A new photoalignment method of controlling the pretilt angle of liquid crystals (LCs) by using organic/inorganic hybrid interpenetrating polymer networks (IPNs) is proposed and demonstrated. In the hybrid IPN alignment layer system, the competition between poly(vinyl cinnamate) (PVCi) favoring planar alignment and poly(dimethyl siloxane) (PDMS) favoring vertical alignment made it possible to achieve pretilt angle in a wide range from 0° to 90°, and adjust pretilt angle as a function of PDMS content. In addition, we achieved the high azimuthal anchoring energy at the intermediate pretilt angle by using PDMS as the vertical-aligning component.
© 2009 OSA
The optically compensated bend (OCB) mode liquid crystal display (LCD), in which LCs normally have pretilt angle of ~25°, have drawn much attention since they can offer a potential of fast response time. The LC transition occurs rapidly because there is no backflow involved in its LC switching . In addition, the viewing angle of OCB mode LCD is large due to its self phase compensation . In the initial stage, LC configuration of OCB mode LCD is a splay state, and it transits to the bend state by applying voltage. Then, the OCB mode LCD presents the image by switching between the bend state and homeotropic state of LC. However, the initial setting voltage, which is the voltage required to set the LC cell into the bend state, is so high that the power consumption becomes higher. A simple method to reduce the initial setting voltage is to use the alignment layer with rather a larger pretilt angle than ~25°. Many works have been focused on this issue and found that pretilt angles of ~50° from the substrate surface could make it possible to fabricate the un-biased pi-cell LCD.
The conventional method to obtain intermediate pretilt angle is silicon oxide (SiOx) evaporation . However, this process is not amenable to mass production and large display panels. New techniques based on surface-induced alignment have been proposed to produce intermediate pretilt angle. These include mechanical rubbing of polyimide having a long alkyl side chain [3–5], nanostructured surface based on random distribution of vertical and horizontal polyimide domains [6–10], ion beam exposure [11,12], photoirradiation of the photosensitive alignment layer containing azobenzene , nanotexture formation by an atomic force microscope local oxidation , use of dual alignment layer [15,16], and nanoparticles-doped planar polyimide film . However, there remain several technical difficulties to be overcome such as high cost and complexity of the manufacturing process, the contact problem resulting from the rubbing process of polyimide. Especially, the contact problem, such as the generation of dust particles and electrostatic charge induced by the friction between the alignment layer and rubbing pad, is a serious drawback to current LCD manufacturing process based on polyimide. To use polyimide as an alignment layer also has another problem of their high-temperature imidization reaction. The imidization process, which changes the poly(amic acid) into the polyimide, needs a high temperature over 200°C that is an obstacle for application in flexible device. Therefore, non-contact method with low temperature process such as the photoalignment, which can generate intermediate pretilt angle, needs to be developed as an alternative to the rubbing process.
Another essential problem to be concerned at intermediate pretilt angle is azimuthal anchoring energy. Usami et al. reported that the defect of LC alignment is observed during the range of the intermediate pretilt angle (5°~70°) . They explained that the content of the alignment layer favoring vertical alignment of LC has no contribution to the azimuthal anchoring energy of the LCs and thus the ability of in-plane LC orientation becomes weaker at the intermediate pretilt angle where vertical-aligning-components have a significant role. They also showed the reduction of the in-plane order parameter with increase of the content of vertical-aligning-components. The in-plane orientation for the polyimide alignment layer was found high enough to define the easy axis of LC alignment because of their own high azimuthal anchoring energy, however, photoalignment materials such as PVCi have an intrinsically low azimuthal anchoring energy [19–21]. To provide a photoalignment layer with an intermediate pretilt angle and also a high azimuthal anchoring energy (or high in-plane order parameter), one should carefully select an alignment layer material capable of aligning LC vertically. Vertical-aligning-components should be used in a small amount within the limits of the possibility to maintain high azimuthal anchoring energy at the intermediate pretilt angle.
To solve these problems, we propose new photoalignment layer which can completely control LC pretilt angle by using organic/inorganic hybrid IPNs. In this system, the component for in-plane LC alignment is PVCi, which is the commonly used as the photoalignment layer, and vertical-aligning-component is PDMS. Since PDMS is extremely hydrophobic and can lead to a very high vertical alignment, the new photoalignment layer can maintain high azimuthal anchoring energy at the intermediate pretilt angle.
2. Hybrid organic/inorganic materials
Chemical formulas of the hybrid organic/inorganic IPN alignment layer are shown in Fig. 1 . 3-(Trimethoxysilyl)propyl methacrylate (TMSPM, Mw = 248.35 g mol−1), Hydroxy terminated poly(dimethyl siloxane) (h-PDMS, typical Mn = 550 g mol−1, typical viscosity = 25,000 cSt), Tin(II) 2-ethylhexanoate (OcSn, Mw = 405.11 g mol−1) and Ethylene glycol dimethacrylate (EGDMA, Mw = 198.22 g mol−1) were purchased from Sigma-Aldrich. Vinyl cinnamate (VC, Mw = 174.20 g mol−1) and α, α’-Azobisisobutyronitrile (AIBN, Mw = 164.21 g mol−1) were obtained from TCI-EP(Tokyo Kasei, Japan) and Kanto Chemical Co., Inc., respectively. All reagents were used as received.
To investigate the effect of concentration of PDMS, we prepared seven formulations with a different inorganic content as summarized in Table 1 . Matrix precursor solutions are composed of TMSPM, h-PDMS, OcSn, EGDMA, VC and AIBN. The number-average molecular weight of h-PDMS is 550 g mol−1, which is low enough to have compatibility with other chemical reagents. TMSPM was used as the cross-linking agent reacting with h-PDMS by sol-gel reaction between hydroxyl group of h-PDMS and methoxy group of TMSPM. OcSn is the catalyst for the polycondensation reaction [22,23]. VC and EGDMA constituted PVCi network by radical polymerization of AIBN. By using TMSPM, radical copolymerization was initiated by thermal decomposition of AIBN and then graft-IPN structure was obtained. In all the cases, the molar ratio of methoxy group of TMSPM to hydroxyl group of h-PDMS was 1:4. For all the synthesis, OcSn was used in an amount corresponding to 2.5% by weight of the polymer. The concentrations of AIBN and EGDMA were 2% and 5% by weight, respectively.
The preparation of the organic/inorganic hybrid IPN alignment layer is performed with the following steps. A mixture of h-PDMS, TMSPM, VC, OcSn, EGDMA and AIBN was poured into a mould manufactured by two glass plates separated by a 75 μm polyimide tape. The upper glass plate was laminated by Mylar film as a releasing layer. Two glass plates were clamped together and then the mixture was injected into sandwiched glass plates. The PDMS network was formed by the sol-gel reaction at room temperature during 1 day while radical polymerization did not occur due to absence of thermal decomposition of AIBN. After then, the temperature was raised to 60°C to initiate the radical copolymerization of the vinyl monomers. The resulting alignment layer was completely cured overnight at 75°C and post-cured under vacuum at 70°C for 3h.
After the formation of the alignment layer, the photoreaction of the PVCi was carried out by irradiating the organic/inorganic hybrid IPN layers with the linear polarized ultraviolet light (LPUVL) at normal incidence. This was conducted by passing light from a 500W high pressure mercury arc (Oriel) through a water bath, a UV band-pass filter (Oriel, 59800) and a UV linear dichroic polarizer (Oriel, 27320) on to the alignment layers. The intensity of the UV irradiation, which is measured by using a UV detector (Delta OHM, HD 9021), was 1.7 mW cm−2.
A homogeneously aligned LC cell was constructed by sandwiching nematic LC (E7) between couples of glass substrate covered with the IPN layer. The thickness of the LC layer was adjusted by spherical polystyrene bead of 10μm diameter. E7 LC was injected into the cell via capillary action under isotropic phase, and the cell was gradually cooled to the temperature corresponding to nematic phase.
3. Synthesis and characterization of the hybrid IPNs
Hybrid organic/inorganic materials have been intensively studied for decades because these composites can combine the versatility of organic polymer with the superior functional properties of inorganic polymers. However, these materials often bring about serious problem of incompatibility. Especially in the case of PDMS, it becomes more significant due to their extremely high hydrophobicity. In our work, we were concerned about this problem, and molecular weights of matrix solutions were carefully controlled to be in the range of 150 to 550 g mol−1 to enhance compatibility and to make precursors the solvent-free system. In the case of using the commercial crosslinker of TEOS, the alignment layer is formed with full-IPN structure however their domain size is bigger than nano-scale, which brings down the transparency . To increase the transparency and get a molecular-scale dispersion, in our work, we introduced TMSPM to make a chemical crosslinking between organic networks and inorganic clusters.
The schematic for development of the alignment layer of the hybrid IPNs is shown in Fig. 2 . The process consists of two reaction steps. As shown in Fig. 2a, the inorganic clusters were predominantly formed by the sol-gel reaction between the hydroxyl group of h-PDMS and the methoxy group of TMSPM under the room temperature. The inorganic networks could not be fully developed and interpenetrated because the content of the inorganic network is not sufficient to allow completion of the PDMS networks over the whole film. As in Fig. 2b, organic networks are cured by thermal decomposition of AIBN. It is important to note that covalent linkage between organic crosslinker (EGDMA) and TMSPM leads to a homogeneous distribution of inorganic clusters in the organic networks, although the two components generally have poor compatibility. Unlike an ideal IPN structure, the more realistic structure of hybrid IPN, in which the inorganic clusters are not necessarily connected but rather dispersed in the organic networks, is shown in Fig. 2c [25,26].
Figure 3 shows the ATR-FTIR spectra of hybrid IPN alignment layer. To evaluate the formation of inorganic clusters within the organic networks, Fourier transform infrared (FTIR) spectra in the attenuated total reflection (ATR) mode were measured by using FTS-7000 spectrometer and UMA-600 microscope (Varian). In Fig. 3a, the absence of the absorption peaks in the range of 3200 to 3400 cm−1 clearly indicates that the alignment layer does not contain any Si-OH bond of h-PDMS. It means that the sol-gel reaction between hydroxyl group of h-PDMS and methoxy group of TMSPM was well performed. Fig. 3b shows the enlarged spectra of Fig. 3a, for the detailed investigation of presence of inorganic clusters. The absorption peaks due to the Si-(CH3)2 bonds in the PDMS appear at 802 and 1261 cm−1. The absorption peak at 1018 and 1087 cm−1 can be assigned to the stretching vibration of Si-O-Si bonds, which are newly formed according to the sol-gel reaction. As the content of inorganic precursors is increased, the corresponding absorption peak intensity is also increased. Based on these ATR-FTIR spectra, it is confirmed that the content of resulting inorganic cluster is increased with increase in the amount of inorganic precursors.
To evaluate the microscopic image of the hybrid IPN alignment layer in detail, we prepared the cross-sectioned sample by ultramicrotoming method. Microscopic images were obtained by using field emission transmission electron microscopy (FE-TEM, FEI, Tecnai G2 F30 S-Twin). Ultramicrotomy method (Leica, EMFCS) was used to prepare specimens for the cross-sectional TEM imaging. An epoxy resin (EpoFix kit, Struers) was employed as a matrix for the ultramicrotoming. Elemental composition analyses were carried out using energy dispersive X-ray (EDX) analysis attached to the microscope.
The morphology of the prepared sample was then observed, as shown in Fig. 4 . To verify the elements within the darker domains and the brighter networks, EDX analysis was also conducted, as shown in Fig. 4c and 4d. As expected, the darker domains proved to be inorganic clusters having PDMS due to their distinguishable amount of Si peak that has not been observed in the white areas. Since the inorganic clusters have the heavier atomic weight than other components, their domains seemed to be darker than other organic networks. In Fig. 4a, transmission electron micrograph is shown for IPN2 and we could notice that the inorganic clusters did not generate networks within the organic networks due to their insufficient amount. Based on the comparison of TEM images with ATR-FTIR results, it is concluded that the addition of small amount of PDMS leads to the homogeneous organic clusters within the organic networks. We could not determine the domain of inorganic clusters by TEM analysis. ATR-FTIR results of IPN2 sample also clearly show that the formation of Si-O-Si bond is too small to get distinguishable peak intensity. In contrast, IPN20 (Fig. 4b) clearly showed larger sized and irregularly distributed inorganic clusters, which is in good agreement with the ATR-FTIR results of IPN20 (Fig. 3b). However, areas of those dark domains were smaller than the estimated areas so that we conjecture that areas of organic networks may contain some of the inorganic clusters of which domain size is invisible. These observations indicate that the inorganic clusters were well produced within the organic networks and domains sizes were increased with increasing PDMS content.
4. Optical and LC aligning property
High transmittance is an important requisite for an alignment layer and with TMSPM, which can lead to a homogeneous dispersion, we could fabricate a highly transparent alignment layer as listed in Table 2 . Fig. 2d represents the real hybrid IPN film with high transparency we have prepared. The light transmittance was determined with He-Ne laser operating at 633nm. The initial intensity of 633nm laser was 227µW and the intensity after passing through the bare glass (I) was 196µW. The transmittance (%) was obtained as the ratio of the intensity of transmitted light (I0) to the intensity after passing through the bare glass (I). Despite of high thickness (75 μm) of the alignment layer, good transmittance was obtained.
Figure 5 shows the dependence of water contact angle of the hybrid IPN alignment layer on the content of PDMS. The contact angle was measured on a contact angle anlayzer (Phoenix 300 Plus, SEO Co., Ltd.). Each of the reported contact angles represents the average of six measurements. In the absence of embedded PDMS, the average water contact angle is 92°, whereas the water contact angle of the sample holding 20 wt% PDMS is 111°. This result confirms that the hydrophobicity of the hybrid IPN alignment layer was increased with increase of the content of PDMS owing to its low surface energy. Figure 6 shows a change of the pretilt angle of the hybrid IPN alignment layer as a function of PDMS content. To observe the hybrid IPN alignment layer effect on the LC pretilt angle, we fabricated an antiparallel cell for measurement of the pretilt angle. The cell gap was 10μm, and the pretilt angle was measured using pretilt angle measurement system (PAMS-200, Sesim photonics technology Co., Ltd.), which operation principle was based on the modified crystal rotation method . As expected from Fig. 5, the pretilt angle shows an increasing tendency with increasing content of PDMS because of high hydrophobicity of PDMS. In the case of IPN0, the pretilt angle is almost zero degree, which is a similar result observed from the conventional photoalignment of PVCi. With increasing the content of PDMS, the pretilt angle gradually increased so that IPN15 can make LC alignment almost vertical. The intermediate pretilt angle could be also generated by addition of only 5 to 8 wt% of PDMS.
The plot of azimuthal anchoring energy vs. content of PDMS is shown in Fig. 7 . In this experiment, the azimuthal anchoring energies were calculated by measurement of the width of the Neel wall . It is found that the azimuthal anchoring energy is gradually decreased with increase of PDMS content. Also, it is noticeable that rather a high azimuthal anchoring energy (~6 x 10−6 J m−2) could be achieved at intermediate pretilt angle. Insets of Fig. 7 are polarized optical microscopic images showing LC alignment at different PDMS content. Even at intermediate pretilt angle, good homogeneity of LC alignment was shown in Fig. 7. Our success stems from the homogeneous dispersion of PDMS within organic photoalignment layer as shown in TEM analysis. Therefore, it is proposed that introduction of PDMS into the conventional PVCi can provide good orientation of LC at intermediate pretilt angle due to its high hydrophobicity only upon small amount of addition.
We have demonstrated that pretilt angle can be controlled in a wide range with a complete success by using organic/inorganic hybrid IPN alignment layer. In addition, high azimuthal anchoring energy at the intermediate pretilt angle is achieved owing to the limited addition of PDMS into the alignment layer. This new hybrid IPN alignment layer has a high potential for various applications including the un-biased pi-cell LCD.
This work was supported by a grant from ‘Center for Nanostructured Materials Technology’ under ‘21st Century Frontier R&D Programs’ of the Ministry of Education, Science, and Technology, Korea. (code# 09K1501-02510).
References and links
1. K. Takatoh, M. Hasegawa, M. Koden, N. Itoh, R. Hasegawa, and M. Sakamoto, “Applications of nematic liquid crystals” in Alignment Technologies and Applications of Liquid Crystal Devices, G. W. Gray, J. W. Goodby, and A. Fukuda, eds. (Taylor & Francis, 2005), pp. 117–121.
2. J. L. Janning, “Thin film surface orientation for liquid crystals,” Appl. Phys. Lett. 21(4), 173–174 (1972). [CrossRef]
3. H. Hatoh, K. Shohara, Y. Kinoshita, and N. Ookoshi, “Molecular tilt direction in a slightly tilted homeotropic aligned liquid crystal cell,” Appl. Phys. Lett. 63(26), 3577–3579 (1993). [CrossRef]
4. G. P. Sinha, B. Wen, and C. Rosenblatt, “Large, continuously controllable nematic pretilt from vertical orientation,” Appl. Phys. Lett. 79(16), 2543–2545 (2001). [CrossRef]
5. H. D. Jayathilake, M. H. Zhu, C. Rosenblatt, A. N. Bordenyuk, C. Weeraman, and A. V. Benderskii, “Rubbing-induced anisotropy of long alkyl side chains at polyimide surfaces,” J. Chem. Phys. 125(6), 064706 (2006). [CrossRef]
6. F. S. Yeung, J. Y. Ho, Y. W. Li, F. C. Xie, O. K. Tsui, P. Sheng, and H. S. Kwok, “Variable liquid crystal pretilt angles by nanostructured surfaces,” Appl. Phys. Lett. 88(5), 051910 (2006). [CrossRef]
7. F. S. Yeung, F. C. Xie, J. T. Wan, F. K. Lee, O. K. Tsui, P. Sheng, and H.-S. Kwok, “Liquid crystal pretilt angle control using nanotextured surfaces,” J. Appl. Phys. 99(12), 124506 (2006). [CrossRef]
8. K. E. Vaughn, M. Sousa, D. Kang, and C. Rosenblatt, “Continuous control of liquid crystal pretilt angle from homeotropic to planar,” Appl. Phys. Lett. 90(19), 194102 (2007). [CrossRef]
9. J. Y. Ho, V. G. Chigrinov, and H. S. Kwok, “Variable liquid crystal pretilt angles generated by photoalignment of a mixed polyimide alignment layer,” Appl. Phys. Lett. 90(24), 243506 (2007). [CrossRef]
10. J.-H. Lee, D.-S. Kang, C. M. Clarke, and C. Rosenblatt, “Full control of nematic pretilt angle using spatially homogeneous mixtures of two polyimide alignment materials,” J. Appl. Phys. 105(2), 023508 (2009). [CrossRef]
11. J. B. Kim, K. C. Kim, H. J. Ahn, B. H. Hwang, D. C. Hyun, and H. K. Baik, “Variable liquid crystal pretilt angles on various compositions of alignment layers,” Appl. Phys. Lett. 90(4), 043515 (2007). [CrossRef]
12. J.-H. Seo, H. J. Jang, S. R. Lee, T.-H. Yoon, J. C. Kim, I.-B. Kang, and C. H. Oh, “Wide Pretilt Angle Control of Liquid Crystal Display Device by Ion Beam Exposure on the Vertical Aligning Layer,” Jpn. J. Appl. Phys. 46, L1074–L1076 (2007). [CrossRef]
13. S. Furumi and K. Ichimura, “Photogeneration of High Pretilt Angles of Nematic Liquid Crystals by Non-Polarized Light Irradiation of Azobenzene-Containing Polymer Films,” Adv. Funct. Mater. 14(3), 247–254 (2004). [CrossRef]
15. J. B. Kim, K. C. Kim, H. J. Ahn, B. H. Hwang, J. T. Kim, S. J. Jo, C. S. Kim, H. K. Baik, C. J. Choi, M. K. Jo, Y. S. Kim, J. S. Park, and D. Kang, “No bias pi cell using a dual alignment layer with an intermediate pretilt angle,” Appl. Phys. Lett. 91(2), 023507 (2007). [CrossRef]
16. Y.-J. Lee, J. S. Gwag, Y.-K. Kim, S. I. Jo, S.-G. Kang, Y. R. Park, and J.-H. Kim, “Control of liquid crystal pretilt angle by anchoring competition of the stacked alignment layers,” Appl. Phys. Lett. 94(4), 041113 (2009). [CrossRef]
18. K. Usami, K. Sakamoto, J. Yokota, Y. Uehara, and S. Ushioda, “Pretilt angle control of liquid crystal molecules by photoaligned films of azobenzene-containing polyimide with a different content of side-chain,” J. Appl. Phys. 104(11), 113528 (2008). [CrossRef]
19. M. Schadt, K. Schmitt, V. Kozinkov, and V. Chigrinov, “Surface-induced parallel alignment of liquid crystals by linearly polymerized photopolymers,” Jpn. J. Appl. Phys. 31(Part 1, No. 7), 2155–2164 (1992). [CrossRef]
20. S. J. Sung, K. Y. Cho, J. H. Yoo, W. S. Kim, H. S. Chang, I.-W. Cho, and J. K. Park, “Dimerization behavior of cinnamate group attached to flexible polymer backbone and its effect on the molecular orientation,” Chem. Phys. Lett. 394(4–6), 238–243 (2004). [CrossRef]
21. S. J. Sung, K. Y. Cho, H.-D. Hah, J.-M. Lee, H. K. Shim, and J. K. Park, “Two different reaction mechanisms of cinnamate side groups attached to the various polymer backbones,” Polymer (Guildf.) 47(7), 2314–2321 (2006). [CrossRef]
22. G. C. Meyer, “The influence of stannous octanolate on the polymerization of acrylic compounds,” Makromol. Chem., Rapid. Commun. 4(4), 221–225 (1983). [CrossRef]
23. X. W. He, J. M. Widmaier, J. E. Herz, and G. C. Meyer, “Polydimethylsiloxane/poly(methylmethacrylate) interpenetrating polymer networks: 1. Efficiency of stannous octoate as catalyst in the formation of polydimethylsiloxane networks in methyl methacrylate,” Polymer (Guildf.) 30(2), 364–368 (1989). [CrossRef]
24. X. W. He, J. M. Widmaier, J. E. Herz, and G. C. Meyer, “Polydimethylsiloxane/poly(methyl methacrylate) interpenetrating polymer networks: 2. Synthesis and properties,” Polymer (Guildf.) 33(4), 866–871 (1992). [CrossRef]
26. S. Lee, Y.-C. Jeong, Y. Heo, S. I. Kim, Y.-S. Choi, and J.-K. Park, “Holographic photopolymers of organic/inorganic hybrid interpenetrating networks for reduced volume shrinkage,” J. Mater. Chem. 19(8), 1105–1115 (2009). [CrossRef]
27. Y. W. Li, J. Y. L. Ho, F. S. Y. Yeung, and H. S. Kwok, “Simultaneous determination of large pretilt angles and cell gap in liquid crystal displays,” J. Disp. Technol. 4(1), 13–17 (2008). [CrossRef]
28. X. T. Li, D. H. Pei, S. Kobayashi, and Y. Iimura, “Measurement of Azimuthal Anchoring Energy at Liquid Crystal/Photopolymer Interface,” Jpn. J. Appl. Phys. 36(Part 2, No. 4A), L432–L434 (1997). [CrossRef]