Selective and mild oxyfunctionalization of model polyolefins

Article abstract of DOI:10.1021/ma0347621

The direct oxyfunctionalization of a model polyolefin, polyethylene-alt-propylene (PEP), was achieved utilizing a straightforward, mild process. In the presence of a manganese complex, manganese meso-tetra-2,6-dichlorophenylporphyrin acetate (Mn(TDCPP)OAc), imidazole, a phase transfer agent, and potassium peroxymonosulfate (Oxone), PEP was functionalized under ambient conditions without chain degradation. The primary oxidation products contained tertiary alcohols and ketones based on IR, ¹H NMR, ¹³C NMR, and DEPT ¹³C NMR spectroscopy of the oxyfunctionalized products. The oxyfunctionalization of squalane, a small molecule, structural analogue of PEP, was also demonstrated. Spectroscopic analyses of the main products from the squalane oxidation were nearly identical with the functional groups identified in the PEP oxidation products. The degree of functionalization was modulated by changing the relative concentration of the oxidant, Oxone. The degree of functionalization and the thermal properties are reported for these new polymeric materials.

Full text of DOI:10.1021/ma0347621

Macromolecules 2003, 36, 7027-7034
7027
Selective and Mild Oxyfunctionalization of Model Polyolefins
Nicole K. Boa en a n d Ma r c A. Hillm yer *
Department of Chemistry, University of Minnesota, 207 Pleasant St. SE,
Minneapolis, Minnesota 55455
Received J une 5, 2003; Revised Manuscript Received J uly 22, 2003
ABSTRACT: The direct oxyfunctionalization of a model polyolefin, polyethylene-alt-propylene (PEP), was
achieved utilizing a straightforward, mild process. In the presence of a manganese complex, manganese
meso-tetra-2,6-dichlorophenylporphyrin acetate (Mn(TDCPP)OAc), imidazole, a phase transfer agent, and
potassium peroxymonosulfate (Oxone), PEP was functionalized under ambient conditions without chain
1
degradation. The primary oxidation products contained tertiary alcohols and ketones based on IR, H
1
3
13
NMR, C NMR, and DEPT C NMR spectroscopy of the oxyfunctionalized products. The oxyfunction-
alization of squalane, a small molecule, structural analogue of PEP, was also demonstrated. Spectroscopic
analyses of the main products from the squalane oxidation were nearly identical with the functional
groups identified in the PEP oxidation products. The degree of functionalization was modulated by
changing the relative concentration of the oxidant, Oxone. The degree of functionalization and the thermal
properties are reported for these new polymeric materials.
In tr od u ction
The chemical modification of polymers is a valuable
method for the preparation of new materials with
tailored properties and useful functionality. A wide-
spread example is the postpolymerization modification
of polydienes, utilizing reactions such as hydrogenation,
epoxidation, chlorosilation, hydroboration, and carbene
addition.1 These reactions apply established trans-
formations of carbon-carbon double bonds to add
functionality within and along a polymer backbone. In
the case of polyolefins (macromolecular alkanes), post-
polymerization modification is challenging due to the
absence of any suitably reactive functional groups. Still,
the search to develop routes to functionalized specialty
materials from inexpensive, commercial polyolefins is
unwavering due to their intrinsic value-added proper-
ties, including improved compatibility with pigments
and other macromolecules, in addition to potentially
-3
F igu r e 1. An effective alkane oxidation catalyst: manganese
meso-tetra-2,6-dichlorophenylporphyrin acetate (Mn(TDCPP)-
OAc).
as molecular oxygen, and a readily available catalyst
active under mild conditions. Previous work has dem-
onstrated that organometallic complexes can induce the
8
activation of alkane C-H bonds. However, to the best
4
,5
useful adhesion, thermal, and barrier properties. Even
the introduction of a small amount of polar functionality
into polyolefins would open new possibilities for macro-
of our knowledge, the only other system reported for the
selective and mild oxyfunctionalization of polyolefins
2
0
was only recently published.
6
,7
molecular grafting and reactive compatibilization.
One promising class of compounds for polyolefin
oxidations are those designed to mimic the activity of
In search of a convenient method for the functional-
ization of polyolefins, we considered previous work
aimed at the functionalization of alkanes.8-16 Histori-
cally, functionalization of alkanes has been achieved via
free radical reactions, where highly reactive species can
effectively functionalize these inert substrates. Unfor-
tunately, these free radical reactions also typically result
in side reactions, such as â-scission, chain transfer, and
coupling reactions, that reduce the efficiency of this
synthetic approach.17 These chain-coupling and chain-
breaking reactions can be especially detrimental to
polymer modification systems, since they disrupt the
molecular weight distribution of the polymer sample,
possibly compromising the desirable mechanical and
21-31
biological systems, such as methane monooxygenase
and cytochrome P-450,
8,32-34
which catalyze the hy-
droxylation of toxic, inert substances in the body to
facilitate their excretion. These model systems provided
novel synthetic approaches for adding polar functional-
33,35-38
ity to alkanes under ambient conditions.
Particu-
larly, manganese meso-tetra-2,6-dichlorophenylpor-
phyrin, Mn(TDCPP)X, where X is a counterion (e.g.,
acetate or chloride), has been recognized as an effective
catalyst for the hydroxylation of alkanes and the ep-
oxidation of alkenes in the presence of various oxygen
donors, including hypochlorite, hydrogen peroxide, per-
oxyacids, sodium periodate, and potassium hydrogen
persulfate from Oxone and a nitrogen base cocatalyst,
such as imidazole or substituted pyridines (Figure 1).
In the presence of one of these oxygen donors and a
1
8,19
processing properties of the starting material.
Al-
ternatively, an ideal functionalization system would
utilize an environmentally benign oxygen donor, such
3
9
nitrogen base, an active oxygenating species is formed
that can transfer an oxygen atom to alkanes or
*
To whom correspondence should be addressed: e-mail
4
0-43
hillmyer@chem.umn.edu.
olefins.
Furthermore, these synthetic halogenated
1
0.1021/ma0347621 CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/29/2003

7
028 Boaen and Hillmyer
Macromolecules, Vol. 36, No. 19, 2003
activity of the previously published Mn(TDCPP)/imid-
4
1
azole system. A typical two-phase reaction consisted
of Mn(TDCPP)OAc, adamantane, and imidazole in
methylene chloride and an aqueous solution of sodium
periodate (NaIO4) (Figure 2). A phase transfer catalyst,
benzyldimethyltetradecylammonium chloride (BDTAC),
was employed to facilitate the transfer of the oxidant
4
1
to the organic phase. The biphasic reaction was
vigorously stirred at room temperature for 24 h, after
which the organic phase was washed with distilled
water, concentrated, and analyzed by gas chromatog-
F igu r e 2. Oxyfunctionalization of alkanes.
1
raphy-mass spectrometry (GC-MS) and H NMR
metalloporphyrin models of cytochrome P-450 are robust
catalysts resistant to oxidative destruction.34,40
spectroscopy. The effects of the NaIO4, imidazole, and
BDTAC concentrations on the oxidation efficiency were
independently investigated to determine the optimal
reaction conditions. While we only examined adaman-
tane reactions using NaIO4, other oxidants including
magnesium monoperoxyphthalate hexahydrate, H2O2,
sodium hypochlorite (NaOCl), and potassium hydrogen
persulfate from Oxone (potassium peroxymonosulfate)
have previously been reported as active oxygen donors
in the oxyfunctionalization of adamantane with man-
Given the desirable features of this system, we
explored Mn(TDCPP)OAc as an oxygenation catalyst for
polyolefins. To effectively probe this reaction, the reac-
tivity of a Mn(TDCPP)OAc/imidazole system was ex-
amined for both a model polymer, poly(ethylene-alt-
propylene) (PEP), and squalane, a small molecule
structurally similar to PEP (Figure 2). The model
polymer was synthesized with controlled molecular
weight and a narrow molecular weight distribution
4
2,43,51,52
ganese porphyrin complexes.
optimal mole ratios for the oxidation reactions were
We found the
using anionic polymerization followed by hydrogena-
tion.4
4,45
Size exclusion chromatography (SEC) of reac-
[
[
NaIO4]0/[Mn(TDCPP)OAc]0 ) 400, [adamantane]0/
Mn(TDCPP)OAc]0 ) 82, [imidazole]0/[Mn(TDCPP)-
tion products from this model material allowed for the
ready identification of any active chain-coupling or
chain-cleavage side reactions. Utilizing this manganese
oxidation system, we were able to effectively incorporate
oxygen functionality into model polyolefins. In this
report, we show that spectral evidence, mass spectrom-
etry, and elemental analyses of oxyfunctionalized prod-
ucts corroborate the successful, direct, mild oxyfunc-
tionalization of model polyolefins under ambient con-
ditions.
OAc]0 ) 10, and [BDTAC]0/[Mn(TDCPP)OAc]0 ) 40,
similar to those mole ratios previously reported.41 These
conditions typically led to high conversion of adaman-
tane (84%).
The two major products were 1-adamantanol and
2
-adamantanone, typically present in a one-to-one mole
ratio. Previous work employing different oxidants and
shorter reactions times reported similar conversions for
the adamantane oxidation but reported 1- and 2-ada-
5
3
Resu lts a n d Discu ssion
mantanol as the principal products. Perhaps the
ketone product in our longer experiments could be
explained by the subsequent oxidation of 2-adamantanol
Syn th esis of Mn (TDCP P )OAc. We prepared the
manganese catalyst shown in Figure 1 using a modified
Lindsey procedure for the synthesis of the sterically
hindered porphyrin ligand, meso-tetra-2,6-dichloro-
5
4
to 2-adamantanone. GC-MS also revealed there were
no coupled products, suggesting the reaction does not
involve long-lived free radical intermediates.54 In a
separate control experiment, we found that no oxyfunc-
tionalized products were formed in the absence of
Mn(TDCPP)OAc. Having effectively generated an active
oxyfunctionalization system, we examined a model
substrate for the targeted polyolefin reaction.
phenylporphyrin (H2TDCPP), followed by metal inser-
tion.4
2,46-48
The yield for the large-scale synthesis of the
1
porphyrin ligand was typical (25%). The H NMR
spectrum, the Soret band at 418 nm in the UV/vis
spectra, and the low-resolution fast atom bombardment
mass spectrometric (FAB-MS) analysis were consistent
with that previously reported.4
2,46
The metalation re-
Oxyfu n ction a liza tion of Squ a la n e. Squalane is a
structurally similar, small molecule analogue of PEP;
thus, application of the manganese oxidation system to
squalane was the next step to explore the potential of
this system toward the oxyfunctionalization of poly-
olefins (Figure 2). The optimized conditions for the
adamantane reactions were applied to the squalane
oxyfunctionalization. Additional experiments using
NaIO4, NaOCl, or Oxone as the oxidant revealed that
similar products were produced for all of the reactions.
A typical oxyfunctionalization using Oxone was moni-
action with manganese(II) acetate to give Mn(III)-
(
(
TDCPP)OAc was monitored by UV/vis spectroscopy
shift in the Soret band to 478 nm). The 1H NMR
spectrum of Mn(TDCPP)OAc consisted of very broad
peaks with chemical shifts significantly outside of the
normal range for organic compounds, presumably due
to the paramagnetic manganese(III) center.4
2,46,49
The
oxidation of Mn(II) to Mn(III) typically occurs in this
type of system when oxygen is present.50 Mn(TDCPP)-
OAc can be purified by Soxhlet extraction with hexanes,
followed by extraction of the desired product with
methylene chloride and subsequent column chromatog-
raphy (54% yield). The crude product could also be
directly purified by column chromatography, followed
by recrystallization in a methylene chloride/hexane
mixture, with a substantial decrease in isolation yield.
High-resolution FAB-MS of the catalyst confirmed the
1
3
tored by C NMR spectroscopy for new chemical shifts
in the range for carbons in close proximity to oxygen-
containing functional groups. After a reaction time of
48 h, the intensity of the new signals appeared virtually
unchanged, and the reaction was stopped. The organic
layer was washed with distilled water, concentrated,
and analyzed by thin-layer chromatography (TLC).
Column chromatography was used to crudely fractionate
the complex mixture. The fractions collected were
divided into groups based on Rf values (Table 1). On
+
presence of the molecular ion [C44H20N4Cl8Mn] .
Oxyfu n ction a liza tion of Ad a m a n ta n e. Oxidation
reactions of adamantane were performed to verify the

Macromolecules, Vol. 36, No. 19, 2003
Oxyfunctionalization of Model Polyolefins 7029
Ta ble 1. Su m m a r y of F u n ction a lized Squ a la n e P r od u ct
group S2 indicates the concentration of any carbonyl-
containing functional group was small, consistent with
the IR analysis. Moreover, the absence of new chemical
shifts for methylene or methine protons R to a hydroxyl
group suggests that the alcohol products identified in
the IR analysis are tertiary alcohols.
a
Mixtu r es
groupb
elution solventc
Rfd
mass (g)
S1
S2
S3
S4
10:1
10:1
5:1
0.58-0.74
0.29-0.45
0.18
0.026
0.058
0.047
0.014
e
CH2Cl2
0.02
Although carbonyl-containing functional groups were
a
1
Isolated from a reaction of Mn(TDCPP)OAc (31.2 mg, 30.9
µmol), imidazole (22.1 mg, 0.320 mmol), benzyldimethyltetradecyl-
ammonium chloride (BDTAC) (0.5083 g, 1.26 mmol), squalane
identified in IR and H NMR spectroscopic analyses, no
13
characteristic resonances were observed in the C NMR
spectra (Figure 4), presumably due to the poor sensitiv-
ity of these quaternary carbons. The following analysis
(
(
0.5263 g, 1.24 mmol), and 5 mL of CH2Cl2. The oxidant, Oxone
1.59 g, 2.59 mmol), was added to the reaction as a solution in 50
1
3
of the C NMR spectra will serve to highlight the
mL of 0.25 M phosphate buffer, pH ) 7. The biphasic mixture was
b
stirred vigorously for 2 days. Groups identified in the Supporting
salient features, identifying the predominant functional
c
d
13
Information. Pentane:Et2O. TLC solvent development solution
groups of the oxidation products. Using a DEPT
C
e
5
:1 pentane/Et2O. Pure methylene chloride was used as the
NMR experiment, all resonances were assigned as
quaternary, methine, methylene, or methyl carbons. The
eluent.
1
3
C NMR spectrum of group S1 exhibited two new
chemical shifts at 54.0 and 46.4 ppm that were very low
in intensity, indicating low concentrations of these
carbons. These new signals were assigned to methylene
and methine carbons R to a carbonyl group, respectively,
based on group additivity calculations using squalane
as a base molecule. Thus, one type of ketone product is
proposed for group S1, having a carbonyl group adjacent
1
3
to a tertiary carbon. The C NMR spectrum of group
S2 revealed new chemical shifts at 73 and 42 ppm.
These two carbons were identified as a quaternary
carbon and a secondary carbon, respectively. Since there
are no quaternary carbons in squalane and the chemical
shift is in the characteristic region for carbons R to a
hydroxyl group, we propose that the products contain
tertiary alcohols. Using group additivity, the additional
new chemical shift at 42 ppm was assigned to methylene
carbon(s) â to the hydroxyl functional group, which is
expected as all tertiary alcohol products would be
adjacent to at least one methylene group in squalane.
F igu r e 3. IR spectra of squalane and functionalized deriva-
tives [thin film/NaCl plate]: (a) squalane, (b) group S1, (c)
group S2, (d) group S3, (e) group S4 from Table 1.
1
3
The C NMR spectrum of group S3 had carbons
resonating in the same range as those identified for
group S2. Similarly, the chemical shifts for a quaternary
carbon at 71 ppm and a methylene carbon at 44 ppm
were ascribed to a tertiary alcohol product. The slightly
shifted values for the signals can be attributed to
different isomers of the tertiary alcohol squalane prod-
uct. For example, the tertiary carbons could be hydroxy-
lated on the end of the chain or within the molecule.
The C NMR spectrum of group S4 contained reso-
nances present in both group S2 and group S3. Since
group S4 is more polar than groups S2 or S3 (see Rf
values in Table 1), it is most likely that group S4
contains both types of hydroxyl groups, as opposed to
being a mixture of products from groups S2 and S3.
the basis of the mass of recovered squalane, approxi-
mately 70% of the starting material was functionalized
(
see Supporting Information). Characterization of the
four isolated product mixtures was accomplished using
1
13
elemental analysis, H NMR, C NMR, distortionless
enhancement by polarization transfer (DEPT) 13C NMR,
and IR spectroscopy.
IR spectra of squalane and oxyfunctionalized products
1
3
(
groups S1-S4) are given in Figure 3. The absorbance
-
1
at 1716 cm in the spectrum of group S1 is character-
istic for a CdO stretch, indicating the presence of
carbonyl-containing functional groups. The absorbances
-
1
-1
-1
at 3376 cm , near 1100 cm , and at 1716 cm in the
spectra for groups S2-S4 were respectively assigned for
an O-H stretch and a C-O stretch in an alcohol and a
CdO stretch of a carbonyl-containing functional group.
Therefore, we conclude that groups S1-S4 are oxyfunc-
tionalized, and the more polar fractions contain both
hydroxyl and carbonyl functionalities.
Elemental analysis (C, H, O) of the four groups (Table
1) allowed the average number of oxygen atoms per
squalane molecule to be determined. Elemental analysis
of group S1 gave an average of 1.4 oxygen atoms per
squalane molecule (5.0% by mass), suggesting there is
a mixture of molecules containing one and two oxygen
atoms. Elemental analysis of groups S2, S3, and S4
revealed there were 1.2, 1.1, and 2.4 oxygen atoms per
squalane molecule, respectively. Importantly, the C, H,
O data did not always equal 100%, indicating there were
some non-C, H, O impurities present. These impurities
could be trace fragments of oxidized catalyst, as sug-
1
H NMR spectra of the oxyfunctionalized products
were largely unchanged from the spectrum of squalane.
However, new chemical shifts were observed near 2.6
ppm in the spectra of groups S1, S3, and S4, which are
consistent with resonances for methylene or methine
protons on a carbon R to a carbonyl group in a ketone.
Furthermore, no additional resonances in the normal
range for carboxylic acids or aldehydes were observed,
providing further support that the carbonyl functional
groups in the isolated products are ketones. The absence
of a resonance at 2.6 ppm in the proton spectrum of
5
5
gested by the colored products and by liquid chroma-
tography-mass spectrometric (LC-MS) analysis. High-
resolution FAB-MS analyses of two products obtained
from separate reactions confirmed the presence of the

7
030 Boaen and Hillmyer
Macromolecules, Vol. 36, No. 19, 2003
F igu r e 4. 13C NMR spectra of oxyfunctionalized products and squalane [ca. 14-50 mg/mL CDCl
; d
) 5 s, n ) 450]: (a) squalane,
3
1
t
(
b) group S1, (c) group S2, (d) group S3, (e) group S4 from Table 1. [*] denotes solvent peaks.
Ta ble 2. Su m m a r y of F u n ction a lized P EP P r od u ctsa
no. of
elution
no. of
OH/100
mass
(g)
groupb solventc
Rfd
OH/chaine carbons
e
P1
P2
P3
P4
P5
10:1
10:1
5:2
5:2, 2:1
2:1
0.71-0.84
0.42-0.52
0.09-0.15
0.03-0.05
0.02
0.7
1.7
1.5
3.1
2.5
0.3
0.6
0.5
1.1
0.9
0.397
0.133
0.108
0.073
0.089
F igu r e 5. Functionalized squalane candidates based on 13C
chemical shifts and IR absorption bands.
a
Isolated from reaction of Mn(TDCPP)OAc (43.6 mg, 43.5 µmol),
imidazole (29.4 mg, 0.435 mmol), benzyldimethyltetradecylammo-
nium chloride (BDTAC) (0.7072 g, 1.75 mmol), PEP (Mn ) 5 kg/
mol) (1.0815 g, 0.235 mmol, 15.45 mmol monomer), and 10 mL of
CH2Cl2. The oxidant, Oxone (1.79 g, 2.91 mmol), was added to the
reaction as a solution in 60 mL of 0.25 M phosphate buffer, pH )
7. The biphasic mixture was stirred vigorously for 5 days.
+
+
molecular ions, [C30H62O + Na] and [C30H62O2 + Na] ,
consistent with the mono- and dihydroxylated products,
respectively (see Supporting Information).
All of the characterization data substantiate the
successful incorporation of oxygen into squalane. Spec-
troscopic evidence suggests the oxygen-containing func-
tional groups are tertiary alcohols and ketones. Some
possible products are presented in Figure 5. Mass
spectrometry and elemental analysis of the products
quantified the degree of functionalization of the products
as containing exclusively one oxygen atom per squalane
up to an average close to two oxygen atoms per
molecule. This trend was consistent with decreasing Rf
values for more highly substituted products.
Oxyfu n ction alization of a Low-Molecu lar -Weigh t
Mod el P EP . Given the success with the oxidation of
squalane, we examined the oxidation of a model poly-
olefin. Specifically, the gram scale postpolymerization
oxyfunctionalization reactions of a low-molecular-weight
b
Corresponds to product identification in the Supporting Informa-
tion. c Pentane:Et2O. TLC solvent development solution 5:1 pen-
tane/Et2O. e Average value determined from 1H NMR analysis of
acetylated products; ca. 70 repeat units (280 backbone carbons)
per chain.58
d
spectra of PEP and two representative fractions are
presented in Figure 6. The absorbance near 3500 cm
-
1
,
characteristic of an O-H stretch in an alcohol, was
observed in the IR spectrum of group P1. The IR spectra
for groups P2-P5 were virtually identical, having a
-
1
broad absorbance near 3400 cm , which was assigned
for an O-H stretch, in addition to a peak near 1710
-
1
cm , which is typical for a CdO stretch.
1H NMR spectra of the oxyfunctionalized products
were similar. New chemical shifts were observed near
2.6 ppm, consistent with methylene or methine protons
of a carbon R to a carbonyl group of a ketone. These
(
5 kg/mol) model PEP was performed utilizing the
manganese system with Mn(TDCPP)OAc, imidazole,
5
6
BDTAC, and Oxone (Figure 2). After 5 days, the
reaction was stopped, washed with distilled water,
dried, and concentrated. The complex product mixture
was crudely purified using column chromatography, and
the fractions collected were combined into groups based
on Rf values (Table 2). The conversion of PEP to
functionalized derivatives was estimated to be 80%
based on recovered PEP. The isolated product mixtures
1
signals are similar to those observed in the H NMR
spectra of squalane oxidation products. No other reso-
nances were observed in the normal range for other
oxyfunctionalized derivatives. Furthermore, the absence
of methylene or methine protons R to a hydroxyl group
strongly supports the alcohol functionality identified by
IR spectroscopic analysis is a tertiary alcohol.
5
7
were characterized by IR and NMR spectroscopy. IR

Macromolecules, Vol. 36, No. 19, 2003
Oxyfunctionalization of Model Polyolefins 7031
functionalization.60 Assuming all of the hydroxyl groups
reacted, an average number of hydroxyl groups per
chain could be determined from integration values of
1
the H NMR spectra. Specifically, the methyl protons
of the acetyl group near 2.0 ppm were integrated
relative to the protons of the polymer chain to determine
the degree of hydroxylation of the functionalized prod-
1
uct. H NMR analysis of the acetylated products was
consistent with the functionalized products having one
to three hydroxyl functional groups per chain or 0.4-
1
.1 hydroxyl groups per 100 backbone carbons.58 From
IR analysis, we also know that carbonyl groups are
present in groups P2-P5 as well.
SEC analyses of the crude reaction and isolated
product mixtures indicated that the reaction conditions
did not significantly alter the number-average molecular
weight (Mn) or the molecular weight distribution (PDI)
of the PEP starting material (Figure 9). Only a very
small amount of coupled product was identified from
the SEC trace, but perhaps most importantly, virtually
no chain degradation was observed. Previous efforts in
this field of research have been characterized by severe
F igu r e 6. IR spectra of 5 kg/mol PEP starting material and
functionalized derivatives [thin film/NaCl plates]: (a) 5 kg/
mol PEP starting material, (b) group P1, (c) group P3 from
Table 2.
The 13C NMR spectra of groups P1-P5 all exhibited
the same resonances, suggesting all of the products
contain similar functional groups. A representative
spectrum is presented in Figure 7. Two new signals
were seen at 73 and 42 ppm for quaternary and
methylene carbons, respectively. All other chemical
shifts were consistent with the PEP starting material.
61,62
chain degradation and loss of mechanical properties.
We investigated the effects of increasing reaction
temperature (50 °C) and varying the oxidant concentra-
tion on the reaction. The effect of temperature was
investigated by performing two parallel experiments in
sealed, high-pressure glass vessels; one was heated to
50 °C, while the second was performed at room tem-
perature. Each reaction was sampled daily, and the
samples were characterized by NMR and IR spectros-
copy as well as SEC. In both cases, hydroxylated product
(
The resonance at 44 ppm is due to the methine carbon
of the minor 4,3-repeat unit (<5%) in the PI precursor
to PEP.) The resonance at 73 ppm is in the typical range
for a carbon R to a hydroxyl group, and the peak at
4
1
9
13
56
was observed after 1 day by C NMR spectroscopy.
2 ppm was assigned to a methylene carbons â to a
IR spectroscopy of the two reactions revealed that both
product mixtures contained hydroxylated products, but
the reaction at 50 °C had a reasonable amount of
carbonyl-containing product as well. The characteristic
hydroxyl group. These chemical shifts were similar to
those reported for group S2 of the oxyfunctionalized
squalane products, consistent with our prediction that
the selectivity of the reaction for PEP would be similar
to that exhibited for squalane, based on the structural
similarity of the molecules. Spectral evidence suggests
the types of functional groups proposed in Figure 8 are
-
1
peak near 1700 cm was not observed for the room
temperature reaction until after 2 days. The sample
aliquots were crudely fractionated to separate the
unfunctionalized PEP from the functionalized product.
Heating does not appear to disrupt the molecular weight
distribution, or PDI, of the crude polymer products, and
relative conversions based on recovered unfunctional-
ized PEP were both 90% or greater after 2 days (see
Supporting Information). Using the aforementioned
acetylation, the average number of hydroxyl groups did
not appear to increase with longer reaction times, with
most products containing between two and four hy-
droxyl groups per polymer chain (ca. 70 repeat units)
1
3
present in the products. Again, the absence of
C
resonances expected from IR analysis for carbonyl
carbons is explained by the poor sensitivity of these
quaternary carbons.59
Derivatization of the hydroxyl groups of the oxyfunc-
tionalized products was performed with acetyl chloride
and pyridine in methylene chloride, providing further
evidence that hydroxyl groups are present in the oxy-
functionalized PEP products and accessible for further
F igu r e 7. 13C NMR spectra of 5 kg/mol PEP and functionalized PEP [ca. 50 mg/mL CDCl
starting material, (b) group P3 from Table 2. [*] designates solvent peaks.
; d
) 5 s, n ) 450]: (a) 5 kg/mol PEP
3
1
t

7
032 Boaen and Hillmyer
Macromolecules, Vol. 36, No. 19, 2003
F igu r e 8. Possible functional groups present in PEP products.
F igu r e 10. Effect of Oxone concentration on the resultant
degree of hydroxylation of 5 kg/mol PEP. Hydroxyl values
1
58
determined by H NMR analysis of the acetylated products.
F igu r e 9. Size exclusion chromatogram [10 mg/mL in THF]
for (a) 5 kg/mol PEP [M ) 8.9 kg/mol, PDI ) 1.1] and the (b)
crude functionalized product [M ) 8.8 kg/mol, PDI ) 1.1].
values relative to polystyrene standards.
n
n
M
n
or 0.7-1.4 hydroxyl groups per 100 backbone carbons.5
8
While the values were similar in both cases, the
products isolated from a few aliquots of the reaction at
5
0 °C contained slightly higher degree of functionaliza-
tion relative to those obtained from the room temper-
ature reaction.
The effect of changing the concentration of Oxone
relative to the concentration of monomer repeat units
was examined at 50 °C. Five parallel reactions were
conducted for approximately 2 days and worked up in
the usual way. Conversions, based on recovered unfunc-
tionalized PEP, increased with higher concentrations of
Oxone. The PDI of the isolated functionalized products
remained narrow, except for the reaction employing the
highest concentration of Oxone ([Oxone]0/[monomer]0 )
F igu r e 11. Relationship between the degree of hydroxylation
and the glass transition temperature (T ) of a series of
g
functionalized 5 kg/mol PEP products. Filled circles: the
oxyfunctionalized products obtained from reactions performed
at room temperature. Open circles: the oxyfunctionalized
products obtained from reactions performed at 50 °C. Filled
square: unfunctionalized PEP starting material. The degree
0
.91). This particular reaction was the first example in
our studies using this system where chain scission has
been identified by the broad PDI and the large decrease
in the number-average molecular weight (see Table S2
in Supporting Information for details). Subsequent
acetylation of the functionalized products revealed a
general trend for the degree of oxyfunctionalization
increasing with higher concentrations of Oxone (see
Figure 10 and Supporting Information).
1
of hydroxylation of PEP was determined from H NMR
analysis of acetylated products.58
g
T of the functionalized
products was determined using DSC.
presence of a variable level of other polar functional
groups (i.e., ketones). These functional groups would be
inert to the acetylation reaction and would therefore
remain unaccounted.
Th er m a l Ch a r a cter iza tion . Introduction of polar
functionality, such as hydroxyl groups, into a nonpolar
polymer can increase the glass transition temperature
Oxyfu n ction alization of a High -Molecu lar -Weigh t
Mod el P EP . Successful application of this system to a
high-molecular-weight model PEP (50 kg/mol) exempli-
fies the practical significance of this approach. The
optimal conditions for oxyfunctionalization were ob-
tained when the [Oxone]0/[monomer]0 was less than 0.25
and reaction was performed at room temperature.
Chloroform was used as the solvent due to the poor
solubility of the high-molecular-weight substrate in
methylene chloride. The reaction was worked up in the
usual way, and the functionalized material was isolated
using column chromatography. Characterization of the
functionalized product by NMR and IR spectroscopy, in
addition to SEC, was consistent with the incorporation
of tertiary alcohol and ketone products, similar to those
identified in the low-molecular-weight example, without
(
Tg) of the polymer.5,20 We measured the Tg of the PEP
starting material and several functionalized products
using differential scanning calorimetry (DSC). The
results are presented in Figure 11 as a function of
hydroxyl groups per chain to demonstrate the effect the
degree of oxyfunctionalization on the Tg. The general
trend is consistent with previously reported examples
whereby the Tg increased with higher levels of func-
tionalization. A near linear trend is observed for the
products from reactions performed at room temperature.
However, Tg’s for the samples obtained from reactions
performed at 50 °C deviate somewhat from this linear
relationship. In the previous section, the formation of
ketone functionality was pronounced in the products
obtained from reactions at 50 °C. Therefore, the vari-
ability in Tg for the samples between 1.5 and 2.5
hydroxyl groups per chain may be explained by the
6
3
chain degradation (Figure 12). While a small amount
of coupling was identified by SEC, it is impressive that

Macromolecules, Vol. 36, No. 19, 2003
Oxyfunctionalization of Model Polyolefins 7033
Ack n ow led gm en t. This research was funded by the
Dow Chemical Company. We thank Dr. J oel M. Harris,
Dr. Elizabeth A. Lewis, Dr. Luis M. Alcazar-Roman,
Stephen F. Hahn, and Dr. Dana Reed for helpful
discussions.
Su p p or tin g In for m a tion Ava ila ble: Experimental sec-
tion and tabular data summarizing the reaction conditions for
the reactions performed at 50 °C and different Oxone concen-
trations. This material is available free of charge via the
Internet at http://pubs.acs.org.
Refer en ces a n d Notes
(
1) Pinazzi, C.; Brosse, J . C.; Pleurdeau, A.; Reyx, D. Appl. Polym.
Symp. 1975, 26, 73.
(
(
2) Shultz, D. N.; Turner, R. S. Rubber Chem. 1982, 55, 809.
3) McGrath, M. P.; Sall, E. D.; Tremont, S. J . Chem. Rev. 1995,
F igu r e 12. Size exclusion chromatogram [10 mg/mL in THF]
9
5, 381.
for (a) 50 kg/mol PEP [M
n
) 62.0 kg/mol, PDI ) 1.1] and the
b) crude functionalized product [M ) 63.7 kg/mol, PDI ) 1.1].
values relative to polystyrene standards.
(
4) Rossi, A.; Icarnato, L.; Tagliaferri, V.; Acierno, D. J . Polym.
Eng. 1995, 14, 191.
5) Chung, T. C.; Raate, B. E.; Shultz, D. N. Macromolecules
(
n
M
n
(
1
988, 21, 1903.
the narrow PDI of the model material (1.1) was retained
through the functionalization. Acetylation of the isolated
product mixture further demonstrated the presence of
hydroxyl groups in the products, and integration values
(6) Halld e´ n, A.; Wessl e´ n, B. J . Appl. Sci. 1996, 60, 2495.
(7) Halld e´ n, A.; Wessl e´ n, B. J . Appl. Sci. 2000, 78, 2416.
(8) Shilov, A. E.; Shul’pin, G. B. Chem. Rev. 1997, 97, 2879.
(9) Crabtree, R. H. Chem. Rev. 1995, 95, 987.
(
(
(
10) Goldsheger, N. F.; Eskova, V.; Shilov, A. E.; Shteinman, A.
A. Zh. Fiz. Khim. 1972, 46, 1353.
11) Escova, V.; Shilov, A. E.; Shteinman, A. A. Kinet. Katal. 1972,
1
from the H NMR spectrum provided an average of 12
hydroxyl groups per chain (730 repeat units). Thermal
analysis of the functionalized material showed a 4 °C
increase in the Tg as compared to the high-molecular-
weight PEP starting material.
1
3, 534.
12) Yamanaka, I.; Akimoto, T.; Otsuka, K. Chem. Lett. 1994,
511.
(13) Yamanaka, I.; Otsuka, K. J . Chem. Soc., Faraday Trans.
994, 90, 451.
1
1
While the oxyfunctionalization of squalane, low-
molecular-weight PEP, and high-molecular-weight PEP
were selective for the formation of tertiary alcohols, the
degree of hydroxylation per 100 backbone carbons
generally decreased on going from squalane to the
polymeric cases. For sterically hindered manganese
porphyrins such as Mn(TDCPP)OAc, regioselectivity in
the oxyfunctionalization of low-molecular-weight n-
(
14) Yamanaka, I.; Nakagaki, K.; Akimoto, T.; Otsuka, K. J .
Chem. Soc., Perkin Trans. 2 1996, 2511.
(15) Yamanaka, I.; Akimoto, T.; Nakagaki, K.; Otsuka, K. J . Mol.
Catal. A: Chem. 1996, 110, 119.
(
16) Neumann, R.; Khenkin, A. M.; Dahan, M. Angew. Chem., Int.
Ed. Engl. 1995, 34, 1587.
(
(
17) Lu, B.; Chung, T. C. J . Polym. Sci., Part A 2000, 38, 1337.
18) Niki, R.; Shiono, T.; Kamiya, Y. J . Appl. Polym. Sci. 1975,
19, 3341.
19) Valliant, D.; Lacoste, J .; Dauphin, G. Polym. Degrad. Stab.
1994, 45, 355.
20) Kondo, Y.; Garcia-Cuadrado, D.; Hartwig, J . F.; Boaen, N.
K.; Wagner, N. L.; Hillmyer, M. A. J . Am. Chem. Soc. 2002,
124, 1164.
(21) Waller, B. J .; Lipscomb, J . D. Chem. Rev. 1996, 96, 2623.
22) Feig, A. L.; Lippard, S. J . Chem. Rev. 1994, 94, 759.
23) Kurtz, D. M. J . Chem. Rev. 1990, 90, 585.
24) Wang, Z.; Martell, A. E.; Monekaitis, R. J . J . Chem. Soc.,
Chem. Commun. 1998, 1523.
(
alkanes is influenced by the degree of steric bulk around
the metal and the length of the n-alkane.4
3,64
The
(
internal carbons are less accessible in the longer al-
kanes. Therefore, the observed decrease in the degree
of hydroxylation with increasing molecular weight of our
substrates could be a result of the increased portion of
the alkane substrate that cannot readily access the
sterically hindered active site.
(
(
(
(
25) Barton, D. H. R.; Boivin, J .; Gastiger, M.; Morzycki, J .; Hay-
Motherwell, R. S.; Motherwell, W. B.; Ozabik, N.; Schwartzen-
trubar, K. J . Chem. Soc., Perkin Trans. 1 1986, 947.
26) Doller, D.; Barton, D. H. R. Acc. Chem. Res. 1992, 25, 504.
Con clu sion s
(
The direct oxyfunctionalization of PEP, a model
polyolefin, and squalane, a small molecule, structural
analogue of PEP, has been demonstrated utilizing a
Mn(TDCPP)OAc/imidazole system in the presence of a
phase transfer agent and an oxygen donor, such as
Oxone. Reasonable yields of functionalized products
were obtained and completely characterized. Both the
low-molecular-weight and high-molecular-weight func-
tionalized polymers exhibit essentially no degradation
by SEC under the reaction conditions. Furthermore, the
reaction products suggest that the oxyfunctionalization
reaction is selective for the hydroxylation of tertiary
carbons. There is also the possibility that secondary
alcohols are formed and then rapidly oxidized to ketone
products. The versatility of this system to employ a
variety of oxygen donors, in conjunction with the pos-
sibility of using a supported Mn(TDCPP) catalyst under
(27) Vincent, J . B.; Huffman, J . C.; Christou, G.; Li, Q.; Nanny,
M. A.; Hendrickson, D. N.; Fong, R. H.; Fish, R. H. J . Am.
Chem. Soc. 1988, 110, 6898.
(
28) Fish, R. H.; Fong, R. H.; Vincent, J . B.; Christou, G. J . Chem.
Soc., Chem. Commun. 1988, 1504.
(29) Kitajima, N.; Fukui, H.; Moro-oka, Y. J . Chem. Soc., Chem.
Commun. 1988, 485.
(30) Kitajima, N.; Fukui, H.; Moro-oka, Y. J . Chem. Soc., Chem.
Commun. 1991, 102.
(
31) Kim, C.; Chen, J .; Kim, J .; Que, L., J r. J . Am. Chem. Soc.
1
997, 119, 5964.
(32) Mansuy, D. Pure Appl. Chem. 1987, 59, 759.
(33) Meunier, B. Bull. Soc. Chim. Fr. 1986, 4, 578.
(34) Meunier, B. Chem. Rev. 1992, 92, 1411.
(35) Stryer, L. Biochemistry, 4th ed.; W. H. Freeman & Co.: New
York, 1995.
(36) Montellano, P. O. D., Ed. Cytochrome P-450: Structure,
Mechanism, and Biochemistry; Plenum Press: New York,
1
985.
(
(
(
37) Ullrich, V. Top. Curr. Chem. 1979, 83, 67.
38) Porter, T. D.; Coon, M. J . J . Biol. Chem. 1991, 266, 13469.
39) Previously, the formation of the active species has been
observed by UV-vis spectroscopy and was reported to form
6
5
mild reaction conditions, makes this system an at-
tractive polyolefin functionalization methodology.

7
034 Boaen and Hillmyer
Macromolecules, Vol. 36, No. 19, 2003
only in the presence of a nitrogen base cocatalyst for most
oxygen donors. See: Campestrini, S.; Furia, F. D.; Labat, G.;
Novello, F. J . Chem. Soc., Perkin Trans. 2 1994, 2175.
resolution and facile identification of products by 13C NMR
spectroscopy.
(57) Reactions were also performed in the absence of PEP to prove
the new chemical shifts were not due to byproducts formed
in the reaction.
(
(
40) Volz, H.; Muller, W. Chem. Ber./ Recl. 1997, 130, 1099.
41) Mohajer, D.; Tayebee, R.; Goudarziafshar, H. J . Chem. Res.,
Synop. 1998, 822.
(58) The error in average number of OH/chain was estimated by
performing three parallel acetylation reactions of the same
functionalized product, working up the reactions in the same
manner described in the Supporting Information, and inte-
grating the resultant 1H NMR spectra. The standard devia-
tion was calculated from these values and used as an estimate
of the error for each sample.
(59) A high-sensitivity spectrum of a functionalized low-molecular-
weight PEP was obtained on a 300 MHz instrument equipped
with a broadband C{H} probe over 12 h using ca. 10 wt % in
CDCl3 and a longer relaxation delay time of 20 s; however,
the carbonyl carbons could still not be distinguished.
(60) Pinazzi, C.; Lenain, J .-C.; Brosse, J .-C.; Legeay, G. Makromol.
Chem. 1976, 177, 3139.
(
(
42) Hoffman, P.; Robert, A.; Meunier, B. Bull. Soc. Chim. Fr.
1
992, 129, 85.
43) Battioni, P.; Renaud, J . P.; Bartoli, J . F.; Reina-Artiles, M.;
Fort, M.; Mansuy, D. J . Am. Chem. Soc. 1988, 110, 8462.
44) Hucul, D. A.; Hahn, S. F. U.S. Patent 5,700,878, 1997.
45) Gotro, J . T.; Graessley, W. W. Macromolecules 1984, 17, 2767.
46) van der Made, A. W.; Hoppenbrouwer, R. J .; Nolte, R. J . M.;
Drenth, W. Recl. Trav. Chim. Pays-Bas 1988, 107, 15.
47) Adler, A. D.; Longo, F. R.; Kampas, F.; Kim, J . J . Inorg. Nucl.
Chem. 1970, 32, 2443.
48) Robert, A.; Loock, B.; Momenteau, M.; Meunier, B. Inorg.
Chem. 1991, 30, 706.
49) Lamar, G. N.; J ensen, F. A. W. The Porphyrins; Academic
Press: New York, 1978; Vol. 4.
(
(
(
(
(
(
(
(61) Gonz a´ lez-Gonz a´ lez, V. A.; Neira-Vel a´ quez, G. N.; Angulo-
S a´ nchez, J . L. Polym. Degrad. Stab. 1998, 60, 33.
(62) Yaykin, O.; Sava s¸ c¸ i, O. T.; Baysal, B. M. J . Appl. Polym. Sci.
1989, 37, 2383.
50) Gonz a´ lez, B.; Kouba, J .; Yee, S.; Reed, C. A.; Kirner, J . F.;
Scheidt, W. R. J . Am. Chem. Soc. 1975, 97, 3247.
51) Groves, J . T.; Neumann, R. J . Org. Chem. 1988, 53, 3891.
52) Sorokin, A.; Robert, A.; Meunier, B. J . Am. Chem. Soc. 1993,
(
(
(
63) Reactions employing concentrations of [Oxone]0/[monomer]0
1
15, 7293.
>
0.25 and reactions performed at 50 °C resulted in products
with a decreased number-average molecular weight and a
broader PDI.
(
(
(
53) Querci, C.; Ricci, M. Tetrahedron Lett. 1990, 31, 1779.
54) Robert, A.; Meunier, B. New J . Chem. 1988, 12, 885.
55) The colored impurity can be reduced by passage through an
alumina column.
56) In a separate experiment, nearly complete conversion of PEP
was observed after 2 days. Higher sample concentrations (70
mg/mL vs 14 mg/mL) were required for maximized peak
(
64) Cook, B. R.; Reinert, T. J .; Suslick, K. S. J . Am. Chem. Soc.
1
986, 108, 7281.
(
(
65) Leadbeater, N. E.; Marco, M. Chem. Rev. 2002, 102, 3217.
MA0347621