only product, isolated in 99% yield, was the doubly cyclized
isomeric dihydropyran 5 (Scheme 2).
and the similar stabilities of 4 and 5 justify the fact that either
can be isolated. A related acylsilane to 2-silyl-[6H]-pyran
isomerization is also known with allenyl-acylsilanes.13
The isomerization of 4 to 5 is rather facile, and even
standing neat at room temperature leads to some isomeriza-
tion. The process is accelerated somewhat by the presence
of protic acid and to a large extent by 1, possibly through a
π-allyl mechanism.14-16 We believe that the effect of water
on the reaction can be explained by proposing that water
acts as a ligand for ruthenium, thus slowing the isomerization
process without overly reducing the rate of the initial
cycloisomerization.17 This claim is supported by the fact that
other labile ligands can function similarly. For example,
DMF slows down the second isomerization as well, although,
in this case, the overall rate is also significantly reduced.
The structures of 4 and 5 were assigned on the basis of
1H NMR, 13C NMR, IR, and HRMS/EA and in analogy to
the HMBC and UV-vis of 30 and 31. The isolated
compounds were clearly not acylsilanes due to absence of
the characteristic 13C carbonyl peak at 230-240 ppm and
the presence of singlets accounting for 6 protons at 1.27 ppm
Scheme 2. Ruthenium-Catalyzed Cyclization of
Alkynyl-silyl-diyne-ols
It was discovered that the amount of water present in the
reaction mixture has a large effect on the product composi-
tion. Increasing the amount of water from 5 equiv to 10 vol
% in acetone led to isolation of a 1:1 mixture of 5 and
isomeric [6H]-pyran 4. Further increasing the percentage of
water led to increased amounts of 4. Addition of 40 vol %
water in acetone allowed isolation of 4 in 94% yield as a
17:1 mixture with 5 (Scheme 3). Higher percentages of water
1
for 4 and 1.08 ppm for 5 in the H NMR. These features
indicated that the two methyl groups in each were equivalent
and not allylic. The 13C NMR revealed four olefinic peaks
and one ether carbon resonance. The HRMS/EA data also
revealed that the mass had not changed from 2, while the
OH stretch disappeared in the IR. These data led us to
consider the isomeric compounds 4 and 5 on the basis of
the known valence isomerism of R,â,γ,δ-unsaturated ke-
tones.18 Utilization of HMBC data along with UV-vis
spectroscopy allowed us to distinguish between the two
isomers. The HMBC of related compound 31 indicated that
the single vinyl proton was 2-3 bonds away from the gem-
diester quaternary carbon and that one of the aliphatic
methylenes was 2-3 bonds away from the gem-dimethyl
quaternary carbon. The HMBC of 30 revealed the opposite
relationship.
Scheme 3. Ruthenium-Catalyzed Cyclization of
Alkynyl-silyl-diyne-ols
or use of only water led to lower yields of 4/5 because
turnovers of 1 were limited. Thus, the regioselectivity of the
cycloisomerization can be controlled by simple change of
solvent.
Additionally, the UV-vis spectra were taken to confirm
the identity of each based upon the well established differ-
ences for s-cis homoannular and s-trans heteroannular 1,3-
dienes.19-22 Dienes with an s-cis/homoannular structure are
known to have smaller extinction coefficients and λmax at
We reasoned that 5 was formed through three consecutive
isomerization reactions. The reaction involves initial cyclo-
isomerization to acylsilane 3 (Scheme 2), spontaneous 6π-
electrocyclization to 4, and ruthenium-catalyzed isomeriza-
tion to 5 (see below). Presumably, these subsequent
isomerizations proceed due to the thermodynamic equilib-
rium between the acylsilane and the 2-silyl-[6H]-pyran and
the strain of the latter relative to the isomerized dihydropyran
5. To support these claims we have calculated the relative
stabilities of the three isomers. Using the Gaussian 98 suite
of programs,11 the gas-phase free energies were computed
using the B3LYP/6-31G(d) level12 of density functional
theory (DFT). The most important term in the free energy
is the electronic energy, and more accurate calculations of
the electronic energies were performed using the improved
B3LYP/6-311G+(2d,p)//B3LYP/6-31G(d) method. Incor-
poration of this larger basis set did not significantly alter
the results, indicating that the present calculations are
sufficiently accurate. The calculations show that 5 is more
stable than 4 by 4 kcal/mol and more stable than 3 by 17
kcal/mol. The relative instability of 3 explains its absence,
(11) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.
W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian,
Inc.: Pittsburgh, PA, 1998.
(12) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(13) Takanishi, K.; Urabe, H.; Kuwajima, I. Tetrahedron Lett. 1987, 28,
2281.
(14) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 1999, 121, 9728.
(15) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 2002, 124, 5025.
(16) Trost, B. M.; Toste, F. D. Angew. Chem., Int. Ed. 2001, 40, 1114.
(17) Rate of the initial cycloisomerization is not greatly affected by the
concentration of water.
(18) Kluge, A. F.; Lillya, C. P. J. Org. Chem. 1971, 36, 1977.
4236
Org. Lett., Vol. 6, No. 23, 2004