Image processing of elastomer blend morphology

An image processing technique has been developed to quantitatively analyze elastomer blend morphology. In this case, the Scanning Electron Microscope (SEM) images of surfaces obtained at various intervals of mixing of binary blends of polyisoprene (IR) and ethylene propylene monomer (EPM) elastomers in a laboratory internal mixer were represented in terms of closed set of orthogonal polynomials. The significant orthogonal effects were measured an d combined to represent the local texture i.e., pronum and global descripto r, prospectrum which is the frequency of occurrence of pronum. Various stat istical parameters were calculated from prospectrum and correlated with ble nd morphology. Time of attainment of final phase morphology beyond a period of mixing for 5 min, was ascertained from this imaging technique. Further, the blends were also annealed for different periods of time up to 20 h. and the SEM pictures, were examined by this method.