Powder-based metal additive manufacturing (AM) has emerged as a competitive, alternative technology to conventional metallurgical techniques—such as casting and forging—for the production of parts used in different industrial sectors, including aerospace, automotive, and biomedical. This success is attributed to the capability of manufacturing near-net-shape parts via selective fusion (or sintering) of the metal powder feedstock, layer by layer. A derivative benefit of this capability is the reduced material waste associated with the additive process. Compared to conventional manufacturing techniques—whereby material is selectively removed from a workpiece to achieve the desired geometry—AM uses only the amount of powder needed to build the final part. As a result, AM offers lower yield losses and is regarded as a sustainable manufacturing technology.
However, state-of-the-art powder-based AM processes are, in practice, far from being sustainable. They require the use of high-quality powders with specific size and shape distribution. Producing powders with such stringent specifications involves processes—such as gas atomisation—which are inherently inefficient, as well as environmentally and economically expensive. Furthermore, AM is in itself an inefficient process. On average, only 30% of the powder feedstock is fused during a build. Although the unfused powder is supposed to be reused for subsequent builds, it often ends up being regarded as waste because it contains a fraction of oxidized and non-spherical particles. This reduction in powder quality is inevitable as powders are exposed to high-temperature processes during AM and this is known to have detrimental effects on the quality of future builds. As a result, the entire batch of unfused powder is typically re-melted through the same inefficient powder production processes.
We propose to develop a “smart” powder-based metal AM technology that is capable of processing low-quality powders to produce high-quality parts. By redesigning the way powders are processed during AM, we aim to relieve the stringent requirements on powder quality and thus improve the sustainability of AM. We call this new paradigm “green” AM of metals (GreAM). The hallmark of GreAM consists in using adaptive algorithms that modulate AM process parameters on the fly based on a real time analysis of the powder feedstock performed at high spatial resolution. Using selective laser melting (SLM) as a representative powder-based AM process, we propose a GreAM technology consisting of:
Inline mapping of powder attributes—such as particle size, shape, flowability, and oxidation state—during the powder re-coating process;
Physics-based predictions of the optimum processing conditions for the analysed powder properties.
This project is an integral part of the new IMPACT Industrial Chair (Innovative Materials and Processes accelerated through Artificial intelligence and Computing Technologies).
The GreAM project will be performed as a Ph. D. work in order to combine the synergetic skills and capabilities of NTU Singapore and the SAMANTA platform (Saclay’s Advanced Manufacturing and Technological Applications) from CEA.
CEA: Hicham MASKROT, firstname.lastname@example.org, 01 69 08 67 98