Ring-expanding olefin metathesis: A route to highly active unsymmetrical macrocyclic oligomeric co-salen catalysts for the hydrolytic kinetic resolution of epoxides

Article abstract of DOI:10.1021/ja0641406

In the presence of the third generation Grubbs catalyst, the ring-expanding olefin metathesis of a monocyclooct-4-en-1-yl functionalized salen ligand and the corresponding Co(II)(salen) complex at low monomer concentrations results in the exclusive formation of macrocyclic oligomeric structures with the salen moieties being attached in an unsymmetrical, flexible, pendent manner. The TOF-MALDI mass spectrometry reveals that the resulting macrocyclic oligomers consist predominantly of dimeric to tetrameric species, with detectable traces of higher homologues up to a decamer. Upon activation under aerobic and acidic conditions, these Co(salen) macrocycles exhibit extremely high reactivities and selectivities in the hydrolytic kinetic resolution (HKR) of a variety of racemic terminal epoxides under neat conditions with very low catalyst loadings. The excellent catalytic properties can be explained in terms of the new catalyst's appealing structural features, namely, the flexible oligomer backbone, the unsymmetrical pendent immobilization motif of the catalytic sites, and the high local concentration of Co(salen) species resulting from the macrocyclic framework. This ring-expanding olefin metathesis is suggested to be a simple way to prepare tethered metal complexes that are endowed with key features - (i) a high local concentration of metal complexes and (ii) a flexible, single point of attachment to the support - that facilitate rapid and efficient catalysis when a bimetallic transition state is required.

Full text of DOI:10.1021/ja0641406

Published on Web 01/12/2007
Ring-Expanding Olefin Metathesis: A Route to Highly Active
Unsymmetrical Macrocyclic Oligomeric Co-Salen Catalysts for
the Hydrolytic Kinetic Resolution of Epoxides
Xiaolai Zheng,† Christopher W. Jones,*,†,‡ and Marcus Weck*,†
Contribution from the School of Chemistry and Biochemistry and the School of Chemical &
Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332
Received June 12, 2006; E-mail: marcus.weck@chemistry.gatech.edu; christopher.jones@chbe.gatech.edu
Abstract: In the presence of the third generation Grubbs catalyst, the ring-expanding olefin metathesis of
a monocyclooct-4-en-1-yl functionalized salen ligand and the corresponding Co(II)(salen) complex at low
monomer concentrations results in the exclusive formation of macrocyclic oligomeric structures with the
salen moieties being attached in an unsymmetrical, flexible, pendent manner. The TOF-MALDI mass
spectrometry reveals that the resulting macrocyclic oligomers consist predominantly of dimeric to tetrameric
species, with detectable traces of higher homologues up to a decamer. Upon activation under aerobic and
acidic conditions, these Co(salen) macrocycles exhibit extremely high reactivities and selectivities in the
hydrolytic kinetic resolution (HKR) of a variety of racemic terminal epoxides under neat conditions with
very low catalyst loadings. The excellent catalytic properties can be explained in terms of the new catalyst’s
appealing structural features, namely, the flexible oligomer backbone, the unsymmetrical pendent
immobilization motif of the catalytic sites, and the high local concentration of Co(salen) species resulting
from the macrocyclic framework. This ring-expanding olefin metathesis is suggested to be a simple way to
prepare tethered metal complexes that are endowed with key featuress(i) a high local concentration of
metal complexes and (ii) a flexible, single point of attachment to the supportsthat facilitate rapid and efficient
catalysis when a bimetallic transition state is required.
the kinetic resolution6-8 have been recognized as the most
practical and eminent approaches. The discovery of the Sharpless
Introduction
Chiral epoxides are among the most versatile building blocks
for modern asymmetric organic synthesis due to the potential
for the facile stereospecific attack of the strained three-
membered ring unit by a wide spectrum of nucleophiles,
radicals, and Lewis acids with the formation of new carbon-
carbon, carbon-nitrogen, or carbon-oxygen bonds.1,2 Among
many available methods for the preparation of enantiomerically
enriched epoxides, the catalytic asymmetric epoxidation3-5 and
epoxidation reaction has provided a straightforward access to
highly enantioenriched epoxy alcohols and, thus, has been
regarded as a cornerstone in the field of asymmetric catalysis.3
The utilization of chiral Mn(III)-salen complexes4,9 and diox-
irane organo-catalysts5 have expanded the scope of the asym-
metric epoxidation to unfunctionalized olefins. As an important
complementary strategy, enzymatic and microbial kinetic reso-
lutions have been developed to yield enantioenriched glycidol
and epichlorohydrin, respectively.6,7
More recently, the hydrolytic kinetic resolution has emerged
as a general and powerful method for the generation of virtually
any terminal epoxide in an enantiopure form.8,10 Using water
as the nucleophile and chiral Co(III)-salen complexes (e.g., 1)
as the catalyst, the hydrolytic kinetic resolution (HKR) features
exceptionally high enantioselectivities, close-to-theoretical yields,
† School of Chemistry and Biochemistry.
‡ School of Chemical & Biomolecular Engineering.
(1) (a) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421-431. (b) Jacobsen,
E. N.; Wu, M. H. In ComprehensiVe Asymmetric Catalysis I-III; Jacobsen,
E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Heidelberg, 1999; Vol.
III, pp 1309-1326.
(2) (a) Pastor, I. M.; Yus, M. Curr. Org. Chem. 2005, 9, 1-29. (b) Paterson,
I.; Berrisford, D. J. Angew. Chem., Int. Ed. Engl. 1992, 31, 1179-1180.
(c) Kasai, N.; Suzuki, T.; Furukawa, Y. J. Mol. Catal. B: Enzym. 1998, 4,
237-252. (d) Besse, P.; Veschambre, H. Tetrahedron 1994, 50, 8885-
8927. (e) Smith, J. G. Synthesis 1984, 629-656. (f) Parker, R. E.; Isaacs,
N. S. Chem. ReV. 1959, 59, 737-799.
(6) Ladner, W. E.; Whitesides, G. M. J. Am. Chem. Soc. 1984, 106, 7250-
7251.
(7) Kasai, N.; Sakaguchi, K. Tetrahedron Lett. 1992, 33, 1211-1212.
(8) (a) Kunaga, M. T.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science
1997, 277, 936-938. (b) Keith, J. M.; Larrow, J. F.; Jacobsen, E. N. AdV.
Synth. Catal. 2001, 343, 5-26.
(3) (a) Sharpless, K. B.; Carl, H.; Katsuki, T.; Lee, A. W. M.; Martin, V. S.;
Takatani, M.; Viti, S. M.; Walker, F. J.; Woodard, S. S. Pure Appl. Chem.
1983, 55, 589-604. (b) Pfenninger, A. Synthesis 1986, 89-116. (c) Katsuki,
T.; Martin, V. S. Org. React. 1996, 48 1-299.
(4) (a) Zhang, W.; Leobach, J. L.; Wilson, S. R.; Jacobsen, E. N. J. Am. Chem.
Soc. 1990, 112, 2801-2803. (b) Irie, R.; Noda, K.; Ito, Y.; Matsumoto,
N.; Katsuki, T. Tetrahedron Lett. 1990, 31, 7345-7348. (c) Jacobsen,
E. N. In ComprehensiVe Organometallic Chemistry II; Wilkinsin, G., Stone,
F. G. A., Abel, E. W., Hegedus, L. S., Eds.; Pegamon: New York, 1995;
Vol. 12, pp 1097-1135. (d) Katsuki, T. Coord. Chem. ReV. 1995, 140,
189-214.
(9) For recent reviews on salen chemistry, see: (a) Canali, L.; Sherrington,
D. C. Chem. Soc. ReV. 1999, 28, 85-93. (b) Katsuki, T. AdV. Synth. Catal.
2002, 344, 131-147. (c) Larrow, J. F.; Jacobsen, E. N. Top. Organomet.
Chem. 2004, 6, 123-152. (d) McMarrigle, E. M.; Gilheany, D. G. Chem.
ReV. 2005, 105, 1563-1602.
(10) Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.; Hansen,
K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N. J. Am. Chem. Soc.
2002, 124, 1307-1315.
(5) Frohn, M.; Shi, Y. Synthesis 2000, 1979-2000.
9
10.1021/ja0641406 CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 1105-1112
1105

A R T I C L E S
Zheng et al.
substituents, respectively, on the two salicylidene rings.17 The
symmetrical approach introduces repeating salen units, in a
bigrafted manner, along the polymer or oligomer main chain.
The most successful catalytic system of this type is the cyclic
oligomeric Co(salen) complexes 2 (X ) OTs, etc.), presented
by Jacobsen and co-workers.13a,b By enforcing intramolecular
bimetallic interactions, 2 exhibited outstanding catalytic proper-
ties in a variety of epoxide ring-opening reactions. In compari-
son, the unsymmetrical approach gives salen units tethered to
the support backbone via a single grafting point. This monoteth-
ered geometry allows for a higher degree of flexibility of the
supported salen moieties in comparison to the symmetrical
approach, making the catalytic centers potentially more acces-
sible to substrates of different steric environments.
Chart 1. Monometallic and Macrocyclic Co-Salen Complexes
In our initial work, we focused on the synthesis of polymers
such as poly(styrene)s and poly(norbornene)s containing un-
symmetrically substituted Co-salen complexes in the side
chain.17,18,22 While these studies clearly demonstrated the
advantages of the approach (the poly(styrene)-based system is
among the most active and selective linear polymer supported
Co-salen catalysts in the literature17), we still needed up to 0.5
mol % catalyst loading for the HKR of common epoxides such
as epichlorohydrin. Herein we report the synthesis and the
catalytic properties of the first unsymmetrical oligomeric Co-
salen complex (4) that overcomes the need for high catalyst
loadings. Complex 4 was generated via the Ru-catalyzed,
sequential ring-opening and ring-closing olefin metatheses of
the monocyclooct-4-en-1-yl functionalized Co(salen) monomer
3 that formally expanded a low-strained eight-membered ring
to a mixture of macrocyclic oligomers. Similar phenomena of
the cyclooligomerization have been known in ring-opening
metathesis polymerization (ROMP)23-25 as well as other revers-
ible polymerization processes26 with the cyclic oligomers
usually, though not invariably,27 as undesired byproducts. In
contrast to 2 in which the Co(salen) moieties have local C2
symmetry,13 4 possesses an unprecedented unsymmetrical
macrocyclic structure with the Co(salen) fragments being
tethered in a flexible, pendent manner. Upon activation with
an appropriate acid in air, 4 exhibited exceptionally high
reactivities and enantioselectivities in the HKR of a library of
racemic terminal epoxides. Catalyst loadings as low as 0.01 mol
% for the HKR can be employed. Catalyst 4 represents the most
active unsymmetrically supported HKR catalyst.
and excellent tolerances toward various functional groups.
Detailed kinetic studies of the HKR reactions have revealed a
second-order dependence of the reaction rates on the Co-salen
species.11 This finding strongly supports a cooperative bimetallic
mechanism for the rate-determining epoxide ring-opening step.
In this context, the incorporation of Co-salen moieties into
dendritic,12 oligomeric,13,14 or polymeric frameworks15-19 not
only serves as a means for recycling the catalysts but also leads
to catalytic systems with substantially enhanced reactivities and/
or enantioselectivities in comparison with the standard mono-
metallic catalyst 1 (Chart 1) due to the close proximity of the
metal centers to each other. Catalytic reactions involving salens
and/or other complexes have been shown to operate via a
bimetallic or bimolecular mechanism.20,21 Thus, methodologies
to prepare catalysts with enhanced intramolecular interactions
of the two cooperative catalytic sites would greatly facilitate
an ever increasing variety of chemistries.
The construction of supported salen derivatives can be
realized by using either symmetrical or unsymmerical salen
monomers or precursors that contain identical or distinct
(11) Nielsen, L. P. C.; Stevenson, C. P.; Blackmond, D. G.; Jacobsen, E. N.
J. Am. Chem. Soc. 2004, 126, 1360-1362.
(12) Breinbauer, R.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2000, 39, 3604-
Results and Discussion
3607.
(13) (a) Ready, J. M.; Jacobsen, E. N. J. Am. Chem. Soc. 2001, 123, 2687-
2688. (b) Ready, J. M.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2002, 41,
1374-1377. (c) White, D. E.; Jacobsen, E. N. Tetrahedron: Asymmetry
2003, 14, 3633-3638.
1. Ring-Expanding Olefin Metathesis. (A) Synthesis of a
Monocyclooct-4-en-1-yl Functionalized, Unsymmetrical Salen
Ligand. The Co(salen) macrocycles 4 can be prepared readily
(14) Kureshy, R. I.; Singh, S.; Khan, N.-U. H.; Abri, S. H. R.; Ahmad, I.; Bhatt,
A.; Jasra, R. V. Chirality 2005, 17, 590-594.
(15) Annis, D. A.; Jacobsen, E. N. J. Am. Chem. Soc. 1999, 121, 4147-4154.
(16) (a) Kwon, M.-A.; Kim, G.-J. Catal. Today 2003, 87, 145-151. (b) Shin,
C.-K.; Kim, S.-J.; Kim, G.-J. Tetrahedron Lett. 2004, 45, 7429-7433. (c)
Song, Y.; Chen, H.; Hu, X.; Bai, C.; Zheng, Z. Tetrahedron Lett. 2003,
44, 7081-7085. (d) Song, Y.; Yao, X.; Chen, H.; Bai, C.; Hu, X.; Zheng,
Z. Tetrahedron Lett. 2002, 43, 6625-6627.
(17) Zheng, X.; Jones, C. W.; Weck, M. Chem.sEur. J. 2006, 12, 576-583.
(18) Holbach, M.; Weck, M. J. Org. Chem. 2006, 71, 1825-1836.
(19) Rossbach, B. M.; Leopold, K.; Weberskirch, R. Angew. Chem., Int. Ed.
2006, 45, 1309-1312.
(20) (a) Sammis, G. M.; Jacobsen, E. N. J. Am. Chem. Soc. 2003, 125, 4442-
4443. (b) Sammis, G. M.; Danjo, H.; Jacobsen, E. N. J. Am. Chem. Soc.
2004, 126, 9928-9929.
(22) Holbach, M.; Zheng, X.; Burd, C.; Jones, C. W.; Weck, M. J. Org. Chem.
2006, 71, 2903-2906.
(23) (23) Olefin Metathesis and Metathesis Polymerization, 2nd ed.; Ivin, J.,
Mol, I. C., Eds.; Academic: New York, 1996.
(24) (a) Thorn-Csa´nyi, E.; Ruhland, K. Macromol. Chem. Phys. 1999, 200,
1662-1671. (b) Thorn-Csa´nyi, E.; Ruhland, K. Macromol. Chem. Phys.
1999, 200, 2245-2249. (c) Thorn-Csa´nyi, E.; Ruhland, K. Macromol.
Chem. Phys. 1999, 200, 2606-2611. (d) Thorn-Csa´nyi, E.; Ruhland, K.
Macromol. Symp. 2000, 153, 145-150. (e) Ho¨cker, H.; Reimann, W.; Reif,
L.; Riebel, K. J. Mol. Catal. 1980, 8, 191-202. (f) Reif, L.; Ho¨cker, H.
Macromolecules 1984, 17, 952-956. (g) Suter, U. H.; Ho¨cker, H.
Makromol. Chem. 1988, 189, 1603-1612.
(25) Hodge, P.; Kamau, S. D. Angew. Chem., Int. Ed. 2003, 42, 2412-2414.
(26) (a) Jacobson, H.; Stockmayer, W. H. J. Chem. Phys. 1950, 18, 1600-
1606. (b) Ercolani, G.; Mandolini, L.; Mencarelli, P.; Roelens, S. J. Am.
Chem. Soc. 1993, 115, 3901-3908. (c) Hodge, P.; Colquhoun, H. M. Poly.
AdV. Technol. 2005, 16, 84-94.
(21) (a) Chen, F. X.; Qin, B.; Feng, X. M.; Zhang, G. L.; Jiang, Y. Z.
Tetrahedron 2004, 60, 10449-10460. (b) Chen, F. X.; Zhou, H.; Liu,
X. H.; Qin, B.; Feng, X. M.; Zhang, G. L.; Jiang, Y. Z. Chem.sEur. J.
2004, 10, 4790-4797. (c) Kim, S. S.; Kwak, J. M. Tetrahedron 2006, 62,
49-53. (d) Kim, S. S.; Song, D. H. Eur. J. Org. Chem. 2005, 1777-1780.
(27) Zhang, W.; Moore, J. S. Angew. Chem., Int. Ed. 2006, 45, 4416-4439.
9
1106 J. AM. CHEM. SOC. VOL. 129, NO. 5, 2007

Ring-Expanding Olefin Metathesis Generating Highly Active Co-Salen Catalysts
A R T I C L E S
Scheme 1. Synthesis of the Cyclooct-4-en-1-yl Substituted Salen
Ligand
on a multigram scale from commercially available, inexpensive
starting materials. The synthetic route starts with the generation
of a monocyclooct-4-en-1-yl functionalized, unsymmetrical salen
ligand22,28 in which the carbon-carbon double bond is prone
to the transformation of the catalytic olefin metathesis. As
shown in Scheme 1, esterification of 3-tert-butyl-2,5-
dihydroxybenzaldehyde13a and cyclooct-4-enecarboxylic acid29
in the presence of 1,3-dicyclohexylcarbodiimide (DCC) and
4-dimethylaminopyridine (DMAP) yielded salicylaldehyde 5 in
excellent yield. We have previously established a general and
straightforward method for the preparation of enantiopure
unsymmetrical salen ligands via a one-pot two-step condensation
of two different salicylaldehydes with a protected chiral di-
amine.22 This method was successfully applied to the current
synthesis. The one-pot stepwise condensation of (R,R)-diami-
nocyclohexane mono(hydrogen chloride) salt 6 with 3,5-di-tert-
butylsalicylaldehyde and 5 in a 1:1:1 molar ratio afforded the
cyclooct-4-en-1-yl substituted unsymmetrical salen ligand 7 as
a bright yellow powder in 84% yield.
(B) Ru-Catalyzed Olefin Metathesis of Monocyclooct-4-
en-1-yl Functionalized Salen Derivatives. Under an inert
atmosphere, salen ligand 7, on treatment with Co(OAc)2‚4H2O
in methanol, was converted to the corresponding Co(II)(salen)
complex 3 (Scheme 2). The desired product precipitated
gradually from the reaction media as a brick red solid and thus
could easily be separated in high yield.
Olefin metathesis reactions of these cyclooct-4-en-1-yl func-
tionalized salen compounds can be tuned through varying
monomer concentrations23-25 to afford either low molecular
weight oligomers or high molecular weight polymers. At a
monomer concentration of 0.1 M in dichloromethane, ligand 7
and complex 3 underwent a ring-opening/ring-closing sequence
in the presence of 2-4 mol % of the third generation Grubbs
catalyst (11),31 affording unsymmetrical macrocyclic oligomers
8 and 4, respectively, in almost quantitative yields. This
sequence can be regarded as an expansion of a low-strained
eight-membered ring to a mixture of oligomeric macrocycles,
referred to as the ring-expanding olefin metathesis in this
contribution. The metathesis reaction was found to be rather
Figure 1. MALDI-TOF spectra of macrocyclic oligomeric salen ligand 8
(top) and Co(salen) complex 4 (bottom) with dithranol as the matrix.
1
fast and clean as monitored by in situ H NMR spectroscopy.
For instance, consumption of monomer 7 approaches completion
in 20 min, and no side reactions were detected. This was
evidenced by an upfield shift of the multiple alkenyl proton
signals from 5.72 ppm for 7 to 5.42 ppm for 8, along with a
moderate line-broadening effect of almost all signals (see
Supporting Information).
The oligomeric nature of 8 was verified by gel-permeation
chromatography (GPC) that indicated a number-average mo-
lecular weight (Mn) of 1000 and a polydispersity index (PDI)
of 1.23. Whereas the oligomeric nature of the sample is clear
according to the GPC data, the use of linear poly(styrene) rather
than structurally analogous cyclic polymers as standards tends
to underestimate the molecular weight with a substantial error.
Grubbs and co-workers32 have found that cyclic polyenes
possess smaller hydrodynamic volumes (i.e., eluting slower) and
lower intrinsic viscosities than their linear analogues. Over a
wide range of molecular weights, they have reported that the
root-mean-square (rms) radius ( R2 0.5) possesses a correlation
g
of R2g cyclic/ R2g linear approaching 0.5 and the ratio of viscosities,
[η]cyclic/[η]linear, approximates to 0.4.32a
We deduced the cyclic structure initially from the absence
of end group signals in the NMR spectra. In addition, matrix-
assisted laser desorption ionization time-of-flight (MALDI-TOF)
mass spectrometry provided unambiguous insight into the
structures of these oligomers at a molecular level (Figure 1).
The spectrum of 4 indicated the exclusive formation of
oligomeric macrocycles as a mixture of predominantly dimeric
to tetrameric species with observable traces of higher homo-
logues up to a decamer (3: m/z ) 700, 4: m/z ) 1400 (dimer),
(28) (a) Campbell, E. J.; Nguyen, S. T. Tetrahedron Lett. 2001, 42, 1221-
1225. (b) Renehan, M. F.; Schanz, H.-J.; MvGarrigle, E. M.; Dalton,
C. T.; Daly, A. M.; Gilheany, D. G. J. Mol. Catal. A: Chem. 2005, 231,
205-220.
(29) (a) Ziegler, K.; Wilms, H. Ann. 1950, 567, 1-43. (b) Ashby, E. C.;
Coleman, D. J. Org. Chem. 1987, 52, 4554-4565.
(30) The position of the ester functionality with respect to the macrocycles 4,
8 or polymers 9, 10 could be either 1,8- or 1,9-, as a result of the head-
to-tail or head-to-head metathesis mode.
(32) (a) Bielawski, C. W.; Benitez, D.; Grubbs, R. H. Science 2002, 297, 2041-
2044. (b) Bielawski, C. W.; Benitez, D.; Grubbs, R. H. J. Am. Chem. Soc.
2003, 125, 8424-8425.
(31) Love, J. A.; Morgan, J. P.; Trnka, T. M.; Grubbs, R. H. Angew. Chem.,
Int. Ed. 2002, 41, 4035-4037.
9
J. AM. CHEM. SOC. VOL. 129, NO. 5, 2007 1107

A R T I C L E S
Zheng et al.
Scheme 2. Olefin Metathesis of Cyclooct-4-en-1-yl Functionalized Salen Monomers30
2099 (trimer), 2799 (tetramer), 3498 (pentamer), and so forth).
From the mass spectrometry data, Mn was estimated to be 2460
with a PDI of 1.3. A similar ionization pattern was observed in
the mass spectrum of oligomeric salen ligand 8 with Mn ) 2050
and PDI ) 1.3 (7: m/z ) 643, 8: m/z ) 1286 (dimer), 1929
(trimer), 2571 (tetramer), 3114 (pentamer), and so forth). Based
on these data, the average degree of oligomerization was
estimated to be 3.2-3.5. As an alternative preparative pathway,
macrocycles 4 can be prepared from the metalation of oligomeric
ligand 8 with Co(OAc)2‚4H2O in methanol. The complex
obtained via this route has a stoichiometric cobalt content
according to elemental analysis (calcd: 8.42%, found: 8.35%)
and exhibits an almost identical mass spectrum to that generated
from the olefin metathesis of 3.
corresponding polymer. Instead, the metalation of polymeric
ligand 9 with Co(OAc)2‚4H2O in methanol and dichloromethane
afforded the polymeric Co(salen) complex 10. The elemental
analysis of 10 gave a cobalt loading of 7.60% (w/w), indicating
that 90.3% of the salen centers were bound with the metal.
(C) Mechanistic Considerations of the Ring-Expanding
Olefin Metathesis. It is well-known that the ring strain and
bulkiness of the chosen cyclic monomers play a key role in
ROMP.23,33 Whereas the ROMP of highly strained, norbornene-
based monomers is essentially enthalpy-driven and often can
be performed in a controlled manner,34 the polymerization of
medium-sized, low-strain rings such as cyclooctene-based
monomers is not “living” and can result in polymers with higher
PDIs.35 The high flexibility of the main chain in the latter case
would significantly increase the possibility of the intramolecular
chain transfer with the elimination of cyclic oligomers. Since
olefin metathesis reactions are reversible in principle and the
resulting cyclic oligomers are involved in the polymerization
process, a ring-chain equilibrium can be established depending
on concentrations.33-35 It is also notable that ring-closing
metathesis (RCM) has been well documented in the literature
as a powerful method for the synthesis of macrocyclic com-
pounds with the elimination of a small olefin such as ethylene.36
To avoid undesired cross-metathesis reactions, RCM reactions
have usually been carried out in high dilution. Using the second
generation Grubbs catalyst, the ring expansion of cyclic
At a higher monomer concentration of 1.0 M in dichlo-
romethane, the metathesis reaction of salen ligand 7 was allowed
to proceed for 30 min and then quenched with ethyl vinyl ether
to give polymer 9, contaminated with ca. 5% oligomeric species,
in 88% yield after repeated precipitations from methanol. The
GPC characterization indicated a number-average molecular
weight of 33 500, corresponding to an average degree of
polymerization of 48 and a PDI of 2.84. Unfortunately, the
solubility of the Co(salen) monomer 3 is not as high in common
organic solvents, and thus, it was not used to generate the
(33) (a) Handbook of Metathesis, Vol. 3 - Application in Polymer Synthesis;
Grubbs, R. H., Ed.; Wiley-VCH: Weinheim, 2003. (b) Trnka, T. M.;
Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-29.
(34) (a) Gilliom, L. R.; Grubbs, R. H. J. Am. Chem. Soc. 1986, 108, 733-742.
(b) Kanaoka, S.; Grubbs, R. H. Macromolecules 1995, 28, 4707-4713.
(c) Weck, M.; Schwab, P.; Grubbs, R. H. Macromolecules 1996, 29, 1789-
1793. (d) Pollino, J. M.; Stubbs, L. P.; Weck, M. Macromolecules 2003,
36, 2230-2234. (e) Stubbs, L. P.; Weck, M. Chem.sEur. J. 2003, 9, 992-
999.
(35) (a) Hillmyer, M. A.; Laredo, W. R.; Grubbs, R. H. Macromolecules 1995,
28, 6311-6316. (b) Demonceau, A.; Stumpf, A. W.; Saive, E.; Noel,
A. F. Macromolecules 1997, 30, 3127-3136. (c) Higley, M. N.; Pollino,
J. M.; Hollembeak, E.; Weck, M. Chem.sEur. J. 2005, 11, 2946-2953.
(d) Hejl, A.; Scherman, O. A.; Grubbs, R. H. Macromolecules 2005, 38,
7214-7218.
9
1108 J. AM. CHEM. SOC. VOL. 129, NO. 5, 2007

Ring-Expanding Olefin Metathesis Generating Highly Active Co-Salen Catalysts
A R T I C L E S
Scheme 3. Proposed Simplified Mechanism for the Formation of Oligomeric Macrocycles during Olefin Metathesis of
Cyclooctene-Functionalized Monomers
Scheme 4. HKR of Racemic Allyl Glycidyl Ether
compounds with acyclic bis-acrylates or bis-vinyl ketones via
sequential ring-opening, cross, and ring-closing metathesis
reactions has been demonstrated.37
The ring-expanding metathesis of a medium-sized ring to
macrocyclic oligomers under low concentrations as described
here has several precedents.23,24,38 Related work includes the
whereas the eliminative cyclization is an intramolecular reaction
that is independent of the monomer. As a result, at a high
monomer concentration the metathesis reaction favors the
formation of polymers, while at a low concentration it tends to
generate macrocyclic oligomers. Somewhat similar results have
been previously reported in the Ru-catalyzed, entropically driven
ROMP of unstrained large rings (ring size g13) that also favors
the formation of macrocycles in high dilution.25 In this case,
the existence of low strain in the eight-membered ring as well
as the bulk pendent salen substituent minimized the chance of
backbiting to the monomer itself.
2. Hydrolytic Kinetic Resolution. (A) Activation of Co-
(II)(salen) Macrocycles. The oligo(cyclooctene)- and poly-
(cyclooctene)-supported Co(II)(salen) complexes were examined
as catalyst precursors for the HKR of terminal epoxides.
Throughout these catalytic studies, precatalyst 4 was used
directly as a mixture of oligomeric macrocycles.41 Two different
oxidation conditions were applied to generate the corresponding
catalytically active Co(III)(salen) species with different coun-
terions.42 Method A involved the aerobic oxidation of Co(II)-
(salen) precatalyst in the presence of an excessive amount of
acetic acid. After the mixture was stirred in dichloromethane
in the open air for 30 min, all volatiles were removed in vacuo
to afford Co(III)(salen)(OAc) as a brown solid. Method B used
1.1 equiv of p-toluenesulfonic acid in THF as the acid reagent.
The oxidation under similar conditions gave catalyst Co(III)-
(salen)(OTs) as a deep green solid.
use of an ill-defined Re2O7/Al2O3 catalyst for the dimerization
of cycloheptene and cyclooctene38 and the Schrock catalyst
inducedcyclooligomerizationofcyclooctadieneandcyclobutene.24a-d
To demonstrate how the macrocyclic oligomers are formed in
the current systems, GPC analysis was employed to monitor
the metathesis reaction of the salen monomer 7 (see Supporting
Information).39 Aliquots of the reaction mixture were taken at
designated intervals, quenched with a solution of 1% (v/v) ethyl
vinyl ether in THF (1 mL), and eluted with THF in a GPC
instrument. Over a wide range of monomer concentrations, we
have observed the existence of an equilibrium between cyclic
oligomers and polymers during the metathesis reaction. At
relatively low monomer concentrations (e.g., 0.1 M), the
polymeric species was formed in the early stage of the
metathesis and then gradually depolymerized completely over
a period of 30 min, furnishing the ring-expanding metathesis
product exclusively. At high monomer concentrations (e.g., 1.0
M), however, polymer formation predominates upon reaching
the equilibrium in coexistence with approximately 5% of
macrocyclic oligomers.
According to these findings as well as the established aspects
of the cyclooligomerization in ROMP,23-25 a simplified mech-
anism is illustrated in Scheme 3. The key lies in the competition
between two reversible processes, the chain propagation and
the eliminative cyclization (or backbiting). The chain propaga-
tion is an intermolecular reaction of first order in the monomer,40
(B) HKR of Racemic Allyl Glycidyl Ether as a Model
Reaction. The HKR of allyl glycidyl ether was chosen as a
model reaction to evaluate the catalytic properties of the new
catalysts (Scheme 4, Figure 2). The resolution was carried out
at ambient temperatures with 0.6 equiv of water and a 0.01 mol
% catalyst loading based on cobalt. Chiral GC analysis showed
(36) (a) Schrock, R. R.; Hoveyda, A. H. Angew. Chem., Int. Ed. 2003, 42, 4592-
4633. (b) Han, S.-Y.; Chang, S. In Handbook of Metathesis, Vol. 2 -
Application in Organic Synthesis; Grubbs, R. H., Ed.; Wiley-VCH:
Weinheim, 2003; pp 5-127. (c) Fu¨rstner, A. Angew. Chem., Int. Ed. 2000,
39, 3012-3034. (d) Fu¨rstner, A. Thiel, O. R.; Ackermann, L. Org. Lett.
2001, 3, 449-451. (e) Lee, C. W.; Grubbs, R. H. Org. Lett. 2000, 2, 2145-
2147.
(37) Lee, C. W.; Choi, T.-L.; Grubbs, R. H. J. Am. Chem. Soc. 2002, 124, 3224-
3225.
(38) (a) Warwel, S.; Asinger, F.; Rauenbusch, C. Angew. Chem., Int. Ed. Engl.
1987, 26, 702-703. (b) Warwel, S.; Ka¨tker, H. Synthesis 1987, 935-937.
(39) Though NMR spectroscopy has been established as the standard method
to investigate the kinetics of ROMP (for example, see: Demel, S.;
Schoefberger, W.; Slugovc, C.; Stelzer, F. J. Mol. Catal. A: Chem. 2003,
200, 11-19), it fails to distinguish between oligomer 8 and polymer 9 in
this case, which is required for estimating the rate of the backbiting
cyclization.
(41) If generated via the metallation of 8, complex 4 is almost Ru-free because
of column chromatographic workup of 8. If generated via the metathesis
of 3, complex 4 contains ca. 0.096% (w/w) residual Ru according to
elemental analysis. In the HKR reactions, the results of catalysts prepared
via the above two methods are almost identical. In control experiments
using the HKR of epichlorohydrin, addition of 1.0 mol % of third generation
Grubbs catalyst with respect to the Jacobsen catalyst 1 (X ) OAc) exhibited
negligible changes in catalytic performance.
(40) Love, J. A.; Sanford, M. S.; Day, M. W.; Grubbs, R. H. J. Am. Chem. Soc.
2003, 125, 10103-10109.
(42) For a detailed study on the effect of counterions on the HKR, see ref 11.
9
J. AM. CHEM. SOC. VOL. 129, NO. 5, 2007 1109

A R T I C L E S
Zheng et al.
Table 1. HKR of Various Racemic Terminal Epoxidesa
loadingc
(mol %)
time
(h)
eed
(%)
yielde
(%)
entry
R
methodb
1
n-Bu
A
A
B
A
B
A
B
A
B
A
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.1
2
>99
>99
>99
>99
>99
>99
>99
>99
>99
98
43
44
43
48
46
46
44
45
48
42
2a
2b
3a
3b
4af
4bf
5ag
5b
6
CH2Cl
CH2Cl
CH2OAllyl
CH2OAllyl
CH2OPh
CH2OPh
Ph
2.5
2.5
12
6
20
6
24
18
48
Ph
t-Bu
0.1
0.25
Figure 2. Kinetic plots of the HKR of racemic allyl glycidyl ether.
a Reactions were performed on 0.05-0.1 mol scales under solvent-free
conditions at rt (except for entry 4). b Method A: 4(OAc) as the catalyst;
Method B: 4(OTs) as the catalyst. c Catalyst loading based on cobalt.
d Determined by chiral GC or HPLC methods. e Isolated yield based on
racemic epoxides; theoretical maximum yield ) 50%. f Resolutions were
carried out at ambient temperature for 1 h and then heated to 40 °C. g 0.01
equiv of AcOH (based on the epoxide) was added.
that the reaction proceeded to completion in 12 h in the presence
of oligomeric 4(OAc) as the catalyst (method A). The enan-
tiomeric excess (ee) of the remaining allyl glycidyl ether was
determined to be over 99% in 51% conversion. In comparison,
the polymeric catalyst 10(OAc) gave 88% ee for the remaining
epoxide. Under the same reaction conditions, however, 3(OAc)
gave less than 1% ee, indicating that the monometallic catalyst
could hardly effect the HKR of allyl glycidyl ether at such a
low catalyst concentration. It is quite impressive that, by a simple
metathesis transformation of monometallic Co(salen) complex
3 into multimetallic species, the catalytic performance can be
improved dramatically for the HKR reactions involving a dual
activation mechanism. The more desirable catalytic performance
of the oligomeric catalyst 4(OAc) than the polymeric analogue
10(OAc) is presumably due to its macrocyclic framework that
may provide a favorable geometry for the intramolecular
bimetallic cooperative interactions. The use of less nucleophilic
tosylate as the counterion for 4 (method B) gave even better
catalytic results in the HKR of allyl glycidyl ether. Although a
short induction period was observed if 4(OTs) was employed,
the total reaction time can be reduced to 6 h to reach 99% ee
with a 51% conversion of the epoxide.
h) to reach >99% ee in this reaction.10,11 At a catalyst loading
level of 0.01 mol %, even the symmetrical cyclic oliogmers 2
(X ) OTs) need 11 h to reach >99% ee.13a
In the HKR of glycidyl phenyl ether under neat conditions,
the corresponding diol product, formed gradually during the
reaction, can precipitate from the reaction mixture and hence
hamper proper stirring. This problem was readily solved by
performing the kinetic resolution at 40 °C after the reaction
mixture was stirred at rt for 1 h. Enantiopure (S)-glycidyl phenyl
ether was obtained in 44% isolated yield in 6 h with 0.01 mol
% of 4(OTs) as the catalyst (Table 1, entry 4). The resolution
of a conjugated epoxide, styrene oxide, was complete in 18-
24 h with 0.1 mol % catalyst loading (Table 1, entry 5). Even
tert-butyloxirane, an epoxide of substantial steric hindrance, can
be resolved in 48 h with a 0.25 mol % cobalt loading (Table 1,
entry 6).
The aforementioned data clearly demonstrate that the new
cyclic oligomeric catalytic system generally possesses outstand-
ing activities and enantioselectivities in the HKR of epoxides.
As outlined in the introduction, we and others have investigated
the HKR of unsymmetrically functionalized Co-salen catalysts
before. In comparison to these known unsymmetrical Co(salen)
analogues supported on dendrimers,12 poly(styrene)s,15,17 or
poly(norbornene)s,18 the cyclic oligomeric catalyst 4 is signifi-
cantly more active. Prior to this work, the best unsymmetrically
functionalized Co-salen catalyst for the HKR was a system
reported by Weberskirch and co-workers.19 This system utilized
amphiphilic block-polymeric micellar aggregates as supports.
In comparison to these micellar aggregate-based catalysts, the
new cyclic oligomers were demonstrated at lower catalyst
loadings (routinely 0.01 mol % catalyst loadings were used here,
while the lowest catalyst loadings reported for the micellar
aggregates were 0.02 mol % with the majority of reactions
employing 0.06-0.1 mol % catalyst). It is suggested the
enhanced activities of complex 4 as evidenced by the lower
catalyst loadings and/or shortened reaction times can be
attributed to the catalyst’s unique flexible macrocyclic structure.
While the search for superior supported salen catalysts for
(C) HKR of Various Racemic Terminal Epoxides. The
macrocyclic precatalyst 4 was employed in the HKR of a variety
of structurally diverse terminal epoxides as outlined in Table
1. Since the activated catalysts showed good solubility in
common epoxides, all resolution reactions were performed in
solvent-free conditions43 that are highly desirable for organic
synthesis. Normal 1-alkene oxides are relatively easy substrates
for the HKR reactions. For instance, the resolution of 1,2-
epoxyhexane was complete in 2 h at ambient temperature with
0.01 mol % cobalt loading of 4(OAc), affording the (R)-
enantiomer of the epoxide in >99% ee and 43% isolated yield
(49% GC yield) after vacuum transfer and dehydration treatment
(Table 1, entry 1). Using the same catalyst loading, racemic
epichlorohydrin was resolved in 2.5 h to give the enatiopure
(S)-epoxide (>99% ee) in 43-44% isolated yields (Table 1,
entry 2). In comparison, the standard monometallic Co(salen)
catalyst 1 (X ) OAc, OTs) required much higher catalyst
loadings (0.2-0.5 mol %) and a prolonged reaction time (16
(43) (a) Tanaka, K.; Toda, F. Chem. ReV. 2000, 100, 1025-1074. (b) Metzger,
J. O. In Organic Synthesis Highlights V; Schmalz, H.-G., Wirth, T., Eds.;
Wiley-VCH: Weinheim, 2003; pp 82-92.
9
1110 J. AM. CHEM. SOC. VOL. 129, NO. 5, 2007

Ring-Expanding Olefin Metathesis Generating Highly Active Co-Salen Catalysts
A R T I C L E S
chemistries that utilize a bimetallic transition state (i.e., HKR,10
enantioselective conjugate additions,20 etc.) is still ongoing, the
current work pushes the catalytic performance of unsymmetrical
Co(salen) complexes to a level matching and sometimes
exceeding the best symmetrical, oligomeric counterparts.
gen chloride) (1.51 g, 10.0 mmol), activated 4 Å molecular sieve (2.0
g), and anhydrous methanol (50 mL). 3,5-Di-tert-butyl-2-hydroxyben-
zaldehyde (2.34 g, 10.0 mmol) was added in one portion and the
reaction mixture was stirred at rt for 4 h. Monitoring the reaction by
TLC showed the complete consumption of the aldehyde during this
time. A solution of cyclooct-4-enecarboxylic acid 3-tert-butyl-5-formyl-
4-hydroxyphenyl ester (3.30 g, 10.0 mmol) in anhydrous dichlo-
romethane (50 mL) was then added to the reaction system, followed
by the slow addition of triethylamine (2.7 mL, 20.0 mmol). After the
reaction mixture was stirred at rt for an additional 4 h, all solvents and
the excessive triethylamine were removed under a vacuum. The residue
was dissolved in dichloromethane (50 mL), washed with water (2 ×
50 mL), and dried with magnesium sulfate. Flash chromatography of
the crude material on silica gel (10/90 ether/hexanes) afforded
compound 7 as a bright yellow solid (5.38 g, 84%). Mp: 101-104
Concluding Remarks
In summary, we have demonstrated that the ring-expanding
olefin metathesis of the cyclooct-4-en-1-yl substituted Co(salen)
derivative 3 at relatively low monomer concentrations afforded
oligomers with a unique macrocyclic structure. The resulting
complex 4 represents the first unsymmetrical macrocyclic salen
oligomer designed for HKR reactions. Upon aerobic oxidation
under acidic conditions, 4 exhibited excellent catalytic properties
in the HKR of a variety of racemic terminal epoxides under
neat conditions with very low catalyst loadings. The HKR using
polymeric analogues (10) as catalysts resulted in lower activities,
demonstrating the superiority of the macrocyclic system. The
high reactivity and enantioselectivity of the macrocycle-based
catalytic system can be explained in terms of its appealing
structural features. The extreme flexibility of the oligomer
backbone as well as the pendent immobilization motif of the
Co(salen) moieties makes the catalytic sites highly accessible
to a diverse spectrum of substrates. Moreover, the macrocyclic
framework of the catalyst increases the local concentration of
the Co(salen) species. This may significantly promote the chance
of the intramolecular bimetallic cooperative interactions that
have played a key role in many salen complex-catalyzed
reactions.
1
°C. H NMR (CDCl3): δ 1.24 (s, 9 H, CMe3), 1.38 (s, 9 H, CMe3),
1.36-1.58 (m, 4 H, cyclooctenyl, cyclohexyl), 1.41 (s, 9 H, CMe3),
1.62-2.49 (m, 14 H, cyclooctenyl), 2.63-2.71 (m, 1 H, CHCO2), 3.32
(m, 2 H, 2 NCHCH2), 5.64-2.78 (m, 2 H, CHdCH), 6.73 (d, J ) 2.7
Hz, 1 H, Ph), 6.91 (d, 7.14, J ) 2.7 Hz, 1 H, Ph), 6.98 (d, J ) 2.8 Hz,
1 H, Ph), 7.32 (d, J ) 2.8 Hz, 1 H, Ph), 8.21 (s, 1 H, CHdN), 8.32 (s,
1 H, CHdN), 13.62 (s, br, 1 H, OH), 13.84 (s, br, 1 H, OH). 13C{1H}
NMR (CDCl3): δ 24.3, 24.5, 26.1, 26.2, 28.0, 29.3, 29.6, 29.7, 31.6,
31.7, 31.8, 33.3, 34.2, 35.0, 35.1, 43.4, 72.4, 72.6, 117.9, 118.4, 121.5,
122.9, 126.2, 127.1, 129.7, 130.9, 136.6, 138.6, 140.2, 141.9, 158.1,
158.2, 165.0 (CdN), 166.1 (CdN), 176.7 (CdO). IR (KBr): ν 1751
(CdO), 1632, 1593 cm-1. UV-vis (THF): λ 256, 331 nm. MS
(FAB+): m/z (Irel) 642 (100, M+), 587 (10, M+ - Bu + H), 506 (30,
M+ - C8H13CO + H). HRMS Calcd for C41H58N2O4: 643.4475.
Found: m/z 643.4485. Anal. Calcd for C41H58N2O4: C, 76.60; H, 9.09;
N, 4.36. Found: C, 76.70; H, 9.25; N, 4.36.
Synthesis of Co(II)(salen) Complex 3. The salen ligand 7 (3.21 g,
5.0 mmol) and cobalt(II) acetate tetrahydrate (1.50 g, 6.0 mmol) were
charged to a 100 mL Schlenk flask. After the system was thoroughly
purged with argon, degassed methanol (50 mL) was added to the system
and a red suspension formed almost immediately. The reaction mixture
was stirred at rt for 4 h. With careful exclusion of the air, the solid
was collected by filtration, washed with methanol (2 × 25 mL), and
dried under a high vacuum overnight to give 3 as a brick red solid
(3.30 g, 94%). Mp: 305-308 °C (decomp., change of color). IR: ν
1744 (CdO), 1599, 1526 cm-1. UV-vis (THF): λ 265, 370, 420 nm.
MS (ESI): m/z (Irel) 699 (100, M+). HRMS Calcd for C41H56N2O4Co:
699.3567. Found: m/z 699.3564. Anal. Calcd for C41H58N2O4Co: C,
70.37; H, 8.07; N, 4.00. Found: C, 70.46; H, 8.18; N 4.18.
Experimental Section
Cyclooct-4-enecarboxylic Acid 3-tert-Butyl-5-formyl-4-hydroxy-
phenyl Ester (5). 3-tert-Butyl-2,5-dihydroxybenzaldehyde (0.97 g, 5.0
mmol), dicyclohexyl carbodiimide (DCC, 1.03 g, 5.0 mmol), and
4-dimethylaminopyridine (DMAP, 30.5 mg, 0.25 mmol) were charged
into a 100 mL two-neck flask equipped with a condenser, an addition
funnel, and an argon outlet. The system was purged with argon, and
anhydrous dichloromethane (10 mL) was added. A solution of cyclooct-
4-enecarboxylic acid (0.75 g, 5.0 mmol) in anhydrous dichloromethane
(10 mL) was added dropwise over a period of 5 min, during which
time a fine white solid was formed gradually. The reaction mixture
was stirred at rt for 30 min and heated at reflux for 16 h. The white
solid was separated by filtration and washed with dichloromethane
(2 × 10 mL). The filtrates were combined and concentrated to give an
oily liquid. The desired product was isolated by flash column
chromatography on silica gel (10/90 to 30/70 ethyl acetate/hexane) as
a pale yellow oil (1.57 g, 95%) that solidified to a waxy solid upon
Synthesis of Oligomeric Salen Ligand 8. The salen ligand 7 (161
mg, 0.25 mmol) was dissolved in degassed dichloromethane (2 mL)
under argon. The third generation Grubbs catalyst (8.8 mg, 0.01 mmol,
2.0 mol %) in dichloromethane (0.5 mL) was added to the reaction.
After the mixture was stirred at rt for 30 min, ethyl vinyl ether (100
µL) was added to quench the reaction. The solvent was removed under
a vacuum, and the residue was purified by flash column chromatography
on silica gel (30/70 ether/hexanes) to give 8 as a yellow powder (154
1
storage in a refrigerator. H NMR (CDCl3): δ 1.40 (s, 9 H, CMe3),
1.40-1.60 (m, 2 H, cyclooctenyl), 1.66-1.82 (m, 3 H, cyclooctenyl),
2.00-2.08 (m, 1 H, cyclooctenyl), 2.15-2.25 (m, 3 H, cyclooctenyl),
2.41-2.51 (m, 1 H, cyclooctenyl), 2.69-2.77 (m, 1 H, cyclooctenyl),
5.65-2.80 (m, 2 H, CHdCH), 7.14 (d, J ) 2.8 Hz, 1 H, Ph), 7.17 (d,
J ) 2.9 Hz, 1 H, Ph), 9.81 (s, 1 H, OH), 11.70 (s, 1 H, CHO). 13C{1H}
NMR (CDCl3): δ 24.2, 26.1, 28.0, 29.2, 29.7, 31.7, 35.2, 43.4, 120.2,
123.3, 128.2, 129.6, 131.0, 140.2, 142.8, 159.0, 176.6, 196.6. IR
(HBr): ν 2743 (CHO), 1751 (CdO), 1659 (CdC) cm-1. UV-vis
(THF): λ 258, 343 nm. MS (EI): m/z (Irel) 330 (15, M+), 194 (100,
M+ - C8H13CO + H), 137 (20, C8H13CO+ - H). HRMS Calcd for
C20H26O4: 330.1831. Found: m/z 330.1829. Anal. Calcd for
C20H26O4: C, 72.70; H, 7.93. Found: C, 72.35, H, 8.10.
1
mg, 96%). H NMR (CDCl3): δ 1.27 (s, 9 H, CMe3), 1.42 (s, 9 H,
CMe3), 1.30-2.30 (m, 18 H), 1.44 (s, 9 H, CMe3), 2.58 (m, 1 H,
CHCO2), 3.33 (m, 2 H, 2 NCHCH2), 5.35-2.55 (m, br, 2 H, CHd
CH), 6.77 (“s”, 1 H, Ph), 6.92 (d, 7.14, J ) 2.7 Hz, 1 H, Ph), 7.01 (d,
J ) 2.0 Hz, 1 H, Ph), 7.34 (d, J ) 1.9 Hz, 1 H, Ph), 8.26 (s, 1 H,
CHdN), 8.33 (s, 1 H, CHdN), 13.64 (s, br, 1 H, OH), 13.91 (s, br, 1
H, OH). 13C{1H} NMR (CDCl3): δ 24.3, 26.8-27.6 (m), 29.3, 29.6,
30.2-30.8 (m), 31.6, 32.1, 32.5, 33.3, 33.6, 34.2, 35.0, 35.1, 43.3-
45.3 (m, 1 C, CHCO2, multiple chemical environments), 72.3, 72.7,
117.9, 118.4, 121.5, 122.9, 126.2, 127.1, 128.7-132.5 (m, 2 C, CHd
CH, multiple chemical environments), 136.6, 138.7, 140.2, 141.7, 158.1,
158.3, 164.9, 166.1, 175.2. IR: ν 1753 (CdO), 1628, 1593 cm-1. UV-
(R,R)-N-(3,5-Di-tert-butylsalicylidene)-N′-[3-tert-butyl-5-(cyclooct-
4′-enecarboxy)salicylidene]-1,2-cyclohexanediamine (7). A 250 mL
flask was charged with (1R,2R)-1,2-diaminocyclohexane mono(hydro-
vis (THF):
λ 258, 330 nm. MS (MALDI-TOF) Calcd for
(C41H58N2O4)n: m/z (Irel) 1286 (100%, M+, n ) 2), 1928 (65%, M+,
9
J. AM. CHEM. SOC. VOL. 129, NO. 5, 2007 1111

A R T I C L E S
Zheng et al.
n ) 3), 2571 (15%, M+, n ) 4), 3213 (4%, M+, n ) 5). Anal. Calcd
for (C41H58N2O4)n: C, 76.60; H, 9.09; N, 4.36. Found: C, 76.69; H,
9.29; N 4.25.
frit, washed with methanol (3 × 5 mL), and dried under a high vacuum
overnight to give 10 as a brick red solid (72 mg, 94%). Elemental
analysis of 10 gave a cobalt content of 7.60%, indicating 90.3% of the
salen centers were functionalized with the metal. IR: ν 1770 (CdO),
1635, 1614 cm-1. UV-vis (THF): λ 259, 344, 415 nm.
Synthesis of Oligomeric Co(II)(salen) Complex 4. Method 1 (via
the Ring-Expanding Olefin Metathesis of 3). The Co(II)(salen)
complex 3 (700 mg, 1.0 mmol) was dissolved in degassed dichlo-
romethane (9 mL) under argon resulting in a deep red solution. The
third generation Grubbs catalyst (33 mg, 0.04 mmol, 4.0 mol %) in
dichloromethane (1 mL) was added to the reaction. After the mixture
was stirred at rt for 60 min, ethyl vinyl ether (200 µL) was added to
quench the reaction. The reaction mixture was concentrated to ca. 5
mL, and degassed methanol (10 mL) was added to precipitate a red
solid. The solid was collected on a frit, washed with methanol (3 × 10
mL) under argon, and dried under a vacuum to afford 4 as a deep red
powder (650 mg, 93%). IR: ν 1747 (CdO), 1600, 1526 cm-1. UV-
vis (THF): λ 264, 370, 420 nm. MS (MALDI-TOF) Calcd for
(C41H56N2O4Co)n: m/z (Irel) 1400 (100%, M+, n ) 2), 2099 (62%, M+,
n ) 3), 2799 (24%, M+, n ) 4), 3498 (3%, M+, n ) 5). Anal. Calcd
for (C41H58N2O4Co)n: C, 70.37; H, 8.07; N, 4.00. Found: C, 69.90;
H, 8.07; N, 4.00.
General Procedures for the Hydrolytic Kinetic Resolution: (S)-
Allyl Glycidyl Ether (Table 1, Entry 3). Method A. Preoxidation of
(R,R)-4 with acetic acid (AcOH). The Co(II)(salen) precatalyst (R,R)-4
(3.5 mg, 0.005 mmol on the basis of cobalt, 0.01 mol %) was dissolved
in dichloromethane (1 mL) in a 25 mL flask. Glacial acetic acid (10
µL) was added to the solution, and the mixture was stirred in the open
air for 30 min, during which time the color changed from deep red to
dark brown. The solvent and the excess acetic acid were roughly
removed by rotary evaporation. The residue was pumped under a
vacuum (1 mbar) for 5 min to give (R,R)-4(OAc) as a dark brown solid.
The activated catalyst was dissolved in racemic allyl glycidyl ether
(5.71 g, 50 mmol) (in the cases of kinetic studies, chlorobenzene was
added as an internal reference for the GC analysis). The flask was
immersed into a water bath at ambient temperature, and deionized water
(0.54 mL, 30 mmol, 0.60 equiv) was added to the system to start the
reaction. The resolution process was monitored by chiral GC method.
After the reaction was stirred at rt for 12 h, the remaining epoxide was
vacuum-transferred (40 °C, 0.15 mbar) to a receiving flask precooled
at -78 °C. The recovered epoxide was passed through a plug of silica
gel packed in a Pasteur pipet to give (S)-allyl glycidyl ether as a clear
colorless liquid (2.73 g, 48%) in >99% ee according to chiral GC
analysis (γ-TA, 80 °C, isothermal, tR(S, major) ) 8.8 min, tR(R, minor)
) 7.7 min). Method B: Preoxidation of (R,R)-4 with p-toluenesulfonic
acid (TsOH). The Co(II)(salen) precatalyst (R,R)-4 (3.5 mg, 0.005 mmol
on the basis of cobalt, 0.01 mol %) was charged to a 25 mL flask. A
solution of p-toluenesulfonic acid monohydrate in THF (0.01 M, 0.55
mL, 0.0055 mmol, 1.1 equiv) was added, and the reaction mixture was
stirred in the open air for 30 min, during which time the color changed
from deep red to deep green. The solvent was roughly removed by
rotary evaporation. The residue was pumped under a vacuum (1 mbar)
for 5 min to give (R,R)-4(OTs) as a deep green solid. The activated
catalyst was dissolved in racemic allyl glycidyl ether (5.71 g, 50 mmol)
(in the cases of kinetic studies, chlorobenzene was added as an internal
reference for the GC analysis). The flask was immersed into a water
bath at ambient temperature, and deionized water (0.54 mL, 30 mmol,
0.60 equiv) was added to the system to start the reaction. The resolution
process was monitored by a chiral GC method. After the reaction was
stirred at rt for 6 h, the remaining epoxide was vacuum-transferred (40
°C, 0.15 mbar) to a receiving flask precooled at -78 °C. The recovered
epoxide was passed through a plug of silica gel packed in a Pasteur
pipet to give (R)-allyl glycidyl ether as a clear colorless liquid (2.62 g,
46%) in >99% ee according to chiral GC analysis.
Method 2 (via the Metalation of 8 with Cobalt(II) Acetate
Tetrahydrate). The oligomeric salen ligand 8 (130 mg, 0.20 mmol)
and cobalt(II) acetate tetrahydrate (60 mg, 0.24 mmol) were charged
to a 50 mL Schlenk flask. After the system was thoroughly purged
with argon, degassed methanol (5 mL) and dichloromethane (5 mL)
were added to the system and a red solid formed gradually. The reaction
mixture was stirred at rt for 24 h. With careful exclusion of the air,
dichloromethane was roughly removed under a vacuum. The solid was
collected by filtration on a frit, washed with methanol (2 × 10 mL),
and dried under a high vacuum overnight to give 4 as a brick red solid
(135 mg, 96%). Complex 4 prepared via Method 2 showed the same
MALDI-TOF MS pattern as that via Method 1.
Synthesis of Polymeric Salen Ligand 9. The salen ligand 7 (161
mg, 0.25 mmol) was dissolved in degassed dichloromethane (0.15 mL)
under argon. A solution of the third generation Grubbs catalyst in
dichloromethane (0.05 M, 100 µL, 0.005 mmol, 2.0 mol %) was injected
into the reaction system via a microsyringe. After the reaction mixture
was stirred at rt for 30 min, ethyl vinyl ether (100 µL) was added to
quench the reaction. Methanol (5 mL) was added to the reaction mixture
to precipitate a yellow solid. The solid was dissolved in dichloromethane
(200 µL) and reprecipitated from methanol (5 mL). The solid was
collected on a frit, washed with methanol, and dried under a vacuum
to give the polymeric salen ligand 9 (contaminated with ca. 5%
1
oligomers) as a yellow powder (141 mg, 88%). H NMR (CDCl3): δ
1.24 (s, 9 H, CMe3), 1.38 (s, 9 H, CMe3), 1.30-2.20 (m, 18 H), 1.40
(s, 9 H, CMe3), 2.58 (m, 1 H, CHCO2), 3.30 (m, 2 H, 2 NCHCH2),
5.33-2.56 (m, br, 2 H, CHdCH), 6.75 (d, J ) 2.2 Hz, 1 H, Ph), 6.89
(d, J ) 2.0 Hz, 1 H, Ph), 6.99 (d, J ) 2.2 Hz, 1 H, Ph), 7.31 (d, J )
0.5 Hz, 1 H, Ph), 8.23 (s, 1 H, CHdN), 8.30 (s, 1 H, CHdN), 13.62
(s, br, 1 H, OH), 13.87 (s, br, 1 H, OH). 13C{1H} NMR (CDCl3): δ
24.5, 27.0-27.7 (m), 29.3, 29.7, 30.75, 31.6, 32.3, 32.7, 33.3, 33.6,
34.2, 35.0, 35.1, 43.5-45.5 (m, 1 C, CHCO2, multiple chemical
environments), 72.3, 72.7, 117.9, 118.4, 121.5, 122.9, 126.2, 127.1,
128.8-132.5 (m, 2 C, CHdCH, multiple chemical environments),
136.6, 138.7, 140.2, 141.7, 158.1, 158.3, 164.9, 166.1, 175.2. IR: ν
1753 (CdO), 1631, 1595 cm-1. UV-vis (THF): λ 262 331 nm. GPC
(THF): Mn ) 33 500, PDI ) 2.84. Anal. Calcd for (C41H58N2O4)n: C,
76.60; H, 9.09; N, 4.36. Found: C, 76.10; H, 9.05; N 4.32.
Acknowledgment. This work is supported by the U.S. DOE
Office of Basic Energy Sciences through the Catalysis Science
Contract No. DE-FG02-03ER15459. M.W. gratefully acknowl-
edges a 3M Untenured Faculty Award, a DuPont Young
Professor Award, an Alfred P. Sloan Fellowship, a Camille
Dreyfus Teacher-Scholar Award, and the Blanchard Assistant
Professorship. C.W.J. acknowledges a DuPont Young Professor
Award, a Shell Young Faculty Award, and the J. Carl and Sheila
Pirkle Faculty Fellowship from ChBE at GT.
Synthesis of Polymeric Co(II)(salen) Complex 10. The polymeric
salen ligand 9 (77 mg, 0.12 mmol) and cobalt(II) acetate tetrahydrate
(36 mg, 0.144 mmol) were charged to a 50 mL Schlenk flask. After
the system was thoroughly purged with argon, degassed methanol
(2 mL) and dichloromethane (2 mL) were added to the system and a
red solid formed gradually. The reaction mixture was stirred at rt for
24 h. With careful exclusion of the air, dichloromethane was roughly
removed under a vacuum. The solid was collected by filtration on a
Supporting Information Available: Additional experimental
1
procedures as well as in situ H NMR spectra and GPC traces
of the ring expanding metathesis reaction. This material is
available free of charge via the Internet at http://pubs.acs.org.
JA0641406
9
1112 J. AM. CHEM. SOC. VOL. 129, NO. 5, 2007