Reagent-controlled regiodivergent resolution of unsymmetrical oxabicyclic alkenes using a cationic rhodium catalyst

Full text of DOI:10.1021/ja807942m

Published on Web 12/29/2008
Reagent-Controlled Regiodivergent Resolution of Unsymmetrical Oxabicyclic
Alkenes Using a Cationic Rhodium Catalyst
Robert Webster, Christian Bo¨ing, and Mark Lautens*
Department of Chemistry, UniVersity of Toronto, Toronto, Ontario, Canada, M5S 3H6
Received October 7, 2008; E-mail: mlautens@chem.utoronto.ca
Scheme 2. Reagent Controlled Ring Opening Using Cationic Rh(I)
Transition metal catalyzed reactions of oxa- and azabicyclic
alkenes have proven extremely useful in the synthesis of materials
ranging from biologically active molecules of pharmaceutical
interest1 to polymer components.2 Many highly efficient asymmetric
transformations involving these unique moieties have emerged in
the literature3 as attractive methods to allow rapid access to difficult-
to-make hydroxy-dihydronapthalene cores. In our continuing effort
to identify useful reactivity profiles in these systems, an enantio-
merically pure bridgehead substituted oxabicycle was synthesized4
and its ring opening chemistry with heteroatom nucleophiles was
explored under rhodium catalysis. From our earlier work, it is
known that a bridgehead substituted oxabenzonorbornene undergoes
Rh-catalyzed addition of methanol with complete substrate control
(Scheme 1).5
Scheme 1. Substrate Control in Rh-Catalyzed Ring Opening5
is produced with very high ee, while the other is formed in lower
selectivity. Because the reaction products have two stereocenters,
one of which is fixed in the starting material, for one of the products
to form in less than perfect ee requires that it originate from the
opposite enantiomer of the starting substrate. In general, (Scheme
4) the examples studied give product type B as virtually a single
enantiomer. This can occur when substrate “R” reacts with the chiral
Rh-ligand complex in a matched fashion and breaks the C-O bond
in agreement with substrate control. Product type A arises from
nucleophillic addition via pathway a, and pathway b (leading to
product type B) is inoperative. The “S” antipode of the starting
material interacts with the chiral catalyst in a mismatched manner,
and the inherent reactivity of the substrate participates to a minor
extent. The substrate controlled pathway a′ leads to product A with
a slight erosion of ee, while ligand control gives rise to product B
in very high enantiopurity. This model conforms to the trend that
the product formed with higher enantioselectivity generally gives
a slightly lower yield. Monitoring reaction progress over time using
chiral HPLC (entry 4, Table 1) has shown that product ee is constant
throughout the reaction, and both enantiomers of the starting
material are consumed at very similar rates.
Josiphos type ligands complexed to Rh promote ring opening
of meso oxabicyclic and azabicyclic alkenes with heteroatom
nucleophiles with very high enantioselectivities. Conventionally,
these transformations are carried out with Rh halide complexes,
and earlier work in our group has shown that in situ generated Rh
iodide complexes6 (via addition of tetrabutylammonium iodide to
the reaction) commonly give the best yields and enantioselectivites
in these types of reactions. However, when oxabicycle 2 (Scheme
2) was subjected to our standard ring opening protocol, it was found
to be essentially inert. We suspected a more electrophillic catalyst
system would be required, and the reaction was attempted using a
cationic Rh(I) triflate catalyst in the presence of the (S,R)-PPF-
PtBu2 Josiphos ligand. Previously we found this complex to be too
reactive. Inherent, efficient opening was observed, and moreover
only 9b was observed which is regioisomeric to our earlier studies
with achiral ligands. Reacting 2 with the enantiomeric ligand yielded
the regioisomer 9a implying the ligand overrides the inherent
preference of the substrate.
In light of the dominant role previously exerted by the substrate,
racemic bridgehead substituted oxabicycles were investigated
(Scheme 3) and the enantiomers were found to give regioisomeric
products implying strong catalyst control.7 The use of the cationic
Rh catalyst has proven to be essential, as replacing the counterion
by chloride or iodide does not allow access to 1b (product type
B), which arises from the unexpected mode of ring opening.
Our findings imply that a matched/mismatched effect during the
oxidative insertion of the Rh ligand complex into the C-O
bridgehead bond is operative and responsible for the observed
stereoselectivity. (Scheme 4) In all cases, one of the regioisomers
It was found that the degree of enantioselectivity results primarily
from the identity of the bridgehead substituent. The unsymmetrical
Scheme 3. Regiodivergent Resolution Resolution Using Cationic
Rh(I)
9
444 J. AM. CHEM. SOC. 2009, 131, 444–445
10.1021/ja807942m CCC: $40.75
2009 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Rh-Catalyzed Divergent Enantioselective Ring Opening
Scheme 5. Regiodivergent Inter/Intramolecular Parallel Resolution
Tethering a nucleophile to the bridgehead position (Scheme 5)
creates a situation wherein one enantiomer undergoes cyclization,
whereas the other enantiomer is unable to do so resulting in
decomposition. If an external nucleophile is added, then intermo-
lecular opening of the complimentary enantiomer of substrate is
observed.
In conclusion, we have demonstrated an unusual mode of reagent
control using cationic Rh(I) complexes to resolve racemic bridge-
head substituted oxabicyclic alkenes into pairs of regiosiomeric
products with high enantioselectivity.
Acknowledgment. We thank NSERC and the University of
Toronto for funding of this research. R.W. would like to thank
Patricia Pere`z-Galan for technical assistance. C.B. thanks the DFG
for a postdoctoral fellowship. We also would like to thank Dr. Alan
Lough and Dr. Thierry Maris (Universite´ de Montre´al) for X-ray
analysis.
a Aromatized naphthol product 1c was isolated. b Reaction performed
at 80 °C. c Reaction performed at 60 °C. d Measured using HPLC
analysis with chiral stationary phase columns; conditions are detailed in
the Supporting Information. e HPLC analysis performed after conversion
of the alcohol to the acetate. f Relative and absolute stereochemistry
assigned by single crystal X-ray analysis.
Supporting Information Available: Experimental procedures,
1
characterization data, H and 13C NMR spectra for new compounds.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
compounds bearing a methyl group (entries 1-6, Table 1) give
product type A in moderate to good ee and universally give product
type B in very high ee. Changing the bridgehead group either leads
to a reversal in which regioisomer is formed in high ee (entries 9
and 10, Table 1) or causes both type A and B products to be formed
with very high stereoselectivity (entry 13, Table 1).
(1) (a) Lautens, M.; Rovis, T. Tetrahedron 1999, 55, 8967. (b) Lautens, M.;
Fagnou, K.; Zunic, V. Org. Lett. 2002, 4, 3465. (c) Fan, E.; Shi, W.; Lowary,
T. J. Org. Chem. 2007, 72, 2917. (d) Madan, S.; Cheng, C. J. Org. Chem.
2006, 71, 8312.
(2) Rehahn, M.; Roth, M.; Seggern, H.; Scmechel, R.; Ahles, M. Ger. Offen.
2006, DE 102005029574. Wegner, S.; Muellen, K.; Macromolecules, 1993,
26, 3037.
(3) (a) Warrener, R. 1,4-Dihydro-1,4-epoxynapthalene; e-EROS; Encyclopedia
of Reagents for Organic Synthesis; Wiley-VCH: 2001. (b) Nishimura, T.;
Kawamoto, T.; Sasaki, K.; Tsurumaki, E.; Hayashi, T. J. Am. Chem. Soc.
2007, 129, 1492. (c) Cabrera, S.; Arrayas, R.; Alonso, I.; Carretero, J. J. Am.
Chem. Soc. 2005, 127, 17938. (d) Bertozzi, F.; Pineschi, M.; Macchia, F.;
Arnold, L.; Minnaard, A.; Feringa, B. Org. Lett. 2002, 4, 2703. (e) Cho, Y.;
Zunic, V.; Senboku, H.; Olsen, M.; Lautens, M. J. Am. Chem. Soc. 2006,
128, 6837.
Scheme 4. Proposed Pathways Allowing for Observed
Stereoselectivity
(4) Prepared using a method described in the Supporting Information, similar
to: Dockendorff, C.; Sahli, S.; Olsen, M.; Milhau, L.; Lautens, M. J. Am.
Chem. Soc. 2005, 127, 15028.
(5) Lautens, M.; Fagnou, K. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5455.
(6) (a) Fagnou, K.; Lautens, M. Angew. Chem., Int. Ed. 2002, 41, 26. (b) Lautens,
M.; Fagnou, K.; Hiebert, S. Acc. Chem. Res. 2003, 36, 48, and references
therein.
(7) Regiodivergent resolutions are relatively rare. Examples of regiodivergent
parallel kinetic resolutions have been reported previously: (a) Pineschi, M.;
Moro, F., Crotti, P. Di Bussolo, V.; Macchia, F. J. Org. Chem. 2004, 69,
2099, and references therein. (b) Tanaka, K.; Fu, G. J. Am. Chem. Soc. 2003,
125, 8078. (c) Jana, C.; Studer, A. Angew. Chem., Int. Ed. 2007, 46, 6542.
(d) Martin, S.; Spaller, M.; Liras, B.; Hartmann, J. Am. Chem. Soc. 1994,
116, 4493. (e) Doyle, M.; Dyatkin, A.; Kalinin, A.; Ruppar, S.; Martin, S.;
Spaller, S.; Liras, S. J. Am. Chem. Soc. 1995, 117, 11021. (f) Visser, M.;
Hoveyda, A. Tetrahedron 1995, 51, 4383. (g) Bolm, C.; Schlingloff, G.
J. Chem. Soc. Chem. Commun. 1995, 1247. (h) Chen, Y.; Deng, L. J. Am.
Chem. Soc. 2001, 123, 11302. Kagan, H. Croat. Chem. Acta 1996, 69, 669.
Vedejs, E.; Chen, X. J. Am. Chem. Soc. 1997, 119, 2584. For a review see:
Dehli, J.; Gotor, V. Chem. Soc. ReV. 2002, 31, 365.
JA807942M
9
J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009 445