The role of phosphine ligands in the catalytic systems of the Heck reaction with aromatic carboxylic anhydrides

Article abstract of DOI:10.1007/s11172-019-2490-7

The results of a comparative study of phosphine-containing and phosphine-free catalytic systems of the Heck reaction using aromatic carboxylic anhydrides as arylating agents are presented. It was demonstrated that the patterns of diff erential selectivity of the reaction under competition of two aromatic anhydrides or two alkenes are independent of the presence of a tertiary phosphine additive in the system. It was established that palladium complexes with no phosphine ligands in their coordination sphere are catalytically active at the step of activation of aromatic carboxylic anhydride and alkene. The patterns of diff erential selectivity for regioisomers of arylated products provide the evidence of the participation of phosphine-containing anionic palladium complexes in the regioselectivity-determining step of the catalytic cycle. The data obtained are in agreement with the phosphine involvement in the catalyst transformations proceeding outside the main catalytic cycle.

Full text of DOI:10.1007/s11172-019-2490-7

Russian Chemical Bulletin, International Edition, Vol. 68, No. 4, pp. 817—824, April, 2019
817
The role of phosphine ligands in the catalytic systems of the Heck reaction
with aromatic carboxylic anhydrides*
N. A. Lagoda, E. V. Larina, E. V. Yarosh, A. A. Kurokhtina, and A. F. Schmidt
Irkutsk State University,
1 ul. K. Marksa, 664003 Irkutsk, Russian Federation.
Fax: +7 (395 2) 24 2238. E-mail: aschmidt@chem.isu.ru
The results of a comparative study of phosphine-containing and phosphine-free catalytic
systems of the Heck reaction using aromatic carboxylic anhydrides as arylating agents are
presented. It was demonstrated that the patterns of differential selectivity of the reaction under
competition of two aromatic anhydrides or two alkenes are independent of the presence of
a tertiary phosphine additive in the system. It was established that palladium complexes with
no phosphine ligands in their coordination sphere are catalytically active at the step of activation
of aromatic carboxylic anhydride and alkene. The patterns of differential selectivity for regio-
isomers of arylated products provide the evidence of the participation of phosphine-containing
anionic palladium complexes in the regioselectivity-determining step of the catalytic cycle. The
data obtained are in agreement with the phosphine involvement in the catalyst transformations
proceeding outside the main catalytic cycle.
Key words: the Heck reaction, palladium, catalysis, kinetics, mechanism, differential
selectivity.
Palladium-catalyzed alkene arylation with aromatic
anhydrides represents a modification of the well-estab-
lished Heck reaction. This reaction offers a promising
trend of the fine organic synthesis, which has found ap-
plication in production of various pharmaceuticals and
agrochemicals as well as semi-products for tailor-made
polymers.1 Despite the numerous studies of the mechanism
of this catalytic process using various substrates,2—4 many
aspects of the main catalytic cycle and related processes
are still unclear. The majority of homogeneous catalytic
systems reported for the Heck reaction contain various
organic ligands (primarily, phosphines).5 At the same time,
the ligand role in the functioning of catalytic systems still
remains an open question. Thus, it is usually believed6 that
there is at least one phosphorus-containing ligand in the
coordination sphere of palladium complexes directly in-
volved in a catalytic cycle which results in formation of
the reaction product. However, the concepts of the ligand
role usually rest on the data of the model experiments
performed either without some or other components of
the real catalytic system, or with a low substrate/catalyst
ratio.7,8 Such deviations from actual catalytic conditions
can cause errors when these conclusions are extended to
real catalytic systems. In this context, convincing evidence
has been accumulated favoring the idea that it is just
phosphine-free Pd(2+) complexes that are catalytically
active in the Heck reaction with aryl bromides using
phosphine-containing catalytic systems.9 To justify this
conclusion, the results of model experiments under stoi-
chiometric conditions were also invoked. Moreover, the
above conclusion concerning the role of phosphine ligands
in the Heck reaction was drawn for a "classic" variant in
which aryl halides are used as arylating agents. Up to now,
the mechanism of functioning of phosphine-containing
catalytic reaction systems for the reaction with aromatic
carboxylic anhydrides was reported only in two studies.
Thus, the model reaction of oxidative addition of an aro-
matic anhydride to palladium using Pd(PPh3)4 phosphine
precursor in a ratio close to that used in catalytic experi-
ments was examined. The formation of neutral and cationic
benzoyl palladium complexes [(PhCO)Pd(OCOPh)(PPh3)2]
and [(PhCO)Pd(solv)(PPh3)2]+ was established in the
absence of other components of the catalytic reaction.10
Using NMR spectroscopy technique under conditions of
real catalytic process, it was shown that the most part of
palladium (catalyst resting state) in the reaction with
benzoic anhydride existed11 in the form of phosphine-
containing complex [(PhCO)PdX(PPh3)2]. In the latter
complex, halide ions introduced into the catalytic sys-
tem to improve its catalytic activity acted as counter-
ion Х. Nevertheless, the experimentally detected presence
of phosphine-containing palladium compounds cannot
* Based on the materials of the V All-Russian Organic Chemistry
Conference (ROCC-V) (September 10—14, 2018, Vladikavkaz,
Russia).
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 4, pp. 0817—0824, April, 2019.
1066-5285/19/6804-0817 © 2019 Springer Science+Business Media, Inc.

818
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 4, April, 2019
Lagoda et al.
provide any information on the role of such compounds
in functioning of the catalytic system.12,13
Results and Discussion
Obviously, the question of involvement of phosphine
ligands in catalytically active species can be only elucidated
when studying reactions proceeding under real catalytic
conditions. Studies of patterns of differential selectivity of
catalytic systems can become the most important tool in
such investigations. In contrast to the value of catalytic
activity, which is generally measured in kinetic studies,
the value of differential selectivity is related exclusively to
the nature of catalytically active compounds. Therefore,
assessing this parameter upon controlled change under the
process conditions (particularly, on adding phosphine
ligands to the catalytic system) will help to get an unam-
biguous answer to the question whether the ligand addition
results in changing the nature of an active catalyst.
Earlier, we proposed a method of qualitative and quan-
titative assessment of the reaction differential selectivity
under conditions of competition between two similar
substrates by using the so-called phase trajectories repre-
senting the dependencies between the yields of the products
obtained from the competing substrates.14,15 In this case
a slope to any point of the phase trajectory is a ratio of
accumulation rates of the products of competing reactions,
which is related unambiguously to the value of the dif-
ferential selectivity.15 Therefore, a change in phase trajec-
tories obtained on varying the reaction conditions, such
as ligand addition, is an unequivocal evidence for a change
of the differential selectivity value of the catalytic system,
and consequently, for a change in the nature of the active
catalyst. In turn, a coincidence of phase trajectories can
be considered as a proof of invariability of the catalytically
active species. This requires, however, an additional ex-
perimental verification with a wider set of variable para-
meters because of possible lower sensitivity of the selectiv-
ity to a change in variable parameters.
In most studies, selectivity is assessed based on the
target and side reaction products formed from the same
substrate. At the same time, an approach based on analy-
sis of differential selectivity using two equitype competing
substrates, enables one to recognize the case wherein
competing reactions proceed on the catalytically active
species of the same type. This enhances the validity of an
interpretation of the observed regularities in the differen-
tial selectivity changes. Moreover, using pairs of various
competing reagents (e.g., a pair of anhydrides or a pair of
alkenes in the Heck reaction) makes it possible to elucidate
the nature of active catalyst directly in the steps of the
catalytic cycles, in which the given reagents operate. The
conventional mechanism of the catalytic cycle of the Heck
reaction, that includes aromatic anhydrides (ligands at-
tached to the palladium atom are omitted) is illustrated in
Scheme 1. In Scheme 2 the mechanism for the Heck reac-
tion with two aromatic anhydrides (a) or two alkenes (b)
competing is shown. Scheme 3 represents catalytically
active complexes the nature of which governs the value of
differential selectivity when a pair of aromatic anhydrides
(a) or a pair of alkenes (b) are involved in competition.
The value of differential regioselectivity (c) is also depen-
dent of the complexes.
Experimental
Samples of the reaction mixture were analyzed on a Chrom-
atech-Crystal 5000 gas-liquid chromatograph (Chromatech,
Russia) (FID, column HP-5 15 m) and Shimadzu GC-MS
QP-2010 Ultra gas chromatograph-mass spectrometer (Shimadzu,
Japan) with electron-impact ionization (ionization energy 70 eV,
column GsBP-5MS 0.25 m×0.25 mm×30 m, helium as carrier
gas) and programmed heating from 100 to 250 С. Mass spectra
obtained were compared to the library ones (Wiley, NIST, and
NIST05 mass spectral libraries). Quantitative sample composi-
tions were determined using the internal standard (naphthalene)
method with calibration on authentic samples.
UV spectra of the solutions under investigation were re-
corded on a SF-2000 spectrophotometer (EDO «Spektr», Russia) in
the wavelength range of 190—600 nm in quartz cells with 0.01 cm
optical path length. Concentrations of complex Pd(PPh3)2Cl2
were measured using the absorption intensity at 350 nm. A cali-
bration curve for this wavelength was pre-constructed based on
the spectra of reference solutions of Pd(PPh3)2Cl2 in N-methyl-
pyrrolidone (NMP). Herewith, also through plotting the cal-
ibration curves, the absorbance maximum at 350 nm from
stilbene, 4-methoxystilbene, n-butyl cinnamate, and chalcone
formed in the process was used with quantitites of the said sub-
stances being determined using chromatography-mass spectro-
metry technique.
Catalytic experiments. All experiments reported herein were
run in inert atmosphere (argon). Reagents and solvents in use
were preliminary degassed and stored under argon. Exper-
iments on arylating alkenes with aromatic anhydrides were per-
formed by mixing a pair of competing anhydrides or alkenes
(2.5 mmol each), a common reagent (styrene or benzoic an-
hydride respectively, 5 mmol), a halide salt additive (0.5 mmol),
and naphthalene (0.5 mmol) as an internal standard, in 5 mL of
NMP. The resulting solution was purged with argon and trans-
ferred into a glass reactor, pre-evacuated and filled with argon,
fitted with a rubber membrane and a magnetic stirrer. The reac-
tor with pre-added PdCl2 (0.08 mmol) and in some cases PPh3
(0.16 mmol), was then placed in an oil bath thermostatized
at 140 С. The experiments were run at constant stirring.
Each experiment was carried out in triplicate for a reproduc-
ibility check.
According to the generally accepted concepts, an aro-
matic anhydride is activatied in a catalytic cycle of the
Heck reaction at the step of its oxidative addition to Pd(0)
complexes10 (see Scheme 1, А). Therefore, evaluation of
differential selectivity patterns based on the products of
alkene arylation with two aromatic anhydrides competing
(see Scheme 2, a) enables one to get insight into the nature
of Pd(0) complexes (see Scheme 3, a) involved in the
oxidative addition step. Experiments with competing al-

Role of phosphine in the Heck reaction
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 4, April, 2019
819
Scheme 1
R = Ar, C(O)OAlk
Scheme 2
a
b
[Pd] = PdCl2, PdCl2 + 2 PPh3
MX = LiCl, LiBr, NaBr, NaCl, NBu4Br

820
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 4, April, 2019
Lagoda et al.
Scheme 3
presence of LiCl and NaBr (see Fig. 1), but also in the
presence of LiBr and NBu4Br. It is assumed that formation
of all products 1—4 requires activation of an anhydride at
the oxidative addition step (see Scheme 1, A). In this case,
coincidence of phase trajectories constructed from the
sums of the products obtained with and without phosphine,
suggests the similar nature of formed catalytically active
compounds reacting with the anhydride. In terms of the
generally accepted concepts of the reaction mechanism
(see Scheme 1, A and Scheme 3, a) these should be Pd(0)
complexes having no phosphine ligands in their coordina-
tion sphere.
a
On going from phosphine-free to the phosphine-
containing catalytic systems, the value of catalytical activ-
ity decreased significantly. The initial consumption rate
of a reagent (styrene) common for competing aromatic
anhydrides showed a 2—2.5-fold decrease. Nevertheless,
on using these two types of catalytic systems, the value of
differential selectivity remained the same thus unequivo-
Ar1 = Ph, Ar2 = 4-MeOC6H4, n = 1—4, m = 0—3;
M = Li+, Na+, NBu4+; X = Cl–, Br–
L are neutral ligands (solvent).
b
C1•102/mol L–1
a
50
40
R1 = Ph, R2 = COOBu, n = 2—3, m = 0—1;
M = Li+, Na+, NBu4+; X = Cl–, Br–
L are neutral ligands (solvent).
1
30
2
20
10
c
0
10
20 C2•102/mol L–1
C1•102/mol L–1
b
R = Ph, COOBu; Ar = Ph, 4-MeOC6H4;
M = Li+, Na+, NBu4+; X = Cl–, Br–
L are neutral ligands (solvent).
50
40
30
20
10
kenes (see Scheme 2, b) provide knowledge about the step
of the catalytic cycle involving alkenes (see Scheme 1, B),
wherein alkenes react with PdII -aryl complex according
to the generally accepted mechanism (see Scheme 3, b).
Carrying out the competing arylation of styrene with
benzoic and 4-methoxybenzoic anhydrides (see Scheme 2, a)
it was found that apart from "Heck´s" products of arylation
at α- and β-positions of alkenes 1 and 2, the formation of
chalcones 3 and dihydrochalcones 4 was observed. Phase
trajectories constructed from the sums of all products 1—4
obtained from each competing anhydride using phosphine-
free catalytic system and a system with triphenylphosphine
additive, were identical (Fig. 1). On phosphine addition,
the phase trajectories remained identical not only in the
1
2
0
10
20 C2•102/mol L–1
Fig. 1. Phase trajectories of competing arylation of styrene
with benzoic and 4-methoxybenzoic anhydrides using catalytic
systems: PdCl2 + LiCl (1) and PdCl2 + LiCl+PPh3 (2) (a);
PdCl2 + NaBr (1) and PdCl2 + NaBr + PPh3 (2) (b); C1
= C(1a+2a+3a+4a), C2 = C(1b+2b+3b+4b)
=
.

Role of phosphine in the Heck reaction
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 4, April, 2019
821
cally suggesting the similar nature of the catalytically active
species. This result demonstrates that studies of the dif-
ferential selectivity patterns can provide more reliable
conclusions than evaluation of the catalytical activity. In
fact, the activity value depends not only on the type of an
active catalyst but also on variations in its amount that can
occur even when the nature of the catalytically active
species remains unchanged.
tion of close ion pairs with catalytically active anionic
palladium species.16
For the Heck reaction with aromatic anhydrides, an-
other type of competing experiments is also possible
wherein two alkenes, rather than an anhydride, will com-
pete for the common catalyst (see Scheme 2, b). In this
case differential selectivity patterns based on the sums of
the yields of the arylation products 1 and 2 makes it pos-
sible to examine the nature of the catalytic species involved
in the activation of alkenes that results in the formation of
these products (see Scheme 3, b). According to the ac-
cepted concepts of the mechanism of the Heck reaction
catalytic cycle, activation of alkene takes place during the
step оf its coordination to ArPdX complexes (where Х is
halide or benzoate anion) (see Scheme 1, B).6 Тhus,
analysis of phase trajectories of the reactions yielding the
traditional arylation products 1 and 2 under competing of
two alkenes, namely, styrene and n-butyl acrylate, can
provide insight into the nature of active species of the
ArPdX type (see Scheme 3, b).
As it follows from the data obtained in our group, dif-
ferential selectivity for competing alkenes in the catalytic
cycle of formation of the "Heck´s" products 1 and 2, as in
the case with competing pair of aromatic anhydrides, does
not depend on the presence of phosphine ligands in the
catalytic system (Fig. 3). This suggests the involvement of
the active -aryl complexes of Pd(2+) without phosphine
ligands in the step of the catalytic cycle wherein competi-
tion of alkenes takes place. However, in this case the
dependence of differential selectivity on the nature of
a cation and an anion of the salt presenting in the catalytic
system is observed. This observation is in agreement with
Earlier, we have demonstrated that Pd(0) complexes,
active at the oxidative addition step in arylation of alkenes
with aromatic anhydrides using phosphine-free catalytic
systems, have an anionic nature.16 In this case, the value
of differential selectivity is governed by the nature of both
an anion and a cation of the salt applied as an additive to
the catalytic system. As evident from our data (Fig. 2), the
value of differential selectivity on competing aromatic
anhydrides in the presence of phosphine-containing
catalytic system also depends on the nature of both an
anion and a cation of the salt additive. An obvious ex-
ample provide phase trajectories shown in Fig. 2 con-
structed using LiCl and LiBr or NBu4Br and NaBr re-
spectively. Effect of the nature of the salt anion on selec-
tivity is explained by its incorporation in the coordination
sphere of Pd(0) catalytically active species. A possibility
of formation of Pd(0) anionic complexes under cross-
coupling conditions was demonstrated using both ex situ
and in situ studies under stoichiometric6,17 and catal-
ytic18,19 conditions. When an alkaline cation is unable to
form a coordinate bond with the palladium atom its effect
on the selectivity is most likely accounted for by the forma-
C1•102/mol L–1
C (1a+2a)•102/mol L–1
45
40
35
30
18
16
14
12
1
2
3
4
1
25
2
3
4
5
6
10
8
20
15
10
5
6
4
2
5
10
15 C2•102/mol L–1
10
20
30 C (1c+2c)•102/mol L–1
Fig. 2. Phase trajectories of competing arylation of styrene
with benzoic and 4-methoxybenzoic anhydrides using catalytic
systems: PdCl2 + LiCl + PPh3 (1), PdCl2 + LiBr + PPh3 (2),
PdCl2 + NaBr + PPh3 (3), PdCl2 + NBu4Br + PPh3 (4); C1 =
Fig. 3. Phase trajectories of competing arylation of styrene and
n-butyl acrylate with benzoic anhydride using catalytic systems:
PdCl2 + LiCl (1), PdCl2 + LiCl + PPh3 (2), PdCl2+NaCl (3),
PdCl2 + NaCl + PPh3 (4), PdCl2 + NaBr (5), PdCl2 + NaBr +
+ Ph3 (6).
= C(1a+2a+3a+4a), C2 = C(1b+2b+3b+4b)
.

822
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 4, April, 2019
Lagoda et al.
C1a•102/mol L–1
the participation of anionic Pd(2+) species in the catalysis
similar to the competition of a pair of aromatic anhydrides
(analogously to a phosphine-free reaction modification).16
The patterns of differential selectivity outlined above
present the evidence that phosphine ligands introduced
into a catalytic system are not incorporated into catalyti-
cally active species on arylation of alkenes with aromatic
anhydrides both at the step of activation of an anhydride
molecule (see Scheme 1, A), and at the step of alkene
activation (see Scheme 1, B).
a
40
30
20
10
1
2
Additional information concerning a possible transfer
of phosphine ligands into intermediates of the Heck reac-
tion catalytic cycle can be derived from patterns of dif-
ferential selectivity of formation of products 1 and 2 from
one of the competing substrates. According to the gener-
ally accepted mechanism, the step determining differential
regioselectivity for such products includes an alkene inser-
tion into the palladium—carbon bond (the step following
the coordination of the alkene to ArPdX complexes) (see
Scheme 1, C). It was established that the differential se-
lectivity value of formation of regioisomer products 1а and
2а changes when passing from phosphine-free to phos-
phine containing catalytic systems. This change is observed
when using both a pair of competing aromatic
anhydrides and a pair of competing alkenes (see Scheme 2).
Thus, for example, with LiCl or LiBr as additives to the
catalytic system, phase trajectories in coordinates of con-
centrations of stilbene 1а (β-regioisomeric product) and
1,1-diphenylethylene 2а (α-regioisomeric product) were
different (Fig. 4). Such result unequivocally suggests
changing the nature of catalytically active compounds
involved in the step of an alkene insertion (see Scheme 1, C).
The change in the nature of catalytically active compounds
can result from incorporation of a phosphine ligand into
the active palladium species, which govern the differential
selectivity of formation of regioisomer "Heck´s" products
(see Scheme 3, c). Regioselectivity in the presence of
a phosphine depends not only on the nature of an anion
of the salt presenting in the system but also on its cation
nature. For example, phase trajectories for phosphine
systems with LiCl and NaCl additives differ significantly
(Fig. 5). Similar to the previously published data,16
this result can be explained by the presence of anionic
complexes in the system forming close ionic pairs with
a salt anion which is unable to adopt coordination to the
palladium atom. However, taking into account the inher-
ent ability of the palladium atom to form tetrahedrally
coordinated saturated complexes, one can suggest that
-aryl complex containing, apart from a halogen anion
and an alkene, at least one phosphine ligand, should be
electroneutral (see Scheme 3, c). Dependence of regiose-
lectivity on concentration of phosphine and cation of the
salt additive testifies that phosphines are incorporated into
an active complex (see Fig. 4). This dependence also sug-
gests the anionic nature of the active complex. All these
0
1
2 C 2a•102/mol L–1
C1a•102/mol L–1
30
b
25
20
15
10
5
1
2
0
1
2
C 2a•102/mol L–1
Fig. 4. Phase trajectories of accumulation of the "Heck´s" re-
gioisomeric products 1а and 2а of arylation of styrene with
benzoic anhydride (under conditions of competing arylation of
styrene with benzoic and 4-methoxybenzoic anhydrides) using
catalytic systems: PdCl2 + LiCl (1) and PdCl2 + LiCl+PPh3 (2) (a),
PdCl2 + LiBr (1) and PdCl2 + LiBr + PPh3 (2) (b).
data strengthen the suggestion that in a real catalytical
system phosphine ate-complexes of PdII are formed (see
Scheme 3, c) similar to those recently detected under
model non-catalytic conditions.17
The data obtained in the present study clearly demon-
strate the advantages offered by measurements of the dif-
ferential selectivity. This method can be used primarily to
solve the problem of participation of phosphine-contain-
ing complexes in the main catalytic cycle of the Heck
reaction. The results thus obtained also enable one to

Role of phosphine in the Heck reaction
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 4, April, 2019
823
C1a•102/mol L–1
well with a decrease of the catalytic reaction rate observed
in all runs with a phosphine added. Thus, phosphine li-
gands play dual role in arylation of alkenes with aromatic
anhydrides. First, they are incorporated into Pd(2+)
complexes, active at the insertion step, that govern the
regioselectivity of formation of traditional products of
α- and β-arylation of an alkene. Secondly, they are in-
volved in the transformations of palladium outside the
main catalytic cycle wherein the reaction products are
formed. At the same time, contrary to the generally
accepted concepts of the Heck reaction mechanism,
catalytically active species participating in the steps of the
substrate activation contain no phosphine ligands.
12
10
8
6
4
2
1
2
Тo summarize, differential selectivity patterns of the
Heck reaction with aromatic anhydrides using phosphine-
free and phosphine-containing catalytic systems were
analyzed. A conclusion has been drawn that complexes of
Pd(0) and Pd(2+) involved in the steps of the catalytic
cycle wherein activation of an aromatic anhydride and
alkene respectively takes place, have no phosphine ligands
in their coordination sphere. Dependence of differential
selectivity on the nature of a cation and an anion of the
salt added to the catalytic system suggests that the anionic
nature of active complexes is not changed in the presence
of a phosphine similar to phosphine-free catalytic systems.
At the same time features of differential selectivity of
formation of traditional “Heck´s” products indicate that
the phosphine ligands are incorporated into catalytically
active species and in this way determine selectivity of the
regioisomeric products formation. Dependence of dif-
ferential selectivity on a phosphine and a cation of a salt
additive to the catalytic system makes it possible to assume
the formation of anionic phosphine complexes represent-
ing probably pentacoordinated ate-complexes Pd(2+).
Moreover, starting from the experimental data and taking
into account the observed effects of the phosphine addition
on the catalytic activity and UV spectroscopy data, one
can make a conclusion that phosphine ligands are also
involved in the transformation of a catalysts outside the
main catalytic cycle wherein the reaction products are
formed.
0
0.5
1.0 C2a•102/mol L–1
Fig. 5. Phase trajectories of accumulation of the "Heck´s" re-
gioisomeric products 1а and 2а of arylation of styrene with
benzoic anhydride (under conditions of competing arylation of
styrene and n-butyl acrylate with benzoic anhydride) using
catalytic systems: PdCl2 + LiCl + PPh3 (1), PdCl2 + NaCl +
+ PPh3 (2).
elucidate the possibility for such ligands to be incorporated
into active intermediates involved in certain steps of the
cycle, namely, activation of anhydrides and alkenes, as
well as the regioselectivity-determining step. The results
obtained therefrom do not depend on a hypothetical
mechanism of these steps. Various experiments performed
under conditions, in which two aromatic anhydrides or
two alkenes compete were directed on evaluation of dif-
ferential selectivity. Investigations of alkene arylation were
aimed at evaluation of differential regioselectivity of forma-
tion of traditional α- and β-products. The obtained data
allows one to distinguish between situations, in which
a phosphine ligand forms a part of catalytically active pal-
ladium π-complexes determining regioselectivity of the
alkene insertion (see Scheme 1, С and Scheme 3, c) and
those, in which catalytically active Pd(0) and Pd(2+)
complexes involved in the steps of activation of an aromatic
anhydride and an alkene respectively (see Scheme 1, A and
B, and Scheme 3, a and b), have a phosphine-free nature.
A conclusion of the absence of phosphine ligands in
the active palladium complexes participating in the oxida-
tive anhydride addition and coordination of alkene appears
to be in conflict with the known ability of tertiary phos-
phines to form stable complexes with palladium including
the cross-coupling conditions.6,11 Actually, we were able
to detect in the UV spectra of the reaction mixtures, e.g., in
the presence of chlorine-containing salt additives, a con-
siderable amount of phosphine complex PdCl2(PPh3)2
(up to 70% of the total amount of palladium added to
the system). However, according to the concepts of the
reaction mechanism this complex is catalytically inactive
and is outside the main catalytic cycle. This concept fits
The study was financially supported by the Russian
Foundation for Basic Research (Project No. 18-33-
00362_mol_а).
References
1. J. P. Corbet, G. Mignani, Chem. Rev., 2006, 106, 2651.
2. J. G. De Vries, in Selective Nanocatalysts and Nanoscience,
Eds A. Zecchina, S. Bordiga, E. Groppo, Wiley-VCH Verlag
GmbH & Co. KGaA, Weinheim, 2011, p. 73.
3. D. B. Eremin, V. P. Ananikov, Coord. Chem. Rev., 2017, 346, 2.
4. A. Biffis, P. Centomo, A. Del Zotto, M. Zecca, Chem. Rev.,
2018, 118, 2249.

824
Russ. Chem. Bull., Int. Ed., Vol. 68, No. 4, April, 2019
Lagoda et al.
5. I. P. Beletskaya, A. V. Cheprakov, in The Mizoroki—Heck
reaction, Ed. M. Oestreich, John Wiley and Sons Ltd.,
Munster, 2009, p. 51.
6. A. Jutand, in The Mizoroki—Heck reaction, Ed. M. Oestreich,
John Wiley and Sons Ltd., Munster, 2009, p. 1.
7. C. Amatore, E. Came, A. Jutand, M. A. M´Barki, G. Meyer,
Organometallics, 1995, 14, 5605.
8. C. Amatore, B. Godin, A. Jutand, F. Lemaitre, Chem. Eur.
J., 2007, 13, 2002.
9. B. P. Carrow, J. F. Hartwig, J. Am. Chem. Soc., 2010, 132, 79.
10. A. Jutand, S. Negri, J. G. de Vries, Eur. J. Inorg. Chem. 2002,
7, 1711.
14. A. F. Schmidt, A. A. Kurokhtina, E. V. Larina, Kinet. Catal.,
2012, 53, 84.
15. A. F. Schmidt, A. A. Kurokhtina, E. V. Larina, Cat. Sci.
Technol., 2014, 4, 3439.
16. A. F. Schmidt, A. A. Kurokhtina, E. V. Larina, E. V. Yarosh,
N. A. Lagoda, Organometallics, 2017, 36, 3382.
17. M. Kolter, K. Boeck, K. Karaghiosoff, K. Koszinowski,
Angew. Chem., Int. Ed., 2017, 56, 13244.
18. F. Schroeter, J. Soellner, T. Strassner, ACS Catal., 2017,
7, 3004.
19. F. Schroeter, T. Strassner, Inorg. Chem., 2018, 57, 5159.
11. A. F. Shmidt, V. V. Smirnov, Kinet. Catal., 2002, 43, 195.
12. A. F. Schmidt, A. A. Kurokhtina, Kinet. Catal., 2012, 53, 714.
13. A. F. Schmidt, A. A. Kurokhtina, E. V. Larina, Mendeleev
Commun., 2017, 27, 213.
Received October 30, 2018;
accepted February 26, 2019