Dynamical and structural aspects of the cold crystallization of poly(dimethylsiloxane) (PDMS)

A detailed study of the dynamical and structural aspects of the cold crystallization behavior of highly flexible poly(dimethylsiloxane)(PDMS) is presented. In order to understand the complete behavior, before, under and after crystallization, a wide range of experimental techniques have been employed. A particular emphasis is made on the interplay between dynamical and structural effects and how the properties of the amorphous phase evolve during the crystallization. This is highlighted by combining mobility sensitive techniques, involving broadband dielectric spectroscopy (BDS) and differential scanning calorimetry, with neutron scattering: wide and small-angle neutron scattering (WANS/SANS) which are sensitive to the relative ordering of the atoms. In this way, we are able to compare the structure associated with crystal formation with the evolution and modification of the amorphous phase. The kinetics deduced from WANS points toward a classical nucleation and growth behavior closely following a Avrami-like growth with an exponent of about n = 3 which is expected for athermal nucleation from fixed centers followed by three-dimensional crystal growth. Furthermore, the amorphous phase (deduced from BDS) decays in parallel with the emergence of the crystalline phase (from WANS/SANS) without any shift in the characteristic relaxation time. However a careful comparison of the crystallization at short times indicates that the amorphous phase seems to be affected before any measurable crystallization is detected by WANS. Although this might be compatible with the existence of mesomorphic phase, it may also be attributed to more simple precursors as initial crystalline "baby-like" nuclei. In this picture, these crystalline nuclei may be formed homogeneously in the system which in turn causes a constraint on the surrounding chains connected to these crystalline nuclei. This is manifested as a distinct relaxation contribution that is drastically slower and heterogeneous than the conventional amorphous (x-relaxation of the melt. It would also explain a signal in SANS before any accompanying crystallization signal in WANS. Once the crystal starts developing, the fraction of the slower amorphous phase (constrained amorphous phase, CAP) grows and the conventional amorphous phase gradually disappears. At the very end the growing crystalline fronts start to overlap and some of the remaining CAP becomes even more constrained due to cross-link strongly manifested in both the dielectric response and the heat capacity.