Epoxy resins (DGEBA): The curing and physical aging process

Differential scanning calorimetry (DSC) and infrared spectroscopy (IR) were used to monitor the degree of cure of partially cured epoxy resin (Epon 828/MDA) samples. The extent of cure, as determined by residual heat of reaction, concurred with that determined by monitoring the infrared radiation absorbance of the epoxide group near 916 cm^?l. The fictive temperature T_{f, g} was found to increase with the degree of cure, increasing rapidly during cure until reaching a value near the cure temperature T_c of 130°C (approximately 80% cure) where the material vitrified. The greatly reduced reaction rate during the final 20% of cure was not only a consequence of vitrification but, as revealed by infrared spectroscopy, the result of the depletion in the number of reactive epoxide groups. The endothermic peak areas and peak temperatures evident during the DSC scans were used as a measure of the extent of “physical aging” which took place during the cure of this resin, and after, fully cured samples were aged 37°C below their ultimate glass temperature for various periods of time. The rate of physical aging slowed as the temperature increment (T_t,g ? T_c) increased. Although an endothermic peak was evident after only 1 h of cure (T_{f, g} = 138.3°C), such a peak did not appear until fully cured samples were aged for 16 h or more. Enthalpy data revealed that for partially cured material, the fictive temperature T_{f, a}, reflecting physical aging, increased with curing time. In contrast, the T_{f, a}, for fully cured samples decreased with sub-T_g aging time. The characteristic jump in the heat capacity ΔC_p which occurred at the T_{f, g} decreased as curing progressed. This decrease appears to be dependent upon the rotational and vibrational degrees of freedom of the glass. Finally, a graphical method of determining the fictive temperature T_{f, a}, of partially and fully cured epoxy material from measured endothermic peak areas was developed.     

A study of the kinetics of the microwave cure of a phenylethynyl‐terminated imide model compound and imide oligomer (PETI‐5)

The kinetic mechanism of the microwave cure of a simple phenylethynyl‐terminated imide model compound, 3,4′‐bis[(4‐phenylethynyl)phthalimido]diphenyl ether (PEPA‐3,4′‐ODA) and a phenylethynyl‐terminated imide oligomer (PETI‐5, M_n 5000 g/mol) was studied. Dielectric properties of the model compound and PETI‐5 were measured in the microwave range from 0.4 GHz to 3 GHz. FTIR was used to follow the cure of the model compound (PEPA‐3,4′‐ODA), while thermal analysis (DSC) was used to follow the cure of the PETI‐5 oligomer. The changes in room temperature IR absorbance of phenylethynyl triple bonds at 2214 cm⁻¹ of PEPA‐3,4′‐ODA as a function of cure time were measured after cure temperatures of 300, 310, 320, and 330 °C. The changes in the glass‐transition temperature, T_g, of PETI‐5 as a function of cure time were measured after cure at 350, 360, 370, and 380 °C, respectively. The T_g 's were determined to calculated the relative extent of cure, x, of the PETI‐5 oligomer according to the DiBenedetto equation. For the model compound, the reaction followed first order kinetics, yielding an activation energy of 27.6 kcal/mol as determined by infrared spectroscopy. For PETI‐5, the reaction followed 1.5th order, yielding an activation energy of 17.1 kcal/mol for the whole cure reaction, as determined by T_g using the DiBenedetto method. © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 2526–2535, 2000     

Cure kinetics of thermosetting bisphenol E cyanate ester

Resin injection repair is a common method to repair delamination damage in polymer matrix composites (PMCs). To repair high-temperature PMCs, the resin should have a very low viscosity, yet cure into a compatible adhesive with high temperature stability. Normally, thermosetting polymers with high glass transition temperatures (T _g) are made from monomers with high room temperature viscosities. Among the high temperature resins, bisphenol E cyanate ester (BECy, 1,1’-bis(4-cyanatophenyl)ethane), is unique because it has an extremely low viscosity of 0.09–0.12 Pa s at room temperature yet polymerizes as a cross-linked thermoset with a high T _g of 274°C. BECy monomer is cured via a trimerization reaction, without volatile products, to form the high T _g amorphous network. In this study, the cure kinetics of BECy is investigated by differential scanning calorimetry (DSC). Both dynamic and isothermal experiments were carried out to obtain the kinetic parameters. An autocatalytic model was successfully used to model isothermal curing. The activation energy from the autocatalytic model is 60.3 kJ mol⁻¹ and the total reaction order is about 2.4. The empirical DiBenedetto equation was used to evaluate the relationship between T _g and conversion. The activation energy of BECy from the dynamic experiments is 66.7 kJ mol⁻¹ based on Kissinger’s method, while isoconversional analysis shows the activation energy changes as the reaction progresses.     

Effects of water on epoxy cure kinetics and glass transition temperature utilizing molecular dynamics simulations

Effects of water on epoxy cure kinetics are investigated. Experimental tests show that absorbed water in an uncured bisphenol‐F/diethyl‐toluene‐diamine epoxy system causes an increase in cure rate at low degrees of cure and a decrease in cure rate at high degrees of cure. Molecular simulations of the same epoxy system indicate that the initial increase in cure rate is due to an increase in molecular self‐diffusion of the epoxy molecules in the presence of water. Effects of water on the glass transition temperature (T_g) of the crosslinked thermoset are also studied. Both experiments and simulations show that water decreases T_g. Both types of results indicate that T_g effects are small below 1% water by weight, but that T_g depression occurs much quickly with increasing water content above 1%. © 2017 Wiley Periodicals, Inc. J. Polym. Sci., Part B: Polym. Phys. 2017 , 55, 1150–1159     

Effect of cyclobutane carbonitrile rings on the glass transition temperature of vinyl copolymers

A wide compositional range of copolymers of 1-cyclobutene carbonitrile (CBCN) and 1-bicyclobutance carbonitrile (BBCN) with common vinyl monomers was prepared in order to study systematically the effect of cyclobutane carbonitrile ring incorporation on the glass transition temperature (T_g). In general, the T_g's of the copolymers are increased in proportion to the weight percent of the small-ring monomer incorporated, but the 1,2-mode of incorporation raises T_g more than the 1,3-mode. Attempts were made to correlate data with quantitative T_g–composition relationships (Fox–Johnston equations). Prediction for nitrile homopolymer T_g's are compared with literature values.     

The evolution of the viscoelastic retardation spectrum during the development of an epoxy resin network

The evolution of the viscoelastic behavior of an epoxy resin at various stages of curing has been followed with the changes in the retardation spectrum. The creep J(t) and recoverable creep compliance J_r(t) curves of the neat epoxy resin Epon l00lF (Shell) were determined at temperatures between 30 and 77°C. The viscosity decreased over 8 orders of magnitude as the temperature was increased. Specimens with eight stages of network development were prepared by reacting all of the epoxy resin's oxirane rings with amine hydrogens from varying ratios of a monofunctional amine (methyl aniline) and a tetrafunctional amine 4,4′-diamino diphenyl sulfone (DDS). Preparations in which 25, 35, and 40% DDS were used did not result in a molecular network, so they were viscoelastic liquids. With 45% DDS, the product had a nascent network and was judged to be just beyond the point of incipient gelation. The remaining preparations from 0.50, 0.60, 0.70, and 1.0 DDS yielded tighter less compliant molecular networks. The creep and recoverable compliance curves were measured over a range of temperatures above the glass transition temperature, T_g. They were reduced to T_g, and retardation spectra L(ln τ) were calculated.     

A study of the thermal cure of a phenylethynyl-terminated imide model compound and a phenylethynyl-terminated imide oligomer (PETI-5)

The kinetic mechanism of the thermal cure of a phenylethynyl-terminated imide model compound, 3,4′-bis[(4-phenylethynyl)phthalimido]diphenyl ether (PEPA-3,4′-ODA) and a phenylethynyl-terminated imide oligomer PETI-5 (MW 5000 g/mol) was studied. FTIR was used to follow the cure of the model compound, while thermal analyses (DSC) was used to follow the cure of the PETI-5 oligomer. The changes in IR absorbance of phenylethynyl triple bonds at 2214 cm⁻¹ of PEPA-3,4′-ODA as a function of cure time were detected at 318, 336, 355, and 373°C, respectively. The changes in the glass transition temperature, T_g, of PETI-5 as a function of time were measured at 350, 360, 370, 380, and 390°C, respectively. The DiBenedetto equation was applied to define the relative extent of cure, x, of the PETI-5 oligomer by T_g. For the model compound, the reaction followed first order kinetics, yielding an activation energy of 40.7 kcal/mol as determined by infrared spectroscopy. For PETI-5, the reaction followed 1.5th order, yielding an activation energy of 33.8 kcal/mol for the whole cure reaction, as determined by T_g using the DiBenedetto method. However, the cure process of PETI-5 just below 90% by this method followed first-order kinetics yielding an activation energy of 37.2 kcal/mol. © 1998 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 36 : 461–470, 1998     

Acceleration of the cationic polymerization of an epoxy with hexanediol

Thin films of 3,4-epoxycyclohexylmethyl 3',4'-epoxycyclohexane carboxylate were UV irradiated (1.1 J cm^-2) under isothermal conditions ranging from 0 to 50°C. Under these conditions the polymerization advanced quickly but only to a conversion level of less than 10% before the reaction rate slowed by more than an order of magnitude. This drop off in rate was not caused by the glass transition temperature, T _g, reaching or exceeding the reaction temperature, T _rxn, since the epoxide's T _g remained at least 40°C below T _rxn. Raising the sample temperature above 60°C caused a sharp increase in the conversion level. At 100°C conversion exceeds 80% and the ultimate T _g approaches 190°C. The addition of 10 mass% 1,6-hexanediol, HD, to the epoxy caused the conversion at room temperature to quintuple over the level obtained without the alcohol present. The heat liberated from this alcohol epoxy blend during cure on a UV conveyor belt system caused the sample's temperature to increase by about 100°C above ambient whereas the epoxy alone under these conditions only experienced a modest temperature rise of about 26°C. If the amount of HD in the blend is increased above 10% the heat of reaction at 23°C decreases due to HD being trapped in a nonreactive crystalline phase. Boosting reaction temperatures above 50°C melts the HD crystals and yields significantly improved conversion ratios. As the level of alcohol blended with the epoxy is raised its ultimate T _g is lowered and when the concentration of alcohol in the blend nears 30 mass%T _g drops below room temperature. This revised version was published online in July 2006 with corrections to the Cover Date.     

Glass transition temperature versus conversion relationships for thermosetting polymers

Owing to enthalpy relaxation, values of the glass transition temperature (T_g) for partially reacted polymers may depend on the thermal history of samples and the heating rate used for measurements. Use of theoretical relations between T_g and the extent of reaction (x) of a thermoset must take this fact into account. The original DiBenedetto equation has been reevaluated as a convenient constitutive equation for expressing T_g versus x. An extension of Couchman's approach for the expression of the compositional variation of T_g enabled us to derive the same functionality as given by the DiBenedetto equation. Thus, the DiBenedetto equation may be regarded as based on entropic considerations applied to a model of the thermosetting polymer consisting of a random mixture of a fully reacted network with the initial monomers in an amount which depends on the particular conversion level. These two equations have been applied with success to different diepoxy-diamine copolymers.     

Recovery of thick shear bands in atactic polystyrene

The DSC thermograms of the shear band material cut out by a diamond saw to include some undeformed material revealed two T_g's clearly separated by about 10°C. The first T_g was at the same temperature as the T_g of the undeformed material. The second T_g, which was at a higher temperature than the first T_g, appeared shortly after the shear strain recovery during the heating of the shear band material in the DSC. When the shear strain in the shear band was partially reversed by mechanical means before taking the DSC thermogram, the ΔT between the two T_g's decreased and when the shear strain was mechanically reversed to almost zero, the second T_g disappeared. The stored energy of shear band material was found to be similar to that of the bulk compressed material for large strains. Dimensional recovery during heating of specimens with thick and fine bands was similar, both taking place above T_g.     

Isothermal cure characterization of dicyclopentadiene

Conversion (α) and the glass transition temperature (T _g) were investigated during the isothermal cure of endo-dicyclopentadiene (DCPD) with a Grubbs catalyst for different temperatures using differential scanning calorimetry. Conversion vs. In (time) data at an arbitrary reference temperature were superposed by horizontal shift and the shift factors were used to calculate an Arrhenius activation energy. Glass transition temperature vs. conversion data fell on a single curve independent of cure temperature, implying that reaction of the norbornene and cyclopentene ring of DCPD proceeds in a sequential fashion. Implications of the isothermal reaction kinetics for self-healing composites are discussed.     

Thermal properties of polyurethanes derived from molasses before and after biodegradation

Polyurethane (PU) foams derived from molasses were placed in soil for various periods from 3 to 12 months. Thermal properties of PU's before and after biodegradation were investigated by differential scanning calorimetry (DSC) and thermogravimetry (TG). Glass transition of PU's after 3 month's degradation was separated into two stages indicating that molecular chains of the original and decomposed portions move independently. Based on variation of glass transition temperature (T_g), heat capacity difference at T_g (δC_p), thermal degradation temperature and mass loss, the degradation mechanism of PU was established.     

Monitoring the Cross-Linking of Epoxide Resins by Thermoanalytical Techniques

The curing reaction of bisphenol A diglycidylether with 4,4′-diaminodiphenylmethane (DDM) was studied by thermoanalytical methods. The overall reaction was monitored through the exothermic heat of reaction by differential scanning calorimetry (DSC), and a method is developed for predicting isothermal conversion-time curves over a wide temperature range from the results of two dynamic DSC scans. The reaction mechanism is not specified but it is assumed not to change with conversion, and the rate is assumed to be controlled by a single rate constant of the Arrhenius form. A series of fully cured resins prepared with varying DDM concentration is characterized by penetrometer, thermal expansion, and DSC methods. The T_g's of these resins are compared with those obtained using the stoichiometric quantity of DDM and reacted to different calorimetric degrees of cure. The T_g of the resin increases by about 70[ddot];C in the final 10% of the curing reaction where ΔH measurements are least sensitive, so that the final stages of cure are best monitored by T_g measurements.     

Sub‐glass‐transition temperature interface formation between an immiscible glass rubber pair

We used neutron reflectivity to measure the interfacial width in the immiscible system polystyrene/poly(n‐butyl methacrylate) (PS/PnBMA). Measurements were made on the same samples at temperatures ranging from below the glass‐transition temperature (T_g) of PS to slightly above. We observed significant broadening of the interface at temperatures below the T_g of PS, indicating chain mobility below the bulk T_g value. The interfacial width exhibited a plateau at a value of 20 Å in the temperature range of 365 K < T < 377 K. A control experiment involving hydrogenated and deuterated PS films (hPS/dPS) showed no such broadening over the same temperature region. The results are consistent with a reduction of the T_g of PS in the interfacial region of ～20 K. © 2001 John Wiley & Sons, Inc. J Polym Sci Part B: Polym Phys 39: 2664–2670, 2001     

Depression of the glass transition temperature of poly(methyl methacrylate) by plasticizers: Conformity with free volume theory

Mixtures of poly(methyl methacrylate) (PMMA) and plasticizer were made by γ-irradiation of mixtures of methyl methacrylate with diethyl phthalate or with dioctyl phthalate. The glass transition temperature T_g was determined by differential scanning calorimetry. Plots of T_g versus volume fraction of PMMA were found to conform to the simple free volume theory of Kelley and Bueche, using physical property values either reported in the literature or else calculated by Bondi's procedures.     

Pseudopoly(amino acid)s: Copolymerization of trans‐4‐hydroxy‐L‐proline and α‐hydroxy acids

A series of novel copolymers of trans‐4‐hydroxy‐L ‐proline (Hpr) and α‐ hydroxy acids [D,L ‐mandelic acid (DLMA) and D,L ‐lactic acid (DLLA)] were synthesized via direct melt copolymerization with stannous octoate as a catalyst. These new copolymers had pendant amine functional groups along the polymer backbone chain. The optimal reaction conditions for the synthesis of the copolymers were obtained with 4 wt % stannous octoate at 140 °C under vacuum for 16 h. The synthesized copolymers were characterized by IR spectrophotometry, proton nuclear magnetic resonance, differential scanning calorimetry, and Ubbelohde viscometry. The effects of the kinds of comonomers and the comonomer molar ratio on the polycondensation and glass‐transition temperature (T_g) were investigated. The T_g's of the copolymers shifted to lower temperatures with an increasing comonomer molar ratio. As expected, the T_g's of the N‐Z‐Hpr/DLMA copolymers were higher than the N‐Z‐Hpr/DLLA copolymers, the pendant groups on the monomers (N‐Z‐Hpr) became larger and more flexible, and the T_g's of the resulting polymers declined. © 2001 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 39: 724–731, 2001     

Influence of carboxyl‐terminated (butadiene‐co‐acrylonitrile) loading on the mechanical and thermal properties of cured epoxy blends

A mixture of epoxy with liquid nitrile rubber, carboxyl‐terminated (butadiene‐co‐acrylonitrile) (CTBN) was cured under various temperatures. The cured resin was a two‐phase system, where spherical rubber domains were dispersed in the matrix of epoxy. The morphology development during cure was investigated by scanning electron microscope (SEM). There was slight reduction in the glass transition temperature of the epoxy matrix (T_g) on the addition of CTBN. It was observed that, for a particular CTBN content, T_g was found to be unaffected by the cure temperature. Bimodal distribution of particles was noted by SEM analysis. The increase in the size of rubber domains with CTBN content is due probably to the coalescence of the rubber particles. The mechanical properties of the cured resin were thoroughly investigated. Although there was a slight reduction in tensile strength and young's modulus, appreciable improvements in impact strength, fracture energy, and fracture toughness were observed. Addition of nitrile rubber above 20 parts per hundred parts of resin (phr) made the epoxy network more flexible. The volume fraction of dispersed rubbery phase and interfacial area were increased with the addition of more CTBN. A two‐phase morphology was further established by dynamic mechanical analysis (DMA). © 2004 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 42: 2531–2544, 2004     

Curing of epoxidized natural rubber with p-phenylenediamine

The curing of epoxidized natural rubber with p-phenylenediamine catalyzed by bisphenol A was investigated. The curing reaction was found to be first order with respect to the amine and to have an activation energy of 67.5 kJ/mol. Vulcanizates of tensile strength 12.5–16.0 MPa were obtained at cure times of 30–75 min at 180°C. The amine-cured vulcanizate was observed to be more rigid than the sulfur-cured one at room temperature consistent with its relatively high T_g. Stress-strain analysis and dynamic mechanical properties indicated that the high T_g of the amine-cured vulcanizate could be attributed to the bulky amine crosslinks and hydrogen bonding effect of bisphenol A. © 1994 John Wiley & Sons, Inc.     

Synthesis and thermal behavior of phenylethynyl‐terminated linear and hyperbranched polyphenylquinoxalines

A phenylquinoxaline (PQ) AB monomer mixture was treated with monofunctional and difunctional end‐capping agents and with and without a coupling agent to afford phenylethynyl‐terminated linear PQ oligomers. The resulting PQ oligomers were soluble in common organic solvents and had intrinsic viscosities (IVs) of 0.21–0.30 dL/g. The glass‐transition temperature (T_g) of the diphenylethynyl‐end‐capped PQ oligomer on both sides increased the most, from 215 °C (before curing) to 251 °C (after curing). The PQ AB₂ monomer, which acted as both a coupling agent and a monomer for the hyperbranched polymer, was treated with an AB monomer and end‐capping agents to afford phenylethynyl‐terminated hyperbranched polyphenylquinoxalines (PPQs). They were also soluble in common organic solvents, had IVs of 1.00–1.65 dL/g and T_g's of 251–253 °C, and underwent exothermic cure with maxima around 412–442 °C. The T_g's of the cured hyperbranched PPQs ranged from 258 to 261 °C, depending on the number of phenylethynyl groups on the surface. After further curing, they displayed a T_g of 316 °C in one sample and turned into a fully crosslinked network. The dynamic melt viscosities of a linear oligomer (IV = 0.21 dL/g), a hyperbranched sample (IV = 1.00 dL/g), and a linear reference PPQ (IV = 1.29 dL/g) were compared with respect to the processing temperature. The PQ oligomer and hyperbranched PPQ had low melt viscosities. © 2004 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 42: 6318–6330, 2004     

Effect of thermoplastic toughening agent on glass transition temperature and cure kinetics of an epoxy prepreg

The isothermal time–temperature-transformation (TTT) cure diagram is developed in this article to investigate the effect of thermoplastic toughening agent on glass transition temperature (T _g) and cure kinetics of an epoxy carbon fiber prepreg, Cycom 977-2 unidirectional (UD) tape. The glass transition temperature was measured using differential scanning calorimetry (DSC) over a wide range of isothermal cure temperatures from 140 to 200 °C. Times to gelation and vitrification were measured using shear rheometry. The glass transition temperature master curve was obtained from the experimental data and the corresponding shift factors were used to calculate the activation energy. The kinetic rate model was utilized to construct iso-T _g contours using the calculated activation energy. It was observed that the iso-T _g contours did not follow the behavior of the neat epoxy resin, since they deviated from the gel time curve. This deviation was believed to be the effect of the thermoplastic toughening agent. The behavior of the neat epoxy resin in 977-2 was shown by constructing the iso-T _g contours using the activation energy obtained from gel time modeling.