Simulation of the Optimized Thermal Conductivity of a Rigid Polyurethane Foam during Its Foaming Process

Abstract

Thermal conductivity (λ) of a rigid polyurethane (PU) foam plays a vital role in determining the performance of thermal insulation. To accurately predict this property for a given recipe in a few seconds is the first step towards striking a balance between compressive strength, λ, and more properties of a foam product in industrial applications during formulation. What experimentally observed from PU closed-cell foams made from PAPI-27 Polymeric Methylene Diphenyl Diisocyanate (PMDI), Voranol 360, and n-pentane were: (1) foam shrinkage and even collapse occurred on both low (typically less than 0.9) and ultra-high (close to 2.0) indexes of isocyanate (I_A); (2) to achieve the lowest λ for the best thermal insulation, it was suggested to use the highest I_A without any foam shrinkage, the fastest stirring rate under safety, and the least physical blowing agent (PBA) until a sharp turn in the trend of λ; (3) there was a definite link between foam density (ρ_f), closed cell content (ccc), and λ: (a) a low λ requested a high ρ_f and ccc (b) a denser foam could be achieved by at least three ways. Increasing the agitation rate had a limit, above which the change in density became insignificant. However, a higher I_A and reducing the amount of PBA had always been linear with ρ_f. (c) ρ_f was the cornerstone of ccc. In low ρ_f range, a growth of density gave rise to a distinct improvement of ccc, beyond which ccc reached a plateau for any higher density. The foam thermal conductivity during polymerization was successfully simulated by Matlab with 0.0157 % deviation. λ was initially dominated by the contribution from the resin mixture until the gel reaction time. Then, the dominant was shifted to the contribution by the gas component. The magnitude of thermal radiation was rather low throughout the reaction.