Orientations and phase transitions in liquid crystals consisting of short linear polymer chains

The system of semiflexible linear polymer liquid crystal (PLC) macromolecul es is studied. Each macromolecule constitutes an alternating chain of flexi ble (F) and rigid (LC) sequences. The distribution function of the chain co nformations is factorized in three terms. The Gibbs distribution is used fo r anisotropically interacting LC sequences; products of the Dirac delta fun ctions represent F sequences modeled by linear chains of freely jointed sta tistical segments; connections of LC and F sequences in a linear chain are controlled by the Dirac delta functions with a proper argument. The general formula for the Helmholtz function A for arbitrary types of anisotropic in teractions between LC sequences and for an arbitrary number of statistical segments per flexible part of linear chain obtained by the present authors [J. Chem. Phys. 105, 4367 (1996)] is applied in numerical calculations perf ormed for some special cases. The cases selected here are (a) the Maier and Saupe mean-field limit formula for LC+LC interactions; and (b) the linear approximation with respect to the inverse value of the total length of the chain flexible part. Mostly one deals with very large (infinite) chain leng th with Gaussian behavior and earlier we have done this also. In this work we investigate the non-Gaussian case with a finite number of statistical se gments per chain. The resulting modifications of the phase diagrams and pha se transition points are discussed taking also into account results of the numerical calculations. (C) 2001 American Institute of Physics.