Additive Manufacturing (AM) is increasingly considered a viable technique for series production of patient-specific medical implants and components with very high strength-to-weight ratio for aerospace applications. Still, valid concerns about quality assurance are hindering the wider adoption of AM techniques in these application areas with extremely stringent standards. One of the main conundrums involves the reuse of feedstock material and the implications for part quality and batch-to-batch consistency in laser powder bed fusion (LPBF), the AM technique most popular for small series production of complex, functional parts.
In traditional subtractive machining, each part is fabricated from a solid block of fresh stock, which is typically validated as a lot prior to use. The cuttings and chips may be recycled to produce new stock, but this is almost never done by those machining the parts.
On the contrary, LPBF is an additive process. This means that the leftovers of the process are essentially in the same form as the stock material; both are powdered. For this reason, “used” powder is reused in subsequent builds. A powder particle can pass through the process cycle dozens of times before it is incorporated into a part. The problem is that during the printing process, the feedstock which is not fused into parts inevitably degrades.
In polymer LPBF (also called selective laser sintering, SLS), degradation results from exposure to elevated temperatures and small amounts of oxygen. The original powder properties can’t be maintained, so a fraction of the unfused powder must be combined with fresh (virgin) feedstock in order to reuse the material. In another Medium post we discussed this problem in detail, and we suggested that a selective powder deposition system like Aerosint’s could be used to fill the non-part areas of the powder bed with an inert, inexpensive support material. For expensive high-performance materials like polyetheretherketone (PEEK) this could represent a vastly more economical route, as these materials often have an even higher scrap rate than standard SLS polymers. Still, the chance for contamination of the part material with the support materials exists and may not be acceptable for the most demanding applications.
In metal LPBF, the increased laser power needed to melt metal creates an extreme environment. Evaporated metal material precipitates inside the chamber and on the powder bed as fumes composed of nanometer-sized particles. The bed is further contaminated by spatter — larger droplets of material ejected from the melt pool by the kinetic expansion of rapidly heated material. Both contaminate the unfused powder. The composition of the powder also slowly changes through pickup of oxygen and / or nitrogen molecules present as residual contamination in the process chamber atmosphere due to elevated temperatures.
To reuse the powder, it is typically passed through a sieve to remove the larger spatter globules, but the fume particles are nearly impossible to remove. Furthermore, as the powder is aerated during sieving this process can add even more oxygen to the powder. End users of the most demanding applications are worried about these complications arising from the re-use of processed powder — this is evident from current standardization actions and discussions during conferences and trade meetings, such as the campfire session on powder recycling during the 2019 meeting of the European Powder Metallurgy Association.
Despite reassurances in the form of white papers from machine makers EOS and Renishaw that demonstrate the high reuse rate for titanium alloys, a typical strategy by end users to counteract any potentially unfavorable effect of this slow degradation is batch control of the feedstock, combined with an often-arbitrary limit on the reuse. This practice effectively nullifies the concept of near zero waste, a benefit which has often been cited to promote additive manufacturing.
However, with selective powder deposition it is possible to place powder exactly where you are going to use it. The first focus of this technology has been to use it in the preparation of multi-material parts, but through interaction with industry we’ve seen that the use of a cheap, non-degrading support material is of nearly equal interest.
This idea can be taken a step further: Instead of introducing a support material with a different composition, one can use powder that has passed through the process several times. This cycled material is therefore used as volume-filling bulk and for the support structures (in the case of metal LPBF), and virgin powder is deposited in part locations in each layer. By depositing a small excess of virgin powder around the part cross-sections, only the reused powder is “contaminated” with clean material, while the cycled material that could contaminate the part is of nearly the same composition. The consumption of reused powder to build support structures is compensated as well.
As long as the powder flows properly, one can keep using it indefinitely. Through this approach, powder waste is minimized and virgin powder is always used for part body material. Costs associated with material waste are therefore decreased, and part properties should be much more predictable and consistent from batch-to-batch than they are in LPBF currently.
by Bram Neirinck, Ph.D. Senior R&D / Applications Engineer @ Aerosint ;
with editing and graphics help from Kevin Eckes