To realize the goal of this study, the proposed grating cell possesses the characteristics of vertical alignment, positive nematic LC (such as E7, Merck) (dielectric anisotropy Δε = 13.8, refractive indices no = 1.52 and ne = 1.75, elastic constants k11, k22, and k33 are 10.3, 7.4, and 16.5 pN, respectively), and octothorp electrode on the bottom substrate. The width, length, and cell gap of the patterned electrode were 2.8, 4, and 20 µm, respectively. In addition, we set the TechWiz LCD 3D options, such as a pretilt angle, azimuthal angle, and wavelength as 90°, 0°, and 543.5 nm, respectively; moreover, we used an optical analysis method with a 2 × 2 extended Jones matrix. The far-field intensity was detected using a photodiode located 30 cm away from the LC cell.
Figure 2a shows the POM images of the proposed grating cell with crossed polarizers at various applied voltages. To verify the direction of rotation of the LCs, a full-wave plate (45°) was inserted between the crossed polarizers. When the voltage was increased, the brightness (retardation) of most regions increased, whereas the virtual wall's brightness (retardation) remained constant, resulting in a spatial phase difference [30–35]. Because of the spontaneous fluctuation of the phase difference, the created defect patterns worked well as 2D diffraction gratings [36]. Green diffraction patterns were detected on a dark screen when an unpolarized laser beam (543.5 nm) passed through the LC cell (Figure 2b). Because most of the laser energy is directed to higher orders, the intensity of the zeroth-order is significantly reduced, regardless of the direction of polarization. We can observe that the diffraction energy is well transported from the zeroth-order to higher orders, regardless of the direction of polarization. Because of the significant spatial phase difference, we can expect the proposed grating cell with an octothorp electrode to switch to an excellent opaque state, regardless of the azimuth angle.
The haze values of the LC grating cells were calculated to determine their haziness. To evaluate the optical performance, we introduced the total, specular, and diffuse transmittance and haze. Specular [diffuse] transmittance Ts [Td] refers to the ratio of the power of the beam that emerges from a sample cell, which is parallel (within a small range of angles of 2.5°) [not parallel] to a beam entering the cell, to the power carried by the beam entering the sample. The total transmittance Tt is the sum of the specular transmittance Ts and the diffuse transmittance Td. The haze H can be calculated as H = Td/Tt. In our calculation, the total transmittance Tt was the same as the specular transmittance Ts (no light absorption). Specular transmittance was calculated by integrating the intensity with a range of 2.5° as shown in Figure 3. The diffuse transmittance Td was calculated by integrating the intensity in the range of 2.5°.
At an applied voltage of 10 V, the 1-D grating cell had a haze of 51.2%, whereas the octothorp grating cell had a higher haze of 76.7% as shown in Figure 4a. This is because the 1-D grating cell has a much larger spatial phase difference, independent of the azimuthal angle. The octothorp grating cells accounted for 25.5% higher haze values than the 1-D grating cell. This is comparable to LC smart windows based on light scattering, such as polymer-dispersed liquid crystal (PDLC) or polymer-networked liquid crystal (PNLC) cells, which have been previously reported. Because the proposed LC cell does not contain any polymer matrices, haze in the opaque state is predominantly caused by the electric-field-induced periodic continuous LC profile diffraction of the white incident light. As a result, when compared to other LC smart windows, the proposed cell offers benefits such as low angle dependence, high stability, low operating voltage, fast response time, and ease of fabrication. Using image analysis in TechWiz LCD 3D, we estimated the images of the LC grating cells placed on top of printed paper (KNU logo) at various applied voltages. When a voltage of 15 V was applied, both the grating cells became opaque. Figures 4b and c show the proposed grating cell was hazier than the 1-D grating cell.
A fast response time is one of the most important requirements for window-display applications. The dynamic switching behavior of the proposed LC cell was investigated (Figure 5). The proposed grating cell had a total response time of 7.57 ms, which is substantially faster than the existing LC smart windows, including cholesteric liquid crystal, polymer-network liquid crystal, and polymer-dispersed liquid crystal cells, which have response times of several hundred milliseconds [19,37,38]. In addition, the response times for the 1-D and 2-D grating cells were examined. The calculated turn-on [turn-off] time for 1-D, 2-D, and octothorp grating cells were 2.23 ms [3.56 ms], 3.23 ms [18.6 ms], and 3.79 ms [3.78 ms], respectively. The top and bottom patterned electrodes were used in the 2-D grating cell, with the top patterned electrode receiving voltage in the x-direction and the bottom patterned electrode receiving voltage in the y-direction. As a result, the LCs in the bulk region of the 2-D grating cell were formed in a random direction, whereas the suggested LC direction of the grating cell had an x- and y-direction owing to the single bottom patterned electrode.
The proposed cell can make a 2-D phase grating effect by consisting of patterned electrodes in only one substrate. We have additionally demonstrated a few more devices that can produce the 2-D grating effect with structures in one substrate (the diamond, spot, and protrusion grating cells). Figure 6a shows schematics of the LC grating cell with octothorp, diamond, and spot-patterned electrodes and the protrusion grating cell without patterned electrodes. The red, blue, and yellow colors in Figure 6a depict a patterned electrode, common electrode, and insulator, respectively. Compared to the proposed grating cell, the diamond grating cell was observed to have different patterned electrodes, but it has the same structure. The only difference was the angle of the patterned electrodes (rotated 45°). The spot grating cell consisted of a circle-patterned electrode. The electrode in the spot grating cell was formed by swapping the common and patterned electrodes, unlike the proposed grating cell. The protrusion grating cell has the same spot structure. It should be noted that the protrusion grating cell does not use a patterned electrode.
Figure 6b shows POM images of crossed polarizers and a full-wave plate (45°) under the same conditions as in Figure 2a. Although we rotated the angle of the patterned electrode, the POM image of the diamond cell was almost the same as that of the octothorp cell because the LCs in both devices tilted down regardless of the azimuthal direction. The POM image of the spot cell was slightly different from that of the octothorp cell owing to the formation of additional virtual walls. This difference results in the decrease of the effective period by half [33]. Therefore, the diffraction angle of the spot grating cell increases owing to the decrease in the effective period (Figure 6c). In the protrusion grating cell, which does not use a patterned electrode, the direction of the electrode field is the same, regardless of the position. In addition, LCs near the protrusion form a pretilt angle, which can provide a direction to other LCs in the bulk region to create the 2-D effect as LCs have randomly lied along the direction [36,39]. By increasing the voltage in the protrusion grating cell, we can observe that the surrounding LCs form new domains by lying in a similar direction, as shown in Figure 6b. It should be noted that the domain size can be changed with time and applied voltage, which can result in low reliability. In the protrusion grating cell, we reduced the period required to achieve a sufficient diffraction effect. We expected that the reduced period would result in a large diffraction angle; however, the diffraction angle was found to be reduced. Because the LC domains were not formed by the electric field from the patterned electrode, the bulk LCs followed the LCs near the protrusion, and the domains were broken and merged by sounding defects [36]. Therefore, it had a large domain size.
We have calculated the haze value of the LC grating cells using the experimental setup as shown in Figure 3. The maximum haze values of the proposed diamond, spot, and protrusion grating cells were 76.7, 76.65, 70.45, and 95.56% at 12.5, 15, 35, and 10 V, respectively as shown in Figure 7. The spot grating cell has a high operating voltage because the area, LCs switched by the elastic energy (blue region in Figure 6a), is larger than the proposed or diamond cells. In addition, the calculated response time profiles of the diamond, spot, and protrusion grating cells. The total response time is 9.59, 474.178, and hundred milliseconds, respectively. In the case of the spot and protrusion grating cell, response time is very slow response time. Because the spot grating cell has many bulk LCs by the circle patterned electrode, and the protrusion grating cell is switched LCs using insulator and rubbing angle without the patterned electrode.