Mechanistic Study of Pd/NHC-Catalyzed Sonogashira Reaction: Discovery of NHC-Ethynyl Coupling Process

Article abstract of DOI:10.1002/chem.202003533

The product of a revealed transformation—NHC-ethynyl coupling—was observed as a catalyst transformation pathway in the Sonogashira cross-coupling, catalyzed by Pd/NHC complexes. The 2-ethynylated azolium salt was isolated in individual form and fully char

Full text of DOI:10.1002/chem.202003533

Accepted Article
Accepted Article
Title: Mechanistic study of Pd/NHC-catalyzed Sonogashira reaction:
discovery of NHC-ethynyl coupling process
Authors: Dmitry Eremin, Daniil Boiko, Alexander Kostyukovich, Julia
Burykina, Ekaterina Denisova, Mariarosa Anania, Jonathan
Martens, Giel Berden, Jos Oomens, Jana Roithová, and
Valentine P. Ananikov
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.202003533
Link to VoR: https://doi.org/10.1002/chem.202003533
01/2020

10.1002/chem.202003533
Chemistry - A European Journal
Mechanistic study of Pd/NHC-catalyzed Sonogashira reaction:
discovery of NHC-ethynyl coupling process
Dmitry B. Eremin,[a],[b],† Daniil A. Boiko, [a],† Alexander Yu. Kostyukovich,[a] Julia V. Burykina,[a]
Ekaterina A. Denisova,[a] Mariarosa Anania,[c] Jonathan Martens,[d] Giel Berden,[d]
Jos Oomens,[d] Jana Roithová,[c] and Valentine P. Ananikov,[a],*
[a]
Dr. Dmitry B. Eremin,† Daniil A. Boiko,† Dr. Alexander Yu. Kostyukovich, Julia V. Burykina, Ekaterina A. Denisova, Prof. Dr. Valentine P. Ananikov*
Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky Pr. 47, Moscow 119991, Russia
† These authors contributed equally.
*E-mail: val@ioc.ac.ru
[b]
[c]
Dr. Dmitry B. Eremin
The Bridge@USC, University of Southern California, 1002 Childs Way, Los Angeles, California 90089-3502, United States
Mariarosa Anania, Prof. Dr. Jana Roithová
Department for Spectroscopy and Catalysis, Institute for Molecules and Materials, Radboud University, Heyendaalseweg 135, 6525 AJ, Nijmegen, the
Netherlands
[d]
Dr. Jonathan Martens, Dr. Giel Berden, Prof. Dr. Jos Oomens
FELIX Laboratory, Institute for Molecules and Materials, Radboud University, Toernooiveld 7, 6525 ED, Nijmegen, the Netherlands
Supporting information for this article is given via a link at the end of the document.
Abstract: The product of a revealed transformation — NHC-ethynyl
coupling — was observed as a catalyst transformation pathway in
the Sonogashira cross-coupling, catalyzed by Pd/NHC complexes.
The 2-ethynylated azolium salt was isolated in individual form and
fully characterized, including the X-ray analysis. A number of
possible intermediates of this transformation with common formulae
(NHC)xPd(C2Ph) (x = 1,2) were observed and subjected to collision-
induced dissociation (CID) and infrared multiphoton dissociation
(IRMPD) experiments to elucidate their structure. Measured bond
dissociation energies (BDEs) and IRMPD spectra were in an
excellent agreement with quantum calculations for coupling product
π-complexes with Pd(0). Molecular dynamics simulations confirmed
the observed multiple CID fragmentation pathways. An
unconventional methodology to study catalyst evolution suggests the
reported transformation to be considered in the development of new
catalytic systems for alkyne functionalization reactions.
Introduction
Scheme 1. Sonogashira coupling catalyzed by Pd/NHC complexes.
Functionalized acetylenes play a key role in natural products,
are actively utilized as synthetic intermediates, and find their
applications in pharmaceuticals and organic materials.[1-3] Over
the last decades, Sonogashira coupling has become one of the
key approaches towards alkyne functionalization in modern
organic synthesis.[4-12] Amongst numerous palladium catalysts
utilized in this transformation Pd/NHC (NHC = N-heterocyclic
carbene) complexes have firmly established their utility as highly
active catalysts which are much more stable towards air and
moisture than popular Pd-phosphine complexes.[13-15] This
important feature makes it possible to perform the reaction in
aqueous media.
The reaction usually requires copper(I) salts as a co-catalyst,
which (i) are unstable toward oxygen and may be easily oxidized
by some organic compounds and (ii) often catalyze Glaser-type
side reactions. These problems may be overcome by application
of other co-catalysts, such as AgOTf, or by avoiding any metal
co-catalysts. The latter option led to use of Pd/NHC catalysts in
Cu-free reaction conditions (Scheme 1).[16-22]
Although
a general rationalization of the mechanism has
enabled activation of different classes of electrophiles and the
development of synthetic protocols, a detailed mechanistic
understanding of the complete reaction pathways and especially
catalyst stability issues remain a challenge. Ligands of metal
complexes may not only facilitate the reactivity and stabilize
intermediates, but react with each other.[23] On the one hand,
these reactions may be crucial for the catalyst performance, if
the ligands participate in the catalyst activation step; on the
other hand, they may be responsible for the catalyst deactivation,
leading to the loss of active species and preventing possible
recycling of the catalysts with well-defined structure.[23]
Our attempts to understand the role of NHC ligands in the Pd-
catalyzed Sonogashira reaction led to discovery of
transformation, which, to the best of our knowledge, was not
reported before: NHC-ethynyl coupling (Scheme 1). The
transformation was extensively studied by means of mass-
spectrometry, DFT calculations, and molecular dynamics; and
a
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
the Pd-intermediates structures were confirmed. We found that
this transformation occurs from one of the key reaction
intermediates and can be responsible for the limited success in
the development of advanced synthetic approaches when
neglected.
substituents in the system. This observation encouraged us to
further study the nature of these species. Noteworthy, 2-
ethynylated azolium cations is a rare and interesting class of
organic moieties.[33-35]
Table 1. Product yields in the Sonogashira coupling reaction under varying
conditions a
Results and Discussion
In the established mechanism of this reaction, prior to the
reductive elimination step leading to the product formation, an
intermediate having an NHC-bound palladium with phenyl and
acetylide groups is proposed.[24-27] According to our previous
study, such an intermediate can undergo R-NHC coupling and
activate NHC-disconnected type of catalysis.[23, 28] In the present
study, we have performed the Sonogashira coupling of
iodobenzene (PhI) and phenylacetylene (PhC2H) using three
different Pd/NHC complexes 1a,c,d (Chart 1). The reactions
were conducted under three different conditions: Cu-free, with
CuBr, and with CuI.
Entry
Pd source
Cu source
Cu-free
Yield of tolane, %
1
2
3
4
5
6
7
8
9
1a
1c
1d
1a
1c
1d
1a
1c
1d
27 (23)[b]
32
38
90
CuBr
CuI
85
85
85
81
85
[a] Reaction conditions: 1 eq. of PhI and PhC2H, 3 eq. of Cs2CO3, 1 mol% of
Pd, 3 mol% of CuX (if stated), in DMSO, 1 h at 80 ºC. [b] Isolated yield (GC-
MS).
Chart 1. Pd/NHC complexes studied for NHC-ethynyl coupling (Py – pyridine,
Mes – 1,3,5-trimethylphenyl, Dipp – 1,3-diisopropylphenyl)
After 1 h product yields were determined by GC-MS (Table 1)
and the reaction mixtures were analyzed by ESI-MS. Depending
on the precatalyst and on the reaction conditions, tolane yields
varied from low to high. Cu-free conditions appeared to be
unselective resulting in 17-31% yields (Table 1, Entries 1-3).
Moreover, tolane was not the major product for the complexes
with BIMe (1a) and IMes (1c) ligands. As a major reaction
product, we have isolated and characterized (Z)-but-1-en-3-yne-
1,2,4-triyltribenzene (50% isolated yield), which has also been
reported previously.[29-31] In the presence of 3 mol% copper, the
reaction appeared to become significantly more efficient with up
to 90% yield, with no significant difference between CuBr and
CuI (Table 1, Entries 4-9).
In order to characterize the product of the NHC-ethynyl coupling,
we performed the reaction between complex 1a and PhC2Li
solution at –78 ºC, but the attempts to directly isolate 2a have
failed. It was not possible to isolate individual compounds due to
rapid polymerization, especially at elevated temperatures. This
problem was overcome by addition of AgOTf to the reaction
mixture (Scheme 2). Silver bound the iodide under the reaction
conditions, and thus increased the coupling rate relative to the
polymerization rate by conversion of the complex to its cationic
form. The ESI-MS analysis of the reaction mixture immediately
after addition of AgOTf confirmed the presence of [BIMe-C2Ph]+
and OTf– and the absence of iodide in the system. The BIMe-
C2Ph+OTf– (2a⋅OTf–) salt was isolated with 10% yield (12% NMR
yield) and characterized by 1H, 13C{1H}, and 19F NMR, FT-IR,
ESI-MS, IRMPD, and X-ray analysis (Figure 1). Even within the
optimized conditions, highly reactive nature and susceptibility to
polymerization challenges isolation of pure 2a⋅OTf– and affects
the yield significantly.
Scheme 2. Transformation of Pd/NHC complex into the NHC-ethynyl coupling
product
Analysis of the mass spectra (ESI) of all reaction mixtures
32]
consistently along with expected [NHC-Ph]+ cations[28,
revealed the formation of an NHC-ethynyl coupling product.
Cations [BIMe-C2Ph]+, [IMes-C2Ph]+, and [IPr-C2Ph]+ were
abundant in the mass spectra and confirmed the presence of
reactive intermediates of Pd with the NHC and ethynyl
Figure 1. Molecular structure of 2a determined by single crystal X-ray analysis.
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
Indeed, unusual structure and side-reactions, easily occurring in
a typical reaction mixture, complicate mechanistic studies and
probably result in these compounds not being reported earlier.
Due to ionic nature of proposed intermediates and catalyst
evolution product, mass spectrometry offers the best capabilities
to solve mechanistic problems as electrospray ionization is a
soft ionization technique. In such a case, utilization of ESI-MS,[24,
the Dipp ligand only at one nitrogen atom, whereas the second
nitrogen atom bears a hydrogen atom or the hydrogen atom
could be transferred to palladium (see Figure 2). This is
supposedly the energetically favored fragmentation pathway, but
disfavored entropically at larger collision energies, where the
simple elimination of bare palladium prevails.
36]
CID[37-38], and IRMPD[39-41] is the best choice to get an insight
Table 2. Appearance energies of NHC-ethynyl cations obtained from CID
MS/MS experiments with helium collision gas and corresponding DFT-
calculated bond dissociation energy, BDE (PBE1PBE/Def2TZVP GD3BJ
theory level)
into the system.
To trap Pd-containing intermediates in the reaction mixtures, we
mixed solutions of 1a-d in dry THF with a solution of PhC2Li in
THF at –78 ºC (PhC2Li was obtained by an NMR controlled
procedure to avoid PhC2H or n-BuLi contamination as described
in the Supporting Information). ESI-MS analysis of these
reaction mixtures revealed abundant signals of the NHC-ethynyl
coupling products. Moreover, for the reaction mixtures of BMIM
(1b), IMes (1c), and IPr (1d) complexes we were able to detect
[(NHC)n(C2Ph)Pd]+ cations, which could be the reactive
intermediates of the NHC-ethynyl coupling. With BIMe as the
NHC, the same ion was only detected at sufficient abundance
after addition of AgOTf to the reaction mixture, binding
excessive amounts of iodine and likely substituting the ligand by
non-coordinating TfO– (see SI).
Collision induced dissociation experiments. We have studied
the fragmentation of the [(NHC)n(C2Ph)Pd]+ (n = 1 or 2) cations.
These complexes can either represent the palladium(II)
intermediates or the palladium(0) coordinated products. In both
alternatives, we expect that the fragmentation should lead to the
elimination of palladium(0) either by the reductive elimination
mechanism or by a simple ligand detachment. Hence, we
decided to measure energy-resolved collision induced
dissociations (CIDs) that permit extraction of bond dissociation
energies[42-48] for the fragmentation channels.[47]
BDEexp
BDEcalc,
Entry
Parent ion
kcal/mol[a]
kcal/mol
1
2
3
4
[(IMes-C2Ph)Pd]+
61 ± 5
63 ± 5
60 ± 5
63 ± 5
54.5
56.0
51.6
56.3
[(IPr-C2Ph)Pd]+
[(BIMe)(BIMe-C2Ph)Pd]+
[(BMIM)(BMIM-C2Ph)Pd]+
[a] The experimental errors are generous based on possible systematic shift
due to the reactivity of the complexes with background gases.
These experimental results suggest that the dominant,
kinetically preferred palladium elimination most likely happens
from simple bond cleavage implying that the isolated complexes
correspond to the product π-complexes. The reductive
elimination might represent a kinetically preferred pathway
compared to hydrogen-rearrangement reactions involving
ligands. The energy demands for π-complex dissociations and
reduction eliminations are similar, albeit the latter processes
usually are slightly higher in energy.
The isolated palladium complexes in the ion trap readily reacted
with background gases (with N2 and H2O) indicating that
palladium in the complexes is highly coordinatively unsaturated.
This interaction with background gases led to artifact signals
during the mass-isolation of the complexes and therefore might
have resulted in a larger experimental error in the determination
of bond dissociation energies (BDEs) from the CID spectra. The
fragmentation of all [(NHC)n(C2Ph)Pd]+ complexes (n = 1 or 2)
led to the product cations [NHC-C2Ph]+ concomitant with the
eliminated palladium(0) (either naked for n = 1 or bearing one
NHC ligand for n = 2). This major fragmentation pathway was
accompanied by a minor or, in the case of [(IPr)(C2Ph)Pd]+, a
substantial fragmentation of the NHC ligand. The energy-
resolved CIDs illustrating the dissociation pathways are given in
the Supporting information (Figures S34-S37). The derived
experimental bond-dissociation energies for the elimination of
palladium(0) are summarized in Table 2. Taking into account
possible limitations in the accuracy of theoretical calculations, a
reasonable correlation between the experimental and calculated
values should be mentioned.
The fragmentation experiments in the ion trap correspond to a
slow heating of the ions in multiple collisions with helium. Hence,
next to the kinetically favored elimination of bare palladium,
fragmentations leading via constrained transition structures
could be also observed. This fact is the most evident in the
fragmentation of [(IPr)(C2Ph)Pd]+. The initial fragmentation of
[(IPr)(C2Ph)Pd]+ leads to elimination of one of the [Dipp – H]
molecules leading to the fragment in which the NHC part bears
Figure 2. Possible kinetically favored fragmentation pathway of
[(IPr-C2Ph)Pd]+.
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
Figure 3. Infrared multiple photon dissociation and computed IR spectra. (a) DFT calculated spectra for cis- or trans-σ- and π-complexes and experimental
spectrum of 8b; (b) DFT calculated spectra for cis- or trans-σ- and π-complexes and experimental spectrum of 5b.
As discussed above, three isomers of studied mono-NHC
intermediate are possible: trans- or [cis-NHC-Pd-C2Ph]+ (3 and 4
correspondingly) or a π-complex [(NHC-C2Ph)Pd]+ (5). Also, for
the bis-NHC intermediates, three isomers are conceivable:
trans- or [cis-(NHC)2-Pd-C2Ph]+ (6 and 7 correspondingly) or π-
complex [(NHC)(NHC-C2Ph)Pd]+ (8). NHC-ethynyl coupling
CID of the relevant Pd containing species. Moreover, the IRMPD
spectrum of 2a matches with its spectrum in the condensed
phase, showing that the condensed phase structure is
conserved when isolated in the mass spectrometer.
Computational modelling. Reductive elimination of the NHC
and ethynyl lgands has a low activation barrier of 14.5 kcal/mol.
This energy barrier is close to the NHC-Ph coupling (13.9
kcal/mol), however higher than the actual reductive elimination
leading to the reaction product (7.7 kcal/mol; see Supporting
Information for the computational details).
cation
2 can be formed after reductive elimination from
complexes 4 and either 6 or 7, followed by dissociation of the
corresponding π-complex. To prove the structures of the
isolated complexes, we have measured their infrared
multiphoton dissociation (IRMPD) spectra.[39]
In this section, we analyze the electronic structure of the
detected palladium complexes. The calculated enthalpy values
( H) for the dissociation reactions 5 → 2 + Pd (for systems with
one NHC) and 8 → 2 + Pd-NHC (for systems with two NHCs)
vary from 44.3 to 56.3 kcal/mol (Figure 4). Compounds 5a and
5b have significantly lower H compared with 8a and 8b. This is
an unusual result since in most cases the ligands in the trans-
position lower the binding energy to the Pd atom, as if they tend
to push each other out of the coordination sphere of the metal.
In this case, the [NHC-C2Ph]+ ligand is a cation; therefore, the
ligand-metal electrostatic interaction is of relevance. The
Mulliken charges on the palladium atom in Pd-NHC are negative,
since NHC is a donor ligand. In the Pd-BMIM and Pd-BIMe
compounds, they are almost equal at −0.35. Accordingly, the
electrostatic interaction between Pd-NHC and [NHC-C2Ph]+ in 8
is stronger than between the bare Pd atom and the same cation
in 2. We emphasize that these arguments are valid only for
processes occurring in the gas phase, since the solvation
energy should be accounted for in solution.
On the other hand, the enthalpy of dissociation of compounds 5
depends on the substituents of the NHC ligand. Systems with
bulky aromatic substituents 5c and 5d have a larger H than
structures 5a and 5b with small alkyl substituents. This is due to
the additional -coordination of aryl substituents with the
palladium atom, which stabilizes the complex (Figure 5). Note
that the palladium-acetylene binding energy in the Pd(0)
coordination complex is 38.6 kcal/mol.[51]
Infrared ion spectroscopy. To confirm the CID fragmentation
hypothesis and in particular the structure of the reactant and
product ions, IR spectra were recorded for species 5 and 8 after
m/z-isolation in the mass spectrometer. Infrared ion
spectroscopy was performed using a modified quadrupole ion
trap mass spectrometer and the FELIX free-electron laser, which
enabled us to cover the entire frequency range from 600 to 2100
cm–1.[49-50] These results are illustrated in Figure 3 which
contains comparisons of experimental versus computationally
predicted IR spectra for structures 5b and 8b. These spectral
comparisons allow us to unambiguously conclude that π-
complexes containing Pd-triple bonds are present in both
systems. This fact is especially evident from the position of the
carbon-carbon triple bond vibration that is redshifted as a result
of Pd binding to approximately 1900 cm–1 in both experimental
IR spectra, matching well only to the predicted IR spectrum of
the π-complex of the (BMIM-C2Ph)-Pd(L) structure. Both
potential σ-complexes (trans- or cis-(BMIM)2-Pd-C2Ph+) have a
much lower red shift of the triple bond vibration in their spectra,
and, thus, do not match the experimental data. Analogous
results were found for all other studied ions: complexes 5a,c,d
and 8a (see the Supporting Information for more details). The π-
bound Pd-alkyne complexes are clearly identified in all cases,
also primarily based on the redshifted positions of the C-C triple
bond vibrations.
Additionally, IR spectra are presented for the cationic reaction
products 2a-d (Figures S30-S33) and compared to
computationally predicted spectra to demonstrate that we indeed
observed the NHC-ethynyl coupling product when generated by
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
Figure 4. Calculated energy profile (ΔH in kcal/mol) for NHC-ethynyl coupling at the PBE1PBE/def2TZVP D3BJ level for (a) mono-NHC palladium complexes and
(b) bis-NHC palladium complexes (NHC = BIMe, BMIM, IMes, IPr). The plots include only thermodynamic energies, no activation barriers were presented.
argon atom. Due to the high cost of AIMD calculations and the
inability to obtain long trajectories, the velocity of the complex
was chosen so that approximately half of the collisions led to
fragmentation in two picoseconds. This approach allows one to
obtain a distribution of the fragmentation products close to the
experiment (whether a single collision can lead to an observed
fragmentation pattern). The energy balance of the process can
be expressed by equation given in Figure S4b. It reflects the
redistribution of energy within the considered molecular system.
Calculated collision paths lead to 6 different product cations
(Figure 6a, Table S5). In 18 trajectories fragmentation did not
occur, although the internal energy increased by 142.3 –
306.6 kcal/mol and the complex heated up to a temperature of
729.8 – 1198.5 K. In 17 MD simulations the Pd-BMIM bond was
Figure 5. Calculated structures of 5c and 5d. Interatomic distances shown in
broken with formation of compound 5b. Another 5 collisions
resulted in the cleavage of the -coordination of Pd and the
formation of 2b, while the Pd-BMIM σ-bond could both be
preserved and broken. The remaining 5 trajectories were
associated with the fragmentation of the butyl substituent of the
heterocycle. The collision-induced internal energy increase Ecol
for different calculated trajectories ranged from 142.3 to
351.6 kcal/mol.
angstrom; hydrogen atoms are omitted for clarity.
High calculated dissociation energies probably have the same
origin: electrostatic interaction of [NHC-C2Ph]+ and Pd-NHC in
compounds 8a – 51.6 kcal/mol and 8b – 56.3 kcal/mol;
additional -coordination in compounds 5c – 54.5 kcal/mol and
5d – 56.0 kcal/mol (Table 2). Based on the calculated data, the
dissociation enthalpy decreases in the order: 8b ~ 5d > 5c > 8a,
which is consistent with the experimental BDE data. The values
of appearance energies, as well as the calculated enthalpies of
dissociation, differ from each other in a consistent fashion (Table
2), although the calculated H values are systematically below
According to the data obtained, the minimum value
, which
have to be transferred to the system in order for fragmentation to
occur, is 223.6 kcal/mol. This value is much higher than the
calculated dissociation enthalpy of the system (56.3 kcal/mol),
since upon impact the translational energy is transferred not to
one degree of freedom, but is distributed over several degrees
of freedom at the collision zone.
compared the experimental data by 10
calculations support the hypothesis that the observed ions
represent [(NHC-C2Ph)Pd]+ π-complexes.
– 12%. Thus,
The more Ecol the system receives, the higher the probability of
fragmentation. This is evidenced by the average values of
collision-induced internal energy increase for various events in
MD simulations (Table S5). Collision and fragmentation events
are clearly visible in the plot of potential energy versus time
(Figure 6b). The figure shows graphs for simulations in which
the Pd-NHC bond is broken.
Molecular dynamics (MD) simulations. To make additional
insight into the fragmentation pathways we have performed ab
initio molecular dynamics (AIMD) simulation of CID process for
[(BMIM)(BMIM-C2Ph)Pd]+ 8b. We were looking for fragmentation
pathways that matches ones observed experimentally after a
single collision. AIMD provides only short-time limit trajectories
and only qualitative result was expected, and no attempt to
address observed ion intensities was made.[52-53] In our model,
complex 8b moving at a velocity of 8.8 km/s collided with a fixed
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
Figure 6. AIMD simulation of CID process for [(BMIM)(BMIM-C2Ph)Pd]+ 8b. (a) A generalized representation of the collision trajectories with speed v1 = 8.8 km/s
and distribution of the events in MD simulations. (b) Evaluation of potential energy with simulation steps for simulations resulting in Pd-NHC bond cleavage.
Trajectories with the minimum (violet curves) and maximum (marine curves) values of Ecol are shown.
6
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
The collision appears in the graphs in the form of peaks of
potential energy. The top of peak corresponds to the closest
distance between 8b and the argon atom (Figure 6, events A
and D). The argon atom then moves away from 8b which leads
to a decrease in potential energy. Gradual increases in potential
energy in the graphs indicate structural changes in the system,
such as isomerization or fragmentation (Figure 6, events B, C,
E).
Quantitative observation in the Sonogashira reaction. We
were able to quantify NHC-ethynyl product concentrations in the
reaction mixture using the isolated coupling product as an added
internal standard. The reaction was performed in Cu-free
conditions, to avoid the possibility of copper catalyzing these
transformations.[34] For further analysis, we used the
Sonogashira reaction with triethylamine as a base due to it
giving decent yields (up to 70%) at ambient conditions (r.t., air).
Analysis of the reaction mixture resulted in an abundant signal of
the NHC-ethynyl coupling product, but with only 1.4% percent
yield, calculated by NMR. With the low activation energy barrier
for the NHC-ethynyl coupling in mind, we questioned if the
amount of product observed is much lower than the amount
actually formed (i.e. due to decomposition of 2a or side-
reactions). Moreover, in mass spectra we observed multiple
compounds, which, we believe, are the products of further
reactions of 2a. To verify this hypothesis, we performed MS-
monitoring of the Sonogashira reaction.
Under studied conditions, Sonogashira reaction was catalyzed
by 5 mol % of deuterium labeled complex 1a-d3. The reaction
mixture also contained small amount of 2a⋅TfO–, which was
used to detect a concentration decrease of forming 2a-d3. Due
to matrix effects, the ionization efficiency may vary during the
course of the reaction. Therefore, we used the 2a-d6⋅TfO–
standard to get accurate 2a and 2a-d3 concentrations (to
account potential consumption of 2a-d3 internal standard during
the reaction course). We assumed its reactivity to be low in
highly-diluted samples, because spectra were measured as
quickly as possible after sampling (<1 min for the first point).
Scheme 3. Formation and transformations of NHC-ethynyl coupling product in the course of the reaction. The reaction was performed in the following conditions:
iodobenzene (0.17 mmol), phenylacetylene (1 eq), triethylamine (1 eq), 5 mol % of the catalyst in 3 mL of DMSO at 18 °C.
7
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
A change of 2a-d3 concentration represents the difference
between the rate of formation and the rate of further reactions of
2a-d3, while a change of 2a concentration is only related to the
rate of further reactions of 2a-d3 (both compounds were
assumed to have the same reactivity, and be represented in the
rate equation by their total concentration). The difference
between relative changes (see SI for complete equation) helps
to calculate cumulative amount of formed 2a-d3 in the course of
the reactions, and the yield of the NHC-ethynyl coupling product
(Figure S1).
Results indicate that a large fraction of the product indeed reacts
with different substances in the solution (formation of possible
Pd-containing intermediate is represented by A in Scheme 3).
The experimental data shows that the concentration of 2a-d3
decreases with time with maximum observed amount
correspond to the yield of 9%. But in assumption of equal
reactivity of 2a and 2a-d3 the concentration should reach a
plateau without any further change when the formation stops.
This observation might be due to formed from the catalyst 2a-d3,
having a much higher reactivity than the product 2a⋅TfO– added
to the reaction mixture. This can be the case if at least one
further reaction is occurring on a palladium atom (see Scheme
3), and k1 is comparable with one of the slowest steps in the
pathway and much smaller than k2. Thus, we can conclude that
the real yield of the NHC-ethynyl coupling product may be
underestimated and is greater than 9%.
palladium species should agglomerate and form nanoparticles.[40,
56-57]
Thereby, we have studied both reaction conditions: (i) with
Cs2CO3 as base at 80 ºC and (ii) with Et3N as base at r.t.
In both cases we have clearly observed palladium nanoparticles
(Figure 7 and Supporting Information). Individual nanoparticles
were clearly detected as well as some aggregates, which are
usually expected for a dynamic system in solution. When the
mixture was heated up in the presence of carbonate the mean
diameter of palladium nanoparticles was 3.2±0.8 nm, while
under mild conditions smaller 0.9±0.3 nm nanoparticles were
observed. Under studied conditions, NHC-ethynyl and NHC-Ph
couplings may contribute to Pd nanoparticles formation.
Although it is difficult to determine which process contributes
more, the observed formation of nanoparticles is in good
agreement with the conclusions made about the nature of the
studied catalytic system.
[R-NHC]X
R
molecular
catalysis
LnPd NHC
"Pd"
LnPd NHC
catalyst precursor
R-NHC coupling
aggregation
X
nanoparticle
catalysis
Pd NPs
catalytic cycle
trapping with
nanofishing
procedure
One may also use another approach for the interpretation of the
data. For the secondary reaction with Pd have to have a
coordination vacancy. The reaction product 2a-d3 is formed in
the reductive elimination process within the Pd coordination
sphere, while the added 2a⋅OTf– has to go through a ligand
exchange on the Pd-atom at least two times. This latter process
is much slower, so that the overall pathway for the added
product 2a is also much slower. This fact may also support the
hypothesis that the NHC-ethynyl coupling product is formed on a
palladium atom.
Figure 7. Formation of NHC-disconnected Pd due to R-NHC coupling,
trapping nanoparticles by nanofishing procedure and experimental TEM
images (scale bar – 20 nm in both cases; see Supporting Information for larger
images).
In order to ensure that this process generally occurs in the
Sonogashira reaction, we have examined several substrates
(under the same conditions but without internal isotope-labeled
standards). Both processes of NHC-Ph and NHC-ethynyl
coupling were observed for electron withdrawing and electron
donating phenyl acetylenes (4-fluoro- and 4-ethoxy-) and
iodobenzenes (4-nitro- and 4-methoxy). Substituents were
chosen in order to avoid possible cross contamination; both
[NHC-C2Ar]+ and [NHC-Ar]+ ions were clearly observed in all
cases. Noteworthy, NHC-ethynyl products were similarly
decreasing in their abundances due to further reactivity, while
NHC-Ph products were accumulating (see Supporting
Information for details).
Interesting to note, that NHC-ethynyl coupling still occurs in the
absence of PhI. This confirms a general nature of the NHC-
ethynyl coupling, which may take place as soon as both ligands
are located on the metal center.
Detection of nanoparticles in the Sonogashira reaction.
Formation of metal nanoparticles from molecular Pd/NHC
complexes is an experimental indication of R-NHC coupling
(Figure 7). As both NHC-Ph and NHC-ethynyl products can be
observed directly in the reaction mixture, we have conducted a
transmission electron microscopy (TEM) nanofishing[54] to
confirm the formation of bare palladium in the reaction
mixture.[55] Under regular reaction conditions, ligandless
Conclusions
To summarize, we reported here experimental evidence for a
novel transformation NHC-ethynyl coupling. This reaction occurs
in acetylide Pd/NHC complexes at a high rate and represents a
pathway in the dynamic catalytic system of the Sonogashira
coupling reactions. The product of NHC-ethynyl coupling was
observed under both Cu-assisted and Cu-free conditions with
NHC ligands of different nature. Moreover, it can be synthesized
by a direct reaction between Pd/NHC and acetylide under mild
conditions.
Detailed MS and IRMPD studies provided
a
valuable
understanding of the mechanism of this process. Reductive
elimination of NHC with an ethynyl substituent occurs easily due
to a low activation barrier, but the dissociation of the NHC-
ethynyl product with Pd center is an endothermic process.
Kinetic studies of the Sonogashira coupling revealed that NHC-
ethynyl coupling occurs from the very beginning of the reaction,
but mostly it is excluded from the reaction by further
transformations such as oligomerization, which was one of the
general limitations to reveal this transformation. We anticipate
more detailed mechanistic studies of the nature of NHC-ethynyl
8
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
[17] R. A. Batey, M. Shen and A. J. Lough, Org. Lett. 2002, 4, 1411-
1414.
[18] P. Dubey, S. Gupta and A. K. Singh, Dalton Trans. 2017, 46,
coupling in the near future to expedite the development of a new
class of recoverable Pd/NHC catalysts for alkyne C-H
functionalization.
13065-13076.
Exceptional importance of R-NHC coupling has been recently
noted.[23] In the presence of R-NHC coupling, cocktail-type NHC-
disconnected catalytic cycle is initiated, which operates on
another concept[57] as compared to regular NHC-connected
molecular catalysis. Different principles should be applied for
catalyst optimization, especially to address recyclability and
sustainability topics, in the cases of molecular NHC-connected
or cocktail-type NHC-disconnected catalysis. In the present
study, we demonstrate that NHC-disconnection can easily took
place during the studied Sonogashira reaction by involvement of
NHC-ethynyl coupling. Such a possibility of accessing NHC-
disconnected catalysis pathway recalls for critical re-thinking of
catalyst design principles.
[19] C. W. Gallop, M. T. Chen and O. Navarro, Org. Lett. 2014, 16,
3724-3727.
[20] Y. Ma, C. Song, W. Jiang, Q. Wu, Y. Wang, X. Liu and M. B.
Andrus, Org. Lett. 2003, 5, 3317-3319.
[21] V. N. Mikhaylov, V. N. Sorokoumov, K. A. Korvinson, A. S.
Novikov and I. A. Balova, Organometallics 2016, 35, 1684-1697.
[22] C. L. Yang and S. P. Nolan, Organometallics 2002, 21, 1020-
1022.
[23] V. M. Chernyshev, E. A. Denisova, D. B. Eremin and V. P.
Ananikov, Chem. Sci. 2020, 11, 6957-6977.
[24] Z. Ahmadi, L. P. Yunker, A. G. Oliver and J. S. McIndoe, Dalton
Trans. 2015, 44, 20367-20375.
[25] R. Chinchilla and C. Najera, Chem. Rev. 2007, 107, 874-922.
[26] R. J. Oeschger, D. H. Ringger and P. Chen, Organometallics
2015, 34, 3888-3892.
[27] E. A. Valishina, M. F. C. G. da Silva, M. A. Kinzhalov, S. A.
Timofeeva, T. M. Buslaeva, M. Haukka, A. J. L. Pombeiro, V. P.
Boyarskiy, V. Y. Kukushkin and K. V. Luzyanin, J. Mol. Catal. A:
Chem. 2014, 395, 162-171.
[28] E. G. Gordeev, D. B. Eremin, V. M. Chernyshev and V. P.
Ananikov, Organometallics 2017, 37, 787-796.
[29] L. Djakovitch and P. Rollet, Adv. Synth. Catal. 2004, 346, 1782-
1792.
[30] E. V. Larina, A. A. Kurokhtina, E. V. Yarosh, N. A. Lagoda and A.
F. Schmidt, Russ. J. Org. Chem. 2016, 52, 1356-1358.
[31] T. Ljungdahl, T. Bennur, A. Dallas, H. Emtenas and J.
Martensson, Organometallics 2008, 27, 2490-2498.
[32] A. V. Astakhov, O. V. Khazipov, A. Y. Chernenko, D. V.
Pasyukov, A. S. Kashin, E. G. Gordeev, V. N. Khrustalev, V. M.
Chernyshev and V. P. Ananikov, Organometallics 2017, 36, 1981-
1992.
[33] B. M. Barry, R. G. Soper, J. Hurmalainen, A. Mansikkamaki, K.
N. Robertson, W. L. McClennan, A. J. Veinot, T. L. Roemmele, U.
Werner-Zwanziger, R. T. Boere, H. M. Tuononen, J. A. C. Clyburne
and J. D. Masuda, Angew. Chem. Int. Ed. 2018, 57, 749-754.
[34] S. Li, J. Tang, Y. Zhao, R. Jiang, T. Wang, G. Gao and J. You,
Chem. Commun. 2017, 53, 3489-3492.
[35] W. M. David, D. Kumar and S. M. Kerwin, Bioorg. Med. Chem.
Lett. 2000, 10, 2509-2512.
[36] K. L. Vikse, Z. Ahmadi and J. Scott McIndoe, Coord. Chem. Rev.
2014, 279, 96-114.
[37] R. J. Oeschger and P. Chen, J. Am. Chem. Soc. 2017, 139,
1069-1072.
[38] E. P. Couzijn, E. Zocher, A. Bach and P. Chen, Chem. Eur. J.
2010, 16, 5408-5415.
Acknowledgements
This work was supported by the Russian Science Foundation
(RSF) grant 17-13-01526. IRMPD study at FELIX laboratory has
been supported by the project CALIPSOplus under the Grant
Agreement 730872 from the EU Frame-work Programme for
Research and Innovation HORIZON 2020. We gratefully
acknowledge
the
Nederlandse
Organisatie
voor
Wetenschappelijk Onderzoek (NWO) for the support of the
FELIX Laboratory. The authors thank Alexey S. Kashin and Vera
A. Cherpanova for TEM study, and Andrey Tsedilin for helpful
discussion.
Keywords: Sonogashira reaction • CID • IRMPD • ESI-MS •
NHC-ethynyl coupling
[1] B. M. Trost and J. S. Tracy, Isr. J. Chem. 2018, 58, 18-27.
[2] B. A. Trofimov and E. Y. Schmidt, Acc. Chem. Res. 2018, 51,
1117-1130.
[3] K. I. Galkin and V. P. Ananikov, Russ. Chem. Rev. 2016, 85, 226-
247.
[4] A. Biffis, P. Centomo, A. Del Zotto and M. Zecca, Chem. Rev.
2018, 118, 2249-2295.
[5] X. Fang, B. Cacherat and B. Morandi, Nat. Chem. 2017, 9, 1105-
1109.
[6] T. O. Ronson, R. J. K. Taylor and I. J. S. Fairlamb, Tetrahedron
2015, 71, 989-1009.
[7] J. Twilton, C. Le, P. Zhang, M. H. Shaw, R. W. Evans and D. W.
C. MacMillan, Nat. Rev. Chem. 2017, 1.
[8] C. X. Zhuo and A. Furstner, J. Am. Chem. Soc. 2018, 140,
10514-10523.
[9] G. C. Fortman and S. P. Nolan, Chem. Soc. Rev. 2011, 40, 5151-
5169.
[10] X. Wang, Y. Song, J. Qu and Y. Luo, Organometallics 2017, 36,
1042-1048.
[11] M. Gazvoda, M. Virant, B. Pinter and J. Kosmrlj, Nat. Commun.
2018, 9, 4814.
[12] O. Rivada-Wheelaghan, A. Comas-Vives, R. R. Fayzullin, A.
Lledos and J. Khusnutdinova, Chem. Eur. J. 2020.
[13] E. A. Kantchev, C. J. O'Brien and M. G. Organ, Angew. Chem.
Int. Ed. 2007, 46, 2768-2813.
[14] S. Diez-Gonzalez, N. Marion and S. P. Nolan, Chem. Rev. 2009,
109, 3612-3676.
[39] L. Jasikova and J. Roithova, Chem. Eur. J. 2018, 24, 3374-3390.
[40] D. B. Eremin, E. A. Denisova, A. Yu Kostyukovich, J. Martens,
G. Berden, J. Oomens, V. N. Khrustalev, V. M. Chernyshev and V. P.
Ananikov, Chem. Eur. J. 2019, 25, 16564-16572.
[41] M. Anania, L. Jašíková, J. Zelenka, E. Shcherbachenko, J. Jašík
and J. Roithová, Chem. Sci. 2020, 11, 980-988.
[42] J. Hyvl and J. Roithova, Org. Lett. 2014, 16, 200-203.
[43] M. Jakl, M. Straka, J. Jaklova Dytrtova and J. Roithova, Int. J.
Mass Spectrom. 2014, 360, 8-14.
[44] D. Schroder, M. Engeser, M. Bronstrup, C. Daniel, J. Spandl
and H. Hartl, Int. J. Mass Spectrom. 2003, 228, 743-757.
[45] J. Schulz, L. Jasikova, A. Skriba and J. Roithova, J. Am. Chem.
Soc. 2014, 136, 11513-11523.
[46] J. Schulz, E. Shcherbachenko and J. Roithova, Organometallics
2015, 34, 3979-3987.
[47] E. L. Zins, C. Pepe and D. Schroder, J. Mass Spectrom. 2010,
45, 1253-1260.
[48] A. Gray, A. Tsybizova and J. Roithova, Chem. Sci. 2015, 6,
5544-5553.
[49] N. C. Polfer and J. Oomens, Phys. Chem. Chem. Phys. 2007, 9,
3804-3817.
[15] M. N. Hopkinson, C. Richter, M. Schedler and F. Glorius, Nature
2014, 510, 485-496.
[16] G. Altenhoff, S. Wurtz and F. Glorius, Tetrahedron Lett. 2006,
47, 2925-2928.
[50] J. Martens, G. Berden, C. R. Gebhardt and J. Oomens, Rev. Sci.
Instrum. 2016, 87, 103108.
[51] X. Wang and L. Andrews, J. Phys. Chem. A 2003, 107, 337-345.
[52] A. Martin-Somer, J. Martens, J. Grzetic, W. L. Hase, J. Oomens
and R. Spezia, J. Phys. Chem. A 2018, 122, 2612-2625.
9
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
[53] R. Spezia, J. Martens, J. Oomens and K. Song, Int. J. Mass
Spectrom. 2015, 388, 40-52.
[54] A. Galushko, E. Gordeev, A. S. Kashin, Y. V. Zubavichus and V.
Ananikov, Faraday Discuss. 2020.
[55] A. Y. Kostyukovich, A. M. Tsedilin, E. D. Sushchenko, D. B.
Eremin, A. S. Kashin, M. A. Topchiy, A. F. Asachenko, M. S.
Nechaev and V. P. Ananikov, Inorg. Chem. Front. 2019, 6, 482-492.
[56] E. A. Denisova, D. B. Eremin, E. G. Gordeev, A. M. Tsedilin and
V. P. Ananikov, Inorg. Chem. 2019, 58, 12218-12227.
[57] D. B. Eremin and V. P. Ananikov, Coord. Chem. Rev. 2017, 346,
2-19.
10
This article is protected by copyright. All rights reserved.

10.1002/chem.202003533
Chemistry - A European Journal
Entry for the Table of Contents
Mass spectrometry insight into the M/NHC-catalyzed Sonogashira cross-coupling reaction resulted in the discovery of NHC-ethynyl
coupling process. Intermediates of this transformations and 2-ethynylated azolium cations were studied with NMR, ESI-MS, CID,
IRMPD, DFT and MD techniques.
Institute and/or researcher Twitter usernames: @AnaikovLab @DMEremin @RoithovaGroup @HFML_FELIX
11
This article is protected by copyright. All rights reserved.