The most common commercial processes for manufacturing prepregs for electronic applications use solvent-based resin systems. Solvents are not environmentally friendly and contribute to voids in prepregs and laminates. The resin impregnation process is performed in an open resin bath. This low-pressure impregnation is conducive to voids in prepregs. Voids cause product variability, which is a major source of scrap in board shops. To eliminate these drawbacks, a solventless process, based on the concept of injection pultrusion, has been developed. The impregnation is performed in a die under pressure to minimize voids. In previous work, chemorheological and kinetic measurements were used to identify a potential epoxy-based resin system. In addition, flow visualization with model fluids was used to establish the basic flow mechanism. Here we use the previous results to develop a mathematical model for the B-staging process. A prototype B-staging die has been built and used to verify the mathematical model. The results show that the model agrees well with the experimental data for low pulling speeds and slightly underpredicts the runs at high pulling speeds. The properties of the prepregs, the dielectric constant (DK) and dielectric loss (DF), have also been measured in this research. The measurements show that the solventless prepregs have acceptable DK and DF values according to the Institute for Printed Circuits FR-4 designation (a permittivity and tangent loss standard). A microscope has been used to observe the void contents of the prepregs. The solventless prepregs have been compared against standard FR-4 prepregs and shown qualitatively to have fewer voids. Based on the mathematical model, two potential process alternatives for the manufacture of solventless prepregs have been developed and analyzed. (C) 2003 Wiley Periodicals, Inc.