Thermally stable exfoliated poly(ethylene terephthalate) (PET) nanocomposites as prepared by selective removal of organic modifiers of layered silicate

Available online at www.sciencedirect.com

Polymer Degradation and Stability 93 (2008) 252e259 www.elsevier.com/locate/polydegstab

Thermally stable exfoliated poly(ethylene terephthalate) (PET) nanocomposites as prepared by selective removal of organic modifiers of layered silicate Jae Woo Chung, Se-Bum Son, Sang-Wook Chun, Tae Jin Kang, Seung-Yeop Kwak* Department of Materials Science and Engineering, Seoul National University, San 56-1, Sillim-dong, Gwanak-gu, Seoul 151-744, Republic of Korea Received 2 July 2007; received in revised form 17 September 2007; accepted 23 September 2007 Available online 7 October 2007

Abstract Exfoliated poly(ethylene terephthalate) (PET) nanocomposite excluding organic modifier (M-PetLSNeom) was successfully prepared by the melt processing via solution method with solventenonsolvent system. PET nanocomposites including organic modifier (M-PetLSNiom and D-PetLSN) as counterpart of M-PetLSNeom were prepared by using the melt processing via solution method without solventenonsolvent system and the only conventional direct melt mixing process, respectively. From elemental analysis (EA) and thermogravimetric analysis (TGA), organic modifier in M-PetLSNeom was confirmed to be well removed by solution method with solventenonsolvent system. Then, it was found that M-PetLSNeom and M-PetLSNiom had exfoliated structure by wide angle X-ray diffraction (WAXD) and high-resolution transmission electron microscopy (HR-TEM), whereas no expansion of gallery height was observed for D-PetLSN. To elucidate the effect of organic modifier on the physical properties of PET nanocomposites, the crystallization behavior, optical transparency, thermal stability, and mechanical properties of M-PetLSNeom, M-PetLSNiom, D-PetLSN, and neat PET were evaluated by differential scanning calorimetry (DSC), UVevisible (UVevis) spectroscopy, TGA, and universal testing machine (UTM). All of the PET nanocomposites exhibited faster crystallization kinetics and better thermal and mechanical properties compared to neat PET due to the presence of silicate layer in PET. However, M-PetLSNiom and D-PetLSN including organic modifier showed lower crystallization constant rates, longer crystallization half times, and poorer optical, thermal, and mechanical properties than M-PetLSNeom. These results were ascribed to the thermal decomposition of the organic modifiers presented in M-PetLSNiom and D-PetLSN during the melt processing. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Exfoliated PET nanocomposites; Organic modifiers; PET; Thermal stability

1. Introduction PET is extensively used for packaging films, beverage bottles, engineering components, and fibers for apparel, on account of its excellent chemical resistance, thermal stability, and mechanical properties [1]. However, the use of PET under more severe conditions requires that its various physical properties, such as optical, thermal, mechanical, and barrier properties, should be much enhanced. Hence, considerable effort

* Corresponding author. Tel.: þ82 2 880 8365; fax: þ82 2 885 1748. E-mail address: [email protected] (S.-Y. Kwak). 0141-3910/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymdegradstab.2007.09.009

has been devoted to improve the various physical properties of PET through blending with high performance polymers [2,3] and mixing with additives such as filler. Recently, the commercial importance of PET in industrial fields has driven intensive investigation into PET-layered silicate nanocomposites (PetLSNs), because these composite materials exhibit far superior physical properties to PET blends with microsized additives [4e9]. Layered silicates have layers with thicknesses in the order of 1 nm and have very high aspect ratios (e.g., 10e1000), and interlayer spacing between the stacked layers of about 1 nm. The distinctive features of the layered silicates result in the nanocomposites having two possible structures, namely intercalated or exfoliated structures. The

J.W. Chung et al. / Polymer Degradation and Stability 93 (2008) 252e259

intercalated nanocomposites show regularly alternating silicate and polymer layers with a repeat distance of a few nanometers, whereas the individual layers in the exfoliated nanocomposites are irregularly delaminated and dispersed in a continuous polymer matrix. Such structural differences play a key role in the enhancement of the nanocomposite properties. Generally, exfoliated nanocomposites have superior mechanical properties to intercalated nanocomposites, because of the larger surface area between the reinforcement phase and the polymer matrix. However, since the substrates of silicate layer are hydrophilic and the gallery height between the silicate layers is very narrow, it is difficult for the hydrophobic PET molecular chain to penetrate into the layered silicates. Therefore, it is necessary to introduce organic modifiers to change the layered silicate substrate from hydrophilic to hydrophobic in order to lead to the insertion of PET molecular chain into the gallery between the layered silicates. This process may provide a favorable way to disperse the layered silicates in the PET matrix [10]. Recently, several methods using organically modified silicate layer have been developed to fabricate the exfoliated polymer-layered silicate nanocomposites (PLSNs) such as in situ polymerization, solution mixing, and direct melt mixing [11e13]. The first two approaches are not appropriate for commercial production on account of difficulties associated with identifying a suitable monomer for the polymerization and a solvent compatible with both the polymer and the silicate. On the other hand, the direct melt mixing method is very useful and simple. In this method, polymers are inserted into the layered silicate at a temperature above the softening or melting of the polymer. However, the main limitations of direct melt mixing method are the localized dispersion of silicate layers into the polymer matrix and the thermal decomposition of the organic modifiers [14e18]. Xie et al. [16,17] reported that the degradation of organic modifiers in PLSNs was correlated with several factors, including the residence time during processing, processing temperature, and the type of ammonium organic modifier used. Davis et al. studied the degradation pathway of PA6/MMT nanocomposites with a focus on the degradation mechanism of organic modifiers in PA6/ MMT nanocomposites at high processing temperatures [18]. Then, it was found that significant thermal degradation occurred in the PA6/MMT nanocomposites during processing at 300  C, and that thermal decomposition may result from hydrolytic peptide scission. Zhang et al. [8,9] developed a new synthetic strategy for the homogeneous dispersion of layered silicate into PET by in situ polymerization, but reported that PetLSNs failed to exhibit optical transparency due to thermal decomposition of the organic modifiers in the PetLSNs during the polymerization procedure at 280  C [7e9], even though the tensile strength increased. Their results revealed that the high processing temperature causes the thermal decomposition of the organic modifiers in PetLSNs. Hence, the main objective of this study is to present a novel approach for the preparation of exfoliated PetLSNs in the absence of the organic modifiers. Additionally, we showed the detailed effects of organic modifiers on the various physical properties of PetLSNs in this study.

253

2. Experimental section 2.1. Materials PET with ca. 255  C of a melting point was obtained from the Huvis Chemical Company (Korea). Organically modified layered silicates (OLS) occupied with organic modifiers (polyoxypropylene methyldiethylammonium cations) between the layered silicate galleries were kindly supplied by Coop Chemical Co. The OLS had approximately 2.8 nm and 4.4 nm of gallery height. The weight ratio of layered silicates/organic modifiers in the OLS was verified to be about 0.4/0.6 by thermogravimetric analysis (TGA). Chloroform (99%) and methanol (99%) were purchased from Daejung Chemicals & Metals (Korea) and trifluoroacetic acid (TFA) used for removing the organic modifier ionically attached on the layered silicate was purchased from SigmaeAldrich Co. All materials were used without any additional purification. 2.2. Sample preparation OLS were added in excess chloroform and stirred at 25  C for 1 h. TFA was poured in OLS/chloroform solution and stirred for 5 min, and then neat PET was added to OLS/chloroform/TFA solution, and dissolved by stirring for an additional 1 h (neat PET/OLS ¼ 92/8, w/w%). PET/OLS/chloroform/ TFA solution was added dropwise to the cold methanol to obtain PET nanocomposite excluding organic modifier (S-PetLSNeom), as shown in Scheme 1. Then, precipitated materials were collected by filtration and dried in a vacuum oven at 80  C for 12 h. On the other hand, PET nanocomposite including organic modifier (S-PetLSNiom) was prepared by removing the solvent from prepared PET/OLS/chloroform/TFA solution in a hood at 30  C for 48 h. Remained white powder was collected and dried in a vacuum oven at 80  C for 12 h. Acquired S-PetLSNiom and S-PetLSNeom, individually, mixed with neat PET by a twin screw extruder (Hakke CTW-100) of co-rotating mode at 270  C using a screw speed of 80 rpm (neat PET/S-PetLSN ¼ 1/0.333, w/w%). Under these conditions, the mean value of the residence time in the extruder for PET was 350 s. After mixing, obtained materials were cooled in a water bath and then pelletized. Melt processed PET nanocomposite obtained from mixing of S-PetLSNiom and S-PetLSNeom with neat PET is denoted as M-PetLSNiom and M-PetLSNeom, respectively. Additionally, PET/OLS (1/0.067, w/w%) mixture (D-PetLSN) was also prepared as counterpart of M-PetLSNs by the direct melt mixing under the same conditions as used for the M-PetLSNs. 2.3. Characterization The presence and absence of organic modifiers in the S-PetLSNs were determined by elemental analysis (EA) and thermogravimetric analysis (TGA). EA was carried out on an EA 1110 CE elemental analyzer, and TGA was carried out using the TA instruments TGA 2050 thermal analyzer from room temperature to 600  C at a heating rate of 2  C min1 under N2

254

J.W. Chung et al. / Polymer Degradation and Stability 93 (2008) 252e259

Scheme 1. Schematic illustration of the preparation of exfoliated PET nanocomposites excluding organic modifier using solvent method with solventenonsolvent system (S-PetLSNeom).

flow. To investigate the silicate structure of PetLSNs, wide angle X-ray diffraction (WAXD) patterns were obtained at room temperature on a MAC/Science MXP 18XHF-22SRA diffractometer with a Cu Ka radiation source (wavelength ¼ 0.154 nm). The supplied voltage and current were set to 40 kV and 40 mA, respectively. The diffraction angle was scanned at a rate of 2q ¼ 1 /min between 2q ¼ 2 and 10 . High-resolution transmission electron microscopic (HRTEM) analysis was performed with a Jeol JEM 2011 instrument at 200 kV accelerating voltage. For the HR-TEM observation, ultra-thin cross-sections of the specimens were obtained by using a Leica Ultracut UCT ultracryomicrotome. Differential scanning calorimetric (DSC) analysis was performed using a TA instruments DSC 2920 under N2 flow. All of the PetSLNs were first heated to 300  C and subsequently kept for 1 min so as to erase the thermal history. To characterize the nonisothermal crystallization behaviors of the PetLSNs, DSC data of neat PET, M-PetLSNs and D-PetLSN measured with different cooling rates, ranging from 2  C min1 to 10  C min1, were employed to the modified Avrami equation [19,20], which extends the isothermal crystallization kinetics model suggested by Avrami to non-isothermal conditions, as follows: 1  XðtÞ ¼ expðKtn Þ

ð1Þ

where X(t) is the relative crystallinity, n is the Avrami exponent, which depends on the shape of the growing spherulites, and K is the growth rate constant of the spherulites. Eq. (1) can be transformed as follows by taking the double logarithmic form: log½  lnð1  XðtÞÞ ¼ log k  n log t

ð2Þ

Cebe [21] and Herrero and Acosta [22] tried to modify the Avrami equation to non-isothermal crystallization conditions using the relationship between crystallization time, t, and cooling rate, R, as follows:

t¼

T0  T R

ð3Þ

where T is the temperature at time t, and T0 is the temperature at the initiation of crystallization, t ¼ 0. By plotting log[ln(1  X(t))] vs. log t at a given temperature, n and K can be obtained from the slope and intercept, respectively. The rate of non-isothermal crystallization depends on the cooling rate; hence, K should be altered to account for this [23]. The corrected crystallization rate constant, k0 , is given by the following equation: log k 0 ¼

log K R

ð4Þ

The half time of crystallization, t1/2, which is the time required for growing 50% of crystalline, is calculated from the measured parameters, n and k0 , as follows: t1=2 ¼

 1=n ln 2 k0

ð5Þ

To observe the optical properties, neat PET, M-PetLSNs, and D-PetLSN were pressed at 280  C on 5000 psi for 5 min and then, quenched by cold water. The yellowness values ascribed to thermal decomposition of organic modifier were calculated by Macbeth Spectrophotometer Color-Eye 3000 using a standard ASTM D-1925 and E-313. The thermal stability of the neat PET, M-PetLSNs, and D-PetLSN was investigated by TGA under a stream of nitrogen gas at a heating rate of 2  C min1. To quantitatively compare their thermal stability, activation energy for pyrolysis was calculated from the HorowitzeMetzger method using the following equation:
Ea ðT  Tmax Þ ln lnð1  aÞ1 ¼ 2 RTmax

ð6Þ

where a is the decomposition fraction, Ea is the activation energy for pyrolysis, T is the temperature, Tmax is the maximum

J.W. Chung et al. / Polymer Degradation and Stability 93 (2008) 252e259

30 25

S-PetLSNeom 20 15

S-PetLSNiom 10

OLS

5 0

3. Results and discussion To investigate the presence and absence of organic modifiers in PetLSNs, it is necessary to evaluate the nitrogen content in S-PetLSNs using EA, because the organic modifier (polyoxypropylene methyldiethylammonium cations) attached on the silicate layers’ surface contains ammonium groups. Table 1 exhibits the nitrogen content in both S-PetLSNs. As shown in Table 1, S-PetLSNiom was found to contain 0.3 wt% of nitrogen, whereas S-PetLSNeom showed no evidence of nitrogen. Lee and Char [25] reported that the de-intercalation of the organic modifiers from layered silicates was affected by solvent acidity, and that de-intercalation occurred to a greater extent in more strongly acidic solvents. Thus, EA results indicated that the organic modifiers in S-PetLSNeom were completely removed by the strong acid (TFA) to break ionic bonding between the organic modifiers and OLS. In addition, the absence of organic modifiers in S-PetLSNeom was verified by the TGA method. The derivative thermogravimetry (DTG) curves (shown in Fig. 1) obtained from the TGA data revealed that S-PetLSNiom had both thermal decomposition peaks at 355  C and 452  C, corresponding to characteristic thermal degradation peak of OLS and PET, respectively. However, S-PetLSNeom was found to have the only thermal decomposition peak of PET [15,18]. From the above results, it is considered that the organic modifiers attached to the layered silicate substrates were completely removed by TFA during the preparation of S-PetLSNeom. The silicate layer nanostructures in PetLSNs were characterized using WAXD and HR-TEM. In the WAXD patterns (Fig. 2), two peaks for the OLS appeared at 2q ¼ 2.27 and 4.56 , which corresponded to 4.4 nm and 2.8 nm of gallery heights, respectively. However, no peak was observed in the WAXD patterns for S-PetLSNs. This implies that periodicity of silicate layers disappears due to sufficient dispersion of individual silicate layer attributed to insertion of PET chain into the gallery of layered silicates, what is called as exfoliation. In addition, M-PetLSNs were also found to have the exfoliated structure by WAXD patterns where pristine peaks of OLS were Table 1 Nitrogen contents of S-PetLSNeom and S-PetLSNiom

S-PetLSNeom S-PetLSNiom

35

dwt/dt (wt% °C–1)

decomposition temperature, and R is the molar gas constant. According to the HorowitzeMetzger method, a plot of ln(ln(1  a)1) vs. (T  Tmax) should give a straight line whose slope is equal to the Ea for pyrolysis [24]. Mechanical properties were evaluated on the basis of tensile tests performed on a LLOYD LR10K universal testing machine (UTM). A tensile load cell of 100 N was used, and a standard ASTM D412-92T dumbbell was employed with a 15.5 mm gauge length and a 20 mm min1 cross-head speed. To obtain more precise and reasonable tensile test results, five specimens were measured and averaged for each sample.

255

Silicate loading (wt%)

Nitrogen content (%)

8 8

e 0.3

300

400

500

600

Temperature (°C) Fig. 1. Derivative thermogravimetry (DTG) curves for OLS, S-PetLSNeom, and S-PetLSNiom.

disappeared, because they were fabricated from S-PetLSNs already having exfoliated structure. In contrast, two basal peaks corresponding to peaks of OLS still remained for D-PetLSN, indicating that PET chain did not insert into gallery of OLS, and gallery height was not expended. Furthermore, as shown in Fig. 3, it was verified by HR-TEM images that the silicate layers in S-PetLSNs and M-PetLSNs were fully delaminated and exfoliated in the PET matrix, which are consistent with WAXD. Therefore, these reveal that our melt processing via solvent method is a valid fabricating method for producing the exfoliated PetLSNs, while conventionally direct melt mixing between PET and silicate failed to obtain the exfoliated PetLSN. To evaluate the effect of organic modifiers in the PetLSNs on the crystallization behaviors, the non-isothermal crystallization behaviors were investigated on the basis of a modified Avrami analysis using DSC. In Fig. 4, the DSC exothermal thermograms as a function of cooling rate were transformed to log[ln(1  X(t))] vs. log t plots using the modified Avrami equation. Then, crystallization kinetic parameters of neat PET, M-PetLSNeom, M-PetLSNiom, and D-PetLSN are summarized in Table 2. As shown in Table 2, the n values were in the range of 2.7 e 3.2 for neat PET and 2.7 e 3.9 for M-PetLSNeom, M-PetLSNiom, and D-PetLSN at a given cooling rate. These values imply that typical three-dimensional spherulitic crystalline was generated in all of the PetLSNs [26e28]. However, M-PetLSNs and D-PetLSN showed a larger k0 and a shorter t1/2 than neat PET at a given cooling rate. This result suggests that the silicate layers in M-PetLSNs and D-PetLSN act as efficient nucleating agents for crystallization [15,28]. In particular, M-PetLSNiom and D-PetLSN exhibited a lower k0 and a shorter t1/2 than M-PetLSNeom at a given cooling rate. This is explained by the fact that thermally decomposed remnants of the organic modifiers in M-PetLSNiom and D-PetLSN resulted from the melt processing at high temperature may act as crystallization inhibitors. Thus, M-PetLSNeom showed the fastest crystallization kinetics [29] among melt processed PET nanocomposites, because M-PetLSNeom only has silicate layers to act as nucleating agents in the PET matrix but not include the organic modifier

J.W. Chung et al. / Polymer Degradation and Stability 93 (2008) 252e259

256

(b)

(a)

Intensity (A.U.)

Intensity (A.U.)

OLS OLS

S-PetLSNeom

M-PetLSNiom

M-PetLSNeom

S-PetLSNiom

2

3

4

5

6

7

D-PetLSN

8

2 (degrees)

2

3

4

5

6

7

8

2 (degrees)

Fig. 2. WAXD patterns of (a) S-PetLSNs, (b) M-PetLSNs and D-PetLSN.

to hinder crystallization. The optical transparencies of neat PET, M-PetLSNs, and D-PetLSN films produced from hot pressing at 280  C for 5 min are exhibited in Fig. 5. For M-PetLSNiom and D-PetLSN, an apparent dark transparency with slight yellowing was observed, while M-PetLSNeom and neat PET showed similar transparencies [8,9,30]. To quantitatively obtain the degree of

yellowness of neat PET, M-PetLSNs, and D-PetLSN films, UVevis spectrophotometry was used. The yellowness values (Table 3) of neat PET and M-PetLSNeom were determined to be 1.26 and 1.40 by ASTM D-1925, respectively (1.38 and 1.50 for ASTM E-313, respectively). In the case of M-PetLSNiom and D-PetLSN, 11.50 and 11.71 of the yellowness values

Fig. 3. HR-TEM images of (a) S-PetLSNeom, (b) S-PetLSNiom, (c) M-PetLSNeom, and (d) M-PetLSNiom.

J.W. Chung et al. / Polymer Degradation and Stability 93 (2008) 252e259

(a)

0.5

0.0

-0.5 Cooling rate 2°C / min 4°C / min 6°C/ min 8°C / min 10°C / min

-1.0

-1.5

log (-ln (1-X(t)))

log (-ln (1-X(t)))

0.5

257

(b)

0.0

Cooling rate

-0.5

2°C / min 4°C / min 6°C / min 8°C / min 10°C / min

-1.0

-1.5 0.4

0.0

-0.4

1.2

0.8

0.0

(c)

0.5

0.0

-0.5

Cooling rate 2°C / min 4°C / min 6°C / min 8°C / min 10°C / min

-1.0

-1.5

log (-ln (1-X(t)))

log (-ln (1-X(t)))

0.5

0.4

0.8

log (t)

log (t)

(d)

0.0

-0.5

Cooling rate

-1.0

2°C / min 4°C / min 6°C / min 8°C / min 10°C / min

-1.5

-0.4

0.0

0.4

0.8

1.2

log (t)

-0.4

0.0

0.4

0.8

1.2

log (t)

Fig. 4. Modified Avrami plots of relative crystallinities (Xt) crystallized at various cooling rates for (a) neat PET, (b) M-PetLSNeom, (c) M-PetLSNiom, and (d) D-PetLSN.

were obtained, respectively (9.66 and 9.84 for ASTM E-313, respectively). These results indicate that organic modifiers in M-PetLSNiom and D-PetLSN are thermally decomposed by high temperature, and then thermally decomposed organic modifiers lead to the change of color. However, color change was observed for M-PetLSNeom, because the organic modifiers Table 2 Modified Avrami parameters of neat PET, M-PetLSNeom, M-PetLSNiom, and D-PetLSN Cooling rate ( C min1)

Sample

2

4

6

8

10

Neat PET

n k0 t1/2

2.7 0.04 2.43

3.2 0.18 1.54

2.9 0.48 1.14

3.2 0.57 1.07

3.1 0.69 1.00

M-PetLSNeom

n k0 t1/2

2.99 0.13 1.72

3.23 0.55 1.07

3.50 0.71 0.99

3.18 0.88 0.92

3.34 0.93 0.91

M-PetLSNiom

n k0 t1/2

3.23 0.10 1.94

3.32 0.40 1.17

3.03 0.69 1.02

3.96 0.82 0.94

3.21 0.89 0.93

D-PetLSN

n k0 t1/2

2.92 0.04 2.43

2.74 0.39 1.22

2.68 0.57 1.06

3.74 0.61 1.03

3.20 0.81 0.95

in M-PetLSNeom were removed by strong acid used in solvent method with solventenonsolvent system [15e18]. Table 4 shows detailed thermal property determined by non-isothermal TGA results, including the temperature at 10% of weight loss (Td10) [31], the maximum decomposition temperature (Tmax) [17,18], and the activation energy for pyrolysis (Ea,p) as the indexes of thermal stability. M-PetLSNs and D-PetLSN exhibited higher Td10, Tmax, and Ea,p values than neat PET, indicating a better thermal stability, since the silicate layers act as thermal barriers in the PET matrix [31]. But, M-PetLSNiom and D-PetLSN exhibited lower Td10, Tmax, and Ea,p values than M-PetLSNeom, on account of the thermal degradation of the organic modifiers in M-PetLSNiom and D-PetLSN. The Young’s modulus, elongation at break, and maximum stress of neat PET, M-PetLSNs, and D-PetLSN are listed in Table 5. The Young’s modulus and maximum stress are in the order of M-PetLSNeom > M-PetLSNiom > D-PetLSN > neat PET. For all of the PetLSNs, mechanical strength increased with relative to neat PET due to the presence of silicate that acts as filler. However, M-PetLSNs exhibited higher mechanical strength than D-PetLSN owing to their unique silicate structure such as exfoliation. In particular, M-PetLSNeom showed significantly higher mechanical strength than M-PetLSNiom, although

J.W. Chung et al. / Polymer Degradation and Stability 93 (2008) 252e259

258

Fig. 5. Transperencies of melt processed films of (a) neat PET, (b) M-PetLSNeom, (c) M-PetLSNiom, and (d) D-PetLSN.

Table 3 Quantitative yellowness values of neat PET, M-PetLSNeom, M-PetLSNiom, and D-PetLSN determined from UVevisible spectrophotometer Samples

ASTM D-1925

ASTM E-313

Neat PET M-PetLSNeom M-PetLSNiom D-PetLSN

1.26 1.40 11.50 11.71

1.38 1.50 9.66 9.84

Table 4 TGA results of neat PET, M-PetLSNeom, M-PetLSNiom, and D-PetLSN Samples

Td10 ( C)

Tmax ( C)

Ea,p (kJ/mol)

Neat PET M-PetLSNeom M-PetLSNiom D-PetLSN

418.2 436.5 429.3 426.3

446.3 471.2 459.8 458.6

193.7 238.2 205.1 200.1

Table 5 Mechanical properties of neat PET, M-PetLSNeom, M-PetLSNiom, and DPetLSN Samples

Young’s modulus (MPa)

Elongation at break (%)

Maximum stress (MPa)

Neat PET M-PetLSNeom M-PetLSNiom D-PetLSN

34  0.2 47  0.4 38  0.2 37  0.5

280  2 200  4 40  3 30  4

102.0  3 183.0  4 122.7  2 115.3  3

both M-PetLSNs had the same exfoliated silicate structure. This can be explained by the decrease of the mechanical properties ascribed to thermally decomposed remnants of the organic modifiers in M-PetLSNiom [8,9,32,33]. 4. Conclusion Exfoliated PET-layered silicate nanocomposites excluding (M-PetLSNeom) or including (M-PetLSNiom) organic modifiers were prepared by melt processing via solution method with solventenonsolvent system and without solventenonsolvent system, respectively. Their physical properties were compared with those of D-PetLSN produced by conventional melt mixing between silicate layer and neat PET. It was found that MPetLSNeom and M-PetLSNiom had exfoliated silicate structure, except D-PetLSN. All of the PetLSNs had faster crystallization and better thermal and mechanical properties compared to neat PET, due to the presence of layered silicates. However, MPetLSNiom and D-PetLSN were found to have poorer crystallization behavior, optical transparency, thermal stability, and mechanical properties compared to M-PetLSNeom because of the thermal decomposition of the organic modifiers during the melt processing at high temperature. Thus, the present study suggests that the exfoliated M-PetLSNeom without organic modifiers has potential features to be exploited in developing PetLSNs, because nanocomposites with these characteristics can withstand the high temperature conditions used in melt processing without weakening the various physical properties.

J.W. Chung et al. / Polymer Degradation and Stability 93 (2008) 252e259

Acknowledgements The authors of this paper would like to thank the Korea Science and Engineering Foundation (KOSEF) for sponsoring this research through the SRC/ERC Program of MOST/KOSEF (R11-2005-065). References [1] Brozenic NJ. Modern plastics encyclopedia. New York: McGraw-Hill; 1986. [2] Bedia EL, Murakami S, Kitade T, Kohjiya S. Polymer 2001;42:7299e 305. [3] Buchner S, Wiswe D, Zachmann HG. Polymer 1989;30:480e8. [4] Ke Y, Long C, Qi Z. J Appl Polym Sci 1999;71:1139e46. [5] Lorenzo ML, Errico ME, Avella M. J Mater Sci 2002;37:2351e8. [6] Yusuke I, Satoshi N, Eiichi A, Hiroshi T, Akimasa A, Akira Y. Chem Mater 2002;14:477e9. [7] Chang JH, Park DK. Polym Bull 2001;47:191e6. [8] Zhang GZ, Shichi T, Tong ZW, Takagi K. Chem Lett 2002;31:410e5. [9] Zhang GZ, Shichi T, Takagi K. Mater Lett 2002;57:1858e62. [10] Pralay M, Kazunobu Y, Masami O, Kazue U, Kazuaki O. Chem Mater 2002;14:4654e61. [11] Vaia TA, Vasudevin S, Krawiec W, Scanlon LG, Giannelis EP. Adv Mater 1994;7:154e6. [12] Giannelis EP. Adv Mater 1996;8:29e35. [13] Lim YT, Park OO. Macromol Rapid Commun 2000;21:231e5.

259

[14] Yano K, Usuki A, Okada A, Kurauchi T, Kamigaito O. J Polym Sci Part A Polym Chem 1993;31:2493e8. [15] Zanetti M, Camino G, Canavese D, Morgan AB, Lamelas FJ, Wilkie CA. Chem Mater 2002;14:189e94. [16] Xie W, Gao Z, Liu K, Pan WP, Vaia RA, Hunter D. Thermochim Acta 2001;367:339e50. [17] Xie W, Gao Z, Hunter D, Singh A, Vaia RA. Chem Mater 2001;13:2979e90. [18] Davis RD, Gilman JW, VanderHart DL. Polym Degrad Stab 2003;79:111e21. [19] Avrami M. J Chem Phys 1939;7:1103e12. [20] Avrami M. J Chem Phys 1940;8:212e24. [21] Cebe P. Polym Compos 1988;9:271e9. [22] Herrero CR, Acosta JL. Polym J 1994;26:786e96. [23] Jeziorny A. Polymer 1971;12:150e8. [24] Horowitz H, Metzger G. Anal Chem 1963;35:1464e8. [25] Lee D, Char K. Langmuir 2002;18:6445e8. [26] Lee SW, Ree M, Park CE, Hung YK, Park CS, Jin YS, et al. Polymer 1999;40:7137e46. [27] Lee SW, Lee B, Ree M. Macromol Chem Phys 2000;201:453e63. [28] Lee JW, Lee SW, Lee B, Ree M. Macromol Chem Phys 2001;202:3072e 80. [29] Stricker F, Bruch M, Mulhaupt R. Polymer 1997;38:5347e53. [30] Chang JH, Park DK. J Polym Sci Part B Polym Phys 2001;39:2581e8. [31] Jin Z, Fawn MU, Alexander BM, Charles AW. Chem Mater 2001;13:4649e54. [32] Manias E, Touny A, Wu L, Strawhecker K, Lu B, Chung TC. Chem Mater 2001;13:3516e23. [33] Strawhecker KE, Manias E. Chem Mater 2000;12:2943e9.