990, 1623. (b) Mikami, K.; Terada, M. Tetrahedron,
1992, 48, 5671.
[65]
Corey, E.J.; Rohde, J.J.; Fisher, A.; Azimioara, M.D.
TetrahedronLett., 1997, 38, 33. (b) Corey, E.J.; Rohde, J.J.
Tetrahedron Lett., 1997, 38, 37. (c) Corey, E.J.;Barnes-Seeman,
D.; Lee, T.W. Tetrahedron Lett., 1997, 38, 1699.
[66]
Corey, E.J.; Barnes-Seeman, D.; Lee, T.W.; Goodman, S.N.
Tetrahedron Lett., 1997, 38, 6513.
[67]
Qian, C.; Huang, T. Tetrahedron Lett., 1997, 38, 6721.
[68]
(a) For a discussion on the coordination chemistry of Cu(II), see:
Hathaway, B.J. In Comprehensive Coordination Chemistry,
Wilkinson, G.; Gillard, R.D.; MaClevarty, J.A.; Eds.; Pergamon
Press: Oxford, Vol. 5, pp. 533, 1989. (b) A recent study on
divalent metal ion catalyst in Diels-Alder reactions in aqueous
media concluded that Cu2+ is superior for bidentate metal ion
dienophilic substrates when compared to Zn2+, Ni2+, and Co2+:
Otto, S.; Bertoncin, F.; Engberts, J.B.F.N. J. Am. Chem. Soc.,
1996, 118, 7702.
[69]
For a very interesting paper about chiral C2-symmetric
semicorrin and related ligands, see: Pfaltz, A. Synlett, 1999, 835.
[70]
(a) For an improved procedure for the preparation of C2symmetric chiral bis(oxazoline)-copper complexes, see: Evans,
D.A.; Peterson, G.S.; Johnson, J.S.; Barnes, D.M.; Campos,
K.R.; Woerpel, K.A. J. Org. Chem., 1998, 63, 4541. (b) For an
excellent review about the use of C2-symmetric chiral
bis(oxazoline)-metal complexes in catalytic asymmetric
synthesis, see: Ghosh, A.K.; Mathiavanen, P.; Cappiello, J.
Tetrahedron:Asymmetry, 1998, 9, 1.
[44]
Mikami, K.; Yajima, T.; Terada, M.; Uchimaru, T. Tetrahedron
Lett., 1993, 34, 7591.
[45]
Terada, M.; Mikami, K. J. Chem. Soc., Chem. Commun., 1994,
833.
[46]
Kitamoto, D.; Imma, H.; Nakai, T. Tetrahedron Lett., 1995, 36,
1861.
[47]
Mikami, K.; Motoyama, Y.; Terada, M. Inorganica Chimica Acta,
1994, 222, 71.
[48]
Terada, M.; Motoyama, Y.; Mikami, K. TetrahedronLett., 1994,
35, 6693.
[49]
It should be noted that the chiral titanium complexes derived
from binaphthol ditrifylamine (R)-49 afforded very low levels of
asymmetric induction (only 3% ee).
[50]
Mikami, K.; Matsukawa, S. Tetrahedron Lett., 1994, 35, 3133,
[51]
Mikami, K.; Yoshida, A. Synlett, 1995, 29.
[52]
Mikami, K.; Narisawa, S.; Shimizu, M.; Terada, M. J. Am.
Chem. Soc., 1992, 114, 9242.
[53]
(a) Mikami, K.; Yoshida, A.; Matsumoto, Y. TetrahedronLett.,
1996, 37, 8515. (b) see also: Mikami, K.; Yoshida, A. Angew.
Chem., Int. Ed. Engl., 1997, 36, 858.
[54]
Faller, J.W.; Liu, X. Tetrahedron Lett., 1996, 37, 3449.
[71]
Johannsen, M.; Jorgensen, K.A. J. Org. Chem., 1995, 60, 5757.
[55]
Mikami, K.; Osawa, A.; Isaka, A.; Sawa, E.; Shimizu, M.;
Terada, M.; Kubodera, N.; Nakagawa, K.; Tsugawa, N.; Okano,
T. Tetrahedron Lett., 1998, 39, 3359.
[72]
Terada, M.; Mikami, K.; Takai, T. TetrahedronLett., 1991, 32,
935.
[73]
[56]
A very similar approach for the synthesis of A ring has been
used by Courtney and coworkers: Courtney, L.F.; Lange, M.;
Uskokovié, M.R.; Noukulich, P.M. Tetrahedron Lett., 1998, 39,
3363.
Gao, Y.; Lane-Bell, P.; Vederas, J.C. J. Org. Chem., 1998, 63,
2133.
[74]
(a) Evans, D.A.; Burgey, C.S.; Paras, N.A.; Vojkovsky, T.;
Tregay, S.W. J. Am. Chem. Soc., 1998, 120, 5824. (b) In a
recent paper, Kanemasa and coworkers described the
preparation and use of transition-metal aqua complexes of 4,6dibenzofurandiyl-2,2'-bis(4-phenyloxazoline).Thecationic aqua
complexes prepared from their enantiopure ligand and various
transition-metal (II) perchlorates are effective catalysts for the
Diels-Alder reaction with excellent enantioselectivity and
extreme chiral amplification, and high tolerance to water,
alcohols, amines, and acids: Kanemasa, S.; Oderaotoshi, Y.;
Sakaguchi, S.; Yamamoto, H.; Tanaka, J.; Wada, E.; Curran,
D.P. J. Am. Chem. Soc., 1998, 120, 3074.
[75]
Evans, D.A.; Johnson, J.S.; Burgey, C.S.; Campos, K.R.
Tetrahedron Lett., 1999, 40, 2879.
[57]
(a) Mikami, K.; Matsukawa, S. J. Am.Chem. Soc., 1993, 115,
7039. (b) Shoda, H.; Nakamura, T.; Tanino, K.; Kuwajima, I.
Tetrahedron Lett. 1993, 34, 6281.
[58]
(a) Mikami, K.; Matsukawa, S. Nature, 1997, 385, 613. (b)
Review on chiral ligand acceleration: Berrisford, D.J.; Bolm, C.;
Sharpless, K.B. Angew. Chem., Int. Ed. Engl., 1995, 34, 1059.
[59]
Mikami, K.; Matsukawa, S.; Volk, T.; Terada, M. Angew. Chem.,
Int. Ed. Engl., 1997, 36, 2768.
342 Current Organic Chemistry, 2000, Vol. 4, No. 3
Luiz Carlos Dias
[76]
Yao, S.; Fang, X.; Jorgensen, K.A. Chem. Commun., 1998, 2547.
[79]
Elder, A.M.; Rich, D.H. Org. Lett., 1999, 1(9), 1443.
[77]
In this same paper, the authors have observed that the use of bis(oxazolines) and different Lewis acids leads only to moderate
ee’s.
[80]
For a very exciting review paper about nonlinear effects in
asymmetric synthesis, see: Girard, C.; Kagan, H.B. Angew.
Chem., Int. Ed. Engl., 1998, 37, 2922.
[78]
Drury, W.J., III.; Ferraris, D.; Cox, C.; Young, B.; Lectka, T. J.
Am. Chem. Soc., 1998, 120, 11006.