Precision vinyl acetate/ethylene (VAE) copolymers by ROMP of acetoxy-substituted cyclic alkenes

Article abstract of DOI:10.1021/ma400092z

Precision linear vinyl acetate/ethylene (VAE) copolymers containing acetoxy groups on precisely every eighth backbone carbon were synthesized by ring-opening metathesis polymerization (ROMP) of racemic 3-acetoxy cyclooctene (3AcCOE) followed by hydrogenation. The use of enantiomerically pure 3AcCOE resulted in an optically active polyalkenamer that afforded isotactic precision VAE materials after hydrogenation. Both of these VAE polymers are semicrystalline (by differential scanning calorimetry and wide-angle X-ray scattering) due to their high degrees of regioregularity and the isotactic VAE samples exhibited a higher apparent degree of crystallinity and melting point compared to the atactic version. In contrast, analogous linear VAE copolymers derived from ROMP-hydrogenation of racemic 4- or 5-acetoxy cyclooctenes were regio-irregular and completely amorphous. The ROMP-hydrogenation of 3-acetoxy cycloheptene also affords precision linear VAE copolymers with acetoxy groups on every seventh carbon, but this polymer was noncrystalline. Mechanical characterization showed that the precision 3AcCOE-derived VAE samples possess improved mechanical properties compared to the compositionally similar commercial VAE copolymers produced by radical copolymerization.

Full text of DOI:10.1021/ma400092z

Article
pubs.acs.org/Macromolecules
Precision Vinyl Acetate/Ethylene (VAE) Copolymers by ROMP of
Acetoxy-Substituted Cyclic Alkenes
Jihua Zhang, Megan E. Matta, Henry Martinez, and Marc A. Hillmyer*
Department of Chemistry, University of Minnesota, Minneapolis, Minnesota 55455, United States
S
* Supporting Information
ABSTRACT: Precision linear vinyl acetate/ethylene (VAE)
copolymers containing acetoxy groups on precisely every
eighth backbone carbon were synthesized by ring-opening
metathesis polymerization (ROMP) of racemic 3-acetoxy
cyclooctene (3AcCOE) followed by hydrogenation. The use
of enantiomerically pure 3AcCOE resulted in an optically
active polyalkenamer that afforded isotactic precision VAE
materials after hydrogenation. Both of these VAE polymers are
semicrystalline (by differential scanning calorimetry and wide-
angle X-ray scattering) due to their high degrees of
regioregularity and the isotactic VAE samples exhibited a higher apparent degree of crystallinity and melting point compared
to the atactic version. In contrast, analogous linear VAE copolymers derived from ROMP−hydrogenation of racemic 4- or 5-
acetoxy cyclooctenes were regio-irregular and completely amorphous. The ROMP−hydrogenation of 3-acetoxy cycloheptene
also affords precision linear VAE copolymers with acetoxy groups on every seventh carbon, but this polymer was noncrystalline.
Mechanical characterization showed that the precision 3AcCOE-derived VAE samples possess improved mechanical properties
compared to the compositionally similar commercial VAE copolymers produced by radical copolymerization.
recision control of polymer structure is important for both
tailoring ultimate properties and gaining a complete
using the well-defined Grubbs second (G2) or third (G3)
generation catalyst leads to highly regio- and stereoregular
polyalkylenamers.11 The polymers possess high levels (>95%)
of head-to-tail (HT) repeating unit connectivity and E double
bond configurations. Hydrogenation of these polyalkylenamers
afforded model linear low density polyethylene (LLDPE) with
precisely positioned branches. In a recent computational
study,12 we proposed that the rate-limiting step for the
polymerization of COE and its derivatives by Grubbs type
catalysts is determined by the barrier for breakdown of the
metallacyclobutane intermediate formed between the catalyst
and the monomer. Consequently, the regio- and stereo-
chemistry seen in the ROMP of 3RCOEs is mainly due to
significant repulsive interaction between the substituent and the
N-heterocyclic carbene (NHC) mesityl ligand.
These reports suggest that any 3-substituted COE having a
sufficiently bulky functional group could also lead to regio- and
stereoregular polymers.13 Furthermore, the use of enantiomeri-
cally pure 3-substituted COEs in such polymerization would
enable the synthesis of isotactic, all-E, and sequence specific
polymers. To test this hypothesis we focused on the acetoxy-
substituted system14 (Figure 1) because of the convenient
accessibility of the monomers, the introduction of a functional
group and the practical value of the final polymers, as they are
P
understanding the structure−property relationships. Biomacro-
molecules often possess high degrees of specificity and
structural complexity not commonly found in synthetic
polymers. In synthetic polymers the control of size (molar
mass) and architecture has been demonstrated in various
approaches during the past several decades.1 Polymer micro-
structures (i.e., comonomer sequence and stereochemistry), on
the other hand, are more challenging to regulate.2 The control
over repeating unit sequence has been achieved mainly through
polymerization of prestructured monomers with built-in
sequence,3 coupling of sequence-specific short oligomers,4
and particular copolymerization processes in which monomers
exhibit different reactivity5 or where external templates are
applied.6 One of the notable examples of the first approach is
the preparation of precision polyethylene by acyclic diene
metathesis (ADMET) polymerization7 of symmetrically sub-
stituted α,ω-dienes.3g−i Alternating copolymerization of 1,3-
butadiene with other vinyl monomer (e.g., methyl methacry-
late) and polymerization of substituted 1,3-butadiene were
reported to eventually result in periodic ethylene copolymers.3c
Use of stereoselective catalysts is by far the most popular
strategy to control backbone stereochemistry which includes
E−Z isomerism and tacticity.8 However, the complete control
over both the sequence and stereochemistry has only been
realized in limited cases.9
Recently, we discovered that ring-opening metathesis
polymerization (ROMP)10 of cis-cyclooctene (COE) substi-
tuted with hydrocarbon groups at the 3-position (3RCOE)
Received: January 14, 2013
Revised: March 11, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 1. Proposed synthesis of regio- and stereoregular polymer by ROMP of 3AcCOE catalyzed by G2 or G3.
comparable to copolymers of vinyl acetate (VA) and ethylene
prepared by radical copolymerization.
impurities complicate the analysis of the structure regularities
of the resulting polymers. Therefore, we employed a three-step
sequence consisting of α-bromination, hydrolysis, and acetox-
ylation from COE to synthesize 3AcCOE23 (Scheme 1). Pure
3AcCOE could be obtained on multigram scale and in overall
yields of ∼60%.
Copolymers of VA and ethylene (EVA when ethylene >50 wt
% or VAE when vinyl acetate >50 wt %) and their hydrolysis
products are widely used as adhesives, coatings and foams.15
Industrial production of EVA/VAE involves high-pressure free
radical conditions similar to those used in the production of
low density polyethylene; branched polymers with difficult-to-
control structures are typically produced. Linear EVA can be
synthesized by palladium-catalyzed coordination copolymeriza-
tion of ethylene and vinyl acetate, but the incorporation of vinyl
acetate is limited to ∼2 mol %.16 Poly(ethylene-alt-vinyl
acetate) (75 wt % VA) was synthesized by Yokota et al17
through cationic polymerization of 1-trimethylsilyloxy-1,3-
butadiene followed by several steps of postpolymerization
modification, but significant branching (∼15%) was unavoid-
able during the polymerization.
Scheme 1. Synthesis of 3-Acetoxy-Substituted Cyclic Alkenes
Combination of metathesis polymerization and hydro-
genation is an attractive way to generate perfectly linear
EVA/VAE copolymers due to the branch-free nature of the
polymerization and the tolerance of certain metathesis catalysts
to esters. For example, in 1995 the Grubbs group18 employed
ROMP of 5-acextoxycyclooctene (5AcCOE) followed by
hydrogenation to synthesize linear VAE with ∼50 wt % VA.
The work of Wagener and his group19 demonstrated the ability
to tune the composition of EVA by performing ROMP or
ADMET polymerization of appropriately functionalized como-
nomers. However, in these examples the EVA/VAE analogues
were regio-irregular and atactic. Using symmetrically sub-
stituted monomers in ADMET, Wagener and co-workers20 also
prepared model linear EVA materials with 20−25 wt % VA that
have precisely positioned acetoxy groups on the backbone.
Precision linear EVA/VAE polymers with >25 wt % VA have
rarely been reported, and the above examples involved only
symmetrically substituted monomers.9c,21
The solution polymerization of 3AcCOE using G2
proceeded smoothly in dry CHCl3; and the monomer
conversion is quantitative within hours at room temperature.
Polymer molar mass can be efficiently controlled by using cis-4-
octene as a chain transfer agent (CTA),24 and the dispersity
values (Đ ∼ 1.7) are similar to that obtained for poly(COE) by
the ROMP of the unsubstituted variant (Figure 2).
The microstructure of poly(3AcCOE) was carefully charac-
terized by 1H and 13C NMR (Figure 3) and by FT-IR
spectroscopies. In the 13C NMR spectrum, the polymer shows
two major olefinic signals (128.51 and 134.35 ppm) as opposed
In this study, we describe the use of tandem ROMP−
hydrogenation of 3-acetoxycyclooctene (3AcCOE) and its
enantiomerically pure version to synthesize “precision” model
polymers that are equivalent to linear VAE materials (50 wt %
VA). We also studied the effect of structural variations
including tacticity, substituent position on COE and ring-size
of 3-acetoxy cyclic alkenes on the regio- and stereoselectivity of
the polymerization as well as on the properties of the final VAE
copolymers.
RESULTS AND DISCUSSION
■
Synthesis and ROMP of 3AcCOE. We attempted the
direct allylic transition metal catalyzed acetoxylation of cis-
cyclooctene, which has been reported in literature with various
cyclic alkenes.22 However, approximately 5% the homoallylic
product (4-acetoxy cyclooctene) was inevitably generated and
proved to be very difficult to separate from the desired allylic
product. Although a convenient synthetic method, such
Figure 2. Molar mass and dispersity of poly(3AcCOE) as a function of
monomer to CTA ratio. Conditions: [M]0 ≈ 1.5 M, [M]0/[G2]0 =
4000. Monomer conversion was quantitative in all cases. The molar
mass was determined by end group analysis using 1H NMR
spectroscopy assuming two end groups from the CTA per chain.
The dispersity was determined by SEC using polystyrene standards in
CHCl3.
B
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
Figure 3. H NMR (top) and 13C NMR (bottom) spectra of poly(3AcCOE). Run 1 in Table 1.
a
Table 1. ROMP of 3AcCOE in Different Solvents at Different Temperatures
c
b
b
c
c
run
solvent
CHCl3
temp (°C)
convn (%)
Mn (kg/mol)
Đ
E (%)
HT (%)
1
2
3
4
5
6
20
20
20
40
60
90
>99
>99
>99
>99
>99
>99
29.5
33.3
35.1
34.2
35.6
1.60
1.71
1.64
1.68
1.67
1.74
95.9
90.1
93.8
90.2
89.4
88.7
99.7
97.7
99.0
98.2
97.6
97.3
cyclohexane
toluene
toluene
toluene
toluene
33.7
a
b
Conditions: [M]0 ≈ 1.5 M, [M]0/[G2]0 = 4000, [M]0/[CTA]0 = 100, 24 h. Determined by SEC using polystyrene standards in CHCl3.
c
1
Determined by H NMR spectroscopy of the reaction mixture.
to eight signals from the six possible regio- and stereoisomeric
structures, and one major signal for the carbon adjacent to the
acetoxy group (75.11 ppm).18 This is strong evidence for high
regio- and stereo(E/Z)-regularity of the polymer. Further
analysis of the NMR (Figures S7−S11, Supporting Informa-
tion) and FT-IR spectra (Figure S62, Supporting Information)
supports a polymer with a HT structure and E configuration of
the double bonds, the same selectivity as shown previously for
other 3RCOE samples.11 The stereoselectivity was slightly
lower than those of the 3RCOE polymers obtained under the
same conditions. This result might be explained by the
differences in sterics and polarity between acetoxy group and
the alkyl/aryl group, which may result in a slightly higher
fraction of Z product of poly(3AcCOE). Very low levels of the
regio- and stereochemical impurities were observed by NMR
spectroscopy as mostly the Z-HT units.
The regio- and stereoselectivity of the polymerization was
affected by both the polymerization solvent and temperature.
As shown in Table 1, polymerization in CHCl3 gave the highest
regio- and stereoselectivity. Performing the polymerization in
less polar solvent (e.g., toluene or cyclohexane) tends to lead to
slight decrease of the regio- and stereoselectivity. This result
suggests the transition state for the rate-determining step
(break down of the metallocycle) that results in the HT and E
product is more polar than the ones leading to other products
and is thus stabilized to a greater extent in solvents with higher
dielectric constants.12 Furthermore, in toluene the stereo-
selectivity of the polymer decreases as the polymerization
temperature increases, and a concomitant decrease of the HT
selectivity was also observed at elevated temperatures. Never-
theless, it can be concluded that poly(3AcCOE) is highly regio-
and stereoregular regardless the polymerization conditions
explored here (Table 1).
To better understand the regioselectivity observed during
ROMP of 3AcCOE, we performed quantum chemical
calculations for the breakdown of the metallocyclobutane
intermediate during the initiation step as we described
previously for the aliphatic and aromatic cycloocetenes.12
Starting from the lowest energy conformers we calculated both
distal and proximal approaches in the gas and solution (CHCl3)
states (Figure 4). The calculations indicated that the distal
approach is thermodynamically more favored by over 12 kcal/
mol. Coordination of the acetoxy group with the Ru metal
center was not observed in either approach. Free energies and
Boltzmann-weighted product distributions are provided in the
Figure 4. Optimized M06-L E TS2 structures for the proximal and
distal reactions of 3AcCOE. Hydrogens were omitted for simplicity.
C
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Supporting Information. We also attempted a 1:1 3AcCOE:Ru
experiment in an attempt to verify this result as we did
previously.11 However, the resultant NMR data was incon-
clusive and we were unable to make definitive determination
concerning the approach of 3AcCOE to the Ru center.
ROMP of other Acetoxy-Substituted Cyclic Alkenes. In
addition to the ROMP of 3AcCOE, we also (re)examined the
polymerization of two regio-isomers of 3AcCOE, namely 4-
acetoxy cyclooctene (4AcCOE) and 5-acetoxy cyclooctene
(5AcCOE) (Scheme 2). NMR kinetic experiments showed that
(3AcCHE) using a procedure similar to the synthesis of
3AcCOE (Scheme 1). No polymerization was observed with
3AcCPE, even when the polymerization was conducted at room
temperature for a week, which is likely due to its low ring
strain.21 As for 3AcCHE, the reactivity was significantly lower
than 3AcCOE, so higher concentration, lower temperature and
longer reaction times were required to obtain high molar mass
polymer ([3AcCHE]0 ≈ 3 M in chloroform, 48 h, 55%
conversion). This reduced reactivity is likely related to the
lower ring strain of 3AcCHE compared to 3AcCOE. The
polymerization of 3AcCHE also shows preference for the
formation of the E-HT polymer as shown by the two major
olefinic signals in the 13C NMR spectra (Figure 5c). However,
several minor olefinic signals are noticeable in comparison to
the 13C NMR spectra of poly(3AcCOE) (Figure 5d), indicating
that the regularity is lower than that of poly(3AcCOE).
Quantitative analysis of the 1H NMR spectrum of poly-
(3AcCHE) (Figure S35, Supporting Information) revealed a
significant decrease of the stereoselectivity (78.4% E) but only a
slight decrease of the regio-selectivity (95.9% HT) relative to
poly(3AcCOE).
Scheme 2. ROMP of 4AcCOE, 5AcCOE, and 3AcCHE
Hydrogenated polymers. All the acetoxy-substituted
polyalkylenamers can be hydrogenated using diimide25 (Table
2), generating polymers equivalent to VAE samples with 50.5
wt % VA (for 3-, 4-, or 5AcCOE) or 55.1 wt % VA (for
3AcCHE). Since all the starting monomers are racemic
compounds, these polymers are expected to be atactic. In all
cases, the extent of hydrogenation was >99% and the acetoxy
group was not affected as indicated by NMR spectroscopy. We
expected that the hydrogenated polymers should retain the
regioregularity from their parent polyalkylenamers, and this is
supported by 13C NMR analysis of the hydrogenated polymers
shown in Figure 6. The polymer obtained from poly(3AcCOE)
(a-HP3AcCOE, “a” for atactic) displays a single signal for both
the carbonyl carbon and the methine carbon in the 13C NMR
spectrum (Figure 6a), supporting a highly regioregular
structure. As poly(3AcCOE) has a HT structure, a-HP3AcCOE
is the equivalent of a model VAE copolymer with pendent
acetoxy group on precisely every eighth backbone carbon.
3AcCOE and 4AcCOE polymerize more slowly than 5AcCOE
(Figure S1, Supporting Information). The reduced reactivity of
3AcCOE and 4AcCOE can be attributed to the steric impact of
the acetoxy group being in close proximity to the metathesis
reactive double bond.
NMR studies revealed that both poly(4AcCOE) and
poly(5AcCOE) are completely regio-irregular as equal
intensities of the resonances corresponding to the HH, HT
and TT polymers can be seen in the 13C NMR spectra (Figure
5a and b). The E/Z ratio of both polymers is approximately 80/
20, which is close to the ratio observed in poly(COE)
generated using G2.
To explore the impact of ring-size on the selectivity in the
ROMP of 3-substituted cyclic olefins, we prepared 3-
acetoxycyclopentene (3AcCPE) and 3-acetoxycycloheptene
Figure 5. 13C NMR spectra (olefinic region) of poly(4AcCOE) (a), poly(5AcCOE) (b) and poly(3AcCHE) (c). For comparison, 13C NMR
spectrum of poly(3AcCOE) in the same region is shown in part d.
D
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 2. Characterization Data of Polymers Derived from Acetoxy-Substituted Cyclic Alkenes
unsaturated polymer
hydrogenated polymer
a
a
b
c
a
a
b
b
b
monomer
Mn (kg/mol)
Đ
Tg (°C)
VA (wt %)
Mn (kg/mol)
Đ
Tg (°C)
Tm (°C)
Tc (°C)
3AcCOE
(R)-3AcCOE
4AcCOE
36.4
36.4
22.1
41.7
18.4
1.64
1.58
1.87
1.74
1.62
−33
−40
−43
−40
50.6
50.6
50.6
50.6
55.1
40.0
37.4
25.9
51.7
18.9
1.60
1.54
1.53
1.57
1.56
−36
−30
−41
−42
−46
53
91
9
59
5AcCOE
3AcCHE
−31
b
a
c
Determined by SEC using polystyrene standards. Determined by DSC on the 2nd heating/cooling cycle at 10 °C/min. Effective vinyl acetate
content in the polymer.
Figure 6. Partial 13C NMR spectra of a-HP3AcCOE (a), a-HP4AcCOE (b), a-HP5AcCOE (c), a-HP3AcCHE (d), and commercial VAE50 (see text
for details) (e). Left: carbonyl carbon. Right: methine carbon.
Scheme 3. Preparation of Enantiomerically Pure 3AcCOE
For comparative purposes the 13C NMR spectra of
hydrogenated poly(4AcCOE) (a-HP4AcCOE) (Figure 6b)
and poly(5AcCOE) (a-HP5AcCOE) (Figure 6c) show multiple
signals for the carbonyl and methine carbons, indicating their
regioirregularity. Polymer a-HP4AcCOE has a more complex
spectrum than a-HP5AcCOE. This is caused by the fact that the
spacing between the pendent acetoxy groups is either 4, 7, or
10 carbons for a-HP4AcCOE, and 6, 7, or 8 carbons for a-
HP5AcCOE, so the difference of chemical environment of the
carbonyl and methine carbons is more exaggerated for a-
HP4AcCOE than for a-HP5AcCOE. On the other hand, a
commercial sample of a VAE copolymer with 50 wt % VA
E
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(VAE50, Mn = 83.7 kg/mol, Mw/Mn = 4.01, purchased from
Scientific Polymer Products, Inc.) exhibits a complex set of 13C
signals (Figure 6e), confirming its irregular and inhomogeneous
structure.
(Figure S64, Supporting Information). Although poly-
(3AcCHE) also shows a comparably high Tg, its higher VA
content may be a more dominating factor than its structure
regularities.
The DSC profile of a-HP3AcCOE displays a broad
endotherm (curve 1 in Figure 7a) during the heating cycle
Isotactic Precision VAE. The regio-selective ROMP of
enantiomerically pure 3AcCOE should give precision polymers
that are regio-selective, E−Z stereoselective and isotactic. To
synthesize a pure enantiomer of 3AcCOE we initially utilized
the enzyme-catalyzed kinetic resolution of secondary alcohols
(or acetates) as this has been shown to give product with good
to excellent enantioselectivity.26 We performed transesterifca-
tion of 3-hydroxy cyclooctene and vinyl acetate catalyzed by a
lipase (Amano PS, from Burkholderia cepacia) in n-hexane
(Scheme S1, Supporting Information). The 3AcCOE obtained
from this reaction was enriched with the (R)-enantiomer, but
the enantiomeric excess (ee) never exceeded 55%, as
determined by comparison of its optical rotation with that of
the pure enantiomer obtained by the method described below.
Gais and co-workers developed a highly efficient method to
prepare various enantiomerically pure allylic compounds using
a chiral palladium complex.27 We utilized a modified version of
this method to prepare enantiomerically pure 3AcCOE. As
shown in Scheme 3, 3-(methoxycarbonyl)cyclooctene (3Moc-
COE) was treated with sodium benzenesulfinate in the
presence of catalytic amount of Pd2(dba)3·CHCl3 and the
chiral Trost ligand (BPA) in CH2Cl2/H2O. After 24 h reaction
at ambient temperature, unreacted 3MocCOE was isolated in
32% yield and >99% ee ((R)-enantiomer) as determined by
comparison of its optical rotation with that of the pure
enantiomer reported in literature.27 Hydrolysis of the
methoxycarbonyl group gave the corresponding 3-hydroxycy-
clooctene, and the optical rotation of which ([α]20D = −52.4°, c
= 1.0, CH2Cl2) was identical to the reported value of an
Figure 7. DSC profiles of various VAE copolymers at 10 °C/min. (a)
Heating curves of a-HP3AcCOE (1), i-HP3AcCOE (2), a-
HP4AcCOE (3), a-HP5AcCOE (4), a-HP3AcCHE (5), and
commercial VAE50 (see text for details) (6). (b) Cooling curves of
a-HP3AcCOE (7), i-HP3AcCOE (8), and commercial VAE50 (see
text for details) (9).
and a broad exotherm (curve 7 in Figure 7b) during the cooling
cycle. These thermal transitions can be related to the melting
(Tm = 53 °C) and crystallization (Tc = 9 °C) of the polymer,
respectively. A glass transition (Tg = −36 °C) was also
observed. The i-HP3AcCOE shows much higher Tm and Tc (91
and 59 °C, respectively). The greater heat of fusion of i-
HP3AcCOE (ΔHm = 67 J/g) relative to that of a-HP3AcCOE
(ΔHm = 39 J/g) as well as its barely distinguished Tg at −30 °C
suggests a considerably higher degree of crystallinity for i-
HP3AcCOE. The shoulder/secondary peak appears on the low
temperature side of the major melting transition suggests there
exist more than one type of crystal structure in i-HP3AcCOE.
Interestingly, the heat of fusion of a-HP3AcCOE or i-
HP3AcCOE is greater than that of the analogous ROMP-
derived precision polymer with methyl substituent (ΔHm = 30
J/g)11 despite the smaller size of the methyl group. This is
likely due to the dipolar nature of the acetoxy groups.
The Tm of a-HP3AcCOE and i-HP3AcCOE are remarkably
high compared to analogous ethylene/vinyl acetate copolymers
produced by radical copolymerization. The Tm of commercial
EVA increases with increasing E content. However, HP3Ac-
COE (50 wt % VA) melts at the temperature close to the range
of Tm of ADMET-derived precision EVA with 20 wt % VA20 or
commercial EVA with 30−40 wt % VA.32 Such distinct and
defined melting and crystallization behaviors have not been
observed in the case of other precision polyethylene derivatives
with even lower frequency of substitution (on every 9th or 15th
carbon).33
enantiomerically pure sample ([α]20 = −52.4°, c = 1.46,
D
CH2Cl2).27a This compound was further converted into the
corresponding (R)-3AcCOE by acetylation ([α]20D = −89.2°, c
= 1.0, n-hexane; lit.28 for the R enantiomer of unknown ee,
[α]25D = −50.5°, c = 5.08, n-hexane; lit.29 for the S enantiomer
of 50% ee, [α]29 = +47.0°, c = 5.01, n-hexane). Since the
D
stereochemistry of the chiral center should not be affected
during the acetylation, we presume that this product is also
enantiomerically pure (i.e., > 99% ee).30
The reactivity of (R)-3AcCOE in ROMP was no different
from the racemic 3AcCOE. Likewise, ROMP of (R)-3AcCOE
was also regio- and E/Z stereoselective, as indicated by the
identical spectroscopic data for the resulting polymer. Also,
poly(R-3AcCOE) is optically active, exhibiting significant
optical rotation ([α]20 = +55°, c = 1.0, CHCl3) that has the
D
opposite sign to the monomer. The optical rotation of poly(R-
3AcCOE) was also dependent linearly on the ee of the
monomer (Figure S2, Supporting Information). Chemical
hydrogenation of poly(R-3AcCOE) gives corresponding VAE
(i-HP3AcCOE, “i” for isotactic) which has a both precision and
isotactic structure. However, the optical rotation of i-
HP3AcCOE is close to zero ([α]20 = +2.4°, c = 1.0,
D
CHCl3) because effective mirror planes of symmetry are
present due to the high levels of regioregularity as in the case of
isotactic and syndiotactic polypropylenes which are optically
inactive.31
Thermal Characterization. DSC was employed to
characterize the thermal behavior of the unsaturated and
saturated polymers (Table 2). All of the unsaturated polymers
are completely amorphous materials with Tg values < −30 °C
Using the conditions of run 1 in Table 1 and the typical
hydrogenation procedure, we synthesized a series of
HP3AcCOE from 3AcCOE samples with different ee’s
prepared by mixing appropriated amounts of racemic
3AcCOE and enantiomerically pure (R)-3AcCOE. DSC
F
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
analysis of this series of polymers shows a linear relationship
between Tm (and Tc) and the ee of monomer (Figure 8a). The
shown in Figure 9. As compared to unstretched samples
(Figure S3, Supporting Information), the 1D WAXS profiles of
Figure 9. Room temperature 1D WAXS patterns of stretched samples
of a-HP3AccOE (top) and i-HP3AcCOE (bottom) with the X-ray
incident beam perpendicular to the stretching direction. Patterns
derived from azimuthal integration of the indicated area in Figure
S5(b) and S6(b), Supporting Information, respectively. Curves are
shifted vertically for clarity.
Figure 8. Dependence of thermal properties of HP3AcCOE on the ee
▲
of 3AcCOE. (a) Temperatures of melting (Tm, ) and crystallization
(Tc, ). (b) Heat of fusion (ΔHm).
these stretch samples exhibited considerably lower intensity
amorphous halo signals, indicating a significant increase of
crystallinity upon stretching. Small angle X-ray scattering
(SAXS) of unaligned samples gave broad primary scattering
peaks with d ∼ 12−14 nm (Figure S4, Supporting
Information), indicating these polymer possesses lamellae
structure of alternating crystalline and noncrystalline regions,
as observed in typical polyethylene.37
▼
heat of fusion also increases with ee (Figure 8b). This result
indicates that the crystallinity and melting point of HP3AcCOE
is influenced by tacticity in a predictable way, which allows for
precise control of crystallinity and related properties.
In contrast, the commercial sample VAE50 shows ambiguous
and broad melting and crystallization transitions that spans
nearly 100 °C (curves 6 and 9 in Figure 7), accompanied by a
much lower heat of fusion (ΔHm = 17 J/g), suggesting much
lower crystallinity than the precision VAE. DSC of the regio-
irregular a-HP4AcCOE or a-HP5AcCOE only show glass
transitions (curves 3 and 4 in Figure 7a). The more randomly
distributed substituents suppress crystallinity in the polymer,
resulting in amorphous material as seen in other substituted
polyethylene (PE) systems.34 This phenomenon was also
observed with some ethylene−ethylene−acrylic copolymers
(e.g., ethylene−ethylene−methyl α-chloroacrylate copolymer),
where the periodic copolymer is semicrystalline but the random
analogue is amorphous.35 Interestingly, we found that a-
HP3AcCHE is amorphous (curve 5 in Figure 7a) even though
it is also highly regioregular (95.8% HT). This result seems to
suggest that a minimum run of 7 methylene units between the
acetoxy groups is required for crystallinity of VAE/EVA
copolymers. This observation also agrees with the result from
model studies of methyl-substituted PE,36 which suggested that
crystallinity only exists with less than an average of 15 methyl
branches per 100 carbon atoms (there are 14.3 acetoxy
branches per 100 carbon atoms for HP3AcHE while 12.5
branches per 100 carbon atoms for HP3AcCOE).
Polyethylene can adopt orthorhombic or triclinic unit
cells36,38 and we posit that the crystal structures adopted by
a-HP3AcCOE and i-HP3AcCOE belong to one of these crystal
families given their structural similarity to polyethylene. For
orthorhombic polyethylene the primary scattering peak is at d =
4.08 Å and for triclinic it is at d = 3.80 Å. Since the acetoxy side
group must occupy space in the unit cell, and thus lead to
expansion relative to linear polyethylene, a triclinic unit cell is
the more likely candidate. In both WAXS patterns the distinct
peaks at 6.25 and 6.71 Å likely arise from acetoxy−acetoxy
correlations in the unit cell of the crystal structure. Given that
these distances are shorter than the distance between adjacent
acetoxy groups on the same chain in the fully extended
conformation, the chains in the unit cell are likely interdigitated
with the polar acetoxy groups on different chains shifted along
one axis in the unit cell. While further studies are needed to
elucidate the crystal structures of a-HP3AcCOE and i-
HP3AcCOE, we can unambiguously conclude that the WAXS
patterns of both a-HP3AcCOE and i-HP3AcCOE are unusual
for EVA/VAE, as only an amorphous halo has been seen when
the VA content >40−45%.39 Therefore, these observations can
be certainly related to the precise microstructure of these
polymers.
X-ray Scattering Measurements. The semicrystalline
structure of both a-HP3AcCOE and i-HP3AcCOE was
supported by wide-angle X-ray scattering (WAXS) measure-
ments. One-dimensional (1D) WAXS of samples that were
stretched to their respective maximum elongations (∼500% for
a-HP3AcCOE and ∼200% for i-HP3AcCOE) at room
temperature (above their glass transition temperatures) are
Mechanical Properties. Figure 10 shows the representa-
tive stress−strain curves for a-HP3AcCOE, i-HP3AcCOE and
the commercial VAE50 at room temperature. The Young’s
modulus of i-HP3AcCOE (133 23 MPa) is about 50% higher
than that of a-HP3AcCOE (90 15 MPa), and both are more
than 30−40 times higher than that of VAE50 (3.1 0.4 MPa).
Between the two regioregular polymers, a-HP3AcCOE has
G
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
here not only create an efficient pathway to approach
functionalized precision polymers but also demonstrate the
achievement of different levels of control over polymer
structure, which provides useful information for understanding
the structure−property relationship of a wider range of
polymers.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details and characterization 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: hillmyer@umn.edu.
■
Figure 10. Representative stress−strain curves of a-HP3AcCOE, i-
HP3AcCOE and the commercial VAE50. Inset: amplification of the
low strain region. The mark “x” denotes failure.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
higher tensile strength (24 5 vs 14 5 MPa) but lower yield
strength (6.9 0.3 vs 11 2 MPa) than i-HP3AcCOE. These
two polymers also exhibit much higher tensile strength than the
random VAE50 (8.6 1.5 MPa). i-HP3AcCOE fails at only
180−220% elongation while a-HP3AcCOE can be elongated to
450−500% its original length, but they are both less ductile
than VAE50 that exhibits ultimate elongations of 900−1000%.
Both a-HP3AcCOE and i-HP3AcCOE behave as typical
semicrystalline thermoplastics. The mechanical behavior of
each polymer is consistent with the degree of crystallinity of the
respective polymer.40 The result of tensile tests highlights the
effect of microstructural regularity (i.e., sequence and tacticity)
on the mechanical properties such as stiffness, toughness and
ductility of compositionally similar polymers.
■
The authors of this paper would like to thank the Abu Dhabi-
Minnesota Institute for Research Excellence (ADMIRE) for
financial support of this research. The tensile tests and X-ray
measurements were carried out at College of Science and
Engineering Characterization Facility, at the University of
Minnesota. We also thank Youngmin Lee for assistance with
NMR experiments. The authors are also grateful for resources
from the University of Minnesota Supercomputing Institute.
REFERENCES
■
(1) (a) Webster, O. W. Science 1991, 251, 887−893. (b) Maty-
jaszewski, K. Prog. Polym. Sci. 2005, 30, 858−875. (c) Hadjichristidis,
N.; Iatrou, H.; Pitsikalis, M.; Mays, J. Prog. Polym. Sci. 2006, 31, 1068−
1132. (d) Ouchi, M.; Terashima, T.; Sawamoto, M. Acc. Chem. Res.
2008, 41, 1120−1132. (e) Ouchi, M.; Terashima, T.; Sawamoto, M.
Chem. Rev. 2009, 109, 4963−5050.
(2) (a) Lutz, J. F.; Pakula, T.; Matyjaszewski, K. ACS Symp. Ser. 2003,
854, 268−282. (b) Badi, N.; Lutz, J. F. Chem. Soc. Rev. 2009, 38,
3383−3390. (c) Lutz, J. F. Polym. Chem. 2010, 1, 55−62.
(d) Hartmann, L. Macromol. Chem. Phys. 2011, 212, 8−13.
(3) (a) Wu, Z.; Grubbs, R. H. J. Mol. Catal. 1994, 90, 39−42. (b) Wu,
Z.; Grubbs, R. H. Macromolecules 1995, 28, 3502−3508. (c) Yokota, K.
Prog. Polym. Sci. 1999, 24, 517−563. (d) Cho, I. Prog. Polym. Sci. 2000,
25, 1043−1087. (e) Cho, I. Macromol. Symp. 2003, 195, 89−94.
(f) Song, A. R.; Lee, J. C.; Parker, K. A.; Sampson, N. S. J. Am. Chem.
Soc. 2010, 132, 10513−10520. (g) Berda, E. B.; Baughman, T. W.;
Wagener, K. B. J. Polym. Sci., Part A: Polym. Chem. 2006, 44, 4981−
4989. (h) Rojas, G.; Berda, E. B.; Wagener, K. B. Polymer 2008, 49,
2985−2995. (i) Li, Z. L.; Li, L.; Deng, X. X.; Zhang, L. J.; Dong, B. T.;
Du, F. S.; Li, Z. C. Macromolecules 2012, 45, 4590−4598. (j) Park, H.;
Choi, T.-L. J. Am. Chem. Soc. 2012, 134, 7270−7273.
(4) (a) Tsarevsky, N. V.; Sumerlin, B. S.; Matyjaszewski, K.
Macromolecules 2005, 38, 3558−3561. (b) Lutz, J. F. Angew. Chem.,
Int. Ed. 2007, 46, 1018−1025. (c) Takizawa, K.; Tang, C.; Hawker, C.
J. J. Am. Chem. Soc. 2008, 130, 1718−1726. (d) Takizawa, K.; Nulwala,
H.; Hu, J.; Yoshinaga, K.; Hawker, C. J. J. Polym. Sci., Part A: Polym.
Chem. 2008, 46, 5977−5990. (e) Yu, T. B.; Bai, J. Z.; Guan, Z. B.
Angew. Chem., Int. Ed. 2009, 48, 1097−1101. (f) Berthet, M. A.;
Zarafshani, Z.; Pfeifer, S.; Lutz, J. F. Macromolecules 2010, 43, 44−50.
(g) Zhang, C. Y.; Ling, J.; Wang, Q. Macromolecules 2011, 44, 8739−
8743.
CONCLUSIONS
■
We have described the synthesis of several model linear VAE
(>50 wt % VA) copolymers by ROMP of acetoxy-substituted
cyclic alkenes and subsequent hydrogenation. All monomers
can be prepared through simple reactions and in high yields.
The polymer obtained from 3AcCOE was found to be highly
regio-and stereoregular, and thus eventually led to a precision
VAE analogue (a-HP3AcCOE). By employing enantiomerically
pure 3AcCOE, the same ROMP procedure afforded an
optically active polymer with both regularly and isotactically
placed side chains, from which an isotactic precision VAE (i-
HP3AcCOE) was obtained. This result showcases a function-
alized example of the precision polymers that we reported
previously by the regio- and stereoselective ROMP of 3-
substitued COE.11 Both a-HP3AcCOE and i-HP3AcCOE
exhibit unusual thermal and mechanical properties due to
their semicrystalline structures, which is the consequence of
their high structure regularity. The presence of substantial
crystallinity is unprecedented in any VAE copolymer with such
high substituent frequency. Regio-irregular counterparts were
synthesized from 4AcCOE and 5AcCOE; and both polymers
have essentially no crystallinity. Furthermore, increase of the
substitution frequency by polymerizing 3AcCHE resulted in a
precision yet amorphous VAE analogue. These ROMP-derived
VAE represent two distinct types of model polymers of the
microstructurally inhomogeneous commercial VAE, one of
which possesses a substantial degree of crystallinity and the
other of which is completely amorphous. The works presented
(5) (a) Inoue, Y.; Watanabe, J.; Takai, M.; Yusa, S.; Ishihara, K. J.
Polym. Sci., Part A: Polym. Chem. 2005, 43, 6073−6083. (b) Song, A.
R.; Parker, K. A.; Sampson, N. S. J. Am. Chem. Soc. 2009, 131, 3444−
3445. (c) Satoh, K.; Ozawa, S.; Mizutani, M.; Nagai, K.; Kamigaito, M.
H
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Nat. Commun. 2010, 1, 1−6. (d) Satoh, K.; Matsuda, M.; Nagai, K.;
Kamigaito, M. J. Am. Chem. Soc. 2010, 132, 10003−10005. (e) Berthet,
M. A.; Zarafshani, Z.; Pfeifer, S.; Lutz, J. F. Macromolecules 2010, 43,
44−50. (f) Natalello, A.; Hall, J. N.; Eccles, E. A. L.; Kimani, S. M.;
Hutchings, L. R. Macromol. Rapid Commun. 2011, 32, 233−237.
(g) Srichan, S.; Oswald, L.; Zamfir, M.; Lutz, J. F. Chem. Commun.
2012, 48, 1517−1519. (h) Srichan, S.; Chan-Seng, D.; Lutz, J. F. ACS
Macro Lett. 2012, 1, 589−592.
(6) (a) Kleiner, R. E.; Brudno, Y.; Birnbaum, M. E.; Liu, D. R. J. Am.
Chem. Soc. 2008, 130, 4646−4659. (b) van Hest, J. C. M. Nat. Chem.
Biol. 2008, 4, 272−273. (c) Ida, S.; Terashima, T.; Ouchi, M.;
Sawamoto, M. J. Am. Chem. Soc. 2009, 131, 10808−10809.
(7) Baughman, T. W.; Wagener, K. B. Adv. Polym. Sci. 2005, 176, 1−
42.
(24) Bielawski, C. W.; Hillmyer, M. A. In Handbook of Metathesis;
Grubbs, R. H., Ed.; Wiley-VCH: Weinham, Germany, 2003; Vol. 3, pp
255−282.
(25) Hahn, S. F. J. Polym. Sci., Part A: Polym. Chem. 1992, 30, 397−
408.
(26) (a) Burgess, K.; Jennings, L. D. J. Am. Chem. Soc. 1991, 113,
6129−6139. (b) Kawasaki, M.; Kakuda, H.; Goto, M.; Kawabata, S.;
Kometani, T. Tetrahedron: Asym. 2003, 14, 1529−1534.
(27) (a) Gais, H. J.; Jagusch, T.; Spalthoff, N.; Gerhards, F.; Frank,
M.; Raabe, G. Chem.Eur. J. 2003, 9, 4202−4221. (b) Lussem, B. J.;
Gais, H. J. J. Am. Chem. Soc. 2003, 125, 6066−6067. (c) Lussem, B. J.;
Gais, H. J. J. Org. Chem. 2004, 69, 4041−4052.
(28) Kasai, M.; Ziffer, H. J. Org. Chem. 1993, 48, 712−715.
(29) Asami, M. Bull. Chem. Soc. Jpn. 1990, 63, 721−727.
(30) Performing the same kinetic resolution with racemic 3AcCOE
as the substrate, which could save two steps, did not give satisfactory
result. After a week of reaction, (R)-3AcCOE was obtained in only
61% ee and 49% yield.
(8) (a) Zambelli, A.; Ammendola, P. Prog. Polym. Sci. 1991, 16, 203−
218. (b) Hatada, K. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 245−
260. (c) Coates, G. W. Chem. Rev. 2000, 100, 1223−1252.
(d) Kamigaito, M.; Satoh, K. Macromolecules 2008, 41, 269−276.
(e) Satoh, K.; Kamigaito, M. Chem. Rev. 2009, 109, 5120−5156.
(f) Schrock, R. R. Dalton Trans. 2011, 40, 7484−7495. (g) Dijkstra, P.
J.; Du, H. Z.; Feijen, J. Polym. Chem. 2011, 2, 520−527.
(9) (a) Sita, L. R. Macromolecules 1995, 28, 656−657. (b) Scherman,
O. A.; Kim, H. M.; Grubbs, R. H. Macromolecules 2002, 35, 5366−
5371. (c) Scherman, O. A.; Walker, R.; Grubbs, R. H. Macromolecules
2005, 38, 9009−9014.
(31) Odian, G. Principles of Polymerization, 4th ed.; Wiley-
Interscience: New York, 2004.
(32) Product information of Dupont Elvax EVA resins.
(33) (a) Baughman, T. W.; Chan, C. D.; Winey, K. I.; Wagener, K. B.
Macromolecules 2007, 40, 6564−6571. (b) Opper, K. L.; Markova, D.;
Klapper, M.; Mullen, K.; Wagener, K. B. Macromolecules 2010, 43,
3690−3698. (3) Leonard, J. K.; Wei, Y. Y.; Wagener, K. B.
Macromolecules 2012, 45, 671−680.
(34) (a) Gospodinova, N.; Zlatkov, T.; Terlemezyan, L. Polymer
1998, 39, 2583−2588. (b) Rojas, G.; Wagener, K. B. Macromolecules
2009, 42, 1934−1947.
(35) Yokota, K.; Kougo, T.; Hirabayashi, T. Polym. J. 1983, 15, 349−
354.
(10) Bielawski, C. W.; Grubbs, R. H. Prog. Polym. Sci. 2007, 32, 1−29.
(11) Kobayashi, S.; Pitet, L. M.; Hillmyer, M. A. J. Am. Chem. Soc.
2011, 133, 5794−5797.
́
(12) Martinez, H.; Miro, P.; Charbonneau, P.; Hillmyer, M. A.;
Cramer, C. J. ACS Catal. 2012, 2, 2547−2556.
(36) (a) Wagener, K. B.; Valenti, D.; Hahn, S. F. Macromolecules
1997, 30, 6688−6690. (b) Baughman, T. W.; Sworen, J. C.; Wagener,
K. B. Macromolecules 2006, 39, 5028−5036.
(13) Zhang, J.; Matta, M. E.; Hillmyer, M. A. ACS Macro Lett. 2012,
1, 1383−1387.
(14) We recently reported using this acetoxy system in the
preparation of block copolymers. See: Pitet, L. M.; Zhang, J.;
Hillmyer, M. A. Dalton Trans. 2013, DOI: 10.1039/C2DT32695G.
(15) (a) Fink, J. K. Handbook of Engineering and Speciality
Thermoplastics: Polyolejns and Styrenics; Scrivener Publishing LLC:
Beverly, MA, 2010. (b) Rinno, H. Poly(Vinyl Esters). Ullmann’s
Encyclopedia of Industrial Chemistry: Wiley-VCH: Weinheim, Germany,
2000. (c) Cordeiro, C. F. and Petrocelli, F. P. Vinyl Acetate Polymers.
Encyclopedia Of Polymer Science and Technology: Wiley-VCH:
Weinheim, 2004.
(37) Peacock, A. Handbook of Polyethylene: Structures: Properties, and
Applications, 1st ed.; CRC Press: Boca Raton, FL, 2000.
(38) (a) Turner-Jones, A. J. Polym. Sci. 1962, 62, S53−S56. (b) Seto,
T.; Hara, T.; Tanaka, K. Jpn. J. Appl. Phys. 1968, 7, 31−42. (c) Russell,
K. E.; Hunter, B. K.; Heyding, R. D. Polymer 1997, 38, 1409−1414.
(d) Androsch, R.; Blackwell, J.; Chvalun, S. N.; Wunderlich, B.
Macromolecules 1999, 32, 3735−3740. (e) Lieser, G.; Wegner, G.;
Smith, J. A.; Wagener, K. B. Colloid Polym. Sci. 2004, 282, 773−781.
(f) Rojas, G.; Inci, B.; Wei, Y. Y.; Wagener, K. B. J. Am. Chem. Soc.
2009, 131, 17376−17386. (g) Buitrago, C. F.; Opper, K. L.; Wagener,
K. B.; Winey, K. I. ACS Macro Lett. 2012, 1, 71−74. (h) Inci, B.;
Lieberwirth, I.; Steffen, W.; Mezger, M.; Graf, R.; Landfester, K.;
Wagener, K. B. Macromolecules 2012, 45, 3367−3376.
(39) Kamath, P. M.; Wakefield, R. W. J. Appl. Polym. Sci. 1965, 9,
3153−3160.
(16) Ito, S.; Munakata, K.; Nakamura, A.; Nozaki, K. J. Am. Chem.
Soc. 2009, 131, 14606−14607.
(17) Mori, Y.; Sumi, H.; Hirabayashi, T.; Inai, Y.; Yokota, K.
Macromolecules 1994, 27, 1051−1056.
(18) Hillmyer, M. A.; Laredo, W. R.; Grubbs, R. H. Macromolecules
1995, 28, 6311−6316.
(40) (a) Peacock, A. J.; Mandelkern, L. J. Polym. Sci., Part B: Polym.
Phys. 1990, 28, 1917−1941. (b) Kennedy, M. A.; Peacock, A. J.;
Mandelkern, L. Macromolecules 1994, 27, 5297−5310. (c) Kennedy,
M. A.; Peacock, A. J.; Failla, M. D.; Lucas, J. C.; Mandelkern, L.
Macromolecules 1995, 28, 1407−1421.
(19) Lehman, S. E.; Wagener, K. B.; Baugh, L. S.; Rucker, S. P.;
Schulz, D. N.; Varma-Nair, M.; Berluche, E. Macromolecules 2007, 40,
2643−2656.
(20) Watson, M. D.; Wagener, K. B. Macromolecules 2000, 33, 5411−
5417.
(21) Hejl, A.; Scherman, O. A.; Grubbs, R. H. Macromolecules 2005,
38, 7214−7218.
(22) (a) Hansson, S.; Heumann, A.; Rein, T.; Akermark, B. J. Org.
Chem. 1990, 55, 975−984. (b) Heumann, A.; Akermark, B.; Hansson,
S.; Rein, T. Org. Synth. 1990, 68, 109−115. (c) Morimoto, T.;
Machida, T.; Hirano, M.; Zhung, X. M. J. Chem. Soc., Perkin Trans. 2
1988, 909−913. (d) Iwaoka, M.; Tomoda, S. J. Chem. Soc., Chem.
Commun. 1992, 17, 1165−1167. (e) Yokota, T.; Fujibayashi, S.;
Nishiyama, Y.; Sakaguchi, S.; Ishii, Y. J. Mol. Catal. A: Chem. 1996,
114, 113−122. (f) Macsari, I.; Szabo, K. J. Tetrahedron Lett. 1998, 39,
6345−6348.
(23) (a) Cope, A. C.; Kinter, M. R.; Keller, R. T. J. Am. Chem. Soc.
1954, 76, 2757−2760. (b) Cope, A. C.; Johnson, H. E. J. Am. Chem.
Soc. 1957, 79, 3889−3892.
I
dx.doi.org/10.1021/ma400092z | Macromolecules XXXX, XXX, XXX−XXX