Composites Science and Technology 63 (2003) 331337

www.elsevier.com/locate/compscitech

Morphology, thermal and mechanical behavior of polyamide

6/layered-silicate nanocomposites
T.X. Liu*, Z.H. Liu, K.X. Ma, L. Shen, K.Y. Zeng, C.B. He
Institute of Materials Research and Engineering, 3 Research Link, Singapore 117602, Singapore

Abstract
The microstructure, morphology, thermal and mechanical properties of polyamide 6/layered-silicate nanocomposites (PLSN)
have been studied using X-ray diraction (XRD), tapping-mode atomic force microscopy (AFM), dierential scanning calorimetry
(DSC) and tensile tests. The XRD and AFM results indicate that the organically-modied layered silicates are intercalated in the
polyamide 6 (PA6) matrix. Also, the crystallization and thermal behavior of PA6 are inuenced by the addition of layered silicates
into polymer matrix. A silicate-induced crystal transformation from the a-form to the g-form of PA6 is conrmed by XRD and
DSC, i.e., the formation of g crystal is enhanced by the presence of silicates. In comparison to PA6, the thermal property of PLSN
has been investigated using DSC. The possible origins of a new silicate-induced endothermic peak are discussed. Tensile tests show
that the tensile modulus and yield strength increase while the strain-at-yield decreases with increasing clay loading.
# 2002 Elsevier Science Ltd. All rights reserved.
Keywords: A. Polymers; A. Nanostructures; B. Thermal properties; B. Mechanical properties

1. Introduction
Compounding polymers with inorganic materials has
long been the interesting topic of scientic research and
industrial applications because inorganic materials can
be employed to improve the mechanical properties and
high temperature durability. It can also provide
enhanced barrier properties and reduce ammability of
the resultant polymer composites [1,2]. Among various
inorganic llers, bers and anisotropic particles with
large aspect ratio (length/diameter ratio) have proven to
be particularly eective in polymer matrix reinforcement, such as mica-type clay or layered silicates (like
montmorillonite, hectorite and saponite) [3]. Usually,
the dispersion of clay particles in a polymer matrix can
result in the formation of three types of composite
materials [1]. The rst type is conventional phase separated composites in which the polymer and the inorganic
host remain immiscible resulting in poor mechanical
properties of the composite material. The second type is
intercalated polymer-clay nanocomposites, which are
formed by the insertion of one or more molecular chains
* Corresponding author. Tel.: +65-68748594; fax: +65-67744657.
E-mail address: [email protected] (T.X. Liu).

of polymer into the interlayer or gallery space. The last

type is exfoliated or delaminated polymer-clay nanocomposites which are formed when the silicate nanolayers are individually dispersed in the continuous
polymer matrix. Exfoliated polymer-clay nanocomposites
are especially desirable for improved properties because
of the large aspect ratio and homogeneous dispersion of
clay and huge interfacial area between polymer and clay
[4].
It has been well documented that the extent of dispersion (intercalation and exfoliation) of organoclay
strongly inuences the properties of the polymer/
layered-silicate nanocomposites [5,6]. Usually, two
complementary characterization techniques used in
polymer/clay nanocomposites are X-ray diraction
(XRD) (which is used to identify intercalated structures)
and transmission electron microscopy (TEM, which is
used to characterize the nanocomposite morphology)
[5]. However, it has been well known that there are disadvantages such as radiation damage and tedious preparation of specimen using TEM. Interestingly, in the
literature there has been only a few reports on the morphology of clay/nanocomposites by non-destructive
atomic force microscopy (AFM) [7,8]. In contrast to
conventional electron microscopies, AFM does not

0266-3538/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0266-3538(02)00226-9

332

T.X. Liu et al. / Composites Science and Technology 63 (2003) 331337

require a conductive coating as for scanning electron

microscopy (SEM) or staining as in the case of TEM.
Additionally, the presence of modied particles (such
as rubber/inorganic ller particles) directly inuences
the morphology, crystallization behavior and consequently the melting behavior of the thermoplastic polymer matrix. Although intensive research eorts have
been recently devoted to the eect of clay on crystallization behavior of polyamide 6 matrix [912], to our
knowledge, the eect of organoclay on melting behavior
of PA6 has not been well identied.
The aim of the present report is to investigate the
structure and morphology of PA6/layered-silicate
nanocomposites using XRD and AFM and to understand the role of nanoller particles in the polymer
matrix. Subsequently, we focus on comparative studies
of the thermal and mechanical properties of neat PA6
and PA6/layered-silicate nanocomposites by DSC and
tensile tests, respectively.

2. Experimental section
The polyamide 6 (PA6)/layered-silicate nanocomposites were prepared with dierent clay loadings (from
2.5 to 10 wt.%) by melt compounding using Brabender
twin-screw extruder. The temperature of the extruder
was maintained at 220, 250, 250 and 230  C from
hopper to die, respectively. The screw speed was maintained at 150 rpm. The organoclay (Nanomer I.30TC)
was purchased from Nanocor. The obtained nanocomposite containing 4 wt.% clay was mainly used in
this study due to greatly enhanced performance at low
loadings, generally 26 wt.% [13]. Before using, the
nanocomposite samples were dried at 90  C for at least
12 h in vacuum oven to remove the absorbed water due
to the high moisture absorption of polyamides. After
that, the samples were melted in metal moulder at 260  C
for 5 min, then pressed into sheets and naturally cooled
down to room temperature. The obtained lms (with
thickness of ca. 1 mm) were used for subsequent X-ray
diraction (XRD) and DSC measurements as well as
AFM observation.
X-ray diraction was performed with a Philips Analytical X-ray (XPert) using nickel-ltered Cu-Ka1
radiation source (l=0.1542 nm) under a voltage of 45
kV and a current of 40 mA. The XRD patterns were
recorded with a step size of 0.05 from 2=2 to 30 .
AFM observations were carried out on an AutoProbe
CP Research model, supplied by ThermoMicroscopes
operating in ambient temperature. Both height (topography) and phase images were obtained simultaneously
under tapping mode using a 100 mm scanner and
AutoProbe mounted Si tip with force constant of 3.2 N/m
and frequency of 90 kHz, respectively. In this intermittent contact mode, the tip was oscillated vertically at

its resonance frequency. The tip-sample interaction

(e.g., attractive and repulsive forces) causes a change in
the amplitude, the phase and the resonance frequency of
the vibrating cantilever. The spatial variations of these
changes can be presented as height or phase shift images
[14]. The images presented here were processed by attening with IP 1.3 Imaging Processing software, provided
by the manufacturer.
DSC experiments were performed in a DSC-2920 from
TA Instruments coupled with a TA-2000 control system.
The samples were heated or cooled at a scanning rate of
10  C/min under nitrogen atmosphere in order to diminish oxidation. For the study of eect of crystallization
temperature (Tc) on the melting behavior of PA6 and
PA6/layered-silicate nanocomposites (PLSN), the samples were isothermally melt-crystallized at dierent
temperatures for 6 h before subsequent heating in
DSC.
Tensile tests were conducted according to ASTM D628
using an Instron testing machine. Youngs modulus,
yield strength and extension at yield were accurately
determined by using an extensometer at a crosshead
speed of 5 mm/min.

3. Results and discussion

3.1. Structure and morphology from XRD and AFM
The periodic intercalated structures of inorganic
crystalline layers can be quantied through powder
X-ray diraction. Fig. 1 illustrates the comparison of
the XRD patterns of pristine clay, melted-annealed neat
PA6 and PLSN. The following structural features can
be seen: (i) for neat PA6, two main reections can be
observed at about 2=20 and 23.7 , which are attributed to the a100 and a002/202 crystal planes, respectively [11,1517]. Therefore, it is evident that the a-form
population is the dominant crystalline phase for the
melt-annealed neat PA6. (ii) Comparatively, besides the
two main reections as in neat PA6, the other three
reections are also observed at around 2=4, 10.7 and
21.4 for the melt-annealed PLSN sample. The reection peaks at 2=10.7 and 21.4 are associated with
g020 and g001 crystal planes of PA6, respectively
[11,1517]. Hence, the addition of silicates induces the
formation of g-form crystals of PA6, probably suggesting that the formation of the g-form crystals of PA6 is a
heterogeneous nucleation mechanism. Moreover, the
reection intensity of a100 (at 2=20 ) decreases but
the intensity of a002/202 reection (at 2=23.7 )
increases, indicating that a signicant change in the
preferential (growth) crystal planes of PA6 occurs upon
addition of silicates into polymer matrix. (iii) The basal
plane of the pristine clay is at 2=6.9 corresponding to
a host gallery d-spacing=1.3 nm. (iv) For PLSN, the

T.X. Liu et al. / Composites Science and Technology 63 (2003) 331337

Fig. 1. XRD patterns of clay, melt-annealed PA6 and PLSN samples.

broad reection peak around 2=4 (d-spacing=2.2

nm) should be assigned to the intercalated structure of
nanocomposites, suggesting that the host galleries of the
pristine clay are eectively swollen by the insertion of
polymer chains (from 1.3 to 2.2 nm). And, the broadening of the peak may indicate that partial exfoliation
has occurred. Accordingly, the XRD results indicate
that the intercalation of polymeric layers in between the
silicate layers occurs probably together with some exfoliation in the PLSN samples.
AFM is used to directly view the structure of nanocomposites, with the emphasis on the dispersion of the
layered silicate llers in PA6 matrix. Here, the tapping
mode AFM is applied to get sucient mechanical (intrinsic) contrast between the two components (hard inorganic
silicate ller and soft polymer matrix). Several typical
AFM height and phase images are shown in Fig. 2. Under
lower magnication [Fig. 2(A)], it can be seen that the
silicate particles (visible as black streaks in phase image) in
the nanocomposites are nely dispersed in the PA6
matrix with an average size (length) from 0.1 to 1 mm.
No signicant exfoliated clay platelets are detected as
was consistent with the XRD measurements. Phase
image under higher magnication [Fig. 2(B)] displays
the existence of intercalated structures of layered clay in
PA6 matrix.
3.2. Thermal behavior of PA6 and PLSN by DSC: a
comparative study
As discussed above, the addition of silicates into PA6
signicantly inuences the crystallization behavior of

333

polymer matrix. Accordingly, this would be reected

from the thermal behavior of PLSN. In this section, the
thermal behavior of PLSN was studied compared to
that of neat PA6 using DSC. For clarity, all the DSC
curves shown here are shifted vertically.
Fig. 3(A) shows the rst and second heating and
cooling DSC curves of neat PA6. From the rst run,
upon heating an endothermic melting peak (noted as
Tm3 here) is observed at 227  C which is associated with
the a-form crystals of PA6 [9,11,17]; and an exothermic
crystallization peak (of a-form) is detected at 186  C
upon cooling. On the second heating curve, the melting
peak of a-form crystals slightly shifts to a lower temperature 223  C due to the melting of imperfect a-form
crystals formed during cooling process of the rst run.
Besides Tm3, another lower endothermic peak (Tm2) is
observed around 214  C as a shoulder of Tm3, which is
related to the melting of g-form crystals of PA6 formed
during the rst cooling process, due to the less thermal
stability of g-modication of PA6 [9,11,18]. The second
cooling curve is almost the same as the rst one. The
eect of addition of silicate particles on thermal behavior of PLSN is presented in Fig. 3(B). Clearly, on
both of rst and second heating curves the melting
peaks (Tm2) of g-form crystals are greatly enhanced.
And, on the cooling curves, two crystallization peaks
are observed at 183 and 189  C, corresponding to
g-form and a-form crystals of PA6, respectively. Most
interestingly, on the heating curves a new endothermic peak (Tm4) is detected at about 237  C, which
has not yet reported in the literature. The possible
origins on this abnormal melting peak will be discussed later.
Fig. 4(A) and (B) illustrates the eect of crystallization or annealing temperature (Tc) on the thermal
behavior of neat PA6 and PLSN melt-crystallized at
dierent temperatures, respectively. It can be observed
that: (i) In the main melting region of PA6 and PLSN,
the positions of Tm2 and Tm3 are relatively constant
with the increase of Tc. It seems that the formation of gcrystals is preferred at low temperature (e.g. 120  C),
whereas the a-form population is dominant at higher
temperatures [see Fig. 4(A)]. (ii) For PA6, an exothermic peak at about 195  C can be observed for
Tc=120  C, probably indicating that some uncrystallizable materials may undergo a reorganization or recrystallization during the heating process in DSC. Such a
recrystallization process during heating is signicantly
prohibited at higher Tc and in PLSN system [Fig. 4(A)
and (B)]. (iii) Another endothermic peak (indicated by
arrows), the so-called annealing peak (Tm1), can be
induced by isothermal crystallization or annealing process for both PA6 and PLSN systems. The position of
the Tm1 shifts towards higher temperatures (always
about 10  C above Tc) accompanied by an increasing
magnitude as Tc increases. It should be noted that this

334

T.X. Liu et al. / Composites Science and Technology 63 (2003) 331337

Fig. 2. Tapping-mode AFM height (left) and phase (right) images of melt-annealed PLSN lm. The elds of view are 1010 mm (A) and 11 mm
(B).

annealing-induced endotherm is a universal phenomenon in most semicrystalline thermoplastics, such as

poly(aryl ether ether ketone), poly(phenylene sulde),
poly(ethylene terephthalate) [19], isotactic polystyrene
[2022]. The possible physical origins of this behavior
are controversial in the literature. In our previous
report, we tentatively ascribed it to an enthalpic relaxation process of an interphase between the crystalline and
amorphous phases [2022]. Some preliminary dynamic
mechanical analysis (DMA) and modulated DSC results
on PA6 and PLSN also support this speculation by
demonstrating another glass transition at about 10  C
above Tc [23]. More detailed work on Ta will be given in
a separate report. (iv) At an even higher annealing temperature (e.g. Tc=210  C), all the melting peaks merge
into one at about 220  C. The abnormal Tm4 is probably
overlapped or it disappears at a higher temperature
[Fig. 4(B)].

Generally, compared to PA6 the most noticeable difference in DSC curves for PLSN is the appearance of
the new higher endothermic peak (Tm4) around 237  C
[see Figs. 3(B) and 4(B)]. Very evidently, such an
abnormal endothermic peak is closely related to the
addition of silicates into polymer matrix. So far, the
exact origin on this new melting peak is still unknown.
The possible origin is speculated as follows. (1) Organoclay itself. But clearly, this is not true because no
thermal transitions can be detected around this melting
region from the DSC curve of clay (not shown here). (2)
Organoclay-induced new ordered or crystalline structure. From the XRD and high-temperature XRD patterns (not shown here), however, no new crystalline
structure is identied. (3) Reorganized population
formed during heating process. From the heating rate
eect on DSC curves (not shown here), this assumption
is not so convincing because it is dicult to conrm the

T.X. Liu et al. / Composites Science and Technology 63 (2003) 331337

335

Fig. 3. DSC heating and cooling curves for melt-annealed PA6 (A) and PLSN (B). Scanning rate: 10  C/min.

Fig. 4. DSC heating curves for PA6 (A) and PLSN (B) melt-crystallized at indicated temperatures for 6 h. Scanning rate: 10  C/min.

existence of Tm4 at higher heating rates, owing to its proximity to Tm2. (4) Melting of crystals or lamellae of PA6
formed in conned environment between the intercalated
layers of organoclay. (5) Other reasons not listed above.
The new melting peak (Tm4) around 237  C reminds
us of some other research work on layered-silicate/
polymer nanocomposites conducted in other groups.
Most recently, it has been documented that the statics
and dynamics of polymers within the host galleries are

dramatically dierent from those observed in the bulk,

not only due to the connement of the polymer chains but
also due to specic polymersurface interaction, normally
not observed in the bulk [1,24,25]. Furthermore, one of
the most striking results (on layered-silicate/polyamide12 nanocomposites) from TEM morphological observation by Kim et al. directly shows that the ne lamellae of
polyamide-12 are oriented with their planes perpendicular
to the injection molding direction owing to nucleation

336

T.X. Liu et al. / Composites Science and Technology 63 (2003) 331337

Fig. 5. Tensile modulus (A), yield strength and strain at yield (B) as a function of clay loading (wt.%).

at the interface between layered silicate and polymer

matrix [26]. Interestingly, the orientation of the polymer
(PA6) lamellae relative to the organoclay sheets has
been also conrmed by small-angle X-ray scattering
(SAXS) [12]. Thus, it can be expected that the intercalated silicate sheets impose stresses on the PA6
lamellae which are tethered on the host layers by strong
(interfacial) ionic interaction. As a consequence, a
superheating eect could be observed as these constrained lamellae are heated in DSC. We believe that the
connement eect on crystallization and/or the orientation of lamellae are reected in the complexity of the
melting behavior in PLSN system. Therefore, the most
possible physical origin of this abnormal endothermic
peak is probably related to the (superheating) melting of
PA6 lamellae formed in the microenvironments of
organoclay particles in the intercalated nanocomposites.
A preliminary result supports our argument in comparison to the case in highly conned intercalated systems,
upon exfoliation of the organoclay in polymer matrix
such an interesting abnormal melting peak becomes
much less or even disappears [23]. The exact origins of
these new melting peaks are being investigated.
3.3. Mechanical properties of PA6 and PLSN
Fig. 5 shows the eect of clay concentration on the
tensile mechanical property of PLSN. Clearly, the addition of organoclay signicantly leads to substantial
improvement in stiness of PA6 [Fig. 5(A)]. A similar
increasing trend with respect to level of organoclay
content is also evident in the yield strength, while the
strain at yield decreases with increasing clay loading
[Fig. 5(B)]. The enhancing eect of organoclay on the
stiness and strength is probably due to the incorporation

of clay platelets with PA6 matrix. In addition, the continued decrease in elongation with clay concentration
indicates that the resulting nanocomposites are very
brittle. Currently, the research work on embrittling,
fracture and toughening mechanisms of PA6/clay
nanocomposites is still in progress in our laboratory.

4. Conclusions
We have investigated the microstructure and morphology of the PA6/layered-silicate nanocomposites, as
well as the phase transformation between a-form and
g-form crystals of PA6 due to the introduction of inorganic llers into polymer matrix. XRD and DSC data
show that the loading of silicates has a heterophase
nucleation eect which is favorable for the formation of
g-form crystals of PA6. Through the comparative DSC
studies, the eect of addition of organoclay on the
thermal behavior of neat PA6 and its layered-silicate
nanocomposites (PLSN) has been studied. For the
nanocomposites, DSC results reveal a new high melting
transition around 237  C, probably suggesting that the
crystallization of polymer chains under a conned
environment is closely associated with this abnormal
melting peak. The tensile tests show that the modulus
and yield strength of PLSN increase with increasing clay
concentration, while the extension at yield decreases
with increasing clay loading.

Acknowledgements
We are thankful for the nancial funding from the US
Air Force Oce of Scientic Research (AFOSR). The

T.X. Liu et al. / Composites Science and Technology 63 (2003) 331337

helpful discussion with Professor T. S. Chung (Department of Chemical and Environmental Engineering, The
National University of Singapore) is also appreciated.

References
[1] Giannelis EP, Krishnamoorti R, Manias E. Polymer-silicate
nanocomposites: model system for conned polymers and polymer brushes. Adv Polym Sci 1999;138:10747.
[2] Kojima Y, Usuki A, Kawasumi M, Okada A, Fukushima Y,
Kurauchi T, et al. Mechanical properties of nylon 6-clay hybrid.
J Mater Res 1993;8:11859.
[3] Cho JW, Paul DR. Nylon 6 nanocomposites by melt compounding. Polymer 2001;42:108394.
[4] Fu X, Qutubuddin S. Polymer-clay nanocomposites: exfoliation
of organophilic montmorillonite nanolayers in polystyrene.
Polymer 2001;42:80713.
[5] Alexandre M, Dubois P. Polymer-layered silicate nanocomposites:
preparation, properties and uses of a new class of materials.
Mater Sci Eng Rep 2000;28:163.
[6] Zanetti M, Lomakin S, Camino G. Polymer layered silicate
nanocomposites. Macromol Mater Eng 2000;279:19.
[7] Reichert P, Nitz H, Klinke S, Brandsch R, Thomann R, Muelhaupt R. Poly(propylene)/organoclay nanocomposite formation:
inuence of compatibilizer functionality and organoclay modication. Macromol Mater Eng 2000;275:817.
[8] Dietsche F, Thomann Y, Thomann R, Muelhaupt R. Translucent
acrylic nanocomposites containing anisotropic laminated nanoparticles derived from intercalated layered silicates. J Appl Polym
Sci 2000;75:396405.
[9] Liu LM, Qi ZN, Zhu XG. Studies on nylon 6/clay nanocomposites by melt-intercalation process. J Appl Polym Sci 1999;71:
11338.
[10] Mathias LJ, Davis RD, Jarrett WL. Observation of a and g
crystal forms and amorphous regions of nylon 6-clay nanocomposites using solid-state 15N nuclear magnetic resonance.
Macromolecules 1999;32:795860.
[11] Wu TM, Liao CS. Polymorphism in nylon 6/clay nanocomposites.
Macromol Chem Phys 2000;201:28205.
[12] Varlot K, Reynaud E, Kloppfer MH, Vigier G, Varlet J. Clay-

[13]
[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]
[22]

[23]
[24]

[25]

[26]

337

reinforced polyamide: preferential orientation of the montmorillonite sheets and the polyamide crystalline lamellae. J Polym Sci
Polym Phys 2001;39:136070.
Nanocor technical data: nylon nanocomposites using nanomer
I.30TC. Available from: http://www.nanocor.com.
Drechsler D, Karbach A, Fuchs H. Morphology and in situ
deformation of polyamide lms investigated with scanning force
microscopy. Surf Inter Anal 1997;25:53742.
Samon JM, Schultz JM, Wu J, Hsiao B, Yeh F, Kolb R. Study of
the structure development during the melt spinning of nylon 6
ber by on-line wide-angle synchrotron X-ray scattering techniques. J Polym Sci Polym Phys 1999;37:127787.
Schultz JM, Hsiao B, Samon JM. Structural development during
the early stages of polymer melt spinning by in-situ synchrotron
X-ray techniques. Polymer 2000;41:888795.
Ramesh C, Bhoje Gowd E. High-temperature X-ray diraction
studies on the crystalline transitions in the a- and g-forms of
nylon-6. Macromolecules 2001;34:330813.
Cheng LP, Lin DJ, Yang KC. Formation of mica-intercalatednylon 6 nanocomposite membranes by phase inversion method.
J Membr Sci 2000;172:15766.
Bonnet M, Rogausch KD, Petermann J. The endothermic
annealing peak of poly(phenylene sulphide) and poly(ethylene
terephthalate). Colloid Polym Sci 1999;277:5138.
Liu TX, Yan SK, Bonnet M, Lieberwirth I, Rogausch KD,
Petermann J. DSC and TEM investigations on multiple melting
phenomena in isotactic polystyrene. J Mater Sci 2000;35:504755.
Liu TX, Petermann J. Multiple melting behavior in isothermally
cold-crystallized isotactic polystyrene. Polymer 2001;42:645361.
Liu TX. Melting behavior of cold-crystallized isotactic polystyrene revealed by thermal analysis and transmission electron
microscopy. Eur Polym J [in press].
Liu TX, Liu ZH, et al. unpublished results.
Vaia RA, Giannelis EP. Lattice model of polymer melt intercalation in organically-modied layered silicates. Macromolecules 1997;30:79909.
Vaia RA, Giannelis EP. Polymer melt intercalation in organicallymodied layered silicates: model predictions and experiment.
Macromolecules 1997;30:80009.
Kim GM, Lee DH, Homann B, Kressler J, Stoeppelmann G.
Inuence of nanollers on the deformation process in layered silicate/polyamide-12 nanocomposites. Polymer 2001;42:1095100.