EUROPEAN

POLYMER
European Polymer Journal 43 (2007) 3113–3127
JOURNAL
www.elsevier.com/locate/europolj

Macromolecular Nanotechnology

The amount of immobilized polymer in PMMA SiO2

nanocomposites determined from calorimetric data
a,b
Albert Sargsyan , Anahit Tonoyan b, Sevan Davtyan b, Christoph Schick a,*

a
University of Rostock, Institute of Physics, Universitätsplatz 3, 18051 Rostock, Germany
b
State Engineering University of Armenia, Department of Chemistry, 105 Teryan Str., 375009 Yerevan, Armenia

MACROMOLECULAR NANOTECHNOLOGY
Received 16 April 2007; received in revised form 8 May 2007; accepted 9 May 2007
Available online 24 May 2007

Abstract

For semicrystalline polymers there is an ongoing debate at what temperature the immobilized or rigid amorphous fraction
(RAF) devitrifies (relaxes). The question if the polymer crystals are melting first and simultaneously the RAF devitrifies or the
RAF devitrifies first and later on the crystals melt cannot be answered easily on the example of semicrystalline polymers. This
is because the crystals, which are the reason for the immobilization of the polymer, often disappear (melt) in the same tem-
perature range as the RAF. For polymer nanocomposites the situation is simpler. Silica nanoparticles do not melt or undergo
other phase transitions altering the polymer–nanoparticle interaction in the temperature range where the polymer is ther-
mally stable (does not degrade). The existence of an immobilized fraction in PMMA SiO2 nanocomposites was shown on
the basis of heat capacity measurements at the glass transition of the polymer. The results were verified by enthalpy relaxation
experiments below the glass transition. The immobilized layer is about 2 nm thick at low filler content if agglomeration is not
dominant. The thickness of the layer is similar to that found in semicrystalline polymers and independent from the shape of
the nanoparticles. Nanocomposites therefore offer a unique opportunity to study the devitrification of the immobilized frac-
tion (RAF) without interference of melting of crystals as in semicrystalline polymers. It was found that the interaction
between the SiO2 nanoparticles and the PMMA is so strong that no devitrification occurs before degradation of the polymer.
No gradual increase of heat capacity or a broadening of the glass transition was found. The cooperatively rearranging regions
(CRR) are either immobilized or mobile. No intermediate states are found. The results obtained for the polymer nanocom-
posites support the view that the reason for the restricted mobility must disappear before the RAF can devitrify. For semi-
crystalline polymers this means that rigid crystals must melt before the RAF can relax.
! 2007 Elsevier Ltd. All rights reserved.

Keywords: Nanocomposites; Immobilized layer; Rigid amorphous fraction (RAF); PMMA; Glass transition; Calorimetry (DSC)

1. Introduction tional properties. Searching in SCOPUSTM [1] for

‘‘polym* nanocompos* OR polymer inorganic
Polymer nanocomposites have attracted a great hybrid’’ yields more than 8000 hits from journals
deal of attention in recent years due to their excep- and more than 35,000 patents [2–17] and references
therein to name a few. Layered silicates [3], ceramic
*
Corresponding author. Fax: +49 381 498 6882. nanoparticles such as silica and titania [18], carbon
E-mail address: [email protected] (C. Schick). [19–21] and others are used as nanofillers. Compared

0014-3057/$ - see front matter ! 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.eurpolymj.2007.05.011
 3114 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127

to conventional micro and macro composites the by the dynamic experiment for the identification of
enormous surface to volume ratio of the nanoparti- an immobilized interfacial layer (Rigid Amorphous
cles is the most important factor. The improved Fraction – RAF) is was demonstrated for semicrys-
properties of nanocomposites are related to the talline poly(ethylene terephthalate) (PET) [51,52].
modification of the structure and dynamics of the For the dynamic glass transition from dielectric,
polymer at and near the particle surface. Because dynamic mechanical and temperature modulated
of the large surface area this fraction of the polymer DSC an immobilized fraction (RAF) was detected.
contributes significantly to the properties of the In contrary, data from the more local secondary
whole nanocomposite, even at low filler content. In b-relaxation process were well described by a two
this respect polymer nanocomposites are somehow phase model not requiring the introduction of a
similar to semicrystalline polymers where the crys- RAF.
tals can be considered as nanofillers too. Even calorimetry, mainly Differential Scanning
The glass transition, calorimetrically measured as Calorimetry (DSC), is routinely used to characterize
well as the dynamic glass transition studied by dif- nanocomposites often the glass transition tempera-
ferent probes (a-relaxation in amorphous poly- ture was reported only. In a few other studies the
MACROMOLECULAR NANOTECHNOLOGY

mers), is often used to detect changes in molecular shape of the glass transition interval was investi-
dynamics in polymers. However, experimental gated too [27,32,46,53] or heat capacity was mea-
results on polymer dynamics and the glass transition sured quantitatively [42]. Privalko recognized very
in polymer nanocomposites are not conclusive con- early the importance of absolute heat capacity mea-
cerning the mechanism and the details of the modi- surements for the thermodynamic characterization
fication near the particle surface. The glass of nanocomposites [25,36]. In this paper we report
transition temperature of the nanocomposite was heat capacity measurements for nanocomposites of
found to increase [22–28], to decrease [27,29–31], polymethylmethacrylate (PMMA) and polystyrene
not to be influenced at all [22,25,27,29,32,33] or (PS) with silicon oxide nanoparticles of different
the glass transition disappeared totally [27,34–36]. shape. To identify an immobilized interfacial frac-
However, there are many experimental results sug- tion of the polymer we apply a formalism well estab-
gesting that the restriction of chain mobility caused lished for the determination of a rigid amorphous
by the nanoparticles does not extend throughout the fraction (RAF) in semicrystalline polymers as
material but affects only the chains within a few described by Wunderlich et al. [54,55].
nanometers of the filler surface. The existence of For semicrystalline polymers there is an ongoing
such an interfacial layer was shown for several filler debate at what temperature the immobilized frac-
polymer combinations by different techniques tion (RAF) devitrifies (relaxes), see e.g. [56–58].
[29,32,37–43]. In some cases the interfacial layer The question if the polymer crystals are melting first
was identified as totally immobilized [32,35,38] and simultaneously the RAF devitrifies or the RAF
while in others a second glass transition [44,45] devitrifies first and later on the crystals melt cannot
was observed at higher temperature or at least a be answered easily on the example of semicrystalline
shoulder at the high temperature flank of the relax- polymers. This is because the crystals, which are the
ation peak [46]. The second peak observed in the reason for the immobilization of the polymer, often
mechanical tand curves by Eisenberg et al. was alter- disappear (melt) in the same temperature range as
natively, as an example to highlight the problem, the RAF. For polymer nanocomposites the situa-
interpreted as an indication for the formation of a tion is simpler. Silica nanoparticles do not melt or
macroscopic gel in the studied nanocomposites undergo other phase transitions altering the poly-
[47] and not as the glass transition of the interfacial mer–nanoparticle interaction in the temperature
layer as discussed in [44,45]. Obviously a peak in range where the polymer is thermally stable (does
dynamic loss curves does not necessarily identify a not degradate). Therefore polymer silica nanocom-
glass transition. Additional criteria must be fulfilled. posites are well suited for a detailed study of the
A straight forward proof of a glass transition is the glass transition of an immobilized layer at the inter-
observation of the typical step in heat capacity. This face between the polymer and the nanoparticle. Sev-
step like change in heat capacity does not occur for eral authors claim to observe such a second glass
local or normal mode relaxation processes because transition, see e.g. [44–46]. In all these cases the sec-
of the missing contribution from entropy fluctua- ond glass transition is detected as a separate peak or
tion [48–50]. How important the length scale probed a shoulder of the a-relaxation peak from dynamic
 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127 3115

measurements. But to the best of our knowledge case of the spherical silica nanoparticles 2 ml of
there is no evidence for a second glass transition the monomer were mixed with the appropriate
in polymer nanocomposites from calorimetric stud- amount of nanopowder and sonicated for 20 min
ies so far. We were therefore interested to obtain at 250 W. At filler loadings above 10 m% chloro-
polymer nanocomposites with a significant amount form was added to the monomer to avoid gelation
of the immobilized fraction and to measure heat of the mixture. Then the monomer nanoparticle dis-
capacity in order to detect a possible second glass persion was emulsified in 80 ml bidistilled and
transition as an increase of heat capacity towards deionized water and heated up to 70 "C. The Lapo-
liquid heat capacity at temperatures above the glass nite RDTM layered silicate was dispersed in 70 ml
transition of the mobile polymer. water by sonication for 10 min. Then the monomer
was emulsified in the water silicate dispersion and
2. Experimental heated to 70 "C. For both systems the aqueous ini-
tiator solution (10 ml) was added to the reaction
For the investigation two polymers with different media and polymerization was carried out with con-
functionality were chosen. One is polystyrene (PS) – tinuous pulsing sonication for 4 h. The polymer

MACROMOLECULAR NANOTECHNOLOGY
a polymer known to dewet on silica surfaces, e.g. latexes obtained were centrifuged at 6000 rpm and
[59], because of a lack of bonds with silica. The sec- the sediment was dried in vacuum (0.1 mbar) at
ond polymer is polymethylmethacrylate (PMMA) 150 "C for 8 h before all experiments.
having a tendency to interact with the nanoparticles
surface. Both polymers and the corresponding 2.3. Characterization of the nanocomposites
nanocomposites were synthesized by surfactant free
microemulsion polymerization. From the surfactant free microemulsion poly-
merization under continuous sonication polymer
2.1. Materials particles of about 150 nm diameter were obtained,
see Fig. 1. The synthesized pure polymer was ana-
The monomers, methylmethacrylate and styrene, lyzed by GPC yielding Mn = 6.1 · 104 g/mol, Mw =
were distilled under reduced pressure; potassium 3.3 · 105 g/mol.
persulfate (initiator) – 99%, silicon dioxide nano- Due to sonication and the not totally tight seal-
powder (catalog #637246, spherical particles with ing of the reaction vessel at 70 "C some monomer
d ! 10 nm) – 99.5%, were used as received from was lost at the beginning of the polymerization reac-
Sigma Aldrich. tion. Therefore an independent determination of the
Laponite RDTM is synthetic clay provided by
Southern Clay Products (Gonzales, Texas). It con-
sists of nearly monodisperse, thin cylindrical plate-
lets with a crystalline unit cell similar to that of
the natural montmorrilonite phyllosilicates [60–62].
A typical Laponite RDTM platelet has the appear-
ance of a flat coin of diameter D ! 25 nm, thickness
d ! 1 nm.

2.2. Microemulsion polymerization

Polymer nanocomposites were prepared avoiding

as much as possible the addition of other compo-
nents possibly effecting on the polymer particle
interaction. Therefore no additional surfactants
were used neither for the dispersion of the nanopar-
ticles nor for microemulsion polymerization. Micro-
emulsion polymerization was carried out at 70 "C
using 2 mass% (m%, in respect to monomer) potas- Fig. 1. Raster electron microscopic image of the dried PMMA
sium persulfate as initiator. A similar synthesis nanocomposite filled with 11 m% Laponite RDTM from micro-
route was used in [63] yielding similar products. In emulsion polymerization.
 3116 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127

filler content of the polymer nanocomposites was located at the surface of the latex particle and they
needed. The mass of the residual silica was therefore prevent interdiffusion of the polymer between the
determined by thermogravimetry using a Setaram particles. The Laponite RDTM platelets are well sep-
Labsys simultaneous TG-DSC. The samples were arated and at 11 m% filler mainly an exfoliated
heated from room temperature to 650 "C at 2 K/ structure is seen in Fig. 3b.
min in air to remove the polymer. The mass percent- The question arises at which filler content a single
age (m%) of the nanoparticle was this way deter- layer of clay will totally cover the separated latex
mined with an uncertainty of about 2%. Laponite particles? For latex particles of diameter 150 nm,
RDTM itself contains about 10 m% volatile fraction, density 0.936 g cm"3, and 1 nm thick clay platelets
which was taken into account for the determination of density 2.4 g cm"3 this is the case at a filler con-
of the Laponite RDTM m% determination from the tent of about 10 m%. The TEM images shown in
TGA residual at 650 "C (see Fig. 2). Fig. 3 are in good agreement with this estimate.
After centrifugation and drying in vacuum the At higher clay loadings either multi layers are
polymer nanocomposites were compacted at formed or the clay is incorporated into the latex
150 "C and 200 kPa. From the obtained tablets of particles. From the heat capacity data presented
MACROMOLECULAR NANOTECHNOLOGY

5 mm diameter and 1 mm thickness ultra microtome below both cases seem to occur for the PMMA/
slices of about 50 nm thickness for TEM investiga- Laponite RDTM nanocomposites synthesized with
tions were prepared. The TEM image, Fig. 3a, higher clay loadings. For the spherical silica nano-
shows the slightly deformed but still separated latex particles some agglomeration is seen in Fig. 4. But
particles. The Laponite RDTM platelets are mainly the agglomerates appear not very densely packed.

a 0
b 0
10 10
PMMA : SiO2
20 20
m% : m%
30 100 : 0 30
Mass loss in %

Mass loss in %

96 : 4
40 85 : 15 40
78 : 22
PMMA : Laponite
50 70 : 30 50 m% : m%
53 : 47 100 : 0
60 34 : 66 60 89 : 11
27 : 73 86 : 14
70 0 :100 70 73 : 27
58 : 42
80 80 41 : 59
0 :100
90 90

100 100
100 200 300 400 500 600 100 200 300 400 500 600
Temperature in ˚C Temperature in ˚C

c 0
10
20
30
Mass loss in %

40
50 PS : SiO 2
m% : m%
60 100 : 0
91 : 9
70 76 : 24
54 : 46
80 0 :100

90
100
100 200 300 400 500 600
Temperature in ˚C

Fig. 2. Mass loss of the nanocomposites of PMMA with spherical SiO2 nanoparticles (a), with Laponite RDTM (b) and PS with spherical
SiO2 nanoparticles (c) with temperature, Labsys TGA, sample mass about 40 mg, air purge.
 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127 3117

MACROMOLECULAR NANOTECHNOLOGY
Fig. 3. TEM images of the sample from Fig. 1 after melt pressing at 150 "C and 200 kPa.

Fig. 4. TEM images of the PMMA nanocomposite with 4 m% spherical SiO2 nanoparticles from microemulsion polymerization. The
single SiO2 particles are 10 nm in diameter.

2.4. Calorimetry and not further considered. Uncertainty of heat

capacity was about 2%.
Precise heat capacity curves were obtained using
a PerkinElmer PYRIS Diamond DSC. Heat capac- 3. Immobilized fraction from heat capacity
ity was determined from Step Scan Differential
Scanning Calorimetry (SSDSC), a special variant For semicrystalline polymers it is well known
of temperature modulated DSC. Measurements that the observed step in heat capacity, Dcp, at the
were performed using samples of about 20 mg, glass transition is often less than expected from crys-
3 K steps at heating rate 6 K/min and isotherms tallinity. Wunderlich et al. therefore introduced the
of about 1 min [64,65]. The temperature range, if concept of a rigid amorphous fraction (RAF), see
not other indicated, was 30–170 "C. The instrument [54] and a recent review [55].
was calibrated as recommended by GEFTA [66] by
RAF ¼ 1 " crystallinity " Dcp =Dcp pure ð1Þ
indium and zinc at zero heating rate for temperature
and sapphire for heat capacity. Due to the low where Dcp and Dcp pure are the heat capacity incre-
underlying heating rate, 3 K/min, deviations from ments at the glass transition for the semicrystalline
the zero rate calibration were less than 0.2 K [67] and the amorphous polymer, respectively. The ratio
 3118 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127

of both step heights represents the fraction of the as purge enhancing the drying process. Sample mass
polymer contributing to the glass transition, which was determined after the measurements.
is called mobile amorphous fraction (MAF). In Fig. 5 the measured specific heat capacities for
Eq. (1) can not only be applied to calorimetric the PMMA and PS based nanocomposites are
relaxation strength data but also for other responses shown. The curves were calculated in respect to
at the dynamic glass transition as shown in the sample mass (polymer + nano particles). Due
[51,52,68]. From the cooperative dynamics at the to the lower specific heat capacity of the SiO2 com-
dynamic glass transition a RAF was detected for pared to that of the polymer the curves are shifted
dielectric, dynamic mechanical and TMDSC mea- with increasing filler content. Such a shift is not
surements. For the secondary more local b-relaxtion observed for semicrystalline polymers where the
process in semicrystalline poly(ethylene terephthal- specific heat capacity of the crystalline, the rigid
ate), as an example, no RAF was obtained applying amorphous and the mobile amorphous fractions
Eq. (1) to the dielectric and mechanic relaxation are very similar below the glass transition. As for
strength. The b-relaxtion process occurs in both semicrystalline polymers a slight increase in glass
the mobile and the rigid amorphous fraction but it transition temperature (Fig. 6) was observed for
MACROMOLECULAR NANOTECHNOLOGY

is not present in the crystalline fraction as deduced all nanocomposites after carefully drying at reduced
from broad band dielectric spectroscopy [51,52,68]. pressure. The glass transition temperature of the
These findings support the idea that the rigid amor- nanocomposites was determined as the half step
phous layer is really immobilized for the cooperative temperature from the StepScanTM DSC measure-
motions, which are needed for the liquid like behav- ments. The values are slightly different compared
ior above the glass transition, and not a gradient in to scan measurements at 10 K/min because of the
mobility from the mobile amorphous fraction to the different time scale of the experiment. Glass transi-
crystal surface causes a continuous decrease in the tion temperature and width of the step in heat
contribution to Dcp. For a discussion of the proper- capacity were only little affected by the addition of
ties of an immobilized layer the length scale of the nanofillers. This is different from semicrystalline
motions probed has therefore to be considered. polymers where always a significant increase in glass
For calorimetric measurements we assume a length transition temperature and a broadening of the
scale of a few nanometers, typical for cooperative transition interval is seen, e.g. [70].
motions [48,69,70]. From the curves presented in Fig. 5 the height of
For polymer nanocomposites sometimes a the step in specific heat capacity at the glass transi-
reduced step height (relaxation strength), Dcp, at tion temperature was determined too. It decreases
glass transition is observed too, see e.g. [42]. An with increasing filler content as expected because
immobilized or rigid amorphous fraction can be the filler does not contribute to the glass transition.
determined from heat capacity according to Eq. If a RAF exists is difficult to see from the measured
(1) replacing the crystalline by the nanoparticle frac- curves without further processing of the data even
tion [25,36]. the PMMA nanocomposites at highest filler con-
tents show a much smaller step in heat capacity than
RAF ¼ 1 " filler content " Dcp =Dcp pure ð2Þ
expected from the filler content.
In order to apply Eq. (2) precise heat capacity data Next, the amount of RAF in the PMMA nano-
in the temperature range of the glass transition of composites was quantified applying Eq. (2). The
the polymer nanocomposite are needed. Even the step in heat capacity at the glass transition for the
samples were carefully dried above Tg at reduced different samples was determined as usual at Tg
pressure there may be some remaining solvent evap- and normalized by the step in heat capacity for
orating during the heat capacity measurement. To the pure polymer. Uncertainty of the step height is
minimize the influence of such events on the final re- mainly due to uncertainties in the slope of the tan-
sults we used a temperature modulated DSC gents needed for the determination. This is dis-
(TMDSC) technique to determine heat capacity. Be- cussed in more detail in [73]. The results for the
cause evaporation is non-reversing it does not con- PMMA and PS nanocomposites are shown in
tribute to complex or reversing heat capacity [71]. Fig. 7.
Additionally, before the TMDSC measurements According to Eq. (2) the diagonal in Fig. 7 repre-
the samples were heated at 20 K/min to 170 "C sents the case when no RAF is present. In other
and annealed for 10 min. Dry nitrogen was used words a two phase system (filler + polymer) which
 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127 3119

a 2.2 PMMA : SiO2 m% : m% b 2.2 PMMA : Laponite m% : m%

100 : 0 ATHAS
Specific heat capacity in J/K*gsample

100 : 0 ATHAS

Specific heat capacity in J/K*gsample

2.0 100 : 0 34 : 66 100 : 0
96 : 4 27 : 73 2.0 89 : 11
85 : 15 0 :100 86 : 14
1.8 78 : 22 1.8 73 : 27
70 : 30 58 : 42
1.6 53 : 47 41 : 59
1.6 0 :100

1.4 1.4

1.2 1.2

1.0 1.0

SiO2 SiO2
0.8 0.8

40 60 80 100 120 140 160 40 60 80 100 120 140 160

Temperature in ˚C Temperature in ˚C

MACROMOLECULAR NANOTECHNOLOGY
2.2 PS : SiO2 m% : m%
Specific heat capacity in J/K*gsample

100 : 0 ATHAS
2.0 100 : 0
91 : 9
76 : 24
1.8
54 : 46

1.6

1.4

1.2

1.0
SiO2
0.8

40 60 80 100 120 140

Temperature in ˚C

Fig. 5. Specific heat capacity of nanocomposites in respect to sample mass. (a) PMMA with spherical SiO2 nanoparticles, (b) PMMA with
Laponite RDTM, (c) PS with spherical SiO2 nanoparticles. Specific heat capacity for SiO2 is shown too. The black lines represent reference
data for PMMA and PS from ATHAS [72]. Pyris Diamond DSC, StepScanTM mode.

8
PMMA/SiO 2
7 PMMA/Laponite
1.0
PS/SiO2 Filler 1.5
6 0.9
ω RAF /ω filler

1.0
5 0.8
0.5
RAF
Tg - Tg pure in K

0.7 0.0
4
pure

0.6 -0.5
∆ cp sample / ∆ cp

3 0 20 40 60 80
0.5
Filler content in mass%
2
0.4
1 MAF
0.3
0 0.2 PS/SiO2
-1 PMMA/SiO2
0.1
PMMA/Laponite
0 10 20 30 40 50 60 70 0.0
Filler content in wt% 0 10 20 30 40 50 60 70 80 90 100
Filler content in mass%
Fig. 6. Glass transition temperature of the different nanocom-
posites as function of filler content. Half step temperature from Fig. 7. Calorimetric relaxation strength as a function of nano-
StepScanTM DSC measuremnts. The Tg of the pure polymers was filler content. The vertical double arrow indicates the amount of
117 "C and 99 "C for PMMA and PS, respectively. RAF for PMMA at 27 m% Laponite RDTM filler. Symbols: n –
PS with spherical SiO2 nanoparticles; h – PMMA with spherical
SiO2 nanoparticles; e – PMMA with Laponite RDTM clay
is expected if there is no interfacial immobilization. nanoparticles. The inset shows the percentage of the RAF versus
The data for the PS nanocomposite are close to this filler content, see text.
 3120 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127

line. For the PMMA nanocomposites the decrease rating during the measurement even after carefully
of the normalized relaxation strength is much stee- drying the sample in vacuum.
per than the diagonal. The glass transition disap- As shown in [73] a simpler correction can be
pears (Dcp = 0) at about 50 m% and 70 m% for the applied. First, specific heat capacity was recalcu-
Laponite RDTM filler and the spherical silica particle, lated in respect to the polymer fraction by dividing
respectively. From the normalized relaxation the measured specific heat capacity by (1 – filler con-
strength the different fractions according to Eq. (2) tent). Next, assuming, as for the semicrystalline
can be obtained. The upper arrow at 27 m% filler polymers, a similar specific heat capacity for the
corresponds to the filler fraction. The lowest indi- possible RAF and the MAF below the glass transi-
cates the mobile amorphous fraction (MAF) con- tion the curves can be aligned to the specific heat
tributing to the calorimetric relaxation strength at capacity of the pure PMMA below the glass transi-
glass transition. The difference between the mea- tion in the temperature range 40–80 "C by shifting
sured values and the diagonal (middle arrow) repre- and rotating. This procedure can be applied because
sents the immobilized (rigid) fraction (RAF) which of the basically linear temperature dependence of
can be calculated according to Eq. (2). the specific heat capacity of the SiO2 in the temper-
MACROMOLECULAR NANOTECHNOLOGY

Straight lines in Fig. 7 correspond to a constant ature region of interest as shown in Fig. 5.
ratio between RAF and filler fraction. In other If the polymer fraction in the nanocomposite
words, it is assumed in ideal case that each filler par- behaves like the pure polymer no differences in the
ticle is covered by the same amount of RAF (all specific heat capacity curves should be visible. This
according mass). This assumption is not well justi- seems to be the case for the PS nanocomposites,
fied by the data of Fig. 7. The mass percentage of except the highest filler content. But as seen from
the RAF normalized by the filler content is shown Fig. 8 the curves for the PMMA based nanocom-
in the inset as a function of the filler mass fraction. posites are different and especially the step height
If all nanoparticles were covered by the same immo- in specific heat capacity at the glass transition
bilized layer one would expect to find a constant decreases with increasing filler content. This can
ratio between the RAF and the filler content. The be interpreted as a RAF in the PMMA nanocom-
slight decrease of the ratio indicates a decrease of posites as already shown in Fig. 7. For the polymer
the RAF per nanoparticle with increasing filler con- fraction of the nanocomposites Eq. (2) can be
centration. This is most probably due to agglomer- rewritten for the immobilized fraction of the poly-
ation of the particles as indicated schematically in mer RAFpolymer [25,36]
the inset.
RAFpolymer ¼ 1 " Dcp polymer =Dcp pure ð3Þ
To allow a direct comparison of the heat capacity
curves of the polymer fraction of the nanocompos- From Fig. 8 the step height in specific heat capacity
ites the contribution of the filler has to be sub- Dcp polymer was determined as usual at Tg and nor-
tracted. Heat capacity (J/K) of the nanocomposite malized by the step in heat capacity for the pure
sample can be obtained from specific heat capacity polymer. Uncertainty of the step height is again
(J/gK) and known sample mass. Multiplying the mainly due to uncertainties in the slope of the tan-
specific heat capacity (J/gK) of the filler (SiO2) by gents needed for the determination. But in the nor-
the mass of filler in each sample yields the heat malized representation (Fig. 8) the tangents should
capacity contribution of the filler. The difference be parallel independent on filler content. Even the
of the heat capacities divided by the polymer mass situation is improved compared to Fig. 5 determina-
finally results in the specific heat capacity of the tion of the tangents is still highly subjective. This is
polymer fraction in the nanocomposite. This discussed in more detail in [73]. Nevertheless
straight forward procedure was applied but the RAFpolymer for PMMA and PS nanocomposites
results are not satisfying because of uncertainties was determined from the data presented in Fig. 8
in composition and in heat capacity of the nano- and are shown in Fig. 9. According to Eq. (3) the ri-
composites [73]. Another problem arises from the gid amorphous fraction of the polymer in the nano-
organic part of the Laponite RDTM. Laponite RDTM composite (RAFpolymer) could be obtained. The
specific heat capacity is therefore different from SiO2 result for the 66 m% SiO2 PMMA nanocomposite
specific heat capacity. To measure Laponite’s spe- is indicated by the vertical arrows in Fig. 9.
cific heat capacity is not an easy task because of According to Eq. (3) a value of unity in Fig. 9
the water adsorbed on the large surface and evapo- represents the case when no RAF is present (two
 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127 3121

a 2.3
PMMA : SiO 2 m% : m%
b 2.3
PMMA : Laponite m% : m%
Specific heat capacity in J/K*gpolymer

2.2 2.2

Specific heat capacity in J/K*g polymer

100 : 0 ATHAS 100 : 0 ATHAS
100 : 0 100 : 0
2.1 96 : 4 2.1 89 : 11
85 : 15 86 : 14
2.0 78 : 22 2.0 73 : 27
70 : 30 58 : 42
1.9 53 : 47 1.9 41 : 59
34 : 66
1.8 27 : 73 1.8
1.7 1.7
1.6 1.6
1.5 1.5
1.4 1.4
1.3 1.3
40 60 80 100 120 140 160 40 60 80 100 120 140 160
Temperature in ˚C Temperature in ˚C

c 2.1

MACROMOLECULAR NANOTECHNOLOGY
PS : SiO2 m% : m%
2.0
Specific heat capacity in J/K*gpolymer

100 : 0 ATHAS
1.9 100 : 0
91 : 9
1.8 76 : 24
54 : 46
1.7

1.6

1.5

1.4

1.3

1.2

1.1
40 60 80 100 120 140
Temperature in ˚C

Fig. 8. Specific heat capacity of the polymer fraction in the nanocomposites. The curves are aligned to the specific heat capacity of the pure
polymers below the glass transition by shifting and rotating as described in the text. (a) PMMA with spherical SiO2 nanoparticles, (b)
PMMA with Laponite RDTM and (c) PS with spherical SiO2 nanoparticles. The black lines represent reference data for PMMA and PS
from ATHAS [72]. Pyris Diamond DSC, StepScanTM mode.

1.0
phase system; filler + polymer). The curved solid
lines result from a model assuming a constant ratio
0.8 between RAF and filler content (equivalent to the
/ ∆cp pure polymer

RAFpolymer straight lines in Fig. 7). The ratio equals 0.1, 0.4
0.6 and 1 for PS, PMMA spherical SiO2 and PMMA
Laponite RDTM nanocomposites, respectively.
Assuming a decrease of the RAF proportional to
p polymer

0.4
the polymer fraction yields the straight lines in
∆c

0.2 MAFpolymer
Fig. 9. Even the determination of Dcp polymer from
PS SiO2
PMMA SiO2 Fig. 8 is a bit more objective than Dcp sample from
0.0
PMMA Laponite RD Fig. 5 it does not give much better results. In both
0 10 20 30 40 50 60 70 80 90 100 cases a RAF is obviously detected for the PMMA
Filler content in mass% nanocomposites. For the PS nanocomposite the
Fig. 9. Calorimetric relaxation strength of the polymer fraction result is not definite. Therefore an independent
as a function of nanofiller content. Symbols: m – PS with determination of the RAF was needed to allow def-
spherical SiO2 nanoparticles; j – PMMA with spherical SiO2 inite conclusions (see Fig. 10).
nanoparticles; r – PMMA with Laponite RDTM; synthesized by
Determination of Dcp for RAF estimation
in situ microemulsion polymerization. The lines through the
measured points are described in the text. The vertical double requires the construction of tangents to the mea-
arrows indicate the amount of RAF and MAF at 70 m% filler for sured heat capacity curves as discussed above. These
PMMA with spherical SiO2 nanoparticles. tangents are sometimes not well defined and in this
 3122 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127

1.0 1.4

Excess specific heat capacity in J/K*gsample

14 PS : SiO 2
pure PMMA
Temperature in ˚C

150 1.2 m% : m%

Enthalpy change in J/gpolymer

600 min 0.8 100 : 0
13 100 1.0 76 : 24
Heat flow in mW, Endo up

50 0.8
12 35 m% SiO2 0.6
0 20 600 650 0.6
Time in min
11 0.4 0.4

0.2
10
0.2 0.0
9 -0.2
0.0
8 -120 -100 -80 -60 -40 -20 0 20 40
70 80 90 100 110 120 130 140 150 Tanneal -Tg in K
Temperature in ˚C
Fig. 11. Enthalpy change (J/gpolymer) during annealing for 10 h as
Fig. 10. Heat flow rate and excess specific heat capacity as a function of the annealing temperature for PS and PS nanocom-
MACROMOLECULAR NANOTECHNOLOGY

function of sample temperature of PMMA annealed at 105 "C for posite with 24 m% spherical SiO2 nanoparticles; Tg PS = 104 "C,
10 h; n, , – heating after annealing, h, e – heating without Tg composite = 106 "C.
annealing, n, , – excess heat capacity for 24 mg pure PMMA
and 25 mg nanocomposite with 35 m% SiO2, respectively.

1.4
PMMA : SiO2
cases RAF determination is uncertain. Molecular 1.2 m% : m%
Enthalpy change in J/gpolymer

mobility at and below the glass transition can be 1.0

100 : 0
65 : 35
tested by annealing experiments too, see [74] for a
0.8
review. If a fraction of the amorphous polymer in
the nanocomposites is immobilized it is expected 0.6

that enthalpy relaxation below glass transition is 0.4

reduced too [75–77]. To check if this provides more 0.2
definite results regarding RAF the following anneal- 0.0
ing experiments were performed.
-0.2
The sample was first heated well above the glass
transition (170 "C) to erase the previous thermal his- -120 -100 -80 -60 -40 -20 0 20
Tanneal -Tg in K
tory. After cooling at 10 K/min to the annealing
temperature the sample was annealed for 10 h. To Fig. 12. Enthalpy change (J/gpolymer) during annealing for 10 h as
detect the enthalpy change due to the annealing function of the annealing temperature for PMMA and PMMA
the sample was cooled to 30 "C and reheated up nanocomposite with 35 m% spherical SiO2 nanoparticles;
Tg PMMA = 111 "C, Tg composite = 117 "C.
to 170 "C. The typical annealing peak is seen in
the first heating, which is not present in the second
heating without annealing. From the difference of From the enthalpy relaxation an independent
both curves excess heat capacity can be obtained proof is available that there is an immobilized frac-
as shown in the bottom curve. Integration from 80 tion (RAF) in PMMA SiO2 nanocomposites, which
to 140 "C finally gives the specific enthalpy change is not present in PS SiO2 nanocomposites (see
Dh during annealing. To allow a direct comparison Fig. 13).
the specific enthalpy change was normalized to the
polymer fraction of the nanocomposites. Because 4. Discussion
of the small shift in glass transition temperature
results are plotted versus Tannealing " Tg. For PS Interaction at the filler polymer interface is con-
there is no difference in the enthalpy relaxation for sidered important for the behavior of polymer
the pure polymer and the nanocomposite as shown nanocomposites. Following the work of Lipatov
in Fig. 11. and Privalko [25,36] the fraction of immobilized
For PMMA nanocomposites the result is differ- polymer (RAF) was quantified for PMMA and PS
ent (Fig. 12). Only a fraction of the expected SiO2 nanocomposites. The occurrence of a RAF
enthalpy relaxation occurs. was confirmed for PMMA nanocomposites by
 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127 3123

a b c
Mobile
polymer
Nano-
particle

Rigid
amorphous

dRAF ca. 2 nm
Fig. 13. Sketch of spherical (a, b) and layered (c) nanoparticles (black or blue) covered by a layer of immobilized polymer (RAF) (dark
grey or green) in a mobile amorphous matrix (light grey or light green). Total deglomeration of the particles is assumed in (a); a more
realistic situation is shown in (b). (c) shows single platelets (lower left and upper right corner) and intercalation (center) as well as stacks of

MACROMOLECULAR NANOTECHNOLOGY
platelets (upper right) or intermediate forms (lower left). (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)

enthalpy relaxation studies too. For the PS SiO2 3.0

PMMA/SiO2
2.8
nanocomposites studied the result from heat capac- PMMA/Laponite RD
2.6
ity is not well defined but no RAF was detected
2.4
from enthalpy relaxation. 2.2
Existence of a RAF was shown for PMMA SiO2
dRAF in nm

2.0
nanocomposites by calorimetric experiments detect- 1.8
ing contributions from liquid like degrees of free- 1.6
dom to heat capacity and enthalpy relaxation. In 1.4
both experiments cooperative motions on a length 1.2

scale of about 2 nm are probed [48,69]. This is much 1.0

0.8
larger than the range of the interaction forces for a
0.6
polymer molecule at the filler surface. The question 0 10 20 30 40 50 60 70 80
arises what the thickness of the immobilized layer Filler content in mass%
around a nanofillers particle is. Fig. 14. Thickness of the immobilized layer around the nano-
From geometric consideration assuming spheri- particles (RAF) as a function of filler content for PMMA
cal particles of 10 nm diameter for the SiO2 filler nanocomposites. No agglomeration is assumed for the calcula-
or platelets with 1 nm thickness for the Laponite tion. The lines are guides to the eyes only.
RDTM filler and a density of 1 g/cm3 for the polymer
and 2.4 g/cm3 for SiO2 a layer thickness ranging with the nanoparticle. Following the idea that coop-
from 2.5 to 1 nm follows from the data shown in eratively rearranging regions (CRR) [48,69,79] are
Figs. 7 and 9. responsible for the liquid like motions near the glass
The relative amount of RAF (RAF/Filler) is sig- transition this observation can be explained as fol-
nificantly larger for the PMMA Laponite RDTM lows. Immobilizing a part of a polymer molecule
nanocomposite compared to the PMMA spherical at the interface affects the movement of all neigh-
SiO2 nanocomposite, Fig. 7 inset and Fig. 9. Despite boring segments within a CRR. Therefore the whole
this the thickness of the RAF layer around the CRR cannot contribute to the liquid like motions
nanoparticles at low filler concentration in Fig. 14, and consequently not to the increment of heat
when agglomeration is not dominant, is nearly the capacity at the glass transition. Assuming molecules
same – about 2.5 nm. A similar value (2 nm) was are bound on one side of the CRR to the interface.
found for the RAF layer at the fold surface of semi- Then at the opposite side, at a distance of about
crystalline PET [70] and 1.5 nm for a filled SBR 2 nm from the interface, there is no local immobili-
1500 rubber [78]. The thickness of the immobilized zation of the polymer chains anymore and the
layer is in all cases much thicker than the range of ‘‘next’’ CRR behaves as in a bulk liquid. Therefore
the forces causing the interaction of the polymer no significant broadening of the glass transition is
 3124 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127

observed for the nanocomposites as it would be capacity. But there is no step like or gradual transi-
expected for a gradual change of mobility between tion observed up to 250 "C where the polymer starts
the interface and the liquid polymer [81]. to degrade at slow heating (see Fig. 16).
If immobilization is due to anchoring polymer Influence of degradation on the heat capacity
molecules at the nanoparticle surface this would determination can be reduced by using high heating
have significant influence on the way of devitrifica- rates. To shift the beginning of degradation to
tion of the immobilized layer. For semicrystalline higher temperatures HyperDSCTM measurements at
polymers it is often argued that devitrification of 400 K/min heating rate were performed [80].
the RAF occurs gradually above the common glass At 400 K/min heating rate the polymer in the
transition [55]. Assuming a local immobilization of nanocomposite does not degrade up to about
the polymer molecules at the nanoparticles interface 360 "C. But there is again no second glass transition
as the reason for the formation of a rather thick visible below 360 "C. This means that the interaction
RAF layer in turn requires a disappearance of the between the polymer matrix and the nanoparticles is
anchoring at the nanoparticle surface. No gradual so strong that heating even up to such high temper-
increase in heat capacity (broad second glass transi- atures is not enough to remove the anchors and to
MACROMOLECULAR NANOTECHNOLOGY

tion) is expected as long as the anchoring persists. allow relaxation and devitrification of the RAF.
Only removing the anchors will allow the immobi- Interaction between PMMA and SiO2 at the
lized layer to relax, to devitrify. In order to check interface of the nanoparticles is expected to be
the behavior of the RAF layer and to detect a pos- weaker than a covalent bond, which is present in
sible second glass transition we performed heat semicrystalline polymers if a polymer chain goes
capacity measurements up to the degradation of from a rigid crystal lamellae through the interface
the polymer. StepScanTM DSC was used first to and the immobilized layer to the mobile amorphous
obtain precise heat capacity data up to the begin- polymer and eventually back into the same or
ning of degradation. Measurements were performed another lamellae. If the non covalent bond between
up to degradation temperature but because of the the inorganic nanoparticle and the PMMA does not
isotherm after each 3 K temperature step heat allow devitrification before degradation of the poly-
capacity could be obtained until the heat flow dur- mer occurs it is very unlikely that in semicrystalline
ing the isotherm was not stable anymore. polymers the RAF devitrifies as long as the polymer
In Fig. 15 the specific heat capacity of PMMA chains are covalently anchored to the rigid polymer
nanocomposite filled with 47 m% spherical SiO2 crystals. Most likely the polymer crystals must melt
nanoparticles as a function of sample temperature before the RAF can relax and devitrify. This was
is shown. RAF devitrification is expected to appear shown to be the case for semicrystalline iPS by
as a second step in heat capacity towards liquid heat applying ultra fast scanning rates suppressing reor-
ganization of the crystals [57].

2.6
Specific heat capacity in J/K*g polymer

2.6
Specific heat capacity in J/K*gpolymer

2.4
2.4
2.2
2.2
2.0
PMMA : SiO2 m% : m% 2.0 PMMA : SiO2 m% : m%
1.8 100 : 0 ATHAS 100 : 0 ATHAS
100 : 0 1.8 100 : 0
1.6 53 : 47 53 : 47
1.6
1.4
60 80 100 120 140 160 180 200 220 240 1.4
Temperature in ˚C 50 100 150 200 250 300 350
Temperature in ˚C
Fig. 15. Specific heat capacity of pure PMMA and the nano-
composite filled with 47 m% spherical SiO2 nanoparticles mea- Fig. 16. HyperDSCTM measurements at 400 K/min heating rate
sured in StepScanTM DSC mode up to the degradation of the for pure PMMA and the nanocomposite filled with 47 m%
polymer. Sample mass 21 mg. The black lines represent reference spherical SiO2 nanoparticles. Sample mass 0.5 mg. The black
data for PMMA from ATHAS [72]. lines represent reference data for PMMA from ATHAS [72].
 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127 3125

5. Conclusion [2] Giannelis EP. Polymer layered silicate nanocomposites. Adv

Mater 1996;8(1):29–35.
[3] Alexandre M, Dubois P. Polymer-layered silicate nanocom-
The existence of an immobilized fraction in posites: preparation, properties and uses of a new class of
PMMA SiO2 nanocomposites was shown on the materials. Mater Sci Eng, R 2000;28(1–2):1–63.
basis of heat capacity measurements at the glass [4] Sinha Ray S, Okamoto M. Polymer/layered silicate nano-
transition of the polymer. The results were verified composites: a review from preparation to processing. Progr
by enthalpy relaxation experiments below the glass Polym Sci (Oxford) 2003;28(11):1539–641.
[5] Fornes TD, Yoon PJ, Keskkula H, Paul DR. Nylon 6
transition. The immobilized layer is about 2 nm nanocomposites: the effect of matrix molecular weight.
thick at low filler content if little agglomeration is Polymer 2001;42(25):9929–40.
present. At higher filler content agglomeration [6] Okamoto M, Morita S, Kim YH, Kotaka T, Tateyama H.
becomes important and the layer thickness cannot Synthesis and structure of smectic clay/poly(methyl meth-
be determined correctly. acrylate) and clay/polystyrene nanocomposites via in situ
intercalative polymerization. Polymer 2000;41(10):3887–90.
The immobilized fraction in polymer inorganic [7] Ishida H, Campbell S, Blackwell J. General approach to
nanocomposites can be determined from heat nanocomposite preparation. Chem Mater 2000;12(5):
capacity as it is common for the determination of 1260–7.

MACROMOLECULAR NANOTECHNOLOGY
the rigid amorphous fraction in semicrystalline [8] Hagrman PJ, Hagrman D, Zubieta J. Organic–inorganic
polymers. The thickness of the layer is also similar hybrid materials: from ‘simple’ coordination polymers to
organodiamine-templated molybdenum oxides. Angew
to that found in semicrystalline polymers and inde- Chem – Int Ed 1999;38(18):2638–84.
pendent from the shape of the nanoparticles. [9] Gilman JW. Flammability and thermal stability studies of
Nanocomposites offer a unique opportunity to polymer layered-silicate (clay) nanocomposites. Appl Clay
study the devitrification of the immobilized fraction Sci 1999;15(1–2):31–49.
(RAF) without interference of melting of crystals as [10] Lebaron PC, Wang Z, Pinnavaia TJ. Polymer-layered
silicate nanocomposites: an overview. Appl Clay Sci 1999;
in semicrystalline polymers. It was found that the 15(1–2):11–29.
interaction between the SiO2 nanoparticles and the [11] Antonietti M, Breulmann M, Göltner CG, Cölfen H, Wong
PMMA is so strong that no devitrification occurs KKW, Walsh D, et al. Inorganic/organic mesostructures
before degradation of the polymer. No gradual with complex architectures: precipitation of calcium phos-
increase of heat capacity or broadening of the glass phate in the presence of double-hydrophilic block copoly-
mers. Chem – A European J 1998;4(12):2493–500.
transition was found. The cooperatively rearranging [12] Krishnamoorti R, Giannelis EP. Rheology of end-tethered
regions (CRR) are either immobilized or mobile. No polymer layered silicate nanocomposites. Macromolecules
intermediate states were found. 1997;30(14):4097–102.
The results obtained for the polymer nanocom- [13] Vaia RA, Sauer BB, Tse OK, Giannelis EP. Relaxations of
posites support the view that the reason for the confined chains in polymer nanocomposites: glass transition
properties of poly(ethylene oxide) intercalated in montmo-
restricted mobility must disappear before the RAF rillonite. J Polym Sci, Part B: Polym Phys 1997;35(1):59–
can devitrify. For semicrystalline polymers this 67.
means that rigid crystals must melt before the [14] Krishnamoorti R, Vaia RA, Giannelis EP. Structure and
RAF can relax. Only for semicrystalline polymers dynamics of polymer-layered silicate nanocomposites. Chem
with significant chain mobility inside the crystals Mater 1996;8(8):1728–34.
[15] Firouzi A, Kumar D, Bull LM, Besier T, Sieger P, Huo Q,
RAF may devitrify before melting. et al. Cooperative organization of inorganic-surfactant and
biomimetic assemblies. Science 1995;267(5201):1138–43.
Acknowledgments [16] Messersmith PB, Giannelis EP. Synthesis and characteriza-
tion of layered silicate-epoxy nanocomposites. Chem Mater
1994;6(10):1719–25.
A.S. gratefully acknowledges financial support
[17] Novak BM. Hybrid nanocomposite materials – between
form DAAD through a ‘‘Sandwich Model’’ scholar- inorganic glasses and organic polymers. Adv Mater
ship. We thank the Electron Microscopy Center of 1993;5(6):422–33.
Rostock University for the electron micrographs [18] Mark JE. Ceriamic-reinforced polymers and polymer-mod-
and D. Pospiech (IPF Dresden) for fruitful discus- ified ceramics. Polym Eng Sci 1996;36(24):2905–20.
[19] Chen G, Wu C, Weng W, Wu D, Yan W. Preparation of
sions and support with sample characterization.
polystyrene/graphite nanosheet composite. Polymer
2003;44:1781–4.
References [20] Kotsilkova R, Nesheva D, Nedkov I, Krusteva E, Stavrev S.
Rheological, electrical, and microwave properties of poly-
[1] Elsevier BV. SCOPUS. See on WWW URL: <http:// mers with nanosized carbon particles. J Appl Polym Sci
www.scopus.com/>. 2004;92:2220–7.
 3126 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127

[21] Thostenson ET, Ren Z, Chou TW. Advances in the science adsorbed poly(dimethylsi1oxane). Macromolecules 1993;
and technology of carbon nanotubes and their composites: a 26:975–80.
review. Compos Sci Technol 2001;61:1899–912. [40] Litvinov VM, Spies HW. 2H NMR study of molecular
[22] Privalko VP, Titov GV. The heat capacity of filled, motions in polydimethylsiloxane and its mixtures with
amorphous polymers. Polym Sci USSR 1978;21:380–7. aerosils. Makromol Chem 1991;192:3005–19.
[23] Liu X, Wu Q. PP/clay nanocomposites prepared by grafting- [41] Lin WY, Blum FD. Segmental dynamics of bulk and silica-
melt intercalation. Polymer 2001;42:10013–9. adsorbed poly(methyl acrylate)-d3 by deuterium NMR: the
[24] Huang X, Brittain WJ. Synthesis and characterization of effect of molecular weight. Macromolecules 1998;31:
PMMA nanocomposites by suspension and emulsion poly- 4135–42.
merization. Macromolecules 2001;34(10):3255–60. [42] Yagrarov MS, Ionikin VS, Gizatullina ZG. Study of the
[25] Lipatov YS, Privalko VP. Glass transition in filled polymer low-temperature thermal behaviour of filled poly-
systems. Polym Sci USSR 1972;14(7):1843–8. dimethylsiloxanes. Polym Sci USSR 1969;11(10):2626.
[26] Kalogeras IM, Neagu NR. Interplay of surface and [43] Haraguchi K, Li HJ. Mechanical properties and structure of
confinement effects on the molecular relaxation dynamics polymer–clay nanocomposite gels with high clay content.
of nanoconfined poly(methyl methacrylate) chains. Eur Phys Macromolecules 2006;39:1898–905.
J E 2004;14:193–204. [44] Tsagaropoulos G, Eisenberg A. Dynamic mechanical study
[27] Bershtein VA, Egorova LM, Yakushev PN, Pissis P, Sysel P, of the factors affecting the two glass transition behavior of
Brozova L. Molecular dynamics in nanostructured poly- filled polymers. similarities and differences with random
MACROMOLECULAR NANOTECHNOLOGY

imide–silica hybrid materials and their thermal stability. J ionomers. Macromolecules 1995;28(18):6067–77.
Polym Sci B: Polym Phys 2002;40:1056–69. [45] Tsagaropoulos G, Eisenberg A. Direct observation of two
[28] Tabtiang A, Lumlong S, Venables RA. The influence of glass transitions in silica-filled polymers. implications for the
preparation method upon the structure and relaxation morphology of random ionomers. Macromolecules 1995;
characteristics of poly(methyl methacrylate)/clay compos- 28:396–8.
ites. Europ Polym J 2000;36(12):2559–68. [46] Fragiadakis D, Pissis P, Bokobza L. Glass transition and
[29] Arrighi V, McEwena IJ, Qiana H, Serrano Prieto MB. The molecular dynamics in poly(dimethylsiloxane)/silica nano-
glass transition and interfacial layer in styrene–butadiene composites. Polymer 2005;46:6001–8.
rubber containing silica nanofiller. Polymer 2003;44: [47] Salaniwal S, Kumar SK, Douglas JF. Amorphous solidifi-
6259–66. cation in polymer–platelet nanocomposites. Phys Rev Lett
[30] Ash BJ, Schadler LS, Siegel RW. Glass transition behavior 2002;89(25):258301–4.
of alumina/polymethylmethacrylate nanocomposites. Mater [48] Donth E. Glass transition. Berlin: Springer; 2001.
Lett 2002;55:83–7. [49] Schick C, Sukhorukov D, Schönhals A. Comparison of the
[31] Ash BJ, Siegel RW, Schadler LR. glass-transition temper- molecular dynamics of a liquid crystalline side group
ature behavior of alumina/PMMA nanocomposites. J Polym polymer revealed from temperature modulated DSC and
Sci B: Polym Phys 2004;42:4371–83. dielectric experiments in the glass transition region. Macro-
[32] Miwa Y, Drews AR, Schlick S. Detection of the direct effect mol Chem Phys 2001;202(8):1398–404.
of clay on polymer dynamics: the case of spin-labeled [50] Huth H, Wang LM, Schick C, Richert R. Comparing
poly(methyl acrylate)/clay nanocomposites studied by ESR, calorimetric and dielectric polarization modes in viscous
XRD, and DSC. Macromolecules 2006;39:3304–11. 2-ethyl-1-hexanol. J Chem Phys 2007;126:104503.
[33] Mijovic J, Lee HK, Kenny J, Mays J. Dynamics in polymer- [51] Schick C, Dobbertin J, Potter M, Dehne H, Hensel A,
silicate nanocomposites as studied by dielectric relaxation Wurm A, et al. Separation of components of different
spectroscopy and dynamic mechanical spectroscopy. Mac- molecular mobility by calorimetry, dynamic mechanical and
romolecules 2006;39:2172–82. dielectric spectroscopy. J Therm Anal 1997;49(1):499–511.
[34] Giannelis EP, Krishnamoorti R, Manias E. Polymer-silicate [52] Dobbertin J, Hensel A, Schick C. Dielectric spectroscopy
nanocomposites: model systems for confined polymers and and calorimetry in the glass transition region of semicrys-
polymer brushes. Adv Polym Sci 1999;138:107–47. talline poly(ethylene terephthalate). J Therm Anal 1996;
[35] Lee DC, Jang LW. Preparation and characterization of 47(4):1027–40.
PMMA-clay hybrid composite by emulsion polymerization. [53] Xia H, Song M. Characteristic length of dynamic glass
J Appl Polym Sci 1996;61:1117–22. transition based on polymer/clay intercalated nanocompos-
[36] Privalko VP, Lipatov YS, Kercha YY. Calorimetric study of ites. Thermochim Acta 2005;429:1–5.
the phase boundary effect on oligo-ethylene glycol adi- [54] Suzuki H, Grebowicz J, Wunderlich B. Glass transition of
pate (Oega) properties*. Polym Sci USSR 1970;12(6): poly(oxymethylene). British Polym J 1985;17(1):1–3.
1520. [55] Wunderlich B. Reversible crystallization and the rigid-
[37] Mammeri F, Rozes L, Le Bourhis E, Sanchez C. Elabora- amorphous phase in semicrystalline macromolecules. Progr
tion and mechanical characterization of nanocomposites Polym Sci 2003;28(3):383–450.
thin films Part II. Correlation between structure and [56] Xu H, Cebe P. Heat capacity study of isotactic polystyrene:
mechanical properties of SiO2–PMMA hybrid materials. J dual reversible crystal melting and relaxation of rigid
Europ Ceram Soc 2006;26:267–72. amorphous fraction. Macromolecules 2004;37:2797–806.
[38] Vieweg S, Unger R, Hempel E, Donth E. Kinetic structure [57] Minakov AA, Mordvintsev DA, Tol R, Schick C. Melting
of glass transition in polymer interfaces between filler and and reorganization of the crystalline fraction and relaxation
SBR matrix. J Non-Cryst Solids 1998;235:470–5. of the rigid amorphous fraction of isotactic polystyrene on
[39] Kirst KU, Kremer F, Litvinov VM. Broad-band dielectric fast heating (30,000 K/min). Thermochim Acta 2006;442:
spectroscopy on the molecular dynamics of bulk and 25–30.
 A. Sargsyan et al. / European Polymer Journal 43 (2007) 3113–3127 3127

[58] Minakov AA, Mordvintsev DA, Schick C. Melting and [68] Ishida Y, Yamafuji K, Ito H, Takayanagi M. Effects of
reorganization of poly(ethylene terephthalate) on fast heat- degree of crystallinity upon dielectric behaviors in some
ing (1000 K/S). Polymer 2004;45(11):3755–63. aromatic polyesters. Kolloid Z Z Polym 1962;184(2):97–108.
[59] Bansal A, Yaang H, Li C, Cho K, Benicewicz BC, Kumar [69] Donth E. The size of cooperatively rearranging regions at
SK, et al. Quantitative equivalence between polymer nano- the glass transition. J Non-Cryst Solids 1982;53(3):325–30.
composites and thin polymer films. Nature Mater [70] Schick C, Donth E. Characteristic length of glass transition:
2005;4:693–8. experimental evidence. Phys Scr 1991;43(4):423–9.
[60] Dejong BHWS, Slaats PGG, Super HTJ, Veldman N, Spek [71] Schubnell M, Schawe JEK. Quantitative determination of
AL. Extended structures in crystalline phyllosilicates – silica the specific heat and the glass transition of moist samples by
ring-systems in lithium, rubidium, cesium, and cesium/ temperature modulated differential scanning calorimetry. Int
lithium phyllosilicate. J Non-Cryst Solids 1994;176(2–3): J Pharm 2001;217(1–2):173–81.
164–171. [72] Wunderlich B. The athas database on heat capacities of
[61] Garciatalegon J, Molin E, Vicente MA. Nature and char- polymers. Pure Appl Chem 1995;67(6):1019–26. see onA-
acteristics of 1/1 phyllosilicates from weathered granite, vailable from: http://athas.prz.edu.pl/databank/intro.html.
Central Spain. Clay Miner 1994;29(5):727–34. [73] Sargsyan A. PhD thesis: quantification of the immobilized
[62] Utracki LA, Sepehr M, Boccaleri E. Synthetic, layered fraction in polymer inorganic nanocomposites rostock,
nanoparticles for polymeric nanocomposites WNCO. Polym University of Rostock; 2007.
Adv Technol 2007;18(1):1–37. [74] Hodge IM. Enthalpy relaxation and recovery in amorphous

MACROMOLECULAR NANOTECHNOLOGY
[63] Choi YS, Choi MH, Wang KH, Kim SO, Kim YK, Chung materials [Review]. J Non-Cryst Solids 1994;169(3):211–66.
IJ. Synthesis of exfoliated PMMA/Na-MMT nanocompos- [75] Hay JN. The physical ageing of amorphous and crystalline
ites via soap-free emulsion polymerization. Macromolecules polymers. Pure Appl Chem 1995;67(11):1855–8.
2001;34(26):8978–85. [76] Lu H, Nutt S. Restricted relaxation in polymer nanocom-
[64] Schick C. Temperature modulated differential scanning posites near the glass transition. Macromolecules 2003;
calorimetry (TMDSC) – basics and applications to poly- 36(11):4010–6.
mers. In: Brown ME, Gallagher PK, editors. Amsterdam, [77] Lu H, Nutt S. Enthalpy relaxation of layered silicate-epoxy
Lausanne, New York, Oxford, Shannon, Singapore: Else- nanocomposite. Macromol Chem Phys 2003;204:1832–41.
vier; 2002. [78] Hempel E, Vieweg S, Huwe A, Otto K, Schick C, Donth E.
[65] Koh YP, McKenna GB, Simon SL. Calorimetric glass Characteristic length of glass transition from calorimetry in
transition temperature and absolute heat capacity of poly- different confinements. J Phys IV France 2000;10:79–82.
styrene ultrathin films. J Polym Sci B: Polym Phys [79] Adam G, Gibbs JH. On the temperature dependence of
2006;44:3518–27. cooperative relaxation properties in glass-forming liquids. J
[66] Sarge SM, Hemminger W, Gmelin E, Hohne GWH, Chem Phys 1965;43(1):139–46.
Cammenga HK, Eysel W. Metrologically based procedures [80] Pijpers MFJ, Mathot VBF, Goderis B, Scherrenberg R, van
for the temperature, heat and heat flow rate calibration of der Vegte E. High-speed calorimetry for the analysis of
DSC. J Therm Anal 1997;49:1125–34. kinetics of vitrification, crystallization and melting of mac-
[67] Neuenfeld S, Schick C. Verifying the symmetry of differen- romolecule. Macromolecules 2002;35(9):3601–13.
tial scanning calorimeters concerning heating and cooling [81] Scheidler P, Kob W, Binder K. Cooperative motion and
using liquid crystal secondary temperature standards. Ther- growing length scales in supercooled confined liquids.
mochim Acta 2006;446(1–2):55–65. Europhys Lett 2002;59(5):701–7.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.