Computational Perspective on Pd-Catalyzed

C
C Cross-Coupling Reaction Mechanisms

MAX GARCÍA-MELCHOR,†, ‡ ATAUALPA A. C. BRAGA,‡, §

† GREGORI UJAQUE,*, † AND
AGUSTÍ LLEDOS,
FELIU MASERAS*, †, ‡
†
Departament de Química, Edifici C.n, Universitat Autonoma de Barcelona,
08193 Cerdanyola del Valles, Catalonia, Spain, ‡Institute of Chemical Research
of Catalonia (ICIQ), Avgda. Països Catalans, 16, 43007 Tarragona, Catalonia,
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Spain, and §Instituto de Química, Universidade de Sa ~o Paulo, Sa

~o Paulo,
SP 05508-900, Brazil
RECEIVED ON MARCH 21, 2013
Downloaded via IISER KOLKATA on November 6, 2020 at 06:22:13 (UTC).

CONSPECTUS

P alladium-catalyzed C
C cross-coupling reactions (Suzuki
Miyaura, Negishi,

Stille, Sonogashira, etc.) are among the most useful reactions in modern
organic synthesis because of their wide scope and selectivity under mild conditions.
The many steps involved and the availability of competing pathways with similar
energy barriers cause the mechanism to be quite complicated. In addition, the short-
lived intermediates are difficult to detect, making it challenging to fully characterize
the mechanism of these reactions using purely experimental techniques. Therefore,
computational chemistry has proven crucial for elucidating the mechanism and
shaping our current understanding of these processes. This mechanistic elucidation
provides an opportunity to further expand these reactions to new substrates and to
refine the selectivity of these reactions.
During the past decade, we have applied computational chemistry, mostly using density functional theory (DFT), to the study of
the mechanism of C
C cross-coupling reactions. This Account summarizes the results of our work, as well as significant
contributions from others. Apart from a few studies on the general features of the catalytic cycles that have highlighted the
existence of manifold competing pathways, most studies have focused on a specific reaction step, leading to the analysis of the
oxidative addition, transmetalation, and reductive elimination steps of these processes. In oxidative addition, computational
studies have clarified the connection between coordination number and selectivity. For transmetalation, computation has increased
the understanding of different issues for the various named reactions: the role of the base in the Suzuki
Miyaura cross-coupling,
the factors distinguishing the cyclic and open mechanisms in the Stille reaction, the identity of the active intermediates in the
Negishi cross-coupling, and the different mechanistic alternatives in the Sonogashira reaction. We have also studied the closely
related direct arylation process and highlighted the role of an external base as proton abstractor. Finally, we have also rationalized
the effect of ligand substitution on the reductive elimination process.
Computational chemistry has improved our understanding of palladium-catalyzed cross-coupling processes, allowing us
to identify the mechanistic complexity of these reactions and, in a few selected cases, to fully clarify their mechanisms. Modern
computational tools can deal with systems of the size and complexity involved in cross-coupling and have a continuing role in
solving specific problems in this field.

Introduction of their main developers, Richard F. Heck, Ei-ichi Negishi, and

Catalyzed C
C cross-coupling reactions are among the Akira Suzuki, were awarded with the Nobel Prize in Chemistry
most important and useful reactions in organic synthesis in 2010 for their work in this field. These reactions consist in the
and organometallic chemistry, because they allow the carbon
carbon bond formation between an organic electro-
selective formation of C
C bonds under mild conditions phile, R1
X, and an organometallic nucleophile, R2
m, in the
and are a primary tool for the synthesis of large variety of presence of a catalyst [M] (Scheme 1). C
C cross-coupling
complex compounds from readily accessible reactants.1 Three reactions are often classified depending on the metal or
2626 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2626–2634 ’ 2013 ’ Vol. 46, No. 11 Published on the Web 07/12/2013 www.pubs.acs.org/accounts
10.1021/ar400080r & 2013 American Chemical Society
 Computational Perspective on Pd-Catalyzed C C Reactions García-Melchor et al.

SCHEME 1. General Scheme for C
C Cross-Coupling Reactions

semimetal present in the nucleophile. For instance, Suzuki

Miyaura2 is boron-mediated, Stille3 reaction tin-mediated,
Negishi4 reaction, zinc-mediated, etc. The catalysts most widely
FIGURE 1. General catalytic cycle for Pd-catalyzed C
C cross-coupling
employed are transition metal complexes from groups 8
10, reactions.
especially complexes of palladium.
Despite the wide use of cross-coupling reactions, their published. The effect of changes in the nature of the phos-
application has been often based on a trial-and-error ap- phine ligands over the different steps in the Suzuki
Miyaura
proach, because mechanistic knowledge has been difficult reaction was analyzed by Harvey et al.12 by means of steric
to acquire from pure experimental techniques. These are and electronic descriptors in conjunction with DFT methods.
multistep processes with many intermediates involved, and The combination of the steric and electronic effects resulted
competing equilibria are often present. The ever increasing in the following order in the transmetalation energy barriers
computational power and the development of more effi- with the different phosphine ligands: P(CF3)3 < PPh3 < PMe3
cient theoretical algorithms have allowed computational < P(t-Bu3). The effect of some of these ligands (i.e., PMe3,
chemistry to enter to this field in the past decade, with PPh3, and P(t-Bu3)) was also theoretically investigated by
significant contributions in terms of mechanistic clarification. Kozuch and Martin for the Suzuki
Miyaura coupling of
In 2008, this journal5 dedicated a special issue to recent Ph
Br and PhB(OH)3
catalyzed by [PdLn] complexes.13
developments in cross-coupling reactions regarding several The full catalytic cycle for the Negishi Csp3
Csp3 coupling
topics: the class of ligands used, the nature of metal cata- between Et
Br and EtZnBr catalyzed by monoligated
lysts, the types of the bonds formed (e.g., C
C, C
N), the kind NHC
Pd (NHC = N-heterocyclic carbene) complexes was
of reactants used, etc. In this Account, we cover the research also investigated by Organ et al.,14 with focus on the role of
on the reaction mechanism for the most known Pd-cata- the NHC ligand. The full catalytic cycle of a selective Heck
lyzed C
C cross-coupling reactions from a computational reaction has been computationally analyzed.15
perspective. These contributions have been collected in a Despite the fact that the three-step mechanism is gener-
number of reviews,6
8 which we are summarizing and ally accepted, each of these main steps has different possi-
updating in this Account, with particular emphasis on our bilities and can be quite subtle and specific for each reaction.
work in the field. The following sections describe the knowledge gained on
each of the three fundamental steps from computational
General Catalytic Cycle chemistry. The section devoted to transmetalation will be
It is generally accepted that cross-coupling reactions follow a the longest one because it is the characteristic reaction step
catalytic cycle consisting of three main steps (Figure 1): for the different cross-coupling reactions.
oxidative addition, transmetalation, and reductive elimina-
tion. There are, however, relatively few computational stud- The Oxidative Addition Step
ies that have attempted to compute the full catalytic cycle. In the oxidative addition step, the bond between the organic
The main reason is highlighted by our work dealing with the R group and the heteroatom X breaks, and two new bonds
full catalytic cycle for the Suzuki
Miyaura reaction, where are formed with the metal, which increases its oxidation
we showed the existence of a manifold of similar energy state by two units (Scheme 2). These reactions can be
pathways, with no clear general preference for one of them.9 reversible, but the equilibrium is expected to be displaced

11
In particular, we showed that both mono- and bis-ligated toward the oxidative addition product when ligands with
catalysts could lead in principle to efficient catalytic cycles. strong electron-donating abilities (that stabilize higher oxi-
Other works on the full catalytic cycle have been also dation states of the metal) are employed. The oxidative

Vol. 46, No. 11 ’ 2013 ’ 2626–2634 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2627

Computational Perspective on Pd-Catalyzed C C Reactions García-Melchor et al.

SCHEME 2. General Scheme for the Oxidative Addition Reaction

addition has been postulated to be rate-limiting in a number

of cross-coupling reactions.16,17
For this step, two main mechanisms have been proposed
(Figure 2). The concerted pathway entails the simultaneous FIGURE 2. Proposed reaction mechanisms for the oxidative addition
formation of the Pd
C and Pd
X bonds in the transition reaction.
state, which leads to retention of configuration at a stereo-
genic carbon atom in the case of chiral electrophiles R
X.
The second pathway is an associative bimolecular process
that consists of two steps: first, the carbon is attacked by the
metal and the anion X
is expelled giving rise to a cationic
species; subsequently, both charged species combine to
yield the product. This second mechanism has been labeled
as SN2 by analogy with the organic chemistry substitution
reaction and gives rise to the inversion of configuration if a
FIGURE 3. Competition experiments for the selective Pd-catalyzed
stereogenic carbon center is involved. Suzuki
Miyaura reaction of R-bromosulfoxides.
The oxidative addition of Ph
X (X = Cl, Br, or I) to Pd(0)
complexes with bidentate phosphines [Pd(PP)] (PP = 1,2-bis- it has been found that the barrier is systematically lower for the
(dimethylphosphino)ethane or (P)-2,20 -bis(dimethylphosphino)- monoligated system9,23
25 and that the prevalence of one or
1,10 -biphenyl) in THF solution was theoretically investigated by the other in the catalytic cycle depends on the ligand dissocia-
Senn and Ziegler.18 Interestingly, the authors could not locate in tion energy from the metal complex.
solution the concerted transition state reported for the gas A representative example of successful application of
phase. Instead, the dissociation of the halide and its subsequent calculations to mechanistic clarification is provided by our
recombination with the cationic phenyl complex (i.e., SN2 study on the relationship between the phosphine nature
mechanism) was found to be a facile process in solution. For and selectivity in the Suzuki
Miyaura reactions shown in
other Pd complexes, however, the concerted transition state has Figure 3.26 The less hindered phosphine, PPh3 in our calcula-
been recently located in a polar solvent described by means of tions, is associated with a bisligated form of the catalyst,
both an implicit and a combined implicit/explicit solvation which favors a SN2 mechanism and activation of the R-
model.19 bromosulfoxide side. The more hindered phosphine, P(1-
Several authors have claimed that anionic Pd species can napthyl)3 in our calculations, is associated with the mono-
be involved in the oxidative addition step when anionic ligated form of the catalyst, which promotes a concerted
additives are present in the reaction medium. Jutand and co- mechanism and activation of the bromoaryl moiety. A
workers20 provided a significant amount of experimental similar connection between selectivity and coordination
evidence suggesting an anionic form for the catalyst, number was also identified by Schoenebeck and Houk on
[Pd(Cl)(L)n]
, while Thiel and co-workers explored the reac- the competition between oxidative addition of Ph
Cl and
tion of [Pd(OAc)(PMe3)2]
with Ph
I.21 The participation of Ph
OTf bonds.27
the anionic or neutral form of catalyst as active species has
been recently found to be significantly affected by the The Transmetalation Step
solvent polarity, especially when both species can coexist Transmetalation is the most characteristic step of C
C cross-
and compete.22 coupling reactions. This is the reaction stage where the
The bisligated form of the catalyst [Pd(L)2] has been organic group R2 bound to an electropositive group m is
postulated in most experimental proposals, but the mono- transferred to the catalyst, as shown in Scheme 3.
ligated form [Pd(L)] can be also envisaged. In fact, in the cases The mechanism for the transmetalation process is ex-
where oxidative addition to both of them has been computed, pected to show differences depending on the particular

2628 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2626–2634 ’ 2013 ’ Vol. 46, No. 11

 Computational Perspective on Pd-Catalyzed C C Reactions García-Melchor et al.

SCHEME 3. General Scheme for the Transmetalation Reaction

cross-coupling reactions since they differ in the nucleophile

used. Therefore, in the following four subsections, we pre-
sent an overview of the reported computational studies on
the transmetalation process for the Suzuki
Miyaura, Stille,
Negishi, and Sonogashira coupling reactions, as well as on FIGURE 4. Roles proposed for the external base in the transmetalation
step of Suzuki
Miyaura reaction.
the closely related direct arylation.
Suzuki
Miyaura Reaction. The Suzuki
Miyaura reac-
tion is one of the most used synthetic methods for the SCHEME 4. General Scheme for the Pd-Catalyzed Suzuki
Miyaura
Reaction
construction of biaryls and substituted aromatic moieties.2
This reaction takes place between an organic halide (or
triflate), R1
X, and a boronic acid, R2
B(OH)2, in the pre-
sence of a base (Scheme 4). valid when the PH3 ligands were replaced by PPh3 and when
The role of the external base in the reaction was initially vinyl groups were replaced by phenyl groups.10 This proposal
not clear, and several proposals had been made.28 In has been recently questioned, because pathway B has been
particular, two main roles for this species were proposed postulated to be preferred in at least some cases.30
33 Ongoing
(Figure 4): either the base binds first to the boronic acid to calculations in our groups on extended models suggest that
form the organoboronate species (pathway A), or the base both pathways may be competitive.
substitutes first the leaving group X in the coordination sphere A recurrent issue in the Suzuki
Miyaura reaction is re-
of the catalyst (pathway B). We evaluated theoretically these lated to the number of phosphine ligands (one or two) in the
two pathways as well as a direct mechanism in the absence of a Pd catalytic species, and this issue affects also the transme-
base (pathway 0) for a model system, using trans-[Pd(CH2dCH- talation step, as shown in studies on the full catalytic cycle.9
)(Br)(PH3)2], CH2dCH
B(OH)2, and OH
species as reactants.29 The role of monophosphine complexes was also discussed
In the absence of the base (pathway 0), the reaction was found by Thiel and co-workers in the palladium-catalyzed Suzuki

to be highly endothermic (ca. 32 kcal 3 mol
1) and has a very Miyaura reaction of acetic anhydride with phenylboronic
high energy barrier (ca. 44 kcal 3 mol
1). These results agree with acid.34
the experimental fact that the reaction does not proceed with- Stille Reaction. The Stille reaction is the second most
out addition of a base. broadly studied C
C cross-coupling reaction.35 In this case,
For pathway A, transmetalation was found to be a multi- the reaction takes place between an organic halide (or triflate)
step process resulting in complex trans-[Pd(CH2dCH)2(PH3)2], R1
X and a stannane R2
SnR3, as shown in Scheme 5. In
organoboronic acid B(OH)3, and free halide Br
. The calcu- contrast to the Suzuki
Miyaura reaction, no base is required
lated energy difference between the lowest and highest for the Stille reaction to take place.
energy points was around 20 kcal 3 mol
1, therefore becom- For the transmetalation process involved in the Stille
ing a suitable pathway explaining the role of the base. Our reaction, two main mechanisms dubbed as cyclic mechanism
exploration of pathway B was not so successful using such a and open mechanism (Figure 5) have been proposed to
model system; we were unable to locate a transition state for account for the reported experimental evidence.35 One of
the direct replacement of the halide by the hydroxyl group in the main contributions of computational chemistry has been
the bisphosphine complex. The optimization attempts led the clarification of the competition between the two me-
either to binding of OH
to a phosphine center or to chanisms. DFT studies in our group,36 followed closely in
replacement of one of the phosphines by OH
in the metal

time by those of Alvarez et al.,37 proved that the open
coordination sphere. Hence, it was concluded that the main mechanism will prevail whenever two ancillary ligands,
catalytic cycle should proceed through pathway A, which usually phosphines or arsines, remain attached to palladium
has low energy barriers and no obvious undesired products. in the key transition state, while the cyclic mechanism will
These computational results were found to remain qualitatively dominate when one ligand is released during the process.

Vol. 46, No. 11 ’ 2013 ’ 2626–2634 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2629

Computational Perspective on Pd-Catalyzed C C Reactions García-Melchor et al.

SCHEME 6. General Scheme for the Pd-Catalyzed Negishi Reaction

combination of carbon type (sp, sp2, or sp3) and that toler-

ates many different functional groups at the reagents. This
reaction takes place between an organic halide (or triflate),
R1
X, and an organozinc compound, as shown in Scheme 6.
The main role of computational chemistry in the study of
the Negishi reaction has been to collaborate in the identifi-
cation of the active organozinc species that undergoes
transmetalation and the role of the experimentally observed
FIGURE 5. Proposed mechanisms for the transmetalation step in the palladium intermediates in the reaction pathway. The orga-
Stille reaction: cyclic (top) and open (bottom) mechanisms. nozinc reactant is usually introduced in the form ZnR2 or
ZnRX, but this species can participate in a variety of chemical
equilibria prior to transmetalation because of the preference
SCHEME 5. General Scheme for the Pd-Catalyzed Stille Reaction
of zinc for coordination number four and its affinity for
halide anions. Thus, the presence of species such as ZnRXS2
(S = solvent) cannot be ruled out a priori. The nature of the
Therefore, any feature favoring phosphine dissociation, active species may affect the reaction rate, the role of
such as bulky phosphines, will favor the cyclic mechanism. additives, and the eventual observation of side products.
The open mechanism will be favored by features favoring The first experimental studies on the mechanism of the
dissociation of the X group, because this will render phosphine transmetalation step in the Negishi coupling were reported
dissociation more difficult. This is favored by the presence of by Espinet et al.40 for the transmetalation reaction between
good leaving groups (i.e., triflate) or polar solvents. The theo- trans-[Pd(Rf)(Cl)(PPh3)2] (Rf = 3,5-dichloro-2,4,6-trifluorophenyl)
retical evaluation of the open mechanism showed that the and the organozinc reagents ZnMe2 and ZnMeCl in THF. Later
highest energy barrier in this mechanism corresponds to the on, we reported in a joint investigation a combined experi-
substitution of the L (or Y) ligand by the stannane. mental and theoretical study on the mechanism of the trans-
The cyclic mechanism has been studied in more detail. metalation reaction between ZnMeCl and trans-[Pd(Me)(Cl)-
The process has been shown to proceed in two steps: (PMePh2)2] in THF.41 The mechanistic conclusions are summar-
associative ligand substitution followed by the transmetala- ized in Figure 6. The transmetalation may follow two compe-
tion reaction through a cyclic transition state (Figure 5). This titive pathways: one producing the dimethylated complex
cyclic transition state was found to have the highest energy trans-[Pd(Me)2(PMePh2)2], which is kinetically preferred but
barrier within the overall cyclic mechanism, thus becoming unproductive for coupling, and the other, about 1 order of
the rate-limiting step. The accelerating effect of CsF additives magnitude slower, that affords the complex cis-[Pd(Me)2-
was examined by Ariafard and Yates for the transmetalation (PMePh2)2] from which coupling will eventually take place,
through the cyclic mechanism between (vinyl)SnMe3 and although at a much slower rate. Furthermore, the rate of the
[Pd(Ph)(Cl)(L)n] (L = P(t-Bu)3, PPh3, PPh2Me, PPhMe2, and direct trans-to-cis isomerization was found to be very low com-
PMe3).38 The enhanced reactivity with this additive was pared with the indirect isomerization via retrotransmetalation.
proven to be associated with the formation of the more We also investigated the reaction between trans-[Pd(Me)-
reactive [Pd(Ph)(F)(L)] intermediate. In a recent work in colla- (Cl)(PMePh2)2] and ZnMe2 in THF,42 which was much faster
boration with the experimental group of Espinet, we have than that with ZnMeCl. This rate acceleration was explained
found that gold cocatalysts can accelerate Stille transmeta- by the presence of a minute amount of the nonobserved
lations of bulky organic groups by effectively reducing steric catalytic intermediate trans-[Pd(Me)(PMePh2)2(THF)]þ, which
crowding in the key transition states.39 opens a faster reaction pathway. Transmetalation involving
Negishi Reaction. The Negishi reaction is a reliable cross- zinc out of a traditional cross-coupling context has been also
coupling process that can be applied to every possible computationally explored.43

2630 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2626–2634 ’ 2013 ’ Vol. 46, No. 11

 Computational Perspective on Pd-Catalyzed C C Reactions García-Melchor et al.

SCHEME 8. General Scheme for the Pd-Catalyzed Direct Arylation

Reaction

FIGURE 6. Mechanistic proposal for the Negishi transmetalation be-

tween ZnMeCl and trans-[Pd(Me)(Cl)(PMePh2)2].

SCHEME 7. General Scheme for the Pd-Catalyzed Sonogashira Reaction

Sonogashira Reaction. The Sonogashira reaction is a

widely used method for preparing arylalkynes and conju-
FIGURE 7. Cationic and anionic alternatives for the transmetalation step
gated enynes.44 The general Sonogashira protocol for the
in the Pd-catalyzed Sonogashira reaction via deprotonation mechanism.
reaction of terminal alkynes with aryl or alkenyl halides (or
triflates) involves a Pd(0)/Cu(I) catalytic system and at least a pathway that we labeled as ionic mechanism. In this mechan-
stoichiometric amount of a base (Scheme 7). The presence of ism the base has two roles: substituting the halide in the
a Cu(I) salt is generally believed to facilitate the transfer of coordination sphere of the catalyst and helping the formation
the alkynyl group to the Pd catalyst by the in situ generation of acetylide species by subtracting the proton. Our theoretical
of a copper acetylide species and the subsequent transme- results showed that this mechanism is competitive with the
talation of this group to Pd. In some cases, however, Cu(I) cationic and anionic alternatives and that may lead to higher
salts have been shown to have an inhibitory effect in the reaction rates with alkynes bearing electron-withdrawing
Sonogashira reaction.45,46 In recent years, many efforts groups, in agreement with experiments.
have been devoted to develop reaction procedures working Direct Arylation. Direct arylation is a reaction formally
in the absence of copper salts; these strategies are com- similar to cross-coupling, with the key difference that one of
monly referred as copper-free Sonogashira reaction. Compu- the carbons involved in the process is bound not to a
tational studies have mainly focused on the mechanism of heteroatom but to hydrogen (Scheme 8). This turns this
this copper-free process. reaction into a very powerful method for the synthesis of
The Sonogashira transmetalation has been recently aryl
aryl bonds, since it allows skipping the introduction of
shown to proceed through the so-called deprotonation the electropositive heteroatom as required in cross-coupling
mechanism (Figure 7).47 An external base is required to reactions.49 Direct arylation is often considered as a different
deprotonate the terminal alkyne, transforming into an alkynyl process from cross-coupling, but we have decided to include
ready to undergo reductive elimination with the other R it in this Account because of their mechanistic similarities.
group attached to the metal. This mechanism is known to From a computational point of view, the key issue is the
entail ligand exchange in the metal coordination sphere, nature of the metalation step, where the R2 group moves
and the nature of these ligand exchanges has been clarified from a R2
H bond to a M
R2 bond. The metalation in
by calculations.48 Depending on the order of these ligand palladium-catalyzed arylation involves the metal in its diva-
exchanges and the deprotonation, two different alternatives lent state. It has been applied in ring closure reactions50 and
had been proposed from the experimental work, labeled as intermolecular arylation.51 Calculations have shown that
cationic and anionic (Figure 7). In our theoretical evaluation of the reaction proceeds without direct interaction between
the reaction mechanisms, we could identify a third reaction the metal and the proton. The hydrogen atom is captured by

Vol. 46, No. 11 ’ 2013 ’ 2626–2634 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2631

Computational Perspective on Pd-Catalyzed C C Reactions García-Melchor et al.

FIGURE 8. Transition state for the metalation step in the Pd-catalyzed direct arylation assisted by HCO3
.

a base (e.g., carbonate, hydrogenocarbonate), with simulta-

neous formal generation of a carbanion that binds to the
metal center. A typical transition state is shown in Figure 8.
The mechanism has been labeled as concerted metala-
tion
deprotonation (CMD).51 Similar mechanisms have
been found to operate in ruthenium-catalyzed52 and in FIGURE 9. Generally accepted mechanism for the reductive elimina-
tion step.
iridium-catalyzed processes.53,54
This mechanism is to a certain extent surprising because a

In collaboration with Espinet and Alvarez et al.,58 we
relatively weak base such as carbonate is able to take away reported a combined experimental and theoretical study
an arylic proton, which has a very low acidity. The presence on the C
C coupling from complexes cis-[Pd(R)2(PMe3)(L)].
of the metal in this case is crucial to stabilize the develop- This work allowed us to characterize the effect of the
ment of a negative charge in the carbon center, as proven ancillary ligand L on the reaction barrier. The barrier was
by the structure of the transition states, where the metal
shown to increase in the order maleic anhydride < “empty” <
carbon bond is almost completely formed. ethylene < PMe3 ≈ MeCN. For four-coordinate complexes,
the barrier decreases with the increasing π-acceptor ability
The Reductive Elimination Step of L; thus additives with strong π-accepting abilities will
In order to complete the catalytic cycle, the reductive elim- accelerate the reaction. The barrier for the three-coordinate
ination step after the transmetalation process is needed. complex (L = “empty”) is lower than that for most four-
This step consists of the coupling of the two organic groups, coordinate systems, and because of that, weakly coordinat-
R1 and R2, in cis position (Figure 9). In the case in which the ing ligands will dissociate prior to reductive elimination. We
complex resulting from transmetalation has the R1 and R2 completed this study with an analysis of reductive elimina-
groups trans to each other, a trans to cis isomerization tion in cis-[Pd(η1-allyl)2(PMe3)(L)],59 where we found the
reaction previous to the reductive elimination is required. Csp2
C0 sp2 elimination to be the most favored one, in agree-
The generally accepted mechanism for this step is concerted ment with previous experimental results.
and features a cyclic three-coordinated transition state that
results in the final C
C coupling and the concomitant Concluding Remarks
regeneration of the catalytic species, [PdL2] (Figure 9). The Cross-coupling is a complex and diverse multistep catalytic
reductive elimination is usually irreversible. process, and because of this, it is difficult to fully characterize
Although the general problem of reductive elimination of the mechanism from pure experimental techniques. Com-
H
H and C
H bonds was among the first topics analyzed in putational chemistry has proven to be a powerful tool in this
the computational study of transition metal complexes,55 context and has helped to shape our current understanding
its direct connection with cross-coupling was not explored of the mechanism. The collaboration between experiment
in detail until the work by Ananikov, Musaev, and and theory has led to the construction of a general scheme
Morokuma56 on cis-[Pd(R1)(R2)(PH3)2] complexes (R1,R2 = where seemingly contradictory results can be explained, a
Me, vinyl, Ph, ethynyl). They found that the barriers increase framework that is already making its way into textbooks.
in the order vinyl < Ph < ethynyl < Me. The effect of the bite Much of this knowledge has been gained from calculations
angle in the case of diphosphine ligands was also analyzed on model systems. However, calculations themselves have
by Bo et al.57 proven that in many cases competitive pathways exist with

2632 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2626–2634 ’ 2013 ’ Vol. 46, No. 11

 Computational Perspective on Pd-Catalyzed C C Reactions García-Melchor et al.

close energy barriers. The choice of a particular pathway will catalysis. Full Professor of Physical Chemistry at UAB since 1994 and
then depend on the particular reaction conditions (e.g., Group Leader of the “Molecular Modeling of Catalytic Systems” group,
ligands, substrates, solvent). Most major mechanistic problems his research focuses on the computational modeling of the organo-
metallic reactivity and homogeneous catalysis processes, with em-
in C
C cross-coupling seem to be already solved, but there is
phasis on reaction mechanisms. Recipient of the “Real Sociedad
still a role for computational chemistry. Challenges remain in Espan~ola de Química” RSEQ-Bruker Award for Physical Chemistry in
the study of particular systems to further expand the scope of 2008, he has been the coordinator of the Chemistry area of the
these fascinating reactions, and there is always the intriguing Spanish Agency of Research Evaluation (ANEP) from 2008 to 2011.
possibility of discovering new reagents for transmetalation. He is the vice-president of the Spanish Chemical Society (RSEQ).
Computational Chemistry is now sufficiently mature to assist Gregori Ujaque was born in Torres de Segre (Lleida, Catalonia,
experimental chemistry in these future developments. Spain) in 1971. He obtained his Ph.D. in Chemistry from the
noma de Barcelona (UAB) in 1999. He then spent
Universitat Auto
two years at the University of California, Los Angeles (UCLA), with
We thank the Spanish MINECO (Grants CTQ2011-23336,
Prof. K.N. Houk working on rational design of catalytic antibodies
CTQ2011-27033, Consolider Ingenio 2010 CSD2006-0003, and
and organic reactivity using computational chemistry. He returned
Consolider Ingenio 2010 CSD2007-0006), Generalitat de Catalu- to UAB contracted by the “Ramon y Cajal” program, obtaining a
nya (Grants 2009SGR-0068, 2009SGR-2059, and Xarxa de permanent position in 2007. His main interests are the application
Referencia en Química Teo rica i Computacional), and the ICIQ of computational methods to the understanding of homogeneous
Foundation for financial support. We sincerely thank all our students catalysis, using QM and QM/MM methods and more recently
applying AIMD and QM/MM-MD methods. Among other catalytic
and other members in our groups that have worked with us in
processes, he enjoys working on cross-coupling, hydrogenation,
this field throughout the years, notably Maria Besora, Christophe
hydroamination, and gold-catalyzed reactions.
Gourlaouen, Carina Ba €cktorp, Ainara Nova, W. M. C. Sameera, Jesu
s
Feliu Maseras was born in Martorelles (Catalonia) in 1962. He
Jover, David Balcells, Torstein Fjermestad, Charles Goehry, Salvador obtained a Ph.D. in Chemistry from the Universitat Auto noma de
Moncho, and Manuel A. Ortun ~o. We also thank all of our experi- Barcelona (UAB) in 1991. He has worked at the UAB, Institute for
mental collaborators, notably the groups of Pablo Espinet, Antonio Molecular Science (Okazaki, Japan), Universite
de Montpellier 2
Echavarren, Gregorio Asensio, Carmen Na
jera, and Pierre Dixneuf. (France) and Emory University (Atlanta, GA, US). He obtained his
current position as Group Leader in the Institute of Chemical
Research of Catalonia (ICIQ) in 2004. His research focuses in the
BIOGRAPHICAL INFORMATION design and application of the methods of quantum mechanics
Max García-Melchor was born in Barcelona (Spain) in 1982. He (QM) and quantum mechanics/molecular mechanics (QM/MM) to
completed his B.Sc. in Chemistry (2006) and his Master in Compu- chemical problems of practical interest, with particular emphasis in
tational Chemistry (2008) at the Universitat Autonoma de Barce- topics of homogeneous catalysis. Receptor of the Chemical Society
lona (UAB). In 2012, he received his Ph.D. in Chemistry from the Reviews Award in 2008, and of the “Real Sociedad Espan ~ola de
UAB under the supervision of Profs A. Lledo
s and G. Ujaque and Química” RSEQ-Bruker Award for Physical Chemistry in 2011.
joined the group of Prof. N. Lo
pez at the Institute of Chemical
Research of Catalonia (ICIQ) as a postdoctoral fellow. His main FOOTNOTES
research interests involve the use of computational methods to *Corresponding authors. E-mail addresses: [email protected]; [email protected].
understand and develop homogeneous and heterogeneous cata- The authors declare no competing financial interest.
lytic processes, such as those involving C
C cross-coupling reac-
tions or the production of chemicals from biomass. REFERENCES
Ataualpa A. C. Braga was born in Foz do Iguac-u (Brazil) in 1976. 1 Metal-Catalyzed Cross-Coupling Reactions, 2nd ed.; de Meijere, A., Diedrich, F., Eds.;
Wiley-VCH: Weinheim, Germany, 2004.
He obtained his Ph.D. in Chemistry from Universidade Estadual de
2 Miyaura, N.; Suzuki, A. Palladium-catalyzed cross-coupling reactions of organoboron
Campinas (UNICAMP) in 2004. In 2005, he joined the Feliu compounds. Chem. Rev. 1995, 95, 2457–2483.
Maseras group at the Institute of Chemical Research of Catalonia 3 Stille, J. K. The palladium-catalyzed cross-coupling reaction of organotin reagents with
(ICIQ). He returned to UNICAMP in 2011, obtaining a permanent organic electrophiles. Angew. Chem., Int. Ed. 1986, 25, 508–524.
position in 2012 at Universidade de Sa~o Paulo (USP). His main field 4 Negishi, E.; King, A. O.; Okukado, N. Selective carbon-carbon bond formation via transition-
metal catalysis 0.3. Highly selective synthesis of unsymmetrical biaryls and diarylmethanes
of interest is the application of methods of computational chem- by nickel-catalyzed or palladium-catalyzed reaction of aryl derivatives and benzylzinc
istry to the theoretical study of homogeneous catalysis. derivatives with aryl halides. J. Org. Chem. 1977, 42, 1821–1823.

s was born in Barcelona (Spain) in 1955. He 5 Buchwald, S. L. Cross-Coupling (Guest Editorial). Acc. Chem. Res. 2008, 41, 1439–1564.
Agustí Lledo
6 Braga, A. A. C.; Ujaque, G.; Maseras, F. The mechanism of palladium-catalyzed cross-
obtained a Ph.D. in Chemistry from the Universitat Auto noma de coupling reactions. In Computational Modeling for Homogeneous and Enzymatic Catalysis.


Barcelona (UAB) in 1984. After a postdoctoral stay at the Universite A Knowledge-Base for Designing Efficient Catalysis; Morokuma, K., Musaev, D. G., Eds.;
Wiley-VCH, Weinheim, Germany, 2008; Chapter 5.
de Paris-Sud (Orsay, France), he returned to UAB where he started a
7 García-Melchor, M.; Ujaque, G.; Maseras, F.; Lled
os, A. Theoretical Evaluation of Phosphine
research group focused on the application of computational chem- Effects in Cross-Coupling Reactions. In Phosphorus Compounds: Advanced Tools in Catalysis
istry to the fields of organometallic chemistry and homogeneous and Material Sciences; Peruzzini, M., Gonsalvi, L., Eds.; Springer: Berlin, 2011; Chapter 3.

Vol. 46, No. 11 ’ 2013 ’ 2626–2634 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2633

Computational Perspective on Pd-Catalyzed C C Reactions García-Melchor et al.

8 Xue, L.; Lin, Z. Theoretical aspects of palladium-catalysed carbon
carbon cross-coupling 36 Nova, A.; Ujaque, G.; Maseras, F.; Lled
os, A.; Espinet, P. A critical analysis of the cyclic and
reactions. Chem. Soc. Rev. 2010, 39, 1692–1705. open alternatives of the transmetalation step in the Stille cross-coupling reaction. J. Am.
9 Braga, A. A. C.; Ujaque, G.; Maseras, F. A DFT study of the full catalytic cycle of the Suzuki
Chem. Soc. 2006, 128, 14571–14578.
Miyaura cross-coupling on a model system. Organometallics 2006, 25, 3647–3658. 37
Alvarez, R.; Faza, O. N.; Lopez, C. S.; de Lera, A. R. Computational characterization of a
10 Braga, A. A. C.; Morgon, N. H.; Ujaque, G.; Lled
os, A.; Maseras, F. Computational study of complete palladium-catalyzed cross-coupling process: The associative transmetalation in
the transmetalation process in the Suzuki
Miyaura cross-coupling of aryls. J. Organomet. the Stille reaction. Org. Lett. 2006, 8, 35–38.
Chem. 2006, 691, 4459–4466. 38 Ariafard, A.; Yates, B. F. Subtle balance of ligand steric effects in Stille transmetalation.
11 Besora, M.; Braga, A. A. C.; Ujaque, G.; Maseras, F.; Lled
os, A. The importance of J. Am. Chem. Soc. 2009, 131, 13981–13991.
conformational search: A test case on the catalytic cycle of the Suzuki
Miyaura cross- 39 delPozo, J.; Carrasco, D.; P
erez-Temprano, M. H.; García-Melchor, M.;
Alvarez, R.;
coupling. Theor. Chem. Acc. 2011, 128, 639–646. Casares, J. A.; Espinet, P. Stille coupling involving bulky groups feasible with gold
12 Jover, J.; Fey, N.; Purdie, M.; Lloyd-Jones, G. C.; Harvey, J. N. A computational study of cocatalyst. Angew. Chem., Int. Ed. 2013, 52, 2189–2193.
phosphine ligand effects in Suzuki
Miyaura coupling. J. Mol. Catal. A 2010, 324, 39–47. 40 Casares, J. A.; Espinet, P.; Fuentes, B.; Salas, G. Insights into the mechanism of
13 Kozuch, S.; Martin, J. M. L. What makes for a bad catalytic cycle? A theoretical study on the the Negishi reaction: ZnRX versus ZnR2 reagents. J. Am. Chem. Soc. 2007, 129,
Suzuki
Miyaura reaction within the energetic span model. ACS Catal. 2011, 1, 246–253. 3508–3509.
14 Chass, G. A.; O'Brien, C. J.; Hadei, N.; Kantchev, E. A. B.; Mu, W.-H.; Fang, D.-C.; 41 Fuentes, B.; García-Melchor, M.; Casares, J. A.; Ujaque, G.; Lled
os, A.; Maseras, F.;
Hopkinson, A. C.; Csizmadia, I. G.; Organ, M. G. Density functional theory investigation of Espinet, P. Palladium round trip in the Negishi coupling of trans-[PdMeCl(PMePh2)2] with
the alkyl-alkyl Negishi cross-coupling reaction catalyzed by N-Heterocyclic carbene (NHC)- ZnMeCl: An experimental and DFT study of the transmetalation step. Chem.;Eur. J. 2010,
Pd complexes. Chem.;Eur. J. 2009, 15, 4281–4288. 16, 8596–8599.
15 Balcells, D.; Maseras, F.; Keay, B. A.; Ziegler, T. Polyene cyclization by a double 42 García-Melchor, M.; Fuentes, B.; Lled
os, A.; Casares, J. A.; Ujaque, G.; Espinet, P. Cationic
intramolecular Heck reaction. A DFT study. Organometallics 2004, 23, 2784–2796. intermediates in the Pd-catalyzed Negishi coupling. Kinetic and DFT study of alternative
transmetalation pathways in the Me-Me coupling of ZnMe2 and trans-[PdMeCl(PMePh2)2].
16 Smith, G. B.; Dezeny, G. C.; Hughes, D. L.; King, A. O.; Verhoeven, T. R. Mechanistic studies
J. Am. Chem. Soc. 2011, 133, 13519–13526.
of the suzuki cross-coupling reaction. J. Org. Chem. 1994, 59, 8151–8156.
43 Jimeno, C.; Sayalero, S.; Fjermestad, T.; Colet, G.; Maseras, F.; Pericas, M. A. Under-
17 Christmann, U.; Vilar, R. Monoligated palladium species as catalysts in cross-coupling
standing the boron-to-zinc transmetallation. Practical implications for the catalytic asym-
reactions. Angew. Chem., Int. Ed. 2005, 44, 366–374.
metric arylation of aldehydes. Angew. Chem., Int. Ed. 2008, 47, 1098–1101.
18 Senn, H. M.; Ziegler, T. Oxidative addition of aryl halides to palladium(0) complexes: A
44 Chinchilla, R.; N
ajera, C. Recent advances in Sonogashira reactions. Chem. Soc. Rev.
density-functional study including solvation. Organometallics 2004, 23, 2980–2988.
2011, 40, 5084–5121.
19 Lyngvi, E.; Schoenebeck, F. Oxidative addition transition states of Pd(0) complexes in polar
45 Gelman, D.; Buchwald, S. L. Efficient palladium-catalyzed coupling of aryl chlorides and
solvent
a DFT study involving implicit and explicit solvation. Tetrahedron 2013, 69,
tosylates with terminal alkynes: Use of a copper cocatalyst inhibits the reaction. Angew.
5715–5718.
Chem., Int. Ed. 2003, 42, 5993–5996.
20 Kozuch, S.; Amatore, C.; Jutand, A.; Shaik, S. What makes for a good catalytic cycle? A
46 Aufiero, M.; Proutiere, F.; Schoenebeck, F. Redox reactions in palladium catalysis: On the
theoretical study of the role of an anionic palladium(0) complex in the cross-coupling of an
accelerating and/or inhibiting effects of copper and silver salt additives in cross-coupling
aryl halide with an anionic nucleophile. Organometallics 2005, 24, 2319–2330.
chemistry involving electron-rich phosphine ligands. Angew. Chem., Int. Ed. 2012, 51,
21 Goossen, L. J.; Koley, D.; Hermann, H. L.; Thiel, W. Mechanistic pathways for oxidative 7226–7230.
addition of aryl halides to palladium(0) complexes: A DFT study. Organometallics 2005, 24,
47 Ljungdahl, T.; Bennur, T.; Dallas, A.; Emten€as, H.; Martensson, J. Two competing
2398–2410.
mechanisms for the copper-free Sonogashira cross-coupling reaction. Organometallics
22 Proutiere, F.; Schoenebeck, F. Solvent effect on palladium-catalyzed cross-coupling 2008, 27, 2490–2498.
reactions and implications on the active catalytic species. Angew. Chem., Int. Ed. 2011, 50,
48 García-Melchor, M.; Pacheco, M. C.; N
ajera, C.; Lled
os, A.; Ujaque, G. Mechanistic
8192–8195.
exploration of the Pd-catalyzed copper-free Sonogashira reaction. ACS Catal. 2012, 2,
23 Gourlaouen, C.; Ujaque, G.; Lled
os, A.; Medio-Sim
on, M.; Asensio, G.; Maseras, F. Why is 135–144.
the Suzuki
Miyaura cross-coupling of sp3 carbons in R-bromo sulfoxide systems fast and
49 Alberico, D.; Scott, M. E.; Lautens, M. Aryl
aryl bond formation by transition-metal-
stereoselective? A DFT study on the mechanism. J. Org. Chem. 2009, 74, 4049–4054.
catalyzed direct arylation. Chem. Rev. 2007, 107, 174–238.
24 Besora, M.; Gourlaouen, C.; Yates, B.; Maseras, F. Phosphine and solvent effects on
50 García-Cuadrado, D.; De Mendoza, P.; Braga, A. A. C.; Maseras, F.; Echavarren, A. M.
oxidative addition of CH3Br to Pd(PR3) and Pd(PR3)2 complexes. Dalton Trans. 2011, 40,
Proton-abstraction mechanism in the palladium-catalyzed intramolecular arylation: Sub-
11089–11094.
stituent effects. J. Am. Chem. Soc. 2007, 129, 6880–6886.
25 McMullin, C. L.; Jover, J.; Harvey, J. N.; Fey, N. Accurate modeling of Pd(0) þ PhX oxidative
51 Lafrance, M.; Rowley, C. N.; Woo, T. K.; Fagnou, K. Catalytic intermolecular direct arylation
addition kinetics. Dalton Trans. 2010, 39, 10833–10836.
of perfluorobenzenes. J. Am. Chem. Soc. 2006, 128, 8754–8756.
26 Mollar, C.; Besora, M.; Maseras, F.; Asensio, G.; Medio-Sim
on, M. Competitive and
52 Ozdemir, I.; Demir, S.; Cetinkaya, B.; Gourlaouen, C.; Maseras, F.; Bruneau, C.; Dixneuf,
selective Csp3-Br versus Csp2-Br bond activation in palladium-catalysed Suzuki cross-
P. H. Direct arylation of arene C
H bonds by cooperative action of
coupling: An experimental and theoretical study of the role of phosphine ligands. Chem.;
NHCarbene
Ruthenium(II) catalyst and carbonate via proton abstraction mechanism.
Eur. J. 2010, 16, 13390–13397.
J. Am. Chem. Soc. 2008, 130, 1156–1157.
27 Schoenebeck, F.; Houk, K. N. Ligand-controlled regioselectivity in palladium-catalyzed
53 Boutadia, Y.; Davies, D. L.; Macgregor, S. A.; Poblador-Bahamonde, A. I. Mechanisms of
cross coupling reactions. J. Am. Chem. Soc. 2010, 132, 2496–2497.
C
H bond activation: Rich synergy between computation and experiment. Dalton Trans.
28 Miyaura, N. Cross-coupling reaction of organoboron compounds via base-assisted 2009, 30, 5820–5831.
transmetalation to palladium(II) complexes. J. Organomet. Chem. 2002, 653, 54–57.
54 García-Melchor, M.; Gorelsky, S. I.; Woo, T. K. Mechanistic analysis of iridium(III) catalyzed
29 Braga, A. A. C.; Morgon, N. H.; Ujaque, G.; Maseras, F. Computational characterization of direct sp2 C
H arylation: A DFT study. Chem.;Eur. J. 2011, 17, 13847–13853.
the role of the base in the Suzuki
Miyaura cross-coupling reaction. J. Am. Chem. Soc.
55 Low, J. J.; Goddard, W. A. Theoretical studies of oxidative addition and reductive
2005, 127, 9298–9307.
elimination. 3. C
H and C
C reductive coupling from palladium and platinum
30 Amatore, C.; Jutand, A.; Le Duc, G. Kinetic data for the transmetalation/reductive bis(phosphine) complexes. J. Am. Chem. Soc. 1986, 108, 6115–6128.
elimination in palladium-catalyzed Suzuki
Miyaura reactions: Unexpected triple role of
56 Ananikov, V. P.; Musaev, D. G.; Morokuma, K. Theoretical insight into the C
C coupling
hydroxide ions used as base. Chem.;Eur. J. 2011, 17, 2492–2503.
reactions of the vinyl, phenyl, ethynyl, and methyl complexes of palladium and platinum.
31 Carrow, B. P.; Hartwig, J. F. Distinguishing between pathways for transmetalation in Organometallics 2005, 24, 715–723.
Suzuki
Miyaura reactions. J. Am. Chem. Soc. 2011, 133, 2116–2119.
57 Zuidema, E.; van Leeuwen, P. W. N. M.; Bo, C. Reductive elimination of organic
32 Butters, M.; Harvey, J. N.; Jover, J.; Lennox, A. J. J.; Lloyd-Jones, G. C.; Murray, P. M. Aryl molecules from palladium
diphosphine complexes. Organometallics 2005, 24,
trifluoroborates in Suzuki
Miyaura coupling: The roles of endogenous aryl boronic acid and 3703–3710.
fluoride. Angew Chem., Int. Ed. 2010, 49, 5156–5160.
58 P
erez-Rodríguez, M.; Braga, A. A. C.; García-Melchor, M.; P
erez-Temprano, M.; Casares,
33 Lennox, A. J. J.; Lloyd-Jones, G. C. Transmetalation in the Suzuki
Miyaura coupling: The J. A.; Ujaque, G.; de Lera, A. R.;
Alvarez, R.; Maseras, F.; Espinet, P. C
C reductive
fork in the trail. Angew. Chem., Int. Ed. 201310.1002/anie.201301737. elimination in palladium complexes, and the role of coupling additives. A DFT study
34 Goossen, L. J.; Koley, D.; Hermann, H. L.; Thiel, W. Palladium monophosphine intermediates supported by experiment. J. Am. Chem. Soc. 2009, 131, 3650–3657.
in catalytic cross-coupling reactions: a DFT study. Organometallics 2006, 25, 54–67. 59 P
erez-Rodríguez, M.; Braga, A. A. C.; de Lera, A.; Maseras, F.;
Alvarez, R.; Espinet, P. A
35 Espinet, P.; Echavarren, A. M. The mechanisms of the Stille reaction. Angew. Chem., Int. DFT study of the effect of the ligands in the reductive elimination from palladium bis(allyl)
Ed. 2004, 43, 4704–4734. complexes. Organometallics 2010, 29, 4983–4991.

2634 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 2626–2634 ’ 2013 ’ Vol. 46, No. 11