V. Casarotto et al. / Tetrahedron Letters 48 (2007) 5561–5564
5563
Table 2. Substrate scope of the LA–LB* catalyzed diethylzinc addition
reaction at room temperature
The intramolecular bifunctional activation displayed by
the LA–LB* catalyst was further demonstrated through
additional control experiments. Replacing the LB* in
ligand 1 with a methoxy group (a nonfunctional
‘dummy’ group) would give a control ligand 10
(Fig. 1) that would generate an achiral LA 11 for cata-
lyzing the diethylzinc addition reaction. The desired
methyl ester 10 was synthesized by solvolysis of ligand
1 in methanol in 89% yield. In the absence of an LB*
tethered intramolecularly, the LA 11 (10 mol %) derived
from this tridentate ligand catalyzed the same reaction
but only gave an incomplete reaction in 3 h (entry 3).
Addition of 10 mol % of BnQ 9 as an external LB* to
10 mol % of LA 11 did not improve the reaction yield
(entry 4), compared to the reaction catalyzed by the
LA alone (entry 3).
O
OH
Me
H
1 (10 mol%)
+ Et2Zn
Toluene
rt, 3 h
X
X
12
13
Entry
ArCHO 12 (–X)
Isolated yield (%)
eea (%)
1
2
3
4
5
4-MeO–
H–
4-Me–
4-F–
56
81
96
92
42
63b
63
73b
90
2-EtO–
Quant.
a ee determined by chiral GC–MS.
b ee determined by chiral GC–MS analysis of the corresponding
acetate.
It should be noted that the presence of the LB* in this
nontethered LA + LB* catalyzed reaction did not result
in any asymmetric induction. In the absence of a linker
between LA and LB*, this is not surprising, presumably
due to the unrestricted LA, LB* interactions. Coordina-
tion of the LB* to the LA generates a mixture of diaste-
reomeric chiral Lewis acids (i.e., random formation of
LA*), each of which would be catalytically active, but
might have opposite asymmetric induction that led to
an overall racemic product. These control experiments
further illustrated the importance of the intramolecular
bifunctional activation promoted by the LA–LB* for
rate acceleration and asymmetric induction. The chiral-
ity at the LB* of the LA–LB* catalyst appeared to be the
only determining factor governing the stereochemical
outcome for this addition reaction because switching
the LB* component in 1 from quinine to its pseudoenan-
tiomeric quinidine completely reversed the enantioselec-
tivity for this reaction (Table 1, ent-1, entry 5). This
reversal of enantioselectivity (i.e., entries 1 and 5) offers
another example of LB*-dependent asymmetric bifunc-
tional catalysis (ABC),4c,d considering all the possible
LA configurations present in 3/4.
chirality is too far away from the LA to create an effec-
tive chiral pocket at the metal center for asymmetric
induction. Therefore, coordination of aldehydes to the
LA would not offer any intrinsic facial discrimination
for the bound carbonyl group. On the other hand,
chelation of diethylzinc by the bridgehead nitrogen of
the LB* converts the diethylzinc reagent into a transient
chiral nucleophile that ultimately differentiates the
enantiomeric faces of the bound carbonyl group and
determines the stereochemical outcome of the reaction.
In addition, the LA–LB* bifunctional catalytic activity
has been demonstrated through the diethylzinc addition
reaction to 2-methoxybenzaldehyde and control experi-
ments. The substrate scope of this new bifunctional
catalyst for the diethylzinc addition reaction has been
illustrated using aromatic aldehydes. The LA–LB*
bifunctional catalyst derived from tridentate ligand
complements other LA*–LB and LA*–LB* catalytic
systems and should be very useful for asymmetric
bifunctional catalysis (ABC).
Acknowledgments
Having established that the Zn(II) complex of tridentate
ligand 1 displays LA–LB* bifunctional catalytic activity
at rt, we subsequently investigated the substrate scope of
the LA–LB* catalyzed diethylzinc addition reaction.
Employing 10 mol % of tridentate ligand 1 for aromatic
aldehydes 12, the addition reaction gave the expected
benzyl alcohols 13 in good to excellent yields (entries
1–5) at rt (Table 2). Among all the substrates examined,
bidentate substrate exhibits the best enantioselectivity in
this reaction (entry 5). This suggests a possible bidentate
coordination of 2-ethoxybenzaldehyde to the LA that
restricts the free rotation of the bound carbonyl group.
This result is also consistent with the excellent ee ob-
served for 2-methoxybenzaldehyde 7 under bifunctional
catalysis conditions (Table 1, entry 1), where a bidentate
chelation by the LA is likely.
Acknowledgment is made to the donors of the American
Chemical Society Petroleum Research Fund (PRF
42754-G1) for partial support of this research. We are
grateful to Ms. Tamam Baiz of our department for assis-
tance with the manuscript.
Supplementary data
Experimental procedures, spectral data, and NMR spec-
tra. Supplementary data associated with this article can
References and notes
In summary, we have designed and synthesized a novel
tridentate ligand by exploiting the readily available
cinchona alkaloid quinine/quinidine as the LB* for
constructing new LA–LB* bifunctional catalysts having
a mixture of the LA configurations. The tethered LB*
1. (a) Lewis Acids in Organic Synthesis; Yamamoto, H., Ed.;
Wiley-VCH: New York, 1999; Vols. 1–2, (b) Comprehensive
Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A., Yama-
moto, H., Eds.; Springer: New York, 1999; Vols. I–III.