Printer Friendly

Solid-state blending of poly(ethylene terephthalate) with polystyrene: extent of compatibilization and its dependence on blend composition.

INTRODUCTION

The blending of immiscible polymers via conventional melt and solution routes is often hindered by technical issues such as thermal degradation, inadequate mixing, undesirable side reactions, and dispersed-phase domain coarsening during thermal shaping [1]. In response to these technical issues, immiscible polymers are often compatibilized, i.e., alloyed into a modified blend that shows stable or reproducibly modifiable morphological characteristics. The prime objectives of compatibilization include enhancement of dispersion and adhesion between the phases in the solid state to facilitate stress transfer and improve performance while ensuring that the alloying stage optimizes blend morphology [2]. Conventional compatibilization strategies include the addition of a small amount of third phase (copolymer or modifier) or reactive compatibilization (in situ reactive polymerization involving graft or block copolymers or mechanochemical bonding). The compatibilization efficiency of these copolymers, however, is limited due to the coupling of diffusion with reaction kinetics that results in the chemical compatibilizer forming a separate phase and/or forming micelles rather than anchoring the phases at the interphase.

In general, the potential synergistic effects of combining a high-modulus thermoplastic with a ductile polymer warrant a cost-effective, robust compatibilization strategy that could achieve favorable cost/performance balances suitable for materials replacement and/or new market applications [3]. In particular, the inherent difficulties in separation of semicrystalline polyethylene terephthalate (PET) from other beverage bottle components have facilitated a need to produce compatibilized mixtures of PET with other polymers for further commercial use [4].

Semicrystalline PET and atactic polystyrene (aPS) are thermodynamically immiscible polymers. Furthermore, PS/PET blends often possess a phase-separated structure in which the interphases between domains are quite fragile and mechanically weak [5]. The dispersed-phase domain size, i.e., the average size of the minority component within the matrix of the majority component, for these immiscible blends is typically 0.1-50 [micro]m [6]. PS does not contain any functionalities necessary for good adhesion with PET, so a typical compatibilization strategy for this system is interphase-reactive compatibilization by in situ polymerization during melt mixing and extrusion. The formation of block or graft copolymers at the interphase between PS and PET is characterized by anhydride or carbamate (masked isocyanate) functionalities [5, 7]. The extent of compatibility of the PS/PET system has not been quantitatively studied. Typical investigations include the determination of an optimal amount of compatibilizer needed for improvement in interphase adhesion and mechanical properties [5]. Qualitative remarks about the changes in thermal transitions are often made, particularly those concerning the crystallization behavior of PET [7]. Quantitative studies of polymer blends that include either PS or PET and another polymer, such as PP/PET and PS/PMMA, have provided insight into the effect of blending on crystallization phenomena but not compatibilization of the amorphous phases of the blend components [8]. In lieu of these shortcomings, the goals of this article are to (1) illustrate the superiority of cryogenic mechanical attrition (CMA) over extrusion in compatibilizing PS/PET blends; (2) show that CMA provides more intimate mixing and compositional control than extrusion; (3) establish and explain the compositional dependence of milled blend compatibility through trends in PET [T.sub.g].

EXPERIMENTAL

Materials and Processing

PET, |[eta]| = 0.58 dL/g, and atactic polystyrene (aPS), [M.sub.w] = 260,000 g/mol, were purchased in pellet form from Scientific Polymer Products (Ontario, NY). The [M.sub.w] for PET is ~43,000 g/mol--an order of magnitude lower than that of PS. The polymers were mechanically attritted individually in pure form or mixed and mechanically alloyed at various compositions (~30, 50, and 70% PET by weight) with a SPEX 6750 Mixer/Mill. The mill was operated at liquid nitrogen temperature (-196[degrees]C) for 30 min total milling time for each sample within polycarbonate milling vials with a stainless steel impactor at a rate of 30 Hz. The mechanism of attrition consists of compressive stresses transferred to the sample during impactor-sample-vial end collisions. The coupling of short-impact times and cryogenic temperatures leads to increases in the probability for elastic fracture [1]. The detailed mechanism of polymer fracture and cryogenic mechanical alloying is described elsewhere [9-11]. Milled powders were immediately stored in a dry environment and then kept in storage vials. A 30/70 PS/PET blend was also milled for 300 min to investigate the effect of milling time on compatibility. The attritted blends were not subjected to further processing, e.g., high-temperature annealing, prior to characterization.

Melt extrusion was performed at 220[degrees]C at 100 rpm in a ThermoHaake Rheomix counter-rotating twin-screw extruder (Waltham, MA). The diameter of the extruder is 25 mm with L/D = 16 and three heating zones. The temperature of all zones was maintained constant for all samples. Pure PS and PET were extruded first to determine the optimal temperature and rotation speed so that the PET crystals were fully molten, and that the viscosity of the melt did not decrease to the extent that the extrudate flowed too fast for adequate cooling and pelletizing. Bulk PS/PET mixtures that were 30, 50, and 70 nominal weight percent PET were fed into the extruder at ~120 g per charge. The extruded pellets were then dried in an oven at 90[degrees]C for 24 h to remove any water incorporated during extrudate cooling.

Thermal Analysis--MDSC and TGA

Thermal properties of blends and pure components were studied by modulated differential scanning calorimetry (MDSC) in a TA Instruments DSC 2920 (Newcastle, DE). Heat capacity and glass transition measurements were made between 40 and 200[degrees]C at a heating rate of 2[degrees]C/min, modulation amplitude of 0.81[degrees]C, and period of 90 s. The samples were first held isothermally at 40[degrees]C for 5 min to equilibrate. These runs were also used to determine [T.sub.g] and [T.sub.g] width for both components in addition to the change in heat capacity at [T.sub.g] ([DELTA][C.sub.p]). Dry nitrogen at a flow rate of 50 mL/min was used as the purge gas. Three samples per nominal composition (in the vicinity of 30, 50, and 70 weight percent PET) were run to obtain a mean value and standard deviation that corresponded to the compositional fluctuations due to processing. Blend compositions and thermal stability were both determined from a TA Instruments TGA 2950 Thermogravitometric Analyzer (Newcastle, DE). Dry nitrogen gas was used as the purge at 60 mL/min. The linear heat ramp was 5[degrees]C/min. and runs were performed from room temperature to 550[degrees]C. Sample sizes ranged from 5 to 10 mg.

Imaging

CMA blend particles were first embedded into a resin and subsequently microtomed into ultrathin (70-100 nm) slices for better viewing and phase contrast during transmission electron microscopy (TEM). The embedding resin used was Embed 812 (Electron Microscopy Sciences. Hatfield, PA), and the resin/blend mixtures were cured at 60[degrees]C for 48 h in an oven. These hardened mixtures were microtomed at room temperature with a Leica Reichert Autocut S microtome (Leica Microsystems, Bannockburn, IL) with a diamond knife. TEM was then performed on blend samples using a JEOL 2011 model TEM at 200 kV accelerating voltage with an Oxford Inca attachment for energy dispersive X-ray spectroscopy (EDS). The EDS spot size was 3 nm. For the TEM investigations, specimens were loaded onto 200-mesh copper grids with Formvar/Carbon support films (Ted Pella).

RESULTS AND DISCUSSION

Blend Composition Determination by TGA

The metric for compatibility in this study is based upon the percent reduction in the heat capacity change at [T.sub.g] for PET, defined as follows:

Percent or extent of compatibility = [100 x ([omega][DELTA][C.sub.p.sup.0] - [DELTA][C.sub.p])]/([omega][DELTA][C.sub.p.sup.0]) (1)

where [omega] is the measured (not nominal) weight fraction of PET in the blend, [DELTA][C.sub.p] is the PET heat capacity change at [T.sub.g] after processing (either by extrusion or milling), and [DELTA][C.sub.p.sup.0] is the heat capacity change of pure neat PET at [T.sub.g]. The numerator in this equation is also known as [delta][DELTA][C.sub.p]--the decrement in the heat capacity change because of process-induced compatibilization [12]. The fully incompatible and compatible limiting cases thus correspond to 0 and 100%, respectively. The heat capacity is associated with the change in degrees of freedom in the glass transition resulting from free volume changes in the region, so smaller free volume changes translate into smaller heat capacity changes [13].

[FIGURE 1 OMITTED]

To perform the calculations for the extent of compatibility, the compositions of the blends had to be first accurately determined because sample-to-sample fluctuations in composition have corresponding fluctuations in heat capacity changes. The composition of the blend is denoted by PET content, i.e., the weight fraction [omega] in Eq. 1. This parameter is more sensitive to underestimation, e.g., more error is introduced in the extent of compatibility if the composition is underestimated by 10 wt% PET than if it is overestimated by the same amount.

TGA plots of neat PS and PET as well as physical, unprocessed mixtures of intermediate compositions are shown in Fig. 1. PET has more cohesive energy density and a more complicated thermal degradation mechanism, so its degradation kinetics are delayed and slower. The boxed region represents the area where analysis of the residual char weight percentage accurately confirms the composition of PS/PET blends (see Fig. 2). The specific residual char amount within this box was determined at the temperature at which the first derivative of the weight loss curve became zero, i.e., the weight loss curve flattened out. This method provided excellent consistency in the determination of the residual amount of char. Blends processed by extrusion and CMA were also subjected to similar TGA runs, and it was found that the final residual char weight did not change significantly relative to unprocessed blends of similar compositions.

In addition to obtaining composition, TGA analysis also provided a comparative investigation into the effect of milling on thermal stability. In Table 1, it is evident that, from the degradation rate, milling prolongs the thermal degradation of PS/PET blends relative to extrusion. The nonisothermal degradation rate is defined as the inverse of the half-time, 1/[t.sub.1/2], i.e., the inverse of the time it takes to lose 50% of the weight of the sample. The time is defined such that time zero is at the temperature of degradation onset [T.sub.o] and is related inversely to the constant heating rate [phi] (see Fig. 1). The milling-induced interactions affect the degradation kinetics, i.e., extra thermal energy is needed to disentangle amorphous PS and PET chains relative to those found in extruded blends. The extrusion of PS with PET occurred at a temperature above the cold crystallization temperature ([T.sub.c]) of PET, resulting in a relative abundance of PET crystallites in melt-extruded samples when compared with CMA blends. The physical significance of this processing difference is that more and weaker amorphous PS/crystalline PET interphases developed within the extruded blends. Since interphase strength is directly proportional to interphase adhesion, it is reasonable to assume that the lower thermal degradation rates of the milled blends relative to the extruded blends are because of better adhesion between PS and PET. This increased adhesion must be due to increased amorphous PET content in the milled blends relative to the extruded blends that subsequently allows for more and stronger amorphous PET/amorphous PS interphases.

[FIGURE 2 OMITTED]

Blend Compatibility Estimation by MDSC

As mentioned previously, the extent of compatibility for a polymer blend is proportional to [delta][DELTA][C.sub.p]--the decrement of the change in heat capacity of PET at its [T.sub.g]. This change in heat capacity at the glass transition can be calculated as the difference between the heat capacities before and after the transition region. Both the reversible heat capacity and its corresponding time derivative can be obtained from MDSC. Thus, evaluating the integral of the derivative heat capacity peak from the onset to the end of the glass transition region gives [DELTA][C.sub.p]. Accurate integration was required because the heat capacity does not change suddenly in the glass transition region and its differential is very sensitive to the glass transition process [14].

Figure 3 illustrates how compatibility is estimated from a differential heat capacity curve in MDSC. The time-derivative reversible heat capacity curve for each sample is obtained by taking the derivative of its corresponding reversible heat capacity with respect to time. Since temperature is directly proportional to time at a constant heating rate, the integration of the peak at the chosen onset and final glass transition temperatures are equivalent to the change in heat capacity at the peak temperature [T.sub.g]. The milled 30/70 PS/PET blend shows a decrease in heat capacity at both component [T.sub.g]s. The reduction in heat capacity change means that bulk PS and PET are being transferred to the interphase. The reason that PET was chosen over PS for this quantitative compatibility determination is because PS is easily trapped due to its side-chain aromatic ring and is less mobile than PET. The PET in these studies has a molecular weight that is about an order of magnitude lower than PS and can subsequently self-diffuse to the PS/PET interphase more efficiently than PS. The values of both bulk unprocessed polymer [DELTA][C.sub.p] agree well with the data of Boller et al., who had extensively studied the kinetics of the glass transition for amorphous PS and PET with MDSC [15].

[FIGURE 3 OMITTED]

Since it is a bulk property, the heat capacity directly correlates to the amount of amorphous PET being lost from the bulk because of interphase entanglements with PS. Free volume restrictions caused by the interphase entanglements subsequently reduce the heat capacity change at [T.sub.g]. Effects such as crystallinity changes during processing are not explicitly included in Eq. 1. The effect of increased crystallinity upon volume relaxation in the glass transition zone for pure PET have been shown to decrease [DELTA][C.sub.p], i.e., any destruction of crystallinity would result in increasing [DELTA][C.sub.p] [16]. For these investigations, however, Eq. 1 will suffice because milled blends are being compared with extruded blends on a relative basis, and any decrement in [DELTA][C.sub.p] caused by CMA or extrusion is solely assumed to be due to the removal of amorphous PET from the bulk. The formation of copolymers at the interphase because of milling-induced mechanoradical formation could also enhance compatibility. However, previous FTIR investigations revealed no changes in peak positions for either PS or PET in milled blends relative to the pure components--a sign that the chemical changes are not occurring during milling [1]. Similarly, gel permeation chromatography results of semicrystalline polyethylene oxide (PEO) and amorphous polyvinylpyrrolidone (PVP) do not show significant molecular weight degradation during cryogenic milling at similar milling times [17]. Neither PS nor PET could form stable mechanoradicals at cryogenic temperatures due to significant structure-related energy barriers. Thus, any compatibilization of PS/PET blends in this work is therefore only due to physical entanglements of amorphous PS and PET at their interphase.

PET Glass Transition Effects

The amorphous phase within PET is certainly affected by the blend technique because the position of PET [T.sub.g] in the blends changes relative to that of pure neat PET (~72[degrees]C). The glass transition region and its position are functions of chain mobility and the various local chain environments. [T.sub.g] changes in polymer blends that are not due to partial or complete miscibility have been attributed to morphological (domain size) and physical interactions. Decreases in [T.sub.g] values are due to a higher percentage of chains with unrestricted motion due to more free volume available for cooperatively rearranging movements, thus reducing the thermal activation required for the glass transition [18]. Interphase interactions, the creation of an immobilized interphase layer, and the presence of a rigid, glassy matrix causing a "wall" effect are reasons for increased [T.sub.g]. In particular, increases in PET [T.sub.g] in PET/PC blends are caused by friction at the "wall"-like interphase between the two phases, and the decreasing PS [T.sub.g] with increased PS content in PS/PP blends is due to a diluency as more PS chain ends are less frequently confined and restricted [19].

[FIGURE 4 OMITTED]

Figure 4 illustrates the effects of processing on the compatibility of PS/PET blends based upon the amorphous phase of PET. The dotted line represents pure neat PET [T.sub.g]. All CMA blends show a [T.sub.g] statistically greater than 72[degrees]C, while all extruded blends show [T.sub.g] less than 72[degrees]C. PS [T.sub.g] did not vary significantly with composition for either CMA or extruded blends. This is not surprising because PS is already highly amorphous and has stiff chains that are not nearly as sensitive to their environments as PET chains. All milled and extruded blends possessed both PET and PS [T.sub.g]s and were thus thermodynamically immiscible, yet the milled blends were characterized by smaller differences between the two [T.sub.g]s ([DELTA][T.sub.g] = [T.sub.g] PS - [T.sub.g] PET) relative to that of extruded blends. The [DELTA][T.sub.g] for the milled blends decreased with increasing PET content--evidence that changes in blend morphology and subsequent increases in physical entanglements because of increased interphase contact with aPS are the reasons for these glass transition shifts rather than partial miscibility of the two polymers. A similar shrinkage of [DELTA][T.sub.g] was observed in both melt-mixed and solution-cast PS/PMMA blends relative to pure PS and PMMA and was attributed to the presence of the PS/PMMA amorphous interphase [20]. The [T.sub.g] for milled PET does not vary significantly relative to the [T.sub.g] of neat PET, so any changes in PET [T.sub.g] during CMA are due to forced restricted conformations of amorphous PET chains. The physical entanglement of amorphous PET chains with aPS at their interphase typifies such a restricted chain environment and thus elevates PET [T.sub.g]. Though the specific mechanism of chain entanglement in CMA of polymers is unknown, the aforementioned compatibilization results support the formation of a compatibilized interphase.

The decrease of PET [T.sub.g] in extruded PS/PET blends relative to unprocessed PET are caused by more chains having unrestricted motion in the melt. This chain extension thereby decreases entropy, so the lowered [T.sub.g] during MDSC is because of the stored entropy decrement in the glassy state [21]. The elevation of PET [T.sub.g] in milled blends is because of the presence of both milling-induced interphase interactions and the mere presence of highly dispersed PS domains. Thirtha et al. attributed similar increases in PET [T.sub.g] in PET/PC blends to friction at the "wall"-like interphase between the two phases and the effect of highly dispersed glass beads on the matrix polymer [22]. This effect is especially true for the PET-rich blends characterized by a matrix of PET with highly dispersed aPS domains. As the amount of PS in the CMA blends increases, the amount of constraint and restriction of the PET cooperative motion decreases as the interphase area decreases and phase inversion occurs.

Blend Morphology and Mixing

The TEM image of a 50/50 PS/PET processed by CMA in Fig. 5 illustrates the highly interphase nature of a blend near the phase inversion. Amorphous PET chains are confined inside PS-rich microdomains with interphase thicknesses ranging from 3 to 5 nm. These thicknesses were confirmed visually and by recording the EDS oxygen peak height for PET. EDS spectra were taken at several points in each phase, traversing the interphase between the two polymers, and the oxygen peak height diminished upon crossing the interphase from PET to PS. The results summarized in Table 2, in addition to the corresponding phase markings in Fig. 5, confirm that the dark phase is PS and the lighter phase is PET. Confinement of PET within such microdomains of PS is one of the reasons for the elevated [T.sub.g] and is attributed to the ability of amorphous PET chains to access more free volume and entangle with more PS chains at the interphase. These regions contain the most amount of interphase relative to bulk and effectively "trap" amorphous PET chains that either originally existed in the virgin material or were expelled from crystalline PET domains during CMA. In order for a large quantity of these amorphous PET chains to access these restricted environments, there must be sufficient self-diffusion from the PET bulk phase. The corresponding compositional trend for glass transition region width of CMA blends is similarly related to the milling-induced physical interactions that result in the increased number of relaxations that are responsible for the spectrum of glass transitions [13, 14].

[FIGURE 5 OMITTED]

The trends shown for the PET glass transition region are reflected in the data for compatibility. Figure 4 also illustrates the compatibilization efficiency of CMA. The compatibilization extent is higher at all concentrations for CMA than for extrusion, and the changes in compatibility with morphology (composition) are more drastic for CMA blends. Compatibilization that results from extrusion is because of random entanglement of amorphous chains present in each phase in the melt assisted by the lowered melt viscosities of polymers [23]. The increase in compatibilization with increased PET content for CMA blends is due to a coupling of two phenomena: the increased interphase area as the PS domains get smaller in the PET-rich blends and the higher mobility of the lower [M.sub.w] and less sterically hindered PET chains which enables them to access the interphase readily. The x-axis error bars show the compositional standard deviation from sample to sample. It is evident from the data that CMA mixes PS and PET more efficiently than extrusion and is subsequently characterized by more sample-to-sample compositional homogeneity than extrusion.

The extruded blends are also increasingly PET-rich, i.e., it is more difficult to keep the actual composition of the blend close to its target composition. The density difference between PET and PS (1.39 g/[cm.sup.3] vs. 1.05 g/[cm.sup.3] respectively) causes a maldistribution in the packing of the polymers in the metering device prior to the entrance to the extruder screws. Both polymers are also in pellet form prior to extrusion, so the coupling of the poor pellet packing with the unevenly distributed pellets due to the density difference reduces the compositional consistency of the extruder. As the concentration of PET in the blends is higher, more PS gets trapped prior to the extrusion in the metering device and does not partake in the extrusion process. The drastic composition fluctuations for extruded blends are not evidenced for CMA blends, thereby illustrating the inherent compatibilization advantage of CMA in its more precise compositional control.

Milling Time Effect on Blend Compatibility

Another CMA processing parameter that was hypothesized to affect compatibilization of PS and PET is milling time. The longer the milling time, the more mechanical energy from impacts is transferred to the polymers. All blends to this point in this study were milled for 30 min. Since the 30/70 PS/PET blend was characterized as having the highest propensity for compatibilization with respect to blend composition, it is also assumed to gain more compatibility from increased milling time. The level of interphase in this blend was thought to be well below its saturation value by amorphous PET entanglements at low milling times. This 30/70 blend was therefore also milled for 300 min.

Table 3 compares the near-30/70 CMA blends at both 30 and 300 min milling time to the extruded sample nearest to that composition as well as literature results for various other systems that include either PS or PET. As expected, the increased milling time led to a corresponding increase in compatibility of PS with PET as more interphase entanglements were physically forced due to the increased input energy of the mill. The extent of compatibility for the PS/PMMA system is based on the PMMA [T.sub.g] x [DELTA][C.sub.p]. As expected, increasing milling time increased the compatibility for PS/PET. In the literature, a PS/PMMA system was compatibilized with PS-b-PMMA diblock copolymer at 3 wt% and 30 min of mixing time (defined as the time when the materials were loaded into the extruder) [6]. Similarly, a PP/PET system was compatibilized with PP grafted with maleic anhydride [12]. The average domain size ranged from 16 [micro]m for the uncompatibilized sample to 2 [micro]m for the blend with the highest compatibilization of 48.7% [24]. The values for the percent compatibility for these three references were calculated from composition and [DELTA][C.sub.p] data via Eq. 1 in precisely the same manner as the data in Fig. 4.

Despite the lack of chemical compatibilizers, the CMA blends show a significant amount of compatibilization based on physical entanglements alone. It is also interesting to note that the composition region of maximum compatibility determined in this work (near-70 wt% PET) is roughly the same as that found in different blend systems [3, 25].

SUMMARY AND CONCLUSIONS

The compatibilization of PET with PS has been quantified in terms of the reduction of the change in heat capacity of PET at its [T.sub.g]. This calculation for percentage compatibilization involves the accurate determination of the blend composition via with the percentage weight loss remaining after nonisothermal degradation. CMA was shown to efficiently compatibilize the PS/PET blend--almost to the same extent of chemical compatibilization in the melt with grafted copolymers. Other benefits of CMA over extrusion include precise compositional control and better mixing. A strong compositional dependence of compatibility on CMA blend morphology mirrors a similar trend for PET [T.sub.g] elevation--more amorphous PET becomes entangled with PS at the interphase at higher PET content and results in a higher percentage of amorphous PET chains being confined locally at the interphase. These forced physical entanglements both increase the amount of interphase in the blend and improve the thermal stability. The interdependence of composition and compatibility that characterizes CMA is surmised to be caused by a decrease of PET crystallinity during milling--a phenomenon that will be discussed in the next part.

ACKNOWLEDGMENTS

The authors thank Dr. Hongxia Zhou for her help with thermal analysis data acquisition.

REFERENCES

1. R.J. Schexnaydre and B.S. Mitchell, Mater. Lett., 61, 2151 (2007).

2. L.A Utracki, Can. J. Chem. Eng., 80, 1008 (2002).

3. L.W. Jang, K.H. Lee. D.S. Lee, J.S. Yoon, I.J. Chin, H.J. Choi, and K.H. Lee, J. Appl. Polym. Sci., 78, 1998 (2000).

4. S. Jonna and J. Lyons, Polym. Test., 24, 428 (2005).

5. J.S. Lee, K.Y. Park, D.J. Yoo, and K.D. Suh, J. Polym. Sci. Part B: Polym Phys., 38, 1396 (2000).

6. C. Chuai, K. Almdal, and L.J. Jorgen. J. Appl. Polym. Sci., 91, 609 (2004).

7. M.Y. Ju and F.C. Chang, Polymer, 41, 1719 (2000).

8. S. Tankhiwale, M.C. Gupta, and S.G. Viswanath, J. Polym. Plast. Technol. Eng., 41(1), 172 (2002).

9. M. Wilczek, D. Hintemann, J. Bertling, and R. Kummel, "PARTEC 2001 International Congress for Particle Technology," in Conference Proceedings, 1 (2001).

10. D.B. Witkin and E.J. Lavernia, Prog. Mater. Sci., 51, 2 (2006).

11. C. Suryanarayana, Prog. Mater. Sci., 46, 4 (2001).

12. Y.X. Pang, D.M. Jia, H.J. Hu, D.J. Hourston, and M. Song, J. Appl. Polym. Sci., 74, 2868 (1999).

13. S.N. Cassu and M. Felisberti, Polymer, 38(15), 3908 (1997).

14. D.J. Hoursion, M. Song, A. Hammiche, H.M. Pollock, and M. Reading, Polymer, 38(1), 1 (1997).

15. A. Boller, I. Okazaki, and B. Wunderlich, Thermochim. Acta, 284, 8 (1996).

16. J. Hadac, P. Slobodian, and P. Saha, J. Mater. Sci., 42, 3724 (2007).

17. R.J. Schexnaydre and B.S. Mitchell, unpublished results (2004).

18. V. Thirtha, R. Lehman, and T. Nosker, Polymer, 47, 5392 (2006).

19. V. Reinsch and L. Rebenfeld, J. Appl. Polym. Sci., 59, 1913 (1996).

20. G.G. Silva, P.M. de Freitas Rocha, P.S. de Oliveira, and B.R.A. Neves. Appl. Surf. Sci., 238, 67 (2004).

21. L.C. Lee and B.G. Min, Polymer, 40, 5445 (1999).

22. V. Thirtha, R. Lehman, and T. Nosker, Polym. Eng. Sci., 45, 1187 (2005).

23. Y. Liu and J. Donovan, Polymer, 36(25), 4797 (1995).

24. Y.X. Pang, D.M. Jia, H.J. Hu, D.J. Hourston, and M. Song, Polymer, 41, 360 (2000).

25. D.H. Kim, K.Y. Park, J.Y. Kim. and K.D. Suh, J. Appl. Polym. Sci., 78, 1022 (2000).

Ryan J. Schexnaydre, Brian S. Mitchell

Department of Chemical and Biomolecular Engineering, Tulane Institute for Macromolecular Engineering and Science, Tulane University, New Orleans, Louisiana 70118

Correspondence to: Brian S. Mitchell; e-mail: brian@tulane.edu

Contract grant sponsor: NASA; contract grant number: NNC06AA18A.
TABLE 1. Nonisothermal degradation rate data from TGA for neat PET and
blends processed by extrusion and CMA.

Processing Weight Degradation rate
type percent PET (1/[t.sub.1/2]) ([min.sup.-1])

None 0 0.09
None 100 0.08
CMA 71 0.05
Extrusion 68 0.07
CMA 49 0.06
Extrusion 47 0.08
CMA 28 0.07
Extrusion 26 0.08

Composition is denoted by weight percent PET, and the overall rate of
degradation is taken as the inverse of the sample half-time.

TABLE 2. Normalized EDS oxygen peak height for positions A, B, and C in
Figure 5.

Spot Normalized oxygen peak height

A 1
B 0.67
C 0

A value of 1 signifies pure PET content.

TABLE 3. A quantitative comparison of CMA with various extrusion
processes in terms of the extent of compatibilization, which is
calculated by Equation 1.

 Weight Mixing or
Method of percent PET milling Extent of
compatibilization (or non-PS) time (min) compatibilization

Extrusion 74 15 6.8
Block copolymer extrusion 70 30 20
 [6]
CMA 71 30 21.9
CMA 70 300 32.7
Grafted copolymer extrusion 75 20 48.7
 [9, 20]
COPYRIGHT 2008 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2008 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Schexnaydre, Ryan J.; Mitchell, Brian S.
Publication:Polymer Engineering and Science
Article Type:Technical report
Geographic Code:1USA
Date:Apr 1, 2008
Words:5097
Previous Article:Oxygen permeability of biaxially oriented polypropylene films.
Next Article:Rheological characterization of room temperature vulcanized silicone sealant: effect of filler particle size.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters