JOURNAL OF MATERIALS SCIENCE 39 (2 0 0 4 ) 1503 – 1506 Effect of adhesive on the strengthening of aluminum foam-ﬁlled circular tubes A. K. TOKSOY Department of Mechanical Engineering, Izmir Institute of Technology, G¨ ulbah ¸ ce K ¨ oy ¨ u, Urla, Izmir, Turkey M. TANOGLU, M. GUDEN Department of Mechanical Engineering and Center for Materials Research, Izmir Institute of Technology, G¨ ulbah ¸ ce K ¨ oy ¨ u, Urla, Izmir, Turkey E-mail: mustafaguden@iyte.edu.tr I. W. HALL Department of Mechanical Engineering, University of Delaware, Newark, DE, USA Studies of the crushing behavior of closed-cell, alu- minum foam-ﬁlled aluminum and steel tubes have shown an interaction effect between tube wall and foam ﬁller [1, 2, 3]. The crushing loads of foam-ﬁlled tubes are, therefore, found to be higher than the sum of the crushing loads of foam (alone) and tube (alone) mainly due to this effect. Santosa et al. [1], based on FEM re- sults, proposed the following equation for the average crushing load of foam-ﬁlled square tubes of length b, P f = P e + C σ f b 2 (1) where P f , P e and σ f are the average crushing loads of the ﬁlled and empty tubes and plateau stress of the ﬁller, respectively. The constant C in Equation 1 is consid- ered to be the strengthening coefﬁcient of the foam ﬁll- ing. The values of C were proposed and experimentally shown to be 1.8 and 2.8 for foam-ﬁlled tubes without and with (epoxy) adhesive, respectively [1]. The study of Santosa and Wierzbicki has also shown that the use of adhesive, although resulting in a relatively small in- crease in the total weight of the tube, <16%, raised the crushing load of the tube by as much as the foam crush- ing load. There has, however, been only this one study on the use and effect of adhesive in foam-ﬁlled tubes and the effect of adhesive in circular tubes has not been investigated yet. The present report is a further investi- gation of the strengthening effect of foam ﬁlling with a bonding layer in circular tubes. The drawn aluminum tubes studied (3003-H14) were 15.88 mm in diameter with a wall thickness of 0.9 mm. 20 mm long empty tubes for compression testing were cut using a slow speed diamond saw. The foam core samples, with a diameter of 14 mm, were prepared by core-drilling. The inner diameter of the tube was almost the same as the diameter of the foam core so that foam samples ﬁtted tightly inside the tubes. The average den- sity of the foam varied between 0.2–0.5 g . cm −3 . The weight and dimensions of the tubes and foams were measured before and after ﬁlling in order to calculate density of the foam and weight of the polyester layer for each individual sample. The empty tube was ﬁlled with polyester resin-curing agent mixture and the foam sample was then inserted inside the tube. Finally, the excessive bonding material was removed, then foam and bonding material were cured at room temperature. The thickness of the polyester layer was predicted to be about 0.1 mm and its weight was about 5% of the weight of the tube. Quasi-static compression tests on empty and ﬁlled tubes and foam samples were conducted using a Testometric Test Machine with a displacement rate of 0.01 mm s −1 . The plateau stress as a function of density for the foams is shown in Fig. 1. The plateau stress (σ pl ) was found to be well-ﬁtted by the power-law of strengthen- ing equation, σ pl = Kρ n (2) where K and n are constants and ρ is the foam density in g . cm −3 . The values of K and n are ∼22.4 (MPa) and ∼1.99, respectively. The foam-crushing load of the 14 mm diameter foam ﬁlling of ﬁlled tube samples was calculated using the above equation by simply inserting the values of the ﬁller density and the cross-sectional area. The deformation modes of the empty and ﬁlled tubes were progressive and axisymmetric (concertina) and typical compression load-displacement curves of the empty and ﬁlled tubes, with various ﬁller densities, are shown in Fig. 2. Compressed empty tubes folded into 3-lobes and each lobe had a length of ∼5 mm. Filled tubes were only compressed to the completion of the second fold and, because of the limited number of folds formed, the effect of foam ﬁlling on the fold length was not precisely determined. Corresponding average crushing loads ( P a ) of the tested tubes were calculated using the following relation: P a =  Pd δ δ (3) where P and δ are the load and displacement, respec- tively. The average load-displacement curves of the empty and foam-ﬁlled tubes are shown in Fig. 3 for 0022–2461 C  2004 Kluwer Academic Publishers 1503