Langmuir 2005, 21, 3613-3618
3613
Preparation of Highly Exfoliated Epoxy/Clay Nanocomposites by “Slurry Compounding”: Process and Mechanisms Ke Wang,† Lei Wang,† Jingshen Wu,‡ Ling Chen,† and Chaobin He*,† Molecular and Performance Materials Cluster, Institute of Materials Research and Engineering, 3 Research Link, Singapore 117602, and Department of Mechanical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China Received September 13, 2004. In Final Form: January 30, 2005 Epoxy/clay nanocomposites with a high degree of exfoliation were achieved using a so-called “slurrycompounding” process with which the dispersed state of clay in water can be successfully transferred to an epoxy matrix. In this process sodium montmorillonite was first exfoliated and suspended in water. This suspension was further treated with acetone to form a clay-acetone slurry followed by chemical modification using silane. The modified slurry was then mixed extensively with epoxy to form epoxy/nanoclay composites. It has been shown that the morphologies of clay before and after curing are quite similar and the exfoliation process is termed “slurry compounding”. Furthermore, the amount of organic modifier used is only 5 wt % of clay, in contrast to conventional organoclays which normally contain at least 25-45 wt % of organic surfactant. The resulting epoxy/nanoclay composites exhibit a high degree of clay exfoliation and a better thermal mechanical property.
1. Introduction One major hurdle preventing polymer/clay nanocomposite from many potential applications is the difficulty to achieve a high degree of exfoliation in polymer matrixes. In recent years, many research groups from both academic institutes and industrial sectors have been working on different methods to facilitate clay exfoliation.1-9 One frequently used strategy is to modify inorganic clay using organic additives in a hope that the compatibility between hydrophobic polymer and hydrophilic clay is improved and clay exfoliation can be facilitated. However, the structure of the clay tactoids often remains even with the aid of organic modifiers and the modified organoclay is more likely intercalated rather than fully exfoliated. As reported in the literature,10-12 the epoxy/organoclay nanocomposites usually have a heterogeneous morphology comprising both exfoliated and intercalated structures. While transmission electron microscopy (TEM) and wideangle X-ray diffraction (XRD) often show that the clay galleries were intercalated with polymer chains and the * Corresponding author. † Institute of Materials Research and Engineering. ‡ Hong Kong University of Science. (1) Usuki, A.; Kawasumi, M.; Kojima, Y.; Okada, A.; Kurauchi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1174-1178. (2) Usuki, A.; Kojima, Y.; Kawasumi, M.; Okada, A.; Fukushima, Y.; Kurauchi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1179-1183. (3) Lan, T.; Pinnavaia, T. J. Chem. Mater. 1994, 6, 2216-2219. (4) Messersmith, P. B.; Giannelis, E. P. Chem. Mater. 1994, 5, 17191725. (5) Lan, T.; Kaviratna, P. D.; Pinnavaia, T. J. Chem. Mater. 1995, 7, 2144-2150. (6) Konmann X.; Lindberg H.; Berglund, L. A. Polymer 2001, 42, 1303-1310. (7) Kong, D.; Park, C. E. Chem. Mater. 2003, 15, 419-424. (8) Giannelis, E. P. Adv. Mater. 1996, 8, 29-35. (9) Ray, S. S.; Okamoto, M. Prog. Polym. Sci. 2003, 28, 1539-1641. (10) Zerda, A. S.; Lesser, A. J. J. Polym. Sci., Part B: Polym. Phys. 2001, 39, 1137-1146. (11) Zilg, C.; Mulhaupt, R.; Finter, J. Macromol. Chem. Phys. 1999, 200, 661-670. (12) Becker, O.; Cheng, Y. B.; Varley, R. J.; Simon, G. P. Macromolecules 2003, 36, 1616-1625.
d-spacing was enlarged to more than 50 Å, large particle aggregates would be seen when examined under lower magnification (scanning electron microscopy or optical microscopy), indicating that clay particles were neither well-exfoliated nor uniformly dispersed in the polymer matrix. Obviously, the poor clay dispersion observed in conventional epoxy/organoclay nanocomposites is mainly due to the preparation process and the intercalation-exfoliation mechanisms involved. In a typical preparation process, epoxy and organoclay powder are mixed and cured with a selected hardener under a designed temperature profile.3-8 At the first stage of preparation, the dispersion of organoclay particles depends on the compatibility between epoxy and organoclay and the intensity of mixing. Because the magnitude of the entropic force responsible for separating the clay layers is very small,13 almost no exfoliation occurs in this step. In a subsequent curing stage, a diffusion-polymerization-exfoliation process occurs.8,13-14 It has been widely accepted that most of the exfoliation occurs in a later stage, because the dominant driving force for exfoliation is the increase of conformation entropy with the progress of cross-linking reaction. However, in the curing stage, the clay tactoids can only be swollen by epoxy and further dispersion of clay tactoids becomes impossible because the mixture is not subjected to shearing and the viscosity of the system increases dramatically with the progress of the curing process. Thus, in most cases, the ordered structure of clay layers is still maintained although the interlayer distance can be enlarged to some extent. Furthermore, for conventional organoclays, the organic modifiers generally amount to 25-45 wt % of the modified clay. While the modified clay does show some advantages in promoting exfoliation and/or dispersion, the use of high content of organic modifier makes organoclay expensive (13) Park, J. H.; Jana, S. C. Macromolecules 2003, 36, 2758-2768. (14) Chen, J. S.; Poliks, M. D.; Ober, C. K.; Zhang, Y.; Wiesner, U.; Giannelis, E. Polymer 2002, 43, 4895-4904.
10.1021/la047709u CCC: $30.25 © 2005 American Chemical Society Published on Web 03/02/2005
3614
Langmuir, Vol. 21, No. 8, 2005
Wang et al.
and matrix-specific. In particular, the small molecular component of organic modifiers, normally surfactants, could lead to low glass transition temperature (Tg) of the resulting nanocomposite and render it thermally unstable. To overcome the drawbacks in the conventional organoclay, we report in this study our new approach in preparing epoxy clay nanocomposites using a “slurry-compounding” technique. This technique leads to a better clay exfoliation with only a very low concentration of organic modifier (4.5 nm (curve 3), which is out of range of our XRD. In fact, a systematic study reported by Morvan et al.18 has shown that the interlayer distance can be 15 nm in a clay-water suspension with 8.5 wt % of clay. When the clay-water suspension was mixed with acetone, clay precipitated to form loose sediments. A weak and broad peak is observed at 2θ ) 5.24° (1.68 nm) in the corresponding XRD curve (curve 4), indicating that some of the clay platelets may reassemble to form a layer structure or the interlayer distance for some of the swollen galleries reduces due to the removal of water. The fact that the d-spacing of clay in the slurry is bigger than that of pristine clay (1.21 nm) reveals that the clay was swollen
Langmuir, Vol. 21, No. 8, 2005 3615
Figure 3. FTIR spectra of (1) pristine clay and (2) dried modified clay.
by acetone or residual water although some of the clay platelets aggregate together. The d-spacing difference is about 0.5 nm, and it is estimated that there may be a single molecular layer between each clay layer. The dispersion state of clay is likely similar to that of clay in salt solution.18 That is to say, clay platelets are not uniformly dispersed in the whole sample and two types of domain coexist: small nematic domains consisting of a few parallel clay platelets and pores of mainly acetone with numerous individual clay platelets. The modification with APTMS did not affect the microstructure much because the (001) diffraction of modified clay/acetone slurry appears at the same position as that before modification (curve 5). After drying, the modified clay shows a peak at 2θ ) 7.20° (1.22 nm, curve 6), which is very close to that of pristine clay, indicating that almost no modifier molecules intercalated into the clay galleries. This is understandable because the silane is preferentially grafted at the edge of the clay platelets rather than the surface of the clay platelets due to existence of hydroxyl group at the edge of the clay platelets.19-21 However, even with such minimal modification, the resulting epoxy nanocomposites exhibit a much better exfoliation morphology. The evidence for the reaction between the silane and clay platelets comes from FTIR and ToF-SIMS studies. The FTIR spectrum of pristine clay (Figure 3) displays typical -OH stretching (3634 cm-1) absorbencies, and broad bands at 3447 and 1640 cm-1 can be attributed to adsorbed water molecules.22 The bands at 1116, 1035, and 914 cm-1 can be collectively attributed to Si-O stretching vibrations.23 The spectrum of silane-modified clay displays almost the same pattern as that of pristine clay except for a new band of -CH stretching at 2937 cm-1, which indicates the grafting of organic groups on the clay mineral surface. (19) Rausell-Colom, J. A.; Serratosa, J. M. In Chemistry of Clays and Clay Minerals; Newman, A. C. D., Ed.; Wiley-Interscience: New York, 1987; p 371. (20) Carrado, K. A.; Xu, L.; Csencsits, R.; Muntean, J. V. Chem. Mater. 2001, 13, 3766-3773. (21) Song, K.; Sandi, G. Clays Clay Miner. 2001, 49, 119-125. (22) Katti, D. R.; Katti, K. S.; Shanmugasundaram, V. Mater. Res. Soc. Symp. Proc. 2002, 704, 257-262. (23) Wypych, F.; Schreiner, W. H.; Mattoso, N.; Mosca, D. H.; Marangoni, R.; da S. Bento, C. A. J. Mater. Chem. 2003, 13, 304-307.
3616
Langmuir, Vol. 21, No. 8, 2005
Wang et al.
Figure 5. XRD diagrams of the epoxy/S-clay nanocomposites with different clay content.
Figure 4. ToF-SIMS spectra of (a) pristine clay and (b) dried modified clay.
ToF-SIMS spectra of pristine clay and the modified clay are shown in Figure 4. For the pristine clay, abundant Na+, Al+, and Si+ are observed, which is consistent with the natural composition of the sodium montmorillonite. Lower abundance cations corresponding to K+ and Mg+ are also observed, which are components of the pristine clay. In addition, the presence of organic adsorbates is observed as well. A relatively abundant example is observed at m/z 41, similar to the organic ions reported by Groenewold and co-workers with clay.24 When the clay was modified with APTMS, new ions with m/z 29 and 30 are observed in the ToF-SIMS spectrum, which corresponds to elemental compositions of CH3N and CH4N, respectively. Evidently, they are derived from the silane molecules that grafted on to the surface of clay sheets. In summary, the highly exfoliated structure of clay in water was partially transferred into acetone media, in which small swollen clay tactoids and single clay platelets coexist. The modification reaction mainly takes place on the surfaces or edges of the single platelets and the tactoids (24) Groenewold, G. S.; Avci, R.; Karahan, C.; Lefebre, K.; Fox, R. V.; Cortez, M. M.; Gianotto, A. K.; Sunner, J.; Manner, W. L. Anal. Chem. 2004, 76, 2893-2901.
rather than intergalleries. It is expected that the structure will be transferred into epoxy in the following steps. 3.2. Morphology. To prepare the epoxy/S-clay nanocomposites, epoxy prepolymer was mixed extensively with the modified clay/acetone slurry, and subsequently the solvents were removed. The dried mixtures are transparent up to 5 wt % clay, indicating good dispersion of clay in epoxy matrix. The XRD curve 6 in Figure 5 represents the typical structure of such epoxy/clay mixtures before curing, in which a weak peak was observed at 2θ ) 6.56° (1.35 nm). By comparison of this scattering (curve 6 in Figure 5) with curve 5 in Figure 2, it is obvious that the layers collapsed slightly but not completely after the removal of solvents and some of the epoxy molecules had intercalated into the clay galleries. The scattering in curve 6 in Figure 5 is in contrast to curve 6 in Figure 2 where a very intense, sharp peak appears at low angle with intensity comparable to the crystalline peak at 2θ ) 20° when the modified clay was dried without addition of epoxy. During curing, the tactoid galleries were expanded further. For samples with 1.0 and 2.5 wt % clay, the (001) diffraction peak is almost invisible (curves 2 and 3 in Figure 5), revealing that a high degree of exfoliation has been achieved in these samples. For the samples containing more than 3.5 wt % clay, a weak scattering peak occurs with a slight large d spacing of 1.42 nm, while the intensity decreases considerably (weaker than that of the amorphous halo). Evidently, the morphology is determined in the preparation steps before curing, whereas the curing step does not affect the microstructure significantly. An optical micrograph shows that, in the epoxy/S-clay nanocomposites, clay particles are uniformly dispersed in the matrix and the visible aggregate size is smaller than 3 µm (Figure 6a). Parts b and c of Figure 6 present TEM micrographs showing that the clay is exfoliated into single layers or thin tactoids that consist of 5 to 10 clay layers. These single layers and thin tactoids disperse uniformly and randomly in the matrix, indicating that our new approach is very effective in promoting both the exfoliation and dispersion of clay. It is interesting to see that the final microstructure of our nanoclay composites is quite similar to the microstructure reported by Morvan et al.18 for clay-water suspensions added with brine. It is understandable because acetone is not as effective as water in swelling clay due to the lower polarity of the former. When the
Epoxy/Clay Nanocomposites
Langmuir, Vol. 21, No. 8, 2005 3617
Figure 6. Morphology of the epoxy/S-clay nanocomposite containing 2.5 wt % of pristine clay: (a) optical micrograph; (b, c) TEM micrographs.
Figure 7. XRD diagrams of Cloisite 93A and the epoxy/93A nanocomposites with different 93A contents.
water in the swollen clay is exchanged by acetone, the clay precipitates from the solution and the interlayer distance is likely to decrease, just as it happens when brine is added in clay/water solution.18 For comparison, a group of epoxy/organoclay nanocomposites derived from commercial available organoclay Cloisite 93A were prepared and characterized. The XRD diagrams in Figure 7 show that the (001) diffraction peak of organoclay is visible except for the sample containing 1 wt % Cloisite 93A. It is noted that the (001) peaks of clay in epoxy/93A nanocomposites shift slightly to low angle
compared with Cloisite 93A, i.e., from 2θ ) 3.72° to 2.50°. The presence and shift of the (001) diffraction indicate that intercalated structure and exfoliated structure coexist in the epoxy/93A nanocomposites. Figure 8a shows the optical micrograph of the epoxy/93A nanocomposite containing 2.5 wt % of Cloisite 93A. The dispersion of the organoclay in epoxy is poor, and the aggregate size ranges from a few micrometerss to more than 20 µm. Under TEM, it can be seen that these aggregates are comprised of hundreds of clay layers (Figure 8b) and the ordered structure of clay is maintained although the d-spacing of the clay layers is as large as about 10 nm (Figure 8c). This morphology is typical for epoxy/organoclay nanocomposites. It is worth mentioning that the dispersion and exfoliation mechanisms in our new approach are different from that widely reported in epoxy/organoclay systems. In our new approach, most of the exfoliation and dispersion process occurs before curing. The key step is “slurry processing”, i.e., the dispersion and exfoliation state of modified clay in the clay/acetone slurry was directly transferred into epoxy. The final microstructure depends on the state of dispersion and exfoliation of clay in water and acetone and how much it can be maintained in epoxy with the aid of silane modification. During the curing stage, more epoxy molecules intercalated into clay galleries and help clay to exfoliate further, but this is not crucial to the final morphology. This is in contrast to epoxy/organoclay systems in which the curing stage is the most important step. Furthermore, compared with most of the commercialized organoclays, in which the amount of organic modifier is between 25 and 45 wt %, the concentration of
Figure 8. Morphology of the epoxy/93A nanocomposite containing 2.0 wt % of Cloisite 93A: (a) optical micrograph; (b, c) TEM micrographs.
3618
Langmuir, Vol. 21, No. 8, 2005
Wang et al.
3.3. Thermal Mechanical Property. The storage modulus (E′) and tan δ of epoxy/S-clay and epoxy/93A nanocomposites are compared with that of the neat epoxy in parts a and b in Figure 9, respectively. The epoxy/Sclay containing 5 wt % of clay has a Tg very close to that of the neat epoxy and shows ∼30% and ∼150% increase in E′ at 50 and 240 °C, respectively. The epoxy/93A containing 5 wt % of commercial organoclay Cloisite 93A shows an E′ between that of the neat epoxy and the epoxy/ S-clay below 150 °C. It is quite reasonable because the Cloisite 93A contains about 40 wt % of organic modifier and thus the true content of clay in the sample is about 3 wt %. However, it is notable that the Tg of the epoxy/93A is considerably lower than that of the neat epoxy (188.0 °C vs 211.1 °C) and the E′ decreases dramatically with the increase of temperature. 4. Conclusions A “slurry-compounding” approach has been developed for epoxy/clay nanocomposite preparation using pristine clay, which involves a new exfoliation mechanism. With this approach, the dispersion state of clay in water has been transferred into epoxy matrix by a solvent exchange step and a surface modification step. The critical step is the replacement of water with organic solvent, which facilitates the surface modification and dispersion of modified clay in epoxy matrix. The most significant feature of the new technique is that very little amount of organic modifier required to facilitate the high exfoliation and well-dispersion of the clay, which leads to a better clay exfoliation and excellent thermal mechanical property of the resulting nanocomposites. It is clear that the dispersion and exfoliation mechanisms in the new approach are different from that widely reported in epoxy/organoclay systems. Figure 9. Storage modulus (a) and tan δ (b) versus temperature for the neat epoxy, the epoxy/S-clay, and the epoxy/93A containing 5 wt % of clay.
silane used in the new approach is quite low (5 wt % of pristine clay). This is very important for both commercialization in terms of cost and the properties of the final product.
Acknowledgment. Funding support from the Air Force Office of Scientific Research (AOARD-02-4019) and the Institute of Materials Research and Engineering (Singapore) is gratefully acknowledged. The authors thank Ms. Doreen Lai Mei Ying for assistance in the ToF-SIMS measurements. LA047709U
