First, the main aspects of liquid crystal materials are discussed. The shape of the molecular structure determines whether the material has an LC phase. The LC phase material structures with dipole moments form an electromagnetic field in the volume of the LC mixture and interact with it. In the nematic LC phase, the dipole moment, μ L C , is directed along the molecule. In the smectic ferroelectric LC phase, the dipole moment is transverse to the molecule and is oriented within the smectic layer.

In nematic LC order parameters, P ₁ = 0 and P ₂ = 0.4–0.8, while in ferroelectric LC order parameters, P ₁ ≠ 0 and P ₂ ≠ 0. Further on, we only consider nematic liquid crystals.

LCs also exhibit elastic orientation properties. When trying to locally change the direction of orientation of molecules, there are forces that return the molecules to their average orientation state. These properties are described by the elastic theory of liquid crystals (Frank, 1958).

In free space, LC molecules are oriented randomly. The external impact (such as the electric or magnetic field) can orient LC molecules in one direction. In the case of the external field, the influence occurs on each of the molecules, and they are all oriented in the same direction given by the field.

Surface LC alignment is the interaction of the liquid crystal continuum with the bounding surface, leading to the orientation of the LC director at the interface in the direction n → θ , φ . It has the meaning of the boundary condition for minimizing the total energy of liquid crystal elastic deformations. This interaction is quantified in terms of the LC deformation energy at the surface that stabilizes the specified alignment direction, n → 0 θ 0 , φ 0 , which is also called anchoring energy, A, having the unit of joule per meter square. The stabilization occurs due to the increase in the anchoring energy part of the system total energy when the LC director deviates from the alignment direction, i.e., n → ≠ n → 0 , which means deviation angles Δθ and Δφ in polar coordinates (Figure 1).

The anchoring energy causes the force applied to the liquid crystal media at the surface from the alignment layer, which tends to minimize the total energy of the system: F → = ∇ → A .

Needless to mention, the same force is applied to the alignment layer surface in the liquid crystal medium. In some rare cases, it may even cause inelastic deformation of the surface of the alignment layer, which is the failure of the alignment material. The nabla operator ∇ indicates the mathematical operation of a gradient in direction. Thus, the change in the anchoring energy makes a contribution, while the constant does not contribute to the LC alignment. The standard formulation of the anchoring energy is the Rapini–Popoular form (Barbero and Durand, 1986): A = − 1 2 W n → ∙ n → 0 2 ,

which, in polar coordinates, is the sum of polar, A _p , and azimuthal, A _a , surface anchoring energies: A p = − 1 2 W p ⁡ cos 2 θ − θ 0 = 1 2 W p ⁡ sin 2 θ − θ 0 − 1 2 W p , A a = − 1 2 W a ⁡ cos 2 φ − φ 0 = 1 2 W a ⁡ sin 2 φ − φ 0 − 1 2 W a , where W _p and W _a are the constants of polar and azimuthal anchoring energies, respectively.

The actual surface LC direction is due to the anisotropy of its properties on the surface. Thus, if the surface of the substrate is treated to provide it with anisotropic properties, then the liquid crystal on such a surface will be oriented in one direction.

Polyimide film is the standard LC alignment layer widely applied in LCD fabrication. After coating with the imide solution, the film undergoes high-temperature baking at typically T ≥ 180°C for >1 h for the thermal imidization process that forms linear polyimide chains. The rubbing step that follows causes inelastic shear deformation of the polyimide film, ordering the chains at the surface along the rubbing direction (Stohr and Samant, 1999). In the ideal case, it is the flat surface alignment. Since the rubbing alignment effect comes not from surface scratching but due to the surface ordering of polyimide chains formed at the hard-baking step (Figure 3), the prior high temperature is essential, while the rubbing conditions and the rubbing cloth are critical (Yamahara et al., 2007). High-temperature imidization is already a standard but a rather slow and limiting step. Can it be changed to an alternative?

Rethinking a standard concept is always challenging. The MTLCD laboratory rethought and developed a low-temperature alignment material (LTAM) that replaced imidization with photo-crosslinking under UV light (Mahilny et al., 2009) (Figure 4).

The 1%–2% solution of methacrylate polymer with photosensitive benzaldehyde groups in butyl acetate is applied for coating a thin film 20–80 nm on the substrate. Soft baking at 60°C–90°C for 5 min is performed to dry the solvent. The polymer film undergoes inelastic shear deformation of the film surface upon the standard rubbing step with cotton clothes, ordering the polymer chains along the rubbing direction. Finally, non-polarized UV-C/UV-B light exposure at wavelengths of 250–330 nm for ∼1 min is applied to induce photo-crosslinking of the side benzaldehyde groups, which fixes the spatial orientation of the polymer chains, making the polymer solvent insoluble and insensitive to rubbing. Such an alignment material does not possess absorption in the UV-A and blue-VIS spectral ranges for excellent photostability of the LC device.

Needless to mention, LTAM, originally intended for plastic substrates, has excellent chemical compatibility with a wide range of standard substrates, i.e., polyethylene terephthalate (PET), triacetyl cellulose (TAC), polycarbonate (PC), and others. The critical aspect is that common organic solvents applied for alignment layer coating, such as dimethylformamide (DMF), N-methyl-2-pyrrolidone (NMP) or even acetone, work well on glass but often fail on plastic substrates. The LTAM alignment layer was successfully applied on top of the TAC film of the internal polarizer inside 42” IPS 3D LCD (Park et al., 2014).

Photo-crosslinking involves modification of the polymer properties. Can light exposure change the liquid crystal pretilt angle at the surface of the photosensitive alignment layer? In fact, the LC pretilt angle is dose-dependent in the case of the alignment layer. Polar anchoring changes upon light exposure (Bezruchenko et al., 2018). Such LTAMs, achieved through the careful designing of polymer chemistry (Bezruchenko et al., 2016), can provide gradient alignment conditions tuning from VA to PL through light exposure control. The main application of pretilt angle gradient alignment materials is the fabrication of polarization-independent liquid crystal lenses (Bezruchenko et al., 2021) and LC lens arrays (Figure 5) (Muravsky et al., 2022).

High polar interaction causes planar liquid crystal alignment with a low pretilt angle, but how do we create and control the azimuthal angle distribution? Classical rubbing causes uniform alignment on the solid substrate, but it is not suitable for the azimuthal pattern on plastic substrates. Can the formation of a profiled surface cause a patterned azimuthal angle at the surface of the LTAM alignment layer?

The holographic embossing process, used in relief-phase hologram manufacturing, was applied for the formation of a patterned profiled surface of a special polymer layer via embossing of a holographic matrix with predetermined directions of diffraction gratings (Mikulich et al., 2019). Such holographic embossing alignment allows roll-to-roll fabrication of patterned quarter- and half-wave retarders from polymerizable liquid crystals on TAC, PET, and PC substrates (Moiseenko et al., 2019) at a roll speed of 20 m/min. The holographic embossing alignment has the highest process temperature of 70°C, which achieves strong azimuthal anchoring over 10⁻⁵ J/m² and is suitable for q-plates and vortex retarder array fabrication (Figure 6).

Flat surface alignment is the opposite of profiled surface alignment. Can the flat surface provide a patterned azimuthal alignment of liquid crystal? Photoalignment provides a direct positive answer verified in the experiment. The photoalignment process (Figure 7) comprises film coating, baking, and polarized light exposure. The latter induces anisotropic changes in the film that form the azimuthal LC alignment direction on the flat surface of the photoalignment material (Chigrinov et al., 2008).

Flat optics and LC photonics based on the liquid crystal-mediated geometrical phase are one of the most advanced and promising applications of contemporary photoalignment technology (Chen et al., 2020). Successful LC device fabrication requires photoalignment materials with azimuthal anchoring energy constant W _a >10⁻⁴ J/cm², which is equivalent to PI rubbing used in LC displays. High azimuthal anchoring is needed to minimize deviations in the azimuthal plane of the LC optical axis from the LC alignment direction given by the alignment layer. For the flat surface case, it requires alignment materials with high dipole moments, like AtA-0042 with μ A L   ∼ 27.7 D (Muravsky et al., 2021).

The largest deviation in the azimuthal plane of achiral nematic LC is obtained in a 90-twist nematic (TN) cell, when the azimuthal orientations given by the alignment layer on top and bottom substrates are orthogonal, giving cell twist angle φ 0 = 90 ° . Such conditions induce the twist deformation of LC inside the cell, causing LC twist angle φ .

What is the principle of azimuthal anchoring energy measurement? Torque balance is the basic principle, and it is achieved when the energies of LC twist deformation and azimuthal anchoring at both substrates are equal: W a = 2 K 22 φ d   sin 2 ∆ φ ,

where K 22 is the LC twist elastic constant, 2 ∆ φ = φ 0 − φ , and d is the LC cell gap. Next, the standard optical method (Konovalov et al., 2000) is used to determine φ and compute W a . What assumption is used by the method? It is assumed that the linearly polarized light transmitted through the twist nematic LC cell in a wave-guiding mode rotates its polarization plane to the angle, γ , that is equivalent to the twist angle of the LC structure, φ . Thus, the model assumption is formulated as follows: γ = φ .

This assumption is not valid for an optically anisotropic alignment layer with retardation δ A L ≠ 0 . What is the effect of alignment material birefringence on azimuthal anchoring energy measurement? In the case of birefringent alignment material, the formula is less trivial and has to be modified as follows (Muravsky and Murauski, 2021): γ = φ + acrtan φ δ tan 2 δ A L ,

where δ = π ∆ n d / λ is the retardation of the liquid crystal layer; δ A L = π ∆ n A L d A L / λ is the retardation of the alignment layer; ∆ n   &   d and ∆ n A L &   d A L are the birefringence and thickness of LC and alignment layers, respectively; and λ is the wavelength of light.

Thus, the alignment layer birefringence, δ A L , contributes to the increase in the light polarization plane rotation angle, γ , leading to overestimated W a values measured by the standard optical method. However, in the case of really high azimuthal anchoring, an unexpected actual result of γ > 90 ° may be obtained.

Azo dyes are one of the most efficient LC photoalignment materials (Chigrinov et al., 2005) that have already been researched for over 2 decades. What experimental facts ground contemporary knowledge about the high-anchoring azo-dye photoalignment? A large set of chemical and physical experiments have been performed to clarify the role of different groups of the structure of azo-dye molecules, optimizing LC photoalignment properties and revealing the physics behind it.

Thus, it is established that when the intermolecular hydrogen bond blocks the cis–trans isomerization of the azo group, the photoalignment properties vanish (Mikulich V. S. et al., 2015). Temperature is an alternative way to block the cis–trans isomerization. It is established that the azo-dye photoalignment layer photosensitivity completely disappears at 103°C and reappears at room temperature (Mikulich et al., 2016a). At the same time, the temperature neither affects the ability of the dye layer to orient liquid crystals nor changes the alignment direction as photoalignment azo-dye molecules are the salts of alkali metals that form intermolecular coordination bonds (Muravsky, 2009). The bonding energy of the coordination bonds, E b o n d s , is higher than the thermal energy, k T , but is lower than the energy of the absorbed light photon, h ν : k T < E b o n d s < h ν ,

where k is the Boltzmann constant; h is the Plank constant; T is the temperature; and ν is the absorbed photon frequency.

Since the azo-dye molecules of the photoalignment layer are bonded by coordination bonds, the photoalignment properties depend on the kind of alkali metals (lithium [Li], sodium [Na], potassium [K], rubidium [Rb], or cesium [Cs]), and as verified in the experiment, the K-salts are preferable (Mikulich et al., 2014). According to the diffusion model of photoaligning in azo-dye layers (Chigrinov et al., 2004), the dye molecules became oriented perpendicular to the polarization of light exposure, increasing the dye-order parameter, P 2 d y e . Two directions are perpendicular to the linear polarization of normal exposure: in the plane and normal to the plane of the photoalignment layer. The preferable direction of molecule orientation during light exposure depends on the dye structure and is subjected to methyl radical position (Mikulich et al., 2016b).

However, investigating the dependences of photoinduced δ A L and W a of the AtA-2 azo-dye layer on the linearly polarized light exposure dose showed that the azimuthal anchoring energy is already high at very low exposure doses, while alignment layer birefringence is close to zero (Muravsky and Murauski, 2021), as well as the dye molecular order parameter, P 2 d y e ≈ 0 . The analysis of experimental data showed that the orientation order parameter of the photoalignment layer determines not the ordering of dye molecules P 2 A L ≠ P 2 d y e but the order of hole dipoles photoinduced inside the photoalignment layer P 2 A L = P 2 h o l e (Muravsky et al., 2020), possibly due to the fixation of molecular orientation by coordination bonds inside the layer. In brief, the hole dipole is the part of the field of the self-consistent dipoles of all dye molecules that remain, when the molecule, having absorbed a photon, is removed.

The popular photoalignment azo-dye Brilliant Yellow (BY) is air humidity-dependent (Wang et al., 2017). Why does the water sensitivity of azo-dye happen? The dye layer is coated with the solution of the dry dimethylformamide (DMF) solvent. However, BY photoalignment properties fail if the layer comes into contacts with water or is exposed to humid air. The major troublemaker is sodium sulfonate groups of the dye structure, i.e., RSO ₃ Na. Due to this group, the dye layer absorbs water from the air, forming crystal hydrates and losing its photosensitivity. In fact, crystal hydrate formation is a common problem of azo dyes with the sulfonate group. However, are there water-friendly photoalignment azo-dyes that are not afraid of water? FbF dye is the first type of water-friendly photoalignment material (Mikulich V. et al., 2015), which yields film that is completely inert to direct water contact and can be rinsed with H₂O.

The recent trends in LC photoalignment follow sustainable development for “GreenLab” with the main idea of applying processes that require harmless solvents and generate less pollution. DMF is the standard solvent for azo-dye photoalignment materials. Can the DMF solvent be changed to water? Rethinking the dye structure, it is possible to refuse using DMF solvent and coat films with water. AtA-2522/AbA-2522 dyes are the second type of water-friendly photoalignment materials that can be coated in a fully functional photoalignment layer from H₂O solution, which provides excellent chemical compatibility. Thus, the AtA-2522 photoalignment layer can be directly coated on top of the polymerizable LC layer for multilayer structure fabrication with advanced properties, i.e., achromatic polymerizable liquid crystal retarder (Muravsky et al., 2023).