Every week it seems as though a new additive manufacturing (AM) machine (or 3D printer) is being announced to the world, leading to a flurry of coverage over what the machine is claimed to do better, cheaper, faster, bigger than existing technology. Whatever the truth of these claims it is clear that current technologies are far from perfect for every application.
Real challenges must be overcome for AM to progress from a niche to mainstream production process.
When it comes to the manufacture of complex metal parts the predominant AM technology is powder bed fusion (PBF) where a high power laser or electron beam is used to selectively fuse a thin layer of powder. Although this approach enables the direct production of highly complex parts the speed is currently limited to a few hundred grams per hour which, in turn, increases the part cost. This is a significant barrier to widespread adoption of the technology.
Recently there has been a noticeable increase in the number alternative approaches to PBF for the production of complex metal parts. Common to these process is the use of a sacrificial binder material to temporarily bond the metal particles to form the shape of the parts before being subjected to a post-processing sintering process to reach full density. Decoupling the shape forming step from the powder sintering provides the potential for much higher productivity.
– Metal binder jetting, where a loose metal powder bed is bound together using a binder agent, e.g. Digital Metal, Desktop Metal (‘Production’ system), ExOne, GE Additive, HP*, Stratasys*.
– Metal material extrusion, where a filament or rod (combination of binder and powders) is extruded from a heated nozzle, (often called fused filament fabrication or fused deposition modelling), e.g. Desktop Metal (‘Studio’ system), Markforged Metal X.
– Metal vat photopolymerisation, where a photocurable resin loaded with powder particles is cured using a laser or other light source (often called stereolithography), e.g. Lithoz.
– Metal material jetting, where ink droplets loaded with powder particles are jetted from an array of nozzles, e.g. XJet
*Assumed to be binder jetting based on snippets of information released.
Following the build process itself, the 3D printed ‘green’ parts undergo a post-processing procedure consisting of a pretty conventional set of steps similar to that used for metal injection moulding (MIM), where the binder is removed and the resulting ‘brown’ part is densified in a sintering stage. The post-processing of AM parts is one of those under-appreciated realities (or ‘dirty little secrets’) of the technology and is not unique to ‘bind-and-sinter’ AM. PBF parts also require a significant amount of post-processing in order to obtain a finished part, and reducing the effort required for this across all AM processes is an active area of research.
A key question when looking at this approach of manufacturing metal parts is: how does the part density, microstructure and material properties compare with PBF processes? To gain insight into this so that the technology can be applied to real applications, the MTC conducted a project with its members which included evaluation of two ‘bind-and-sinter’ AM processes: the Digital Metal binder jetting system and the Lithoz metal stereolithography system. Both systems were found to achieve a density of around 95-97%, which can be compared with a typical density of >99% with PBF., Hot isostatic pressing (HIP),where parts are “squeezed” under high temperature and pressure, can be used to fully densify parts.
Unlike PBF, in metal binder jetting (and metal stereolithography) a support structure is not required which opens up the range of geometries that can be manufactured. Supports in PBF often need to be reduced through changes in the part geometry and any remaining supports need to be removed at the end of the build – a slow, manual process which add cost and time as well as wasting material. Moreover, the presence of supports inhibits the ability to stack parts on top of each other in the machine.
In addition, the increased design freedom can mean metal binder jetting is the only way to achieve the required geometry and some example parts made using the four ‘bind-and-sinter’ AM processes are shown below. Some machine manufacturers have targeted large parts (e.g. GE Additive and ExOne), whilst others have targeted small parts (e.g. Digital Metal). Others have prioritised build speed such as the Desktop Metal ‘production’ system due to be released in 2019.
For PBF the material needs to be weldable. Moreover, difficult-to-weld materials often require changes to the process, including preheating. With ‘bind-and-sinter’ AM, the material range can be much wider and includes other materials such as ceramics. However, there are some challenges with using ‘bind-and-sinter’ AM. The thermal treatment process required for the part to reach full density increases the lead time compared to PBF and requires furnaces that are capable of handling organic residues of extracted binders. In addition, the parts are relatively fragile during the intermediate ‘green’ and ‘brown’ stages, which makes them prone to failure during handling or de-powdering. Also, the sintering process could cause deformation to particular geometries, such as slumping of unsupported structures, which makes it necessary to design innovative support solutions during sintering. Shrinkage is generally repeatable and can be compensated for, although not necessarily isotropic, and for some geometries temporary ceramic supports may be required to achieve the required geometric tolerances.
In terms of potential application areas, bind-and-sinter AM technologies are likely to find use in industries where small, highly complex parts with tight tolerances are required on a batch production scale. These include traditional MIM sectors, such as medical equipment, automotive, portable electronics and luxury watches, glasses and other consumer goods. Considering the difference in production scale capabilities, binder-and-sinter AM is likely to complement MIM rather than compete with it, as current MIM producers already have the infrastructure required to process green parts and will see the advantage of using AM for producing small batches or prototypes before investing in expensive MIM tools for large scale production. Binder-and-sinter AM surpasses MIM in terms of higher geometrical complexity and design flexibility, which will enable it to penetrate new application areas that are currently not viable for MIM.
So to summarise, ‘bind-and-sinter’ AM has some characteristics that are different to PBF which can make AM applicable to a wider range of components, with a particular intention of increasing component geometric complexity, increasing build speed, reducing costs and broadening the range of materials available. As the home for the National Centre for AM, the MTC has recently installed a Digital Metal binder jetting system and it is inviting companies that require small and highly complex metal parts to get in touch to see the potential this process offers for their business.