Date of Award

Fall 2024

Document Type

Open Access Dissertation

Department

Mechanical Engineering

First Advisor

Lang Yuan

Abstract

Laser Powder Bed Fusion (L-PBF), a widely used metal additive manufacturing (AM) method, is emerging as a key technology in the modern manufacturing industry. Its applications span across aerospace, transportation, energy, and biomedical industries, aiming to maximize product functionality, quality, and manufacturing efficiency.

Surface roughness, an important measure of product quality, impacts geometrical tolerances and plays a critical role in mechanical and functional properties. Although post-process techniques such as machining, shot peening, and chemical processing can be used to refine surfaces, internal surfaces in complex geometries remain difficult to treat, limiting part performance due to undesired surface features. The formation of surface topography in L-PBF is the result of complex, intertwined cause-effect relationships between physical phenomena—such as heat transfer, solidification, fluid dynamics, and surface tension—and process parameters, including powder size distribution, laser settings, layer thickness, and scanning patterns.

Investigating surface roughness enhances our understanding of melt pool dynamics and morphology. It is generally accepted that surface roughness is influenced by factors such as partially remelted particles, spatter, balling effects, and geometrical stair effect. However, the relationship between melt pool morphology and surface roughness across all surfaces has not been thoroughly studied. Additionally, the dominant surface characteristics and their formation mechanisms, especially concerning laser-powder interactions and melt pool solidification behavior, remain insufficiently explored.

In this research, systematic experiments were conducted to examine how process parameters (including laser power, scanning speed, and contour scanning offset) affect surface roughness on both top and vertical surfaces. Surface roughness (arithmetic average, Sa) was measured using confocal microscopy, while scanning electron microscopy (SEM) was used to investigate surface characteristics and solidification microstructures near the surfaces. Firstly, the relationship between melt pool dimensions and processing parameters was established. Without keyholing, high energy density or high power and speed under the same energy density level results in wide and deep melt pools. The surface features, including humps and balling, were analyzed based on laser power, speed and hatch spacing. Secondly, the top surface roughness is correlated with the melt pool behavior, including overlaps between melt pool tracks, powder denudation and surface tension induced instability. A process map for the top surface was proposed and the preferred melt pool morphology was revealed based on the porosity level in all printed samples. Thirdly, a new mechanism for dross formation on the vertical surfaces was revealed based on multi-layer melt pool instability and detailed solidification microstructure near the vertical surfaces. This study also implies that vertical surface roughness is positively correlated to melt pool width and depth. Controlling melt pool morphology for contour parameters provides a solution to address the surface roughness on vertical surfaces. Fourthly, to achieve the best vertical surface roughness, the selection of contour parameters, including laser power, scanning speed and offset to inner hatches, addresses the melt pool motion and instability at the surfaces, which is found to depend on the hatch parameters. A guideline for choosing the contour parameters is provided based on experimental observations supported by numerical simulations.

Overall, this research deepens the understanding of melt pool behavior, especially in terms of melt pool migration at the powder-bulk boundary and its effect on surface roughness. It proposes effective methods—such as optimizing print energy density of contour/hatch scans and the contour offset distance—to precisely control and improve surface quality in L-PBF processes.

Rights

© 2025, Tianyu Zhang

Download

Share

COinS