1. Introduction

Owing to their potential applications in medical imaging, scientific visualization, surgical training or model prototyping, three-dimensional (3D) devices have generated an enormous interest among the scientific community in the recent years. In 1968, Sutherland provided the first demonstration of a stereoscopic 3D image using a head-mounted eye-wear, however it was not referred as stereoscopic image during its discovery [1]. The conventional stereoscopic image is device with a specialized optics that provides a separate imagery for each eye or a shifted view. However, the stereoscopic displays cause distortions like the accommodation-vergence mismatch which results in visual fatigue to the user [2–5]. Volumetric devices provide an unique solution to this problem with the ability to project a real-time spatial 3D image [6–11]. Geng demonstrates a volumetric 3D display based on digital light processing (DLP) engine that illuminates each voxel in the 3D space that can be isotropically viewed from all directions [12]. A solid-state and volumetric 3D display with interfering laser beams was demonstrated by Downing. In spite of coherent light sources, the optical efficiency of the interfering lasers was found to be as low as 1% and hence this type of volumetric 3D display could not be used for commercial applications [6]. More recently, Sadovnik and associates have patented their technology to visualize a 3D image with multi-layered liquid crystal (LC) based shutters that consists of polymer dispersed liquid crystals (PDLCs). The multi-layered optics forms a see-through device when it is inactive and individual layer scatters light on their electrical activation [13]. Similarly, Sullivan has demonstrated a volumetric 3D display with multi-planar LC based shutters to generate a 3D image [14,15]. However, the design prototypes for volumetric 3D display discussed in the literature so far requires extremely high operational voltage (>100V). Here we discuss the design of electro-optical shutters based on enhanced composition of reverse-mode polymer stabilized cholesteric texture that allows switching of only one shutter at once, unlike switching all but one in normal-mode polymer stabilized cholesteric texture, as discussed by Sullivan et al. This modification makes the volumetric 3D device more energy efficient. In addition, we designed the prototype of volumetric 3D device with low operational voltage.

In this work, we have presented a hardware prototype of a 3D display that generates volume-filled images using multi-planar optical element with time-multiplexed series of two-dimensional images. The optical element consists of layers of cholesteric liquid crystal (CLC) based optical shutter which are transparent in its off-state, and scatters light in its on-state. Figure 1 shows the layout of the volumetric 3-D display prototype built with a stack of 15 layers of optical shutters. The shutters are sequentially driven continuously across all the layers with time-multiplexed series of the 2-D images projected on the shutters with a high-speed DLP Pico projector (1). A neutral density filter (2) is used to attenuate the light intensity. As the projector does not have focusing lens, we have added a lens (3) to focus the projected image on the center of stack i.e. on the 8th layer. The stack of 15 shutters (4) are placed accordingly to achieve a focused image on its 8th layer and connected to the input pins of Arduino Uno (5) microcontroller to supply AC voltage to the shutters. The shutters are sequentially driven using Arduino Uno that controls the time and amplitude of the applied voltage. The projector is synchronized with the microelectronics controller to project each of the 15 images for a length of 4 ms. Simultaneously, all the 2-D images are projected on the shutters in a synchronized way so that the image is projected on the light-scattering layer, so the viewer (6) observes a single 3D image (7). The Visualization 1 in the supplementary data shows the video of 3D image observed from the volumetric 3D device illustrated in Fig. 1.

 

Fig. 1 The prototype of volumetric 3D display having projector (1), a neutral density filter (2), a focusing lens (3), a stack of optical shutters (4), Arduino Uno (5) with a viewer (6) that observes the generated 3D image (7).

Download Full Size | PDF

The optical shutters for volumetric 3D prototype are designed using transparent cholesteric films which exhibits high transmittance in its off-state, while it scatters light in its on-state. The transparent cholesteric film consists of a CLC with long helical pitch (> 1 µm) in planar alignment that is supported by polymer network. In off-state, CLC forms a Grandjean texture with the reflection band-gap in infra-red region and hence appears transparent. As the electric field is applied, the CLC switches to focal conic state that scatters incident light. On removal of the electric field, the films regain their initial transparent state. This technology is also known as reverse-mode polymer stabilized cholesteric texture (R-PSCT) [16]. Figure 2 shows the schematic diagram of a R-PSCT test cell with the LC alignment in transparent and scattered state.

 

Fig. 2 Schematic representation of a R-PSCT film in transparent (left) and scattered state (right).

Download Full Size | PDF

2. Materials and Methods

2.1. Material optimization for optical shutters

The formulated mixture consists of 93.2 wt.% nematic host HTG18500-100 (Merck), 0.4 wt.% chiral dopant R5011 (Merck), 6.1 wt.% reactive monomer and 0.3 wt.% photoinitiator Irgacure 651 (Ciba). According to the datasheet, the dielectric anisotropy of the nematic host is + 57.2 at 1 kHz and 25°C. The birefringence of the nematic host is 0.204 measured at 589 nm and 20°C. The LC used in this study has nematic-isotropic transition temperature of 97°C. The helical pitch of the CLC is 2.7 ± 0.1 µm. The mixture of two reactive monomers were prepared, that includes the commercially available RM257 (Merck) and laboratory synthesized RM6. The mixtures were prepared in proportions of 0/100, 20/80, 40/60, 60/40, 80/20 and 100/0, where the former number denotes the proportion of RM6 and later number denotes the proportion of RM257, in the total quantity of reactive monomer. The homogeneous mixture is capillary-filled in the LC test cell in their isotropic state. The test cell consists of two glass substrates with their coated with conductive indium-tin-oxide (ITO) and a mechanically rubbed layer of the spin coated polyimide PI2555 (Nissan) for the planar alignment of LC molecules, as shown in Fig. 2. The gap between the two glass substrates is controlled with the glass spacers to achieve the cell gap of 13.5 ± 0.1 µm. As the mixture is cooled, the CLC aligns to form a Grandjean texture. Later, the test cell is exposed to UV light using a metal halide lamp (Loctite) having intensity of 0.05 mW/cm² for 12 hours, to uniformly polymerize the reactive monomers within the test cell. The UV polymerization is conducted under the ambient conditions.

2.2. Fabrication of volumetric 3D

The optical shutters are prepared with two glass substrates having dimensions 2” X 2” X 1.1mm (thick). The optimized CLC mixture is sandwiched between the two ITO-coated glass substrates and UV photopolymerized as described above. The two substrates of optical shutters are connected to the positive and negative input pins in the Arduino. A series of 15 images are projected from the projector using a USB flash drive. Arduino is programmed for sequential driving of optical shutters that switch to scattering state. Synchronized with sequential driving, 15 time-multiplexed images of ‘red circle’ are projected on the shutters. The image quality of projection is controlled using a neutral density filter and a focusing lens.

3. Results and Discussions

3.1. Static electro-optic response of R-PSCT

The static electro-optic responses of R-PSCT cells with 100% RM6 and 100% RM257 are shown in Fig. 2. The R-PSCT test cell is exposed to coherent beam from He-Ne Laser (λ = 633 nm) with a photodetector used to collect the uniaxially transmitted photons. The output light from the laser source was inherently unpolarized. The R-PSCT test cells are initially in Grandjean state with high transmittance (> 85%). The test cells are 20 cm away from the photodetector. The photodetector has the aperture diameter of 10 mm. Based on the geometry of aperture, the solid angle over which photons are collected is 3°. As the applied voltage increases, the cells are switched to the low transmittance focal conic state. On reverse traversing of the applied voltage, electro-response follows the similar path and exhibits a low hysteresis of the test cell. With ~6% concentration of polymer, surface area of the network per unit volume is large enough to relax the CLC back to planar alignment during the ramp down of the applied voltage, as shown in Fig. 3a & b. Hysteresis of the R-PSCT cells is calculated using Eq. (1):

 

Fig. 3 Static electro-optic response of R-PSCT containing 100% RM6 [a] and 100% RM257 [b]. Transmittance at 0 V and contrast ratio of the R-PSCT test cell as a function of proportion of RM257 in the reactive monomer mixture [c].

Download Full Size | PDF

3.2. Morphology of polymer network in R-PSCT

Figure 4[a] and [b] are the scanning electron microscopic (SEM) photographs of the polymer network structures of RM6 and RM257 respectively. The RM6-based network typifies a discrete rice-grain like structure with large pores and less-dense network. The RM257-based network, in contrast, forms a dense network with finely stranded and highly interconnected chains. In terms of their chemical structure, RM6 has a long flexible alkyl chain at the terminal ends, as compared to RM257. Hence, the polymer morphology that forms after photopolymerization are distinctly different for both. Here, we also want to mention that, we achieved rice-grain like morphology with RM6 only with long exposure times and low UV intensity. Thus, the R-PSCT with 100% RM6 has a low driving voltage (~11V) and the R-PSCT with 100% RM257 has a much higher driving voltage (~25V). The anchoring strength of the RM6 polymer network is also much lower that allows a randomized focal-conic structure to realize a higher CR. Whereas, a dense polymer network in RM257 limits randomization of the focal-conic structure and hence exhibits lower CR in the R-PSCT with 100% RM257. With inclusion of RM257 in the reactive monomer mixture, the density of polymer network increases and subsequently, the CR decreases, as shown in Fig. 3[c].

 

Fig. 4 SEM photographs of polymer network formed with 100% RM6 [a] and 100% RM257 [b].

Download Full Size | PDF

3.3. Electro-mechanical stress analysis of R-PSCT test cells

The prepared R-PSCT test cells were exposed to the electro-mechanical (E-M) stress test to study the mechanical stability of the polymer network. In E-M stress test, the R-PSCT cells were applied with driving voltage of the individual test cells (voltage at minimum transmittance) at 1 Hz frequency, so that the configuration of the CLC changes from planar-to-focal conic-to-planar, as shown in Fig. 5[a]. The test was carried out for 72 hours and during which the R-PSCT cell switches for more than 150,000 times. This experimental design allows us to analyze the lifetime of the R-PSCT cells for their application in multi-layered 3D visualization. The performance of the R-PSCT test cells was based on the comparison of the loss in transmittance at 0V and optical contrast that is measured before and after the E-M stress test, as shown in Fig. 5[b]. The transmittance at 0V in R-PSCT with 100% RM6 reduced to ~60% after E-M stress test from ~85% before E-M stress test. However, upon inclusion of RM257 in the polymer network in the other test cells, the loss in transmittance at 0V was almost negligible even after the E-M stress test. During the E-M stress test, the polymer network in R-PSCT with 100% RM6 possibly lost the planar alignment and that allowed only partial recovery of CLC to its Grandjean state. This lead to a low transmittance in off-state for the R-PSCT test cell with 100% RM6. Due to the lower transmittance at 0V, the CR of the test cell also decreases to 7 from ~14 before the E-M stress test. Figure 5[c] shows the cell images of R-PSCT test cell in off-state with 100% RM6 which has a hazy texture, whereas the test cell with 20% RM257 + 80% RM6 has a transparent texture. Although there is a trade-off in the optical contrast, we found that optimum composition of the reactive monomers for electro-mechanically stable R-PSCT cell is 20% RM257 + 80% RM6.

 

Fig. 5 [a] Driving scheme for electro-mechanical stress test. [b] Transmittance at 0V and contrast ratio as a function of proportion of RM257 measured before and after electro-mechanical stress test. [c] Cell images of R-PSCT test cell with 100% RM6 (left) and 80% RM6 + 20% RM257 (right) at 0V after E-M stress test.

Download Full Size | PDF

3.4. 3-D imaging from volumetric 3D display

The stack of R-PSCT films used as the multi-planar optical element in the volumetric 3D display prototype comprises of CLC and the mixture of reactive monomers that includes 80% RM6 + 20% RM257. The response time for the individual R-PSCT film was measured between 0 V and 24 V at 250 Hz, as shown in Fig. 6[a]. The turn-on time for the R-PSCT test cell measured between 10% and 90% transmittance of the cell was 1.1 ms and turn-off time was 1.8 ms. As the response time of the R-PSCT films is less than < 4 ms, the R-PSCT films can be driven at 250 Hz. The Arduino used to program the sequential switching of the shutters was functionalized in a way to have a time lag of around 4ms between switching of two consecutive shutters, so that one shutters can turn on and turn off before the next shutter turns on. The Arduino drove the shutters to switch from 1 to 15 and back to 1, in the same order. For the 3-D display system, wide viewing angle is one of the most important performance parameters. Hence, the optical contrast i.e. ratio of transmittance at 0V and 24V was measured from −60° to + 60° (left to right), as shown in Fig. 6[b]. The optical contrast shows a negligible variation across a wide viewing angle for the R-PSCT film. With almost unaltered optical contrast in the cone of 120° for R-PSCT film, the stack of R-PSCT film can be utilized as the optical element for 3-D display system for volumetric imaging in the near-field and far-field devices. Figure 6[c] show the plot of transmittance as a function of wavelength for optical shutters used in volumetric 3D device. The color capability shows that off-configuration has no wavelength dependence for the transmittance, however in on-state, the optical shutters show slightly higher transmittance at longer wavelength as compared to shorter wavelength. There is a possibility of absorbance of light at shorter wavelength by some cholesteric domains in the cell. However, this effect do not contribute significantly in color breakup of the display, hence can be ignore for this application.

 

Fig. 6 [a] Time response for the optical shutter designed for volumetric 3D display measured between 0V and 24 V. [b] Optical contrast of the shutter between transparent state (0 V) and light-scattering state (24 V) at wide viewing angle. [c] A plot of transmittance vs wavelength measured for optical shutter designed for volumetric 3D display.

Download Full Size | PDF

4. Conclusions

In this article, we have studied the electro-optical switching in reverse-mode polymer stabilized cholesteric texture comprising of a polymer network formed using a combination of two reactive monomers RM6 and RM257 in different compositions. We have observed the strong influence of a polymer network on the optical contrast of the R-PSCT films. From this study, we found that RM6 polymer network shows a weak electro-mechanical stability that leads to hazy film after multiple switching cycles. However, the addition of RM257 in the mixture imparts the mechanical stability that maintains high transmittance (> 85%) of R-PSCT films in the off-state after multiple switching cycles. The optimum composition of reactive monomer for R-PSCT was found to be 80% RM6 + 20% RM6 that yields the optical contrast of ~7, turn-on time 1.1 ms, turn-off time 1.8 ms and wide viewing angle with a cone of 120°. Further, we also designed a 3-D display system for volumetric 3D imaging that consists of 15 layers of R-PSCT films. We have successfully achieved the generation of a 3D image that can be perceived without any eyewear and under ambient light. A volumetric 3-D image can be created from time-multiplexed series of 2-D images using a sequential switching stack of R-PSCT films and digitally synchronized with 2-D image projection. We wish to continue our work on developing the prototype for generation of high resolution imaging with pixelated layer structure. The prototype in its present form limits the fast projection of 2-D images to create a blur-free 3-D image due to the use of commercial DLP projector. Using parallelly-programmed algorithm, a better synchronization can be developed between the projected images and switching of the individual shutters, which is currently lacking.