1. Introduction

An axially symmetric liquid crystal (LC) structure can be adopted as spatial polarization converter and axial waveplate. Such waveplates present a valuable photonics tool for beam shaping [1–3], imaging [4,5], laser material processing [6], spectral filtering [7] and optical tweezer applications [8,9]. An attractive characteristic of the axially-symmetric LC structure is that the device is independent of a linearly polarized incident beam, because the LC directors are oriented symmetrically in the azimuthal directions. The potential for developing space-variant polarized light and axially symmetry devices is increasing [10–12].

Many approaches for producing axially symmetric polarization have been studied .They include the use of special LC cells with circular rubbing [10, 11] or subwavelength gratings [12]. However, these approaches depend on a complex fabricating procedure, such as circular rubbing or micro-fabrication. Tuning the polarization state of these axially symmetric devices is difficult. Another approach for fabricating an axial waveplate is based on a liquid crystal polymer (LCP) [13]. The major disadvantage of this method is that the phase retardation cannot be tuned using one cell, and the fabrication process is too complex to match to a particular wavelength.

This work presents an axially symmetric twisted nematic liquid crystal (ASTNLC) device that is based on axially symmetrical photoalignment in azo dye-doped liquid crystal films. The structure of the ASTNLC is analyzed and compared with a simulation result. Such ASTNLC devices can be adopted as tunable polarization converters or waveplates for axially, azimuthally and vertically polarized light.

2. Device fabrication

The LC and azo dye that are adopted in this experiment were E7 (Merck) and Methyl Red (MR; Aldrich), respectively. The MR:E7 mixing ratio was 1:99 wt%. Two indium-tin-oxide (ITO)-coated glass slides, separated by 12um ball spacers, were adopted to fabricate an empty cell. No surface treatment was applied to the two cleaned glass slides. The homogeneously mixed MR/E7 compound was then injected into an empty cell in the isotropic state to generate a dye-doped liquid crystal (DDLC) sample.

In this work, the non-contact photoalignment method was adopted to produce an axial symmetric LC conformation. As described elsewhere in a previous work [14], the MR dyes undergo trans–cis isomerization, and then molecular reorientation occurs continuously after they are pumped by green-blue light. Finally, the excited MR dyes are diffused and adsorbed onto the un-treated ITO surface and LC molecules are aligned with their long axes perpendicular to the polarization of the pump beam. If the cell is maintained at room temperature, then the photoalignment effect occurs primarily on the substrate that faces the incident pump beam. This approach is called the single-side photo-alignment approach. However, if a DDLC cell is heated and maintained at a temperature just above the clear point of the LCs, then MR dyes are adsorbed onto two substrates of the cell. This process is referred to as double-side photoalignment [15].

In this work, photo-alignment was performed using a linearly polarized DPSS (diode-pump solid state) laser (λ=532 nm), whose wavelength was close to the peak of the MR absorption spectrum [16]. Figure 1 shows the experimental setup. The pump laser beam, propagating along the z-axis, with an intensity of ~0.361 W/cm², was expanded into a collimated beam with a diameter of ~21 mm. It then passed through a linear mask with a line-width of ~200 um, and was focused using a cylindrical lens onto the cell. The sample was attached to a rotating motor, and thermally controlled at a temperature of ~65°C (which exceeds the clear temperature of E7 of ~61°C) during pumping to ensure double-side photoalignment. The angle, θ, made between the polarization of the pump beam and the x-axis (Fig. 1) can be controlled using the rotator at a rotating speed ~140 rpm. The period of illumination was ~60 minutes. As stated above, the LC molecules on the surfaces of the double-side substrate are photo-aligned perpendicular to the polarization. Therefore, a DDLC cell with a double-side axially symmetric structure was formed. Notably, a reliable double-side photoalignment can be performed by rotating the sample at a speed of ~60 to 800 rpm under illumination for 60 minutes.

 

Fig. 1 Sample fabrication setup.

Download Full Size | PDF

3. Results and discussion

Figure 2 schematically depicts the conformation and optical images, obtained under a polarized optical microscope (POM), of axially symmetric LC samples. Initially, the pumping laser irradiates the sample with a polarization angle of θ = 90° (with polarization along the y-axis, as presented in Fig. 1). The excited dyes undergo trans–cis isomerization, molecular reorientation, diffusion and, finally, adsorption onto the ITO surfaces and the LC molecules become aligned with their long axes perpendicular to the polarization of the pump light. The dyes can be adsorbed by both of the substrates of the cell when the cell is optically excited at a cell temperature that just exceeds the clear temperature of the LC [15]. Since the sample is rotated, double-side photo-alignment causes the formation of a double-side axial symmetric azimuthal LC cell, as presented in Fig. 2(a). The diameter of the pattern is ~20 mm. The polarization direction of the pumping light was then changed from y-axis to x-axis (θ =0°) to fabricate another axially symmetric radial LC structure, as presented in Fig. 2(b). Figures 2(d) and 3(e) show images of the axially symmetric azimuthal and radial LC samples, respectively, under crossed POM. To fabricate a hybrid azimuthal-radial cell, two substrates were disassembled from both azimuthal and radial samples were combined to fabricate a new azimuthal-radial cell, as presented in Fig. 2(c). The coincidence of the symmetrical centers of the two substrates is important. This step was performed carefully under a POM. One substrate was fixed on the object-stage of the POM, and another substrate was adjusted by the micro-travel stage. After being coincided, two substrates were glued together and a hybrid azimuthal-radial cell was formed. Injection with LC yielded an LC cell with an axially symmetric twisted structure, because of the orthogonal alignment between the two substrates. Figures 2(f) and 2(g) present images of the ASTNLC samples under crossed and parallel POM, respectively. In contrast, Figs. 2(d) and 2(e), which present the dark regions, show that the polarizer and analyzer are converted to the bright state under crossed POM in the ASTNLC sample, as presented in Fig. 2(f). This change of state results from the 90° polarization rotation effect (under Mauguin’s condition) for a lineally polarized beam through the twisted nematic structure. When the sample was rotated under POM, a stationary image was obtained because of the axially symmetric conformation, indicating directly that the ASTNLC device is polarization-independent.

 

Fig. 2 Axially symmetric (a) azimuthal, (b) radial and (c) twisted nematic LC structures. Images of axially symmetric (d) azimuthal, (e) radial and (f) and (g) twisted nematic LC devices under a polarized optical microscope. P: polarizer, A: analyzer.

Download Full Size | PDF

 

Fig. 3 Measuring transmittance of ASTNLC cell; (a) β angle in front view and (b) testing setup.

Download Full Size | PDF

The transmittance of the device was plotted as a function of β angle (T-β) at various positions to obtain a structural model of the axially-symmetric twisted nematic LC film. Here, β is the angle between the polarization axis and the front LC director. The device was placed between crossed polarizers and probed using an He-Ne laser, λ=634 nm. Figure 3 schematically depicts the measurement setup. Figure 4 plots the measured T-β curves. Clearly, the transmittance has maxima at β=0° and β=90° under the cross-polarizer condition because Mauguin’s condition is satisfied. The transmittance is lowest at β=45° because of the bisector effect of the 90° twisted nematic structure. Figure 4 plots the simulation results obtained using DIMOS software. The experimental results are highly consistent with simulated results. The scattering loss and surface reflection in the experiment show that the experimental curves are slightly lower than the simulated curves.

 

Fig. 4 T-β curves of ASTNLC device

Download Full Size | PDF

To confirm the conformation and tunability of an ASTNLC, the transmittance-voltage (T-V) curve of the device was measured under cross-polarizer condition at different positions of the ASTNLC (A, B, C, marked in Fig. 3(a)). Figures 5(a) , 5(b), and 5(c) show the experimental and simulated results, obtained using DIMOS software, at positions of A (β=0°), B (β=45°) and C (β=90°), respectively. The T-V curves are the same as a standard TN device under the same β angle. The variation between simulated and experimental results mainly comes from the large spot size of probe beam compared to the ASTNLC conformation.

 

Fig. 5 The transmittance-voltage curves of an ASTNLC at positions of (a) A (β=0°), (b) B (β=45°) and (c) C (β=90°) marked in Fig. 3(a).

Download Full Size | PDF

Figure 6 presents the polarization converter effects based on ASTNLC devices in an axially symmetric optical system. Initially, a homogeneous-radial LC film (polarization converter) [17] was adopted to transform linearly polarized light into radially polarized light. The radial polarization was analyzed using an analyzer with transmission along the y-axis; Fig. 6(b) presents the corresponding image. The region in which the polarization is perpendicular to the analyzer is in the dark state. When an ASTNLC was placed behind the homogeneous-radial LC film, it converted radially polarized light to azimuthally polarized light. Figure 6(c) presents the analyzed (using a y-axis analyzer) image. This image is the opposite of the image of radial polarization (Fig. 6(b)) in transmission.

 

Fig. 6 (a) Setup for converting polarization of linearly polarized beam using axially symmetric LC devices; (b) radially polarized light and (c) azimuthally polarized light analyzed using an analyzer (y-axis) under POM.

Download Full Size | PDF

Light in a particular axially symmetric polarized state, such as axially symmetric circular polarized light, can also be produced using this method. A particular LC device can be designed and fabricated by combining axially symmetric radial and vortex LC [17] alignment substrates as presented in Fig. 7(a) . Figure 8 presents the image of a radial-vortex alignment sample with α=45°, Δnd=1.9820 μm observed under a POM. The polarization of incident light was modulated after passing through the radial-vortex alignment sample. Simulated results, obtained using Matlab software are also presented in Figs. 8(c), 8(d) for comparison. As seen, they agree quite well with each other. The converted polarization state is determined by Δnd, the angle β and the angle α (Δn: birefringence, d: cell gap, α: twisted angle). For suitable angles α and β, axially symmetric radial polarized light can be converted into axially symmetric circularly polarized light by propagation through the designed axially symmetric LC device. According to calculation performed in Matlab, axially symmetric circular polarized light is obtained when axially symmetric radial polarized light passes through the specific axially symmetric LC device with β=30°, α=30°, and Δnd=1.7891 μm, as presented in Fig. 7(b).Additionally, unlike that of devices fabricated from liquid crystal polymers [13], the phase retardation of ASTNLC devices with fixed β and α can also be controlled electrically.

 

Fig. 7 (a) Specific axially symmetric LC device; the top substrate exhibits radial alignment and the bottom substrate exhibits vortex alignment [17]. The angle between the entrance LC direction and the exit LC direction is α; (b) axially symmetric circularly polarized light was obtained by the conversion of a radially polarized light via a particular axially symmetric LC device (β=30°, α=30°, and Δnd=1.7891 μm)

Download Full Size | PDF

 

Fig. 8 (a), (b) Images of radial-vortex alignment sample observed under a POM; (c), (d) simulated results. The analyzer makes an angle of (a) 90°, (b) and (d) 45° with the polarizer.

Download Full Size | PDF

4. Conclusion

In conclusion, the fabrication of an axially symmetric twisted nematic liquid crystal device based on axially symmetrical photoalignment in a dye-doped liquid crystal film was demonstrated. Its conformation was confirmed by measuring T-β curves, and the transmission of light through it agrees well with the simulated transmission. Additionally, ASTNLC devices in an axially symmetric optical system, based on the polarization converter effect, were also demonstrated. Notably, given suitable device parameters, an electrically tunable axially symmetric waveplate can be formed from radial, azimuthal and vortex LC films. Therefore, various axially symmetric polarization converters can be fabricated. The device is very convenient to use. It therefore has great potential in practical applications.