Thick silicon films with good electronic quality have been prepared by glow discharge of He-diluted SiH4 at a substrate temperature ∼150°C and subsequent annealing at 160°C for about 100 hours. The stress in the films obtained this way decreased to ∼100 MPa compared to the 350 MPa in conventional a-Si:H. The post-annealing helped to reduce the ionized dangling bond density from 2.5 × 1015 cm−3 to 7 × 1014 cm−3 without changing the internal stress. IR spectroscopy and hydrogen effusion measurements implied the existence of microvoids and tiny crystallites in the material showing satisfactory electronic properties. P-I-N diodes for radiation detection applications have been realized out of the new material.

A novel family of three-terminal three-color (blue, green, red) detectors based on stacked a-SiC:H/a-Si:H heterostructures is presented: TCO/PIN/TCO/PINIP/TCO/metal and TCO/PINIP/TCO/PIN/TCO/metal structure. The analysis of stacked photodetectors and the optimization of their geometrical dimensions is performed using the adopted ASPIN simulation program. Both structures are mutually compared with regard to calculated current-voltage characteristics and spectral responsivity. They both exhibit linear photocurrent/generation-rate relationship for all three colors at peak wavelengths 430, 530, 630 nm, applying ±1 V or more. This linearity allows that all three colors can be detected with high rejection ratio using simple system electronics.

We present the Adjustable Threshold Color Detectors (ATCDs) based on stacked heterojunctions of amorphous silicon (a-Si:H) and silicon carbide (a-SiC:H). These devices work as bias-controlled photocurrent sources with different values of the voltage threshold depending on the wavelength of the incident radiation. The ATCDs detect the three fundamental chromatic components (blue, green, red) by varying the external voltage within a few Volts.

The aim of this work is to provide the basis for the interpretation of the steady state lateral photoeffect observed in p-i-n a-Si:H ID Thin Film Position Sensitive Detectors (ID TFPSD). The experimental data recorded in ID TFPSD devices with different performances are compared with the predicted curves and the obtained correlation's discussed.

A Linear array Thin Film Position Sensitive Detector (LTFPSD) based on hydrogenated amorphous silicon (a-Si:H) is proposed for the first time, taking advantage of the optical properties presented by a-Si:H devices we have developed a LTFPSD with 128 integrated elements able to be used in 3D inspections/measurements. Each element consists on an one-dimensional TFPSD, based on a p.i.n diode produced in a conventional PECVD system, where the doped layers are coated with thin resistive layers to establish the required device equipotentials. By proper incorporation of the LTFPSD into an optical inspection camera it will be possible to acquire information about an object/surface, through the optical cross-section method1. The main advantages of this system, when compared with the conventional CCDs, are the low complexity of hardware and software used and that the information can be continuously processed (analogue detection).

The Schottky barrier device with a metal/a-Si (n+) /a-Si alloy/metal structure has been widely used as an alternative evaluation tool for the photovoltaic performance of a-Si alloy material since it more reliably reflects the carrier transport in a solar cell than the conventional material characterization tool such as PDS, CPM, and SSPG, and is easier to be fabricated compared with a complete nip solar cell. However, a multiple chamber device making system is still needed to fabricate such a device since one does not want to deposit the a-Si intrinsic material to be studied together with an n+ layer in the same chamber. We have explored the use of a Schottky barrier device deposited on heavily doped n-type crystalline wafer substrate, c-Si (n+) /a-Si alloy/metal, as an evaluation tool for a-Si alloy materials. In this device, besides the evaporation of a thin semi-transparent metal layer, only the active a-Si alloy layer needs to be deposited using the plasma enhanced or other deposition techniques. We have compared the performance of such a device with that of reference n-i-p solar cells deposited at the same time and demonstrated that the FF measured under weak red light show a good correlation between these two types of devices. Therefore the c-Si (n+) /a-Si alloy/metal device can be used as a convenient technique to reliably evaluate the material performance in a solar cell device.

A study of the electroluminescence degradation of a-SiC:H based light emitting devices (LED) is presented for the first time. The best initial peak brightness obtained is 4.2 cd/m2. All LEDs reported in this paper emit a red light which, when operated under continuous bias in a not fully darkened room, is visible for several minutes, depending on degradation rate. The time dependence of LED degradation, which is reversible upon annealing, can be explained if self-annealing is taken into account. There is evidence of an improved LED performance for lower temperature operation. Pulsed operation, with respect to dc operation, produces a markedly lower defect production rate, associated to a higher brightness after degradation. The possibility of some optimization of the operation parameters (peak current, duty cycle) is discussed.

Band gap and defect engineered amorphous silicon based nipin photo diodes with bias controlled spectral response have been fabricated successfully. The devices exhibit good linearity over a wide illumination range and linearly independent spectral response curves which are required to generate a standard RGB-signal. In the bias range from -1.5 V to 1.5 V a dynamic range exceeding 90 dB for two color sensors and 80 dB for three color sensors has been observed. The general operation principle of the multispectral photo diode is discussed using a numerical simulation program. The model describes the defect state distribution of dangling bonds according to the defect-pool model and uses coherent wave propagation in the device to calculate the profile of photo generated carriers. Additionally, an analytical model has been developed to be included into standard circuit simulation programs like SPICE (Simulation Program with Integrated Circuit Emphasis). The analytical model uses linear field approximations in both i-layers of the device.

Utilizing the high optical absorption of amorphous silicon in conjunction with advanced signal processing capabilities of modern ASICs offers the possibility of simply realizing smart color sensors with random pixel access and without any losses in active area due to the signal processing circuitry. Thin-film photodiodes have to be incorporated into a layer sequence of ASIC/insulator/metal/a-Si:H based multilayer/TCO for this so-called Thin Film on ASIC (TFA) concept[l]. Amorphous silicon and silicon carbide photodiodes have been studied, and discrimination of the fundamental colors (red, green and blue) has been obtained from 2-terminal n-i-p-i-n devices at small bias voltages within ±2 V. Changing of the bias voltage applied to the two back-to-back p-i-n junctions allows for adjusting the location of predominant photo-carrier collection, and therefore enables color readout according to the energy dependent absorption profile of the incident light. Film thickness and optical bandgap of the individual layers have been optimized to achieve optimum color sensitivity. Switching experiments in the voltage mode give some information about conceivable frame rates these TFA sensors could be operated at. Light-induced degradation does not hamper the performance of the sensors thanks to low light intensity, reverse bias conditions, and small i-layer thickness.

The human visual system perceives much smaller spatial steps in edges between high contrast regions than equivalents fine, periodic features. This characteristic is known as hyperacuity. We have designed, simulated, fabricated and characterized amorphous silicon sensors which provide hyperacuity information. The individual pixels are position sensitive detectors, the outputs of which provide the x and y first moments of the cell illumination pattern as well as the average gray level. In the simplest case the top electrode of a standard p-i-n diode sensor is replaced by four edge strip electrodes. Both quadrilateral cells (having all four lateral electrodes on the same side of the p-i-n diode) and duolateral cells (having x-electrodes on top and y-electrodes on bottom) have been tested. Results of probing the cells with rastered spots show that both types provide usable linearity and sensitivity. The duolateral structure provides greater orthogonality of the x and y information. One μm spatial resolution can be achieved with devices compatible with standard amorphous silicon sensor processing.

In the past we have developed a transient technique, called the Flying Spot Technique (FST). FST allows, not only to infer the ambipolar diffusion length but also the effective lifetime of the photogenerated carriers once the light spot velocity and geometry of the structure were known.

In this paper, we propose to apply this technique backwards in order to detect the path and velocity of an object that is moving in the direction of a light source. The light reflected back from the object is analyzed through a p.i.n structure being the transient transverse photovoltage dependent on the movement of the object (position and velocity). Assuming that the transport properties of the material and the geometry of the device are known and using a triangulation method we show that it is possible to map the movement of the object. Details concerning material characterization, simulation and device geometry are presented.

Active addressing is focused on TFTs and MIMs as pixel-switches. The manufacture of a-Si-TFTs tends to settle down at a three-mask process as long as risky steps can be avoided. Fabrication throughput is being enhanced by increasing the deposition rate for a-Si- and poly-Si-layers processed by PECV.D. For wider process windows and for building row drivers with a-Si-TFTs effective mobility will have to be increased beyond 1 cm2/Vs especially by new TFT-electrodes, a protective sandwich around the a-Si-layer and modified process techniques. Poly-Si-TFTs should be fabricated at temperatures below 450 °C and preferably without laser annealing before they become an economic solution. MIMs still require more materials research to find a dielectric with a larger trap density enhancing speed of operation, lower addressing voltages and electrodes for improved long-term stability. Low-temperature processes are required for LCDs with plastic substrates.

In this paper, feasibility of crystalline silicon (c-Si) substrate for transmissive active matrix liquid crystal displays (AM-LCDs) has been investigated. The transparent pixel areas of AM-LCD were formed by vertical etching of (110) silicon substrate using anisotropie etching property of aqueous KOH. Combining this vertical etching process with the conventional MOSFET fabrication process, the pixel switching devices, peripheral circuits and transparent apertures were successfully integrated on the same c-Si substrate. The pixel NMOS devices exhibit an electron mobility of about 600cm2/V-s, a subthreshold slope of 65mV/decade and ON/OFF current ratio of 9 decades at 5V drain voltage. And the gate delay time is 3.3ns at 10V power voltage, measured from a ring oscillator which has enhancement-load type NMOS inverters having the (W/L) iload- (W/L) driverer of 25μm/15μm - 50μm/10μm.

Ferrorlrtric liquid crystals (FLCs) have shown great potential for use in electro-optic and display technology due to theri inherently fast switching speeds. Recently within this area a great deal of attention has also been given to FLCs dispersed within a polymer networks. Adding the polymer may act to enhance certain electro-optic properties and will substantially increase the mechanical strenth on the FLC system. This study examines the effects of adding either a diacrylate monomer or a polymer network to a FLC mixture of known composition. The monomer depresses the phase transition temperatures to more orderd phases for both first and second order transitions and causes a marked decrease in the amount of liquid crystal which exhibits typical transitions behavior. During polymerization the network phase separates forming two cocontinuous phases and allows the liquid crystal transitions to return close to values seen in polymer systems. As a result of this decrease, the rotational viscosity decrease for these same samples. Maximum double bond conversions and polymerization rate maxima increase with monomer concentration until saturation on monomer in the liquid crystal is reached. The rate maxima then decreases as the monomer as the monomer must dissolute into the liquid crystal and diffuse to the reactive sites.

A major force for change in substrate requirements in the late 90's may well be the commercialization of poly-silicon thin film transistor (TFT) Active Matrix Liquid Crystal Display (AMLCDs) technology. The processes necessary for “poly-Si” occur at temperatures that are 150–300°C higher than the current amorphous-Si LCD processes. This impacts the thermal shrinkage and thermal gravimetric warp requirements of the glass, particularly as display resolutions tighten, as enabled by poly-Si. In addition, the expected integration of more components (e.g. chip-on-glass) impacts the requirements for the thermal expansion of the substrate.

One approach for meeting the poly-Si demands for greater thermal-dimensional stability is to use glasses with higher temperature capability. A new glass, Code 1737, with the highest strain point commercially available at over 660°C, now enables poly-Si processing with acceptable sag and shrinkage after annealing. A logical goal for the next significant glass advancement would be to eliminate annealing altogether, but it is unclear what temperature capability is required. In this study, various glasses with strain points ranging from 600–800°C have been evaluated in terms of their density, thermal expansion, and thermal shrinkage following poly-Si thermal process simulations. It has been confirmed that the magnitude of shrinkage decreases with increasing strain point for glasses in this compositional family. In addition, future new insight into the effect of thermal expansion coefficient has been developed; the lower the thermal expansion coefficient (for a given strain point), the lower the magnitude of the shrinkage for a given strain point and high temperature thermal cycle. This is important new learning in the area of substrates for flat panel displays that will help in further design and development of glasses for future AMLCDs.

This paper describes the excimer laser annealed (ELA) poly-Si TFT technologies in terms of excimer laser annealing of Si films, the leakage current, and the TFT stability. A laser energy density and a shot dependencies of TFT characteristics was analyzed by TEM, SEM, and Raman. The mobility increases with increasing not only the energy density but also the shot density. The mobility increase with the energy density is due to the grain size enlargement. On the other hand, the mobility increase up to 10 to 20 shots is due to a decrease of defects, including small grains, grain boundaries and defects inside grains. The contribution of grain-growth is small. The ELA TFT has a micro-offset structure to reduce the leakage current. Moreover, we have proposed a dynamic leakage current reduction structure. The combination of these technologies provides a sufficiently small leakage current for AMLCDs. The stability of the gate insulator was analyzed. The TFT shows negative threshold voltage shift under gate bias stress. This is due to water penetration and the subsequent field activated chemical reaction in the gate insulator. The dissociation of Si-OH bonds with hydrogen-bonded water was a fundamental contributor. The shift was suppressed sufficiently by hydrogen passivation. Obtained ELA TFTs;s have mobilities of over 100 cm2/Vsec, threshold voltages of less than 3 V, effective leakage currents of less than 10−13 A, and are stable more than 10 years.