Developing Alloys for AM Processes
Much of the current research and development in additive manufacturing is devoted to making parts from just a few specific alloy compositions. The focus on these compositions and their behavior is justified because they are most familiar to designers and fabricators.
For this reason, there is some expectation that we can understand how parts made from these alloys will behave. However, additive processes produce microstructures that are in the as-cast state and can often be different from those found in wrought alloys, even after post-processing steps. A major challenge to adoption of additive methods is ensuring that all parts will behave as expected once they are in service.
We can create even more challenges by considering new alloys. Additive processes typically produce structures that are unattainable by conventional processing. It’s likely that there are many new alloys with desirable properties that could only be useful through additive methods. How do we assess these new alloys in the context of additive manufacturing, particularly for popular methods like powder-bed processes? It’s not practical to make small amounts of powder to assess behavior.
At Carnegie Mellon, we have been investigating what we can learn using the energy source of an additive process (e.g., the laser in laser powder-bed fusion) to create melt pools on bulk alloys. This is illustrated schematically in the figure.
In these no-powder experiments, we see similarities between melt-pool shapes and microstructures compared to melt pools made with a layer of powder. Although this doesn’t simulate the effects of overlapping layers when building a finalized part, we can start to get an understanding of how these alloys might behave in additive processes. The real benefit is that we can easily create small amounts of novel bulk alloys of virtually any composition to screen both how composition and process variables (e.g., beam power and travel speed) affect properties.
Recently, my group has been using this method to investigate how boron additions to titanium alloys change the material behavior. Boron has been added to conventionally processed titanium alloys to refine grain size and improve mechanical properties but generally at levels less than 1% by weight. Above that threshold, enough titanium-boride phases form to negatively influence mechanical properties.
The rapid solidification achievable by additive methods like laser powder-bed fusion should result in much finer dispersions of the titanium borides, even when the boron addition amounts are high. We tested this by using a small arc-melting setup to make buttons of Ti6Al4V with varying boron additions – 1-10% by weight. We then made a large number of single laser tracks in a laser powder-bed machine to study how the microstructure changes under different laser-beam powers and travel speeds.
We extracted a significant amount of data after examining each of these single tracks and could identify where defects might occur. We observed a variety of microstructures. The example in the figure is of a structure containing a network of titanium boride (TiB) with spacing on the order of a few microns.
The network microstructure is encouraging because it is uniform in the melt pool and has high hardness. No networks of boride phase this fine have been created using conventional processing. This structure is promising to explore further for actual parts. The microstructures in the figure were produced from laser remelt tracks on totally solid samples (i.e., no powder involved).
We’re in the process of obtaining some powder batches of compositions that show potential and will turn our focus to building three-dimensional parts to fully assess the alloy properties. Although the powderless method is only part of the story, we are looking at even more alloy compositions to design new materials for the future of additive manufacturing.