Palladium-Catalyzed Ethylene/Methyl Acrylate Cooligomerization: Effect of a New Nonsymmetric α-Diimine

Article abstract of DOI:10.1002/cctc.201200520

A new nonsymmetric bis(aryl-imino)acenaphthene (Ar-BIAN) ligand, featured by a subtle steric and electronic unbalance of the N-donor atoms, is reported. With the new ligand and the corresponding symmetrically substituted derivatives, both neutral and monocationic Pd-CH₃ compounds have been synthesized and characterized. The series of the monocationic complexes [Pd(CH₃)(L)(Ar-BIAN)][PF₆] (L=CH₃CN, dmso) has been extended to dimethyl sulfoxide derivatives. The monocationic complexes are tested as precatalysts for the ethylene/methyl acrylate cooligomerization under mild reaction conditions of temperature and ethylene pressure. The catalytic product is a mixture of ethylene/acrylate cooligomers and higher alkenes. The catalysts containing the new nonsymmetric ligand are found to be more productive than those with the symmetric Ar-BIANs. The Pd-dmso catalysts are more productive and show a longer lifetime than their Pd-NCCH₃ counterparts.

Full text of DOI:10.1002/cctc.201200520

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201200520
Palladium-Catalyzed Ethylene/Methyl Acrylate
Cooligomerization: Effect of a New Nonsymmetric
a-Diimine
Angelo Meduri,[a] Tiziano Montini,[a] Fabio Ragaini,[b] Paolo Fornasiero,[a] Ennio Zangrando,[a]
and Barbara Milani*[a]
A new nonsymmetric bis(aryl-imino)acenaphthene (Ar-BIAN)
ligand, featured by a subtle steric and electronic unbalance of
the N-donor atoms, is reported. With the new ligand and the
corresponding symmetrically substituted derivatives, both neu-
tral and monocationic PdꢀCH3 compounds have been synthe-
sized and characterized. The series of the monocationic com-
plexes [Pd(CH3)(L)(Ar-BIAN)][PF6] (L=CH3CN, dmso) has been
extended to dimethyl sulfoxide derivatives. The monocationic
complexes are tested as precatalysts for the ethylene/methyl
acrylate cooligomerization under mild reaction conditions of
temperature and ethylene pressure. The catalytic product is
a mixture of ethylene/acrylate cooligomers and higher alkenes.
The catalysts containing the new nonsymmetric ligand are
found to be more productive than those with the symmetric
Ar-BIANs. The Pd–dmso catalysts are more productive and
show a longer lifetime than their Pd–NCCH3 counterparts.
Introduction
Functionalized polymers featured by a polyolefin skeleton
modified with polar groups represent highly desirable materi-
als that should express surface properties, such as adhesion,
dye-ability, printability, and compatibility, that are insufficiently
present in unmodified polyolefins.[1–6] Commercial processes
for the incorporation of polar monomers are currently based
either on radical polymerization or on post-polymerization
functionalization—two technologies that suffer from high
energy consumption, low cost-efficiency, and poor control over
the polymer microstructure. To overcome these problems, the
most straightforward approach used is based on the develop-
ment of highly efficient catalysts for the controlled, direct co-
polymerization of terminal alkenes with polar monomers. The
discovery of such a catalyst has been indicated as one of the
“holy grails” in the field of olefin polymerization.[7]
Figure 1. The three classes of complexes reported in the literature: a) a-di-
imino derivatives; b) phosphine sulfonate derivative; c) bisphosphine mon-
oxide derivatives.
polymerization[8,9] opens the way to apply them in the ethyl-
ene/polar monomer copolymerization. The two main catalytic
systems reported in the literature are based on palladium com-
plexes and differ in the nature of the ancillary ligand present
in the metal coordination sphere, which in one case is a neutral
a-diimine[10,11] and in the other an anionic phosphine sulfonate
derivative (PꢀO) (Figure 1).[12]
Although the use of early transition metals, applied usually
as catalysts for polyolefin synthesis, is hampered because of
their oxophilicity, Brookhart’s discovery that compounds based
on late transition metals are efficient catalysts for ethylene
In the copolymerization of ethylene with methyl acrylate
(MA), the model reaction, both catalytic systems lead to real
copolymers and not to a mixture of two homopolymers. The
two catalytic systems are operative under quite different reac-
tion conditions that limit a direct comparison of the catalytic
results. Nevertheless, some differences are evident: the a-dii-
mine system leads to amorphous copolymers with a number
average molecular weight of 300–88000, with an incorporation
of the polar monomer content from 1.0 to 12.1% into the
branches of the polymer chain;[10] the PꢀO system leads to the
copolymer with a number average molecular weight of 3900–
12800, with an incorporation of the polar monomer content
from 3 to 17% into the linear polyethylene backbone.[12] In the
case of the PꢀO system, a remarkable improvement in the cat-
alytic performances is realized moving from the in situ catalytic
system to preformed complexes;[13] for example, [Pd(CH3)-
[a] A. Meduri, Dr. T. Montini, Prof. Dr. P. Fornasiero, Prof. E. Zangrando,
Dr. B. Milani
Department of Chemical and Pharmaceutical Sciences
University of Trieste, Via Licio Giorgieri 1, 34127 Trieste (Italy)
Fax: (+39)0405583903
E-mail: milaniba@units.it
[b] Prof. Dr. F. Ragaini
Department of Chemistry
University of Milano, Via C. Golgi 19, 20133 Milano (Italy)
Supporting information for this article contains crystal data, details of
the structure solution, and refinements, NMR spectra of complexes and
ethylene/MA cooligomers, in situ reactivity investigations, GC–MS analy-
ses and their Poisson fitting, and modeling of the oligomer growth rate,
and is available on the WWW under http://dx.doi.org/10.1002/
cctc.201200520.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1170

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
(dmso)(PꢀO)] yields the copolymer with a content of MA up to
52%.[14]
gands,[14,27,28] this is the first report in which Pd–dmso com-
plexes are used in conjunction with a-diimine ligands. The cat-
alytic behavior of these complexes in ethylene/MA cooligome-
rization has been investigated in detail along with mechanistic
studies.
In addition, the catalysts based on the PꢀO system are quite
versatile because they can catalyze the copolymerization of
ethylene with various polar monomers, such as vinyl acetate,[15]
acrylonitrile,[16] allyl monomers,[17] alkyl vinyl ether,[18] and vinyl
fluoride.[19] They were also used for emulsion copolymeriza-
tion[20] and for the copolymerization of ethylene with nitrogen-
containing polar monomers.[21]
The importance of having a nonsymmetric ligand on palladi-
um for this catalysis has been highlighted by the catalytic be-
havior of palladium complexes with the PꢀO ligand. Here it is
of interest to evaluate whether a subtle differentiation of the
two N-donor atoms could result in remarkable effects on the
catalytic performances.
Contrarily, the literature on the a-diimine system is rather
limited and the system has been extended to a peculiar cyclo-
phane-modified ligand[22] that yields an ethylene/MA copoly-
mer containing 21.8% of the polar monomer. The Pd–a-dii-
mine system also catalyzes the copolymerization of 1-hexene
with silyl vinyl ether[23] and of ethylene with 2-hydroxyethyl
acrylate.[24]
Results and Discussion
Synthesis and characterization of ligand 1 and of its
palladium complexes
A palladium catalytic system with a bisphosphine monoxide
ligand catalyzes the copolymerization of ethylene with many
polar monomers but MA, which leads to the highly linear co-
polymers with a random distribution of the polar functional
groups into the polymer chain (Figure 1c).[25]
Ligand 1 [(2,6-dimethyl-phenyl),(3,5-bis(trifluoromethyl)phenyl)
diiminoacenaphthene] was synthesized by using a method
slightly modified with respect to that reported previously for
the nonsymmetrically substituted mAr-BIANs.[26,29] In the transi-
mination reaction, the zinc chlorido derivative coordinated
with ligand 4 was reacted with 2,6-dimethyl aniline (Scheme 1)
at room temperature for several days. The oil obtained after
zinc decoordination contained ligand 1 together with the un-
reacted aniline and 3,5-bis(trifluoromethyl) aniline.
For these catalytic systems, the productivity is not practical
and the discovery of highly efficient and inexpensive catalysts
is a crucial challenge.[7]
We studied a series of a-diimine ligands featured by the ace-
naphthene skeleton and the aryl rings substituted in the meta
position (mAr-BIAN).[26] The corresponding palladium com-
plexes, [Pd(CH3)(NCCH3)(mAr-BIAN)][PF6], are efficient catalysts
for the CO/vinyl arene copolymerization, which leads to poly-
ketones with an atactic microstructure. The precatalysts with
the nonsymmetrically substituted mAr-BIAN were found to be
more productive than those with the related symmetrically
substituted ligands, which thus indicates that the unbalance of
the electronic properties of the ligand has a positive effect on
the catalytic performances. On the basis of these considera-
tions and with the aim of creating both a steric and an elec-
tronic unbalance of the two N-donor atoms, the series of non-
symmetric Ar-BIANs has been extended to a new compound
featured by an aryl ring with an electron-withdrawing group
on the meta positions and the other ring with an electron-re-
leasing group on the ortho positions (1; Figure 2).
Scheme 1. Synthesis of ligand 1.
With ligand 1 and the corresponding symmetrically substi-
tuted derivatives 2 and 4, two series of monocationic palladi-
um compounds were synthesized: [Pd(CH3)(L)(Ar-BIAN)][PF6]
(L=CH3CN, dmso). Although the complexes with dimethyl sulf-
oxide have been investigated extensively with PꢀO li-
Ligand 1 was isolated in good yield (up to 45%) from the
crude reaction mixture by using column chromatography. As
an alternative methodology, the crude oil can be reacted di-
rectly with [Pd(CH3)Cl(cod)] (cod=1,5-cyclooctadiene), which
leads to pure [Pd(CH3)Cl(1)] (1a).
In general, the synthesis of nonsymmetrically substituted Ar-
BIANs is not so trivial. Several attempts to synthesize the steri-
cally more hindered nonsymmetric ligand [(2,6-di-isopropyl-
phenyl),(3,5-bis(trifluoromethyl)phenyl)] diiminoacenaphthene
either through a transimination reaction or in two steps have
failed. The mixtures of products were obtained, from which
only the symmetric Ar-BIAN ligand with four isopropyl groups
could be isolated in a pure form.
Figure 2. The studied a-diimines, 1–5, and their numbering scheme.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1171

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Ligand 1 was characterized both in the solid state and in so-
lution. For BIAN ligands, (E,E) and (E,Z) isomers are possible de-
pending on the relative configuration of the aryl rings with re-
spect to C=N imine bonds.[30] For Ar-BIAN ligands, the (E,E)
isomer is preferentially observed in the solid state,[30–34] except
for few, such as the ligands featured by biphenyl or naphthyl
substituents showing the (E,Z) isomer as the only molecule
present in the unit cell.[35,36]
Scheme 2. Synthesis of palladium complexes 1a–5a, 1b–5b, 1c, 2c, and
4c.
Suitable crystals for X-ray analysis of 1 were obtained by
slow diffusion of hexane into a dichloromethane solution at
48C (Figure 3).
Figure 4. Schematic drawing of cis and trans isomers and of the complexes
with the symmetric Ar-BIAN ligands. For R, see Figure 2.
of the cis isomer are observed. This is in agreement with the
stronger trans influence of the methyl group with respect to
chlorido and the weaker coordinating strength of the nitrogen
atom bearing the trifluoromethyl-substituted aryl group with
respect to that containing the methyl groups. The two isomers
are in equilibrium as demonstrated by both NOESY and ROESY
experiments.
Figure 3. ORTEP drawing (thermal ellipsoids at the 30% probability level) of
ligand 1. Of the disordered F1-3 group, only flourines at higher occupancy
are shown.
In agreement with the NMR data of analogous Pd–(Ar-BIAN)
complexes,[26,37] owing to coordination to palladium the signals
of H3 and H10 are shifted in the opposite direction with respect
to the same signals in the free ligand, which is the doublet of
H3 at a frequency higher than that of H10. This shift is attribut-
ed to its position cis to the chlorido ligand.[38] The singlet of
PdꢀCH3 is sensitive to the nature of Ar-BIAN coordinated to
the metal center: in complex 4a it resonates at 0.78 ppm, in
1a at 0.72 ppm, and in 2a at 0.61 ppm, moving to shielding
direction on increasing the electron density on the N-donor
atoms and therefore on palladium.
Ligand 1 shows a (E,Z) isomeric preference over the (E,E) pat-
tern adopted on coordination. The different nature and posi-
tion of substituents at imino units do not lead to important
differences in bond lengths involving the imino nitrogens
[N(1)ꢀC(14)=1.414(3),
N(2)ꢀC(20)=1.426(3),
N(1)ꢀC(2)=
1.275(3), N(2)ꢀC(3)=1.277(3) ꢁ], and these values are also
maintained within their estimated standard deviations upon
coordination in complex 1a. The dihedral angle between the
phenyl ring plane and the acenaphthene plane is 74.22(7)8 for
the (Z)-3,5-bis(trifluoromethyl)phenyl ring and 84.53(7)8 for the
(E)-2,6-dimethyl-phenyl ring.
Suitable crystals for complex 1a were obtained by using
a crystallization method based on slow diffusion of hexane
into a chloroform solution of the complex at 48C. The crystal
structures of the neutral complexes 2a[39] and 4a[40] with the
corresponding symmetric ligands, 2 and 4, have already been
reported, which thus allows a comparison of their relevant
structural features with those of 1a.
1
The H NMR spectrum, at room temperature, of a dichloro-
methane solution of 1 shows the signals of both (E,E) and (E,Z)
isomers in equimolar ratio; the composition of the isomeric
mixture varies with the solvent: (E,E)/(E,Z)=2:1 in CDCl3 solu-
tion. The two isomers are in equilibrium with slow rate on the
NMR timescale, as usually observed for these molecules.[30,31]
Ligand 1 was reacted with [Pd(CH3)Cl(cod)], which yielded
the corresponding neutral complex 1a (Scheme 2). The NMR
characterization in CD2Cl2 solution reveals the presence of ex-
clusively one species that, by a NOESY experiment, has been
identified as the isomer with the PdꢀCH3 moiety trans to the
PdꢀN bond of the meta-substituted ring. For the sake of clarity,
this species is designated as the trans isomer (Figure 4). Traces
The structural analysis of 1a evidences the presence of only
the trans isomer in the unit cell (Figure 5), in which the palladi-
um ion attains a distorted square planar geometry with a rela-
tively small bite angle (N1ꢀPdꢀN2 of 78.67(15)8).
In agreement with the trans influence of the PdꢀCH3 frag-
ment, in all the three complexes 1a, 2a, and 4a the PdꢀN
bond length trans to the PdꢀC bond is longer than the other
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1172

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
room temperature (Figure 6a). The chemical shift of the singlet
of the methyl groups of dmso (at 2.89 ppm) indicates its coor-
dination through the sulfur atom; other relevant signals are
Figure 5. ORTEP drawing (thermal ellipsoids at the 40% probability level) of
complex 1a·2CHCl3. Of the disordered F1-3 group, only fluorines at higher
occupancy are shown. Selected bond lengths (ꢁ) and angles (8): PdꢀC(1)
2.010(5), PdꢀCl(1) 2.2991(14), PdꢀN(1) 2.201(4), PdꢀN(2) 2.063(4), N(1)ꢀPdꢀ
N(2) 78.67(15), C(1)ꢀPdꢀCl(1) 89.37(18), C(1)ꢀPdꢀN(1) 172.4(2), C(1)ꢀPdꢀN(2)
94.0(2), Cl(1)ꢀPdꢀN(1) 98.05(11), Cl(1)ꢀPdꢀN(2) 175.28(11).
Figure 6. 1H NMR spectra in CD2Cl2 of a) 2c at T=298 K, b) 1c at T=233 K,
and c) 4c at T=223 K.
PdꢀN bond length. Within the standard deviation, the PdꢀN1
bond lengths in 1a and 2a are similar (2.201(4) and 2.203(5) ꢁ,
respectively) and shorter than that in 4a (2.227(4) ꢁ), which
suggests the lower coordinating capability of 4; this is in
agreement with the presence of electron-withdrawing groups
on both the aryl rings. The PdꢀN2 bond lengths in the three
complexes are in the range of 2.052(3)–2.088(6) ꢁ.
the two singlets at 2.34 and 2.29 ppm owing to the methyl
groups on the aryl rings and the singlet at 0.83 ppm assigned
to the PdꢀCH3 fragment. For complexes 1c and 4c, broad sig-
nals are observed at room temperature and they become
sharper with the decrease in temperature. In the case of 4c,
the decoalescence of signals due to methylic protons was ach-
ieved at 223 K, which proved the presence of two species in
the solution that differ in the coordination mode of dmso: the
singlet at 3.04 ppm is attributed to S-bonded dmso and the
singlet at 2.71 ppm is attributed to O-bonded dmso (Figur-
es 6c and S1). Two singlets are also observed for the PdꢀCH3
group at 0.48 and 0.67 ppm. The two species are in the ratio
3:2, and the isomer with S-bonded dmso is the major species.
For complex 1c, at the decoalescence temperature (233 K;
Figures 6b and S2) the signals indicate the presence of three
species in solution: one major and two minor species in 13:3:2
ratio.
The orientation of the substituted phenyl rings in the three
complexes is worth noting. The dihedral angles formed by 3,5-
bis(trifluoromethyl)phenyl with the BIAN mean plane on the
chlorido side are 64.5(1)8 in 1a and 53.3(1)8 in 4a, whereas the
corresponding values for the phenyl cis to the PdꢀCH3 frag-
ment are 80.1(1) and 78.3(2)8, respectively, which indicates
a greater conformational freedom in the former case owing to
a smaller bulk for the chlorido than for methyl. Contrarily, in
2a the measured dihedral angles for the two 2,6-dimethyl-
phenyl rings are very similar, 76.1(2) and 80.1(2)8, which thus
confirms that the substituents in the ortho positions of both
aryl rings hinder a tilt beyond a certain limit to avoid steric
clashes.
In the NOESY spectrum of 1c recorded at 233 K, cross peaks
due to both exchange processes and Overhauser effect are ob-
served (Figure S3). For the major species, the cross peak be-
tween the signal of the methyl group bonded to palladium (at
0.51 ppm) and the singlet of the methyl groups on the aryl
ring of the ligand (at 2.24 ppm) indicates that this species is
the trans isomer. The signal of the methyls of dmso at
3.07 ppm is indicative of dmso bonded to palladium through
sulfur. Thus, the major species is featured by S-bonded dmso
and the PdꢀCH3 fragment trans to the PdꢀN bond of the aryl
ring substituted with the CF3 groups (species i in Scheme 3).
The characterization of two minor species was achieved by
comparison with the spectra of 2c and 4c (Figure 6). The sig-
nals of 1c at 2.88 ppm (CH3 of dmso) and 2.29 ppm (aryl CH3)
have similar chemical shifts to those of 2c and are therefore
assigned to the species with the PdꢀCH3 group trans to the
The monocationic complex with acetonitrile, 1b, was ob-
tained by following the well-known method (Scheme 2).[41] The
synthesis of the dimethyl sulfoxide derivatives, 1c, 2c, and 4c,
was achieved by a slight modification of the methodology ap-
plied for the acetonitrile complexes (Scheme 2; see the Experi-
mental Section). Both the series of complexes were character-
1
ized by using H and 13C NMR spectroscopy in CD2Cl2.
1
The H NMR spectrum of 1b shows, at room temperature,
sharp signals, as observed for Pd–acetonitrile derivatives with
Ar-BIAN ligands[26]. The number of signals and their integration
indicate the prevalence of the trans isomer; for 1b the trans/
cis ratio is 10:1, whereas for 1a it is 19:1.
For the dimethyl sulfoxide derivatives, different NMR spectra
are obtained depending on the nature of the Ar-BIAN ligand.
In the 1H NMR spectrum of 2c, sharp signals are present at
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1173

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
gen atom with lower Lewis basicity. Contrarily, isomer iv
(Scheme 3), with dmso bonded through oxygen (hard base) to
Pd2+ (soft acid) and the coordinated methyl competing with
the trans located N-donor atom of higher Lewis basicity,
should be the least stable and was not detected. The stability
of the other two species, ii and iii, should lie in between that
of i and iv.
The observation that the trans isomer is the prevailing spe-
cies in all the cases is in agreement with the literature data,
which indicate a preferential tendency for the methyl group to
be coordinated trans to the lower basic N atom in Pd–methyl
complexes
of
electronically
unequivalent
N-donor
atoms.[41,43–46] The opposite situation is observed for palladium
complexes in which steric requirements overcome the elec-
tronically driven coordination.[47]
Scheme 3. Schematic representation of the four possible isomers for com-
plex 1c.
CH3-substituted ring (cis isomer) and S-bonded dmso (specie-
s iii in Scheme 3). The remaining singlets at 2.86 ppm (CH3 of
dmso) and 2.20 ppm (aryl CH3) are attributed to the trans
isomer with O-bonded dmso (species ii in Scheme 3). These at-
Ethylene/MA cooligomerization and mechanistic studies
Complexes 1b–5b, 1c, 2c, and 4c were tested as precatalysts
for ethylene/MA cooligomerization by performing the reaction
1
tributions are unambiguously confirmed by H–13C HSQC ex-
in 2,2,2-trifluoroethanol as solvent at T=308 K and PC H
=
4
2
periments (Figure S4). Compared to free dmso, the carbon
atoms resonate at a higher frequency if coordinated to a transi-
tion metal through sulfur and at a lower frequency if bonded
2.5 bar (1 bar=0.1 MPa) for 24 h (Scheme 4). The catalytic
1
through oxygen.[42] In the H–13C HSQC spectrum of 1c, both
the major and one of the two minor species have the cross
peak of dmso at higher frequency than that of free dmso
whereas for the other minor isomer the dmso cross peak falls
at a lower frequency, which thus confirms the previous
assignments.
In addition, the singlets due to the PdꢀCH3 groups in the
three isomers were assigned on the basis of the results of com-
parison of their 13C chemical shift with that of the related com-
plexes 2c and 4c with the symmetric Ar-BIANs.
Scheme 4. Ethylene/methyl acrylate cooligomerization.
product was a yellow-red oil, which was characterized by using
1H and 13C NMR spectroscopy. In addition, a small portion of
the reaction mixture, before the workup, was analyzed by
using GC–MS to determine whether higher alkenes were also
formed.
Precatalysts 3b–5b with symmetrically substituted mAr-
BIANs coordinated to palladium were found practically inac-
tive, which yielded an oil that contained palladium derivatives,
ethylene/MA cooligomers, and higher alkenes (Table 1, runs 1–
3). This inactivity is related to the fast decomposition of the
catalyst to inactive palladium metal that formed within the
first 4h of the reaction. Contrarily, precatalyst 1b with the non-
symmetric ligand 1 was found more productive than 2b,
Finally, as demonstrated by the cross peaks caused by ex-
change processes present in the NOESY spectrum, the three
isomers of 1c are in equilibrium and at room temperature the
rate of this exchange process is comparable to the NMR time-
scale. In the 1H NMR spectrum of 1c recorded in [D6]DMSO,
sharp signals are already present at 298 K and indicate the
presence of only one species (Figure S5). The chemical shift of
the PdꢀCH3 signal indicates that this species is species i
(Scheme 3). This experiment indicates that the dynamic pro-
cess that exchanges species i, ii, and iii occurs through an as-
sociative mechanism.
The different coordination mode of dmso was also con-
firmed in the solid state by the stretching frequency of the Sꢀ which is considered as the reference compound with the sym-
O bond in the IR spectra: one band typical of S-bonded dmso
is observed in the IR spectrum of 2c, whereas two bands at
the bands of S- and O-bonded dmso, respectively, are present
in the IR spectra of 1c and 4c.
Thus, for complex 1c, only three of the four possible isomers
are observed (Scheme 3).
metric Ar-BIAN substituted on the ortho positions; the ob-
tained catalytic product contains a slightly higher amount of
polar monomer (Table 1, runs 4 and 5). An increase in produc-
tivity was achieved with the Pd–dmso derivatives; on average,
the productivity of 1b and 1c was two times higher than that
of 2b and 2c, respectively (Table 1, runs 4–8). A slight decrease
in the content of the polar monomer incorporated into the
catalytic product was found with use of the Pd–dmso deriva-
tives instead of the Pd–acetonitrile compounds. However, at
least in some cases, this decrease can be attributed to the
By simple considerations about coordination chemistry, the
major isomer (species i in Scheme 3) should be the most
stable: dmso is bonded through sulfur (soft base) to Pd2+ (soft
acid); the PdꢀCH3 bond is trans to the PdꢀN bond of the nitro-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1174

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
A catalytic experiment was performed by using a 1:1 mixture
of precatalysts 2b and 4b, which yielded a productivity much
lower than that obtained with 1b; this indicates that no scram-
bling process occurs during catalysis with the latter (Table 1,
run 9). The value of productivity, the acrylate content, and the
lack of higher alkenes in the product indicate that active spe-
cies is originated by 2b.
By applying precatalysts with ligand 1, the effect of reaction
parameters, such as temperature, ethylene pressure, and reac-
tion time, was investigated. Productivity increases on increas-
ing the temperature from 298 to 308 K for 1c, whereas it de-
creases slightly for 1b. For both precatalysts, an increase in
temperature up to 318 K results in a sharp decrease in produc-
tivity owing to the decomposition of the catalysts to inactive
palladium metal, which indicates their low thermal stability
(Figure 7).
Table 1. Ethylene/methyl acrylate cooligomerization: effect of the Ar-
BIAN ligand.[a]
Run Precatalyst
L
Yield
[mg]
g P/g Pd[b] Mol% MA[c] Bd[c,d] Alkenes
1
2
3
4
5
6
7
8
3b
4b
5b
2b
1b
2c
4c
1c
CH3CN 56.7[e]
CH3CN 37.1[e]
CH3CN 19.3[e]
CH3CN 171.4
CH3CN 297.0 133.2
dmso 221.8 99.5
dmso
55.0[e]
dmso 520.8 233.5
CH3CN 73.6 33.0
C4–C6
C4–C6
C4–C6
79.4
10.4
14.7
6.6
101
–
137 C4–C16
117
–
C4–C6
12.5
8.1
146 C4–C16
119
9[f] 2b+4b
–
[a] Precatalyst: [Pd(CH3)(L)(Ar-BIAN)][PF6]. Reaction conditions: nPd =2.1ꢂ
10ꢀ5 mol, 2,2,2-trifluoroethanol: V=21 mL, T=308 K, PC H =2.5 bar;
2
4
methyl acrylate (MA) V=1.13 mL, [MA]/[Pd]=616, t=24 h; [b] Isolated
yield; g P/g Pd=grams of product per gram of palladium; [c] Calculated
1
by using H NMR spectroscopy (see the Supporting Information); [d] Bd=
degree of branching as branches per 1000 carbon atoms; C(O)OCH3
carbon atoms are excluded; [e] Mixture of palladium derivatives, ethyl-
ene/methyl acrylate cooligomers and higher alkenes; [f] n2b =n4b =1.05ꢂ
10ꢀ5 mol.
higher MA conversion in the reaction of the former catalysts
than in the reaction of the latter ones, as discussed later in
more detail.
In agreement with Brookhart’s system,[10,11] for precatalysts
2b and 2c, no higher alkenes were formed whereas in all
other cases butenes and hexenes were also obtained, and for
complexes 1b and 1c the produced alkenes ranged from C4
to C12 with traces of C14 and C16 (Table 1, Figures S6–S8). The
formed alkenes are a complex mixture of isomers as the result
of the chain walking mechanism.[8] The observed chain length
distribution, expressed in terms of molar fraction of each
alkene, can be reasonably fitted by the Poisson distribution
(see the Supporting Information). With precatalysts 1b and 1c,
the product with the higher concentration is C6 whereas the
product with the higher weight fraction is C8.
Figure 7. Ethylene/methyl acrylate cooligomerization: effect of temperature.
Precatalyst: [Pd(CH3)(L)(1)][PF6] (L=CH3CN, dmso). Reaction conditions: see
Table 1. Filled symbols: productivity data; open symbols: mol% MA:
^ ^
* *
, ).
1b (
,
); 1c (
The observed formation of higher alkenes with the symmet-
rically substituted mAr-BIANs is due to the lack of steric hin-
drance on the ortho positions of the aryl rings that does not
avoid the b-hydrogen elimination.[8,48] In the case of catalysts
with ligand 1 with one aryl ring substituted on both ortho po-
sitions and the other ring substituted on both meta positions,
b-hydrogen elimination is slowed down slightly and longer al-
kenes were produced. These data highlight the subtle role
played by the position of the substituents on the ancillary
ligand in regulating the length of the synthesized macro-
molecules.
An increase in the content of the incorporated polar mono-
mer into the catalytic product is achieved by increasing the
temperature for both catalysts. Even the production of alkenes
shows a progressive decrease, more pronounced for 1c,
moving from 298 to 318 K (Figure S6).
The effect of ethylene pressure was studied in the range
1.5–7.0 bar (Figure 8).
For both precatalysts, productivity increases on increasing
ethylene pressure up to 5.0 bar; however, an increase in pres-
sure up to 7.0 bar results in a less pronounced increase in pro-
ductivity. In the case of 1c, the value of 400 g P/g Pd is
reached at 7.0 bar. As expected,[10,11] the increase in ethylene
pressure results in a progressive decrease in the content of the
polar monomer incorporated in the catalytic product
(Figure 8). The ethylene pressure significantly affects the pro-
duction of higher alkenes (Figure S7). Their amount increases
continuously from 1.5 to 7.0 bar, with a slight shift to higher al-
kenes in the Poisson distribution.
As far as the effect of the nature of the Ar-BIAN ligand on
the productivity is concerned, which is in agreement with the
data reported for the CO/vinyl arene copolymerization,[26] even
for the currently investigated reaction, the catalyst with the
nonsymmetric ligand bonded to the metal center is more pro-
ductive than those with the corresponding symmetric Ar-
BIANs for both the Pd–NCCH3 and Pd–dmso derivatives
(Table 1, 1b vs. 2b and 4b, 1c vs. 2c and 4c).
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1175

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
to CH3CN, but also due to additional reasons (described
below). As also shown in Figure 9, the percentage of MA incor-
porated in the oligomer decreases as the reaction proceeds.
This can also be explained, at least qualitatively, on the basis
of the reaction kinetics (described below). An unusual effect of
the reaction time was observed on the production of higher al-
kenes (Figure S8). A progressive increase in their amounts, to-
gether with an increase in the average chain length k, was ob-
served up to 16 h. When the reaction was stopped after 24 h,
the concentration of alkenes decreased drastically; however,
when the reaction was continued up to 48 h, it became higher
again, but with low k values. This peculiar behavior could be
rationalized by considering the main processes operative in
solutions. The ethylene oligomerization results in the increase
in concentration and in the lengthening of the alkene chains
(k values). At the same time, the reaction with MA results in
the consumption of alkenes and in the formation of cooligom-
ers. As the concentration of alkenes increases, the possible in-
sertion of higher alkenes instead of ethylene cannot be exclud-
ed. At the lower reaction times, the predominant processes are
ethylene oligomerization and ethylene/MA cooligomerization.
As the reaction proceeds, formed alkenes participate in the
cooligomerization process, which results in the decrease in
their concentration. With the increase in the reaction time, MA
concentration decreases and, below a critical value, ethylene
oligomerization becomes the predominant process, which pro-
duces a higher amount of higher alkenes.
Figure 8. Ethylene/methyl acrylate cooligomerization: effect of pressure. Pre-
catalyst: [Pd(CH3)(L)(1)][PF6]. Reaction conditions: see Table 1. Filled symbols:
^ ^
* *
, ).
productivity data; open symbols: mol% MA: 1b (
,
); 1c (
A different catalytic behavior of the two precatalysts 1c and
1b is observed by evaluating the effect of the reaction time by
conducting the catalytic experiments at PC H =2.5 bar and T=
2
4
308 K (Figure 9).
The characterization of the isolated catalytic product was
1
performed by using H and 13C NMR spectroscopy in CDCl3 so-
lution at room temperature. The signals in the spectra were
identified by comparison with literature data[8,12,14,49–52] and by
multidimensional homo- and heteronuclear NMR experiments
1
(Figures S9–S12). In the H NMR spectra of the oils synthesized
with precatalysts 1b and 1c, the following signals are recog-
nized: the signals at 5.60–4.90 ppm assigned to the vinylic pro-
tons, the singlet at 3.60 ppm assigned to the methoxy group,
the broad signals between 2.50 and 1.50 ppm assigned to allyl-
ic protons and to methylenic groups close to MA units, the
broad signal centered at 1.23 ppm assigned to the methylenic
and methynic moieties of ethylene units, and the signals be-
tween 1.0 and 0.6 ppm assigned to methyl groups at the end
of branches (Figure 10). No signal due to methynic carbon
atoms of MA units was observed.
Figure 9. Ethylene/methyl acrylate cooligomerization: effect of reaction
time. Precatalyst: [Pd(CH3)(L)(1)][PF6]. Reaction conditions: see Table 1. Filled
The NMR analysis indicates that both ethylene oligomers
and ethylene/MA cooligomers are highly branched
(
^ ^
* *
, ).
symbols: productivity data; open symbols: mol% MA: 1b (
,
); 1c (
ꢁ100 branches per 1000 carbon atoms; Table 1) and the polar
monomer is inserted at the end of the branches as the ꢀCH2ꢀ
CH2ꢀC(O)OCH3 fragment, which is the result of the chain walk-
ing mechanism; this is in agreement with Brookhart’s
system.[10,11] From the NMR analysis, the presence of higher al-
kenes in the isolated catalytic product cannot be excluded.
To gain some insights into the reactions involved in the cata-
lytic cycle, the independent reactivity of the two comonomers
with complexes 1b, 2b, and 4b was investigated by using
in situ NMR spectroscopy. A 10 mm CD2Cl2 solution of 1b was
reacted with ethylene at room temperature. In the 1H NMR
spectrum recorded after 10 min of the bubbling of the apolar
Although precatalyst 1b, with acetonitrile as the labile
ligand, is deactivated within 16 h of the reaction, precatalyst
1c is active for at least 48 h, which reaches a productivity of
approximately 350 g P/g Pd. Only a partial decrease in produc-
tivity is apparent at a long reaction time, which can be partly
explained by the kinetics of the reaction, as detailed in the fol-
lowing. The longer catalyst lifetime for 1c indicates that the
better performing catalytic behavior of the precatalyst with di-
methyl sulfoxide than that with acetonitrile could be not only
due to the lower coordinating capability of dmso with respect
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1176

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Figure 10. 1H NMR spectrum in CDCl3 at 298 K of the ethylene/methyl acry-
late cooligomers obtained with 1c (analogous spectrum for ethylene/methyl
acrylate cooligomers obtained with 1b).
monomer, no signal of the precursor was present and the sig-
nals of butenes and hexenes were observed. The fast reactivity
of 1b with alkene is in agreement with Brookhart’s system.[11,48]
The reactivity with MA was investigated by adding 2 equiv
of the polar monomer to the 10 mm CD2Cl2 solution of the
1
complex at room temperature. The H NMR spectra were re-
corded every 5 min for the first 20 min after the addition of
the polar monomer. The three complexes showed the same
behavior: in the first spectrum recorded after the addition of
MA, the signals of the precatalyst were present together with
the new signals assigned to the five- and six-membered palla-
dacycles B“ and C” (Scheme 5; Figures S13–S15). In addition,
the signal of free acetonitrile was present and the resonances
of A’ were observed at longer reaction times. The reactivity of
the three complexes differs with the rate of the reaction,
which decreases in the order 4b>2b>1b (Figures 11 and
S16). The different trend of reactivity with respect to that of
productivity is related to the different stabilities of the three
catalysts: 4b is the fastest but least stable, and 1b is the slow-
est but very stable.
Scheme 5. Proposed reaction mechanism for the reactivity of complex 1b
with methyl acrylate.
The rate of the reaction of the precatalyst with MA follows
a first-order kinetic. The same is likely to hold for the related
insertion of MA into the growing oligomer chain. Because
during the catalytic reactions ethylene pressure is kept con-
stant, the apparent rate dependence on ethylene pressure will
be pseudo zero order for any single reaction. (For ethylene
polymerization catalyzed by Brookhart’s system, a zero-order
rate of ethylene disappearance was modeled.)[8] We plan to
perform a series of kinetic experiments to prove or disprove
these assumptions, and those will be published in a future
paper. However, we note that these reasonable assumptions
allow us to justify some of the trends evidenced previously.
1. The productivity of the catalyst decreases with reaction time
even in the case of the Pd–dmso complex 1c. From the amount
of the product obtained and the percentage of MA measured
by using NMR spectroscopy, it is easy to calculate the moles of
MA and ethylene consumed for each reaction (see Supporting
Information for the details). This allows us to treat the MA and
Figure 11. First-order rate for the consumption of PdꢀCH3 by insertion of
methyl acrylate ([Pd]=10 mm in CD2Cl2, T=298 K, A=[PdꢀCH3]t,
A0 =[PdꢀCH3]t=0).
ethylene consumption separately, which is in agreement with
the presence of two processes: ethylene oligomerization and
ethylene/MA cooligomerization. The data for the reactions cat-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1177

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
alyzed by 1c can be fitted quite well by a kinetic model, which
is first order in MA and pseudo zero order in ethylene. A small
decrease in activity can be observed, which must be attributed
to some catalyst deactivation; however, most of the decrease
in productivity at long reaction times can be attributed to the
slowing down of the MA insertion owing to the consumption
of the MA. Contrarily, in the case of the reactions catalyzed by
1b, the model fails completely. In this case, catalyst deactiva-
tion is clearly the main reason for the decrease in productivity
with time.
ed to the two metallacycles B“ and C” and the appearance of
the resonances of methyl crotonate (A’) and of D’ (Figure S17).
Methyl crotonate is the result of b-hydrogen elimination on
the five-membered palladacycle B“ (Scheme 5). C” and B“ are
in equilibrium as probed by the cross peaks due to exchange
processes present in the NOESY spectrum (Figure S18, cross
peak in green circle). D’ is the result of the migratory insertion
of MA into the PdꢀH bond. This intermediate was also ob-
served in Brookhart’s system.[10]
Notably, in the case of the intermediates with the nonsym-
metric ligand 1, cis and trans isomers are possible. As reported
above, the trans isomer is featured by the PdꢀC bond trans to
2. The percentage of MA in the product decreases as the reac-
tion proceeds. The insertion reactions of MA and ethylene com-
pete with each other. To a first approximation, it can be as-
sumed that the relative insertion rate is proportional to the
MA/ethylene ratio in solution. Because the concentration of
MA decreases during the reaction but that of ethylene does
not, it is to be expected that the percentage of inserted MA
will decrease at higher MA conversions, such as those that
occur if the reaction time is increased, at least for the Pd–
dmso catalysts. Although we tried to model quantitatively the
product composition on the basis of this principle, the limited
amount of data makes the conclusions unreliable. However, on
a semiquantitative basis, we can confirm that the product
composition roughly follows the expected trend.
1
the PdꢀN bond of the meta-substituted ring. In the H NMR
spectra, the signals of only one isomer are evident for C“ (Fig-
ure S13; the concentrations of A’ and B” are too low to allow
the detection of their other possible isomers). Their assignment
was based on the following criteria: (1) The intensity of the sin-
glet of the aryl CH3 of 1b decreases as the reaction proceeds,
and this singlet is replaced by a new singlet at a higher fre-
quency (2.32 ppm; Figure S13). (2) The signal of H3 on the ace-
naphthene skeleton is also shifted at a higher frequency
(7.24 ppm). (3) The signals of H14,16,18 overlap, whereas they are
separated in the precursor. (4) A clear cross peak between the
a
aryl CH3 and the PdꢀCH2 of C“ is present in the NOESY spec-
3. The rate does not increase linearly with ethylene pressure,
and the percentage of MA insertion decreases along the series.
The reason for this is immediately clear from what is said
above. An increase in ethylene pressure increases only the rate
of ethylene insertion. It does not increase the rate of MA inser-
tion. Because the productivity of the catalyst is due to the sum
of two insertions, a linear increase in productivity with ethyl-
ene pressure is not to be expected under any circumstance.
Contrarily, a higher ethylene pressure increases the ethylene/
MA ratio in solution and makes the MA insertion less
competitive.
trum (Figure S18, cross peak in black circle). (5) No variation is
observed in the chemical shift of H10. All these NMR data indi-
cate that the trans isomer (C”trans) is the only one present,
which thus suggests that after the migratory insertion of MA,
the isomerization of the growing chain occurs.
During catalysis, in the growing of the cooligomer chain the
metallacycle C“trans could also be responsible for the formation
of the ꢀCH2ꢀCH2ꢀC(O)OCH3 fragment. In a preliminary in situ
NMR experiment, complex 1c, with dmso bonded to palladi-
1
um, is reacted with 2 equiv of MA. In the H NMR spectrum re-
corded after 2 min of the addition of MA, no signal of 1c is
present and the signals of B” and C“trans are observed together
with a broad peak at 2.63 ppm; no signal of free dmso is evi-
dent. Following the reaction with time, the signal at 2.63 ppm
increases in intensity and becomes sharper. In addition, the
typical signals of the metallacycles B” and C“trans fall at a fre-
quency slightly different from that of the same signals ob-
served for the reactivity of 1b (Figure S19). These signals are
tentatively assigned to an open intermediate analogous to
C”trans with dmso bonded to palladium instead of the oxygen
atom of the inserted MA unit. In a paper by Mecking on Pd–
phosphine sulfonate complexes,[28] the effect of coordinating
ligand L, such as dmso and various phosphine oxides, on the
ethylene/MA copolymerization activity and on the rate of mo-
nomer insertion has been investigated. It has been demon-
strated that the coordination strength of dmso exceeds that of
the oxygen atom of the carbonyl group of the polar monomer
in a four-membered intermediate analogous to A“. The open
intermediate resulting from the insertion of MA into the Pdꢀ
CH3 bond and with coordinated dmso, [Pd{CH-
(COOCH3)CH2CH3}(dmso)(PꢀO)], that corresponds to the open
form of A” was isolated from the reaction of MA with [Pd(CH3)-
(dmso)(PꢀO)].[14] In the Pdꢀ(PꢀO) catalytic system, no chain
The NMR studies described above also allow us to propose
a more detailed hypothesis for the mechanism of the reactivity
with MA (Scheme 5). The substitution reaction of acetonitrile
with the polar monomer takes place on the precatalyst, which
leads to free acetonitrile and two intermediates that differ in
MA orientation. On these intermediates, the migratory inser-
tion of the polar monomer into the PdꢀCH3 bond takes place
with primary and secondary regiochemistry, respectively; the
latter is remarkably favored. The insertion with primary regio-
chemistry leads directly to the five-membered palladacycle A’,
whereas in the case of the insertion with secondary regio-
chemistry, the detected six-membered metallacycle C“ is the
result of the chain walking mechanism. The insertion of the
polar monomer into the PdꢀCH3 bond with secondary regio-
chemistry is typical for MA, which is observed in palladium
complexes with a-diimines,[10] phosphine sulfonate derivatives,
phosphino acetamido-derived ligands,[53] and (phosphinome-
thyl)oxazoline PꢀN ligands.[54] The insertion with exclusively pri-
mary regiochemistry was instead observed for methyl metha-
crylate on Pd–a-diimine complexes.[55]
For complex 1b, the reactivity with MA was followed up to
17 h, which indicated the disappearance of the signals attribut-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1178

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
walking mechanism occurs and, thus, dmso competes favora-
bly with the oxygen atom of the four-membered palladacycle
analogous to A“ whereas in the Pd–a-diimine system, dmso
has to compete with the oxygen atom of the six-membered
metallacycle C”, which is more difficult to open than A“.
Our preliminary investigation indicates that the different cat-
alytic behavior of Pd–NCCH3 and Pd–dmso derivatives could
be attributed to the capability of dmso to remain coordinated
to palladium in the apical position of a pentacoordinated inter-
mediate, which thus competes with the oxygen atom of the
last inserted MA unit and favors the opening of the metallacy-
cle C“trans—the resting state of the catalytic cycle—and the in-
sertion of the next incoming monomer. The DFT calculation
showed that in the palladium-catalyzed carbon monoxide/eth-
ylene copolymerization, carbon monoxide and ethylene ex-
change through an associative mechanism via a trigonal bipyr-
amidal intermediate.[56] Further investigations are in progress.
could be related to the possibility of dimethyl sulfoxide com-
peting with the oxygen atom of the last inserted MA unit,
which favors the cleavage of the PdꢀO bond and the insertion
of the new incoming monomer.
Further investigations aiming at evaluating the capability of
dmso to stabilize the catalytically active species are currently
in progress.
Experimental Section
General considerations
All complex manipulations were performed by using standard
Schlenk techniques under argon. Anhydrous dichloromethane was
obtained by distilling it over CaH2 and under argon. Before use,
ZnCl2 and basic Al2O3 were stored in an oven at 1108C overnight.
Ligands 2–4,[30,57] 5,[26] [Pd(CH3)Cl(cod)], their neutral palladium
complexes, and acetonitrile derivatives[41] were synthesized accord-
ing to methods given in the literature. [Pd(OAc)2] was donated by
Engelhard Italia and used as received. Ethylene (purityꢂ99.9%)
supplied by SIAD and MA (99.9%, with 0.02% of hydroquinone
monomethyl ether) supplied by Aldrich were used as received.
Deuterated solvents were stored as recommended by Cambridge
Isotope Laboratories. All the other reagents and solvents were pur-
chased from Sigma–Aldrich and used without further purification
for synthetic, spectroscopic, and catalytic purposes.
Conclusions
The synthesis and characterization of a peculiar nonsymmetri-
cally substituted bis(aryl-imino)acenaphthene (Ar-BIAN) ligand
is reported. This ligand and the corresponding symmetrically
substituted derivatives are used to synthesize the relevant neu-
tral and monocationic Pd–methyl complexes. The series of the
monocationic complexes has been extended to the dimethyl
sulfoxide derivatives, and this is the first report on Pd–dmso
compounds with a-diimines as ancillary ligands. The characteri-
zation of the Pd–dmso compounds probes the presence of dif-
ferent species in solution depending on Ar-BIAN: for ligand 2,
only the compound with S-bonded dmso is observed; for
ligand 4, both isomers with S- and O-bonded dmso are pres-
ent; for ligand 1, three species are observed that differ in the
coordination mode of dmso and the relative position of Pdꢀ
CH3 with respect to the two halves of Ar-BIAN.
The NMR spectra of ligands, complexes, and catalytic products and
of the in situ reactivity investigations were recorded by using
a Varian 500 MHz spectrometer at the following frequencies:
500 MHz (1H) and 126 MHz (13C); the resonances (d) were reported
in ppm and referenced to the residual solvent peak versus Si(CH3)4:
CDCl3 at d 7.26 (1H) and d 77.0 (13C), CD2Cl2 at d 5.32 (1H) and d
54.0 (13C), and [D6]DMSO at d 2.50 (1H) and d 39.5 (13C). NMR ex-
periments were performed by using the automatic software pa-
rameters; in the case of NOESY and ROESY experiments, a mixing
time of 500 and 250 ms was used, respectively. In situ kinetic inves-
tigations were performed by using the pre-acquisition delay (pad)
Varian experiment.
Both the Pd–NCCH3 and Pd–dmso complexes with the new
nonsymmetric Ar-BIAN ligand are tested as precatalysts in the
ethylene/methyl acrylate cooligomerization with regard to the
catalytic behavior of the complexes with the corresponding
symmetric Ar-BIANs under mild reaction conditions of temper-
ature and ethylene pressure. The catalytic product is a mixture
of ethylene/methyl acrylate cooligomers and higher alkenes.
The catalysts containing the nonsymmetric Ar-BIAN ligand are
found to be more productive than those containing the sym-
metric counterparts, yielding the cooligomer with a higher
content of the polar monomer, which thus supports the idea
that a subtle unbalance of the steric and electronic properties
of the ancillary ligand has a positive effect on the catalytic
performances.
IR spectra were recorded in Nujol by using a Perkin–Elmer System
2000 FT-IR spectrometer. Elemental analyses were performed in the
analytical laboratories of the University of Udine. The mass spectra
of ligand 1 and complexes were run by ESI ion trap by using
a Bruker Esquire 4000. GC–MS analyses were performed with an
Agilent 7890 GC using a DB-225 ms column (J&W: 60 m, 0.25 mm
inner diameter, 0.25 mm film) and helium as carrier coupled with
a 5975 Series MSD. Before analysis, samples were diluted with
methanol and nonane was added as the internal standard.
Synthesis of ligand 1
Ligand 1 was synthesized through the transimination reaction by
using a method slightly different from that given in the litera-
ture.[29] [ZnCl2(4)] (1.00 g, 1.35 mmol) was dissolved in CH3OH
(100 mL). To the orange solution 2,6-dimethyl aniline (366 mL,
2.97 mmol) was added, and the reaction mixture was stirred at RT
for at least 7 days. The reaction was followed by TLC (alumina,
hexane/diethyl ether=2:1), and a final check was performed by
using 1H NMR spectroscopy. To do so, an aliquot of the solution
was withdrawn and ligand decoordination was performed on it
(see later). The reaction did not proceed over 50% of conversion.
When it was reached, the yellow precipitate of [ZnCl2(2)] and
For all the tested ligands, the catalysts obtained from the
Pd–dmso derivatives were found to be more productive than
those obtained from the Pd–NCCH3 derivatives. The higher
productivity of the Pd–dmso catalysts was observed in both
the isolated product and the higher alkenes production. The
Pd–dmso catalyst with ligand 1 showed a longer lifetime than
the corresponding Pd–NCCH3 derivative. In situ NMR investiga-
tions indicate that the higher stability of the Pd–dmso catalyst
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1179

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[ZnCl2(4)] was filtered off and the solution was concentrated to
half volume and poured into a separating funnel together with
CH2Cl2 (30 mL) and a water solution supersaturated with Na2C2O4
(10 mL). The two phases were shaken for approximately 1 min
until the organic phase color changed from yellow-orange to red-
orange. The water phase was washed with CH2Cl2 (10 mL), and the
two organic phases were combined, washed with water (2ꢂ
10 mL), dried over Na2SO4, and concentrated at reduced pressure
to yield a red oil. The oil contained 2,6-dimethyl aniline, 3,5-bis(tri-
fluoromethyl) aniline, and minor byproducts (e.g., monoketoimine),
and it was purified by using column chromatography over basic
alumina with hexane/diethyl ether (2:1) as eluent. Ligand 1 eluted
first as an orange band (316.3 mg, 0.64 mmol).
Upon addition of petroleum ether (bp: 40–608C), a red solid was
obtained (242.6 mg, 0.37 mmol).
Quantification of 1 in the crude oil: In addition to 1, 2,6-dimethyl
aniline (ortho) and 3,5-bis(trifluoromethyl) aniline (meta) were pres-
ent in the crude oil. The following system of equations [Eqs. (1)–
(3)] was used to quantify the amount of these compounds in the
mixture:
Oil weight ¼ northoðFWorthoÞ þ nmetaðFWmetaÞ þ n1ðFW1Þ
ð1Þ
ð2Þ
ð3Þ
northo ¼ an1
nmeta ¼ bn1
For which a and b are the ratios between the integrals of both ani-
line peaks and the integrals of the 1 peak divided by the correct
amount of protons. The following peaks in the H NMR spectrum
[3,5-(CF3)2C6H3],[2,6-(CH3)2C6H3]BIAN (1)
1
Yield: 47%; MS (ESI): m/z (%): 519.2 (100) [M+Na]+, 497.2 (80)
[M+H]+; found: C 66.94, H 3.45, N 5.24; elemental analysis calcd
(%) for C28H18N2F6: C 67.74, H 3.65, N 5.64; isomeric ratio (E,E)/(E,Z):
CDCl3, 298 K: 2/1; CD2Cl2, 298 K: 1/1.
of the crude oil recorded in CD2Cl2 were chosen for the integra-
tion: 7.06 (3,5-bis(trifluoromethyl) aniline), 2.15 (2,6-dimethyl ani-
line), and 2.10 ppm (1).
1H NMR (500 MHz, CDCl3, 298 K): Major isomer (E,E): d=7.98 (d, J=
3
[Pd(CH3)Cl(1)] (1a)
3
8.3 Hz, 1H; H5), 7.94 (d, J=8.3 Hz, 1H; H8), 7.79 (s, 1H; H16), 7.65
(s, 2H; H14,18), 7.47–7.41 (m, 2H; H4,9), 7.19–7.15 (m, 2H; H15’,17’),
Yield: 89.5%. MS (ESI): m/z (%): 654.0 [M+H]+; found: C 53.82, H
3.41, N 4.41; elemental analysis calcd (%) for PdC29H21N2ClF6: C
53.31, H 3.24, N 4.29; isomeric ratio trans/cis: CD2Cl2, 298 K: 1/19.
7.13–7.07 (m, 1H; H16’), 6.84 (d, 3J=7.2, 1H; H3), 6.73 (d, 3J=7.2,
3
1H; H10), 2.13 ppm (s, 3H; CH3). Minor isomer (E,Z): d=8.16 (d, J=
3
3
7.1, 1H; H3), 8.07 (d, J=8.2, 1H; H5), 7.96 (d, J=8.2, 1H; H8), 7.82
1H NMR (500 MHz, CD2Cl2, 298 K): Major isomer (trans): d=8.16 (dd,
(dd, J=8.2, 7.1, 1H; H4), 7.53 (s, 1H; H16), 7.45 (s, 2H; H14,18), 7.35
3
3J=8.3, 2H; H5,8), 7.98 (s, 3H; H14,16,18), 7.57 (t, J=7.8, 1H; H4), 7.49
3
(dd, J=8.2, 7.1, 1H; H9), 7.07–7.02 (m, 1H; H16’), 7.01–6.94 (m, 2H;
3
3
3
(t, J=7.8, 1H; H9), 7.39–7.29 (m, 3H; H15’,16’,17’), 7.22 (d, J=7.3, 1H;
H3), 6.57 (d, 3J=7.3, 1H; H10), 2.30 (s, 6H; ArꢀCH3), 0.72 ppm (s, 3H;
PdꢀCH3). Minor isomer (cis): d=8.08 (s, 1H; H16), 7.88 (s, 2H; H14,18),
7.26 (m, 3H; H15’,16’,17’), 6.79 (d, J=7.8, 1H; H10), 6.65 (d, J=7.8, 1H;
H3), 2.33 (s, 6H; ArꢀCH3), 0.68 ppm (s, 3H; PdꢀCH3).
H
15’,17’), 6.53 (d, 3J=7.1, 1H; H10), 1.82 ppm (s, 3H; CH3).
1H NMR (500 MHz, CD2Cl2, 298 K): (E,E) isomer: d=7.99 (d, 3J=
3
8.3 Hz, 1H; H5), 7.96 (d, J=8.3 Hz, 1H; H8), 7.87–7.80 (m, 1H; H16),
7.66 (s, 2H; H14,18), 7.43 (m, 2H; H4,9), 7.19 (d, 3J=7.4 Hz, 2H;
H15’,17’), 7.10 (t, 3J=7.4 Hz, 1H; H16’), 6.81 (d, 3J=7.3 Hz, 1H; H3),
6.72 (d, 3J=7.3 Hz, 1H; H10), 2.10 ppm (s, 3H; CH3). (E,Z) isomer:
13C NMR (126 MHz, CD2Cl2, 298 K): Major isomer (trans): d=132.2
(C5 or C8), 132.0 (C5 or C8), 129.7 (C9), 129.3 (C16’), 128.9 (C4), 127.9
(C15’,17’), 124.9 (C3), 124.7 (C10), 123.5 (C16), 121.4 (C14,18), 18.2 (Arꢀ
CH3), 1.8 ppm (PdꢀCH3).
3
3
d=8.16 (d, J=7.0 Hz, 1H; H3), 8.10 (d, J=8.3 Hz, 1H; H5), 7.99 (d,
3J=8.3 Hz, 1H; H8), 7.87–7.80 (m, 1H; H4), 7.56 (s, 1H; H16), 7.48 (s,
2H; H14,18), 7.37 (t, J=7.8 Hz, 1H; H9), 7.04 (m, 2H; H15’,17’), 6.98 (m,
3
1H; H16’), 6.55 (d, J=7.1 Hz, 1H; H10), 1.81 ppm (s, 3H; CH3).
3
13C NMR (126 MHz, CD2Cl2, 298 K): d=130.0 (C5,8E,Z), 129.8 (C5,8E,E),
Synthesis of [Pd(CH3)(NCCH3)(1)][PF6] (1b)
or
9
129.1 (C4E,Z), 129.0 (C9 and C15’,17’E,E), 128.7 (C15’,17’ and C4 E,E),
E,Z
E,Z
128.2 (C4
E,E), 124.5 (C16’ and C16’E,E), 124.1 (C3E,E), 123.4 (C10E,E),
123.3 (C10E,Z), 120.9 (C3E,Z), 119.6 (C14,18E,E), 119.0 (C14,18E,Z), 118.3
(C16E,E), 116.9 (C16E,Z), 18.2 (CH3 E,E), 17.8 ppm (CH3 E,Z).
or
9
A solution of AgPF6 (101.5 mg, 0.401 mmol) in CH3CN (1 mL) was
added to a solution of complex 1a (228.2 mg, 0.349 mmol) in
CH2Cl2 (2 mL). The solution was stirred at RT for 30 min, and AgCl
was filtered over celite; the solution was then concentrated at half
volume under vacuum. Upon addition of diethyl ether, the product
precipitated as a yellow solid (249.6 mg, 0.310 mmol).
E,Z
Synthesis of [Pd(CH3)Cl(1)] (1a)
[Pd(CH3)Cl(cod)] (201.5 mg, 0.76 mmol) was kept in a Schlenk flask
and dissolved in CH2Cl2 (3 mL). A solution of ligand 1 (417.0 mg,
0.84 mmol) in CH2Cl2 (3 mL) was added. The reaction mixture was
protected from light, stirred at RT for 45 min, and then concentrat-
ed at half volume under reduced pressure. Upon addition of petro-
leum ether (bp: 40–608C), a red solid was obtained (444.9 mg,
0.68 mmol).
[Pd(CH3)(NCCH3)(1)][PF6] (1b)
Yield: 88.8%; found: C 46.77, H 3.24, N 5.19; elemental analysis
calcd (%) for PdC31H24N3PF12: C 46.32, H 3.01, N 5.23; isomeric ratio
trans/cis: CD2Cl2, 298 K: 1/10.
1H NMR (500 MHz, CD2Cl2, 298 K): Major isomer (trans): d=8.27 (d,
J=8.8, 2H; H5), 8.25 (d, J=8.8, 2H; H8), 8.11 (s, 1H; H16), 8.04 (s,
2H; H14,18), 7.66 (t, J=7.9, 1H; H4), 7.57 (t, J=7.9, 1H; H9), 7.41 (m,
1H; H16’), 7.38–7.31 (m, 3H; H14’,18’,3), 6.56 (d, J=7.3, 1H; H10), 2.28
(s, 6H; ArꢀCH3), 2.19 (s, 3H; PdꢀNCCH3), 0.85 ppm (s, 3H; PdꢀCH3).
Minor isomer (cis): 8.15 (s, 1H; H16), 7.89 (s, 2H; H14,18), 7.01 (d, J=
7.2, 1H; H10), 6.66 (d, J=7.6, 1H; H3), 2.39 (s, 6H; ArꢀCH3),
0.79 ppm (s, 3H, PdꢀCH3).
Alternative synthetic method: [Pd(CH3)Cl(cod)] (119.9 mg,
0.45 mmol) was kept in a Schlenk flask and dissolved in CH2Cl2
(1.5 mL). A solution of the crude oil (containing 1.1 equiv of 1; for
exact quantities of 1 in the oil, see later) in CH2Cl2 (3 mL) was
added. The reaction mixture was protected from light, stirred at RT
for 45 min, and then concentrated at half volume under vacuum.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1180

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
or
13C NMR (126 MHz, CD2Cl2, 298 K): Major isomer (trans): d=133.8
(C5), 133.6 (C8), 130.2 (C9), 129.7 (C14’,18’), 129.6 (C4), 125.8 (C3), 125.7
(C10), 122.7 (C14,18), 122.5 (C16), 18.0 (ArꢀCH3), 3.1 (PdꢀNCCH3),
7.8 ppm (PdꢀCH3).
1H NMR (500 MHz, CD2Cl2, 298 K): d=8.23 (d, J=8.3, 1H; H5 8),
8.20 (d, J=8.3, 1H; H5 or 8), 7.57 (t, J=7.8, 2H; H4,9), 7.47–7.42 (m,
1H; ArꢀH), 7.41–7.36 (m, 3H; ArꢀH), 7.36–7.32 (m, 2H; ArꢀH), 6.71
(d, J=7.3, 1H; H3), 6.61 (d, J=7.3, 1H; H10), 2.89 (s, 6H; (CH3)2SO),
2.34 (s, 6H; ArꢀCH3), 2.29 (s, 6H; ArꢀCH3), 0.83 ppm (s, 3H; Pdꢀ
CH3).
IR (Nujol): n˜ =846 (s), 559 cmꢀ1 (s) (PF6ꢀ).
13C NMR (126 MHz, CD2Cl2, 298 K): d=134.1 (C5 8), 133.3 (C5 8),
130.0 (C4,9), 129.3 (CAr), 129.7 (CAr), 129.8 (CAr), 126.0 (C3), 126.3 (C10),
45.1 ((CH3)2SO), 18.2 (ArꢀCH3), 14.6 ppm (PdꢀCH3).
or
or
Synthesis of dmso derivatives [Pd(CH3)(dmso)(Ar-BIAN)][PF6]
(1c, 2c, 4c)
IR (Nujol): n˜ =1123 cmꢀ1 (S=O), 868 (s) and 556 cmꢀ1 (s) (PF6ꢀ).
AgPF6 (0.132 mmol) was added as a solid to a solution of the neu-
tral complex (0.115 mmol) in CH2Cl2 (2 mL) containing dmso
(0.132 mmol). The solution was stirred at RT for 30 min, and AgCl
was filtered over Celite; the solution was then concentrated at half
volume under vacuum. Upon addition of diethyl ether, the product
precipitated as a yellow solid (average yield: 62.0%).
[Pd(CH3)(dmso)(4)][PF6] (4c)
Yield: 65.5%; found: C 39.04, H 2.20, N 2.93; elemental analysis
calcd (%) for PdC31H21N2PF18SO: C 39.24, H 2.23, N 2.95; isomeric
ratio (dmso-O)/(dmso-S): CD2Cl2, 223 K: 2/3.
1H NMR (500 MHz, CD2Cl2, 223 K): Major isomer (dmso-S): d=8.33–
[Pd(CH3)(dmso)(1)][PF6] (1c)
8.17 (m, 2H; H5,8), 7.90 (s, 5H; H14,18,14’,18’ and H16 or 16’), 7.79 (s, 1H;
or
H16 16’), 7.64–7.50 (m, 2H; H4,9), 7.23–7.15 (m, 1H; H3), 6.67–6.46
Yield: 58.3%; found: C 44.16, H 3.18, N 2.28; elemental analysis
calcd (%) for PdC31H27N2PF12SO: C 44.27, H 3.24, N 3.33; isomeric
ratio (cis, dmso-S)/(trans, dmso-O)/(trans, dmso-S): CD2Cl2, 233 K:
2/3/13.
(m, 1H; H10), 3.04 (s, 6H; (CH3)2SO), 0.48 ppm (s, 3H; PdꢀCH3).
Minor isomer (dmso-O): d=8.33–8.17 (m, 2H; H5,8), 8.17–7.96 (m,
or
or
5H; H14,18,14’,18’ and H16 16’), 7.93 (m, 1H; H16 16’), 7.64–7.50 (m,
2H; H4,9), 7.23–7.15 (m, 1H; H3), 6.67–6.46 (m, 1H; H10), 2.71 (s, 6H;
(CH3)2SO), 0.67 ppm (s, 3H; PdꢀCH3).
1H NMR (500 MHz, CD2Cl2, 233 K): Major isomer (trans, dmso-S): d=
3
8.26 (d, J=8.3, 2H; H5,8), 8.04 (s, 1H; H16), 7.92 (s, 2H; H14,18), 7.61–
13C NMR (126 MHz, CD2Cl2, 223 K). Major isomer (dmso-S): d=132.8
7.53 (m, 2H; H4,9), 7.40 (dd, J=8.3, 6.8, 1H; H16’), 7.34 (d, J=7.5,
2H; H15’,17’), 6.73 (d, J=7.4, 1H; H3), 6.54 (d, J=7.3, 1H; H10), 3.07
(s, 6H; (CH3)2SO), 2.24 (s, 6H; ArꢀCH3), 0.51 ppm (s, 3H; PdꢀCH3).
Minor isomer (cis, dmso-S): d=8.21–8.18 (m, 2H; H5,8), 8.16 (s, 1H;
H16), 7.95 (s, 2H; H14,18), 6.64 (d, J=7.4, 1H; H10), 6.58 (d, J=7.4,
1H; H3), 2.88 (s, 6H; (CH3)2SO), 2.29 (s, 6H; ArꢀCH3), 0.73 ppm (s,
3H; PdꢀCH3). Minor isomer (trans, dmso-O): d=8.26 (d, J=8.3, 1H;
H5), 8.21–8.18 (m, 1H; H8), 7.99 (s, 1H; H16), 7.79 (s, 2H; H15,17),
7.53–7.48 (m, 1H; H9), 7.16 (d, J=7.3, 1H; H3), 6.48 (d, J=7.5, 1H;
H10), 2.86 (s, 6H; (CH3)2SO), 2.20 (s, 6H; ArꢀCH3), 0.84 ppm (s, 3H;
PdꢀCH3).
(C5 8), 132.3 (C5 8), 129.1 (C4,9), 125.3 (C10), 125.1 (C3), 122.4
(C16,16’), 122.2 (C14,18,14’,18’), 45.2 ((CH3)2SO), 11.23 ppm (PdꢀCH3).
Minor isomer (dmso-O): d=126.5 (C10), 122.0 (C14,18,14’,18’), 121.4
(C16,16’), 38.83 ((CH3)2SO), 3.05 ppm (PdꢀCH3).
or
or
IR (Nujol): n˜ =1131 (s) and 984 cmꢀ1 (m) (S=O), 868 (s) and
556 cmꢀ1 (s) (PF6ꢀ).
Cooligomerization reactions
All catalytic experiments were performed in a Bꢃchi tinyclave glass
reactor equipped with an interchangeable 50 mL glass vessel. The
vessel was loaded with the desired complex (21 mmol), 2,2,2-tri-
fluoroethanol (21 mL), and MA (1.13 mL). The reactor was then
placed in a preheated oil bath, connected to the ethylene tank,
and pressurized. The reaction mixture was stirred at constant tem-
perature and pressure. After the appropriate time, the reactor was
cooled to RT and vented. An aliquot (200 mL) of the reaction mix-
ture was withdrawn and diluted in CH3OH (1 mL) for GC–MS analy-
ses. The reaction mixture was poured in a 50 mL round flask, to-
gether with dichloromethane (3ꢂ2 mL) used to wash the glass
vessel. No separation of palladium black was performed. Volatiles
were removed under reduced pressure, and the residual oil was
dried at constant weight and analyzed by using 1H NMR
spectroscopy.
1H NMR (500 MHz, [D6]DMSO, 298 K): d=8.40 (d, J=8.3, 1H; H5 or 8),
or
8.37 (d, J=8.3, 1H; H5 8), 8.32 (s, 2H; H14,18), 8.28 (s, 1H; H16),
7.78–7.66 (m, 2H; H4,9), 7.41 (s, 3H; H15’,16’,17’), 6.71 (d, J=7.1, 1H;
H
3 or 10), 6.48 (d, J=7.1, 1H; H3 or10), 2.29 (s, 6H; ArꢀCH3), 0.52 ppm
(s, 3H; PdꢀCH3).
13C NMR (126 MHz, CD2Cl2, 233 K): Major isomer (trans, dmso-S): d=
133.5 (C5 or C8), 133.4 (C5 or C8), 121.7 (C16), 122.2 (C14,18), 129.0 (C4),
129.5 (C9), 128.6 (C16’), 129.3 (C15’,17’), 126.3 (C3), 126.0 (C10), 45.1
((CH3)2SO), 18.1 ppm (ArꢀCH3), 10.9 (PdꢀCH3). Minor isomer (cis,
dmso-S): d=122.5 (C16), 122.7 (C14,18), 125.8 (C10), 126.4 (C3), 44.5
((CH3)2SO), 18.1 (ArꢀCH3), 15.2 ppm (PdꢀCH3). Minor isomer (cis,
dmso-O): d=134.2 (C5), 121.6 (C16), 122.6 (C14,18), 129.6 (C9), 125.0
(C3), 125.1 (C10), 37.5 ((CH3)2SO), 17.9 (ArꢀCH3), 7.6 ppm (PdꢀCH3).
13C NMR (126 MHz, [D6]DMSO, 298 K): d=133.1 (C5 or 8), 132.3
(C5 or 8), 123.2 (C14,18), 121.1 (C16), 129.4 (C4,9), 128.8 (C15’,16’,17’), 125.2
(C3 or 10), 124.3 (C3 or 10), 17.6 (ArꢀCH3), 9.4 ppm (PdꢀCH3).
Ethylene/MA cooligomer
IR (Nujol): n˜ =1139 (s) and 975 (m) (S=O), 847 (s) and 558 cmꢀ1 (s)
1H NMR (500 MHz, CDCl3, 298 K): d=5.57–4.90 (m; CH=CH), 3.60 (s;
OCH3), 2.24 (t, J=7.5; CH2C(O)), 2.09–1.79 (m; CH2CH=CH), 1.68–
1.46 (m; CH2CH2C(O), CH3CH=CH), 1.46–0.99 (m; ꢀ(CH2)ꢀ, ꢀCH(R)ꢀ),
0.99–0.86 (m; CH3CH2CH=CH, ꢀCH3), 0.86–0.63 ppm (m, ꢀCH3).
13C NMR (126 MHz, CDCl3, 223 K): Characteristic resonances due to
the functionalized cooligomer: d=173.9 (C(O)), 51.1 (OCH3), 33.9
(CH2C(O)), 24.8 ppm (CH2CH2C(O)).
(PF6ꢀ).
[Pd(CH3)(dmso-S)(2)][PF6] (2c)
Yield: 62.3%; found: C 50.76, H 4.48, N 3.84; elemental analysis
calcd (%) for PdC31H33N2PF6SO: C 50.79, H 4.54, N 3.82.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1181

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[8] L. K. Johnson, C. M. Killian, M. Brookhart, J. Am. Chem. Soc. 1995, 117,
6414–6415.
In situ NMR investigations
A 10 mm CD2Cl2 solution of the complex in the NMR tube was re-
acted with the monomer (see later) kept in the spectrometer at
298 K, and the Varian pad experiment was performed on it, with
[9] S. D. Ittel, L. K. Johnson, M. Brookhart, Chem. Rev. 2000, 100, 1169–
1204.
[10] L. K. Johnson, S. Mecking, M. Brookhart, J. Am. Chem. Soc. 1996, 118,
267–268.
1
time delay set to 300 s. The required time for each H NMR experi-
[11] S. Mecking, L. K. Johnson, L. Wang, M. Brookhart, J. Am. Chem. Soc.
1998, 120, 888–899.
[12] E. Drent, R. van Dijk, R. van Ginkel, B. van Oort, R. I. Pugh, Chem.
Commun. 2002, 744–745.
[13] T. Kochi, K. Yoshimura, K. Nozaki, Dalton Trans. 2006, 25–27.
[14] D. Guironnet, P. Roesle, T. Runzi, I. Gottker-Schnetmann, S. Mecking, J.
Am. Chem. Soc. 2009, 131, 422–423.
[15] S. Ito, K. Munakata, A. Nakamura, K. Nozaki, J. Am. Chem. Soc. 2009, 131,
14606–14607.
ment (60 s) was taken into account when analyzing the kinetic
data.
Ethylene: In the case of the apolar monomer, the NMR tube was
kept in a suitable Schlenk flask and ethylene was bubbled through
a cannula coaxial to the tube and connected to the ethylene tank.
The tip of the cannula was kept 2 cm above the interphase to
avoid any loss of solution.
[16] T. Kochi, S. Noda, K. Yoshimura, K. Nozaki, J. Am. Chem. Soc. 2007, 129,
8948–8949.
Methyl acrylate: In the case of the polar monomer, MA (2 equiv)
was added to the solution through a 10 mL syringe.
[17] S. Ito, M. Kanazawa, K. Munakata, J. Kuroda, Y. Okumura, K. Nozaki, J.
Am. Chem. Soc. 2011, 133, 1232–1235.
[18] S. Luo, J. Vela, G. R. Lief, R. F. Jordan, J. Am. Chem. Soc. 2007, 129, 8946–
8947.
Detected intermediates
1H NMR (500 MHz, CD2Cl2, 298 K): 1b(C“): d=8.25 (m, 2H; H5,8),
8.06 (s, 1H; H16), 8.04 (s, 2H; H14,18), 7.64 (t, J=7.8, 1H; H4), 7.55 (t,
J=7.8, 1H; H9), 7.44–7.38 (m, 1H; H16’), 7.38–7.33 (m, 2H; H14’,18’),
7.24 (d, J=7.3, 1H; H3), 6.56 (d, J=7.3, 1H; H10), 3.42 (s, 3H; OCH3),
2.64–2.56 (m, 2H; CH2C(O)), 2.32 (s, 6H; ArꢀCH3), 1.71 (t, J=5.9,
2H; PdꢀCH2CH2CH2C(O)), 0.74 ppm (pentet, J=5.9, 2H;
CH2CH2C(O)); 1b(B”): d=1.81 (m, 1H; PdꢀCH(CH3)), 0.43 ppm (d,
J=7.1, 3H; PdꢀCH(CH3)); 1b(A’): d=2.33 (m, 1H; CH(CH3)), 0.69
(m, 3H; CH(CH3)); 1b(D’): d=3.75 (s, 3H; OCH3), 2.49 (t, J=6.9, 1H;
CH2C(O)), 1.65 ppm (t, J=6.9, 1H; PdꢀCH2).
13C NMR (126 MHz, CD2Cl2, 298 K): 1b(C“): d=183.5 (C(O)), 133.4
(C5,8), 130.0 (C9), 129.8 (C14’,18’), 129.3 (C4), 129.0 (C16’), 125.8 (C3),
125.6 (C10), 122.5 (C14,18), 122.1 (C16), 54.9 (OCH3), 36.1 (CH2C(O)),
31.7 (PdCH2CH2CH2C(O)), 24.0 (CH2CH2C(O)), 18.0 ppm (ArꢀCH3);
1b(B”): d=34.7 (PdꢀCH(CH3)), 21.9 ppm (PdꢀCH(CH3)); 1b(A’): d=
36.5 (CH(CH3)), 17.7 (CH(CH3)); 1b(D’): d=191.3 (C(O)), 55.5
(C(O)OCH3), 37.9 ppm (CH2C(O)).
[19] W. Weng, Z. Shen, R. F. Jordan, J. Am. Chem. Soc. 2007, 129, 15450–
15451.
[20] K. M. Skupov, J. Hobbs, P. Marella, D. Conner, S. Golisz, B. L. Goodall, J. P.
Claverie, Macromolecules 2009, 42, 6953–6963.
[21] K. M. Skupov, L. Piche, J. P. Claverie, Macromolecules 2008, 41, 2309–
2310.
[22] C. S. Popeney, D. H. Camacho, Z. Guan, J. Am. Chem. Soc. 2007, 129,
10062–10063.
[23] S. Luo, R. F. Jordan, J. Am. Chem. Soc. 2006, 128, 12072–12073.
[24] X. B. Shi, Y. Zhao, H. Y. Gao, L. Zhang, F. M. Zhu, Q. Wu, Macromol. Rapid
Commun. 2012, 33, 374–379.
[25] B. P. Carrow, K. Nozaki, J. Am. Chem. Soc. 2012, 134, 8802–8805.
[26] A. Scarel, M. R. Axet, F. Amoroso, F. Ragaini, C. J. Elsevier, A. Holuigue, C.
Carfagna, L. Mosca, B. Milani, Organometallics 2008, 27, 1486–1494.
[27] T. Rꢃnzi, D. Guironnet, I. Gçttker-Schnetmann, S. Mecking, J. Am. Chem.
Soc. 2010, 132, 16623–16630.
[28] B. Neuwald, F. ꢄlscher, I. Gçttker-Schnetmann, S. Mecking, Organometal-
lics 2012, 31, 3128–3137.
[29] M. Gasperini, F. Ragaini, E. Gazzola, A. Caselli, P. Macchi, Dalton Trans.
2004, 3376–3382.
[30] R. van Asselt, C. J. Elsevier, W. J. J. Smeets, A. L. Spek, R. Benedix, Recl.
Trav. Chim. Pays-Bas 1994, 113, 88–98.
[31] N. J. Hill, I. Vargas-Baca, A. H. Cowley, Dalton Trans. 2009, 240–253.
[32] J. Kovach, M. Peralta, W. W. Brennessel, W. D. Jones, J. Mol. Struct. 2011,
992, 33–38.
[33] U. El-Ayaan, F. Murata, S. El-Derby, Y. Fukuda, J. Mol. Struct. 2004, 692,
209–216.
[34] D. N. Coventry, A. S. Batsanov, A. E. Goeta, J. A. K. Howard, T. B. Marder,
Polyhedron 2004, 23, 2789–2795.
[35] I. L. Fedushkin, V. A. Chudakova, A. A. Skatova, N. M. Khvoinova, Y. A.
Kurskii, T. A. Glukhova, G. K. Fukin, S. Dechert, M. Hummert, H. Schu-
mann, Z. Anorg. Allg. Chem. 2004, 630, 501–507.
[36] V. Rosa, T. Aviles, G. Aullon, B. Covelo, C. Lodeiro, Inorg. Chem. 2008, 47,
7734–7744.
Acknowledgements
This work was financially supported by the Italian MIUR (PRIN N8
207HMTJWP 002). Fondazione CRTrieste and Fondazione Benefi-
ca Kathleen Foreman Casali are gratefully acknowledged for the
generous donations of a Varian 500 MHz spectrometer and
a Bꢀchi tinyclave glass reactor, respectively. P.F. and T.M. acknowl-
edge Consorzio Interuniversitario Nazionale per la Scienza e Tec-
nologia dei Materiali for financial support.
[37] R. van Asselt, E. E. C. G. Gielens, R. E. Rulke, C. J. Elsevier, J. Chem. Soc.
Chem. Commun. 1993, 1203–1205.
Keywords: ethylene · N ligands · oligomerization · palladium ·
polar monomer
[38] R. E. Rꢃlke, J. M. Ernsting, A. L. Spek, C. J. Elsevier, P. W. N. M. Van Leeu-
wen, K. Vrieze, Inorg. Chem. 1993, 32, 5769–5778.
[39] V. Khlebnikov, A. Meduri, H. Mueller-Bunz, T. Montini, P. Fornasiero, E.
Zangrando, B. Milani, M. Albrecht, Organometallics 2012, 31, 976–986.
[40] J. Zhou, H. Sun, K. Harms, J. Sundermeyer, Z. Anorg. Allg. Chem. 2008,
634, 1517–1521.
[41] J. Durand, E. Zangrando, M. Stener, G. Fronzoni, C. Carfagna, B. Binotti,
P. C. J. Kamer, C. Muller, M. Caporali, P. W. N. M. van Leeuwen, D. Vogt, B.
Milani, Chem. Eur. J. 2006, 12, 7639–7651.
[1] A. Nakamura, S. Ito, K. Nozaki, Chem. Rev. 2009, 109, 5215–5244.
[2] A. Berkefeld, S. Mecking, Angew. Chem. 2008, 120, 2572–2576; Angew.
Chem. Int. Ed. 2008, 47, 2538–2542.
[3] A. Sen, S. Borkar, J. Organomet. Chem. 2007, 692, 3291–3299.
[4] J.-Y. Dong, Y. Hu, Coord. Chem. Rev. 2006, 250, 47–65.
[5] L. S. Boffa, B. M. Novak, Chem. Rev. 2000, 100, 1479–1494.
[6] S. Ito, K. Nozaki, Chem. Rec. 2010, 10, 315–325.
[42] J. D. Fotheringham, G. A. Heath, A. J. Lindsay, T. A. Stephenson, J. Chem.
Res. (S) 1986, 82–83.
[43] A. Bastero, A. Ruiz, C. Claver, B. Milani, E. Zangrando, Organometallics
2002, 21, 5820–5829.
[7] J. A. Gladysz, Z. T. Ball, G. Bertrand, S. A. Blum, V. M. Dong, R. Dorta, F. E.
Hahn, M. G. Humphrey, W. D. Jones, J. Klosin, I. Manners, T. J. Marks,
J. M. Mayer, B. Rieger, J. C. Ritter, A. P. Sattelberger, J. M. Schomaker,
V. W.-W. Yam, Organometallics 2012, 31, 1–18.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1182

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[44] A. Bastero, C. Claver, A. Ruiz, S. Castillon, E. Daura, C. Bo, E. Zangrando,
Chem. Eur. J. 2004, 10, 3747–3760.
[45] A. Scarel, J. Durand, D. Franchi, E. Zangrando, G. Mestroni, B. Milani, S.
Gladiali, C. Carfagna, B. Binotti, S. Bronco, T. Gragnoli, J. Organomet.
Chem. 2005, 690, 2106–2120.
[51] T. Usami, S. Takayama, Macromolecules 1984, 17, 1756–1761.
[52] D. E. Axelson, G. C. Levy, L. Mandelkern, Macromolecules 1979, 12, 41–
52.
[53] P. Braunstein, C. Frison, X. Morise, Angew. Chem. 2000, 112, 2989–2992;
Angew. Chem. Int. Ed. 2000, 39, 2867–2870.
[46] M. R. Axet, F. Amoroso, G. Bottari, A. D’Amora, E. Zangrando, F. Faraone,
D. Drommi, M. Saporita, C. Carfagna, P. Natanti, R. Seraglia, B. Milani, Or-
ganometallics 2009, 28, 4464–4474.
[47] A. D’Amora, L. Fanfoni, D. Cozzula, N. Guidolin, E. Zangrando, F. Felluga,
S. Gladiali, F. Benedetti, B. Milani, Organometallics 2010, 29, 4472–4485.
[48] D. J. Tempel, L. K. Johnson, R. L. Huff, P. S. White, M. Brookhart, J. Am.
Chem. Soc. 2000, 122, 6686–6700.
[54] M. Agostinho, P. Braunstein, Chem. Commun. 2007, 58–60.
[55] S. Borkar, H. Yennawar, A. Sen, Organometallics 2007, 26, 4711–4714.
[56] E. Zuidema, C. Bo, P. W. N. M. van Leeuwen, J. Am. Chem. Soc. 2007, 129,
3989–4000.
[57] M. Gasperini, F. Ragaini, S. Cenini, Organometallics 2002, 21, 2950–
2957.
[49] E. F. McCord, S. J. McLain, L. T. J. Nelson, S. D. Ittel, D. Tempel, C. M. Kill-
ian, L. K. Johnson, M. Brookhart, Macromolecules 2007, 40, 410–420.
[50] E. F. McCord, S. J. McLain, L. T. J. Nelson, S. D. Arthur, E. B. Coughlin, S. D.
Ittel, L. K. Johnson, D. Tempel, C. M. Killian, M. Brookhart, Macromole-
cules 2001, 34, 362–371.
Received: July 31, 2012
Published online on November 5, 2012
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 1170 – 1183 1183