Structures and responses of ferroelectric liquid crystals in the surface-stabilized geometry

Abstract

We present experimental and theoretical studies of the structures and transitions of the molecular orientation field of a smectic C* liquid crystal in the surface-stabilized geometry. The experimental results show that structures having from two to six stable orientation configurations are possible. The optical properties of these states, electric-field-induced sequences of transitions between them, and defects which appear in them are analyzed to infer the states' structure. These studies show that in the surface-stabilized geometry: (1) the smectic layers can be tilted relative to the bounding plates; (2) the boundary conditions can stabilize molecular orientations tilted relative to the bounding plates. Evidence is also presented for the coupling of smectic layer deformation and the director field.The theoretical results show that the simplest models of the interactions of a ferroelectric liquid crystal with bounding surfaces and applied electric fields explain a variety of phenomena. First, a simple, low-order surface interaction preferring molecular orientations parallel to bounding surfaces produces an unwinding of the intrinsic helix when the bounding plates are spaced closely enough. The resulting unwound liquid crystal exhibits two states with the ferroelectric polarization uniformly perpendicular to the bounding plates. A third state with the polarization splayed across the sample thickness is also possible, and is stabilized either by interactions that favor a specific orientation of the polarization to the surface normal or by an intrinsic elastic distortion complementary to the one that produces the helix. The addition of the simplest coupling of the ferroelectric dipoles to an applied electric field predicts first-order transitions between the various surface-stabilized states. The transitions proceed by the motion of domain walls, whose structure may be calculated in the most simple instance. In the absence of predicted domain wall structures, dimensional arguments provide scaling behaviors for the dynamics of domain wall motion that agree with experiment.