Defect classification and dimensional estimation in single-point exposure laser powder bed fusion of lattice struts using optical emission spectroscopy

Single-point exposure (SPE) allows generating strut dimensions close to the size of the laser beam and the powder particles in laser powder bed fusion, making it appealing for producing lattice structures based on thin struts as well as biomedical devices like stents. The SPE strategy demands careful attention concerning the defect formation due to shorter interaction times and smaller spatial extents compared to conventional vectorial scanning methods. Monitoring the presence or absence of powder and the change of material type due to multi-material processing is crucial to prevent structural failure. Hence, a monitoring technique able to analyze each released laser pulse should be integrated to the process. Optical emission spectroscopy (OES) is a method that can accommodate the needs for sensing the process characteristics at the level of a single exposure. The emission spectra change as a function of the material type and also based on both powder and bulk form. Moreover, with adequate training, the emission spectra can be used to estimate the consolidated region dimension. This work demonstrates the use of a coaxial OES approach to reveal the mentioned defect types, as well as to provide a geometrical size of the circular lattice strut. Initially, the approach is described along with the definition of the monitoring setup, followed by the experimental campaign to generate training data. By means of machine learning algorithms, the approach is capable of indicating the presence or absence of powder, the material type, and the consolidated region size with regard to the AISI316L and Ti–6Al–4V struts at each exposure to the single laser pulse.