Some landmarks
The first display of a miniature torsion-mirror (~2x2mm)
It was carried out by Kurt Peterson at IBM in 1982.
If we take this innovation as a reference, Peterson anticipated that “silicon mechanical Microsystems could find a practical application for display screens (most particularly if we managed to integrate the silicon command circuit on the same chip by sending electronically on a matrix way the two-dimensional mirror network)”. This assertion turned out to be excellent judging by the success of Microsystems to make screens such as the Texas Instrument DMD (Digital Micro-mirror Device).
Rappel :
See casework in Introduction to micro-optics.
We can quickly see that the screens using micro-mirrors gratings are the oldest scope of optical Microsystems. But we will see that the trend led to many other applications for micro-mirror networks.
The basic idea of digital micro-mirror systems consists in making a bistable micro-mirrors grating in order to intercept a beam of light so that one of the positions of a micro-mirror directs light on a screen while another position directs it to an absorbent block. For this, each micro-mirror activates and modulates a different pixel. We can also establishing a gray scale while making a dithering on the mirror. Color screens can be reached by standard methods: color wheel (the most compact and cheapest, and therefore, the most usual) or triple-mirrors gratings for the three fundamental colors (red, green and blue) combined with dichroic optical mirrors and displaying on the same screen. We will go into the details in the fourth part.
An alternative concept of micro-mirror screens
This concept was suggested in 1994 by a team led by David M. Bloom, from Stanford University. The modulation method, in contrast with micro-mirrors, consisted in using light diffraction by miniature grating, one for each pixel on a screen. Its inventor called it “GLV Grating Light Valve”. More details about the GLV structure and micro-mechanics are in the fourth part. In most cases of GLV technology implementation, light is emitted through a Schlieren system that blocks diffraction at order zero and lets diffraction pass at ±1order. The result: when the grating is inactive, light is not transmitted on the screen whereas during the actuation of the grating, the corresponding pixel on the screen is lit up by the interference pattern constituted by -1 and +1 orders. The main difference between GLV and micro-mirror made screens is that the GLV actuators produce long (~ 0,5 mm) and thin (a few microns) beams, varying between two vertical positions and that can be actuated faster (a tenth of a MHz versus some kHz). For this, the GLV needs a 2D pixel matrix, but is implemented on a simple line that is read quickly in the vertical direction to produce the complete effect. For the same reason, the GLV line can have many more pixels than micro-mirror screens of the same generation with the linear dimension. But there is a problem: as light modulation is based on diffraction, its efficiency is reduced and needs high power laser sources to have enough shine. This can explain why GLV is only for high technology display markets, which can bear laser sources cost and bulk in return for an excellent resolution of thousands of pixels by dimension on a large projection display surface (for instance for a digital movie). On the other hand, micro-mirror screens were a success on the mass computer and video projector market in which XGA resolution is sufficient and price, bulk and weight are decisive factors.
Remarque :
It's interesting to compare GLV[1] and DMD[2] technologies because we can see how overlapping between optical principles (reflection vs diffraction) on the one hand and mobile elements (revolving rectangular blades vs thin beams moving vertically) on the other hand can influence the other system extrinsic parameters(choice of light source, resolution, bulk, price) and completely change the result of micro-systems production.
Remarque :
We can notice that Texas Instruments has considered (since 2004) that the laser printer market can be an alternative for its DMD products.
" Silicon micro-optical bench "
Nowadays, a natural extension of micro-mirrors is the instruments known as “silicon micro-optical bench” for which optical devices are miniaturized in a silicon chip, whereas with traditional optics, several cm² are necessary on an optical table. Many ingenious systems have been implemented to treat light at one small scale, among them automatic micro-mirrors (pop-up), diffraction micro-lenses (blades with a Fresnel zone). Such systems have a strong potential for a large number of applications in optical treatment of signals and to set up captors.