Loughborough University
PhD THESIS_Mushtaq Khan Oline 2010.pdf (16.94 MB)

Selective Laser Melting (SLM) of gold (Au)

Download (16.94 MB)
posted on 2010-05-10, 08:05 authored by Mushtaq Khan
Selective Laser Melting (SLM) is a laser based Solid Freeform Fabrication (SFF process which uses laser energy to melt a thin layer of metal powder. This process is repeated to produce a 3-dimensional metallic part. SLM is capable of producing intricate parts which are otherwise difficult to produce with conventional manufacturing techniques. As compared to traditional manufacturing processes, SLM can also produce parts with higher density. Before a material is processed using SLM, suitable processing parameters are first identified. Over the years, different materials have been processed using the SLM process. However, very little work has been done on SLM of bio-compatible precious metals such as gold and its alloys. Gold and its alloys have been used for manufacturing of dental crowns for centuries. The SLM process could be used to produce intricate metallic substructures for porcelain fused to metal dental restorations. This research work was focused on understanding the processing parameters for SLM of 24 carat gold powder. The gold powder was analyzed for Particle Size Distribution (PSD), apparent density and tap density before identifying suitable processing parameters for SLM. The gold powder particles were found to be spherical in nature but smaller particles stuck to each other and formed larger powder agglomerates. From the apparentdensity experiments, the gold powder was found to be cohesive and non-flowing in nature which hindered powder flowability during the powder deposition process with the existing system. This issue was resolved by designing a new powder deposition system which could allow the gold powder to flow evenly over the substrate. The tap density of the gold powder was found by Constant Weight Tap Density (CWTD) and Constant Volume Tap Density (CVTD) techniques. The difference in results from these two techniques was negligible. Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) of gold powder showed it to be more than twice as reflective as other commonly processed metal powders such as stainless steel and H13 tool steel. This analysis proved useful in understanding the laser processing of gold powder. Due to the high cost and small quantity of material available for this work, a very small build platform was designed to optimise material utilisation and reduce wastage. Single scans were performed on a single layer of gold powder to identify the good melting region. Five different regions i.e. balling, good melting, unstable melt, weak sintering and very little sintering were observed in the processing window. The balling phenomenon was observed at low and high scan speeds, which was due to the melt pool instability at these parameter settings. The size of droplets (balling) also increased with decreasing scan speed and increasing laser power which was due to an increase in the break up time of the molten metal. In the good melting region, the gold powder was found to be completely melted and continuous beads were successfully produced. The unstable melt region showed the melt pool spreading unevenly in different directions whereas in the weak sintering and very little sintering regions the gold powder did not melt completely. Single layers were produced on a layer of gold powder, which showed the parameters in the good melting regions to be suitable for multiple layer parts manufacturing. Gold cubes were produced using the suitable processing parameters identified from single scan and single layer experiments and then analyzed for their internal porosity. The porosity in the gold cubes was found to be at a minimum for parameters obtained in the good melting region. The internal porosity was found to be mostly inter-layer porosity; this indicated less heat transferred to the region between the two layers which could be associated with the high reflectivity of gold. The inter-layer porosity in gold cubes was further reduced by reducing the layer thickness. This could be due to the thinner layers requiring less energy to melt and be fused to the previous layers. The hatch distance had a negligible effect on the inter-layer porosity of gold cubes. The reduction in hatch distance increased the energy delivered but it was still not enough to completely melt the gold powder and fuse it to the previous layer. A pre-scan technique was also tested to be used for pre-heating the powder bed. However, due to the rapid drop in temperature, this technique was not found suitable to be used as a powder bed pre-heating technique. The gold cubes were checked for their mechanical properties i.e. hardness and modulus. The hardness of gold cubes was found to be higher than expected for 24 carat gold. The modulus was found to be less than 24 carat gold. This variation in the mechanical properties of gold cubes could be due to the rapid heating and cooling of material during the laser processing or presence of internal porosity in these gold cubes. After single scans and single layers manufacturing, gold dental parts (premolar and molar) were also manufactured using the optimum processing parameters. These gold dental parts were also analyzed for their internal porosity, which was found to be less than that observed in gold cubes. This difference in porosity could be due to the difference in structure of gold cubes and premolar part, where the latter was a thin wall structure.



  • Mechanical, Electrical and Manufacturing Engineering


© Mushtaq Khan

Publication date



A Doctoral Thesis. Submitted in partial fulfillment of the requirements for the award of Doctor of Philosophy of Loughborough University.

EThOS Persistent ID



  • en

Usage metrics

    Mechanical, Electrical and Manufacturing Engineering Theses


    Ref. manager