Communications to the Editor
J. Am. Chem. Soc., Vol. 118, No. 39, 1996 9451
Table 2. Cyclopentenones from Enynes Containing Disubstituted
Olefins
olefin substrates is shown in entry 3. By utilizing a geo-
metrically pure cis substrate, it was found that cyclization occurs
with considerable olefin isomerization16 (entry 4). The reactions
of 1,2-disubstituted olefin substrates did not go to completion
at 18 psig of CO, but lowering the pressure to 5 psig led to
complete consumption of substrate. Presumably, competition
between CO and the 1,2-disubstituted olefin for coordination
to the titanium center gives rise to the observed pressure
dependence.
Despite the differences in substrate scope between the two
titanium-catalyzed methodologies noted above, we initially
assumed that the carbonylation proceeds Via a mechanism
similar to our previous iminocyclopentene synthesis, with CO
replacing the isocyanide (Scheme 1). However, further evidence
indicates the two systems are mechanistically distinct. Cyclo-
carbonylations employing Cp2Ti(PMe3)2 or metallacycle 217 as
the precatalyst require twice the amount of catalyst utilized by
reactions employing Cp2Ti(CO)2. This implies that the car-
bonylation of enynes using Cp2Ti(CO)2 does not proceed Via
formation of metallacycle 2. There is also a significant
stereochemical variance between the two systems when chiral
enyne substrates are cyclized (Table 1, entries 7, 8).18 If both
systems proceed through the same intermediate metallacycle,
the diastereoselectivities would be expected to be similar if both
reactions are under thermodynamic (or kinetic) control. Further
evidence against 2 as an intermediate is demonstrated in the
cyclization of allenynes. While enynes which are carbonylated
with Cp2Ti(CO)2 form metallacycles like 2 when treated with
Cp2Ti(PMe3)2, 1,5- and 1,4-allenynes show no evidence for
metallacycle formation, although they are cleanly carbonylated
by Cp2Ti(CO)2. We are currently seeking to clarify the
mechanistic discrepancies between the two catalytic reactions.
In conclusion, we have developed the first early transition
metal catalyzed carbonylative route to cyclopentenones. Besides
demonstrating a novel reactivity mode for an early transition
metal carbonyl, the Cp2Ti(CO)2 system displays an unexpectedly
increased level of functional group compatibility as compared
to other group 4 metallocene cyclizations and intriguing
mechanistic differences from the related iminocyclopentene
synthesis. Additionally, it shows a greater tolerance for
substituted alkenes than related cobalt procedures.
a Required 18 psig of CO for conversion. b Required 5 psig of CO
for complete conversion. c Unreacted substrate by GC (30%).
Recently, the Mo(CO)6-promoted conversion of 1,5-allenynes
to R-methylene cyclopentenones has been reported.12 In that
report, Co2(CO)8 failed to effect cyclization, presumably due
to competing allene polymerization. While Cp2Ti(CO)2 also
promotes the cyclization of 1,5-allenynes,10 it catalyzes the
cyclization of a 1,4-allenyne to a bicyclic dienone (entry 10).
This represents the first transition metal catalyzed cyclocar-
bonylation of an allenyne.
One of the most significant features of this catalyst system
is the ability to cyclize substrates containing substituted olefins.
The earlier Cp2Ti(PMe3)2-catalyzed isocyanide cycloconden-
sation is only effective for the cyclization of 1,1-disubstituted
olefins, and then only when 20 mol % of catalyst is used.4 Only
a single example of the successful catalytic Pauson-Khand-
type transformation of an enyne containing a 1,2-disubstituted
olefin has been reported.3a,d This substrate contains both an
R-acetoxy group capable of acting as a chelating ligand13a and
is a terminal alkyne.13b,14 Table 2 summarizes our work with
substrates containing both substituted olefins and internal
alkynes. Entry 1 shows an example of a 1,1-disubstituted olefin
substrate which is cyclized using 5 mol % of catalyst. As shown
in entry 2, tricyclic cyclopentenones are produced from the
cyclic 1,2-disubstituted olefin substrates in excellent yields.
Although stoichiometric Pauson-Khand cyclizations of this
class of substrate have been realized,15 this is the first report of
a catalytic variant of this transformation. The viability of this
catalyst for the transformation of simple acyclic 1,2-disubstituted
Supporting Information Available: Complete experimental pro-
cedures as well as spectroscopic data for all new compounds and
analytical data for most new compounds (8 pages). See any current
masthead page for ordering and Internet access instructions.
Acknowledgment. We thank the NIH (GM 34917), Pfizer, and Dow
for their support of this work. F.A.H. thanks the NSF for a predoctoral
Fellowship. N.M.K. is a National Cancer Institute Predoctoral Trainee
supported by NIH Cancer Training Grant CI T32CA09112. N.M.K.
thanks Boehringer Ingelheim, Inc., the Division of Organic Chemistry
of the ACS, and Smith-Kline Beecham for fellowships. We also
acknowledge Dr. Minghui Zhang for supplying several substrates.
(12) Kent, J. L.; Wan, H.; Brummond, K. M. Tetrahedron Lett. 1995,
36, 2407.
JA9621509
(13) (a) Enynes with pendant chelating groups undergo the Pauson-
Khand cyclization with accelerated rates and improved yields: Krafft, M.
E.; Scott, I. L.; Romero, R. H.; Feibelmann, S.; Van Pelt, C. E. J. Am.
Chem. Soc. 1993, 115, 7199. (b) The preponderance of Pauson-Khand
reactions employ substrates containing terminal alkynes (see ref 1).
(14) Examples of the stoichiometric Pauson-Khand employing an enyne
with both an internal alkyne and a 1,2-disubstituted olefin are rare (cf.,
Camps, F.; Moreto, J. M.; Ricart, S.; Vinas, J. M. Angew. Chem., Int. Ed.
Engl. 1991, 30, 1470. Almansa, C.; Carceller, E.; Garcia, E.; Serratosa, F.
Synth. Commun. 1988, 18, 1079).
(15) (a) Carceller, E.; Centellas, V.; Moyano, A.; Pericas, M. A.;
Serratosa, F. Tetrahedron Lett. 1985, 26, 2475. (b) Takano, S.; Inomata,
K.; Ogasawara, K. Chem. Lett. 1992, 443.
(16) Negishi, E.-i.; Choueiry, D.; Nguyen, T. B.; Swanson, D. R.; Suzuki,
N.; Takahashi, T. J. Am. Chem. Soc. 1994, 116, 9751.
(17) Hicks, F. A.; Berk, S. C.; Buchwald, S. L. J. Org. Chem. 1996, 61,
2713.
(18) A comparison of the diastereoselectivities [old(new)]: entry 7, 12:
1(3.5:1); entry 8, 1.6:1(8:1).