Utility of the Tandem Pauson-Khand Reaction in the Construction of Tetracycles

Article abstract of DOI:10.1021/jo9914014

The scope of the tandem Pauson-Khand reaction has been explored for the regiospecific construction of [5.5.5.5]- and [5.6.6.5]tetracyclic systems via the photolytic method of Livinghouse. The rapid regiospecific entry into the two dicyclopentapentanoid systems 17 and 29 was accomplished from the key diene-diynes 11 and 19b. A photochemically mediated catalytic tandem Pauson-Khand cyclization was employed to prepare the parent ring systems of dicyclopenta[a,e]pentalene (from 19b) and dicyclopenta[a,f]pentalene (from 11) in regiospecific fashion in a one-pot process. Under these conditions, conversion of acyclic diene-diyne 16 into tetracyclic system 17 was achieved in 74% yield, while a similar process was employed to convert 28 into tetracycle 29 in 90% yield. This is much improved over the previous conditions that employed NMO. Six carbon-carbon bonds were generated in this process constituting up to 98% yield for each carbon-carbon bond so formed. Furthermore, tetracyclic [5.6.6.5] systems such as dicyclopenta[b,g]decalins 37, 38, and 40 were prepared from similar diene-diyne precursors via the tandem Pauson-Khand cyclization. Importantly, acetal 36 provided the desired cis-fused [5.6.6.5] system 38a (via 40a/b) in stereospecific fashion. This reaction is unique in that it provides a cis-decalin ring system; moreover, the yield of each of the six carbon-carbon bonds formed in this process was at least 89%. The structure of cis diol 38a was confirmed by X-ray crystallography.

Full text of DOI:10.1021/jo9914014

J . Org. Chem. 2000, 65, 1957-1971
1957
Utility of th e Ta n d em P a u son -Kh a n d Rea ction in th e
Con str u ction of Tetr a cycles
Scott G. Van Ornum, Michelle M. Bruendl, Hui Cao, Mundala Reddy, Desiree S. Grubisha,
Dennis W. Bennett, and J ames M. Cook*
Department of Chemistry, University of WisconsinsMilwaukee, Milwaukee, Wisconsin 53201
Received September 7, 1999
The scope of the tandem Pauson-Khand reaction has been explored for the regiospecific construction
of [5.5.5.5]- and [5.6.6.5]tetracyclic systems via the photolytic method of Livinghouse. The rapid
regiospecific entry into the two dicyclopentapentanoid systems 17 and 29 was accomplished from
the key diene-diynes 11 and 19b. A photochemically mediated catalytic tandem Pauson-Khand
cyclization was employed to prepare the parent ring systems of dicyclopenta[a,e]pentalene (from
19b) and dicyclopenta[a,f]pentalene (from 11) in regiospecific fashion in a one-pot process. Under
these conditions, conversion of acyclic diene-diyne 16 into tetracyclic system 17 was achieved in
74% yield, while a similar process was employed to convert 28 into tetracycle 29 in 90% yield. This
is much improved over the previous conditions that employed NMO. Six carbon-carbon bonds were
generated in this process constituting up to 98% yield for each carbon-carbon bond so formed.
Furthermore, tetracyclic [5.6.6.5] systems such as dicyclopenta[b,g]decalins 37, 38, and 40 were
prepared from similar diene-diyne precursors via the tandem Pauson-Khand cyclization. Impor-
tantly, acetal 36 provided the desired cis-fused [5.6.6.5] system 38a (via 40a /b) in stereospecific
fashion. This reaction is unique in that it provides a cis-decalin ring system; moreover, the yield of
each of the six carbon-carbon bonds formed in this process was at least 89%. The structure of cis
diol 38a was confirmed by X-ray crystallography.
In tr od u ction
During the last two decades interest in the study of
molecules that contain significant amounts of strain
energy in their molecular structure has increased. The
F igu r e 1. Dicyclopenta[a,e]pentalene 1 and dicyclopenta[a,f]-
pentalene 2.
questions of bonding character, π overlap, and Hu¨ckel
stabilization are of great importance to computational
and organic chemists.1-6 The desire to better understand
the bonding character of carbon has prompted the study
of the synthesis of a number of strained molecules termed
polyquinenes.7 Dicyclopenta[a,e]pentalene (1) and dicyclo-
penta[a,f]pentalene8 (2) (Figure 1) have not been syn-
thesized and have been discussed only from a computa-
tional point of view. Controversy exists as to whether
these 14π annulenes are delocalized, exhibit aromatic
Hu¨ckel-type stability, or exist as nonalternant hydro-
carbons that behave as highly reactive olefins.9-14
The parent ring systems of 1 and 2 have been prepared
by Eaton,15,16 McKervey,17 Hafner,18 Kotha,19 Mehta,20,21
and others, as well as in our laboratory;22 however, the
present approach was designed to employ a tandem
Pauson-Khand reaction23-27 to regiospecifically generate
the tetracyclic framework of annulenes 1 and 2. This
would provide systems required for the construction of
1 and 2 in a higher oxidation state than previously
(13) Glidewell, C.; Lloyd, D. Chemica Scripta 1988, 28, 385.
(14) Zhou, Z.; Parr, R. G. J . Am. Chem. Soc. 1989, 111, 7371.
(15) Eaton, P. E.; Mu¨ller, R. H.; Carlson, G. R.; Cullison, D. A.;
Cooper, G. F.; Chou, T. C.; Krebs, E. P. J . Am. Chem. Soc. 1977, 99,
2751.
(16) Eaton, P. E.; Srikrishna, A.; Uggeri, F. J . Org. Chem. 1984,
49, 1728.
* Phone: (414) 229-5856. Fax: (414) 229-5530. Email: capncook@
csd.uwm.edu.
(1) Keese, R.; Pfenninger, A.; Roesle, A. Helv. Chim. Acta 1979, 62,
326.
(2) Hoffman, R.; Alder, R. W.; Wilcox, C. F. J . Am. Chem. Soc. 1970,
92, 4992.
(3) Lendvai, T.; Friedl, T.; Butenscho¨n, H.; Clark, T.; de Meijere, A.
Angew. Chem. 1986, 98, 734.
(4) Kuck, D.; Schuster, A.; Olhorst, B.; Sinnwell, V.; de Meijere, A.
Angew. Chem. 1989, 101, 626.
(17) McKervey, M. A.; Vibuljan, P.; Ferguson, G.; Siew, P. J . J .
Chem. Soc., Chem. Commun. 1981, 912.
(18) Stowasser, B.; Hafner, K. Angew. Chem., Int. Ed. Engl. 1986,
25, 466.
(5) Kuck, D.; Schuster, A.; Olhorst, B.; Sinnwell, V.; de Meijere, A.
Angew. Chem., Int. Ed. Engl. 1989, 28, 595.
(6) Gutman, I.; Milun, M.; Trinajstic, N. J . Am. Chem. Soc. 1977,
99, 1692.
(7) Gupta, A. K.; Fu, X.; Snyder, J . P.; Cook, J . M. Tetrahedron 1991,
47, 3665.
(8) Before the systematic method of nomenclature developed by
Chemical Abstracts Service, dicyclopenta[a,f]pentalene was known as
dicyclopenta[a,d]pentalene.
(9) Toyota, A.; Nakajima, T. Tetrahedron 1981, 37, 2572.
(10) Nakajima, T.; Toyota, A.; Kataoka, M. J . Am. Chem. Soc. 1982,
104, 5610.
(19) Kotha, S.; Brahmachary, E.; Sivakumar, R.; J oseph, A.;
Sreenivasachary, N. Tetrahedron Lett. 1997, 38, 4497.
(20) Mehta, G.; Rao, K. S. J . Org. Chem. 1985, 50, 5537.
(21) Mehta, G.; Krishnamurthy, N. J . Chem. Soc., Chem. Commun.
1986, 1319.
(22) Lannoye, G.; Sambasivarao, K.; Wehrli, S.; Weiss, U.; Cook, J .
M. J . Org. Chem. 1988, 53, 2327.
(23) Van Ornum, S. G.; Cook, J . M. Tetrahedron Lett. 1996, 37, 7185.
(24) Van Ornum, S. G.; Cook, J . M. Tetrahedron Lett. 1997, 38, 3657.
(25) Thommen, M.; Gerber, P.; Keese, R. Chimia 1991, 45, 21.
(26) Thommen, M.; Veretenov, A. L.; Guidetti-Grept, R.; Keese, R.
Helv. Chim. Acta 1996, 79, 461.
(11) Hess, B. A., J r.; Schaad, L. J . J . Org. Chem. 1971, 36, 3418.
(12) Hess, B. A., J r.; Schaad, L. J . J . Am. Chem. Soc. 1971, 93, 305.
(27) Van Ornum, S. G.; Bruendl, M. M.; Cook, J . M. Tetrahedron
Lett. 1998, 39, 6649.
10.1021/jo9914014 CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/14/2000

1958 J . Org. Chem., Vol. 65, No. 7, 2000
Van Ornum et al.
Sch em e 1
synthetic sequence for its rapid construction from readily
available diethyl malonate (6) was designed. As illus-
trated in Scheme 2, the addition of the sodium salt of
diethyl malonate,39 which was generated in situ by
treating diethyl malonate (6) with sodium ethoxide in
ethanol, to allyl bromide at 70 °C resulted in diallylation
and furnished diethyl-2,2-diallylmalonate 7. Monoester
5 was prepared directly from diester 7 by employing the
conditions of Krapcho.40 Diester 7 was stirred in dimethyl
sulfoxide at reflux in the presence of lithium chloride.
Treatment of ester 5 with excess lithium trimethylsilyl-
acetylide smoothly furnished the desired alcohol 8, and
the silyl groups were removed with tetrabutylammonium
fluoride to provide diene-diyne 4.
Under the modified conditions of the Pauson-Khand
reaction of Schreiber,41 diene-diyne 4 was treated with
3.0 equiv of dicobalt octacarbonyl [Co2(CO)8] in CH2Cl2,
and this was followed by addition of excess N-methyl-
morpholine-N-oxide monohydrate (NMO) to provide
the tetracyclic diastereomers represented by 10 in 20%
yield. Attempts to cyclize diene-diyne 8 directly under
thermal conditions, as well as using NMO as a promoter,
were not successful, presumably as a result of steric
interactions between the bulky trimethylsilyl groups in
the dicobalt hexacarbonyl functionalized intermediate
and the corresponding dione (see 9). Since this model
sequence was successful, an approach to the more highly
functionalized diol related to monoalcohol 4 was con-
ceived. A modified retrosynthetic analysis was developed
in which the diene-diyne 11 was functionalized as a diol,
as shown in Scheme 3. The key diol 11 was synthesized
from diethyl oxalate (12).
As shown in Scheme 4, a route to diol 11 from diethyl
oxalate (12) by a two-step process was developed. Nu-
merous attempts to synthesize 13 via the literature
method of Saytzeff42,43 proved too difficult to control on
a large scale because of the exothermic nature of the
process; consequently, milder conditions were sought. A
modified procedure was developed utilizing THF as a
solvent. In this vein, diethyl oxalate (12) in THF was
added to a mixture of allyl bromide and zinc in THF to
afford the R-hydroxy ester 13.
achieved.22 The newly incorporated functionality would
permit further transformations toward the eventual
synthesis of 1 and 2.
Resu lts a n d Discu ssion
The initial retrosynthetic strategy employed to gain
entrance into the angular tetraquinane system is shown
in Scheme 1. In the design of this approach the inherent
symmetry present in both annulenes 1 and 2 was
recognized. Because the Pauson-Khand reaction28-33 has
proven to be an effective means to generate fused five-
membered rings, an appropriately functionalized sub-
strate was constructed to attempt a tandem reaction. The
tandem Pauson-Khand reaction,23-27,34,35 which gener-
ates six carbon-carbon bonds in a one-pot process, is an
ideal way to quickly increase molecular complexity,
analogous to studies on the Diels-Alder and Weiss
reactions reported by Bertz.36,37 Moreover, the photo-
chemically mediated catalytic Pauson-Khand cyclization
developed by Livinghouse38 was an attractive protocol for
preparative synthesis of bicyclic enones. The key feature
of this approach was the ease with which one can
assemble an appropriately functionalized acyclic system
such as diene-diyne 4 to set up a tandem Pauson-Khand
reaction to generate the dicyclopentapentalene frame-
work.
Ester 13 was stirred with freshly prepared excess
lithium trimethylsilylacetylide to provide the desired diol
11. To determine if the tandem Pauson-Khand reaction
would take place in the absence of a conformational
restraint (i.e., an acetonide to provide a constrained
geometry), the trimethylsilyl groups were removed from
Consequently, after identification of diene-diyne 4 as
the key substrate for the first model reaction, a simple
44
diol 11 on stirring with K2CO3 in methanol to provide
(28) Khand, I. U.; Knox, G. R.; Pauson, P. L.; Watts, W. E.; Foreman,
M. I. J . Chem. Soc., Perkin Trans. 1 1973, 977.
(29) Schore, N. E. In Comprehensive Organometallic Chemistry II;
Hegedus, L. S., Ed.; Elsevier: Oxford, 1995; Vol. 12, p 703.
(30) Magnus, P.; Exon, C.; Albaugh-Robertson, P. Tetrahedron 1985,
41, 5861.
(31) Ingate, S. T.; Contelles-Marco, J . Org. Prep. Proced. Int. 1998,
30, 121.
(32) Schore, N. E. Org. React. 1991, 40, 1.
(33) Gybin, A. S.; Smit, W. A.; Caple, R.; Veretenov, A. L.; Shashkov,
A. S.; Vorontsova, L. G.; Kurella, M. G.; Chertkov, V. S.; Carapetyan,
A. A.; Kosnikov, A. Y.; Alexanyan, M. S.; Lindeman, S. V.; Panov, V.
N.; Maleev, A. V.; Struchkov, Y. T.; Sharpe, S. M. J . Am. Chem. Soc.
1992, 114, 5555.
(34) Bruendl, M. M.; Van Ornum, S. G.; Chan, T.-M.; Cook, J . M.
Tetrahedron Lett. 1999, 40, 1113.
diene-diyne 14. An attempt to cyclize diene-diyne 14 with
2.5 equiv of Co2(CO)8 in CH2Cl2 followed by addition of
excess NMO41 was unsuccessful. The lack of a conforma-
tional restraint to eclipse the diene-diyne moieties was
felt to be a major factor that prevented the tandem
cyclization toward the tetracyclic dione 15 from taking
place. Constraint of the diol 11 in an eclipsed geometry,
as shown in Scheme 4, was accomplished when diol 11
was treated with 2,2′-dimethoxypropane in refluxing
chloroform in the presence of pTSA45 to afford the desired
(35) Grossman, R. B. Tetrahedron 1999, 55, 919.
(36) Bertz, S. H.; Sommer, T. J . Chem. Commun. 1997, 2409.
(37) Bertz, S.; Sommer, T. Applications of Graph Theory to Synthesis
Planning: Complexity, Reflexivity, and Vulnerability. In Organic
Synthesis: Theory and Applications; Hudlicky, T., Ed.; J AI Press Inc:
Greenwich, CT, 1993; Vol. 2, p 67.
(39) Beaulieu, N.; Deslongchamps, P. Can. J . Chem. 1980, 58, 875.
(40) Battersby, A. R.; Westwood, S. W. J . Chem. Soc., Perkin Trans.
1 1987, 1679.
(41) Shambayati, S.; Crowe, W. E.; Schreiber, S. L. Tetrahedron Lett.
1990, 31, 5289.
(42) Saytzeff, M. J ustus Liebegs Ann. Chem. 1877, 185, 183.
(43) Schatzky, E. Ann. Chem. 1886, 185, 184.
(44) Hurst, D. T.; McInnes, A. G. Can. J . Chem. 1965, 43, 2004.
(38) Pagenkopf, B. L.; Livinghouse, T. J . Am. Chem. Soc. 1996, 118,
2285.

Pauson-Khand for the Construction of Tetracycles
J . Org. Chem., Vol. 65, No. 7, 2000 1959
Sch em e 2
Sch em e 3
F igu r e 2. NOE enhancements observed for tetracycles 17a
and 17b.
acetonide followed by removal of the silyl groups to
provide diene-diyne 16. Recently Keese25,26 has reported
a tandem Pauson-Khand reaction to furnish a [5.5.5.5]-
fenestrane system obtained in a complex mixture of
products under the modified conditions of Schreiber.41
Using this procedure, acetonide 16 was stirred with 2.5
equiv of dicobalt octacarbonyl in CH2Cl2, followed by
addition of excess N-methylmorpholine-N-oxide to pro-
vide the dicyclopentapentanoid systems represented by
17 in a one-pot process in 67% yield. The two major
components (17a and 17b) (Figure 2) were separated via
selective recrystallization from heptane. A trace of an-
other stereoisomer 17c was found in the mother liquor,
obtained in only small amounts.
The stereochemistry of each stereogenic center of these
tetracycles was not considered critical. Both isomers 17a
and 17b of the tetracyclic system should serve as func-
tionalized precursors to an annulene such as 2. The struc-
tures of all three isomers were assigned by high-resolu-
tion NMR spectroscopy including NOESY and 2D COSY
experiments (see Supporting Information for details).
To rationalize the observed stereoselectivity of the
three isomers formed in this process, MM246 calculations
F igu r e 3. Calculated relative energies of tetracycles 17a -c.
were employed to determine the relative energies of the
three diastereomers 17a -c (see Figure 3). The fact that
the calculated energies do not correlate directly with the
experimental results suggests that in the steps of the
tandem Pauson-Khand reaction, steric effects in the
metallocyclic intermediates30,47-49 and kinetics of the
insertion process are responsible for this observed stereo-
selectivity (see below).50
Attempts to effect the tandem cyclization of acetonide
16 under thermal conditions32 led to yields of the tetra-
cyclic systems represented by 17 in the range of 45-50%.
The tandem Pauson-Khand reaction can be carried out
on a 3 g level; however, large quantities of Co2(CO)8 (i.e.,
13 g) were needed under these early stoichiometric
conditions. Catalytic conditions were sought to generate
the tetracyclic system on a preparative scale. This would
render it much easier to purify the products from this
process. Recently Livinghouse38 has reported a photo-
(47) Magnus, P.; Becker, D. P. J . Am. Chem. Soc. 1987, 109, 7495.
(48) Krafft, M. E.; Chirico, X. Tetrahedron Lett. 1994, 35, 4511.
(49) Castro, J .; Moyano, A.; Perica`s, M. A.; Riera, A. Tetrahedron
1995, 51, 6541.
(50) Van Ornum, S. G. Ph.D. Thesis, University of Wisconsins
Milwaukee, 1998.
(45) Rosini, C.; Scamuzzi, S.; Focati, M. P.; Salvadori, P. J . Org.
Chem. 1995, 60, 8289.
(46) Huang, Q.; Van Ornum, S. G.; Cook, J . M., unpublished results.

1960 J . Org. Chem., Vol. 65, No. 7, 2000
Van Ornum et al.
Sch em e 4
Sch em e 5
Sch em e 6
chemically mediated catalytic intramolecular Pauson-
Khand reaction in which 5-10 mol % of Co2(CO)8 was
employed in the cyclization. Diene-diyne 16 (300-mg
scale) was stirred with 9 mol % of Co2(CO)8 and irradiated
with a Q-Beam MAX MILLION 100 W xenon spotlight51
for 72 h at 50-55 °C under an atmosphere of CO. This
process provided a mixture of tetracyclic systems repre-
sented by 17 in 74% yield. Attempts to carry out the
reaction on multigram levels with 10 mol % of Co2(CO)8
resulted in prolonged reaction times. For preparative
purposes, the photochemical cyclization was best carried
out on a 6-g scale of diene-diyne 16 with 1 equiv (50 mol
%) of Co2(CO)8 to provide the three tetracyclic systems
17a -c in 69% yield in a ratio of 4.7:5.9:1, respectively
(Scheme 4).27 The ratio of dicyclopentapentanoids was
determined by proton NMR and 13C inverse gate NMR
experiments. The observance of dione 17c in slightly
higher yield than with the cyclization carried out in the
presence of NMO may result from a slower reaction rate
resulting in greater thermodynamic control.50
chemistry. It was felt the diol 19 could be synthesized
by a pinacol coupling reaction52,53 from ketone 20, which
was readily available from propargyl alcohol 22.
An initial literature search for ketone 20 resulted in a
four-step synthesis illustrated in Scheme 6.54 Propargyl
alcohol 22 was oxidized with pyridinium chlorochromate
(PCC) to furnish the desired aldehyde, and this was
followed by cyclic acetal formation with 1,3-propanediol
in the presence of pTSA to give the 1,3-dioxane. Depro-
tonation of the acetal with n-BuLi was followed by
addition of allyl bromide to provide the ene-yne 23.
Hydrolysis of the ketal in 0.2 M aqueous H2SO4 in
refluxing acetone, however, did not provide the reported54
desired ketone 20 but furnished the R,â-unsaturated
ketone 24 instead (Scheme 6). The structure of this
This regiospecific catalytic tandem Pauson-Khand
process has been expanded to include the synthesis of
the parent ring system contained in the 14π annulene
dicyclopenta[a,e]pentalene (1). The retrosynthetic analy-
sis for the synthesis of diene-diyne 19 is illustrated in
Scheme 5. The key diene-diyne 19 required for tandem
Pauson-Khand cyclization must contain the dl stereo-
(52) Lenoir, D. Synthesis 1989, 883.
(53) McMurry, J . E. Chem. Rev. 1989, 89, 1513.
(54) Kruithof, K. J . H.; Schmitz, R. F.; Klumpp, G. W. Tetrahedron
1983, 39, 3073.
(51) Q-Beam MAX MILLION 106 candlepower spotlight distributed
by Brinkmann, Inc.

Pauson-Khand for the Construction of Tetracycles
J . Org. Chem., Vol. 65, No. 7, 2000 1961
Sch em e 7
ketone 24 was characterized;50 however, numerous at-
tempts to modify the conditions still did not provide the
desired ketone 20. It was believed the first component
formed was the ketone 20, but this olefin underwent
rapid isomerization into the more stable enone system
contained in R,â-unsaturated ketone 24. Consequently,
another route to this key ketone 20 was developed.
As illustrated in Scheme 7, alcohol 22 was oxidized
with pyridinium chlorochromate (PCC) on a 50-g scale
to provide the propargyl aldehyde, which was then
reacted with allylmagnesium chloride to furnish the
secondary alcohol 21.55 Under mild conditions at room
temperature, the alcohol 21 was treated with Na2Cr2O7‚
2H2O in a biphasic system (ether/water) to afford ketone
20. Attempts to purify this ketone on a silica gel wash
column resulted in isomerization to the R,â-unsaturated
enone 24. The lability of this system and the cleanliness
of the previous step prompted use of the crude ketone
directly without purification.
The pinacol coupling reaction52,53 has been demon-
strated to be a versatile reaction in organic synthesis to
prepare 1,2-diols. Typical metals employed in this process
include titanium,56,57 zinc,58 and magnesium,59 although
this scope is expanding as newer metals are being
exploited. The simplest conditions were sought to attempt
a pinacol coupling of ketone 20. The sensitivity of the
functional groups contained in 20 were taken into ac-
count during evaluation of the various conditions. Con-
sequently, as shown in Scheme 7, the ketone 20 was
stirred in the presence of excess zinc dust and trimethyl-
silyl chloride.58 This process furnished a mixture of meso
19a and dl 19b diols in a 3:1 ratio, respectively, in 85%
yield, which were separated by flash column chromatog-
raphy. The dl isomer 19b was required for the future
tandem Pauson-Khand cyclization (after acetonide for-
mation) as opposed to the meso isomer 19a , which would
carry the alkene and alkyne moieties cis to one another
when the diol was constrained.
If the meso diol was in the staggered conformation as
expected, the diene-diyne units might be close enough
to each other for the tandem cyclization to occur. meso-
Diol 19a was treated with K2CO3 in methanol to remove
the trimethylsilyl groups, and the colbalt-mediated
cyclization was attempted with (Co)2(CO)8/NMO. The
Pauson-Khand cyclization in the meso case was presum-
ably retarded as a result of the formation of a higher
energy trans ring junction60 of the tetraquinane 30 which
would result (Scheme 7). Another possible explanation
for the lack of success in the meso system is the lability
of the propargylic functionality associated with diol 19a .
Since the earlier diene-diyne 4 employed for the synthesis
of tetracyclic system 10 cyclized in poor yield (Scheme
2), it was not surprising that meso-diol 19a did not afford
tetracycle 30. Similar observations with propargylic
alcohols that did not undergo a Pauson-Khand cycliza-
tion were reported by Keese.25,26
The poor selectivity of dl to meso isomers may be
rationalized on the basis of the bulkiness of the radical
intermediates.50 Alteration of the alkynyl trimethylsilyl
groups to bulky triisopropylsilyl (TIPS) groups (25 f 26
f 27) to increase the stereoselectivity of dl formation did
not help direct the pinacol coupling reaction to the
desired dl product. Attempts to improve the stereoselec-
tivity of 19b to 19a employing Cp2TiCl2/Zn,61 SmI2,62,63
(55) Darresh, S.; Grant, A.; MaGee, D.; Valenta, Z. Can. J . Chem.
1991, 69, 712.
(56) McMurry, J . E.; Fleming, M. P. J . Am. Chem. Soc. 1974, 96,
4708.
(57) Mukaiyama, T.; Sato, T.; Hanna, J . Chem. Lett. 1973, 1041.
(58) Hodge, P.; Khan, M. N. J . Chem. Soc., Perkin Trans. 1 1975,
809.
(59) Csuk, R.; Fu¨rstner, A.; Weidmann, H. J . Chem. Soc., Chem.
Commun. 1986, 1802.
(60) Barret, J . W.; Linstead, R. P. J . Chem. Soc. 1936, 611.

1962 J . Org. Chem., Vol. 65, No. 7, 2000
Van Ornum et al.
Sch em e 8
Sch em e 9
or other metals were not successful or resulted only in
low yields of the diols. However, recently an alternative
pathway to the desired 19b (TIPS analogue) has been
developed that provides the two isomers in a ratio of 5.5:1
(dl/meso).64
As shown in Scheme 7, the desired dl diol 19b was
stirred with 2,2′-dimethoxypropane in refluxing chloro-
form in the presence of pTSA45 to furnish the acetonide.
The silyl groups were removed from the acetonide on
exposure to tetrabutylammonium fluoride to provide
diene-diyne 28. Acetonide 28 was stirred with 3 equiv of
Co2(CO)8 in CH2Cl2. This process was followed by addition
of NMO to provide the tetracyclic systems represented
by 29 in 62% yield as a 1:1 mixture of diastereomers.
Both stereoisomers 29a and 29b of the tetracyclic system
should serve as functionalized precursors to a planar
annulene such as 1.
To explore the utility of the photochemically mediated
catalytic tandem Pauson-Khand cyclization,27,38 diene-
diyne 28 was treated with 20 mol % of dicobalt octacar-
bonyl followed by irradiation51 for 24 h at 50-55 °C under
an atmosphere of CO. This process provided a mixture
of tetracycles 29a and 29b in 90% yield in a ratio of 2:3,
respectively.27
Recently, the tandem Pauson-Khand cyclization has
been extended to generate the tetracyclic framework of
the [5.6.6.5] carbon systems 37 or 38.34 Analogous to
the method employed for the preparation of the previous
[5.5.5.5]dicyclopentapentalene skeletons, this tandem
Pauson-Khand strategy began from acyclic molecules
(Scheme 8) but provided an internal cis-decalin within a
[5.6.6.5]dicyclopenta[b,g]decalin tetracycle (Scheme 9).
A straightforward sequence was developed for rapid
construction of the diene-diyne required for this extension
of the tandem Pauson-Khand reaction. Bisalkyne 31 was
readily available by addition of diethyl oxalate to a stirred
suspension of trimethylsilyl propargyl bromide and zinc
in THF with a catalytic amount of mercury(II) chloride.50
When the dialkyne ester 31 was treated with 3 equiv of
(61) Handa, Y.; Inanaga, J . Tetrahedron Lett. 1987, 28, 5717.
(62) Namy, J . L.; Souppe, J .; Kagan, H. B. Tetrahedron Lett. 1983,
24, 765.
(63) Namy, J . L.; Souppe, J .; Collin, J .; Kagan, H. B. J . Org. Chem.
1984, 49, 2045.
(64) Cao, H.; Cook, J . M., manuscript in preparation.

Pauson-Khand for the Construction of Tetracycles
J . Org. Chem., Vol. 65, No. 7, 2000 1963
products. Although the tetracycle with a trans-fused 6-6
ring juncture 37b was slightly lower in energy than the
cis counterpart 37a (2.5 kcal/mol), the cis compound was
preferentially formed in a 3.2:1 ratio. While an unequivo-
cal mechanistic hypothesis for the Pauson-Khand reac-
tion has not been reported to date, the currently accepted
pathway has relied on steric interactions in the metal-
locycle intermediate to determine product formation.30,47-49
As mentioned in the case of the [5.5.5.5]dicyclopenta[a,f]-
pentalene systems 17a -c discussed previously, kinetics
of the insertion process are also believed to be important,
as evidenced by the isolation of 17a as the predominant
product rather than the lower energy 17b or 17c.
Numerous factors may contribute to the observed ster-
eochemical outcome for this [5.6.6.5] tandem Pauson-
Khand system, but steric interactions in the proposed
metallocycle intermediates are believed to play a signifi-
cant role.
F igu r e 4.
Recently, Castro et al.49 have employed MMX force
field calculations to evaluate the relative energies of the
intermediary cobaltacycles in the Pauson-Khand reac-
tion, which provides new evidence to support the pro-
posed mechanism.30,47,48 The configuration of the product
formed was correlated to that of the most stable (product-
determining) cobaltacycle. The intermediate cobalta-
cyclopentane ring fusion with a trans arrangement, which
appeared to be severely strained, was shown to be more
stable than the cis isomer.49
The studies of Castro et al.49 permit the qualitative
prediction of the stereochemical outcome of this tandem
Pauson-Khand reaction. An illustration of the steric
components involved in the metallocycle intermediates
39a -f is shown in Scheme 10, accompanied by the six
possible [5.6.6.5]dicyclopenta[b,g]decalin products 37a -
f. The hydrogen atom at the juncture between the two
rings that are formed (here the 5-6 ring fusion) is shown
on the face opposite of the bulky cobalt species. In this
tandem Pauson-Khand reaction, formation of two sets
of rings at once would create an unfavorable steric
interaction between the two cobaltacycle moieties. This
situation is further exacerbated by the presence of the
two trimethylsiloxy groups. Examination of molecular
models of the six diastereomers indicated that the two
isomers that formed 37a and 37b (which have the lowest
energies) should proceed through the least hindered
metallocycle intermediates 39a and 39b.
Since cis-fused decalins were preferentially formed in
this process, it was felt the trimethylsilyl groups in diene-
diyne 34 may have retarded the reaction progress,
resulting in low yields. Steric interactions from the bulky
silyl groups would further promote a staggered arrange-
ment of diene-diyne 34, limiting the success of the
tandem Pauson-Khand reaction. Proper alignment of the
diene-diyne moieties would require an eclipsed conforma-
tion, which would appear to be much less favored in the
presence of the trimethylsilyl protecting groups. For this
reason, the tandem Pauson-Khand reaction of the
unprotected diol 33 was then carried out (Scheme 9).
Unfortunately, the method of catalytic photolysis fur-
nished low yields of the [5.6.6.5]dicyclopenta[b,g]decalin
tetracycles 38a /b [(26%), Table 1, entry b] in a 2.2:1 ratio.
This is in contrast to the [5.5.5.5]tetracyclic system en
route to the dicyclopentapentalene framework, in which
the tandem Pauson-Khand reaction of unconstrained
diol 14 did not provide the desired tetracyclic product 15.
The structure of tetracyclic cis-decalin diol 38a was
F igu r e 5.
allylmagnesium chloride, diene-diyne 32 was obtained
(Scheme 8). Removal of the silyl protecting groups from
the alkyne functions in 32 occurred readily in the
presence of K2CO3/MeOH44 to afford the desired diol 33.
Protection of the hydroxyl groups via excess DMAP and
TMSCl furnished the penultimate Pauson-Khand inter-
mediate 34.
The application of the Livinghouse photochemically
mediated Pauson-Khand protocol38 for the synthesis of
the parent ring systems of the dicyclopentapentalenes27
had previously provided a convenient, catalytic tandem
reaction process. This success prompted the use of similar
reaction conditions here for construction of the desired
[5.6.6.5]tetracycle. Conversion of diene-diyne 34 into
tetracyclic decalins 37a /b was met with somewhat lim-
ited initial success (38% yield). The two tetracyclic diones
37a /b (Figures 4 and 5) were obtained in a 3:1 ratio,
easily separated by flash column chromatography, and
fully characterized (see Supporting Information). The
spectral data for the two [5.6.6.5]dicyclopenta[b,g]deca-
lins was similar to that of the [5.5.5.5]dicyclopenta-
[a,f]pentalene systems discussed earlier (see Suporting
Information for structural assignments based on NMR
spectroscopy).
It is interesting to note that a cis-fused decalin system
was preferentially formed in this tandem Pauson-Khand
reaction. To rationalize the observed stereoselectivity of
the two isomers formed in this process, MM265,66 calcula-
tions were employed to determine the relative energies
of the potential diastereomeric [5.6.6.5] systems. The
relative energies of tetracycles 37a /b are shown in
Scheme 10, as well as those for the other four possible
isomers 37c-f (not observed). Examination of the cal-
culated energies indicated the two tetracycles obtained
are clearly lower in energy than the other potential
(65) MacroModel 6.0 was the program employed for calculations.
MM2 force field and Monte Carlo methods were employed to conduct
the conformational search.
(66) Harris, N.; Gajewski, J . J . Am. Chem. Soc. 1994, 116, 6121.

1964 J . Org. Chem., Vol. 65, No. 7, 2000
Van Ornum et al.
Sch em e 10
Ta ble 1.
must be effected by removal of this species. In the
metallocycle related to minor isomer 38b, where two
cobalt species are on one face with a hydroxyl group,
it was believed that steric interactions are somewhat
reduced, and a slightly higher amount of the trans
product was observed.
Removal of the silyl groups from the hydroxyl moieties
in tetracyclic dione 37a with tetrabutylammonium fluo-
ride hydrate in THF provided diol 38a in 89% yield. The
spectral data obtained for this product were identical in
all respects to the diol 38a obtained via the tandem
Pauson-Khand reaction of diol 33.
As shown in Table 1, a variety of reaction conditions
were investigated in attempts to increase the reaction
yield from the initial photochemical process. Recently,
Livinghouse67 has reported that the thermal (60 °C)
version of the catalytic Pauson-Khand procedure (entry
c, Table 1) occurred in yields comparable to the photolytic
catalytic method.38 In the study outlined below, yields
for the tetracyclic products were also similar to the light-
promoted reactions. The modified conditions of Schreiber41
yield
entry substrate reaction conditionsa
(%) product ratiob
a
b
c
d
e
f
OTMS 34 DME, hν
38
26
31
53
52
44
33
37a /b
38a /b
38a /b
37a /b
38a /b
38a /b
38a /b
38a /b
3.2:1
2.2:1
2.2:1
4.0:1c
2.3:1
2.2:1
2.1:1
diol 33
diol 33
DME, hν
DME, 65 °C
OTMS 34 CH2Cl2, NMO
d iol 33
diol 33
diol 33
diol 33
CH2Cl2, NMO
CH3CN, NMO
CF3CH2OH, NMO
1:1 heptane/CH2Cl2, <5
NMO
g
h
i
j
k
diol 33
heptane, NMO
0
43
45
38a /b
40a /b
40a /b
acetal 36 CH2Cl2, NMO
acetal 36 DME, hν
1.2:1
1.1:1
a
All reactions were performed under an atmosphere of carbon
b
monoxide. Ratios are approximate and were determined by
integration of the NMR spectra. c Isolated yield.
confirmed by a single-crystal X-ray analysis (see Sup-
porting Information for details).
The cis:trans ratio of products 38a /b from this reaction
series (2.2:1) was slightly lower than that of the case of
the bistrimethylsilyl-protected series (37a /b, 3.2:1). The
tandem Pauson-Khand reaction of the diene-diol 33
(67) Belanger, D. B.; O’Mahony, D. J . R.; Livinghouse, T. Tetra-
hedron Lett. 1998, 39, 7637.

Pauson-Khand for the Construction of Tetracycles
J . Org. Chem., Vol. 65, No. 7, 2000 1965
Sch em e 11
employed earlier in the route to the dicyclopentapenta-
lene framework were successfully adapted here, and the
yield of tetracycles 37a /b was increased to 53% (4:1 ratio
of diastereomers) (Scheme 9). Importantly, diol 33 was
also submitted to the same reaction conditions, which
provided tetracycle 38a /b in 52% yield. Success with this
approach prompted the study of other solvents (see Table
1, entries f-i), but CH2Cl2 remained the solvent of choice.
It was felt the polarity of CH2Cl2 contributed to the near-
homogeneous nature of the NMO reaction in contrast to
the heterogeneous mixture present when heptane was
employed.
Finally, the success of the tandem Pauson-Khand
reaction (53%) in the sterically hindered case, coupled
with no yield increase when the free diol 33 itself was
employed, prompted the construction of a constrained
diene-diyne conformation via a benzylidene acetal, which
substantially reduced rotation about the diol carbon-
carbon bond. Treatment of diol 32 with p-anisaldehyde
dimethyl acetal in refluxing toluene with a catalytic
amount of pTSA readily provided acetal 35 (Scheme 8).
The trimethylsilyl groups were removed on stirring with
K2CO3/MeOH44 to afford diene-diyne 36. This con-
strained, eclipsed diene-diyne was subjected to the condi-
tions of NMO for the tandem Pauson-Khand reaction
to provide the cis-fused decalin 40a /b in 43% yield
(Scheme 11), isolated as a 1.2:1 inseparable mixture of
acetal isomers (Table 1, entry j). The spectral data
provided evidence that only one tetracyclic core was
formed but the material was diastereomeric at the
benzylic carbon atom (see Supporting Information for
details).
Reaction of p-methoxybenzylidene acetal 36 with di-
cobalt octacarbonyl under the conditions of the photo-
chemically mediated Pauson-Khand process afforded
45% of the desired cis-decalin 40a /b in a 1.1:1 ratio (Table
1, entry k and Scheme 11). Deprotection of the p-methoxy-
benzylidene acetal in this isomeric mixture was ac-
complished by heating the mixture of decalins 40a /b in
water (∼70 °C) in the presence of a catalytic amount of
zinc chloride. This reaction exclusively afforded a single
[5.6.6.5]diol 38a in high yield, which confirmed that the
diastereomers 40a /b were isomeric at the benzylic carbon
atom and not two diastereomeric tetracycles at Hb at the
5-6 ring fusion. When this process was scaled up, 40a /b
was accompanied by a trace amount of the symmetrical
diastereomer (Ha and Hb on the same face) in an ap-
proximately 19:1 ratio. The Pauson-Khand reaction of
diene-diyne acetal 36 has been shown to be a process
that proceeds with high diastereoselectivity. This result
again lends evidence to the steric arguments for metal-
locycle intermediates believed to play a role in product
formation. The p-methoxybenzylidene protecting group
occupies a larger volume on one face of the molecule than
the trimethylsiloxy groups, which can rotate out of the
way. The presence of two cobalt moieties on the same
face (required for observation of a symmetrical isomer)
was greatly reduced in the presence of the p-methoxy-
benzylidene protecting group; consequently only one
diastereomer predominated. The process with the acetal
protecting group is unique for it provides only the cis-
fused decalin system, as expected.
Con clu sion
The regiospecific entry into the two functionalized
tetraquinanes, cyclopentapentalenes 17 and 29, was
made possible by the synthesis of the key diene-diynes
11 and 19b, followed by a photochemically mediated
tandem Pauson-Khand process. This permitted the facile
regiospecific preparation of functionalized ring systems
related to dicyclopenta[a,e]pentalene 1 and dicyclopenta-
[a,f]pentalene 2 in 74% and 90% yields, respectively. This
reaction is catalytic in some cases and stoichiometric in
others, as illustrated here. Six carbon-carbon bonds were
generated in this process, constituting a 95% and 98%
yield for each carbon-carbon bond formed in 17 and 29,
respectively. In addition to the preparation of tetracyclic
systems 17 and 29, the scope of the tandem Pauson-
Khand reaction has been extended to the [5.6.6.5] dicyclo-
penta[b,g]decalin ring system. A cis-fused decalin core
38a was obtained in stereospecific fashion from the
formation of six carbon-carbon bonds in this one-pot
process when acetal 36 was employed. The tandem
Pauson-Khand reaction is unique in that it provides a
cis-decalin ring system. Moreover, the yield of each of
the six carbon-carbon bonds formed in this procedure
in some examples (see 37a /b) was at least 89%. The
extension of the cobalt-mediated process to this [5.6.6.5]
system extends the utility of the tandem Pauson-Khand
reaction for the generation of other novel tetracyclic diene
diones.
Exp er im en ta l Section
Melting points are uncorrected. Tetrahydrofuran was dis-
tilled from sodium-benzophenone ketyl. Diisopropylamine was
dried by distillation over KOH. Dicobalt octacarbonyl was
purchased from Strem Chemical Company. All air-sensitive
reactions were kept under an inert atmosphere of nitrogen or
argon. A Q-Beam MAX MILLION 100 W xenon spotlight was
purchased from K-Mart and is distributed by Brinkmann, Inc.
4-(2-P r op en yl)-1-(t r im et h ylsilyl)-3-[(t r im et h ylsilyl)-
eth yn yl]-6-h ep ten -1-yn e-3-ol (8). A solution of trimethyl-
silylacetylene (11.5 g, 0.117 mol) in THF (50 mL) was cooled
to 0° followed by addition of n-BuLi (43.0 mL, 2.5 M in
hexanes). After 45 min, the ester39 5 (6.00 g, 0.035 mol) in THF
(55 mL) was added dropwise. The solution that resulted was
stirred for 1 h and quenched with an aqueous solution of
saturated NH4Cl (50 mL). The organic layer was washed with
brine, dried (MgSO4), and concentrated under reduced pres-
sure. The crude material was purified by flash chromatography

1966 J . Org. Chem., Vol. 65, No. 7, 2000
Van Ornum et al.
(silica gel, eluent 10% EtOAc/hexanes) to give diene-diyne 8
(10.0 g, 88%): 1H NMR (250 MHz, CDCl3) δ 5.98-5.82 (m,
2H), 5.10-4.98 (m, 4H), 2.58-2.48 (m, 2H), 2.28-2.16 (m, 2H),
1.96-1.89 (m, 1H), 0.17 (s, 18H) hydroxy signal not resolved;
13C NMR (63 MHz, CDCl3) δ 137.58, 116.09, 104.69, 89.42,
oxalate remained unreacted. An additional amount of zinc (24
g) was added in one portion followed by slow addition of allyl
bromide (30 mL neat), and the proton spectrum was rechecked
after 1 h. Examination of the proton spectrum at this time
indicated the presence of only 8% unreacted diethyl oxalate
remaining. Additional zinc (20 g) and allyl bromide (20 mL
neat) were added again, and the slurry was stirred for 1 h.
Examination again of the proton spectrum indicated complete
consumption of diethyl oxalate. The reaction was then quenched
slowly at 0-5 °C by dropwise addition of a saturated aqueous
solution of NH4Cl. A heavy precipitate formed, and ethyl
acetate (400 mL) was added followed by rapid stirring for 10
min. The gelatinous material was removed by vacuum filtra-
tion and washed thoroughly with ethyl acetate. The solid
gelatinous salts were dissolved in 5% aqueous HCl and
extracted with EtOAc to remove additional product that had
been trapped in the solids. The combined organic extracts were
neutralized with aqueous saturated NaHCO3, washed with
brine and dried (MgSO4). Concentration of the solvent under
reduced pressure followed by vacuum distillation afforded 13
(145.0 g, 84%) as a colorless liquid: bp 62-63 °C, 3.5 mmHg
(lit.42 208-209, 760 mmHg). The spectral characteristics for
13 are in agreement with the literature.50
68.02, 49.34, 34.51, -0.34; IR (neat) 3519, 3073, 1636 cm-1
;
LRMS (CI) m/z relative intensity 319 (M + 1, 4), 223 (100).
Anal. Calcd for C18H30OSi2: C, 67.85; H, 9.50. Found: C, 67.53;
H, 9.79.
3-Eth yn yl-4-(2-p r op en yl)-6-h ep ten -1-yn e-3-ol (4). A so-
lution of the diene-diyne 8 (0.300 g, 0.94 mmol) was dissolved
in THF (30 mL), treated with tetrabutylammonium fluoride
(2.8 mL, 1.0 M solution in THF), and stirred at room temper-
ature. After 20 min, the mixture was poured into water (30
mL) and extracted with EtOAc (2 × 25 mL). The organic layer
was washed with brine, dried (MgSO4), and concentrated
under reduced pressure to give the crude alcohol. Flash
chromatography (silica gel, eluent EtOAc) provided alcohol 4
(0.155 g, 94%) as a golden colored oil: 1H NMR (250 MHz,
CDCl3) δ 5.96-5.80 (m, 2H), 5.12-4.99 (m, 4H), 2.75 (broad
s, 1H), 2.60 (s, 2H), 2.59-2.48 (m, 2H), 2.30-2.18 (m, 2H),
2.02-1.94 (m, 1H); 13C NMR (63 MHz, CDCl3) δ 137.20, 116.52,
83.09, 73.05, 67.46, 48.78, 34.33; IR (neat) 3515, 3293, 3074,
1639 cm-1. This material was used directly in the next step
without further purification.
4-(2-P r op en yl)-1-(t r im et h ylsilyl)-3-[(t r im et h ylsilyl)-
et h yn yl]-6-h ep t en -1-yn e-3,4-d iol (11). A solution of tri-
methylsilylacetylene (0.968 g, 9.88 mmol) in THF (25 mL) was
stirred at 0 °C (ice bath). n-BuLi (3.64 mL, 2.5 M in hexanes)
was added, and the mixture that resulted was allowed to stir
for 30 min. A second flask that contained ester 13 (0.413 g,
2.25 mmol) in THF (10 mL) was cooled to 0 °C. The freshly
prepared lithium trimethylsilylacetylide solution was added
by syringe into the second flask. After 2 h, the reaction mixture
was quenched with a saturated aqueous solution of NH4Cl (15
mL). The aqueous layer was extracted with ether (2 × 50 mL).
The combined extracts were neutralized with a saturated
aqueous solution of NaHCO3, dried (MgSO4), and concentrated
under reduced pressure. The crude diol was purified by flash
chromatography (silica gel, eluent 15% EtOAc/hexanes) to give
11 (0.565 g, 75.4%): mp 50.5-51.5 °C; 1H NMR (250 MHz,
CDCl3) δ 6.07-5.90 (m, 2H), 5.15-5.08 (m, 4H), 2.92 (s, 1H),
2.59 (sept, J ) 7.2 Hz, 4H), 2.26 (s, 1H), 0.18 (s, 18H); 13C
NMR (63 MHz, CDCl3) δ 134.12, 118.32, 103.15, 90.98, 77.78,
Rea ction of Dien e-Diyn e 4 w ith Co2(CO)8 to P r ovid e
Tetr a cyclic Dion es 10. To a stirred solution of the diene-
diyne 4 (0.297 g, 1.71 mmol) in CH2Cl2 (15 mL) was added
Co2(CO)8 (0.649 g, 1.90 mmol, 1.1 equiv) in one portion. A new
nonpolar red spot was observed by TLC (silica gel) and
assumed to be the monometallocycle complex. An additional
1 equiv of Co2(CO)8 (0.650 g) was added to the mixture, and a
new upper red spot appeared on TLC that was believed to be
the bismetallocycle complex. N-Methylmorpholine-N-oxide
monohydrate (2.40 g, 17.8 mmol)41 was then added in one
portion, and the solution that resulted was stirred overnight.
The mixture was passed through a plug of silica gel (eluent
EtOAc) to provide a mixture of the tetracyclic diones repre-
sented by 10. Attempts to scale this reaction to preparative
levels were not successful. 10 (symmetrical isomer): 1H NMR
(250 MHz, CDCl3) δ 6.21 (d, J ) 2.2 Hz, 2H), 3.24-3.11 (m,
2H), 2.96 (t, J ) 0.2 Hz, 1H), 2.74 (d, J ) 6.4 Hz, 1H), 2.68 (d,
J ) 6.4 Hz, 1H), 2.20 (d, 2.7 Hz, 1H), 2.14-1.83 (m, 5H),
hydroxy signal not resolved; 13C NMR (63 MHz, CDCl3) δ
208.84, 188.60, 125.33, 83.82, 56.16, 43.73, 43.69, 38.78; LRMS
(CI) m/z relative intensity 231 (M + 1, 100); HRMS (EI+) calcd
for C14H14O3 (M+) 230.0942, found 230.0939. 10 (unsym-
metrical isomer): 1H NMR (250 MHz, CDCl3) δ 6.23 (d, J )
2.0 Hz, 1H), 5.87 (d, J ) 2.2 Hz, 1H), 3.75 (broad s, 1H), 3.44
(m, 1H), 3.02 (m, 2H), 2.74-2.58 (m, 3H), 2.20-2.11 (m, 2H),
71.07, 39.82, -0.50; IR (KBr) 3500, 3077, 2170, 1642 cm-1
;
LRMS (CI) m/z relative intensity 335(2), 223 (100). Anal. Calcd
for C18H30O2Si2: C, 64.61; H, 9.04. Found: C, 64.64; H, 9.09.
4-(2-P r op en yl)-1-(t r im et h ylsilyl)-3-[(t r im et h ylsilyl)-
eth yn yl]-6-h ep ten -1-yn e-3,4-d iol (11) (p r ep a r a tive sca le).
A solution of trimethylsilylacetylene (100.02 g, 1.02 mol) in
THF (500 mL) was stirred at 0 °C. n-BuLi (416 mL, 2.5 M in
hexanes) was added via a pressure equalizing dropping funnel,
and the mixture that resulted was stirred for 60 min. A
solution of the ester 13 (48.08 g, 0.258 mol in 125 mL THF)
was added dropwise via a pressure equalizing dropping funnel
over a period of 60 min, and the mixture that resulted was
allowed to stir for 2 h. An aqueous solution of saturated NH4-
Cl (125 mL) was added dropwise. The mixture was diluted with
ether (200 mL). The organic layer was washed with brine,
dried (MgSO4), and concentrated under reduced pressure. The
mixture was recrystallized from heptane to furnish a first crop
of diol 11 (29.85 g). A second crop of the diol was recrystallized
from the mother liquor to provide an additional 14.05 g. The
mother liquor can be flash chromatographed (silica gel, eluent
25% EtOAc/hexanes) to provide additional diol if desired. The
spectral characteristics of 11 were in agreement with those
reported above in a previous experiment.
2.07-1.88 (m, 1H), 1.60 (m, 1H), 1.03 (q, J ) 9.5 Hz, 1H); 13
C
NMR (63 MHz, CDCl3) δ 209.92, 209.44, 188.36, 185.58,
125.60, 124.80, 81.07, 56.70, 45.03, 42.60, 42.50, 41.74, 38.02,
35.25; LRMS (CI) m/z relative intensity 231 (M + 1, 100);
HRMS (EI+) calcd for C14H14O3 (M+) 230.0942, found 230.0940.
Eth yl-2-(2-p r op en yl)-2-h yd r oxy-4-p en ten oa te (13). To
a three-neck 3-L round-bottom flask equipped with a mechan-
ical stirrer, pressure equalizing dropping funnel, and reflux
condenser was placed granular zinc metal (186 g, 2.84 mol,
20 mesh) in dry THF (600 mL) under nitrogen. Pure allyl
bromide (202 mL in 200 mL of THF) was added dropwise
(slowly), and about 100 mL of the solution was dispensed until
the formation of the zinc Grignard was observed. This process
is very exothermic, and great care should be taken to control
the temperature. The remainder of the allyl bromide solution
was then added slowly to maintain the internal temperature
at 45 °C or lower. After complete addition (1.5 h) of the allyl
bromide solution, the mixture was stirred for 1 h, followed by
cooling in an ice bath to 0-5 °C. Diethyl oxalate (12) (138 g,
0.94 mol) in THF (100 mL) was added dropwise (slowly) to
maintain the internal temperature at 30 °C or less. After
stirring for 45 min, approximately 1.0 mL of the reaction
mixture was removed and quenched with an aqueous solution
of saturated NH4Cl for analysis by NMR. Examination of the
proton spectrum of the crude material revealed 30% of diethyl
3-(Eth yn yl)-4-(2-p r op en yl)-6-h ep ten -1-yn e-3,4-d iol (14).
A solution of the diol 11 (1.01 g, 3.01 mmol) in MeOH (25 mL)
was stirred at room temperature. Anhydrous K2CO3 (0.827 g,
5.98 mmol) was added portionwise. After 1 h, the mixture was
diluted with ether (50 mL) and H2O (50 mL). The aqueous
layer was extracted with ether (2 × 50 mL). The combined
extracts were washed with brine, dried (MgSO4), and concen-
trated under reduced pressure to afford 14 (0.567 g, 98%): 1H
NMR (250 MHz, CDCl3) δ 6.05-5.89 (m, 2H), 5.15 (d, J ) 3.8
Hz, 2H), 5.10 (s, 2H), 3.20 (broad s, 1H), 2.71-2.51 (m, 6H),

Pauson-Khand for the Construction of Tetracycles
J . Org. Chem., Vol. 65, No. 7, 2000 1967
2.4 (broad s, 1H); 13C NMR (63 MHz, CDCl3) δ 133.64, 118.93,
81.80, 77.65, 74.30, 70.53, 39.66; IR (neat) 3505, 3297, 3077,
2116, 1639 cm-1; LRMS (EI) m/z relative intensity 149 (10),
69 (100). This material was employed directly in the next step.
found 286.1201. Anal. Calcd for C17H18O4‚1/8H2O: C, 70.76; H,
6.38. Found: C, 70.85, H, 6.41.
17b: 1H NMR (250 MHz, CDCl3) δ 6.36 (d, J ) 2.2 Hz, 1H),
5.94 (d, J ) 2.0 Hz, 1H), 3.4 (m, 1H), 2.94 (m, 1H), 2.79-2.64
(m, 3H), 2.48 (dd, J ) 7.8 Hz, 1H), 2.27-2.13 (m, 2H), 1.71 (t,
J ) 12.5 Hz, 1H), 1.50 (s, 3H), 1.33 (s, 3H), 1.21 (t, J ) 12.5
Hz, 1H); 13C NMR (63 MHz, CDCl3) δ 208.73, 208.17, 182.90,
181.48, 127.14, 126.65, 117.60, 104.13, 89.21, 44.14, 43.58,
42.96, 42.64, 42.39, 41.91, 28.62, 28.02; IR (KBr) 1713, 1641
cm-1; LRMS (CI) m/z relative intensity 287 (M + 1, 100);
HRMS (EI+) calcd for C17H18O4 (M+) 286.1206, found 286.1195.
17c: 1H NMR (250 MHz, CDCl3) δ 6.28 (d, J ) 2.3 Hz, 2H),
3.40-3.28 (m, 2H), 2.74-2.54 (m, 4H), 2.23-2.13 (m, 2H), 1.85
(t, J ) 12.7 Hz, 2H), 1.36 (s, 6H); 13C NMR (63 MHz, CDCl3)
δ 207.53, 183.40, 127.06, 120.88, 102.82, 92.31, 46.83, 44.82,
43.28, 29.26; LRMS (CI) m/z relative intensity 287 (M + 1,
100); HRMS (EI+) calcd for C17H18O4 (M+) 286.1205, found
286.1198. This process could be scaled up to the 3-g level with
no loss in yield.
At t em p t ed P a u son -Kh a n d R ea ct ion w it h Dien e-
Diyn e 14. A solution of the diene-diyne 14 (0.238 g, 1.25 mmol)
in CH2Cl2 (15 mL) was stirred at room temperature. Dicobalt
octacarbonyl (1.02 g, 2.98 mmol) was added in one portion,
and the mixture that resulted was stirred for 1.5 h. The
4-methylmorpholine-N-oxide (1.50 g, 12.75 mmol) was added
to the mixture portionwise. When the reaction mixture was
monitored by TLC (silica gel, EtOAc), no evidence of product
formation was observed, and only baseline material was
visualized. The reaction was stirred for an additional 24 h and
rechecked with no change.
4,4-(Dieth yn yl)-2,2-d im eth yl-5,5-(d i-2-p r op en yl)-1,3-d i-
oxola n e (16). A solution of the diol 11 (23.02 g, 68.9 mmol),
2,2′-dimethoxypropane (16 mL), and dry pTSA (322 mg) was
dissolved in CHCl3 (1.4 L) and heated to reflux in a Soxhlet
apparatus filled with 4 Å molecular sieves for 16 h. Analysis
by TLC at this time indicated the presence of a small amount
of unreacted starting material. The reaction mixture was
cooled and concentrated under reduced pressure, followed by
flash chromatography (silica gel, eluent 5% EtOAc/hexanes)
to afford the acetonide (20.98 g, 81% yield): 1H NMR (250
MHz, CDCl3) δ 6.0-5.8 (m, 2H), 5.15 (d, J ) 2.5 Hz, 2H), 5.09
(d, J ) 1.0 Hz, 2H), 2.60 (m, 4H), 1.51 (s, 6H), 0.16 (s, 18H);
13C NMR (63 MHz, CDCl3) δ 133.17, 118.50, 110.70, 101.82,
92.23, 87.07, 75.19, 38.64, 28.63, -0.60; IR (neat) 3076, 2127,
1638 cm-1; LRMS (EI+) m/z relative intensity 374 (8), 206
(100); HRMS (EI+) calcd for C20H31O2Si2 (M+) 374.2097, found
374.2045. The recovered diol may be recycled for future use.
The acetonide was employed directly in the next step.
To a stirred solution of the acetonide (31.29 g, 83.7 mmol)
in MeOH (450 mL) was added K2CO3 (17.4 g, 0.129 mol) in
one portion. The solution that resulted was stirred at room
temperature for 30 min. The mixture was concentrated to one-
half of its original volume under reduced pressure, diluted with
CHCl3 (200 mL), and washed with water (100 mL). The aque-
ous layer was extracted with CHCl3 (2 × 50 mL). The combined
organic extracts were washed with brine (100 mL) and dried
(MgSO4). The solvent was concentrated under reduced pres-
sure followed by storing in a freezer, which resulted in a tan
solid. This material was recrystallized from heptane to afford
diene-diyne 16 (13.3 g, 70%): mp 55-56 °C; 1H NMR (250
MHz, CDCl3) δ 5.94-5.78 (m, 2H), 5.19-5.12 (m, 4H), 2.69 (s,
2H), 2.64 (d, J ) 7.3 Hz, 4H), 1.53 (s, 6H); 13C NMR (63 MHz,
CDCl3) δ 132.56, 119.03, 111.04, 86.97, 80.23, 75.47, 74.34,
38.59, 28.58; IR (neat) 3273, 3072, 2121, 1641 cm-1; LRMS
(EI+) m/z relative intensity 215 (10), 189 (100). Anal. Calcd
for C15H18O2: C, 78.21; H, 7.89. Found: C, 77.82; H, 7.92.
cis-6a,7,8,8a-Tetr ah ydr o-10,10-dim eth yl-(epoxym eth an -
oxy)-d icyclop en ta [a ,f]p en ta len e-2,5(1H,6H)-d ion e (17a ),
tr a n s-6a,7,8,8a-Tetr ah ydr o-10,10-dim eth yl-(epoxym eth an -
oxy)-d icyclop en ta [a ,f]p en ta len e-2,5(1H,6H)-d ion e (17b),
an d cis-6a,7,8,8a-Tetr ah ydr o-10,10-dim eth yl-(epoxym eth -
an oxy)-dicyclopen ta[a ,f]pen talen e-2,5(1H,6H)-dion e (17c).
A solution of the diene-diyne 16 (0.900 g, 3.91 mmol) and Co2-
(CO)8 (3.34 g, 9.76 mmol) in CH2Cl2 (45 mL) was stirred at
room temperature under an atmosphere of CO. After 12 h,
N-methylmorpholine-N-oxide (3.30 g, 28.01 mmol) was added
portionwise to the mixture. Additional NMO was later added
until no metal complexes were observed by TLC (silica gel).
The purple solution was concentrated to one-half of its original
volume under reduced pressure, loaded onto a silica gel
column, and eluted (flash chromatography) with EtOAc to
P h otoch em ica lly Med ia ted P a u son -Kh a n d Rea ction
of Aceton id e 16 to P r ovid e Tetr a cyclic System 17. A
solution of the diene-diyne 16 (6.00 g 26.07 mmol) and Co2-
(CO)8 (8.94 g, 26.13 mmol, 1 mol equiv) in degassed 1,2-DME
(300 mL) was stirred at room temperature under an atmo-
sphere of CO. After 1 h, the reaction mixture was irradiated
with a Q-Beam MAX MILLION spotlight.51 During the course
of the irradiation, the position of the lamp was adjusted such
that the internal reaction temperature was maintained be-
tween 50 and 55 °C (approximately 2.5 ft distance between
lamp and reaction flask). After 25 h, irradiation was discon-
tinued, the mixture was cooled, and the solvent was concen-
trated under reduced pressure. The residual material that was
isolated was purified by flash chromatography (silica gel,
eluted twice, 70% EtOAc/hexane) to afford 5.15 g (69%) of a
mixture of tetracyclic diones 17a , 17b, and 17c in a ratio of
4.7:5.9:1, respectively. The spectral characteristics of these
cyclopentapentalenes were identical to those reported in the
previous experiment. When this process was run on 300-mg
scale, a 74% yield of 17a -c was realized.50
1-Tr im eth ylsilyl-5-h exen -2-yn e-3-ol (21). A solution of
allylmagnesium chloride (190 mL, 2.0 M solution in THF) was
cooled to 5 °C (ice bath) followed by addition of 3-trimethylsilyl-
2-propynal66 (32.9 g, 0.261 mol) dissolved in THF (60 mL).
After stirring for 1 h, the mixture was quenched carefully with
a saturated aqueous solution of NH4Cl (125 mL) while the
internal temperature was carefully kept at 25 °C or less. The
mixture was diluted with ether (200 mL), and the aqueous
layer was extracted with ether (2 × 50 mL). The combined
extracts were washed with brine, dried (MgSO4), and concen-
trated under reduced pressure to give the crude alcohol 21.
Vacuum distillation (water aspirator) of this material afforded
21 (42.37 g, 97%) as a colorless liquid: bp 86-87 °C, 15 mmHg;
1H NMR (250 MHz, CDCl3) δ 5.92-5.76 (m, 1H), 5.18-5.11
(m, 2H), 4.37 (t, J ) 6.2 Hz, 1H), 2.43 (t, J ) 6.5 Hz, 2H), 2.12
(broad s, 1H), 0.12 (s, 9H). The spectral characteristics of 21
are in agreement with those published earlier.55
1-Tr im eth ylsilyl-5-h exen -2-yn e-3-on e (20). A solution of
1-trimethylsilyl-5-hexen-2-yne-3-ol 21 (5.00 g, 29.76 mmol) in
ether (130 mL) was cooled to 0-5 °C and stirred vigorously.
Another solution, which had been prepared from Na2Cr2O7‚
2H2O (10.0 g, 33.56 mmol) and concentrated H2SO4 (6 mL)
dissolved in water (45 mL), was added dropwise via a dropping
funnel. The temperature was maintained at 0 °C. After the
addition was complete, the ice bath was removed, and the
mixture was allowed to warm to room temperature. After 2 h,
the reaction mixture was diluted with ether (100 mL). The
organic layer was washed with a solution of saturated aqueous
NaHCO3, washed with brine, and dried (MgSO4). Concentra-
tion of the solvent under reduced pressure furnished ketone
20 (4.17 g, 84%) as a pale yellow oil, and this material was
used directly in the next step. [Note: avoid flash chromatog-
raphy at this point; even a wash column on silica gel will
isomerize the allyl group to the conjugated enone]. 1H NMR
(250 MHz, CDCl3) δ 6.01-5.84 (m, 1H), 5.23-5.13 (m, 2H),
3.29 (d, J ) 7.0 Hz, 2H), 0.22 (s, 9H); 13C NMR (63 MHz,
provide 17a and 17b (0.743 g, 67%) in
respectively, accompanied by a trace of 17c.
a ratio of 2:1,
17a : 1H NMR (250 MHz, CDCl3) δ 6.20 (d, J ) 2.3 Hz, 2H),
3.75 (m, 2H), 2.73 (dd, J ) 6.7 Hz, 2H) 2.60 (dd, J ) 8.5 Hz,
2H), 2.13 (dd, J ) 3.2 Hz, 2H), 1.54 (s, 6H), 1.32 (dd, J ) 11.5
Hz, 2H); 13C NMR (63 MHz, CDCl3) δ 208.30, 179.54, 128.57,
117.30, 106.97, 88.27, 46.05, 43.09, 42.95, 28.93; IR (KBr) 3084,
1709, 1635 cm-1; HRMS (EI+) calcd for C17H18O4 (M+) 286.1205,

1968 J . Org. Chem., Vol. 65, No. 7, 2000
Van Ornum et al.
CDCl3) δ 184.84, 129.16, 119.52, 101.69, 98.91, 49.67, -0.85;
IR (neat) 3082, 2149, 1678, 1639 cm-1; LRMS (EI+) m/z relative
intensity 125 (100).
was diluted with ether (100 mL). The organic layer was neu-
tralized with a saturated aqueous solution of NaHCO3, washed
with brine, and dried (MgSO4). Concentration of the solvent
under reduced pressure furnished ketone 27 (0.92 g, 96%) as
a pale yellow oil: 1H NMR (250 MHz, CDCl3) δ 6.02-5.86 (m,
1H), 5.24-5.13 (m, 2H), 3.30 (d, J ) 6.9 Hz, 2H), 1.10 (s, 3H),
1.09 (s, 18H); 13C NMR (63 MHz, CDCl3) δ 184.81, 129.35,
119.65, 103.96, 96.90, 50.10, 18.45, 11.04; IR (neat) 2147, 1678,
1636 cm-1; LRMS (EI+) m/z relative intensity 209 (96), 109
(100). This material was employed directly in the next step.
Attem p ted P in a col Cou p lin g of Keton e 27. A solution
of the ketone 27 (0.92 g, 3.68 mmol), activated zinc dust (2.40
g, 36.70 mmol), and trimethylsilyl chloride (2.50 g, 23.15 mmol)
dissolved in THF (20 mL) was cooled to 5 °C (ice bath) and
stirred under nitrogen. The mixture was allowed to warm to
room temperature. After 4 h, H2O was added dropwise very
carefully to quench the process one drop at a time. Ether was
added to the mixture, and the aqueous layer was extracted
with additional ether (2 × 50 mL). The combined extracts were
washed with brine, dried (MgSO4), and concentrated under
reduced pressure. Analysis by TLC (silica gel) indicated a com-
plex mixture of four to five components with the major spot
exhibiting a similar Rf to alcohol 26. Analysis of the proton
NMR of this material indicated that reduction of the starting
ketone to the alcohol 26 occurred, which suggested the inter-
mediate transition state may be too bulky for pinacol coupling.
Attem p ted P a u son -Kh a n d Cycliza tion w ith m eso-
Diol 19a . A solution of the meso-diol 19a (1.018 g, 3.00 mmol)
in 5% aqueous THF (20 mL) was stirred at room temperature.
Tetrabutylammonium fluoride (9.0 mL, 1.0 M solution in THF)
was added dropwise, and the mixture was stirred for 30 min.
The reaction solution was washed with water (30 mL) and
brine (30 mL) and dried (MgSO4). Concentration of the solvent
under reduced pressure afforded the crude diol, which was
flash chromatographed (silica gel, eluent 25% EtOAc/hexanes)
to provide the diol (0.454 g, 79%): 1H NMR (250 MHz, CDCl3)
δ 6.13-5.97 (m, 2H), 5.25-5.18 (m, 4H), 2.68 (s, 2H) 2.61 (m,
4H), 2.56 (s, 2H); 13C NMR (63 MHz, CDCl3) δ 132.90, 119.58,
83.26, 75.36, 53.94, 40.36; IR (neat) 3527, 3292, 2113, 1641
cm-1; LRMS (CI+) m/z relative intensity 173 (50), 131 (100).
This material was used directly in the next step. A solution of
the diene-diyne diol (0.193 g, 1.01 mmol) in CH2Cl2 (11 mL)
was stirred at room temperature. Dicobalt octacarbonyl (1.04
g, 3.04 mmol) was added in one portion, and the mixture that
resulted was stirred for 2 h. The 4-methylmorpholine-N-oxide
(1.50 g, 12.75 mmol) was then added to the mixture portion-
wise. When the reaction mixture was monitored by TLC (silica
gel, EtOAc), no evidence of product formation was observed;
only baseline material was visualized. The reaction was stirred
for an additional 24 h and rechecked with no change.
4-(2-P r op en yl)-1-(t r im et h ylsilyl)-3-[(t r im et h ylsilyl)-
eth yn yl)]-6-h ep ten -1-yn e-3,4-d iol (m eso-19a a n d d l-19b).
Ketone 20 (4.97 g, 29.94 mmol), activated zinc dust (19.6 g,
0.299 mol), and trimethylsilyl chloride (19.4 g, 0.299 mol) were
added to THF (85 mL) at 5 °C (ice bath) and stirred under
nitrogen. The mixture was warmed to room temperature. After
4 h, H2O was added dropwise very carefully one drop at a time,
which resulted in a moderate temperature rise. Ether (100 mL)
was added, and the aqueous layer was extracted with ether
(3 × 50 mL). The combined extracts were washed with brine,
dried (MgSO4), and concentrated under reduced pressure to
give a mixture of meso-19a and dl-19b isomers (4.49 g, 89%)
in a ratio of 3:1, respectively. Separation of the two isomers
was achieved with flash chromatography (silica gel, eluent 30%
EtOAc/hexanes).
19a (meso): 1H NMR (250 MHz, CDCl3) δ 6.12-5.95 (m,
2H), 5.20-5.13 (m, 4H), 2.60 (s, 2H), 2.54 (m, 4H), 0.16 (s,
18H); 13C NMR (63 MHz, CDCl3) δ 133.52, 118.60, 105.22,
92.10, 76.13, 40.31, -0.22; LRMS (CI) m/z relative intensity
335 (M + 1, 15), 317 (100). Anal. Calcd for C18H30O2Si2; C,
64.61; H, 9.04. Found: C, 64.68; H, 9.13.
19b (dl): 1H NMR (250 MHz, CDCl3) δ 6.12-5.96 (m, 2H),
5.23-5.14 (m, 4H), 2.65 (m, 4H), 2.53 (s, 2H), 0.16 (s, 18H);
13C NMR (63 MHz, CDCl3) δ 133.51, 119.26, 104.69, 92.12,
75.62, 41.74, -0.21; LRMS (CI) m/z relative intensity 335
(M + 1, 10), 317 (100). Anal. Calcd for C18H30O2Si2; C, 64.61;
H, 9.04. Found: C, 64.75; H, 9.05.
1-Tr iisop r op ylsilyl-5-h exen -2-yn e-3-ol (26). The 3-tri-
isopropylsilylpropargyl alcohol 25 (16.0 g, 75.47 mmol), which
had been dissolved in CH2Cl2 (75 mL), was added to a slurry
of PCC (24.4 g, 113.0 mmol) in CH2Cl2 (330 mL) and stirred
with an overhead stirrer at room temperature. After 1.5 h,
observation of TLC (silica gel) indicated starting alcohol
remained. Additional PCC (13.0 g) was added, and the mixture
was allowed to stir for 3 h. The mixture was transferred to a
wash column (silica gel) and eluted with CH2Cl2. The solvent
was removed under reduced pressure while the bath temper-
ature was kept at 25 °C or less until approximately 20 mL of
solvent remained. Vacuum distillation (water aspirator) of the
1
material afforded the aldehyde (11.4 g, 71%): bp 105 °C; H
NMR (250 MHz, CDCl3) δ 9.18 (s, 1H), 1.09 (s, 21H); 13C NMR
(63 MHz, CDCl3) δ 176.28, 104.60, 100.57, 18.40, 11.00; IR
(neat) 2147, 1667 cm-1; HRMS (EI+) calcd for C12H22OSi (M+)
210.1440, found 210.1441. This material was employed directly
in the next step.
A solution of allylmagnesium chloride (50.0 mL, 2.0 M
solution in THF) was cooled to 5 °C (ice bath) followed by
addition of the aldehyde (11.0 g, 52.38 mmol), which had been
dissolved in THF (30 mL). After 1 h, the mixture was quenched
with a saturated solution of aqueous NH4Cl (40 mL) while the
internal temperature was carefully kept at 25 °C or less. The
reaction mixture was diluted with ether (100 mL). The aqueous
layer was extracted with additional ether (2 × 50 mL). The
combined extracts were washed with brine, dried (MgSO4), and
concentrated under reduced pressure to provide the crude
alcohol 26. Purification by vacuum distillation (water aspira-
tor) afforded 26 (12.97 g, 97%) as a pale yellow oil: 1H NMR
(250 MHz, CDCl3) δ 5.97-5.80 (m, 1H), 5.21-5.13 (m, 2H),
4.42 (t, J ) 6.0 Hz, 1H), 2.46 (t, J ) 6.0 Hz, 2H), 1.81 (broad
s, 1H), 1.04 (s, 21H); 13C NMR (63 MHz, CDCl3) δ 133.04,
118.70, 108.23, 66.03, 62.17, 42.41, 18.56, 11.22; IR (neat) 3332,
3078, 2169, 1642 cm-1; HRMS (EI+) calcd for C15H28OSi (M+)
252.1909, found 252.1919. This material was employed directly
in the next step.
1-Tr iisop r op ylsilyl-5-h exen -2-yn e-3-on e (27). A solution
of 1-trimethylsilyl-5-hexen-2-yne-3-ol 26 (1.003 g, 3.98 mmol)
in ether (20 mL) was cooled to 0-5 °C and stirred vigorously.
A solution that had been prepared from Na2Cr2O7‚2H2O (3.01
g, 10.10 mmol) and concentrated H2SO4 (2.3 mL) dissolved in
water (15 mL) was added dropwise. After the addition was
complete, the ice bath was removed, and the mixture was al-
lowed to warm to room temperature. After 1 h, the mixture
tr a n s-4,5-Dieth yn yl-2,2-d im eth yl-4,5-d i-2-p r op en yl-1,3-
d ioxola n e (28). A solution of the threo-diol 19b (0.399 g, 1.19
mmol), 2,2′-dimethoxypropane (2.0 g mL), and dry pTSA (25
mg) in CHCl3 (125 mL) was heated to reflux in a Soxhlet
apparatus filled with 4 Å molecular sieves. After 6 h, analysis
by TLC (silica gel) indicated an approximate ratio of acetonide
to diol of 3:1. Additional 2,2′-dimethoxypropane (2.0 g) was
added, and the reaction was heated for an additional 2 h. After
13 h of heating (total reaction time) the reaction mixture was
cooled to room temperature and washed with a saturated
aqueous solution of NaHCO3 and brine and dried (MgSO4).
Concentration of the solvent under reduced pressure afforded
the crude acetonide, which was chromatographed by a gravity
column (silica gel, eluent 4% EtOAc/hexanes) to give the
acetonide (0.288 g, 65% yield): 1H NMR (250 MHz, CDCl3) δ
6.02-5.86 (m, 2H), 5.19-5.12 (m, 4H), 2.77 (dd, J ) 6.6 Hz,
2H), 2.53 (dd, J ) 7.6 Hz, 2H), 1.52 (s, 6H), 0.15 (s, 18H); 13
NMR (63 MHz, CDCl3) δ 132.82, 118.48, 110.64, 104.23, 94.33,
C
81.99, 42.60, 27.80, -0.34; IR (neat) 3079, 2167, 1642 cm-1
,
LRMS (EI+) m/z relative intensity 359 (8), 135 (100). This
material was employed directly in the next step. A solution of
the acetonide (0.278 g, 0.743 mmol) in 5% aqueous THF (5
mL) was stirred at room temperature. Tetrabutylammonium
fluoride (3.71 mL, 1.0 M solution in THF) was added dropwise,
and the mixture was stirred for 30 min. The reaction solution

Pauson-Khand for the Construction of Tetracycles
J . Org. Chem., Vol. 65, No. 7, 2000 1969
was diluted with ether (40 mL), washed with water (20 mL)
and brine (20 mL), and dried (MgSO4). Concentration of the
solvent under reduced pressure afforded the diene-diyne 28
(0.169 g, 98%) as a pale yellow oil: 1H NMR (250 MHz, CDCl3)
δ 6.05-5.88 (m, 2H), 5.23-5.15 (m, 4H), 2.86 (d, J ) 6.6 Hz,
4H), 2.66 (s, 2H), 1.54 (s, 6H); 13C NMR (63 MHz, C6D6) δ
132.98, 118.76, 111.13, 82.71, 82.23, 77.65, 43.03, 28.03; IR
(neat) 3293, 3077, 2109, 1641 cm-1; LRMS (EI+) m/z relative
intensity 215 (100); HRMS (EI+) calcd for C14H15O2 (M - 15)
215.1072, found 215.1071. This material was used directly in
the next step.
J ) 7.4 Hz, 4H), 0.15 (s, 18H); 13C NMR (63 MHz, CDCl3)
δ 134.47, 117.64, 103.31, 69.35, 77.64, 76.81, 40.20, 28.08,
-0.07; IR (thin film) 3529, 3074, 2174 cm-1; LRMS (CI+) m/z
relative intensity 363 (M + 1, 27), 347 (100). Anal. Calcd for
C
20H34O2Si2: C, 66.24; H, 9.45. Found: C, 66.06; H, 9.40.
Dep r otection of Dien e-Diyn e 32 to P r ovid e Dien e-
Diyn e 33. A solution of the diol 32 (7.86 g, 21.7 mmol) in
MeOH (220 mL) was stirred at room temperature. Anhydrous
K2CO3 (4.50 g, 32.6 mmol, 1.5 equiv) was added. After 1 h,
the mixture was diluted with CHCl3 (200 mL). Water was
added to dissolve the excess K2CO3 (100 mL), and the aqueous
layer was extracted with additional CHCl3 (3 × 150 mL). The
combined extracts were washed with brine, dried (MgSO4), and
concentrated under reduced pressure to afford crude 33. Flash
chromatography (silica gel, eluent 15% EtOAc/hexanes) af-
forded 33 as a pale yellow oil (4.50 g, 95%): 1H NMR (250
MHz, CDCl3) δ 6.02-5.85 (m, 2H), 5.15-5.09 (m, 4H), 2.72
(d, J ) 2.6 Hz, 4H), 2.49 (d, J ) 7.5 Hz, 4H), 2.12 (t, J ) 2.6
Hz, 2H), OH signals were not resolved; 13C NMR (65 MHz,
CDCl3) δ 134.20, 118.40, 80.61, 76.70, 72.19, 40.01, 26.24; IR
(thin film) 3532, 2118, 1638 cm-1; LRMS (CI+) m/z relative
intensity 219 (M + 1, 8), 201 (100); HRMS (EI+) calcd for
(3b .r.,4a .r.,7b .r.,8a .r.)-4a ,5,8,8a -Tet r a h yd r o-10,10-d i-
m eth yl-3b,7b-(epoxym eth an oxy)-dicyclopen ta[a ,e]pen ta-
len e-2,6(1H,4H)-d ion e (29a ) a n d (3b.r.,4a .r.,7b.r.,8a .â.)-
4a ,5,8,8a -Tetr a h yd r o-10,10-d im eth yl-3b,7b-(ep oxym eth -
an oxy)-dicyclopen ta[a ,e]pen talen e-2,6(1H,4H)-dion e (29b).
A solution of the diene-diyne 28 (0.058 g, 0.257 mmol) and
Co2(CO)8 (0.282 g, 0.826 mmol) in CH2Cl2 (10 mL) was stirred
at room temperature under an atmosphere of CO. After 1.5 h,
N-methylmorpholine-N-oxide (1.0 g, 8.56 mmol) was added to
the reaction mixture portionwise. Additional NMO was added
until the metal complexes were no longer observed by TLC.
The purple solution that remained was concentrated under
reduced pressure to one-half of its original volume, and this
material was loaded onto a silica gel column and eluted with
EtOAc to provide 29a and 29b (0.044 g, 62%) in a ratio of 2:3,
respectively: 1H NMR (250 MHz, CDCl3) δ 6.20 (d, J ) 2.2
Hz, 1H), 6.16 (d, J ) 2.0 Hz, 2H), 6.07 (d, J ) 2.3 Hz, 1H),
3.79-3.72 (m, 2H), 3.56-3.45 (m, 1H), 2.97-2.90 (m, 1H),
2.83-2.61 (m, 6H), 1.80 (t, J ) 12.7 Hz, 1H), 1.56 (s, 6H), 1.53
(s, 3H), 1.33 (s, 3H), 1.21 (dd, J ) 7.1 Hz, 12.05 Hz, 2H); 13C
NMR (63 MHz, CDCl3) δ 208.87, 208.57, 208.09, 185.98,
184.28, 183.18, 127.48, 126.09, 125.01, 117.43, 117.14, 98.21,
97.19, 94.25, 46.11, 42.70, 42.28, 42.23, 41.87, 40.72, 40.10,
39.18, 29.04, 28.48, 28.04 (two signals could not be resolved);
LRMS (CI+) m/z relative intensity 287 (M + 1, 100); HRMS
(EI+) calcd for C17H18O4 (M+) 286.1205, found 286.1194.
P h otoch em ica lly Med ia ted P a u son -Kh a n d Rea ction
of Aceton id e 28 to P r ovid e Tetr a cyclic System 29. A
solution of the diene-diyne 28 (0.196 g, 0.853 mmol) and Co2-
(CO)8 (0.060 g, 0.176 mmol, 0.2 mole equiv) in degassed 1,2-
DME (7 mL) was stirred at room temperature under an
atmosphere of CO. The reaction mixture was irradiated with
a Q-Beam MAX MILLION spotlight.51 During the course of
the irradiation, the position of the lamp was adjusted such
that the internal reaction temperature was maintained be-
tween 50 and 55 °C (approximately 2.5 ft distance between
lamp and reaction flask). After 24 h, irradiation was discon-
tinued, the mixture was cooled, and the solvent was concen-
trated under reduced pressure. The residual material that was
isolated was purified by flash chromatography (silica gel,
eluent 55% EtOAc/hexane) to afford (0.219 g, 90%) of a mixture
of tetracyclic diones 29a and 29b in a ratio of 2:3, respectively.
The spectral characteristics of these cyclopentapentalenes were
identical to those reported in the previous experiment.
Syn th esis of Dien e-Diyn e 32. A solution of ester 3150
(8.218 g, 25.36 mmol) in THF (100 mL) was stirred under N2
and cooled with a water bath. Allylmagnesium chloride (2.0
M in THF, 38.0 mL, 76.0 mmol, 3 equiv) was added quickly
via syringe, and the cooling bath was removed. After 1 h, a
0.25 mL aliquot was removed and quenched by an aqueous
solution of saturated NH4Cl for analysis by NMR. A small
amount of starting material remained. An additional 6.0 mL
of allylmagnesium chloride (12.0 mmol, 0.5 equiv) was added,
and the mixture was stirred for 1 h to complete the reaction.
The flask was cooled to 0 °C, and the reaction was quenched
via dropwise addition of aqueous saturated NH4Cl. The
reaction mixture was diluted with 100 mL of H2O and 100 mL
of Et2O. The aqueous layer was extracted with Et2O (3 × 100
mL), and the combined extracts were washed with brine, dried
(MgSO4), and concentrated under reduced pressure to provide
the crude diene-diyne. Flash chromatography (silica gel, eluent
5% EtOAc/hexanes) afforded 32 as a pale yellow oil (8.037 g,
88%): 1H NMR (250 MHz, CDCl3) δ 6.04-5.87 (m, 2H), 5.12-
5.06 (m, 4H), 2.97 (s, 1H), 2.86 (s, 1H), 2.69 (s, 4H), 2.51 (d,
C
C
14H18O2 (M+) 218.1307, found 218.1312. Anal. Calcd for
14H18O2: C, 77.03; H, 8.31. Found: C, 77.04; H, 8.31.
P r otection of Diol 33 to Give Tr im eth ylsilyl Der iva -
tive 34. To a solution of diol 33 (3.33 g, 15.3 mmol) in CH2Cl2
(100 mL) were added 4-(dimethylamino)pyridine (18.7 g, 153.4
mmol, 10 equiv) and trimethylsilyl chloride (18.5 mL, 145.8
mmol, 9.5 equiv). The reaction mixture was stirred for 20 h,
and TLC indicated the absence of unreacted starting material
and the presence of monoprotected diol, in an approximate
ratio of 1:2 with diprotected product. An additional amount of
DMAP (9.35 g, 76.6 mmol, 5 equiv) and TMSCl (9.2 mL, 72.5
mmol, 4.7 equiv) was added, and reaction mixture was stirred
for 12 h. TLC indicated the presence of only the higher Rf
diprotected diol. The reaction mixture was cooled to 0 °C and
quenched via dropwise addition of H2O (15 mL), followed by
filtration through a pad of silica gel. The filter cake was rinsed
with excess EtOAc, and the filtrates were concentrated under
reduced pressure to provide crude 34. Purification by flash
chromatography (silica gel, eluent hexanes) afforded 34 as a
colorless oil (5.46 g, 99%): 1H NMR (250 MHz, CDCl3) δ 5.99-
5.82 (m, 2H), 5.06-4.98 (m, 4H), 2.86 (dd, J ) 17.0, 2.7 Hz,
2H), 2.66 (dd, J ) 17.0, 2.7 Hz, 2H), 2.63-2.43 (m, 4H), 2.04
(t J ) 2.8 Hz, 2H), 0.19 (s, 9H), 0.14 (s, 9H); 13C NMR (63
MHz, CDCl3) δ 135.65, 116.63, 82.79, 82.63, 82.17, 71.54, 41.34,
26.62, 3.03, 2.65; IR (thin film) 3081, 3075, 2119 cm-1; LRMS
(EI+) m/z relative intensity 321 (M+ - C3H5, 9), 183 (100);
HRMS (EI+) calcd for C20H34O2Si2 (M+ - C3H5) 321.1677, found
321.1706. Anal. Calcd for C20H34O2Si2: C, 66.24; H, 9.45.
Found: C, 66.08; H, 9.47.
P r otection of Dien e-Diyn e 32 To Give p-Meth oxyben z-
ylid en e Aceta l 35. A solution of diol 32 (2.91 g, 8.05 mmol),
p-anisaldehyde dimethyl acetal (2.70 mL, 2 equiv) and dry
pTSA (123 mg, 0.09 equiv) in toluene (260 mL) was heated to
reflux in a Soxhlet apparatus filled with 4 Å molecular sieve
pellets. After 40 min, the reaction mixture was cooled to room
temperature, quenched via addition of aqueous saturated
NaHCO3 (100 mL), and diluted with Et2O (200 mL). The
aqueous layer was extracted with Et2O (3 × 150 mL), and the
combined extracts were washed with brine, dried (MgSO4), and
concentrated under reduced pressure to provide crude pro-
tected diol 35. Purification by flash chromatography (silica gel,
eluent 4% Et2O/hexanes) afforded the benzylidene acetal 35
as an almost colorless oil (3.47 g, 90% yield): 1H NMR (250
MHz, CDCl3) δ 7.45 (d, J ) 8.6 Hz, 2H), 6.88 (d, J ) 8.7 Hz,
2H), 6.08-5.87 (m, 2H), 5.99 (s, 1H), 5.18-5.03 (m, 4H), 3.80
(s, 3H), 3.06 (d, J ) 17.1 Hz, 1H), 2.85-2.66 (m, 4H), 2.59-
2.50 (m, 3H), 0.17 (s, 18H); 13C NMR (75 MHz, CDCl3) δ 160.02,
133.72, 133.36, 131.02, 127.64 (2), 118.31, 118.03, 113.60 (2),
102.94, 102.34, 100.41, 88.00, 87.77, 85.54, 84.74, 55.27, 38.76,
36.74, 27.00, 24.83, 0.00; IR (thin film) 3076, 2177 cm-1; LRMS
(CI+) m/z relative intensity 481 (M + 1, 45), 345 (100); HRMS
(EI+) calcd for C28H40O3Si2 (M+) 480.2539, found 480.2516.

1970 J . Org. Chem., Vol. 65, No. 7, 2000
Van Ornum et al.
Anal. Calcd for C28H40O3Si2: C, 69.95; H, 8.39. Found: C,
69.65; H, 8.32.
39.66, 39.29, 38.37, 37.37, 37.18, 36.96, 35.42; minor isomer δ
213.03, 184.66, 127.15, 75.09, 70.94, 39.80, 38.84, 37.66, 35.42;
LRMS (CI) m/z relative intensity 275 (M + 1, 100). Anal. Calcd
for C16H18O4‚1/3H2O: C, 68.55; H, 6.71. Found: C, 68.46; H,
6.71. Major isomer 38a was partially separated and character-
ized: 1H NMR (250 MHz, CDCl3) δ 5.96 (s, 2H), 3.75 (bs, 1H),
3.29 (m, 2H), 3.02 (d, J ) 13.2 Hz, 1H), 2.79-2.52 (m, 5H),
2.40 (m, 1H), 2.20 (dd, J ) 13.6, 6.1 Hz, 1H), 2.07-2.00 (m,
3H), 1.89-1.69 (m, 2H); 13C NMR (75 MHz, CDCl3) 208.88,
208.44, 180.44, 179.53, 130.70, 129.00, 76.44, 73.06, 42.71,
42.02, 41.55, 41.17, 40.16, 38.94, 38.67, 36.78; LRMS (CI) m/z
relative intensity 275 (M + 1, 100).
Rem ova l of Tr im eth ylsilyl Gr ou p s of p-Meth oxyben z-
ylid en e Acet a l 35 To Affor d 36. A solution of the benz-
ylidene acetal 35 (1.06 g, 2.22 mmol) in MeOH (30 mL) was
stirred at room temperature. Anhydrous K2CO3 (0.49 g, 3.56
mmol, 1.6 equiv) was added. After 2 h, TLC analysis of the
reaction mixture indicated that a small amount of monopro-
tected alkyne remained. Additional K2CO3 (0.40 g, 1.3 equiv)
was added, and the reaction mixture was stirred for 1 h. The
mixture was concentrated under reduced pressure to a white
paste, which was dissolved in 100 mL each of H2O and Et2O.
The aqueous layer was extracted with Et2O (3 × 100 mL). The
combined extracts were washed with H2O and brine, dried
(MgSO4), and concentrated under reduced pressure to afford
crude 36, which was purified by flash chromatography (silica
gel, eluent 7% EtOAc/hexanes) to afford 36 as a colorless oil
(0.73 g, 98%): 1H NMR (300 MHz, CDCl3) δ 7.42 (d, J ) 8.7
Hz, 2H), 6.88 (d, J ) 8.7 Hz, 2H), 6.06-5.85 (m, 3H), 5.20-
5.06 (m, 4H), 3.79 (s, 3H), 3.04 (dd, J ) 17.0, 2.7 Hz, 1H), 2.82-
2.70 (m, 4H), 2.57-2.53 (m, 3H), 2.11-2.07 (m, 2H); 13C NMR
(75 MHz, CDCl3) δ 159.15, 132.37, 132.07, 129.68, 126.73 (2),
117.61, 117.44, 122.72 (2), 99.56, 84.51, 83.65, 79.12, 78.75,
70.59, 70.51, 54.30, 37.66, 35.57, 24.50, 22.23; IR (thin film)
3291, 3073, 2119, 1612 cm-1; LRMS (CI+) m/z relative intensity
337 (M + 1, 27), 137 (100); HRMS (EI+) calcd for C22H24O3
(M+) 336.1711, found 376.1725. Anal. Calcd for C22H24O3: C,
78.54; H, 7.19. Found: C, 78.57; H, 7.16.
Gen er a l P r oced u r e A: P h otoch em ica l Ca ta lytic P a u -
son -Kh a n d Rea ction of Dien e-Diyn e w ith Co2(CO)8 To
P r ovid e a [5.6.6.5]Tetr a cyclic Dion e. A solution of the
diene-diyne 34, 33, or 36 (1 equiv) and dicobalt octacarbonyl
(1.05 equiv) in degassed 1,2-DME (∼0.10 M) was stirred at
room temperature under an atmosphere of CO. After 1 h
the reaction mixture was irradiated with a Q-Beam MAX
MILLION spotlight.51 During the course of the irradiation, the
position of the lamp was adjusted such that the internal
reaction temperature was maintained between 50 and 55 °C
(approximately 2.5 ft distance between lamp and reaction
flask). After 20 h, irradiation was discontinued, the mixture
was cooled, and the solvent was concentrated under reduced
pressure. The residual material that was isolated was purified
by flash chromatography on silica gel to provide tetracycles
37a /b, 38a /b, or 40a /b, respectively.
40a /b: 1H NMR (300 MHz, CDCl3) δ 7.43 (d, J ) 8.6 Hz,
2H), 7.32 (d, J ) 8.7 Hz, 2H), 6.93-6.87 (m, 4H), 6.25 (s, 1H),
6.16 (s, 1H), 6.04 (s, 3H), 5.87 (s, 1H), 3.81 (s, 3H), 3.80 (s,
3H), 3.34-3.18 (m, 5H), 3.05 (d, J ) 13.5, 1H), 2.85 (d, J )
13.0 Hz, 1H), 2.65-2.49 (m, 13H), 2.09-2.03 (m, 4H), 2.03-
1.60 (m, 4H); 13C NMR (75 MHz, CDCl3) δ 206.10, 205.97,
205.72, 205.50, 176.52, 176.24, 175.33, 175.06, 157.75 (2),
129.32, 128.90, 128.39, 127.94, 127.74, 126.98, 124.74 (2),
124.61 (2), 111.86 (2), 111.77 (2), 98.36, 98.01, 82.30, 80.50,
79.61; LRMS (CI) m/z relative intensity 393 (M + 1, 57), 137
(100); HRMS (EI+) calcd for C24H24O5 (M+) 392.1624, found
392.1613. When this reaction was run on a larger scale, 40a /b
was accompanied by a trace amount of the symmetrical
diastereomer (Ha and Hb on the same face) in an approximately
19:1 ratio.
Th er m a l P a u son -Kh a n d Rea ction of Dien e-Diyn e
w ith Co2(CO)8 To P r ovid e a [5.6.6.5]Tetr a cyclic Dion e.
A solution of the diene-diyne 33 (0.413 g, 1.89 mmol) and
dicobalt octacarbonyl (0.681 g, 1.99 mmol, 1.05 equiv) in
degassed 1,2-DME (18 mL) was stirred at room temperature
under an atmosphere of CO. After 1 h, the reaction mixture
was heated to an internal temperature of 65 °C via an oil bath
and stirred for 17 h, at which time the reaction mixture was
cooled, and the solvent was concentrated under reduced
pressure. The residual material that remained was purified
by flash chromatography (silica gel, eluent 5% MeOH/EtOAc)
to afford of a mixture of tetracyclic diones 38a and 38b in a
ratio of 2.2:1, respectively (0.161 g, 31%). The spectral char-
acteristics of these tetracyclic diones were identical to those
reported in a previous experiment.
Gen er a l P r oced u r e B: P a u son -Kh a n d Rea ction of
Dien e-Diyn e w ith Co2(CO)8 a n d NMO To P r ovid e a
[5.6.6.5]Tetr a cyclic Dion e. A solution of the diene-diyne 34,
33, or 36 (1 equiv) and dicobalt octacarbonyl (2.5 equiv) in the
solvent indicated in Table 1 (∼0.07 M) was stirred at room
temperature under an atmosphere of CO. After stirring for 2
h, the reaction mixture was diluted with additional solvent,
and N-methylmorpholine-N-oxide was added to the reaction
mixture portionwise. After stirring several hours, additional
NMO was added until no metal complexes were observed by
TLC (silica gel). The reaction solution was concentrated to one-
half of its original volume, and the mother liquor was loaded
onto a silica gel column and purified by flash chromatography
to provide tetracycles 37a /b, 38a /b, or 40a /b, respectively.
Dep r otection of Silyl-P r otected [5.6.6.5]Tetr a cycle 37a
To P r ovid e Tetr a cyclic Diol 38a . To a solution of [5.6.6.5]-
tetracycle 37a (42.9 mg, 0.103 mmol) in THF (1.0 mL) was
added tetrabutylammonium fluoride hydrate (56.0 mg, 0.214
mmol, 2.1 equiv), turning the reaction solution deep red. The
reaction mixture was stirred for 10 min and diluted with
15 mL each of H2O, CHCl3, and MeOH. The aqueous layer
was extracted with additional CHCl3 and MeOH (1:1 v/v, 4 ×
20 mL). The combined extracts were dried (MgSO4) and
concentrated under reduced pressure to afford crude diol 38a ,
which was purified by flash chromatography (silica gel, eluent
6% MeOH/EtOAc) to afford 38a as a white foam (24.2 mg,
89%). The spectral characteristics of diol 38a were identical
to those reported for diol 38a in a previous Pauson-Khand
experiment.
37a : 1H NMR (500 MHz, CDCl3) δ 5.95 (s, 1H), 5.90 (s, 1H),
3.24-3.18 (m, 1H), 3.15 (d, J ) 12.4 Hz, 1H), 2.81-2.74 (m,
1H), 2.68 (d, J ) 12.4 Hz, 1H), 2.65 (d, J ) 14.8 Hz, 1H), 2.64
(dd, J ) 18.6, 6.6 Hz, 1H), 2.59 (dd, J ) 18.6, 6.6 Hz, 1H),
2.48 (d, J ) 14.8 Hz, 1H), 2.08 (dd, J ) 12.6, 6.0 Hz, 1H), 2.03
(dd, J ) 18.6, 3.0 Hz, 1H), 2.03 (dd, J ) 12.0, 12.0 Hz, 1H),
2.02 (dd, J ) 18.6, 3.0 Hz, 1H), 1.95 (dd, J ) 12.0, 6.0 Hz,
1H), 1.73 (dd, J ) 13.5, 12.5 Hz, 1H), 0.20 (s, 9H), 0.16 (s,
9H); 13C NMR (125 MHz, CDCl3) δ 208.35, 207.71, 180.37,
179.63, 130.42, 129.16, 80.36, 77.20, 42.36, 41.58, 41.51, 41.48,
40.65, 40.31, 38.70, 36.95, 2.71, 2.53; LRMS (CI+) m/z relative
intensity 419 (M + 1, 100). Anal. Calcd for C22H34O4Si2‚1/4H2O;
C, 62.44; H, 8.20. Found: C, 62.43; H, 8.23.
1
37b: H NMR (250 MHz, CDCl3) δ 5.92 (s, 2H), 3.08-3.00
(m, 2H), 2.97 (d, J ) 14.1 Hz, 2H), 2.69 (d, J ) 14.1 Hz, 2H),
2.57 (dd, J ) 18.5, 6.5 Hz, 2H), 1.98 (dd, J ) 18.5, 2.8 Hz,
2H), 1.90 (dd, J ) 13.5, 6.0 Hz, 2H), 1.68 (dd, J ) 13.0, 13.0
Hz, 2H), 0.23 (s, 9H), 0.02 (s, 9H); 13C NMR (75 MHz, CDCl3)
δ 207.93, 180.39, 130.03, 81.41, 77.53, 41.61, 40.26, 39.17,
37.02, 2.77, 2.49; LRMS (CI) m/z relative intensity 419 (M +
1, 100); HRMS (EI+) calcd for C22H34O4Si2 (M+) 418.2047, found
418.1996. When this reaction was run on a larger scale,
sufficient quantities of 37b were obtained that permitted the
accurate structural assignment of the minor isomer as trans
rather than the cis system previously implied.34
38a /b: 1H NMR (300 MHz, D2O) δ 6.08-6.04 (m, 4H), 3.34-
3.23 (m, 2H), 3.04-3.00 (m, 2H), 3.02 (d, J ) 13.3 Hz, 2H),
2.87 (d, J ) 13.3 Hz, 2H), 2.81-2.63 (m, 8H), 2.21-1.91 (m,
10H), 1.79 (dd, J ) 12.8, 12.8 Hz, 1H), 1.60 (dd, J ) 13.1, 13.1
Hz, 1H); 13C NMR (75 MHz, D2O) major isomer δ 213.24,
212.50, 183.62, 183.39, 127.59, 126.56, 75.23, 71.58, 40.14,
Dep r otection of Ben zylid en e Aceta l [5.6.6.5]Tetr a -
cycle 40a /b To Affor d Tetr a cyclic Diol 38a . Approximately
6 mg of acetal 40a /b and 0.5 mL of H2O were placed in a micro
test tube, and a few crystals of zinc chloride were added. The

Pauson-Khand for the Construction of Tetracycles
J . Org. Chem., Vol. 65, No. 7, 2000 1971
heterogeneous reaction mixture was heated on a sand bath at
70 °C. The reaction mixture was monitored by TLC (silica gel,
10% MeOH/EtOAc), and evidence of product formation of lower
Rf was observed within 30 min, identical to the tetracyclic diol
38a by comparison to an authentic sample (Rf value ) 0.23).
Consumption of the starting material and conversion to diol
38a was essentially complete after 10 h.
Su p p or t in g In for m a t ion Ava ila b le: Detailed charac-
terization and assignment of the structures of 17a , 17b; 29a ,
29b; 37a , 37b; and 40a , 40b by NMR spectroscopy. Tables of
crystallographic data, structure solution and refinement,
atomic coordinates, bond lengths and angles, and anisotropic
thermal parameters for 38a . Detailed experimental procedures
for the syntheses of compounds 37a /b, 38a /b, and 40a /b
illustrated in Table 1. An X-ray crystallographic file is avail-
able in CIF format. This material is available free of charge
via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. The authors wish to thank Mr.
Frank Laib for high resolution mass spectral data; Dr.
Eugene DeRose, Dr. Tze-Ming Chan, and Dr. F. Holger
Fo¨rsterling for NMR data; and Mr. Keith Krumnow for
CHN analysis. We thank Dr. Brian Pagenkopf for
helpful discussions.
J O9914014