Chemo- and Regioselective Functionalization of Isotactic Polypropylene: A Mechanistic and Structure-Property Study

Article abstract of DOI:10.1021/jacs.9b05799

Polyolefins represent a high-volume class of polymers prized for their attractive thermomechanical properties, but the lack of chemical functionality on polyolefins makes them inadequate for many high-performance engineering applications. We report a metal-free postpolymerization modification approach to impart functionality onto branched polyolefins without the deleterious chain-coupling or chain-scission side reactions inherent to previous methods. The identification of conditions for thermally initiated polyolefin C-H functionalization combined with the development of new reagents enabled the addition of xanthates, trithiocarbonates, and dithiocarbamates to a variety of commercially available branched polyolefins. Systematic experimental and kinetic studies led to a mechanistic hypothesis that facilitated the rational design of reagents and reaction conditions for the thermally initiated C-H xanthylation of isotactic polypropylene (iPP) within a twin-screw extruder. A structure-property study showed that the functionalized iPP adheres to polar surfaces twice as strongly as commercial iPP while demonstrating similar tensile properties. The fundamental understanding of the elementary steps in amidyl radical-mediated polyolefin functionalization provided herein reveals key structure-reactivity relationships for the design of improved reagents, while the demonstration of chemoselective and scalable iPP functionalization to realize a material with improved adhesion properties indicates the translational potential of this method.

Full text of DOI:10.1021/jacs.9b05799

Article
pubs.acs.org/JACS
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Chemo- and Regioselective Functionalization of Isotactic
Polypropylene: A Mechanistic and Structure−Property Study
Jill B. Williamson,† Christina G. Na,† Robert R. Johnson, III,† William F. M. Daniel,‡
Erik J. Alexanian,*,† and Frank A. Leibfarth*,†
†Department of Chemistry, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, United States
‡Department of Applied Physical Sciences, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599,
United States
S
* Supporting Information
ABSTRACT: Polyolefins represent a high-volume class of
polymers prized for their attractive thermomechanical proper-
ties, but the lack of chemical functionality on polyolefins makes
them inadequate for many high-performance engineering
applications. We report a metal-free postpolymerization
modification approach to impart functionality onto branched
polyolefins without the deleterious chain-coupling or chain-scission side reactions inherent to previous methods. The
identification of conditions for thermally initiated polyolefin C−H functionalization combined with the development of new
reagents enabled the addition of xanthates, trithiocarbonates, and dithiocarbamates to a variety of commercially available
branched polyolefins. Systematic experimental and kinetic studies led to a mechanistic hypothesis that facilitated the rational
design of reagents and reaction conditions for the thermally initiated C−H xanthylation of isotactic polypropylene (iPP) within
a twin-screw extruder. A structure−property study showed that the functionalized iPP adheres to polar surfaces twice as strongly
as commercial iPP while demonstrating similar tensile properties. The fundamental understanding of the elementary steps in
amidyl radical-mediated polyolefin functionalization provided herein reveals key structure−reactivity relationships for the design
of improved reagents, while the demonstration of chemoselective and scalable iPP functionalization to realize a material with
improved adhesion properties indicates the translational potential of this method.
used for coordination−insertion polymerization and hinder
their activity.6 In response, late transition metal catalysts that
are more tolerant to polar functionality have been
developed.5,13,14 Despite significant effort, state-of-the-art
methods exhibit insufficient catalyst activity or result in
inadequate material properties for translation.15−19
The PPM of polyolefins leverages established industrial
production capacity while installing a tunable degree of
functionality randomly along the polymer backbone. The
radical-mediated addition of maleic anhydride to polyethylene
(PE) and isotactic polypropylene (iPP) is practiced
commercially via reactive extrusion.20,21 The lack of chemo-
selectivity in this strategy results in deleterious side reactions
including β-scission and polymer chain coupling that adversely
affect the molar mass, molecular weight distribution (MWD),
and thermomechanical properties of the polymers.22−25
Consequently, maleic anhydride functionalization of iPP and
other branched polyolefins results in low molecular weight
species with minimal incorporation of polar functional-
ity.24,26−28
INTRODUCTION
■
The annual global market for polyolefins is over 150 million
metric tons and projected to grow at an annual rate of 6.7%
through 2025.1 The continued demand for these high-volume,
low-cost thermoplastics is attributed to their high tensile
strength, ductility, thermal stability, and chemical resistance.2
These excellent thermomechanical properties along with the
versatility of polyolefins make them useful in diverse industries,
ranging from commodity packaging or automotive applications
to the medical device market.3,4 Despite their omnipresence,
these hydrocarbons are hindered by their inability to interface
with polar additives, fillers, or other materials.5 This intrinsic
limitation renders polyolefins inadequate for use in many high-
performance engineering applications. The addition of polar
functionality to polyolefins has been a long-standing goal in
polymer chemistry due to the promise of materials that
combine the thermomechanical properties of polyolefins with
the desirable properties of high-value polymers used in
composites, adhesives, and blends.6,7
The two most common synthetic approaches in pursuit of
polar polyolefins include copolymerization and postpolymeri-
zation modification (PPM).7−10 Methods for the copolymer-
ization of α-olefins with polar monomers have been pursued
aggressively for decades.7,11,12 An intrinsic problem of
installing polarity in this fashion is that Lewis basic
comonomers bind strongly to early transition metal catalysts
A number of modern polyolefin PPM methods have been
developed. Metal-catalyzed approaches have been success-
ful,29−31 most notably the ruthenium-catalyzed hydroxylation
Received: May 30, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Amidyl radical-mediated polyolefin C−H functionalization via thermal initiation.
of branched polyolefins.32−34 Trace metal in the final material,
however, results in oxidative degradation of the polymer
sample over time.35 For this reason, as well as the cost and
accessibility challenges associated with precious metals, a
metal-free route would be preferred. Previously reported metal-
free methods for PPM of polyolefins rely on hydrogen atom
abstraction by electron-rich radicals with moderate bond
dissociation free energies (BDFEs), resulting in a decrease in
the molecular weight for branched polyolefins due to β-
scission.36−39 Furthermore, none of these modern approaches
have been optimized for and implemented in an extruder,
which is a key requirement for translation.
small-molecule surrogate to facilitate rapid and accurate
characterization. Heating cyclooctane to 80 °C in benzene
(0.15 M) with xanthylamide 1a and 10 mol % azobis-
(isobutyronitrile) (AIBN) successfully yielded xanthylated
cyclooctane in a 22% yield (Figure 2). These conditions
In an effort to develop a metal-free PPM reaction that
retains the beneficial thermomechanical properties of the
parent polymer, we recently identified a sterically encumbered
N-xanthylamide reagent that, upon photolysis, provided
regioselective xanthyl group transfer to branched polyolefins
without deleterious chain scission reactions. In contrast to
previously reported metal-free PPM approaches that rely on
radicals generated from the thermal decomposition of
peroxides to abstract hydrogen atoms, this reagent did not
require an exogenous initiator, which suggests that the
transformation proceeds through a fundamentally distinct
pathway. The development of a thermal PPM process
combined with a deeper understanding of the mechanism of
this unique reagent is required to design improved reagents
and identify reaction conditions that enable functionalization
via reactive extrusion.
Our current mechanistic hypothesis is that an amidyl radical
is responsible for hydrogen atom abstraction, followed by rapid
trapping of the polymer-centered radical by the thiocarbonyl
group of the reagent (Figure 1). This results in a set of
complex equilibria that we probe herein through systematic
reagent design, crossover experiments, and kinetic studies. The
culmination of the data presented provides the rational design
of reagents and reaction conditions for the thermally initiated
functionalization of branched polyolefins without coincident
chain scission. Key principles identified from systematic
experiments facilitated the C−H xanthylation of iPP within a
twin-screw extruder. The resulting functionalized iPP demon-
strated similar mechanical properties to the virgin polymer
while exhibiting twice the adhesion strength to polar
substrates. The fundamental understanding of the elementary
steps in amidyl radical-mediated polyolefin functionalization
provided in this report reveals key structure−reactivity
relationships and serves as a platform for the design of
improved methods and translational technologies for polymer
C−H functionalization.
Figure 2. Optimization of the thermally initiated xanthylation of
cyclooctane with 1b. (tBuO)2 = di-tert-butyl peroxide. N−S cleavage
and percent yield were determined by 1H NMR analysis using
hexamethyldisiloxane as an internal standard.
resulted in only 24% homolysis of the N−S bond in 1a, which
was monitored by conversion of 1a to its respective parent
amide by 1H NMR. Alternative radical initiators, such as
dilauroyl peroxide (DLP) and benzoyl peroxide (BPO),
yielded similar results at 80 °C. In the case of BPO, increasing
the reaction temperature resulted in an increase in yield.
Further optimization found that the reaction performed best
with dicumyl peroxide (DCP) at 130 °C. Upon product
isolation, only xanthylated cyclooctane, O,O-diethyl dithiobis-
(thioformate), and the parent amide were observed. This
chemoselective C−H functionalization is attractive for trans-
lation to polymer substrates, since multiple products attached
to the same polymer chain cannot be separated from one
another.11
The electronic properties of thiocarbonylthio functional
groups are known to play a large role in their reactivity. For
example, these groups demonstrate different rates of chain
transfer in reversible addition−fragmentation chain-transfer
(RAFT) radical polymerization.40−43 To probe the structure−
reactivity properties of amidyl reagents with different
thiocarbonylthio groups, we designed and synthesized 1a−c
(Figure 3).44−46 Similar to RAFT polymerization, thiocarbo-
nylthio groups can be conveniently described with respect to
their Z groups: −OEt for 1a, −SEt for 1b, and −NEt2 for 1c.
Despite differing only in the heteroatom of the Z group,
reagents 1b and 1c provided lower yields of functionalized
cyclooctane under analogous conditions to those of 1a.
Trithiocarbonylamide 1b was observed to undergo complete
RESULTS AND DISCUSSION
■
Structure−Reactivity Studies of Amidyl Reagents. To
develop structure−reactivity relationships for the thermally
initiated C−H functionalization, cyclooctane was chosen as a
B
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Polymer C−H functionalization was quantified by 1H NMR
integration of the methylene protons of the Z group (Figure
5A). Theoretical maximum functionalization represents the
Figure 3. Reagents 1a−c demonstrate initial structure−reactivity
relationships for cyclooctane C−H functionalization. N−S cleavage
and percent yield were determined by 1H NMR analysis using
hexamethyldisiloxane as an internal standard.
N−S cleavage of 1b to parent amide, but only a 47% yield of
trithiocarbonylated cyclooctane was obtained. Dithiocarbamy-
lamide 1c did not completely convert to parent amide, but
demonstrated a similar yield to that of reagent 1b.
Figure 5. (A) 1H NMR of 1 mol % functionalized HBPE was taken in
CDCl3 and used to determine percent functionalization. The methine
proton of dithiocarbamylated HBPE likely overlaps with the
methylene protons of the diethylamine Z group. (B) SEC traces
obtained by the refractive index detector of 1 mol % functionalized
HBPE; (C) TGA−MS of HBPE.
Functionalization of Polyolefins. To translate the results
on cyclooctane to polymeric substrates, hyperbranched
polyethylene (HBPE) was chosen as a model branched
polyolefin for reaction optimization. HBPE with 10 methyl
branches per 100 carbons was synthesized using a previously
reported Pd(II) α-diimine catalyst.47 Polymers with different
number-average molar mass (Mn) and narrow dispersities (Đ)
were synthesized in order to determine the influence of
reaction conditions on molecular weight and MWD. The use
of amorphous HBPE enabled facile characterization of a
branched polyolefin at room temperature. Under homoge-
neous conditions, HBPE was heated at 130 °C for 6−24 h in
the presence of 1a−c and 10 mol % DCP in chlorobenzene at a
concentration of 0.2 M with respect to reagent (Figure 4). The
stoichiometry of reagents 1a−c was varied relative to the
repeat unit, with functionalization reported as mol % compared
to the polymer repeat unit (Figure 4).
stoichiometry of reagent relative to repeat unit (i.e., a
theoretical maximum of 10 mol % functionalization is addition
of 1 equiv of reagent per 10 repeat units). Exposing HBPE
along with reagent 1a at a theoretical maximum functionaliza-
tion of 10 mol % resulted in 4 mol % polymer xanthylation
(entry 1, Figure 4). By increasing the mol % theoretical
maximum xanthylation, C−H functionalization of HBPE
proved tunable in a range of 1−7 mol %.
Despite the addition of more reagent, the ultimate mol %
functionalization plateaued around 8 mol % for reagent 1a.
Reagents 1b and 1c also successfully functionalized HBPE, but
the plateau of functionalization appeared at 3 and 1 mol %,
respectively. This plateau is in part due to differences in the
conversion of 1a−c to the parent amide, with lower conversion
being observed through the series 1a−c (Figure 4). We note
that even low degrees of functionalization can have a
significant influence on the structure and properties of a high
molecular weight polymer; for example, 1 mol % functionaliza-
tion installs an average of 9 thiocarbonylthio groups on a
polymer chain with an Mn of 25 kg/mol. In agreement with
previous work, regioselectivity of amidyl radical hydrogen atom
1
transfer (HAT) was determined by H NMR to be selective
toward secondary carbon sites on HBPE.48,49
Changes in the MWD were analyzed by size exclusion
chromatography (SEC). A sample of HBPE was functionalized
with 1 mol % of each of the thiocarbonylthio groups, and the
SEC trace of the polymers is shown in Figure 5B. Xanthylated
(from 1a) and trithiocarbonylated (from 1b) HBPE
demonstrate MWD similar to that of the parent polymer,
but dithiocarbamylated (from 1c) HBPE shows evidence of
polymer-chain coupling reactions, with a peak appearing at a
13.6 min retention time upon functionalization (Figure 5B).
To assess the influence of the Z group on polymer thermal
properties, 1 mol % functionalized HBPE of each Z group
variant was analyzed via thermal gravimetric analysis (TGA)
and differential scanning calorimetry (DSC). All the polymers
Figure 4. Reagents 1a−c were reacted with HBPE in the presence of
a
DCP. Determined by 1H NMR in CDCl3. bDetermined by SEC
against polystyrene standards in tetrahydrofuran.
C
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
demonstrated thermal stability up to 250 °C, at which point
they underwent a partial mass loss before reaching a plateau
and fully degrading at >400 °C (Figure 5C). We hypothesized
that this partial mass loss was the result of a Chugaev-type
elimination.50,51 To support this hypothesis, the evolution of
volatile compounds accompanying the partial mass loss was
analyzed by mass spectrometry (TGA/MS). A prominent peak
at a mass to charge (m/z) of 76.1 was evident in all of the
samples, which we hypothesize is due to expulsion of carbon
disulfide. On the basis of these data, we hypothesize the
thiocarbonylthio groups are undergoing a [1,5]-sigmatropic
Chugaev-like rearrangement to yield an olefin on the polymer
backbone (Figure S9).52 This hypothesis is also supported by
the magnitude of mass loss for each of the functional polymers
in TGA, which correlates to the loss of the entire
thiocarbonylthio functional group. The DSC spectra for
functional HBPE demonstrate that, with identical degrees of
functionalization, the Z groups did not have a significant
impact on the thermal properties. The parent polymer
underwent glass transition at −69 °C, while the HBPEs with
1 mol % functionalization had glass transition temperatures
(Tg) of −66 and −67 °C (Figure S5A). HBPE also
demonstrates a melting temperature at −42 °C, which does
not change considerably upon 1 mol % functionalization. This
melting exotherm, however, does disappear at higher mol %
functionalization of xanthate and trithiocarbonate (Figure
S5B).
Crossover and Kinetic Experiments. The structure−
reactivity experiments for polymer functionalization (Figure 4)
revealed key differences among the three reagents. One of the
most instructive was that each reagent reached a maximum mol
% functionalization despite further addition of reagent relative
to polymer repeat unit. For example, reagent 1a achieved 8 mol
% xanthylation of HBPE when adding 50 mol % reagent, but
doubling the amount of reagent did not further increase
functionalization. In the case of 1c, the maximum amount of
dithiocarbamylation plateaued at 1 mol % regardless of reagent
stoichiometry. We observed a similar trend in our previous
work using a photochemical initiation strategy.48 We
hypothesized that this phenomenon was the result of
degenerative chain transfer of the thiocarbonylthio groups
between polymer chains.53,54 The equilibria of this chain-
transfer process, therefore, would result in reversible
functionalization.
To test this hypothesis, we designed and conducted a
number of crossover experiments using polymer samples that
had previously been functionalized with thiocarbonylthio
groups. Initially, an HBPE sample with 5.4 mol % xanthylation
was heated in the presence of DCP as an initiator with no
additional reagent. A decrease in functionalization to 2.3 mol %
was observed (Figure 6A, entry 1). Subsequently, reagents
containing Z groups not found on the functionalized polymer
(1b and 1c, respectively) were reacted with xanthylated HBPE
(Figure 6A, entries 2, 3). In both cases, the mol %
functionalization of xanthate decreased and the other
thiocarbonylthio group was added to the polymer. The same
trends were observed if the HBPE was initially functionalized
with trithiocarbonate (entry 4) or dithiocarbamate (entry 5).
These experiments demonstrate that polymer functionalization
is reversible under the reaction conditions and the ultimate
mol % functionalization is affected by the rate of degenerative
chain transfer.
Figure 6. (A) Functionalized HBPE was resubjected to the reaction
conditions with an amide reagent, yielding polyolefins with two
different thiocarbonylthio moieties appended. aDetermined by 1H
b
c
NMR in CDCl3. Repeat unit. No functionalized amide reagent was
added to the reaction. (B) Kinetics of reagent conversion during
functionalization of HBPE at a theoretical maximum of 10 mol %
functionalization (10 mol % DCP, 130 °C, [0.2 M] in
1
chlorobenzene). Reaction progress was determined by H NMR.
The observation of reversible functionalization under the
reaction conditions led us to probe the kinetics of group
transfer for reagents 1a−c under the reaction conditions. A
solution of HBPE, reagent 1, and DCP in chlorobenzene was
separated into aliquots and reacted for different amounts of
time to monitor the conversion of 1a−c to the parent amide.
As shown in Figure 6B, reagent 1a demonstrated considerably
faster conversion than that of either 1b or 1c, resulting in a rate
constant of 9 times and 6 times larger, respectively. The trend
for rate of reagent consumption does follow the pattern of mol
% polymer functionalization, with reagent 1a demonstrating
both the fastest rate and highest mol % polymer functionaliza-
tion and 1c representing the slowest rate and lowest mol %
functionalization.
Mechanistic Hypothesis. To develop a more compre-
hensive understanding of this C−H functionalization method,
our experimental observations were used to formulate a
mechanistic hypothesis (Figure 7). Supported by previous
literature, we hypothesize that thermolysis of dicumyl peroxide
yields a methyl radical,55,56 which is quickly trapped by the
functionalized amide reagent to form the captodatively
stabilized radical 2. Fragmentation of intermediate 2 yields
amidyl radical 4.44 We hypothesize that 4 is responsible for
HAT from the polymer backbone. HAT of aliphatic C−H
bonds by amidyl radicals is exergonic due to the large
difference in BDFE of an amidyl radical (BDFE of 107−110
kcal/mol) relative to C−H bonds (BDFE of 96−100 kcal/
mol).57 Furthermore, polarity matching between the electro-
philic amidyl radical and electron-rich aliphatic C−H bonds is
proposed to lower the kinetic barrier toward HAT.58,59
D
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Previous work understanding the rate of chain transfer of
thiocarbonylthio groups in RAFT polymerization helps to
conceptualize the experimental observations.60,61 In RAFT
polymerization, the choice of Z groups had pronounced effects
on the rates of radical addition to thiocarbonylthio functional
groups and the rate of fragmentation of the captodatively
stabilized species.40 For the Z groups studied herein, the rate of
radical addition (k1, k−2, k4, k−5, and k6) decreases in the series
SEt > OEt > NEt2, whereas the rate of radical fragmentation
(k−1, k2, k−4, k5, and k−6) decreases in the series NEt2 > OEt >
SEt (Figure 8). For the dithiocarbamate functional group in 1c,
Figure 7. Proposed mechanism of amidyl radical-mediated polyolefin
C−H functionalization. Polymer side chains were omitted for clarity.
Figure 8. Relative rates of radical addition and fragmentation help
conceptualize experimental observations for polyolefin C−H
functionalization.
A hydrogen atom from polymer 5 undergoes HAT to furnish
the parent amide 6 concomitant with carbon-centered radical
7. The production of 6 was monitored (Figure 6B) to
determine the overall rate of conversion of 1 → 6 (Figure 6B).
We hypothesize that step 3 is an irreversible transformation
due to the thermodynamics of this step. In the productive
reaction pathway, carbon-centered radical 7 reacts with reagent
1 to form the captodative radical 8. Fragmentation of 8 is not
degenerative, and, based on the kinetic experiments, the rate of
fragmentation depends largely on the Z group. C−S homolysis
can revert compound 8 back to the carbon-centered radical 7
(k−4) or facilitate productive cleavage of intermediate 8 to
yield functionalized polymer (9) and amidyl radical 4 (k−5).
The crossover experiments detailed in Figure 6A demonstrate
that both trapping of the polymer-centered radical (7 → 8)
and fragmentation to yield functionalized polymer (8 → 9) are
reversible. Even after formation of desired product 9, the
functionalized polymer can react with amidyl radical 4 and
revert back to the carbon-centered radical 7 (k−5). It should be
noted that, despite the myriad of radical chain-transfer
processes occurring in the system, we did not observe cross-
linked material in any of the experiments with HBPE.
In a separate experiment, we isolated xanthylated HBPE and
added a radical initiator in the presence of a lower molecular
weight polyolefin (Figure S6). After the reaction, both
polymers exhibited UV−vis absorptions consistent with
xanthate functionalization, indicating xanthate group transfer.
From this experiment we confirmed that an “off-cycle”
degenerative radical chain-transfer process is likely occurring
that sequesters a portion of the polymer-centered radical into
intermediate 10. Furthermore, control experiments conducted
by irradiation of reagents 1a−c excluding the polymer
substrate demonstrated that there is most likely a self-
immolative radical chain-process that consumes a portion of
the reagent without participating in HAT from the polymer
(Figure S10). The culmination of these experiments describes
the complexity of this C−H functionalization reaction and the
many equilibria that must be considered when optimizing
reactivity and/or designing new reagents.
we hypothesize that the slower rate of radical addition leads to
a longer lifetime of the polymer-centered radical 7 and an
increased potential for radical−radical coupling. Significant
chain coupling observed in the SEC traces (Figure 5B)
supports this hypothesis. The trithiocarbonate group in 1b
presumably results in an increase in both k4 and k−5. The faster
rate of radical addition leads to better control over the MWD
of the functionalized polyolefin, which agrees with our
experimental results. The faster rate of addition, however,
results in a lower amount of overall polymer functionalization
than 1a even at high reagent loadings, presumably due to a
faster rate for k−5 even after product 9 is formed. We
hypothesize that the reagent 1a is successful because it
balances the rates of addition and fragmentation. In the
conditions described herein, this enables the addition of 8 mol
% xanthyl groups to HBPE.
Functionalization of Semicrystalline Branched Poly-
olefins. With a more complete understanding of reagent
design principles and reaction mechanism, the thermal
functionalization of commercially available semicrystalline
branched polyolefins was explored (Figure 9). Reactions
were conducted at 180 °C, a temperature typically used for
the commercial PPM of polyolefins via reactive extrusion.62−65
Xanthylation of Dow DNDA-1081 NT 7 linear low density
polyethylene resin (LLDPE, 19 branches per 100 carbons)
using reagent 1a resulted in 3 mol % functionalization upon
addition of 5 mol % of the reagent. At higher equivalents of
reagent 1a, mol % functionalization increased along with an
observed increase in MWD. On the basis of the mechanistic
understanding gained herein, we hypothesized that reagent 1b,
which has a faster rate of radical addition, would decrease the
amount of deleterious chain coupling and chain scission side
reactions. The experimental results confirmed our hypothesis;
trithiocarbonate reagent 1b led to moderate functionalization
of LLDPE without significantly altering Mnor the Đ.
Functionalization of Dow polyethylene 4012 low density
(LDPE, 49 branches per 100 carbons) is challenging due to its
undefined structure and broad MWD (Đ = 9.54). Regardless,
the trends observed in our mechanistic experiments held true,
E
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 9. Functionalization of commercial branched polyolefins.
b
1
c
aRepeat unit. Determined by H NMR at 110 °C. Determined by
d
SEC at 120 °C. Determined by DSC in the second heating cycle.
Figure 10. Reactive extrusion of iPP with reagent 1a yielded iPP with
1 mol % xanthate functionality. The polymer was characterized by (A)
1H NMR at 110 °C in C2D2Cl4 and (B) SEC at 120 °C in
trichlorobenzene. (C) Visualization of the large-scale functionalization
of iPP by reactive extrusion.
with reagent 1a providing more efficient functionalization. The
chemoselective functionalization provided by these reagents is
notable, as even small amounts of chain coupling side reactions
are known to cause gelation on substrates with high weight-
average molecular weights (Mw). No functionalization was
imparted using reagent 1c, consistent with results on the
HBPE model system.
Basell Profax 6301 polypropylene homopolymer (iPP)
represents the most challenging semicrystalline polyolefin
substrate for C−H functionalization, as it has a high melting
temperature, high degree of crystallinity, high branch content
(50 methyl branches per 100 carbons), and is prone to β-
scission reactions. For this challenging substrate, polymer
functionalization in solution was not observed in any of the
conditions we tested using reagents 1a−c. Using reagent 1a
under neat conditions (i.e., without solvent), however, resulted
in 1 mol % xanthylation at a theoretical maximum
functionalization of 5 mol %. High-temperature 1H NMR
spectroscopy indicated that functionalization was highly
regioselective for the primary C−H bonds on iPP. Most
importantly, no evidence of β-scission of iPP was observed and
the material maintained a high melting temperature.
The culmination of mechanistic experiments, structure−
reactivity studies, and functionalization experiments led us to
design conditions for the functionalization of iPP by reactive
extrusion. A powder formulation of iPP mixed with reagent 1a
and 10 mol % DCP at a stoichiometry corresponding to a
theoretical maximum functionalization of 5 mol % was injected
into a twin-screw extruder at 180 °C and allowed to circulate
for 30 min. The extrusion process resulted in the C−H
functionalization of iPP with 1 mol % xanthate groups and no
observed β-scission (Figure 10B). In fact, the shear mixing
provided by the twin-screw extruder led to an MWD that
better represented that of the parent iPP than analogous
reactions conducted using mechanical stirring alone (Figure
S16).21
remove small-molecule impurities. The crystalline nature of
xanthylated iPP was confirmed by DSC. The xanthylated iPP
had a high melting point (137 °C) and a percent crystallinity
of 28%. Both of these values are modest decreases from the
commercial Profax 6301 material tested under analogous
conditions (Tm = 155 °C, 41% crystalline), which is commonly
observed for semicrystalline polymers upon functionaliza-
tion.30,33
The large thermal processing window (above the Tm and
below the Td) of the xanthylated iPP enabled us to melt-press
films of the material to probe the impact of functionalization
on mechanical and adhesive properties. Tensile testing of dog-
bone-shaped samples cut from melt-pressed films of the parent
Basell Profax 6301 iPP and 1 mol % xanthylated iPP by
dynamic mechanical analysis in linear film tension mode
yielded stress−strain curves that showed plastic deformation
behavior consistent with semicrystalline thermoplastics (Figure
11A). All quantitative values reported are an average of three
or more samples. The yield strength (σy) of iPP (σy = 35
1
MPa) was slightly higher than that of xanthylated iPP (σy = 26
3 MPa). The elongation at break value of iPP (εB = 570
50%) was congruent with results with xanthylated iPP (εB =
430 120%). Strain stiffening was observed in films of iPP and
xanthylated iPP, generating a stress at break of σB = 31
MPa and σB = 24 6 MPa, respectively.
2
In addition to retaining the desirable thermomechanical
properties of commercial polyolefin materials, we expected
xanthylated iPP to exhibit significantly different adhesive
properties due to the intrinsic polarity of the xanthate group.
On this basis, we reasoned that xanthylated iPP should display
superior adhesion to polar surfaces (i.e., glass) relative to
polyolefin materials. To test this hypothesis, we prepared a
single-lap joint between two glass slides, using iPP or
xanthylated iPP, and subjected it to lap shear analysis (Figure
11B). Xanthylated iPP demonstrated more than twice the
Scaling up the functionalization in an extruder achieved
multigram quantities of xanthylated iPP and provided sufficient
material for further structure−property evaluation. Prior to
further analysis, the xanthylated iPP was precipitated to
F
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
AUTHOR INFORMATION
■
Corresponding Authors
*eja@email.unc.edu
*FrankL@email.unc.edu
ORCID
Erik J. Alexanian: 0000-0002-0256-2133
Frank A. Leibfarth: 0000-0001-7737-0331
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work is supported by the Air Force Office of Scientific
Research under the Young Investigator Program
(#17RT0487). J.B.W. thanks the Henry G. Luce Foundation
and UNC for a Clare Boothe Luce Graduate Student
Fellowship. C.G.N. thanks the NSF for a Graduate Research
Fellowship and the Graduate School for a Royster Fellowship.
A provisional patent application has been filed (U.S. Prov.62/
567,460). We acknowledge the Polymer and Molecular
Characterization Facility in the Center for the Science and
Technology of Advanced Materials and Interfaces (STAMI) at
the Georgia Institute of Technology (Atlanta, GA, USA) for
the high-temperature gel permeation chromatography. This
work made use of the Integrated Molecular Structure
Education and Research Center (IMSERC) at Northwestern
University with thermal analysis with tandem mass spectrom-
etry (TGA−MS) results. We thank Prof. Theo Dingemans and
Dr. Maruti Hedge for thoughtful discussions.
Figure 11. Structure−property studies of iPP and 1 mol %
xanthylated iPP (XiPP). (A) Representative stress−strain curves for
iPP and XiPP measured by tensile testing (0.1 mm/s, room
temperature, break point indicated by arrow). (B) Representative
relative shear stress−strain curves for iPP and XiPP single-lap joints
measured by tensile testing (5 mm/min, room temperature).
adhesion strength to glass than did the iPP material, with
apparent lap shear strengths of 120 30 and 48 4 MPa,
respectively. These results prove the significant impact of
adding even small amounts of polar functionality onto
polyolefins.
REFERENCES
■
(1) Grand View Research. Polyolefin Market Size, Share & Trends
Analysis By Product Type (PE, PP, Ethylene Vinyl Acetate,
Thermoplastic Olefin), By Application (Film & Sheet, Injection
Molding, Blow Molding, Profile Extrusion), and Segment Forecasts,
2018−2025; 2018.
(2) Galli, P.; Vecellio, G. Polyolefins: The most promising large-
volume materials for the 21st century. J. Polym. Sci., Part A: Polym.
Chem. 2004, 42 (3), 396−415.
(3) Seymour, R. B. History of Polyolefins: The World’s Most Widely
Used Polymers; D. Reidel: Dordrecht, 1986.
CONCLUSION
■
A thermally initiated, metal-free C−H functionalization of
branched polyolefins is demonstrated using thiocarbonylthio
amide reagents. Comprehensive experimental studies on both
small molecule and polyolefin substrates revealed key
structure−reactivity relationships. Crossover and kinetic
experiments provide information on the many competing
equilibria that combine to determine the efficiency and
magnitude of polymer functionalization. From these exper-
imental studies, we developed a mechanistic hypothesis that
informed the translation of these reagents for the successful
chemoselective functionalization of commercial samples of
semicrystalline polyolefins. Ultimately, the reactive extrusion of
iPP on a decagram scale was demonstrated and resulted in
functionalized iPP that adheres to polar substrates twice as
strong as commercial iPP. We envision the results of this
experimental and mechanistic study will serve as a
comprehensive launching point for the continued improve-
ment and implementation of polyolefin C−H functionalization
in practical settings.
(4) Tolinski, M. Additives for Polyolefins, 2nd ed.; WIlliam Andrew
Publishing, 2015.
(5) Domski, G. J.; Rose, J. M.; Coates, G. W.; Bolig, A. D.;
Brookhart, M. Living alkene polymerization: New methods for the
precision synthesis of polyolefins. Prog. Polym. Sci. 2007, 32 (1), 30−
92.
(6) Chung, T. C. Synthesis of functional polyolefin copolymers with
graft and block structures. Prog. Polym. Sci. 2002, 27 (1), 39−85.
(7) Boaen, N. K.; Hillmyer, M. A. Post-polymerization functionaliza-
tion of polyolefins. Chem. Soc. Rev. 2005, 34 (3), 267.
(8) Franssen, N. M. G.; Reek, J. N. H.; De Bruin, B. A different route
to functional polyolefins: Olefin-carbene copolymerisation. Dalton
Trans 2013, 42 (25), 9058−9068.
(9) Vasile, C. Handbook of Polyolefins, 2nd ed.; CRC Press: New
York, 2000.
(10) Aima’adeed, A.-A.; Krupa, I. Polyolefin Compounds and
Materials: Fundamentals and Industrial Applications; Springer Interna-
tional Publishing, 2016.
(11) Williamson, J. B.; Lewis, S. E.; Johnson III, R. R.; Manning, I.
M.; Leibfarth, F. C-H Functionalization of Commodity Polymers.
Angew. Chem., Int. Ed. 2019, 58, 8654−8668.
(12) Dong, J. Design and synthesis of structurally well-defined
functional polypropylenes via transition metal-mediated olefin
polymerization chemistry. Front. Chem. China 2006, 1 (2), 145−157.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.9b05799.
■
S
Experimental procedures and spectral data for all new
compounds (PDF)
G
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(13) Carrow, B. P.; Nozaki, K. Synthesis of Functional Polyolefins
Using Cationic Bisphosphine Monoxide−Palladium Complexes. J.
Am. Chem. Soc. 2012, 134 (21), 8802−8805.
Low-Density Polyethylene with Polar Groups. Angew. Chem., Int. Ed.
2005, 44 (39), 6410−6413.
(33) Bae, C.; Hartwig, J. F.; Boaen Harris, N. K.; Long, R. O.;
Anderson, K. S.; Hillmyer, M. A. Catalytic Hydroxylation of
Polypropylenes. J. Am. Chem. Soc. 2005, 127 (2), 767−776.
(34) Kondo, Y.; García-Cuadrado, D.; Hartwig, J. F.; Boaen, N. K.;
Wagner, N. L.; Hillmyer, M. A. Rhodium-Catalyzed, Regiospecific
Functionalization of Polyolefins in the Melt. J. Am. Chem. Soc. 2002,
124 (7), 1164−1165.
(14) Nakamura, A.; Ito, S.; Nozaki, K. Coordination-Insertion
Copolymerization of Fundamental Polar Monomers. Chem. Rev.
2009, 109, 5215−5244.
(15) Zhang, W.; Waddell, P. M.; Tiedemann, M. A.; Padilla, C. E.;
Mei, J.; Chen, L.; Carrow, B. P. Electron-Rich Metal Cations Enable
Synthesis of High Molecular Weight, Linear Functional Polyethylenes.
J. Am. Chem. Soc. 2018, 140 (28), 8841−8850.
(35) Foster, G. N.; Wasserman, S. H.; Yacka, D. J. Oxidation
behavior and stabilization of metallocene and other polyolefins.
Angew. Makromol. Chem. 1997, 252, 11−32.
(36) Liu, D.; Bielawski, C. W. Direct azidation of isotactic
polypropylene and synthesis of ‘grafted to’ derivatives thereof using
azide−alkyne cycloaddition chemistry. Polym. Int. 2017, 66 (1), 70−
76.
(37) Li, H.; Ma, P.; Chen, Y.; Zhou, H.; Huang, H.; Liu, L.; Li, L.;
Plummer, C. M. Regioselective post-functionalization of isotactic
polypropylene by amination in the presence of N-hydroxyphthali-
mide. Polym. Chem. 2019, 10 (5), 619−626.
(38) Zhou, H.; Wang, S.; Huang, H.; Li, Z.; Plummer, C. M.; Wang,
S.; Sun, W. H.; Chen, Y. Direct Amination of Polyethylene by Metal-
Free Reaction. Macromolecules 2017, 50 (9), 3510−3515.
(39) Coiai, S.; Augier, S.; Pinzino, C.; Passaglia, E. Control of
degradation of polypropylene during its radical functionalisation with
furan and thiophene derivatives. Polym. Degrad. Stab. 2010, 95, 298−
305.
(40) Moad, G.; Rizzardo, E.; Thang, S. H. Radical addition−
fragmentation chemistry in polymer synthesis. Polymer 2008, 49 (5),
1079−1131.
(41) Chiefari, J.; Mayadunne, R. T. A.; Moad, C. L.; Moad, G.;
Rizzardo, E.; Postma, A.; Skidmore, M. A.; Thang, S. H.
Thiocarbonylthio Compounds (SC(Z)S-R) in Free Radical Polymer-
ization with Reversible Addition-Fragmentation Chain Transfer
(RAFT Polymerization). Effect of the Activating Group Z. Macro-
molecules 2003, 36, 2273−2283.
(42) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Krstina, J.;
Moad, G.; Postma, A.; Thang, S. H. Living Polymers by the Use of
Trithiocarbonates as Reversible Addition-Fragmentation Chain
Transfer (RAFT) Agents: ABA Triblock Copolymers by Radical
Polymerization in Two Steps. Macromolecules 2000, 33 (2), 243−245.
(43) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Chong, Y. K.;
Moad, G.; Thang, S. H. Living Radical Polymerization with Reversible
Addition-Fragmentation Chain Transfer (RAFT Polymerization)
Using Dithiocarbamates as Chain Transfer Agents. Macromolecules
1999, 32 (21), 6977−6980.
(44) Czaplyski, W. L.; Na, C. G.; Alexanian, E. J. C−H Xanthylation:
A Synthetic Platform for Alkane Functionalization. J. Am. Chem. Soc.
2016, 138 (42), 13854−13857.
(45) Na, C. G.; Alexanian, E. J. A General Approach to Site-Specific,
Intramolecular C-H Functionalization Using Dithiocarbamates.
Angew. Chem., Int. Ed. 2018, 57 (40), 13106−13109.
(46) Smith, G. E. P.; Alliger, G.; Carr, E. L.; Young, K. C.
Thiocarbamylsulfenamides. J. Org. Chem. 1949, 14 (6), 935−945.
(47) Gottfried, A. C.; Brookhart, M. Living Polymerization of
Ethylene Using Pd(II) α-Diimine Catalysts. Macromolecules 2001, 34
(5), 1140−1142.
(48) Williamson, J. B.; Czaplyski, W. L.; Alexanian, E. J.; Leibfarth,
F. A. Regioselective C-H Xanthylation as a Platform for Polyolefin
Functionalization. Angew. Chem., Int. Ed. 2018, 57, 6261−6265.
(49) Plummer, C. M.; Zhou, H.; Zhu, W.; Huang, H.; Liu, L.; Chen,
Y. Mild halogenation of polyolefins using an N-haloamide reagent.
Polym. Chem. 2018, 9 (11), 1309−1317.
(50) Benkeser, R. A.; Hazdra, J. J. Factors Influencing the Direction
of Elimination in the Chugaev Reaction. J. Am. Chem. Soc. 1959, 81
(1), 228−231.
(51) Betou, M.; Male, L.; Steed, J. W.; Grainger, R. S. Carbamoyl
Radical-Mediated Synthesis and Semipinacol Rearrangement of r-
Lactam Diols. Chem. - Eur. J. 2014, 20, 6505−6517.
(16) Chen, Z.; Leatherman, M. D.; Daugulis, O.; Brookhart, M.
Nickel-Catalyzed Copolymerization of Ethylene and Vinyltrialkox-
ysilanes: Catalytic Production of Cross-Linkable Polyethylene and
Elucidation of the Chain-Growth Mechanism. J. Am. Chem. Soc. 2017,
139 (44), 16013−16022.
̈
(17) Leicht, H.; Gottker-Schnetmann, I.; Mecking, S. Synergetic
Effect of Monomer Functional Group Coordination in Catalytic
Insertion Polymerization. J. Am. Chem. Soc. 2017, 139 (20), 6823−
6826.
(18) Diaz-Requejo, M. M.; Wehrmann, P.; Leatherman, M. D.;
Trofimenko, S.; Mecking, S.; Brookhart, M.; Perez, P. J. Controlled,
Copper-Catalyzed Functionalization of Polyolefins. Macromolecules
2005, 38 (12), 4966−4969.
(19) Aglietto, M.; Alterio, R.; Bertani, R.; Galleschi, F.; Ruggeri, G.
Polyolefin functionalization by carbene insertion for polymer blends.
Polymer 1989, 30, 1133−1136.
(20) Shi, D.; Yang, J.; Yao, Z.; Wang, Y.; Huang, H.; Jing, W.; Yin, J.;
Costa, G. Functionalization of isotactic polypropylene with maleic
anhydride by reactive extrusion: mechanism of melt grafting. Polymer
2001, 42 (13), 5549−5557.
(21) Hopmann, C.; Adamy, M.; Cohnen, A. In Reactive Extrusion;
Beyer, G., Hopmann, C., Eds.; Wiley-VCH Verlag GmbH & Co.
KGaA: Weinheim, Germany, 2017; pp 1−10.
(22) Gloor, P. E.; Tang, Y.; Kostanska, A. E.; Hamielec, A. E.
Chemical modification of polyolefins by free radical mechanisms: a
modelling and experimental study of simultaneous random scission,
branching and crosslinking. Polymer 1994, 35 (5), 1012−1030.
(23) Lu, B.; Chung, T. C. Synthesis of maleic anhydride grafted
polyethylene and polypropylene, with controlled molecular structures.
J. Polym. Sci., Part A: Polym. Chem. 2000, 38 (8), 1337−1343.
(24) Zhang, M.; Colby, R. H.; Milner, S. T.; Huang, T.; deGroot,
W.; Chung, T. C. M. Synthesis and Characterization of Maleic
Anhydride Grafted Polypropylene with a Well-Defined Molecular
Structure. Macromolecules 2013, 46, 4313−4323.
(25) Hamielec, A. E.; Gloor, P. E.; Zhu, S. Kinetics of, free radical
modification of polyolefins in extruders − chain scission, crosslinking
and grafting. Can. J. Chem. Eng. 1991, 69 (3), 611−618.
(26) Bettini, S. H. P.; Agnelli, J. A. M. Grafting of maleic anhydride
onto polypropylene by reactive extrusion. J. Appl. Polym. Sci. 2002, 85
(13), 2706−2717.
(27) Porejko, S.; Gabara, W.; Kulesza, J. Grafting of maleic
anhydride on polyethylene. II. Mechanism of grafting in a
homogeneous medium in the presence of radical initiators. J. Polym.
Sci., Part A: Polym. Chem. 1967, 5 (7), 1563−1571.
(28) He, X.; Zheng, S.; Huang, G.; Rong, Y. Solution Grafting of
Maleic Anhydride on Low-Density Polyethylene: Effect on Crystal-
lization Behavior. J. Macromol. Sci., Part B: Phys. 2013, 52 (9), 1265−
1282.
(29) Ray, A.; Zhu, K.; Kissin, Y. V.; Cherian, A. E.; Coates, G. W.;
Goldman, A. S. Dehydrogenation of aliphatic polyolefins catalyzed by
pincer-ligated iridium complexes. Chem. Commun. 2005, 3388−3390.
(30) Bunescu, A.; Lee, S.; Li, Q.; Hartwig, J. F. Catalytic
Hydroxylation of Polyethylenes. ACS Cent. Sci. 2017, 3 (8), 895−903.
(31) Boaen, N. K.; Hillmyer, M. A. Selective and Mild Oxy-
functionalization of Model Polyolefins. Macromolecules 2003, 36 (19),
7027−7034.
(32) Bae, C.; Hartwig, J. F.; Chung, H.; Harris, N. K.; Switek, K. A.;
Hillmyer, M. A. Regiospecific Side-Chain Functionalization of Linear
H
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(52) Chugaev, L. A new method for the preparation of unsaturated
hydrocarbons. Ber. Dtsch. Chem. Ges. 1899, 32 (3), 3332−3335.
(53) Braunecker, W. A.; Matyjaszewski, K. Controlled/living radical
polymerization: Features, developments, and perspectives. Prog.
Polym. Sci. 2007, 32 (1), 93−146.
(54) Zard, S. Z. In Handbook of RAFT Polymerization; Barner-
Kowollik, C., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA:
Weinheim, Germany, 2008.
(55) Inoue, S.; Kumagai, T.; Tamezawa, H.; Aota, H.; Matsumoto,
A.; Yokoyama, K.; Matoba, Y.; Shibano, M. Predominant methyl
radical initiation preceded by β-scission of alkoxy radicals in allyl
polymerization with organic peroxide initiators at elevated temper-
atures. Polym. J. 2010, 42 (9), 716−721.
(56) Norrish, R. G. W.; Searby, M. H. The photochemical
decomposition of dicumyl peroxide and cumene hydroperoxide in
solution. Proc. R. Soc. London A Math. Phys. Sci. 1956, 237 (1211),
464−475.
(57) Blanksby, S. J.; Ellison, G. B. Bond Dissociation Energies of
Organic Molecules. Acc. Chem. Res. 2003, 36 (4), 255−263.
(58) Roberts, B. Polarity-reversal catalysis of hydrogen-atom
abstraction reactions: concepts and applications in organic chemistry.
Chem. Soc. Rev. 1999, 28 (1), 25−35.
(59) Hioe, J.; Zipse, H. Radical stability and its role in synthesis and
catalysis. Org. Biomol. Chem. 2010, 8 (16), 3609.
(60) Jeffery, J.; Ercole, F.; Mayadunne, R. T. A.; Moad, G.; Rizzardo,
E.; Le, T. P. T.; Chiefari, J.; Krstina, J.; Chong, Y. K.; Moad, C. L.;
Thang, S. H.; Meijs, G. F. Living Free-Radical Polymerization by
Reversible Addition-Fragmentation Chain Transfer: The RAFT
Process. Macromolecules 2002, 31 (16), 5559−5562.
(61) Zard, S. Z. Some intriguing mechanistic aspects of the radical
chemistry of xanthates. J. Phys. Org. Chem. 2012, 25 (11), 953−964.
(62) Manning, S. C.; Moore, R. B. Carboxylation of Polypropylene
by Reactive Extrusion With Functionalized Peroxides for Use as a
Compatibilizer in Polypropylene/Polyamide-6,6 Blends. J. Vinyl Addit.
Technol. 1997, 3 (2), 184−189.
(63) Berzin, F.; Flat, J. J.; Vergnes, B. Grafting of maleic anhydride
on polypropyleneby reactive extrusion: effect of maleic anhydride and
peroxide concentrations on reaction yield and products character-
istics. J. Polym. Eng. 2013, 33 (8), 673−682.
(64) Cicogna, F.; Coiai, S.; Passaglia, E.; Tucci, I.; Ricci, L.;
Ciardelli, F.; Batistini, A. Grafting of functional nitroxyl free radicals to
polyolefins as a tool to postreactor modification of polyethylene-based
materials with control of macromolecular architecture. J. Polym. Sci.,
Part A: Polym. Chem. 2011, 49 (1), 781−796.
(65) Assoun, L.; Manning, S. C.; Moore, R. B. Carboxylation of
polypropylene by reactive extrusion with functionalised peroxides.
Polymer 1998, 39 (12), 2571−2577.
I
DOI: 10.1021/jacs.9b05799
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX