The asymmetric catalytic aldol reaction of allenolates with aldehydes using N-fluoroacyl oxazaborolidine as the catalyst

Article abstract of DOI:10.1021/ol000377+

matrix presented The asymmetric catalytic aldol reaction of silyl allenolates with aldehydes has been achieved by using N-C₃F₇CO oxazaborolidine as the catalyst. The fluoroacyl group of the catalyst was found to be crucial for control of enantioselectivity. The reaction provides the first enantioselective approach to β-halo Baylis-Hillman-type adducts.

Full text of DOI:10.1021/ol000377+

ORGANIC
LETTERS
2001
Vol. 3, No. 6
823-826
The Asymmetric Catalytic Aldol
Reaction of Allenolates with Aldehydes
Using N-Fluoroacyl Oxazaborolidine as
the Catalyst
Guigen Li,* Han-Xun Wei, Brian S. Phelps, David W. Purkiss, and Sun Hee Kim
Department of Chemistry and Biochemistry, Texas Tech UniVersity,
Lubbock, Texas 79409-1061
qeggl@ttu.edu
Received December 7, 2000
ABSTRACT
The asymmetric catalytic aldol reaction of silyl allenolates with aldehydes has been achieved by using N-C F7CO oxazaborolidine as the
3
catalyst. The fluoroacyl group of the catalyst was found to be crucial for control of enantioselectivity. The reaction provides the first
enantioselective approach to â-halo Baylis−Hillman-type adducts.
The asymmetric catalytic aldol reaction is among the most
important C-C bond-forming reactions in organic chemistry
and has been an active research topic for several decades.1-3
Unfortunately, an asymmetric catalytic aldol reaction using
allenolates as nucleophiles to form C(sp3)-C(sp2) bonds has
not been established thus far. Such a reaction could conceiv-
ably act as a powerful tool for the asymmetric synthesis of
multifunctionalized organic compounds, particularly R-1-
hydroxyalkyl-â-iodo enones which belong to Baylis-Hill-
man-type adducts and can be used as versatile building blocks
for a variety of chemical and biological purposes.4,5 Baylis-
Hillman-type adducts were found to directly inhibit the
growth of tumor cells.6 In fact, the Baylis-Hillman reaction
has been used for the synthesis of these important compounds
and has attracted considerable interest in the synthetic
community in the past few years.4-8
Several asymmetric C(sp3)-C(sp2) formation processes
have been carried out under asymmetric Baylis-Hillman
conditions using chiral auxiliary-anchored substrates5a or by
replacing DABCO catalysts with special chiral amines,5b,c
BINAP,6a and chalcogenide-TiCl4 complexes.6b Very re-
cently, the Pd complex-catalyzed deracemization was de-
(1) (a) Evans, D. A.; Nelson, J. V.; Taber, T. R. Top. Stereochem. 1982,
13, 1. (b) Heathcock, C. H. In ComprehensiVe Carbonion Chemistry; Buncel,
E., Durst, T., Eds.; Elsevier, New York, 1984; Vol. 5B, p 177. (c) Kim, B.
M.; Williams, S. F.; Masumune, S. In ComprehensiVe Organic Synthesis;
Trost, B. M., Fleming, I., Heathcock, C. H., Eds.; Pergamon: Oxford, 1991;
Vol. 2, Chapter 1.7, p 239.
(4) For recent reviews regarding the Baylis-Hillman reaction, see: (a)
Ciganek, E. Org. React. 1997, 51, 201. (b) Basavaiah, D.; Rao, P. D.; Hyma,
R. S. Tetrahedron 1996, 52, 8001. (c) Langer, P. Angew. Chem., Int. Ed.
2000, 39, 3049.
(2) For several recent reviews, see: (a) Nelson, S. G. Tetrahedron:
Asymmetry 1998, 9, 357. (b) Mahrwald, R. Chem. ReV. 1999, 99, 1095. (c)
Arya, P.; Qin, H. Tetrahedron 2000, 56, 917.
(5) (a) Brzezinski, L. J.; Rafel, S.; Leahy, J. W. J. Am. Chem. Soc. 1997,
119, 4317. (b) Iwabuchi, Y.; Nakatani, M.; Yokoyama, N.; Hatakeyama,
S. J. Am. Chem. Soc. 1999, 121, 10219. (c) Barrett, A. G. C.; Cook, A. S.;
Kamimura, A. Chem. Commun. 1998, 2533. (d) Kawamura, M.; Kobayashi,
S. Tetrahedron Lett. 1999, 40, 1539. (e) Li, G.; Wei, H.-X.; Caputo, T. D.
Tetrahedron Lett. 2000, 41, 1.
(6) (a) Li, G.; Wei, H.-X.; Whittlesey, B.; Batrice, N. N. J. Org. Chem.
1999, 64, 1061. (b) Wei, H.-X.; Hook, J. D.; Fitzgerald, K. A.; Li, G.
Tetrahedron: Asymmetry 1999, 10, 661. (c) Li, G.; Wei, H.-X.; Hook, J.
D. Tetrahedron Lett. 1999, 40, 4611. (d) Li, G.; Narayanan, V. L. NIH
bioassay NSC number: 714074-Z.
(3) For representative references on asymmetric catalytic Aldol reactions,
see: (a) Parmee, E. R.; Tempkin, O.; Masumune, S.; Abiko, A. J. Am.
Chem. Soc. 1991, 113, 9365. (b) Corey, E. J.; Cywin, C. L.; Roper, T. D.
Tetrahedron Lett. 1992, 33, 6907. (c) Evans, D. A.; Kozlowski, M. C.;
Murray, J. A.; Burgey, C. S.; Campos, B. T.; Staples, R. J. J. Am. Chem.
Soc. 1999, 121, 669. (d) Denmark, S. E.; Stavenger, R. A.; Wong, K.-T.;
Su, X. J. Am. Chem. Soc. 1999, 121, 4982. (e) Taylor, S.; Duffey, M. O.;
Morken, J. P. J. Am. Chem. Soc. 2000, 122, 4528.
10.1021/ol000377+ CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/24/2001

veloped by Trost and co-workers for asymmetric synthesis
of Baylis-Hillman-type adducts.8 Unfortunately, only a lim-
ited number of Baylis-Hillman-type adducts can be prepared
by using these known methods due to serious scope limita-
tions. In this Letter, we present a novel asymmetric carbon-
carbon bond formation between silyl allenolates and alde-
hydes using N-heptafluorobutyryl oxazaborolidines as catalysts,
as represented in Scheme 1. The reaction provides a straight-
Table 1. Effects of N-Protecting Groups on Enantioselectivity
in CH2Cl2 at -78 °C
etc.2,14 which have been widely used in asymmetric aldol
reactions, were tested to activate aldehydes. Unfortunately,
all of these Lewis acids failed to give enantiomeric excesses
even when they were used in a stoichiometric amount. We
then studied another aldol catalyst series, oxazaborolidines,
under similar conditions.3a-b,15 Still no enantioselectivity was
observed when Ti and Al metal allenolates were employed
as the nucleophiles to react with benzaldehyde. Our attention
was thereafter turned to the use of Kishi’s â-iodo alleno-
lates.9a-c The effort to render this asymmetric process became
extremely challenging because its racemic version can
proceed smoothly at 0 °C in CH2Cl2 in the absence of any
catalysts. We were also challenged by the fact that a low-
temperature system should be found for in situ generation
of both allenolates and catalysts, as well as for the subsequent
carbonyl addition. Our next attempt was a thorough search
to find and modify suitable catalysts, carefully choose certain
allenolates, catalysts, and solvents or cosolvents, and concur-
rently decrease the reaction rate through slow addition of
reactants with a syringe pump.
While other Lewis acid catalysts failed to give any
enantioselectivity for the carbonyl reaction of Kishi’s â-iodo
allenolates with aldehydes, N-sulfonyl oxazaborolidine cata-
lysts15,17 gave promising results. The first encouraging experi-
ments (entry 1 of Table 3) were performed in dichloro-
methane at -78 °C using two N-sulfonyl oxazaborolidines
which were derived from (S)-tryptophane and (S)-phenyl-
alanine. The enantioselectivity of the E isomer controlled
by these two catalysts was determined to be 11 and 40 ee
%, respectively. These results led us to favor phenylalanine-
derived oxazaborolidine as the candidate for further refine-
ments of the catalyst and catalytic system. The N-tosyl group
of phenylalanine-derived oxazaborolidine was then replaced
with several other electron-deficient counterparts such as
N-trifluoromethanesulfonyl16a and 4-nosyl and 2-nosyl
groups.16b,c In fact, N-subunits of oxazaborolidines have
already been proven by Corey and co-workers to be
important in controlling the orientation of boron-aldehyde
coordination.17 To our surprise, the ee % was diminished
Scheme 1
forward approach to â-halo Baylis-Hillman-type adducts.9
These products can be readily converted into â-alkyl Baylis-
Hillman-type adducts and other useful building blocks.10
As mentioned earlier, the aldol reaction acts as the key
step of the original Baylis-Hillman reaction, which indeed
inspired us to develop alternative approaches to Baylis-
Hillman-type adducts. The use of silyl allenolates, metal
allenolates or allenoates, and vinyl anionic species pioneered
by Kishi,9 Marino,11 and Tsuda12 has previously been proven
to be successful for the synthesis of a variety of â-substituted
Baylis-Hillman-type adducts. The two-step Michael-type
addition followed by elimination has also been utilized for
the synthesis of these adducts.13 In the past few years, we
have been involved in exploring reactions of vinyl anions
and/or metal allenolates with aldehydes7,10c-d and believe that
their asymmetric catalytic versions could be achieved by
using chiral Lewis acids to activate the aldehydes or by using
Lewis bases to push anionic species onto aldehydes.
At first, enantiomerically pure titanium(VI) Lewis acids
such as Ti(OiPr)2[BINOL], TiCl2[BINOL], TiO[BINOL],
(7) (a) Hayase, T.; Shibata, T.; Soai, K.; Wakatsuki, Y. Chem. Commun.
1998, 1271. (b) Kataoka, T.; Iwama, T.; Tsujiyama, S.; Watanable, S. Chem.
Lett. 1999, 3, 257.
(8) Trost, B. M.; Tsui, H.-C.; Toste, F. D. J. Am. Chem. Soc. 2000, 122,
3534.
(9) (a) Cheon, S. H.; Christ, W. J.; Hawkins, L. D.; Jin, H. L.; Kishi,
Y.; Taniguchi, M. Tetrahedron Lett. 1986, 39, 4759. (b) Taniguchi, M.;
Kobayashi, S.; Nakagawa, M.; Hino, T.; Kishi, Y. Tetrahedron Lett. 1986,
39, 4763. (c) Taniguchi, M.; Hino, T.; Kishi, Y. Tetrahedron Lett. 1986,
39, 4767. (d) Wei, H. X.; Caputo, T. D.; Purkiss, D. W.; Li, G. Tetrahedron
2000, 56, 2397. (e) Kataoka, T.; Kinoshita, H.; Kinoshita, S.; Iwamura, T.;
Watanable, S. Angew. Chem., Int. Ed. 2000, 39, 2358.
(10) (a) Ayed, T. B.; Amri, H.; Villieras, J. Tetrahedron 2000, 56, 805.
(b) Ramachandran, P. V.; Reddy, M. V. R.; Rudd, M. T. Chem. Commun.
1999, 1979. (c) Li, G.; Wei, H.-X.; Hook, J. D. Tetrahedron Lett. 1999,
40, 4611-4614. (d) Li, G.; Wei, H.-X.; Willis, S. Tetrahedron Lett. 1998,
39, 4607.
Table 2. Solvent Effects on ee % Using N-Trifluoroacetyl
Oxazaborolidine as the Catalyst
(11) (a) Marino, J. P.; Linderman, R. J. J. Org. Chem. 1981, 52, 1624.
(b) Marino, J. P.; Linderman, R. J. J. Org. Chem. 1983, 48, 4621.
(12) (a) Tsuda, T.; Yoshida, T.; Saegusa, T. J. Org. Chem. 1988, 53,
1037. (b) Tsuda, T.; Yoshida, T.; Kawamoto, T.; Saegusa, T. J. Org. Chem.
1987, 52, 1624.
(13) (a) Nagaoka, Y.; Tomioka, K. J. Org. Chem. 1998, 63, 6468. (b)
Barrett, A. G. M.; Kamimura, A. Chem. Commun. 1995, 1755.
824
Org. Lett., Vol. 3, No. 6, 2001

Table 3. Results of the New Asymmetric Catalytic C-C Bond Formationa
a Reactions were run at -78 °C by adding aldehyde into the preformed mixture of catalyst and silyl allenolate in propionitrile over a period of 10 h via
a syringe pump. After addition was finished, the reaction was stirred for additional 7 h before being quenched with dilute HCl. The catalyst was generated
by adding BH3‚THF into the propionitrile solution of N-heptafluoropropyl phenylalanine at rt.3a,b TMS-allenolates were obtained by adding TMS-I to acetylenic
1
ketones also in propionitrile at -78 °C stirring for 1 h. b Estimated by crude H NMR determination. c Both Z and E isomers were determined to have S
chirality which is shown in ref 19. Enantiomeric excesses were determined by Chiralcel OD-H, AD, and OG HPLC columns (see Supporting Information).
d The combined yields of Z and E isomers after column chromatography; 20 mol % of catalyst was used for all cases. Similar ee % and yields were obtained
with 50 mol % of catalyst for case 1.
with the two nosyl replacements but retained with the
N-trifluoromethanesulfonyl modification. We then evaluated
the N-carbonyl moieties instead of N-sulfonyl groups for
possible enantioselectivity enhancement. It was found that
the N-acetyl oxazaborolidine did not give any aldol products.
However, N-trifluoroacetyl oxazaborolidine catalyst did
increase enantioselectivity to 60 and 27 ee % for E and Z
isomers, respectively (Table 1).
The reason to use dichloromethane as the first solvent is
that it has been proven to be effective for in situ preparation
of â-iodo TMS-allenolates and the subsequent carbonyl
additions. However, for the oxazaborolidine-catalyzed aldol
reaction, propionitrile and nitroethane are normally chosen,3a,b,18
even though they have not been utilized for in situ prepara-
tion of â-iodo TMS-allenolates as yet. To further improve
the reaction, parallel optimization experiments were then set
up with three solvents, CH2Cl2, EtCN, and EtNO2. As shown
in Table 2, the use of propionitrile resulted in improved
(14) Mikami, K.; Terada, M. In Lewis Acid Reagents; Yamamoto, H.,
Ed.; Oxford University Press: Oxford, UK, 1999; Chapter 6, p 93, and
references therein.
(15) (a) Takasu, M.; Yamamoto, H. Synlett 1990, 194. (b) Sartor, D.;
Saffrich, J.; Helmchen, G. Synlett 1990, 197.
(16) (a) Boger, D. L.; Yohannes, D. J. Org. Chem. 1989, 54, 2498. (b)
Fukuyama, T.; Jow, C.-K.; Cheung, M. Tetrahedron Lett. 1995, 36, 6373.
(c) Nelson, S. G.; Spencer, K. L. Angew. Chem., Int. Ed. 2000, 39, 1323.
(d) Wipf, P.; Henninger, T. C. J. Org. Chem. 1997, 62, 1586. (e) Miller, S.
C.; Scanlan, T. S. J. Am. Chem. Soc. 1997, 119, 2301.
(17) (a) Corey, E. J.; Imwinkelried, R.; Pikul, S.; Xiang, Y. B. J. Am.
Chem. Soc. 1989, 111, 5493. (b) Corey, E. J.; Loh, T.-P.; Roper, T. D.;
Azimioara, M. D.; Noe, M. C. J. Am. Chem. Soc. 1992, 114, 8290.
Org. Lett., Vol. 3, No. 6, 2001
825

enantioselectivity for both E and Z isomers (75 ee % for E
and 63 ee % for Z, respectively). We then reconsidered the
importance of the N-substituent of the catalyst and replaced
the N-trifluoroacetyl group with its N-heptafluorobutyryl
counterpart. We found that this replacement can give greater
enantioselectivity for both Z and E isomers (85 ee % for E
and 70 ee % for Z, respectively). Obviously, it is crucial
that all three in situ preparations for catalysts, â-iodo TMS-
allenolates, as well as the subsequent carbonyl additions can
be performed in propionitrile at -78 °C.
Scheme 2
With these conditions in hand, a series of substrates were
thus extensively investigated. The preliminary results listed
in Table 3 indicate that a broad scope of substrates can be
used for this new reaction. Both aromatic and aliphatic
aldehydes can serve as the electrophilic acceptors. Further-
more, both aromatic and aliphatic R,â-acetylenic ketones can
be employed as Michael-type acceptors to generate TMS-
allenolates prior to carbonyl additions. Modest to excellent
Z/E selectivity (>20:1-1.8:1) has been obtained for all cased
examined. The resulting Z/E isomers of each case can be
readily separated by routine column chromatography. The
Z/E selectivity was measured by 1H NMR analysis of crude
products. The geometry was determined by NOE experiments
In conclusion, the first asymmetric carbonyl addition of
allenolates to aldehydes has been established using N-
heptafluoropropyl oxazaborolidine as the catalyst. The reac-
tion provides an unprecedented enantioselective approach to
individual Z and E isomers of â-halo Baylis-Hillman-type
adducts. Further optimizations of this new asymmetric
reaction and scope extension as well as its synthetic
applications will be studied in the future.
1
of H NMR spectroscopy in which 5% NOE was observed
between the signals of vinyl proton and methyl protons of
the E isomer of product 1. In contrast to N-trifluoroacetyl
oxazaborolidine, the N-pentafluorobutyryl-modified catalyst
resulted in consistent and reproducible Z/E selectivity.
Increasing reaction temperature or adding aldehyde in one
portion diminished Z/E selectivity and enantioselectivity.
Using deuterium solvents enabled us to directly observe
â-iodo TMS-allenolate. After the reaction, no such interme-
Acknowledgment. We gratefully acknowledge the Na-
tional Institutes of Health, General Medical Sciences (GM-
60261), and the Robert A. Welch Foundation (D-1361) for
generous support of this work and the National Science
Foundation (CHE-9808436) for partial funding of the 500
MHz NMR spectrometer. We thank Professor David M.
Birney for helpful suggestions and Mr. Subramanian Karur
and Jason D. Hook for their assistance.
1
diate was detected in the reaction mixture by H NMR
determination, which means that both enantiomers of silyl
allenolate were consumed. The facial selectivity of the
carbonyl addition is obviously controlled by the catalyst.
Further study will be conducted on the use of chiral
aldehydes or alternative catalysts to ascertain the asymmetric
kinetic resolution of â-iodo silyl allenolates. Since the present
reaction was carried out at -78 °C, the resulting isomeric
products are believed to result from kinetical control. Two
possible transition states (A and B) for forming individual
Z and E isomers are shown in Scheme 2. Transition state A
is obviously the predominant one; the steric effect between
the iodine and the hydrogen of the aldehyde is smaller than
the corresponding interaction in transition state B.
Supporting Information Available: Experimental, HPLC
1
analysis and H and 13C NMR spectra for all pure Z and E
isomers. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL000377+
(19) Synthetic transformations for absolute structure determination: (a)
MeI, Ag2O, MeCN, reflux. (b) i, KOCl, 0 °C, 12 h then rt 3 h; ii, H3O+.
(c) RuCl2‚3H2O (cat.), H5IO6 (3 equiv), CCl4/MeCN/H2O (1/1/1.4, v/v/v).
(d) Me3SiCl, MeOH, rt, overnight.
The absolute structure was determined by transforming
the E isomer of product 1 to methyl (S)-R-methoxyphenyl-
acetate which is derived from an authentic sample of (S)-
menthyl mandelate.18-21 The resulting chirality controlled by
oxazaborolidines derived from N-trifluoroacetyl and N-
heptafluoropropyl phenyalanine as well as N-trifluoro acetyl
tryptophane is consistent with that of the aldol reaction
catalyzed by Corey’s oxazaborolidine catalyst.3b,22 Therefore,
the asymmetric induction of the current reaction can be
suggested by using Corey’s oxazaborolidine-aldehhyde
complexation model.3b
(20) Harding, K. E.; Clement, K. S.; Gilbert, I. C.; Wiechman, B. J.
Org. Chem. 1984, 49, 2049.
(21) (a) Carlsen, P. H. J.; Katsuki, T.; Martin, S. V.; Sharpless, K. B. J.
Org. Chem. 1981, 46, 3936. (b) Martin, S. V.; Nunez, M. T. J. Org. Chem.
1990, 55, 1928.
(18) Kiooka, S.; Kaneko, Y.; Kume, K. Tetrahedron Lett. 1992, 33, 4927.
(22) Ishihara, K.; Kondo, S.; Yamamoto, H. Synlett 1999, 8, 1283.
826
Org. Lett., Vol. 3, No. 6, 2001