Discrete Element Method Investigation of Stainless Steel 316l Powder Flow in Vacuum Conditions during Additive Manufacturing

Burgess, A (2025) Discrete Element Method Investigation of Stainless Steel 316l Powder Flow in Vacuum Conditions during Additive Manufacturing. Doctoral thesis, Liverpool John Moores University.

Preview

Text 2025BurgessPhD.pdf - Published Version Available under License Creative Commons Attribution Non-commercial. Download (10MB) | Preview

Abstract

Additive Manufacturing is an increasingly popular approach for manufacturers to generate complex components more efficiently. As a flagship process of Industry 4.0, researchers are leveraging digital modelling to optimise the process without incurring the costs of practical trial and error experiments. This project pursues methods to optimise the powder spreading process and deliver novel solutions for commercial practices, using Discrete Element Methods to investigate how the spreading parameters and powder characteristics influence the quality of the formed powder bed. Three discrete metrics were identified to determine the powder bed quality: the packing density, surface roughness, and dispersion of polydisperse powder elements within the spread layer.
A key knowledge gap exists in the current research landscape. The majority of contemporary simulations insert powder in a user-defined volume which then falls to the substrate under gravity, a process referred to as the “rainfall” method. This misrepresents powder deposition in commercial Additive Manufacturing systems, where the powder is inserted using various techniques such as by a moving funnel, or a piston-operated supply table. This project addresses this knowledge gap by inserting Stainless Steel 316l powder with a moving funnel to provide a more realistic deposition approach than existing methods in the literature. Stainless Steel 316l is a metallic material, widely used in Powder Bed Fusion applications for its processability in generating high-quality and dense parts with complex geometries, excellent corrosion resistance, and mechanical properties such as strength, ductility, and toughness.
The novel simulation approach has been benchmarked against values set by the rainfall approach, with multiple powder sets inserted for comparison. These sets included uniform powder consisting of same-sized particles, and polydisperse sets with varying fractions of relatively coarse, fine, and intermediately sized particles within the size range in commercial Additive Manufacturing. The veracity of the digital model was ascertained by comparison to practical powder experiments, and confirmed the fidelity of the model to physical powder flow analysis.
Research highlighted that depositing the powder with the moving funnel engenders a significant difference in the spread layer. The funnel generally lowered the packing density by between 1-2% for all inserted sets and incurred a rougher surface in the spread layer ranging from 4.77% rougher to 72.34% rougher depending on the inserted particle size ranges. These results were significant to the existing research landscape, as they imply that rainfall models may artificially increase the quality of the formed powder beds. The results showed a suitable powder size range for Stainless Steel 316l, delivered by the moving funnel in the spreading conditions tested, would be a 60% population of 15-25 μm particles, 25% of the population between 25-40 μm, and no more than 15% of the particles ≥ 40 μm in diameter.
The key outcomes of the thesis, specifically that existing deposition methods currently misrepresent the techniques observed in industrial Powder Bed Fusion machinery, lay the foundation for future research. This work provides the basis for the realistic deposition of powder, advancing existing knowledge by increasing the accuracy of the Discrete Element Methods for Additive Manufacturing investigations, and more accurately reflecting commercial methods. Thus, giving a foundation for researchers in the industrial and academic spheres to adapt parameters and engender conditions which serve to optimise the powder bed. This contribution is further reinforced by the suggested size distribution of particles that optimise spreading conditions. For contemporaries in the simulation and modelling of Additive Manufacturing processes, a significant contribution has been made by establishing a direct conversion between properties known as the Surface Energy and Cohesion Energy Density. To the best of the author's knowledge this relationship, although explored in literature and within the wider Discrete Element Method community, had not previously been numerically established.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	Additive Manufacturing; Discrete Element Method; Stainless Steel 316l; Powder Spreading; Digital Modelling
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Engineering
Date of acceptance:	3 June 2025
Date of first compliant Open Access:	25 June 2025
Date Deposited:	25 Jun 2025 11:23
Last Modified:	25 Jun 2025 11:23
DOI or ID number:	10.24377/LJMU.t.00026573
Supervisors:	Malkeson, S, Falkingham, P, Sharp, M and Darlington, R
URI:	https://researchonline.ljmu.ac.uk/id/eprint/26573

View Item