(38f) Prediction of the Spreadability of Metal Powders: The Last Developements

The spreadability of powders, i.e. their ability to produce smooth and homogeneous layers, is an essential property to guarantee the good quality of the parts produced in powder-bed based AM processes. However, evaluating the spreadability of new materials during production is usually not feasible due to the cost associated with the minimum batch volume to fill in the machine and the time required to empty and clean the machine between each test.

The Cohesive Index metric (GranuDrum, Granutools, Belgium) has been demonstrated to be well correlated with the homogeneity of the deposited layers in an SLM printer (ISO/ASTM TR 52952:2023). This indicates the strong relationship between powder cohesiveness and spreadability. These results opened up new perspectives to predict the spreadability of materials a the formulation stage, without having to produce large batches. Also, suppressing the need to fill the machine to evaluate the spreadability is nice to have to reduce the material cost and machine time. Recently, the study has been extended to broaden the range of types of material and particle sizes and confirmed the good correlation between the Cohesive Index and layer spatial homogeneity (Francqui et al., ICAM2023).

Investigating the spreadability directly in the SLM machine has the advantage of being close to the real configuration in production. Indeed, the actual spreadability is influenced by the powder property but also by the recoater geometry (blade/roller), the wiper shape and material (polymer, metal), and the bed roughness especially for the first layers. However, the instrumentation is restricted by the build chamber volume and the moving parts. In the previous studies, the built-in video camera system was used to acquire pictures of the powder bed. The pictures were then analyzed with a custom image processing algorithm to get a measure of the layer spatial homogeneity, the interface fluctuation metric.

To tackle some of the limitations of the in-situ evaluation, a test bench has been developed and instrumented (LEM3, Metz, France). The mechanical parts (bed, recoater) of the test bench are taken from the real machine (SLM280, SLM Solutions, Germany). Therefore, even if the evaluation is not performed inside the machine, the experimental conditions are still very close. Powder layers are sequentially deposited, and the layer quality is evaluated between each layer. The device has been instrumented with a confocal microscope used to precisely measure profiles of the height variation across the layer. The layer local topology is thus directly accessible. In addition, powder samples are taken from different locations of the spred and the particle size distribution is analyzed.

This measure is complemented with optical images taken by a camera placed orthogonally to the powder bed. Moreover, the lighting system is designed so that the light orientation can be varied relative to the recoater direction. Indeed, we show that depending on the light direction, the observation of defects attributed to the powder cohesiveness can be amplified compared to the ones attributed to the interaction with the recoater.