Asymmetric aldol reaction via a dinuclear zinc catalyst: α-hydroxyketones as donors [1]

Full text of DOI:10.1021/ja003871h

J. Am. Chem. Soc. 2001, 123, 3367-3368
3367
Communications to the Editor
support for the proposed structures.6 At the same time, higher
aggregates can also form. Thus, direct examination of the catalyst
without exposure to acetic acid showed a series of peaks in the
electrospray mass spectrometer between 1543 and 1567 consistent
with formula C86H87N4O7Zn4 that corresponds to the M + H+
peak of the oxo-bridged dimer 3b. This dimer likely derives from
reaction of 2 with trace amounts of adventitious water. Indeed,
deliberate addition of trace amounts of water does not adversely
affect the selectivity.
Asymmetric Aldol Reaction via a Dinuclear Zinc
Catalyst: r-Hydroxyketones as Donors
Barry M. Trost,* Hisanaka Ito, and Elliad R. Silcoff
Department of Chemistry, Stanford UniVersity
Stanford, California 94305-5080
Adding a mixture of 1.5 equiv of hydroxyacetophenone (4)
and 1.0 equiv cyclohexanecarboxaldehyde (5) to 5 mol % of the
catalyst 2 in THF in the presence of 4 Å MS and Ph3PS at room
temperature, a catalyst system optimized for acetophenone,5 gave
a high yield of the desired aldol product with good diastereose-
lectivity favoring the syn adduct 67 (eq 1 and Table 1, entry 1).
ReceiVed NoVember 6, 2000
The ability to control the stereoselectivity of the directed aldol
condensation has raised this process to prominence in the synthesis
of complex molecular targets shared by few reactions.1 However,
these reactions almost invariably require preactivation of the
nucleophilic or donor partner. R-Hydroxyketone donors are
particularly interesting because of the utility of the polyoxygenated
products, yet represent one of the most troublesome donors
because of chemoselectivity issues. Only recently have several
reports addressed the simple aldol addition involving both chemo-
and enantioselectivity using both biological-type (e.g., catalytic
antibodies)2 and nonbiological-type3,4 catalysis and, in some cases,
included R-hydroxyacetone and related derivatives.2,4b In these
cases, significant excesses of the donor must be employed. We
recently reported the development of a new type of asymmetric
catalyst which we postulated involves a dinuclear zinc complex.5
In this paper, we communicate the effectiveness of this catalyst
with R-hydroxyketones that permits use of nearly stoichiometric
amounts of both partners in the asymmetric aldol reaction and
the surprising effect of the donor on facial selectivity with respect
to the aldehyde.
The catalyst is prepared by reacting the phenol 1 with
diethylzinc in THF as in eq 1. Exposure of the complex to acetic
acid in the inlet of an electrospray mass spectrometer shows a
series of peaks between m/e 823-833 consistent with the formula
C45H47N2O5Zn2 that corresponds to the M + H+ peak of 3a. The
combination of these data with our earlier observation regarding
the 2:1 stoichiometry of diethylzinc to ligand provides good
Disappointingly, the ee was only 30%. However, temperature had
a strong effect; as the temperature was lowered to -35°, the ee
progressively increased to 90% (Table 1, entries 2-5). Remark-
ably, the dr was invariant with respect to temperature. Removing
Ph3PS at -35° led to the same result (Table 1, entry 6), that is,
no enhancement in conversion in contrast to the opposite
observation in the case of acetophenone. Further lowering the
temperature to -55° saw a drop in conversion and consequently
in yield, but a small increase in ee (Table 1, entry 7). Remarkably,
lowering the catalyst loading to 2.5 mol % showed a significant
increase in diastereoselectivity with just a modest loss in yield.
The conditions of entries 6 and 8 were adopted for further
reactions as a compromise between rate/yield and ee. The relative
and absolute stereochemistry of the major diastereomer was
established by comparison to the product derived from an
asymmetric dihydroxylation.8 Strikingly, the absolute configura-
(1) For the general review on enantioselective Mukaiyama aldol reactions,
see: Carreira, E. M. In ComprehensiVe Asymmetric Catalysis; Jacobsen, E.
N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Heidelberg, 1999; Vol. 3, p
998; Mahrwald, R. Chem. ReV. 1999, 99, 1095; Gro¨ger, H.; Vogl, E. M.;
Shibasaki, M. Chem. Eur. J. 1998, 4, 1137; Nelson, S. G. Tetrahedron:
Asymmetry 1998, 9, 357; Bach, T. Angew. Chem., Int. Ed. Engl. 1994, 33,
417. Also see: Johnson, J. S.; Evans, D. A. Acc. Chem. Res. 2000, 33, 325.
(2) (a) Machajewski, T. D.; Wong, C.-H. Angew. Chem., Int. Ed. 2000,
39, 1352; Takayama, S.; McGarvey, G. J.; Wong, C. H. Chem. Soc. ReV.
1997, 26, 407; Rasor, J. P. Chim. Oggi. 1995, 13, 9; Barbas, C. F., III; Heine,
A.; Zhong, G.; Hoffmann, T.; Gramatikova, S.; Bjo¨rnestedt, R.; List, B.;
Anderson, J.; Stura, E. A.; Wilson, E. A.; Lerner, R. A. Science 1997, 278,
2085. Also see: Kajimoto, T. Yakugaku Zasshi 2000, 120, 42; Chem. Abstr.
2000, 132, 194551; Hiratake, J.; Oda, J. Yuki Gosei Kagaku Kyokaishi 1997,
55, 452; Chem. Abstr. 1997, 127, 17218. (b) Also see: Hoffmann, T.; Zhong,
G.; List, B.; Shabat, D.; Anderson, J.; Gramatikova, S.; Lerner, R. A.; Barbas,
C. F., III. J. Am. Chem. Soc. 1998, 120, 2768.
(6) For other dinuclear zinc complexes, see: Sakiyama, H.; Mochizuki,
R.; Sugawara, A.; Sakamoto, M.; Nishida, Y.; Yamasaki, M. J. Chem. Soc.,
Dalton Trans. 1999, 997. For other zinc complexes derived from 2,6-di-
(dialkylaminomethyl)-p-cresol, see: Uhlenbrock, S.; Wegner, R.; Krebs, B.
J. Chem. Soc., Dalton Trans. 1996, 3731. For bis-ligated dinuclear zinc
complexes, see: Fahrni, C. J.; Pfaltz, A.; Neuburger, M.; Zehnder, M. HelV.
Chim. Acta 1998, 81, 507.
(3) Yamada, Y. M. A.; Yoshikawa, N.; Sasai, H.; Shibasaki, M. Angew.
Chem., Int. Ed. Engl. 1997, 36, 1871; Yamada, Y. M. A.; Shibasaki, M.
Tetrahedron Lett. 1998, 39, 5561; Yoshikawa, N.; Yamada, Y. M. A.; Das,
J.; Sasai, H.; Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 4168; For a review,
see: Shibasaki, M., Sasai, H. Top. Stereochem. 1999, 22, 201.
(7) This compound has been satisfactorily characterized spectroscopically
and elemental composition established.
(8) Jacobsen, E. N.; Marko, I.; Mungall, W. S.; Schro¨der, G.; Sharpless,
K. B. J. Am. Chem. Soc. 1995, 34, 1059. For a review, see: Kolb, H. C.; Van
Nieuwenhze, M. S.; Sharpless, K. B. Chem. ReV. 1994, 94, 2483. The syn
and anti diols for entries 2 and 5 are also known, see: Miyoshi, N.; Fukuma,
T.; Wada, M. Chem. Lett. 1995, 999 and referencess therein. Clerici, A. Porta,
O. J. Org. Chem. 1989, 54, 3872; Mukaiyama, T.; Yamaguchi, M. Chem.
Lett. 1982, 509.
(4) (a) List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem. Soc. 2000,
122, 2395. (b) Notz, W.; List, B. J. Am. Chem. Soc. 2000, 122, 7386. (c) For
an intramolecular variant differentiating prochiral carbonyl groups, see: Hajos,
Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39, 1615; Agami, C.; Platzer, N.;
Sevestre, H. Bull. Soc. Chim. Fr. 1987, 358.
(5) Trost, B. M.; Ito, H. J. Am. Chem. Soc. 2000, 122, 12003.
10.1021/ja003871h CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/15/2001

3368 J. Am. Chem. Soc., Vol. 123, No. 14, 2001
Communications to the Editor
Table 1. Optimization of Aldol Reaction of
Frequently, yields and selectivities plummet with substrates such
as those in entries 6-8.3,4a,b As noted previously, lowering the
temperature for dihydrocinnamaldehyde from -35° (entry 5) to
-55° (entry 6) saw an increase in ee from 90% to 96% albeit at
the expense of conversion.
Hydroxyacetophenonea
mol % 15 mol % temp time isolated yield
ee
entry catalyst
Ph3PS
°C
(h)
(%)
dr (%)
1
2
3
4
5
6
7
8
5.0
5.0
5.0
5.0
5.0
5.0
5.0
2.50
yes
yes
yes
yes
yes
no
r.t.
5
-5
-25 48
-35 24
-35 24
-55 48
-40 24
48
15
48
>90
82
>90
>90
94
97
77
5:1
30
45
76
88
90
90
93
Switching hydroxyacetophenone to 2-hydroxyacetylfuran gave
excellent results (entries 9 and 10). In these cases, the ratio of
substrates was decreased to 1.3:1 with no deleterious effect on
conversion or chemoselectivity. While the diastereoselectivity was
similar to that of hydroxyacetophenone, a significant increase in
ee was observed (entries 1 and 5 vs 9 and 10). The success of a
ratio of ketone to aldehyde of only 1.3:1.0 led to our exploration
of a further reduction. As shown in entry 2b versus 2c, dropping
this ratio to 1.1:1.0 decreased the conversion, resulting in a lower
isolated yield of aldol adduct. Nevertheless, both the diastereo-
selectivity and enantioselectivity increased. A similar trend was
observed with 3-methylbutanal (entry 4). This new set of
conditions gave the highest ee observed, 98%, in entry 9b. In all
cases examined (entries 1-5), dropping the catalyst loading from
5 to 2.5 mol % increased diastereoselectivity and, in some cases
(entries 2-4), ee significantly.
5:1
5:1
5:1
5:1
5:1
5:1
no
no
83
30:1 92
a All reactions were run on 0.5 mmol scale at 0.3 M in aldehyde in
the presence of 100 mg of 4 Å MS. b Enantiomeric excess determined
by chiral HPLC using chiralcel OD column.
Table 2. Asymmetric Aldol Reactiona
The catalyst system reported herein comes closest in reaching
the ideal atom economical version of the asymmetric aldol
condensation, that is, the ideal being stoichiometric amounts of
both reactants and anything else needed only catalytically.
Excellent conversion and chemoselectivity was observed with
ratios as low as 1.1:1.0 even with simple aldehydes such as
dihydrocinnamaldehyde. As pointed out by others,4b an advantage
of such an asymmetric aldol condensation over the asymmetric
dihydroxylation is the formation of both stereocenters simulta-
neous with carbon-carbon bond formation. Further, chemose-
lectivity issues may arise in the AD reaction with substrates such
as that of Table 2, entry 8. On the basis of our proposed model,
it is reasonable to assume that the enolate of the R-hydroxyketone
serves as a bidentate ligand bridging the two zincs as depicted in
7 based upon the manner in which carboxylates bridge some
a All reactions as in eq 3 using a 1.5:1.0 ratio of hydroxyketone to
aldehyde using 2.5 mol % catalyst unless noted otherwise. b See ref 7.
c Determined by 1H NMR spectroscopy on the crude mixture. d Deter-
mined by chiral HPLC on a Chiracel OD or OJ column. e For this run,
5.0 mol % catalyst was employed. f Reactions performed using a 1.1:
1.0 ratio of hydroxyketone to aldehyde. g Reaction performed at -55
°C. h Reactions performed using 1.3:1.0 ratio of hydroxyketone to
aldehyde.
related dinuclear zinc complexes.6 Coordination of the aldehyde
as shown then delivers the major observed product. The picture
that emerges nicely accommodates the unexpected change in facial
selectivity with respect to the aldehyde. The enhanced selectivity
as a result of lowering the catalyst loading may speak to a role
of dimeric type species such as 3b as well as the monomer with
the latter giving higher selectivities.
tion of the stereocenter derived from the aldehyde is opposite to
that obtained using acetophenone as donor.5
Table 2 and eq 2 summarize the results for a number of
Acknowledgment. We thank the National Science Foundation and
the National Institutes of Health, General Medical Sciences for their
generous support of our programs. H.I. thanks the Uehara Memorial
Foundation for a partial support of his postdoctoral studies. Mass spectra
were provided by the Mass Spectrometry Regional Center of the
University California-San Francisco supported by the NIH Division of
Research Resources.
Supporting Information Available: Experimental details (PDF). This
material is available free of charge via the Internet at http://pubs.acs.org.
examples. Increasing the size of the R-substituents increases the
dr (entry 3). Moving the branch point (entries 4 and 5) or
removing it (entries 7 and 8) had no effect on the reaction.
JA003871H