Data-Driven Design Space Analysis for Multimaterial Thermoplastic Composites Manufactured by Fused Filament Fabrication

Seda Oturak, Melinda McKeehan, Callie Zawaski, Wesley F. Reinhart

Research output: Contribution to journalArticlepeer-review

Abstract

Multimaterial additive manufacturing has enabled the fabrication of components with highly tailored mechanical responses. However, as both manufacturing processes and constituent materials become more sophisticated, the large number of process variables makes design increasingly challenging. Here, we investigate the design space for multimaterial thermoplastic composites produced by a commercially available fused filament fabrication printer. We consider the uniaxial compression of neat materials with sample geometry and toolpath variations, as well as composites comprised of a soft elastic matrix with stiff reinforcement material in different reinforcement fractions and geometries. We find that some changes to the toolpath can have a significant impact on the compressive behavior of the samples due to the high anisotropy of the filaments. The composite geometries were found to exhibit different specific strengths relative to reinforcement fraction, and their compressive behavior matched qualitatively but not quantitatively to predictions from finite element analysis. To understand the performance space, we analyzed the experimental dataset with truncated singular value decomposition. Surprisingly, despite the complexity of the system, 97.8% of the variance in stress-strain curves of our samples was captured by the first component and 99.9% by the first two. The shape of the components indicates that while the stress-strain curves of samples may vary quantitatively, very limited modes are controllable with the design variables considered here. In effect, the strength of the composite could be controlled by manipulating reinforcement mass fraction, but the shape of the nonlinear behavior was largely baked into the constituent materials despite changing the reinforcement geometry. This result has important consequences that must be considered early in the design process when developing new multimaterial systems to achieve tailored mechanical responses.

Original languageEnglish (US)
Journal3D Printing and Additive Manufacturing
DOIs
StateAccepted/In press - 2024

All Science Journal Classification (ASJC) codes

  • Materials Science (miscellaneous)
  • Industrial and Manufacturing Engineering

Fingerprint

Dive into the research topics of 'Data-Driven Design Space Analysis for Multimaterial Thermoplastic Composites Manufactured by Fused Filament Fabrication'. Together they form a unique fingerprint.

Cite this