Flat panel displays made on plastic substrates are envisioned for use in certain commercial and military systems because they are more rugged and lightweight than displays made on glass substrates. High information content can be attained for such displays using an active matrix array of thin film transistors (TFTs) for the pixels and high current TFTs for the drivers. In this work the fabrication of high performance polysilicon TFTs on flexible plastic substrates is presented along with corresponding electrical characteristics. Plastic substrates pose severe temperature constraints on the fabrication process. To overcome electrical characteristics. Plastic substrates pose sever temperature constraints on the fabrication process. To overcome these constraints, our group at LLNL has used low temperature silicon, oxide, and aluminum thin film deposition steps and pulsed excimer laser processing to perform the TFT channel crystallization and the source/drain doping. Sheet resistance values below 1k(Omega) /$DAL are obtained using our laser doping technique for 900 angstrom thick polysilicon films. Our n-channel polysilicon TFT electrical performance on plastic shows mobilities up to 50 cm²/V-sec and ON current to OFF current ratios of up to 1 X 10⁶ for gate voltages from -1 to +35 V.

THe progress made recently in display technology covers many fields of application. The specification of radiance, colorimetry and lighting efficiency creates some new challenges for designers. Photometric design is limited by the capability of correctly predicting the result of a lighting system, to save on the costs and time taken to build multiple prototypes or bread board benches. The second step of the research carried out by company OPTIS is to propose an optimization method to be applied to the lighting system, developed in the software SPEOS. The main features of the tool requires include the CAD interface, to enable fast and efficient transfer between mechanical and light design software, the source modeling, the light transfer model and an optimization tool. The CAD interface is mainly a prototype of transfer, which is not the subjects here. Photometric simulation is efficiently achieved by using the measured source encoding and a simulation by the Monte Carlo method. Today, the advantages and the limitations of the Monte Carlo method are well known. The noise reduction requires a long calculation time, which increases with the complexity of the display panel. A successful optimization is difficult to achieve, due to the long calculation time required for each optimization pass including a Monte Carlo simulation. The problem was initially defined as an engineering method of study. The experience shows that good understanding and mastering of the phenomenon of light transfer is limited by the complexity of non sequential propagation. The engineer must call for the help of a simulation and optimization tool. The main point needed to be able to perform an efficient optimization is a quick method for simulating light transfer. Much work has been done in this area and some interesting results can be observed. It must be said that the Monte Carlo method wastes time calculating some results and information which are not required for the needs of the simulation. Low efficiency transfer system cost a lot of lost time. More generally, the light transfer simulation can be treated efficiently when the integrated result is composed of elementary sub results that include quick analytical calculated intersections. The first axis of research appear. The quick integration research and the quick calculation of geometric intersections. The first axis of research brings some general solutions also valid for multi-reflection systems. The second axis requires some deep thinking on the intersection calculation. An interesting way is the subdivision of space in VOXELS. This is an adapted method of 3D division of space according to the objects and their location. An experimental software has been developed to provide a validation of the method. The gain is particularly high in complex systems. An important reduction in the calculation time has been achieved.

Organic light emitting diodes are a new flat panel display technology that offers high luminous efficiencies. In this paper, a VGA format polysilicon active matrix organic light emitting diode display will be presented. The display design and pixel structure will be discussed as well as the integration of the polysilicon TFT with the OLED display process. The method used to drive the display will be presented along with the active matrix display performance.

The emergence of wearable electronics is leading away form glass substrates for the display backplane, to plastic and metal. At the same time the substrate thickness is reduced to make displays lighter. These two trends cooperate toward the development of compliant substrates, which are designed to off load mechanical stress from the active circuit onto the substrate. Compliant substrates made the circuit particularly rugged against rolling and bending. Design principles for compliant substrates include: (a) Moving the circuit p;lane as close as possible to the neutral plane of the structure, and (b) Using substrate and encapsulation materials with low stiffness. Design principle (a) is demonstrated on thin-film transistors made on thin steel foil. Such transistors function well after the foils are rolled to small radii of curvature. Principle (b) of compliant substrates is demonstrated with bending experiments of a-Si:H TFTs made on thin substrates of polyimide foil. TFTs on 25-micrometers thick polyimide foil may be bent to radii of curvature as low as 0.5 mm without failing. The reduction in bending radius, from R-2 mm on same- thickness steel foil, agrees with the theoretical prediction that changing from a stiff to a compliant substrate reduces the bending strain in the device plane by a factor of up to 5.

A thin layer of the biological photochromic material bacteriorhodopsin (BR) sandwiched between dielectric layers and absorptive filters mounted onto glass substrates forms a direct-view display. Information is written to this BR- display via a scanning laser beam. Back-illumination of the BR-display with a white light source allows to observe the information in high contrast and with a high resolution which is limited only by the parameters of the laser scanner. Since BR thermally decays back into the initial state a periodic refresh is required every 15-30 secs in order to keep 90 percent of the maximal contrast. This low refresh frequency is achieve distance the genetically modified BR variant D96N is employed in the photoactive layer. Photochemical erasure with intense blue light allows to clear the whole display immediately. Weak homogeneous illumination with blue light is used to maximize the contrast by compensating the bleaching effect of the light from the back-illumination. The BR-displays described here have diameters of 2 inches but much larger displays can be manufactured by the same technology.

Current AMLCD panel pixels are divided into sub-pixels each covered by red, green, or blue absorptive color filters to transmit each of the color components. This method discards 2/3 of available light and causes these displays to be highly inefficient. Using a diffractive color separation filter, DCSF, a much higher percentage of energy from the back-light is used in the display, which can be translated into higher brightness and lower power consumption. Such a DCSF is designed to separate the colors and focus the desired bands onto the apertures of the color pixels. The black matrix is used to block the undesired wavebands. Two basic prototype models were designed and fabricated. The first filter design constraints focusing elements and the second filter contains no focusing elements. This paper present testing results from the two prototype diffractive color separation filter designs.

Miniature displays based on scanning a low power beam directly onto the viewer's retina can offer high spatial and color resolution and very high luminance.For scanning display systems, the resolution is primarily determined by the product of scan-angle and mirror-size. Once (theta) D is determined based on resolution requirements, it then remains to choose D and (theta) . Once choice of D and (theta) has a big impact in scanner design and many factors need to be taken into account. This paper discuses how D and (theta) should be chosen considering the limitations due to dynamic mirror deformation, stress in flexures, scanner frequency, optomechanical design, size, and cost.

Surface plasmon tunable filter is a new technology under development at JPL. I had discussed the theoretical model and possible applications of using surface plasmon tunable filters for flat panel display device in several SPIE meetings. Now a prototype of surface plasmon tunable filter has been fabricated at JPL, and the experiment measurement will be discussed in this paper.

Recent developments in improving the performance of SrS based blue thin film EL phosphors are reported. A model approach to optimizing EL material is presented, which considers activator incorporation, charge compensation, defect reduction and energy transfer processes. We demonstrate how the flexibility and accuracy of physical vapor deposition and modern characterization techniques, used in a systematic and cost effective research methodology, can improve the understanding of these issues and be used to develop more efficient thin film EL phosphors. For example, the growth of specialized materials system coupled with PL and PLE, EPR measurements were successfully used to understand the defect structure due to the valence mismatch from the doping of 3⁺ Ce onto the 2⁺ Sr site. Successful incorporation of 1⁺ Ag into SeS:Ce was shown not only to improve both the blue chromaticity from x equals 0.31, y equals 0.52, to x equals 0.26, y equals 0.46, and the EL brightness, but also to eliminate the thermal quenching of the emission intensity. QDLTS was used to identify the trap states and density to help the development of a low temperature annealing process for SeS:Cu. A reduction in the thermal quenching behavior due to codoping was also observed in this material.

We have evaluated the trap states in SrS:Cu thin films using thermally stimulated luminescence (TSL). The TSL experimental curves were fit to a general order kinetics equation to estimate the mean trap energies, escape frequencies and order parameters. SrS films doped with Cu have been prepared on Si and AlTiO (ALO) substrates by sputtering at an elevated substrate temperature. A BaTa₂O₆ film was then sputtered on a SrS:Cu film grown on the ALO substrate, which represents a typical thin film electroluminescence device structure. The trap states are measured as a function of the three different film structures. The three films all have a main trap state at approximately 0.4 eV, which shows small shifts for the three structures. In addition to the single TSL peak,the SrS:Cu based films have a distribution of more shallow traps. When the phosphor films are capped with BTO an additional trap state is observed at approximately 0.08eV. From the SIMS data, the BTO/SrS:Cu interface is extended after annealing, which may contribute to these additional trap states. There is large body of work on the physical processes in SrS:Cu using electrical and optical methods, which shows that the trap states play an important role in the electroluminescence from these devices.

We have developed a spraying process for coating individual phosphor particles with electrically conducting, non- conducting, and hybrid films. The phosphor is suspended in either aqueous or non-aqueous solvent containing the appropriate coating materials precursors. Depositions via gelation or precipitation of the precursor material on the particles is prevented before spraying. The suspension is sprayed into a hot zone where the coating either gels or precipitates on the particle surface in-flight. Coatings developed in-flight are more uniform than coatings developed in solution. Typical coating thicknesses of 10 to 90 nm have been achieved in a single pass. Multiple coatings are applied using multiple passes. This process is scalable to industrial sizes and is easily adaptable for coating particles for other applications.

This report reviews flat panel technologies and presents our recent breakthrough results on phosphors and phosphor screens for mid-voltage FED displays. Mid-voltage FED displays promise superior performance as compared to other flat panel display technologies, if significant improvements in cathodoluminescent efficiency can be made. The blue and green improved efficiencies of RGB non-sulfur triad and aluminized screens have been obtained and are the highest reported. The plasmon cluster approach was used to select phosphor hosts for preparation of highly efficient phosphors. This allowed us to convert phosphor synthesis and FED front plate fabrication form an art to a predictable chemical technology.

Cathodoluminescence (CL) characterization in a demountable vacuum chamber is an important benchmarking tool for flat- panel display phosphors and screens. The proper way to perform these measurement is to minimize the effects of secondary electrons, excite the phosphor/screen with a uniform beam profile, and maintain a clean vacuum environment. CL measurements are important for preliminary evaluation and lifetesting of phosphor powders and screens prior to incorporation into the FPD. A survey of many CL characterization systems currently in use revealed the myriad of spectroradiometers, colorimeters, electron guns, vacuum pumps, mass spectrometers, etc. that introduce many avenues for error that are often difficult to isolate. A preliminary round-robin experiment was coordinated by Sandia and involved five other research groups. The purpose of this experiment was to obtain an indication of equipment capabilities and instrument variations, as well as reliability and consistency of results. Each group was asked to measure the luminance and chromaticity coordinates of a Y₃Al₂Ga₃O₁₂:Tb pellet and calculate the luminous efficiency. Pellets were chosen in order to reduce errors associated with processing and handling of powders or screens. Some of the data reported in this experiment were in good agreement, while others differed significantly. Determining sources of error in CL measurements is an ongoing effort. By performing this experiment, we were able to identify some of the causes of error and develop a characterization protocol for display phosphors.

We have synthesized oxide and sulfide based nanoparticles doped with Mn for potential use as cathodoluminescent phosphors in field-emission displays. Earlier work ion nanophosphors had raised the question whether it is possible to achieve high efficiency using particles that are less than approximately 1 micrometers in size. Although high photoluminescent efficiency had been achieved in particles as small as 3 nm, high cathodoluminescent efficiency has not been reported in nanoparticles. We show, for the first time, that the cathodoluminescent efficiency of nanoparticles is comparable to that of commercial phosphors especially for electron beam energy ranges between 500 V and 3 kV. For example, the luminescence of ZnS:Mn nanoparticles is within a factor of 2.5 of the commercial product at 650V. And, for the Zn₂SiO₄:Mn nanophosphor we have indeed exceeded the luminance of the commercial product. We stress the importance of a capping layer on the sulfide-based nanophosphors for improved stability and performance.

Candescent's full color, full video ThinCRT is an FED-based display which allows the advantages of high voltage phosphors to be realized. These advantages include good color gamut, availability, long life, and high efficiency. A major constraint associated with the use of HV phosphor includes the requirement for a larger separation between faceplate and backplate in order to hold-off the higher voltage. This in turn a) requires the use of high aspect ratio internal support structures, and b) requires that an electron focusing structure be incorporated. The internal support structure, including methods of controlling the vertical voltage distribution on the walls, is discussed. This paper also gives data on the molybdenum Spindt emitters, including I-V curves showing the low switching voltage obtained in pat by the small diameter gate holes.

A new field emitter array (FEA) electron source has been developed which includes a hollow cylindrical multi-layer thin-film-edge low work function emitter and a low capacitance horizontal extraction gate. The field emitter is made by sandwiching a low work function lithium thin film between two high work function noble metal thin films to create a lithium dispenser field emitter in each FEA cell. Our fabrication method, based on conformal chemical beam deposition, is self-aligned and consists of very few processing steps compared to conical or pyramidal FEAs. The structure is fundamentally tolerant to back ion-bombardment and insensitive to poisoning as determined by cycling the operating emitter from 10^-8 to 10^-6 Torr of room air for several hours with no significant change in emission current. Using an extraction voltage of 62 volts we measured 16 microamps from 1-3 FEA cells. No special activation is required either for initial turn-on or when taking the FEA out to air and back again to vacuum. We believe this new FEA should be scalable and inexpensive for manufacturing low-cost FEDs. The chemistry, fabrication, and emission characteristics will be discussed in this talk.

This paper reports for the first time the electrical conduction and breakdown characteristics of thin-wall ceramic spacers bridging two thin-film electrodes, which represent the FEA cathode and the phosphor anode in a field emission display (FED). Techniques to set up a high aspect- ratio thin-wall spacer were specially developed, without use of glue. An extra-low light detection 3D-imaging system using an intensified CCD camera was developed, without use of glue. An extra-low light detection 3D-imaging system using an intensified CCD camera was developed which was able to identify the location of any light activity in the stressed vacuum-gap, indicative of imminent device failure. The result of this work are highly encouraging in that an approximately 1000 micrometers tall spacer can support approximately 18 kV, at least 80 percent above the expected operational voltage of high voltage FEDs. However, it is likely that the breakdown strength can degrade in the presence of electronic bombardment from the field emitted array in an FED - a subject for future investigations in our laboratory.

For more than a decade the Image Applications department at IBMs Watson Research Center has been involved in cultural and commercial imaging projects that demand high-fidelity color reproduction of precious objects like paintings, illuminated manuscripts or jewelry. Our primary display media have been high-resolution cathode ray tubes (CRT), but for the last three years our customers have been replacing them with liquid crystal displays (LCD). The color calibration model we have been using for the CRT is the one described in the literature. It assumes that the chromas of the primaries are independent of intensity, that the colors produced from them are additive and that the intensity of black is almost zero. We measured several models of LCDs and observed that they poorly satisfied these assumptions at medium to low intensities. This becomes noticeable if the image has dark areas or if the display is viewed under a weak ambient light. In this paper, we use a modified version of the CRT model to calibrate the LCD. First we measure four sets of red, green, blue and gray patches. THen we determine the correction factor needed to make, at each level,the sum of the primaries equal to the corresponding gray. Finally, we use these factors to modify the data of red, green and blue.

A calibration facility has been developed at the National Institute of Standards and Technology (NIST) to address the need for high accuracy color measurements of displays. Calibration services are planned for colorimeters and spectroradiometers, tailored to display measurements. A key component of the facility, a reference spectroradiometer, has been developed and its uncertainty for display measurements estimated using a series of computer simulations. The simulations predict that the reference spectroradiometer - corrected for wavelength error and variable bandpass - can measure any color of a cathode ray tube or liquid crystal display with a combined standard uncertainty of approximately 0.001 in chromaticity (x,y) and 1 percent in luminance (Y). In addition a new matrix correction technique has been developed as a means to transfer the calibration from the reference instrument to a test instrument. Using the four-color method, the residual errors with the calibrated instrument for one type of display are reduced to within 0.001 in x,y with respect to the reference instrument. To evaluate the overall performance of the system, commercial instruments - spectroradiometers and tristimulus colorimeter - were calibrated against the reference instrument, measuring both a CRT and an LCD display. The results show that calibrated target instruments can measure various colors of a particular display with a combined standard uncertainty of approximately 0.002 in x,y and 2 percent in Y.

In order to assure safe, comfortable and thus effective working conditions, LCD-screens and monitors have to fulfill certain minimum performance requirements that have been established that have been established by suitable experiments and fixed in the respective ergonomic standards.

Light-measurement instrumentation based upon high-quality charge-coupled-devices is currently in use for measuring the characteristics of electronic displays. When such array detectors are used to measure scenes having high contrast or wide color variations, they can suffer from the effects of veiling glare or lens flare and thereby inaccurately measure the darker luminances because of a mixing of the scene luminances or colors. The simulated-dye-design camera attempts to reduce the effects of unwanted light contamination by copying some of the characteristics of the eye. This first prototype shows an improvement of a factor of 2.7 in its ability to measure high contrasts over a similar camera that is not filled with liquid.

A new uniformity measurement method and the associated equipment is described. This method avoids the viewing angle dependence of the uniformity measurement results, which is observed with conventional area photometer methods. After describing and analyzing currently used techniques and their merits and drawbacks, the modified method is introduced and evaluated. Results of measurements carried out on a conventional TFT LCDs are analyzed.

Capacitance crosstalk and TFT photoleakage affect the transmission-voltage array characteristics in high- resolution TFTLCDs. These effects depend upon the drive inversion scheme use, and are image-dependent. Photoleakage can also be a cause of flicker in TFTLCDs at low frame rates. One characterization method utilizes comparison of front-of-screen measurements with either measured test cells or theoretical cell characteristics. A second method utilizes digital crosstalk compensation, in which the image data provided to the panel is modified to offset the effects of crosstalk in various test image patterns.

We are investigating the important physical factors for the reproduction of the high-order sensation, such as 'the sense of existence' and 'the naturalness', in a display of images. In our research, a key assessment word, 'image depth', is a useful indicator for the discovery of new physical factors. Focusing on the faithful reproduction of 'image depth', some physical factors of a display have been discovered. In this report, we will describe the newly discovered physical factor, the halation and the degradation of image quality. After observation of the images on two displays which have different types of coating on the surface of CRT, we have hypothesized that the halation around the surface of CRT might cause a serious degradation of image quality such as 'image depth'. Also we have considered that the coating might be one of important factors related to the halation characteristic. To confirm the above assumption, we prepared three types of CRT picture monitor which have different types of coating: 13 layer AR coating, 4 layer AR + non- glare coating and no-coating. The increase of luminance around a test target of white rectangle caused by the halation have been measured and the relationship between the halation characteristic and the reproduction of 'image depth' has been assessed.

The visibility of flicker on a display depends upon many factors, including the observer's sensitivity to flicker. Whenever flicker is observed, it is probably undesirable and often unacceptable. Much has been written about flicker, its perception, and its variability. Methods have been presented to the industry that use frequency domain analysis of measured flicker response. Here we prose an alternate method to quantify flicker in the time domain, just as people see it, and we will try to understand how the measurements relate to what people see. Both the frequency domain and time domain flicker measurements can be found in the Video Electronics Standards Association Flat Panel Display Measurements Standard. An attempt is made to compare the two methods and show how the simpler measurement can be employed for many display technologies.

Electronic front-projection display specifications are often based on measurements made in ideal darkroom conditions. However, not everyone has access to such a facility. In many environments, ambient light from other sources in the room illuminates the screen. This includes room lights directly illuminating the screen and the reflection of these light sources off of walls, floors, furniture, and other objects. Additionally, back-reflections from the projection screen must be considered. These stray light components contribute to the measured value, giving an inaccurate measurement of the projector light output. Thus, these conditions may make the task of adequately comparing and evaluating different projection systems difficult. We can better verify whether the projector is operating according to its specifications or compare its performance with other projectors by compensating for stray light. A simple projection mask constructed from black plastic and a stray-light elimination tube are presented as solutions that can provide an accurate measurement of projector light output in many ambient light conditions.

Impressive advances are being made in the field of display metrology, as illustrated in standards such as the Video Electronics Standards Association's Flat Panel Display Measurement standard. However, issues remain regarding the technology-dependent response of display to large-scale signal eros such as noise-induced errors on the data and control lines. The NIST Display Interface Testbed can be used in conjunction with a graphics controller, test generator, or other standard signal source, to inject a user-specified controlled sequence of errors into the interface of a display under test, permitting observation and measurement of the response of the display. In addition to its function as an interface tester, the Testbed effectively serves as a non-intrusive probe into the operation of the display. the programmed control structure of the Testbed permits the convenience of automated test sequences. Development of a feedback mechanisms which allows the test stimulus to be modified in real time, as a function of measurement result and user commands is planned for 1999.