Progress toward a liquid crystal contact lens display

Progress toward a liquid crystal contact lens display Jelle De Smet (SID Student Member) Aykut Avci (SID Student Member) Pankaj Joshi (SID Student Member) David Schaubroeck Dieter Cuypers (SID Member) Herbert De Smet (SID Member) Abstract — A contact lens embeddable display using electro-optic modulation was designed and fabricated. Using a guest–host liquid crystal configuration, a spherically deformed liquid crystal cell was fabricated comprising poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT : PSS) as a conductive layer and obliquely evaporated SiO2 as an alignment layer. An additional SiO2 buffer layer was evaporated on top of the PEDOT : PSS to overcome compatibility problems with the patterning of the photolithographically defined spacers. Although the contrast is modest, a patterned modulation could clearly be observed, indicating that our approach and fabrication process could eventually lead to a fully pixelated contact lens display. Keywords — contact lens, flexible display, guest–host, liquid crystal cell, PEDOT : PSS, oblique SiO2 evaporation. DOI # 10.1002/jsid.188 1 Introduction Today’s modern society cannot be contemplated without noticing the prevalent use of displays in our daily lives. With each technical evolution they underwent in the last decades, displays evolved from a rare and bulky piece of electronics to slim and ergonomic devices that are omnipresent. The rise of liquid crystal displays (LCDs) and, more recently, organic light-emitting diode displays also revolutionized our possibilities to interact with each other and the way we consume entertainment. Their integration into smartphones, tablets, and portable media devices allows people to remotely communicate in new ways (e.g., video calling), have instant and easy access to any kind of high content information, and watch high definition movies at anytime and anyplace they want. A clear trend can thus be identified in which displays are becoming a highly personalized gateway, allowing for a closer interaction with many aspects of our daily lives. One can wonder what the next step in this evolution will be. When seeking to further integrate displays in our daily routine, one should strive for benefits such as a hands-free, instant access to our display, a lightweight and almost unnoticeable device, and even a display providing a superimposed view onto our real view of the world. Although some of these benefits are being realized by head-mounted displays, even the most modern implementations require goggles that are significantly heavier than common glasses. Moreover, most people are reluctant to wear glasses if they normally do not need them, which is also one of the reasons why the adoption of stereoscopic 3D TV is slow. One possible way of overcoming these hurdles is bringing the display even closer to the eye and integrating it into a contact lens. Such a contact lens display aims for a bigger coalescence of the man–machine interface and would pursue an enhancement of the real view, rather than feeling as just an add-on (which is the case for head-mounted displays). Of course, there are many technological challenges in realizing a contact lens display. Aside from power management, focusing issues, and biocompatibility requirements, the main question arises, however, which display technology would be best suited for integration into a contact lens. A considerable amount of research has been performed by the group of B. Parviz1,2 on the integration and powering of micro-lightemitting diodes into a contact lens. Although a single-pixel wireless contact lens display has been demonstrated, the emissive nature of the technology might hamper the development of a high resolution display because of its considerable power consumption. Even if the energy budget can be met, the fabrication and placing of each micro-light-emitting diode can be quite cumbersome. Alternatively, one can incorporate an electro-optic element capable of modulating incoming light, rather than emitting new light, thus possibly reducing the required power. An additional benefit of this approach is that the number of pixels and their surface area can easily be changed by the patterning process of the conductive layer present in many of these electro-optic elements. This versatility allows the development of applications with a broad range of pixel counts and sizes, ranging from a one pixel, fully covered contact lens to a highly pixelated contact lens display. In this paper, we present our research on the design and fabrication of a modulating contact lens display, focusing on the achievement of the display effect. Note that, although technically speaking the integration of a pixelated optical modulator in the lens constitutes the realization of a contact lens display, this in itself does not suffice to generate a controllable image on the retina of its wearer because the eye is not capable of focusing on objects closer than about 25 cm. In what follows, we will briefly present some suggestions to achieve that goal. Received 12/10/12; accepted 10/09/13. J. De Smet, A. Avci, P. Joshi and H. De Smet are with Ghent University, Department of Electronics and Information Systems, Centre for Microsystems Technology (CMST), Technologiepark 914 A, 9052 Ghent, Belgium; e-mail: [email protected]. D. Schaubroeck, D. Cuypers and H. De Smet are with Imec–CMST, Technologiepark 914 A, 9052 Ghent, Belgium. © Copyright 2013 Society for Information Display 1071-0922/14/2109-0188$1.00. Journal of the SID 21/9, 2014 399 2 2.1 Design Preliminary considerations Prior to the selection of the electro-optic effect on which the display will be based on and how it will be fabricated, one has to consider the constraints that concur with embedding the display in a contact lens. Firstly, the outer material must be biocompatible, as it will be in direct contact with the epithelial cells of the cornea. However, the inner materials are not bound by this restriction and can be chosen on the basis of their manufacturability, as long as the final fabrication step consists of embedding in a biocompatible material. Secondly, the total thickness should be restricted to the bare minimum. Although thickness is not a regulated parameter in lens design (as, e.g., by the International Organization for Standardization (ISO)3 or the Association of Optometric Contact Lens Educators (AOCLE)4) and is mostly a consequence of the desired dioptre, it is generally kept as small as possible to enhance user comfort. Moreover, the edge thickness of the contact lens should not exceed 140 μm as this would create discomfort for the lens wearer while blinking.5 Finally, in the long term, one has to assure that a sufficient oxygen transmission through and beneath the contact lens is present, as this is required to keep the surface of the eye in a healthy state.6 2.2 Selection of the electro-optic effect Although there is a variety of electro-optic effects capable of modulating the incoming light (e.g., electrowetting, electrochromism, and electrophoresis), we investigated a liquid crystal (LC)-based technology because of its low driving voltage and power consumption. Because the total thickness is limited, a polarizer-free approach is needed as no sufficiently thin polarizers are commercially available at present. Therefore, a transmissive guest–host configuration as proposed by White and Taylor7 was chosen with a homeotropic alignment and a twist angle of 180°. The homeotropic alignment leads to a normally white mode, which is preferred because in case of a power malfunction, the vision of the contact lens wearer should be preserved. Furthermore, the 180° twist angle was chosen to achieve a higher contrast while avoiding the known hysteresis effect.8 A conceptual drawing of how such an LC-based contact lens display may look like is presented in Fig. 1. The central part of the contact lens display comprises a spherically shaped LC cell, filled with the guest–host LC mixture, and represents the most challenging part. FIGURE 1 — Conceptual drawing of the contact lens display. 400 De Smet et al. / A liquid crystal contact lens display 2.3 Design of the liquid crystal cell Inspired by existing flexible displays, our LC cell basically consists of two polymer films that are glued together and cut in circles, bearing photolithographically defined spacers in between. These circular cells are then molded using a spherical aluminum mold, leading to the desired shape of the LC cell. In previous work, we studied the mechanical design of such a spherically shaped LC cell and investigated the correlation between the thicknesses of the polymer films being used and the cell’s final smoothness.9 This design included features in the photolithographic layer (Fig. 2) addressing some problems specifically related to the spherical shape and dimensions of the cell. Besides the spacers that are keeping the cell gap (10 μm) fixed, a circular barrier was added to control the inward glue flow during processing, and the entrance of the cell through which it is filled with LC was reinforced using an interlaced finger structure. When using polyethylene terephthalate (PET) as the polymer material, we found that a 50 μm top film (convex side) in combination with a 75 μm bottom film (concave side) provided the best results in terms of smoothness and reduced final thickness. PET was chosen because its chemical properties along with its strength and easy moldability made it a good candidate for our fabrication process. However, because the oxygen transmission of PET is low in comparison with existing contact lens materials, it should be replaced by a more suitable material in the future. As for the remaining dimensions such as diameter and curvature, we chose them to be as generic as possible to allow for embedding of the LC cell in a variety of contact lens materials and shapes. Rigid gas permeable (RGP) contact lenses generally have an overall diameter of 9 mm and a curvature of 7.8 mm,4 which is the mean diameter of a human cornea. Because of their flexibility, soft contact lenses tend to have larger diameters, between 13 and 15 mm, and larger curvatures. Our LC cell was therefore designed with a curvature of 7.8 mm FIGURE 2 — Mask pattern of our photolithographic layer. It comprises a circular area filled with a matrix of vertical cylinders, a 100 μm barrier to control the glue flow, and an interlaced finger structure to reinforce the entrance of the liquid crystal cell. and a diameter of 8 mm. A cell with a curvature of 7.8 mm allows for embedding in most RGP lenses and in practically all soft lenses, because larger curvatures are less demanding for the LC cell. A diameter of 8 mm also allows for both RGP as soft contact lens embedding with the additional remark that having a larger diameter might be less useful as the maximal pupil size for most humans is 8 mm.10 3 Active cell layers As in most LCDs, a transparent conductive layer and an alignment layer need to be incorporated into the LC cell. The inclusion of these layers should be compatible with the current cell design and fabrication, in particular with the processing of the photolithographic layer. In the following paragraphs, the material selection and their integration into the existing process flow will be discussed. The exact fabrication process will be described in the subsequent section. 3.1 Alignment layer Application of the homeotropic alignment layer should preferentially be contact-free, that is, without a rubbing step, because this would likely harm the PET film surface. Moreover, because the PET films should not be heated above 150°C before molding, many known methods using spin coated materials are excluded as they require a heating step above this temperature. Therefore, we chose oblique SiO2 evaporation as a contact-free, low-temperature alternative.11 This process results in a homeotropic alignment for the LC molecules, with a small tilt angle to increase the switching speed. Normally, the spacers of an LC cell are put on top of the alignment layer, but because of the nature of the SiO2 layer, the development process of the photolithographic spacers is likely to interfere with the quality of the alignment. Instead, the SiO2 layer can be applied after deposition and patterning of the photolithographic spacers. 3.2 Transparent conductive layer Indium tin oxide (ITO) is the standard transparent conductive layer used in rigid LCDs, but because of its brittle nature, its use on flexible films is limited. Many studies about mechanical deformation of ITO layers on flexible films reveal an increase in their resistance with decreasing bending radius or increasing cyclic loading number, mainly attributed to crack formation and propagation. An overview is provided in Reference 12 and references therein. Because of the extreme deformation process and small radius of our contact lens-shaped LC cells (7.8 mm), the cracking of an embedded ITO layer in these cells seemed inevitable. Nevertheless, preliminary test cells with ITO were fabricated because its use in conformable displays, such as our LC cell, has been suggested by other authors.13 As can been seen in Fig. 3, the test cells indeed revealed a cracked ITO layer, FIGURE 3 — Misalignment of the liquid crystals in some areas that were delineated by crack lines. but more worrisome was the misalignment of the LCs in some areas that were delineated by crack lines. Although this suggests that the ITO and/or the alignment layer delaminate during the molding step, the exact nature of this phenomenon was not studied in detail. Nevertheless, it is clearly detrimental to the overall optical quality of the LC cell, prompting the use of other flexible conductive materials over ITO. We therefore selected poly(3,4ethylenedioxythiophene) : poly(styrenesulfonate) (PEDOT: PSS) as an alternative, being a conductive polymer with a proven flexibility.14 A commercial solution (OrgaconTM S305 by AGFA, Mortsel, Belgium) was chosen, and the deposition parameters were optimized, resulting in a 110 nm thin layer with a conductivity of about 200 ohms/square. 3.3 Photolithographic layer and barrier layer The choice for SiO2 evaporation and PEDOT: PSS resulted in the requirement to develop the photolithographic layer on top of the PEDOT: PSS. We had selected SU8 3010 (MicroChem, Newton, Massachusetts, USA) in our study of the mechanical design of our LC cell, as this is a chemically and thermally stable photoresist suitable for permanent application. However, the creation of a patterned SU8 (photoacid generator containing photoresist) layer15 directly on PEDOT: PSS is not straightforward because of two main reasons.16,17 First, PEDOT: PSS films can be damaged by aqueous solutions or certain solvents that are used as developers in conventional photolithography. Second, damage of the unexposed photoresist layer is observed when a photoacid generator containing photoresist is used because of acid diffusion. The acid diffusion phenomenon is described as the proton diffusion from the PEDOT: PSS layer into the unexposed photoresist layer causing decomposition of the acidlabile photoresist. This affects the resolution and the overall quality of the patterned image. To circumvent this problem, an additional SiO2 barrier layer was evaporated on top of the PEDOT: PSS. A thickness of 50 nm proved to be sufficient to prevent diffusion from the PEDOT: PSS to the SU8 3010 layer and does not interfere with the flexibility of the stack. Journal of the SID 21/9, 2014 401 4 Fabrication 4.1 Poly(3,4-ethylenedioxythiophene) laser patterning The design and realization of an active matrix multipixel display is of course beyond the scope of this explorative research, but a simple patterned PEDOT: PSS layer in combination with passive addressing can already provide a proof of concept. Out of the many patterning methods for organic materials, such as photolithography, embossing, or inkjet printing,18 we chose laser ablation as our preferred means. For this purpose, the output power of an excimer laser was carefully tuned (1 pulse, 3–7 ns full width at half maximum pulse width, 248 nm wavelength, 125 mJ/cm2 fluence) in order to selectively pattern the PEDOT: PSS layer, while minimizing the damage to the underlying PET film. As shown in Fig. 4, a pattern can be formed by firing individual pulses next to each other, thus electrically isolating different areas. However, because the minimum beam size attainable by our equipment was 40 μm, only coarse patterns can be formed in this stage. Furthermore, the homeotropic alignment seems to be affected, as the ablated line is slightly visible in the voltage off stage (Fig. 7). Some preliminary scanning electron microscope analyses suggested that this misalignment was due to topological defects in the PET film introduced by the laser ablation process, but a more detailed description will be given in ensuing work. Nevertheless, this maskless approach was deemed beneficial in this exploratory phase because of its flexibility, whereas the resolution can be further improved by using more advanced laser ablation instruments. 4.2 Process flow The fabrication process starts with the lamination of two cleaned PET films (50 and 75 μm) onto a glass carrier using a temporary adhesive (Fig. 5), after which they were plasma FIGURE 4 — Patterning of the poly(3,4-ethylenedioxythiophene):poly (styrenesulfonate) layer can be performed by firing individual pulses next to each other, thereby electrically isolating the different areas. Each ablated circle has a diameter of about 40 μm. 402 De Smet et al. / A liquid crystal contact lens display FIGURE 5 — Process flow. (a) The polyethylene terephthalate (PET) films are laminated onto a glass carrier; (b) poly(3,4-ethylenedioxythiophene) (PEDOT) is spin coated on the PET films and is patterned by excimer laser ablation; (c) the buffer layer is vertically evaporated on top; (d) the photolithographic layer is applied on the 75 μm PET film; (e) the alignment layer is obliquely evaporated; (f) the glue is deposited just next to the glue barrier and a 50 μm PET film including the PEDOT, the buffer layer. and the alignment layer is put on top; (g) the total stack is pressed together and illuminated with ultraviolet (UV) light; (h) after removal of the glass carriers, the lenses are cut out through the glue joint with a CO2 laser and the entrance is reopened with an excimer laser; (i) the lenses are molded with a spherical aluminum mold; and (j) after molding, a spherically conformed liquid crystal cell is formed, which can then be filled and sealed. treated to improve the wettability of the PEDOT: PSS solution. Next, PEDOT: PSS was spin coated on top of the PET films and was patterned by excimer laser ablation. A 50 nm thick buffer layer of conventionally evaporated SiO2 was deposited onto the PEDOT: PSS, and a 10 μm thick photoresist (SU8 3010) was spin coated and patterned on top of the 75 μm PET film, using the design mentioned earlier. Both PET films then received the alignment layer by obliquely evaporating SiO2. A flexible, ultraviolet-curable glue (UVS 91, Norland Products, Cranbury, New Jersey, USA) was deposited just next to the circular barrier with an automatic dispenser, creating a circular glue joint with a small interruption at the interlaced finger structure. Afterward, the 50 μm PET film was put on top, and the total stack was then pressed together, ensuring a close contact as the stack was subsequently illuminated with ultraviolet light to cure the glue joint. After curing, the glass carriers were removed, and the LC cells were cut out through the glue joint with a CO2 laser. Additional rectangular contact areas were provided at the side of the LC cell that, after fabrication, were connected to a power source with a copper tape. Because of the thermal ablation mechanism of the CO2 laser, the opening of the LC cell was inadvertently closed in most of the samples and needed reopening. An excimer laser was accordingly set to ablate material away at the edge of the contact lens, going through the first PET film but stopping 10 μm deep in the second film, thus reopening the LC cell entrance. The flat, cut-out lenses were then molded in an aluminum mold with a curvature radius of 7.8 mm. The molding process started with heating the whole mold to a temperature of 175°C, well above the glass transition temperature of PET (75°C). When the mold reached this temperature, the lenses were placed onto the bottom (concave) part of the mold, and a soaking time of several minutes was applied. Afterward, the upper part of the mold was placed on top and pressed together with a 0.1 kg/cm2 pressure. The total ensemble was then allowed to gently cool down until it was assured that the temperature of the mold was again well below the glass transition temperature of PET before removing the lenses from the mold. After molding, the LC cells were filled with the guest–host LC mixture with a vacuum filling method and were sealed with the same glue used for the glue joint. An example of such an active LC cell is displayed in Fig. 6, where it is placed next to a commercial RGP contact lens for comparison. 5 5.1 FIGURE 7 — Patterned modulation of our contact lens display. Top: voltage off; bottom: voltage on. Contact lens display Active switching Modulation of the incoming light corresponding to the designed pattern could clearly be observed (Fig. 7) when a voltage was applied to the LC cell, suggesting that our approach and our fabrication process could lead to a contact lens display. A few minor points still need improvement. A closer look at the spacers in the voltage off state (Fig. 8) reveals that the alignment in the surroundings of the spacers is not FIGURE 6 — An active liquid crystal cell placed next to a commercial rigid gas permeable contact lens. FIGURE 8 — The matrix of vertical spacers after filling the cell with liquid crystal. Top right: a close-up of such a spacer. The arrow represents the evaporation direction. perfect. Because of the oblique evaporation, a shadow zone is present behind the spacers, as this area cannot be reached by the SiO2 molecules, and as a result, the LC molecules will not align homeotropically. The dark band observed at the front side of the spacers stems from the conflicting alignment directions imposed at the front side of the spacers and the bottom of the LC cell. The LC molecules will align perpendicular onto the spacer wall, and a transient zone can be observed until the LC molecules are again homeotropically aligned with respect to the bottom of the LC cell. The Journal of the SID 21/9, 2014 403 potential negative impact of this effect will depend on the application or, more in particular, on the area of the affected area versus the pixel size needed for the application. 5.2 Contrast measurement Contrast measurements (Fig. 9) were made by focusing the light of a xenon arc lamp onto a small spot on an unpatterned LC cell and reading out the modulated light with an integrating sphere and an Ocean Optics (Dunedin, Florida, USA) HR 2000+ spectrometer. The final contrast is determined by the thickness of the cell (10 μm), the dichroic ratio of the dye (14:1), the weight concentration of the dye in the LC (3 wt %), and the twist angle of the LC in the cell (180°). The values of these parameters resulted in a peak contrast of 2.2:1, when the cells were driven near to their maximum attainable modulation (square wave, f = 30 Hz, Vpp = 10 V). Next to this, the absolute transmission in the voltage off state was found to be around 50%, which also includes both PET films. 6 Discussion The main goal of this work was to fabricate a modulating contact lens display, effectively constituted by a spherically shaped LC cell capable of showing a patterned transmission. Apart from a good mechanical design, an equally important material selection was made for the active layers of the cell. The choice of a conductive polymer such as PEDOT: PSS over ITO was indispensible because of the extreme deformation during the molding step. The PEDOT: PSS shows no optical defects after the molding process, while its function as a conductive layer is kept intact. However, its integration into our process flow was not trivial, because there is no standardized way of patterning PEDOT: PSS on top of PET, and photolithographic patterning on top of PEDOT: PSS is not straightforward. Our approach using laser ablation has the advantage of being versatile, but the current limited resolution FIGURE 9 — Transmission and contrast measurement of an unpatterned liquid crystal cell, revealing a peak contrast of 2.2:1 and an absolute transmission of about 50% at 550 nm. 404 De Smet et al. / A liquid crystal contact lens display and its influence on the homeotropic alignment indicates that other solutions should also be investigated in the future. Of course, alternatives for the SiO2 alignment layer could alleviate the problems concerning the disturbed alignment in the ablated areas and around the spacers. For instance, an approach using no alignment layer by mixing nanoparticles in the guest–host mixture19 could be considered. Another possibility might be the use of an alignment layer that simultaneously acts as a buffer layer between PEDOT: PSS and SU8 3010. Given the limited temperature budget, only a few alignment techniques are possible, with one of the most prominent being photoalignment.20 Next to this, photoresists are being developed that can be directly patterned on top of PEDOT: PSS,16 but as they are not designed for permanent application, the use of a buffer layer in combination with SU8 3010 seems preferential at this moment. When assessing the properties of the display, we notice that the contrast of the display is rather low. Although some parameters affecting the contrast could be improved, no major increase is expected without a negative impact on the switching speed when the use of polarizers is excluded. As mentioned earlier, a polarizer-free approach was chosen because commercial polyvinyl alcohol-based polarizers were deemed too thick to be used in a contact lens. Alternatively, thin-film polarizers could be used, as they generally have a thickness below 1 μm and could directly be applied on top of our PET film.21,22 Although many of them do not perform as well as commercial polyvinyl alcohol polarizers, their increased thermal stability22 or easy photo-patternability23 could prove advantageous for integration in a contact lens display. With an improved contrast, some direct applications can already be envisioned, although these would be mainly medically related. When patterned in concentric circles, the lens could be used as an artificial iris, replacing the functionality of a defective iris by regulating the light transmission toward the retina. Alternatively, it could be used as an iris with tunable color, being a more cosmetic application. Notwithstanding that these applications are not the end goal being a fully autonomous contact lens display, solving the common remaining problems such as biocompatibility and energy supply and storage can certainly pave the way toward it. As mentioned in the Section 1, the integrated display cannot be observed clearly by a person wearing this contact lens because the eye cannot focus on an object so close to its surface. Hence, an additional optical structure should refocus the incident light onto the retina. Possible solutions by placing a lens-like structure behind each pixel such as Fresnel zone plates24 or LC lenses25 have been suggested, but simulations26 indicate that such a discrete approach will not result in a satisfactory resolution on the retina. Likely, an integral approach using, for example, holographic components is necessary, but this will require further research. Regarding biocompatibility, apart from sufficient oxygenation of the cornea, diffusion of species from the LC cell to the eye or from the surroundings into the LC cell is a major concern. Potentially hazardous materials should be confined within the cell, while the cell should be protected from the detrimental effects of water and oxygen. Most likely, an approach using various barrier layers will be necessary and should be the focus of a follow-up study aiming for full biocompatibility. Providing autonomy to the lens is also an important issue, and although Pandey et al.1 proved that an electronic contact lens can be powered wirelessly by electromagnetic induction, the continuous need of a secondary coil does not constitute full autonomy. This approach should be combined with energy harvesters, such as solar cells, and energy storage components, such as microbatteries, so the lenses can be charged when not inserted in the eye, much like electronic devices such as electric toothbrushes or mobile phones. 7 Conclusion A contact lens embeddable display using electro-optic modulation was designed and fabricated. Because our design was based on a guest–host LC configuration, a spherically deformed LC cell was fabricated, giving special attention to the selection of the active layers and their inclusion into the fabrication process. Although PEDOT: PSS performed well as a conductive layer, compatibility problems with the processing of SU8 on top of this layer required the use of an additional SiO2 buffer layer. Notwithstanding the modest contrast, a patterned modulation could clearly be observed. Subsequent research could be aimed at increasing the contrast by, for example, including thin-film polarizers. Next to this, increasing the number of pixels and investigating on how to focus the image onto the retina are the most prominent display related research topics that need to be addressed. 8 Acknowledgment The authors would like to thank Steven Van Put for his assistance on the laser ablation instruments. References 1 J. Pandey et al., “A fully integrated RF-powered contact lens with a single element display,” IEEE Trans. Biomed. Circuits Syst. 4, No. 6, 454–461 (2010). 2 A. R. Lingley et al., “A single-pixel wireless contact lens display,” J. Micromech. Microeng. 21, 125014 (2011). 3 ISO 14534:2011, ISO 18369–1:2009, ISO 18369–2:2006, ISO 18369–3:2006, ISO 11978:2000, ISO 11980:2009, ISO 19979:2004; ISO standards on contact lenses. 4 Contact lens clinical pocket guide, 4th ed., AOCLE, http://www.aocle.org (2012). 5 N. Soubry, Belgian optical supply, Noordlaan 24, Kuurne, Belgium, private communication, (2011). 6 M. M. Hom and A. S. Bruce, Manual of contact lens prescribing and fitting, 3rd ed., Elsevier Health Sciences, St. Louis, Missouri, USA (2006). 7 D. L. White and G. H. Taylor, App. Phys. Lett. 45, 4718 (1974). 8 H. Ikeno et al., “A reflective guest-host LCD with 4096-color display capability,” SID Tech. Digest 28, 1015 (1997). 9 J. De Smet et al., “Design and wrinkling behavior of a contact lens with an integrated liquid crystal light modulator,” J. Disp. Technol. 8, No. 5, 299–305 (2012). 10 A. B. Watson et al., “A unified formula for light-adapted pupil size,” J. Vision 12, No. 10, 1–16 (2012). 11 K. Hiroshima, “Controlled high-tilt-angle nematic alignment compatible with glass frit sealing,” Jpn. J. Appl. Phys. 21, No. 12, L761–L763 (1982). 12 G. P. Crawford, Flexible flat panel displays, Wiley-SID Series in Display Technology, Chichester, West Sussex, England, Ch. 6–9, (2005). 13 D. R. Cairns et al., “Conformable displays based on polymer-dispersed liquid-crystal materials on flexible substrates,” J. Soc. Inf. Disp. 11, 289–295 (2003). 14 S. J. Gorkhali et al., “Reliability of transparent conducting substrates for rollable displays: a cyclic loading investigation,” J. Soc. Inf. Disp. 12, 45–49 (2004). 15 J. M. Shaw et al., “Negative photoresists for optical lithography,” IBM J. Res. Dev. 41, No. 1/2, 81–94 (1997). 16 P. G. Taylor et al., “Orthogonal patterning of PEDOT:PSS for organic electronics using hydrofluoroether solvents,” Adv. Mater. 21, 2314–2317 (2009). 17 J.-K. Lee et al., “Acid-diffusion behaviour in organic thin films and its effect on patterning,” J. Mater. Chem. 19, 2986–2992 (2009). 18 E. Menard et al., “Micro- and nanopatterning techniques for organic electronic and optoelectronic systems,” Chem. Rev. 107, No. 4, 1117–1160 (2007). 19 W.-T. Yeng et al., “Nanoparticle-doped guest-host liquid crystal displays,” Opt. Lett. 33, No. 15, 1663–1665 (2008). 20 V. G. Chigrinov et al., “Photoalignment of liquid crystalline materials,” John Wiley & Sons, Chichester, West Sussex, England, (2008). 21 W. C. Yip et al., “Photo-patterned e-wavepolarizer,” Displays 22, 27–32 (2001). 22 Y. Brobov et al., “Environmental and optical testing of thin crystal film polarizers,” J. Soc. Inf. Disp. 11, No. 1, 63–70 (2003). 23 X. Zhao et al., “High-resolution thin ‘guest-host’ micropolarizer arrays for visible imaging polarimetry,” Opt. Express 19, No. 6, 5565–5573 (2011). 24 R. Mirjalili and B. A. Parviz, “Microlight-emitting diode with integrated fresnel zone plate for contact lens embedded display”, J. Micro/ Nanolithogr., MEMS MOEMS 11, No. 3, 033010 (2012). 25 S. Valyukh et al., “A liquid-crystal-lens-array based projection system for near eye displays,” Proc. 19th IDW/AD, 1305–1308 (2012). 26 H. De Smet et al., “A contact lens with built-in display: science fiction or not?” Proceedings of the 33rd IDRC – EuroDisplay 2013, (2013), 8–11. Jelle De Smet graduated with a degree in physics engineering at Ghent University in 2007. Since January 2009, he is pursuing a PhD at that same university at the Department of Electronics and Information Systems under the supervision of Prof. Herbert De Smet, where his research is focused on electronic contact lenses and contact lens display technology. Jelle De Smet is a Society for Information Display (SID) student member, vice-president of the SID Ghent-Lowlands Student Branch, and author or co-author of five publications in international journals and 10 publications in international conference proceedings Aykut Avci received his BS degree in computer engineering from Canakkale 18 Mart University at Canakkale, Turkey, in 2002. He obtained the MS degree from Karadeniz Technical University at Trabzon, Turkey, in 2006. Since 2008, he is pursuing his doctoral studies in the Electronics and Information Systems department, Ghent University, Ghent, Belgium. His research interest includes multiview video coding and 3D imaging systems. He has six publications in international journals and nine publications in international conference proceedings. Journal of the SID 21/9, 2014 405 Pankaj Joshi received the Bachelor of Technology degree from Indian Institute of Technology, New Delhi, in 2006. He earned the Master of Photonics degree jointly from KTH Sweden and Ghent University, Belgium, in 2011. He is currently pursuing his doctoral research, in Prof. Herbert De Smet’s group at Ghent University. His current research includes blue phase liquid crystals and their application in photonic devices. Dieter Cuypers graduated with a degree in electromechanical engineering in 1997 at Ghent University. In 2005, he obtained his doctoral degree at that same university. He has been working at the Electronics and Information Systems department since his graduation, doing research for both the university and Imec. His research is largely situated in the field of visualization technology, specifically microdisplays, and also stretches into other areas like optics and thin-film processing. He is a member of SID and SPIE and author or co-author of over 40 papers. David Schaubroeck was born in 1980 in Belgium. He obtained an MSc degree in chemistry at the University of Ghent (Belgium) in 2004. Until 2006, he worked as a scientific researcher for a spin-off company of Ghent University specialized in the development of olefin metathesis catalysts. Thereafter, he joined the research group of Prof. P. Van Der Voort at Ghent University where he developed ordered mesoporous polymers. Currently, he is a PhD student at the Electronics and Information department (Ghent University, Belgium) under the supervision of Prof. André Van Calster and Prof. Peter Dubruel. His research interests include polymer materials for electronics, biomaterials, adhesion, and polymer surface modification. Herbert De Smet graduated with a degree in physics engineering at Ghent University in Belgium in 1988. In 1994, he obtained his PhD degree from the same university, on the subject of integrated drivers for flat-panel displays. Since 1995, he is working for Imec, where he has been responsible for several national and European scale (FP4-5-6-7) research projects concerning displays, microdisplays, and projection. He is currently heading a research group that focuses on display technology, smart lenses, and multiview 3D projection. Since 2000, he has been a Professor at Ghent University, presently in the rank of Senior Lecturer, teaching courses in the field of microsystems, sensors, and actuators and photonics. He has (co-)authored more than 200 publications, 87 of which are listed in the Web of Science and is the (co)-inventor of three patents. He is a member of SPIE and the director of the Mid-Europe Chapter of SID. 406 De Smet et al. / A liquid crystal contact lens display