Control of degradation reactions during radical functionalization of polypropylene in the melt

Article abstract of DOI:10.1021/ma0400315

Isotactic polypropylene (iPP) gives very extensive degradation when treated with peroxides above its melting temperature in mechanical mixers or extruders. This undesired reaction is very modestly affected by maleate molecules which on the contrary actively compete with side reactions of macroradicals in the case of ethylene polymers. In this work iPP was treated in a Brabender mixer at 180°C with peroxide and different selected molecules capable of promptly reacting with the macroradical formed on iPP chain and converting it into a more stable free radical. Furan derivatives, successfully used for iPP cross-linking without any remarkable increase of MFR, were used as free radical removers and maleic anhydride as functionalizing monomer. The results indicate a detectable improvement with respect to the use of maleic monomers and peroxide only, allowing to us get a significant grafting of functional groups and only partial degradation. Moreover, furan derivatives bearing various reactive substituents were used as functionalizing molecules. The results are discussed in the frame of the general mechanism proposed for the free radical functionalization of polyolefin in the melt.

Full text of DOI:10.1021/ma0400315

8414
Macromolecules 2004, 37, 8414-8423
Control of Degradation Reactions during Radical Functionalization of
Polypropylene in the Melt
Ser en a Coia i,† Elisa P a ssa glia ,‡ Ma u r o Aglietto,§ a n d F r a n cesco Cia r d elli*,‡,§
Scuola Normale Superiore, Piazza dei Cavalieri 7, 56126 Pisa, Italy; CNR-ICCOM Pisa Section,
Department of Chemistry and Industrial Chemistry, University of Pisa, Via Risorgimento 35,
56126 Pisa, Italy
Received February 4, 2004; Revised Manuscript Received J une 28, 2004
ABSTRACT: Isotactic polypropylene (iPP) gives very extensive degradation when treated with peroxides
above its melting temperature in mechanical mixers or extruders. This undesired reaction is very modestly
affected by maleate molecules which on the contrary actively compete with side reactions of macroradicals
in the case of ethylene polymers. In this work iPP was treated in a Brabender mixer at 180 °C with
peroxide and different selected molecules capable of promptly reacting with the macroradical formed on
iPP chain and converting it into a more stable free radical. Furan derivatives, successfully used for iPP
cross-linking without any remarkable increase of MFR, were used as free radical removers and maleic
anhydride as functionalizing monomer. The results indicate a detectable improvement with respect to
the use of maleic monomers and peroxide only, allowing to us get a significant grafting of functional
groups and only partial degradation. Moreover, furan derivatives bearing various reactive substituents
were used as functionalizing molecules. The results are discussed in the frame of the general mechanism
proposed for the free radical functionalization of polyolefin in the melt.
In tr od u ction
reaction occurs by the addition of the terminal macro-
radical, formed through â-scission, to the monomer
double bond. As for polyethylene,21 diethyl maleate
(DEM) or a mixture of DEM/MAH 2:1 mol:mol with
dicumyl peroxide (DCP) gave PP with good functional-
ization degrees (FDs), ranging from 0.2 to 1.0 mol %,
but with a remarkable decrement of PP molecular
weight (MW).22 The competition of the projected grafting
reaction against chain scission can be improved through
the formation of stabilized macroradicals. The recent
researches on this topic concern the control of the
degradation reactions. With this aim molecules or
molecular systems (X) have been proposed, able to
create a new stabilized macroradical not involved in
â-scission, which gives place to transfer and/or coupling
reactions.23 Cartier et al.24 studied the radical function-
alization of PP with MAH or GMA and a suitable
comonomer such as the styrene (STY). The styrene
reacts with macroradicals creating more stabilized
radicals which subsequently react with the polar mono-
mer. The reactivity of GMA toward the styryl radical is
believed to be greater than its reactivity toward the PP
macroradical; hence, the use of this comonomer pro-
duces a synergistic effect on the grafting. In this way,
it is possible to limit the degradation and simulta-
neously to obtain functionalized products.
The chemical modification of already existing poly-
mers is widely used for the conversion of low cost
commodity polymers, such as the polyolefins (PO), into
useful reactive polymers. The functionalization reaction
of POs has been studied intensively by grafting of polar
groups onto the polymer backbone.1-5 Indeed these
products act as in situ compatibilizer6 of polymer blends
and composites.7 Free radical initiators give rise, in the
bulk and at the high-temperature necessary for good
rheology, to scission, grafting, branching, and cross-
linking reactions.8 The polymer degradation through
â-scission reactions4,9 predominates over the other reac-
tions in the case of isotactic polypropylene (iPP). Prob-
ably hydrogen abstraction from PP backbone proceeds
preferably at the level of the tertiary carbon atom, even
though the abstraction from secondary carbon atoms or
from methyl groups is possible as well.10,11 Primary and
secondary radicals are reported to give more easily the
coupling reaction than the tertiary ones which instead
undergo â-scission reactions or disproportionation reac-
tions. Moreover, the reactive radical formed by the
addition of the macroradical to the double bond of the
functionalizing monomers gives transfer reaction to the
polymer backbone, thus producing again tertiary mac-
roradicals which lead again to â-scission.12-14 In many
works, different reactive monomers and organic perox-
ides have been employed for PP radical functionaliza-
tion in the melt. The most studied monomers are
maleic anhydride (MAH)15-17 or glycidyl methacrylate
(GMA).18-20 A careful analysis of the products shows,
in any reaction conditions, the decreasing of the average
molecular weight of the polymer correlating to an
increase of the melt flow rate (MFR). The grafting
Recently the chemical cross-linking of polypropylene25
has been studied and performed under dynamic condi-
tions using a peroxide and a furan or bis(maleimide)
based coagent, such as bis(2-furanyl)aldazine (VP3) or
N,N-p-phenylenebis(maleimide) (BMI), as the cross-
linking promoter.26-32 These coagents are able to react
properly with the macroradicals, produced by the pri-
mary radicals, thus spacing the free radical from the
backbone and promoting chain extension and/or pre-
venting the â-scission (Scheme 1).
* To whom correspondence should be addressed. Telephone:
+39-50-2219-229. Fax: +39-50-2219-320. E-mail: fciard@
dcci.unipi.it.
The furan derivatives, having the heterocyclic ring
conjugated with a double bond bearing an electron
attracting group substituent, display a high addition
rate to the macroradical with the formation of resonance
† Scuola Normale Superiore di Pisa.
‡ CNR-ICCOM Pisa Section.
§ University of Pisa.
10.1021/ma0400315 CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/08/2004

Macromolecules, Vol. 37, No. 22, 2004
Control of Polypropylene Radical Functionalization 8415
Sch em e 1. P P Ma cr or a d ica ls Rea ction s w ith F u r a n Coa gen ts
acetate (21.4 g, 0.189 mol). Afterward, 2.4 mL of a solution
piperidine-acetic acid (1:3 vol) was added. The solution was
maintained at reflux of the solvent for 24 h. The solvent was
distilled off at reduced pressure. Then the remainder was
dissolved in dichloromethane and washed with water. The
organic phase was dried on anhydrous Na2SO4 and then
evaporated. The resulting oil was separated in its components
through chromatography using silica as stationary phase and
chloroform as eluent. The cyano 3-(2-furanyl) propenoate (a
yellow solid) was recovered through crystallization from a
solution in ethyl acetate-hexane (1:1 vol). A 19.3 g (64.8%)
stabilized radicals. These molecules are now (Scheme
1) proposed as functionalization coagents able to control
the intrachain transfer reaction, responsible of PP
extensive degradation, and able to introduce onto the
polymer backbone suitable polar groups.
On this basis, the iPP radical functionalization reac-
tions in the melt by peroxide, maleate derivatives, and
furan coagents have been studied as a function of the
feed composition, aiming at obtaining of high grafting
values (FDs), keeping unaltered the MW of PP and
trying to rationalize the complex mechanism of the
process in order to check the capability of furan coagents
to control the PP degradation. Two series of experiments
have been carried out: in the former case maleic
anhydride (MAH) and a furan derivative, butyl 3-(2-
furanyl)propenoate (BFA), have been employed; in the
latter case, furan coagents bearing suitable functional
groups and able to control the PP â-scission reaction
have been used in the absence of any maleic derivatives.
yield was obtained. IR main peaks (cm-1): 3130-3040 νH-Cd
2997-2970 νH-C-, 2223 νCN, 1716 νCdO, 1621 νCdC, 1464
δCH -O-CO, 932 γCH ring, 877 γCH
1H NMR assign-
,
.
furan ring
furan
2
ments: 1.2 ppm, 3 protons, triplet (CH3 ethyl groups); 4.2 ppm,
4 protons, quartet, (O-CH2 ethyl); 6.5 ppm, double doublet, 1
proton (furane ring position 4); 7.2 ppm, doublet, 1 proton,
(furane ring position 3); 7.6 ppm, doublet,1 proton (olefin
position 3); 7.9 ppm, doublet 1 proton (furane ring position 5).
In str u m en ts. PP samples have been prepared using a
Brabender Plastograph PL2100 mixer, torque and temperature
data have been acquired by Brabender Mixing software
WinMix ver.1.0. FT-Infrared spectra were performed with a
Fourier Transform spectrometer “Perkin-Elmer FT-IR 1760-
X”. Spectra of chemicals have been made on films produced
by evaporation of CHCl3 solutions on KBr windows. PP sample
spectra have been acquired on films made by compression
molding. Proton magnetic resonance spectra were performed
with a spectrometer “Varian Gemini 200 MHz”; the spectra of
chemicals were obtained on solutions of the compounds in
deuterochloroform, and chemical shifts were assigned in ppm
using tetramethylsilane (TMS) as internal standard.
Melt F u n ction a liza tion of P P Sa m p les. All samples were
prepared, with the same procedure, in the Brabender mixer
by melt-mixing the components at the temperature of 180 °C
and rotor speed of 50 rpm. Then, 20 g of iPP were introduced
into the Brabender chamber. After complete fusion of PP,
functionalizing molecules and peroxide were added to the
molten bulk. The reaction was carried out for 25 min and then
stopped. The recovered polymer was sequentially extracted
with boiling acetone and toluene. In each case, the residues
were dried until constant weight.
Exp er im en ta l Section
Ma ter ia ls a n d Rea gen ts. Isotactic polypropylene (iPP)
homopolymer having a melt flow rate (MFR) of 0.24 g/10 min
supplied by Borealis BE 50/E1271, and propylene homopoly-
mer Moplen Q30P, MFR of 1.0 g/10 min, have been used. 2,5-
Dimethyl-2,5-di-tert-butylperoxi hexane (Akzo-Nobel) and di-
cumyl peroxide (Sigma-Aldrich) have been used as provided
without further purification. Maleic anhydride (Sigma-Aldrich)
has been recrystallized from benzene. 2-Furaldehyde (Sigma-
Aldrich), diethyl malonate (Sigma-Aldrich), and ethyl cyano-
acetate (Sigma-Aldrich) have been purified by distillation
under reduced pressure. Pyridine and piperidine (Carlo Erba)
have been purified by distillation on KOH pellets. Solvents
have been purified by standard procedures.
Syn t h esis of F u r a n yl Der iva t ives. Butyl 3-(2-furanyl)
propenoate (BFA) and ethyl 2-carboxyethyl 3-(2-furanyl) pro-
penoate (CEFA) were prepared as reported in the litera-
ture.25,32 The structural characteristics of these two composites
are summarized in Appendix 1.
Syn t h esis of Cya n o 3-(2-F u r a n yl)p r op en oa t e (CF A).
Into a three-neck 500 mL flask, equipped with a charge funnel,
refrigerant, and mechanical agitator, were introduced, in a
nitrogen atmosphere, reactants in the following order: dioxane
(440 mL), furfuraldehyde (15 g, 0.156 mol), and ethyl cyano-
Th eor etica l Ca lcu la tion for Extin ction Coefficien t
Deter m in a tion . Quantum mechanical calculations of the
extinction coefficients were performed by using density func-
tional theory (DFT)33 with the B3LYP34 hybrid functional and

8416 Coiai et al.
Macromolecules, Vol. 37, No. 22, 2004
Ta ble 1. Mod ifica tion of P P Sa m p les w ith MAH, P er oxid e a n d Va r ia ble Am ou n ts of BF A
end-torque
values
(N‚m)
c
d
peroxide
(% mol)
BFA
(% mol)
MAH
(% mol)
FDSAH, SA
(% mol)
FDBFA
FYe
(%)
samplea
Rb
(% mol)
PPAMC1.0
PPAMC1.1
PPAMC1.2
PPAMC1.3
PPAMC1.4
0.06
0.06
0.06
0.06
0.06
4.28
4.28
4.28
4.28
4.28
2.9
4.2
4.8
5.1
5.4
0.46
0.48
0.51
0.54
0.85
10.7
11.8
12.4
13.2
21.6
0.09
0.13
0.17
0.22
47.5
33
25
0.08
0.06
0.07
0.20
19
a
b
The PP used is Moplen Q30P; MFR ) 1 g/10 min. R is the MAH/BFA molar ratio in the feed. c Number of succinic anhydride (SAH)
d
and succinic acid (SA) inserted groups per 100 monomeric units of the polymer. Number of BFA inserted groups per 100 monomeric
units of the polymer. e FY ) functionalization yield, percent of PP attached carbonyl groups with respect to their total initial number.
[FY ) 2FDSAH,SA + FDBFA/BFA + 2MAH × 100].
F igu r e 1. Behavior of the mixing torque for the samples modified with BFA, peroxide and a constant amount of MAH: (A)
PPAMC1.0; (B) PPAMC1.1; (C) PPAMC1.2; (D) PPAMC1.3; (E) PPAMC1.4.
the 6-31G* basis set.35 The Gaussian program36 was exploited.
The frequencies and extinction coefficients were calculated for
the geometries of all the species optimized at the same
quantum mechanical level. The quantum mechanical model
which was exploited here has been proven to give reliable
results for both frequencies and extinction coefficients of
carbonyl stretching modes: see, for example, ref 37.
at 135 °C on decahydronaphthalene solution of samples
previously washed with warm acetone. For each analyzed
samples three solutions at different concentrations have been
prepared adding a polymer amount ranging from 5 to 35 mg
to 10 mL of solvent. Through the evaluation of the flow time
for the solvent and for the three solutions, for each sample,
the intrinsic viscosity [η] values have been determined by
linear fitting of experimental points. The average viscometric
molecular weight (PMη) was determined by the Mark-Hou-
wink equation (a ) 0.8 and Kη ) 11.0 × 10-3 mL/g39).
Decon volu tion Meth od s. The deconvolution procedure
applied to the bands in the CdO stretching region of FT-IR
spectra of MAH/BFA functionalized PP samples was the same
described for PE samples functionalized by using a MAH/
diethyl maleate (DEM) mixture.38 Six Lorentzian-shaped
bands were considered in the range between 2000 and 1600
cm-1: (1) symmetric νCdO of free anhydride (1790 cm-1); (2)
symmetric νCdO of hydrogen bonded anhydride (1770 cm-1);
(3) asymmetric νCdO of free anhydride (1860 cm-1); (4) asym-
metric νCdO of hydrogen bonded anhydride (1840 cm-1); (5) νCd
O of the ester (1736 cm-1); (6) νCdO of the carboxylic acid (1710
cm-1), whose absorbances vary by increasing the functional-
ization degree. As an example, in Figure 6 (Appendix 2), we
show the results of the deconvolution procedure for PPAMC1.1
functionalized with the BFA/MAH system. The deconvolution
was performed by using a NLSF method (nonlinear least
squares fitter) (Origin 6.0 software) by optimizing for each
peak the values of the area, the half-width of the band, and
the maximum of the peak starting from determined values and
by varying them between a predetermined range.
Gel P er m ea tion Ch r om a togr a p h y. A Waters Alliance
GPCV-2000 Series System apparatus equipped with three
Waters Styragel HT 6E columns (MW 5000-1000000) and one
Waters Styragel HT 3 (MW 500-30000) column, with an
average particle size of 10 µm, a differential refractive index
(DRI) detector, and a differential viscometer was used. Polymer
solutions were prepared with amounts of 4-5 mg of polymer
in 8 mL of 1,2,4-trichlorobenzene (TCB) containing a little
amount of antioxidant (BHT) to prevent any degradation and
eluted at 160 °C and 1 mL/min flow rate. The calibration was
made with narrow MWD standards polystyrene, and calcula-
tions were carried out by Millennium software.
Resu lts a n d Discu ssion
A. F u n ction a lizin g System Ba sed on P er oxid e,
Ma leic An h yd r id e, a n d Coa gen t. As already men-
tioned, the previously described selected furan deriva-
tives favor polypropylene cross-linking and chain ex-
Viscosity Mea su r em en ts. A capillary Ubbehlode viscom-
eter has been used. Viscosity measurements have been achieved

Macromolecules, Vol. 37, No. 22, 2004
Control of Polypropylene Radical Functionalization 8417
Ta ble 2. Mod ifica tion of P P Sa m p les w ith BF A, P er oxid e a n d Va r ia ble Am ou n ts of MAH
end-torque
values
(N‚m)
c
d
peroxide
(% mol)
BFA
(% mol)
MAH
(% mol)
FDSAH, SA
(% mol)
FDBFA
FYe
(%)
samplea
Rb
(% mol)
PPAMC2.0
PPAMC2.1
PPAMC2.2
PPAMC2.3
0.06
0.06
0.06
0.06
0.17
0.17
0.17
0.17
2.6
3.5
4.7
4.5
0.15
0.16
0.14
0.16
0.86
2.57
3.43
5
15
20
0.20
0.36
0.47
29.6
16.2
15.6
a
b
The PP used is Moplen Q30P; MFR ) 1 g/10 min. R is the MAH/BFA molar ratio in the feed. c Number of succinic anhydride (SAH)
d
and succinic acid (SA) inserted groups per 100 monomeric units of the polymer. Number of BFA inserted groups per 100 monomeric
units of the polymer. e FY ) functionalization yield, percent of PP attached carbonyl groups with respect to their total initial number [FY
) 2FDSAH,SA + FDBFA/ BFA + 2MAH × 100].
Ta ble 3. Molecu la r Weigh t of F u n ction a lized IP P
Sa m p les
b
peroxide
BFA
MAH
FDTOT
[η]
MWη
samplea
PP
PPAMC1.0
PPAMC1.3
PPAMC2.0
(% mol) (% mol) (% mol) (% mol) (dL/g) (g/mol)
0
1.6
0.9
2.1
1.1
161 000
78 000
224 000
100 000
0.06
0.06
0.06
4.28
4.28
0.46
0.61
0.15
0.17
0.17
a
b
The PP used is Moplen Q30P; MFR ) 1 g/10 min. Number
of total functional inserted groups per 100 monomeric units of the
polymer.
and content of the various functional groups derived
from MAH and BFA grafting, respectively (Appendix
2). The calculated FDs show (Table 1) that the increase
of BFA concentration in the feed, while MAH and
peroxide are kept constant, gives an increase not only
of FDBFA but also of FDSAH,SA. It appears therefore that
the presence of BFA coagent during the PP radical
functionalization with peroxide and MAH, beyond limit-
ing the polymer degradation, favors the grafting reac-
tion of MAH, with larger monomer conversion (func-
tionalization yield) by increasing the amount of BFA
(thanks to the increase of efficiency observed in the
experiments with MAH/BFA ) 19) (Table 1).
A second set of experiments was performed by keep-
ing constant the BFA feed content and peroxide amount
while varying the MAH amount (Table 2). The MAH/
BFA ratio was kept below 20, as this value had given
the best FY in the former set of experiments.
The mixing torque shows a similar trend as in the
previous case, however the end-torque values increase
with increasing the amount of starting MAH. The FDs
indicate that in the selected conditions the grafting of
BFA is almost complete and the amount of grafted MAH
increases with its presence in the feed, while the
functionalization yield (FY) goes in the opposite direc-
tion.
After the functionalization reactions all the first and
second set samples were completely soluble in toluene,
as the starting iPP; therefore, it is possible to assume
that no cross-linking has taken place, but degradation
reactions cannot be excluded. To estimate the possible
PP degradation, the viscometric molecular weight (MWη)
was evaluated for a few selected samples. In the case
of the functionalization of PP with MAH and peroxide
(run PPAMC1.0) a remarkable decrease of MWη was
observed (Table 3).
The replacement of MAH with BFA (run PPAMC2.0)
gives still a decrease of MWη but to a lower extent, in
spite the BFA/peroxide ratio is only 2.5. In the presence
of both BFA and MAH (run PPAMC1.3) the MWη
becomes even higher than that of not-treated PP. The
values as such are not directly comparable as the
viscosity is certainly increased by the presence of ester
groups. A very rough correction can be introduced by
F igu r e 2. FT-IR spectra in the range 1950-1250 cm-1 of PP
samples modified with peroxide, MAH and variable amounts
of BFA.
tension (or branching) toward chain scission. The
formation of cross-linking points is shown by the mixing
torque growth after addition of the reagents and also
by the high gel content and low values of the melt flow
rate (MFR).25 Among the above furan derivatives, the
butyl 3-(2-furanyl)propenoate (BFA) was used in the
present work during PP functionalization reactions
together with MAH and peroxide. Indeed BFA grafts
on PP, maintaining a low MFR value without cross-
linking.25
A first set of experiments was performed starting with
constant amounts of 2,5-dimethyl-2,5-di-tert-butylper-
oxihexane and maleic anhydride (MAH) while varying
the amount of BFA (Table 1).
The mixing torque (Figure 1) shows in the last
experiments a decrease after the addition (180 s) of
the reagents to the molten polymer, but with in-
creasing amounts of BFA in the feed (runs PPAMC1.1-
PPAMC1.4), the effect is reduced thus suggesting a
lower degradation (Table 1).
The final samples were analyzed by FT-IR after
washing with boiling acetone in order to remove the
unreacted monomers and low MW products. FT-IR
spectra present superposition of diagnostic bands (ester,
anhydride, and acid obtained by hydrolysis of anhydride
groups) (Figure 2) with peaks at 1710 cm-1 stretching
CdO (νCO) for the carboxylic group, at 1780 and 1865
cm-1 νCO symmetrical and νCO asymmetrical, respec-
tively, for the anhydride groups, and at 1736 cm-1 νCO
for BFA ester group. This indicates that different types
of groups are grafted onto PP chains.
By a nonlinear fitting of experimental points, a
deconvolution of the spectra has been performed. In this
way it is possible to obtain separately the diagnostic
band areas, thus allowing the determination of the type

8418 Coiai et al.
Macromolecules, Vol. 37, No. 22, 2004
Ta ble 4. Molecu la r Weigh t Aver a ges a n d P olyd isp er sity In d ex of F u n ction a lized iP P Sa m p les
SEC measurements (g/mol)
b
peroxide
(% mol)
BFA
(% mol)
MAH
(% mol)
FDTOT
samplea
PPtq
Mn
Mw
Dn
(% mol)
121 000
45 400
113 000
128 000
97 400
102 000
131 000
102 000
525 000
124 000
319 000
386 000
282 000
396 000
340 000
299 000
4.3
2.7
2.8
3.0
2.9
3.9
2.6
2.9
PPAMC1.0
PPAMC1.2
PPAMC1.3
PPAMC1.4
PPAMC2.0
PPAMC2.2
PPAMC2.3
0.06
0.06
0.06
0.06
0.06
0.06
0.06
4.28
4.28
4.28
4.28
0.46
0.57
0.61
1.05
0.15
0.50
0.63
0.13
0.17
0.22
0.17
0.17
0.17
2.57
3.43
a
b
The PP used is Moplen Q30P; MFR ) 1 g/10 min. Number of total functional inserted groups per 100 monomeric units of the polymer.
Sch em e 2. P r oba ble Rea ction s of th e P P Ma cr or a d ica l Usin g th e BF A/MAH P er oxid e System
dividing MWη by the FD; the values of MWη/FD ×
10-3 are respectively 170, 367, and 667 for the runs
PPAMC1.0, PPAMC1.3, and PPAMC2.0, thus indicating
a remarkable effect of BFA to hinder the chain scission.
These data indicate that the BFA/MAH combination
allows one to improve the functionalization degree and
control the degradation.
PPAMC1.3) the Mw goes from a maximum value of
396 000 (sample PPAMC2.0) to lower values, except for
the sample PPAMC1.3 prepared by using the higher
amount of MAH, for which the Mw was higher than for
the samples PPAMC2.2 and PPAMC2.3.
On the basis of these results, the related mechanism
can be tentatively described as follows. The primary
radicals formed by decomposition of the peroxide can
produce PP macroradicals (PP•) by hydrogen abstrac-
tion. These PP• macroradicals can now either undergo
â-scission or react with BFA or MAH. In the latter case,
the free radical is removed from the chain backbone (as
shown in the Scheme 2) to give 1 and 2, respectively.
The molecular weight distributions were also ana-
lyzed by size exclusion chromatography (SEC). The
calculated values of the major molecular weight aver-
ages, Mn and Mw, and the polydispersity index, Dn
Mw/Mn, are given in Table 4.
)
By using MAH and peroxide alone (PPAMC1.0 sample)
the molecular weight distribution is narrowed with
respect to the neat PP, and the percentage of high
molecular weight species is also decreased. Besides, the
degradation of the polymer is less significant for all the
sample treated with BFA, as evinced by the higher Mw
and Dn values. Moreover this effect seems major by
increasing the BFA concentration in the feed. In fact,
with the MAH kept constant and by the BFA con-
centration increased (samples PPAMC1.0, PPAMC1.2,
1 is stabilized by resonance and has low tendency to
give H-abstraction; it should react predominantly with
itself or other macroradicals to give chain extension and
to a minor extent the disproportionation reactions.
Actually Romani et al.25 in their experiments about PP
radical cross-linking by using BFA showed through FT-
IR evidences, supported by rheological experiments, that
this coagent is able to prevent the â-scission process not
promoting the cross-linking by coupling of macroradi-
cals. In particular, the FT-IR analysis of the PP samples
treated with BFA pointed out the presence of a peak at
1646 cm-1, visible also in few our BFA/MAH modified
samples, which is assignable to a CdC stretching as
PPAMC1.3, and PPAMC1.4), the Mw grows until
reaching a maximum value equal to 386 000 g/mol.
Further, for the same amount of BFA with rising the
MAH (samples PPAMC2.0, PPAMC2.2, PPAMC2.3,

Macromolecules, Vol. 37, No. 22, 2004
Control of Polypropylene Radical Functionalization 8419
Sch em e 3. P r eva len t Rea ction s of th e Va r iou s Ma cr or a d ica ls in th e P P /P er oxid e/BF A/MAH System
described in the paper. They concluded admitting that
there is a new organic group chemically grafted to the
PP bearing a CdC double bond and suggesting the
occurrence of the disproportionation reaction reported
in Scheme 2.
On the other side 2 gives, as in the case of PE,40 the
transfer reaction with hydrogen abstraction from the
PP backbone recreating the corresponding macroradi-
cals (PP•). In this way, the instantaneous PP• concentra-
tion is lower with BFA than with MAH (Scheme 3). The
relative probability of these reactions clearly depends
on the MAH/BFA ratio, which should be optimized
considering also that BFA reacts more promptly than
MAH with PP•.
Starting from a commercially available PP (Borealis
BE 50/E1271), two series of experiments were performed
using CEFA or CFA and DCP as peroxide (Table 4). The
mixing torque graph for the two series of samples
(Figure 4) shows a different behavior if compared with
that obtained for PP treated in the melt with maleate
molecules, peroxide and the coagent BFA (e.g., Figure
1).
Indeed the mixing torque increases after the addition
of the peroxide, probably because of a significant extent
of chain extension reactions, and successively decreases.
This result can be correlated to the successive formation
of a PP cross-linked fraction which is dispersed in the
non-cross-linked matrix during the reaction course.25
A
This hypothesis provides a possible explanation for
the role of BFA in reducing the chain degradation.
However as shown by the MW decrease even in the
conditions of BFA/peroxide ) 2.5, the presence of BFA
in a molecular amount larger than the number of
primary radicals does not cancel degradation and the
grafting of BFA is practically complete. Therefore, it
appears that PPBFA• undergoes also some H-transfer
even if to a modest extent. Finally the chain termination
by free radical coupling in the presence of BFA occurs
between longer (not degradated) macroradicals, thus
maintaining a high molecular weight.
B. F u n ction a liza tion w ith F u r a n Der iva tives.
Taking into account the above observations and the in
someway competitive action of BFA and MAH, we have
attempted PP functionalization without MAH but using
furan derivatives with two functional groups, thus
increasing their functionalization capacity. The known
ethyl 2-carboxyethyl 3-(2 furanyl)propenoate (CEFA)
and the newly synthesized cyano 3-(2-furanyl)pro-
penoate (CFA) (Figure 3) were used as degradation
stabilizers and functionalizing molecules at the same
time.
considerable amount of boiling toluene insoluble poly-
mer was observed (Table 5) for samples prepared with
coagent/initiator ratio larger than 6-7, but it can be
markedly reduced with lower values of the above ratio.
The FT-IR spectra of the modified polypropylene
samples, after washing with acetone, show the presence
of the absorption peak at 1737 cm-1 in the case of CEFA
(Figure 5) and at 1743 cm-1 in the case of CFA
correlated with the carboxyethyl groups CdO stretching
of the grafted molecules.
To achieve a suitable comparison for the different
functionalized samples prepared, the functionalization
degree (FD) was determined for each sample evaluating,
like in the previously discussed case, the ratio between
the band area of the CdO functional groups stretching
and the PP area band at 1166 cm-1 (A1/A2) (Appendix
3) and by knowing the integral absorptivity (ꢀ) for the
two bands. In the case of PP functionalized with CEFA
as well as with CFA, the FD values increase, both
increasing the coagent/PP ratio and the DCP/PP ratio.
The FD depends clearly also on the coagent/initiator
ratio according to the fact that the grafting of CEFA or
CFA requires first the formation of the PP•. Thus, the
maximum number of grafted CEFA or CFA units should
be equal to the double of the peroxide molecules (initia-
tor efficiency ) 1). In general the FD for CEFA, CFA,
and BFA is larger than this number, and some H-
transfer by PP-coagent macroradicals (3) (Scheme 4)
must occur.
F igu r e 3. CEFA and CFA coagents.

8420 Coiai et al.
Macromolecules, Vol. 37, No. 22, 2004
F igu r e 4. Behavior of the mixing torque for a few samples modified with CEFA: (A) PP01; (B) PP32; (C) PP34; (D) PP33; (E)
PP35.
Ta ble 5. P olyp r op ylen e Mod ified w ith P er oxid e a n d CEF A or CF A
residue
toluene
DCP
(% mol)
CEFA
(% mol)
CFA
(% mol)
coagent/
initiator
end-torque
values (N‚m)
FDc
(% mol)
FYd
(%)
extractione
(wt %)
samplea
PP01b
PP02
PP16
PP23
PP27
PP32
PP33
PP34
PP35
PP15
PP19
PP20
PP22
PP28
PP29
PP30
PP31
10.9
1.4
4.4
5.3
3.8
3.2
3.7
3.8
5.3
5.0
4.8
3.7
3.3
3.6
4.4
4.9
4.1
82.9
1.1
26.1
45.9
58.0
6.0
1.4
5.0
5.1
22.7
46.2
49.6
30.6
2.3
16.5
4.3
1.7
0.08
0.08
0.15
0.30
0.30
0.30
0.15
0.15
0.08
0.15
0.15
0.15
0.30
0.15
0.30
0.30
1.76
4.41
4.41
0.88
1.76
0.88
1.76
22.0
29.4
14.7
2.9
5.9
5.9
11.7
27.5
14.7
22.0
29.3
14.7
7.3
0.41
0.52
0.93
0.75
0.74
0.30
0.41
0.37
0.51
0.86
0.49
1.10
0.56
0.86
0.86
23
12
21
85
42
34
23
17
23
25
11
25
51
78
39
2.20
2.20
3.30
4.40
4.40
1.10
1.10
2.20
3.7
7.3
a
b
The PP used is isotactic polypropylene (iPP) homopolymer by Borealis BE 50/E1271; MFR ) 0.24 g/10 min. Polypropylene treated
in Brabender at 180 °C without peroxide and functionalizing monomers. c Number of functional inserted groups per 100 monomeric units
of the polymer. FY ) funtionalization yield, percent of PP attached carbonyl groups with respect to their total initial number. e Weight
d
percent of polymer insoluble in boiling toluene.
If an excess of coagent is present all PP• numerically
formed are converted into the macroradical (3) and the
higher concentration of this last favors cross-linking.
Also under these conditions (3) may react with larger
probability with the coagent with formation of very short
coagent blocks. Consequently the grafting remains of
the same extent but the measured FD increases, and
the highest FDs are obtained with the maximum
amount of coagent in the feed (Table 5). As far as the
functionalization yield FY is concerned, the highest
values are observed with the lowest coagent/peroxide
ratio according to the fact that the grafting needs, as
with MAH, the formation of PP•. In any case the values
of FY are certainly larger than expected for a stoichio-
metric (1:1) reaction of PP• with CEFA or CFA, and this
implies a certain occurrence of the transfer reactions.
The lower FY appears to be flanked by a higher fraction
of insoluble (cross-linked) PP, but a univocal interpreta-
tion is at present not easy due to the evident contem-
porary occurrence of several competitive reactions.
Con clu sion s
Furan derivatives, having a ring conjugated with a
double bond substituted with electron attracting groups,
promote, thanks to high addition rate to the macroradi-
cal, the formation of a stabilized radical due to the
presence of the heteroaromatic ring. For this reason
these molecules act as good functionalization coagents
able to control intrachain hydrogen transfer reaction
responsible of PP degradation. Moreover, the coagents
molecular structure can be opportunely modulated to
improve their capability to prevent the â-scission reac-
tion and at the same time to limit the cross-linking
reaction. In this work we demonstrated that with the
use of the above furan derivatives the polypropylene
radical functionalization can be achieved with limited

Macromolecules, Vol. 37, No. 22, 2004
Control of Polypropylene Radical Functionalization 8421
F igu r e 5. FT-IR spectra of a few samples modified with CEFA: (A) PP27; (B) PP23; (C) PP16.
Sch em e 4. H-Tr a n sfer Rea ction s by P P -Coa gen t
Ma cr or a d ica l
Ack n ow led gm en t. The authors gratefully acknowl-
edge Dr. Chiara Cappelli, from INFM Pisa Section, Pisa,
Italy, for theoretical calculations.
Ap p en d ix
Ap p en d ix 1: Str u ctu r a l Ch a r a cter istics of th e
Bu tyl 3-(2-F u r a n yl)p r op en oa te (BF A) a n d Eth yl
2-Ca r boxyet h yl-3-(2-fu r a n yl)p r op en oa t e (CE F A).
The BFA coagent was synthesized as reported in the
literature25,32 and a colorless viscous oil with a bp ) 138
°C at 10-1 mmHg was obtained with a yield of 75 wt %.
reduction of the MW: by combining the above coagents
with maleic derivatives or by using specially designed
coagents with suitable functional groups.
IR main peaks (cm-1): 3015-3005 νH-Cd, 2997-2970
1
ν
H-C-, 1715 νCdO, 1640 νCdC, 1158 νC-C()O-O). H NMR
assignments: 0.9 ppm, 3 protons, triplet (CH3 butyl);
1.3 ppm, 2 protons, multiplet; 1.6 ppm 2 protons,
multiplet, (CH2-CH2 butyl); 4.2 ppm triplet 2 protons
(O-CH2 butyl); 6,2-6,3 ppm doublet 1 proton, (olefin
position 2); 6.4 ppm double doublet, 1 proton, (furan ring
position 4); 6.6 ppm doublet 1 proton, (furan ring
position 3); 7.4-7.5 doublet,1 proton, (olefin position 3);
7.5 doublet 1 proton, (furan ring position 5). The CEFA
coagent was synthesized as reported in the litera-
ture,25,32 and a colorless solid having a melting point of
49,1 °C was prepared with a yield of 43.3 wt %. IR main
peaks: 3015-3005 νH-Cd, 2997-2970 νH-C-, 1720 νCd
In the former case, the furan coagent prompt reaction
with the tertiary polypropylene macroradicals prevents
the â-scission while MAH gives grafting. By using the
BFA coagent together with maleic anhydride, good
functionalization degrees have been obtained and a good
control of the degradation reaction has been realized.
Low amounts of BFA suffice to limit the PP degradation
and are responsible for increasing MAH functionaliza-
tion degrees. The use of BFA favors the functionaliza-
tion (FD) values and also the functional monomer
conversion. The monofunctional ethyl 2-carboxyethyl
3-(2-furanyl)propenoate (CEFA) and the bifunctional
cyano 3-(2-furanyl)propenoate (CFA) coagents have been
used for PP radical functionalization reactions without
MAH. The role of the different species can be rational-
ized with a modified mechanism of the grafting process
(Scheme 4), where the role of the relative stability of
the various free radicals present in the system has been
taken into account. Further studies in due course are
devoted to the described structural analysis of the
products for a better quantitative evaluation of the
various factors including a mathematical modeling of
the process.
O, 1643 νCdC, 1479 δCH -O-CO, 934 γCH furan ring, 881
2
γCH furan ring
.
1H NMR assignments: 1.1-1.2 ppm, 6
protons, 2 near triplets (CH3 ethyl groups); 4.1-4.2 ppm,
4 protons, 2 near quartets, (O-CH2 ethyl); 6.3 ppm,
double doublet, 1 proton (furan ring position 4); 6.6 ppm
doublet, 1 proton, (furan ring position 3); 7.2 doublet,1
proton (olefin position 3); 7.3 doublet 1 proton (furan
ring position 5).
Ap p en d ix 2: Eva lu a tion of Gr a fted Gr ou p s by
F T-IR Sp ectr a Decon volu tion . By a nonlinear fitting
of experimental points, a deconvolution of the spectra

8422 Coiai et al.
Macromolecules, Vol. 37, No. 22, 2004
F igu r e 6. Deconvolution PPAMC1.1 spectrum in the range 1900-1650 cm-1
Ta ble 6. Extin ction Coefficien ts for th e CdO Str etch in g
.
Ap p en d ix 3: Eva lu a tion of F u n ction a liza tion
Degr ees by F T-IR for Sa m p les F u n ction a led w ith
CEF A or CF A. The functionalization degrees were
calculated in accordance with the following reactions:
of th e F u n ction a lizin g Molecu les a n d th e P P Ba n d a t
1166 cm -1
extinction
grafted molecules
coeff (km/mol)a)
diethyl succinate
succinic acid
succinic anhydride
BFA grafted
392
491
691
209
377
197
348
ꢀ1166 ACEFA
FDCEFA
)
)
‚
ꢀCEFA
A1166
CEFA grafted
ꢀ1166 ACFA
CFA grafted
FDCFA
‚
PP band at 1166 cm-1
ꢀCFA
A1166
a
Calculated at the B3LYP/6-31G* level.
The extinction coefficient for all the bands used has been
calculated in accord to the computational procedure
reported in the experimental part. In Table 5, the values
obtained are showed.
has been performed obtaining the diagnostic band
(Figure 6).
Comparing the above absorption bands with the
typical absorption band of PP at 1166 cm-1 (stretching
of C-CH3 groups on PP chains) and using the integral
absorptivity (ꢀ) for all the bands used allowed the FD
values related to BFA and MAH to be calculated
according to the following equations:
Refer en ces a n d Notes
(1) Xanthos, M. Reactive Extrusion; Oxford University Press:
New York, 1992.
ꢀ1166 ABFA
(2) Moad, G. Prog. Polym. Sci. 1999, 24, 81-142.
(3) J ois, Y. H. R.; Harrison, J . B. J . Macromol. Sci. Rev.
Macromol. Chem. Phys. 1996, C36 (3), 433-455.
(4) Gaylord, N. G.; Mishra. M. K. J . Polym. Sci., Polym. Lett.
Ed. 1983, 21, 23-30.
(5) Ruggeri, G.; Aglietto, M.; Petragnani, A.; Ciardelli, F. Eur.
Polym. J . 1983, 19, 863-866.
FDBFA
)
‚
ꢀBFA
A1166
A
∑
SAH,i
ꢀ1166 ASA
ꢀ1166
i
FDSAH,SA
)
‚
+
‚
(6) Sabu, T.; Groeninckx, G. Polymer 1999, 40, 5799-5819.
(7) Gloor, P. E.; Tang, Y.; Kostanska, A. E.; Hamielec, A. E.
Polymer 1994, 35, 1012-1030.
ꢀSA
ꢀSAH
A1166
A1166
(8) Kwei, T. K.; Pearce, E. M.; Schlecht, M. F.; Cheung, W. J .
Appl. Polym. Sci. 1991, 42, 1939-1941.
The extinction coefficient for all the bands used has been
calculated by following the computational procedure
reported in the experimental part and described in a
recent work of our group.41 In Table 6, the values
obtained are reported.
The extinction coefficient for the PP band at 1166
cm-1 was calculated by considering a linear correlation,
found in a previous work,22 between the FD, calculated
by 1H NMR, for PP samples functionalized with diethyl
maleate and the peak areas ratio (A1/A2), where A1 is
the band at 1736 cm-1 (νCO dithyl succinate groups
grafted onto the PP chains) and A2 at 1166 cm-1. By
knowing the theoretical value of the ꢀDES, obtained by
the computational procedure, ꢀ1166 cm-1 was evaluated.
(9) Gaylord, N. G.; Metha, M.; Metha, R. J . Appl. Polym. Sci.
1987, 33, 2549-2558.
(10) Chodak, I.; Lazar, M. J . Appl. Polym. Sci. 1986, 32, 5431-
5437.
(11) Lazar, M.; Hrcˇkova´, L’.; Borsig, E.; Marcincˇin, A.; Reichelt,
N.; Ratzsch, M. J . Appl. Polym. Sci. 2000, 78, 886-893.
(12) De Roover, B.; Sclavons, M.; Carlier, V.; Devaux, J .; Legras,
R.; Momtaz, A. J . Polym. Sci., Polym. Chem. 1995, 33, 829-
842.
(13) De Roover, B.; Devaux, J .; Legras, R. J . Polym. Sci., Polym.
Chem. 1996, 34, 1195-1202.
(14) Bettini, S. H. P.; Agnelli, J . A. M. J . Appl. Polym. Sci. 1999,
74, 247-255.
(15) Garcia-Martinez, J . M.; Laguna, O.; Collar, E. P. J . Appl.
Polym. Sci. 1997, 65, 1333-1347.

Macromolecules, Vol. 37, No. 22, 2004
Control of Polypropylene Radical Functionalization 8423
(16) Garcia-Martinez, J . M.; Laguna, O.; Collar, E. P. J . Appl.
Polym. Sci. 1998, 68, 483-495.
(34) (a) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38,
3098-3100. (b) Becke, A. D. J . Chem. Phys. 1993, 98, 5648-
5652.
(35) Hehre, W. J .; Radom, L.; Schleyer, P. V.; Pople, J . Ab initio
Molecular Orbital Theory; Wiley-Interscience: New York,
1986.
(17) Garcia-Martinez, J . M.; Laguna, O.; Areso, S.; Collar, E. P.
J . Appl. Polym. Sci. 1998, 70, 689-696.
(18) Sun, Y. J .; Hu, G. H.; Lambla, M. Makromol. Chem. 1995,
229, 1-13.
(19) Sun, Y. J .; Hu, G. H.; Lambla, M. J . Appl. Polym. Sci. 1995,
57, 1043-1054.
(36) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgom-
ery, J . A.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.;
Millam, J . M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.;
Farkas, O.; Tomasi, J .; Barone, V.; Cossi, M.; Cammi, R.;
Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski,
J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Salvador, P.; Dannenberg, J . J .; Malick, D. K.; Rabuck, A.
D.; Raghavachari, K.; Foresman, J . B.; Cioslowski, J .; Ortiz,
J . V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox,
D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Challacombe, M.; Gill, P. M. W.; J ohnson, B.; Chen, W.;
Wong, M. W.; Andres, J . L.; Gonzalez, C.; Head-Gordon, M.;
Replogle, E. S.; Pople, J . A. Gaussian 98 (Revision A.10);
Gaussian, Inc.: Pittsburgh, PA, 2001.
(37) Cappelli, C.; Corni, S.; Cammi, R.; Mennucci, B.; Tomasi, J .
J . Chem. Phys. 2000, 113, 11270-11279.
(38) Passaglia, E.; Corsi, L.; Aglietto, M.; Ciardelli, F.; Michelotti,
M.; Suffredini, G. J . Appl. Polym. Sci. 2003, 87, 14-23.
(39) Brandrup, J .; Immergut, E. H. Polymer Handbook, 3rd ed.;
Wiley-Interscience: New York, 1989.
(40) Ciardelli, F.; Aglietto, M.; Passaglia, E.; Ruggeri, G. Macro-
mol. Symp. 1998, 129, 79-88.
(41) Bertoldo, M.; Bronco, S.; Cappelli, C.; Gragnoli, T.; Andreotti,
L. J . Phys. Chem. B 2003, 107, 11880-11888.
(20) Hu, G. H.; Sun, J .; Lambla, M. J . Appl. Polym. Sci. 1996, 61,
1039-1047.
(21) Passaglia, E.; Marrucci, M.; Ruggeri, G.; Aglietto, M. Gazz.
Chim. Ital. 1997, 127, 91-95.
(22) Passaglia, E.; Coiai, S.; Aglietto, M.; Ruggeri, G.; Ruberta`,
M.; Ciardelli, F. Macromol. Symp. 2003, 198, 147-159.
(23) Fritz, H. G.; Cai, Q.; Boelz, U. Kunststoffe 1993, 86 (6), 439-
444.
(24) Cartier, H.; Hu, G. H. J . Appl. Polym. Sci. 1998, 36, 1053-
1063.
(25) Romani, F.; Corrieri, R.; Braga, V.; Ciardelli, F. Polymer 2001,
43, 1115-1131.
(26) Chodak, I.; Fabianova, K.; Borsig, E.; Lazar, M. Makromol.
Chem. 1978, 69, 107-115.
(27) Robinson, A. E.; Marra, J . V.; Amberg, L. O. Ind. Eng. Chem.
Prod. Res. Develop. 1962, 1, 78-82.
(28) Lenas, L. P. Ind. Eng. Chem. Prod. Res. Develop. 1963, 2,
202-208.
(29) Bartorelli, G.; Falcone, L.; Pregaglia, G. (Montecatini Edison,
S.p.A.); U.S. Patent 3 301 387, 1967.
(30) Braga, V.; Ghisellini, R. Ausimont, S. r.l. Brev. B-742-03,
1990.
(31) Wong, B.; Baker, W. E. Polymer 1987, 38, 2781-2789.
(32) Romani, F.; Braga, V.; European Patent EP 1 216 999 A1,
2002.
(33) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density
Functional Theory, 2nd ed.; Wiley-VCH: Berlin, 2001.
MA0400315