Polymer 51 (2010) 4238e4242

Contents lists available at ScienceDirect

Polymer
journal homepage: www.elsevier.com/locate/polymer

Polypropylene-based nano-composite formation: Delamination of organically

modied layered ller via solid-state processing
Tomotaka Saito a, Masami Okamoto b, *
a
b

Research Laboratories, Toyota Boshoku Co. Ltd., 1-1, Toyoda-Cho, Kariya 448-8651, Japan
Advanced Polymeric Nanostructured Materials Engineering, Graduate School of Engineering, Toyota Technological Institute, Hisakata 2-12-1, Tempaku, Nagoya 468-8511, Japan

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 20 April 2010
Received in revised form
15 June 2010
Accepted 25 June 2010

Solid-state processing for the preparation of polypropylene (PP)-based nano-composites having nely
dispersed layered llers was conducted. The mixture of PP and organically modied layered ller (OMLF)
(95:5 wt./wt.) was subjected to the processing using alumina mortar heated 65
C, below Tm of PP (i.e., PP
is still at the solid-state), and ground for 8 h before melt compounding. On X-ray diffraction, the d(001)
peak of OMLF was broaden and peak position shifted slightly. The mixture prepared by solid-state
processing exhibited disorder and delaminated layer structure with the thickness of 3e7 nm into PP
matrix through TEM observations. On the contrary, nano-composite prepared by melt compounding at
180
C for 3 min (without solid-state processing) showed the large stacked silicate layers in the PP
matrix. Furthermore, instead of using alumina mortar, we carried out solid-state processing using
internal mixer. X-ray diffraction pattern and TEM observation exhibited similar results. The solid-state
processing led to delaminate of the silicate layers and attained the discrete dispersion.
2010 Elsevier Ltd. All rights reserved.

Keywords:
Delamination
Solid-state processing
Layered llers

1. Introduction
A signicant amount of work has already done on various
aspects of polymeric nano-composite containing organically
modied layered llers (OMLFs) [1]. However, complete delamination of OMLFs in continuous polymer matrix is still challenging
issue because it could not be satisfactorily attained. Gardolinski and
Lagaly [2] described a very distinctive denition of delamination
and exfoliation in aim to avoid the controversial use of these terms.
Exfoliation is dened as the decomposition of large aggregates into
smaller particles, whereas delamination denotes the process of
separation of the individual layers of the particles at the nanoscale.
To the best of our knowledge, so far, the complete delamination
is not feasible after melt intercalation with appropriate shear. Only
a few examples of this type can be found in the literature [3,4], but
many published photographs show very small regions in the melt
compound where partially exfoliation occurred [1]. Delamination
of stacked layered llers in polymeric nano-composite is the ultimate target for controlling better overall materials properties.
Thus, we are far from the goal of understanding the mechanisms
of the nano-structure control and the preparation of the

* Corresponding author. Fax: 81 (0) 52 809 1864.

E-mail address: [email protected] (M. Okamoto).
0032-3861/$ e see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2010.06.049

nano-composite with discrete dispersion of the nano-llers. From

this reason, a novel preparation method is currently in progress.
Some methods for the delamination of OMLFs were conducted by
using supercritical CO2 [5,6]. The effect of the supercritical CO2 (scCO2) fed to the tandem extruder on the dispersion of organically
modied montmorillonite (MMT) with different intercalants into
Nylon 6 matrix was examined [6]. In the absence of sc-CO2, pressure
improved the MMT-clay delamination by reducing the free volume
of the polymer and increasing the interaction between chains and
ultimately increasing the viscosity. Using sc-CO2 did not improve the
clay dispersion due to the decreasing the melt viscosity. Another
interesting approach for the delamination of OMLFs is an ultrasound
in the preparation of nano-composites. The effect of the in-situ
ultrasound on the polymer/MMT melt phase is reported [7].
An effective method to enhance the dispersion, intercalation
and exfoliation of OMLFs in thermoplastic-based nano-composites
is reported. The same experiment was done in another report for
polypropylene (PP)-base nano-composite preparation [8]. The
maximum power output and frequency of the ultrasonic generator
are 300W and 20 kHz, respectively. They described the ne
dispersion of silicate layers in PP matrix after ultrasonic treatment
(100W). However, the ultrasonic oscillations exhibited a little effect
on the delamination of OMLFs as revealed by transmission electron
microscope (TEM) observation. Thus, the compounding with an
assist from sc-CO2 uids and ultra-sonication did not improve the

T. Saito, M. Okamoto / Polymer 51 (2010) 4238e4242

state of the nano-ller dispersion once a critical morphology was

established. That is, the dispersion of the nano-ller in the polymer
matrix is governed by judicious choice of OMLF.
Although the intercalation technology of the polymer melt is
developed along with the current industrial process, such as
extrusion and injection molding, we have to develop more innovative compounding process, especially in the preparation of the
nano-composites possessing discrete dispersion of the nano-llers.
In this regard, we have reported solid-state processing of poly(pphenylenesulde) (PPS)-based nano-composites to delaminate the
stacked, layered ller in the polymer matrix [9,10]. The mixture of
PPS and organically modied layered ller (OMLF) (95:5 wt./wt.)
was subjected to the processing using thermostatted hot-press at
ambient temperature and 150
C, below Tm of PPS (i.e., PPS matrix is
still at the solid-state), and applying pressures of 7, 14 and 33 MPa
for 30s. The mixture exhibited disorder and delaminated layer
structure with the thickness of 40e80 nm into PPS matrix. The
solid-state processing led to delaminate of the silicate layers and
attained the discrete dispersion. Similarly, Wang et al. [11e13]
reported the exfoliation of talc llers by solid-state shear processing using pan-type mill, to prepare PP/talc nano-composites, in
which the delamination of talc llers was not achieved in the nanocomposite as revealed by TEM images.
Wakabayashi et al. [14] demonstrated that a continuous scalable
solid-state shear pulverization (SSSP) could result in well-dispersed
unmodied graphite in PP, leading to a 100% increase in modulus as
compared with neat PP. High shear and compressive forces result in
repeated fragmentation and fusion of polymer in the solid-state
producing excellent mixing and dispersion of nano-llers in the
nano-composites [15].
The solid-state shear processing may be an innovative technique
to delaminate the layered llers in overcoming the pressure drop
(Dp) within the nano-galleries [10]. Therefore, successful delamination of OMLFs could broaden the scope of application of this
procedure. The objective of this work is to gain insight into the
solid-state processing for the preparation of nano-composites
based on PP having nely dispersed layered llers.
2. Experimental
Organically modied layered ller (OMLF) used in this study
was montmorillonite (MMT, Hojun Co., Ltd.) intercalated with nhexadecyl tri-n-butyl phosphonium (C16TBP) cation. A polypropylene (PP) ne powder (average particle size w5 mm,
Tm 151
C), purchased from Seishin enterprise Co., Ltd. The
homogeneous mixture of PP and OMLF in the weight ratio 95/5 was
prepared. We have ground the mixture with pestle and mortar for
8 h at 65
C (well below Tm of PP matrix powder). We call the
process solid-state processing because PP matrix is solid-state.
After solid-state processing, the mixture of PP and OMLF was melt
compounded using miniature mixer of gram scale (MINI-MAX
Molder CS-183, Custom Scientic Instruments Inc.) and to convert
into sheet for the characterization.
The nano-structure analyses of wide-angle X-ray diffraction
(WAXD) were performed for the PP/OMLF mixture and corresponding nano-composite sheets by using MXlabo X-ray diffractometer (MAC Science Co., CuKa radiation, wavelength
l 0.154 nm) and Ultima IV X-ray diffractometer (Rigaku Co., CuKa
radiation, wavelength l 0.154 nm), and transmission electron
microscopy (TEM) were carried out using the same apparatus as in
the previous articles [9]. To investigate the micro-scale morphology
of the nano-composites, we also used polarizing optical microscope
(POM). The detail of POM observation is in our previous paper [10].
For analyzing the features of the micrographs, we carried out fast
Fourier transform (FFT) analysis on digitally saved images of POM

4239

Fig. 1. WAXD patterns of (a) mixture of PP and OMLF (95:5 wt./wt.) before solid-state
processing, (b), (c) and (d) processed mixture of PP/OMLF at 65
C for 1, 2 and 8 h,
respectively. Arrows in each panel indicate d(001) peak of OMLF. The strong diffraction
peaks at 2Q 15e25
are assigned to the monoclinic (a-form)/triclinic (g-form) unit
cells of the crystallized PP.

and TEM micrographs using the commercial image analysis software (Ultimage, Graftek, France) [16], which allowed to provide us
information equivalent to the scattering analyses.
For comparison, the PP-based nano-composite preparation was
conducted via conventional melt compounding operated at 180
C

Fig. 2. Time variation of d(001) and corresponding intensity during processing. Intensity is normalized by a(110)/g(111) peak.

4240

T. Saito, M. Okamoto / Polymer 51 (2010) 4238e4242

Fig. 3. POM photographs of the mixture (a) unprocessed sample and (b) after solidstate processing for 8 h. Both micrographs were taken at 180
C just after annealing for
30 s. The inset in each image is a computed FFT spectrum of the micrograph.

for 3 min [17]. The ejected samples were dried under vacuum at
80
C for 6 h to remove water. The dried nano-composite was then
converted into sheets with a thickness of 0.7e2 mm by pressing
with z1.5 MPa at 180
C for 1 min using a laboratory hot-press
(Mini Test Press-10, Toyo Seiki Seisaku-sho, Ltd.). Furthermore,
instead of using alumina mortar, we carried out solid-state processing using internal mixer (Laboplastomill V4C150, Toyo Seiki
Seisaku-sho, Ltd.) with a rotating speed of 50 rpm at 50
C for 5 h.
3. Results and discussion
3.1. Variation of WAXD proles
WAXD patterns for the mixture powder of PP and OMLF (95:5
wt./wt.) are presented in Fig. 1(a). The mean interlayer spacing of
the (001) plane (d(001)) for the MMT that was modied by C16TBP
[MMT-C16TBP] obtained by WAXD measurements is 2.26 nm
(diffraction angle, 2Q 3.90
). Actually, there is a large anisotropy
of the stacked silicate layers. The size of the some of the stacked
silicate layers appears to reach about 50e100 nm in lengths and the
interlayer distance as revealed by TEM observation is about 3 nm
(see Fig. 4(a and c)). This value is virtually same as compared to
WAXD data.
After solid-state processing for 2 h, a small remnant shoulder is
observed around 2Q y 4.00
(Fig. 1(c)). In the following processing
from 3 to 8 h, the peak was almost featureless diffraction, only

Fig. 4. Bright led TEM images of (a) unprocessed sample, (b) and (c) sample prepared
by solid-state processing for 8 h. The dark entities are the cross section and/or face of
intercalated-and-stacked silicate layers, and the bright areas are the matrix. The inset
in (a) and (c) are a computed FFT spectrum of the micrograph.

T. Saito, M. Okamoto / Polymer 51 (2010) 4238e4242

4241

Table 1
Form factors of two nano-composites obtained from TEM and FFT analyses.

Fig. 5. One-dimensional scattering pattern of processed PP/OMLF for 8 h obtained

from FFT analysis. The solid line was drawn by the power low of I(q) y q1 at the q
range of 0.02e0.2 nm1.

exhibiting the monoclinic (a-form)/triclinic (g-form) unit cells of

the crystallized PP [18].
Fig. 2 shows the time variation of d(001) and corresponding
intensity during processing. With increasing processing time the
intensity from d(001) gradually decreases with peak shift to lower
angle (layer shrinkage). Beyond 4 h, the layer spacing nally
reached a constant value (w2.05 nm) possibly due to the change of
the interdigitated layer structure of the MMT-C16TBP [19]. The
structure may suggest that the different orientation angle could
adopt, giving a decreasing of the basal spacing during processing.
3.2. Morphology
To elucidate the morphologies before and after solid-state processing, we conducted POM observation at 180
C Fig. 3(a) shows
the POM photograph of the mixtures prepared by melt compounding and annealing at 180
C for 30 s. It is clear from the POM

Parameters

TEM image

FFT with DB theory

LMMT/nm
tMMT/nm
xMMT/nm

67  6
5.8  0.5
45  2

e
3.7
157

photographs that stacked-and-agglomerated structure of layers is

evident in the melt compounded sample, whereas a good dispersion appears in the processed sample for 8 h (Fig. 3(b)). The FFT
pattern shows weak scattering with isotropy (halo) as compared
with that of unprocessed sample (i.e., melt compounded sample).
This indicates that the particles size of the dispersed nano-ller
becomes smaller during solid-state processing. The dispersion state
in the nanometer scale was directly observed via TEM analysis.
Fig. 4 shows the results of TEM bright eld images and their FFT
patterns of the melt compounded mixtures corresponding to the
POM experiments, in which dark entities are the cross section of the
layered nano-llers. The large agglomerated tactoids of about 3 mm
thickness are seen in Fig. 4(a) (unprocessed sample). On the other
hand, in Fig. 4(b), nanometer sized thickness layers were straggled
in the observation area. Fig. 4(c) shows disorder and delaminated
silicate layer structure with the thickness of 3e7 nm and length of
50e200 nm (average thickness of 5.8 nm and length of 67 nm). This
is a very interesting observation of the discrete silicate layers.
Fig. 5 shows the one-dimensional scattering pattern obtained from
FFT analysis. Here, the scattering intensity (I(q)) is shown as a function
of the magnitude of the scattering vector (q). We found a broad
shoulder around q w 0.02 nm1 and small remnant peaks in the q
range from 0.04 to 0.09 nm1. The broad ness is due to the fact that the
dispersed OMLF particles are rather disordered and short-range
ordered. The decay of intensity in the q range from 0.02 to 0.1 nm1can
be expected by a power law in the form I(q) y q1, suggesting the form
factor is the randomly oriented rod. Apparently, this is consistent with
the shape of the dispersed OMLF particles in the TEM image (Fig. 4(c)).
We estimated the form factors obtained from TEM images and
FFT analysis combined with DebyeeBueche (DB) theory, i.e., average
value of the particle length (LMMT), thickness (tMMT), of the dispersed
particles and the correlation length (xMMT) between the particles.
The details of the evaluation were described in our previous paper
[20]. The DebyeeBueche equation [21] is applicable in isotropic
dense system [22]. The detail procedure for the DebyeeBueche plots
was described in our previous paper [23]. To estimate the value of
xMMT, we employed the DebyeeBueche equation:

Iq

1=2


3 1=2
8phr2 ix


2
1 x q2

(1)

where x is the long-range structure correlation distance, q is the

magnitude of scattering vector, I(q) is the intensity of the scattered light
at q and hr2 i is the mean-square uctuation of the refractive index. [24].
Fig. 6 shows the DebyeeBueche plot for FFT analysis. The plot is
observed to be linear. From the slope and the intercept, we can
estimate x (slope/intercept)1/2. Once the value of x (3.65 nm) is
given, other morphological parameters are obtained by

xMMT
tMMT

Fig. 6. DebyeeBueche plot for FFT analysis. The solid line is calculated by linear
regression. The dashed line indicates the intercept of the plot.

x
fMMT
x
1  fMMT

(2)
(3)

where fMMT is the volume fraction of the dispersed phase (MMT

particles). The details of the evaluation were described in our
previous paper [23]. The results are presented in Table 1.

4242

T. Saito, M. Okamoto / Polymer 51 (2010) 4238e4242

To understand the kinetics of the solid-state processing, we

attempt to apply a phenomenological formulation for the breakup
of mineral particles (Rittingers law):

 
1
dE b n dD
D

(4)

where E is the energy for breakup, D is the mean diameter of

particles and the parameter b and n are the breakup coefcients,
which depend on processing condition and materials [25]. The
mixing torque T is constant throughout the processing so that the
input energy E is proportional to residence time t, i.e., E fT,t.
Hence, eq. (4) can be rewritten as

1
T$tw n1
D

(5)

The effect of the different applying torque and residence time on

the delamination behavior of the OMLF will be claried shortly [26].
4. Conclusions
Fig. 7. Bright led TEM image of PP/OMLF (95:5 wt./wt.) prepared by internal mixer
with a rotating speed of 50 rpm at 50
C for 5 h. The dark entities are the cross section
and/or face of intercalated-and-stacked silicate layers, and the bright areas are the
matrix.

For the sample prepared by annealing without solid-state

processing, LMMT and tMMT are in the range of (6e10) mm and
(1e3) mm, respectively as revealed by TEM image (Fig. 4(a)). On
the other hand, PP/OMLF prepared by solid-state processing
exhibits a small value of LMMT (67  6 nm) with almost double of
stacking of the silicate layers (tMMT 5.8  0.5 nm). xMMT value of
PP/OMLF with processing (45  2 nm) suggests that a more
uniform dispersion of the silicate layers seems to be attained due
to the solid-state processing. The absolute value of xMMT shows
much lower as compared with that of DebyeeBueche plot. The
reason is not obvious at present. However, the value of tMMT is
virtually same as compared with that of DebyeeBueche plot. After
solid-state processing, a stacked silicate layers contain two
discrete silicate layers with ner dispersion. This is a unique
observation of the discrete silicate layers. The intercalants into
nano-galleries act as a lubricant at processing temperature. Since
the melting temperature of the intercalant (C16TBP) into the
nano-galleries was 23.5
C, we can observe the effect of layer
delamination at 65
C.
Furthermore, we conducted solid-state processing by using
internal mixer instead of alumina mortar. For processing with
a rotating speed of 50 rpm at 50
C for 5 h, the dispersed
morphology of the silicate layers exhibits same trend with applying
pressure (corresponding to the shear stress of 34 MPa) (see Fig. 7).
The disordered and delaminated silicate layer structure is observed
in TEM image. LMMT and tMMT are in the range of (372  38) nm and
(33.2  3.8) nm, respectively. From these facts, the solid-state
processing is extremely effective method to collapse the stacked
structure in overcoming the pressure drop (Dp w 24 MPa [10])
within the nano-galleries.

We have described a novel and feasible method for the nanoscale control of the dispersed layered llers via solid-state processing by using common experimental tools. This processing led to
delaminate of the silicate layers and attained the discrete dispersion. This approach can be extended to prepare polymeric nanocomposites with delamination of the nano-llers in overcoming
the pressure drop within the nano-galleries.
References
[1] Sinha Ray S, Okamoto M. Prog Polym Sci 2003;28:1539.
[2] Gardolinski JEFC, Lagaly G. Clay Miner 2005;40:547.
[3] Usuki A, Kawasumi M, Kojima Y, Okada A, Kurauchi T, Kamigaito O. J Mater
Res 1993;8:1179.
[4] Katoh Y, Okamoto M. Polymer 2009;50:4718.
[5] Yang K, Ozisik R. Polymer 2006;47:2849.
[6] Bellair RJ, Manitiu M, Gulari E, Kannan RM. J Polym Sci Polym Phys
2010;48:823.
[7] Lee EC, Mielewski DF, Baird RJ. Polym Eng Sci 2004;44:1773.
[8] Zhao L, Li J, Guo S, Du Q. Polymer 2006;47:2460.
[9] Saito T, Okamoto M, Hiroi R, Yamamoto M, Shiroi T. Macromole Rappid
Commun 2006;27:1472.
[10] Saito T, Okamoto M, Hiroi R, Yamamoto M, Shiroi T. Polymer 2007;48:4143.
[11] Shao W, Wang Q, Li K. Polym Eng Sci 2005;45:451.
[12] Shao W, Wang Q, Ma H. Polym Int 2005;54:336.
[13] Shao W, Wang Q, Wang F, Chen Y. J Polym Sci Polym Phys 2003;44:249.
[14] Wakabayashi K, Pierre C, Dikin DA, Ruoff RS, Ramanathan T, Brinson IC, et al.
Macromolecules 2008;41:1905.
[15] Masuda J, Torkelson JM. Macromolecules 2008;41:5974.
[16] Kim YH, Okamoto M, Kotaka T. Macromolecules 2000;33:8114.
[17] Saito T, Okamoto M, Hiroi R, Yamamoto M, Shiroi T. Macromol Mater Eng
2006;291:1367.
[18] Meille S, Bruckner S, Porzio W. Macromolecules 1990;23:4114.
[19] Yoshida O, Okamoto M. Macromol Rapid Commun 2006;27:751.
[20] Sinha Ray S, Yamada K, Okamoto M, Ogami A, Ueda K. Chem Mater
2003;15:1456.
[21] Debye P, Bueche AM. J Appl Phys 1949;20:518.
[22] Hsiao BS, Stein RS, Deutscher K, Winter HH. J Polym Phys 1990;28:1571.
[23] Okamoto M, Inoue T. Polym Eng Sci 1993;33:175.
[24] Okamoto M, Inoue T. Polymer 1995;36:2736.
[25] Bond FC. Mining Eng 1952;4:484.
[26] Saito T, Okamoto M, in preparation.