In this study, we examined the performance of two core-shell acrylic-based impact modifiers (AIM) prepared by emulsion polymerization. The rubber core was prepared from ethyl hexyl acrylate (EHA) and noctyl acrylate (n-OA). In such as process, the particle size and particle-size distribution of the modifiers were precisely controlled, so that performance differences observed in polybutylene terephthlate (PBT), used as matrix resin, Could only be interpreted in terms of the nature of the elastomeric component of the modifiers. When isolated, the rubber core of the modifiers showed identical glass transition temperatures (T-g) by differential scanning calorimetry (DSC) and dynamic mechanical analysis (DMA) despite the fact that they were made from two different acrylic monomers. Temperature-frequency superposition principle inferred from the classical WLF equation showed that the rubber components exhibit the same T-g at all frequencies including at the time scale at which mechanical impact typically Occurs. However, significant differences in low temperature impact performance measured at -30 degrees C using notched Izod impact test according to ASTM D 256 were obtained even though their rubber components had identical T-g Such differences were attributed to the dynamic relaxation behavior of the rubber components algid identified as inherent properties of the elastomers due to the structure of the monomers' repeat units. (c) 2007 Wiley Periodicals, Inc.