Development of a microscale numerical modelling method for the simulation of 3D powder bed printing of Bioceramics

Qureshi, Mohammad Ali Ausaf ORCID: 0000-0001-6954-2463 (2023). Development of a microscale numerical modelling method for the simulation of 3D powder bed printing of Bioceramics. University of Birmingham. Ph.D.

Preview

Qureshi2023PhD.pdf
Text - Accepted Version
Available under License All rights reserved.
Download (10MB) | Preview

Abstract

Significant progress has been made in powder bed selective laser processing (PBSLP) of 3D printing of metals and polymers. Researchers have developed numerical models to simulate these processes to develop a better understanding, both at macro and microscales. However, due to their high thermal resistance, laser processing of ceramic materials presents unique challenges. Therefore, this resea rch focused on modelling and simulating the PBSLP of bioceramics to develop a closer insight into its processing at the microscale. Hydroxyapatite (HAP) and alumina powders were particularly addressed, due to their broad applications in the medical industry, including orthopaedic extremities.

The printing process consists of a pre-processing phase involving the powder bed deposition and layer formation, followed by a laser processing phase that selectively scans the powder region to generate heat for sintering. Therefore, we adopted a discrete element method (DEM) tool, which could model the dynamics of discrete spherical particles virtually at any scale (as low as nano).

In the preprocessing phase, the impact of the circular profile blade coater was investigated on packing density and surface roughness for 5 different size distributions. For each sample, 4 layers were deposited with varying layer thicknesses of 100, 150, and 200 μm. The 100 and 150 μm thicknesses exhibited considerable voids in the initial layers, indicating loose packing densities—however, the effect of layer size diminished by the fourth layer, resulting in comparable final packing densities. As the DEM methodology enabled a closer examination of powders' temporal and dynamic behaviour during the deposition process, it was observed that voids in the smaller layer thicknesses were caused by the smaller ground clearance of the blade, which swept away the larger particles. The visualisation of the deposition process simulations also demonstrated how those voids closed as more layers were added. Furthermore, a ray tracing code was also developed to assess each sample's surface quality, revealing that distributions with smaller average particle sizes yielded finer surfaces.

In the laser processing phase, ray tracing laser heating and particles sintering codes were developed and coupled with the DEM solver to implement a limited-scale PBSLP model. The impact of various fibre laser parameters, such as power (30-54 W) and scan speed (25-125 mm s-1), on the sintering of HAP powder was investigated. Simulations correctly predicted the sintering responses for 72% of the parameters compared to the experiments. 12% of the laser parameters in simulations overestimated the results by predicting melting temperatures instead of sintering, while 16% underestimated the results by predicting no sintering while they occurred in the experiments. Combining DEM models with laser ray tracing and sintering codes provided valuable insights into the PBSLP. The heat distribution field resulting from the laser scan could be precisely modelled and investigated within the random-sized powders. The sintering processes against different laser parameters were better understood. The slower laser scan speeds for the same power improved the densification of powders. Higher power and lower speeds resulted in powder melting, while lower powers and speeds produced no sintering response. This methodology could also identify the combination of laser parameters that produced a solid-state sintering phase.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):