Catalytic asymmetric 1,4-addition reactions using α,β- unsaturated N-acylpyrroles as highly reactive monodentate α,β- unsaturated ester surrogates

Article abstract of DOI:10.1021/ja0485917

Synthesis and application of α,β-unsaturated N-acylpyrroles as highly reactive, monodentate ester surrogates in the catalytic asymmetric epoxidation and Michael reactions are described. α,β-Unsaturated N-acylpyrroles with various functional groups were synthesized by the Wittig reaction using ylide 2. A Sm(O-i-Pr)₃/H₈-BINOL complex was the most effective catalyst for the epoxidation to afford pyrrolyl epoxides in up to 100% yield and >99% ee. Catalyst loading was successfully reduced to as little as 0.02 mol % (substrate/catalyst = 5000). The high turnover frequency and high volumetric productivity of the present reaction are also noteworthy. In addition, a sequential Wittig olefination-catatytic asymmetric epoxidation reaction was developed, providing efficient one-pot access to optically active epoxides from various aldehydes in high yield and ee (96→99%). In a direct catalytic asymmetric Michael reaction of hydroxyketone promoted by the Et ₂Zn/linked-BINOL complex, Michael adducts were obtained in good yield (74-97%), dr (69/31-95/5), and ee (73-95%). This represents the first direct catalytic asymmetric Michael reaction of unmodified ketone to an α,β-unsaturated carboxylic acid derivative. The properties of α,β-unsaturated N-acylpyrrole are also discussed. Finally, the utility of the N-acylpyrrole unit for further transformations is demonstrated.

Full text of DOI:10.1021/ja0485917

Published on Web 05/26/2004
Catalytic Asymmetric 1,4-Addition Reactions Using
r,â-Unsaturated N-Acylpyrroles as Highly Reactive
Monodentate r,â-Unsaturated Ester Surrogates
Shigeki Matsunaga, Tomofumi Kinoshita, Shigemitsu Okada, Shinji Harada, and
Masakatsu Shibasaki*
Contribution from the Graduate School of Pharmaceutical Sciences, The UniVersity of Tokyo,
Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
Received March 11, 2004; E-mail: mshibasa@mol.f.u-tokyo.ac.jp
Abstract: Synthesis and application of R,â-unsaturated N-acylpyrroles as highly reactive, monodentate
ester surrogates in the catalytic asymmetric epoxidation and Michael reactions are described. R,â-
Unsaturated N-acylpyrroles with various functional groups were synthesized by the Wittig reaction using
ylide 2. A Sm(O-i-Pr)3/H8-BINOL complex was the most effective catalyst for the epoxidation to afford pyrrolyl
epoxides in up to 100% yield and >99% ee. Catalyst loading was successfully reduced to as little as 0.02
mol % (substrate/catalyst ) 5000). The high turnover frequency and high volumetric productivity of the
present reaction are also noteworthy. In addition, a sequential Wittig olefination-catalytic asymmetric
epoxidation reaction was developed, providing efficient one-pot access to optically active epoxides from
various aldehydes in high yield and ee (96f99%). In a direct catalytic asymmetric Michael reaction of
hydroxyketone promoted by the Et2Zn/linked-BINOL complex, Michael adducts were obtained in good yield
(74-97%), dr (69/31-95/5), and ee (73-95%). This represents the first direct catalytic asymmetric Michael
reaction of unmodified ketone to an R,â-unsaturated carboxylic acid derivative. The properties of R,â-
unsaturated N-acylpyrrole are also discussed. Finally, the utility of the N-acylpyrrole unit for further
transformations is demonstrated.
Introduction
been intensively investigated (Figure 1). With these templates,
Lewis acidic chiral catalysts are in many cases thought to
Enantioselective 1,4-additions of nucleophiles to R,â-unsatur-
ated carbonyl compounds is an important field of study in
asymmetric synthesis.1 Catalytic asymmetric 1,4-additions to
R,â-unsaturated esters afford versatile chiral building blocks,
because the ester moiety can be easily transformed into various
functional groups. Despite great achievements with R,â-unsatur-
ated ketones as substrates,1 however, there is room for improve-
ment in catalytic asymmetric 1,4-additions to R,â-unsaturated
esters, probably due to an intrinsic reactivity of R,â-unsaturated
ester that is lower than that of R,â-unsaturated ketone. To
address this issue, several functional groups acting as ester
surrogates, such as oxazolidinones,2 pyrrolidinone,2 acyl phos-
phonates,3 acyl pyrrazoles,4 pyrazolidinone,5 and imides,6 have
coordinate in a bidentate manner and activate substrates,7 leading
to high reactivity and excellent asymmetric induction. In some
chiral catalyses, however, bidentate coordination of the sub-
strates is unfavorable and the above-mentioned ester surrogates
are not at all suitable. In particular, difficulty arises when chiral
catalysts developed for monodentate R,â-unsaturated ketones
are applied for bidentate carboxylic acid derivatives.8 For
example, during our ongoing research project on catalytic
asymmetric Michael reactions of acyclic substrates9 and epoxi-
dation of electron deficient olefins,10 wherein the reaction is
(3) (a) Evans, D. A.; Scheidt, K. A.; Fandrick, K. R.; Lam, H.-W.; Wu, J. J.
Am. Chem. Soc. 2003, 125, 10780. For other applications, see also: (b)
Evans, D. A.; Johnson, J. S.; Olhava, E. J. J. Am. Chem. Soc. 2000, 122,
1635. (c) Telan, L. A.; Poon, C.-D.; Evans, S. A., Jr. J. Org. Chem. 1996,
61, 7455.
(1) For recent review for catalytic asymmetric 1,4-addition reaction, see: (a)
Krause, N.; Hoffmann-Ro¨der, A. Synthesis 2001, 171. (b) Sibi, M. P.;
Manyem, S. Tetrahedron 2000, 56, 8033. (c) ComprehensiVe Asymmetric
Catalysis; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin,
1999; Chapter 31. (d) Catalytic Asymmetric Synthesis, 2nd ed.; Ojima, I.,
Wiley: New York, 2000; Chapter 8D.
(2) (a) Ager, D. J.; East, M. B. Asymmetric Synthetic Methodology; CRC
Press: Boca Raton, 1996. For selected recent examples using oxazolidinone
and pyrrolidinone as achiral template in catalytic asymmetric 1,4-addition
reactions, see: (b) Evans, D. A.; Scheidt, K. A.; Johnston, J. N.; Willis,
M. C. J. Am. Chem. Soc. 2001, 123, 4480. (c) Sibi, M. P.; Manyem, S.;
Zimmerman, J. Chem. ReV. 2003, 103, 3263 and references therein. (d)
Zhuang, W.; Hazell, R. G.; Jørgensen, K. A. Chem. Commun. 2001, 1240.
(e) Kobayashi, S.; Ogawa, C.; Kawamura, M.; Sugiura, M. Synlett 2001,
983. (f) Miller, S. J.; Guerin, D. J. J. Am. Chem. Soc. 2002, 124, 2134. (g)
Hird, A. W.; Hoveyda, A. H. Angew. Chem., Int. Ed. 2003, 42, 1276. See
also ref 4c.
(4) For selected examples, see: (a) Sibi, M. P.; Shay, J. J.; Liu, M.; Jasperse,
C. P. J. Am. Chem. Soc. 1998, 120, 6615. (b) Itoh, K.; Kanemasa, S. J.
Am. Chem. Soc. 2002, 124, 13394. (c) Kanemasa, S. J. Synth. Org. Chem.
Jpn. 2003, 61, 1073 and references therein.
(5) (a) Sibi, M. P.; Liu, M. Org. Lett. 2001, 3, 4181. For other applications,
see an excellent review on chiral relay using achiral template: (b)
Corminboeuf, O.; Quaranta, L.; Renaud, P.; Liu, M.; Jasperse, C. P.; Sibi,
M. P. Chem.sEur. J. 2003, 9, 28. See also: (c) Sibi, M. P.; Ma, Z.-H.;
Jasperse, C. P. J. Am. Chem. Soc. 2004, 126, 718 and references therein.
(6) (a) Taylor, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 2003, 125, 11204.
(b) Sammis, G. M.; Jacobsen, E. N. J. Am. Chem. Soc. 2003, 125, 4442.
(c) Myers, J. K.; Jacobsen, E. N. J. Am. Chem. Soc. 1999, 121, 8959. (d)
Goodman, S. N.; Jacobsen, E. N. AdV. Synth. Catal. 2002, 344, 953. (e)
Sibi, M. P.; Prabagaran, N.; Ghorpade, S. G.; Jasperse, C. P. J. Am. Chem.
Soc. 2003, 125, 11796.
(7) For imides as substrate, both monodentate and bidentate transition states
were proposed, depending on the catalyst and reaction. See ref 6c and e.
9
10.1021/ja0485917 CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 7559-7570
7559

A R T I C L E S
Matsunaga et al.
Figure 2. Structure of R,â-unsaturated N-acylpyrrole 1, ylide 2, (R)-H8-
BINOL 3, hydroxyketone 4, and (S,S)-linked-BINOL 5.
high ee (96f99.5%).11 The utility of R,â-unsaturated N-
acylpyrrole 1 was also demonstrated in the first direct catalytic
asymmetric Michael reaction of an unmodified ketone to R,â-
unsaturated carboxylic acid derivatives. The Et2Zn/linked-
BINOL 5 catalyst,12 which was originally optimized for R,â-
unsaturated ketones,9a,9b promoted the Michael reaction of
hydroxyketone 4 smoothly to afford products in good yield (73-
97%), dr (69/31-95/5), and ee (73-95%). The properties of
R,â-unsaturated N-acylpyrrole transformations of an N-acylpyr-
role moiety and application to the synthesis of fragments of
natural products are also described.
Figure 1. (a) Bidentate R,â-unsaturated ester surrogates; (b) Monodentate
R,â-unsaturated ketone and ester surrogate.
considered to proceed through the 1,4-addition of peroxide such
as tert-butyl hydroperoxide (TBHP), it was often difficult to
apply our catalysts to the above-mentioned carboxylic acid
derivatives, which favored bidentate coordination. The mis-
matched coordination mode of the substrates to the chiral
catalysts is assumed to result in significantly decreased enan-
tioselectivity and/or reactivity. Thus, the development of a highly
reactive, monodentate ester surrogate is desirable to broaden
the substrate generality of asymmetric catalysis, which was
originally developed for R,â-unsaturated ketones.
Here, we report the utility of R,â-unsaturated N-acylpyrrole
as a monodentate R,â-unsaturated ester surrogate, which is a
highly reactive and versatile substrate. R,â-Unsaturated N-
acylpyrroles 1 were easily prepared from various aldehydes
using a pyrrolylmethylenetriphenylphosphorane (2, Figure 2).
High reactivity of the R,â-unsaturated N-acylpyrrole 1 was first
demonstrated in catalytic asymmetric epoxidations with a novel
Sm(O-i-Pr)3/H8-BINOL complex. Catalyst loading was suc-
cessfully reduced to as little as 0.02 mol % (substrate/catalyst
) 5000), while maintaining high enantiomeric excess (up to
99% ee). With reduced catalyst loading, the reaction was
performed under concentrated conditions (up to 3 M). A one-
pot sequential Wittig-catalytic asymmetric epoxidation process
provided efficient one-pot access to optically active pyrrolyl
epoxides from various aldehydes in high yield (72-100%) and
Results and Discussion
(A) Synthesis of r,â-Unsaturated N-Acylpyrrole: Although
the potential of N-acylpyrroles to function as ester surrogates
was reported two decades ago,13 it was only recently that
N-acylpyrroles were used in fine organic synthesis, such as
asymmetric synthesis.14 Inspired by recent reports on the unique
reactivity of N-acylpyrroles by Evans et al.,15a we investigated
the properties of R,â-unsaturated N-acylpyrroles in detail.16
Although Arai reported an efficient synthesis of R,â-unsaturated
N-acyl-2-substituted-pyrroles,14d the method was not applicable
to an unsubstituted pyrrole, probably due to the high reactivity
of pyrrole at the 2-position. The reaction of lithiated pyrrole
with R,â-unsaturated acyl halide gave unsatisfactory results. In
the literature, the most efficient synthesis of R,â-unsaturated
(11) A portion of this article was communicated previously. Kinoshita, T.; Okada,
S.; Park, S.-R.; Matsunaga, S.; Shibasaki, M. Angew. Chem., Int. Ed. 2003,
42, 4680.
(12) For the synthesis of linked-BINOL 5, see: (a) Matsunaga, S.; Das, J.; Roels,
J.; Vogl, E. M.; Yamamoto, N.; Iida, T.; Yamaguchi, K.; Shibasaki, M. J.
Am. Chem. Soc. 2000, 122, 2252. (b) Matsunaga, S.; Ohshima, T.;
Shibasaki, M. AdV. Synth. Catal. 2002, 344, 3. Both enantiomers of linked-
BINOL are also commercially available from Wako Pure Chemical
Industries, Ltd. Catalog No. for (S,S)-5 is No. 152-02431 and (R,R)-5, No.
155-02421.
(13) (a) Lee, S.-D.; Brook, M. A.; Chan, T.-H. Tetrahedron Lett. 1983, 24, 1569.
(b) Brandange, S.; Rodriguez, B. Acta Chem. Scand., Ser. B 1987, 41, 740.
(14) Application to catalytic asymmetric reactions: (a) Evans, D. A.; Johnson,
D. S. Org. Lett. 1999, 1, 595. (b) Evans, D. A.; Scheidt, K. A.; Johnston,
J. N.; Willis, M. C. J. Am. Chem. Soc. 2001, 123, 4480. (c) Hodous, B. L.;
Fu, G. C. J. Am. Chem. Soc. 2002, 124, 10006. Application to diastereo-
selective reactions: (d) Arai, Y.; Matsuda, T.; Masaki, Y. Chem. Lett. 1997,
145. (e) Arai, Y.; Ueda, K.; Xie, J.; Masaki, Y. Chem. Pharm. Bull. 2001,
49, 1609. (f) Arai, Y.; Kasai, M.; Ueda, K.; Masaki, Y. Synthesis 2003,
1511 and references therein.
(15) (a) Evans, D. A.; Borg, G.; Scheidt, K. A.Angew. Chem., Int. Ed. 2002,
41, 3188. For other applications of pyrrole carbinol as useful building
blocks, see: (b) Dixon, D. J.; Scott, M. S.; Luckhurst, C. A. Synlett 2003,
2317.
(8) For example, catalytic asymmetric addition of organozinc reagents to R,â-
unsaturated carboxylic acid derivatives remained problematic until Hov-
eyda’s recent excellent report (ref 2g). Hoveyda et al. developed a noVel
chiral ligand to achieve catalytic asymmetric 1,4-addition of R2Zn to
bidentate substrates in high selectivity, suggesting that conventional various
chiral ligands developed for enones were not suitable for 1,4-addition of
R2Zn to bidentate substrates with chiral metal catalysts. Appropriate tuning
of chiral ligands was necessary. For representative recent examples of
catalytic asymmetric 1,4-addition of organozinc reagents, see ref 1a and
2g and references therein.
(9) For selected recent examples of 1,4-addition reactions of acyclic R,â-
unsaturated ketone from our group, see: (a) Harada, S.; Kumagai, N.;
Kinoshita, T.; Matsunaga, S.; Shibasaki, M. J. Am. Chem. Soc. 2003, 125,
2582. (b) Kumagai, N.; Matsunaga, S.; Shibasaki, M. Org. Lett. 2001, 3,
4251. (c) Funabashi, K.; Saida, Y.; Kanai, M.; Arai, T.; Sasai, H.; Shibasaki,
M. Tetrahedron Lett., 1998, 39, 7557. (d) Yamagiwa, N.; Matsunaga, S.;
Shibasaki, M. J. Am. Chem. Soc. 2003, 125, 16178.
(10) Epoxidation of R,â-unsaturated ketone: (a) Bougauchi, M.; Watanabe, T.;
Arai, T.; Sasai, H.; Shibasaki, M. J. Am. Chem. Soc. 1997, 119, 2329. (b)
Nemoto, T.; Ohshima, T.; Yamaguchi, K.; Shibasaki, M. J. Am. Chem.
Soc. 2001, 123, 2725. See, also (c) Daikai, K.; Kamaura, M.; Inanaga, J.
Tetrahedron Lett. 1998, 39, 7321. For other examples, see ref 20.
(16) Although Arai et al. reported the utility of R,â-unsaturated N-acyl-2-
substituted-pyrrole as a Michael acceptor, it might not function as a
monodentate substrate because of the coordinating functional group at
2-position. See ref 14e and f.
9
7560 J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004

R,â-Unsaturated N-Acylpyrroles
A R T I C L E S
Scheme 1. Preparation of Pyrrolylmethylenetriphenylphosphorane
(Ylide 2)a
Table 1. Synthesis of Various R,â-Unsaturated N-Acylpyrrole 1
a Conditions: (i) 2,5-dimethoxytetrahydrofuran, AcOH, 100 °C; (ii) PPh3,
toluene, 100 °C; (iii) 2 M aq NaOH, CH2Cl2/H2O; (iv) THF/Et2O, -78 to
25 °C.
N-acylpyrrole was the condensation of cinnamamide and 2,5-
15a
dimethoxytetrahydrofuran in acetic acid at 100 °C
(or with
SOCl2 at rt 80 °C),17 affording 1a in 60-68% yield. For the
synthesis of R,â-unsaturated N-acylpyrroles with various func-
tional groups, however, a milder method was necessary. One
reason that R,â-unsaturated N-acylpyrrole has not been widely
used as an acceptor for 1,4-addition reactions might be the lack
of efficient preparation methods under mild conditions.
We examined the Wittig reaction using pyrrolylmethylene-
triphenylphosphorane (ylide 2). Initially, ylide 2 was prepared
from chloroacetamide in three steps as shown in Scheme 1a.
Low reproducibility of the first condensation step and the
lachrymatory property of intermediate N-chloroacetyl pyrrole
prompted us to examine a different route for ylide 2. Following
the reported procedure for the reaction of N-acylimidazole and
methylenetriphenylphosphorane 9 to afford alkanoylmethylene-
triphenylphosphorane,18 the reaction of carbonyldipyrrole 819
with 9 was examined (Scheme 1b). For the preparation of 9
from methyltriphenylphosphonium bromide, PhLi was used
instead of BuLi. With BuLi as the base, ylide containing a
pyrrole moiety and one butyl group derived from nucleophilic
substitution at the phosphorus atom was obtained as a side
product. We obtained 2 as a colorless to pale brown powder in
98% yield when 3 equiv of 9 were used, and 2 could be stored
at ambient temperature under air for at least 5 months.
The substrate scope and limitations of the Wittig reaction
using ylide 2 and various aldehydes 10 are summarized in Table
1. With aromatic (entries 1-9) and heteroaromatic (entries 10-
12) aldehydes, both chemical yield and E/Z ratio were good.
Both the electron donating (entries 3 and 4) and withdrawing
(entries 5 and 6) substituents at the para-position gave products
in high yield and E/Z ratio (>20/1). With ortho-substituents,
the E/Z ratio decreased to 7/1 (entry 8) and 10/1 (entry 9).
Various linear and branched aliphatic aldehydes were also
applicable (entries 13-25). The E/Z ratio was dependent on
the substrate, ranging from 3/1 to >20/1. Sterically crowded
aldehydes (entries 16 and 17) required a high reaction temper-
ature and a long reaction time but still afforded products in
moderate yield (entry 16, 63%, E/Z ) >20/1 and entry 17, 63%,
E/Z ) >20/1). Benzyl ether (entry 17), p-methoxybenzyl ether
(entry 18), benzyloxymethyl ether (entry 19), tert-butyldimeth-
ylsilyl ether (entry 20), methyl ketone (entry 21), olefin (entry
22), N-Boc (entry 24), and acetonide (entry 25) were compatible
under the reaction conditions. R,â-Unsaturated aldehyde also
gave product in good yield (entry 23).
(B) Application to Catalytic Asymmetric Epoxidation: To
evaluate the reactivity of R,â-unsaturated N-acylpyrroles, we
examined catalytic asymmetric epoxidation using R,â-unsatur-
ated N-acylpyrrole 1a as a substrate. Catalytic asymmetric
epoxidation of R,â-unsaturated carbonyl compounds is one of
the most important transformations in organic synthesis.20
Although we10 and others10,20 have achieved efficient catalytic
asymmetric epoxidation of R,â-unsaturated ketones, there are
(20) Recent reviews: (a) Porter, M. J.; Skidmore, J. Chem. Commun. 2000,
1215. (b) Nemoto, T.; Ohshima, T.; Shibasaki, M. J. Synth. Org. Chem.
Jpn. 2002, 60, 94. Polyamino acid catalysis: (c) Porter, M. J.; Roberts, S.
M.; Skidmore, J. Bioorg. Med. Chem. 1999, 7, 2145. Chiral ketone as
catalyst: (d) Frohn, M.; Shi, Y. Synthesis 2000, 1979. For selected leading
references, see also: (e) Enders, D.; Zhu, J.; Raabe, G. Angew. Chem., Int.
Ed. Engl. 1996, 35, 1725. (f) Elston, C. L.; Jackson, R. F. W.; MacDonald,
S. J. F.; Murray, P. J. Angew. Chem., Int. Ed. Engl. 1997, 36, 410.
(17) Ekkati, A. R.; Bates, D. K. Synthesis 2003, 1959 and references therein.
The scope of the reaction was limited to aromatic amides and cinnamamide.
(18) Miyano, M.; Stealey, M. A. J. Org. Chem. 1975, 40, 2840.
(19) Synthesis and application of carbonyldipyrrole 8 was reported by Evans et
al. See ref 15a.
9
J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004 7561

A R T I C L E S
Matsunaga et al.
substituents (imidazolide, 79-94% ee; amide, 99% ee), many
problems remain from a practical viewpoint. (a) Catalyst loading
and reaction rate: 5-10 mol % catalyst loading was essential
for good conversion; in many cases it was impossible to
complete the conversion with less than 5 mol %. Turnover
frequency of the catalyst was, in most cases, approximately 1-3
h-1. (b) Oxidant: explosive TBHP was essential for good
reactivity. From a practical viewpoint, it is desirable to use less
explosive (less reactive) cumene hydroperoxide (CMHP). (c)
The preparation of â-alkyl R,â-unsaturated acid imidazolides
from aldehydes was lengthy. (d) Some of the â-alkyl R,â-
unsaturated acid imidazolides were very unstable. (e) Volumetric
productivity of the reported epoxidation was rather low (0.1
M), which is not suitable for practical synthesis on a large scale.
Because the concentration of the lanthanide catalyst was
maintained at less than 1-10 mM to achieve high enantio-
selectivity, it was difficult to improve volumetric productivity
unless catalyst loading was improved.25 The low solubility of
acid imidazolides is also problematic for improving volumetric
productivity.
The catalytic asymmetric epoxidation of 1a proceeded
smoothly, as summarized in Table 2. With 10 mol % of the
Sm(O-i-Pr)3/(R)-BINOL complex26 and 10 mol % of Ph3As-
(O),10 the reaction was complete within 0.5 h and afforded 11a
in 93% yield and 94% ee (entry 1). The reaction rate was much
faster than when using acid imidazolide and morpholinyl amide
and as fast as that using R,â-unsaturated ketone.27 The reaction
also proceeded smoothly with 5 mol % catalyst (entry 2: y.
85%, 96% ee).28 To improve the enantioselectivity, various
BINOL derivatives were screened to determine that H8-BINOL
functioned best. The H8-BINOL complex gave better results
than the BINOL complex, probably due to the large bite angle.29
With a novel Sm(O-i-Pr)3/(R)-H8-BINOL complex, 11a was
obtained in 99% ee (entry 3). Ph3P(O) was also effective for
the present substrate10 (entries 4-6). THF/toluene mixed solvent
produced better results than THF alone, and 11a was obtained
in 96-99% ee depending on the amount of Ph3P(O) (entries
7-9). Both the reaction rate and selectivity were highest with
100 mol % of Ph3P(O) (entry 9). Under the best conditions
(THF/toluene, Ph3P(O): 100 mol %), less explosive and less
reactive cumene hydroperoxide (CMHP) was also applicable
and the reaction reached completion within 0.2 h with 5 mol %
catalyst (entry 10, y. 91%, >99.5% ee).30 The result in entry
10 gave additional practical benefits to the present system
compared with previous reports.23,24
Figure 3. Postulated reaction mechanism of catalytic asymmetric epoxi-
dation promoted by lanthanide-BINOL complex.
only a few examples of R,â-unsaturated esters as substrates using
a salen-Mn complex21 or chiral ketones22 as the catalysts. In
both cases, except for Shi’s recent report,22a only â-aryl-
substituted R,â-unsaturated esters were used. Although Shi et
al. reported excellent results with a few â-alkyl substituted
substrates, trans-â-alkyl-substituted R,â-unsaturated ester was
not applicable, thus leaving room for improvement in substrate
generality.
The proposed catalytic cycle of the asymmetric epoxidation
reaction is shown in Figure 3. A lanthanide alkoxide moiety
would change to lanthanide-peroxide through proton exchange
(I). The lanthanide-BINOL complex also functions as a Lewis
acid to activate electron deficient olefins through monodentate
coordination (II). Enantioselective 1,4-addition of lanthanide-
peroxide gave intermediate enolate (III), followed by epoxide
formation to regenerate catalyst (IV). In the transition state for
the 1,4-addition step, the lanthanide-BINOL complex is pos-
tulated to favor the monodentate coordination mode for high
enantioselectivity. Although lanthanide-BINOL complexes were
highly reactive and enantioselective for the catalytic asymmetric
epoxidation of R,â-unsaturated ketones,10 it was rather difficult
to apply lanthanide-BINOL complexes to R,â-unsaturated
esters due to its low reactivity. When using bidentate R,â-
unsaturated oxazolidinone amide as a substrate, epoxide was
obtained in only 73% yield and 87% ee, using as much as 20
mol % catalyst loading after 24 h.23 Recently, we partially solved
this substrate limitation problem by using R,â-unsaturated
carboxylic acid imidazolide23 and R,â-unsaturated morpholinyl
amide24 as substrates. Although moderate to high enantiomeric
excess was achieved using substrates with â-aryl- and â-alkyl-
(25) Our mechanistic studies suggested that several species exist in equilibrium
in the mixture of the Ln(O-i-Pr)3/BINOL complex. Catalyst concentration
should be kept low (<5-10 mM) to achieve the best enantioselectivity,
because undesired dimeric and oligomeric species would increase under
concentrated conditions. See ref 10b for mechanistic studies.
(26) Standard condition for catalytic asymmetric epoxidation is lanthanide metal
alkoxide/BINOL derivative ) 1/1 with Ph3As(O) or Ph3P(O) as an additive.
Among lanthanide metals screened (La, Pr, Nd, Sm, Gd, Er, and Yb), Sm
had best reactivity and selectivity. La, Pr, Nd, Gd, and Er complexes also
afforded 11a in >90% ee and >80% yield. Yb was not suitable for 1a.
Sm(O-i-Pr)3 was purchased from Kojundo Chemical Laboratory Co., Ltd.
(Fax: +81-492-84-1351. sales@kojundo.co.jp).
(21) Jacobsen, E. N.; Deng, L.; Furukawa, Y.; Martinez, L. E. Tetrahedron 1994,
50, 4323.
(22) (a) Wu, X.-Y.; She, X.; Shi, Y. J. Am. Chem. Soc. 2002, 124, 8792. (b)
Armstrong, A.; Hayter, B. R. Chem. Commun. 1998, 621. (c) Solladie´-
Cavallo, A.; Boue´rat, L. Org. Lett. 2000, 2, 3531. (d) Seki, M.; Furutani,
T.; Imashiro, R.; Kuroda, T.; Yamanaka, T.; Harada, N.; Arakawa, H.;
Kusama, M.; Hashiyama T. Tetrahedron Lett. 2001, 42, 8201. See also:
(e) Yang, D.; Yip, Y.-C.; Tang, M.-W.; Wong, M.-K.; Zheng, J.-H.; Cheung,
K.-K. J. Am. Chem. Soc. 1996, 118, 491.
(27) Reaction time for imidazolide: with 10 mol % of La-BINOL catalyst, at
room temperature, 3.5 h, 86% yield and 5 mol % of La-BINOL catalyst,
at room temperature, 12 h, 73% yield. Reaction time for amide: 5-10
mol % Sm-BINOL catalyst, at room temperature, 3-24 h, >90% yield.
See refs 23 and 24.
(23) R,â-Unsaturated carboxylic acid imidazolide: (a) Nemoto, T.; Ohshima,
T.; Shibasaki, M. J. Am. Chem. Soc. 2001, 123, 9474. (b) Ohshima, T.;
Nemoto, T.; Tosaki, S.-y.; Kakei, H.; Gnanadesikan, V.; Shibasaki, M.
Tetrahedron 2003, 59, 10485.
(24) (a) R,â-Unsaturated amide: Nemoto, T.; Kakei, H.; Gnanadesikan, V.;
Tosaki, S.-y.; Ohshima, T.; Shibasaki, M. J. Am. Chem. Soc. 2002, 124,
14544. (b) Tosaki, S.-y.; Nemoto, T.; Ohshima, M.; Shibasaki, M. Org.
Lett. 2003, 5, 495. Application to 1,3-polyol synthesis: (c) Tosaki, S.-y.;
Horiuchi, T.; Nemoto, T.; Ohshima, M.; Shibasaki, M. Chem.sEur. J. 2004,
10, 1527.
(28) With imidazolide or amide as substrate, 5-10 mol % catalyst loading was
essential for completion of the reaction.
(29) Review for H8-BINOL: Terry, T.-L. A.-Y.; Chan, S.-S.; Chan, A. S. C.
AdV. Synth. Catal. 2003, 345, 537.
(30) With less reactive CMHP as oxidant, acid imidazolide resulted in low
chemical yield (18 h, 47% yield with 10 mol % catalyst). See ref 23.
9
7562 J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004

R,â-Unsaturated N-Acylpyrroles
A R T I C L E S
Table 2. Catalytic Asymmetric Epoxidation Reaction of R,â-Unsaturated N-Acylpyrrole 1a
c
Sm(O-i-Pr)3
(x mol %)
ligand
(x mol %)
additive
(y mol %)
time
(h)
yieldb
(%)
ee
entry
solvent
THF
THF
THF
THF
THF
THF
THF/toluene
THF/toluene
THF/toluene
THF/toluene
oxidanta
(%)
1
2
3
4
5
6
7
8
9
10
10
5
5
5
5
5
5
5
5
5
BINOL (10)
BINOL (5)
Ph3As(O) (10)
Ph3As(O) (5)
Ph3As(O) (5)
Ph3P(O) (15)
Ph3P(O) (50)
Ph3P(O) (100)
Ph3P(O) (15)
Ph3P(O) (50)
Ph3P(O) (100)
Ph3P(O) (100)
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
CMHP
0.5
0.5
0.5
0.5
0.5
0.5
0.4
0.5
0.2
0.2
93
85
94
84
88
85
85
92
97
91
94
96
99
94
98
97
96
99
99
H8-BINOL (5)
H8-BINOL(5)
H8-BINOL (5)
H8-BINOL(5)
H8-BINOL (5)
H8-BINOL(5)
H8-BINOL (5)
H8-BINOL(5)
>99.5
a TBHP in decane or CMHP in toluene was used. b Isolated yield. c Determined by chiral HPLC analysis.
Table 3. Trials to Reduce Catalyst Loading in Catalytic Asymmetric Epoxidation Reaction of R,â-Unsaturated N-Acylpyrrole 1
Sm(O-i-Pr)3
(x mol %)
H -BINOL
(x mol %)
additive
(y mol %)
MS4Å
(mg/mmol of 1a)
concn of
[1a] (M)
time
(h)
yielda
(%)
eeb
(%)
8
entry
1c
2c
3d
4d
5d
6d
7d
8d
5
1
5
1
Ph3P(O) (100)
Ph3P(O) (100)
Ph3P(O) (100)
Ph3P(O) (100)
Ph3P(O) (100)
Ph3As(O) (0.1)
Ph3As(O) (0.05)
Ph3As(O) (0.02)
1000
500
250
100
100
100
100
100
0.1
1
1
2
2
3
3
3
0.2
0.3
0.6
1
2
0.6
1
97
94
100
99
99
99
97
97
96
99
98
99
0.5
0.2
0.1
0.1
0.05
0.02
0.5
0.2
0.1
0.1
0.05
0.02
90
100
100
94
1.5
a Isolated yield. b Determined by chiral HPLC analysis. c TBHP in decane was used. d Anhydrous TBHP in toluene (dried with MS 4A) was used.
We then tried to reduce the catalyst loading. As expected,
based on the high reaction rate with 5 mol % catalyst loading
in Table 2, catalyst loading was easily reduced. As summarized
in Table 3, the epoxidation reaction of 1a completed using as
little as 1, 0.5, 0.2 mol % of Sm(O-i-Pr)3/H8-BINOL complex
with 100 mol % of Ph3P(O), giving product 11a in high yield
(94-100%) and high enantiomeric excess (96-99% ee) after
0.3 h (entry 2), 0.6 h (entry 3), and 1 h (entry 4). The reaction
proceeded well with as little as 0.1 mol % catalyst loading
(substrate/catalyst ) 1,000), affording product in 90.4% yield
(TON ) 904) and 96% ee (entry 5). With the Sm(O-i-Pr)3/
H8-BINOL/Ph3As(O) ) 1/1/1 complex, catalyst loading was
further reduced to 0.1, 0.05, and 0.02 mol % (substrate/catalyst
) up to 5000) as summarized in entries 6-8. After 1.5 h, 2.01
g of 11a (94.2% yield) was obtained in 99% ee using 0.589
mg of H8-BINOL (0.02 mol % catalyst loading) and 0.644 mg
of Ph3As(O) in entry 8. The high catalyst turnover number (TON
) 4710 based on the metal and chiral ligand used) and the
catalyst turnover frequency (TOF ) up to >3000 h-1) in the
present system were far better compared with previous reports
from our group with carboxylic acid imidazolide (substrate/
catalyst ) <10, TOF ) 2.5 h-1 with the La-BINOL catalyst)23
and morpholinyl amide (substrate/catalyst ) 10-20, TOF )
<2 h-1 with the Sm-BINOL catalyst).24 For precise comparison
of the reactivity using the same Sm-H8-BINOL catalyst, see
section E.31 Because the commercially available TBHP solution
in decane (5-6 M) contains up to 4% water, the use of the wet
TBHP solution in decane was not suitable for reduced catalyst
loading (entries 3-8). Thus, TBHP in toluene (ca. 6 M) dried
with MS 4A (water content e 0.4%) was alternatively used to
avoid decomposition of the Sm(O-i-Pr)3/H8-BINOL catalyst by
too much H2O. TBHP in toluene was added slowly to the
mixture of the Sm(O-i-Pr)3/H8-BINOL catalyst, either Ph3P(O)
or Ph3As(O), 1a, and MS 4A in THF/toluene. To reduce the
catalyst loading, it was also important to keep the concentration
of the catalyst within 1-5 mM, which is similar to the best
conditions with 5 mol % catalyst loading ([Sm(O-i-Pr)3/H8-
BINOL] ) 5 mM). Thus, the concentration of substrate 1a
increased accordingly when the reaction shown in Table 3 was
performed. For example, [1a] was 2 M in entry 5 and 3 M in
entries 6-8. In previous reports from our group on the catalytic
asymmetric epoxidation reaction, the standard substrate con-
centration for substrates was approximately 0.1 M, because the
concentration of the catalyst needed to be kept low ([cat] )
<5-10 mM) to achieve best enantioselectivity due to the
equilibrium between many catalyst species.25 Thus, the much
(31) Difference in reaction rate is in part ascribed to the difference in metal
and chiral ligand used in previous reports (refs 23 and 24). For the precise
comparison of reactivity of R,â-unsaturated N-acylpyrrole, N-acylimidazole,
ketone, and amide under exactly the same conditions, see section E. R,â-
Unsaturated N-acylimidazole showed the best reaction rate as long as 5
mol % Sm-H8-BINOL catalyst was used, although the reaction did not
proceed well with 1 mol % catalyst loading in the case of N-acylimidazole
as substrate (section E, Scheme 4).
9
J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004 7563

A R T I C L E S
Matsunaga et al.
Wittig reaction is generally considered less efficient due to the
production of waste Ph3P(O). We hypothesized that reusing
waste Ph3P(O) in the epoxidation reaction as a useful modulator
for the Sm-H8-BINOL complex would partially compensate for
the disadvantage of the Wittig reaction. Thus, a sequential
Wittig-catalytic asymmetric epoxidation reaction was designed
in which the waste Ph3P(O) from the first Wittig reaction was
reused as an additive in the second epoxidation reaction. A
sequential Wittig-catalytic asymmetric epoxidation of 10a
afforded 11a in 96% yield (2 steps) and 99.8% ee (er 999/1;
Scheme 2A). As expected, the selectivity and reactivity were
much lower when the reaction was performed step-by-step
without adding Ph3P(O) in the second step (none 75.2% ee, er
7.06/1; Scheme 2B). By adding external Ph3P(O) (15 mol %
and 50 mol %) in the second step after isolation of intermediate
1a and Ph3P(O) (1 equiv) as waste, 11a was obtained in 96.8%
ee (er 61.5/1) and 98.8% ee (er 166/1), respectively (Scheme
2B). These results suggested that (a) Ph3P(O) functioned as an
effective additive to improve yield and ee but (b) only a portion
of 1 equiv of waste Ph3P(O) was necessary to achieve high ee
in the second reaction. Comparison of the purification process
(Scheme 2A, once vs Scheme 2B, twice), however, indicated
that the total efficiency of the process from aldehyde to epoxide
11a was highest in the sequential process (A). Total efficiency
of the present system is also better than the previous results
with acid imidazolide23 and morpholinyl amide,24 which were
synthesized and purified prior to use.
Figure 4. Effects of H2O on chemical yield and ee.
higher volumetric productivity of the present system together
with reduced catalyst loading is noteworthy for large scale
synthesis. Because the amount of MS 4A needed depends on
the amount of solvent used, the amount of MS 4A was also
successfully reduced from 1000 mg/mmol substrate to 100 mg/
mmol under the optimized conditions (entries 4-8), which is
also a practical advantage of the present system. Further
experiments to reduce catalyst loading were unsuccessful,
because it was difficult to maintain the appropriate concentration
of Sm catalyst due to solubility problems of product 11a.32
The amount of H2O present during preparation of the Sm
catalyst greatly affected the reaction rate and enantioselectivity
in the epoxidation reactions. The amount of H2O in the reaction
mixture depended on the conditions of MS 4A used. The
relations between H2O concentration and reactivity, yield, and
ee are summarized in Figure 4. The highest ee and reactivity
were achieved when TBHP in decane was added to the Sm(O-
i-Pr)3/H8-BINOL complex and MS 4A with an H2O concentra-
tion of less than 10 ppm. As summarized in Figure 4, both
reactivity and selectivity of the catalyst decreased significantly
with an H2O concentration of more than 10 ppm. Because a
small increase in the H2O concentration drastically retarded the
reaction, it is recommended to dry MS 4A at 160 °C for 1-3
h under 0.7 KPa prior to use to achieve reliable and reproducible
results. When using activated MS 4A and 5 mol % catalyst,
commercially available TBHP in decane, which might contain
up to 4% H2O, was used directly as received without any
problems. The results indicate that the H2O content during
catalyst formation from Sm(O-i-Pr)3 and H8-BINOL is important
and is not very problematic in the presence of peroxide as long
as 5 mol % catalyst was used.
As summarized in Table 4, the present sequential reaction
has broad substrate generality, affording epoxides in excellent
enantiomeric excess (96f99.5% ee) and good yield (entries
1-14: y. 72-96% from aldehydes). In the sequential reactions,
1.3 equiv of ylide 2 were employed to complete the Wittig
reaction. The remaining ylide 2 had no adverse effects on the
subsequent epoxidation step. Aromatic aldehydes with various
substituents (entries 1-7), aliphatic linear and branched alde-
hydes with functional groups (entries 8-13), and R,â-unsatur-
ated aldehyde (entry 14) were applicable to the present catalysis.
Because the epoxidation reaction proceeded through 1,4-addition
of peroxide, the reaction proceeded chemoselectively in the
presence of an unactivated carbon-carbon double bond (entry
13). The reaction proceeded chemoselectively with aldehyde
10u possessing a methyl ketone unit (entry 12). In entry 14,
the epoxidation reaction selectively occurred at the 2,3-position.
4,5-Epoxide was not detected. In many examples, less explosive
CMHP was used as an oxidant instead of the explosive TBHP,
providing additional practical benefits to the present methodol-
ogy. The epoxidation reaction of 1q derived from R,R-
disubstituted aldehyde 10q as the substrate was not successful,
probably due to steric hindrance. The E/Z ratio in the first Wittig
olefination step partially affected the total chemical yield in the
sequential reactions (two steps) with aliphatic aldehydes (entries
8-14) and aromatic aldehydes with an ortho-substituent (entries
6-7). The epoxidation of Z-R,â-unsaturated N-acylpyrrole
proceeded rather slowly. For example, isolated Z-R,â-unsatur-
ated N-acylpyrrole 12 gave cis-epoxide 13 as the major product
in 32% yield and 86% ee after 1 h using 10 mol % Sm catalyst,
5 mol % Ph3As(O) as an additive, and TBHP as an oxidant
(Scheme 3). The chemical yield was not improved, even after
a longer reaction time. When 12 was used as the substrate, no
trans-epoxide was observed. Under the sequential reaction
(C) Sequential Wittig Olefination-Catalytic Asymmetric
Epoxidation: Although the functional group compatibility of
the Wittig reaction is good because of its mild reactivity, the
(32) With 0.01 mol % Sm(O-i-Pr)3 ([Sm-cat] ) 0.3 mM), epoxide was obtained
in less than 10% yield.
9
7564 J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004

R,â-Unsaturated N-Acylpyrroles
A R T I C L E S
Scheme 2. Wittig-Catalytic Asymmetric Epoxidation Reaction with (A) One-Pot Sequential Process and (B) Step-by-Step Process with
External Ph3P(O)
conditions, that is with 100 mol % of Ph3P(O), most of the
Z-isomer remained after the second epoxidation reaction as
observed by TLC analysis. Only trace amounts of the cis-
epoxide were obtained under the sequential reaction conditions.
With chiral aldehyde 10z, the sequential reaction proceeded
in an almost completely catalyst-controlled manner (Table 5).
The epoxide was obtained in a ratio of 14/15 ) >99/1 with the
(R)-catalyst (matched pair, entry 1) and 14/15 ) 1/36 with the
(S)-catalyst (mismatched pair, entry 3). In the case of the
mismatched pair, TBHP was required to attain good reactivity
(entry 2 vs entry 3). With achiral oxidating reagents, there was
much lower selectivity (dr ) 54/46).33
(D) Application of r,â-Unsaturated N-Acylpyrrole to a
Direct Catalytic Asymmetric Michael Reaction: To further
demonstrate the utility of R,â-unsaturated N-acylpyrrole as a
monodentate ester surrogate, we examined the direct catalytic
asymmetric Michael reaction of hydroxyketone 4 promoted by
the Et2Zn/linked-BINOL complex. Because the Et2Zn/linked-
BINOL complex was effective for the Michael reaction of R,â-
unsaturated ketone,9a,b,34 the system was chosen as a model
reaction. Direct catalytic asymmetric Michael addition of
unmodified ketone or aldehyde was recently studied inten-
sively.35 Although a few efficient catalyses were reported
recently using R,â-unsaturated ketone and nitro-olefin, several
critical problems remain to be addressed. One is the extension
of a highly enantioselective direct catalytic asymmetric Michael
addition of unmodified ketone or aldehyde to R,â-unsaturated
carboxylic acid derivatives.
slightly lower than that of the R,â-unsaturated ketone, all the
reactions were performed at 0 °C. Michael adducts were
obtained in good dr (81/19-95/5), yield (74-97%), and ee (88-
95%) from â-aromatic and heteroaromatic substituted R,â-
unsaturated N-acylpyrrole. The slightly lower enantioselectivity
than that with R,â-unsaturated ketone can be ascribed to the
reaction temperature. In the case of â-alkyl substituents (entry
10), the selectivity tendency was similar to that obtained with
enone, that is, only modest dr and ee were achieved. To the
best of our knowledge, this is the first direct catalytic asymmetric
Michael reaction of unmodified ketone to R,â-unsaturated
carboxylic acid derivatives. On the basis of our mechanistic
studies on Et2Zn/linked-BINOL 5 catalysis, enantioselectivity
in the present reaction would mainly be determined by Re-face
shielding of zinc enolate generated from hydroxyketone 4 and
Et2Zn/linked-BINOL 5 complex.9a,34b On the other hand, the
diastereomeric ratio is supposed to be dependent on the facial
selectivity of electrophiles by the zinc catalyst. The good
diastereomeric ratio (69/31-95/5, Table 6) observed in the
present Michael reaction suggested that facial selectivity of R,â-
unsaturated N-acylpyrrole with Et2Zn/linked-BINOL complex
is similar to that of R,â-unsaturated phenyl ketone.36 These
results also supported our assumption that R,â-unsaturated
N-acylpyrrole enabled application of the catalyst, which was
optimized for R,â-unsaturated ketone, to R,â-unsaturated car-
boxylic acid derivatives. On the other hand, R,â-unsaturated
carboxylic acid imidazolide was not applicable in the Michael
reaction. Only trace, if any, product was observed under the
same reaction conditions as Table 6.
The Michael reaction proceeded smoothly at 0 °C (Table 6).
Because the reactivity of the R,â-unsaturated N-acylpyrrole was
(E) Electronic Properties of r,â-Unsaturated N-Acylpyr-
role: The high reactivity of R,â-unsaturated N-acylpyrrole can
be attributed to the delocalization of the nitrogen lone pair in
the aromatic system, reducing donation into the carbonyl group.
The reactivity of R,â-unsaturated N-acylpyrrole is thus expected
to be much higher than that of simple amide and ester. To
support the hypothesis, the LUMO energy level was calculated37
and the reaction rates in the epoxidation of various electron
deficient olefins were compared. As summarized in Figure 5,
the LUMO energy level of R,â-unsaturated N-acylpyrrole
(-2.06 eV) was lower than those of morpholinyl amide (-1.56
eV) and ester (-1.72 eV) and was rather close to that of phenyl
ketone (-2.09 eV), oxazolidinone (-2.05 eV), and methyl
(33) Et2Zn, o-MeO-phenol, t-BuOOH afforded epoxide in 76% yield after 18 h
at 25 °C, albeit in low selectivity (dr ) 54/46). Other conditions, such as
H2O2-urea with base, mCPBA, and aqNaClO with phase-transfer catalyst,
either resulted in decompositions of enone or afforded product in low yield
(<20%).
(34) For other applications of Et2Zn/linked-BINOL catalyst, see Mannich-type
reaction: (a) Matsunaga, S.; Kumagai, N.; Harada, S.; Shibasaki, M. J.
Am. Chem. Soc. 2003, 125, 4712. Aldol reaction: (b) Kumagai, N.;
Matsunaga, S.; Kinoshita, T.; Harada, S.; Okada, S.; Sakamoto, S.;
Yamaguchi, K.; Shibasaki, M. J. Am. Chem. Soc. 2003, 125, 2169. (c)
Kumagai, N.; Matsunaga, S.; Yoshikawa, N.; Ohshima, T.; Shibasaki, M.
Org. Lett. 2001, 3, 1539. (d) Yoshikawa, N.; Kumagai, N.; Matsunaga, S.;
Moll, G.; Ohshima, T.; Suzuki, T.; Shibasaki, M. J. Am. Chem. Soc. 2001,
123, 2466.
(35) (a) Zhang, F.-Y.; Corey, E. J. Org. Lett. 2000, 2, 1097. (b) Betancort, J.
M.; Sakthivel, K.; Thayumanavan, R.; Barbas, C. F., III. Tetrahedron Lett.
2001, 42, 4441. (c) List, B.; Pojarliev, P.; Martin, H. J. Org. Lett. 2001, 3,
2423. (d) Enders, D.; Seki, A. Synlett 2002, 26. (e) Andrey, O.; Alexakis,
A.; Bernardinelli, G. Org. Lett. 2003, 5, 2559. (f) Betancort, J. M.; Barbas,
C. F., III. Org. Lett. 2001, 3, 3737. (g) Alexakis, A.; Andrey, O. Org. Lett.
2002, 4, 3611. (h) Melchiorre, P.; Jørgensen, K. A. J. Org. Chem. 2003,
68, 4151.
(36) Observed diastereomeric ratio for â-aryl substituted phenyl ketone (chalcone
type) was 76/24-85/15, and for â-BOMCH2CH2 substituted phenyl ketone
was 77/23. See ref 9a.
9
J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004 7565

A R T I C L E S
Matsunaga et al.
Table 4. Sequential Wittig-Catalytic Asymmetric Epoxidation Reaction with Achiral Aldehydes
a Isolated yield. b Determined by chiral HPLC analysis. c TBHP in decane was used as oxidant. d 10 mol % of catalyst was used.
Scheme 3. Catalytic Asymmetric Epoxidation of Z-R,â-Unsaturated
N-Acylpyrrole
reactivity was necessary under the same conditions using the
same catalyst.
To precisely compare the reaction rates in the epoxidation
of R,â-unsaturated phenyl ketone 18, R,â-unsaturated methyl
ketone 19, R,â-unsaturated N-acylpyrrole 1a, R,â-unsaturated
(37) LUMO energy was calculated using the B3LYP method with 6-31G* as
basis sets. (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C. T.;
Yang, W. T.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (c) Kong, J.; White,
C. A.; Krylov, A. I.; Sherrill, C. D.; Adamson, R. D.; Furlani, T. R.; Lee,
M. S.; Lee, A. M.; Gwaltney, S. R.; Adams, T. R.; Ochsenfeld, C.; Gilbert,
A. T. B.; Kedziora, G. S.; Rassolov, V. A.; Maurice, D. R.; Nair, N.; Shao,
Y.; Besley, N. A.; Maslen, P. E.; Dombroski, J. P.; Daschel, H.; Zhang,
W.; Korambath, P. P.; Baker, J.; Byrd, E. F. C.; Van Voorhis, T.; Oumi,
M.; Hirata, S.; Hsu, C.-P.; Ishikawa, N.; Florian, J.; Warshel, A.; Johnson,
B. G.; Gill, P. M. W.; Head-Gordon, M.; Pople, J. A. J. Computational
Chem. 2000, 21, 1532. Spartan’02 Wavefunction, Inc.; Irvine, CA.
ketone (-1.88 eV). The LUMO energy level was lowest with
4-phenylimidazolide (-2.37 eV). The LUMO energy tendency
seemed closely related to the observed reactivity tendency except
4-phenylimidazolide. Because we previously utilized the La-
BINOL catalyst for 4-phenylimidazolide,23 reevaluation of its
9
7566 J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004

R,â-Unsaturated N-Acylpyrroles
A R T I C L E S
Table 5. Sequential Wittig-Catalytic Asymmetric Epoxidation
Reaction with Chiral Aldehyde
Figure 5. LUMO energy level of (a) R,â-unsaturated ketones and (b) R,â-
unsaturated carboxylic acid derivatives.
time
(h)
yielda
(%)
drb
(14/15)
entry
ligand
oxidant
1
2
3
(R)-H8-BINOL
(S)-H8-BINOL
(S)-H8-BINOL
0.7
0.9
0.7
CMHP
CMHP
TBHP
80
60
78
>99/1
1/56
1/36
a Isolated yield. b Determined by HPLC analysis.
Table 6. Direct Catalytic Asymmetric Michael Reaction of
Hydroxyketone 4 Promoted by Et2Zn/Linked-BINOL 5 Complex
Figure 6. Epoxidationreaction profile of R,â-unsaturated ketones and R,â-
unsaturated carboxylic acid derivatives with 5 mol % of Sm (O-i-Pr)3, H8-
BINOL, and Ph3As(O).
reaction profile was observed at -10 °C in THF ([substrate] )
0.05 M) using 5 mol % Sm(O-i-Pr)3, 5 mol % of H8-BINOL,
and 5 mol % of Ph3As(O). The reaction rate tendency was R,â-
unsaturated carboxylic acid imidazolide 17> R,â-unsaturated
phenyl ketone 18> R,â-unsaturated N-acylpyrrole 1a > R,â-
unsaturated methyl ketone 19 >> R,â-unsaturated amide 20
(Figure 6). The reaction profiles indicated that the reactivity of
R,â-unsaturated N-acylpyrrole is slightly better than R,â-
unsaturated methyl ketone 19 and slightly worse than phenyl
ketone 18, which was also proportional to the calculated LUMO
energy level. As expected from the LUMO energy level, the
initial reaction rate of 17 was fastest among those examined
under identical conditions, as long as 5 mol % catalyst loading
was used (vide infra). The observed high reaction rate of 17
was quite different from our previous results using the La-
BINOL catalyst for 17. The difference is ascribed to the
combination of the lanthanide metal and chiral ligand. With
BINOL as a chiral ligand, the La complex gave better results
than the Sm complex using R,â-unsaturated carboxylic acid
imidazolide.23b With the H8-BINOL ligand, the Sm complex
had the highest reactivity and selectivity.
a Isolated yield. b Determined by chiral HPLC analysis. c 15 mol % of
(S,S)-linked-BINOL and 3 equiv of ketone were used.
carboxylic acid imidazolide 17, and R,â-unsaturated amide 20,
reaction profiles with these substrates were observed under
identical conditions. Because the reaction of R,â-unsaturated
N-acylpyrrole 1a and enones 18 and 19 was expected to proceed
too fast under standard conditions at room temperature, the
9
J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004 7567

A R T I C L E S
Matsunaga et al.
Scheme 4. Catalytic Asymmetric Epoxidation Reaction of (a)
R,â-Unsaturated N-Acylpyrrole 1a and (b) N-Acylimidazole 17 with
5 and 1 Mol % Sm Catalyst
Figure 7. R,â-Unsaturated N-acylpyrrole as a monodentate R,â-unsaturated
ester surrogate.
Scheme 5. Transformations of Pyrrolyl Epoxidesa
The results in Figure 6 suggested that R,â-unsaturated
N-acylpyrrole is as reactive as R,â-unsaturated ketone. On the
other hand, the observed highest reactivity with N-acylimidazole
17 was rather unexpected on the basis of our previous experi-
ments with various lanthanide metal-BINOL catalysts.23b Thus,
we focused on the comparison of N-acylpyrrole 1a and
N-acylimidazole 17 to evaluate which is more suitable as a
monodentate R,â-unsaturated ester surrogate. Although R,â-
unsaturated carboxylic acid imidazolide was most reactive with
5 mol % catalyst loading (Figure 6), the opposite tendency was
observed with 1 mol % catalyst loading (Scheme 4). In contrast
to the case with R,â-unsaturated N-acylpyrrole, it was impossible
to reduce catalyst loading from 5 mol % using 17 as the
substrate. When the epoxidation of 17 was examined with 1
mol % of Sm(O-i-Pr)3 and 1 mol % of H8-BINOL, epoxide
was obtained in only 17% yield (Scheme 4b). On the other hand,
the reaction proceeded without any problems with 1 mol % of
catalyst loading when using R,â-unsaturated N-acylpyrrole 1a
(Scheme 4a, 91% yield, 99% ee). The opposite tendencies can
be ascribed to the difference in strength of the C-N amide bond
of the N-acylimidazole and N-acylpyrrole unit. As shown in
Scheme 4, the reaction of 17 affords epoxy peroxy ester 21
and liberated 4-phenylimidazole. On the other hand, the
N-acylpyrrole moiety remains unchanged under the reaction
conditions, probably because the leaving ability of pyrrole is
relatively poorer. The liberated free imidazole had adverse
effects on the reactivity of the Sm catalyst by coordinating to
the Sm metal. Thus, the reaction rate dropped suddenly with 1
mol % catalyst loading using N-acylimidazole. In fact, in the
presence of additional 100 mol % of 4-phenylimidazole, epoxide
was obtained in only 22% yield after 30 min at -10 °C using
5 mol % of Sm(O-i-Pr)3, H8-BINOL, and Ph3As(O) (same
conditions as Figure 6).38
a Conditions: (i) tert-butyl acetate, BuLi, THF, -78 °C, 10 min; then
DBU, CH2Cl2, 25 °C, 20 min, y. 74% (two steps); (ii) PhLi, THF, -78 °C,
10 min; then DBU, CH2Cl2, 25 °C, 20 min, y. 88% (two steps); (iii) BuLi,
1-pentyne,THF, -78 °C, 10 min; then DBU, CH2Cl2, 0 °C, 10 min, y.
84%(2 steps); (iv) LiBH4, THF, 0 to 25 °C, 1 h; then NaBH4, 25 °C, 4 h,
y. 72% (two steps); (v) LiBH4, THF, 25 °C, 5 min; then (EtO)2P(O)CH2-
CO2Et, LiCl, DBU, 25 °C, 4 h, y. 69% (two steps).
N-acylpyrrole as a monodentate R,â-unsaturated ester surrogate.
The characteristic aspects of R,â-unsaturated N-acylpyrrole as
a monodentate ester surrogate are summarized in Figure 7. The
difference in the solubility of N-acylimidazole and N-acylpyrrole
in organic solvent should be also mentioned. Because the direct
catalytic asymmetric Michael reaction with Et2Zn/linked-BINOL
5 had first-order dependency on the concentration of Michael
acceptor,9a it was difficult to utilize the R,â-unsaturated car-
boxylic acid imidazolide 17 due to its poor solubility in THF.
Michael reaction of 17 and ketone 4 gave trace, if any, product.
Although highly concentrated conditions would be required to
reduce the catalyst loading in many catalytic asymmetric
reactions, it would be impossible to improve the volumetric
productivity when using 17 for other 1,4-addition reactions.
(F) Transformation of N-Acylpyrrole Unit: As discussed
in section E, the robust C-N bond of N-acylpyrrole under the
conditions for catalytic asymmetric reactions was crucial for
achieving good catalyst turnover number in the epoxidation
reaction. On the other hand, for synthetic utility, an achiral
template should be efficiently cleaved under mild conditions.
Too robust a template is not suitable as an ester surrogate for
further functional group manipulations. To demonstrate the
utility of the N-acylpyrrole unit as the ester surrogate,39 several
transformations were performed (Scheme 5). Using the proce-
dure reported by Evans,15a reactions with carbon nucleophiles
These results indicated that both the relatively low LUMO
energy level, close to that of R,â-unsaturated ketone, and robust
property of N-acylpyrrole moiety contributed to the high
reactivity and high catalyst turnover number of R,â-unsaturated
(38) In addition to the adverse effects of the liberated 4-phenylimidazole, product
inhibition by epoxy peroxy ester is also postulated to explain the low
reactivity with 1 mol % catalyst (Scheme 4b). Epoxy peroxy ester 21 would
function as a bidentate ligand toward the Sm catalyst. Adverse effects of
the bidentate coordination were also observed when using an R,â-
unsaturated Weinreb amide as a substrate. See ref 24c.
(39) For the conversion of N-acylpyrrole unit, see refs 13 and 15 and references
therein.
9
7568 J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004

R,â-Unsaturated N-Acylpyrroles
A R T I C L E S
Scheme 6. Transformation of Michael Adducta
Scheme 8. Synthesis of Intermediate 33 in Smith’s Total Synthesis
of Phorboxazole Aa
a Conditions: (i) EtSLi, EtOH, 25 °C, 2 h, y. 96%.
Scheme 7. Preparation of Synthetic Intermediate for Antifungal
Natural Producta
a Conditions: (i) ylide 2, toluene, 85 °C, 39 h; then Sm(O-i-Pr)3 (5 mol
%), (S)-H8-BINOL (5 mol %), MS 4A, THF/toluene, CMHP, 25 °C, 0.8 h,
y. 83% (from 10n), 96% ee; (ii) CH3C(O)N(OCH3)CH3, LHMDS, THF,
-78 °C, 20 min; then DBU, CH2Cl2, 25 °C, 40 min, y. 63% (two steps).
a Conditions: (i) PhSeSePh, NaBH4, EtOH/AcOH, 25 °C, 15 min, y.
94%; (ii) EtSLi, EtOH, 25 °C, 40 min, y. 92%; (iii) CH3I, Ag2O, MS 3A,
toluene, 45 °C, 36 h, y. 93%; (iv) LiAlH4, Et2O, 25 °C for 30 min, then
reflux for 40 min, y. 85%; (v) TBSCl, imidazole, CH2Cl2, 25 °C, 50 min,
y. 86%; (vi) H2 (5 atm), Pd(OH)2, AcOEt/EtOH, NaHCO3, 25 °C, 18 h, y.
96%; (vii) PCC, AcONa, MS 3A, CH2Cl2, 25 °C, 20 min; then
(CH3O)2P(O)C(N2)COCH3, K2CO3, CH3OH, 25 °C, 80 min, y. 57% (2
steps); (viii) BuLi, THF, -78 °C; then CH3I, -78 °C to 25 °C, y. 91%;
(ix) Bu4N+F-, THF, 25 °C, 2.5 h, y. 88%.
were examined. Pyrrolyl epoxide 11a was converted to â-ke-
toester 22 in 74% yield by the addition of lithium enolate
prepared from tert-butyl acetate, followed by treatment with
DBU. R,â-Epoxy ketone 23 was obtained in 88% yield using
PhLi and DBU. The addition of lithiated alkyne gave 24 in 84%
yield after treatment with DBU. In two steps, 11m was reduced
to epoxyalcohol 25; successive treatment with LiBH4 and
NaBH4 gave 25 in 72% yield. The pyrrol carbinol intermediate
obtained from the reduction of 11r with LiBH4 was also
converted into 26 through the Horner-Wadsworth-Emmons
reaction using Masamune-Roush conditions.15b,40 The addition
of EtSLi efficiently promoted alcoholysis of N-acylpyrrole in
the Michael adduct (Scheme 6). With EtSLi in ethanol,
ethanolysis of 27 proceeded within 2 h at 25 °C to give ethyl
ester 28 in 96% yield.
We then applied the present method for the synthesis of
known intermediates in the total synthesis of natural products.
As shown in Scheme 7, a sequential Wittig reaction-catalytic
asymmetric epoxidation reaction of 10n with (S)-H8-BINOL as
the ligand afforded ent-11n in 83% yield and 96% ee in one-
pot. Then, 11n was easily converted to 29, the synthetic
intermediate for antifungal natural products reported by our
group,24b with Weinreb acetamide and LiHMDS followed by
addition of DBU.
Smith et al. reported 33 in Scheme 8 as a fragment in their
total synthesis of phorboxazole A.41 Sequential Wittig reaction-
catalytic asymmetric epoxidation of 10z afforded 14 in >99/1
dr as determined by HPLC analysis (see, Table 5). Then, 30
was obtained through reductive epoxide opening of 14 using
diselenide and NaBH4.42 EtSLi in ethanol was effective for
ethanolysis of 30. Ethanolysis of 30 proceeded smoothly at room
temperature, and ethyl ester 31 was obtained in 92% yield after
40 min. More harsh reaction conditions are required when using
metal ethoxide directly in the absence of ethanethiol.13 Simple
functional group manipulation of 31 gave 32 in four steps.
Oxidation, alkynylation of aldehyde,43 methylation of terminal
alkyne, and removal of the TBS group gave Smith’s intermediate
33.
Summary
In summary, we demonstrated the utility of R,â-unsaturated
N-acylpyrrole as a monodentate ester surrogate in the catalytic
asymmetric epoxidation reaction and direct catalytic asymmetric
Michael reaction. The results suggested that the substrate scope
of asymmetric catalysts originally developed for R,â-unsaturated
ketone could be broadened to R,â-unsaturated carboxylic acid
derivatives with only small, if any, modifications in the reaction
conditions using R,â-unsaturated N-acylpyrrole. Both a relatively
low LUMO energy level and robust pyrrole amide bond were
essential to achieve high reactivity. In the epoxidation reactions,
the following advantages are noteworthy: (i) Catalyst loading
(0.02-5 mol %) and the reaction rate (0.2-2.5 h, TOF up to
>3000 h-1) were improved compared with the previously
reported reaction with acid imidazolide and morpholinyl amide
(5-10 mol %, 1-24 h). The high volumetric productivity (up
to 3 M) with reduced catalyst loading (0.02 mol %) is also
noteworthy. (ii) The less explosive, thus, more practical CMHP
was applicable for R,â-unsaturated N-acylpyrrole. (iii) The
sequential Wittig-olefination-catalytic asymmetric epoxidation
process provided an efficient one-pot access to optically active
epoxides from aldehydes. The total efficiency of the asymmetric
epoxidation reaction was improved, especially from the view-
point of purification, including synthesis of the substrate for
the epoxidation step. Good yield (y. 72-100% from aldehyde)
and excellent ee (96f99.5%) were achieved with broad
substrate generality. In the Michael reaction, we succeeded in
the first direct catalytic asymmetric Michael reaction of unmodi-
fied ketone to R,â-unsaturated carboxylic acid derivatives in
(40) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essefeld, A. P.; Masamune, S.;
Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183.
(41) Smith, A. B., III; Verhoest, P. R.; Minbiole, K. P.; Schelhaas, M. J. Am.
Chem. Soc. 2001, 123, 4834.
(42) Miyashita, M.; Hoshino, M.; Suzuki, T.; Yoshikoshi, A. Chem. Lett. 1988,
507.
(43) Mu¨ller, S.; Liepold, B.; Roth, G. J.; Bestmann, H. J. Synlett 1996, 521.
9
J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004 7569

A R T I C L E S
Matsunaga et al.
good yield (74-97%), dr (69/31-95/5), and ee (73-95%). The
application of the N-acylpyrrole unit to other catalytic asym-
metric reactions is ongoing in our group.
JSPS and MEXT. We thank Ms. S.-R. Park for preparation of
substrates.
Supporting Information Available: Experimental procedures,
characterization of the products, detail data for reaction kinetic
profiles. This material is available free of charge via the Internet
at http://pubs.acs.org.
Acknowledgment. We thank financial support by Grant-in-
Aid for Specially Promoted Research, RFTF, and Grant-in-Aid
for Encouragements for Young Scientists (B) (for S.M.) from
JA0485917
9
7570 J. AM. CHEM. SOC. VOL. 126, NO. 24, 2004