Calorimetric and microwave dielectric monitoring of epoxy resin cure

The cure process of a BADGE (diglycidyl ether of bisphenol-A) resin (Epon 828) and ethylenediamine has been investigated by means of calorimetry and dielectrometry in the microwave region (10⁷–10¹⁰ Hz) in the temperature range 50 to 70°C. Kinetic data from calorimetry were analyzed in detail. An overall kinetic order of 2.5 has been obtained. The time domain reflectometry (TDR) has been used to characterize pure components and their mixtures. Cure monitoring was carried out with both TDR and a cavity method at fixed frequency (9.5 × 10⁹ Hz). A very good agreement was obtained between the reaction rate as measured by calorimetry and the rate of decrease of dielectric constants up to very high conversion. This was explained by admitting that the rates of disappearance of dielectric dipoles and of reactive species coincided.     

Modeling the dielectric response of unsaturated polyester resin during cure

The dielectric relaxation of unsaturated polyester resin during cure was modeled using a modified Williams–Watts decay function. The dielectric response was studied using dielectric measurements coupled with two dynamic mechanical measurement methods. It was found that the variation of the relaxation time during cure is a WLF process using T_g (α) (α-conversion) as the varied temperature. It was shown that this process fits the Williams-Watts decay function Φ(t)=exp(-(t/τ)^β) where τ-relaxation time and β-empirical parameter are time dependent. It was found that τ obeys a time dependent power law rule and β depends linearly on log(time). Using this modified decay function, it was shown that the experimental dielectric response measured during cure agrees well with the computed response. Relaxation times above and below the dielectric measurement system capability were computed thus, demonstrating the capability of yielding the entire relaxation times spectrum during cure, out of a single limited frequency dielectric measurement.     

The effect of state of cure on bond performance for an epoxy resin system

A study of the influence of the state and mode of cure of an epoxy resin, diglycidyl ether of bisphenol A cured with 4,4′-diaminodiphenylmethane, has been made. The state of cure was assessed by differential scanning calorimetry as residual heat of reaction and a transition temperature T^* corresponding to either the melting point or the glass transition temperature. The state of cure was varied by controlling both the time and temperature of cure. Bond strengths can be related to either T^g or residual heat of reaction, but the different cure schedules produce different relations; simultaneous measurement of both parameters is necessary to specify the state of cure. Some additional work has been carried out on the effect of delayed cure on bond performance and on the effect of varying the cure schedule for a phenol-formaldehyde adhesive. Both single- and double-overlap joints have been used and a constant single/double bond strength ratio has been found. This constant is different for the two adhesives, reflecting their different behavior in shear.     

A thermoanalytical study of the cure characteristics of an epoxy resin system

Rates and extents of cure at different times and temperatures are determined for a modified epoxy resin system and related to gel time measurements. An empirical equation based on time-temperature superposition determines cure time as a function of degree and temperature of cure. The dependence of glass transition temperature on degree of cure, and heat capacity measurements on the cured resin, are discussed.     

Improved models of viscosity and relaxation modulus for epoxy resin during cure

In this work, epoxy resin viscosity was calculated by using reference viscosity and “viscosity‐cure” shift factor which was established to research the dependence of viscosity on curing degree. The predicted results were compared with experimental data and the agreements between them were better than those in reference. Generalized Maxwell model was used to characterize the viscoelastic behavior of epoxy resin and the relaxation modulus was modeled by Prony series. A simple model of stress relaxation times was derived as functions of viscosity and stiffness. The stress relaxation times at different curing degrees were calculated after acquiring the stress relaxation times at the reference curing degree. The relaxation modulus predicted by this method agreed well with experimental data. The above results showed that these models could well be used to predict the viscosity and relaxation modulus of thermosetting resin during cure. POLYM. ENG. SCI., 56:617–621, 2016. © 2016 Society of Plastics Engineers     

Monitoring the cure of an epoxy-anhydride resin

The cure of an epoxy-anhydride resin system used in pultrusion was characterized to develop an understanding of the cure behavior and to determine potential process controls parameters. Isothermal cures of neat resin formulations were monitored by differential scanning calorimetry (DSC), torsional braid analysis (TBA), and microdielectrometry (MDE). The processing conditions define a time/ temperature region in which monitoring would be applicable. Both DSC and MDE were found to yield useful information in this region, however, the events typically monitored by TBA either did not occur or occurred too quickly to be monitored. Significant ionic conductivity was observed in the fully cured resins at temperatures above the glass transition temperature and could possibly be used as a control parameter. This study revealed an apparent change in reaction mechanism with increasing cure temperature. DSC showed a change in activation energy with extent of reaction and a decrease in heat of reaction at the higher isothermal cure temperatures. The formation of a different network structure was indicated by a decreasing glass transition temperature of the cured resin with increasing cure temperature by both DSC and TBA.     

The relationship between dielectric and rheological behavior of epoxy resin during cure

Measurement of the frequency-dependent, vector voltage (V_c) provided an in-situ and non-destructive technique to measure continuously the rheological change of a resin due to polymerization, and can be used as the basis of real-time control. The vector voltage depends on the degree of polarization of the dipolar molecules and on the change of viscosity during cure; both result from the modified structure of the epoxy resin during cure, The initial stage of curing, represented by the former portion Of the Vc curve (divided at the minimum of the Vc curve), was caused mainly by the effects of temperature and viscosity. During the latter stage of the cure reaction, Vc alters because of the effect of the lightened matrix structure that inhibits alignment of dipoles. The duration of reaction. temperature of curing and degree of conversion all have the same effects on both vector voltage and viscosity, The minimum value of vector voltage is correlated to the minimum viscosity, and there is a nearly quantitative relationship between them, One can determine the viscosity of the epoxy resin during cure from reading of the vector voltage. Various reaction mechanisms may be explained based on the graphs of vector voltage of various types.     

Sensor system for monitoring impregnation and cure during resin transfer molding

In the resin transfer molding (RTM) fabrication of composites, knowledge of the position of a moving resin front during impregnation is important for process optimization. We describe here a simple, inexpensive, multi-point sensor system based on DC conductometry for determination of resin position in an RTM mold. This Resin Position Sensor (RPS) system consists of a matrix of small sensors embedded in the RTM tool, whose combined output can be used to produce a resin flow pattern at any given time after the start of impregnation. As it cures, the resin resistance increases and the sensor can then function as a cure monitor. A large, 24-sensor RTM tool was fabricated for demonstration of the RPS. Flow contour maps generated from sensor data during impregnation of both E-glass and carbon fiber preforms are shown.     

湿固化聚氨酯/环氧树脂固化体系的研究

以聚丙二醇(PEG)、甲基六氢苯酐(MHHPA)、异佛尔酮二异氰酸酯(IPDI)、二苯基甲烷二异氰酸酯(MDI)以及甲苯二异氰酸酯(TDI)为原料,分别合成出了三种末端含有异氰酸酯的反应性聚氨酯。评价了这三种反应性聚氨酯的物理机械性能,并将这三种聚氨酯分别与低黏度双酚A型环氧树脂按不同的比例混合,考察了三个不同的聚氨酯/环氧树脂固化体系的力学性能。研究表明,当反应性聚氨酯预聚体与环氧树脂的质量比为95∶5,固化条件为室温下7d或者室温下12h后,再在80℃放置4h时,这三种共混体系均能达到理想的黏结效果。     

Effects of variation in composition and temperature on the amine cure of an epoxy resin model system

Prior liquid chromatographic studies have shown that the reactions in epoxy resin model system phenyl glycidyl ether, p-chloroaniline, and Monuron include amine addition to epoxy, homopolymerization of the epoxy, and a chain-transfer reaction involving the hydroxy groups of the addition products. The present work examines the effect of variation in concentration of the accelerator Monuron, the amine-to-epoxy ratio, and the temperature on the competitive reaction mechanisms. The fraction of phenyl glycidyl ether reacting by homopolymerization increases with accelerator concentration and decreases with increasing amine-to-epoxy ratio and increasing temperature. The estimated contribution from chain transfer is much smaller and appears to parallel the homopolymerization reaction, as might be expected.     

Real-time dielectric studies of network formation in thermally activated epoxy-amine and isocyanate-triol systems

Dielectric measurements are reported on the changes that occur in the nature of the dipole relaxation processes during cure of an epoxy-amine and a diisocyanate-triol system. The epoxy resin forms a vitrified solid in the final cured state, whereas the urethane retains its elastomeric properties. The initial behaviour for the epoxy resin is dominated by ionic conduction processes. Subtraction of the conductivity contribution reveals a dipolar relaxation process, which is analysed using the Havriliak-Negami (HN) equation. The characteristic exponent 1 - n of the HN equation changes in a similar manner to that found with other epoxy-amine systems. However, the dipolar process in the urethane occurs at very high frequency and a different form of the 1 - n dependence is observed. Sensitivity of the dielectric method for the detection of vitrification and gelation is critically assessed. If the vitrification is allowed to occur, then detection of a gel point may not necessarily be inferred from the dielectric data.     

Variation in dielectric properties of an epoxy-novolac molding compound during dynamic cure

The variation in dielectric characteristics of an epoxy-novolac molding compound during dynamic cure was studied by means of dielectrometry (DE). Dielectric parameters such as permittivity (?′), loss factor (?″), and dissipation (tan δ) were observed to depend on various factors including temperature, frequency, the extent of cure (α, measured previously by using differential scanning calorimetry, DSC), as well as contributions from ionic conductivity (σ) and electrode polarization. The characteristic relaxation time (τ) and the relaxed permittivity (?_r), suggested in the literature as possible parameters for cure monitoring purposes, was found difficult to determine because of interfernce from ionic conduction and electrode polarization. In comparison, σ could be measured throughout the entire cure process and was observed to depend only on temperature and α. In combination with our previous DSC results, an empirical function relating α to temperature and α was constructed. Our analysis also indicated that the strong maximum in the ?′ and the ?″ curves during the course of dynamic cure was a direct result of the α-and the temperature-dependence of σ; the strong maximum of ?″ is directly related neither to gelation nor to vitrification.     

Phase diagram of different epoxy-amine precursors modified with a thermoplastic: Effect of structure of epoxy-amine system on miscibility

The miscibility of a thermoplastic with the precursors of different epoxy-amine systems was analyzed thermodynamically in which the dependence of interaction parameter on temperature and composition χ (T, ?) and the polydispersity of components were considered. The epoxy-amine precursors were different only in the nature of amino groups, which were provided by a monoamine and a diamine in different proportions. Cloud-point curves were measured for five unreacted modified systems resulting that miscibility of the system increased with the proportion of monoamine. The thermodynamic analysis was realized in two steps: first, the model was applied to each system individually and secondly, a general equation for χ (T, ?) depending on the monoamine-diamine proportion was searched and used to analyze all systems together. Theoretical calculations of cloud-point curves, shadow curves, spinodal curves, critical points, vitrification curves and species distributions were realized and discussed for those systems.     

Effect of cure history on dynamic mechanical properties of an epoxy resin

The effect of cure history on the dynamic thermomechanical properties of a high temperature curing epoxy resin has been studied using torsional braid analysis. In isothermal cures “full cure” is not possible except at temperatures above the maximum glass transition temperature (T_g) of the cured resin, hence the necessity of a “post-cure” after lower temperature isothermal cures. The highest T_g and maximum cross-linking in the cured resin was for a linear heating rate of 0.05°C/min from 30 to 200°C; higher heating rates lead to lower glass transition temperatures.     

Modelling the cure of a commercial epoxy resin for applications in resin transfer moulding

An analytical procedure has been developed for modelling the kinetics of the cure process of a commercial epoxy resin for resin transfer moulding (RTM) applications, using differential scanning calorimetry (DSC) in the isothermal and dynamic modes to obtain the experimental database. The overall reaction rate of the epoxide groups with amines was determined and fitted by an autocatalytic kinetic model. An improvement of the model to allow for diffusion limitation effects results in a good agreement between experimentally determined and predicted reaction rates. A non-linear least squares regression analysis method based on Marquardt's algorithm was used to fit the DSC reaction rate data with an appropriate model and to evaluate the activation energies and the reaction orders for this particular resin system. The Di Benedetto equation was utilised to establish the relationship between conversion and glass transition temperature (T_g), required to develop the diffusion-dominated part of the model.     

A chemorheological model for the cure of unsaturated polyester resin

A chemorheological model is developed, using the free volume concept, for the prediction of viscosity during the cure of unsaturated polyester resin. We have incorporated into the development of the chemorheological model a mechanistic kinetic model of curing kinetics that predicts the degree of cure as a function of cure time. The mechanistic kinetic model uses an approach of free-radical polymerization that takes into account diffusion-controlled curing reactions, In order to test the usefulness of the chemorheological model developed, we have conducted cure experiments and measured viscosities of partially cured resin samples, using a general-purpose unsaturated polyester resin. Specifically, the following measurements were taken: (1) the quantity of ethylenic double bonds in the resin system before and after the cure reaction by infrared spectroscopy, (2) the glass transition temperature by differential scanning calorimetry (DSC) and (3) the viscosity as a function of shear rate, at several temperatures, using a cone-and-plate rheometer. It is concluded that the chemorheological model developed is very useful for predicting the variation of viscosity during the cure of unsaturated polyester resin.     

Analysis of corrosion products beneath an epoxy-amine varnish film

In situ Raman analysis was applied to identify the corrosion products formed on iron coated with epoxy-amine varnish and exposed to sodium chloride solution saturated by hydrogen sulfide or carbon dioxide. The results showed the formation of iron sulfide or iron carbonate layer while electrochemical impedance spectrum predicted the polymer coating exhibiting an almost perfect protecting barrier property. It is concluded therefore, that the corrosion process involves the permeation of water and hydrogen sulfide at molecular state through polymer interstices whereas no ionic conduction is allowed to take place in the coating film. The corrosion process induces the complete delamination of the exposed coating area but as far as its integrity is respected (no macroscopic pores or crevices) the corrosion rate is very low, far from that expected from permeation measurements of H₂S.