ABS/clay nanocomposites obtained by a solution technique: Influence of clay organic modifiers

Available online at www.sciencedirect.com

Polymer Degradation and Stability 92 (2007) 2206e2213www.elsevier.com/locate/polydegstab

ABS/clay nanocomposites obtained by a solution technique:Influence of clay organic modifiers

M. Modesti a,*, S. Besco a, A. Lorenzetti a, V. Causin b, C. Marega b, J.W. Gilman c,D.M. Fox d, P.C. Trulove e, H.C. De Long f, M. Zammarano c

a Department of Chemical Process Engineering, University of Padova, Via Marzolo 9, 35131 Padova, Italyb Department of Chemical Sciences, University of Padova, Via Marzolo 1, 35131 Padova, Italy

c Building and Fire Research Laboratory, National Institute of Standards and Technology, Gaithersburg, MD 20899, USAd Department of Chemistry, American University, Washington, DC 20016-8014, USA

e Chemistry Department, US Naval Academy 572M Holloway Road, Annapolis, MD 21402-5026, USAf Directorate of Chemistry and Life Sciences, Air Force Office of Scientific Research, Arlington, VA 22203-1768, USA

Received 10 December 2006; received in revised form 16 January 2007; accepted 20 January 2007

Available online 15 August 2007

Abstract

Acrylonitrile-butadiene-styrene (ABS) polymer/clay nanocomposites were produced using an intercalationeadsorption technique from poly-mer in solution: polymer/clay suspensions were subjected to ultrasonic processing to increase the effectiveness of mixing. Several kinds of or-ganically modified layered silicates (OMLS) were used to understand the influence of the surfactant nature on the intercalationeexfoliationmechanism. We show that only imidazolium-treated montmorillonite (DMHDIM-MMT) is stable at the processing temperature of 200 �C,used for hot-pressing, whereas alkyl-ammonium modified clays show significant degradation.

The morphology of ABS based polymer nanocomposites prepared in this work was characterized by means of wide angle X-ray diffraction(WAXD) and transmission electron microscopy (TEM). Dynamic-mechanical analysis (DMA) was used to determine the storage modulus anddamping coefficient as a function of temperature, and to investigate the correlations between mechanical properties and morphology of the nano-composites. The thermal stability was assessed by means of thermogravimetric analysis (TGA). DMA and TGA show that the nanocompositesbased on imidazolium-modified clay out-perform the nanocomposites based on quaternary-ammonium-modified clays in terms of mechanicalproperties and thermal stability.� 2007 Elsevier Ltd. All rights reserved.

Keywords: ABS; Nanocomposites; Solution; Organic modifiers

1. Introduction

The addition of small fractions of a nanometric filler toa polymer matrix has a large potential to improve the polymerproperties. Smectite-type clays with a layered structure, suchas montmorillonite (MMT) are commonly employed nanomet-ric fillers. MMT has a layered structure in which each layer is1 nm thickness and has lateral dimensions ranging from 30 nm

* Corresponding author. Tel.: þ39 049 8275541; fax: þ39 049 8275555.

E-mail address: [email protected] (M. Modesti).

0141-3910/$ - see front matter � 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.polymdegradstab.2007.01.036

to several microns [1]. The performance of polymereclaynanocomposites strongly depends on clayepolymer interac-tions, which can be improved through dispersion and exfolia-tion of the clay [2,3]. Three typologies of composites can bedefined depending on the degree of fillerematrix interactions:conventional composite, intercalated nanocomposite and exfo-liated nanocomposite. In a conventional composite the claysheets remain stacked in micrometric structures called tactoidswith no increase in the layer-to-layer distance (d-spacing) ascompared to the pristine clay; in an intercalated nanocompo-site polymer chains penetrate into the interlayer region andincrease the d-spacing; a further increase in d-spacing gives

2207M. Modesti et al. / Polymer Degradation and Stability 92 (2007) 2206e2213

formation to an exfoliated nanocomposite, characterized bya disordered and homogeneous dispersion of single clay sheets.The last system is the most desirable because it maximizes theinterfacial region between the filler and polymer matrix, witha subsequent improvement in reinforcement effect.

Acrylonitrile-butadiene-styrene (ABS) is a widely usedengineering thermoplastic owing to its desirable propertieswhich include good mechanical behaviour and chemical resis-tance. However, there are still only a few reports about thepreparation of ABS/MMT nanocomposites. Lee and co-workers [4] reported a study on ABS/clay systems obtainedby an emulsion technique while Wang and Hu [5] producedand characterized intercalatededelaminated nanocompositesusing direct melt intercalation. Stretz et al. [6] discussed thedispersion of clay particles in an ABS matrix comparing thissystem with a styrene-acrylonitrile (SAN) based nanocompo-site, where it was observed that the ABS/MMT compositeclay resides in the SAN phase of ABS and accumulates pref-erentially at the rubber particles surface. Pourabas and Raeesi[7] prepared an ABS/clay nanocomposite developing a sol-vent/non-solvent method and working with different kinds ofmixers while Jang and Wilkie [8] studied the effect of theclay on the thermal degradation behaviour of ABS.

The aim of this work is to investigate the effect of the claysurfactant on the morphology and performance of ABS/claynanocomposites. Several kinds of organically modified layeredsilicates were used for the preparation of nanocomposites bya solution-intercalation technique. We have focused our atten-tion on surfactants based on imidazolium and quaternaryammonium salts. The good thermal stability of imidazolium-modified clays [9] suggests their huge potential in the prepara-tion of nanocomposites suitable for elevated processing and/oroperating temperatures.

2. Experimental1

2.1. Raw materials

The ABS grade chosen (Magnum 3904, Dow Chemicals)had a melt flow index of 4.7 g/10 min (220 �C/10 kg, UNI-ISO 1133). Several OMLS were used in order to investigatethe influence of the organic surfactants on the morphologyand properties of the composites. The main chemical andphysical characteristics of OMLS are listed in Table 1.

Two commercially available OMLS were used, B108(Bentone 108 e Elementis Specialities Inc., USA) and D43B(Dellite 43B e Laviosa Chimica Mineraria, Italy). The former

1 Part of this work was carried out by the National Institute of Standards and

Technology (NIST), an agency of the US government, and by statute is not

subject to copyright in the United States. Certain commercial equipment, in-

struments, materials, services, or companies are identified in this paper in or-

der to specify adequately the experimental procedure. This in no way implies

endorsement or recommendation by NIST. The policy of NIST is to use metric

units of measurement in all its publications, and to provide statements of un-

certainty for all original measurements. In this document, however, data from

organizations outside NIST are shown, which may include measurements in

non-metric units or measurements without uncertainty statements.

derives from a natural hectorite modified with alkyl-ammoniumsalts while the latter is based on a benzyl-alkyl-ammonium saltsmodified montmorillonite. Dimethyl-hexadecyl-imidazolium-modified montmorillonite (DMHDIM-MMT) is experimentalfiller with higher thermal stability as compared to traditionalammonium salt based OMLS (Fig. 6). DMHDIM-MMTwas pre-pared by a standard cationic-exchange procedure [10]. Sodiummontmorillonite with an ion-exchange capacity of 92 meq/100 g was obtained from Southern Clay Products (Gonzales,Texas) and exchanged with 1,2-dimethyl-3-hexadecylimidazo-lium (DMHDIM) bromide in water/ethanol (1/1 volume ratio).DMHDIM was prepared and purified as previously reported [9].

2.2. Preparation of materials

ABS was dissolved in boiling acetone by mechanical stir-ring in a three neck reflux flask. When the polymer was com-pletely dissolved, a dispersion of organically modified clay inthe same solvent was added to the solution to obtain 5 and15 wt.%2 concentrations of OMLS in the polymer/clay com-posite after solvent removal.

In order to improve the mixing process, ultrasonic waves(Bransons 1510, maximum power of 70 W at 72 kHz) were ap-plied for 6 h to the dispersion of organically modified clay inacetone and for 6 h to the polymer solution after the additionof the OMLS dispersion.

The solvent was then evaporated and recovered by distilla-tion under vacuum at 50 �C; the composite was dried undervacuum at 100 �C for 4 h to remove the residues of solventfrom the solid. The polymereclay composites were compres-sion moulded with a Collin P200 press at 200 �C and 30 barduring a 600-s cycle (with a cooling rate of 0.5 �C/s) to obtainspecimens for morphological and mechanical characterization.

2.3. Wide angle X-ray diffraction (WAXD)

WAXD patterns were recorded in a 2q angular range of1.5e40� on a Philips X’Pert PRO diffractometer, working inreflection geometry and equipped with a graphite monochro-mator on the diffracted beam (CuKa radiation). The uncer-tainty in terms of d-spacings was 0.05 nm (2s).

2.4. Transmission electron microscopy (TEM)

The level and degree of dispersion was investigated by highmagnification transmission electron microscopy images (TEM,Philips mod. EM 208) using an acceleration voltage of 100 kV.Specimens were microtomed using a Leica Ultracut UCT.

2.5. Dynamic-mechanical analysis (DMA)

Dynamic-mechanical properties of the samples were mea-sured using DMA 2980 (TA Instruments). Analyses were

2 % (or wt.%) is used throughout this manuscript and is identical to mass-

fraction %.

2208 M. Modesti et al. / Polymer Degradation and Stability 92 (2007) 2206e2213

Table 1

Chemical and physical characteristics of OMLS

OMLS Structure Surfactant Residue 900 �C/air

TGA (wt.%)

d-spacing

WAXD (nm)

Dd/dPa (%)

B108 Hectorite MX(Mg6�XLiX)

Si8O20(OH)4

Dimethyl dihydrogenated tallow 65.9 2.31 50.9

D43B Montmorillonite MX(Al4�XMgX)

Si8O20(OH)4

Benzyl dimethyl

hydrogenated tallow

72.3 1.77 47.5

DMHDIM-MMT Montmorillonite MX(Al4�XMgX)

Si8O20(OH)4

Dimethyl-hexadecyl-imidazolium 70.7 1.76 46.6

a Dd/dP¼ (dOMLS� dP)/dP, where dOMLS corresponds to the d-spacing of the modified clay while dP identifies the d-spacing of pristine unmodified clay. In par-

ticular: dP, hectorite¼ 1.53 nm [26]; dP, montmorillonite¼ 1.2 nm [27]; Mx¼ exchangeable cations.

performed using a single-cantilever configuration between�100 �C and þ100 �C with a heating rate of 5 �C/min, fre-quency of 1 Hz and amplitude of 15 mm. Glass transition tem-peratures have been evaluated from the peaks of tan d functionwith an uncertainty of 1.5 �C (2s) as determined runninga polystyrene standard sample five times.

2.6. Thermal gravimetric analysis (TGA)

The thermal stability of the ABS matrix and the polymereclay composites were studied on a TA Instruments Q5000 an-alyzer operating from ambient temperature to 1000 �C ata heating rate of 20 �C/min under nitrogen and air atmo-spheres. Isothermal tests at 200 �C in air were performed in or-der to investigate the thermal stability of the pristine OMLS inan oxidative environment. An uncertainty of 0.1 wt.% (2s)was determined running five replicates of a standard calciumoxalate sample.

3. Results and discussion

3.1. Morphology

The dispersion of the clay in the polymeric matrix wasstudied by WAXD and TEM. A summary of the original

Fig. 1. WAXD patterns for ABS/D43B composites.

d-spacings for all the pristine OMLS used is reported inTable 1. It can be observed that the OMLS denoted as B108is characterized by the largest increase in d-spacing withrespect to the pristine natural clay (hectorite); this is probablydue to its high content of organic surfactant as pointed out bythe low residue (approximately 66 wt.%) reported in Table 1.

WAXD data (Figs. 1e3) show that OMLS exhibit an in-crease in d-spacing in all composites, suggesting the formationof intercalated nanocomposites. Up to three orders of basal re-flections can be detected. This is consistent with a well orderedsystem of stacked clay layers. It is quite interesting to note thatthe interlayer d-spacing in the intercalated tactoids is indepen-dent of the structure of the surfactants and clay loadings. Inall the samples the d-spacing has been shifted to values ofapproximately 3 nm. However, the extent of intercalation ismaximum for DMHDIM-MMT as shown by the increase ind-spacing observed for the nanocomposites as compared tothe pristine clays (Table 2). It is generally reported that a largequantity of nanometric filler prevents a complete intercalation/exfoliation leading to the formation of residue tactoids[11,12]. Despite this, a slight increase of 0.2 nm in d-spacingis observed when the clay loading is increased from 5 to15 wt.% in ABS/DMHDIM-MMT composites.

Fig. 2. WAXD patterns for ABS/B108 composites.

2209M. Modesti et al. / Polymer Degradation and Stability 92 (2007) 2206e2213

The degree of dispersion has been investigated for the ABS/OMLS composites with TEM. As shown in Fig. 4, stacks offew intercalated clay layers were generally observed for allthe materials. However, a greater degree of dispersion is re-vealed by the presence of several single delaminated particlesfor ABS/DMHDIM-MMT sample.

Significant effects of layered silicates on the mechanicalproperties, even in the absence of extensive exfoliation, havealready been reported [3,13,14]. Intercalation, together witha reduction in the thickness of the pristine clay tactoids con-taining hundreds of stacked layers, can be sufficient to enlargethe interfacial region, permitting the exploitation of the filler’sreinforcing effect [15e17].

As observed in previous studies [6,18,19] the clay particlesin the ABS/OMLS nanocomposites reside in the SAN matrixphase with some accumulation at the rubber particle surfaces.There are numerous influences, thermodynamic and kinetic,which could cause the clay particles to reside at the rubber par-ticle surface. The addition of a solid S to a blend AeB of twoimmiscible polymers can stabilize the system, i.e. the solidacts as a compatibilizer by adsorbing A and/or B on its surface[20]. Evidently, the stabilizing energy gain originates from theadsorption of polymeric components on the solid surface. Theobvious consequence is that the solid particles must reside atthe interface between the two polymeric phases.

In particular, the presence of rubbery particles in the micronrange (about 0.2 mm for emulsion made rubbers [4]) might

Table 2

Effect of organic surfactant and clay loading on the extent of intercalation as

determined by WAXD

Composite OMLS content (wt.%) Dd (A)

ABS/B108 5 8.3

15 8.6

ABS/D43B 5 11.3

15 11.4

ABS/DMHDIM-MMT 5 11.7

15 13.7

Fig. 3. WAXD patterns for ABS/DMHDIM-MMT composites.

interfere with clay orientation as previously observed by Stretzet al. [6]: in their comparative study they observed that SAN/MMT and ABS/MMT nanocomposites have very similar mor-phologies even if TEM analyses shows that the efficiency ofparticles orientation does not appear to be as great for theABS/MMT composites because of the interference of the rub-ber particles.

Fig. 4. TEM images of ABS/OMLS composites: ABS/DMHDIM-MMT

5 wt.% (a), ABS/D43B 5 wt.% (b), ABS/B108 5 wt.% (c).

2210 M. Modesti et al. / Polymer Degradation and Stability 92 (2007) 2206e2213

118.65°C

-89.06°C0.0

0.5

1.0

1.5

2.0

Tan Delta

0

1000

2000

3000

4000

Stor

age

Mod

ulus

(MPa

)

-150 -100 -50 0 50 100 150

Temperature (°C)

ABS/B108 (5 wt. %)ABS/D43B (5 wt. %)ABS/DMHDIM-MMT (5 wt. %)

Pristine ABS

Universal V4.2ETA Instruments

Fig. 5. Dynamic-mechanical properties of pure ABS and polymer/OMLS (5 wt.%). Tan d function is referred to unfilled polymer.

3.2. Dynamic-mechanical properties

In order to obtain more information on the polymer/claynanocomposites, all the materials were compared using dy-namic-mechanical (DMA) tests. It is well known that for filledpolymers, DMA can yield information on clay dispersion,fillerematrix and fillerefiller interactions [12,21].

In particular, the elastic modulus of polymer based nano-composites is strongly influenced by the aspect ratio of the dis-persed organic-modified clay: high aspect ratios are related toan increase of the contact surface between matrix and fillerand hence to an enhanced reinforcing action.

It is evident (Fig. 5) that in all nanocomposites the storagemodulus is slightly higher with respect to the neat polymer inthe entire temperature range. The reinforcing effect is influ-enced by the OMLS nature. ABS/DMHDIM-MMT 5 wt.%shows, by far, the largest increase in stiffness. This is probablydue to the higher extent of delamination observed in thisformulation.

The pristine polymer exhibits two glass transition tempera-tures (Fig. 5): a low temperature transition (Tg1), in the rub-bery phase, and a high temperature transition (Tg2), typicalof the rigid SAN matrix.

The glass transition temperatures of the polymer (Table 3)are not affected by the presence of OMLS. In fact, as observedin previous studies [6], the filler acts as a reinforcing agent,limiting the chain mobility, only at temperatures below Tg2,whereas, at higher temperatures (near and beyond Tg2) the to-tal softening of the SAN phase inhibits the action exerted bythe clay.

3.3. Thermal properties

The TGA plots for OMLS and pure polymer tested at200 �C (isothermal conditions) in an oxidative environmentare shown in Fig. 6. The higher thermal stability of

DMHDIM-MMT compared to the ammonium based clays isevident. These data show that ammonium based OMLS mighthave been partially degraded by the hot-pressing procedure at200 �C for 10 min, used for the preparation of the specimens,while imidazolium based OMLS are not affected.

The TGA plots for ABS and polymer/clay composites, ac-quired in air (Figs. 7 and 8) and nitrogen (Figs. 9 and 10) ata rate of 20 �C/min, are shown. Two main steps in the degra-dation pathway of ABS and its clay composites are observed.In the first step between 300 �C and 450 �C, the major massloss occurs. It is attributed to the evolution of volatile productsderiving from polybutadiene followed closely by the aromaticsof the styrenic fraction, that lead to the formation of a charredresidue [22]. The second step occurs above 450 �C and it is as-signed to the degradation of the carbonaceous products formedduring the first step. No significant increase in the charred res-idue at 500 �C in air or nitrogen is observed for the nanocom-posites as compared to the neat ABS once the inorganiccontent of the OMLS in the residue is accounted for. At tem-peratures higher than 600 �C the experimental residue is equalto the calculated inorganic content of the OMLS. A similar de-composition pathway is observed both in oxidative (Figs. 7and 8) and inert (Figs. 9 and 10) environment. Thus, the pres-ence of oxygen is not a main factor in the decomposition ofABS and ABS nanocomposites.

The onset of thermal degradation occurs at a lower temper-ature in presence of OMLS. It might be argued that the

Table 3

Glass transition temperatures for the pure polymer and ABS/OMLS nanocom-

posites as determined by the peak of tan d

Composite Tg1 (�C) Tg2 (�C)

ABS/B108 �89 118

ABS/D43B �91 118

ABS/DMHDIM-MMT �91 119

ABS (pure) �89 118

2211M. Modesti et al. / Polymer Degradation and Stability 92 (2007) 2206e2213

Fig. 6. Thermal stability of OMLS and pure polymer (isothermal conditions, 200 �C e air).

decrease in the onset is due to the partial degradation of theclay surfactant. However, the weight loss at 375 �C in nitrogenis about 2 wt.% for all nanocomposites with 5 wt.% OMLS(Fig. 9). The same weight loss in nitrogen at 375 �C and a load-ing level of 15 wt.% OMLS is 2 wt.%, 2 wt.% and 4 wt.% forABS/B108, ABS/DMHDIM-MMT and ABS/D43B, respec-tively (Fig. 10). The weight loss does not increase proportion-ally with the OMLS loading level and, thus, it cannot be dueonly to the decomposition of the organic surfactant. IsothermalTGA data at 200 �C in nitrogen for 30 min have been collectedin order to further investigate this phenomenon. The weightloss for the OMLS, neat ABS and both the experimental andcalculated weight loss for the nanocomposites are reportedin Table 4. The theoretical and experimental residues for

ABS/DMHDIM 5 wt.% are identical, therefore there is nochemicalephysical interaction between imidazolium basedOMLS and ABS that affects the thermal stability at 200 �C.Instead, the experimental residue is always lower than the cal-culated one for all ammonium based OMLS nanocompositesand in particular for ABS/B108 5 wt.%. These data suggestthat ammonium based OMLS exert a catalytic effect on thepolymer degradation. Similar results have been previously re-ported when the onset of thermal degradation for the polymeris higher than the onset of decomposition for the ammoniumbased OMLS [23]. In fact, in this case the decomposition ofthe ammonium salt via Hoffmann elimination or nucleophilicsubstitution reaction could generate moieties able to catalyzethe degradation of the polymer [24]. In particular, the Hoffman

-10

0

10

20

30

40

50

60

70

80

90

100

110

Wei

ght (

%)

ABS/B108 5 wt. %_AIRABS/D43B 5 wt. %_AIRABS/DMHDIM-MMT 5 wt. %_AIRABS_AIR

0 100 200 300 400 500 600 700 800Temperature (°C) Universal V4.2E TA Instruments

Fig. 7. Thermal stability of ABS/OMLS 5 wt.% nanocomposites (air e 20 �C/min).

2212 M. Modesti et al. / Polymer Degradation and Stability 92 (2007) 2206e2213

-10

0

10

20

30

40

50

60

70

80

90

100

110

Wei

ght (

%)

ABS/B108 15 wt. %_AIRABS/D43B 15 wt. %_AIRABS/DMHDIM-MMT 15 wt. %_AIRABS_AIR

0 100 200 300 400 500 600 700 800

Temperature (°C) Universal V4.2E TA Instruments

Fig. 8. Thermal stability of ABS/OMLS 15 wt.% nanocomposites (air e 20 �C/min).

decomposition of ammonium based OMLS gives protonatedlayered silicates that act as a protonic acid catalyst [25]. Inthis work the DMHDIM-MMT is the only OMLS used thatis stable up to 200 �C. Thus, a partial degradation of theABS matrix in the ammonium based nanocomposites mighthave occurred because all samples were exposed to a tempera-ture of 200 �C for 10 min during hot-pressing. This, togetherwith the extent of exfoliation previously mentioned, mightbe another possible explanation for the superior reinforcementin terms of elastic modulus observed in imidazolium basednanocomposites as compared to ammonium based OMLS.

4. Conclusions

ABS/clay nanocomposites were obtained using OMLSwith different kinds of organic surfactants in a solution pro-cess. The behaviour of alkyl-ammonium (B108), benzyl-alkyl-ammonium (D43B) and alkyl-imidazolium based OMLS aftertheir dispersion in the polymer was investigated through mor-phological (WAXD, TEM), dynamic-mechanical (DMA) andthermal analyses (TGA).

WAXD showed the presence of intercalated tactoids char-acterized by an average interlayer spacing of about 3 nm in

-10

10

0

20

30

40

50

60

70

80

90

100

110

Wei

ght (

%)

ABS/B108 5 wt. %_NITROGENABS/D43B 5 wt. %_NITROGENABS/DMHDIM-MMT 5 wt. %_NITROGENABS_NITROGEN

0 100 200 300 400 500 600 700 800

Temperature (°C) Universal V4.2E TA Instruments

Fig. 9. Thermal stability of ABS/OMLS 5 wt.% nanocomposites (nitrogen e 20 �C/min).

2213M. Modesti et al. / Polymer Degradation and Stability 92 (2007) 2206e2213

-10

10

30

50

70

90

110

Wei

ght (

%)

ABS/B108 15 wt. %_NITROGENABS/D43B 15 wt. %_NITROGENABS/DMHDIM-MMT 15 wt. %_NITROGENABS_NITROGEN

0 100 200 300 400 500 600 700 800Temperature (°C) Universal V4.2E TA Instruments

Fig. 10. Thermal stability of ABS/OMLS 15 wt.% nanocomposites (nitrogen e 20 �C/min).

all the OMLS, independently of the amount of filler (5 and15 wt.%). As observed by TEM, single delaminated lamellaewere also present and the highest extent of exfoliation amongthe prepared composites was obtained with ABS/DMHDIM-MMT.

All OMLS exerted a reinforcing action on the polymermatrix in terms of stiffness. This effect was particularly evi-dent for ABS/DMHDIM 5 wt.% composite that showed anincrease of about 40% at 25 �C in the storage modulus as com-pared with the pristine polymer. The superior performance ofDMHDIM-MMT composite might be due to both the level ofdispersion and the thermal stability. In fact, TGA data showedthat the processing temperature influences the behaviour of theclays modified with quaternary ammonium salts that mightdegrade and catalyze the decomposition of ABS, whereas inimidazolium based nanocomposites no catalytic decomposi-tion is observed.

Table 4

Comparison between real and theoretical residue of polymer composites

(isothermal conditions e 1800 s at 200 �C e nitrogen)

Residue after 1800 s

isothermal

200 �C e N2 [wt.%]

Theoretical residuea

after 1800 s isothermal

200 �C e N2 [wt.%]

ABS/DMHDIM 5 wt.% 99.6 99.6

ABS/D43B 5 wt.% 98.5 99.3

ABS/B108 5 wt.% 96.8 99.5

ABS (pristine) 99.6 e

DMHDIM 98.9 e

D43B 93.0 eB108 98.5 e

a Taking into account the values of the experimental weight loss for the pure

polymer (EWLABS) and clay (EWLOMLS) reported in the table, the theoretical

residue has been calculated as follow: theoretical residue

[wt.%]¼ 100� 0.95(EWLABS)� 0.05 (EWLOMLS).

References

[1] Shao W, Wang Q, Ma H. Polym Int 2005;54:336.

[2] Vaia RA, Giannelis EP. Macromolecules 1997;30:8000.

[3] Mehta S, Mirabella FM, Rufener K, Bafna A. J Appl Polym Sci

2004;92:928.

[4] Jang LW, Kang CM, Lee DC. J Polym Sci Part B Polym Phys

2001;39:719.

[5] Wang S, Hu Y. Polym Degrad Stab 2002;77:423.

[6] Stretz HA, Paul DR, Cassidy PE. Polymer 2005;46:3818.

[7] Pourabas B, Raeesi V. Polymer 2005;46:5533.

[8] Jang BN, Wilkie CA. Polymer 2005;46:9702.

[9] Gilman JW, Awad WH, Davis RD, Shields J, Kashiwagi T, Van der

Hart DL, et al. Chem Mater 2002;14:3776.

[10] Vaia RA, Teukolsky RK, Giannelis EP. Chem Mater 1994;6:1017.

[11] Shah RK, Paul DR. Polymer 2006;47:4075.

[12] Lim JW, Hassan A, Rahmat AR, Wahit MU. Polym Int 2006;55:204.

[13] Fasulo PD, Rodgers WR, Ottaviani RA, Hunter DL. Polym Sci Eng

2004;44:1036.

[14] Causin V, Marega C, Marigo A, Ferrara G, Idiyatullina G, Fantinel F.

Polymer 2006;47:4773.

[15] Vaia RA, Wagner HD. Mater Today 2004;7:32.

[16] Pukanszky B. Eur Polym J 2005;41:645.

[17] Eastwood E, Viswanathan S, O’Brien CP, Kumar D, Dadmun MD.

Polymer 2005;46:3957.

[18] Cheng TW, Paul DR, Keskkula H. Polymer 1992;33:1606.

[19] Bousmina M, Utracki LA, Suprakas SR, Pouliot S. Polymer

2004;45:8403.

[20] Lipatov YS. Polymer reinforcement. Canada: ChemTec Publishing; 1995.

[21] Chow WS, Mohd Ishak ZA, Karger-Kocsis J, Apostolov AA, Ishiaku US.

Polymer 2003;44:7427.

[22] Wang S, Hu Y, Song L, Wang Z, Chen Z, Fan W. Polym Degrad Stab

2002;77:423.

[23] Zammarano M. In: Morgan A, Wilkie CA. Flame retardant polymer

nanocomposites. John Wiley & Sons, in press.

[24] Xie W, Gao Z, Pan WP, Hunter D, Singh A, Vaia S. Chem Mater

2001;13:2979.

[25] Camino G, Tartaglione G, Frache A, Manferti C, Costa G. Polym Degrad

Stab 2005;90:354.

[26] http://webmineral.com/.

[27] Yu ZZ, Yang M, Zhang Q, Zhao C, Mai YW. J Polym Sci Part B Polym

Phys 2003;41:1234.