Friction extrusion is a thermo-mechanical process that can be used to form fully consolidated wire, rods, tubes, or other non-circular metal shapes directly from a variety of precursor charges including metal powder, flake, machining waste (chips or swarf) or solid billet. The process imparts unique, and potentially, highly desirable microstructures to the resulting products. Friction extrusion was invented at The Welding Institute in the UK and patented in 1991. It was originally intended primarily as a method for production of homogeneous microstructures and particle distributions in metal matrix composite materials.[1]
As in conventional extrusion processes, in friction extrusion, a shape change is enforced on the charge by forcing its passage through a die. However, friction extrusion differs from conventional extrusion in several key ways. Critically, in the friction extrusion process, the extrusion charge (billet or other precursor) rotates relative to the extrusion die. In addition, similar to conventional extrusion, an extrusion force is applied so as to push the charge against the die. In practice either the die or the charge may rotate or they may be counter-rotating. The relative rotary motion between the charge and the die has several significant effects on the process. First, the relative motion in the plane of rotation leads to large shear stresses, hence, plastic deformation in the layer of charge in contact with and near the die. This plastic deformation is dissipated by recovery and recrystallization processes leading to substantial heating of the deforming charge. Because of the deformation heating, friction extrusion does not generally require preheating of the charge by auxiliary means potentially resulting in a more energy efficient process. Second, the substantial level of plastic deformation in the region of relative rotary motion can promote solid state welding of powders or other finely divided precursors, such as flakes and chips, effectively consolidating the charge (friction consolidation) prior to extrusion.[2] Scrolled features on the face of the die aid material flow into the extrusion orifice which can lead to order of magnitude reduction in extrusion force compared to conventional extrusions of equivalent cross section.[3] Third, the combined effects of elevated temperature and large levels of deformation normally lead the extrudate to have a relatively fine, equiaxed, grain structure which results from recrystallization after the conclusion of deformation: desirable crystallographic textures may also be created by the process and formation of nanocomposite structures are also possible.[4]
Based on the foregoing, it can be said that the essential controlled parameters in friction extrusion are typically:
The corresponding response parameters include:
In principle, friction extrusion can be performed on any machine which can produce the required rotary and linear motions between the die and charge. Examples include machines built for friction stir welding, milling machines modified to accommodate the extrusion forces, and purpose built friction extrusion equipment such as the shear assisted processing and extrusion (ShAPE™) machine at the Pacific Northwest National Laboratory. Figures 1-3 show examples of friction extrusion equipment and extruded products. Figure 4 shows typical friction extrusion dies designed for production of wire, rod and tube. Dies are rotated in the direction which enhances material flow toward the extrusion orifice during the process.
In conventional extrusion, the strain imparted to the charge is loosely defined by the extrusion ratio.[5] The extrusion ratio is simply the cross-sectional area of the extrusion billet, A0, divided by the cross sectional area of the extrudate, Af. The extrusion strain is then e=ln(A0/Af).
In friction extrusion there is an additional strain component which will arise from the shearing motion of the rotating die as it contacts the charge. The strain produced by the rotation of the die results in redundant work as it does not accomplish a shape change. In order to investigate the strain due to shearing, studies have been performed with marker materials embedded in the material to be extruded.[6] After extrusion, these materials are detected by metallographic methods and provide insight regarding the way in which material flows during the extrusion process. Figure 5 shows an example of how the amount of shear strain changes with changing ratios of extrusion rate to die rotation rate. In the limit of very high extrusion rates, the friction extrusion process closely mimics the conventional extrusion process with respect to strain levels.
Figure 6 shows the cross section and microstructure of a titanium wire produced by friction extrusion of Ti-6-4 powder. Notably, the cross section is fully consolidated and the transformed b microstructure indicates that extrusion likely occurred near 1000 °C (above the beta transus for the alloy). Figure 7 shows grain size and crystallographic orientation typical of thin walled tubing extruded from AZ91 melt spun flake.[7] Grains are refined to less than 5 mm and orientation of the (0001) planes are off-normal due to the rotational shear component. Figure 8 shows examples of friction extruded magnesium alloy tubes. Friction consolidation has also been used to refine grain size and preferentially orient texture in functional materials such as bismuth-telluride thermoelectrics [8] and iron-silicon magnets.[9] Examples of the effect of friction extrusion of microstructure have been reported for AZ31,[10][11][12] various aluminum alloys [13][14][15][16] and pure copper.[17]