Predictive Multiphysics for Reaction-Programmed Support-Free Direct Foam Writing of Thermosets

Mahdi Baniasadi1, Professor Mostafa Yourdkhani2
1School of Manufacturing Systems and Networks Arizona State University, Polytechnic Campus, Mesa, AZ, USA
2Associate Professor School of Manufacturing Systems and Networks Arizona State University, Polytechnic Campus, Mesa, AZ, USA
Published in 2026

Lightweight polymeric cellular solids are central to transportation, impact protection, thermal isolation, packaging, soft robotics, and biomedical systems, yet conventional foam manufacturing remains constrained by molds, multistep processing, high external energy demand, and limited geometric freedom. Additive manufacturing has enabled architected cellular materials by defining macroscopic porosity through printed lattices; however, direct, rapid, and energy-efficient generation of material-level micro- to sub-microscale porosity during printing remains difficult. Existing approaches often separate shape formation, foaming, and solidification into sequential or power-intensive steps, limiting process speed, structural fidelity, and requiring intensive support for complex 3D geometries. Here, we introduce reaction-programmed direct foam writing as a single-step, energy-efficient manufacturing strategy that couples frontal polymerization with in-situ tandem foaming to fabricate thermoset cellular solids. In this process, a self-propagating exothermic reaction front simultaneously polymerizes a deposited thermoset ink and activates endothermic expansion of embedded thermally expandable microspheres. This internally powered coupling transforms a reactive ink into a mechanically stable closed-cell foam at the reaction front, enabling rapid, support-free fabrication of freeform (mid-air) architectures without continuous external heating, ultraviolet irradiation, support materials, or post-curing. To make this reaction-programmed process predictive and designable, we developed a 3D deposition-aware thermo-chemo-mechanical multiphysics model in COMSOL Multiphysics. The framework resolves an intrinsically highly coupled transient system in which 3D material deposition, frontal polymerization, heat transfer, microsphere activation, density evolution, cure-dependent stiffening, foaming-induced expansion, and gravity-driven deformation occur simultaneously within a moving and continuously growing domain. The model couples Heat Transfer in Solids, Solid Mechanics, Domain ODEs and DAEs for Arrhenius-type autocatalytic cure kinetics with mobility suppression, and Coefficient Form PDEs for deposition-path tracking. A deposition indicator activates the governing equations only in regions reached by the moving nozzle, directly coupling deposition history to reaction-front propagation, foaming, and mechanical stabilization. The formulation captures conduction, convection, internal polymerization heat generation scaled by monomer fraction and microsphere-related endothermic heat absorption, temperature-, conversion-, and composition-dependent stiffness, and foaming-induced inelastic expansion. A curvilinear coordinate system along the printing path, combined with a local cylindrical frame, links density reduction to radial filament expansion while preserving axial deposition history. This model was validated against synchronized optical and thermal imaging experiments across microsphere concentrations in reactive DCPD micro-composite inks. Simulations reproduced reaction speed and temperature, density evolution, foaming expansion, reaction-front localization, stiffness evolution, and degree of cure across substantial porosity variation. Beyond the validation and prediction capabilities of the mentioned variables, the framework predicts mechanical performance and stability for unsupported filaments during printing for large overhangs and complex 3D helical structures, with precise prediction of deflections below 500 µm at 120 mm unsupported length. By tuning thermal conditions, printing parameters, microsphere chemistry, and particle concentration, the process enables programmable control over density, feature size, fabrication time, and structural fidelity, with densities ranging from ~1.03 to ~0.17 g cm^-3, corresponding to ~84% porosity. This digital process-design platform enables prediction and design of graded density, spatially programmed porosity, expansion, and geometry, guiding rapid, energy-efficient, support-free manufacturing of lightweight thermoset foam architectures.

Download