Development of a method for the reductive cyclization of enones by a titanium catalyst

Article abstract of DOI:10.1021/ja954192n

An effective protocol in which bis(trimethylphosphine)titanocene is used to catalyze the reductive cyclization of enones to cyclopentanols via a metallacyclic intermediate has been developed. The key step in the process is the cleavage of the titanium-oxygen bond in the metallacycle by a silane to regenerate the catalyst. Mechanistic aspects of the reaction are discussed and the diastereoselectivity of the transformation is studied using both achiral and chiral substrates. The scope and limitations of the procedure are described. An in situ protocol for the generation of the air- and moisture-sensitive catalyst has also been developed. This work demonstrates, for the first time, the viability of using an early transition metal complex to catalyze the reductive cyclization of an alkene with a heteroatom-containing functional group.An effective protocol in which bis(trimethylphosphine) titanocene is used to catalyze the reductive cyclization of enones to cyclopentanols via a metallacyclic intermediate has been developed. The key step in the process is the cleavage of the titanium-oxygen bond in the metallacycle by a silane to regenerate the catalyst. Mechanistic aspects of the reaction are discussed and the diastereoselectivity of the transformation is studied using both achiral and chiral substrates. The scope and limitations of the procedure are described. An in situ protocol for the generation of the air-and moisture-sensitive catalyst has also been developed. This work demonstrates, for the first time, the viability of using an early transition metal complex to catalyze the reductive cyclization of an alkene with a heteroatom containing functional group.

Full text of DOI:10.1021/ja954192n

3182
J. Am. Chem. Soc. 1996, 118, 3182-3191
Development of a Method for the Reductive Cyclization of
Enones by a Titanium Catalyst
Natasha M. Kablaoui† and Stephen L. Buchwald*
Contribution from the Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
ReceiVed December 13, 1995X
Abstract: An effective protocol in which bis(trimethylphosphine)titanocene is used to catalyze the reductive cyclization
of enones to cyclopentanols via a metallacyclic intermediate has been developed. The key step in the process is the
cleavage of the titanium-oxygen bond in the metallacycle by a silane to regenerate the catalyst. Mechanistic aspects
of the reaction are discussed and the diastereoselectivity of the transformation is studied using both achiral and
chiral substrates. The scope and limitations of the procedure are described. An in situ protocol for the generation
of the air- and moisture-sensitive catalyst has also been developed. This work demonstrates, for the first time, the
viability of using an early transition metal complex to catalyze the reductive cyclization of an alkene with a heteroatom-
containing functional group.
Introduction
Scheme 1
Scheme 2
The early transition metal-mediated reductive cyclization of
unsaturated organic fragments encompasses a well-documented
class of reactions in which a metallacycle containing a new
carbon-carbon bond is formed (Scheme 1).1 In many cases,
processes which employ these reactions represent interesting
net organic transformations that are difficult or impossible to
achieve using traditional methods of organic synthesis. The
stoichiometric zirconocene-mediated reductive cyclizations of
diynes,2 dienes,3 and enynes4 were originally explored by Nugent
and Negishi (Scheme 2). These methodologies have been
extended to other group 4 metals5 and to the reductive
cyclization of heteroatom-containing unsaturated fragments
including the intramolecular cyclizations of hydrazone/alkenes
(or alkynes),6 enones, and ynones.7,8
procedures, which would not only decrease the amount of the
metal required but also make use of more expensive chiral
catalysts for enantioselective synthesis more practical.
Several groups have developed catalytic methods for the
reductive cyclization of dienes and enynes. The initial formation
of the metallacyclic intermediates of these reactions parallels
that seen in the stoichiometric procedures. In order to develop
a viable catalytic reaction, a number of strategies have been
employed to liberate the organic product and regenerate an active
form of the catalyst. For the catalytic reductive cyclization of
dienes, Waymouth and co-workers employ n-butylmagnesium
chloride to affect a transmetalation of the intermediate zircona-
cycle to generate the diGrignard organic product and dibutylzir-
conocene, which eliminates butane to reform the zirconocene-
butene adduct (Scheme 3).9 Mori and co-workers use a similar
strategy for the catalytic formation of heterocycles from dienes.10
In our laboratory, a catalytic technique for the reductive
cyclization of enynes to bicyclic cyclopentenones was developed
in which an isonitrile reacts with the intermediate metallacycle
(Scheme 4). Reductive elimination of the iminocyclopentene
reforms the Ti(II) catalyst.11 In each case, the intermediate
metallacycle is converted into a reactive metal complex which
either is or can be converted to the active form of the catalyst.
In order to make these methods more accessible to synthetic
organic chemists, there is a need for the development of simpler,
more efficient protocols for carrying out these reactions. A
major improvement would be the development of catalytic
† Fellow of the Organic Chemistry Division of the American Chemical
Society, sponsored by Smith-Kline Beecham. Boehringer Ingelheim
Pharmaceuticals, Inc., Fellow.
X Abstract published in AdVance ACS Abstracts, March 15, 1996.
(1) For lead references, see: (a) Buchwald, S. L.; Nielsen, R. B. Chem.
ReV 1988, 88, 1047. (b) Broene, R. D.; Buchwald, S. L. Science 1993,
261, 1696. (c) Broene, R. D. In ComprehensiVe Organometallic Chemistry
II; Hegedus, L. S., Ed.; Pergamon: Oxford, 1995; Vol. 12, Chapter 3.7, p
323. (d) Buchwald, S. L.; Broene, R. D. In ComprehensiVe Organometallic
Chemistry II; Hegedus, L. S., Ed.; Pergamon: Oxford, 1995; Vol. 12,
Chapter 7.4, p 771 and references within.
(2) (a) Nugent, W. A.; Calabrese, J. C. J. Am. Chem. Soc. 1984, 106,
6422. (b) Nugent, W. A.; Thorn, D. L.; Harlow, R. L. J. Am. Chem. Soc.
1987, 109, 2788.
(3) Nugent, W. A.; Taber, D. F. J. Am. Chem. Soc. 1989, 111, 6435.
(4) (a) Rajanbabu, T. V.; Nugent, W. A.; Taber, D. F.; Fagan, P. J. J.
Am. Chem. Soc. 1988, 110, 7128. (b) Negishi, E.-I.; Holmes, S. J.; Tour, J.
M.; Miller, J. A.; Cederbaum, F. E.; Swanson, D. R.; Takahashi, T. J. Am.
Chem. Soc. 1989, 111, 3336. (c) For a review of this chemistry, see:
Negishi, E.-I. In ComprehensiVe Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon: Oxford, 1991; Vol. 5, Chapter 9.5, p 1163.
(5) (a) Grossman, R. B.; Buchwald, S. L. J. Org. Chem. 1992, 57, 5803.
(b) Urabe, H.; Hata,T.; Sato, F. Tetrahedron Lett. 1995, 36, 4261.
(6) Jensen, M.; Livinghouse, T J. Am. Chem. Soc. 1989, 111, 4495.
(7) Hewlett, D. F.; Whitby, R. J. J. Chem. Soc., Chem. Commun. 1990,
1684.
(9) (a) Knight, K. S.; Waymouth, R. M. J. Am. Chem. Soc. 1991, 113,
6268. (b) Knight, K. S.; Wang, D.; Waymouth, R. M.; Ziller, J. J. Am.
Chem. Soc. 1994, 116, 1845. For a similar catalytic process, see: (c) Knight,
K. S.; Waymouth, R. M. Organometallics 1994, 13, 2575.
(10) Uesaka, N.; Mori, M.; Okamura, K.; Date, T. J. Org. Chem. 1994,
59, 4542.
(8) For stoichiometric enone cyclizations using a late transition metal
see: Bryan, J. C.; Arterburn, J. B.; Cook, G. K.; Mayer, J. M. Organome-
tallics 1992, 11, 3965.
(11) Berk, S. C.; Grossman, R. B.; Buchwald, S. L. J. Am. Chem. Soc.
1993, 115, 4912; 1994, 116, 8593.
0002-7863/96/1518-3182$12.00/0 © 1996 American Chemical Society

ReductiVe Cyclization of Enones by a Titanium Catalyst
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3183
Scheme 3
Scheme 6
the same key reactions; the initial work from both laboratories
was recently communicated.15
Scheme 4
Results and Discussion
In initial experiments, the substrate, diphenylsilane, and 10
mol % 1 were combined in toluene under anhydrous conditions.
While this protocol was found to work extremely well for
aldehyde-containing substrates (Table 1, entry 4), the cyclization
of ketone-containing substrates under these conditions resulted
in the formation of a mixture of products (Scheme 6). In
addition to the desired silylated cis-cyclopentanol 4, a small
quantity of the trans isomer 6 and a large amount of acyclic
silyl ether 7, resulting from the simple reduction of the carbonyl,
were also produced. We first set out to determine the cause of
the carbonyl reduction in an attempt to eliminate side product
7.
When the reaction is monitored by gas chromatography,
initially only cyclized isomers 4 and 6 are observed. As the
reaction progresses, the relative amount of 7 is observed to
increase rapidly. This suggests that the species responsible for
the carbonyl reduction is different than the catalyst for the
reductive cyclization; presumably, this species is generated by
decomposition of the cyclization catalyst. We surmise that over
the course of the reaction, some of the Ti(II) complex which
acts as the cyclization catalyst is converted to a Ti(III) hydride,
which merely reduces the carbonyl.17 Other experimental
observations support this hypothesis. For example, the addition
of excess trimethylphosphine to the reaction mixture reduces
the amount of 7 produced, possibly by stabilizing the Ti(II)
complex. Additionally, running the reaction at lower temper-
atures also decreases the amount of 7 which is observed (see
Table 2).
Experimentally it was found that 10 mol % 1, 60 mol %
trimethylphosphine, and 1.0 equiv of diphenylsiliane at -20
°C will convert an enone to the cyclopentanol silyl ethers 4
and 6, while virtually eliminating the production of the acyclic
silyl ether 7. The results of the reaction under these conditions
are shown in Table 1. It should be noted that under conditions
employing excess PMe3, reduction is the main pathway for
aldehyde substrates. This contrasting behavior can be explained
if the reduction can also occur through a second pathway as
shown in Scheme 7. Here, the silane cleaves the titanium-
oxygen bond in titanocene-carbonyl complexes B or C,
followed by reductive elimination of the acyclic reduced product.
The addition of excess trimethylphosphine will favor the
formation of the PMe3 adducts of B and C. In both instances,
as is shown in Scheme 7a, two geometric isomers of the adducts
Scheme 5
Whitby and Hewlett have demonstrated that a stoichiometric
quantity of Cp2Ti(PMe3)2, 1, can be used to convert 1,6-enones
to oxatitanabicyclopentanes in good yields.7 We became
interested in developing a catalytic variant of this process.
Cleaving the exceptionally strong titanium-oxygen bond in
these metallacycles in such a way as to lead to the regeneration
of a catalytically active species posed a significant challenge.
Titanium-oxygen bond energies are approximately 104 kcal/
mol and are significantly stronger than the 40 to 50 kcal/mol
strength of the titanium- and zirconium-carbon bonds which
are broken in the catalytic reactions mentioned earlier.12 We
have previously found that silanes will readily cleave titanium-
oxygen bonds with concomitant formation of Ti-H and Si-O
bonds13 via a σ-bond metathesis process. Using this key
reaction, we envisioned the catalytic process for the conversion
of enones to cyclopentanols shown in Scheme 5. Formation
of titanacycle 2 proceeds by insertion of the coordinated olefin
into the Ti-C bond of the nascent titanocene-ketone complex
A. The silane then cleaves the Ti-O bond to form the
titanocene alkyl hydride 3, which undergoes ligand-induced
reductive elimination14 to afford the silyl-protected cyclopen-
tanol 4, while regenerating the catalyst. Hydrolysis of the silyl
ether produces the cyclopentanol 5. We note that Crowe and
Rachita independently developed a similar protocol based on
(15) (a) Kablaoui, N. M.; Buchwald, S. L. J. Am. Chem. Soc. 1995, 117,
6785. (b) Crowe, W. E.; Rachita, M. J. J. Am. Chem. Soc. 1995, 117, 6787.
(16) Jencks, W. P. Acc. Chem. Res. 1980, 13, 161.
(17) (a) Aitken, C. T.; Harrod, J. F.; Samuel, E. J. Am. Chem. Soc. 1986,
108, 4059. (b) Harrod, J. F.; Yun, S. S. Organometallics 1987, 6, 1381. (c)
Harrod, J. F. In Inorganic and Organometallic Polymers; Zeldin, M., Wayne,
K. J., Allcock, H. R., Eds.; ACS Symposium Series 360; American Chemical
Society: Washington, DC, 1988; Chapter 7. Titanium hydrides have been
previously postulated as carbonyl reduction catalysts in ref 13 and in: (d)
Nakamo, T.; Nagai, Y. Chem. Lett. 1988, 481.
(12) Connor, J. A. Top. Curr. Chem. 1977, 71, 71.
(13) Berk, S. C.; Kreutzer, K. A.; Buchwald, S. L. J. Am. Chem. Soc.
1991, 113, 5093.
(14) Gell, K. I.; Schwartz, J. J. Am. Chem. Soc. 1981, 103, 2687.

3184 J. Am. Chem. Soc., Vol. 118, No. 13, 1996
Kablaoui and Buchwald

ReductiVe Cyclization of Enones by a Titanium Catalyst
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3185
Scheme 7
(a)
(b)
(c)
Table 2
Scheme 8
fused metallacycle, are also detected in most cases (Scheme
8). In the stoichiometric reaction, the steps leading to metal-
lacycle formation are reversible, so only the thermodynamically
favored product is formed. During the catalytic process, the
organic fragment can be cleaved from the intermediate titanium
complex by the silane before complete equilibration to the
thermodynamic product can occur, and a mixture of isomers is
produced.18 For substrates containing a heteroatom in the
backbone (Table 1, entries 8 and 9), little or no diastereoselec-
tivity is observed. The reactions of these substrates showed
no kinetic selectivity when transformed to the metallacycles
stoichiometrically but equilibrated to the more stable cis-isomer
over several days (Scheme 9).19,20 Hence, the catalytic reactions
employing these substrates would not be expected to be
diastereoselective
are possible. We speculate that only the isomer with the PMe3
group coordinated anti to the oxygen can undergo the σ-bond
metathesis reaction. A possible explanation for this is that the
silane precoordinates with the complex syn to the oxygen before
σ-bond metathesis occurs (see Scheme 7b).16 For enal com-
plexes, both B1 and B2 are viable and the reaction may proceed
via a σ-bond metathesis process through B2, leading to side
product 7a. For enones, C2 is destabilized for steric reasons
(see Scheme 7c) and the quantity of 7b produced is minimal.
Diastereoselectivity of CyclizationsEffect of Substituents.
In addition to exploring the mechanism of this cyclization
reaction, we have also investigated the cyclization of a wide
range of substrates in order to study the diastereoselectivity of
the transformation. Two aspects of the diastereoselectivity will
be discussed: the formation of new chiral centers at the ring
junction of the intermediate metallacycles, and the effect of
preexisting chiral centers on the diastereoselectivity of the
cyclization.
The interplay of kinetic and thermodynamic factors also
contributes to the slight decrease in diastereoselectivity for the
cyclization of enones that are disubstituted in the â-position
Whitby found that the stoichiometric reaction of achiral
substrates is completely diastereoselective; only the thermody-
namically favored cis-fused metallacycle is formed.7 The
diastereoselectivity of the catalytic reaction for achiral substrates,
while not complete, is generally very good (Table 1, entries
1-4 and 10). Though cis-cyclopentanol 5, which is formed
via a cis-fused metallacycle, is the major product formed in the
catalytic reaction, small amounts of isomer 8, formed via a trans-
(18) Poor diastereoselectivity in catalytic transformations compared to
the corresponding stoichiometric processes has previously been observed,
see ref 10.
(19) A similar effect upon heteroatom substitution has been noted: (a)
Davis, J. M.; Whitby, R. J.; Jaxa-Chamiec, A. Tetrahedron Lett. 1992, 33,
5655. (b) Taber, D. F.; Louey, J. P.; Wang, Y.; Nugent, W. A.; Dixon, D.
A.; Harlow, R. L. J. Am. Chem. Soc. 1994, 116, 9457.
(20) The equilibration of these substrates to the cis isomer was monitored
using 1H NMR verses an internal standard, and the stereochemistry was
determined by nOe analysis of the resulting silyl ether of entry 9, Table 1.

3186 J. Am. Chem. Soc., Vol. 118, No. 13, 1996
Kablaoui and Buchwald
Scheme 9
Scheme 12
Scheme 10
Scheme 11
Scheme 13
(Table 1, entries 5 and 6). Scheme 10 shows that the “chair-
like” intermediate D which leads to the cis-fused metallacycle
contains a destabilizing pseudo-1,3-diaxial interaction between
the substituent on the â-position and the methyl group. This
decreases the energy difference between D and the “boat-like”
intermediate E, leading to increased formation of the trans-
product. For an aldehyde that is disubstituted in the â-position
(Table 1, entry 4), there is a significantly less severe pseudo-
1,3-diaxial interaction in intermediate F, and excellent selectivity
is observed.
The acetophenone derivative (Table 1, entry 10) is interesting
in that it is cyclized by this protocol very selectively to give
only one observable silyl ether product. Depending on the
method of workup employed, either the bicyclic cyclopentanol
or the dimethylindene can be produced in good yield.
Chiral substrates with single substituents on the backbone
(Table 1, entries 7, 11, 12, and 13) can form four isomers.
Substrates with substituents R, â, and γ to the carbonyl were
studied. In these cases, modest to excellent diastereoselectivity
is observed.
Substrates with a single substituent in the position R to the
carbonyl (Table 1, entries 11 and 12) proceed with good to
excellent diasteroselectivity, and one isomeric product is
primarily observed. As shown in Scheme 11, the substituent
is placed preferentially in the equatorial position of the chair-
like intermediate. Cleavage of the organic fragment from the
resulting metallacycle followed by hydrodesilylation provides
the observed cyclopentanol. If the R substituent is an ethyl ester
(Table 1, entry 11), cyclization occurs with a 90:1 diastereo-
selectiviy; if it is a benzyl group (Table 1, entry 12), the
diastereoselectivity is 20:1. Crowe and Rachita reported the
cyclization of an aldehyde with a methyl group R to the carbonyl
proceeded with the formation of two diastereomers with a 4:1
selectivity.15b
Scheme 12 shows that in the reaction of 4-phenyl-6-hepten-
2-one (Table 1, entry 7), which has a phenyl group in the
â-position, two of the four possible isomers are formed
(assignment by nOe analysis). Additionally, as shown in Table
2, the ratio of the isomers which are observed is temperature
dependent. A possible explanation for these findings is that
the titanocene moiety can initially bind to either of the two
diastereotopic faces of the carbonyl. If the titanium binds to
the si face, reaction via the chair-like intermediate G is favored,
and the major isomer 10 is formed. If the titanium binds to the
re face, reaction via the boat-like intermediate H is favored,
and the minor isomer 11 is formed. The temperature depen-
dence of the isomeric ratio may result from the sensitivity of
the rate of the equilibration between the re and si faces on
temperature. At higher temperatures, the equilibration between
the diastereomeric complexes is fast, so that the proportion of
the isomer formed via the thermodynamically favored interme-
diate increases.21 At lower temperatures, the organic fragment
is cleaved from the metal before equilibration is complete, so
that a mixture of isomers is produced.
Cyclizations of substrates which have γ substituents pro-
ceeded with very low levels of diastereoselectivity. Several
substrates were cyclized using a stiochiometric quantity of 1;
in each case an ca. 1:1 mixture of cis-metallacycles was
observed. For substrates with smaller γ substituents, such as
OAc and OBn, the product with the substituent on the exo-face
was slightly favored. When the substituent was the larger
triisopropylsilyl group, the metallacycle with the endo substitu-
ent was slightly favored (Scheme 13). The benzyl-substituted
substrate (Table 1, entry 13) was cyclized under catalytic
conditions, and at room temperature two products were produced
which corresponded to the intermediate cis-metallacycles (Scheme
14). When the reaction was run at -20 °C, a third isomer could
be identified as shown in Scheme 14.
Scope and Limitations. The scope of this cyclization
methodology was explored, and it was found that these reactions
are in general very sensitive to the steric bulk of the substrates.
While methyl and n-propyl ketones (Table 1, entries 1 and 2)
cyclized smoothly under the conditions outlined above, more
sterically congested ketones such as entry 3 required a smaller
silane for catalysis. However, small silanes such as methyl-
phenylsilane also reacted faster with the catalyst to form Ti-
(III) complexes,17a resulting in an increase in the production of
acyclic reduction products. Enones with isopropyl and phenyl
(21) It is assumed that the rate of the σ-bond metathesis is relatively
insensitive to temperature changes.

ReductiVe Cyclization of Enones by a Titanium Catalyst
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3187
Table 3
Scheme 14
utilizing the inexpensive, air- and moisture-stable Cp2TiCl2 was
developed. Whitby originally synthesized the oxatitanacycles
from Cp2Ti(PMe3)2 formed in situ from Cp2TiCl2 and PMe3.
We found that a variation of this method, which is experimen-
tally less complicated, forms a viable catalyst that is active at
15 mol %.23 Treating finely ground Cp2TiCl2 with n-BuLi and
excess PMe3 in toluene produced Cp2Ti(PMe3)2, which was then
cannula filtered into a separate vessel containing 7.5 equiv of
the enone. This solution was then cooled to the desired
temperature and the silane was added. Because the formation
of Cp2Ti(PMe3)2 is not quantitative (upon filtration, some
titanium residue is left behind), a slightly higher catalyst loading
is necessary. Without the filtration step, products resulting from
carbonyl reduction are observed. Attempts to form the oxa-
metallacycles directly from Cp2TiCl2 and n-BuLi were unsuc-
cessful. Table 3 compares the results obtained using the two
methods for catalyst generation with several substrates.
Scheme 15
substituents on the ketone (12 and 13) formed metallacycles
stoichiometrically, but they were too large to undergo facile
σ-bond metathesis without significant carbonyl reduction under
catalytic conditions. Both Whitby and Crowe found that
substrates that would give rise to cyclohexanols do not cyclize
at all, even stoichiometrically; we found that 7-hexen-2-one,
14, failed to cyclize. We attempted to favor ring closure by
incorporating substituents on the backbone; however, com-
pounds 15, 16, and 17 also failed to form metallacycles when
treated with a stoichiometric amount of 1. Furthermore,
substitution on the alkene was not tolerated by this protocol;
enones 18, 19, and 20 did not form metallacycles when treated
with a stoichiometric quantity of 1. The moderate degree of
functional group tolerance of this methodology should be noted.
Often early transition metal catalysts are incompatible with polar
functional groups; however, in the transformation reported here,
enones containing esters and allyl ethers are tolerated (Table 1,
entries 4, 5, and 9).5a Additionally, the cyclization of the â-keto
ester (Table 1, entry 11) shows that the acidic proton is not
detrimental to cyclization. In related studies of McMurry
couplings, Fu¨rstner has shown that functional group compat-
ibility is extremely sensitive to the exact nature of the low-
valent titanium species employed; he has described very good
functional group compatibility using both Ti(I) and Ti(II)
complexes.22
Conclusion
In summary, we have developed a titanium-catalyzed reduc-
tive cyclization of enones. We have explored the scope,
limitations, and the diastereoselectivity of the catalytic cycliza-
tion, and we have developed a method for the in situ generation
of an active catalyst. The net transformation described here
resembles the conversion of enones to cyclopentanols by
techniques which proceed Via radical pathways,24 but several
differences should be noted. The titanium-catalyzed method
is complementary, in that it affords the opposite isomer than
that obtained in the radical cyclizations. Furthermore, the
intermediacy of titanacycle 1 provides the opportunity for further
elaboration of the method, such as the use of a chiral titanocene
catalyst or functionalization of the Ti-C bond.
It is instructive to compare the two recent independent studies
of this type of reaction. The protocol of Crowe and co-workers,
although it uses twice the catalyst loading (20 mol %), works
well for aldehyde-containing substrates. The use of triethoxy-
silane25 lends itself to ease in isolation of the products, since
the resulting silyl ethers are stable to silica gel chromatography.
Diphenylsilyl ethers produced by the method described here are
(23) A similar protocol was also developed for catalytic enyne cycliza-
tions in: Hicks, F. A.; Berk, S. C.; Buchwald, S. L. J. Org. Chem. In press.
(24) For leading references see: (a) Shono, T.; Nishiguchi, I.; Ohmizu,
H.; Mitani, M. J. Am. Chem. Soc. 1978, 100, 545. (b) Lee, G. H.; Choi, E.
B.; Lee, E.; Pak, C. S. J. Org. Chem. 1994, 59, 1428. (c) Shono, T.; Kise,
N.; Fujimoto, T.; Yamanami, A.; Nomura, R. J. Org. Chem. 1994, 59, 1730.
For related samarium-mediated cyclizations that also give trans selectivity,
see: (d) Molander, G. A.; Kenny, C. Tetrahedron Lett. 1987, 28, 4367.
(25) Safety note: Triethoxysilane vapors can cause blindness. Silicon
Compounds Register and ReView; Anderson, R., Larson, G. L, Smith, C.,
Eds.; Hu¨ls America, Inc.; Piscataway, NJ, 1991; p 5, 190. Additionally,
under inert atmosphere, triethoxysilane is disproportionated by titanium
reagents to form SiH4, a pyrophoric gas. See: Xin, S.; Aitken, C.; Harrod,
J. F.; Mu, Y.; Samuel, E. Can. J. Chem. 1990, 68, 471.
In Situ Generation of the Catalyst. A limitation to the use
of this methodology is that the catalyst, 1, is a pyrophoric, air-
and moisture-sensitive complex that must be stored and handled
in a glovebox under argon. In order to make this methodology
more practical, a protocol for the in situ generation of the catalyst
(22) (a) Fu¨rstner, A.; Hupperts, A.; Ptock, A.; Janssen, E. J. Org. Chem.
1994, 59, 5215. (b) Fu¨rstner, A.; Hupperts, A. J. Am. Chem. Soc. 1995,
117, 4468. (c) Fu¨rstner, A.; Jumbam, D. N.; Shi, N. Z. Nacturforsch. 1995,
55b, 326.

3188 J. Am. Chem. Soc., Vol. 118, No. 13, 1996
Kablaoui and Buchwald
chromatograph with a 3392A integrator and FID detector using a 25-m
capillary column with cross-linked SE-30 as a stationary phase.
Preparation of Enone Starting Materials. Diethyl 2-Oxo-6-
heptene-4,4-dicarboxylate (Table 1, Entry 5). Using a modified
Wacker oxidation,34 palladium(II) chloride (1.06 g, 6 mmol), copper-
(I) chloride (2.97 g, 30 mmol), dimethylformamide (15 mL), and water
(2.1 mL) were added to a flask and stirred under a balloon of oxygen
for 2 h (until the solution turned green in color). Diethyl allylmalonate
(6 mL, 30 mmol) was added and the solution was stirred for 18 h.
Following the addition of H2O (30 mL), the mixture was extracted with
3 × 30 mL Et2O, the combined organic layers were washed with brine
and dried over MgSO4, and the solvent was removed in Vacuo.
Purification of the resulting yellow oil by flash chromatography
(hexane-ethyl acetate 4:1) yielded 3.7 g (56% yield) of a clear oil. Of
this, 2 g (9 mmol) was added to an oven-dried Schlenk flask containing
a slurry of NaH (0.5 g, 14 mmol) and toluene (50 mL) and the mixture
was heated to 65 °C for 0.5 h. The mixture was cooled to room
temperature and allyl bromide (0.66 mL, 11 mmol) was added. The
solution was stirred at 85 °C for 12 h. After cooling the solution to
room temperature, p-toluenesulfonic acid (0.6 g, 3 mmol) was added
and the mixture was stirred for 10 min and then filtered through Celite.
The solvent was removed in Vacuo and purification by flash chroma-
tography (hexane-ethyl acetate 9:1) afforded 0.7 g (34% yield) of a
colorless oil. 1H NMR (300 MHz, C6D6): δ 5.71 (m, 1 H); 4.92 (m,
2 H); 3.99 (q, J ) 7.2 Hz, 4 H); 3.05 (s, 2 H); 3.03 (d, J ) 7.3 Hz, 2
H); 1.64 (s, 3 H); 0.94 (t, J ) 7.2 Hz, 6 H). 13C NMR (75 MHz,
C6D6): δ 203.8, 170.2, 133.8, 118.9, 61.4, 55.5, 45.8, 38.1, 29.8, 14.0.
IR (neat): 2892, 2938, 1732, 1640, 1466, 1407, 1366, 1287, 1200,
1095, 925. Anal. Calcd for C13H20O5: C, 60.91; H, 7.87. Found: C,
61.06; H, 8.13.
N-Allyl-N-acetonylaniline (Table 1, Entry 8). A modified version
of Watanabe’s procedure35 was used. Under an argon atmosphere,
N-allylaniline (4,4 mL, 25 mmol), propargyl alcohol (1.5 mL, 25 mmol),
cadmium acetate dihydrate (12 mg, 0.04 mmol), and zinc acetate
dihydrate (12 mg, 0.05 mmol) were added to a Schlenk flask fitted
with a reflux condenser and heated to 80 °C for 3 days. The reaction
products were first purified by vacuum distillation, then the fraction
containing the title compound was further purified by flash chroma-
tography (hexane-ethyl acetate 9:1), which afforded 0.95 g (20% yield)
of a pale yellow oil. 1H NMR (300 MHz, CDCl3): δ 7.19 (dd, J )
8.9 Hz, J ) 7.3 Hz, 2 H); 6.73 (t, J ) 7.3 Hz, 1 H); 6.58 (d, J ) 8.9
Hz, 2 H); 5.85 (m, 1 H); 5.17 (m, 2 H); 3.99 (d, J ) 4.2 Hz, 2 H); 3.98
(s 2 H); 2.15 (s, 3 H). 13C NMR (75 MHz, CDCl3): δ 203.2, 148.0,
133.4, 129.2, 117.3, 116.7, 112.2, 60.7, 54.3, 26.9. IR (neat): 3040,
2917, 1726, 1675, 1599, 1506, 1383, 1354, 1233, 1165, 994, 692, 748.
Anal. Calcd for C12H15NO: C, 76.14; H, 7.99. Found: C, 76.32; H,
8.03.
3-Benzyl-6-hepten-2-one (Table 1, Entry 12). To a flame-dried
Schlenk flask containing NaH (0.26g, 11 mmol) and THF (20 mL),
tert-butyl 2-acetyl-5-hexenoate36 (2.3g, 11 mmol) was added. The
mixture was stirred at room temperature for 1 h, then benzyl bromide
(1.3 mL, 11 mmol) was added and the flask was fitted with a reflux
condenser and refluxed for 15 h. The mixture was cooled, quenched
with H2O (50 mL), and extracted with 3 × 30 mL of ethyl acetate.
The combined organic layers were washed with brine, dried over
MgSO4, and concentrated, and the 3 benzyl-3-(tert-butoxycarbonyl)-
6-heptenoate was purified by Kugelrohr distillation. A solution of
p-toluenesulfonic acid monohydrate (65 mg, 0.34 mmol) in toluene
(30 mL) under argon in a 50-mL round-bottom flask equipped with a
Dean-Stark trap and reflux condenser was heated at reflux for 1.5 h.
The solution was cooled to room temperature and all of the 3 benzyl-
3-(tert-butoxycarbonyl)-6-heptenoate from the previous step was added.
The mixture was heated at reflux for 5 h, then it was cooled to room
temperature overnight. The solution was diluted with 30 mL of diethyl
ether, washed with 30 mL of saturated NaHCO3 solution, and dried
over MgSO4. The solvent was removed in Vacuo, and the crude
material was purified by flash chromatography (hexane-ethyl acetate
not stable to silica gel and must be hydrodesilylated before
analytically pure products can be isolated. The protocol
described here is preferred for the cyclization of ketone-
containing substrates, since excessive carbonyl reduction is
avoided. We have also shown that aldehydes can be cyclized
under similar conditions using only a 10 mol % catalyst loading.
This methodology is the first example of an early transition
metal-catalyzed reductive cyclization of an alkene with a
heteroatom-containing functional group. Efforts toward the
development of other synthetically useful transformations based
on this and related catalytic processes are in progress.
Experimental Section
General Considerations. All manipulations involving air-sensitive
materials were conducted in a Vacuum Atmospheres glovebox under
an atmosphere of argon or using standard Schlenk techniques under
argon. THF was distilled under argon from sodium/benzophenone ketyl
before use. Toluene was distilled under argon from molten sodium,
and CH2Cl2 was distilled under nitrogen from CaH2. Bis(trimeth-
ylphosphine)titanocene, Cp2Ti(PMe3)2, 1, was prepared from titanocene
dichloride (obtained from Boulder Scientific, Boulder, CO) by the
procedure of Binger et al.,26 and was stored in a glovebox under argon.
The enone ethyl 2-oxo-6-heptene-3-carboxylate (Table 1, entry 11)27
was prepared from ethyl acetoacetate and 4-bromobutene (NaOEt,
HOEt, reflux), and the enones 6-hepten-2-one,28 8-nonen-4-one,29 and
2-homoallylcyclohexanone28 (Table 1, entries 1, 2, and 3) were prepared
using the same procedure with the appropriate ethyl acetate followed
by decarboxylation. Enones 4,4-dimethyl-6-hepten-2-one and 4-phenyl-
6-hepten-2-one (Table 1, entries 6 and 7) were prepared from the
allylation of the appropriate R,â-unsaturated ketone with allyltrimeth-
ylsilane and TiCl4.30 Enone diethyl 1-oxo-5-hexene-3,3-dicarboxylate
(Table 1, entry 4) was prepared by the procedure of Bernard et al.31
Allyl acetonyl ether (Table 1, entry 9) was prepared according to the
procedure of Kachinsky and Salomon32 (NaH, allyl alcohol, and
propylene oxide, followed by PCC oxidation). o-Allylacetophenone
(Table 1, entry 10) was prepared by a Stille coupling of allyltributyltin
and o-bromoacetophenone.33 Syntheses of previously unreported enone
substrates are described below. All other reagents were available from
commercial sources and were used without further purification, unless
otherwise noted.
Flash chromatography was performed on E. M. Science Kieselgel
60 (230-400 mesh). Yields, unless otherwise stated, refer to isolated
yields of compounds of greater than 95% purity as estimated by
capillary GC and 1H NMR analysis, and in the cases of unknown
compounds, elemental analysis. Yields indicated in this section refer
to a single experiment, while those reported in the tables are an average
of two or more runs, so the numbers may differ slightly. All compounds
were characterized by 1H NMR, 13C NMR, and IR spectroscopies.
Previously unreported compounds were also characterized by elemental
analysis (E & R Analytical Laboratory, Inc.). Nuclear magnetic
resonance (NMR) spectra were recorded on a Varian XL-300, a Varian
XL-500, or a Varian Unity 300. Splitting patterns are designated as
follows: s, singlet; d, doublet; dd, doublet of doublets; t, triplet; q,
quartet; qd, quartet of doublets; m, multiplet. All 1H NMR spectra are
reported in δ units, parts per million (ppm) downfield from tetra-
methylsilane. All 13C NMR spectra are reported in ppm relative to
deuteriochloroform. Infrared (IR) spectra were recorded on a Perkin-
Elmer 1600 Series Fourier transform spectrometer. Gas chromatog-
raphy (GC) analyses were performed on a Hewlet-Packard 5890 gas
(26) Binger, P.; Mu¨ller, P.; Benn, R.; Rufinska, A.; Gabor, B.; Kru¨ger,
C.; Betz, P. Chem. Ber. 1989, 122, 1035.
(27) Barbry, D.; Faven, C.; Ajana, A. Org. Prep. Proced. Int. 1994, 26,
469.
(28) Molander, G. A.; McKie, J. A. J. Org. Chem.. 1992, 57, 3132.
(29) O’Shea, M. G.; Kitching, W. Tetrahedron 1989, 45, 1177.
(30) (a) Hosomi, A.; Sakurai, H. J. Am. Chem. Soc. 1977. 99, 1673. (b)
Sakurai, H.; Hosomi, A.; Hayashi, J. Organic Synthesis; Freedan, J. P.,
Ed.; Wiley & Sons: New York, 1990; Collect. Vol. VII, p 443.
(31) Barnard, A. M.; Piras, P. P.; Toriggia, P. Synthesis 1990, 6, 527.
(32) Kachinsky, J. L. C.; Solomon, R. G J. Org. Chem. 1986, 51, 1393.
(33) Yasuo, F.; Suzuki, Y.; Tanaka, Y.; Tominaga, T.; Takeda, H.;
Sekine, H.; Morito, N.; Miyaoca, Y. Heterocycles 1977, 6, 1604.
(34) Tsuji, J.; Shimizu, I.; Yamamoto, K. Tetrahedron Lett. 1976, 34,
2975.
(35) Watanabe, Y.; Idei, M.; Takegami, Y. Tetrahedron Lett. 1979, 37,
3523.
(36) Connolly, P. J.; Heathcock, C. H. J. Org. Chem. 1985, 50, 4135.

ReductiVe Cyclization of Enones by a Titanium Catalyst
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3189
19:1) to yield 1.1 g (50% yield) of a colorless oil. 1H NMR (300 MHz,
CDCl3): δ 7.2 (m, 5 H); 5.72 (m, 1 H); 4.99 (m, 2 H); 2.74 (m, 2 H);
2.69 (m, 1 H); 2.03 (m, 2 H); 1.98 (s, 3 H); 1.75 (m, 1 H); 1.51 (m, 1
H). 13C NMR (125 MHz, CDCl3): δ 212.2, 139.4, 137.8, 128.8, 128.5,
126.3, 115.3, 53.8, 38.0, 31.4, 30.5, 30.4. IR (neat): 3027, 2926, 1713,
1496, 1453, 1351, 1162, 914, 753, 700. Anal. Calcd for C14H18O: C,
83.11; H, 8.97. Found: C, 83.35; H, 8.92.
The aqueous layer was extracted with 30-mL portions of ethyl ether
and ethyl acetate, then the combined organic layers were washed with
brine and dried over MgSO4 to afford the crude product.
General Procedure C. Cp2Ti(PMe3)2 (0.1 equiv), toluene (2-3
mL), PMe3 (0.6 equiv), and enone (1.0 equiv, thoroughly dried by
passage through a column of activated alumina) were added to a dry
Schlenk tube in a glovebox under argon. The solution was cooled at
-40 °C for 20 min, then Ph2SiH2 (1.0 equiv, also thoroughly dried by
passage through a column of activated alumina) was added. The flask
was then sealed, removed from the glovebox, and placed in a -20 °C
bath to be stirred for 16-48 h. After this time, the solution was allowed
to warm to room temperature. The toluene was removed in Vacuo to
afford the crude product.
General Procedure D. Cp2Ti(PMe3)2 (0.1 equiv), toluene (2-3
mL), PMe3 (0.6 equiv), and enone (1.0 equiv, thoroughly dried by
passage through a column of activated alumina) were added to a dry
Schlenk tube in a glovebox under argon. The solution was cooled at
-40 °C for 20 min, then Ph2SiH2 (1.0 equiv, also thoroughly dried by
passage through a column of activated alumina) was added. The flask
was then sealed, removed from the glovebox, and placed in a -20 °C
bath to be stirred for 16-48 h. After this time, the solution was allowed
to warm to room temperature. The toluene was removed in Vacuo and
the residue was put under an atmosphere of argon. THF (5 mL) and
then 1 N TBAF in THF (5 mL) were added and the reaction was stirred
for 15 min. The THF was removed in Vacuo and residue was taken
up in ether and water (30 mL each). The organic layer was washed
with brine and dried over MgSO4 to afford the crude product.
In Situ Generation of the Catalyst. General Procedure E. Finely
crushed Cp2TiCl2 (0.15 equiv) was added to a dry Schlenk flask under
argon. The flask was evacuated and backfilled three times with argon,
then 2 mL of toluene was added. The suspension was cooled to -78
°C and 0.3 equiv n-BuLi added, then after 10 min, 1 equiv of PMe3
was added. The red-colored mixture was stirred for 1 h at -78 °C
and then for 1 h at 0 °C, during which time the solution turned brown.
The reaction mixture was then cannula filtered into a dry Schlenk tube
containing 1 equiv of enone. The resulting red solution was cooled to
-20 °C, Ph2SiH2 was added, and the reaction was stirred at low
temperature for 16-48 h. After this time, the solution was allowed to
warm to room temperature. The toluene was removed in Vacuo and
acetone (10 mL) and 1 N HCl (1 mL) were added. The mixture was
stirred for 1-4 h and then diluted with 30 mL of ether and 30 mL of
saturated NH4Cl solution. The organic layer was washed with brine
and dried over MgSO4 to afford the crude product.
1,2-Dimethylcyclopentanol (Table 1, Entry 1).38 Procedure A was
used to convert 6-hepten-2-one (0.230 g, 2.34 mmol) to the desired
product. Purification by Kugelrohr distillation followed by flash
chromatography (pentane-ethyl ether 4:1) afforded 0.150 g (65% yield)
of a colorless oil. 1H NMR (300 MHz, CDCl3): δ 1.8-1.3 (m, 7 H),
1.25 (s, 3 H), 1.13 (s, 1 H), 0.94 (d, J ) 6.7 Hz, 3 H). 13C NMR (75
MHz, CDCl3): δ 79.8, 43.8, 41.0, 32.1, 25.9, 20.8, 12.4. IR (neat):
3418, 2959, 2873, 1454, 1374, 1292, 1213, 1151, 1088, 1030, 916,
875, 841, 734.
5-(Benzyloxy)-6-hepten-2-one (Table 1, Entry 13). 1,6-Heptadien-
3-ol was prepared by a method adapted from that of Ireland37 by the
addition of vinylmagnesium bromide to the crude aldehyde product
generated by the Swern oxidation of 4-penten-1-ol. The 1,6-heptadien-
3-ol (1.6 g, 14 mmol) was added to a flame-dried Schlenk flask under
argon containing a slurry of NaH (0.5 g, 20 mmol) and THF (20 mL)
and fitted with a reflux condenser. The mixture was heated to reflux
for 20 min, benzyl bromide (1.62 mL, 14 mmol) was added, and the
solution was heated at reflux for 10 h. The mixture was quenched
with 20 mL of H2O, acidified with several drops of 4 N HCl solution,
and extracted with 3 × 50 mL of diethyl ether. The combined organic
layers were washed with 30 mL of saturated NaHCO3 solution and 30
mL of brine, then dried over MgSO4. The solvent was removed in
Vacuo and the resulting oil was purified to 90% purity by flash
chromatography (hexane-ethyl acetate 48:1) to afford 1.2 g (6 mmol)
of material. Without further purification, this material was added to a
suspension of palladium(II) chloride (0.1 g, 0.6 mmol), copper(I)
chloride (0.6 g, 6 mmol), dimethylformamide (5 mL), and H2O (0.6
mL) that had stirred under a balloon of oxygen for 1 h.34 The mixture
was stirred for 4 h and then quenched with water and extracted with 3
× 20 mL of diethyl ether. The combined organic layers were washed
with 20 mL of brine and dried over MgSO4. The solvent was removed
in Vacuo, and the resulting oil was purified by flash chromatography
(hexane-ethyl acetate 9:1) to afford 0.6 g (10% yield from 4-penten-
1-ol) of a colorless oil.1H NMR (300 MHz, CDCl3): δ 7.28 (d, J )
8.1 Hz, 2 H); 7.18 (t, J ) 7.2 Hz, 2 H); 7.10 (t, J ) 7.2 Hz, 1 H); 5.58
(m, 1 H); 5.08 (m, 2 H); 4.47 (d, J ) 11.9 Hz, 1 H); 4.18 (d, J ) 11.9
Hz, 1 H); 3.61 (q, J ) 7.3 Hz, 1 H); 2.13 (t, J ) 6.5 Hz, 2 H); 1.83 (q,
J ) 7.0 Hz, 2 H); 1.61 (s, 3 H). 13C NMR (125 MHz, CDCl3): δ
208.5, 138.3, 128.3, 127.7, 127.6, 127.4, 117.4, 79.3, 70.0, 39.2, 29.9,
29.2. IR (neat): 3030, 2926, 1717, 1452, 1422, 1356, 1166, 1071,
1028, 993, 928, 736, 699. Anal. Calcd for C14H18O2: C, 77.03; H,
8.31. Found: C, 76.97; H, 8.40.
Conversion of Enones to Cyclopentanones. General Procedure
A. Cp2Ti(PMe3)2 (0.1 equiv), toluene (2-3 mL), PMe3 (0.6 equiv),
and enone (1.0 equiv, thoroughly dried by passage through a column
of activated alumina) were added to a dry Schlenk tube in a glovebox
under argon. The solution was cooled at -40 °C for 20 min, then
Ph2SiH2 (1.0 equiv, also thoroughly dried by passage through a column
of activated alumina) was added. The flask was then sealed, removed
from the glovebox, and placed in a -20 °C bath to be stirred for 16-
48 h. After this time, the solution was allowed to warm to room
temperature. The toluene was removed in Vacuo and acetone (10 mL)
and 1 N HCl (1 mL) were added. The mixture was stirred for 1-4 h
and then was diluted with 30 mL of ether and 30 mL of saturated NH4-
Cl solution. The organic layer was washed with brine and dried over
MgSO4 to afford the crude product.
1,2-Dimethyl-1-n-propylcyclopentanone (Table 1, Entry 2).39
Procedure A was used to convert 8-nonen-4-one (0.254 g, 1.8 mmol)
to the desired product. Purification by Kugelrohr distillation followed
by flash chromatography (ethyl ether-pentane 1:6) afforded 0.170 g
(67% yield) of the desired compound as a colorless oil. 1H NMR (300
MHz, CDCl3): δ 1.85-1.25 (m, 11 H), 0.96 (t, J ) 7.9 Hz, 3 H), 0.92
(d, J ) 7.0 Hz, 3 H). 13C NMR (75 MHz, CDCl3): δ 82.2, 42.9, 41.9,
38.3, 32.1, 21.0, 18.0, 14.8, 12.5. IR (neat): 3478, 2956, 2872, 1456,
1378, 942, 735.
9-Methylbicyclo[4.3.0]nonan-1-ol (Table 1, Entry 3). Procedure
A was used to convert 2-homoallylcyclohexanone (0.138 g, 0.9 mmol)
to the desired product. Purification by Kugelrohr distillation followed
by flash chromatography afforded 0.10 g (72% yield) of a 9:1 mixture
of cyclized product and the carbonyl reduction product as a colorless
oil. 1H NMR (300 MHz, CDCl3) mixture: δ 2.1-1.0 (m, 15 H), 0.90
(d, J ) 6.8 Hz, 3 H). 13C NMR (75 MHz, CDCl3): δ 81.0, 47.1, 36.2,
33.7, 31.1, 29.6, 27.8, 25.0, 23.2, 12.8. IR (neat) mixture: 3443, 2928,
General Procedure B. Cp2Ti(PMe3)2 (0.1 equiv), toluene (2-3
mL), PMe3 (0.6 equiv), and enone (1.0 equiv, thoroughly dried by
passage through a column of activated alumina) were added to a dry
Schlenk tube in a glovebox under argon. The solution was cooled at
-40 °C for 20 min, then Ph2SiH2 (1.0 equiv, also thoroughly dried by
passage through a column of activated alumina) was added. The flask
was then sealed, removed from the glovebox, and placed in a -20 °C
bath to be stirred for 16-48 h. After this time, the solution was allowed
to warm to room temperature. The toluene was removed in Vacuo and
the residue was put under an atmosphere of argon. THF (10 mL) and
CH2Cl2 (10 mL) were added and the solution was cooled to 0 °C.
Trifluoroacetic acid (3 mL) and H2O (0.3 mL) were added, and the
reaction was stirred vigorously as the ice bath warmed to room
temperature. After 16 h, saturated NaHCO3 solution (30 mL) was added
slowly, and after bubbling ceased, the reaction mixture was poured into
a seperatory funnel containing 30 mL each of ethyl ether and H2O.
(38) Saturnin, C.; Tebacchi, R.; Saxer, A. Chimia 1993, 47, 221.
(39) Molander, G. A.; Etter, J. B. Synth. Commun. 1987, 17, 901.
(37) Ireland, R. E.; Norbeck, D. W. J. Org. Chem. 1985, 50, 2198.

3190 J. Am. Chem. Soc., Vol. 118, No. 13, 1996
Kablaoui and Buchwald
2856, 1448, 1120, 1076, 997, 952, 939. Anal. Calcd for C10H18O: C,
77.87; H, 11.76. Found: C, 77.77; H, 11.73.
distillation followed by separation on a chromatatron (hexane-ethyl
acetate 9:1) afforded 61 mg of isomer A and 69 mg of isomer B (36%
and 40% yields, respectively) as colorless oils which turn to a blue
color over time if not stored at low temperature. 1H NMR (300 MHz,
CDCl3): Isomer A: δ 7.21 (t, J ) 8.8 Hz, 2 H), 6.65 (t, J ) 8.3 Hz,
1 H), 6.50 (d, J ) 8.8 Hz, 2 H), 3.42 (dd, J ) 8.7 Hz, 1 H), 3.29 (dd,
J ) 10.3 Hz, J ) 17.1 Hz, 1 H), 3.07 (dd, J ) 9.5 Hz, 1 H), 2.09 (m,
1 H), 1.68 (s, 1 H), 1.34 (s, 3 H), 1.05 (d, J ) 6.7 Hz, 3 H). Isomer
B: δ 7.22 (t, J ) 7.7 Hz, 2 H), 6.66 (t, J ) 7.3 Hz, 1 H), 6.50 (d, J
) 7.8 Hz, 2 H), 3.63 (dd, J ) 7.2 Hz, J ) 9.3 Hz, 1 H), 3.28 (dd, J
) 9.8 Hz, J ) 21.0 Hz, 2 H), 2.94 (dd, J ) 5.5 Hz, J ) 9.3 Hz, 1 H),
2.22 (m, 1 H), 1.94 (s, 1 H), 1.29 (s, 3 H), 1.0 (d, J ) 7.0 Hz, 3 H).
13C NMR (75 MHz, CDCl3): Isomer A: δ 129.1, 115.8, 111.4, 102.2,
77.4, 61.6, 53.7, 42.2, 23.0, 10.1. Isomer B: δ 129.1, 115.8, 111.3,
102.2, 78.6, 59.9, 54.0, 43.4, 21.3, 14.5. IR (neat): Isomer A: 3433,
2965, 2833, 1919, 1810, 1599, 1509, 1472, 1386, 1194, 1120, 938,
873, 750, 691. Isomer B: 3396, 2967, 2841, 1912, 1711, 1662, 1599,
1505, 1481, 1372, 1184, 1141, 999, 747, 692. Anal. Calcd for C12H17-
NO (A): C, 75.35; H, 8.96. Found: C, 75.32; H, 8.86.
1,2-Dimethyl-4-oxacyclopentyl Diphenylsilyl Ether (Table 1,
Entry 9). Procedure D was used to convert allyl acetonyl ether (0.103
g, 0.9 mmol) to the desired product. Purification by Kugelrohr
distillation afforded 0.210 g (77% yield) of a 3:2 mixture of the desired
compounds as a colorless oil. 1H NMR (300 MHz, CDCl3) mixture:
Isomer A: δ 7.62 (m, 4 H), 7.39 (m, 6 H), 5.55 (s, 1 H), 4.03 (dd, J
) 7.8 Hz, J ) 15.8 Hz, 2 H), 3.62 (dd, J ) 10.2 Hz, J ) 20.0 Hz, 2
H), 2.01 (m, 1 H), 1.34 (s, 3 H), 1.03 (d, J ) 6.9, 3 H). Isomer B: δ
7.62 (m, 4 H), 7.39 (m, 6 H), 5.55 (s, 1 H), 4.17 (t, J ) 8.1 Hz, 1 H),
3.88 (d, J ) 9.0 Hz, 1 H), 3.66 (d, J ) 10.4 Hz, 1 H), 3.43 (t, J ) 7.8
Hz, 1 H), 2.32 (m, 1 H), 1.33 (s, 3 H), 0.92 (d, J ) 7.5 Hz, 3 H). 13C
NMR (75 MHz, CDCl3) mixture: δ 134.4 (2), 130.0, 130.1, 127.9 (4),
83.6, 82.1, 79.3, 78.3, 74.9, 74.2, 49.7, 44.8, 22.3, 20.6, 14.2, 9.0. IR
(neat) mixture: 3135, 3068, 3049, 3000, 2969, 2930, 2867, 2123, 1959,
1889, 1823, 1589, 1455, 1428, 1383, 1324, 1241, 1154, 1112, 1058,
1036, 1012, 926, 890, 824, 734, 699. Anal. Calcd for C18H22O2Si
(mixture): C, 72.44; H, 7.43. Found: C, 72.68; H, 7.55. A nuclear
Overhauser enhancement study was undertaken to determine the relative
configurations of the two isomers observed. For the major isomer,
irradiation of the Si hydrogen at δ 5.55 gave no enhancement at the
C-2 hydrogen, while for the minor isomer, irradiation of the Si hydrogen
at δ 5.55 gave a 4% enhancement at the C-2 hydrogen. Based on this
observation, the relative configurations of the two isomers were assigned
as shown:
Diethyl 1-Hydroxy-2-methylcyclopentane-4,4-dicarboxylate (Table
1, Entry 4). Procedure B was used to convert diethyl 1-oxo-5-hexene-
3,3-dicarboxylate (0.183 g, 0.75 mmol) to the desired product.
Purification by Kugelrohr distillation followed by flash chromatography
(pentane-ethyl ether 7:3) afforded 0.121 g (66% yield) of a colorless
oil. 1H NMR (300 MHz, CDCl3): δ 4.22 (m, 4 H), 4.08 (m, 1 H),
2.48 (m, 1 H), 2.45 (dd, J ) 14.9 Hz, J ) 16.1 Hz, 1 H), 2.34 (dd, J
) 4.4 Hz, J ) 14.9 Hz, 1 H), 2.03 (m, 1 H), 2.01 (s, 1 H), 1.98 (m, 1
H), 1.24 (td, J ) 7.1 Hz, J ) 2.5 Hz, 6 H), 1.06 (d, J ) 6.4 Hz, 3 H).
13C NMR (75 MHz, C6D6): δ 173.7, 173.0, 75.8, 61.9, 61.6, 59.5,
44.1, 40.7, 39.9, 14.34, 14.30, 13.6. IR (neat): 3534, 2979, 1731, 1446,
1367, 1259, 1181, 1146, 1096, 1038, 961, 862, 756. Anal. Calcd for
C12H20O5: C, 59.0; H, 8.25. Found: C, 59.23; H, 8.19.
Diethyl 1-Hydroxy-1,2-dimethylcyclopentane-4,4-dicarboxylate
(Table 1, Entry 5). Procedure B was used to convert diethyl 2-oxo-
6-heptene-4,4-dicarboxylate (0.232 g, 0.75 mmol) to the desired product.
Purification by Kugelrohr distillation followed by flash chromatography
(hexane-ethyl acetate 9:1) afforded 0.161 g (69% yield) of a colorless
oil. 1H NMR (300 MHz, CDCl3): δ 4.19 (m, 4 H), 2.54 (dd, J ) 8.0
Hz, J ) 13.8 Hz, 1 H), 2.50 (d, J ) 14.8 Hz, 1 H), 2.21 (d, J ) 14.8
Hz, 1 H), 2.11 (s, 1 H), 2.03 (dd, J ) 12.2 Hz, 1 H), 1.82 (m, 1 H),
1.259 (m, 9 H), 0.97 (d, J ) 7.1 Hz, 3 H). 13C NMR (75 MHz,
CDCl3): δ 174.1, 172.6, 79.5, 61.8, 61.5, 57.1, 48.9, 43.9, 40.5, 24.5,
14.0, 13.9, 11.4. IR (neat): 3533, 2976, 1731, 1447, 1368, 1259, 1153,
1061, 930, 867. Anal. Calcd for C13H22O5: C, 60.45; H, 8.58.
Found: C, 60.58; H, 8.38.
1,2,4,4-Tetramethylcyclopentanol (Table 1, Entry 6).40 Procedure
A was used to convert 4,4-dimethyl-6-hepten-2-one (0.140 g, 1.0 mmol)
to the desired product. Purification by Kugelrohr distillation followed
by flash chromatography (pentane-ethyl ether 4:1) afforded 85 mg
(61% yield) of a colorless oil. 1H NMR (300 MHz, CDCl3): δ 1.84
(m, 1 H), 1.7-1.4 (m, 4 H), 1.22 (s, 3 H), 1.10 (s, 3 H), 1.03 (s, 1 H),
1.00 (s, 3 H), 0.92 (d, J ) 6.6 Hz, 3 H). 13C NMR (75 MHz, CDCl3):
δ 81.1, 56.6, 48.5, 43.4, 35.2, 31.9, 31.8, 26.8, 11.8. IR (neat): 3472,
2953, 2866, 2361, 1456, 1372, 1303, 1236, 1210, 1079, 1009, 933,
910, 847.
1,2-Dimethyl-4-phenylcyclopentanol (Table 1, Entry 7). Proce-
dure A was used to convert 4-phenyl-6-hepten-2-one (0.188 g, 1.0
mmol) to the desired product. Purification by Kugelrohr distillation
followed by flash chromatography (hexane-ethyl acetate 5.7: 1)
afforded 0.117 g (62% yield) of a colorless oil. 1H NMR (300 MHz,
CDCl3): δ 7.28 (m, 4 H), 7.15 (m, 1 H), 3.09 (m, 1H), 2.30 (dd, J )
10.2 Hz, J ) 14.1 Hz, 1 H), 2.12 (m, 1 H), 1.90 (dd, J ) 7.1 Hz, J )
15.0 Hz, 1 H), 1.78 (m, 1 H), 1.69 (m, 1 H), 1.32 (s, 3 H), 1.20 (s, 1
H), 1.01 (d, J ) 6.2 Hz, 3 H). 13C NMR (75 MHz, CDCl3): δ 146.6,
128.3, 127.3, 125.7, 79.6, 49.7, 45.2, 42.7, 42.2, 27.3, 12.1. IR (neat):
3576, 3461, 2958, 2931, 2871, 1945, 1871, 1804, 1602, 1493, 1455,
1373, 1121, 1031, 922, 847, 759, 700. Anal. Calcd for C13H18O: C,
82.06; H, 9.53. Found: C, 81.84; H, 9.70. A nuclear Overhauser
enhancement study was undertaken to determine the relative config-
uration of the two isomers. For the major isomer, irradiation of the
C-3 hydrogen at δ 3.09 gave a 3% enhancement at the C-4 hydrogen,
and irradiation of the C-1 methyl at δ 1.32 gave a 3% enhancement at
the C-4 hydrogen. For the minor isomer, irradiation of the C-3
hydrogen at δ 3.36 also gave a 3% enhancement at the C-4 hydrogen,
while irradiation of the C-1 methyl at δ 1.01 gave no enhancement at
the C-4 hydrogen. Based on these observations, the configuration of
the isomers were assigned as shown:
1,2-Dimethyl-1-hydroxy-2,3-dihydroindene (Table 1, Entry 10a).
Procedure B was used to convert o-allylacetophenone (0.160 g, 1.0
mmol) to the desired product. Purification by flash chromatography
(ethyl acetate-hexane 1:4) followed by Kugelrohr distillation afforded
0.144 g (89% yield) of the desired compound as a colorless oil. 1H
NMR (300 MHz, CDCl3): δ 7.38 (m, 1 H), 7.24 (m, 3 H), 2.96 (dd,
J ) 7.2 Hz, J ) 15.6 Hz, 1 H), 2.66 (dd, J ) 9.0 Hz, J ) 15.6 Hz, 1
H), 2.25 (m, 1 H), 1.56 (s, 3 H), 1.38 (s, 1 H), 1.16 (d, J ) 7.7 Hz, 3
H). 13C NMR (75 MHz, CDCl3): δ 148.1, 142.6, 128.3, 126.7, 124.9,
122.6, 80.7, 45.0, 37.9, 25.1, 12.9. IR (neat): 3422, 3068, 2967, 1913,
1707, 1606, 1477, 1375, 1290, 1215, 1183, 1076, 912, 841, 761, cm-1
.
Anal. Calcd for C11H14O: C, 81.4; H, 8.7. Found: C, 81.19; H, 8.66.
1,2-Dimethylindene (Table 1, Entry 10b).41 Procedure A was used
to convert o-allylacetophenone (0.146 g, 0.91 mmol) to the desired
product. Purification by Kugelrohr distillation followed by flash
chromatography (hexane) afforded 93 mg (71% yield) of the desired
compound as a pale yellow oil. 1H NMR (300 MHz, CDCl3): δ 7.36
(d, J ) 8.3 Hz, 1 H), 7.23 (m, 2 H), 7.10 (t, J ) 6.5 Hz, 1 H), 3.25 (s,
2 H), 2.05 (s, 3 H), 2.02 (s, 3 H). 13C NMR (75 MHz, CDCl3): δ
147.5, 142.3, 137.9, 132.4, 126.0, 123.6, 123.0, 117.9, 42.4, 13.8, 10.1.
1,2-Dimethyl-4-phenyl-4-azacyclopentanol (Table 1, Entry 8).
Procedure A was used to convert N-allyl-N-acetonylaniline (0.177 g,
9.3 mmol) to the mixture of desired products. Purification by Kugelrohr
(40) Rei, M.-H. J. Org. Chem. 1978, 43, 2173.
(41) Yeung, L. L.; Yip, Y. C.; Luh, T. Y. J. Org. Chem. 1990, 55, 1874.

ReductiVe Cyclization of Enones by a Titanium Catalyst
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3191
IR (neat): 3066, 3042, 2911, 1933, 1897, 1782, 1636, 1607, 1467,
1458, 1395, 1226, 1205, 1015, 757, 717. Anal. Calcd for C11H12: C,
91.61; H, 8.39. Found: C, 91.49; H, 8.64.
2,3-cis: 1H NMR (500 MHz, CDCl3): δ 7.32 (m, 5 H); 4.64 (d, J )
12.2 Hz, 1 H); 4.38 (d, J ) 12.2 Hz, 1 H); 3.90 (t, J ) 4.9 Hz, 1 H);
3.19 (s, 1 H); 2.01 (m, 2 H); 1.78 (m, 2 H); 1.60 (m, 1 H); 1.23 (s, 3
H); 1.10 (d, J ) 6.8 Hz, 3 H). 13C NMR (75 MHz, CDCl3): δ 134.3,
128.3, 127.5, 127.3, 84.3, 79.6, 70.8, 48.2, 40.0, 28.1, 24.7, 7.5. IR
(neat): 3448, 2962, 2871, 1453, 1207, 1091, 1027, 917, 734, 697. Anal.
Calcd for C14H20O2: C, 76.31; H, 9.16. Found: C, 76.30; H, 9.28. A
nuclear Overhauser enhancement study was undertaken to determine
the relative configuration of the three isomers observed. For the 1,2-
cis-2,3-trans isomer (major isomer), irradiation of the C-1 methyl at δ
1.27 gave a 5% enhancement at the C-2 hydrogen, and irradiation of
the C-2 methyl at δ 1.04 gave a 4% enhancement at the C-3 hydrogen.
For the 1,2-cis-2,3-cis isomer (minor isomer), irradiation of the C-2
methyl at δ 1.10 gave a 2% enhancement at the C-1 hydroxyl, but
only a 1% enhancement of the C-3 hydrogen. For the 1,2-trans-2,3-
cis isomer (not isolated, but observed when reaction run at -20 °C),
irradiation of the C-1 methyl at δ 1.21 gave a 2% enhancement of the
C-2 methyl, but only a 1% enhancement at the C-2 hydrogen.
Irradiation of the C-1 methyl at δ 0.87 gave a 5% enhancement at the
C-3 hydrogen and no enhancement at the C-1 hydroxyl. Based on
these observations, the configurations of the three isomers were assigned
as shown:
Ethyl 1-Hydroxy-1,2-dimethylcyclopentanol-5-carboxylate (Table
1, Entry 11). Procedure B was used to convert ethyl 2-oxo-6-heptene-
3-carboxylate (0.166 g, 0.9 mmol) to the title compound. Purification
by Kugelrohr distillation followed by flash chromatography (pentane-
ethyl acetate 3:1) afforded 0.132 g (79% yield) of a colorless oil. 1H
NMR (300 MHz, CDCl3): δ 4.14 (qd, J ) 7.0 Hz, J ) 2.1 Hz, 2 H);
2.83 (t, J ) 6.5 Hz, 1 H); 1.91 (m, 4 H); 1.66 (s, 1 H); 1.21 (m, 1 H);
1.28 (t, J ) 7.1 Hz, 3 H); 1.21 (s, 3 H); 0.96 (d, J ) 6.7 Hz, 3 H). 13
C
NMR (75 MHz, CDCl3): δ 174.9, 81.6, 60.3, 55.8, 43.1, 30.9, 24.9,
23.7, 14.3, 13.0. IR (neat): 3508, 2966, 2906, 2875, 1784, 1716, 1455,
1373, 1342, 1299, 1253, 1188, 1096, 1039, 917, 857. Anal. Calcd
for C10H18O3: C, 64.49; H, 9.74. Found: C, 64.36; H, 9.86.
5-Benzyl-1,2-dimethylcyclopentanol (Table 1, Entry 12). Proce-
dure A was used to convert 3-benzyl-6-hepten-2-one (0.153 g, 0.76
mmol) to the title compound. Purification by Kugelrohr distillation
followed by flash chromatography (hexane-ethyl acetate 12:1) afforded
0.120 g (77% yield) of a colorless oil. 1H NMR (300 MHz, CDCl3):
δ 7.27 (m, 2 H); 7.24 (m, 3 H); 2.93 (dd, J ) 3.4 Hz, J ) 12.6 Hz, 1
H); 2.23 (t, J ) 11.7 Hz, 1 H); 2.15 (m, 1 H); 1.75 (m, 3 H); 1.23 (s,
3 H); 1.22 (m, 2 H); 1.19 (s, 1 H); 0.96 (d, J ) 6.5 Hz, 3 H). 13C
NMR (75 MHz, CDCl3): δ 141.5, 128.9, 128.3, 125.7, 81.1, 51.4, 42.9,
37.8, 30.3, 27.6, 23.4, 14.0. IR (neat): 3449, 3028, 2956, 2871, 1583,
1495, 1452, 1372, 910, 698. Anal. Calcd for C14H20O: C, 82.29; H,
9.87. Found: C, 82.50; H, 9.82.
3-(Benzyloxy)-1,2-dimethylcyclopentanol (1,2-cis-2,3-trans and
1,2-cis-2,3-cis) (Table 1, Entry 13). Procedure B was used to convert
5-(benzyloxy)-6-hepten-2-one (99 mg, 0.45 mmol) to the title com-
pound. Purification by Kugelrohr distillation followed by flash
chromatography (hexane-ethyl acetate 9:1 (200 mL), 2.5:1 (100 mL))
afforded 38 mg of 1,2-cis-2,3-trans title compound and 14 mg of 1,2-
cis-2,3-trans title compound (52% combined yield) as colorless oils.
1,2-cis-2,3-trans: 1H NMR (300 MHz, CDCl3): δ 7.30 (m, 5 H); 4.58
(d, J ) 11.7 Hz, 1 H); 4.45 (d, J ) 11.7 Hz, 1 H); 3.72 (m, 1 H); 2.13
(m, 1 H); 1.89 (m, 1 H); 1.70 (m, 3 H); 1.27 (s, 3 H); 1.10 (s 1 H);
1.04 (d, J ) 7.0 Hz, 3 H). 13C NMR (75 MHz, CDCl3): δ 138.8,
128.3, 127.7, 127.4, 86.2, 79.1, 71.7, 50.1, 38.5, 28.3, 26.7, 10.6. IR
(neat): 3528, 2965, 1453, 1406, 1061, 1028, 734, 696. Anal. Calcd
for C14H20O2: C, 76.31; H, 9.16. Found: C, 76.04; H, 9.30. 1,2-cis-
Acknowledgment. We thank the National Institutes of
Health (GM 34917) for their generous support of this work.
N.M.K. is an National Cancer Institute Predoctoral Trainee
supported by NIH Cancer Training Grant CI T32CAO9112.
Additional support from the Dow Chemical Company is also
gratefully acknowledged. N.M.K. also thanks the American
Chemical Society organic division and Smith-Kline Beecham
and Boehringer Ingelheim Pharmaceuticals, Inc., for fellowships.
JA954192N