C. C. Riccardi

Miscibility of epoxy monomers with carboxyl-terminated butadiene-acrylonitrile random copolymers

Abstract Cloud-point curves for mixtures comprising epoxy monomers of the DGEBA type (diglycidyl ether of bisphenol A), rubbers of the CTBN type (carboxyl-terminated acrylonitrile-butadiene random copolymers) and, eventually, hardeners such as diamines are reported. It is shown that the miscibility of epoxy monomers with a particular rubber is very sensitive to the molar mass of the epoxy molecule. The application of a simple Flory-Huggins lattice model, regarding both components as monodisperse, leads to the location of the critical point, coexistence curves and the estimation of the interaction parameter per unit volume (Λ). A correlation of the type Λ=Λ0 + ΛTT was found, with a negative ΛT. The decrease in miscibility observed when using a CTBN containing less acrylonitrile is explained from the change in the solubility parameter. Using CTBN as an adduct with the epoxy monomer leads to an increase in miscibility, which is explained by the copolymer effect. Systems based on epoxy-diamine copolymers prepared in the presence of CTBN adducts showed a complex behaviour including two maxima in cloud-point curves.

Thermodynamic analysis of the phase separation in polyetherimide-modified epoxies

The miscibility of polyetherimides (PEIs) with epoxy monomers based on diglycidylether of bisphenol-A (DGEBA), and with reactive mixtures based on stoichiometric amounts of DGEBA and an aromatic diamine (DA) {either 4,4′-diaminodiphenylsulfone (DDS) or 4,4′-methylenebis[3-chloro 2,6-diethylaniline] (MCDEA)}; was experimentally studied. Cloud-point curves (temperature vs. composition) are reported for PEI-DGEBA and PEI-DGEBA-DA initial mixtures. Cloud-point conversions are reported for the reactive mixtures, for various PEI amounts and polycondensation temperatures. A thermodynamic model based on the Flory-Huggins-Staverman approach, taking polydispersity of both components into account, was used to analyze the experimental information. A single relationship between the interaction parameter and temperature, χ(T), could fit experimental results of mixtures of two commercial PEIs with DGEBA. The addition of DDS led to a decrease in miscibility whereas MCDEA improved the initial miscibility. In both cases, the interaction parameter decreased with conversion, meaning that PEI was more compatible with oligomeric species than with the mixture of starting monomers. The phase separation process in initially miscible rubber- or thermoplastic-modified thermosetting polymers is the result of two factors: increase in the average molar size of the thermosetting oligomer (main driving force favoring demixing), and variation of the interaction parameter with conversion, which may act to increase or decrease the cloud-point conversion determined by the first factor.

Statistical structural model for the build-up of epoxy-amine networks with simultaneous etherification

Abstract A statistical structural model is developed to describe a diepoxy-diamine cure, taking into account the possibility of simultaneous epoxy-hydroxy reaction (etherification). Expressions for the number- and weight-average molecular weights, gel conversion, sol fraction, mass fraction of pendant and elastically active network chains (EANC) and concentration of EANC are derived. The different reactivity of primary and secondary amine hydrogens is taken into account, but intramolecular reactions in the pregel stage are neglected. The model is applied to the cure of bisphenol A diglycidyl ether with diaminodiphenyl sulphone, where a previous kinetic analysis showed the presence of etherification. It is shown that, for stoichiometric formulations, etherification acts to decrease both the gel conversion and the concentration of EANC at full epoxy conversion. However, the elastic modulus of the material is not expected to change significantly. Instead, for formulations containing a 100% epoxy excess, the predicted elastic modulus at full epoxy conversion is 50% higher than that predicted for a stoichiometric mixture.

Rubber-modified cyanate esters : thermodynamic analysis of phase separation

Abstract Cloud-point curves were determined for blends of a rubber (butadiene—acrylonitrile random copolymer terminated in non-functional groups, NFBN) and a cyanate ester (4,4′-dicyanate-1,1′-diphenylethane, CE). Measurements were made in the initial blend and in partially polycondensed blends (thermal polycyclotrimerization of the CE monomer). Cloud-point conversions and temperatures were determined for different rubber fractions in the initial formulation. A thermodynamic analysis based on a Flory-Huggins equation, taking polydispersity of both components into account, led to the following conclusions. (1) Phase separation was the result of the change in two different contributions to the free energy of mixing: (a) a decrease in the entropic contribution due to the increase in the oligomer size; (b) a decrease in the enthalpic contribution as a result of the decrease in the cohesive energy density of the CE-oligomer in the course of polycondensation. (2) Spinodal demixing was excluded as a possible mechanism of phase separation during polycondensation in solutions containing less than 12% rubber by volume.

Phase separation induced by a chain polymerization : Polysulfone-modified epoxy/anhydride systems

The reaction-induced phase separation in a blend of a commercial polysulfone (PSu) with diepoxide-cyclic anhydride monomers, was studied. The diepoxide was based on diglycidylether of bisphenol A (DGEBA) and the hardener was methyl tetrahydrophthalic anhydride (MTHPA), used in stoichiometric proportion. Benzyldimethylamine (BDMA) was used as initiator. PSu had no influence on the polymerization kinetics, the gel conversion, and the overall heat of reaction per epoxy equivalent. A kinetic model including initiation, propagation, and termination steps was used to estimate the distribution of linear and branched species in the first stages of the chain-wise copolymerization. This distribution, together with the PSu distribution, were taken into account in a thermodynamic model of the blend. The interaction parameter was fitted from experimental determinations of conversions at the start of phase separation, obtained under different conditions. The thermodynamic model was used to explain the complex morphologies developed in materials containing different PSu concentrations as well as their dynamic mechanical response. The shift in glass transition temperatures was explained by the fractionation of different species during the phase separation process. Phase inversion produced a significant decrease of the elastic modulus in the glassy state and a thermoplastic-like behavior of the material in the rubbery region.

Design of morphology in PMMA‐modified epoxy resins by control of curing conditions. I. Phase behavior

Rheokinetic and phase separation behavior of diglycidylether of bisphenol-A–4,4′-diaminodiphenyl methane epoxy mixtures, modified with a constant amount (15 wt %) of poly(methyl methacrylate) (PMMA), have been investigated. Stoichiometric epoxy/amine mixtures precured at 80°C several times presented various levels of miscibility. Differential scanning calorimetry (DSC) and dynamic mechanic thermal analysis were used for rheokinetic studies of curing and also for testing the thermal behavior of the fully cured mixtures. Phase separation, through curing, was simultaneously studied by transmission optical microscopy and DSC, showing an excellent correlation between the results obtained with both techniques.

Influence of curing conditions on the morphologies of a PMMA-modified epoxy matrix

The effect of curing conditions such as time and temperature on the morphology developed in a diglycidyl ether of bisphenol-A epoxy resin cured with diamino diphenyl methane, and modified with 15 wt% poly(methyl methacrylate), has been investigated. The reacting mixtures were precured at 80°C for a period of time ranging from 2 to 7 h, afterwards they were cured at 140°C and finally postcured at 200°C. The mixtures were opaque or transparent depending on the precuring time. Dynamic mechanical thermal analysis suggested that all mixtures were heterogeneous. However, phase separation occurred for all precuring times but to a lesser extent for samples precured for 5, 6 or 7 h than for those precured for shorter time intervals at 80°C. Two phases were clearly distinguished by atomic force microscopy, in all mixtures. The phase size was controlled, on one hand, by the time the phases had to grow, i.e. the interval between the cloud and gel points, and on the other hand, by the viscosity of the reacting mixture at the moment of phase separation.

Thermodynamic analysis of phase separation in rubber-modified thermosetting polymers: influence of the reactive polymer polydispersity

A thermodynamic simulation of the phase separation process in a modified thermosetting polymer was carried out. The polydispersity of the generated polymeric species was taken into account in the frame of a conventional Flory-Huggins equation. The example considered in the simulation was a diglycidyl ether of bisphenol-A (DGEBA)-ethylenediamine (EDA), epoxy-amine polymer, modified by the addition of 15 wt% castor oil (monodisperse modifier). The size increase of the oligomeric species and the corresponding decrease of the entropic contribution to the free energy of mixing made a modifier-rich phase (β-phase) segregate from the matrix (α-phase) at a particular conversion level. The β-phase is enriched in monomers and low-molecular-weight species of the polymer distribution. This produces a significant decrease of the β-phase conversion with respect to the overall conversion. The monomer with the smaller size and functionality is preferentially segregated into the β-phase, leading to a stoichiometric imbalance. When a semipermeable β-phase is assumed, i.e. no oligomeric species are allowed to transfer to the α-phase, a secondary phase separation inside the β-phase is generated. This leads to a sub-matrix (δ-phase) which is rich in modifier, and a sub-segregated phase (γ-phase) which is rich in thermosetting polymer. This process may continue well beyond the gelation of the α-phase, due to the low conversion level of the β-phase at the time the α-phase gels. The thermodynamic simulation explains some recent experimental observations in systems of commercial interest.

UV-MALDI-TOF and ESI-TOF mass spectrometry characterization of silsesquioxanes obtained by the hydrolytic condensation of (3-glycidoxypropyl)-trimethoxysilane in an epoxidized solvent

Silsesquioxanes obtained by the hydrolytic condensation of (3-glycidoxypropyl)trimethoxysilane (GPMS) in diglycidyl ether of bisphenol A (DGEBA) were characterized by electrospray ionization time-of-flight mass spectrometry (ESI-TOF MS) and matrix-assisted ultraviolet laser desorption/ionization time-of-flight mass spectrometry (UV-MALDI-TOF MS), employing two different matrices and both positive and negative ion modes. A bimodal distribution of molar masses, in the 1300-6400 m/z range, was observed in MALDI mass spectra. This distribution accounted for oligomers formed in two successive generations but did not include a cluster of higher molar-mass species present in SEC chromatograms. Most of the peaks present in ESI and MALDI mass spectra could be described by the generic formula T n -(OCH 3 ) m , with m = 0, 2, and 4 for n = even, m = 1, 3, 5 for n = odd, and T = RSiO (3n-m)/2n . This corresponds to completely condensed polydedra (m = 0), incompletely hydrolyzed polyhedra (m = 1 to 3), and their precursors (m = 4). Predominant species in the first cluster contained 10 to 14 Si atoms whereas those in the second cluster had 18 to 23 Si atoms. Small amounts of the following species: T 8 (OH)(OCH 3 ), T 9 (OH), T 10 (OH)(OCH 3 ), and T 11 (OH) could be identified in MALDI MS, using 9H-pyrido[3,4-b]indole (nor-harmane) as matrix in the negative ion mode. It was inferred that some of these species had a relevant participation in the generation of the second cluster of higher molar masses. The stability of the silsesquioxane solution in DGEBA was the result of the very small concentration of free SiOH groups available for further condensation.

Build-up of polymer networks by initiated polyreactions

SummaryTheoretical treatment of network formation with participation of initiated reactions is to be based on the kinetic (coagulation) theory, because the application of the statistical network build-up from monomer units (cascade substitution) can be a source of serious deviations. This comparison is demonstrated by the degree-of-polymerization distribution obtained in the linear living polymerization and the gel point conversion in the multifunctional polymerization involving a monomer with two groups of independent reactivity