Comparison of surface strain for stamp formed aluminum and an aluminum-polypropylene laminate

Laminate structures incorporating thin layers of metal and polymer, or polymer composite, can offer significant weight savings for engineering structures, while retaining excellent mechanical and impact performance. Laminates based on thin layers of aluminum and glassfiber/polypropylene thermoplastic have been the subject of recent study [1, 2], and have exhibited excellent specific mechanical properties and superior specific impact behavior compared to monolithic aluminum. Such materials, therefore, have great potential for widespread application in engineering structures. One such potential area is the automotive industry where weight reduction and impact performance are pertinent issues. Lighter vehicles will result in improved fuel efficiency, and greater energy absorption capability may contribute to improved crash performance. However, for the automotive industry it is necessary to produce components using a high-volume manufacturing process such as stamping. Thermoplastic-based materials and sandwich structures are good candidates for stamp forming as they can be heated to conform to the mold, and then rapidly cooled for removal from the mold. Mosse et al. [3, 4] investigated the effects of blankholder force, laminate preheat temperature, tooling temperature, and tool radii on FML formability. It was found that significantly lower levels of springback could be achieved over aluminum, and forming defects could be eliminated by restricting process variables to a given range. In particular, it was found that delamination at the bimaterial interface and within the composite layer was eliminated when the laminate was pre-heated to 160 ◦C then formed in a heated die. This is significant as delamination would adversely affect the mechanical performance of a formed component. Further, Kim and Thomson [5] found that high forming speed increased the transverse stiffness of polymer-metal laminates, in turn reducing the inter-laminar shear and the degree of springback. They also found that laminates forming at elevated temperatures decreased the rigidity but improved the springback characteristics. This letter presents some preliminary results from research into stamp-forming aluminum-thermoplastic sandwich materials. Here, the permanent strain on the surface of a channel-formed aluminum-polypropylene laminate is compared to monolithic aluminum. Characterization of the strain is significant as it provides insight into the behavior of the material during formation and assists in the production of parameters for subsequent formation methodologies. The materials used in this study were 5005-H34 aluminum and a self-reinforced polypropylene (Curv, BP). An aluminum-Curv laminate was made in a 2/1 configuration in a 200 × 200 mm picture frame mold. A 0.9 mm thick layer of Curv was sandwiched between two layers of 0.5 mm thick aluminum cleaned with a solvent (isopropanol). A 50 μm thick layer of a hot-melt polypropylene adhesive (Gluco Ltd., UK) was placed at each bi-material interface. The laminate was consolidated by heating to 160 ◦C in a platen press followed by rapid water cooling under a pressure of approximately 1 MPa. The nominal laminate thickness was 2.2 mm. Samples of 19 mm width were sectioned from the laminate and from a plain sheet of 2 mm thick aluminum. A 3 mm circular grid etched onto the surfaces enabled post-forming major strain measurements, that is in the direction of the sample length, to be made. Channel sections were stamped in an open die. Plain aluminum was stamped cold whereas the aluminumCurv laminates were pre-heated to 160 ◦C then immediately transferred to the die, which was pre-heated to 80 ◦C. This enabled a temperature window of 125– 140 ◦C to be maintained during the stamping operation. The channel sections were stamped in an Enerpac 30 tonne press using two tool radii of 3 and 7 mm. The blank holder force was 3.5 kN. Surface strain measurements were taken from ten grids around the mid-point of the sidewall area of the channel section, shown in Fig. 1, using an optical microscope with a graticule scale of 20 μm resolution. Measurements were taken from the sidewall area as it is likely to undergo significant tensile strain during formation. Microscope examination of the sidewall edge, prior to taking the strain measurements, confirmed the absence of delamination. The average major surface strain for the aluminum and aluminum-Curv samples is plotted in Fig. 2. (The