INVESTIGATING THE INFLUENCE OF MICROENCAPSULATED HEALING AGENTS ON THE PROPERTIES OF EPOXY THERMOSETS

Abstract

The primary motivation behind the present research work is to study the effect of inclusion of healant loaded microcapsules on specific properties (thermal, structural and mechanical) of a representative epoxy thermoset. In addition, we also explore alternate chemistry for introducing self healing functionality in epoxy composites. Two distinct healing systems have been investigated, namely cycloaliphatic epoxy and unsaturated polyester. The healants were encapsulated in urea-formaldehyde microcapsules by adopting an in-situ dispersion polymerization route and in polystyrene shell through solvent evaporation process. The effect of operating parameters particularly stirring speed on the particle size distribution has been studied. Under optimal conditions, the core content of the epoxy loaded microcapsuleswas found to be 65 ± 4%for microcapsules prepared by dispersion polymerization route and 38 ± 2%for microcapsules prepared using solvent evaporation route. It is to be noted that the healing efficiency is strongly influenced by the internal microstructure of the microcapsule and we also developed an analytical model for predicting the amount of healant released in the event of microcapsule rupture. In microcapsules possessing “reservoir” type microstructure, the healant exists as a single droplet, where the entire content is expected to be released upon rupture. On the other hand, in monolithic microcapsules, the healant is dispersed in the form of discrete micro-droplets, and only the healant available within the cracked microcapsule is expected to be released and cause healing. Our model predicted that significantly lower amounts of healant is released in monolithic microspheres in comparison to reservoir microcapsules, especially when the micro-droplet dimensions and core content, both are low. V

Triethylenetetramine (TETA) hardener was encapsulated by adopting two methods, namely interfacial polymerization and physical entrapment technique. The effects of experimental parameters, namely reaction temperature (50-75°C), stirring speed (400-700 rpm) and epoxy: amine concentration ratio (10:1.2-10:4.3) on the microcapsule properties was investigated. A polymeric surfactant was used to stabilize the suspension in order to modulate the particle size distribution of the resultant microcapsules. Highest encapsulation efficiencies resulted when the reaction medium was maintained at 70°C under a stirring speed of 600 rpm, while maintaining an epoxy: amine ratio of 10:3.2. The microcapsule dimensions and core content could be tailored, following the interfacial polymerisation route. Under optimal conditions, spherical microcapsules with 100 % yield and 12% core content were obtained. Physical entrapment approach was also explored for the immobilisation of amine hardener in mesoporous silica. For this purpose, mesoporous silica (SBA-15) was synthesised using polymer-templated technique and employed as a substrate for immobilization. Vacuum infiltration of TETA led to its entrapment within the porous structure of SBA-15 with loadings as high as 5g/g, which could be attributed to hydrogen bonding and acid–base interactions. The curing kinetics of self-healing epoxy compositions was investigated by non-isothermal differential scanning calorimetric (DSC) studies. Epoxy loaded microcapsules and immobilised amine was dispersed into epoxy resin, and cured using TETA. DSC studies revealed the autocatalytic nature of epoxy curing, which remained unaltered due to addition of the fillers responsible for introducing self healing functionality. The kinetic parameters of the curing process were determined using both Friedman and Kissinger–Akahira–Sunose (KAS) method. The activation VI

energy at different degree of conversion (E  ) was found to decrease with increasing degree of cure (  ). Although urea-formaldehyde possess secondary amine functionalities, which have the potential to react with the epoxy groups, no significant differences in the curing kinetics of the base resin were observed.Kinetic parameters were used to predict the curing behaviour of compositions at higher heating rates using KAS method. As expected, the onset curing temperature (Tonset) and peak exotherm temperature (Tpeak) of epoxy shifted towards higher temperatures with increased heating rate; however introduction of fillers do not affect these characteristic temperatures significantly. Also, the overall order of reaction does not vary significantly. The results suggest that although 2° amino groups are available with the urea-formaldehyde (UF) resin, these do not directly participate in the curing reaction, as the primary amino groups in TETA are more easily accessible. To evaluate the effect of self healing additives on the mechanical properties and healing efficiency, epoxy composites containing UF and PS microcapsules (5 30%, w/w) were prepared by room temperature curing and their mechanical behaviour and healing efficiency was studied under both quasi-static and dynamic loadings. The tensile strength, modulus and impact resistance of the matrix was found to decrease with increasing amount of microcapsule in the formulation, the detrimental effect being less pronounced for polystyrene microcapsules due to its monolithic internal microstructure. Morphological investigations on the cracked surface revealed features like crack pinning, crack bowing, microcracking and crack path deflection, which were used to explain the toughened nature of microcapsule containing epoxy composites. VII

Healing efficiency was quantified in terms of the ratio of impact strength before and after healing. For the purpose of validation of the developed analytical model, composites were prepared using epoxy encapsulated microcapsules with varied internal structures. Ni and Cu-imidazole complexes were prepared for use as latent hardeners for epoxy. Both the imidazole-metal complexes could effectively cure the epoxy released from within the microcapsules in the event of damage followed by thermal treatment. In line with our predictions, the extent of healing was much lower in the case of samples containing monolithic microcapsules. At 20% w/w microcapsule loadings, healing efficiencies close to 60% was observed upon introduction of reservoir type microcapsules, while under similar loadings, only 10% healing could be evidenced in the presence of monolithic microcapsules. For reservoir type microcapsules, complete healing (efficiency ~ 100±2%) could be effected at 30% microcapsule loading, in the presence of metal imidazole complexes. In comparison, the complete healing was evidenced at relatively lower microcapsule loading (20%, w/w) when amine immobilised SBA-15 was used. The potential of encapsulated unsaturated polyester resin (USP) towards introduction of healing functionality was also explored. USP was encapsulated in urea-formaldehyde shell and polystyrene shell by dispersion polymerization and solvent evaporation technique respectively both resulting in the formation of free flowing microcapsules. Calorimetric studies confirmed the chemical activity of the encapsulated USP, which spontaneously polymerised in the presence of a free radical initiator, 2,2‟-Azobis(2-methylpropionitrile) (AIBN), at temperature as low as 80°C. Temperature triggered healing of epoxy-microcapsule composites was performed at 110° C and healing efficiency was quantified as the ratio of impact strength of healed and virgin specimens. The same was found to increase with increasing amount of VIII

microcapsule in the formulation and reached a maximum (100 ± 2%) at 20% (w/w) loading. Fractographic analysis of the surface revealed the flow pattern of chemically active polyester resin from the ruptured microcapsules, which subsequently cured in the presence of AIBN available within the matrix.