Terminal functionalization of polypropylene by radical-mediated thiol-ene addition

Article abstract of DOI:10.1021/ma050283l

The radical-mediated addition of alkanethiols to the terminal unsaturation of polypropylene (PP) is used to prepare functional polyolefin derivatives without altering the molecular weight of the parent material. The sulfide product of thiol-ene addition, as well as the internal olefin resulting from a remarkable thiyl radical-mediated olefin migration, is characterized using a model hydrocarbon, 2,4-dimethylhept-1-ene. The yield of 3- mercaptopropyltrimethoxysilane (MPTMS) addition to atactic-PP is examined in detail, and the functionalization of high molecular weight isotactic-PP is used to produce moisture-curing resins that bind to siliceous fillers. The limited unsaturation content of PP, along with the high temperatures needed to modify the resin in its melt state, is shown to impact negatively on the reaction conversion due to the reversibility of thiyl radical attack on olefin.

Full text of DOI:10.1021/ma050283l

5538
Macromolecules 2005, 38, 5538-5544
Terminal Functionalization of Polypropylene by Radical-Mediated
Thiol-Ene Addition
J. Scott Parent* and Saurav S. Sengupta
Department of Chemical Engineering, Queen’s University, Kingston, Ontario, Canada K7L 3N6
Received February 8, 2005; Revised Manuscript Received April 25, 2005
ABSTRACT: The radical-mediated addition of alkanethiols to the terminal unsaturation of polypropylene
(
PP) is used to prepare functional polyolefin derivatives without altering the molecular weight of the
parent material. The sulfide product of thiol-ene addition, as well as the internal olefin resulting from
a remarkable thiyl radical-mediated olefin migration, is characterized using a model hydrocarbon, 2,4-
dimethylhept-1-ene. The yield of 3-mercaptopropyltrimethoxysilane (MPTMS) addition to atactic-PP is
examined in detail, and the functionalization of high molecular weight isotactic-PP is used to produce
moisture-curing resins that bind to siliceous fillers. The limited unsaturation content of PP, along with
the high temperatures needed to modify the resin in its melt state, is shown to impact negatively on the
reaction conversion due to the reversibility of thiyl radical attack on olefin.
Introduction
Scheme 1. Radical-Mediated Thiol-Ene Addition to
DMH
Radical-mediated polymer modifications are robust
and inexpensive methods for preparing functionalized
commodity materials of added value. A leading example
1
is the grafting of vinyltrialkoxysilanes to polyethylene
to create moisture-curing resins that adhere covalently
2
to siliceous fillers. Unfortunately, the selectivity of this
process is inadequate for polypropylene (PP) modifica-
tions. Because of the propensity of tertiary macroradi-
3
cals to undergo â-scission, a high degree of PP degra-
dation is typically incurred to attain a significant
4,5
amount of monomer incorporation. The products have
very low melt strength, and they cannot be moisture-
cured to the point of large-scale gel formation.6
We have explored a synthetic approach that has the
potential to be free of such limitations, which involves
the radical-mediated addition of functional thiols to the
terminal unsaturation of the polymer.7,8 This process
exploits the well-established principles of thiol addition
strated by silica-binding and moisture-curing experi-
ments.
Experimental Section
9
to olefins (Scheme 1), wherein rapid hydrogen atom
Materials. Dicumyl peroxide (DCP, 98%), 2,6-di-tert-butyl-
-methylphenol (BHT, 99%), 1-dodecanethiol (98%), (3-mer-
4
donation by RSH is exploited in combination with
•
captopropyl)trimethoxysilane (MPTMS, 95%), mercaptosuc-
cinic acid (MSA, 97%), 11-mercaptoundecanoic acid (MUA,
favorable thiyl radical (RS ) addition kinetics to generate
1
0,11
a closed reaction sequence.
Because the chain-
9
5%), and vinyltrimethoxysilane (VTMS, 98%) were used as
carrying species is a thiyl radical as opposed to the
polymeric alkyl radical species that support conven-
tional vinylsilane grafting processes, the functionaliza-
tion of PP by a thiol-ene process is not expected to
received from Sigma-Aldrich. Dibutyltin dilaurate (94%, Alfa
Aesar), 2,4-dimethylhept-1-ene (DMH, 99%, ChemSampCo),
2
2
,4-dimethylheptan-2-ol (5, 98%, ChemSampCo), and Hi-Sil
33 (PPG) were used without purification. 2,2′-Azobis(isobu-
1
2
suffer from polymer fragmentation.
tyronitrile) (AIBN), 2,5-dimethyl-2,5-di(tert-butylperoxyl)hex-
ane (L-101, Elf Atochem), and 3,3,5-trimethyl-1,1-bis(tert-
butylperoxy)cyclohexane (L-231, Elf Atochem) were stored
In this report we describe the addition of alkyl
mercaptans to polypropylene homopolymers and com-
pare this modification strategy to conventional monomer
grafting processes. Unambiguous characterization of
polymer reaction products is provided by analyses of the
products derived from 2,4-dimethyl-1-heptene (DMH),
which serves as a model for the terminal unsaturation
of PP. These structural investigations are extended to
studies of low molecular weight atactic-PP modifications
in which issues related to kinetic chain length and
polymer fragmentation are addressed. The functional-
ization of high molecular weight isotactic-PP is demon-
n
under refrigeration. Isotactic polypropylene (i-PP, M ) 50 000,
polydispersity ) 3.8, Sigma-Aldrich) was used as received for
thiol addition experiments but purified by dissolving in hot
xylenes and precipitating from solution by the addition of
acetone for conventional vinylsilane grafting work. Atactic
polypropylene (a-PP, M ) 3300, polydispersity ) 3.0, [R Cd
n
2
2
CH ] ) 0.29 mmol/g, Scientific Polymers Products) was puri-
fied by dissolution-precipitation (toluene-acetone) prior to
use.
[
3-(2,4-Dimethylheptylthio)propyl]trimethoxysilane
(
1a). DMH (1 g, 7.9 mmol), MPTMS (1.56 g, 7.9 mmol), and
L-231 (0.001 g, 3.3 µmol) were degassed by three freeze-
pump-thaw cycles and heated to 125 °C for 60 min under a
nitrogen atmosphere. Unconsumed reagents were removed by
Kugelrohr distillation (0.03 mmHg, 70 °C) to yield a pale
*
Author for correspondence: phone (613) 533-6266; fax (613)
5
33-6637; e-mail parent@chee.queensu.ca.
1
0.1021/ma050283l CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/28/2005

Macromolecules, Vol. 38, No. 13, 2005
Terminal Functionalization of Polypropylene 5539
GC analysis employed a Supelco SPB-1 microbore column,
with injector and detector temperatures of 225 and 300 °C,
respectively. The oven temperature profile involved 40 °C for
6
min, ramp to 150 °C at 1 °C/min, ramp to 280 °C at 12 °C/
min, and hold for 5 min. Helium carrier gas was used at 2
mL/min. Retention times for the diastereomers of a 2,4-
dimethyl-1-heptanol (4) standard were 22.492 and 23.003 min,
while those of a 2,4-dimethyl-3-heptanol (5) standard were
1
4.083 and 14.803 min. Retention times observed for the
hydrated mixture of DMH + 3 were 13.819, 14.730 as well as
2
2.443, 22.950 min.
Sulfide Stability. DMH-S-dodecane (1b) (1.61 g, 5 mmol)
was heated with dodecanethiol (1.01 g, 5 mmol) at 150 °C for
1
h with DCP (2.093 mg, 7.74 µmol or 6.27 mg, 0.002 mmol).
1
H NMR revealed the unreacted sulfide and dodecyl disulfide
as the principal reaction products.
a-PP-g-MPTMS. a-PP (1 g), MPTMS (0.094 g, 0.48 mmol),
and AIBN (0.004 g, 30.4 µmol) were heated under a nitrogen
atmosphere to 90 °C for 60 min. Residual thiol was removed
1
Figure 1. Downfield H NMR spectra (CDCl
3
): (A) 1a; (B)
a-PP-g-MPTMS (X ) MeOH).
by Kugelrohr distillation (0.03 mmHg, 100 °C) to yield a sticky,
1
pale yellow product. H NMR (CDCl
3
, Figure 1): δ 3.5 (s, 9H,
-S-), δ 2.50-2.38 (m, 1H
, -CH -S-), δ 1.7-1.5
), δ 1.5-0.5 (m, -CH-, -CH -, -CH ).
-
OCH
CH
3
), δ 2.55-2.45 (m, 2H, -CH
2
a
,
-
(
a
H
b
-S-), δ 2.35-2.20 (m, 1H
b
a b
H
m, 3H, -CH-, -CH
2
2
3
Studies of a-PP conversion to sulfide were conducted as
described above, using the reaction conditions listed in Table
1
1
and in Figure 3. Normalized integration of H NMR spectra
provided the relative concentration of residual olefin (δ 4.6-
4
4
.8, 2H, dCH
H, -CH -S-CH
Poisoning of Thiol-Ene Additions. The model compound
2
) and the sulfide analogous to 1a (δ 2.45-2.55,
2
2
-) to within (5%.
Figure 2. JMOD 13C NMR spectrum of 1a; assignments as
shown in Figure 1.
formulations listed in Table 2 were degassed by three freeze-
pump-thaw cycles and heated to 125 °C for 60 min under
nitrogen. Analogous polymer formulations were similarly
treated at 150 °C for 60 min.
i-PP-g-MPTMS. Ground i-PP (0.75 g) was tumble-mixed
with L-101 (0.001 g, 3.4 µmol) and the required amount of
MPTMS. Reactions were conducted in the cavity of an Atlas
Laboratory mixing molder at 180 °C for 15 min, yielding i-PP-
g-MPTMS. Radical-mediated i-PP degradations were con-
ducted by mixing i-PP (45 g) and the desired amount of DCP
at 180 °C within a Haake Polylab R600 internal batch mixer
for 15 min at 60 rpm.
H+
yellow liquid, 1a. MS analysis: required for C15
m/e 322.199; found m/e 322.192. H NMR (CDCl
H
3
34SSiO
3
1
, Figure 1):
δ 3.55 (s, 9H, -OCH
3
), δ 2.52-2.44 (t, 2 H, -CH
-S-), δ 2.35-2.22 (m, 1H
), δ 1.5-1.0 (m, 7H,
), δ 1.0-0.5 (m, 11H, -CH , -CH
-). C NMR (CDCl , Figure 2): δ 8.8 (CH ), δ
), δ 19.4 (CH ), δ 20.1 (CH ), δ 20.2 (CH ), δ 23.2
), δ 30.0 (CH), δ 31.0 (CH), δ 36.0 (CH ), δ 39.6 (CH ), δ
0.5 (CH ), δ 44.3 (CH ), δ 50.7 (OCH ).
2
-S-), δ 2.50-
2
.38 (m, 1H , -CH
a
a
H
b
b
, -CH
a b
H -
S-), δ 1.8-1.6 (m, 3H, -CH-, -CH
2
-
CH
CH
2
-, -CH
2
-, -CH
3
3
3
,
1
3
-
3
, -CH
2
3
2
1
4.4(CH
3
3
2
3
(
CH
2
2
2
4
2
2
3
Reaction yield studies were conducted as described above,
Moisture Curing of i-PP-g-MPTMS. Polymer (1.0 g) and
xylenes (20 mL) were heated to reflux prior to the addition
dibutyltin dilaurate (10 µL, 20.2 µmol) and water (0.5 mL).
The mixture was maintained at a reflux condition for 20 min,
after which the polymer was recovered from solution by
precipitation with acetone (150 mL) and dried in vacuo. Gel
content was determined by extracting cured products with
refluxing xylenes from 120 mesh sieve cloth. Extraction
solutions were stabilized with 100 ppm of 2,6-di-tert-butyl-4-
methylphenol (BHT), and the procedure was conducted for a
minimum of 2 h, with longer times having no effect on the
results. Unextracted material was dried under vacuum to
constant weight, and gel content was calculated as the weight
percent of insoluble polymer.
Silica Immobilization of i-PP-g-MPTMS. Polymer (1.0
g), xylenes (20 mL), and precipitated silica (0.4 g) were heated
to reflux for 30 min, after which the polymer and silica were
recovered from solution by precipitation with acetone (150 mL)
and dried in vacuo. Gel content was determined as described
above, with the data presented as weight percent of insoluble
polymer after correcting for the silica content of the sample.
using thiol and initiator concentrations listed in Table 1.
1
Normalized integration of H NMR spectra provided the
relative concentration of residual DMH (δ 4.55-4.75, 2H, d
CH
2
), the sulfide 1 (δ 2.5-2.2, 4H, -S-CH
2
-), and the
disulfide 2 (δ 2.6, 4H, -S
2
-CH -) to within (5%.
2
Disulfide (2a). MPTMS (1 g, 5.09 mmol) was heated with
AIBN (0.83 g, 5.09 mmol) to 90 °C for 60 min. Unconsumed
reagents were removed by Kugelrohr distillation (0.03 mmHg,
1
6
2
0
0 °C). H NMR (CDCl
.63 (t, 4H, -CH -S-), δ 1.64-1.69 (m, 4H, -CH
.66 (t, 4H, -CH -Si-).
,4-Dimethylhept-2-ene (3). A solution containing a 2:1
3
): δ 3.55 (s, 18H, -OCH
3
), δ 2.59-
2
2
-), δ 0.54-
2
2
ratio of dodecanethiol (0.64 g, 3 mmol) and 1,1-di(tert-butyl-
peroxy)-3,3,5-trimethylcyclohexane (L-231, 0.48 g, 1.5 mmol)
was added dropwise via a syringe pump to DMH (1 g, 7.9
mmol) at 140 °C over 8 h. An olefin mixture containing DMH
(
89%) and 2,4-dimethylhept-2-ene (11%) was recovered from
the distillate of a Kugelrohr distillation (0.03 mmHg, 40 °C).
1
3
Selective Olefin Hydration. The desired olefin (DMH,
or DMH + 3 mixtures, 0.075 g, 0.6 mmol) was mixed a 1 M
solution of borane in tetrahydrofuran (1 mL) under nitrogen
for 30 min at 25 °C before quenching with water (1 mL). The
resulting organoborane was oxidized by addition of 1 mL of 3
N sodium hydroxide and 1 mL of 30% aqueous hydrogen
peroxide. The solution was stirred for 30 min at 25 °C prior to
Analysis. NMR spectra were recorded with a Bruker AM-
1
13
400 spectrometer (400.13 MHz H, 100.62 MHz C) in CDCl
3
with chemical shifts referenced to tetramethylsilane. Mass
spectra were recorded with a Fisons VG Quattro triple-
4
quadrupole instrument with chemical ionization (i-C H10). Melt
the addition of CH
with saturated NaCl solution (3 × 20 mL), and the aqueous
phase was extracted with CH Cl (20 mL). The organic extracts
were combined and dried over MgSO , and the resulting
alcohols were isolated by rotary evaporation.
Cl
2 2
(20 mL). The organic layer was washed
flow rate (MFR) values are reported as grams of resin extruded
in 10 min, as determined using a Tinius Olsen apparatus at
230 °C with a 2.16 kg load, unless otherwise noted. The
relative viscosity of toluene solutions containing 5 g/100 mL
of polymer was determined at 25 °C using an Eubelohde
2
2
4

5540 Parent and Sengupta
Macromolecules, Vol. 38, No. 13, 2005
Scheme 2
Scheme 3. Thiyl Radical-Mediated Olefin Migration
Scheme 4
capillary viscometer. Since elution times exceeded 200 s for
all solutions, kinetic energy corrections were not required.
Results
Model Thiol-Ene Additions. The products result-
ing from thiol addition to PP have been characterized
by comparisons to a model compound, 2,4-dimethyl-1-
heptene (DMH), whose structure is consistent with the
terminal unsaturation of the polymer (Scheme 1). Two
principal reaction products were isolated from the
modification process: the desired sulfide (1) and a
disulfide resulting from thiol oxidation (2).
tion by alkyl thiyl radicals has been documented for
The sulfide generated by reaction of DMH and 3-mer-
captopropyltrimethoxysilane (MPTMS) is the expected
14,15
thermodynamically favorable benzylic
and allylic
16
1
systems as well as for other C-H activated com-
pounds.17 The present case involves a similarly favor-
able allylic hydrogen abstraction, which is followed by
trapping of the intermediate allyl radical by thiol to
generate the observed internal olefin.
anti-Markovnikov addition product (1a) whose H and
1
3
C NMR spectra are illustrated in Figures 1a and 2,
respectively. Of particular analytical value is the down-
field H NMR resonances derived from residual terminal
1
unsaturation and from methylene groups adjacent to
sulfur (-CH2-S-CH2-). However, the two asymmetric
centers within 1a complicate the multiplicity of the
In contrast to conventional grafting reactions that
accommodate relatively insoluble reagents such as
maleic anhydride, thiol additions are highly sensitive
1
sulfide H NMR resonances for Ha and Hb (Figure 1).
8
Given that the protons of this methylene group are
diastereotopic, they exhibit geminal coupling. Addition-
ally, the four stereoisomers of compound 1a constitute
two pairs of diastereomers, with each pair being rep-
resented in an NMR spectrum. As a result, two sets of
resonances for each methylene proton are observed by
to modifier solubility. Soluble thiols such as MPTMS
could be added to DMH with conversions exceeding 90%.
On the other hand, poorly soluble reagents such as
mercaptoundecanoic acid (MUA) and mercaptosuccinic
acid (MSA) were completely inactive in the absence of
a cosolvent. Only when DMH and MUA were rendered
miscible by dissolving both reagents in chlorobenzene
was a modest olefin conversion achieved. Therefore, the
scope of solvent-free additions of thiols to polyolefins
may be considerably less than anticipated, given the
apparent incompatibility of thiol-ene reactions involv-
ing nonpolar hydrocarbons and polar modifiers.
1
H NMR, and two sets of signals are recorded for each
1
3
carbon in the C NMR spectrum that is proximal to an
asymmetric center. Peak assignments were confirmed
by 2-dimensional COSY H NMR and HSQC H- C
NMR analysis.
1
1
13
In addition to the desired sulfide, we were surprised
to discover small quantities of 2,4-dimethyl-2-heptene
Stability of Sulfide Addition Products. The tem-
peratures required to chemically modify and process
isotactic polypropylene are well in excess of those
commonly used for thiol-ene additions, which may raise
concerns regarding sulfide stability. In fact, there are
many reports of sulfide decomposition, three of which
(3) in the products of model reactions that employed low
thiol concentrations. Much higher yields of this isomer-
ization product were achieved by introducing thiyl
radicals to DMH in the absence of significant amounts
of free thiol (Scheme 2). These conditions generated
dodecyl disulfide (2b) as the principal sulfur-containing
product and resulted in an 11% conversion of DMH to
18
are illustrated in Scheme 4. The studies of Ohno et al.
and Kampmeir et al.19 initiated fragmentation by
abstracting â-hydrogen from the sulfide, leading to a
rapid elimination of a thiyl radical. These decomposi-
tions possessed no chain character, presumably due to
the inability of the eliminated thiyl radical to abstract
hydrogen in the manner of tert-butoxyl and phenyl
radicals. However, Huyser and Kellogg20 succeeded in
producing a chain sulfide decomposition by exploiting
the C-H bond activation brought on by an R-hydroxyl
group. Quantitative decomposition of their hydroxy
thioether was achieved due to the tautomerization of
3
. Unambiguous characterization of 3 required a selec-
1
3
tive hydration of the olefin mixture, followed by gas
chromatography analysis of the resulting secondary and
primary alcohols.
While the extent olefin migration may not be sub-
stantial under standard thiol-ene addition conditions,
its occurrence is of considerable fundamental impor-
tance, since it involves hydrogen atom transfer between
thiyl and allylic species (Scheme 3). Although thiols are
widely recognized as efficient hydrogen donors, abstrac-

Macromolecules, Vol. 38, No. 13, 2005
Terminal Functionalization of Polypropylene 5541
Table 1. Silane Modification of a-PP
initiator loading
modifier loading
silane graft
content (mmol/g)
olefin
conversion (%)
relative
viscosity
a
b
modifier
(µmol/g)
(mmol/g)
MPTMS
MPTMS
MPTMS
MPTMS
MPTMS
MPTMS
VTMS
30.4
30.4
16.5
16.5
18.5
18.5
18.0
0.48
0.96
0.48
0.96
0.48
0.96
0.48
0.16
0.23
0.12
0.19
0.08
0.17
0.34
57
82
43
66
34
58
1.74
1.78
1.75
1.76
1.86
1.75
1.04
a
.48 and 0.96 mmol/g represent 2.0 and 4.0 thiol equivalents, respectively. b Toluene; 5 g of polymer/100 mL; 25 °C; unmodified a-PP
0
relative viscosity ) 1.75.
the enol product, which prevented the readdition of
eliminated thiyl radicals.
to produce the derivative. Furthermore, the silane-
modified polymer was generated without altering its
relative viscosity, meaning that no gross changes to
molecular weight were inflicted by the process. In
contrast, the conventional radical-mediated addition of
vinyltrimethoxysilane (VTMS) to a-PP resulted in se-
vere viscosity losses due to macroradical fragmentation
(Table 1).
The stability of the sulfides of interest under the
reaction conditions used in this work was confirmed by
exposing the compounds to a variety of treatments, none
of which led to the regeneration of DMH. As an
illustrative example, sulfide 1b was maintained at 125
°C for 8 h while a solution containing 0.25 equiv of
dodecanethiol and 0.12 equiv of L-231 was added with
a syringe pump at a uniform rate. This treatment
exposed 1b to a steady concentration of thiyl radicals
under thiol-starved reaction conditions, thereby favoring
sulfide decomposition. Nevertheless, dodecyl disulfide
was the only significant reaction product. The sulfide
Figure 3 illustrates the influence of thiol loading,
temperature, and initiator concentration on sulfide
yields. As expected, increased thiol concentrations im-
proved olefin conversions under all the conditions
investigated. However, this sensitivity to thiol concen-
tration precluded the use of poorly soluble mercaptans
such as MUA and MSA. Just as they were ineffective
in the model compound system, these modifiers were
completely inactive for a-PP functionalization under
solvent-free conditions.
The effect of reaction temperature is of particular
interest since isotactic polypropylene modifications must
operate above the 160 °C melting point of the material.
The data plotted in Figure 3 show a sharp decline in
olefin conversion on moving from 90 to 125 °C and
further to 150 °C. We note that the half-lives of AIBN,
L-231, and DCP are nearly equivalent at the tempera-
tures employed (t1/2 ) 14.6, 18.6, and 15.0 min, respec-
tively) and that initiation efficiencies should be excep-
tionally high given the hydrogen-donating properties of
thiols. Therefore, pseudo-steady-state radical concentra-
tions are not expected to vary considerably between the
three series of experiments. Since the decline in yield
with temperature cannot be attributed to radical con-
centration effects on kinetic chain length, this effect
likely stems from the reversibility of thiyl radical attack
on olefins, as argued in the Discussion section.
Although the initiation of thiol-ene addition reactions
is commonly written as shown in Scheme 1, knowledge
of the details of the initiation process is lacking. This
may be due, in part, to the particular sensitivity of thiol
additions to low-yielding radical generation reactions,
given their exceptional kinetic chain lengths. In fact,
we observed substantial MPTMS addition to a-PP in the
absence of initiator (Figure 3), which is not seen for
conventional vinylsilane and maleic anhydride additions
to polyethylene.21 Since the sulfide adduct is an anti-
Markovnikov addition product of a radical-mediated
reaction, this baseline radical activity could be at-
tributed to trace amounts of peroxide/hydroperoxide
within a-PP, as others have proposed for the addition
of thiols to unpurified olefins.22 It should be noted,
however, that spontaneous radical formation has been
reported for thiol-ene mixtures held in the absence of
daylight and oxygen,23 possibly due to a molecule-
assisted homolysis mechanism, as proposed by Pryor
and co-workers.24
1b was recovered in very high yield, and only trace
amounts of olefins that did not include DMH were
1
detected by H NMR analysis.
Further evidence of stability was provided by the
observed inactivity of 1 with respect to sulfide exchange.
Heating 1a with 1 equiv of dodecanethiol and L-231 to
1
25 °C had no effect. The only significant product of this
treatment was dodecyl disulfide, which was accompa-
nied by trace amounts of unisolable olefinic byproducts.
Similar behavior was observed at 150 °C using DCP as
initiator.
We attribute the high-temperature stability of com-
pound 1 its polymer analogues to the inefficiency of the
â-hydrogen abstraction that facilitates sulfide cleavage.
Given that this hydrogen transfer involves an alkyl
mercaptan (RS-H bond dissociation energy ≈86 kcal/
mol) and a tertiary alkyl hydrocarbon (R3C-H ≈91 kcal/
mol), the process energetically unfavorable. Indeed,
Walling and Rabinowitz have reported limited yields for
n-butyl thiyl radical abstraction from isooctane,14 and
one would, therefore, expect the decomposition of sul-
fides such as 1a to demonstrate a limited kinetic chain
length, as reported for similarly unactivated sys-
1
8,19
tems.
Atactic-PP Modifications. Toward the synthesis of
silane-modified, isotactic materials of industrial interest,
we extended our model compound investigations to an
atactic polypropylene material (a-PP) having a Mn of
approximately 3300 g/mol. This polymer was amor-
phous, and its substantial terminal olefin content (0.29
mmol/g) allowed the extent of olefin conversion to sulfide
1
to be monitored by solution H NMR. Moreover, any
polymer degradation that accompanied the modification
process could be detected by dilute solution viscosity.
1
H NMR analysis of a-PP-g-MPTMS (Figure 1b)
confirmed that the structure of its functional moiety is
consistent with that of the model sulfide (1a). The yields
of the desired product were acceptable given the low
olefin concentrations imposed by the polymeric sub-
strate (Table 1), and the concentration of olefin migra-
tion products was negligible under the conditions used

5542 Parent and Sengupta
Macromolecules, Vol. 38, No. 13, 2005
Figure 3. a-PP conversion of terminal vinylidene to sulfide as a function of initiator and thiol loadings.
Table 2. Thiol-Ene Poisoning Trials
Table 3. Silane Modification of i-PPa
thiol
[thiol]
[poison]
modifier
loading
product
cured
a
a
b
modifier
(equiv)
potential poison
(equiv)
conv (%)
MFR
gel content bound polymer
modifier (mmol/g) (g/10 min)
(wt %)
(wt %)
DMH; [L-231] ) 0.1 wt %; T ) 125 °C; 60 min
MPTMS
MPTMS
MPTMS
MPTMS
1.0
1.0
1.0
1.0
none
71
73
70
73
0.00
0.10
0.20
0.20
1.00
0.20
0.33
>700
38
0
0
0
5
8
14
35
38
45
sulfide 1a
sulfide 1a
disulfide 2a
0.19
0.46
0.15
MPTMS
MPTMS
MPTMS
MPTMS
VTMS
37
10
12
32
0
37
40
a-PP; [DCP] ) 0.24 wt %; T ) 150 °C; 60 min
>700
>700
n-C12H25SH
n-C12H25SH
n-C12H25SH
3.0
3.0
3.0
none
a-PP-g-C12H25SH
disulfide 2b
67
65
68
VTMS
0
0.44
0.42
a [L-101] ) 3.4 µmol/g; T ) 170 °C; 15 min. b MFR measured at
230 °C using a 2.16 kg load. MFR of unmodified i-PP ) 35 g/10
a
Molar equivalents relative to olefin.
min.
Further anomalies are observed at high initiator
loadings, as the incremental response of sulfide yield
to additional initiator decreased dramatically beyond a
relatively low threshold concentration. This observation
is consistent with data reported by Ciardelli et al. for
AIBN-initiated additions of functional thiols to diene-
based elastomers.12 On the basis of the simple mecha-
nism presented in Scheme 1, one would expect to
observe a half-order dependence of the sulfide formation
rate on the concentration of initiator. As a result, the
sulfide yield should continue to increase with increasing
initiator loading as long as both olefin and thiol are
available. In an effort to define this initiator issue in
greater detail, we have considered potential complica-
tions derived from the interaction of major reaction
products with the initiator or propagating radical
intermediates. Table 2 lists the olefin conversions
recorded for a series of model compound and a-PP
experiments wherein different amounts of sulfide or
disulfide were included in the reaction mixture. Neither
reaction product influenced sulfide yields, meaning that
further work is required to characterize the yield
plateau behavior illustrated in Figure 3.
Isotactic-PP Modifications. The principles dem-
onstrated with the model compound and a-PP can be
applied to commercial grades of high-molecular-weight
isotactic polypropylene (i-PP). For this substrate, the
extent of MPTMS functionalization was determined by
coupling the modified polymer to precipitated silica and
by moisture-curing the modified resin. Given that only
silane-functionalized chains participate in these reac-
tions, these assays fractionate the polymer into its
grafted and unmodified components and provide an
accurate and meaningful measure of the extent of PP
modification.
The data presented in Table 3 confirm that radical
chemistry can be used to functionalize i-PP while
retaining the melt viscosity of the parent resin. The
MPTMS addition products reacted extensively with
precipitated silica, providing bound polymer contents of
up to 35 wt %. However, the MPTMS loadings required
to produce this degree of polymer functionalization were
high when compared to a conventional VTMS grafting
process, which provided relatively high bound polymer
contents at low modifier loadings. The key difference
between the two modification approaches is revealed by
2
5

Macromolecules, Vol. 38, No. 13, 2005
Terminal Functionalization of Polypropylene 5543
Table 4. MPTMS Addition to Degraded i-PPa
ics of MPTMS additions to PP depends on the ratio kf/
kRSH[RSH], which reflects the rate of thiyl radical
[
DCP] for
degraded
sulfide
cured
bound
c
c
•
degradation i-PP MFR
MFR
gel content polymer
elimination from the carbon-centered radical adduct (A )
b
(
µmol/g)
(g/10 min) (g/10 min)
(wt %)
(wt %)
relative to the rate at which thiol traps this adduct
(
Scheme 1). The small amount of quantitative data on
d
9
57
170
420
11
59
170
430
32
53
72
86
35
48
75
83
0
1
3
.0
.8
.7
this subject has been discussed by Chatgilialoglu et al.
in the course of studying ambient temperature reactions
28
of monounsaturated fatty acid esters. A ratio of kf/kRSH
-
1
a
) 15.0 M was reported for thiyl radical elimination
[
MPTMS] ) 1.0 mmol/g; [L-101] ) 3.4 µmol/g; T ) 170 °C; 15
b
-1
min. Controlled i-PP degradation prior to MPTMS modifications;
to E-isomers of this system, and a value of 2.2 M was
c
T ) 180 °C; 15 min. MFR measured at 190 °C using 2.16 kg load.
assigned to fragmentations leading to the corresponding
d
No degradation.
Z-isomers. Walling and Helmreich have recorded values
-
1
of kf/kRSH ) 84.9 and 19.8 M
at 60 °C for the
melt flow rate (MFR) and gel content analyses. Samples
of i-PP-g-VTMS demonstrated grossly reduced melt
viscosities and an inability to moisture-cure to the gel
point, while the MPTMS-grafted products retained the
MFR of the parent resin and could be cured to a
significant extent.
•
elimination of CH3S to generate the E- and Z-isomers
29
of 2-butene, respectively.
Both studies confirm that the rate of adduct radical
fragmentation greatly exceeds that of hydrogen transfer
when low thiol concentrations are employed. However,
these values reflect the reactivity of adducts carrying
â-substituents, which is not the case in the present
system. McPhee et al. derived a very crude estimate of
It is obvious that the transformation of i-PP into a
moisture-curable polymer by thiol-ene addition is af-
fected by the vinylidene content of the resin. The
catalyst technology used to prepare the polymer dictates
the end-group functionality, thereby establishing an
upper limit on the sulfide concentration that one can
-
1
kf/kRSH ) 0.034 M for reactions of 1-octene and tert-
butyl thiyl radicals at 25 °C.30 This value suggests that
fragmentation may have a minor effect on the addition
of thiols to unsubstituted terminal olefins. However,
given the expected differences between the activation
energies for thiyl radical elimination vs that of hydrogen
transfer, we propose that fragmentation is much more
favorable at the high temperatures employed for PP
modifications.
7
achieve through a thiol-ene addition approach. In a
final series of experiments, the radical-mediated deg-
radation of i-PP prior to MPTMS addition was explored
as a means of increasing the olefin content of the resin26
and, by extension, the amount of thiol addition to the
polymer. Heating i-PP with various amounts of DCP in
the absence of thiol resulted in a substantial degree of
polymer fragmentation, as revealed by increases in MFR
If this premise is valid, then the addition and elimi-
nation of thiyl radicals should maintain a pseudo-
equilibrium condition throughout the grafting process.
The position of this equilibrium is dictated by K1 ) ka/
kf, whose dependence on temperature is defined by a
negative enthalpy of reaction (Scheme 1).30 Conse-
quently, K1 declines as the reaction temperature is
increased, and the equilibrium is displaced away from
the desired adduct toward thiyl radical + olefin. The
net result of this displacement is a decline in sulfide
production with temperature, as observed throughout
our studies of the a-PP-g-MPTMS system.
The reversibility of thiyl radical addition to the
unsaturation within PP also creates a pronounced
sensitivity to thiol concentration. Mercaptans of low
solubility in PP (such as MUA and MSA) have a limited
potential to trap carbon-centered adduct radicals. With-
out efficient hydrogen donation, the rate of sulfide
formation declines to such an extent that disulfide
becomes the predominant reaction product. Conven-
tional graft additions of maleic anhydride proceed quite
well despite limited modifier solubility because alkyl
radical attack on the monomer is irreversible, and the
hydrogen donor in these systems is not the poorly
soluble modifier, but the polymeric substrate.
It is interesting to contemplate a further potential
consequence of the reversibility of thiyl radical attack
on carbon-carbon double bonds. We note that the
overall thiol-ene addition process could be reversible
if the hydrogen atom transfer component of the propa-
gation cycle shown in Scheme 1 is also reversible. This
possibility has been alluded to by Walling, who depicted
thiol addition in this manner without exploring the
potential consequences of reversible hydrogen transfer
reactions on the overall dynamics of the process.31 Since
thiol-ene additions to PP are high-temperature, exo-
thermic processes conducted under dilute conditions,
(
Table 4). Despite reducing the molecular weight of the
resin, degradation increased the olefin content of the
polymer substantially, thereby leading to a higher
bound polymer contents in silica-coupling reactions and
to a more extensive siloxane network in a moisture-
curing process.
Discussion
The functionalization of PP with MPTMS would be a
straightforward application of the thiol-ene addition
reaction if not for the extraordinary conditions that are
imposed by the substrate. For example, a commercial
grade of i-PP with a Mn of 50 000 may contain a
terminal unsaturation content of 20 µmol/g, which is far
less than the concentrations employed for standard
sulfide preparations. The high melting temperature of
i-PP also imposes stress on the process since the
reaction must be carried out in the melt state. Aside
from unresolved issues regarding initiation, the mech-
anism illustrated in Scheme 1 adequately accounts for
the consequences of these imposed conditions on a
thiol-ene derivatization. Thermodynamic and kinetic
analyses must, however, be limited to qualitative argu-
ments, given a lack of thermochemical data and rate
constants for the compounds and temperatures that are
relevant to PP functionalization.
What distinguishes the thiol-ene approach from
conventional monomer grafting methods is the revers-
ibility of thiyl radical addition to olefin. This reversibil-
ity underlies well-established methods for the E,Z-
rearrangement of internal olefins, wherein catalytic
amounts of thiol are activated by standard radical
initiators to isomerize large quantities of olefin at the
expense of trace amounts of sulfide addition product.27
The extent to which reversibility influences the dynam-

5544 Parent and Sengupta
Macromolecules, Vol. 38, No. 13, 2005
these reactions are reasonable candidates for an equi-
librium limitation. However, our studies have shown the
sulfide addition product is stable with respect to radical-
mediated fragmentation. Since the kinetics of sulfide
cleavage are not competitive with those of sulfide
formation, a mechanism wherein sulfide formation and
fragmentation constitute a dynamic equilibrium is inap-
plicable to this process.
Preparative Organic Chemistry; MacMillan: New York, 1964;
pp 62-119.
(
(
11) Back, R.; Trick, G.; McDonald, C.; Sivertz, C. Can. J. Chem.
1954, 32, 1078-1091.
12) Ciardelli, F.; Aglietto, M.; Passaglia, E.; Picchioni, F. Polym.
Adv. Technol. 2000, 11, 371-376.
(13) Zweifel, G.; Brown, H. C. Org. React. 1963, 13, 33.
(
14) Walling, C.; Rabinowitz, R. J. Am. Chem. Soc. 1959, 81,
137-1143.
1
(
15) Bickel, A. F.; Kooijman, E. C. Nature (London) 1952, 170,
2
11-212. (b) Wang, C. H.; Cohen, S. G. J. Am. Chem. Soc.
Conclusions
1
957, 79, 1924-1929. (c) van Zwet, H.; Kooyman, E. C. Recl.
Trav. Chim. 1968, 87, 45-48. (d) Pryor, W. A.; Gojon, G.;
Stanley, J. P. J. Am. Chem. Soc. 1973, 95, 945-946. (e) Pryor,
W. A.; Gojon, G.; Church, D. F. J. Org. Chem. 1978, 43, 793-
800.
The addition of alkanethiols to the terminal unsat-
uration of polypropylene yields functional derivatives
without affecting molecular weight. However, the re-
versibility of thiyl radical attack on olefin results in an
inverse relationship between temperature and reaction
yield and restricts the scope of this chemistry to soluble
modifiers that trap intermediate adduct radicals ef-
ficiently. MPTMS is particularly well-suited to this
process and yields moisture-curing products that bind
to siliceous fillers.
(16) Lunazzi, L.; Placucci, G.; Grossi, L. J. Chem. Soc., Chem.
Commun. 1979, 12, 533-534. (b) Schoeneich, C.; Asmus, K.-
D.; Dillinger, U.; Bruchhausen, F. Biochem. Biophys. Res.
Commun. 1989, 161, 113-120.
(
17) Zhao, R.; Lind, J.; Merenyi, G.; Eriksen, T. E. J. Am. Chem.
Soc. 1994, 116, 12010-12015. (b) Robins, M. J.; Ewing, G. J.
J. Am. Chem. Soc. 1999, 121, 5823-5824. (c) Nauser, T.;
Schoneich, C. J. Am. Chem. Soc. 2003, 125, 2042-2043.
18) Ohno, A.; Kito, N.; Ohnishi, Y. Bull. Chem. Soc. Jpn. 1971,
(
(
44, 463-467.
Acknowledgment. Financial support from the Natu-
ral Sciences and Engineering Research Council is
greatly appreciated. The authors thank Professor Emeri-
tus Kenneth E. Russell for valuable discussions con-
cerning this work.
19) Kampmeier, J. A.; Geer, R. P.; Meskin, A. J.; D’Silva, R. M.
J. Am. Chem. Soc. 1966, 88, 1257-1265.
(20) Huyser, E. S.; Kellogg, R. M. J. Org. Chem. 1966, 31, 3366-
3
369.
21) Parent, J. S.; Tripp, M.; Dupont, J. Polym. Eng. Sci. 2003,
3, 234-242.
22) Jones, S. O.; Reid, E. E. J. Am. Chem. Soc. 1938, 60, 2452-
(
(
4
References and Notes
2
455. (b) Overberger, C. G.; Mulvaney, J. E.; Marshall, B. J.
Org. Chem. 1956, 21, 1311-1312.
(
1) Russell, K. E. Prog. Polym. Sci. 2002, 27, 1007-1038.
2) Munteanu, D. In Reactive Modifiers for Polymers; Al-Malaika,
S., Ed.; Blackie Academic and Professional: London, 1997;
pp 196-265.
(
(
23) Sensfuss, S.; Friedrich, M.; Klemm, E. Makromol. Chem.
(
1
991, 192, 2895-2900.
24) Pryor, W. A.; Coco, J. H.; Daly, W. H.; Houk, K. N. J. Am.
Chem. Soc. 1974, 96, 5591-5593. (b) Pryor, W. A.; Lasswell,
L. D. In Advances in Free-Radical Chemistry; Williams, G.
H., Ed.; Academic Press: New York, Vol. 5, pp 27-99.
25) This assumes a steady-state concentration of thiyl radicals
and an equilibrium distribution between thiyl- and carbon-
centered adduct radical species.
(
3) Moad, G. Prog. Polym. Sci. 1999, 24, 81-142.
4) Hu, G. H.; Flat, J. J.; Lambla, M. In Reactive Modifiers for
Polymers; Al-Malaika, S., Ed.; Blackie Academic and Profes-
sional: London, 1997; pp 196-265.
(
(
(
(
(
(
5) Nachtigall, S. M. B.; Stedile, F. C.; Felix, A. H. O.; Mauler,
R. S. J. Appl. Polym. Sci. 1999, 72, 1313-1319.
6) Huang, H.; Lu, H. H.; Liu, N. C. J. Appl. Polym. Sci. 2000,
(26) Thompson, M. R.; Tzoganakis, C.; Rempel, G. L. J. Polym.
78, 1233-1238.
Sci., Polym. Chem. 1997, 35, 3083-3086.
7) Mulhaupt, R.; Duschek, T.; Rieger, B. Makromol. Chem.,
Macromol. Symp. 1991, 48/49, 317-332.
8) Boileau, S.; Mazeaud-Henri, B.; Blackborow, R. Eur. Polym.
J. 2003, 39, 1395-1404. (b) Gorski, U.; Maenz, K.; Stader-
mann, D. Angew. Makromol. Chem. 1997, 253, 51-64.
9) Kharasch, M. S.; Read, A. T.; Mayo, F. R. Chem. Ind. 1938,
(27) Golub, M. A. J. Polym. Sci. 1957, 25, 373-377. (b) Huyser,
E. S.; Kellogg, R. M. J. Org. Chem. 1965, 30, 2867-2868.
(28) Chatgilialoglu, C.; Altieri, A.; Fischer, H. J. Am. Chem. Soc.
2002, 124, 12816-12823.
(29) Walling, C.; Helmreich, W. J. Am. Chem. Soc. 1959, 81,
1144-1148.
(
7
52. (b) Sivertz, C.; Andrews, W.; Elsdon, W.; Graham, K. J.
(30) McPhee, D. J.; Campredon, M.; Lesage, M.; Griller, D. J. Am.
Chem. Soc. 1989, 111, 7563-7567.
Polym. Sci. 1956, 19, 587-588.
(
10) Griesbaum, K. Angew. Chem., Int. Ed. 1970, 9, 273-287. (b)
(
31) Walling, C. In Fifty Years of Free Radicals/Cheves Walling;
Walling, C. In Free Radicals in Solution; Wiley: New York,
American Chemical Society: Washington, DC, 1995; p 57.
1
957. (c) Stacey, F. W.; Harris, J. F. Org. React. 1963, 13,
50-376. (d) Sosnovsky, G. In Free Radical Reactions in
1
MA050283L