Evidence for a Cooperative Mechanism Involving Two Palladium(0) Centers in the Oxidative Addition of Iodoarenes

Article abstract of DOI:10.1002/chem.201704899

Oxidative addition of iodoarenes (ArI) to Pd⁰ ligated by 1-methyl-1H-imidazole (mim) in polar solvents leads to cationic arylpalladium(II) complexes [ArPd(mim)₃]⁺. Kinetic analyses evidence that this reaction is second order with respect to the concentration of Pd⁰, and a mechanism involving the cooperative intervention of two Pd⁰ centers has been postulated to explain this finding. This unusual behavior is also observed with other nitrogen-containing ligands and it is general for iodobenzenes substituted with electron-donating or weakly electron-withdrawing groups. In contrast, bromoarenes and electron-poor iodoarenes display first-order kinetics with respect to Pd⁰. Theoretical calculations performed at the density functional theory (DFT) level suggest the existence of mim-ligated ArI-Pd⁰ complexes, in which the iodoarene is bound to the metal in a halogen-bond-like fashion. Coordination weakens the C?I bond and facilitates the oxidative insertion of another Pd⁰ center across this C?I bond. This conceptually novel mechanism, involving the cooperative participation of two distinct metal centers, allows a full explanation of the experimentally observed kinetics.

Full text of DOI:10.1002/chem.201704899

A Journal of
Accepted Article
Title: Evidence for a Cooperative Mechanism Involving two
Palladium(0) Centers in the Oxidative Addition of Iodoarenes
Authors: Laurence Grimaud, Ilaria Ciofini, Luca Alessandro Perego,
Pierre-Adrien Payard, and Baptiste Haddou
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201704899
Link to VoR: http://dx.doi.org/10.1002/chem.201704899
Supported by

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
Evidence for a Cooperative Mechanism Involving two
Palladium(0) Centers in the Oxidative Addition of Iodoarenes
Luca Alessandro Perego,*[a,b] Pierre-Adrien Payard,[a] Baptiste Haddou,[a] Ilaria Ciofini,*[b] and Laurence
Grimaud*[a]
Abstract: Oxidative addition of iodoarenes (ArI) to Pd(0) ligated by
1-methyl-1H-imidazole (mim) in polar solvents leads to cationic
arylpalladium(II) complexes [ArPd(mim)3]+. Kinetic analyses evidence
that this reaction is second order with respect to the concentration of
Pd(0) itself, and a mechanism involving the cooperative intervention
of two Pd(0) centers has been postulated to explain this finding. This
unusual behavior is also observed with other nitrogen-containing
ligands and it is general for iodobenzenes substituted with electron-
donating or weak electron-withdrawing groups. By contrast,
bromoarenes and electron-poor iodoarenes display first order kinetics
with respect to Pd(0). Theoretical calculations performed at the
density functional theory (DFT) level suggest the existence of mim-
ligated ArI-Pd(0) complexes in which the iodoarene is bound to the
metal in a halogen bond-like fashion. Coordination weakens the C-I
bond and facilitates the oxidative insertion of another Pd(0) center
across the C-I bond itself. This conceptually novel mechanism,
involving the cooperative participation of two distinct metal centers,
allows to fully explain the experimentally observed kinetics.
except for the special case of very hindered phosphine ligands,
such as P(o-Tol)3 or P(t-Bu)3, for which the active species is the
mono-ligated [Pd(0)L].[7] Analogous results have been obtained
for bidentate phosphines.[8]
Other species commonly present in catalytic cross-coupling
protocols can associate with Pd(0) and influence the rate and the
mechanism of OA. Electron-deficient olefins, which are the typical
substrates of Heck reactions, decrease the rate of the OA of aryl
halides.[9] Similarly, trans,trans-dibenzylidene acetone (dba),
introduced with the popular air-stable pre-catalyst Pd2(dba)3,
slows down OA by the formation of poorly reactive heteroleptic
dba-phosphine Pd(0) complexes.[10] Common anions, such as
Cl–,[11] AcO–[12] and CF3COO–[13] can also have an influence by
associating with Pd(0).[14] The OA involving Pd(0) complexes
supported by ligands other than phosphines, such as N-
heterocyclic carbenes[15] and isocyanides[16] has also been
studied.
Despite this profusion of very thorough studies, nitrogen
ligands have been little considered in this field.[17] However,
nitrogen-containing heteroarenes, such as imidazoles, have been
used as ligands for some Pd-catalyzed cross-couplings.[18]
Moreover, amines are classically used in a stoichiometric amount
as bases in several Pd-catalyzed transformations, such as Heck
or Sonogashira couplings, being therefore potential ligands for the
palladium center. Besides, nitrogen heterocycles may be used as
substrates for direct arylation reactions.[19] In the context of a
mechanistic study on the C-H activation of imidazoles under
phosphine-free conditions,[20] we recently became interested in
the OA reaction of imidazole-ligated Pd(0) complexes with aryl
halides. We report here a detailed analysis of the latter reaction
(Scheme 1), which evidences an unprecedented second-order
dependence of reaction rate with respect to the concentration of
Pd(0). This observation could be generalized to several other
nitrogen-based ligands. Density functional theory (DFT) studies
on the mechanism of this reaction allowed interpreting the
experimental kinetics in term of the cooperative intervention of two
Pd(0) centers: one weakens the C-I bond by coordination in a
halogen-bond fashion, while the second is responsible for the
actual oxidative insertion (Scheme 1).
Introduction
Oxidative addition (OA) is an elementary step that lies at the very
heart of the transition metal-mediated activation of organic
electrophiles.[1] OA is a fundamental part of the catalytic cycles of
several widely used Pd-catalyzed transformations, such as
catalytic cross-couplings.[2]
Mechanistic studies on the Pd(0)-to-Pd(II) OA have a decade-
long history. Building on Fitton’s pioneering studies on the
reaction of halogenoarenes with Pd(PPh3)4,[1d, 3] Fauvarque first
analyzed in detail the kinetics of this transformation[4] and
introduced the use of electroanalytical techniques in the study of
OA, which later proved to be very useful in this field.[5] The main
deduction drawn from these early researches is that the active
species is usually the 14-electron complex [Pd(0)L2] (L =
monodentate phosphine).[4] This conclusion is generally valid,[6]
[a]
[b]
L. A. Perego, P.-A. Payard, B. Haddou, Dr. L. Grimaud
PASTEUR, Département de chimie
École normale supérieure, PSL Research University, Sorbonne
Universités, UPMC Univ. Paris 06, CNRS
75005 Paris (France)
E-mail: luca.perego@ens.fr, laurence.grimaud@ens.fr
L. A. Perego, Dr. I.Ciofini
PSL Research University, Institut de Recherche de Chimie Paris
IRCP, CNRS-Chimie ParisTech
11 rue P. et M. Curie, F-75005 Paris 05 (France)
E-mail: ilaria.ciofini@chimie-paristech.fr
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
with [Pd(0)] the concentration of Pd(0) species in the mixture, and
α and β the partial orders with respect to Pd(0) and ArI.
Under the experimental conditions (excess ArI), [ArI] does not
vary much and it can be considered constant, thus the rate law
can be simplified as follows:
훼
β
[
]
[
]
푟 ≈ 푘app Pd(0)
with 푘app = 푘 ArI
Given that the conductivity (κ) is proportional to the
concentration of the ionic products, simple calculations (see
Supporting information, §2.1) lead to the following laws for the
time evolution of conductivity:[23]
Scheme 1. Oxidative addition of aryl halides to Pd(0) by a cooperative and a
non-cooperative mechanism and associated expressions for the reaction rate
(r).
1st order in Pd(0) (α = 1):
휅 = 휅∞(1 − 푒−푘
푡) (i)
app,1
푐 푘
푡
0
app,2
2nd order in Pd(0) (α = 2):
휅 = 휅∞
(ii)
1+푐 푘
푡
0
app,2
in which 휅∞ is the final conductivity and c0 the initial
concentration of Pd(0) species.
A typical kinetic curve obtained for the model system
described in Scheme 2 is presented in Figure 1.
Results and Discussion
Studies on the model system
The typical experiment performed in this study is presented in
Scheme 2. As in our earlier investigations,[20] imidazole-ligated
Pd(0) was generated in situ by treating Pd(dba)2[21] in acetone with
an excess of an imidazole derivative. The heterocycle displaces
dba from the coordination sphere of Pd(0) within a few minutes,
as we verified previously for 1,2-dimethyl-1H-imidazole.[20]
Kinetic experiments were performed by the flooding method: a
large excess of the aryl iodide (ArI, typically >10 equivalents with
respect to the palladium complex) was added to a solution
containing Pd(dba)2 and 40 equivalents of 1-methyl-1H-imidazole
(mim), which we chose as the reference ligand.
Experimental data
Fitting according to equation (ii) (2nd order)
180
Fitting according to equation (i) (1st order)
160
140
120
100
80
50
40
30
The [Pd(mim)n] species so formed are able to react with ArX (X =
I, Br) to give cationic organometallic complexes [ArPd(mim)3]+ and
free halide in polar solvents. The formation of these products was
already reported by us and we isolated and characterized an
analogous complex featuring 1,2-dimethyl-1H-imidazole as
ligand.[20]
Since ionic species are formed, we resorted to conductivity
measurements to follow the kinetics of the reaction: after the
addition of ArX to the Pd(dba)2/mim solution, conductivity was
recorded versus time in a temperature-controlled cell under Ar
atmosphere.[22] For experimental convenience, the kinetic runs
were performed at 268 K (if not differently stated). At this
temperature, the transformations were usually complete in the
time frame of a few minutes.
60

20

40
Experimental data
Linear fit
10
0
20
0
0.00
600
0.05
800
0.10
0.15
0.20
t-1 (s-1)
-20
0
200
400
1000
1200
1400
t (s)
Figure 1. Kinetics of the reaction of Pd(dba)2 (2.0 mM in acetone) with 4-
iodoanisole (20 mM, 10 equiv) in the presence of mim (80 mM, 40 equiv), as
monitored by conductometry. Experimental conductivities are reported as a
function of time (○), as well as curves fitted according to equations (i) and (ii).
Inset: reciprocal plot and linear fitting, illustrating the validity of equation (ii).[24]
Fitting of equations (i) and (ii) to the experimental kinetic curve
demonstrated that only equation (ii) is suitable for the case under
study, i.e. the reaction is second-order with respect to Pd(0)
(Figure 1, see the Supporting Information, §2.2, for analysis of the
residuals). Moreover, a plot of κ–1 as a function of t–1 proved to be
linear, in agreement with equation (ii) (Figure 1, inset).[24] This
result was unexpected, given that to the best of our knowledge all
the oxidative addition of organic electrophiles to Pd(0) reported in
the literature so far are first-order in [Pd(0)] itself.
The order of reaction in the iodoarene was also determined
from the apparent second-order rate constant kapp,2 obtained with
various amounts of ArI (Ar = 4-MeO-C6H4) (Figure 2, left). The
expected first order in ArI was obtained. Consequently, only one
Scheme 2. Typical kinetic experiment.
The rate of the reaction (r) is expected to depend on the
concentration of the reactants according to an equation of the
form:
β
[
]α[
]
푟 = 푘 Pd(0) ArI
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
molecule of ArI should be involved in the rate-determining
elementary step of the reaction under study.
Table 1. Apparent second order rate constant kapp,2 for different solvents.[a]
Entry
Solvent
Dipolar
moment
(D)[b]
Dielectric
constant[b]
kapp,2
(M–1 s–1
)
25
20
15
10
5
25
20
15
10
5
1
2
3
Acetone
2.88
3.82
1.60
24.2
43.2
10.1
9.4
6.0
4.5
DMF
Dichloromethane
[a] Experimental conditions: Pd(dba)2 (2 mM), mim (80 mM, 40 equiv), ArI
(Ar = 4-MeO-C6H4, 20 mM, 10 equiv). [b] Values estimated at 268 K.[25]
The apparent rate constants found for these three solvents
have no obvious relationship with the polarity of the solvent itself.
Even in the case of less dissociative dichloromethane, the
reaction proceeds at a rate comparable to that found for DMF and
acetone.
0
0
0
10
20
30
40
100
200
300
[ArI] (mM)
[mim] (mM)
To probe the generality of our results with respect to the ligand,
several monodentate nitrogen ligands of various nature were
tested, as reported in Figure 3 (for experimental conditions and
kinetic constants, see the Supporting Information, §2.5). All the
nitrogen-containing heteroarenes tested (imidazoles L1 and L2,
benzimidazole L3, pyrazole L4, and pyridine L5) behave similarly
to mim: Pd(dba)2 was readily dissolved and the reaction with 4-
iodoanisole proceeded with a second order with respect to Pd(0).
Similar results were also obtained for representative primary and
secondary aliphatic amines (piperidine L6 and butylamine L7),
suggesting that the observed behavior is general for unhindered
nitrogen ligands. By contrast, aniline (L8) and the tertiary amine
N-methylpiperidine (L9) display low affinity for Pd(0). Indeed, with
these amines most of Pd(dba)2, which has scant solubility in
acetone, remained undissolved and no reaction was observed
when the iodoarene was added at 268 K.
Figure 2. Left: dependence of the second order apparent rate constant kapp,2 on
[ArI] (Ar = 4-MeO-C6H4) and linear regression; for all experiments: Pd(dba)2 (2
mM in acetone), mim (80 mM, 40 equiv). Right: dependence of kapp,2 on [mim],
for all experiments: Pd(dba)2 (2 mM in acetone), ArI (Ar = 4-MeO-C6H4, 20 mM,
10 equiv).
Similarly, the influence of mim on the kinetics of the model
reaction was evaluated by measuring kapp,2 for different excess
amounts of mim itself (Figure 2, right). When increasing the
amount of mim in the reaction mixture, kapp,2 decreases slightly
with no apparent integer reaction order in mim. In this respect, it
is worth noting that mim can be involved in different steps of the
oxidative addition that can influence the overall rate: formation of
the initial [Pd(mim)n] complex and possibly dissociation to allow
the coordination of ArI. Furthermore, at these high concentrations
mim constitutes a sizeable fraction of the bulk of the reaction
mixture, leading to an even more complex picture that could be
compatible with the observed behavior.
Summarizing all the experimental observations presented
above, the rate law at fixed [mim] is:
[
]2[
]
푟 = 푘 Pd(0) ArI
(iii)
Variation of the experimental conditions
Acetone was chosen as a model solvent due to its high polarity,
allowing a strong dissociation of the oxidative addition product
and easy conductometric measurements. To confirm that the
findings reported above can be generalized to other solvents,
similar experiments were performed in DMF (N,N-
dimethylformamide) and in dichloromethane. For all solvents,
second order in Pd(0) has been observed and the results are
summarized in Table 1.
Figure 3. Nitrogen ligands tested for the reaction of Pd(dba)2 with 4-iodoanisole,
classified according to the outcome of the experiment.
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
The experiments described so far all involved 4-iodoanisole
as a substrate. To ensure that this unexpected kinetic law applies
also to other substrates, experiments have been performed on a
wide range of substituted iodobenzenes (Figure 4).
from simple calculations under standard pre-equilibrium
hypotheses (see the Supporting Information, § 2.1).[29]
The second-order in Pd(0) rate law fitted correctly the OA
kinetics observed for various substituted iodoarenes bearing
electron-donating groups (EDGs – HO, MeO, Me) or mild
electron-withdrawing groups (EWGs, such as halogens). A
satisfactory correlation was found with Hammett’s σ parameters
(Figure 4).[26] The slope of this plot (ρ = 1.50) indicates a lower
sensitivity to electronic effects on the haloarene substrate than for
the well-studied Pd(PPh3)4 (ρ = 2.3 ± 0.2), thus suggesting that in
this case the build-up of negative charge on the aromatic moiety
in the transition state is less significant.[5c]
Iodoarenes having strong electron-withdrawing group, such
as CN and NO2, behave differently and the first-order rate
equation (i) fitted better the experimental kinetic curves (see
Supporting Information, §2.4). Similarly, the same experiments
performed with bromoarenes instead of iodoarenes gave kinetics
that were first-order in Pd(0) and in agreement with equation (i),
regardless of the ring substituent.[27, 28]
Scheme 3. Proposed mechanistic models for oxidative addition of haloarenes
to imidazole-ligated Pd(0).
0.6
4-Cl
0.4
To better interpret these experimental findings, theoretical
calculations were performed focusing on some important points
concerning the nature of the Pd0 species in solution, their
interaction with the ArX substrate and the actual OA pathway.
4-Br
0.2
0.0
4-F
H
The structure of [Pd0(mim)n] complexes
4-Me
In our previous report,[20] we suggested on the basis of NMR data
that imidazole-ligated Pd(0) complexes are tricoordinate. Any
attempt to isolate this compound or to get high-resolution NMR
spectra by decreasing the temperature invariably failed.[20]
Given the difficulties encountered in the experimental
characterization of this material, we turned our attention to
theoretical calculations at the DFT level to shed light on the
structure of this complex. Despite thorough searches, it was not
possible to locate a minimum for a [Pd(mim)3] complex featuring
all the three mim ligands coordinating in a κ1 fashion through the
pyridine-like nitrogen. One of the mim molecules was invariably
ejected during geometry optimizations, giving a linear [Pd(κ1-
mim)2] structure.
Imidazoles can also act as η2 ligands through the C4-C5
bond.[30] Stable geometries were found for tricoordinate
[Pd(κ1-mim)2(η2-mim)], [Pd(κ1-mim)(η2-mim)2] and [Pd(η2-mim)3],
in addition to the dicoordinate [Pd(κ1-mim)2], [Pd(κ1-mim)(η2-
mim)] and [Pd(η2-mim)2].
Computed thermodynamic parameters for these complexes
are presented in the Supporting Information (§3.1). The most
stable configuration among the dicoordinate complexes, which we
used as a reference for energies and free energies (ΔH = 0 and
ΔG = 0), is the one featuring one mim coordinating in η2 fashion
and one in κ1 mode. Its κ1,κ1 isomer is only marginally less stable
(ΔH = +7.5 kJ mol–1 and ΔG = +4.4 kJ mol–1), in the same way as
the η2,η2 isomer (ΔH = +7.4 kJ mol–1 and ΔG = +9.2 kJ mol–1).
-0.2
4-MeO
-0.4
 1.50 ± 0.17
4-HO
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
0.1
0.2
0.3

Figure 4. Hammett correlation of the third-order kinetic constant k (see Equation
(iii)) with respect to kH for substituted iodobenzenes. For all experiments:
Pd(dba)2 (2 mM in acetone), mim (80 mM, 40 equiv), ArI (10 mM or 20 mM, 5
or 10 equiv). Standard deviation of the mean (determined for R = 4-MeO, n = 6
measures) is shown as error bars.
These results highlight that the mechanism leading to second-
order kinetics with respect to Pd(0) is not favored for very
electron-poor iodoarenes and for bromoarenes.
A mechanistic model that could explain experimental kinetics
is presented in Scheme 3. For bromoarenes and iodoarenes
having strong EWGs (Scheme 3, a) an ordinary concerted
insertion of Pd(0) should be operative. In the case of iodoarenes
substituted with EDGs or mild EWGs, an additional Pd(0) center
would get involved by coordinating the departing I atom, in
agreement with the observed second-order rate law, as resulting
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
Addition of a third ligand is predicted to be exothermic but not
exergonic, irrespective of the coordination mode (κ1 or η2).[31]
The preference for η2 coordination can be explained by the π-
acceptor character of mim coordinated in this configuration, which
is expected to stabilize the zerovalent palladium complex by
withdrawing electron density from the metal center. This
observation can be confirmed by considering the charge
distribution within the complexes. [Pd(κ1-mim)2] has a very small
dipole moment (1.4 D) lying almost perpendicular to the NPdN
axis, while [Pd(κ1-mim)(η2-mim)] has a dipole moment of 9.4 D
pointing towards the κ1-coordinating mim molecule (Figure 5). As
assessed by the Natural Population Analysis (NPA),[32] in the κ1,κ1
isomer the Pd center bears a negative partial charge (-0.315 |e|)
while the two ligands have positive charges of almost equal
magnitude (+0.157 |e| and +0.158 |e|). In the κ1,η2 complex, the
partial charge on the metal center is diminished in absolute value
(-0.116 |e|) and the η2 ligand has a negative charge (-0.047 |e|),
while the mim coordinated in the κ1 mode has a partial charge
(+0.163 |e|) comparable to that of the ligands of the κ1,κ1 isomer.
Analogously, [Pd(κ1-mim)2(η2-mim)] has a calculated dipole
moment of 11.8 D with positive charge accumulating in the
direction of the mim molecule coordinating in a κ1 fashion.
By contrast, η2 coordination is not favorable for Pd(II). For
example, passing from trans-[(4-MeO-C6H4)Pd(κ1-mim)2I] to
trans-[(4-MeO-C6H4)Pd(κ1-mim)(η2-mim)I], ΔH = +31.0 kJ mol–1
and ΔG = +31.8 kJ mol–1. This observation can be rationalized by
considering that removal of electron density from the Pd(II) center
is not as favorable as for Pd(0).
In summary, calculations suggest that mim-ligated Pd0
complexes may have coordination number 2 or 3, and that mim
can act either as a κ1 or an η2 ligand for Pd(0), while only κ1
coordination is favorable for Pd(II).
Interaction of [Pd0(mim)2] with ArX
The interaction of haloarenes with [Pd(mim)2] was next
examined.[33] Two sets of ArX were considered (for X = I, Br, Cl):
one with Ar = 4-(MeO)-C6H4 as a representative of electron-rich
derivatives and a second one featuring Ar = 4-(O2N)-C6H4 as a
prototypical electron-poor moiety. In every case, a stable
[(ArX)Pd(κ1-mim)2] structure could be found. Similarly to what
reported by Thiel and Goossen for anionic [(PhX)Pd(PMe3)Y]– (Y
= AcO–, Cl–) on the basis of DFT calculations,[34] a linear C-I-Pd
arrangement was found, giving rise to a T-shape coordination
environment around Pd. Selected geometric parameters for the
complex involving 4-(MeO)-C6H4I are reported in Figure 6.
Analogous [(ArX)Pd(κ1-mim)(η2-mim)] complexes were predicted
to exist, and they are about as stable as the corresponding
[(ArX)Pd(κ1-mim)2] isomers for X = I and slightly more stable for
X = Br, Cl. For homogeneity and clarity, only κ1-mim complexes
are reported here (see the Supporting Information, §3.2 for more
data on κ1, η2 isomers).
Figure 6. Structure of complex [(4-(MeO)-C6H4I)Pd(κ1-mim)2], as optimized at
the DFT level. Selected bond distances (in Å) and angles (in degrees) are
reported.
Computed enthalpies of formation (ΔH) for the said
complexes are given in Table 2. In every case, complexation is
exothermic and the absolute value of ΔH increases according to
the nature of X in the order Cl<Br<I, and it is greater for 4-nitro
derivatives than for 4-methoxyhaloarenes. In accordance with the
thermodynamic stability of this interaction, the Pd-X distance
d(Pd-X) increases in the order I<Br<Cl, i.e. in the opposite sense
of atomic radii. The C-X bond is slightly lengthened upon
complexation. Even if in the elongations of the C-X bonds Δd(C-
X) are not greater than 0.05 Å, a trend is clear, since for both R =
OMe and R = NO2 they vary in the order Cl<Br<I. This bond
weakening can be interpreted in terms of charge transfer from the
metal to an antibonding orbital of the C-X moiety. This view is
confirmed by NPA charge analysis, as negative partial charge is
built up on ArX, as reported in Table 2. Again, even if figures are
small, the periodic trend is also evident, the absolute values
increasing in the order Cl<Br<I. Electron-poor 4-nitrohaloarenes,
Figure 5. Structures of [Pd(κ1-mim)2], [Pd(κ1-mim)2(η2-mim)] and [Pd(κ1-
mim)(η2-mim)], as resulting from DFT calculations. NPA partial charges on each
mim ring and on the Pd center are reported (in units of |e|). Dipole moments are
shown as vectors.
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
as expected, are better electron acceptors than 4-haloanisoles
and they have a higher partial charge (in absolute value).[35]
These observations mirror the fact that the second-order-in-
Pd OA mechanism, which can be attributed to cooperative
assistance of a second Pd(0) center by coordination to the
departing X atom and thus weakening the C-X bond, is only
observed for X = I, since for X = Br the interaction is much weaker
and it is practically negligible for X = Cl.
Table 2. Complex formation between haloarenes and [Pd(κ1-mim)2], as
studied by DFT calculations.
Partial
charge on
ArX[c]
ΔH
(kJ mol–1
d(Pd-X)
(Å)
Δd(C-X)
(Å)[b]
ArX
[a]
)
4-(MeO)-C6H4I
4-(MeO)-C6H4Br
4-(MeO)-C6H4Cl
4-(O2N)-C6H4I
4-(O2N)-C6H4Br
4-(O2N)-C6H4Cl
–34.9
–19.1
–11.5
–45.7
–23.9
–13.2
3.13
3.19
3.31
3.04
3.10
3.24
0.04
0.02
0.01
0.05
0.03
0.01
–0.068
–0.027
–0.001
–0.121
–0.052
–0.011
Figure 7. Computed structures of the transition states for the OA reaction
involving 4-iodoanisole and 4-iodonitrobenzene according to a non-cooperative
[a] Enthalpy of complex formation from ArX and [Pd(κ1-mim)2]. [b] Increase
of the C-X bond length passing from uncoordinated ArX to [(ArX)Pd(κ1-
mim)2]. [c] As determined by natural population analysis (NBO method)[32]
and summing all the charges associated to the atoms constituting ArX.
(TS1_OMe and TS1_NO2)
a cooperative mechanism (TS2_OMe and
TS2_NO2). Selected bond distances (in Å) angles (in degrees) are reported.
All the four representative TSs have a similar arrangement
involving the reactive centers, with the C(1)-Pd-I atoms lying in a
plane almost perpendicular to the aromatic ring (Figure 7). To get
a qualitative understanding of these processes, two meaningful
geometric parameters can be analyzed: the length of the
dissociating C(1)-I bond and that of the forming C(1)-Pd bond. In
the case of Ar = 4-(MeO)-C6H4 the C(1)-Pd distance is almost the
same in TS1_OMe as in TS2_OMe, while the C(1)-I bond is 0.06
Å longer in TS2_OMe. This lengthening is about 0.02 Å greater
than the elongation associated to the formation of [(ArI)Pd(κ1-
mim)2] (Δd(C-X) in Table 2). This means that coordination of the
[Pd(κ1-mim)2] alters the transition state structure in such a way
that TS2_OMe is slightly less reactant-like than TS1_OMe only
with respect to the breaking of the C-I, i.e. by weakening the latter.
Analogous considerations hold in the case of Ar = 4-(O2N)-C6H4,
since passing from TS1_NO2 to TS2_NO2 the C(1)-I bond is
lengthened by 0.10 Å, while the C(1)-Pd distances increases by
0.04 Å.
As stated previously, experimental support of the cooperative
mechanism involving two Pd(0) centers has not been found for
electron-poor iodoarenes, such as 4-iodonitrobenzene. Aiming at
finding an interpretation of this contrasting behavior according to
the electronic properties of the iodoarene, the enthalpy of binding
ΔrH of the [Pd(κ1-mim)2] moiety to ArI through the I atom has been
calculated (Table 3), either for isolated ArI or for the TSs for the
OA reaction (TS1_OMe and TS1_NO2). Comparison of these
thermodynamic parameters gives an appreciation of the relative
Oxidative addition pathways
To completely characterize the reaction pathway, the transition
states (TS) of the OA reaction involving either the iodoarene alone
– thus associated with the first-order-in-Pd mechanism – or the
aryl halide-[Pd(κ1-mim)2] complexes described previously – thus
related to the cooperative pathway corresponding to the second-
order-in-Pd kinetic model
– have been located by DFT
calculations. As previously, 4-iodoanisole has been considered as
a representative electron-rich iodoarene (Figure 7, TS1_OMe and
TS2_OMe), and 4-iodonitrobenzene as a prototypical electron-
poor derivative (Figure 7, TS1_NO2 and TS2_NO2). The full
mechanistic pathways associated to the reaction passing through
the abovementioned TSs have been analyzed at the DFT level
and the associated energy profiles are reported in the Supporting
Information (§3.4). All transition states feature only one mim
ligand on the Pd(0) center actually inserting into the C(1)-I bond.
A third pathway featuring two mim ligands on the Pd(0) atom
involved in OA has also been considered, but it has been found
too high in energy (see the Supporting Information, §3.4).
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
The following basis sets were employed: 6-31+G* for N; 6-31G* for H, C
and O; 6-311+G* for Cl, a triple zeta quality correlation-consistent
polarized basis set augmented with diffuse functions (aug-cc-pVTZ-PP)
designed for the small-core relativistic Stuttgart-Dresden pseudopotential
(ECP10MDF) for Br;[39] and a basis set of the same quality as the latter
(aug-cc-pVTZ-PP)[40] in conjunction with the Stuttgart-Dresden
pseudopotential (ECP28MDF) for Pd and I.[41] To simulate experimental
conditions, bulk solvent effects were accounted for by using an implicit
solvation model (IEF-PCM) as implemented in Gaussian.[42] The default
spheres radii, static and optical dielectric constants for acetone were used.
All stationary points were characterized as minima or first-order transition
states by analytical frequency calculations. IRC calculations were used to
confirm that transition states linked the proper minima. Computed
harmonic frequencies (without any scaling factor) were used to calculate
the thermal contributions to the thermodynamic functions under standard
conditions with the usual approximations.
stabilization upon coordination of the second Pd(0) center of the
TS with respect to the reactant. It is here appropriate to compare
reaction enthalpies, since the molecularity of the reactions
reported in Table 3 is the same and the entropic contribution to
free energy is thus similar.[36] Indeed, while for R = OMe ΔrH is
unchanged passing from free 4-iodoanisole to TS1_OMe, binding
[Pd(κ1-mim)2] to TS1_NO2 is less favorable by 5.7 kJ mol–1
compared to free 4-iodonitrobenzene. The resulting picture is
compatible with the experimental observation that the cooperative
mechanism is not favorable for electron-poor iodoarenes.
Table 3. Coordination of the Pd(0) to ArI and to TSs for the oxidative
addition, as resulting from DFT calculations. All mim ligands are coordinating
in κ1 mode.
ΔrH (kJ
Reaction
R
mol–1
)
Acknowledgements
MeO
O2N
–34.9
–45.7
We thank CNRS, ENSCP, and ENS, as well as the ANR program
CD2I (project CuFeCCBond) for financial support.
MeO
O2N
–34.9
–40.0
Keywords: palladium • reaction mechanisms • oxidative addition
• halogen bond • nitrogen heterocycles
[1]
A recent tutorial review on OA: a) J. A. Labinger, Organometallics 2015,
34, 4784–4795; see also: b) R. H. Crabtree, The Organometallic
Chemistry of the Transition Metals, John Wiley, Hoboken, 2005; c) J. F.
Hartwig, Organotransition Metal Chemistry: From Bonding to Catalysis,
University Science Books, Susalito, 2010; d) J. Kochi, Organometallic
Mechanisms and Catalysis. The Role of Reactive Intermediates in
Organic Processes, Academic Press, New York, 1978.
Conclusions
[2]
[3]
[4]
[5]
E. Negishi, Ed., Handbook of Organopalladium Chemistry for Organic
Synthesis, John Wiley & Sons, New York, 2002.
We performed a detailed mechanistic study on the OA reaction of
iodoarenes to imidazole-ligated Pd(0). Kinetic analyses
demonstrated an unexpected second-order dependence of the
reaction rate with respect to the concentration of Pd(0). This
a) P. Fitton, M. P. Johnson, J. E. McKeon, Chem. Commun. 1968, 6–7;
b) P. Fitton, E. A. Rick, J. Organomet. Chem. 1971, 28, 287–291.
J.-F. Fauvarque, F. Pflüger, M. Troupel, J. Organomet. Chem. 1981, 208,
419–427.
a) A. Jutand, Chem. Rev. 2008, 108, 2300–2347; b) A. Jutand, S. Négri,
Organometallics 2003, 22, 4229–4237; c) C. Amatore, F. Pflüger,
Organometallics 1990, 9, 2276–2282; d) C. Amatore, E. Carré, A. Jutand,
M. A. M’Barki, Organometallics 1995, 14, 1818–1826.
behavior is general across
a range of electron-rich and
moderately electron-poor substituted iodobenzenes and it is also
observed for other monodentate nitrogen ligands (e.g. amines)
that may be present in stoichiometric amounts in the conditions of
Pd-catalyzed cross-couplings. These experimental results have
been interpreted by proposing a mechanism involving the
cooperative intervention of two Pd(0) centers, one of which effects
the oxidative insertion into the C-I bond and the other weakens
this linkage by binding to the same molecule of iodoarene in a
halogen bond-like fashion. To the best of our knowledge, such a
cooperative involvement of two metal centers in the activation of
an organic halogen derivative has never been proposed before.
DFT calculations corroborated this mechanistic hypothesis and
allowed interpreting experimental trends.
[6]
[7]
a) M. R. Netherton, G. C. Fu, Angew. Chem. Int. Ed. 2002, 41, 3910–
3912; b) I. D. Hills, M. R. Netherton, G. C. Fu, Angew. Chem. Int. Ed.
2003, 42, 5749–5752.
a) F. Barrios-Landeros, J. F. Hartwig, J. Am. Chem. Soc. 2005, 127,
6944–6945; b) F. Barrios-Landeros, B. P. Carrow, J. F. Hartwig, J. Am.
Chem. Soc. 2009, 131, 8141–8154; c) E. Lyngvi, F. Schoenebeck,
Tetrahedron 2013, 69, 5715–5718; d) A. Kurbangalieva, D. Carmichael,
K. K. M. Hii, A. Jutand, J. M. Brown, Chem. Eur. J. 2014, 20, 1116–1125;
e) C. L. McMullin, N. Fey, J. N. Harvey, Dalton Trans. 2014, 43, 13545-
13556; f) S. Borjian, M. C. Baird, Organometallics 2014, 33, 3936–3940;
g) J. F. Hartwig, F. Paul, J. Am. Chem. Soc. 1995, 117, 5373–5374; h)
F. Proutiere, F. Schoenebeck, Angew. Chem. Int. Ed. 2011, 50, 8192–
8195.
[8]
[9]
a) M. Portnoy, D. Milstein, Organometallics 1993, 12, 1665–1673; b) L.
M. Alcazar-Roman, J. F. Hartwig, A. L. Rheingold, L. M. Liable-Sands, I.
A. Guzei, J. Am. Chem. Soc. 2000, 122, 4618–4630; c) L. M. Alcazar-
Roman, J. F. Hartwig, Organometallics 2002, 21, 491–502.
a) A. Jutand, K. K. (Mimi) Hii, M. Thornton-Pett, J. M. Brown,
Organometallics 1999, 18, 5367–5374; b) C. Amatore, E. Carré, A.
Computational details
All DFT calculations were performed with the Gaussian09 program (Rev.
E.01).[37] The structures of all minima and transition states were fully
optimized using the M06 functional without any symmetry constraint.[38]
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
Jutand, Y. Medjour, Organometallics 2002, 21, 4540–4545; c) J. B.
Johnson, T. Rovis, Angew. Chem. Int. Ed. 2008, 47, 840–871.
[10] Reviews on this topic: a) I. J. S. Fairlamb, Org. Biomol. Chem. 2008, 6,
3645-3656; b) C. Amatore, A. Jutand, Coord. Chem. Rev. 1998, 178–
180, 511–528; see also: c) C. Amatore, A. Jutand, F. Khalil, M. A. M’Barki,
L. Mottier, Organometallics 1993, 12, 3168–3178; d) C. Amatore, G.
Broeker, A. Jutand, F. Khalil, J. Am. Chem. Soc. 1997, 119, 5176–5185;
e) C. Amatore, A. Fuxa, A. Jutand, Chem. Eur. J. 2000, 6, 1474–1482; f)
C. Amatore, A. Bucaille, A. Fuxa, A. Jutand, G. Meyer, A. Ndedi Ntepe,
Chem. Eur. J. 2001, 7, 2134–2142; g) C. Amatore, A. Jutand, F. Khalilb,
ARKIVOC 2006, 2006, 38–48.
[23] The proportionality of conductivity with respect to the concentration of
ionic products has been verified in our experimental conditions. See the
Supporting Information, §2.3.
–
[24] Equation (ii) can be rearranged as follows: κ – = κ∞ 1(1+at–1) with
1
–1
a = (c0kapp,2
) .
[25] Values were calculated according to the formulae and parameters
reported in: D. R. Lide, CRC Handbook of Chemistry and Physics, CRC
Press, 2003, pp. 6-153–6-175, 9-45–9-51.
[26] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165–195.
[27] We reported first-order dependence on [Pd(0)] for 4-bromobenzonitrile in
our previous paper, see ref. [19].
[11] a) C. Amatore, M. Azzabi, A. Jutand, J. Organomet. Chem. 1989, 363,
C41–C45; b) C. Amatore, A. Jutand, A. Suarez, J. Am. Chem. Soc. 1993,
115, 9531–9541.
[28] For kinetic data concerning 4-bromoanisole, see the Supporting
Information, §2.6. The influence of additional I– has been also
investigated, and no effect on the order of reaction was observed, see
the Supporting Information, §2.7.
[12] C. Amatore, E. Carre, A. Jutand, M. A. M’Barki, G. Meyer,
Organometallics 1995, 14, 5605–5614.
[29] Kinetic curve fitting for electron poor iodoarenes (R = CN, NO2) is not
entirely satisfactory neither using equation (i) (first order), nor using
equation (ii) (second order), albeit the former is the most suitable, as
already stated (see Supporting Information, §2.4). We cannot exclude
that in some cases both mechanisms are operative at the same time.
[30] Other coordination modes involving the other couples of contiguous
atoms are not possible or less favorable, see the Supporting Information,
§3.1.
[31] For example ΔH = –21.6 kJ mol–1 and ΔG = +30.8 kJ mol–1 for the κ1,η2,η2
isomer, but it is also to be considered that the entropic cost of this
association reaction is overestimated by the usual approximations
involved in the calculation of Gibbs free energy.
[13] C. Amatore, A. Jutand, F. Lemaître, J. L. Ricard, S. Kozuch, S. Shaik, J.
Organomet. Chem. 2004, 689, 3728–3734.
[14] For a general overview, see: a) C. Amatore, A. Jutand, J. Organomet.
Chem. 1999, 576, 254–278; b) C. Amatore, A. Jutand, Acc. Chem. Res.
2000, 33, 314–321.
[15] a) J. C. Green, B. J. Herbert, R. Lonsdale, J. Organomet. Chem. 2005,
690, 6054–6067; b) A. Jutand, J. Pytkowicz, S. Roland, P. Mangeney,
Pure Appl. Chem. 2010, 82, 1393–1402.
[16] a) S. Otsuka, A. Nakamura, T. Yoshida, J. Am. Chem. Soc. 1969, 91,
7196–7198; b) L. A. Perego, P. Fleurat-Lessard, L. El Kaïm, I. Ciofini, L.
Grimaud, Chem. Eur. J. 2016, 22, 15491–15500.
[17] To the best of our knowledge, only one published study addresses
oxidative addition in the presence of amines: A. Tougerti, S. Negri, A.
Jutand, Chem. Eur J. 2007, 13, 666–676.
[32] a) J. P. Foster, F. Weinhold, J. Am. Chem. Soc. 1980, 102, 7211–7218;
b) A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899–
926.
[18] a) M. C. Done, T. Rüther, K. J. Cavell, M. Kilner, E. J. Peacock, N.
Braussaud, B. W. Skelton, A. White, J. Organomet. Chem. 2000, 607,
78–92; b) C. J. Mathews, P. J. Smith, T. Welton, J. Mol. Catal. A: Chem.
2003, 206, 77–82; c) C. J. Mathews, P. J. Smith, T. Welton, J. Mol.
Catal.A: Chem. 2004, 214, 27–32; d) İ. Özdemir, N. Şahin, Y. Gök, S.
Demir, B. Çetinkaya, J. Mol. Catal. A: Chem. 2005, 234, 181–185; e) S.
Haneda, C. Ueba, K. Eda, M. Hayashi, Adv. Synth. Catal. 2007, 349,
833–835; f) C. Xi, Y. Wu, X. Yan, J. Organomet. Chem. 2008, 693, 3842–
3846; g) M. S. Szulmanowicz, W. Zawartka, A. Gniewek, A. M. Trzeciak,
Inorg. Chim. Acta 2010, 363, 4346–4354; h) C. Xu, Z. Li, L.-M. Duan, Z.-
Q. Wang, W.-J. Fu, X.-Q. Hao, M.-P. Song, Transition Met. Chem. 2012,
37, 373–378; i) M. Trivedi, G. Singh, R. Nagarajan, N. P. Rath, Inorg.
Chim. Acta 2013, 394, 107–116; j) C. Liu, G. Liu, H. Zhao, Chin. J. Chem.
2016, 34, 1048–1052; k) W. Zawartka, A. Gniewek, A. M. Trzeciak, Inorg.
Chim. Acta 2017, 455, 595–599; for a very recent review: l) A. Kumbhar,
J. Organomet. Chem. 2017, 848, 22–88.
[33] For a general review on halogenated compounds as ligands for transition
metals, see: R. J. Kulawiec, R. H. Crabtree, Coord. Chem. Rev. 1990,
99, 89–115. For previous experimental evidence of the complexation of
haloarenes with Pd(0) complexes, see ref. [7d] and ref. [8c].
[34] L. J. Goossen, D. Koley, H. L. Hermann, W. Thiel, Organometallics 2005,
24, 2398–2410.
[35] For a more detailed analysis on the ArX-Pd0 bond, see the Supporting
Information, §3.3.
[36] Direct comparison of computed free energy activation barriers is not
possible, since with the usual approximations involved in computing ΔG,
the entropic cost of molecular association is largely overestimated. See
the Supporting Information, §3.4 for an extended discussion.
[37] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino,
G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T.
Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J.
J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.
Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J.
Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J.
V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT., 2009
[38] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215–241.
[39] K. A. Peterson, D. Figgen, E. Goll, H. Stoll, M. Dolg, J. Chem. Phys. 2003,
119, 11113.
[19] Selected reviews: a) L. Ackermann, R. Vicente, A. Kapdi, Angew. Chem.
Int. Ed. 2009, 48, 9792–9826; b) F. Bellina, R. Rossi, Tetrahedron 2009,
65, 10269–10310; c) M. Livendahl, A. M. Echavarren, Isr. J. Chem. 2010,
50, 630–651; d) C. Verrier, P. Lassalas, L. Théveau, G. Quéguiner, F.
Trécourt, F. Marsais, C. Hoarau, Beilstein J. Org. Chem. 2011, 7, 1584–
1601; e) F. Shibahara, T. Murai, Asian J. Org. Chem. 2013, 2, 624–636;
f) I. Hussain, T. Singh, Adv. Synth. Catal. 2014, 356, 1661–1696; g) F.
Bellina, in C-H Bond Activation and Catalytic Functionalization I (Eds.:
P.H. Dixneuf, H. Doucet), Springer, Cham, 2015, pp. 77–102.
[20] L. A. Perego, L. Grimaud, F. Bellina, Adv. Synth. Catal. 2016, 358, 597–
609.
[21] Pd(dba)2 is to be understood as an empirical formula. It is known that this
complex behaves as Pd2(dba)3 in solution, so it is sometimes written as
Pd2(dba)3·(dba). For a thorough NMR characterization of Pd2(dba)3, see:
A. R. Kapdi, A. C. Whitwood, D. C. Williamson, J. M. Lynam, M. J. Burns,
T. J. Williams, A. J. Reay, J. Holmes, I. J. S. Fairlamb, J. Am. Chem. Soc.
2013, 135, 8388–8399.
[40] K. A. Peterson, B. C. Shepler, D. Figgen, H. Stoll, J. Phys. Chem. A 2006,
110, 13877–13883.
[41] K. A. Peterson, D. Figgen, M. Dolg, H. Stoll, J. Chem. Phys. 2007, 126,
124101.
[22] Imidazole-ligated Pd(0) species are air-sensitive.
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
[42] a) J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999–
3094; b) M. Cossi, G. Scalmani, N. Rega, V. Barone, J. Chem. Phys.
2002, 117, 43–54.
This article is protected by copyright. All rights reserved.

10.1002/chem.201704899
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Luca Alessandro Perego,* Pierre-Adrien
Payard, Baptiste Haddou, Ilaria Ciofini,*
and Laurence Grimaud*
Page No. – Page No.
Evidence for a Cooperative
Mechanism Involving two
Palladium(0) Centers in the Oxidative
Addition of Iodoarenes
Oxidative addition of iodoarenes to Pd(0) complexes with nitrogen ligands is second
order in Pd(0) itself, and a mechanism involving the cooperative intervention of two
metal centres is proposed to explain this finding. Theoretical calculations suggest
the existence ArI-Pd(0) complexes in which the iodoarene is bound to the metal in a
halogen bond-like fashion, thus weakening the C-I bond and facilitating the insertion
of another Pd(0) centre.
This article is protected by copyright. All rights reserved.