Palladium-catalyzed asymmetric tandem allylic substitution using chiral 2-(phosphinophenyl)pyridine ligand

Article abstract of DOI:10.1016/j.tetlet.2004.08.014

Palladium-catalyzed asymmetric tandem allylic substitution of (Z)-1,4-diacyloxy- and (Z)-1,4-bis(alkoxycarbonyloxy)-2-butene (2a-c) using 2-(phosphinophenyl)pyridine 1 as chiral ligand provided optically active six-membered 2-vinyl-1,4-diheterocyclic comp

Full text of DOI:10.1016/j.tetlet.2004.08.014

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 7277–7281
Palladium-catalyzed asymmetric tandem allylic substitution using
chiral 2-(phosphinophenyl)pyridine ligand
a
a
a
Katsuji Ito,a,c, Yoshikatsu Imahayashi, Tomomi Kuroda, Shuuichiro Eno,
*
Bunnai Saitob,c and Tsutomu Katsukib,c,
*
aDepartment of Chemistry, Fukuoka University of Education, Akama, Munakata, Fukuoka 811-4192, Japan
bDepartment of Chemistry, Faculty of Science, Graduate School, Kyushu University 33, 6-10-1 Hakozaki,
Highasi-ku, Fukuoka 812-8581, Japan
cCREST, Japan Science and Technology Agency (JST), Japan
Received 20 July 2004; revised 30 July 2004; accepted 4 August 2004
Available online 21 August 2004
Abstract—Palladium-catalyzed asymmetric tandem allylic substitution of (Z)-1,4-diacyloxy- and (Z)-1,4-bis(alkoxycarbonyloxy)-2-
butene (2a–c) using 2-(phosphinophenyl)pyridine 1 as chiral ligand provided optically active six-membered 2-vinyl-1,4-dihetero-
cyclic compounds with good to high enantioselectivity. For example, the reactions with catechol, 2-(benzylamino)phenol, or
1,2-bis(benzylamino)ethane as a nucleophile gave 2-vinyl-1,4-benzodioxane (71% ee), 4-benzyl-2-vinyl-1,4-benzoxazine (86% ee),
and 1,4-dibenzyl-2-vinylpiperazine (86% ee), respectively.
Ó 2004 Elsevier Ltd. All rights reserved.
Palladium-catalyzed asymmetric allylic substitution is a
major topic in catalytic asymmetric synthesis and vari-
ous chiral ligands for this reaction have been developed
to date.1 In 1993, 2-(phosphinophenyl)oxazolines (P,N
ligands) were reported to be excellent chiral ligands for
asymmetric allylic alkylation.2 The ensuing introduction
of various types of P,N ligands has expanded the scope
of asymmetric allylic alkylation.3 We have demonstrated
that 2-(phosphinophenyl)pyridine 1 bearing an isopro-
pyl group at C7 is an efficient chiral auxiliary for palla-
dium-catalyzed allylic alkylation of both acyclic and
cyclic alkenyl substrates (Scheme 1).4a,b Furthermore,
we have found that a palladium–1 complex serves as a
good catalyst for asymmetric intramolecular allylic
amination.4c
butene or 1,4-bis(alkoxycarbonyloxy)-2-butene using a
1,2-heterofunctionalized compound as nucleophile is a
usefulmethod for the synthesis of these types of hetero-
cyclic compounds.5 Thus, much effort has been directed
toward asymmetrization of the tandem allylic substitu-
tions. Although many chiral ligands have been applied
to these tandem reactions (Scheme 2),6 Nakano et al.
have recently demonstrated that the xylofuranose-based
chiralP,N ligand induces good to high asymmetry in the
synthesis of 2-vinylmorpholine and 2-vinylpiperazine
derivatives, indicating that the chiralP,N ilgand is a
promising chiral ligand for catalytic asymmetric tandem
allylic substitution.6h
In PTAS, the first substitution reaction gives a x-hetero
substituted allyl ester derivative that undergoes intramo-
lecular allylic substitution. Since compound 1 is a potent
chiral ligand for palladium-mediated intramolecular
allylic amination (Scheme 1),4c we expected that asym-
metric PTAS would be effected by using 1 as the chiral
ligand.
On the other hand, most heterocyclic compounds pos-
sessing heteroatoms at C1 and C4 show unique thera-
peutical and biological activities. Palladium-catalyzed
tandem allylic substitution (PTAS) of 1,4-diacyloxy-2-
Keywords: Palladium; Asymmetric catalysis; P,N ligand; Tandem
allylic substitution.
We first examined the reaction of (Z)-1,4-diacetoxy-2-
butene (2a) with catecholin the presence of the Pd– 1
complex as catalyst and K2CO3 as a base (Table 1).
The reaction was completed after 7days and was moder-
ately enantioselective (entry 1). Replacement of the
*
Corresponding authors. Tel.: +81 940 35 1367; fax: +81 940 35 1711
(K.I.); tel.: +81 926 42 2586; fax: +81 926 42 2607 (T.K.); e-mail
addresses: itokat@fukuoka-edu.ac.jp; katsuscc@mbox.nc.kyushu-u.
ac.jp
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.08.014

7278
K. Ito et al. / Tetrahedron Letters 45 (2004) 7277–7281
Scheme 1.
Scheme 2.
Table 1. Asymmetric PTAS of 2a–c with catechol, giving 3a as the producta
Entry
Substrate
Base
Time (d)
Yield (%)
%eeb
Confignc
1
2
3
4
5
6
7
8e
9
2a
2b
2c
2a
2a
2a
2a
2a
2c
K2CO3
K2CO3
K2CO3
Cs2CO3
Na2CO3
CsF
7
19
8
97
65d
74d
91
62
64
65
24
––
56
71
66
71
R
R
R
R
––
R
R
R
R
1
––
1
––
77
KF
KF
9
87
14
14
17d
85
KF
a All reactions were carried out in CH2Cl2 at room temperature with a 2/catechol/base/[Pd(C3H5)Cl]2/1 molar ratio = 1:3:3:0.025:0.05, unless
otherwise mentioned.
b Determined by HPLC analysis using chiral stationary phase column (Daicel Chiralcel OJ-H; hexane–i-PrOH = 99.5:0.5).
c Absolute configuration was determined by chiroptical comparison with the published value (Ref. 6c).
d Reaction was not completed.
e Reaction was performed at 0°C.
acetylgroup in 2a with an i-propoxycarbonyl( 2b) or a
pivaloyl group (2c) slowed the reaction, though the
enantioselectivities were slightly improved (entries 2
and 3). Thus, the reaction conditions were optimized
by using 2a as the substrate. It is well known that the
base affects the reaction rate and/or the enantioselectiv-
ity in allylic substitution. Thus, the effect of the base was
next examined. Use of Cs2CO3 accelerated the reaction
but adversely affected enantioselectivity (entry 4). On
the other hand, use of Na2CO3 badly decelerated the
reaction (entry 5). Use of CsF also accelerated the reac-
tion but somewhat diminished enantioselectivity (entry
6). Use of an amine base such as triethylamine or Proton
spongeÒ reduced the reaction rate seriously. Finally, the
highest enantioselectivity of 71% ee was observed when
KF was used as the base (entry 7).7
Lowering the reaction temperature to 0°C diminished
the enantioselectivity and the reaction rate (entry 8).
We examined the reactions of 2b and 2c under the
optimized conditions: the reaction of 2c proceeded
somewhat slowly with the same enantioselectivity as
that of 2a (entry 9), but that of 2b was much slower.
The reaction using 2-(benzylamino)phenol was next
examined with KF as the base (Table 2). The reaction
of 2a was fast but moderately enantioselective (entry
1). Fortunately, the reaction of 2b proceeded smoothly

K. Ito et al. / Tetrahedron Letters 45 (2004) 7277–7281
7279
Table 2. Asymmetric PTAS of 2a–c with other nucleophiles, giving 3c,e,g as the productsa
Entry
Substrate
Nucleophile
Product
Base
Time (h)
Yield (%)
%eeb
Confign
1
2a
2b
2b
2c
2b
2-(Benzylamino)phenol
3c
3c
3c
3c
3c
KF
KF
24
24
98
89
96
92
72e
66
72
86
67
69
Rc
Rc
Rc
Rc
Rc
2
3d
4
KF
KF
72
24
5
K2CO3
144
6
2a
2b
2c
2c
2-(Benzylamino)ethanol
3e
3e
3e
3e
KF
KF
KF
KF
24
48
24
72
67
51e
91
74
69
75
75
81
Rf
Rf
Rf
Rf
7
8
9d
10
11
12
13d
14
15
2a
2a
2b
2b
2b
2c
1,2-Bis(benzylamino)ethane
3g
3g
3g
3g
3g
3g
K2CO3
KF
24
24
99
85
88
88
81
83
64
54
80
86
76
73
Rf
Rf
Rf
Rf
Rf
Rf
K2CO3
K2CO3
KF
24
72
24
120
K2CO3
a All reactions were carried out in CH2Cl2 at room temperature with a 2/nucleophile/base/[Pd(C3H5)Cl]2/1 molar ratio = 1:3:3:0.025:0.05, unless
otherwise mentioned.
b Determined by HPLC analysis using chiral stationary phase column (Daicel Chiralpak AD-H; hexane–i-PrOH = 99:1).
c Absolute configuration was determined by chiroptical comparison with the published value (Ref. 6c).
d Reaction was performed at 0°C.
e Reaction was not completed.
f Absolute configuration was determined by chiroptical comparison with the published value (Ref. 6f).
with improved enantioselectivity (entry 2). Lowering
the reaction temperature to 0°C further improved the
enantioselectivity to 86% ee (entry 3). The reaction of
2c also proceeded smoothly but the enantioselectivity
was diminished (entry 4). The reaction of 2b with
K2CO3 was slow.
designed 1 with the expectation that the 7-isopropyl
group would direct inward to regulate the coordination
sphere around the palladium ion efficiently and induce
high asymmetry.4 To clarify the structure of the Pd–1
complex, we performed its X-ray analysis (Fig. 1).8
Although the crystal of the PdCl2–1 complex possessed
two molecules in a crystal lattice, their structural fea-
tures are almost identical and the discussion is limited
to one of them. The six-membered chelate ring adopts
an envelope-like form and the palladium atom is out
of the plane. One of the phenyl groups on the phospho-
rus atom, which is cis to the isopropylgroup, occupies a
pseudoaxialposition, and the other occupies a pseudo-
equatorialone. The Pd–Clbond trans to the P atom is
longer than that trans to the N atom, reflecting the trans
effects of the phosphorus and nitrogen atoms. These
structuralfeatures are common to other P,N-ilgand–
Pd complexes.9 However, the PdCl2–1 complex has the
following unique features. The geometry around the pal-
ladium ion is a distorted square planar configuration, in
which the Cl1 atom is located below the N–Pd–P plane.
This distortion is probably caused by the repulsion be-
tween the Cl1 and the C7 atoms. One methyl group in
We also examined the reaction with 2-(benzylamino)eth-
anol(entries 6–9). Of 2a–c, 2c was the best substrate in
terms of enantioselectivity and chemical yield. The reac-
tion at 0°C showed 81% ee.
The reactions of 1,2-bis(benzylamino)ethane were, how-
ever, the best effected with K2CO3 as the base and enan-
tioselectivity of 80% ee was achieved when 2b was used
as the substrate (entry 12). Lowering the reaction tem-
perature to 0°C improved the enantioselectivity up to
86% ee (entry 13).
As discussed above, the best enantioselectivities ever
have been achieved in three of the reactions of four dif-
ferent nucleophiles with 1 as the chiralligand, except for
the reaction of 2-(benzylamino)ethanol.6h We have
Figure 1. Structure of PdCl2–1 complex (hydrogen atoms are omitted for clarity). (a) Front view from chloro ligand; (b) top view from isopropyl
group.

7280
K. Ito et al. / Tetrahedron Letters 45 (2004) 7277–7281
Scheme 3.
Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A., Yama-
moto, H., Eds.; Springer: Berlin, 1999; Vol. 2, pp 833–886.
2. (a) von Matt, P.; Pfaltz, A. Angew. Chem., Int. Ed. Engl.
1993, 32, 566–568; (b) Sprinz, J.; Helmchen, G. Tetrahe-
dron Lett. 1993, 34, 1769–1772; (c) Dawson, G. J.; Frost,
C. G.; Williams, J. M. J.; Coote, S. J. Tetrahedron Lett.
1993, 34, 3149–3150.
3. Reviews: Trost, B. M.; Lee, C. In Catalytic Asymmetric
Synthesis; 2nd ed.; Ojima, I., Ed.; VCH: New York, 2000;
pp 593–649; see also Ref. 1.
4. (a) Ito, K.; Kashiwagi, R.; Iwasaki, K.; Katsuki, T. Synlett
1999, 1563–1566; (b) Ito, K.; Kashiwagi, R.; Hayashi, S.;
Uchida, T.; Katsuki, T. Synlett 2001, 284–286; (c) Ito, K.;
Akashi, S.; Saito, B.; Katsuki, T. Synlett 2003, 1809–1812.
5. Tsuda, T.; Kiyoi, T.; Saegusa, T. J. Org. Chem. 1990, 55,
3388–3390.
6. (a) Massacret, M.; Goux, C.; Lhoste, P.; Sinou, D.
Tetrahedron Lett. 1994, 33, 6093–6095; (b) Lhoste, P.;
Massacret, M.; Sinou, D. Bull. Soc. Chim. Fr. 1997, 134,
343–347; (c) Massacret, M.; Lhoste, P.; Lakhmiri, R.;
Parella, T.; Sinou, D. Eur. J. Org. Chem. 1999, 2665–2673;
(d) Masacret, M.; Lakhmiri, R.; Lhoste, P.; Nguefack, C.;
Abdelouahab, F. B. B.; Fadel, R.; Sinou, D. Tetrahedron:
Asymmetry 2000, 11, 3561–3568; (e) Yamazaki, A.;
Achiwa, I.; Achiwa, K. Tetrahedron: Asymmetry 1996, 7,
403–406; (f) Uozumi, Y.; Tanahashi, A.; Hayashi, T. J.
Org. Chem. 1993, 58, 6826–6832; (g) Yamazaki, A.;
Achiwa, K. Tetrahedron: Asymmetry 1995, 6, 1021–1024;
(h) Nakano, H.; Yokoyama, J.; Fujita, R.; Hongo, H.
Tetrahedron Lett. 2002, 43, 7761–7764.
the isopropylgroup is col se to the pseudoaxialphenyl
˚
group at ca. 3.5A, probably due to an attrac-
tive CH–p interaction.10 This directs the other methyl
group over the Cl1 atom, constituting an asymmetric
coordination sphere, as we expected (see also Scheme
3). This also explains why the isopropyl group is supe-
rior to other groups as the C7 substituent of 1.4
PTAS is a four-step reaction:5,6d,f (i) oxidative addition;
(ii) intermolecular allylic substitution at the terminal
carbon; (iii) the second p-allyl palladium formation;
and (iv) intramolecular ring formation giving a dihetero-
cyclic compound. The final ring formation should be the
step determining the stereochemistry. In the step, the
possible diastereomeric p-allyl palladium complexes
(A, B, D, and E) have been reported to be in a rapid
equilibrium (Scheme 3).1 A nucleophile has been
reported to preferentially attack the carbon trans to
the P-atom, reflecting the strong trans-effect of phospho-
rus.1b Furthermore, it has been proposed that the allyl
system rotates in the direction causing less steric repul-
sion, when the nucleophile attacks the p-allyl complex
to give a Pd(0)–olefin complex. Taking into account
the above proposals1b together with the X-ray structure,
the reaction is considered to give the R-isomer as the
major product via C.
In conclusion, we have demonstrated that 2-(phosphino-
phenyl)pyridine 1 was an efficient chirailgland for
PTAS and discussed asymmetric induction by 1 based
on the X-ray structure of PdCl2–1 complex.
7. Typicalexperimentalprocedure was exempilfied by asym-
metric PTAS of 2a and catechol: Allylpalladium(II)
chloride dimer (1.3mg, 3.6lmol) and 1 (3.1mg, 7.3lmol)
were dissolved in dichloromethane (1.5mL) under nitro-
gen and stirred for 1h at room temperature. Compound 2a
(25.0mg, 0.145mmol), catechol (16.0mg, 0.145mmol),
and potassium fluoride (25.3mg, 0.435mmol) were suc-
cessively added to the solution and stirred for 9days at the
temperature. The mixture was quenched with water and
extracted with dichloromethane. The extract was dried
over anhydrous MgSO4 and concentrated. Silica gel
chromatography of the residue (pentane–diethyl
ether = 30:1) gave 2-vinyl-1,4-benzodioxane (20.4mg,
87%) as an oil. Its ee was determined as described in the
footnote to Table 1.
Acknowledgements
We thank associate Prof. Ryo Irie, this Department,
for X-ray structuralanaylsis of 1. B.S. acknowledges
Research Fellowships of the Japan Society for the Pro-
motion of Science for Young Scientists.
8. Crystallographic data for PdCl2–1: Recrystallized from
dichloromethane–ethyl acetate, C29H28NPCl2Pd, M =
References and notes
˚
598.83, monoclinic, space group P21, a = 10.0767A,
˚
˚
1. Reviews: (a) Hayashi, T. In Catalytic Asymmetric Synthe-
sis; Ojima, I., Ed.; VCH: Weinheim, 1993; pp 325–365; (b)
Trost, B. M.; van Vranken, D. L. Chem. Rev. 1996, 96,
395–422; (c) Pfaltz, A.; Lautens, M. In Comprehensive
b = 18.5150A, c = 14.8170A,
˚
b = 102.3170°,
V =
l(Mo-Ka) =
3
2700.7800A , Z = 4, Dc = 1.473gcmÀ3
,
7.107cmÀ1, R1 = 0.0385, wR2 = 0.1119 for 9006 reflec-
tions and 614 variables, GOF = 0.890. Data were collected

K. Ito et al. / Tetrahedron Letters 45 (2004) 7277–7281
7281
on a Rigaku RAXIS RAPID imaging plate area detector
with graphite monochromated Mo-Ka at 20°C. Structural
analysis was performed using the teXsan crystallographic
software package. The structure was solved by direct
methods (SIR92) and expanded using Fourier techniques
(DIRDIF94). Some non-hydrogen atoms were refined
anisotropically, while the rest were refined isotropically.
Hydrogen atoms were refined using the riding model. The
Crystallographic data (excluding structure factors) have
been deposited with the Cambridge Crystallographic Data
Centre as supplementary publication numbers CCDC
244644.
9. (a) Sprinz, J.; Kiefer, M.; Helmchen, G. Tetrahedron Lett.
1994, 35, 1523–1526; (b) Kollmar, M.; Steinhagen, H.;
Janssen, J. P.; Goldfuss, B.; Malinovskaya, S. A.;
´
Vazquez, J.; Rominger, F.; Helmchen, G. Chem. Eur. J.
2002, 8, 3103–3114.
10. Nishio, M.; Hirota, M. Tetrahedron 1989, 45, 7201–7245.