Microelectromechanical systems must include large-deflection actuators to create devices which can interact mechanically with their surroundings. Electrostatic actuators with large nonresonant motion require high voltages and large surface areas to produce useful forces. Large-deflection thermal actuators have been fabricated in electroplated metal processes, but the choice of materials is limited, and the high conductivity of metal means these devices need a high current to generate the required heat. Thus, there is a need for a low voltage actuator with a large nonresonant deflection which can be fabricated in integrated circuit compatible surface-micromachining processes. The device described in this paper has a simple, flexible, and reliable design which overcomes the limitations of other actuators. It is a low voltage, medium current thermal actuator which can be designed to move laterally in a controllable, nonresonant motion. Typical actuators are 200 micrometers long, 18 micrometers wide, and deflect 10 micrometers at the tip with a drive current of 5 mA at 5 volts. A wide range of layout geometries allows this type of actuator to be fabricated in any surface-micromachining process that includes a releasable, current-carrying layer. An empirical model is presented to describe the devices fabricate in a commercially available surface-micromachining process. Arrays of actuators were fabricated to test the effect of different device dimensions on maximum achievable deflection as a funciton of applied current. Actuators can be combined to produce more force and have been integrated with other micromechanical structures. Applications of these actuators include linear and rotating motors, compliant micromechanisms, latches, micro-relays, variable capacitators and tweezers.

Microelectromechanical actuators with large vertical deflections are desirable for microactuation and sensing applications, such as tactile stimulators for telerobotics, micropositioners, and microflaps for airflow control. Thermally actuated cantilever beams provide large deflections at low voltages and low power. Cantilever beams using silicon micromachining have been fabricated in a complementary metal-oxide-semiconductor process. The beams were tested using a microscope-based laser interferometer and a weight-balance force tester, both of which were specifically designed and built for analysis of microelectromechanical devices. Measurments were taken of cantilever beam deflection, maximum operating frequency, and operation under load. Cantilevers 300 micrometers long and 100 micrometers wide with a 2.6 K(Omega) polycrystalline silicon heating resistor were capable of deflecting loads of up to 15 mg with a 6 volt drive. The same cantilevers were tested for fatigue endurance under a 5 mg load, producing a 5 micrometers deflection that remained unchanged after 75 hours of continuous operation with a 6 volt peak, 0.33 Hz square wave drive signal.

A design is presented and tested to build and actuate reflection based 3D micro-opto- mechancial systems in standard CMOS using only a single maskless post-processing step. We have built two generations of structures and actuated them at various frequencies. The initial design uses the various components of CMOS to make a rigid box being pulled against the substrate. Results and photos are presented to show the actuation. Possible developments which take advantage of the CMOS underlying the structure are discussed.

Force measurements from 3D polysilicon mechanisms under various loading conditions are presented. Force was measured using the deflection of polysilicon spring gauges with known dimensions and estimated elastic properties. Force measurements were made to determine the force required to overcome static friction when rotating hinged structures with and without external loading using a sliding mechanism. Both loaded and unloaded mechanisms display a wide statistical variation in friction force. Unloaded mechanisms require on average 1-5 (mu) N of force to move, whereas mechanisms manipulating 3.5 (mu) N loads require on average 1-10 (mu) N of force. These measurements are motivated by an effort to produce articulated microrobots, and progress in this area is also discussed.

IFAs are MEMS actuators which are powered by the electrostatic forces between the plates of many microscopic deformable capacitors arranged in monolithic arrays. IFAs are fabricated using standard techniques of VLSI electronics. The IFAs reported here resemble thin, flexible plastic membranes 10 mm long and either 1 or 3 mm wide, which contain from 75,000 to 200,000 cells. They are low-weight, high-efficiency actuators with low power consumption, silent operation, and absence of sliding friction. Testing methods applicable to these free- standing MEMS structures are discussed, along with experimental observations and measurements of forces and displacements. We have measured forces in excess of 6 dynes and displacements of over 700 micrometers . The force/cross-sectional area of this MEMS structure is 2800 dynes/mm², and the work done by the IFA divided by its volume is in excess of 7 ergs/mm³. A rate over 20,000 contractions/second has been observed, as well as lifetimes of greater than 10⁸ contractions. The metalization of the IFA strongly affects the performance. Experimental results are presented demonstrating the improvements in the performance with a Cr/Au metal system compared to the Cr/Pd metal system originally used.

This paper describes the development of electrostatically-actuated nickel microrelays. Device design is based on lateral comb-finger drives for actuation of the relay contact. The contact geometry is intended to reduce the 'ON' resistance and also prevent spontaneous closing of the relay. Planar microrelays, up to 15 micrometers thick, are batch fabricated using nickel surface micromachining which uses electroless plated nickel as the structural material, high-aspect- ratio positive resist lithography for the plating mold, and polysilicon as the sacrificial layer. The microfabricated relays were successfully actuated to investigate electrical contact attributes. Initial experiments reveal contact resistances to be around 15(Omega) and current densities up to 2mA/micrometers ² without degradation.

Presented here are preliminary results on a resonant gyroscope that is very simple to fabricate. It is made using both silicon and glass substrates, and involves the minimum of mask steps. Actuation and detection are all achieved via capacitive means. The minimum rotation rate that has been able to be sensed is 300 degrees/s; improvements in the manufacturing process could reduce this to as low as 0.07 degrees/s before the rate detection would become limited by electrical noise.

This paper describes the concept, the design and measurement results of a surface micromachined accelerometer. For the accelerometer presented here a polysilicon high rate deposition process was used to fabricate polysilicon layers with a thickness of 10 micrometers . The sensor is designed as an interdigital finger structure forming a differential capacitor, where the moveable fingers are mounted on a moveable mass and the fixed electrodes are anchored on the substrate. The air gap between the fingers is 1 micrometers . A force-feedback operation mode is realized to increase the sensor performance. A two chip solution with sensing element and signal processing circuit on separate chips was chosen as first approach towards a monolithic integrated accelerometer. The evaluation circuit was realized in BiCMOS technology. Sensors with a closed-loop sensitivity of 8mV/g and a bandwidth of about 10kHz have been fabricated. A resolution of 0.1g and a linearity error of less than 1% could be achieved for a measurement range of +/- 100 g.

Using a single maskless postprocessing step we have developed an accelerometer in a standard commercial CMOS process capable of a sensitive axis parallel or perpendicular to the die surface. Out postprocess is realized using xenon difluoride (XeF₂) as a bulk etchant. The combination of this etchant and the standard CMOS process allows realization of cantilevers with piezoresistive sensors in all spacial coordinates from a widely-accessible source and at a minimal cost. Fabrication of accelerometers for all three axes and associated electronics on a single piece of silicon reduces the cost of 3D acceleration detection while increasing sensor reliability.

Gas sensors have been used and available for many years. While there are many different types of gas sensors and detecting materials, metal oxides are commonly used for the detection of various gases, such as carbon monoxide (CO), methane (CH₄), and other hydrocarbons. Such sensors are used in environmental, commercial HVAC, home, and industrial applications. One of these metal oxides, tin oxide, is very stable in air at temperatures > 500 degrees C, and is readily processed using ceramic or thin-film techniques. The thin films on silicon can be doped with inpurities using standard diffusion or spin-on methods, which influence the sensitivity, selectivity, and stability of the sensors. Thin- film tin oxide on silicon CO gas sensors have been characterized. The structure, chemistry, and response results are reported.

A miniature gas chromatography (GC) system has been designed and fabricated using silicon micromachining and integrated circuit (IC) processing techniques. The silicon micromachined gas chromatography system (SMGCS) is composed of a miniature sample injector that incorporates a 10 (mu) l sample loop; a 0.9-m long, rectangular-shaped capillary column coated with a 0.2-micrometers thick copper phthalocyanine (CuPc) stationary-phase; and a dual- detector scheme based upon a CuPc-coated chemiresistor and a commercially available, 125- micrometers diameter thermal conductivity detector bead. Silicon micromachining was employed to fabricate the interface between the sample injector and the GC column, the column itself, and the dual-detector cavity. A novel IC thin-film processing technique was developed to sublime the CuPc stationary-phase coating on the column walls that were micromachined in the host silicon wafer substrate and Pyrex cover plate, which were then electrostatically bonded together. The SMGCS can separate binary gas mixtures composed of parts-per-million (ppm) concentrations of ammonia and nitrogen dioxide when isothermally operated (55-80 degrees C). With a helium carrier gas and nitrogen diluent, a 10 (mu) l sample volume containing ammonia and nitrogen dioxide injected at 40 psi (2.8 X 10⁵ Pa) can be separated in less than 30 minutes.

We report the technology for the design and fabrication of bulk-micromachined hydrocarbon based gas sensors in standard two micron n-well CMOS process. The hydrocarbon based gas sensor is a palladium-oxide-polysilicon type MOS structure. The gas sensor is realized in three steps: the layout design in CMOS technology using VLSI CAD tools, post-processing on fabricated CMOS devices followed by the deposition of palladium. The design includes additional layer in CMOS called 'open' which enables the formation of a 'cavity' in the silicon substrate and results in a micromachined structure. After the fabrication of CMOS devices a single maskless etch in an aqueous solution of ethylenediamine-pyrocatechol or xenon difluoride is done to create a 'cavity' followed by gas in palladium changes the C-V characteristics of the MOS structure and is detected through an integrated readout CMOS electronics.

A microelectromechanical systems (MEMS) force transducer, with a volume less than one cubic millimeter, is being developed to measure forces generated by living, isolated cardiac muscle cells in order to resolve the complex mechanisms of muscle contraction. The force transducer consists of two movable clamps facing each other. Each clamp contains two vertical, parallel hinged polysilicon plates attached to a moveable shuttle, and the entire structure is suspended 2 micrometers above the substrate via support beams attached to the substrate at one end. Each end of a living rat heart cell is glued between a pair of vertical plates. Calcium is then introduced into the cell's nutrient bath and stimulates the cell to contract. Upon contraction the support beams bend, and the amount of deflection is translated to force via the known spring constant in the beams. Typcially the 70 micrometers long central portion of a 120 micrometers long cell will contract approximately 6-7 micrometers in full activating solution, resulting in forces up to 16 (mu) N. The average value obtained for F_max per cross-sectional area was 21.8mN/mm² which is comparable to the value found in other laboratories using standard transducer technology.

Public awareness of the risks of exposure to electromagnetic radiation has grown over the past ten yeras. The effects of power lines on human and animal health have drawn particular attention. Some longitudinal studies of cancer rates near power lines show a significant correlation, while others show a null result. The studies have suffered from inadequate sensors for the measurement of electromagnetic radiation in vivo. In this work, we describe the design, construction, and testing of electrically passive, microfabricated single-pole antennas and coils. These sensors will be used in vivo to study the effects of electromagnetic radiation on animals. Our testing to date has been limited to in vitro studies of the magnetic field probes. Magnetic field pickup coils were fabricated with up to 100 turns, over a length of up to 1000 micrometers . Measurements were carried out with the sensors in air, and in water of various saline concentrations. Magnetic fields were applied using a Helmholtz coil. Both dc and ac fields were applied. The results indicate that small-area measurements of electromagnetic fields in vitro can be made successfully, provided adequate shielding and amplification are used.

The mixing of chemical components becomes more and more important. Especially in the field of micro analysis systems a mixing procedure of different solutions is often required. We have developed a static micromixer built up in silicon. The design and fabrication procedures of the micromixer allow the integration of other microcomponents like pumps, valves or sensors. The static micromixer consists of two structured silicon wafers. The structured surfaces are bonded together by using a low temperature silicon direct bonding process. The structure of the mixer consists of several identical mixing elements. The size of one element is about 500 micrometers ^*500 micrometers ^* 720 micrometers . The internal volume of one element is about 0.0001 ml. Each element consists of two or more microchannels for different liquids to be mixed. In a first step of the mixing procedure the liquid streams are arranged separately in one horizontal layer. This can be achieved by a connection of the microchannels. In a second step step the whole arate microchannels. These microchannels act as inlet for the next mixing element. We found a homogenous mixing after about 5 mixing elements for mixable liquids. Not mixable liquids will be dispensed after 16-20 mixing elements.

A microelectrohydrodynamic (EHD) injection pump has been developed using laser micromachining technoogy. Two desings have been fabricated, tested, and evaluated. The first design has two silicon pieces with KOH-etched wells which are stacked on the tpo of each other. The wells are etched on one side of the wafer and gold is deposited on the other side to serve as the pump electrodes. A Nd:YAG laser is used to drill an array of circular holes in the well region of both silicon parts. This creates a grid distribution with a square pattern. Next the well regions of the silicon parts are aligned, and the parts are bonded together using a Staystik thermoplastic. The pump unit is then mounted into a ceramic package with a large hole drilled in the bottom of the package to permit fluid flow. Aluminum ribbon wire bonds are used to connect the pump electrodes to the package leads. Isolation of metallization and wires is achieved by filling the package well and coating the wires with polyimide. When a voltage is applied at the electrodes, ions are injected into the working fluid, such as an organic solvent, thus inducing flow. The second design has the silicon parts oriented 'back-to-back' and bonded together with Stayform. A 'back-to-back' design will decrease the grid distance so that a smaller voltage is required for pumping. Preliminary results have demonstrated that this micropump can achieve a pressure head of about 287 Pa with an applied voltage of 120 Volt.

In this paper, smart electronics and MEMS are employed to sense and control the drag in aircraft structures. The sensors are fabricated with interdigital transducers printed on a piezoelectric polymer. They in turn are mounted onto an ultra thin Penn State's novel RF antenna (Patent field). The sensor are designed to measure both pressure and shear of the fluid flow on aerospace structures. The wave form measurements may be monitored at a remote location either at the cockpit or elsewhere via the antennas in the sensors and an outside antenna. The integrated MEMS actuators which comprise of cantilever-, diaphram- and microbridge-based MEMS with suitable smart electronics etched onto the structure are controlled by the built-in antennas through feedback and feedforward control architecture. The integration of such materials and smart electronics into the skin of airfoil is ideal for sensing and controlling drag. The basic idea of this concept involves detection of the point of transition from laminar to turbulent flow and transmitting acoustical energy into the boundary layer so that the low energy fluid particles accelerate in the transverse direction and mix with the high energy flow outside of the boundary layer. 3D microriblets can be fabricated using stereo lithography and UV curable conducting polymers. The control of drag using these active microriblets are outlined.

Here we present results for a new inductive proximity sensor that makes use of two coupled coils, forming a planar transformer, fabricated on a micromachined membrane. Macroscopic proximity sensors based on eddy-current techniques are currently commercially available. However, their design is not conducive to scaling inductor. When such an inductor is scaled down and made planar, the inductance decreases while the resistance of the coil increases dramatically. This dictates that the sensor be operated at much higher frequencies for which the inductance of the coil dominates its resistance. In our new two-coil design a signal is applied across the primary coil and a significant sensitivity to the gap to the 'target' plate is seen in the phase difference between the primary and secondary output. This phase change is largely invariant with change in coil resistance (due to either temperature change or change due to scaling to IC levels), allowing operation at lower frequencies than that required for a single coil proximity sensor. By fabricating the transformer on the back side of a micromachined membrane, the coil is protected from the environment, while the deleterious effect of the silicon substrate on inductance is eliminated.

The subsystems of an integrated MEMS 3D magnetic sensor are described and created using a standard CMOS fabrication process. For the completed sensor, after fabrication, the chip is micromachined and the sensor is assembled. The sensing is done by a two layer, aluminum coil whose equivalent area is specific to being sent off chip. The sensitivity and resolution of the transducer is found and shown to be a function of frequency. The frequency response is shown to be linear. Both higher sensitivity and finer resolution can be achieved by increasing the equivalent area of the micro-coil or by using a higher input frequency. The next phase is to fabricate the complete three axis system with two perpendicular coils rotated off the substrate and one remaining on the substrate.

A general purpose finite element method is developed for the direct numerical solution of small amplitude vibrations of crystal structures which are superposed on a static biasing state. The use of resonant structures for sensing applications is widespread in the development of microelectromechanical systems (MEMS) such as accelerometers, pressure and force sensors, rate of rotation sensors, and resonant strain gauges. Central to the design of such devices is the problem of calculating the change in resonant frequency of a vibrating structure under an imposed quasi-static mechanical loading. The direct solution to this problem, except for very simple cases, is generally intractable since the formulation usually results in a set of nonlinear partial differential equations. Perturbation analysis has been used to obtain solutions to problems of simple geometry which are valid over a small range of magnitudes of the static biasing variables. When the structural geometry becomes complicated or when the change in frequency over a large range of the biasing magnitude is needed, as it would be in modeling the performance of a sensor design over its full operational range, a powerful numerical technique is necessary. A general class of finite elements will be presented which are based on the theory of Baumhauer and Tiersten for small fields superposed on a bias. The solution of the discretized problem is handled in two parts: 1) the solution of a static problem under a loading from the variable being sensed, and 2) the construction of a vibration eigen-value problem which incorporates the static solution vector into the element stiffness matrices via finite element shape functions. Relevant examples of the modeling of various resonant microsensors mentioned above are presented using materials such as Quartz, Gallium Arsenide, and Silicon.

Capacitive sensors are essential elements in many instrumentation circuits. The widespread use of these instruments and the need for integrating sensors on the same chip as the signal processing circuitry has motivated the developement of linearization techniques for the sensor output. Presented here are four methods which can be used to 'linearize' the essentially nonlinear nature of capacitive transducers. Described first is the polynomial linearization scheme, in which the Taylor series expansion of the sensor output is linearized by Khachab's method. An improvement in accuracy is obtained by the direct linearization scheme which scales the sensor output by the nonlinear denominator. Though this scheme is useful in cases where the signal processing circuitry must be separated from the interface circuitry, the hardware requirements are pretty intense. An alternative approach is the Summer-Divide scheme, wherein the difference and the sum of the top and bottom capacitances are ratioed to yield the linearized output. The last method is the constant charge injection scheme. Here a constant charge is put on top of the top of the plates and the subsequent voltages are differenced to yield a linear output. Nonlinearity associated with parasitic capacitances can be easily removed by feedback schemes.

Besides foundry facilities, CAD-tools are also required to move microsystems from research prototypes to an industrial market. CAD tools of microelectronics have been developed for more than 20 years, both in the field of circuit design tools and in the area of TCAD tools. Usually a microelectronics engineer is involved only in one side of the design: either he deals with application design or he participates in the manufacturing design, but not in both. This is one point that is to be followed in case of microsystem design, if higher level of design productivity is expected. Another point is that certain standards should also be established in case of microsystem design too: based on selected technologies a set of standard components should be predesigned and collected in a standard component library. This component library should be available from within microsystem design frameworks which might well be established by a proper configuration and extension of existing IC design frameworks. A very important point is the development of proper simulation models of microsystem components that are based on e.g. the FEM results of the predesign phase and are provided in the form of an analog VHDL script. After detailing the above mentioned considerations we discuss the development work concerning a microsystem design framework. Its goal is to provide a set of powerful tools for microsystem application designers. This future framework will be composed of different industry-standard CAD programs and different design databases which in certain cases are completed with special interfaces and special purpose simulation tools.

Microgyroscope applications are increasing in many fields, creating a need to develop low cost, compact gyroscopes. A full understanding of the operation principles and limitations of a rate gyroscope are essential in creating a small and low-cost device. Currently, there is no simulation tool available that can simulate both the electronic and nonelectronic components of a complete device. The work described in this paper presents a method to simulate a complete comb-drive rate microgyroscope using the SPICE electronic simulation tool. The simulation method given here can be generalized to a wide range of MEMS devices. An equivalent circuit was developed and simulated using SPICE. Results of the frequency response, and transient analysis are given. The results obtained reproduce the reported device operation. A set of important new findings are revealed from the analysis using this new tool. Using the transient analysis, for example, we can see that the requirement of a large quality factor of the sensor causes a very slow (open-loop) response and that a closed-loop operation is essential in this device. In addition, the nonideal factors are described and are analyzed using SPICE.

A finite element formulation is presented for modeling microelectromechanical systems (MEMS) to predict the behavior of dynamic response of piezoelectric ceramics subjected to both mechanical and electrical loadings. The FEM formulation presented is based on variational principle using the concept of virtual work. An unconditionally stable method for direct integration of the equations of piezoelectric material is introduced and can be put in the form of 3-step linear multistep method of second order equations. A thin cantilever plate with PZT sensor mounted on it, is investigated to show the feasibility of analysis using 20 node isoparametric 3D piezoelectric elements, flat-shell elements, transition elements which are implemented for the numerical analysis. Under a given initial loading, a structure would deform accordingly, and the distributed sensor outputs could be calculated at each time step. Such calculation can be used as design tool for MEMS structures containing sensors and actuators.

A monolithically integrated Fabry-Perot cavity pressure transducer has been fabricated using micromachining techniques that are fully compatible with silicon IC processing technology. Surface micromachining can produce a high quality micro cavity without the wafer bonding process that has been commonly used in previously fabricate Fabry-Perot pressure transducers. Dielectric film stacks consisting of layers of silicon nitride and silicon dioxide were used as mirrors. An air gap cavity was formed by selectively etching a sacrifical polysilicon layer. Remote pressure measurement was demonstrated using the micromachined Fabry-Perot cavity and a multimode optical fiber interconnect. The measured response of the sensor has been compared with theoretical simulation which takes into account the averaging effect caused by the shape of the deflected mirror in the cavity.

In this paper, we present a review of micromechanical Mach-Zehnder interferometer pressure sensors based on silicon integrated optics. The sensors are based on the ARROW waveguide structure and fabricated using IC compatible dielectrics.

A surface micromachined pressure sensor array has been designed and fabricated. The sensors are based upon deformable, silicon nitride diaphragms with polysilicon piezoresistors. Absolute pressure is detected by virtue of reference pressure cavities underneath the diaphragms. For this type of sensor, design tradeoffs must be made among allowable diaphragm deflection, diaphragm size, and desireable pressure ranges. Several fabrication issues were observed and addressed. Offset voltage, sensitivity, and nonlinearity of 100 micrometers diameter sensors were measured.

For consistent optimum electrical performance of a pressure sensor device, the die-attach process is very critical. A polymer adhesive is used for attaching the pressure sensor die to the plastic package. The polymer dispense process depends on the machine (die bonder) set up variables. The significant machine set up variables were the time of dispense, the pressure of dispense, and the gap between the dispense nozzle and the surface of dispense. A response surface design model was used to find the relation between the machine ste up variables and the response surface variables. It was found that the time of dispense, the interaction of the time and gap, square of time were the important variables that affect the result of dispense. A process window for machine set up variables were found to achieve optimum and consistent adhesive dispense.

This paper examines issues in the packaging of silicon-micromachined devices. Standard microelectonics packaging seeks to physically isolate the integrated circuits from harmful elements in the environment, to provide mechanical strength for the die, to facilitate thermal dissipation, and to sustain electrical communication with outside circuits. However, due to the many novel applications of MEMS devices, new sets of packaging requirements need to be met and further research on packaging technology to meet non-traditional requirements are needed.

A microvalve system made of silicon for use in hydrocephalus therapy is presented, which will provide an excellent intracranial pressure stabilization. Design and processing are described. Testing results (in vitro) are presented.

A 3D color graphics visualization package has been developed for MEMS design. The package takes input in CIF format and translates the 2D layout information into a 3D rendering of the design. The user can pan, zoom, and rotate the entire design, or select any individual object and translate, rotate, or scale it. 3Dmicrometers V allows interactive assembly and 3D modeling of hinged, out-of-plane structures. It is particularly useful when interacting parts need to be modeled. A kinematic modeler (collision detection and simulation) has been incorporated into the package. The user can easily manipulate parts and assembly structures, determine clearances, and check for design errors and unexpected inconsistences. Processes supported currently are two-poly/two-sacrificial surface processes (such as the process offered by MCNC in the US), and high aspect ratio processes such as LIGA. Depending on process complexity, new processes can be added to the system with very little effort. As a part of the UCLA MEMS synthesis and analysis CAD toolset, 3Dmicrometers V is public domain, and is available by anonynous FTP from synergy.icsl.ucla.edu. 3Dmicrometers V weas developed on a Sun Sparcstation and written in C with X11R5 Xt and PHIGS-SI graphics library.

Projection displays and microelectromechanical systems (MEMS) have evolved independently, occasionally crossing paths as early as the 1950s. But the commercially viable use of MEMS for projection displays has been illusive until the recent invention of Texas Instruments Digital Light Processing ^TM (DLP) technology. DLP technology is based on the Digital Micromirror Device^TM (DMD) microchip, a MEMS technology that is a semiconductor digital light switch that precisely controls a light source for projection display and hardcopy applications. DLP technology provides a unique business opportunity because of the timely convergence of market needs and technology advances. The world is rapidly moving to an all- digital communications and entertainment infrastructure. In the near future, most of the technologies necessary for this infrastrucutre will be available at the right performance and price levels. This will make commercially viable an all-digital chain (capture, compression, transmission, reception decompression, hearing, and viewing). Unfortunately, the digital images received today must be translated into analog signals for viewing on today's televisions. Digital video is the final link in the all-digital infrastructure and DLP technoogy provides that link. DLP technology is an enabler for digital, high-resolution, color projection displays that have high contrast, are bright, seamless, and have the accuracy of color and grayscale that can be achieved only by digital control. This paper contains an introduction to DMD and DLP technology, including the historical context from which to view their developemnt. The architecture, projection operation, and fabrication are presented. Finally, the paper includes an update about current DMD business opportunities in projection displays and hardcopy.

The rapid expansion in number and scope of research projects in the general area of micromachining technology makes the field especially interesting. Advances, particularly from universities and research institutes, suggest that increased commercial development is likely in a number of new fields. The nature of the fabrication technologies used to make these prototype parts, however, lead to difficulties in quickly reaping commercial benefits from these technolgoical advancements. This is in contrast to the nature of commercial development in the integrated circuit industry, where standardized processes result in rapid development of systems and products which use new designs.