A preview of some of the most interesting papers

A preview of some of the most interesting papers appearing in the
February 2010 issue of the Journal of the SID.
To obtain access to these articles on-line, please go towww.sid.org
Edited by Jay Morreale
Investigation of oil-motion non-uniformity in a reflective display based on electrowetting
Li Chen (SID Student Member)
Yan Tu
Lili Wang (SID Student Member)
Qiaofen Li
Xin Yang
Jun Xia (SID Member)
B. Johan Feenstra (SID Member)
Andrea Giraldo (SID Member)
Nadin Rösler
Southeast University
Liquavista
Abstract - A novel reflective display based on electrowetting technology has gained much attention because of its readability in sunlight. This technology can provide high-efficiency control of the display pixel reflectivity. In the on-state, the oil droplet is expected to move to the same corner regularly. To realize uniform motion of the oil, it is possible to provide a symmetry-breaking mechanism in the pixel. However, depending on the uniformity of the processing and the size of the built-in symmetry breaking, it is possible that the oil actually moves to a different position. The visibility of defects by the representation of oil-motion non-uniformity, depending on different notch pattern, notch size, and defect type has been investigated in the perception experiments. Results indicate that the influence of the defects is more obvious for the more preferred pattern. The larger the notch sizes, the easier the defects can be observed. The best notch pattern is pointed out. When designing an electrowetting display, these results need to be considered to eliminate the influence of oil-motion non-uniformity on the image quality by making the droplet as small as possible or by adding optical layers such as diffusers films.
The basic principle of the reflective display based on electrowetting technology is demonstrated in Fig. 1. The pixel unit in an electrowetting display (EWD) consists of a colored oil droplet, water, a hydrophobic insulator, an electrode, and a white diffuse reflective substrate. Pixels are partitioned with a barrier. In the off-state, no voltage is applied and the entire pixel is covered with oil. The light reflected from the white substrate covered by the colored oil can be perceived as shown in Fig. 1(a). When a DC voltage is applied across the hydrophobic insulator, the oil shrinks because of the modified surface tension. The underlying substrate with high reflectivity is exposed. A mixture of light reflected through the colored oil and the substrate can be observed as illustrated in Fig. 1(b).

FIGURE 1 - Schematic diagram of the pixel unit in an electrowetting display. (a) In the off-state, no voltage is added and oil covers the entire pixel. (b) In the on-state, oil moves to the corner with an additional DC voltage. The photographs in the two insets demonstrate the corresponding situation in practical EWDs.
Active-matrix organic light-emitting diode using inverse-staggered poly-Si TFTs with a center-offset gated structure
Dong Han Kang (SID Student Member)
Mi Kyung Park
Jin Jang (SID Fellow)
Young Jin Chang
Jae Hwan Oh
Jae Beom Choi
Chi Woo Kim
Kyung Hee University
Samsung Mobile Display
Abstract - A low-cost active-matrix backplane using non-laser polycrystalline silicon (poly-Si) having inverse-staggered thin-film transistors (TFTs) with amorphous-silicon (a-Si) n+ contacts has been developed. The TFTs have a center-offset gated structure to reduce the leakage current without scarifying the ON-currents. The leakage current of the center-offset TFTs at Vg = -10 V is two orders of magnitude lower than those of the non-offset TFTs. The center-offset length of the TFTs was 3 mm for both the switching and driving TFTs. A 2.2-in. QQVGA (160 x 120) active-matrix organic light-emitting-diode (AMOLED) display was demonstrated using conventional 2T + 1C pixel circuits.
Figure 1(a) is a schematic cross-sectional view of an inverse-staggered poly-Si TFT with n+ a-Si:H source/drain contacts and a center-offset gated structure and Fig. 1(b) is a SEM image. The inverse-staggered TFT using a poly-Si by metal-induced crystallization of a-Si through a cap layer (MICC) was reported. The MICC can reduce metal contamination in poly-Si and provide a clean and smooth surface. It is found that the Ni could be diffused through the plasma-enhanced chemical-vapor-deposition (PECVD) SiNx layer into the a-Si and the Ni atoms in the a-Si form NiSi2 nuclei. Then, disk-shaped grain growth takes place from the NiSi2 nuclei until the neighboring grains collide. A conventional dual-gate LTPS TFT has a highly doped region between the two gates in an active channel. However, the center-offset gated TFT has an undoped poly-Si region between the dual-gate electrodes. The inverse staggered poly-Si TFT with a center-offset gated structure can be fabricated with a relatively simple process and has a symmetric planar TFT structure, which does not require any additional photomask steps and is immune to the misalignment of gate patterns caused by lithography. This means that both channel and offset lengths of a center-offset gated TFT do not change because of a small misalignment. But the offset length can be changed in the drain-offset TFT by a small amount of misalignment.

FIGURE 1-(a) A schematic cross-sectional view of an inverse-staggered poly-Si TFT with a center-offset gated structure and (b) its SEM image.
Ultra-fast-switching flexoelectric liquid-crystal display with high contrast
Flynn Castles (SID Student Member)
Stephen M. Morris
Damian J. Gardiner
Qasim M. Malik
Harry J. Coles
University of Cambridge
Abstract - The flexoelectro-optic effect provides a fast-switching mechanism (0.01-0.1 msec), suitable for use in field-sequential-color full-motion-video displays. An in-plane electric field is applied to a short-pitch chiral nematic liquid crystal aligned in the uniform standing helix (or Grandjean) texture. The switching mechanism is experimentally demonstrated in a single-pixel test cell, and the display performance is investigated as a function of device parameters. A contrast ratio of 2000:1 is predicted.
The structure of the device is shown in Fig. 1(a). The N* is aligned in the USH configuration using anti-planar alignment on the glass surfaces. With no field applied, the helical structure is undistorted, and the device is non-transmissive between crossed polarizers. When an in-plane electric field is applied, the optic axis of the N* rotates, and a birefringence is induced. For a sufficient tilt angle, the device becomes transmissive (the required angle depends on the birefringence of the LC). The short-pitch N* may be considered as a uniaxially birefringent structure with the optic axis along the helical axis. The viewing characteristics of the device are analogous in many respects to that of the vertically aligned nematic (VA) mode LCD [Fig. 1(b)].

FIGURE 1 - (a) The chiral-nematic liquid crystal is aligned in the uniform standing helix configuration. An in-plane electric field is applied in the same direction as the surface alignment, which rotates the optic axis via the flexoelectro-optic effect. Crossed polarizers are positioned at ±45° to the direction of the electric field. (b) The short-pitch chiral nematic is effectively a uniaxially birefringent structure, with the optic axis along the helical axis. The viewing characteristics of the device are analogous to a VA-mode LCD.
Mechanism of image-sticking after long-term AC field driving of IPS mode
Yuichi Momoi (SID Member)
Kazuhiko Tamai
Kaoru Furuta
Tae-Rim Lee
Kyeong-Jin Kim
Chang-Ho Oh
Tomonori Koda
LG Display Co., Ltd.
Yamagata University
Abstract - An AC electric field to drive the in-plane-switching (IPS) mode of a liquid-crystal display (LCD) causes a reduction in the contrast after a long period of display operation. This phenomenon is refered to as the AC image-sticking problem caused by long-term driving. Thus far, there is no useful method of quantitatively evaluating AC image sticking. LCD panel products that use the IPS mode have been evaluated for a decade. In this paper, a new evaluation parameter (??), which was recently proposed by Suzuki et al., is introduced. It was calculated from the slight difference in the deviation angle of LC molecules from the rubbing direction. Results from several conditions of test samples are presented in this paper as a phenomena that reflect the interaction between the surface of the PI alignment and the LC molecules. The results and discussions describe reasons for azimuthal gliding after long display operation for weak AC voltage driving. It is explained by suitably adopting the Kelvin-Voigt model which is used to discuss the rheology of viscoelastic material. It is concluded that the surface rheology of PI alignment is one of the most important factors for the contrast reduction of the AC image-sticking problem.
Test cells of a simple structure that had bended-comb-teeth ITO electrodes is illustrated in Fig. 1. The electrode area was 1 x 1 cm. The spacing between electrodes was 10 mm. The angle of the electrodes and LC alignment direction was 20°. We set one glass substrate having electrodes on another plain non-conductive substrate using a column spacer with a 3-mm height. Figure 1 also describes the measurement scheme used in this work. The left part of Fig. 1 shows the initial state after preparing the samples. The rubbing direction is from the bottom to the top. The LC molecules shown in the figure align along the rubbing direction. The middle part of Fig. 1 shows the state under AC driving. For these bended-comb-teeth ITO electrodes, the twist direction of the LC molecule along the electric field on the upper electrode area is always opposite the direction of the lower electrode area. The right part of Fig. 1 shows the slightly deviated LC molecules from the initial state after a long period of AC driving. It is easy to measure the exact deviation of the LC molecule by adding the difference in the deviations of the upper and lower electrode areas because the direction of the deviation is opposite to each other.

FIGURE 1 - Measurement scheme.
Super-high-resolution transfer printing for full-color OLED display patterning
Hongzheng Jin (SID Student Member)
James C. Sturm
Princeton University
Abstract - A transfer-printing method for the patterning of thin polymer layers is described. A hard stamp with a raised feature is brought into contact with a spin-coated organic film under elevated pressure and temperature to break the films. The patterned film is then transfer printed onto the devices. This method is used to print RGB subpixel arrays with a pattern size as small as 12 mm at a resolution of 530 ppi to demonstrate its ability for full-color organic light-emitting-display fabrication. Devices with printed organic layers have similar performance to spin-coated controls under optimized printing temperature and pressure settings. The critical physical parameters include a soft intermediate plate for the sharp breaking of edge patterns, control of surface energies, and printing at moderate temperature and pressure to achieve intimate contact between the printed layer and the underlying substrate.
Figure 1 illustrates the procedures of our transfer-printing method. The process begins with the fabrication of a hard stamp [Fig. 1(a)], where the raised features correspond to the final regions without the polymer on the final device plate. A silicon wafer is used for the hard stamp and conventional lithography and etching is used. In this work, rectangularly shaped patterns of two dimensions (12 x 40 mm, 24 x 80 mm) are etched into the stamp to a depth of ~700 nm. A second flat plate, also silicon for convenience, is coated with a poly(dimethylsiloxane) (PDMS) layer and used as a backing layer for the active polymer layer to be patterned. The PDMS is spin-coated from a PDMS prepolymer which is a mixture of Sylgard 184 and its hardener (Dow Corning) dissolved into cyclohexane solvent. The PDMS prepolymer solution is syringe-filtered by using a 0.1-mm filter before use. The PDMS film after spin-coating is baked at 150°C for 2 hours on a hotplate for cross-linking. The final PDMS film is about 1.2 mm in thickness after baking.

FIGURE 1 - Schematic illustration of the transfer-printing method for full-color display patterning. (a) A hard stamp with raised features and a polymer-coated (e.g., red-light-emitting) PDMS/Si substrate. (b) The stamp is pressed onto the polymer layer at elevated temperature (p and T). Upon separation, two patterned polymer layers are formed on each side. (c) The patterned polymer on the PDMS side is used as the printing plate and (d) is pressed onto a device plate (p and T) to transfer the polymer layer. Steps (a-d) are repeated for (e) green- and (f) blue-light-emitting polymer to finish the pattern (g) required for a full-color display.
LED backlight driving system with local dimming and scanning driver IC
Ki-Chan Lee (SID Member)
Kyoung-Uk Choi (SID Member)
Sang-Gil Lee
Yun-Jae Park
Hyun-Seok Ko
Seung-Hwan Moon
Brian H. Berkeley (SID Senior Member)
Intersil Corp.
Samsung Electronics Co., Ltd.
Samsung Mobile Display
Abstract - Novel LED backlight driving technology with a fully embedded LED driver IC is presented. This driving system and IC feature a high speed reduced swing differential signaling (RSDS) interface, an independent PWM controller with high-resolution programming, and embedded scan logic. To attain high efficiency from the boost converter, highly advanced dynamic headroom control technology has been implemented onto the IC. The embedded current boosting function showed almost no brightness reduction when operated in scanning mode. The scanning function of the system improved motion blur values 26-44% compared to that of a non-scanning LCD panel.
It is well known that hold-type driving is one of the main causes of motion blur in LCD TVs. Efforts to reduce motion blur have been focused on reducing the hold time. Fast-responding liquid crystal alone can not resolve this blurring problem. In our previous work, we developed unique motion blur reduction technique, motion compensated frame interpolation (McFi), for the best motion picture quality using the simplest system. By doubling the number of frame updates, motion picture response time (MPRT) values have been reduced from 15 to 8 msec. To achieve even better MPRT and dynamic contrast ratio (CR), a new local dimming and scanning LED backlight system has been developed as shown in Fig. 1. In the timing controller of this system, localized gray data is extracted after the frame interpolation (rate doubling) process. This local dimming data is then transferred to an LED controller (L-CON).

FIGURE 1 - 120 Hz LCD panel driving system with timing controller providing motion picture interpolation and localized gray data extraction for LED backlights.
A reduced-voltage differential signaling (RVDS) interface for chip-on-glass TFT-LCD applications
Jung-Pil Lim (SID Member)
Donghoon Baek
Jae-Youl Lee
Yoon-Kyung Choi
Myunghee Lee
Samsung Electronics Co., Ltd.
Abstract - Reduced-voltage differential signaling (RVDS) is a novel interface for TFT-LCD panels with a chip-on-glass (COG) structure, which has a point-to-point topology and a voltage mode differential signaling scheme. The voltage-driving interface scheme has advantages in high-speed operation owing to its relatively small time constant for the resistive channel condition. And reduced-voltage signaling can reduce the power consumption of a transmitter. The display source driver IC with an RVDS interface, which is fabricated by using a 0.25-mm CMOS process with a 2.5-V logic supply voltage, offers a high data rate up to 500 Mbps, low-current consumption of 2.2 mA, and good EMI characteristics. Also, an RVDS interface has programmable options that control the bandwidth, system power, and EMI performance. Therefore, the RVDS interface is a competitive solution for low-power, low-cost, and slim notebook applications.
TFT-LCDs are widely used in notebooks, desktop monitors, and TV applications because it offers many advantages, such as slim profile, light design, high brightness, low power consumption, etc. Despite many of the advantages as described above, the TFT-LCD module market is becoming saturated and thus the price keeps coming down. Hence, display makers introduced chip-on-glass (COG) technology in order to reduce manufacturing cost of TFT-LCDs. Also, the demand for a COG-type source driver (SD) is increasing with the trend of using an LED backlight unit to reduce module thickness. Because a PCB, which holds supporting components, can be placed in line with an LCD panel and does not have to be folded to the back side of the LCD panel. And the film and packaging cost of chip-on-film (COF) or tape-carrier-package (TCP) can be saved. Additionally, the slim bezel design of notebooks and monitors is possible by minimizing the size of the printed circuit board (PCB) for which the timing controller (TCON) and passive elements are placed on it as shown in Fig. 1(b).

FIGURE 1 - Block diagram of TFT-LCD panel using (a) COF and (b) COG.
Parameters influencing exo-electron emission currents from MgO film of ACPDPs
Sang-Hoon Yoon
Heesun Yang
Yong-Seog Kim (SID Member)
Hongik University
Abstract - A theoretical model of exo-electron emission kinetics was developed by considering back-diffusion and the gas-amplification phenomena. By using the model, the effects of temperature, trapped electron concentration, trap energy level, and trap concentration on the exo-electron currents were predicted and compared with experimental results. The theoretically predicted values agreed reasonably well with the trends of the measured results.
The exo-electron emission phenomenon has been explained by primarily using a thermally stimulated exo-electron emission model schematically illustrated in Fig. 1. In this model, a fraction of the trapped electrons at shallow trap levels located beneath the conduction band edge of MgO are thermally excited to the conduction band. The trap levels in MgO film have been estimated or measured to be in the range of 0.1-0.5 eV. Therefore, the direct excitation of such trapped electrons above vacuum level by the thermal excitation process has been all but ruled out in this study because they have to overcome an energy barrier that consists of the trap energy plus the electron affinity of MgO (~0.8 eV). Once the free electrons are created by the thermal excitation process, part of them would become trapped back to the traps and the remaining part would react with the holes at recombination centers. Upon the recombination reaction, the reaction energy will be released. This energy could either be dissipated as a photon radiation process or be transferred to an electron at the trap levels to excite above the vacuum level. The later reaction is termed the Auger excitation mechanism of exo-electron emission. It should be noted that there are several other proposed exo-electron emission mechanisms that include direct thermal excitation, field emission, tunneling, etc. In this study, we attempted to modify the Auger excitation mechanism of the exo-electron emission model in order to incorporate several features that are characteristic
of the exo-electrons emission of ACPDPs.

FIGURE 1 - Auger excitation mechanism of thermally stimulated exo-electron emission.
A high-contrast front-projection display system optimizing the projected light-angle range
Baku Katagiri (SID Member)
Masanobu Ooike
Tatsuo Uchida (SID Fellow)
Tohoku University
Abstract - The problem with front-projection displays is that the screen contrast ratio decreases under bright-ambient conditions. To overcome this problem, the design of a special screen, composed of the diffuser whose diffusing property shows top-hat characteristics and a sawtooth reflector, is proposed. The screen diffuses the incident image light arriving at a projection-angle range that is a lower-angle range than the viewing-angle range, and reflects the ambient light out of the viewing-angle range. In this paper, the projection-angle range and the viewing-angle range was optimized to improve the contrast ratio of a front-projection display. As a result, a special screen with the above-mentioned diffusing property was realized, and a high-quality front-projection display with a high contrast ratio, even in a bright room, was achieved.
Several methods have been proposed to improve the contrast ratio of front-projection displays by suppressing the reflection of the ambient light by using different polarization methods, spectra, or incident directions between the projector and ambient light. As for the former two methods, screens which reflect only one type of polarized light or narrow red, green, and blue spectra adjusted to the projector can be used. However, an increase in the contrast of these methods is a factor of 2 or so under typical conditions and changes depending on the ambient-light conditions. As for the last method, the use of a different incident direction can be accomplished by using a silver screen. A silver screen has a Gaussian distribution of the diffusing property, and therefore the diffusing direction of the projected light and ambient light can be separated if the diffusing angle is designed to be rather narrow. However, this method is accompanied by the disadvantage of having a rather narrow viewing angle and spatial non-uniformity of luminance in the screen, referred to as the "hot-spot." To solve this problem, we previously proposed a novel front-projection display that uses diffused light control, comprised of a diffusing film and reflector as shown in Fig. 2, which is referred to as DLC-1.

FIGURE 2 - Proposed front-projection-display system decreases the influence of the ambient light from light sources on the ceiling.
Display motion blur: Comparison of measurement methods
Andrew B. Watson (SID Senior Member)
NASA Ames Research Center
Abstract - Motion blur is a significant display property for which accurate, valid, and robust measurement methods are needed. Recent motion-blur measurements of a set of eight displays by a set of six measurement devices provided an opportunity to evaluate techniques of measurement and analysis. Both the raw data waveforms collected by each device and the metrics derived from those waveforms were examined. Significant discrepancies between instruments and variability within instruments were found. A new motion-blur metric (GET) that exhibits increased robustness and reduced variability relative to existing metrics is proposed.
There have been a number of attempts to characterize motion blur in a systematic and meaningful way. Most of these involve estimating the width of an edge subjected to motion blur. This edge can be captured in any of three ways. A total of 36 data sets were obtained, each consisting of a particular LMD applied to a particular DUT. Altogether, these comprised 1360METP and 1281 BET estimates (some teams submitted waveforms but not BET data for certain conditions). An example of one METP is shown in Fig. 1. This particular METP exhibits considerable noise. This is largely because this is a small transition between nearby gray levels {139, 150}. Note that the curve is expressed (like all METPs) as a function of time in frames.

FIGURE 1 - Example of METP. Details: LMDID = 616000, DUTID = 001, gray levels = {139,150}, speed = 16 pixels/frame.