Texas A&M University researchers have used machine learning to further refine the process of creating superior metal parts using laser powder bed fusion additive manufacturing.
Laser powder bed fusion builds 3D metal parts layer by layer through fusing metal alloy powders together with a laser.
These powders contain a mixture of metals such as nickel, aluminium and magnesium at different concentrations.
During printing, these powders cool rapidly after being heated by the laser beam. Since the individual metals in the alloy powder have very different cooling properties and consequently solidify at different rates, this mismatch can create a type of microscopic flaw called microsegregation.
‘When the alloy powder cools, the individual metals can precipitate out,’ explained Raiyan Seede, a doctoral student in the university’s Department of Materials Science and Engineering. ‘Imagine pouring salt in water. It dissolves right away when the amount of salt is small, but as you pour more salt, the excess salt particles that do not dissolve start precipitating out as crystals. In essence, that’s what is happening in our metal alloys when they cool quickly after printing.’
This defect appears as tiny pockets containing a slightly different concentration of the metal ingredients than other regions of the printed part. These inconsistencies compromise the mechanical properties of the printed object.
To rectify these microdefects, the researchers investigated the solidification of four alloys containing nickel and one other metal ingredient. In particular, for each of these alloys, they studied the physical states or phases present at different temperatures for increasing concentrations of the other metal in the nickel-based alloy.
From detailed phase diagrams, they could determine the chemical composition of the alloy that would lead to minimum microsegregation during additive manufacturing.
Next, they melted a single track of the alloy metal powder for different laser settings and determined the process parameters that would yield porosity-free parts. Then they combined the information gathered from the phase diagrams with that from the single-track experiments to get a consolidated view of the laser settings and nickel alloy compositions that would yield a porosity-free printed part without microsegregation.
A scanning electron microscope image of a single laser scan cross-section of a nickel and zinc alloy. Here, dark, nickel-rich phases interleave lighter phases with uniform microstructure. A pore can also be observed in the melt pool structure. (Image: Raiyan Seede)
Lastly, the researchers went a step further and trained machine-learning models to identify patterns in their single-track experiment data and phase diagrams to develop an equation for microsegregation applicable to any other alloy. The equation is designed to predict the extent of segregation given the solidification range, material properties, and laser power and speed.
'Our methodology eases the successful use of alloys of different compositions for additive manufacturing without the concern of introducing defects, even at the microscale,' said Ibrahim Karaman, head of the Department of Materials Science and Engineering. 'This work will be of great benefit to the aerospace, automotive and defence industries that are constantly looking for better ways to build custom metal parts.'
The researchers explained that the uniqueness of their methodology lies in its simplicity, which can easily be adapted by industries to build sturdy, defect-free parts with an alloy of choice. They noted that their approach contrasts prior efforts that have primarily relied on expensive, time-consuming experiments for optimising processing conditions.
'Our original challenge was making sure there are no pores in the printed parts because that’s the obvious killer for creating objects with enhanced mechanical properties,' said Seede. 'But having addressed that challenge in our previous work, in this study, we took deep dives into fine-tuning the microstructure of alloys so that there is more control over the properties of the final printed object at a much finer scale than before.'
The research has been published in Additive Manufacturing and was supported by the United States Army Research Office and the National Science Foundation.
Related article: Tailoring microstructures in metal additive manufacturing