Decreasing the alkyl branch frequency in precision polyethylene: Effect of alkyl branch size on nanoscale morphology

Article abstract of DOI:10.1021/ma3002577

Synthesis and morphological characterization are reported for a series of 13 precision branched polyethylene structures, the branch being placed on every 39th carbon and varying in size from methyl to pentadecyl group. A recently established synthetic scheme for preparation of the symmetrical α,ω-diene monomer was employed to increase the number of methylene carbons between the branch points from 20 to 38, yielding polymers with 5.26 mol % α-olefin incorporation. The morphology of these polymers was investigated using differential scanning calorimetry (DSC), wide-angle X-ray diffraction (WAXD), and transmission electron microscopy (TEM). Methyl branching significantly reduces the melting point and single crystal lamellae thickness of unbranched polyethylene. On the other hand, all further branches from ethyl to pentadecyl produce polymers that have similar melting points and single crystal lamellae thicknesses. A clear change in the morphology of both solution and melt-grown crystals of these polymers was observed from a situation where the methyl branch is incorporated in the polymer's unit cell to one where branches of greater mass are mostly expelled from the unit cell.

Full text of DOI:10.1021/ma3002577

Article
pubs.acs.org/Macromolecules
Decreasing the Alkyl Branch Frequency in Precision Polyethylene:
Effect of Alkyl Branch Size on Nanoscale Morphology
Bora Inci,†,§ Ingo Lieberwirth,‡ Werner Steffen,‡ Markus Mezger,‡ Robert Graf,‡ Katharina Landfester,‡
and Kenneth B. Wagener*,†
†Center for Macromolecular Science and Engineering, The George and Josephine Butler Polymer Research Laboratory, Department
of Chemistry, University of Florida, Gainesville, Florida 32611-7200, United States
‡Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
S
* Supporting Information
ABSTRACT: Synthesis and morphological characterization are
reported for a series of 13 precision branched polyethylene
structures, the branch being placed on every 39th carbon and
varying in size from methyl to pentadecyl group. A recently
established synthetic scheme for preparation of the symmetrical
α,ω-diene monomer was employed to increase the number of
methylene carbons between the branch points from 20 to 38,
yielding polymers with 5.26 mol % α-olefin incorporation. The
morphology of these polymers was investigated using differential
scanning calorimetry (DSC), wide-angle X-ray diffraction (WAXD), and transmission electron microscopy (TEM). Methyl
branching significantly reduces the melting point and single crystal lamellae thickness of unbranched polyethylene. On the other
hand, all further branches from ethyl to pentadecyl produce polymers that have similar melting points and single crystal lamellae
thicknesses. A clear change in the morphology of both solution and melt-grown crystals of these polymers was observed from a
situation where the methyl branch is incorporated in the polymer′s unit cell to one where branches of greater mass are mostly
expelled from the unit cell.
locene,21−23 and late transition metal24 catalytic systems. By its
nature, chain-propagation chemistry incorporates structural
INTRODUCTION
The structure of polyethylene has been extensively inves-
■
tigated1,2 and is considered as the model for polymer
defects via inevitable chain transfer and chain walking processes
crystallization studies.3−6 Linear polyethylene (high density
which produce alkyl branches of varying lengths randomly
spaced along the polyethylene backbone.25
polyethylene) crystallized from solution or the melt forms
folded chain crystals having the orthorhombic unit cell.7 The
Recently, we have employed step polycondensation chem-
istry (the ADMET reaction) to generate precisely branched
branched versions of polyethylene (low density and linear low
density polyethylene) have attracted more attention due to
ethylene copolymers.26−31 Symmetrical α,ω-diene monomers
(having built-in branches) are condensed into unsaturated
their wide-ranging thermal and mechanical properties governed
by branching.8,9 The effect of branching on crystal structure of
polymers, which upon saturation generate precision branched
polyethylenes.32,33 This approach circumvents the random
polyethylene has been examined in chain polymerized ethyl-
ene/α-olefin copolymers with comonomers propylene,10,11 1-
nature of branching in polyethylene with the elimination of
butene,10,12,13 and 1-octene.14 It is known that the ethylene/
chain transfer and chain walking processes. The design of the
symmetrical α,ω-diene monomer obviates the ethylene/
propylene copolymers show evidence for methyl branch
comonomer reactivity ratio problems and ensures the precise
inclusion into the crystal unit cell, whereas ethylene/1-butene
placement of branches along the polymer backbone. We are
and ethylene/1-octene polymers display the exclusion of ethyl
and hexyl branches due to their larger branch sizes.9,15 The
now able to prepare polymers where both the identity of the
branch and its position are known without equivocation.
validity of this generalization is limited, and partial ethyl branch
Systematic alterations in branch identity and position on the
inclusion is suggested for some of the ethylene/1-butene
copolymer systems.16,17 For the case of ethylene/1-octene
polymer backbone can be made during monomer synthesis to
yield predictable changes in physical properties.31
copolymers, X-ray investigations show the existence of a
Linear polyethylene prepared via ADMET by polycondensa-
complex multiphase morphology which hinders the deeper
tion of 1,9-decadiene melts at 134 °C. Introduction of branches
examination of hexyl branch placement within the polymer
structure.14
The ambiguity in morphology determination of ethylene
copolymers is governed by the heterogeneously distributed
nature of branches using either Ziegler−Natta,18−20 metal-
Received: February 6, 2012
Revised: April 4, 2012
Published: April 15, 2012
© 2012 American Chemical Society
3367
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
Scheme 1. Synthesis of 21-Alkylhentetraconta-1,40-dienes (8a−m)
Scheme 2. Synthesis of Precisely Sequenced Polymers
onto the polyethylene backbone disrupts the crystallinity and
results in a decrease in melting point.26,34 For example,
insertion of butyl branches on every 21st carbon (i.e., 20
methylene carbons between consecutive branch points)
generates a semicrystalline polymer that melts at 12 °C.29
More frequent butyl branch placement on the polymer
backbone can also be achieved with this technique to prepare
polymers with lower melting points and, in some cases,
amorphous materials (with as few as four methylene groups
between branches).35 A recently developed synthetic scheme
for preparation of the symmetrical α,ω-diene monomer has
been employed to decrease the total concentration of butyl
branches by increasing the number of methylene carbons
between the branch points from 20 to 38. This synthetic
methodology yields a semicrystalline polymer that melts at 75
°C.36 The thermal and morphological behavior of this polymer
was compared to that of an ethylene/1-hexene copolymer
having the same net concentration of butyl branches along the
polymer backbone but prepared via chain polymerization with a
metallocene catalyst.37 Both polymers show the existence of
(110) and (200) orthorhombic unit cell reflection planes, with
the ADMET polymer having a sharper melting transition and
narrower lamella thickness distribution.
Herein, we report the synthesis and the characterization of
precisely sequenced polyethylenes containing 13 different
branches, each being placed on every 39th carbon on the
polymer backbone. This paper extends the scope of the
precision polyethylene study to provide a better understanding
for the effect of the branch identity on polymer properties.
3368
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
Figure 1. (A) 1H 500 MHz NMR spectra of monomer 8h-(hexyl), unsaturated 9h-(hexyl), and saturated 10h-(hexyl) polymers. (B) Olefinic region
shown in higher magnification.
Monomers 8a−m were condensed to form the correspond-
ing unsaturated ADMET polymers (Scheme 2). Because of the
relatively high melt viscosity of the resulting polymers (9a−m),
the polymerization temperature was set to 85 °C. At that
elevated temperature, Ru-based catalysts are prone to have low
turnover numbers and to generate Ru−H species,41−43 which
would cause isomerization problems and disrupt the sym-
metrical nature of the monomer.44,45
RESULTS AND DISCUSSION
■
A. Synthesis and Primary Structure Characterization
of Precisely Branched Polymers. Successful preparation of
symmetrical α,ω-diene monomers is the key in this research,
where a multistep synthetic route was required for each
monomer (8a−m). Scheme 1 illustrates the elaborate synthetic
effort, which consisted of a six-step procedure including alkenyl
bromide (5) preparation,38 dialkylation of nitriles (6a−m),39
and decyanation of alkylcyano α,ω-dienes (7a−m).40
The spacing between the two consecutive branch points
along the polyethylene backbone (see structures of polymers
10a−m in Scheme 2) was set in the dialkylation step of nitriles,
where the alkenyl bromide run length (in this case 18
methylene units) played a crucial role. Systematic increase of
run length from 9 to 18 required challenging purification steps.
For example, in the dehydrohalogenation of alkyl dibromide
(4) with t-BuOK, the crude mixture contained the desired
alkenyl bromide (20-bromoicos-1-ene 5), the dielimination
product, and unreacted starting material, all with close
retardation factors by thin layer chromatography. Consecutive
column chromatography passes and the relatively low room
temperature solubility of compound 5 in hexane (used as an
eluent for the separation) made the purification of the desired
compound, 20-bromoicos-1-ene (5), tedious.
Therefore, monomers 8a−m were condensed to form
unsaturated ADMET polymers using Schrock’s [Mo] catalyst
for clean metathesis chemistry. Because of the oxophilic nature
of [Mo] catalysts, all the manipulations prior to polymerization
and catalyst addition (catalyst to monomer ratio 1:500) were
performed in a glovebox. Polymerization was initiated by
melting the monomer at 50 °C, and the temperature was set to
85 °C to be able to stir the viscous polymer melt. After 24 h of
reaction, the polymer was cooled to RT, another portion of
[Mo] catalyst (catalyst to monomer ratio 1:500) was added in
the glovebox, and the temperature was set back to 85 °C.
ADMET polymerization proceeded smoothly to give the
desired unsaturated linear polymers 9a−m with no detectable
1
side reactions. Disappearance of terminal olefin signals in H
NMR spectra (Figure 1) proved the complete conversion,
which is necessary for any step-growth polymerization.
Unsaturated polymers (9a−m) were hydrogenated using
Wilkinson's catalyst (catalyst to monomer ratio 1:250) to
yield the precision polyethylenes (polymers 10a−m) having
branches from methyl to pentadecyl on every 39th carbon.
The polymerization and hydrogenation steps were followed
by 1H nuclear magnetic resonance (NMR) spectroscopy. As an
Alkenylation of nitriles 6a−m in the presence of lithium
diisopropylamide (LDA) and 20-bromoicos-1-ene produced
the alkylcyano α,ω-dienes 7a−m in high yields. Decyanation of
nitriles 7a−m was achieved with potassium metal via radical
chemistry. The resulting tertiary radical after decyanation was
further quenched by abstraction of hydrogen from t-BuOH to
give α,ω-diene monomers 8a−m in moderate to high yields.
1
example, Figure 1 illustrates the H NMR spectra of hexyl
branched polymer, 10h-(hexyl), and its precursors. ADMET
3369
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
polymerization of monomer 8h-(hexyl) yielded the unsaturated
polymer, 9h-(hexyl). Formation of the ADMET polymer
resulted in loss of the terminal olefin signals (5.0 and 5.8 ppm)
and the appearance of the internal olefin at 5.4 ppm (Figure 1).
Exhaustive hydrogenation of the internal olefins with
Wilkinson's catalyst generated 10h-(hexyl), corresponding to
polyethylene with hexyl branches on every 39th carbon, with
absence of CC in the polymer backbone and proves the full
conversion for the hydrogenation step.
After hydrogenation, solutions of saturated polymers were
concentrated and precipitated into methanol. It is important to
note that the solubility characteristics of polymers 10a−m
(soluble in toluene, dichlorobenzene, trichlorobenzene, etc., at
high temperatures) are similar to those of polyethylenes
prepared by chain polymerization. Molecular weight data were
obtained using high temperature gel permeation chromatog-
raphy (GPC) in 1,2,4-trichlorobenzene at 135 °C relative to
polystyrene standards. Table 1 illustrates the weight-average
molecular weights for the precisely branched unsaturated and
saturated polymers.
1
complete loss of the olefinic signals in H NMR.
Polymerization and hydrogenation of monomer 8h-(hexyl)
was also monitored with 13C NMR spectroscopy (Figure 2).
Table 1. Molecular Weights and Thermal Data for Precisely
Branched Polymers
a
b
M
(kg/mol) (PDI )
̅
w
branch identity on every
39th carbon
Tm (°C)
(peak)
ΔHm
(J/g)
unsaturated
saturated
no branch
methyl
ethyl
134
92
76
78
77
75
73
74
73
74
74
73
71
70
211
137
93
71
74
66
51
88
85
85
73
84
76
83
70.2 (2.7)
33.7 (2.2)
56.7 (2.2)
217 (3.1)
107 (2.0)
60.2 (2.3)
22.6 (2.5)
29.3 (2.2)
30.5 (2.3)
76.6 (3.0)
200 (3.6)
35.9 (2.4)
28.0 (2.4)
58.0 (2.4)
70.2 (2.7)
92.7 (2.0)
53.1 (2.4)
225 (3.0)
144 (3.5)
66.5 (2.5)
54.8 (2.4)
30.2 (2.0)
30.5 (1.9)
74.2 (2.9)
181 (3.3)
34.3 (2.2)
27.7 (1.8)
55.9 (2.4)
propyl
isopropyl
butyl
isobutyl
pentyl
hexyl
Figure 2. 13C 126 MHz NMR spectra of monomer 8h-(hexyl),
unsaturated 9h-(hexyl), and saturated 10h-(hexyl) polymers.
heptyl
octyl
nonyl
Comparison of the 13C NMR spectra for the monomer 8h-
(hexyl) and unsaturated polymer 9h-(hexyl) indicates the
disappearance of the signals belonging to the terminal olefin at
114.31 and 139.44 ppm and formation of the new internal
olefin (cis olefin at 130.12 ppm, minor product, and trans olefin
at 130.58 ppm, major product) generated by the metathesis
chemistry. Hydrogenation of the internal olefin with
Wilkinson's catalyst yielded the saturated polymer 10h-
(hexyl) with no detectable trace of olefins.
decyl
pentadecyl
a
Molecular weight data were collected by GPC in 1,2,4-trichlor-
b
obenzene at 135 °C relative to polystyrene standards. PDI,
polydispersity index M /M .
̅
̅
w
n
Control over the primary structure of the final precision
polymer is governed by the precision established on the
molecular level. The purity of the monomers and the absence
of any side reactions during the polymerization ensure the
successful polycondensation chemistry and the level of control.
Upon close inspection of the 13C NMR data for the polymers, it
can be concluded that the branches are precisely placed along
the polyethylene backbone with none of the undesired
branches due to chain transfer typically observed during
chain-growth chemistry. As an example, Figure 4 shows a
portion (10−55 ppm) of the 13C NMR spectra for precise
polymers having branches from methyl to propyl. All spectra
are dominated by a singlet at 29.99 ppm corresponding to
methylenes on the main polyethylene chain. Note that the
presence of alkyl branches precisely placed along the main
chain affects the chemical shifts of carbons located within three
CH2 units from an individual branch.26
Infrared (IR) spectroscopy was also used to monitor this
transformation (Figure 3). Disappearance of the out-of-plane
C−H bend absorption at 969 cm−1 indicates the complete
In the spectrum for 10a-(methyl), which is polyethylene
containing methyl branches on every 39th backbone carbon
(Figure 4A), the resonances belonging to the methyl branch,
19.84 ppm, as well as main chain carbons, 37.45, 27.98, 30.34,
and 29.99 ppm and the carbon at the branch point (33.12 ppm)
indicate that only methyl branches are present, in good
agreement with previously reported26 experimental data and
predicted values.46
Figure 3. Infrared spectra of the precisely branched polymers 10a-
(methyl), 10b-(ethyl), 10c-(propyl), 10d-(iso-propyl), 10e-(butyl),
10f-(iso-butyl), 10g-(pentyl), 10h-(hexyl), 10i-(heptyl), 10j-(octyl),
10k-(nonyl), 10l-(decyl), 10m-(pentadecyl), and ADMET PE.
3370
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
Figure 5. Differential scanning calorimetry thermograms for second
heating cycles of ADMET PE, 10a-(methyl), 10b-(ethyl), 10c-
(propyl), 10d-(iso-propyl), 10e-(butyl), and 10f-(iso-butyl) with 10
°C/min heating rate.
Figure 6. Differential scanning calorimetry thermograms for second
heating cycles of precision polymers 10g-(pentyl), 10h-(hexyl), 10i-
(heptyl), 10j-(octyl), 10k-(nonyl) and 10l-(decyl), and 10m-
(pentadecyl) with 10 °C/min heating rate.
(DSC) thermograms of linear ADMET PE and precise
polymers having branches from methyl to pentadecyl increasing
in both length and bulkiness.
Figure 4. Comparison of 13C NMR spectra for precision polymers (A)
10a-(methyl), (B) 10b-(ethyl), and (C) 10c-(propyl).
It is immediately obvious that methyl branching significantly
reduces the melting point of ADMET polyethylene. On the
other hand, all further branches, from ethyl to pentadecyl,
produce polymers that have very similar (in some cases
identical) melting points. As discussed further below, this
behavior may be attributed to a change in morphology from a
situation where the methyl branch is incorporated into the
polymer crystal to one where larger branches are expelled from
the crystal. It is important to note that the ethyl branch is
incorporated into the polymer crystal when the branches are
placed on every 21st carbon.31 Increasing the distance between
two consecutive branches from 20 carbons (10.0 mol % α-
olefin incorporation) to 38 carbons (5.26 mol % α-olefin
incorporation) expels the ethyl branches from the crystal lattice
into the amorphous phase. Similar observations were also
reported for chain polymerized ethylene/1-butene copolymers,
where the ethyl branch is mostly found in the amorphous
region.9,54
Introduction of ethyl branches changes the chemical shifts of
both backbone and side chain carbons (Figure 4B). The
terminal methyl carbon is now more shielded by the presence
of a methylene unit, which results in a predictable shift upfield
(11.19 ppm). On the other hand, the methine carbon is
deshielded with more electron delocalization and shifts
downfield (39.41 ppm) compared to the methyl branched
polymer. All the chemical shifts shown in Figure 4B for the
polymer 10b-(ethyl) are in good agreement with previously
reported ethylene/1-butene and hydrogenated polybutadiene
systems.47−50
In the spectrum for 10d-(propyl) (Figure 4C), the
resonances belonging to the propyl branch, 14.55, 20.20, and
36.66 ppm, as well as main chain carbons, 34.25, 27.16, 30.49,
and 29.99 ppm and the methine carbon (37.70 ppm) indicate
that only propyl branches are present, in good agreement with
previously reported data on chain-growth polymers obtained by
copolymerization of ethylene with 1-pentene.51−53
As shown in Figure 5, unbranched ADMET PE displays
thermal behavior virtually the same as that of high density
polyethylene (Tm = 134 °C, ΔHm = 211 J/g). Incorporation of
B. Melting Behavior of Precisely Branched Polymers.
Figures 5 and 6 present the differential scanning calorimetry
3371
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
methyl branches precisely placed on every 39th carbon,
polymer 10a-(methyl), disrupts the crystal structure and
decreases the melting temperature to 92 °C and the enthalpy
of fusion to 137 J/g. A similar trend is observed when the
branch is an ethyl group, 10b-(ethyl); the melting point
decreases below that of 10a-(methyl) (Tm = 76 °C with a heat
of fusion ΔHm = 93 J/g). Depression of both the melting
temperature and the heat of fusion suggests a reduction of
crystal thickness for polymer 10b-(ethyl). Transmission
electron microscopy (TEM) measurements of solution-grown
single crystals for polymer 10b-(ethyl) also support this
observation (see section C). Extension of the branch size
from ethyl to propyl, 10c-(propyl), does not lead to further
decreases in melting temperature and enthalpy of fusion (Tm =
78 °C and ΔHm = 71 J/g). Similar melting behavior is observed
for precision polyethylene possessing longer branches (Figure
7).
the broader lamella thickness distribution of chain-polymerized
ethylene/1-hexene copolymer due to random positioning of
butyl branches along the polymer backbone.37
C. Morphology and Crystal Structure Determination
of Precisely Branched Polymers. Wide-angle X-ray and
selected area electron diffraction measurements further support
the observation of a change in unit cell as a function of branch
size. WAXD diffractograms of precision polymers having
branches from methyl to pentadecyl are shown in Figures 8
and 9.
Figure 7. Effect of branch identity on melting point and heat of fusion.
Incorporation of defects in polyethylene crystal along with
the melting behavior of chain-polymerized ethylene copolymers
has been extensively studied.9,15,54 Alamo et al. reported that
only small groups, such as CH3, Cl, O, and OH, are
incorporated into the crystal lattice, whereas the bigger defects
are expelled from the crystal.9 Hosoda et al. prepared various
ethylene/α-olefin copolymers with a vanadium catalyst
system.54 They determined that the probability of branch
inclusion into the crystal is strongly dependent on branch
identity and is in the order of methyl > ethyl > butyl = hexyl =
decyl > isobutyl. Interestingly, ethylene/1-octene and ethylene/
1-dodecene copolymers reported by the same group with 5.14
and 4.54 mol % α-olefin incorporation ratios exhibit melting
points (70−75 °C) very similar to those of precision polymers
10h-(hexyl) and 10l-(decyl) with 5.26 mol % α-olefin
incorporation. More recently, Hosoda et al. compared the
melting behavior of precision polymer 10e-(butyl) with the
ethylene/1-hexene copolymer having the same net concen-
tration of butyl branches along the polymer backbone but
prepared via chain polymerization with a metallocene catalyst.37
The melting profile of polymer 10e-(butyl) displays a melting
curve at 75 °C with width less than 20 °C. Even though the two
polymers have the same net concentration of butyl branches,
ethylene/1-hexene copolymer melts over a broader range from
50 to 105 °C with a melting point of 99 °C. The distinctly
different melting behavior of these polymers was attributed to
Figure 8. Wide-angle X-ray diffraction (WAXD) patterns for precision
polymers 10a-(methyl), 10b-(ethyl), 10c-(propyl), 10d-(iso-propyl),
10e-(butyl), 10f-(iso-butyl), and ADMET PE obtained at RT. Prior
to data acquisition, all samples were heated to 20 °C above the melting
temperature of each polymer to remove the thermal history and then
cooled to room temperature at a rate of 50 °C/min.
WAXD data were collected using the Cu Kα radiation (λ =
0.154 17 nm) induced by a generator operating at 40 kV and
150 mA. Diffraction patterns were recorded for 2(theta) values
ranging from 10° to 40°. The peak at 38.5° observed in all
WAXD diffractograms is the (111) reflection of the Al cubic
closed packed structure. Aluminum foil was used to sandwich
the polymer sample between copper holders.
For the sake of comparison, ADMET PE is displayed at the
bottom of Figure 8. It exhibits the typical orthorhombic crystal
structure (lattice parameters a = 7.40 Å, b = 4.93 Å, and c =
2.54 Å) with two characteristic crystalline peaks, exactly the
same as for high density polyethylene made by chain
propagation chemistry.55 The more intense peak at scattering
angle 21.7° and the less intense one at 24.0° correspond to
reflection planes (110) and (200), respectively.
Introduction of precisely placed methyl branches on every
39th carbon (see polymer 10a-(methyl) in Figure 8) disturbs
3372
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
Figure 10. Scattering angles of two strong reflections for alkyl
branched precision polymers. The blue line corresponds to reflection
at higher angle, while the black line corresponds to reflection at lower
angle.
exhibits a distinct morphological behavior, in which the
relatively long pentadecyl branch presumably cocrystallizes
with the main polymer chain, thereby increasing the d-spacing
between the diffraction planes. Cocrystallization of branches
longer than 10 carbon atoms in length with the main chain was
previously reported for chain polymerized polyolefins.56
An interesting aspect of the WAXD patterns for the precision
polymers ranging from 10b-(ethyl) to 10m-(pentadecyl) is
that the 19.6° peak, which is sharper than that of the
amorphous halo, suggests the possible coexistence of a
dominant orthorhombic crystal and another metastable crystal
structure as a minor component. Selected area electron
diffraction (SAED) measurements on solution-grown polymer
10b-(ethyl) crystals confirm the formation of a monoclinic
crystal with lattice parameters of a = 8.04 Å, c = 4.80 Å, and β =
109.4°. The metastable monoclinic polyethylene structure was
previously reported by Seto57 with unit cell parameters a = 8.08
Å, b = 2.54 Å, c = 4.79 Å, and β = 107.9°. Please note that the
notation of crystallographic directions for the monoclinic
crystal structure differs from the orthorhombic case. Here the
crystalline b-direction is the direction parallel to the polymer
chain.
Formation of monoclinic crystalline phase is known for
chain-polymerized ethylene/α-olefin copolymers (having
branches larger than methyl) with high comonomer content.58
Hu et al. reported ethylene/1-butene and ethylene/1-octene
copolymers with α-olefin concentrations ranging from 7.50 to
19.7% prepared with single site metallocene catalyst.58
Monoclinic crystals were found to coexist with the dominant
orthorhombic crystalline phase for polymers having como-
nomer content higher than 9.00 mol %. The authors proposed
that the formation of monoclinic crystals is strongly dependent
on the nature of the crystalline−amorphous interface.58
Polymers with high comonomer content display low crystal-
linity by populating the crystalline−amorphous interfacial area
and favor the formation of monoclinic crystals. It is interesting
that the precision polymers discussed here also exhibit
monoclinic crystals even with 5.26 mol % comonomer
incorporation. Because the defects larger than methyl are
expelled from the crystal, they presumably “agglomerate” at the
fold surface of the crystal. This might lead to a packing problem
Figure 9. Wide-angle X-ray diffraction patterns for precision polymers
10g-(pentyl), 10h-(hexyl), 10i-(heptyl), 10j-(octyl), 10k-(nonyl),
10l-(decyl), and 10m-(pentadecyl) obtained at RT. Prior to data
acquisition, all samples were heated to 20 °C above the melting
temperature of each polymer and then cooled to room temperature at
a rate of 50 °C/min.
the unit cell of ADMET PE and shifts the two strong scattering
peaks to 21.1° and 23.0°, corresponding to (110) (d = 4.19 Å)
and (200) (d = 3.87 Å) reflection planes with lattice parameters
a = 7.74 Å and b = 5.00 Å. An expansion of the unit cell by
4.6% and 1.4% along the a- and b-axes proves the distortion in
the orthorhombic unit cell governed by the incorporation of the
methyl branches in the unit cell. A similar expansion in unit cell
parameters was also reported for ethylene/propylene copoly-
mers prepared via chain polymerization, thus confirming the
methyl branch inclusion into the unit cell.11 In the case of the
ethyl branched polymer 10b-(ethyl), WAXD and electron
diffraction data suggest the existence of the orthorhombic
crystal structure with lattice parameters of a = 7.58 Å and b =
4.98 Å. WAXD diffractograms of precision polymers having
branches larger than ethyl (from propyl to pentadecyl) show
nearly identical scattering patterns, indicating the formation of
the orthorhombic crystal structure.
To better understand these observations, the scattering
angles of two strong diffraction peaks are plotted as a function
of branch identity in Figure 10. The decrease of scattering
angles for methyl branched polymer 10a-(methyl) indicates the
larger d-spacing between the diffraction planes, suggesting the
inclusion of the methyl branches into the unit cell. A clear
structural change for polymers ranging from 10b-(ethyl) to
10l-(decyl) is evidenced by the increase of scattering angles,
confirming the exclusion of branches from the unit cell. It is
important to note that precision polymer 10m-(pentadecyl)
3373
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
Table 2. Observed WAXD and SAED Lattice Plane Distances for 10a-(methyl) and Their Corresponding Allocation to
a
Orthorhombic and Monoclinic Crystal Planes
WAXDobs d [Å]
SAEDobs d [Å]
intensityobs WAXD/SAED
index (hkl)
(001)mono
d(hkl)calc [Å]
4.58
4.19
3.87
3.58
2.50
4.61
4.19
3.88
3.62
2.52
w/s
vs/vs
s/s
4.58
(110)ortho
4.20
(200)ortho/(200)mono
3.88/3.87
3.60
w/s
w/s
(−201)mono
(020)ortho/(201)mono
2.50/2.57
a
Abbreviations: obs, observed; calc, calculated; mono, monoclinic; ortho, orthorhombic; w, weak; s, strong; vs, very strong.
with too high a density of bulky defects and finally cause the
crystal lattice to shear to a metastable, in this case, monoclinic
lattice. There should be a limit in metastable crystal phase
formation for precisely branch polyethylene structures. That
limit was recently reached by positioning branches on every
75th carbon (2.70 mol % comonomer incorporation) where the
butyl branches are fully expelled from the crystalline phase,
yielding pure orthorhombic crystals with none of the
metastable crystal structure formation.59
SAED measurements were performed for solution-grown
crystals of precision polymer 10a-(methyl). Along with the
orthorhombic crystalline phase, monoclinic crystals were also
found with lattice parameters a = 8.22 Å, c = 4.85 Å, and β =
109.3°. Table 2 illustrates WAXD and SAED analysis of
polymer 10a-(methyl). A strong reflection corresponding to a
d-spacing of 4.19 Å indicates that the orthorhombic crystal
structure is the dominant form. However, some of the single
crystals show a nonorthorhombic electron diffraction pattern
which can be indexed as a monoclinic unit cell. Methyl
branches in chain-polymerized ethylene/1-propene copolymers
drive the change of orthorhombic crystal toward a hexagonal/
rotator crystalline phase instead of monoclinic phase.10
Observation of this unusual behavior in precision polymer
10a-(methyl) is subject to further investigation.
shaped rhombohedral single crystals were observed for all
polymers, consistent with previous reports on solution-grown
HDPE single crystals.3,4 It is noteworthy that, although the
crystallization conditions were similar for all the branched
polymers, the morphology of solution-grown crystals changed
based on the branch size.
Besides the ADMET PE single crystals, which clearly show
secondary nucleation via screw dislocations, only the single
crystals of polymer 10a-(methyl) exhibit secondary nucleation.
This may be an indication of a well-defined crystalline fold
surface and hence the incorporation of methyl branches into
the polymer crystal.
The lamella thicknesses of the solution-grown single crystals
were determined using energy-filtered TEM measurements.60,61
Taking two consecutive images from the same TEM sample,
one with the slit positioned at the zero-loss peak and one
without the slit, yielded two images, the elastically filtered
image I0 and the inelastic image It. After aligning both images
for possible sample drift, the relative thickness, t, at each point
of the images was given by t/λ = ln(I0/It), in which λ is the
mean free path of the respective material for the respective
electron energy (200 kV). Calculation of the mean free path
scales the image to an absolute thickness mapping.61 Table 3
illustrates such single crystal lamella thicknesses of ADMET PE
and precision polymers.
Figure 11 shows the TEM images of solution-grown single
crystals for precision polymers and ADMET PE. Lozenge-
Table 3. Lamellae Thicknesses of ADMET PE and Precision
Polymers Determined by TEM
polymer
thickness (nm)
polymer
thickness (nm)
ADMET PE
10a-(methyl)
10b-(ethyl)
10e-(butyl)
10g-(pentyl)
10h-(hexyl)
9.75
7.09
4.79
4.25
5.03
4.17
10i-(heptyl)
10j-(octyl)
4.39
4.06
4.67
4.16
5.33
10k-(nonyl)
10l-(decyl)
10m-(pentadecyl)
Methyl branching reduces the lamella thickness of
unbranched polyethylene; on the other hand, all further
branches produce polymers that have very similar lamella
thicknesses. Single crystal thickness data provide good evidence
for the positioning of branches. Assuming the all-trans
conformation of methylene units in the crystalline phase
(lattice constant c = 2.54 Å), the distance between the two
consecutive branch points (38 methylene units) can be
estimated to be 4.8 nm. The length matches quite well with
the observed crystal thickness of polymer 10b-(ethyl) and
precision polymers having branches longer than ethyl,
suggesting the exclusion of such branches from the crystal
phase. However, the single crystal thickness of polymer 10m-
(pentadecyl) is slightly increased to a value of ∼5.3 nm, which
Figure 11. Transmission electron microscopy (TEM) micrographs of
(1) ADMET PE, (2) 10a-(methyl), (3) 10b-(ethyl), (4) 10c-
(propyl), (5) 10e-(butyl), (6) 10g-(pentyl), (7) 10h-(hexyl), (8)
10i-(heptyl), (9) 10j-(octyl), (10) 10k-(nonyl), (11) 10l-(decyl),
and (12) 10m-(pentadecyl). ADMET PE crystals were grown from
0.03 wt % C2Cl4 solution, and all the other polymer single crystals
were grown from 0.03% o-xylene solution.
3374
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
(6) Ungar, G.; Stejny, J.; Keller, A.; Bidd, I.; Whiting, M. Science
1985, 229 (4711), 386−389.
may be attributed to cocrystallization of the pentadecyl branch
onto the crystal surface.
(7) Bunn, C. Trans. Faraday Soc. 1939, 35 (1), 0482−0490.
(8) Alamo, R.; Mandelkern, L. Macromolecules 1989, 22 (3), 1273−
1277.
Solution-grown single crystal lamella thickness measure-
ments provide additional evidence of a clear change in the
morphology of these polymers from a situation where the
methyl branch is incorporated into the solution-grown single
crystal to one where branches of greater mass are mostly
expelled from the crystal.
(9) Alamo, R.; Mandelkern, L. Thermochim. Acta 1994, 238, 155−
201.
(10) Hu, W.; Srinivas, S.; Sirota, E. Macromolecules 2002, 35 (13),
5013−5024.
(11) deBallesteros, O.; Auriemma, F.; Guerra, G.; Corradini, P.
Macromolecules 1996, 29 (22), 7141−7148.
CONCLUSIONS
■
(12) Crist, B.; Howard, P. Macromolecules 1999, 32 (9), 3057−3067.
(13) Crist, B.; Williams, D. J. Macromol. Sci., Phys. 2000, B39 (1), 1−
13.
Metathesis polycondensation chemistry has been employed to
control the crystalline morphology of a series of 13 precision
branched polyethylene structures, the branch being placed on
every 39th carbon and varying in size from methyl to
pentadecyl. The primary structures of both monomers and
polymers have been confirmed via 1H NMR, 13C NMR, and IR
spectroscopy to prove the purity of the monomers and the
absence of any undesired side reactions during polycondensa-
tion and hydrogenation steps. Methyl branches, being
incorporated into the polymer crystal, decrease the melting
point and the heat of fusion of ADMET PE. On the other
hand, increasing the size of the branch to ethyl and larger yields
a set of polymers with a completely different morphology, with
the branches mostly expelled from the polymer crystal. WAXD
patterns of melt-grown crystals and lamella thicknesses of the
solution-grown single crystals also support this argument.
(14) Androsch, R.; Blackwell, J.; Chvalun, S.; Wunderlich, B.
Macromolecules 1999, 32 (11), 3735−3740.
(15) Richardson, M.; Flory, P.; Jackson, J. Polymer 1963, 4 (2), 221−
236.
(16) Perez, E.; Vanderhart, D.; Crist, B.; Howard, P. Macromolecules
1987, 20 (1), 78−87.
(17) Howard, P.; Crist, B. J. Polym. Sci., Part B: Polym. Phys. 1989, 27
(11), 2269−2282.
(18) Ziegler, K.; Holzkamp, E.; Breil, H.; Martin, H. Angew. Chem.,
Int. Ed. 1955, 67 (16), 426−426.
(19) Ziegler, K.; Holzkamp, E.; Breil, H.; Martin, H. Angew. Chem.,
Int. Ed. 1955, 67 (19−2), 541−547.
(20) Tait, P. CHEMTECH 1975, 5 (11), 688−692.
(21) Sinn, H.; Kaminsky, W.; Vollmer, H.; Woldt, R. Angew. Chem.,
Int. Ed. Engl. 1980, 19 (5), 390−392.
(22) Kaminsky, W.; Miri, M.; Sinn, H.; Woldt, R. Makromol. Chem.,
Rapid Commun. 1983, 4 (6), 417−421.
(23) Kaminsky, W.; Hoff, M.; Derlin, S. Macromol. Chem. Phys. 2007,
208 (13), 1341−1348.
(24) Gates, D.; Svejda, S.; Onate, E.; Killian, C.; Johnson, L.; White,
P.; Brookhart, M. Macromolecules 2000, 33 (7), 2320−2334.
(25) Mattice, W. Macromolecules 1983, 16 (3), 487−490.
(26) Sworen, J.; Smith, J.; Wagener, K.; Baugh, L.; Rucker, S. J. Am.
Chem. Soc. 2003, 125 (8), 2228−2240.
(27) Sworen, J.; Smith, J.; Berg, J.; Wagener, K. J. Am. Chem. Soc.
2004, 126 (36), 11238−11246.
(28) Sworen, J.; Wagener, K. Macromolecules 2007, 40 (13), 4414−
4423.
(29) Rojast, G.; Wagener, K. Macromolecules 2009, 42 (6), 1934−
1947.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures and spectral data for all monomers
and polymers. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: wagener@chem.ufl.edu.
Present Address
§Department of Chemistry and the Beckman Institute for
Advanced Science and Technology, University of Illinois at
Urbana−Champaign, Urbana, IL 61801.
(30) Rojas, G.; Berda, E.; Wagener, K. Polymer 2008, 49 (13−14),
2985−2995.
Notes
(31) Rojas, G.; Inci, B.; Wei, Y. Y.; Wagener, K. B. J. Am. Chem. Soc.
2009, 131 (47), 17376−17386.
The authors declare no competing financial interest.
(32) Wagener, K.; Valenti, D.; Hahn, S. Macromolecules 1997, 30
(21), 6688−6690.
ACKNOWLEDGMENTS
■
(33) Smith, J.; Brzezinska, K.; Valenti, D.; Wagener, K. Macro-
molecules 2000, 33 (10), 3781−3794.
The authors thank the National Science Foundation (DMR-
0703261) and the International Max Planck Research School
for Polymer Materials (IMPRS) for financial support. The
Army Research Office contributed to the catalyst work
necessary to make proper selections. We also thank Michael
Bach at Max Planck Institute for Polymer Research for his help
with X-ray measurements and Sandra Seywald at Max Planck
Institute for Polymer Research for her help with GPC
measurements.
(34) Lieser, G.; Wegner, G.; Smith, J.; Wagener, K. Colloid Polym. Sci.
2004, 282 (8), 773−781.
(35) Baughman, T.; Sworen, J.; Wagener, K. Macromolecules 2006, 39
(15), 5028−5036.
(36) Zuluaga, F.; Inci, B.; Nozue, Y.; Hosoda, S.; Wagener, K.
Macromolecules 2009, 42 (14), 4953−4955.
(37) Hosoda, S.; Nozue, Y.; Kawashima, Y.; Suita, K.; Seno, S.;
Nagamatsu, T.; Wagener, K.; Inci, B.; Zuluaga, F.; Rojas, G.; Leonard,
J. Macromolecules 2011, 44 (2), 313−319.
(38) Baughman, T.; Sworen, J.; Wagener, K. Tetrahedron 2004, 60
(48), 10943−10948.
REFERENCES
■
(39) Rojas, G.; Baughman, T.; Wagener, K. Synth. Commun. 2007, 37
(22−24), 3923−3931.
(1) Till, P. J. Polym. Sci. 1957, 24 (106), 301−306.
(2) Strobl, G. Progress in Understanding of Polymer Crystallization;
Springer: Heidelberg, 2007.
(40) Rojas, G.; Wagener, K. J. Org. Chem. 2008, 73 (13), 4962−4970.
(41) Ulman, M.; Grubbs, R. J. Org. Chem. 1999, 64 (19), 7202−7207.
(42) Hong, S.; Day, M.; Grubbs, R. J. Am. Chem. Soc. 2004, 126 (24),
7414−7415.
(3) Organ, S.; Keller, A. J. Mater. Sci. 1985, 20 (5), 1571−1585.
(4) Organ, S.; Keller, A. J. Mater. Sci. 1985, 20 (5), 1586−1601.
(5) Organ, S.; Keller, A. J. Mater. Sci. 1985, 20 (5), 1602−1615.
3375
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376

Macromolecules
Article
(43) Hong, S.; Wenzel, A.; Salguero, T.; Day, M.; Grubbs, R. J. Am.
Chem. Soc. 2007, 129 (25), 7961−7968.
(44) Courchay, F.; Sworen, J.; Ghiviriga, I.; Abboud, K.; Wagener, K.
Organometallics 2006, 25 (26), 6074−6086.
(45) Fokou, P.; Meier, M. J. Am. Chem. Soc. 2009, 131 (5), 1664−
1665.
(46) Grant, D.; Paul, E. J. Am. Chem. Soc. 1964, 86 (15), 2984−2990.
(47) Liu, W.; Ray, D.; Rinaldi, P. Macromolecules 1999, 32 (11),
3817−3819.
(48) Axelson, D.; Levy, G.; Mandelkern, L. Macromolecules 1979, 12
(1), 41−52.
(49) Kuroda, N.; Nishikitani, Y.; Matsuura, K.; Ikegami, N.
Macromolecules 1992, 25 (11), 2820−2827.
(50) Rossi, A.; Zhang, J.; Odian, G. Macromolecules 1996, 29 (7),
2331−2338.
(51) Luruli, N.; Heinz, L.; Grumel, V.; Brull, R.; Pasch, H.;
Raubenheimer, H. Polymer 2006, 47 (1), 56−66.
(52) Randall, J. C. J. Polym. Sci., Part B: Polym. Phys. 1973, 11 (2),
275−287.
(53) Striegel, A. M.; Krejsa, M. R. J. Polym. Sci., Part B: Polym. Phys.
2000, 38 (23), 3120−3135.
(54) Hosoda, S.; Nomura, H.; Gotoh, Y.; Kihara, H. Polymer 1990,
31 (10), 1999−2005.
(55) Butler, M.; Donald, A.; Bras, W.; Mant, G.; Derbyshire, G.;
Ryan, A. Macromolecules 1995, 28 (19), 6383−6393.
(56) Russell, K.; McFaddin, D.; Hunter, B.; Heyding, R. J. Polym. Sci.,
Part B: Polym. Phys. 1996, 34 (14), 2447−2458.
(57) Seto, T.; Hara, T.; Tanaka, K. Jpn. J. Appl. Phys. 1968, 7 (1),
31−42.
(58) Hu, W.; Sirota, E. Macromolecules 2003, 36 (14), 5144−5149.
(59) Inci, B.; Wagener, K. J. Am. Chem. Soc. 2011, 133 (31), 11872−
11875.
(60) Iakoubovskii, K.; Mitsuishi, K.; Nakayama, Y.; Furuya, K.
Microsc. Res. Tech. 2008, 71 (8), 626−631.
(61) Egerton, R. F. Electron Energy Loss Spectroscopy in the Electron
Microscope; Plenum Press: New York, 1996.
3376
dx.doi.org/10.1021/ma3002577 | Macromolecules 2012, 45, 3367−3376