Effects of bidentate phosphine ligands on syn - Anti isomerization in π-allylpalladium complexes

Article abstract of DOI:10.1021/om020680+

The rates of syn-anti isoinerization in a series of cationic π-allylpalladium complexes [Pd(η³-Me₂CCHCHD)(bisphosphine)]BAr^F₄ [Ar^F = 3,5-(CF₃)₂C₆H₃] were measured by ¹H NMR spin-saturation transfer techniques. For the complexes with α,ω-bis(diphenylphosphino)-alkanes (Ph₂P(CH₂)_nPPh₂, n = 2-4), a correlation between the bisphosphine's bite angles and the syn-anti isomerization rates was observed, the bisphosphine with smaller bite angle accelerating the syn-anti isomerization. The isomerization rate was also dependent on the electronic characteristics of the bisphosphine ligands. Of the substituted dppf ligands (Fe(η⁵-C₅H₄PAr₂)₂, Ar = 4-CF₃C₆H₄, Ph, 4-MeOC₆H₄), it was fastest with the electron-withdrawing 4-CF₃C₆H₄ group and slowest with the electron-releasing 4-MeOC₆H₄ group on the phosphorus atoms. These bisphosphines were used as supporting ligands for the palladium-catalyzed allylic alkylation of (Z)-cinnamyl acetate and (Z)-2-hexenyl acetate with the sodium salt of dimethyl methylmalonate. The alkylation products with E-geometry, which result from the isomerization of π-allylpalladium intermediates from anti to syn, were formed more with the bisphosphines of faster syn-anti isomerization.

Full text of DOI:10.1021/om020680+

Organometallics 2002, 21, 4853-4861
4853
Effects of Bid en ta te P h osp h in e Liga n d s on syn -a n ti
Isom er iza tion in π-Allylp a lla d iu m Com p lexes
Masamichi Ogasawara, Ken-ichi Takizawa, and Tamio Hayashi*
Department of Chemistry, Faculty of Science, Kyoto University, Sakyo, Kyoto 606-8502, J apan
Received August 19, 2002
The rates of syn-anti isomerization in a series of cationic π-allylpalladium complexes
[Pd(η3-Me2CCHCHD)(bisphosphine)]BArF4 [ArF ) 3,5-(CF3)2C6H3] were measured by 1H NMR
spin-saturation transfer techniques. For the complexes with R,ω-bis(diphenylphosphino)-
alkanes (Ph2P(CH2)nPPh2, n ) 2-4), a correlation between the bisphosphine’s bite angles
and the syn-anti isomerization rates was observed, the bisphosphine with smaller bite angle
accelerating the syn-anti isomerization. The isomerization rate was also dependent on the
electronic characteristics of the bisphosphine ligands. Of the substituted dppf ligands (Fe(η5-
C5H4PAr2)2, Ar ) 4-CF3C6H4, Ph, 4-MeOC6H4), it was fastest with the electron-withdrawing
4-CF3C6H4 group and slowest with the electron-releasing 4-MeOC6H4 group on the
phosphorus atoms. These bisphosphines were used as supporting ligands for the palladium-
catalyzed allylic alkylation of (Z)-cinnamyl acetate and (Z)-2-hexenyl acetate with the sodium
salt of dimethyl methylmalonate. The alkylation products with E-geometry, which result
from the isomerization of π-allylpalladium intermediates from anti to syn, were formed more
with the bisphosphines of faster syn-anti isomerization.
In tr od u ction
isomerization of the olefin geometry from Z to E, which
is caused by anti-syn isomerization in the π-allylpal-
ladium intermediates5,6 (Scheme 1). In the asymmetric
substitution reactions of racemic allylic esters using
chiral ligands, epimerization is required in the π-al-
lylpalladium intermediates, for which the π-σ-π pro-
cess is responsible7 (Scheme 2). In these reactions, the
relative rate between the isomerization of the π-al-
lylpalladium intermediates and the nucleophilic attack
giving the final products is a key factor determining the
stereochemical outcome.
When a palladium complex coordinated with a phos-
phine ligand is used as a catalyst, the influence of the
phosphine ligand on stereoselectivity in the catalytic
allylic substitution reactions is very large,8 although the
effect is not yet fully established. Thus, it is expected
Palladium-catalyzed allylic alkylation is one of the
most widely and the most frequently used carbon-
carbon bond forming reactions catalyzed by transition
metal complexes because of its wide range of reactivity,
high catalytic activity, and easy manipulation,1 and the
allylic alkylation reaction has been successfully ex-
tended to catalytic asymmetric synthesis by use of chiral
ligands.2 The reaction proceeds via a π-allylpalladium-
(II) intermediate, and the stereochemical outcome of the
catalytic reaction is highly dependent on the stereo-
chemical structure of the π-allylpalladium(II) interme-
diate.3 The π-allylpalladium complex is well known to
undergo isomerization via a so-called π-σ-π process,
and the isomerization plays a decisive role in the
reactions that involve the π-allylpalladium intermediate
in the catalytic cycle.1-4 Thus, for example, the allylic
alkylation of (Z)-allyl esters is usually accompanied by
(4) (a) Cuvigny, T.; J ulia, M. Rolande, C. J . Organomet. Chem. 1985,
285, 395. (b) Åkermark, B.; Krakenberger, B.; Hansson, S.; Vitagliano,
A. Organometallics 1987, 6, 620. (c) Åkermark, B.; Zetterberg, K.;
Hansson, S.; Krakenberger, B.; Vitagliano, A. J . Organomet. Chem.
1987, 335, 133. (d) Åkermark, B.; Hansson, S.; Vigagliano, A. J . Am.
Chem. Soc. 1990, 112, 4587. (e) Sjo¨gren, M.; Hansson, S.; Norrby,
P.-O.; Åkermark, B.; Cucciolito, M. E.; Vitagliano, A. Organometallics
1992, 11, 3954. (f) Sjo¨gren, M. P. T.; Hansson, S.; Åkermark, B.;
Vitagliano, A. Organometallics 1994, 13, 1963.
(5) As an example of the palladium-catalyzed allylic alkylation of
(Z)-allyl acetates giving E-products: Hayashi, T,; Yamamoto, A.;
Hagihara, T. J . Org. Chem. 1986, 51, 723.
(6) As an example of the nucleophilic attack before the isomeriza-
tion: Kazmaier, U.; Zumpe, F. L. Angew. Chem., Int. Ed. 2000, 39,
802.
(1) For reviews: (a) Tsuji, J . Acc. Chem. Res. 1969, 2, 144. (b) Tsuji,
J . Pure Appl. Chem. 1982, 54, 197. (c) Trost, B. M.; Verhoeven, T. R.
In Comprehensive Organometallic Chemistry, Vol. 8; Wilkinson, G.,
Stone, F. G. A., Abel, E. W., Eds.; Pergamon: Oxford, 1982; pp 799-
938. (d) Heck, R. F. Palladium Reagents in Organic Synthesis;
Academic Press: New York, 1985. (e) Tsuji, J . J . Organomet. Chem.
1986, 300, 281. (f) Tsuji, J .; Minami, I. Acc. Chem. Res. 1987, 20, 140.
(g) Tsuji, J . Organic Synthesis with Palladium Compounds; Springer-
Verlag: New York, 1990. (h) Godleski, S. In Comprehensive Organic
Synthesis, Vol. 4; Trost, B. M., Fleming, I., Semmelhack, M. F., Eds.;
Pergamon: Oxford, 1991; pp 585-659. (i) Tsuji, J . Palladium Reagents
and Catalysts; J ohn Wiley and Sons: Chichester, 1995.
(2) For reviews: (a) Trost, B. M.; Chulbom, L. In Catalytic Asym-
metric Synthesis, 2nd ed.; Ojima, I., Ed.; VCH: New York, 2000; p 593.
(b) Pfaltz, A.; Lautens, M. In Comprehensive Asymmetric Catalysis;
J acobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin, 1999;
Vol. 2, Chapter 24. (c) Trost, B. M.; Van Vranken, D. L. Chem. Rev.
1996, 96, 395. (d) Hayashi, T. In Catalytic Asymmetric Synthesis;
Ojima, I., Ed.; VCH: New York, 1993; p 325. (e) Frost, C. G.; Howarth,
J .; Williams, J . M. J . Tetrahedron Asymmetry 1992, 3, 1089.
(3) For reviews: (a) Consiglio, G.; Waymouth, M. Chem. Rev. 1989,
89, 257. (b) Pregosin, P. S.; Salzmann, R. Coord. Chem. Rev. 1996,
155, 35.
(7) ) (a) Auburn, P. R.; Mackenzie, P. B.; Bosnich, B. J . Am. Chem.
Soc. 1985, 107, 2033. (b) MacKenzie, P. B.; Whelan, J .; Bosnich, B. J .
Am. Chem. Soc. 1985, 107, 2046. (c) Hayashi, T.; Kishi, K.; Yamamoto,
A.; Ito, Y. Tetrahedron Lett. 1990, 31, 1743. (d) Dawson, G. J .; Williams,
J . M. J .; Coote, S. J . Tetrahedron Asymmetry 1995, 6, 2535. (e) Hayashi,
T.; Kawatsura, M.; Uozumi, Y. Chem. Commun. 1997, 561. (f) Pre´tot,
R.; Pfaltz, A. Angew. Chem., Int. Ed. 1998, 37, 323.
(8) For recent examples: (a) Kranenburg, M.; Kamer, P. C. J .; van
Leeuwen, P. W. N. M. Eur. J . Inorg. Chem. 1998, 25. (b) van Haaren,
R. J .; van Strijdonck, G. P. F.; Oevering, H.; Reek, J . N. H.; Kramer,
P. C. J .; van Leeuwen, P. W. N. M. Eur. J . Inorg. Chem. 2001, 837.
10.1021/om020680+ CCC: $22.00 © 2002 American Chemical Society
Publication on Web 09/24/2002

4854 Organometallics, Vol. 21, No. 22, 2002
Ogasawara et al.
Sch em e 1 a n ti-syn Isom er iza tion in th e Allylic
Alk yla tion of (Z)-Allyl Ester s
Sch em e 2. Ep im er iza tion via th e π-σ-π P r ocess
in Asym m etr ic Allylic Alk yla tion of Ra cem ic Allyl
Ester s
Sch em e 3
that the rate of the isomerization via the π-σ-π process
in π-allylpalladium complexes can be controlled by
proper choice of the phosphine ligand. To the best of
our knowledge, however, there have been only a few
reports on the measurements of the absolute values of
the syn-anti isomerization rates, which is in particular
palladium complexes coordinated with chiral phosphine
ligands.9-11 We have measured absolute values of the
isomerization rates in a series of cationic π-allylpalla-
dium-bisphosphine complexes [Pd(η3-Me2CCHCHD)-
phine ligands were used in the present studies. The first
series is R,ω-bis(diphenylphosphino)alkanes, where the
number of methylene units between the two phosphorus
atoms is two (dppe), three (dppp), or four (dppb). These
three bisphosphines have different bite angles, while
their electronic characteristics can be assumed to be
almost the same. The second series is 1,1′-bis(di-
arylphosphino)ferrocenes, where the aryl substituents
are phenyl (dppf), 4-MeOC6H4 (MeO-dppf), or 4-CF3C6H4
(CF3-dppf). The electron-releasing MeO groups or the
electron-withdrawing CF3 groups are introduced at the
p-position of the phenyl groups. They can alter only the
electronic characteristics of the phosphines without
changing the steric environment around the phosphorus
atoms. All of the bisphosphine ligands chosen above are
highly symmetric. They are at least C2 symmetric, the
two phosphino groups being equivalent, which will not
increase the number of the isomers of π-allylpalladium
complexes.
Design of Com p lexes. For the accurate measure-
ment of the rates of the syn-anti isomerization, we
designed a π-allylpalladium system shown in Scheme
3, which should show simple fluxional behavior as
described below.
First, two methyl groups were introduced at one end
of the π-allyl moiety, which is expected to make the
isomerization through a σ-allyl intermediate containing
a tertiary alkyl-palladium bond negligibly slow.3,14,15
Actually, the 1,1-dimethyl-π-allyl ligand showed the
(bisphosphine)]BArF using NMR techniques, where
4
bisphosphine is dppe,12 dppp,12 dppb,12 dppf,12 MeO-
dppf,12 or CF3-dppf.12 This study has clarified the effects
of the bidentate phosphines on the isomerization that
proceeds through the π-σ-π process in the π-allylpal-
ladium complexes. These π-allylpalladium complexes
have been used for the catalytic allylic alkylation of (Z)-
allyl acetates5 and showed a good correlation between
the E/Z-selectivity in the catalytic reactions and the
rates of the syn-anti isomerization in the complexes.
Here we report our observations.
Resu lts a n d Discu ssion
P h osp h in e Liga n d s. Two main factors character-
izing the nature of bidentate phosphine ligands are their
bite angles and electronic effects based on the substit-
uents on the phosphorus atoms.13 These two factors are
fundamentally different; thus, two series of bisphos-
(9) Breutel, C.; Pregosin, P.; Salzmann, R.; Togni, A. J . Am. Chem.
Soc. 1994, 116, 4067.
(10) Kawatsura, M.; Uozumi, Y.; Ogasawara, M.; Hayashi, T.
Tetrahedron 2000, 56, 2247.
(11) For the isomerization related to a memory effect in a π-allylpal-
ladium intermediate: Hayashi, T.; Kawatsura, M.; Uozumi, Y. J . Am.
Chem. Soc. 1998, 120, 1681.
(12) Abbreviations of bisphosphines: dppe ) 1,2-bis(diphenylphos-
phino)ethane, dppp ) 1,3-bis(diphenylphosphino)propane, dppb ) 1,4-
bis(diphenylphosphino)butane, dppf ) 1,1′-bis(diphenylphosphino)-
ferrocene, MeO-dppf ) 1,1′-bis[di(4-methoxyphenyl)phosphino]ferrocene,
CF3-dppf ) 1,1′-bis[bis(4-trifluoromethylphenyl)phosphino]ferrocene.
(13) Tolman, C. A. Chem. Rev. 1977, 77, 313.
(14) During the studies on palladium-catalyzed asymmetric reduc-
tion of allylic esters with formic acid, it was observed that this type of
π-σ-π isomerization is very slow: (a) Hayashi, T.; Iwamura, H.; Naito,
M.; Matsumoto, Y.; Uozumi, Y.; Miki, M.; Yanagi, K. J . Am. Chem.
Soc. 1994, 116, 775. (b) Hayashi, T.; Iwamura, H.; Uozumi, Y.;
Matsumoto, Y.; Ozawa, F. Synthesis 1994, 526. (c) Hayashi, T.;
Kawatsura, M.; Iwamura, H.; Yamaura, Y.; Uozumi, Y. Chem. Com-
mun. 1996, 1767.
(15) The alkyl substituent effects on the π-σ-π isomerization have
been studied by applying a density functional theory: Solin, N.; Szabo´.
Organometallics 2001, 20, 5464.

Effects of Bidentate Phosphine Ligands
Organometallics, Vol. 21, No. 22, 2002 4855
Sch em e 4
cationic complexes are air stable and can be purified
by recrystallization or column chromatography over
silica gel (with chloroform). All the complexes 1-6 exist
as a mixture of two isotopomers in solution. In any case,
the two isotopomers, the Hsyn-Danti isomer and the Hanti
-
Dsyn isomer, exist with equal abundance in solution and
show no thermodynamic preference for one of the two
isomers.
16
Recrystallization of the dppb-d8 complex 3 from
dichloromethane/pentane gave prismatic crystals, and
the solid state structure was determined by single-
crystal X-ray crystallography (Figure 1). The overall
molecular structure of the complex can be seen as
distorted four-coordinate square planar. Pd(1), P(1),
P(2), C(1), and C(3) atoms are located on the same
plane: the sum of the four angles at the Pd center
involving the other four atoms is 359.9°. The Pd(1)-
C(3) bond is longer (i.e., weaker) than the Pd(1)-C(1)
bond, which is consistent with formation of the σ-allyl
intermediate at the C(1) center (vide supra). An ex-
pected outcome from these is a stronger trans influence
from C(1) than that from C(3). Indeed the Pd(1)-P(1)
bond is longer than the Pd(1)-P(2) bond. No bonding
interactions were observed between the cationic pal-
syn-anti isomerization only at the nonsubstituted CH2
terminus. The selective isomerization is demonstrated
by the NOESY spectrum of the complex where the
exchange is observed between the two terminal hydro-
gens, but not between the two methyl groups. Second,
the monodeuterated π-allyl ligand was used for the
following reasons. The exchange rates between the two
hydrogen nuclei were calculated from the T1 values of
the hydrogen resonances and the magnitude of the spin-
saturation transfer between the two resonances (see
Experimental Section). With a standard pulse sequence
for the spin-saturation transfer measurement, the nuclear
Overhauser effect is also induced in addition to the spin-
saturation transfer effect. Thus, the intensity change
is always observed as an aggregate of the two funda-
mentally different phenomena. In our system, the two
hydrogens investigated are located on the same carbon;
therefore, the NOE might not be negligible. Our strategy
to solve this problem is deuterium labeling of one of the
terminal hydrogens, as shown in Scheme 3. With the
monodeuterated π-allyl complex, the syn-anti isomer-
ization is observed as an intermolecular process by
NMR, and thus NOE should not be induced. The third
device in the design of the palladium complex is partial
deuterium labeling of the bisphosphine ligands. For the
ladium moiety and the BArF anion.
4
Mea su r em en t of syn -a n ti Isom er iza tion Ra te.
The syn-anti isomerization shown in Scheme 3 was so
slow that no line broadening of the Hsyn and Hanti signals
1
was detected in the H NMR spectra in CDCl3 in the
temperature up to 55 °C for all the complexes. However,
1
in the H-1H NOESY spectra of the deuterium-labeled
complexes, cross-peaks were observed between the Hsyn
and Hanti resonances as expected, which is direct evi-
dence for the relatively slow exchange between the two
nuclei; thus, the measurement of the exchange rates
was carried out by a spin-saturation transfer technique
using the Forse´n-Hoffman method.18 We have exam-
ined several solvents and observed that the exchange
rates in less polar solvents such as CDCl3 and CD2Cl2
are highly dependent on a small amount of impurities
such as acetone or ether, which was brought to the
samples during the purification processes. We found
that acetone-d6 is the solvent of choice for the rate
measurement because the results are most reproducible
and reliable. Thus, the spin-saturation transfer experi-
ments were performed in acetone-d6 by irradiating
either of the Hsyn and Hanti signals of the π-allyl
1
dppp and dppb complexes, H NMR signals from the
alkylene backbone of the phosphines are overlapped
with the resonances from the π-allyl ligand. Therefore,
the partially deuterated phosphines Ph2P(CD2)nPPh2 (n
) 3 (dppp-d6) and 4 (dppb-d8)) were synthesized from
the corresponding deuterated R,ω-diols.16
16
complexes. For example, in the dppb-d8 complex 3,
saturation of the Hsyn resonance at δ 3.75 led to a
considerable decrease in the intensity of the Hanti
resonance at δ 2.83 (see Figure 2). The results obtained
for the π-allylpalladium complexes 1-6 are summarized
in Table 1. The exchange rates from Hsyn to Hanti and
those of reverse directions were determined for all the
complexes; that is, two exchange rates were measured
for each complex. As described above, all the complexes
exist as a 1:1 mixture of the two isotopomers (Hsyn-Danti
isomer and Hanti-Dsyn isomer) in solution; thus the two
exchange rates obtained for each complex should be
identical. The inequality of the two values in Table 1
P r ep a r a tion of Com p lexes. Preparation of the
deuterium-labeled π-allylpalladium complexes are shown
in Scheme 4. Reduction of 3-methyl-2-butenal with
LiAlD4 gave a quantitative yield of 1-deuterio-3-methyl-
2-butenol, which was treated with Li2PdCl4 and carbon
monoxide in hydrochloric acid according to the standard
procedure7a to give a dichloro-bridged dimeric π-al-
lylpalladium complex. Treatment of the dimer with 2
equiv of the bidentate phosphines (1 equiv to Pd)
17
followed by anion exchange with NaB[3,5-(CF3)2C6H3]4
gave a series of cationic palladium complexes 1-6. The
(18) (a) Forse´n, S.; Hoffman, R. A. J . Chem. Phys. 1963, 39, 2892.
(b) Forse´n, S.; Hoffman, R. A. J . Chem. Phys. 1964, 40, 1189. (c)
Sandstro¨m, J . Dynamic NMR Spectroscopy; Academic: New York,
1982.
(16) Ogasawara, M.; Saburi, M. Organometallics 1994, 13, 1911.
(17) Brookhart, M.; Grant, B.; Volpe, A. F., J r. Organometallics 1992,
11, 3920.

4856 Organometallics, Vol. 21, No. 22, 2002
Ogasawara et al.
F igu r e 1. Molecular structure of [Pd(η3-CMe2CHCHD)(dppb-d8)][B(C6H3-3,5-(CF3)2)4] (3). Displacement ellipsoids are shown
at the 30% probability level. Selected bond distances (Å) and angles (deg): Pd(1)-P(1) ) 2.331(3), Pd(1)-P(2) ) 2.287(3),
Pd(1)-C(1) ) 2.12(1), Pd(1)-C(2) ) 2.15(1), Pd(1)-C(3) ) 2.36(1), P(1)-Pd(1)-P(2) ) 103.5(1), P(1)-Pd(1)-C(3) ) 101.0-
(3), P(2)-Pd(1)-C(1) ) 89.5(4), C(1)-Pd(1)-C(3) ) 65.9(5).
between the bisphosphine’s bite angles and the syn-
anti isomerization rates obtained here in the π-allylpal-
ladium complexes. Thus, the exchange rates from Hanti
complex to Hsyn complex in complexes 1, 2, and 3 were
determined to be 3.8, 0.42, and 0.38 s-1, respectively
(entries 1-3). The order of the exchange rates in the
complexes 1-3 is as follows.
1 (dppe) > 2 (dppp) g 3 (dppb)
Although the difference of the exchange rates between
those in 2 and 3 is not large, the syn-anti isomerization
in 1 is much faster than in 2 or 3, 1 having the
bisphosphine with the smallest bite angle. As a general
trend, it can be seen that bisphosphine with a smaller
bite angle accelerates the syn-anti isomerization in the
π-allylpalladium complex.
The electronic effect of the phosphine ligands on the
syn-anti isomerization is distinct. Comparison of the
complexes 4-6, which are the series of dppf derivatives,
clarified that the syn-anti isomerization in the π-al-
lylpalladium species is faster with the electron-with-
drawing phosphine ligands. Introduction of the electron-
withdrawing CF3 group on the diphenylphosphino group
of dppf accelerates the isomerization 3-4 times faster,
while the electron-releasing OMe group decelerates the
isomerization 4-5 times slower.
F igu r e 2. 1H NMR (500 MHz) spectra of the complex 3 in
the π-allyl region at 50 °C in acetone-d6: (a) normal
spectrum; (b) spin-saturation transfer experiment with
saturation of the syn-H resonance at δ 3.75; (c) difference
spectrum between (a) and (b).
Ta ble 1. T1 Va lu es a n d Ra te Con sta n ts, k, for
syn -a n ti Isom er iza tion in
[P d (η3-Me2CCHCHD)(d ip h osp h in e)]BAr F in
4
Aceton e-d 6 a t 50 °C
entry complexa
T1(syn)/sb T1(anti)/sb
k
sfa/s-1
k
afs/s-1
1
2
3
4
5
6
1 (85.8°)c
2 (90.6°)c
3 (94.5°)d
4 (99.1°)e
5
3.02
2.93
2.58
2.34
1.84
1.73
3.04
2.67
2.56
2.31
1.77
1.76
3.8
3.8
0.39
0.37
1.2
0.26
4.3
0.42
0.38
1.2
0.27
4.6
6 (CF3-dppf) > 4 (dppf) > 5 (MeO-dppf)
6
a
Bite angles of the diphosphine ligands in PdCl2(diphosphine)
Effects of syn -a n ti Isom er iza tion Ra te on P a l-
la d iu m -Ca ta lyzed Allylic Alk yla tion . The results
described in the previous section suggest that the choice
of a phosphine ligand should have a great influence on
catalyst behavior in the palladium-catalyzed allylic
substitution reactions where the syn-anti isomerization
of the π-allylpalladium intermediates is a crucial factor
determining the selectivity. As a model reaction dem-
onstrating the influence of the phosphine ligands, we
chose a palladium-catalyzed allylic alkylation of (Z)-allyl
acetates. The outline of the system is illustrated in
Scheme 5.
b
in parentheses. Measured by an inversion-recovery method.
c Taken from ref 19a. Taken from ref 19b. e Taken from ref 19c.
d
(up to 8% difference in entry 2) can be attributed to
technical errors during the experiments.
As the representative bite angles of the bisphos-
phines, the P-Pd-P values in palladium complexes
PdCl2(bisphosphine),19 which have been reported for
dppe,19a dppp,19a dppb,19b and dppf,19c are also listed in
Table 1. It was found that there is a good correlation
(19) (a) Steffen, W. L.; Palenik, G. J . Inorg. Chem. 1976, 15, 2432.
(b) Makhaev, V. D.; Dzhabieva, Z. M.; Konovalikhin, S. V.; D’yachenko,
O. A.; Belov, G. P. Russ. J . Coord. Chem. 1996, 22, 563. (c) Hayashi,
T.; Konishi, M.; Kobori, Y.; Kumada, M.; Higuchi, T.; Hirotsu, K. J .
Am. Chem. Soc. 1984, 106, 158.
The oxidative addition of (Z)-allyl acetate ((Z)-RCHd
CHCH2OAc) to a palladium(0) species forms a π-allyl-
palladium intermediate where the R substituent is

Effects of Bidentate Phosphine Ligands
Organometallics, Vol. 21, No. 22, 2002 4857
Sch em e 5
Sch em e 6
located anti with respect to the central hydrogen. The
anti-π-allylpalladium possibly undergoes isomerization
into a syn-π-allylpalladium by the π-σ-π process,
because this type of syn-isomer which does not have a
sterically bulky substituent at the central position is
usually much more thermodynamically stable than the
anti-isomer.20 The nucleophilic attack on the nonsub-
stituted terminus of the anti-π-allylpalladium and syn-
π-allylpalladium gives the corresponding Z-product and
E-product, respectively. It follows that the E/Z ratio of
the final product in the palladium-catalyzed allylic
alkylation of (Z)-allyl acetate is dependent on the rate
of syn-anti isomerization (anti to syn isomerization).
Thus, the reaction of (Z)-allyl acetate gives the Z-
product if the isomerization is slow compared with the
nucleophilic attack, while it gives the E-product if the
isomerization is fast. Another important factor deter-
mining the E/Z ratio of the product is lifetime of the
π-allylpalladium intermediates, which is related to the
relative rate of the oxidative addition forming the
π-allylpalladium intermediate and the nucleophilic at-
tack,21 the longer lifetime offering the opportunity of
more isomerization. Although we do not have a quan-
titative measurement of the lifetime, we do have the
absolute rate of the syn-anti isomerization of a series
of π-allylpalladium complexes [(π-allyl)Pd(P-P)]+. We
found that the isomerization rate is a decisive factor for
the E/Z ratio of the products in the palladium-catalyzed
allylic alkylation of (Z)-allyl acetates (Scheme 6).
The allylic alkylation of (Z)-cinnamyl acetate and (Z)-
2-hexenyl acetate was carried out with a large excess
(5 equiv to the allyl esters) of the sodium salt of dimethyl
methylmalonate in the presence of 2 mol % (Pd) of a
catalyst generated in situ from [PdCl(π-C3H5)]2 and one
of the bisphosphines in acetone, which is the solvent
used for the measurement of the syn-anti isomeriza-
tion. The catalytic reactions at 20 °C for 1 h gave a high
(>91%) yield of a mixture of isomeric allylation products
consisting of the Z-isomer, branch isomer, and E-isomer.
The ratios of the isomers are shown as a%, b%, and c%,
respectively in Table 2. The formation of the branch
isomer makes it difficult to see the ratio of an anti-π-
allylpalladium intermediate to a syn-π-allylpalladium
intermediate at the moment of the nucleophilic attack
by the malonate, because the branch isomer can be
produced from both of the π-allylpalladium intermedi-
ates.22 The anti/syn ratio was conveniently estimated
by use of the results obtained for the allylic alkylation
of (E)-allyl acetates under the same conditions (Table
3). The (E)-acetates gave the E-isomer (d%) and the
branched isomer (e%), and only a trace amount of the
Z-products was detected, indicating that the reaction
starting from (E)-acetates proceeded via the syn-
intermediate almost exclusively. This is as expected
because the syn-isomer which does not have a sterically
bulky substituent at the central position is usually much
more thermodynamically stable than the anti-isomer.20
The higher stability of the syn-isomer was confirmed
by the 1H NMR analysis of the (1-phenyl-π-allyl)-
palladium complex coordinated with dppb 7, where the
syn-species is the sole observable species in acetone-d6
and the corresponding anti-species was not detected at
all (Scheme 7).
Assuming that the syn-intermediate always gives the
corresponding E and branched products with a fixed
ratio, a distribution between the anti-intermediate (x%)
and the syn-intermediate (y%) at the nucleophilic attack
can be estimated from the following equations.
x + y ) 100
a + b + c ) 100
d + e ) 100
y ) c(1 + d/e) ) 100c/e
As shown in Table 2, the ratio of the anti-intermediate
to syn-intermediate (x% to y%) at the nucleophilic attack
was strongly dependent on the bisphosphine ligand on
the palladium catalyst. General trends are similar for
the two (Z)-allyl acetates. In the reaction with the R,ω-
bis(diphenylphosphino)alkanes (dppe, dppp, and dppb),
the ratio of the syn-intermediate decreased as the length
(20) Cotton, F. A.; Faller, J . W.; Musco, A. Inorg. Chem. 1967, 6,
179.
(21) The rate constants of the reaction of π-allylpalladium-phoshine
complexes with some nucleophiles have been reported. Kuhn, O.; Mayr,
H. Angew. Chem., Int. Ed. 1999, 38, 343.
(22) The anti-isomers of the monosubstituted π-allylpalladium spe-
cies have a tendency to give relatively high degrees of the correspond-
ing branched products by the reactions with carbon soft nucleophiles.
See refs 4d and 8b.

4858 Organometallics, Vol. 21, No. 22, 2002
Ogasawara et al.
Ta ble 2. Allylic Alk yla tion of (Z)-Cin n a m yl Aceta te a n d (Z)-2-Hexen yl Aceta te Ca ta lyzed by
P a lla d iu m -Bisp h osp h in e Com p lexesa
Z
a (%)
branched
b (%)
E
c (%)
entry
R
phosphine
total yield (%)
anti x (%)b
syn y (%)b
1
2
3
4
5
6
7
8
Ph
dppe
dppp
dppb
dppf
MeO-dppf
CF3-dppf
dppe
dppp
dppb
24
58
51
24
33
38
8
39
80
73
72
81
14
30
46
75
65
57
2
61
13
3
1
2
>99
97
99
>99
>99
93
99
93
96
91
32
87
97
99
98
68
13
3
1
2
5
92
8
n-Pr
90
53
3
<1
<1
<1
9c
91c
54
3
8
46
97
>99
>99
>99
9
17
27
28
18
10
11
12
dppf
MeO-dppf
CF3-dppf
<1
<1
<1
99
97
a
The reactions were carried out with allyl acetate (0.3 mmol) and NaCMe(COOMe)2 (1.5 mmol) in the presence of 2 mol % (Pd) of the
b
catalyst generated from [PdCl(π-C3H5)]2 and bisphosphine in a mixture of acetone (1 mL) and THF (1 mL) at 20 °C for 1 h. The estimated
ratio of the intermediates at the nucleophilic attack. c Estimated in disregard of the Z-product obtained from the E-substrate (see Table
3).
an electronic effect of the dppf ligand.23 The electronic
Ta ble 3. Allylic Alk yla tion of (E)-Cin n a m yl Aceta te
a n d (E)-2-Hexen yl Aceta te Ca ta lyzed by
effect of the ferrocenylbisphosphine, which is a less basic
P a lla d iu m -Bisp h osp h in e Com p lexesa
triarylphosphine,23 decreases the reactivity at the oxi-
Z
branched
E
dative addition forming the π-allylpalladium intermedi-
ate and enhances the reactivity of the π-allylpalladium
intermediate toward the nucleophile attack. It follows
that the less basic character of dppf shortens the lifetime
of the π-allypalladium intermediate, and as a result, the
intermediate undergoes nucleophilic attack before its
isomerization from Z to E. In the reaction of (Z)-hexenyl
acetate, almost no isomerization of the π-allypalladium
intermediate was detected with either dppf, MeO-dppf,
or CF3-dppf (entries 10-12). The contribution of the syn-
intermediate was too small to compare the influence of
the dppf ligands. In the catalytic reaction of (Z)-
cinnamyl acetate, we were able to observe the isomer-
ization from Z to E at 8% with the CF3-dppf ligand
(entry 6), whose π-allylpalladium complex is fastest in
the π-σ-π isomerization. It is interesting that the
isomerization of the π-allylpalladium intermediates was
observed in the catalytic reaction with the CF3-dppf
catalyst system because the lifetime of its π-allylpalla-
dium intermediates is considered to be shortest of the
complexes examined, due to the substitution with the
electron-withdrawing CF3 groups on the diphenylphos-
phino groups.21
entry
R
phosphine (%)
d (%)
e (%) total yield (%)
1
2
3
4
5
6
7
8
Ph
dppe
dppp
dppb
dppf
MeO-dppf <1
CF3-dppf
<1
<1
<1
<1
10
3
8
12
8
57
2
1
1
1
1
90
97
92
88
92
43
95
98
98
99
99
98
94
96
97
98
97
<1
3
98
n-Pr dppe
dppp
>99
>99
>99
98
>99
>99
1
9
dppb
dppf
MeO-dppf <1
CF3-dppf
<1
10
11
12
<1
<1
1
a
The reactions were carried out with allyl acetate (0.3 mmol)
and NaCMe(COOMe)2 (1.5 mmol) in the presence of 2 mol % (Pd)
of the catalyst generated from [PdCl(π-C3H5)]2 and bisphosphine
in a mixture of acetone (1 mL) and THF (1 mL) at 20 °C for 1 h.
Sch em e 7
of polymethylene chains in the bisphosphines increases
(entries 1-3 and 7-9), which is in good agreement with
the syn-anti isomerization rates measured in the
π-allylpalladium complexes coordinated with dppe, dppp,
and dppb (see Table 1). It may be regarded that the
dppe-, dppp-, and dppb-palladium catalysts have similar
reactivity toward the nucleophile, because their elec-
tronic characteristics are not very different. The more
isomerization from Z to E with dppe than with dppb
has been reported by van Leeuwen,8b and it is in good
agreement with our present results.
In the allylic alkylation catalyzed by the dppf-pal-
ladium complex (entries 4 and 10), the contribution of
the syn-intermediate at the nucleophilic attack is much
smaller than expected from the rate of the syn-anti
isomerization of the dppf complex 4. The isomerization
rate of 4 was measured to be between those of the dppe
complex 1 and dppp complex 2, but the catalytic reaction
with the dppf complex gave only a trace amount of the
E-isomers. A probable explanation for these results is
Con clu sion s
The syn-anti isomerization rates in the π-allylpalla-
dium complexes [Pd(π-Me2CCHCHD)(bisphosphine)]-
BArF were measured for a variety of bisphosphine
4
ligands. It was revealed that the rates of the syn-anti
isomerization are dependent on both the bite angles and
electronic characteristics of the phosphine ligands. As
we have demonstrated in this report, the syn-anti
isomerization rate plays an important role in the
catalytic allylic alkylation which proceeds through π-al-
lylpalladium intermediates. Although the syn-anti
isomerization rate is not a sole factor determining the
selectivity of the reaction, it is still a decisive factor and
constitutes a very important part of the π-allylpalladium
chemistry. The results reported here suggest that
selectivity of π-allylpalladium-mediated reactions can
(23) Gan, K.-S.; Hor, T. S. A. In Ferrocenes; Togni, A., Hayashi, T.,
Eds.; VCH: Weinheim, 1995; pp 3-104.

Effects of Bidentate Phosphine Ligands
Organometallics, Vol. 21, No. 22, 2002 4859
in vacuo, and then vacuum-transferred to give 1.7 g (44% yield)
be controlled by selection of an appropriate phosphine
ligand, if the syn-anti isomerization is the crucial factor
determining the selectivity. This report will be a useful
guideline for choice of suitable phosphine ligands in
palladium-catalyzed allylic substitution reactions in-
cluding asymmetric ones.
1
2
of propanediol-d6. H NMR (CDCl3): δ 2.43 (br, 2H). H{1H}
NMR (CHCl3): δ 1.75 (br, 2D), 3.80 (br, 4D).
TsO(CD2)3OTs. To a solution of HO(CD2)3OH (1.00 g, 12.1
mmol) in pyridine (20 mL) was added p-toluenesulfonyl
chloride (5.03 g, 26.4 mmol). After stirring the mixture for 5
min, the cooling bath was removed and the reaction mixture
was vigorously stirred for 3 h. The volume of the mixture was
reduced to about half of the initial volume, and the mixture
was poured into ice-water. The ditosylate, deposited as a
white solid, was filtered, washed with water and hexane, and
then dried under reduced pressure. The residual solid was
recrystallized from CHCl3/hexane to give colorless crystals
Exp er im en ta l Section
Gen er a l P r oced u r es. All anaerobic and/or moisture-sensi-
tive manipulations were carried out with standard Schlenk
techniques under predried nitrogen or with glovebox tech-
niques under prepurified argon. NMR spectra were recorded
on a J EOL J NM LA500 spectrometer (1H, 500 MHz; 13C, 125
1
(1.97 g, 42% yield). H NMR (CDCl3): δ 2.46 (s, 6H), 7.35 (d,
J ) 8.3 Hz, 4H), 7.75 (d, J ) 8.3 Hz, 4H). 2H{1H} NMR
(CHCl3): δ 1.97 (br, 2D), 3.65 (br, 4D). Anal. Calcd for
2
1
Hz; 31P, 202 MHz; 19F, 470 MHz; H, 76.5 MHz). H and 13C-
{1H} chemical shifts are reported in ppm downfield of internal
tetramethylsilane. 31P{1H} and 19F NMR chemical shifts are
externally referenced to 85% H3PO4 and C6F6 (δ -162),
respectively. 2D NMR chemical shifts are reported in ppm
using CDCl3 (δ 7.24) as an internal standard. Tetrahydrofuran,
Et2O, and toluene were distilled from benzophenone-ketyl
under nitrogen prior to use. Dichloromethane was distilled
from CaH2 under nitrogen prior to use. Methanol was dried
over magnesium methoxide, distilled, and stored in a glass
flask with a Teflon stopcock under nitrogen. P(NMe2)Cl2,24 Na-
[B(C6H3-3,5-(CF3)2)4],17 Ph2P(CD2)4PPh2,16 Fe[η5-C5H4P(C6H4-
C
17H14D6O6S2: C, 52.29; H+D, 6.71. Found: C, 52.09; H+D,
6.42.
P h 2P (CD2)3P P h 2. A solution of LiPPh2, which was prepared
from PPh3 (2.55 g, 9.7 mmol) and Li (182 mg, 26.2 g atom) in
THF (8.3 mL) followed by t-BuCl (1.1 mL, 9.7 mmol) treatment,
was added to a THF (5 mL) solution of TsO(CD2)3OTs (1.89 g,
4.8 mmol) at 0 °C, and the mixture was refluxed for 1 h. After
cooling to room temperature, the solvent was removed under
reduced pressure and the residue was chromatographed on
silica gel with Et2O as an eluent under nitrogen atmosphere.
The NMR analysis of the product showed a small amount of
impurities; however, it was sufficiently pure for the synthesis
4-CF3)2]2,24a,25 [PdCl(η3-allyl)]2,26 and [PdCl(η3-1-phenylallyl)]2
7a
were synthesized as reported. Allyl acetates (cis- and trans-
2-hexenyl acetates, cis- and trans-3-phenyl-2-propenyl ace-
tates) were prepared from corresponding allyl alcohols and
acetic anhydride.
1
of the palladium complex. H NMR (CDCl3): δ 7.24-7.38 (m,
20H). 2H{1H} NMR (CHCl3): δ 1.56 (br, 2D), 2,16 (br, 4D).
31P{1H} NMR (CHCl3): δ -17.32.
(4-MeOC6H4)2P Cl. An ether solution of p-methoxyphenyl-
magnesium bromide, which was prepared from p-methoxy-
bromobenzene (25.3 g, 0.14 mol) and Mg (3.6 g, 0.15 mol) in
dry ether (70 mL), was added dropwise to a cold (-78 °C)
solution of Me2NPCl2 (7.3 g, 0.050 mol) and pyridine (20 mL,
0.25 mol) in dry ether (150 mL). The reaction mixture was
stirred for 2 h at room temperature and heated to reflux for a
further 2 h. The solution was cooled to room temperature, and
anhydrous HCl was bubbled through the reaction mixture for
1 h. The mixture consisted of two layers. The upper one was
collected and concentrated under reduced pressure to give 5.2
g of the crude (4-MeOC6H4)2PCl. This crude product was used
in the next reaction without further purification.
Mea su r em en ts of Exch a n ge Ra tes betw een syn - a n d
a n ti-Hyd r ogen s in th e P a lla d iu m Com p lexes. The solvent,
acetone-d6, was dried over activated 4 Å molecular sieves and
vacuum transferred prior to use. The NMR samples for the
exchange rate measurements were degassed by three freeze-
pump-thaw cycles and flame-sealed in the NMR tubes.
Determination of the exchange rates between the two isoto-
pomers was carried out according to the Forse´n-Hoffman
method.18 Spin-saturation transfer experiments were per-
formed by irradiating either of the Hsyn or Hanti signals of the
π-allyl complexes. The exchange rates, k, were calculated from
the following equation:
F e[η5-C5H4P (C6H4-4-OMe)2]2. To a suspension of ferrocene
(0.63 g, 3.4 mmol) in hexane (17.2 mL) were quickly added
nBuLi in hexane (1.54 M, 4.8 mL, 7.4 mmol) and TMEDA (0.81
g, 7.1 mmol), and the resulting orange solution was stirred at
60 °C for 1 h. The heating bath was removed, and 6.9 mL of
dry THF was added. The dark brown suspension was cooled
to -50 °C, and a solution of (4-MeOC6H4)2PCl (2.1 g, 7.5 mmol)
in THF (3.5 mL) was added dropwise over 5 min. The reaction
mixture was stirred overnight at room temperature. The
reaction mixture was diluted with Et2O, and insoluble solid
was removed by filtration. The filtrate was washed with dilute
HCl, saturated NaHCO3, and H2O, then dried over Na2SO4.
The resulting crude product was chromatographed on silica
gel (hexane/Et2O ) 1) to give the title compound in pure form.
Yield: 1.1 g (46%). 1H NMR (CDCl3): δ 3.79 (s, 12H), 3.97 (dd,
J ) 3.8 and 1.8 Hz, 4H), 4.25 (t, J ) 1.8 Hz, 4H), 6.82 (d, J )
8.8 Hz, 8H), 7.22 (d, J ) 8.8 Hz, 8H). 31P{1H} NMR (CHCl3):
δ -20.2.
1-Deu ter io-3-m eth yl-2-bu ten ol. To a suspension of LiAlD4
(5.0 g, 120 mmol) in Et2O (160 mL) was added dropwise
3-methyl-2-butenal (13.6 g, 160 mmol) over 30 min by means
of syringe. The reaction mixture was stirred at room temper-
ature for 1 h and heated under reflux for 1.5 h. The mixture
was cooled with ice-water and quenched with saturated
sodium sulfate. The mixture was filtered through a pad of
Celite, and the filter cake was rinsed with Et2O. The combined
organic solution was dried over MgSO4 and concentrated. The
I′/I ) τ/(τ + T1)
where I and I′ are the signal intensities of the nonirradiated
H (syn-H or anti-H) without and with saturation of the other
H resonance, respectively. T1 is the spin-lattice relaxation
time of the nonirradiated H resonance and τ ()1/k) is the pre-
exchange lifetime of this exchange system. The ratio I′/I was
calculated from the difference spectrum recorded by subtract-
ing the irradiated spectrum from the reference (nonirradiated)
spectrum. H NMR T1 determinations were performed with a
standard 180°-τ-90° pulse sequence by the inversion-
recovery method.
HO(CD2)3OH. Malonic acid-d4 (5.0 g, 46 mmol) in tetrahy-
drofuran (50 mL) was added dropwise to a suspension of
LiAlD4 (5.0 g, 0.12 mol) in tetrahydrofuran (100 mL) at 0 °C
by means of syringe. The reaction mixture was warmed to
room temperature gradually with stirring, then refluxed for 1
h. The mixture was quenched with aqueous NaOH (5% in H2O,
25 mL), and it was heated to reflux for 1 h. The cooled mixture
was filtered through a pad of Celite, and the filter cake was
rinsed with Et2O. The filtrate was dried (MgSO4), concentrated
1
(24) (a) Unruh, J . D.; Christenson, J . R. J . Mol. Catal. 1982, 14, 19.
(b) Issleib, K.; Seidel, W. Chem. Ber. 1959, 92, 2681.
(25) (a) Hamann, B. C.; Hartwing, J . F. J . Am. Chem. Soc. 1998,
120, 3694. (b) de Lang, R.-J .; van Soolinger, J .; Verkruijsse, H. D.;
Brandsma, L. Synth. Commun. 1995, 25, 2989.
(26) Tatsuno, A.; Yoshida, T.; Otsuka, S. Inorg. Synth. 1979, 19, 220.

4860 Organometallics, Vol. 21, No. 22, 2002
Ogasawara et al.
residue was vacuum-transferred to give the title compound
with a small amount of residual solvents. This crude material
was sufficiently pure for the next step and used without
further purification. Yield: 13.7 g (99%). H NMR (CDCl3): δ
1.21 (br, 1H), 1.69 (s, 3H), 1.75 (s, 3H), 4.11 (br, 1H), 5.43 (d,
(CHCl3): δ 1.61 (br, 4D), 2.51 (br, 4.5D), 3.59 (br, 0.5D). 19F
NMR (CDCl3): δ -63.57. 31P{1H} NMR (CDCl3): δ 20.53 (d, J
) 49.1 Hz, 1P), 22.87 (d, J ) 49.1 Hz, 1P). Anal. Calcd for
1
C65H40D8BF24P2Pd: C, 52.96; H+D, 3.96. Found: C, 52.75;
H+D, 3.67.
J ) 7.1 Hz, 1H).
Dp p f com p lex (4) was recrystallized from CHCl3/hexane
[P d Cl(η3-3-d eu ter io-1,1-d im eth yla llyl)]2. A mixture of
palladium(II) chloride (27.7 g, 156 mmol) and lithium chloride
(22.2 g, 523 mmol) was dissolved in hot water (35 mL), and to
this were added ethanol (300 mL), 1-deuterio-3-methyl-2-
butenol (13.7 g, 157 mmol) in tetrahydrofuran (100 mL), and
concentrated hydrochloric acid (100 mL), successively. Carbon
monoxide was passed through the solution for 4 h at room
temperature with vigorous stirring. During this period, the
solution became clear orange. After removing the solvents
under reduced pressure, the residue was dissolved in CHCl3
and the solution was washed with water, then dried over
MgSO4. Evaporation of the solvent gave a yellow solid. This
crude product was purified by chromatography over alumina
(with CHCl3), then by recrystallization from CH2Cl2/hexane.
Yield: 1.73 g (52%), mp 110-111 °C. 1H NMR (CDCl3): δ 1.25
(s, 3H), 1.44 (s, 3H), 3.08 (d, J ) 12.7 Hz, 0.5H), 3.84 (d, J )
7.4 Hz, 0.5H), 5.08 (m, 1H). 2H{1H} NMR (CHCl3): δ 3.09 (br,
0.5D), 3.85 (br, 0.5D). 13C{1H} NMR (CDCl3): δ 21.85, 21.86,
27.09, 55.57 (t, J ) 24.6 Hz), 95.08, 106.27, and 106.30. Anal.
Calcd for C10H16D2Cl2Pd2: C, 28.33; H+D, 4.75. Found: C,
28.18; H+D, 4.61.
(53% yield). Mp: 206-207 °C. H NMR (CDCl3): δ 1.07 (t, J
1
) 6.1 Hz, 3H), 1.09 (dd, J ) 10.8 and 6.1 Hz, 3H), 2.76 (dd, J
) 13.7 and 10.0 Hz, 0.5H), 3.60 (t, J ) 7.9 Hz, 0.5H), 3.87 (br,
1H), 4.13 (br, 1H), 4.19 (br, 1H), 4.34 (br, 1H), 4.40 (br, 1H),
4.41 (br, 1H), 4.45 (br, 1H), 4.51 (br, 1H), 5.19 (d, J ) 7.9 Hz,
0.5H), 5.20 (d, J ) 13.7 Hz, 0.5H), 7.34-7.63 (m, 20H), 7.50
(br, 4H), 7.71 (br, 8H). 2H{1H} NMR (CHCl3): δ 2.76 (br, 0.5D),
3.63 (br, 0.5D). 19F NMR (CDCl3): δ -63.56. 31P{1H} NMR
(CDCl3): δ 22.17 (d, J ) 45.2 Hz, 0.5P), 22.19 (d, J ) 45.2 Hz,
0.5P), 26.95 (d, J ) 45.2 Hz, 1P). Anal. Calcd for C71H48DBF24
-
FeP2Pd: C, 53.49; H+D, 3.16. Found: C, 53.25; H+D, 3.27.
F e[η5-C5H4P (C6H4-4-OMe)2]2 Com p lex (5) was purified by
preparative TLC (94% yield). 1H NMR (CDCl3): δ 1.06 (t, J )
6.1 Hz, 3H), 1.09 (dd, J ) 11.0 and 6.4 Hz, 3H), 2.76 (dd, J )
13.5 and 10.1 Hz, 0.5H), 3.56 (t, J ) 7.6 Hz, 0.5H), 3.83 (br,
6H), 3.83 (br, 3H), 3.88 (br, 1H), 4.11 (br, 1H), 4.14 (br, 1H),
4.31 (br, 1H), 4.35 (br, 2H), 4.42 (br, 1H), 4.45 (br, 1H), 5.17
(d, J ) 7.6 Hz, 0.5H), 5.18 (d, J ) 13.5 Hz, 0.5H), 6.86-7.05
(m, 8H), 7.28-7.44 (m, 8H), 7.51 (br, 4H), 7.70 (br, 8H). 2H-
{1H} NMR (CHCl3): δ 2.78 (br, 0.5D), 3.60 (br, 0.5D). 19F NMR
(CDCl3): δ -63.57. 31P{1H} NMR (CDCl3): δ 19.49 (d, J ) 47.9
Hz, 0.5P), 19.51 (d, J ) 47.9 Hz, 0.5P), 24.08 (d, J ) 47.9 Hz,
1P). Anal. Calcd for C75H56DBF24FeO4P2Pd: C, 52.55; H+D,
3.41. Found: C, 52.29; H+D, 3.32.
[P d (η3-3-d eu t er io-1,1-d im et h yla llyl)(b isp h osp h in e)]-
[B(C6H3-3,5-(CF 3)2)4] (1-6): Gen er a l P r oced u r e. A mixture
of [PdCl(3-deuterio-1,1-dimethylallyl)]2 (0.165 mmol) and the
bisphosphine (0.30 mmol) was suspended in methanol (0.6
mL). The solids dissolved after 10 min to give a yellow or
faintly orange solution. To the solution was added dropwise
Na[B(C6H3-3,5-(CF3)2)4] (316.9 mg, 0.33 mmol) in methanol (0.5
mL). The mixture was allowed to stand for 30 min at ambient
temperature and then at 0 °C overnight. For 1-4, the crude
complexes were crystallized from the solution. The precipitates
were collected and were washed with methanol/water (1:1) and
then with hexane. These materials were further purified by
recrystallization. For 5 and 6, which were not crystalline, the
reaction solutions were evaporated to dryness and the residual
complexes were purified by preparative thin-layer chromatog-
raphy over SiO2 (with CHCl3).
F e[η5-C5H4P (C6H4-4-CF 3)2]2 com p lex (6) was purified by
preparative TLC (66% yield). 1H NMR (CDCl3): δ 1.04 (t, J )
5.9 Hz, 3H), 1.13 (dd, J ) 11.5 and 6.6 Hz, 3H), 2.79 (dd, J )
13.7 and 10.0 Hz, 0.5H), 3.66 (t, J ) 7.4 Hz, 0.5H), 3.91 (br,
1H), 4.15 (br, 1H), 4.19 (br, 1H), 4.38 (br, 1H), 4.41 (br, 1H),
4.43 (br, 1H), 4.52 (br, 1H), 4.54 (br, 1H), 5.21 (d, J ) 7.4 Hz,
0.5H), 5.22 (d, J ) 13.7 Hz, 0.5H), 7.45-7.88 (m, 16H), 7.48
(br, 4H), 7.68 (br, 8H). 2H{1H} NMR (CHCl3): δ 2.83 (br, 0.5D),
3.72 (br, 0.5D). 19F NMR (CDCl3): δ -64.68 (s, 6F), -64.66 (s,
3F), -64.61 (s, 3F), -63.55 (s, 24F). 31P{1H} NMR (CDCl3): δ
22.05 (d, J ) 45.2 Hz, 0.5P), 22.07 (d, J ) 45.2 Hz, 0.5P), 27.01
(d, J ) 45.2 Hz, 1P). Anal. Calcd for C75H44DBF36FeP2Pd: C,
48.27; H+D, 2.48. Found: C, 48.04; H+D, 2.71.
Dp p e com p lex (1) was recrystallized from methanol (34%
Gen er a l P r oced u r es for P a lla d iu m -Ca ta lyzed Allylic
Alk yla tion . A typical procedure is given for the reaction of
(E)-2-hexenyl acetate with Pd-dppe complex. To a solution of
[PdCl(η3-allyl)]2 (1.1 mg, 6 µmol of Pd) and dppe (2.7 mg, 6.6
µmol) in acetone (1 mL) was added a THF solution of Na[MeC-
(COOMe)2] (1.5 mmol), which was prepared from dimethyl
methylmalonate and sodium hydride, at -78 °C. To the
solution was added (E)-2-hexenyl acetate (42.7 mg, 0.3 mmol)
at the same temperature, and the reaction mixture was stirred
at 20 °C for 1 h. The mixture was quenched with water and
extracted with ether. The combined organic layer was washed
with saturated NaCl, dried over MgSO4, and evaporated. The
residue was chromatographed on silica gel (EtOAc/hexane, 1:5)
to give a mixture of alkylated products (68.4 mg, 99.8% yield).
The ratio of the isomers in the mixture was determined by a
1
yield). Mp: 141-142 °C. H NMR (CDCl3): δ 1.01 (t, J ) 6.3
Hz, 3H), 1.84 (dd, J ) 9.8 and 8.3 Hz, 3H), 2.05-2.19 (m, 1H),
2.26-2.45 (m, 1H), 2.57-2.75 (m, 2H), 3.21 (dd, J ) 14.2 and
10.4 Hz, 0.5H), 4.21 (t, J ) 7.1 Hz, 0.5H), 5.30 (d, J ) 7.1 Hz,
0.5H), 5.31 (d, J ) 14.2 Hz, 0.5H), 7.18-7.63 (m, 20H), 7.49
(br, 4H), 7.70 (br, 8H). 2H{1H} NMR (CHCl3): δ 3.25 (br, 0.5D),
4.27 (br, 0.5D). 19F NMR (CDCl3): δ -63.59. 31P{1H} NMR
(CDCl3): δ 51.27 (d, J ) 2.6 Hz, 1P), 51.45 (d, J ) 2.6 Hz, 1P).
Anal. Calcd for
Found: C, 52.51; H+D, 3.10.
C63H44DBF24P2Pd: C, 52.61; H+D, 3.22.
Dp p p -d 6 com p lex (2) was recrystallized from CHCl3/
hexane (71% yield). Mp: 177-179 °C. 1H NMR (CDCl3): δ 1.04
(t, J ) 6.2 Hz, 3H), 1.32 (dd, J ) 10.3 and 7.9 Hz, 3H), 2.80
(dd, J ) 14.2 and 10.3 Hz, 0.5H), 3.71 (t, J ) 7.6 Hz, 0.5H),
5.16 (d, J ) 7.6 Hz, 0.5H), 5.18 (d, J ) 14.2 Hz, 0.5H), 7.15-
7.54 (m, 20H), 7.51 (br, 4H), 7.71 (br, 8H). 2H{1H} NMR
(CHCl3): δ 1.74 (br, 1D), 2.06 (br, 1D) 2.58 (br, 4.5D), 3.68
(br, 0.5D). 19F NMR (CDCl3): δ -63.56. 31P{1H} NMR
(CDCl3): δ 5.89 (d, J ) 63.4 Hz, 1P), 10.30 (d, J ) 63.4 Hz,
1P). Anal. Calcd for C64H40D7BF24P2Pd: C, 52.71; H+D, 3.73.
Found: C, 52.47; H+D, 3.80.
1
GC analysis, and the results are listed in Tables 2 and 3. H
NMR spectra of the products are shown below.
Dim eth yl 2-((Z)-2-h exen yl)-2-m eth ylp r op a n e-1,3-d io-
a te.27 1H NMR (CDCl3): δ 0.87 (t, J ) 7.4 Hz, 3H), 1.38 (sext,
J ) 7.4 Hz, 2H), 1.40 (s, 3H), 2.02 (q, J ) 7.4 Hz, 2H), 2.64 (d,
J ) 7.6 Hz, 2H), 3.72 (s, 6H), 5.24 (dt, J ) 11.0 and 7.6 Hz,
1H), and 5.53 (dt, J ) 11.0 and 7.4 Hz, 1H).
Dp p b-d 8 com p lex (3) was recrystallized from CH2Cl2/
pentane (70% yield). Mp: 189-191 °C. 1H NMR (CDCl3): δ
0.81 (t, J ) 5.9 Hz, 3H), 1.29 (dd, J ) 10.5 and 6.9 Hz, 3H),
2.73 (dd, J ) 13.7 and 10.1 Hz, 0.5H), 3.55 (t, J ) 7.8 Hz,
0.5H), 5.16 (d, J ) 7.8 Hz, 0.5H), 5.17 (d, J ) 13.7 Hz, 0.5H),
7.20-7.58 (m, 20H), 7.50 (br, 4H), 7.71 (br, 8H). 2H{1H} NMR
Dim eth yl 2-((E)-2-h exen yl)-2-m eth ylp r op a n e-1,3-d io-
a te.27 1H NMR (CDCl3): δ 0.87 (t, J ) 7.4 Hz, 3H), 1.35 (sext,
J ) 7.4 Hz, 2H), 1.38 (s, 3H), 1.96 (q, J ) 7.4 Hz, 2H), 2.55 (d,
(27) Sjo¨gren, M. P. T.; Frisell, H.; Åkermark, B.; Norrby, P.-O.;
Eriksson, L.; Vitagliano, A. Organometallics 1997, 16, 942.

Effects of Bidentate Phosphine Ligands
Organometallics, Vol. 21, No. 22, 2002 4861
J ) 6.9 Hz, 2H), 3.71 (s, 6H), 5.28 (dt, J ) 15.0 and 7.4 Hz,
1H), and 5.50 (dt, J ) 15.0 and 6.9 Hz, 1H).
added Na[B(C6H3-3,5-(CF3)2)4] (223 mg, 251 mmol) in methanol
(0.3 mL) dropwise. The mixture was stirred for 30 min at
ambient temperature, then the solvent was removed under
reduced pressure. The residue was purified by chromatography
on silica gel with CHCl3 to give the title complex in pure form.
Yield: 106 mg (57%). 1H NMR (CDCl3): δ 1.65-2.05 (br, 4H),
2.25-2.74 (br, 4H), 2.91 (t, J ) 12.5 Hz, 1H), 3.81 (t, J ) 7.8
Hz, 1H), 4.84 (t, J ) 12.5 Hz, 1H), 5.99 (td, J ) 12.5 and 7.8
Hz, 1H), 6.64-6.77 (m, 3H), 6.82-6.96 (m, 2H), 7.03-7.60 (m,
Dim et h yl 2-(1-h exen -3-yl)-2-m et h ylp r op a n e-1,3-d io-
a te.27 1H NMR (CDCl3): δ 0.87 (t, J ) 7.4 Hz, 3H), 1.38 (sext,
J ) 7.4 Hz, 2H), 1.39 (s, 3H), 2.02 (q, J ) 7.6 Hz, 2H), 2.72 (t,
J ) 7.6 Hz, 1H), 3.68 (s, 3H), 3.71 (s, 3H), 5.05 (dd, J ) 16.9
and 1.9 Hz, 1H), 5.09 (dd, J ) 10.2 and 1.9 Hz, 1H), and 5.57
(dd, J ) 16.9 and 10.2 Hz, 1H).
Dim eth yl 2-((Z)-3-ph en yl-2-pr open yl)-2-m eth ylpr opan e-
1
20H), 7.51 (s, 4H), and 7.73 (s, 8H). 19F NMR (CDCl3):
δ
1,3-d ioa te. H NMR (CDCl3): δ 1.41 (s, 3H), 2.93 (d, J ) 7.2
-63.53. 31P{1H} NMR (CDCl3): δ 19.47 (d, J ) 53.1 Hz, 1P)
and 25.21 (d, J ) 53.1 Hz, 1P). Anal. Calcd for C69H49BF24P2-
Pd: C, 52.71; H, 3.73. Found: C, 52.47; H, 3.80.
Hz, 2H), 3.68 (s, 6H), 5.53 (dt, J ) 11.7 and 7.2 Hz, 1H), 6.57
(d, J ) 11.7 Hz, 1H), and 7.18-7.37 (m, 5H).
Dim eth yl 2-((E)-3-ph en yl-2-pr open yl)-2-m eth ylpr opan e-
1,3-d ioa te.11 1H NMR (CDCl3): δ 1.46 (s, 3H), 2.77 (d, J )
7.5 Hz, 2H), 3.73 (s, 6H), 6.08 (dt, J ) 15.8 and 7.5 Hz, 1H),
6.45 (d, J ) 15.8 Hz, 1H), and 7.18-7.37 (m, 5H).
Ack n ow led gm en t. This work was supported in part
by a Grant-in-Aid for Scientific Research from the
Ministry of Education, Science, Sports, and Culture,
J apan.
Dim eth yl (1-p h en yl-2-p r op en yl)-2-m eth ylp r op a n e-1,3-
d ioa te.11 1H NMR (CDCl3): δ 1.43 (s, 3H), 3.62 (s, 3H), 3.68
(s, 3H), 4.15 (d, J ) 8.8 Hz, 1H), 5.11 (d, J ) 16.9 Hz, 1H),
5.14 (d, J ) 10.3 Hz, 1H), 6.32 (ddd, J ) 16.9, 10.3, and 8.8
Hz, 1H), and 7.16-7.37 (m, 5H).
Su p p or tin g In for m a tion Ava ila ble: Tables of crystal-
lographic data for complex 3. This material is available free
of charge via the Internet at http://pubs.acs.org.
[P d(η3-1-ph en ylallyl)(dppb)][B(C6H3-3,5-(CF3)2)4]. A mix-
ture of [PdCl(1-phenylallyl)]2 (30 mg, 58 mmol) and dppb (49
mg, 116 mmol) was dissolved in methanol (0.9 mL) and stirred
for 10 min to give a pale orange solution. To the solution was
OM020680+