norbornadiene for 2 days at 65 °C provided the corresponding
levorotatory product 4a in good yield but low enantioselec-
tivity (40:60 er). When the same reaction was performed
under N-oxide-promoted conditions (NMO, CH2Cl2, rt), the
reaction time increased to 17 days and the yield was only
23%; however, the product showed excellent optical purity
(2:98 er). We subsequently explored the p-tolylsulfinyl PNSO
ligand 3 in the same process (Figure 2, Scheme 2). For
reasons of stability, ligand 3 is better stored as its borane
complex.16 This is not an inconvenience since borane
removal and ligand exchange reaction can be performed in
a one-pot procedure. Complex 3a was obtained from
diphenylacetylene in an improved 83% yield. Gratifyingly,
N-oxide-promoted reaction led to 4a in good yield (71%)
and excellent enantiomeric excess (96:4 er). Again, thermal
conditions provided the PKR product in lower enantiomeric
purity (78:22 er) than the N-oxide-mediated reaction.
Most interestingly, the stereochemistry of the sulfinyl
moiety in ligands 2 and 3 determined the absolute config-
uration of the PKR adducts. Thus, RS ligand 2 provided
levorotatory cyclopentenones, while SS ligand 3 afforded
dextrorotatory ones. The absolute configuration of the
disubstituted adducts was determined by means of X-ray
crystallography of a dichloro analogue (Figure 3).17
Figure 4. Novel PNSO ligands with a range of diarylphosphine
groups. Yield in parentheses corresponds to the one-pot phosphinyla-
tion-borane protection of the corresponding N-isobutyl-p-tolyl-
sulfinamide (n-BuLi, THF, -78 °C, Ar2PCl, then BH3-SMe2 at
-30 °C).
With the novel ligands in hand, we proceeded to check
their efficiency toward the PKR of diphenylacetylene (Table
1, entries 2-5). While ligands 3, 6, and 8 afforded the
corresponding bridged complexes in excellent yield, the
PNSO ligand 7 failed to produce the desired complex 7a.
Again, it is highly probable that the steric encumbrance
caused by the ortho substituents on the phosphine moiety
obstructed the ligand-exchange reaction. The novel PNSO
ligands exhibited small differences in selectivity, all of them
giving higher than 95:5 er. Among these, compound 6,
holding a bis(p-methoxyphenyl)phosphine group, afforded
the final PKR adduct in an enhanced 97:3 enantiomeric ratio
(Table 1, entry 3). We then proceeded to explore the scope
of the reaction with respect to the nature of the alkyne
component. We found that neither electron-releasing nor
electron-withdrawing groups on the aromatic rings attached
to the alkyne have any effect on the outcome of the reaction
(Table 1, entries 6-10). Again, for both methoxy- and
fluorine-substituted substrates, ligand 6 provided a better
enantiomeric ratio than ligand 3. Alkynes with nonaromatic
substituents were also studied. Triisopropylsilyl (TIPS)-
protected 2-butyne-1,4-diol provided a good yield of the
bridged complex and an enantiomeric ratio of 93:7 (Table
1, entry 12). Finally, complex 6f derived from 4-octyne
provided the cyclization product in low yield (27%) but
excellent enantioselectivity (94:6 er).
Figure 3. Dihydro derivative of (+)-4d and its X-ray structure.
Despite a slightly lower selectivity, overall, the p-tolyl
ligand 3 was more effective than the tert-butyl analogue since
it afforded higher yields in the ligand exchange and cycload-
dition steps. At this point, in an attempt to improve
selectivity, we undertook the optimization of the phosphine
moiety in ligand 3. For this purpose, ligands 5-8 were
synthesized from the corresponding N-isobutyl-p-tolylsulfi-
namide. Reaction of lithium sulfinamide with the corre-
sponding Ar2PCl reagent followed by treatment with
BH3-SMe2 in a one-pot procedure afforded the desired
borane-protected ligands in good to excellent yield (Figure
4). In the case of ligand 7, increased steric hindrance around
the phosphorus atom prevented borane protection allowing
the isolation of 7 as a free phosphine.
Previous research on intermolecular PKR of terminal
alkynes has shown that directing olefin insertion to a specific
Co-C bond is essential to accomplish stereoselective cy-
clization.9 With this in mind, a feasible mechanistic scenario
that would account for the observed results is that in the
presence of N-oxide the PNSO ligand works mainly as a
(18) For a review on sulfoxide-metal bonding, see: Calligaris, M.
Coord. Chem. ReV. 2004, 248, 351.
(19) (a) Shambayati, S.; Crowe, W. E.; Schreiber, S. L. Tetrahedron
Lett. 1990, 31, 5289. (b) Verdaguer, X.; Moyano, A.; Perica`s, M. A.; Riera,
A.; Bernardes, V.; Greene, A. E.; Alvarez-Larena, A.; Piniella, J. F. J. Am.
Chem. Soc. 1994, 116, 2153. (c) Montenegro, E.; Poch, M.; Moyano, A.;
Perica`s, M. A.; Riera, A. Tetrahedron Lett. 1998, 39, 335. (d) Verdaguer,
X.; Vazquez, J.; Fuster, G.; Bernardes-Genisson, V.; Greene, A. E.; Moyano,
A.; Perica`s, M. A.; Riera, A. J. Org. Chem. 1998, 63, 7037.
(17) Compound (+)-4d was partially hydrogenated (H2, Pd/C, MeOH).
The dihydro derivative provided single crystals suitable for X-ray analysis.
The absolute configuration was determined by the anomalous dispersion
method.
4348
Org. Lett., Vol. 11, No. 19, 2009