Numerical and experimental studies of laser powder bed fusion processes
Yousef Shaheen is a PhD student in the department Multi Scale Mechanics. (Co)Promotors are dr. T. Weinhart, prof.dr. S. Luding and dr.ir. T.H.J. Vaneker from the faculty of Engineering Technology.
Additive Manufacturing (AM), commonly known as 3D printing, stands as a transformative technology poised to reshape the approaches to product design, development, and manufacturing. Diverging from conventional subtractive manufacturing techniques, which involve removing material from a solid block to form a part, AM constructs parts layer by layer, utilizing various materials like plastics, metals and ceramics.
This approach facilitates the fabrication of complex geometries and structures that were previously very difficult or economically impractical. In recent years, AM has experienced significant growth, finding applications across diverse industries such as aerospace, automotive, healthcare and consumer goods.
There are seven unique methods in additive manufacturing, we focus on laser powder bed fusion (LPBF) method in this thesis. Laser powder bed fusion employs a high-energy laser to selectively melt and fuse consecutive layers of powdered material resulting in the formation of a three-dimensional object. Some of the main challenges of LPBF are parts of high porosity, anisotropic properties and divergence from design to execution.
To address these challenges, it is essential to optimize process parameters, a task that typically involves expensive experimental trials. This thesis aims to explore the feasibility of utilizing numerical methods for optimizing the laser powder bed fusion (LPBF) process. Furthermore, experiments are carried out to calibrate and validate the developed numerical models, enhancing their practical utility. The specific emphasis lies on two critical aspects: powder spreading and the processes of powder melting, coalescence and consolidation.
In the first part of this study, we investigate the powder spreading process numerically and experimentally.
First, we employ computer simulations utilizing the discrete particle method (DPM). This methodology enables the establishment of metrics to assess the quality of powder layers, enabling direct comparisons across different tools and parameters. In the simulations, we emulate the influence of complex particle shape and surface roughness by incorporating factors such as rolling resistance and inter-particle sliding friction. Furthermore, we investigate the impact of particle cohesion, the choice of spreading tool shape and direction on the overall results.
Next, we employ an experimental setup for powder spreading to investigate and characterize the spreadability of powder during the spreading process. Three distinct powder materials are considered: two metals (Ti6Al4V and Inconel 718) and one polymer (PA12). Each powder is sieved based on its D50 particle diameter, resulting in three samples for each powder: as received, coarse (powder retained in the sieve) and fine (powder passed through the sieve). Additionally, we examine three different spreading process parameters, namely spreading tool speed, spreading tool geometry (blade and a counter-rotating roller) and spreading tool gap height. Two metrics, powder layer area density fraction and powder layer uniformity, are defined to characterize the quality of the spread powder layer.
We showed that various particle characteristics such as coarse/large, strongly cohesive, non-spherical, or rough particles contribute to defects in the spread powder layer, thereby reducing its quality. Moreover, we found a unique correlation between quality criteria uniformity and mass fraction/area density fraction, facilitating the easy assessment of layer quality through experimental measurements.
In the second part, we introduce the developed thermal discrete particle and contact model, which accurately captures fundamental phenomena such as melting, coalescence and consolidation within laser powder bed fusion (LPBF). We proposes in our model that, influenced by heat, solid particles in contact initially undergo melting, subsequently coalesce and solidify to create bonds during the cooling phase. The rate of coalescence during melting is determined by the material's surface tension and viscosity. To simulate the phase transition, we employ an apparent heat capacity method. Initially, we present our contact model and validate it against analytical solutions for a two-particle system. Finally, we demonstrate the effectiveness of our model by applying it to a multi-particle scenario. Our model allows for the optimization of the laser powder bed fusion process parameters.
