Linear Friction Welding (LFW) of metal matrix composites

Metal matrix composites (MMCs) offer higher specific stiffness and strength, better wear resistance and greater thermal stability compared to the corresponding unreinforced alloys. One of the outstanding challenges in the use of aluminium based MMCs concerns their joining, since traditional fusion welding rocesses generally lead to microstructural defects and, consequently, to a general decrease in their mechanical properties. These problems are significantly reduced by the use of solid state joining techniques, such as Friction Stir Welding (FSW), that has recently been successfully applied to particles reinforced aluminium matrix composites. The aim of the present work is to evaluate the possibility of using the Linear Friction Welding (LFW) process to join an aluminium matrix composite reinforced with SiC particles. In this solid state process (Fig.1), the frictional heating, generated during the rubbing of metallic blocks, is used to join metal components. Under controlled conditions of pressure and relative frequency of motion, significant heat is generated at the components interface, thus causing plasticization of the rubbing layers. Once sufficient plasticization has occurred, the reciprocal motion is stopped and a forging force is applied to produce a consolidated joint. The material used in this study was the AMC225xe composite, consisting of the 2124 aluminium alloy matrix (Tab. I) reinforced with 25vol% of fine (≤3 μm) SiC particles. Billets were produced by Aerospace Metal Composites Ltd (Farnborough, UK) by means of a proprietary powder metallurgy production route, then forged and finally heat treated at the T4 condition. LFW joints were manufactured at TWI (The Welding Institute, Cambridge, UK) through careful optimisation of the welding parameters (Tab. II); no post-weld heat treatment was carried out. The microstructural characterization of the joints was carried out by optical and scanning electron microscopes (OM, SEM). Tensile and hardness tests were performed on the joined specimens. Residual stresses, induced by the LFW, were evaluated separately for the matrix and the reinforcement, by means of neutron diffraction measurements carried out at the ENGIN-X station at ISIS (Fig. 2), Rutherford Appleton Laboratory, UK. The welded joint exhibited plastically deformed material expelled due to the high compressive pressure (Fig. 3). Microstructural analyses showed substantially defect-free joints (Figs. 4-5) with four characteristic regions: A central zone, with an ultra-fine microstructure and a uniform particles distribution; a thermo-mechanically affected zone (TMAZ) were it is possible to identify the plastic flow modification that the material underwent during the process; a heat-affected zone (HAZ), without noticeable microstructural modification but decreased hardness; the parent material, unaffected by the process. Statistical analysis of SEM images showed how LFW didn't affect the particles size distribution (Figs. 6-7). Maximum residual stress resulted lower compared to those reached with TIG arc fusion welding (Fig. 8). The hardness decrease, in the welded zone, was less than 10%, although sensible fluctuation of the data were observed, due to the complex microstructural modification introduced by the LFW (Fig. 9). The joint efficiency, evaluated in respect to the ultimate tensile strength (UTS), was higher than 80% (Tab. III), while the decrease of the elongation to failure (40% of the nominal values) was probably related to the orientation of the plastic flow in the TMAZ, where generally the fracture occurred (Fig. 10). SEM analyses of the fracture surfaces showed that the fracture was mainly due to the ductile failure of the matrix, rather than the interfacial decohesion at the particles-matrix interface or the cracking of the reinforcing particles (Fig. 11)