The simplest nematic liquid crystal phase (N, Fig 1a) is a three-dimensional orientationally ordered liquid. The local average direction of the molecules is denoted by the director n, andcharacteristic for the N phase is thateven if the constituent (usually rodlike) molecules may be polar themselves, n is invariant under sign reversal, i.e., n = -n. The structure and non-polar symmetry of the N phase is well-known since more than a century and the electrooptic effects of nematic liquid crystals are the basis for the multi-billion dollar liquid crystal display technology.
In 2020, Clark and co-workers reported the first experimental evidence of a new liquid crystal phase [i], the ferroelectric nematic (NF) phase, cf. Fig 1b, in a material (RM734) first synthesized in 2017 by Mandle, Cowling and Goodby[ii],[iii]. The possible existence of such a spontaneously polar nematic phase, driven by dipolar ordering of molecular dipoles in highly polar materials, was proposed already by Born in 1916 [iv], but evidence for its existence has not been presented before. The discovery of the NF phase “opens another chapter in condensed matter science and technology” (Lavrentovich [v]) and has sparked an enormous interest, both from fundamental physics and from applicational points of view.
In order to understand the mechanisms behind the formation of the new NF phase, and to explore its future application potential, there is an immediate need to design, synthesize and investigate a vast number of similar materials, and rapidly scan these for the NF phase. In this letter I present a surprisingly simple and fast method for direct experimental identification of the NF phase, utilizing only a single sample cell and an optical microscope (or even the bare eye). No external electric or magnetic fields, no second harmonic generation experiments, nor any spectroscopic methods are needed. The study solely treats the experimental identification of the NF phase, with no ambition at this stage, to further elaborate on the mechanisms behind the formation of the NF phase.
In a nematic liquid crystal display cell the quiescent state is governed by the boundary conditions at the two cell substrates. For homogeneous planar anchoring of n mechanically rubbed polyimide layers can be used, forcing n to align with the rubbing direction at the surfaces. In the conventional N phase, parallel, as well as antiparallel, rubbing at the two surfaces produce a non-twisted director field, Fig 2a, and if the rubbing directions are not colinear we get a director twist between the plates.
For the polar NF phase the situation is fundamentally different: on a rubbed surface the NF phase aligns with the spontaneous polarization parallel to the rubbing direction[vi]. The striking consequence is that when the rubbing directions are anti-parallel, we get a ±180° director twist between the surfaces, cf. Fig 2b. When the angle between the two rubbing directions is neither 0 nor 180°, the NF phase adopts the state with the smallest twist matching the rubbing directions. When the two surfaces are rubbed in parallel we get a non-twisted state (just as in the N phase).
Let us now make a cell where the bottom surface is linearly rubbed, and the top surface is circularly rubbed. Such circularly rubbed cells (CRCs) have been used by Suh et al. to measure the pitch in long pitch cholesteric materials[i]. Linear alignment is obtained by gently rubbing the polyimide coated surface with a velvet cloth in a single direction. Circular alignment can be realized by putting the polyimide-coated substrate on a spinner chuck and gently press the rubbing cloth towards the spinning substrate for a few seconds. Alternatively, a small piece of rubbing cloth can be attached to a flat rotating chuck, that is gently pressed towards to a then fixed substrate. In a CRC the N phase adopts the configuration illustrated in Fig 3a7. If we use a clockface notation and the bottom substrate is rubbed along the 3 -9 direction, we get a non-twisted director field at “12” and “6”. Away from the “12” and “6” positions we get an increasing twist and at 3 and at 9 the twist
changes sign, to minimize the twist elastic energy. The discontinuous change from +90° to -90° twist occurs across two sharp disclination lines running horizontally, i.e. in the “9”-“3” orientation through the center of circular rubbing.
Let us now consider the case of the NF
phase in the same cell, cf. Fig. 3b
. The circular rubbing is made counterclockwise, and the rubbing directions are therefore parallel at “12” and antiparallel at “6”. The polar NF
phase orients with the polarization along the rubbing directions and to match the boundary conditions, we expect a non-twisted director at “12” where the rubbing directions are in parallel. When going away from “12” towards “6”, clockwise or anticlockwise, there is a twist that increases continuously all the way to the “6” position, where the rubbing directions become perfectly antiparallel. To minimize the elastic energy, the twist discontinuously changes from 180° right-handed to 180° left-handed at “6” across a disclination line, while the structure is continuous at all other positions in the cell. To summarize: in the CRC cell the N phase should make two disclination lines along the linear rubbing direction
, while the NF
should make one disclination line normal to the linear rubbing direction
, starting from the center of rubbing.
Fig. 4 shows polarized optical microscope photographs of a 9mm thick CRC filled with the archetype NF material RM7341,2, having a phase transitions between the ferroelectric NF and the conventional N phase at about 133°C, and between N and isotropic at 139°C, respectively. The polyimide aligning layer on both surfaces is PI2610 from DuPont, and the bottom linearly rubbed substrate is the warmer substrate (in direct contact with the hot-plate). The cell was capillary filled with RM734 in the isotropic phase at 150°C. At 140°C (a and l)) we have two horizontal disclination lines emanating from the center of circular rubbing, as expected from the non-polar N phase. At “12” and “6” on the clockface we get extinction of light as the director here is parallel to one of the crossed polarizers. At 130°C (f) the two horizontal disclination lines have been replaced by one disclination line at “6” which proves that the sample is now in the NF phase, cf. Fig .3. Figs. 4a-f shows the evolution of the cell on cooling from 140° to 130°C at a rate of 3°C/min. Soon after the phase transition (b) the two disclination lines at “9” and at “3” start to move downwards and become one disclination at “6”. On heating from 130°C to 140°C (Figs. 4g-l) the single disclination line at “6”, soon after the transition to the N phase (h), splits close to the center of circular rubbing and branches symmetrically into the two horizontal disclinations characteristic of the N phase at the “9” and “3” positions (l). The course of events shown in Fig 4 is perfectly reversible.
To conclude, circularly rubbed cells and the position of the disclination line(s) in relation to the linear rubbing direction comprise a simple and yet powerful tool to probe the polar order of nematic liquid crystals, and in particular to unambiguously distinguish the NF phase from the N phase.
The data that supports the findings of this study are available within the article.