The regioselectivity of the asymmetric intermolecular Heck reaction with planar chiral diphosphine-oxazoline ferrocenyl ligands

Article abstract of DOI:10.1002/chem.200204450

A series of novel planar chiral diphosphine-oxazoline ferrocenyl ligands were synthesized and used efficiently in the palladium-catalyzed asymmetric intermolecular Heck reaction of 2,3-dihydrofuran with aryl triflate and cyclohexenyl triflate. The tuning of the regioselectivity was realized by means of different palladium precursors and by changing the electronic factor of the ligands. A plausible rationale based on the existed mechanism is provided.

Full text of DOI:10.1002/chem.200204450

FULL PAPER
The Regioselectivity of the Asymmetric Intermolecular Heck Reaction with
Planar Chiral Diphosphine Oxazoline Ferrocenyl Ligands
Tao Tu,[a] Wei-Ping Deng,[a] Xue-Long Hou,*[a] Li-Xin Dai,*[a] and Xi-Cheng Dong[b]
Abstract: A series of novel planar chiral diphosphine-oxazoline ferrocenyl ligands
were synthesized and used efficiently in the palladium-catalyzed asymmetric
Keywords: electronic effects ¥ fer-
intermolecular Heck reaction of 2,3-dihydrofuran with aryl triflate and cyclohexenyl
rocenyl ligands ¥ Heck reaction ¥
triflate. The tuning of the regioselectivity was realized by means of different
palladium ¥ regioselectivity
palladium precursors and by changing the electronic factor of the ligands. A plausible
rationale based on the existed mechanism is provided.
Introduction
Since the first reports in the late
1980s,[1] the asymmetric Heck re-
action has received considerable
attention,[2] in particular, the intra-
molecular mode has been widely
applied in organic synthesis.[3] The
asymmetric intermolecular reac-
tion between an olefin and an aryl
source has been devised mainly for
cyclic substrates (predominantly
dihydrofuran), which was first re-
ported by Hayashi and co-workers
Scheme 1. The asymmetric intermolecular Heck reaction with 2,3-dihydrofuran and phenyl triflate.
in 1991.[4] High enantioselectivities have been achieved by the
use of bidentate ligands, usually diphosphines or phosphine-
oxazolines.[4, 5] When these ligands are employed in the
reaction of dihydrofuran (1) with phenyl triflate (2), two
different products 3 and 4 are obtained (Scheme 1) as a result
of the possibility of double-bond isomerization. For diphos-
phine-type ligands, such as BINAP,[4] TMBTP,[6, 7] and oth-
ers,[8, 9] the compound with the migrated double bond,
2-phenyl-2,3-dihydrofuran (4), is obtained as the major
product;[4, 10] while for P,N-type ligands, such as PHOX,[5]
1,1'-P,N-ferrocene ligand[11] and others,[12, 13, 14] the 2-phenyl-
2,5-dihydrofuran (3) is formed predominantly. However, the
reason for these strikingly different regioselectivities or how
to control the regioselectivity is not clear.[15, 16] To understand
the controlling or tuning of the regioselectivity, it is important
to gain an insight into the mechanism of the asymmetric
intermolecular Heck reaction.
In our previous studies of chiral ferrocenyl ligands[17, 18] and
the role played by their planar chirality,[11] we designed and
synthesized a series of new diphosphine oxazoline ferrocenyl
ligands, which belong to a system synthetically convenient for
modification,[19] and applied them to the asymmetric Heck
reaction. Herein, we disclose the results of the use of these
diphosphine oxazoline ferrocenyl ligands in the asymmetric
intermolecular Heck reaction. The regioselectivity of this
reaction depends on different palladium sources as well as the
electronic factor of the ligands.
[a] Prof. X.-L. Hou, Prof. L.-X. Dai, T. Tu, Dr. W.-P. Deng
State Key Laboratory of Organometallic Chemistry
Shanghai Institute of Organic Chemistry
Chinese Academy of Sciences
354 Fenglin Lu, Shanghai 200032 (China)
Fax : (‡86)21-6416-7519
E-mail: Xlhou@pub.sioc.ac.cn., Dailx@pub.sioc.ac.cn.
[b] Dr. X.-C. Dong
Key Laboratory of Computer Chemistry
Shanghai Institute of Organic Chemistry
Chinese Academy of Sciences
354 Fenglin Lu, Shanghai 200032 (China)
Fax : (‡86)21-6416-7519
Results and Discussion
Synthesis of ligand (S,Sp)-7a: As shown in Scheme 2, ferro-
cenyl oxazoline derivative 5, that bears a tert-butyl group on
E-mail: Dongxc@pub.sioc.ac.cn
Chem. Eur. J. 2003, 9, 3073 3081
DOI: 10.1002/chem.200204450
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3073

X.-L. Hou, L.-X. Dai et al.
FULL PAPER
[Pd2(dba)3 ¥ dba] in the reaction.
Interestingly, a reversed ratio
67:33 for the two isomers 3 and
4 was obtained, whereby isomer
3 was formed with 96.8% ee
(Table 2, entries 2 and 1). Fur-
thermore, when the nonpolar
solvent toluene was employed,
Scheme 2. Synthesis of the planar chiral diphosphine oxazoline ferrocenyl ligand (S,Sp)-7a.
the oxazoline ring,[20] was treated with nBuLi in THF at
À788C for 0.5 h, followed by trapping with Ph2PCl to afford
1-diphenyl-phosphino-1'-oxazolinylferrocene (6a). Directed
diastereoselective ortho-lithiation of 6a according to Ri-
chard×s and our procedure,[11a, 21] followed by trapping with
chlorodiphenylphosphine, gave diphosphine oxazoline fer-
rocenyl ligands (S,Sp)-7a, which contained not only a diphos-
phine fragment but also a phosphine oxazoline fragment.
an exciting ratio 95:5 was observed for 3 and 4 (Table 2,
entry 3). When a proton sponge and trietheylamine were used
instead of diisopropylethylamine, the observed regioselectiv-
Table 2. Effect of palladium precursors on the regioselectivity of Heck
reaction.[a]
Entry Palladium
Solvent Base
Conversion Ratio
[%][b] of 3/4[b] 3(R) 4(S)
ee[%][c]
1
2
3
4
5
6
7
8[d]
9[d]
10[e]
11[f]
[Pd(dba)3 ¥ dba] THF
iPr2NEt 98
iPr2NEt 89
toluene iPr2NEt 85
28:72
67:33
95:5
83:17
91:9
84:16
55:45
37:63
60:40
61:39
60:40
97
97
97
94
95
97
99
86
95
58
95
29
33
26
39
27
nd
nd
28
27
26
16
Asymmetric Heck reaction: With 1.5 mol% of catalyst,
prepared in situ from [Pd2(dba)3 ¥ dba] and 3 mol% of ligand
(S,Sp)-7a, the reaction of 1 with 2 in THF was completed in
36 h with a high conversion to afford a mixture of regioisom-
ers 3 and 4 in a ratio of 28:72 (Table 1, entry 1). The
enantiomeric excess (ee) of product 3 was 97.1% ee, while
that of the double-bond migration product 4 was low for a
kinetic resolution process in the reaction.[10]
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(OAc)2]
THF
toluene PS
toluene Et3N
70
70
[Pd(O2CCF3)2] toluene iPr2NEt 65
[Pd(O2CCF3)2] THF
[Pd(dba)3 ¥ dba] THF
[Pd(dba)3 ¥ dba] toluene iPr2NEt 84
[Pd(dba)3 ¥ dba] toluene iPr2NEt 70
[Pd(dba)3 ¥ dba] toluene iPr2NEt 87
iPr2NEt 65
iPr2NEt 95
[a] The reaction was carried out with 1.5 mol% palladium precursor and
3 mol% ligand (S,Sp)-7a at 608C for 36 h. [b] The conversion and ratio of 3/4
were determined by GC with n-undecane as the internal standard based on
Table 1. Regioselective and enantioselective Heck reaction of 2,3-dihydro-
furan.
phenyl triflate (2). [c] Determined by GC on
a Chiral G-PN column.
[d] Addition of 3 mol% (nBu)4NOAc (2 equiv/Pd) as additive. [e] Addition
of 7.5 mol% (nBu)4NOAc (5 equiv/Pd) as additive. [f] Addition of 3 mol%
LiOAc (2 equiv/Pd) as additive.
Entry Base
Solvent Conversion[%][a] Ratio of 3/4[a]
ee[%][b]
3(R)[c] 4(S)[c]
ities did not improved (Table 2, entries 4 and 5). When
[Pd(O2CCF3)2] was used in toluene, a ratio of 84:16 was
obtained with a conversion of 65% (Table 2, entry 6). Again,
The polar solvent favored the formation of isomer 4 (Table 2,
entry 7). It seems that [Pd(dba)3 ¥ dba] favors the formation
of product 4, while [Pd(OAc)2] prefers the unisomerized
product 3.
1
iPr2NEt THF
98
98
99
80
28:72
40:60
67:33
55:44
40:60
35:65
25:75
97
96
98
98
91
93
98
29
30
13
29
10
nd
nd
2
3
4
5
6
7
PS
Et3N
THF
THF
NaOAc THF
iPr2NEt toluene 96
iPr2NEt DMF 90
iPr2NEt (CH2Cl)2 80
Ligand (S,Sp)-7a has three binding sites which means that
there are three possible bidentate coordination modes:
N,P, N,P' and P,P' (Scheme 3). We originally thought that the
[a] The conversion and ratio of 3/4 were determined by GC with n-undecane
as the internal standard based on phenyl triflate (2). [b] Determined by GC on
a Chiraldex G-PN column. [c] The absolute configuration of the products was
assigned by comparison of the sign of specific rotations with literature
data.[4, 5] nd ˆ not determined.
To improve the regioselectivity of the reaction, the effect of
bases and solvents was studied.[5, 10] With a Proton Sponge as
the base, the ratio of 3:4 increased to 40:60 (Table 1, entry 2).
However, when triethylamine was employed, product 3 was
obtained as major product with 98.3% ee (Table 1, entry 3).
Inorganic bases, such as NaOAc, gave higher ratios than
diisopropylethylamine, but with lower conversion (Table 1,
entry 4). Solvents more polar than toluene, such as DMF or
1,2-dichorlethane, favored the formation of isomer 4 (Table 1,
entries 6 and 7).
Scheme 3. Three possible coordination modes of (S,Sp)-7a with Pd0.
difference in regioselectivity might be caused by the different
coordination modes of the ligand with different palladium
precursors, and the very high ratio of isomer 3 with [Pd(OAc)2]
is probably caused by a P,N coordination mode.[5] Therefore,
the 31P NMR spectroscopy of ligand (S,Sp)-7a with
In view of the fact that [Pd(OAc)2] and [Pd2(dba)3 ¥ dba] are
two commonly used precursors of palladium species in the
asymmetric Heck reaction, [Pd(OAc)2] was used instead of
3074
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 3073 3081

Asymmetric Intermolecular Heck Reaction
3073 3081
[Pd2(dba)3 ¥ dba] and [Pd(OAc)2] in C6D6 were examined to
assess what kind of coordination mode was taken in the reaction.
The chemical shifts of the two phosphorus nuclei in the ligand
(S,Sp)-7a were d ˆ À16.85 (s) and À17.16 ppm (s) in C6D6.
After reacting (S,Sp)-7a with [Pd2(dba)3 ¥ dba] for 30 minutes,
the corresponding phosphorus signals were shifted to down-
field regions (d ˆ 25.97 (brs) and 25.05 ppm (brs)). When
phenyl triflate was added to the mixture, no obvious change of
the phosphorus signals was found. These results indicate that
ligand (S,Sp)-7a might function as a diphosphine ligand in this
case (C, Scheme 3). It has been reported that diphosphine
ligands with [Pd(OAc)2] usually afford product 4 predom-
inantly;[10] however, when ligand (S,Sp)-7a and [Pd(OAc)2]
were applied in this reaction, product 3 was obtained as the
major product, although chelation of two phosphorus atoms
with the palladium center was again revealed by 31P NMR
spectroscopy (the two phosphorus signals were also at d ˆ
25.97 (brs) and 25.05 ppm (brs)) in this case.[22]
To gain further structural information on palladium species
coordinated to the diphosphine oxazoline ferrocenyl ligand,
the crystal structure of complex 9, which was synthesized from
ligand 8 and [PdCl2(CH3CN)2] in benzene, was examined by
X-ray diffraction (Scheme 4 and Table 3). Ligand 8 bearing
Figure 1. Structure of complex 9 (ORTEP diagram). Selected bond lengths
À
À
À
[ä] and angles [8]: Pd P1 2.2850(13), Pd P2 2.2913(15), Pd Cl2
À
À
À
À
2.3375(14), Pd Cl1 2.3457(16), P1 C1 1.813(6), P1 C23 1.809(6), P1 C17
À
À
À
1.833(5), P2 C6 1.816(6), P2 C29 1.821(6), P2 C35 1.823(6); P1-Pd-P2
99.75(5), P1-Pd-Cl2 170.65(5), P1-Pd-Cl1 86.35(5), P2-Pd-Cl2 86.80(5), P2-
Pd-Cl1 162.22(6), Cl1-Pd-Cl2 89.35(6), C1-P1-Pd 115.08(18), C6-P2-Pd
122.75(18), C1-Fe-C6 109.1(2).
Scheme 4.
Table 3. Selected crystal and structure refinement data for complex 9 ¥
H2O.
caused by the different coordination modes of ligand with
palladium.
formula
fw
C40H39NO2P2Cl2FePd
860.81
The six-membered heterometallocyclic ring in 9 formed by
chelate coordination of ligand 8 is highly skewed (Figure 1). It
can be seen that considerable steric hindrance exists between
the oxazoline ring and the upper phosphino-phenyl ring. The
P1-Pd-P2 bite angle of 99.75(5)8 was larger than that
measured for [Cl2Pd(dppf)] (99.078)[23] and [Cl2Pd(BINAP)]
(92.78).[10] Hence, the diphosphine-oxazoline ferrocenyl ligand
has higher reactivity than BINAP: a 1.5 mol% palladium
source and 3 mol% (S,Sp)-7a were needed to complete the
reaction within 36 h.[16]
T[K]
293
monoclinic
P2(1)
9.3941(11)
20.155(2)
10.4647(12)
113.055(2)
1823.1(4)
2
crystal system
space group
a[ä]
b[ä]
c[ä]
b[8]
V[ä3]
Z
1calcd [gcmÀ3
]
1.568
1.162
876
m[mmÀ1
F(000)
]
cryst size[mm]
Tmin/Tmax
q range[8]
parameters
R1[a], Rw
0.355 Â 0.270 Â 0.179
0.55423/1.00000
2.02 28.24
459
0.0433, 0.0810
0.910
The generally accepted mechanism of the asymmetric
intermolecular Heck reaction is depicted in Scheme 5.[10]
Hayashi and co-workers explained the high enantioselectivity
of product 4 through a kinetic resolution process and studied
the relationship between the enantiomeric induction of
product 4 and the ratio of 3/4 with this mechanism. Amatore
and co-workers have provided a deep insight into the
oxidative addition step of different palladium catalytic
systems, and the different reactivities of various diphosphine
ligands were compared.[16] However, there is no detailed
discussion on the regioselectivity of the reaction, especially
the electronic nature of the ligand and the palladium
precursors.
GOF
[a] [I > 2s(I)]
an isopropyl group on the oxazoline ring was similarly
prepared as ligand 7a. As can be seen in Figure 1, two
phosphorus atoms chelate with the palladium center in 9. This
was further confirmation of the results from the 31P NMR
study. Thus, the difference in regioselectivity may be not
Chem. Eur. J. 2003, 9, 3073 3081
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3075

X.-L. Hou, L.-X. Dai et al.
FULL PAPER
(Table 2, entry 9), which is con-
sistent with the results for
[Pd(OAc)2]. A larger amount
of (nBu)4NOAc did not change
the ratio further (Table 2, en-
try 10). When two equivalents
of LiOAc per equivalent of Pd
was added to the reaction in the
presence of [Pd(dba)3 ¥ dba] and
toluene, a similar result was
obtained as that with (nBu)4-
NOAc (Table 2, entry 11).
If this explanation is true, the
electrophilicity of the cationic
Pd atom of complex C might
also influence the path toward 3
or 4. Hence, ligands 7b i bear-
ing different electron-donating
or electron-withdrawing groups
on phenyl rings of the phos-
phine were prepared with the
same procedures as ligand
(S,Sp)-7a (Scheme 6) and ap-
plied to the intermolecular
asymmetric Heck reaction.
The results are summarized in
Table 4. All these ligands with
Scheme 5. Proposed mechanism for the catalytic arylation of 2,3-dihydrofuran with phenyl triflate.
Herein, we attempt to rationalize the different regioselec-
tivities of [Pd(OAc)2] and [Pd(dba)3 ¥ dba]. Notably, the
hydride olefin complex C is a point of demarcation directed
either to 3 or to the isomerized product 4. Since the hydride
olefin complex C has a 16-electron square-planar structure,
dissociation of the coordinated olefin should proceed by
means of an associative mechanism involving an 18-electron
transition state formed by nucleophilic attack of an incoming
ligand at the palladium center. When [Pd(OAc)2] was used as
a palladium source, acetate anions in the reaction system
should possess sufficient nucleophilicity toward the cationic
palladium center in C to cause the dissociation to give product
3. However, when [Pd(dba)3 ¥ dba] was used, a similar
nucleophilic attack could not occur, so that palladium would
re-insert into the olefin to give complex D. A subsequent b-
hydride elimination therefore proceeded more easily to give
4. The nucleophilicity of an acetate anion should be more
pronounced in a nonpolar media, such as toluene.[10, 24] Thus,
when toluene was used as a solvent instead of THF, a higher
regioselectivity could be observed (Table 2, entries 2 and 3).
When palladium bistrifluoroacetate was used, the ratio of 3
and 4 was lower than that of [Pd(OAc)2] for the weaker
nucleophilicity of the trifluoroacetate anion, despite the fact
that 3 was the major product (Table 2, entry 6 versus 3). To
investigate the effect of acetate anion as a nucleophile on the
regioselectivity, the effect of the addition other acetate salts
was also tested (Table 2, entries 8, 9, 10, and 11). When two
equivalents of (nBu)4NOAc per equivalent of Pd were added
to the reaction in the presence of [Pd(dba)3 ¥ dba] and THF,
the ratio of 3/4 increased to 37:63 (compare the ratio of 28:72
for entry 1, Table 2). Furthermore, when toluene was used
instead of THF, a higher ratio, namely 60:40, was obtained
[Pd(OAc)2] were able to catalyze the reaction of 1 and 2 in
good conversions within 36 h and isomer 3 was formed with
Scheme 6. Planar chiral diphosphine oxazoline ferrocenyl ligands modi-
fied with different electronic groups.
Table 4. Electronic effects of ligands on the regioselectity of Heck Reaction.[a]
Entry Ligand Palladium
Solvent
Conversion Ratio
ee[%][c]
of 3/4[b] 3(R) 4(S)
[%][b]
1
2
3
4
5
6
7
8
7b
7c
7c
7d
7e
7 f
7g
7h
7i
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(dba)3 ¥ dba] THF
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(OAc)2]
toluene
toluene
72
67
86
77
73
98
80
67
65
100
94:6
99:1
89:11
90:10
93:7
75
92
94
98
96
97
83
98
86
94
nd
13
nd
20
nd
nd
29
29
29
27
48
19
toluene
toluene
toluene
toluene
toluene
toluene
95:5
64:36
83:17
14:86
15:85
8:92
9
10
11
7i
7i
[Pd(dba)3 ¥ dba] THF
[Pd(dba)3 ¥ dba] (CH2Cl)2 68
[a] The reaction was carried out with 1.5 mol% palladium precursor and
3 mol% ligand in the presence of iPr2NEt as base at 608C for 36 h. [b] The
conversion and the ratio of 3/4 were determined by GC with n-undecane as the
internal standard based on phenyl triflate (2). [c] Determined by GC on a Chiral
g-PG column. nd ˆ not determined.
3076
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Wein-
heim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 3073 3081

Asymmetric Intermolecular Heck Reaction
3073 3081
a high ee value. The best regioselectivity of 99:1 for 3 and 4
was observed with ligand 7c, which has four strongly electron-
withdrawing trifluoromethyl groups on the phenyl rings
of the phosphine attached to the lower Cp ring (Table 4,
entry 2); while a reversed ratio of 14:86 for the two
isomers was found for ligand 7i, which contains two elec-
tron-donating methoxy groups on the phenyl rings of phos-
phine (Table 4, entry 9). The diphosphine mode of coordi-
nation was supported by 31P NMR spectroscopic studies in
these two cases: when ligand 7c and 7i coordinate with Pd0 in
C6D6, the chemical shifts of the two phosphorus atoms all
shifted to downfield (7c: d ˆ 23.63 and 22.52 from À16.54 and
À17.99 ppm; 7i: d ˆ 25.48 and 19.55 from À17.23 and
À20.48 ppm). When ligand 7c coordinated with Pd to form
the hydride olefin complex C, the electron-withdrawing
trifluoromethyl groups could, in principle, lower the electron
Scheme 7. The regioselectivities of Heck reaction catalyzed by ligands 10
and 11 with 1.5 mol% [Pd(OAc)2] in the presence of iPr2NEt and toluene at
608C.
The literature shows that different regioselectivities are
produced by P,P ligands and P,N ligands.[4, 5] Could this
difference be explained by a similar rationale? To understand
‡
density on Pd , thereby making it more vulnerable to attack
‡
by an acetate anion and leading eventually to product 3.
However, because of the electron-donating effect of the
methoxy group, ligand 7i was able to enhance the electron
density at the palladium center, leading to its re-insertion
into the olefin to give isomer 4. When [Pd2(dba)3 ¥ dba] was
used instead of [Pd(OAc)2] in THF, the regioselectivity of
ligand 7c was lower than that obtained with [Pd(OAc)2] in
toluene (Table 4, entry 2 versus 3); while less change was
found in the ratio of the two isomers when ligand 7i was used
(Table 4, entry 9 versus 10). When the more polar solvent 1,2-
dichloroethane was used instead of THF, a better ratio of 8:92
for the two isomers was observed for ligand 7i (Table 4,
entry 11).
These results suggested that the different electronic nature
of ligating atoms greatly affected the regioselectivity. How-
ever, when the modified phosphine group was attached to the
upper Cp ring, the ring bearing an oxazoline group, the
regioselectivity of the reaction changed a little compared to
that with ligand 7a. That is, the effect of the electronic nature
of such ligand is not as significant as the ligand with modified
groups on the lower Cp ring. It is possible that the oxazoline
ring also acts as a electron-withdrawing group, therefore, the
additional attached group at upper Cp ring has less effect than
that at the lower Cp ring. To check the electronic effect of the
ligand on the regioselectivity in more detail, the dppf
analogues ligands 10 and 11 were synthesized. These ligands
contain electron-withdrawing
it, the atomic (Mulliken) charge of [PdH{(R)-BINAP}] and
‡
[PdH{(S)-PHOX}] were calculated at the PM3 level with
Spartan5.0.[25] The atomic (Mulliken) charge on the palladium
‡
center of [PdH{(R)-BINAP}] is À0.041975, while that
‡
of [PdH{(S)-PHOX}] is ‡0.020161. The electron density
‡
on Pd of hydride olefin complex C with PHOX should
be much lower than that with BINAP. So, unisomerized
isomer 3 was obtained as main product with PHOX, while
with BINAP, isomerized product 4 was afforded predom-
inantly owing to the rich electron density on the palladium
center.
With these results in hand, other aryl and cyclohexenyl
triflates were used to examine the above effect. Reactions of
2,3-dihydrofuran and these triflates with ligands 7a, 7c, and 7i
gave good enantioselectivities for 2,5-dihydrofuran deriva-
tives and high conversion at 608C within 24 36 h (Table 5
and Table 6). In spite of the steric or electronic properties of
triflates, the regioselectivities of the Heck reaction with these
ligands also agree with our explanation. When 1-cyclohex-
enyltriflate was used with ligand 7i, the regioselectivity was
poor; however, isomer 14 was observed as the major product
(Table 5, entry 3); When 4-nitrophenyl triflate was employed
with ligand 7c, the ratio of 16:17 was lower, but isomer 16 was
still the main product (Table 6, entry 7). However, when
ligand 7i was employed, no 2,5-dihydrofuran derivatives were
observed (Table 6, entry 9).
trifluoromethyl groups and the
Table 5. Enantioselective and regioselective Heck reaction of 2,3-dihydrofuran with 1-cyclohexenyltriflate.
electron-donating
methoxy
group, respectively. When they
were employed as the ligand
with [Pd(OAc)2] and toluene in
the Heck reaction, different
regioselectivities were observed
(Scheme 7). Ligand 10 provides
product 3 predominantly, while
4 is the major product with
ligand 11. These results support
our hypothesis that the elec-
tronic nature of the ligating
atoms does affect the regiose-
lectivity.
Entry
Ligand
Palladium
Solvent
Conversion
[%][a]
13/14[a]
ee[%][b]
13(R)
14
1
2
3
7c
7a
7i
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(dba)3 ¥ dba]
toluene
toluene
(CH2Cl)2
99
98
87
97:3
92:8
35:65
80[c]
90[c]
40
nd
nd
nd
[a] The conversion and the ratio of 13/14 were determined by GC with n-undecane as the internal standard based
on cyclohexenyl triflate (12). [b] Determined by GC on a Chiraldex B-PH column. [c] The absolute configuration
of products was assigned by comparison of the sign of specific rotations with literature data.[5] nd ˆ not
determined.
Chem. Eur. J. 2003, 9, 3073 3081
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3077

X.-L. Hou, L.-X. Dai et al.
FULL PAPER
Table 6. Enantioselective and regioselective Heck reaction of 2,3-dihydrofuran with different aryl triflates.
further purification. The following
starting materials were prepared ac-
cording to literature procedures.
Synthesis of 1-diphenylphosphino-1'-
[(S)-tert-butyl-2,5-oxazolinyl]ferro-
cene (6a):[11] Compound 5 (205 mg,
0.5 mmol) was dissolved in freshly
distilled THF (4 mL) under argon
and cooled to À788C. At this temper-
ature, nBuLi (0.38 mL, 0.6 mmol, 1.6m
in n-hexane) was added, and then the
resulting deep red solution was stirred
for 20 min. Chlorodiphenylphosphine
(0.13 mL, 0.7 mmol) was then added,
and the resulting mixture was contin-
ually stirred and allow to warm to
room temperature over 30 min. The
reaction mixture was diluted with
ether (20 mL), washed with saturated
aqueous NaHCO3, and dried over
Na2SO4. The solvent was removed
under reduced pressure, and the re-
sulting residue was purified by chro-
matography (silica gel, ethyl acetate/
petroleum 1:5) to give 227 mg of 5
Entry
Ligand
Triflate
Palladium
Solvent
Conversion[%]
16/17
ee[%][f]
16(R)[f]
17(S)
1
2
3
4
5
6
7
8
9
7c
7a
7i
7c
7a
7i
7c
7a
7i
15a
15a
15a
15b
15b
15b
15c
15c
15c
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(dba)3 ¥ dba]
[Pd(OAc)2]
[Pd(OAc)2]
[Pd(dba)3 ¥ dba]
[Pd(OAc)2]
[Pd(OAc)2]
toluene
toluene
(CH2Cl)2
toluene
toluene
(CH2Cl)2
toluene
toluene
(CH2Cl)2
73[a]
99[a]
83[a]
91[b]
67[b]
86[b]
100[b]
68[b]
82[b]
90:10
83:17
17:83
97:3
66[c]
78[c]
94[c]
96[d]
66[d]
10[c]
9[c]
14[c]
nd
nd
9[d]
nd
94:6
0:100
60:40
49:51
0:100
> 99[e]
96[e]
nd
[Pd(dba)3 ¥ dba]
nd
[a] Conversion determined by GC with n-tridecane as the internal standard based on 2-naphthyl triflate (15a).
[b] Conversion determined by GC with n-undecane as the internal standard based on aryl triflate (15b or 15c).
[c] Determined by HPLC on a Chiralcel OD-H column. [d] Determined by GC on a Rt-bDex column.
[e] Determined by GC on a G-TA column. [f] The absolute configuration of the products was assigned by
comparison of the sign of specific rotations with literature data.[4, 5] nd ˆ not determined.
(88% yield) as an orange oil. [a]D20
ˆ
À132.6 (c ˆ 0.35, CHCl3); 1H NMR
(CDCl3): d ˆ 0.92 (s, 9H), 3.87 (dd,
J ˆ 10.1, 7.7 Hz, 1H), 4.09 4.24 (m,
6H), 4.40 (brs, 2H), 4.65 4.68 (m,
2H), 7.30 7.40 ppm (m, 10H).
Conclusion
Synthesis of 1-bis(3,5-bistrifluoromethylphenyl)phosphino-1'-[(S)-tert-bu-
tyl-2,5-oxazolinyl]ferrocene (6b): Prepared from 5 in 73% yield as a yellow
powder. M.p. 408C; [a]D20 ˆ À91.2 (c ˆ 0.75, CHCl3); 1H NMR (CDCl3):
d ˆ 0.93 (s, 9H), 3.90 (dd, J ˆ 9.78, 7.64 Hz, 1H), 3.89 4.18 (m, 3H), 4.21
(m, 3H), 4.53 (m, 2H), 4.68 (d, J ˆ 12.83 Hz, 2H), 7.77 (dd, J ˆ 11.0, 6.4 Hz,
4H), 7.90 ppm (s, 2H); 31P NMR (CDCl3): d ˆ À14.32 ppm (s, 1P); 19F
A study of the regioselectivity of the asymmetric Heck
reaction was carried out utilizing a series of diphosphine-
oxazoline ferrocenyl ligands and different palladium precur-
sors. All these ligands can catalyze the reaction in high
conversion and with 1.5 mol% catalyst loading. The tuning of
the regioselectivity was realized by the use of palladium
precursors and by varying the electronic nature of the ligands.
A plausible mechanistic rationale was also provided. Fur-
thermore, for 2,5-dihydrofuran derivatives, up to 99% ee was
obtained. This is the first time that dramatic electronic effects
have been observed and regioselectivity has been controlled
in an asymmetric Heck reaction. These results may improve
our understanding of the mechanisms of the asymmetric Heck
reaction and provide a guide toward the design of ligands.
‡
NMR (CDCl3): d ˆ 13.05 ppm (s, 12F); MS (EI): m/z (%): 767 ([M] , 100),
‡
‡
768 ([M‡1] , 43.22), 766 ([M À 1] , 15.97), 253 (50.40); IR (KBr): nÄ ˆ 2960,
1662, 1480, 1355, 1280, 1185, 1136, 970, 899, 844, 705, 682, 508 cmÀ1
;
elemental analysis calcd (%) for C33H26F12FeNOP: C 51.65, H 3.42, N 1.83;
found: C 51.74, H 3.68, N 1.66.
Synthesis of 1-bis(3,5-dimethylphenyl)phosphino-1'-[(S)-tert-butyl-2,5-ox-
azolinyl]ferrocene (6c): Prepared from 5 in 80% yield as a yellow powder.
1
M.p. 458C; [a]2D0 ˆ À143.4 (c ˆ 0.745, CHCl3); H NMR (CDCl3): d ˆ 0.92
(s, 9H), 2.27 (s, 12H), 3.87 (dd, J ˆ 9.77, 7.64 Hz, 1H), 4.12 (d, J ˆ 7.95 Hz,
3H), 4.19 (m, 3H), 4.37 (m, 2H), 4.65 (m, 2H), 6.97 ppm (t, J ˆ 7.33 Hz,
6H); 31P NMR (CDCl3): d ˆ À16.74 ppm (s, 1P); MS (EI): m/z (%): 552
‡
‡
([M‡1] , 100), 553 ([M‡2] , 40.06), 468 (43.17), 253(33.90); IR (KBr): nÄ ˆ
2951, 1658,1581, 1464, 1299, 1260, 1115, 1026, 969, 846, 693, 492 cmÀ1
;
elemental analysis calcd (%) for C33H38FeNOP: C 71.87, H 6.95, N 2.54;
found: C 71.47, H 6.99, N 2.43.
Synthesis of 1-bis(4-trifluoromethylphenyl)phosphino-1'-[(S)-tert-butyl-
2,5-oxazolinyl]ferrocene (6d): Prepared from 5 in 85% yield as a yellow
powder. M.p. 398C; [a]D20 ˆ À105.5 (c ˆ 1.39, CHCl3); 1H NMR (CDCl3):
d ˆ 0.92 (s, 9H), 3.87 (dd, J ˆ 10.38, 7.64 Hz, 1H), 4.08 (m, 3H), 4.19 (m,
3H), 4.47 (t, J ˆ 1.22 Hz, 2H), 4.70 (d, J ˆ 10.99 Hz, 2H), 7.45 (t, J ˆ
7.64 Hz, 4H), 7.58 ppm (d, J ˆ 7.33 Hz, 4H); 31P NMR (CDCl3): d ˆ
À16.30 ppm (s, 1P); 19F NMR (CDCl3): d ˆ 13.18 ppm (s, 6F); MS (EI):
Experimental Section
General: All reactions and manipulations were performed in an argon
atmosphere by means of standard Schlenk techniques. Anhydrous solvents
were transferred by oven-dried syringes. Flaskware was flame-dried under
a stream of argon. Melting points are uncorrected. NMR spectra were
recorded at room temperature in CDCl3 or C6D6 with 300 MHz (1H),
121 MHz (31P), and 288 MHz (19F) instruments. The chemical shifts are
given relative to TMS (as an internal reference) for 1H NMR, 85% H3PO4
(as an external reference) for 31P NMR, and CF3COOH (as an external
reference) for 19F NMR spectroscopy. IR spectra were recorded in KBr
with a Shimadze IR-440 infrared spectrophotometer. Mass spectra were
recorded on a HP5989A mass spectrometer. Elemental analyses were
carried out on a Foss-Heraus Vario instrument by the Analytic and Testing
Center of SIOC, Chinese Academy of Sciences.
‡
‡
‡
m/z (%): 631 ([M] , 100), 632 ([M‡1] , 39.32), 633 ([M‡2] , 12.43), 630
‡
([M À 1] , 50.40), 253 (67.15); IR (KBr): nÄ ˆ 2958, 1660, 1486, 1396, 1324,
1166, 1128, 1061, 1017, 970, 832, 698, 502 cmÀ1; elemental analysis calcd (%)
for C31H28F6FeNOP: C 58.97, H 4.47, N 2.22; found: C 59.28, H 4.77, N 2.17.
Synthesis of 1-bis(4-methyoxylphenyl)phosphino-1'-[(S)-tert-butyl-2,5-ox-
azolinyl]ferrocene (6e): Prepared from 5 in 40% yield as a yellow powder.
1
M.p. 428C; [a]2D0 ˆ À117.7 (c ˆ 0.525, CHCl3); H NMR (CDCl3): d ˆ 0.92
(s, 9H), 3.80 (s, 6H), 3.87 (dd, J ˆ 9.77, 7.33 Hz, 1H), 4.10 (m, 3H), 4.20 (m,
3H), 4.36 (s, 2H), 4.67 (d, J ˆ 11.61 Hz, 2H), 6.85 (d, J ˆ 7.49 Hz, 4H),
7.27 ppm (m, 4H); 31P NMR (CDCl3): d ˆ À20.40 ppm (s, 1P); MS (EI):
Materials: Tetrahydrofuran (THF), diethyl ether, benzene and toluene
were distilled from sodium/benzophenone prior to use; CH3CN and CH2Cl2
were distilled from CaH2. All purchased reagents were used without
‡
‡
m/z (%): 555([M] , 100), 556 ([M‡1] , 27.75), 556 (27.75), 245(14.01); IR
3078
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 3073 3081

Asymmetric Intermolecular Heck Reaction
3073 3081
(KBr): nÄ ˆ 2953, 1658, 1593, 1498, 1284, 1246, 1176, 1116, 1095, 1029, 968,
826, 796, 491 cmÀ1 cmÀ1; elemental analysis calcd (%) for C31H34FeNO3P: C
67.04, H 6.17, N 2.52; found: C 66.55, H 6.27, N 2.40
eoselective ortho-lithiation of 6c, the resulting mixture was treated with
chlorodiphenylphosphine as described above. After chromatography (ethyl
acetate/petroleum 1:10) of the crude product, 7e was obtained as a yellow
powder in 93% yield. M.p. 658C; [a]2D0 ˆ À126.5 (c ˆ 0.61, CHCl3);
1H NMR (CDCl3): d ˆ 0.74 (s, 9H), 2.22 (d, J ˆ 8.56 Hz, 12H), 3.45 (t,
J ˆ 1.22 Hz, 1H), 3.70 (dd, J ˆ 9.78, 7.95 Hz, 1H), 3.78 (dd, J ˆ 15.89,
9.48 Hz, 1H), 3.88(t, J ˆ 1.84 Hz, 1H), 4.16 (dd, J ˆ 9.78, 8.55 Hz, 1H), 4.22
(t, J ˆ 2.45 Hz, 1H), 4.28 (t, J ˆ 1.22 Hz, 1H), 4.48 (t, J ˆ 1.22 Hz, 2H),
4.90(t, J ˆ 1.22 Hz, 1H), 6.88 (m, 6H), 7.15 7.42 ppm (m, 10H); 31P NMR
(CDCl3): d ˆ À16.94 (s, 1P), À17.61 ppm (s, 1P); MS (EI): m/z (%): 736
Synthesis of 1-diphenylphosphino-1'-[(S)-tert-butyl-2,5-oxazolinyl]-2'-(Sp)-
diphenylphosphinoferrocene (7a): A solution of 6a (258 mg, 0.5 mmol)
and TMEDA (0.1 mL, 0.7 mmol) in diethyl ether (6 mL) under argon was
cooled to À788C. To this solution was added nBuLi (0.4 mL, 0.64 mmol)
and the mixture was stirred at À788C for 2 h. Chlorodiphenylphosphine
(0.13 mL, 0.7 mmol) was then added and the dry-ice bath was removed. The
resulting mixture was continually stirred for 20 min, and then quenched
with saturated NaHCO3, diluted with diethyl ether, washed with brine,
dried over MgSO4, filtered and evaporated under reduced pressure. The
resulting residue was purified by column chromatography (ethyl acetate/
petroleum 1:10) to afford 309 mg of compound 7a as a yellow powder
(yield 88%). M.p. 708C; [a]2D0 ˆ À130.5 (c ˆ 1.485, CHCl3); 1H NMR
(CDCl3): d ˆ 0.75 (s, 9H), 3.43 (s, 1H), 3.75 (dt, J ˆ 39.0, 8.1 Hz, 2H), 3.94
(t, J ˆ 1.5 Hz, 1H), 4.13 4.21 (m, 3H), 4.26 (t, J ˆ 1.6 Hz, 1H), 4.49 (t, J ˆ
1.7 Hz, 1H), 4.89 (s, 1H), 7.10 7.40 ppm (m, 20H); 31P NMR (CDCl3): d ˆ
‡
‡
‡
([M] , 100), 737 ([M‡1] , 47.31), 735 ([M À 1] , 89.25), 679 (33.74); IR
(KBr): nÄ ˆ 2951, 1662, 1583, 1478, 1434, 1141, 1027, 979, 846, 742, 694,
501 cmÀ1; elemental analysis calcd (%) for C45H47FeNOP2: C 73.47, H 6.44,
N 1.90; found: C 73.59, H 6.22, N 1.71.
Synthesis of 1-diphenylphosphino-1'-[(S)-tert-butyl-2,5-oxazolinyl]-2'-(Sp)-
bis(4-trifluoromethyl)phenylphosphinoferrocene (7 f): After directed dia-
stereoselective ortho-lithiation of 6a, the resulting mixture was treated with
chlorobis(4-trifluoromethyl)phenylphosphine as described above. After
chromatography (ethyl acetate/petroleum 1:10) of the crude product, 7 f
was obtained as a yellow powder in 94% yield. M.p. 1658C; [a]2D0 ˆ À210.0
(c ˆ 0.635, CHCl3); 1H NMR (CDCl3): d ˆ 0.80 (s, 9H), 3.29 (d, J ˆ 1.23 Hz,
1H), 3.68 (dd, J ˆ 9.78, 7.33 Hz, 1H), 3.85 (m, 1H), 3.92 (t, J ˆ 7.34 Hz, 3H),
4.16 (dd, J ˆ 18.33, 9.17 Hz, 1H), 4.30 (t, J ˆ 3.06 Hz, 1H), 4.36 (s, 1H), 4.47
(m, 2H), 4.95 (t, J ˆ 1.22 Hz, 1H), 7.10 7.30 (m, 12H), 7.40 7.57 ppm (m,
4H); 31P NMR (CDCl3): d ˆ À17.47 (s, 1P), À17.69 ppm (s, 1P); 19F NMR
‡
À17.24 (s, 1P), À17.47 ppm (s, 1P); MS (EI): m/z (%): 679 ([M] , 100), 680
‡
([M‡1] , 44.33), 622 (34.59); IR (KBr): nÄ ˆ 2953, 1662, 1479, 1434, 1164,
1141, 1027, 979, 831, 742, 702, 696, 501 cmÀ1; elemental analysis calcd (%)
for C41H39FeNOP2: C 72.47, H 5.78, N 2.06; found: C 72.39, H 5.77, N 2.24.
Synthesis of 1-diphenylphosphino-1'-[(S)-tert-butyl-2,5-oxazolinyl]-2'-
(Sp)-bis(3,5-bistrifluoromethyl)phenylphosphinoferrocene (7b): After di-
rected diastereoselective ortho-lithiation of 6a, the resulting mixture was
treated with chloro-bis-3,5-bistrifluoromethylphenylphosphine as descri-
bed above. After chromatography (ethyl acetate/petroleum 1:10) of the
crude product, 7b was obtained as a yellow powder in 68% yield. M.p.
1838C; [a]2D0 ˆ À262.5 (c ˆ 0.75, CHCl3); 1H NMR (CDCl3): d ˆ 0.86 (s,
9H), 3.14 (t, J ˆ 2.16 Hz, 1H), 3.60 (dd, J ˆ 9.77, 7.33 Hz, 1H), 3.74 (s, 1H),
4.07 (dd, J ˆ 15.36, 8.86 Hz, 1H), 4.13(dd, J ˆ 18.33, 8.86 Hz, 1H), 4.41 (m,
3H), 4.49 (s, 1H), 4.98(t, J ˆ 1.22 Hz, 1H), 7.14 7.28 (m, 10H), 7.58 (d, J ˆ
6.72 Hz, 2H), 7.77 (s, 2H), 7.79 (s, 1H), 7.88 ppm (s, 1H); 31P NMR
(CDCl3): d ˆ À16.54 (s, 1P), À17.99 ppm (s, 1P); 19F NMR (CDCl3): d ˆ
‡
(CDCl3): d ˆ 13.27 ppm (s, 6F); MS (EI): m/z (%): 815 ([M] , 100), 816
‡
([M‡1] , 40.77), 758 (28.83); IR (KBr): nÄ ˆ 2955, 1659, 1606, 1480, 1434,
1395, 1323, 1168, 1133, 1059, 1015, 977, 833, 745, 697, 599, 502 cmÀ1
;
elemental analysis calcd (%) for C43H37F6FeNOP2: C 63.33, H 4.57, N 1.72;
found: C 63.34, H 4.82, N 1.69.
Synthesis of 1-bis(4-trifluoromethyl)phenylphosphino-1'-[(S)-tert-butyl-
2,5-oxazolinyl]-2'-(Sp)-diphenylphosphinoferrocene (7g): After directed
diastereoselective ortho-lithiation of 6d, the resulting mixture was treated
with chlorodiphenylphosphine as described above. After chromatography
(ethyl acetate/petroleum 1:10) of the crude product, 7g was obtained as a
yellow powder in 80% yield. M.p. 808C; [a]2D0 ˆ À145.0 (c ˆ 0.60, CHCl3);
1H NMR (CDCl3): d ˆ 0.76 (s, 9H), 3.37 (d, J ˆ 1.22 Hz, 1H), 3.71 (dd, J ˆ
20.77, 9.48 Hz, 1H), 3.78 (dd, J ˆ 15.89, 9.51 Hz, 1H), 3.89 (t, J ˆ 1.83 Hz,
3H), 4.15 (dd, J ˆ 18.33, 9.77 Hz, 1H), 4.20 (m, 1H), 4.29 (t, J ˆ 1.22 Hz,
1H), 4.55 (m, 2H), 4.93 (t, J ˆ 1.22 Hz, 1H), 7.12 7.55 ppm (m, 16H); 31P
NMR (CDCl3): d ˆ À16.78 (s, 1P), À17.37 ppm (s, 1P); 19F NMR (CDCl3):
‡
13.12 ppm (d, J ˆ 52.85 Hz, 12F); MS (EI): m/z (%): 951 ([M] , 100), 952
‡
‡
([M‡1] , 47.55), 950 ([M À 1] , 22.30), 893 (23.28); IR (KBr): nÄ ˆ 2958,
1660, 1480, 1436, 1353, 1278, 1181, 1141, 983, 895, 844, 699, 682, 518,
501 cmÀ1; elemental analysis calcd (%) for C45H35F12FeNOP2: C 56.80, H
3.71, N 1.47; found: C 56.78, H 4.02, N 1.33.
Synthesis of 1-bis(3,5-bistrifluoromethyl)phenylphosphino-1'-[(S)-tert-bu-
tyl-2,5-oxazolinyl]-2'-(Sp)-diphenylphosphinoferrocene (7c): After direct-
ed diastereoselective ortho-lithiation of 6b, the resulting mixture was
treated with chlorodiphenylphosphine as described above. After chroma-
tography (ethyl acetate/petroleum 1:10) of the crude product, 7c was
obtained as a yellow powder in 71% yield. M.p. 678C; [a]2D0 ˆ À146.5 (c ˆ
0.75, CHCl3); 1H NMR (CDCl3): d ˆ 0.77 (s, 9H), 3.46 (s, 1H), 3.75 (m,
3H), 4.22 (m, 3H), 4.53 (m, 1H), 4.63 (s, 1H), 4.91(s, 1H), 7.16 7.44 (m,
‡
d ˆ 13.22 ppm (s, 6F); MS (EI) m/z (%): 815 ([M] , 81.17), 759 (28.74), 739
(27.13), 574 (33.47), 265 (30.57); IR (KBr): nÄ ˆ 2955, 1664, 1607, 1479, 1435,
1396, 1324, 1166, 1128, 1060, 1016, 979, 831, 743, 696, 499 cmÀ1; elemental
analysis calcd (%) for C43H37F6FeNOP2: C 63.33, H 4.57, N 1.72; found: C
63.62, H 5.07, N 1.89.
Synthesis of 1-diphenylphosphino-1'-[(S)-tert-butyl-2,5-oxazolinyl]-2'-(Sp)-
bis(4-methoxyl)phenylphosphinoferrocene (7h): After directed diastereo-
selective ortho-lithiation of 6a, the resulting mixture was treated with
chlorobis(4-methoxyl)phenylphosphine as described above. After chroma-
tography (ethyl acetate/petroleum 1:10) of the crude product, 7h was
obtained as a yellow powder in 83% yield. M.p. 778C; [a]2D0 ˆ À129.2 (c ˆ
0.905, CHCl3); 1H NMR (CDCl3): d ˆ 0.74 (s, 9H), 3.42 (m, 1H), 3.74 (s,
3H), 3.77 (m, 3H), 3.83 (s, 3H), 3.91 (m, 1H), 4.13 (m, 1H), 4.21 (t, J ˆ
2.44 Hz, 1H), 4.48 (m, 2H), 4.88 (t, J ˆ 1.22 Hz, 1H), 6.80 (dd, J ˆ 25.28,
10H), 7.66 (dd, J ˆ 18.33, 6.11 Hz, 4H), 7.69 ppm (d, J ˆ 11.61 Hz, 2H); 31
P
NMR (CDCl3): d ˆ À15.17 (s, 1P), À17.62 ppm (s, 1P); 19F NMR (CDCl3):
‡
d ˆ 13.05 ppm (d, J ˆ 0.75 Hz, 12F); MS (EI): m/z (%): 951 ([M] , 100),
‡
952 ([M‡1] , 48.48), 874 (26.96); IR (KBr): nÄ ˆ 2958, 1666, 1480, 1435,
1354, 1279, 1186, 1138, 979, 898, 844, 742, 702, 682, 501 cmÀ1; elemental
analysis calcd (%) for C45H35F12FeNOP2: C 56.80, H 3.71, N 1.47; found: C
56.36, H 3.96, N 1.37.
7.95 Hz, 4H), 7.09 (dd, J ˆ 8.55, 7.33 Hz, 2H), 7.22 7.41 ppm (m, 12H); 31
P
Synthesis of 1-diphenylphosphino-1'-[(S)-tert-butyl-2,5-oxazolinyl]-2'-(Sp)-
bis(3,5-dimethyl)phenylphosphinoferrocene (7d): After directed diaster-
eoselective ortho-lithiation of 6a, the resulting mixture was treated with
chloro-bis-[3,5-dimethyl]-phenylphosphine as described above. After chro-
matography (ethyl acetate/petroleum 1:10) of the crude product, 7d was
obtained as a yellow powder in 68% yield. M.p. 758C; [a]2D0 ˆ À51.9 (c ˆ
0.875, CHCl3); 1H NMR (CDCl3): d ˆ 0.74 (s, 9H), 2.24 (d, J ˆ 20.17 Hz,
12H), 3.45 (s, 1H), 3.73 (m, 2H), 3.93(s, 1H), 4.18 (m, 2H), 4.26 (s, 1H),
NMR (CDCl3): d ˆ À17.23 (s, 1P), À20.48 ppm (s, 1P); MS (EI): m/z (%):
‡
‡
740 ([M] , 100), 741 ([M‡1] , 37.00), 683 (55.65); IR (KBr): nÄ ˆ 2952, 1737,
1662, 1594, 1498, 1434, 1363, 1283, 1246, 1176, 1094, 1031, 979, 825, 743, 697,
494 cmÀ1; elemental analysis calcd (%) for C43H43FeNOP2: C 69.83, H 5.86,
N 1.89; found: C 69.40, H 5.99, N 1.85.
Synthesis of 1-bis(4-methoxyl)phenylphosphino-1'-[(S)-tert-butyl-2,5-oxa-
zolinyl]-2'-(Sp)-diphenylphosphinoferrocene (7i): After directed diaster-
eoselective ortho-lithiation 6e, the resulting mixture was treated with
chlorodiphenylphosphine as described above. After chromatography (ethyl
acetate/petroleum 1:10) of the crude product, 7i was obtained as a yellow
powder in 48% yield. M.p. 708C; [a]2D0 ˆ À159.7 (c ˆ 0.595, CHCl3);
1H NMR (CDCl3): d ˆ 0.74 (s, 9H), 3.41 (s, 1H), 3.79 (s, 6H), 3.60 3.90 (m,
3H), 4.16 (t, J ˆ 9.17 Hz, 1H), 4.25 (s, 2H), 4.48 (d, J ˆ 9.17 Hz, 2H), 4.90(s,
1H), 6.79 (dd, J ˆ 17.11, 8.56 Hz, 4H), 7.14 7.41 ppm (m, 14H); 31P NMR
(CDCl3): d ˆ À16.96 (s, 1P), À21.06 ppm (s, 1P); MS (EI): m/z (%): 739
4.50 (s, 2H), 4.92(s, 1H), 6.73 (m, 3H), 7.01 (m, 3H), 7.25 ppm (m, 10H); 31
P
NMR (CDCl3): d ˆ À16.42 (s, 1P), À17.30 ppm (s, 1P); MS (EI): m/z (%):
‡
‡
736 ([M] , 36.75), 735 ([M À 1] , 100), 678 (75.44); IR (KBr): nÄ ˆ 2951,
1663, 1583, 1478, 1434, 1126, 1027, 979, 846, 742, 695, 502 cmÀ1; elemental
analysis calcd (%) for C45H47FeNOP2: C 73.47, H 6.44, N 1.90; found: C
73.74, H 6.90, N 1.83.
Synthesis of 1-bis(3,5-dimethyl)phenylphosphino-1'-[(S)-tert-butyl-2,5-ox-
azolinyl]-2'-(Sp)-diphenylphosphinoferrocene (7e): After directed diaster-
Chem. Eur. J. 2003, 9, 3073 3081
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3079

X.-L. Hou, L.-X. Dai et al.
FULL PAPER
‡
‡
‡
([M] , 100), 738 ([M À 1] , 90.77), 740 ([M‡1] , 36.95), 682 (61.77); IR
(KBr): nÄ ˆ 2951, 1661, 1593, 1497, 1434, 1352, 1284, 1246, 1176, 1095, 1029,
978, 825, 743, 696, 530, 500 cmÀ1; elemental analysis calcd (%) for
C43H43FeNOP2: C 69.83, H 5.86, N 1.89; found: C 69.76, H 5.94, N 1.90.
allowed to sitr for 30 min, cyclohexenyl or aryl triflate (1 mmol) and n-
undecane or n-tridecane (20 mg) (an internal GC standard) were added,
followed by the addition of 2,3-dihydrofuran (1, 0.38 mL, 4 mmol) and base
(2 mmol). The mixture was stirred at 608C under argon until the reaction
was complete according to GC analysis. The reaction mixture was diluted
with additional petroleum ether and the resulting red suspension was
filtered through Celite and washed with Et2O. The combined organic
solutions were concentrated to give a red oil, which was purified by flash
chromatography on silica gel (petroleum ether/ethyl acetate 15:1) to afford
2,5-dihydrofuran and 2,3-dihydrofuran derivatives as colorless oils.
Synthesis of 1-diphenylphosphino-1'-[(S)-isopropyl-2,5-oxazolinyl]-2'-(Sp)-
diphenylphosphinoferrocene (8): After directed diastereoselective ortho-
lithiation of 1-[(S)-isopropyl-2,5-oxazolinyl]-1'-bromoferrocene, which was
synthesized according to a literature procedure,[11a] the resulting mixture
was treated with chlorodiphenylphosphine as described above. After
chromatography (ethyl acetate/petroleum 1:10) of the crude product, 8 was
obtained as a yellow powder in 84% yield. M.p. 708C; [a]2D0 ˆ À54 (c ˆ 1.5,
CHCl3); 1H NMR (CDCl3): d ˆ 0.64 (d, J ˆ 6.8 Hz, 3H), 0.80 (d, J ˆ 6.8 Hz,
3H), 1.63 (m, 1H), 3.44 (m, 1H), 3.70 (t, J ˆ 8.2 Hz, 1H), 3.80 3.91 (m,
1H), 3.96 (m, 1H), 4.18 4.30 (m, 3H), 4.48 4.51 (m, 2H), 4.92(t, J ˆ
1.2 Hz, 1H), 7.18 7.48 ppm (m, 20H); 31P NMR (CDCl3): d ˆ À16.49 (s,
All conversions of different triflates were determined by GC and ee values
were determined by chiral GC or HPLC. Absolute configuration of
products was assigned through comparison of the sign of specific rotations
with literature data.[4, 5]
Phenyl triflate (2): GC (PE-XL, EC-5 (Altech); 30m  0.32 mm, 90
1108C, N2: 16 psi): tR ˆ 2.5 min (2), tR ˆ 3.8 min (n-undecane), tR ˆ 6.6 min
(4), tR ˆ 8.0 min (3); GC (PE-XL, Chiraldex G-PN (Astec); 20 m Â
0.25 mm, 90 1158C, 1 8CminÀ1, N2: 10 psi): 3: 11.9 min (S), 12.2 (R); 4:
28.4 min (S), 28.9 (R).
‡
1P), À17.28 ppm (s, 1P); MS (EI): m/z (%): 666 ([M] , 100), 480 (70.40),
393 (38.65), 183 (39.60); IR (KBr): nÄ ˆ 2955, 1658, 1478, 1433, 1026, 980,
741, 695, 498 cmÀ1; elemental analysis calcd (%) for C40H37FeNOP2: C
72.19, H 5.60, N 2.11; found: C 72.10, H 5.60, N 1.94.
X-ray diffraction study of complex 9: A solution of ligand 8 (66.6 mg,
0.1 mmol) in benzene (2.0 mL) was added with stirring to a mixture of
[PdCl2(CH3CN)2] (26 mg, 0.1 mmol) in benzene (1.0 mL), and the mixture
was stirred overnight. The orange precipitate that formed was collected by
filtration, washed with benzene, and dried under vacuum. The crude
product was dissolved in CH2Cl2, layered with hexane, and allowed to stand
at room temperature to give red crystals (76 mg, 90% yield): anal. calcd for
C40H37Cl2FeNOP2Pd ¥ H2O: C 55.81, H 4.75, N 1.63; found: C 55.80, H 4.44,
N 1.47.
Cyclohexenyl triflate (12): GC (PE-XL, EC-5 (Altech); 30m  0.32 mm,
120 1408C, N2: 15.7 psi): tR ˆ 1.0 min (12), tR ˆ 2.0 min (n-undecane), tR ˆ
3.5 min (14), tR ˆ 3.6 min (13); GC (PE-XL, Chiraldex B-PH (Astec);
20 m  0.25 mm, 1008C, 20 min, N2: 6 psi): 13: 11.9 min (S), 12.1 (R).
2-Naphthyl triflate (15a): GC (PE-XL, EC-5 (Altech); 30m  0.32 mm,
120 2508C, N2: 16 psi): tR ˆ 4.0 min (15a), tR ˆ 9.4 min (n-tridecane), tR ˆ
23.5 min (17a), tR ˆ 26.5 min, (16a); HPLC (Chiralcel OD-H; 0.46 m Â
25 cm, 0.5 mLminÀ1, heptane/isopropanol 90:10, detection at 220 nm):
16a: 13.7 min (S), 14.31 (R); 17a: 14.6 min (S), 16.0 (R).
CCDC-204150 (complex 9) contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crys-
tallographic Data Centre, 12 Union Road, Cambridge CB21EZ, UK; fax:
(‡44)1223-336033; or deposit@ccdc.cam.uk).
4-Methoxyphenyl triflate (15b): GC (PE-XL, EC-5 (Altech); 30m Â
0.32 mm, 120 2508C, N2: 16 psi): tR ˆ 3.4 min (15b), tR ˆ 2.0 min (n-
undecane), tR ˆ 8.3 min (17b), tR ˆ 9.7 min (16b); GC (PE-XL, Rt-b-Dex
(Restek); 30 m  0.25 mm, 100 1808C, 1 8CminÀ1
, N2: 10 psi): 16b:
71.3 min (S), 71.8 min (R); 17b: 65.4 min (S), 65.9 min (R).
Synthesis of 1-bis(3,5-bistrifluoromethyl)phenylphosphino-1'-diphenyl-
phosphinoferrocene (10): 1-Bromo-1'-diphenylphosphinoferrocene[26]
(225 mg, 0.5 mmol) was dissolved in fresh distilled THF (4 mL) under
argon and cooled to À788C. At this temperature, nBuLi (0.38 mL,
0.6 mmol, 1.6m in n-hexane) was added. The resulting deep red solution
was stirred for 20 min. Chlorobis(3,5-dimethyl)phenylphosphine (300 mg,
0.6 mmol) was then added, and resulting mixture was continually stirred
and allowed to warm to room temperature over 30 min. The reaction
mixture was diluted with ether (20 mL), washed with saturated aqueous
NaHCO3, and dried over Na2SO4. The solvent was removed under reduced
pressure, and the resulting residue was purified by chromatography (silica
gel, petroleum) to give 10 (330 mg, 80% yield) as a yellow powder. 1H NMR
(CDCl3): d ˆ 3.90 (dd, J ˆ 3.8, 1.8 Hz, 2H), 4.09 (dd, J ˆ 3.8, 1.8 Hz, 2H),
4.28 (t, J ˆ 1.5 Hz, 2H), 4.43 (t, J ˆ 1.5 Hz, 2H), 7.22 7.40 (m, 10H), 7.66 (d,
J ˆ 6.5 Hz, 4H), 7.86 ppm (s, 2H); 31P NMR (CDCl3): d ˆ À13.66 (s, 1P),
À17.12 ppm (s, 1P); 19F NMR (CDCl3): d ˆ 13.07 ppm (s, 12F); MS (EI):
4-Nitrophenyl triflate (15c): GC (PE-XL, EC-5 (Altech); 30m  0.32 mm,
120 2508C, N2: 16 psi): tR ˆ 2.0 min (n-undecane), tR ˆ 5.1 min (15c), tR ˆ
16.6 min (17c), tR ˆ 19.4 min (16c); GC (PE-XL, Chiraldex G-TA (Astec);
20 m  0.25 mm, 120 1708C, 1 8CminÀ1, N2: 12 psi): 16c: 45.0 min (R),
46.1 min (S).
Acknowledgement
Financial support from the National Natural Science Foundation of China,
Chinese Academy of Sciences, the Major Basic Research Development
Program (Grant No. G2000077506), and the National Outstanding Youth
Fund. We thank Professor H. N. C. Wong and Professor Z.-G. Zhang for
inspiring discussions.
‡
‡
‡
m/z (%): 826 ([M] , 100), 828 ([M‡2] , 37.78), 827 ([M‡1] , 63.15), 825
‡
([M À 1] , 81.16), 613 (24.21); IR (KBr): nÄ ˆ 3074, 2932, 1355, 1279, 1135,
1029, 898, 829, 743, 682, 495 cmÀ1; elemental analysis calcd (%) for
C38H24F12FeP2: C 55.23, H 2.93; found: C 55.53, H 3.00.
[1] a) Y. Sato, M. Sodeoka, M. Shibasaki, J. Org. Chem. 1989, 54, 4738;
b) N. E. Carpenter, D. J. Kucera, L. E. Overman, J. Org. Chem. 1989,
54, 5846.
Synthesis of 1-bis(4-methoxy)phenylphosphino-1'-diphenylphosphinofer-
rocene (11): After directed lithiation of 1-bromo-1'-diphenylphosphino-
ferrocene at À788C, the resulting mixture was treated with chlorobis(p-
methoxy)phenylphosphine as in the procedure for ligand 10. After
chromatography (petroleum) of the crude product, 11 was obtained as an
orange oil in 53% yield. 1H NMR (CDCl3): d ˆ 3.79 (s, 6H), 3.97 (ddd, J ˆ
9.4, 3.8, 1.9 Hz, 2H), 4.25 (dt, J ˆ 14.1, 1.8 Hz, 2H), 6.82 (d, J ˆ 8.9 Hz, 2H),
7.12 (dd, J ˆ 9.8, 7.1 Hz, 2H), 7.21 7.29 ppm (m, 18H); 31P NMR (CDCl3):
ƒ
[2] Reviews: a) C. Bolm, J. P. Hildebrand, K. Muniz, N. Hermanns,
Angew. Chem. 2001, 113, 3382; Angew. Chem. Int. Ed. 2001, 40, 3284;
b) M. Shibasaki, E. M. Vogl in Comprehensive Asymmetric Catalysis
(Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, Heidelberg,
1999, 458; c) O. Loiseleur, M. Hayashi, M. Keenan, N. Schmees, A.
Pfaltz, J. Organomet. Chem. 1999. 576, 16; d) M. Beller, T. H.
Riermeier, G. Stark in Transition Metals for Organic Synthesis (Eds:
M. Beller, C. Bolm), Wiley-VCH, Weinheim, 1998, Vol. 1, 208.
‡
d ˆ À16.26 (s, 1P), À20.05 ppm (s, 1P); MS (EI): m/z (%): 614 ([M] , 100),
‡
‡
616 ([M‡2] , 26.78), 615 ([M‡1] , 93.64), 538 (20.59), 537 (21.52); IR
(KBr): nÄ ˆ 2833, 1592, 1496, 1283, 1245, 1176, 1094, 1028, 825, 742, 696,
490 cmÀ1; elemental analysis calcd (%) for C36H32FeO2P2: C 70.37, H 5.25;
found: C 70.18, H 5.08.
[3] a) F. Diederich, P. J. Stang, Metal Metal Cross-Coupling Reaction,
Wiley-VCH, 1999; b) M. Shibasaki, C. D. J. Boden, A. Kojima,
Tetrahedron, 1997, 53, 7371; c) S. E. Gibson, R. J. Middleton, Contemp.
¬
Org. Synth. 1996, 3, 447; d) E. Negishi, C. Coperet, S. Ma, S.-Y. Liou, F.
General procedure of the asymmetric intermolecular Heck reaction: The
palladium species (15 mmol), diphosphineoxazoline ferrocenyl ligand
(30 mmol), and solvent (5 mL) were placed under argon in an Schlenk
tube with a magnetic stirring bar. After the reaction mixture had been
Liu, Chem. Rev. 1996, 96, 365.
[4] a) F. Ozawa, Y. Kobatake, T. Hayashi, Tetrahedron Lett. 1993, 34,
2505; b) F. Ozawa, T. Hayashi, J. Organomet. Chem. 1992, 428, 267;
c) F. Ozawa, A. Kubo, T. Hayashi, J. Am. Chem. Soc. 1991, 113, 1417.
3080
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 3073 3081

Asymmetric Intermolecular Heck Reaction
3073 3081
[5] a) O. Loiseleur, M. Hayashi, N. Schmees, A. Pfaltz, Synthesis 1997,
1338; b) O. Loiseleur, P. Meier, A. Pfaltz, Angew. Chem. 1996, 108,
218; Angew. Chem. Int. Ed. Engl. 1996, 35, 200.
Int. Ed. Engl. 1997, 36, 984; b) M. Ludwig, S. Strˆmberg, M. Svensson,
B. äkermark, Organometallics 1999, 18, 970.
[16] C. Amatore, G. Broeker, A. Jutand, F. Khalil, J. Am. Chem. Soc. 1997,
119, 5176.
¡
[6] a) L. F. Tietze, K. Thede, R. Schimpf, F. Sannicolo, Chem. Commun.
¡
2000, 583; b) T. Benincori, O. Piccolo, S. Rizzo, F. Sannicolo, J. Org.
[17] a) S.-L. You, X.-Z. Zhu, Y.-M. Luo, X.-L. Hou, L.-X. Dai, J. Am.
Chem. Soc. 2001, 123, 7471; b) S.-L. You, X.-L. Hou, L.-X. Dai, X.-Z.
Zhu, Org. Lett. 2001, 3, 149; c) S.-L. You, X.-Z. Zhu, X.-L. Hou, L.-X.
Dai, Acta Chimica Sinica 2001, 10, 1667; d) S.-L.You, X.-L. Hou, L.-X.
Dai, B.-X. Cao, J. Sun, Chem. Commun. 2000, 1933; e) S.-L.You, X.-L.
Hou, L.-X. Dai, Tetrahedron: Asymmetry 2000, 11, 1495.
[18] a) X.-L. Hou, L.-X. Dai, W.-P. Deng, S.-L.You, Y.-G. Zhou, Pure Appl.
Chem. 1999, 71, 1401; b) W.-P. Deng, X.-L. Hou, L.-X. Dai, Tetrahe-
dron: Asymmetry 1999, 10, 4689; c) S.-L. You, Y.-G. Zhou, X.-L. Hou,
L.-X. Dai, Chem. Commun.1998, 24, 2765; d) X.-D. Du, L.-X. Dai, X.-
L. Hou, L.-J. Xia, M.-H. Tang, Chin. J. Chem. 1998, 16, 90.
[19] a) A. Tongi, C. Breutel, A. Schnyder, F. Spindler, H. Landert,
A. Tijani, J. Am. Chem. Soc. 1994, 116, 4062; b) A. Tongi, Angew.
Chem. Int. Ed. Engl. 1996, 35, 1475; Angew. Chem. 1996, 108, 1581.
[20] A. Chesney, M. R. Bryce, R. W. J. Chubb, A. S. Batsanov, J. A. K.
Howard, Synthesis 1998, 413.
[21] C. J. Richards, A. W. Mulvaney, Tetrahedron: Asymmetry 1996, 7,
1419.
[22] F. Ozawa, A. Kubo, T. Hayashi, Chem. Lett. 1992, 2177.
[23] T. Hayashi, M. Konishi, Y. Kobori, M. Kumada, T. Higuchi, K.
Hirotsu, J. Am. Chem. Soc. 1984, 106, 158.
[24] For the nucleophilicity of anions in different solvents, see: M. B.
Smith, J. March, March×s Advanced Organic Chemistry: Reaction,
Mechanism and Structure, Wiley, 2001, 438 445.
[25] The structural parameters of the [Pd{(R)-BINAP}] moiety are taken
from the X-ray structure of [PdCl2{(R)-BINAP}];[10] and the structural
parameters of the [Pd{(R)-PHOX}] moiety are taken from the X-ray
structure of [Pd(allyl){(R)-PHOX}]: J. Sprinz, M. Kiefer, G. Helm-
chen, Tetrahedron Lett. 1994, 35, 1523.
¡
Chem. 2000, 65, 8340; c) L. F. Tietze, K. Thede, F. Sannicolo, Chem.
Commun. 1999, 1811.
[7] Contrary to usual expectations, diphosphine TMBTP gave an
isomerized product with 2,3-dihydrofuran; when this ligand was
employed in the reaction with dihydropyrrole mainly an unisomerized
product was obtained: L. F. Tietze, K. Thede, Synlett 2000, 1470.
[8] a) K. Selvakumar, M. Valentini, P. S. Pregosin, A. Albinati, F.
Eisentr‰ger, Organometallics 2000, 19, 1299; b) A. Albinati, P. S.
Pregosin, M. Tschoerner, Organometallics 1999, 18, 670; c) G. Trabe-
singer, A. Albinati, N. Feiken, R. W. Kunz, P. S. Pregosin, M.
Tschoerner, J. Am. Chem. Soc. 1997, 119, 6315.
[9] N. G. Andersen, R. McDonald, B. A. Keay, Tetrahedron: Asymmetry
2001, 12, 263.
[10] a) F. Ozawa, A. Kubo, Y. Matsumoto, T. Hayashi, Organometallics
1993, 12, 4188; b) T. Hayashi, A. Kubo, F. Ozawa, Pure Appl. Chem.
1992, 64, 421.
[11] a) W.-P. Deng, S.-L. You, X.-L. Hou, L.-X. Dai, Y.-H. Yu, W. Xia, J.
Sun, J. Am. Chem. Soc. 2001, 123, 6508; b) W.-P. Deng, X.-L. Hou, L.-
X. Dai, X.-W. Dong, Chem. Commun. 2000, 1483; b) W.-P. Deng, X.-
L. Hou, L.-X. Dai, Y.-H Yu, W. Xia, Chem. Commun. 2000, 4, 285.
[12] a) S. R. Gilberston, D. Xie, Z. Fu, J. Org. Chem. 2001, 66, 7240; b) S. R.
Gilberston, Z. Fu, Org. Lett. 2001, 3, 161.
[13] a) A. J. Hennessy, D. J. Connolly, Y. M. Malone, P. J. Guiry, Tetrahe-
dron Lett. 2000, 41, 7757; b) A. J. Hennessy, Y. M. Malone, P. J. Guiry,
Tetrahedron Lett. 1999, 40, 9163.
[14] a) A. V. Malkov, M. Bella, I. G. Stara, P. Kocovsky, Tetrahedron Lett.
2001, 42, 3045; b) Y. Hashimoto, Y. Horie, M. Hayashi, K. Saigo,
Tetrahedron: Asymmetry, 2000, 11, 2205; c) M. Ogasawara, K.Yoshida,
T. Hayashi, Heterocycles 2000, 52, 195; d) J. Priego, J. C. Carretero,
Synlett 1999, 1603.
¬
œ
¬
[26] T.-Y. Dong, L.-L. Lai, J. Organomet. Chem. 1996, 509, 131.
Received: September 25, 2002
Revised: February 24, 2003 [F4450]
[15] For recent mechanistic advances, see: a) K. K. S. Hii, T. D. W.
Claridge, J. M. Brown, Angew. Chem. 1997, 109, 1033; Angew. Chem.
Chem. Eur. J. 2003, 9, 3073 3081
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3081