Defect structure correlation length

Systematic studies of well-defined materials in which specific structural variations have been made, provide the basis for structure/property relationships. These variations may include the effect of charge, hybridization, delocalization length, defect sites, quantum confinement and anharmonicity (symmetric and asymmetric). However, since NLO effects have their origins in small perturbations of ground-state electron density distributions, correlations of NLO properties with only the ground state properties leads to an incomplete understanding of the phenomena. One must also consider the various excited-state electron density distributions and transitions. 

The high temporal resolution of SFM imaging uncovered elementary dynamic processes of structural rearrangements. We observed short-term interfacial undulations [111], fast repetitive transitions between distinct defect configurations [112], their spatio-temporal correlations on a length scale of several microdomains [112], and unexpected defect annihilation pathways via formation of temporal excited states [51, 111]. 

The description of a network structure is based on such parameters as chemical crosslink density and functionality, average chain length between crosslinks and length distribution of these chains, concentration of elastically active chains and structural defects like unreacted ends and elastically inactive cycles. However, many properties of a network depend not only on the above-mentioned characteristics but also on the order of the chemical crosslink connection — the network topology. So, the complete description of a network structure should include all these parameters. It is difficult to measure many of these characteristics experimentally and we must have an appropriate theory which could describe all these structural parameters on the basis of a physical model of network formation. At present, there are only two types of theoretical approaches which can describe the growth of network structures up to late post-gel stages of cure. One is based on tree-like models as developed by Dusek7 I0-26,1 The other uses computer-simulation of network structure on a lattice this model was developed by Topolkaraev, Berlin, Oshmyan 9,3l) (a review of the theoretical models may be found in Ref.7) and in this volume by Dusek). Both approaches are statistical and correlate well with experiments 6,7 9 10 13,26,31). They differ mainly mathematically. However, each of them emphasizes some different details of a network structure. 

---