Selective laser melting of aluminium components

Previous work has shown that the processing of aluminium alloys by selective laser melting (SLM) is difficult, with reasonable components only being produced with high laser powers (minimum 150W) and slow laser scanning speeds. The high laser power is a significant problem a:3 it is higher than that used in many SLM machines. Also, the combination of high power and low speed creates a large melt pool that is difficult to control, leading to balling of the melt and possible damage to the powder distribution system. Even when processing is carried out successfully, the high power and slow scan speed significantly increase build time and the manufacturing costs. This paper considers the changes that can be made to the SLM process so as to reduce the laser power required and increase the laser scanning rates, while still producing components with a high relative density. It also considers why aluminium and its alloys are much more difficult to process than stainless steels and commercially pure titanium. Two MCP Realizer machines were used to process 6061 and AlSi12 alloys, one processing at SOW and the other 100W laser power. Even with an optimum combination of process parameters a maximum relative density of only 89.5% was possible (achieved with 100W). The major confounding factor for processing aluminium and its alloys was found to be oxidation due to the presence of oxygen within the build chamber. This formed thin oxide films on both the solid and molten materials. It was observed that the oxide on the top of the melt pool vaporised under the laser creating a fume of oxide particles, while melt pool stirring, probably due to Marangoni forces, tended to break the oxide at the base of the melt pool allowing fusion to the underlying tracks. However, the oxides at the sides of the melt pool remained intact creating regions of weakness and porosity, as the melt pool failed to wet the surrounding material. Therefore, if 100% dense aluminium components are to be produced by SLM, using low laser powers, methods need to be developed that can either disrupt these oxide films or avoid their formation. (C) 2010 Elsevier B.V. All rights reserved.