Chemorheological characterization of thermoset cure

Thermosetting resin/glass cloth composites find extensive use in the packaging of electrical circuits into multi-layered circuit boards. To determine optimum processing conditions, it is necessary to understand the rheology of the resin as it cures. In this study, a squeezing flow geometry was used to determine the shear viscosity of a high performance epoxy resin during cure; Viscosity profiles were obtained during rising temperature cure. The results were compared with the complex viscosities obtained using the dynamic oscillatory parallel plate geometry. A numerical optimization algorithm was used to obtain the dual Arrhenius viscosity model parameters from the experimental viscosity data. The sensitivities of the model parameters and their effects on the predicted viscosity profiles were also determined.     

A simple dilatometer for thermoset cure shrinkage and thermal expansion measurements

A simple capillary and bulb mercury dilatometer designed for specific volume measurements on thermoset resin systems during the curing reaction and as a function of temperature is described. The design, calibration, operation, data treatment, and error analysis are presented in detail, with data on the bisphenol A dicyanate resin system used as an example. Particular attention is directed at experimental difficulties such as monomer/prepolymer degassing, filling the dilatometer under vacuum, adhesive distortion of the curing resin on the dilatometer bulb, and the dilatometer bulb to capillary connection problem. © 1994 John Wiley & Sons, Inc.

1 This article is a US Government work and, as such, is in the public domain in the United States of America.

     

Variation of thermal conductivity and specific heat during cure of thermoset polyesters

Variation of thermal conductivity and specific heat of a general-purpose unsaturated polyester during cure was investigated experimentally. Measurements were made on the cured, partially cured, and uncured samples, and the results are presented. Both thermal conductivity and specific heat of the cured samples increase nearly linearly with increasing temperature. Effect of degree of cure on the thermal properties is evident during the earlier stages of cure.     

A fiber optic thermoset cure monitoring sensor

Curing determines the chemical and physical properties of a reacting resin. Among these, the optical properties strongly correlate with the structural features of the developing polymeric network. By monitoring changes of the refractive index, it is possible to analyze the polymerization of thermoset resin. In this work, a fiber optic sensor system has been designed and developed (based on the optical time domain reflectometry principle) to measure the reflection coefficient at the interface between the fiber optic and the resin during a curing process. Correlation between the sensor output and conversion has been proposed, following the Lorentz‐Lorenz law. Isothermal data from the sensor have been compared with calorimetric analysis of an epoxy‐based resin.     

A thermal and rheological investigation during the complex cure of a two‐component thermoset polyurethane

The complex cure kinetics of the reaction between oligomeric diphenylmethane diisocyanate (PMDI) and glycerol was characterized through thermal and rheological techniques. Isoconversional analysis of Differential scanning calorimetry (DSC) data resulted in the activation energy varying with conversion. Isothermal analysis gave activation energies ranging from 5 kJ/mol to 33.7 kJ/ mol, whereas nonisothermal data gave values for the activation energy ranging from 49.5 to 55 kJ/mol. Incomplete cure was evident in isothermal DSC, becoming diffusion controlled in the final stages of cure. DMA analysis on the cured material gave a glass transition temperature of 104 ± 3°C, which was evidence for vitrification of the curing system. The primary and secondary hydroxyl group reactivity was dependant on the isothermal cure temperature. Rheological studies of viscosity increase and tan δ changes with time revealed a complex cure process, with primary and secondary hydroxyl reactivity also showing dependence on isothermal cure temperatures, reflecting similar results obtained from isothermal DSC studies. The independence of tan δ on frequency was used to determine the point where the polymer formed an infinite network and was no longer able to flow, providing an overall activation energy attained at the gel point of 77.4 ± 4.4 kJ/mol. © 2009 Wiley Periodicals, Inc. J Appl Polym Sci, 2009     

A comparison of chemorheological models for thermoset cure

A chemorheological study of a thermoset system consisting of diglycidyl ether of bisphenol-A (DGEBA) epoxy resin and a mixture of two armoatic amine curing agents; 4,4′ methylenedianiline (MDA) and m-phenylene diamine (m-PDA), was conducted. Experimentally obtained viscosity data were checked against the predictions of two different viscosity models; one based on a phenomenological equation obtained by modification of the classical Williams–Landel–Ferry (WLF) equation and the other based on an extension of the branching theory originally proposed by Flory. In general, the predictions of both models were in excellent agreement with experimentally obtained isothermal and dynamic viscosity data. The branching theory model was found to have a slight advantage over the phenomenological equation model in describing the viscosity prior to gelation in a fast heating cure cycle.     

Kinetics modeling and time-temperature-transformation diagram of microwave and thermal cure of epoxy resins

Stoichimetric mixtures of a diglycidyl ether of bisphenol A (DGEBA)/ diaminodiphenyl sulfone (DDS) and a DGEBA/meta phenylene diamine (mPDA) were cured using both microwave and thermal energy. Fourier transform infrared (FTIR) was used for the measurement of the extent of cure and thermal mechanical analysis (TMA) was used for the determination of the glass transition temperature (T_g). The cure kinetics of the DGEBA/mPDA and DGEBA/DDS systems were described by an autocatalytic kinetic model up to vitrification in both the microwave and thermal cure. For the DGEBA/mPDA system, the reaction rate constants of the primary amine-epoxy reaction are equal to those of the secondary amine-epoxy reaction, and the etherification reaction is negligible for both microwave and thermal cure. For the DGEBA/DDS system, the reaction rate constants of the primary amine-epoxy reaction are greater than those of the secondary amine-epoxy reaction and the etherification reaction is only negligible at low cure temperatures for both microwave and thermal cure. Microwave radiation decreases the reaction rate constant ratio of the secondary amine-epoxy reaction to the primary amine-epxy reaction and the ratio of the etherification reaction to the primary amine-epoxy reaction. T_g data were fitted to the DiBenedetto model. A master curve and a time-temperature-transformation (TTT) diagram were constructed. The vitrification time is shorter in microwave cure than in thermal cure, especially at higher isothermal cure temperatures. For the DGEBA/mPDA system, the minimum vitrification time is two to five times shorter in the microwave cure than in the thermal cure. For the DGEBA/DDS system, the minimum vitrification time is 44 times shorter in the microwave cure than in the thermal cure.     

Kinetic studies of thermoset cure reactions: A review

This article presents a review of the kinetic studies on the cure reactions of thermosetting resins. The emphasis is placed on those conducted using the thermal analysis by differential scanning calorimetry. Two important categories of kinetic models are discussed and some existing parameter estimation techniques are presented. A variety of factors affecting the cure reactions, including the formulation of cure and process variables, are discussed. At the end, two different approaches in modeling the chemoviscosity of polymeric reactive systems are presented and some existing chemorheological models developed based on these approaches are briefly reviewed.     

Synthesis,characterization, and cure of allyl and propargyl functionalized indene as a thermoset composite matrix resin

Two highly functionalized resins were synthesized by the phase transfer reaction of indene with propargyl bromide or allyl chloride in the presence of strong base. The resins consisted of a mixture of tri- and tetrafunctional indenes with 60–80% of the product being tetrafunctional. The allylated (AL) and propargylated (PL) indene resins were thermally cured without added catalysts. Both resins exhibited a broad, highly exothermic cure with a peak energy at 320°C for AL resin and 282°C for PL resin. Thermal degradation of cured AL resin was found to begin at approximately 400°C with a carbon yield of 20% of its initial weight at 1000°C. Carbon yields for cured PL resin were excellent, with 68% retention of weight at 1000°C. Unidirectional, carbon fiber composites were fabricated from the substituted indene resins. AL–carbon fiber composites gave modulus values of 126 GPa and strength values of 967 MPa, while PL–carbon fiber composites gave modulus values of 116 GPa and strength values of 935 MPa in three-point bending tests. © 1998 John Wiley & Sons, Inc. J Appl Polym Sci 68: 475–482, 1998     

Experimental characterization of a curing thermoset epoxy‐anhydride system—Isothermal and nonisothermal cure kinetics

This work describes in detail the kinetic model for the cure of an epoxy‐anhydride thermoset matrix resin system. The cure kinetics in both nonisothermal and isothermal modes has been characterized using differential scanning calorimetry. The Sestak–Berggren two‐parameter autocatalytic model was used to describe the nonisothermal cure behavior of the resin satisfactorily. The isothermal cure data was fitted with Kamal's four‐parameter autocatalytic model, coupled with a diffusion factor. These characterization data will form material property inputs for a multiscale modeling framework for the estimation of cure‐induced residual stresses in thick thermoset matrix composites. © 2012 Wiley Periodicals, Inc. J. Appl. Polym. Sci., 2013     

The characterization of thermoset cure behavior by differential scanning calorimetry. Part I: Isothermal and Dynamic Study

A study of the isothermal cure kinetics of an unsaturated polyester resin by differential scanning calorimetry is described. An autocatalyzed second order kinetic model is adopted to elucidate the cure reaction and also assess the kinetic parameters. The rate constant, the maximum cure rate, the extent of cure, and the degree of conversion at the maximum cure rate, all increase with increasing cure temperature, while the half-life and the time required to reach the maximum cure rate both decrease. Discrepancies between the experimental results and the predicted values of some of the kinetic parameters, especially at high degrees of conversion, are attributed to the highly different-controlled cure reaction following the gel point. The activation energy (E) and the pre-exponential factor (1n A) of the polyester cure reaction were estimated, using differential graphical techniques, to be 131 ± 4 kJ/mol and 39 ± 1 respectively. An ASTM method (E698) produces erroneously low values of the Arrhenius parameters, suggesting that the assumptions of the method may be overly simplified.     

Effect of melamine-formaldehyde structure on cure response of thermoset coatings

Reactivity of etherified melamine-formaldehyde (MF) resins with polyol binders depends on the pattern of substitution on off-ring nitrogen atoms. These sites can be doubly substituted [-N(CH₂OCH₃)₂] or singly substituted [-N(H)(CH₂OCH₃)]. The (N-H) portion of singly substituted sites is referred to as an imino group. The main finding of this work is that acid catalyzed cure response is improved greatly when the imino content is reduced to very low levels. The effect of degree of substitution on MF basicity and the site of protonation, ring, or side chain are key elements in the explanation of this cure response behavior. Properties were determined for films cured for 30 min at various temperatures with constant catalyst level and type (0.5 phr paratoluenesulfonic acid). Resins of varying imino content were used. With the MF resin of lowest imino content, extensive cure of acrylic and polyester polyols was observed at cure temperatures as low as 180°F (82°C). Paint tests of hardness and solvent resistance were useful for determining cure response. Dynamic mechanical analysis (DMA) on free films also indicated cure response differences. Storage modulus values from DMA were used to calculate crosslink densities of the thermoset films. Presented at Coatings for Asia ’99, August 30–September 1, 1999, Singapore. 730 Worcester St., Springfield, MA 01151.     

Addition polyimides. I. Kinetics of cure reaction and thermal decomposition of bismaleimides

The thermal polymerization of four structurally different bismaleimide resins, prepared by reacting maleic anhydride with four aromatic diamines, viz., 4,4′-diaminodiphenyl methane, 4,4′-diamino diphenyl ether, 4,4′-diamino diphenyl sulfone, and 3,3′-diamino diphenyl sulfone, was followed by differential scanning calorimetry (DSC). The enthalpy change and the kinetic constants for the polymerization reactions were evaluated from the DSC curves. Thermal stability of the cured polymers was studied by thermogravimetry (TG). The kinetic parameters, viz., activation energy E and preexponential factor A, for the thermal decomposition of the cured bismaleimides were calculated from the TG curves using three nonmechanistic integral equations. The kinetic constants (E and A) follow a trend similar to the thermal stability of the polymers.     

Modeling of chemorheology of thermoset cure by modified WLF equation

An analysis of chemorheology of cure of a formulation composed of diglycidyl ether of bisphenol-A (DGEBA) type epoxy resin and a mixture of aromatic amines was performed. A series of kinetic and rheological tests were run, and the experimentally obtained results were checked against the proposed model. A modified form of the classical Williams–Landel–Ferry (WLF) equation was used to model the chemorheology of the formulation. To arrive at the appropriate form of the modified WLF equation, expressions were developed for (i) the cure kinetics, (ii) the glass transition temperature as a function of the degree of cure, and (iii) the changes in parameters C₁ and C₂ as a function of temperature. By using these correlations in conjunction with the WLF equation, good agreement between predicted and experimental findings was observed.     

Kinetics and thermal characterization of epoxy‐amine systems

The curing behavior of epoxy resins was analyzed based on a simple kinetic model. We simulated the curing kinetics and found that it fits the experimental data well for both diglycidylether of bisphenol A–4,4′‐methylene dianiline and diglycidylether of bisphenol A–carboxyl‐terminated butadiene acrylonitrile–4,4′‐methylene dianiline systems. The kinetic results showed the curing of epoxy resins involves different reactive process and reaction stages, and the value of activation energy is dependent on the degree of conversion. By analyzing the effect of vitrification, at low curing temperature, we found the curing reaction at the later stage was practically diffusion‐controlled for unmodified resin, and the rubber component did not markedly decrease T_g at the early stage of reaction as would be expected. © 1999 John Wiley & Sons, Inc. J Appl Polym Sci 71: 2401–2408, 1999     

Process modelling of thermal spraying for thermoset coatings

Thermal spraying has the ability to coat large components and structures on site. It has been used to deposit thermoplastic coatings but little work has been done on thermosets. To form a thermoset coating by thermal spraying, the feedstock particles have to melt in the flame without degradation but also the coating itself must then be cured by the end of the deposition process. The particle residence time in a typical thermal-spray flame during deposition is less than 0.01 s, whereas conventional thermoset powders need orders of magnitude more time at the baking temperature to achieve a sufficient degree of cure. However, the freshly deposited coating also receives heat from the scanning flame during thermal-spray deposition, which should contribute to the curing process. This effect is complicated by the fact that the heat input during scanning varies with time. To control the curing process, therefore, a full understanding of the interaction between the flame, coating and substrate is essential. In this paper, experimental trials and computer simulations were carried out aimed at controlling the temperature profiles of the thermoset deposit to enable sufficient cure to take place. Software developed by the authors was used to simulate flame scanning and post heat treatment. Commercial software was also applied to simulate steady-state infrared heating. The results indicate that process parameters have a critical effect on the properties of the coatings and can be optimized with the aid of computer simulation.     

Thermoset injection molding dynamics and the distribution of cure in thermoset molded parts

A detailed characterization of a commercial-filled unsaturated polyester molding compound has been carried out to determine the kinetics of cure and the rheological behavior of the material at various temperatures and shear rates. Molding experiments were conducted in a 2 1/3 oz, 68 ton reciprocating screw injection molding machine, in conjunction with a simple rectangular cavity. The cavity and nozzle were equipped with pressure transducers to determine, the variation of pressure with position throughout the injection molding cycle. The injection speed was determined with the help of a position transducer. Finally, the moldings were analyzed to determine the distribution of cure states and tensile properties in the molding at various cure times. Significant differences have been observed. It is expected that studies of this type should be helpful in obtaining a better understanding of the thermoset injection molding process and the development of mathematical models to simulate this process.     

Effect of temperature gradients on cure and stress gradients in thick thermoset castings

The temperature gradient developed during casting of an unsaturated polyester for which the kinetics of reaction and the heat capacities, thermal conductivities, and densities were known from earlier work was measured experimentally as a function of time in a thick casting and compared with predicted values calculated from a mathematical model. Agreement was excellent. The castings were sectioned and the sections were examined by birefringence. These examinations show good agreement between the temperature gradient developed during cure and the residual stress gradient. This work demonstrates that the extent of cure as a function of position can be predicted from heat transfer calculations if the reaction kinetics and thermal properties are known and that the residual stress gradient is dependent upon the temperature gradient developed.