Synthesis of Poly (methyl methacrylate) Nanocomposites via Emulsion

Mar 10, 2004 - Department of Chemical Engineering, Case Western Reserve University, University Circle,. A. W. Smith, Room 111C, Cleveland, Ohio 44106...
0 downloads 0 Views 191KB Size
Download PDF
3424

Langmuir 2004, 20, 3424-3430

Synthesis of Poly(methyl methacrylate) Nanocomposites via Emulsion Polymerization Using a Zwitterionic Surfactant Paulo Meneghetti and Syed Qutubuddin* Department of Chemical Engineering, Case Western Reserve University, University Circle, A. W. Smith, Room 111C, Cleveland, Ohio 44106 Received September 12, 2003. In Final Form: December 31, 2003 The synthesis of nanocomposites via emulsion polymerization was investigated using methyl methacrylate (MMA) monomer, 10 wt % montmorillonite (MMT) clay, and a zwitterionic surfactant octadecyl dimethyl betaine (C18DMB). The particle size of the diluted polymer emulsion was about 550 nm, as determined by light scattering, while the sample without clay had a diameter of about 350 nm. The increase in the droplet size suggests that clay was present in the emulsion droplets. X-ray diffraction indicated no peak in the nanocomposites. Transmission electron microscopy showed that emulsion polymerization of MMA in the presence of C18DMB and MMT formed partially exfoliated nanocomposites. Differential scanning calorimetry showed an increase of 18 °C in the glass transition temperature (Tg) of the nanocomposites. A dynamic mechanical thermal analyzer also verified a similar Tg increase, 16 °C, for the partially exfoliated nanocomposites over poly(methyl methacrylate) (PMMA). Thermogravimetric analysis indicated a 37 °C increase in the decomposition temperature for a 20 wt % loss. A PMMA nanocomposite with 10 wt % C18DMB-MMT was also synthesized via in situ polymerization. This nanocomposite was intercalated and had a Tg 10° lower than the emulsion nanocomposite. The storage modulus of the partially exfoliated emulsion nanocomposite was superior to the intercalated structure at higher temperatures and to the pure polymer. The rubbery plateau modulus was over 30 times higher for the emulsion product versus pure PMMA. The emulsion technique produced nanocomposites of the highest molecular weight with a bimodal distribution. This reinstates that exfoliated structures have enhanced thermal and mechanical properties over intercalated hybrids.

* Author to whom correspondence should be addressed. E-mail: [email protected].

a variety of surfactants, dispersion techniques, and polymerization conditions.10-17 Emulsion polymerization has the advantage that all the ingredients are added to a single reactor, avoiding the two steps that are necessary when using an in situ polymerization technique: clay treatment and then a separate polymerization after dispersing the treated clay in the monomer. Lee and Jang10 were the first to report the synthesis of PMMA nanocomposites via emulsion polymerization, which was carried out at 70 °C for 12 h. Their results, using an anionic surfactant, sodium dodecyl sulfate (SDS), indicated an intercalated hybrid. Chen et al.11 synthesized PMMA nanocomposites with 4 wt % clay treated with a cationic surfactant via in situ polymerization at 200 °C for 30 min. The microstructure was claimed to be exfoliated, although transmission electron microscopy (TEM) showed tactoids. Also via an in situ technique, Okamoto et al.12 synthesized intercalated nanocomposites with 10 wt % organoclay. In a particular system, in which the clay was treated with methyltrioctylammonium chloride [CH3(C8H17)3N+]Cl-, the storage modulus and the glass transition temperature (Tg) were higher than those of pure PMMA by 34% and 10 °C, respectively. Bandyopadhyay et al.13,14 synthesized

(1) Messersmith, P. B.; Giannelis, E. P. Chem. Mater. 1996, 6, 1719. (2) Lan, T.; Pinnavaia, T. J. Chem. Mater. 1994, 6, 2216. (3) Usuki, A.; Kawasumi, M.; Kojima, Y.; Okada, A.; Fukushima, Y.; Kurachi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1179. (4) Kojima, Y.; Kawasumi, M.; Usuki, A.; Okada, A.; Fukushima, Y.; Kurachi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1185. (5) Vaia, R. A.; Ishii, H.; Giannelis, E. P. Chem. Mater. 1993, 5, 1694. (6) Qutubuddin, Q.; Fu, X. In Nano-Surface Chemistry; Rosoff, M., Ed.; Mercel Dekker: New York, 2002; p 653. (7) Biasci, L.; Aglieto, M.; Ruggeri, G.; Ciardelli, F. Polymer 1994, 35, 3296. (8) Kelly, P.; Akelah, A.; Qutubuddin, S.; Moet, A. J. Mater. Sci. 1994, 29, 2274. (9) Akelah, A.; Kelly, P.; Qutubuddin, S.; Moet, A. Clay Miner. 1994, 29, 169.

(10) Lee, D. C.; Jang, L. W. J. Appl. Polym. Sci. 1996, 61, 1117. (11) Chen, G.; Chen, X.; Lin, Z.; Ye, W.; Yao, K. J. Mater. Sci. Lett. 1999, 18, 1761. (12) Okamoto, M.; Morita, S.; Taguchi, H.; Kim, Y. H.; Kotaka, T.; Tateyama, H. Polymer 2000, 41, 3887. (13) Bandyopadhyay, S.; Giannelis, E.; Hsieh, A. Polym. Mater. Sci. Eng. 2000, 82, 208. (14) Bandyopadhyay, S.; Giannelis, E.; Hsieh, A. In Polymer Nanocomposites: Synthesis, Characterization, and Modeling; Krishnamoorti, R., Vaia, R., Eds.; Oxford University Press: San Francisco, 2002, p 15. (15) Huang, X.; Brittain, W. Polym. Prepr. 2000, 41, 521. (16) Huang, X.; Brittain, W. Macromolecules 2001, 34, 3255. (17) Zeng, C.; Lee, L. Macromolecules 2001, 34, 4098.

Introduction Polymer composites are widely used in applications such as transportation, construction, electronics, and consumer products. The properties of particle-reinforced polymer composites are strongly influenced by the dimensions and microstructure of the dispersed phase. Recently, there has been a growing interest in the development of polymer-clay nanocomposites.1-6 Nanocomposites constitute a new class of material that involves nanoscale dispersion in a matrix. Nanocomposites have at least one ultrafine phase dimension, typically in the range of 1-100 nm, and exhibit improved properties when compared to micro- and macrocomposites.1 Strong interfacial interactions between the dispersed clay layers and the polymer matrix lead to enhanced mechanical, thermal, and barrier properties of the virgin polymer.2-4 Because clay is hydrophilic, it is necessary to make it organophilic via cation exchange, typically with alkylammonium cations.5-9 The synthesis of poly(methyl methacrylate) (PMMA)clay nanocomposites has been reported very recently using

10.1021/la0357099 CCC: $27.50 © 2004 American Chemical Society Published on Web 03/10/2004

Synthesis of PMMA Nanocomposites

PMMA nanocomposites via emulsion polymerization at 80 °C with SDS as the emulsifier and 5 wt % montmorillonite (MMT). The nanocomposites were well dispersed and showed enhanced thermal stability, an increase in Tg by 6 °C with the MMT system, and a slightly higher molecular weight than pure PMMA. Huang and Brittain15,16 developed PMMA nanocomposite using 5 wt % MMT and a polymerizable surfactant, [2-(methacryloyloxy)ethyl]trimethylammonium chloride, via emulsion polymerization at 80 °C for 12 h. X-ray spectra showed no peaks indicating exfoliation. However, no TEM evidence was presented by Huang and Brittain. Thermal degradation for a 20 wt % loss was enhanced significantly up to 341 °C, Tg increased by 15 °C over the pure polymer, and molecular weight increased to 565 000 for the nanocomposite. Zeng and Lee17 prepared PMMA nanocomposites via in situ polymerization with 5 wt % MMT treated with a similar reactive cationic surfactant, [2-(methacryloyloxy)ethyl]hexadecyldimethylammonium bromide. The microstructure appeared to be exfoliated on the basis of TEM. Wang et al.18 prepared intercalated PMMA nanocomposites via emulsion polymerization using a polymerizable surfactant, dimethyl-n-hexadecyl-(4-vinylbenzyl)ammonium chloride. The molecular weight of the nanocomposites was lower than that of the virgin polymer. This polymerization, however, was a two-step process involving clay functionalization and polymerization of the aqueous dispersion. Previously, the PMMA nanocomposite was synthesized via in situ polymerization.19 In this paper, we describe the synthesis of the PMMA-clay nanocomposite with 10 wt % clay via emulsion polymerization at 60 °C with a focus on how the zwitterionic surfactant, octadecyl dimethyl betaine (C18DMB), affects the final product. A cationic surfactant, benzalkonium chloride (BAC), and an anionic surfactant, SDS, were also used for qualitative comparison of the emulsion product. The charge distribution in C18DMB depends on the pH of the medium. Thus, the morphology and properties of the nanocomposite will be affected by the different interactions between the surfactant and the clay depending on whether the surfactant has cationic, neutral, or anionic behavior. Experimental Section Materials. MMT used in this study was provided by ECC America, Inc. (under the trade name of Mineral Colloid BP), and contains exchangeable cations of primarily Na+. Mineral Colloid BP is a fine powder with an average particle size of 75 µm in the dry state and a cation exchange capacity (CEC) of 0.9 mol/Kg. Methyl methacrylate (MMA) was purchased from Aldrich Chemical and purified by distillation under a reduced pressure at 35 °C. The free radical initiator, 2,2-azobis(isobutyronitrile) (AIBN), was obtained from DuPont and purified by recrystallization twice in methanol. BAC and SDS were purchased from Aldrich Products, and C18DMB was synthesized in the laboratory.20 Methods. Synthesis of PMMA-Clay Nanocomposites. In Situ Nanocomposite C18DMB-MMT-PMMA. Hydrophilic MMT was ion-exchanged with C18DMB as described in a previous paper.17 The organophilic clay was dispersed in MMA, mixed by vortex, sonicated for 3 h, and allowed to swell overnight. Approximately 0.1 wt % initiator AIBN was added to the sample, which was then purged with nitrogen for 15 min and polymerized in an oil bath at 55 °C for at least 24 h. Emulsion Nanocomposite BAC-MMT-PMMA. A total of 0.04 g of BAC was added to a dispersion containing of 0.282 g of MMT (18) Wang, D.; Zhu, J.; Yao, Q.; Wilkie, C. Chem. Mater. 2002, 14, 3837. (19) Meneghetti, P.; Qutubuddin, S. Chem. Eng. Commun. 2001, 188, 81. (20) Qutubuddin, S.; Lin, C. S. Polymer 1994, 35, 4120.

Langmuir, Vol. 20, No. 8, 2004 3425 in 30 mL of deionized water. Continuous stirring was applied for 5 h, and the mixture was left to idle overnight. A total of 3 mL of MMA monomer was added along with 0.1 wt % AIBN. The mixture was purged with nitrogen for 3 min and then polymerized in an oil bath at 60 °C for 48 h with constant stirring. Emulsion Nanocomposite C18DMB-MMT-PMMA. A total of 0.04 g of C18DMB was added to a dispersion containing 0.282 g of MMT in 30 mL of deionized water. Continuous stirring was applied for 5 h at a temperature of 40 °C, and the mixture was left to idle overnight in the oil bath. A total of 3 mL of MMA monomer was added along with 0.1 wt % AIBN. The mixture was purged with nitrogen for 3 min and then polymerized in an oil bath at 60 °C for 48 h with constant stirring. Characterization of Emulsion Droplets and Dry Composites. The droplet size of polymerized emulsions was determined by using light scattering measurements. Polymerized emulsion samples with clay and without clay were prepared. For the samples with clay, the emulsion was centrifuged to remove larger particles, and then the supernatant was diluted with deionized water. For the samples without clay, the emulsion was simply diluted with deionized water. The light scattering experiments were performed on a Brookhaven Instrument (BI) Corp. spectrometer with a Spectra Physics 15-mW He/Ne laser (λ ) 632.8 nm). The intensity autocorrelation function was measured at a 30° scattering angle with a 264-channel BI 2030 AT 4-bit correlator. X-ray diffraction (XRD) measurements were made using a Philips XRG 3100 X-ray generator equipped with a Ni-filtered Cu KR (1.5418 Å) source that was connected to a Phillips APD 3520 type PW 1710 diffractometer controller. The PMMA nanocomposite samples containing BAC and C18DMB were analyzed using XRD. Scanning electron microscopy (JEOL JSM 840) and TEM (JEOL 1200 EX TEM) were used to investigate the microstructure of PMMA nanocomposites. Samples for TEM were compression-molded and cut to 80-nm-thick sections with a diamond knife. Differential scanning calorimetry (DSC) was performed on a Perkin-Elmer DSC 7 from 50 to 150 °C at a heating rate of 10 °C/min. Thermogravimetric analysis (TGA) was performed on a TA Instruments TGA 2950. Samples of 10 to 15 mg were heated to 500 °C at a rate of 20 °C/min under a nitrogen atmosphere. The samples used in DSC and TGA were the dry powder from the emulsion and were not compression-molded. Dynamic mechanical properties were measured using a dynamic mechanical thermal analyzer (DMTA) from Rheometric Scientific. The samples (0.2 × 10 × 25 mm3) were swept at 3 °C min-1 from 25 to 170 °C at a frequency of 1 Hz. The storage modulus (E′), the loss modulus (E′′), and the mechanical loss factor (tan δ) were determined. Molecular weights analysis was performed by gel permeation chromatography (GPC) with a Varian model 350 RI detector and a PSS-SDV column (8 × 300 mm and pore size of 5 µm). All measurements were carried out at a flow rate of 1 mL/min and at 25 °C. Polystyrene standards were used for calibration. The polymer nanocomposites were dissolved in tetrahydrofuran (THF), filtered, and used for GPC measurements.

Results and Discussion Surfactant Effect. The clay, MMT, is hydrophilic; however, it contains sodium and calcium cations that can be ion-exchanged with organic surfactant cations. In an aqueous medium, the ion exchange takes place in the inner Helmholtz plane of the silica surface21 when a cationic surfactant is used. A cationic surfactant is expected to provide stronger interaction with the clay than any other surfactant (nonionic or anionic) because it will replace the sodium or calcium ions. In an emulsion, the surfactant type will affect the final product. The amount of surfactant used was about one-third of the CEC of MMT, to avoid coagulation and reduce foaming of the emulsion. The emulsion polymerization with cationic surfactant BAC resulted in the formation of small pellets of about (21) Dekany, I.; Szekeres, M.; Marosi, T.; Balazs, J.; Tombacz, E. Prog. Colloid Polym. Sci. 1994, 95, 73.

3426

Langmuir, Vol. 20, No. 8, 2004

Meneghetti and Qutubuddin

Figure 1. XRD patterns of dried emulsion nanocomposite samples with BAC surfactant and C18DMB and, also, pattern of unmodified MMT at the bottom.

2 mm in diameter. After monomer addition and during polymerization, the droplets are attracted by van der Waals forces and coalesce to form macroscopic pellets. The stirring during polymerization plays an important role in the kinetics of the aggregation process. XRD of the dried polymer pellets as shown in Figure 1 indicates a peak around 5°. This peak corresponds to a d spacing of 1.82 nm and is similar to the 5° peak of MMT ionexchanged with BAC,19 indicating a poorly intercalated polymer composite. In the case of emulsion polymerization with zwitterionic C18DMB, there was no coalescence or coagulation of latex particles to form pellets. XRD of dried emulsion powder (Figure 1) shows no apparent peak and is an indication that the sample may be either exfoliated or intercalated. X-ray alone cannot substantiate either claim. However, the absence of coagulation and observed difference in X-ray patterns indicate that C18DMB behaves differently than BAC. Thus, the nanocomposite properties are expected to be different. A possible explanation for the difference in the nanostructure when comparing surfactant types is that, for BAC, the positive charge of the cationic surfactant neutralizes the negative charge of the MMT surface, which leads to coagulation. Thus, no emulsion droplets are present at the end of polymerization. In the case of C18DMB, the electrostatic interaction is pH-dependent because the charge on the zwitterionic surfactant can be positive, zero, or negative. A schematic diagram of MMA emulsion droplets with C18DMB and MMT in water is shown in Figure 2. Three different structures are possible: intercalated, partially exfoliated, and exfoliated. In the present case, the pH of the emulsion was about 8.50, assuming that the pH electrode is not affected on the surface by the clay and the monomer. The electrostatic interactions also depend on the CEC of clay. The pH of MMT dispersed in water was 9.30. Thus, the pH of the emulsion with C18DMB is influenced by the charge balance between the surfactant and the clay. There is definitely some penetration of the surfactant inside the

Figure 2. Schematic diagram of MMA emulsion droplets with C18DMB and MMT. The charge can be positive, negative, or zero depending on the pH, and this affects the structure.

clay layers, as observed in the functionalized clay reported in a previous paper.19 But some C18DMB is also present at the monomer-water interface, stabilizing the emulsion and avoiding coagulation. After polymerization, however, the emulsion settles in about 1 day because it is thermodynamically unstable. To understand more of the surfactant effect, an emulsion with a 1:1 ratio of BAC/SDS was prepared. SDS provides the anionic part, and it promotes repulsion. Because the SDS will not intercalate into the clay layers, it goes primarily to the interface with the negative headgroup facing the water medium. This provides electrostatic repulsion between the droplets, and the final latex had similar visual characteristics as the C18DMB emulsion with no sign of coagulation. Another emulsion was prepared with C18DMB only, in which a small amount of HCl was added to lower the pH of the solution, so that

Synthesis of PMMA Nanocomposites

Langmuir, Vol. 20, No. 8, 2004 3427

Table 1. DSC, TGA, and DMTA Analysis of Nanocomposites type of nanocomposite pure PMMA PMMA with 10 wt % C18DMB-MMT (in situ) PMMA with 10 wt % MMT and 1.5 wt % C18DMB (emulsion)

Tg, °C (DSC)

Tg, °C (DMTA)

T, °C, for 20 wt % loss

storage modulus at 80 °C (GPa)

intercalated

104.6 112.0

129.3 137.1

321.6 382.6

1.81 2.84

partially exfoliated

122.6

145.2

358.2

2.50

microstructure

Figure 3. SEM micrograph of a dry C18DMB polymer emulsion.

C18DMB would act as a cationic surfactant. Coagulation occurred, forming particles of the same size as in the BAC emulsion. Light scattering was performed on the polymerized C18DMB emulsion after centrifuging it and also on the same emulsion but without clay. The particle size of the diluted polymer emulsion with C18DMB was about 550 nm ( 3%, while the sample without clay had a diameter of about 350 nm ( 3%. The increase in the droplet size suggests that clay was present in the emulsion droplets. Nanocomposites. SEM was performed on the dried C18DMB emulsion. Figure 3 indicates spherical aggregates with a diameter smaller than 1 µm. The dry sample was molded to form a thin film for observation under TEM. The TEM micrograph in Figure 4 indicates that the final PMMA-clay nanocomposite was partially exfoliated. The morphology resembles that presented in refs 17 and 19. For comparison, a PMMA-clay nanocomposite was prepared via in situ polymerization with same clay concentration of 10 wt %. The nanostructure observed by TEM in Figure 5 is only intercalated, as observed by various investigators. DSC showed an 18 °C increase in the Tg of the nanocomposite when prepared by the emulsion technique when compared to that of the pure polymer (Table 1, Figure 6). This increase was slightly higher than that observed by Huang and Brittain16 with a sample containing 5 wt % MMT that was melt-pressed. Also, the increase in Tg was higher than for the intercalated nanocomposite with 10 wt % clay produced by the in situ method.19 In addition, the Tg was verified by DMTA through tan δ (Table 1, Figure 7), showing an increase similar to that measured by DSC. The Tg increased by 16 and 8° for the partially exfoliated and intercalated nanocomposites, respectively. Bandyopadhyay et al.,13,14 using a dynamic mechanical analyzer, measured the Tg of the PMMA-clay nanocomposite with 5 wt % MMT to be 121 °C, which was 6° higher than that

Figure 4. TEM micrograph of a partially exfoliated PMMAclay nanocomposite after the dry powder from emulsion polymerization was molded (10 wt % clay).

Figure 5. TEM micrograph of an intercalated PMMA-clay nanocomposite via in situ polymerization (10 wt % clay).

of the pure polymer. However, this increase was not as significant as that presented in Table 1. Thermal degradation can be monitored by TGA, and the behavior is shown in Figure 8 for pure PMMA and the nanocomposites. Hirata et al.22 reported that two main reaction stages take place during the degradation of PMMA in a nitrogen atmosphere. The first stage, which can be divided into two steps, represents decomposition of weak head-to-head linkages and impurities for the range between 160 and 240 °C and decomposition of the PMMA chain end around 290 °C. The second stage, between 300 and 400 °C, represents random scission of the polymer chains. In Figure 8, pure PMMA displays these two (22) Hirata, T.; Kashiwagi, T.; Brown, J. Macromolecules 1985, 18, 1410.

3428

Langmuir, Vol. 20, No. 8, 2004

Meneghetti and Qutubuddin

Figure 9. DMTA scans of (a) pure PMMA and nanocomposites (b) PMMA-C18MMT 10% (in situ) and (c) PMMA-MMTC18DMB 10% (emulsion). Figure 6. DSC scans of (a) pure PMMA and nanocomposites (b) PMMA-C18MMT 10% (in situ) and (c) PMMA-MMTC18DMB 10% (emulsion).

Figure 7. tan δ curves calculated from DMTA of (a) pure PMMA, (b) PMMA-C18MMT 10% (in situ), and (c) PMMAMMT-C18DMB 10% (emulsion).

Figure 8. TGA of (a) pure PMMA and nanocomposites (b) PMMA-C18MMT 10% (in situ) and (c) PMMA-MMTC18DMB 10% (emulsion).

reaction stages, while the nanocomposites display only the second stage, indicating random scission decomposition. PMMA-clay nanocomposites presented much enhanced thermal decomposition when compared to the pure polymer. TGA indicated an increase of 37 and 61 °C in the decomposition temperature of 20 wt % for the nanocomposites prepared by emulsion and in situ polymerization,

respectively (Table 1). The reason for the lower increase in the emulsion nanocomposites is probably due to excess surfactant present with the polymer emulsion powder. Also, the emulsion nanocomposite was finer in terms of grain size than the in situ nanocomposite. Smaller grains will allow faster decomposition. In fact, after compression molding of the nanocomposite emulsion powder, TGA revealed an increase of 45 °C instead of 37 °C for the partially exfoliated nanocomposite. Dynamic mechanical analysis was used to measure the viscoelastic properties of the polymer nanocomposites as a function of the temperature. The storage modulus, E′, is shown in Figure 9 for pure PMMA and PMMA-clay nanocomposites formed via in situ or emulsion polymerization. The modulus increased with the addition of clay as expected: at 80 °C, the storage modulus increased from 1.81 GPa for PMMA to 2.50 GPa and to 2.84 GPa for emulsion and in situ nanocomposites, respectively (Table 1). The lower modulus for the emulsion nanocomposite versus the in situ can, again, be due to excess surfactant present in the emulsion product, which acts as a plasticizer. At temperatures higher than 100 °C, the partially exfoliated sample showed a higher storage modulus than the intercalated structure. The storage modulus was also higher than that of Wang et al.18 by almost 3 orders of magnitude. The rubbery plateau modulus for the partially exfoliated nanocomposite was over 30 times higher than that of pure PMMA and 10 times higher than the intercalated nanocomposites. These results were higher than those reported by Bandyopadhyay et al.:13,14 for the nanocomposite with 5 wt % clay, the rubbery plateau modulus was 10 times higher than that of PMMA. Therefore, the dispersion of the clay nanostructure in the polymer matrix affects the resulting thermo/mechanical properties of the nanocomposite. For further understanding of the effects of clay addition and the type of polymerization route to the nanocomposite, GPC measurements were performed as shown in Figure 10. The average molecular weight (Mw) for pure PMMA (curve a) was 334 000, while for the PMMA nanocomposite containing 10% clay using an in situ technique (curve b) it was 548 000 and for the PMMA nanocomposite containing 10% clay using the emulsion technique (curve c) it was 930 000. However, the curves actually have the shape of a bimodal distribution. By fitting such a distribution to Figure 10, pure PMMA would have Mw’s of 59 700 and 1 170 000, the in situ nanocomposite would have Mw’s of 31 400 and 1 480 000, and the emulsion nanocomposite would have Mw’s of 30 600 and 1 560 000. The nanocomposites showed a higher molecular weight

Synthesis of PMMA Nanocomposites

Langmuir, Vol. 20, No. 8, 2004 3429

molecular weight strongly increases; as a result, the MWD becomes broader and a bimodal distribution is obtained. The glass effect leads to a decrease in the average molecular weight in the final phase of the reaction.23 Operating conditions such as the temperature, monomer, and initiator concentration have strong influence on the values of the molecular weight and MWD. Balke and Hamielec24 reported bulk polymerization of MMA at 50 °C, chemically initiated with AIBN, in which the PMMA MWD was bimodal at high conversions, as measured by GPC. In this paper, pure PMMA was synthesized in bulk polymerization at 55 °C, similarly as in the work by Balke and Hamielec. Thus, high conversion was probably obtained and the MWD was bimodal. In the case of nanocomposites, the molecular weight was much higher than for pure PMMA. This demonstrated that the clay inhibits free radical polymerization, specifically for PMMA, as described by Solomon and Swift.25 Also, the bimodal distribution is more clearly observed in the nanocomposites because the lower molecular weight PMMA becomes trapped between the clay galleries, while the higher molecular weight of PMMA corresponds to the amorphous matrix. The observed increase in Tg for the nanocomposites compared with pure PMMA may be attributed to their higher molecular weight. O’Driscoll and Sanayei26 described the change of Tg for the low molecular weight of PMMA by a nonlinear regression up to Mw of 14 000. However, for the higher molecular weight, Dibenedetto and Dilandro27 illustrated that 1/Tg follows a linear correlation with 1/X, where X is the number-average chain length. In a simplified form, their equation reduces to the Ueberreiter-Kanig equation:

T(∞) g Tg(X)

Figure 10. GPC curves of (a) pure PMMA (bulk polymerization) and nanocomposites (b) PMMA-C18MMT 10% (in situ), and (c) PMMA-MMT-C18DMB 10% (emulsion) dissolved in THF.

than the pure polymer, and the emulsion technique gave the highest molecular weight between the nanocomposites. These values were, in fact, even higher than the molecular weights reported in the literature for pure PMMA and its nanocomposites.10-17 To understand the nature of bimodal distribution, it is necessary to analyze the reaction mechanism. Free radical polymerization is described by three processes: initiation, propagation, and termination. In batch reactors, these reaction mechanisms become more complicated because of the significant changes in the physical properties of the reacting system with conversion during the reaction; as conversion increases, the main effect of the diffuse phenomena is on the termination rate because the long polymer chains move more slowly as a result of the increase in viscosity (gel effect). At a very high conversion, also, the propagation rate is affected by diffusion (glass effect).23 The reaction kinetics directly affects the chain length of the polymer produced during polymerization, and as a consequence, the average molecular weight and the molecular weight distribution (MWD) are affected by the diffusion phenomena. During the gel effect, the average (23) Maschio, G.; Scali, C. Macromol. Chem. Phys. 1999, 200, 1708.

[ ()]

) 1+

ce 1 cm X

(1)

where ce/cm is the ratio of excess end segment mobility to that of the midchain segment. In fact, for their calculathat was used for PMMA was 395 K or tion, the T(∞) g 122 °C. This Tg value was the same as that obtained by the partially exfoliated PMMA nanocomposite described in this paper. Conclusions The synthesis of the partially exfoliated nanocomposite via emulsion polymerization of MMA in the presence of 10 wt % clay and using zwitterionic surfactant was described. C18DMB provided some ion exchange with the clay and led to a stable emulsion without pellets, which were obtained when using a cationic surfactant. Light scattering on the polymerized C18DMB emulsion suggested that the clay was present inside the emulsion droplets. The dry powder from the polymer emulsion with C18DMB had no X-ray peak. TEM of the melt-processed sample indicated a partially exfoliated nanocomposite. DSC indicated an increase in Tg of the nanocomposite by 18 °C above that of pure polymer. DMTA also verified a similar Tg increase, 16 °C, for the partially exfoliated nanocomposite over PMMA. The temperature for decomposition of the 20 wt % loss increased by 37 °C to 358 °C. The storage modulus of the partially exfoliated nanocomposite was superior to the intercalated structure at (24) Balke, S.; Hamielec, A. J. Appl. Polym. Sci. 1973, 17, 905. (25) Solomon, D.; Swift, J. J. Appl. Polym. Sci. 1967, 11, 2567. (26) O’Driscoll, K.; Sanayei, R. Macromolecules 1991, 24, 4479. (27) Dibenedetto, A.; Dilandro, L. J. Polym. Sci., Part B: Polym. Phys. 1989, 27, 1405.

3430

Langmuir, Vol. 20, No. 8, 2004

higher temperatures and to the pure polymer. The rubbery plateau modulus was over 30 times higher for the emulsion product versus for pure PMMA. The emulsion technique produced nanocomposites of the highest molecular weight with a bimodal distribution. Acknowledgment. Partial financial support of this research by Edison Polymer Innovation (EPIC) is grate-

Meneghetti and Qutubuddin

fully acknowledged. The authors thank Professor Blackwell for the use of X-ray equipment, Professor A. Jamieson for the use of the light scattering instrument, the Macromolecular Sciences Department for the use of GPC equipment, John Davidson for helping with SEM and TEM, and Dr. Nazarenko for the use of DMTA. LA0357099