Morphology And Rheology Of Polystyrene Nano Composites Based Upon Clay

57

Macromol. Rapid Commun. 21, 57–61 (2000)

A correlation between morphology development and rheology of polystyrene nanocomposites based upon organophilic layered silicates (organoclay) such as fluoromicas was found as a function of the silicate modification. Organoclay was obtained by means of ion exchange of clay with protonated— amine-terminated polystyrenes with molar mass of Mn = 121 and 5 800 g/mol. Only when applying shear forces during melt compounding of organoclay modified with high molar mass polystyrene (PS), individual silicate platelets of 1 nm diameter and 600 nm length were obtained. Dispersions of such in-situ formed nanoparticles with aspect ratio of 600 accounted for unique elastic properties observed in the low frequency range of the dynamic modulus, whereas organoclay modified with low molecular PS did not exfoliate and exhibited rheological behavior very similar to that of conventional fillers.

TEM micrograph of an exfoliated PS nanocomposite. Silicate loading: 5 wt.-%

Communication

Morphology and rheology of polystyrene nanocomposites based upon organoclay Botho Hoffmann, Christoph Dietricha, Ralf Thomann, Christian Friedrich*, Rolf Mu¨lhaupt Freiburger Materialforschungszentrum und Institut fu¨r Makromolekulare Chemie, Stefan-Meier-Str. 21, D-79104 Freiburg i. Br., Germany [email protected] (Received: August 18, 1999; revised: October 04, 1999)

Introduction It is well established that effective dispersion of anisotropic particles with high aspect ratio such as short fibers, platelets and whiskers within the continuous polymer matrix in combination with adequate interfacial adhesion between filler and polymer can account for substantially improved reinforcement of the polymer matrix1). Pioneering advances at Toyota research2) during the early 1990’s has stimulated the development of various polymer/organoclay nanocomposites with attractive property profiles such as improved stiffness combined with elevated dimensional stability, barrier resistance, improved thermal stability and inherent flame retardency3–8). Several synthetic routes were examined to produce polystyrene/organoclay nanocomposites, aiming at improving exfoliation of orgaa

Current address: Freiburger Materialforschungszentrum und Institut fu¨r Mikrosystemtechnik, Stefan-Meier-Str. 21; D-79104 Freiburg i. Br., Germany.

Macromol. Rapid Commun. 21, No. 1

noclay. Moet and coworkers5, 9–11) reported in-situ bulk and solution polymerization of styrene using coreactive organophilic montmorillonite, obtained via ion exchange of sodium montmorillonite with vinylbenzyltrimethyl ammonium chloride, in order to achieve interfacial grafting of polystyrene onto clay and to promote swelling of clay in styrene and various solvents. In-situ bulk and solution styrene polymerization as well as compounding of organoclay in polystyrene melts12, 13), including the use of poly(styrene-co-vinyloxazoline)14) and polystyrene-blockpoly(ethylene oxide)15) as compatibilizers, afforded intercalation but failed to achieve complete exfoliation. In aqueous emulsion polymerization processes water swelling of clays was combined with intergallery polymerization to improve exfoliation16, 17). In a recent advance Sogah and coworkers18) converted clay into initiators for TEMPOmediated controlled radical polymerization, thus improving control of polystyrene grafting onto silicates and substantially improving exfoliation. In the past incomplete

i WILEY-VCH Verlag GmbH, D-69451 Weinheim 2000

1022-1336/2000/0101–0057$17.50+.50/0

58

B. Hoffmann, Ch. Dietrich, R. Thomann, Chr. Friedrich, R. Mu¨lhaupt

Tab. 1.

Properties of used materials —

—

Material

Abbreviation

Function

Mn g=mol

Mw /Mn

Tg 8C

polystyrene amine-terminated polystyrene 2-phenylethylamine

PS100 AT-PS8 PEA

matrix modifier modifier

106 000 5 800 121

1.07 1.33 1

100 78 –

exfoliation and formation of complex nanoparticle mixtures has hampered investigation of rheological behavior although is was well recognized that nanocomposites exhibit unusual rheological properties. The rheological properties of in-situ polymerized nanocomposites with end-tethered polymer chains were at first described by Giannelis et al.19, 20) The flow behavior of poly(e-caprolactone) and polyamide-6 nanocomposites differed extremely from that of the neat matrices, whereas the thermorheological properties (Arrhenius activation energy of flow) of the composites were entirely determined by that behavior of the matrix20). The slope of the storage modulus G9 and the loss modulus G99 versus the frequency x in the terminal region was smaller than 2 and 1, respectively. Values of 2 and 1 are expected for melts of linear monodisperse polymers and large deviations, especially for small amounts of silicate loading in the percentage range may be indicative of network formation. However, such nanocomposites based on the in-situ polymerization technique exhibit fairly broad molar mass distribution of the polymer matrix which hides the structure relevant information and impedes the interpretation of the results. Therefore, quantitative correlations between the viscoelasticity and the morphology of the composites cannot be established. For a better understanding of the relationship of viscoelasticity and structure well defined model materials are necessary. From early research by Weiss21) and Vansant22) it is well known that the interlayer distance of organoclays increases with increasing chain length of alkyl groups used as substituents of alkyl ammonium cations, which represent typical modifiers of organoclays. Similar to observations by Okada23), who used x-amino acids as modifiers, interlayer distance can increase from 0.95 nm up to 2 nm when using such small molecular weight modifiers. Also high molecular weight modifiers have been used. For example, an industrial grade difunctional amine-terminated poly(butadieneco-acrylonitrile) with number-average molecular weight — — — (Mn) of 1 090 g/mol and Mw /Mn = 6.75 was applied as modifier5). Although intercalation was observed, all these modifiers failed to afford complete exfoliation. In our research we have used polystyrenes with narrow molecular weight distributions as continuous matrix and as ammonium-functional modifier of organoclay in order to establish correlations between morphology development and rheological behavior of polystyrene/organoclay nanocomposites.

Experimental part Materials The synthetic layered silicate SOMASIFTM ME100 (ME100), which represents a fluromica and was prepared by heating talcum in the presence of Na2SiF6 , was supplied by CO-OP Ltd., Japan. The negative charge of the layers is compensated by Na+-ions in the interlayer galleries. The cation exchange capacity (CEC) was in a range of 0.7 to 0.8 meq/g for ME100. The interlayer spacing calculated from the wide angle X-ray scattering (WAXS) d001-reflection was 0.95 nm. The cation exchange was performed with two different swelling agents: 2-phenylethylamine (PEA, — Mn = 121 g/mol) and amine-terminated polystyrene (AT-PS8) — with Mn = 5 800 g/mol and degree of polymerization of 56, which was prepared by means of anionic polymerization24). Polystyrene (PS100), prepared by means of anionic polymerization, was used as continuous matrix. The properties of the polystyrenes are listed in Tab. 1. Swelling of the layered silicate A mixture of 5.6 mmol/l swelling agent (AT-PS8 or PEA, respectively) and 6.7 mmol/l hydrochloric acid were suspended in a THF/water (4 vol.-%/1 vol.-%) mixture at 40 8C. Then 8 g/l of layered silicate was added to the hot solution and white precipitates were obtained. The precipitates were filtered after 30 min and washed twice: first with a hot THF/ water mixture and thereafter with hot water. The modified layered silicate was dried in vacuum at 45 8C for 72 h. Fluoromica modified with AT-PS8 will be denoted as organoclayPS, and fluoromica modified with 2-phenylethylamine will be denoted organoclay-PEA. Compounding The composites were obtained by compounding the organoclays with PS100 at 200 8C in a DACA Microcompounder for 5 min. The compound of PS100 with organoclay-PS is called C-PS, the compound of PS100 with organoclay-PEA is called C-PEA. The overall silicate loading was 5 wt.-% for the composites. Morphological characterization Ultrathin sections of the composites with a thickness of approximately 50 nm were prepared with an ultramicrotome (Ultracut E, Reichert & Jung) equipped with a diamond knife. Transmission electron microscopy (TEM) was carried out with a Zeiss CEM 912 transmission electron microscope using an acceleration voltage of 120 keV. Due to the high electron density differences between silicate and polymer staining of the samples was not necessary.

Morphology and rheology of polystyrene nanocomposites based upon organoclay

WAXS measurements The interlayer distance of the organoclays was studied by means of wide angle X-ray scattering (WAXS) using a Siemens D500 apparatus with CuKa radiation (k = 0.154 nm) and a scanning rate of 0.3 8/min. Rheological characterization The neat PS100 and the composites C-PS and C-PEA were dried for 48 h at 60 8C in vacuum and molten for 20 min under vacuum at a temperature of 200 8C in a Collin PCS2 vacuum press. The samples were pressed for 10 min and subsequently quenched to ambient temperature. The plates had a diameter of 25 mm and a thickness of approximately 1 mm. For the rheological measurements the Rheometrics RMS 800, a strain controlled rheometer, with parallel plate geometry was used. All measurements were carried out under nitrogen. The strain region in which the material can be regarded as linear viscoelastic was determined by amplitude sweep experiments and is in the order of 10%. Isothermal frequency sweeps were taken at different temperatures ranging from 140 8C to 220 8C. The master curves were shifted with the program LSSHIFT18). The temperature of 180 8C was chosen as reference temperature.

Results and discussion The synthetic fluoromica (ME100), which represents a hectorite obtained by heating talcum with Na2SiF6 , was rendered organophilic by means of ion exchange of the sodium intergallery cations for protonated amine-termiScheme 1:

Fig. 1. WAXS traces of neat ME100 (a), organoclay-PEA (b), and organoclay-PS

nated polystyrene (AT-PS8, Tab. 1) with number average molecular weight of 5 800, which corresponds to a degree of polymerization of 56, and 2-phenylethylamine (PEA), which corresponds to a polystyrene degree of polymerization of 1. As illustrated in Scheme 1, the ammonium-terminated polystyrene modifier was prepared by means of anionic polymerization followed by chain termination with dimethylchlorosilane, silylation with allyl amine and protonation of the resulting amine-terminated polystyrene. Fig. 1 shows the WAXS traces of neat layered silicate (ME100), ME100 modified with PEA (organoclay-PEA) and ME100 modified with AT-PS8 (organoclay-PS). The

Synthetic route for the preparation of AT-PS8 and organoclay-PS

59

60

B. Hoffmann, Ch. Dietrich, R. Thomann, Chr. Friedrich, R. Mu¨lhaupt

Fig. 3. G9 versus reduced frequency for PS100 matrix (black line), C-PS (F), and C-PEA (9). Silicate loading: 5 wt.-%

—

Fig. 2. TEM micrographs of the melt compounds C-PEA (a) and C-PS (b). Silicate loading: 5 wt.-%

interlayer distances of the organoclays were obtained from the peak position (d001-reflection) of WAXS traces. The d001-reflection for neat ME100 was found at a 2H value of 9.3 8, which corresponds to an interlayer distance of 0.95 nm (Fig. 1, trace a). Organoclay modification afforded substantially increased interlayer distances. The reflection of organoclay-PEA (trace b in Fig. 1) was found at 2H of 6 8, corresponding to an interlayer distance of 1.4 nm. In contrast, organoclay-PS (trace c in Fig. 1) exhibited no d001-reflection in the relevant angle region, thus indicating the existence of interlayer layer distances larger than 4 nm! Transmission electron microscopy (TEM) was performed to examine morphology development of composites, abbreviated as C, prepared by means of melt com-

pounding polystyrene with Mn = 106 000 in the presence of organoclay-PEA (compound C-PEA) or organoclay-PS (compound C-PS). The overall silicate loading was 5 wt.-% for both composites. The TEM micrographs are displayed in Fig. 2 where the dark lines represent the silicate layers in the polystyrene matrix (bright). From Fig. 2 (a) it is apparent that in the case of compound CPEA the PEA modification caused intercalation but failed to exfoliate the intercalated particles. In sharp contrast, compound C-PS exhibited a very fine dispersion of individual platelets being composed of individual silicate layers. Typically such layers were 600 nm long and 1 nm thick. From TEM tilting experiments the layers width was estimated to be approximately 100 nm. Obviously, the large interlayer distance in organoclay-PS (larger than 4 nm) was accompanied by weakening of the silicate interlayer interactions and afforded complete exfoliation when applying shear forces during melt compounding with matrix polystyrene (Fig. 2 (b)). Rheological properties were determined to examine the presence of particle network formation via interparticle interactions and self-assembly. We found that time temperature superposition (TTS) holds for all investigated samples. There was no influence of a silicate loading of 5 wt.-% on TTS because of the high compatibility between organophilic modified ME100 and PS100. Fig. 3 shows the storage modulus (G9) versus the reduced frequency (x aT) for PS100 (black line), C-PS (F), and C-PEA (9). PS100 showed the normal material response of a narrow distributed thermoplastic polymer. The curve for C-PEA was slightly shifted to higher moduli values, as expected for the filler effect. The slope in the terminal regime was approximately 2 for both curves. The material response of C-PS was found to be very different from that of CPEA. At the lowest realized frequencies, which correspond to the marked region III, strongly increased values

Morphology and rheology of polystyrene nanocomposites based upon organoclay

of storage modulus were found and the slope of the curve approached zero. Such a behavior is an indication of network formation involving assembly of individual platelets being composed of silicate layers. In the regime of intermediate frequencies (region II in Fig. 3) the moduli values were reduced in comparison of that of the neat matrix. This might be a dilution effect caused by the amine-terminated polystyrene having a molecular weight lower than the entanglement molecular weight of polystyrene. A validation of the thinning effect will be described elsewhere25). At high frequencies (region I) an increase in the storage modulus was detected with respect to that of the neat matrix. This modulus increase in this frequency range was independent of the state of dispersion and represents the combination of elasticity of both components, the matrix and the silicate.

Acknowledgement: The authors gratefully acknowledge financial support by the Sonderforschungsbereich SFB 428 of the Deutsche Forschungsgemeinschaft, and thank K. Cra¨mer for the synthesis of At-758.

1) 2)

3) 4)

5) 6) 7)

Conclusions Organophilic modification of clay with amine-terminated polystyrene, obtained by means of anionic polymerization, affords completely exfoliated polystyrene nanocomposites containing silicate nanoparticles with an aspect ratio exceeding 600, when such organoclays were melt compounded together with polystyrene. In contrast, small molecular weight modifiers, such as 2-phenylethylamine, only promoted intercalation and failed to exfoliate the silicate particles during melt compounding. Since both the polystyrene continuous matrix and the ammonium-terminated polystyrene attached to the silicate nanoparticle surface exhibited very narrow molecular weight distributions it became possible to examine rheological behavior and to correlate superstructure formation with rheological properties. Only when high molecular weight polystyrene was grafted onto the dispersed silicate layers, rheological investigation revealed the formation of networks via assembly of in-situ formed silicate nanoparticles. It should be possible to control silicate/polymer interaction of such “hairy” silicate nanoparticles by varying the chain length of polystyrenes which are attached to the silicate nanoparticles surface via their ammonium end group. Fully exfoliated nanocomposites based upon organoclay and polystyrene represent very effective model systems to achieve better understanding of morphology development, processing and structure/property correlations of nanocomposites based upon layered silicates.

8) 9) 10)

11) 12) 13) 14) 15) 16) 17) 18) 19) 20) 21) 22) 23)

24) 25)

H. S. Katz, J. V. Milewski, Eds., Handbook of Fillers for Plastics, Van Nostrand Reinhold Publ., New York 1987 G. Lagaly, “Smectic Clays as Ionic Macromolecules” in: Developments in Ionic Polymers, A. D. Wilson, H. T. Posser, Eds., Applied Science Publishers, London 1986, p. 77 ff. A. Okada, Y. Kojima, M. Kawasumi, Y. Fukushima, T. Kurauchi, O. Kamigaito, J. Mater. Res. 8, 1179 (1993) A. Akelah, Polymers and Other Advanced Materials, P. N. Prasad, J. E. Mark, T. J. Fai, Eds., Plenum Press, New York 1995, p. 625 ff. A. Akelah, N. Salahuddin, A. Hiltner, E. Baer, A. Moet, Nanostruct. Mater. 4, 965 (1994) E. P. Giannelis, Adv. Mater. 8, 29 (1996) E. P. Giannelis, Appl. Organomet. Chem. 3(5), 490 (1998) C. Zilg, P. Reichert, F. Dietsche, T. Engelhardt, R. Mu¨lhaupt, Kunststoffe 88, 1812 (1998) A. Moet, A. Akelah, Mater. Lett. 18, 97 (1993) A. Moet, A. Akelah, A. Hiltner, E. Baer, Molecularly Designed Ultrafine/Nanostructured Materials, K. E. Gonsalves, G. M. Chow, T. D. Xiao, R. C. Cammarata, Eds., MRS Symp. Proc. 351, 91 (1994) A. Akelah, A. Moet, J. Mater. Sci. 31(13), 3589 (1996) R. A. Vaia, K. D. Jandt, E. J. Kramer, E. P. Giannelis, Macromolecules 28, 8080 (1995) M. Sikka, L. N. Cerini, S. S. Ghosh, K. I. Winey, J. Polym. Sci., Part B: Polym. Phys. 34(8), 1443 (1996) N. Hasegawa, H. Okamoto, M. Kawasumi, A. Usuki, Polym. Mater. Sci Eng. 80, 353 (1999) H. R. Fischer, L. H. Gielgens, T. P. M. Koster, Acta Polym. 50, 122 (1999) M. Laus, M. Camerani, M. Lelli, K. Sparnacci, F. Sandrolini, F. Francesangeli, J. Mater. Sci. 33(11), 2883 (1998) M. W. Noh, D. C. Lee, Polym. Bull. (Berlin) 42, 619 (1999) M. W. Weimer, H. Chen, E. P., Giannelis, D. Y. Sogah, J. Am. Chem. Soc. 121, 1615 (1999) R. Krishnamoorti, E. P. Giannelis, Macromolecules 30, 4097 (1997) E. P. Giannelis, R. Krishnamoorti, E. Manias, Adv. Polym. Sci. 138, 107 (1999) A. Weiss, Clays Clay Min. 10, 191 (1963) E. F. Vansant, J. B. Uytterhoeven, Clays Clay Min. 20, 47 (1972) A. Okada, A. Usuki, T. Kurauchi, O. Kamigaito, Hybrid Organic-Inorganic Composites, J. E. Mark, C. Y. Lee, P. A. Bianci, Eds., ACS Symp. Ser. 585, 55 (1995) K. Cra¨mer, H. A. Schneider, New Polymeric Mater 4, 153 (1994) B. Hoffmann, C. Dietrich, Ch. Friedrich, J. Rheol., submitted

61