J. M. Berg, T. Dallas, W. Dayawansa, R. Gale, M. Holtz, and H. Temkin
Chips containing large numbers (~10 6) of individually addressable digital micromirrors are now commonly used in projections displays. Arrays of analog micromirrors are used in fiber-to-fiber switching in optical telecommunication systems. However, despite rapid advances in technology, some of the basic problems of micromirror fabrication and control are not well understood. A well-known problem known as stiction, where the tilting mirror sticks to the underlying base structure, appears to be caused by a combination of capillary forces, electrostatic attraction, or direct chemical bonding. Chemical passivation coatings are used to empirically control stiction.
Fluorinated fatty acid self-assembled monolayers (SAM) are used on the aluminum oxide surface of digital micromirrors fabricated by Texas Instruments (our collaborator in this project). Another approach is to coat the mirror with silicone polymers. All these processes require good understanding of surfaces exposed in the fabrication process and cross-disciplinary approach. We have designed and built a contact angle measurement system that allows us to use different coatings to determine the reactivity of the surface. We also have developed wet process and plasma treatments to alter the nature of the surface, using the established surface passivation treatment as a baseline. We have developed an optical system and software to exercise individual micromirrors and generate the statistical analysis of the results. The goal is to identify a robust passivant that would eliminate the need for hermetic packaging of micromirror chips.
Another fundamental problem of microsystems in general and micromirrors in particular is their tendency to exhibit non-linear frequency dependence, known as snap-through, which makes them difficult to control. Nonlinear control theory is needed to deal with this problem and the implementation of this theory in electronic circuits is of particular interest. One approach that we are pursuing is to include low level interactivity of the elements of an array of such spatial light modulators.
Our program also includes the design and fabrication of analog micromirrors. The novelty of our design lies in inclusion of local interactions between individual mirrors of an array.
This leads to collective behavior that offers advantages in mimicking vision, adaptive optics, and beam steering. A mathematical model of the control algorithm needed to create this behavior has been constructed and is now being implemented in an electronic circuit. Capacitive feedback sensing for on-chip implementation of this concept is also underway.
Research support: NSF, Applied MEMS, and Texas Instruments