вход по аккаунту


Ruthenium-Catalyzed ReactionsЧA Treasure Trove of Atom-Economic Transformations.

код для вставкиСкачать
B. M. Trost et al.
DOI: 10.1002/anie.200500136
Ruthenium Catalysis
Ruthenium-Catalyzed Reactions—A Treasure Trove of
Atom-Economic Transformations
Barry M. Trost,* Mathias U. Frederiksen, and Michael T. Rudd
atom economy · CC coupling ·
cycloisomerization · green
chemistry · ruthenium
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
The demand for new chemicals spanning the fields of health care to
materials science combined with the pressure to produce these
substances in an environmentally benign fashion pose great challenges
to the synthetic chemical community. The maximization of synthetic
efficiency by the conversion of simple building blocks into complex
targets remains a fundamental goal. In this context, ruthenium
complexes catalyze a number of non-metathesis conversions and allow
the rapid assembly of complex molecules with high selectivity and
atom economy. These complexes often exhibit unusual reactivity.
Careful consideration of the mechanistic underpinnings of the transformations can lead to the design of new reactions and the discovery of
new reactivity.
From the Contents
1. Introduction
2. Additions via Vinylidene
3. Additions via Allenylidene
4. Redox Isomerization
5. The Ruthenium-Catalyzed
Alkene–Alkyne Coupling
6. The Allene–Alkene Addition via
1. Introduction
The synthetic chemical community has been put under
increased pressure to produce, in an environmentally benign
fashion, the myriad of substances required by society. Thus,
green chemistry has emerged as a discipline that permeates all
aspects of synthetic chemistry. A major goal of this endeavor
must be to maximize the efficient use of raw materials and
simultaneously to minimize waste. Thus, synthetic efficiency
has to address not only selectivity (i.e. chemo-, regio-,
diastereo-, and enenatioselectivity) but also atom economy.[1, 2] That is, in chemical transformations of the general
form A + B ! C + D, if the desired product is C, then the byproduct D must be as small and innocuous as possible if no
use for it can be found. In an ideal scenario D would vanish all
together, and the general scheme would simplify to A + B !
C, with anything else needed only catalytically. In other
words, the use of simple addition reactions must be emphasized if we are to make chemical manufacturing more
Several reactions of this type have been developed, with
the Diels–Alder reaction perhaps representing the ideal
reaction in that CC bonds may be formed, not only with a
high degree of chemo-, regio-, diastereo-, and enantioselectivity, but also atom economically. Few industrial processes
make use of the exquisite power of the Diels–Alder transformation, even though it is an invaluable tool in the synthesis
of complex molecules.[3] Hydroformylation, on the other
hand, is an atom economic reaction that enjoys widespread
industrial use, but sees only limited use in typical research
laboratories.[4, 5] Finally, catalytic hydrogenation represents an
almost ideal reaction that is extensively used both industrially
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
7. Ruthenium-Catalyzed [5+2]
8. Summary and Outlook
and academically. Unfortunately, few reactions in multistep
synthetic sequences are simple addition reactions, and as a
result few types of molecular transformations are possible in
an atom-economic fashion. Expanding the synthetic toolbox
to include more addition reactions will certainly change this
picture. In this context transition-metal catalysts have occupied and continue to occupy a central role, both in improving
existing processes and, more importantly, in discovering new
ones. In this regard our research program directed at the
development of non-metathesis ruthenium-catalyzed reactions has been a particularly fruitful avenue of investigation.
A comprehensive treatment of this area goes well beyond
the scope of this Review.[6] Rather, we illustrate that by
approaching the development of reactions in a semirational
way, one can discover, indeed invent, new reactions guided by
the mechanistic underpinnings of these transformations. An
understanding of the mechanisms of these reactions continues
to provide the basis and impetus for the discovery of new
reactions, even if the proposed hypothesis does not always
turn out to be the actual mechanism! This Review is roughly
divided into sections along the lines of the proposed operating
mechanisms, with special care being taken to guide the reader
through the reasoning behind the design of the reactions.
[*] Prof. B. M. Trost, Dr. M. U. Frederiksen, Dr. M. T. Rudd
Department of Chemistry
Stanford University
Stanford, CA 94305-5080
Fax: (+ 1) 650-725-0002
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
2. Additions via Vinylidene Complexes
A number of reactions have been reported in which
alkynes were reported to be converted either thermally into
carbenes,[7, 8] or in the presence of a metal into the corresponding metal complexed carbenes.[9] There were, however,
several issues with the generation of these carbenes. First, the
thermal route [Eq. (1), path A], required extremely high
temperatures, which made the synthetic usefulness of this
method doubtful. The formation of metal vinylidenes
[Eq. (1), path B], on the other hand, proceeds under
extremely mild conditions, but unfortunately these complexes
were generally too stable to be synthetically useful.[10–16]
The work disclosed by Bruce and co-workers in which
various ruthenium vinylidene complexes (e.g. 2) were reportedly synthesized from [CpRu(PPh3)2Cl] (1) [Eq. (2); Cp =
cyclopentadienyl],[17–21] stimulated a search for how such
complexes may function as reactive intermediates in a
catalytic cycle. These complexes are remarkably stable
towards nucleophiles, for example, complex 2 must be
heated at reflux in methanol to obtain the addition product.
Moreover, higher alcohols do not react with this complex at
To overcome the inherent low reactivity of 2 towards
alcohols, the possibility of precoordinating the nucleophile
may offer a viable option. In such a scenario the double bond
of an allylic alcohol would coordinate to the metal, and hence
deliver the hydroxy group in an intramolecular fashion
(Scheme 1). Thus, the reaction of a coordinatively unsatu-
Barry M. Trost was born in Philadelphia, PA
in 1941 and studied at the University of
Pennsylvania (BA, 1962). He obtained his
PhD in 1965 at MIT. He moved to the University of Wisconsin where he was made Professor in 1969 and subsequently Vilas
Research Professor in 1982. He moved to
Stanford University in 1987 and became
Tamaki Professor of Humanities and Sciences in 1990. In addition to holding Visiting
Professorships at several universities worldwide, he has been awarded numerous prizes.
His interests span the entire field of organic
synthesis, particularly in the development of
novel methodology.
Michael T. Rudd was born in 1976 and studied chemistry at Purdue University. After
obtaining his BS in 1999, he joined the
research group of Professor Barry M. Trost at
Stanford University, where he obtained his
PhD in 2004. He is currently a medicinal
chemist at Merck Research Laboratories in
West Point, PA.
Scheme 1. Mechanistic rationale for the reconstitutive addition of
alkynes and allylic alcohols.
Mathias U. Frederiksen was born in 1975 in
Luxemburg. He studied chemistry at Mount
Allison University, Canada. After the completion of his undergraduate studies, he joined
the research group of Professor Tony Barrett
at Imperial College in London, UK, where
he obtained his PhD in 2002. After postdoctoral studies in the group of Professor
Barry M. Trost, he is now a medicinal chemist at the Novartis Institutes for Biomedical
Research in Basel, Switzerland.
rated Ru complex such as 3 with a terminal alkyne was
expected to generate the corresponding vinylidene complex 4.
Ligand exchange would form the complex 5 in which the
nucleophilic hydroxy group is proximal to the vinylidene
carbon atom. Nucleophilic attack then generates ruthenadiene 6, which may undergo sigmatropic rearrangement
followed by reductive elimination to give the b,g-unsaturated
ketone 8 via allyl complex 7.
Initial studies in this program focused on the reaction
between complex 1 and various terminal alkynes. These
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
revealed that in contrast to other aliphatic alcohols, various
allylic alcohols reacted cleanly with terminal acetylenes in the
presence of catalytic amounts of complex 1 and ammonium
hexafluorophosphate to form the desired b,g-unsaturated
ketones in good yield [Eq. (3)].[22] The reaction tolerates
branching at the allylic position, but not substituents on the
double bond, presumably due to steric factors. Mechanistic
studies lent strong support for the proposed mechanism.[23]
First, the coordination of the allylic alcohol is proposed to
take place by ligand exchange with triphenylphosphane in the
second step of the catalytic cycle. Several results support this
rationale. For example, the rate of reaction between 9 and
allyl alcohol is significantly retarded in the presence of excess
triphenylphosphane (> 1 equiv/Ru). Moreover, the use of a
bidentate ligand such as dppe (1,2-bis(diphenylphosphanyl)ethane) instead of triphenylphosphane completely shuts
down the reaction, presumably due to the reluctance of the
dppe ligand to dissociate from the ruthenium center, which in
turn precludes coordination of the olefin.
Labeling experiments cast light on other aspects of the
proposed mechanism [Eq. (4) and Eq. (5)]. Treatment of 1-
ethynylnaphthalene (12) with deuterated allylic alcohol 13 led
to addition products 14 and 15 ( 7:1). The carbon atom that
bears the allylic hydroxy group preferentially formed the new
CC bond with the terminal alkyne carbon atom [Eq. (4)].
Similarly, treatment of 12 with trans-dideuterated allylic
alcohol 16 gave the with a 11.5:1 ratio [Eq. (5)]. The retention
of the geometry of the alkene demonstrates the intervention
of a p-allyl species (e.g. 7) in the catalytic cycle, in which
isomerization is slower than reductive elimination. Morehindered allylic alcohols show complete scrambling of the
olefin geometry, thus indicating that h3~·h1~·h3 isomerization
now competes with reductive elimination.
This reaction exhibits excellent chemoselectivity and
allows the presence of acetals, esters, conjugated ketones,
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
internal alkynes, as well as alcohols to be present in the
acetylenic coupling partner.[24] Moreover, it lends itself well to
synthesis. For example rosefuran (22), the essence of rose oil,
was synthesized in 23 % overall yield in a concise six-step
sequence starting with acetylene 19 and allylic alcohol 20. In
this case, the Ru-catalyzed reaction generates the 1:1 adduct
21, which is readily converted into the target furan 22
(Scheme 2).[25]
Scheme 2. Reconstitutive condensation in the total synthesis of rosefuran (22).
Later, a second-generation catalyst system with In(OTf)3
as a halophilic cocatalyst allowed a decrease in the amounts of
allylic alcohol required for complete conversion into a nearly
equimolar ratio with respect to the alkyne coupling partner.[24, 26] The role of the cocatalyst is probably to remove
chloride ion, thus generating a coordinatively more unsaturated Ru cation.
The potential susceptibility of vinylidene complexes
towards nucleophilic attack by precoordinated nucleophiles
prompted an examination of the possibility of intercepting
such species by a tethered nucleophile [Eq. (6)]. In this
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
scenario a homopropargylic (23, n = 1) or higher alcohol (24,
n = 2) was expected to attack the vinylidene complex 25 to
form the oxacarbene species 26. An oxygen heterocycle may
then be released upon oxidative decomplexation, for example, in the form of a lactone 27.
The synthesis of Fischer oxacarbene species such as 26 has been known since the early
1970s, and systems containing five-,[27] six-,[28]
and seven-membered rings[29] have been prepared. The development of preparatively
useful methods that make use of the Fischer
oxacarbene complexes has been hampered by
the perceived difficulty of the decomplexation
step.[30] Many useful transition-metal-mediated methods for alkynol cycloisomerization have nevertheless been developed.[31]
The logical extension of this work has been the development of viable catalytic methods for alkynyl cycloisomerization-type reactions. In this regard McDonald and co-workers
developed elegant catalytic versions of their tungsten-mediated endo cycloisomerization, which provides access to
structurally diverse dihydropyrans in a highly efficient
manner [Eq. (7)].[32–36] This methodology has been extensively
utilized in the synthesis of biologically significant glycosides.[37–40]
The key to using the ability of Ru vinylidene complexes to
be trapped by a homopropargylic alcohol is the design of a
decomplexation pathway of Ru–oxacarbene complexes. One
approach considers a subsequent oxidative decomplexation
step, which would lead to g-butyrolactones (see 27, n = 1
[Eq. (6)]). Literature precedent indicated that the latter step
had been effected by strong oxidants such as cerium
ammonium nitrate[41, 42] and dimethyl dioxirane.[43] The critical
issue seemed to be the discovery of an oxidant that would be
chemoselective and would maintain the catalytic cycle. The
choice of oxidant (N-hydroxysuccinimide) and a small
electron-poor phosphine (trifurylphosphane) was critical for
achieving the catalytic reaction as shown in the transformation of substrate 30 into product 32 catalyzed by Ru complex
31 [Eq. (8)].[44]
This methodology was applied in a short asymmetric
synthesis of ()-muricatacin, a natural product with antitumor properties (Scheme 3).[45, 46] The chemoselectivity of this
oxidative cyclization of 35 for the formation of a fivepreferentially to a six-membered ring is noteworthy.
This synthetic methodology can be extended to 5hydroxy-1-alkynes. Initial forays in this direction met with
limited success owing to significant competitive exo cyclization, rather than cyclization in the desired endo mode.
Considerable experimentation revealed a most intriguing
divergence. When a 5-hydroxy-1-alkyne (e.g. 37) was exposed
to catalytic [CpRu(Ar3P)2Cl] with electron-donating phosphine ligands (e.g. tris(para-methoxyphenyl)phosphane) the
lactone 39 formed exclusively [Eq. (9)]. In contrast, the
catalyst bearing electron-withdrawing phosphine ligands (e.g.
tris(para-fluorophenyl)phosphane) led to clean conversion
into the dihydropyran 41 [Eq. (10)].[47] Slightly higher catalyst
loadings are necessary to reach complete conversion in the
former case.
This interesting dichotomy can be explained by considering the proposed mechanism (Scheme 4). The initial stages of
the catalytic cycle are common for both pathways. Thus,
vinylidene 43 is formed, and nucleophilic attack of the
hydroxy group at the vinylidene carbon atom leads to the
pivotal intermediate 44. In the case of the cycloisomerization
(Scheme 4, cycle A), the electron-deficient ligand favors
ligand exchange to form anionic intermediate 45, which
upon protonation liberates the dihydropyran 46 and Rucomplex 47. Simple ligand exchange liberates N-hydroxysuccinimidate and regenerates the catalyst 42. An electron-rich
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
The divergent reaction lends itself well to
synthetic applications. For example, iterative cycloisomerization, akin to the elegant application of
tungsten-catalyzed cycloisomerization by McDonald and Wu,[39] opened up an approach to narbosine A (51), an antiviral bistetrahydropyran natural
product isolated from Streptomyces (Scheme 5).[50]
Thus, alcohol 52 was converted into the dihydropyran 53 under the standard conditions. Stereoselective formation of the anomeric ether 54
preceded the second cycloisomerization, which
afforded the glycal 55 in very good yield. Finally,
acid-catalyzed addition of methanol followed by
deprotection gave narbosine A (51).
Scheme 3. Total synthesis of ()-muricatacin (36) by Ru-catalyzed cycloisomerization/oxidation; cod = 1,5-cyclooctadiene.
Likewise, the oxidative cyclization is also useful
for the synthesis of natural products, for example, in
the concise formal synthesis of the mosquito
pheromone 56 (Scheme 6).[51, 52] In this case, oxidative cycliphosphine is used in the oxidative cyclization (Scheme 4,
cycle B), in which protonation of the ruthenadihydropyran 44
zation of 57 under the standard conditions revealed the
is preferred and affords the cationic complex 48. Attack at the
hydroxylactone 58 (64 % yield), an intermediate in the
anomeric center by the nucleophilic oxidant gives complex 49,
synthesis by Kotsuki et al.[53]
followed by protonation to liberate the lactone 50 and
The marine ladder toxins represent a structurally fasciregenerate the catalyst 42. The formation of the dihydropyran
nating class of natural products.[54] With the efficient Ruis particularly interesting as some groups have reported the
catalyzed cycloisomerization in hand, an iterative approach to
attempted formation of dihydropyrans from the ruthenium
the trans-fused tetrahydropyran ring systems found in the
intermediates and noncoordinating bases.[48] In the case at
BCD ring fragment of yessotoxin,[55] a polyether toxin
hand the choice of a coordinating base, specifically Nimplicated in diarrhetic shellfish poisoning, was envisaged
hydroxysuccinimide, is critical for turnover, as other bases
(Scheme 7). Readily available alcohol 59 was subjected to the
failed entirely to give the hydropyrans.[49]
standard cycloisomerization conditions to afford glycal 60 in
Scheme 4. Postulated mechanism for the cycloisomerization (cycle A) and oxidative cyclization (cycle B).
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
Scheme 5. Iterative cycloisomerization in the total synthesis of narbosine A (51); PMB = para-methoxybenzyl; TBS = tert-butyldimethylsilyl;
Ts = para-toluenesulfonyl; CSA = 10-camphorsulfonic acid; DDQ = 2,3-dichloro-5,6-dicyano-1,4-benzoquinone.
catalytic cycles. For example, Liu and co-workers recently
described an efficient Ru-catalyzed synthesis of functionalized furans (e.g. 65) from propargylic epoxides (e.g. 66) in
very good yield [Eq. (11); Tp = tris(pyrazolyl)borate].[56]
Scheme 6. Formal total synthesis of insect pheromone 56 by a Ru-catalyzed oxidative cyclization reaction.
excellent yield. This was readily converted into the alkynyl
alcohol 61 in two steps, thus setting the stage for a second
cycloisomerization reaction, which proceeded uneventfully to
give bicycle 62. Straightforward manipulations again delivered the requisite 5-hydroxy-1-alkyne 63. A third cycloisomerization then gave tricyclic glycal 64 in excellent yield.
Alternative methods for the interception of ruthenium
vinylidene complexes by nucleophiles have also led to
Ruthenium vinylidene 67 is proposed to be the key intermediate in this reaction, and evidence for this mechanism was
supported by deuterium-labeling experiments. Similarly,
carbon nucleophiles have also been used to trap ruthenium
vinylidenes. Merlic and co-workers demonstrated that dienyl
Scheme 7. Iterative Ru-catalyzed cycloisomerization in the synthesis of trans-fused tetrahydropyrans.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
alkynes (e.g. 68) could be cyclized to the corresponding furan
(e.g. 69) in high yield in the presence of [RuCl2(p-cymene)PPh3] and NaPF6 [Eq. (12)].[57]
electron-poor and electron-rich alkynes functioned equally
well in this cross-coupling reaction.
3. Additions via Allenylidene Complexes
A ruthenium vinylidene species is also invoked as an
intermediate in the cyclization of 1-alkyne-6-allyl silanes 70 to
produce 1,4-dienes 71. Echavarren demonstrated that enyne
70 cyclized smoothly to diene 71 in the presence of the
[CpRu(PPh3)2Cl] precatalyst and NaPF6. The silane is proposed to stabilize the positive charge that develops during
ring closure [Eq. (13)].[58] Ruthenium catalysis was also used
to mediate the alkenylation of pyridine to give 2-alkenyl
pyridines 72–74 in high yield. The pyridine species is thought
to undergo ligand exchange with one phosphine ligand,
thereby placing the heterocycle proximal to the vinylidene
moiety and thus enabling regioselective alkenylation
[Eq. (14)]. Evidence for this mechanism was found by switching to a bidentate phosphine, dppe, which led to no
Ruthenium vinylidenes can also participate in carbametalation reactions, and a number of these have emerged in the
literature.[60] A particularly interesting example was revealed
by Yi and co-workers: vinylidene complex 75 was found to
catalyze the cross-coupling of tert-butyl acetylene (76) with a
variety of alkynes (e.g. 77) to give the addition products (e.g.
78 and 79) in excellent yield [Eq. (15)].[61] Notably both
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Transition-metal–allenylidene complexes have attracted a
great deal of attention in recent years owing to their
interesting structural features as well the multitude of transformations they undergo.[62, 63] Most syntheses are based on
the spontaneous dehydration of propargylic alcohols first
reported by Selegue, or on variations thereof.[64] Several
stoichiometric metal–allenylidene complexes have been described, and their chemistry has been well documented.[13]
Their use in non-metathesis catalytic applications, however,
are less developed.
It was the potential for the activation of three carbon
atoms that primarily evoked our interest in the use of Ru–
allenylidene complexes in catalytic processes. At the outset,
complexes had been utilized in stoichiometric amounts in
several interesting settings, but no catalytic applications had
been developed. The use of ruthenium–vinylidene complexes
in the reconstitutive condensation (see Section 2) encouraged
exploratory studies with allenylidene complexes. Specifically,
the use of tethered nucleophiles could potentially trap the
allenylidene complex, thereby generating a vinylidene complex that could undergo the standard reconstitutive addition,
thus giving rise to a tandem cyclization–reconstitutive addition process (Scheme 8). The allenylidene complex 80 is
expected to form readily from Ru complex 3 and the
propargylic alcohol. The tethered hydroxy group would lead
to concomitant formation of the tetrahydrofuran 81 as well as
the vinylidene complex, which would undergo a standard
reconstitutive addition (via 82–85) as described previously.
w-Hydroxypropargylic alcohols participated well to form
both five- and six-membered cyclized products in good yield
when exposed to catalytic [CpRu(Ph3P)2Cl] and NH4PF6
[Eq. (16)]). For example, diol 86 cyclized smoothly to give
bicycle 87 in very good yield (57 %).[65] Similarly, the
spiroketal subunit of calyculin A was accessed readily by
subjecting propargylic alcohol 88 to the tandem cyclization–
reconstitutive addition protocol [Eq. (17); Piv = pivaloyl][66]
to give ketone 89, a precursor for 90.
More recently, Ru–allenylidene intermediates have been
invoked in several fascinating reactions. For example, Hidai,
Uemura, and co-workers have published extensively on the
use of thiolate-bridged bimetallic ruthenium complex 91 in
several Nicholas-type addition reactions to terminal propargylic alcohols.[67] They isolated the allenylidene complex 92
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
Scheme 8. Tandem cyclization/reconstitutive addition of propargyl alcohols with allyl alcohols.
dicarbonyl compounds [Eq. (22)],[70] as well as
amides, amines, phosphine oxides,[67] phenols,[71]
and olefins.[72] Hidai, Uemura, and co-workers
expanded the scope of this reaction to include
thiols as nucleophiles as well as internal propargylic alcohols as electrophiles. In this case, a
different mechanism to that proceeding through
allenylidene complex (such as 92) has been
invoked.[73, 74]
Another notable application by this group of
researchers is the one-pot sequential Ru- and Ptcatalyzed formation of furans and pyrroles. Propargylic alcohols undergo Ru-catalyzed propargylic
substitution by a ketone followed by Pt-catalyzed
hydration/cyclization to give the desired heterocycle (e.g. 99) in good yield [Eq. (23)].[74] This
methodology is also amenable to the synthesis of
pyrroles (e.g. 100) by the addition of 1 equivalent
of aniline [Eq. (24)]. In the latter case, a higher
catalyst loading is required for complete conversions.
Notably, the synthetic chemist is not limited to
Ru catalysis, as several groups have reported
metal-catalyzed propargylic substitutions. For
example, Mahrwald and co-workers developed a
in 84 % yield by treating 91 with 1,1-ditolylpropargylic
alcohol. Complex 92, in turn, underwent substitution at the
Cg position when heated in ethanol to afford ether 93 in 89 %
[Eq. (18); Cp* = pentamethylcyclopentyl, Tol = tolyl].
The precatalyst 91 was screened in a number of different
propargylic substitution reactions and found to be a very
efficient and general catalyst system for the addition of a
number of nucleophiles, including alcohols [Eq. (19)],
ketones [Eq. (20)],[68] aromatic groups [Eq. (21)],[69] 1,3-
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
titanium-catalyzed propargylic substitution.[75] Toste and coworkers described the rhenium-catalyzed addition of alcohols,[76] allyl silanes,[77] and electron-rich aryl groups[78] to
terminal and internal alkynes in excellent yields. Finally,
Matsuda and co-workers reported the iridium-catalyzed
alkylation of propargylic esters with enoxysilanes.[79]
4. Redox Isomerization
The adjustments of oxidation levels are transformations of
fundamental importance in organic synthesis. These processes
often suffer from poor atom economy, as they frequently
require multiple steps and stoichiometric amounts of
reagents. Thus, much effort has been expended on developing
more-atom-economical ways to effect this transformation.
The transformation of unsaturated alcohols (e.g. allylic
alcohols) into their corresponding ketones or aldehydes
represents an attractive one-step alternative to traditional
reduction–oxidation procedures. Many catalyst systems based
on transition metals have been developed.[80, 81] Among these,
Ru-catalysis has occupied a central role ever since early work
by Strohmeier and Weigelt[82] and later Pascal and Dedieu[83]
demonstrated the utility of RuCl3·3H2O as a catalyst for the
isomerization of allylic alcohols (e.g. 101) into the corresponding aldehydes (e.g. 102) [Eq. (25); TFE = trifluoroethanol].
The studies on the reconstitutive addition reaction
revealed that some allylic alcohols were isomerized to
saturated ketones by certain Ru complexes. This interesting
observation warranted further examination, especially as the
complexes in question had exhibited high chemoselectivity,
which might overcome the inherent problems of chemoselectivity usually observed in transition-metal-catalyzed
isomerization reactions. Indeed, compounds that contain
isolated olefins (such as 103) [Eq. (26)] undergo clean
isomerization to yield the corresponding ketones 104 in
excellent yield simply by exposing the allylic alcohols to
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
catalytic [CpRu(Ph3P)2Cl] (1) and triethylammonium hexafluorophosphate in hot dioxane; the isolated double bond is
not affected.[84] Chemoselectivity was also demonstrated by
the compatibility with free alcohols, esters, carbonyl groups,
alkynes, and terminal alkenes.
Cyclic substrates along with sterically encumbered allylic
alcohols reacted sluggishly with the cyclopentadienyl complex. The lower rates were attributed to the slower coordination of a sterically demanding olefin to the catalyst as well
as the to the requirement that a coordination site must be
liberated for b-hydrogen elimination. The dissociation of a
phosphine group to free such a coordination site would be
disfavored in this case as phosphine–Ru bond is stronger than
the olefin–Ru bond. A decrease in the hapticity of the
cyclopentadienyl ligand from h5 to h3 would also achieve the
same result [Eq. (27)]. Thus, changing to a ligand that would
more readily undergo a switch of hapticity was expected to
enhance the reactivity in the isomerization of more-difficult
substrates.[85] Indeed, the indenyl complex 105[86–88] proved to
be much more reactive and afforded the isomerized product
108 in excellent yield in only 2 h [Eq. (28)]. However, the
increased reactivity is offset by some loss in chemoselectivity.
Scheme 9 outlines the mechanism for the redox isomerization of allylic alcohols. Initial coordination of the olefin
and the allylic hydroxy group would give rise to a complex
such as 109. Subsequent b-hydride elimination leads to the
ruthenium hydride 110. Migratory insertion gives the ruthenaenolate 111, which is readily protonated to give the
aldehyde 112 with regeneration of the catalyst 3. Deuteriumlabeling experiments support this mechanism.[89]
Ru complexes that give the critical 14-electron species 3
more readily would probably serve as more-reactive isomerization catalysts. Indeed, complexes that bear more-labile
ligands (e.g. [RuCp(PR3)(MeCN)2][PF6]) have been examined and found to be more reactive; they allow the reaction to
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
The apparent analogy of the isomerization of propargyl
alcohols into enones with that of allyl alcohols into saturated
ketones is completely misleading. Mechanistic work indicates
that an intramolecular 1,2-hydride migration is the critical
step (Scheme 10) that gives rise to the presumed vinyl–
Scheme 9. Postulated mechanism for the redox isomerization of allylic
be carried out at lower temperatures.[90] Concomitant with
increased reactivity, however, is lower chemoselectivity.
Finally, other Ru complexes, including several Ru hydride
species,[80] tetrapropylammonium perruthenate (TPAP),[91]
and the Grubbs vinylidene-metathesis catalyst[92] have been
found to catalyze the redox isomerization of allylic alcohols
The Ru-catalyzed isomerization distinguishes itself from
other isomerization reactions as it is not limited to allylic
alcohols. In fact, propargylic alcohols function equally well in
the redox isomerization to form a,b-unsaturated carbonyl
compounds. This transformation is normally effected by the
Meyer–Schuster rearrangement, in which a propargylic
hydroxy group is transposed.[93] Transition-metal-catalyzed
isomerizations based on Pd,[94–96] Rh,[97] Ir,[98] and Ru[99] have
been described. The latter employs tris(triphenylphosphane)ruthenium dichloride precatalyst in combination with a
trialkyl phosphine. This reaction did not show good reproducibility in our hands. Inspired by this and our success in the
isomerization of allylic alcohols catalyzed by [CpRu(Ph3P)2Cl] and triethylammonium hexafluorophosphate, we
applied this methodology to the isomerization of propargylic
alcohols. In the event, the indenyl catalyst 105 effected the
redox isomerization of propargylic alcohols (e.g. 113), cocatalyzed by indium(iii) chloride and sodium hexaflurophosphate. Enals 114 [Eq. (29)], and enones 116 [Eq. (30)] are
readily available by using this methodology.[100] The reaction
is highly chemoselective and allows the presence of carbonyl,
hydroxy, alkyne, and alkene groups.
Scheme 10. Redox isomerization of propargylic alcohols.
ruthenium species 118 from 117, in stark contrast to the
mechanism of the redox isomerization of allyl alcohols. The
exact role of the indium(iii) salt is not entirely clear; most
likely it functions as a halophile, and enables the formation of
the coordinatively unsaturated ruthenium cation. The mechanistic proposal has been supported by deuterium-labeling
5. The Ruthenium-Catalyzed Alkene–Alkyne
Coupling Reaction
5.1. Background
The Alder-ene reaction[101, 102] is a prime example of an
atom-economic reaction that only involves a simple addition
of an alkene 119 with allylic protons to an enophile acceptor,
which is typically another unsaturated compound such as an
alkyne 120, to give the corresponding 1,4-diene 121
[Eq. (31)].
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
In a classical sense, this reaction is promoted by heat and
is believed to take place through a supra-suprafacial endo or
exo six-electron electrocyclization.[103] Electron-withdrawing
groups on the enophile accelerate the coupling, and strong
Lewis acids accentuate this effect.[104] Conversely, electronrich alkenes can also greatly accelerate this reaction.[105–107]
Although the thermal ene reaction has found some use in
synthesis,[105, 106] the generally harsh conditions required have
limited its synthetic utility.
The expansion of the synthetic utility of this reaction
requires the development of a catalyst that could selectivity
carry out this type of transformation. It has long been known
that palladium can mediate[108] and even catalyze[109] sigmatropic reactions, and similar ene-type reactions were known
with dienes catalyzed by rhodium[110–113] and palladium.[114] A
major advance arose from discovery of the palladiumcatalyzed formal Alder-ene-type reaction of enynes
[Eq. (32)],[115, 116] the alkene–alkyne addition.
isomer in 91 % yield. In a related reaction, 1,3-dienes can also
be obtained by using ruthenium catalysis in an intermolecular
fashion from alkynes and a,b-unsaturated olefins
[Eq. (34)].[125]
Given the factors listed above, the development of a
general catalytic system that carries out the inter- and
intramolecular alkene–alkyne coupling reaction with control
of olefin geometry and regiochemistry would constitute an
important synthetic development.
5.2. Intermolecular Alkene–Alkyne Coupling Reactions To Form
For example, enyne 122 is readily cyclized to 1,4-diene 123
in 71 % yield in the presence of [(Ph3P)2Pd(OAc)2] (5 mol %).
Previous to this work, several references to the oligimerization of alkynes and alkenes had appeared,[117–119] along with
one case of allyl halide–alkyne coupling, which results in a 1,4diene.[120] Related palladium-catalyzed alkene–alkyne coupling to yield 1,3-dienes have been noted,[121] as well as
ruthenium-catalyzed alkyne–diene coupling to yield 1,3enynes.[122] The mechanism of the palladium-catalyzed
alkene–alkyne coupling reaction is clearly different from
that of the thermal Alder-ene reaction, and thus selectivity
can often be different.[123] The conditions of the catalytic
reaction are generally mild, and its utility has been demonstrated in synthesis.[116] Palladium can be quite effective for
the catalytic cycloisomerization of enynes, but, importantly,
the intermolecular alkene–alkyne coupling reaction is not
feasible under these conditions. Furthermore, 1,4- versus 1,3diene formation is often substrate-controlled.[124] For example, a 1,3-diene is generally the major product if an allylic
oxygen functionality or a tertiary carbon atom is present at
the allylic position [Eq. (33)]. Catalytic palladium rapidly
converts enyne 124 into the diene 125 as a single geometrical
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Inspired by our successful development of the Rucatalyzed reconstitutive addition reaction (see Section 2) in
which a coordinatively unsaturated Ru complex derived from
[CpRu(PPh3)2Cl] catalyzes the addition of an alkyne and an
allylic alcohol,[22] we were interested to examine the effect of
replacing the phosphine ligands with more-labile ligands,
thereby generating a coordinatively more-unsaturated
{CpRu} cationic complex that can catalyze alkene–alkyne
coupling reactions.[126] This reaction could give rise to two
regioisomeric products, a “branched” (131) and a “linear” 1,4diene (132) [Eq. (35)].
Mechanistically, two scenarios may account for the
addition reaction. First, an insertion of an alkyne into a pallyl ruthenium hydride species generated through CH
activation could lead to the coupled products 131 and 132.
Alternatively, a mechanism that involves a ruthenacyclopentene intermediate, analogously to the well-known Pauson–
Khand reaction[127] and the [2+2] addition of strained olefins
and alkynes,[128] could be involved. Both mechanisms appear
reasonable. Although the CH activation mechanism was
originally proposed,[129] it has since been found that the
ruthenacycle mechanism explains many key features of this
reaction more accurately (Scheme 11).
As alkynes are better ligands than alkenes, coordination
of two molecules of alkyne and formation of ruthenacyclopentadiene is both possible and likely. However, no product
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
Scheme 11. Mechanistic rationale for the alkene–alkyne coupling reaction. L = solvent.
formation is seen though this pathway, presumably due to the
reversible nature of the oxidative cyclization and the lack of
productive pathways leading from the ruthenacyclopentadiene. Following coordination and cyclization to either of two
ruthenium(iv) species, 136 and 137, syn b-hydride elimination
yields two different regioisomeric vinyl ruthenium species 138
and 139, which undergo reductive elimination to yield the
observed products 140 and 141 along with regeneration of the
catalyst 133. The coordination and subsequent cyclization of
the alkyne can proceed in two distinct ways: 1) Coordination
such that the R group on the alkyne is adjacent to ruthenium
(i.e. 134) leads to ruthenacyclopentene intermediate 136,
which thus results in “linear” product 140. 2) Coordination
and cyclization in the opposite regioisomeric sense (i.e. 135)
leads to “branched” product 141. Presumably the olefin may
coordinate in two different orientations, but the product
derived from ruthenacyclopentene A is not observed. The
principal reason for this is that the geometrical constraints of
the endocyclic syn b-hydride elimination step are difficult to
achieve in such a cyclic species.
To render this reaction truly synthetically useful, control
of the two regioisomeric products is necessary. Indeed, when
simple unfunctionalized monosubstituted alkenes and
alkynes are subjected to a catalytic amount of {CpRu}, the
“branched” isomer is typically favored, with ratios ranging
from 5:1 to 10:1.[129] 1-Octyne (142) is coupled to 1-octene
(143) to give “branched” 1,4-diene 144 selectively in a 5:1
ratio with “linear” 1,4-diene 145 [Eq. (36)]. Remarkably, no
homocoupling products from either alkyne 142 or alkene 143
partner are isolated.
There are several factors that explain
the observed isomeric products (i.e. 144 and
145), of which the simplest is based on steric
arguments (Scheme 12). Thus, in the path
to branched products, steric interactions
occur during the CC bond-forming event,
146, whereas in the pathway to linear
products, steric interactions occur between
the {CpRu}+ moiety and the R group on a
monosubstituted alkyne in the intermediate
138 following b-hydride elimination, rather
than in the transition state 147. Thus, the
kinetically formed ruthenacycle is 136,
whereas 137 can lead to the major product
under thermodynamically controlled conditions (i.e. when b-hydride elimination is
slow). The inherent selectivity for branched
products with simple alkenes and alkynes
can be explained by assuming that the
ruthenacyclopentene formation is fast and
reversible when both isomers 137 and 136
(Scheme 12) are accessible (as should be
the case with simple alkenes and alkynes).
Transition state 146 contains significant
steric interaction in the formation of the
ruthenacyclopentene 137. The product formation, however, is not determined until
the b-hydride-elimination step, which is slower than the
ruthenacyclopentene formation. As b-hydride elimination is
disfavored in the path towards linear products owing to the
accumulating steric strain resulting from b-elimination (i.e.
138), the branched product is favored with simple unbranched
alkenes and alkynes (Curtin–Hammet situation). However,
this inherent selectivity can be modified through various
pathways. Most simply, if the rate of b-hydride elimination is
increased relative to ruthenacycle formation, then the kinetically formed ruthenacyclopentene 136 leads to the formation
of predominantly “linear” products. Other factors that
influence the inherent selectivity are discussed in this section.
The original catalyst discovered for this reaction was
[CpRu(cod)Cl], which is a precatalyst for the presumed active
species (CpRuL2Cl or [CpRuL3]Cl; L = solvent). The typical
reaction conditions for this catalyst system involve the
ruthenium catalyst (5 mol %) in a mixture of DMF (N,Ndimethylformamide) and water at 100 8C. Before any alkene–
alkyne coupling takes place, cyclooctadiene reacts with one
molecule of alkyne in an unusual ruthenium-catalyzed
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
Scheme 12. Steric interactions in alkene–alkyne coupling.
[2+2+2] cycloaddition reaction.[130] This reaction
results in a coordinatively more-unsaturated and
thus more reactive ruthenium catalyst. The secondgeneration catalyst [CpRu(CH3CN)3]PF6[131, 132] is
generally much more active, and under typical
conditions the reaction is carried out at room
temperature in acetone or DMF. Presumably the
enhanced reactivity is due to the greater coordinative
unsaturation provided by the absence of a chloride
ion in the precatalyst and the rate of ligand exchange
of the bound acetonitrile molecules.
Scheme 13. Formal synthesis of alternaric acid (148). Fmoc = 9-fluorenylmethoxycarbonyl.
The power of this methodology to assemble
complex building blocks rapidly from simple alkenes
and alkynes was demonstrated in the formal synthesis of
alternaric acid (148)[133] (Scheme 13). In this example, the
coupling of the unbranched alkyne 149 and the highly
functionalized alkene 150 under standard conditions leads
to predominantly branched product 151 (4.9:1 ratio). Interestingly, this ratio increased to 8.9:1 when the reaction was
conducted under higher pressure.
The introduction of propargyl substituents leads to lower
amounts of branched products. For example, propargylic
oxygen substituents effectively reverse the observed regiochemistry such that linear products are favored. This result
can be rationalized by assuming that both ruthenacyclopentanes must be readily accessible in order to produce branched
products. If the rate of ruthenacyclopentene formation is
retarded significantly owing to the presence of extreme steric
thought of as a coordination effect, the fact that a noninteractions during CC bond formation (i.e. 146), then the bcoordinating propargylic oxygen functionality (OTIPS) also
hydride elimination is no longer the rate-limiting step, and the
promotes the formation of the linear product reinforces the
major product is determined by the kinetic ratio of ruthenanotion of a primarily steric effect.[129]
cyclopentene intermediates 136 and 137. An illustration of
this effect is presented in Equation (37) (Bn = benzyl, TIPS =
Nearly exclusive formation of the linear product is
triisopropylsilyl) in the coupling of differentially protected
possible when a tetrasubstituted center is adjacent to the
alcohols 152 and 153 with alkene 154.
alkyne. For example, 2-methyl-3-butyn-2-ol (157) reacts with
In this case, changing the steric bulk of the propargylic
monsubstituted olefin 158 in the presence of the secondprotecting group from benzyl (i.e. 152) to TIPS (i.e. 153) alters
generation catalyst to give the linear product 159 in excellent
the ratio of branched to linear products from 1:2 to 1:4. While
yield (91 %) in a 32:1 ratio with respect to the corresponding
the explanation for the propargylic oxygen effect could be
branched isomer [Eq. (38)].[134]
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
Not surprisingly, the choice of solvent also plays a major
role in determining the product ratios. For example, DMF
generally favors production of the branched isomer, whereas
acetone tends to produce a large fraction of linear product.
This disparate behavior is due to the greater ligating ability of
DMF relative to acetone. First, this would make the effective
size of the ruthenium center larger, thus enhancing the steric
effect present in 138 (Scheme 12). Furthermore, a more fully
ligated ruthenium center decreases the rate of b-hydride
elimination, as the latter requires an open coordination site.
Essentially, the use of DMF as solvent decreases the rate of bhydride elimination to the extent that the thermodynamic
pathway (Scheme 12) becomes the major route. This effect is
further enhanced in the presence of the chloride-free catalyst
[CpRu(CH3CN)3]PF6. For example, the reaction between 5hexynenitrile (160) and alkene 161 in acetone with this
catalyst leads to branched and linear products in a 1:1 ratio.
However, switching the solvent to DMF resulted in a much
improved ratio of 8:1 [Eq. (39)].[132]
The possibility of using disubstituted alkynes would
dramatically expand the power and scope of this method.[135]
The selectivity issue becomes even more complicated with
unsymmetrical alkynes, as the regiodifferentiation with
respect to the alkyne is greatly diminished. As was seen
with monosubstituted alkynes, the selectivity is dominated by
the presence of propargylic substituents. In the case at hand,
the determining factor in product formation is not the
propargylic alcohol moiety but rather the tert-butyl group of
the alkyne 164; the linear product 165 is formed almost
exclusively [Eq. (40)].[135] This example demonstrates that a
tert-butyl group is a more powerful director than an unhindered propargylic alcohol.[135] This is plausible due to the
greater steric bulk of the tert-butyl group and again reiterates
that coordination is not a major factor in determining the
The propargylic alcohol may still be a significant directing
group in simple disubstituted alkynes. For example, when but2-yn-1-ol (166) is coupled to a monosubstituted olefin 161,
there is seemingly little steric difference between the two
propargylic positions. Still the propargylic hydroxy group
directs the CC bond formation at the alkyne carbon atom
distal to the hydroxy group and results in a 6:1 ratio in favor of
the isomer 167 [Eq. (41)].
Clearly, the steric difference between a methyl group and
a primary propargylic alcohol is minimal, and thus the
selectivity arises from the differing electronic properties.
The electron-withdrawing nature of the hydroxy group
preferentially promotes the formation of linear products.
This effect has been exploited successfully in a key step in the
total synthesis of callipeltoside A (168) (Scheme 14).[136, 137]
The functionalized propargylic ether 169 was treated with
protected alcohol 170 to give the 1,4-diene 171 as a single
regioisomer in 85 % yield, as a result of carbon–carbon bond
formation distal to the propargylic oxygen atom.
The carbon–carbon bond formation can also be directed
to the distal carbon atom to a lesser degree by homopropargylic alcohols. The ratio for simple homopropargylic alcohols
is generally approximately 4:1 in favor of linear products. If
some steric bias is included (secondary or tertiary alcohol),
then this selectivity can be raised to excellent levels
[Eq. (42)].[135] For example, cyclohexanol 172 gave almost
exclusively linear product 174 when coupled to alkene 173.
In general, the effect of simple steric and electronic
factors can be summarized as follows: In the absence of
propargylic substituents, branched isomers are favored.
However, when propargylic substituents are present, then
the steric demand at the forming CC bond leads to larger
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
Scheme 14. Total synthesis of callipeltoside A (168). Troc = trichloroethoxycarbonyl.
proportions of linear products. When proximal polar functional groups are present on the alkyne, linear products are
Yet another way that the selectivity rules can be modified
is with a change to the more sterically demanding catalyst
[Cp*Ru(CH3CN)3]PF6, in which the cyclopentadienyl ligand
has been replaced with the sterically encumbered pentamethylcyclopentadienyl (Cp*) ligand. This was elegantly illustrated in the synthesis of the proposed structure of amphidinolide A (175).[138, 139] The target molecule was rapidly assembled through an intermolecular followed by an intramolecular
alkene–alkyne coupling reaction (Scheme 15).[140]
As one would anticipate from the above discussion, with
either of the standard catalysts, the selectivity in the first
alkene–alkyne coupling between 176 and 177 tends towards
the linear product because of the propargylic oxygen substituent. The branched product 178 can be favored, however,
if a bulky catalyst is used, which disfavors alkyne substituents
near the ruthenium center (see 136). The steric interaction of
the {Cp*Ru} moiety is more severe in the potential intermediate 136; thus the pathway towards branched products
predominates. The [Cp*Ru(CH3CN)3]PF6 catalyst resulted in
a 3.5:1 ratio of branched to linear isomers (46 % yield of
isolated product, 76 % based on recovered 178). Although it is
unclear that the same factors that govern intermolecular
reaction would also be applicable to a macrocyclization, the
second alkene–alkyne coupling reaction of 179 proceeds to
favor the branched product 180 in 58 % yield; this result was
anticipated given that neither alkyne nor alkene partner has
severe bulk or functionality that could readily coordinate. The
efficiency of this macrocyclization is noteworthy, as it is
significantly more effective than either Stille cross-coupling[141] or olefin metathesis[142] in the formation of the
same ring system. More recently, the robust nature of the
ruthenium-catalyzed alkene–alkyne coupling reaction as well
as the extraordinary chemoselectivity was demonstrated in
the determination of the correct structure of amphidinolide A.[143] Nine different amphidinolide diastereomers were
synthesized by using this chemistry, finally resulting in the
identification of the natural isomer.
Scheme 15. Total synthesis of amphidinolide A (175). Fm = fluorenylmethyl; brsm = based on recovered starting material.
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
Another member of the amphidinolide family of natural
products, namely apmhidinolide P (181), was recently synthesized by using an alkene–alkyne coupling reaction
(Scheme 16). In the key step, monosubstituted olefin 182
was coupled with enyne 183 with complete selectivity for
Scheme 16. Synthesis of amphidinolide P (181).
formation of the branched product 184, which was isolated in
75 % yield. Apparently, conjugation increases the rate of the
equilibration of the ruthenacycles and thus results in a greater
selectivity for the branched product. The synthesis of
amphidinolide P (181) highlights several features of the Rucatalyzed alkene–alkyne coupling: 1) the remarkable chemoselectivity of the Ru catalyst—four double bonds are present
in the two partners, but only the monosubstituted alkene
function reacts; 2) the sensitive b-lactone is also tolerated
under the reaction conditions; 3) an enyne may participate in
the coupling reaction.
The propargylic steric effects alluded to earlier can be
utilized to an even greater extent through the use of alkynyl
silanes. The bulky silyl group on the alkyne 185 acts in a
similar fashion to the tert-butyl group, but the vinyl silane
products have further synthetic utility owing to the variety of
subsequent transformations possible, including cross-coupling,[144] ipso substitution with halides,[145] and oxidation.[146]
A wide variety of silyl groups can be used to give exclusively
products resulting from carbon–carbon bond formation distal
to the silyl group 187 [Eq. (43)].[147] This effect can be thought
of in purely steric terms, but an electronic factor could also be
contributing: Silicon is well known to stabilize positive charge
in the b position, and in this case one could visualize attack of
the olefin at this partially positive alkyne terminus (i.e. 188),
which would lead to the branched regioisomer 189.
The ability of alkynyl silanes to give exclusively branched
products opens up a wide range of potential synthetic
applications, exemplified by the formal synthesis of ()mycalamide A (190) (Scheme 17).[148] In the key step for the
construction of the pyran moiety 191, the functionalized
alkyne 192 was coupled to b,g-unsaturated alkene 193 to
afford the 1,4-diene 194 in 63 % yield as a single regioisomer.
The coupling is noteworthy as the b,g-unsaturation is
isomerized into conjugation with the ester as a result of the
reaction. In the case at hand, the trimethylsilyl group was
simply protodesilylated, thereby unmasking the delicate exo
methylene functionality present in the target molecule 190.
However, the trimethylsilyl group served two important roles
in this synthesis. First, the TMS group controlled the
regioselectivity of the coupling reaction, and secondly it also
stabilized the olefin in the desired exo position until this
sensitive functionality could be revealed later in the synthesis.
The highly selective coupling reaction with alkynyl silanes
may be combined with other metal-catalyzed processes, such
as the palladium-catalyzed cyclization of nitrogen and oxygen
nucleophiles onto p-allyl species. This tandem process could
give access to a wide variety of heterocycles in an expeditious
manner.[145] For example, the N-protected alkynylsilane 195 is
readily coupled to para-nitrophenyl-protected homoallyl
alcohol 196 to give 197 as a single regioisomer [Eq. (44);
Ns = 4-nitrophenylsulfonyl, PNP = 2,6-bis(diphenylphosphanylmethyl)pyridine, DBU = 1,8-diazabicyclo[5.4.0]undec-7ene]. The diene 197 may be added without isolation to an
asymmetric palladium catalyst (derived from ligand 198
under basic conditions) to afford the pyrrolidine 199 in
90 % yield with 91 % ee. This process also gave access to sixmembered nitrogen heterocycles as well as five- and sixmembered oxygen heterocycles.[145] In this case, the alkyne–
Scheme 17. Synthesis of ()-mycalamide A (190).
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
alkene coupling isomerizes the homoallyl ether 196 to the
allylic ether 197, thereby setting the stage for Pd-catalyzed pallyl formation and subsequent trapping with the tethered
If a chiral homopropargylic alcohol (200, Scheme 18) is
utilized as the coupling partner, either diastereomer 201 and
202 of the target heterocycles may be accessed, by simply
switching the chirality of the ligand, 198 or ent-198, used in the
second Pd-catalyzed step (Scheme 18). The ruthenium-catalyzed alkene–alkyne coupling reaction has also been coupled
by means of Prins and Friedel–Crafts cyclizations to form
various medium-sized rings in an expeditious fashion.[149] For
example, silylpropargyl alcohol 203 was coupled to allyl ether
204 with complete regioselectivity to yield acetal 205 in good
yield. Subsequently, a Sn-mediated Friedel–Crafts-type cyclization gave tricycle 206 in excellent yield [Eq. (45)].
Scheme 18. Diastereoselective formation of heterocycles.
The convergence of steric and electronic effects observed
with alkynyl silanes can also be seen when alkynoates are
used in the alkene–alkyne coupling reaction. Normally,
addition of these Michael acceptors occurs at the b-carbon
atom. This is also true for Lewis acid catalyzed ene reactions.
In contrast, the major products in the presence of a ruthenium
catalyst result from addition to the a-carbon atom. Moreover,
in the case of g-hydroxybutynoates such as 207 or 208, the
product reacts further under the conditions of the reaction to
give the unsaturated lactone 209 [Eq. (46)].[150]
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Both the ester and the hydroxy
group contribute to the observed selectivity. The ester group potentially stabilizes a partial negative charge b to
the ruthenium center in the initially
formed ruthenacyclopentene 137
(Scheme 12). Whereas a simple unhindered substrate 207 leads to moderate
levels of selectivity (2.9:1) in favor of
addition a to the ester when coupled to
olefin 161, an increase in steric bulk at
the propargylic position with two
geminal methyl groups (i.e. 208)
results in excellent levels of selectivity
This strategy was used in the synthesis and assignment of
configuration of (+)-ancepsenolide (211; Scheme 19).[150, 151]
Its expedient assembly utilized a bidirectional coupling of
propargylic alcohol 212 and diene 213 to afford the bislactone
214 in excellent yield. The latter, in turn, was transformed into
the target molecule 211 by hydrogenation. A similar alkynoate–alkene coupling was also used in the synthesis of the
squamocin E
Scheme 20).[152, 153] Alkene 216 was coupled with alkynoate
212 regioselectively to produce 217, which was readily
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
the remarkable chemoselectivity for monosubstituted olefins.
No reaction was seen at the 1,3-diene portion of 220, even
though 1,3-dienes are generally more reactive than isolated
olefins [Eq. (47)].
When the seemingly related a,b-unsaturated ynone 222 is
used in this reaction, no 1,4-diene product is isolated.
Interestingly, the half-open ruthenocene compound 223 is
the only isolable product [Eq. (48)].[156] It is unclear why there
Scheme 19. Total synthesis of (+)-ancepsenolide (211).
converted into the target molecule 215 in only four steps.
Similarly, squamocin K was produced in a single step from 217
by simple hydrogenation of the disubstituted double bond.
Another related butenolide natural product, (+)-solamin, was
also synthesized by using this methodology.[154]
Pyrrolinones such as 218 [Eq. (47)] are also readily
accessible in a two-step protocol by this method.[155] In this
case, propargylic amine 219 was coupled to triene 220, which
gave the unsaturated amine 221 in good yield. Lactamization
of 221 upon treatment with a suitable Lewis acid gave lactam
218. This example demonstrates the regioselectivity (4.2:1
ratio of a:b regioisomers) obtained with alkynoates, as well as
is such a marked difference between alkynoates and alkynones. Not only is catalytic turnover not seen, but the
regiochemistry of the coupling is reversed as well.
With the first-generation catalyst, disubstituted olefins are
unreactive; however, the cationic catalyst [CpRu(CH3CN)3]PF6 enables these substrates to react well. In a
similar fashion for the propargyl alcohols for which steric bulk
at the propargylic position is key for high regioselectivity, the
more sterically hindered 1,1-disubstituted alkenes lead predominately to linear products. The steric bulk that directs the
reactivity towards the linear product in this case comes from
the olefin partner and not from the alkyne [Eq. (49)].[157]
Unhindered alkynoate 224 is cleanly coupled with 1,1disubstituted olefin 225 to furnish the 1,4-diene 226 in 79 %
yield as a mixture of regioisomers (4.4:1 ratio).
The 1,4-diene products of alkene–alkyne coupling reaction can also be obtained in related, but less-atom-economic
reactions, for example, indium-mediated allyl halide addition
Scheme 20. Total synthesis of (+)-squamocin E (215).
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
to alkynes,[158–161] Lewis acid catalyzed addition of allyl silanes/
stannanes,[162–164] or palladium-catalyzed addition of allyl
halides to alkynes.[165]
231, exclusive formation of the linear silyl enol ether 232 is
observed [Eq. (52)].[168] This contrasting product distribution
is quite remarkable. In a further extension of this principle,
enamides can be constructed in an expeditious fashion from
allyl amides [Eq. (53); Boc = tert-butoxycarbonyl].[169] Func-
5.3. Ruthenium-Catalyzed Alkene–Alkyne Coupling Reactions
That Do Not Result in Simple 1,4-dienes
Besides all the above-mentioned reactions and applications, the principles of alkene–alkyne coupling can be applied
to several more specialized systems that give rise to products
other than 1,4-dienes. Exposure of allyl alcohols to a
phosphine-free {CpRu} catalyst system results in an alkene–
alkyne coupling reaction to give g,d-unsaturated ketones or
aldehydes.[126] When this coupling reaction is attempted with
simple alkynes, mixtures of branched and linear products
are obtained; however, the same principles discussed
above can be used to tune the reaction to give largely a
single product. An explanation for the increased production of linear isomer when allyl alcohols and ethers are
used may lie in the weaker allylic CH bond in these
substrates which should increase the rate of b-hydride
elimination, thus resulting in the formation of more
linear/kinetic product (see Section 5.2). A combination
of this effect with the tendency of tertiary propargylic alcohols
227 to promote linear product as well, leads to the generation
of a single product 228 [Eq. (50)].[126] In this case, both
electronic and steric factors lead primarily to the linear
product. Dixneuf and co-workers also developed such a
tionalized propargyl ether 233 reacts with Boc-protected allyl
amine 234 to yield carbamate 235 in 65 % yield as a mixture of
regioisomers (2.5:1). In the presence of the standard {CpRu}
catalyst used in these cases, the propargylic heteroatom
functions as the directing element for the generation of
primarily linear products. The ratio is lower mainly because
the alkyne substrate is a secondary propargylic ether.
5.4. Multicomponent Alkene–alkyne Coupling Reactions
As detailed in Section 5.2, this inherent selectivity for
linear product in the case of propargyl alcohols can be
overcome with the more sterically bulky [Cp*Ru(cod)Cl]
catalyst.[167] For example, the propargyl alcohol 229 can be
coupled with excess allyl alcohol to yield tetrahydropyran 230
in excellent yield [Eq. (51)]. In contrast, when the less
sterically encumbered [CpRu(MeCN3)3]PF6 catalyst system
is utilized in combination with a silyl-protected allyl alcohol
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
The indication that ruthenacyclopentenes are key intermediates in alkene–alkyne coupling reactions led to the
investigation of whether nucleophiles could participate in
addition reactions when no protons were available for bhydride elimination. These reactions were designed to differ
from the previously mentioned examples in that the proposed
ruthenacyclopentene would be trapped by the addition of an
external nucleophile (Scheme 21, path A) analogously to the
alkyne/carboxylate addition chemistry to form vinyl esters
developed by Dixneuf and co-workers.[170] Either water or HX
could attack the ruthenacyclopentene 236 formed from an
alkyne and a vinyl ketone to give ruthenacyclopentane
intermediate 237; b-hydride elimination and reductive elimination would then give ketone 238 or vinyl halide 239
(Scheme 21, path A).
The same products could also be produced through a cis or
trans Wacker-type addition to the alkyne 240, either by water
to produce intermediate 241 or by halides to give intermediates 242 or 243. Subsequent 1,4-addition to an acceptor alkene
(Scheme 21, path B) would then yield 238 or 239. The first of
these processes explored was a three-component addition of a
vinyl ketone, an alkyne, and water to form 1,5-diketones
[Eq. (54)].[171] 4-Pentyn-1-ol (245) is coupled with methyl
vinyl ketone (MVK) to yield 1,5-diketone 246 as the sole
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
Scheme 21. Mechanistic rationale for three-component coupling reactions.
identifiable product. This reaction works with a variety of
terminal alkynes and vinyl ketones to serve as a versatile
synthesis of 1,5-diketones. A related addition reaction with
enones and propargylic alcohols (e.g. 247) was developed to
provide a,b-unsaturated 1,5-diketone products 248 [Eq. (55)].
The absence of any branched product, which would be
anticipated at least in minor amounts, may lead one to
propose a tandem Wacker/Michael non-ruthenacycle mechanism.
During the course of the development of the intermolecular hydrative alkyne/alkene coupling, side products containing vinyl chlorides were isolated. It was postulated that these
products arose from competitive addition of chloride, which
originates from the precatalyst, to the alkyne. This observation was explored, and the process was optimized
(Scheme 22).[172, 173] This reaction works well for a variety of
mono- and disubstituted alkynes and monosubstituted vinyl
ketones to give E vinyl chlorides and Z vinyl bromides in
generally good yields. [CpRu(cod)Cl] was used as the catalyst
for optimization of the chloride addition, but [CpRu(CH3CN)3]PF6 can also be used and leads to similar yields
and E/Z selectivities.
For the chloride addition reaction, the observed isomeric
mixtures can be rationalized by assuming that both cis- and
trans-chlororuthenation are possible in a
Wacker-type mechanism. A less-polar solvent such as acetone leads predominately
to Z products—presumably due to the
lower polarization of the RuCl bond,
thus leading to more internal attack and
therefore the observed Z products. Even
though the RuCl bond is stronger than the
RuBr bond,[174] increased amounts of Z
products during bromide addition can be
explained by the weaker nucleophilicity of
bromide relative to that of chloride in
aprotic solvents.[175] As a consequence,
excellent Z/E ratios can be attained with
(Scheme 22).[176] The potential utility of
this methodology has been demonstrated
in an expeditious synthesis of various
cyclopentanoids, for example, that of tetrahydrodicranenone (252) from the vinyl ketone 253 and
alkyne 254 via vinyl bromide 255 (Scheme 23).[177, 178]
We were drawn towards the possibility of expanding the
scope of the three-component coupling to include a fourth
component. Indeed, if the proposed tandem Wacker/
Michael mechanism (Scheme 21, path B) is at play, then
the addition of a vinyl ketone to the vinyl ruthenium species
256 should lead to a ruthenium enolate 257 [Eq. (56)].
Support for this mechanism could be gained if the enolate
could be trapped with any species other than a proton, for
example, an aldehyde. In fact, when alkyne 142, halide source
(Me4NCl), vinyl ketone, and aldehyde 158 were combined
under the standard reaction conditions the corresponding keto alcohol 259 was obtained
[Eq. (57)].[179] The diastereoselectivity in the aldol
addition step is generally quite high. The main
limitation is that the inability to suppress completely
the proton-quenched three-component coupling
Scheme 22. Vinyl halides through ruthenium-catalyzed three-component coupling.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
have been thoroughly reviewed.[6] Furthermore, many metals
have been shown to promote ene-type cyclizations or the
related isomerizations to 1,3-dienes: palladium,[115] rhodium,[186, 187] iridium,[188] titanium,[189] and other ruthenium
systems.[190, 191]
In fact, the ruthenium catalyzed alkene–alkyne coupling
reactions to form 1,4-dienes can be applied very successfully
in an intramolecular sense,[192] and a wide range of substrates
function extremely well, including trisubstituted olefins 262 as
well as electron-poor alkynes, to give the corresponding
cycloisomerized product 263 in good yield [Eq. (59)]. This
Scheme 23. Synthesis of tetrahydrodicranenone B (252).
leads to slightly lower yields. The resulting syn relationship
following the aldol reaction indicates the preferential formation of the Z enolate.
More recently, it was shown that terminal alkynes can be
added to vinyl ketones in a 1,4-fashion, thus showing the
mechanistic diversity that can be obtained through the use of
various ruthenium catalysts.[180, 181] Furthermore, recent work
demonstrated that treatment of strained olefins (such as 260)
with propargyl alcohol in the presence of a ruthenium catalyst
give the corresponding cyclopropanes 261 in excellent yield
[Eq. (58)].[182] The mechanism of this reaction is proposed to
proceed via a ruthenacyclopentene, and diverges from the
known [2+2][128]and [4+2] reactions[130] because of the presence of the propargyl alcohol moiety.
5.5. Cycloisomerization of Enynes To Form 1,4-Dienes
In a test of the synthetic versatility of the alkene–alkyne
coupling reaction, the possibility of an intramolecular coupling reaction was also examined. Ruthenium has been shown
to mediate a large variety of ring-closing reactions, including
the widely used metathesis reactions,[183–185] many of which
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
method is complementary to the palladium-catalyzed reaction as selectivities are often different, and 1,4-dienes are
almost always obtained with the {CpRu} catalysts. One of the
biggest differences between the palladium- and the ruthenium-catalyzed systems is that allylic silyloxy functionality
(i.e. 264) leads to geometrically defined 1,4-dienol silyl ethers
265 [Eq. (60)],[193] in a similar manner to the intermolecular
version (see [Eq. (52)]).
When unsymmetrical trisubstituted olefins are used, two
possible b-hydride abstractions are possible which lead to two
different products. However, in this system, one pathway is
preferred to a large degree based on the original geometry of
the olefin. As illustrated in Scheme 24, E-olefin 266 undergoes cycloisomerization to 1,4-diene 267 in moderate yield
and good selectivity for formation of the trisubstituted double
bond.[192] On the other hand, Z olefin 269 produces 1,1disubstituted olefin 270 selectively.
This selectivity can be explained through analysis of
intermediates 268 and 271. In both cases, the pseudoequatorial substituent, which is determined by the starting geometry
of the olefin, has a b-hydrogen atom adjacent to the metal
such that the required overlap is easily obtained. These results
are in contrast to the thermal ene reaction in which the olefin
geometry has little effect on the allylic proton abstracted.[102]
Generally, this cycloisomerization to 1,4-dienes is effective for
the formation of five- and six-membered rings. Although the
asymmetric version of the enyne cycloisomerization is not yet
feasible with ruthenium catalysis, Zhang and co-workers have
demonstrated that rhodium in combination with chiral
phosphines can be quite effective for this task with a limited
range of substrates.[194, 195] The groups of Trost,[196] Ito,[197] and
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
rium-labeling experiments, which indicated that only protons from the cis
methyl group are incorporated in the
The hydrative alkyne vinyl ketone
reaction (see Section 5.4) can also be
carried out in an intramolecular fashion
to form five- and six-membered rings.
Various alkynes and vinyl ketones can be
used, and it is noteworthy that, again, the
use of the more active catalyst [CpRu(MeCN3)3]PF6
alkenes to be used as is required in this
(Scheme 26).[204] In the first case, ynenone 278 undergoes the hydrative reac-
Scheme 24. Regioselectivity control in the cycloisomerization of enynes.
Mikami[198–200] have also developed palladiumcatalyzed systems for asymmetric enyne cycloisomerization.[201]
When the Ru-catalyzed enyne cycloisomerization is attempted with alkynoate substrates
272 that contain a tetrasubstituted propargylic
center, an alternate reaction takes place to
[Eq. (61)].[202, 203] In general, this reaction functions well for 1,6-enynoates with a cis methyl
group and a tetrasubstituted propargylic
carbon atom. 1,7-Enynoates give “normal”
enyne cycloisomerization products (six-membered-ring 1,4-dienes), whereas 1,6-enynoates
that lack a cis methyl group do not react well or
result in “normal” five-membered-ring 1,4dienes (if a trans-1,2-olefin is used).
Scheme 25. Postulated mechanism for the formation of seven-membered rings.
tion just as in the intermolecular case to yield 1,5-diketone
279 in good yield. However, when the reaction is carried out
under anhydrous conditions, [4+2] cycloaddition takes place
to yield the pyran 280. The production of the pyran products
supports the ruthenacyle mechanism. The formation of
ruthenacyclopentene 281 followed by isomerization to the
The proposed mechanism of this reaction involves p-allyl
formation through CH activation (Scheme 25). Enynoate
272 coordinates to the ruthenium catalyst to form intermediate 274. Subsequent CH activation of the cis allylic hydrogen
atoms gives intermediate 275, which can undergo conjugate
addition to the alkynoate to produce intermediates 276 and
277 as an equilibrium mixture. To produce a cis double bond
in the seven-membered-ring product, 275 must be present as
the anti p-allyl species. The addition could also form a fivemembered ring from the opposite end of the p-allyl system;
however, this addition is only seen in the formation of sixmembered rings. The product 273 is then released following bhydride elimination from 276. No product is seen from 277
owing to the severe allylic strain that would occur after bhydride elimination. This mechanism is supported by deute-
Scheme 26. Hydrative cyclization and [4+2] cycloaddition of ynenones.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
oxygen-bound enolate (282) and reductive elimination
account for the observed product 280 (Scheme 27).
Such pyran products can sometimes also be isolated
through intermolecular hydrative cyclization. This observation suggests that all the products result from this type of
mechanism, with diketones simply arising by hydrolysis of the
initially formed pyrans. However, a different mechanism
cannot be excluded for reactions carried out under anhydrous
tion of the substrate and catalyst to control the direction of
the reaction, different ruthenium catalysts can be used to alter
the reaction pathway.
Under an atmosphere of CO and in the presence of a
dimeric ruthenium carbonyl catalyst, an enyne (e.g. 287) that
bears a terminal alkyne and an appropriately substituted
alkene undergoes an enyne-metathesis reaction to give
rearranged 1,3-dienes 288 [Eq. (64)].[207] The mechanism of
this reaction is unknown, but, analogously to the alkene–
alkyne coupling reaction, is presumed to go through a
ruthenacycle intermediate. In another related reaction, an
appropriately substituted dienyne 289 can be transformed
into a complex tetracyclic product 290, which contains two
cyclopropane units [Eq. (65)].[208]
Scheme 27. Postulated mechanism for the [4+2] cycloaddition
5.6. Related Ruthenium-Catalyzed Cycloisomerizations of
Ruthenium can also catalyze many other reactions of
enynes through ruthenacyclopentene intermediates, and
several comprehensive reviews are available.[6, 205] Some of
the more recent applications are detailed in this section. If no
b-hydride elimination pathway is available to an enyne, 1,3dienes can be formed by utilizing ruthenium hydride catalysts,[190, 206] in analogy to palladium-catalyzed chemistry.[116]
Monosubstituted enyne 283 [Eq. (62)] undergoes cycloisomerization to 1,3-diene 284 in the presence of [Cp*Ru-
(cod)Cl] in acetic acid, whereas highly functionalized ynenoate 285 is cyclized to 1,3-diene 286 when treated with a
ruthenium hydride catalyst [Eq. (63)]. Besides this modifica-
The proposed ruthenacyclopentane may be intercepted by
small, coordinating molecules such as CO; in this case, the
reaction is a ruthenium-catalyzed version of the venerable
Pauson–Khand reaction [Eq. (66); N,N-dimethylacetamide]
and affords a variety of ketones (e.g. 292).[209] Similarly,
catalytic ruthenium complexes can also promote heteroPauson–Khand reactions in which a tethered aldehyde (e.g.
293) acts as one unsaturated component to produce, after
carbonylation, unsaturated lactones 294 [Eq. (67)].[210]
5.7. Ruthenium-Catalyzed Alkyne–Alkyne and
Diyne-Coupling Reactions
Whereas Section 5.6 dealt with the coupling
of an alkene and an alkyne, Section 5.7 describes
various ruthenium-catalyzed alkyne–alkyne coupling reactions. Ruthenacyclopentenes are
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
believed to be key intermediates in the former case, whereas
the corresponding ruthenacyclopentadienes are frequently
invoked in the latter processes. The simplest example of an
alkyne–alkyne addition reaction is the metal-catalyzed dimerization of terminal alkynes 295 to produce enynes[211–213] or
1,3-dienol esters 296 when carried out in the presence of a
carboxylic acid (e.g. acetic acid) [Eq. (68)].[214, 215]
Two alkynes can also be coupled by the addition of a third
unsaturated component through a [2+2+2] cycloaddition.
Reactions of this type can be catalyzed by many different
metals.[216–219] The use of ruthenium in a wide range of
intramolecular [2+2+2] cycloadditions is also well established. The mechanism proposed for these reactions involves
the formation of a ruthenacyclopentadiene 298 from a diyne
297 [Eq. (69)], followed by cycloaddition or insertion of
another unsaturated molecule to yield, after expulsion of
ruthenium, cyclized products 299.
When all three components are alkynes, benzenoid compounds are formed.[220] A particularly
intriguing application of this concept was recently
described by Yamamoto and co-workers [Eq. (70);
Cy = cyclohexyl, dba = trans,trans-dibenzylideneacetone]:[221] To achieve complete regioselectivity in
the cycloaddition, a boronic acid 300 was used. It is
thought that a borononate ester is initially formed
in situ from 300 and 301. The boronate tethered
diyne then participates in a cycloaddition with 1-hexyne (302)
to yield bicycle 303, which was used directly in a Suzuki–
Miyaura coupling with 4-iodoacetophenone to give 304 in
63 % yield over two steps. In this example, the known[222] meta
selectivity of the {Cp*Ru} catalyst was used to yield 304 as a
single regioisomer.
The ruthenacyclopentadiene can also be trapped with
various other unsaturated molecules. If an olefin[223] or a
diene[224] is substituted for the third alkyne moiety, cyclic
dienes and trienes can be formed [Eq. (71)]. The use of diyne
305 and diene 306 is key to the isolation of cyclooctatriene 307
in good yields. Less-substituted 1,3-dienes give significant
amounts of vinyl cyclohexadiene products as well. The
mechanism of this reaction is proposed to involve insertion
of one double bond of the diene into ruthenacyclopentadiene
298 [Eq. (69)], b-hydride elimination, reductive elimination,
and conrotatory 8p-electrocyclization. A variety of heterocycles can also be accessed through this general concept. For
example, activated nitriles 308 undergo cycloaddition with
diyne 305 to yield substituted pyridines 309 [Eq. (72)].[225–228]
Simple nitriles generally do not function well in this
reaction, and require activation by an electron-withdrawing
group or adjacent coordinating nitrile. It has been proposed
that reactions with electron-deficient nitriles proceed via an
azaruthenacyclopentadiene instead of the all-carbon analogue. Heterocumulenes also participate readily to produce
pyridones 310 from diynes 311 and isocyanates 312
[Eq. (73)].[229] 2H-Thiopyran-2-imines are formed from isothiocyanates,[230] and dithiopyrones from carbon disulfide.[230]
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
Other quite different ruthenium-catalyzed alkyne–alkyne
cycloaddition reactions are also known. For example, a
versatile synthesis of bicyclic catechol derivatives proceeds
in the presence of a ruthenium carbonyl complex
[Eq. (74)].[231] In this transformation, two molecules of
carbon monoxide are incorporated to form catechol derivative 315, reportedly through the formation of a hydrido
ruthenium carbyne in situ. Along with the well-known alkene
and enyne metathesis reactions, ruthenium complexes can
also promote metathesis-type processes with diynes.
A recent example employs a tandem diyne cycloisomerization/cross-olefin metathesis to yield 1,3,5-trienes
[Eq. (75)].[232] Various diynes (e.g. 311) and olefins (e.g. 231)
may be utilized. The E/Z selectivity, however, is low
unless the olefin bears a bulky group (e.g. TBSO).
Activated carbonyl compounds participate in a
cycloaddition-type reaction to form highly unsaturated carbonyl compounds [Eq. (76)]:[233] The unsymmetrical diyne 317 undergoes coupling with diethyl
ketomalonate (318) to yield unsaturated ketone 320.
Although the mechanism is unknown, a cycloaddition
is proposed that presumably would give 2H-pyran 319;
an electrocyclic opening would then yield the product
Analogously, a range of propargylic alcohols (e.g.
321) dimerize in the presence of a {CpRu} catalyst to
produce a,b,g,d-unsaturated ketones 322 in good to
excellent yields [Eq. (77)].[234] The conditions of the
reaction have been optimized to yield exclusively the
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
a,b-Z olefin isomer, as well as a single g,d geometrical isomer
if there is branching in the alkyl substituent.
A mechanistic rationale for this unusual dimerization is
represented in Scheme 28. Initial coordination of two alkyne
molecules to the coordinatively unsaturated Ru catalyst forms
the complex 323. After cyclization to ruthenacyclopentadiene
324, a molecule of water is expelled to give carbene
intermediate 326. Water then presumably adds again to the
carbene carbon atom, thus resulting in an overall 1,4-shift of
the hydroxy group to give intermediate 327, which upon bhydride elimination and reductive elimination releases product 329 (via vinyl ruthenium species 328). The expulsion of a
molecule of water is depicted as a simple elimination, but
could also be viewed as a series of 1,2-shifts. The degree of
reactivity of the propargyl alcohols in this reaction is linked to
the leaving-group ability of the alcohol, with tertiary or
Scheme 28. Mechanistic rationale for the dimerization of propargyl alcohols.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
secondary benzylic being the best. This mechanism is
supported by the formation of the thermodynamically less
stable Z-olefin isomer, as well as by the isolation, under
different reaction conditions, of other products that should
result from alternate ruthenacyclopentadiene isomer 325.
Additionally, a vinylidene-based mechanism is unlikely given
the fact that internal alkynes also react, albeit with a lower
efficiency. The more-electron-rich and sterically encumbered
complex [Cp*Ru(CH3CN)3]PF6 may also be used in this
reaction to form unsaturated aldehyde hemiacetal products
from ruthenacyclopentadiene 325 [Eq. (78)].
When this type of reaction is carried out in DMF with the
1-ethynylcyclobutanol 333, then a different reaction manifold
is accessed to produce the 1,3-diene 334 [Eq. (79)].[235] The
mechanism of this ring expansion/dimerization is proposed to
proceed via a ruthenacyclopentadiene similar to 325
(Scheme 25), but prior to elimination, a 1,2-alkyl shift
occurs to expand the four-membered ring into a fivemembered ring (Scheme 29).
The reason for this type of reaction occurring and not the
previously mentioned elimination-type reaction (see
[Eq. (77)]) is likely related to the change in solvent as well
as the presence of the cyclobutane ring. Attempts to perform
the same transformation with alkyne 330 in DMF led to a very
sluggish reaction, which indicates that the elimination is
disfavored in this solvent. Thus, with a slow elimination, the
ring expansion can occur. Notably, higher temperatures are
required for the ring expansion/dimerization to proceed.
A completely different, but related, dimerization of
similar tertiary propargylic alcohol substrates is possible in
the presence of [CpRu(cod)Cl] catalyst in isoprene
[Eq. (80)]:[236] Tertiary propargylic alcohols (e.g. 340) dimerize
Scheme 29. Mechanistic rationale for the ring-expansion/dimerization
of acetylenylcyclobutanols.
upon addition of a carboxylic acid to form alkylidene
cyclobutenes (e.g. 341) in moderate to good yields. The
mechanism of this reaction is proposed to proceed through
ruthenium cyclobutadiene complexes, one of which (344,
Scheme 30) was isolated and characterized by X-ray crystallography. The reason for the observed difference in reactivity
is likely to be related to the use of a {Cp*Ru} complex in
combination with the very nonpolar solvent.
A similar dimerization of allenyl alcohols in the presence
of the same ruthenium catalyst to produce 1,4-dienol esters
was recently reported by Ihara and co-workers.[237] Based on
these mechanisms, we believed that only a single propargylic
alcohol should be required for a reaction to take place, and a
cross-coupling between an unfunctionalized alkyne and a
propargylic alcohol should therefore be possible. Indeed, an
intramolecular version works well to form a wide range of
five- and six-membered rings [Eq. (81)].[238] In this case, a
bicyclic product 348 is formed in very high yield in the
presence of only 1 mol % of catalyst. In contrast to the
intermolecular dimerization, even secondary and primary
propargyl alcohols function well in this cycloisomerization
The utility of this method was demonstrated in the total
synthesis of (+)-a-kainic acid (350; Scheme 31).[239] In this
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
which resulted from the addition of a molecule of water to the starting diyne 353
[Eq. (82)]. This result led us to believe that a
hydrative diyne cyclization could be possible,
thus enabling the transformation of diynes
into enones. We envisioned a mechanism for
the reaction based on the addition of water to
a ruthenacyclopentadiene/ruthenacyclopentatriene intermediate in analogy to the mechanism proposed by Kirchner and co-workers[240]
and by Dixneuf and co-workers[214, 215]
(Scheme 32).
Attack of a molecule of water on ruthenacyclopentadiene 356, or its tautomer 357,
would give carbene 358, which could, in turn,
rearrange to vinyl ruthenium species 359.
Protonation would yield dienol 360, which
would readily tautomerize to the observed
product 361. This process works for a variety
of diynes. For example, simple, symmetrical
diynes 362 are converted into unsaturated
Scheme 30. Postulated mechanism for the dimerization of propargyl alcohols upon addiketones 363 in high yield and selectivity in the
tion of carboxylic acids.
presence of only 5 mol % of catalyst
[Eq. (83)].[241]
This hydrative cyclization works well for the formation of
a broad array of five- and six-membered ring enones and is
tolerant of most functionalities, including epoxides. Even
more dramatic is the chemoselective addition of water to
unsymmetrical diynes (e.g. 364 and 366) [Eq. (84) and
Eq. (85)], almost exclusively at the least sterically hindered
alkyne. Thus, regioselective addition is possible if the steric
differentiation (i.e. the branch site) is a to the alkyne (e.g.
364) [Eq. (85)]. Excellent selectivity (~ 20:1) is realized if the
steric element is b to the alkyne (e.g. 366) [Eq. (85)]; even a
hydroxy group may act as a branch site. The fact that the
methanol-addition product was isolated when water was
Scheme 31. Synthesis of (+)-a-kainic acid (350).
replaced with methanol lends further support to the proposed
mechanism [Eq. (86)].
case, protected aminodiyne
351 underwent regioselective
cycloisomerization to unsaturated ketone 352 in high yield
and with no loss of enantiopurity at the propargylic
center. This intermediate
could be elaborated into the
neuroexcitatory natural product (+)-a-kainic acid in several
steps. The ability of the catalyst to cleave unhindered TBS
ethers is exploited to cyclize
the protected propargyl alcohol directly.
During the ruthenium-catalyzed reaction in the aforementioned synthesis of (+)kainic acid (Scheme 31), we
isolated a side product 354,
Scheme 32. Mechanistic rationale for the hydrative diyne cyclization.
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
6. The Allene–Alkene Addition via
Ruthenacyclopentene Intermediates
Remarkably, effects other than steric differentiation may
be utilized to direct attack of water to either alkyne function
in the diyne. For example, in the case of diyne 370, which
contain an alkene function conjugated to one of the alkyne
groups, water adds to the sterically most encumbered alkyne
to give ketone 372 (Scheme 33). If, on the other hand, the
olefin is replaced with a methyl group (e.g. 371), water is
added, as expected, to the sterically least hindered alkyne to
afford ketone 373. The potential utility of the chemoselective
hydrative diyne was demonstrated in the total synthesis of
several cyclindricine alkaloid natural products, for example,
cylindricine C (376; Scheme 33).[242]
Allenes constitute important building
blocks in organic synthesis.[243] Recent developments have been extensively investigated in
a wide array of transition-metal-catalyzed
reactions of allenes in intra-[244] or intermolec[245, 246]
ular fashion.
In the latter case, several elegant PdIIcatalyzed studies, in which allenes that contain several
tethered nucleophiles were coupled with other unsaturated
groups to give interesting heterocycles, were described by Ma
and co-workers.[247–250]
After our successful application of alkynes in the alkene–
alkyne coupling reactions it was tempting to pose the question
as to whether other sp-hybridized species, such as allenes, may
enter into similar mechanistic manifolds under Ru catalysis.
Such a coupling reaction would give rise to 1,3-dienes, an
important structural motif in organic synthesis.[251, 252] The
mechanistic proposal based on the alkene–alkyne reaction is
detailed in Scheme 34. The active catalyst is envisioned to be
the coordinatively unsaturated cationic ruthenium complex,
which facilitates the coordination of the two reaction partners.
Thus, initial bonding to the allenic C(sp) atom in 377 and the
double bond of the enone 378 leads to the ruthenacyclopentane 380 via 379. The stage is now set for b-hydride
elimination, which generates a 1,3-diene 381, with concomitant formation of a ruthenium hydride species. The steric
constraints of the vinylic substituent and the exo methylene
group would presumably favor the depicted conformation,
which leads to the E isomer. Reductive elimination completes
the catalytic cycle and regenerates the RuII catalyst. Importantly, a second runthenacyclopentane 383 may emanate from
the initial coordination (step 1), but only 380 is capable of
further reaction. Thus the formation of 383 is probably
reversible, which enables equilibration to 380 and leads to a
productive reaction.
The initial studies examined the reaction between allene
384 and MVK in the presence of [CpRu(cod)Cl] (10 mol %)
and hydrated cerium(iii) chloride in DMF. MVK was chosen
to avoid b-elimination. In any event, the reaction gave the
desired 1,3-diene 385 in satisfactory yield (66 %). We had
discovered earlier that the activation of the [CpRu(cod)Cl]
complex stemmed from a [2+2+2]-type cycloaddition of cod
and the alkyne substrate. We reasoned that the allene moiety
might be a poorer activator of ruthenium, and added a
catalytic amount of an alkyne “activator” 386, which gratifyingly increased the yield of the desired product 385 to 81 %
(Scheme 35).[253] The 1,3-dienes obtained in this manner are
useful intermediates; for example, treatment with maleic
anhydride and acidic methanol gave bicyclic lactone 387 with
excellent chemo- and diasteroselectivity as well as atom
The reaction is not limited to monosubstituted allenes; in
fact, polysubstituted systems may also be used in the twocomponent coupling. In these cases, regioselectivity issues
may occur, but synthetically useful ratios are attained with
suitable directing functionality, that is, either electronic or
steric factors may control the regiochemical outcome of the
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
reaction.[254] The use of a 1,1disubstituted allene, for example,
raises the question of regioselectivity in the b-hydrogen elimination, which may be influenced by
placing an “activating” functional
group proximally to the hydrogen
atom. When a 1,1-disubstituted
allene such as 388 is exposed to
MVK and the cationic catalyst
system, only the depicted
regioisomer 389 is obtained. In
this case, b-hydride elimination of
390 is greatly aided by the conjugated ester functionality, thus
leading to exclusive elimination
of HB over HA. When this functionality is reduced to the corresponding alcohol, no selectivity is
observed, leading to a 1:1 ratio of
Scheme 33. Chemoselective hydrative diyne cyclization and application in the total synthesis of cyclindriregioisomers [Eq. (87)].
cine C (376).
Steric strain is a useful way to
induce regiocontrol in trisubstituted allenes. For example, allene
391 undergoes smooth reaction to afford
regioisomer 392 in a 8.3:1 ratio with the
other regioisomer in 70 % combined yield
(Scheme 36). The origins of the regioselectivity become apparent when considering
the possible ruthenacyclopentanes emerging
in the mechanism of this reaction: Two
metallacycles 393 and 394, which are formed
reversibly, can be envisioned. The fact that
the apparently more hindered ruthenacyclopentane 393 leads to the observed major
product should not be surprising seeing that
the reaction is under Curtin–Hammett control. As b-hydride elimination, not ruthenacyclopentane formation, is the productdetermining step, the formation of 392
should be favored as b-hydride elimination
relieves strain in the spirometallacycle 394.
The latter, on the other hand, develops
considerable strain in the b-elimination step
in which a tetrasubstituted exo double bond
is formed. The synthetic utility of this
substrate is evident when exposed to Nphenylmaleimide to generate the highly
Scheme 34. Proposed mechanism for allene–alkene coupling.
functionalized polycyclic system 396 in
Scheme 35. Ru-catalyzed two-component coupling of allenes with vinyl ketones.
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
Scheme 36. Coupling of trisubstituted allenes with vinyl ketones.
excellent yield. The potential for this reaction to generate
complexity rapidly should render it useful for diversityoriented synthesis.[255, 256]
The postulated mechanism for the formation of 1,3-dienes
from the two-component coupling between allenes and vinyl
ketones suggests the intermediacy of a ruthenacyclopentane.
This ruthenacycle undergoes b-hydride elimination on the
route towards the 1,3-diene (Scheme 31, step 2). The suggested presence of the allyl ruthenium moiety enticed us to
question whether this could serve in chemical transformations
typical of other allyl metal complexes. The known susceptibility of such complexes towards nucleophilic attack suggested that a suitably tethered nucleophile might trap the
incipient Ru allyl species faster than it underwent the bhydride elimination.[257] At the time, relatively little was
known about ruthenium-catalyzed allylic substitutions.[258]
The scant precedence, nevertheless, encouraged us to explore
the feasibility of this reaction. We were pleased to discover
that tethered hydroxy groups and amines work
well for the formation of bicyclic ethers (e.g.
398) [Eq. (88)] and amines (e.g. 400), respectively [Eq. (89)]. These reactions were run in
the presence of the cationic catalyst and a
Lewis acidic cocatalyst and enabled access to a
wide range of five- and six-membered fused
heterocycles with good diastereoselectivity.
Carboxylates 401 are also competent nucleophiles in this reaction and generate structurally
diverse lactones 402 in very good yields
[Eq. (90)].[259]
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
Our working hypothesis for the mechanism is detailed in
Scheme 37. The ruthenacyclopentane 405 intermediate can
exist in an equilibrium with the p-allyl species 406. In the
absence of the tethered nucleophile, the intermediate 405
undergoes b-hydride elimination to reveal the 1,3-diene as
described earlier (see Scheme 34). In the presence of a
7. Ruthenium-Catalyzed [5+2] Cycloaddition
A common theme for many of the reactions described
herein invokes various ruthenacycles (e.g. 136) as intermediates in our working mechanisms, the fate of which has been bScheme 37. Postulated mechanism for Ru-catalyzed alkylative cyclization.
tethered nucleophile, however, the p-allyl 406 is trapped,
resulting in the ring-closed neutral species 407. Subsequent
protonation releases the cyclized product 408 and regenerates
the coordinatively unsaturated complex. Thus, product formation is critically dependent on the relative rates of bhydride elimination and nucleophilic attack. In accord with
this hypothesis, the formation of small rings should be
favored, whereas medium-sized rings should be disfavored
owing to the slower rates of cyclization. Similarly, when bhydride elimination is favored, for example, by conjugation of
the resulting double bond to an ester, no nucleophilic trapping
is observed.
Ihara and co-workers reported an interesting alkylative
ring expansion catalyzed by ruthenium that also intercepts the
proposed p-allyl intermediate.[260] In this reaction, allenocyclobutanes such as 409 undergo alkylation with an activated
olefin 410, which in turn expands presumably via 412 to form
the a-substituted cyclopentanone 411 in good to excellent
yield [Eq. (91)].
Fujiwhara et al. recently disclosed a related cross-dimerization.[261] Although an allene is not involved as a coupling
partner, it is nevertheless proposed to proceed via a ruthenacyclopentane intermediate. In the event, Fujiwhara et al.
found that when isoprene and vinyl acetate was warmed in
methanol in the presence of [CpRu(cod)Cl], the coupled
products 413 and 414 were formed in 95 % overall yield (96:4
ratio) [Eq. (92)].
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
hydride elimination to give the desired products. Spurred by
our desire to extend our understanding of these intermediates, we pondered the possibility of placing a small, strained
ring proximal to the double bond (e.g. 415), analogous to the
highly successful rhodium-catalyzed system pioneered by
Wender and co-workers.[262–281] The key question in this
mechanistic scenario would be whether the presence of a
small, strained ring would lead to cleavage of the cyclopropyl
moiety rather than to b-hydride elimination. If the latter
could be suppressed, intermediates such as 415 should prove
to be useful in higher-order cycloaddition reactions, and as
such would provide an attractive alternative to the aforementioned Rh-catalyzed cycloadditions championed by
Wender and co-workers.
Thus, inspired by the mechanistic underpinnings of the
Ru-catalyzed alkyne–alkene coupling reaction and the Rhcatalyzed [5+2] cycloaddition, we proposed the mechanism
illustrated below (Scheme 38). Initial complexation (see 417)
of the cationic ruthenium complex 416 and the vinyl cyclopropane leads to the ruthenacyclopentene 418. This cyclopropylcarbinyl intermediate was expected to undergo a
cyclopropylcarbinyl-to-homoallyl rearrangement to 419,
driven by the release of ring strain. Competitive b-elimination
is suppressed in 418, as it would necessarily lead to a highly
strained alkylidenecyclopropane species. Finally, reductive
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
Scheme 38. Proposed mechanism for Ru-catalyzed [5+2] cycloaddition.
elimination would complete the
catalytic cycle by formation of the
cycloheptadiene 420 and regeneration of the Ru catalyst 416.
Initial experiments in this
program revealed that the Rucatalyzed [5+2] cycloaddition
proceeds with extraordinary efficiency. For example, catalyzed by
[CpRu(MeCN)3]PF6 (10 mol %)
in acetone at ambient temperature, the cyclopropylenyne 421
underwent cyclization to give the hexahydroazulene 422 in
excellent yield after only 2 h [Eq. (93)].[282]
The investigation of substrates that contain more highly
substituted backbones revealed that the 1,1-disubstituted
cyclopropane 423 cyclized smoothly to give the corresponding
bicycle 424 in excellent yield as a single diastereomer
[Eq. (94)]. The reaction strongly favors the hydrogen atom
at the newly formed bridgehead stereocenter to be anti with
respect to the homoallylic hydroxy group.[283] This apparently
general trend is in agreement with the Stork–Houk–JRger
“inside-alkoxy” model.[284–286]
The examination of substrates that contain 1,2-disubstituted cyclopropanes revealed several interesting trends. For
example, treatment of the 1,2-cis-disubstituted cyclopropane 425 with the Ru catalyst under standard
conditions produced exclusively the diastereomer 426
[Eq. (95)]. Changing the group at C2 to an electronwithdrawing group (e.g. CHO; 427) reveals a complete
switch in regioselectivity [Eq. (96)] to give 428 as a single
These observations show that the regioselectivity of
the reaction of cis-disubstituted cyclopropanes may be
controlled by simply switching the electronic nature of
the substituent at C2; electron-withdrawing substituents
show a total preference for cleaving the more-substituted
cyclopropyl bond. In the case of electron-neutral or
-donating substituents a preference for the less-substituted cyclopropyl bond is observed. For 1,2-trans-disubstituted cyclopropanes poorer selectivities are generally
observed. In these cases, the best strategy for obtaining
regioselective cyclopropyl cleavage is to alter the steric
demands of the substituents at C2. Notably, the reaction is
completely diastereoselective in all the aforementioned cases.
The observed regio- and diastereoselectivities provide
circumstantial evidence for the proposed mechanism and act
as a guiding principle for applications of this chemistry. Thus,
when considering the possible ruthenacyclopentenes 429 and
430, it appears evident that the cis substituent is better
accommodated in structure 429 than in 430, in which severe
steric interactions lead to exclusive cleavage of the leasthindered cyclopropane bond. In the case of electron-withdrawing groups, the steric effect is offset by the strong
preference for the cleavage of the more-substituted (i.e.
weakest) bond. In the case of trans-cyclopropanes, it is clear
that the Rtrans substituent is easily accommodated in both
The extraordinarily mild conditions of the Ru-catalyzed
[5+2] cycloaddition, combined with the regio- and diaster-
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
eocontrol that can be readily invoked, render this methodology highly suitable for applications to the total synthesis of
bioactive natural products that contain the polyhydroazulene
motif concealed in more-complex architectures. Indeed,
complex polycyclic frameworks have been accessed by
fusing the cyclopropane to various carbocyles (e.g. 431–433)
to give the desired tricycles (e.g. 434–436) in excellent yields;
these ring systems are present in many bioactive natural
products [Eq. (97)].[288]
8. Summary and Outlook
Transition-metal catalysis provides a rich avenue of
research opportunities that address key questions of selectivity and atom economy. In this regard, the extraordinary
potential of Ru complexes to catalyze highly unusual and
powerful transformations has added valuable processes to the
synthetic toolbox. Yet one can only have a feeling that we
have barely begun to scratch the surface of the immense
possibilities afforded by Ru catalysis, particularly when
considering the vast number of ruthenium complexes that
may be synthesized combined with the incredible array of
possible oxidation states possible for ruthenium. Indeed, only
a handful of ruthenium species have been described herein,
yet an impressive number of unusual and powerful transformations have been discovered. Thus, by careful design of
novel catalysts combined with a semirational approach to
reaction design, we should be able to tap into the virtually
boundless opportunities for research provided by Ru catalysis
in particular and transition metals in general. It is therefore
with a great degree of confidence that we can assure that new
reactions and processes will be discovered that address the
goals of simultaneously decreasing the depletion of raw
materials as well as the generation of waste, while satisfying
the ever-expanding demand for sophisticated molecular
The work emanating in our laboratories was derived from an
exceptionally talented group of co-workers who are identified
individually in the references. This work was supported by the
National Science Foundation and the National Institutes of
Health, General Medical Sciences Institute. M.U.F. thanks
GlaxoSmithKline Research and Development for partial
postdoctoral fellowship support.
Received: January 13, 2005
Published online: October 5, 2005
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
[1] B. M. Trost, Science 1991, 254, 1471.
[2] B. M. Trost, Angew. Chem. 1995, 107, 285; Angew. Chem. Int.
Ed. Engl. 1995, 34, 259.
[3] K. C. Nicolaou, S. A. Snyder, T. Montagnon, G. Vassilikogiannakis, Angew. Chem. 2002, 114, 1742; Angew. Chem. Int. Ed.
2002, 41, 1668.
[4] B. Breit, W. Seiche, Synthesis 2001, 1.
[5] B. Breit, Acc. Chem. Res. 2003, 36, 264.
[6] B. M. Trost, F. D. Toste, A. B. Pinkerton, Chem. Rev. 2001, 101,
[7] B. Solaja, J. Huguet, M. Karpf, A. S. Dreiding, Tetrahedron
1987, 43, 4875.
[8] J. Ackroyd, M. Karpf, A. S. Dreiding, Helv. Chim. Acta 1984,
67, 1963.
[9] W. A. Nugent, J. M. Mayer, Metal–Ligand Multiple Bonds,
Wiley, New York, 1988.
[10] Y. Wakatsuki, Z. M. Hou, M. Tokunaga, Chem. Rec. 2003, 3,
[11] C. Menendez, D. Morales, J. Perez, V. Riera, D. Miguel,
Organometallics 2001, 20, 2775.
[12] Y. C. Lin, J. Organomet. Chem. 2001, 617, 141.
[13] V. Cadierno, M. P. Gamasa, J. Gimeno, Eur. J. Inorg. Chem.
2001, 571.
[14] B. Weyershausen, K. H. Dotz, Eur. J. Inorg. Chem. 1999, 1057.
[15] M. C. Puerta, P. Valerga, Coord. Chem. Rev. 1999, 195, 977.
[16] C. Bruneau, P. H. Dixneuf, Acc. Chem. Res. 1999, 32, 311.
[17] M. I. Bruce, R. C. Wallis, J. Organomet. Chem. 1978, 161, C1.
[18] M. I. Bruce, A. G. Swincer, R. C. Wallis, J. Organomet. Chem.
1979, 171, C5.
[19] M. I. Bruce, A. G. Swincer, Aust. J. Chem. 1980, 33, 1471.
[20] M. I. Bruce, R. C. Wallis, Aust. J. Chem. 1981, 34, 209.
[21] M. I. Bruce, F. S. Wong, B. W. Skelton, A. H. White, J. Chem.
Soc. Dalton Trans. 1982, 2203.
[22] B. M. Trost, G. Dyker, R. J. Kulawiec, J. Am. Chem. Soc. 1990,
112, 7809.
[23] B. M. Trost, R. J. Kulawiec, J. Am. Chem. Soc. 1992, 114, 5579.
[24] B. M. Trost, R. J. Kulawiec, A. Hammes, Tetrahedron Lett.
1993, 34, 587.
[25] B. M. Trost, J. A. Flygare, J. Org. Chem. 1994, 59, 1078.
[26] B. M. Trost, M. Thommen, unpublished results.
[27] M. H. Chisholm, H. C. Clark, J. Am. Chem. Soc. 1972, 94, 1532.
[28] M. I. Bruce, A. G. Swincer, B. J. Thomson, R. C. Wallis, Aust. J.
Chem. 1980, 33, 2605.
[29] C. Bianchini, A. Marchi, N. Mantovani, L. Marvelli, D. Masi, M.
Peruzzini, R. Rossi, Eur. J. Inorg. Chem. 1998, 211.
[30] G. Davies, J. P. Mcnally, A. J. Smallbridge, Adv. Organomet.
Chem. 1990, 30, 1.
[31] F. E. Mcdonald, Chem. Eur. J. 1999, 5, 3103.
[32] S. J. Danishefsky, M. T. Bilodeau, Angew. Chem. 1996, 108,
1482; Angew. Chem. Int. Ed. Engl. 1996, 35, 1380.
[33] P. H. Seeberger, M. T. Bilodeau, S. J. Danishefsky, Aldrichimica
Acta 1997, 30, 75.
[34] F. E. Mcdonald, K. S. Reddy, Y. Diaz, J. Am. Chem. Soc. 2000,
122, 4304.
[35] F. E. Mcdonald, K. S. Reddy, J. Organomet. Chem. 2001, 617,
[36] Y. H. Sheng, D. G. Musaev, K. S. Reddy, F. E. Mcdonald, K.
Morokuma, J. Am. Chem. Soc. 2002, 124, 4149.
[37] M. H. Davidson, F. E. Mcdonald, Org. Lett. 2004, 6, 1601.
[38] W. W. Cutchins, F. E. Mcdonald, Org. Lett. 2002, 4, 749.
[39] F. E. Mcdonald, M. L. Wu, Org. Lett. 2002, 4, 3979.
[40] F. E. Mcdonald, K. S. Reddy, Angew. Chem. 2001, 113, 3765;
Angew. Chem. Int. Ed. 2001, 40, 3653.
[41] P. Quayle, S. Rahman, E. L. M. Ward, Tetrahedron Lett. 1994,
35, 3801.
[42] P. Quayle, S. Rahman, Tetrahedron Lett. 1995, 36, 8087.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
[43] M. Gilbert, M. Ferrer, A.-M. Lluch, F. Sanchez-Baeza, A.
Messeguer, J. Org. Chem. 1999, 64, 1591.
[44] B. M. Trost, Y. H. Rhee, J. Am. Chem. Soc. 1999, 121, 11 680.
[45] S. H. Yoon, H. S. Moon, S. K. Hwang, S. R. Choi, S. K. Kang,
Bioorg. Med. Chem. 1998, 6, 1043.
[46] M. J. Rieser, J. F. Kozlowski, K. V. Wood, J. L. Mclaughlin,
Tetrahedron Lett. 1991, 32, 1137.
[47] B. M. Trost, Y. H. Rhee, J. Am. Chem. Soc. 2002, 124, 2528.
[48] J. L. Bowman, F. E. Mcdonald, J. Org. Chem. 1998, 63, 3680.
[49] B. M. Trost, Y. H. Rhee, unpublished results.
[50] T. Henkel, S. Breidingmack, A. Zeeck, S. Grabley, P. E.
Hammann, K. Hutter, G. Till, R. Thiericke, J. Wink, Liebigs
Ann. Chem. 1991, 575.
[51] G. Q. Lin, H. J. Xu, B. C. Wu, G. Z. Guo, W. S. Zhou,
Tetrahedron Lett. 1985, 26, 1233.
[52] K. Machiya, I. Ichimoto, M. Kirihata, H. Ueda, Agric. Biol.
Chem. 1985, 49, 643.
[53] H. Kotsuki, I. Kadota, M. Ochi, J. Org. Chem. 1990, 55, 4417.
[54] T. Yasumoto, M. Murata, Chem. Rev. 1993, 93, 1897.
[55] M. Murata, M. Kumagai, J. S. Lee, T. Yasumoto, Tetrahedron
Lett. 1987, 28, 5869.
[56] C.-Y. Lo, H. Guo, J.-J. Lian, F.-M. Shen, R.-S. Liu, J. Org. Chem.
2002, 67, 3930.
[57] C. A. Merlic, M. E. Pauly, J. Am. Chem. Soc. 1996, 118, 11 319.
[58] C. Fernandez-Rivas, M. Mendez, A. M. Echavarren, J. Am.
Chem. Soc. 2000, 122, 1221.
[59] M. Murakami, S. Hori, J. Am. Chem. Soc. 2003, 125, 4720.
[60] C. S. Yi, N. Liu, Synlett 1999, 281.
[61] C. S. Yi, N. Liu, Organometallics 1998, 17, 3158.
[62] D. Touchard, P. H. Dixneuf, Coord. Chem. Rev. 1998, 178–180,
[63] M. I. Bruce, Chem. Rev. 1998, 98, 2797.
[64] J. P. Selegue, Organometallics 1982, 1, 119.
[65] B. M. Trost, J. A. Flygare, J. Am. Chem. Soc. 1992, 114, 5476.
[66] B. M. Trost, J. A. Flygare, Tetrahedron Lett. 1994, 35, 4059.
[67] Y. Nishibayashi, I. Wakiji, M. Hidai, J. Am. Chem. Soc. 2000,
122, 11 019; Y. Nishibayashi, M. D. Milton, Y. Inada, M.
Yoshikawa, I. Wakiji, M. Hidai, S. Uemura. Chem. Eur. J.
2005, 11, 1433.
[68] Y. Nishibayashi, I. Wakiji, Y. Ishii, S. Uemura, M. Hidai, J. Am.
Chem. Soc. 2001, 123, 3393.
[69] Y. Nishibayashi, M. Yoshikawa, Y. Inada, M. Hidai, S. Uemura,
J. Am. Chem. Soc. 2002, 124, 11 846.
[70] Y. Nishibayashi, M. Yoshikawa, Y. Inada, M. Hidai, S. Uemura,
J. Org. Chem. 2004, 69, 3408.
[71] Y. Nishibayashi, Y. Inada, M. Hidai, S. Uemura, J. Am. Chem.
Soc. 2002, 124, 7900.
[72] Y. Nishibayashi, Y. Inada, M. Hidai, S. Uemura, J. Am. Chem.
Soc. 2003, 125, 6060.
[73] Y. Inada, Y. Nishibayashi, M. Hidai, S. Uemura, J. Am. Chem.
Soc. 2002, 124, 15 172.
[74] Y. Nishibayashi, Y. Inada, M. Yoshikawa, M. Hidai, S. Uemura,
Angew. Chem. 2003, 115, 1533; Angew. Chem. Int. Ed. 2003, 42,
1495. .
[75] R. Mahrwald, S. Quint, S. Scholtis, Tetrahedron 2002, 58, 9847.
[76] B. D. Sherry, A. T. Radosevich, F. D. Toste, J. Am. Chem. Soc.
2003, 125, 6076.
[77] M. R. Luzung, F. D. Toste, J. Am. Chem. Soc. 2003, 125, 15 760.
[78] J. J. Kennedy-Smith, L. A. Young, F. D. Toste, Org. Lett. 2004,
6, 1325.
[79] I. Matsuda, K. Komori, K. Itoh, J. Am. Chem. Soc. 2002, 124,
[80] R. Uma, C. Crevisy, R. Gree, Chem. Rev. 2003, 103, 27.
[81] R. C. Van Der Drift, E. Bouwman, E. Drent, J. Organomet.
Chem. 2002, 650, 1.
[82] W. Strohmeier, L. Weigelt, J. Organomet. Chem. 1975, 86, C17.
[83] M. Dedieu, Y. Pascal, J. Mol. Catal. 1980, 9, 59.
[84] B. M. Trost, R. J. Kulawiec, Tetrahedron Lett. 1991, 32, 3039.
[85] B. M. Trost, R. J. Kulawiec, J. Am. Chem. Soc. 1993, 115, 2027.
[86] L. Oro, M. A. Ciriano, M. Campo, C. Foces-Foces, F. H. Cano, J.
Organomet. Chem. 1985, 289, 117.
[87] T. B. Marder, D. C. Roe, D. Milstein, Organometallics 1988, 7,
[88] C. P. Casey, J. M. OTConnor, Organometallics 1985, 4, 384.
[89] B. M. Trost, R. J. Kulawiec, J. Am. Chem. Soc. 1993, 115, 2027.
[90] C. Slugovc, E. Ruba, R. Schmid, K. Kirchner, Organometallics
1999, 18, 4230.
[91] I. E. MarkU, A. Gautier, M. Tsukazaki, A. Llobet, E. Plantalech-Mir, C. J. Urch, S. M. Brown, Angew. Chem. 1999, 111,
2126; Angew. Chem. Int. Ed. 1999, 38, 1960.
[92] M. K. Gurjar, P. Yakambram, Tetrahedron Lett. 2001, 42, 3633.
[93] S. Swaminathan, K. V. Narayanan, Chem. Rev. 1971, 71, 429.
[94] D. Ma, X. Lu, J. Organomet. Chem. 1989, 359, 259.
[95] K. Minn, Synlett 1991, 115.
[96] T. J. J. Mueller, M. Ansorge, D. Aktah, Angew. Chem. 2000,
112, 1323; Angew. Chem. Int. Ed. 2000, 39, 1253.
[97] M. K. Eddine Saiah, R. Pellicciari, Tetrahedron Lett. 1995, 36,
[98] D. W. Ma, X. Y. Lu, Tetrahedron Lett. 1989, 30, 2109.
[99] D. Ma, X. Y. Lu, J. Chem. Soc. Chem. Commun. 1989, 890.
[100] B. M. Trost, R. C. Livingston, J. Am. Chem. Soc. 1995, 117,
[101] H. M. R. Hoffman, Angew. Chem. 1969, 81, 597; Angew. Chem.
Int. Ed. Engl. 1969, 8, 557.
[102] D. F. Taber,
Intramolecular Diels–Alder and Alder Ene
Reactions, Springer, Berlin, 1984.
[103] W. Oppolzer, V. Snieckus, Angew. Chem. 1978, 90, 506; Angew.
Chem. Int. Ed. Engl. 1978, 17, 476.
[104] B. B. Snider, Acc. Chem. Res. 1980, 13, 426.
[105] W. Oppolzer, H. F. Strauss, D. P. Simmons, Tetrahedron Lett.
1982, 23, 4673.
[106] W. Oppolzer, K. Battig, Tetrahedron Lett. 1982, 23, 4669.
[107] H. Felkin, J. D. Umpleby, Tetrahedron Lett. 1972, 13, 2285.
[108] J. C. Trebella, J. R. Olechows, H. B. Jonassen, J. Organomet.
Chem. 1966, 6, 412.
[109] L. E. Overman, F. M. Knoll, J. Am. Chem. Soc. 1980, 102, 865.
[110] A. Bright, J. K. Nicholson, J. Powell, B. L. Shaw, J. F. Malone, J.
Chem. Soc. D 1971, 712.
[111] E. Schmitz, U. Heuck, D. Habisch, J. Prakt. Chem. 1976, 318,
[112] E. Schmitz, R. Urban, U. Heuck, G. Zimmermann, E.
Grundemann, J. Prakt. Chem. 1976, 318, 185.
[113] R. Grigg, T. R. B. Mitchell, A. Ramasubbu, J. Chem. Soc.
Chem. Commun. 1980, 27.
[114] R. Grigg, T. R. B. Mitchell, A. Ramasubbu, J. Chem. Soc.
Chem. Commun. 1979, 669.
[115] B. M. Trost, M. Lautens, J. Am. Chem. Soc. 1985, 107, 1781.
[116] B. M. Trost, Janssen Chim. Acta 1991, 9, 3.
[117] T. L. Cairns, V. A. Engelhardt, H. L. Jackson, G. H. Kalb, J. C.
Sauer, J. Am. Chem. Soc. 1952, 74, 5636.
[118] D. M. Singleton, Tetrahedron Lett. 1973, 14, 1245.
[119] H. Suzuki, K. Itoh, Y. Ishii, K. Simon, J. A. Ibers, J. Am. Chem.
Soc. 1976, 98, 8494.
[120] K. Kaneda, T. Uchiyama, Y. Fujiwara, T. Imanaka, S. Teranishi,
J. Org. Chem. 1979, 44, 55.
[121] K. Kaneda, T. Uchiyama, H. Kobayashi, Y. Fujiwara, T.
Imanaka, S. Teranishi, Tetrahedron Lett. 1977, 18, 2005.
[122] T. Mitsudo, Y. Nakagawa, H. Watanabe, K. Watanabe, H.
Misawa, Y. Watanabe, J. Chem. Soc. Chem. Commun. 1981,
[123] B. M. Trost, M. Lautens, Tetrahedron Lett. 1985, 26, 4887.
[124] B. M. Trost, J. Y. L. Chung, J. Am. Chem. Soc. 1985, 107, 4586.
[125] T. Mitsudo, S. W. Zhang, M. Nagao, Y. Watanabe, J. Chem. Soc.
Chem. Commun. 1991, 598.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Ruthenium Catalysis
[126] B. M. Trost, J. A. Martinez, R. J. Kulawiec, A. F. Indolese, J.
Am. Chem. Soc. 1993, 115, 10 402.
[127] I. U. Khand, G. R. Knox, P. L. Pauson, W. E. Watts, M. I.
Foreman, J. Chem. Soc. Perkin Trans. 1 1973, 977.
[128] T. Mitsudo, K. Kokuryo, T. Shinsugi, Y. Nakagawa, Y.
Watanabe, Y. Takegami, J. Org. Chem. 1979, 44, 4492.
[129] B. M. Trost, A. Indolese, J. Am. Chem. Soc. 1993, 115, 4361.
[130] B. M. Trost, K. Imi, A. F. Indolese, J. Am. Chem. Soc. 1993, 115,
[131] B. M. Trost, C. M. Older, Organometallics 2002, 21, 2544.
[132] B. M. Trost, F. D. Toste, Tetrahedron Lett. 1999, 40, 7739.
[133] B. M. Trost, G. D. Probst, A. Schoop, J. Am. Chem. Soc. 1998,
120, 9228.
[134] B. M. Trost, A. B. Pinkerton, F. D. Toste, M. Sperrle, J. Am.
Chem. Soc. 2001, 123, 12 504.
[135] B. M. Trost, H. C. Shen, A. B. Pinkerton, Chem. Eur. J. 2002, 8,
[136] B. M. Trost, J. L. Gunzner, J. Am. Chem. Soc. 2001, 123, 9449.
[137] B. M. Trost, J. L. Gunzner, O. Dirat, Y. H. Rhee, J. Am. Chem.
Soc. 2002, 124, 10 396.
[138] J. Kobayashi, M. Ishibashi, H. J. Hirota, J. Nat. Prod. 1991, 54,
[139] J. Kobayashi, M. Ishibashi, H. Nakamura, Y. Ohizumi, T.
Yamasu, T. Sasaki, Y. Hirata, Tetrahedron Lett. 1986, 27, 5755.
[140] B. M. Trost, J. D. Chisholm, S. T. Wrobleski, M. Jung, J. Am.
Chem. Soc. 2002, 124, 12 420.
[141] H. W. Lam, G. Pattenden, Angew. Chem. 2002, 114, 4887;
Angew. Chem. Int. Ed. 2002, 41, 508.
[142] R. E. Maleczka, L. R. Terrell, F. Geng, J. S. Ward, Org. Lett.
2002, 4, 2841.
[143] B. M. Trost, Chem. Soc. Rev. 1982, 11, 141.
[144] B. M. Trost, M. R. Machacek, Z. T. Ball, Org. Lett. 2003, 5,
[145] B. M. Trost, M. R. Machacek, Angew. Chem. 2002, 114, 4887;
Angew. Chem. Int. Ed. 2002, 41, 4693.
[146] K. Tamao, M. Kumada, K. Maeda, Tetrahedron Lett. 1984, 25,
[147] B. M. Trost, M. Machacek, M. J. Schnaderbeck, Org. Lett. 2000,
2, 1761.
[148] B. M. Trost, H. Yang, G. D. Probst, J. Am. Chem. Soc. 2004, 126,
[149] F. Lopez, L. Castedo, J. L. Mascarenas, J. Am. Chem. Soc. 2002,
124, 4218.
[150] B. M. Trost, T. J. J. Muller, J. Am. Chem. Soc. 1994, 116, 4985.
[151] B. M. Trost, T. J. J. Muller, J. Martinez, J. Am. Chem. Soc. 1995,
117, 1888.
[152] B. M. Trost, T. L. Calkins, C. G. Bochet, Angew. Chem. 1997,
109, 2749; Angew. Chem. Int. Ed. Engl. 1997, 36, 2632.
[153] B. M. Trost, T. L. Calkins, Tetrahedron Lett. 1995, 36, 6021.
[154] B. M. Trost, Z. P. Shi, J. Am. Chem. Soc. 1994, 116, 7459.
[155] B. M. Trost, G. J. Roth, Org. Lett. 1999, 1, 67.
[156] F. D. Toste, PhD thesis, Stanford University, Stanford, 2001.
[157] B. M. Trost, F. D. Toste, Tetrahedron Lett. 1999, 40, 7739.
[158] B. C. Ranu, A. Majee, Chem. Commun. 1997, 1225.
[159] S. Araki, A. Imai, K. Shimizu, M. Yamada, A. Mori, Y.
Butsugan, J. Org. Chem. 1995, 60, 1841.
[160] M. M. Salter, S. Sardo-Inffiri, Synlett 2002, 2068.
[161] E. Klaps, W. Schmid, J. Org. Chem. 1999, 64, 7537.
[162] N. Asao, J. X. Liu, T. Sudoh, Y. Yamamoto, J. Org. Chem. 1996,
61, 4568.
[163] N. Asao, E. Yoshikawa, Y. Yamamoto, J. Org. Chem. 1996, 61,
[164] N. Asao, Y. Matsukawa, Y. Yamamoto, Chem. Commun. 1996,
[165] A. N. Thadani, V. H. Rawal, Org. Lett. 2002, 4, 4317.
[166] S. Derien, P. H. Dixneuf, J. Chem. Soc. Chem. Commun. 1994,
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
[167] S. Derien, L. Ropartz, J. Le Paih, P. H. Dixneuf, J. Org. Chem.
1999, 64, 3524.
[168] B. M. Trost, J. P. Surivet, F. D. Toste, J. Am. Chem. Soc. 2001,
123, 2897.
[169] B. M. Trost, J. P. Surivet, Angew. Chem. 2001, 113, 1516; Angew.
Chem. Int. Ed. 2001, 40, 1468.
[170] H. Doucet, B. Martin-Vaca, C. Bruneau, P. H. Dixneuf, J. Org.
Chem. 1995, 60, 7247.
[171] B. M. Trost, M. Portnoy, H. Kurihara, J. Am. Chem. Soc. 1997,
119, 836.
[172] B. M. Trost, A. B. Pinkerton, J. Am. Chem. Soc. 1999, 121, 1988.
[173] B. M. Trost, A. B. Pinkerton, J. Am. Chem. Soc. 2002, 124, 7376.
[174] L. Luo, C. Li, M. E. Cucullu, S. P. Nolan, Organometallics 1995,
14, 1333.
[175] F. A. Carey, R. J. Sunberg,
Advanced Organic Chemistry,
Part B, 3rd ed., Plenum, New York, 1990.
[176] B. M. Trost, A. B. Pinkerton, Angew. Chem. 2000, 112, 368;
Angew. Chem. Int. Ed. 2000, 39, 360.
[177] B. M. Trost, A. B. Pinkerton, J. Org. Chem. 2001, 66, 7714.
[178] B. M. Trost, A. B. Pinkerton, Org. Lett. 2000, 2, 1601.
[179] B. M. Trost, A. B. Pinkerton, J. Am. Chem. Soc. 2000, 122, 8081.
[180] M. Picquet, C. Bruneau, P. H. Dixneuf, Tetrahedron 1999, 55,
[181] S. Chang, Y. Na, E. Choi, S. Kim, Org. Lett. 2001, 3, 2089.
[182] H. Kikuchi, M. Uno, S. Takahashi, Chem. Lett. 1997, 1273.
[183] R. H. Grubbs, S. Chang, Tetrahedron 1998, 54, 4413.
[184] A. Furstner, Alkene Metathesis on Organic Synthesis, Springer,
New York, 1998.
[185] S. T. Diver, A. J. Giessert, Chem. Rev. 2004, 104, 1317.
[186] P. Cao, B. Wang, X. M. Zhang, J. Am. Chem. Soc. 2000, 122,
[187] K. M. Brummond, H. F. Chen, P. Sill, L. F. You, J. Am. Chem.
Soc. 2002, 124, 15 186.
[188] N. Chatani, H. Inoue, T. Morimoto, T. Muto, S. Murai, J. Org.
Chem. 2001, 66, 4433.
[189] S. J. Sturla, N. M. Kablaoui, S. L. Buchwald, J. Am. Chem. Soc.
1999, 121, 1976.
[190] J. Le Paih, D. C. Rodriguez, S. Derien, P. H. Dixneuf, Synlett
2000, 95.
[191] Y. Yamamoto, Y. Nakagai, N. Ohkoshi, K. Itoh, J. Am. Chem.
Soc. 2001, 123, 6372.
[192] B. M. Trost, F. D. Toste, J. Am. Chem. Soc. 2000, 122, 714.
[193] B. M. Trost, J. P. Surivet, F. D. Toste, J. Am. Chem. Soc. 2004,
126, 15 592.
[194] A. W. Lei, M. S. He, S. L. Wu, X. M. Zhang, Angew. Chem.
2002, 114, 3607; Angew. Chem. Int. Ed. 2002, 41, 3457.
[195] P. Cao, X. M. Zhang, Angew. Chem. 2000, 112, 4270; Angew.
Chem. Int. Ed. 2000, 39, 4104.
[196] B. M. Trost, B. A. Czeskis, Tetrahedron Lett. 1994, 35, 211.
[197] A. Goeke, M. Sawamura, R. Kuwano, Y. Ito, Angew. Chem.
1996, 108, 686; Angew. Chem. Int. Ed. Engl. 1996, 35, 662.
[198] M. Hatano, M. Terada, K. Mikami, Angew. Chem. 2001, 113,
255; Angew. Chem. Int. Ed. 2001, 40, 249.
[199] M. Hatano, M. Yamanaka, K. Mikami, Eur. J. Org. Chem. 2003,
[200] M. Hatano, K. Mikami, J. Am. Chem. Soc. 2003, 125, 4704.
[201] I. J. S. Fairlamb, Angew. Chem. 2004, 116, 255; Angew. Chem.
Int. Ed. 2004, 43, 1048.
[202] B. M. Trost, F. D. Toste, J. Am. Chem. Soc. 1999, 121, 9728.
[203] B. M. Trost, F. D. Toste, J. Am. Chem. Soc. 2002, 124, 5025.
[204] B. M. Trost, R. E. Brown, F. D. Toste, J. Am. Chem. Soc. 2000,
122, 5877.
[205] T. Naota, H. Takaya, S. I. Murahashi, Chem. Rev. 1998, 98, 2599.
[206] M. Mori, Y. Kozawa, M. Nishida, M. Kanamaru, K. Onozuka,
M. Takimoto, Org. Lett. 2000, 2, 3245.
[207] N. Chatani, T. Morimoto, T. Muto, S. Murai, J. Am. Chem. Soc.
1994, 116, 6049.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
B. M. Trost et al.
[208] N. Chatani, K. Kataoka, S. Murai, N. Furukawa, Y. Seki, J. Am.
Chem. Soc. 1998, 120, 9104.
[209] T. Kondo, N. Suzuki, T. Okada, T. Mitsudo, J. Am. Chem. Soc.
1997, 119, 6187.
[210] N. Chatani, T. Morimoto, Y. Fukumoto, S. Murai, J. Am. Chem.
Soc. 1998, 120, 5335.
[211] A. K. Dash, M. S. Eisen, Org. Lett. 2000, 2, 737.
[212] T. Ohmura, S. Yorozuya, Y. Yamamoto, N. Miyaura, Organometallics 2000, 19, 365.
[213] H. Schmitt, H. Singer, J. Organomet. Chem. 1978, 153, 165.
[214] J. Le Paih, S. Derien, P. H. Dixneuf, Chem. Commun. 1999,
[215] J. Le Paih, F. Monnier, S. Derien, P. H. Dixneuf, E. Clot, O.
Eisenstein, J. Am. Chem. Soc. 2003, 125, 11 964.
[216] D. B. Grotjahn, K. P. C. Vollhardt, J. Am. Chem. Soc. 1986, 108,
[217] M. Lautens, W. Klute, W. Tam, Chem. Rev. 1996, 96, 49.
[218] N. E. Schore, Chem. Rev. 1988, 88, 1081.
[219] S. Saito, Y. Yamamoto, Chem. Rev. 2000, 100, 2901.
[220] Y. Yamamoto, R. Ogawa, K. Itoh, Chem. Commun. 2000, 549.
[221] Y. Yamamoto, J. I. Ishii, H. Nishiyama, K. Itoh, J. Am. Chem.
Soc. 2004, 126, 3712.
[222] Y. Yamamoto, T. Arakawa, R. Ogawa, K. Itoh, J. Am. Chem.
Soc. 2003, 125, 12 143.
[223] Y. Yamamoto, H. Kitahara, R. Ogawa, K. Itoh, J. Org. Chem.
1998, 63, 9610.
[224] J. A. Varela, L. Castedo, C. Saa, Org. Lett. 2003, 5, 2841.
[225] J. A. Varela, L. Castedo, C. Saa, J. Org. Chem. 2003, 68, 8595.
[226] J. A. Varela, C. Saa, Chem. Rev. 2003, 103, 3787.
[227] Y. Yamamoto, S. Okuda, K. Itoh, Chem. Commun. 2001, 1102.
[228] Y. Yamamoto, R. Ogawa, K. Itoh, J. Am. Chem. Soc. 2001, 123,
[229] Y. Yamamoto, H. Takagishi, K. Itoh, Org. Lett. 2001, 3, 2117.
[230] Y. Yamamoto, H. Takagishi, K. Itoh, J. Am. Chem. Soc. 2002,
124, 28.
[231] N. Chatani, Y. Fukumoto, T. Ida, S. Murai, J. Am. Chem. Soc.
1993, 115, 11 614.
[232] R. Stragies, M. Schuster, S. Blechert, Chem. Commun. 1999,
[233] Y. Yamamoto, H. Takagishi, K. Itoh, J. Am. Chem. Soc. 2002,
124, 6844.
[234] B. M. Trost, M. T. Rudd, J. Am. Chem. Soc. 2001, 123, 8862.
[235] M. Yoshida, K. Sugimoto, M. Ihara, Arkivoc 2003, 8, 35.
[236] J. Le Paih, S. Derien, C. Bruneau, B. Demerseman, L. Toupet,
P. H. Dixneuf, Angew. Chem, 2001, 113, 2996; Angew. Chem.
Int. Ed. 2001, 40, 2912.
[237] M. Yoshida, T. Gotou, M. Ihara, Tetrahedron Lett. 2003, 44,
[238] B. M. Trost, M. T. Rudd, J. Am. Chem. Soc. 2002, 124, 4178.
[239] B. M. Trost, M. T. Rudd, Org. Lett. 2003, 5, 1467.
[240] K. Mauthner, K. M. Soldouzi, K. Mereiter, R. Schmid, K.
Kirchner, Organometallics 1999, 18, 4681.
[241] B. M. Trost, M. T. Rudd, J. Am. Chem. Soc. 2003, 125, 11 516.
[242] B. M. Trost, M. T. Rudd, Org. Lett. 2003, 5, 4599.
[243] H. F. Schuster, Allenes in Organic Synthesis, Wiley, New York,
[244] T. Makino, K. Itoh, J. Org. Chem. 2004, 69, 395.
[245] A. S. K. Hashmi, Angew. Chem. 2000, 112, 3737; Angew. Chem.
Int. Ed. 2000, 39, 3590.
[246] R. Zimmer, C. U. Dinesh, E. Nandanan, F. A. Khan, Chem.
Rev. 2000, 100, 3067.
[247] S. Ma, Z. Yu, Angew. Chem. 2002, 114, 1853; Angew. Chem. Int.
Ed. 2002, 41, 1775.
[248] S. Ma, W. Gao, J. Org. Chem. 2002, 67, 6104.
[249] S. Ma, W. Gao, J. Org. Chem. 2003, 68, 5943.
[250] S. Ma, Z. Yu, Chem. Eur. J. 2004, 10, 2078.
[251] W. Oppolzer in Comprehensive Organic Synthesis, Vol. 5 (Eds.:
B. M. Trost, I. Fleming), Pergamon, Oxford, 1991, p. 315.
[252] W. R. Roush in Comprehensive Organic Chemistry, Vol. 5
(Eds.: B. M. Trost, I. Fleming), Pergamon, Oxford, 1991,
p. 513.
[253] B. M. Trost, A. B. Pinkerton, J. Am. Chem. Soc. 1999, 121, 4068.
[254] B. M. Trost, A. B. Pinkerton, M. Seidel, J. Am. Chem. Soc. 2001,
123, 12 466.
[255] S. L. Schreiber, Science 2000, 287, 1964.
[256] S. L. Schreiber, Chem. Eng. News 2003, 81, 51.
[257] R. W. Bates, V. Satcharoen, Chem. Soc. Rev. 2002, 31, 12.
[258] S. K. Kang, D. Y. Kim, R. K. Hong, P. S. Ho, Synth. Commun.
1996, 26, 3225.
[259] B. M. Trost, A. McClory, A. B. Pinkerton, unpublished results.
[260] M. Yoshida, K. Sugimoto, M. Ihara, Tetrahedron Lett. 2001, 42,
[261] M. Fujiwhara, T. Nishikawa, Y. Hori, Org. Lett. 1999, 1, 1635.
[262] P. A. Wender, J. A. Love, T. J. Williams, Synlett 2003, 1295.
[263] P. A. Wender, T. M. Pedersen, M. J. C. Scanio, J. Am. Chem.
Soc. 2002, 124, 15 154.
[264] P. A. Wender, T. J. Williams, Angew. Chem. 2002, 114, 4732;
Angew. Chem. Int. Ed. 2002, 41, 4550.
[265] P. A. Wender, G. G. Gamber, R. D. Hubbard, L. Zhang, J. Am.
Chem. Soc. 2002, 124, 2876.
[266] P. A. Wender, G. G. Gamber, M. J. C. Scanio, Angew. Chem.
2001, 113, 4013; Angew. Chem. Int. Ed. 2001, 40, 3895.
[267] P. A. Wender, F. C. Bi, M. A. Brodney, F. Gosselin, Org. Lett.
2001, 3, 2105.
[268] P. A. Wender, C. M. Barzilay, A. J. Dyckman, J. Am. Chem.
Soc. 2001, 123, 179.
[269] P. A. Wender, A. G. Correa, Y. Sato, R. Sun, J. Am. Chem. Soc.
2000, 122, 7815.
[270] P. A. Wender, L. Zhang, Org. Lett. 2000, 2, 2323.
[271] P. A. Wender, A. J. Dyckman, C. O. Husfeld, M. J. C. Scanio,
Org. Lett. 2000, 2, 1609.
[272] P. A. Wender, A. J. Dyckman, Org. Lett. 1999, 1, 2089.
[273] P. A. Wender, M. Fuji, C. O. Husfeld, J. A. Love, Org. Lett.
1999, 1, 137.
[274] P. A. Wender, A. J. Dyckman, C. O. Husfeld, D. Kadereit, J. A.
Love, H. Rieck, J. Am. Chem. Soc. 1999, 121, 10 442.
[275] P. A. Wender, F. Glorius, C. O. Husfeld, E. Langkopf, J. A.
Love, J. Am. Chem. Soc. 1999, 121, 5348.
[276] P. A. Wender, H. Rieck, M. Fuji, J. Am. Chem. Soc. 1998, 120,
10 976.
[277] P. A. Wender, D. Sperandio, J. Org. Chem. 1998, 63, 4164.
[278] P. A. Wender, C. O. Husfeld, E. Langkopf, J. A. Love, N. Pleuss,
Tetrahedron 1998, 54, 7203.
[279] P. A. Wender, C. O. Husfeld, E. Langkopf, J. A. Love, J. Am.
Chem. Soc. 1998, 120, 1940.
[280] P. A. Wender, T. E. Smith, Tetrahedron 1998, 54, 1255.
[281] P. A. Wender, H. Takahashi, B. Witulski, J. Am. Chem. Soc.
1995, 117, 4720.
[282] B. M. Trost, F. D. Toste, H. Shen, J. Am. Chem. Soc. 2000, 122,
[283] B. M. Trost, H. C. Shen, T. Schulz, C. Koradin, H. Schirok, Org.
Lett. 2003, 5, 4149.
[284] J. Haller, T. Strassner, K. N. Houk, J. Am. Chem. Soc. 1997, 119,
[285] K. N. Houk, S. R. Moses, Y. D. Wu, N. G. Rondan, V. JRger, R.
Schohe, F. R. Fronczek, J. Am. Chem. Soc. 1984, 106, 3880.
[286] G. Stork, M. Kahn, Tetrahedron Lett. 1983, 24, 3951.
[287] B. M. Trost, H. C. Shen, Org. Lett. 2000, 2, 2523.
[288] B. M. Trost, H. C. Shen, Angew. Chem. 2001, 113, 2375; Angew.
Chem. Int. Ed. 2001, 40, 2313.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6630 – 6666
Без категории
Размер файла
923 Кб
transformation, treasure, atom, economic, reactionsчa, trove, ruthenium, catalyzed
Пожаловаться на содержимое документа