ABSTRACT:A modified glycolysis reaction of recycled poly(ethylene terephthalate) (PET) bottles by ethylene glycol (EG) was investigated. Influences of the glycolysis temperature, the glycolysis time, and the amount of catalysts (per kg of recycled PET) were illustrated in this study. The manganese acetate was used as a glycolysis catalyst in this study. Bis-2-hydroxyethyl terephthalate (BHET) and its dimer were predominately glycolysis products. It was found the optimum glycolysis temperature is 190°C. And the best glycolysis condition is 190°C of glycolysis temperature, 1.5 h of glycolysis time, and 0.025 moles of manganese acetate based on per kg of recycled PET. If the best glycolysis condition is conducted, the glycolysis conversion may be as high as 100%. For a given reaction time (1.0 h), the ln(% glycolysis conversion) is linear to 1/T (K Ϫ1 ) and the activation energy (E) of glycolysis reaction is around 92.175 kJ/(g mole). The glycolysis conversion rate increases significantly with increasing the glycolysis temperature, the glycolysis time, or the amount of manganese acetate (glycolysis catalyst). Thermal analyses of glycolysis products were examined by a differential scanning calorimetry (DSC) and a thermogravimetric analysis (TGA). According to the definition of a 2 3 factorial experimental design, the sequence of the main effects on the glycolysis conversion of the recycled PET, in ascending order, is the glycolysis time (0.18) Ͻ the amount of catalyst per kg of the recycled PET (0.34) Ͻ the glycolysis temperature (0.40). Meanwhile, the prediction equation of glycolysis conversion from the result of a 2 3 factorial experimental design is Ŷ ϭ 0.259ϩ0

In this research a model to simulate both the filling the curing stages of a reaction injection molding (RIM) process in complex three-dimensional molds is developed. This model can be used to predict not only the temperature and conversion changes with time but also the front position during fllling. Using given physical and chemical properties of the RIM system, moldability can be determined in advance. The numerical techniques used in this research include adaptation of the SIMPLE algorithm developed by Patankar for a moving-front, two-phase system with non-negligible inertial effects, and exothermic chemical reaction. The model predictions of temperature and conversion compare favorably with available data on simple two-dimensional molds. The ability of the model to predict the dynamics of filling in more complicated molds was verified by comparison to mold filling experiments with water and a polyurethane foam.

No abstract

The objective of this research is to demonstrate the application of a general model of reaction injection molding (RIM) to a complex shaped mold part. Using the fundamental physical and chemical properties of the RIM system, prediction of RIM moldability can be made. The numerical techniques used in this research are presented in a previous paper (1). The rigorous modeling of the transport processes and the dynamics of the filling front for a particular application are described in this work. MTRODUCTIONhe objective of this research is to develop and T apply a tool for the evaluation of flow, temperature. and conversion in thick-walled or complex polymeric part prepared by reaction injection molding (RIM). Essentially, all parts currently produced by the molding of reactive polymers are characterized by thin walls. The Wing of such molds is often essentially two-dimensional. Dicyclopentadiene (DCPD) (2) based systems, however, can be prepared to delay the onset of polymerization. Thus, large complex shaped or thick-walled parts can be filled before gelling. The description of the filling process for purpose of optimization and control requires a model with three-dimensional capability.Because monomers and low-molecular-weight polymers are used as feed materials in RIM, the viscosity of the reactants is low. Because of the low viscosity of the reactants, inertial effects are often significant during filling. This, combined with the rapid and ex@ thermic reaction, results in a very complicated filling behavior. Complex shaped or thick-walled parts may exhibit significant gradients in physical and chemical properties and may be sensitive to the filling dynamics and operating conditions. In addition, trial and error approaches to mold design are not economically viable if several-hundred-pound parts are produced in thick-walled molds. These complications require sophisticated computer-aided tools for process simulation, design, and control. A model developed in our laboratories (1). however, demonstrates the capability for predicting flow, reaction, and heat transfer behavior in a complicated three-dimensional mold. The basic capacity of the model to simulate flow, reaction, and heat transfer phenomena in a large, complex RIM mold and operating condition effects on product properties is demonstrated in this paper.The numerical techniques used in this research include adaptation of the SIMPLE algorithm developed by Patankar (3) for a moving-front, two-phase system with non-negligible inertial effects, and exothermic chemical reaction. The model predictions of temperature compare favorably with available data on simple two-dimensional molds. In a previous paper, the data of Castro and Macosko (4) were used to test the model's ability to predict concentration and temperature in a simple mold. The ability of the model to predict the dynamics of the filling front in more complicated molds was evaluated by comparison to mold filling experiments with water and polyurethane foam (1).In this work, the applicability of the mod...

No abstract