Parameretizing the Friction Stir Welding of Aluminum Metal Matrix Composities

In his seminal text Friction and Wear of Materials, Ernest Rabinowicz defines wear as “the removal of material from solid surfaces as a result of mechanical action” [1]. The wear problem is incredibly important from an economic perspective; in the manufacturing industry, cost is in direct proportion to the frequency with which a particular component must be replaced. Wear as a phenomenon is undertheorized, a deficiency Rabinowicz attributes to the pervading idea among researchers that wear is a chaotic process, so complex that it cannot be characterized or predicted. This undercharacterization of wear is especially true for Friction Stir Welding (FSW), a solid-state joining process developed at The Welding Institute (TWI) and currently used by NASA for structural welds on the Space Shuttle external tank, Ares I upper stage and Orion crew exploration vehicle. The commercial aerospace sector also makes extensive use of FSW, as major components of the Boeing’s Delta and SpaceX’s Falcon series of rockets are joined using this technique. Traditionally fabricated from tool steels, FSW tools have a nearly infinite life when used to join Aluminum alloys. Tool wear in FSW is a phenomenon unique to high strength alloys, having been documented in the literature only for FSW of steels and metal composites [2].

Metal Matrix Composites (MMCs) consist of two separate phases. The continuous phase, termed the matrix, is a conventional metal alloy (usually Aluminum). Embedded within the matrix is a reinforcement phase, which contains abrasive material in the form of fibers or particles. Materials commonly used as reinforcements include Aluminum (III) Oxide (Al2O3), Silicon Carbide (SiC), and Boron Carbide (B4C). MMCs are classified according to the materials which comprise the matrix and reinforcement phases, the shape of the reinforcement,
and the concentration in which the reinforcement is present. While both fiber and particulate reinforced MMCs find applications in industry, the isotropic properties of particulate composites make them easier to characterize and model. The degree of enhancement in mechanical properties attained through use of a metal composite (such as increased strength, temperature resistance, and hardness) increases with the reinforcement percentage. For particle-based composites, where the reinforcement is added in the form of a powder, this translates to a
significant enhancement in strength with only a nominal increase in weight. The high strength to weight ratio of this class of composites makes them ideal for use in aerospace structures. Some “weight-saving” structures in which MMCs have been successfully implemented include the Space Shuttle Orbiter’s structural tubing, the Hubble Space Telescope’s antenna mast, control surfaces and propulsion systems for aircraft, tank armors, and braking systems for roller coasters [3-4]. Although the constituents of metal matrix composites are relatively inexpensive materials, the price of these composites is driven upward by costs associated with their manufacture
and machining. MMCs are thus reserved for use in structures where increased strength and weight reduction are of critical importance.

An additional barrier to the widespread use of MMCs is the difficulty encountered when joining them to other MMCs or unreinforced alloys in a larger structure. The most problematic aspect of fusion welding Aluminum MMCs to one another is the formation of a deleterious theta phase induced by the reaction of molten Aluminum with reinforcement particles. In the case of Silicon Carbide reinforcement, liquid Aluminum reacts with SiC to produce Aluminum Carbide (Al4C3). Stojohann et al. indicate that theta phase formation, initiated when the weld temperature exceeds the melting point of Aluminum, can be somewhat mitigated through careful control of temperature but not eliminated altogether in joints produced using fusion techniques [5]. This phase can segregate into a molten region, leaving behind a phase-depleted, low-strength region in the joint. A solid-state joining process, in which the weld temperature remains below the melting temperature of the workpiece, is a more viable alternative. A comparison of metal composite joints produced using friction stir welding with those joined using fusion techniques reveals that the deleterious theta phase is absent in the former and unavoidable in the latter [5]. While friction stir welding precludes the theta phase dispersion which contributes to significant reductions in joint strength, it is not a panacea. FSW of MMCs is characterized by rapid and severe wear of the tool; this wear is caused by contact between the tool and the comparatively harder abrasive reinforcement which gives the material its enhanced strength. This wear creates an engineering conundrum: though FSW is the only process capable of joining MMCs without theta phase disturbance, the consumption of tools associated with severe wear discourages its use for this application.
The objective of this research is to characterize the dependence of wear on process parameters and develop a predictive process model for wear incurred in FSW of MMCs. It is anticipated that results of this work will point the way toward the development of tool designs and/or selection of tool materials which can reduce or combat wear in this process.

References:

1. Rabinowicz, Ernest. Friction and Wear of Materials. New York: John Wiley & Sons,1965.

2. Stellwag, W.L. and T.J. Lienert. EWI. Friction Stir Welding of Aluminum Metal Matrix Composites Progress Report. Columbus: 2001.

3. Kunze, J.M. and C.C. Bamptom. “Challenges to developing and producing MMCs for space
applications.” Journal of the Minerals, Metals and Materials Society 53 (2001): 22-25.

4. DWA Aluminum Composites. 2009. DWA-DRA. Oct. 2009 <http://www.dwa-dra.com>.

5. Storjohann, D., O.M. Barabash, S.S. Babu and S.A. David, et al. “Fusion and Friction Stir Welding of Aluminum Metal Matrix Composites.” Metallurgical and Materials Transactions: A: Physical Metallurgy and Materials Science 36A (2005): 3237-3247.