International Communications in Heat and Mass Transfer 37 (2010) 1036–1043

Contents lists available at ScienceDirect

International Communications in Heat and Mass Transfer

j ourna l homepage: www.e lsev ie r.com/ locate / ichmt

Effect of organoclay on the mechanical/thermal properties of microcellular injectionmolded PBT–clay nanocomposites☆

Shyh-shin Hwang a,⁎, Sung-po Liu a, Peming P. Hsu b, Jui-ming Yeh c, Kung-chin Chang d, Ying-zhong Lai a

a Department of Mechanical Engineering, Ching-Yun University, Chung-Li, 32097, Taiwan, ROCb Department of Computer Application Engineering, Far-East University, Tainan, 74420, Taiwan, ROCc Department of Chemistry, Chung-Yuan Christian University, Chung-Li, 32023, Taiwan, ROCd Taiwan Textile Research Institute, Tucheng, 23674, Taiwan, ROC

☆ Communicated by W.J. Minkowycz.⁎ Corresponding author.

E-mail address: stanhwang@cyu.edu.tw (S. Hwang).

0735-1933/$ – see front matter © 2010 Elsevier Ltd. Aldoi:10.1016/j.icheatmasstransfer.2010.06.010

a b s t r a c t

a r t i c l e i n f o

Available online 9 July 2010

Keywords:CompoundOrganoclayNanocompositesMicrocellularTwin-screw extruderPBT

An organically modified montmorillonite (MMT) was compounded with polybutylene terephthalate (PBT) in atwin-screw extruder. The organoclay PBT nanocomposites were then injection molded by conventional andmicrocellularmethods. Nitrogenwas used as the blowing agent. The effect of organoclay content, organoclay size (8and 35 μm), and speed of the screw (80 and 100 rpm) on themechanical and thermal propertieswere investigated.The results showed that when the MMT content was 1.0 wt.%, the nanocomposites have maximum tensilestrength, wear resistance, and cell density. Moreover, the larger the particle size, the greater the tensile strength.The screw speed during compounding also affected the mechanical strength. The higher speed of the screwincreased the tensile strength of the nanocomposites. The addition ofMMTalsohelped the thermal stability of thePBT/clay nanocomposites. The WXRD results showed that when MMT loading is 1.0 wt.%, the nanocompositeshavemaximumd-spacing structure. TEM results showed thatMMT iswell dispersed on the nanocomposites at aMMT loading of 1.0 wt.%.

l rights reserved.

© 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Polybutylene terephthalate (PBT) is commonly used as an engineer-ing plastic. It has many advantageous properties, including good flowproperties, excellent thermal stability, and good chemical resistance.However, PBT has a low heat distortion temperature and poormechanical properties. Many attempts have been made to improvethese properties of PBT by blending it with different fillers. Thesecomposites are mostly polymer clay nanocomposites (PCN). Adding asmall amount of clay to the polymer gives PCN enhanced physical andmechanical properties in aspects such as Young's modulus, tensilestrength [1], and flame retardancy [2], by comparison with neatpolymers. For example, Zhou et al. [3] studied the effect of the epoxyon the mechanical properties and melt viscoelastic behavior of PBT/MMT nanocomposites. The results showed that with epoxy as thecompatibilizer, the clay is easily intercalated and dispersed well on thePBTmatrix. Therefore, the PCN demonstrated better tensile and impactproperties than neat PBT. However, the excess epoxy reduced thematerial performance. Amendola et al. [4] employed two commercialorganoclays melt compounded with PBT with different clay loadingsand screw speeds. It was found that PCN exhibiting mixed intercalated/

exfoliated clay layer structure and the extent of exfoliation increasedwith extrusion rate for low amounts of clay. Chow [5] studied themechanical/thermal properties of cyclic extrusion PBT/MMT nanocom-posites followedby injectionmoldingwith3 wt.% clay loading.He foundthat organoclay improved the crystallinity of PBT. The meltingtemperature, crystallization temperature, and degree of crystallinity ofPCN extruded once were comparable with that extruded twice. Thisrevealed that PBT/MMT intercalated nanocomposites exhibited goodretention of mechanical/thermal properties after being extruded twicefollowed by injection molding.

Wu. et al. [6] investigated the immiscible morphology of PBT/PEblend nanocomposites by melt compounding. They concluded that PEdomain size was reduced when the PBT phase was continuous (PBT/PE=60/40). The clay prevented the agglomeration of the PE particles.Chang and Mun [7] used PBT/MMT nanocomposite fiber to investigatethe thermal/mechanical properties of the nanocomposite. Fiber wasspun for different draw ratios. The tensile properties of the nanocom-posite fibers increased with increasing organoclay contents at drawratio=1. But the tensile properties of the fibers decreased withincreasing draw ratio.

Several methods for the preparation of the PCNs have beendeveloped, including in-situ polymerization, melt compounding, andtemplate synthesis [8]. Many papers have reported the preparation ofpolymer nanocomposites by in-situ polymerization [9–11]. Thistechnique gives the PCN a good degree of exfoliation, which cannotbe achieved by melt compounding of polymer and clay. A more

Table 1Effects of organoclay content, clay size, and screw speed on the tensile strength of thePBT/MMT nanocomposites.

Organoclaycontent (wt.%)

Screw speed 100,Clay size 8 μmSolid/foamed(kgf/mm2)

Screw speed 100,Clay size 35 μmSolid/foamed(kgf/mm2)

Screw speed 80,Clay size 35 μmSolid/foamed(kgf/mm2)

Neat PBT 4.908/3.195 4.908/3.195 4.908/3.1950.5 wt.% 4.967/3.391 5.166/3.395 5.012/–1.0 wt.% 5.082/3.405 5.222/3.651 5.143/3.2192.0 wt.% 5.056/3.387 5.123/2.917 4.986/1.1893.0 wt.% 4.935/3.198 4.718/2.303 4.745/0.9824.0 wt.% – 4.372/1.780 4.65/1.297

Table 2Effects of organoclay content, clay size, and screw speed on the size of the PBT/MMTfoams.

Organoclaycontent (wt.%)

Screw speed 100,Clay size 8 μm(μm)

Screw speed 100,Clay size 35 μm(μm)

Screw speed 80,Clay size 35 μm(μm)

Neat PBT 164 164 1640.5 wt.% 157 134 –

1.0 wt.% 105 78 932.0 wt.% 200 90.4 1593.0 wt.% 219 106 2104.0 wt.% – 273 220

Fig. 2. Wide-angle XRD patterns of solid MMT/PBT nanocomposites at 1.0 wt.% clayloading with 35 μm clay size at different screw speeds.

1037S. Hwang et al. / International Communications in Heat and Mass Transfer 37 (2010) 1036–1043

economic method that uses conventional polymer processing is meltcompounding of the polymer with the organoclay [12,13].

PCN is also applied during the foaming process where the clayserves as the nucleation agent, leading to a small cell size [14] that isuseful in acoustic and thermal insulator applications [8]. Microcellularfoaming blends polymer and supercritical fluid, creating millions ofmicrocells whose size is less than 100 μm [15], improving the part'sdimensional stability [16]. However, few reports associated with PBT–clay on the microcellular foaming have been published. In this paper,we investigate the effect of the organoclay content, size and screwspeed of the melt compounding on the mechanical/thermal proper-ties of the microcellular injection molded PBT/MMT nanocomposites,based on the studies of thermogravimetric analysis (TGA), differentialscanning calorimetry (DSC), tensile test, wear resistance, and SEM,respectively.

Fig. 1.Wide-angle XRD patterns of solid MMT/PBT nanocomposites with 8 μm clay size.

2. Materials and experimental

MMTwith a cation-exchange capacity of 90 mequiv/100 g was usedas the clay, while stearylbenzyldime-ammonium chloride (STAC),supplied by Süd-Chemie (Germany), was used as an organic modifierof MMT. Two different organoclay sizes, Nanofil 9 and 919 (8 μm and35 μm), were used in this study to investigate the effect of clay size onthe mechanical/thermal properties of the PCN. Pure PBT, DHK220, wassupplied by Shin-Kong Co. Ltd. and PBT/MMT nanocomposites wereprepared by twin-screw extruder (KTX-30, Cobel Co., Japan). To studythe effect of screw speed on the properties of the PBT/MMTnanocomposites, two different speeds, 80 and 100 rpm, were used inthis study. The mixing ratios of MMT/PBT were 0 wt.%, 0.5 wt.%, 1 wt.%,2 wt.%, and 3 wt.% respectively.

Fig. 3. TEM image of solid 1 wt.% MMT/PBT nanocomposites with 8 μm clay size.

Fig. 4. TEM image of solid 3 wt.% MMT/PBT nanocomposites with 8 μm clay size.

Fig. 6. DSC curves of the melting temperature in solid MMT/PBT nanocomposites forclay size 35 μm and screw speed 100 rpm.

1038 S. Hwang et al. / International Communications in Heat and Mass Transfer 37 (2010) 1036–1043

2.1. Injection molding machine

The injection-molding machine used was the Arburg 420C Allroun-der 1000-350, equipped with Mucell capability. Nitrogen was used asthe gas source. In this study, process conditions were kept constant inorder to examine the effect of the organoclay contents, organoclay sizeson the mechanical and thermal properties. A mold of ASTM standardtensile and impact sample was used in this study.

2.2. Instrumentations

AnSEMwasused toobserve themorphologyof the cell structure in thefoamedpolymer–claynanocompositesmaterial. TheSEMimageof thecellstructure was taken on a JEOL JSM6360. Specimens were cut into smallerpieces and goldwas sputtered onto the surface. Theywere then inspectedusing the SEM. The samples for TEM study were prepared by placing the

Fig. 5. TGA curves of solid MMT/PBT nanocomposites for clay size 35 μm and screwspeed 100 rpm.

bulkof thePCNmaterials into epoxy resin capsules, followedby curing theepoxy resin at 100 °C for 24 h in a vacuum oven. The cured epoxy resincontainingPCNmaterialswasmicrotomedwithaReichert-JumgUltracut-E into slices 60–90 nm thick. Subsequently, one layer of carbon, about10 nm thick, was deposited on these slices on 300-mesh copper nets forTEM observations on a JEOL JEM-2010 with an acceleration voltage of200 kV.Wide-angle X-ray diffraction study of the sampleswas performedon a Panalytical X'Pert Pro X-ray diffractometer with copper target andNifilter at a scanning rate of 2°/min. A DuPont thermal analysis systemequipped with model TA DSC-Q10 and TA TGA-Q50 was used for thethermal analyses under N2 and air flow. The tensile test of the ASTMstandard sample was carried out using a tensile test analyzer, an HT-9102 M, manufactured by Hong-Da Company.

A ball-on-plane Schwingung Reibung Verschkiss (SRV) oscillationfriction wear tester was used to carry out the wear measurement. Thediameter of the steel ball was 10 mm and the hardness of the steel ballwas 62±2 HRc. The applied load was 60 N.

3. Results and discussion

The effects of organoclay content, clay size, and screw speed on thetensile strength and cell size of the PBT–clay nanocomposite materials

Fig. 7. Load–displacement curves of solid MMT/PBT nanocomposites with 8 μm claysize at 100 rpm screw speed.

Fig. 8. Load–displacement curves of foamed MMT/PBT nanocomposites with 8 μm claysize at 100 rpm screw speed.

Fig. 10. Tensile strength comparison of solid PCN at clay loading of 1.0 wt.% withdiffering clay sizes and screw speeds.

1039S. Hwang et al. / International Communications in Heat and Mass Transfer 37 (2010) 1036–1043

varied from 0 to 4 wt.% of organoclay with respect to the PBT, assummarized in Tables 1 and 2 respectively.

3.1. Characterization

Fig. 1 shows the wide-angle powder XRD patterns for a series ofsolid PCN materials with the clay size of 8 μm at the screw speed of100 rpm. The diffraction peaks of the PCN materials at the [001]direction for clay loadings of 0.5 wt.%, 1 wt.%, 2 wt.% and 3 wt.% werefound to be 2θ=2.49, 2.36, 2.45, and 2.64 respectively, correspondingto a d-spacing of 3.35 nm, 3.74 nm, 3.61 nm and 3.55, respectively.

Generally, the d-spacing of the organoclay is around 1.89 and2.03 nm for clay sizes of 8 and 35 μm, respectively. However, slightaggregation of the clay occurred at clay loadings of 2 wt.% and 3 wt.%,which caused d-spacing narrowing. Fig. 2 shows thewide-angle powderXRD patterns for the solid PCNmaterials with a clay size of 35 μmat clayloading of 1.0 wt.% at different screw speeds (80 and 100 rpm). It isobserved that the PCN at high speed suffered high shear rates, whichmade the clay layer more dispersed. Moreover, the larger clay sizeresults in longer d-spacing layers than the smaller clay size. For the twodifferent clay sizes used, it is seen that the diffraction peaks appear

Fig. 9. Tensile strength comparison of solid PCN at differing clay loadings, clay sizes, andscrew speeds.

around 2°–3°, implying that the clay has an intercalated layer structure,as shown in Figs. 3 and 4. For themorphological image study (Fig. 3), theTEMmicrograph of the PCNmaterial with clay size 8 μm atMMT 1 wt.%content shows the dispersion of intercalated MMT within the PBTmatrix. In themicrographs, the bright field shows the PBTmatrix, whilethe dark lines show the transverse section of the MMT layers. Some ofthe clay layers slightly aggregate when the MMT content is 3 wt.%, asshown in Fig. 4.

3.2. Thermal stability

Fig. 5 shows the TGA curves of the MMT/PBT nanocomposites forthe clay size 35 μm and a screw speed of 100 rpm, as well as the neatPBT, under atmosphere. In general, one major stage of weight lossappears, starting from 390 °C, and ending at 430 °C. This correspondsto the structural decomposition of the polymer. For PCN materials,clay has been found to enhance the thermal stability of the polymer[17,18]. We found that the beginning of thermal decomposition inthese nanocomposites shifted slightly, to a higher temperature, thanthat of neat PBT. DSC measurements showed that the addition of theorganoclay results in an increase in the melting temperature, relativeto neat PBT, as shown in Fig. 6.

Fig. 11. Wear resistance comparison of solid PCN with differing organoclay contents,sizes, and screw speeds.

1040 S. Hwang et al. / International Communications in Heat and Mass Transfer 37 (2010) 1036–1043

3.3. Mechanical properties

ASTM standards D 638-99 and D 256-99 were used as the tensileand impact tests, respectively. To investigate the effect of clay size andscrew speed of twin-screw extruder on the tensile strength of solidmolding, two different clay sizes and screw speeds were used in thisstudy. The clay sizes were 8 μm (Nanofil 9) and 35 μm (Nanofil 919)and screw speeds of the twin-extruder were 80 rpm and 100 rpm. Therelationship between the mechanical properties and the clay content

Fig. 12. SEM of worn surfaces of solid PCN with differ

of solid and foamed PCN with 8 μm clay (Nanofil 9) at 100 rpm screwspeed is shown in Figs. 7–8 respectively. Neat PBT is a very ductilematerial. Adding a small amount of MMT can increase its tensilestrength but decreases its ductility. The tensile strength reaches itsmaximum at a clay loading of 1 wt.% and then levels off at loadings of2 wt.% and 3 wt.%. Foamed PBT/MMT nanocomposites have similartrends in tensile strength as the clay loading increases, as observed insolid molding, but the elongation of the foamed PCN is worse thanthat of the solid PCN.

ing organoclay contents, sizes, and screw speeds.

1041S. Hwang et al. / International Communications in Heat and Mass Transfer 37 (2010) 1036–1043

Fig. 9 shows the tensile strength comparison of solid PCN atdifferent screw speeds (100 and 80 rpm) and clay sizes (8 and 35 μm).It is observed that neat PCN material at high shear rate (high screwspeed) has high tensile strength. The high shear rate made the d-spacing of the layer clay widen (Fig. 2), increasing the tensile strength.Furthermore, PCN with large clay size (35 μm) has greater tensilestrength than that of the small clay size (8 μm). While one research

Fig. 13. Cell structure of PBT/MMT nanocomposite foams wit

group reported that particle size has a major effect on the wear rate,with larger sizes resulting in poorer wear resistance, there does notappear to be a significant effect on the friction coefficient [17].

Among the three PCNs, the maximum tensile strength is attainedat the clay loading of 1.0 wt.% no matter what the clay sizes and screwspeeds are. Fig. 10 shows the load displacement curves of the threePCNs with 1.0 wt.% clay loading. PCN with large clay size has larger

h differing organoclay contents, sizes, and screw speeds.

1042 S. Hwang et al. / International Communications in Heat and Mass Transfer 37 (2010) 1036–1043

tensile strength but is more brittle than PCN with small clay size.Furthermore, PCN at the same clay size extruded at high speed has agreater tensile strength than that extruded at low speed. PCNextruded at high speed suffers higher shear rates during processing,which creates a more dispersed clay layer and enhances the tensileproperties of the PCN [4].

3.4. Wear resistance

A ball-on-plane Schwingung Reibung Verschkiss (SRV) oscillationfriction wear tester was used to carry out the wear measurement. Thediameter of the steel ball was 10 mm and the hardness of the steel ballwas 62±2 HRc. The applied load was 90 N. It is well known that wearresistance is proportional to the tensile strength, meaning thatmaterial with high tensile strength also has high hardness and wearresistance. Fig. 11 shows the wear resistance comparison of solid PCNwith differing organoclay contents, sizes and screw speeds. As clayloading of the PCN increases, weight loss decreases (high wearresistance) due to the strength enhancement. As the clay loadingincreases up to 1 wt.%, weight loss decreases, and then levels off dueto the strength decay of the PCN. Among the threematerials, PCNwith35 μm clay size and a screw speed of 100 rpm has the maximumtensile strength and wear resistance.

Fig. 12 shows an SEM image of the worn surfaces of the solid PCNsat differing organoclay contents, sizes and screw speeds. We can seethat as the strength of the PCN increases (Fig. 12a), the material ismore rigid and its worn surfaces are smoother than those of the highloading clay PCN (Fig. 12d). The worn surfaces in Fig. 12d, e, and f arerough, showing obvious signs of abrasive and adhesive wear. Bycontrast, a relatively smooth and uniform worn surface is formed onthe PCN (Fig. 12a, b, and c). This is consistent with previous frictionand wear studies on PA6/SiO2 nanocomposites [19].

3.5. Effect of clay content, clay size and screw speed on PBTnanocomposites foams

Cell size, volume expansion ratio (VER), and cell density, are thethree main parameters for judging a foam structure [20]. VER can becalculated from Eq. (1) [15].

VER = ρp = ρf ð1Þ

Fig. 14. Cell size and cell density vs. clay content in nanocomp

Where ρf is the foam density and ρp is the density of solid polymer.The void fraction (ψ) or weight reduction of polymer foams can becalculated from Eq. (2) [20].

ψ = 1–1= VER ð2Þ

The foam cell density (Nf) can be calculated from Eq. (3) [21].

Nf = n=Að Þ3=2 × VER ð3Þ

where n is the number of cells and A is the area taken from the SEMpicture. Table 2 shows the cell size of themicrocellular injectionmoldedMMT/PBT nanocomposites at different organoclay contents, sizes andscrew speeds. The neat PBT and PCN materials were microcellularinjection foamed with 0.8 wt.% N2. Evidently, as the content of MMTincreases, the cell size decreases, leveling off when clay content exceeds1.0 wt.% nomatter what the clay sizes and screw speeds. Further, in thePCNs, when the clay size increases, the cell size decreases. This is inagreement with most previous foaming results [14,21], where welldispersed clay particles in polymer nanocomposites were shown tohave smaller cell sizes than in neat polymer. This can also be explainedfrom the tensile strength of the foamedMMT/PBT nanocomposites. Thefoamed PCNhas themaximum tensile strength at an organoclay loadingof 1.0 wt.%, a clay size of 35 μm, and a screw speed of 100 rpm. In turn,foamed PCNs has the smallest cell size at the clay loading of 1.0 wt.%, aclay size of 35 μm, and a screw speed of 100 rpm. Fig. 13 shows the cellstructure of foamed MMT/PBT nanocomposites with different organo-clay contents, sizes and screw speeds.

Pure PBT foam had the largest cell size of the 3 batch foams studiedhere, on the order of 164 μm, as shown in Table 2. Foamed cell sizewas approximately 105, 78 and 93 μm for the PBT nanocompositefoam with 1.0 wt.% organoclay under identical process conditions forclay size/screw speeds equaling 80 μm/100 rpm, 35 μm/100 rpm, and35 μm/80 rpm respectively. As the organoclay content in the PBTnanocomposites increased, the cell size in the nanocomposite foamsdecreased. The cell size levels off when the clay content exceeds 1 wt.%. This is caused by the clay layer aggregation, as shown in Fig. 4.

Fig. 14 also demonstrates the foam cell density changes with purePBT and its nanocomposites of clay size 35 μmand screwspeed 100 rpmat different organoclay contents. The foaming process decreases the cellsize and increases the cell density. In general, microcellular nanocom-posite foams have higher tensile strength but lower impact strength

osite foams of clay size 35 μm and screw speed 100 rpm.

1043S. Hwang et al. / International Communications in Heat and Mass Transfer 37 (2010) 1036–1043

than their neat resin counterparts. The effect of the nano-clays on themechanical properties is also related to the weight reduction [22]. Theaddition of clay to polymer can createmore nucleation sites for foamingand can produce smaller cell sizes by comparison with neat polymer. Inturn, a foamedpolymerwith smaller cell sizes has lessweight reduction,which enhances the tensile strength [23]. In addition, the foamedpolymerwith large cell sizes has a thinner cell wall thickness and hencetends to have lower tensile strength.

4. Conclusions

In this study, a series of PCN materials consisting of commercial PBTand organoclay were successfully prepared using melt compounding.The as-prepared PCNmaterialswere then subjected tomicrocellular andconventional injectionmolding. The sampleswere characterized byXRDand TEM.We found that addition of organoclay to the PBT resulted in anincrease in mechanical strength and an increase in thermal stability,based on analysis by tensile, TGA, and DSC tests. Furthermore, theincorporation of organoclay into the PBT matrix showed a decrease inthe cell size based on the cellmorphologymeasurement. The organoclaysize and screw speed during melt compounding affected the tensilestrength of the PCN as well.

Acknowledgment

This work was supported by the National Science Council grantNSC 96-2221-E-231-011.

References

[1] Z. Wang, T.J. Pinnavaia, Nanolayer reinforcement of elastomeric polyurethane,Chem. Mater. 10 (1998) 3769.

[2] J.W. Gilman, C.L. Jackson, A.B. Morgan, R. Hayyis, E. Manias, E.P. Giannelis, M.Wuthenow, D. Hilton, S.H. Philips, Flammability properties of polymer–layered-silicate nanocomposites. Polypropylene and polystyrene nanocomposites, Chem.Mater. 12 (2000) 1866.

[3] F. Wu, C.X. Zhou, X. Fan, D.L. Mao, Z. Bian, Effect of epoxy resin on the mechanicalproperties and melt viscoelastic behaviour of poly(butylene terephthalate)/mont-morillonite nanocomposites, Polymers & Polymer Composites 13 (6) (2005) 571.

[4] E. Amendola, G. Nocerino, G. Costa, Preparation and characterization of PBTnanocomposites compounded with different montmorillonites, Polym. Eng. Sci.44 (6) (2004) 1012.

[5] W.S. Chow, Cyclic extrusion of poly(butylene terephthalate)/organo-montmoril-lonite nanocomposites: thermal and mechanical retention properties, J. Appl.Polym. Sci. 110 (2008) 1642.

[6] D.F. Wu, C.X. Zhou, M. Zhang, Effect of clay on immiscible morphology of poly(butylenes terephthalate)/polyethylene blend nanocomposites, J. Appl. Polym.Sci. 102 (2006) 3628.

[7] J.-H. Chang, M.K. Mun, Synthesis and characterization of poly(butylenesterephthalate) nanocomposite fibers: thermo-mechanical properties and mor-phology, J. Appl. Polym. Sci. 100 (2006) 1247.

[8] M. Alexandre, P. Dubois, Polymer-layered silicate nanocomposites: preparation,properties and uses of a new class of materials, Mater. Sci. Eng. R, 28 (2000) 1–63.

[9] A. Usuki, Y. Kojima, M. Kawasumi, A. Okada, Y. Fukushima, T. Kurauchi, Synthesisof nylon 6-clay hybrid, J. Mater. Res. 8 (1993) 1197.

[10] X. Huang, S. Lewis, W.J. Britain, R.A. Vaia, Synthesis of polycarbonate-layeredsilicate nanocomposites via cyclic oligomers, Macromolecules 33 (6) (2000).

[11] P.B. Messersmith, E.P. Giannelis, Synthesis and barrier properties of poly(e-capro1ac-tone)-layered silicate nanocomposites, J. Polym. Sci. Part A: Polym. Chem. 33 (7) (1995)1047.

[12] M. Kato, A. Usuki, A. Okada, Synthesis of polypropylene oligomer–clayintercalation compounds, J. Appl. Polym. Sci. 66 (1997) 1781.

[13] R.A. Vaia, E.P. Giannelis, Polymer melt intercalation in organically-modified layeredsilicates: model predictions and experiment, Macromolecules 30 (25) (1997) 8000.

[14] Y.F. Fujimoto, S.S. Ray, M. Okamoto, A. Ogami, K. Yamada, K. Ueda, Well-controlledbiodegradable nano nanocomposite foams: from microcellular to nanocellular,Macromol. Rapid Commun. 24 (2003) 457–461.

[15] V. Martini, J. E., Nam. P. Suh, and F. A. Waldman, U.S. Patent No. 4,473,665, 1984[16] S.S. Hwang, Z.S. Ke, The dimensional stability of a microcellular injection molded

gear shaft, Int. Commun. Heat Mass Transfer 35 (2008) 263–275.[17] S. Mcelwain, T. Blanchet, L. Schadler, Effect of particle size on the wear resistance

of alumina-filled PTFE micro- and nanocomposites, Tribol. Trans. 51 (2008) 247.[18] J.M. Yeh, C.L. Chen, T.H. Kuo, W.F. Su, H.Y. Huang, Preparation and properties of

(BATB–ODPA) polyimide–clay nanocomposite materials, J. Appl. Polym. Sci. 92(2004) 1072.

[19] M. Garcia, M. de Rooij, L. Winnubst, W.E. van Zyl, H. Verweij, Friction and wearstudies on nylon-6/SiO2 nanocomposites, J. Appl. Polym. Sci. 92 (2004) 1855.

[20] S.T. Lee, Foam Extrusion: Principles and Practice, CRC, Lancaster, 2000.[21] X.M. Han, K.W. Koelling, D.L. Tomasko, J.L. Lee, Continuous microcellular

polystyrene foam extrusion with supercritical CO2, Polym. Eng. Sci. 42 (2002)2094–2106.

[22] M. Yuan, L. S. Turng, S. Gong, D. Caulfield, C. Hunt, and R. Spindler, Study ofinjection molded microcellular polyamide-6 nanocomposites, Polym. Eng. Sci., 44(2004), 44, 4, 673.

[23] A. Chandra, M.Y. Shaoqin Gong, Lih-Sheng Turng, Microstructure and crystallog-raphy in microcellular injection-molded polyamide-6 nanocomposite and neatresin, Polym. Eng. Sci. 45 (2005) 52.