Allyl sulfones as precursors to allylzincs in the palladium-catalyzed zinc-ene cyclization: Highly efficient synthesis of enantiopure (-)-erythrodiene

Article abstract of DOI:10.1021/ol051049i

(Chemical Equation Presented) Easily prepared allyl phenyl sulfones, capable of introduction of the alkene by electrophilic α-substitution, are superior to allyl acetates as substrates for Pd-catalyzed Zn-ene cyclizations, providing C5 or C₄N r

Full text of DOI:10.1021/ol051049i

ORGANIC
LETTERS
2005
Vol. 7, No. 17
3637-3640
Allyl Sulfones as Precursors to
Allylzincs in the Palladium-Catalyzed
Zinc-ene Cyclization: Highly Efficient
Synthesis of Enantiopure
(−
)-Erythrodiene†
Kai Deng, Justin Chalker, Ao Yang, and Theodore Cohen*
Department of Chemistry, UniVersity of Pittsburgh, Pittsburgh, PennsylVania 15260
cohen@pitt.edu
Received May 6, 2005
ABSTRACT
Easily prepared allyl phenyl sulfones, capable of introduction of the alkene by electrophilic
r-substitution, are superior to allyl acetates as
substrates for Pd-catalyzed Zn-ene cyclizations, providing C5 or C4N rings with cis-1,2-vinyl and -CH2Zn substituents. Several examples, with
different methods of substrate preparation, are presented.
Magnesium-ene cyclizations (Scheme 1) have been used in
a number of elegant syntheses, especially by Oppolzer.1 They
Scheme 1. Mg-Ene Cyclization
are highly stereoselective with vinyl and CH2Mg substituents
oriented in a cis fashion on the new rings. Li-ene cyclizations
occur at lower temperatures but are thermodynamically
unfavorable.2,3
There are several clues that carbometalations via the Zn-
ene process are far more facile than those proceeding by the
Mg-ene process.4 Furthermore, many functional groups that
are unstable in the presence of Grignard reagents can survive
† Taken in part from the Ph.D. Thesis of Kai Deng, University of
in the presence of organozinc compounds.5
Pittsburgh, Pittsburgh, PA, 2004.
Zn-ene cyclizations had not been well studied until the
appearance of Oppolzer’s work on Pd-catalyzed Zn-ene
cyclizations using allyl acetate precursors.6 The mechanism
is thought to involve Pd(0) insertion into the allyl acetate to
generate a π-allyl palladium intermediate that undergoes
transmetalation with diethylzinc to give the corresponding
(1) (a) Oppolzer, W. Angew. Chem., Int. Ed. Engl. 1989, 28, 38-52. (b)
Oppolzer, W. Pure Appl. Chem. 1990, 62, 1941-1948. (c) Oppolzer, W.
In ComprehensiVe Organic Synthesis; Trost, B. M., Ed.; Pergamon Press:
New York, 1991; Vol. 5, pp 29-61.
(2) Cheng, D.; Zhu, S.; Liu, X.; Norton, S. H.; Cohen, T. J. Am. Chem.
Soc. 1999, 121, 10241-10242.
(3) Cheng, D.; Knox, K. R.; Cohen, T. J. Am. Chem. Soc. 2000, 122,
412-13.
10.1021/ol051049i CCC: $30.25
© 2005 American Chemical Society
Published on Web 07/20/2005

allylzinc intermediate together with an ethyl palladium
species (Scheme 2). The allylzinc undergoes the Zn-ene
allylzincation could be constructed in a highly efficient,
connective manner. The use of allyl phenyl sulfones as
precursors for Pd-catalyzed allylzincation of unactivated
alkenes has not been reported.
We now present four syntheses involving four different
concepts for preparing in a highly efficient manner the allyl
sulfones required for R-alkylation and intramolecular car-
bozincation. Furthermore, we provide the first definitive
evidence that the Pd-catalyzed Zn-ene cyclization provides
cis-disubstituted five-membered rings as in the Mg case; this
fact has some bearing on the question of whether this is a
Zn-ene rather than a Pd-ene cyclization.
Scheme 2. Oppolzer’s Mechanism for the
Palladium-Catalyzed Zinc-ene Cyclization
One example of the production of a simple allyl phenyl
sulfone is illustrated by the production of methallyl phenyl
sulfone 112 by treatment of methallyl chloride with com-
mercial C6H5SO2Na (Scheme 3). The allyl phenyl sulfone
cyclization to produce a species capable of being trapped
by electrophiles to give functionalized products. The Pd atom
attached to the ethyl group undergoes PdH elimination to
release ethylene and regenerate the Pd(0) catalyst. The
stereochemistry of the products was thought6a to be cis on
the basis of the fact that the 13C chemical shifts of the ring
carbon atoms bearing the substituents in the electrophile-
trapped products were within the range found for the 3- and
4-carbon atoms in 3,4-dimethyl and 3,4-diethyl-N-meth-
ylpyrrolidines and in a different range than those of the
corresponding trans-disubstituted pyrrolidines.7,8
Scheme 3. Zn-Ene Cyclization Utilizing an Allyl Sulfone
Prepared from an Allyl Chloride
On the basis of this elegant work and the extensions of it
described below, we envisioned a great increase in the
versatility of Pd-catalyzed Zn-ene cyclizations by replacing
the allyl acetates with allyl phenyl sulfones as precursors.
Trost9 had demonstrated that allyl phenyl sulfones could
serve as substrates for allyl palladium formation, while Julia10
had shown that allyl palladiums generated in this way could
be reduced by diethylzinc to generate allylzincs capable of
undergoing intermolecular addition to aldehydes. Due to the
wide variety of methods available for producing allyl phenyl
sulfones (see below) along with their valuable feature of
readily losing an R-proton to form an allylic anion that reacts
with alkylating agents virtually exclusively at the R-posi-
tion,11 we expected that many substrates for Pd-catalyzed
anion, generated by treatment of 1 with n-butyllithium, is
alkylated exclusively at the R-position by 1-bromo-4-pentene
2 to provide a high yield of cyclization substrate 3. Subjecting
3 to 10 mol % Pd(PPh3)4 and 6 equiv of Et2Zn affords the
cyclopentane derivative 4, after iodination, in excellent yield
as the only detectable product. The cis stereochemistry of 4
was established by converting a sample to the corresponding
1
sulfone 5; both the 13C and H NMR spectra of the latter
were identical with those reported for the same compound,
the X-ray crystal structure of which had been determined.13
This definitive evidence for the stereochemistry of the
cyclization product lends strong credence to the suggestions
of Oppolzer and Schro¨der that the main products of the Pd-
catalyzed Zn-ene cyclizations are indeed rings bearing
adjacent cis substituents. It also provides some support, albeit
weak, for their contention that this is a Zn-ene cyclization
rather than a Pd-ene cyclization, followed by Zn-Pd
exchange; the Pd-ene cyclization in an analogous system,
although under somewhat different conditions, yields a trans
product.6a,14
(4) Intermolecular comparisons: Lehmkuhl, H.; Reinehr, D.; Hennenberg,
D.; Schomburg, G.; Schroth, G. Ann. Chem. 1975, 119-144. Lehmkuhl,
H.; Nehl, H. J. Organomet. Chem. 1973, 60, 1-10. Lehmkuhl, H.; Doring,
I.; Nehl, H. J. Organomet. Chem. 1981, 221, 123-130. Intramolecular
comparisons: Courtemanche, G.; Normant, J.-F. Tetrahedron Lett. 1991,
32, 5317-5320. Meyer, C.; Marek, I.; Normant, J.-F.; Platzer, N.
Tetrahedron Lett. 1994, 35, 5645-5648. Lorthiois, E.; Marek, I.; Meyer,
C.; Normant, J.-F. Tetrahedron Lett. 1995, 36, 1263-1266. Lorthiois, E.;
Marek, I.; Normant, J.-F. Tetrahedron Lett. 1997, 38, 89-92.
(5) Review: Knochel, P.; Perea, J. J. A.; Jones, P. Tetrahedron 1998,
54, 8275-8319.
(6) (a) Oppolzer, W.; Schro¨der, F. Tetrahedron Lett. 1994, 35, 7939-
7942. (b) In the previous year, it had been shown that allylzinc species,
capable of adding to an aldehyde, could be generated from allyl esters in
the presence of 1 equiv of diethylzinc and a catalytic quantity of Pd(PPh3)4;
Yasui, K.; Goto, Y.; Yajima, T.; Taniseki, Y.; Fugami, K.; Tanaka, A.;
Tamaru, T. Tetrahedron Lett. 1993, 34, 7619-7622.
Since allyl phenyl sulfides are readily available by a wide
variety of procedures15,16 and since they are easily oxidized
to sulfones by a variety of selective reagents,17 such oxidation
(7) Hawthorne, D. G.; Johns, S. R.; Willing, R. I. Aust. J. Chem. 1976,
29, 315-326.
(8) Since the ring substituents in the cyclizations described in ref 6a are
very different from simple alkyl groups of ref 7 and in most cases the rings
are quite different than in the model systems,7 we consider that there is
considerable uncertainty about the stereochemistry and even the mechanism
of the Pd-catalyzed Zn-ene cyclization; see below.
(9) Trost, B. M.; Schmuff, N. R.; Miller, M. J. J. Am. Chem. Soc. 1980,
102, 5979-5981.
(10) Clayden, J.; Julia, M. J. Chem. Soc., Chem. Commun. 1994, 1905-
1906.
(11) (a) Simpkins, N. S. Sulphones in Organic Synthesis; Pergamon: New
York, 1993; pp 106, 116. (b) Kocienski, P. J. Tetrahedron Lett. 1979, 20,
441-442. (c) Phillips, E. D.; Warren, E. S.; Whitham, G. H. Tetrahedron
1997, 53, 307-320. (d) Savoia, D.; Trombini, C.; Umani-Ronchi, A. J.
Chem. Soc., Perkin Trans. 1 1977, 123-125.
(12) Trost, B. M.; Schmuff, N. R. J. Am. Chem. Soc. 1985, 107, 396-
405.
(13) Liu, C. Q.; Kudo, K.; Hashimoto, Y.; Saigo, K. J. Org. Chem. 1996,
61, 494-502.
3638
Org. Lett., Vol. 7, No. 17, 2005

is a very versatile procedure for preparing allyl phenyl
sulfones. One example, used in the preparation of the
spirocyclic alkyl iodide 10, is shown in Scheme 4. Oxidation
able synthetic effort over the past decade, and two syntheses
of this molecule have been reported to date. The first
synthesis22 used an intramolecular alkyne carbomercuration
reaction to construct the spirocarbon center with moderate
diastereoselectivity. In total, 10 steps were performed with
an overall yield of 35%. The second synthesis23 used an allyl
acetate as the substrate for the Pd-catalyzed Zn-ene cycliza-
tion to construct the spiro carbon center with excellent
diastereoselectivity. A total of 11 steps were performed with
an overall yield of 24%.
Scheme 4. Use of an Allyl Sulfone Prepared by Oxidation of
an Allyl Phenyl Sulfide
A main purpose of our attempt to synthesize 18 was to
determine any advantages of using an allyl phenyl sulfone
instead of an allyl acetate to synthesize the same target from
the same starting material, natural (-)-perillyl alcohol 11,
utilizing the same Pd-catalyzed Zn-ene cyclization strategy
as that used by the Oppolzer group23 but with an allyl phenyl
sulfone rather than an allyl acetate as a cyclization substrate.
The allylic alcohol 12 was prepared by hydrogenation of 11
(Scheme 5).22,23 Treating 12 with PhSCl24 and Et3N produced
of the readily prepared allyl phenyl sulfide 716 with m-CPBA
gave the corresponding allyl phenyl sulfone 8 in 91% yield.
Alkylation18 of the sulfone-stabilized allyl anion with
1-iodo-4-pentene19 occurred in 84% yield. Treating the
alkylation product 9 with 5 mol % Pd(PPh3)4, followed by
the addition of 6 equiv of Et2Zn, led to the generation of the
allylzinc, which readily attacked the tethered olefin. After
18 h at 25 °C, I2 was added to quench the reaction. A single
diastereomer20 of 10 was obtained in 96% yield.
Scheme 5. Synthesis of (-)-Erythrodiene Utilizing an Allyl
Sulfone Prepared by Oxidation of an Allyl Phenyl Sulfoxide
Derived from a [2,3]-Sigmatropic Rearrangement
Encouraged by the easy assembly of spirobicyclic mol-
ecule 10, we turned our attention to the synthesis of (-)-
erythrodiene 18, a sesquiterpene isolated from the Caribbean
gorgonian coral Erythropodium caribaeorum.21 The rare
spirobicylo[4.5]decane skeleton of 18 has attracted consider-
(14) Limitations of this cyclization procedure became evident when
attempts to cyclize the analogues of 3 bearing methyl groups on either
terminus of the nonallylic alkene were unsuccessful. It is conceivable that
special reaction conditions such as those required in a literature synthesis
of (-)-erythrodiene (see below) would be successful. Other conditions were
not tried.
(15) For examples, see: (a) Hopkins, P. B.; Fuchs, P. L. J. Org. Chem.
1978, 43, 1208-1213. (b) Giese, B.; Mazundar, P. Chem. Ber. 1981, 114,
2859-2865. (c) Denmark, S. E.; Weber, E. J. HelV. Chim. Acta 1983, 66,
1655-1659. (d) Binns, M. R.; Haynes, R. K.; Lambert, D. E.; Schober, P.
A. Tetrahedron Lett. 1985, 26, 3385-3388. (e) Guo, B.-S.; Doubleday,
W.; Cohen, T. J. Am. Chem. Soc. 1987, 109, 4710-4711. (f) Hannaby,
M.; Warren, S. J. Chem. Soc., Perkin Trans. 1 1989, 303-311. (g) Sato,
T.; Hiramura, Y.; Otera, J.; Nozaki, H. Tetrahedron Lett. 1989, 30, 2821-
2824. (h) Behrens, K.; Kneisel, B. O.; Noltemeyer, M.; Bruckner, R. Liebigs
Ann. Chem. 1995, 385-400. (i) Wang, X. Z.; Wu, Y. L.; Jiang, S. D.;
Singh, G. Tetrahedron Lett. 1999, 40, 8911-8914.
(16) Cohen, T.; Guo, B. S. Tetrahedron 1986, 42, 2803-2808.
(17) (a) Trost, B. M.; Curran, D. P. Tetrahedron Lett. 1981, 22, 1287-
1290. (b) Bortolini, O.; Di Furia, F.; Modena, G.; Seraglia, R. J. Org. Chem.
1985, 50, 2688-2690. (c) Williams, D. R.; Brooks, D. A.; Berliner, M. A.
J. Am. Chem. Soc. 1999, 121, 4924-4925. (d) Sato, K.; Hyodo, M.; Aoki,
M.; Zheng, X. Q.; Noyori, R. Tetrahedron 2001, 57, 2469-2476. (e) Liu,
P.; Jacobsen, E. N. J. Am. Chem. Soc. 2001, 123, 10772-10773. (f)
Choudary, B. M.; Reddy, C. R. V.; Prakash, B. V.; Kantam, M. L.; Sreedhar,
B. J. Chem. Soc., Chem. Commun. 2003, 754-755.
(18) Gais, H. J.; Gumpel, M. V.; Raabe, G.; Mueller, J.; Braun, S.;
Lindner, H. J.; Rohs, S.; Runsink, J. Eur. J. Org. Chem. 1999, 7, 1627-
1652.
(19) Padwa, A.; Kamigata, N. J. Am. Chem. Soc. 1977, 99, 1871-1880.
The use of 1-bromo-4-pentene as an alternative electrophile provided no
desired product under the same reaction conditions.
(20) Stereochemistry is assumed on the basis of the results in Scheme 3
and the reported structures of all other kinetic products of metallo-ene
cyclizations using main group organometallics.
(21) Pathirana, C.; Fenical, W.; Corcoran, E.; Clardy, J. Tetrahedron
Lett. 1993, 34, 3371-3372.
intermediate 13, which underwent a facile [2,3]-sigmatropic
rearrangement to afford allyl phenyl sulfoxide 14 in 89%
yield.24,25 Oxidation of 14 afforded the corresponding allyl
phenyl sulfone 15 in 86% yield. Alkylation of 15 as in
Scheme 4 afforded the allyl phenyl sulfone cyclization
substrate 16 in 89% yield as two diastereomers in a ratio of
4:1 (NMR). Treating 16 with Pd(PPh3)4/Et2Zn generated the
allylzinc intermediate that smoothly executed carbozincation
of the terminal alkene to afford, after an iodine quench, a
94% yield of a mixture of two diastereomers in a ratio of
95:5, a selectivity similar to that observed in the Oppolzer
synthesis using an allyl acetate as the precursor of the
allylzinc species.23 However, the use of the most common
Pd(0) source, Pd(PPh3)4, in his reaction gave a poor yield; a
special combination of Pd(OAc)2/PBu3 was required, and the
(22) Huang, H.; Forsyth, C. J. J. Org. Chem. 1995, 60, 2773-2779.
(23) Oppolzer, W.; Flachsmann, F. HelV. Chim. Acta 2001, 84, 416-
430.
(24) Hua, D. H.; Venkaraman, S.; Ostrander, R. A.; Sinai, G. Z.; McCann,
P. J.; Coulter, M. J.; Xu, M. R. J. Org. Chem. 1988, 53, 507-515.
(25) Evans, D. A.; Andrews, G. C. Acc. Chem. Res. 1974, 7, 147-155.
Org. Lett., Vol. 7, No. 17, 2005
3639

use of 20 equiv of Et2Zn was crucial in obtaining a high
yield of the cyclization product. Of considerable significance
for the comparison of allyl sulfones with allyl acetates as
substrates in these cyclizations is the finding that in our (-)-
erythrodiene synthesis (Scheme 5), using 16 as the substrate
for the Zn-ene cyclization and Pd(PPh3)4 together with 6
equiv of Et2Zn was highly successful. No effort was made
to optimize the amount of Et2Zn used.
Scheme 6. Synthesis of a Cis-3,4-disubstituted Pyrrolidine
from an Allyl Sulfone Derived by 1,2-Addition of an Amine to
a 2-Phenylsulfonyl-1,3-butadiene
Dehydroiodination of the mixture of iodides produced a
95:5 mixture of (-)-erythrodiene 18 and its diastereomer in
almost quantitative yield. A pure sample of 18, obtained from
the mixture of diastereomers by chromatography on silica
impregnated with AgNO3, had identical NMR spectroscopic
properties and rotation ([R]D -112°) to those reported.22,23
The same diastereoselectivity between iodide 17 and (-)-
erythrodiene 18 derived from it shows that the chiral center
at C7 must have been generated in a completely diastereo-
selective manner during cyclization and that the stereochem-
ical divergence comes from the relationship between the spiro
carbon center at C1 and the carbon atom C5 bearing the
isopropyl group. In total, six linear steps were used to obtain
(-)-erythrodiene starting from commercially available (-)-
perillyl alcohol with an overall yield of 60%. The synthesis
utilizing the allyl acetate23 required 11 steps starting with
the same alcohol, and the overall yield was 24%. Such a
dramatic improvement in yield and efficiency is a testament
to the power of the allyl phenyl sulfone approach to (-)-
erythrodiene and more generally demonstrates the connective
power and synthetic utility of allyl phenyl sulfones in these
reactions.
with high stereoselectivity (>40:1) as determined by GC.29
The ease of assembly of 22 and the exquisite stereoselectivity
bode well for a synthesis of the highly biologically active
(-)-kainic acid30 that is under way in this laboratory.
Because of the ease of preparation of allyl phenyl sulfones,
they are considerably superior to allyl acetates as substrates
in Pd-catalyzed Zn-ene cyclizations. In the synthesis of (-)-
erythrodiene, the sulfone method required about half of the
number of steps and led to a doubling of the yield as
compared to use of the allyl acetate substrate; in addition,
special cyclization conditions were not required as was the
case with the acetate. Several other five-membered rings,
including a pyrrolidine, were prepared from other allyl phenyl
sulfone substrates that were generated by diverse methods.
Acknowledgment. We thank the US-Israel Binational
Science Foundation (Grant 2000155), the US National
Science Foundation (Grant 0102289), and The Dreyfus
Foundation (Senior Scientist Mentor Initiative SI-03-007) for
support of this work. We also thank Professor Ilan Marek
for useful suggestions.
We envisioned that the addition of a primary amine to
2-phenylsulfonyl-3-methyl-1,3-butadiene 20 would be espe-
cially useful because the resulting product is an allyl phenyl
sulfone,26 which can be further elaborated to synthesize a
precursor for the Zn-ene cyclization. As illustrated in Scheme
6, 2027 added benzylamine smoothly to afford the desired
allyl phenyl sulfone 21 in 90% yield. N-Allylation28 provided
the cyclization substrate 22 in 98% yield. When 22 was
treated with Pd(PPh3)4/Et2Zn, the Zn-ene cyclization pro-
ceeded to afford cyclization product 23 in a moderate
unoptimized yield of 60% (68% based on consumed reactant)
Supporting Information Available: Experimental pro-
cedures and compound characterizations, including NMR
spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL051049I
(29) Lowered yield in this case is probably the result of elimination of
zinc allyl amide in the allylzinc intermediate. The latter presumably
undergoes competitive cyclization and elimination as noted in an analogous
case in which an allyl acetate substrate was used.6a Corresponding
allylmagnesium intermediates give only elimination: Oppolzer, W.; Gaudin,
J.-M.; Bedoya-Zurita, M.; Hueso-Rodriguez, J.; Robyr, C. Tetrahedron Lett.
1988, 29, 4709-4712.
(26) The largely 1,2-addition of a secondary amine to 2-phenylsulfonyl-
1,3-butadiene has been reported: Ba¨ckvall, J.-E.; Juntunen, S. K. J. Am.
Chem. Soc. 1987, 109, 6396-6403.
(27) Ba¨ckvall, J. E.; Ericsson, A. J. Org. Chem. 1994, 59, 5850-5851.
(28) Kemp, M. I.; Whitby, R. J.; Coote, S. J. Synthesis 1998, 557-568.
(30) Trost, B. M.; Rudd, M. T. Org. Lett. 2003, 5, 1467-1470 and
reference therein.
3640
Org. Lett., Vol. 7, No. 17, 2005