Volume 4 Issue 4 - May 9, 2008
Melting Behavior and Polymorphic Forms of Poly(hexamethylene terephthalate) in Exfoliated Silicate Interlayers
Arup K. Ghosh 1, Kai-Cheng Yen 2 and Eamor M. Woo 2,*

1 GE India Technology Center Pvt. Ltd., Bangalore - 560066, INDIA.
2 Department of Chemical Engineering, National Cheng Kung University, Tainan, 701-01, Taiwan
E-mail: emwoo@mail.ncku.edu.tw

Font Normal   Font Enlarge
Polymer-layered silicate nanocomposites (PLSN) have shown an explosive growth from the perspectives of both fundamental research interests and industrial applications [1,2]. Incorporation of organo-clays in semicrystalline polymers often leads to alteration of the crystalline morphology through transformation in their polymorphic crystal forms.

The complex, yet interesting, multiple melting behavior of poly(hexamethylene terephthalate) (PHT) has been analyzed in terms of two polymorphic crystalline forms α and β[3]. The origin of multiple endotherms has been mostly explained by two mechanistic approaches dual/multiple modification and reorganization. The α cell in PHT is preferentially formed by stress-induced crystallization with a monoclinic chain packing, whereas β forms is characterized by a triclinic cell and usually found on high temperatures annealing (~140°C) of PHT [4].  However, a mixture of α and β cells is always formed at moderate temperature of crystallization (<140°C).  PHT in a neat bulk form exhibits complex melting peaks and as many as five peaks were evident when melt crystallized at 90ºC. Incorporation of silicate layers (in either intercalation or exfoliation layers) in PHT to form a nanocomposite might induce changes in crystal forms, and this information can be used as useful evidence for more in-depth interpretation on the complex polymorphism and melting behavior in PHT.  An earlier paper has shed some light on the origin of melting peaks P1 and P3 [3] in the neat PHT. However, correlations between the polymorphic morphology of two unit-cell crystals and melting peaks in PHT have yet to be further understood. In extension of our previously published work on effects of intercalation of silicates (organo-clays) incorporation in semicrystalline PHT [5], this short article attempted to address effects of exfoliation of silicate interlayers on polymorphism behavior in PHT.

Dispersion of C30B in polymer

Figure 1 shows the TEM images of PHT/C30B hybrids having different content of C30B (graphs-a, b, c) where the dark lines represent silicate layers. It is evident from the micrograph that the silicate layers are almost completely delaminated in the PHT matrix containing 1-wt% of C30B to produce a highly intercalated/exfoliated structure (graph-a). Intercalated clay tactoids with multilayer thickness of about 35 nm are also apparent. A mixed morphology of intercalation/exfoliation is observed in PHT/C30B (5wt%) nanocomposites (graph-b). It shows more ordered distribution of silicate layers. However, in PHT/C30B (10wt%) nanocomposites (graph-c) an intercalated morphology is observed predominantly. Some unintercalated clay tactoids are also apparent as dark spots. Thus, the pictorial evidences as obtained from TEM suggests the formation of PHT-C30B nanocomposites for all composition used in this study.
Fig. 1.  TEM images of PHT/C30B composites: (a) C30B (1%), (b) C30B (5%), and (c) C30B (10%).

Melting characteristics and crystal forms in PHT/clay composites

WAXD analysis was performed to identify the crystal forms of the melt-crystallized PHT/ C30B nanocomposites.
Fig. 2.  WAXD patterns of PHT- C30B nanocomposites melt crystallized at 130ºC for 2h.
Figure 2 shows the WAXD patterns for neat PHT and PHT/C30B nanocomposites isothermally melt-crystallized at 130ºC for 2h.  In all the WAXD traces characteristic peaks of α form (2θ= 15.9, 20.5, 21.2 and 25.4º) are apparent in conjunction with β crystals (2θ= 7.2, 17.9 and 23.6º) indicating the coexistence of both types of unit cells.  The relative intensities of two most significant peaks at 2θ= 21.2º (α) and 23.6º (β) could be considered for a qualitative estimation of α-to-β phase formation.  It is apparent from the diffractogram of neat PHT (Trace-a) that α crystal is predominantly formed over the β cells.  On the other hand, the PHT/C30B nanocomposites exhibit different trend in XRD patterns (Traces b-f). It should be noted that, when comparing the nanocomposites the peak due to α crystals (2θ= 21.2º) appears sharper as the C30B content is increased from 1-wt% to 3-wt% (comparing Trace-b with Trace-c) implying higher α/β fraction.  However, in samples containing more than 3-wt% of C30B the peak assigned to α forms (2θ= 21.2º) is weakened as compared that of the β forms (2θ= 23.6º), which implies reduced yield of α phase with higher loading of C30B.  The β phase (2θ= 23.6º) is the most significant in PHT/C30B (5wt%) nanocomposites.  On further increase in C30B loading (to 7 and 10 wt%, respectively) the diffraction peaks for both α and β forms are weakened (2θ= 21.2º (α) and 23.6º (β)) probably due to the lowering of crystallinity in PHT matrix.  However, β crystals are yielded predominantly with the increase in C30B content.  In other words, the presence of silicate layers favors the transformation of α crystals to the β forms at a lower temperature of 130ºC.

In our previous study, the endothermic peak P2 is reported to be associated with the β crystals [5].  This can be substantiated further in the present study.  It is already shown above that formation of β form is favored in the presence of layered silicate as compared to the neat PHT and at higher Tc (>130ºC) this crystalline form predominates over the α crystals. The DSC analysis was also done on the PHT/C30B nano-composites. In the DSC trace of the nanocomposite melt crystallized at 130ºC, P2 is distinctly visible as a separate endotherm. The increase in peak intensity of P2 in the nanocomposites with increase in Tc (120º-130ºC) indicates greater population of β cells in PHT matrix as expected.  Thus, association of melting endotherm P2 with the β crystalline form of PHT can be justified from respective WAXD pattern and the thermal analysis obtained form DSC (not shown for brevity).

Furthermore, WAXD and DSC analysis were done on the PHT/C30B (5%) nanocomposites to identify the correlation of the P3 with α-crystals. Figure 3 shows the WAXD patterns of the PHT/C30B (5wt%) nanocomposites subjected to different thermal treatments.  The corresponding DSC thermograms are also shown inset.  The WAXD patterns of the sample melt crystallized at 130ºC (for 2h) (Trace-a) and 140ºC (for 8h) (Trace-d) are completely different as discussed earlier.  While at 130ºC the peaks of α form are seen in conjunction with β crystals, but at 140ºC only the peaks of β cells are apparent.  At 140ºC the DSC thermogram exhibits only one significant peak at ~ 147.8ºC indicating a single type of crystalline lamellae. The sample melt crystallized at 130ºC for 2h and then dynamically heated to 140ºC at a scan rate of 10ºC/min and subsequently annealed for 2h (Trace-b) to melt the crystalline lamellae associated with endotherm P1, exhibit WAXD pattern corresponding to mixed α and β forms similar to the sample crystallized at 130ºC.  DSC thermogram of the annealed sample looks very similar to that of the sample crystallized at 140ºC, but the peak at lower temperature (~143.3ºC) is more significant. The presence of α cells further implies that the endotherm P3 is most likely to be associated with the former as both P2 and P4 are ascribed to the β forms. Further increase in the annealing temperature to 143ºC to melt the crystals associated with P2 endotherm (Trace-c) resulted in only the β form as evident in WAXD pattern.  The DSC thermogram exhibits a very broad peak, which is actually a combination of two overlapped peaks (at 146.7ºC and 148.7ºC) revealed from the deconvolution of the same (inset).  It seems that even after annealing at 143ºC, two different types of crystalline lamellae associated with P3 and P4 endotherms exist separately, but only melting peak of the β type crystal is observed.
Fig. 4.   DSC thermograms of PHT/ C30B (5%) nanocomposites (a) melt crystallized at 130ºC for 2h, (b) melt crystallized at 130ºC for 2h and annealed at 140ºC for 2h, (c) melt crystallized at 130ºC for 2h and annealed at 143ºC for 2h and (d) melt crystallized at 140ºC for 8h (scan rate = 2.5ºC/min).
Fig. 3.   WAXD patterns of PHT/C30B (5%) nanocomposites (a) melt crystallized at 130ºC for 2h, (b) melt crystallized at 130ºC for 2h and annealed at 140ºC for 2h, (c) melt crystallized at 130ºC for 2h and annealed at 143ºC for 2h and (d) melt crystallized at 140ºC for 8h.  (Corresponding DSC thermograms (scan rate = 10ºC/min) shown inset).

DSC scanning of the samples at a slower heating rate (2.5ºC/min) was performed to further clarify the endothermic peaks (higher resolution).  Figure 4 shows that the melting peak P3 is prominent (at 143.6ºC) in the sample melt crystallized at 130ºC (Trace-a).  The thermogram of the sample melt crystallized at 140ºC (Trace-d) is almost similar to that obtained from normal scanning rate (10ºC/min) (inset, Trace-3d).  The high intensity of P3 in the sample annealed at 140ºC can further be attributed to the large population of α form.  However, on annealing at 140ºC, P3 is shifted to the higher temperature of 146ºC due to the formation of thicker and more perfect crystalline lamella of α cells via reorganization of the initial thinner lamella (Trace-b).  Similarly, P2 and P4 are also shifted to 141.8ºC and 146.9ºC from ~139-140ºC and 145ºC respectively, by annealing.  Sample annealed at 143ºC exhibits two distinct endothermic peaks at ~145ºC (Px) and 147.5ºC (Py) (Trace-c).  The higher endotherm (Py) may arise due to the shifting of P4 on annealing, but the endotherm at lower temperature (Px) is presumably a new one and not to correspond P3 as lowering of the peak temperature (from 146ºC) is unlikely with increase in annealing temperature.  However, this new endothermic peak (Px) is possibly formed due to yield of more perfect β-type lamellae by melting and recrystallization of the crystals associated with P2 on annealing at 143ºC.  Nevertheless, P3 seems to merge with P4 to give rise (Py) and α crystalline form is transformed to thermodynamically more stable β forms by annealing at high temperature, especially in presence of layered silicates, which facilitates β cell formation.


  • Alexandre M, Dubois P. 2000. Polymer-layered silicate nanocomposites: preparation, properties and uses of a new class of materials. Mater Sci Eng 28(1-2): 1-63.

  • Ray SS, Okamoto M. 2003. Polymer/layered silicate nanocomposites: a review from preparation to processing. Prog Polym Sci 28(11): 1539-1641.

  • Woo EM, Wu PL, Chiang CP, Liu HL. 2004. Analysis of Polymorphism and Dual Crystalline Morphologies in Poly(hexamethylene terephthalate). Macromol Rapid Commun 25(9): 942-948.

  • Hall IH, Ibrahim BA. 1982. Structure and properties of poly(hexamethylene terephthalate)-1. The preparation, morphology and unit cells of three allomorphs. Polymer 23(6):805-816.

  • Ghosh AK, Woo EM. 2004. Effect of layered silicates on the confined crystalline morphology of Poly(hexamethylene terephthalate). J Mater Chem 14(20): 3034-3042.
< previousnext >
Copyright National Cheng Kung University