Investigations of enantioreversal in both direct and directed enantioselective aldol reactions catalyzed by CuCl2[(-)-sparteine] and NiCl2[(-)-sparteine] complexes

Article abstract of DOI:10.1016/j.tetasy.2010.07.008

The use of structurally well-defined chiral [CuCl₂(sparteine)] 1 and [NiCl₂(sparteine)] 2 complexes as catalysts under fluoride anion-promoted double catalytic activation (DCA) conditions cause enantioreversal in the asymmetric Mukaiyama aldol reaction of 1-phenyl-1-trimethylsiloxyethylene with various aromatic aldehydes. A similar enantioreversal also occurs in the direct aldol reaction between methyl vinyl ketone and various aromatic aldehydes under Et₃N promoted DCA conditions. We have quantitatively analyzed using group theoretical analysis via symmetry deformation coordinates, the molecular structures of 1 and 2 from the X-ray structural data; the results from the study show that the innate stereochemical differences that are present in their molecular structures, form the basis for the enantioreversal phenomenon in the asymmetric C-C bond-forming aldol reactions.

Full text of DOI:10.1016/j.tetasy.2010.07.008

Tetrahedron: Asymmetry 21 (2010) 2158–2166
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Investigations of enantioreversal in both direct and directed
enantioselective aldol reactions catalyzed by CuCl2[(ꢀ)-sparteine]
and NiCl2[(ꢀ)-sparteine] complexes
*
H. Maheswaran , P. J. Amal Joseph, K. Leon Prasanth, S. Priyadarshini, P. Satyanarayana, Praveen R. Likhar,
M. Lakshmi Kantam
Inorganic and Physical Chemistry Division, Indian Institute of Chemical Technology, Hyderabad, India
a r t i c l e i n f o
a b s t r a c t
Article history:
The use of structurally well-defined chiral [CuCl2(sparteine)] 1 and [NiCl2(sparteine)] 2 complexes as cat-
alysts under fluoride anion-promoted double catalytic activation (DCA) conditions cause enantioreversal
in the asymmetric Mukaiyama aldol reaction of 1-phenyl-1-trimethylsiloxyethylene with various aro-
matic aldehydes. A similar enantioreversal also occurs in the direct aldol reaction between methyl vinyl
ketone and various aromatic aldehydes under Et3N promoted DCA conditions. We have quantitatively
analyzed using group theoretical analysis via symmetry deformation coordinates, the molecular struc-
tures of 1 and 2 from the X-ray structural data; the results from the study show that the innate stereo-
chemical differences that are present in their molecular structures, form the basis for the enantioreversal
phenomenon in the asymmetric C–C bond-forming aldol reactions.
Received 8 June 2010
Accepted 6 July 2010
Available online 31 July 2010
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
late 1980s. However, it is only with the recent publication of the
three reviews2 related to the subject that the focus has been aimed
The demand for enantiomerically pure single enantiomer com-
pounds has been steadily increasing within the pharmaceutical
industry.1 Despite significant advances in asymmetric synthesis/
catalytic technologies, the preparation of both enantiomers of a
target molecule remains a challenging endeavor because many chi-
ral building blocks, chiral ligands, and chiral Lewis acid catalysts
are often derived from natural sources (chiral pool), and thus are
often only readily available in one absolute configuration. The
other antipode, either in D- or L-form, may have a cost premium
associated with it, may require an expensive and lengthy synthesis.
Thus, the strategies involving enantiodivergent asymmetric cataly-
sis that enables dual or reversible enantiocontrol using the same
chiral ligands for obtaining two antipodal products merits serious
consideration.2,3
at the specificity of this topic of increasing interest.
In our previous paper, we have disclosed that chelation of steri-
cally demanding (ꢀ)-sparteine ligand to copper(II) results in the
precise counter anion-dependent stereochemical outcomes, and
that contributes to significant differences in their reactivities and
selectivities in asymmetric C–C bond-forming reactions.4 Herein,
we report that the reversal of enantioselectivity with the same chi-
ral ligand, (ꢀ)-sparteine is readily achieved by changing the Lewis
acidic center from copper(II) to nickel(II) in dichloro[(ꢀ)-sparteine-
N,N0]metal(II) catalysts for both direct aldol and directed aldol
reactions that are promoted under double catalytic activation
(DCA) conditions. The details of these interesting observations
are the focus of this paper.
In the literature, enantiodivergent asymmetric catalysis has
been achieved by the addition of an achiral ligand or additive, by
changing the counter ion in the chiral organometallic complex, or
the reaction conditions, such as temperature, pressure, and sol-
vent.2 After a seminal paper in the early 1970s by Mosher, where
enantioreversal of asymmetric reduction of carbonyl compounds
depended upon aging of the stoichiometric chiral alkoxy-alumini-
umhydride reducing agent employed, a range of intriguing exam-
ples have been slowly appeared in the literature beginning in the
2. Results and discussion
First,amixtureof4-nitrobenzaldehyde(1.0 mmol)and/or2-nitro-
benzaldehyde (1.0 mmol) and silyl nucleophile (1.4 mmol) was
added in one portion to DMF (2 mL) in the presence of 0.20 mmol
(20 mol % relative to aldehyde) of preformed [CuCl2[(ꢀ)-sparte-
ine,N,N0] 1 or [NiCl2[(ꢀ)-sparteine,N,N0] 2 as catalysts and the result-
ing mixture was stirred for 24 h (Scheme 1). After routine
acidic–aqueous work-up, and silica-gel column purification, these
model reactions gave directly the corresponding aldol adducts in
low to moderate yields (25–48%). The most interesting aspect of the
reaction is present in the stereoselectivity (observed sign of the
* Corresponding author. Tel.: +91 040 27191532; fax: +91 040 27160921.
E-mail address: maheswaran_dr@yahoo.com (H. Maheswaran).
0957-4166/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2010.07.008

H. Maheswaran et al. / Tetrahedron: Asymmetry 21 (2010) 2158–2166
2159
OSiMe3
20 mol% 1 or 2
1 = CuCl2[(-)-sparteine]
OH
O
CHO
+
Ph
2 = NiCl2[(-)-sparteine]
DMF (2mL); 30oC; 24h
R
R
1.4 mmol
1.0 mmol
44% yield; (+)-(R)-isomer
57% ee with 1
R = 4-NO2
R = 4-NO2
R = 2-NO2
38% yield; (-)-(S)-isomer
50% ee with 2
25% yield; (-)-(S)-isomer
48% ee with 1
R = 2-NO2
44% yield; (+)-(R)-isomer
68% ee with 2
Scheme 1. Directed Mukaiyama aldol reactions of nitrobenzaldehydes with 1-phenyl-1-trimethylsiloxyethylene catalyzed by 1 and 2.
specific rotation) of the model reactions, which depend solely on the
nature of the aldehyde substrate employed. For example, for aldol ad-
ducts derived from 4-nitrobenzaldehyde, a strong preference for the
(R)-isomer with 57% ee was obtained with catalyst 1; while those of
one catalyzed by 2 showed a preference for the (S)-isomer with 50%
ee.5 In contrast, for aldol adducts derived from 2-nitrobenzaldehyde,
catalyst 1 shows preference for (S)-isomer with 48% ee; while the use
ofcatalyst 2 gives (R)-isomer with 68% ee.5 To our surprise, this model
reaction (Scheme 1) gives aldol products ininsignificant yields (<10%)
when other aromatic aldehyde substrates that do not contain a
strongly electron-withdrawing –NO2 group were evaluated as sub-
strates. The results are very distinct from those reported with dicat-
ionic and unsaturated square-planar [(ꢀ)-sparteine)Pd-[SbF6]2
complex as a catalyst with various aldehyde substrates, which have
been shown to yield aldol products with a strong preference for the
(S)-isomer of the aldol adducts in DMF.6
Apparently, the course of the reaction under a DCA protocol is very
sensitive to substitution at the ortho-position of the aldehyde sub-
strate. For example, the reaction of 2-nitrobenzaldehyde with a silyl
ether in the presence of 1ꢁKF or 2ꢁKF adducts gave not only low yields
but also imparted low enantioselectivity for the aldol adducts with a
preference for (R)-isomer without enantioreversal in their aldol ad-
ducts (Table 2, entries 11 and 12). In contrast, the reaction of 2-nitro-
benzaldehyde with a silyl ether, results in enantioreversal in the
aldol adducts, in particular, when the reactions are carried out in
the absence of KF with catalysts 1 and 2, respectively (Table 2, en-
tries 13 and 14 and Scheme 1). It appears that the enantioreversal
in the aldol reaction is also sensitive to the lack of heteroatom-based
functionalities in the aldehyde substrates. For instance, the reac-
tions of the silyl ether with benzaldehyde and propanaldehyde give
aldol adducts in moderate to good yields with modest enantioselec-
tivity with a preference for the (S)-isomer but no enantioreversal
was observed (Table 2, entries 15–18).
To enable a variety of aldehyde substrates applicable for the
asymmetric Mukaiyama aldol reaction with catalysts 1 and 2, we
have exploited double catalytic activation (DCA)7 protocols using
potassium fluoride (KF)8,9 as an activator for a silyl ether nucleo-
phile. Notably, when KF was used directly as additives for the mod-
el reaction with 4-nitrobenzaldehyde depicted in Scheme 1, only
poor yields for the aldol products were obtained with the forma-
tion of acetophenone as the major product.8,9 Interestingly, how-
ever, when KF and catalysts 1 or 2 were heated in methanolic
solutions at 55 °C for 12 h, and the complexes that precipitated
upon cooling (here after called 1ꢁKF or 2ꢁKF adducts) were directly
employed as catalysts, excellent yields for the aldol products were
obtained. We have optimized the reaction conditions for the model
reaction with 4-nitrobenzaldehyde under such DCA conditions
with varying amounts of KF and catalysts 1 and 2 (see Table 1).
The best results [86% yield and 61% ee for the (R)-isomer with 1;
86% yield and 63% ee for the (S)-isomer with 2] were obtained
when the reaction was carried out under reasonably dilute condi-
tions in DMF (12 mL) with an equimolar amounts (20 mol %) of KF
with 1 or 2 (1ꢁKF or 2ꢁKF adducts) as catalysts (see Table 2, entries 1
and 2). With the optimized conditions in hand (see Table 1, entries
6 and 12), the scope of the reaction was explored with 1ꢁKF and
2ꢁKF adducts as catalysts using various aldehyde substrates. As
can be seen in Table 2, the aromatic aldehyde substrates containing
functional groups such as –NO2, (both at a para-position and a
meta-position; see Table 2, entries 1–4) –CF3, and –CN (at para-
positions; see Table 2, entries 5–8) gives good yields with modest
enantioselectivities with concomitant enantioreversal for the aldol
adducts 5a–5d with catalysts 1ꢁKF and 2ꢁKF, respectively.
To verify whether the observed enantioreversal with catalysts 1
and 2 is an accidental phenomenon, and is limited to directed
Mukaiyama aldol reactions, we have studied the direct aldol reac-
tion of various aromatic aldehydes with methyl vinyl ketone10
(MVK) under Et3N-promoted DCA conditions by using optimized
reaction conditions with 4-nitrobenzaldehyde (see Scheme 2 and
Table 3 for optimization data). As was the case with Scheme 1, the
most interesting aspect of the direct aldol reaction (Scheme 2) is
present in the stereoselectivity (observed sign of the specific rota-
tion) of the product, which depends on the aldehyde substrate em-
ployed. For direct aldol adducts derived from 4-nitrobenzaldehyde
3a, a strong preference for the (R)-isomer with 79% ee was obtained
with catalyst 1; while those of one catalyzed by 2 showed a prefer-
ence for the (S)-isomer with 56% ee in DMF (76% ee in MeOH).11 In
contrast, for the direct aldol adducts derived from 2-nitrobenzalde-
hyde 3f, catalyst 1 shows a preference for the (S)-isomer with 37% ee
in DMF; while the use of catalyst 2 gives the (R)-isomer with 52% ee.
Similarly, activated aldehyde substrates such as 3-nitrobenzalde-
hyde 3b, 4-cyanobenzaldehyde 3c, and 4-chloro-3-nitrobenzalde-
hyde 3i also underwent direct aldol reaction with MVK under the
optimized experimental conditions, and gave good yields with a
concomitant enantioreversal with moderate enantioselectivity for
the aldol adducts (see Table 4; entries 1–10).
The above results strongly demonstrate that the observed
enantioreversal phenomenon is rather general in nature for the
aldol reactions, and the cause of the phenomenon is pertinent to
the nature of the distinct stereochemical characteristics of cata-
lysts 1 and 2. It should be noted that the results presented in Tables
2 and 4 also illustrate the fact that the obtained ee values with var-
ious aldehyde substrates in both direct and directed aldol reactions
are only somewhat moderate to good with the highest ee values
We have also observed similar results with heterocyclic aldehyde
substrate 3e (4-pyridaldehyde) under our experimental conditions,
albeit with lower enantioselectivity (Table 2, entries 9 and 10).

2160
H. Maheswaran et al. / Tetrahedron: Asymmetry 21 (2010) 2158–2166
Table 1
Optimization of the reaction parameters for the reaction of 4-nitrobenzaldehyde with 1-phenyl-1-trimethylsilyloxyethene under double catalytic conditionsa
OSiMe3
OH
O
CHO
dichloro[(-)-sparteine-N,N']Metal(II).KF
Ph
(20 mol%)
+
DMF, RT
O2N
O2N
1.0 mmol
1.4 mmol
KFb (mmol)
DMF (mL)
Yieldc (%)
Time (h)
eee (%)
½
a 2D5
ꢂ
(%)
d
Entry
Metal
1
2
3
Cu
Cu
Cu
Cu
Cu
Cu
Cu
Cu
Ni
0.20
0.40
0.20
0.20
0.20
0.20
0.10
0.30
0.20
0.20
0.20
0.20
2
2
3
6
9
12
12
12
3
6
9
80
26
77
64
90
86
68
60
67
70
79
81
12
17
12
12
12
12
12
12
12
12
12
12
+19.6
+21.5
+19.5
+21.5
+27.8
+35.8
+28.4
+24.6
ꢀ14.9
ꢀ19.5
ꢀ26.5
ꢀ36.9
30 (R)
33 (R)
30 (R)
33 (R)
4
5
46 (R)
6g
7
55 (R) (61)f
47 (R)
8
9
10
11
12g
42 (R)
23 (R)
30 (S)
Ni
Ni
Ni
45 (S)
12
57 (S) (63)f
a
b
c
All reactions were performed on a 1.0 mmol scale with 0.20 mmol or 20 mol % of catalyst 1 or 2 and 1.4 equiv of 1-phenyl-1-trimethylsilyloxyethene at 30 °C.
The amount of KF employed is expressed in mmol (mol %) relative to aldehyde substrate.
Values are isolated yields after chromatographic purification.
Specific rotation was measured in CHCl3 with c = 1 at 25 °C.
Enantiomeric excess and the absolute configuration [(R)-isomer or (S)-isomer] of the aldol adduct was assigned by comparison with the specific rotation data of a known
d
e
compound. Lit. ½a D22
ꢂ
¼ þ65:3 (c 1, CHCl3).
f
For entries 6 and 12, enantiomeric excess (ee) was determined by HPLC using Chiralcel AS-H column using (80:20 hexane/isopropanol, with flow rate of 0.7 mL/min, with
UV detection at 254 nm).
g
Entries 6 and 12 correspond to the optimized reaction conditions for reaction under DCA conditions.
being obtained in the range of 77–79% for selected substrates.
However, to the best of our knowledge, this is the first report
involving chiral (ꢀ)-sparteine ligand, that is, showing an enantior-
eversal phenomenon for the asymmetric C–C bond-forming reac-
tions. It is noteworthy that the (ꢀ)-sparteine ligand does not
have a readily available antipode (+)-sparteine, even though (+)-
sparteine surrogates are well-known.12 Hence, a better under-
standing of underlying stereochemical principles that lead to
enantioreversal in these reactions with 1 and 2 in the aldol reac-
tions are very desirable because the results from the study could
enable the rational design of catalysts for enantiodivergent appli-
cations. In order to gain insights into the nature of distinct stereo-
chemical characteristics that are present in the catalysts 1 and 2,
we have quantitatively studied the X-ray crystallographic data of
these complexes using group theoretical analysis via symmetry
deformation coordinates, and the details of our study are discussed
below.
to a rectangular planar arrangement, which degenerates more to-
ward linear arrangements.15 Notably, such distortions are more
pronounced in catalyst 1 than catalyst 2, which is consistent with
the balance of the electronic effect of the d9 system (Jahn–Teller
distortion) in the former with the steric effect of the bulky
(ꢀ)-sparteine ligand.18 Thus, the crystallographically observed
molecular structures of 1 and 2 are indeed very distinct and are
best described as intermediate structures that lie along the spread
pathway for the interconversion of the tetrahedron and rectangu-
lar-planar arrangements.15,19 The molecular structures of both
the complexes 1 and 2 have also been characterized by various
spectroscopic techniques in solution. The results from these stud-
ies indicate crystallographically observed molecular structures
around the copper(II) and nickel(II) sites are retained in the solu-
tion-state as well.20
A root-mean-square (rms) overlay of Cu, Ni, N1, and N2 atoms
of X-ray structures of complexes 1 and 2 is presented in Figure 2.21
As can be seen there are two different ways of rms overlay of these
structures can be carried out. First, the structure presented in top
(A) correspond to rms overlay of Cu(1)–Ni(2) with N1(1)–N1(2),
N2(1)–N2(2) atoms of sparteine ligand in these complexes. The
structure (B) corresponds to rms overlay of Cu(1)–Ni(2) with
N1(1)–N2(2) and N2(1)–N1(2) atoms of sparteine ligand in these
complexes. In structure A, the conformations of all four alkaloid
rings (A–D) of sparteine framework fit in a nearly identical fashion
with significant differences present in the relative orientation of
chloride counter anions. In contrast, in the overlay structure (B),
only B and C alkaloid ring conformations overlay in good fashion,
while terminal alkaloid rings (A and D) adopt distinct and opposite
non-superimposable configurations (exo–endo fashion) towards
the metal center.
The X-ray structures of both [CuCl2[(ꢀ)-sparteine,N,N0] 1 and
[NiCl2[(ꢀ)-sparteine,N,N0] 2 have been reported,13 and the results
from these studies show that the metal atoms (Cu and Ni) in these
complexes are four-coordinate, and have a distorted tetrahedral
geometry which show severe distortions or deformations in their
bond angles.14 As discussed above, symmetry deformation coordi-
nates via group theoretical analysis for both the complexes 1 and 2
have been computed with reference to ideal Td reference geome-
try.15–17 This symmetry analyses indicate angular distortions or
deformations in these complexes occur primarily via degenerate
S2a(E), and S2b(E) coordinates18 (see Fig. 1), which describe com-
pression and twist-type deformations, respectively, leading from
tetrahedral to planar MX2X0 fragments. For complex 1, the com-
2
puted S2a(E) and S2b(E) vectors are ꢀ19.5° and ꢀ29.4°, respectively;
and for complex 2, the corresponding values are ꢀ7.2° and ꢀ12.6°,
respectively. Both the magnitude and the signs of S2a/S2b show both
molecular structures 1 and 2 undergoes tetragonal compression
along S2a(E) followed by diagonal twist-type elongation (S2b(E))
Naturally, the question arises as to which one of these two over-
layed structures shown in Figure 2 adequately represents the crys-
tallographically observed molecular structures of 1 and 2. We
show that such precise distinctions can be made when angular

H. Maheswaran et al. / Tetrahedron: Asymmetry 21 (2010) 2158–2166
2161
Table 2
Asymmetric Mukaiyama aldol reaction of 1-phenyl-1-trimethylsilyloxyethene with various aldehydes catalyzed by 1ꢁKF and 2ꢁKF adductsa
CuCl2[(-)-sparteine].KF 1
OSiMe3
or
OH
O
NiCl2[(-)-sparteine].KF 2
RCHO
+
R
Ph
°
DMF (12 mL); 30 C; 12h
3
5
4
R = 4-NO2-C6H4; 3a
R = 3-NO2-C6H4; 3b
R = 4-CN-C6H4; 3c
R = 4-CF3-C6H4; 3d
R = C5H4N(4-pyridyl); 3e
R = 2-NO2-C6H4; 3f
R = C6H5; 3g
R = 4-NO2-C6H4; 5a
R = 3-NO2-C6H4; 5b
R = 4-CN-C6H4; 5c
R = 4-CF3-C6H4; 5d
R = C5H4N(4-pyridyl); 5e
R = 2-NO2-C6H4; 5f
R = C6H5; 5g
R = C3H7; 3h
R = C3H7; 5h
R = 4-Cl,3-NO2-C6H4; 3i
Entry
Catalyst
Substrate
Product
Yieldb (%)
eec,d (%)
1
2
3
4
5
6
7
8
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
3a
3a
3b
3b
3c
3c
3d
3d
3e
3e
3f
3f
3f
3f
3g
3g
3h
3h
5a
5a
5b
5b
5c
5c
5d
5d
5e
5e
5f
5f
5f
5f
5g
5g
5h
5h
86
86
82
80
64
69
78
76
60
67
34
33
25
44
35
77
61
51
61 (R)
63 (S)
61 (R)
77 (S)
23 (R)
55 (S)
24 (R)
43 (S)
26 (R)
26 (S)
12 (R)
27 (R)
48 (S)
68 (R)
8 (S)
9
10
11
12
13e
14e
15
16
17
18
52 (S)
40 (S)
40 (S)
a
All reactions were performed on a 1.0 mmol scale with 0.20 mmol or 20 mol % of catalyst 1ꢁKF or 2ꢁKF and 1.4 equiv of 1-phenyl-1-trimethylsiloxyethylene at 30 °C in
12 mL DMF for 12 h.
b
Values are isolated yields after chromatographic purification.
Enantiomeric excess (ee) was determined by HPLC using Chiralcel AS-H and OD-H columns using isopropanol and hexanes.
Absolute configuration (see Ref. 5) of the aldol adduct was tentatively assigned by comparison of the signs of the specific rotation data of the aldol with those reported for
c
d
5a and 5g as reference compounds.
e
The results of the data obtained when the reactions were carried out using 20 mol % (relative to aldehyde) of preformed catalysts 1 and 2 in the absence of KF with 2 mL of
DMF.
20 mol% 1 or 2
1 = CuCl2[(-)-sparteine]
OH
O
CHO
+
O
2 = NiCl2[(-)-sparteine]
Et3N(10 mol%)
R
DMF (1mL); 30 o
24 - 48 hrs
C
R
1.0 mmol
10 mmol
61 % yield; (+)-(R)-isomer
79 % ee with 1
R = 4-NO2
R = 2-NO2
R = 4-NO2
R = 2-NO2
72 % yield; (-)-(S)-isomer
56 % ee with 2
58 % yield; (-)-(S)-isomer
37 % ee with 1
75 % yield; (+)-(R)-isomer
52 % ee with 2
Scheme 2. Direct aldol reactions of nitrobenzaldehydes with methyl vinyl ketone under DCA conditions.
deformation vectors that occur along S4(T2) deformation coordi-
such a reversal of signs in S4b and S4c deformation vectors are tech-
nically only feasible when two enantiomeric non-superimposable
molecular structures are compared. Hence, precisely speaking,
the (ꢀ)-sparteine ligand configuration in complex 1 in a sense mi-
mic its antipodal structure (+)-sparteine in complex 2 or vice versa.
Thus, overlay structure (B) is the most reasonable representation of
nates are analyzed. For complex 1, the computed S4a, S4b, and S4c
vectors are 11.6°, 1.18°, and 5.4°, respectively; and for complex 2,
the corresponding values are 21.3°, ꢀ2.0°, and ꢀ9.6°, respectively.
As can be seen S4a remains positive in both 1 and 2 complexes;
while S4b and S4c have opposite signs. As can be seen in Figure 3,

2162
H. Maheswaran et al. / Tetrahedron: Asymmetry 21 (2010) 2158–2166
Table 3
Table 4
Optimization of reaction parameters for the reaction of 4-nitrobenzaldehyde with
Asymmetric direct aldol reaction of methyl vinyl ketone with various aldehydes
methyl vinylketone (MVK) under double catalytic conditions (for Scheme 2)a
catalyzed by 1 and 2 promoted by Et3N under DCA conditionsa
OH
O
[1] or [2] (20 mol%)
O
OH
O
O
Catalyst, Additive
Solvent, RT
Et3N (5mol%-10mol%)
+
RCHO +
RCHO
R
R
MeOH or DMF
6
7
Entryb,c
Additive
Mol (%)
Solvent
Time (h)
Yield (%)
eef (%)
Entry Catalystb Substrate
R
Product R Time (h) Yieldd (%) eee,f (%)
1
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
(iPr)2EtN
NaHCO3
K2CO3
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
10
10
10
10
10
10
10
10
10
20
40
10
15
20
10
120
20
10
10
10
10
10
10
10
10
5
MeOH
DMF
DCM
ACN
60
45
55
60
60
45
48
65
55
15
11
40
25
15
52
55
48
3
22
25
22
30
31
21
20
5
25
61
35
39
34
56
53
32
38
73
71
57
62
68
51
44
65
78
72
67
71
63
69
73
70
80
82
67
71
35 (R)
79 (R)
65 (R)
72 (R)
69 (R)
64 (R)
75 (R)
42 (R)
59 (R)
54 (R)
53 (R)
72 (R)
55 (R)
48 (R)
62 (R)
58 (R)
72 (R)
55 (S)
56 (S)
31 (S)
42 (S)
26 (S)
8 (S)
2g
1
2
1
2
3a
3a
7a
7a
45
5
61
80,
72c
55
75
48
65
58
81
75c
54
71
79 (R)
76 (S)
56 (S)c
31 (R)
35 (S)
24 (R)
49 (S)
37 (S)
65 (R)
52 (R)c
44 (R)
54 (S)
3
4
5
6
7
8
9
THF
3
4
5
6
7
8
1
2
1
2
1
2
3b
3b
3c
3c
3f
7b
7b
7c
7c
7f
48
6
52
10
42
5
DMSO
CHCl3
C6H6
Neat
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
MeOH
DMF
DCM
ACN
10
11
12
13
14d
15
16
17
18e
19e
20e
21e
22e
23e
24e
25e
26e,g
27e
28e
29e
3f
7f
9
10
1
2
3i
3i
7i
7i
48
7
a
All reactions were performed at 1.0 mmol scale with 0.20 mmol or 20 mol % of
catalyst 1 or 2 and 10 equiv of methyl vinyl ketone at 30 °C. All reactions with
catalyst 1 were conducted in DMF (1 mL) and all reactions with catalyst 2 were
conducted in MeOH (1 mL).
For the protocol with Cu(sp)cl2[1], 10 mol % of Et3N and 1 ml of DMF and for
Ni(sp)cl2[2], 5 mol % of Et3N and 1 ml of methanol were used.
b
c
Reaction carried out with catalyst 2 in DMF.
Values are isolated yields after chromatographic purification.
The absolute configuration [(R)-isomer or (S)-isomer] of the aldol adduct was
THF
C6H6
IPA
d
e
28 (S)
32 (S)
76 (S)
49 (S)
38 (S)
70 (S)
tentatively assigned as (R) for (+)-rotation and (S) for (ꢀ)-rotation.
neat
f
Enantiomeric excess (ee) was determined by HPLC using Chiralcel AS-H, OD-H
MeOH
MeOH
MeOH
MeOH
and AD-H columns.
Et3N
Et3N
(iPr)2EtN
20
40
5
2
1.5
6
Mukaiyama aldol reaction of 1-phenyl-1-trimethylsiloxyethylene
with various aldehydes. A similar enantioreversal has also been
shown to occur in the direct aldol reaction between various alde-
hydes and methyl vinyl ketone under Et3N-promoted DCA condi-
tions. A quantitative analysis of the X-ray structures of catalysts
1 and 2 has been carried out using group theoretical analysis via
symmetry deformation coordinates, and the results from the study
unambiguously show (ꢀ)-sparteine ligand configuration in com-
plex 1 mimics its antipodal structure (+)-sparteine in complex 2
or vice versa. These innate stereochemical differences that are
present in their molecular structures of catalysts form the basis
for the observed enantioreversal phenomenon during the aldol
reaction with these catalysts.
a
All reactions were performed on a 1.0 mmol scale with 1 mmol of 4-nitro-
benzaldehyde and 10 mmol of methyl vinyl ketone at 30 °C.
b
Reactions (entries 1–17 except 12–14) were carried out using 0.20 mmol of
Cu(sp)cl2[1].
c
Reactions (entries 12 and 13) were carried out with 0.30 mmol Cu(sp)cl2[1].
Reaction (entry 14) was carried out with 0.40 mmol Cu(sp)cl2[1].
Reactions (entries 18–29) were carried out using 0.20 mmol of Ni(sp)cl2[2].
Enantiomeric excess (ee) was determined by HPLC using Chiralcel AS-H column
d
e
f
using isopropanol and hexanes.
g
Entries 2 and 26 correspond to the optimized reaction conditions for the
reaction under DCA conditions.
crystallographically observed molecular structures of complexes 1
and 2.
The above result is mainly due to differences in the nature and
types of distortions (both electronic and steric contributions) that
occurs in the molecular structures of these Cu and Ni complexes.18b
It is well known that the chelation of the sterically demanding (ꢀ)-
sparteine ligand to a metal ion occurs via specific inversion of one
nitrogen atom configuration. Our studies show that the inverted
nitrogen atom configuration both in complex 1 and complex 2 are
not crystallographically identical in their molecular structures
(overlay structure B), and hence (ꢀ)-sparteine structure in complex
1 mimics its antipodal structure, (+)-sparteine, in complex 2. As a
consequence, the use of these complexes as catalysts in C–C bond-
forming aldol reactions invariably results in the enantioreversal of
stereogenic center because chiral recognition process for enantiore-
versal require two antipodal structures of the chiral ligand.
4. Experimental
4.1. General remarks
All chemicals were purchased from Aldrich and were used as
received. All solvents used were analytical grade and were used
as received from Merck India Pvt. Ltd. Commercial column chroma-
tography grade silica gel (100–200 mesh) was first calcined at
200 °C for 6 h prior to use. Purification of the reaction products
was carried out by flash chromatography using calcined silica
gel, using a mixture of ethyl acetate and hexanes as eluting agent.
TLC was performed with glass-backed plates coated with 0.2 mm
silica (Merck, DC-Platten, Kieselgel: 60F254). All products were
characterized by 1H NMR spectroscopy. The 1H spectra of samples
3. Conclusion
were recorded on a Varian, Unity Innova 500 MHz, Innova
400 MHz, and Bruker-Avance-300 MHz Spectrometer using TMS
as an internal standard in CDCl3. The 13C spectra of samples were
recorded on a Bruker-Avance-75 MHz Spectrometer using TMS as
an internal standard in CDCl3. High performance liquid chromatog-
raphy (HPLC) was performed using an SHIMADZU 10AT series
We have shown that the use of structurally well-defined chiral
[CuCl2(sparteine)] 1 and [NiCl2(sparteine)] 2 complexes as cata-
lysts under fluoride anion-promoted double catalytic activation
(DCA) conditions causes enantioreversal in the asymmetric

H. Maheswaran et al. / Tetrahedron: Asymmetry 21 (2010) 2158–2166
2163
Z-axis
Cl1
Cl2
Cl2
Cl1
2
1
2
1
N3
N4
N4
N3
M
M
Enantiomer 2
4
Enantiomer 1
1
4
2
θ12
3
3
θ34
4
θ12
4
θ34
3
S4a(T2) ≅ θ12 − θ34
12 > θ34
S4a(T2) = + ve
S
4a(T2) ≅ θ12 − θ34
θ12 > θ34
3
θ
S4a(T2) = + ve
2
S2a(E)
S2b(E)
1
Figure 1. Pictorial representation of the bond-angle distortions represented as two
orthogonal symmetry displacement coordinates of the compression type (S2a(E), on
the left) and twist-type (S2b(E), on the right). Deformation along S2b(E) coordinate
(twist deformation) describes a twist of the two MX2 fragments (retaining local D2
symmetry). Along this coordinate h12 and h34 remain constant while two opposing
angles open up and two others close by an equivalent amount. In its limits, the
geometry becomes rectangular planar. If all angles adopt identical values (109.5°,
S2a(E) = S2b(E) = 0); the fragment is a tetrahedron. Deformation along S2a(E) coor-
dinates describes mutual opening and closing of a pair of opposite angles at the
metal center (retaining a local D2d symmetry), at the expense of the remaining four
bond angles. In its limit at S2a(E) = 103.9°, it describes a square planar geometry, at
1
2
θ13
θ24
S4b(T2) ≅ θ13 − θ24
θ13 > θ24
S4b(T2) = + ve
S4b(T2) ≅ θ24 − θ13
θ13 > θ24
S4b(T2) = - ve
3
4
4
3
θ13
2
2
θ24
1
1
1
2
S2a(E) = 0°;
a tetrahedron, and for negative values, the fragment degenerates
θ23
θ14
towards linear arrangements
S4c(T2) ≅ θ23 − θ14
θ14 > θ23
S4a(T2) = - ve
S4c(T2) ≅ θ14 − θ23
θ14 > θ23
S4c(T2) = + ve
3
θ23
4
θ14
4
3
Mirror plane
Figure 3. Angular deformations along S4(T2) coordinates in a tetrahedral structure:
a case for comparison of enantiomers in the distorted tetrahedral structures of
CuCl2(sparteine) 1 and NiCl2(sparteine) 2. In this figure, the bond angles that
describe the deformations along S4a(T2), S4b(T2), S4c(T2) coordinates are shown,
which are computed as
a
difference between two angles; S4a(T2) ꢃ h12 ꢀ h34;
S4b(T2) ꢃ h13 ꢀ h24 S4c(T2) ꢃ h14 ꢀ h23
;
.
As illustrated above, only positive S4(T2)
values will be obtained for one of the enantiomeric structure shown (left image to
imaginary mirror plane) because h12 > h34, h13 > h24, and h14 > h23. In contrast, for the
opposite enantiomeric structure (on the right side of imaginary mirror plane), two
opposite angles are inverted due to enantio-inversion; hence, S4a(T2) remains
positive; while S4b(T2) and S4c(T2) vectors are negative.
4.1.1. Preparation of dichloro[(ꢀ)-sparteine-N,N0]copper(II) 1
Copper(II) chloride (342 mg, 2 mmol) was first dehydrated by
heating at 110 °C under vacuum for several hours, which was then
taken in an oven dried 25 mL RB flask containing a solution of (ꢀ)-
sparteine (469 mg, 2 mmol) in methanol (6 ml). This mixture was
stirred under nitrogen atmosphere for 4 h. An olive green precipi-
tate quickly precipitated out, which was filtered to yield 692 mg
(89% yield) of dichloro[(ꢀ)-sparteine-N,N0]copper(II) complex. An
analytically pure compound was obtained by recrystallization from
hot methanol. Anal. Calcd for C15H26N2Cl2Cu: C, 48.87; H, 7.05; N,
7.59. Found: C, 48.28; H, 7.22; N, 7.43.
Figure 2. A rms overlay of Cu, Ni, N1, and N2 atoms from X-ray structures
of dichloro[(ꢀ)-sparteine-N,N0]copper(II) 1 (red) and dichloro[(ꢀ)-sparteine-N,N0]
nickel(II) complexes 2 (blue). Top picture (A) corresponds to overlay of Cu(1)–Ni(2),
with N1(1)–N1(2), N2(1)–N2(2) atoms of sparteine ligand (rms deviation =
0.00134 Å) Bottom picture (B) corresponds to an overlay of Cu(1)–Ni(2), with
N1(1)–N2(2), N2(1)–N1(2) atoms of sparteine ligand (rms deviation = 0.0124 Å).
liquid chromatograph equipped with UV detector using a CHIRAL-
CEL AS-H and OD-H columns with 2-propanol/hexanes as eluting
agent. Optical rotations were obtained on an automated JASCO
P-1020 polarimeter, and the values were reported in absolute rota-
4.1.2. Preparation of dichloro[(ꢀ)-sparteine-N,N0]nickel(II) 2
Nickel(II) chloride (476 mg, 2 mmol) was first dehydrated by
heating at 110 °C under vacuum for several hours, which was then
taken in an oven dried 25 mL RB flask containing 2 mL of methanol.
This was stirred under nitrogen atmosphere until it was com-
pletely dissolved. To this was added a solution of (ꢀ)-sparteine
(469 mg, 2 mmol) in 8 ml of ethyl acetate after which it was re-
fluxed at 72 °C for 24 h. A purple colored precipitate settled out,
which was filtered to yield 597 mg (82%) of dichloro[(ꢀ)-sparte-
ine-N,N0]nickel(II) complex. The analytically pure compound was
obtained by recrystallization from dichloromethane–triethylortho-
tions: ½a 2D8
ꢂ
(c 1, CHCl3) 1 mL of solvent]. Infrared (IR) absorption
data were acquired on a Thermo Nicolet Nexus 670 FT-IR spec-
trometer with DTGS KBr detector. Elemental Analysis of complexes
was carried out on ELEMENTAR VARIO-EL CHNS analyzer. Mass
Spectrometry of the samples was acquired on a Micromass, Quat-
tro LC Mass or Shimadzu, QP 2010+ spectrometer. The rms overlay
of X-ray structures 1 and 2 was obtained by calculations involving
published X-ray crystal structure data using Mercury software
(Version 1.4.1).

2164
H. Maheswaran et al. / Tetrahedron: Asymmetry 21 (2010) 2158–2166
formate (5:1) solution. Anal. Calcd for C15H26N2Cl2Ni: C, 49.49; H,
7.20; N, 7.70. Found: C, 49.50; H, 7.14; N, 7.74.
4.2.1.2. 3-Hydroxy-3-(3-nitrophenyl)-1-phenyl-propan-1-one 5b.
1H NMR (300 MHz, CDCl3) d: 3.12–3.50 (m, 2H), 3.85 (br, 1H), 5.30–
5.50 (br, 1H), 7.30–8.30 (br, 9H); 13C NMR (75 MHz, CDCl3) d:
199.7, 148.6, 145.2, 136.2, 134.0, 131.8, 129.5, 128.8, 128.2,
122.6, 120.9, 69.0, 47.1; EI-MS (m/z): 271 (M+); enantiomeric ex-
cess was determined by HPLC with a Chiralcel AS-H column
(80:20) hexane/isopropanol, 0.7 mL/min, UV detection at 254 nm
HPLC data for aldol adducts derived from CuCl2[(ꢀ)-sparte-
ine](1)ꢁKF catalysts: major enantiomer (R) tr = 20.8 min, minor
4.1.3. Preparation of dichloro[(ꢀ)-sparteine-N,N0]copper(II)ꢁKF
adduct 1ꢁKF
Anhydrous potassium fluoride, (1 mmol, 58 mg), is taken in an
oven dried 25 mL RB flask containing a solution of preformed
dichloro[(ꢀ)-sparteine-N,N0]copper(II) 1, complex (1 mmol, 368
mg), in methanol (10 ml). This mixture was stirred under a nitro-
gen atmosphere at 55 °C for 12 h. Upon cooling in an ice-bath, a
deep green colored 1ꢁKF compound precipitated out, which was fil-
tered and washed with cold methanol, and dried to yield 366 mg
(86%). Anal. Calcd for C15H26N2Cl2CuꢁKF: C, 42.22; H, 6.09; N,
6.56. Found: C, 40.96; H, 6.01; N, 6.18. Without further purification,
enantiomer (S) tr = 17.8 min; 61% ee ½a D25
¼ þ48:7 (c 1.0, CHCl3).
ꢂ
HPLC data for aldol adducts derived from NiCl2[(ꢀ)-sparte-
ine](2)ꢁKF catalysts: major enantiomer (S) tr = 17.5 min, minor
enantiomer (R) tr = 21.1 min; 77% ee; ½a D25
¼ ꢀ49:5 (c 1.0, CHCl3).
ꢂ
this compound was used directly as
conditions.
a
catalyst under DCA
4.2.1.3. 3-Hydroxy-3-(4-cyanophenyl)-1-phenyl-propan-1-one 5c.
1H NMR (300 MHz, CDCl3) d: 3.17–3.43 (m, 2H), 3.75 (br, 1H), 5.36
(br, 1H), 7.40–7.70 (m, 7H), 7.88–7.96 (m, 2H); EI-MS (m/z): 251
(M+); enantiomeric excess was determined by HPLC with a Chiral-
cel AS-H column (70:30) hexane/isopropanol, 1.0 mL/min, UV
detection at 254 nm. HPLC data for aldol adducts derived from
CuCl2[(ꢀ)-sparteine](1)ꢁKF catalysts: major enantiomer (R) tr =
4.1.4. Preparation of dichloro[(ꢀ)-sparteine-N,N0]nickel(II)ꢁKF
adduct 2ꢁKF
Anhydrous potassium fluoride, (1 mmol, 58 mg), is taken in an
oven dried 25 mL RB flask containing a solution of preformed
dichloro[(ꢀ)-sparteine-N,N0]nickel(II) complex 2, (1 mmol, 364
mg), in methanol (10 ml). This mixture was stirred under a nitro-
gen atmosphere at 55 °C for 12 h. Upon cooling a violet colored
compound precipitated out, which was filtered and washed with
cold methanol, and dried to yield 359 mg (85%) of homogenous
dichloro[(ꢀ)-sparteine-N,N0]nickel(II)ꢁKF adduct. Anal. Calcd for
10.4 min, minor enantiomer (S) tr = 19.8 min; 23% ee; ½a D25
¼
ꢂ
þ19:0 (c 1.0, CHCl3). HPLC data for aldol adducts derived from
NiCl2[(ꢀ)-sparteine](2)ꢁKF catalysts: major enantiomer (S) tr =
19.8 min, minor enantiomer (R) tr = 10.4 min; 55% ee; ½a D25
¼
ꢂ
ꢀ34:0 (c 1.0, CHCl3).
C
15H26N2Cl2NiꢁKF: C, 42.70; H, 6.16; N, 6.64. Found: C, 41.69; H,
4.2.1.4. 3-Hydroxy-3-(4-trifluromethylphenyl)-1-phenylpro pan-
1-one 5d. 1H NMR (300 MHz, CDCl3) d: 3.22–3.38 (m, 2H), 3.66
(br, 1H), 5.36 (br, 1H), 7.50–7.68 (m, 5H), 7.41–7.49 (m, 2H),
7.90–7.97 (m, 2H); EI-MS (m/z): 294 (M+); enantiomeric excess
was determined by HPLC with a Chiralcel AS-H column (80:20)
hexane/isopropanol, 0.7 mL/min, UV detection at 254 nm. HPLC
data for aldol adducts derived from CuCl2[(ꢀ)-sparteine](1)ꢁKF
complexes: major enantiomer (R) tr = 12.6 min, minor enantiomer
6.41; N, 6.39. Without further purification, this compound was
used directly as catalyst under DCA conditions.
4.2. General reaction procedure for aldol reactions
4.2.1. General reaction procedure for Mukaiyama aldol reaction
A dry and nitrogen flushed 5 mL flask, equipped with a mag-
netic stirring bar, was charged with 1ꢁKF or 2ꢁKF adducts
(0.2 mmol), and DMF (12 ml) at room temperature (30 °C). The cor-
responding aldehyde (1 mmol), and 1-phenyl-1-trimethylsiloxy-
ethylene (0.3 mL, 1.4 mmol) were then added successively, and
the resulting solution was stirred at room temperature till the
completion of the reaction. The reaction was quenched with a sat-
urated aqueous solution of NH4Cl (10 mL), extracted with ethyl
acetate (3 ꢄ 20 mL). The combined organic layers were dried using
anhydrous Na2SO4, filtered, and the solvent was removed by evap-
oration. The crude product was purified by column chromatogra-
phy (silica gel: 100–200 mesh using ethyl acetate and hexanes)
to give the corresponding pure aldol adducts. Enantiomeric excess
was determined by HPLC using Chiralcel AS-H, and OD-H columns
using isopropanol and hexanes as eluting agent. The absolute con-
figuration of the aldol products were assigned by comparison with
specific rotation data of known compounds.5
(S) tr = 11.5 min; 24% ee; ½a D25
¼ þ20:0 (c 1.0, CHCl3). HPLC data
ꢂ
for aldol adducts derived from NiCl2[(ꢀ)-sparteine](2)ꢁKF com-
plexes: major enantiomer (S) tr = 12.8 min, minor enantiomer (R)
tr = 11.7 min; 43% ee; ½a D25
¼ ꢀ39:0 (c 1.0, CHCl3).
ꢂ
4.2.1.5. 3-Hydroxy-1-phenyl-3-pyridinyl-propan-1-one 5e. 1H
NMR (300 MHz, CDCl3) d: 3.12–3.42 (m, 2H), 5.32 (m, 1H), 7.30–
7.39 (m, 2H), 7.41–7.53 (m, 2H), 7.55–7.65 (m, 1H), 7.91–8.0 (m,
2H), 8.58 (br, 2H); EI-MS (m/z): 227 (M+); enantiomeric excess
was determined by HPLC with a Chiralcel OD-H column (80:20)
hexane/isopropanol, 0.5 mL/min, 254 nm. HPLC data for aldol ad-
ducts derived from CuCl2[(ꢀ)-sparteine](1)ꢁKF: major enantiomer
(R) tr = 22.0 min, minor enantiomer (S) tr = 24.4 min; 26% ee;
½
a 2D5
ꢂ
¼ þ8:0 (c 1.0, CHCl3). HPLC data for aldol adducts derived
from NiCl2[(ꢀ)-sparteine](2)ꢁKF catalyst: major enantiomer (S)
tr = 24.1 min, minor enantiomer (R) tr = 22.0 min; 26% ee; ½a D25
¼
ꢂ
ꢀ10 (c 1.0, CHCl3).
4.2.1.1. 3-Hydroxy-3-(4-nitrophenyl)-1-phenyl-propan-1-one 5a.
1H NMR (300 MHz, CDCl3) d: 3.13 (dd, J = 9.8, 18.1 Hz, 1H), 3.73 (dd,
J = 1.9, 18.1 Hz, 1H), 5.83 (dd, J = 1.9, 9.8 Hz, 1H), 7.47–8.04 (m,
9H); EI-MS (m/z): 271; enantiomeric excess was determined by
HPLC with a Chiralcel AS-H column (80:20) hexane/isopropanol,
0.7 mL/min, detection at 254 nm. HPLC data for aldol adducts de-
rived from CuCl2[(ꢀ)-sparteine](1)ꢁKF catalysts: major enantiomer
(R) tr = 27.3 min, minor enantiomer (S) tr = 39.7 min; 61% ee;
4.2.1.6. 3-Hydroxy-3-(2-nitrophenyl)-1-phenyl-propan-1-one 5f.
1H NMR (300 MHz, CDCl3) d: 3.02–3.20 (dd, J = 9.6, 17.6 Hz, 1H),
3.66–3.78 (dd, J = 2.2, 17.6 Hz, 1H), 3.85–4.10 (br, 1H), 5.70–5.95
(dd, J = 2.2, 9.6 Hz, 1H), 7.35–7.75 (m, 5H), 7.85–8.15 (m, 4H); EI-
MS (m/z): 271 (M+); enantiomeric excess was determined by HPLC
with a Chiralcel AS-H column (80:20) hexane/isopropanol, 0.7 mL/
min, 254 nm. HPLC data for aldol adducts derived from CuCl2[(ꢀ)-
sparteine](1)ꢁKF catalyst: major enantiomer (R) tr = 17.4 min, min-
½
a 2D5
ꢂ
¼ þ35:8 (c 1.0, CHCl3). Literature data: ½a D25
¼ þ65:3 (c 1.0,
ꢂ
CHCl3; 99% ee, (R)-isomer).5 HPLC data for aldol adducts derived
from NiCl2[(ꢀ)-sparteine](2). KF catalysts: major enantiomer (S)
tr = 37.4 min, minor enantiomer (R) tr = 27.0 min; 63% ee;
or enantiomer (S) tr = 18.8 min; 12% ee; ½a D25
¼ þ7:1 (c 1.0, CHCl3).
ꢂ
HPLC data for aldol adducts derived from NiCl2[(ꢀ)-sparte-
ine](2)ꢁKF catalyst: major enantiomer (R) tr = 17.1 min, minor
½
a 2D5
ꢂ
¼ ꢀ36:9 (c 1.0, CHCl3).
enantiomer (S) tr = 18.5 min; 27% ee; ½a D25
¼ þ15:0 (c 1.0, CHCl3).
ꢂ

H. Maheswaran et al. / Tetrahedron: Asymmetry 21 (2010) 2158–2166
2165
4.2.1.7. 3-Hydroxy-1,3-diphenylpropan-1-one 5g. 1H NMR
(300 MHz, CDCl3) d: 3.23–3.37 (dd, J = 5.9, 14.7 Hz, 2H), 3.39–
3.70 (br, 1H), 5.26–5.36 (m, 1H), 7.20–7.65 (m, 8H), 7.80–8.07
(m, 2H); EI-MS (m/z): 226 (M+). Enantiomeric excess was deter-
mined by HPLC with a Chiralcel AS-H column (80:20) hexane/iso-
propanol, 0.5 mL/min, 254 nm. HPLC data for aldol adducts derived
from CuCl2[(ꢀ)-sparteine](1)ꢁKF catalyst: major enantiomer (S)
tr = 11.0 min, minor enantiomer (R) tr = 13.1 min; 8% ee;
4.2.3.1. 5-Hydroxy-5-(4-nitrophenyl)pent-1-en-3-one 7a. IR
(KBr): 3477, 3107, 2903, 1977, 1673, 1602, 1512, 1400, 1340, 1219,
1080, 1027, 987, 888, 849, 784, 748, 699, 596; 1H NMR (400 MHz,
CDCl3) d: 2.87–3.01 (m, 2H), 3.62 (br s, 1H), 5.29 (br s, 1H), 5.94 (d,
J = 10.25 Hz, 1H), 6.25 (d, J = 17.58 Hz, 1H), 6.35 (dd, J = 10.25,
17.58 Hz, 1H), 7.54 (d, J = 8.79 Hz, 2H), 8.19 (d, J = 8.79 Hz, 2H); 13C
NMR (75 MHz, CDCl3) d: 199.9, 150.3, 147.3, 136.2, 130.3, 126.5,
123.7, 68.9, 47.5. ESI-MS (m/z):244(M+Na), 222(M+H). Enantiomeric
excess wasdeterminedbyHPLCwitha ChiralcelAS-Hcolumn(70:30)
hexane/isopropanol, 1.5 mL/min, detection at 254 nm. HPLC data for
aldol adducts derived from CuCl2[(ꢀ)-sparteine](1) catalyst: major
enantiomer (R) tr = 7.34 min, minor enantiomer (S) tr = 12.5 min;
½
a 2D5
ꢂ
¼ ꢀ8:1 (c 1.0, CHCl3). HPLC data for aldol adducts derived
from NiCl2[(ꢀ)-sparteine](2)ꢁKF catalyst: major enantiomer (S)
tr = 11.4 min, minor enantiomer (R) tr = 13.6 min; 52% ee;
½
a 2D5
ꢂ
¼ ꢀ49:2 (c 1.0, CHCl3).
79% ee; ½a 2D5
¼ þ30:2 (c 1.0, CHCl3). HPLC data for aldol adducts de-
ꢂ
4.2.1.8. 3-Hydroxy-1-phenylhexan-1-one 5h. 1H NMR (300 MHz,
CDCl3) d: 0.98–1.22 (m, 3H), 1.48–1.70 (m, 2H), 2.93–3.05 (dd,
J = 8.9, 17.6 Hz, 1H), 3.08–3.18 (dd, J = 2.3, 17.6 Hz, 1H), 4.11 (br
s, 1H), 7.40–7.6 (m, 3H), 7.88–8.10 (m, 2H); EI-MS (m/z): 178
(M+); enantiomeric excess was determined by HPLC with a Chiral-
cel AS-H column (80:20) hexane/isopropanol, 0.7 mL/min, 254 nm.
HPLC data for aldol adducts derived from both CuCl2[(ꢀ)-sparte-
ine](1) ꢁKF and NiCl2[(ꢀ)-sparteine](2) ꢁKF catalyst: major enantio-
mer (S) tr = 8.2 min, minor enantiomer (R) tr = 9.2 min; 40% ee;
rived from NiCl2[(ꢀ)-sparteine](2) catalyst: major enantiomer (S)
tr = 11.07 min, minor enantiomer (R) tr = 7.43 min; 76% ee;
½
a 2D5
ꢂ
¼ ꢀ33:1 (c 1.0, CHCl3).
4.2.3.2. 5-Hydroxy-5-(3-nitrophenyl)pent-1-en-3-one 7b. IR
(KBr): 3446, 3090, 2926, 1955, 1679, 1613, 1529, 1480, 1405, 1351,
1204, 1064, 983, 809, 738, 688, 560; 1H NMR (500 MHz, CDCl3) d:
2.91–3.05 (m, 2H), 3.57 (br s, 1H), 5.29 (br s, 1H), 5.93 (d,
J = 10.73 Hz, 1H), 6.27 (d, J = 17.56 Hz, 1H), 6.37 (dd, J = 10.73,
17.56 Hz, 1H), 7.52 (t, J = 7.81 Hz), 7.73 (d, J = 7.81 Hz), 8.13 (d,
J = 7.81 Hz, 1H), 8.22 (s, 1H); EI-MS (m/z): 244 (M+Na), 222 (M+H);
enantiomeric excess was determined by HPLC with a Chiralcel OD-
H column (85:15) hexane/isopropanol, 0.5 mL/min, UV detection at
254 nm. HPLC data for aldol adducts derived from CuCl2[(ꢀ)-sparte-
ine](1) catalyst: major enantiomer (R) tr = 23.80 min, minor enantio-
½
a 2D5
ꢂ
¼ ꢀ26:0 (c 1.0, CHCl3) for both aldol adducts obtained from
CuCl2[(ꢀ)-sparteine](1) ꢁKF and NiCl2[(ꢀ)-sparteine](2) ꢁKF cata-
lysts .
4.2.2. 2a. General reaction procedure for direct aldol reaction
with methyl vinyl ketone (MVK) under DCA conditions:
(procedure using catalyst 1)
mer (S) tr = 21.20 min; 31% ee; ½a D25
¼ þ29:2(c 1.0, CHCl3). HPLC data
ꢂ
A dry and nitrogen flushed 5 mL flask, equipped with a mag-
netic stirring bar, was charged with dichloro[(ꢀ)-sparteine-
N,N0]copper(II) catalyst (0.2 mmol), triethylamine (0.1 mmol) and
DMF (1 ml) at 0 °C. The corresponding aldehyde (1 mmol) and
methyl vinyl ketone (10 mmol) were then added successively,
and the resulting solution was stirred at 0 °C for an hour. It is then
gradually brought to room temperature (30 °C), and stirred until
the completion of the reaction. The reaction was quenched with
ice cold water (5 mL), and extracted with ethyl acetate (3 x
20 mL). The combined organic layers were dried using anhydrous
Na2SO4, filtered, and the solvent was removed by evaporation.
The crude product was purified by column chromatography (silica
gel: 100–200 mesh using ethyl acetate and hexane as eluent) to
give the corresponding pure aldol adducts. Enantiomeric excess
was determined by HPLC using Chiralcel AS-H, ADH, and OD-H col-
umns using isopropanol and hexanes as eluting agent.
foraldol adductsderivedfromNiCl2[(ꢀ)-sparteine](2)catalyst:major
enantiomer (S) tr = 21.05 min, minor enantiomer (R) tr = 23.78 min;
35% ee; ½a 2D5
¼ ꢀ34:0 (c 1.0, CHCl3).
ꢂ
4.2.3.3. 4-(1-Hydroxy-3-oxopent-4-enyl)benzonitrile 7c. IR
(KBr): 3461, 3102, 2910, 2229, 1934, 1677, 1611, 1504, 1405,
1293, 1201, 1066, 978, 884, 837, 756, 567; 1H NMR (300 MHz,
CDCl3) d: 2.85–3.01 (m, 2H), 3.53 (br s, 1H), 5.24 (br s, 1H), 5.93
(d, J = 9.82 Hz, 1H), 6.25 (d, J = 17.37 Hz, 1H), 6.36 (dd, J = 9.82,
17.37 Hz, 1H), 7.49 (d, J = 8.31 Hz, 2H),7.64 (d, J = 8.31 Hz, 2H);
13C NMR (75 MHz, CDCl3) d: 200.0, 148.1, 136.3, 132.5, 132.2,
130.1, 126.4, 118.8, 69.2, 47.5. ESI-MS (m/z): 224 (M+Na), 202
(M+H); enantiomeric excess was determined by HPLC with a Chi-
ralcel AS-H column (70:30) hexane/isopropanol, 1.0 mL/min,
detection at 254 nm. HPLC data for aldol adducts derived from
CuCl2[(ꢀ)-sparteine](1)
catalyst:
major
enantiomer
(R)
tr = 10.20 min, minor enantiomer (S) tr = 24.90 min; 24% ee;
4.2.3. 2b. General reaction procedure for the direct aldol reaction
with methyl vinyl ketone (MVK) under DCA conditions:
(procedure using catalyst 2)
½
a 2D5
ꢂ
¼ þ20:0 (c 1.0, CHCl3). HPLC data for aldol adducts derived
from NiCl2[(ꢀ)-sparteine](2) catalyst: major enantiomer (S)
tr = 23.64 min, minor enantiomer (R) tr = 10.20 min; 49% ee;
A dry and nitrogen flushed 5 mL flask, equipped with a mag-
netic stirring bar, was charged with dichloro[(ꢀ)-sparteine-
N,N0]nickel(II) catalyst (0.2 mmol), triethylamine (0.05 mmol),
and methanol (1 ml) at 0 °C temperature. The corresponding alde-
hyde (1 mmol) and methyl vinyl ketone (10 mmol) were then
added successively, and the resulting solution was stirred at 0 °C
for an hour. It was then gradually brought to room temperature
(30 °C), and stirred until the completion of the reaction. The reac-
tion was quenched with ice cold water (5 mL), and extracted with
ethyl acetate (3 x 20 mL). The combined organic layers were dried
using anhydrous Na2SO4, filtered, and the solvent was removed by
evaporation. The crude product was purified by column chroma-
tography (silica gel: 100–200 mesh using ethyl acetate and hexane
as eluent) to give the corresponding pure aldol adducts. Enantio-
meric excess was determined by HPLC using Chiralcel AS-H, ADH,
and OD-H columns using isopropanol and hexanes as eluting
agent.
½
a 2D5
ꢂ
¼ ꢀ34:0 (c 1.0, CHCl3).
4.2.3.4. 5-Hydroxy-5-(2-nitrophenyl)pent-1-en-3-one 7f. IR
(KBr): 3449, 3101, 2926, 1945, 1680, 1612, 1577, 1525, 1404,
1348, 1189, 1062, 973, 856, 788, 747, 706, 550; 1H NMR
(300 MHz, CDCl3) d: 2.76 (dd, J = 9.07, 17.37 Hz, 1H), 3.31 (d,
J = 17.37, 1H), 3.85 (br s, 1H), 5.67 (d, J = 9.07, 1H), 5.93 (d,
J = 9.82 Hz, 1H), 6.32 (d, J = 17.37 Hz, 1H), 6.38 (dd, J = 9.82,
17.37 Hz, 1H), 7.43 (t, J = 6.80 Hz, 15.11 Hz, 1H), 7.65 (t,
J = 7.55 Hz, 15.11 Hz, 1H), 7.95 (t, J = 6.80 Hz, 7.55 Hz, 2H); 13C
NMR (75 MHz, CDCl3) d: 200.2, 147.2, 138.6, 136.2, 133.8, 130.1,
128.3, 124.4, 65.6, 47.1; ESI-MS (m/z): 244 (M+Na), 222 (M+H);
enantiomeric excess was determined by HPLC with a Chiralcel
AD-H column (93:7) hexane/isopropanol, 0.7 mL/min, detection
at 254 nm. HPLC data for aldol adducts derived from CuCl2[(ꢀ)-
sparteine](1) catalyst: major enantiomer (S) tr = 17.08 min, minor
enantiomer (R) tr = 18.85 min; 37% ee; ½a D25
¼ ꢀ44:2 (c 1.0, CHCl3).
ꢂ

2166
H. Maheswaran et al. / Tetrahedron: Asymmetry 21 (2010) 2158–2166
ed.; Cornell University Press: New York. Chapter 3. This property enables its
routine use in organic synthesis as deprotecting agents for silyl groups.
10. To the best of our knowledge until now there is no report for the
enantioselective direct aldol reaction of aromatic aldehydes and
methylvinylketone (MVK) under catalytic conditions. For direct aldol
reaction of aliphatic aldehydes with methylvinylketone (MVK); see: Trost, B.
M.; Shin, S.; Sclafani, J. A. J. Am. Chem. Soc. 2005, 127, 8602.
HPLC data for aldol adducts derived from NiCl2[(ꢀ)-sparteine](2)
catalyst: major enantiomer (R) tr = 18.85 min, minor enantiomer
(S) tr = 17.08 min; 65% ee; ½a D25
¼ þ67:0 (c 1.0, CHCl3).
ꢂ
4.2.3.5. 5-hydroxy-5-(4-chloro-3-nitrophenyl)pent-1-en-3-one
7i. IR (KBr): 3448, 3100, 2909, 1949, 1678, 1611, 1535, 1479,
1403, 1353, 1200, 1050, 975, 900, 834, 799, 754, 709, 688, 537;
1H NMR (500 MHz, CDCl3) d: 2.90–3.01 (m, 2H), 3.71 (br s, 1H),
5.23 (br s, 1H), 5.94 (d, J = 10.73 Hz, 1H), 6.26 (d, J = 17.56 Hz,
1H), 6.36 (dd, J = 10.73, 17.56 Hz, 1H), 7.47–7.55 (m, 2H), 7.87 (s,
1H); 13C NMR (75 MHz, CDCl3) d: 199.8, 148.0, 143.6, 136.2,
131.8, 130.5, 130.3, 125.8, 122.9, 68.2, 47.4; ESI-MS (m/z): 278
(M+Na), 256 (M+H); enantiomeric excess was determined by HPLC
with a Chiralcel AD-H column (93:07) hexane/isopropanol, 0.7 mL/
min, UV detection at 220 nm. HPLC data for aldol adducts derived
from CuCl2[(ꢀ)-sparteine](1) catalyst: major enantiomer (R)
tr = 20.4 min, minor enantiomer (S) tr = 22.52 min; 44% ee;
11. For direct aldol reaction of methylvinylketone (MVK) using catalyst 2, better
yields and enantioselectivities were obtained in methanol. In contrast, for
catalyst 1, the best solvent for the reaction is DMF. It should be noted that
solvent independent enantioreversal with identical sign of the specific rotation
occurs in both the solvents (DMF and MeOH) with catalysts 1 and 2.
12. It is a popular misconception that the (+)-sparteine is not a natural product;
however, although more narrowly distributed than its better known and
commercially available antipode (ꢀ)-sparteine. It is well known that (+)-
sparteine is biosynthesized by numerous plant species. Traditionally, (+)-
sparteine has been most conveniently obtained from natural ( )-lupanine (2-
oxosparteine), extracted from the seeds of Lupinus albus, by classical resolution
followed by reduction of the resulting (ꢀ)-lupanine. For references see (a)
Norcross, N. R.; Melbardis, J. P.; Solera, M. F.; Sephton, M. A.; Kilner, C.;
Zakharov, L. N.; Astles, P. C.; Warriner, S. L.; Blakemore, P. R. J. Org. Chem. 2008,
73, 7939; (b) Orekhov, A. P.; Norkina, S. S.; Maksimova, T. Arch. Pharm. 1935,
273, 369; (c) The cost of (+)-sparteine (as pachycarpine) is ca. three times
greater than that typical for (ꢀ)-sparteine.; More recently, (+)-sparteine itself
has been discovered to be a minor alkaloid of Lupinus albus, see: (d) Wysocka,
W. Sci. Legumes 1995, 2, 137–140; (e) Ebner, T.; Eichelbaum, M.; Fischer, P.;
Meese, C. O. Arch. Pharm. 1989, 322, 399.
13. (a) Lopez, S.; Muravyov, I.; Pulley, S. R.; Keller, S. W. Acta Crystallogr., Sect. C
1998, 54, 355; (b) Kang, S. K.; Choi, S.-N.; Lee, Y.-M. Acta Crystallogr., Sect. C
2004, 60, m174.
14. Degree of distortion from the planarity is reflected by the magnitude of
dihedral angle between the MN2 plane and MCl2 plane, which is 67.0°and
81.3° for 1 and 2, respectively.
15. (a) Klebe, G.; Weber, F. Acta Crystallogr., Sect. B 1994, 50, 50; (b) Murray-Rust,
P.; Bürgi, H. B.; Dunitz, J. D. Acta Crystallogr., Sect. B 1978, 34, 1787; (c) Murray-
Rust, P.; Bürgi, H. B.; Dunitz, J. D. Acta Crystallogr., Sect. B 1978, 34, 1793; (d)
Raithby, P. R.; Shields, G. P.; Allen, F. H.; Motherwell, W. D. S. Acta Crystallogr.,
Sect. B 2000, 56, 444.
½
a 2D5
ꢂ
¼ þ29:2 (c 1.0, CHCl3). HPLC data for aldol adducts derived
from NiCl2[(ꢀ)-sparteine](2) catalyst: major enantiomer (S)
tr = 22.57 min, minor enantiomer (R) tr = 20.4 min; 54% ee;
½
a 2D5
ꢂ
¼ ꢀ38:0 (c 1.0, CHCl3).
Acknowledgments
PJAJ thank UGC, New Delhi for the award of fellowship, and KLP
thanks CSIR, New Delhi for the award of fellowship.
References
1. Agranat, I.; Caner, H.; Caldwell, J. J. Nat. Rev. Drug Disc. 2002, 1, 753.
2. For recent reviews, see: (a) Zanoni, G.; Castronovo, F.; Franzini, M.; Vidari, G.;
Giannini, E. Chem. Soc. Rev. 2003, 32, 115; (b) Kim, Y. H. Acc. Chem. Res. 2001, 34,
955; (c) Sibi, M. P.; Liu, M. Curr. Org. Chem. 2001, 5, 719; (d) Bartók, M. Chem.
Rev. 2010, 110, 1663.
16. The internal coordinates, r1, r2, r3, r4, h12, h13, h14, h23, h24, and h34 transform as
2A1 + E +2T2 irreducible representation for perfect Td symmetry. For the
CuCl2[(ꢀ)-sparteine]
1 complex, the crystallographically observed internal
coordinates around the tetrahedron are: r1 = 2.258 Å; r2 = 2.238 Å, r3 = 2.021 Å,
r4 = 2.003 Å, h12 = 106.93°; h13 = 101.73°; h14 = 134.12°; h23 = 126.53°;
3. Some classic examples of enantio-divergent catalytic synthesis: see: in
asymmetric aldol reaction: (a) Evans, D. A.; Murry, J. A.; Kozlowski, M. C. J.
Am. Chem. Soc. 1996, 118, 5814; (b) Evans, D. A.; Kozlowski, M. C.; Murry, J. A.;
Burgey, C. S.; Campos, K. R.; Connell, B. T.; Staples, R. J. J. Am. Chem. Soc. 1999,
121, 669; (c) Evans, D. A.; Burgey, C. S.; Kozlowski, M. C.; Tregay, S. W. J. Am.
Chem. Soc. 1999, 121, 686; (d) Matsunaga, H.; Yamada, U.; Ide, T.; Ishizuka, T.;
Kunieda, T. Tetrahedron: Asymmetry 1999, 10, 3095; (e) Kobayashi, S.; Horibe,
M. J. Am. Chem. Soc. 1994, 116, 9805; (f) Kobayashi, S.; Horibe, M. Tetrahedron
1996, 52, 7277; in asymmetric ene reactions: see: (g) Johannsen, M.; Jørgensen,
K. A. J. Org. Chem. 1995, 60, 5757; (h) Evans, D. A.; Burgey, C. S.; Paras, N. A.;
Vojkovsky, T.; Tregay, S. W. J. Am. Chem. Soc. 1998, 120, 5824; in asymmetric
cycloaddition reaction, see: (i) Corey, E. J.; Imai, N.; Zhang, H.-Y. J. Am. Chem.
Soc. 1991, 113, 728; (j) Evans, D. A.; Murry, J. A.; von Matt, P.; Norcross, R. D.;
Miller, S. J. Angew. Chem., Int. Ed. 1995, 34, 798; (k) Evans, D. A.; Miller, S. J.;
Lectka, T.; von Matt, P. J. Am. Chem. Soc. 1999, 121, 7559; (l) Ghosh, A. K.;
Mathivanan, P.; Cappiello, J. Tetrahedron Lett. 1996, 37, 3815; (m) Corey, E. J.;
Ishihara, K. Tetrahedron Lett. 1992, 33, 6807; for an excellent comprehensive
account on asymmetric Diels–Alder, aldol, ene, and Michael additions see: (n)
Johnson, J. S.; Evans, D. A. Acc. Chem. Res. 2000, 33, 325.
h24 = 100.06°; h34 = 90.52°; for the NiCl2[(ꢀ)-sparteine]
2 complex, the
crystallographically observed internal coordinates around tetrahedron are:
r1 = 2.236 Å; r2 = 2.234 Å, r3 = 2.032 Å, r4 = 2.037 Å, h12 = 119.67°; h13 = 103.10°;
h14 = 110.34°; h23 = 123.85°; h24 = 105.99°; h34 = 89.52°. Note: The internal
coordinates r1, r2, r3 and r4 describe crystallographically observed M–Cl1, M–
Cl2, M–N1 and M–N2 bond lengths while hij represents corresponding
crystallographically observed bond angles. The atoms are labelled (1–4) in
the order of decreasing bond lengths so that the corresponding bond angles
between displacement vectors lie in the same asymmetric unit of the vector
space defined by symmetry coordinates.
17. Forthe CuCl2[(ꢀ)-sparteine]1 complex,thecomputeddisplacement vectorsalong
respective symmetry coordinates are: S1(A1) = 4.26 Å, S3a(T2) = 0.235 Å, S3b(T2) =
ꢀ0.001 Å, S3c(T2) = ꢀ0.019 Å, S2a(E) = ꢀ19.50°, S2b(E) = ꢀ29.43°, S4a(T2) = 11.60°,
S4b(T2) = 1.18°, S4c(T2) = 5.37°, S5(A1) = 269.4°; for the NiCl2[(ꢀ)-sparteine]
2
complex, computed displacement vectors along respective symmetry
coordinates are: S1(A1) = 4.26 Å, S3a(T2) = 0.206 Å, S3b(T2) = ꢀ0.0015 Å, S3c(T2) =
0.0035 Å, S2a(E) = ꢀ7.19°, S2b(E) = ꢀ12.55°, S4a(T2) = 21.30°, S4b(T2) = ꢀ2.04°,
S4c(T2) = ꢀ9.55°, S5(A1) = 266.4°.
18. (a) Figgis, B. N. Introduction to Ligand Fields; John Wiley and Sons: New York,
1966; (b) The balance of electronic and steric factors in 1 and 2 complexes are
directly reflected in the asymmetry of M–N bond lengths in their X-ray
structures. The Cu–N1 facing the endo-ring has a shorter Cu–N bond length
(Cu–N1 = 2.003 (3) Å) compared to 2.021 (3) Å for Cu–N2 facing the exo-ring.
In complex 2, both Ni–N1 and Ni–N2 bond distances correspond to a more
symmetric configuration with 2.032 (4) Å and 2.027 (4) Å, respectively, which
is slightly longer than those observed in complex 1.
4. Maheswaran, H.; Prasanth, K. L.; Krishna, G. G.; Ravikumar, K.; Sridhar, B.;
Lakshmi Kantam, M. Chem. Commun. 2006, 39, 4066.
5. Absolute configuration of the aldol adducts are tentatively assigned (R) or (S)
by comparison with relative signs of the specific rotations data of aldol with
those reported for 5a and 5g as reference compounds, see: Sugasawa, T.;
Toyoda, T. Tetrahedron Lett. 1979, 16, 1423. By this analogy, a (+)-aldol adduct is
tentatively assigned as the (R)-isomer while a (ꢀ)-aldol adduct is assigned as
the (S)-isomer.
19. Cirera, J.; Alemany, P.; Alvarez, S. Chem. Eur. J. 2004, 10, 190.
6. Kiyooka, S.; Takeshita, Y.; Tanaka, Y.; Higaki, T.; Wada, Y. Tetrahedron Lett. 2006,
47, 4453.
20. (a) Jasiewicz, B.; Sikorska, E.; Khmelinskii, I. V.; Warzajtis, B.; Rychlewska, U.;
Boczon, W. l.; Sikorski, M. J. Mol. Struct. 2004, 707, 89; (b) Boschmann, E.;
Weinstock, L. M.; Carmack, M. Inorg. Chem. 1974, 13, 1297; (c) Lee, Y. M.; Kwon,
M.-A.; Kang, S. K.; Jeong, J. H.; Choi, S.-N. Inorg. Chem. Commun. 2003, 6, 197.
21. The rms overlay of X-ray structures (1 and 2) was obtained by calculations
involving published X-ray crystal structure data (see Ref. 13) using Mercury
software (Version 1.4.1). For further details see: Bruno, I.; Cole, J. J. C.;
Edgington, P. R.; Kessler, M. K.; Macrae, C. F.; McCabe, P.; Pearson, J.; Taylor, R.
Acta. Cryst., Sect. B 2002, 58, 389–397.
7. For
a recent review on double catalytic activation (DCA) protocol: see
Kanemasa, S.; Ito, K. Eur. J. Org. Chem. 2004, 4741.
8. Fluoride anions are excellent catalysts for Mukaiyama aldol reaction: see (a)
Nakamura, E.; Shimizu, M.; Kuwajima, I.; Sakata, J.; Yokoyama, K.; Noyori, R. J.
Org. Chem. 1983, 48, 932; (b) Denmark, S. E.; Lee, W. Chem. Asian J. 2008, 3, 327;
(c) Denmark, S. E.; Lee, W. J. Org. Chem. 1994, 59, 707.
9. Fluoride anions show an extremely high affinity for the silicon atom due to
high homolytic bond energy; see: Pauling, L. The Nature of Chemical Bond, 3rd