Monofunctional hyperbranched ethylene oligomers

Article abstract of DOI:10.1021/ja411945n

The neutral κ²N,O-salicylaldiminato Ni(II) complexes [κ²N,O-{(2,6-(3′,5′-R₂C ₆H₃)₂C₆H₃-Ni -C(H)-(3,5-I₂-2-O-C₆H₂)}]NiCH ₃(pyridine)] (1a-pyr, R = Me; 1b-pyr, R = Et; 1c-pyr, R = iPr) convert ethylene to hyperbranched low-molecular-weight oligomers (M_n ca. 1000 g mol^-1) with high productivities. While all three catalysts are capable of generating hyperbranched structures, branching densities decrease significantly with the nature of the remote substituent along Me > Et > iPr and oligomer molecular weights increase. Consequently, only 1a-pyr forms hyperbranched structures over a wide range of reaction conditions (ethylene pressure 5-30 atm and 20-70 C). An in situ catalyst system achieves similar activities and identical highly branched oligomer microstructures, eliminating the bottleneck given by the preparation and isolation of Ni-Me catalyst precursor species. Selective introduction of one primary carboxylic acid ester functional group per highly branched oligoethylene molecule was achieved by isomerizing ethoxycarbonylation and alternatively cross metathesis with ethyl acrylate followed by hydrogenation. The latter approach results in complete functionalization and no essential loss of branched oligomer material and molecular weight, as the reacting double bonds are close to a chain end. Reduction yielded a monoalcohol-functionalized oligomer. Introduction of one reactive epoxide group per branched oligomer occurs completely and selectively under mild conditions. All reaction steps involved in oligomerization and monofunctionalization are efficient and readily scalable.

Full text of DOI:10.1021/ja411945n

Article
pubs.acs.org/JACS
Monofunctional Hyperbranched Ethylene Oligomers
Thomas Wiedemann, Gregor Voit, Alexandra Tchernook, Philipp Roesle, Inigo Gottker-Schnetmann,
̈
and Stefan Mecking*
Chair of Chemical Materials Science, Department of Chemistry, University of Konstanz, 78464 Konstanz, Germany
S
* Supporting Information
ABSTRACT: The neutral κ2N,O-salicylaldiminato Ni(II) com-
plexes [κ2N,O-{(2,6-(3′,5′-R2C6H3)2C6H3-NC(H)-(3,5-I2-2-
O-C6H2)}]NiCH3(pyridine)] (1a-pyr, R = Me; 1b-pyr, R = Et;
1c-pyr, R = iPr) convert ethylene to hyperbranched low-
molecular-weight oligomers (Mn ca. 1000 g mol−1) with high
productivities. While all three catalysts are capable of generating
hyperbranched structures, branching densities decrease signifi-
cantly with the nature of the remote substituent along Me > Et
> iPr and oligomer molecular weights increase. Consequently,
only 1a-pyr forms hyperbranched structures over a wide range of reaction conditions (ethylene pressure 5−30 atm and 20−70
°C). An in situ catalyst system achieves similar activities and identical highly branched oligomer microstructures, eliminating the
bottleneck given by the preparation and isolation of Ni−Me catalyst precursor species. Selective introduction of one primary
carboxylic acid ester functional group per highly branched oligoethylene molecule was achieved by isomerizing
ethoxycarbonylation and alternatively cross metathesis with ethyl acrylate followed by hydrogenation. The latter approach
results in complete functionalization and no essential loss of branched oligomer material and molecular weight, as the reacting
double bonds are close to a chain end. Reduction yielded a monoalcohol-functionalized oligomer. Introduction of one reactive
epoxide group per branched oligomer occurs completely and selectively under mild conditions. All reaction steps involved in
oligomerization and monofunctionalization are efficient and readily scalable.
polymerization.5 Thus, particularly with the Pd catalysts
ethylene is polymerized to a highly branched, high-molecular-
INTRODUCTION
■
Beyond the well-known applications of polyethylenes such as
weight, entirely amorphous rubbery material. These even
HDPE and LDPE as thermoplastic materials, lower molecular
contain branches on branches,: that is, a hyperbranched6
weight ethylene oligomers can serve numerous functions.
structure (also cf. Figure 2 and Scheme 2, left).5,7,8 In these
Concerning their synthesis, the Ziegler Alfol process is an
illustrative and practically important example.1 Chain growth
catalysts, bulky substituents on the diimine ligand are
responsible for the formation of high-molecular-weight
on aluminum yields longer chain aluminum alkyls. Subsequent
polymer.9 Shielding of the apical positions of the square-planar
oxidation and hydrolysis affords linear long-chain terminal
alcohols, stoichiometric in aluminum. Chain growth on
metal centers suppresses chain transfer. At the same time, these
bulky substituents also enhance chain growth rates. This has
metallocenes with simultaneous chain transfer reactions is
employed for the production of linear polyethylene waxes.2
been ascribed to a destabilization of the ground state of the
resting state.5,10 Consequently, less bulky substituted Pd
These are also functionalized by oxidation under severe
conditions, to yield linear oligoethylenes with an undefined
diimine catalysts produce lower molecular weight branched
oligomers with low activities.11
number of a mixture of different functional groups (primarily
carboxylic acids) per chain.2 Precisely chain end functionalized
In addition to the low activity being a limitation of these
Pd(II) oligomerization catalysts, catalysts based on nickel are
often more favorable in terms of abundant availability of the
metal. This is illustrated by the Shell higher olefin process.
linear oligoethylenes are accessible by “chain shuttling”
processes.3,4 As in the Ziegler Alfol process, functionalization
occurs in a second step from main-group-metal or zinc alkyls in
a stoichiometric fashion. Chain growth occurs catalytically on a
smaller amount of transition-metal catalyst.
All of these reported procedures afford linear waxlike
oligoethylenes. Highly branched oligoethylenes would also be
of interest as functional additives: for example, in lubricants or
surface modifiers. Considering possible approaches for their
synthesis, the known capability of late-transition-metal catalysts
Here, neutral Ni(II) catalysts convert ethylene to strictly linear
longer α-olefins.12 Cationic Ni(II) diimine catalysts lacking
bulky o-aryl substituents also oligomerize ethylene to the linear
α-olefins.13 Thus, for this type of catalyst the synthesis of highly
branched material is usually limited to high molecular weights.
For the generation of branched oligoethylenes with neutral
Ni(II) catalysts, κ2N,O-salicylaldiminato complexes are promis-
to produce highly branched high-molecular-weight poly-
ethylenes appears attractive. Cationic Pd− and Ni−α-diimine
catalysts can undergo extensive “chain walking” during
Received: November 28, 2013
Published: January 22, 2014
© 2014 American Chemical Society
2078
dx.doi.org/10.1021/ja411945n | J. Am. Chem. Soc. 2014, 136, 2078−2085

Journal of the American Chemical Society
Article
ing candidates. Complexes with the 2,6-diisopropyl N-phenyl
motif inspired by the aforementioned diimines polymerize
ethylene to moderately branched, high-molecular-weight ma-
terial.14,15 In N-terphenyl-substituted catalysts, substituents (R,
cf. Scheme 1) on the peripheral aromatic rings have a
remarkable effect on the catalytic properties, despite their
remoteness from the active sites.16 Depending on the
substituents, high-molecular-weight linear polyethylene (for R
= CF3) or low-molecular-weight, highly branched oligomers
(for R = CH3) are formed.16 More electron donating
substituents favor branch formation and chain transfer, which
both occur through β-hydride elimination as the underlying
step.16−18 The observed effect of substituents can be related to
very similar barriers of β-hydride elimination (ΔG⧧β‑elim) and
ethylene insertion chain growth (ΔG⧧ins).19 Small relative
changes in ΔG⧧β‑elim and ΔG⧧ins exerted by the electronics of R
can then alter the ratio ΔG⧧ins/ΔG⧧β‑elim to a noticeable extent,
resulting in entirely different materials obtained.18b
complex (9 mg) in toluene (0.2 mL) with pentane. The trans
coordination geometry of the methyl group and the oxygen
donor in Figure 1 agrees with the structures of other
In general, there are very few examples of syntheses of
ethylene oligomers (<5000 g mol−1) with high degrees of
branching, and such materials have been little characterized and
studied.20 Functionalized, much less precisely (mono-)
functionalized materials have not been reported. We herein
give a full account on the efficient synthesis of low-molecular-
weight ethylene oligomers with a hyperbranched micro-
structure and their functionalization to monofunctional
compounds.
Figure 1. X-ray diffraction analysis of the complex 1a-pyr with 50%
probability ellipsoids. Hydrogen atoms are omitted for clarity.
salicylaldiminato complexes reported.14b,16 The deviation of
the nickel atom from the root-mean-square plane defined by
O1, N1, C35, and N2 amounts to 0.0117(15) Å, which
indicates an only minuscule deviation of the coordination
geometry of nickel from planarity.
RESULTS AND DISCUSSION
■
Synthesis of Complexes. Terphenyl amines with different
remote 3′,5′-alkyl substituents were prepared by Suzuki
coupling of the corresponding 3,5-substituted phenylboronic
acid with dibromoaniline. Condensation with 3,5-diiodosalicy-
laldehyde afforded the salicylaldimines. Reaction with 1.2 equiv
of [(tmeda)Ni(CH3)2] and excess pyridine in benzene at
ambient temperature yielded complexes 1a-pyr, 1b-pyr, and 1c-
pyr (Scheme 1). Indicative NMR resonances are found at
Catalytic Oligomerization and Microstructure Anal-
ysis. A study of the influence of reaction conditions on the
oligomerization with 1a-pyr (Table 1) reveals that the
polymerization temperature has a much more pronounced
influence on the molecular weights than the ethylene pressure.
As expected, molecular weights increase for low temperatures
and high pressures, while the degrees of branching decrease
under the same conditions. Molecular weight distributions of
Mw/Mn ≈ 2 indicate a well-behaved single-site polymerization
behavior. Note that GPC analyses are against linear polystyrene
standards; thus, they yield apparent molecular weights. As
expected, they exceed the true Mn value from NMR
spectroscopy, with a ca. 2-fold deviation. Overall, degrees of
branching vary only to a small extent with reaction conditions
in the pressure and temperature range studied, from 74 to 81
branches per 1000 C atoms.
Scheme 1. Complexes 1a-pyr, 1b-pyr, and 1c-pyr with
Different Remote Alkyl Substituents
However, oligomer yields are strongly dependent on reaction
conditions. The highest productivity was observed at
intermediate pressures (20−25 bar) and slightly elevated
temperatures (40−50 °C). Ethylene mass flow traces reveal
that although the activity increases for higher temperatures,
significant catalyst degradation occurs, resulting in lower yields
(cf. the Supporting Information). For lower temperatures the
activity is significantly reduced, but no observable catalyst
decomposition occurred over the period of 5 h studied, while at
70 °C the polymerization came to an end after 3 h (Figure S1,
Supporting Information).
−0.56, −0.54, and −0.54 ppm for the Ni−Me protons for the
catalyst precursors 1a-pyr, 1b-pyr, and 1c-pyr, respectively.
The presence of only one signal for these groups proves the
formation of a single isomer. A single set of resonances is
observed for all remote substituents R in 1a-pyr, 1b-pyr, and
1c-pyr, giving no evidence of hindered rotation in the terphenyl
moiety in solution on the NMR time scale (for full
characterization data cf. the Supporting Information). Single
crystals of 1a-pyr suitable for X-ray diffraction analysis could be
obtained within 1 day at −60 °C after layering a solution of the
Microstructure analysis by 13C NMR revealed the branching
patterns of the oligomers produced (Table 2 and Figure 2).
Remarkably, they contain hyperbranched structures. This is
clearly evidenced by significant amounts of sec-butyl groups as
2079
dx.doi.org/10.1021/ja411945n | J. Am. Chem. Soc. 2014, 136, 2078−2085

Journal of the American Chemical Society
Article
Table 1. Polymerization Results with Complex 1a-pyr under Different Oligomerization Conditions
a
b
c
c
d
entry
p (bar)
T (°C)
yield (g)
TON
Mn(NMR)
Mn(GPC)
Mw/Mn(GPC)
branches/1000 C
1
30
25
20
15
10
5
50
50
50
50
50
50
70
60
50
40
30
20
60
54.2
56.6
51.3
45.5
35.0
22.6
12.8
29.6
51.3
55.0
33.1
13.2
57.2
48400
50500
45800
40600
31300
20200
11400
26400
45800
49100
29600
11800
19300
1300
1300
1300
1200
1100
1000
850
2900
2600
2600
2500
2500
2200
1800
2300
2600
3500
4300
5900
1.7
1.8
1.7
1.7
1.7
1.6
1.5
1.6
1.7
1.7
1.8
1.7
2.0
75
75
77
78
78
81
80
77
77
77
76
74
2
3
4
5
6
7
20
20
20
20
20
20
20
8
1000
1300
1600
2100
2600
1000
9
10
11
12
e
13
1600
77
a
b
Reaction conditions (unless stated otherwise): 40 μmol of 1a-pyr in 200 mL of toluene for 5 h. In units of mol of C2H4 (mol of Ni)−1. Molecular
c
d
1
weights calculated from H NMR intensity ratio of unsaturated end groups vs overall integral. In THF vs polystyrene standards. Degree of
branching calculated from H NMR intensity ratio of methyl groups (corrected for saturated end groups) vs overall integral. Reaction conditions:
e
1
106 μmol of 1a-pyr in 600 mL of toluene for 5 h. 1a-pyr prepared by in situ catalyst synthesis (cf. the Supporting Information).
a
Table 2. Microstructure Analysis with Fractional Amounts of Different Branch Lengths
a
a
a
a
a
entry
p (bar)
T (°C)
branches/1000 C
methyl (%)
ethyl (%)
propyl (%)
C4+ (%)
sec-butyl (%)
1
30
25
20
15
10
5
50
50
50
50
50
50
70
60
50
40
30
20
60
75
75
77
78
78
81
80
77
77
77
76
74
77
68
70
70
68
67
63
57
69
70
71
80
81
65
12
9
4
5
5
4
3
4
4
4
5
3
3
4
5
8
7
7
9
2
3
10
11
10
10
14
7
8
8
4
7
11
10
14
15
8
5
10
9
6
7
20
20
20
20
20
20
20
9
8
12
8
9
10
8
8
10
11
12
12
7
6
6
4
5
6
4
b
13
9
9
9
a
Percentages of different branch lengths can be calculated from the relative intensity ratios of the methyl (1B1, 1B2, 1B3,, Bsec‑butyl) or methine (*b4+)
b
signals of the respective branches in the 13C NMR spectra. 1a-pyr prepared by in situ catalyst synthesis (cf. the Supporting Information).
Figure 2. Typical 13C NMR spectrum of a hyperbranched oligomer. Assignments are numbered according to ref 8,21. Chain ends are assigned with
S1−S4. Branches are labeled as xBy, where y is the branch length and x is the carbon, starting from the methyl end with 1. The methine groups for the
different branch lengths are labeled with *By. A and B are the methyl groups of a sec-butyl branch.
2080
dx.doi.org/10.1021/ja411945n | J. Am. Chem. Soc. 2014, 136, 2078−2085

Journal of the American Chemical Society
Article
the smallest and detectable branch-on-branch motif. Although
the oligomers have similar overall degrees of branching, their
microscopic chain architectures differ significantly. While
oligomers obtained at low temperatures or high pressures
comprise mainly methyl branches and only low amounts of
longer chain branches or branch-on-branch structures, those
obtained at higher temperatures and low pressures show
significantly reduced amounts of these methyl branches in favor
of hyperbranched structures, which are found with up to 15% of
all branches. As already observed for the molecular weights, the
oligomerization temperature also has a more pronounced
influence on chain architecture than ethylene pressure. To our
knowledge, such hyperbranched microstructures are unprece-
dented in this low-molecular-weight regime. These low-
molecular-weight hyperbranched oligomers are viscous liquids
and do not show a melt or crystallization transition during DSC
measurements.
A significant simplification of the catalyst precursor could be
achieved by eliminating the need for isolated, sensitive
[(tmeda)NiMe2] used in the preparation of 1a-pyr to 1c-pyr.
A reaction sequence comprising the reaction of [(tmeda)Ni-
(OAc)2] with methyllithium followed by addition of
salicylaldimine 2a and pyridine and simple evaporation yielded
a catalyst which displayed 70% of the activity with regard to
isolated 1a-pyr and yielded an identical oligomer micro-
structure (cf. entry 13 vs 8 in Tables 1 and 2).
behavior, highly branched oligomers are formed with all of
these electron-donating substituents. In detail, higher molecular
weights and lower degrees of branching are obtained on going
from methyl to ethyl and isopropyl substituents. This is in
agreement with the finding that sterics have an additional,
smaller influence.18b
All three catalysts are capable of producing hyperbranched
oligomer structures, as indicated by the presence of sec-butyl
branches (Table 4). Notwithstanding, catalysts 1b-pyr and 1c-
pyr in particular produce oligomers with distinctively higher
percentages of methyl branches (>80%) and minor amounts of
longer chain branches. sec-Butyl branches, as indicators for a
hyperbranched microstructure, can only be obtained at high
reaction temperatures using these two catalysts. These findings
also show that molecular weights correlate with the overall
branch density as well as the average branch length. As a result,
higher molecular weight oligomers always come with lower
branch densities and almost exclusively comprise methyl
branches. Beneficially, desirable features such as high degrees
of branching and hyperbranched structures as well as low
oligomer molecular weight (ca. 1000) go along with one
another.
Oligomer Functionalization. Considering that chain
termination occurs via β-hydride elimination with the Ni−
salicylaldiminato complexes employed, every oligomer chain
comprises an unsaturated end group.22
Dependence of Oligomer Branching Pattern on
Catalyst Structure. The effect of remote alkyl substituents
on the catalyst and its potential utility for fine tuning of
oligomer microstructures was probed by oligomerizations with
1b-pyr and 1c-pyr as catalyst precurors (Tables 3 and 4).
Therefore, these oligomers exhibit potential for the synthesis
of monofunctional, highly branched ethylene oligomers by
postpolymerization functionalization. Due to the extensive
chain walking of the catalysts the major end groups found for
these ethylene oligomers are internal double bonds and only
minor amounts of terminal double bonds (Figure 3).
Table 3. Comparison of Molecular Weights and Branch
Distributions for Oligomers Obtained with Catalyst
Precursors 1a-pyr, 1b-pyr, and 1c-pyr
a
b
catalyst
p (bar)
T (°C)
Mn(NMR)
branches/1000 C
1a-pyr
1b-pyr
1c-pyr
1a-pyr
1b-pyr
1c-pyr
20
20
20
20
20
20
60
60
60
20
20
20
800
80
73
57
74
65
44
1500
3300
2600
3700
13300
a
Molecular weights calculated from 1H NMR intensity ratios of
b
unsaturated end groups vs overall integrals. Degree of branching
calculated from H NMR intensity ratios of methyl groups (corrected
Figure 3. Representative 1H NMR spectrum of unfunctionalized
1
ethylene oligomer.
for saturated end groups) vs overall integrals.
In terms of a uniform and high reactivity of the products,
terminal functional groups are more preferable than a mixture
of primary, secondary, and tertiary functional groups. A possible
In agreement with the findings that the electronic character-
istics of the remote substituents govern the polymerization
Table 4. Branching Distribution of Ethylene Oligomers Prepared with Catalyst Precursors 1a-pyr, 1b-pyr, and 1c-pyr
a
b
b
b
b
b
catalyst
p (bar)
T (°C)
branches/1000 C
methyl (%)
ethyl (%)
propyl (%)
C4+ (%)
sec-butyl (%)
1a-pyr
1b-pyr
1c-pyr
1a-pyr
1b-pyr
1c-pyr
20
20
20
20
20
20
60
60
60
20
20
20
80
73
57
74
65
44
69
82
90
81
93
95
7
8
6
5
4
2
4
3
1
4
1
2
12
2
8
4
2
4
0
0
1
6
2
1
a
b
1
Degree of branching calculated from H NMR intensity ratios of methyl groups vs complete integrals. Percentage of total branches; calculated
from intensity ratios of the corresponding signals in 13C NMR spectra.
2081
dx.doi.org/10.1021/ja411945n | J. Am. Chem. Soc. 2014, 136, 2078−2085

Journal of the American Chemical Society
Article
approach toward this issue is an isomerization of the internal
double bonds to the chain ends. Isomerizing alkoxycarbonyla-
tion allows for converting internal double bonds very selectively
to terminal ester groups in a one-step reaction in the presence
of carbon monoxide; thus, it appears attractive for the synthesis
of end-functionalized oligomers (Scheme 2).23
Scheme 2. Monofunctionalization by Isomerizing
Alkoxycarbonylation
Figure 4. Proton NMR spectrum of ester-functionalized oligomer,
with the olefinic region before (a) and after functionalization (b)
highlighted.
bonds to occur, since a complete lack of reactivity should rather
be expected to result in a further enrichment by isomerization.
In summary, isomerizing alkoxycarbonylation allows for a
selective terminal postfunctionalization of highly branched
oligomers, though quantitative conversion could not be
achieved.
An alternative approach to ester-functionalized oligomers is
cross metathesis26,27 of the double bonds with functionalized
olefins such as acrylates and subsequent hydrogenation of the
double bond. This is a promising reaction also in terms of large-
scale synthesis, considering the low catalyst loadings required.
However, due to the high amount of internal double bonds
present in the oligomers, the reduction of molecular weights is
a possible complication.
Typically, the methanol or ethanol coreagent also serves as a
solvent. However, this was not suitable for the functionalization
of the apolar oligomers due to their lack of solubility. A mixture
of pentane and ethanol (1:1) proved to be most suited.
Using [(dtbpx)Pd(OTf)2] (dtbpx = 1,2-bis[(di-tert-
butylphosphino)methyl]benzene) as a catalyst precursor,24
functionalization is evidenced by new signals appearing in the
1H NMR spectra at 4.14 and 2.29 ppm (Table 5) using
Table 5. Conversions in the Isomerizing
Alkoxycarbonylation of Different Oligomers
a
b
c
oligomer
Mn(NMR)
branches/1000 C
conversion (%)
Cross metathesis of an ethylene oligomer (molecular weight
1600, 78 branches/1000 C atoms) was studied with 10 equiv of
ethyl acrylate in the presence of 0.1 mol % of Hoveyda-Grubbs
II catalyst (Scheme 3). Quantitative conversion occurred after 4
h at 80 °C, as evidenced by key resonances at 6.97, 5.81, 4.18,
and 2.19 ppm in the 1H NMR spectrum and the absence of any
olefinic resonances of the starting material (cf. the Supporting
Information). Identical results in terms of complete function-
alization were observed for various other oligomers, independ-
ent of their molecular weights and branching densities. The
synthesis was readily scalable and was conducted in 50 g
batches. From these experiments, no limitation for further
scaleup is evident, such as a necessity of excessive solvent
7
850
80
81
75
74
50
35
35
17
6
1000
1300
2600
1
12
a
Reaction conditions: 1 g of oligomer, 4 mol % of [(dtbpx)Pd(OTf)2],
10 mL of pentane/EtOH (1/1), 20 bar of CO, 90 °C, 4 days. Entry
number from Table 1. To the primary ethyl ester.
b
c
different molecular weight oligomers as starting materials.
However, conversion is not complete, as indicated by the
observation of residual double bonds. The highest conversion
to the ester could be obtained using low-molecular-weight
oligomers (850). For higher molecular weights the degrees of
functionalization are significantly lower.
However, the reaction is very selective and the primary ester
is formed exclusively. No secondary esters are formed, as
evidenced by 2D NMR spectroscopy (cf. the Supporting
Information). Residual internal and trisubstituted double bonds
are found after the reaction, while the terminal vinyl groups are
fully converted. Notably, trisubstituted double bonds at a
branching point are enriched during the reaction due to the
high extent of isomerization (Figure 4). They appear poorly
reactive for further conversion and thus represent a limiting
factor for the degree of functionalization. Additionally, the
lower conversions for higher molecular weights indicate
solubility problems of the oligomers in the solvent mixture
used.25
1
volumes or complicated workup procedures. H NMR spectra
and GPC measurements of the oligomers before and after
metathesis reveal that molecular weights did not decrease,
indicating that the internal double bonds are located close to a
chain end. Correspondingly, small α,β-unsaturated ester
fragments should be expected as additional metathesis
products. To further illuminate this issue, all volatiles were
removed after the metathesis reaction under vacuum and
collected by condensation at −196 °C. GC-MS analysis
showed, aside from the self-metathesis product diethyl
fumarate, the formation of the expected fragments from ethyl
but-2-enoate to ethyl oct-2-enoate, with the latter being the
longest fragment observed (Figure 5). These findings show
that, despite the distinct tendency of the catalyst for chain
walking, the major amount of internal double bonds is still
located near the chain end and is not isomerized deep into the
oligomer chain.
A second alkoxycarbonylation of the partially functionalized
products resulted in further conversion of internal double
bonds with conversions similar to those for the first step, while
the amount of the trisubstituted double bonds remains
unaltered. This suggests a slow conversion of these double
Hydrogenation of the double bonds was readily achieved
with an ethyl vinyl ether quenched29 Grubbs I catalyst in
quantitative yield at 40 bar of H2 and 75 °C after 24 h (Scheme
2082
dx.doi.org/10.1021/ja411945n | J. Am. Chem. Soc. 2014, 136, 2078−2085

Journal of the American Chemical Society
Article
Scheme 3. Monofunctionalization with an Ester or Hydroxyl Group, Respectively, via Cross Metathesis
catalyst 3 (Scheme 3), reported by Gusev et al. to convert α,β-
unsaturated esters into the saturated alcohols, was found to be
suited for the hydrogenation of our ester-functionalized
oligomers.30 The hydroxyl-functionalized oligomers could be
obtained quantitatively in a one-step hydrogenation of the
unsaturated ester products of the cross-metathesis reaction after
1 day at 50 bar of H2, as evidenced by NMR spectroscopy.
In terms of further conversion or application of the highly
branched oligomers, for example as macromonomers, reactive
epoxide functional groups are also of strong interest. They can
be reacted with a range of different nucleophiles, giving access
to various different functionalities. Furthermore, epoxidation of
double bonds is well documented and has already been applied
to polymers. In contrast to the aforementioned functionaliza-
tion reactions with terminal functional groups, epoxidation
mainly yields disubstituted (“internal”) epoxides corresponding
to the distribution of double bonds present in the oligomers.
Cobalt-catalyzed epoxidation31,32 with mCPBA or peracetic
acid, respectively, was studied (Scheme 4). After 60 min at
Figure 5. GC trace from GC-MS analysis of the volatile fragments
after cross metathesis.28
3). Metathesis and hydrogenation of the oligomers could also
be performed convienently in a one-pot reaction. An oligomer
(molecular weight 1000, 78 branches/1000 C) was subjected to
cross metathesis with ethyl acrylate in a stirred pressure reactor
at 80 °C. After 4 h the reaction was quenched with ethyl vinyl
ether and pressurized to 40 bar with H2 for 1 day. The saturated
primary carboxylic acid ester product is formed as the only
product, as evidenced by proton NMR spectroscopy (Figure 6,
Scheme 4. Cobalt-Catalyzed Epoxy Functionalization with
mCPBA
room temperature virtually complete epoxidation of the
internal as well as terminal double bonds was achieved,
corresponding to degrees of functionalization of >95%. Key 1H
NMR resonances from the epoxide products arise at 3.1 to 2.5
ppm, while only small amounts (<5%) of terminal double
bonds remain unreacted (Figure S14, Supporting Information).
Epoxide formation is also evidenced by a new resonance for the
C−O stretching of the epoxide at 894.9 cm−1 in the IR
spectrum. The reaction yielded a mixture of internal and
terminal epoxides, reflecting the distribution of the double
bonds generated in the branching ethylene oligomerization.
Epoxidation of the oligomers proved to be very robust and
easily scalable. Functionalization could be performed on the
branched oligoethylenes as obtained from oligomerization
without further workup (involving e.g. removal of residual
catalyst or its deactivation products) on a 50 g scale or larger.
1
Figure 6. H NMR spectra of the unsaturated ester (bottom), the
saturated ester (center), and the hydroxyl-functionalized oligomer
(top).
center). An incomplete metathesis functionalization step would
result in an apparent increase of molecular weight, as
determined from NMR after hydrogenation (cf. the Supporting
Information). A found molecular weight of Mn(NMR) of 950 is
further evidence of complete metathesis.
Further functionalization of the ester-substituted apolar
branched oligomers was exemplified by conversion to the
alcohol. Reduction with LiAlH4 quantitatively yielded the
corresponding hydroxyl-terminated oligomers (Figure 6, top).
However, for a large-scale synthesis the utilization of LiAlH4
would be prohibitive due to its cost and the tedious workup
required for the removal of inorganic salts. The ruthenium
2083
dx.doi.org/10.1021/ja411945n | J. Am. Chem. Soc. 2014, 136, 2078−2085

Journal of the American Chemical Society
Article
(7) Guan, Z.; Cotts, P. M.; McCord, E. F.; McLain, S. J. Science 1999,
283, 2059.
(8) Cotts, P. M.; Guan, Z.; McCord, E.; McLain, S. Macromolecules
2000, 33, 6945.
SUMMARY AND CONCLUSION
■
Low-molecular-weight products with a hyperbranched micro-
structure are accessible by oligomerization of ethylene with
Ni(II)−salicylaldiminato catalysts. A key to this efficient
formation of highly branched oligomers is an N-terphenyl
motif with remote alkyl substituents. The oligomerization to
hyperbranched products proceeds with high rates under
moderate conditions of pressure and temperature (≤20 atm,
50 °C). An in situ prepared catalyst eliminates the necessity of
tedious preparation of isolated Ni−dimethyl precursors and
thereby resolves an essential limitation of this chemistry.
Functionalization of the unsaturated end group of the oligomer
chains via cross metathesis or epoxidation, respectively, gives
access to monofunctional highly branched ethylene oligomers
in essentially quantitative yield. Notably, also in cross
metathesis molecular weights are essentially retained and very
little material is lost. Overall, all reactions involved are
straightforward and readily scalable, as demonstrated by the
preparation of 50 g batches. Due to their high reactivity, the
primary carboxy or alcohol or disubstituted epoxide groups
generated lend themselves to further functionalization. To our
knowledge, this is the first report of such hyperbranched
monofunctional materials.
(9) Gates, D. P.; Svejda, S. A.; Onate, E.; Killian, C. M.; Johnson, L.
̃
K.; White, P. S.; Brookhart, M. Macromolecules 2000, 33, 2320.
(10) Deng, L.; Woo, T. K.; Cavallo, L.; Margl, P. M.; Ziegler, T. J.
Am. Chem. Soc. 1997, 119, 6177.
(11) Xiang, P.; Ye, Z.; Subramanian, R. Polymer 2011, 52, 5027.
(12) (a) Keim, W. Chem. Ing. Tech. 1984, 56, 850. (b) Pietsch, J.;
Braunstein, P.; Chauvin, Y. New J. Chem. 1998, 22, 467. (c) Vogt, D.:
Organic-organic biphasic catalysis. In Multiphase Homogeneous
Catalysis; Cornils, B., Herrmann, W. A., Horvath, I. T., Leitner, W.,
Mecking, S., Olivier-Bourbigou, H., Vogt, D., Eds.; Wiley-VCH:
Weinheim, Germany, 2005; pp 330−335. (d) Keim, W. Angew. Chem.
Int. Ed. 2013, 52, 12492.
(13) (a) Killian, C. M.; Johnson, L. K.; Brookhart, M. Organometallics
1997, 16, 2005. (b) Svejda, S. A.; Brookhart, M. Organometallics 1999,
18, 65.
(14) (a) Wang, C.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs, R.
H.; Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3149.
(b) Younkin, T. R.; Connor, E. F.; Henderson, J. I.; Friedrich, S. K.;
Grubbs, R. H.; Bansleben, D. A. Science 2000, 287, 460. (c) Johnson,
L. K.; Bennett, A. M. A.; Ittel, S. D.; Wang, L.; Parthasarathy, A.;
Hauptman, E.; Simpson, R. D.; Feldman, J.; Coughlin, E. B. (DuPont)
WO98/30609, 1998.
(15) (a) Song, D.-P.; Wang, Y.-X.; Mu, H.-L.; Li, B.-X.; Li, Y.-S.
Organometallics 2011, 30, 925−934. (b) Radlauer, M. R.; Buckley, A.
K.; Henling, L. M.; Agapie, T. J. Am. Chem. Soc. 2013, 135, 3784−
3787.
ASSOCIATED CONTENT
* Supporting Information
Text, tables, figures, and CIF files giving complete experimental
procedures and analytical data, molecular weight and branching
degree calculations, and crystallographic data/processing
parameters. This material is available free of charge via the
Internet at http://pubs.acs.org.
■
S
(16) Zuideveld, M. A.; Wehrmann, P.; Rohr, C.; Mecking, S. Angew.
̈
Chem., Int. Ed. 2004, 43, 869.
(17) Osichow, A.; Rabe, C.; Vogtt, K.; Narayanan, T.; Harnau, L.;
Drechsler, M.; Ballauff, M.; Mecking, S. J. Am. Chem. Soc. 2013, 135,
11645.
(18) (a) Gottker-Schnetmann, I.; Wehrmann, P.; Rohr, C.; Mecking,
̈
̈
AUTHOR INFORMATION
Corresponding Author
stefan.mecking@uni-konstanz.de
■
S. Organometallics 2007, 26, 2348. (b) Bastero, A.; Gottker-
̈
Schnetmann, I.; Rohr, C.; Mecking, S. Adv. Synth. Catal. 2007, 349,
̈
2307. (c) Weberski, M. P.; Chen, C.; Delferro, M.; Zuccaccia, C.;
Macchioni, A.; Marks, T. J. Organometallics 2012, 31, 3773−3789.
Notes
(d) Osichow, A.; Gottker-Schnetmann, I.; Mecking, S. Organometallics
̈
The authors declare no competing financial interest.
2013, 32, 5239−5242. (e) Soshnikov, I. E.; Semikolenova, N. V.;
̈
Zakharov, V. A.; Moller, H. M.; Olscher, F.; Osichow, A.; Gottker-
̈
̈
ACKNOWLEDGMENTS
We thank Jurgen Omeis, Michael Bessel, and Dominika Bernert
for fruitful discussions. Financial support by Byk is gratefully
■
Schnetmann, I.; Mecking, S.; Talsi, E. P.; Bryliakov, K. P. Chem. Eur. J.
2013, 19, 11409−11417.
̈
(19) Jenkins, J. C.; Brookhart, M. J. Am. Chem. Soc. 2004, 126, 5827.
(20) (a) Daugulis, O.; Brookhart, M. Organometallics 2002, 21, 5926.
(b) Desjardins, S. Y.; Way, A. A.; Murray, M. C.; Adirim, D.; Baird, M.
C. Organometallics 1998, 17, 2382.
(21) (a) Randall, J. C. J. Macromol. Sci., Rev. Macromol. Chem. Phys.
1989, C29, 201. (b) Galland, G. B.; de Souza, R. F.; Mauler, R. S.;
Nunes, F. F. Macromolecules 1999, 32, 1620.
(22) A possible route of catalyst deactivation involves reductive
coupling of a growing Ni−alkyl species with an Ni−H or another Ni−
alkyl; cf.: Berkefeld, A.; Mecking, S. J. Am. Chem. Soc. 2009, 131,
1565−1574. This would result in chains with two saturated end
groups. Given that in the oligomerizations studied here ca. 100
oligomer chains are formed per Ni(II) center, any such deactivation
reaction would be negligible in terms of oligomer microstructure and
analysis.
acknowledged.
REFERENCES
■
(1) Behr, A.: Ziegler processes. In Ullmann’s Encyclopedia of Industrial
Chemistry; Wiley-VCH: Weinheim, Germany, 2000.
(2) Wolfmeier, U.; Schmidt, H.; Heinrichs, F.-L.; Michalczyk, G.;
Payer, W.; Dietsche, W.; Boehlke, K.; Hohner, G.; Wildgruber, J.:
Waxes. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH:
Weinheim, Germany, 2000.
(3) Amin, S. B.; Marks, T. J. Angew. Chem., Int. Ed. 2008, 47, 2006.
(4) (a) Britovsek, G. J. P.; Cohen, S. A.; Gibson, V. C.; Maddox, P. J.;
van Meurs, M. Angew. Chem., Int. Ed. 2002, 41, 489. (b) Arriola, D. J.;
Carnahan, E. M.; Hustad, P. D.; Kuhlman, R. L.; Wenzel, T. T. Science
2006, 312, 714. (c) Mazzolini, J.; Espinosa, E.; D’Agosto, F.; Boisson,
C. Polym. Chem. 2010, 1, 793. (d) Gibson, V. C. Science 2006, 312,
703.
(23) (a) Pugh, R. I.; Drent, E.; Pringle, P. G. Chem. Commun. 2001,
1476. (b) Jimenez Rodriguez, C.; Foster, D. F.; Eastham, G. R.; Cole-
Hamilton, D. J. Chem. Commun. 2004, 1720.
(24) Stempfle, F.; Quinzler, D.; Heckler, I.; Mecking, S. Macro-
molecules 2011, 44, 4159.
(5) (a) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc.
1995, 117, 6414. (b) Rose, J. M.; Cherian, A. E.; Coates, G. W. J. Am.
Chem. Soc. 2006, 128, 4186.
(25) Decreasing the amount of ethanol resulted in decreased degrees
of functionalization. Also, higher dilution or utilization of other solvent
mixtures such as pentane/methanol and toluene/ethanol did not
enhance conversion.
(6) (a) Hyperbranched Polymers Eds. Yan, D.; Gao, C.; Frey, H.
Wiley: Hoboken, 2011. (b) Voit, B. I.; Lederer, A. Chem. Rev. 2009,
109, 5924−5973. (c) Calderon, M.; Quadir, M. A.; Sharma, S. K.;
Haag, R. Adv. Mater. 2010, 22, 190−218.
2084
dx.doi.org/10.1021/ja411945n | J. Am. Chem. Soc. 2014, 136, 2078−2085

Journal of the American Chemical Society
Article
(26) Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. J.
Am. Chem. Soc. 2003, 125, 11360.
(27) Abbas, M.; Slugovc, C. Monatsh. Chem. 2012, 143, 669.
(28) Minor peaks not assigned are background originating from the
column material.
́
(29) (a) Olivan, M.; Caulton, K. G. Inorg. Chem. 1999, 38, 566.
(b) Bielawski, C. W.; Louie, J.; Grubbs, R. H. J. Am. Chem. Soc. 2000,
122, 12872. (c) Drouin, S. D.; Zamanian, F.; Fogg, D. E.
Organometallics 2001, 20, 5495.
(30) Spasyuk, D.; Smith, S.; Gusev, D. G. Angew. Chem., Int. Ed.
2012, 51, 2772.
(31) Matoishi, K.; Nakai, K.; Nagai, N.; Terao, H.; Fujita, T. Catal.
Today 2011, 164, 2.
(32) Hyun, M. Y.; Kim, S. H.; Song, Y. J.; Lee, H. G.; Jo, Y. D.; Kim,
J. H.; Hwang, I. H.; Noh, J. Y.; Kang, J.; Kim, C. J. Org. Chem. 2012,
77, 7307.
2085
dx.doi.org/10.1021/ja411945n | J. Am. Chem. Soc. 2014, 136, 2078−2085