1. Introduction

The photoalignment of liquid crystals (LCs) is favorable because it is a non-contact approach. The most significant application of photoalignment, typically by irradiating a surface that contains photoreactive photodimerization molecules with polarized light (visible or ultraviolet), is to fabricate liquid crystal displays (LCDs) and LC devices. Recently, light-induced molecular reorientation in dye-doped LCs (DDLCs), including nematic LCs (NLCs) [1–3], cholesteric LCs (CLCs) [4–5], polymer-dispersed LCs (PDLCs) [6–7] and ferroelectric LCs (FLCs) [8], have attracted substantial attention because of both their practical and fundamental interests. One of the effects of photoalignment involves the light-induced reorientation of the azobenzene chromophore, which is promoted by trans-cis isomerization cycles. Many works have been performed on photoalignment in LC cells, using the azo dye-adsorption effect in LC cells that are doped with a small fraction of azo dye, such as methyl red (MR). MR dye is employed in this work, and some of their properties are given in Experiment section. Voloschenko et al. initially addressed the effect of adsorption in homogeneously aligned dye-doped NLCs (DDNLCs) films [9]. Simoni et al. reported on this effect in similar systems [10–11]. Fedorenko et al. also described the evolution of light-induced anchoring in DDNLCs upon irradiation with a polarized light [12]. It should be noted that adsorption effect of MR dye is dependent on the substrate surface significantly. Refer to Refs. 3 and 9–12, photo-induced MR adsorption reported in these work was associated with the substrate coated with a photopolymer film, fluorinated polyvinylcinnamate (PVCN-F). Briefly, the adsorptions include the dark adsorption, light-induced adsorption and desorption effects. With irradiation of a linearly polarized light, the dark adsorbed and the light-excited dyes are, respectively, desorbed and adsorbed anisotropically from the PVCN-F layer coated substrate. Finally, the easy axis of LCs is determined by the competition of desorption of dark adsorbed dyes and light-induced adsorption of dyes. The former generates an easy axis perpendicular to the polarization of impinging light, while the latter yields a parallel one. Without coating such a photopolymer layer, PVCN-F, the dark adsorption effect does not take place [2, 4–5, 7–8, 13]. Rather, the adsorbed dyes on the substrate dominate the photoalignment process, and induce an easy axis perpendicular to the polarization of the pump beam. Such a mechanism is summarized briefly as follow. When MR molecules are excited by the absorption of blue-green light, they undergo a series of transformations, including photoisomerization, three-dimensional reorientation, diffusion and finally adsorption onto the substrate that faces the incident beam. Notably, the long axes of the adsorbed dyes, excited by the linearly polarized light, are oriented perpendicularly to both the polarization and propagation directions of the impinging light [13]. The adsorbed dyes in turn align the LCs via the guest-host effect [14].

Additionally, dye-doped PDLCs (DD-PDLCs) have attracted increasing interest because of their potential for use in displays, light switching devices, modulators and other devices. The authors’ earlier works introduced two DD-PDLCs - dye-doped polymer-ball-type (PBT) and LC-droplet-type (LCDT) PDLCs [6–7]. The LCDT- and PBT-PDLCs comprise, respectively, LC droplets dispersed in a continuous polymer matrix, and polymer balls dispersed in a continuous LC matrix. The differences between these two types have been described elsewhere [15–16]. We demonstrated that doped azo-dyes, when excited by linearly polarized light, were adsorbed onto the surface of the UV-cured polymer film with their long axes perpendicular to the directions of both the polarization and the propagation of the polarized light in DD-PDLCs [7].

This work demonstrates a polarizer-free, electrically switchable and optically rewritable display based on dye-doped polymer-dispersed liquid crystals (DD-PDLCs). The excited dyes in the illuminated region are homeotropically adsorbed onto a polymer film when a suitable AC voltage is applied during patterning. Finally, after the applied voltage is switched off, the adsorbed dyes can align the LCs homeotropically. To our knowledge, this study is the first report that the azo dyes may be adsorbed with their long axes being parallel to the direction of the propagation of the impinging light. The LCs used herein are nematic LCs with a positive dielectric anisotropy. The ordinary refractive index of LCs (n_o) is approximately the index of the used polymer (n_p). Therefore, the illuminated area of homeotropically adsorbed dyes is patterned, and displayed permanently in the scattering background without the use of any polarizer after the applied voltage is switched off. Experimentally, the displays can be erased and rewritten by thermal and optical treatments, respectively. The measured rise- and fall-times (10–90%) for the LCD are 1 and 60 ms, respectively, when an AC (1 KHz) voltage of 100 V is applied to the sample; the contrast ratio is ~ 30.

2. Experiments

A DD-PDLC sample was fabricated from a mixed compound of the LC (50.2 wt%, E7, Merck), the monomer di-pentaerythritol pentaacrylate (42.2 wt%, DPPA, Polysciences), the cross-linking agent 1-vinyl-2-pyrrolidinone (8 wt%, NVP, Aldrich) and the azo dye methyl red (1.6 wt%, MR, Aldrich). MR is suited to light-induced molecular reorientation; it must be excited with blue-green light since it has a broad absorption spectrum from 440 to 550 nm, peaking at ~ 530 nm [4]. The homogeneously mixed compound was heated into the isotropic state, and then injected into an empty cell that consists of two indium-tin-oxide (ITO)-coated glass slides, separated by two 30 μm-thick spacers. One of the inner-surfaces of these glass substrates was coated with a homeotropic alignment film of N, N-dimethyl-N-octadecyl-3-aminoprophyltrimethoxy-ailyl chloride (DMOAP), while the other was untreated. The surface with (without) the DMOAP film was called the reference (command) surface, S_R (S_C). After filling, the edges of the cell were sealed with epoxy to produce a sample. To generate the polymer film on the S_C, an unpolarized UV light with an optimized intensity of 0.14 W/cm² was used to cure the filled sample from the SC for 50 minutes. The monomers in a sample were polymerized using an unpolarized UV light. Notably, the formed polymer film was adhered mainly onto the S_C substrate because of the diffusion of the LCs and monomers during photo-polymerization. Restated, the LCs and monomers, respectively, diffuse outward and toward the UV source [17–19].

Figure 1(a) depicts the use of UV-curing to produce a polymer film on S_C, as described above. Samples were prepared by UV curing with various durations. Figure 1(b) presents the experimental setup for obtaining the transmittance versus voltage (T-V) curve of the cured sample, and for investigating the reorientation of the LCs and adsorbed dyes. An unpolarized red probe beam with a weak intensity of 1.2 mW/cm² (E_R, from a He-Ne laser, λ = 632.8 nm) is normally incident (along z-axis) on the sample from S_C side. Figure 1(c) presents the experimental setup for recording the patterns onto the polymer films using a linearly polarized diode-pumped solid-state (E _G along y, DPSS, λ = 532 nm) laser light through a mask. The abbreviations, P, BE and AP, stand for the polarizer, the beam expander and the aperture, respectively. Notably, BE, comprising two convex lenses that are separated by a distance that equals the sum of their two focal lengths, is used to expand the beam size of the DPSS laser, and AP is used to block the undesired light. A DD-PDLC sample is irradiated under a linearly polarized DPSS laser beam, which is expanded and collimated to a beam diameter of 1 cm using the BE. A suitable AC (1 KHz) voltage is applied during patterning. The selected amplitude of the applied voltage will be discussed below.

 

Fig. 1. Experimental setups (a) for curing sample under UV; (b) for measuring the transmittance of the cured sample with or without the application of an AC voltage; (c) for patterning a DD-PDLC sample. P, BE and AP represent polarizer, beam expander and aperture, respectively. The mask in (c) is a home-made transparent mask of the two letters, “LC”.

Download Full Size | PDF

3. Results and discussion

To elucidate the morphologies of SC and SR of the cured sample, the sample was taken apart carefully. Figures 2(a) and 2(b) display, respectively, photographs of S_R and S_C, placed on a piece of printed paper. It is clear to see that the S_R is clear, while S_C is opaque. Figures 2(c) and 2(d) present the top-view morphologies of the S_R and S_C substrates, respectively, captured under a scanning electron microscope (SEM). The cause of Figs. 2(a) and 2(b) is due the fact that the monomers in a sample are diffused towards the UV source and polymerized on the command surface, S_C, primarily [17–19]. Also, as demonstrated by the authors’ earlier work and the references therein [7], the formed polymer films were polymer-ball-type PDLC films, rather than LC-droplet-type PDLC films.

 

Fig. 2. Images of (a) reference surface, S_R, and (b) command surface, S_C, of cell after UV-curing (intensity ~ 0.14 W/cm²) for 50 minutes on a piece of printed paper. Top-view SEM images of (c) DMOAP-coated S_R, and (d) polymer film formed on S_C.

Download Full Size | PDF

The transmittance of the probe red light through the samples that were cured for various durations using the experimental setup in Fig. 1(b) without applying a voltage is measured to determine the optimal duration of UV curing (0.14 W/cm²). Figure 3(a) presents the results, and indicates clearly that the sample exhibits the lowest (highest) transmittance (scattering) when it is cured for ~ 50 minutes. The transmittances of the probe red light through the samples that were cured for more than 50 minutes were almost equal to those of the sample that had been cured for 50 minutes. Therefore, the UV-curing duration in the experiments described below was 50 minutes. Additionally, the optimal conditions of applied voltage, illumination duration and intensity of the DPSS laser were obtained experimentally. These values were ~ 100 V, 60 minutes and 0.5 W/cm², respectively. The T-V curve plotted in Fig. 3(b), obtained using the experimental setup in Fig. 1(b) without a polarizer, clearly indicates that the LCs were aligned homeotropically when a voltage of ~ 100 V (3.3 V/μm) was applied. The sample was therefore patterned using the setup in Fig. 1(c) with an applied voltage of 100 V, and the transmittance of the patterned region was then measured using the setup in Fig. 1(b) without an applied voltage. Figure 3(c) plots the variation of transmittance of the patterned region of a sample with duration of illumination by green light. Notably, an AC voltage of 100 V was applied during green-light illuminating the sample first, and then the measurement of Fig. 3(c) was performed when the green light and the applied voltage were switched off. It is clear to see from Fig. 3(c) that the transmittance reaches a stable value after illumination for about 60 minutes, and the evaluated contrast ratio is ~ 30. Therefore, the optimal duration of illumination by green light, used to pattern the sample, was set to be 60 minutes. Figure 3(d) plots the transmittance versus the polarization of the probe red light of the patterned region obtained using the experimental setup presented in Fig. 1(b) without an applied voltage. The figure indicates that the transmittance of the patterned area is independent of polarization. The experimental results (not shown) show that a rare (rough) dye-adsorption layer is formed if the intensity of the writing DPSS laser beam is less (higher) than 0.5 W/cm². Notably, a rare dye-adsorption layer cannot apply a sufficiently effective torque to align the LCs, and a rough dye-adsorption layer may cause the LCs to reorient randomly rather than homeotropically. Comparing the intensity of the DPSS laser used in the present experiment with that in typical experiments of photoalignment in MR-doped LCs [4, 13], the intensity used herein is rather high. It is reasonable, and is understood as follows. The dichroic ratio, D, defined as A_∥/A_⊥, of MR is around six for visible light, where A_∥ and A_⊥ are the dye absorbance when the pump-beam polarization is parallel and perpendicular, respectively, to the director axis of the LCs in a homogeneously aligned MR-doped LC cell. Moreover, the LCs (Δε > 0) and MRs are oriented with their long axes perpendicular to the polarization, but parallel to the propagation of the impinging light, when the sample is applied with a suitable AC voltage during patterning. Therefore, the absorbance of linearly polarized light by MR in this case is only one sixth of that when the long axes of MRs are parallel to the polarization of the linearly polarized light in a typical photoalignment of MR-doped LCs.

 

Fig. 3. Experimental results; (a) measured transmittance of UV-cured sample as function of UV curing duration; (b) transmittance versus voltage curve of cured sample; (c) variation of transmittance of cured sample with duration of illumination with green light; (d) transmittances of patterned region versus polarization direction of probe light.

Download Full Size | PDF

A polarizer-free, electrically switchable and optically rewritable display based on DD-PDLC was successfully demonstrated, using the experimental setup in Fig. 1(c) and the aforementioned optimal conditions. A home-made mask having two letters-“LC” was placed in contact with the S_C of a DD-PDLC sample. Figures 4(a) and 4(b) present the states of the DD-PDLC sample before and after patterning, respectively. Figure 4(b) indicates the illuminated letters -“LC” when dyes are adsorbed. The letters of “LC” have been patterned and displayed in the scattering background, because the light-induced adsorbed dyes are homeotropically aligned in the patterned “LC” letters, and align the LCs perpendicular to the surface of the substrate after the applied voltage is switched off. Restated, the ordinary refractive index of the LCs (n_o) matches the refractive index of the polymer film (n_p) is achieved in the patterned “LC” letters for a normally incident probe beam. The homeotropic alignment of the light-induced adsorbed dyes in the patterned “LC” letters can be understood as follows. When a suitable AC voltage is applied, the LCs (Δε > 0) are aligned by the applied electric field, causing the dyes to be reoriented by the guest-host effect with their long axes parallel the LC director. Irradiation with a green light causes the dyes to undergo photoisomerization, and they are eventually adsorbed onto the command surface with their long axes normal to the surface. The measured contrast ratio is ~ 30. Additionally, the observed homeotropically adsorbed dye layer on the polymer film in the patterned “LC” letters, presented in Fig. 4(b), can be erased and rewritten by thermal and optical treatments, respectively. Figures 4(c) and 4(d) present the respective results. The DD-PDLC sample with adsorbed dyes is heated to ~ 100 °C, which temperature is maintained for 60 minutes, and then cooled naturally to room temperature. The thermal disturbance causes desorption of the adsorbed MR from the sample [4], the patterned “LC” can be erased, and the scattered mode re-appears, as presented in Fig. 4(c). New patterns can be rewritten on the thermally treated sample using the setup in Fig. 1(c), and yielding the results presented in Fig. 4(d). Notably, the recorded patterns exhibit no significant age effect after three months.

 

Fig. 4. Photographs of DD-PDLC sample taken using a digital camera (a) before and (b) after the sample had been illuminated with a green laser beam through a mask of the two letters “LC”; (c) and (d) photographs of sample after it had been thermally erased and optically rewritten, respectively.

Download Full Size | PDF

In a separate experiment using the setup shown in Fig. 1(c), we examined the dye adsorption onto an ITO-coated glass slide spin-coated with a flat polymer (NOA81 from Norland) film and cured by an unpolarized UV light. An empty cell similar to that for making a DD-PDLC cell except for the Sc was fabricated. In this case, Sc substrate was replaced by the NOA81 polymer-coated one. A homogeneously mixed compound of the LC (98.4 wt%, E7) and the azo dye (1.6 wt%, MR) was injected into the empty cell. The result (figure is not shown) shows that the MR dyes are homeotropically adsorbed onto the flat polymer film in the region illuminated with a linearly polarized DPSS laser beam with the sample being applied with an AC (1 KHz) voltage of ~ 15 V as verified using a conoscope.

Additionally, the patterns are electrically switchable, as presented in Fig. 5. Figure 5(a) depicts the recorded pattern -“LC”. When an AC (1 KHz) voltage of 100 V is applied to the sample, the LCs in the cell are homeotropically aligned, and the recorded pattern disappears, as presented in Fig. 5(b). Notably, the sample is placed on a piece of printed paper. When the applied voltage is switched off, the sample returns to its original state, which is presented in Fig. 5(a). The switching times, τ_rise and τ_decay, defined as the times required to change the transmittance of the patterned letters from 10 to 90% and from 90 to 10%, respectively, are measured at τ_rise ~ 1 and τ_decay ~ 60 ms when an AC (1 KHz) voltage of 100 V is applied to the sample. The decay time is much longer than the rise time, because when the AC applied voltage is switched off, the LCs and dyes return to their initial state in the absence of any electric field.

 

Fig. 5. Recorded pattern, “LC”, on DD-PDLC film with applied AC voltage (1 KHz) of (a) 0 V and (b) 100 V. The pattern returns to (a) when the applied voltage is switched off. The background is a piece of printed paper.

Download Full Size | PDF

4. Conclusion

In conclusion, this work demonstrates the light-induced dye-adsorption in dye-doped polymer-dispersed liquid crystals. When a suitable AC voltage is applied during patterning, the dyes are adsorbed onto the UV-cured polymer films, and their long axes are perpendicular to the substrate surface. The adsorbed dyes then align the LCs homeotropically after the applied voltage is switched off. The written patterns are displayed without any polarizer. Additionally, the recorded patterns can be erased and rewritten by thermal and optical treatments, respectively. An application of a polarizer-free, electrically switchable and optically rewritable display is finally demonstrated. The switching time is of the order of milliseconds, and the contrast ratio is measured as ~ 30.