ENG-EAM: Parallax Additive Manufacturing of Sustainable Electrical Interconnects

This project will improve the fabrication precision of volume additive manufacturing by controlling residual stress through improved CAD tools, mechanical modeling, and materials. Volume additive manufacturing is a relatively new polymer additive manufacturing method that projects hundreds of three-dimensional light fields into a container of photosensitive resin. These light fields overlap within the resin, solidifying the part by exceeding a total exposure threshold in the desired shape. This process is orders of magnitude faster and provides better material uniformity than traditional layer-by-layer processes. However, parts fabricated by volumetric exposure have lower stiffness before post processing and are thus subject to greater deformation, reducing shape accuracy. This project will improve part fidelity by studying residual stress development, focusing on one volumetric manufacturing architecture, parallax volume additive manufacturing. Improved dimensional accuracy and reduced stress will extend the applicability of volumetric additive manufacturing to meet the needs of high precision, high volume industrial production such as high bandwidth electronic connectors. The hundreds of images projected into the resin container are found by solving a very large inverse problem using the mathematics of computed tomography. To make this problem computationally tractable, current algorithms ignore the inevitable stresses that develop during polymerization and post-processing steps. This project will build a finite element model as a digital twin, comparing this to the Virtual Volumetric Additive Manufacturing model created at Lawrence Livermore National Laboratory. Simplified models of viscoelasticity will be implemented to find a minimal description of the fabrication process. This model will be implemented in a new image generation algorithm that provides greater computational efficiency by representing fields in basis sets developed for computer image generation. These algorithms solve an analogous problem of mapping three-dimensional radiance fields to two-dimensional images with orders of magnitude efficiency gains. Those gains will be exploited here to incorporate more complex materials models while maintaining tractable computational cost. To validate this process, custom resins will be formulated with distinctive stress development characteristics. These will compare step to chain growth monomers to manipulate polymerization shrinkage and covalent adaptable networks to relax stress during post processing. The expected outcome of this program is a computational tool that optimizes image sets for final shape after post processing, significantly improving shape fidelity relative to current tools which optimize only for monomer conversion during exposure. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2430936 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Robert McLeod | Institution: University of Colorado at Boulder, Boulder, CO | Award Amount: $400,000 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2430936 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2430936.html