The Direct Non-Perturbing Leaching Test in the Phosphine-Free Suzuki-Miyaura Reaction Catalyzed by Palladium Nanoparticles

Article abstract of DOI:10.1002/cctc.201500329

The aryl halides normally unreactive in the reaction with phenylboronic acid in the presence of a palladium nanoparticle catalyst were shown to become involved, if a more reactive aryl iodide is simultaneously participating in the process. The involvement phenomenon can be interpreted as a direct evidence of the formation of smaller and more reactive palladium nanoparticles through leaching from the initial nanocatalyst. The involvement test was applied both to free unsupported Pd nanoparticles and supported species, thus exhibiting essentially the same character of evolutions of various nanomaterials during the catalytic transformation associated with leaching of reactive species from the initial particles. The system allows for concurrent processing of less reactive aryl bromides and even chlorides together with much more reactive aryl iodides. The interpretation of the observed phenomena and implications for practical realization of nanoparticle catalysts are discussed.

Full text of DOI:10.1002/cctc.201500329

DOI: 10.1002/cctc.201500329
Full Papers
The Direct Non-Perturbing Leaching Test in the
Phosphine-Free Suzuki–Miyaura Reaction Catalyzed by
Palladium Nanoparticles
Alexander N. Kashin, Olga G. Ganina, Andrey V. Cheprakov, and Irina P. Beletskaya*[a]
The aryl halides normally unreactive in the reaction with phe-
nylboronic acid in the presence of a palladium nanoparticle
catalyst were shown to become involved, if a more reactive
aryl iodide is simultaneously participating in the process. The
involvement phenomenon can be interpreted as a direct evi-
dence of the formation of smaller and more reactive palladium
nanoparticles through leaching from the initial nanocatalyst.
The involvement test was applied both to free unsupported
Pd nanoparticles and supported species, thus exhibiting essen-
tially the same character of evolutions of various nanomaterials
during the catalytic transformation associated with leaching of
reactive species from the initial particles. The system allows for
concurrent processing of less reactive aryl bromides and even
chlorides together with much more reactive aryl iodides. The
interpretation of the observed phenomena and implications
for practical realization of nanoparticle catalysts are discussed.
Introduction
The palladium-catalyzed reactions have caused revolutionary
changes in synthetic organic chemistry, providing a simple and
efficient method of creation of carbon–carbon and carbon–
heteroatom bonds, which is evidenced by a Nobel Prize award
in 2010.[1] An explosive expansion of the new methodology de-
pended on a rapid and mostly chaotic development and intro-
duction of new catalytic systems, based on various ligands and
palladium pre-catalysts, ranging from simple salts to palladacy-
cles, pincer complexes, and complex multicomponent mixtures
(“catalytic cocktails”[2]). Some of these systems exhibited
record-making catalytic efficiency, being able to serve at very
low concentrations, which is highly important for large-scale
implementation of this chemistry given a high cost of Pd
metal and other components. However, such a huge efficiency
measured by TON values of 104–107 and higher is observed
almost only for the reactions of the most reactive substrates,
aryl iodides and some activated aryl bromides. In the case of
more practically important, but less reactive substrates, the
TON values observed rarely exceed modest levels. The only
way to improve the efficiency is to take advantage of various
means to extend the effective life-time of catalysts through
heterogenization and design of recyclable catalytic systems.[3]
Common methods of heterogenization through the attach-
ment of palladium complexes, already proven as good homo-
geneous catalysts, to various organic or inorganic supports are
however rather ineffective. The operation of such immobilized
complexes, as well as any other so called phosphine-free cata-
lytic systems is known to be accounted for by leaching of cata-
lytically active species to the solution. Moreover, all such sys-
tems, be it homogeneous or supported complex, are now be-
lieved to give rise to palladium nanoparticles, which actually
serve as real source of Pd0 in the phosphine-free systems.
The recognition of the actual role of nanoparticles in the
phosphine-free palladium-catalyzed cross-coupling reactions
gave a new dimension to the development of palladium cata-
lysts.[4] The nanocatalysts thus serve as a bridge between the
homogeneous and heterogeneous catalysis. A number of free
and supported palladium nanomaterials have been described
and promoted as effective, and in many cases recyclable cata-
lysts for Suzuki–Miyaura and related reactions. Since our early
studies on aqueous phosphine-free palladium catalysis we
have paid attention to the involvement of palladium sols, and
further developed a series of effective and recyclable systems
based on supported nanoparticles, some of which endured as
many as 11 reuses without apparent deterioration.[5] The prep-
arative potential of palladium nanocatalysts is impressive.
Though the activity of Pd species cannot be controlled and en-
hanced by appropriate ancillary ligands such as phosphines,
there are other means of tuning up the performance and reac-
tivity through controlling the size distribution, the nature of
supports, and other factors governing the availability and ac-
tivity of palladium atoms at nanoparticle surface. For a long
time such systems, as overall all other phosphine-free systems,
were used only for the most reactive aryl iodides and activated
aryl bromides. However, after a few seminal reports on good
results with less reactive electrophiles,[6] showing the high cata-
lytic efficiency comparable to the best phosphine-assisted
methods,[7] such results started to appear at an increasing
pace. Various preparations of nanoparticles and sols were re-
ported to be competent for aryl bromides and even aryl chlo-
[a] Dr. A. N. Kashin, Dr. O. G. Ganina, Dr. A. V. Cheprakov,
Prof. Dr. I. P. Beletskaya
Chemistry Department, Lomonosov Moscow State University
1 Leninskie gory, Moscow 119991 (Russia)
E-mail: beletska@org.chem.msu.ru
This publication is part of a Special Issue on Palladium Catalysis. To view
the complete issue, visit:
http://onlinelibrary.wiley.com/doi/10.1002/cctc.v7.14/issuetoc
ChemCatChem 2015, 7, 2113 – 2121
2113
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
rides,[5b,8] demonstrating that the enhanced reactivity is not
a fortuitous deviation, but rather the systematic behavior
rooted in the structure of nanoparticles.
cles, while an immobilized boronic acid would be equally reac-
tive disregarding whether there is or there is no leaching. The
hot filtration test would give ambiguous results as no common
filtering media can reliably separate primary and secondary
nanoparticles.
In spite of such an impressive progress, the establishment of
nanoparticle catalysts in common preparative practice is im-
peded by incomplete understanding of the mechanism of
functioning of such systems, and the idea that nanocatalysts
also actually operate via leaching is a mainstream concept.[9]
Therefore, any researcher active in the area of developing
nanocatalysts should be prepared to meet and deal with the
paradoxical situation as the leaching is the driving force of cat-
alysis, and at the same time the main limiting factor of it, as it
destroys the initial fine structure of the nanocatalysts. Thus,
this phenomenon undermines the efforts to apply Pd catalysts
in the modern continuous-flow apparata, as in this case, the
leaching is irreversible and results both in rapid deterioration
of packed-bed catalysts and contamination of products.[10] The
detachment of Pd from the initial nanoparticles and the associ-
ated redistribution of the metal between the old pre-catalyst
and new Pd particles formed during the catalytic reaction in
the absence of good stabilizing ancillary ligands, result in the
Oswald ripening phenomenon which leads to gradual disap-
pearance of active sites and formation of large particles with
strongly diminished and uniform surface. Designing sophisti-
cated supports can partly help when such a support is de-
clared to be able not only to release, but also to catch back
the metal is a self-sustained mode,[11] but even in the most
spectacular recent systems,[12] partial loss of catalyst and deteri-
oration, though slower, is admitted to be still inevitable. The
leaching accounts for the so-called “homeopathic” catalysis, as
only a small fraction of Pd loaded as a pre-catalyst is actually
engaged in the catalysis.[13]
Thus, an unresolved ambiguity in the understanding of the
operation of catalytic systems employing various kinds of pal-
ladium nanoparticles as catalysts, in our opinion, hampers the
progress in this area. What should be done to improve recycla-
bility and activity of such systems? The answer to this critical
question should be dependent on whether the nanocatalysts
operate directly, or through leaching and formation of secon-
dary nanoparticles. If the latter is true, then the development
of increasingly complex supports is likely not to be the best
strategy of development, and the focus is best to be shifted to
intimate understanding of the processes involved in the mor-
phological evolution of nanomaterials regarded not as station-
ary solids, but rather as reactive particles engaged in the ex-
change of matter with reaction media and undergoing contin-
uous transformation. In the current paper we describe unam-
biguous evidence that such processes are indeed taking place
leading to dramatic change of nanocatalyst activity during the
reaction.
Results and Discussion
While studying the Suzuki–Miyaura reaction in the presence of
various palladium nanocatalysts, we have discovered an intri-
guing phenomenon—under the conditions of competitive re-
actions, the reactivity of less reactive aryl halides may be dra-
matically increased in the presence of more reactive aryl hal-
ides. We hypothesized that this phenomenon may bring light
on the phenomenon of leaching, and undertook a more de-
tailed study.
Notably, the detection of leaching from the nanoparticles re-
quires a more careful understanding of what is exactly going
on in such systems, and not as simple as in more conventional
supported catalysts involving immobilized complexes.
To be able to observe this phenomenon unambiguously and
not be misled by reaction caused by fortuitous presence of
more active fraction in the initial nanocatalyst, an initial nano-
particle pre-catalyst of modest reactivity, totally unreactive to-
wards the less reactive substrate, is required. The nanoparticle
pre-catalysts reported in the literature often are sophisticated
materials employing polymeric or solid supports or various
agents used to protect the particles from aggregation. The
effect of such agents on the catalytic activity and behavior of
palladium species may be unpredictably complex, and the task
to decouple these effects from the effects associated with the
innate reactivity of nanostructured palladium metal can hardly
be solved. To avoid the task to separate the factors associated
with stabilizing additives and the innate reactivity of nano-
structured Pd metal, a few years ago we succeeded in develop-
ing a technique allowing for the generation of stable palladi-
um sol without supports and low-molecular stabilizing agents.
Vacuum deposition of metallic palladium onto a polyvinylalco-
hol support followed by hot water extraction of the polymer
gave a nearly monodisperse palladium sol with average parti-
cle size of 10 nm (Figure 1). It is of prime importance for the
present study that the thus freshly prepared sol contained no
fractions of substantially smaller particles.[5c] The vacuum depo-
The main problem is associated with the expected fate of
leached Pd0 species. In the absence of good ancillary ligands
stabilizing the Pd0 state, one could not expect the accumula-
tion of substantial amounts of Pd0 species capable of driving
the catalytic cycle. In the recent paper we revealed the evi-
dence that the only reasonable resting state involving Pd0 in
the absence of stabilizing ancillary ligands is the nanosized
metal particles.[14] In the case when the initial form of Pd cata-
lyst (the pre-catalyst) also consists of nanoparticles, the most
probable result of leaching would be the formation of new
nanoparticles, which probably can be distinguished from the
initial material by size, morphology, or reactivity. At the same
time, the so called leaching tests[15] commonly used to recog-
nize the leaching and formation of nanoparticles from non-
nanoparticle pre-catalysts, are useless or prone to give errone-
ous results if applied for systems with initial loading of nano-
particle catalysts. Indeed, the mercury test would inactivate
both the primary and secondary nanoparticles and give no
clue to understanding the real cause of suppression. The 3-
phase test is altogether useless, as an immobilized aryl halide
would not react with either primary or secondary nanoparti-
ChemCatChem 2015, 7, 2113 – 2121
www.chemcatchem.org
2114
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Indeed, if the reaction with reactive substrate takes place via
clipping palladium atoms from the surface, after a full turn of
catalytic cycle such atoms would not be reattached to the sur-
face but instead would form either monoatomic Pd0 species or
form new and smaller clusters, purportedly having a different
reactivity.
In the first experiments we added the less reactive iodoar-
ene, p-iodotoluene, to the reaction mixture with the reactive
substrate p-iodoanisole. While p-iodotoluene, taken alone, was
completely unreactive, in the equimolar mixture both sub-
strates gave the respective biphenyls in roughly equal yields
(Table 1). Before going into further discussion, it should be
Table 1. The Suzuki–Miyaura reaction of aryl halides with phenylboronic
acid with and without involvement.
Figure 1. TEM image of the PdNPs obtained by vacuum deposition of Pd
with subsequent withdrawal of the support. Pd nanoparticles size ꢀ10 nm.
sition technique can be also expected to afford particles with
roughly uniform surface, relatively free from crystal defects,
which are inevitably formed when the techniques based on
chemical reduction of soluble Pd salts are used.
Entry
Hal-C6H4-X
I-C6H4-Y
T
[h]
Yield [%]
X-C6H4-Ph[b] Y-C6H4-Ph[b]
1
2
3
4[c]
5[c]
6
7
8
9
I-C6H4-Me
I-C6H4-Me
–
3
5
5
5
5
5
3
3
5
4.5
60
<1
17
0
–
50
–
91
–
I-C6H4-OMe
–
Br-C6H4-OMe
Br-C6H4-OMe
Br-C6H4-Ac
Br-C6H4-Ac
Br-C6H4-Ac
Cl-C6H4-Ac
Cl-C6H4-Ac
When we tested the behavior of these nanoparticles in
a Suzuki–Miyaura reaction of aryl iodides with phenylboronic
acid, the material exhibited a modest performance being capa-
ble of catalyzing only the reactions of more reactive aryl io-
dides, but failing with activated aryl bromides or strongly deac-
tivated electron-rich aryl iodides, surprisingly excluding p-io-
doanisole, which exhibited a fair reactivity in a sharp contrast
to p-iodotoluene which was almost completely unreactive.
Such a behavior could be expected as the obtained nanoparti-
cle are rather big, consequently there are relatively few atoms
at the surface capable of taking part in the oxidative addition
allowing to observe the reaction only with the most reactive
substrates. Moreover, such a dramatic discrimination between
two iodoarenes is not likely to be accounted for by the rate of
oxidative addition alone—there are no such precedents in the
cross-coupling chemistry of any kind. Besides, p-MeO-substitu-
ent is commonly believed to behave as a more strongly deacti-
vating substituent than p-methyl in the oxidative addition pro-
cess, the transition state of which is believed to resemble s-
complex of the nucleophilic aromatic substitution, though
with a much smaller negative charge delocalized in the ring.[16]
The discrimination is more likely to be caused by differential
interaction of substrates with the surface of palladium metal
particle, which should take place in order to enable the reac-
tion with surface-bound atom of Pd, as such centers in a big
particle are too few to ensure reasonable reaction rate with
a freely diffusing substrate molecule. In fact, a precedent of un-
usual reactivity of p-methoxy-substituted substrate in a reaction
with phenylboronic acid in the presence of palladium nanopar-
ticles, inconsistent with the routine electronic effect considera-
tions were reported by Gomez and co-workers.[17] A sort of se-
lective chemisorption facilitated by methoxy-group[18] but dis-
abled by simple methyl group is likely to be involved.
I-C6H4-Ac
–
I-C6H4-OMe
I-C6H4-Me
–
83
12
<1
12
47
4
–
I-C6H4-OMe
97
[a] Reaction conditions: 0.075 mmol of Hal-C6H4-X, 0.075 mmol of I-C6H4-Y,
0.113 mmol PhB(OH)2, 0.15 mol% Pd, 0.225 mmolK3PO4, 0.3 mL DMA,
1208C. [b] Yield determined by GC. [c] Reaction temperature 1408C.
noted that we have not performed true competitive runs, as
the reactions in phosphine-free systems are characterized by
complex, substrate-sensitive kinetics often involving the induc-
tion periods of varied length, so performing these reactions
competitively with two substrates and fixed reaction time
would give ambiguous results strongly dependent on the
time-slice chosen. Instead, we performed preparative runs
giving the substrates time for reactions to reach a certain con-
version, always limiting the reaction time to the same duration
and keeping identical all other conditions, and then measuring
the distribution of the products formed. Such data on the
yields and TON values are cumulative by nature, and allow for
no conclusions about the true relative reactivity of substrates,
as could have been measured by relative rate constants. There-
fore, for example, a roughly equal ratio of products from both
substrates should not be confused for the evidence that the
initially unreactive substrate somehow increased its reactivity
to match the one of the initially more reactive substrate. What
is actually shown by the results discussed below is that the
nature of catalytic species changes during the reaction, and
that a more reactive form of Pd0 as compared to the initial big
ChemCatChem 2015, 7, 2113 – 2121
www.chemcatchem.org
2115
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
and lazy nanoparticles, does form in the reaction mixture, and
is capable to involve the less reactive substrates.
in the reaction of reactive substrates was additionally tested
using the derivatives of the same aryl group - a mixture of io-
dobenzene and perdeuteriobromobenzene (Table 2, entries 1–
4). The involvement indeed took place, with twice as much
We hypothesized that “leaching”, as a result of reaction with
p-iodoanisole, leads to formation of more reactive catalytic
species, which, in accordance with the concept of the nanopar-
ticle-enabled pathway[14] should be smaller and more reactive
clusters of palladium, capable of participating in oxidative ad-
dition with another haloarene, not active if taken alone under
the same reaction conditions. Thus, the reaction of a more re-
active substrate (a “trigger” substrate) would be expected to
“involve” the unreactive substrate and enable its reaction.
The suggestion was first tested in the concurrent reaction of
PhB(OH)2 with two iodoarenes, p-iodoanisole and p-iodo-
toluene (in ratio 1.5:1:1 respectively). We indeed observed the
conversion of both substrates with somewhat higher yield of
p-methylbiphenyl than p-methoxybiphenyl (Table 1, entries 1
vs 2). This observation was tested in several reaction runs,
always with the same result. If taken alone in the presence of
nanoparticle pre-catalyst the substrate was unreactive, but
could be awakened by the addition of the more reactive sub-
strate, and, moreover, further the reaction with both substrates
present in the reaction mixture from the beginning proceeded
almost quantitatively exhibiting roughly the same apparent cu-
mulative TON.
Table 2. The Suzuki–Miyaura concurrent reaction of iodo- and bromo-
benzene with phenylboronic acid.
Entry
Haloarene
PhB(OH)2
[equiv]
Yield [%]
[a]
Ph2
Ph-Ar
1
2
3
4
PhBr
PhI
PhI+C6D5Br[b]
PhI+PhBr[c,d]
1.5
1.5
1.5
3
0
99
80
–
–
40
100
100
[a] Reaction conditions: 0.075 mmol of Hal-C6H5, 0.113 mmol PhB(OH)2,
0.15 mol% Pd, 0.225 mmolK3PO4, 0.3 mL DMA, 1208C. [b] 0.075 mmol of
[D5]PhBr was added. [c] 0.075 mmol of PhBr was added [d] additional
0.113 mmol PhB(OH)2 was used.
product formed from PhI than that formed from PhDBr ob-
served at incomplete conversion. With increased reaction time
both substrates could be made to undergo a complete conver-
sion, making obvious that although the iodoarene reacts
faster, it leaves behind a sufficiently reactive and stable catalyt-
ic Pd species to fully process the bromoarene, and that no de-
activation takes place upon the depletion of iodoarene. This
finding gives a hint for further optimization of the involvement
approach by increasing the ratio of bromo- to iodoarene. The
involvement works not only with ArI/ArBr pairs, but also with
activated aryl chlorides (Table 1, entries 8–9), though the con-
version of chloroarene is substantially lower and is associated
with practically quantitative conversion of aryl iodide. Thus, it
is evident that active species born in the reaction of the initial
large nanoparticles with the more reactive aryl iodide are capa-
ble of sustaining the reaction of aryl chloride at a reasonable
rate within the same period of time as used in the other reac-
tions under discussion performed with substrates of much
higher reactivity.
Encouraged by these results, we studied several other pairs
of substrates of much stronger difference in reactivity, and
tried if the reaction of bromoarenes can be triggered by io-
doarenes. Thus, p-bromoanisole was involved in the reaction
using the mixture with p-iodoacetophenone (Table 1, entries 3
vs 4). Given a supposedly huge difference of reactivity of an
activated iodoarene vs deactivated bromoarene a roughly 5:1
ratio of products obtained shows that the conversion of bro-
moarene took place mostly after major part of iodoarene took
part in the reaction. However, when a mixture of p-iodoanisole
and p-bromoacetophenone was taken, the product from the
latter was surprisingly formed in a higher yield than the prod-
uct from the former. It should be noted that when speaking
about “higher” or “lower” reactivity, no reliable data could be
given in support. The reactivity of various haloarenes in cross-
coupling reactions involving the oxidative addition, generally
assumed as an only rate-limiting step, seems to have never
been rigorously assayed, so all discussions have to use prepa-
rative yields obtained within a roughly equal times and condi-
tions, as a rough measure of reactivity.[19] In the “involvement”
test discussed in this paper, active species are generated
during the initial phase, when only the more reactive substrate
is processed, followed by the phase in which the relative reac-
tivity continuously changes because of the variation of relative
concentrations of the substrates and accumulation of iodide
and bromide ions in the reaction mixture. In the case when
the more reactive nanoparticles are taken as pre-catalyst (e.g.
sol formed by reduction of h3-allylpalladium chloride dimer
containing nanoparticles of ca. 2 nm diameter, furnishing
active catalyst for bromoarenes[20]), the ratio of products
formed from p-iodoanisole and p-bromoacetophenone is
higher in favor of the more reactive substrate. The involvement
of unreactive substrates due to the formation of active species
To show that the “involvement” phenomenon is not limited
to a certain kind of nanoparticles, which could be denoted as
neat nanoparticles, that is the ones lacking solid support and
protective agents, we performed the tests with a much more
sophisticated nanomaterial, palladium supported by TiO2
coated multi-walled carbon nanotubes (Pd/MWCNT/TiO2)[21]
(Figure 2) in the same set of reactions (Table 3). This nanomate-
rial also showed a modest activity and was not able to serve as
active catalyst in the reactions of the less active substrates.
The “involvement” reactions meanwhile gave similar results as
those performed with unsupported nanoparticles. A somewhat
lower activity of species generated using the supported nano-
material is evident, as no reaction of aryl chloride was ob-
served in the “involvement” test using p-iodoanisole as a trig-
ger substrate.
ChemCatChem 2015, 7, 2113 – 2121
www.chemcatchem.org
2116
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Nevertheless, in other palladium catalyzed reactions, the
higher reactivity of bromoarenes with respect to iodoarenes is
a rather common phenomenon. Most often it is encountered
in C-heteroatom cross coupling, such as amination, amidation,
or etheration. In these reactions the net reactivity of iodoar-
enes, measured by cumulative TON values is often lower than
the reactivity of the respective aryl bromides, which is account-
ed for by suppression of ligand exchange by iodide ligand, as
compared to more labile bromide.[22] It should be clearly stated
that in all such cases the reactivity is understood in terms of
the observed conversion and cumulative yields measured after
the completion of a given reaction. At the same time even in
these cases CÀI bonds are more reactive than CÀBr bond, and
this is clearly seen in the reactions of substrates containing
both iodine and bromine, the former is always and without ex-
ceptions substituted first, and the latter enters the reaction
after and only after iodine is fully substituted. Lower reactivity
of aryl iodides is thus manifested only at substantial depths of
conversion, when the catalyst is literally suffocated by the li-
berated iodide, which is formed obviously in huge excess over
the catalyst, and forms less labile Pd iodide complexes.
Figure 2. SEM and TEM image of the sample Pd/MWCNT/TiO2. The TiO2
coated MWCNTs, prepared via slow hydrolysis of titanium isopropoxide
(Ti[OCH(CH3)2]4, TTIP) on the surface of MWCNTs ; palladium was deposited
on TiO2-MWCNTs by precipitation followed by reduction resulting Pd/
MWCNT/TiO2. Pd0 nanoparticles size obtained 10–20 nm.
Table 3. The Suzuki–Miyaura reaction of aryl halides with phenylboronic
acid catalyzed by Pd/MWCNT/TiO2 with or without involvement.
The source of this controversy is in the hierarchy of stages in
the catalytic cycle. The oxidative addition of CÀhalogen bonds
is always the first step, and the reactivity of CÀI bonds is
always higher than the reactivity of CÀBr bonds. However, in
many reactions the transmetalation or ligand exchange are the
limiting steps defining the overall efficiency of the catalytic
cycle, and aryl bromides exhibit higher TON values, deeper
conversions and higher overall yields. Recently, this phenomen-
on was revealed and confirmed in numerous studies as
a common feature for the directed CH-arylation reactions,
which can be regarded as a special case of CÀC cross coupling
in which the key intermediate is formed not via transmetala-
tion (as in the classical cross-coupling reactions such as
Suzuki–Miyaura reaction) nor through the ligand exchange as
in C-heteroatom cross coupling, but via a sort of electrophilic
metalation, in which it is vital to maintain a good level of elec-
trophilicity of the primary intermediates formed via the oxida-
tive addition, and the effect of retardation by iodide ions in
particularly pronounced. Thus, the higher reactivity of ArBr in
CH-arylation reactions was revealed by Fagnou and co-work-
ers[23] and reliably confirmed in a number of similar meth-
ods.[24]
Entry
Hal-C6H4-X
I-C6H4-Y
Yield [%]
X-C6H4-Ph[b]
Y-C6H4-Ph[b]
1
2
3
4[c]
5
6
7
8
Br-C6H4-Ac
–
Br-C6H4-Ac
Br-C6H4-Ac
–
Br-C6H4-OMe
Br-C6H4-OMe
Cl-C6H4-Ac
–
5
–
98
90
–
3
12
0
–
98
49
40
98
–
I-C6H4-OMe
I-C6H4-OMe
I-C6H4-OMe
I-C6H4-Ac
–
I-C6H4-Ac
I-C6H4-OMe
99
98
[a] Reaction conditions: 0.1 mmol of Hal-C6H4-X, 0.1 mmol of I-C6H4-Y,
0.15 mmol PhB(OH)2, 0.5 mol% Pd, 0.25 mmolK2CO3, 0.5 mL of mixture
EtOH/H2O 2:3, 608C. [b] Yield determined by GC. [c] 0.05 mmol of
p-iodoanisole was used.
An abnormal relative reactivity of “involved” bromoarenes
The anomalous cumulative reactivity of the involved bromoar-
enes is an unexpected side effect of the involvement phenom-
enon deserving a discussion of its possible sources.
Thus, the abnormal reactivity of aryl bromides in our case is
a well-established trend in the catalytic processes limited by
reactivity of the oxidative adducts in the steps required to
bring the second coupling partner into the coordination shell
of palladium. Can the same option operate in the classical
cross coupling, where such step is the transmetalation? We be-
lieve that this is exactly the case in the present study.
In our experiments the yields of products formed from the
“involved” bromoarenes were consistently higher than what
could have been expected from the estimates of reactivity
based on the assumption that the reactivity in the cross cou-
pling is governed by the only rate-limiting step, the oxidative
addition. Such behavior should be considered an interesting
anomaly, as there are no known examples of higher of compa-
rable reactivity in Suzuki–Miyaura or other classical CÀC cross-
coupling reactions of bromoarenes with respect to similar io-
doarenes, when the reactions are performed with a sole sub-
strate in a non-competitive mode, and this state of things is
evidenced in thousands of published papers, and, as far as we
know, has been never questioned.
The main difference between the classical CÀC cross-cou-
pling reactions and C-heteroatom cross coupling or CH-aryla-
tion is that the transmetalation is usually perceived as a facile
fast step not requiring any special measures for activation.
Therefore, such reactions are limited by the requirements of
the oxidative addition, which is the sole process defining che-
moselectivity patterns. In this case, it would be extremely diffi-
ChemCatChem 2015, 7, 2113 – 2121
www.chemcatchem.org
2117
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
cult to find a rationale for a markedly enhanced reactivity of
aryl bromides in a competitive reaction in which both aryl
iodide and bromide are present from the start. If the oxidative
addition is the sole limiting stage, the outcome of such runs
would always be the same—aryl bromide would not partici-
pate until the conversion of aryl iodide is near quantitative.
However, there is evidence that Suzuki–Miyaura reaction is
not as indifferent to the transmetalation, at least under some
conditions. Recent studies, performed independently and si-
multaneously by three groups of high authority in the chemis-
try of palladium-catalyzed reactions (Amatore, Jutand, and Le
Duc,[25] Hartwig and co-workers,[26] Schmidt and co-workers[27]
)
Scheme 1. Concurrent pathways in reactions of a mixture of iodo- and
bromoarene with phenylboronic acid (OZ=phosphate or any other oxygen-
centered ligand derived from the base used).
established a key role of the transmetalation step in the
Suzuki–Miyaura reaction, and the nature of base therein, and
revealed that the reaction of the oxidative addition intermedi-
ate ArPdX with arylboronic acid requires prior exchange of
halide ligand for hydroxide or other oxygen-centered nucleo-
phile derived from the inorganic base used.[28] Though this
concept has been recently challenged basing on high-level
theoretical computations of the alternative reaction paths,[29] it
remains a highly relevant mode of reasoning in the discussion
of the roles of various factors in Suzuki–Miyaura and related
chemistry, at least until disproved with more solid experimen-
tal evidence. The relevance of this process for Suzuki–Miyaura
reaction in the presence of palladium nanoparticles was also
demonstrated.[30] The evidence on the dependence of ligand
exchange rate on the nature of halide is somewhat controver-
sial—iodide was found to be the most labile halide, but only
in the reactions in which phosphine complexes of palladium
were used as catalysts. On the contrary, Schmidt and co-work-
ers studied a phosphine-free reaction, and found a substantial
retardation of the exchange by iodide ions.[27] These observa-
tions are more relevant in the context of this research, than
the trends observed with phosphine complexes, as it is well
known that strong donor ligands, such as phosphines, can
exert a substantial influence on the rates of ligand exchange.
In our case, we are dealing with the phosphine-free process,
served by labile complexes of palladium with partially unde-
fined coordination shell, so we may expect to observe a nega-
tive influence of iodide-ions on the rates of ligand exchange
with oxygen centered anions due to several reasons: 1) reser-
vation of palladium in relatively stable dimeric resting states;
2) the same involving anionic ate-complexes; 3) inhibition of
ligand exchange of halide ligand by iodide-ions. Possible path-
ways are depicted in Scheme 1. The formation of dimeric
states including the cross-dimer[31] may substantially influence
the reactivity of palladium catalyst. Unfortunately, the stability
of such complexes cannot be assessed quantitatively, and par-
ticularly there is no reliable data on such complexes lacking at
least a single coordinatively stable ligand such as phosphine.[32]
It is well known that bulky ancillaries of such kind strongly de-
stabilize such complexes,[33] from which it could be reversely
inferred that altogether in the absence of such (that is in the
phosphine-free case, as discussed in the present work) the sta-
bility should be very high. Taking into consideration what is
known about the ligand exchange in the intermediates of the
oxidative addition, it would thus not be unreasonable to spec-
ulate that the dimers gather all palladium species after the oxi-
dative addition and before the ligand exchange for oxygen-
centered base-ligand and further reaction with the boronic
acid. As soon as the iodide ligand in such complexes is likely
to be dramatically less prone for ligand exchange, we could
expect to have more of the aryl iodide oxidative adduct to
hang around as such a dimeric resting state, than the aryl bro-
mide adduct, which undergoes further ligand exchange and
transmetalation to deliver the respective product in a competi-
tive manner disregarding a higher rate of initial oxidative addi-
tion in the case of aryl iodide. Thus, aryl bromides can com-
pete with aryl iodides and be involved in the reaction far
ahead of full consumption of the more reactive electrophile.
The hypothetical mechanism of the involvement
The involvement phenomenon can be rationalized by the ap-
pearance of more reactive species through the initial reaction
of the more reactive substrate (a trigger substrate) with palla-
dium pre-catalyst of relatively low activity. It should be clearly
understood that nanoparticles involved in the catalytic reac-
tion can only undergo gradual deterioration during the reac-
tion because of the Ostwald ripening phenomenon which re-
moves more reactive smaller and defective particles in favor of
large and lazy metal chunks. So how could it be that small par-
ticles be born from the larger ones, is this a paradox and a to-
tally impossible option? Certainly not, as Oswald ripening itself
is not required to take place unidirectionally at any moment of
the lifetime of a given catalytic system. In fact, the essence of
the Oswald ripening is a tendency for such an evolution of
a dispersion system, which leads to a net decrease of total sur-
face area. The mechanism of such an evolution is the redistrib-
ution of matter between particles, involving the processes of
withdrawal and re-attachment from/to the surface. In our case,
the withdrawal is served by the oxidative addition of a mole-
cule of substrate to an atom on the surface (“clipping”) leading
to soluble complexes taking part in the catalytic cycle. The
cross-coupling reaction gives Pd0 species, which, in the phos-
phine-free mode, are improperly ligated and prone to forming,
first, clusters, and then new nanoparticles. The reattachment of
Pd0 born in the catalytic cycle to the surface of its parent nano-
ChemCatChem 2015, 7, 2113 – 2121
www.chemcatchem.org
2118
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
particle, apparently, might seem as a more probable scenario,
than the appearance of new small clusters, at least by proba-
bility reasons. In order for the reattachment to big particles to
happen, however, the parent surface should be formed of
mainly or totally of the reduced Pd0 atoms. It is, however, well
known that the surfaces of metal nanoparticles, including
those of Pd, are commonly oxidized, unless the sol is obtained,
manipulated, and kept under a rigorously anaerobic atmos-
phere and reducing conditions. Otherwise, an essential part of
surface atoms is likely to be oxidized by traces of oxygen, im-
purities, etc. We know of no proven examples that any mature
Pd nanoparticles, formed by a certain procedure, and trans-
ferred to a different system with or without intermediate
ageing, and brought under conditions when dissolution/repre-
cipitation takes place, would continue to grow via the addition
of new layers of atoms over the surface of the “old” particles.
The evidence reported so far shows that very small nanopar-
ticles may possess a strongly enhanced reactivity, as was
shown by Astruc and co-workers.[7,34] Very small particles
should have more Pd centers immediately available for reac-
tion, than do bigger particles in which most atoms are either
in the interior or lie on the crystal planes unavailable for reac-
tion until the leaching would cleave off the neighboring atoms
at vertices and defective sites.[35] A similar observation was
done by Corma and co-workers, who showed that certain ma-
nipulations with pre-formed and rather big and therefore rela-
tively unreactive Pd nanoparticles lead to spontaneous forma-
tion of very small Pd clusters possessing a strongly enhanced
reactivity and exhibiting high TON values.[36] The transforma-
tion of supported complex into supported nanoparticles, fur-
ther operating via the leaching mechanism was recently exper-
imentally demonstrated by Trzeciak and co-workers.[37]
Such a system would operate as if the initial nanoparticles
were the pre-catalyst, and the reaction with the more reactive
substrate—as the pre-catalyst activation process. Certainly,
such an approach would seem to be impractical, as the trigger
substrate performs a sacrificial service, and aryl iodides are too
expensive, though the amount can probably be optimized and
reduced. Moreover, some other reaction can probably be
made to serve this role. In the present study we did not offer
any practical solutions, but rather are willing to draw attention
to several issues of dramatic importance for further application
of nanoparticle-driven systems for phosphine-free palladium
catalyzed reactions. As leaching is likely to be the only mecha-
nism of functioning of such systems, the change of morpholo-
gy during the reaction is inevitable, which makes any ap-
proaches based on the design of more and more sophisticated
supports and nanoparticle preparations poorly justifiable. The
longevity of such nanoparticles cannot be good enough to jus-
tify efforts spent for their preparation and maintenance.
Indeed, these are the most probable reasons why such systems
failed to gain any popularity in the practical area in spite of
the announced performance records. The present work gives
a sketch of quite different an approach, to use a simpler palla-
dium nanoparticle material without any sophisticated supports,
and some conjugated reaction generating a more active Pd0
species in situ in the presence of main substrates ready to
catch on as soon as reactive catalyst is generated in the
system. In this manner, leaching can be exploited as a useful
process, while otherwise this is a fatal inherent feature of
nanoheterogeneous pre-catalysts the only role of which is to
ruin the carefully planned state-of-the-art nanoparticle sys-
tems.
Apparently, this is what happens in our case, as is schemati-
cally depicted in Scheme 2. The initial big nanoparticles, which
apparently do not contain a sufficient number of Pd0 atoms lo-
cated at reactive sites (defects or higher order surface fac-
ets[34b]), are capable only of reaction with the most reactive aryl
iodides. This reaction results in leaching Pd species into reac-
tion media, and Pd0 species lacking stable ancillary ligands,
would form new small nanoparticles, capable of involving the
less reactive substrate.
Conclusions
Thus, we believe that the observation of “involvement”, that is
the appearance of catalytic species of markedly higher activity
in the system initially charged with relatively unreactive nano-
particles and a highly reactive substrate, can be regarded as
a reliable and straightforward demonstration of the leaching
mechanism in the phosphine-free catalytic systems served by
Pd nanoparticle materials. Unlike the well-known leaching
tests, this phenomenon is observed in reaction mixtures undis-
turbed by the addition of alien reagents (mercury), dramatic
change of one of substrates (as in the 3-phase tests), or violent
mechanical perturbation unavoidable in the hot filtration test.
The involvement thus can be regarded as the first non-per-
turbing diagnosis of the redistribution of palladium from the
initial heterogeneous material through leaching and further re-
assembly of new and more reactive nanoparticles. Besides un-
ambiguously proving this mechanism, the involvement phe-
nomenon can be of practical value giving an important clue as
to how the catalytic activity of palladium nanoparticles could
be controlled and tuned up. Indeed, the current development
of nanoparticle pre-catalysts clearly follows the path of sophis-
tication of supports which has led to the introduction of more
and more intricate means of the preparation of Pd nanoparti-
cle materials and sustaining their reactivity, that is equivalent
Scheme 2. The involvement of the less reactive substrate via leaching effect-
ed by the more reactive ArI substrate using large nanoparticles as the initial
pre-catalyst.
ChemCatChem 2015, 7, 2113 – 2121
www.chemcatchem.org
2119
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
at least 5 T1. To suppress NOE-effect in 13C NMR spectra the inverse
gated-decoupling technique was used. Integral intensities of sig-
nals with accuracy of ꢁ1% were determined by an iteration analy-
sis of the total line shape taking into account the residual field in-
homogeneity and phase distortions (INTSPECT2 program.[17]). The
ratio of isotopomers C6H5–C6H5 and C6H5–C6D5 was determined by
comparing of the signal integral intensities for ortho-carbon atoms
for which deuterium induced isotope shift[18] is 17.3 ppb. Relative
contents of deuterated compounds were evaluated from proton-
to attempts to control leaching and the associated Ostwald
ripening. In spite of many spectacular achievements on this
path, it is perfectly evident that such approaches are becoming
more expensive than the modern systems relying on the state-
of-the-art phosphine ligands, while still being vastly inferior in
comparison with the latter in what concerns their reactivity
and reliability. Indeed, the leaching and ripening, if altogether
could be controlled in a limited way, cannot be fully subdued.
The involvement phenomenon can give us a totally different
means—if the leaching and ripening cannot be controlled in
a reliable and economical way, they can be used to generate
the more reactive species from the less reactive and cheaper
ones in a real-time regime, right inside of the catalytic reaction
mixture. Certainly, to make this approach economical we’d
have to find a cheaper reaction responsible for primary leach-
ing than the cross coupling of the expensive iodoarenes. Apart
from the observation of involvement of aryl bromides in the
reaction initiated in the presence of more reactive aryl iodides,
we have observed an unusually high reactivity of the former.
In spite of the commonly known huge difference of the oxida-
tive addition rates between CÀI and CÀBr bonds, under the
“involvement” conditions aryl bromides start to react and are
processed to deep conversions far ahead of the full depletion
of aryl iodide. This observation in our opinion presents the first
proof of the limiting nature of the transmetalation step in the
catalytic cycle, well known in the CÀheteroatom cross coupling
and CH-arylation reactions, but so far unobserved in the classi-
cal CÀC cross-coupling reactions.
2
decoupled H NMR spectrum data. As a result, the following reac-
tion
product
ratio
was
found
[C6H5–C6H5]:[C6H5–
C6D5]:[C6D5Br]:[C6H5I]:[C6D5I]=54.3:30.5:7.7:6.8:0.7.
FE-STEM and FE-SEM Measurements
A target-oriented approach was utilized for the optimization of the
analytic measurements.[19] Before measurements, the samples were
mounted on a 3 mm copper grid and fixed in a grid holder.
Sample morphology was studied by using a Hitachi SU8000 field-
emission scanning electron microscope (FE-SEM). Images were ac-
quired in bright-field STEM mode at 30 kV accelerating voltage.
Samples morphology was studied by scanning electron microscopy
(FE-SEM) under native conditions to exclude metal coating surface
effects.[20] Images were acquired in secondary electron mode at
10 kV accelerating voltage and at working distance 8–10 mm.
Acknowledgements
This work was supported by a grant of Russian Science Founda-
tion N14-23-001186. Electron microscopy characterization was
performed in the Department of Structural Studies of Zelinsky In-
stitute of Organic Chemistry, Moscow. The authors thank Dr. Yury
K. Grishin for registration of NMR spectra.
Experimental Section
Pretreatment of PdNPs
Keywords: leaching
palladium · Suzuki–Miyaura reaction
·
nanocatalysis
·
nanoparticles
·
A glass 5 mL test tube equipped with a magnetic stirrer was
charged with a sample of palladium-coated PVA film (ꢀ4 mg) and
water (2 mL). To attain complete dissolution of polyvinyl alcohol,
the mixture was stirred for 10 min at ꢀ1008C. Palladium was sepa-
rated from the aqueous phase by centrifugation, the solution was
discarded, and this procedure was repeated three times. Traces of
water were removed under reduced pressure.
[1] N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457–2483.
[2] a) I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000, 100, 3009–3066;
b) V. P. Ananikov, I. P. Beletskaya, Organometallics 2012, 31, 1595–1604.
[3] D. Astruc, F. Lu, J. R. Aranzaes, Angew. Chem. Int. Ed. 2005, 44, 7852–
7872; Angew. Chem. 2005, 117, 8062–8083.
[4] a) M. T. Reetz, E. Westermann, Angew. Chem. Int. Ed. 2000, 39, 165;
Angew. Chem. 2000, 112, 170; b) A. Balanta, C. Godard, C. Claver, Chem.
Soc. Rev. 2011, 40, 4973–4985; c) M. Lamblin, L. Nassar-Hardy, J. C.
Hierso, E. Fouquet, F. X. Felpin, Adv. Synth. Catal. 2010, 352, 33–79;
d) A. Fihri, M. Bouhrara, B. Nekoueishahraki, J. M. Basset, V. Polshettiwar,
Chem. Soc. Rev. 2011, 40, 5181–5203.
[5] a) A. V. Selivanova, V. S. Tyurin, I. P. Beletskaya, ChemPlusChem 2014, 79,
1278–1283; b) A. N. Kashin, I. P. Beletskaya, Russ. J. Org. Chem. 2011, 47,
475–479; c) A. N. Kashin, D. A. Panchuk, L. M. Yarysheva, A. L. Volynskii,
I. P. Beletskaya, Russ. J. Org. Chem. 2011, 47, 48–53; d) I. P. Beletskaya,
A. R. Khokhlov, E. A. Tarasenko, V. S. Tyurin, J. Organomet. Chem. 2007,
692, 4402–4406; e) V. P. Ananikov, K. A. Gayduk, I. P. Beletskaya, V. N.
Khrustalev, M. Y. Antipin, Eur. J. Inorg. Chem. 2009, 1149–1161; f) I. P. Be-
letskaya, O. G. Ganina, React. Kinet. Mech. Catal. 2010, 99, 1–4.
[6] C. R. LeBlond, A. T. Andrews, Y. K. Sun, J. R. Sowa, Org. Lett. 2001, 3,
1555–1557.
General procedure for the Suzuki arylation of haloarenes
The test tube was charged in air with palladium catalyst, the corre-
sponding aryl halide (0.075 mmol), the phenylboronic acid
(0.113 mmol), the potassium phosphate (0.225 mmol), and solvent
(0.3 mL). When the reaction was complete, the mixture was diluted
with a saturated aqueous solution of sodium chloride (2 mL), hexa-
decane C16H34 (internal standard, 5–7 mL) was added, and the mix-
ture was extracted with CH2Cl2 (3”2 mL). The extract was dried
over anhydrous sodium sulfate and analyzed by GLC.
NMR measurements
[7] C. Deraedt, L. Salmon, D. Astruc, Adv. Synth. Catal. 2014, 356, 2525–
2538.
[8] a) W. Han, C. Liu, Z. Jin, Adv. Synth. Catal. 2008, 350, 501–508; b) W.
Han, C. Liu, Z.-L. Jin, Org. Lett. 2007, 9, 4005–4007; c) B. Z. Yuan, Y. Y.
Pan, Y. W. Li, B. L. Yin, H. F. Jiang, Angew. Chem. Int. Ed. 2010, 49, 4054–
4058; Angew. Chem. 2010, 122, 4148–4152; d) C. B. Putta, S. Ghosh,
The quantitative analysis of the (C6D5Br) reaction products was
1
2
made using NMR spectroscopy. The H (400.0 MHz), H (61.4 MHz),
and 13C (100.6 MHz) NMR spectra were recorded by using an Agi-
lent 400-MR spectrometer. The relaxation delay between repeti-
tions of the pulse sequence during the signals accumulation was
ChemCatChem 2015, 7, 2113 – 2121
www.chemcatchem.org
2120
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Adv. Synth. Catal. 2011, 353, 1889–1896; e) B. Sreedhar, D. Yada, P. S.
Reddy, Adv. Synth. Catal. 2011, 353, 2823–2836; f) M. Samarasimhared-
dy, G. Prabhu, T. M. Vishwanatha, V. V. Sureshbabu, Synthesis 2013, 45,
1201–1206; g) R. Ghorbani-Vaghei, S. Hemmati, H. Veisi, J. Mol. Catal. A
2014, 393, 240–247; h) X. Liu, X. H. Zhao, M. Lu, J. Organomet. Chem.
2014, 768, 23–27; i) B. Liu, Y. Niu, Y. Li, F. Yang, J. Guo, Q. Wang, P. Jing,
J. Zhang, G. Yun, Chem. Commun. 2014, 50, 12356–12359; j) H. Veisi, D.
Kordestani, S. Hemmati, A. R. Faraji, H. Veisi, Tetrahedron Lett. 2014, 55,
5311–5314; k) V. V. Singh, U. Kumar, S. N. Tripathi, A. K. Singh, Dalton
Trans. 2014, 43, 12555–12563; l) P. Li, P. P. Huang, F. F. Wei, Y. B. Sun,
C. Y. Cao, W. G. Song, J. Mater. Chem. A 2014, 2, 12739–12745; m) Y. S.
Feng, X. Y. Lin, J. Hao, H. J. Xu, Tetrahedron 2014, 70, 5249–5253; n) S.
Yamamoto, H. Kinoshita, H. Hashimoto, Y. Nishina, Nanoscale 2014, 6,
6501–6505; o) G. B. B. Varadwaj, S. Rana, K. Parida, J. Phys. Chem. C
2014, 118, 1640–1651; p) Y. Monguchi, K. Sakai, K. Endo, Y. Fujita, M. Nii-
mura, M. Yoshimura, T. Mizusaki, Y. Sawama, H. Sajiki, ChemCatChem
2012, 4, 546–558; q) Q. Zhou, S. H. Wei, W. Han, J. Org. Chem. 2014, 79,
1454–1460; r) L. Y. Li, T. Wu, J. Y. Wang, R. H. Wang, ChemPlusChem
2014, 79, 257–265; s) D. Dey, T. Bhattacharya, B. Majumdar, S. Mandani,
B. Sharma, T. K. Sarma, Dalton Trans. 2013, 42, 13821–13825; t) V. Pol-
shettiwar, B. Baruwati, R. S. Varma, Acs Nano 2009, 3, 728–736; u) S. A.
Zhou, M. Johnson, J. G. C. Veinot, Chem. Commun. 2010, 46, 2411–
2413; v) F. Lu, J. Ruiz, D. Astruc, Tetrahedron Lett. 2004, 45, 9443–9445.
[9] P.-P. Fang, A. Jutand, Z.-Q. Tian, C. Amatore, Angew. Chem. Int. Ed. 2011,
50, 12184–12188; Angew. Chem. 2011, 123, 12392–12396.
[20] E. Ramirez, S. Jansat, K. Philippot, P. Lecante, M. Gomez, A. M. Masdeu-
Bulto, B. Chaudret, J. Organomet. Chem. 2004, 689, 4601–4610.
[21] G. N. Bondarenko, O. G. Ganina, R. K. Sharma, I. P. Beletskaya, Russ.
Chem. Bull. 2014, in press.
[22] B. P. Fors, N. R. Davis, S. L. Buchwald, J. Am. Chem. Soc. 2009, 131, 5766–
5768.
[23] a) L.-C. Campeau, M. Parisien, A. Jean, K. Fagnou, J. Am. Chem. Soc.
2006, 128, 581–590; b) M. Lafrance, K. Fagnou, J. Am. Chem. Soc. 2006,
128, 16496–16497; c) B. LiØgault, D. Lapointe, L. Caron, A. Vlassova, K.
Fagnou, J. Org. Chem. 2009, 74, 1826–1834.
[24] a) S. K. Guchhait, S. Kandekar, M. Kashyap, N. Taxak, P. V. Bharatam, J.
Org. Chem. 2012, 77, 8321–8328; b) Y.-S. Zhu, B. Shi, R. Fang, X. Wang,
H. Jing, Org. Biomol. Chem. 2014, 12, 5773–5780; c) J. X. Wang, J. A.
McCubbin, M. Z. Jin, R. S. Laufer, Y. Y. Mao, A. P. Crew, M. J. Mulvihill, V.
Snieckus, Org. Lett. 2008, 10, 2923–2926.
[25] a) C. Amatore, A. Jutand, G. Le Duc, Chem. Eur. J. 2011, 17, 2492–2503;
b) C. Amatore, G. Le Duc, A. Jutand, Chem. Eur. J. 2013, 19, 10082–
10093; c) C. Amatore, A. Jutand, G. Le Duc, Chem. Eur. J. 2012, 18,
6616–6625.
[26] B. P. Carrow, J. F. Hartwig, J. Am. Chem. Soc. 2011, 133, 2116–2119.
[27] A. F. Schmidt, A. A. Kurokhtina, E. V. Larina, Russ. J. Gen. Chem. 2011, 81,
1573–1574.
[28] A. J. J. Lennox, G. C. Lloyd-Jones, Angew. Chem. Int. Ed. 2013, 52, 7362–
7370; Angew. Chem. 2013, 125, 7506–7515.
[29] M. A. OrtuÇo, A. Lledós, F. Maseras, G. Ujaque, ChemCatChem 2014, 6,
3132–3138.
[10] D. Cantillo, C. O. Kappe, ChemCatChem 2014, 6, 3286–3305.
[11] F. Tinnis, O. Verho, K. P. J. Gustafson, C.-W. Tai, J.-E. Bäckvall, H. Adolfsson,
Chem. Eur. J. 2014, 20, 5885–5889.
[30] B. D. Briggs, R. T. Pekarek, M. R. Knecht, J. Phys. Chem. C 2014, 118,
18543–18553.
[31] a) A. H. Roy, J. F. Hartwig, Organometallics 2004, 23, 1533–1541; b) A.
Kurbangalieva, D. Carmichael, K. K. Hii, A. Jutand, J. M. Brown, Chem.
Eur. J. 2014, 20, 1116–1125.
[32] a) F. Paul, J. Patt, J. F. Hartwig, J. Am. Chem. Soc. 1994, 116, 5969–5970;
b) V. V. Grushin, H. Alper, Organometallics 1993, 12, 1890–1901.
[33] a) F. Barrios-Landeros, B. P. Carrow, J. F. Hartwig, J. Am. Chem. Soc. 2009,
131, 8141–8154; b) C. L. McMullin, N. Fey, J. N. Harvey, Dalton Trans.
2014, 43, 13545–13556.
[34] a) A. K. Diallo, C. Ornelas, L. Salmon, J. R. Aranzaes, D. Astruc, Angew.
Chem. Int. Ed. 2007, 46, 8644–8648; Angew. Chem. 2007, 119, 8798–
8802; b) G. Collins, M. Schmidt, C. O’Dwyer, G. McGlacken, J. D. Holmes,
ACS Catal. 2014, 4, 3105–3111.
[35] H. Ramezani-Dakhel, P. A. Mirau, R. R. Naik, M. R. Knecht, H. Heinz, Phys.
Chem. Chem. Phys. 2013, 15, 5488–5492.
[12] a) E. V. Johnston, O. Verho, M. D. Kärkäs, M. Shakeri, C.-W. Tai, P. Palmg-
ren, K. Eriksson, S. Oscarsson, J.-E. Bäckvall, Chem. Eur. J. 2012, 18,
12202–12206; b) J. Malmgren, A. Nagendiran, C.-W. Tai, J.-E. Bäckvall, B.
Olofsson, Chem. Eur. J. 2014, 20, 13531–13535.
[13] a) A. H. M. de Vries, J. Mulders, J. H. M. Mommers, H. J. W. Henderickx,
J. G. de Vries, Org. Lett. 2003, 5, 3285–3288; b) C. Deraedt, D. Astruc,
Acc. Chem. Res. 2014, 47, 494–503.
[14] A. S. Sigeev, A. S. Peregudov, A. V. Cheprakov, I. P. Beletskaya, Adv. Synth.
Catal. 2015, 357, 417–429.
[15] a) N. T. S. Phan, M. Van Der Sluys, C. W. Jones, Adv. Synth. Catal. 2006,
348, 609–679; b) K. Kçhler, R. G. Heidenreich, J. G. E. Krauter, M. Pietsch,
Chem. Eur. J. 2002, 8, 622–631; c) I. W. Davies, L. Matty, D. L. Hughes,
P. J. Reider, J. Am. Chem. Soc. 2001, 123, 10139–10140; d) Z. Dong, Z.
Ye, Adv. Synth. Catal. 2014, 356, 3401–3414.
[36] A. Leyva-PØrez, J. Oliver-Meseguer, P. Rubio-Marques, A. Corma, Angew.
Chem. Int. Ed. 2013, 52, 11554–11559; Angew. Chem. 2013, 125, 11768–
11773.
[16] A. Jutand, A. Mosleh, Organometallics 1995, 14, 1810–1817.
[17] F. Fernµndez, B. Cordero, J. Durand, G. Muller, F. Malbosc, Y. Kihn, E.
Teuma, M. Gomez, Dalton Trans. 2007, 5572–5581.
[18] D. Sanhes, E. Raluy, S. Retory, N. Saffon, E. Teuma, M. Gomez, Dalton
Trans. 2010, 39, 9719–9726.
´
[37] E. Mieczynska, T. Borkowski, M. Cypryk, P. Pospiech, A. M. Trzeciak, Appl.
Catal. A 2014, 470, 24–30.
[19] a) Q.-L. Luo, J.-P. Tan, Z.-F. Li, W.-H. Nan, D.-R. Xiao, J. Org. Chem. 2012,
77, 8332–8337; b) M. Murata, T. Oyama, S. Watanabe, Y. Masuda, J. Org.
Chem. 2000, 65, 164–168.
Received: March 27, 2015
Published online on June 17, 2015
ChemCatChem 2015, 7, 2113 – 2121
www.chemcatchem.org
2121
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim