Palladium-(S,pR)-FerroNPS-catalyzed asymmetric allylic etherification: Electronic effect of nonconjugated substituents on benzylic alcohols on enantioselectivity

Article abstract of DOI:10.1002/anie.200703955

(Chemical Equation Presented) An enantioselective C-O bond-forming reaction proceeds well under palladium catalysis with newly developed N-P,S ligands with a ferrocene motif (see scheme). Nonconjugated substituents on the benzylic alcohol substrate lead to an increase in enantioselectivity with increasing electron-donating ability in the title reaction with racemic 1,3-diphenyl-2-propenyl acetate. Cy = cyclohexyl.

Full text of DOI:10.1002/anie.200703955

Communications
DOI: 10.1002/anie.200703955
Asymmetric Catalysis
Palladium–(S,pR)-FerroNPS-Catalyzed Asymmetric Allylic
Etherification: Electronic Effect of Nonconjugated Substituents on
Benzylic Alcohols on Enantioselectivity**
Fuk Loi Lam, Terry Tin-Lok Au-Yeung, Fuk Yee Kwong, Zhongyuan Zhou, Kwok Yin Wong,
and Albert S. C. Chan*
The development of efficient methods for enantioselective
synthesis remains at the center of modern-day organic
chemistry, as such methods have many important applica-
tions, from the total synthesis of natural products[1] to the
preparation of analogues of lead compounds in the pharma-
ceutical industry. The ability to prepare compounds by a
carbon–heteroatom bond-forming process from a common
intermediate is of great significance to the drug-discovery
process. In particular, the stereoselective construction of an
ether linkage adjacent to a stereogenic carbon center is
important for the synthesis of many biologically active
targets.[2] However, this process requires further develop-
Enantioselective iridium-catalyzed[5] allylic substitution
reactions with a broad range of phenols (relatively soft
nucleophiles) have been reported. They generally proceed
with good selectivity with monodentate phosphoramidite
À
ligands. Asymmetric palladium-catalyzed C O bond forma-
tion between phenols and various allylic substrates to give
ethers has also been studied.[6–8] In a separate study, Kim and
Lee demonstrated that the palladium-catalyzed etherification
of allylic acetates with aliphatic alcohols afforded achiral
ethers by using zinc alkoxides generated from diethyl zinc and
an alcohol.[9] Haight et al. reported an asymmetric variant of
the protocol described by Kim and Lee. However, the more
reactive allylic carbonate and harsher conditions (reflux in
THF) were required, and the observed enantioselectivities
were rather poor.[7c] Spurred by these findings, we undertook
the challenge to develop an efficient etherification process
that can proceed under mild reaction conditions with good
stereoselectivity. Herein, we report a general palladium-
catalyzed asymmetric allylic substitution of racemic 1,3-
diphenyl-2-propenyl acetate with aliphatic alcohols in the
presence of newly developed fine-tunable phosphinamidite–
thioether ligands with a ferrocene motif (Scheme 1) to
generate chiral ethers in high yields with excellent enantio-
selectivities.
À
ment. For example, the conventional formation of a C O
bond by a direct SN2-type O alkylation (Williamson ether
synthesis) is sometimes impractical synthetically owing to the
strong basicity of the alkoxide anion, which may be incom-
patible with other functional groups present in the system. It
À
would clearly be advantageous to construct C O bonds in a
catalytic manner under mild conditions rather than through
traditional organic synthesis. Enantioselective transition-
metal-catalyzed allylic substitution[3] has become one of the
most powerful tools for the generation of carbon–carbon and
carbon–heteroatom bonds with various nucleophiles. The
development of the synthesis of chiral compounds containing
carbon–carbon or carbon–nitrogen bonds from racemic allylic
electrophiles has been documented well [Eq. (1)]. In contrast,
the enantioselective allylic substitution of unactivated allylic
acetates with relatively hard oxygen nucleophiles has only
been studied sporadically.[4]
We recently developed a convenient synthesis of the
versatile Ugi amine[10] in optically pure form with a view to
using it as a building block for the development of novel and
highly modular chiral ligands.[11] The chiral intermediate
aminothioether 2 of FerroNPS was synthesized by diastereo-
selective ortho lithiation of the Ugi amine by treatment with
sBuLi in Et2O followed by quenching with the appropriate
[*] F. L. Lam, Dr. T. T.-L. Au-Yeung, Dr. F. Y. Kwong, Prof. Dr. Z. Zhou,
Prof. Dr. K. Y. Wong, Prof. Dr. A. S. C. Chan
Open Laboratory of Chirotechnology of the
Institute of Molecular Technology for Drug Discovery and Synthesis
and
Department of Applied Biology and Chemical Technology
The HongKongPolytechnic University
HungHom, Kowloon, HongKongSAR (China)
Fax: (+852)2364-9932
E-mail: bcachan@polyu.edu.hk
[**] We thank the University Grants Committee Area of Excellence
Scheme (AoE/P-10/01) and The HongKongPolytechnic University
(Area of Strategic Development) for financial support of this study.
FerroNPS refers to a series of P,S ligands with a ferrocenyl motif and
an amine linkage.
Supportinginformation for this article is available on the WWW
under http://www.angewandte.org or from the author.
Scheme 1. Preparation of (S,pR)-FerroNPS ligands. Cy = cyclohexyl.
1280
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 1280 –1283

Angewandte
Chemie
Table 1: Initial screeningof the asymmetric allylic etherification of 1,3-
diphenyl-2-propenyl acetate with benzyl alcohol.
disulfide (Scheme 1). The treatment of 2 with hot acetic
anhydride and an aqueous solution of methylamine gave the
ferrocenyl methylamine 3. A simple phosphanylation step
with Ph2PCl under basic conditions then afforded the
enantiomerically pure ferrocenyl-scaffolded ligands L1–L5.
This route has the potential to offer ready access to a large
array of ligands with modular steric and electronic properties
at both the thioether and the phosphanyl moieties.[12] In fact, a
small modification at the thioether group of FerroNPS ligands
leads to improved reactivity or enantioselectivity.[13]
Entry
L (mol%)
Solvent
Yield [%][a]
ee [%][b]
1[c]
2[c]
3[c]
4[c]
5
6
7
8
L5 (4)
L5 (4)
L5 (4)
L5 (4)
L1 (4)
L2 (4)
L3 (4)
L4 (4)
L5 (4)
L5 (4)
CH2Cl2
THF
91
83
93
83
93
92
92
93
96
95
88.9 (S)
91.0 (S)
90.7 (S)
88.5 (S)
81.8 (S)
87.8 (S)
86.0 (S)
88.6 (S)
91.5 (S)
95.5 (S)
toluene
CH3CN
toluene
toluene
toluene
toluene
toluene
toluene
The absolute configuration of L2 was determined unam-
biguously to be S,pR by single-crystal X-ray crystallography
(Figure 1). The ORTEP representation of the structure shows
9
10[d]
[a] Yield of the isolated product after chromatography. [b] The ee value
was determined by HPLC on a chiral stationary phase. The absolute
configuration was established by correlation with literature data.[7c]
[c] Cs2CO3 was weighed in air rather than under nitrogen in a dry box.
[d] The reaction was performed at 08C for 17 h. Bn = benzyl.
(Table 1, entries 5–9). Thioether ligands with a tert-butyl,
phenyl, or isopropyl group showed somewhat lower enantio-
selectivities (Table 1, entries 6–8) than that of L5. Higher
enantioselectivity was observed when the reaction temper-
ature was lowered to 08C (Table 1, entry 10).
To test the effectiveness of the Pd–L5 catalytic system, we
examined a series of substituted benzylic alcohols (Table 2).
Figure 1. X-ray structure of L2 (ORTEP view; thermal ellipsoids at 30 %
probability).
Table 2: Enantioselective allylic etherification of 1,3-diphenyl-2-propenyl
acetate with 4-substituted benzylic alcohols under the catalysis of a Pd–
L5 complex.
that the sulfur and stereogenic carbon atoms lie in the plane
of the Cp ligand C6–C10. In the free ligand, the tert-butyl
group and the methyl group are oriented such that they point
away from each other, presumably to minimize steric
congestion. The even bulkier diphenylphosphanyl moiety is
located further away from the thioether group.
Entry
R
4, 5
t [h]
Yield [%][a]
ee [%][b]
1
2
3
4
5
6
CH2Ph
a
b
c
d
e
f
2
98
94
94
95
92
93
91.6 (S)
94.5
93.4
89.6
88.8
CH2C6H4-p-OMe
CH2C6H4-p-Me
CH2C6H4-p-F
CH2C6H4-p-Cl
CH2C6H4-p-CF3
2.5
2.5
2.5
2.5
2.5
We first tested the reactivity of chiral (S,pR)-FerroNPS-Cy
(L5) in conjunction with a palladium precatalyst in the
asymmetric allylic etherification (AAE; Table 1). Benzyl
alcohol and racemic 1,3-diphenyl-2-propenyl acetate were
chosen as model substrates and treated with allylpalladium
dimer and Cs2CO3 in CH2Cl2 at room temperature. The
desired ether product was obtained in 91% yield with 89% ee
(Table 1, entry 1). Other bases that are commonly used for
allylic substitution, such as N,O-bis(trimethylsilyl)acetamide
(BSA),[14] were tested, and the use of Cs2CO3 was found to be
crucial to the success of the reaction. The polarity of the
solvent was not found to have a significant effect on product
formation or enantioselectivity. Although the highest enan-
tioselectivity was observed when THF was used as the
solvent, the best result in terms of the yield of the product
(93%) was observed with toluene, along with only a minor
decrease in enantioselectivity (90.7% ee; Table 1, entries 2
and 3). To investigate the efficiency of other ligands in the
family, we studied the influence of the thioether substituent
77.4
[a] Yield of the isolated product after chromatography. [b] The ee value
was determined by HPLC on a chiral stationary phase.
In general, the reaction proceeded smoothly under the
optimized conditions to afford the desired product in
excellent yield. We observed an intriguing relationship
between the enantioselectivity of the reaction and the
electronic nature of the substituted benzylic alcohol. Higher
enantioselectivity was observed when the benzylic alcohol
contained an electron-rich para substituent, and the selectiv-
ity diminished gradually as the substituent became more
electron deficient.
The aromatic electronic effect can be represented by a
Hammett relationship.[15] The Hammett plot of log(ratio of
Angew. Chem. Int. Ed. 2008, 47, 1280 –1283
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1281

Communications
Table 3: Enantioselective allylic etherification of 1,3-diphenyl-2-propenyl
acetate with a variety of alcohols under the catalysis of a Pd–L5 complex.
enantiomers) against sp shows a linear free-energy relation-
ship (Figure 2; 1 = À0.77, r = 0.975) between enantiose-
lectivity and the electronic character of the substituent. This
electronic effect appears to be significant in this nonconju-
Entry
1
ROH
4, 5
t [h]
Yield [%][a]
87
ee [%][b]
g
2.5
94.7
2
3
h
i
2.5
2.5
88
94
93.8
92.6
4
5
j
24
21
33[c]
71
n.d.
Figure 2. Hammett plot for the Pd-catalyzed asymmetric allylic ether-
ification with L5 (product: 77–94% ee).
k
82.9
6
7
8
l
2.5
21
96
82
78
89.4
93[d]
m
n
gated system. To the best of our knowledge, such an electronic
effect of a nonconjugated substrate on enantioselectivity has
not been reported to date. This observation may provide a
useful tool for predicting the enantioselectivity of asymmetric
allylic substitution when substituted benzylic alcohols are
used as nucleophiles. It is complementary to the well-known
electronic effect on stereoselectivity of either a substrate or a
ligand composed of conjugated aromatic systems.[16]
21
93.2[d]
9
o
20
96
93.5
10
11
p
q
3
6
98
98
92.7
94.1
12
r
3
94
93.4
To further extend the scope of the reaction with respect to
the alcohol substrate, ortho- and meta-substituted benzylic
alcohols (Table 3, entries 1–3) were applied successfully in the
AAE to give the ether products in high yield with high
enantioselectivity. The Pd–L5 catalytic system was also found
to be compatible with heterocycles (Table 3, entries 5–8). The
reaction of the potentially problematic substrate 2-pyridine-
methanol, the nitrogen atom of which might coordinate
competitively to the metal center of the catalyst, proceeded
smoothly to give the corresponding ether, although the
reaction time had to be extended to 21 h (Table 3, entry 5).
Primary aliphatic alcohols, such as allyl alcohol and n-butanol,
underwent the desired reaction to provide the product in
excellent yield with high enantioselectivity (Table 3,
entries 10 and 11). The use of the secondary alcohol 4r led
to the desired product in good yield with 93.4% ee, whereas
only moderate conversion was observed with the less strained
secondary alcohol 4s under the same reaction conditions
(Table 3, entries 12and 13). tert-Butanol was found to be a
poor substrate for this transformation.
In summary, we have reported the synthesis of the new
ferrocenyl ligand L5 and derivatives thereof. This scaffold has
several beneficial features, including ease of accessibility and
the possibility of fine-tuning the steric and electronic proper-
ties of the donor atoms. The corresponding palladium
complexes of the FerroNPS ligands L1–L5 were employed
effectively in the palladium-catalyzed asymmetric allylic
etherification of racemic 1,3-diphenyl-2-propenyl acetate
with a wide array of aliphatic alcohols with good to excellent
13
14
s
t
29
24
58 (73)[e]
trace
96.2
n.d.
[a] Yield of the isolated product after chromatography. [b] The ee value
was determined by HPLC on a chiral stationary phase. [c] The ether
product decomposed significantly on a TLC plate. [d] The de value is
given instead of an ee value. [e] The conversion (given in brackets) was
1
determined by H NMR spectroscopy. n.d.=not determined.
enantioselectivities. To the best of our knowledge, we have
described the broadest substrate scope reported to date for
AAE. We also observed the first example of an electronic
effect of a nonconjugated substituent on a benzylic alcohol on
the enantioselectivity of a reaction. This finding may aid in
predicting the enantioselectivity of certain reactions. Further
mechanistic investigations into AAE and the application of
N–P,S ligands with a ferrocene-motif to other enantioselective
reactions are under way.
Received: August 28, 2007
Revised: November 10, 2007
Published online: January 3, 2008
Keywords: asymmetric catalysis · electronic effects ·
.
etherification · P,S ligands · palladium
1282
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 1280 –1283

Angewandte
Chemie
nium: $490, platinum: $1237 ($ per oz, August 2007; 1 oz =
28.35 g); http://www.platinum.matthey.com.
[9] H. Kim, C. Lee, Org. Lett. 2002, 4, 4369.
[10] D. Marquarding, H. Klusacek, G. Gokel, P. Hoffmann, I. Ugi, J.
Am. Chem. Soc. 1970, 92, 5389.
[1] K. C. Nicolaou, E. J. Sorensen, Classics in Total Synthesis:
Targets, Strategies, and Methods, VCH, Weinheim, 1996.
[2] J. Buckingham, Dictionary of Natural Products, University Press,
Cambridge, Cambridge, MA, 1994.
[3] a) B. M. Trost, C. Lee in Catalytic Asymmetric Synthesis, 2nd ed.
(Ed.: I. Ojima), Wiley-VCH, Weinheim, 2000, chap. 8; b) B. M.
Trost, M. L. Crawley, Chem. Rev. 2003, 103, 2921; c) J. Tsuji,
Palladium Reagentsand Catalysis , Wiley, New York, 2004; d) A.
Pfaltz, M. Lautens in Comprehensive Asymmetric Catalysis I–III,
Vol. 2, Springer, New York, 1999, p. 833.
[4] For the effect of substituents on phenols on the enantioselectiv-
ity of etherification in toluene, see: B. M. Trost, F. D. Toste, J.
Am. Chem. Soc. 1999, 121, 4545.
[11] For the large-scale of production of the chiral Ugi amine by
asymmetric hydrogenation, see: a) W. S. Lam, S. H. L. Kok,
T. T.-L. Au-Yeung, J. Wu, H. Y. Cheung, F. L. Lam, C. H. Yeung,
A. S. C. Chan, Adv. Synth. Catal. 2006, 348, 370; for the
development of our new ligands, see: b) F. L. Lam, T. T.-L. Au-
Yeung, H. Y. Cheung, S. H. L. Kok, W. S. Lam, K. Y. Wong,
A. S. C. Chan, Tetrahedron: Asymmetry 2006, 17, 497; c) H. Y.
Cheung, W.-Y. Yu, F. L. Lam, T. T.-L. Au-Yeung, Z. Zhou, T. H.
Chan, A. S. C. Chan, Org. Lett. 2007, 9, 4295.
[5] a) C. Shu, J. F. Hartwig, Angew. Chem. 2004, 116, 4898; Angew.
Chem. Int. Ed. 2004, 43, 4794; b) C. Fischer, C. Defieber, T.
Suzuki, E. M. Carreira, J. Am. Chem. Soc. 2004, 126, 1628; c) F.
Lopez, T. Ohmura, J. F. Hartwig, J. Am. Chem. Soc. 2003, 125,
3426; d) I. Lyothier, C. Defieber, E. M. Carreira, Angew. Chem.
2006, 118, 6350; Angew. Chem. Int. Ed. 2006, 45, 62 04; e) C.
Welter, A. Dahnz, B. Brunner, S. Streiff, P. Dubon, G. Helmchen,
Org. Lett. 2005, 7, 1239.
[12] M. O. Okoroafor, D. L. Ward, C. H. Brubaker, Organometallics
1988, 7, 1504.
[13] N. Khiar, I. Fernandez, V. Valdivia, B. Gori, F. Alcudia, E.
Alvarez, Org. Lett. 2005, 7, 1307.
[14] When BSA was used, a complex mixture of products was
observed.
[15] For reviews on Hammett studies, see: a) C. Hansch, A. Leo,
R. W. Taft, Chem. Rev. 1991, 91, 165; b) A. J. Gordon, R. A.
Ford, The Chemistꢀs Companion, Wiley, New York, 1972.
[16] For recent references on aromatic electronic effects in asym-
metric catalysis, see: a) E. N. Jacobsen, W. Zhang, M. L. Güler, J.
Am. Chem. Soc. 1991, 113, 6703; b) H. C. Zhang, F. Xue, T. C. W.
Mak, K. S. Chan, J. Org. Chem. 1996, 61, 8002; c) M. M. C. Lo,
G. C. Fu, J. Am. Chem. Soc. 1998, 120, 10270; d) H. Doucet, E.
Fernandez, T. P. Layzell, J. M. Brown, Chem. Eur. J. 1999, 5,
1320; e) W. C. Lo, C. M. Che, K. F. Cheng, T. C. W. Mak, Chem.
Commun. 1997, 1205; f) F. Y. Kwong, Q. Yang, T. C. W. Mak,
A. S. C. Chan, K. S. Chan, J. Org. Chem. 2002, 67, 2769; g) F. Y.
Kwong, Y.-M. Li, W. H. Lam, L. Qiu, H. W. Lee, K. S. Chan,
C. H. Yeung, A. S. C. Chan, Chem. Eur. J. 2005, 11, 3872.
[6] B. M. Trost, H. C. Shen, L. Dong, J.-P. Surivet, J. Am. Chem. Soc.
2003, 125, 9276.
[7] For Pd-catalyzed enantioselective etherification with a limited
range of O nucleophiles, see: a) B. M. Trost, F. D. Toste, J. Am.
Chem. Soc. 1998, 120, 815; b) B. M. Trost, F. D. Toste, J. Am.
Chem. Soc. 2000, 122, 11262; c) A. R. Haight, E. J. Stoner, M. J.
Peterson, V. K. Grover, J. Org. Chem. 2003, 68, 8092; d) M.
Kimura, Y. Uozumi, J. Org. Chem. 2007, 72, 707; e) Y. Uozumi,
M. Kimura, Tetrahedron: Asymmetry 2006, 17, 161; f) L. F.
Tietze, J. K. Lohmann, C. Stadler, Synlett 2004, 1113; g) A.
Iourtchenko, D. Sinou, J. Mol. Catal. A 1997, 122, 91.
[8] For industrial use, the price of the metal may be an important
factor: palladium: $331, iridium: $450, rhodium $6100, ruthe-
Angew. Chem. Int. Ed. 2008, 47, 1280 –1283
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1283