You are here

Methods to rapidly optimize materials for Additive Manufacturing processes


OBJECTIVE: Develop a model-guided experimental method that will determine an optimal set of Additive Manufacturing process control parameters for any given powder sample. DESCRIPTION: Additive Manufacturing (AM), the set of related techniques that sequentially fuse layers of material to one another in order to build up a part, are rapidly maturing aerospace manufacturing methods whose key strengths are in minimizing the cost, and often more importantly, the time needed to transition from a CAD design to an end-use part. The lack of required tooling, design flexibility, and high part quality are changing the way companies approach the production of low volume or high value parts. A key challenge that remains to realizing the full potential that AM processes offer is the relatively small number and high cost of feedstock materials suitable for use in additive manufacturing. The cost of the AM optimized powder feedstock is often as much as an order of magnitude higher than the bulk commodity version of the materials, particularly in plastics AM. Due to the high cost and low availability, powder-based AM parts manufacturers are currently dependent on a handful of feedstocks which were often optimized for an entirely different process, such as powder metallurgy or powder coating. The morphology, crystallinity, and size distribution of these powders are rarely optimal for the AM process, and a costly, iterative, and empirical process is required to determine if a new feedstock is even able to produce high-quality parts via AM. For powders of the same alloy or molecular composition, differences in the shape and size distribution of the powder particles can dramatically change the process control parameters that need to be employed in order to achieve high-quality parts. The selection of appropriate process control parameters becomes even more complicated when fillers and fibers are added to the base powder to improve or modify the bulk material properties delivered in the end use part. For powder bed AM processes such as Electron Beam Melting (EBM), Direct Metal Laser Sintering (DMLS), and Selective Laser Sintering (SLS), excellent powder flow and high packing density are essential when applying fresh powder layers to the powder bed to ensure uniform and consistent part production. For powder feed processes such as Laser Engineered Net Shaping (LENS), stable powder flow ensures that parts are uniformly built. What is needed are Integrated Computational Materials Engineering (ICME) tools to help rapidly screen new materials by employing model-guided Design of Experiment (DOE) approaches to rapidly find an optimal set of process control parameters for a given AM process. It is anticipated that novel approaches will be needed to create these ICME tools as off-the-shelf continuum fluid dynamics models fail to adequately capture important powder based phenomena such as arching, friction, segregation, stick-slip dynamics, and dilatancy. In particular, it will be essential to model the discrete granular nature of powders and the complex, collective behavior that they exhibit, including densification mechanisms, densification kinetics, and powder-beam interactions. Beyond the added flexibility of having a larger palette of materials to choose from, and the potential for greatly reduced costs, better understanding of the granular processing properties of powder AM feedstocks will deliver other measurable benefits such as decreasing part porosity, maximizing the recyclability of used powder, and reducing anisotropy introduced due to the layering process that most AM processes employ. PHASE I: Demonstrate proof-of-concept capability that integrates (metals and/or polymer) powder material properties, as well as shape and size distribution information, into a model that predicts AM processing properties including, but not limited to, powder bed density, powder flow, thermal conductivity, and laser/powder interaction. Particular attention should be given to the validation plan for the model. PHASE II: The capability developed in Phase I would be extended to include prediction of (metals and/or polymer) powder specific optimal process control parameters for a selected AM process. Demonstrate the capability to improve the processing of a current AM powder feedstock due to variations in feedstock uniformity. This will be based on the developed model or develop an optimized process control parameter set for a powder feedstock that has not been utilized for AM previously. PHASE III: Technologies should be directly implemented into AM production lines or licensed to AM machine vendors to improve the dimensional accuracy, performance, cost, and quality of production parts and offer a greatly expedited capability for introducing new materials into the appropriate AM process. REFERENCES: 1. B. Liu et al."Investigation of the effect of particle size distribution on processing parameters optimization in selective laser melting process"Proc of Solid Freeform Fabrication Symposium (2011) 227-238. 2. R. McVey et al."Absorption of laser irradiation in a porous powder layer"J Laser Applications 19 (2007) 214-224. 3. A. Simchi"Direct laser sintering of metal powders: Mechanism, kinetics, and microstructural features"Mat Science and Engineering A 428 (2006) 148-158.
US Flag An Official Website of the United States Government