Chemical
Science
ISSN 2041-6539
rsc.li/chemical-science
Volume 8 Number 3 March 2017 Pages 1671–2466
PERSPECTIVE
Dominic R. Pye and Neal P. Mankad
Bimetallic catalysis for C–C and C–X coupling reactions
Bimetallic catalysis for C–C and C–X coupling
reactions
Dominic R. Pye and Neal P. Mankad
*
Bimetallic catalysis represents an alternative paradigm for coupling chemistry that complements the more
traditional single-site catalysis approach. In this perspective, recent advances in bimetallic systems for
catalytic C–C and C–X coupling reactions are reviewed. Behavior which complements that of
established single-site catalysts is highlighted. Two major reaction classes are covered. First, generation
of catalytic amounts of organometallic species of e.g. Cu, Au, or Ni capable of transmetallation to a Pd
co-catalyst (or other traditional cross-coupling catalyst) has allowed important new C–C coupling
technologies to emerge. Second, catalytic transformations involving binuclear bond-breaking and/or
bond-forming steps, in some cases involving metal–metal bonds, represent a frontier area for C–C and
C–X coupling processes.
1. Introduction
Coupling reactions that allow for catalytic C–CorC–Xbond
formation (X ¼ e.g. B, N, O) have revolutionized synthetic chem-
istry by allowing complex organic structures to be created from
simpler building blocks, even at late stages of multistep synthetic
sequences.
1–5
The dominant paradigm in coupling chemistry is to
utilize single-site homogeneous catalysts and tailor reactivity and
selectivitypatternsusingliganddesign.Forexample,muchof
cross-coupling catalysis involves palladium–phosphine systems
that operate by a canonical oxidative addition/reductive elimina-
tion cycling mechanism. Decades have been spent designing
elaborate phosphine ligands to provide reactivity suitable for
modern applications.
6–9
Exquisite levels of catalytic activity,
regioselectivity, and/or stereoselectivity ultimately have been
achieved using this paradigm.
Pursuing alternative catalytic paradigms that go beyond this
single-site approach has the potential to uncover complemen-
tary reactivity and selectivity regimes.
10
In addition, in some
cases catalytic reactivity can become accessible with inexpensive
and earth-abundant metals not typically utilized extensively in
cross-coupling catalysis. This perspective highlights one such
alternative approach: the use of bimetallic catalysis for C–C and
Dominic studied chemistry at
the University of Manchester,
where he obtained an MChem in
2010 aer completing his nal
year project with Dr Peter
Quayle. He then moved to the
University of Bristol where he
undertook his PhD on iron-cat-
alysed cross-coupling reactions
under the supervision of Profes-
sors Robin Bedford and Timothy
Gallagher. In 2015 he moved to
the University of Illinois at Chi-
cago to work as a post-doctoral research associate investigating
heterobimetallic-catalysed carbonylation reactions with Professor
Neal Mankad.
Before starting his independent
career at the University of Illi-
nois at Chicago (UIC) in 2012,
Neal conducted undergraduate
research at MIT with Joseph
Sadighi, graduate research at
Caltech with Jonas Peters, and
postdoctoral research at the
University of California-Berkeley
with Dean Toste. At UIC, his
group conducts research on
bimetallic catalysis in organic
systems, multimetallic catalytic
sites in biology, and other topics related to energy and health
sciences. Neal's recent awards include an Alfred P. Sloan Research
Fellowship, a Thieme Chemistry Journal Award, and the UIC
Rising Star Researcher of the Year. He likes exercising, eating,
drinking, and traveling (sometimes all on the same day).
Department of Chemistry, University of Illinois at Chicago, 845 W. Taylor St., Chicago,
IL 60607, USA. E-mail: npm@uic.edu
Cite this: Chem. Sci.,2017,8,1705
Received 19th December 2016
Accepted 14th January 2017
DOI: 10.1039/c6sc05556g
www.rsc.org/chemicalscience
This journal is © The Royal Society of Chemistry 2017 Chem. Sci.,2017,8,1705–1718 | 1705
Chemical
Science
PERSPECTIVE
Open Access Article. Published on 16 January 2017. Downloaded on 8/26/2022 11:02:28 AM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online
View Journal
| View Issue
C–X bond forming reactions.
11
The focus of the perspective is on
recent developments in bimetallic catalysis as applied to catalytic
C–C and C–X bond formation in molecular organic systems. All of
the included examples involve catalytic amounts of two d-block
metals cooperating during catalysis. The following topics are
excluded from this perspective: reactions involving d-block
metals cooperating with main-group elements (e.g. Na, Al, B, P)
during catalysis;
12,13
cases with one of the two metals not
participating directly in bond-breaking/forming events (e.g. pho-
toredox catalysis,
14
Wacker oxidation
15
); cases with one of the two
metals present in stoichiometric quantities; tandem, cascade, or
domino reactions
16
that do not involve direct communication
between codependent catalytic metals; and cluster systems where
mechanistic understanding is limited.
17
Also excluded from the
perspective are catalytic polymerization reactions,
18–20
reductions
of unsaturated organics (e.g. hydrogenation, hydrosilylation),
21–25
and transformations of small-molecule inorganics (e.g. H
2
,H
2
O,
CO, CO
2
,N
2
,O
2
, etc.),
26–30
all of which have had recent advances in
their own right using bimetallic approaches.
The contribution of this perspective is timely, as bimetallic
catalysis for C–C and C–X coupling is a burgeoning area that
stands to make important contributions to the synthetic toolkit.
In this perspective, these contributions are categorized into two
broad subdivisions, with representative mechanistic schemes
shown in Scheme 1. First are bimetallic systems involving
catalytic generation of an organometallic nucleophile that
undergoes transmetallation with a “traditional” coupling cata-
lyst that operates using its canonical single-site mechanism
(Scheme 1a). Here, no binuclear steps are utilized for breaking
the bonds of the coupling partners or forming the bonds of the
products.
31
Second are bimetallic systems that involve binuclear
bond activation and/or bond elimination events (Scheme 1b).
Here, metal–metal bonds oen (though not always) play key
roles during catalysis.
32
2. Bimetallic catalysis with
mononuclear bond breaking & forming
mechanisms
2.1 Organocopper nucleophiles
It was demonstrated in 1966 by Owsley and co-workers that
copper-acetylides react in a stoichiometric fashion with aryl
halides.
33
The harsh reaction conditions (reux in pyridine) lead
to the formation of side products, such as aryl halide reduction
and acetylene dimerization. Monometallic, palladium-catalysed
coupling of aryl halides and terminal alkynes was reported by
both Cassar
34
and Heck
35
in 1975; however both catalyst systems
still required elevated temperature for high conversion. In 1975
Sonogashira and co-workers reported that the addition of
copper salts greatly accelerated the reaction, leading to room
temperature reactivity, greater functional group tolerance, and
a generally more useful procedure.
36
It is now accepted that the
role of copper in these reactions is to react with the alkyne in the
presence of a base to give a copper acetylide.
37
The acetylide
moiety then undergoes transmetallation from copper to palla-
dium to give the key arylpalladium(II) acetylide (Scheme 2),
which subsequently releases the desired product by reductive
elimination. The key step, transmetallation of an organic frag-
ment to palladium, has proven the inspiration for all the
copper–palladium bimetallic-catalysis described in this section.
2.1.1 Borylcupration and hydrocupration of unsaturated
C–C bonds. Traditional cross-coupling requires the pre-
synthesis of organometallic or organo-main group nucleophiles
for use. Organometallic nucleophiles are oen unstable, and
require synthesis immediately prior to use. Organo-main group
nucleophiles are usually bench stable, but also require
synthesis and purication prior to use. In addition to these
drawbacks, stoichiometric metal or main group byproducts are
unavoidable in such couplings.
An alternative mechanistic paradigm is the catalytic, in situ
formation of organometallic species from catalytic amounts of
a precatalyst and an organic pro-nucleophile. This avoids the pre-
formation and purication of the nucleophilic component, and
the organic pro-nucleophiles are generally more stable and readily
available than organometallic or main group nucleophiles. One
example of this approach is the insertion of unsaturated organic
compounds into copper-element bonds (Scheme 3). The product
of this insertion is a reactive organo-copper nucleophile ready for
further reaction. A further advantage in catalytic formation of the
nucleophile is that low concentrations of reactive species leads to
fewer undesired side reactions. The functionalisation of an
Scheme 1 Representative mechanistic schemes for bimetallic catal-
ysis (a) without and (b) with binuclear bond breaking and forming
events.
Scheme 2 Cu/Pd cooperation in Sonogashira coupling.
1706 | Chem. Sci.,2017,8,1705–1718
This journal is © The Royal Society of Chemistry 2017
Chemical Science Perspective
Open Access Article. Published on 16 January 2017. Downloaded on 8/26/2022 11:02:28 AM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online
unsaturated bond also leads to further complexity build up in
asinglestep.
The insertion of alkenes in to copper–boron bonds has been
implicated in many copper-catalysed C–B bond forming
processes, such as copper-catalysed hydroboration of alkenes.
In 2006 Sadighi proved this unambiguously by the isolation of
b-boryl alkyl copper species from the reaction of IPrCu(Bpin)
with styrenes (Scheme 4).
38
In 2014 Semba/Nakao
39
and Brown
40
independently showed
that this intermediate could be used as a transmetallating reagent
from copper to an arylpalladium(II), which reductively eliminates
to form an carboborylated product and Pd(0). The copper boryl
species is regenerated through alkoxide assisted transmetallation
with B
2
(pin)
2
, and palladium(0) undergoes oxidative addition
with an aryl bromide to complete two synergistic catalytic cycles.
Both groups reported (NHC)copper(I)andPd(II)/dicyclohex-
ylbiarylphosphine precatalyst mixtures (Scheme 5).
Brown and co-workers also showed that when cyclic styrene
derivatives (such as 1,2-dihydronapthalene) were employed as
substrates, in most cases the reaction gave the trans diaste-
reomer with high selectivity. As the addition of (SIMes)CuB(pin)
across the alkene is likely to proceed in a syn-fashion, it was
stated that the transmetallation Cu–Pd must proceed with
inversion of stereochemistry to give the trans isomer aer
stereoretentive reductive elimination. Under the described
conditions, acyclic, 1,2-disubstituted styrenes gave low diaster-
eoselectivity. Brown later reported modied conditions under
which both the syn- and anti-carboboration diastereomers of
acyclic disubstituted styrenes could be obtained, selectivity
being determined by a change in solvent and ligand (Scheme
6).
41
It was found that the use of THF and RuPhos would
selectively provide the syn-product, presumably through a ster-
eoretentive transmetallation of the putative syn-Cu-alkyl inter-
mediate, followed by reductive elimination. Diastereoselectivity
was found to be reversed if the solvent was changed to toluene,
and the Pd ligand to triisobutylphosphine. It should be noted
that both the change of solvent and ligand were required; both
Pd-RuPhos in toluene and Pd-PiBu
3
in THF gave low
diastereoselectivity.
Liao and co-workers reported an enantioselective variant of
the Cu/Pd carboboration reaction in 2015 (Scheme 7).
42
Here
they use copper(
I) acetate with a chiral sulfoxide– phosphine
ligand to achieve enantioselective borylcupration of a styrene,
followed by transmetallation to a palladium-allyl species
generated from the oxidative addition of allyl-tert-butylcar-
bonates to Pd(0). Products were generated with ee > 90%. When
racemic, cyclic allylic carbonates were used as electrophiles,
diastereomeric control of the two contiguous chiral centres was
achieved. It was also demonstrated that iodobenzene could be
used in place of the allylic carbonate to give an enantioselective
version of the Nakao–Brown carboboration, albeit in lower
yield.
In 2016 Semba/Nakao reported a procedure for the carbo-
boration of alkenes, this time using nickel rather than palla-
dium to activate the aryl electrophile.
43
As a rst-row transition
metal, nickel is preferable to palladium in terms of cost and
earth abundance. It was also found that the Nakao–Brown
conditions were not suitable for the use of aryl chlorides, or
phenol derived electrophiles (other than triates). Semba/
Nakao demonstrated that styrenes could be converted to their
carboborylated product using a precatalyst system of CuCl,
Ni(acac)
2
, and tricyclopentylphosphine, whilst using aryl chlo-
rides or tosylates as the electrophilic carbon source (Scheme 8).
Drawbacks, in comparison with the copper/palladium-catalysed
Scheme 3 General scheme for organocopper nucleophiles generated
by element-cupration.
Scheme 4 Borylcupration of styrenes demonstrated by Sadighi.
Scheme 5 Cu/Pd catalysis for arylboration developed by the groups of
Semba/Nakao and Brown.
Scheme 6 Diastereoselective carboboration developed by Brown.
This journal is © The Royal Society of Chemistry 2017 Chem. Sci.,2017,8,1705–1718 | 1707
Perspective Chemical Science
Open Access Article. Published on 16 January 2017. Downloaded on 8/26/2022 11:02:28 AM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online
systems are the requirement for higher temperatures (80
C vs.
room temperature) and the poor reactivity of b-substituted
styrenes (30–40% yield).
In 2016 Semba/Nakao reported a system wherein borylcup-
ration could be replaced with hydrocupration, thereby furnishing
1,1-diarylalkanes (Scheme 9).
44
(NHC)CuH is generated in situ by
the reaction of (NHC)CuOtBu with HSi(OEt)
3
. The use of
a deuterium labelled silane gave conrmation of syn-addition
across the double bond, and subsequent stereoinversion upon
transmetallation. Theoretical calculations suggest an S
E
2(back)
mechanism, which is consistent with stereoinversion.
Later that year Buchwald used a chiral bisphosphine Cu(I)
precatalyst to give an enantioselective Cu/Pd hydroarylation of
styrenes (Scheme 10).
45
The enantiodetermining step is the
addition of a chiral copper hydride across the double bond.
Riant and co-workers have described an asymmetric reduc-
tion/allylation of a,b-unsaturated ketones using a copper–
palladium bimetallic catalyst system, a silane, and allyl
carbonates.
46
The products of these reactions are highly valu-
able, chiral all-carbon quaternary centres. In this instance the
enantioselective step is governed by the palladium catalyst,
bearing a chiral PHOX ligand, as opposed to a chiral copper
hydride vide supra. Copper-catalysed 1,4-reduction of the
unsaturated ketone gives a copper-enolate, which will then
transmetallate to palladium. Mechanistic studies suggest that
this transmetallation can give both the C- and O-bound Pd-
enolate, which, aer reductive elimination, release both the
desired product and an allyl-enol ether. The allyl-enol ether
converts to the desired product under the reaction conditions
via a palladium-catalysed Cope rearrangement (Scheme 11).
The reaction of copper(
I) complexes with silyl boranes is
known to give copper-silyl species, which Riant and co-workers
showed will add across electron decient alkynes to catalytically
generate a vinyl-copper species to be used in palladium-cata-
lysed cross-coupling (Scheme 12).
47
In this instance allylic
carbonates were used as the electrophiles. The authors found
that the regioselectivity of the double bond could be controlled,
switching between the Z- and E-isomers by omitting triphenyl-
phosphine from the reaction conditions, and switching from
simple copper chloride to NHC ligated IMesCu(DBM) (DBM ¼
dibenzoylmethanate).
The effect of triphenyl phosphine on the regioselectivity is
ascribed to the relative reactivity of the palladium(
II) allyl
present, and the steric hindrance of the initial syn-silylcupration
product compared to its tautomeric allenolate. In the absence of
Scheme 7 Enantioselective carboboration developed by Liao.
Scheme 8 Cu/Ni catalysis developed by Semba/Nakao.
Scheme 9 Hydroarylation catalysis developed by Semba/Nakao.
Scheme 10 Enantioselective hydroarylation developed by Buchwald.
Scheme 11 Reductive allylation of a,b-unsaturated ketones devel-
oped by Riant.
Scheme 12 Silylation catalysis developed by Riant.
1708 | Chem. Sci.,2017,8,1705–1718
This journal is © The Royal Society of Chemistry 2017
Chemical Science Perspective
Open Access Article. Published on 16 January 2017. Downloaded on 8/26/2022 11:02:28 AM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online