Role of Radical Species in Salicylaldiminato Ni(II) Mediated Polymer Chain Growth: A Case Study for the Migratory Insertion Polymerization of Ethylene in the Presence of Methyl Methacrylate

Article abstract of DOI:10.1021/jacs.5b08612

To date, an inconclusive and partially contradictive picture exists on the behavior of neutral Ni(II) insertion polymerization catalysts toward methyl methacrylate (MMA). We shed light on this issue by a combination of comprehensive mechanistic NMR and EPR studies, isolation of a key Ni(I) intermediate, and pressure reactor studies with ethylene and MMA, followed by detailed polymer analysis. An interlocking mechanistic picture of an insertion and a free radical polymerization is revealed. Both polymerizations run simultaneously (25 bar ethylene, neat MMA, 70 °C); however, the chain growth cycles are independent of each other, and therefore exclusively a physical mixture of homo-PE and homo-PMMA is obtained. A Ni-C bond cleavage was excluded as a free radical source. Rather a homolytic P-C bond cleavage in the labile aryl phosphine ligand and the reaction of low-valent Ni(0/I) species with specific iodo substituted N^O (Ar-I) ligands were shown to initiate radical MMA polymerizations. Several reductive elimination decomposition pathways of catalyst precursor or active intermediates were shown to form low-valent Ni species. One of those pathways is a bimolecular reductive coupling via intermediate (N^O)Ni(I) formation. These intermediate Ni(I) species can be prevented from ultimate decomposition by capturing with organic radical sources, forming insertion polymerization active [(N^O)Ni(II)-R] species and prolonging the ethylene polymerization activity.

Full text of DOI:10.1021/jacs.5b08612

Article
pubs.acs.org/JACS
Role of Radical Species in Salicylaldiminato Ni(II) Mediated
Polymer Chain Growth: A Case Study for the Migratory Insertion
Polymerization of Ethylene in the Presence of Methyl Methacrylate
†
†
‡
,†
̈
Franz Olscher, Inigo Gottker-Schnetmann, Vincent Monteil, and Stefan Mecking*
̈
†Chair of Chemical Materials Science, Department of Chemistry, University of Konstanz, D-78464 Konstanz, Germany
‡Universite
́
de Lyon, Univ. Lyon 1, CPE Lyon, CNRS UMR 5265 Laboratoire de Chimie Catalyse Polymeres et Procedes (C2P2),
LCPP team Bat 308F, 43 Bd du 11 novembre 1918, F-69616 Villeurbanne, France
̀
́
́
S
* Supporting Information
ABSTRACT: To date, an inconclusive and partially contra-
dictive picture exists on the behavior of neutral Ni(II)
insertion polymerization catalysts toward methyl methacrylate
(MMA). We shed light on this issue by a combination of
comprehensive mechanistic NMR and EPR studies, isolation
of a key Ni(I) intermediate, and pressure reactor studies with
ethylene and MMA, followed by detailed polymer analysis. An
interlocking mechanistic picture of an insertion and a free
radical polymerization is revealed. Both polymerizations run
simultaneously (25 bar ethylene, neat MMA, 70 °C); however,
the chain growth cycles are independent of each other,
and therefore exclusively a physical mixture of homo-PE
and homo-PMMA is obtained. A Ni−C bond cleavage was
excluded as a free radical source. Rather a homolytic P−C bond cleavage in the labile aryl phosphine ligand and the reaction
of low-valent Ni(0/I) species with specific iodo substituted N^O (Ar−I) ligands were shown to initiate radical MMA
polymerizations. Several reductive elimination decomposition pathways of catalyst precursor or active intermediates were shown
to form low-valent Ni species. One of those pathways is a bimolecular reductive coupling via intermediate (N^O)Ni(I)
formation. These intermediate Ni(I) species can be prevented from ultimate decomposition by capturing with organic
radical sources, forming insertion polymerization active [(N^O)Ni(II)−R] species and prolonging the ethylene polymerization
activity.
Unique catalyst systems in the general context of functional
group tolerant polymerizations are κ2-N^O-salicylaldiminato
Ni(II) complexes.10 Remarkably, these are tolerant toward
polymerizations in aqueous media.11 Additionally, copolymer-
ization of ethylene with substituted norbonenes or α-olefins,
respectively, with a functional group in a remote position to the
monomers’ double bond have been reported.12 Concerning
fundamental monomers, like acrylates, Grubbs et al. reported
that no polymers are formed in attempted copolymerizations of
MA with ethylene applying salicylaldiminato Ni(II) complexes.13
Mechanistic studies provide an explanation for this finding. It has
been shown that MA inserts into the Ni−C bond of [Ni−Ph]14
or [Ni−Me]15 complexes and also into reactive [Ni−H]15
species. However, further migratory chain growth is hindered
due to chelate formation by the ester group of the inserted MA
and β-H elimination occurs instead. The resulting [Ni−H]
species reacts with the residual MA insertion product in a
bimolecular elimination at temperatures ≥0 °C, resulting in
INTRODUCTION
■
While catalytic polymerizations of ethylene (E) and propylene
are applied on a very large scale, an incorporation of polar vinyl
monomers in such reactions is challenging. The common early
transition metal catalysts employed are deactivated by polar
compounds due to their high oxophilicity. This problem can be
addressed by late transition metal catalysts that are less oxophilic
and more functional group tolerant.1 For example, a copoly-
merization of ethylene with acrylates was achieved with
Brookhart’s cationic α-diimine Pd(II) complexes. As a result of
a chain walking mechanism, highly branched polyethylene is
formed with the acrylate-derived repeat units located at the ends
of branches preferentially.2 By contrast, neutral κ2-phosphine-
sulfonato Pd(II) complexes form linear copolymers of ethylene
and acrylates.3 These catalysts4 are compatible with a range of
polar vinyl monomers, including difficult candidates like
acrylonitrile,5 vinyl acetate,6 and acrylic acid.7 High incorpo-
rations of methyl acrylate (MA) of up to 52 mol % were achieved.
In the absence of ethylene, MA is homopolymerized by an
insertion mechanism.8 By contrast, methyl methacrylate (MMA)
is not incorporated by these catalysts.9
Received: October 22, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
formation of saturated organic substrate and irreversible deacti-
vation of the catalyst ([Ni−H] + [Ni−alkyl] → H−alkyl + “Ni
species”).15,16
were selected for their robustness and high activity in ethylene
polymerizations,11,22 and (N^O)iPr,I−I was chosen because the
−
Ph
PPh3
corresponding phenyl PPh3 complex (2 ) was described to
mediate insertion and radical polymerizations.20 Furthermore,
complexes with ligand (N^O)iPr,tBu‑H− (3LR) do not contain iodo
substituents on the salicylaldehyde (R2 and R3 in 1 and 2), which
allows the determination of a specific influence of Ar−I
functionalities. Besides [Ni−Ph] complexes as most common
polymerization catalyst precursors, we used [Ni−Me] complexes
as a closer mimic of the growing [Ni−alkyl] species in the
ethylene polymerization. Additionally, [Ni−H] complexes were
applied in this study because they are reactive intermediates
formed after β-H elimination during insertion polymerization. In
addition to phosphines, which were reported to promote the
radical polymerization process,20 dimethyl sulfoxide (DMSO)
was used as a weakly coordinating ligand, enabling insertion
reactions into the [Ni−C] moiety at milder conditions (for de-
tails, see Supporting Information).
However, repetitive reports on the formation of polymers
containing incorporated methyl methacrylate (MMA) with
similar catalysts generate an inconclusive and partially contra-
dictive picture. [Bis(β-ketoiminato)Ni(II)]/MAO systems were
used for homopolymerizations of MMA. Here, a radical
mechanism was proposed, involving a reversible Ni−C homo-
lytic bond cleavage.17 Furthermore, the efficient copolymeriza-
tion of ethylene and MMA mediated by [(β-ketoiminato)Ni-
(II)Ph(PPh3)]/MMAO systems was reported, claiming an
insertion mechanism pathway.18 A similar approach using
bimolecular salicylaldiminato N^O Ni(II) complexes to
copolymerize ethylene and MMA via migratory insertion was
reported; however, some results were not reproducible.19 In
addition, a novel approach for the simultaneous polymerization
of ethylene with MMA was described, combining a migratory
insertion mechanism for ethylene and a radical mechanism for
MMA, resulting in block copolymers.20 Neutral [(N^O)iPr,I−INi-
(II)Ph(PPh3)] and [(κ2-phosphineenolate)Ni(II)Ph(PPh3)]
catalyst precursors were shown to be active for both types of
polymerization mechanisms, with a postulated reversible homo-
lytic Ni−C bond cleavage as transfer reaction between insertion
and radical polymerization. Addition of PPh3 to the nickel
complexes promotes free radical MMA homopolymerizations
and influences the MMA content in simultaneous polymer-
izations of ethylene with MMA.
Reactivity of Different Ni(II)-Catalyst Precursors to-
ward MMA. Leblanc et al. reported that free radical MMA
polymerizations occur in the presence of [(N^O)iPr,I−INi(II)Ph-
(PPh3)] (2 ), which are promoted by additional PPh3.20
Ph
PPh3
However, the role of PPh3 remained unclear. Herein, we provide
further insights by performing homopolymerizations of MMA,
labeling experiments, and detailed polymer analysis.
Homopolymerization of MMA Initiated with [(N^O)Ni(II)-
R(PPh3)] Complexes. To elucidate the effect of the salicylaldi-
minato ligand’s substitution pattern and additional phosphine in
[Ni] initiated MMA homopolymerizations, [(N^O)CF,I−INiPh-
This mixed picture indicated that an elucidation of underlying
reactions is of general interest to reveal relevant pathways
not recognized to date in polar monomer polymerizations.
We now report possible formation pathways and multiple roles
of radicals in Ni(II) catalyzed polymerizations, as concluded
from reactivities of intermediates observed under NMR and
pressure reactor conditions and from detailed analysis of formed
polymers.
Ph
PPh3
Ph
PPh3
(PPh3)] (1 ) and [(N^O)iPr,tBu‑HNiPh(PPh3)] (3 ) were
reacted in neat MMA at 70 °C with different concentrations of
PPh3. A significant decrease in molecular weights was observed
with increasing concentration of PPh3 (Table 1, entries 1−3).
Table 1. MMA Homopolymerizations Initiated with
Ph
Ph
Complexes 1 , 1Me and 3
with or without
PPh3 DMSO
PPh3
a
Additional PPh3
RESULTS AND DISCUSSION
■
b
c
General Considerations. Various systems containing Ni(II)
complexes have been shown to promote the simultaneous
polymerization of ethylene and polar vinyl monomers.17,18,20,21
In our present study, we have focused on polymerizations of
ethylene in the presence of MMA using neutral (N^O)
salicylaldiminato Ni(II) aryl-phosphine complexes. For these
systems, detailed mechanistic studies on the reaction of
ethylene16 and MA14,15 have been reported previously (vide
supra).
entry complex equiv add. yield, % Mn, 103 g·mol−1 chains/[Ni]
Ph
1
2
3
1
1
1
2.8
6.2
6.8
803
610
463
0.05
0.14
0.26
PPh3
Ph
PPh3
3PPh3
5PPh3
3PPh3
Ph
PPh3
Me
4
5
1
3
7.2
1.1
563
270
0.21
0.06
DMSO
Ph
PPh3
Ph
PPh3
6
3
3PPh3
1.8
251
0.10
a
Reaction conditions: 70 °C, neat MMA, c([Ni]) = 0.5 mmol·L−1. Details
b
We have employed a number of different (N^O) Ni(II)
see Supporting Information. Determined by SEC (40 °C, THF) vs PS
standards. Calculated by m(PMMA)/[Mn(PMMA)n([Ni])].
complexes (Chart 1): Complexes with ligand (N^O)CF,I−I− (1LR)
c
Chart 1. General Notation of (N^O)Ni(II)R(L) Complexes
Used in This Study
This observation can be rationalized by a higher steady state
concentration of free radicals and therefore a higher rate of
bimolecular termination reactions of the growing radical
polymeric species (by radical recombination or disproportiona-
tion).23 An accurate measure for the efficiency of the polymer-
ization initiation reaction is the amount of PMMA chains formed
per Ni center (chains/[Ni]). In the absence of additional PPh3,
0.05 chains/[Ni] were formed with complex [(N^O)CF,I−INiPh-
(PPh3)], while addition of 3 or 5 equiv of PPh3 increased the
chains/[Ni] ratio to 0.14 and 0.26, respectively. Similarly,
for [(N^O)iPr,tBu‑HNiPh(PPh3)], a ratio of 0.06 chains/[Ni]
was obtained, which increased to 0.10 with 3 equiv of PPh3
B
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(Table 1, entries 5 and 6). This already points out a crucial role of
polymerized PMMA initiated with the radical initiator
azo-bis(isobutyronitrile) (AIBN), applying similar reaction
conditions (Figure S18−21). In addition, unsaturated end
the phosphine for the generation of free radicals. Beyond
Me
DMSO
[Ni−Ph] complexes, a [Ni−Me] species (1
with 3 equiv of
1
PPh3), as a model compound for the growing [Ni−PE] species in
ethylene polymerization, was shown to initiate radical MMA
polymerizations, enabling comparable yields, molecular weights,
and chains/[Ni] ratios (Table 1, entry 2 vs 4) and resembling
similar kinetic polymerization behavior as the corresponding
[Ni−Ph] complexes (Figure S16, red vs blue).
groups were detected by H NMR spectroscopy (5.45 and
6.15 ppm in CD2Cl2, Figure 1), which were formed via a
bimolecular disproportionation reaction of two radical PMMA
polymeryl chains.23
In addition to the effect of ligand substitution and the influence
of additional phosphine, we investigated the effect of the con-
centration of initially supplied [(N^O)Ni(II)R(L)] on the
efficiency of radical MMA polymerizations. This can give
valuable insights into the initiation reaction, for example, if a
key reaction step herein is mononuclear or bimolecular. To this
Ph
PPh3
end, [(N^O)iPr,I−INiPh(PPh3)] (2 ), which was previously
described to be an efficient radical polymerization initiator,20 was
reacted at different [Ni] concentrations with MMA (Table 2).
Figure 1. 1H NMR spectra (400 MHz, CD2Cl2, RT) of the olefinic and
Table 2. MMA Homopolymerizations Initiated with Complex
13CH3
Ph
Ph
aromatic region of PMMA initiated with 1PPh (a), 2 (b), phenylazo-
2
PPh3
with Additional 3 equiv of PPh3 in Dependency of the
PPh3
3
Me
P(mTol)3
a
triphenylmethane (c), and 2
(d). Details see Table S3.
[Ni] Concentration
c([Ni]),
chains/
c
b
entry complex
mmol·L−1
0.5
yield, % Mn, 103 g·mol−1
[Ni]
To gain further insights into the initiation process, 13CH3
13CH3
Ph
1
2
3
2
2
2
4.3
16.0
27.1
321
35
0.18
0.63
1.52
PPh3
labeled 1PPh was employed as an initiator (60 °C, 3 equiv
3
13CH3
Ph
PPh3
4.1
of PPh3). The PMMA formed by initiation with 1PPh should
3
Ph
PPh3
contain a 13C labeled end group if the release of an organic free
radical occurs by homolytic Ni−C bond cleavage after insertion
of MMA into the Ni−C bond or directly from the [Ni−Me]
precursor. However, no 13CH3 label incorporation was detected
8.4
12
a
Reaction conditions: 3 equiv of PPh3, 70 °C, c(MMA) = 9.389 M for
entry 1 and 5.633 M for entries 2 and 3, with toluene as a solvent.
Determined by SEC (40 °C, THF) vs PS standards. Calculated by
m(PMMA)/[Mn(PMMA)n([Ni])].
b
c
by 13C NMR spectroscopy in the PMMA formed (Mn,NMR
=
30 × 103 g/mol, Figure S22; quantification of products, see
Supporting Information).25 Therefore, both anticipated path-
ways for the initiation of the radical MMA polymerization,
homolytic Ni−C bond cleavage of an MMA insertion product or
the [Ni−Me] precursor, can be excluded. This also indicates that
in the simultaneous polymerization of ethylene and MMA the
alkyl moiety at the Ni center (e.g., PE polymeryl) is unlikely to
initiate free radical MMA chain growth. This precludes the
formation of PE-block-PMMA polymers via a simple homolytic
Ni−C bond cleavage in [Ni−PE/alkyl] complexes.
Ph
When the initial concentration of 2 was raised from 0.5 to 4.1
PPh3
and finally to 8.4 mmol·L−1, a significant increase in the amount
of PMMA chains per [Ni] center (0.18, 0.63, and 1.52,
respectively) was observed. That is, the radical polymerization
initiation efficiency depends on the initial [Ni] concentration.
This suggests a radical initiation process, which includes a
reaction step involving more than one [Ni] center.
In analogy to, for example, cobalt mediated radical polymer-
izations,24 one could expect a reversible trapping of the free
radical PMMA chains by [Ni] species formed during polymeri-
Surprisingly, in the 1H NMR spectrum of the PMMA formed
13CH3
by initiation with 1PPh , aromatic end groups were clearly
3
Ph
PPh3
Ph
PPh3
zation. However, for 1
and 2
(with addition of PPh3), a
identified (0.20 equiv with respect to Hb, Figure 1a). The same
Ph
PPh3
plot of Mn(PMMA) vs conversion is not linear, and poly-
dispersities (Mw/Mn) are in the range of 2−5 (Figure S17).
This indicates that no controlled polymerization was operative
here.
end groups could be identified when 2
with 3 equiv of PPh3
(0.25 equiv with respect to Hb, Figure 1b) or phenylazo-
triphenylmethane (PAT) as a phenyl radical source (Figure 1c)
were used as initiators. The assignment of phenyl end groups was
further confirmed by matrix-assisted laser desorption/ionization
time-of-flight mass spectroscopy (MALDI-TOF MS, Figure S66),
The MMA homopolymerizations revealed that complexes
Ph
Ph
1PPhPh , 2 , and 3
effectively promote radical MMA poly-
PPh3
PPh3
3
1
and the H NMR spectrum is in agreement with reported
merizations, with a strong dependency on the phosphine and the
initial [Ni] concentration. However, the nature of the initiating
organic radicals remains unclear. Therefore, the PMMA
polymers formed were studied in more detail to provide insights
into the initiation process.
Microstructure and End Group Analysis of PMMA
Homopolymers. According to literature reports,20 a radical
polymerization mechanism is active in the aforementioned
polymerizations. This is confirmed, because PMMA homopol-
ymers obtained by initiation with Ni(II) precursors show glass
transition temperatures (Tg’s) of ca. 124 °C and are slightly
syndio enriched (about 60% rr). This agrees with free radically
resonances for phenyl terminated PMMA.26 Since PPh3 is the sole
13CH3
phenyl source in 1PPh , the aryl end group of the resulting PMMA
3
must originate from the phosphine compound. This was further
confirmed by using P(mTol)3 as a phosphine ligand in the initiator
Me
P(mTol)3
complex (2
), indeed resulting in different 1H NMR
resonances of the aromatic end groups in the PMMA obtained
from MMA homopolymerization (Figure 1d). Possible pathways
to this end group formation are discussed later on in the section on
sources of organic free radicals.
Reactivity of MMA toward Insertion Chain Growth
Species. In the previous section, it was shown that the initiation
C
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
■
Figure 2. Time dependent plots of the overall productivity (mg(polymer)/μmol([Ni]), , left axis), together with mol % MMA-based repeat units
Me
Me
Ph
▼
(
, right axis) for the simultaneous polymerization of ethylene and MMA with complexes 1PPh (a), 2 (b), and 2 (c). The lines were included as
PPh3 PPh3
3
guidance for the eye. Reaction conditions: 70 °C, 3 equiv of PPh3, 50 mL of MMA, 25 bar of E. For details and additional data, see Supporting
Information.
of radical MMA polymerizations can be effectively initiated with
[(N^O)Ni(II)R(PPh3)] complexes. In addition, these com-
plexes are known as highly active catalyst precursors for ethylene
insertion polymerization,22 and even simultaneous insertion and
radical polymerizations were reported.20 However, the results of
the aforementioned homopolymerizations of MMA, do not
allow a prediction of the nature of the polymers forming during
insertion polymerization of ethylene in the presence of MMA.
Different scenarios are conceivable, depending on the interaction
of both polymerization modes: Statistical or block copolymers
could be formed, or without any interaction, the formation of
homo-PE and homo-PMMA mixtures would be anticipated.
Therefore, we conducted insertion polymerizations of ethylene
mediated by [(N^O)Ni(II)R(PPh3)] complexes in the presence
of MMA and analyzed the polymers carefully.
formed per [Ni] center (details, see Supporting Information).
Even though a lower initial [Ni] concentration was supplied in
simultaneous ethylene and MMA polymerizations in comparison
to the polymerization of MMA alone, a substantially higher
PMMA chains per [Ni] center ratio of 0.35 was already
established after 5 min. In polymerizations of MMA alone under
similar conditions, such high ratios were not reached even after 2
h of reaction time (Table S2). Apparently, the presence of
ethylene seems to accelerate the generation of free radicals.
The progress of initiation of the ethylene polymerization was
studied by using 13C labeled [Ni−Me] and [Ni−Ph] catalyst
precursors, which introduce a detectable and quantifiable 13CH3
or phenyl group at the PE chain end. Both catalyst precursors,
13CH3
Ph
2PPh and 2
(Table S6, entries 2 vs 5), were activated by
PPh3
3
ethylene to ca. 50% after 10 min; therefore, the activation rate by
Polymerization of Ethylene in the Presence of MMA. To
provide kinetic insights, time dependent polymerizations with
ethylene seems to be similar for the [Ni−Me] and [Ni−Ph]
complexes here. The extent of activation eventually further
Me
Ph
1MPPeh , 2 , and 2 with an additional 3 equiv of PPh3 at 70 °C
Ph
increased up to 62% after 120 min for complex 2
(Table S6,
PPh3
PPh3
PPh3
3
and 25 bar of ethylene were carried out in neat MMA (Figure 2).
In every case, the resulting polymer product contained an almost
equal amount of repeat units from ethylene and MMA.
Generally, a reduced productivity in the insertion polymerization
of ethylene was observed compared with previously described
polymerizations in a noncoordinating solvent such as toluene.22
This can be accounted for by the coordination of additional PPh3
or coordination of the ester group of the MMA solvent to a
entries 1−4). At the same time, the overall MMA content of the
polymer composition formed was 40% (mol % MMA-based
repeat units) after 5 min, increased to 56% after 10 min, and
finally decreased to 46% after 120 min. This implies that the free
radical polymerization is mainly active during the reaction
period, when a high concentration of active species of the
ethylene polymerization is also present in the reaction mixture
(for further discussion, see Supporting Information). At first
sight, this indicates an interaction of both polymerization modes.
Microstructure Analysis of the Polymers Formed in
Simultaneous Ethylene and MMA Polymerizations. Multiple
techniques, such as differential scanning calorimetry (DSC), size
exclusion chromatography (SEC) and diffusion ordered NMR
spectroscopy (DOSY) were used for characterization of the
polymers formed by simultaneous ethylene and MMA polymer-
ization to clearly identify the nature of the resulting polymer
products.
vacant coordination site on the Ni center, hindering the
Me
coordination of ethylene. Complex 1
was almost twice as
PPh3
productive as 2RPPh (R = Me, Ph) regarding the overall pro-
3
ductivity (Figure 2, panel a vs b and c). The mol % content of
MMA-based repeat units in the polymer product for all
complexes evolved similarly over time (Figure 2, red). Most of
the MMA was polymerized on the same time scale in which the
insertion polymerization was mainly active; after ca. 40 min, the
mol % MMA content in the product mixture decreased slightly
NMR analysis showed that the tacticity of the PMMA obtained
from polymerizations in the presence of ethylene is comparable
to PMMA obtained from free radical MMA homopolymeriza-
(ethylene polymerization was more productive in comparison).
Me
No significant further conversion was observed for 1 after ca.
PPh3
Ph
PPh3
40 min (Figure 2a) and for 2
after ca. 20 min (Figure 2b) in
tions (62−65% rr, Table S7, Figure S26). For PMMA obtained
the presence of ethylene, while in homopolymerizations of
R
PPh3
from polymerizations with 1
(R = Me, Ph) at high ethylene
MMA, significant further conversion was observed until 300 min
Ph
PPh3
Ph
PPh3
for 1 and up to 200 min for 2 (Figure S16). Therefore, we
pressures (100 bar, 15 mol % MMA, Figure S27, green), the Tg of
PMMA overlaps with the intensive melting endotherm for PE at
a Tm of about 120 °C. Therefore, the Tg of PMMA was not
observable. For experiments at lower ethylene pressures, PEs
degree of branching (ca. 25/1000 C atoms, see Table S7) and the
conclude that in the presence of ethylene, the free radical
polymerization of MMA proceeds much faster. A further
indication for a faster and more efficient generation of free
radicals in the presence of ethylene is the ratio of PMMA chains
D
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
MMA content (33 mol % at 50 bar, 58 mol % at 25 bar)
increased, resulting in separate observable melting and glass
transitions of PE (ca. 110 °C) and PMMA (ca. 124 °C),
respectively, in the DSC traces (Figure S27). Tm and Tg are in the
range of homo-PE and -PMMA, respectively. This indicates that
either homopolymer mixtures or block copolymers, containing
long block segments derived from ethylene or MMA, were
formed.
Scheme 1. (a) Schematic Representation of the Activation
and Termination of the Ethylene Insertion Chain Growth and
(b) Observed Relative Amounts of Unsaturated (Nunsat) and
c
Saturated (Nsat) End Groups per PE Chain
However, in 1H and 13C{1H} NMR spectra, no characteristic
resonances for a transition between PE and PMMA blocks were
detected, as would be expected for block copolymers.27 Instead,
detailed polymer analysis by extraction experiments, DOSY, and
Me
SEC strongly indicate that under the conditions applied (1
PPh3
or 2RPPh with R = Me, Ph, 70 °C, 25 bar of ethylene, neat MMA,
3
15 μmol of [Ni], and 3 equiv of PPh3) homopolymer mixtures of
PE and PMMA were formed (Figure 3, Figure S30−S34).
a
Arithmetic mean for polymerizations stopped after 5, 10, 30, 60, and
b
c
120 min. Polymerization time of 10 min. Reaction conditions:
solvent volume = 50 mL, 25 bar of E, 15 μmol of [Ni], 3 equiv of
PPh3, 70 °C. Details see Supporting Information.
polymerizations of MMA alone. Therefore, we anticipate that
reaction products of catalyst precursors are involved in the
initiation of the free radical polymerization.
Stoichiometric Reactivity of Ethylene and MMA
toward (N^O)Ni(II)-Catalyst Precursors and Intermediate
Ni Species. To identify the aforementioned Ni species,
reactions of ethylene and MMA with Ni species that are
anticipated to be involved in the polymerization process were
monitored by spectroscopic methods (Scheme 2, NMR and
EPR; detailed description, see Supporting Information).
Figure 3. 1H DOSY (600 MHz, 130 °C, 1,1,2,2-tetrachloroethane-d2) of
a polymer mixture (52 mol % MMA-based repeat units) polymerized
Me
PPh3
with 1 plus 3 equiv of PPh3 in 50 mL of MMA, 25 bar of E, 70 °C, and
30 min reaction time. PMMA resonances (D = 7.8 × 10−10 m2·s−1) and
PE backbone resonance (D = 2.0 × 10−10 m2·s−1) show separate
diffusion traces.
Scheme 2. Summarized Reactivity of [(N^O)Ni(II)R(L)]
Catalyst Precursors and Intermediate Ni Species as Derived
from Stoichiometric Experiments
Molecular weights estimated by SEC (40 °C, THF, vs PS stan-
dards) of the PE fractions are in the range of 3000−6000 g·mol−1
(Mw /Mn ≈ 1.2−1.3) and 60000−110000 g·mol−1
(Mw/Mn ≈ 2−3) of the PMMA fractions. Similar values are
expected for polymers obtained from polymerizations of each
monomer alone under comparable conditions.
a
In addition, detailed NMR studies showed that the relative
amount of saturated end groups per PE chain (Nsat, normalized
to 1 group per PE chain) introduced upon the start of a new chain
in insertion chain growth are almost equal to the relative amount
of unsaturated chain ends (Nunsat) formed via β-H elimination
(Scheme 1). This suggests that the PE chain growth starts by
insertion of ethylene into [Ni−Me]/[Ni−Ph] catalyst precursors
or intermediate [Ni−H] species and is terminated by β-H
elimination, without any impact of MMA. In agreement, the
same ratio of Nunsat to Nsat is obtained in the absence of MMA
(Scheme 1b, entries 1−3 vs 4). This suggests that the insertion
chain growth species [Ni−PE] does not participate in the
generation of free radicals and is not impacted by the growing
PMMA radicals, because this would result in a significantly
different ratio of unsaturated and saturated end groups.
In summary, the key reaction steps of the insertion poly-
merization of ethylene (activation, chain growth, and termination)
seem to proceed without interaction of the simultaneous free
radical MMA polymerization. Both ethylene insertion and radical
polymerization run simultaneously and are not influenced by each
other. However, it was shown that the generation of free radicals is
more feasible in the presence of ethylene, in comparison to
a
For details, see Supporting Information. Gray structures were not
directly observed.
Ph
Ph
[Ni−Ph] complexes (1PPPhh , 2 , and 3 ) inserted MMA
PPh3
PPh3
3
in a 2,1 fashion at 70 °C, followed by a rapid β-H elimination.
[Ni−Me] complexes showed no reactivity toward MMA under
Me
PPh3
the conditions applied (1MDMe SO and 1 , 60 °C). This agrees with
the observed undisturbed ethylene polymerization in the
presence of MMA (vide supra). Therefore, it is concluded that
[Ni−alkyl] complexes do not insert MMA. Reactive [Ni−H]
complex, 1HPMe , which is formed via β-H elimination from
3
[Ni−alkyl] complexes after ethylene or MMA insertion into the
catalyst precursors, is readily able to reversibly insert MMA in a
E
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
1,2 fashion at low temperatures (−35 °C, THF-d8), while at
higher temperatures (greater than −2 °C), β-H elimination as
the reverse reaction is favored. Kinetic studies revealed that the
second order rates for the insertion of MMA and ethylene15 into
intermediate [Ni−H] species are on a similar order and can
compete under polymerization conditions. No consecutive
insertions of MMA were observed. This was accounted for by
a fast β-H elimination in comparison to a second insertion of
MMA into [Ni−alkyl] complexes.
The stoichiometric experiments also provide insights into
decomposition pathways: Reductive elimination of methyl
moieties to the labile phosphine ligand (PPh3) led to methyl
phosphonium salt formation. Furthermore, reductive elimina-
tion from intermediately formed [Ni−H] species to the N^O
ligand was observed. Previously, similar decomposition reactions
for other [Ni−H] species were described by Jenkins and
Brookhart28 and Grubbs et al.14 In both reductive elimination
pathways, Ni(0) species are expected to be formed. In addition,
decomposition via bimolecular reductive coupling was observed,
which proceeded even under mild conditions in the presence of
[Ni−H] species, or under more harsh conditions from catalyst
precursor complexes (ethane formation from [Ni−Me]).
Herein, we conclude from observed reactivities, that two Ni
complexes, [Ni(II)−R] and [Ni(II)−R′], react to form two
intermediate [Ni(I)] species and release the coupled product
R−R′ (Scheme 2, center). This intermediate species was directly
observed in a reaction mixture of [(N^O)iPr,tBu‑HNiPh(PPh3)]
and ethylene by spectroscopic methods (Figure 4, right).
Sources of Organic Free Radicals in (N^O)Ni(II)R(L)
Mediated Polymerizations. We have elucidated that neither
the organic moiety of the catalyst precursor nor the growing alkyl
PE chain at the Ni center is involved in the initiation process of
the free radical polymerization. Two possible initiation reactions
for free radical polymerizations, which are supposed from an
overall view of the observed reactivity, will be discussed in the
following:
As a starting point, end group analysis of PMMA formed
13CH3
with 1PPh and PPh3 revealed phenyl end groups instead of the
3
anticipated methyl end groups and thus indicated the aryl
phosphine as a source of initiating organic radicals (Figure 1).
Also, additional phosphine was shown to have a crucial impact on
the efficiency of the radical polymerization (vide supra). Tertiary
phosphines are widely used as labile ligands and are often
employed in homogeneous catalysis. They are usually considered
to remain intact during catalysis; however, a number of examples
have been reported where this is not the case.29 Various pathways
for a transfer of an aryl group from P(aryl)3 to the PMMA formed
are conceivable, and we propose a homolytic P−C bond cleavage
as the most plausible pathway (additional pathways are discussed
in the Supporting Information).
Photodriven P−C bond cleavage was reported for PPh3,30 and
tri-o-tolylphosphine was even described as a rapid free radical
photopolymerization initiator.31 Ru mediated homolytic cleav-
age of a P−C bond driven by light was reported recently and used
for the initiation of a free radical polymerization of MMA.32
A thermal P−C bond cleavage is considered, since all poly-
merizations were conducted under exclusion of light. Thermal
Li-mediated homolytic P−C bond cleavages are known for
P−aryl and P−alkyl bonds, and mechanistic studies indicate that
prior to the P−C bond cleavage the metal is oxidized (Li0 to Li+)
and the phosphine is reduced.33 A similar reactivity for Cr(0)
complexes has been explored by computational methods by
Espinosa et al.34 A reduction of PPh3 by Ni species requires a
ready oxidation of these metal species. Therefore, we investigated
the efficiency of low-valent Ni(0/I) species, which can form
during polymerization, to initiate free radical polymerizations in
the presence of phosphines. However, MMA homopolymeriza-
tions initiated with [Ni(0)(PPh3)4] did not result in the for-
mation of Ph end groups and an induction period was observed,
which was not the case when [(N^O)Ni(II)R(PPh3)] complexes
were employed as initiators (Figure S65 vs Figure S16). Also,
[(N^O)iPr,tBu‑HNi(I)(PPh3)2] did not efficiently initiate a radical
polymerization (16.3 mmol·L−1 of [Ni], neat MMA, 70 °C, 1 h,
Table S9, entry 6). It can be concluded, that low-valent Ni species
are not involved in the homolytic P−C bond cleavage.
To elucidate the nature of the initial free radical more directly,
spin trapping experiments were performed with 1MDMe SO, an excess of
PPh3, and α-phenyl-N-tert-butylnitrone (PBN) at 70 °C in toluene
(Figure 5, top). After 18 h a distinct EPR resonance of a nitrosyl
radical was detected. Comparison with separately synthesized Me−
and Ph−PBN spin adducts35 indicates that the phenyl spin adduct
was formed (Figure S61 and S62). The phosphine free system,
1MDMe SO, under otherwise identical conditions led to formation of a
different, unassigned PBN spin adduct (Figure S64). In summary,
we believe this pathway is the most likely and propose a thermal
metal-mediated homolytic P−C bond cleavage for aryl phosphine
ligands in [(N^O)Ni(II)R(PAr3)] complexes.
Figure 4. Identification of the paramagnetic [Ni(I)] complex
Ph
PPh3
[(N^O)iPr,tBu‑HNi(I)(PPh3)2] formed during the reaction of 3
with
ethylene. ORTEP (50% probability ellipsoids, hydrogens are omitted for
clarity) plot determined by X-ray diffraction is shown left, and X band
EPR spectra (9.36 GHz, toluene, −170 °C) of the reaction mixture of
Ph
3
PPh3
(27 mM) and 9 equiv of ethylene after 48 h at room temperature
PPh3
(black) and of separately synthesized 3
(blue) on the right.
PPh3
The corresponding [Ni(I)] species, [(N^O)iPr,tBu-HNi(I)(PPh3)2],
which was independently synthesized and characterized,
represents the first isolated [(N^O)Ni(I)] complex. This
complex underwent further disproportionation to finally form
a bis-chelated [(N^O)2Ni(II)] complex and the corresponding
[Ni(0)] species. In the presence of MMA and PPh3 complex
[(PPh3)2Ni(0)(η2-H2CC(CH3)COOMe)] was identified as a
main [Ni(0)] species (Figure S49−51). These results give useful
insights into possible reaction pathways and products that can be
formed during the simultaneous ethylene and MMA polymer-
ization. However, the source of organic free radicals still remains
elusive at this point.
A second possible pathway to initiate a free radical
polymerization of MMA is indicated by a higher efficiency of
Ph
(N^O)Ni complexes with iodo substituted N^O ligands (1
PPh3
Ph
Ph
and 2 ) in comparison to iodo free complexes (3 , Tables 1
PPh3
PPh3
F
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
reaction of [Ni(0)]/[Ni(I)] with iodo substituted N^O ligands
(1 and 2). This is a rather specific case, because these ligands
Ph
are not widely used; however 2
was the subject of a previous
PPh3
study on a dual radical/insertion polymerization.20 A possible
P−C bond cleavage in aryl phosphine ligands as an organic
radical releasing reaction is a somewhat more general path-
way since phosphines are widely used in organometallic catalytic
systems.
Two possible pathways for the initiation of a free radical
polymerization were suggested; however, we want to emphasize
that these do not represent the only possible pathways for an
initiation of a radical polymerization. For example, iodine and
phosphine free systems were also found to be able to initiate
radical polymerizations ([(N^O)CF,Ant‑HNiMe(pyridine)],
Figure S67, red). These further pathways were not subject of
the present study.
Figure 5. CW X-band (9.37 MHz, RT) EPR spectra of the trapped
Me
radical of the reaction of 1
(4 mM) with 4 equiv of PPh3 and 13
DMSO
equiv of PBN at 70 °C after 18 h in toluene (1) and of the separately
synthesized phenyl PBN spin adduct (2).
Reactivity of Ni(I) toward Organic Radicals. Although the
intermediately formed [(N^O)Ni(I)] species do not initiate free
radical polymerizations, their reactivity toward organic radicals is
of interest, since metal centered radicals are reactive species in
and 2). A significant discrepancy was also noticed in stoichiometric
NMR experiments. Observing the reactivity of [(N^O)iPr,tBu‑H
-
Ni(II)Ph(PPh3)] toward MMA ultimately led to formation of
[{(N^O)iPr,tBu‑H}2Ni(II)] and [(PPh3)2Ni(0)η2-MMA] (vide
supra, Figure S49). In contrast, during NMR experiments with
iodo substituted N^O ligands (1 and 2), no [Ni(0)] species
could be detected, although various decomposition pathways
should lead to the formation of [Ni(0)] species. Therefore, the
reactivity of Ar−I functionalities and [Ni(0)] species was
examined more closely. Tsou and Kochi36 found that the
oxidative addition of aryl halides to [Ni(0)] can be separated into
two distinct one electron transfer steps. In the first step, the aryl
halide is reduced. In the following step, either the oxidative
addition product [(L)nNi(II)X(Ar)] is formed or an aryl radical
escapes the solvation sphere and [(Ln)Ni(I)X] is formed along
with an organic free radical. Free radical polymerizations initiated
by low-valent Ni species and organic halides are also known.37
Indeed, [Ni(0)(PPh3)4] with addition of iodo aryls effectively
initiated a radical polymerization of MMA (neat MMA, 70 °C,
0.5 mmol·L−1, 64 equiv of PhI, 1.2 chains/[Ni], 5 h, Table S9,
entry 2). [(PPh3)3Ni(I)Cl] showed a similar reactivity (neat
MMA, 70 °C, 1.1 mmol·L−1, 33 equiv of PhI, 1.5 chains/[Ni],
7 h, Table S9, entry 5) under comparable reaction conditions.
These results clearly show that isolated low-valent Ni(0/I)
species effectively initiate free radical MMA polymerizations in
the presence of Ar−I functionalities. To verify whether this
pathway is also accessible from [(N^O)Ni(II)R(L)] systems,
40 equiv of PhI was added to a polymerization mixture of iodo
controlled radical polymerizations.24
PPh3
PPh3
Upon reaction of [(N^O)iPr,tBu‑HNi(I)(PPh3)2] (3 ) with
15 equiv of phenylazo-triphenylmethane (PAT, Scheme 3), an
Scheme 3. Combination of Organic and Metal Centered
Radical Species
immediate formation of [(N^O)iPr,tBu‑HNi(II)Ph(PPh3)] and
noncoordinated PPh3 was observed by NMR spectroscopy
PPh3
PPh3
(RT, C6D6, Figures S68 and S69). 3 also reacted rapidly with
AIBN; however, no [Ni−alkyl] species could be identified, which
was accounted for by rapid β-H elimination. These reactions
were much faster than the expected thermally induced decom-
position of PAT39 or AIBN40 to free radicals, and therefore the
reaction seems to be metal mediated. We hypothesized that this
observed reactivity can be used to capture intermediate [Ni(I)]
complexes, formed via bimolecular reductive coupling, to
reestablish a catalytically active species directly in the insertion
polymerization mixture by the addition of a free radical source
(e.g., azo compounds) to prevent further decomposition.
As a proof of principle, we carried out polymerizations
free Ni(II) salicylaldiminato complex [(N^O)CF,Ant‑H
-
NiMe(pyridine)] (4 equiv of PPh3 and neat MMA). Indeed,
increased conversions from ca. 3% to 5% were observed (70 °C,
c([Ni]) = 0.5 mmol·L−1, 5 h, Figure S67). Hence, it seems likely
that the in situ generated low-valent Ni-species together with
iodo substituted N^O ligands contribute to the formation of
organic free radicals. According to MALDI-TOF MS and NMR
analysis, no end groups were transferred to the formed PMMA
polymer by this initiation pathway and the formed end group is
independent of Ar and X (Figure S66, red = 3,5-dimethoxy-
bromobenzene and black = (N^O)iPr,I−I as Ar−X). We believe,
that the intermediate radical formation proceeds in the ligand
sphere of [(PPh3)2Ni(η2-H2CC(CH3)COOMe)] (vide
supra), and here, for example, a hydrogen abstraction from the
α-CH3 group of MMA38 is more likely than the addition of the
organic radical to the double bond of MMA.
of ethylene in the presence of azo compounds with complex
Ph
PPh3
2
(Table 3). Addition of TAP or AIBN led to a significant
Table 3. Ethylene Polymerizations with Catalyst Precursor
a
Ph
2
with or without Addition of Azo Compounds at 25 °C
PPh3
b
c
d
entry
equiv add.
yield, g
lifetime, min
TOF
Tm, °C
1
2
3
4
5
0
0.92
1.87
2.28
5.77
6.57
30
50
8750
10670
10840
9140
132
132
132
132
132
2 PAT
2 AIBN
4 AIBN
6 AIBN
60
>180
>190
9860
a
Reaction conditions: 100 mL of toluene, 7.5 μmol of [Ni], 25 bar of
b
In summary, the major initiation reaction for radical
polymerizations in the (N^O)Ni(II)R(P(aryl)3) systems studied
in the simultaneous ethylene and MMA polymerization is the
E, 25 °C. Lifetime is determined when no further consumption of
c
ethylene is detected by mass flow meter. n(E)·n([Ni])−1·h−1
.
d
Determined by DSC.
G
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 4. Schematic Overview of the Reactivity of [(N^O)Ni(II)R(L)] Complexes in the Presence of Monomers (Ethylene and
a
MMA)
a
Different pathways are color-coded. Blue describes migratory insertion polymerization, orange describes formation of [Ni(I)] and [Ni(0)] species,
green describes homolytic P−C-bond cleavage, red describes one electron oxidation of [Ni(0/I)] species, and purple describes formation of
[Ni(II)(alkyl/aryl)] from Ni(I) species.
increase in yield and lifetime in the ethylene polymerization.
The lifetime increased from 30 to >190 min in case of 0 or 6 equiv
of AIBN added (Table 3, entry 1 vs 5). At the same time, the
turnover frequencies (TOF) and the microstructure of the
resulting PE polymers (concluded from Tm) remained
unchanged, indicating that the same active species was present
in the reaction mixtures (for a detailed discussion, see Supporting
Information).
In summary, [(N^O)Ni(I)] species that are formed during the
polymerization via bimolecular reductive coupling can be
transformed into the corresponding catalytically active [(N^O)-
Ni(II)R] species by reaction with organic radicals.
ization was observed, resulting in homopolymer mixtures of
PE and PMMA. However, the presence of MMA led to de-
creased productivities in the ethylene polymerizations, which is
accounted for by an inhibition of insertion polymerization by
MMA as a weakly coordinating solvent. (2) Concerning the
reactivity of [(N^O)Ni(II)R(L)] precursors (R = Me, Ph) and
active species (R = H) toward MMA, it was found that [Ni−H]
(1,2 insertion mode) and [Ni−Ph] (2,1 insertion mode)
complexes insert MMA to form thermally unstable [Ni−alkyl]
complexes. These undergo rapid β-H eliminations instead of a
further insertion of monomer species, which is believed to be
suppressed by chelate formation by coordination of the inserted
monomers’ carbonyl group.14−16 Interestingly, [Ni−Me] cata-
lyst precursors, which are also a model for the growing PE chain,
did not insert MMA under the conditions applied. Instead,
these complexes undergo decomposition by either reductive
CONCLUDING REMARKS
■
To date, reported observations of the reactivity of MMA in
Ni(II)-catalyzed polymerizations provided an inconclusive
picture. To elucidate the role of free radicals in these reactions,
we have performed (a) synthesis of different (N^O)Ni(II)R(L)
catalyst precursors (R = Ph, Me), intermediate species, and
decomposition products, (b) homopolymerizations of MMA
and polymerizations of ethylene in the presence of MMA, (c)
detailed analysis of the polymers formed (labeling experiments,
end group analysis, microstructure analysis), and (d) stoichio-
metric experiments monitored via spectroscopic methods (NMR
and EPR). Combining these methods, the following picture
emerged (Scheme 4). (1) The migratory insertion polymeri-
zation of ethylene (Scheme 4, blue) is virtually unaffected by a
simultaneous radical polymerization of MMA. Ethylene chain
growth is initiated by insertion of ethylene into [Ni−Me]/
[Ni−Ph] catalyst precursors or into intermediately formed
[Ni−H] species and is mainly terminated by β-H elimina-
tion (conclusive with end group analysis). The growing
[Ni−polymer] species is not relevant for the formation of free
radicals that initiate the radical chain growth of MMA.
Furthermore, no interaction of insertion and radical polymer-
+
elimination to the labile phosphine ligand (to form MePPh3 ) or
bimolecular reductive coupling with formation of ethane
(Scheme 4, orange). In addition, supplying 13C labeled
[Ni−13CH3] as an initiator in MMA homopolymerizations did
not lead to an incorporation of the labeled group into the PMMA
polymer chain. [Ni−Me] and [Ni−Ph] complexes showed very
similar kinetic behavior for MMA homopolymerizations,
indicating that an insertion of MMA into the Ni−C bond is
not a mandatory reaction in the generation of radicals that initiate
MMA free radical chain growth. (3) As one source for organic
free radicals, we suggest a homolytic P−C bond cleavage in labile
aryl phosphine ligands (Scheme 4, green). Phenyl terminated
PMMA polymers were formed when the radical polymerization
was initiated with [(N^O)CF,I−INi(13CH3)PPh3] and 3 equiv of
PPh3. The sole possible phenyl source herein is PPh3. By
variation of the phosphine ligand to P(mTol)3, a variation of the
1
polymer end group pattern in the H NMR spectrum was
observed. In addition, spin trapping EPR experiments have
Me
shown that 1
with PBN led to the Ph−PBN spin adduct
PPh3
(Figure 5). Although this aspect was not studied in detail, we
H
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Ph
believe the P−C bond cleavage is metal-mediated and occurs
from the initial [(N^O)Ni(II)R(PPh3)] complex. Also, P−C
bond cleavage could be an explanation for higher efficiency for
radical polymerizations with [Ni] complexes when additional
phosphine is added, which was reported repeatedly previ-
ously20,41 and was also observed in our experiments. (4) A
second radical source, much more specific to the nickel
complexes under investigation, is one electron oxidation/single
electron transfer from intermediately formed [Ni(0)] species to
aryl halides, for example, the aryl iodide moiety of the N^O
ligand, whereby metal-centered Ni(I) and ligand-based aryl
radicals are formed (Scheme 4, red). (5) With respect to the
generation of [Ni(0)] species, we have identified several reaction
pathways: (a) reductive elimination of (N^O)H from (N^O)Ni
hydride complexes formed during chain transfer in the ethylene
Structure of 3
(CCDC Number: 1432841) (CIF)
PPh3
Synthetic procedures, characterizations of complexes,
spectroscopic data for key experiments, and experimental
and calculation details (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: stefan.mecking@uni-konstanz.de.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was financially supported by the DFG (Grant Me
1388/10-2). The authors thank Michael Gießl and Julia
Zimmerer for performing pressure reactor experiments, Anke
Friemel and Ullrich Haunz for technical assistance with NMR
measurements, Lars Bolk for DSC and SEC measurements, and
Evelyn Wuttke for technical support with EPR measurements.
+
polymerization; (b) reductive elimination of MeP(aryl)3 from
(N^O)Ni methyl phosphine complexes; (c) a relevant
alternative pathway, especially at high [Ni] concentrations as
present in NMR experiments, is a bimolecular reductive coupling
([Ni−R] + [Ni−R′] → R-R′ + 2 [Ni(I)]). We have studied this
reaction in detail for [Ni−H] species (Scheme 4, orange top
pathway) and identified the intermediate formation of [(N^O)-
Ni(I)(PPh3)2], which underwent a further disproportionation
reaction ultimately leading to the observed formation of
[(N^O)2Ni(II)] and [Ni(0)(PPh3)4]. This pathway can proceed
under mild conditions with involvement of [Ni−H] species and
is believed to proceed in a similar manner under more harsh
conditions also for [Ni−Me] and [Ni−Ph] complexes. In
simultaneous polymerizations of ethylene and MMA, the
initiation process was found to be significantly faster than in
polymerizations of MMA alone. This is traced to a fast ethylene
insertion into the catalyst precursors, which yields [Ni−alkyl]
complexes. These readily undergo β-H elimination, and reactive
[Ni−H] species form, which decompose by bimolecular
reductive coupling and finally form [Ni(0)] species. These
[Ni(0)] species react with aryl halide functionalities in the
specific (N^O) ligands and initiate a free radical chain growth.
(6) Intermediately formed [Ni(I)] species may be transformed
into insertion polymerization active Ni(II) (aryl/alkyl) prior to
disproportionation into Ni(0) complexes and polymerization
inactive bis(chelate) nickel(II) by addition of azo-compounds
(Scheme 4, purple). By using this strategy, the overall
productivity of an exemplified catalyst studied here has been
increased by a factor of 6 in ethylene polymerizations due to
regeneration of insertion polymerization active species. It has to
be emphasized that an analogous recombination of [Ni(I)]
species with growing PMMA radicals cannot be excluded.
However, since the insertion products of MMA into [Ni−H]
and [Ni−Ph] undergo rapid β-hydride elimination rather than
ethylene insertion, the formation of PMMA−PE block
copolymers is precluded.
REFERENCES
■
(1) Nakamura, A.; Ito, S.; Nozaki, K. Chem. Rev. 2009, 109, 5215−44.
(2) (a) Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc.
1996, 118, 267−8. (b) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart,
M. J. Am. Chem. Soc. 1998, 120, 888−99.
(3) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh, R. I.
Chem. Commun. 2002, 744−5.
(4) Nakamura, A.; Anselment, T. M. J.; Claverie, J.; Goodall, B.; Jordan,
R. F.; Mecking, S.; Rieger, B.; Sen, A.; van Leeuwen, P. W. N. M.;
Nozaki, K. Acc. Chem. Res. 2013, 46, 1438−49.
(5) Kochi, T.; Noda, S.; Yoshimura, K.; Nozaki, K. J. Am. Chem. Soc.
2007, 129, 8948−9.
(6) Ito, S.; Munakata, K.; Nakamura, A.; Nozaki, K. J. Am. Chem. Soc.
2009, 131, 14606−7.
(7) Runzi, T.; Frohlich, D.; Mecking, S. J. Am. Chem. Soc. 2010, 132,
̈
̈
17690−1.
(8) Guironnet, D.; Roesle, P.; Runzi, T.; Gottker-Schnetmann, I.;
Mecking, S. J. Am. Chem. Soc. 2009, 131, 422−3.
(9) Runzi, T.; Guironnet, D.; Gottker-Schnetmann, I.; Mecking, S. J.
̈
̈
̈
̈
Am. Chem. Soc. 2010, 132, 16623−30.
(10) (a) Wang, C. M.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs,
R. H.; Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3149−51.
(b) Johnson, L. K.; Bennett, A. M. A.; Ittel, S. D.; Wang, L.;
Parthasarathy, A.; Hauptman, E.; Simpson, R. D.; Feldman, J.; Coughlin,
E. B. (E. I. du Pont de Nemours & Co., USA) International Patent WO
98/30609, 1998.
(11) (a) Zuideveld, M. A.; Wehrmann, P.; Rohr, C.; Mecking, S. Angew.
Chem. 2004, 116, 887−91. (b) Tong, Q.; Krumova, M.; Gottker-
̈
̈
Schnetmann, I.; Mecking, S. Langmuir 2008, 24, 2341−7. (c) Osichow,
A.; Rabe, C.; Vogtt, K.; Narayanan, T.; Harnau, L.; Drechsler, M.;
Ballauff, M.; Mecking, S. J. Am. Chem. Soc. 2013, 135, 11645−50.
(12) Younkin, T. R.; Connor, E. F.; Henderson, J. I.; Friedrich, S. K.;
Grubbs, R. H.; Bansleben, D. A. Science 2000, 287, 460−2.
(13) Connor, E. F.; Younkin, T. R.; Henderson, J. I.; Hwang, S.;
Grubbs, R. H.; Roberts, W. P.; Litzau, J. J. J. Polym. Sci., Part A: Polym.
Chem. 2002, 40, 2842−54.
ASSOCIATED CONTENT
* Supporting Information
■
S
(14) Waltman, A. W.; Younkin, T. R.; Grubbs, R. H. Organometallics
2004, 23, 5121−3.
(15) Berkefeld, A.; Drexler, M.; Moller, H. M.; Mecking, S. J. Am.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.5b08612.
̈
Chem. Soc. 2009, 131, 12613−22.
Structure of 1PPPhh (CCDC Number: 1432837) (CIF)
(16) Berkefeld, A.; Mecking, S. J. Am. Chem. Soc. 2009, 131, 1565−74.
(17) He, X.; Wu, Q. Appl. Organomet. Chem. 2006, 20, 264−71.
(18) Li, X.-F.; Li, Y.-G.; Li, Y.-S.; Chen, Y.-X.; Hu, N.-H.
Organometallics 2005, 24, 2502−10.
3
Ph
Structure of 2 (CCDC Number: 1432838) (CIF)
PPh3
Structure of {3-3} (CCDC Number: 1432839) (CIF)
Structure of [(N^O)iPr,tBu‑HNi(μ2-OH)2Ni(N^O)iPr,tBu‑H
]
(19) (a) Rodriguez, B. A.; Delferro, M.; Marks, T. J. J. Am. Chem. Soc.
2009, 131, 5902−19. (b) Rodriguez, B. A.; Delferro, M.; Marks, T. J. J.
Am. Chem. Soc. 2013, 135, 17651.
(CCDC Number: 1432840) (CIF)
PPh3
Structure of 3 (CCDC Number: 1432842) (CIF)
PPh3
I
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(20) (a) Leblanc, A.; Grau, E.; Broyer, J.-P.; Boisson, C.; Spitz, R.;
Monteil, V. Macromolecules 2011, 44, 3293−301. (b) Leblanc, A.;
Broyer, J.-P.; Boisson, C.; Spitz, R.; Monteil, V. Pure Appl. Chem. 2012,
84, 2113−20.
(21) (a) Carlini, C.; Martinelli, M.; Raspolli Galletti, A. M.; Sbrana, G.
J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 2117−24. (b) Carlini, C.;
De Luise, V.; Martinelli, M.; Raspolli Galletti, A. M.; Sbrana, G. J. Polym.
Sci., Part A: Polym. Chem. 2006, 44, 620−33.
(22) (a) Gottker-Schnetmann, I.; Wehrmann, P.; Rohr, C.; Mecking, S.
̈
̈
Organometallics 2007, 26, 2348−62. (b) Wehrmann, P.; Mecking, S.
Organometallics 2008, 27, 1399−408.
(23) (a) Bevington, J. C.; Melville, H. W.; Taylor, R. P. J. Polym. Sci.
1954, 12, 449−59. (b) Bevington, J. C.; Melville, H. W.; Taylor, R. P. J.
Polym. Sci. 1954, 14, 463−76.
(24) (a) Ouchi, M.; Terashima, T.; Sawamoto, M. Chem. Rev. 2009,
109, 4963−5050. (b) Poli, R. Eur. J. Inorg. Chem. 2011, 1513−30.
(25) Collins, S.; Ward, D. G. J. Am. Chem. Soc. 1992, 114, 5460−2.
(26) (a) Hatada, K.; Kitayama, T.; Masuda, E. Polym. J. 1986, 18, 395−
402. (b) Kashiwagi, T.; Inaba, A.; Brown, J. E.; Hatada, K.; Kitayama, T.;
Masuda, E. Macromolecules 1986, 19, 2160−8.
(27) Hiller, W.; Pasch, H.; Macko, T.; Hofmann, M.; Ganz, J.; Spraul,
M.; Braumann, U.; Streck, R.; Mason, J.; Van Damme, F. J. Magn. Reson.
2006, 183, 290−302.
(28) Jenkins, J. C.; Brookhart, M. J. Am. Chem. Soc. 2004, 126, 5827−
42.
(29) (a) Garrou, P. E. Chem. Rev. 1985, 85, 171−85. (b) Parkins, A. W.
Coord. Chem. Rev. 2006, 250, 449−67. (c) Macgregor, S. A. Chem. Soc.
Rev. 2007, 36, 67−76. (d) Albrecht, M. Chem. Rev. 2010, 110, 576−623.
(e) Crabtree, R. H. Chem. Rev. 2015, 115, 127−50.
(30) (a) Horner, L.; Dorges, J. Tetrahedron Lett. 1965, 6, 763−7.
̈
(b) Kaufman, M. L.; Griffin, C. E. Tetrahedron Lett. 1965, 6, 769−72.
(c) Hwang, S. J.; Powers, D. C.; Maher, A. G.; Nocera, D. G. Chem. Sci.
2015, 6, 917−22.
(31) Eldred, R. J. J. Polym. Sci., Part A-1: Polym. Chem. 1969, 7, 265−8.
(32) Versace, D.-L.; Cerezo Bastida, J.; Lorenzini, C.; Cachet-Vivier,
́
C.; Renard, E.; Langlois, V.; Malval, J.-P.; Fouassier, J.-P.; Lalevee, J.
Macromolecules 2013, 46, 8808−15.
(33) Dogan, J.; Schulte, J. B.; Swiegers, G. F.; Wild, S. B. J. Org. Chem.
2000, 65, 951−7.
(34) Espinosa, A.; Gom
7250−6.
́
ez, C.; Streubel, R. Inorg. Chem. 2012, 51,
(35) Kotake, Y.; Janzen, E. G. J. Am. Chem. Soc. 1988, 110, 3699−701.
(36) Tsou, T. T.; Kochi, J. K. J. Am. Chem. Soc. 1979, 101, 6319−32.
(37) (a) Uegaki, H.; Kamigaito, M.; Sawamoto, M. J. Polym. Sci., Part A:
Polym. Chem. 1999, 37, 3003−9. (b) Grishin, D. F.; Valetova, N. B.;
Il’ichev, I. S.; Prokhorova, M. G.; Beletskaya, I. P. Russ. Chem. Bull. 2006,
55, 2106−8. (c) Valetova, N. B.; Il’ichev, I. S.; Grishin, D. F. Russ. J. Appl.
Chem. 2010, 83, 895−900.
(38) Mavroudakis, E.; Cuccato, D.; Moscatelli, D. J. Phys. Chem. A
2014, 118, 1799−806.
(39) Alder, M. G.; Leffler, J. E. J. Am. Chem. Soc. 1954, 76, 1425−7.
(40) Sasaki, H.; Nagayama, M. J. Appl. Polym. Sci. 1967, 11, 2097−108.
́
(41) Moineau, G.; Minet, M.; Dubois, P.; Teyssie, P.; Senninger, T.;
́ ̂
Jerome, R. Macromolecules 1999, 32, 27−35.
J
DOI: 10.1021/jacs.5b08612
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX