Synthesis, resolution, and aldol reactions of a planar-chiral Lewis acid complex

Article abstract of DOI:10.1021/ja0552702

The synthesis and resolution of a new planar-chiral Lewis acid complex is described. The compound is applied to asymmetric Mukaiyama aldol reactions, leading to the formation of the desired product with good stereoselectivity. The sense of stereoselection is as predicted by the design, which exploits the ability of the Lewis acid to simultaneously serve as a σ and a π acceptor. Mechanistic observations are consistent with rate-determining formation of an aldehyde-Lewis acid complex, followed by rapid nucleophilic addition. Copyright

Full text of DOI:10.1021/ja0552702

Published on Web 10/14/2005
Synthesis, Resolution, and Aldol Reactions of a Planar-Chiral Lewis Acid Complex
Shih-Yuan Liu, Ivory D. Hills,1 and Gregory C. Fu*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
Received August 3, 2005; E-mail: gcf@mit.edu
Chiral Lewis acids serve as useful catalysts for a wide variety
of powerful transformations, particularly addition reactions to
aldehydes.2,3 Several years ago, we outlined a new approach to the
design of a chiral Lewis acid, the distinguishing feature of which
was a π interaction between the Lewis acid and the carbonyl group
(Figure 1). Since maximizing this overlap should enhance the
electrophilicity of the aldehyde, this interaction could define the
most reactive conformation of the Lewis acid-aldehyde complex.4
Because this design does not require a second functional group on
Figure 1. (top) The challenge: Controlling the conformational preference
of a Lewis acid-aldehyde complex. (bottom) A possible solution: A Lewis
acid that can accept electron density from a σ and a π orbital of the aldehyde.
the substrate in order to achieve conformational control of the
complex,5-7 we anticipated that this approach might furnish an
unusually versatile catalyst.
We speculated that a planar-chiral boron heterocycle (e.g., see
A in eq 1) might serve as a suitable framework for this π-activation
strategy, with the aromatic system of the boracycle functioning as
the π acceptor. Consideration of steric interactions leads to the
prediction that the depicted conformation should be favored for
Figure 2. Synthesis of an enantiopure planar-chiral Lewis acid.
nucleophilic addition reactions, which would therefore generate the
illustrated stereoisomer (eq 1; Nu adds opposite to the face blocked
by MLn).8,9
We next turned our attention to the application of azaborolyl
complex 1 in a stereoselective process, specifically the Mukaiyama
In an earlier study, we described crystallographic and spectro-
scopic evidence for the suggested π overlap in the case of an elec-
tron-rich aldehyde and an (η6-borabenzene)Cr(CO)3 moiety.4a Al-
though this initial investigation provided proof-of-principle for a
critical aspect of our design, we were not able to achieve addition
reactions with this system. We therefore turned our attention to
other frameworks, including (η5-1,2-azaborolyl)iron derivatives. In
this Communication, we report the generation of an enantiopure
(η5-1,2-azaborolyl)iron adduct,10 and we establish that it promotes
highly stereoselective aldol reactions in which the sense of asym-
metric induction is that predicted by our design (eq 2 versus eq 1).
aldol reaction.13 We were pleased to establish that, in the presence
of 1, silyl ketene acetals react with aldehydes to generate the desired
aldol product with excellent stereoselectivity (Table 1, entries
1-3).14 Somewhat lower selectivity is observed when a less
sterically demanding nucleophile derived from a thioester is
employed (entry 4).15 To date, attempts to achieve turnover have
not been successful, presumably due to the stability of the B-O
bond of the aldol product.
We have established that the sense of stereoselection is as
predicted by the general model illustrated (for a borabenzene
complex) in eq 1. Thus, for the boron aldolate product of entry 1
of Table 1, hydrolysis of the B-O bond leads to the formation of
a â-hydroxyester of known absolute configuration (eq 3).16 For the
The synthesis of our target complex (1) began with previously
described (η5-1,2-azaborolyl)iron adduct 2 (Figure 2).10c Selective
reduction of the Fe-I bond with sodium amalgam, followed by
quenching with TMSCl, furnishes chloroborane 3. Hydrolysis of
the B-Cl bond generates 4, the enantiomers of which can be
separated by preparative chiral HPLC.11 The hydroxy group is then
converted into a better leaving group by sequential reaction with
oxalyl chloride and AgOTs,12 thereby producing the desired
enantiopure Lewis acid, (+)-1.
aldol adducts of the reactions depicted in entries 3 and 4, X-ray
crystallographic analysis confirms that the relative stereochemistry
is as anticipated (Figure 3).
Our current working hypothesis is that these aldol reactions
proceed through the pathway illustrated in Figure 4. For the addition
depicted in entry 4 of Table 1, we have determined that the rate
9
15352
J. AM. CHEM. SOC. 2005, 127, 15352-15353
10.1021/ja0552702 CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Stereoselective Aldol Reactions in the Presence of
Planar-Chiral Lewis Acid 1
in the ground state. Finally, the addition of [Bu4N]OTs leads to a
slower reaction.17 All of these observations are accommodated by
the proposed mechanism (Figure 4).
In summary, we have presented evidence that a new chiral Lewis
acid design, based on a π interaction with the substrate that
simultaneously provides activation and organization, can furnish
high stereoselectivity in addition reactions to aldehydes. Current
efforts are directed at second-generation designs that will lead to
catalyst turnover.
Acknowledgment. Support has been provided by Merck
Research Laboratories and Novartis. Funding for the MIT Depart-
ment of Chemistry Instrumentation Facility has been furnished in
part by the National Science Foundation (CHE-9808061 and DBI-
9729592).
Supporting Information Available: Experimental procedures and
compound characterization data (PDF, CIF). This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Correspondence concerning X-ray crystallography should be directed to
I. D. Hills.
(2) (a) For an early review of chiral Lewis acid catalysts, see: Ishihara, K.;
Yamamoto, H. In AdVances in Catalytic Processes; Doyle, M. P., Ed.;
JAI Press: Greenwich, CT, 1995; pp 29-59. (b) For other examples,
see: ComprehensiVe Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A.,
Yamamoto, H., Eds.; Springer: New York, 1999.
(3) For reviews of carbonyl-Lewis acid complexes, see: (a) Ooi, T.; Maruoka,
K. In Modern Carbonyl Chemistry; Otera, J., Ed.; Wiley-VCH: New York,
2000; Chapter 1. (b) Saito, S.; Yamamoto, H. In Modern Carbonyl
Chemistry; Otera, J., Ed.; Wiley-VCH: New York, 2000; Chapter 2. (c)
Lewis Acids in Organic Synthesis; Yamamoto, H., Ed.; Wiley-VCH: New
York, 2000.
a Isolated yield of the major diastereomer. b Determined by 1H NMR.
The value in parentheses is the ee determined after hydrolysis of the B-O
bond. c Racemic 1 was used. d Acetonitrile was employed as the solvent.
TIPS ) triisopropylsilyl.
(4) (a) Amendola, M. C.; Stockman, K. E.; Hoic, D. A.; Davis, W. M.; Fu,
G. C. Angew. Chem., Int. Ed. Engl. 1997, 36, 267-269. (b) Fu, G. C. J.
Org. Chem. 2004, 69, 3245-3249.
(5) For examples of chiral Lewis acids that exploit a second functional group
to provide organization, see: (a) Johnson, J. S.; Evans, D. A. Acc. Chem.
Res. 2000, 33, 325-335. (b) Jørgensen, K. A.; Johannsen, M.; Yao, S.;
Audrain, H.; Thorhauge, J. Acc. Chem. Res. 1999, 32, 605-613.
(6) For examples of chiral Lewis acids that employ two-point binding but do
not require an “extra” functional group, see: (a) Hawkins, J. M.; Loren,
S.; Nambu, M. J. Am. Chem. Soc. 1994, 116, 1657-1660. (b) Corey, E.
J.; Lee, T. W. Chem. Commun. 2001, 1321-1329. See also: Gladysz, J.
A.; Boone, B. J. Angew. Chem., Int. Ed. Engl. 1997, 36, 551-583.
(7) For reviews of multidentate Lewis acids, see: (a) Vaugeois, J.; Simard,
M.; Wuest, J. D. Coord. Chem. ReV. 1995, 145, 55-73. (b) Maruoka, K.
Catal. Today 2001, 66, 33-45. (c) Piers, W. E.; Irvine, G. J.; Williams,
V. C. Eur. J. Inorg. Chem. 2000, 2131-2142.
(8) For details of our analysis of the reactive conformation of these Lewis
acid-aldehyde adducts, see ref 4b.
Figure 3. ORTEP illustrations, with thermal ellipsoids drawn at the 35%
probability level, of two products from Table 1: (a) entry 3 and (b) entry 4.
(9) For NMR and reactivity studies of an isoelectronic system that furnish
support for the analysis outlined in eq 1, see: Bloem, P.; David, D. M.;
Kane-Maguire, L. A. P.; Pyne, S. G.; Skelton, B. W.; White, A. H. J.
Organomet. Chem. 1991, 407, C19-C22.
(10) For leading references to (η5-1,2-azaborolyl)iron chemistry, see: (a)
Schmid, G. In ComprehensiVe Heterocyclic Chemistry II; Shinkai, I., Ed.;
Elsevier: Oxford, 1996; Vol. 3, Chapter 3.17. (b) Schmid, G. Comments
Inorg. Chem. 1985, 4, 17-32. (c) Liu, S.-Y.; Lo, M. M.-C.; Fu, G. C.
Angew. Chem., Int. Ed. 2002, 41, 174-176.
(11) Notes: (a) We were not able to resolve (()-3 by preparative chiral HPLC.
(b) The absolute configuration of (-)-4 was determined by X-ray
crystallography (for details, see the Supporting Information).
(12) For a related example of halide abstraction by a silver salt, see ref 10c.
(13) For a review, see: Carreira, E. M. In ComprehensiVe Asymmetric Catalysis;
Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: New York,
1999; Chapter 29.1.
(14) If the aldehyde is not electron-rich, the aldol reaction is either slow or
does not proceed at all, presumably due to difficulty in displacing the
tosylate group from 1 (for our proposed mechanism for these aldol
reactions, see Figure 4).
(15) This reaction also proceeds in CH2Cl2, but with somewhat lower
stereoselectivity.
Figure 4. Possible mechanism for aldol reactions in the presence of Lewis
acid 1.
(16) Ishitani, H.; Yamashita, Y.; Shimizu, H.; Kobayashi, S. J. Am. Chem.
Soc. 2000, 122, 5403-5404.
(17) Other observations: (a) When we monitor an aldol reaction by 1H NMR,
we do not detect the build-up of any intermediates. (b) In the absence of
any one of the three reaction components, there is no detectable reaction/
interaction between the other two species.
law is first-order in the aldehyde, first-order in 1, and zero-order
in the nucleophile. Furthermore, the rate is higher in solvents with
higher dielectric constants (CH3CN > CH2Cl2 > THF, benzene),
consistent with greater charge separation in the transition state than
JA0552702
9
J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005 15353