Polymer Degradation and Stability 94 (2009) 443–449

Contents lists available at ScienceDirect

Polymer Degradation and Stability

journal homepage: www.elsevier.com/locate/polydegstab

Aging behavior and thermal degradation of fluoroelastomer reactive blends

with poly-phenol hydroxy EPDM
Yaming Wang, Lan Liu*, Yuanfang Luo, Demin Jia
College of Materials Science and Engineering, South China University of Technology, Guangzhou 510640, People’s Republic of China

a r t i c l e i n f o a b s t r a c t

Article history: In this paper, the aging behavior of the reactive blends of fluoroelastomer (FPM) with poly-phenol
Received 5 August 2008 hydroxy ethylene propylene diene monomer rubber (PHEPDM) in hot air was firstly investigated. The
Received in revised form aging mechanism was analyzed by the swelling experiment, attenuated total-reflectance Fourier-
27 October 2008
transform infrared (FTIR-ATR) spectroscopy and X-ray photoelectron spectroscopy (XPS). The results
Accepted 3 November 2008
showed that the aging process increased the crosslinking density and the content of double bond. The O/
Available online 19 November 2008
F or O/C ratios increased and then decreased during aging because of the oxidation reaction of molecular
chain and the surface migration of fluoro group. Secondly, thermogravimetric analysis (TGA) was used to
Keywords:
Fluoroelastomer study the thermal degradation behavior of the reactive blends. The apparent degradation activation
PHEPDM energy (E) of FPM/PHEPDM reactive blends was calculated by the Kissinger and Coats–Redfern methods,
Aging respectively. The results showed that the FPM/PHEPDM reactive blends had higher thermal degradation
Thermal degradation temperature but lower E than FPM. Both the thermal degradation process of FPM/PHEPDM reactive
blends and FPM were determined by nucleation and growth mechanism (Am). The general mechanism
function was [ln(1  a)]1/m. The optimum value of m was between 1/3 and 1/2 for FPM/PHEPDM
reactive blends, but 1/2 for FPM. From the results above, it was deduced that the special structure of
PHEPDM made itself surrounded by fluoroelastomer and protected from hot-air aging and thermal
degradation.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction ozone [16–19]. Blending EPDM into fluoroelastomer can be

a potential measure to prepare the materials with better overall
Fluoroelastomers are widely used in many industrial applica- properties. However, the high incompatibility and non-co-vulca-
tions due to their excellent resistance to heat, oil and solvent [1–5]. nization between FPM and EPDM make it difficult to obtain
The increasing use of such polymers in automobile, aerospace, a blending material with better overall properties [20,21]. In order
offshore, and energy-related industries imposes them stringent to improve the compatibility and solve the co-vulcanization
product performance standards under inclement temperatures and problem between FPM and EPDM, a PHEPDM (as shown in
in hostile chemical environments [6–11]. Recently, their use in cold Scheme 1) with pliant molecular chain and phenol hydroxyl groups
districts is especially increasing. Although the conservation of is prepared.
elasticity in low-temperature environment is necessary for fluo- Fluorocarbon elastomer (vinylidenefluoride-co-hexafluoro
roelastomer material, one disadvantage of fluoroelastomer is its propylene) can be vulcanized by bisnucleophiles such as bisph-
lack of low-temperature resistance [12,13]. enols [11]. According to the current acceptable mechanism, the
It is reported that blending fluoroelastomers with rubbers PHEPDM with phenol hydroxyl groups can act as crosslinking
having low glass transition temperature (Tg) may enhance the low- agent for FPM. Thus the molecules of FPM and PHEPDM are
temperature properties of fluoroelastomer [14,15]. Ethylene connected through chemical bonds and some specific properties
propylene diene monomer rubber (EPDM) is the material that has will be obtained for the vulcanized materials because of the
excellent performances in low-temperature flexibility, thermal particular structure of PHEPDM. Researches on the vulcanization
stability, weatherability and aging resistance to oxidation and procedure, low-temperature resistance and mechanical proper-
ties of the FPM/PHEPDM reactive blends have been reported
elsewhere by our research group. In this paper, we attempt to
study the aging behavior and thermal degradation behavior of the
* Corresponding author. Tel.: þ86 020 87110272; fax: þ86 020 87110273. FPM/PHEPDM reactive blends and find out the corresponding
E-mail address: psliulan@scut.edu.cn (L. Liu). mechanisms.

0141-3910/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymdegradstab.2008.11.007
444 Y. Wang et al. / Polymer Degradation and Stability 94 (2009) 443–449

Scheme 1. The structure of PHEPDM.

Scheme 3. Chemical bonds between PHEPDM and ultrafine silicone dioxide.

2. Experimental
2.3.3. Swelling tests
2.1. Materials
The samples were weighed accurately before immersing into
butanone and kept immersing into butanone at 30  C for 48 h. After
Fluoroelastomer (copolymer of vinylidene fluoride and hexa-
the swelling, the sample was taken out from the solvent and
fluoropropylene), FPM-2603, (66% F; Mooney viscosity, ML (1 þ10)
weighed after removing surface solvent by filter paper. Swelling
at 121  C: 80; density: 1.81–1.84 g/cm3) was obtained from Chen-
ratio was calculated using the following equation:
guang Chemical Co. Ltd., Sichuan, China. PHEPDM was self-
prepared [22]. The bisphenol AF and benzyltriphenylphosphonium W2  W1
chloride (BPP) were supplied by Dupont Dow, USA. The structures q ¼  100%
W1
of PHEPDM and bisphenol AF are shown in Schemes 1 and 2. The
Ca(OH)2, MgO and ultrafine silicone dioxide were supplied by where q is the ratio of swollen weight to original weight; W1 and
Guangzhou Rubber Institute, China. W2 are the weights of the sample before and after swelling,
respectively.
2.2. Preparation of the blends
2.3.4. X-ray photoelectron spectroscopic (XPS) analysis
The rubber blends were prepared in a 4160  330 open two-roll XPS spectra of the FPM/PHEPDM reactive blends before and
mill. The rollers ran at a speed ratio of 1 (front roller): 1.22 (back after aging were recorded by a Kratos Axis UltraDLD XPS instrument
roller). The raw materials were charged in the following order: with an Aluminum (mono) Ka source (1486.6 eV). The Aluminum
100 g of fluoroelastomer, 6 g of Ca(OH)2, 3 g of MgO, 14.5 g of Ka source was operated at 15 kV and 10 mA at a take-off angle of
ultrafine silicone dioxide, 0.4 g of BPP, and variable PHEPDM of 5 g, 90 , resulting in a maximum probe depth of around 5 nm. The
7.5 g and 10 g. The PHEPDM was substituted by bisphenol AF for the samples were mounted onto a holder with double-sided conduc-
control FPM. The blends were vulcanized in a hydraulic press at tive adhesive tape and placed in a vacuum of 109 torr. The
a pressure of 5 MPa for 10 min (10 min was in excess of the analyzed sample area was approximately 0.7  0.3 mm. For all the
optimum cure time about 3 min, which was determined from the samples, a low-resolution survey run (0–1100 eV, pass ener-
oscillating disc rheometer). The molded samples were post-cured gy ¼ 160 eV) and a high-resolution survey (pass energy ¼ 40 eV)
at 200  C for 4 h. were performed at spectral regions relating to carbon, oxygen and
fluorine.
2.3. Testing
2.3.5. Thermal analysis
2.3.1. FTIR-ATR spectra Thermogravimetric analysis (TGA) was carried out in a NETZSCH
In order to study the aging mechanism, the FTIR-ATR spectra of 209 (NETZSCH Instrument, Germany) thermogravimetric analyzer
the samples before and after aging were measured on a Nicolet over a temperature range from room temperature to 900  C at
Fourier-transform infrared spectrophotometer from 4000 cm1 to a heating rate of 5, 10, 20 and 30  C/min, respectively. Nitrogen flow
550 cm1 in 4 cm1 resolution. For each sample, three scans were (80 ml/min) was employed in order to remove all corrosive gases
taken to identify the peaks. involved in the degraded materials and to avoid thermo-oxidative
degradation.
2.3.2. Hot-air aging tests For non-isothermal degradation, the apparent activation energy
Hot-air aging was performed in an air-circulating oven at 200  C (E) of degradation could be calculated by using the Kissinger
for 48 h, 96 h and 144 h, respectively. method [23] and Coats–Redfern method [24] through equations (1)
and (2), respectively.

!
bi Ak R Ek 1
ln 2
¼ ln  ; i ¼ 1; 2; ,,,4 (1)
Tpi Ek R Tpi

gðaÞ AR Ea
ln 2 ¼ ln  (2)
Scheme 2. The structure of bisphenol AF. T bR RT
 Y. Wang et al. / Polymer Degradation and Stability 94 (2009) 443–449 445

In equation (1), b ¼ dT/dt is the heating rate, Tpi is the temper-

ature at which there is a maximum weight loss, Ak is a pre-
exponential factor, Ek is the activation energy and R is the universal
gas constant for equation (1). A plot of ln(b/T2pi) versus 1/Tpi gives
the activation energy without a specific assumption on the degra-
dation dependent function. This enables a single value of Ek to be
determined by a model free way.
In equation (2), a is the mass conversion rate, T is the temper-
ature, b ¼ dT/dt is the heating rate, A is the pre-exponential factor,
Ea is the activation energy and R is the universal gas constant. Given
a thermogravimetic function g(a), a plot of ln g(a)/T2 versus 1/T for
different a also gives the activation energy.
Several thermogravimetic functions were used in this article, as
shown in Table 1. The thermogravimetic activation energy Ea was
calculated based on different g(a). The thermal degradation
mechanism was deduced by the comparison of Ea to Ek.

3. Results and discussion

3.1. Hot-air aging behavior of FPM/PHEPDM reactive blends

3.1.1. FTIR-ATR analysis

In the aging procedure, it is observed that the colour of both FPM
and FPM/PHEPDM reactive blends change from yellow to brown
due to thermo-oxidative reactions. In order to study the aging
mechanism of such reactions, the representative samples are
analyzed by using FTIR-ATR spectroscopy. Fig. 1a and b shows the
FTIR-ATR spectra of FPM/PHEPDM (10%) reactive blend and FPM
aged at 200  C, respectively. The assignments of various peaks for
FPM/PHEPDM (10%) reactive blend and FPM are given in Fig. 1a and
b. The FTIR-ATR spectra show the significant variations of the
absorbance peaks after hot-air aging, the absorbance at 1605 cm1
corresponding to carbon double bond stretching, the absorbance at
1750 cm1 corresponding to carbonyl group and the absorbance at
2870 cm1 and 2970 cm1 corresponding to the –CH stretching of
methylene. It is observed that the absorbance peak for carbon
double bond increases because of the dehydrofluorination which
forms C]C, and the absorbance peak of methylene decreases
because the dehydrofluorination reaction and surface migration of
fluoro group reduces the content of CH2. A new peak about the
C]O stretching which is induced by thermal oxidation of C]C
appears at 1750 cm1 after hot-air aging for 48 h and 144 h. The Fig. 1. FTIR spectra of FPM/PHEPDM (10%) reactive blends (a) and FPM (b) before and
suggested mechanism for dehydrofluorination reaction is similar to after hot-air aging.
the vulcanization procedure of FPM, just like that reported by Kader
[16]. Compared with FPM, the absorbance peak of methylene is not
enhanced for FPM/PHEPDM reactive blends, the reason for which is
3.1.2. XPS analysis
related to the structure and crosslinking agent role of PHEPDM, it
X-ray photoelectron spectroscopy is one of the most powerful
will be explained in detailed in the next paragraph.
surface analytical techniques that probes the composition and
Table 1
structure of sample surface to a depth of about 5 nm, thereby allows
The mechanisms of solid-state thermal degradation reaction and corresponding for the elucidation of exterior chemical degradation phenomena
thermal degradation functions g(a). [6,10,26].
Mode g(a) Thermal degradation process
Table 2 shows the atomic ratios (C/F, O/F and O/C) of FPM/
PHEPDM (10%) reactive blend which has undergone a dramatic
Sigmoidal curves
A1/3 [ln(1  a)]3 Nucleation and growth (Avrami equation) chemical change after different hot-air aging periods at 200  C. The
A1/2 [ln(1  a)]2 Nucleation and growth (Avrami equation) O/C ratio increases from 0.160 up to 0.226 after aging for 48 h due to
A2/3 [ln(1  a)]3/2 Nucleation and growth (Avrami equation) the thermal oxidation reaction, and then reduced to 0.217 after
A2 [ln(1  a)]1/2 Nucleation and growth (Avrami equation) s aging for 144 h because of the surface migration of fluoro groups.
form a-t figure
Though the content of fluorin is reduced by the elimination of
Deceleration curves hydrogen fluoride (HF) [25] at high temperature, the same
R1 a Phase interface controlled reaction (linear
phenomenon is also detected for the O/F ratio just for the same
dimension diffusion)
R2 1(1  a) 1/2
Phase interface controlled reaction (contraction reason. The C/F ratio decreases dramatically after aging for 48 h and
area) 144 h compared to the unaged sample due to the surface migration
D1 a2 One-dimensional diffusion of fluoro groups. Just as Table 2 shows, the atomic ratios (C/F, O/F
D2 aþ(1  a)ln(1  a) Two-dimensional diffusion and O/C) of FPM have the same phenomena as FPM/PHEPDM (10%)
D3 [1(1  a)1/3]2 Three-dimensional diffusion (Jander equation)
reactive blend after hot-air aging, the difference is the contents of
446 Y. Wang et al. / Polymer Degradation and Stability 94 (2009) 443–449

Table 2 reason why the content of fluorin is high but the content of oxygen
Ratios of atomic percentages of C, F and O of the surface of FPM/PHEPDM reaction is low for FPM/PHEPDM (10%) reactive blend.
blends and FPM before and after aging at 200  C obtained from XPS.
In order to gain more detailed information about the chemical
FPM/PHEPDM (10%) reaction blends (h) C/F O/F O/C changes occurring in FPM/PHEPDM reactive blends and FPM during
0 1.560 0.250 0.160 hot-air aging, high-resolution XPS analysis was performed. Fig. 3a
48 1.519 0.344 0.226 and b shows the C1s spectra of the reactive blends and FPM after
168 1.382 0.299 0.217
hot-air aging for 144 h at 200  C. The C1s spectra for the FPM/
FPM PHEPDM (10%) reactive blend and FPM in both Fig. 3a and b can be
0 3.68 1.55 0.42 fitted to five components corresponding to the C–C/C–H (284.6 eV),
48 9.27 3.97 0.43
168 2.95 1.44 0.48
C–CF2 (286.9 eV), CF (289.9 eV), CF2 (291.6 eV) and CF3 (293.9 eV),
the results are consistent with those reported by Mitra et al. [10].
From the position point of view, the C1s spectra (which are not
depicted in the article) for the FPM/PHEPDM (10%) reactive blend
oxygen and fluorin. As Fig. 2a and b shows, the content of oxygen is
and FPM before and after aging for 48 h are the same with that
much higher and the content of fluorin is much lower for FPM. This
aging for 144 h, the change is the group content. As Fig. 2a and
result can be explained from the structure of PHEPDM. The Si–OCH3
b shows, the most significant changes are the evident variation in
groups of PHEPDM can form hydrogen bonds with the Si–OH of
the amount of the CF, CF2, and CF3 groups after hot-air aging for
ultrafine silicon dioxide, thus much filler is attracted and dispersed
48 h and 144 h, respectively. The amount of these groups decrease
in PHEPDM. On the other hand, the crosslinking agent role of
drastically after aging for 48 h, associated with dehydrofluorination
PHEPDM makes it surrounded by fluoroelastomer molecule. As
[10], but increase dramatically after aging for 144 h owing to the
Scheme 4 shows, the fluoroelastomer and PHEPDM are connected
surface migration of fluoro groups. From Fig. 2a and b, we can see
through chemical bond, thus the crosslinking site of –CH2–CF2– is
that the FPM/PHEPDM (10%) reactive blend contains much more
drawn to the inside and the group of CF3 is left outside. That is the
fluorine on the surface than FPM, the reason for which has been
explained in the previous paragraph. For FPM/PHEPDM reactive
blends, more CF3 groups transfer to the surface because of the
crosslinking reaction during hot-air aging.

3.1.3. Swelling studies

The equilibrium-swelling ratio represents the crosslinking
density for the vulcanized rubbers. It is known that rubbers with
higher crosslinking density have smaller equilibrium-swelling
ratio. Fig. 4 shows the equilibrium-swelling ratios of FPM/PHEPDM
reactive blends. It is clear from Fig. 4 that the FPM/PHEPDM reactive
blends have higher equilibrium-swelling ratio than FPM before
aging. The equilibrium-swelling ratio decreases as the content of
PHEPDM increases. An possible explanation is that the fluoroelas-
tomer molecular chains are connected through pliant molecular
chains of PHEPDM in the FPM/PHEPDM reactive blends. The long
pliant PHEPDM chains facilitate the movement of fluoroelastomer
molecules and induce the solvent to swell the blended fluoroelas-
tomer. Especially when the content of PHEPDM is low, the cross-
linking density is low and the equilibrium-swelling ratio is
increased.
After aging, the equilibrium-swelling ratios of FPM/PHEPDM
reactive blends decrease much more than that of FPM, and the FPM/
PHEPDM (10%) reactive blend even has smaller equilibrium-
swelling ratio than FPM. This phenomenon is related to the
vulcanization procedure of rubbers. The bisphenol AF is a kind of
small molecular material, which diffuses and reacts quickly within
fluoroelastomer matrix at the vulcanization temperature, 170  C
and 200  C. But the phenol hydroxy groups of PHEPDM diffuse and
react slowly because of the high molecular weight and steric
hindrance of PHEPDM itself. There are some unreacted phenol
hydroxy groups embedded in the PHEPDM matrix during
mechanical mix. The embedded phenol hydroxy groups are set free
and react with FPM in the aging process because of the chain
movement of PHEPDM. Thus the FPM/PHEPDM reactive blends are
vulcanized for the second time and the equilibrium-swelling ratio
is reduced much more after aging than that of FPM.

3.2. Thermal stability of FPM/PHEPDM reactive blends

The thermal degradation of rubber results in the scission of

crosslinking network and the chain cleavage, and in the decrease of
Fig. 2. Content of % O (a) and % F (b) from C1s curve fitting date with hot-air aging time the crosslinking density and molecular weight. This reduces the
from XPS analysis. mechanical strength and the application value of the rubbers. It has
 Y. Wang et al. / Polymer Degradation and Stability 94 (2009) 443–449 447

Scheme 4. The procedure of crosslinking reaction.

been confirmed that the polymers like fluorocarbon rubbers from other thermal kinetic method, but cannot affirm the thermal
[25,27] degrade in several steps including chain cleavage, end- degradation mechanism itself. The Coats–Redfern method is
initiated depolymerization and random scission. The degradation correlative to the thermogravimetic function g(a). Given a ther-
of FPM/PHEPDM reactive blends shows some interesting steps mogravimetic function g(a), the apparent thermal degradation
which are different from that of the conventional rubbers. It may activation energy E and degradation mechanism can be calculated
occur in individual or blended form. To determine the degradation by the Coats–Redfern method. Different thermogravimetric func-
behavior of FPM/PHEPDM reactive blends and find its distinction tion g(a) will give different E. To approve E achieved from the
from FPM, two kinetic methods such as the Kissinger method and Coats–Redfern method, other kinetic methods can be used. In this
Coats–Redfern method are introduced. For the Coats–Redfern article, we use the Kissinger method to approve E calculated by the
method, nine thermal degradation functions are adopted. Coats–Redfern method, and combined these two methods to
From Fig. 5, it is obvious that the onset degradation temperature deduce the degradation kinetic mechanism. The results are listed in
(T0), 5% weight loss temperature, 10% weight loss temperature, 20% Tables 3 and 4. From Table 3, it is observed that the FPM/PHEPDM
weight loss temperature and DTG peak temperature (Tmax) of the reactive blends have lower apparent activation energy E than FPM.
FPM/PHEPDM reactive blends increase slightly compared to those It means that the FPM has better thermal stability than the FPM/
of FPM. The apparent activation energy E for FPM and FPM/ PHEPDM reactive blends, which is inconsistent with the results
PHEPDM reactive blends is calculated through the Kissinger shown in Fig. 5. To further study the thermal degradation, the
method and Coats–Redfern method. The results are listed in Tables thermal degradation kinetic mechanism is deduced by comparing
3 and 4, respectively. The Kissinger method is a differential method, the values of Table 4 to that of Table 3. The compared results show
and E is deduced from a plot of the temperature at maximum rate of that both the thermal degradation process of FPM/PHEPDM reac-
degradation versus the heating rate. And the temperature at tive blends and FPM are determined by the nucleation and growth
maximum rate of degradation is determined by the differential mechanism, which accords with the sigmoidal curves (Am, just as
quotient thermal analysis plots. This method can avoid the error shown in Table 1). The general mechanism function is [ln(1  a)]1/m,
induced by different thermal degradation mechanism hypotheses and the value of m can be 1/3, 1/2, 2/3, 2, etc. It is observed that
and are usually used to verify apparent activation energy obtained the optimum value of m is 1/2 for FPM and between 1/3 and 1/2 for
448 Y. Wang et al. / Polymer Degradation and Stability 94 (2009) 443–449

Fig. 5. TG and DTG plots of FPM/PHEPDM reactive blends and FPM.

Table 3
Apparent activation energy E for FPM and FPM/PHEPDM reactive blends by Kis-
singer method.

Ek (kJ/mol) r
FPM 251.74 0.9995
FPM/PHEPDM (5%) 244.98 0.9992
FPM/PHEPDM (7.5%) 219.60 0.9958

all FPM/PHEPDM reactive blends under research. Equation (2)

shows that smaller value of m gives bigger apparent activation
energy for a certain thermal degradation procedure, which means
that smaller value of m corresponds to more difficult thermal
degradation process. Obviously, the FPM/PHEPDM reactive blends
will have higher thermal stability than FPM if they have the same
content ratio of fluorin. PHEPDM plays an important role in the
thermal degradation process of FPM/PHEPDM reactive blends.
Multi-phenol hydroxy groups make the network degradation
Fig. 3. XPS C1S spectra of (a) FPM/PHEPDM (10%) reactive blend and (b) FPM after
difficult, but the repolymerization or recrosslinking reaction easy at
aging for 144 h.
high temperature. Meanwhile, the reaction of hydroxyl and silica-
methoxyl groups between the ultrafine silicone dioxide and
PHEPDM also improves the thermal stability of FPM/PHEPDM
reactive blends. The reaction mechanisms are shown in Schemes 3
and 4, respectively. Since the backbone of PHEPDM is free of fluo-
rine, the thermal stability of PHEPDM is not as excellent as that of
fluoroelastomer. That is the reason why FPM/PHEPDM reactive
blends have higher thermal degradation temperature but lower
apparent activation energy than FPM.

Table 4
Apparent activation energy E for FPM and FPM/PHEPDM reactive blends by Coats–
Redfern method.

FPM FPM/PHEPDM (5%) FPM/PHEPDM (7.5%)

E r E r E r
A1/3 381.88 0.9963 265.46 0.9950 271.31 0.9961
A1/2 250.75 0.9962 173.13 0.9948 177.00 0.9959
A2/3 185.18 0.9961 133.02 0.9947 129.84 0.9957
A2 54.04 0.9947 34.65 0.9915 35.54 0.9935
R1 113.99 0.9965 82.29 0.9948 80.46 0.9956
R2 116.78 0.9962 79.55 0.9942 81.57 0.9954
D1 239.50 0.9969 168.12 0.9952 172.54 0.9962
D2 243.19 0.9967 169.76 0.9951 174.01 0.9961
D3 246.96 0.9964 171.44 0.9949 175.50 0.9960
Fig. 4. Swelling ratio of FPM/PHEPDM reactive blends before and after aging.
 Y. Wang et al. / Polymer Degradation and Stability 94 (2009) 443–449 449

4. Conclusions surrounded by fluoroelastomer. The outside fluoro groups

protect the inner PHEPDM from hot-air aging and thermal
C FTIR-ATR and XPS show that both the FPM/PHEPDM reactive degradation. This is the reason why FPM/PHEPDM reactive
blends and FPM undergo thermal oxidation and dehydro- blends have a good hot-air aging resistance and thermal
fluorination reaction during aging. The carbonyl group and stability as FPM.
carbon double band are formed, and the fluoro group shifts to
the surface at the aging temperature of 200  C. On the surface,
the FPM/PHEPDM reactive blends contain more fluorin but Acknowledgements
less oxygen than FPM because of the special structure and
crosslinking agent role of PHEPDM. The PHEPDM is sur- The authors want to express their sincere appreciation to
rounded by fluoroelastomer molecule in the reactive blends. National natural science foundation of China (contract grant
C FPM/PHEPDM reactive blends have higher equilibrium- number: 50873036) for the financial support in undertaking the
swelling ratio than FPM before hot-air aging. The possible project.
reason is that the fluoroelastomer molecular chains are con-
nected through pliant molecular chains of PHEPDM in the References
FPM/PHEPDM reactive blends. The long pliant PHEPDM
[1] Kader MA, Bhowmick AK. J Appl Polym Sci 2003;89:1442–52.
chains facilitate the movement of fluoroelastomer molecules [2] Boutevin B, Caporiccio G, Pietrasanta FG, Ratsimihety A. J Fluorine Chem
and induce the solvent to swell the blended fluoroelastomer. 2003;124:131–8.
C After aging, the equilibrium-swelling ratios of FPM/PHEPDM [3] Maiti M, Bhowmick A. J Polym Sci Part B Polym Phys 2006;44:162–76.
[4] Maiti M, Bhowmick AK. Polym Eng Sci 2007;47:1777–87.
reactive blends decrease much more than that of FPM. The
[5] Wang P, Schneider NS, Sung NH. J Appl Polym Sci 1999;71:1525–35.
reason is that the embedded phenol hydroxy groups of [6] Mitra S, Ghanbari-Siahkalia A, Kingshott P, Hvilsted S, Almdal K. J Polym Sci
PHEPDM are set free and react with FPM in the aging process Part A Polym Chem 2004;42:6216–29.
and the FPM/PHEPDM reactive blends are vulcanized for the [7] De Angelis MG, Sarti GC, Sanguineti A, Maccone P. J Polym Sci Part B Polym
Phys 2004;42:1987–2006.
second time. [8] Valsecchi R, Vigano M, Levi M, Turri S. J Appl Polym Sci 2006;102:4484–7.
C Multi-phenol hydroxy groups of PHEPDM make the network [9] Ghosh A, De SK. Rubber Chem Technol 2004;77:856–72.
degradation difficult, but the repolymerization or recros- [10] Mitra S, Ghanbari-Siahkalia A, Kingshotta P, Almdal K, Rehmeier HK,
Christensen AG. Polym Degrad Stab 2004;83:195–206.
slinking reaction easy at high temperature. Meanwhile, the [11] Ameduri B, Boutevin B, Kostov G. Prog Polym Sci 2001;26:105–87.
reaction of hydroxyl and silica-methoxyl groups between [12] Hirano K, Suzuki K, Nakano K, Tosaka M. J Appl Polym Sci 2005;95:149–56.
the ultrafine silicone dioxide and PHEPDM also improve the [13] Ghosh A, Naskar AK, Khastgir D, De SK. Polymer 2001;42:9849–53.
[14] Gomaa E. J Appl Polym Sci 2007;105:2564–70.
thermal stability of FPM/PHEPDM reactive blends. Since the [15] Chawdhury R, Banerji MS, Shivakumar K. J Appl Polym Sci 2006;100:1008–12.
backbone of PHEPDM is free of fluorine, the thermal stability [16] Sirqueira AS, Soares BG. Macromol Mater Eng 2007;292:62–9.
of PHEPDM is not as excellent as that of fluoroelastomer. That [17] Wl Wu, Chen DJ. Polym Compos 2006;27:621–6.
[18] Deniz V, Karaagac B, Ceyhan N. J Appl Polym Sci 2007;103:557–63.
is the reason why FPM/PHEPDM reactive blends have higher [19] Mitra S, Ghanbari-Siahkali A, Kingshott P, Hvilsted S, Almdal K. Mater Chem
thermal degradation temperature but lower apparent acti- Phys 2006;98:248–55.
vation energy than FPM. The thermal degradation process is [20] Kanu RC, Shaw MT. Polym Eng Sci 1982;22:507–11.
[21] Mathew AP, Thomas S. Polym Degrad Stab 2001;72:423–39.
determined by the nucleation and growth mechanism (Am),
[22] Luo YF, Wang YM, Liu L, Jia DM. China patent, 200810027894.8.
and the general function is [ln(1  a)]1/m. The optimum [23] Kissinger HE. Anal Chem 1957;29:1702–6.
value of m is between 1/3 and 1/2 for FPM/PHEPDM reactive [24] Coats AW, Redfern JP. Nature 1964;201:68–9.
blends but 1/2 for FPM. [25] Kader MA, Bhowmick AK. Polym Degrad Stab 2003;79:283–95.
[26] Mohai M. Surf Interface Anal 2006;38:640–3.
C From the results analysis of FTIR-ATR, XPS and TGA, it is [27] Banik I, Bhowmick AK, Raghavan SV, Majali AB, Tikku VK. Polym Degrad Stab
concluded that the special structure of PHEPDM makes itself 1999;63:413–21.