Use of welded titanium alloy to aluminium alloy structures in the aerospace industry has a number of potential benefits for both cost and weight saving by enabling titanium to be used only in the most critical parts, with the cheaper and lighter aluminum alloy making up the rest of the structure. However, due to the formation of brittle intermetallic compounds (IMC) at interface and the enormous gap in melting point, the welding of titanium to aluminium remains a major challenge. Solid state welding processes are most likely to be successful since they do not involve any melting, and so issues associated with the large difference in melting point and the high reaction rate of the liquid phase are avoided. In this study, an emerging low energy input solid state welding process - high-power ultrasonic spot welding (USW) was applied to weld Al and Ti (AA6111-T4/Ti6Al4V and AA2139-T8/Ti6Al4V combinations). No obvious intermetallic reaction layer was observed on the Al/Ti interface even using transmission electron microscopy. As a result, the maximum joint strength measured reached the same level as similar Al-Al (AA6111) welds and greatly exceeded those observed in Al-Fe and Al-Mg joints made using the same technique, in which a brittle reaction layer forms rapidly. However, the Al/Ti welds always failed at the weld interface after natural ageing, which is not desirable due to the low fracture energy associated with interfacial fracture mode. By using high resolution STEM-EDS, residual oxides and Si segregation were detected on the as-welded Al/Ti interface, which are thought to be factors that result in the no reaction layer Al/Ti interface. The Si segregation is predicted to be able to increase the weld interface cohesion through thermodynamic calculation. A series of prolonged heat treatment experiments were performed to understand the Al-Ti reaction layer growth kinetics and to explain the lack of reaction layer in as-welded Al-Ti joint. Al3Ti (D022 structure) was the only Al-Ti intermetallic phase observed in the reaction layer (IMC layer). In pure Al/Ti joints, it is found that the very long slow-growth stage of IMC layer is probably caused by the residual oxides on the interface. Calculations show that grain boundary (GB) diffusion makes the major contribution to the effective diffusion coefficient in the Al3Ti layer. In AA2139/Ti joints, the IMC layer growth is significantly slower than that in pure Al/Ti joints. The effects of alloying elements on the IMC layer growth was studied in detail. Cu was observed to segregate on both the Al3Ti grain boundaries and the Al3Ti/Ti interface. Si also segregated on the the Al3Ti/Ti interface and enriched in the Al3Ti layer. Both Cu and Si are thought to retard IMC layer growth. Interestingly small patches of Al were found trapped in the IMC layer; its formation mechanism is discussed. In pure Al/Ti6Al4V joints, the IMC layer growth rate did not change significantly. The presence of V greatly retarded the Al3Ti grain growth at high annealing temperature (630 °C) and suppressed the anisotropic growth of Al3Ti at 600 °C. Overall this study successfully joined Al/Ti by USW and systematically investigated the grain size effect and alloying effects on the Al3Ti layer growth. The present study for the first time: (a) observed the no-IMC-layer Al/Ti weld interface; (b) observed Cu segeration on Al3Ti GBs; (c) quantitatively studied the grain size effect on Al3Ti layer growth kinetics; (d) observed the orientation relationship between trapped Al islands and the adjacent Al3Ti grains; (e) observed that V greatly retarded the anisotropic growth of Al3Ti grains.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:666849 |
| Date | January 2015 |
| Creators | Zhang, Chaoqun |
| Publisher | University of Manchester |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | https://www.research.manchester.ac.uk/portal/en/theses/ultrasonic-welding-of-aluminium-to-titanium-microstructure-properties-and-alloying-effects(1aaec7d7-19dd-4524-bde4-e6ffef5217a0).html |
Page generated in 0.002 seconds