Backbone effect of MAP ligands on their coordination patterns with palladium(II)

Article abstract of DOI:10.1021/om020475u

The allylpalladium complexes (R)-4 and (R)-5 of chiral P, N auxiliaries (R)-1 and (R)-3 have been prepared. The X-ray structures of the complexes have been determined, and the results clearly show that two different chelating modes (P, N vs P, C_σ) are involved. The coordination patterns of donating atoms were dependent on the binaphthyl backbone of the chiral ligands. The isodesymmetrization revealed that the reaction occurs without regiochemical memory effect, which demonstrates a full dynamic equilibrium has been attained before nucleophilic attack. These results, combined with the solution behaviors of the complexes revealed by NMR studies, provided a further understanding of their asymmetric induction in allylic substitution of 1,3-diphenylprop-2-en-1-yl acetate with dimethyl malonate. The dramatic enhancement of enantioselectivity with (R)-3 as a chiral inducer can be related to the different chelating mode of catalyst (R)-5 from that of (R)-4. The change of bite angles of the two donors with palladium in the complexes (R)-4 and (R)-5 was considered as a direct reason for their dramatically different enanatioselectivities in the catalysis.

Full text of DOI:10.1021/om020475u

1856
Organometallics 2003, 22, 1856-1862
Ba ck bon e Effect of MAP Liga n d s on Th eir Coor d in a tion
P a tter n s w ith P a lla d iu m (II)
Yi Wang, Xin Li, J ie Sun, and Kuiling Ding*
The State Key Laboratory of Organometallic Chemistry, Shanghai Institute of
Organic Chemistry, Chinese Academy of Sciences, 354 Fenglin Lu, Shanghai 200032, China
Received J une 17, 2002
The allylpalladium complexes (R)-4 and (R)-5 of chiral P, N auxiliaries (R)-1 and (R)-3
have been prepared. The X-ray structures of the complexes have been determined, and the
results clearly show that two different chelating modes (P, N vs P, Cσ) are involved. The
coordination patterns of donating atoms were dependent on the binaphthyl backbone of the
chiral ligands. The isodesymmetrization revealed that the reaction occurs without regio-
chemical memory effect, which demonstrates a full dynamic equilibrium has been attained
before nucleophilic attack. These results, combined with the solution behaviors of the
complexes revealed by NMR studies, provided a further understanding of their asymmetric
induction in allylic substitution of 1,3-diphenylprop-2-en-1-yl acetate with dimethyl malonate.
The dramatic enhancement of enantioselectivity with (R)-3 as a chiral inducer can be related
to the different chelating mode of catalyst (R)-5 from that of (R)-4. The change of bite angles
of the two donors with palladium in the complexes (R)-4 and (R)-5 was considered as a direct
reason for their dramatically different enanatioselectivities in the catalysis.
In tr od u ction
was developed by Kocovsky’s and our group indepen-
dently.7 This ligand exhibited a dramatic acceleration
of the amination of aryl halide,8 but only moderate
asymmetric induction in Pd(0)-catalyzed allylic substi-
tution of 1,3-diphenylprop-2-en-1-yl acetate with di-
methyl malonate.7a Kocovsky also demonstrated by
X-ray crystal structural analysis that MAP and MOP
ligands coordinated to Pd via unusual P, Cσ-chelation
and P, η2-coordination modes,9,10 respectively, rather
than a P, N- or P, O-chelation pattern. Buchwald et al.
reported that the palladium complex of the similar
aminophosphine ligand 2 showed excellent reactivity
and enantioselectivity in asymmetric cross-coupling
reactions.11 Our previous work on the backbone effect
of MAP ligands in Pd-catalyzed allylic substitution
revealed that an octahydro analogue (H8-MAP, 3) of 1
provided much better asymmetric induction (82.5% ee
vs 73% ee).12a We speculated that an P, N-chelated Pd
complex was responsible for the catalysis and the
enhanced enantioselectivity may be attributed to the
Homogeneous enantioselective catalysis with chiral
transition metal complexes is an attractive synthetic
methodology since it is able to generate large quantities
of enantiopure products using only a small amount of
chiral catalyst.1 Many well-known chiral metal catalysts
containing palladium salts and chiral P, N auxiliaries
show good asymmetric induction in the reactions such
as Heck,2 Michael,3 cross-coupling,4 and allylic substitu-
tion.5 Recently, a novel class of aminophosphine ligand
1 (MAP) containing an sp3 nitrogen donor, which could
be regarded as a nitrogen analogue of Hayashi’s MOP,6
* Corresponding author. E-mail: kding@pub.sioc.ac.cn. Fax: 86-21-
6416-6128.
(1) (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis;
Wiley: New York, 1994. (b) Catalytic Asymmetric Synthesis, 2nd ed.;
Ojima, I., Ed.; VCH: New York, 2000. (c) Comprehensive Asymmetric
Catalysis; J acobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer-
Verlag: Berlin, 1999. (d) Lin, G. Q.; Li, Y. M.; Chan, A. S. C. Principles
and Applications of Asymmetric Synthesis; Wiley: New York, 2001.
(2) (a) Loiseleur, O.; Pfaltz, A. Angew. Chem., Int. Ed. Engl. 1996,
35, 200. (b) Loiseleur, O.; Hayashi, M.; Schmees, N.; Pfaltz, A. Synthesis
1997, 1338. (c) Loiseleur, O.; Hayashi, M.; Keenan, M.; Schmees, N.;
Pfaltz, A. J . Organomet. Chem. 1999, 576, 16. (d) Deng, W.; Hou, X.;
Dai, L.; Dong, X. Chem. Commun. 2000, 1483.
(6) (a) Uozumi, Y.; Hayashi, T. J . Am. Chem. Soc. 1991, 113, 9887.
(b) Hayashi, T. Acta Chem. Scand. 1996, 50, 259.
(7) (a) Vyskocil, S.; Smrcina, M.; Hanus, V.; Polasek, M.; Kocovsky,
P. J . Org. Chem. 1998, 63, 7738. (b) Ding, K.; Wang, Y.; Yun, H.; Liu,
J .; Wu, Y.; Terada, M.; Okubo, Y.; Mikami, K. Chem. Eur. J . 1999, 5,
1734.
(3) Hu, X.; Chen, H.; Zhang, X. Angew. Chem., Int. Ed. 1999, 38,
3518.
(4) (a) Uemura, M.; Nishimura, H.; Hayashi, T. Tetrahedron Lett.
1993, 34, 107. (b) Hayashi, T.; Niizuma, S.; Kamikawa, T.; Suzuki,
N.; Uozumi, Y. J . Am. Chem. Soc. 1995, 117, 9101.
(8) Vyskocil, S.; Smrcina, M.; Kocovsky, P. Tetrahedron Lett. 1998,
39, 9289.
(9) Kocovsky, P.; Vyskocil, S.; Cisarova, I.; Sejbal, J .; Tislerova, I.;
Smrcina, M.; Lloyd-J ones, G. C.; Stephen, S. C.; Butts, C. P.; Murray,
M.; Langer, V. J . Am. Chem. Soc. 1999, 121, 7714.
(10) (a) Lloyd-J ones, G. C.; Stephen, S. C.; Murray, M.; Butts, C.
P.; Vyskocil, S.; Kocovsky, P. Chem. Eur. J . 2000, 6, 4348. (b) Fairlamb,
I. J . S.; Lloyd-J ones, G. C.; Vyskocil, S.; Kocovsky, P. Chem. Eur. J .
2002, 8, 4443.
(11) (a) Yin, J .; Buchwald, S. L. J . Am. Chem. Soc. 2000, 122, 12051.
(b) Chieffi, A.; Kamikawa, K.; Ahman, J .; Fox, J . M.; Buchwald, S. L.
Org. Lett. 2001, 3, 1897-1900. (c) Hamada, T.; Chieffi, A.; Ahman, J .;
Buchwald, S. L. J . Am. Chem. Soc. 2002, 124, 1261.
(12) (a) Wang, Y.; Guo, H.; Ding, K. Tetrahedron: Asymmetry 2000,
11, 4153. (b) Guo, H.; Wang, Y.; Ding, K. J . Chin. Chem. 2001, 19, 52.
(5) (a) Sprinz, J .; Helmchen, G. Tetrahedron Lett. 1993, 34, 1769.
(b) Matt, P. V., Pfaltz, Angew. Chem., Int. Ed. Engl. 1993, 32, 566. (c)
Dawson, G. J .; Frost, C. G.; Williams, J . M. J .; Coote, S. J . Tetrahedron
Lett. 1993, 34, 3149. (d) Ogasawara, M.; Yoshida, K.; Kamei, H.; Kato,
K.; Uozumi, Y.; Hayashi, T. Tetrahedron: Asymmetry 1998, 9, 1779.
(e) Imai, Y.; Zhang, W.; Kida, T.; Nakatsuji, Y.; Ikeda, I. Tetrahedron
Lett. 1998, 39, 4343. (f) Selvakumar, K.; Valentini, M.; Woerle, M.;
Pregosin, P. S.; Albinati, A. Organometallics 1999, 18, 1207. (g) Deng,
W.-P.; Hou, X.-L.; Dai, L.-X.; Yu, Y.-H.; Xia, W. Chem. Commun. 2000,
285. (h) Deng, W.-P.; You, S.-L.; Hou, X.-L.; Dai, L.-X.; Yu, Y.-H.; Xia,
W.; Sun, J . J . Am. Chem. Soc. 2001, 123, 6508. (i) You, S.; Zhu, X.;
Hou, X.; Dai, L. J . Am. Chem. Soc. 2001, 123, 7471.
10.1021/om020475u CCC: $25.00 © 2003 American Chemical Society
Publication on Web 03/27/2003

Backbone Effect of MAP Ligands
Organometallics, Vol. 22, No. 9, 2003 1857
change of bite angle of the Pd complex after partial
reduction of binaphthyl backbone. All these distinct
results prompted us to elucidate what happened to our
catalyst system containing (R)-3. In this article, we
report our results on the structural study of allyl
complexes of (R)-1/Pd and (R)-3/Pd. The possible path-
way of asymmetric induction was proposed on the basis
1
of structural and H NMR information of catalysts and
of the results of stoichiometric alkylation.
Resu lts a n d Discu ssion
F igu r e 1. Molecular structure of (R)-4, showing the atom-
numbering scheme. PF6 has been omitted for clarity.
-
P r ep a r a tion of th e Allyl Com p lexe (R)-4. The
reactions of ligand (R)-1 with [PdCl2 (PhCN)2] and
[Pd(η3-C3H3) (MeCN)2]+OTf- have been demonstrated
by Kocovsky to give P, Cσ-chelated palladium complexes.
However, the structure of the real catalyst for asym-
metric allylation of 1,3-diphenylprop-2-en-1-yl acetate
with dimethyl malonate remains unknown. To mimic
the catalytic system, a diphenylallyl Pd precursor,
[Pd(Ph(η3-C3H3)Ph)Cl]2, was reacted with ligand (R)-1.
As shown in Scheme 1, (R)-1 and palladium dimer
Thermal ellipsoids are drawn at the 30% probability level.
Ta ble 1. Bon d Len gth s [Å] a n d An gles [d eg]
for (R)-4
bond lengths
bond angles
Pd-C41 2.214(12)
Pd-C42 2.182(10)
Pd-C43 2.320 (10)
Pd-C11 2.338(11)
Pd-P1 2.283(3)
C41-Pd-C43 66.3(4)
P1-Pd-C11 82.5(2)
C43-Pd-C11 111.8(4)
C41-Pd-P1 99.0(3)
N-C20 1.365(15)
N-C21 1.437(18)
C6-C7 1.39(2)
Sch em e 1a
C11-C12 1.454(15)
C11-C20 1.439(15)
C18-C19 1.36(2)
C19-C20 1.491(18)
of C18-C19 is only 1.36 Å, which is responsible for a
conjugated CdC double bond. Particularly, the bond
length of C20-N (1.365(15) Å) is much shorter than a
normal N-C single bond (1.40 Å), which suggests a
partial CdN double-bond feature. Moreover, the planar
arrangement of the NMe2 group (Figure 1) is an
alternative demonstration of the sp2 hybridization at the
nitrogen, which again supports the presence of a CdN
double bond. The significant difference of bond lengths
between C11-Pd (2.338(11) Å) and C12-Pd (2.863 Å)
excluded the possibility for the formation of an η2-
coordinated complex. All these results again support the
observation by Kocovsky in allylic Pd/1 complex.9,10 The
stronger trans effect of the P atom than that of the C11
atom can be observed by comparison of the bond length
of C43-Pd (2.320 (10) Å) with that of C41-Pd (2.214 (12)
Å). A similar phenomenon was also observed in an
analogous allyl Pd complex by Kocovsky.9,10 The distinct
structural feature is that the distances of allyl moiety
carbon atoms from Pd in (R)-4 increase by g0.1 Å in
comparison with those observed in the unsubstituted
allyl complex.9 These differences may be attributed to
the increased steric demand of the 1,3-diphenylallyl
moiety in the complex. Accordingly, the nucleophile will
attack at the allylic C43 carbon occupying the position
trans to the P atom to give the S product, as observed
experimentally.12a
a
(i) [Pd(Ph(η3-C3H3)Ph)Cl]2, CH2Cl2, rt; (ii) AgPF6, 62%.
[Pd(Ph(η3-C3H3)Ph)Cl]2 were dissolved in dichloro-
methane and stirred at room temperature for 10 min.
After the addition of AgPF6, the reaction mixture was
stirred in the dark at room temperature for an ad-
ditional 1 h. The resulting AgCl was filtered off through
Celite and washed with dichloromethane. The filtrate
was concentrated under reduced pressure, and the
residue was recrystallized from CH2Cl2/heptane to af-
ford (R)-4 as dark red crystals in 62% yield.
X-r a y a n d NMR Stu d ies on Com p lex (R)-4. A
single crystal of (R)-4 was selected from the crystals
obtained above and submitted to X-ray diffraction
analysis. A view of the molecular structure is shown in
Figure 1, and selected bond lengths and angles are listed
in Table 1.
The results of the X-ray diffraction analysis clearly
demonstrated that ligand (R)-1 acts as a P, C-ligand in
the complex (R)-4 with an unusual Cσ-Pd bonding
mode. The distance between Pd and C11 was 2.338(11)
Å, which is an indication of single C11-Pd bond forma-
tion. Moreover, the bond lengths of C11-C12 and C11-
C20 are 1.454 and 1.439 Å, respectively, which show
some features of a C-C single bond. The bond length
The behavior of (R)-4 in solution was also investigated
1
by H NMR and 31P NMR in CDCl3. A mixture of three
1
species (11:9:1) was detected by H NMR spectroscopy.

1858 Organometallics, Vol. 22, No. 9, 2003
Wang et al.
F igu r e 2. Different orientations (M and W types) of
diphenylallyl unit in the complex (R)-4.
Two abundant species with the singlet peaks of NMe2
at δ 2.80 and 2.60 can be attributed to the P, Cσ-chelated
Pd complex with the “M” and “W” rotamers of diphen-
ylallyl moieties (Figure 2), respectively. Two groups of
signals at δ 4.71-4.87 and 5.43-5.60 are indicative of
allyl protons. Each of them included two broad peaks
with a ratio of 11:9, which again demonstrated the
presence of different orientations of diphenylallyl moi-
eties in the complex. The minor species with the singlet
of NMe2 at δ 3.45 can be assigned to a P, N-chelated Pd
complex, because the coordination of the nitrogen atom
to Pd will significantly decrease the shielding of protons
at the N-methyl group. This can be further verified by
the crystal structure and 1H NMR spectrum of (R)-5.
Two singlets in the 31P NMR spectrum at δ 40.21 and
37.23 also confirmed the presence of P-Pd coordination
and provided circumstantial evidence for the presence
of two rotamers (M and W type) in the complex. These
results mirror those observed by Kocovsky and co-
workers, although they did not detect the minor P,N-
chelate species.9,10 This may be attributed to the steric
bulk of the 1,3-diphenylallyl moiety in (R)-4, for which
the more crowded P,C-chelate may not be as favorable
as in the case of the unsubstituted allyl Pd complex of
1.
F igu r e 3. Molecular structure of (R)-5, showing the atom-
numbering scheme. OTf- and hydrogen atoms have been
omitted for clarity. Thermal ellipsoids are drawn at the
30% probability level.
controlled by the reactivity difference between M and
W types of (R)-4 complex toward the nucleophile rather
than their ratio in solution, because an M/W equilibrium
has been reached before the nuleophilic attack occurs
(this will be discussed below on the basis of the absence
of memory effect in the catalytic system disclosed by
isotopical experiment).
P r ep a r a tion of Allyl Com p lex (R)-5. In our previ-
ous work, it was postulated that the Pd complex of (R)-3
would adopt a P, N-chelating pattern because of a
higher energy level of the intermediate involved in TS
1 than that in TS 2 (1,3-butadiene vs benzene struc-
ture).12a The significant enhancement of enantioselec-
tivity of the reaction with the catalysis of Pd/(R)-3 in
comparison with Pd/(R)-1 may be attributed to the
change of bite angle of the Pd complex after partial
reduction of the binaphthyl backbone. Therefore, the
preparation and X-ray diffraction characterization of the
Pd complex of (R)-3 will provide direct evidence for
elucidating this problem. Following a similar procedure
for the preparation of (R)-4, complex (R)-5 was obtained
by the reaction of (R)-3 with [Pd(η3-C3H5)(CH3CN)2]+OTf-
in dichloromethane at room temperature. Removal of
the solvent and recrystallization of the residue from
CH2Cl2/EtOAc afforded (R)-5 as yellow prisms in 51%
yield.
Asym m etr ic In d u ction w ith Ca ta lytic a n d Sto-
ich iom etr ic Am ou n ts of (R)-4 in Allylic Su bstitu -
tion . To correlate the crystal and solution structures
of complex (R)-4 with its asymmetric induction, the
reaction was carried out in CHCl3 in the presence of a
stoichiometric amout of (R)-4 (Scheme 2). It was found
Sch em e 2
that (R)-4 reacted with malonate smoothly in the
presence of Cs2CO3 to give the product (S)-6 with 50.1%
ee. This result is comparable to that (40.5% ee, S)
obtained using the catalyst (5 mol %) prepared in situ
from the reaction of (R)-1 and [Pd(C3H5)Cl]2. This result
indicated that the allylic substitution of 1,3-diphenyl-
prop-2-en-1-yl acetate with dimethyl malonate most
probably took place through a Cσ, Pd-coordinated com-
plex, and the enantioselectivity of the reaction was
X-r a y a n d NMR Stu d ies on Com p lex (R)-5. A
single crystal of (R)-5 was selected from the crystals
obtained above and submitted to X-ray diffraction
analysis. A view of the molecular structure is shown in
Figure 3, and selected bond lengths and angles are listed
in Table 2. The results clearly show that ligand (R)-3
acts as a P, N-ligand in the complex (R)-5 to form a
N-Pd coordination bond rather than the case of (R)-4,

Backbone Effect of MAP Ligands
Organometallics, Vol. 22, No. 9, 2003 1859
Ta ble 2. Bon d Len gth s [Å] a n d An gles [d eg]
demonstrates the formation of N-Pd coordination in
(R)-5 in solution.
for (R)-5
An experiment using the isolated P, N-connected
bond length
bond angle
-
[Pd(Ph(C3H3)Ph)((R)-3)]+PF6 complex ((R)-7) of H8-
Pd-N 2.193(3)
Pd-C35 2.191(5)
Pd-C36 2.174 (10)
Pd-C37 2.105(6)
Pd-P 2.2981(10)
N-C20 1.473(5)
N-C21 1.460(6)
C6-C7 1.479(9)
C11-C12 1.413(5)
C11-C20 1.419(5)
C18-C19 1.355(6)
C19-C20 1.373(6)
N-Pd-P 98.90(9)
C35-Pd-N 101.4(2)
C37-Pd-P 92.6(2)
C37-Pd-C35 67.2(3)
MAP as substrate to react with malonate was carried
out in CDCl3 at room temperature in order to compare
its asymmetric induction with the case of using a
catalytic amount of (R)-3/Pd complex (Scheme 4). The
Sch em e 3a
with Cσ-Pd bonding. The bond lengths of P-Pd and
N-Pd are 2.2981(10) and 2.139(3) Å, respectively, which
are indicative of a normal coordination bond between
Pd and the P or N atom.13 The bond distance of N-C20
(1.473(5) Å) is longer than that observed in (R)-4
(1.365(15) Å) or that of a normal N-C single bond (1.40
Å) in the 2-amino-2′-hydroxy-1,1′-binaphthyl (NOBIN)
ligand.7b This phenomenon is probably caused by the
coordination of lone-pair electrons of N to Pd, which
prevents n-π conjugation between the NMe2 group and
the aromatic ring. The bond length of Pd-C35 (2.191(5)
Å) is longer than that of Pd-C37 (2.105(6) Å), which
again indicates that the P-donor has a stronger trans-
influence than the N-donor. Accordingly, the allylation
reaction will take place by nucleophilic attack at the
carbon atom trans to the P-donor in principle. Interest-
ingly, the bite angle of the two donors with palladium
(98.90°) was much larger than that of the allyl pal-
ladium complex of (R)-1 (84.21°).9 Although we did not
obtain the crystal structure of the 1,3-diphenylallyl
palladium complex of (R)-3 (H8-MAP), it is reasonable
to assume that changing the ligand backbone from MAP
to H8-MAP triggers the coordination change from the
P, C- to the P, N-pattern, independently of the allyl used
(allyl or diphenyl allyl). Therefore, the change of bite
angles between the two donors and palladium in (R)-1
and (R)-3 could be an important reason for the dramatic
enhancement of the asymmetric induction level with
(R)-3 in allylic substitution.14
a
(i) [Pd(η3-C3H5)(CH3CN)2]+OTf-, CH2Cl2.
Sch em e 4
allylation product (S)-6 was obtained in 54% yield with
80.4% ee, which is essentially identical to that (82.1%
ee) achieved by using 5 mol % of the catalyst prepared
in situ from the reaction of (R)-3 and [Pd(C3H5)Cl]2.
These results reflect that the reaction involves the same
asymmetric induction pathway using either catalytic or
stoichiometric amount of H8-MAP/Pd complexes for the
reaction.
Regioch em ica l Mem or y Effect of th e Ca ta lysis.
An important question arises regarding the regiochemi-
cal memory effect in the catalysis of the allylic substitu-
tion. The study of such memory effect can be an indica-
tion of the situation of dynamic equilibrium before
nucleophilic attack. Kocovsky reported that Pd com-
plexes of MAP did not exhibit a regiochemical memory
effect in the reaction of isocinnamyl acetate with [NaCR-
(CO2Me)2] (R ) H or Me). However, the reaction of
cyclopentenyl pivalate proved that the reaction occurs
with powerful stereochemical memory effects. To inves-
tigate whether the memory effect is involved in the
present catalytic system, the R-D labeled (()-1,3-diphen-
ylprop-2-en-1-yl acetate, (()-8, was taken as the sub-
strate for the reaction (Scheme 5). The reaction was
carried out in dichloromethane with 5 mol % of Pd
catalyst bearing (R)-1 or (R)-3 prepared in situ in the
presence of Cs2CO3 as the base at room temperature
(25 °C). The reaction proceeded smoothly to give allylic
substitution product (S)-6 in quantitative yields. The
1H NMR spectrum of the product (Figure 4) shows that
(S)-R-D-6 and (S)-γ-D-6 was formed in a 1:1 molar ratio,
which clearly demonstrates that there is no regiochemi-
cal memory effect in the present catalytic system. The
reaction of isocinnamyl acetate (()-9 with dimethyl
malonate with the catalysis of (R)-3/Pd in the presence
of Cs2CO3 gives the branched product 10 as the major
isomer, which mirrors the result obtained by Kocovsky
using (R)-1/Pd catalyst (i.e., no memory effect).10 All
Solu tion Beh a vior of Com p lex (R)-5 a n d Its
Asym m etr ic In d u ction w ith Ca ta lytic a n d Sto-
ich iom etr ic Am ou n ts of Ca ta lysts in Allylic Su b-
stitu tion . The solution behavior of complex (R)-5 was
also studied by 1H NMR, 31P NMR, and 13C NMR
spectra. The 31P NMR and 13C NMR spectra of (R)-5
indicate that the chemical shifts of 31P (19.90 and 21.23
ppm) and 13C (52.69 ppm for NMe2) are much higher
than those of its free ligand (-14.3 ppm for 31P and
43.32 ppm for NMe2, respectively), which strongly
supports the formation of P, N-chelation with Pd in
1
solution. The signals of NMe2 in the H NMR spectrum
of (R)-5 appearing at 3.27 and 3.33 ppm in almost a 1:1
ratio show the presence of two diastereoisomers of (R)-5
in solution because of the spatial orientation of allyl
unit. The significant downfield shift of Me2N in the
complex relative to that of free ligand (2.23 ppm) again
(13) Cahil, J . P.; Bohnen, F. M.; Goddard, R.; Kruger, C.; Guiry, P.
J . Tetrahedron: Asymmetry 1998, 9, 3831.
(14) (a) Trost, B. M.; Breit, B.; Peukert, S.; Zambrano, J .; Ziller, J .
W. Angew. Chem., Int. Ed. Engl. 1995, 34, 2386. (b) Trost. B. M.; Van
Vranken, D. L.; Bingel, C. J . Am. Chem. Soc. 1992, 114, 9327.

1860 Organometallics, Vol. 22, No. 9, 2003
Wang et al.
palladium in the complexes (R)-4 and (R)-5 was consid-
ered as a direct reason for their dramatically different
enanatioselectivities in the catalysis.
Exp er im en ta l Section
1H NMR spectra were recorded in CDCl3 on a Varian
Mercury 300 spectrometer at 25 °C. Chemical shifts are
expressed in ppm with TMS as an internal standard (δ ) 0
ppm). The 31P NMR spectra were recorded on a Varian
Mercury 300 instrument in CDCl3 with 85% H3PO4 as an
external reference. Optical rotation was measured with a PE-
341 automatic polarimeter. Liquid chromatographic analyses
were conducted on a J ASCO 1580 system. The IR spectra were
measured on a Rio-Rad FTS-185 spectrometer in KBr pellets.
ESI spectra were obtained on a Mariner LC-TOF spectrometer.
Elemental analysis was performed with an Elemental VARIO
EL apparatus. Commercial reagents were used as received
without further purification unless otherwise noted. Dichlo-
romethane was freshly distilled from calcium hydride. (R)-1
and (R)-3 were prepared according to the literature methods.7b,12
Syn th esis of (R)-4. [Pd(Ph(η3-C3H3)Ph)Cl]2 (100 mg, 0.148
mmol) and (R)-1 (156.8 mg, 0.326 mmol) in dried dichlo-
romethane (10 mL) were stirred at room temperature for 10
min. After the addition of AgPF6 (70.5 mg, 0.729 mmol), the
reaction mixture was stirred in the dark at room temperature
for an additional 1 h. The AgCl formed was filtered through
Celite and washed with dichloromethane. The filtrate was
concentrated under reduced pressure, and the residue was
F igu r e 4. 1H NMR spectrum of product (S)-6 showing a
1:1 mixture of (S)-R-D-6 and (S)-γ-D-6.
Sch em e 5
recrystallized from CH2Cl2/heptane to afford (R)-4 as a dark
20
red crystal (93.2 mg, 62% yield). Mp: 180 °C (dec). [R]D
)
1
-2355° (c 0.03, CH2Cl2). H NMR (300 MHz, CDCl3): δ 2.60
(br, 3H), 2.80 (br, 3H), 4.71-4.87 (br, 1H), 5.29 (s, 2H), 5.43-
5.60 (br, 1H), 5.89-6.20 (m, 2H), 6.74-7.73 (m, 29H), 7.83 (d,
J ) 8.1 Hz, 1H), 7.91 (d, J ) 8.6 Hz, 1H). 13C NMR (100.6
MHz, CDCl3): δ 29.65, 53.43, (other peaks cannot be inter-
preted clearly). 31P NMR (161.9 MHz): δ 37.2, 40.2. IR (KBr
pellet): ν 3053 (w), 1611 (m), 1599 (m), 1539 (m), 1514 (m),
1436 (m), 1097 (m), 838 (vs), 752 (m), 691 (s), 557 (s). ESIMS
(m/z): 780.2 ([M - PF6 - CH2Cl2]+, 100%). Anal. Calcd for
these results indicated that in the reaction of (()-1,3-
diphenylprop-2-en-1-yl acetate with dimethyl malonate
in the presence of (R)-1- or (R)-3-modified Pd complex
the nuleophilic attack takes place after the full dynamic
equilibrium has been attained. The enantioselectivity
of the reaction was controlled by the reactivity difference
between M- and W-type rotamers of the 1,3-diphenyl-
allyl palladium complexes (R)-4 and (R)-5 with the
nucleophile rather than their ratio in solution.
C
49H41F6NP2Pd‚CH2Cl2: C, 59.39; H, 4.29; N, 1.39. Found: C,
59.56; H, 4.32; N, 1.36.
Syn th esis of (R)-5. [Pd(η3-C3H5)(CH3CN)2]+OTf- (94.5 mg,
0.25 mmol) and (R)-3 (122.3 mg, 0.25 mmol) in dried dichlo-
romethane (4 mL) were stirred at room temperature for 10
min. The solution was concentrated under reduced pressure.
The crude product was recrystallized from CH2Cl2/EtOAc to
give (R)-5 as yellow prisms (100 mg, 51% yield). Mp: 286-
Con clu sion
20
1
287 °C. [R]D ) -31.9° (c 0.51, CH2Cl2). H NMR (300 MHz,
CDCl3): δ 0.74 (br, 1H), 0.89-0.94 (m, 1H), 1.24-1.32 (m, 3H),
1.63-1.90 (m, 6H), 2.05-2.29 (m, 3H), 2.50 (br, 1H), 2.84-
2.95 (m, 3H), 3.27 (s, 3H), 3.33 (s, 3H), 3.95-4.54 (m, 1H),
4.74-4.77 (m, 1H), 5.76-5.93 (m, 1H), 6.89-7.53 (m, 14H).
13C NMR (100.6 MHz, CDCl3): δ 21.86, 22.36, 22.61, 27.52,
27.73, 29.16, 29.85, 52.69, 123.00, 123.18, 128.81, 128.95,
129.09, 129.47, 129.58, 130.34, 130.64, 130.67, 130.93, 131.67,
131.70, 132.42, 132.58, 134.74, 134.94, 135.29, 137.81, 142.62,
142.65, 145.52. 31P NMR (161.9 MHz): δ 19.9, 21.2. IR (KBr
pellet): ν 3059 (w), 2926 (s), 1589 (w), 1576 (w), 1479 (s), 1461
(m), 1438 (s), 1263 (vs), 1221 (s), 1149 (vs), 1031 (vs), 745 (m),
719 (m), 693 (s), 637 (vs). ESIMS (m/z): 636.2 ([M - OTf]+,
100%). Anal. Calcd for C38H41F3NO3PPdS: C, 58.05; H, 5.26;
N, 1.78. Found: C, 58.13; H, 5.46; N, 1.72.
X-r a y Cr ysta l Str u ctu r e Deter m in a tion s of Com p lexes
(R)-4 a n d (R)-5. A single crystal of (R)-4 and (R)-5 suitable
for X-ray diffraction study was obtained by recrystallization
at room temperature from heptane/CH2Cl2 and EtOAc/CH2-
Cl2 solutions, respectively. The X-ray diffraction intensity data
were collected on a Bruker Smart CCD diffractometer at 20
°C using Mo KR radiation (λ ) 0.71073 Å) with a ω-2θ scan
mode.
In conclusion, the allylpalladium complexes (R)-1 and
(R)-3 have been prepared and characterized by X-ray
single crystal diffraction analyses. The results clearly
demonstrate that two kinds of chelating modes (P, N
vs P, Cσ) of coordination donors with the Pd atom have
been involved in two kinds of complexes. The coordina-
tion patterns of donating atoms were dependent on the
binaphthyl backbone of the chiral ligands. The isode-
symmetrization experiments revealed that the reaction
occurs without regiochemical memory effect, which
demonstrates a full dynamic equilibrium has been
attained before nucleophilic attack. These results, com-
bined with the solution behaviors of the complexes
revealed by NMR studies, provide a further understand-
ing of their asymmetric induction in allylic substitution
of 1,3-diphenylprop-2-en-1-yl acetate with dimethyl
malonate. The dramatic enhancement of enantioselec-
tivity with (R)-3 as a chiral inducer can be related to
the different chelating mode of catalyst (R)-5 from that
of (R)-4. The change of bite angle of the two donors with

Backbone Effect of MAP Ligands
Organometallics, Vol. 22, No. 9, 2003 1861
Ta ble 3. Cr ysta l Da ta a n d Exp er im en ta l Deta ils
Cl. The mixture was extracted with EtOAc (3 × 20 mL), and
the organic phase was washed with saturated aqueous NaH-
CO3, water, and brine, dried over Na2SO4, and concentrated
under reduced pressure. The residue was purified by flash
chromatograghy on silica gel with hexane/EtOAc (4:1) as
eluent to give (S)-6 (14.5 mg, 24%) as a colorless oil with
40.5% ee.
for Com p lexes (R)-4 a n d (R)-5
(R)-4
(R)-5
formula
fw
C
49H41F6NP2Pd
C38H41F3NO3PPdS
786.15
926.17
space group
a (Å)
b (Å)
P21
P212121
9.717(18)
23.259(4)
10.320(19)
10.2167(4)
13.3715(5)
28.7307(12)
Asym m etr ic Allylic Su bstitu tion of [P d (P h (C3H3)P h )-
((R)-3)] Com p lex ((R)-7) w ith Dim eth yl Ma lon a te. [Pd(Ph-
(η3-C3H3)Ph)Cl]2 (62 mg, 0.093 mmol) and (R)-3 (100 mg, 0.204
mmol) in dried dichloromethane (5 mL) were stirred at room
temperature for 10 min. After the addition of AgPF6 (45 mg,
0.180 mmol), the reaction mixture was stirred in the dark at
room temperature for an additional 1 h. The AgCl formed was
filtered through Celite and washed with dichloromethane. The
filtrate was concentrated under reduced pressure to give
orange solids of (R)-7 in quantitative yield. This complex was
taken as a substrate directly without further purification to
react with dimethyl malonate. To a solution of (R)-7 (45.0 mg,
0.045 mmol) in dried CHCl3 (1 mL) were added Cs2CO3 (30.0
mg, 0.092 mmol) and dimethyl malonate (12.2 mg, 0.092
mmol), respectively. The reaction mixture was stirred at room
temperature for 12 h. The mixture was filtered through Celite,
and the filtrate was concentrated under reduced pressure. The
residue was submitted to chromatographic separation on silica
gel with hexane/EtOAc (4:1) as eluent to give (S)-6 as a
colorless oil (8.5 mg, 54%) with 80.4% ee.
c (Å)
R (deg)
â (deg)
92.517(3)
γ (deg)
V (Å3)
2330.1(7)
2
1.320
3631.4(3)
4
1.438
Z
Dcalcd (g/cm3)
cryst size (mm)
µ(Mo KR) (cm-1
orientation reflns:
range (2θ) (deg)
0.15 × 0.20 × 0.20 0.25 × 0.20 × 0.30
)
5.23
6.65
2.84-56.60
3.50-50.00
no. of unique data, total 6106
8565
with I > 2.00σ(I)
no. of params refined
correct. factors,
max. min.
4712
547
0.4695-1.000
7742
454
0.9366-1.000
Ra
0.0785
0.1849
1.052
0.0469
0.1124
1.057
b
Rw
quality of fit indicatorc
max. shift/esd final cycle 0.034
0.110
0.677
-0.728
largest peak, e-/Å3
min. peak, e-/Å3
1.895
-0.696
En a n tioselective Allyla tion Ca ta lyzed by in Situ -
Gen er a ted (R)-3/P d Ca ta lyst. To a Schlenk tube containing
1,3-diphenylprop-2-en-l-yl acetate (54 mg, 0.23 mmol), [Pd-
(C3H5)Cl]2 (2.0 mg, 0.0055 mmol), (R)-3 (7.8 mg, 0.016 mmol),
and Cs2CO3 (139 mg, 0.37 mmol) was added 2 mL of CHCl3.
The mixture was stirred at room temperature for 20 min, and
dimethyl malonate (48.9 mg, 0.37 mmol) was then added. The
reaction mixture was stirred at room temperature for 24 h
before it was quenched with saturated aqueous NH4Cl. The
mixture was extracted with EtOAc (3 × 20 mL), and the
organic phase was washed with saturated aqueous NaHCO3,
water, and brine, dried over Na2SO4, and concentrated under
reduced pressure. The residue was purified by flash chro-
matograghy on silica gel with hexane/EtOAc (4:1) as eluent
to give (S)-6 (64 mg, 86%) as a colorless oil with 82.1% ee.
Syn th esis of (()-1-[2H1]-1,3-Dip h en ylp r op -2-en -1-yl Ac-
eta te (()-8. 1,3-Diphenylpropenone (3.70 g, 17.8 mmol) in
Et2O (20 mL) was added dropwise to a suspension of LiAl-
[2H4] (0.36 g, 8.60 mmol) in Et2O (10 mL). After complete
addition, water (20 mL) was added dropwise (CAUTION!). The
mixture was extracted with Et2O (7 × 20 mL), and the organic
phase was washed with saturated aqueous NaHCO3, water,
and brine, dried over Na2SO4, and concentrated under reduced
pressure. The residue was used for the next step without
further purification. To a stirred solution of the residue
obtained above (2.12 g, 10 mmol) in Et3N (10 mL) was added
(Ac)2O (5.60 mL, 0.06 mmol). After 12 h, the reaction mixture
was poured into water, extracted with Et2O (3 × 50 mL), dried
over MgSO4, and concentrated in vacuo. The residue was
submitted to flash chromatography separation on silica gel
with ethyl acetate/hexane (1:4) as eluent to give a 1:1 mixture
of (()-1-[2H1]-1,3-diphenylprop-2-en-1-yl acetate, (()-8, and
(()-1,2,3-[2H3]-1,3-diphenylpropyl acetate in >99% yield as a
colorless oil. 1H NMR (CDCl3, 300 MHz): δ 2.13(s, 1.5H), 2.17
(s, 1.5H), 2.56-2.70 (m, 1H), 6.34 (d, J ) 16.2 Hz, 0.5 H), 6.64
(d, J ) 16.2 Hz, 0.5 Hz), 7.13-7.43 (m, 10H).
a
b
2
R ) ∑||Fo| - |Fc||/∑|Fo|. Rw ) [∑w(|Fo| - |Fc|)2/∑w|Fo| ]1/2
;
.
w ) I/σ2(|Fo|). c Quality-of-fit ) [∑w(|Fo| - |Fc|)2/(Nobs - Nparams)]1/2
The structures of (R)-4 and (R)-5 were solved by direct
methods (SHELX-97) and refined by full-matrix least squares.
The non-hydrogen atoms were refined anisotropically, while
the hydrogen atoms were included but not refined. The final
cycle of full-matrix least-squares refinement gave agreement
factors of R ) 0.0785 and Rw ) 0.1849 for (R)-4 and R ) 0.0469
and Rw ) 0.1124 for (R)-5.
The details of the crystallographic data and the procedures
used for data collection and reduction information for (R)-4
and (R)-5 are given in Table 3. Selected bond lengths and
angles are listed in Tables 1 and 2. The atomic coordinates
and Biso/Beq, anisotropic displacement parameters, complete
bond lengths and angles, and least-squares planes for them
are given in the Supporting Information. The molecular
structures of (R)-4 and (R)-5 are given in Figures 1 and 2,
respectively.
Asym m etr ic Allylic Su bstitu tion of (R)-4 w ith Di-
m eth yl Ma lon a te. To a solution of (R)-4 (45.0 mg, 0.046
mmol) in dried CHCl3 (1 mL) were added Cs2CO3 (30.0 mg,
0.092 mmol) and dimethyl malonate (12.2 mg, 0.092 mmol),
respectively. The reaction mixture was stirred at room tem-
perature for 12 h. The mixture was filtered through Celite,
and the filtrate was concentrated under reduced pressure. The
residue was submitted to chromatographic separation on silica
gel with hexane/EtOAc (4:1) as eluent to give (S)-6 as a
colorless oil (9.5 mg, 64%) with 50.1% ee (determined by HPLC
on Chiralpak AD column, hexane/i-PrOH, 90:10, flow rate )
1 mL/min, tR ) 12.6 min, tS ) 18.9 min). 1H NMR (CDCl3, 300
MHz): δ 3.52 (s, 3H), 3.69 (s, 3H), 3.97 (d, J ) 10.9 Hz, 1H),
4.27 (dd, J ) 10.9, 8.7 Hz, 1 H), 6.34 (dd, J ) 15.4, 8.4 Hz,
1H), 6.48 (d, J ) 15.8 Hz, 1H), 7.17-7.24 (m, 10H).
En a n tioselective Allyla tion Ca ta lyzed by in Situ -
Gen er a ted (R)-1/P d Ca ta lyst. To a Schlenk tube containing
1,3-diphenylprop-2-en-l-yl acetate (46.5 mg, 0.20 mmol),
[Pd(C3H5)Cl]2 (1.7 mg, 0.0046 mmol), (R)-1 (6.7 mg, 0.014
mmol), and Cs2CO3 (120.3 mg, 0.370 mmol) was added 2 mL
of CHCl3. The mixture was stirred at room temperature for
20 min, and dimethyl malonate (48.9 mg, 0.37 mmol) was then
added. The reaction mixture was stirred at room temperature
for 24 h before it was quenched with saturated aqueous NH4-
Sea r ch for th e Mem or y Effect in P a lla d iu m -Ca ta lyzed
Allylic Su bstitu tion . The following procedure is typical: To
a stirred solution containing (()-1-[2H1]-1,3-diphenylprop-2-
en-1-yl acetate, (()-8 (43.5 mg, 0.18 mmol), [Pd(C3H5)Cl]2 (1.6
mg, 0.0044 mmol), (R)-3 (6.3 mg, 0.013 mmol), and Cs2CO3
(112.5 mg, 0.34 mmol) was added 2 mL of CH2Cl2. The mixture
was stirred at room temperature for 20 min, and dimethyl
malonate (45.0 mg, 0.34 mmol) was then added. The reaction

1862 Organometallics, Vol. 22, No. 9, 2003
Wang et al.
mixture was stirred at room temperature for 24 h before it
was quenched with saturated aqueous NH4Cl. The mixture
was extracted with EtOAc (3 × 20 mL), and the organic phase
was washed with saturated aqueous NaHCO3, water, and
brine, dried over Na2SO4, and concentrated under reduced
pressure. The residue was purified by flash chromatograghy
on silica gel with hexane/EtOAc (4:1) as eluent to give (S)-6
(55 mg, >99%) as a colorless oil with 80% ee. 1H NMR (CDCl3,
300 MHz): δ 3.51 (s, 3H), 3.70 (s, 3H), 3.93-3.97 (m, 1H),
4.23-4.30 (m, 0.5H), 6.20-6.34 (m, 1H), 6.47 (d, J ) 15.9 Hz,
0.5H), 7.19-7.34 (m, 10H). 1H NMR spectrum shows that (S)-6
is a 1:1 mixture of (S)-R-D-6 and (S)-γ-D-6.
the organic phase was washed with saturated aqueous NaH-
CO3, water, and brine, dried over Na2SO4, and concentrated
under reduced pressure. The residue was purified by flash
chromatograghy on silica gel with hexane/EtOAc (4:1) as
eluent to give 10 (75 mg, 88%) as a colorless oil. 1H NMR
(CDCl3, 300 MHz): δ 2.81 (td, J ) 1.2, 7.5 Hz, 2H), 3.53 (m,
1H), 3.75 (s, 6H), 6.15 (m, 1H), 6.48 (d, J ) 16.5 Hz, 1H), 7.35-
7.21 (m, 5H).
Ack n ow led gm en t. Financial support from the Na-
tional Natural Science Foundation of China, the Chinese
Academy of Sciences, the Major Basic Research Devel-
opment Program of China (Grant No. G2000077506),
and the Science and Technology Commission of Shang-
hai Municipality is gratefully acknowledged.
Following the same procedure mentioned above, 5 mol % of
(R)-1/Pd was taken as the catalyst for the same reaction, and
a 1:1 mixture of (S)-R-D-6 and (S)-γ-D-6 was obtained in 93%
yield with 43.9% ee.
Rea ction of Isocin n a m yl Aceta te (()-9 w ith Dim eth yl
Ma lon a te w ith th e Ca ta lysis of (R)-3/P d . To a Schlenk tube
containing isocinnamyl acetate (()-910 (60.6 mg, 0.344 mmol),
[Pd(C3H5)Cl]2 (3.2 mg, 0.0086 mmol), (R)-3 (12.7 mg, 0.026
mmol), and Cs2CO3 (225 mg, 0.688 mmol) was added 2 mL of
CH2Cl2. The mixture was stirred at room temperature for 20
min, and dimethyl malonate (90.8 mg, 0.688 mmol) was then
added. The reaction mixture was stirred at room temperature
for 24 h before it was quenched with saturated aqueous NH4-
Cl. The mixture was extracted with EtOAc (3 × 20 mL), and
Su p p or tin g In for m a tion Ava ila ble: Text giving a full
listing of crystallographic data for (R)-4 and (R)-5, including
tables of positional and isotropic equivalent displacement
parameters, calculated positions of the hydrogen atoms, aniso-
tropic displacement parameters, and bond distances and
angles. This material is available free of charge via the
Internet at http://pubs.acs.org.
OM020475U