Oriented Quiescent Crystallization of Polyethylene Studied by USAXS.

Summary: In Part 1 of this series, the results of an in situ small‐angle X‐ray scattering (SAXS) study of polyethylene crystallization were presented. They showed that crystallite placement was basically a random process, from which some order is growing. This paper presents a first survey concerning the required change of paradigm. No distortion of a lattice is to be studied, but order grown on the nanometer scale must be distinguished from the stochastic case, i.e. the “car parking problem” from the field of random sequential adsorption (RSA). RSA is explored by computer simulation. The results concerning the corresponding liquid scattering are required to identify short‐range order (cf. Part 3). Processing of the simulated scattering patterns verified that the features of quasi‐random arrangement are preserved in the interface distribution function (IDF), if only the crystals are shielded by some transition layer. In a condensed random nanostructure, almost only next neighbor correlations are present. The distribution of the widths of amorphous gaps between the crystals, h a , is a truncated exponential distribution. In the scattering pattern the stochastic nanostructure can hardly be distinguished from a system with short‐range order. On the other hand, in the IDF the features of order become clear. In random systems there is no convolution polynomial based on crystalline and amorphous distributions. Only two shifted images of h a are occurring. We find that packing correlations collapse, if the crystallite thickness distribution is wide enough. In this case the pure particle scattering is a fair approximation in the IDF. This criterion is, in general, valid for technical polymer materials.IDFs computed from the nanostructure obtained in a computer simulation of ideal random isothermal crystallization after infinite time (random car parking process). Bold line: No transition zone. Thin line: Transition zone introduced ( $\mathop {\bar p}\nolimits_t /\mathop {\bar p}\nolimits_c = \mathop {\bar t}\nolimits_t /\mathop {\bar t}\nolimits_c = 0.2$ ). Dashed‐dotted line: A Gaussian crystalline layer thickness distribution with a standard deviation of σ t  = 0.2.