A review on polymer–layered silicate nanocomposites

Progress in Polymer Science 33 (2008) 1119–1198

Contents lists available at ScienceDirect

Progress in Polymer Science

journa l homepage: www.e lsev ier .com/ locate /ppolysc i

A review on polymer–layered silicate nanocomposites

S. Pavlidoua, C.D. Papaspyridesb,∗

a CLOTEFI, Clothing, Textile and Fibre Technological Development, 4 El. Venizelou Str., Kallithea, Athens 176 76, Greeceb Laboratory of Polymer Technology, Department of Chemical Engineering, National Technical University of Athens, Zographou, Athens 157 80, Greece

a r t i c l e i n f o

Article history:Received 25 June 2007Received in revised form 11 July 2008Accepted 25 July 2008Available online 25 September 2008

Keywords:

a b s t r a c t

This review reports recent advances in the field of polymer–layered silicate nanocompos-ites. These materials have attracted both academic and industrial attention because theyexhibit dramatic improvement in properties at very low filler contents. Herein, the struc-ture, preparation and properties of polymer–layered silicate nanocomposites are discussedin general, and detailed examples are also drawn from the scientific literature.

© 2008 Elsevier Ltd. All rights reserved.

NanocompositesPolymersLayered silicatesC

C

0

lays

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11202. Milestones in the research on polymer–layered silicate nanocomposites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11213. Layered silicates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1122

3.1. Structure and characteristics of layered silicates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11223.2. Organic modification of layered silicates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1123

4. Nanocomposite structures and characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11264.1. Nanocomposite structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11264.2. Nanocomposite structural characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1127

5. Preparation of nanocomposites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11295.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1129

5.1.1. Template synthesis (sol–gel technology) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11295.1.2. Intercalation of polymer or prepolymer from solution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11305.1.3. In situ intercalative polymerization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11305.1.4. Melt intercalation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1130

5.2. Intercalation of polymer from solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11305.3. In situ intercalative polymerization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1131

5.3.1. In situ intercalative polymerization of thermoplastic polymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1131
5.3.2. In situ intercalative polymerization of thermosett
5.4. Polymer melt intercalation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5.4.1. Introduction and advantages of the technique . . . .5.4.2. Factors affecting polymer melt intercalation . . . . .

∗ Corresponding author.E-mail address: [email protected] (C.D. Papaspyrides).

079-6700/$ – see front matter © 2008 Elsevier Ltd. All rights reserved.doi:10.1016/j.progpolymsci.2008.07.008

ing polymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1137. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1142. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1142. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1144

http://www.sciencedirect.com/science/journal/00796700

http://www.elsevier.com/locate/ppolysci

mailto:[email protected]

dx.doi.org/10.1016/j.progpolymsci.2008.07.008

1120 S. Pavlidou, C.D. Papaspyrides / Progress in Polymer Science 33 (2008) 1119–1198

5.4.3. Compatibility issues in non-polar polymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11515.4.4. Degradation problems encountered during melt intercalation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1154

6. Nanocomposite properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11566.1. Mechanical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1156

6.1.1. The reinforcing mechanism of layered silicates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11566.1.2. Modulus and strength. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11576.1.3. Toughness and strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11646.1.4. Comparison and synergistic effects of clays and conventional reinforcements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1168

6.2. Dynamic mechanical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11696.3. Barrier properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11706.4. Thermal stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11736.5. Flame retardance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1178

6.5.1. Flame retardance of polymer–layered silicate nanocomposites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11786.5.2. Synergism between nanocomposites and flame retardants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1183

6.6. Heat distortion temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11856.7. Rheological properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11856.8. Crystallinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11866.9. Biodegradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11886.10. Photo-degradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11896.11. Optical clarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1190

7. Nanocomposites: advantages and applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11908. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1193

. . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction

Traditionally, polymeric materials have been filled withsynthetic or natural inorganic compounds in order toimprove their properties, or simply to reduce cost. Con-ventional fillers are materials in the form of particles (e.g.calcium carbonate), fibers (e.g. glass fibers) or plate-shapedparticles (e.g. mica). However, although conventionallyfilled or reinforced polymeric materials are widely usedin various fields, it is often reported that the addition ofthese fillers imparts drawbacks to the resulting materi-als, such as weight increase, brittleness and opacity [1–5].Nanocomposites, on the other hand, are a new class ofcomposites, for which at least one dimension of the dis-persed particles is in the nanometer range. Depending onhow many dimensions are in the nanometer range, one candistinguish isodimensional nanoparticles when the threedimensions are on the order of nanometers, nanotubes orwhiskers when two dimensions are on the nanometer scaleand the third is larger, thus forming an elongated structure,and, finally, layered crystals or clays, present in the form ofsheets of one to a few nanometers thick and hundreds tothousands nanometers in extent [1,4]. Among all the poten-tial nanocomposite precursors, those based on clay andlayered silicates have been most widely investigated, prob-ably because the starting clay materials are easily availableand because their intercalation chemistry has been studiedfor a long time [6].

Polymer–layered silicate nanocomposites, which arethe subject of the present contribution, are prepared byincorporating finely dispersed layered silicate materials in a
polymer matrix [2]. However, the nanolayers are not easilydispersed in most polymers due to their preferred face-to-face stacking in agglomerated tactoids. Dispersion ofthe tactoids into discrete monolayers is further hinderedby the intrinsic incompatibility of hydrophilic layered sil-
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1193

icates and hydrophobic engineering plastics. Therefore,layered silicates first need to be organically modified toproduce polymer–compatible clay (organoclay). In fact, ithas been well-demonstrated that the replacement of theinorganic exchange cations in the cavities or “galleries” ofthe native clay silicate structure by alkylammonium sur-factants can compatibilize the surface chemistry of the clayand a hydrophobic polymer matrix [7].

Thereafter, different approaches can be applied to incor-porate the ion-exchanged layered silicates in polymer hostsby in situ polymerization, solution intercalation or sim-ple melt mixing. In any case, nanoparticles are addedto the matrix or matrix precursors as 1–100 �m pow-ders, containing associated nanoparticles. Engineering thecorrect interfacial chemistry between nanoparticles andthe polymer host, as described previously, is critical butnot sufficient to transform the micron-scale composi-tional heterogeneity of the initial powder into nanoscalehomogenization of nanoparticles within a polymericnanocomposite [8]. Therefore, appropriate conditions haveto be established during the nanocomposite preparationstage.

The resulting polymer–layered silicates hybrids possessunique properties – typically not shared by their more con-ventional microscopic counterparts – which are attributedto their nanometer size features and the extraordinarilyhigh surface area of the dispersed clay [1,4]. In fact, it iswell established that dramatic improvements in physicalproperties, such as tensile strength and modulus, heat dis-tortion temperature (HDT) and gas permeability, can beachieved by adding just a small fraction of clay to a poly-
mer matrix, without impairing the optical homogeneity ofthe material. Most notable are the unexpected propertiesobtained from the addition of stiff filler to a polymer matrix,e.g. the often reported retention (or even improvement) ofthe impact strength. Since the weight fraction of the inor-

S. Pavlidou, C.D. Papaspyrides / Progress in Polymer Science 33 (2008) 1119–1198 1121

Nomenclature

AIBN N,N′-azobis(isobutyronitrile)CEC cation exchange capacityCX-clay X is the number of carbon atoms in clay

organic modifierd spacing between diffractional lattice planesD diffusivityDETDA diethyl toluene diamineDGEBA diglycidyl ether of bisphenol ADM dioctadecyldimethyl ammonium chlorideDMA dynamic mechanical analysisDSC differential scanning calorimetryDTGA differential thermogravimetric analysisE′ storage modulus under bending modeE′′ loss modulus under bending modeEVA ethyl-vinyl-acetate copolymerFTIR Fourier transform infrared spectroscopyfwhm full-width-at-half-maximumG′ storage modulus under tensile modeG′′ loss modulus under tensile modeGPC gas-permeation chromatographyHDPE high density polyethyleneHDT heat distortion temperatureHRR heat release rateMMT montmorilloniteNMR nuclear magnetic resonanceo-Clay organo-modified clayOMLS organo-modified layered silicate, organosil-

icate, or organoclayP permeabilityPA polyamidePAA poly(acrylic acid)PBO polybenzoxalePCL polycaprolactonePCN polymer–clay nanocompositePDMS polydimethylsiloxanePE polyethylenePEI poly(ether imide)PEO poly(ethylene oxide)PET poly(ethylene terephthalate)PHB poly(3-hydroxybutyrate)PHRR peak heat release ratePI polyimidePA polylactidePLS polymer–layered silicate nanocompositePMMA polymethyl methacrylatePP polypropylenePP-MA or PP-g-MA maleic anhydride-grafted

polypropylenePS polystyrenePSF polysulfonePU polyurethanePVA poly(vinyl acetate)PVC poly(vinyl chloride)PVE poly(vinyl ethylene)PVOH poly(vinyl alcohol)PVP poly(vinyl pyrrilidone)S solubility

SBS poly(styrene-butadiene-styrene)SEA specific extinction areaTan ı G′/G′′

Tc crystallization temperatureTEM transmission electron microscopyTg glass transition temperatureTGA thermogravimetric analysisTGAP triglycidyl p-aminophenolTGDDM tetrafunctional tetraglycidyldiaminodiphe-

nylmethaneTPO thermoplastic olefinUP unsaturated polyesterWAXD or WAXS wide angle X-ray diffraction or scat-

teringXRD X-ray diffraction

ganic additive is typically below 10%, the materials are alsolighter than most conventional composites [2,9–12]. Theseunique properties make the nanocomposites ideal mate-rials for products ranging from high-barrier packaging forfood and electronics to strong, heat-resistant automotivecomponents [11]. Additionally, polymer–layered silicatenanocomposites have been proposed as model systems toexamine polymer structure and dynamics in confined envi-ronments [12,13].

However, despite the recent progress in polymernanocomposite technology, there are many fundamentalquestions that have not been answered. For example, howdo changes in polymer crystalline structure induced bythe clay affect overall composite properties? How doesone tailor organoclay chemistry to achieve high degreesof exfoliation reproducibility for a given polymer system?How do process parameters and fabrication affect compos-ite properties? Further research is needed that addressessuch issues [14]. The objective of this work is to reviewrecent scientific and technological advances in the field ofpolymer–layered silicate nanocomposite materials and todevelop a better understanding of how superior nanocom-posites are formed.

2. Milestones in the research on polymer–layeredsilicate nanocomposites

The incorporation of layered silicates into polymermatrices has been known for over 50 years [15]. In fact,one of the earliest systematic studies of the interactionbetween a clay mineral and a macromolecule dates backto 1949, when Bower [16] described the absorption ofDNA by montmorillonite. Even in the absence of X-raydiffraction (XRD) evidence, this finding implied insertionof the macromolecule in the lamellar structure of the sili-cate. In 1950, Carter et al. [17] developed organoclays withseveral organic onium bases to reinforce latex-based elas-
tomers and in 1958, Hauser and Kollman [18] were granteda patent for “clay complexes with conjugated unsatu-rated aliphatic compounds of four to five carbon atoms”.Uskov [19], in 1960, found that the softening point ofpoly(methyl methacrylate) derived by polymerization of

ess in Po
1122 S. Pavlidou, C.D. Papaspyrides / Progr
methyl methacrylate was raised by montmorillonite modi-fied with octadecylammonium, while in the following yearBlumstein [20] obtained a polymer inserted in the structureof a montmorillonite by polymerizing a previously insertedvinyl monomer. Two years later, Greenland [21] used apoly(vinyl alcohol)/montmorillonite system to show thata polymer could be directly inserted in a clay in an aqueoussolution. The same year, the incorporation of organoclayinto a thermoplastic polyolefin matrix was disclosed byNahin and Backlund [22] of Union Oil Co. They obtainedorganoclay composites with strong solvent resistance andhigh tensile strength by irradiation-induced cross linking.However, they did not focus on the intercalation char-acteristics of the organoclay or the potential propertiesof the composites. In 1975, Tanihara and Nakagawa [23]reached a similar result by intercalating polyacrylamideand poly(ethylene oxide) from an aqueous solution. In 1976Fujiwara and Sakamoto [24] of the Unichika Co. describedthe first organoclay hybrid polyamide nanocomposite.

However, it was not until Toyota researchers began adetailed examination of polymer–layered silicate compos-ites that nanocomposites became more widely studied inacademic, government and industrial laboratories [25–28].The Toyota research group disclosed improved methods forproducing nylon 6 clay nanocomposites using in situ poly-merization similar to the Unichika process. They reportedthat these polymer–clay nanocomposites exhibit superiorstrength, modulus, heat distortion temperature, water andgas barrier properties, with comparable impact strength asneat nylon 6. They also reported on various other types ofpolymer–clay hybrid nanocomposites based on epoxy resinand polystyrene, acrylic polymer, rubber, and polyimidesformed using a similar approach.

On the other hand, work by Giannelis and co-workers[29,30] revealed that intercalation of polymer chains intothe galleries of an organoclay can occur spontaneously onheating a mixture of polymer and silicate clay powderabove the polymer glass transition or melt temperature.Once sufficient polymer mobility is achieved, chains dif-fuse into the host silicate clay galleries, thereby producingan expanded polymer–silicate structure.

To summarize: although the intercalation chemistry ofpolymers when mixed with appropriately modified lay-ered silicates and synthetic layered silicates has long beenknown, two major findings have stimulated the revival ofinterest in polymer–layered silicate nanocomposite mate-rials. First, the report from the Toyota research groupon a nylon 6/montmorillonite nanocomposite, in whichvery small amounts of layered silicate loadings resultedin pronounced improvements of thermal and mechanicalproperties; and second, the observation by Giannelis andhis co-workers that it is possible to melt-mix polymers withlayered silicates, without the use of organic solvents.

Since then, the high promise for industrial applicationshas motivated vigorous research, and today efforts arebeing conducted globally, using almost all types of polymer
matrices. In fact, nanocomposites have been demonstratedwith many thermoplastic and thermosetting polymers ofdifferent polarities including, among others, polystyrene,polycaprolactone, polypropylene, poly(ethylene oxide),epoxy resin, polysiloxane and polyurethane [7,14,15,31–33].
lymer Science 33 (2008) 1119–1198

It must be noted, however, that so far most of these mate-rials have been produced only on the laboratory scale; anduntil a short time ago, research tended to center on proofof exfoliation of the clay [32].

3. Layered silicates

3.1. Structure and characteristics of layered silicates

Layered silicates used in the synthesis of nanocompos-ites are natural or synthetic minerals, consisting of very thinlayers that are usually bound together with counter-ions.Their basic building blocks are tetrahedral sheets in whichsilicon is surrounded by four oxygen atoms, and octahe-dral sheets in which a metal like aluminum is surroundedby eight oxygen atoms. Therefore, in 1:1 layered structures(e.g. in kaolinite) a tetrahedral sheet is fused with an octa-hedral sheet, whereby the oxygen atoms are shared [34].

On the other hand, the crystal lattice of 2:1 layered sil-icates (or 2:1 phyllosilicates), consists of two-dimensionallayers where a central octahedral sheet of alumina is fusedto two external silica tetrahedra by the tip, so that theoxygen ions of the octahedral sheet also belong to thetetrahedral sheets, as shown in Fig. 1. The layer thick-ness is around 1 nm and the lateral dimensions may varyfrom 300 Å to several microns, and even larger, depend-ing on the particulate silicate, the source of the clay andthe method of preparation (e.g. clays prepared by millingtypically have lateral platelet dimensions of approximately0.1–1.0 �m). Therefore, the aspect ratio of these layers(ratio length/thickness) is particularly high, with valuesgreater than 1000 [1,35–37].

The basic 2:1 structure with silicon in the tetrahedralsheets and aluminum in the octahedral sheet, without anysubstitution of atoms, is called pyrophyllite. Since the layersdo not expand in water, pyrophyllite has only an externalsurface area and essentially no internal one. When sili-con in the tetrahedral sheet is substituted by aluminum,the resulting structure is called mica. Due to this substi-tution the mineral is characterized by a negative surfacecharge, which is balanced by interlayer potassium cations.However, because the size of the potassium ions matchesthe hexagonal hole created by the Si/Al tetrahedral layer,it is able to fit very tightly between the layers. Conse-quently, the interlayers collapse and the layers are heldtogether by the electrostatic attraction between the nega-tively charged tetrahedral layer and the potassium cations.Therefore, micas do not swell in water and, like pyrophyl-lite, have no internal surface area [38]. On the other hand, ifin the original pyrophyllite structure the trivalent Al-cationin the octahedral layer is partially substituted by the diva-lent Mg-cation, the structure of montmorillonite is formed,which is the best-known member of a group of clay min-erals, called “smectites” or “smectite clays”. In this case theoverall negative charge is balanced by sodium and calciumions, which exist hydrated in the interlayer [39]. A particu-
lar feature of the resulting structure is that, since these ionsdo not fit in the tetrahedral layer, as in mica, and the layersare held together by relatively weak forces, water and otherpolar molecules can enter between the unit layers, causingthe lattice to expand [33].


oduced

ani

dmic

teodpvucih

TC

2

MHS

(

Fig. 1. The structure of a 2:1 layered silicate [35]. Repr

Along with montmorillonite, hectorite and saponitere the layered silicates that are most commonly used inanocomposite materials. Their chemical formula is given

n Table 1 [1].The reason why these materials have received a great

eal of attention recently, as reinforcing materials for poly-ers, is their potentially high aspect ratio and the unique

ntercalation/exfoliation characteristics that will be dis-ussed later [15].

In general, it is well established that structural perfec-ion is more and more nearly reached as the reinforcinglements become smaller and that the ultimate propertiesf reinforcing composite elements may be expected if theirimensions reach atomic or molecular levels. For exam-le, carbon nanotubes display the so far highest known
alues of elastic modulus (ca. 1.7 TPa!). Similarly, individ-al clay sheets, being only 1 nm thick, display a perfectrystalline structure. However, the smaller the reinforc-ng elements are, the larger is their internal surface andence their tendency to agglomerate rather than to dis-
able 1hemical structure of commonly used 2:1 phyllosilicatesa [1].

:1 Phyllosilicates General formula

ontmorillonite Mx(Al4−xMgx)Si8O20(OH)4

ectorite Mx(Mg6−xLix)Si8O20(OH)4

aponite MxMg6(Si8−xAlx)O20(OH)4

a M: monovalent cation; x: degree of isomorphous substitutionbetween 0.5 and 1.3).

from Beyer by permission of Elsevier Science Ltd., UK.

perse homogeneously in a matrix [2]. In fact, the silicatelayers have the tendency to organize themselves to formstacks with a regular van der Waals gap between them,called an “interlayer” or “gallery” [1,35,36]. The interlayerdimension is determined by the crystal structure of the sil-icate (for dehydrated Na–montmorillonite this dimensionis approximately 1 nm) [37].

Analysis of layered silicates has shown that thereare several levels of organization within the clay miner-als. The smallest particles, primary particles, are on theorder of 10 nm and are composed of stacks of parallellamellae. Micro-aggregates are formed by lateral joiningof several primary particles, and aggregates are com-posed of several primary particles and micro-aggregates[40].

3.2. Organic modification of layered silicates

Since, in their pristine state layered silicates are onlymiscible with hydrophilic polymers, such as poly(ethyleneoxide) and poly(vinyl alcohol), in order to render themmiscible with other polymers, one must exchange thealkali counter-ions with a cationic-organic surfactant, asshown in Fig. 2. Alkylammonium ions are mostly used,
although other “onium” salts can be used, such as sul-fonium and phosphonium [1,39,41]. This can be readilyachieved through ion-exchange reactions that render theclay organophilic [42]. In order to obtain the exchange ofthe onium ions with the cations in the galleries, water


ively smap betwre chan

Fig. 2. Schematic picture of an ion-exchange reaction. The inorganic, relatcations. This ion-exchange reaction has two consequences: firstly, the gbetween them and secondly, the surface properties of each single sheet aby permission of Elsevier Science Ltd., UK.

swelling of the silicate is needed. For this reason alkalications are preferred in the galleries because 2-valent andhigher valent cations prevent swelling by water. Indeed,the hydrate formation of monovalent intergallery cations isthe driving force for water swelling. Natural clays may con-tain divalent cations such as calcium and require exchangeprocedures with sodium prior to further treatment withonium salts [41]. The alkali cations, as they are not struc-tural, can be easily replaced by other positively chargedatoms or molecules, and thus are called exchangeablecations [43].

The organic cations lower the surface energy of the sili-cate surface and improve wetting with the polymer matrix[4,42]. Moreover, the long organic chains of such surfac-tants, with positively charged ends, are tethered to thesurface of the negatively charged silicate layers, resultingin an increase of the gallery height [44]. It then becomespossible for organic species (i.e. polymers or prepolymers)to diffuse between the layers and eventually separate them[42,45]. Sometimes, the alkylammonium cations may evenprovide functional groups that can react with the poly-mer or initiate polymerization of monomers [33]. Themicrochemical environment in the galleries is, therefore,appropriate to the intercalation of polymer molecules[46]. Conclusively, the surface modification both increasesthe basal spacing of clays and serves as a compatibilizerbetween the hydrophilic clay and the hydrophobic polymer[45].

The excess negative charge of layered silicates and theircapability to exchange ions can be quantified by a spe-cific property known as the cation-exchange capacity (CEC)and expressed in mequiv./g [1,39]. This property is highlydependent on the nature of the isomorphous substitutionsin the tetrahedral and octahedral layers and therefore onthe nature of the soil where the clay was formed. Thisexplains, for example, why montmorillonites from different
origins show differences in CEC, ranging from approxi-mately 0.9–1.2 mequiv./g [39,42]. The charge of the layer isnot locally constant, as it varies from layer to layer, and mustrather be considered as an average value over the wholecrystal. Proportionally, even if a small part of the charge
all (sodium) ions are exchanged against more voluminous organic oniumeen the single sheets is widened, enabling polymer chains to move in

ged from being hydrophilic to hydrophobic [2]. Reproduced from Fischer

balancing cations is located on the external crystallite sur-face, the majority of these exchangeable cations are locatedinside the galleries [1].

Depending on the functionality, packing density, andlength of the organic modifiers, the organo-modified lay-ered silicates (OMLSs, organosilicates or organoclays) maybe engineered to optimize their compatibility with a givenpolymer [43,47]. It is worth noticing that, on the basis ofthe CEC of the clay, the content of the surfactant is usuallyabout 35–45 wt.% [48]. Actually, one way to measure theclay CEC is by determining the amount of alkylammoniumsalt retained by the organoclays. That is, dried clays thathave been subjected to organo-modification along with asample of the corresponding untreated clay are ignited at1000 ◦C. From the differences in the loss on ignition of thesample and blank and the molecular weight of the alkylam-monium salt, the milliequivalents of the organic substanceretained by the clays are calculated and those values aretaken as their CEC. Alternatively, chemical analysis of theclay can also be applied for CEC determination [42].

In general, the longer the surfactant chain length, andthe higher the charge density of the clay, the further apartthe clay layers will be forced. This is expected since bothof these parameters contribute to increasing the volumeoccupied by the intragallery surfactant [7]. For example,Wang et al. prepared organoclays with different alkylam-monium chain lengths and also used an organophilic clay,Cloisite 20A, which has two long alkyl chains. They foundthat the interlayer spacing increases with the increase insize of alkylamine chain length. The interlayer spacings ofC12 M, C16 M, C18 M (with 12, 16 and 18 carbon atoms inthe alkylammonium chain) and 20 A were 1.36, 1.79, 1.85and 2.47 nm, respectively [49].

However, the interlayer distance also depends on theway the onium ion chains organize themselves in theorganoclay. In order to describe the structure of the
interlayer in organoclays, one has to know that, as the neg-ative charge originates in the silicate layer, the cationichead group of the alkylammonium molecule preferentiallyresides at the layer surface, leaving the organic tail radiatingaway from the surface [1,41].


F (b) lateral bilayer; (c) paraffin-type monolayer and (d) paraffin-type bilayer [1].R Ltd., UK.

dteotaF

Vfbftdatossdmrt

op

FiA

ig. 3. Alkyl chain aggregation in layered silicates: (a) lateral monolayer;eproduced from Alexandre and Dubois by permission of Elsevier Science

Initially, the orientation of surfactant chains waseduced from infrared and XRD measurements, accordingo which the organic chains have been long thought to lieither parallel to the silicate layer, forming mono or bilayersr, depending on the packing density and the chain length,o radiate away from the surface, forming mono or evenbimolecular tilted “paraffinic” arrangement, as shown in

ig. 3 [1,7].A more realistic description has been proposed by

aia et al., based on FTIR experiments. By monitoringrequency shifts of the asymmetric CH2 stretching andending vibrations, they showed that alkyl chains can varyrom liquid-like to solid-like, with the liquid-like struc-ure dominating as the interlayer density or chain lengthecreases, or as the temperature increases. When the avail-ble surface area per molecule is within a certain range,he chains are not completely disordered but retain somerientational order similar to that in the liquid crystallinetate (Fig. 4). As the chain length increases, the interlayertructure appears to evolve in a stepwise fashion, from aisordered to more ordered monolayer, then “jumping” to aore disordered pseudo-bilayer. In addition, an NMR study

eported by Wang et al. indicated the coexistence of ordered
rans and disordered gauche conformations [50].
Fornes et al. conducted WAXS scans for differentrganoclays and for pristine sodium montmorillonite, andlotted basal spacing values obtained vs. the mass of

ig. 4. Alkyl chain aggregation models: (a) short alkyl chains: isolated molecules,nterdigitation to form quasi bilayers and (c) longer chain length: increased intlexandre and Dubois by permission of Elsevier Science Ltd., UK.

Fig. 5. WAXS results for organoclays: mass of organic per unit mass ofmontmorillonite [51]. Reproduced from Fornes, Yoon, Hunter, Keskkulaand Paul by permission of Elsevier Science Ltd., UK.

organic component per unit mass of inorganic MMT foreach organoclay, as shown in Fig. 5. They further analyzed
the data by expressing the mass of organic material per unitvolume of gallery, or gallery density, as
�gallery= mass organicgallery volume

= 1d − d0

mass organic/mass MMT(area/side)/mass MMT

lateral monolayer; (b) intermediate chain lengths: in-plane disorder anderlayer order, liquid crystalline-type environment [1]. Reproduced from

ess in Po
where d − d0 is the gallery height, as indicated by Fig. 5.The slope of the linear relation between gallery height andthe second quantity in the above equation gives the den-sity of the organic material in the gallery, �gallery. Forcingthe fit through an intercept d0 = 9.6 Å, which is the basalspacing of pristine sodium montmorillonite, produces acalculated density of 1.07 g/cm3. This range of densitiesencompasses what might be expected for organic liquidsor solids. Therefore, the densities calculated in that workagree with conclusions made by Vaia et al. about the con-formation and structure of the organic interlayer of similarorganoclays, suggesting that organoclay galleries exhibitmolecular environments ranging from solid- to liquid-like[51].

Summarizing this section, there are two particularcharacteristics of layered silicates that are exploited inpolymer–layered silicate nanocomposites. The first is theability of the silicate particles to disperse into individuallayers. Since dispersing a layered silicate can be picturedlike opening a book, an aspect ratio as high as 1000 forfully dispersed individual layers can be obtained (contrastthat to an aspect ratio of about 10 for undispersed or poorlydispersed particles). The second characteristic is the abilityto fine-tune their surface chemistry through ion exchangereactions with organic and inorganic cations. These twocharacteristics are, of course, interrelated since the degreeof dispersion in a given matrix that, in turn, determinesaspect ratio, depends on the interlayer cation [4,40].

4. Nanocomposite structures and characterization

4.1. Nanocomposite structures

Any physical mixture of a polymer and silicate (orinorganic material in general) does not necessarily forma nanocomposite. The situation is analogous to polymerblends. In most cases, separation into discrete phasesnormally takes place. In immiscible systems, the poorphysical attraction between the organic and the inorganiccomponents leads to relatively poor mechanical proper-ties. Furthermore, particle agglomeration tends to reducestrength and produce weaker materials [4]. Thus, when thepolymer is unable to intercalate between the silicate sheets,a phase-separated composite is obtained, whose propertiesare in the same range as for traditional microcomposites[1,35].

Beyond this traditional class of polymer-filler com-posites, two types of nanocomposites can be obtained,depending on the preparation method and the natureof the components used, including polymer matrix, lay-ered silicate and organic cation [1,35]. These two types ofpolymer–layered silicate nanocomposites are depicted inFig. 6 [50].

Intercalated structures are formed when a single (orsometimes more) extended polymer chain is intercalated
between the silicate layers. The result is a well orderedmultilayer structure of alternating polymeric and inorganiclayers, with a repeat distance between them. Intercalation
causes less than 20–30 ´̊A separation between the platelets[1,4,33,35,44,52].

lymer Science 33 (2008) 1119–1198

On the other hand, exfoliated or delaminated structuresare obtained when the clay layers are well separated fromone another and individually dispersed in the continuouspolymer matrix [1,4,33,44]. In this case, the polymer sepa-

rates the clay platelets by 80–100 ´̊A or more [52]. That is, theinterlayer expansion is comparable to the radius of gyrationof the polymer rather than that of an extended chain, as inthe case of intercalated hybrids [4].

The exfoliation or delamination configuration is of par-ticular interest because it maximizes the polymer–clayinteractions making the entire surface of layers availablefor the polymer. This should lead to the most significantchanges in mechanical and physical properties [35]. In fact,it is generally accepted that exfoliated systems give bet-ter mechanical properties than intercalated ones [5,33].The complete dispersion of clay nanolayers in a polymeroptimizes the number of available reinforcing elements forcarrying an applied load and deflecting cracks. The cou-pling between the tremendous surface area of the clayand the polymer matrix facilitates stress transfer to thereinforcement phase, allowing for mechanical propertyimprovements [35,53].

However, it is not easy to achieve complete exfolia-tion of clays and, indeed with few exceptions, the majorityof the polymer nanocomposites reported in the literaturewere found to have intercalated or mixed intercalated-exfoliated nanostructures [33]. This is because the silicatelayers are highly anisotropic, with lateral dimensions rang-ing from 100 to 1000 nm, and even when separated by largedistances (i.e. when delaminated) cannot be placed com-pletely randomly in the sea of polymer. Furthermore, themajority of the polymer chains in the hybrids are teth-ered to the surface of the silicate layers. Thus, it can beexpected that there are domains in these materials, evenabove the melting temperature of the constituent poly-mers, wherein some long-range order is preserved and thesilicate layers are oriented in some preferred direction. Thislong-range order and domain structure is likely to becomebetter defined at the higher silicate contents, where thegeometrically imposed mean distance between the layersbecomes less than the lateral dimensions of the silicatelayers, thus forcing some preferential orientation betweenthe layers. However, there might be considerable polydis-persity effects in terms of the orientation and the distancebetween the silicate layers. Many such randomly orientedgrains make up the entire sample leading to the presenceof disordered material. Thus, in general the material pos-sesses a layered structure, with grains wherein the silicatelayers are oriented in a preferred direction leading to thepresence of grain boundaries and concomitant defects [54].

At this point, it is worth mentioning six interrelatedstructural characteristics, distinguishing polymer–silicatenanocomposites from conventional filled systems. Thesecharacteristics, which are attributed to the nanoscopicdimensions and the extreme aspect ratios of layered sili-cates, are [8]:

• Low percolation threshold (ca. 0,1–2 vol.%).• Particle–particle correlation (orientation and position)

arising at low volume fractions.


F achievaa

•

•

•

•

4

ct

coctBlia[tn

oohuttItssdlmelso

layered silicates initially do not exhibit well-defined basalreflections. Thus, peak broadening and intensity decreasesare very difficult to study systematically. Therefore, con-clusions concerning the mechanism of nanocomposite

ig. 6. Schematic illustration of two different types of thermodynamicallynd Bousima by permission of Elsevier Science Ltd., UK.

Large number of particles per particle volume (106–108

particles/�m3).Extensive interfacial area per volume of particles(103–104 m2/ml).Short distances between particles (10–50 nm at� ∼ 1–8 vol.%).Comparable size scales among the rigid nanoparticleinclusions, distance between particles, and the relaxationvolume of polymer chains.

.2. Nanocomposite structural characterization

Two complementary techniques are generally used toharacterize the structures of nanocomposites: XRD andransmission electron microscopy (TEM) [1,35,55,56].

Due to its ease of use and availability, XRD is mostommonly used to probe the nanocomposite structure andccasionally to study the kinetics of polymer melt inter-alation [55]. This technique allows the determination ofhe spaces between structural layers of the silicate utilizingragg’s law: sin � = n�/2d, where � corresponds to the wave

ength of the X-ray radiation used in the diffraction exper-ment, d the spacing between diffractional lattice planesnd � is the measured diffraction angle or glancing angle1,56]. By monitoring the position, shape and intensity ofhe basal reflections from the distributed silicate layers, theanocomposite structure may be identified [55].

For immiscible polymer/OMLS mixtures, the structuref the silicate is not affected, and thus, the characteristicsf the OMLS basal reflections do not change. On the otherand, in comparison with the spacing of the organoclaysed, the intercalation of the polymer chains increaseshe interlayer spacing, leading to a shift of the diffrac-ion peak towards lower angle, according to Bragg’s law.n such intercalated nanocomposites, the repetitive mul-ilayer structure is well preserved, allowing the interlayerpacing to be determined (Fig. 7). In contrast, the exten-ive layer separation associated with exfoliated structuresisrupts the coherent layer stacking and results in a feature-

ess diffraction pattern. Thus, for exfoliated structures no
ore diffraction peaks are visible in the XRD diffractograms
ither because of a much too large spacing between theayers (i.e. exceeding 8 nm in the case of ordered exfoliatedtructure) or because the nanocomposite does not presentrdering [1,35,47].

ble polymer/layered silicate nanocomposites [50]. Reproduced from Ray

The influence of polymer intercalation on the order ofthe OMLS layers may be monitored by changes in the full-width-at-half-maximum (fwhm) and intensity of the basalreflections. An increase in the degree of coherent layerstacking (i.e. a more ordered system) results in a relativedecrease in the fwhm of the basal reflections upon hybridformation. On the other hand, a decrease in the degreeof coherent layer stacking (i.e. a more disordered system)results in peak broadening and intensity loss [47].

However, although XRD offers a conventional methodto determine the interlayer spacing of the silicate layers inthe original layered silicates and the intercalated nanocom-posites (within 1–4 nm), little can be said about thespatial distribution of the silicate layers or any structuralinhomogeneities in nanocomposites. Additionally, some

Fig. 7. Typical XRD patterns from polymer/layered sili-cates: (a) PE + organoclay → no formation of a nanocomposite,(b) PS + organoclay → intercalated nanocomposite, (c) silox-ane + organoclay → delaminated nanocomposite [35]. Reproducedfrom Beyer by permission of Elsevier Science Ltd., UK.

ess in Po
formation and structure based solely on XRD patterns areonly tentative. On the other hand, TEM allows a qualitativeunderstanding of the internal structure and can directlyprovide information in real space, in a localized area, onmorphology and defect structures [57,58].

Since the silicate layers are composed of heavier ele-ments (Al, Si and O) than the interlayer and surroundingmatrix (C, H and N), they appear darker in bright-fieldimages. Therefore, when nanocomposites are formed, theintersections of the silicate sheets are seen as dark lineswhich are the cross sections of the silicate layers, measur-ing 1 nm thick. However, special care must be exercisedto guarantee a representative cross-section of the sample[56,57]. Fig. 8 shows the TEM micrographs obtained for anintercalated and an exfoliated nanocomposite. As alreadymentioned, besides these two well defined structures otherintermediate organizations can exist presenting both inter-calation and exfoliation. In this case, a broadening of thediffraction peak is often observed and one must rely on TEMobservation to define the overall structure [1].

Characterization of polymer/layered silicate nanocom-posites by 13C solid-state nuclear magnetic resonance (13CNMR) has also been proposed. VanderHart et al. first usedthis technique as a tool for gaining greater insight aboutthe morphology, surface chemistry, and to a very limitedextent, the dynamics of exfoliated polymer clay nanocom-posites. They were especially interested in developing NMRmethods to quantify the level of clay exfoliation, a veryimportant facet of nanocomposite characterization [59].
The main objective in solid-state NMR measurement is toconnect the measured longitudinal relaxations, T1s, of pro-ton (and 13C nuclei) with the quality of clay dispersion [55].
The surfaces of naturally occurring layered silicates suchas MMT are mainly made of tetrahedral silica, while the

Fig. 8. TEM micrographs of poly(styrene)-based nanocomposites: (a) intercalatedAlexandre and Dubois by permission of Elsevier Science Ltd., UK.

lymer Science 33 (2008) 1119–1198

central plane of the layers contains octahedrally coordi-nated Al3+ with frequent non-stoichiometric substitutions,where an Al3+ is replaced by Mg2+ and, somewhat less fre-quently, by Fe3+. The concentration of the later ion is veryimportant because Fe3+ is strongly paramagnetic in this dis-torted octahedral environment. Typical concentrations ofFe3+ in naturally occurring clays produce nearest-neighborFe–Fe distances of about 1.0–1.4 nm, and at such distances,the spin-exchange interaction between the unpaired elec-trons on different Fe atoms is expected to produce magneticfluctuations in the vicinity of the Larmor frequencies forprotons or 13C nuclei. The spectral density of these fluc-tuations is important because the T1

H of protons (and13C nuclei) within about 1 nm of the clay surface canbe directly shortened. For protons, if that mechanism isefficient, relaxation will also propagate into the bulk ofthe polymer by spin diffusion. Thus the paramagneticallyinduced relaxation will influence the overall measured T1

H

to an extent that will depend both on the Fe concentra-tion in the clay layer and, more importantly, on the averagedistances between clay layers. The latter dependence sug-gests a potential relationship between measured T1

H valuesand the quality of the clay dispersion. If the clay particlesare stacked and poorly dispersed in the polymer matrix,the average distances between polymer/clay interfaces aregreater, and the average paramagnetic contribution to T1

H

is weaker. VanderHart et al. also employed the same argu-ments in order to understand the stability of a particularOMLS under different processing conditions [55,60,61].

Some authors also used Fourier transform infraredspectroscopy (FTIR) to elucidate the structure of thenanocomposites [62,63]. FTIR may be able to identify differ-ences between the bonding in a mixture and the bondingin a related nanocomposite, but as these variations are

nanocomposite and (b) exfoliated nanocomposite [1]. Reproduced from

S. Pavlidou, C.D. Papaspyrides / Progress in Po

Fip

mFc

fiwtaiitsDimtibsoasmpemcbatn

saan

layers. Therefore, this technique, although widely used for

ig. 9. DSC traces of pure PS, a physical mixture of PS/OLS, and PS-ntercalated OLS [41]. Reproduced from Zanetti, Lomakin and Camino byermission of Wiley-VCH, Germany.

inute, even when intercalation has taken place, at presentTIR is an unreliable method of characterization in mostases [56].

Finally, differential scanning calorimetry (DSC) providesurther information concerning intercalation. The manynteractions the intercalated chains of the polymer form

ith the host species greatly reduce their rotational andranslational mobility. The situation is similar to that in

reticulated polymer, where restrictions on its mobilityncrease its glass transition temperature (Tg). A similarncrease is anticipated to occur in a nanocomposite dueo elevation of the energy threshold needed for the tran-ition. This effect is readily detected by DSC. Fig. 9 presentsSC traces of polystyrene (PS), a PS/OMLS mixture and an

ntercalated PS/OMLS nanocomposite. The PS and PS/OMLSixture curves clearly display the characteristic peak due

o glass transition of the polymer. The presence of this peakn the mixture is evidence of the absence of interactionsetween the organic and the inorganic phases. The tran-ition is absent in the nanocomposite curve and in factccurs at temperatures higher than those shown in Fig. 9. Inddition to being an interesting analytical datum, the con-iderable increase in Tg is an important property of theseaterials that enables them to be employed at higher tem-

eratures compared with the original polymer and thusxtends their fields of application [41]. To date, the afore-entioned subsidiary methods have only been used to

onfirm the evidence from the primary methods. However,uilding a clearer picture of the changes that occur whennanocomposite forms is important, as it not only helps

o characterize the material, but in principle could indicateovel methods of synthesis [56].

Concerning the evaluation of other nanocomposites
tructural characteristics, it should be noted that themount of clay present in a sample may be estimated,s for conventional composites, i.e. by placing pre-driedanocomposite pellets in a furnace at ca. 900 ◦C for approx-
lymer Science 33 (2008) 1119–1198 1129

imately 45 min. The resulting ash is then weighed andcorrected for loss of structural water [64].

However, unlike conventional fiber composites, thedetermination of filler aspect ratio for layered silicatenanocomposites is not straightforward. Good estimatesrequire a thorough analysis of TEM photomicrographsat different magnifications. Fig. 10 depicts various com-plications of calculating an aspect ratio from TEMphotomicrographs that arise from variations in bothlength/diameter, and thickness. Clay platelets intrinsicallyhave a distribution of lateral dimensions. The recovery,refinement, chemical treatment, and post-treatment ofthese clays may contribute to the variation in filler geom-etry. Furthermore, extrusion of these clays with polymerand any additional melt processing steps that follow, e.g.injection molding, will amplify the range of particle shapesand sizes, particularly when the layered silicate is notcompletely exfoliated, as illustrated in Fig. 10. Finally,microtoming of the nanocomposite sample into thin sec-tions for TEM analysis will also result in an apparentdistribution of observed particle sizes even if all disk-likeplatelets were the same size [64].

It becomes obvious from this section that a major issuewhen synthesizing polymer–layered silicate nanocompos-ites is the characterization of the product. In fact, many ofthe studies conducted so far in this field are solely dedi-cated to structural characterization of the nanocomposites,without reporting properties of the products.

5. Preparation of nanocomposites

5.1. Introduction

At present there are four principal methods for produc-ing polymer–layered silicate nanocomposites: (1) in situtemplate synthesis, (2) intercalation of polymer or prepoly-mer from solution, (3) in situ intercalative polymerizationand (4) melt intercalation [1,14,35,44,52,65].

5.1.1. Template synthesis (sol–gel technology)In this technique, the clay minerals are synthesized

within the polymer matrix, using an aqueous solution (orgel) containing the polymer and the silicate building blocks.As precursors for the clay silica sol, magnesium hydroxidesol and lithium fluoride are used. During the process, thepolymer aids the nucleation and growth of the inorganichost crystals and gets trapped within the layers as theygrow. Although theoretically this method has the poten-tial of promoting the dispersion of the silicate layers ina one-step process, without needing the presence of theonium ion, it presents serious disadvantages. First of all,the synthesis of clay minerals generally requires high tem-peratures, which decompose the polymers. An exception isthe synthesis of hectorite-type clay minerals which can beperformed under relatively mild conditions. Another prob-lem is the aggregation tendency of the growing silicate

the synthesis of double-layer hydroxide-based nanocom-posites, is far less developed for layered silicates and willnot be considered in the following discussion. However,it should be mentioned that several workers have suc-


o of laye
Fig. 10. Examples of complications in the determination of the aspect ratiFornes and Paul by permission of Elsevier Science Ltd., UK.
cessfully applied it for the preparation of nanocompositematerials. For example, Carrado et al. and Carrado andXu synthesized hectorites from gels consisting of silica,magnesium hydroxide, lithium fluoride and polymers likepoly(vinyl alcohol), polyaniline and polyacrylonitrile. Evi-dently, some silicate layers aggregated, but most of themremained uniformly distributed in the polymer matrix[1,3,41]

5.1.2. Intercalation of polymer or prepolymer fromsolution

Following this technique, the layered silicate is exfoli-ated into single layers using a solvent in which the polymer(or prepolymer in case of insoluble polymers, such as poly-imide) is soluble. It is well known that such layered silicates,owing to the weak forces that stack the layers together canbe easily dispersed in an adequate solvent. After the organ-oclay has swollen in the solvent, the polymer is added tothe solution and intercalates between the clay layers. Thefinal step consists of removing the solvent, either by vapor-ization, usually under vacuum, or by precipitation. Uponsolvent removal the sheets reassemble, sandwiching thepolymer to form a nanocomposite structure. Under thisprocess are also gathered the nanocomposites obtainedthrough emulsion polymerization where the layered sili-cate is dispersed in the aqueous phase. The major advantageof this method is that intercalated nanocomposites can besynthesized that are based on polymers with low or evenno polarity. However, the solvent approach is difficult toapply in industry owing to problems associated with theuse of large quantities of solvents [1,35].

5.1.3. In situ intercalative polymerization
In situ-polymerization was the first method used to syn-
thesize polymer–clay nanocomposites based on polyamide6. In this technique, the modified layered silicate is swollenby a liquid monomer or a monomer solution. The monomermigrates into the galleries of the layered silicate, so that

red silicate fillers within polymer nanocomposites [64]. Reproduced from

the polymerization reaction can occur between the inter-calated sheets. The reaction can be initiated either by heator radiation, by the diffusion of a suitable initiator orby an organic initiator or catalyst fixed through cationicexchange inside the interlayer before the swelling step bythe monomer. Polymerization produces long-chain poly-mers within the clay galleries. Under conditions in whichintra- and extra-gallery polymerization rates are properlybalanced, the clay layers are delaminated and the resultingmaterial possesses a disordered structure [1,35,37].

5.1.4. Melt intercalationThis technique consists of blending the layered silicate

with the polymer matrix in the molten state. Under suchconditions – if the layer surfaces are sufficiently compati-ble with the chosen polymer – the polymer can crawl intothe interlayer space and form either an intercalated or anexfoliated nanocomposite [1,35,37].

Among the aforementioned methods, in situ poly-merization and melt intercalation are considered ascommercially attractive approaches for preparing poly-mer/clay nanocomposites. Melt intercalation, in particular,is especially of practical interest, since it presents signifi-cant advantages that will be discussed in the correspondingparagraph.

5.2. Intercalation of polymer from solution

Intercalation of a polymer from a solution is a two-stageprocess in which the polymer replaces an appropriate, pre-viously intercalated solvent, as shown in Fig. 11. Such areplacement requires a negative variation in the Gibbs freeenergy. It is thought that the diminished entropy due to the
confinement of the polymer is compensated by an increasedue to desorption of intercalated solvent molecules. Inother words, the entropy gained by desorption of solventmolecules is the driving force for polymer intercalationfrom solution [66–71].


PLS obta

w[aniOthtassTpmtcKuOlofttCni

itmsd

nsitmc

the silicate platelets [51]. Fig. 13 represents the concep-tual view of the swelling behavior of �,�-amino acidmodified Na+-MMT by �-caprolactam [93]. In a typical syn-thesis, 12-aminolauric acid-modified MMT (12-MMT) was

Table 2PLS nanocomposites prepared by intercalation from solution.

Nanocomposite Solvent(s) Ref.

PVOH/Na+-MMT Water [86]PVA/Na+-MMT Water [87]TPU/OMLS Toluene/DMAc [88]PEO/Na+-MMT or

Na+-hectoriteAcetonitrile [89]

PEO/MMT Chloroform [90]PLA/OMLS Dichloromethane [83]PLA/OMLS DMAc [84]HDPE/protonated Xylene/benzonitrile [91]

Fig. 11. Schematic representation of

Even though this technique has been mostly used withater soluble polymers, such as PEO, PVE, PVP and PAA

3,21,72–77], intercalation from non-aqueous solutions haslso been reported [78–81]. For example, HDPE-basedanocomposites have been prepared by dissolving HDPE

n a mixture of xylene and benzonitrile with dispersedMLS. The nanocomposite was then recovered by precipi-

ation from THF [79]. PS/OMLS exfoliated nanocompositesave also been prepared by the solution intercalationechnique, by mixing pure PS and organophilic clay withdsorbed cetyl pyridium chloride [82]. Similarly, severaltudies have focused on the preparation of PLA-layeredilicate nanocomposites using intercalation from solution.he first attempts by Ogata [78], involved dissolving theolymer in hot chloroform in the presence of organo-odified MMT. However, TEM and WAXD analyses revealed

hat only microcomposites were formed and that an inter-alated morphology was not achieved. In a later study,rikorian and Pochan [83] prepared PLA nanocompositessing dichloromethane as the polymer solvent and as theMLS dispersion medium. The authors obtained interca-

ated or exfoliated nanocomposites, depending on the typef OMLS used. That is, exfoliated nanocomposites wereormed when diols were present in the organic modifier ofhe clay, due to the favorable enthalpic interaction betweenhese diols and the C O bonds in the PLA backbone.hang et al. [84] reported the preparation of PLA-basedanocomposites with different kinds of OMLS via solution

ntercalation using N,N′-dimethylacetamide (DMA).In the case of polymeric materials that are infusible and

nsoluble even in organic solvents, the only possible routeo produce nanocomposites with this method is to use poly-

eric precursors that can be intercalated in the layeredilicate and then thermally or chemically converted to theesired polymer [1,85].

Summarizing the above: although a number ofanocomposites have been produced by intercalation from
olution (representative examples are presented in Table 2),t is important to note that, in using this method, intercala-ion only occurs for certain polymer/clay/solvent systems,
eaning that for a given polymer one has to find the rightlay, organic modifier and solvents [1,50]. Moreover, from

ined by intercalation from solution.

the industrial point of view, this method may involve thecopious use of organic solvents, which is usually environ-mentally unfriendly and economically prohibitive [50].

5.3. In situ intercalative polymerization

5.3.1. In situ intercalative polymerization ofthermoplastic polymers

The Toyota research group first reported the abilityof �,�-amino acid (COOH–(CH2)n−1–NH2

+, with n = 2, 3,4, 5, 6, 8, 11, 12, 18) modified Na+-MMT to be swollenby �-caprolactam monomer at 100 ◦C and subsequentlyinitiate ring opening polymerization to obtain PA6/MMTnanocomposites [25]. The number of carbon atoms in the�,�-amino acid was found to have a strong effect on theswelling behavior as reported in Fig. 12, indicating thatthe extent of intercalation of �-caprolactam monomer ishigh when the number of carbon atoms in the �-aminoacid is large [93]. Moreover, it was found from a com-parison of different types of inorganic silicates that clayshaving higher CEC lead to more efficient exfoliation of

dodecylamine modifiedMMT

(80/20 wt.%)

PSF/OMLS DMAC [92]PI/dodecylammonium

modified MMTDMAC [85]


Table 3Peak intensity (Im) and interlayer spacing (d) of nylon-6-based nanocom-posites prepared in presence of different acid derivatives by the one-pottechnique [95].

Acid Im (cps) d (Å)

Phosphoric acid 0 0Hydrochloric acid 200 21.7Isophtalic acid 255 20.2

both the layered silicate modifier and the monomer. Theyfirst studied by XRD the dependence of the clay swellingprocess on ALA concentration in HCl, and found that it

Fig. 12. XRD patterns of �-amino acid [NH2(CH2)n−1COOH] modified Na+-MMT [93]. Reproduced from Usuki, Kawasumi, Kojima, Okada, Kurauchiand Kamigaito by permission of Materials Research Society, USA.

mixed with �-caprolactam and the mixture was heated at250–270 ◦C for 48 h to polymerize �-caprolactam, using 12-MMT as a catalyst (when the relative amount of 12-MMTin the mixture was less than 8 wt.%, 6-aminocaproic acidwas added as a polymerization accelerator and the heatingprofile was slightly modified). Depending on the amountof 12-MMT introduced, either exfoliated (for less then15 wt.%) or intercalated structures (from 15 to 70 wt.%) wereobtained, as evidenced by XRD and TEM. Comparison of thetitrated amounts of COOH and NH2 end groups present inthe synthesized nanocomposites with given values, such asthe CEC of the montmorillonite used (119 mequiv./100 g),have led to the conclusion that the COOH end groups
present along the 12-MMT surface are responsible for thepolymerization initiation [25].
Further work demonstrated that intercalative poly-merization of �-caprolactam could be realized without

Fig. 13. Swelling behavior of �-amino acid modified MMT by �-caprolactam [93]. Reproduced from Usuki, Kawasumi, Kojima, Okada,Kurauchi and Kamigaito by permission of Materials Research Society, USA.

Benzenesulfonic acid 280 19.3Acetic acid 555 20.3Trichloroacetic acid 585 21.3No acid 1840 18.6

modifying the MMT surface. Indeed, this monomer wasable to directly intercalate the Na+-MMT in water in thepresence of hydrochloric acid, as proved by the increase ininterlayer spacing from 10 to 15.1 Å. At high temperature(200 ◦C), in the presence of excess �-caprolactam, the clayso modified can be swollen again, allowing the ring openingpolymerization to proceed at 260 ◦C when 6-aminocaproicacid is added as an accelerator. The resulting compositedoes not present a diffraction peak in XRD, and TEM obser-vation agrees with a molecular dispersion of the silicatesheets [94].

In attempts to carry out the whole synthesis in one pot,the system proved to be sensitive to the nature of the acidused to promote the intercalation of �-caprolactam. Table 3gives results obtained for different acids in relation to theintensity (Im) of the XRD intercalation peak that might bepresent in the nanocomposites obtained (Fig. 14). Theseresults show that, for unclear reasons, only phosphoric acidallows for the preparation of a truly exfoliated nanocom-posite, the other acids tending to promote the formationof partially exfoliated-partially intercalated structures. Onecan also point out that an intercalated structure is obtainedeven if no acid is added [95].

Another polyamide, nylon 12, has also been reported toform nanocomposites via in situ intercalative polymeriza-tion. Reichert et al. [96] used 12-aminolauric acid (ALA) as

can be separated in two regimes: a cation-exchange ofinorganic cations by protonated ALA at low ALA concentra-

Fig. 14. XRD intensity curve of injection molded nylon-6 nanocompositeas obtained by the one-pot intercalation polymerization process in thepresence of acetic acid [95]. Reproduced from Kojima, Usuki, Kawasumi,Okada, Kurauchi and Kamigaito by permission of John Wiley & Sons, Inc.


FoiS

taeTttAaoscl

fnrtoiwTssdr

the preparation of PET-based nanocomposites. Because of

FR

ig. 15. Interlayer distance of fluoro-modified talc (ME 100) in functionf an increasing amount of aminolauric acid used as the organic mod-fier [96]. Reproduced from Reichert, Kressler, Thomann, Mulhaupt andtoppelmann by permission of Wiley-VCH, Germany.

ion and a further diffusion of zwitterionic 12-aminolauriccid into the interlayer space, when the ALA concentrationxceeds the amount of HCl in the medium (Figs. 15 and 16).he swelling was found to be independent of the swellingemperature, the layered silicate concentration and theype of acid used to protonate ALA (HCl, H2SO4, H3PO4).LA was then polymerized at high temperature (280 ◦C)nd under elevated pressure (ca. 20 bar) with both typesf swollen clay. XRD and TEM, coupled with energy disper-ive X-ray (EDX), as well as atomic force microscopy (AFM),onfirmed that the resulting structures were partially exfo-iated and otherwise intercalated nanocomposites.

However, although in situ polymerization was success-ully applied for the preparation of PA6 and PA12/clayanocomposites, few publications focused on the prepa-ation of polyamide from diamine and diacid. In one ofhese studies, Wu et al. [97] investigated the preparationf PA1012 nanocomposite by polycondensation polymer-zation. A dispersion of organoclay in absolute alcoholas added to 1,10 diaminodecane in absolute alcohol.

hen, this mixture was added to an absolute alcohol
olution of 1,10-decanedicarboxylic acid under vigoroustirring, resulting in the immediate precipitation of aiaminodecane–decanedicarboxylic acid salt. The salt wasecrystallized from a mixture of alcohol and water and
ig. 16. Schematic representation of the swelling behavior of the fluoro-modifieichert, Kressler, Thomann, Mulhaupt and Stoppelmann by permission of Wiley

lymer Science 33 (2008) 1119–1198 1133

was obtained as a white powder. It was then added witha slight excess of diaminodecane to a U-shaped glass tubewhich was purged with nitrogen before the reaction. Thetube was immersed in an oil bath and the temperature wasquickly raised to 200 ◦C to start the reaction. After main-taining the autoclave for 2 h at 200 ◦C, the temperature wasincreased to 215 ◦C and held for 1.5 h under these condi-tions. The glass tube was flushed with nitrogen each time toremove the water produced in polycondensation. In the laststep a vacuum (<0.1 atm) was applied to remove the waterand residual monomer, the temperature was increased to225 ◦C, and the reaction was continued for another 2 h.The glass tube was then cooled to room temperature andthe resulting PA1012/organoclay hybrid was obtained as awhite solid. The absence of peaks in the XRD pattern indi-cated the exfoliation of the clay platelets in the PA1012matrix.

The in situ polymerization technique has also beenapplied for the preparation of nanocomposites based onthermoplastic polymers other than polyamides, includingpolymethyl methacrylate (PMMA) [98,99], polystyrene (PS)[100], polybenzoxale (PBO) [101], polyolefins (PP and PE)[102–105], and polyethylene terephthalate (PET) [106].

For example, in a study discussing the synthesis of PETnanocomposites using in situ polymerization, the organo-modified montmorillonite is reported to react with PETcomonomers (ethylene glycol and terephthalic acid deriva-tives) to form an intercalated nanocomposite [106].

However, according to Lee et al. [107] endeavors toprepare PET nanocomposites using direct condensationpolymerization of diol and diacid result in formation ofoligomers with significantly low molecular weight, dueto ineffective control of stoichiometry; and thus, a largeincrease in intragallery distance is hard to obtain. Onthe other hand, attempts to prepare PET nanocompositesthrough melt intercalation have resulted in limited inter-calation of guest molecules, presumably due to the highviscosity of PET polymer. Therefore, the authors proposedand successfully applied the ring-opening polymerizationof ethylene terephthalate cyclic oligomers (ETCs) withorganically modified MMT, as an alternative approach to

low molecular weight and cyclic molecular architecture,cyclic oligomers of PET have much lower solution and meltviscosities than the corresponding polymer; so it may beexpected that, when clay intercalation is intended by mix-

ed talc ME 100 in presence of aminolauric acid [96]. Reproduced from-VCH, Germany.

1134 S. Pavlidou, C.D. Papaspyrides / Progress in Po

Fig. 17. Schematic representation of nanocomposite formation by ring-opening reaction of cyclic oligomers in between silicate layers [107].Reproduced from Lee, Ma, Rhee and Kim J by permission of Elsevier ScienceLtd., UK.

ing with cyclic oligomer instead of linear polymer, easierdiffusion and a higher degree of intercalation or exfoliationwould be obtained. In addition, problems such as precisecontrol of stoichiometry and high vacuum requirements,strictly required in preparation of PET by conventional con-densation polymerization of difunctional monomers, can
be effectively avoided through ring opening reaction ofcyclic oligomers (Fig. 17).
HDPE nanocomposites have been synthesized by theso-called polymerization-filling technique (PFT), whichinvolves anchoring in a first step a Ziegler-Natta type or any

Table 4Synthesis and composition of PE-based nanocomposite produced by in situ intmodified layered silicatesa [1].

Filler MAO (10−3 mol) Catalyst (10−6 mol) P(H

h 33.00 15.6 0m 27.20 12.5 0h 23.75 16.2 0.3

a h: hectorite and m: montmorillonite.b Hydrogen partial pressure at start.c Measured by thermogravimetric analysis (TGA).d Insoluble UHMWPE that cannot be eluted by SEC.

lymer Science 33 (2008) 1119–1198

other coordination catalyst onto a filler surface and thenin situ polymerizing ethylene directly from the surface-treated fillers. In order to apply this technique to nanofillers,layered silicates in aqueous colloidal suspension weremade less hydrophilic through elimination of water byfreeze drying. The fluffy materials obtained could then benicely dispersed in non-polar solvents such as heptaneor toluene. The clay dispersion was then surface-treatedwith MAO, and after solvent removal by evaporation, ahigh temperature treatment at 150 ◦C was applied to mod-ify the layered silicate. After removal of unreacted MAO,the silicate layers were contacted with a metallocenepre-catalyst, i.e. (tert-butylamido)dimethyl(tetramethyl-Z5-cyclopentadienyl) silane titanium dimethyl, for 1 h toform the active catalyst species. The polymerization wascarried out by adding ethylene to the medium. It hasto be pointed out that no ion exchange reaction wasrequired. Rather, this strategy relied on the immobiliza-tion of the active species through electrostatic interactionswith surface anchored MAO. Some typical in situ interca-lation experiments are gathered in Table 4. When ethylenepolymerization was carried out in the absence of molec-ular hydrogen (thus without any transfer agent), layeredsilicate/UHMWPE was produced, which is an extremely vis-cous material, very difficult to melt (Table 4, entries 1 and2). Addition of hydrogen to the polymerization mediumallows molecular weight to be reduced with substantialimprovement of melt processability (Table 4, entry 3).Examination of the TEM pictures reveals partial exfoliation,even at high filler content (ca. 30 wt.% montmorillonite)[108–110].

Similarly, Bergman et al. [103] used a palladium-basedcomplex and synthetic fluorohectorite as the polymeriza-tion catalyst and inorganic component respectively for thesynthesis of polyethylene nanocomposites. They first inter-calated palladium catalyst into the galleries of modifiedfluorohectorite (C14N-2) and exposed the dry powder toethylene gas. Over a 2 h period, they observed monomerconsumption and a dramatic increase in the size of thesilicate-catalyst composite, while after 12 h a large mass ofcolorless, rubbery polymer formed. Analysis of the toluene-extracted polyethylene by GPC revealed a high molecularweight. Moreover, the complete absence of diffractionpeaks in the XRD patterns strongly suggests the formation
of an exfoliated nanocomposite.
Jin et al. [105] also applied the in situ exfoliationmethod during ethylene polymerization by fixing a Ti-based Ziegler-Natta catalyst at the inner surface of MMT.They used organic salts with hydroxyl groups for the mod-

ercalative polymerization of ethylene (P(C2H4) = 10 bar) in non-organo-

2)b (bar) Filler loadingc (wt.%) HDPE Mn (g/mol)

4.2 –d

3.3 –d

3.4 77.000


FiK

iicfiTiAqmatat

iamemetmenps

moflacp

occurs in this case also; however this is not evident from

Fd

ig. 18. Mechanistic representation of the fixing of TiCl4 between the sil-cate layers of MMT–OH [105]. Reproduced from Jin, Park, Im, Kwak andwak by permission of Wiley-VCH, Germany.

fication of MMT (MMT–OH) since hydroxyl groups inntercalation agents offer facile reactive sites for anchoringatalysts between silicate layers. Fig. 18 represents the TiCl4xation mechanism between silicate layers of MMT–OH.he polymerization of ethylene was conducted by inject-ng ethylene into the catalyst slurry (30–50 ◦C, 4 bars).fter predetermined reaction times, polymerization wasuenched with a dilute HCl solution in methanol. The poly-er was precipitated in methanol, separated by filtration

nd dried in vacuum. The powder was then used as the mas-er batch and mixed with HDPE by melt extrusion. WAXDnalysis again revealed exfoliation of MMT layers in bothhe powdery reaction product and the HDPE matrix.

In a study by Shin et al. [111] montmorillonite wasntercalated with triisobutylaluminum and �-undecyleny-lcohol. The intercalation process allows the transitionetal catalyst and the activator methylaluminoxane to

nter the clay galleries and polymerize ethylene, while for-ation of polyethylene inside the galleries leads to the

xfoliation of layered silicates. The authors noted that twoypes of polymerization are possible: (a) the homopoly-

erization of ethylene, and (b) the copolymerization ofthylene and the vinyl ends of alcohol modifier con-ected to the surface during intercalation, which producesolyethylene chains, chemically connected to the silicateurface.

Heinemann et al. [112] carried out ethylene poly-erizations in the presence of layered silicates, such as

rgano-modified bentonite and unmodified hectorite and
uoromica, using 1-octene as comonomer. Different cat-lysts were used, affording HDPE and ethylene–octeneopolymers (zirconium-based catalyst) and branchedolyethylene (nickel and palladium-based catalysts). It
ig. 19. Schematic illustration of the modification and ion exchange of laponite wuced from Tudor, Willington, O’Hare and Royan by permission of The Royal Socie

lymer Science 33 (2008) 1119–1198 1135

appeared that the modified bentonites had a dramaticnegative effect on the polymerization activity of thezirconium-based catalyst due to its high sensitivity towardsany kind of polar functionality, while Ni-and Pd-basedcatalysts were much less affected by the nature of theclay. However, nanocomposites formed only in the case oforgano-modified clays, while in situ polymerization withunmodified silicates gave microcomposite structures.

PP/clay nanocomposites have also been prepared by insitu intercalative polymerization. Tudor et al. [104] treateda synthetic hectorite with methylaluminoxane (MAO) toremove the acidic protons and to prepare the interlayerspacing to receive the transition metal catalyst. Details ofthe preparation route are shown in Fig. 19. Using a syntheticfluorinated mica-type layered silicate deprived of protonsin the galleries, the catalyst was intercalated directly withinthe silicate layers, without the need of MAO treatment. Sunand Garces [102] also reported the preparation of PPCN byin situ polymerization with metallocene/clay catalysts.

Following the in situ intercalative polymerization tech-nique, Doh and Cho [113] compared the ability of severaltetraalkylammonium cations incorporated into Na+-MMTto promote the intercalation of PS through the free radicalpolymerization of styrene, initiated by AIBN at 50 ◦C. Threetetraalkylammonium cations were tested, all based on thefollowing formula: (CH3)2N·(hydrogenated tallow alkyl)R,where hydrogenated tallow alkyl corresponds to a mixtureof mainly octadecyl chains together with small amountsof lower linear homologs and R may be either anotherhydrogenated tallow alkyl (Ta), 2-ethyl hexyl (Eh) or benzyl(Bz) group. The so-modified o-MMTs were coded as Ta-MMT, Eh-MMT and Bz-MMT, respectively. Layer spacingsobtained for the three MMTs and corresponding compos-ites are presented in Table 5 and reveal that the bestintercalation occurs for Bz-MMT. This is probably due toa better affinity between styrene and benzyl groups spreadall along the layered MMT surfaces. It is worth mentioningthat even though the interlayer spacing for Ta-MMT doesnot change much, the authors assumed that intercalation

the d-spacings because such hydrogenated tallow alkylchains should be long enough (mainly C18 chains) to easilyaccommodate the PS. Even though this technique allows forextensive intercalation of PS chains through the choice of

ith [Zr(n–C5H5)2Me(thf)]BPh4 and propene polymerization [104]. Repro-ty of Chemistry, UK.


Table 5Interlayer spacings of organo-modified montmorillonites (X-MMT) andas-obtained PS-based nanocomposites and the clay dispersibility withinthe polymerization medium [113].

Xa-MMT Interlayer spacing (Å) Dispersibilityb

In X-MMT In PS/X-MMT

Sodium MMT 11.8 14.2 −BZ-MMT 19.1 34.0 ++Eh-MMT 20.4 28.5 +Ta-MMT 32.7 32.9 +

a Organo-modifiers: (CH3)2N+ (hydrogenated tallow alkyl)R with R = Bz(benzyl), Eh (2-ethylhexyl), or Ta (hydrogenated tallow alkyl).

show any significant increase in the layer spacing (13.6 Å).

b It was judged by the appearance of the montmorillonite dispersionin styrene monomer (++) fully dispersible, (+) partly dispersible and (−)non-dispersible.

an appropriate alkylammonium cation, neither exfoliationnor control over the molecular weight of PS produced hasbeen observed.

Akelah and Moet [100] modified MMT with vinylben-zyltrimethyl ammonium cation, and they dispersed andswelled the modified clay in various solvent and co-solventmixtures, such as acetonitrile or acetonitrile/toluene underN2. Styrene polymerizations were carried out in presenceof AIBN at 80 ◦C for 5 h. In this way, intercalated PS/MMTnanocomposites were produced, with the extent of inter-calation depending upon the nature of the solvent used.However, although PS is well intercalated, one drawbackof this procedure is that the macromolecule produced isnot a pure PS, but rather a copolymer between styrene andsurfactant.

A similar approach has been applied by Weimer et al.[114], who modified Na+-MMT by anchoring an ammoniumcation bearing a nitroxide moiety known for its ability tomediate the free radical polymerization of styrene in bulk.The absence of WAXD peaks together with TEM observa-tions of silicate layers randomly dispersed within the PSmatrix attest to the complete exfoliation of the layered sil-icate.

In another study, Yei et al. [115] prepared PS/claynanocomposites through emulsion polymerization, by sus-pending the clay in styrene monomer. The clay was treatedwith either cetyl-pyridinium chloride (CPC) or the inclusioncomplex of CPC in �-cyclodextrin (CPC/�-CD). This was thefirst study reporting cyclodextrins (CDs) as surfactants forintercalation in clay. CDs comprise a series of �-1,4-linkedcyclic oligosaccharides with shapes resembling hollowtruncated cones. The cavities of CDs are hydrophobic; thusCDs have the ability to include hydrophobic moleculeswithin their cavities. For the preparation of an inclusioncomplex, CPC was mixed with a saturated solution of �-CDin water and the complex was obtained as a white crys-talline precipitate. The modified clays were also obtainedas white precipitate after adding a solution of the surfac-tant (CPC or CPC/�-CD inclusion complex) in a Na+-MMTsuspension. Emulsion polymerization was performed asfollows: a surfactant solution (CPC or CPC/�-CD inclusion
complex) was added to an aqueous suspension of clay. Afterstirring for 4 h, KOH and SDS were added to the solutionand the temperature was raised to 50 ◦C. Styrene monomerand K2S2O8 were added slowly and polymerization was
lymer Science 33 (2008) 1119–1198

performed at 50 ◦C for 8 h. After cooling, 2.5% aqueous alu-minum sulfate was added to the polymerized emulsion,followed by dilute HCl with stirring. Finally, acetone wasadded to break the emulsion. XRD patterns indicate that theCPC surfactant is intercalated successfully into the galleriesof the clay, increasing their spacing from 1.43 to 2.27 nm.Furthermore, the d-spacing caused by the inclusion com-plex is 5.12 nm, i.e. substantially higher than that caused byCPC alone. This was explained by the authors consideringthat the linear, aliphatic chain within the CPC/�-CD cannotbend within the galleries of the clay. On the other hand, nopeak was detected in the XRD patterns for polymer/claynanocomposites prepared from the CPC and CPC/�-CDtreated clays which implies that they all possess exfoliatedstructures.

In another study, exfoliated PLA/clay nanocompositeswere prepared by in situ coordination-insertion polymer-ization method [116]. The authors used two different kindsof OMLS (C30B and C25A) for the preparation of nanocom-posites. In a typical synthetic procedure, the clay wasthoroughly dried and placed in the polymerization vial.l,l-Lactide solution in dried THF was then transferred tothe vial under nitrogen and the solvent was eliminatedunder reduced pressure. Polymerizations were conductedin bulk at 120 ◦C for 48 h, after 1 h of clay swelling in themonomer melt. When C30B was used, the polymeriza-tion was co-initiated by AlEt3, while Sn(Oct)2 was used tocatalyze the polymerization of l,l-lactide in the presenceof C25A. The clay C30B led to fully exfoliated structure,whereas C25A-based nanocomposites exhibited an inter-calated morphology.

Messersmith and Giannelis [117] reported the firstpreparation of PCL nanocomposites, in 1993, by in situintercalative polymerization, using Cr3+ exchanged fluo-rohectorite (FH). In a typical synthesis, a mixture of Cr3+

FH and CL was heated at 100 ◦C for 48 h. Upon cooling toroom temperature the reaction mixture solidified. Inter-calation of the CL monomer was revealed by XRD, whichshowed an increase in the silicate d-spacing from 1.28 to1.46 nm. The d0 0 1 spacing observed prior to polymerizationwas found to be consistent with the orientation of the CLring perpendicular to the silicate layers. XRD analysis of thenanocomposites after polymerization indicates a reductionin the silicate d-spacing from 1.46 to 1.37 nm, as presentedin Fig. 20. The decrease in the d-spacing is consistent withthe dimensional change accompanying polymerization ofCL monomer.

The same researchers also modified a Na+-MMT bythe protonated aminolauric acid and dispersed this mod-ified clay in liquid �-caprolactone before polymerizing itat high temperature. The PCL nanocomposites were pre-pared by mixing up to 30 wt.% of the modified clay with�-caprolactone at room temperature for two h, followed bythe ring opening polymerization at 170 ◦C for 48 h. Inter-estingly enough, XRD patterns of the modified clay aftercontact with �-caprolactone at room temperature do not

The authors assumed that the monomer intercalates in thegaps between the aminolauric acid chains so that no galleryexpansion could be seen, in contrast to what is usuallyobserved in in situ intercalative polymerization, where the


Fig. 20. Powder XRD of the composite before (solid line) and after (dashedlspp

mibhtp

ipTgrtodCdtpeaa

tpbntapoipisc

ine) polymerization. Insets are schematic illustrations (not drawn tocale) corresponding to the intercalated monomer (left) and intercalatedolymer (right) [117]. Reproduced from Messersmith and Giannelis byermission of the American Chemical Society, USA.

onomer insertion within the silicate gallery induces anncrease in the interlayer spacing. Another possibility maye that intercalation of the monomer occurs only during theeating step of the solution. After polymerization, XRD pat-erns of the obtained composites did not show a diffractioneak, indicating that exfoliation occurred [118].

Pantoustier et al. [119,120] compared PCL nanocompos-tes prepared by intercalative polymerization using bothristine MMT and �-aminododecanoic acid modified MMT.he polymerization of CL with pristine and modified MMTives PCL with a molar mass of 4800 and 7800 g/mol,espectively, and a narrow distribution. For comparison,he authors also conducted the same experiment with-ut MMT, but found no CL polymerization. These resultsemonstrate the ability of MMT to catalyze and controlL polymerization, at least in terms of a molecular weightistribution that remains remarkably narrow. Even thoughhe polymerization mechanism of �-caprolactone in theresence of clay generally remains unclear and differ-nt assumptions have been made [121–123], the authorsssumed that CL is activated through interaction with thecidic site on the clay surface.

On the other hand, Gorrasi et al. [6] suggested thathe low molecular weight of PCL synthesized by in situolymerization in the presence of layered silicate maye responsible for the high brittleness of the resultinganocomposites. Therefore, they proposed the prepara-ion of blends of high molecular weight PCL with differentmounts of o-MMT/PCL nanocomposite prepared via in situolymerization and containing a high amount (30 wt.%)f clay. More specifically, they added a suspension of an
ntercalated PCL nanocomposite (prepared by intercalativeolymerization of �-CL and having a basal spacing of 24 Å)
n chloroform to a chloroform solution of PCL and aftertirring for 15 h, they obtained the final composite by pre-ipitation into hexane. The XRD patterns of this product

lymer Science 33 (2008) 1119–1198 1137

showed a peak corresponding to a basal spacing of 18 Å,suggesting that the PCL is indeed intercalated within thesilicate layers of the clay in the blend that, however, under-went partial restructuring during the blend formation.

At this point, it is worth mentioning that, even thoughin situ intercalative polymerization has proved success-ful in the preparation of various polymer–layered silicatenanocomposites, important drawbacks of this techniquehave also been pointed out: (1) it is a time-consumingpreparation route (the polymerization reaction may takemore than 24 h); (2) exfoliation is not always thermody-namically stable; and the platelets may re-aggregate duringsubsequent processing steps; and (3) the process is avail-able only to the resin manufacturer who is able to dedicatea production line for this purpose [124].

5.3.2. In situ intercalative polymerization ofthermosetting polymers

Despite the aforementioned disadvantages of in situintercalative polymerization, this is the only viable tech-nique for the preparation of thermoset-based nanocom-posites, since such nanocomposites obviously cannot besynthesized by melt intercalation, which is the other com-mercially important preparation method [42,125–127].

In this case, the exfoliation ability of the organoclays isdetermined by their nature, including the catalytic effect onthe curing reaction, the miscibility with the curing agent,etc. Since there is a curing competition between intra-gallery and extragallery resin, as long as the intragallerypolymerization occurs at a rate comparable to the extra-gallery polymerization, the curing heat produced is enoughto overcome the attractive forces between the silicate layersand an exfoliated nanocomposite structure can be formed.In contrast, if the extragallery polymerization is more rapidthan the intragallery diffusion and polymerization or ifintragallery polymerization is retarded, the extragalleryresin will gel before the intragallery resin produces enoughcuring heat to drive the clay to exfoliate; consequently,exfoliation will not be reached. It can be inferred, therefore,that factors promoting the curing reaction of intragalleryresin will facilitate the exfoliation of the clay. Such factorsinclude the catalytic effect of organoclay on the curing reac-tion, the good penetrating ability of curing agent to clay, thelong alkyl-chain of the organo-cation, meaning a greateramount of intragallery resin preload and a completed orga-nization of the clay, and meaning weaker attractive forcesbetween the silicate layers [33,128].

In fact, a number of research groups have studied theeffect of various parameters on the exfoliation of claysin epoxy resins. Pioneering studies by Pinnavaia and co-workers [129] on MMT/epoxy systems established theinitial conceptual methodology. Interfacial modifiers, suchas primary ammonium alkyls are intercalated betweenthe MMT layers, not only to compatibilize the inorganicaluminosilicate and organic resin, but also to accelerate
the crosslinking reaction between the layers through acidcatalysis. That is, as the curing agent is mixed into theclay/epoxy mixture, it is thought that the modifiers intro-duced into the galleries of the clay sheets would promotethe reaction between the epoxy in the gallery with the cur-


Table 6d-Spacing of tixogel and OPTC18 at 5 phr in epoxy/amine systems at the beginning and at the end of polymerization.

Epoxy system + 5 phr Tixogel Epoxy system + 5 phr OPTC18

DGEBA/MCDEA

d-Spacing when reaction starts (Å) 33.2d-Spacing at the end of the reaction (Å) 33.9

ing agent. This would make the intragallery curing reactionfaster than the extragallery reaction, thus facilitating theexpansion of the clay sheets and helping to achieve exfoli-ation [130].

For the diglycidyl ether of bisphenol A (DGEBA),crosslinked with m-phenylene diamine (MPDA) and con-taining 5% modified MMT, Lan et al. [131] found that theclays with primary and secondary onium ions formed exfo-liated nanocomposites, whereas those with tertiary andquaternary onium ions retained an intercalated structure.It was argued that acidic alkyl ammonium ions tend to favorexfoliation by catalyzing homopolymerization of DGEBAmolecules inside the clay galleries. It was also found thatthe length of alkyl chains of modified MMT may determinewhether an intercalated and partially exfoliated or a totallyexfoliated nanocomposite will be obtained. The same kindof study was also conducted by Zilg et al. [132] who curedDGEBA with hexahydrophthalic acid anhydride in the pres-ence of different types of clays, again modified with a widevariety of surfactants.

Another parameter of the clay that greatly affectsthe outcome of nanocomposite processing is the cationexchange capacity (CEC), which determines the amount ofsurfactant ions present between the clay layers and there-fore controls the space available for diffusion of epoxymolecules during mixing with the organoclay. It has beenestablished that the highest CEC provides the minimumspace. In this context, the swelling phase is of critical impor-tance to the final nanocomposite structure. An MMT with alow CEC is exfoliated already during swelling in the epoxyresin prior to curing. A possible mechanism explaining thisphenomenon is homopolymerization of the epoxy resinduring the swelling phase, causing diffusion of new epoxymolecules into the clay galleries. The large amount of spaceavailable between the layers favors the diffusion. On theother hand, the duration of swelling of the clay with highCEC is shown to be critical for the synthesis of an exfoliatednanocomposite [42].

Other researchers investigated the effect of the polymerresin. For example, Becker et al. [133] prepared nanocom-posites of three different epoxy resins: diglycidyl ether ofbisphenol A (DGEBA), triglycidyl p-aminophenol (TGAP)and tetrafunctional tetraglycidyldiamino diphenylmethane(TGDDM), using a mixture of two diethyltoluene diamine(DETDA) isomers as the hardener and a commercially avail-able octadecyl ammonium ion modified MMT as the clay.All epoxy resin systems intercalated the organically mod-ified layered silicate and increased the d-spacing from 23
up to 80 Å. Similarly, Hackman and Hollaway [134] notedthat the epoxy resin component of the nanocomposite haslittle effect on the exfoliation of the clay layers; althoughit is the basic unit, the curing agent controls the rate ofcure. Lower viscosity resins lead to faster pre-intercalation,
DGEBA/D2000 DGEBA/MCDEA DGEBA/D2000

33.7 33.0 54.535.0 70 110

but they do not seem to offer any significant long-termadvantage.

Interestingly, much research has focused on the effectof curing agent on the intercalation/exfoliation of clays inepoxy resins. Messersmith and Giannelis [135] analyzedthe effect of different curing agents on the formation ofnanocomposites based on DGEBA and a montmorillonitemodified by bis(2-hydroxyethyl)methyl hydrogenated tal-low alkyl ammonium cation. They found that when primaryand secondary amines, such as methylene dianiline, wereused, only intercalated epoxy–clay structures could beobtained. This was attributed to either the bridging ofthe silicate layers by the bifunctional amine molecules,which prevents further expansion of the layers from tak-ing place, or to the strong polarity of the N–H groupsin the primary and secondary amines that causes a re-aggregation of dispersed silicate layers. When other curingagents, such as nadic methyl anhydride (NMA), boron triflu-oride monomethylamine (BTFA) or benzyldimethylamine(BDMA) were added, delamination during heating of thereaction mixture occurred. Addition of the curing agentinduced first an increase of the interlayer spacing from 36to 39 Å, indicating some partial intercalation. With furtherheating, disappearance of the interlayer spacing reflectionindicated that delamination had occurred. Study of thecuring reactions tended to prove that the particular alky-lammonium used (that bears two hydroxyl functions) couldplay an active role, especially when BDMA or NMA wereadded as the curing agents. For example, BDMA can catalyzethe reaction between the hydroxyl groups of the alkylam-monium and the oxirane of the monomer, producing a newhydroxyl that subsequently reacts with free DGEBA via asimilar base-catalyzed oxirane ring opening to build up theepoxy network.

Recently, Le Pluart et al. [136] investigated the influ-ence of curing agent and clay organophilic treatment onthe reactivity and cure behavior of epoxy networks andon the morphology of the final composites. They used twodifferent curing agents: an aliphatic diamine with a poly-oxypropylene backbone (D2000) and 4,4′-methylenebis[3-chloro-2,6-diethylaniline] (MCDEA) as well as two mont-morillonites modified with different alkyl ammonium ionshaving the same chain length. The first MMT is Tixo-gel MP250, a benzyl dimethyl tallow alkyl ammoniumMMT, while the second is OPTC18, an octadecylammo-nium ion modified MMT. The authors noticed that geltime was decreased in the presence of Tixogel for boththe DGEBA/D2000 and the DGEBA/MCDEA systems. How-
ever, for unclear reasons the authors could not determinea gel time in the presence of OPTC18. Concerning the influ-ence of network formation on organoclay dispersion, thed-spacings of the clays at the beginning and the end ofpolymerization are presented in Table 6. In the case of

ess in Po

TodmcooOdbidnoitr

mawlai

hldcatwfifarat

itfipibpfh

ipbpaiU(fiwi

S. Pavlidou, C.D. Papaspyrides / Progr

ixogen, both reactive mixtures have swelled the organ-clay already at the beginning of the reaction, since the-spacing increased from 20.2 up to 34 Å. During the poly-erizations the d-spacing remains the same, whatever the

uring agent used, demonstrating that the polymerizationf the epoxy/amine systems does not modify the dispersionf Tixogen. The observations are very different in the case ofPTC18. Again the initial state of dispersion is good, but the-spacing further increases during isothermal reaction ofoth reactive mixtures. This demonstrates the possibility of

mproving the quality of dispersion at the nanometer scaleuring the polymerization of the network. The authorsoted that such improvement is not only due to the kind ofrganoclay used nor to the epoxy system chosen, but rathers linked with interactions and chemical affinities betweenhe organoclay and the network precursors, as well as witheaction and diffusion kinetics of the reactive systems.

In another study, Chin et al. [33] confirmed the for-ation of exfoliated nanocomposites when DGEBA was

uto-polymerized with MMTs. Exfoliated nanocompositesere also observed with DGEBA cured with MPDA of

ess than equimolar concentration. However, as the curinggent concentration increased, the extragallery crosslink-ng dominated, resulting in intercalated nanocomposites.

Among the various parameters, the effect of processingas also been investigated. Jiankun et al. [128] interca-

ated an organically modified MMT by epoxy resin by bothirect or solution mixing. XRD patterns indicated that inter-alation was realized irrespective of the mixing methodpplied. It was also concluded that prolonging the stirringime above a certain level or using solution intercalationould not further improve the intercalation. An importantnding was that the intercalated hybrids were quite stable

or storage. In order to prepare nanocomposites, a curinggent was added into the hybrid and mixed thoroughly. Theesults indicated that if the organoclays can be exfoliatedt all, the exfoliation will be finished before the gel point ofhe epoxy.

Koerner et al. [137] studied the impact of shear dur-ng epoxy nanocomposite processing and demonstratedhat with proper mechanical processing conditions, uni-orm dispersion and a high degree of exfoliation is possiblen systems that typically only show intercalated mor-hologies after traditional cure cycles. Conceptually, this

s achieved by maximizing epoxy viscosity by halting cureefore gelation and by compounding at sub-ambient tem-eratures near the resin’s glass transition. High shearorces, due to the high viscosity of the system, facilitateomogenization of the layered silicate nanocomposite.

The so-called “high pressure mixing” (HPM) method,nvolving clay processing in solvent (acetone) under highressure and also against solid obstacles, has been appliedy Liu et al. [130] for the preparation of epoxy nanocom-osites. Using this method, nanoclays can be dispersed incetone and an epoxy solution to form a stable suspension,n which the basal spacing of the nanoclay is increased.
sing TGDDM as the matrix, 4,4′-diaminodiphenyl sulfone
DDS) as the hardener and an octadecyl amine modi-ed MMT, nanocomposites of up to 7.5 wt.% clay loadingere successfully synthesized with the HPM method. TEM

mages show that the agglomerates of nanoclays were bro-

lymer Science 33 (2008) 1119–1198 1139

ken down to form small particles consisting of several clayplatelets.

Similarly, the “slurry compounding” approach hasbeen developed for epoxy/clay nanocomposite preparationusing sodium MMT [138]. The most significant feature ofthis technique is that very little (<5 wt.%) organic modi-fier is required to facilitate the exfoliation and dispersionof the clay, reducing the cost of the nanocomposites. Tofurther reduce the cost of polymer/clay nanocompositesWang et al. prepared epoxy/crude clay nanocompositesusing the “slurry-compounding” technique. A clay–acetoneslurry was mixed with required quantities of epoxy resin at50 ◦C. Acetone was then evaporated and a stoichiometricquantity of the curing agent was added. The mixture wasdegassed and dried, leading to the formation of orderedexfoliated nanocomposites.

As for epoxy-based nanocomposites, the synthesis ofunsaturated polyester (UP)/layered silicate nanocompos-ites involves two steps: first the mixing process, whereinthe UP linear chains are mixed with the curing agentand layered silicate and second the curing process, duringwhich the crosslinking reaction takes place by decompos-ing the initiators [139].

The synthesis of nanocomposites based on montmo-rillonite and UP has been reported by different researchgroups. In a typical example, Bharadwaj et al. [140]described the preparation of crosslinked polyester/claynanocomposites by dispersing organically modifiedMMT in pre-promoted polyester resin and subsequentlycrosslinking the system using methyl ethyl ketone per-oxide (MEKP) catalyst at several clay concentrations. Theformation of exfoliated nanocomposites was confirmed byXRD and TEM.

In another study, montmorillonite was treated withmethacrylate-silane coupling agent in order to rend thefiller hydrophilic and reactive. Then, UP was polymerized bythe free radical polymerization with the modified montmo-rillonite dispersed in it. The authors claimed the formationof exfoliated structure, based on XRD and TEM findings[127].

Furthermore, Suh et al. [139] reported on the formationmechanism of UP nanocomposites. They used two differ-ent kinds of MMT: a dodecyl ammonium bromide MMTand Cloisite 20A, containing dimethyl dehydrogenated tal-low ammonium as an organic modifier, and used twodifferent ways of mixing in order to prepare UP-basednanocomposites, i.e. simultaneous mixing or sequentialmixing, where in the first step, pre-intercalates of the UPand MMT nanocomposites (i.e. the mixtures of UP and mod-ified MMT) are prepared; and then the styrene monomer,acting as the curing agent, is added with varying mixingtimes from 15 to 180 min. Finally, all mixtures were curedat 80 ◦C for 3 h and post-cured at 120 ◦C for 4 h. The struc-tures of UP/MMT nanocomposites were investigated byXRD and TEM, whereas in order to investigate the forma-tion mechanism of UP/MMT nanocomposites, the authors
used DMTA, solution rheometry and melt rheometry. Theresults led the authors to suggest the following mecha-nism of UP/silicate nanocomposite formation: The styrenemonomer moves more easily than uncured UP chains. Thismay generate higher styrene monomer concentration in


Table 7Polyurethane characteristics.

PU reference PU nanocomposite Polyol

Name Calculated Mn Functionalitya Wt.% EOb end

I NC-I Acclaim 2220 2000 2 15II NC-II Acclaim 4220 4000 2 15III NC-III Daltocel F435 4000 2.5 17
IV NC-IV Arcol 1374
a Approximated.b EO = ethylene oxide.

the MMT gallery than in any other part in a simultaneousmixing system. If polymerization occurs in these condi-tions, the total crosslinking density of the sample decreases,because of the low concentration of styrene in uncuredUP linear chains. In the sequential mixing method, thestyrene monomer diffuses into the gallery of MMT interca-lated with UP as time goes on. Therefore, it is thought thatcrosslinking density and Tg of UP/silicate nanocompositeincrease to some extent. Hence, the styrene monomers aremore easily dispersed inside and outside the silicate lay-ers as mixing time increases. Therefore, the crosslinkingreaction takes place homogeneously inside and outside thesilicate layers, and crosslinking density reaches the degreeof crosslinking density of the cured pure UP.

A number of researchers have synthesized polyurethane(PU)-based nanocomposites by the in situ polymeriza-tion method. The first examples of elastomeric PU/claynanocomposites with improved properties compared tothe pristine polymer were reported by Wang and Pinnavaia[141]. Conventional PU microcomposites are usually for-mulated by premixing the inorganic component with thepolyol and then curing the mixture with the diisocyanate.The approach of Wang and Pinnavaia to form polyurethanenanocomposites, therefore, focused on the solvation of theorganoclay by polyols. Interestingly, they found that mont-morillonites exchanged with long chain onium ions (carbonnumber > 12) are easily solvated by several polyols that are
commonly used in polyurethane chemistry.
More recently, Yao et al. [142] reported the preparationof a novel kind of PU/MMT nanocomposite using a mixtureof modified 4,4′-diphenyl methylate diisocyanate (MMDI),modified polyether polyol (MPP) and Na+-MMT. In a typical

Fig. 21. Preparation of PU/clay nanocomposites [144]. Reproduced from Berta, Lin

6000 2.3 15

synthetic route, a known amount of Na+-MMT was blendedwith a known amount of MMDI and cured at 78 ◦C for 168 h.As measured through XRD, the gallery spacing of the lay-ered clay is 1.1 nm and increases to 1.6 nm for the PU/clay(21.5%) nanocomposites, indicating that the PU chains wereintercalated between the layers of the clay.

On the other hand, Mulhaupt and co-workers [143]prepared PU nanocomposites from modified reactivefluoromica clay. The dried organophilic mica was dis-persed by means of a high shear mixer in trihydroxy-terminated oligo-propylene-oxide. Stable and transparentpolyol dispersions were obtained and then cured withdiisocyanatophenylmethane and accelerated with 0.6 wt.%N,N-dimethyl-benzyl-amine.

Berta et al. [144] synthesized elastomer polyurethanenanocomposites showing several degrees of dispersion,using polyols with different molecular weight and func-tionality (see Table 7) as well as methylenediphenylenediisocyanate (MDI). For the preparation of nanocompos-ites, the desired weights of polyol and organoclay weremixed and then butanediol (chain extender), catalyst andMDI were added in appropriate amounts. The system wascured at 120 ◦C for 3 h and 80 ◦C for 24 h. The synthesisof the PU/clay nanocomposites is schematically illustratedin Fig. 21. Table 8 summarizes the d-spacings determinedby SAXS for the PU/clay nanocomposites. It is clear thatall the PU/clay materials are nanocomposites, as the 18 Å
d-spacing, associated with the base organoclay has dis-appeared. No peaks associated with clay gallery spacingswere observed for NC-II and NC-IV, indicating that in thesenanocomposites there is no longer sufficient ordering ofthe clay platelets to produce a scattering peak. The authors
dsay, Pans and Camino by permission of Elsevier Science Ltd., UK.


Table 8d-Spacings for organoclay and PU/clay nanocomposites.

Material Clay d-spacing

Cloisite 30B (organoclay) 18NNNN

p(Asct

iawodop4poCtat

Fo

C-I 65C-II NoneC-III 102C-IV none

ostulated that this is due to the higher equivalent Mw

Mn/functionality) of the polyols in these nanocomposites.ccording to the authors, this is of particular importance,ince it means that the dimensions of the polyol moleculeontrol the gallery spacing in the initial step of dispersinghe clay in polyol.

Finally, Gao et al. [145] prepared PU nanocompos-tes, as well as their foams by in situ polymerizationnd batch foaming with different modified MMTs. MMTas modified by dibutyldimethoxytin (DBDMT) and, thus,rganophilic montmorillonite with a catalytic function,enoted as MMT–Tin, was obtained. For the synthesisf PU nanocomposites and foams, the authors used aolymeric aromatic isocyanate based on diphenylmethane,4′-diisocyanate (MDI) and two trifunctional polyesterolyols. For PU nanocomposites, clay was first mixed with
ne monomer and then the second monomer was added.atalyst was always added with polyol. After polymeriza-ion under ambient conditions, the hybrid was post-curedt 100 ◦C for 4 h. For reactive foaming of PU, a surfac-ant as well as pentane (blowing agent) were also used.
ig. 22. SEM micrographs of PU foams at cross-sections parallel to foam rising diref Elsevier Science Ltd., UK.

lymer Science 33 (2008) 1119–1198 1141

The mixture of all ingredients was mixed and foamingoccurred in a closed plastic container with fixed volumeat ambient temperature, followed by curing at 100 ◦C. ByXRD analysis it was found that the basal spacing of theorganoclay increased compared to MMT (d0 0 1 = 1.16 nm),because the gallery of MMT was expanded by molecularchains of the modifier. However, the d-spacing decreasedfrom 1.77 to 1.43 nm when MMT–OH was further mod-ified by DBDMT—an effect attributed to conjugation ofhydroxyl groups from different layers. The authors stud-ied the effect of the clay–monomer mixing sequenceon clay dispersion and found that the two-step processoffered better clay dispersion than the one-step approach,wherein all ingredients were mixed simultaneously. Espe-cially, premixing the functional clays with isocyanateprovides better clay dispersion, which was attributed tothe reaction between the isocyanate monomers and thehydroxyl groups on alkyl chains of MMT–OH, causing anincrease of gallery spacing of clay. In the XRD spectra nodiffraction peak was observed for PU nanocomposite withMMT–Tin. In fact, TEM analysis revealed that nanocom-posites containing both MMT–OH and MMT–Tin exhibitedgood clay dispersion, however MMT–Tin showed betterexfoliation and more uniform dispersion, probably dueto the intragallery catalysis of organotin. Turning to the
reactive foaming of PU nanocomposites, Fig. 22 showsSEM images of the freeze fractured surface of PU foams.It is clearly observed that the neat PU foam has fewercells and a larger cell size than PU nanocomposite foamswith 5 wt.% organoclay, whereas there is little difference
ction [145]. Reproduced from Gao, Lee, Widya and Macosko by permission


Table 9PLS nanocomposites prepared by in situ polymerization.

Nanocomposite Monomer Polymerization conditions Ref.

PA6/12-aminolauric acid modifiedMMT

�-Caprolactam Ring opening, 250–270 ◦C, 48 h [25]

PA12/Somasif ME 100 �-Amino dodecanoic acid (ADA) 280 ◦C, 20 bar, 9.5 h (after an initial swelling stage) [44]PCL/aminolauric acid modified MMT �-Caprolactone Ring opening, 170 ◦C, 48 h [117]PCL/Cr3+ exchanged fluorohectorite �-Caprolactone 100 ◦C, 48 h [118]PE/MMT with fixed Ti-based

Ziegler-Natta catalystEthylene 30–50 ◦C, 4 bars [105]

PMMA/OMLS MMA Free radical, 80 ◦C, 5 h [98,99]PS/OMLS Styrene Free radical, 100 ◦C, 16 h [98,99]PS/vinylbenzyl-trimethyl-ammonium-

MMTStyrene Free radical, 80 ◦C, 5 h, AIBN [100]

PET/o-MMT Ethylene glycol, terephthalic acid derivatives [106]Epoxy/o-MMT DGEBA Crosslinking, diamine curing agent [125]
Epoxy/octadecylammonium modified
MMTDGEBA, TGAP, TGDDM

UP/MMT treated with amathacrylate-silane coupling agent

Unsaturated polyester

between MMT–Tin/PU and MMT–OH/PU nanocompositefoams in terms of cell size and density. The appearanceof a shoulder at very low angle in the XRD patterns ofnanocomposite foams implies that clay orientation anddispersion is somewhat affected by the foaming process.However, as the authors noted, the detailed mechanism ofhow nanoparticles influence cell morphology needs furtherinvestigation.

A list of representative thermoset and thermoplastic –based nanocomposites prepared through in situ polymer-ization is given in Table 9.

5.4. Polymer melt intercalation

5.4.1. Introduction and advantages of the techniqueFor most technologically important polymers, both

in situ polymerization and intercalation from solutionare limited because neither a suitable monomer nora compatible polymer–silicate solvent system is alwaysavailable. Moreover, they are not always compatible withcurrent polymer processing techniques. These disadvan-tages drive the researchers to the direct melt intercalationmethod, which is the most versatile and environmentallybenign among all the methods of preparing polymer–claynanocomposites (PCNs) [4,146].

As already mentioned, nanocomposite synthesis viapolymer melt intercalation involves annealing, usuallyunder shear, of a mixture of polymer and layered silicateabove the softening point of the polymer. During anneal-ing, polymer chains diffuse from the bulk polymer melt into

Fig. 23. The ‘melt intercalation’ process [35]. Reproduced fro

DETDA curing agent [126]

Free radical crosslinking [127]

the galleries between the silicate layers, as shown in Fig. 23[13,15,47].

The advantages of forming nanocomposites by melt pro-cessing are quite appealing, rendering this technique apromising new approach that would greatly expand thecommercial opportunities for nanocomposites technology[14,15,46,47]. If technically possible, melt compoundingwould be significantly more economical and simpler thanin situ polymerization. It minimizes capital costs because ofits compatibility with existing processes. That is, melt pro-cessing allows nanocomposites to be formulated directlyusing ordinary compounding devices such as extrudersor mixers, without the necessary involvement of resinproduction. Therefore, it shifts nanocomposite productiondownstream, giving end-use manufacturers many degreesof freedom with regard to final product specifications (e.g.selection of polymer grade, choice of organoclay, level ofreinforcement, etc.). At the same time, melt processingis environmentally sound since no solvents are required[14,15]; and it enhances the specificity for the intercalationof polymer, by eliminating the competing host-solvent andpolymer–solvent interactions [29].

Thus, the majority of thermoplastic polymers, includ-ing PA [14,36,51,147], PET [148] (and recycled PET [149]),EVA [81,150], thermoplastic polyurethane [88], polyolefins
[111,151], PLA [152–154], PCL [155,156], etc., have been usedto study nanocomposite formation by melt intercalation.
Before discussing in detail the factors affecting claydelamination during melt blending as well as the degrada-tion issues involved in this technique, it is worth describing

m Beyer by permission of Elsevier Science Ltd., UK.

ess in Po

atp

abasppcwmiflip2tSpstbidittiotlbTEWl

(sspti

FL


slightly modified approach aiming to facilitate exfolia-ion and consisting of melt processing the polymer with are-intercalated clay or slurry.

Following such an approach, Liu and Wu [146] preparedco-intercalated clay, by absorbing epoxide compound

etween the silicate layers. They expected strong inter-ction between PA66 and this new kind of modified clay,ince Ishida had successfully prepared PA6, PA12 and otherolymer nanocomposites using similarly modified clay. Thereparation of the new kind of co-intercalated organophiliclay used was as follows. Na+-MMT was dispersed in hotater using a homogenizer. Then, hexadecyltrimethylam-onium bromide, dissolved in hot water, was poured

nto the Na+-MMT–water solution with vigorous stirringor 30 min to yield a white precipitate, which was col-ected and washed with hot water. After thorough dryingn a vacuum oven, the precipitate was ground into aroduct termed PrEMMT. PrEMMT and epoxy resin GY40 were mixed in a Haake Reocorder 40 mixer for 1 h;hus the co-intercalated clay, termed EMMT was obtained.ubsequently, a twin-screw extruder was used for thereparation of nanocomposites. In XRD patterns, Na-MMThowed a characteristic diffraction peak corresponding tohe (0 0 1) plane at 1.24 nm. PrE-MMT showed a 1.96 nmasal spacing in the XRD pattern, while the basal spac-

ng of E-MMT was 3.77 nm. The obviously larger layeristance of E-MMT demonstrates the advantage of co-

ntercalated organophilic clay. As the authors pointed out,he alkylammonium ion exchange enables conversion ofhe hydrophilic interior clay surface to hydrophobic andncreases the layer distance as well. This is the conditionf PrEMMT. In this organophilic environment, epoxy resinhen diffuses into the clay galleries to further increase theayer distance. In addition, the co-intercalated clay alsorings the active functional group into the PA66 system.herefore, a better dispersion effect can be expected forMMT. In fact, using the aforementioned procedure, Liu andu obtained well exfoliated nanocomposites when clay

oading was less than 7 wt.%.Also in the case of polyolefins, the use of a swelling agent

a monomer or polymer known to intercalate/exfoliate
mectite clay) next to the surfactant placed at the clayurfaces after an ion-exchange reaction has allowed thereparation of nanocomposites. Present in small amounts,he swelling agent serves to swell the clay layers, allow-ng the organic matrix to be virtually any polymer [2].
ig. 24. The synthetic route for the formation of the terpolymer MAST [159]. Reptd., UK.

lymer Science 33 (2008) 1119–1198 1143

As an example, Wolf et al. [157] modified a commerciallyavailable organo-ammonium-exchanged montmorillonite,using an organic swelling agent (with boiling pointbetween 100 and 200 ◦C, such as ethylene glycol, naph-tha or heptane) in order to increase the interlayer spacing.The swollen organo-modified clay was then compoundedwith PP in a twin-screw extruder at 250 ◦C. The swellingagent was volatilized during extrusion, leading to the for-mation of nanocomposites. Similarly, Liu and Wu [158]reported the preparation of PP nanocomposites via melt-compounding, using a type of co-intercalated organophilicclay. One of the co-intercalation monomers is unsaturated,so it could tether on the PP backbone through a graftingreaction. The co-intercalated organophilic clay (EM-MMT)was prepared as follows. Hexadecylammonium modifiedMMT (C16-MMT) was mixed with epoxypropyl methacry-late in a Haake mixer for 1 h. Before mixing with clay,the initiator for the grafting reaction, dibenzoyl peroxide(BPO) and a donor agent were dissolved in epoxypropylmethacrylate. The nanocomposites were prepared usinga twin-screw extruder with a screw speed of 180 rpmoperating at 200 ◦C. WAXD patterns and TEM observationsestablished that the larger interlayer spacing and the stronginteraction caused by grafting can improve the dispersioneffect of silicate layers in the PP matrix.

In another study, Zheng et al. [159] used an oligomer-ically modified clay, prepared by ion-exchange withthe oligomer prepared from maleic anhydride (MA),styrene (ST) and vinylbenzyltrimethylammonium chlo-ride (VBTACl) terpolymer, herein called MAST, to preparePS/clay nanocomposites by melt blending. The syntheticroute for the formation of the terpolymer MAST is depictedin Fig. 24. Thereafter, a portion of MAST oligomer, dis-solved in acetone was added drop-wise to a dispersionof clay in distilled water and acetone. A precipitate(MAST hectorite clay) formed immediately. Nanocom-posites were subsequently prepared by melt blendingin a Brabender Plasticorder at 60 rpm and 190 ◦C for15 min. XRD measurements indicated a mixed interca-lated/delaminated structure for the MAST modified clay,whereas no peaks were observed for the PS/MAST. By com-
bining XRD and TEM analyses the authors concluded thatthe hybrids formed were characterized by a mixed immis-cible/intercalated/delaminated structure.
Recently, Hasegawa et al. [160] reported a novelcompounding process for the preparation of PA6/MMT

roduced from Zheng, Jiang and Wilkie by permission of Elsevier Science

ess in Po
nanocomposites, using a Na+-MMT water slurry as an alter-native for organically modified MMT. In this process, theNa+-MMT slurry was blended with PA6 using an extruder,followed by removal of the water. WAXD patterns and TEMobservations clearly indicate the exfoliation of MMT layersin the PA6 matrix but the final properties of PA6/Na+-MMTnanocomposites were nearly equal to those of conventionalPA6/MMT nanocomposites prepared by dry compoundingwith MMT. Fig. 25 shows schematically dispersion of theNa+-MMT silicate layers of the clay slurry into the PA6matrix during compounding by an extruder. According tothis study, the exfoliation of silicate layers into the PA6matrix occurs as follows: (a) the clay slurry is first pumpedinto the melting matrix under vigorous shear; (b and c) theclay slurry reduces to finer drops during blending and, atthe same time, the water of the slurry drops begins to evap-orate in contact with the PA6 melt; (d) the evaporated wateris removed under vacuum, and silicate layers are dispersedinto the PA6 melt as monolayer or as a few layers. Thedispersion of silicate layers in this process is quite differ-ent from that of conventional compounding process usingorganophilic clay, where polymer chains first intercalateinto the stacked silicate galleries and then exfoliate into
the matrix. In this process, the exfoliated silicate layers aredirectly fixed in the polymer matrix without aggregation ofthe silicate layers.
Fig. 25. Schematic figures depicting dispersion of the Na+-MMT slurry into nyloKato, Usuki and Sato by permission of Elsevier Science Ltd., UK.

lymer Science 33 (2008) 1119–1198

5.4.2. Factors affecting polymer melt intercalationPredicting whether or not a polymer–silicate nanocom-

posite will form through melt compounding is notstraightforward, as a wide variety of factors influence theoutcome. These include energy changes, arising from theconfinement of the polymer within the silicate, the expan-sion of the spaces between the layers of the silicate, andthose associated with intermolecular interaction amongsilicate surface, tethered chain and polymer. With a viewto improving predictability, attempts have been made tomodel the behavior of hybrids that form as a result of directmelt intercalation with organically modified clays and toassess the parameters required to favor intercalation [56].Some of the questions that need to be addressed are, forexample, why do certain polymer–silicate systems favorintercalated hybrids, others delamination, and yet othersare immiscible, leading to microcomposites? How doespacking density and chain length of the alkylammoniumchains in the organosilicate layer, charge of the silicate orspecific groups on the polymer (or the alkyl chains) affecthybrid formation and miscibility? How does temperatureor shear affect processing? And finally, how does the typeof bonding at the polymer/silicate interface (i.e. hydro-
gen, dipole–dipole, van der Waals or covalent in which thealkylammonium chains become part of the polymer chain)affect the properties of the hybrid [4].
n-6 during compounding [160]. Reproduced from Hasegawa, Okamoto,

ess in Po

5qitmtcpitnoHltofiaatfiptem[

Fmodm


.4.2.1. Thermodynamic aspects. To address some of theseuestions Giannelis et al. focused on the thermodynam-cs governing nanocomposite formation. To that end,hey developed a mean-field, lattice-based thermodynamic

odel. Assuming the configurations and interactions ofhe various constituents are independent, the free energyhange of hybrid formation can be separated into inde-endent enthalpic and entropic terms. The entropic term

s the sum of the configurational changes associated withhe polymer and the silicate (including the alkylammo-ium chains in organosilicates). Configurational changesf the silicate are determined using a modified Flory-uggins lattice model in which the occupation of the

attice is weighted to simulate the preferred orienta-ions of the alkylammonium cations in the presencef two impenetrable surfaces (silicate layers). The con-nement of the intercalated polymer chains is similarlypproximated using a self-consistent field treatment ofrandom-flight polymer with excluded volume between

wo surfaces. For the enthalpic term a modified mean-eld, site-fraction approach, where the number of contacts
er lattice site is replaced by an interaction area per lat-ice site, is used. This modified approach allows one toxpress the interaction parameter as energy per area anday be approximated by interfacial or surface energies
1,4].

ig. 26. (a) Molecular structure and nomenclature of amine salts used to organiethyl, T: tallow (predominantly composed of chains with 18 carbons (∼65%)), HT:

f chains with 22 carbons (∼45%)), C*: coco (product made from coconut oil, consiesignate the substituents on the nitrogen. (b) Organoclays used to evaluate thorphology and properties [51]. Reproduced from Fornes, Yoon, Hunter, Keskkul

lymer Science 33 (2008) 1119–1198 1145

In general, the conclusions of the mean-field modeldeveloped, which are widely accepted by other researchers,may be summarized as follows. Since the spacing (or“gallery”) between the sheets is on the order of 1 nm, whichis smaller than the radius of gyration of typical polymers,there is obviously a large entropic barrier that inhibits thepolymer from penetrating this gap and intermixing withthe clay [9,11,147]. In this case, of course, unlike the solu-tion intercalation method, the decreased entropy due tothe confinement of the polymer is not compensated byan increase due to desorption of solvent molecules. How-ever, this entropy loss, associated with the confinementof a polymer melt is not prohibitive to nanocompos-ites formation, because an entropy gain, associated withlayer separation and greater conformational energy of thealiphatic chains of the alkylammonium cations, balancesthe entropy loss of polymeric intercalation, resulting ina net entropy change near zero. Thus, from the theoreti-cal model, the outcome of hybrid formation via polymermelt intercalation depends on energetic factors which maybe determined from the surface energies of the polymer
and OMLS. Thus, whether a mixture of polymer and OMLSproduces an exfoliated or intercalated nanocomposite or aconventional microcomposite depends critically upon thecharacteristics of the polymer and the OMLS, including thenature of the polymer as well as the type, packing den-
cally modify sodium montmorillonite by ion exchange. The symbols M:hydrogenated tallow, HE: 2-hydroxy-ethyl, R: rapeseed (consisting largelysting predominantly of chains with 12 carbons (∼48%)), and H: hydrogene effect of structural variations of the amine cations on nanocompositea and Paul by permission of Elsevier Science Ltd., UK.


Table 10Interlayer spacing of various modified montmorillonites and the resulting composites obtained with EVA (10.76 mol% vinyl acetate).

Code Cation Interlayer spacing

In modified clay In EVA composite

Mont-Na Na+ 12.2 12.6
Mont-2CN2C18 (CH3)2N+(C18H37)2
Mont-NC11COOH H3N+C11H22COOHMont-3CNC21COOH (CH3)3N+C21H42COOH

sity and size of the organic modifiers on the silicate surface[1,2,4,41,47,147,161].

It is worth mentioning that even when the surfactantchains are miscible with the polymer matrix, a completelayer separation depends on the establishment of veryfavorable polymer–surface interactions to overcome thepenalty of polymer confinement. If this is not the case,good dispersion of the particles may be achieved by thehelp of strong shear forces during the preparation andprocessing of the nanocomposite materials; the system,however, remains thermodynamically unstable. This can beobserved, e.g. in the case of PP or PE nanocomposite mate-rials prepared by melt mixing of the polymers with surfacemodified clay using high shear forces. If such a mixture isheated (e.g. during processing) to temperatures above themelting temperature, a (partial) re-agglomeration of theparticles takes place [2] immediately.

5.4.2.2. The effect of layered silicates and their organic mod-ification. In order to achieve clay exfoliation, the interlayerstructure of the OMLS should be optimized to maximizethe configurational freedom of the functionalizing chainsupon layer separation and to maximize potential interac-tion sites at the interlayer surface [47]. Therefore, as alreadymentioned in Section 3.2, the organic modification of theclay is a very important factor affecting the resulting struc-tures. In this respect, the type of surfactant, the chain lengthand the packing density may play an important role.

In a detailed study, Fornes et al. [51] used various aminecompounds to exchange the sodium ion of native mont-morillonite clay. The selection of amines shown in Fig. 26apermitted the authors to make six structural comparisons,as presented in Fig. 26b. Among different variables, threesurfactant structural issues were found to significantlyaffect nylon 6 nanocomposite morphology and properties:decreasing the number of long alkyl tails from two tallowsto one, use of methyl rather than hydroxy-ethyl groups, anduse of an equivalent amount of surfactant with the mont-morillonite, as opposed to adding an excess, led to greaterextents of silicate platelet exfoliation. However, the authorsemphasized that these effects may be specific to nylon 6matrices.

In fact, Hotta and Paul [162] prepared PE/clay nanocom-posites by melt compounding various combinations ofLLDPE-g-MA, LLDPE and two organoclays in a co-rotatingtwin screw extruder with a barrel temperature set at 200 ◦C
and a screw speed of 280 rpm. Nanocomposites based onthe organoclay having two alkyl chains are superior to thenanocomposites based on the organoclay having one alkylchain, in terms of clay dispersion. This was attributed tothe relatively better affinity of LLDPE for the alkyl chains
31.9 40.316.3 16.720.1 20.1

than for the silicate surface. Therefore, it is reasonable that,in this case, increasing the number of alkyl chains shouldlead to better dispersion of the organoclay. As deduced fromthese findings, the conditions favoring exfoliation may bequite different depending on the specific system investi-gated.

The effect of clay organo-modification on the mor-phology of EVA-based nanocomposites has also beeninvestigated. In a relevant study EVA was melt mixedwith two clays: Cloisite Na+ and Cloisite 30B (modifiedwith methyl-tallow-bis-2-hydroxyethyl ammonium ions).Mixing was performed in a Brabender Laboratory Mixerat 160 ◦C. Even though, quite surprisingly, XRD measure-ments revealed a decrease in clay interlayer spacing afterblending with EVA, for both montmorillonites, the authorsclaimed that nanocomposites were formed in the case ofCloisite 30B, on the basis of TEM observations. However,microstructures were obtained in the case of Cloisite Na+

[163].Zhang et al. [164] also synthesized EVA/clay nanocom-

posites through a melt blending method, using differ-ent EVAs and octadecyltrimethyl ammonium, dioctade-cyldimethyl ammonium, and tricetadecymethyl ammo-nium ion exchange MMTs. Again, EVA chains intercalatedinto the organo-modified MMT sheets, but in the case ofsodium montmorillonite (Na+-MMT) there was no suchintercalation.

Moreover, differences in the intercalation behavior havealso been observed between clays modified with differ-ent surfactants. For example, two different EVAs containing12 and 19% vinyl acetate, abbreviated as EVA-12 and EVA-19, respectively, were processed into their nanocompositeswith synthetic clay fluorohectorite (FH) organo-modifiedby octadecyl ammonium ion (ODA) and aminododecanoicacid. These materials were melt-blended at 120 ◦C ina twin screw microcompounder to obtain nanocompos-ites. Octadecyl ammonium ion intercalated FH (FH-ODA)formed a delaminated nanostructure, whereas ammoniumdodecanoic acid intercalated FH formed a microcompos-ite. This means that the octadecyl ammonium ion is morecompatible than ammoniumdodecanoic acid when interca-lated in FH during the synthesis of EVA/FH nanocomposites[165].

In another study by Alexandre et al. [166], severalexchanging cations bearing either simple alkyl chainsor aliphatic chains terminated by a carboxylic group
were studied for modifying montmorillonites as describedin Table 10. Nanocomposites were only formed whenEVA copolymers were melt mixed with unfunctional-ized organo-montmorillonites, such as montmorilloniteexchanged with dimethyldioctadecyl ammonium (Mont-


Table 11Characteristics of the various studied clays.

Filler Clay type Interlayer cations Ammonium contenta (wt.%) Interlayer distanceb (Å)

Cloisite Na Montmorillonite Na+ 0 12.1Cloisite 20A Montmorillonite (CH3)2N+ (hydrogenated tallow)2 29.2 22.1Cloisite 25A Montmorillonite (CH3)2N+ (hydrogenated tallow)(2-ethylhexyl) 26.9 20.7Cloisite 30B Montmorillonite (CH3)2N+ (tallow)(CH2CH2OH)2 20.3 18.5Nanofil 757 Montmorillonite Na+ 0 12.2Nanofil 15 Montmorillonite (CH3)2N+ (hydrogenated tallow)2 28.9 ∼29 (broad)Somasif ME100 Fluoromica Na+ 0 12.2Somasif MAE Fluoromica (CH3)2N+ (hydrogenated tallow)2 40.8 31.1

T o-unsa.

2adid

cpcowTdbpaomttw(asmNiocu

gtmt

TI

F

CCCCNNSS

allow: linear alkyl chains (C18 (65%), C16 (30%), C14 (5%) with ∼40% mona Determined by thermogravimetric analysis under helium atmosphereb Determined by X-ray diffraction on as-received clays.

CNC18), whereas with the same EVA matrix, the use ofmmonium cations functionalized with carboxylic groupsid not lead to the formation of an intercalated structure,

ndicating that functionalization of the clay interlayer isetrimental to the intercalation process.

In order to determine the effect of the nature of thelay and clay organic modifier on nanocomposite mor-hology and properties, Peeterbroeck et al. [167] used EVAopolymer containing 27 wt.% VA and various commercialrganoclays, presented in Table 11. EVA and clay (5 wt.%)ere compounded in a two-roll mill for 12 min at 140 ◦C.

he results of XRD analysis are reported in Table 12. Evi-ently, while all the tested organo-modified clays result inoth intercalated and exfoliated structures, the nanocom-osites based on Cloisite 30B display the highest exfoliationnd clay stacking destruction, characterized by the absencef a characteristic XRD peak. This better filler dispersionight arise from interactions between the acetate func-

ions of EVA and the hydroxyl-bearing ammonium cationshat modify Cloisite 30B. It is also worth noticing that,hatever the clay nature, dispersion of unmodified clays

Cloisite Na, Nanofil 757 or Somasif ME100) in EVA is char-cterized by the formation of microcomposites, since noignificant increase in the basal spacings recorded for theseaterials can be observed. When comparing Cloisite 20A,anofil 15 and Somasif MAE, characterized by clays of var-

ous origins but modified by the same ammonium cation,ne can remark that the final interlayer spacings are verylose, independent of the interlayer spacing of the clayssed or the amount of modifier.

Concerning the conformation of surfactant chains, it isenerally accepted that at low interlayer packing densi-ies of the organic modifier, the chains adopt a disordered

onolayer arrangement. As the packing density increases,he chains adopt more extended conformations (and thus

able 12nterlayer spacing variation as obtained from diffraction peaks measured by XRD

iller Filler interlayer distance (Å) Compo

loisite Na 12.1 12.2loisite 20A 22.4 38.7loisite 25A 20.7 36.8loisite 30B 18.5 –anofil 757 12.2 12.2anofil 15 Ca 29 (broad) 40.2omasif ME100 12.2 12.3omasif MAE 31.1 40.4

turated chains.

larger initial gallery heights), ultimately resulting in a solid-like paraffinic arrangement of the chains [47].

According to Vaia and Giannelis, the optimal structureappears to exhibit a chain arrangement slightly greater thana pseudo-bilayer. That is, there is an optimum interlayerstructure favoring hybrid formation that is intermedi-ate between a disordered monolayer and a solid-likeparaffinic arrangement of aliphatic chains. The differencebetween primary and quaternary ammonium head groupsdid not appear to be a predominant factor, at least for thepolystyrene intercalation, which they examined [47].

Ginzburg et al. calculated phase diagrams and showedthat as the length of the grafted chains and/or their densityis increased, the miscibility between the clay sheets and thepolymer is improved and the resulting mixture can exhibitexfoliated structure for a range of clay volume fractions.According to their work, for short surfactant molecules, thepolymer is unable to penetrate the gallery between the claysurfaces, and the equilibrium morphology becomes immis-cible (two-phase) for most values of the Flory-Hugginsparameter and the clay volume fraction. Only in the limit oflarge negative � (strong attraction between grafted chainsand polymer melt) can such a composite become exfoliated[9].

Next to these studies, Balazs and co-workers proposed atheoretical model that uses numerical self-consistent fieldcalculations to study the effect on morphological behav-ior of varying the surfactant–matrix enthalpic interaction,surfactant coverage, and surfactant length. The modelindicated that longer surfactants (clay organic modifiers)
promote intercalation of matrix molecules in the modi-fied clay galleries by providing a reduction in the entropicpenalty to the intercalating polymer. However, very highdegrees of surfactant coverage on the silicate surface makeintercalation and exfoliation unfavorable [168].
on clays and the resulting melt-blended EVA composites.

site interlayer distance(Å) Interlayer distance variation (Å)

0.116.316.1–0Ca 110.19.3


me onpermiss

Fig. 27. Illustration, for a bimodal surfactant brush, of one possible outcosurfactant [169]. Reproduced from Kurian, Dasgupta, Beyer and Galvin by

This last conclusion is consistent with that from theexperimental work of Kurian et al. These authors ini-tially prepared organically modified clays by exchangingmost cation exchange positions in Na+-MMT with custom-made quaternary ammonium ion terminated PS surfactantsof five different molecular weights. They showed thathigh levels of modification resulted in dense polymerbrushes on the clay surfaces, preventing intercalation ofPS homopolymer molecules. Phase-separated morpholo-gies were observed in all cases, regardless of surfactantmolecular weight. This was explained within the frame-work of well-established theories of dewetting from densepolymer brushes [169].

Thus, in a later study the authors described a newscheme termed mixed coverage, wherein the silicate is mod-ified with a mixture of PS-based surfactants of differentlengths, to create a silicate surface grafted with a bimodalbrush. Texturing of the silicate surface was expected toallow intercalation of matrix molecules in the galleriesby providing a favorable entropic gain to the intercalat-ing polymer through interaction with the longer graftedmolecules, and reduced enthalpic interaction with the claysurface. However, XRD data indicated that the samplesremained in a phase-separated morphology, with no indi-cation of intercalation or exfoliation of the layered silicatein the PS matrix. To explain this result, the authors sug-gested that, despite the bimodal nature of the brushes inthis experiment, it is still possible that the brush is stilleffectively dense (only 17% of the exchangeable cationswere replaced by the longer surfactant and the remaining83% by the short surfactant), making the long surfactantfraction ineffective in fostering wetting and intercalation.Fig. 27 shows a schematic of this scenario. In Fig. 27a, along surfactant is illustrated at low coverage over a surfacethat is repulsive to the polymer matrix (PS homopolymer).In this case, the surfactant will stretch away from the sur-face until the energy required to stretch is no longer lessthan that of the repulsive enthalpic interactions. This is thescenario for typical surfactant-modified clay at very lowcoverage. In the bimodal brush, the shorter surfactant atfull coverage (83%) effectively replaces the silicate surface.As illustrated in Fig. 27b, the surface now can be thought
of as being comprised of the dense brush. When a smallfraction of those surfactants are replaced with longer sur-factants, the situation may become something like thatillustrated in Fig. 27c. The length of the longer surfactantis still important, but the conformation of the longer sur-
longer surfactant conformation of the densely packed brush of a shorterion of John Wiley & Sons Inc.

factant must also be considered. If the mixed surfactantsused to create the textured surface do not differ much intheir lengths, one has effectively a short surfactant at highcoverage. As the molecular weight difference between thetwo surfactants increases, the surfactant present in lesseramount becomes effectively longer, but still it is possiblethat it does not actually assist in the wetting and intercala-tion processes, because there is now no enthalpic repulsioncausing the surfactant to stretch away from the silicate sur-face. This would result in the scenario illustrated in Fig. 27c,where the longer surfactant now remains close to the brushof chemically similar molecules and does not aid interca-lation by interacting with matrix polymer molecules [169].This work clearly demonstrates that when tailoring the sil-icate characteristics that favor intercalation or exfoliation,one has to simultaneously consider different factors, e.g.the surfactant chain length and packing density.

Summarizing the above, the simple organic modi-fication of the clay through ion-exchange reactions isnot always enough to achieve nanocomposite formation.This is because an ideal compatibilization agent betweentwo intrinsically incompatible components should have(combined in one molecule) parts which mix thermody-namically stable and easily with both components [2]. Infact, Balazs et al. likened the situation found in nanocom-posites to the behavior of fiber-reinforced composites. Theclay substrate represents the surface of a fiber and the “sur-factants” represent a coating, which is applied to enhancethe adhesion between the fiber and polymer matrix [11].Unfortunately, surfactants fulfill only partly this require-ment. That is, the ionic part interacts certainly in a favorableway with the charged surface of the sheet-like clay par-ticles; however, the long alkyl tail displays only a limitedcompatibility with the polymer chains. Second, as statedabove, polymer–surface interactions are also important fora complete and stable dispersion. Therefore, better compat-ibilization should be expected from “macro-surfactants”like block or graft copolymers combining blocks which caninteract with the solid particle surface and with the matrixpolymer – thus meeting most of the requirements listedabove (Fig. 28) [2]. For example, the clay compatible blockcould be PEO. The PEO block acts as a complexing group for
the sodium ions located between the sheets of the inorganicclay. Therefore, an ion-exchange reaction may not even benecessary. Alternatively, the block copolymer may also beused as a “co-surfactant” with the surfactant on the sur-face of organo-clays, thus fine-tuning the thermodynamic


Ftb

cmataobqro

cmoaoci[

5pi

hshtcfPdaimst

potr

the preparation of EVA-based nanocomposites. For exam-

ig. 28. Schematic picture of the action of block copolymer for an exfolia-ion of clay sheets within a polymeric matrix [2]. Reproduced from Fischery permission of Elsevier Science Ltd., UK.

haracteristics of the clay surface with respect to the poly-er matrix material. Toyota, for example, developed suchprocess for the incorporation of MMT sheets into PP. In

his procedure, a first ion-exchange step with double tailedmmonium cations is followed by a further incorporationf end-group modified (grafted) oligomeric polypropyleneetween the organically modified sheets and a subse-uent mixing with the matrix polymer. This processelies on the compatibilizing action of the functionalizedligomers [2].

Finally, it is important to note that, apart from the clayhemistry, the amount of clay incorporated in the poly-er matrix also plays a determining role. In fact, it is

ften reported that while low clay loadings favor exfoli-tion, higher amounts of clay (usually above 10 wt.%) allownly intercalation of polymer chains in the layered sili-ate galleries to occur. Further dispersion of clay plateletss hindered, most probably due to percolation phenomena2,170].

.4.2.3. The effect of the polymer matrix. Polymer matrixarameters may also determine the outcome of melt blend-

ng a polymer and a layered silicate.In this context, the effect of the matrix molecular weight

as been considered. Early work by Vaia and Giannelishowed that, for statically annealed PS samples, the finalybrid structure is independent of the molecular weight ofhe polymer. Only the time needed for intercalation to pro-eed was different, going from 6 h for Mw of 30,000 to 24 hor 90,000 and 48 h for 400,000 at 160 ◦C. Clearly, high Mw

S decreases the kinetics of intercalation by decreasing theiffusivity of the polymer in the interlayer. However, theuthors noted that additional experimental work examin-ng the intercalation behavior of a broader range of polymer

olecular weights coupled with dynamic blending of con-tituents to enhance equilibrium mixing would be requiredo further explore this issue [47].

In another study, Fornes et al. [14] prepared nanocom-
osites based on three different molecular weight gradesf nylon 6 (low, medium and high) using a co-rotatingwin screw extruder. WAXD and TEM results collectivelyeveal a mixed structure for the LMW based nanocom-
lymer Science 33 (2008) 1119–1198 1149

posites, having regions of intercalated and exfoliated clayplatelets. Qualitative TEM observations were supported bya quantitative analysis of high magnification TEM images.The average number of platelets per stacks was shownto decrease with increasing molecular weight, therebyrevealing larger extents of clay platelet exfoliation forthe nanocomposites in the order HMW > MMW > LMWcomposites. As a result, tensile tests revealed superiorperformance for the higher molecular weight nylon 6composites, particularly those based on HMW, as will bediscussed in Section 6.1.2.

However, probably the most critical condition for theformation of intercalated and especially exfoliated hybridsvia polymer melt intercalation, is the presence of polartype interactions (i.e. other than Van der Waals forces).Therefore, polar polymers containing groups capable ofassociative type interactions, such as Lewis acid-base inter-actions or hydrogen bonding, favor the intercalation ofmacromolecular chains into the silicate galleries [47,49],while in the case of apolar polymer matrices, clay delam-ination typically requires the use of compatibilizers, asdiscussed in the following paragraph.

A good example demonstrating the importance of polarinteractions is the formation of EVA-based nanocompos-ites via melt intercalation. It is well established that thepresence of polar groups (ester groups of the vinyl acetatemoieties) all along the chains improves the ability of thesepolymers to intercalate in organo-modified montmoril-lonites [1]. Therefore, several studies have focused on theeffect of the vinyl acetate content on the dispersion ofclay nanoplatelets. In general, it has been observed thatthe higher the vinyl acetate content the larger is the basalspacing increase of the clay, inducing the formation of inter-calated to exfoliated nanostructures.

In a representative study, Chaudhary et al. [171]prepared EVA-based nanocomposites via melt intercala-tion using an intermeshing counter-rotating twin screwextruder. They used EVA copolymers with 9, 18 and 28% VA(vinyl acetate) and two organo-modified clays: Cloisite 15A(C15A), which is more suitable for the less polar EVA9 dueto long aliphatic chains in C15A, and Cloisite 30B (C30B),suitable for the more polar polymers, like EVA18 or EVA28.They prepared composites with filler level varying from 2.5to 7.5 wt.%, and subsequently characterized the structuresobtained by WAXD and TEM. The results indicated that onlyintercalation occurred in the case of the less polar EVA9,while the clay was exfoliated in the more polar EVA18 andEVA28. Therefore, the authors concluded that an increase inthe content of polar VA groups in EVA 18 and EVA28 as com-pared to EVA9, which lowered the thermodynamic energybarrier for clay–polymer interaction, possibly allowed a rel-atively higher number of polymer chains to migrate andstabilize within the clay platelet and form partially exfoli-ated and/or disordered intercalated states.

It is worth mentioning that despite the presence of polargroups in EVAs, polymer compatibilizers have been used in

ple, Li and Ha [172] selected a maleic anhydride grafted EVAcontaining 18 mol% VA to process its nanocomposites withorgano-modified Cloisite through melt blending at 175 ◦Cand found that the dispersion of Cloisite in the maleic anhy-

ess in Po
dride grafted EVA was much better than in the simple EVAmatrix.

5.4.2.4. The effect of melt intercalation processing conditions.Melt processing conditions play a key role in achievinghigh levels of exfoliation. Indeed, nanocomposites havebeen formed using a variety of shear devices (e.g. extrud-ers, mixers, ultrasonicators, etc.), among which twin screwextruders have proven to be most effective for the exfolia-tion and dispersion of silicate layers [14].

The screws in twin screw extruders intermesh so thatthe relative motion of the flight of one screw inside thechannel of the other acts as a paddle that pushes the mate-rial from screw to screw and from flight to flight. Twodifferent patterns for intermeshing twin-screw extrudersare possible. In the co-rotating pattern the screws rotatein the same direction and the material is passed from onescrew to another and follows a path over and under thescrews. This gives high contact with the extruder barrel,which improves the efficiency of heating. The path alsoensures that most of the resin will be subjected to the sameamount of shear as it passes between the screws and thebarrel. The self-wiping nature of the co-rotating screws ismuch more complete than in the counter-rotating system,thus in the co-rotating case there is less likelihood thatmaterial will become stagnant. In the counter-rotating pat-tern, on the other hand, the screws rotate counter to eachother and the material is brought to the junction of the twoscrews, building up in what is called the material bank onthe top of the junction. This buildup of material is con-veyed along the length of the screw by the screw flights.As the material passes between the screws, high shear iscreated, but shear elsewhere is very low. Since only a smallamount of material passes between the screws, total shearis lower than in single-screw extruders and in co-rotatingtwin-screw extruders. Therefore, co-rotating systems aremore effective than either counter-rotating or single-screwextruders [173]. On the other hand, although it is oftenstated that twin screw extruders favor intercalation whencompared to single screw systems, this may not be the casefor counter-rotating extrusion systems, for the aforemen-tioned reason.

Focusing on the effect of the extrusion system on thedegree of intercalation, Cho and Paul [15] prepared nylon6/o-MMT nanocomposites using either an intermeshing co-rotating twin screw extruder or a single screw extruder.They found that for the composite prepared by singlescrew extrusion, full exfoliation is not achieved, which wasattributed to insufficient amount of shear and short resi-dence time. On the other hand, by mixing in the twin screwextruder, the organoclay is uniformly dispersed into nylon6 and the individual layers are aligned along the flow axis.

However, other researchers have reported on nanocom-posite preparation using single-screw extruders. For exam-ple, McNally et al. [36] successfully prepared PA12/claynanocomposites using conventional single-screw melt
blending.
Moreover, it is important to notice that, even for a givenextruder, processing conditions may determine the out-come. More specifically, increasing the mean residencetime in the extruder generally improves the delamination

lymer Science 33 (2008) 1119–1198

and dispersion. However, there appears to be an optimumextent of back mixing and an optimum shear intensity;excessive shear intensity or backmixing apparently causespoorer delamination and dispersion [52]. Often, specialscrew designs, including provisions for additional mixing,or static mixers at the end of the screw are used to enhancemixing and thus the silicate dispersion [173].

Li et al. [174] developed an ultrasonic extrusion technol-ogy, which organically combines extruder and ultrasoundpower. The authors claimed that the introduction of ultra-sonic irradiation in extrusion processing of polymer canimprove the processibility of polymer materials, and alsoreduce the size and size distribution of dispersed parti-cles in polymer blends. They used the ultrasonic oscillationextrusion system developed to prepare polymer/MMTnanocomposites. The system, consisting of an extruder anda cylinder die connected to a generator of ultrasonic oscil-lations in the direction parallel to the flow of the polymermelt, was found to improve the exfoliation of the clay,though only for specific matrices.

It should be noted at this point that, apart of the vari-ous extrusion systems, internal mixers (i.e. batch mixingdevices where mixing occurs in a closed chamber) mayalso be successfully used for the preparation of exfoliatednanocomposites, as demonstrated, for example, in the caseof PEI matrix [46,52]. However, these devices appear to bemuch less popular in nanocomposite preparation.

Another factor affecting the resulting structures in thecase of crystalline polymer matrices, is the crystalliza-tion temperature. For example, Okamoto and co-workers[175,176] observed through X-ray analyses that the inter-gallery spacing of PP-MA based nanocomposites increaseswith the crystallization temperature Tc for any amount ofclay content in the nanocomposites. The microstructureof the nanocomposites, observed directly by TEM, showedthat the clay particles are well dispersed at low Tc andthat segregation of silicate layers occurs at high Tc. Thisimplies that, by controlling intercalation through crystal-lization at a suitable temperature, one can control the finestructure, the morphology—and thus the properties of crys-talline polymer/clay nanocomposites.

Conclusively, a number of factors affect the outcome ofmelt intercalation. In this context, Dennis et al. [52] pre-sented a simplified scheme to underline the conditionsunder which clay exfoliation into a polymer occurs dur-ing melt blending. The proposed scheme (Fig. 29a) is basedon the relationship between the compatibility of the chem-istry of the clay treatment and the matrix and the processconditions used to make a nanocomposite. It distinguishesthree cases. The first case is chemistry-dependent. Whenthe clay chemical treatment and the resin are compati-ble, almost any set of processing conditions can be used toform an exfoliated nanocomposite. In case 2, clay chemi-cal treatment and polymer are marginally compatible. Inthis situation, the process conditions can be optimizedto give an exfoliated nanocomposite. That is, the organ-
oclay chemical treatment and the matrix are compatibleenough that processing conditions can be tailored to opti-mize delamination and dispersion. Finally, in case 3, thereis no apparent compatibility of the clay chemical treatmentand the polymer. Processing conditions can be optimized to


Fig. 29. Proposed mechanism of how the organoclay particles disperseinto polymers during melt processing. Part (a) shows three cases involvingtdtW

ge

ti1pstpdpottts

5

ccpPb

he interplay between chemistry and process conditions in the mixingevice. Part (b) illustrates schematically how platelets peel apart underhe action of shear [52]. Reproduced from Dennis, Hunter, Chang, Kim,

hite, Cho and Paul DR by permission of Elsevier Science Ltd., UK.

ive intercalants or tactoids that are minimized in size, butven partial exfoliation does not occur.

These authors also presented possible clay delamina-ion paths (Fig. 29b) to demonstrate that increasing shearntensity is not enough to achieve exfoliation. In pathway, stacks of platelets are decreased in height by slidinglatelets apart from each other, a pathway that requireshear intensity. Pathway 2 shows polymer chains enteringhe clay galleries pushing the ends of platelets apart. Thisathway does not require high shear intensity, but involvesiffusion of polymer into the clay galleries (driven by eitherhysical or chemical affinity of the polymer for the organ-clay surface) and is, thus, facilitated by residence time inhe mixing device. As more polymer enters and goes fur-her in between clay platelets, especially near the edge ofhe clay galleries, the platelets appear to peel from the edge,ince they are able to bend [14,52].

.4.3. Compatibility issues in non-polar polymersIn contrast to polar polymers, like polyamides, that

an effectively exfoliate organically modified clays usingonventional melt processing techniques, for non-polarolymers, such as the most widely used polyolefins, PE orP, synthesis of well exfoliated nanocomposites appears toe more difficult, because these polymers are so hydropho-

lymer Science 33 (2008) 1119–1198 1151

bic and lack suitable interactions with the clay surface, evenafter it has been organically modified [49,162,177–179].However, the development of polyolefin/clay nanocompos-ites is a field of rapidly growing industrial relevance due totheir promise of improved performance in packaging andengineering applications [111]. Therefore, ways to resolvethe difference in polarity between polyolefins and clays,in order to prepare nanocomposites by conventional meltcompounding, have been proposed.

More specifically, initial attempts to create non-polarpolymer/clay nanocomposites by simple melt mixing werebased on the introduction of a modified oligomer to medi-ate the polarity between the clay surface and the polymer[2,49,177]. The most promising strategy at the present timeis to add a small amount of a maleic anhydride grafted poly-olefin that is miscible with the base polyolefin. It is believedthat the polar character of the anhydride has an affinityfor the clay materials, such that the maleated polyolefincan serve as a “compatibilizer” between the matrix and thefiller [162,180,181].

In fact, Zhai et al. [179] showed that two kinds ofhybrids are formed by melt mixing: an o-MMT with PEand with PE-g-MA, respectively. For the PE/o-MMT sys-tem the intercalate effect is limited and the dispersion ofclay is unsatisfactory. However, for the PE-g-MA/o-MMTnanocomposites, MMT was exfoliated in the matrix, as tes-tified by both XRD and TEM.

Wang et al. [49] prepared several types of nanocompos-ites with different compositions of the organically modifiedclay and maleated polyethylene by melt compounding at140 ◦C using a Brabender mixer operating at 60 rpm for20 min. They found that the alkylammonium chain lengthmay change the degree of interaction between clay andpolyethylene and that the original basal reflection peak ofthe clay disappeared completely above a certain graftinglevel of MA, about 0.1 wt.%.

Quite interestingly, Zhang and Wilkie [182] obtained PE-organoclay nanocomposites by adding maleic anhydridedirectly as a compatibilizer during the melt blending. As theauthors suggested, the maleic anhydride probably reactedwith PE during the high temperature blending in air, lead-ing up to the formation of a graft copolymer in which maleicanhydride units are attached to the PE chains.

In another study, Tang et al. [183] described the prepa-ration of PP-based nanocomposites through a successfulcombination of clay modification and intercalation in onestep. The authors mixed pristine MMT with the surfac-tant (C16) and PP with or without PP-MA using a highspeed mixer. The mixed powder was then processed in atwin screw extruder and nanocomposites were obtained.The results of this study showed that the structure of PP-clay nanocomposites is sensitive to the compatibilizer andthe surfactant, since their increasing concentrations willreduce the free energy of the system, which is favorablefor thermodynamic stability. The dispersion mechanismproposed is the following. At first, some surfactant chains
diffuse into the interlayer under physical absorption andshear, rendering the clay organophilic. However, some sur-factant remains in the polymer matrix, which may enhancethe compatibility when the matrix is intercalated into theinterlayer. In fact, the authors suggested that there is some


Table 13Structures of the polar polymers used as composite matrices.

Polymer Structure

LDPE

EMA (y = 0.215)

EVA −18, −28 (x = 0.18, 0.28)

EMAAA (y = 0.18, z = 0.06)

PE-g-MAH (y ≈ 0.02)

interaction between the polymer matrix and the surfactant,just as in the interaction between the surfactant and thesilicates. On the other hand, some PP-MA may be interca-lated into the interlayer of MMT, after the surfactant makesthe clay sufficiently organophilic. At the same time, PP-MAmay act as a high molecular weight surfactant. The inter-layer spacing of the clay increases and, if the miscibility ofPP-MA with PP is good enough to allow dispersion at themolecular level, the exfoliation of intercalated MMT shouldtake place.

In addition to maleic anhydride and maleic anhydridegrafted PE, EVA has also been used as a compatibilizer toprepare PE-based nanocomposites. For example, Zanettiand Costa [177] prepared several types of composites withdifferent PE/EVA ratios and 5 wt.% organoclay by meltcompounding at 150 ◦C using a Brabender internal mixerwith a screw speed of 60 rpm for 10 min. The polymerEVA contained 19 wt.% VA. No interaction was obtained bycompounding the PE with the clay in absence of a com-patibilizer. However, 1 wt.% EVA was enough to intercalateall the organoclay. Further increasing the amount of EVAabove 10 wt.% caused a decrease in the degree of coherentlayer stacking (i.e. a more disordered system).

Zhao et al. [184] investigated chlorosilane-modifiedmontmorillonites and their results showed that inter-calated PE nanocomposites were obtained by meltintercalation using common alkylammonium intercalatedclay, which was pretreated with chlorosilane. In a laterwork, the authors used directly a reactive intercalatingagent (N-�-trimethoxyl-silanepropyl) octadecyldimethy-
lammonium chloride (abbreviated JSAc) to modify themontmorillonite clay, so that the chemical reaction withhydroxyl groups at the edge of the clay layers and the inter-layer ion exchange were carried out in one step. PE/claynanocomposites were then directly prepared by melt inter-
calating PE and the above mentioned clay in a twin screwextruder at 180 ◦C and 200 rpm, whereas only microcom-posites were formed when using common intercalatingagent.

It is also worth mentioning the work of Preston et al.[178], who prepared nanocomposites using the followingmatrices: two EVAs with different vinyl acetate contents,poly(ethylene-co-methyl acrylate) (EMA), poly(ethylene-co-methyl acrylate-co-acrylic acid) (EMAAA), and a blendof LDPE with maleated ethylene copolymer (PE-g-MA).Structures for each of these materials are given in Table 13.The organoclay they used was organically modified ben-tonite clay. Composites were prepared by melt mixing ina twin screw extruder operating at 380 rpm with a screwconfigured for intensive mixing. Through XRD measure-ments, the authors concluded that no interaction was likelybetween the LDPE and the silicate, whereas intercalation ofthe organoclay occurred in the presence of the four polarpolymers.

As in the case of PE, it is difficult to get exfoliatedand homogenous dispersion of the silicate layer at thenanometer level in polypropylene, due to its low polar-ity. Consequently, PP also is usually modified with polaroligomers prior to introduction of modified clay, in orderto achieve nanometric dispersion of the clay [161].

One typical example is the PP/clay nanocompositesystem described by Toyota. The Toyota research groupprepared PP/clay nanocomposites by direct melt com-pounding of PP with organo-modified MMT, in the presenceof a maleic anhydride grafted PP (PP-g-MA). They added
three times as much PP-g-MA as the clay by weight toprepare well mixed PP/clay nanocomposites, and pointedout that the miscibility between maleated oligomer andmatrix polymer played a key role in composite properties[2,49,177,185]. In fact, it has been suggested that the relative


F trix [187U

cictsrP

oPasnob

bmiwHccbnc

P(oowdc

am

ig. 30. Schematic illustration of OMLS dispersion process in PP-g-MA masuki by permission of Wiley-VCH, Germany.

ontent in maleic anhydride cannot exceed a given value,n order to retain some miscibility between PP-MA and PPhains. When too many carboxyl groups were spread alonghe polyolefin chains, no further increase in the interlayerpacing was obtained in PP/PP-g-MA/clay blends, leadingather to the dispersion of PP-g-MA intercalated clay in theP matrix [1].

Similarly, Hasegawa et al. [186] reported the preparationf exfoliated PP-based nanocomposite by melt blendingP-g-MA and organically modified MMT at 200 ◦C, usingtwin screw extruder. Fig. 30 shows a schematic repre-

entation of the clay dispersion process in PP-MA-basedanocomposites. According to the authors, the driving forcef exfoliation originates from the strong hydrogen bondingetween the MA groups and the polar clay surface.

Kato et al. [188] prepared PP-based nanocompositesy the melt intercalation of PP chains modified by eitheraleic anhydride (PP-g-MA) or hydroxyl groups (PP-OH)

n o-MMT. For both matrices, intercalated nanocompositesere recovered after melt blending at 200 ◦C for 15 min.owever, a PP-g-MA matrix with a lower maleic anhydrideontent did not intercalate under the same conditions, indi-ating that a minimal functionalization of PP chains has toe reached for intercalation to proceed. The authors alsooticed that intercalation increased with the polymer-to-lay ratio, i.e. when the PP-g-MA fraction was increased.

Using the same method, Okamoto et al. [189] preparedP/MMT nanocomposites. The authors mixed PP-g-MA0.2 wt.% MA) and different amounts (2, 4 and 7.5 wt.%)f C18-MMT in a twin screw extruder at 200 ◦C andbtained nearly exfoliated structures when 2 wt.% clayas added. However, addition of 4 and 7.5 wt.% clay led to
isordered intercalated nanocomposites and ordered inter-alated structures, respectively.
Lopez et al. [161] used two different polar couplinggents, diethyl maleate grafted PP (PP-g-DEM) and com-ercial maleic anhydride grafted PP (PP-g-MA) to prepare

]. Reproduced from Hasegawa, Okamoto, Kawasumi, Kato, Tsukigase and

PP-based nanocomposites. DEM was chosen as the compat-ibilizing agent, because of its high thermal stability, highboiling point, and good compatibilization with polyolefins,compared to other compatibilizing agents. Furthermore,the low homopolymerization behavior of DEM allowsbetter control of the functionalization reaction. Maleicanhydride was used as reference, since it has been widelyused as compatibilizer for this kind of system. The PP/clayhybrids were prepared by melt compounding with twodifferent clays, commercial modified montmorillonite andsodium bentonite modified with octadecylammoniumions. The results showed that although the commercial clayoutperforms the octadecylammonium treated bentonite,differences in mechanical properties when using differentclays are smaller if DEM is used instead of MAH. This isa consequence of the very low degree of compatibiliza-tion between the polymer matrix and the clay. In fact, thisstudy proves that clay dispersion and interfacial adhesionare greatly affected by the kind of matrix modification. DEMhas a lower polarity compared to MAH, providing a lesseffective interaction with the polar components of the clay.The authors therefore, concluded that clay and matrix mod-ification are synergistic factors which need to be properlymodulated in order to obtain the desired final properties inthis kind of non-polar polymer-based nanocomposite.

Finally, as in the case of PE and PP matrices, compat-ibilization is a critical issue also for other polymers, suchas PS. Therefore, Wang and Wilkie [190] prepared PS/claynanocomposites by in situ reactive blending with boththe organically modified clays and the pristine inorganicclay in the presence of maleic anhydride, and found thatmaleic anhydride increases the possibility of nanocompos-
ite formation. Also, Hasegawa et al. [191] produced partiallyexfoliated PS/clay nanocomposites by compounding in atwin-screw extruder organically modified MMT with anblend of PS and ≥50% of another compatibilizer, namelypoly(styrene-co-methyl vinyl oxazoline).


Table 14Molecular weight of host polymers.

SPU HPU

Mn PDI Mn PDI

As-received PU 216.000 1.8 85.000 2.0Solvent cast 121.000 1.7 82.000 2.1Melt compounded 66.000 2.0 59.000 1.9

5.4.4. Degradation problems encountered during meltintercalation

Despite the aforementioned advantages of polymermelt intercalation, this technique may involve polymerdegradation problems since, when preparing clay/polymernanocomposites using melt blending, a certain temper-ature is needed in the processing. Also, apart from thepolymer matrix degradation, if the processing tempera-ture to make the PLS is beyond the thermal stability ofthe organic treatment on the OMLS, some decompositionwill take place. The onset temperature of decompositionof the organic modifier is, therefore, important in the pro-cess to make a polymer/clay nanocomposite, since polymerprocessing is normally done above 150 ◦C. Moreover, inaddition to common detrimental aspects of degradation,the resulting products may play a major but yet to be deter-mined role in the formation of exfoliated nanostructures[43].

Therefore, the degradation issues encountered duringmelt intercalation have been addressed in several studies.For example, Finnigan et al. [88] prepared TPU nanocom-posites by both twin screw extrusion and solvent casting,in order to compare the outcomes of these methods. Theauthors employed two TPUs: a soft elastomer (SPU) and ahard elastomer (HPU) consisting of the same soft and hardsegments, but in different relative amounts. As demon-strated by WAXD analysis, both processing routes led todelaminated structures, which illustrates that if there is agood driving force for intercalation between the polymerand organosilicate, the need for an optimized processingroute is diminished. Although melt compounding offeredslightly better silicate dispersion than solvent casting,the authors suggested that solvent casting must be thepreferred processing route for these materials, owing toelimination of PU and surfactant degradation. In fact, asshown in Table 14, it was found that the number averagemolecular weight (Mn) of PU significantly decreases duringmelt compounding and, to a smaller extent, during solventcasting (due to an ultrasonic probe that was applied). In thisparticular case, the authors identified as additional causesof thermal degradation the small size of the extruder (andthus the large surface to volume ratio) as well as the absenceof additives to reduce degradation.

Xie et al. [43] focused on the effect of organic modi-fiers on the thermal decomposition of OMLSs and foundthat, while different long alkyl substituents have no effector very little effect, the thermal degradation of organi-
cally modified montmorillonite is quite different comparedto that of pure montmorillonite. The DTGA thermal curveshown in Fig. 31 for the OMLS was considered in fourparts: (a) the free water region below 200 ◦C; (b) theregion where organic substances evolve in the tempera-
Fig. 31. Comparison of DTGA curves of various OMLS [43]. Reproducedfrom Xie, Gao, Liu, Pan, Vaia, Hunter and Singh by permission of ElsevierScience Ltd., UK.

ture range 200–500 ◦C; (c) the structural water region inthe temperature range 500–800 ◦C; (d) a region between800 and 1000 ◦C where organic carbon reacts in some yetunknown way. In OMLS sample the free water disappearsby 40 ◦C. There is no interlayer water in OMLS as the quater-nary ammonium salt has been exchanged for the hydratedsodium cation. The most distinguishing difference betweenthe sodium montmorillonite and the organically modifiedmontmorillonite is in the temperature range 200–500 ◦C,as the organic constituent in the organo-clay starts todecompose somewhat above 200 ◦C. Another distinguish-ing difference between sodium montmorillonite and theorganically modified montmorillonite is in the tempera-ture range from 800 to 1000 ◦C. Sodium montmorilloniteis very stable when the temperature is higher than 800 ◦C,however, the OMLS continues to lose weight and a largeramount of CO2 is released at temperatures over 800 ◦C.

Davis et al. [192] found that during melt blendingMMT/PA6 nanocomposites in a twin-screw extruder at240 ◦C, a particular quaternary alkyl ammonium treat-ment of the montmorillonite clay degraded it to an extentcorrelated with extruder residence time. To address thisissue, they conducted an investigation on the processingdegradation of PA6/MMT nanocomposites and clay organicmodifier. The results led them to the following conclusions:

1. PA6 nanocomposites significantly degraded during pro-cessing at 300 ◦C. Within experimental uncertainty,drying at 120 ◦C rather than 80 ◦C prior to processing hadlittle effect on the degree of degradation. Virgin PA6 didnot degrade under identical processing conditions.

2. Thermal decomposition of PA6 nanocomposite may haveresulted from hydrolytic peptide scission. The catalyticactivity of MMT was not investigated in this particu-lar study; however, on the basis of previous research, itappeared that MMT could be involved in PA6 thermaldegradation.

3. Heating at 120 ◦C for 4 h thoroughly dried virgin PA6; butdrying at 80 ◦C resulted in no water loss. The amount ofvolatile water in PA6 nanocomposites was greater thanwas observed in virgin PA-6. Longer drying times and


ced from

4

5

mimtlmtoptt

TP

N

PPPPPPPPPEP

Fig. 32. Synthesis of the thermally stable organoclay [193]. Reprodu

higher temperatures resulted in drier PA6 nanocompos-ites.

. MMT and water are responsible for the degradation ofPA6 nanocomposites.

. When PA6 was processed at 300 ◦C some water waspresent, however, little degradation was observed. Thismeans that: (a) water itself may not be sufficient to causedegradation, (b) water escapes from PA-6 faster thanfrom the nanocomposite, (c) clay and water are a spe-cial catalyst combination and/or (d) clay is a source ofhigh-temperature reactive water or hydroxyls.

In addition to the aforementioned studies, which focusainly on the degradation mechanism, others are explor-

ng ways to overcome or limit degradation during polymerelt intercalation. In general, when a material is subjected

o extrusion, degradation is detected as discoloration andowered physical or mechanical properties. A strong odor

ay also indicate degradation. If the degradation is general,
hat is, the entire extrudate is affected, as shown by discol-ration throughout, although darker streaks may also beresent, the most likely cause is that the heat is too high forhe speed of extrusion. The obvious solutions are to reducehe heat or to increase the extrusion speed. Some combina-
able 15LS nanocomposites prepared by melt intercalation.

anocomposite

A6/[(HE)2M1R1] modified MMTA6/Na+-MMT water slurryS/alkylammonium modified MMTEI/hexadecylamine modified MMTEO/Li+ or Na+-MMTLA/C18-MMTP/stearylammonium modified clayP-MA/C18-MMTP/o-MMT modified using an organic swelling agent (Tb = 100–200 ◦C)VA/dimethyl-dioctadecyl ammonium modified MMTET/1,2-dimethyl-3-N-alkyl imidazolium salt modified MMT

Chang, Kim, Joo and Im by permission of Elsevier Science Ltd., UK.

tion of these two variables are likely feasible since the speedof the extruder affects mechanical heating of the mate-rial [173]. However, this needs to be done carefully since,as mentioned above, processing conditions may affect themorphology of the resulting material.

On the other hand, a number of researchers havedeveloped and applied clays exhibiting high thermalstability. In this context, Chang et al. [193] developed athermally stable montmorillonite through an ion exchangereaction between Na+-MMT and dodecyl triphenyl phos-phonium chloride (C12PPh-Cl−) (Fig. 32). Gilman et al.[194] described the preparation of PA6-based nanocom-posites of MMT modified with trialkylimidazoliumcations to obtain high stability OMLS at high processingtemperatures.

A surprising result reported in another study was thatpoly(3-hydroxybutyrate) (PHB) nanocomposites preparedvia melt intercalation showed severe degradation as testi-fied by GPC, when an organically modified MMT was used,
whereas no degradation was found with nanocompositesbased on organically modified fluoromica. Even thoughthere is no explanation on how organically modified flu-oromica acted to protect the system, the authors suggestthat the presence of Al Lewis acid sites, which catalyze the
Mixing device and conditions Ref.

Co-rotating twin screw extruder, 240 ◦C, 280 rpm [14]Extrusion and drying [160]Statically heating at 165 ◦C, vacuum, 25 h [29]Internal mixer, 370 ◦C, 30 min [46]Statically annealing, 80 ◦C, 6 h [196]Twin screw extruder, 190 ◦C [153]Twin screw extruder, PP-MA compatibilizer [197]Twin screw extruder, 200 ◦C [186]Twin screw extruder, 250 ◦C [166]130 ◦C [166]Co-rotating mini twin screw extruder, 285 ◦C [148]


Table 16Predicted reinforcing factors per number of platelets per stack.

No. of platelets per stack Reinforcing factor (RF) Halpin–Tsai equation Mori–Tanaka theory

d0 0 1 = 0.96 nm d0 0 1 = 1.8 nm d0 0 1 = 0.96 nm d0 0 1 = 1.8 nm

1 49.2 49.2 34.8 34.82 39.7 27.5 23.8 16.6
3 33.5 21.14 29.0 17.55 25.6 15.1
10 16.3 9.3

hydrolysis of ester linkages at high temperature, may beone reason [195].

Finally, it is worth noting at this point that, despite theaforementioned degradation problems encountered duringmelt intercalation, very few authors have used stabilizationsystems in the preparation of polymeric nanocomposites.

In Table 15 several PLS nanocomposites prepared viamelt intercalation are presented as typical examples.

6. Nanocomposite properties

6.1. Mechanical properties

6.1.1. The reinforcing mechanism of layered silicatesThe first mechanism that has been put forward to

explain the reinforcing action of layered silicates is one alsovalid for conventional reinforcements, such as fibers, whichis schematically depicted in Fig. 33. That is, rigid fillersare naturally resistant to straining due to their high mod-uli. Therefore, when a relatively softer matrix is reinforcedwith such fillers, the polymer, particularly that adjacent tothe filler particles, becomes highly restrained mechanically.This enables a significant portion of an applied load to becarried by the filler, assuming, of course, that the bondingbetween the two phases is adequate [64]. From this mech-anism it becomes obvious that the larger the surface of thefiller in contact with the polymer, the greater the reinforc-ing effect will be. This could partly explain why layered
silicates, having an extremely high specific surface area(on the order of 800 m2/g) impart dramatic improvementsof modulus even when present in very small amountsin a polymer. In fact, the low silicate loading requiredin nanocomposites to effect significant property improve-
Fig. 33. Reinforcement mechanism in composite materials.

18.3 11.714.9 9.312.7 7.87.5 4.6

ments, is probably their most distinguishing characteristic.In most conventionally filled polymer systems, the mod-ulus increases linearly with the filler volume fraction,whereas for nanocomposites much lower filler concentra-tions increase the modulus sharply and to a much largerextent [55].

However, some authors have argued that the dramaticimprovement of modulus for such extremely low clay con-centrations (i.e. 2–5 wt.%) cannot be attributed simply tothe introduction of the higher modulus inorganic filler lay-ers. A proposed theoretical approach assumes a layer ofaffected polymer on the filler surface, with a much highermodulus than the bulk equivalent polymer. This affectedpolymer can be thought of as a region of the polymermatrix that is physisorbed on the silicate surface, and isthus stiffened through its affinity for and adhesion to thefiller surface. Obviously, for such high aspect ratio fillers asthe layered silicate layers, the surface area exposed to thepolymer is huge and, therefore, the significant increases inthe modulus with very low filler content are not surprising.Furthermore, beyond the percolation limit, the additionalsilicate layers are incorporated in polymer regions thatare already affected by other silicate layers, and thus it isexpected that the enhancement of modulus will becomemuch less dramatic [198].

In order to prove the effect of degree of exfoliationon nanocomposite mechanical properties, Fornes and Paul[64] used an analytical approach to elucidate how incom-plete exfoliation influences nanocomposite stiffness. Theyexpressed the modulus of a simple clay stack in the direc-tion parallel to its platelets, by using the rule of mixtures:

Estack = �MMTEMMT + �galleryEgallery

where �MMT is the volume fraction of silicate layers in thestack, EMMT is the modulus of MMT, �gallery is the volumefraction of gallery space and Egallery is the modulus of thematerial in the gallery, which is expected to be much lessthan EMMT. The volume fraction occupied by gallery space,�gallery can be expressed in terms of X-ray d-spacings, as

�gallery = (n − 1)(d0 0 1 − tplatelet)d0 0 1(n − 1) + tplatelet

where n is the number of platelets per stack, d0 0 1 is the
repeat spacing between silicate particles, and tplatelet is thethickness of a silicate platelet. Obviously, when the numberof platelets in a stack is equal to one, the system repre-sents an individual exfoliated platelet. Table 16 shows howthe number of platelets in a stack affects the reinforce-


F and (b)f

m(mscittorfis

6

eiTptas4ioimwwwop

bpw2ecsapt

Most polymer–clay nanocomposite studies report ten-sile properties, such as modulus, as a function of claycontent [31], as in Fig. 36. This plot of Young’s modulus ofnylon 6 nanocomposite vs. filler weight content, shows a

ig. 34. The effect of the number of platelets per stack on the (a) modulusrom Fornes and Paul by permission of Elsevier Science Ltd., UK.

ent factor (RF in an unexchanged, non-expandable clayd0 0 1 = 0.96 nm) as well as in an intercalated or organically

odified clay (d0 0 1 = 1.8 nm). Increasing n in both stackingcenarios leads to lower reinforcement efficiencies, espe-ially for the intercalated clay. Interestingly, the largest dropn reinforcement is experienced when going from one towo platelets per stack. Overall, the trends in Table 16 showhe high sensitivity of nanocomposite stiffness to the levelf exfoliation. Fornes et al. concluded that stacks of plateletseduce stiffness of nanocomposites through lower effectiveller moduli and reduced aspect ratio, the effects showneparately in Fig. 34.

.1.2. Modulus and strengthIn general, the addition of an organically modified lay-

red silicate in a polymer matrix results in significantmprovements of Young’s modulus, as can be seen inable 17 for a number of different materials. For exam-le, Gorrasi et al. [156] reported an increase from 216o 390 MPa for a PCL nanocomposite containing 10 wt.%mmonium-treated montmorillonite, while in anothertudy [201], Young’s modulus was increased from 120 to45 MPa with addition of 8 wt.% ammonium treated clay

n PCL. Similarly, in the case of nylon 6 nanocompositesbtained through the intercalative ring opening polymer-zation of �-caprolactam, a large increase in the Young’s

odulus at rather low filler content has been reported,hatever the method of preparation: polymerizationithin organo-modified montmorillonite, polymerizationithin protonated �-caprolactam swollen montmorillonite

r polymerization within natural montmorillonite in theresence of �-caprolactam and an acid catalyst [45].

However, exceptions to this general trend haveeen reported. As shown in Fig. 35, in crosslinkedolyester/OMLS nanocomposites, the modulus decreasesith increasing clay content; in fact, the drop for the

.5 wt.% nanocomposite was greater than expected. Toxplain this phenomenon, it was proposed that the inter-
alation and exfoliation of the clay in the polyester resinerve to effectively decrease the number of crosslinks fromtopological perspective. The origin of the greater drop inroperties of the 2.5 wt.% nanocomposites may be tracedo the morphology; i.e. it was observed that the sam-
aspect ratio for the simplified arrangement of platelets [64]. Reproduced

ple showed exfoliation on a global scale compared to thenanocomposite containing 10 wt.% clay, indicating that thecrosslinking density is inversely proportional to the degreeof exfoliation [140].

Apart from the modulus, the addition of OMLS in apolymer matrix usually also increases the tensile strengthcompared to that of the neat polymer material. For exam-ple, Shelley et al. [32] reported a 175% improvement inyield stress accompanied by a 200% increase in tensilemodulus for a nylon 6 nanocomposite containing 5 wt.%clay. However, it should be emphasized that the effect ofnanocomposite formation on tensile strength is not as clearas in the case of the modulus since reductions of tensilestrength upon nanocomposite formation have also beenreported. Such examples are included in Table 18, whichlists the tensile strengths of a number of nanocompositematerials and compares them with the corresponding val-ues for the neat polymers.

Fig. 35. Tensile modulus vs. clay concentration for crosslinked polyesternanocomposites [140]. Reproduced from Bharadawaj, Mehrabi, Hamilton,Trujillo, Murga, Fan, Chavira and Thompson by permission of Elsevier Sci-ence Ltd., UK.


Table 17Young modulus of various PLS nanocomposites.

Nanocomposite Clay content (wt.%) Young modulus (GPa) Ref.

PA6/MMT (in situpolymerization)

0 1.11 [199]4.7 1.875.3 2.04

PA6(LMW)/MMT (meltintercalation)

0 2.82 [14]3.2 3.656.4 4.92

PA6(MMW)/MMT(melt intercalation)

0 2.71 [14]3.1 3.667.1 5.61

PA6(HMW)/MMT (meltintercalation)

0 2.75 [14]3.2 3.927.2 5.70

PP(7.2%PP-g-MA)/OMLS

0 0.714 [186]7.2 0.838

PP(21.6%PP-g-MA)/OMLS

0 0.760 [186]7.2 1.010

EVA/Cloisite Na 0 0.0122 [167]3 0.0135

EVA/Cloisite 20A 0 0.0122 [167]3 0.0249

EVA/Cloisite 25A 0 0.0122 [167]3 0.0220

EVA/Cloisite 30B 0 0.0122 [167]3 0.0228

EVA/Nanofil 757 0 0.0122 [167]3 0.0116

EVA/Nanofil 15 0 0.0122 [167]3 0.0240

EVA/Somasif ME100 0 0.0122 [167]3 0.0124

EVA/Somasif MAE 0 0.0122 [167]3 0.021

Soft PU/30B (solutionintercalation)

0 0.0075 [88]3 0.01387 0.024

Soft PU/30B (meltintercalation)

0 0.0072 [88]3 0.01147 0.0193

Hard PU/30B (solutionintercalation)

0 0.050 [88]3 0.0867 0.134

Hard PU/30B (meltintercalation)

0 0.061 [88]3 0.0817 0.119

HDPE/o-MMT 0 1.020 [200]
0.91.82.84.0
constant large rate of increase of modulus up to ca. 10 wt.%of nanoclay, whereas above this threshold the aforemen-
tioned levelling-off of Young’s modulus is observed. Thischange corresponds to the passage from totally exfoliatedstructure (below 10 wt.%) to partially exfoliated—partiallyintercalated structure (for 10 wt.% clay and above), as deter-mined by XRD and TEM [1,55].
1.0601.2501.3801.360

In another study, Liu and Wu [146] studied the mechan-ical performance of PA66 nanocomposites prepared via
melt intercalation, using epoxy co-intercalated clay. Thetensile strength increases rapidly from 78 MPa for PA66up to 98 MPa for PA66CN5, but the increasing amplitudedecreases when the clay content is above 5 wt.%. A simi-lar phenomenon is observed in the dependence of tensile


Table 18Tensile strength of various PLS nanocomposites.

Nanocomposite Clay content (wt.%) Tensile strength (MPa) Ref.

PA6/MMT (in situpolymerization)

0 68.6 [199]4.7 97.25.3 97.3

PA6(LMW)/MMT (meltintercalation)

0 69.2 [14]3.2 78.96.4 83.6

PA6(MMW)/MMT(melt intercalation)

0 70.2 [14]3.1 86.67.1 95.2


0 69.7 [14]3.2 84.97.2 97.6

PMMA/OMLS 0 53.9 [1]12.6 62.0

PS/OMLS 0 28.7 [1]17.2 23.424.6 16.6

EVA 0 28.4 [167]EVA/Cloisite Na 3 25.9EVA/Cloisite 20A 3 25.8EVA/Cloisite 25A 3 26.2EVA/Cloisite 30B 3 30.7EVA/Nanofil 757 3 27.6EVA/Nanofil 15 3 26.7EVA/Somasif ME100 3 24.5EVA/Somasif MAE 3 25.1


0 45 [88]3 317 21


0 58 [88]3 447 34

PU/MMT 0 5.9 [142]5 6.2

10 6.521.5 8.3

PE/JS 0 22 [184]5 25

10 2715 28

PE/DM 0 22 [184]5 21

10 2315 24

HDPE/o-MMT 0 27 [200]

miaa

mnitb

0.91.82.84.0

odulus of PA66CN on clay content. The smaller increasen amplitude observed with a clay loading above 5 wt.% wasgain attributed to the inevitable aggregation of the layerst high clay content.

Another example (Fig. 37) shows both the tensile
odulus and the yield strength of neat PA12 and the
anocomposites, which increased steadily with increas-ng organoclay loading. Compared to the virgin polymer,he tensile modulus of PA12/clay systems was improvedy about 40% upon adding only 5 wt.% of clay, while lim-

26262625

ited improvement of the tensile strength was observedby incorporating clay in the matrix. Again, it was sug-gested that there is an optimum clay concentration fornanocomposite tensile strength improvement. With fur-ther increase in clay loading a moderate decrease of tensile
strength was observed, suggesting that the relative amountof intercalation/exfoliation of the clay morphology gradu-ally increases with increasing clay content, since the tensilestrength is usually sensitive to the degree of dispersion[147].


Table 19XRD peak intensity (Im) and Young’s modulus of various nylon-6-basednanocomposites obtained by a one-step in situ intercalative polymer-ization of �-caprolactam with Na–montmorillonite in the presence ofdifferent acids.

Acid Im (cps) Young’s modulus (GPa)

Phosphoric acid 0 2.25Hydrochloric acid 200 2.05

Fig. 36. Effect of clay content on tensile modulus, measured atroom temperature, of organo-modified montmorillonite/nylon-6-basednanocomposite obtained by melt intercalation [170]. Reproduced from Liu,Qi and Zhu by permission of John Wiley & Sons, Inc.

Similarly, other factors that influence the degree of exfo-liation, apart from the clay content, also have an impact onnanocomposite modulus and strength.

This explains the variations observed in moduli of PA6nanocomposites prepared by intercalative ring openingpolymerization of �-caprolactam, with different kinds ofacids to catalyze the polymerization (Table 19). The WAXDpeak intensity Im, which is inversely related to the amountof exfoliated layers in the nanocomposite, also depends onthe nature of the acid used to catalyze the polymerizationprocess. For an increase in Im, a parallel decrease in Young’smodulus is observed, indicating that exfoliated layers arethe main factor responsible for the improvement in stiff-
ness, while intercalated particles, having a smaller aspectratio, play a rather minor role [1,55].
Cho and Paul [15] studied the effect of mixing deviceand processing parameters on the mechanical properties

Fig. 37. Tensile modulus and yield strength of PA12/clay nanocompositesas a function of clay concentration [147]. Reproduced from Phang, Liu,Mohamed, Pramoda, Chen, Shen, Chow, He, Lu and Hu by permission ofJohn Wiley & Sons Ltd. on behalf of the SCI.

Isophtalic acid 255 1.74Benzenesulfonic acid 280 1.74Acetic acid 555 1.63Trichloroacetic acid 585 1.67

of polyamide nanocomposites. In the case of compositesformed by single-screw extrusion, the exfoliation of theclay platelets is not extensive. Even after a second passthrough this extruder, undispersed tactoids are still easilyobserved with naked eye. However, the tensile strength andmodulus were slightly improved by the second pass. On theother hand, nylon 6 nanocomposites with good propertiescan be obtained over a broad range of processing condi-tions in the twin screw extruder. The final nanocompositeproperties are almost independent of the barrel tempera-ture over the range of typical nylon 6 processing, but theyare slightly improved by increasing the screw speed or bya second pass through the extruder. Therefore, processingconditions need to be optimized to allow greater exfolia-tion of the clay platelets and, thus, greater improvement inmechanical properties.

The effect of PA6 molecular weight and MMT contenton nanocomposite tensile modulus is shown in Fig. 38.The addition of organoclay leads to a substantial improve-ment in stiffness for the composites based on each of thethree PA6 samples examined, i.e. LMW, MMW and HMW(low, medium and high molecular weight, respectively).Interestingly, the stiffness increases with increasing matrixmolecular weight at any given concentration, even though
the moduli of the neat PA6s are all quite similar. Similartrends with respect to the level of organoclay content andmolecular weight are evident in the yield strength results(Fig. 39). Yield strength increases with MMT content; how-
Fig. 38. Effect of MMT content on tensile modulus for LMW, MMW,and HMW based nanocomposites [14]. Reproduced from Fornes, Yoon,Keskkula and Paul by permission of Elsevier Science Ltd., UK.


Table 20Influence of maleic anhydride-modified polypropylene content on the stiffness of PP matrices and PP/clay nanocompositesa.

Sample Filler content (wt.%) PP-MA content (wt.%) Young’s modulus (MPa)

PP 0 0 780PP/PP-MA 7 0 7.2 714PP/PP-MA 22 0 21.6 760PPCC 6.9 0 830PPCH 1/1 7.2 7.2 838PPCH 1/2 7.2PPCH 1/3 7.2

a PP = polypropylene; PP-MA x: polypropylene modified by maleic anhydride (posite, PPCH y/z = polypropylene based nanocomposite (y/z = weight ratio betwee

FaK

esteah

mopepcpas

mally stable, aromatic amine modifier containing active

TM

S

PPPPPPP

ig. 39. Effect of MMT content on yield strength for LMW, MMW,nd HMW based nanocomposites [14]. Reproduced from Fornes, Yoon,eskkula and Paul by permission of Elsevier Science Ltd., UK.

ver, while the HMW and MMW-based nanocompositeshow a steady increase in strength with clay content,he LMW-based nanocomposites show a less pronouncedffect. These differences with respect to molecular weightre attributed to the better exfoliation achieved for theigher molecular weight matrices [14].

Other factors that may play a crucial role in improve-ent of nanocomposite mechanical properties include the

rganic modification of the clay and the addition of com-atibilizers to the polymer matrix. As a representativexample, Young’s modulus values of PP/PP-MA nanocom-osites are listed in Table 20 and compared with the
orresponding microcomposite as well as simple PP-MA/PPolymer blends. It is readily observed that increasing themount of PP-MA increases the modulus, while compari-on of PP with the simple PP-MA/PP blends rules out any
able 21echanical properties of PE and PE/clay composites.

ample Tensile strength (MPa) Flexural strength (MPa)

E 22 26E/JS5 25 28E/JS10 27 33E/JS15 28 38E/DM5 21 26E/DM10 23 31E/DM15 24 30

14.4 96421.6 1010

x = wt.% of PP-MA in the blend); PPCC = polypropylene-based microcom-n y parts of filler and z parts of PP-MA).

possible effect of matrix modification due to the presenceof increasing amounts of PP-MA [1].

In another study, Hotta and Paul [162] performed ten-sile tests on various PE and PE-MA nanocomposites basedon organoclays with one or two alkyl tails. The increasein modulus with addition of MMT is much stronger for theorganoclay with two alkyl tails than for the one with a singletail, as would be expected on the basis of the much betterdispersion of clay platelets for the surfactant with two alkyltails. Similar trends were observed also for nanocompositeyield strength. Interestingly, the authors noted that thereis no advantage in adding PE-MA for building modulus orstrength at low MMT content (≤2.5 wt.%), in spite of themorphological differences seen. On the contrary, there is aclear advantage in adding PE-MA at higher MMT contents.Even though the benefit for modulus is not as great as mightbe expected, in the absence of PE-MA, the yield strengthactually decreases on addition of MMT beyond 2.5 wt.%.

Table 21 lists the strength and modulus values forPE-based nanocomposites, in which the initial montmo-rillonite was modified by two intercalating agents: thecommonly used dioctadecyldimethyl ammonium chloride(DM) and the reactive N-�-trimethoxysilanepropyl octade-cyldimethyl ammonium chloride (JS). Both strength andmodulus are higher in the case of the reactive intercalat-ing agent, owing to the better dispersion of the organoclay[184].

The effect of clay organic modification on nanocom-posite mechanical properties is also demonstrated inFig. 40, which presents the ultimate strength of PU-nanocomposites with different contents of two organicallytreated montmorillonites: MO-MMT, treated with a ther-

groups, and C16-MMT, treated with a quaternary alkylammonium salt. As can be seen the ultimate strengthincreased dramatically with clay content and reached amaximum at 5 wt.% MMT, where the ultimate strength of

Flexural modulus (MPa) Izod impact strength (J/m)

710 20780 16

1050 161330 12750 22980 16

1030 14


Table 22Tensile test results of Polypox H205 nanocomposites processed with low shear.

Specimen Average ultimate tensile stress (N/mm2) Average tensile modulus (N/mm2) Relative ultimate stress Relative tensile modulus

0% I30E 59.0 ± 0.6 2565 ± 115% I30E 54.3 ± 9.2 2796 ± 3110% I30E 53.6 ± 6.5 3075 ± 56

The authors attributed this phenomenon to the fact that

Fig. 40. Effect of organic-MMT loading on the tensile strength of (a)PU/MO-MMT and (b) PU/C16-MMT [202]. Reproduced from Xiong, Liu,Yang and Wang by permission of Elsevier Science Ltd., UK.

the nanocomposites increased by about 450% for C16-MMTand 600% for MO-MMT, compared with that of pure PU,indicating that the improved mechanical strength dependson the characteristics of the modifier [202].

The extent of improvement of nanocomposite mechan-ical properties will also depend directly upon the averagelength of the dispersed clay particles, since this determinestheir aspect ratio and, hence, their surface area [55,203].

At this point we note that several authors have alsopointed out factors that have an adverse effect on nanocom-posite modulus and/or strength and need to be taken intoconsideration when preparing nanocomposite materials.

Quite interestingly, Gopakumar et al. [151] found thatthe exfoliation of 5 and 10 wt.% clay in PE-MA increasedYoung’s modulus by 30 and 53%, respectively, whereas thetensile stress at yield showed only a marginal increase, upto a maximum of 15% for the 10 wt.% clay composition. Theauthors noted that the greatly enhanced interfacial area
derived from exfoliation of the clay improves the mechani-cal reinforcement potential of the filler. However, given thatthe mechanical properties of a filled system depend on twoprincipal factors, i.e. crystallinity of the polymer matrix and
Table 23Flexural test results of Polypox H205 nanocomposites processed with low shear.

Specimen Average flexural modulus (N/mm2) Average ultimate flexural s

0% I30E 2755 ± 83 95.0 ± 1.85% I30E 2966 ± 90 97.3 ± 2.07.5% I30E 3101 ± 85 102.2 ± 3.210% I30E 3294 ± 76 102.4 ± 3.1

1.000 1.0000.920 1.0900.909 1.199

the extent of filler reinforcement, the degree of crystallinitymust also be considered.

In another study dealing with the effect of matrixvariations on mechanical properties of nanocomposites,Chaudhary et al. [171] studied the tensile properties ofnanocomposites based on EVAs with various VA contentsand two alternative organoclays. Since in EVA with increas-ing VA content the crystallinity of the polymer decreases(and will lower the stiffness), while the polarity increases(and will increase the intercalation), the authors suggestedthat in their system, the stiffness and toughness responseswould reflect an interplay of two factors: (a) an increase inthe “rigid” amorphous phase due to polymer–clay interca-lation and (b) an increase in the “mobile” amorphous phasedue to the increasing VA content. Experimental resultsshowed that the influence of increasing clay concentrationon the tensile behavior of EVA matrices was significant onlywith a low or moderately polar EVA matrix (9 and 18% VA).Thus, a linear proportionality was found between clay con-centration and tensile modulus for EVA-9 and EVA-18, arelation not observed with EVA-28. In fact, it is very dif-ficult to compare the extent of the improvement of themechanical properties of different EVA/clay nanocompos-ites reported so far, because EVAs of different vinyl acetatecontents have been processed into the nanocompositeswith different clays and different modifying agents by dif-ferent methods [81].

In the case of high Tg thermosets, it has been suggestedthat neither intercalated nor exfoliated nanosilicates leadto an improvement of the tensile stress at break, but rathermake the materials more brittle. This effect appears to begenerally more pronounced for intercalated structures thanfor exfoliated ones [1].

The results of tensile tests conducted by Hackman andHollaway [134] on epoxy nanocomposites conventionallyprepared under low-shear (stirring for 5 h at 90 ◦C) arehighlighted in Table 22. The tensile modulus increasedby 9.0 and 19.9% with 5 and 10 wt.% clay loading respec-tively. However, the ultimate tensile stress decreased withincreasing clay content, although the variation was large.

large clay particles act as impurities and increase stressconcentrations. Flexural tests were also conducted and theresults are outlined in Table 23. As can be seen, the flexuralmodulus and ultimate flexural stress increased by 19.6 and

tress (N/mm2) Relative flexural modulus Relative flexural stress

1.000 1.0001.076 1.0341.126 1.0551.196 1.077


Table 24Summary of tensile properties of thermoplastic PU-based nanocomposites.

30B content (wt.%) Solvent cast Melt compounded

Young’s modulus (MPa) Tensile strength (MPa) Fail strain (%) Young’s modulus (MPa) Tensile strength (MPa) Fail strain (%)

SPU0 7.5 45 1136 7.2 21 14453 13.8 31 1109 11.4 22 11637 24 21 1030 19.3 7 568

H0 8983 8087 704

7Fm9Iwc

arccsnSBiis

piwtab

TT

O

0

1

2

3

PU50 5886 44

134 34

.7%, respectively, for specimens containing 10 wt.% clay.or nanocomposites processed under high shear, the tensileodulus and ultimate tensile stress increased by 18.7 and

.3%, respectively, when 5 wt.% clay loading was applied.n this case, the improvement in ultimate tensile strengthas attributed to the smaller particles not generating stress

oncentrations leading to premature failure.A summary of the tensile properties of soft (SPU)

nd hard (HPU) polyurethane elastomers and of the cor-esponding nanocomposites, prepared by either solventasting or melt compounding, is provided in Table 24. Asan be seen, upon silicate addition large improvements intiffness were observed, which however were accompa-ied by a decrease in tensile strength and elongation [88].imilar trends have been reported by Tortora et al. [204].oth exfoliated and intercalated PU/o-MMT nanocompos-

tes showed an improvement in the elastic modulus uponncreasing the clay content, but a decrease in the stress andtrain at break.

In general, it has been argued that in the presence ofolar or ionic interactions between the polymer and the sil-

cate layers, the stress at break is usually increased, whereashen there is lack of interfacial adhesion, no or very slight

ensile strength enhancement is recorded [1]. Pegoretti etl. [149] found that the yield strength was not reducedy the addition of clay to recycled PET and considered

able 25ensile properties of PET hybrid fibers.

rganoclay (wt.%) DRa Ultimate str

(Pure PET) 1 463 47

10 5116 51

1 583 56

10 5016 48

1 683 55

10 5416 51

1 713 68

10 6216 55

a Draw ratio.b Elongation percent at break.

61 44 77681 20 283

119 15 100

this to be a sign of good interfacial adhesion; however, inthe same study, a slight decrease of stress at break and adramatic reduction of strain were reported. On the otherhand, in PS intercalated nanocomposites the ultimate ten-sile stress was found to decrease compared to that of thePS matrix and dropped further at higher filler content.This lack of strength was attributed to the fact that onlyweak interactions exist at the PS/clay interface, contrary toother compositions in which polar interactions may prevail,strengthening the matrix interface [205].

It should be noted that several authors have reportedinability to measure nanocomposite yield stress, becausethe materials often become brittle and fail before reach-ing the yield point. Such remarks were made by Gorrasiet al. [6], who conducted tensile tests on PCL nanocom-posite, containing 30 wt.% clay, and also on blends of thisnanocomposite with HMW PCL. For the blend containing15 wt.% clay only the elastic modulus could be evaluatedsince the sample did not reach the yield point, while lowerclay concentrations in the blend led to better mechanicalproperties in terms of flexibility and drawability. For the
initial nanocomposite, however, it was not even possible todraw the sample because of its brittleness.
An interesting study was performed by Chang et al.[193] who prepared PET-based nanocomposites throughin situ intercalative polymerization, and subsequently pro-

ength (MPa) Modulus (GPa) E.B.b

2.21 32.24 32.28 32.39 2

2.88 32.80 32.63 32.47 3

3.31 32.63 32.51 32.29 3

4.10 33.40 33.12 23.08 3


Table 26Bending modulus of various PLS nanocomposites.

Nanocomposite Clay content (wt.%) Bending modulus (GPa) Ref.

PLA/OMLS 0 4.8 [208]4 5.55 5.67 5.8

PE/JS 0 0.71 [184]5 0.78

10 1.05015 1.330

PE/DM 05

1015

duced nano-hybrid fibers by extrusion through the die ofa capillary rheometer. The hot extrudates were stretchedthrough the die of a capillary rheometer at 270 ◦C andimmediately drawn to various draw ratios (DR). As is evi-dent from Table 25, the tensile properties of the fibersformed increased with increasing amount of organoclayat DR = 1. When the organoclay was increased from 0 to3 wt.% in hybrids at DR = 1, the strength linearly improvedfrom 46 to 71 MPa, and the modulus from 2.21 to 4.10 GPa.On the other hand, it is quite interesting to note the effectof DR on the tensile strength and modulus of PET and PETnanocomposite fibers. As shown in Table 25, for pure PET,the strength and modulus increased from 46 to 51 MPa and2.21 to 2.39 GPa, respectively, as the DR was increased from1 to 16. However, the ultimate strength and modulus of thehybrid fibers decreased markedly with increasing DR. Anincrease in the mechanical tensile strength with increasingDR is very common for engineering plastics and is usuallyobserved in flexible coil-like polymers. However, nanocom-posite fibers did not follow this trend. Chang et al. suggestedthat higher stretching of the fiber leads to debonding andcreation of voids in the hybrid, which reduce the tensilemechanical properties. This study clearly illustrates thatnanocomposite materials may have a different responseto mechanical loads than the corresponding neat polymer
matrices.
Finally, even though nanocomposite researchers aregenerally interested in the tensile properties of the finalmaterials, there are a few reports concerning the flexuralproperties of PLS nanocomposites [206,207]. Some results

Table 27Bending strength of various PLS nanocomposites.

Nanocomposite Clay content (wt.%)

PLA/OMLS 0457

PE/JS 05

1015

PE/DM 05

1015

0.71 [184]0.750.981.030

obtained by bending tests on nanocomposite materials arepresented in Tables 26 and 27.

6.1.3. Toughness and strainThe brittle behavior often exhibited by nanocompos-

ites probably originates from the formation of microvoidsdue to debonding of clay platelets from the polymermatrix upon failure. This has been testified through care-ful inspection of fracture surfaces and is also correlated toobservations by in situ deformation experiments using TEM[147,181]. In fact, the observation of nanocomposite frac-ture surfaces is quite interesting. Fig. 41(a) shows a typicalfracture morphology in virgin nylon 12 and a ductile frac-ture as evidenced by plastic deformation. Fig. 41(b) and (c)show fracture surfaces of the nanocomposites containing1 and 5 wt.% clay, respectively. No distinct clay agglomer-ates are observed by scanning electron microscopy (SEM)even at high magnification, as shown in Fig. 41(d). For1 wt.% clay addition (Fig. 41(b)), the fracture surface becamesmoother compared with that of neat PA12; an evenmore brittle feature for clay concentration of 5 wt.% wasobserved in Fig. 41(c). Careful inspection of the fracturesurface at higher magnification of nanocomposite with5 wt.% clay (Fig. 41(d)) verifies the formation of microvoidsdue to the debonding of clay platelets from the matrix.
Usually, microvoids are formed around the large inhomo-geneities, which become evident especially at high clayloadings. These microvoids will coalesce with formation oflarger cracks causing embrittlement, ultimately resultingin reduced toughness [147].
Bending strength (MPa) Ref.

86 [208]134122105

26 [184]283338

26 [184]263130


F PA12; (b( , PramoL

ttsd

FtPS

ig. 41. SEM images showing fracture surfaces after impact tests: (a) neatd) high magnification of (c) [147]. Reproduced from Phang, Liu, Mohamedtd. on behalf of the SCI.

In the case of nylon 12 nanocomposites, Fig. 42 showshat the Izod impact strength monotonically decreases ashe clay concentration increases. The toughness (repre-enting the energy absorption during the fracture process)ecreases by about 25% with 5 wt.% of clay. Similar obser-

ig. 42. Izod impact strength of PA12/clay nanocomposites as a func-ion of clay concentration [147]. Reproduced from Phang, Liu, Mohamed,ramoda, Chen, Shen, Chow, He, Lu and Hu by permission of John Wiley &ons Ltd. on behalf of the SCI.

) and (c) PA12 nanocomposites containing 1 and 5 wt.% clay, respectively;da, Chen, Shen, Chow, He, Lu and Hu by permission of John Wiley & Sons

vations of reduction in impact strength are also reportedin nylon 6/clay nanocomposites and PE-based nanocom-posites, indicating that the incorporation of clay intosemicrystalline thermoplastics usually results in toughnessreduction, i.e. the aforementioned embrittlement effectfrom clay addition [147].

On the other hand, some studies report little or nochange of toughness upon clay intercalation/exfoliation.For example, while the tensile strength and modulus ofPP nanocomposites increased rapidly with increasing claycontent from 0 to 5 wt.%, the notched Izod impact strengthwas constant, within experimental error, in the clay con-tent range between 0 and 7 wt.% [158]. Another studyreports the impact properties for exfoliated nylon 6-basednanocomposites prepared either by in situ intercalativepolymerization or by melt intercalation. In that studymarginal reductions in impact properties are reported,whatever the exfoliation process used. In the case of insitu intercalative polymerization, the Izod impact strengthis reduced from 20.6 to 18.1 J/m when 4.7 wt.% clay is incor-porated. Charpy impact tests show similar reduction in theimpact strength, with a drop from 6.21 kJ/m2 for the fillerfree matrix, down to 6.06 kJ/m2 for the 4.7 wt.% nanocom-posite. Fig. 43 shows that the decrease in the Izod impact
strength of melt-intercalated nylon 6 nanocomposites isnot very pronounced over a relatively large range of fillercontent [170].
Furthermore, toughness improvements upon clay dis-persion have also been reported—a remarkable result,


Fig. 43. Effect of clay content on notched Izod impact strength ofnylon-6/clay nanocomposites obtained trough melt intercalation [170].Reproduced from Liu, Qi and Zhu by permission of John Wiley & Sons, Inc.

considering that conventional polymer–clay compos-ites, containing aggregated nanolayer tactoids ordinarilyimprove rigidity but sacrifice toughness and elongation [7].

As an example, Liu and Wu [146] observed a toughen-ing effect in PA66CN. The notched Izod impact strengthincreased from 96 to 146 J/m upon 5 wt.% clay addition, andremained higher than that of PA66, even with higher claycontent.

Such results are particularly surprising, considering thatfrom length-scale arguments it is known that toughen-ing occurs over a specific size range; effective tougheningnecessitates a filler size greater than 0.1 �m and maynot be energetically favorable at nano-length scales. Also,the sizes of the nanoparticles are generally too small toprovide toughening through a crack-bridging mechanismand cannot effectively enhance crack-trajectory tortuosity.Therefore, the extremely reduced scale of a fully exfoli-ated nanocomposite does not lend itself to a tougheningapplication. However, in an intercalated system there isconsiderable interaction between silicate layers that mightalleviate this concern [45].

For example, Zerda and Lesser [45] showed that thegross yielding behavior of a glassy thermoset was sub-stantially modified upon the formation of intercalatednanocomposites, with void formation within clay aggre-gates leading to the evolution of a visible shear-banded

Fig. 44. Effect of MMT content on elongation at break for LMW, MMW, and HMW5.1 cm/min [14]. Reproduced from Fornes, Yoon, Keskkula and Paul by permission

lymer Science 33 (2008) 1119–1198

zone in compression samples. The fracture behaviorappears to be most dramatically improved in the interca-lated system. The fracture energy of the composites wasincreased by 100% at clay concentration of 5 wt.%. By inves-tigating the surface roughness and crack propagation undersubcritical loading, it has been hypothesized that the cre-ation of additional surface area on crack propagation is theprimary means for toughening intercalated systems. Themorphology of the system plays an important role in thetoughening mechanism because the spacing of regions ofintercalated clay is important to toughening. It is believed,therefore, that the intercalated morphology can affordsome property improvements that are unavailable to thefully exfoliated systems.

Concerning the fracture behavior of EVA-basednanocomposites, Peeterbroeck et al. [167] concluded thatit is independent of the origin of the clay, while it appearsto be related to the nature of the clay organo-modifier andthe state of nanocomposite dispersion. On the other hand,Kim et al. [44] attributed the enhanced toughness theyobserved for intercalated PA12 nanocomposites to thefact that some amount of applied energy is dissipated bysplitting, sliding or opening of the separated bundles in thestacked layers. Also, LePluart et al. [136] reported that theincorporation of a benzyl dimethyl tallow alkyl ammoniummontmorillonite in rubbery and glassy epoxy matricesleads to promising improvement of mechanical properties.They obtained an interesting stiffness/toughness balancefor very low filler contents and without reducing the Tg

of the matrix, which is particularly interesting consid-ering how the brittleness of epoxies limits their use intechnological areas where their high Tg is often highlyappreciated.

Quite interestingly, Fornes et al. [14] investigated howthe matrix molecular weight as well as the extensionrate during tensile tests affect the ductility of PA6-basednanocomposites. Fig. 44 presents the relationship betweenMMT content and elongation at break for PA6 matrices
of different molecular weights, for two different rates ofextension. As shown in Fig. 44a, the virgin polyamides arevery ductile at a test rate of 0.51 cm/min. With increasingclay content, the ductility gradually decreases, however, theHMW and MMW-based composites attain reasonable lev-
based nanocomposites at a crosshead speed of (a) 0.51 cm/min and (b)of Elsevier Science Ltd., UK.


Table 28Elongation at break of various PLS nanocomposites.

Nanocomposite Clay content (wt.%) Elongation at break (%) Ref.

PA6(LMW)/MMT 0 28 [14]3.2 116.4 4.8

PA6(MMW)/MMT 0 101 [14]3.1 187.1 5.0


0 129 [14]3.2 277.2 6.1

EVA 0 1406 [167]EVA/Cloisite Na 3 1403EVA/Cloisite 20A 3 1231EVA/Cloisite 25A 3 1259EVA/Cloisite 30B 3 1266EVA/Nanofil 757 3 1358EVA/Nanofil 15 3 1291EVA/Somasif ME100 3 1312EVA/Somasif MAE 3 1270


0 1136 [88]3 11097 1030

Soft PU/30B (meltintercalation)

0 1445 [88]3 11637 568


0 898 [88]3 8087 704

Hard PU/30B (meltintercalation)

0 776 [88]3 2837 100

PU/MMT 0 950 [142]5 1020

10 106521.5 1160

HDPE/o-MMT 0 36 [200]

ewp1acrtToblilccMa

a

0.91.82.84.0

ls of ductility at MMT concentrations as high as 3.5 wt.%,hile the elongation at break for the LMW based nanocom-osites decreases rapidly at low MMT content (aroundwt.%). Even though the opposite result could have beennticipated, considering that high molecular weight matri-es favor clay exfoliation, the authors attribute the largereduction of elongation at break in the LMW-based sys-ems to the presence of stacked silicate layers, as seen inEM photographs. On the other hand, the higher testing ratef 5.1 cm/min yields similar trends, as shown in Fig. 44b,ut the absolute level of elongation at break is significantly

ower. Interestingly, the strain at break for LMW compositess relatively independent of the rate of extension, simi-ar to what has been observed in glass fiber reinforcedomposites. Even at the highest clay content, the HMW
omposite exhibits ductile fracture, whereas the LMW andMW based nanocomposites fracture in a brittle manner
t the highest clay content.As in the case of toughness, contradictory results have

lso been presented concerning the effect of nanocom-

25201415

posite formation on the elongation at break, as can beseen in Table 28. Even though in most cases this propertydeteriorates when a layered silicate is dispersed into a poly-mer matrix, nanocomposites have been reported exhibitingsimilar or even higher elongations at break than the neatmatrices.

For example, in the case of EVA-12/MMT nanocompos-ites, a significant increase of both the strength and theelongation has been reported with the introduction of theorganoclay into the EVA-12 matrix. However, this enhance-ment is a maximum when the clay concentration is only2 wt.%. Further increase of clay content causes reduction inmechanical properties, probably due to aggregation of claylayers, as already discussed [209]. Thellen et al. [152] con-ducted tensile tests on PLA-based nanocomposite blown
films and recorded improvements up to 40% for both themodulus and the elongation. Yao et al. [142] also reportedimprovements in strain at break. Data of tensile strengthand strain-at-break vs. clay content are shown in Table 29.The authors suggested that the improved elasticity is due


Table 29Tensile strength and strain-at-break vs. the loading of layered clay.

Weight fractionof the clay (%)

Tensile strength (MPa) Strain-at-break (%)

0 5.9 9505 6.2 1020

10 6.5 106521.5 8.3 1160

in part to the plasticizing effect of gallery onium ions,which contribute to dangling chain formation in the matrix.Accordingly, Chen and Evans [210] observed a dramaticimprovement in tensile elongation at break in the pres-ence of clay. At low clay loadings, test pieces underwentyielding during tension, similar to pristine PCL, but withdramatic increases in ductility, quite the opposite of theusual effect of adding a particulate filler to a polymer. Whenthe clay loading was high, typically higher than 20 wt.%, thecomposites became brittle and did not reach the yield point.

Finally, it is worth summarizing the work of Hong etal. [185] on PP-based RTPO/clay nanocomposites, preparedby using PP-MA as a compatibilizer. PP-based RTPO (orin reactor made TPO) is a blend of PP and poly(ethylene-co-propylene) (EPR), produced by the bulk polymerizationof propylene, followed by gas-phase copolymerization ofethylene and propylene driven by the TiCl4/MgCl2-basedcatalyst system. Such materials, like the conventionalblends of PP/EPR prepared by mechanical blending, exhibitimproved flexibility and toughness compared to neat PP.Moreover, because the rubber phase can be dispersed uni-formly and reach a high degree of dispersion in these in situblends, it is possible to achieve more intimate interactionbetween the matrix and the rubber phase. The com-positions and tensile properties of polypropylene-basedRTPO/clay nanocomposites are reported in Table 30. As canbe seen, the tensile moduli of the nanocomposites becamehigher as the clay content increases. On the other hand, theelongation at break decreases as the clay content increases,but the value of nanocomposites containing 10 wt.% clay is437%, which is much higher than that of PP/clay nanocom-posites reported elsewhere. As the authors claim, theseelongational properties of PP based RTPO/clay nanocom-posites are unique and promising for many applications.In fact, for reasons of comparison, Hong et al. also preparedand tested nanocomposites using PP/EPR mechanical blendmatrix, modified with PP-MA. For these materials, the elon-gation at break values were about 50%, which are much
lower than those of RTPO clay nanocomposites and is notsuitable for industrial application. The authors attributedthis discrepancy to the difference of dispersion homogene-ity and domain size of ethylene copolymer between RTPOand PP/EPR mechanical blends.
Table 30Compounding formulations and tensile properties of PP-based RTPO/PpgMA/clay

Sample PP-based RTPO (wt.%) PpgMA (wt.%) Clay (wt.%) Tensile streat yield (M

RTPO 100 0 0 5.1RTPO NC3 88 9 3 6.4RTPO NC5 80 15 5 8.1RTPO NC10 60 30 10 14.2

Fig. 45. Stress–strain curves for nylon 6 and 95/05 composites at acrosshead speed of 5.08 cm/min [15]. Reproduced from Cho and Paul bypermission of Elsevier Science Ltd., UK.

6.1.4. Comparison and synergistic effects of clays andconventional reinforcements

Typically, layered silicates are incorporated in polymericmaterials as the sole reinforcing element. However, severalstudies have investigated the potential synergistic effects ofclays and conventional reinforcements, such as glass fibers.

In this context, Wu et al. [52] studied the effect ofadding glass fibers to PA6 and PA6-based nanocompos-ites containing 3 wt.% montmorillonite. They found that thetensile strength of PA6/clay containing 30 wt.% glass fibersis 11% higher than that of PA6 containing 30 wt.% glassfibers, while the tensile modulus of the nanocompositesincreases by 42%. Bending strength and bending modulusof neat PA6/clay are similar to PA6 reinforced with 20% glassfibers. However, the notched Izod impact strength of thenanocomposite is lower than that of neat polyamide 6, andis further decreased with the addition of fibers.

In another study, typical stress-strain diagrams forPA6 and composites containing 5 wt.% of fillers are com-pared, as shown in Fig. 45 (at 5.08 cm/min) and Fig. 46(at 0.5 cm/min). A summary of the mechanical proper-ties of these materials is shown in Table 31. As can beseen from the table, regardless of the type of filler, the
strength and modulus are substantially increased relativeto the neat PA6, without significant variation in toughnessor impact strength, as measured by the standard Izod test.Furthermore, nanocomposites show superior mechanicalproperties, especially modulus, as compared with conven-
nanocomposites.

ngthPa)

Tensile strengthat break (MPa)

Elongation atbreak (%)

Tensile modulus(MPa)

20.6 1390 46.016.6 980 71.216.8 859 78.316.6 437 251


Fcp

tfinPdtsocutoimttgmotgleso

TM

P

PPPPP

ig. 46. Stress–strain curves for nylon 6 and 95/05 composites at arosshead speed of 0.51 cm/min [15]. Reproduced from Cho and Paul byermission of Elsevier Science Ltd., UK.

ional PA6 composites formed by compounding with glassbers or untreated clay. The elongation at break for theanocomposites is more or less the same as that of the neatA6, whereas, values for the conventional composites areramatically decreased. Also, the elongation at break forhe nanocomposites is greatly affected by the crossheadpeed, as is the case for neat PA6. On the other hand, ratef extension has little effect on the elongation of glass fiberomposites. It is noteworthy that the composite of PA6 withntreated Na+-MMT shows higher strength and modulushan neat PA6, which is quite contrary to the results fromther investigators, who claim that untreated clay compos-tes with PA6 are inferior to neat polymer in terms of some

echanical properties. Interestingly, a synergistic effect onhe tensile strength and modulus is again observed whenhe exfoliated nanocomposite is used as the matrix for alass fiber reinforced composite. As shown in Table 31, theodulus of the nanocomposite with 5 wt.% loading of the

rganoclay is improved about 38% relative to neat PA6 andhe glass fiber composite shows a 22% improvement. When
lass fibers are added to the nanocomposite, the modu-us is 81% higher than that of PA6. This exceeds what isxpected on the basis of simple additivity. Stiffness andtrength are dramatically improved as the amount of organ-clay increases. On the other hand, the impact strength and
able 31echanical properties of polyamide 6 composites.

olyamide composites Clay content (%) Izod impactstrength (J/m)

Modulu

olyamide 6 0 38 ± 4 2.66 ± 0A6/glass fiber 5 53 ± 8 3.26 ± 0A6/MMT 5 40 ± 2 3.01 ± 0A6/organoclay 3.16 38 ± 3 3.66 ± 0A6/organoclay/glass fiber 8 44 ± 3 4.82 ± 0

lymer Science 33 (2008) 1119–1198 1169

elongation at break remain at the levels of neat PA6 up toabout 5 wt.% of the organoclay, and decrease thereafter [15].

6.2. Dynamic mechanical properties

Dynamic mechanical analysis (DMA) measures theresponse of a material to a cyclic deformation (usually ten-sion or three-point bending type deformation) as a functionof the temperature. DMA results are expressed by threemain parameters: (i) the storage modulus (E′ or G′), cor-responding to the elastic response to the deformation; (ii)the loss modulus (E′′ or G′′), corresponding to the plasticresponse to the deformation and (iii) tan ı, that is, the E′/E′′

(or G′/G′′) ratio, useful for determining the occurrence ofmolecular mobility transitions such as the glass transitiontemperature [1].

Indicatively, the temperature dependence of G′, G′′ andtan ı of a nylon 6 matrix and various nanocomposites ispresented in Fig. 47. In the case of nanocomposites, themain conclusion derived from dynamic mechanical stud-ies is that the storage modulus increases upon dispersionof a layered silicate in a polymer. This increase is generallylarger above the glass transition temperature, and for exfo-liated PLS nanocomposite structures is probably due to thecreation of a three-dimensional network of interconnectedlong silicate layers, strengthening the material throughmechanical percolation [1]. Above the glass transition tem-perature, when materials become soft, the reinforcementeffect of the clay particles becomes more prominent, due tothe restricted movement of the polymer chains. This resultsin the observed enhancement of G′ [55]. For example, anepoxy-based nanocomposite, containing 4 vol.% silicates,showed a 60% increase in G′ in the glassy region, com-pared to the unfilled epoxy, while the equivalent increasein the rubbery region was 450% [135]. Similar results havealso been reported in the case of PP- [189], PCL- [80], SBS-[211], PA- [64,212], PLA- [83,153,208,213], and epoxy-basednanocomposites [135].

Enhancement of the loss modulus, G′′, has also beenreported for nanocomposite materials, however this aspectof dynamic mechanical performance is far less discussed inthe literature.

Finally, the tan ı values are affected in different waysby nanocomposite formation, depending on the polymer
matrix. For example, in PS based nanocomposites, a shiftof tan ı to higher temperatures has been observed, accom-panied by a broadening of this transition [205], whilethe opposite effect was reported in the case of PP-basednanocomposites [189]. Some authors observed a decrease
s (GPa) Yield strength(MPa)

Elongation atbreak (%)

Crosshead speed0.51 cm/min

Crosshead speed5.08 cm/min

.2 64.2 ± 0.8 200 ± 30 40 ± 8

.1 72.6 ± 0.8 18 ± 1.3 14 ± 4

.1 75.4 ± 0.3 22 ± 6.0 14 ± 3

.1 83.4 ± 0.7 126 ± 25 38 ± 19

.1 95.0 ± 0.9 8 ± 0.5 7 ± 4


ites, they have also been used in reproducing experimental

Fig. 47. . Temperature dependence of G′; G′′ and tan ı for N6 matrix andvarious N6CNs [55]. Reproduced from Ray and Okamoto by permission ofElsevier Science Ltd., UK.

of tan ı peaks, and considered this indicative of a glass tran-sition suppression by the presence of the clay. However,Fornes and Paul [64] pointed out that this conclusion is amisinterpretation, since the low values for the nanocom-posites are simply the result of dividing the relativelyconstant loss modulus values in the Tg region, by larger andlarger storage modulus values.

Quite surprisingly, DMA showed that above Tg, the mod-uli for the pure PU and the PU/o-MMT nanocompositesshow no obvious difference, while below Tg, addition ofo-MMT strongly influences the modulus values. Interest-ingly, the authors found that E′ and E′′ of the PU/o-MMTdecrease in comparison with values for the PU, for unclearreasons. On the other hand, significant enhancements ofE′ and E′′ were seen for the nanocomposite prepared usinga particular modified clay [202]. In the case of PLA-basednanocomposites, it was observed that PLACNs with a verysmall amount of o-PCL as a compatibilizer exhibited a verylarge enhancement of mechanical properties compared tothat of PLACN with comparable clay loading [153]. Kriko-
rian and Pochan [83] also studied the dynamic mechanicalproperties of neat PLA and nanocomposites prepared withOMLS. These authors found that at high temperatures thereinforcement effect of OMLS weakens, and suggested that
Fig. 48. Proposed model for the torturous zigzag diffusion path in anexfoliated polymer–clay nanocomposite when used as a gas barrier [85].Reproduced from Yano, Usuki, Okada, Kurauchi and Kamigaito by permis-sion of John Wiley & Sons Inc.

this indicates a weakening of the thermomechanical stabil-ity of these materials at high temperature.

6.3. Barrier properties

Generally, polymer/layered silicate nanocomposites arecharacterized by very strong enhancements of their bar-rier properties. Polymers ranging from epoxies and goodsealants (like siloxanes) to semi-permeable (e.g. polyureas)and highly hydrophilic (e.g. PVA) are all improved up to anorder of magnitude by low clay loadings [31].

The dramatic improvement of barrier properties can beexplained by the concept of tortuous paths. That is, whenimpermeable nanoparticles are incorporated into a poly-mer, the permeating molecules are forced to wiggle aroundthem in a random walk, and hence diffuse by a tortuouspathway, as shown in Fig. 48 [4,7,55,206,214–216].

The tortuosity factor is defined as the ratio of the actualdistance, d′, that the penetrant must travel to the shortestdistance d that it would travel in the absence of barriers. Itis expressed in terms of the length L, the width W and thevolume fraction of the sheets �s as

� = d′

d= 1 + L

2W�s

It becomes obvious from this expression that a sheet-likemorphology is particularly efficient at maximizing the pathlength, due to the large length-to-width ratio, as comparedto other filler shapes [1,50,55].

According to the model proposed by Nielsen, the effectof tortuosity on the permeability may, in turn, be expressedas

PPCN

Pp= 1 − �s

�

where PPCN and PP represent the permeability of thenanocomposite and the pure polymer, respectively and �s

is the clay content [50,55,217].Although the above equations were developed to model

the diffusion of small molecules in conventional compos-

results for the relative permeability in PLS nanocompos-ites. Discrepancies between the experimental data and thetheoretical line may be attributed either to inadequaciesof the model or to incomplete orientation of the particles

ess in Polymer Science 33 (2008) 1119–1198 1171

wkpfaab

Ntfonemniac

aoiTmv

otsttttoAiataceatb

wbmmt

osttsatdr


ithin the nanocomposite film plane [50,162]. In fact, theey assumption of the Nielsen model is that the sheets arelaced in an arrangement such that the direction of dif-usion is normal to the direction of the sheets. Clearly, thisrrangement results in the highest tortuosity, and any devi-tion from it would, in fact, lead to deterioration of thearrier properties [50,55].

Moreover, the tortuous path theory, including theielsen equation as well as other phenomenological rela-

ions (e.g. the Cussler [218] formula, the Barrel [219]ormula and the power law equation [220]), is groundedn the assumption that the presence of nanoparticles doesot affect the diffusivity of the polymer matrix. However,xperimental observations demonstrate that molecularobility in a polymer matrix, which is intimately con-

ected to the mass transport properties, diminished by clayncorporation. This reduction should be accompanied by

decrease in diffusivity of small molecules, which is notonsidered in the concept of tortuous paths.

Messersmith and Giannelis [118] studied the perme-bility of liquids and gases in nanocomposites and theybserved that water permeability in PCL nanocompositess dramatically reduced compared to the unfilled polymer.hey also noted how the decrease in permeability is muchore pronounced in the nanocomposites compared to con-

entionally filled polymers with much higher filler content.Liu and Wu [146], recorded the water absorption curves

f PA66 and corresponding nanocomposites. They foundhat with increasing clay content, the water absorption ataturation decreases rapidly from 7.6% for PA66 to 5.2% forhe nanocomposite containing 5 wt.% clay. They attributedhis reduction to the presence of immobilized polymer inhe amorphous phase. However, above 5 wt.% clay content,he decrease in the saturation amount of water is not sobvious, probably because of aggregation of silicate layers.lso, the diffusion coefficient values decrease greatly with

ncreasing clay loading, but after 5 wt.% clay content, themplitude of the decrease is obviously slower. In addition tohe immobilized phase explanation, and the increased aver-ge diffusion path length, since an epoxy-co-intercalatedlay was used in their study, the authors assumed that thepoxy groups between silicate layers have a strong inter-ction with amino and amide groups of the PA66 matrix,o some extent preventing them from forming hydrogenonds with water.

Significant reductions in diffusivity and maximumater uptake were also reported by Liu et al. [130] in epoxy-ased nanocomposites. Here, however, the decreasedaximum water uptake was attributed to the low maxi-um water uptake of the nanoclays (ca. 2.8 wt.%) compared

o the epoxy resin system (ca. 7.5 wt.%).Drozdov et al. [221] conducted moisture diffusion tests

n vinylester resin-MMT clay nanocomposites and demon-trated that the clay content affects in a similar wayhe diffusion coefficient and the constants expressinghe elastoplastic behavior, indicating that moisture diffu-
ion and elastoplasticity may reflect the same phenomenat the microlevel, associated with molecular mobility ofhe polymeric matrix. Moreover, their experimental dataemonstrated that moisture diffusion in the neat polymeresin is nearly Fickian, but is transformed to an anoma-
Fig. 49. Equilibrium concentration of water vapor, Ceq (g/100 g), as a func-tion of activity a = p/p0 for samples OMont ( ), NPU0 (�), NPU4 (�), NPU20(), NPU40 (♦) [204]. Reproduced from Tortora, Gorrasi, Vittoria, Galli,Ritrovati and Chiellini by permission of Elsevier Science Ltd., UK.

lous mode of transport of the penetrant molecules withan increase in clay concentration. The authors explainedthe anomalous moisture uptake by immobilization of watermolecules on the surfaces of the hydrophilic MMT clay lay-ers. In fact, they pointed out that, after a nanocompositeplate is immersed in water, three processes occur in thenanocomposite: (1) sorption of water molecules on thesample surfaces, (2) diffusion of water into the plate, and (3)adsorption of water molecules on the hydrophilic surfacesof clay layers, where these molecules become immobilized.

Many studies reported in the literature have focused onnanocomposite barrier properties against gases and vapors.As an example, Tortora et al. [204] measured the trans-port properties of PU/o-MMT nanocomposites (preparedusing a PCL nanocomposite “master-batch”) using watervapor as hydrophilic permeant and dichloromethane ashydrophobic one. For both vapors, the sorption behaviorchanged in the presence of the clay, as can be seen forexample in Fig. 49, where the equilibrium concentration,Ceq (g/100 g), of water vapor is represented as a functionof the vapor activity for all nanocomposites and for the o-MMT. The sorption curve of water vapor for o-MMT followsthe Langmuir sorption isotherm, in which the sorption ofsolvent molecules occurs at specific sites; therefore, whenall the sites are saturated, a plateau is reached. On the otherhand, the sorption of neat PU shows a linear dependenceof equilibrium concentration on activity, while nanocom-posites show a dual sorption shape, that is a downwardconcavity, an inflection point and an upward curvature.The prevailing mechanism in the first zone is the sorptionof solvent molecules on specific sites, due to interactinggroups. Tortora et al. inferred that this type of sorption
is due to the presence of clay in the polymers. At higheractivities, the plasticization of the polymeric matrix deter-mines a more than linear increase of vapor concentrationand a transition in the curve is observed, from a dual typeto a Flory-Huggins behavior. From the calculated values of


Fig. 51. Effect of montmorillonite weight fraction on diffusivity D of mont-morillonite/epoxy nanocomposites [223]. Reproduced from Ogasawara,Ishida, Ishikawa, Aoki and Ogura by permission of Elsevier Science Ltd.,UK.

Fig. 52. Effect of montmorillonite weight fraction on solubility, S, ofmontmorillonite/epoxy nanocomposites. The numerical curve based onrule-of-mixture is superimposed [223]. Reproduced from Ogasawara,Ishida, Ishikawa, Aoki and Ogura by permission of Elsevier Science Ltd.,UK.

Fig. 50. The O2 permeability of PET/o-MMT [222]. Reproduced from Keand Yongping by permission of Elsevier Science Ltd., UK.

the sorption parameters, defined as: S = d(Ceq)/dp, and thezero-concentration diffusion coefficients for water sorp-tion and dichloromethane vapor, the authors concludedthat the sorption did not drastically change on increasingthe clay content, whereas the zero-concentration diffusioncoefficient D0 strongly decreased with increasing inor-ganic content. The permeability calculated as the productSD0, was largely dominated by the diffusion parameter; itshowed a remarkable decrease up to 20 wt.% of clay and alevelling off at higher contents.

Ke and Yongping [222] tested the O2 permeabilityof intercalated PET nanocomposites. As demonstrated inFig. 50, a small amount of clay effectively reduced the per-meability of the PET film. When the content of o-MMTreached 3 wt.% the permeation of O2 was reduced to halfthat of the pure PET film. Further examples of barrier prop-erty improvement for PET nanocomposites designated forpackaging applications are given in Table 32 [216].

Ogasawara et al. [223] reported on improved helium gasbarrier properties of epoxy/MMT nanocomposites, com-pared to the pure resin. The estimated diffusivity, D,solubility, S, and permeability P are shown in Figs. 51–53,as functions of montmorillonite weight fraction. Dispers-ing MMT particles in the epoxy decreased the diffusioncoefficient D. For example, the diffusion coefficient of thenanocomposite with 6 wt.% clay was approximately one-tenth that of the base polymer. On the other hand, the
solubility increased with montmorillonite dispersion andpermeability remained almost constant due to the balanceof diffusivity and solubility.
On the other hand, Ray et al. [208] found that the O2 gaspermeability of PLA nanocomposites with 4, 5 and 7 wt.%

Table 32Relative barrier performance of newly developed rigid packaging basedon PET.

Container composition (supplier) Relative oxygentransmission rateat 23 ◦C 50% RH

PET 1PET nanocomposite (Tetrapak) <0.3PET/PA nanocomposite (Eastman) 0.4–0.7

Fig. 53. Effect of montmorillonite weight fraction on permeability, P,of montmorillonite/epoxy nanocomposites [223]. Reproduced from Oga-sawara, Ishida, Ishikawa, Aoki and Ogura by permission of Elsevier ScienceLtd., UK.


Table 33Oxygen gas permeabilities of the PLA/OMLS hybrid films.

Clay (wt.%) O2 gas (cm3/(m2 day))

C16MMT DTA-MMT C25A

0 777 777 777

cn

pdnpnc1hO

mab1npao

FmF

4 449 455 –6 340 353 430

10 327 330 340

lay, was reduced to 88, 85 and 81% compared to that ofeat PLA.

Similarly, Chang et al. [84] reported the oxygen gasermeability of PLA nanocomposites prepared with threeifferent kinds of OMLS using a melt intercalation tech-ique. Table 33 summarizes the results for O2 gasermeability. The results show that O2 gas permeability ofanocomposites systematically decreased with increasinglay content; and when the clay loading was as much as0 wt.%, the permeability for nanocomposite decreased toalf the PLA permeability, regardless of the nature of theMLS employed for the nanocomposite preparation.

Finally, Fig. 54 presents water vapor and oxygen per-eation results for PLA-based nanocomposite films. In

ll cases the nanocomposite films were better oxygenarriers than the pure PLA films (Fig. 54a), exhibiting a5–48% reduction in oxygen permeation rate. It is worth
oticing that the oxygen permeation rate of the nanocom-osite films was essentially independent of screw speednd feed rate, whereas the oxygen barrier propertiesf the neat PLA homopolymer were quite sensitive to
ig. 54. MOCON permeation data (a) oxygen and (b) water vapor trans-ission [152]. Reproduced from Thellen, Orroth, Froio, Ziegler, Lucciarini,

arrell, D’ Souza and Ratto by permission of Elsevier Science Ltd., UK.

lymer Science 33 (2008) 1119–1198 1173

processing, suggesting that the need to optimize process-ing parameters is more critical when working with thePLA homopolymer than with the PLA/OMLS nanocompos-ites. The PLA/OMLS nanocomposites also exhibited muchimproved barrier properties to water vapor relative tothe neat films (Fig. 54b). In general, clay incorporationdecreased the permeation rate of the resulting films towater vapor by about 40–50%, again independently of pro-cessing [152].

Summarizing: although a decrease of diffusivity is awell-established result of nanocomposite formation, con-tradictory results are reported concerning the saturationuptake values of various solvents or gases. Increases ofthe saturation uptake level are usually attributed to clus-tering phenomena. It is worth noticing, however, that innanocomposites the coexistence of phases with differentpermeabilities can cause complex transport phenomena.On the one hand, the organophilic clay gives rise to super-ficial adsorption and to specific interactions with thesolvents. In turn, the polymer phase can be considered,in most cases, as a two-phase, crystalline-amorphous sys-tem, the crystalline regions being generally impermeableto penetrant molecules. The presence of the silicate lay-ers may be expected to cause a decrease in permeability,due to the more tortuous path for the diffusing moleculesthat must bypass impenetrable platelets [133]. Simultane-ously, the influence of changes in matrix crystallinity andchain mobility, induced by the presence of the filler, shouldalways be taken into consideration [10].

6.4. Thermal stability

The thermal stability of polymeric materials is usuallystudied by thermogravimetric analysis (TGA). The weightloss due to the formation of volatile products after degra-dation at high temperature is monitored as a function oftemperature (and/or time). When heating occurs under aninert gas flow, a non-oxidative degradation occurs, whilethe use of air or oxygen allows oxidative degradation of thesamples [50,55].

Generally, the incorporation of clay into the polymermatrix was found to enhance thermal stability by acting as asuperior insulator and mass transport barrier to the volatileproducts generated during decomposition, as well as byassisting in the formation of char after thermal decompo-sition [50,55,133,224].

Vyazovkin et al. [225] compared the thermal degrada-tion of a PS nanocomposite with that of the virgin polymerunder nitrogen and air. As seen in Fig. 55 in both nitrogenand air the decomposition temperature of nanocompos-ites increased by 30–40 ◦C. The authors also observed thatthe virgin polymer degrades without forming any residue,whereas the nanocomposite (as expected) leaves someresidue.

Zanetti et al. [226] reported TGA curves of a nanocom-posite PE/EVA/o-MMT and the corresponding matrix
PE/EVA. Under nitrogen, these samples do not show greatdifferences of stability. However, in air, the PE/EVA blendis subject to a marked weight loss above 350 ◦C, to form a5 wt.% residue at 450 ◦C, which is completely oxidized tovolatile products between 470 and 550 ◦C. The nanocom-


Fig. 55. TGA curves of the degradation of PS100 and nPS90 at a heating rate5 ◦C min−1 in air and nitrogen [225]. Reproduced from Vyazovkin, Dranka,Fan and Advincula by permission of Wiley-VCH, Germany.

posite, on the other hand, displays a different pattern. Thepresence of 5 wt.% o-MMT is enough to change the poly-mer’s thermo-oxidative behavior and between 350 and480 ◦C the amount of residue is higher to that observed ina nitrogen flow. According to the authors, the organoclayshields the polymer from the action of oxygen, dramaticallyincreasing the thermal stability under oxidative conditions.

Bandyopadhyay et al. [227] reported the first improvedthermal stability of biodegradable nanocomposites thatcombined PLA and organically modified fluorohectorite ormontmorillonite. They showed that the PLA intercalatedbetween the galleries of FH or MMT clay resisted the ther-mal degradation under conditions that would otherwisecompletely degrade pure PLA. This conclusion has been
verified by a number of researchers in subsequent stud-ies. Thellen et al. [152] presented TGA curves for the neatpolymer and corresponding nanocomposites (Fig. 56) andreported that the onset of thermal degradation was approx-
Fig. 56. TGA curves of neat PLA and nano-PLA/MLS [152]. Reproducedfrom Thellen, Orroth, Froio, Ziegler, Lucciarini, Farrell, D’ Souza and Rattoby permission of Elsevier Science Ltd., UK.

lymer Science 33 (2008) 1119–1198

imately 9 ◦C higher for the nanocomposite than for the neatPLA.

The thermal stability of PCL-based nanocomposites hasalso been studied by TGA. Generally, the degradation ofPCL fits a two-step mechanism. First, random chain scis-sion through pyrolysis of the ester groups, with the releaseof CO2, H2O and hexanoic acid, and in the second step, �-caprolactone (cyclic monomer) formation as a result of anunzipping depolymerization process. It has been reportedthat the thermal stability of PCL/o-MMT nanocompositessystematically increases with increasing clay, up to a load-ing of 5 wt.% [1,228].

On the other hand, contradictory results are found inthe literature concerning the thermal degradation of PA6-based nanocomposites. Pramoda et al. observed that thedegradation onset temperature is 12 ◦C higher for PA6 with2.5% clay loading than that of virgin PA6 and that theonset temperature for the higher clay loading remainedunchanged. Also, Dabrowski et al. [229] showed that pro-tective barriers are formed during thermal degradation ofpolyamide 6/clay nanocomposite, which slow down therate of degradation via a diffusion process (hindering theescape of volatiles). However, TGA experiments by otherworkers did not show significant changes in the onset ofdegradation. For example, according to Jang and Wilkie[230] the mass loss behavior of PA6/clay nanocompositesis not significantly different from that of virgin PA6. Irre-spective of formulation, the temperatures at 50% mass losswere 471–476 ◦C, which was within the error range of theTGA instrument used. Moreover, other researchers foundthat PA6 nanocomposites have somewhat lower stabilitythan neat nylon 6, and attributed such observations tothe degradation effect of the quaternary alkylammoniumtreatment on the montmorillonite [15]. For example, Daviset al. [192] studied the thermal stability of PA6 and PA6nanocomposites, injection molded at 300 ◦C, by 13C NMR.They found that PA6 does not degrade at processing tem-perature, whereas there is significant decrease in molecularweight in nanocomposites under the same conditions. Theauthors noted that the degradation might depend uponwater in the nanocomposites, which may cause hydrolyticcleavage. The thermal degradation mechanism of PA6, pro-posed by Levchik et al., is shown in Fig. 57 [231].

In fact, despite the general improvement of thermalstability, decreases in the thermal stability of polymersupon nanocomposite formation have also been reported,and various mechanisms have been put forward to explainthe results. It has been argued, for example, that after theearly stages of thermal decomposition the stacked sili-cate layers could hold accumulated heat, acting as a heatsource to accelerate the decomposition process, in con-junction with the heat flow supplied by the outside heatsource [55]. Also, the alkylammonium cations in the organ-oclay could suffer decomposition following the Hoffmannelimination reaction, and the product could catalyze thedegradation of polymer matrices. Moreover, the clay itself
can also catalyze the degradation of polymer matrices.Thus, it becomes obvious that the organoclay may have twoopposing functions in thermal stability of nanocomposites:a barrier effect, which should improve the thermal stabilityand a catalytic effect on the degradation of the poly-


F a nucleR r Scienc

m[

sspunwt4Phiapian

cn

ig. 57. Dominant PA-6 thermal degradation products in the absence ofeproduced from Davis, Gilman and VanderHart by permission of Elsevie

er matrix, which should decrease the thermal stability184].

For example, Zhao et al. [184] investigated the thermaltability of PE-based nanocomposites in a nitrogen atmo-phere and recorded the TGA and DTGA (derivative curve)rofiles presented in Fig. 58. As can be seen from this fig-re, at the initial stage of degradation (before 400 ◦C), theanocomposites degrade faster than the pure matrix; thisas attributed to the Hoffmann elimination reaction and

he clay-catalyzed degradation. On the other hand, above00 ◦C nanocomposites appear to be more stable than pureE. The onset temperatures for nanocomposites are alligher than that of pure PE, but decrease with increas-

ng clay loading. Thus, the authors suggested that whenlow clay fraction is added to the polymer, the clay dis-

erses well and the barrier effect is predominant, but withncreasing loading, the catalyzing effect rapidly increases
nd becomes dominant, so that the thermal stability of theanocomposite decreases.
Other researchers have studied the effect of clayoncentration on the thermal stability of EVA-basedanocomposites. It has been found, for example, that the

ophile (a) and in the presence of a nucleophile, such as water (b) [192].e Ltd., UK.

thermal stability of EVA-12 (12 wt.% VA) increases withthe introduction of o-MMT. With an increase in the o-MMT loading over 2 wt.%, however, the hybrids show adecreasing trend in their initial thermal decompositiontemperature. These findings were attributed to the fact that,at low filler contents homogenous exfoliation and randomdispersion of clay is achieved on a nanometer level, whereasthe higher filler loading destabilizes the matrix, becauseof the aggregation of silicate layers [232]. In fact, EVA/claynanocomposites containing more vinyl acetate maintainthis thermal stability improvement up to 6–8 wt.% fillerloading, probably because of the degree of dispersion ofclay in the polymer matrix [209].

Similarly, Paul et al. [233] observed an increase in ther-mal stability of PLA nanocomposites with increasing claycontent, with a maximum at a clay loading of 5 wt.%. Withfurther increase of filler content, a decrease in thermal sta-
bility was observed—an effect explained by the relativeextent of exfoliation as a function of the amount of OMLS.
Phang et al. [147] found that the thermal stability ofPA12 is significantly enhanced as the clay concentrationincreases; for instance, by about 20 ◦C with incorporation


Table 34Thermal properties of PET hybrid fibers (DRa = 1).

Organoclay (wt.%) ninhb Tm (◦C) �Hm

c (J/g) T iD

d (◦C) wt600R

e

0 (pure PET) 1.02 245 32 370 11 1.26 247 32 375 82 0.98 245 33 384 153 1.23 246 32 386 21

a Draw ratio.in a phe
b Inherent viscosities measured at 30 ◦C by using 0.1 g/100 ml solutions
c Enthalpy change of fusion.d Initial weight reduction onset temperature.e Weight percent of residue at 600 ◦C.

of only 5 wt.% clay into the matrix. For PA12/organoclaynanocomposites with lower clay contents (<2 wt.%) onlylimited improvement in thermal stability was observed.This was attributed to the fact that the exfoliated struc-ture formed at low clay content is probably not enough totrigger the thermal and gas barrier properties effectivelyin the matrix, because of the short tortuous path formed.This result is consistent with reports claiming that opti-mum thermal stability is usually achieved for clay loadingsbetween 2.5 and 5 wt.%. For lower clay fractions, exfoli-ated morphology dominates, but this low clay content doesnot enable the barrier effect. At much higher clay content,on the other hand, particle agglomeration occurs. There-
fore, intercalated and exfoliated structures usually coexist,which again does not allow maximization of the tortuouspath.
As shown in Table 34, PET nanocomposites exhibitimproved thermal stability, with the initial thermal degra-

Fig. 58. TGA and DTGA curves of PE and PE/clay nanocomposites in nitro-gen atmosphere [184]. Reproduced from Zhao, Qin, Gong, Feng, Zhang andYang by permission of Elsevier Science Ltd., UK.

nol/1,1,2,2-tetrachloroethane (w/w) mixture.

dation temperature T iD increasing with the amount of

organoclay. A maximum increase of 16 ◦C was recorded inthe case of 3 wt.% clay. Also, the weight of residue at 600 ◦Cincreased, ranging from 1 to 21%, with clay loading from 0to 3 wt.% [193].

An optimum clay loading for thermal stability enhance-ment was also reported for PS-based nanocomposites.When dimethylbenzyloctadecyl-ammonium cation wasused for MMT modification, the threshold was reached at asurprisingly low MMT content of only 0.3 wt.% [35].

Berta et al. [144] did TGA under nitrogen and air forPU and the corresponding nanocomposite. Under nitrogen,the nanocomposite shows the same TGA and DTGA profilesas PU, but displaced by 10 ◦C, a change attributed to thebarrier effect. On the contrary, when tested under air, thenanocomposites showed very different behavior than purePU. For pure PU, most of the polymer volatilizes in the firststep of PU decomposition and the second step is diminishedin the TGA/DTGA curves in air compared to those obtainedin nitrogen; thus only a relatively small shoulder appearsat 350 ◦C in air. However, in the nanocomposites the lowertemperature degradation is significantly suppressed andthe second DTGA peak around 395 ◦C is clearly present. Thisbehavior was again attributed to greatly retarded thermaloxidation due to the shielding of the material from oxygen.

The thermal properties of EVA-based nanocompositeshave been widely studied, mainly by means of TGA. Ithas been well established by different research groupsthat EVAs exhibit two-step decomposition. The first step,which is identical in both oxidative and non-oxidativeconditions, occurs from 350 to 400 ◦C and correspondsto the deacetylation reaction, with production of gaseousacetic acid and formation of carbon–carbon double bondsalong the polymer backbone. The second step, between400 and 500 ◦C, involves thermal decomposition of theunsaturated backbone, either by further radical scission(non-oxidative decomposition) or by thermal combustion(oxidative decomposition) [35,81,167].

Riva et al. [234] observed accelerated acetic acid lossof EVA–clay nanocomposites and speculated that this pro-cess can be catalyzed by acidic sites of the nanodispersedclay. For the second step of EVA degradation, Maurin et al.[235] found as products 1-butene, carbon dioxide, ethylene,
methane and carbon monoxide, while McGrattan [236]identified hydrocarbons ranging from C8 to C26, groupedin a series of �,�-dienes, 1-alkenes and n-alkenes. By thor-oughly studying both the first and second steps, Costacheet al. [237] investigated the possibility that the presence of


FiE

noopGetiaablIteh

nalEplsaric

cptt

TMao

O

ig. 59. Possible radical recombination reactions for EVA nanocompos-tes [237]. Reproduced from Costache, Jiang and Wilkie by permission oflsevier Science Ltd., UK.

anodispersed clay can change the degradation pathwayf EVA. In the early stages of the EVA degradation, the lossf acetic acid seems to be catalyzed by the hydroxyl groupsresent on the edges of the clay lamellae. The TGA–FTIR andC–MS results of thermal degradation of EVA in the pres-nce and in the absence of clay show that even though thewo processes are very similar, subtle changes occur, lead-ng to the formation of products that differ both in quantitynd identity. It has been suggested that these products forms a result of radical recombination reactions that occurecause the degrading polymer is contained within the clay

ayers for sufficient time to permit the reactions (Fig. 59).n cases where there are multiple degradation pathways,he presence of the clay can promote one of these at thexpense of another, and thus lead to different products andence a different rate of volatilization.

In another study, the thermal decomposition of EVAanocomposites has been investigated using TGA in heliumnd in air. In helium, EVA nanocomposites exhibit a neg-igible reduction in thermal stability compared to virginVA or EVA microcomposite. In contrast, when decom-osed under air, the same nanocomposites exhibit a rather

arge increase in thermal stability, as the maximum of theecond degradation peak is shifted 40 ◦C to higher temper-ture, while the maximum of the first decomposition peakemains unchanged (see Table 35). The explanation for themproved thermal stability is the formation of protectivehar under oxidative conditions [238].

Finally, several studies have focused on the effect of
lay organic modification on thermal stability of nanocom-osites. More specifically, it has been suggested that thehermal stability of the nanocomposites is directly relatedo the stability of the OMLS used for their preparation [84].
able 35aximum temperature at the main degradation peak as measured under

ir at 20 ◦C/min for EVA and EVA-nanocomposites with different organ-clay contents.

rganoclay content (wt.%) Maximum temperature at themain degradation peak (◦C)

0 452.01 453.42.5 489.25 493.5

10 472.015 454.0

Fig. 60. TGA curves of (a) pure PU, (b) PU/3 wt.% C16-MMT and (c)PU/3 wt.% MO-MMT [202]. Reproduced from Xiong, Liu, Yang and Wangby permission of Elsevier Science Ltd., UK.

In a typical example, Xiong et al. [202] studied the ther-mal stability of PU and nanocomposites prepared usingMMT modified either by a quaternary alkyl ammonium salt(C16-MMT) or by an aromatic modifier (MO-MMT). The TGAcurves obtained (Fig. 60) show that the degradation ratesof the nanocomposites were slightly slower compared tothat of PU, indicating an improvement of thermal stability.Comparing the quaternary alkyl ammonium salt and thearomatic amine modifier they used, the authors found thedegradation onset temperature of PU/C16-MMT at 316.3 ◦C,thus lower than that of the PU/MO-MMT (331.6 ◦C), show-ing the latter to have the higher thermal stability. The mainreason is that the aromatic chain of the modifier in MO-MMT has higher thermal stability than the alkyl chain of themodifier in C16-MMT. The authors also noted that the aro-matic amine modifier can react with the pre-polyurethanematrix used, further strengthening the interaction betweeninorganic and organic phases.

In another interesting study, Yei et al. [115] used TGA tocharacterize PS-based nanocomposites prepared from claytreated with cetylpyridinium chloride (CPC) and CPC/�-cyclodextrin (CPC/�-CD) inclusion complex, as alreadydescribed in Section 5.3.1. Fig. 61 shows the TGA curvesof pure CPC and the CPC/�-CD inclusion complex. As isclearly evident, the CPC/�-CD inclusion complex decom-poses at higher temperature (284 ◦C) than the pure CPC(220 ◦C). Thus, the formation of an inclusion complexbetween CPD and �-CD improved the thermal stability ofthe CPC surfactant; the presence of the �-CD protects CPCfrom earlier decomposition. Table 36 summarizes the TGAresults for the nanocomposites and the pure polymer. Bothnanocomposites display higher decomposition tempera-tures than the virgin PS, with the CPC/�-CD intercalated
clay nanocomposite being the most thermally stable of thethree samples.
As deduced from the previous examples, even thoughcontradictory results are sometimes found in the lit-


Table 36Results of TGA and DSC data for the polystyrene nanocomposites.

Sample Clay content (wt.%) Tg (◦C)a T0.05 (◦C)b T0.5 (◦C)c Char at 600 ◦C (%)

PS 0 100 390 424 0CPC/Clay/PS 3 102 408 424 2.9a-CD/CPC/Clay/PS 3 106 423 452 5.8

a Glass transition temperature (Tg).b 5% degradation temperature (T0.05).c 50% degradation temperature (T0.5).

Fig. 61. TGA curves of (a) pure CPC and (b) the CPC/a-CD inclusion complex
[115]. Reproduced from Yei, Kuo, Fu and Chang by permission of ElsevierScience Ltd., UK.
Fig. 62. Schematic view of the cone calorimeter [35]. Reproduce

erature concerning the thermal stability of polymericnanocomposites, the opportunity of achieving a significantimprovement in thermal stability through low filler con-tent is particularly attractive because end-products can bemade cheaper, lighter and easier to process [35].

6.5. Flame retardance

6.5.1. Flame retardance of polymer–layered silicatenanocomposites

Polymers are being used in more and more applicationswhere flame retardant behavior is of critical importance.Traditionally, flame retardancy has been achieved either byusing intrinsically flame retardant polymers, such as fluo-ropolymers or PVC, or by incorporating flame retardants
(FRs), such as aluminum trihydrate, magnesium hydroxide,organic brominated compounds or intumescent systems.However, such FRs exhibit significant disadvantages. Forexample, aluminum trihydrate and magnesium hydroxideneed to be applied at very high loadings to be effec-
d from Beyer by permission of Elsevier Science Ltd., UK.


Table 37Cone calorimeter data of various polymers and nanocomposites with OMLS.

Sample PHRR (kW/m2) HRRave (kW/m2) SEAave (m2/kg) Ref.

PA6 1010 603 197 [240]PA6/2% OMLS 686 390 271PA6/5% OMLS 378 304 296

PS 1120 703 1460 [240]PS silicate mix 3% 1080 715 1840PS nanocomposite 3% 567 444 1730PSw/DBDPO/Sb2O3 30%) 491 318 2580

PS 1230 1315 [159]PS/MAST-Hect 1% 1011 1336PS/MAST-Hect 3% 894 1332PS/MAST-Hect 5% 728 1327

PPgMA 1525 536 704 [240]PPgMA nanocomposite 2% 450 322 1028PPgMA nanocomposite 4% 381 275 968

EVA/Na+ 5% 1200 [163]EVA/Cloisite 30B 3% 860EVA/Cloisite 30B 5% 780EVA/Cloisite 30B 10% 630

EVA 2303 430 [237]EVA/30B 1174 670EVA/30BHect 1289 593EVA/30BMag 2010 476EVA/MMT 1959 517

PU (I) 2561 741 176 [144]PU (I)/o-MMT 918 344 305

PU (II) 2254 637 235 [144]PU (II)/o-MMT 641 363 412

PU (III) 2647 768 165 [144]PU (III)/o-MMT 848 444 172

PU (IV) 2664 775 235 [144]PU (IV)/o-MMT 797 435 412

PE 1470 510 [184]PPPP

ttacctpta[

flbvrttsfmto

E/JS 2% 670 440E/JS 5% 320 450E/JS 10% 540 380E/JS 15% 390 280

ive, resulting in high density and lack of flexibility ofhe end-products, as well as low mechanical propertiesnd problems in compounding. On the other hand, con-erns over the environmental impact have made halogenontaining materials a less popular option in many coun-ries. Moreover, the addition of many FRs increases theroduction of soot and carbon monoxide during combus-ion. Finally, intumescent systems are relatively expensivend electrical requirements can restrict their application35,56].

Among test methods developed for the evaluation ofame retardant properties, the one most commonly usedy researchers is cone calorimetry (Fig. 62), as it providesaluable information and may even indicate the flameetardancy mechanism. The measuring principle in thisest, which is standardized as ASTM E 1354 and ISO 5660, ishat of oxygen consumption. This states that there is a con-
tant relationship between the mass of oxygen consumedrom the air and the amount of heat released during poly-
er combustion. In a typical cone calorimeter experimenthe sample is exposed to a defined heat flux, usually 35r 50 kW/m2; properties such as heat release rate (HRR),

peak of heat release (PHRR), time to ignition (TTI), totalheat released (THR), mass loss rate (MLR), mean CO yieldand mean specific extinction area can be simultaneouslymeasured [35,41]. The HRR is considered to be the mostimportant variable characterizing a fire. A high HRR causesfast ignition and flame spread, while the PHRR representsthe point in a fire where heat is likely to propagate further,or ignite adjacent objects [35,239].

As can be seen from Table 37, where cone calorime-ter data are presented for various nanocomposites andthe corresponding neat polymers, the incorporation of lay-ered silicates results in significant reductions of PHRR andaverage HRR. Moreover, it has been demonstrated that thereduction in PHRR is proportional to the fraction of clay,and it also depends on its aspect ratio and surface chargedensity.

In the literature it is reported that the primary param-
eter responsible for the lower HRR of nanocomposites isthe MLR during combustion, which, in turn, is also sig-nificantly reduced compared to the values observed forthe pure polymer. This difference comes into effect shortlyafter the initial combustion, when the nanocomposite has


nd EVA
Fig. 63. Residues after cone calorimeter experiment for EVA, EVA/30B-5 aand Gloaguen by permission of Elsevier Science Ltd., UK.
had time to form a char on its surface [56,177]. In fact, itis generally considered that the formation of a thermalinsulating and low permeability char on the outer sur-face of a nanocomposite during combustion is the key tounderstanding improvement in flame retardant properties[35,56]. More specifically, heat transfer from an externalsource or from a flame promotes thermal decomposition ofthe organoclay and the polymer. This results in accumula-tion and reassembly of clay platelets on the surface of theburning material [226]. Therefore, the carbonaceous charformed superficially during combustion is rich in silicatesand can be viewed as a sort of ceramic char–layered silicatenanocomposite [31,178,226]. In fact, XRD and TEM exami-nation of such residues has revealed intercalated structures[241]. Actually, it is exactly its own nanocomposite struc-ture that allows the residue formed to act as a protectivebarrier by reducing the heat and mass transfer between theflame and the polymer. That is, the char insulates the under-lying polymer from heat and also slows oxygen uptake andthe escape of volatile gases produced by polymer degra-dation. Both these actions interfere with the combustioncycle by reducing the amount of fuel available for burning[31,35,56,178,241,242].

This mechanism has been put forward in moststudies reporting on the flame retardant properties ofnanocomposites. For example, combustion experiments onEVA-based nanocomposites showed that the HRR and MLRwere reduced by 70–80% in a nanocomposite with low sil-icate loadings (2–5%) presumably because of a refractorychar-clay surface layer formed by the reassembling of theclay layers and catalyzed charring of the polymer [177].

Duquesne et al. [163] also reported on the fire retardancy
of EVA-based nanocomposites, using two montmoril-lonites: Cloisite Na+ and Cloisite 30B. While the PHRRwas clearly reduced when either clay was added to thepolymer (relative decreases of 25% for Cloisite Na+ and
Table 38Cone calorimetry results for PU nanocomposites and reference materials.

PU TTI (s) PHRR (kW/m2) PI (PHRR/TTI

PU NC-PU PU NC-PU PU

I 29 29 2561 918 88II 35 33 2254 641 64III 26 17 2647 848 102IV 22 25 2664 797 102

/Na+-5 [163]. Reproduced from Duquesne, Jama, Le Bras, Delobel, Recourt

50% for Cloisite 30B), the authors noted that the TTI wasalso reduced and that the THR was similar for pure poly-mer and the clay-containing polymer. Comparison of theresidues after the cone calorimeter experiment (Fig. 63)demonstrates the different behavior of the three materi-als. The pure polymer does not give a residue, whereas theEVA/Na+-5 gives a powdery grey “ash,” and the EVA/30B-5system gives a fragile, carbonaceous residue around 3 mmthick.

In the case of PE/clay nanocomposites significant reduc-tions of PHRR were reported, whereas the TTI results weresomewhat more complicated. For low clay contents, TTIwas increased, due to the barrier effect of the clay. How-ever, with increasing clay content a reduction of TTI wasattributed to catalytic effects. Again, a large reduction offlammability in nanocomposites compared to pure PE wasattributed to the formation of char [184].

Table 38 summarizes the results of the cone calorime-try tests performed on PU and PU-based nanocomposites.The data show that the average HRR and PHRR weregreatly reduced for the nanocomposites as compared withthe reference materials. However the TTI was similar orslightly reduced for the nanocomposites and the initialrate of weight loss/heat release was higher, probably dueto early decomposition of the organic modifier of theclay and subsequent clay-catalyzed polymer degradation.As shown by the “fire performance index” (PHRR/TTI),the nanocomposites perform much better than pure PUs.However, they showed higher rates of smoke emissionthan their pure PU counterparts, presumably because ofa greater degree of secondary condensed phase chemistryin an oxygen-depleted environment, leading to the forma-
tion and evolution of aromatic and carbonaceous species.Finally, Berta et al. [144] conducted vertical burn tests,where a flame was applied twice for 10 s to the base ofthe test specimens and the burning behavior was observed.
) HRRave (kW/m2) SEAave (m2/kg)

NC-PU PU NC-PU PU NC-PU

32 741 344 176 30519 637 363 235 41250 768 444 165 17250 775 435 235 412


Fig. 64. HRR measured with a cone calorimeter (heat flux 35 kW/m2) forvarious PP-based RTPO materials: (a) PP-based RTPO (neat resin), (b) PP-based RTPO/clay microcomposite, containing 10 wt.% clay, (c) PP-basedR[s

Pfp

antrtilmarrtcccdmo

pamd

aflctwgel3[

that the observed reduction in HRR (and MLR) tends to bedue to chemical and physical processes in the condensedphase, rather than in the gas phase. To prove this conjec-ture, the authors exposed samples to the same externalflux as in the cone calorimeter, but in nitrogen, to avoid

TPO/PPgMA/clay nanocomposite, containing 10 wt.% clay, 30 wt.% PPgMA185]. Reproduced from Hong, Lee, Bae, Jho, Nam, Nam and Lee by permis-ion of John Wiley & Sons, Inc.

ure PU materials dripped heavily upon ignition, whereasor the PU nanocomposites dripping was strongly sup-ressed or eliminated.

HRR plots of PP based RTPO neat resin, microcompositesnd nanocomposites are shown in Fig. 64. The PHRR of theanocomposite containing 10 wt.% clay is 37% lower thanhat of neat resin, while the microcomposite and the neatesin show very similar behavior. At the end of combustion,he neat resin leaves no residue and the microcompos-te leaves only a little powder, while the nanocompositeeaves a consistent char-like residue. Generally, PP ther-

ally degrades to volatile products above 250 ◦C throughradical chain process, propagated by carbon centered

adicals created by carbon–carbon bond scission. In neatesin, thermal degradation continues during the combus-ion, and all intermediate degradation products volatilizeompletely. On the other hand, there may be physicochemi-al adsorption of the intermediate degradation products onlay surfaces in clay nanocomposites. This adsorption mayelay volatilization of the degradation products and pro-ote accumulation of the incomplete degradation products

n the clay [185].Zanetti and Costa [177] also found that the nanodis-

ersed morphology in PE/EVA nanocomposites leads tosubstantial decrease of combustion rate of the poly-er matrix, whereas the microcomposite shows a smaller

ecrease of the rate of combustion.However, a quite interesting observation is that,

lthough the clay must be nanodispersed to affect theammability of nanocomposites, it does not need to beompletely delaminated. In other words, it has been arguedhat the flame retardancy obtained in nanocompositeshen merely intercalation has occurred is at least as

ood as when complete exfoliation is achieved. In fact,
xcellent performance has been observed when the clayayers have remained separated by only approximatelynm, which is considered to be in the “intercalation” zone
55,135,231,243].

Fig. 65. Effects of clay content on mass burning rate of PA6 (8 mm thick) at50 kW/m2 [242]. Reproduced from Kashiwagi, Harris, Zhang, Briber, Cipri-ano, Ragharan, Awad and Shields by permission of Elsevier Science Ltd.,UK.

At this point, it is worth summarizing a very inter-esting study by Kashiwagi et al. [242], aiming to furtherelucidate the nanocomposite flame retardancy mechanism.The authors performed cone calorimeter tests on PA6 andnanocomposites containing 2 and 5% clay. They noticed thatthe MLR (burning) curve (Fig. 65) of each sample is propor-tional to the HRR curve (Fig. 66). Thus, the specific heatof combustion (Hc), obtained from the HRR divided by theMLR remains unchanged for the three samples, implying

Fig. 66. Effects of clay content on HRR of PA6 (8 mm thick) at 50 kW/m2

[242]. Reproduced from Kashiwagi, Harris, Zhang, Briber, Cipriano, Ragha-ran, Awad and Shields by permission of Elsevier Science Ltd., UK.


/m2 [242
Fig. 67. Selected video images at 100, 200, and 400 s in nitrogen at 50 kWAwad and Shields by permission of Elsevier Science Ltd., UK.
any gas phase effects. Despite the quantitative difference inMLR between the two cases, the overall differences amongthe three samples were very similar between the burn-ing case and the gasification case. Given that the burningbehavior depends on processes in both the gas and the con-densed phase, while the gasification behavior depends onlyon those in the condensed phase, the authors suggestedthat the observed improvement in the flammability resis-tance of nanocomposites is due to chemical and physicalprocesses in the condensed phase. During gasification, thenanocomposites appeared to be more viscous than the neatPA6 sample and dark floccules appeared on their surfaces,grew with time (Fig. 67) and were left at the bottom ofthe container at the end of the test. Similar carbonaceousfloccules were also observed in the residues of the burntsamples tested in the cone calorimeter. Further analysisshowed that up to 80% of the protective floccules consistof clay particles, while the remaining 20% consist of ther-mally stable organic components with possible graphiticstructure. The authors attributed the accumulation of clayparticles on the burning/gasifying sample surface to twopossible mechanisms. One is the recession of the polymerresin from the surface by pyrolysis with dewetted clay par-ticles left behind, and the other is the transportation of clayparticles pushed by numerous rising bubbles of degrada-tion products.

The development of the aforementioned radiative gasi-fication technique (Fig. 68) is expected to increase theunderstanding of the condensed phase decomposition pro-cesses of pyrolysis, since the mechanism of nanocompositeflame retardancy can be looked at in more detail through

]. Reproduced from Kashiwagi, Harris, Zhang, Briber, Cipriano, Ragharan,

such techniques. With apparatus similar to the to conecalorimeter, this method allows pyrolysis in a nitrogenatmosphere at heat fluxes similar to those found in fires[35,56,242].

Summarizing: nanocomposites could offer significantadvantages in the area of polymers flame retardancyand these advantages become even more evident whennanocomposites are compared with conventional FRs. Firstof all, only very low concentrations of silicate are necessaryin nanocomposites, resulting in commercial advantagessuch as low density, lower cost and ease of preparation.Moreover, these materials are an environmentally friendlyalternative to some types of fire retardants, as they con-tain no halogens, phosphates or aromatics other than thosethat may be present in the polymer matrix; and theydo not produce increases in the carbon monoxide andsoot levels during combustion like those associated withconventional FRs. Also, while traditional fillers very oftenseverely degrade the physical properties of the polymeror discolor it, an important feature of nanocomposites isthe simultaneous improvement in many physical proper-ties, without any change in the polymer color [3,35,56,178].Finally, unlike some FRs, silicates provide physical integrityto the material burning in configurations (e.g. verticalupward combustion) in which dripping of flaming mate-rial could occur, which constitutes an additional hazard of
fire propagation to surrounding materials [231].
However, even though flame retardancy is one of themost promising properties of nanocomposites, and despitethe fact that these materials have shown to perform verywell in laboratory tests concerning their resistance to fire,


s [35]. R

mtbwe

6r

irHatmiowflmnsInttvth

TC

F

PPPP

Fig. 68. A schematic view of the “radiative gasification” apparatu

uch more research needs to be carried out to establishheir applications and limitations. The potential synergyetween traditional flame retardants and nanocomposites,hich is discussed in the following paragraph, could further

xtend the uses of nanocomposites in this field [56].

.5.2. Synergism between nanocomposites and flameetardants

From the point of view of fire retardancy, the most signif-cant advantage offered by nanocomposite formation is theeduction in PHRR (from cone calorimeter measurements).owever, compared to the virgin polymer, the TTI is usu-lly lowered while the THR remains the same. This meanshat the nanocomposite ignites faster than the virgin poly-

er and that, like the virgin polymer, the nanocomposites eventually completely combusted [239,244,245]. More-ver, a major limitation of nanocomposites is that they onlyork in the condensed phase and do nothing to inhibit theame in the gas phase [246]. Therefore, even though theseaterials can be considered to be flame retarded by defi-

ition, they perform poorly when tested under industriallyignificant or regulatory fire safety tests, such as UL-94V.n fact, it is believed that the excess quaternary ammo-ium surfactants used to disperse the clays, also increase
he probability of early ignition [239,246]. Considering alsohe disadvantages of traditional FRs, described in the pre-ious paragraph, it becomes obvious that it is necessaryo develop novel synergistic flame retardant systems withigh efficiency and acceptable environmental impact.
able 39one calorimeter data (35 kW/m2) for PP nanocomposites [247].

ormulation PHRR (

P-MA + 22 wt.% DBE + 5 wt.% Sb2O3 300, 65P-MA + 22 wt.% DBE + 5 wt.% Sb2O3 + 5 wt.% organoclay 200P-MA + organoclay 350P-MA 600

eproduced from Beyer by permission of Elsevier Science Ltd., UK.

In fact, as will be demonstrated in the followingexamples, polymer nanocomposites work in a synergisticmanner with other FRs. Since a portion of the FR is replacedby the nanocomposite in a less than 1:1 replacement ratio,products with improved fire performance and a better bal-ance of properties can be obtained [239].

Despite the existing regulatory concerns in the area ofhalogenated FRs, polymer nanocomposites combined withsuch additives are reported in the literature. For example,Zanetti et al. [247] described a PP-based nanocompositeincorporating decabromodiphenyl ether (DBE) and anti-mony oxide (AO) as the FR system. The cone calorimeterresults obtained are presented in Table 39. It is worth notic-ing that while the reference system (PP-MA + DBE + AO)showed two peaks in HRR (one at 90 s followed by a sec-ond larger one at 170 s), the addition of 5 wt.% organoclayeliminated the secondary peak, indicating a more uni-form flammability behavior. The average HRR was loweredby 58% and the PHRR by 70%. Also, the nanocompos-ite significantly increased the burn time for the samplewhen compared to the flame retardant control sam-ple.

Si et al. [246] demonstrated that self-extinguishingPMMA-based nanocomposites, which can pass the strin-
gent UL 94 V0 standard, can be successfully prepared bycombining organoclays with halogenated FRs. The authorsmixed PMMA, a highly combustible polymer, with DBE,AO and o-MMT. Both neat PMMA and the control flameretarded material failed the UL 94 V0 standard, while
kW/m2) PHRR, time (s) Average HRR

0 90, 170 25485 10785 188180 279


Fig. 69. Optical images of PMMA composites after the combustion of UL

pared, either by melt intercalation or intercalation from

94 V0 test: (a) PMMA/DB/AO (75/20/5) and (b) PMMA/DB/AO/Cloisite 20A(70/20/5/5) [246]. Reproduced from Si, Zaitsev, Goldman, Frenkel, Peiffer,Weil, Sokolov and Rafailovich by permission of Elsevier Science Ltd., UK.

the sample where both FR and clay were added self-extinguished in 1 s after the application of the flamewithout dripping. Optical images of the samples are shownin Fig. 69, where it can be seen that the PMMA sample con-taining both FR and clay is blackened by the smoke of theflame, but remains unchanged in shape. It was also noticedthat the samples with all three components, clay, DBE andAO show a lower PHRR and average MLR than those withonly clay or the FRs.

Chigwada et al. [244] conducted a study aiming toexplore if any improvements in the TTI and THR, whilemaintaining the reduction in PHRR, can be achievedby the incorporation of a small amount of brominechemically attached to the clay cation. The authors firstprepared an organically modified clay using ammoniumsalts with added oligomeric material consisting of vinyl-benzyl chloride, styrene and dibromostyrene. Then theyprepared intercalated nanocomposites by bulk polymer-ization. They found that the dibromostyrene enhances theflame retardancy of PS nanocomposites as compared withboth the virgin polymer and nanocomposites preparedfrom non-halogen containing, organically modified clays.More specifically, with the bromine-containing organoclay,nanocomposites were obtained exhibiting reduced PHRR,THR and improved thermal stability. The reduction in PHRRis due to nanocomposite formation and not to the presenceof bromine, which, however, is important in the reductionof the THR, i.e. the presence of bromine, even at less than
4%, prevents the PS burning.
Another interesting study was carried out by Tang etal. [248]. They did clay modification and intercalationsimultaneously during melt mixing. More specifically, they

lymer Science 33 (2008) 1119–1198

melt compounded PP, Na-MMT and hexadecyltrimethy-lammonium bromide to yield a nanocomposite, suggestingthat clay modification occurred in the molten polymermatrix. Therefore, the sodium bromide produced by theion-exchange reaction remained in the nanocomposite.Subsequently, the authors combined the nanocompos-ites obtained with an intumescent system consistingof pentaerythritol and ammonium polyphosphate, withmelamine phosphate sometimes added. Even though syn-ergism was observed between the nanocomposites and theintumescent FRs, the interpretation of the data is com-plicated by the presence of NaBr, which acts as a vaporphase FR. In fact, it remains unclear whether the improvedflammability properties reported are due to the nanocom-posite or to NaBr. In any case, as Morgan [239] noted inhis review on nanocomposite–FR synergism, this novelmethod of nanocomposite formation should be investi-gated further, especially if the NaBr can be removed, toproduce a truly non-halogenated flame retardant formu-lation.

Tang et al. [183] also observed that when MMT andintumescent FRs are combined in a polymer matrix, theirsynergistic effect is related to their composition ratio. Itwas anticipated that the reassembling of silicate layerson the surface of the nanocomposite during combustionmay hinder NH3 from swelling, leading to a negativeeffect on the fire retarding properties. NH3 is the maingaseous product; it volatilizes and makes the mixture ofthe carbonaceous residue and phosphocarbonaceous mate-rials swell, leading to the formation of the intumescentresidue char. Therefore, when the mass of MMT is increased,the negative effect may exceed the positive effect on thefire retarding property, which explains why the syner-gistic effect was no longer demonstrated at high MMTloadings.

In another study dealing with the synergism ofnanocomposites and intumescent systems, the polyol(carbonization agent) was replaced by PA6. More specif-ically, a PA6 nanocomposite was compounded with EVAand ammonium polyphosphate (acid source). The resultsshowed that the PHRR and THR were decreased by the PA6nanocomposite, while the mechanical performance wassignificantly improved [249].

Another important group of FRs which exhibit syner-gism with nanocomposites are the phosphorous-based FRs.For example, Chigwada et al. [245] observed reduction inthe PHRR, THR and MLR, when phosphorous-containingFRs were incorporated in vinylester nanocomposite. Theauthors reported that these reductions were directly pro-portional to the amount of phosphate added.

Zheng and Wilkie [250] investigated the synergisticaction of phosphates and nanocomposites by incorporat-ing the phosphate in the organic treatment of the clay.They treated Na-MMT with a styrenic oligomer that con-tained ammonium pendant groups and copolymerizedvinylphenyl phosphate. Then, PS nanocomposites were pre-

solution. TGA/FTIR showed that the phosphate is releasedduring thermal decomposition, and acts in the gas phase asan FR. The cone calorimeter data, on the other hand, showedthat in the presence of the phosphate, the average HRR and


Pi

emssppr

owowtestl

mflgfln

ciitb

6

admi

nia[

opdsinnida

bo1a

Table 40HDT of PP/MMT nanocomposites and the respective unfilled PP.

Organically modified MMT (wt.%) HDT

PP-f-MMT PP/alkyl-MMT

0 109 1093 144 130a

6 152 141b


HRR can be significantly reduced while the time to PHRRs increased up to ca. 40 s.

Quite similarly, Kim et al. [251] intercalated triph-nylphosphate (TPP) in the galleries of an organo-modifiedontmorillonite (Cloisite 30B). The clay obtained was sub-

equently melt mixed with ABS yielding nanocompositetructures. The key advantage of intercalating the phos-hate in the organoclay is that the normally volatile TPP isrotected during melt compounding, allowing for a graterange of melt processing without TPP loss.

On the other hand, Hussain et al. [252] introducedrganophosphinate into the backbone of an epoxy resin,hich was subsequently crosslinked in the presence of

-MMT. Quite surprisingly, flame retardant antagonismas observed with the PHRR and THR increasing for

he nanocomposite + FR system, as compared with thepoxy + FR control sample. No clear reason was given in thistudy; but in a later review article, Morgan [239] indicatedhat the observed antagonism could have been due to theack of uniform clay dispersion.

Another example where flame retardant antagonismay be observed is the case of FR systems that achieve

ame retardancy by dripping away from the flame. As Mor-an [239] noticed, since nanocomposites inhibit polymerow and dripping during flaming combustion, such FRs willot work with a polymer nanocomposite system.

In general, however, the synergism reported betweenlays and conventional FRs provides an alternative formproving fire performance without seriously deteriorat-ng appearance and mechanical behavior, and even openshe possibility of formulating self-extinguishing polymer-ased materials.

.6. Heat distortion temperature

Heat distortion temperature or heat deflection temper-ture (HDT) is the temperature at which a polymer sampleeforms under a specified load. Thus, it is an index of a poly-eric material’s heat resistance towards applied load and

s assessed by the procedure given in ASTM D-648 [55].In general, improvements of HDT are reported by

anocomposite formation. Usually, a significant increases achieved for clay contents of approximately 5 wt.%,nd then HDT values level off for higher clay loadings94,208,253].

For example, Kojima et al. [94] showed that the HDTf pure PA6 increases up to 90 ◦C after nanocompositereparation with OMLS, and they reported the clay contentependence of HDT of PA6/MMT nanocomposites. Morepecifically, they showed that there is a marked increasen HDT from 65 ◦C for neat PA6 to 152 ◦C for 4.7 wt.%anocomposite. Beyond that MMT loading, the HDT of theanocomposites levels off. Moreover, the HDT values of var-

ous PA6 nanocomposites prepared with clay lamellae ofifferent lengths showed that the HDT also depends on thespect ratio of dispersed clay particles.

Similarly, nanodispersion of MMT in a PP matrix haseen found to promote a higher HDT (Table 40). In the casef PP/f-MMT there is a marked increase of the HDT, from09 ◦C for the neat polymer to 152 ◦C for a 6 wt.% of clay,fter which the HDT of the nanocomposites levels off. When

9 153

a C18-MMT filler, extruder processed.b 2C18-MMT filler, twin head mixer.

the same neat PP polymer is filled with alkylammoniummodified MMT, the HDT is also increased but to a smallerextent, reflecting the lower exfoliation level of the inorganicfillers [254].

It should be emphasized that the increase of HDT dueto clay dispersion is very important, not only from applica-tion or industrial point of view, but also because it is verydifficult to achieve similar HDT enhancements by chem-ical modification or reinforcement by conventional fillers[31,50,55].

6.7. Rheological properties

The measurement of the rheological properties of anypolymeric material is crucial to gain fundamental under-standing of the processibility of that material and isusually conducted by either dynamic oscillatory shearor steady shear measurements [50,55]. In the case ofpolymer–layered silicate nanocomposites, the study of rhe-ological properties is instructive for two reasons: First,these properties are indicative of melt processing behaviorin unit operations, such as injection molding. Second, sincethe rheological properties of particle-filled materials aresensitive to the structure, particle size, shape and surfacecharacteristics of the dispersed phase, rheology potentiallyoffers a means to assess the state of dispersion in nanocom-posites, directly in the melt state. Thus, rheology can beenvisaged as a tool that is complementary to traditionalmethods of materials characterization [15,37,50].

It is generally established that when nanocompositesare formed, the viscosity at low shear rates increases withfiller concentration [15]. Very often, solid-like behavior isobserved, which is attributed to the physical jamming orpercolation of the randomly distributed silicate layers, atsurprisingly low volume fraction, due to their anisotropy[55]. On the other hand, at high shear rates, shear thinningbehavior is usually observed [15]. It has been suggested thatthis is the result of the alignment of silicate layers towardsthe direction of flow at high shear rates. Such observa-tions support the percolation argument used in the case ofnanocomposite rheological behavior under low shear [55].Typical shear viscosity curves as a function of shear arepresented in Fig. 70 for PBS-based nanocomposites withvarious clay loadings.

Shear thinning behavior at high shear rates has also beenobserved for a PA6 nanocomposite. In fact, this behavioris similar to that of neat PA6 and the composites with thesame amount of glass fiber or unmodified MMT. It is of greatinterest to note that the absolute value of the melt viscos-


Fig. 71. DSC thermograms of PCL and of selected PCL/clay nanocompos-

containing 10 wt.% clay. Gelfer et al. considered this to be ananticipated result, since it is known that in ethylene-basedcopolymers side chains (short branches) larger than themethyl group are excluded from the crystalline domain andare aggregated into the amorphous phase. Consequently,

Fig. 70. Shear viscosity as a function of shear rates [55]. Reproduced fromRay and Okamoto by permission of Elsevier Science Ltd., UK.

ity of the nanocomposite is significantly lower than that ofneat nylon 6 or the other composites—which implies goodmelt processibility over a wide range of processing condi-tions. One possible reason for reduction of melt viscosityin the nanocomposite is slip between the PA6 matrix andthe exfoliated organoclay platelets during high shear flow.Another possibility is a reduced molecular weight of thenylon 6 due to degradation (e.g. hydrolysis) in the presenceof clay [15]. Similarly, capillary data revealed that nanocom-posites based on medium and low molecular weight PA6exhibited lower viscosities than the corresponding purepolymers [14].

6.8. Crystallinity

Crystal formation includes two stages, namely nucle-ation and crystal growth. However, although it is wellestablished that nanometer sized clay platelets are effec-tive nucleating agents, different effects have been reportedon the linear growth rate and the overall crystallization rate,depending on the type of polymer [97].

For example, Maiti et al. [175] found that although clayparticles act as nucleating agents for the crystallization ofa PP-MA matrix, the linear growth rate and the overallcrystallization rate are not significantly influenced by thepresence of clay.

Similarly, Di Maio et al. [255] studied the isothermalcrystallization of PCL/clay nanocomposites and noticed thatthe dispersed clay platelets act as nucleating agents in thePCL matrix, remarkably reducing the crystallization half-time t1/2. By DSC analysis after isothermal crystallization(Fig. 71), the authors observed a reduction of the meltingtemperature with the increase of clay content, indicat-ing a reduced degree of crystals perfection and degreeof crystallinity. This was attributed to the confinement ofchains and segments in the presence of clay, hinderingthe segmental rearrangement during crystallization andrestricting the formation of perfect crystals in the polymer
matrix.
Ke and Yongping [222] conducted DSC analysis onintercalated PET/o-MMT nanocomposites. They found areduction of Tg in the composite compared to the purematrix, which they attributed to the plasticizing effect of o-

ites after isothermal crystallization at 45 ◦C. Number refers to clay weightconcentration; 0* is the DSC thermogram of non-isothermal crystallized(−10 ◦C/min) pure PCL [255]. Reproduced from Di Maio, Iannace, Sorren-tini and Nicolais by permission of Elsevier Science Ltd., UK.

MMT. However, they noticed that by increasing the o-MMTcontent the Tg is increased, as shown in Fig. 72. Further,they observed that the cold crystallizing point of pure PET is150 ◦C, while for the nanocomposite it decreases to 130 ◦C.This result shows that adding o-MMT into PET is favorableto its crystallization.

For the effect of nanocomposite preparation on EVAcrystallinity, contradictory results have been reported. Forexample, Chaudhary et al. [171] found that the crystal-lization process for EVA-28 is significantly affected by thepresence of clays, while Gelfer et al. [256] argued thatthe effect of clay on the crystalline phase in EVA-3 isinsignificant. They observed the same crystalline structure(orthorhombic) and similar melting behavior, as well astransition temperatures in pure EVA-3 and nanocomposites

Fig. 72. The relationship between o-MMT content and Tg of composite[222]. Reproduced from Ke and Yongping by permission of Elsevier ScienceLtd., UK.


tVdptot

cdpfcwptpb

tmpspF

ntem�dtclatiooai

FW


he amorphous domains in EVA are enriched by the polarA units. It was hypothesized that the organoclay particlesispersed in the EVA matrix have higher affinity with theolar VA units; hence they are predominantly confined tohe amorphous phase where the VA content is high, with-ut significantly affecting the crystalline domain formed byhe non-polar ethylene segments.

While further work needs to be done to elucidaterystallization phenomena in nanocomposites based onifferent polymers, the effect of layered silicates onolyamide crystallinity has been far more investigated. In

act, a number of researchers have focused on the effect oflay addition on the crystal morphology of polyamides, asell as on the melting, glass transition, crystallization tem-erature and crystallization rate. In semi-crystalline PA6,wo phases generally exist at room temperature. The �-hase and the �-phase are monoclinic, but the latter cane represented as a pseudo-orthotropic lattice.

In general, it has been observed that upon prepara-ion of nanocomposites the crystallography of the PA6

atrix changes. Maiti and Okamoto [257] found that in theresence of clay particles, PA6 crystallizes faster and exclu-ively in the �-phase (Fig. 73). The nucleation and growthrocesses of PA6 in the nanocomposites are presented inig. 74, as deduced from direct observation by TEM.

Similarly, Lincoln et al. [12] showed through simulta-eous SAXS and WAXS analyses at elevated temperaturehat the presence of silicate layers provides a confinednvironment for crystallization and disrupts crystallite for-ation, ultimately resulting in the predominance of the

-phase in quenched samples. They noted that the moreuctile �-phase that is formed may contribute greatly tohe observed changes in properties of the material, specifi-ally the toughness. Evidence has also been found that theayers affect not only the formation of the lamellae, butlso of spherulites, which may also be the root of some ofhe property enhancements. Moreover, certain processes
nvolved in nanocomposite preparation affect the globalrientation of the crystal phase. That is, X-ray diffractionf PA/MMT nanocomposites has indicated that the chainxes are normal to the silicate layers in the interior of annjection molded component, while they are parallel to the
Fig. 73. Typical TEM images of PA6CN3.7 crystallized at (a) 170 and 210 ◦C.The enlarged part shown (a) is form the indicated lamella in the originalimage. The black strip inside the white part is clay. (b) Shows the typicalshish-kebab type of structure [257]. Reproduced from Maiti and Okamotoby permission of Wiley-VCH, Germany.

ig. 74. Schematic view of the nucleation and growth mechanism in nylon-6 nanocomposite [257]. Reproduced from Maiti and Okamoto by permission ofiley-VCH, Germany.


Fig. 75. The heat flow vs. the time during isothermal crystallization ofnylon 1012 (spectrum a) and nylon 1012/clay nanocomposite (spectrum

isms into the bulk and hindered their diffusion. Similarly,Maiti et al. [195] attributed the suppressed biodegradation

b) by DSC [97]. Reproduced from Wu, Zhou, Qi and Zhang by permissionof John Wiley & Sons, Inc.

layers in the near surface region of molded and extrudedcomponents.

Varlot et al. [5] suggested that the overall crystallinity ofthe polyamide matrix remained unchanged with the addi-tion of montmorillonite, even though the proportion of the�-phase was higher.

Concerning the glass transition, a decrease in Tg

has been found for PA6/clay nanocomposites; this wasattributed to an increase in the amorphous volume frac-tion due to the surface/volume ratio of the crystallites, thenumber of nuclei and the perfection of the crystal lamel-lae [139]. However, DSC measurements conducted by otherresearchers, indicate that the presence of layered silicatefiller does not affect the Tg of the PA6 matrix, which occursat approximately 53 ◦C. The melting peak of the compos-ites is at a slightly lower temperature than that of neat PA6,due to a slight reduction of crystallite size in the presence offillers. All the fillers cause an increase in crystallization tem-perature relative to neat PA6. To various degrees, fillers mayact as nucleating agents, causing a higher crystallizationrate than that of the neat PA6 [15]. The same phenomenahave also been observed in PA66 based nanocomposites[146].

Finally, Wu et al. [97] reported DSC curves for isothermalcrystallization of PA1012 and PA1012/clay nanocompos-ites (Fig. 75). The crystallization exothermic peak ofthe PA1012/clay nanocomposites is narrower than thatof PA1012, indicating that the nanocomposites have ahigher crystallization rate. The heat of crystallizationof the nanocomposite is less than that of the PA1012,implying that the clay platelets decrease the polymer crys-tallinity. The higher crystallization rate recorded for PA1012nanocomposites was attributed to the nucleating action ofclay platelets. On the other hand, the authors suggested thatthe confinement of the chains and segment movement will
hinder segmental rearrangement during crystallization andrestrict the formation of perfect crystals in the PA1012/claynanocomposite, thus resulting in decreased crystallinity.
lymer Science 33 (2008) 1119–1198

6.9. Biodegradation

An interesting aspect of nanocomposite technology isthe enhancement in biodegradability, often reported afternanocomposite formation. Tetto et al. [258] first presentedresults on the biodegradability of nanocomposites based onPCL, reporting that the PCL/OMLS nanocomposites showedimproved biodegradability compared with pure PCL.

Biodegradability improvements have also been reportedin the case of PLA-based nanocomposites. As an example,Ray et al. [208] presented photographs of the recoveredsamples of neat PLA and PLACN4 from compost with time(Fig. 76(a)). The decreased Mw and residual weight Rw per-centage of the initial test samples with time is also shownin Fig. 76(b). Obviously, the biodegradability of neat PLAis significantly enhanced after nanocomposite preparation.For a month, both the extent of Mw loss and weight lossare almost the same for neat PLA and PLACN4.5 However,after 1 month, a sharp change occurs in weight loss ofPLACN4, and within 2 months, it is completely degraded bycompost. The degradation of PLA in compost is a complexprocess involving: water absorption, ester cleavage and for-mation of oligomer fragments, solubilization of oligomerfragments, and finally diffusion of soluble oligomers. There-fore, any factor that increases the hydrolysis tendency ofneat PLA, ultimately controls the degradation. However,from Fig. 76(b) it is evident that the hydrolysis tendenciesof PLA and PLACN4 are almost the same.

Ray et al. suggest that the presence of terminal hydrox-ylated edge groups of the silicate layers may be one of thefactors responsible for this behavior. In PLACN4, the stackedand intercalated silicate layers are homogeneously dis-persed in the PLA matrix, and these hydroxy groups initiateheterogeneous hydrolysis of the PLA matrix after absorbingwater from the compost. This process takes some time tostart. For this reason, according to these authors, the weightloss and degree of hydrolysis of PLA and PLACN4 is almostsame up to 1 month (Fig. 76(b)). However, after 1 monththere is a sharp weight loss in the case of PLACN4 as com-pared with that of PLA. This means that 1 month is a criticaltime to start heterogeneous hydrolysis, and due to this typeof hydrolysis the matrix degrades into very small fragmentsand is eliminated with the compost. This assumption wasconfirmed by conducting the same type of experiment withPLACN prepared by using dimethyl dioctadecylammoniumsalt modified synthetic mica which has no terminal hydrox-ylated edge group; in that case the degradation tendencywas almost the same as that of PLA.

However, contradictory results concerning the effectof clay dispersion on polymer biodegradability are alsofound in the literature. In fact, Lee et al. [259] preparednanocomposites based on aliphatic unsaturated polyesterand found a decrease in biodegradability under compost-ing with intercalation. They assumed that due to the highaspect ratio and better dispersion of clay in the matrix amore tortuous path formed for penetration of microorgan-

tendency, which they observed in the case of PHB nanocom-posites, to an improvement of the barrier properties ofthe matrix after nanocomposite formation, even though


F covered3 of matriY

tcri

bnno

6

ittr

optm

ig. 76. (a) Real pictures of biodegradability of neat PLA and PLACN4 recm × 10 cm × 0.1 cm. (b) Time dependence of residual weight, Rw andamada, Okamoto and Ueda by permission of Elsevier Science Ltd., UK.

hey did not report on the permeability. This explanationontradicts the results of Ray et al. [260], who found noelation between biodegradability and barrier propertiesn PLA/OMLS nanocomposites.

From the aforementioned contradictory results, itecomes obvious that the increase or the decrease inanocomposite biodegradation is still under discussion ando conclusion can be driven about mechanisms on the basisf the current literature [231].

.10. Photo-degradation

The degradability of nanocomposites under UV lights a serious problem that may limit their applications. Inhe few studies addressing this issue, it has been foundhat nanocomposites exhibit lower stability than the cor-esponding neat polymers [231].

For example, Huaili et al. [261] studied the photo-xidative degradation of PE/MMT nanocomposites com-ared with neat polyethylene. Since it is well establishedhat the degradation of hydrocarbon chains leads to the for-

ation of hydroxyl and keto groups, they studied the extent

from compost with time. Initial shape of the crystallized samples wasx, Mw of PLA and PLACN4 under compost [208]. Reproduced from Ray,

of photo-degradation by FTIR observations. As is shown inFig. 77, the degradation of PE/o-MMT nanocomposite wasgreater than that of pure PE polymer after 200 h irradia-tion. Fig. 78 shows FTIR spectra in the carbonyl region uponUV irradiation. Obviously, there is a considerable increasein the intensity of the carbonyl region with an increase ofirradiation time in PE/o-MMT, which means that the mate-rial is undergoing degradation. In the pure PE, on the otherhand, the intensity in the carbonyl region was significantlyless, which indicates less degradation.

Morlat [262] and Mailhot et al. [263] studied theeffect of compatibilizers on photo-degradation and itskinetics by comparing PP nanocomposites with neat poly-mer. The increase in the absorbance at 3200–3600 and1600–1800 cm−1 was rapid in nanocomposites in compar-ison with neat polymer. It was observed that the inductionperiod decreased from 8 to 4 h by using PP-g-MA as com-
patibilizer and a two-phase degradation mechanism wasobserved. In the first stage (up to 40 h) there was no evi-dence for hydroxyl band formation in the IR spectra, whichimplies the absence of degradation on polymer backbone,whereas in the second stage a dramatic increase in the rate


Fig. 77. (a) FTIR spectra of PE/o-MMT nanocomposites before and after

nanocomposites are far lighter in weight than a conven-

200 h irradiation. (b) FTIR spectra of pure PE before and after 200 h UVirradiation [261]. Reproduced from Huaili, Chungui, Shimin, Guangmingand Mingshu by permission of Elsevier Science Ltd., UK.

of photo-oxidation was found. The degradation productswere the same in the composite and the neat polymer.

According to Patterson et al. [264], the incorporationof layered silicates in polycarbonate appeared to increasethe rate at which chain scission occurs. Furthermore, thesecarbonate scissions produced a yellowing of the polycar-bonate, which could inhibit its use in applications whereoptical clarity is important.

However, in another study, the effect of acceleratedweathering of polycarbonate nanocomposites was inves-tigated. The silicate content used ranged from 0 to 3.5 wt.%.A UV-accelerated weathering tester programmed to cyclefor 8 h of UV radiation and 4 h of dark condensation was
selected for the exposure study. The materials were charac-terized by UV/vis spectroscopy and FTIR spectroscopy and itwas concluded that the degradation of the nanocompositewas less than that of the neat polymer [265].
Fig. 78. FTIR spectra of PE/nanocomposite at carbonyl region duringphoto-degradation [261]. Reproduced from Huaili, Chungui, Shimin,Guangming and Mingshu by permission of Elsevier Science Ltd., UK.

In any case, although no explanation has so far beengiven for the differences in photo-degradation stabilitybetween the nanocomposites and the pure polymers, ithas been suggested that the best way to increase the out-door durability would be to develop nanocomposites bymodification of the clay rather than functionalization ofthermoplastics [231].

6.11. Optical clarity

Microsized particles used as reinforcing agents scatterlight, thus reducing light transmittance and optical clarity[266]. On the other hand, layered silicate platelets, albeittheir micron lateral size, are just 1 nm thick. Thus, whensingle layers are dispersed in a polymer matrix, the result-ing nanocomposite is optically clear in the visible region,whereas, there is a loss of intensity in the UV region (for� < 250 nm) mostly due to scattering by the MMT parti-cles. There is no marked decrease in the clarity due to thenanodispersed fillers (e.g. one has to load 20 wt.% of C18-MMT in 3 mm thick film of PP before there develops hazeobservable by the eye). This is a general behavior in UV/vistransmittance for thick films (3–5 mm) of polymer/MMTnanocomposites based on PVA, PP, and several epoxies [31].

7. Nanocomposites: advantages and applications

As described above, polymer nanocomposite materialsoften exhibit properties superior to conventional com-posites, such as strength, stiffness, thermal and oxidativestability, barrier properties, as well as flame retardantbehavior [267]. These improved properties are generallyattained at lower filler content in comparison with conven-tionally filled systems. Therefore, polymer–layered silicate

tional composite, which makes them quite competitive forspecific applications [55].

Moreover, for systems with favorable thermodynamicsof mixing, the organoclay can be incorporated in the final


Table 41Selected commercial nanoclays.

Product Characteristics Applications Producer

Cloisite Organophilic Additives to enhance flexural and tensilemodulus, barrier properties and flameretardancy of thermoplastics

Southern Clay Products

Nanomers Microfine powder Nylon, epoxy, unsaturated polyester,engineering resins

Nanocor

Masterbatches Pellet Thermoplastic olefin and urethane,styrene-ethylene butylene-styrene, ethylenevinyl acetate

PolyOne Corporation, ClariantCorporation, RTP Company

Bentone With a broad range of polarity Additives to enhance mechanical, flameretardancy and barrier properties ofthermosets and thermoplastics

Elementis Specialties

Nanofil Improve the mechanical, thermal andbarrier properties

Thermoplastics and thermosets Sud-Chemie

Planomers Additive, enhance mechanical barrierproperties, thermal stability and flameresistance

Electric and electronic, medical and healthcare, adhesive, building and constructionmaterials

TNO

P rative col free co

P parent pgs, tran

scTtpc

ofmHrttp

pa

TS

P

N

P

M

D

A

A

S

F

lanoColors Nanopigments, e.g., blue, red, green,yellow, high UV-stability

Decometa

lanoCoatings Additive, excellent transparency andimproved barrier properties

Transcoatin

tages of polymer processing (e.g. extrusion, injection orompression molding) to obtain nanocomposite materials.hus, polymer nanocomposites are amenable to most ofhe common processing techniques in today’s industrialractice—which will lower the barriers to their commer-ialization [31].

Another unique aspect of nanocomposites is the lackf property trade-offs. Traditionally, blend or compositeormulations require trade-offs between desired perfor-

ance, mechanical properties, cost and processibility.owever, polymer nanocomposite technology provides a

oute around these traditional limitations, and offers, forhe first time, the opportunity to design materials without
he compromises typically found in conventionally filledolymers [267,268].
The aforementioned attractive characteristics ofolymer–layered silicate nanocomposites already suggest

variety of possible industrial applications: automo-

able 42elected commercial polymer nanocomposites.

roduct Characteristics Applicatio

ylon nanocomposites Improved modulus, strength, heatdistort temperature, barrier properties

Automotivcover, engpackagingmedical, e

olyolefinnanocomposites

Stiffer, stronger, less brittle, lighter,more easily recycled, improved flameretardancy

Step-assischevroletelectrical e

9TM High barrier properties Juice or becontainers

urethan KU2-2601(nylon 6)

Doubling of stiffness, high gloss andclarity, reduced oxygen transmissionrate, improved barrier properties

Barrier film

egisTM NC (nylon6/barrier nylon)

Doubling of stiffness, higher heatdistort temperature, improved clarity

Medium b

egisTM OX Highly reduced oxygen transmissionrate, improved clarity

High barri

ETTM nanocompositenylon 12

Improved stiffness, permeability, fireretardancy, transparency and recycling

Catheter sfilm and b

orteTM nanocomposite Improved temperature resistance andstiffness, very good impact properties

Automotiv

loring, UV-stable coloring, heavyloring

TNO

ackaging materials, protectivesparent barrier coatings

TNO

tive (gas tanks, bumpers, interior and exterior panels),construction (building sections and structural panels),aerospace (flame retardant panels and high performancecomponents), food packaging, textiles, etc. [269].

It is for this reason that many companies have taken astrong interest and have invested in developing nanoclays(Table 41) and polymer nanocomposites (Table 42) [270].

Among these, the first commercial product of clay-basedpolymer nanocomposites was the timing-belt cover madefrom PA6 nanocomposites by Toyota Motors in the early1990s. This timing-belt cover exhibited good rigidity, excel-lent thermal stability and no wrap. It also saved weight byup to 25% [270]. Later, General Motors and partners Basell,
Southern Clay Products and Blackhawk Automotive Plasticsannounced external automotive body parts (step-assist)made from thermoplastic, olefin layered silicate nanocom-posites. A TPO nanocomposite with as little as 2.5% layeredsilicate is as stiff and much lighter than parts with 10 times
ns Producer

e parts (e.g., timing-beltine cover, barrier fuel line),(e.g., cosmetics, food,lectronics), barrier film

Bayer, Honeywell Polymer, RTPCompany, Toyota Motors, Ube, Unitika

t for GMC Safari andAstro vans, heavy-dutynclosure

Basell, Blackhawk Automotive Plastics,General Motors, Gitto GlobalCorporation, Southern Clay Products

er bottles, multi-layer films, Mitsubishi Gas Chemical Company

s, paper coating Bayer

arrier bottles and films Honeywell Polymer

er beer bottles Honeywell Polymer

hafts and balloons, tubing,arriers, flexible devices

Foster Corporation

e, furniture, appliance Noble Polymer

ess in Po
the amount of conventional talc filler. Thus, the weight sav-ings can reach 20%, depending on the part and the materialthat is being replaced by the TPO nanocomposite [268].William Windscheif, Basell’s Global Business Vice Presidentfor Advanced Polyolefins, called this application “a smallstep, but a giant one for nanocomposites” and added that itheralds a broader shift to nano-PP in automotive technol-ogy [271].

Consultant Kenneth Sinclair, head of STA Research inSndiomish, Wash., estimated that in the near future theapplication of PP-based nanocomposites in the automotiveindustry will expand, with nano-PP mostly cannibilizingexisting PP applications. However, replacement of metalsand engineering thermoplastics will follow. In this context,a long term goal for Dow Plastics is in-reactor compound-ing of nano-PP by using nano-clays as the catalyst supportfor in-situ polymerization of PP homopolymer. Dow’s effortis focused on highly loaded (up to 10% clay) nano-PPs forsemi-structural automotive uses [271].

Meanwhile, a role for nanocomposites in polycarbonateautomotive glazing is being explored by Exatec of Wixon,Mich., a joint venture of Bayer and GE Plastics, that isdedicated to PC auto-glazing development, considered forthe exterior coating needed to achieve weatherability andabrasion resistance without reducing clarity. A Bayer coat-ing containing nanoparticles is one of several promisingapproaches being pursued [271].

It is worth noticing that the weight advantage of poly-mer nanocomposites could have a significant impact onenvironmental protection and material recycling. It is pre-dicted that widespread use of polymer nanocompositeswould save 1.5 billion liters of gasoline over the life of 1year’s production of vehicles and reduce related CO2 emis-sions by more than 5 billion kilograms [270].

Next to the automotive industry, polymer nanocom-
posites are expected to find wide applications as barriermaterials. In fact, the excellent barrier properties ofclay-based polymer nanocomposites could result in con-siderable enhancement of shelf-life for many types of
Fig. 79. Possible growth paths for na

lymer Science 33 (2008) 1119–1198

packaged food. Meanwhile, the optical transparency ofpolymer nanocomposite film is generally similar to theirpristine counterparts, which is impossible with conven-tional polymer composites. Therefore, the above propertyadvantages would make them widely acceptable in packag-ing industries as wrapping films and beverage containers.For example, Bayer has developed a new grade of plasticfilms for food packaging, which are made from PA6 exfoli-ated nanocomposites [270]. Honeywell is also developingnano-PA materials that can beat the cost of high-barrierplastics or even glass. Nanocor’s Imperm compound sup-plements the inherent gas barrier of amorphous MDX6nylon from Mitsubishi Gas Chemical with the addition ofnanoclay. Used as the core of a three-layer PET bottle,Imperm reportedly ensured a 28.5-week beer shelf-life.Imperm is said to adhere to PET without tie layers, whilesufficient clarity is retained to meet requirements for theamber bottle [271].

Meanwhile, Ube America is developing nanocompositebarriers for automotive fuel systems, using up to 5% nan-oclay in PA6 and PA6/66 blends [271].

Also quite interesting are potential applications ofnanocomposites based on biodegradable polymers. Suchpolymers have become indispensable in a wide rangeof applications. However, despite their attractive degra-dation characteristics and significant demand for suchmaterials, the lack of structural and functional stabilityprevents currently available biodegradable polymers fromhaving widespread commercial impact [267]. Therefore,biodegradable polymer-based nanocomposites appear tohave a very bright future for a wide range of applicationsas high-performance biodegradable materials [55].

Layered silicate nanoparticles, when distributed withinthe matrix of a fiber-reinforced polymer (FRP), can retardthe diffusion of environmental moisture and other chemi-
cals to the fiber–matrix interface where their presence canresult in delamination and fiber weakening. Thus, the use ofnanoparticles helps to preserve the integrity of FRPs and toprolong the service life of composites when these are used
nocomposites applications.

ess in Po

i[

nfp[

ipficrpecb

tmlla

8

ktporwismcwwetfiIwmrnrotcsswfi

R


n outdoor applications such as bridges and utility poles272].

Other potential nanocomposite applications includeano-pigments that are believed to be an environment-

riendly alternative to toxic cadmium and palladiumigments, as well as controlled drug delivery systems, etc.270].

However, despite the current optimism surround-ng polymer–nanocomposite materials, the mechanicalerformance benchmarks set by highly loaded short-ber composites are still beyond the capabilities of lowoncentration nanocomposites. Manufacturing costs alsoemain a significant factor restricting the growth ofolymer–nanocomposite applications [200]. In fact, somearly application development programs have lapsed forost reasons. Such casualties include the timing-belt coverased on nylon 6 nanocomposite [271].

According to Silberglitt, there are two possible paths orrends—a high-growth path, under which nanocomposite

aterials are pervasively applied throughout society and aow-growth path, under which the use of nanocompositeseads to incremental improvements in specific technologyreas (Fig. 79) [269].

. Summary

Polymer–layered silicate nanocomposites, althoughnown for many years, have attracted recent attention dueo the report of the Toyota research group on the improvedroperties of PA6 nanocomposites and also due to thebservation by Giannelis and co-workers that their prepa-ation is possible by simple melt-mixing of the polymerith the layered silicate. Other preparation routes include

ntercalation of polymer or prepolymer from solution, initu intercalative polymerization and template synthesis. Inost cases, layered silicates first need to be modified with

ationic-organic surfactants, in order to become miscibleith polymeric matrices. Then, whether a nanocompositeill form or not, and whether this will be intercalated or

xfoliated, depends on a variety of factors. These includehe type of polymer, layered silicate and organic modi-er, the preparation technique and processing conditions.

n general, nanocomposite materials, particularly thoseith exfoliated structures present significant improve-ents of modulus and strength, whereas contradictory

esults are reported concerning their elongation and tough-ess. Improvements of storage and loss moduli are alsoeported by many authors. Other interesting characteristicsf this class of materials include improved barrier proper-ies, thermal stability and flame retardance. Despite someontradictory results reported in the literature and pre-ented here, concerning certain aspects of polymer–layeredilicate nanocomposite technology, we hope this reviewill be a useful tool for those conducting research in thiseld.

eferences

[1] Alexandre M, Dubois P. Polymer–layered silicate nanocomposites:preparation, properties and uses of a new class of materials. MaterSci Eng R 2000;28:1–63.

lymer Science 33 (2008) 1119–1198 1193

[2] Fischer H. Polymer nanocomposites: from fundamental research tospecific applications. Mater Sci Eng C 2003;23:763–72.

[3] Lagaly G. Introduction: from clay mineral-polymer interactions toclay mineral-polymer nanocomposites. Appl Clay Sci 1999;15:1–9.

[4] Giannelis EP. Polymer layered silicate nanocomposites. Adv Mater1996;8:29–35.

[5] Varlot K, Reynaud E, Kloppfer MH, Vigier G, Varlet J. Clay–reinforcedpolyamide: preferential orientation of the montmorillonite sheetsand the polyamide crystalline lamellae. J Polym Sci Polym Phys2001;39:1360–70.

[6] Gorrasi G, Tortora M, Vittoria V, Galli G, Chiellini E. Transportand mechanical properties of blends of poly(�-caprolactone) and amodified montmorillonite–poly(�-caprolactone) nanocomposite. JPolym Sci Polym Phys 2002;40:1118–24.

[7] LeBaron PC, Wang Z, Pinnavaia TJ. Polymer–layered silicatenanocomposites: an overview. Appl Clay Sci 1999;15:11–29.

[8] Vaia RA, Wagner HD. Framework for nanocomposites. Mater Today2004;7:32–7.

[9] Ginzburg VV, Singh C, Balazs AC. Theoretical phase diagrams ofpolymer/clay composites: the role of grafted organic modifiers.Macromolecules 2000;33:1089–99.

[10] Osman MA, Mittal V, Lusti HR. The aspect ratio and gas perme-ation in polymer–layered silicate nanocomposites. Macromol RapidCommun 2004;25:1145–9.

[11] Balazs AC, Singh C, Zhulina E, Lyatskaya Y. Modeling thephase behavior of polymer/clay nanocomposites. Acc Chem Res1999;32:651–7.

[12] Lincoln DM, Vaia RA, Wang Z-G, Hsiao BS. Secondary structure andelevated temperature crystallite morphology of nylon 6/layered sil-icate nanocomposites. Polymer 2001;42:1621–31.

[13] Vaia RA, Giannelis EP. Liquid crystal polymer nanocomposites:direct intercalation of thermotropic liquid crystalline polymers intolayered silicates. Polymer 2001;42:1281–5.

[14] Fornes TD, Yoon PJ, Keskkula H, Paul DR. Nylon 6 nanocompos-ites: the effect of matrix molecular weight. Polymer 2001;42:9929–40.

[15] Cho JW, Paul DR. Nylon 6 nanocomposites by melt compounding.Polymer 2001;42:1083–94.

[16] Bower CA. Studies on the form and availability of organic soil phos-phorous. IOWA Agric Exp Station Res Bull 1949;362:39–42.

[17] Carter LW, Hendricks JG, Bolley DS. Elastomer reinforced with mod-ified clay (assigned to National Lead Co.). United States Patent No.2,531,396; 1950.

[18] Hauser EA, Kollman RC. Clay complexes with conjugated unsatu-rated aliphatic compounds. US Patent No. 2,951,087; 1960.

[19] Uskov IA. Filled polymers. III. Polymerization of methyl methacry-late during dispersion of sodium bentonite. Vysokomol Soed1960;2:926–30.

[20] Blumstein A. Etude des polymerisations en couche adsorbee. BullChim Soc 1961:899–906.

[21] Greenland DJ. Adsorption of poly(vinyl alcohols) by montmoril-lonite. J Colloid Sci 1963;18:647–64.

[22] Nahin PG, Backlund PS. Organoclay–polyolefin compositions(assigned to Union Oil Co.). United States Patent No. 3,084,117;1963.

[23] Tanihara K, Nakagawa M. Flocculation treatment of waste waterscontaining montmorillonite. IV. Interlamellar complex formationbetween various ion forms of montmorillonite and poly(ethyleneoxide) or polyacrylamide. Nippon Kagaku Kaishi 1975;5:782–9.

[24] Fujiwara S, Sakamoto T. Method for manufacturing aclay/polyamide composite (assigned to Unichika K. K., Japan).Japanese Kokai Patent Application No. 109,998; 1976.

[25] Usuki A, Kojima Y, Kawasumi M, Okada A, Fukushima Y, KurauchiT, et al. Synthesis of nylon 6–clay hybrid. J Mater Res 1993;8:1179–84.

[26] Usuki A, Koiwai A, Kojima Y, Kawasumi M, Okada A, Kurauchi T, etal. Interaction of nylon 6–clay surface and mechanical properties ofnylon 6–clay hybrid. J Appl Polym Sci 1995;55:119–23.

[27] Okada A, Usuki A. The chemistry of polymer–clay hybrids. MaterSci Eng C-Bio S 1995;3:109–15.

[28] Okada A, Fukushima Y, Kawasumi M, Inagaki S, Usuki A, SugiyamaS, Kurauchi T, Kamigatio O. Composite material and its prepara-tion (assigned to Toyota Motor Co., Japan). United States Patent No.4,739,007; 1987.

[29] Vaia RA, Ishii H, Giannelis EP. Synthesis and properties of two-
dimensional nanostructures by direct intercalation of polymermelts in layered silicates. Chem Mater 1993;5:1694–6.
[30] Mehrotra V, Giannelis EP. Conducting molecular multilayers: inter-calation of conjugated polymers in layered media. Mater Res SocSymp Proc 1990;171:39–44.

ess in Po
[31] Manias E. Origins of the materials properties enhancements in poly-mer/clay nanocomposites, http://raman.plmsc.psu.edu/∼manias/pdfs/nano2001b.pdf; 2001.

[32] Shelley JS, Mather PT, DeVries KL. Reinforcement and environ-mental degradation of nylon 6/clay nanocomposites. Polymer2002;42:5849–58.

[33] Chin I-J, Thurn-Albrecht T, Kim H-C, Russell TP, Wang J. On exfolia-tion of montmorillonite in epoxy. Polymer 2001;42:5947–52.

[34] Miranda-Trevino JC, Coles CA. Kaolinite properties, structure andinfluence of metal retention on pH. Appl Clay Sci 2003;23:133–9.

[35] Beyer G. Nanocomposites: a new class of flame retardants for poly-mers. Plast Addit Compound 2002;4(10):22–7.

[36] McNally T, Murphy WR, Lew CY, Turner RJ, Brennan GP. Polyamide-12 layered silicate nanocomposites by melt compounding. Polymer2003;44:2761–72.

[37] Solomon MJ, Almusallam AS, Seefeldt KF, Somwangthanaroj A,Varadan P. Rheology of polypropylene/clay hybrid materials. Macro-molecules 2001;34:1864–72.

[38] Dixon JB. Roles of clays in soils. Appl Clay Sci 1991;5:489–503.[39] Manias E, Touny A, Wu L, Strawhecker K, Lu B, Chung TC. Polypropy-

lene/montmorillonite nanocomposites. Review of the syntheticroutes and materials properties. Chem Mater 2001;13:3516–23.

[40] Ishida H, Campbell S, Blackwell J. General approach to nanocom-posite preparation. Chem Mater 2000;12:1260–7.

[41] Zanetti M, Lomakin S, Camino G. Polymer layered silicate nanocom-posites. Macromol Mater Eng 2000;279:1–9.

[42] Kornmann X, Lindberg H, Berglund LA. Synthesis of epoxy–claynanocomposites: influence of the nature of the clay on structure.Polymer 2001;42:1303–10.

[43] Xie W, Gao Z, Liu K, Pan W-P, Vaia R, Hunter D, et al. Thermal char-acterization of organically modified montmorillonite. ThermochimActa 2001;367/368:339–50.

[44] Kim C-M, Lee D-H, Hoffmann B, Kressler J, Stoppelmann G.Influence of nanofillers on the deformation process in layeredsilicate/polyamide 12 nanocomposites. Polymer 2001;42:1095–100.

[45] Zerda AS, Lesser AJ. Intercalated clay nanocomposites: morphol-ogy, mechanics and fracture behavior. J Polym Sci Polym Phys2001;39:1137–46.

[46] Huang J-C, Zhu Z-K, Yin J, Qian X-F, Sun Y-Y. Poly(etherimide)/montmorillonite nanocomposites prepared by melt intercalation:morphology, solvent resistance properties and thermal properties.Polymer 2001;42:873–7.

[47] Vaia RA, Giannelis EP. Polymer melt intercalation in organically-modified layered silicates: model predictions and experiment.Macromolecules 1997;30:8000–9.

[48] Yu ZZ, Yang M, Zhang Q, Zhao C, Mai YW. Dispersion and distribu-tion of organically modified montmorillonite in nylon-66 matrix. JPolym Sci Polym Phys 2003;41:1234–43.

[49] Wang KH, Choi MH, Koo CM, Choi CM, Chung IJ. Synthesis andcharacterization of maleated polyethylene/clay nanocomposites.Polymer 2001;42:9819–26.

[50] Ray SS, Bousima M. Biodegradable polymers and their layered sili-cate nanocomposites: in greening the 21st century materials world.Prog Mater Sci 2005;50:962–1079.

[51] Fornes TD, Yoon PJ, Hunter DL, Keskkula H, Paul DR. Effect oforganoclay structure on nylon 6 nanocomposite morphology andproperties. Polymer 2002;43:5915–33.

[52] Dennis HR, Hunter DL, Chang D, Kim S, White JL, Cho JW, etal. Effect of melt processing conditions on the extent of exfolia-tion in organoclay-based nanocomposites. Polymer 2001;42:9513–22.

[53] Wu S-H, Wang F-Y, Ma C-CM, Chang W-C, Kuo C-T, Kuan H-C, et al.Mechanical, thermal and morphological properties of glass fiberand carbon fiber reinforced polyamide 6 and polyamide 6/claynanocomposites. Mater Lett 2001;49:327–33.

[54] Krishnamoorti R, Giannelis EP. Rheology of end-tetheredpolymer layered silicate nanocomposites. Macromolecules1997;30:4097–102.

[55] Ray SS, Okamoto M. Polymer–layered silicate nanocompos-ite: a review from preparation to processing. Prog Polym Sci2003;28:1539–641.

[56] Porter D, Metcalfe E, Thomas MJK. Nanocomposite fire retardants—areview. Fire Mater 2000;24:45–52.

[57] Ma J, Xu J, Ren J-H, Yu Z-Z, Mai Y-W. A new approach to poly-
mer/montmorillonite nanocomposites. Polymer 2003;44:4619–24.
[58] Morgan AB, Gilman JW. Characterization of polymer–layered sil-icate (clay) nanocomposites by transmission electron microscopyand X-ray diffraction: a comparative study. J Appl Polym Sci2003;87:1329–38.

lymer Science 33 (2008) 1119–1198

[59] VanderHart DL, Asano A, Gilman JW. NMR measurements relatedto clay dispersion quality and organic-modifier stability innylon 6/clay nanocomposites. Macromolecules 2001;34:3819–22.

[60] VanderHart DL, Asano A, Gilman JW. Solid-state NMR investigationof paramagnetic nylon 6 clay nanocomposites. 2. Measurement ofclay dispersion, crystal stratification, and stability of organic mod-ifiers. Chem Mater 2001;13:3796–809.

[61] VanderHart DL, Asano A, Gilman JW. Solid-state NMR investigationof paramagnetic nylon 6 clay nanocomposites. 1. Crystallinity, mor-phology, and the direct influence of Fe3+ on nuclear spins. ChemMater 2001;13:3781–95.

[62] Loo LS, Gleason KK. Fourier transforms infrared investigation ofthe deformation behavior of montmorillonite in nylon 6/nanoclaynanocomposite. Macromolecules 2003;36:2587–90.

[63] Wu HD, Tseng CR, Chang FC. Chain conformation and crystal-lization behavior of the syndiotactic polystyrene nanocompositesstudied using Fourier transform infrared analysis. Macromolecules2001;34:2992–9.

[64] Fornes TD, Paul DR. Modelling properties of nylon 6/clay nanocom-posites using composite theories. Polymer 2003;44:4993–5013.

[65] Kim SW, Jo WH, Lee MS, Ko MB, Jho JY. Preparation of clay–dispersed poly(styrene-co-acrylonitrile) nanocomposites usingpoly(�-caprolactone) as a compatibilizer. Polymer 2001;42:9837–42.

[66] Vaia RA, Giannelis EP. Lattice of polymer melt intercalation in organ-ically modified layered silicates. Macromolecules 1997;30:7990–9.

[67] Arada P, Ruiz-Hitzky E. Polymer–salt intercalation complexes inlayer silicates. Adv Mater 1990;2:545–7.

[68] Arada P, Ruiz-Hitzky E. Poly(ethylene oxide)–silicate intercalationmaterials. Chem Mater 1992;4:1395–403.

[69] Tunney JJ, Detellier C. Poly(ethylene glycol)–kaolinite intercalates.Chem Mater 1996;8:927–35.

[70] Fischer HR, Gielgens LH, Koster TPM. Nanocomposites from poly-mers and layered minerals. Acta Polym 1999;50:122–6.

[71] Theng BKG. Formation and properties of clay–polymer complexes.Amsterdam: Elsevier; 1979.

[72] Ogata N, Kawakage S, Ogihara T. Poly(vinyl alcohol)/clay andpoly(ethylene oxide)/clay blend prepared using water as solvent.J Appl Polym Sci 1997;66:573–81.

[73] Parfitt RL, Greenland DJ. Adsorption of poly(ethylene glycols) onmontmorillonites. Clay Miner 1970;8:305–15.

[74] Ruiz-Hitzky E, Aranda P, Casal B, Galvan JC. Nanocomposite materi-als with controlled ion mobility. Adv Mater 1995;7:180–4.

[75] Levy R, Francis CW. Interlayer adsorption of polyvinylpyrrolidoneon montmorillonite. J Colloid Interface Sci 1975;50:442–50.

[76] Billingham J, Breen C, Yarwood J. Adsorption of polyamine, poly-acrylic acid and polyethylene glycol on montmorillonite: an in situstudy using ATR-FTIR. Vib Spectrosc 1997;14:19–34.

[77] Zhao X, Urano K, Ogasawara S. Adsorption of poly(ethylene vinylalcohol) from aqueous solution on montmorillonite clays. ColloidPolym Sci 1989;267:899–906.

[78] Ogata N, Jimenez G, Kawai H, Ogihara T. Structure and ther-mal/mechanical properties of poly(l-lactide)–clay blend. J PolymSci Polym Phys 1997;35:389–96.

[79] Jeon HG, Jung HT, Lee SW, Hudson SD. Morphology of polymersilicate nanocomposites. High density polyethylene and a nitrile.Polym Bull 1998;41:107–13.

[80] Jimenez G, Ogata N, Kawai H, Ogihara T. Structure and ther-mal/mechanical properties of poly(�-caprolactone)–clay blend. JAppl Polym Sci 1997;64:2211–20.

[81] Srivastava SK, Pramanik M, Acharya H. Ethylene/vinyl acetatecopolymer/clay nanocomposites. J Polym Sci Polym Phys2006;44:471–80.

[82] Tseng C-R, Wu J-Y, Lee H-Y, Chang F-C. Preparation and crystal-lization behavior of syndiotactic polystyrene–clay nanocomposites.Polymer 2001;42:10063–70.

[83] Krikorian V, Pochan D. Poly(l-lactide acid)/layered silicatenanocomposite: fabrication, characterization, and properties.Chem Mater 2003;15:4317–24.

[84] Chang J-H, Uk-An Y, Sur GS. Poly(lactic acid) nanocomposites withvarious organoclays. I. thermomechanical properties, morphology,and gas permeability. J Polym Sci Polym Phys 2003;41:94–103.

[85] Yano K, Usuki A, Okada A, Kurauchi T, Kamigaito O. Synthesis
and properties of polyimide/clay hybrid. J Polym Sci Polym Chem1993;31:2493–8.
[86] Manias E, Chen E, Krishnamoorti R, Genzer J, Kramer EJ, GiannelisEP. Intercalation kinetics of long polymers in 2 nm confinements.Macromolecules 2000;33:7955–66.

http://raman.plmsc.psu.edu/~manias/pdfs/nano2001b.pdf

http://raman.plmsc.psu.edu/~manias/pdfs/nano2001b.pdf

ess in Po

[135] Messersmith PB, Giannelis EP. Synthesis and characterization oflayered silicate–epoxy nanocomposites. Chem Mater 1994;6:1719–


[87] Strawhecker KE, Manias E. Structure and properties of poly(vinylalcohol)/Na+–montmorillonite nanocomposites. Chem Mater2000;12:2943–9.

[88] Finnigan B, Martin D, Halley P, Truss R, Campell K. Morphology andproperties of thermoplastic polyurethane nanocomposites incor-porating hydrophilic layered silicates. Polymer 2004;45:2249–60.

[89] Ginsburg VV, Balazs AC. Calculating phase diagrams for nanocom-posites: the effect of adding end-functionalized chains topolymer/clay mixture. Adv Mater 2000;12:1805–9.

[90] Choi HJ, Kim SG, Hyun YH, Jhon MS. Preparation andrheological characteristics of solvent-cast poly(ethyleneoxide)/montmorillonite nanocomposites. Macromol RapidCommun 2001;22:320–5.

[91] Hackett E, Manias E, Giannelis EP. Molecular dynamics simu-lations of organically modified layered silicates. J Chem Phys1998;108:7410–5.

[92] Sur GS, Sun HL, Lyu SG, Mark JE. Synthesis, structure, mechanicalproperties, and thermal stability of some polysulfone/organoclaynanocomposites. Polymer 2001;42:9783–9.

[93] Usuki A, Kawasumi M, Kojima Y, Okada A, Kurauchi T, KamigaitoO. Swelling behavior of montmorillonite cation exchanged for �-amino acid by �-caprolactam. J Mater Res 1993;8:1174–8.

[94] Kojima Y, Usuki A, Kawasumi M, Okada A, Kurauchi T, Kamigaito O.Synthesis of nylon-6 hybrid by montmorillonite intercalated with�-caprolactam. J Polym Sci Polym Chem 1993;31:983–6.

[95] Kojima Y, Usuki A, Kawasumi M, Okada A, Kurauchi T, Kamigaito O.One-pot synthesis of nylon-6–clay hybrid. J Polym Sci Polym Chem1993;31:1755–8.

[96] Reichert P, Kressler J, Thomann R, Mulhaupt R, Stoppelmann G.Nanocomposites based on a synthetic layer silicate and polyamide-12. Acta Polym 1998;49:116–23.

[97] Wu Z, Zhou C, Qi R, Zhang H. Synthesis and characterization of nylon1012/clay nanocomposite. J Appl Polym Sci 2002;83:2403–10.

[98] Okamoto M, Morita S, Taguchi H, Kim YH, Kotaka T, Tateyama H.Synthesis and structure of smectic clay/poly(methyl methacrylate)and clay polystyrene nanocomposites via in situ intercalative poly-merization. Polymer 2000;41:3887–90.

[99] Okamoto M, Morita S, Kotaka T. Dispersed structure and ionicconductivity of smectic clay/polymer nanocomposites. Polymer2001;42:2685–8.

[100] Akelah A, Moet M. Polymer–clay nanocomposites: free radical graft-ing of polystyrene on to organophilic montmorillonite interlayers.J Mater Sci 1996;31:3589–96.

[101] Hsu SLC, Chang KC. Synthesis and properties of polybenzo-xazole–clay nanocomposites. Polymer 2002;43:4097–101.

[102] Sun T, Garces JM. High-performance polypropylene–clay nanocom-posites by in-situ polymerization with metallocene/clay catalysts.Adv Mater 2002;14:128–30.

[103] Bergman JS, Chen H, Giannelis EP, Thomas MG, Coates GW. Syn-thesis and characterization of polyolefin–silicate nanocomposites:a catalyst intercalation and in situ polymerization approach. J ChemSoc Chem Commun 1999;21:2179–80.

[104] Tudor J, Willington L, O’Hare D, Royan B. Intercalation of catalyticallyactive metal complexes in phyllosilicates and their application aspropene polymerization catalyst. Chem Commun 1996;17:2031–2.

[105] Jin Y-H, Park H-J, Im S-S, Kwak S-Y, Kwak S. Polyethylene/claynanocomposite by in situ exfoliation of montmorillonite duringZiegler–Natta polymerization of ethylene. Macromol Rapid Com-mun 2002;23:135–40.

[106] Ke YC, Long CF, Qi ZN. Crystallization, properties, and crystal andnanoscale morphology of PET–clay nanocomposites. J Appl PolymSci 1999;71:1139–46.

[107] Lee S-S, Ma YT, Rhee H-W, Kim J. Exfoliation of layered silicate byring opening reaction of cyclic oligomers in PET–clay nanocompos-ites. Polymer 2005;46:2201–10.

[108] Alexandre M, Dubois P, Jerome R, Garcia-Marti M, Sun T, GarcesJM, Millar DM, Kuperman A. Polyolefin nanocomposites. WO:947598A1; 1999.

[109] Alexandre M, Dubois P, Garces JM, Sun T, Jerome R.Polyethylene–layered silicate nanocomposites prepared bythe polymerization-filling technique: synthesis and mechanicalproperties. Polymer 2002;43:2123–32.

[110] Dubois P, Alexandre M, Hindryckx F, Jerome R. Homogeneouspolyolefin-based composites. J Macromol Sci Rev Macromol Chem
Phys 1998;C38:511–65.
[111] Shin S-YA, Simon LC, Soares JBP, Scholz G. Polyethylene–clay hybridnanocomposites: in situ polymerization using bifunctional organicmodifiers. Polymer 2003;44:5317–21.

lymer Science 33 (2008) 1119–1198 1195

[112] Heinemann J, Reichert P, Thomann R, Mulhaupt R. Polyolefinnanocomposites formed by melt compounding and transition metalcatalyzed ethene homo- and copolymerization in the presence oflayered silicates. Macromol Rapid Commun 1999;20:423–30.

[113] Doh JG, Cho I. Synthesis and properties of polystyrene–organoammonium montmorillonite hybrid. Polym Bull 1998;41:511–8.

[114] Weimer MW, Chen H, Giannelis EP, Sogah DY. Direct synthesis of dis-persed nanocomposites by in situ living free radical polymerizationusing a silicate-anchored initiator. J Am Chem Soc 1999;121:1615–6.

[115] Yei D-R, Kuo S-W, Fu H-K, Chang F-C. Enhanced thermalproperties of PS nanocomposites formed from montmorillonitetreated with a surfactant/cyclodextrin inclusion complex. Polymer2005;46:741–50.

[116] Paul MA, Alexandre M, Degee P, Calberg C, Jerome R, DuboisP. Exfoliated polylactide/clay nanocomposites by in situcoordination–insertion polymerization. Macromol Rapid Commun2003;24:561–6.

[117] Messersmith PB, Giannelis EP. Polymer–layered silicate nanocom-posites: in-situ intercalative polymerization of �-caprolactone inlayered silicates. Chem Mater 1993;5:1064–6.

[118] Messersmith PB, Giannelis EP. Synthesis and barrier properties ofpoly(�-caprolactone)–layered silicate nanocomposites. J Polym SciPolym Chem 1995;33:1047–57.

[119] Pantoustier N, Alexandre M, Degee P, Calberg C, Jerome R, Hen-rist C, et al. Poly(�-caprolactone) layered silicate nanocomposites:effect of clay surface modifiers on the melt intercalation process.e-Polymer 2001;9:1–9.

[120] Pantoustier N, Lepoittevin B, Alexandre M, Kubies D, CalbergC, Jerome R, et al. Biodegradable polyester layered silicatenanocomposites based on poly(�-caprolactone). Polym Eng Sci2002;42:1928–37.

[121] Rozenberg BA. Kinetics and mechanism of �-caprolactone anionicpolymerization under the influence of amines. Pure Appl Chem1981;53:1715–22.

[122] Mecerreyes D, Jerome R, Dubois P. Novel macromoleculararchitectures based on aliphatic polyesters: relevance of thecoordination–insertion ring opening polymerization. Adv Polym Sci1999;147:1–59.

[123] Lofgren A, Albertsson A-C, Dubois P, Jerome R. Recent advances inring-opening polymerization of lactones and related compounds. JMacromol Sci Rev Macromol Chem Phys C 1995;35:379–418.

[124] Sepehr M, Utracki LA, Zheng X, Wilkie CA. Polystyrenes with macro-intercalated organoclay. Part I. Compounding and characterization.Polymer 2005;46:11557–68.

[125] Lan T, Pinnavaia TJ. Clay–reinforced epoxy nanocomposites. ChemMater 1994;6:2216–9.

[126] Becker O, Varley R, Simon G. Morphology, thermal relaxations andmechanical properties of layered silicate nanocomposites basedupon high-functionality epoxy resins. Polymer 2002;43:4365–73.

[127] Kornmann X, Berglund LA, Sterte J. Nanocomposite basedon montmorillonite and unsaturated polyester. Polym Eng Sci1998;38:1351–8.

[128] Jiankun L, Yucai K, Zongneng Q, Xiao-Su Y. Study on intercalationand exfoliation behavior of organoclays in epoxy resin. J Polym SciPolym Phys 2001;39:115–20.

[129] Lan T, Kaviratna PD, Pinnavaia TJ. Synthesis, characterization andmechanical properties of epoxy–clay nanocomposites. Polym MaterSci Eng 1994;71:527–8.

[130] Liu W, Hoa SV, Pugh M. Fracture toughness and water uptake ofhigh-performance epoxy/nanoclays nanocomposites. Compos SciTechnol 2005;65:2364–73.

[131] Lan T, Kaviratna PD, Pinnavaia TJ. Mechanism of clay tactoid exfoli-ation in epoxy–clay nanocomposites. Chem Mater 1995;7:2144–50.

[132] Zilg C, Mulhaupt R, Finter J. Morphology and toughness/stiffnessbalance of nanocomposites based upon anhydride curedepoxy resins and layered silicates. Macromol Chem Phys1999;200:661–70.

[133] Becker O, Varley RJ, Simon GP. Thermal stability and water uptakeof high performance epoxy layered silicate nanocomposites. EurPolym J 2004;40:187–95.

[134] Hackman I, Hollaway L. Epoxy–layered silicate nanocomposites incivil engineering. Compos Part A-Appl S 2006;37:1161–70.

25.[136] LePluart L, Duchet J, Sautereau H. Epoxy/montmorillonite nanocom-

posites: influence of organophilic treatment on reactivity, morphol-ogy and fracture properties. Polymer 2005;46:12267–78.

ess in Po
[137] Koerner H, Misra D, Tan A, Drummy L, Mirau P, VaiaR. Montmorillonite–thermoset nanocomposites via cryo-compounding. Polymer 2006;47:3426–35.

[138] Wang K, Chen L, Kotaki M, He C. Preparation, microstructure andthermal mechanical properties of epoxy/crude clay nanocompos-ites. Compos Part A-Appl S 2007;38:192–7.

[139] Suh D, Lim Y, Park O. The property and formation mechanism ofunsaturated polyester–layered silicate nanocomposite dependingon the fabrication methods. Polymer 2000;41:8557–63.

[140] Bharadwaj RK, Mehrabi AR, Hamilton C, Trujillo C, MurgaMF, Chavira A. Structure–property relationships in cross-linkedpolyester–clay nanocomposites. Polymer 2002;43:3699–705.

[141] Wang Z, Pinnavaia TJ. Nanolayer reinforcement of elastomericpolyurethane. Chem Mater 1998;10:3769–71.

[142] Yao KJ, Song M, Hourston DJ, Luo DZ. Polymer/layered claynanocomposites: 2 polyurethane nanocomposites. Polymer2002;43:1017–20.

[143] Zilg C, Thomann R, Mulhaupt R, Finter J. Polyurethane nanocompos-ites containing laminated anisotropic nanoparticles derived fromorganophilic layered silicates. Adv Mater 1999;11:49–52.

[144] Berta M, Lindsay C, Pans G, Camino G. Effect of chemical struc-ture on combustion and thermal behaviour of polyurethaneelastomer layered silicate nanocomposites. Polym Degrad Stabil2006;91:1179–91.

[145] Gao X, Lee J, Widya T, Macosko C. Polyurethane/clay nanocom-posites foams: processing, structure and properties. Polymer2005;46:775–83.

[146] Liu X, Wu Q. Polyamide 66/clay nanocomposites via melt intercala-tion. Macromol Mater Eng 2002;287:180–6.

[147] Phang IY, Liu T, Mohamed A, Pramoda KP, Chen L, Shen L, et al.Morphology, thermal and mechanical properties of nylon 12/organ-oclay nanocomposites prepared by melt compounding. Polym Int2005;54:456–64.

[148] Davis CH, Mathias LJ, Gilman JW, Schiraldi DA, Shields JR, TruloveP, et al. Effects of melt-processing conditions on the quality ofpoly(ethylene terephthalate) montmorillonite clay nanocompos-ites. J Polym Sci Polym Phys 2002;40:2661–6.

[149] Pegoretti A, Kolarik J, Peronic C, Migliaresi C. Recycled poly(ethyleneterephthalate) layered silicate nanocomposites: morphology andtensile mechanical properties. Polymer 2004;45:2751–9.

[150] Tang Y, Hu Y, Wang J, Zong R, Gui Z, Chen Z, et al. Influ-ence of organophilic clay and preparation methods onEVA/montmorillonite nanocomposites. J Appl Polym Sci2004;91:2416–21.

[151] Gopakumar TG, Lee JA, Kontopoulou M, Parent JS. Influenceof clay exfoliation on the physical properties of montmoril-lonite/polyethylene composites. Polymer 2002;43:5483–91.

[152] Thellen C, Orroth C, Froio D, Ziegler D, Lucciarini J, Farrell R, et al.Influence of montmorillonite layered silicate on plasticized poly(l-lactide) blown films. Polymer 2005;46:11716–27.

[153] Ray SS, Maiti P, Okamoto M, Yamada K, Ueda K. New polylac-tide/layered silicate nanocomposites. 1. Preparation, characteriza-tion and properties. Macromolecules 2002;35:3104–10.

[154] Pluta M. Morphology and properties of polylactide modified bythermal treatment, filling with layered silicates and plasticization.Polymer 2004;45:8239–51.

[155] Di Y, Iannace S, Maio ED, Nicolais L. Nanocomposites by melt inter-calation based on polycaprolactone and organoclay. J Polym SciPolym Phys 2003;41:670–8.

[156] Gorrasi G, Tortora M, Vittoria V, Pollet E, Lepoittevin B, AlexandreM, et al. Vapor barrier properties of polycaprolactone mont-morillonite nanocomposites: effect of clay dispersion. Polymer2003;44:2271–9.

[157] Wolf D, Fuchs A, Wagenknecht U, Kretzschmar B, Jehnichen D, Haus-sler L. Nanocomposites of polyolefin clay hybrids. In: Proceedingsof Eurofiller ’99. 1999. p. 6–9.

[158] Liu X, Wu Q. PP/clay nanocomposites prepared by grafting meltintercalation. Polymer 2001;42:10013–9.

[159] Zheng X, Jiang DD, Wilkie CA. Polystyrene nanocomposites basedon an oligomerically-modified clay containing maleic anhydride.Polym Degrad Stabil 2006;91:108–13.

[160] Hasegawa N, Okamoto H, Kato M, Usuki A, Sato N. Nylon6–montmorillonite nanocomposites prepared by compounding
nylon 6 with Na–montmorillonite slurry. Polymer 2003;44:2933–7.
[161] Garcia-Lopez D, Picazo O, Merino JC, Pastor JM. Polypropylene–claynanocomposites: effect of compatibilizing agents on clay disper-sion. Eur Polym J 2003;39:945–50.

[162] Hotta S, Paul DR. Nanocomposites formed from linear low densitypolyethylene and organoclays. Polymer 2004;45:7639–54.

lymer Science 33 (2008) 1119–1198

[163] Duquesne S, Jama C, Le Bras M, Delobel R, Recourt P, GloaguenJM. Elaboration of EVA-nanoclay systems—characterization,thermal behavior and fire performance. Compos Sci Technol2003;63:1141–8.

[164] Zhang W, Chen D, Zhao Q, Fang Y. Effects of different kinds of clayand different vinyl acetate content on the morphology and proper-ties of EVA/clay nanocomposites. Polymer 2003;44:7953–61.

[165] Zanetti M, Camino G, Thomann R, Mulhaupt R. Synthesis and ther-mal behavior of layered silicate–EVA nanocomposites. Polymer2001;42:4501–7.

[166] Alexandre M, Beyer G, Henrist C, Cloots R, Rulmont A, Jérôme R,et al. Preparation and properties of layered silicate nanocompos-ites based on ethylene vinyl acetate copolymers. Macromol RapidCommun 2001;22:643–6.

[167] Peeterbroeck S, Alexandre M, Jerome R, Dubois Ph. Poly(ethylene-co-vinyl acetate)/clay nanocomposites: effect of clay nature andorganic modifiers on morphology, mechanical and thermal prop-erties. Polym Degrad Stabil 2005;90:288–94.

[168] Balazs AC, Singh C, Zhulina. Modeling the interactions betweenpolymers and clay surfaces through self-consistent field theory.Macromolecules 1998;31:8370–81.

[169] Kurian M, Dasgupta A, Beyer FL, Galvin ME. Investigation ofthe effects of silicate modification on polymer–layered sil-icate nanocomposite morphology. J Polym Sci Polym Phys2004;42:4075–83.

[170] Liu LM, Qi ZN, Zhu XG. Studies on nylon-6 clay nanocomposites bymelt-intercalation process. J Appl Polym Sci 1999;71:1133–8.

[171] Chaudhary DS, Prasad R, Gupta RK, Bhattacharya SN. Clayintercalation and influence on crystallinity of EVA-based claynanocomposites. Thermochim Acta 2005;433:187–95.

[172] Li X, Ha CS. Nanostructure of EVA/organoclay nanocomposites:effects of kinds of organoclays and grafting of maleic anhydride ontoEVA. J Appl Polym Sci 2003;87:1901–9.

[173] Strong AB. Plastics-Materials and Processing. 2nd edition PrenticeHall, Inc.; 2000.

[174] Li J, Zhao L, Guo S. Ultrasonic preparation of polymer/layered silicatenanocomposites during extrusion. Polym Bull 2005;55:217–23.

[175] Maiti P, Nam PH, Okamoto M, Kotaka T, Hasegawa N, UsukiA. Influence of crystallization on intercalation, morphology, andmechanical properties of propylene/clay nanocomposites. Macro-molecules 2002;35:2042–9.

[176] Maiti P, Nam PH, Okamoto M, Kotaka T, Hasegawa N, UsukiA. The effect of crystallization on the structure and mor-phology of polypropylene/clay nanocomposites. Polym Eng Sci2002;42:1864–71.

[177] Zanetti M, Costa L. Preparation and combustion behavior of poly-mer/layered silicate nanocomposites based upon PE and EVA.Polymer 2004;45:4367–73.

[178] Preston CML, Amarasinghe G, Hopewell JL, Shanks RA, Mathys Z.Evaluation of polar ethylene copolymers as fire retardant nanocom-posite matrices. Polym Degrad Stabil 2004;84:533–44.

[179] Zhai H, Xu W, Guo H, Zhou Z, Shen S, Song Q. Preparation and charac-terization of PE and PE-g-MAH/montmorillonite nanocomposites.Eur Polym J 2004;40:2539–45.

[180] Lu H, Hu Y, Xiao J, Kong Q, Chen Z, Fan W. The influence of irradiationon morphology evolution and flammability properties of maleatedpolyethylene/clay nanocomposite. Mater Lett 2005;59(6):648–51.

[181] Morawiec J, Pawlak A, Slouf M, Galeski A, Piorkowska E,Kransnikowa N. Preparation and properties of compatibilizedLDPE/organo-modified montmorillonite nanocomposites. EurPolym J 2005;41:1115–22.

[182] Zhang J, Wilkie CA. Preparation and flammability proper-ties of polyethylene–clay nanocomposites. Polym Degrad Stabil2003;80:163–9.

[183] Tang Y, Hu Y, Li B, Liu L, Wang Z, Chen Z, et al. Polypropy-lene/montmorillonite nanocomposites and intumescent, flame-retardant montmorillonite synergism in polypropylene nanocom-posites. J Polym Sci Polym Chem 2004;42:6163–73.

[184] Zhao C, Qin H, Gong F, Feng M, Zhang S, Yang M. Mechanical, thermaland flammability properties of polyethylene/clay nanocomposites.Polym Degrad Stabil 2005;87:183–9.

[185] Hong HC, Lee YB, Bae JW, Jho YJ, Nam BU, Nam GJ, et al. Ten-sile and flammability properties of polypropylene-based RTPO/clay
nanocomposites for cable insulating material. J Appl Polym Sci2005;97:2375–81.
[186] Hasegawa N, Kawasumi M, Kato M, Usuki A, Okada A. Preparationand mechanical properties of polypropylene–clay hybrids using amaleic anhydride-modified polypropylene oligomer. J Appl PolymSci 1998;67:87–92.

ess in Po

[


[187] Hasegawa N, Okamoto H, Kawasumi M, Kato M, Tsukigase A, UsukiA. Polyolefin–clay hybrids based on modified polyolefins and organ-oclay. Macromol Mater Eng 2000;280/281:76–9.

[188] Kato M, Usuki A, Okada A. Synthesis of polypropylene oligomer–clayintercalation compounds. J Appl Polym Sci 1997;66:1781–5.

[189] Nam PH, Maiti P, Okamoto M, Kotaka T, Hasegawa N, Usuki A. A hier-archical structure and properties of intercalated polypropylene/claynanocomposites. Polymer 2001;42:9633–40.

[190] Wang D, Wilkie CA. In-situ reactive blending to preparepolystyrene–clay and polypropylene–clay nanocomposites. PolymDegrad Stabil 2003;80:171–82.

[191] Hasegawa N, Okamoto H, Kawasumi M, Usuki A. Preparation andmechanical properties of polystyrene–clay hybrids. J Appl PolymSci 1999;74:3359–64.

[192] Davis RD, Gilman JW, VanderHart DL. Processing degradationof polyamide 6/montmorillonite clay nanocomposites and clayorganic modifier. Polym Degrad Stabil 2003;79:111–21.

[193] Chang J-H, Kim SJ, Joo YL, Im S. Poly(ethylene terephthalate)nanocomposites by in situ interlayer polymerization: the thermo-mechanical properties and morphology of the hybrid fiber. Polymer2004;45:919–26.

[194] Gilman JW, Awad WH, Davis RD, Shields J, Harris Jr RH, DavisC, et al. Polymer/layered silicate nanocomposites from thermallystable trialkylimidazolium-treated montmorillonite. Chem Mater2002;14:3776–85.

[195] Maiti P, Batt CA, Giannelis EP. Renewable plastics: synthesisand properties of PHB nanocomposites. J Macromol Sci Rev2003;88:58–9.

[196] Vaia RA, Vasudevan S, Krawiec W, Scanlon LG, GiannelisEP. New polymer electrolyte nanocomposites: melt intercala-tion of poly(ethylene oxide) in mica-type silicates. Adv Mater1995;7:154–6.

[197] Kawasumi M, Hasegawa N, Kato M, Usuki A, Okada A. Preparationand mechanical properties of polypropylene–clay hybrids. Macro-molecules 1997;30:6333–8.

[198] Shia D, Hui CY, Burnside SD, Giannelis EP. An interface model forthe prediction of Young’s modulus of layered silicate–elastomernanocomposites. Polym Compos 1998;19:608–17.

[199] Kojima Y, Usuki A, Kawasumi M, Okada A, Kurauchi T, KamigaitoO. Sorption of water in nylon 6–clay hybrid. J Appl Polym Sci1993;49:1259–64.

200] Goettler LA. Overview of property development in layered sil-icate polymer nanocomposites. Ann Tech Confr Soc Plast Eng2005:1980–2.

[201] Tortora M, Vittoria V, Galli G, Ritrovati S, Chiellini E. Trans-port properties of modified montmorillonite–poly(�-caprolactone)nanocomposites. Macromol Mater Eng 2002;287:243–9.

[202] Xiong J, Liu Y, Yang X, Wang X. Thermal and mechanical propertiesof polyurethane/montmorillonite nanocomposites based on a novelreactive modifier. Polym Degrad Stabil 2004;86:549–55.

[203] Kojima Y, Usuki A, Kawasumi M, Okada A, Fukushima Y, KarauchiT, et al. Mechanical properties of nylon-6/clay hybrid. J Mater Res1993;6:1185–9.

[204] Tortora M, Gorrasi G, Vittoria V, Galli G, Ritrovati S, Chiellini E.Structural characterization and transport properties of organicallymodified montmorillonite/polyurethane nanocomposites. Polymer2002;43:6147–57.

[205] Noh MW, Lee DC. Synthesis and characterization of PS–claynanocomposite by emulsion polymerization. Polym Bull1999;42:619–26.

[206] Ray SS, Yamada K, Okamoto M, Ogami A, Ueda K. New poly-lactide/layered silicate nanocomposites. 3. High performancebiodegradable materials. Chem Mater 2003;15:1456–65.

[207] Ray SS, Okamoto M. Biodegradable polylactide and its nanocom-posites: opening a new dimension for plastics and composites.Macromol Rapid Commun 2003;24:815–40.

[208] Ray SS, Yamada K, Okamoto M, Ueda K. New polylactide/layeredsilicate nanocomposites. 2. Concurrent improvements of mate-rial properties, biodegradability and melt rheology. Polymer2003;44:857–66.

[209] Pramanik M, Srivastava SK, Samantaray BK, Bhowmick AK. EVA/claynanocomposite by solution blending: effect of aluminosilicatelayers on mechanical and thermal properties. Macromol Res2003;11:260–6.

[210] Chen B, Evans JRG. Poly(�-caprolactone)–clay nanocompos-
ites: structure and mechanical properties. Macromolecules2006;39:747–54.
[211] Laus M, Francesangeli O, Sandrolini F. New hybrid nanocompos-ites based on an organophilic clay and poly(styrene-b-butadiene)copolymers. J Mater Res 1997;12:3134–9.

lymer Science 33 (2008) 1119–1198 1197

[212] Okamoto M, Nam PH, Maiti P, Kotaka T, Nakayama T, Takada M, etal. Biaxial flow-induced alignment of silicate layers in polypropy-lene/clay nanocomposite foam. Nano Lett 2001;1(9):503–5.

[213] Nielsen LE. Mechanical properties of polymer and composites, vol.2. New York: Marcel Dekker; 1981.

[214] Fredrickson GH, Bicerano J. Barrier properties of oriented disk com-posites. J Chem Phys 1999;110:2181–8.

[215] Burnside SD, Giannelis EP. Synthesis and properties ofnew poly(dimethylsiloxane) nanocomposites. Chem Mater1995;7:1597–600.

[216] Lange J, Wyser Y. Recent innovations in barrier technologies forplastic packaging—a review. Packag Technol Sci 2003;16:149–58.

[217] Nielsen LE. Models for the permeability of filled polymer systems.J Macromol Sci 1967;A1:929–42.

[218] Cussler EL, Hughes SE, Ward WJ, Aris R. Barrier membranes. J MembrSci 1988;38:161–74.

[219] Petricova E, Knight R, Schadler LS, Twardowski TE. Nylon 11/sil-ica nanocomposite coatings applied by the HVOF process. II.Mechanical and barrier properties. J Appl Polym Sci 2000;78:2272–89.

[220] Fukuda M, Kuwajima S. Molecular-dynamics simulation of mois-ture diffusion in polyethylene beyond 10 ns duration. J Chem Phys1997;107:2149–59.

[221] Drozdov AD, Christiansen JdeC, Gupta RK, Shah AP. Model foranomalous moisture diffusion through a polymer–clay nanocom-posite. J Polym Sci Polym Phys 2003;41:476–92.

[222] Ke Z, Yongping B. Improve gas barrier property of PET film withmontmorillonite by in situ interlayer polymerization. Mater Lett2005;59:3348–51.

[223] Ogasawara T, Ishida Y, Ishikawa T, Aoki T, Ogura T. Helium gas per-meability of montmorillonite/epoxy nanocomposites. Compos PartA-Appl S 2006;37:2236–40.

[224] Zhu J, Uhl FM, Morgan AB, Wilkie CA. Studies on the mechanism bywhich the formation of nanocomposites enhances thermal stability.Chem Mater 2001;13:4649–54.

[225] Vyazovkin S, Dranka I, Fan X, Advincula R. Kinetics of the thermaland thermo-oxidative degradation of a polystyrene–clay nanocom-posite. Macromol Rapid Commun 2004;25:498–503.

[226] Zanetti M, Bracco P, Costa L. Thermal degradation behavior ofPE/clay nanocomposites. Polym Degrad Stabil 2004;85:657–65.

[227] Bandyopadhyay S, Chen R, Giannelis EP. Biodegradableorganic–inorganic hybrids based on poly(l-lactide). J Macromol SciRev 1999;81:159–60.

[228] Lepoittevin B, Pantoustier N, Devalckenaere M, Alexandre M, KubiesD, Calderg C. Poly(�-caprolactone)/clay nanocomposites by in situintercalative polymerization catalyzed by dibutyltindimethoxide.Macromolecules 2002;35:8385–90.

[229] Dabrowski F, Bourbigot S, Delobel R, LeBras M. Kinetic modeling ofthe thermal degradation of polyamide 6 nanocomposite. Eur PolymJ 2000;36:273–84.

[230] Jang BN, Wilkie CA. The effect of clay on the thermal degrada-tion of polyamide 6 in polyamide 6/clay nanocomposites. Polymer2005;46:3264–74.

[231] Pandey JK, Reddy KR, Kumar AP, Singh RP. An overview of thedegradability of polymer nanocomposites. Polym Degrad Stabil2005;88:234–50.

[232] Dutta S, Bhowmick AK, Mukunda PG, Chaki TK. Thermal degradationstudies of electron beam cured ethylene-vinyl acetate copolymer.Polym Degrad Stabil 1995;50:75–82.

[233] Paul M-A, Alexandre M, Degee P, Henrist C, Rulmont A, Dubois P.New nanocomposite materials based on plasticized poly(l-lactide)and organo-modified montmorillonites: thermal and morphologi-cal study. Polymer 2003;44:443–50.

[234] Riva A, Zanetti M, Braglia M, Camino G, Falqui L. Thermaldegradation and rheological behaviour of EVA/montmorillonitenanocomposites. Polym Degrad Stabil 2002;77:299–304.

[235] Maurin MB, Dittert LW, Hussain AA. Thermogravimetric analy-sis of ethylene-vinyl acetate copolymers with Fourier transforminfrared analysis of the pyrolysis products. Thermochim Acta1991;186:97–102.

[236] McGrattan BJ. Examining the decomposition of ethylene-vinyl acetate copolymers using TG/GC/IR. Appl Spectrosc1994;48:1472–6.

[237] Costache MC, Jiang DD, Wilkie CA. Thermal degradation of
ethylene-vinyl acetate copolymer nanocomposites. Polymer2005;63:6947–58.
[238] Beyer G. Flame retardant properties of EVA-nanocomposites andimprovements by combination of nanofillers with aluminium tri-hydrate. Fire Mater 2002;25:193–7.

ess in Po
[239] Morgan AB. Flame retarded polymer layered silicate nanocompos-ites: a review of commercial and open literature systems. PolymAdv Technol 2006;17:206–17.

[240] Gilman JW, Jackson CL, Morgan AB, Harris Jr R, Manias E,Giannelis EP, et al. Flammability properties of polymer–layered sil-icate nanocomposites. Propylene and polystyrene nanocomposites.Chem Mater 2000;12:1866–73.

[241] Gilman JW, Kashiwagi TCL, Giannelis EP, Manias E, Lomakin S,Lichtenhan JD, et al. Radiative gasification and vinyl polymerflammability. In: Le Bras M, Camino G, Bourbigot S, Delobel R, edi-tors. Fire retardancy of polymers. Cambridge: The Royal Society ofChemistry; 1998. p. 203–21.

[242] Kashiwagi T, Harris Jr RH, Zhang X, Briber RM, Cipriano BH, Ragha-ran SR, et al. Flame retardant mechanism of polyamide 6–claynanocomposites. Polymer 2004;45:881–91.

[243] Lee J, Giannelis EP. Synthesis and characterization of unsaturatedpolyester and phenolic resin nanocomposites. Polym Preprints1997;38:688–9.

[244] Chigwada G, Jash P, Jiang DD, Wilkie CA. Synergy between nanocom-posite formation and low levels of bromine on fire retardancy inpolystyrenes. Polym Degrad Stabil 2005;88:382–93.

[245] Chigwada G, Jash P, Jiang DD, Wilkie CA. Fire retardancy of vinyl esternanocomposites: synergy with phosphorus-based fire retardants.Polym Degrad Stabil 2005;89:85–100.

[246] Si M, Zaitsev V, Goldman M, Frenkel A, Peiffer DG, Weil E, et al. Self-extinguishing polymer/organoclay nanocomposites. Polym DegradStabil 2007;92:86–93.

[247] Zanetti M, Camino G, Canavese D, Morgan AB, Lamelas FJ, WilkieCA. Fire retardant halogen-antimony-clay synergism in polypropy-lene layered silicate nanocomposites. Chem Mater 2002;14:189–93.

[248] Tang Y, Hu Y, Wang S, Gui Z, Chen Z, Fan W. Intumescent flameretardant-montmorillonite synergism in polypropylene–layeredsilicate nanocomposites. Polym Int 2003;52:1396–400.

[249] Dabrowski F, Le Bras M, Cartier L, Bourbigot S. The use of clayin an EVA-based intumescent formulation. Comparison with theintumescent formulation using polyamide-6 clay nanocompositeas carbonisation agent. J Fire Sci 2001;19:219–41.

[250] Zheng X, Wilkie CA. Flame retardancy of polystyrene nanocompos-ites based on an oligomeric organically-modified clay containingphosphate. Polym Degrad Stabil 2003;81:539–50.

[251] Kim J, Lee K, Lee K, Bae J, Yang J, Hong S. Studies on the thermalstabilization enhancement of ABS; synergistic effect of triphenylphosphate nanocomposite, epoxy resin, and silane coupling agentmixtures. Polym Degrad Stabil 2003;79:201–7.

[252] Hussain M, Varley RJ, Mathys Z, Cheng YB, Simon GP. Effect oforgano-phosphorus and nano-clay materials on the thermal andfire performance of epoxy resins. J Appl Polym Sci 2004;91:1233–
53.
[253] Nam PH, Maiti P, Okamoto M, Kotaka T, Nakayama T, Takada M,et al. Foam processing and cellular structure of polypropylene/claynanocomposites. Polym Eng Sci 2001;42:1907–18.

[254] Karian HG. Handbook of polypropylene and polypropylene com-posites. New York: Marcel Dekker; 1999.

lymer Science 33 (2008) 1119–1198

[255] Di Maio E, Iannace S, Sorrentini L, Nicolais L. Isothermal crystal-lization in PCL/clay nanocomposites investigated with thermal andrheometric methods. Polymer 2004;45:8893–900.

[256] Gelfer MY, Burger C, Chu B, Hsiao BS, Drozdov AD, Si M,et al. Relationships between structure and rheology in modelnanocomposites of ethylene-vinyl-based copolymers and organ-oclays. Macromolecules 2005;38:3765–75.

[257] Maiti P, Okamoto M. Crystallization controlled by silicate sur-faces in nylon 6–clay nanocomposites. Macromol Mater Eng2003;288:440–5.

[258] Tetto JA, Steeves DM, Welsh EA, Powell BE. Biodegradable poly(�-caprolactone)/clay nanocomposites. Ann Tech Confr Soc Plast Eng1999:1628–32.

[259] Lee SR, Park HM, Lim H, Kang T, Li X, Cho WJ, et al. Microstructure,tensile properties, and biodegradability of aliphatic polyester/claynanocomposites. Polymer 2002;43:2495–500.

[260] Ray SS, Yamada K, Okamoto M, Fujimoto Y, Ogami A, Ueda K. Newpolylactide/layered silicate nanocomposites. 5. Designing of mate-rials with desired properties. Polymer 2003;44:6633–46.

[261] Huaili Q, Chungui Z, Shimin Z, Guangming C, Mingshu Y. Photo-oxidative degradation of polyethylene/montmorillonite nanocom-posite. Polym Degrad Stabil 2003;81:497–500.

[262] Morlat S, Mailhot B, Gonzalez D, Gardett J. Photo-oxidationof polypropylene/montmorillonite nanocomposites. Chem Mater2004;16:377–83.

[263] Mailhot B, Morlat S, Gardett J, Boucard S, Duchet J, Gerard J. Pho-todegradation of polypropylene nanocomposites. Polym DegradStabil 2003;82:163–7.

[264] Patterson PH, Sloan JM, Hsieh AJ. Photo degradation mechanismsof layered silicate/polycarbonate nanocomposites. Ann Tech ConfrSoc Plast Eng 2002:3936–8.

[265] Sloan JM, Patterson P, Hsieh A. Mechanisms of photo degradationfor layered silicate–polycarbonate nanocomposites. Polym MaterSci Eng 2003;88:354–5.

[266] Schmidt G, Malwitz MM. Properties of polymer–nanoparticle com-posites. Curr Opin Colloid Interface Sci 2003;8:103–8.

[267] Schmidt D, Shah D, Giannelis EP. New advances in poly-mer/layered silicate nanocomposites. Curr Opin Solid State Mater2002;6:205–12.

[268] Kim JH, Koo CM, Choi YS, Wang KH, Chung IJ. Preparation andcharacterization of polypropylene/layered silicate nanocompositesusing an antioxidant. Polymer 2004;45:7719–27.

[269] Gacitua WE, Ballerini AA, Zhang J. Polymer nanocomposites: syn-thetic and natural fillers. A review. Maderas Ciencia y tecnología2005;7:159–78.

[270] Zeng QH, Yu AB, Lu GQ, Paul DR. Clay-based polymer nanocompos-ites: research and commercial development. J Nanosci Nanotechnol2005;5:1574–92.

[271] Leaversuch R. Nanocomposites broaden roles in automotive, bar-rier packaging. http://www.ptonline.com/articles/200110fa3.html;2001.

[272] Rana HT, Gupta RK, GangaRao HVS, Sridhar LN. Measurementof moisture diffusivity through layered-silicate nanocomposites.Mater Interfaces Electrochem Phenom 2005;51:3249–56.

http://www.ptonline.com/articles/200110fa3.html

A review on polymer–layered silicate nanocomposites

Documents

papaspyrides progress

john wiley

situ intercalative polymerization

degradation problems encountered

low clay loadings

heat distortion temperature

polymerlayered silicate nanocomposites

twin screw extruder