Clay

Langmuir 2007, 23, 1701-1710

1701

Polymorphism Behavior of Poly(ethylene naphthalate)/Clay Nanocomposites: Role of Clay Surface Modification Yang Choo Chua†,‡ and Xuehong Lu*,† School of Materials Science and Engineering, Nanyang Technological UniVersity, Nanyang AVenue, Singapore 639798, and Institute of Materials Research and Engineering, 3 Research Link, Singapore 117602 ReceiVed September 6, 2006. In Final Form: NoVember 8, 2006 The influence of clay surface modification on the polymorphism behavior of poly(ethylene naphthalate) (PEN)/clay nanocomposites was investigated via in situ Fourier transform infrared (FTIR) spectroscopy and X-ray diffraction. The results show that untreated clay has a heterogeneous nucleating effect on PEN and favors the β-crystal form, while the surfactant 1-hexadecyl-2,3-dimethylimidazolium (IMC16) has a plasticization effect and tends to enhance the kinetically favored R-phase instead. In contrast, the nanocomposite (PEN/IMC16-MMT) formed from IMC16-treated clay (IMC16-MMT) exhibits a strong temperature-dependent polymorphic behavior, with the β-phase being more favored at 200 °C, but the R-phase being preferred instead at 180 °C. In situ FTIR spectroscopy of PEN/IMC16-MMT reveals an abrupt change in the concentration of R- and β-“crystalline conformers” between the two temperatures during the induction period of crystallization. This is attributed to the hindered formation of stable nuclei at the organoclay surface. In addition, surfactant degradation gives rise to a highly plasticized polymer/organoclay interface. The combination of the hindered heterogeneous nucleation and plasticization effects gives rise to the unique temperaturedependent polymorphism behavior in PEN/IMC16-MMT.

Introduction The development of polymer/clay nanocomposites has attracted much attention in recent years, driven mainly by the dramatic enhancements in physical properties that these materials often exhibit at relatively low inorganic loadings. To disperse the hydrophilic clay layers in hydrophobic polymers at the nanoscale, surfactant molecules often need to be ion-exchanged into the clay galleries to impart compatibility between the two components. The dispersion of the high-aspect-ratio clay platelets leads to an extremely large interfacial area which, combined with the surfactant-modified clay surface, often results in unique structural modifications to the polymer matrix. For polymers that exhibit polymorphism in particular, the presence of a nanoclay has often been found to exert an important effect on the type of crystal phase developed.1-5 The phenomenon of clay-induced polymorphism has often been cited as an important factor contributing to the unique combination of properties exhibited by polymer/clay nanocomposites.4,5 In spite of the extensive investigations that have been carried out in this area, the underlying mechanisms involved, in particular the role played by the surfactant-modified clay surface, are still not well understood. For polymers which require higher processing temperatures, beyond what could typically be sustained by the organoclays, the occurrence of organoclay degradation could further complicate the issue by altering the elevated * To whom correspondence should be addressed. E-mail: asxhlu@ ntu.edu.sg. † Nanyang Technological University. ‡ Institute of Materials Research and Engineering. (1) Zheng, W.; Lu, X.; Toh, C. L.; Zheng, T. H.; He, C. J. Polym. Sci., Part B: Polym. Phys. 2004, 42, 1810. (2) Wu, H.-D.; Tseng, C.-R.; Chang, F.-C. Macromolecules 2001, 34, 2992. (3) Mathias, L. J.; Davis, R. D.; Jarrett, W. L. Macromolecules 1999, 32, 7958. (4) Kojima, Y.; Usuki, A.; Kawasumi, M.; Okada, O.; Fukushima, Y.; Kurachi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1185. (5) Shah, D.; Maiti, P.; Gunn, E.; Schmidt, D. F.; Jiang, D. D.; Batt, C. A.; Giannelis, E. P. AdV. Mater. 2004, 16, 1173.

temperature interfacial chemistry.6 This effect has however often been overlooked, as most studies on the polymorphism of polymer/clay nanocomposites have implicitly assumed that the original surface structure of the organoclay is maintained throughout the various high-temperature excursions. Poly(ethylene naphthalate) (PEN) is a high-performance polyester that is attracting increasing commercial interest. Despite considerable interest in the development of poly(ethylene terephthalate) (PET)/clay nanocomposites in recent years,7 the PEN/clay system has been relatively unexplored and thus presents significant potential for discovery. Unlike PET, PEN is known to crystallize in two different forms, R and β, depending on the thermal treatment.8-10 The R-phase, which adopts an all-trans conformation, is formed on annealing amorphous PEN in the solid state, while the β-phase, which adopts a conformation with both trans and gauche character, can develop if crystallization is carried out directly from the melt state at higher temperatures.9 The β-phase is believed to be more thermodynamically stable, while the R-phase is more kinetically favored;10 this has been found to be true for both pure PEN and PEN/clay nanocomposites.11 In an earlier work, we investigated the effect of the melt crystallization temperature on the polymorphic behavior of PEN/ clay nanocomposites.12 Under melt crystallization at 200 °C, pristine montmorillonite (Na-MMT) and two types of surfactant(6) Frost, R. L.; He, H.; Kloprogge, T.; Bostrom, T. Langmuir 2005, 21, 8675. Xie, W.; Gao, Z. M.; Pan, W. P.; Hunter, D.; Singh, A.; Vaia, R. Chem. Mater. 2001, 13, 2979. Xie, W.; Xie, R.; Pan, W. P.; Hunter, D.; Koene, B.; Tan, L.-S.; Vaia, R. Chem. Mater. 2002, 14, 4837. (7) Wan, T.; Chen, L.; Chua, Y. C.; Lu, X. J. Appl. Polym. Sci. 2004, 94, 1381. Hao, J.; Lu, X.; Liu, S. L.; Lau, S. K.; Chua, Y. C. J. Nanosci. Nanotechnol. 2006, 6, 3981. Hao, J.; Lu, X.; Liu, S. L.; Lau, S. K.; Chua, Y. C. J. Appl. Polym. Sci. 2006, 101, 1057. (8) Mencik, Z. Chem. Prum. 1976, 17, 78. Buchner, S.; Wiswe, D.; Zachmann, H. G. Polymer 1989, 30, 480. (9) Vasanthan, N.; Salem, D. R. Macromolecules 1999, 32, 6319. (10) Blanton, T. N. Powder Diffr. 2002, 17, 125. (11) Wu, T.-M.; Liu, C.-Y. J. Macromol. Sci., Phys. 2004, B43, 1171. (12) Chua, Y. C.; Lu, X.; Wan, T. J. Polym. Sci., Part B: Polym. Phys. 2006, 44, 1040. Chua, Y. C.; Wu, S.; Lu, X. J. Nanosci. Nanotechnol. 2006, 6, 3985.

10.1021/la0626048 CCC: $37.00 © 2007 American Chemical Society Published on Web 01/04/2007

1702 Langmuir, Vol. 23, No. 4, 2007

modified MMTs were found to enhance the formation of the β-phase over the R-crystal form, with the organoclays exerting a more pronounced β-enhancement effect. Reducing the melt crystallization temperature to 180 °C however caused the R-crystal form to be more favored in the nanocomposites with the surfactantmodified MMT, while Na-MMT continued to enhance the β-phase. This study was undertaken to elucidate the underlying mechanisms for the unusual temperature-dependent polymorphism behavior exhibited by the PEN/organoclay nanocomposites. It was found that the type of crystal phase formed in the PEN/organoclay nanocomposites depends on the interplay between the hindered heterogeneous nucleating effect of the organoclay and the plasticizing effect of the surfactant molecules and their degradation products at the polymer/organoclay interface. Experimental Section Materials. PGW grade sodium montmorillonite (Na-MMT), with a cation exchange capacity (CEC) of 145 mequiv/100 g, was supplied by Nanocor, Inc. 1-Hexadecyl-2,3-dimethylimidazolium (IMC16) chloride was purchased from Merck, while additive-free PEN granules were supplied by Goodfellow Cambridge Ltd. All chemicals were used without further purification. The organically modified clay (IMC16-MMT) was prepared by a standard ion-exchange procedure, and the PEN/clay hybrids (PEN/Na-MMT and PEN/IMC16-MMT) were prepared by melt compounding PEN and the clays as detailed in our earlier papers.12 Both PEN/clay hybrids have a nominal 2% inorganic content by mass. Pure PEN was also extruded under the same conditions to be used as a reference. For comparison purposes, a PEN/IMC16 blend containing the same amount of IMC16 surfactant (ca. 0.8 wt %) as PEN/IMC16-MMT was also prepared using the same method. The amount of IMC16 surfactant that has been incorporated into IMC16-MMT was evaluated by thermogravimetric analysis (TGA) of IMC16-MMT in air at a heating rate of 10 °C/ min.12 All materials were dried under vacuum at 80 °C for at least 8 h prior to melt compounding. Preparation of Crystallized Samples. The materials were melted at 290 °C for 5 min in a convection oven to completely remove their thermal history. The melted samples were then immediately transferred to an oil bath set at the desired melt crystallization temperature and allowed to crystallize for 15 min (at 180 and 190 °C) or 30 min (at 200 °C) before they were air-cooled to ambient temperatures. To eliminate effects due to preferred orientation of the crystallites, all the materials were ground into a fine powder using a Fritsch Pulverisette 14 and an 80 µm sieve. X-ray Diffraction (XRD). Room-temperature XRD for the meltcrystallized samples was performed with a Shimadzu X-ray diffractometer with Cu KR radiation. The samples were scanned from 2θ ) 2.5° to 2θ ) 40.0° at a scan rate of 0.25 deg/min. Deconvolution of the superposed crystalline and amorphous peaks in the X-ray diffraction pattern was performed using the profile fitting program XFIT.12 High-temperature X-ray diffraction patterns were recorded with a Bruker GADDS X-ray diffractometer equipped with a twodimensional area detector using Cu KR radiation. The powdered samples were packed into glass capillaries (Hampton Research, HR6194) before they were placed into the temperature chamber of the X-ray diffractometer. The glass capillaries are specially manufactured for X-ray diffraction experiments, and they do not exhibit any peak in the range of 2θ ) 2.5-10.0°. The samples were then heated at a rate of 20 °C/min to 300 °C and allowed to melt for 3 min before they were cooled at 20 °C/min to 200 °C. A data acquisition time of 200 s was employed, and the X-ray diffraction patterns were collected while the sample was held isothermal at 200 °C, until no further changes in the diffraction patterns were observed. The twodimensional scattering patterns were integrated radially to obtain intensity against 2θ plots using the GADDS software package. Fourier Transform Infrared (FTIR) Spectroscopy. FTIR spectra were collected with a Perkin-Elmer AutoImage FTIR

Chua and Lu microscope equipped with a mercury cadmium telluride detector. For the in situ monitoring of the melt crystallization process, thin polymer films on gold-coated glass cover slips were used. The samples were placed on top of the FTIR 600 Linkam hot stage and melted at 280 °C for 3 min before they were cooled at a rate of 50 °C/min to the desired melt crystallization temperatures. FTIR spectra were collected during the melt crystallization process in the reflection mode with a mean collection time of 45 s per spectrum. The spectra were obtained by coadding 32 scans. For IMC16 and IMC16-MMT, the spectra were collected in transmission mode at a resolution of 2 cm-1 from KBr pellets placed on top of the FTIR 600 Linkam hot stage. The clay was heated using the same temperature profile as that used for the melt crystallization process, and the respective IR spectra were collected at various temperature points of the melt crystallization process by averaging 32 scans. Differential Scanning Calorimetry (DSC). The dynamic crystallization exotherms of the materials were measured using a TA Instruments MDSC 2920. The sample was first heated to 300 °C and held for 5 min to remove its thermal history. It was then cooled at a controlled rate of 5 °C/min to 25 °C. All experiments were performed under a nitrogen purge. Thermogravimetric Analysis (TGA). The degradation behaviors of IMC16-MMT during melt compounding and melt crystallization were examined by using a TA Instruments TGA 2050 thermogravimetric analyzer. The clays were first predried at 140 °C for 1 h to remove physiabsorbed water and gases. To simulate the melt compounding process, the clay was heated at 10 °C/min to 290 °C and held for 5 min. The clay residues were then collected and subjected to a second heating cycle, simulating the melt crystallization process: the clays were reheated at 20 °C/min to 300 °C and kept isothermal for 3 min, before they were cooled at a rate of 20 °C/min to 180 or 200 °C and held at that temperature for 30 min. All the TGA procedures were carried out under an air atmosphere.

Results and Discussion The states of dispersion achieved in the PEN/clay hybrids have been reported in a separate paper.12 An intercalated nanocomposite has been formed in PEN/IMC16-MMT, as evidenced by an increase of ∼0.65 nm in the d spacing of IMC16MMT after melt compounding with PEN. On the other hand, the inorganic nature of the Na-MMT surface causes it to be incompatible with PEN, and a microcomposite is formed in the case of PEN/Na-MMT. At the melt compounding temperature, the surfactant IMC16 exists in the liquid state (Figure S1, Supporting Information) and is able to mix with PEN at the molecular level. PEN/IMC16 is therefore a miscible blend, or better called plasticized PEN, as a reduction in Tg has been observed in PEN/IMC16 (Table S1, Supporting Information). Polymorphism Behaviors of PEN and the Hybrids. Figure 1 displays the XRD patterns of PEN and the PEN/clay hybrids that were melt crystallized at 200, 190, and 180 °C. As a comparison, the XRD patterns of PEN/IMC16 that were melt crystallized at 200 and 180 °C are also shown. The relative amount of the β-crystalline form is measured by an index, Kβ, which is given by Kβ )

Aβ(111) + Aβ(020) + Aβ(242)

AR(010) + AR(100) + AR(110) + Aβ(111) + Aβ(020) + Aβ(242)

× 100%

where A refers to the integrated intensity associated with the crystal peak of interest, based on the peak deconvolution results.12 The variation of Kβ with the melt crystallization temperature is illustrated in Figure 2. The results presented above indicate that pristine Na-MMT tends to promote the formation of the β-phase in PEN at all

Polymorphism BehaVior of PEN/Clay Nanocomposites

Langmuir, Vol. 23, No. 4, 2007 1703

Figure 2. Variation of the Kβ index with the melt crystallization temperature. Table 1. Band Assignments of PEN in the Amorphous, r- and β-Crystalline Phasesa IR frequencies (cm-1) amorphous 1477 m 1452 m

R-crystalline

β-crystalline

1477 m 1004 m

964 m 931 sh

931 s

974 sh 917 s

838 m, 811 m

832 m

916 s 822 m a

Figure 1. XRD patterns of PEN and its hybrids melt crystallized at (a) 200 °C, (b) 190 °C, and (c) 180 °C.

temperatures. On the other hand, the presence of the IMC16 surfactant alone in PEN/IMC16 enhances the R-phase instead. In contrast, the effect of IMC16-MMT is found to vary with temperature. At 200 °C, the presence of IMC16-MMT strongly

assignments CH2 bending (trans) CH2 bending (gauche) C-O stretching (trans) C-O stretching (gauche) OdC-O out-of-plane bending CH2 rocking (amorphous) CH2 rocking (crystalline)

Abbreviations: s ) strong, m ) medium, sh ) shoulder.

enhances the formation of the β-phase. The β-phase enhancement effect however diminishes rapidly with reducing temperatures, and at 180 °C, PEN/IMC16-MMT is found to preferentially form the R-crystal form instead, even though a small amount of β-crystals is still observed in pure PEN. The polymorphism behaviors of the materials were corroborated by FTIR spectroscopy. Figure 3 shows representative spectra of PEN, PEN/IMC16, and the hybrids at the quiescent melt state and after complete crystallization at 200 and 180 °C. The band frequencies that are highly sensitive to the structural changes occurring during crystallization are labeled and their assignments provided in Table 1.9,13 It is observed that as the polymer passes from the melt state to the crystal phase, new absorption bands appear in the IR spectrum, or splitting occurs in the bands present in the spectra of the melt. These changes arise because of the changes in conformations and/or intermolecular interactions of the polymer chains as they pack into the ordered crystal phase. In addition, the relative intensities of the R-phase-related bands at 1477, 1004, 931, and 811 cm-1, together with the position of the crystal band at ∼832-838 cm-1, confirm that, under melt crystallization at 200 °C, the β-phase is enhanced in PEN/IMC16-MMT, but at 180 °C, the R-phase is enhanced instead. In Situ Crystallization Behaviors of PEN and the Hybrids. The conformational changes of the PEN chains during the course of crystallization were probed via high-temperature FTIR (13) Ouchi, I.; Hosoi, M.; Shimotsuma, S. J. Appl. Polym. Sci. 1977, 21, 3445.

1704 Langmuir, Vol. 23, No. 4, 2007

Chua and Lu

Figure 3. FTIR spectra of (a) PEN, (b) PEN/IMC16, (c) PEN/Na-MMT, and (d) PEN/IMC16-MMT in the (A) melt state and after melt crystallization at (B) 200 °C and (C) 180 °C: (i) 1485-1420 cm-1, (ii) 1010-800 cm-1.

spectroscopy. To monitor the conformational changes, the evolution of the crystalline- and amorphous-dependent bands in the region of 850-800 cm-1 was plotted as a function of crystallization time at 200 and 180 °C (Figure 4). In this frequency range, the band associated with the CH2 rocking mode is split into two (for the β-crystal form) or three (for the R-crystal form) bands. The additional bands are at around 832 cm-1 (for the β-crystal form) and 838/811 cm-1 (for the R-crystal form). Because the bands at 832 and 838 cm-1 are located very close to each other, only one band is observed in the

FTIR spectra, even if the sample contains a mixture of R- and β-crystal forms; however, the relative proportions of each phase can still be deduced from the position of this band. To qualitatively compare the conformational changes in the polymer backbone over the course of crystallization, the normalized peak heights of the crystalline-sensitive band at ∼832-838 cm-1 are plotted as a function of crystallization time (Figure 5). To directly monitor the development of the two polymorphic forms during crystallization, high-temperature XRD was also

Polymorphism BehaVior of PEN/Clay Nanocomposites

Langmuir, Vol. 23, No. 4, 2007 1705

Figure 4. Time-resolved spectra of (a) PEN, (b) PEN/IMC16, (c) PEN/Na-MMT, and (d) PEN/IMC16-MMT in the range of 850-800 cm-1 during melt crystallization at (i) 200 °C and (ii) 180 °C. The spectra were arranged with a 90 s interval.

performed. The development of the X-ray diffractograms with time during the initial 2000 s of the melt crystallization is displayed in Figure 6. The curves at t ) 0 s correspond to the scans obtained while the samples were cooling from ∼240 °C to the desired crystallization temperature. Due to the limited cooling rate (20 °C/min) that can be achieved by the high-temperature setup in the X-ray diffractometer, the materials

tend to crystallize substantially before they reach 180 °C so that the initial stages of melt crystallization at 180 °C could not be captured properly. The X-ray data at 180 °C are thus not reported here. The development of the R- and β-crystal forms with time is followed by monitoring the intensities of the (010) and (020) peaks of the R- and β-phases, respectively (Figure 7). The relative

1706 Langmuir, Vol. 23, No. 4, 2007

Chua and Lu Table 2. Hot Crystallization Temperatures of PEN and Its Hybrids under Dynamic Cooling at 5 °C/min

hot crystallization temperature (°C)

PEN

PEN/ IMC16

PEN/ Na-MMT

PEN/ IMC16-MMT

224

235

232

217

Table 3. Wavenumber at the Approximate Peak Position of the υas(CH2) Band in IMC16-MMT and IMC16 at Various Temperature Points of the Melt Crystallization Processa υas (cm-1) temp (°C)

IMC16-MMT

IMC16

180 200 280

2927.5 ( 0.4 2927.6 ( 0.3 2927.3 ( 0.2

2925.4 ( 0.2 2925.1 ( 0.2

a Errors given represent 1 standard deviation from the mean, based on three independent measurements.

Table 4. Extent of Thermal Degradation in IMC16-MMT during the Simulated Melt Compounding and Melt Crystallization Conditions, as Evaluated by TGA weight lossa (%)

condition melt compounding melt crystallization

200 °C 180 °C

5.9 4.7 4.6

a The amount of degradation is expressed in terms of the percentage weight loss in the original IMC16-MMT. All values stated are the average obtained from three independent measurements.

Figure 5. Normalized peak heights of the crystalline-sensitive band at ∼832-838 cm-1 as a function of crystallization time at (a) 200 °C and (b) 180 °C.

amounts of R- and β-phases formed at a particular time t are measured by an index, χ(t), where

χR(t) )

χβ(t) )

IR(010)(t) IR(010)(∞) + Iβ(020)(∞) Iβ(020)(t) IR(010)(∞) + Iβ(020)(∞)

I(t) refers to the intensity of the respective peaks at time t after subtraction of the amorphous background, and I(∞) refers to the final intensity of the peak after crystallization has reached completion. In the following sections, the in situ crystallization behaviors of each system will be examined toward providing insights into the underlying mechanisms by which clay alters the type of crystalline phase formed in the PEN matrix, in particular the role played by the surfactant-modified clay surface. PEN/IMC16. In situ XRD confirms that the R-phase is preferentially formed in PEN/IMC16 during melt crystallization at 200 °C. In addition, it reveals that PEN/IMC16 exhibits the highest initial crystallinity at t ) 0 s (Figure 6). In situ FTIR also

shows that the system exhibits more rapid conformational changes with respect to pure PEN. The higher crystallization rate in PEN/IMC16 is consistent with the higher hot crystallization temperature observed from DSC cooling experiments (Table 2). The faster overall crystallization rates in PEN/IMC16 can be attributed to the plasticizing effect caused by the dispersion of the IMC16 molecules in PEN at the molecular level. The plasticizing effect did not change the thermodynamics of the system however (the β-phase is still enhanced at higher temperatures and hence remains as the more thermodynamically stable phase); rather, it has altered the kinetics of the system such that the R-phase now grows much faster than the β-phase in comparison to those of pure PEN (Figure 7a,b). This is because the R-phase is the kinetically favored phase with a smaller free energy of activation for a chain crossing the phase boundary to the crystals.10 The increase in chain mobility would therefore increase the crystallization rate of the R-phase to a greater extent than for the β-phase. The addition of IMC16 thus gives rise to an R-enhancement effect in PEN. PEN/Na-MMT. The crystallization behavior of PEN/Na-MMT reveals that pristine Na-MMT exerts a heterogeneous nucleating effect on PEN, with the β-crystal phase being preferentially nucleated at both crystallization temperatures. This is evidenced by the rapid conformational changes PEN/Na-MMT displayed in real-time FTIR spectroscopy (Figure 5), its fast initial crystallization rates in high-temperature XRD (Figure 7c), and its higher hot crystallization temperature (Table 2). It is worth noting that heterogeneous nucleation-induced β-phase enhancement in PEN has also been observed with other types of nucleating agents.14 The effect is believed to be related to the surface (14) Gao, X.; Liu, R.; Yu, Y.; Jin, M.; Bu, H. J. Polym. Sci., Part B: Polym. Phys. 2004, 42, 296.

Polymorphism BehaVior of PEN/Clay Nanocomposites

Langmuir, Vol. 23, No. 4, 2007 1707

Figure 6. X-ray diffractograms of (a) PEN, (b) PEN/IMC16, (c) PEN/Na-MMT, and (d) PEN/IMC16-MMT obtained during melt crystallization at 200 °C.

geometry or crystal structure of the nucleating agents, although further studies are required to provide experimental evidence for the claims. PEN/IMC16-MMT. Figure 5 shows that the intensity of the crystalline-sensitive band increases most rapidly in PEN/IMC16MMT. This indicates that the surfactant-modified surface can induce more rapid conformational changes of PEN chains than the untreated clay surface. In addition, the polymer chain conformations that are induced by the presence of IMC16-MMT tend to vary with temperature. At 200 °C, the favored chain conformations are closer to that of the β-crystal form (as indicated by the band position at 832 cm-1 in Figure 4d, panel i); at 180 °C, however, the R-crystal conformations are favored instead (as suggested by the band position at 838 cm-1 in Figure 4d, panel ii). High-temperature XRD confirms that the formation of the β-crystalline phase is favored during melt crystallization at 200 °C. In contrast to the case of PEN/Na-MMT or PEN/IMC16, however, the rapid conformational changes observed in realtime FTIR spectroscopy are not accompanied by a corresponding appearance of significant Bragg peaks in the XRD patterns during the very initial stage of the crystallization process (at t ) 0 s). The slower overall crystallization rate of PEN/IMC16-MMT is also supported by its lower hot crystallization temperature. The apparent different observations obtained with in situ FTIR, XRD,

and DSC measurements can be rationalized by understanding that the “crystallinities” detectable by the three techniques are in fact different. FTIR measures the concentration of the “crystalline isomer”, which could be a short-range order or intramolecular phenomenon. In contrast, both XRD and DSC require order to be present on a longer range before it can be detected. The results thus indicate that although local skeletal conformational changes are induced much faster in PEN/IMC16MMT, the development of long-range order is somehow hindered on the surfactant-modified clay surface in comparison to the untreated clay surface. As the initial conformational changes observed in PEN/IMC16MMT are not accompanied by the formation of three-dimensional order, they in fact indicate the ordering present in the PEN chains during the induction period of crystallization. The abrupt change in the concentration of R- and β-“crystalline conformers” between 200 and 180 °C thus reflects a change in the equilibrium nuclei distribution for the two phases. To investigate the origins of this phenomenon, the thermally induced structural changes of the surfactant-modified clay surface and their influence on the polymer/organoclay interface were further examined. Effect of the Melt Crystallization Temperature on the Surfactant Conformation in IMC16-MMT. To understand how the structure of IMC16-MMT changes with temperature, hightemperature FTIR was employed to probe the molecular

1708 Langmuir, Vol. 23, No. 4, 2007

Chua and Lu

Figure 7. Development of the R- and β-crystallinities with time for (a) PEN, (b) PEN/IMC16, (c) PEN/Na-MMT, and (d) PEN/IMC16MMT melt crystallized at 200 °C.

Figure 8. Representative FTIR spectra of IMC16-MMT in the region 2950-2900 cm-1 acquired at (a) 280 °C and after subsequent cooling to (b) 200 °C and (c) 180 °C.

conformation of IMC16 surfactant molecules at the various melt crystallization temperatures.15 Figure 8 shows the high-wavenumber CH stretching region of IMC16-MMT at various temperatures of the melt crystallization process. The band at ∼2920 cm-1 arises from the CH2 asymmetric, (15) Zhao, Z.; Tang, T.; Qin, Y.; Huang, B. Langmuir 2003, 19, 9260. Vaia, R. A.; Teukolsky, R. K.; Giannelis, E. P. Chem. Mater. 1994, 6, 1017.

υas(CH2) stretch. In general, the bandwidth and band position of υas(CH2) are sensitive to the trans/gauche conformer ratio, as well as the packing density of methylene chains. A broadening and shift of this band to a higher frequency signify that the intercalated methylene chains change from a more ordered structure with more trans conformations to a more disordered structure with more gauche conformations.16 The frequencies of the IMC16-MMT υas(CH2) band at the various temperatures of interest are summarized in Table 3. Table 3 shows that there exists no significant difference in the conformations of the surfactant molecules in IMC16-MMT at 200 and 180 °C. To ascertain whether the minimal conformational change experienced by the surfactant molecules at these two temperatures is due to their restricted conformational freedom, as a result of their confinement within the clay galleries (before the galleries are expanded by the intercalation of the polymer chains), the FTIR spectra of the IMC16 salt were also acquired at 200 and 180 °C and the spectral data listed in Table 3. From Table 3, it is observed that, even without the geometric constraining effect of the clay layers, the IMC16 chains still undergo very few conformational changes between 200 and 180 °C. This indicates that the difference in the concentration of Rand β-crystalline conformers observed in PEN/IMC16-MMT between the two melt crystallization temperatures was not induced by the conformational changes experienced by the surfactant chains, as has been previously suspected.12 (16) Snyder, R. G.; Strauss, H. L.; Elliger, C. A. J. Phys. Chem. 1982, 86, 5145. Li, Y.; Ishida, H. Langmuir 2003, 19, 2479. Gelfer, M.; Burger, C.; Fadeev, A.; Sics, I.; Chu, B.; Hsiao, B. S.; Heintz, A.; Kojo, K.; Hsu, S.-L.; Si, M.; Rafailovich M. Langmuir 2004, 20, 3746.

Polymorphism BehaVior of PEN/Clay Nanocomposites

Figure 9. TGA profiles of IMC16-MMT under the simulated conditions of melt crystallization at (i) 200 °C and (ii) 180 °C. Time ) 0 min corresponds to the end of the predrying step at 140 °C.

Thermal Degradation Behaviors of IMC16-MMT. To examine the weight loss behaviors of IMC16-MMT under the melt compounding and melt crystallization conditions, TGA was performed, simulating the temperature profiles of these two processes. Table 4 summarizes the extent of degradation that was observed in IMC16-MMT under the simulated melt compounding and melt crystallization conditions. From Table 4, it is observed that significant degradation has occurred in IMC16-MMT, during both melt compounding and melt crystallization. There is, however, no significant difference in the extent of degradation between melt crystallization at 200 and 180 °C. In fact, the degradation that occurs during melt crystallization takes place during the melting process at 300 °C, and no further degradation occurs during crystallization at 200 or 180 °C, as shown in Figure 9. The finding thus rules out the possibility that the abrupt change in the concentration of R- and β-crystalline conformers formed between 200 and 180 °C is due to a difference in the degradation behavior of IMC16-MMT at these two temperatures. Role of the Clay Surface Modification. As discussed earlier, although short-range molecular ordering is induced very quickly during the induction period of crystallization in PEN/IMC16MMT, the development of long-range order is somehow hindered; this indicates that the formation of stable nuclei is more difficult at the PEN/IMC16-MMT interface than at the PEN/Na-MMT interface. The hindered formation of stable nuclei on the surfactant-modified surface is likely to be the major cause of the large change in equilibrium nuclei distribution for the two phases

Langmuir, Vol. 23, No. 4, 2007 1709

between 200 and 180 °C, as this effect may tend to have a more pronounced influence on the thermodynamically less stable R-phase. At 200 °C, the near absence of the 811 cm-1 band at the initial stage of crystallization (Figure 4d panel i) indicates that most crystalline conformers of the less stable R-phase are unable to develop into stable nuclei of a critical size on the surfactant-modified clay surface, so that its concentration remains extremely low. At 180 °C on the other hand, both types of nuclei could reach their critical size; diffusion thus becomes the controlling factor, so the kinetically favored R-phase becomes dominant. The above discussion opens up an important question as to what constrains the formation of stable nuclei at the PEN/IMC16MMT interface. One hypothesis is that the crystals are initiated from polymer chains which are intercalated in clay galleries of nanometer thicknesses.17 To clarify the possible role that the intercalated polymer chains may play in the nucleation of the PEN crystallites, a comparison is made between the morphologies and polymorphism behaviors exhibited by PEN/IMC16-MMT and a closely related system, PEN/IM2C10-MMT.12 Surfactant 1,3-didecyl-2-methylimidazolium (IM2C10) has two decyl tails, which makes it less compatible with PEN as compared to IMC16. The clay basal spacing in PEN/IM2C10-MMT thus remained relatively unchanged after melt compounding with PEN. PEN/ IM2C10-MMT shows a similar crystallization behavior to that of PEN/IMC16-MMT, but its Kβ is even more sensitive to the crystallization temperature (it changes from 78% at 200 °C to ∼0% at 180 °C).12 This indicates that the hindered formation of stable nuclei at the PEN/IMC16-MMT interface does not originate from the confinement of PEN chains in the clay galleries. The hindrance to the formation of stable nuclei should thus come from the flexible nature of the surfactant-modified clay surface, which has a poorer structural match with the crystal phases in comparison to the untreated clay surface.18 In polymer/clay nanocomposites, the cationic surfactants typically (1) reside at the intergallery spaces (intercalated) or (2) are ionically bound on the surface of the clay layers.19 On the basis of the results presented above, however, it is found that significant degradation and/or desorption of the surfactants has occurred during both melt compounding and melt crystallization. As a result, significant amounts of small molecules from the degradation and desorption of the surfactants may also be found (3) in the bulk of the PEN matrix (mainly from the surfactants that degrade or desorb during melt compounding) and (4) at close Vicinity to the clay surface (mainly from the surfactants that degrade during melt crystallization). The concentration of the small molecules (either the surfactants or their degradation products) in the bulk of PEN/IMC16-MMT might be much lower than that found in the bulk of PEN/IMC16; however, their concentration at the polymer/clay interfacial regions could be fairly significant. The effect of the intercalated surfactant molecules (1) on the polymorphism behavior of PEN may be ignored, while the small molecules in the bulk (3) may lead to a plasticizing effect in the bulk of the PEN matrix. On the other hand, the combination of the small molecules located at (2) and (4) may give rise to a unique interface, which is different from the interface formed by the untreated Na-MMT surface and the unplasticized melt. First, on the surfactant-modified clay surface, although the nucleation is still “heterogeneous” in nature, the formation of stable nuclei (17) Nam, P. H.; Maiti, P.; Okamoto, M.; Kotaka, T.; Hasegawa, N.; Usuki, A. Polymer 2001, 42, 9633. (18) Liu, X. Y. Langmuir 2000, 16, 7337. (19) Ka´da´r, F.; Sza´zdi, L.; Fekete, E.; Puka´nszky, B. Langmuir 2006, 22, 7848. Hanley, H. J. M.; Muzny, C. D.; Butler, B. D. Langmuir 1997, 13, 5276.

1710 Langmuir, Vol. 23, No. 4, 2007

is hindered. Second, with a highly plasticized interfacial region, the difference between the crystallization rates of the two phases could be larger. The combination of the heterogeneous nucleating and plasticizing effects causes the interface to be more sensitive to temperature changes than the interface formed by Na-MMT and the unplasticized melt. At 200 °C, the heterogeneous nucleating effect is dominant, but the R-phase nuclei are largely unstable due to the flexible nature of the surfactant-modified clay surface. This is supported by the fact that increasing the flexibility of the surfactant, for example, in the case of PEN/ IM2C10-MMT, results in an increase in Kβ at 200 °C. At 180 °C, on the other hand, both types of nuclei are stable, so the plasticizing effect becomes dominant. The kinetically favored R-crystal phase thus preferentially forms at both the interface and the bulk.

Conclusion In situ FTIR spectroscopy and XRD have helped to provide new insight into the origins of the unusual temperature-dependent polymorphic behavior observed from PEN/IMC16-MMT. In situ FTIR spectroscopy of PEN/IMC16-MMT reveals an abrupt change in the concentration of R- and β-crystalline conformers

Chua and Lu

with crystallization temperature during the induction period of crystallization. This phenomenon is attributed to the hindered formation of stable nuclei at the surfactant-modified clay surface. In addition, the highly plasticized PEN/IMC16-MMT interface caused by surfactant degradation could increase the crystallization rate of the kinetically favored R-phase more than that of the β-phase. It is this combination of the hindered heterogeneous nucleating and plasticizing effects which causes the PEN/ organoclay interface to be more sensitive to temperature. Acknowledgment. We thank Nanyang Technological University (NTU) for issuing Academic Research Fund RG20/02 to support this work. Y.C.C. thanks the Agency for Science, Technology and Research, Singapore, for providing her Ph.D. scholarship in the course of this work. Supporting Information Available: DSC thermograms of IMC16 and IMC16-MMT, glass transition temperatures of PEN and the hybrids, and TGA profiles of IMC16-MMT under the simulated melt compounding conditions. This material is available free of charge via the Internet at http://pubs.acs.org. LA0626048