Effects of the carboxyl concentration on the solid‐state polymerization of poly(ethylene terephthalate)

There are two types of polycondensation reactions in the solid‐state polymerization (SSP) of poly(ethylene terephthalate) (PET), namely, transesterification and esterification. Transesterification is the reaction between two hydroxyl ends with ethylene glycol as the byproduct, and esterification is the reaction between a carboxyl end and a hydroxyl end with water as the byproduct. The SSP of powdered PET in a fluid bed is practically a reaction‐controlled process because of negligible or very small diffusion resistance. It can be proved mathematically that an optimal carboxyl concentration for reaction‐controlled SSP exists only if k₂/k₁ > 2, where k₂ and k₁ are the forward reaction rate constants of esterification and transesterification, respectively. Several interesting observations were made in fluid‐bed SSP experiments of powdered PET: (1) the SSP rate increases monotonously with decreasing carboxyl concentration, (2) k₂ < k₁ in the presence of sufficient catalyst, (3) k₁ decreases with increasing carboxyl concentration if the catalyst concentration is insufficient, and (4) the minimum catalyst concentration required to achieve the highest SSP rate decreases with decreasing carboxyl concentration. In the SSP of pelletized PET, both reaction and diffusion are important, and there exists an optimal carboxyl concentration for the fastest SSP rate because esterification, which generates the faster diffusing byproduct, is retarded less than transesterification in the presence of substantial diffusion resistance. The optimal prepolymer carboxyl concentration, which ranges from 25 to 40% of the total end‐group concentration in most commercial SSP processes, increases with increasing pellet size and product molecular weight. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 83: 1288–1304, 2002     

Solid‐state polymerization of poly(trimethylene terephthalate)

The solid‐state polymerization (SSP) of poly(trimethylene terephthalate) (PTT) has been studied and compared with that of poly(ethylene terephthalate) (PET). Because PTT and PET share the same SSP mechanism, the modified second‐order kinetic model, which has successfully been used to describe the SSP behaviors of PET, also fits the SSP data of PTT prepolymers with intrinsic viscosities (IVs) ranging from 0.445 to 0.660 dL/g. According to this model, the overall SSP rate is ?dC/dt = 2k_a(C ? C_ai)², where C is the total end group concentration, t is the SSP time, k_a is the apparent reaction rate constant, and C_ai is the apparent inactive end group concentration. With this equation, the effects of all factors that influence the SSP rate are implicitly and conveniently incorporated into two parameters, k_a and C_ai. k_a increases, whereas C_ai decreases, with increasing SSP temperature, increasing prepolymer IV, and decreasing pellet size, just as for the SSP of PET. Therefore, the SSP rate increases with increasing prepolymer IV and increasing SSP temperature. The apparent activation energy is about 26 kcal/mol, and the average SSP rate about doubles with each 10°C increase in temperature within the temperature range of 200–225°C. The SSP rate increases by about 30% when the pellet size is decreased from 0.025 to 0.015 g/pellet. Compared with PET, PTT has a much lower sticking tendency and a much higher SSP rate (more than twice as high). Therefore, the SSP process for PTT can be made much simpler and more efficient than that for PET. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 89: 3188–3200, 2003     

Effects of crystallinity on solid‐state polymerization of poly(ethylene terephthalate)

There has been a widely held assumption that the solid‐state polymerization (SSP) rate of poly(ethylene terephthalate) (PET) decreases with increasing crystallinity. Several published articles that purported to prove this assumption were based on faulty experiments. Therefore, a proper experimental procedure has been used to study the true effects of crystallinity on the SSP of PET. The results show that, for PET in pellet and powder forms, the SSP rate increases with increasing crystallinity. This is because an increase in the crystallinity results in increased end‐group concentration in the amorphous phase, where SSP reactions take place, and decreased concentrations of inactive end groups trapped inside the crystals, thereby increasing the rates of end‐group collision and reactions. These positive effects outweigh the negative effect of the increased byproduct‐diffusion resistance because of the increase in crystallinity. As the particle size of PET is increased beyond a critical value of about 7 mm, the SSP rate actually decreases with increasing crystallinity because of the excessively increased byproduct‐diffusion resistance within the PET particles. However, this critical particle size is far greater than the pellet sizes of commercial PET resins. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 102: 623–632, 2006     

Solid‐state polymerization of poly(ethylene terephthalate). II. Modeling study of the reaction kinetics and properties

Comprehensive modeling studies were used to describe the kinetics of the solid‐state polymerization (SSP) of poly(ethylene terephthalate). The validity of the model was confirmed by the successful fitting of the experimental results for molecular weight increases, at temperatures ranging from 180 to 230°C and for times up to 12 h, with one fitting parameter. The changes in the concentrations for hydroxyl end groups ([? OH]), carboxyl end groups ([? COOH]), vinyl end groups, and terephthalic acid (TPA) were simulated with the model. During SSP, the contents of not only hydroxyl and carboxyl end groups but also vinyl ester end groups and TPA monomer were predicted to decrease as a function of the SSP time and temperature. The effects of the pellet size and the molar ratio of carboxyl end groups to hydroxyl end groups were also calculated. At an end‐group molar ratio ([? COOH]/[? OH]) of around 0.7, a maximum SSP rate was obtained. As the [? COOH]/[? OH] ratio increased, the contents of the vinyl end groups and TPA monomer were predicted to increase. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 89: 213–227, 2003     

Morphological changes of poly(ethylene terephthalate‐co‐isophthalate) during solid state polymerization

Poly(ethylene terephthalate‐co‐isophthalate) (PETI) prepolymer was submitted to solid state polymerization (SSP) at 184–230°C in a fixed bed reactor, to study the evolution of morphological changes during the process. Short reaction times were selected to investigate crystallization phenomena during nonisothermal (heating) and isothermal SSP phases. More specifically, multiple PETI melting behavior was observed and attributed to secondary crystallization, the rate of which increased significantly with SSP temperature. Reaction time was also found to exert a positive effect on solid‐phase perfection of secondary crystals, leading at each temperature to melting points close to the value of bottle‐grade poly(ethylene terephthalate). Finally, the mass fraction crystallinity of the SSP grades was found to comply with the crystal morphology encountered. © 2011 Wiley Periodicals, Inc. J Appl Polym Sci, 2011     

Properties of poly(ethylene terephthalate)–poly(ethylene naphthalene 2,6‐dicarboxylate) blends with montmorillonite clay

The production and properties of blends of poly(ethylene terephthalate) (PET) and poly(ethylene naphthalene 2,6‐dicarboxylate) (PEN) with three modified clays are reported. Octadecylammonium chloride and maleic anhydride (MAH) are used to modify the surface of the montmorillonite–Na⁺ clay particles (clay–Na⁺) to produce clay–C18 and clay–MAH, respectively, before they are mixed with the PET/PEN system. The transesterification degree, hydrophobicity and the effect of the clays on the mechanical, rheological and thermal properties are analysed. The PET–PEN/clay–C18 system does not show any improvements in the mechanical properties, which is attributed to poor exfoliation. On the other hand, in the PET–PEN/clay–MAH blends, the modified clay restricts crystallization of the matrix, as evidenced in the low value of the crystallization enthalpy. The process‐induced PET–PEN transesterification reaction is affected by the clay particles. Clay–C18 induces the largest proportion of naphthalate–ethylene–terephthalate (NET) blocks, as opposed to clay–Na⁺ which renders the lowest proportion. The clay readily incorporates in the bulk polymer, but receding contact‐angle measurements reveal a small influence of the particles on the surface properties of the sample. The clay–Na⁺ blend shows a predominant solid‐like behaviour, as evidenced by the magnitude of the storage modulus in the low‐frequency range, which reflects a high entanglement density and a substantial degree of polymer–particle interactions. Copyright © 2005 Society of Chemical Industry     

Effect of poly(ethylene glycol) on the solid‐state polymerization of poly(ethylene terephthalate)

Poly(ethylene glycol) (PEG) and end‐capped poly(ethylene glycol) (poly(ethylene glycol) dimethyl ether (PEGDME)) of number average molecular weight 1000 g mol^?1 was melt blended with poly(ethylene terephthalate) (PET) oligomer. NMR, DSC and WAXS techniques characterized the structure and morphology of the blends. Both these samples show reduction in T_g and similar crystallization behavior. Solid‐state polymerization (SSP) was performed on these blend samples using Sb₂O₃ as catalyst under reduced pressure at temperatures below the melting point of the samples. Inherent viscosity data indicate that for the blend sample with PEG there is enhancement of SSP rate, while for the sample with PEGDME the SSP rate is suppressed. NMR data showed that PEG is incorporated into the PET chain, while PEGDME does not react with PET. Copyright © 2005 Society of Chemical Industry     

Transesterification reactions in triphenyl phosphite additivated‐poly(ethylene terephthalate)/poly(ethylene naphthalate) blends

The occurrence of transesterification reactions in poly(ethylene terephthalate) (PET)/poly(ethylene naphthalate) (PEN) blends prepared in presence of triphenyl phosphite (TPP) was investigated. When PEN was processed with TPP, which is a known chain extender for PET, chain extension reactions also took place. Torqueprocessing time curves obtained during preparation of 75/25 PET/PEN blends containing TPP, showed a build‐up profile followed by a fast decrease that was interpreted as chain extension between blend components and degradation due to phosphite residues formation, respectively. Although transesterification inhibition was expected, this type of reaction was not suppressed by TPP.     

A comprehensive single‐particle model for solid‐state polymerization of poly(L‐lactic acid)

We propose here, a comprehensive model for the solid‐state polymerization (SSP) of a low to moderate molecular weight (MW) prepolymer of lactic acid, to produce high MW poly(L ‐lactic acid) (PLLA). The reactions are rationally assumed to occur only in the amorphous region, and effective concentrations of end groups, vary with crystalinity, X_c, during SSP. We estimate byproduct diffusivities, D, using free volume theory. The effects of various parameters on the SSP of PLLA prepolymer have been examined with respect to the optimum MW, X_c and D. We introduce self‐consistently, scaling factors of ～ 0.27, in the experimental procedure, to determine via ¹⁹F‐NMR, concentrations of the end groups, after converting them to fluorinated ester groups. The relevant reaction rate constants are obtained by fitting to early time data from representative SSP experiments at 150°C, under high vacuum, on PLLA prepolymer powder (i.e., spherical geometry) of number average MW, M_n0 ～ 10,200 Da, which attains M_n ～ 150,000 Da, via SSP. The subsequent successful comparison of the model predictions with experimental data throughout the entire SSP duration indicates that the model is comprehensive and accounts for all the relevant phenomena occurring during the SSP to synthesize high MW PLLA. © 2011 Wiley Periodicals, Inc. J Appl Polym Sci, 2011     

Continuous production of polyester‐poly(ethylene terephthalate) resins in melt‐phase and solid‐state reactors

A simple reaction model has applied net polycondensation rates to predict the steady‐state performance of three distinct continuous processes for manufacturing polyester‐PET resins. A net melt‐phase polycondensation rate was described by the simple second‐order kinetics. A net solid‐state polycondensation rate was assumed to follow the modified second‐order kinetics with respect to active end group concentration. A moving‐packed bed requires a longer residence time to deal with the diffusion‐limited SSP of standard pellets or challenging pastilles. The calculations and data showed low IV pastilles to have much slower diffusion‐controlled SSP rates than medium IV pellets. The tanks‐in‐series model demonstrated a narrow RTD in a gas fluidization bed with five mixing stages. Higher reaction temperatures may significantly increase the low diffusion resistance SSP rates of smaller beads or micro‐pellets in a gas‐fluidized reactor. The reaction‐controlled SSP of micro‐beads becomes apparent at 230°C. The high IV melt resins may challenge the slow reaction rates of Ti or Al‐catalyzed SSP resins. The efficacy of catalyst promoters on Ti activity enhancement may depend upon various ligands in Ti glycolate, Ti citrate, or titanic acid. The thermo‐oxidative stability of Ti or Al‐catalyzed resins may decrease at higher hot air drying temperatures (188°C or above). POLYM. ENG. SCI., 57:505–519, 2017. © 2016 Society of Plastics Engineers     

Solid‐state polymerization of poly(ethylene terephthalate). I. Experimental study of the reaction kinetics and properties

The solid‐state polymerization (SSP) reaction kinetics of poly(ethylene terephthalate) were investigated in connection with the initial precursor intrinsic viscosity (IV; molecular weight). Evaluations were performed with otherwise equivalent precursors melt‐polymerized to IVs of 0.50, 0.56, and 0.64 dL/g. The changes in the molecular weight and other properties were monitored as functions of the reaction times at solid‐state temperatures of 160–230°C. Precursors with higher initial molecular weights exhibited higher rates of SSP than those with lower initial values, as discussed in connection with the levels of crystallinity and the carboxyl and hydroxyl end‐group composition. Activation energies decreased at temperatures above 200°C, and this indicated a change in the SSP reaction mechanism. At temperatures of 200–230°C, similar activation energies were required for the polymerization of all three precursors. Lower temperature polymerizations, from 160 to 200°C, required higher activation energies for all precursors, with the 0.50‐IV material requirement almost twice as high as that calculated for the higher IV precursors. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 89: 197–212, 2003     

End‐group determination in poly(ethylene terephthalate) by infrared spectroscopy

A complete infrared (IR) spectroscopy experimental procedure was used to determine end‐group concentrations of poly(ethylene terephthalate) (PET). The correlations of the hydroxyl and carboxyl units were derived independently for accurate calibration results. The intermediate monomer of PET, bis(hydroxyethyl terephthalate), was used to prepare hydroxyl end‐group standards and titration measurements were used to determine the carboxyl content for the carboxyl end‐group standards. A double‐Gaussian form equation was defined to account for the interference between the hydroxyl and the carboxyl absorbance peaks in the PET IR spectrum. Some deviation was found from the assumption traditionally used for end‐group determination, stating that carboxyl and hydroxyl are the only end‐group units available in PET. © 2002 Society of Chemical Industry     

Crystallization kinetics of poly(ethylene terephthalate)/poly(ethylene 2,6‐naphthalate) blends

The crystallization kinetics of poly(ethylene terephthalate)/poly(ethylene 2,6‐naphthalate) (PET/PEN) blends were investigated by DSC as functions of crystallization temperature, blend composition, and PET and PEN source. Isothermal crystallization kinetics were evaluated in terms of the Avrami equation. The Avrami exponent (n) is different for PET, PEN, and the blends, indicating different crystallization mechanisms occurring in blends than those in pure PET and PEN. Activation energies of crystallization were calculated from the rate constants, using an Arrhenius‐type expression. Regime theory was used to elucidate the crystallization course of PET/PEN blends as well as that of unblended PET and PEN. The transition from regime II to regime III was clearly observed for each blend sample as the crystallization temperature was decreased. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 81: 23–37, 2001     

Rheological properties of hyperbranched polymer/poly(ethylene terephthalate) reactive blends

The rheological properties in solution, in shear and in uniaxial elongation of poly(ethylene terephthalate) (PET) reacted together with hyperbranched polymers (HBPs) were investigated. Two different PET grades, of low and high molecular weights, were compounded with sub‐ to over‐stoichiometric concentrations of HBPs of second and fourth pseudo‐generation, and subsequently subjected to a solid‐state polycondensation (SSP). The formation of microgels, which occurs at high HBP concentration, gave rise to a large increase in melt elasticity and a related decrease in melt strength. At low HBP concentrations, the complex viscosity of the unreacted HBP/PET was considerably reduced, thus demonstrating a lubrication effect of the HBP molecules. During SSP, the intrinsic and shear viscosities exhibited a gradual increase, which was similar for both PET and HBP/PET blends, and was correlated to an increase in molecular weight, through linear‐chain extension and branching reactions. The elongational viscosity of the reactive blends was also increased as a function of reaction time, and this increase was much larger in the case of the HBP/PET blends. A 400% increase in melt strength of the PET was obtained by combining SSP and trace amounts of an HBP of second generation, without any decrease in drawability.     

Chain extension reaction in solid‐state polymerization of recycled PET: The influence of 2,2′‐bis‐2‐oxazoline and pyromellitic anhydride

Conventional and chain extended‐modified solid‐state polymerization (SSP) of postconsumer poly(ethylene terephthalate) (PET) from beverage bottles was investigated. SSP was carried out at several temperatures, reaction times, and 2,2′‐bis‐2‐oxazoline (OXZ) or pyromellitic anhydride (ANP) concentrations. The OXZ was added by impregnation with chloroform or acetone solution. Higher molecular weights were reached when the reaction was carried out with OXZ, resulting in bimodal distribution. The molecular weights of the flakes reacted at 230°C for 4 h were 85,000, 95,000, and 100,000 for samples impregnated with 0, 0.5, and 1.25 wt % OXZ solution, respectively. In the case of reactions with ANP, branched chains were obtained. The thermal and thermal‐mechanical‐dynamic properties of these high‐molecular‐weight recycled PET were determined. For OXZ‐reacted samples, the reduction of crystallinity was observed as the reaction time was increased, becoming evident the destruction of the crystalline phase. The chain extended samples did not show changes in thermal relaxations or thermal degradation behavior. © 2009 Wiley Periodicals, Inc. J Appl Polym Sci, 2010     

Morphology and barrier mechanism of biaxially oriented poly(ethylene terephthalate)/poly(ethylene 2,6‐naphthalate) blends

To improve the barrier properties of poly(ethylene terephthalate) (PET), PET/poly(ethylene 2,6‐naphthalate) (PEN) blends with different concentrations of PEN were prepared and were then processed into biaxially oriented PET/PEN films. The air permeability of bioriented films of pure PET, pure PEN, and PET/PEN blends were tested by the differential pressure method. The morphology of the blends was studied by scanning electron microscopy (SEM) observation of the impact fracture surfaces of extruded PET/PEN samples, and the morphology of the films was also investigated by SEM. The results of the study indicated that PEN could effectively improve the barrier properties of PET, and the barrier properties of the PET/PEN blends improved with increasing PEN concentration. When the PEN concentration was equal to or less than 30%, as in this study, the PET/PEN blends were phase‐separated; that is, PET formed the continuous phase, whereas PEN formed a dispersed phase of particles, and the interface was firmly integrated because of transesterification. After the PET/PEN blends were bioriented, the PET matrix contained a PEN microstructure consisting of parallel and extended, separate layers. This multilayer microstructure was characterized by microcontinuity, which resulted in improved barrier properties because air permeation was delayed as the air had to detour around the PEN layer structure. At a constant PEN concentration, the more extended the PEN layers were, the better the barrier properties were of the PET/PEN blends. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 101: 1309–1316, 2006     

Solid-state polymerization of poly(ethylene terephthalate)

Solid -state polymerization of poly(ethylene terephthalate) (PET) is carried out by heating the low molecular weight prepolymer at temperatures below its melting point but above its glass transition temperature. Postcondensation occurs and the condensation byproducts can be removed by applying vacuum or inert gas. Polymers obtained usually have high molecular weight, low carboxyl and acetaldehyde content, and can be used for beverage bottle or industrial yarns. Polyesters for textile purposes are manufactured by a melt process. Chemical reactions involved in the solid state polymerization are transesterification, esterification, as well as the diffusion of byproducts. Overall reaction rate was governed by the molecular weight, carboxyl content of prepolymer, crystallinity, particle size, reaction temperature, and time. Prepolymer for solid state polymerization should have intrinsic viscosity 0.4 dL/g or more, density 1.38 g/mL, and minimum dimension 3 mm or less. The reaction temperature could be 200–250°C. When textile grade PET is used as prepolymer, crystallization at 180–190°C for 1–2 h increases the density to 1.38 g/mL. Polymerization at 240–245°C for 3–5 h can raise the intrinsic viscosity to 0.72 dL/g and carboxyl content less than 20 meq/kg. Appropriate reaction conditions are subject to the properties of prepolymers and the design of reactors. Reactor used for solid state polymerization could be vacuum dryer type or stationary bed. The former is suitable for a small capacity and is run batchwise. The latter is a continuous process and is economically feasible for large -scale production.     

Solid‐state polymerization of poly(ethylene terephthalate). III. Thermal stabilities in terms of the vinyl ester end group and acetaldehyde

Poly(ethylene terephthalate) (PET), precursors, and solid‐stated samples were evaluated in terms of changing vinyl ester (VE) concentrations. The results obtained through the application of reaction kinetics gave VE contents ranging from 0.7 to 5.5 mmol/kg of PET. As the initial intrinsic viscosity (IV) of the precursor increased, the VE content also increased, representing the different thermal histories of the samples during melt‐phase polymerization. The VE contents decreased as the solid‐state polymerization (SSP) time increased from 0 to 12 h and as the temperature of SSP increased up to 220°C. A series of acetaldehyde (AA) generation experiments were conducted from 270 to 300°C with samples solid‐stated from three precursors with different initial IVs. The rate of AA generation decreased as the final IV of the solid‐stated PET increased, and this showed that the SSP process improved the thermal stability of PET. The AA generation rates of samples that had similar final IVs but were solid‐stated from different IV precursors were also compared. When the heating temperature was low, the amount and rate of AA generation were higher for samples with higher initial precursor IVs. This tendency, however, became less clear as the generation temperatures increased, probably because interference from the dissociation reactions (occurring between the polymer chains to produce VE) increased with increasing temperature. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 89: 228–237, 2003     

Effects of electron beam irradiation on poly(ethylene 2,6‐naphthalate)/poly(ethylene terephthalate) blends

The effect of electron beam (EB) irradiation on the properties and compatibility of poly(ethylene 2,6‐naphthalate) (PEN)/poly(ethylene terephthalate) (PET) blends was investigated. Upon EB irradiation, PEN/PET blends underwent transesterification reactions, resulting in the formation of more random copolymers from the original binary pair. The degree of transesterification increased with dose rate, and all of the irradiated blends exhibited a single glass transition temperature. This indicated that transesterification reactions promoted by EB irradiation led to the formation of a single phase. Transesterification reactions promoted by EB irradiation led to more random copolymers, and the reduced regularity in the irradiated blends decreased the melting temperature. A higher degree of randomness and lower number‐average sequence lengths for the blend systems indicated that a more random chain structure was formed in the blends. The rheological measurements demonstrated that the irradiated PEN/PET blends were miscible. EB irradiation could promote transesterification reaction, thus enhancing the compatibility of PEN/PET blends.     

Glass‐transition and melting behavior of poly(ethylene terephthalate)/poly(ethylene 2,6‐naphthalate) blends

The glass‐transition temperatures and melting behaviors of poly(ethylene terephthalate)/poly(ethylene 2,6‐naphthalate) (PET/PEN) blends were studied. Two blend systems were used for this work, with PET and PEN of different grades. It was found that T_g increases almost linearly with blend composition. Both the Gibbs–DiMarzio equation and the Fox equation fit experimental data very well, indicating copolymer‐like behavior of the blend systems. Multiple melting peaks were observed for all blend samples as well as for PET and PEN. The equilibrium melting point was obtained using the Hoffman–Weeks method. The melting points of PET and PEN were depressed as a result of the formation of miscible blends and copolymers. The Flory–Huggins theory was used to study the melting‐point depression for the blend system, and the Nishi–Wang equation was used to calculate the interaction parameter (χ₁₂). The calculated χ₁₂ is a small negative number, indicating the formation of thermodynamically stable, miscible blends. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 81: 11–22, 2001