Regioselectivity in the palladium-catalyzed addition of carbon nucleophiles to carbocyclic derivatives

Article abstract of DOI:10.1021/jo010672n

The regioselectivity of Pd-catalyzed malonate additions and arylations to cycloalkenyl esters can be predicted by completing a stereochemical analysis of the Pd-π-allyl complex. The Pd-catalyzed malonate additions which have the greatest degree of regioselectivity are in which substituents have a steric influence in blocking the incoming nucleophile. Cyclopentenyl substrates displayed lower regioselectivity than the cyclohexyl counterparts presumably due to increased planarity of the system. Arylations using tin and hypervalent silicon reagents were compared.

Full text of DOI:10.1021/jo010672n

VOLUME 67, NUMBER 2
J ANUARY 25, 2002
© Copyright 2002 by the American Chemical Society
Articles
Regioselectivity in th e P a lla d iu m -Ca ta lyzed Ad d ition of Ca r bon
Nu cleop h iles to Ca r bocyclic Der iva tives‡
Molly E. Hoke, Marc-Raleigh Brescia, Suzanne Bogaczyk, and Philip DeShong*
Department of Chemistry and Biochemistry, The University of Maryland, College Park, Maryland 20742
Bryan W. King and Michael T. Crimmins§
Department of Chemistry and Biochemistry, The University of North Carolina at Chapel Hill,
Venable and Kenan Laboratories CB 3290, Chapel Hill, North Carolina 27599-3290
pd10@umail.umd.edu
Received J uly 3, 2001
The regioselectivity of Pd-catalyzed malonate additions and arylations to cycloalkenyl esters can
be predicted by completing a stereochemical analysis of the Pd-π-allyl complex. The Pd-catalyzed
malonate additions which have the greatest degree of regioselectivity are in which substituents
have a steric influence in blocking the incoming nucleophile. Cyclopentenyl substrates displayed
lower regioselectivity than the cyclohexyl counterparts presumably due to increased planarity of
the system. Arylations using tin and hypervalent silicon reagents were compared.
In tr od u ction
of configuration at the least hindered site of the inter-
mediate π-allyl complex.8
Palladium-catalyzed allylic substitution is an efficient
and highly stereoselective method for formation of carbon-
carbon bonds, and the alkylation with stabilized carbon
nucleophiles has been widely applied in organic syn-
thesis.1-8 The major characteristic of alkylation using
stabilized nucleophiles is that it proceeds with retention
Our group9 and others3,10-12 have investigated factors
controlling the regioselectivity of Pd(0)-catalyzed alkyl-
ations of heterocyclic (i.e., carbohydrate and pyranose)
derivatives. In many cases, it was found that both elec-
(6) (a) Stille, J . K. Pure Appl. Chem. 1985, 57, 1771-1780. (b) Stille,
J . K.; Hegedus, L. S.; Del Valle, L. J . Org. Chem. 1990, 55, 3019-
3023.
(7) Farina, V.; Krishnamurthy, V.; Scott, W. J . Organic Reactions;
J ohn Wiley & Sons: New York, 1997; Vol. 50, pp 1-652.
(8) (a) Trost, B. M. Tetrahedron 1977, 33, 2615-2649. (b) Trost, B.
M. Pure Appl. Chem. 1979, 51, 787-800. (c) Trost, B. M. Acc. Chem.
Res. 1980, 13, 385-393. (d) Trost, B. M.; Van Vranken, D. L. Chem.
Rev. 1996, 96, 395-422.
(9) (a) Brescia, M.-R.; Shimshock, Y. C.; DeShong, P. J . Org. Chem.
1997, 62, 1257-1263. (b) Class, Y. J .; DeShong, P. Tetrahedron Lett.
1995, 36, 7631-7634.
(10) Curran, D. P.; Suh, Y.-G. Carbohydr. Res. 1987, 171, 161-191.
(11) Dunkerton, L. V.; Serino, A. J . J . Org. Chem. 1982, 47, 2812-
2814.
* To whom correspondence should be addressed. Phone: 301-405-
1892. Fax: 301-314-9121.
‡
This paper is dedicated to Richard W. Franck on the occasion of
his 65th birthday.
§ Phone: 919-962-2172. Fax: 919-962-2388. E-mail: crimmins@
email.unc.edu.
(1) Diederich, F.; Stang, P. J .; Eds. In Metal-catalyzed Cross-coupling
Reactions; Wiley-VCH: Weinheim, 1998.
(2) Tsuji, J . Palladium Reagents and Catalysts. Innovations in
Organic Synthesis; J ohn Wiley & Sons: New York, 1995; pp 290-422.
(3) Godleski, S. A. In Comprehensive Organic Synthesis; Trost, B.
M.; Ed.; Pergamon Press: New York, 1991; Vol. 4, pp 585-661.
(4) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457-2483.
(5) Hegedus, L. S. Coord. Chem. Rev. 1996, 147, 443-545.
(12) Baer, H. H.; Hanna, Z. S. Can. J . Chem. 1981, 59, 889-906.
10.1021/jo010672n CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/20/2001

328 J . Org. Chem., Vol. 67, No. 2, 2002
Hoke et al.
Sch em e 1
Sch em e 3
Sch em e 2
meric substitution products. Allylic benzoates such as 5
and 6 are known to provide “symmetrical” Pd-π-allyl
complex 7.2 That is, the metal resides symmetrically over
the allyl function due to the substituent residing on the
opposite face of the complex.Subsequent malonate addi-
tion to this “symmetrical” complex should occur with high
stereoselectivity at C-1 due to the steric bias provided
by the methyl substituent during approach of the nu-
cleophile. Arylation of this “symmetrical” complex, on the
other hand, should occur with low stereoselectivity due
to the altered mechanism involved in this process, i.e.,
aryl transfer to the metal followed by reductive elimina-
tion. Since the substituent is on the opposite face of the
ring, it does not bias the arylation process.
tronic and steric factors influenced the regioselectivity
of the reaction. However, a systematic investigation of
the regioselectivity of the substitution of cyclohexenyl and
cyclopentenyl systems has not been reported. In this
paper, we report a systematic study of the regioselectivity
and stereoselectivity of Pd(0)-catalyzed alkylation reac-
tions using both stabilized nucleophiles and silyl deriva-
tives. In addition, the effectiveness of tin and hypervalent
silicon reagents as arylation agents in the substitution
reaction will be compared and contrasted.13
Pd-catalyzed substitution of a cyclohexenyl allylic ester
such as 1 with stabilized nucleophiles (i.e., malonate) is
known to occur with overall retention of configuration.2
Employing the standard mechanism for this process, the
reaction proceeds via displacement of the benzoate by Pd
with inversion of configuration to provide Pd-π-allyl
complex 2. Attack of the stabilized nucleophile onto this
intermediate occurs on the face opposite to Pd resulting
in overall retention of configuration (Scheme 1). On the
other hand, arylation of complex 2 using silyl (or stannyl)
derivatives occurs with overall inversion presumably due
to aryl transfer from silicon to Pd followed by reductive
elimination.
Resu lts a n d Discu ssion
Benzoates 5 and 6 were individually subjected to a
standard set of alkylation conditions employing malonate,
and, as predicted by the model, gave diester 8 as the
major product along with a small amount of diester 9.
The structures of the regioisomers produced in each
alkylation reaction were confirmed by homonuclear de-
coupling experiments.
In the arylation studies, cis-1,4-allylic benzoate 5
(Scheme 3) again gives trans Pd-π-allyl intermediate 7
(Scheme 2) and therefore it would be expected that
reductive elimination of the phenyl group could occur
equally toward C-1 or C-3. Using standard Stille protocol6b
(top, Scheme 3) gave an equal distribution of regioisomers
10 and 11 (resulting from attack at C-1 and C-3,
respectively). The structures of the regioisomers of each
arylation product were confirmed by homonuclear de-
coupling experiments. The stereochemistries of the aryl-
ation products were determined by reduction (5% Pd/C)
and comparing the experimental spectral data to pub-
lished spectral data.19
The arylation reactions were also performed using tetra-
butylammonium triphenyldifluorosilicate (TBAT, Figure
The complex formed from benzoate 1 is symmetrical
with respect to the methyl group and Pd and there is no
regiochemistry associated with the substitution. Substi-
tution of unsymmetrical complexes such as complex 7
(Scheme 2) is expected to provide a mixture of regioiso-
(14) It should be noted that structures depicted in the cyclohexenyl
series in this article are not absolute stereochemistries, but are relative
stereochemistries. For the cyclopentenyl series, the structures are
absolute, and the compounds were enantiomerically pure.
(15) DeShong, P.; Handy, C. J .; Lam, Y.-F. J . Org. Chem. 2000, 65,
3542-3543.
(13) For more information on the use of silicon reagents as sources
of arylation, see: (a) Denmark, S. E.; Wu, Z. Org. Lett. 1999, 1, 1495-
1498. (b) Shibata, K.; Miyazawa, K.; Goto, Y. Chem. Commun. 1997,
1309-1310. (c) Corriu, R. J . P.; Chuit, C.; Reye, C.; Young, J . C. Chem.
Rev. 1993, 93, 1371-1448. (d) Horn, A. H. Chem. Rev. 1995, 95, 1317-
1350. (e) Nolan, S. P.; Lee, H. M. Org. Lett. 2000, 2, 2053-2055. (f)
Correia, R.; DeShong, P. J . Org. Chem. 2001, 66, 7159-7165.
(16) Brescia, M.-R.; DeShong, P. J . Org. Chem. 1998, 63, 3156-
3157.
(17) Mowery, M. E.; DeShong, P. J . Org. Chem. 1999, 64, 3266-
3270.
(18) Definitive information as to why TBAT gives a more regio-
selective reaction is not available as of yet.

Addition of C-Nucleophiles to Carbocycles
J . Org. Chem., Vol. 67, No. 2, 2002 329
Sch em e 5
F igu r e 1. Tetrabutylammonium triphenyldifluorosilicate
(TBAT).
Sch em e 4
methyl group had an indirect role by distorting the
position of the metal in the Pd-π-allyl complex.
Regioselectivity of the arylation of trans-1,4-allylic
benzoate 12 (Scheme 5) was also investigated and found
to be significantly more regioselective than the cis-isomer.
Employing the proposed model for regioselectivity in
malonate additions, the Pd-π-allyl complex formed from
benzoate 12 is distorted due to steric interactions be-
tween the methyl group and the metal-ligand moiety
(14, Scheme 4). Accordingly, reductive elimination of the
aryl group from Pd in complex 14 would occur preferen-
tially distal to the methyl group at C-3 to provide alkene
17 as the major regioisomer (Scheme 5). The arylation
was also performed using Stille conditions to give a 5:1
ratio of regioisomers 17 and 18, respectively (Scheme 5).
Again, the reaction employing TBAT gave higher yield
of the adducts, in addition to being more regioselective
than the stannane.18
As part of our ongoing project to apply this methodol-
ogy to the synthesis of alkaloidal natural products, we
investigated the coupling reaction with carbonate 19.
This more complex cyclohexenyl system was studied to
determine whether the established model was valid with
heterosubstituents. Carbonate 19 was synthesized by
Diels-Alder cycloaddition of R-chloronitrosocyclohexane
with 1,4-cyclohexadiene followed by reduction of the N,O-
bond and functionalization of the resulting amino and
hydroxyl groups.21 On the basis of the analogues dis-
cussed above, it was anticipated that coupling reactions
would proceed via “symmetrical” π-allyl complex 20
(Scheme 6). Malonate addition to carbonate 19 occurred
regiospecifically to afford diester 21. In this instance, the
reaction was even more regioselective than the methyl
counterpart (compare to benzoate 5 in Scheme 2). In our
systems, steric factors are apparently the dominant factor
that controls whether “distorted” or “symmetrical” π-allyl
complexes are formed. It should be noted that Szabo´ and
Ba¨ckvall have demonstrated that electronic effects of
substitutents on the face opposite the palladium also play
a role in the outcome of the nucleophilic attack on π-allyl
1).15 Previous studies in our lab have shown that TBAT
is an effective cross coupling reagent for allylic benzo-
ates16 as well as aryl iodides, bromides, and triflates.17
The arylation of allylic benzoate 5 (bottom, Scheme 3)
was performed with TBAT, and in this case a ratio of
1:2 of phenylated adducts 10 and 11 were obtained. In
this case, there is a slight preference for formation of the
regioisomer resulting from attack at C-3.18 It should be
noted that the TBAT reaction is higher yielding and more
regioselective than the Stille counterpart.
The corresponding set of trans-allylic benzoates 12 and
13 (Scheme 4) was also synthesized.20 In this case, the
intermediate Pd-π-allyl complex 14 is syn. It was
anticipated that there would be a steric interaction
between the methyl group and Pd, and that the Pd-π-
allyl complex would be distorted away from the methyl
group in an attempt to minimize this steric interaction
(Scheme 4). The complex is best described as σ-η2. It was
predicted that attack of the distorted Pd-π-allyl complex
14 would occur in an SN2′-like fashion, and attack at C-3
would be favored to give the 1,2-regioisomer 16 as the
major product. Benzoates 12 and 13 were individually
subjected to a standard set of alkylation conditions, and
both gave a 1:4 ratio of diesters 15 and 16. The major
regioisomer resulted from attack at C-3, as expected from
the proposed model. Malonate additions in the cis series
of allylic benzoates gave a higher degree of regioselec-
tivity (Scheme 2) than the corresponding trans series
(Scheme 4). In the cis series, the methyl group played a
direct role in influencing the regioselectivity of attack by
the incoming malonate anion. In the trans series, the
(21) For the synthesis of R-chloronitrosocyclohexane using hypochlo-
rous acid, see Corey, E. J .; Estreicher, H. Tetrahedron Lett. 1980, 21,
1117-1120. The preparation of the amino alcohol derivative preceding
carbamate 19 is described in Nelsen, S. F.; Thompson-Colon, J . A.;
Kirste, B.; Rosenhouse, A.; Kaftory, M. J . Am. Chem. Soc. 1987, 109,
7128-7136. Protection of the amino alcohol derivative is performed
using ethyl chloroformate as described in Hall, A.; Bailey, P. D.; Rees,
D. C.; Rosair, G. M.; Wightman, R. H. J . Chem. Soc., Perkin Trans. 1
2000, 329-343.
(19) 1H and 13C NMR spectral data for cis- and trans-4-methyl-1-
phenylcyclohexanes can be found in Eliel, E. K.; Manoharan, M. J .
Org. Chem. 1981, 46, 1959-1962. Actual spectral data for these
compounds is shown in Garbisch, E. E., J r.; Patterson, D. B. J . Am.
Chem. Soc. 1963, 85, 3228-3231.
(20) The stereochemistries of allylic benzoates
5 and 12 were
indirectly determined by comparing the preparation methods with the
preparation method of crystalline 1,4-trans allylic p-nitrobenzoate.

330 J . Org. Chem., Vol. 67, No. 2, 2002
Hoke et al.
Sch em e 6
Sch em e 7
complexes.22 They too have proposed the intermediacy
of “distorted” palladium complexes.
Having established a model that explained the regio-
selectivity of Pd(0)-catalyzed alkylation and arylation of
cyclohexenyl derivatives, we studied cyclopentenyl sub-
strates to determine if the model would also be applicable
to these systems. On the basis of the model, cis-allylic
benzoate 26 was predicted to produce “symmetrical” Pd-
π-allyl complex 27 (Scheme 7). Subsequent malonate
addition to 27 was expected to occur selectively at C-1 to
avoid steric interactions with the bulky ester side chain
at C-4. In fact, malonate addition to dibenzoate 26 gave
exclusively the cis-1,4-cyclopentene adduct 28.
On the other hand, arylation of carbonate 19 with a
variety of silicon-based aryl transfer reagents23 displayed
no selectivity and gave a 1:1 mixture of regioisomeric
adducts in all cases (Scheme 6). The lack of regioselec-
tivity in the arylation reaction was consistent with the
intermediacy of the “symmetrical” Pd-complex.
(22) J onasson, C.; Kritikos, M.; Ba¨ckvall, J .-E.; Szabo´, K. Chem. Eur.
J . 2000, 6, 432-436, and references therein.
(23) (a) Manoso, A. S.; DeShong, P. J . Org. Chem. 2001, 66, 7449-
7455. (b) DeShong, P.; Mowery, M. E. J . Org. Chem. 1999, 64, 1684-
1688. (c) DeShong, P.; Mowery, M. E. Org. Lett. 1999, 1, 2137-2140.
(d) DeShong, P.; Handy, C. J .; Mowery, M. E. Pure Appl. Chem. 2000,
72, 1655-1658.
Arylation reactions of cis-dibenzoate 26 were expected
to display limited regioselectivity due to the “sym-
metrical” Pd-complex (27, Scheme 7). Arylation of 26

Addition of C-Nucleophiles to Carbocycles
J . Org. Chem., Vol. 67, No. 2, 2002 331
Sch em e 8
Sch em e 9
using standard Stille protocol6b gave a 1:1 ratio of
arylated adducts 29 and 30. Similarly, arylation of
dibenzoate 26 utilizing TBAT gave an identical distribu-
tion of products (Scheme 7).
with strong regioselectivity for coupling at C-1 since
reductive elimination of the aryl group from the metal to
the substrate must proceed distal to the bulky ester side
chain at C-4, irrespective of whether the benzoate group
is ligated to Pd. In other words, Pd-complex 38 is a
“symmetrical” complex with regard to malonate addition,
but is “distorted” when considering the arylation reaction.
This hypothesis was supported by the subsequent results.
Arylation with TBAT showed excellent regioselectivity
with the C-1 adduct as the major product (Scheme 8).
Stille arylation of dibenzoates 31 and 32 (Scheme 8)
also gave a 1:20 ratio of arenes 36 and 37 (arylation at
C-3 and C-1, respectively). However, the Stille protocol
also gave a mixture of arenes 29 and 30 resulting from
a loss of stereochemical integrity of the Pd-π-allyl
complex according to the process outlined in Scheme 9.24
On the basis of the precedents of Cook and Ba¨ckvall, it
is proposed that Pd(0) attacks complex 33 to give dia-
stereomeric complex 27 (Scheme 9). If the rate of aryla-
tion of complex 33 is faster than the rate of conversion
to 27, then the arylation will occur stereospecifically to
afford cis-isomers 29 and 30, respectively. However, if
the rates of arylation and Pd attack are comparable, then
mixtures of regio- and stereoisomers are obtained. These
Unexpected results were obtained when coupling reac-
tions of trans allylic benzoates were studied. Coupling
reactions of trans-dibenzoates 31 and 32 proceeded via
the “distorted” Pd-π-allyl complex 33 (Scheme 8). Ac-
cordingly, it was anticipated that malonate addition
would occur with high regioselectivity at C-3. To our
surprise, the mixture of benzoates 31 and 32 gave a 1:1
mixture of adducts 34 and 35 (Scheme 8). This result is
consistent with a “symmetrical”, rather than “distorted”,
intermediate Pd-π-allyl complex. A possible explanation
for this unusual result involves ligation of the Lewis-basic
ester functionality to the electron-deficient palladium to
form a “symmetrical” Pd-π-allyl complex as depicted in
Figure 2.
Even though malonate addition to trans allylic ben-
zoates 31 and 32 appeared to occur via a “symmetical”
Pd-π-allyl complex (38, Figure 2), we anticipated that
arylation of benzoates 31 and 32 would afford adducts
(24) This loss of stereochemical integrity is best explained by a
second Pd coming in and attacking the Pd-π-allyl complex. Fore more
information regarding the phenomenon, please see: (a) Cook, G. R.;
Shanker, P. S.; Pararajasingham, K. Angew. Chem., Int. Ed. 1999, 38,
110-112. (b) Granberg, K. L.; Ba¨ckvall, J .-E. J . Am. Chem. Soc. 1992,
114, 6858-6863. (c) Ba¨ckvall, J .-E.; Granberg, K.; Heumann, A, Isr.
J . Chem. 1991, 31, 17-24.
F igu r e 2. “Symmetrical” Pd-π-allyl complex.

332 J . Org. Chem., Vol. 67, No. 2, 2002
Hoke et al.
Sch em e 10
results indicate that the aryl transfer from silicates to
Pd proceeded faster than the tin counterparts.
Exp er im en ta l Section
Gen er a l Meth od s. Nuclear magnetic resonance (1H and
13C NMR) spectra were recorded on a 200 or 400 MHz
spectrometer in CDCl3. Chemical shifts are reported in parts
per million (∂) relative to the nondeuterated solvent peak.
Coupling constants (J values) are reported in hertz (Hz), and
spin multiplicities are indicated by the following symbols: s
(singlet), d (doublet), t (triplet), q (quartet), m (multiplet), br
s (broad singlet). Infrared spectra were recorded as solutions
in CCl4. Band positions are given in reciprocal centimeters
(cm-1), and relative intensities are listed as: br (broad), s
(strong), m (medium), or w (weak). Thin-layer chromatography
(TLC) was performed with the compounds being identified in
one or more of the following manners: UV (254 nm), iodine,
or vanillin/sulfuric acid charring. Flash chromatography data
is reported as (column diameter in mm, column height in cm,
solvent). Et2O was distilled from sodium/benzophenone ketyl
while methylene chloride (CH2Cl2) was distilled from calcium
hydride. Dimethylformamide (DMF) was distilled from 4Å
Linde-type molecular sieves. Reagent grade methanol was
stored over 4Å Linde-type molecular sieves. Absolute ethanol
(EtOH) was used without further purification. Glassware used
in the reactions described below was dried for a minimum of
12 h in an oven at 120 °C. All reactions were run under an
atmosphere of N2 at room temperature unless otherwise noted.
All ratios given were determined by GC or 1H NMR unless
otherwise indicated.
We had proposed that ligation of the benzoate func-
tionality in complex 38 was responsible for the lack of
regioselectivity observed in malonate addition to this
complex (see Scheme 8 and Figure 2). Silyl-protected
primary alcohol derivatives 39 and 40, derivatives that
should not be able to serve as ligands for the metal, were
subjected to additions, and the results are summarized
in Scheme 10. With the side chain unable to participate
via coordination to Pd, the Pd-π-allyl intermediate
predicted from the model should be “distorted” (Scheme
10) and malonate anion should couple preferentially at
C-3. However, reaction with malonate afforded a 1:1
mixture of adducts 42 and 43, exactly the same results
that were obtained with the benzoate derivative (Scheme
8).25 These results suggest that malonate addition in the
cyclopentenyl system does not proceed through a “dis-
torted” Pd-π-allyl complex as anticipated. Therefore,
conformational factors of the intermediate cyclopentenyl
complex must play a more significant role than has been
anticipated in the model outlined above.
Con clu sion s
Benzoic acid (PhCO2H), benzoyl chloride (BzCl), ethyl chlo-
roformate, bromine (Br2), tert-butyldimethylsilyl chloride (TB-
DMS-Cl), diethyl malonate, diphenylphosphinoethane (dppe),
lithium chloride (LiCl), 4-methylcylcohexanone, p-nitrobenzoyl
chloride (p-NBzCl), palladium acetate (Pd(OAc)2), palladium
on carbon (5%, Pd/C), sodium borohydride (NaBH4), sodium
hydride (60% dispersion in oil, NaH), tetrakis(triphenylphos-
phine)palladium(0) (Pd(PPh3)4), and trimethyl(phenyl)tin (Ph-
SnMe3) were purchased from Aldrich and used as received.
Bis(dibenzylidene acetone)palladium (Pd(dba)2) and diethyl
azodicarboxylate (DEAD) were purchased from Acros and used
as received. Lithium carbonate (Li2CO3) was purchased from
Baker and used as received. Fluorosilicic acid (H2SiF6) was
purchased from Fischer and used as received. Triphenylphos-
phine (PPh3) was purchased from Aldrich and recrystallized
from pentane prior to use. Tetrabutylammonium triphenyldi-
fluorosilicate (TBAT) was synthesized according to the litera-
ture procedure.15 The corresponding diol precursor for diben-
zoate 26 was provided by Prof. Michael Crimmins.
Diest er s 8 a n d 9. Usin g P d (P P h 3)4. Su bst r a t e: cis-4-
Meth ylcycloh ex-2-en -1-yl Ben zoa te (5). Sodium hydride
(0.12 g of a 60% dispersion in oil, 2.9 mmol) was washed with
2 × 2 mL of hexane and 2 mL of THF. To a suspension of oil-
free NaH in 3 mL of THF was added 0.90 mL (5.9 mmol) of
diethyl malonate. The resulting diethyl malonate anion was
added via syringe to a degassed solution of 0.10 g (0.46 mmol)
of allylic benzoate 5, 48 mg (0.042 mmol) of Pd(PPh3)4, and
0.11 g (0.42 mmol) of PPh3 in 11 mL of THF. The resulting
yellow solution was heated at 78 °C. The reaction mixture was
The results summarized above indicate that steric
factors in the intermediate Pd-π-allyl complexes exert
significant control of the regioselectivity of Pd(0)-cata-
lyzed malonate additions and arylations of cyclohexenyl
and cyclopentenyl derivatives. Substituents on the carbo-
cyclic systems lead to formation of either “symmetrical”
or “distorted” Pd-complexes. The regioselectivity of the
subsequent attack on the complex (i.e., malonate addition
or arylation) can be correctly predicted in most instances
using mechanistic considerations. The resulting model
performs well with cyclohexenyl derivatives, but is
limited with cyclopentenyl analogues due to conforma-
tional factors that have yet to be fully understood. We
are currently investigating the conformations of the inter-
mediate cyclohexenyl and cyclopentenyl Pd-complexes
computationally, and the results will be reported at a
later date.
(25) The regiochemistries of the products 42 and 43 formed from
the malonate addition shown in Scheme 10 were determined in the
following manner: Diesters 42 and 43 were deprotected using HF and
H2SiF6 to give a 1:1 ratio of alcohols. The alcohols were then esterified
with benzoyl chloride to give a 1:1 mixture of previously characterized
diesters 34 and 35 (Scheme 8). These diesters were identical by 1H
NMR to the diesters previously synthesized in Scheme 8. The
procedures and yields are fully explained in the Experimental Section
under the heading “Diesters 42 and 43 using Pd(PPh3)4”.

Addition of C-Nucleophiles to Carbocycles
J . Org. Chem., Vol. 67, No. 2, 2002 333
quenched after 25 h. with 40 mL of H2O. The layers were
separated, and the aqueous phase was extracted with 3 × 50
mL of Et2O. The combined organic layers were dried over Na2-
SO4 and concentrated in vacuo. Purification of the residue by
flash chromatography (15 mm, 20 cm, 5% EtOAc/hexane) gave
65 mg (55%) of diesters 8 and 9 as a colorless oil. The ratio of
diesters 8 to 9 was 14:1. Approximately 8% of diester 16 was
obtained and this was attributed to the presence of 8% of
benzoate 12 in the starting material. TLC Rf ) 0.62 (20%
EtOAc/hexane); IR (CCl4) 3025 (w), 2988 (s), 2963 (s), 2931
(s), 2868 (s), 1750 (s), 1738 (s), 1656 (w), 1481 (m), 1463 (m),
1400 (m), 1369 (s), 1338 (m), 1300 (s), 1275 (s), 1244 (s), 1231
(s), 1156 (s), 1181 (s), 1100 (s), 1038 (s); 1H NMR (CDCl3, major
isomer 8) 0.95 (d, 3H, J ) 7.1), 1.01-2.15 (m, 11H), 2.83-
2.85 (br m, 1H), 3.24 (d, 1H, J ) 9.8), 4.17 (q, 4H, J ) 7.1),
5.51 (B of ABq, 1H, J AB) 10.2), 5.66 (A of ABq, 1H, J AB)10.2);
13C NMR (CDCl3, major isomer 8) 14.1, 21.0, 23.0, 23.9, 27.9,
29.7, 57.1, 61.0, 126.6, 135.6, 168.5, 168.6; GC/MS (diester 8)
254 (M+, 6), 180 (100), 161 (41), 151 (47), 134 (24), 115 (44),
107 (28), 95 (73), 94 (75), 91 (24), 79 (56), 77 (29); (diester 9)
254 (M+, 5), 180 (100), 161 (30), 152 (37), 115 (35), 107 (30),
95 (71), 79 (37).
Su bstr a te: cis-6-Meth ylcycloh ex-2-en -1-yl Ben zoa te
(6). The preparation of diesters 8 and 9 was performed in the
same manner as described for the cis-1,4-benzoate 5. The
Pd(0)-catalyzed addition of the diethyl malonate anion to 12
mg (0.055 mmol) of allylic benzoate 6 gave 10 mg (71%) of
diesters 8 and 9 as a colorless oil following, and the ratio of
diesters 8 to 9 was 20:1. Approximately 4% of diester 16 was
obtained, and this was attributed to the presence of 4% of
benzoate 13 in the starting material. The spectral data was
identical to the above-reported results.
Approximately 5% of diester 8 was observed due to the
presence of 5% of benzoate 5 in the starting material. Integra-
tion of the doublets at 3.17 (diester 15) and 3.46 (diester 16)
1
of the malonate proton in the H NMR spectrum gave a ratio
of 1:3.8. TLC Rf ) 0.77 (33% EtOAc/hexane); IR (CDCl3) 3038
(w), 2988 (m), 2962 (m), 2931 (m), 2875 (m), 2856 (w), 1759
(s), 1731 (s), 1606 (w), 1462 (w), 1369 (m), 1361 (m), 1269 (m),
1
1231 (m), 1181 (s), 1156 (s), 1106 (s); H NMR (CDCl3, major
isomer 16) 0.99 (d, 3H, J ) 6.6), 1.17-1.99 (m, 11H), 2.53 (br
s, 1H), 3.47 (d, 1H, J ) 7.1), 4.14-4.42 (m, 4H), 5.58 (B of
ABq, 1H, J AB) 10.2), 5.74 (A of ABq, 1H, J AB)10.2) 13C NMR
(CDCl3, major isomer 16) 14.1, 19.3, 23.0, 27.8, 30.2, 41.8, 55.2,
61.3, 126.3, 128.7, 168.4, 169.1; GC/MS (diester 16) 254 (M+,
5), 209 (3), 180 (100), 161 (40), 151 (40), 133 (21), 115 (54), 95
(94), 94 (47), 79 (74); (diester 15) 254 (M+, 4), 209 (1), 180 (37),
161 (34), 151 (20), 133 (13), 115 (30), 95 (58), 94 (100), 79 (61),
77 (18).
Su bstr a te: tr a n s-6-Meth ylcycloh ex-2-en -1-yl Ben zoa te
(13). The preparation of diesters 15 and 16 was performed in
the same manner as for diesters 8 and 9 (benzoate 5). The
Pd(0)-catalyzed addition of the diethyl malonate anion to 20
mg (0.092 mmol) of allylic benzoate 13 gave 16 mg (67%) of
diesters 15 and 16 as a colorless oil following purification by
column chromatography (15 mm, 17 cm, 12.5% EtOAc/hexane).
The ratio of diesters 15 to 16 was 1:4 by GC. Approximately
11% of diester 8 was observed due to a 5% impurity of benzoate
6 in the starting material. Integration of the doublets at 3.18
(diester 15) and 3.47 (diester 16) of the malonate proton gave
a ratio of 1:5.0. The spectral data was identical to the above-
reported results.
Ar en es 17 a n d 18. Usin g P h Sn Me3. Using the previously
described procedure for the synthesis of arenes 10 and 11 using
PhSnMe3, but substituting benzoate 12 gave 15 mg (64%) of
arenes 17 and 18 as a 5:1 ratio by GC. TLC Rf ) 0.58 (hexanes);
IR (CCl4) 3088 (w), 3063 (w), 3025 (m), 2963 (s), 2925 (s), 2875
(m), 2856 (m), 1563 (s), 1544 (s), 1456 (w), 1250 (m), 1219 (m),
1138 (m), 1117 (m), 1081 (m), 1006 (s), 981 (m), 838 (s); 1H
NMR (CDCl3)1.04 (d, 3H, J ) 7.1), 1.63-1.69 (m, 2H), 1.87-
1.93 (m, 2H), 2.24 (br s, 1H), 3.37 (br s, 1H), 5.73 (dt, 2H, J )
39.8, 10.0, 2.5) 7.15-7.35 (m, 5H); 13C NMR 21.4, 27.7, 29.6,
29.7, 41.1, 125.9, 128.0, 128.2, 128.6, 134.7, 151.0; GCMS
(arene 18) 172 (M+, 42), 157 (11), 131 (19), 130 (100), 129 (72),
128 (23), 115 (38), 91 (21); (arene 17) 172 (M+, 81), 157 (33),
143 (29), 130 (85), 129 (92), 128 (30), 117 (18), 115 (52), 104
(100), 91 (37), 77 (19), 51 (21).
Ar en es 10 a n d 11. Usin g P h Sn Me3. To a yellow mixture
of 30 mg (0.14 mmol) of allylic benzoate 5, 6 mg (10 µmol) of
Pd(dba)2, and 24 mg (0.57 mmol) of LiCl in 3 mL of DMF was
added 40 µL (0.22 mmol) of PhSnMe3. The mixture was
degassed via a single freeze-pump-thaw cycle. The brown
solution was heated at 55 °C for 21 h. The resulting dark-
brown suspension was quenched by the addition of 50 mL of
H2O and extracted with 4 × 50 mL of Et2O. The combined
organics were dried over MgSO4 and concentrated in vacuo.
Column chromatography (15 mm, 21 cm, hexanes) gave 17 mg
(72%) of arenes 10 and 11 with a ratio of 1:1. TLC Rf ) 0.53
(hexanes); IR (CCl4, arene 10 only) 3063 (w), 3025 (m), 2963
(s), 2931 (s), 2863 (s), 1681 (m), 1514 (s), 1456 (m), 1256 (m),
1
1113 (m), 1013 (m), 900 (s); H NMR (arene 10 only) 1.01 (d,
Usin g TBAT. Using the previously described procedure for
the synthesis of arenes 10 and 11 using TBAT, but substituting
benzoate 12, gave 43 mg (91%) of arenes 17 and 18 as a 10:1
mixture. The spectral data was identical to the preceding
results.
3H, J ) 7.1), 1.17-2.24 (m, 5H), 3.32 (br s, 1H), 5.66 (dd, 2H,
J ) 19.0, 10.6), 7.15-7.35 (m, 5H); 13C NMR (arene 10 only)
21.9, 30.4, 31.4, 32.7, 42.4, 126.0, 127.6, 128.3, 129.7, 134.5,
146.8; 13C NMR (arene 11 only) 20.2, 25.3, 30.6, 36.7, 50.6,
128.1, 128.2, 128.3, 130.9, 135.8, 145.9; GCMS (arene 11) 173
172 (M+, 37), 157 (8), 131 (17), 130 (100), 129 (68), 128 (23),
115 (36), 91 (21), 77 (10), 51 (12); (arene 10) 172 (M+, 66), 157
(28), 143 (28), 130 (85), 124 (86), 128 (30), 117 (18), 115 (51),
104 (100), 91 (39), 79 (12), 78 (11), 77 (19), 65 (13), 51 (19).
Usin g TBAT. A solution of 17 mg (30 µmol) of Pd(dba)2,
9.0 mg (34 µmol) of PPh3, and 60 mg (0.28 mmol) of allylic
benzoate 5 in 5 mL of THF was degassed via a single freeze-
pump-thaw cycle. Then, 0.32 g (0.60 mmol) of TBAT was
added to the reaction which was degassed by a second freeze-
pump-thaw cycle. The dark yellow solution was heated at 70
°C for 17 h. The reaction was quenched by the addition of 50
mL of H2O and extracted with 4 × 50 mL of Et2O. The
combined organics were dried over MgSO4 and concentrated
in vacuo. Purification of the residue by column chromatogra-
phy (20 mm, 20 cm, hexanes) gave 45 mg (94%) of arenes 10
and 11 as a 1:2 mixture. The spectral data was identical to
the preceding results.
Diester 21. Sodium hydride (93 mg of a 60% dispersion in
oil, 2.33 mmol) was washed with 2 × 2 mL of hexane and 2
mL of THF. To a suspension of oil-free NaH in 3 mL of THF
was added 355 µL (2.33 mmol) of diethyl malonate. The
resulting diethyl malonate anion was added via syringe to a
degassed solution of 200 mg (0.778 mmol) of carbonate 19 and
40 mg (0.039 mmol) of Pd2(dba)3‚CHCl3 in 10 mL of THF. The
resulting yellow solution was heated at 50 °C. The reaction
mixture was quenched after 25 h with 40 mL of H2O. The
layers were separated, and the aqueous phase was extracted
with 3 × 40 mL of Et2O. The combined organic layers were
dried over Na2SO4 and concentrated in vacuo. Purification of
the residue by flash chromatography (15 mm, 20 cm, 3:1
hexanes:EtOAc) gave 150 mg (60%) of diester 21 as a colorless
oil. TLC Rf ) 0.19 (3:1 hexanes:EtOAc); IR (CCl4) 3455 (w) 2960
(m), 2922 (m), 2871 (m), 1736 (m), 1459 (m), 1242 (m); 1H NMR
(CDCl3) 1.18-1.24 (m, 9H), 1.43-1.49 (m, 1H), 1.67-1.73 (m,
3H), 2.76-2.82 (m, 1H), 3.25 (d, 1H, J ) 7.4), 4.04 (q, 2H, J )
Diester s 15 a n d 16. Usin g P d (P P h 3)4. Su bstr a te: tr a n s-
4-Meth ylcycloh ex-2-en -1-yl Ben zoa te (12). The preparation
of diesters 15 and 16 was performed in the same manner as
described for diesters 8 and 9 (benzoate 5). The Pd(0)-catalyzed
addition of the diethyl malonate anion to 0.11 g (0.527 mmol)
of allylic benzoate 12 gave 0.11 g (87%) of diesters 15 and 16
as a colorless oil. The ratio of diesters 15 to 16 was 1:4.
7.0), 4.06-4.17 (m, 5H), 4.76 (d, 1H, J ) 8.0), 5.70 (s, 2H); 13
C
NMR (CDCl3) 14.1, 14.6, 22.2, 27.7, 35.0, 44.7, 56.1, 60.6, 61.4,
128.8, 131.5, 155.7, 168.1, 168.2; FAB mass spectrum m/z
(relative intensity) 328 ((M + H), 9), 239 (28), 161 (100), 79
(38); HRMS (FAB) calcd for C16H26O6N (M + H) 328.1760,
found 328.1773.

334 J . Org. Chem., Vol. 67, No. 2, 2002
Hoke et al.
Ar en es 22 a n d 23. Usin g P h Si(OEt)3. To a solution of 100
mg (0.389 mmol) of carbonate 19 and 20 mg (0.020 mmol)
Pd2(dba)3‚CHCl3 in 10 mL THF was added 190 µL (0.778
mmol) PhSi(OEt)3, followed by 778 µL (0.778 mmol, 1.0 M
solution in THF) TBAF. The solution was degassed via a single
freeze-pump-thaw cycle. After 10 min., a color change from
maroon to amber was noticed. The reaction mixture was
quenched after 48 h. with 40 mL H2O. The layers were
separated and the aqueous phase was extracted with 3 × 40
mL Et2O. The combined organic layers were dried over Na2SO4
and concentrated in vacuo. Purification of the residue by flash
chromatography (20 mm, 20 cm, 5% EtOAc/ hexanes) gave 48
mg (51%) of arenes 22:23 as a white solid. The ratio of diesters
22 to 23 was 1:1. The pure regioisomers were separated using
preparative HPLC (3:1 hexanes:EtOAc). Arene 22: TLC Rf )
0.35 (3:1 hexanes:EtOAc); IR (CCl4) 3454 (w), 3026 (w), 2943
5.66 (d, 1H, 10.0), 5.87 (d, 1H, 10.0), 5.91 (s, 2H), 6.67-6.73
(m, 3H); 13C NMR (CDCl3) 14.5, 22.9, 25.5, 47.6, 52.7, 60.6,
100.9, 108.0, 108.7, 121.5, 127.9, 136.5, 146.2, 147.6, 156.0;
FAB mass spectrum m/z (relative intensity) 290 ((M + H), 21),
201 (100), 174 (31), 135 (81), 73 (90); HRMS (FAB) calcd for
C16H20O4N (M + H) 290.1392, found 290.1379.
Diester 28. Usin g P d (P P h 3)4. The preparation of diester
28 was performed in the same manner as for diesters 8 and 9
(benzoate 5). The Pd(0)-catalyzed addition of the diethyl
malonate anion to 30 mg (0.085 mmol) of allylic benzoate 26
gave 32 mg (97%) of diester 28 as a colorless oil. TLC Rf )
0.44 (10% EtOAc/hexane); IR (CCl4) 3069 (w), 2988 (s), 2963
(s), 2944 (m), 2906 (w), 2889 (w), 2875 (w), 1756 (s), 1731 (s),
1606 (w); 1H NMR (CDCl3) 1.23 (t, 6, J ) 7.1), 1.37 (dt, 1, J )
7.5, 13.5), 2.39 (dt, 1, J ) 8.3, 13.5), 3.14-3.17 (m, 1), 3.26 (d,
1, J ) 9.6), 3.37-3.42 (m, 1), 4.15-4.27 (m, 6), 5.77-5.80 (m,
2), 7.42 (t, 2, J ) 7.5), 7.53 (t, 1, J ) 7.5), 8.01 (d, 2, J ) 7.5);
13C NMR (CDCl3) 14.1, 31.7, 45.0, 45.1, 57.3, 61.3, 68.1, 128.3,
129.6, 130.2, 132.8, 132.9, 133.6, 166.5, 168.4; LRMS (EI) 361
((M + 1), 3), 238 (45), 164 (87), 105 (100), 77 (35); HRMS (EI)
calcd for C20H25O6 (M + 1) 361.1651, found 361.1641.
1
(w), 1728 (s), 1500 (s); H NMR (CDCl3) 1.25 (t, 3H, J ) 6.7),
1.47-1.52 (m, 1H), 1.58-1.63 (m, 1H), 2.07-2.09 (m, 2H),
3.36-3.38 (m, 1H), 4.12 (q, 2H, J ) 6.7), 4.32-4.34 (m, 1H),
4.63 (br s, 1H), 5.77 (d, 1H, J ) 9.9), 5.82 (d, 1H, 9.9), 7.16-
7.31 (m, 5H); 13C NMR (CDCl3) 14.6, 29.6, 31.0, 41.8, 47.0, 60.7,
126.3, 127.5, 128.4, 129.6, 133.1, 145.2, 156.0; FAB mass
spectrum m/z (relative intensity) 246 ((M + H), 17), 157 (66),
91 (100); HRMS (FAB) calcd for C15H20O2N (M + H) 246.1494,
found 246.1491. Arene 23: Rf ) 0.35 (3:1 hexanes:EtOAc); IR
(CCl4) 3447 (w), 3029 (w), 2933 (w), 1729 (s), 1552 (s); 1H NMR
(CDCl3) 1.16 (t, 3H, J ) 6.8), 1.58-1.62 (m, 1H), 1.87-1.92
(m, 1H), 2.14-2.24 (m, 2H), 3.29-3.33 (m, 1H), 3.78-3.83 (m,
1H), 4.01 (q, 2H, J ) 6.8), 4.80 (br s, 1H), 5.62 (ddd, 1H, J )
2.0, 3.6, 9.9), 5.89 (ddd, 1H, J ) 2.0, 4.0, 9.9), 7.21-7.29 (m,
5H); 13C NMR (CDCl3) 14.5, 23.0, 25.5, 48.0, 52.6, 60.6, 126.7,
127.9, 128.3, 128.4 (2C), 142.6, 156.0; FAB mass spectrum m/z
(relative intensity) 246 ((M + H), 17), 102 (100), 91 (25); HRMS
(FAB) calcd for C15H20O2N (M + H) 246.1494, found 246.1491.
Usin g TBAT. To a solution of 100 mg (0.389 mmol) of
carbonate 19 and 20 mg (0.020 mmol) of Pd2(dba)3‚CHCl3 in
10 mL of THF was added 420 mg (0.778 mmol) of TBAT. The
solution was degassed via a single freeze-pump-thaw cycle
and then heated to 55 °C. The reaction was quenched by the
addition of 20 mL of H2O and extracted with 3 × 20 mL of
Et2O. The combined organics were dried over MgSO4 and
concentrated in vacuo. Purification of the residue by column
chromatography (20 mm, 20 cm, 9:1 hexanes:EtOAc) gave 50
mg (53%) of arenes 22 and 23 as a 1:1 mixture. The spectral
data was identical to the preceding results.
Ar en es 29 a n d 30. Usin g P h Sn Me3. To a yellow mixture
of 30 mg (0.093 mmol) of allylic benzoate 26, 3 mg (5 µmol) of
Pd(dba)2 and 12 mg (0.28 mmol) of LiCl in 3 mL of DMF was
added 34 mg (0.14 mmol) of phenyltrimethyltin. The mixture
was degassed and stirred at room temperature for 72 h. The
reaction mixture was then partitioned between 15 mL of
distilled H2O and 15 mL of Et2O. The organic layer washed
with an additional 15 mL of distilled H2O, dried over MgSO4,
and concentrated in vacuo. Purification of the residue by flash
chromatography (10 mm, 16 cm, 0-10% CH2Cl2/hexane) gave
23 mg (88%) of a 1:1 ratio of arenes 29 and 30 as a yellow oil.
TLC Rf ) 0.33 (25% EtOAc/hexane); IR (CCl4) 3063 (w), 3038
1
(w), 2956 (s), 2899 (s), 2869 (s), 1725 (s), 1606 (w); H NMR
(CDCl3) for arene 29 1.97-2.05 (m, 1), 2.26-2.31 (m, 1), 3.33-
3.35 (m, 1) 4.01-4.05 (m, 1), 5.72-5.96 (m, 2), 7.18-7.58 (m,
6), 7.96-8.06 (m, 2); 1H NMR (CDCl3) for arene 30 2.32-2.35
(m, 1), 2.60-2.66 (m, 1), 2.70-2.75 (m, 1) 3.74-3.76 (m, 1),
5.72-5.96 (m, 2), 7.18-7.58 (m, 6), 7.96-8.06 (m, 2); GCMS
(arene) 130 (100), 129 (98), 128 (84), 115 (95), 91 (67), 77 (46),
51 (59); (arene) 160 (40), 115 (51), 91 (100), 77 (47), 51 (72).
Usin g TBAT. A solution of 3.0 mg (5.0 mmol) of Pd(dba)2,
1.0 mg (5.0 mmol) of PPh3, and 30 mg (0.093 mmol) of allylic
benzoate 26 in 5 mL of THF was degassed by three freeze-
pump-thaw cycles. The solution stirred at room temperature
for 30 min before 97 mg (0.18 mmol) of TBAT was added. The
reaction mixture was then heated at reflux for 12 h. The
resulting black mixture was filtered through a silica gel plug
before it was partitioned between 5 mL of water and 5 mL of
Et2O. The aqueous layer was then extracted with 3 × 5 mL of
Et2O. The organic layers were combined, dried over MgSO4,
and concentrated in vacuo. Purification of the residue by
column chromatography using pentane gave 8 mg (33%) of a
1:1 ratio of arenes 29 and 30 as a pale yellow oil.
Ar en es 24 a n d 25. To a solution of 220 mg (0.778 mmol) of
1,2-methylenedioxyphenyl siloxane in 10 mL of THF were
added 100 mg (0.389 mmol) of carbonate 19, 20 mg (0.019
mmol) of Pd2(dba)3‚CHCl3, and 780 µL (1.05 mmol, 1.0 M
solution in THF) of TBAF. The reaction mixture was heated
to 55 °C. After 10 min, a color change from maroon to amber
was noticed. The reaction mixture was quenched after 20 h
with 40 mL of H2O. The layers were separated, and the
aqueous phase was extracted with 3 × 40 mL of Et2O. The
combined organic layers were dried over Na2SO4 and concen-
trated in vacuo. Purification of the residue by flash chroma-
tography (20 mm, 20 cm, 9:1 hexanes:EtOAc) gave 75 mg (67%)
of a 1:1 mixture of arenes 24:25 as a white solid. The ratio of
diesters 24 to 25 was 1:1. The pure regioisomers were
separated using preparative HPLC (3:1 hexanes:EtOAc). Arene
24: recrystallized from CH2Cl2/ hexanes, mp ) 90-92 °C. TLC
Rf ) 0.22 (3:1 hexanes:EtOAc); IR (CCl4) 3441 (w), 3033 (w),
2926 (w), 1721 (s), 1542 (s); 1H NMR (CDCl3) 1.23 (t, 3H, J )
7.0), 1.41-1.54 (m, 2H), 2.01-2.07 (m, 2H), 3.28-3.29 (m, 1H),
4.10 (q, 2H, 7.0), 4.28-4.30 (m, 1H), 4.63 (br s, 1H), 5.75 (m,
Diester s 34 a n d 35. Usin g P d (P P h 3)4. The preparation
of diesters 34 and 35 was performed in the same manner as
for diesters 8 and 9 (benzoate 5). The Pd(0)-catalyzed addition
of the diethyl malonate anion to 30 mg (0.085 mmol) of allylic
benzoates 31 and 32 gave 32 mg (97%) diesters 34 and 35 as
a 1:1 ratio of a colorless oil. TLC Rf ) 0.44 (10% EtOAc/
hexane); IR (CCl4) 3069 (w), 2988 (s), 2963 (s), 2944 (m), 2906
1
(w), 2889 (w), 2875 (w), 1756 (s), 1731 (s), 1606 (w); H NMR
(CDCl3) 1.23 (t, 6, J ) 7.1), 1.37 (dt, 1, J ) 7.5, 13.5), 2.39 (dt,
1, J ) 8.3, 13.5), 3.14-3.17 (m, 1), 3.26 (d, 1, J ) 9.6), 3.37-
3.42 (m, 1), 4.15-4.27 (m, 6), 5.77-5.80 (m, 2), 7.42 (t, 2, J )
7.5), 7.53 (t, 1, J ) 7.5), 8.01 (d, 2, J ) 7.5); GCMS (diester)
238 (29), 165 (69), 164 (88), 118 (59), 105 (96), 91 (81), 79 (97),
78 (74), 77 (100), 51 (49); (diester) 238 (60), 165 (81), 164 (95),
118 (45), 105 (98), 91 (61), 79 (94), 78 (72), 77 (100), 51 (54).
Ar en es 36 a n d 37. Usin g P h Sn Me3. Using the previously
described procedure for the synthesis of arenes 29 and 30 via
PhSnMe3, but substituting a 2:1 mixture of benzoates 31 and
32 gave 20 mg (77%) of a 1:20:2:2 ratio of arenes 36, 37, 29,
and 30 as yellow oil.
2H), 5.90 (s, 2H), 6.60-6.64 (m, 2H), 6.72 (d, 1H, J ) 7.9); 13
C
NMR (CDCl3) 14.6, 29.5, 31.1, 41.5, 46.9, 60.7, 100.8, 108.0,
108.2, 120.4, 129.6, 133.2, 139.2, 146.0, 147.6, 156.1; FAB mass
spectrum m/z (relative intensity) 290 ((M + H), 46), 201 (100),
174 (24), 135 (60), 73 (48); HRMS (FAB) calcd for C16H20O4N
(M + H) 290.1392, found 290.1390. Arene 25: Rf ) 0.22 (3:1
hexanes:EtOAc); IR (CCl4) 3447 (w), 3029 (w), 2933 (w), 1729
(s), 1552 (s); 1H NMR (CDCl3) 1.18 (t, 3H, J ) 7.2), 1.53-1.62
(m, 1H), 1.86-1.90 (m, 1H), 2.04-2.20 (m, 2H), 3.21-3.23 (m,
1H), 3.71-3.74 (m, 1H), 4.03 (q, 2H, 7.2), 4.74-4.76 (m, 1H),

Addition of C-Nucleophiles to Carbocycles
J . Org. Chem., Vol. 67, No. 2, 2002 335
Usin g TBAT. Using the previously described procedure for
the synthesis of arenes 29 and 30 via TBAT, but substituting
a 2:1 mixture of benzoates 31 and 32 gave 11 mg (42%) of a
1:20 ratio of arenes 36 and 37 as a light yellow oil. TLC Rf )
0.47 (10% EtOAc/hexane); IR (CCl4) 3063 (w), 3031 (w), 2956
(s), 2938 (s), 2898 (s), 2868 (s), 1725 (s), 1600 (w); 1H NMR
(CDCl3) 1.52-1.61 (m, 2), 2.63-2.78 (m, 1), 3.21-3.36 (m, 1),
3.97-4.15 (m, 1), 4.32-4.44 (m, 2), 5.88-5.94 (m, 2), 7.18-
7.72 (m, 8), 7.95-8.11 (m, 2); 13C NMR (CDCl3) 37.6, 45.7, 51.2,
68.2, 126.2, 127.1, 127.3, 128.3, 128.4, 128.5, 129.6, 132.4,
132.9, 136.0, 166.6; LRMS (EI) 278 (M+, 3), 201 (17), 156 (100),
105 (53), 77 (39); HRMS calcd for C19H18O2 (M+) 278.1307,
found 278.1308. The experiment was repeated with dppe as
the ligand, and a 1:21 ratio of arenes 36 and 37 was obtained.
Diester s 42 a n d 43. Usin g P d (P P h 3)4. The preparation
of diesters 42 and 43 was performed in the same manner as
for diesters 8 and 9 (benzoate 5). The Pd(0)-catalyzed addition
of the diethyl malonate anion to 10 mg (31 mmol) of allylic
benzoates 39 and 40 gave an oil which was transferred to a
polyethylene test tube in 5 mL of CH3CN. To this solution were
added 10 µL (49% aq solution, 0.31 mmol) of HF and 15 µL
(25% aq solution, 0.03 mmol) of H2SiF6 via Eppendorf pipet.
Within 5 min, the solution became light yellow. The reaction
mixture was stirred for 20 min and quenched by the addition
of 1 mL of saturated of K2CO3 solution and 1 mL of saturated
NaCl solution solution. The aqueous phase was extracted with
3 × 2 mL EtOAc. The combined organic layers were dried over
Na2SO4 and concentrated in vacuo. Purification of the residue
by flash chromatography (5 mm, 3 cm, 40% EtOAc/hexane)
gave 6 mg (80%) of a 1:1 ratio of deprotected diesters as a
colorless oil. The deprotected diesters were then esterified with
BzCl and pyridine using conditions previously described
(benzoate 12). The esterification gave 4 mg (50%) of a 1:1 ratio
of diesters 42 and 43 as a colorless oil. The overall yield of
transformations was 40%. TLC Rf ) 0.23 (40% EtOAc/hexane);
IR (CCl4) 3038 (m), 3550 (m), 2988 (m), 2962 (m), 2938 (m),
1
2875 (m), 1756 (s), 1738 (s); H NMR (CDCl3) 1.25 (t, 6, J )
6.9), 1.79-1.83 (m, 1), 1.90-1.93 (m, 1), 2.85-3.02 (m, 1), 3.22
(d, 1, J ) 9.5), 3.42-3.44 (m, 1), 3.54 (d, 2, J ) 5.6), 4.17 (q, 4,
J ) 6.9), 5.76 (d, 1, J ) 5.8), 5.81 (d, 1, J ) 5.8); 13C NMR
(CDCl3) 14.1, 30.9, 44.9 47.7, 57.0, 61.3, 66.1, 133.4, 134.0,
168.6; LRMS (EI) 256 (M+, 2), 238 (12), 226(11), 166 (23), 164
(55), 151 (87), 114 (30), 79 (100); HRMS (EI) calcd for C13H20H5
(M+) 256.1311, found 256.1302.
Ack n ow led gm en t . We thank Dr. Yiu-Fai Lam,
Ms. Caroline Ladd, and Dr. J im Fettinger for their
assistance in obtaining NMR, mass spectral, and X-ray
data, respectively. We also thank Professor Michael
Crimmins (University of North Carolina at Chapel Hill)
for providing us with the diol precursor for dibenzoate
26. The generous financial support of the University of
Maryland and the National Institutes of Health is
acknowledged. M.H. thanks the Organic Division of the
American Chemical Society for a Predoctoral Fellowship
(sponsored by Pfizer, Inc.).
Su p p or tin g In for m a tion Ava ila ble: Synthetic proce-
dures and spectroscopic data for compounds 5, 6, 12, 13, 19,
26, 31, 32, 39, and 40 are contained in the Supporting
Information, along with 1H and 13C NMR spectra for com-
pounds 8, 21, 22, 23, 24, 25, 28, and 37. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O010672N