Novel use of ring strain to control regioselectivity: Alkene-directed, palladium-catalyzed allylation [2]

Full text of DOI:10.1021/ja016446c

9174
J. Am. Chem. Soc. 2001, 123, 9174-9175
Novel Use of Ring Strain to Control Regioselectivity:
Alkene-Directed, Palladium-Catalyzed Allylation
Marie E. Krafft,*,† Masaharu Sugiura,† and Khalil A. Abboud‡
Department of Chemistry, Florida State UniVersity
Tallahassee, Florida 32306-4390
opposite regioselectivity of allylation is obtained when a Pd/P
ratio of 1 is used (vide infra), suggesting an influential role of
the alkene. Coordination of a pendant alkene to the metal center
in the π-allyl intermediate resulting from inter- or intramolecular
diene dimerizations has been demonstrated.7-9 Interestingly, the
presence of a remote double bond has been proposed to influence
the direction of â-elimination of a palladacycle, whereas in the
presence of phosphines the opposite alkene isomer predomi-
nated.10
The results of the palladium-catalyzed reaction of malonate
anion with allylic acetates 2-4 show a surprising change in
regioselectivity as the tether length between the alkene and allylic
acetate increases from one to three atoms. In the reactions of 2
Department of Chemistry, UniVersity of Florida
GainesVille, Florida 32611-7200
ReceiVed June 19, 2001
ReVised Manuscript ReceiVed July 23, 2001
Palladium-catalyzed allylations of soft nucleophiles are a
powerful synthetic transformation.1,2 While the stereo- and
regiochemical consequences of the allylation have been well-
studied, questions still remain. The major factors influencing
regioselectivity are nucleophile type, steric bulk of the groups at
the allylic termini, electronic nature of allyl substituents, stability
of the η2-alkene palladium(0) complex resulting from the nu-
cleophilic addition to the π-allyl moiety, and steric and electronic
effects from the other ligands on the metal center. Much attention
has recently been given to external ligand control over the
regiochemical outcome, in particular, as it pertains to asymmetric
catalysis of the transformation.3,4 We have demonstrated that
allylic acetates possessing a thioether or dimethylamino substituent
in the homoallylic position directed the addition of malonate to
the allylic terminus proximal to the heteroatom.5 We proposed
that chelation of the heteroatom played an integral role in
determining the regiochemical outcome. This internal (intramo-
lecular) ligand control represents an additional factor in the
regioselectivity-defining process (Scheme 1). We anticipated that
other nonheteroatom containing functional groups capable of
coordination to a metal would direct the regiochemistry of
allylation. Herein we describe the noVel use of an alkene as a
directing group in Pd-catalyzed allylations, the stereochemistry
of the process, and mechanistic insight proVided by a crystal
structure of the putatiVe alkene-bound Pd(+2) intermediate.
and 4 the major products arise from substitution at the unsubsti-
tuted terminus of the allylic acetate. However, the only isomer
obtained from the reaction of allylic acetate 3 was diene 6 where
substitution had taken place on the more substituted allylic
position proximal to the tethered alkene. In light of our results
with heteroatom-directed allylation, it is evident that alkene
coordination to the metal center is responsible for the observed
high selectivity. Support for the influential capacity of the alkene
is derived from observation of the regioselectivity of reaction of
3 in the presence of DPPE in which almost complete disruption
of the remote alkene effect is observed, giving a 14:48:38 ratio
of 6a:6b:6c in 90% yield. The dichotomy of results when the
reaction is performed in the presence of 1 versus 2 or more equiv
of phosphine per palladium atom strongly implicates coordination
of the alkene as the regiochemical determining factor. Reactions
of allylic benzoates 8, 9, and 10 suggest the generality of the
directing effect. Substituents at the allylic terminus distal to the
Scheme 1
A number of examples of palladium-catalyzed reactions of
alkene bearing substrates exist. One example is very relevant to
the results described herein (eq 1). In the presence of a large
excess of phosphine, Pd(0) catalyzes allylic alkylation at the less
substituted terminus of the allyl moiety despite the presence of
an alkene in a position to bind to the metal center (eq 1).6 The
directing alkene or at the central allylic carbon have no effect on
the regiochemical outcome. Tertiary allylic acetate 10 undergoes
reaction to generate a quaternary stereocenter in high yield with
complete regiocontrol, although disubstitution at one allylic
terminus may impact the result.11,12 It is striking that, even when
the distal terminus is unsubstituted, reaction occurs at the more
substituted end proximal to the tether which is contrary to the
normal steric guidance over substitution. Alkyl substitution at the
internal position of the directing alkene has a minimal effect on
the regioselectivity; however, 1,2-disubstituted and 1,1,2-trisub-
† Florida State University.
‡ University of Florida.
(1) For leading references, see: Harrington, P. J. In ComprehensiVe
Organometallic Chemistry II; Abel, E. W., Stone, F. G. A., Wilkinson, G.,
Eds.; Pergamon: Oxford, 1995; Vol. 12, Chapter 8.2, p 798. Godleski, S. A.
In ComprehensiVe Organic Synthesis; Trost, B. M.; Fleming, I., Semmelhack,
M. F., Eds.; Pergamon: Oxford, 1991; Vol. 4, Chapter 3.3, pp 585-661.
(2) Blacker, A. J.; Clarke, M. L.; Loft, M. S.; Williams, J. M. J. Org. Lett.
1999, 1, 1969. For regioselectivity in Rh-catalyzed allylic alkylations, see:
Evans, P. A.; Nelson, J. D. J. Am. Chem. Soc. 1998, 120, 5581. Evans, P. A.;
Robinson, J. E.; Nelson, J. D. J. Am. Chem. Soc. 1999, 121, 6761.
(3) Pfaltz, A.; Lautens, M. In ComprehensiVe Asymmetric Catalysis II;
Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin, 1999;
Chapter 24, pp 833
(5) Krafft, M. E.; Fu, Z.; Procter, M. J.; Wilson, A. M.; Hirosawa, C. Pure
Appl. Chem. 1998, 70, 1083-1090. Krafft, M. E.; Wilson, A. M.; Fu, Z.;
Procter, M. J.; Dasse, O. A. J. Org. Chem. 1998, 63, 1748.
(6) Trost, B. M.; Tanoury, G. J.; Lautens, M.; Chan, C.; MacPherson, D.
T. J. Am. Chem. Soc. 1994, 116, 4255.
(7) Jolly, P. W.; Mynott, R.; Raspel, B.; Schick, K.-P. Organometallics
1986, 5, 473.
(4) For several examples, see: Trost, B. M.; Van Vranken, D. L. Chem.
ReV. 1996, 96, 395. Frost, C. G.; Howarth, J.; Williams, J. M. J. Tetrahe-
dron: Asymmetry 1992, 3, 1089. Moberg, C.; Bremberg, U.; Hallman, K.
Svenson, M.; Norrby, P.-O.; Hallberg, A.; Larhed, M.; Cso¨regh, I. Pure Appl.
Chem. 1999, 71, 1477. Tenaglia, A.; Heumann, A. Angew. Chem., Int. Ed.
1999, 38, 2180. Hayashi, T.; Kawatsura, M.; Uozumi, Y. J. Am. Chem. Soc.
1998, 120, 1681. Lloyd-Jones, G. C.; Stephen, S. C. Chem. Eur. J. 1998, 4,
2539.
(8) Takacs, J. M.; Chandramouli, S. V. J. Org. Chem. 1993, 58, 7315.
(9) Takacs J. M.; Zhu, J. J. Org. Chem. 1989, 54, 5193. See also, Takacs,
J. M.; Zhu, J. Tetrahedron Lett. 1990, 31, 1117.
(10) Trost, B. M.; Lautens, M. Tetrahedron Lett. 1985, 26, 4887.
10.1021/ja016446c CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/22/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 37, 2001 9175
Table 1
Scheme 2
PR3
time (h) yield 6a (%)
PR3
time (h) yield 6a (%)
P(C6F5)3
PCy3
1
16
79a
44b
P(OPh)3
PPh3
3
17
91
69a
a Small amounts of starting material remained. b Starting material
and other unidentified compounds were present.
hydrogens (δ 4.67 and 4.26 ppm), were evident. 1H NOE
enhancements between H5 and H3R and between H1â and H3R
suggest the major conformer resides in a chairlike conformation
in solution. Not surprisingly, reaction of one equiv of 18 with
one equiv of sodio dimethylmalonate at room temperature resulted
in a 65% yield of diene 11 along with 5% of the regioisomeric
addition product. Thus, the results are identical when the reaction
is performed under either catalytic or stoichiometric conditions.
X-ray diffraction analysis revealed disorder in the tethered alkene
part of the ligand. A thermal ellipsoid drawing of one of the two
allyl complexes is shown in Figure 1.
Figure 1. Thermal ellipsoids drawing of η3,η2-allyl ene complex 18; H
atoms of the phenyl rings were removed for clarity. Drawing with part 1
of the disorder (C1-C3). Selected bond lengths (Å): Pd-C5 2.189, Pd-
C6 2.148, Pd-C7 2.177, Pd-C2 2.274, Pd-C1 2.316.
A number of factors are known to contribute to regioselectivity
in Pd-catalyzed allylations.1,3 However, in these cases an influ-
ential role of the alkene, presumably ligated, must be an integral
part of the explanation. With due consideration given to all aspects
of the mechanistic manifold, ring strain-induced selective activa-
tion of one of the allylic termini with concomitant influence in
the direction of rotation of the allyl ligand is the factor most likely
required in a rationalization of the regiochemical outcome.3
Nucleophilic addition at C5 (18, eq 3) associated with a
counterclockwise rotation of the allyl moiety out of the coordina-
tion plane during the change in hapticity, with simultaneous
shifting of the π-system to generate the elusive alkene complex,17
relieves ring strain as the chelate ring is effectively enlarged
(Scheme 2, A). Reaction at C7 would require a clockwise rotation
of the allyl moiety during hapticity change with simultaneous shift
of the π-system to the right, thus causing an increase in ring strain
as the effective ring size is decreased (Scheme 2, B).
In summary, a novel use of internal ring strain to drive reaction
selectivity has been described. Using a tethered alkene as a
nontraditional control element, regioselective addition to π-allyl
Pd complexes is possible. The directing effect overcomes the
normal steric bias, and reaction can occur at the more substituted
terminus of a monofunctionalized π-allyl moiety. X-ray diffraction
analysis of the alkene-bound intermediate sheds light on the
mechanism by providing structural evidence for alkene binding.
Further work in this area is currently underway.
stituted alkenes are detrimental to the outcome of the highly
selective allylation. Reaction of the sodium salt of diethyl
methylmalonate with 8 gave rise to a 93:7 mixture of regioisomers
where substitution at the allylic terminus proximal to the directing
alkene predominated.
In an attempt to ascertain the role of the external ligand in
influencing the regiochemical outcome, phosphorus ligands of
different electronic nature were employed (Table 1, allylic acetate
3, [(η3-allyl)PdCl]2 (2.5 mol %), PR3 (5 mol %), LiCH(CO2Me)2,
THF, 75 °C). Evidently, a trans influence from the phosphine
ligand12 is not responsible for the observed regioselectivity as
only one isomer was obtained regardless of the electronic nature
of the ligand.
The stereoselectivity of the alkene-directed substitution is high
and has been demonstrated to be retention of configuration,
suggesting a double inversion process as is normally observed in
Pd-catalyzed allylations (eq 2). Allylic acetate (R,E)-14 underwent
allylation to give (S,E)-15 which, after ring-closing metathesis
using Grubbs’ catalyst,13 provided the known cyclopentene (S)-
16.14
Acknowledgment. K.A.A. acknowledges the National Science Foun-
dation and the University of Florida for funding of the purchase of the
X-ray equipment. M.E.K. acknowledges the National Science Foundation
and donors to the Krafft Research Fund for support of this work. We
thank Professor Takacs (Nebraska) for helpful comments.
In an attempt to probe the mechanism and provide support for
alkene coordination, cationic complex 18 was synthesized and
1
characterized by H NOE spectroscopy and X-ray diffraction
Supporting Information Available: Experimental information for
all compounds; X-ray crystallographic information for 18 (PDF). This
material is available free of charge via the Internet at http://pubs.acs.org.
analysis (eq 3).15,16 The 1H NMR spectrum of allylic complex 17
showed broad peaks suggesting an equilibrium between mono-
meric and oligomeric complexes. However, the 1H NMR spectrum
of 18 revealed a distinct 4:1 mixture of two compounds where
conformational rigidity and binding of the alkene to the metal
center, as determined by an upfield shift of the terminal alkene
JA016446C
(15) NMR spectra of η5-alkenyl, allylic palladium complexes: Hughes,
R. P.; Jack, T.; Powell, J. J. Organomet. Chem. 1973, 63, 451. Jolly, P. W.;
Mynott, R.; Raspel, B.; Schick, K.-P. Organometallics 1986, 5, 473. Benn,
R.; Jolly, P. W.; Mynott, R.; Raspel, B.; Schenker, G.; Schick, K.-P.; Schroth,
G. Organometallics 1985, 4, 1945. GomezBenogoa, E.; Cuerva, J. M.;
Echavarren, A. S. M. Angew. Chem., Int. Ed. Engl. 1997, 36, 767.
(16) Crystal structures of η5-alkenyl, allylic palladium complexes: Ciajolo,
R.; Jama, M. A.; Tuzi, A.; Vitagliano, A. J. Organomet. Chem. 1985, 295,
233. Pregosin, P. S.; Salzmann, R. Coord. Chem. ReV. 1996, 155, 35.
(17) Steinhagen, H.; Reggelin, M.; Helmchen, G. Angew. Chem., Int. Ed.
1997, 36, 2108.
(11) Blacker, A. J.; Clarke, M. L.; Loft, M. S.; Williams, J. M. J. Org.
Lett. 1999, 1, 1969. Cuvigny, T.; Julia, M.; Rolando, C. J. Organomet. Chem.
1985, 285, 395.
(12) A° kermark, B.; Vitagliano, A. Organometallics 1985, 4, 1275.
(13) For recent reviews, see: Grubbs, R. H.; Chang, S. Tetrahedron 1998,
54, 4413. Fu¨rstner, A. Angew. Chem., Int. Ed. 2000, 39, 3013. Yet, L. Chem.
ReV. 2000, 100, 2963.
(14) Trost, B. M.; Bunt, R. C. J. Am. Chem. Soc. 1994, 116, 4089.