Static and dynamic fracture toughness of epoxy/alumina composite with submicron inclusions

Abstract

The influence of the particle size and the volume fraction of submicron spherical alumina particles in an epoxy matrix on the fracture toughness of the composite is investigated experimentally. Three particle sizes, 50 nm, 500 nm and 5 μm are used. The static and the dynamic fracture toughnesses are evaluated from the 3-point bend tests. It is found that the particle size significantly affects the static and the dynamic fracture toughness. Particulate composites with either nanosize particles or carbon nanotubes as reinforcements are getting increased attention because of the expectation of orders of magnitude improvement [1] in their strength and elastic moduli over conventional reinforcements. Niihara and coworkers [2, 3] have reported a dramatic increase in the mechanical performance of ceramic matrices containing ceramic nanoparticles. However, Todd et al. [4] and Cambier et al. [5] measured a modest increase in the fracture toughness of a nanocomposite over that of alumina, and Sternitzke et al. [6] found that the fracture toughness of a nanocomposite decreased. Warren et al.’s [7] work suggests that the deflection of the crack path is a key contributor to the toughening mechanisms in alumina composites. There appears to be no clear consensus on the effect of particle size, volume fraction of reinforcements and mechanisms influencing the toughness of nanocomposites. For polymer matrix composites, a general trend is that the fracture toughness increases with an increase in the volume fraction of reinforcing particles before reaching a plateau at around 50% volume fraction [8]. Mechanisms contributing to the enhancement in the strength of particulate composites are the change in the fracture mode to transcrystalline, flaw size reduction due to smaller grains, crack deflection at inclusions, and crack bridging. Several investigations [2–4, 9] support the change in the fracture mode from intergranular to transgranular. For nanosize particles, the fracture may not be transgranular. For instance, Birringer et al. [10] found that nominally brittle materials become ductile at room temperature when the grain size is about 100 nm. Ash et al. [11] have reported a 400% increase in the ductility of 5 wt% nanophase alumina filled epoxy over that of pure epoxy. The particulate composites were prepared by dispersing solid alumina particles in a slow curing epoxy EPOTHIN supplied by Buehler. Spherical alumina particles of mean diameters 50 nm (AKP-G008, Sumitomo Chemical Co., Japan), 500 nm (AKP-3000, Sumitomo Chemical Co., Japan) and 5000 nm (Buehler) were used as inclusions. Values of Young’s modulus E, Poisson’s ratio ν and mass density ρ for the epoxy (alumina) are 2.85 GPa (400 GPa), 0.36 (0.22) and 1149 kg/m3 (3990 kg/m3) respectively. The epoxy is prepared by mixing 73.5 wt% of epoxy resin and 26.5% of hardener. The mixture is gently stirred for 10 min at room temperature and allowed to degas for 10 min. Then an appropriate quantity of alumina particles is added to the resin/hardener mixture which is again stirred slowly till particles have been uniformly mixed with the resin and air bubbles have escaped. The mixture is then poured carefully into a mold made of PMMA plates held together by four bolts. The edges of the mold are first lined with a cellophane tape and then coated twice with a polyurethane release agent to facilitate the separation of the specimen from the mold after curing at room temperature for 24 h. Thin sheets of the particulate composite are allowed to stabilize for at least seven days before they are cut and machined into rectangular test specimens. The initial plan was to fabricate composites with nominal volume fractions of 2%, 5%, 10%, 20% and 40% alumina particles. However, good quality specimens with a large volume fraction of smaller size alumina particles could not be prepared. Therefore, the maximum volume fractions of 500 nm and 50 nm alumina particles were limited to 30% and 10% respectively. For each particle size and each volume fraction of the inclusions, five specimens were machined with the span (S)/width (W ) and the crack-length (a)/width equal to 4 and 0.25 respectively. Edge notches were machined with a Buehler diamond wafer blade resulting in a nominal crack tip radius of 75 μm. For non-zero radius of the notch-root, the dependence of fracture toughness on the notch-root radius should be considered. However, no such relations are available for particulate composites with submicron size particles. The static 3-point bend tests were performed in an Instron testing machine at a cross-head speed of 0.2 mm/min, and the load-displacement data was recorded. Fracture is assumed to initiate at the peak load Pc, and the fracture toughness, KIc, is computed