The shear test specimen consisted of an Al 6082 rectangular plate (240 mm × 160 mm × 3 mm). In order to increase the roughness of the original aluminum plate (with an average surface roughness Sa of about 0.5 μm), the bonding surface (a centered rectangular portion of 160 mm × 95 mm) was shot peened with granular corundum of different grain diameters, ranging from 0.1 to 0.6 mm on average. Among the technologies available for modifying the sheet metal finishing shot peening was selected because it is suitable for complex geometries at low cost, it can be used for treating selected fractions of the entire surface and it promotes the part resistance to fatigue stress.
A special injection mold was designed to create shear lap joint specimens (Fig. 1a). A centre-gated polymer cavity with a thickness of 3 mm was used to facilitate mold filling and to contrast polymer wrinkling under shear test conditions. The injection molding was carried out on an all-electric 1000 kN Engel e-motion 440/100 injection molding machine with a 40 mm diameter screw. A polypropylene (PP) produced by Ticona under the trade name of Celstran® and reinforced with long strand glass fiber was used. The limits of the ranges of several process parameters were explored in preliminary experiments. Aluminum plates without preheating were also used for creating lap joints. However, the bond broke upon ejection from the mold. As a result of these preliminary experiments, the following two parameters were selected for a two-level full factorial study: (1) aluminum surface roughness, with an Sa of about 2 and 6 μm and (2) glass fiber content, varying from 30 to 50%. Furthermore, each of the 4 treatments was replicated 3 times. The aluminum plates were heated in an oven prior to insertion into the mold. The oven temperature was set to 260 °C. The aluminum samples remained in the oven for a period of 15 min to allow the entire specimen to achieve thermal equilibrium with the oven. After preheating, the aluminum plates were then inserted into the mold and held at a constant temperature of 80 °C. The adherend surface temperature was recorded using a Flir Systems model ThermoVision A40 infrared camera (Fig. 1b). The adherend surface temperature at the time of mold filling (5 s after the mold began to close) was extrapolated from recorded data.