Chinese Acad Sci, Inst High Energy Phys, Beijing Synchrotron Radiat Lab, Beijing 100039, Peoples R China
; China Petr & Chem Corp, Beijing Res Inst Chem Ind, Beijing 100013, Peoples R China
The three-phase structure model of crystal line polymer is used to calculate the one-dimensional electron density correlation function. The conclusion is that the definitions of lamellar thickness and invariant are not the same as those in the two-phase structure. Then utilizing this result to analyze some specimens, we found that the crystalline transition-layer thicknesses obtained by the correlation function method agree with those by the Ruland method, but the long periods obtained by correlation function are much different from the Bragg spacings. The reason is that the two methods to calculate crystalline transition-layer thicknesses are all based on the non-homogeneous area of the specimen electron density and have the same nature. However, the Bragg spacing and the con-elation-function long period represent different regions of the non-homogeneous area.