A prediction model using response surface methodology based on cell size and foam density to predict thermal conductivity of polystyrene foams

Abstract

Polymeric foams are one of the most important insulating materials due to their low thermal conductivity. The comprehensive investigation of the thermal behavior of these materials is experimentally difficult, costly, and time-consuming. In this study, a step-wise approach is used in order to predict thermal conductivity of the polymeric foams. The results demonstrate that two different models estimate the thermal conductivity of the polystyrene foams with cell sizes smaller than 100 μm and larger than 100 μm with an error almost smaller than 10%. According to these theoretical models, a comprehensive investigation is performed on the thermal-insulation performance of polystyrene foams. The results indicate that the thermal conductivity reduces by decreasing the cell size but there is an optimum foam density for achieving the smallest thermal conductivity. The effects of foam density and cell size are studied on the different mechanisms of thermal conductivity. In the following, a new approach is offered based on the cell size and the foam density using response surface method (RSM). The reliability of the proposed regression model is checked not only using analysis of variance (ANOVA) tool of RSM but also in comparison to empirical results.