close

Вход

Забыли?

вход по аккаунту

?

Nickel-Catalyzed Reductive Cyclizations and Couplings.

код для вставкиСкачать
Reviews
J. Montgomery
Multicomponent Coupling Reactions
Nickel-Catalyzed Reductive Cyclizations and Couplings
John Montgomery*
Keywords:
metallacycles · multicomponent
reactions · nickel · reductive
coupling · reductive
cyclization
In memory of Norman A. LeBel
Angewandte
Chemie
3890
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200300634
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
For over 50 years, nickel catalysis has been applied in cycloaddition
processes. Nickel-catalyzed reductive couplings and cyclizations,
however, have only recently attracted a high level of interest. This
group of new reactions allows a broad range of multicomponent
couplings involving two or more p components with a main-group or
transition-metal reagent. These processes allow the assembly of
important organic substructures from widely available reaction
components. Multiple contiguous stereocenters, polycyclic ring
systems, and novel arrays of complex functionality may often be
prepared from simple, achiral, acyclic precursors. With three or more
reactive functional groups participating in the catalytic processes,
many mechanistic questions abound, including the precise timing of
bond constructions and the nature of reactive intermediates. This
Review is thus aimed at providing a critical evaluation of recent
progress in this rapidly developing field.
1. Introduction
Oligomerizations of acetylene and butadiene were among
the earliest reports of synthetically useful nickel-catalyzed
reactions, and contemporary applications of these processes
have resulted in methodological advances of substantial
importance.[1] For instance, nickel-catalyzed [4+4],[2]
[4+2],[3] and [2+2+2][4] cycloadditions have received considerable attention. Other important nickel-catalyzed processes
include olefin polymerizations,[2b, 5] dimerizations,[1a, 6] hydrocyanations,[7] and hydrometallations.[8] One particular grouping of nickel-catalyzed reactions that has received substantial
attention recently is the reductive coupling of two p components with one main-group organometallic reagent or metal
hydride (Scheme 1).[9] Within this grouping of contemporary
Scheme 1. Intermolecular nickel-catalyzed coupling of p components
in the presence of a main-group organometallic complex.
studies, many fundamental advances in reaction discovery
and complex synthetic applications have been presented. The
purpose of this Review is to highlight recent discoveries and
applications in nickel-catalyzed reductive couplings and
cyclizations as well as mechanistic hypotheses that have
grown from studies in this area.
2. General Mechanistic Considerations
Despite the considerable attention that has been devoted
to three-component couplings of two p components and a
main-group organometallic reagent or metal hydride, many
interesting mechanistic questions still remain. Of the various
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
From the Contents
1. Introduction
3891
2. General Mechanistic
Considerations
3891
3. Choice of Catalysts
3892
4. Three-Component Couplings
3892
5. Applications in the Synthesis of
Complex Molecules
3903
6. Summary and Outlook
3906
mechanistic pathways that have been
advanced, most fall into three general
groupings. These three groupings can
be characterized according to the oxidative transformation
that initiates the overall coupling process (Scheme 2). The
first general mechanism is initiated by oxidative cyclization of
nickel(0) with two p components A=B and C=D to form a
metallacycle 1. Transmetallation of a metal alkyl MR to
afford 2 followed by reductive elimination affords product 3.
The second general mechanism is initiated by oxidative
addition of nickel(0) to MR (a metal hydride or metal alkyl) to
form a reactive nickel hydride or nickel alkyl 4. Sequential
migratory insertions of the two p components C=D and A=B,
followed by reductive elimination of 5, affords product 3. The
third general mechanism is initiated by oxidative addition of
nickel (0) to one of the p components A=B, often facilitated
by a Lewis acid (M’X), to form a reactive nickel alkyl 6 (most
often a p-allyl complex). Migratory insertion of the second
p component C=D, followed by transmetallation of MR, and
finally reductive elimination affords product 3.
These are undoubtedly oversimplified descriptions, and
many variations on the three mechanism classes summarized
above are possible. Issues such as metal coordination number,
prior association of reactive components, and changes in the
hapticity of unsaturated reactive ligands provide many
reasonable variations in the mechanisms highlighted above.
Furthermore, electron-transfer processes are certainly possible and have been well documented and thoroughly studied in
other classes of nickel-catalyzed reactions.[10] The involvement of electron-transfer pathways could be important in the
individual steps of the three mechanism classes described
above. Alternatively, entirely distinct mechanisms that
involve cyclizations of free radicals, radical anions, or paramagnetic nickel intermediates are also possible. This Review
[*] Prof. J. Montgomery
Department of Chemistry
Wayne State University
Detroit, MI 48 202-3489 (USA)
Fax (+ 1) 313-577-2554
E-mail: jwm@chem.wayne.edu
DOI: 10.1002/anie.200300634
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3891
Reviews
J. Montgomery
3. Choice of Catalysts
Scheme 2. Possible mechanisms for the three-component coupling of
two p components with a main-group organometallic complex or a
main-group-metal hydride: a) oxidative cyclization of two p components; b) oxidative addition to a reducing agent and subsequent insertion of the p components; c) oxidative addition to one p component
and subsequent insertion of the second component.
primarily presents the mechanistic descriptions provided by
the authors who initially reported the work, but readers
should be aware that most mechanistic proposals in this area
are largely speculative. Further study may involve revision of
many of the mechanistic proposals summarized in this report.
The vast majority of reactions developed that fall within
the scope of this Review utilize either [Ni(cod)2] (cod = 1,5cyclooctadiene) or [Ni(acac)2] (acac = acetyl acetonate) as
the commercially available source of the active catalyst. It has
generally been proposed that the active oxidation state of
nickel in the catalytic processes is 0, and thus [Ni(cod)2]
requires no prior activation for catalysis to occur. Whereas
various complexes of nickel(0) with phosphanes and amines
(and related derivatives) may be prepared, the typical
strategy involves preparation of various catalysts in situ
simply by premixing [Ni(cod)2] with the appropriate ligand.
A description of the relative reactivities of various ligands will
be provided throughout this Review.
[Ni(acac)2] may often be used as the catalyst source,
although this nickel(ii) species typically requires reduction for
catalysis to occur. Some of the more nucleophilic reducing
agents such as dialkyl zinc reagents can reduce [Ni(acac)2] to a
lower oxidation state, but a reliable procedure that is often
employed involves prior reduction of [Ni(acac)2] with
DIBAL-H (diisobutyl aluminum hydride). This procedure
was studied by Schwartz and co-workers in the context of
nickel-catalyzed conjugate additions, and those original
studies included electrochemical evidence that DIBAL-H
promotes the formation of a paramagnetic nickel(i) species
when [Ni(acac)2] and DIBAL-H in THF are employed in a 1:1
ratio.[10a–b] Later studies by Mackenzie and Krysan demonstrated that a 1:2 ratio of [Ni(acac)2]:DIBAL-H in THF in the
presence of cod provides a convenient preparation of
[Ni(cod)2],[11] so it appears that there is little doubt that a
nickel(0) species may be produced under the appropriate
conditions. The advantages of [Ni(acac)2] over of [Ni(cod)2]
include the lower cost and air stability of [Ni(acac)2]
([Ni(cod)2] requires storage and handling under inert atmosphere). In addition to the stability and cost issues, it was
shown by Mori that DIBAL(acac), which is produced by
reduction of [Ni(acac)2] with DIBAL-H, can have a substantial impact on some reactions. Hence [Ni(acac)2]/DIBAL-H
should not be viewed as rigorously equivalent to [Ni(cod)2].[12]
4. Three-Component Couplings
4.1. Couplings of Alkenes with Alkynes
John Montgomery was born in 1965 in Concord, NC. He studied chemistry at the University of North Carolina with Prof. Joe Templeton and Prof. Maurice Brookhart (1987).
He received his PhD at Colorado State University with Prof. Louis Hegedus (1991). He
was an American Cancer Society Postdoctoral Fellow at the University of California at
Irvine (1991–1993) with Prof. Larry Overman. In 1993, he moved to Wayne State
University where he is now Professor of
Chemistry. His work is focused on transition
metals in reaction discovery, synthetic methodology development, mechanistic chemistry, and complex molecule synthesis.
3892
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
The nickel-catalyzed coupling of an alkene, an alkyne, and
a main-group organometallic reagent has been studied
intensively in a variety of contexts (Scheme 3). This process
provides an excellent way to control the stereochemistry of
Scheme 3. Intramolecular coupling of enynes with main-group organometallic reagents.
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
challenging tri- and tetrasubstituted alkenes. An apparent
requirement for the coupling or cyclization to proceed
effectively is that the alkene unit must be electron deficient.
Intramolecular versions of this process, largely developed in
our laboratories, have been demonstrated with organozinc
reagents (as in the production of 7 and 8),[13] organoaluminum
reagents,[14] and alkenyl zirconium reagents (as in the
production of 9, Scheme 4).[15] Both internal and terminal
incorporation occurs. This feature adds considerable flexibility to the method. It was proposed that nickel species 10 is an
intermediate for both alkylative and reductive manifolds, and
that the s-donating ability of the ligands controls the reaction
outcome.[13a,b]
The corresponding intermolecular couplings of enones,
alkynes, and main-group organometallic reagents also proceed efficiently to generate acyclic structures, which may be
further elaborated by a variety of procedures. The first
reported examples of couplings of this type from Ikeda et al.
involved acetylenic tin reagents to generate conjugated
enynes such as 11 and 12 (Scheme 6).[16] Both internal and
Scheme 4. Examples for syntheses through coupling of enynes and
main-group organometallic reagents. TBS = tert-butyldimethylsilyl.
Scheme 6. Examples of the first intermolecular couplings of enones,
alkynes, and main-group organometallic reagents.
alkynes are cleanly tolerated as the acetylenic component,
and enones, alkylidene malonates, nitroalkenes, and unsaturated imides are tolerated as the electron-deficient alkene
component. The scope of the reaction is very broad
(Scheme 4). Several total synthesis applications of this
reaction class have been demonstrated (Sections 5.1 and 5.2).
Organozinc reagents that bear b-hydrogen atoms are
tolerated in the process, although ligand effects become
important (Scheme 5). In the absence of phosphanes, alkylgroup transfer is observed, but when the nickel(0) catalyst is
pretreated with triphenylphosphane, selective hydrogen-atom
Scheme 5. Effect of phosphane additives on the reactivity of organozinc reagents that bear b-hydrogen atoms.
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
terminal alkynes participate in the couplings, and regioselectivities are very high with terminal alkynes to generate
products derived from coupling of the enone with the
unsubstituted alkyne terminus. Acetylenic zinc reagents
participate in the couplings, although yields and regioselectivities of alkyne insertion are lower than in the corresponding
alkynyl tin reagent couplings.[17] Alkyl zinc reagents, either in
pure form or generated from the corresponding organolithium and zinc chloride, also participate cleanly in the threecomponent couplings, as demonstrated by the production of
13.[18]
An asymmetric version of the intermolecular couplings
was developed by Ikeda et al., who used a nickel catalyst
modified with a chiral monodentate oxazoline (Scheme 7).[19]
Yields and enantioselectivities ranged from good to modest in
this procedure. A variety of bidentate ligands including
bisphosphanes, bisoxazolines, phospanyloxazolines, and pyridinyloxazolines afforded poorer results both in efficiency and
enantioselectivity.
A related process developed by Ikeda et al. involves
couplings of allylic halides and acetates with alkynes in the
presence of acetylenic stannanes (Scheme 8).[20] This procedure provides a very attractive entry to conjugated enynes.
The reactions can be entirely intermolecular (!14) or
partially intramolecular (!15). Although the process was
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3893
Reviews
J. Montgomery
nation of 17 provides the observed product 18, whereas b-H
elimination occurs when phosphanes are used (Scheme 5). By
omitting the organozinc reagent, metallacycle 19 was isolated
and fully characterized as the h1, O-bound enolate
(Scheme 10).[22] Treatment of metallacycle 19 with dimethyl-
Scheme 7. Asymmetric variants of intermolecular coupling.
Scheme 10. Intramolecular enone–alkyne coupling without organozinc
reagents. tmeda = N,N,N’,N’-tetramethylethylenediamine.
most extensively developed with acetylenic stannanes, organozinc and organoaluminum reagents are also effective
participants.[21]
The mechanistic proposals for the enone/alkyne couplings
have largely focused on the involvement of nickel metallacycles derived from the oxidative cyclization of an enone and
an alkyne with nickel(0) (Scheme 9).[9] Once the initial
oxidative cyclization occurs to produce metallacycle 16,
transmetallation of the main-group organometallic reagent
occurs to generate intermediate 17. Direct reductive elimi-
zinc affords the same product 20, which may be obtained from
the catalytic reaction of an alkynyl enal and dimethyl zinc.
This observation in no way proves that catalytic reactions
proceed by this mechanism, but the sequence does provide
direct precedent for each of the key individual steps of the
metallacycle mechanism. Studies in our laboratory are aimed
at directly investigating the kinetic competence of 19 in
catalytic reactions, and a detailed full report of that study
involving both experimental and computational chemistry is
forthcoming. Further synthetic applications of the proposed
nickel metallacycle and related metallacycles that are beyond
the scope of this Review have also been reported.[23]
An alternate mechanism for this class of transformations
has been discussed. If the reaction is initiated by Lewis acid
promoted oxidative addition of nickel(0) to the enone, then pallyl intermediate 21 would result (Scheme 11). Alkyne
insertion, transmetallation of the organozinc reagent, and
finally reductive elimination would afford the observed
Scheme 9. Proposed mechanism for intramolecular enone–alkyne coupling.
Scheme 11. Alternative mechanism for enone–alkyne coupling in the
presence of Ni0.
Scheme 8. Coupling of allyl halides or acetates with alkynes in the
presence of alkynyl stannanes.
3894
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
product 22. The formation of p-allyl complexes from enones,
nickel(0), and trimethylsilyl chloride is well precedented,[24]
but direct precedent for the conversion of p-allyl complex 21
into products such as 22 has not been obtained. Furthermore,
trimethylsilyl chloride is not a required additive in most
variants. Therefore, the metallacycle-based mechanism is
probably more reasonable based on data currently available.
The allyl chloride couplings by Ikeda et al. (Scheme 8) were
reported to proceed through a similar p-allyl-based pathway,
although a metallacycle-based mechanism may be operative
with that group of reactions as well.
Additional mechanistic pathways, including alkyne carbometallation pathways with organozinc or organonickel as
well as radical cyclization pathways, have been considered.[25]
Various probe substrates were examined and provided data
that appeared to be most consistent with the metallacycle or
p-allyl mechanisms described above.
the polymerization of ethylene with highly active catalysts
under similar conditions.[5]
In contrast to the well-studied dimerization and cycloisomerization of olefins by highly active cationic nickel
hydrides, much less is known about the corresponding
nickel-catalyzed reductive cyclization of dienes catalyzed by
electron-rich nickel(0) catalysts. The cyclization of bis-enones
has been developed in our laboratories by using [Ni(cod)2] as
catalyst and an organozinc complex as the reducing agent
(Scheme 13).[13b, 26] For instance, bis-enones undergo efficient
4.2. Coupling of Two Alkenes
The nickel-catalyzed dimerization of olefins has a rich
history dating back to the elegant early studies of Wilke[1a]
into the dimerization of simple a-olefins such as propylene as
well as activated olefins such as methyl acrylate. A general
theme of these early studies was that electrophilic nickel(ii)
catalysts were employed. For instance, h3-allyl nickel(ii)
halide catalysts were activated by Lewis acids to generate a
cationic nickel(ii) species (which can also be viewed as a Lewis
acid activated neutral species), which catalyzes the dimerization of propylene. Similarly, treatment of [Ni(cod)2] with
HBF4 or an h3-allyl nickel(ii) halide species with AgBF4
generates a cationic nickel(ii) tetrafluoroborate species that
catalyzes the dimerization of methyl acrylate. These highly
active species almost certainly operate by formation of a
catalytically active nickel hydride that undergoes sequential
olefin insertions prior to b-hydride elimination to produce the
dimeric product and regenerate the active nickel hydride
(Scheme 12). This general mode of reactivity has been further
illustrated in the nickel-catalyzed cyclization of a,w-dienes,[6b]
in the heterodimerization of ethylene and styrenes,[6a] and in
Scheme 12. Catalytic cycle of olefin dimerization in the presence of a
NiII complex.
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Scheme 13. Reductive cyclization of bis-enones with [Ni(cod)2] catalyst.
cyclization with BuLi/ZnCl2 to afford [3.3.0]bicyclooctanol
products. Bis-enone 23 (R = Ph) undergoes efficient cyclization to afford a single isomer of bicycloctanol 24 in 90 % yield,
whereas 23 (R = Me) provides 25 (3:1 ratio of diastereomers)
in 71 % combined yield. The organozinc structure plays a
critical role, since more reactive sp2-hybridized organozinc
reagents undergo direct conjugate addition to provide
products derived from tandem conjugate additions, as illustrated by the formation of 26 with PhLi/ZnCl2 (Scheme 13).
Little is known about the mechanism of the potentially
interesting bis-enone cyclization, which involves a reductive
cyclization/aldol addition sequence. The original report of this
reaction suggested that the cyclization is initiated by oxidative
addition of nickel(0) to a single enone (in analogy to
Scheme 11),[26] although organozinc-promoted formation of
a metallacycle is also possible. In contrast to alkynyl enones,
which cleanly afford metallacycles upon treatment with
[Ni(cod)2] and tmeda, treatment of a bis-enone with a
stoichiometric quantity of nickel(0) leads to simple coordination of the alkenes to nickel without cyclization. Thus, if
metallacycles are involved in the catalytic cyclization of bisenones, they are either generated in a small equilibrium
concentration, or their formation is promoted by the organozinc reagent. Bis-enones are good substrates for free radical
cyclizations, and the process could be initiated by electron
transfer from a low-valent Ni species, although this apparently
does not happen in the absence of an organozinc reagent or
another Lewis acidic reducing agent. Free radical cyclizations
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3895
Reviews
J. Montgomery
of bis-enones to afford [3.3.0]bicyclooctanols are well documented.[27]
Whereas the intramolecular couplings of two enones
described above require that both alkenes bear activating
substituents, an intermolecular process was developed by
Ikeda et al. that involves coupling of an electron-deficient
olefin with a strained olefin (Scheme 14).[28] Upon treatment
Scheme 14. Intermolecular coupling of an electron-deficient olefin with
a strained olefin.
of an enone with norbornene or norbornadiene and an
acetylenic stannane in the presence of [Ni(acac)2]/DIBAL-H/
TMSCl and a pyridinyloxazoline ligand, efficient coupling
occurs to generate products with up to five contiguous
stereocenters in a highly diastereoselective sense. Both acyclic
and cyclic enones participate in the process. This particular
variant was proposed to proceed by the mechanism described
in Section 4.1 for related alkyne couplings.
Scheme 15. Intramolecular coupling of 1,7-diynes with silanes.
4.3. Coupling of Two Alkynes
The coupling of two alkynes with silyl hydrides is among
the earliest examples of the general reactivity mode covered
in this Review. In the original reports of this process, Lappert
et al. discussed the fully intermolecular variant, but the scope
and stereoselectivity of the process was not studied.[29] The
real synthetic utility was not realized until the intramolecular
variant with diynes was examined by Tamao, Ito, and coworkers.[30] The coupling of 1,7-diynes with a variety of silanes
proceeds cleanly to produce six-membered ring products with
a Z-configured vinyl silane moiety, as illustrated by the
production of 27 (Scheme 15). With mixed terminal/internal
diyne substrates, the silyl unit was chemoselectively introduced at the terminal alkyne to give product 28, and yields
with tethered internal alkynes were poor. Cyclizations of
unsymmetrical diynes with basic nitrogen atoms in the tether
chain proceeded with modest regioselectivity to afford
products 29 a and 29 b (71:29).
Both inter- and intramolecular couplings involving silylboranes provide a useful entry to interesting 1-silyl-4-boryl1,3-dienes (Scheme 16).[31] Intermolecular couplings are moderately selective, with regioisomers 30 a and 30 b being
produced in a 3:1 ratio in couplings of 1-hexyne. A small
amount of product 31, derived from silylboration of a single
alkyne, was also obtained. Yields were best when a large
excess (6 equiv) of alkyne was used. The corresponding
intramolecular coupling of 1,7-octadiyne proceeded in 55 %
yield to generate product 32. Attempts to effect intermolecular cross-dimerizations were modestly successful, and for-
3896
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Scheme 16. Inter and intramolecular coupling of alkynes with silylboranes. pin = pinacolyl.
mation of the homodimeric products was difficult to suppress.
Related “germaboration” couplings of 1-hexyne proceeded
smoothly to generate dimeric products 33 a and 33 b, whereas
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
Pd and Pt catalysts produced substantial amounts of the
simple germaborated product 34.
The authors proposed mechanisms that involve initial
oxidative addition of the SiH, SiB, or GeB bond to
nickel(0) to afford intermediates such as 35 (Scheme 17).[31]
Scheme 18. Intra and intermolecular alkylative alkyne–aldehyde couplings. TMS = trimethylsilyl, Ts = p-toluenesulfonyl.
Scheme 17. Proposed mechanism for the intermolecular coupling of
alkynes with silylboranes.
Several different options were considered for the timing of
the following insertions, but the proposal for the intermolecular silylborative couplings is representative. An initial
insertion of one alkyne into the NiB bond and the second
alkyne into the NiSi bond of 35 would afford the divinyl
nickel species 36. Direct reductive elimination of this species
would afford the observed product 37. The analogous
mechanism that involves insertion of the second alkyne into
the initially formed vinyl nickel species is deemed unlikely in
intermolecular couplings on the basis of regiochemical
considerations. The authors also argue against the potential
involvement of metallacycles since there is little precedent for
the cleavage of a nickel metallacyclopentadiene with silanes
or silylboranes. (Most metallacycle-based mechanistic proposals described herein involve metallacycles that bear a
ligand activated for displacement, such as an alkoxide or
enolate, whereas the metallacyclopentadiene derived from
nickel(0) and two alkynes would likely have a much higher
activation barrier towards cleavage by a reducing agent.)
Scheme 19. Ligand dependence of the intramolecular alkylative alkyne–
aldehyde coupling.
Although alkylative cyclizations could be accessed with a
variety of organozinc reagents, undesired 1,2-addition to the
aldehyde is problematic with alkenyl zinc reagents. To avoid
this limitation and to expand the scope of readily available
alkenyl units, we examined the addition of alkenyl zirconium
reagents in both intra- (e.g., 42) and intermolecular (e.g., 43)
additions (Scheme 20).[15] Both processes proceed efficiently
and avoid many of the limitations derived from the high
reactivity of alkenyl zinc reagents. However, a surprising
observation in the intermolecular couplings is that a regio-
4.4. Coupling of Carbonyl Compounds or Imines with Alkynes
The coupling of aldehydes with alkynes has been extensively developed in both reductive and alkylative processes in
our laboratory. Couplings of this type provide access to
structurally diverse and synthetically useful allylic alcohols.
Alkylative couplings with organozinc reagents were established as both intra- (e.g., 38 and 39) and intermolecular (e.g.,
40) processes (Scheme 18).[32] In a similar fashion to that
described with enone/alkyne cyclizations, a strong ligand
dependence was noted (Scheme 19): [Ni(cod)2] alone catalyzes alkylative cyclizations effectively with a variety of
organozinc reagents to give product 41 a, and [Ni(cod)2]/PBu3
catalyzes reductive cyclizations with diethylzinc to give
product 41 b.
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Scheme 20. Intra and intermolecular addition of alkenyl zirconium
complexes to aldehydes and alkynes.
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3897
Reviews
J. Montgomery
chemical reversal of the mode of terminal alkyne insertion
occurs with alkenyl zirconium reagents compared with our
earlier observations with alkyl zinc reagents.
Recent developments from Mori and co-workers illustrated that couplings employing CO2 instead of an aldehyde
directly provide trisubstituted acrylic acid derivatives
(Scheme 21).[33] Although couplings involving CO2 were
Scheme 23. Top: reductive coupling of alkynes to aldehydes with Et3B
as the reducing agent and tributylphosphane as ligand; bottom: asymmetric variant with a menthylphosphane derivative as ligand.
Scheme 21. Reductive coupling of alkynes and organozinc reagents
with CO2. DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene.
restricted to the stoichiometric use of [Ni(cod)2], functionalized organozinc reagents derived from active zinc insertion to
alkyl and aryl halides were efficient participants in the
process.
Although simple reductive cyclizations of aldehydes and
alkynes were effective with diethylzinc, we noted two problems with more complex substrates. First, selectivity between
hydrogen-atom and ethyl-group incorporation eroded with
substrate complexity. Second, as the cyclizations become
more demanding, direct 1,2-addition of diethylzinc to the
aldehyde became a significant problem. To avoid these
complexities, a very efficient procedure employing triethylsilane as the reducing agent and tributylphosphane as ligand
was developed (Scheme 22).[34] These simple changes avoid
Scheme 22. Reductive intramolecular coupling of alkynes and carbonyl
groups with triethylsilane as the reducing agent and tributylphosphane
as ligand.
the need for selectivity for H vs. ethyl incorporation, as a
hydrogen atom is directly transferred, and undesired aldehyde reduction does not occur. The scope of reductive
cyclizations that proceed under this set of conditions is very
broad, and the process has been used in several applications in
complex molecule synthesis (Section 5.4).
Despite the efficiency of a broad range of cyclizations that
proceed with [Ni(cod)2]/PBu3/Et3SiH, the corresponding
intermolecular process is not effective under these conditions.
Jamison, however, found that using the same catalyst/ligand
combination, but changing the reducing agent to Et3B, allows
the intermolecular process to proceed efficiently
(Scheme 23).[35] A range of intermolecular couplings was
thus developed, and couplings of enynes were recently found
to be particularly regioselective.[32d, 35d] Additionally, a very
attractive asymmetric variant employing a menthyl-based
monodentate phosphane proceeded with excellent enantioselectivities.[36]
3898
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
The corresponding alkylative intermolecular couplings of
imines and alkynes was also developed by Jamison and Patel
(Scheme 24).[37] Interestingly, a similar set of conditions that
Scheme 24. Alkylative intermolecular coupling of imines with alkynes
and alkenyl boronic acids.
led to reductive couplings with aldehydes predominantly led
to the alkylative manifold with imines, although methanol is a
required cosolvent in imine couplings. Aryl and alkenyl
boronic acids were used to prepare a broad range of 1,3 dienes
and styrene derivatives.
The fundamental mechanistic theme that we originally
suggested for this set of transformations involves oxidative
coupling of an alkyne and aldehyde with nickel(0) to afford
oxametallacycle 44, followed by a transmetallation/reductive
elimination sequence (Scheme 25).[32a] The initial oxidative
cyclization is clearly promoted by the reducing agent. The
structures of the ligand, substrate, and reducing agent all play
a role in controlling the b-hydride elimination/reductive
elimination selectivity. Subsequent reports by our group and
others on related processes suggested the same fundamental
mechanism or similar variations. A number of issues have not
been explained, including the regiochemical reversal of
alkyne insertions with organozinc reagents versus alkenyl
zirconium reagents, and the crossover from reductive to
alkylative manifolds based on subtle changes in substrate
structure, reducing agent, or reaction conditions. In addition
to the oxidative cyclization mechanism depicted (Scheme 25),
hydrometallation or silylation mechanisms may be operative
in some instances. Indeed, it is very likely that the different
variants of aldehyde/alkyne couplings proceed by different
mechanisms. In recent developments from our group, a
crossover deuterium-labeling study unambiguously demonstrated that two fundamentally different mechanisms are
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
Scheme 27. Selective intramolecular coupling of aldehydes with 1,3dienes: The selectivity of the reaction depends on the reducing agent.
Scheme 25. Proposed mechanism for alkyne–aldehyde coupling with a
nickel(0) complex.
was demonstrated, and related couplings involving
Me3SiSnBu3 were reported.[40] Further applications of these
processes in complex synthetic applications are described in
Sections 5.4 and 5.5. An attractive asymmetric version was
developed with a chiral monodentate phosphane ligand
(Scheme 28).[41]
operative in reductive couplings of aldehydes and alkynes
involving triethylsilane, and that the change in mechanism is
ligand-dependent. This insight is likely applicable to other
related processes described in this Review.[32d]
4.5. Couplings of Carbonyls with Dienes
The reductive coupling of aldehydes with 1,3-dienes,
largely developed by Mori and co-workers and by Tamaru,
Kimura, and co-workers has been one of the most thoroughly
developed reaction classes covered within this Review.
Reactions of this type may proceed in either the 1,4 or 1,2
sense to afford either homoallylic or bis-homoallylic alcohols
(Scheme 26). A variety of reducing agents have been
employed, with triethylsilane, triethylborane, diethylzinc, or
DIBAL(acac) being most common.
Scheme 28. Asymmetric variant of the intramolecular coupling of aldehydes with 1,3-dienes.
A proposal was advanced by Mori and co-workers that
two mechanisms are operative in reductive cyclizations of
diene aldehydes to produce internal and terminal alkene
isomers 48 and 50 (Scheme 29).[12] With triethylsilane as
reducing agent, the reaction is initiated by oxidative addition
of nickel(0) to the silane. Hydrometallation of the diene gives
p-allyl intermediate 47, and subsequent carbonyl insertion
and OSi reductive elimination affords internal alkene 48.
Alternatively, with DIBAL(acac) as the reducing agent,
oxidative cyclization to metallacycle 49 is followed by a
Scheme 26. Reductive coupling of aldehydes with 1,3-dienes.
Catalytic intramolecular variants developed by Mori and
co-workers employ either triethylsilane or DIBAL(acac) as
the reducing agent, although earlier examples employed
stoichiometric quantities of [Ni(acac)2] and DIBAL-H.[12, 38]
Alternatively, cyclizations developed by Tamaru and coworkers involved either diethylzinc or triethylborane as the
reducing agent.[39] In the studies by Mori and co-workers, the
selectivity for formation of homoallylic alcohols 45 or bishomoallylic alcohols 46 completely switches between experiments that employ triethylsilane and DIBAL(acac) as the
reducing agent (Scheme 27). A broad range of cyclizations
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Scheme 29. Different mechanisms for the reductive cyclization of
diene aldehydes lead to internal or terminal alkenes.
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3899
Reviews
J. Montgomery
transmetallation/reductive elimination sequence to afford
terminal alkene product 50.
It was proposed that the metallacycle mechanism was
probably not operative in the formation of internal alkene
product 48, largely on the basis that treatment of a diene/
aldehyde substrate with stoichiometric amounts of [Ni(cod)2]
did not generate substantial quantities of cyclized product.
However, the catalytic generation of a reactive equilibrium
quantity of a metallacycle or the promotion of metallacycle
formation by the reducing agent (triethylsilane) are both fully
consistent with this observation. Given that ligand substitutions and other modifications of reaction conditions are
known to alter regioselectivities substantially in reductive
eliminations of metal–allyl complexes,[42] a unified mechanism
involving reductive elimination from either allyl terminus of
p-allyl 51 (derived from metallacycle 49) is an alternative that
should be considered (Scheme 30). However, in the studies of
Scheme 31. Intermolecular diene–aldehyde coupling with triethylborane
or diethylzinc.
Scheme 30. Proposed mechanism for the reductive elimination
through a common intermediate.
asymmetric catalysis described above (Scheme 28), the internal and terminal alkene isomers of an asymmetric dienal
cyclization were obtained with different ee values, thus
providing evidence that a common intermediate after the
enantioselectivity-determining step is likely not involved.
Additional mechanisms analogous to those recently proposed
in ynal cyclizations are also possible.[32d]
Tamaru, Kimura and co-workers developed the intermolecular version of this process through the use of triethylborane and diethylzinc as reducing agents (Scheme 31).[43] The
scope of both variants is broad, and of particular interest is
that triethylborane-mediated couplings work best for couplings of aromatic and unsaturated aldehydes, as demonstrated in the production of 52 and 53, whereas diethylzincpromoted couplings work best for aliphatic aldehydes and
ketones, as shown by the synthesis of 54. These studies
provided the first clear illustration of the complementary
behavior of these two reducing agents. The process works
very nicely for solving problems in 1,2- and 1,3-acyclic
stereocontrol, as the examples illustrate. Remarkably, the
process was illustrated to be effective in water and alcohols,
thus allowing aqueous solutions of glutaraldehyde and cyclic
hemiacetals to participate in the couplings, as illustrated by
the production of 55 (Scheme 31).[44] The mechanism proposed for this class of reactions involves metallacycle
formation and transmetallation of Et3B, followed by a
b-hydride elimination/reductive elimination sequence
(Scheme 32). Related intermolecular procedures were developed by Mori and co-workers[45] and Loh et al.[46]
The corresponding alkylative cyclizations and couplings
of carbonyls and dienes have been less well-developed
3900
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Scheme 32. Proposed mechanism for the intermolecular diene–aldehyde coupling with triethylborane.
(Scheme 33).[47, 48] Several transmetallating agents that lack
b-hydrogen atoms (e.g., dimethylzinc and diphenylzinc) are
effective participants in catalytic reactions, as demonstrated
in the production of 56, and Grignard reagents are participants in stoichiometric reactions in which metallacycle 57 is
preformed prior to the introduction of the reducing agent, as
illustrated in the formation of 58 a and 58 b.
Scheme 33. Examples of alkylative cyclizations and coupling reactions
of carbonyl compounds with dienes.
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
Surprisingly, attempts by Mori and co-workers to develop
the analogous stoichiometric procedure with CO2 in place of
the aldehyde component led to the discovery of a double
carboxylation of dienes (Scheme 34).[49] Treatment of a diene
the allene/aldehyde cyclization process were concurrently
developed by Kang and Yoon[52] and by us.[53]
It was demonstrated that both aldehydes and ketones
participate in the cyclization process as do monosubstituted
and 1,3-disubstituted allenes (Scheme 36). In most cases, the
substituted cyclopentane rings were formed with high cis
selectivities, and good selectivities in favor of the Z alkene
were typically observed.
Scheme 34. Diene–CO2 coupling with double carboxylation.
and CO2 in the presence of [Ni(cod)2] and DBU afforded a
solution of metallacycles 59 a and 59 b, and the subsequent
addition of acid affords carboxylic acids 60 a and 60 b.
However, treatment of 59 a and 59 b with dimethylzinc led
to the incorporation of a second equivalent of CO2 to afford
product 61. Although the mechanism of this latter process is
unclear, it was shown that the 2 equivalents of CO2 add in an
anti sense across cyclic dienes.
Scheme 36. Cyclization through coupling of aldehydes and ketones
with monosubstituted and 1,3-disubstituted allene groups
4.6. Coupling of Alkenes or Carbonyl Compounds with Allenes
4.7. Coupling of Aldehydes with Epoxides
Alkylative coupling reactions of allenes with either
aldehydes or electron-deficient alkenes have been developed
as an approach to prepare homoallylic alcohols 62 and d,eunsaturated carbonyl compounds 63 (Scheme 35). A metallacycle-based mechanism involving intermediates 64 a and
64 b was proposed for these processes in direct analogy to the
proposals made in the corresponding enone/alkyne and
aldehyde/alkyne couplings (Sections 4.1 and 4.4). These
processes were first reported by our group in the total
syntheses of kainic acid (Section 5.1)[50] and testudinariol A
(Section 5.6),[51] and more extensive methodology studies of
A very recent development by Jamison and Molinaro was
the inter- and intramolecular reductive coupling of epoxides
and alkynes with triethylborane as the reducing agent
(Scheme 37).[54] Intermolecular coupling, as illustrated by
Scheme 37. Examples of alkyne–epoxide coupling reactions
Scheme 35. Syntheses of homoallylic alcohols and d,e-unsaturated carbonyl compounds through alkylative coupling of allenes with aldehydes
or alkenes.
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
the production of 65, is effective with internal alkynes, with
most examples involving aryl alkynes or enynes. A variety of
cyclizations to form five- and six-membered rings were
observed. All examples involved monosubstituted epoxides,
and addition always occurred at the unsubstituted epoxide
position (e.g., !67).
The regioselectivity of the addition to the unsubstituted
position of the epoxide in intramolecular versions has
interesting mechanistic implications, as a metallacycle-based
mechanism would likely require addition at the substituted
position. The mechanism thus likely falls within the classification involving oxidative addition to one of the reaction
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3901
Reviews
J. Montgomery
components. As suggested by Jamison and Molinaro, direct
oxidative addition to the epoxide would afford a fourmembered oxametallacycle 68 (Scheme 38). Alkyne inser-
Scheme 38. Proposed mechanism for the alkyne–epoxide coupling.
tion, followed by reduction of the CNi bond by ethyl transfer
from boron to nickel, and then a b-hydride elimination/
reductive elimination sequence would afford the observed
product 67. The formulation of a mechanism that differs from
that of most processes covered in this Review is reasonable
because the epoxide opening involves the cleavage of a single
bond during the catalytic process, unlike the other reactions
covered herein.
Scheme 39. Sequential coupling and cyclization of an alkynyl enal with
an acetylenic stannane.
4.8. Combinations and Domino Reactions
Whereas the previous sections largely focused on threecomponent couplings involving two p systems and a maingroup organometallic reagent or metal hydride, several
processes have been developed that involve four or more
components through combinations of the previously reported
methods.
In one example, our group recognized that strong
similarities exist between the ynal cyclizations and alkynyl
enone cyclizations developed in our laboratories.[55] Thus, we
examined the reactivity of enals with the idea that both the
C=C and C=O bonds could participate in sequential couplings
or cyclizations (Scheme 39). Coupling of an enal with an
alkyne and acetylenic stannane affords conjugated enynes
such as 69 and 71 with a tethered aldehyde, with the observed
reaction taking place by addition to the C=C bond of the enal.
Further treatment of the initial products 69 and 71 under
similar conditions then results in cyclization of the aldehyde
and alkyne units to afford products such as 70 and 72. In the
second coupling event, organozinc reagents lead exclusively
to the introduction of carbon substituents, whereas triethylborane leads exclusively to the introduction of a hydrogen
atom. Fully intermolecular four-component couplings and
partially intramolecular variants are possible.
Tamaru and co-workers developed a one-pot combination
sequence involving a 1,3-diene, an alkyne, an organozinc
reagent, and an aldehyde (Scheme 40).[56] Impressive chemoselectivity is illustrated in this process, which allows the rapid
assembly of complex structures such as 74. It was proposed
that the diene, alkyne, and aldehyde form metallacycle 73
prior to the transmetallation event.
Mori and Takimoto developed a different four-component
coupling of two 1,3-dienes, CO2, and an organozinc reagent
(Scheme 41).[57] The catalytic process allows densely functionalized ring systems such as 76 to be assembled in a
3902
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Scheme 40. Chemoselective one-pot synthesis of 74.
Scheme 41. Example of a four-component reaction for the preparation
of highly functionalized rings.
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
straightforward fashion from simple precursors. A very recent
report describes asymmetric variants of this process.[57b] A
related multicomponent coupling of two dienes, a silyl
chloride, and a Grignard reagent was developed by Kambe
and co-workers to afford products such as 78 (Scheme 42).[58]
Scheme 42. Example of a multicomponent coupling reaction with a
Grignard reagent.
The mechanism of these processes likely involves the
formation of metallacycle 75 or 77 by oxidative cyclization
of two dienes, followed by alkylation and transmetallation/
reductive-elimination steps. Kambe and co-workers demonstrated that transmetallation likely precedes alkylation.
Ikeda et al. also developed a multicomponent coupling
that involves an enone, an alkyne, an alkene, ZnCl2, and Zn
dust as the reducing agent (Scheme 43).[59] This interesting
pathway. A similar effect was elucidated by Negishi and coworkers in apparent 6-endo Heck cyclizations.[60]
5. Applications in the Synthesis of Complex
Molecules
As many of the methods described above constitute very
recent developments, application of these methods in the
synthesis of complex molecules are just beginning to appear.
In several examples below, the Ni-catalyzed process is critical
to the complete synthetic plan. The overall synthetic plans
will only be described briefly, and the discussion that follows
will focus on how a key nickel-catalyzed step is critical in the
assembly of important structural features of the target
molecules. Additionally, the reader is directed to the important advances that appeared since submission of this
Review.[35b,c]
5.1. Kainoid Amino Acids
The kainoid amino acids comprise a large class of natural
products ranging from the simplest members, kainic acid and
allokainic acid, up to more complex members such as
isodomoic acid G (Schemes 44–46). Through the use of
Scheme 44. Key step in the total synthesis of kainic acid.
Scheme 43. Proposed mechanism for the multicomponent coupling
reaction to form 80.
process was proposed to involve the formation of metallacycle 79, followed by ZnCl2-mediated metallacycle cleavage,
5-exo and 3-exo cyclization, b-CC cleavage, and b-hydride
elimination to afford product 80. The latter portion of this
mechanism explains the stereochemistry reversal that would
have occurred if the process proceeded by a simpler 6-endo
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
nickel-catalyzed unsaturated imide/alkyne and unsaturated
imide/allene cyclizations, our group recently completed the
total syntheses of each of these three natural products. Kainic
acid was prepared by the nickel-catalyzed cyclization of allene
81 with dimethylzinc (Scheme 44).[50] This key cyclization
directly assembled the pyrrolidine nucleus and set the relative
stereochemistry about the five-membered ring. The epimeric
structure allokainic acid was prepared by the nickel-catalyzed
cyclization of alkyne 82 with dimethylzinc, followed by a Tsuji
rearrangement to install the C4 stereocenter (Scheme 45).[14]
This complementary sequence allows allene cyclizations and
alkyne cyclizations to provide a stereodivergent strategy for
the preparation of these two epimeric natural products.
Isodomoic acid G contains an exocyclic alkene at C4
which is more structurally complex than the corresponding
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3903
Reviews
J. Montgomery
Scheme 47. Construction of the framework of isogeissoschizine.
Scheme 45. Key step in the total synthesis of allokainic acid.
prepared by ozonolysis and double reductive amination of
cyclopentene precursor 86, was cleanly effected in 84 % yield
with dimethylzinc and [Ni(cod)2] (10 mol %). A sequence
involving a Fischer indole synthesis proceeded with epimerization to allow the preparation of the isogeissoschizine
framework 88. This sequence allowed the construction of
the alkaloid D-ring and the installation of the exocyclic
ethylidene in a single step.
5.3. Pentalenene Triquinanes
The angularly-fused triquinane unit is a common structural motif in a variety of naturally occurring terpenes,
including pentalenene, pentalenic acid, and deoxypentalenic
acid. Through the use of a nickel-catalyzed bis-enone
cyclization, our group prepared a triquinane structure by
sequential reductive cyclization and Dieckmann condensation (Scheme 48).[63] 1,2-Addition of an alkyllithium to
Scheme 46. Key step in the total synthesis of isodomoic acid G.
TIPS = triisopropylsilyl.
side chain of kainic and allokainic acids (Scheme 46). The
nickel-catalyzed cyclization of alkyne 85 with the vinyl
zirconium reagent 84 derived from alkyne 83 was used to
prepare directly the isodomoic acid G core structure.[61]
Notably, the pyrrolidine unit, the C2/C3 relative stereochemistry, and the complete densely functionalized 1,3-diene were
assembled in a single operation in a completely selective
fashion to allow an efficient total synthesis of this natural
product.
5.2. Indole Alkaloids
Geissoschizine, which has been isolated from a variety of
plant species, is an important biosynthetic precursor to a large
number of polycyclic indole alkaloids. Through the use of a
nickel-catalyzed unsaturated imide/alkyne cyclization, our
group recently prepared the isogeissoschizoid skeleton
(Scheme 47).[62] Cyclization of substrate 87, which was
3904
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Scheme 48. Triquinane synthesis through nickel-catalyzed bis-enone
cyclization.
dimethylcyclopentenone, followed by oxidative transposition
with pyridinium chlorochromate allowed efficient preparation of cyclization substrate 89. Treatment of 89 with ZnEt2/
ZnCl2 and [Ni(cod)2] (10 mol %) afforded triquinane 90 in
56 % yield as a mixture of epimers. Compound 90, which was
thus prepared in only five steps from dimethylcyclopentenone, had previously been converted into each of the
triquinane natural products noted above.[64]
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
5.4. Indolizidine Alkaloids
The indolizidine skeleton is present in a diverse range of
alkaloids. Through the use of a nickel-catalyzed ynal cyclization, our group completed the total syntheses of three
members of the pumiliotoxin family.[34] In a representative
example, (+)-allopumiliotoxin 339A was prepared from
structurally complex ynal 91, which was assembled from
proline and threonine (Scheme 49). Cyclization of ynal 91
features being controlled in a completely selective manner.
Simple deprotection of the nickel-mediated cyclization product allowed completion of the synthesis.
The indolizidine alkaloid ()-elaeokanine C was prepared by Mori and co-workers through a nickel-catalyzed
aldehyde–diene cyclization (Scheme 50).[65] Cyclization precursor 94, which was prepared from compound 93, was
treated with triethylsilane and catalytic quantities of
[Ni(cod)2]/PPh3. Although a nearly 1:1 ratio of diastereomers
was obtained, the undesired isomer 95 b was recycled by a
Mitsunobu reaction to 95 a, which was then converted into an
advanced intermediate, thus completing the formal synthesis
of elaeokanine C.
5.5. Prostaglandins
The prostaglandins comprise a very large class of natural
products with diverse biological activities. Through the use of
a nickel-catalyzed aldehyde/diene cyclization, Mori and coworkers synthesized a representative member of this class of
natural products, prostaglandin F2a (Scheme 51).[66] Substrate
Scheme 49. Total synthesis of (+)-allopumiliotoxin 339A. SEM = trimethylsilylethoxymethyl.
with triethylsilane and a catalytic quantity of [Ni(cod)2]/PBu3
proceeded in a remarkably efficient manner, furnishing
bicycle 92 as a single diastereomer in 93 %. This single step
assembles the six-membered ring of the indolizidine core,
controls the relative stereochemistry adjacent to a quaternary
center, and assembles the alkylidene unit, with each of these
Scheme 51. Total synthesis of prostaglandin F2a (PGF2) through aldehyde–diene cyclization.
97 was efficiently prepared by a sequence involving the
addition of vinylmagnesium bromide to enantioenriched
epoxide 96. Treatment of 97 with DIBAL(acac), catalytic
[Ni(cod)2]/PPh3, and 1,3-cyclohexadiene (as a critical additive
to control the position and stereochemistry of the desired
Z alkene) resulted in the selective formation of product 98,
with complete control of the contiguous stereocenters and
alkene stereochemistry. Compound 98 was then converted
into prostaglandin F2a in a straightforward fashion to complete a very attractive synthesis of this natural product.
5.6. Testudinariol A
Scheme 50. Total synthesis of ()-elaeokanine C through nickel-catalyzed aldehyde–diene cyclization.
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Testudinariol A is a member of a small family of C2symmetric natural products with an internal butylene core.
Through the use of a nickel-catalyzed aldehyde/allene
cyclization, our group recently completed an asymmetric
total synthesis of this natural product (Scheme 52).[51] The
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3905
Reviews
J. Montgomery
the transition metals in terms of reactivity trends, functionalgroup tolerance, and catalytic activity. The reductive cyclizations and couplings described in this Review provide a clear
illustration of this unique position.
I thank the National Institutes of Health, National Science
Foundation, Petroleum Research Fund, Arthur C. Cope
Foundation, Camille and Henry Dreyfus Foundation, Johnson
and Johnson, Pfizer, and 3M Pharmaceuticals for support of
our research in the area of developing nickel-catalyzed
reactions.
Received: September 9, 2003 [A634]
Published Online: June 17, 2004
Scheme 52. Asymmetric total synthesis of testudinariol A through
nickel-catalyzed aldehyde–allene cyclization. Mes = 2,4,6-trimethylphenyl, MEM = methoxyethoxymethyl.
synthesis began with an Abiko–Masamune asymmetric anti
aldol reaction to assemble compound 99, which was converted
into cyclization precursor 100. Treatment of 100 with dimethylzinc and catalytic [Ni(cod)2]/PBu3, with Ti(OiPr)4 as a
coadditive, resulted in the selective production of 101 as a
single diastereomer in 62 % yield. Conversion of 101 into 102
followed by a two-directional oxocarbenium ion cyclization
resulted in a very direct and efficient synthesis of (+)testudinariol A. Notably, the nickel-catalyzed cyclization
allowed the stereoselective introduction of four contiguous
stereocenters and directly assembled the requisite functionality needed to complete the synthesis.
6. Summary and Outlook
Over the past decade, nickel-catalyzed reductive cyclizations and couplings have evolved into a broadly useful
strategy for assembling synthetically versatile substructures
as well as complex molecules. The three-component nature of
the processes lends itself very well to the rapid generation of
molecular complexity from simple p components and maingroup organometallic reagents. Although many different
classes of reaction components have been demonstrated to
participate in the processes, new variants are rapidly being
discovered, and that trend will likely continue for some time.
Furthermore, useful asymmetric versions are just beginning to
emerge. Many complex mechanistic questions surrounding
these processes exist, and future work will likely clarify many
of these questions. The complexity of problems that will be
addressed by these methods, as well as the frequency with
which these methods are utilized, will also undoubtedly
continue to increase. Nickel maintains a unique place among
3906
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[1] a) G. Wilke, Angew. Chem. 1988, 100, 189; Angew. Chem. Int.
Ed. Engl. 1988, 27, 185; b) P. W. Jolly, Comprehensive Organometallic Chemistry, Vol. 8 (Ed.: G. Wilkinson), Pergamon,
Oxford, 1982, p. 615; c) “Organometallic Complexes of
Nickel”: J. Montgomery in Science of Synthesis (Houben-Weyl
Methods of Molecular Transformations), Vol. 1 (Eds.: B. M.
Trost, M. Lautens), Thieme, Stuttgart, 2001, p. 11.
[2] a) R. Benn, B. BLssemeier, S. Holle, P. W. Jolly, R. Mynott, I.
Tkatchenko, G. Wilke, J. Organomet. Chem. 1985, 279, 63; b) W.
Keim, Angew. Chem. 1990, 102, 251; Angew. Chem. Int. Ed.
Engl. 1990, 29, 235; c) P. A. Wender, N. C. Ihle, Tetrahedron Lett.
1987, 28, 2451; d) P. A. Wender, M. L. Snapper, Tetrahedron Lett.
1987, 28, 2221; e) P. A. Wender, N. C. Ihle, C. R. D. Correia, J.
Am. Chem. Soc. 1988, 110, 5904; f) P. A. Wender, M. J. Tebbe,
Synthesis 1991, 1089; g) P. A. Wender, J. M. Nuss, D. B. Smith, A.
SuMrez-Sobrino, J. Vagberg, D. Decosta, J. Bordner, J. Org.
Chem. 1997, 62, 4908; h) P. A. Wender, N. C. Ihle, J. Am. Chem.
Soc. 1986, 108, 4678.
[3] a) P. A. Wender, T. E. Jenkins, J. Am. Chem. Soc. 1989, 111,
6432; b) P. A. Wender, T. E. Smith, J. Org. Chem. 1995, 60, 2962;
c) P. A. Wender, T. E. Jenkins, S. Suzuki, J. Am. Chem. Soc. 1995,
117, 1843; d) P. A. Wender, T. E. Smith, J. Org. Chem. 1996, 61,
824; e) P. A. Wender, T. E. Smith, Tetrahedron 1998, 54, 1255.
[4] a) M. Lautens, W. Klute, W. Tam, Chem. Rev. 1996, 96, 49; b) P.
Bhatarah, E. H. Smith, J. Chem. Soc. Perkin Trans. 1 1992, 2163;
c) Y. Sato, T. Nishimata, M. Mori, J. Org. Chem. 1994, 59, 6133;
d) S. Ikeda, H. Watanabe, Y. Sato, J. Org. Chem. 1998, 63, 7026;
e) M. Shanmugasundaram, M. S. Wu, M. Jeganmohan, C.-W.
Huang, C.-H. Cheng, J. Org. Chem. 2002, 67, 7724; f) J. Louie,
J. E. Gibby, M. V. Farnworth, T. N. Tekavec, J. Am. Chem. Soc.
2002, 124, 15 188.
[5] a) L. K. Johnson, C. M. Killian, M. Brookhart, J. Am. Chem. Soc.
1995, 117, 6414; b) C. M. Killian, D. J. Tempel, L. K. Johnson, M.
Brookhart, J. Am. Chem. Soc. 1996, 118, 11 664; c) C. M. Killian,
L. K. Johnson, M. Brookhart, Organometallics 1997, 16, 2005;
d) S. A. Svejda, M. Brookhart, Organometallics 1999, 18, 65.
[6] a) N. Nomura, J. Jin, H. Park, T. V. RajanBabu, J. Am. Chem.
Soc. 1998, 120, 459; b) B. Radetich, T. V. RajanBabu, J. Am.
Chem. Soc. 1998, 120, 8007.
[7] a) R. J. McKinney, W. A. Nugent, Organometallics 1989, 8, 2871;
b) A. L. Casalnuovo, T. V. RajanBabu, T. A. Ayers, T. H.
Warren, J. Am. Chem. Soc. 1994, 116, 9869; c) T. Horiuchi, E.
Shirakawa, K. Nozaki, H. Takaya, Tetrahedron: Asymmetry
1997, 8, 57.
[8] a) Y. Kiso, M. Kumada, K. Tamao, M. Umeno, J. Organomet.
Chem. 1973, 50, 297; b) M. Lautens, P. Chiu, S. Ma, T. Rovis, J.
Am. Chem. Soc. 1995, 117, 532; c) M. Lautens, T. Rovis, J. Am.
Chem. Soc. 1997, 119, 11 090; d) M. Lautens, S. Ma, P. Chiu, J.
Am. Chem. Soc. 1997, 119, 6478; e) S. Vettel, A. Vaupel, P.
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Angewandte
Chemie
Nickel-Catalyzed Couplings
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
Knochel, Tetrahedron Lett. 1995, 36, 1023; f) S. Vettel, A.
Vaupel, P. Knochel, J. Org. Chem. 1996, 61, 7473.
For reviews of segments of this area, see: a) J. Montgomery, Acc.
Chem. Res. 2000, 33, 467; b) J. Montgomery, K. K. D. Amarasinghe, S. K. Chowdhury, E. Oblinger, J. Seo, A. V. Savchenko,
Pure Appl. Chem. 2002, 74, 129; c) S. Ikeda, Acc. Chem. Res.
2000, 33, 511; d) S. Ikeda, Angew. Chem. 2003, 115, 5276; Angew.
Chem. Int. Ed. 2003, 42, 5120; e) K. M. Miller, C. Molinaro, T. F.
Jamison, Tetrahedron: Asymmetry 2003, 14, 3619.
a) F. M. Dayrit, D. E. Gladkowski, J. Schwartz, J. Am. Chem.
Soc. 1980, 102, 3976; b) F. M. Dayrit, J. Schwartz, J. Am. Chem.
Soc. 1981, 103, 4466; c) D. G. Morrell, J. K. Kochi, J. Am. Chem.
Soc. 1975, 97, 7262; d) C. Amatore, A. Jutand, Acta. Chem.
Scand. 1990, 44, 755.
D. J. Krysan, P. B. Mackenzie, J. Org. Chem. 1990, 55, 4229.
Y. Sato, M. Takimoto, M. Mori, J. Am. Chem. Soc. 2000, 122,
1624.
a) J. Montgomery, A. V. Savchenko, J. Am. Chem. Soc. 1996, 118,
2099; b) J. Montgomery, E. Oblinger, A. V. Savchenko, J. Am.
Chem. Soc. 1997, 119, 4911; c) J. Montgomery, M. V. Chevliakov,
H. L. Brielmann, Tetrahedron 1997, 53, 16 449.
M. V. Chevliakov, J. Montgomery, Angew. Chem. 1998, 110,
3346; Angew. Chem. Int. Ed. 1998, 37, 3144.
Y. Ni, K. K. D. Amarasinghe, J. Montgomery, Org. Lett. 2002, 4,
1743.
a) S. Ikeda, Y. Sato, J. Am. Chem. Soc. 1994, 116, 5975; b) S.
Ikeda, K. Kondo, Y. Sato, J. Org. Chem. 1996, 61, 8248.
S. Ikeda, K. Kondo, Y. Sato, Chem. Lett. 1999, 1227.
a) S. Ikeda, H. Yamamoto, K. Kondo, Y. Sato, Organometallics
1995, 14, 5015; b) J. Montgomery, J. Seo, H. M. P. Chui,
Tetrahedron Lett. 1996, 37, 6839.
S. Ikeda, D.-M. Cui, Y. Sato, J. Am. Chem. Soc. 1999, 121, 4712.
a) S. Ikeda, D.-M. Cui, Y. Sato, J. Org. Chem. 1994, 59, 6877;
b) D.-M. Cui, T. Tsuzuki, K. Miyake, S. Ikeda, Y. Sato,
Tetrahedron 1998, 54, 1063.
S. Ikeda, H. Miyashita, Y. Sato, Organometallics 1998, 17, 4316.
K. K. D. Amarasinghe, S. K. Chowdhury, M. J. Heeg, J. Montgomery, Organometallics 2001, 20, 370.
a) G. M. Mahandru, A. R. L. Skauge, S. K. Chowdhury, K. K. D.
Amarasinghe, M. J. Heeg, J. Montgomery, J. Am. Chem. Soc.
2003, 125, 13 481; b) S. K. Chowdhury, K. K. D. Amarasinghe,
M. J. Heeg, J. Montgomery, J. Am. Chem. Soc. 2000, 122, 6775.
a) J. R. Johnson, P. S. Tully, P. B. Mackenzie, M. Sabat, J. Am.
Chem. Soc. 1991, 113, 6172; b) B. A. Grisso, J. R. Johnson, P. B.
Mackenzie, J. Am. Chem. Soc. 1992, 114, 5160.
J. Montgomery, J. Seo, Tetrahedron 1998, 54, 1131.
A. V. Savchenko, J. Montgomery, J. Org. Chem. 1996, 61, 1562.
a) E. J. Enholm, K. S. Kinter, J. Org. Chem. 1995, 60, 4850;
b) E. J. Enholm, K. S. Kinter, J. Am. Chem. Soc. 1991, 113, 7784;
c) D. S. Hays, G. C. Fu, J. Org. Chem. 1996, 61, 4.
D.-M. Cui, H. Yamamoto, S. Ikeda, K. Hatano, Y. Sato, J. Org.
Chem. 1998, 63, 2782.
a) M. F. Lappert, S. Takahashi, J. Chem. Soc. Chem. Commun.
1972, 1272; b) M. F. Lappert, T. A. Nile, S. Takahashi, J.
Organomet. Chem. 1974, 72, 425.
a) K. Tamao, K. Kobayashi, Y. Ito, J. Am. Chem. Soc. 1989, 111,
6478; b) K. Tamao, K. Kobayashi, Y. Ito, Synlett 1992, 539.
M. Suginome, T. Matsuda, Y. Ito, Organometallics 1998, 17, 5233.
a) E. Oblinger, J. Montgomery, J. Am. Chem. Soc. 1997, 119,
9065; b) M. Lozanov, J. Montgomery, Tetrahedron Lett. 2001, 42,
3259; c) X. Qi, J. Montgomery, J. Org. Chem. 1999, 64, 9310;
d) G. M. Mahandru, G. Liu, J. Montgomery, J. Am. Chem. Soc,
2004, 126, 3698.
a) M. Takimoto, K. Shimizu, M. Mori, Org. Lett. 2001, 3, 3345;
b) K. Shimizu, M. Takimoto, M. Mori, Org. Lett. 2003, 5, 2323.
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
[34] a) X.-Q. Tang, J. Montgomery, J. Am. Chem. Soc. 1999, 121,
6098; b) X.-Q. Tang, J. Montgomery, J. Am. Chem. Soc. 2000,
122, 6950.
[35] a) W.-S. Huang, J. Chan, T. F. Jamison, Org. Lett. 2000, 2, 4221.
New reports are frequently appearing in this very active field,
and many developments have been published since submission
of this Review. Of particular note, two reports in total synthesis
from the Jamison group on the methods described in Sections 4.4
and 4.7 greatly expand the scope of the processes and
demonstrate significant applications in complex molecule synthesis: b) J. Chan, T. F. Jamison, J. Am. Chem. Soc. 2003, 125,
11 514; c) E. A. Colby, K. C. OOBrien, T. F. Jamison, J. Am.
Chem. Soc. 2004, 126, 998; d) K. M. Miller, T. Luanphaisarnnont,
C. Molinaro, T. F. Jamison, J. Am. Chem. Soc. 2004, 126, 4130.
[36] a) K. M. Miller, W.-S. Huang, T. F. Jamison, J. Am. Chem. Soc.
2003, 125, 3442; b) E. A. Colby, T. F. Jamison, J. Org. Chem.
2003, 68, 156.
[37] S. J. Patel, T. F. Jamison, Angew. Chem. 2003, 115, 1402; Angew.
Chem. Int. Ed. 2003, 42, 1364.
[38] a) Y. Sato, M. Takimoto, K. Hayashi, T. Katsuhara, K. Takagi, M.
Mori, J. Am. Chem. Soc. 1994, 116, 9771; b) Y. Sato, M.
Takimoto, M. Mori, Tetrahedron Lett. 1996, 37, 887; c) Y. Sato,
T. Takanashi, M. Hoshiba, M. Mori, Tetrahedron Lett. 1998, 39,
5579; d) Y. Sato, R. Sawaki, M. Mori, Organometallics 2001, 20,
5510.
[39] K. Shibata, M. Kimura, M. Shimizu, Y. Tamaru, Org. Lett. 2001,
3, 2181.
[40] Y. Sato, N. Saito, M. Mori, Chem. Lett. 2002, 18.
[41] a) Y. Sato, N. Saito, M. Mori, J. Am. Chem. Soc. 2000, 122, 2371;
b) Y. Sato, N. Saito, M. Mori, J. Org. Chem. 2002, 67, 9310.
[42] a) J. S. Temple, J. Schwartz, J. Am. Chem. Soc. 1980, 102, 7381;
b) Y. Hayasi, M. Riediker, J. S. Temple, J. Schwartz, Tetrahedron
Lett. 1981, 22, 2629; c) J. S. Temple, M. Riediker, J. Schwartz, J.
Am. Chem. Soc. 1982, 104, 1310.
[43] a) M. Kimura, A. Ezoe, K. Shibata, Y. Tamaru, J. Am. Chem.
Soc. 1998, 120, 4033; b) M. Kimura, H. Fujimatsu, A. Ezoe, K.
Shibata, M. Shimizu, S. Matsumoto, Y. Tamaru, Angew. Chem.
1999, 111, 410; Angew. Chem. Int. Ed. 1999, 38, 397.
[44] M. Kimura, A. Ezoe, S. Tanaka, Y. Tamaru, Angew. Chem. 2001,
113, 3712; Angew. Chem. Int. Ed. 2001, 40, 3600.
[45] a) M. Takimoto, Y. Hiraga, Y. Sato, M. Mori, Tetrahedron Lett.
1998, 39, 4543; b) Y. Sato, R. Sawaki, N. Saito, M. Mori, J. Org.
Chem. 2002, 67, 656.
[46] T.-P. Loh, H.-Y. Song, Y. Zhou, Org. Lett. 2002, 4, 2715.
[47] a) M. Kimura, S. Matsuo, K. Shibata, Y. Tamaru, Angew. Chem.
1999, 111, 3586; Angew. Chem. Int. Ed. 1999, 38, 3386; b) M.
Kimura, K. Shibata, Y. Koudahashi, Y. Tamaru, Tetrahedron
Lett. 2000, 41, 6789; c) K. Shibata, M. Kimura, K. Kojima, S.
Tanaka, Y. Tamaru, J. Organomet. Chem. 2001, 624, 348.
[48] Y. Sato, T. Takanashi, M. Mori, Organometallics 1999, 18, 4891.
[49] M. Takimoto, M. Mori, J. Am. Chem. Soc. 2001, 123, 2895.
[50] M. V. Chevliakov, J. Montgomery, J. Am. Chem. Soc. 1999, 121,
11 139.
[51] K. K. D. Amarasinghe, J. Montgomery, J. Am. Chem. Soc. 2002,
124, 9366.
[52] S.-K. Kang, S.-K. Yoon, Chem. Commun. 2002, 2634.
[53] J. Montgomery, M. Song, Org. Lett. 2002, 4, 4009.
[54] C. Molinaro, T. F. Jamison, J. Am. Chem. Soc. 2003, 125, 8076;
see also references [35 c, d].
[55] M. Lozanov, J. Montgomery, J. Am. Chem. Soc. 2002, 124, 2106.
[56] A. Ezoe, M. Kimura, T. Inoue, M. Mori, Y. Tamaru, Angew.
Chem. 2002, 114, 2908; Angew. Chem. Int. Ed. 2002, 41, 2784.
[57] a) M. Takimoto, M. Mori, J. Am. Chem. Soc. 2002, 124, 10 008;
b) M. Takimoto, Y. Nakamura, K. Kimura, M. Mori, J. Am.
Chem. Soc. 2004, 126, 5956.
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3907
Reviews
J. Montgomery
[58] J. Terao, A. Oda, A. Ikumi, A. Nakamura, H. Kuniyasu, N.
Kambe, Angew. Chem. 2003, 115, 3534; Angew. Chem. Int. Ed.
2003, 42, 3412.
[59] S. Ikeda, H. Miyashita, M. Taniguchi, H. Kondo, M. Okano, Y.
Sato, K. Odashima, J. Am. Chem. Soc. 2002, 124, 12 060.
[60] Z. Owczarczyk, F. Lamaty, E. J. Vawter, E. Negishi, J. Am.
Chem. Soc. 1992, 114, 10 091.
[61] Y. Ni, K. K. D. Amarasinghe, B. Ksebati, J. Montgomery, Org.
Lett. 2003, 5, 3771.
3908
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[62] R. S. Fornicola, K. Subburaj, J. Montgomery, Org. Lett. 2002, 4,
615.
[63] J. Seo, H. Fain, J.-B. Blanc, J. Montgomery, J. Org. Chem. 1999,
64, 6060.
[64] M. T. Crimmins, J. A. DeLoach, J. Am. Chem. Soc. 1986, 108,
800.
[65] a) Y. Sato, N. Saito, M. Mori, Tetrahedron Lett. 1997, 38, 3931;
b) Y. Sato, N. Saito, M. Mori, Tetrahedron 1998, 54, 1153.
[66] a) Y. Sato, M. Takimoto, M. Mori, Synlett 1997, 734; b) Y. Sato,
M. Takimoto, M. Mori, Chem. Pharm. Bull. 2000, 48, 1753.
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3890 – 3908
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 773 Кб
Теги
nickell, couplings, cyclization, reductive, catalyzed
1/--страниц
Пожаловаться на содержимое документа