First planar chiral bidentate ligand based on a (η5- cyclopentadienyl)(η4-cyclobutadiene) cobalt backbone: High efficiency in enantioselective palladium-catalyzed allylic substitutions

Article abstract of DOI:10.1039/b405342g

The synthesis of the planar chiral (R)-[η⁵-(1- diphenylphosphino-2-tert-butylsulfenyl)cyclopentadienyl](η⁴- tetraphenylcyclobutadiene) cobalt and its high efficiency as P,S-bidentate ligand in Pd-catalyzed allylic substitutions is described.

Full text of DOI:10.1039/b405342g

View Article Online / Journal Homepage / Table of Contents for this issue
First planar chiral bidentate ligand based on a
(h5-cyclopentadienyl)(h4-cyclobutadiene) cobalt backbone: high efficiency
in enantioselective palladium-catalyzed allylic substitutions†
Ramón Gómez Arrayás, Olga García Mancheño and Juan C. Carretero*
Departamento de Química Orgánica, Facultad de Ciencias, Universidad Autónoma de Madrid,
Cantoblanco, 28049 Madrid, Spain. E-mail: juancarlos.carretero@uam.es; Fax: 34 913973966; Tel: 34
913973925
Received (in Cambridge, UK) 8th April 2004, Accepted 19th May 2004
First published as an Advance Article on the web 17th June 2004
The synthesis of the planar chiral (R)-[h5-(1-diphenylphos-
phino-2-tert-butylsulfenyl)cyclopentadienyl](h4-tetraphenylcy-
clobutadiene) cobalt and its high efficiency as P,S-bidentate
ligand in Pd-catalyzed allylic substitutions is described.
ortho-lithiation of oxazolines of type I with n-BuLi, sec-BuLi and
tert-BuLi has also been reported to fail, despite the use of additives
such as TMEDA and prolonged reaction times.8b
Initial efforts towards deprotonation of complex 211 with n-BuLi,
sec-BuLi or tert-BuLi, both alone or in combination with activating
additives such as TMEDA or DMPU, followed by trapping with
various electrophiles were unsuccessful. Gratifyingly, the function-
alization of 2 at the Cp ring was achieved with the superbasic
reagent combination of tert-BuLi (1.0 equiv.) and potassium tert-
butoxide (1.0 equiv.) in a 3 : 1 mixture of THF–pentane at 0 °C.12,13
Quenching with enantiopure (R)-S-tert-butyl tert-butanethiosulfi-
nate14 at 278 °C produced the (R)-tert-butylsulfoxide 3 in 36%
yield (64% of starting complex 2 was recovered) and extremely
high enantiopurity (ee > 99.5%, HPLC, Chiralcel OD column) as a
result of the complete inversion of the configuration at sulfur
(Scheme 1).
In recent years the use of chiral bidentate ligands having strong and
weak donor heteroatomic groups has proved to be a very useful tool
in enantioselective catalysis.1 Among the wide variety of chiral
skeletons currently used in ligand design, planar chiral 1,2-dis-
ubstituted ferrocenes with P,P- and P,N-coordination modes have
become increasingly popular,2 and even industrial applications
have been developed.3 In this field we have recently described that
1-phosphino-2-sulfenylferrocene ligands (Fesulphos), possessing
planar chirality only, act as very efficient P,S-ligands in Pd-
catalyzed allylic substitutions4 and in Cu-catalyzed formal hetero
Diels–Alder cycloadditions.5 Taking into account the great interest
in ferrocenes in enantioselective catalysis, it is surprising that few
studies dealing with non-metallocene planar chiral ligands have
been reported.6 One of these appealing organometallic backbones is
Interestingly, the diastereoselective functionalization at C-2 was
smoothly achieved by simple treatment of (R)-3 with 1.2 equiva-
lents of tert-BuLi (THF, 278 °C) and subsequent addition of
Ph2PCl. The phosphine (RS,Rp)-4 was obtained in 84% yield as the
only product detected by NMR in the crude reaction mixture. Once
the central (R) chirality of the sulfinyl directing group was applied
to the generation of the (R) planar chirality, the sulfoxide 4 was
reduced to the desired sulfide ligand (R)-1 by treatment with
HSiCl3–Et3N in refluxing toluene (86% yield). As expected, optical
purity of (R)-1 was proved to be higher than 99.5% ee (HPLC,
Chiralpak AD column).
5
the readily available and chemically robust (h -cyclopentadie-
4
nyl)(h -cyclobutadiene)cobalt moiety. With these types of com-
pounds, as far as we are aware, only the chiral oxazoline-based
complexes I7 and a palladacycle derivative II,8 reported by
Richards et al. (Fig. 1), have been used in enantioselective
catalysis.
5
This paper reports the synthesis of the (R)-[h -(1-phosphino-
4
2-sulfenyl)cyclopentadienyl](h -tetraphenylcyclobutadiene)cobalt
complex 1 (Cosulphos ligand), having solely planar chirality, and
its high asymmetric efficiency as a P,S-bidentate ligand in Pd-
catalyzed allylic substitutions.
The efficiency of this new ligand was tested in the asymmetric
Pd-catalyzed allylic substitution reaction. Table 1 shows the results
obtained in the reaction of 1,3-diphenylpropenyl acetate with a
variety of nucleophiles in the presence of (R)-1. Typically the
We envisaged that the introduction of planar chirality at the
4
3
Cp(h -C4Ph4)Co unit could be accomplished, as previously
reactions were performed using a combination of [Pd(h -C3H5)Cl]2
developed by Kagan et al. in the case of ferrocenes,9 by
diastereoselective ortho-lithiation of a suitable sulfinyl derivative
such as 3, which in turn could be accessed by lithiation of complex
(2 mol%) and ligand 1 (4 mol%) as catalyst at temperatures from
220 °C to room temperature. After a brief search of solvents, it was
found that THF and CHCl3 provided the best results in terms of
both reactivity and enantioselectivity, the former being optimal in
the case of malonate (entries 1–5) and benzylamine, and the latter
was found to be the best solvent for the reaction with potassium
phthalimide (entries 6–7). Interestingly, good yields and excellent
enantioselectivities (97–99% ee) were achieved with all nucleo-
philes.
5
4
(h -C5H5)(h -C4Ph4)Co (2) and subsequent treatment with a chiral
sulfinylating agent. However, as others have noted previously,10
the functionalization of this type of cobalt complex at the Cp ring
is a challenge. For example, it has been reported that direct
lithiation of complex 2 under usual conditions fails, although this
metallocene was successfully functionalized through electrophilic
aromatic substitution.10 On the other hand, the diastereoselective
The bidentate character of 1 as a P,S-ligand was demonstrated by
an X-ray crystallographic study of the complex (1)PdCl215 (Fig. 2),
readily obtained as a single epimer at sulfur by the reaction of 1
with Pd(CH3CN)2Cl2 in CH2Cl2 at room temperature (82% yield).
5
4
Fig. 1 Chiral (h -C5H5)(h -C4Ph4)Co-based catalysts used in enantiose-
lective catalysis.
Scheme 1 Reagents and conditions: i. t-BuLi (1.0 equiv.)–K-t-BuO (1.0
equiv.), THF–pentane (3 : 1), 278 °C; (R)-t-BuS(O)-S-t-Bu. ii. t-BuLi (1.0
equiv.), THF, 278 °C; Ph2PCl. iii. HSiCl3, Et3N, toluene, 110 °C.
† Electronic supplementary information (ESI) available: crystal data and
experimental details. See http://www.rsc.org/suppdata/cc/b4/b405342g/
1654
C h e m . C o m m u n . , 2 0 0 4 , 1 6 5 4 – 1 6 5 5
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Article Online
The crystal structure of (1)PdCl2 confirmed the (R) planar
configuration previously established on chemical considerations.
To obtain additional mechanistic evidence of the reasons for the
high asymmetric performance of 1, the key cationic p-allyl
palladium intermediate complex 5 was prepared by reaction of
stoichiometric amounts of the (trans-1,3-diphenylpropenyl) palla-
dium dichloride dimer and 1 in the presence of AgSbF6 (85% yield,
as a 3 : 1 mixture of isomers). X-Ray quality crystals of complex 5
were obtained from an EtOAc solution at room temperature. Fig. 2
shows the ORTEP drawing of complex 5a,15 in which the allylic
moiety adopts a W-type configuration. As expected, the bond
between Pd and the terminal allylic carbon atom which is trans to
phosphorus was longer (Pd–C = 2.259 Å) than that which is trans
to sulfur (Pd–C = 2.183 Å), indicating the stronger trans effect of
the phosphine moiety compared to the thioether.4,16 NMR studies
of complex 5 in CD2Cl2 allowed the unambiguous assignment of
the major isomer 5a as W-configuration, mainly based on several
critical NOE contacts.17 Besides, the terminal allylic carbon atom
trans to phosphorus appears at much lower field (101.0 ppm) than
that which is trans to sulfur (77.7 ppm), reflecting the much higher
electrophilicity of the former.4,16 Therefore, the high enantiose-
lectivity achieved in these reactions (97–99% ee) could be readily
explained by the attack of the nucleophile trans to phosphorus on
the complex 5a (Fig. 3), which should exist in a fast equilibrium
with the isomer of M-configuration 5b (Curtin–Hammet condi-
tions).
phenylcyclobutadiene)cobalt complex has been prepared. The
readily available P,S-bidentate ligand (R)-1 (Cosulphos) provides
very high enantioselectivities in Pd-catalyzed allylic substitutions
with both carbon and nitrogen nucleophiles. The extension of this
methodology to the preparation of related planar cosulphos ligands
is underway.
Financial support of this work by the M. C. Y. T. is gratefully
acknowledged (project BQU2000-0266). R. G. A. thanks the M. C.
Y. T. for a Ramón y Cajal contract. O. G. M. thanks the M. E. C.
for a predoctoral fellowship. We thank Johnson Matthey PLC for a
generous loan of PdCl2.
Notes and references
† Electronic supplementary information (ESI) available: crystal data and
experimental details. See http://www.rsc.org/suppdata/cc/b4/b405342g/
1 Catalytic Asymmetric Synthesis, ed. I. Ojima, Wiley-VCH, New York,
2nd edn., 2000.
2 Recent reviews: C. J. Richards and A. J. Locke, Tetrahedron:
Asymmetry, 1998, 9, 2377; A. Togni, in Metallocenes: Synthesis
Reactivity Applications, eds. A. Togni, and R. L. Halterman, Wiley-
VCH, Weinheim, Germany, 1998, vol. 2, p. 685.
3 F. Spindler, Top. Catal., 1997, 4, 275.
4 O. García Mancheño, J. Priego, S. Cabrera, R. Gómez Arrayás, T.
Llamas and J. C. Carretero, J. Org. Chem., 2003, 68, 3679.
5 O. García Mancheño, R. Gómez Arrayás and J. C. Carretero, J. Am.
Chem. Soc., 2004, 126, 456.
6 For a review on non-metallocene chiral ligands: O. Delacroix and J. A.
Gladysz, Chem. Commun., 2003, 665.
7 (a) G. Jones and C. J. Richards, Tetrahedron Lett., 2001, 42, 5553; (b)
G. Jones, D. C. D. Butler and C. J. Richards, Tetrahedron Lett., 2000,
41, 9351.
8 (a) L. E. Overman, C. E. Owen, M. M. Pavan and C. J. Richards, Org.
Lett., 2003, 5, 1809; (b) C. E. Anderson and L. E. Overman, J. Am.
Chem. Soc., 2003, 125, 12412; (c) A. M. Stevens and C. J. Richards,
Organometallics, 1999, 18, 1346.
In summary, the first bidentate chiral ligand based on a
4
stereoselective 1,2-heteroatomic disubstitution on a Cp(h -tetra-
Table 1 Pd-catalyzed reaction of 1,3-diphenyl-2-propenyl acetate with
nucleophiles in the presence of ligand (R)-1
9 F. Rebière, O. Riant, L. Ricard and H. B. Kagan, Angew. Chem., Int. Ed.
Engl., 1993, 32, 568.
10 (a) M. D. Rausch and R. A. Genetti, J. Org. Chem., 1970, 35, 3888; (b)
D. C. D. Butler and C. J. Richards, Organometallics, 2002, 21, 5433.
11 (a) A. Nakamura and N. Hagihara, Bull. Chem. Soc. Jpn., 1961, 34, 452;
(b) J. L. Boston, D. W. Sharpe and G. Wilkinson, J. Chem. Soc., 1962,
3488; (c) P. M. Maitlis and M. L. Games, J. Am. Chem. Soc., 1963, 85,
1887.
12 (a) M. Schlosser, Angew. Chem., Int. Ed. Engl., 1974, 13, 701; (b) M.
Schlosser and H. Geneste, Chem.–Eur. J., 1998, 4, 1969.
13 The use of other superbasic reagents such as n-BuLi–K-t-BuO or s-
BuLi–K-t-BuO in THF–pentane resulted in lower yield of the sulfoxide
4 (18% and 23%, respectively).
Yield
Entry
Nu
Solvent
T/h
(%)a
Ee (%)b
1c
2c
3
(CO2Me)2CH2
(CO2Me)2CH2
BnNH2
BnNH2
BnNH2
CHCl3
THF
CHCl3
THF
THF
CHCl3
CHCl3
48
1
48
3
48
48
48
61
99
65e
96
45e
70
80
95
98
90
97
98
98
99
4
5d
6
KPhth
KPhth
7d
14 D. A. Cogan, G. Liu, K. Kim, B. J. Backes and J. A. Ellman, J. Am.
Chem. Soc., 1998, 120, 8011.
a In pure product after chromatography. b Determined by HPLC (Chiralcel
OD and AD). c In the presence of BSA and 10 mol% of Bu4NCl. d Reaction
performed at 220 °C. e Conversion yield (determined by NMR).
15 Crystal data for C49H42Cl2CoPPdS [(1)PdCl2]: crystal size 0.20 3 0.09
3 0.04 mm3, Mw = 930.09, orthorhombic, space group P2(1)/2(1)/2(1),
a = 13.21140(10), b = 17.1528(2), c = 17.7440(2) Å, a = 90°, b =
90°, g = 90°, V = 4021.01(7) Å3, Z = 4, Dc = 1.536 g cm23, m = 9.188
mm21, T = 100(2) K, CuKa radiation (l = 1.54178 Å), 25737
reflections measured, 7462 independent (Rint = 0.0645). Refinement on
F2 for 7462 reflections and 664 parameters gave GOF = 1.032, R =
0.0315, Rw = 0.0744 for I > 2s(I). CCDC reference number 235288.
Crystal data for C68H61CoF6O2PPdSSb (5a): crystal size 0.20 3 0.15 3
0.15 mm3, Mw
41.0772(4), b
=
1374.28, monoclinic, space group C2/c, a
=
=
=
15.8565(2), c 21.8384(3) Å, a 90°, b
=
=
117.1300(10)°, g = 90°, V = 12659.2(2) Å3, Z = 8, Dc = 1.442 g
cm23, m = 8.702 mm21, T = 100(2) K, CuKa radiation (l = 1.54178
Å), 34812 reflections measured, 10954 independent (Rint = 0.0358).
Refinement on F2 for 10954 reflections and 729 parameters gave GOF
= 1.057, R = 0.0432, Rw = 0.1168 for I > 2s(I). CCDC reference
number 235289. See http://www.rsc.org/suppdata/cc/b4/b405342g/ for
crystallographic data in .cif or other electronic format.
2
Fig. 2 ORTEP drawings of complexes (1)PdCl2 (left) and 5a (right, SbF6
anion has been omitted for clarity).
16 See for instance: T. Tu, Y.-G. Zhou, X.-L. Hou, L.-X. Dai, X.-C. Dong,
Y.-H. Yu and J. Sun, Organometallics, 2003, 22, 1255.
17 Stereochemical assignment of p-allyl complexes 5a and 5b.
Fig. 3 Proposed model for the asymmetric allylic substitutions.
C h e m . C o m m u n . , 2 0 0 4 , 1 6 5 4 – 1 6 5 5
1655