An efficient catalyst for asymmetric epoxidation of terminal olefins

Full text of DOI:10.1021/ja9818699

460
J. Am. Chem. Soc. 1999, 121, 460-461
Scheme 1
An Efficient Catalyst for Asymmetric Epoxidation of
Terminal Olefins
James P. Collman,*,† Zhong Wang, Andrei Straumanis, and
Me´lanie Quelquejeu
Department of Chemistry, Stanford UniVersity
Stanford, California 94305-5080
Eric Rose*,‡
Laboratoire de Synthe`se Organique et Organome´tallique
UMR CNRS 7611, Tour 44, 4, Place Jussieu
75252 Paris Cedex 05, France
Table 1. Epoxidation of Unfunctionalized Olefins Catalyzed
with 1a
ReceiVed May 29, 1998
Prodigious effort has been devoted to the development of
catalytic enantioselective olefin epoxidation because chiral epox-
ides are appealing synthetic intermediates.1 Highly selective
asymmetric epoxidation of allylic alcohols has been accomplished
with titanium tartrate complexes.2 Mn(salen) complexes derived
from chiral C2 symmetrical 1, 2-diamines have generated good
optical selectivities for 1, 2-cis alkenes and a number of tri- and
tetrasubstituted olefins.3 However, development of a highly
selective epoxidation catalyst for terminal olefins has been far
less successful, despite the fact that the target chiral epoxides
have broad applications. Herein we present a highly efficient
catalyst based on a novel chiral iron porphyrin. This system gives
very high enantioselectivities and large turnover numbers for
styrene derivatives and nonconjugated terminal alkenes.
Studies with metalloporphyrins as oxygenating catalysts were
stimulated by the attempts to model the reactivity of the
cytochrome P-450 family of heme enzymes.4 The rigid macrocylic
core and alterable periphery of porphyrins make them attractive
templates for building asymmetric catalysts. Chiral groups have
been attached to porphyrins in many different geometries, aiming
at systems which might give high enantioselectivities and large
turnover numbers, but as yet no porphyrin-based catalyst has been
sufficiently refined to find application in general synthesis.5 The
porphyrin catalyst we present, 1, has a previously overlooked
RRââ geometry, with one pseudo-C2 axis within the porphyrin
plane.6 An important feature of this geometry is that it provides
open space for substrate access, but at the same time imposes
substantial steric bulk in the proximity of the metal center. This
feature contributes to high catalytic activity and selectivity.
a Reaction conditions: 1 (1.0 µmol), substrate (1.0 mmol), and PhIO
(0.10 mmol) react at room temperature in CH2Cl2 (2 mL). b Yields are
based on consumed PhIO. c Determined by GC with use of a Cyclo-
dex-B chiral column. d Assigned by comparing the GC retention time
with standard samples. e Reaction conditions: PhIO (ca. 1.2 mmol) is
added in 10 portions at room tempetature to a mixture of 1 (1.0 µmol)
and substrate (1.0 mmol) in CH2Cl2 (2 mL); each subsequent portion
is added when the previous batch of oxidant has been consumed.
f Isolated yield based on the substrate. g Determined by 1H NMR with
Eu(hfc)3 as a chiral shift agent. h The absolute configuration was not
determined.
† E-mail: jpc@chem.stanford.edu.
‡ E-mail: rose@ccr.jussieu.fr.
Complex 1 is synthesized from RRââ tetrakis(aminophenyl)-
porphyrin (TAPP, 2)7 (Scheme 1), in 53% overall yield in two
steps.8
(1) (a) Ojima, I. Catalytic Asymmetric Synthesis; VCH: New York, 1993.
Other catalytic approaches to obtain chiral nonracemic epoxides include
carbene transfer to carbonyl compounds and kinetic resolution of a racemic
mixture. (b) Aggarwal, V. K.; Ford, J. G.; Thompson, A.; Jones, R. V. H.;
Standen, M. J. Am. Chem. Soc. 1996, 118, 7004. (c) Tokunaga, M.; Larrow,
J. F.; Kakiuchi, F.; Jacobsen, E. N. Science 1997, 277, 936.
(2) (a) Johnson, R. A.; Sharpless, K. B. in ComprehensiVe Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: New York, 1991;
Vol. 7, p 389. (b) Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102,
5974. (c) Hanson, R. M.; Sharpless, K. B. J. Org. Chem. 1986, 51, 1922. (d)
Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune, H.; Sharpless,
K. B. J. Am. Chem. Soc. 1987, 109, 5765.
Epoxidation reactions catalyzed with 1 were first examined with
iodosylbenzene as the oxidant and excess olefin substrates (1:
PhIO:substrate ) 1:100:1000). These epoxidations show an
unprecedented selectivity pattern for a number of unfunctionalized
olefins. As shown in Table 1, this catalyst yields high enantiose-
lectivities for epoxidation of styrene (83% ee), pentafluorostyrene
(88% ee), and m-chlorostyrene (82% ee). It is significant that
this system manifests unusual chiral induction for nonconjugated
terminal olefins such as 3,3-dimethylbutene and vinyltrimethyl-
silane. The ee values obtained for these two olefins exceed the
highest values from any previously reported catalytic systems,
including the remarkable Mn(salen) derivatives.3
(3) (a) Zhang, W.; Loebach, J. L.; Wilson, S. R.; Jacobsen, E. N. J. Am.
Chem. Soc. 1990, 112, 2801. (b) Irie, R.; Noda, K.; Ito, Y.; Matsumoto, N.;
Katsuki, T. Tetrahedron Lett. 1990, 31, 7345. (c) Katsuki, T. J. Mol. Catal.
A: Chem. 1996, 113, 87.
(4) Canter, M. J.; Tarner, P. Coord. Chem. ReV. 1991, 108, 115.
(5) (a) Collman, J. P.; Zhang, X.; Lee, V. J.; Uffelman, E. S.; Brauman, J.
I. Science 1993, 261, 1404. (b) Campbell, L. A.; Kodadek, T. J. Mol. Catal.
A: Chem. 1996, 113, 293.
(6) There was only one previous example of chiral Fe porphyrin with an
RRââ geometry, which gave no chiral induction due to the flexible conforma-
tions of the chiral pickets. Groves, J. T.; Myers, R. S. J. Am. Chem. Soc.
1983, 105, 5791.
(7) Rose, E.; Cardon-Pilotaz, A.; Quelquejeu, M.; Bernard, N.; Kossanyi,
A.; Desmazie`res, B. J. Org. Chem. 1995, 60, 3919.
(8) The identity of 1 was established by MS, UV-visible spectra, and
conversion to its metal-free form.
10.1021/ja9818699 CCC: $18.00 © 1999 American Chemical Society
Published on Web 12/29/1998

Communications to the Editor
J. Am. Chem. Soc., Vol. 121, No. 2, 1999 461
Scheme 2
Figure 1. ([) Epoxidation catalyzed with 1. (9) Epoxidation catalyzed
with 1a.
Other useful features of 1 include its exceptional activity and
stability. For example, 1 catalyzes the epoxidation of styrene at
a rate of 40 turnovers/min and affords up to 5500 turnovers while
maintaining a reasonable ee of 75%. Epoxidations of some
substrates with olefin as the limiting reagent have also been tested
and the results are given in Table 1. Although the catalyst degrades
in the presence of excess oxidant, portionwise addition of
iodosylbenzene allows complete conversion of the substrates with
0.1 mol % catalyst. Styrene, m-chlorostyrene, and m-nitrostyrene
are epoxidized in good isolated yields and slightly lower but
reasonable ee values (Table 1, entries 2, 5, and 8, cf. entries 1, 4,
and 7). A typical problem with previously studied epoxidation
systems is that selectivity is gained at the expense of activity and
catalyst life, or Vice Versa; the high enantioselectivity and activity
obtained with 1 clearly show the advantage of the (open access)
+ (critical directing bulk) strategy.9 It is noteworthy that the same
strategy has been successfully used in the development of chiral
Mn(salen) catalysts.
complex A′. The larger energy gap between the two diastereo-
meric complexes translates into increased selectivity for 5.
This catalyst modification has been confirmed by comparing
mass spectra of the recovered catalyst with that of the original
complex. Because the mass of 5 coincides with its precursor, 1,
we failed in our first attempt to identify these structural changes.
However, the CD3O-analogue 1a shows loss of 3 and 6 amu after
one reaction run, corresponding to the transformation to a quinone-
like structure on one or both sides, respectively. An epoxidation
study with 1a reveals a more subtle piece of information (Figure
1): the ee-turnover plot of catalyst 1a also shows an initial rise,
but a slower one, before it reaches the 82-83% ee plateau. This
means the rate of the modification process is slower for the
deuterated complex; a similar isotope effect is observed in our
recent studies of iron porphyrin catalyzed oxidations of C6H5-
CH3 and C6D5CD3.11 We believe this, combined with the mass
spectroscopic studies, is strong evidence that 5 is the actual
catalyst giving high enantioselectivities.12
A detailed investigation of the ee vs turnover plot reveals an
unexpected trend in the ee-turnover relationship: there is a rise
in ee values during the initial period of the reaction, from 65%
at 10 turnovers to 83% at 200 turnovers (Figure 1). From 100 to
2000 turnovers, the reaction maintains a maximum ee of 82-
83%.
We believe that oxidative modification of the catalyst is
responsible for these observations. Previous studies10 and CPK
models indicate a close contact between the methoxy group on
one naphthyl lobe and the oxo-metal center in metalloporphyrins
fitted with a binaphthyl strap. We propose that in the initial stage
of the reaction, the catalyst is self-modified via oxidative
demethylation and subsequent reaction to form 5, which is a more
selective catalyst (Scheme 2). The critical geometry that governs
the enantiofacial selectivity is consistent with the observed
direction of stereochemical induction and with our hypothesis that
5 should exhibit a better enantioselectivity.
We have also studied the ligand effect on the stereoselectivity
in epoxidation of styrene, and found that using DMSO as a ligand
raises the ee to 88%, the highest reported for any catalytic system
to date.13
In conclusion, we have shown that the C2 symmetric iron
porphyrin, 1, is an efficient catalyst for asymmetric epoxidation
of terminal olefins. The high ee values and turnover numbers
obtained with some simple terminal olefins are exceptional
compared to previously reported catalytic systems, while the ready
availability of 1 makes it a potential catalyst for practical
applications.
As shown in Scheme 2, with the (R)-binaphthyl strapped
porphyrin catalyst, complex A with the larger RL group close to
the inward leaning lobe of binaphthyl is less favored. In contrast,
complex B, with RL in the space next to the outward leaning lobe,
is the favored low-energy path, which leads to the observed major
enantiomer, (S)-epoxide. The increased enantioselectivity going
from 1 to 5 can also be rationalized by the same scheme:
decreasing the steric bulk of the inward pointing methoxy would
make more room to accommodate RL in complex B′, making it
a more favored conformation, while not affecting the energy of
Acknowledgment. We thank the NSF (Grant No. CHE-9612725) and
NATO (Grant No. 960485) for financial support. We also thank the Mass
Spectrometry Facility, University of California, San Francisco, supported
by the NIH (Grant Nos. RR 04112 and RR 01614).
Supporting Information Available: Experimental procedures and
characterizations of new compounds along with the GC and NMR data
for the determination of the enantiomeric excesses of the formed epoxides
(PDF). See any current masthead page for Web access instructions.
JA9818699
(11) Collman, J. P.; Chien, A.; Eberspacher, T. Unpublished results.
(12) Epoxidation of styrene with a recovered catalyst or a catalyst
premodified by stirring with cyclopentene and PhIO does give higher ee (83%)
even at low turnover numbers.
(13) A slightly higher ee % value was obtained in the initial stages of
epoxidation of styrene catalyzed by a chiral Fe porphyrin. Groves, J. T.;
Crowley, S. J.; Shalyaev, K. V. Chirality 1998, 10, 106.
(9) A comparison has been made between 1 and a closely related Fe
porphyrin complex, amide-linked “twin-coronet” porphyrin derived from
tetrakis(2,6-diamino-4-tert-butylphenyl)porphyrin. 1 gives both a higher %
ee and a greater reaction rate. See also: Naruta, Y.; Tani, F.; Maruyama, K.
Tetrahedron Lett. 1992, 33, 6323.
(10) Naruta, Y.; Maruyama, K. Tetrahedron Lett. 1987, 28, 4553.