Effects of thermal history in the ring opening polymerization of CBT and its mixtures with montmorillonite on the crystallization of the resulting poly(butylene terephthalate)

Abstract

Differential scanning calorimetry was used to study the thermal characteristics and morphological structure of species produced during the ring opening polymerization of cyclic butylene terephthalate (CBT). Thermal programs consisting of a first ramp heating scan and an isothermal step, followed by cooling and a second ramp heating step, were used to study the effects of thermal history, catalyst (butyl chlorotin dihydroxide) at concentrations between 0.1 and 1.3% (w/w), and the presence of a layered silicate nanofiller (montmorillonite at 4.0%, w/w) on the structure of the resulting polymer (poly(butylene terephthalate), pCBT). Wide angle X-ray diffraction was used to monitor the degree of exfoliation of the nanocomposites.

It was found that pCBT is formed in the amorphous state, and crystallizes during the heating step or during the isothermal step at temperatures lower than the equilibrium melting temperature of the polymer ($T_{\text{m}}^0$). When premixed with the nanofiller, irrespective of whether this was previously intercalated with a tallow surfactant or used in its pristine form, polymerization took place at higher temperatures and most of the crystallization was found to occur during the cooling stage. In those cases where crystallization took place during either the first heating scan, or during a prolonged isothermal step below the $T_{\text{m}}^0$ of the polymer, the resulting crystals were found to have a higher lamellar thickness, as compared with the same polymer crystallized from the melt during the cooling step from temperatures above the polymer $T_{\text{m}}^0$.

Introduction

Studies on polymer/layered silicate (PLS) nanocomposites have attracted, in recent years, increasing interest in both academic institutions and industry, owing to the remarkable improvements in properties that can be achieved relative to the pristine polymer or conventional particulate or short-fiber composites. At equal filler content, PLS nanocomposites have a higher modulus and strength, together with a higher thermal oxidative stability, hence a lower flammability and gas permeability than the corresponding microcomposites [1].

To enhance the “compatibility” between the nanofiller and the polymer, thereby improving the extent of exfoliation and the related nanofiller dispersion within the polymer, it is often necessary to modify the chemical structure of the surface layers of natural occurring silicate fillers.

For the case of thermoplastic matrix nanocomposites the intercalation of the nanofiller is usually carried out by conventional melt-compounding methods, which produce intensive mixing and often require high temperatures. Melt intercalation of the layered silicate carried out at high temperatures can induce considerable degradation of the organic modifier through chemical decomposition reactions, which may result in extensive deterioration of the interfacial interactions between the filler and the polymer [1], [2], [3]. There are a few cases, however, where polymerization can be carried out during compounding, at substantially lower temperatures or with shorter exposures at higher temperatures, than the corresponding melt-compounding techniques.

There has been considerable interest also on the in situ polymerization of cyclic butylene terephthalate (CBT), which is a low viscosity cyclic oligoester, to produce poly(butylene terephthalate) (pCBT) matrix composites capable of being processed at relatively low temperatures [4], [5], [6], [7]. These composites, therefore, combine the typical advantages of thermoplastic polymers, particularly toughness, with the low viscosities of thermoset resins which can provide thorough impregnation of fibers, thereby making it possible to produce composites with high fiber contents [8], [9], [10], [11].

Another advantage of pCBT matrix composites is the expected higher toughness, relative to the corresponding PBT-based systems, in so far as the polymerization of CBT does not produce low molecular weight by-products [12].

Not surprisingly CBT has also been used as a precursor for the production of pCBT-based nanocomposite [13], [14]. In this case, the low melting temperature of the cyclic oligomer and the very low viscosity of the resulting melt can be exploited to achieve an efficient exfoliation of the nanofiller and dispersion in the polymer matrix, and reduce the degradation susceptibility of any organic surface modifier present.

Despite the great attention paid to CBT and the resulting pCBT, only few of the works reported in literature deal with the characterization of the ring opening polymerization of CBT with a thermal method of analysis [6], [11], [15]. In the majority of cases the authors have reported results on one-component commercial CBT, usually containing an unknown amount of catalyst. Other workers have reported data for a widely different range of temperatures. For instance, Mohd Ishak et al. [15] have reported data on the TMDSC analysis of CBT polymerization at temperatures that are respectively much higher (260 °C) and much lower (200 °C) than the melting temperature of PBT. No study has been reported on the effect of polymerization temperature within a closer range of the melting temperature of PBT. Furthermore, only one study has been reported in the literature on the thermal characterization of nanocomposites derived from CBT [14]. Even so, the analysis was performed on samples heated to very high temperatures (260 °C), which is not realistic for CBT nanocomposites, due to susceptibility of the modifier to thermally degrade at such temperatures.

The aim of this work is to study the effect of different conditions, particularly catalyst content, polymerization temperature and surface modifier used for the exfoliation of the nanofiller on the structure of the resulting pCBT polymer.

To this end DSC analysis was used to analyze the influence of the different variables on the melting and crystallization of the pCBT obtained. It is worth noting that DSC is particularly suitable for these studies as the ring opening polymerization of CBT is virtually athermal [7] and, therefore, the crystallization and melting transitions can be easily followed without significant interference from the associated heat of conversion.