Advancing Vat Photopolymerization of Photosensitive Polyimides: Process-Structure-Property Relationships

Abstract

Vat photopolymerization (VPP) is a robust additive manufacturing (AM) technology which can produce products with high resolution features, smooth service finish, and at relatively high throughput. These advantages are leveraged in applications like jewelry manufacturing as well as dental and medical device production. The primary challenge that limits VPP's adoption in broader applications is its limited material selection. Typical photocurable resins amenable to VPP are brittle, have limited thermomechanical performance, and are highly flammable. Aromatic polyimides have found broad adoption in aerospace, automotive, and electronics applications due to their exceptional thermal stability, mechanical strength, and chemical resistance. Their excellent dielectric properties are particularly advantageous for use in flexible circuits, insulating films, and semiconductor manufacturing. However, their excellent properties also make them challenging to process. For this reason, they are typically only found in films, tapes, rods, and plaques. Combining the geometric freedom of VPP with the material properties of aromatic polyimides would open a wide array of new applications. Existing efforts to add photosensitivity to polyimide precursor resins have been limited by challenging synthetic steps, limited solubility, high viscosity, and significant part shrinkage. Prior to this work, the state of the art approach was a method termed the supramolecular salt or "polysalt" approach. This strategy utilized electrostatic interactions between small molecule precursors to aromatic polyimides that had been modified to photopolymerize. Following templating the small molecule precursors with VPP, thermal post-processing was used to convert the small molecules first into a higher molecular weight poly(amic acid), then a cyclized polyimide, and finally to degrade the photopolymer scaffold. The aim of this work is to build on the existing understanding of the polysalt approach to evaluate how the VPP process, resin structure, and final part properties interact. The morphology of polysalt parts is explored extensively throughout the thermal post-process . PMDA/ODA and PMDA/DDS polysalt parts are manufactured and thermally post-processed to temperatures ranging from 100 to 400 °C in vacuum to capture morphology snapshots before and after imidization as well as acrylate scaffold degradation. It is found that PMDA/ODA yields highly ordered crystalline morphologies and strong interchain interactions via the polysalt approach while PMDA/DDS does not. This increased order also yields greater thermal stability and reduced porosity in PMDA/ODA polysalt parts. This work also provides a potential framework for the design of new photosensitive polyimide resins with tailored properties. It is also found that dimethyl esters of dianhydrides can be used in aromatic polyimide resin systems (specifically PMDA/ODA and PMDA/DDS) to adjust the acrylate content in what is termed the "mixed polysalt" approach. While reducing acrylate content leads to reduced gel stiffness making printing more challenging, it also reduces the buildup of stress during shrinkage in post-processing, leading to reduced part cracking. PMDA/DDS with 50% reduced acrylate content and PMDA/ODA with 20% reduced acrylate content are shown to be printable with gel moduli of 45 kPa and 62 kPa respectively. Further, the decrease in acrylate content leads to less evolution of pore forming gases during acrylate degradation leading to a significant improvement in final part density. Characterization of these factors enables balancing of these tradeoffs to successfully manufacture PMDA/ODA lattices with 0.50% porosity and an HDT of 438.0 °C. Finally, this work extends the resin characterization efforts introduced in earlier chapters by establishing a statistical framework for analyzing working curves. F-tests are established as a means to test whether Jacob's equation describes the relationship between cure depth and exposure for a given resin. In cases where lack-of-fit is determined, it is shown that the prediction of values such as Ec and Dp from the working curve is unreliable. Error expressions for derived values of the working curve such as Dp and Ec are also established. The impact of experimental design considerations such as exposure range, exposure average value, and unexplained variations from experimental setup can impact the error expressions for these derived values. This statistical framework not only advances the rigor of resin evaluation but also provides practical diagnostic tools that complement the polysalt and mixed polysalt strategies, ensuring that new photosensitive polyimide resins can be both effectively designed and reliably characterized for advanced VPP applications.