Microphotograph of the liquid crystal texture in twisted nematic cells under crossed polarizers. The cells were aligned between an EPP film with various UV exposure times: ͑ a ͒ 10, ͑ b ͒ 20, and ͑ c ͒ 60 min and a Rubbed PI film and filled with 5CB parallel to the electric field of the EPP film in the nematic phase. The stripes for short unpolarized UV exposure times disappear with increasing time. 

Context in source publication

Context 1

... of the key issues of liquid crystal displays LCDs is the uniform alignment of the liquid crystal ͑ LC ͒ molecules on an aligning surface. In practice this can be achieved by using a thin polymer film that provides defect free and anisotropic anchoring for the LCs. There are several well known methods 1–3 to introduce a surface anisotropy onto a polymer film. Among them the rubbing by a velvet cloth is most popular due to its simplicity and the possibility of controllable anchoring energy and pretilt angle. However, rubbing easily introduces dust and charges that may lead to serious damage in the final LCD production. 4 Therefore there has been a lot of research to develop rubbing free methods. 5,6 As a promising alternative, Schadt et al. 6 have introduced a photopolymer film exposed to linearly polarized ultraviolet light ͑ LPUV ͒ . The LPUV method has the additional advantage that it can be applied to fine pixels, which is advantageous in obtaining wide view angles. Its disadvantages are unstable anchoring and no pretilt angle. Note that though on a molecular level there is a pretilt angle, this averages out to zero because of the two fold degeneracy of such systems ͑ see Refs. 6 and 7 ͒ . We have recently demonstrated an electrically poled photopolymer ͑ EPP ͒ film 8 –10 as a rubbing free method to improve the anchoring properties of LPUV films. A fairly weak electric field in combination with a fixation using ultraviolet induced polymerization was shown to lead to an anisotropy of photopolymers that can be used for LC alignment. 8,9 In this article we report on the mechanisms of nematic liquid crystal alignment on such an EPP film. In particular the observation of the azimuthal anchoring energy in twisted nematic ͑ TN ͒ cells and the pretilt angle in planar cells are presented and their alignment mechanisms are explained via the molecular structure of the EPP film. The preparation of the EPP film has been discussed before. 8 –10 In short, thin films of polyvinylcinnamate ͑ PVCN ͒ were prepared by spin- or dip-coating a solution of 1 wt % PVCN ͑ Aldrich ͒ in chloroform on a glass substrate between two indium–tin–oxide ͑ ITO ͒ coated electrodes that were 3 mm apart. The thickness of the film, measured using a thickness profiler and ellipsometry, was about 400 Ϯ 100 nm. The coated films were dried for 1 h at 100 °C in an oven to remove the solvent. The samples were poled at 55 Ϯ 1 °C by applying an electric field of 2 kV/cm for more than 30 min and exposed to unpolarized UV light for various times from 0 to 60 min. The UV light source was a 150 W Xe-arc lamp and was placed 22 cm from the sample. Before cell preparation, the optical anisotropy of the EPP films was measured by a sensitive birefringence setup. 8 To determine the azimuthal anchoring energy, TN cells were prepared using an EPP film and a rubbed polyimide ͑ PI ͒ film as a reference surface ͑ 5-Cyano biphenyl on the rubbed PI surface was assumed to have a strong anchoring energy ͒ . The cell thickness was adjusted to 5.6 ␮ m using polyethyleneterephthalate ͑ PET ͒ films and was determined by the interference method. 11 The cells were filled with 5CB in the nematic or isotropic phase. In order to investigate the effect of the electric field on the alignment, the field in a TN cell was oriented parallel, antiparallel, or perpendicular to the rubbing direction of the PI. To study the presence of a flow effect on EPP films, the cell was filled parallel, antiparallel, or perpendicular to the electric field. The twist angle in a TN cell was measured using the method given in Ref. 12 and used to calculate the azimuthal anchoring energy. For pretilt angle measurements, planar cells were prepared using an EPP film and an ITO surface, that was rubbed five times by a homemade rubbing machine with a velvet cloth. The cell thickness was adjusted and measured using the same method as for the TN cells. The cells were filled parallel and antiparallel to the electric field with 5CB in the nematic phase. For the light source a He–Ne laser was used without focusing in order to measure a macroarea. The textures of the cells were observed under a microscope with crossed polarizers. The twist angle in a cell with EPP and PI is very sensitive to the LC filling direction. Table I shows the obtained twist angles ␾ depending on the combination of three factors: the rubbing of PI, the electric field of the EPP, and the filling direction in the nematic phase of 5CB. The cells in Table I are all aligned with PI and an EPP film exposed to UV for 10 min except R ͑ with a PVCN film instead of an EPP one ͒ . The LC on PI has a strong anchoring energy (10 Ϫ 4 J/m 2 ). R has a twist angle of 72 Ϯ 2° due to the strong flow effect on PVCN. The flow direction in a cell with an EPP film is seen to determine the LC director on the film, while the electric field plays a secondary role in the alignment. Depending on the filling A parallel to the electric field achieves a TN structure and B , perpendicular to the field is only half twisted by rotating the flow alignment to the field direction, while C , with the three factors parallel to each other shows a planar structure. The above comparisons lead to the conclusion that the anchoring energy on an EPP film is determined by the flow and the electric field of the EPP film and its maximum is obtained by having these two factors in parallel. The LC alignment texture of A was studied with changing UV-exposure times under a crossed polarized microscope ͑ see Fig. 1 ͒ . A cell exposed for less than 30 min shows stripes of about a 100 ␮ m width and parallel to the filling direction. With increasing the exposure time to more than 40 min, the cells become homogeneous and defect free. Figure 2 shows the azimuthal anchoring energy W A of 5CB in a TN cell as a function of unpolarized UV exposure time. W A is calculated from the twist angle of the TN cell by using 12 2 K 22 W A ϭ L sin 2 ␾ , ͑ 1 ͒ where L is the cell gap and K 22 is the twist elastic constant of 5CB. The azimuthal anchoring energy shown in Fig. 2 increases first rapidly with UV exposure and after a maximum slowly reduces to a constant with extending UV exposure time. In order to explain this observation, the azimuthal anchoring energy on an EPP film can be expressed via the van der Waals’ interaction between the LCs and an EPP film and the anisotropy of an EPP film. 13 W A ϳ W s • S 2 , ͑ 2 ͒ with W s ϭ the van der Waals’ interaction and S ϭ the LC order parameter. Starting with the van der Waals’ interaction, the interaction energy W s on an EPP film can be described by a contribution from the cinnamates ( W c ) and ␤ -truxinates ( W t ) W s ϭ W c ϩ W t Ϸ W c . ͑ 3 ͒ Due to the dominant interaction of the cinnamate groups with the LCs, 14,15 W s is approximated by W c . Figure 3 ͑ a ͒ shows the relative concentration of cinnamates C c as a function of UV exposure obtained from the UV spectra ͑ Fig. 3, inset ͒ . The reduction of the interaction energy as a function of UV exposure time ␶ can be written as W s ϭ A ͑ 1 ϩ exp ͑ Ϫ ␣␶ ͒͒ , ͑ 4 ͒ where A is a constant defined as the interaction energy per area between the LCs and the PVCN film and ␣ is the pho- toreactivity obtained as 0.08 via fitting C c as a function of ␶ . The increase of the optical anisotropy ⌬ s ͓ Fig. 3 ͑ b ͔͒ of a poled PVCN film during the UV exposure can be described with a scalar order parameter S s . 16 The anisotropic refractive index is equal to ⌬ n s ϳ ␳ 1/2 S s , ͑ 5 ͒ where ␳ is the density of the film that remains constant during the photoreaction. ⌬ s shown in Fig. 3 can be fitted as a function of ␶ ...