Investigation of surface quality, metallurgical defects and mechanical properties of Laser powder bed fusion processed 316L stainless steel and Ti6Al4V titanium alloys

  • Shubhavardhan Ramadurga Narasimharaju

Student thesis: Doctoral Thesis


Additive manufacturing (AM) is making a significant breakthrough including osseointegration implants used in biomedical applications and heat transfer-cooling channels used in aerospace applications. Most functional osseointegrated implants and cooling channels require a surface roughness of 0.5 µm -1.5 µm, with the current state-of-the-art of AM, it is non-viable to achieve this extent surface finish. However, early-stage research revealed that these implants' biocompatibility and load-bearing capabilities could be enhanced by designing/modifying and/or tailoring the surface roughness to greater than 4 µm (up to 100 µm in dental implants). Similar certain (higher) surface roughness induced by intentional porosities adds more benefits for extracting heat from cooling and maintaining the structural integrity of rocket engines and other aerospace-related applications. Additionally, the surface quality of acetabular and tibial knee augments ortho implant's external surface is different from internal surfaces. To design or tailor the desired roughness levels, it is imperative to understand the complex surface topography fingerprint that emerges on various external and internal AM surfaces. The laser powder bed fusion (LPBF) process is a type of metal AM process which uses a laser energy source to selectively melt the predefined contours of thousands of minuscule layers of the powder bed layer-by-layer; generating complex surface topography. This complex surface topography of metal AM presents a new challenge for conventional surface quality characterisation or surface texture metrology as the surface quality of AM surfaces is not trivial like conventional surfaces. The complex surface topography of various metal AM surfaces is formed by the virtue of diverse metallurgical defects and surface asperities, such as balling, porosities, cracks, staircase effect, spatters and adhered un-melted or partially melted particles which impart poor surface quality. As a result of this substandard final part quality, LPBF components fail to meet compliance with industrial standards. It is paramount to address the emergence of these defects and asperities, but very limited research is available addressing the emergence of metallurgical defects and surface asperities. Conventional surface texture metrology is predominantly focused to characterise the overall surface topographical asperities data of the whole measured region by a limited set of scalar values. Ra/Sa lacks characterising the diverse surface topographical features (metallurgical defects and surface asperities) and fails to interpret the functional performance. Also, the conventional texture parameters (Ra/Sa) are not intending to provide information about the individualities present on the surface nor the spatial distributions. Also, good practice to characterise and correlate the individual defect or asperity (including quantification of particle features) with a suitable surface texture parameter is clearly missing. More importantly, it is incomprehensible to characterise and correlate this plethora of defects and anomalies with just one profile/areal surface texture height parameter (Ra/Sa). For example, a widely used Ra/Sa parameter is not suitable or applicable to characterise, correlate and or quantify the particle features (spatters, un-melted or partially melted particles) formed on various LPBF surfaces. Furthermore, the elaborate geometries of internal surfaces promote a significant challenge for measurement and characterisation, concurrently post-processing alone is not sufficient to improve the internal surface quality. Establishing an advanced surface texture characterisation tool for complex metal AM surfaces is a prerequisite. The research developed within this thesis contributes to the novel characterisation and quantification of diverse metallurgical defects and various surface asperities which emerge on both external and internal surfaces of bespoke metal AM artefacts built with varying inclination angles. The individual surface defects and asperities are then correlated with an array of suitable areal surface texture parameters described in ISO 25178-2, and newly proposed particle analysis for quantifying the particle features. Based on the areal surface texture characterisation, height parameters Sa and Sq, and spatial parameter Sal were suitable to characterise the staircase effect, while hybrid parameters Sdq, Sdr, feature parameter Spd and functional parameter Vmp were appropriate to characterise the particle features (spatters, un-melted/partially melted particles). Particle analysis is applicable to quantify the particle features. The experimental investigation and prediction model established a strong correlation between final surface quality and varying inclination angles. The research also found that the location of AM part on the build platform and the laser incidence angle significantly influence the final surface quality. A comprehensive statistical analysis of variance (ANOVA) revealed the presence of a statistical significance of inclinations angles, build orientations and the type of surfaces on the resultant surface quality. The critical LPBF parameters deemed mainly responsible for designing/optimising the final surface quality are laser power, hatch spacing, layer thickness, point distance, and exposure time. Therefore, sincere efforts have been made to design and optimise the critical parameters to achieve (tailor) lower or desired final surface quality by using the Taguchi design of experiments (TDOE). The impact (percentage contribution) of individual critical parameters on the resultant surface quality was examined by ANOVA. The results found that the hatch spacing and point distance were the main contributing factors to the top surface quality, while exposure time and point distance were the dominant factors responsible for the side surface quality. Confirmation experiments were conducted to validate the statistically predicted parametric combinations and compare them with the industrial default settings. The results demonstrated that the lower or desired final surface quality can be achieved/tailored by optimising the critical parameters. Additionally, the diverse metallurgical defects and surface asperities that emerged on 25 different top and side surfaces were systematically characterised. To complement surface texture characterisation and TDOE optimisation, the influence of different build orientations on the surface quality, microstructure and mechanical properties was studied. Horizontally-oriented samples displayed higher tensile strength and superior fatigue life, due to refined microstructure and smaller defects whereas the presence of oxide inclusions, bigger defects and asperities in vertically-oriented samples resulted in poor performance under both static and dynamic loading conditions. This complimentary research investigation concluded that build orientations significantly influenced the surface quality and mechanical properties especially the tensile and fatigue performance of LPBF dog-bone samples.
Date of Award17 Aug 2023
Original languageEnglish
SupervisorShan Lou (Main Supervisor), Wenhan Zeng (Co-Supervisor) & Jane Jiang (Co-Supervisor)

Cite this