Highly regioselective ruthenium-catalyzed allylic alkylations of chelated enolates

Article abstract of DOI:10.1021/ol102106v

Ru-catalyzed allylic alkylations are a highly interesting alternative to Pd-catalyzed reactions. Ru complexes show a high tendency for regioretention, especially for branched and (Z)-configured substrates, and they do not undergo isomerization of the ally

Full text of DOI:10.1021/ol102106v

ORGANIC
LETTERS
2010
Vol. 12, No. 21
4960-4963
Highly Regioselective
Ruthenium-Catalyzed Allylic Alkylations
of Chelated Enolates†
Anton Bayer and Uli Kazmaier*
UniVersita¨t des Saarlandes, Institut fu¨r Organische Chemie, Im Stadtwald, Geb. 23.2,
D-66123 Saarbru¨cken, Germany
u.kazmaier@mx.uni-saarland.de
Received September 3, 2010
ABSTRACT
Ru-catalyzed allylic alkylations are a highly interesting alternative to Pd-catalyzed reactions. Ru complexes show a high tendency for regioretention,
especially for branched and (Z)-configured substrates, and they do not undergo isomerization of the allyl intermediates formed. Therefore,
(Z)-substrates conserve their olefin geometry, and a perfect chirality transfer is observed with optically active substrates.
π-Allyl metal complexes play a major role in modern organic
synthesis as key intermediates of a wide range of C-C and
Scheme 1. Transition Metal Catalyzed Allylic Alkylations
C-heteroatom couplings.1 The scenery is clearly dominated
by Pd catalysts, but during the last years a range of other
metals, especially late transition metals, made their way into
the limelight.2 This clearly increases the potential of allylation
chemistry because these metals show significantly different
reaction behavior compared to palladium (Scheme 1). For
example, if terminal π-allyl complexes (3) are formed either
from linear (1) or from branched substrate (2), the Pd-
catalyzed allylation provides the linear substitution product
(4) preferentially, while Mo, W, and Ir give rise to the
branched product (5).2 However, especially with Pd catalyst,
the control of the regioselectivity (rs) is not a trivial issue.3
On the other hand, Rh shows a high degree of regioretention,
although with a tendency toward the branched product.4 A
similar behavior is also observed for Ru complexes. With
respect, that with these catalysts allylic alcohols can be used
without further activation5 and that besides C-nucleophiles
also amines,6 alcohols,7 or thiols8 can be allylated, the Ru-
catalyzed allylations are still rather underdeveloped.9
Trost et al.10 and Bruneau et al.11 reported regio- and
stereoselective allylations of various nucleophiles at room
temperature using [Cp*Ru(NCCH3)]PF6 as a catalyst, which
† Dedicated to Prof. Dr. G. Helmchen on the occasion of his 70th
birthday.
(1) Recent reviews: (a) Trost, B. M.; Crawley, M. L. Chem. ReV. 2003,
103, 2921–2943. (b) Lu, Z.; Ma, S. Angew. Chem., Int. Ed. 2008, 47, 258–
297.
(4) (a) Minami, I.; Shimizu, I.; Tsuji, J. J. Organomet. Chem. 1985,
296, 269–280. (b) Evans, P. A.; Nelson, J. D. J. Am. Chem. Soc. 1998,
120, 5581–5582.
(2) Recent reviews: (a) Kazmaier, U.; Pohlman In Metal Catalyzed C-C
and C-N Coupling Reactions; de Meijere, A., Diederich, F., Eds.; Wiley-
VCH: Weinheim, Germany, 2004; pp 531-583. (b) Helmchen, G.;
Kazmaier, U.; Fo¨rster, S. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.;
Wiley-VCH, Weinheim, 2009, pp 497-641 and references cited therein.
(3) (a) Miyabe, H.; Takemoto, Y. Synlett 2005, 1641–1655. (b)
Norsikian, S.; Chang, C.-W. Curr. Org. Chem. 2009, 6, 264–289.
(5) Nieves, I. F.; Schott, D.; Gruber, S.; Pregosin, P. S. HelV. Chim.
Acta 2007, 90, 271–276.
(6) (a) Morisaki, Y.; Kondo, T.; Mitsudo, T. Organometallics 1999, 18,
4742–4746. (b) Kawatsura, M.; Ata, F.; Hirakawa, T.; Hayase, S.; Itoh, T.
Tetrahedron Lett. 2008, 49, 4873–4875.
10.1021/ol102106v  2010 American Chemical Society
Published on Web 09/24/2010

proved to be superior compared to the sterically less
demanding [CpRu(NCCH3)]PF6 catalyst. Comparable prod-
uct ratios (4/5) were obtained, independent of the substrate
used (1 or 2), indicating that the reaction proceeds via a
π-allyl complex, comparable to the Pd-catalyzed process. The
formation of π-allyl intermediates was also confirmed by
detailed mechanistic work from the Pregosin group.12 The
stereochemical outcome of the reaction can be explained by
a double inversion mechanism, as discussed for Pd com-
plexes,1 but in contrast to the Pd-catalyzed process, a
complete chirality transfer can be observed in reactions of
optically pure substrates 2. Obviously, the Ru-π-allyl
complexes do not undergo π-σ-π-isomerization as do the
Pd complexes.12 Although Tunge et al. observed a partial
racemization in decarboxylative allylations of optically active
ꢀ-keto allylic esters, in this case an isomerization of the
allylic substrate (2 f 1) was responsible for the fading ee
and not a π-σ-π isomerization.13
Obviously, the effect of ligands is tremendous in this
reaction. The product ratio can be shifted nearly completely
to the linear product 4, if o-phosphinobenzoic acid/Ru3(CO)12
is used as catalyst,14 while [RuCl2(p-cymene)]2 shows
excellent regioretention.15 This allows a clean conversion
of the linear substrate 1 into the linear product 4 and branched
2 into 5. This outcome can not be explained via a common
π-allyl intermediate (3) but is an indication for a σ-allyl- or
a σ-enyl-complex, as also discussed in Rh-catalyzed pro-
cesses.4
Scheme 2
.
Transition Metal Catalyzed Allylic Alkylations of
Chelated Enolates
-78 °C the π-σ-π isomerization of the π-allyl Pd
intermediates can be suppressed almost completely,19 al-
lowing regioselective attack at the different allylic positions20
and the more or less isomerization-free reaction of (Z)-
substrates.21 Nevertheless, substrates forming terminal allyl
complexes are critical candidates. If optically active sub-
strates 2 are used, complete epimerization is observed, as a
consequence of the fast π-σ-π-isomerization.
This forced us to focus our efforts also on the Rh-catalyzed
version. And indeed, with Wilkinson’s catalyst the branched
product 8 was obtained preferentially, and a nearly perfect
chirality transfer was obtained with optically active sub-
strates.22 However, herewith, only terminal, monosubstituted
allyl substrates 2 showed good conversion, and good
branched-selectivities were only observed for allylic sub-
strates with small substituents, such as 2a. With sterically
more demanding substrates also the linear product is formed,
clearly indicating that in Rh-catalyzed a π-allyl complex
formation can not be neglected.
For some time, our group has investigated Pd-catalyzed
allylic alkylations of chelated enolates such as 6 (Scheme
2), obtained from amino acid esters16 and peptides.17 On the
basis of their high reactivity, these enolates react under much
milder conditions compared to the generally used stabilized
enolates, offering new synthetic options.18 For example, at
Therefore, we were interested to find an alternative catalytic
system showing a broader substrate spectrum but with properties
similar to the Rh catalysts, and we investigated the allylic
alkylation of our chelated enolates in the presence of various
Ru catalysts using racemic butenyl-3-benzoate 2a as a model
(7) (a) Mbaye, M. D.; Demerseman, B.; Renaud, J.-L.; Toupet, L.;
Bruneau, C. AdV. Synth. Catal. 2004, 346, 835–841. (b) Hermatschweiler,
R.; Ferna´ndez, I.; Pregosin, P. S. Organometallics 2006, 25, 1440–1447.
(c) Onitsuka, K.; Okuda, H.; Sasai, H. Angew. Chem., Int. Ed. 2006, 47,
1454–1457. (d) Achard, M.; Derrien, N.; Zhang, H.-J.; Demerseman, B.;
Bruneau, C. Org. Lett. 2009, 11, 185–188.
(8) (a) Kondo, T.; Morisaki, Y.; Uenoyama, S.; Wada, K.; Mitsudo, T.
J. Am. Chem. Soc. 1999, 121, 8657–8658. (b) Zaitsev, A. B.; Caldwell,
H. F.; Pregosin, P. S.; Veiros, L. F. Chem.sEur. J. 2009, 15, 6468–6477.
(9) Kondo, T.; Mitsudo, T. In Ruthenium in Organic Synthesis;
Murahashi, S.-I., Ed.; Wiley-VCH, Weinheim, 2004; pp 129-151, and
references cited therein.
Table 1. Optimization of Ru-Catalyzed Allylic Alkylations
(10) Trost, B. M.; Fraisse, P. L.; Ball, Z. T. Angew. Chem., Int. Ed.
2002, 41, 1059–1061.
(11) (a) Bruneau, C.; Renaud, J. L.; Demerseman, B. Chem.sEur. J.
2006, 12, 5178–5187. (b) Bruneau, C.; Renaud, J.-L.; Demerseman, B. Pure
Appl. Chem. 2008, 80, 861–871. (c) Zhang, H.-J.; Demerseman, B.; Toupet,
L.; Xi, Z.; Bruneau, C. AdV. Synth. Catal. 2008, 350, 1601–1609.
(12) (a) Hermatschweiler, R.; Fernandez, I.; Pregosin, P. S.; Watson,
E. J.; Albinati, A.; Rizzato, S.; Veiros, L. F.; Calhorda, M. J. Organome-
tallics 2005, 24, 1809–1812. (b) Hermatschweiler, R.; Fernandez, I.; Breher,
F.; Pregosin, P. S.; Veiros, L. F.; Calhorda, M. J. Angew. Chem., Int. Ed.
2005, 44, 4397–4400. (c) Fernandez, I.; Hermatschweiler, R.; Breher, F.;
Pregosin, P. S.; Veiros, L. F.; Calhorda, M. J. Angew. Chem., Int. Ed. 2006,
45, 6386–6391.
Ru cat./ yield ratio
ratio
entry subs.
X
La
[%] 7a:8a (8) anti:syn
1
2
3
2a
2a
2a
OBz
OBz
OBz
4% cat. A 36
2% cat. B 92
2% cat. B 98
2% PPh3
10:90
27:73
22:78
54:46
82:18
79:21
(13) Burger, E. C.; Tunge, J. A. Chem. Commun. 2005, 2835–2837.
(14) Kawatsura, M.; Ata, F.; Wada, S.; Hayase, S.; Uni, H.; Itoh, T.
Chem. Commun. 2007, 298–300.
4
5
6
2a
2b
2c
OBz
OAc
2% cat. C 83
2% cat. C 88
OCOOtBu 2% cat. C 81
2:98
4:96
1:99
88:12
89:11
90:10
(15) Kawatsure, M.; Ata, F.; Hayase, S.; Itoh, T. Chem. Commun. 2007,
4283–4285.
a Catalyst systems: cat. A: RuCl2(PPh3)3; cat. B: [Cp*Ru(MeCN)3]PF6;
cat. C: [(p-cymene)RuCl2]2/2 PPh3.
(16) (a) Kazmaier, U.; Zumpe, F. L. Angew. Chem., Int. Ed. Engl. 1999,
38, 1468–1470. (b) Kazmaier, U. Curr. Org. Chem. 2003, 317–328. (e)
Bauer, M.; Kazmaier, U. Recent Res. DeV. Org. Chem. 2005, 9, 49–69.
Org. Lett., Vol. 12, No. 21, 2010
4961

Table 2. Ru-Catalyzed Allylic Alkylations of Chelated Enolates 6
substrate (Table 1). A slight excess of the enolate was used to
allow a complete consumption of the allylic substrate. First,
reactions were carried out with RuCl2(PPh3)3 (cat. A) which
showed that, in principle, the reaction is possible and that the
branched product is formed preferentially, but the yield and
diastereoselectivity were low (entry 1). Therefore, we next
switched to the Cp*-Ru complex (cat. B) introduced by
Trost,10 which provided the allylation product in excellent yield
and good diastereoselectivity, although in this case the regi-
oselectivity dropped to 73% (entry 2). Addition of phosphines
such as PPh3 had only a marginal effect, both on the yield and
selectivities (entry 3).
The best results were obtained with [(p-cymene)RuCl2]2/
2 PPh3 (cat. C) which provided the branched product nearly
exclusively with high anti-selectivity and in good yield (entry
4). The yield was a little lower compared to the Cp*
complexes (cat B), probably because of a slightly lower
reactivity of the cymene complex. Other leaving groups, such
as the corresponding acetate (2b) or carbonate (2c) can be
used as well, without significant influence on the yield and
selectivity (entries 5 and 6). It should be mentioned that in
all examples investigated the (E)-configured linear product
(E)-7a was formed exclusively.
To prove the generality of these observations and to
evaluate the scope and limitations of this protocol, we
subjected a wide range of other allylic substrates 2 (including
optically active ones) to our optimized reaction conditions
(Table 2). With all optically active substrates used, a perfect
chirality transfer was observed. This clearly indicates that
no epimerization occurs under the reaction conditions used.
This is a great advantage compared to the Pd-catalyzed
processes and comparable to the Rh-catalyzed reactions.
Prolonging the side chain had no great influence on the yield
and selectivities (entries 1-3). Excellent results were
obtained with aryl-substituted substrates (entries 4 and 5)
providing the branched product nearly exclusively in almost
quantitative yield. A comparable regioselectivity was also
observed with substrates with the leaving group at a
quaternary center (entries 6 and 7). In this case, the yield
dropped to 50-60%, probably because of sterical hindrance
in the Ru-coordination step. However, it should be mentioned
that no reactions are observed with these substrates under
Rh-catalyzed conditions.
(17) (a) Kazmaier, U.; Deska, J.; Watzke, A. Angew. Chem., Int. Ed.
2006, 45, 4855–4858. (b) Deska, J.; Kazmaier, U. Angew. Chem., Int. Ed.
2007, 46, 4570–4573. (c) Deska, J.; Kazmaier, U. Chem.sEur. J. 2007,
13, 6204–6211.
As another interesting substrate, we investigated dienoate
2k (Scheme 3). This type of substrate gives mixtures of
products under Pd-catalyzed conditions because both double
bonds can form π-allyl complexes and the complexes can
easily isomerize. Therefore, in general the conjugated dienes
are formed preferentially, even with the highly reactive
chelated enolates.23 However, with the Ru catalyst also a
(18) (a) Pohlman, M.; Kazmaier, U.; Lindner, T. J. Org. Chem. 2004,
69, 6909–6912. (b) Kazmaier, U.; Lindner, T. Angew. Chem. Int Ed. 2005,
44, 3303–3306. (c) Lindner, T.; Kazmaier, U. AdV. Synth. Catal. 2005,
1687–1695.
(19) (a) Kazmaier, U.; Zumpe, F. L. Angew. Chem., Int. Ed. Engl. 2000,
39, 802–804. (b) Kazmaier, U.; Zumpe, F. L. Eur. J. Org. Chem. 2001,
4067–4076.
(20) Kazmaier, U.; Pohlman, M. Synlett 2004, 623–626.
(21) (a) Kazmaier, U.; Kra¨mer, K. J. Org. Chem. 2006, 71, 8950–8953.
(b) Kra¨mer, K.; Deska, J.; Hebach, C.; Kazmaier, U. Org. Biomol. Chem.
2009, 7, 103–110.
(22) (a) Kazmaier, U.; Stolz, D. Angew. Chem., Int. Ed. 2006, 45, 3072–
3075. (b) Stolz, D.; Kazmaier, U. Synthesis 2008, 2288–2292.
(23) (a) Basak, S.; Kazmaier, U. Org. Lett. 2008, 10, 501–504. (b) Basak,
S.; Kazmaier, U. Eur. J. Org. Chem. 2008, 4169–4177.
4962
Org. Lett., Vol. 12, No. 21, 2010

Scheme 3. Ru-Catalyzed Allylic Alkylations Using Dienoates
Table 3. Ru-Catalyzed Allylic Alkylations Using Linear
Substrates 9
high degree of regioretention was observed, combined with
an excellent yield. The conjugated product 9k was formed
only in 14% yield. The unbranched, linear product (7k) was
not formed at all.
Last but not least, we focused our interest also on linear
allylic substrates 1. In general, under Pd-catalyzed conditions
these substrates give the linear substitution products pref-
erentially as an (E/Z)-mixture, with the (E)-product as the
major one. Under certain circumstances, we were able to
conserve the (Z)-olefin geometry (at least in part) in reactions
of (Z)-substrates,21 but this is still not a trivial issue and
requires careful optimizations of the reaction conditions.
Therefore, the Ru-catalyzed process would fill an important
synthetic weak spot.
On the basis of the results obtained with the benzoates,
we subjected linear (E)- and (Z)-configured substrates 1a to
the usual reaction conditions (Table 3, entries 1 and 2).
Compared to the terminal substrate 2a, the reactions were
much slower, and the yields dropped dramatically, down to
15% in the case of the (Z)-substrate (entry 2). However, out
of three possible products, only two were formed in each
reaction. The substitution product obtained by regioretention
was the major product. This clearly indicates that no
isomerization had occurred. With respect to our previous
good experience in Rh-catalyzed reactions with allylic
phosphates, we tried to increase the yields by using these
leaving groups. Indeed, the corresponding phosphates 1b
gave the required substitution product 7a in much better
yields and selectivities compared to the benzoates (entries 3
and 4). Especially the (Z)-substrate (Z)-1b provided nearly
exclusively the product (Z)-7a with complete retention of
the substitution position and olefin geometry.24 This general
trend could be confirmed with further, also functionalized,
yield
[%] (E)-7:(Z)-7:8
ratio
entry substrate
R
X
1
2
3
4
6
7
(E)-1a
(Z)-1a
(E)-1b
(Z)-1b
(Z)-1l
Me
Me
Me
Me
OBz
OBz
32
15
93
85
95
98
68:0:32
0:84:16
80:0:20
1:98:1
0:98:2
0:99:1
OPO(OEt)2
OPO(OEt)2
CH2OPMP OPO(OEt)2
CH2OBn OPO(OEt)2
(Z)-1m
examples, such as the p-methoxyphenyl (pmp) ether 1l or
the benzylated derivative 1m (entries 5 and 6).
In conclusion, we could show that Ru-catalyzed allylic
alkylations are a highly interesting alternative to Pd- or Rh-
catalyzed reactions. The investigated Ru complexes show a
high tendency for regioretention, especially with (Z)-
configured substrates, which are critical candidates in Pd-
catalyzed reactions. The allyl intermediates formed do not
undergo isomerization, comparable to Rh complexes, but the
Ru complexes show a larger substrate spectrum. Further
investigations with other allylic substrates as well as synthetic
applications are currently underway.
Acknowledgment. This work was supported by the
Deutsche Forschungsgemeinschaft (Ka 880/8-2) and by the
Fonds der Chemischen Industrie.
Supporting Information Available: Experimental pro-
cedures as well as analytical and spectroscopic data of all
allylation products. This material is available free of charge
via the Internet at http://pubs.acs.org.
(24) All isomeric ratios and ee values were determined by GC using
the chiral column Chira-Si-L-Val.
OL102106V
Org. Lett., Vol. 12, No. 21, 2010
4963