Experimental and simulation study of resin infiltration in carbon fiber rovings

TL;DR

This study investigates the dynamic wetting of carbon fiber rovings by an epoxy resin through a combined experimental and mesoscale modeling approach. Optical measurements capture velocity-dependent advancing contact angles and capillary rise inside the roving, while mesoscale geometries inform a two-phase Darcy framework to compute capillary pressure and relative permeabilities. The results show that roving microstructure and porosity evolution, influenced by sizing and fiber arrangement, primarily govern resin impregnation, with dynamic contact angles playing a secondary role in the capillary-rise dynamics under the tested conditions. Overall, the work provides a quantitative link between microscale roving geometry and macroscale infiltration behavior, offering a pathway to optimize manufacturing processes and improve composite quality.

Abstract

Continuous-fiber-reinforced polymers are vital for lightweight structural applications, where fiber-matrix wetting critically influences composite performance. Understanding dynamic wetting behavior during resin infiltration remains challenging, especially under realistic processing conditions. Here we investigate the dynamic wetting of commercial carbon fiber rovings by a commonly used epoxy resin through combined optical experiments and two-phase flow simulations informed by microscale roving geometries. We measure velocity-dependent advancing contact angles and observe roving geometry changes during capillary-driven resin infiltration. Simulations using microscale-derived capillary pressure and permeability parameters quantitatively reproduce the time-dependent infiltration dynamics. Our findings demonstrate that while dynamic contact angles vary with velocity, the microscale roving structure predominantly governs resin impregnation behavior. This integrated experimental and modeling approach enhances insight into fiber-resin interactions, offering a pathway to optimize composite manufacturing processes and improve material quality.

Figures (16)

• Figure 1: The roving and the resin bath were homogeneously illuminated through a light source (KL 1500, Schott, Germany) and a diffuser sheet. We controlled the velocity of the roving with a dip coater with velocities up to 1.4m m s.
• Figure 2: Comparison of roving geometry, roving integrity as well as single fiber surface morphology and coverage by sizing of unsized and differently sized samples including fibers from a textile.
• Figure 3: Average roughnesses Ra for unsized and differently sized fiber samples.
• Figure 4: Cross sectional images of carbon fiber roving EP-R at different positions (see figure \ref{['fig:cross-sections']}), analyzed for area fraction of the resin (porosity of the roving).
• Figure 5: Viscosity vs. time of a 100:30 mixture of RIMR135 and RIMH137 in a temperature profile that was inspired by a typical application situation. The two overlapping curves indicate the reproducibility of two independent measurements.