Multiple Roles of Isocyanides in Palladium-Catalyzed Imidoylative Couplings: A Mechanistic Study

Article abstract of DOI:10.1002/chem.201602913

Kinetic, spectroscopic and computational studies examining a palladium-catalyzed imidoylative coupling highlight the dual role of isocyanides as both substrates and ligands for this class of transformations. The synthesis of secondary amides from aryl halides and water is presented as a case study. The kinetics of the oxidative addition of ArI with RNC-ligated Pd⁰species have been studied and the resulting imidoyl complex [(ArC=NR)Pd(CNR)₂I] (Ar=4-F-C₆H₄, R=tBu) has been isolated and characterized by X-ray diffraction. The unprecedented ability of this RNC-ligated imidoyl-Pd complex to undergo reductive elimination at room temperature to give the amide in the presence of water and an F^?/HF buffer is demonstrated. Its behavior in solution has also been characterized, revealing an unexpected strong tendency to give cationic complexes, and notably [(ArC=NR)Pd(CNR)₃]⁺with excess isocyanide and [(ArC=NR)Pd(PP)(CNR)]⁺with bidentate phosphines (PP). These species may be responsible for catalyst deactivation and side-reactions. Ab initio calculations performed at the DFT level allowed us to rationalize the multiple roles of RNC in the different steps of the catalytic cycle.

Full text of DOI:10.1002/chem.201602913

DOI: 10.1002/chem.201602913
Full Paper
&
Reaction Mechanisms
Multiple Roles of Isocyanides in Palladium-Catalyzed Imidoylative
Couplings: A Mechanistic Study
[
a, b, c]
[d]
[e]
[c]
Luca A. Perego,*
Laurence Grimaud*
Paul Fleurat-Lessard, Laurent El Ka ¨ı m, Ilaria Ciofini,* and
[a, b]
Abstract: Kinetic, spectroscopic and computational studies
examining a palladium-catalyzed imidoylative coupling high-
light the dual role of isocyanides as both substrates and li-
gands for this class of transformations. The synthesis of sec-
ondary amides from aryl halides and water is presented as
a case study. The kinetics of the oxidative addition of ArI
perature to give the amide in the presence of water and an
ꢀ
F /HF buffer is demonstrated. Its behavior in solution has
also been characterized, revealing an unexpected strong
tendency to give cationic complexes, and notably [(ArC=
+
NR)Pd(CNR) ]
with excess isocyanide and [(ArC=
3
_
_
+
NR)Pd(PP)(CNR)] with bidentate phosphines (PP). These
species may be responsible for catalyst deactivation and
side-reactions. Ab initio calculations performed at the DFT
level allowed us to rationalize the multiple roles of RNC in
the different steps of the catalytic cycle.
0
with RNC-ligated Pd species have been studied and the re-
sulting imidoyl complex [(ArC=NR)Pd(CNR) I] (Ar=4-F-C H ,
2
6
4
R=tBu) has been isolated and characterized by X-ray diffrac-
tion. The unprecedented ability of this RNC-ligated imidoyl-
Pd complex to undergo reductive elimination at room tem-
[
2]
Introduction
CꢀPd bond has been known since 1969, and the develop-
ment of Pd-catalyzed reactions with CO incorporation (i.e., car-
bonylative couplings, Scheme 1, a) started towards the end of
the last century, approximately at the same time as the ordi-
Isocyanides are recognized as immensely valuable building
blocks in present-day organic chemistry, mainly for their key
role in the Passerini and Ugi multicomponent reactions with
[
3]
nary Pd-catalyzed cross couplings. However, the develop-
ment of Pd-catalyzed transformations involving isocyanide in-
sertion (i.e., imidoylative couplings, Scheme 1, b) has taken
place mainly after 2000. The efforts in this field are summar-
[
1]
their countless variations.
Their distinctive feature, in
common perhaps only with CO, carbenes, and carbenoid spe-
cies, is the presence of a carbon atom that can sequentially act
as a nucleophile and an electrophile and give insertion reac-
tions across carbon–metal bonds. Isocyanide insertion in the
[4]
ized in some excellent recent reviews.
[a] L. A. Perego, Dr. L. Grimaud
Ecole Normale Supꢀrieure, PSL Research University
UPMC Univ Paris 06, CNRS, Dꢀpartement de Chimie, PASTEUR
2
4, rue Lhomond, 75005 Paris (France)
[
b] L. A. Perego, Dr. L. Grimaud
Sorbonne Universitꢀs, UPMC Univ Paris 06, ENS
CNRS, PASTEUR, 75005 Paris (France)
E-mail: luca.perego@ens.fr
Scheme 1. Pd-catalyzed carbonylative and imidoylative couplings.
laurence.grimaud@ens.fr
[c] L. A. Perego, Dr. I. Ciofini
Institut de Recherche de Chimie Paris, CNRS-Chimie
ParisTech, UMR 8247, 11 rue Pierre et Marie Curie, 75005 Paris (France)
E-mail: ilaria.ciofini@chimie-paristech.fr
Broadly speaking, isocyanides interact with metals rather like
an electron-rich variant of CO. They allow the construction of
molecular complexity by offering one more point of variability,
in comparison to CO, at the N-substituent. Unlike CO, isocya-
nides have the tendency to undergo poly-insertion across the
CꢀPd bond and to polymerize in the presence of Pd cata-
[
d] Prof. P. Fleurat-Lessard
Institut de Chimie Molꢀculaire de l’Universitꢀ de Bourgogne
Universitꢀ de Bourgogne, UMR-CNRS 6302
Universitꢀ de Bourgogne Franche-Comtꢀ
9
, avenue Alain Savary, 21078 Dijon (France)
[4b,5]
[
e] Prof. L. El Ka ¨ı m
Laboratoire LSO ENSTA-Polytechnique-CNRS, UMR 7652
Ecole Nationale Supꢀrieure de Techniques Avancꢀes
lysts.
For this reason, many of the imidoylative couplings
previously described involve sterically hindered isocyanides,
which are intrinsically less prone to this kind of parasitic reac-
tions, or involve an intramolecular ring-closure with a tethered
6
28 Bd des Marꢀchaux, 91120 Palaiseau (France)
Supporting information for this article can be found under
http://dx.doi.org/10.1002/chem.201602913.
[4]
nucleophile. The latter strategy most likely shortens the life-
Chem. Eur. J. 2016, 22, 1 – 11
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
time of the imidoylpalladium intermediate and then disfavors
multiple additions of isocyanide.
not essential for the reaction to proceed. Indeed, the analo-
gous “ligandless” protocol employing only PdCl also gives the
2
[
10g]
The extent of the aforementioned side reactions depends
critically on the nature of the solvent and ligand, and in gener-
al, imidoylative couplings are known to be extremely sensitive
to the reaction conditions, which may be one of the reasons
expected amide in satisfactory yield (Scheme 3).
Isocyanides
0
II
are extremely good ligands for both Pd and Pd , to the point
that isocyanide-functionalized silica gel has been recently pro-
posed as an effective scavenger to reduce the Pd content of
[
4b,c]
[11]
for their slow development.
across the CꢀPd bond was studied in detail from a mechanistic
point of view a long time ago. Surprisingly, the reductive
The insertion of isocyanides
active pharmaceutical ingredients down to sub-ppm levels.
Thus, we expected that isocyanides play the role of both sub-
[
5e,f,6]
[12]
strate and ligand in these transformations.
elimination from the imidoylpalladium complexes has received
comparatively little attention. In contrast, reductive elimination
from acylpalladium intermediates formed in carbonylative reac-
[
7]
tions has been analyzed in detail. Several studies describe
the synthesis and reactivity of a variety of structurally diverse
[
5g,8]
cyclometalated imidoylpalladium complexes.
Our research efforts aim to gather information that is poten-
tially useful for the rationalization of empirical findings and to
ease the development of new synthetically valuable transfor-
mations. Thus, we undertook a detailed mechanistic study on
[
10g]
3
Scheme 3. Pd-catalyzed amidation of PhBr with and without PPh .
a
representative prototypical imidoylative coupling, with
a broad focus on the entire catalytic cycle. Apart from a very
When PPh is used as a ligand, one equivalent is readily oxi-
3
[
9]
0
recent computational report, to the best of our knowledge,
no study has appeared in the literature concerning this kind of
transformation. Among the comparatively few literature exam-
ples of intermolecular imidoylative couplings with simple nu-
dized to OPPh , along with the generation of Pd , at room
3
[13]
temperature in the presence of water and a base. In the ab-
sence of PPh , the action of hydroxide anions on isocyanide
3
II
complexes of Pd with an excess of the isocyanide itself may
[
10]
0
cleophiles, the study case we chose for our work is the ele-
gantly simple reaction reported in 2011 by Jiang and co-work-
ers for the Pd-catalyzed synthesis of secondary amides from an
aryl halide, an isocyanide, and water in the presence of CsF as
generate Pd with concurrent oxidation of the isocyanide to
[14,15]
isocyanate, as first proposed by Malatesta in 1955.
0
Oxidative addition with isocyanide-ligated Pd
[
10g]
a base (Scheme 2).
The oxidative addition of an organic iodide with [Pd(CNtBu) ]
2
[2a]
has been first reported as early as 1969, but to the best of
our knowledge, no kinetic data are available. Considering that
0
the oxidative addition to isocyanide-ligated Pd is expected to
be the first step of the catalytic reaction of interest when per-
formed in the absence of PPh , we studied its kinetic profile in
3
detail. We chose 4-fluoroiodobenzene as a model compound
Scheme 2. Pd-catalyzed amidation of aryl halides reported by Jiang and co-
workers.
19
to follow the progress of the reaction conveniently by F NMR.
[
10g]
When tBuNC (10 equiv) was added to a purple-violet solu-
tion of [Pd(dba) ] (20 mm in CDCl , dba=dibenzylideneace-
2
3
We present here a kinetic study of the oxidative addition
tone), the color rapidly changed to orange-yellow and the
0
1
step with RNC-ligated Pd species. We also report a thorough
H NMR spectrum of the resulting solution showed the pres-
characterization of the complex resulting from the oxidative
addition, with particular attention to the equilibria in which it
engages with other species normally present in the reaction
environment. On the basis of experimental observations and
theoretical computations, we propose a catalytic cycle and an-
alyze the multiple roles played by RNC in this transformation.
ence of only uncoordinated dba, without the characteristic sig-
[16]
nature peaks of ligated dba. This preliminary result prompt-
ed us to believe that the [Pd(dba) ]/tBuNC is a good model
2
system to study the oxidative addition of organic halides to
0
the isocyanide-coordinated Pd , similar to what we have previ-
[12]
ously performed for a Pd/imidazole system.
When a solution of [Pd(dba) ] was treated with 4-fluoroiodo-
2
benzene (1.0 equiv) in the presence of excess tBuNC (25 equiv)
Results and Discussion
at room temperature, a comparatively fast reaction ensued.
19
The resonance of the organic iodide in the F NMR spectrum
0
II
Generation of Pd from Pd precursors
(
d=ꢀ114.5 ppm) gradually decreased with time and a new
During their optimization studies, Jiang and co-workers estab-
lished that a mixture of PdCl and PPh in 1:2 molar ratio gave
peak appeared downfield (d=ꢀ113.3 ppm). The kinetics of
this reaction could be readily obtained and the reciprocal of
the aryl iodide concentration fitted a straight line with respect
to time, indicating overall second-order behavior (Figure 1),
2
3
the best yields among all other conditions tested. However,
they also realized that the addition of a phosphine ligand is
&
&
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
well-studied oxidative addition of aryl halides to [Pd(PPh ) ],
3
4
for which dissociation of PPh to give the 14-electron complex
3
[
17]
[
Pd(PPh ) ] is required for the oxidative addition to proceed.
3 2
0
The exact nature of RNC-ligated Pd in solution in the pres-
ence of excess isocyanide, which is relevant for catalytic reac-
tions, has not been investigated experimentally. Isolated com-
plexes of stoichiometry [Pd(CNR)2] have long been
[
2a,14,16a,18]
known.
solid state
Usually, they show a trimeric structure in the
[16a,18b,c]
(except for the case of extremely hindered
[18d]
isocyanides, which give rise to monomers ), but molecular
weight determination in solution has not been conclu-
Figure 1. Kinetics of the consumption of 4-fluoroiodobenzene (18 mm in
benzene) in the presence of [Pd(dba) ] (1.0 equiv) and tBuNC (25 equiv) at
[16a,18a,c]
sive.
Moreover, formation of complexes featuring more
2
than two isocyanide ligands is conceivable in the presence of
excess isocyanide. To shed more light on this point, we esti-
1
9
ꢀ1
2
58C, as determined by F NMR spectroscopy. [ArI] (*, left scale) and [ArI]
(
&
, right scale) are reported as a function of time, demonstrating second-
order behavior.
mated the relative stabilities of [Pd(CNtBu) ] complexes (n=0–
n
4) by means of DFT calculations (Figure 3). The most stable
species appears to be the linearly dicoordinated [Pd(CNtBu)2],
and the addition of a third ligand is disfavored (DG8=
with the second-order kinetic constant k as the slope (k=
ꢀ
1
ꢀ1
ꢀ1
[19]
0.59 Lmol min ).
+7.7 kcalmol at 258C). Although the manner in which the
entropic contribution is taken into account in our DFT calcula-
tions may lead to an overestimation of the difference in stabili-
ty of the species, this result is in agreement with the experi-
mentally observed independence of the reaction rate on the
concentration of isocyanide, because the most stable species is
also the reactive one.
The observed second-order rate constant did not change
significantly when the experiment was repeated with different
excess amounts of tBuNC (up to 50 equiv), thus proving
zeroth-order with respect to the isocyanide (Figure 2). Data col-
lected with excess aryl iodide (see the Supporting Information,
Section 1.1 for detailed data) confirmed a simple second-order
0
rate law: dP/dt=k[ArI][Pd ], in which P stands for the concen-
0
tration of the formed product and [Pd ] is the total concentra-
0
tion of the Pd species present in solution. It is noteworthy
ꢀ
1
ꢀ1
that the observed value (k=0.59 Lmol min at 258C, relative
standard deviation 6%) indicates a comparatively fast reaction
at room temperature. Thus, at least for aryl iodides, it is unlike-
ly that the oxidative addition is the turnover-limiting step of
[
9]
the catalytic cycle, as previously suggested for PhBr.
0
Figure 3. Relative stabilities of mononuclear tBuNC-ligated Pd complexes,
as determined by DFT calculations.
Characterization of the product of oxidative addition
Figure 2. Determination of the order of reaction at 258C with respect to
tBuNC for the reaction of 4-fluoroiodobenzene (20 mm in benzene) with
When the reaction of [Pd(dba) ] with 4-fluoroiodobenzene in
2
the presence of tBuNC (4.0 equiv) was done on a preparative
scale, the air-stable imidoylpalladium complex 1 could be iso-
lated by flash chromatography in excellent yield (Scheme 4).
Its structure was confirmed by standard spectroscopic tech-
2
[Pd(dba) ] (1.0 equiv) in the presence of excess tBuNC (see the Supporting
Information, Section 1.1 for raw data).
[20]
The behavior we observed is consistent with the complete
niques and single-crystal X-ray diffraction analysis.
In the
0
displacement of dba ligands from Pd , thus experimentally vali-
solid state, the four ligands are in the expected trans square
planar geometry around the Pd center and the imidoyl moiety
features a trans arrangement of the aromatic ring and the tBu
substituent.
0
dating our model, and with the reactivity of tBuNC-ligated Pd
towards the iodoarene. The observed zeroth-order behavior
with respect to tBuNC suggests that the dissociation of one of
the coordinated isocyanide ligands is not necessary for the re-
action to take place. This picture is very different from the
1
19
Comparison of H and F NMR data proved that the oxida-
tive addition of ArI with [Pd(dba) ]/tBuNC described in the pre-
2
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
2
Scheme 4. Synthesis of trans-[(ArC=NtBu)Pd(CNtBu) I] (1).
vious section, apparently yields 1 without any observable non-
inserted intermediate. This demonstrates that under these re-
action conditions (excess isocyanide), insertion of tBuNC across
the CꢀPd bond is significantly faster than the already facile oxi-
1
9
Figure 4. Left: fitting of the observed F NMR chemical shift of 1 (20 mm in
]DMSO, at 258C) in the presence of n equiv of isocyanide. The continuous
[D
6
dative addition. A solution of 1 (20 mm in [D ]DMSO) and
6
line represents the fitting function (see the Supporting Information, Section
1.2). Right: F NMR spectra from which the experimental data are derived.
19
19
excess tBuNC (10 equiv) showed a broadened F NMR peak,
19
but no significant time evolution occurred in both F and
1
H NMR spectra within 72 h at room temperature. This means
that complex 1 is rather stable towards poly-insertion of tBuNC
at room temperature.
The observed line broadening prompted us to do further in-
vestigations. When more tBuNC was added, the peak sharp-
ened and moved further downfield, in a pattern typical for fast
exchange on the NMR timescale (Figure 4). This equilibrium
ꢀ
was interpreted as the substitution of coordinated I with
tBuNC, resulting in the generation of a cationic complex
(Scheme 5). Fitting of the observed chemical shift with the ex-
pected function (see the Supporting Information, Section 1.2)
gave access to the equilibrium constant K=0.066 at 258C
(Figure 4). Electrical conductivity measurements confirmed this
hypothesis and the thermodynamic constant for this equilibri-
um has been determined by fitting the conductivity titration
Figure 5. Conductometric titration of complex 1 (2 mm in DMSO at 258C)
with increasing amounts of tBuNC. The conductivity k is plotted as a function
of the equiv of tBuNC added. The continuous line shows the fitting function
(see the Supporting Information, Section 1.3). For the inset, the same fitting
is reported in a linearized plot.
curve according to
a method we previously described
[12]
(Figure 5, see the Supporting Information, Section 1.3). Con-
ductivity measurements (Table 1) demonstrate that this ligand
exchange readily takes place in most of the common polar sol-
vents. The value found for DMSO is in agreement with NMR
data.
Table 1. Equilibrium constants for the reaction represented in Scheme 5
in different solvents and solvent mixtures, as determined by conducto-
metric titration at 258C.
The cationic complex was synthesized in good yield with
[
a]
ꢀ
TfO as a counter-anion by treating the parent iodide complex
[
21]
with Tl(OTf) (Scheme 6), and characterizing by multinuclear
NMR and IR spectra. The NMR data agreed with the assign-
ment for the previously observed species. IR in the solid state
showed that the main band corresponding to isocyanide
Solvent
K
Asymptotic
standard error [%]
DMSO
DMF
MeCN
0.0672
0.0118
0.080
0.118
1.2
1.7
2.4
3.7
ꢀ
1
ꢀ1
stretching moved from 2190 cm (for neutral 1) to 2217 cm
for 2-OTf), consistent with the expected reduction of back-
2
DMSO/THF/H O (5:1:1)
(
[
a] See the Supporting Information, Section 1.3 for detailed data treat-
donation for the charged species. The formation of such a cat-
ionic complex under conditions relevant for the catalytic reac-
ment.
Scheme 5. Ligand exchange of 1 in the presence of excess isocyanide.
Scheme 6. Synthesis of ionic complex 2-OTf.
&
&
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
tion, devoid of any halide trapping reagent, to the best of our
knowledge, has never been suggested before. In the related
case of phosphine-ligated acylpalladium complexes, the dis-
quaternary tert-butyl carbon) and one molecule of uncoordi-
14
13
1
nated isocyanide (featuring N, C J-coupling) were apparent.
The CN stretching frequency in the IR spectrum in DMSO
ꢀ
ꢀ1
ꢀ1
placement of I is possible only under very high pressure of
moved from 2195 cm for 1 to 2203 cm for 3. Electrical con-
2
ꢀ1
carbon monoxide, but it has an important role in the double
ductivity measurements (molar conductivity: 28 Scm mol at
258C) were in the range expected for a 1:1 electrolyte in such
a viscous solvent. These measurements also agreed with the
values found for the tris-isocyanide cationic complex 2 (molar
conductivity from extrapolation of the titration curve of 1 with
carbonylation of aryl halides to give a-ketoamide deriva-
[
7b,d,e]
tives.
Interaction of 1 with phosphine ligands commonly used in
Pd-catalyzed imidoylative reactions
2
ꢀ1
tBuNC in DMSO: 24 Scm mol ) and for the cationic species
2
ꢀ1
formed with PPh (28 Scm mol ).
3
Considering that phosphines are often used successfully to ac-
complish Pd-catalyzed transformations with isocyanide inser-
[
4b,c,10d–f]
tion,
we studied the interaction of the imidoylpalladium
complex 1 with representative ligands of this class. In this way,
we aimed to gain insight into the relative affinity of phos-
phines and isocyanides for palladium and to gather useful in-
formation about the resting state of the catalyst.
When a solution of 1 (20 mm in DMSO) was treated with
one equivalent of PPh , a very complex mixture resulted. At
3
Scheme 7. Reaction of complex 1 with dppe.
19
least eight resonances were present in the F NMR spectrum
31
and seven signals in the P NMR spectrum, some of which
were broad at room temperature, indicating multiple species
[
22]
in equilibrium. Addition of more PPh (up to 5.0 equiv) did
3
not simplify the spectra. To probe the formation of cationic
complexes, the interaction of PPh with 1 was also studied by
3
conductivity measurements. Assuming proportionality be-
tween the measured conductivity and the concentration of
ionic species in solution, about 61% of the total palladium is
present in the form of cationic complexes with one equivalent
of PPh , and 82% with two equivalents. The measured appar-
3
ꢀ
ent constant for the displacement of I by PPh is K=2.8 (see
3
the Supporting Information, Section 1.3). Given the complexity
of the system, no further attempt was made to identify the
species formed. Nevertheless, under the catalytic conditions
[
10g]
described by Jiang,
there is at least one equivalent of PPh
3
[
23]
with respect to Pd and excess isocyanide. As our data indi-
cates, a significant fraction of any imidoyl-Pd species formed is
thus of ionic nature. This observation should be considered
when interpreting the results recently proposed on the basis
of DFT calculations for the mechanism of the transformation
1
3
1
Figure 6. Section of the C{ H} NMR spectrum of complex 1 (62.5 mm in
]DMSO) in the presence of dppe (1.0 equiv) showing the region of the
[D
6
tert-butyl group quaternary carbon. Three peaks are evident: one for the
imine nitrogen substituent [d=59.5 ppm (s)], one for the coordinated isocy-
anide of 3 [d=57.7 ppm (dd, JC-P =11.5, 2.0 Hz)], and one for uncoordinated
[
9]
4
under study, because only PPh -ligated complexes were eval-
3
1
isocyanide [d=54.1 ppm (1:1:1 triplet, JC-N =4.5 Hz)].
uated and the formation of RNC-ligated species possibly over-
looked.
Similarly, bidentate phosphines are often useful in Pd-cata-
[
10d,e,24]
lyzed imidoylative couplings.
We selected dppe (1,2-di-
In [D ]acetone, which is less polar than DMSO, a mixture of
6
[25]
phenylphosphinoethane) as a representative example. Only
one species was formed quantitatively by interaction of equi-
molar amounts of complex 1 and dppe in DMSO (20 mm);
a sharp F NMR resonance and two doublets in the P NMR
spectrum were indicative. Surprisingly, dppe did not displace
the two isocyanide ligands, but instead displaced one mole-
two complexes, that we can tentatively identify as [(ArC=
+
NtBu)Pd(dppe)I] and [(ArC=NtBu)Pd(dppe)(CNtBu)]
was
formed with an approximate 30:70 ratio, as inferred by NMR
and IR data (see the Supporting Information, Section 3.2). The
stability of the cationic complex 3 towards excess isocyanide
was also tested. When it was left at room temperature in the
presence of ten equivalents of tBuNC for 72 h (as a 20 mm so-
lution in DMSO), about 80% of the material remained un-
19
31
ꢀ
cule of tBuNC and I , thus giving the cationic compound 3
[
26]
13
1
(Scheme 7).
This fact was confirmed by C{ H} NMR data
1
9
(
Figure 6, see the Supporting Information, Section 3.2 for
changed and several new resonances appeared in the F NMR
spectrum. Thus, this complex seems to be less stable towards
excess isocyanide than the tBuNC-ligated analogue 2.
a more detailed characterization), from which one molecule of
3
1
13
4
coordinated isocyanide (showing P, C J-coupling for the
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
In summary, imidoylpalladium complexes are prone to give
ionic species in polar solvents, both in the presence of excess
isocyanide, which necessarily happens when
a catalytic
amount of Pd is used, and even more strongly with common
phosphine ligands. To the best of our knowledge, this fact has
never been reported before. This should be taken into account
to design new imidoylative coupling reactions and to rational-
ize the experimentally observed reactivity. The cationic species
may also be implicated in the multiple insertion of isocyanide,
[
7b,d,e]
which occurs for the CO analogues.
This is relevant to cat-
alysis because poly-insertion and polymerization are side reac-
[
4b]
tions that often plague imidoylative couplings.
Figure 7. Kinetics for the appearance of the amide 4 by the reaction of
ꢀ
Reductive elimination in the presence of OH
1
(14 mm in a DMSO/THF/H
2
O mixture in 5:1:1 ratio by volume) in the pres-
NF 8.5 equiv, HF 1.5 equiv) at 258C, as deter-
ꢀ
ence of a F /HF buffer (nBu
mined by F NMR spectroscopy. In the inset: same data, in a semi-logarith-
mic plot (C and C are the concentrations of 4 at any time and at reaction
4
To validate our hypothesis that complex 1 is an intermediate
of the catalytic synthesis of secondary amides under ligandless
19
1
conditions (Scheme 3), we studied its reactivity with different
completion, respectively) evidencing first-order behavior.
ꢀ
sources of OH . When complex 1 (20 mm in [D ]DMSO) was
6
treated with tetrabutylammonium hydroxide (1.0–2.0 equiv,
dispensed as a 1.54m solution in water), instantaneous decom-
position to a variety of different fluorine-containing products
ensued, together with the generation of a dark red air-sensitive
Computational study of the catalytic cycle
To corroborate the kinetic data observed and to get insight
into the nature of the non-isolable intermediates in the catalyt-
ic cycle, we investigated the complete mechanism of the reac-
tion by means of DFT. We used the experimental observations
as a guide to explore the potential energy landscape; the over-
all cycle is reported in Figure 8. Oxidative addition of ArI (Ar=
0
material, which we speculate to be a Pd species. No effort
was made to characterize this complex mixture, because no
19
trace of the expected amide 4 was identified by F NMR. Simi-
lar results were obtained with KOH as the base in DMSO/H O
2
[
27]
°
ꢀ1
mixtures.
The reaction of 1 with stoichiometric or excess
4-F-C H ) with [Pd(CNtBu) ] (DG =22.1 kcalmol ) first gives
6 4 2 1
ꢀ
OH is thus not representative of the catalytic process.
the cis-complex I1 that can rearrange to the thermodynamical-
ly more stable trans isomer I2. Isocyanide insertion into the Cꢀ
Pd bond has a smaller activation barrier than the previous step
ꢀ
When a F / HF buffer was employed instead of strong
bases, in an aqueous-organic solvent mixture (nBu NF
4
°
ꢀ1
8.5 equiv and HF 1.5 equiv were added to 1 to give a final con-
(DG2 =19.5 kcalmol ), in agreement with the fact that inter-
mediate I2 is not observed experimentally. The coordinatively
unsaturated insertion product I3 can add a further molecule of
tBuNC to give 1, which is an isolable intermediate. The X-ray
structure of 1 compares well with the corresponding opti-
centration of 14 mm in a DMSO/THF/H O mixture in 5:1:1 ratio
2
[
28]
by volume ) a smooth reaction occurred that gave the ex-
19
1
pected amide 4 exclusively, as confirmed by F and H NMR
spectroscopic analyses of the reaction mixture. Notably, this re-
action proceeds at room temperature, albeit slowly, whereas
the catalytic reaction (see Scheme 3) is commonly performed
[29]
mized geometry (without implicit solvation, Kabsch RMSD:
[29]
0.71 ꢁ, excluding the Ar ring: 0.36 ꢁ), thus implicitly validat-
[
10g]
ꢀ
at 908C.
The kinetics of this transformation were readily ac-
ing the chosen level of theory. Iodide exchange with OH is
19
cessible by F NMR spectroscopy (Figure 7) and apparent first-
order behavior (t1/2 =26 h at 258C) was found.
energetically favorable and gives rise to I4. The greater affinity
+
ꢀ
ꢀ
of the [(ArC=NtBu)Pd(CNtBu)2] moiety for OH than for I ex-
ꢀ
The influence of other species present under catalytic condi-
tions in the reaction mixture was then evaluated. The isocya-
nide displayed a strong inhibitory effect; when the experiment
under the previously described conditions was repeated with
excess tBuNC (10 equiv), only 20% of the starting material was
converted after 42 h (instead of more than 60% previously). In
the presence of [(PhCN) PdCl ] (5 equiv), which acts as an iso-
plains the experimental observation that I does not strongly
inhibit the reaction. Complex I4 can easily convert into the cis
isomer I5. The latter, in turn, can undergo CꢀO bond-forming
°
ꢀ1
reductive elimination (DG3 =15.4 kcalmol ) that delivers the
amide tautomer I6. In the end, an energetically favorable keto-
enol tautomerization yields the amide 4 and regenerates
[Pd(CNtBu) ]. In a byproduct-forming pathway, 1 can otherwise
2
2
2
cyanide scavenger, complete conversion to 4 was obtained in
less than 3 h. Iodide slowed the reaction down slightly; when
undergo a second insertion of one of the coordinated tBuNC
into the CꢀPd bond through TS4 (not shown in Figure 8,
°
4
ꢀ1
the reaction was performed with the addition of nBu NI
DG =22.0 kcalmol ) to give the bis-inserted complex
4
ꢀ
(
10 equiv), the yield of 4 was 25% after 42 h. When the F /HF
[{Ar(C=NtBu) }Pd(CNtBu)I] (I7).
2
ratio was changed from 85:15 to about 1:1 the reaction
It is important to note that free energies calculated as usual
for all the species at 1 atm pressure (i.e., at a concentration of
became slower (t1/2 =44 h), thus suggesting that the reaction
ꢀ
ꢀ1
rate is an increasing function of [OH ] (but, as we already
about 0.04 molL within the ideal gas approximation) are not
ꢀ
noted, too high of a [OH ] causes degradation of the com-
representative of the experimental conditions, because the
ꢀ
ꢀ
plex).
concentration of OH secured by a F /HF buffer is much small-
&
&
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 8. Computed reaction path for the amidation of 4-fluoroiodobenzene with tBuNC. Solid line: free energies reported for all the species at concentration
ꢀ
ꢀ11
0
.04 m (1 atm). Dots: free energies for all the species at 0.04 m except [OH ]=10 m.
ꢀ
er. To give a rough estimate of [OH ], we can consider that the Conclusions
pK of HF in water is approximately 3.2, so that the order of
a
ꢀ
ꢀ
magnitude of [OH ] in an F /HF (1:1 ratio) aqueous buffer is
We have performed a mechanistic study on the Pd-catalyzed
amidation of aryl halides with isocyanides and water under
phosphine ligand-free conditions. We have established that
the isocyanide can play the role of ligand for Pd complexes in
all the steps of the catalytic cycle and it modulates their reac-
tivity in several ways.
ꢀ
11
ꢀ1
ꢀ
ꢀ11
ꢀ1
10
molL . Assuming [OH ]=10 molL , the difference in
ꢀ
1
free energy at 258C is approximately +13.1 kcalmol . Apply-
ing this correction, the potential energy profile of the last
steps changes considerably (Figure 8). More specifically, I4 is
no longer stable with respect to 1, which is in accordance with
the fact that I4 is not observed experimentally under condi-
tions close to the catalytic reaction.
We have studied the kinetics of the oxidative addition of
0
aryl iodides with RNC-ligated Pd . It was unexpected that
Considering that the reductive elimination is arguably the
most critical step of the catalytic cycle, we studied it in more
detail. Actually, several complexes featuring a hydroxo ligand
cis to the imidoyl moiety can be envisaged and take part into
this process. Thus, we evaluated their free energy with respect
to I5a taken as a reference, as well as the respective activation
barrier for CꢀO bond-forming reductive elimination. Results
obtained are summarized in Figure 9. We have considered the
two possible tricoordinated, T-shaped complexes I5b,c and
the four solvent-ligated species I5d–g. In accordance with the
excess RNC does not inhibit this step with the formation of 16-
electron complexes. We disclosed that the [(ArC=NR)Pd(CNR)2I]
species formed after insertion of RNC into the CꢀPd bond
ꢀ
have a surprisingly strong tendency to expel I and give cat-
ionic complexes with excess RNC or added phosphine ligands
in polar solvents. These RNC-ligated species undergo CꢀO
bond-forming reductive elimination at room temperature in
ꢀ
the presence of H O and a F /HF buffer. RNC hinders this last
2
step, most likely by formation of inactive cationic species and
preventing the formation of more active coordinatively unsatu-
rated complexes. Moreover, cationic complexes can probably
participate in side reactions due to poly-insertion of RNC. Fur-
ther studies are underway to apply our findings to the rational
development of new Pd-catalyzed reactions involving isocya-
nide insertion.
[
30]
thermodynamic trans effect, the tBuNC ligand trans to the
imidoyl moiety is remarkably easy to remove from the coordi-
ꢀ
1
nation sphere of Pd (DG=0.3 kcalmol ). The reductive elimi-
nation proceeds from complex I5b, formed with an activation
barrier that is smaller than that computed for I5a,c–g. It has
already been observed for other Pd complexes that CꢀC reduc-
tive elimination is easier for coordinatively unsaturated spe-
[
31]
cies. This observation may contribute to explain the inhibi-
tion of reductive elimination exerted by excess isocyanide and
the rate enhancement in the presence of [(PhCN) PdCl ] as an
Experimental Section
Synthesis of trans-[(4-F-C H C=NtBu)Pd(CNtBu) I] (1): A 25-mL
6
4
2
2
2
round-bottom flask with a gas inlet tube and a magnetic stirrer
isocyanide scavenger. On the basis of our findings, we can pro-
pose the revised mechanism depicted in Scheme 8, featuring
RNC-ligated complexes only. Intermediates and resting states
that have been isolated and characterized are highlighted in
rectangles.
was charged with [Pd(dba) ] (460 mg, 0.80 mmol) and conditioned
2
under an atmosphere of Ar. Degassed toluene (8 mL) and tert-butyl
isocyanide (360 mL, 3.2 mmol) were added sequentially by syringe.
The initially formed violet-black suspension changed to an orange-
yellow solution and the starting material dissolved completely in
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 9. Relative stabilities (calculated DG at 258C) of different imidoyl-palladium cis-hydroxo complexes I5a–g and of the respective transition states (TS)
for reductive elimination TS3a,b,d–g. It is not possible to locate a TS corresponding to I5c, because isomerization to I5b occurs.
Scheme 8. Revised catalytic cycle for the Pd-catalyzed synthesis of secondary amides from aryl iodides and isocyanides.
1
3
a few minutes. 4-Fluoroiodobenzene (185 mL, 1.6 mmol) was added
and the mixture was left overnight at room temperature. Workup
was done without special precautions to exclude air. Chromatog-
raphic grade silica gel (approx. 2 g) was added to the crude reac-
tion mixture and volatile components were removed by evapora-
tion under reduced pressure, while keeping the temperature at or
below 258C. The free-flowing powder obtained was then dry-
loaded onto a flash-chromatography column (diameter: 25 mm,
silica gel) and eluted with petroleum ether/EtOAc (gradient from
CDCl ): d=166.5, 163.0 (d, JCF =248.3 Hz), 141.6, 130.8 (d, J =
3 CF
2
8.2 Hz), 114.8 (d, J =21.5 Hz), 58.1, 57.4, 31.4, 29.9 ppm (the iso-
cyanide carbons bound to Pd gave no apparent peak); F{ H} NMR
CF
19
1
(282 MHz, CDCl ): d=ꢀ113.6 ppm; IR (ATR): n˜ =3054 (w), 3035 (w),
3
2983 (m), 2929 (w), 2865 (w), 2190 (s), 1645 (m), 1494 (m), 1215 (s),
ꢀ1
1190 (s), 1090 (m), 886 (s), 869 (m), 839 (s), 794 cm (s); ESI-MS
(positive ion mode, MeCN): m/z (%): 583.3 (7), 580.3 (96), 578.4
(100), 577.4 (80), 576.4 (48) [M+H] ; 526.4 (15), 524.4 (36), 522.3
+
+
(38), 521.3 (29), 520.4 (14) [M+HꢀC H ] ; 454.4 (22), 452.4 (44),
4
8
+
7
5:25 to 50:50 ratio). The fractions containing the product were
450.5 (54), 449.5 (44), 448.5 (18) [MꢀI] ; 371.2 (15), 369.2 (39),
+
combined and a few drops of tert-butyl isocyanide were added.
Evaporation at reduced pressure (at room temperature or below)
gave the title compound in the form of a yellow-orange solid;
yield: 441 mg (95%). X-ray quality crystals were obtained by slow
367.4 (40), 366.2 (26), 365.0 (19) [MꢀIꢀtBuNC] , 178.3 (18) [4-F-
+
C H C=NtBu] .
6
4
Synthesis of [(4-F-C H C=NtBu)Pd(CNtBu) ]OTf (2-OTf): To a solu-
6
4
3
evaporation of a concentrated solution in EtOAc/heptane (1:1).
tion of 1 (115 mg, 0.20 mmol) and tBuNC (24 mL, 0.21 mmol) in dry
1
H NMR (300 MHz, CDCl ): d=7.84 (dd, J=8.6, 5.7 Hz, 2H), 7.00 (t,
CH Cl2 (20 mL) was added finely powdered Tl(OTf) (176 mg,
3
2
13
J=8.6 Hz, 2H), 1.56 (s, 9H), 1.39 ppm (s, 18H); C NMR (75.5 MHz,
0.50 mmol) and the resulting suspension was stirred for 30 min at
&
&
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
room temperature. The mixture was filtered through a short plug
[3] For some selected reviews, see: a) J.-J. Brunet, R. Chauvin, Chem. Soc.
Rev. 1995, 24, 89–95; b) M. Beller, B. Cornils, C. D. Frohning, C. W. Kohl-
paintner, J. Mol. Catal. A. 1995, 104, 17–85; c) B. El Ali, H. Alper, Synlett
of diatomaceous earth and the solid was washed with more CH Cl2
2
(
3ꢂ5 mL). The combined filtrates were evaporated at reduced
2
000, 161–171; d) R. Skoda-Foldes, L. Kollar, Curr. Org. Chem. 2002, 6,
pressure, keeping the temperature below room temperature. The
title compound was obtained in the form of a pale yellow solid;
yield: 128 mg (94%). The compound was fairly stable in the solid
1
5
097–1119; e) T. Morimoto, K. Kakiuchi, Angew. Chem. Int. Ed. 2004, 43,
580–5588; Angew. Chem. 2004, 116, 5698–5706; f) A. M. Trzeciak, J. J.
Ziꢃłkowski, Coord. Chem. Rev. 2005, 249, 2308–2322; g) C. F. J. Barnard,
Organometallics 2008, 27, 5402–5422; h) R. Grigg, S. P. Mutton, Tetrahe-
dron 2010, 66, 5515–5548; i) X.-F. Wu, H. Neumann, M. Beller, Chem.
Soc. Rev. 2011, 40, 4986–5009; j) I. Omae, Coord. Chem. Rev. 2011, 255,
139–160; k) X.-F. Wu, H. Neumann, M. Beller, Chem. Rev. 2013, 113, 1–
state, but decomposed in a few hours when dissolved in common
1
organic solvents. H NMR (300 MHz, [D ]DMSO): d=7.69 (dd, J=
6
8
1
.7, 5.7 Hz, 2H), 7.25 (t, J=8.8 Hz, 2H), 1.53 (s, 9H), 1.48 (s, 9H),
1
9
.39 ppm (s, 18H); F NMR (282 MHz, [D ]DMSO): d=ꢀ77.8 ppm
6
13
35.
(
s, 3F), ꢀ112.6 (br s, 1F); C NMR (75.5 MHz, CD Cl ): d=163.8 (d,
2
2
1
3
1
[4] a) G. Qiu, Q. Ding, J. Wu, Chem. Soc. Rev. 2013, 42, 5257–8569; b) T.
Vlaar, E. Ruijter, B. U. W. Maes, R. V. A. Orru, Angew. Chem. Int. Ed. 2013,
J =249.4 Hz), 140.6, 131.3 (d, JCF =8.2 Hz), 121.5 (q, J =
CF
CF
2
3
2
21.3 Hz), 115.5 (d, JCF =21.6 Hz), 60.2, 59.5, 58.1, 31.7, 30.2,
5
2, 7084–7097; c) S. Lang, Chem. Soc. Rev. 2013, 42, 4867–4880; d) S.
9.8 ppm (carbons directly bound to Pd gave no visible signal); IR
Chakrabarty, S. Choudhary, A. Doshi, F.-Q. Q. Liu, R. Mohan, M. P. Ravin-
dra, D. Shah, X. Yang, F. F. Fleming, Adv. Synth. Catal. 2014, 356, 2135–
(
1
ATR): n˜ =3070 (w), 2983 (w), 2943 (w), 2241 (w), 2217 (s), 1617 (m),
593 (m), 1375 (m), 1265 (s), 1148 (s), 1030 (s), 893 (m), 849 cm
ꢀ
1
2196; e) V. P. Boyarskiy, N. A. Bokach, K. V. Luzyanin, V. Y. Kukushkin,
(
(
3
3
m); ESI-MS (positive ion mode, MeOH): m/z (%): 454.3 (35), 452.3
Chem. Rev. 2015, 115, 2698–2779.
+
72), 450.4 (80), 449.5 (68), 448.5 (30) [MꢀOTfꢀtBuNC] ; 371.2 (46),
[5] a) F. Millich, Chem. Rev. 1972, 72, 101–113; b) Y. Yamamoto, H. Yamazaki,
Coord. Chem. Rev. 1972, 8, 225–239; c) Y. Ito, M. Suginome, in Handbook
of Organopalladium Chemistry for Organic Synthesis (Ed.: E. Negishi),
Wiley, New York, 2003, pp. 2705–2712; d) M. Suginome, Y. Ito, in Advan-
ces in Polymer Science Vol. 171, Springer, Heidelberg, 2004, pp. 77–136;
e) P. Veya, C. Floriani, A. Chiesi-Villa, C. Rizzoli, Organometallics 1993, 12,
+
69.2 (95), 367.3 (100), 366.3 (98), 365.2 (41) [MꢀOTfꢀ2tBuNC] ;
15.3 (33), 313.3 (72), 311.3 (71), 310.3 (54), 309.3 (28) [MꢀOTfꢀ2t-
+
BuNCꢀC H ] ; ESI-MS (negative ion mode, MeOH): m/z (%): 149.3
4
8
ꢀ
(
100) [OTf] .
4899–4907; f) G. R. Owen, R. Vilar, A. J. P. White, D. J. Williams, Organo-
Computational details
metallics 2002, 21, 4799–4807; g) J. Vicente, J.-A. Abad, E. Martꢄnez-Viv-
iente, P. G. Jones, Organometallics 2002, 21, 4454–4467.
All DFT calculations were performed with the Gaussian09 program
[6] a) S. Otsuka, K. Ataka, J. Chem. Soc., Dalton Trans. 1976, 327–334; b) A.
Campagnaro, A. Mantovani, P. Uguagliati, Inorg. Chim. Acta 1985, 99,
L15–L19; c) R. Bertani, A. Berton, F. Di Bianca, B. Crociani, J. Organomet.
Chem. 1986, 303, 283–299; d) J. G. P. Delis, P. G. Aubel, P. W. N. M. van
Leeuwen, K. Vrieze, N. Veldman, A. L. Spek, J. Chem. Soc., Chem.
Commun. 1995, 2233–2234; e) J. G. P. Delis, P. G. Aubel, K. Vrieze,
P. W. N. M. van Leeuwen, N. Veldman, A. L. Spek, F. J. R. van Neer, Orga-
nometallics 1997, 16, 2948–2957; f) Y. Kayaki, I. Shimizu, A. Yamamoto,
Bull. Chem. Soc. Jpn. 1997, 70, 917–927; g) L. Canovese, F. Visentin, C.
Santo, C. Levi, A. Dolmella, Organometallics 2007, 26, 5590–5601; h) L.
Canovese, F. Visentin, C. Santo, C. Levi, Organometallics 2009, 28, 6762–
6770; i) L. Canovese, G. Chessa, F. Visentin, Inorg. Chim. Acta 2010, 363,
3426–3431; j) L. Canovese, F. Visentin, C. Santo, V. Bertolasi, Organome-
tallics 2014, 33, 1700–1709.
[32]
(
Rev. A02).
The structures of all minima and transition states
[33]
were fully optimized using the M06-2X functional and the fol-
lowing basis set: LANL2DZ with the associate effective core poten-
tial for Pd, 6-31+G(d) for N, O, S, P, and C atoms directly bound
to Pd, 6-311G* for I, and 6-31G for all the others atoms. Bulk sol-
vent effects were accounted for by using an implicit solvation
[34]
[35]
model (IEF-PCM) as implemented in Gaussian.
The default
spheres radii, static, and optical dielectric constants were used. All
stationary points were characterized as minima or first-order transi-
tion states by analytical frequency calculations. IRC (intrinsic reac-
tion coordinate) calculations were used to confirm that transition
states are linked to the proper minima. Computed harmonic fre-
quencies were also used to calculate the free energy of all mole-
cules at T=298 K and P=1 atm assuming an ideal gas behavior.
[
7] a) J. T. Chen, A. Sen, J. Am. Chem. Soc. 1984, 106, 1506–1507; b) F.
Ozawa, H. Soyama, H. Yanagihara, I. Aoyama, H. Takino, K. Izawa, T. Ya-
mamoto, A. Yamamoto, J. Am. Chem. Soc. 1985, 107, 3235–3245; c) F.
Ozawa, N. Kawasaki, H. Okamoto, T. Yamamoto, A. Yamamoto, Organo-
metallics 1987, 6, 1640–1651; d) L. Huang, F. Ozawa, A. Yamamoto, Or-
ganometallics 1990, 9, 2603–2611; e) A. Yamamoto, F. Ozawa, K. Osaka-
da, L. Huang, T. Son, N. Kawasaki, M.-K. Doh, Pure Appl. Chem. 1991, 63,
Acknowledgements
6
3
87–696; f) Y.-S. Lin, A. Yamamoto, Organometallics 1998, 17, 3466–
478; g) P. W. N. M. van Leeuwen, M. A. Zuideveld, B. H. G. Swennenhuis,
We thank Dr. Anny Jutand for fruitful discussions and Dr. Pat-
rick Herson (UPMC) for X-ray diffraction analysis. We thank
CNRS, ENSCP, and ENS, as well as the ANR program CD2I (proj-
ect CuFeCCBond) for financial support. National computing
center (IDRIS) is acknowledged for computing time granted.
Z. Freixa, P. C. J. Kamer, K. Goubitz, J. Fraanje, M. Lutz, A. L. Spek, J. Am.
Chem. Soc. 2003, 125, 5523–5539.
[8] a) R. D. O’Sullivan, A. W. Parkins, J. Chem. Soc. Chem., Commun. 1984,
165–1166; b) J. Dupont, M. Pfeffer, J. Chem. Soc., Dalton Trans. 1990,
193–3319; c) J. Vicente, I. Saura-Llamas, J. Turpin, M. C. Ramirez de Ar-
1
3
ellano, P. G. Jones, Organometallics 1999, 18, 2683–2693; d) J. Vicente,
J.-A. Abad, W. Foertsch, P. G. Jones, A. K. Fischer, Organometallics 2001,
Keywords: coupling · insertion · isocyanide · palladium ·
reaction mechanisms
2
0, 2704–2715; e) J. Vicente, J.-A. Abad, A. D. Frankland, J. Lopez-Serra-
no, M. C. Ramirez de Arellano, P. G. Jones, Organometallics 2002, 21,
72–282; f) J. Vicente, J.-A. Abad, E. Martinez-Viviente, P. G. Jones, Orga-
2
[
1] For some selected reviews: a) A. Dçmling, Chem. Rev. 2006, 106, 17–89;
b) J. E. Biggs-Houck, A. Younai, J. T. Shaw, Curr. Opin. Chem. Biol. 2010,
nometallics 2003, 22, 1967–1978; g) G. R. Owen, R. Vilar, A. J. P. White,
D. J. Williams, Organometallics 2003, 22, 4511–4521; h) J. Vicente, J.-A.
Abad, W. Fçrtsch, M.-J. Lꢃpez-Sꢅez, P. G. Jones, Organometallics 2004,
23, 4414–4429; i) J. Vicente, J.-A. Abad, F. S. Hernandez-Mata, B. Rink,
P. G. Jones, M. C. Ramirez de Arellano, Organometallics 2004, 23, 1292–
1304; j) J. Vicente, A. Arcas, J. M. Fernꢅndez-Hernꢅndez, D. Bautista, P. G.
Jones, Organometallics 2005, 24, 2516–2527; k) J. Vicente, M. T. Chicote,
A. J. Martinez-Martinez, P. G. Jones, D. Bautista, Organometallics 2008,
27, 3254–3271; l) J. Vicente, M. T. Chicote, A. J. Martinez-Martinez, D.
Bautista, Organometallics 2009, 28, 5915–5924; m) J. Vicente, I. Saura-
14, 371–382; c) L. El Kaim, L. Grimaud, Tetrahedron 2009, 65, 2153–
2171; d) A. V. Gulevich, A. G. Zhdanko, R. V. A. Orru, V. G. Nenajdenko,
Chem. Rev. 2010, 110, 5235–5331; e) Isocyanide Chemistry: Applications
in Synthesis and Material Science (Ed.: V. G. Nenajdenko), Wiley-WCH,
Weinheim, 2012.
[2] a) S. Otsuka, A. Nakamura, T. Yoshida, J. Am. Chem. Soc. 1969, 91, 7196–
7198; b) Y. Yamamoto, H. Yamazaki, Bull. Chem. Soc. Jpn. 1970, 43,
2653–2653.
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Llamas, J.-A. Garcia-Lopez, D. Bautista, Organometallics 2009, 28, 448–
64; n) J. Vicente, M. T. Chicote, A. J. Martinez-Martinez, A. Abellan-
cies is predicted to be significantly less stable than the monomer (DG8
for the trimerization: +26.0 kcalmol ).
ꢀ
1
4
Lopez, D. Bautista, Organometallics 2010, 29, 5693–5707; o) J. Vicente,
A. Arcas, F. Julia-Hernandez, D. Bautista, P. G. Jones, Organometallics
[20] a) CCDC 1485600 (1) contains the supplementary crystallographic data
for this paper. These data are provided free of charge by The Cam-
bridge Crystallographic Data Centre. For the synthesis of some similar
complexes, see: b) M. Morishita, H. Amii, J. Organomet. Chem. 2007,
692, 620–624.
[21] The anion exchange with silver salts invariably failed despite trying dif-
ferent reaction conditions, due to the great affinity of isocyanides for
AgI. AgI is soluble in organic solvents in the presence of excess tBuNC.
[22] An analogously complex mixture was obtained when excess tBuNC was
2010, 29, 3066–3076; p) J. Vicente, J. A. Abad, R.-M. Lopez-Nicolas, P. G.
Jones, Organometallics 2011, 30, 4983–4998; q) J. Vicente, P. Gonzalez-
Herrero, R. Frutos-Pedreno, M.-T. Chicote, P. G. Jones, D. Bautista, Orga-
nometallics 2011, 30, 1079–1093; r) M.-T. Chicote, I. Vicente-Hernandez,
P. G. Jones, J. Vicente, Organometallics 2012, 31, 6252–6261; s) M.-J.
Oliva-Madrid, J.-A. Garcia-Lopez, I. Saura-Llamas, D. Bautista, J. Vicente,
Organometallics 2012, 31, 3647–3660; t) R. Frutos-PedreÇo, P. Gonzꢅlez-
Herrero, J. Vicente, P. G. Jones, Organometallics 2012, 31, 3361–3372;
u) J. Vicente, M.-T. Chicote, A. Abellꢅn-Lꢃpez, D. Bautista, Dalton Trans.
added to [ArPd(PPh
[23] As stated previously, one of the two equivalents of PPh
dized to PPh O.
3 2
) I] complexes.
3
can be oxi-
2012, 41, 752–762; v) A.-J. Martꢄnez-Martꢄnez, J. Vicente, M.-T. Chicote,
3
D. Bautista, Organometallics 2012, 31, 2697–2708; w) A. Abellꢅn-Lꢃpez,
M.-T. Chicote, D. Bautista, J. Vicente, Organometallics 2013, 32, 7612–
[24] a) T. Vlaar, E. Ruijter, A. Znabet, E. Janssen, F. J. J. de Kanter, B. U. W.
Maes, R. V. A. Orru, Org. Lett. 2011, 13, 6496–6499; b) K. Onitsuka, S.
Suzuki, S. Takahashi, Tetrahedron Lett. 2002, 43, 6197–6199; c) S.
Kamijo, T. Jin, Y. Yamamoto, J. Am. Chem. Soc. 2001, 123, 9453–9454;
d) X.-D. Fei, Z.-Y. Ge, T. Tang, Y.-M. Zhu, S.-J. Ji, J. Org. Chem. 2012, 77,
7624; x) M.-J. Oliva-Madrid, I. Saura-Llamas, D. Bautista, J. Vicente, Chem.
Commun. 2013, 49, 7997–7999; y) R. Frutos-PedreÇo, P. Gonzꢅlez-Her-
rero, J. Vicente, P. G. Jones, Organometallics 2013, 32, 4664–4676; z) R.
Frutos-PedreÇo, P. Gonzꢅlez-Herrero, J. Vicente, P. G. Jones, Organome-
tallics 2013, 32, 1892–1904; aa) M.-J. Oliva-Madrid, J.-A. Garcia-Lopez, I.
Saura-Llamas, D. Bautista, J. Vicente, Organometallics 2014, 33, 6420–
1
0321–10328.
25] An analogous behavior was found for the wider bite-angle phosphine
,1’-bis(diphenylphosphino)ferrocene (dppf).
[
1
[
[
26] For a partly related literature precedent, see reference [6b].
27] A 20% water content was necessary to maintain the 10 equiv of KOH
6
430; ab) M.-J. Fernꢅndez-Rodrꢄguez, E. Martinez-Viviente, J. Vicente,
P. G. Jones, Organometallics 2015, 34, 3282–3291; ac) I. Vicente-Hernꢅn-
dez, M.-T. Chicote, J. Vicente, D. Bautista, Chem. Commun. 2016, 52,
(
dispensed as a 1.0m aqueous solution) completely dissolved. With less
water, the base precipitated in colloidal form and the required product
actually formed, because the slight dissolution generated a low concen-
5
94–596.
9] Y. Liang, Y. Ren, J. Jia, H.-S. Wu, J. Mol. Model. 2016, 22, 53.
10] a) T. Kajimoto, H. Takahashi, J. Tsuji, J. Organomet. Chem. 1970, 23, 275–
80; b) M. Kosugi, T. Ogata, H. Tamura, H. Sano, T. Migita, Chem. Lett.
986, 1197–1200; c) T. Ishiyama, T. Oh-e, N. Miyaura, A. Suzuki, Tetrahe-
[
ꢀ
ꢀ
tration of OH , in the same way as the F /HF buffer (vide infra).
28] The choice of this ternary mixture was done to mimic the catalytic con-
ditions as close as possible (refer to Scheme 3) and to ensure the com-
plete solubility of all the components. Water was essential for the reac-
tion to proceed towards the desired amide 4, rapid decomposition was
observed in its absence.
[
[
2
1
dron Lett. 1992, 33, 4465–4468; d) C. G. Saluste, R. J. Whitby, M. Furber,
Angew. Chem. Int. Ed. 2000, 39, 4156–4158; Angew. Chem. 2000, 112,
4326–4328; e) C. G. Saluste, R. J. Whitby, M. Furber, Tetrahedron Lett.
[
[
29] W. Kabsch, Acta Crystallogr. Sect. A 1976, 32, 922–923.
30] T. G. Appleton, H. C. Clark, L. E. Manzer, Coord. Chem. Rev. 1973, 10,
2
001, 42, 6191–6194; f) K. K. R. Tetala, R. J. Whitby, M. E. Light, M. B.
Hurtshouse, Tetrahedron Lett. 2004, 45, 6991–6994; g) H. Jiang, B. Liu, Y.
Li, A. Wang, H. Huang, Org. Lett. 2011, 13, 1028–1031.
335–422.
[
[
31] M. Pꢇrez-Rodrꢄguez, A. A. C. Braga, M. Garcia-Melchor, M. H. Pꢇrez-Tem-
prano, J. A. Casares, G. Ujaque, A. R. de Lera, R. ꢈlvarez, F. Maseras, P. Es-
pinet, J. Am. Chem. Soc. 2009, 131, 3650–3657.
[
[
[
11] J. M. French, J. R. Griffiths, S. T. Diver, Adv. Synth. Catal. 2015, 357, 361–
3
65.
12] L. A. Perego, L. Grimaud, F. Bellina, Adv. Synth. Catal. 2016, 358, 597–
09.
13] a) C. Amatore, A. Jutand, M. A. M’Barki, Organometallics 1992, 11, 3009–
013; b) C. Amatore, E. Carre, A. Jutand, M. A. M’Barki, Organometallics
995, 14, 1818–1826.
14] L. Malatesta, J. Chem. Soc. 1955, 3924–3926.
15] Actually, a colorless solution of PdCl (0.2m in DMSO) and CyNC
4.0 equiv, Cy=cyclohexyl) changed to yellow upon treatment with
32] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, 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. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, 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, ꢉ. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, D. J. Fox, Gaussian, Inc., Wallingford CT., 2009.
6
3
1
[
[
2
(
2
NaOH (1m in H O, 1.0 equiv) and CyNCO was detected by GC analysis
after one minute.
[
16] a) A. Christofides, J. Organomet. Chem. 1983, 259, 355–365; b) 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,
1
35, 8388–8399.
17] J.-F. Fauvarque, F. Pflꢆger, M. Troupel, J. Organomet. Chem. 1981, 208,
19–427.
[
[
[
[
33] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215–241.
34] a) P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299; b) P. J. Hay, W. R.
Wadt, J. Chem. Phys. 1985, 82, 270.
35] a) J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999–3093;
b) M. Cossi, G. Scalmani, N. Rega, V. Barone, J. Chem. Phys. 2002, 117,
4
18] a) E. O. Fischer, H. Werner, Chem. Ber. 1962, 95, 703–708; b) C. G. Fran-
cis, S. I. Khan, P. R. Morton, Inorg. Chem. 1984, 23, 3680–3681; c) Y. Ya-
mamoto, H. Yamazaki, J. Chem. Soc., Dalton Trans. 1989, 2161–2166;
d) L. A. Labios, M. D. Millard, A. L. Rheingold, J. S. Figueroa, J. Am. Chem.
Soc. 2009, 131, 11318–11319.
[
43–54.
[19] The existence of a stable triangulo-[Pd(m
2
-CNtBu)(CNtBu)]
3
trimer in so-
lution, analogous to [Pd(m -CNCy)(CNCy)]
2
3
which has been characterized
Received: June 17, 2016
Published online on && &&, 0000
[
18b]
in the solid state,
has also been examined computationally. This spe-
&
&
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
Reaction Mechanisms
L. A. Perego,* P. Fleurat-Lessard,
L. El Ka ¨ı m, I. Ciofini,* L. Grimaud*
&& – &&
Multiple Roles of Isocyanides in
Palladium-Catalyzed Imidoylative
Couplings: A Mechanistic Study
Double duty: Isocyanides are substrates
and ligands at the same time in Pd-cata-
lyzed imidoylative couplings. A detailed
mechanistic study of the amidation of
aryl halides explains how the substrate
tunes the reactivity of intermediates by
forming unexpected cationic resting
states, and rationalizes the proper
choice of the base in these couplings
(see scheme).
Chem. Eur. J. 2016, 22, 1 – 11
www.chemeurj.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ