As electronic and mechanical sensors become microscopic, our world-view of contamination issues has to adjust. A few years ago, MEMS was a catchword. MEMS, or Micro-Electro-Mechanical Systems, combines mechanical elements, sensors, actuators, and electronics on a common silicon substrate. The technology is already used for pressure sensors, accelerometers and more. In a recent column, we discussed a SAW contamination surface film detector based on MEMS technology.1
More recently, the term NEMS has been introduced, with the îMicroî becoming ìNano.î These smaller devices have the promise of even greater sensitivity, and much lower power consumption. As detection sensitivity increases, an array of new or rarely used words are likely to become commonplace. Milli-, micro-, and nano- prefixes are being joined by atto- (10-18), zepto- (10-21), and yocto- (10-24›. NEMS-based mass sensors have been proposed as part of a toolkit to be used for the inexpensive real-time analysis or detection of toxics or explosives. Protein molecules or viruses could be distinguished from one another by weight or by the affinity for a particular surface (e.g. body-antibody).2
Nanotribology takes the study of friction and adhesion and lubrication down to the molecular or atomic level. New contamination sensors, which are sensitive to extremely small quantities (including a balance with potential to weigh a single atom!), are under development. We will discuss some of these new sensors in greater detail next month.
An inherent consequence of the expected trend toward NEMS-based devices will be increased significance of surface attributes including cleanliness, potential contamination sources, and surface quality or surface attributes. As dimensions decrease, the surface-to-volume ratio increases. At the ultimate limit one could envision a situation where the entire device is a monolayer surface with an analogous power and delicacy that biologists recognize in membrane structures.
In such situations, surface contamination is an inherently important issue. Along the same line of thought, contamination particles that may be insignificant specks on a larger device may be virtual boulders on a NEMS device. Microscopy reveals that even the smoothest surface has roughness or structure at high magnification. Both MEMS and NEMS involve both structural and mechanical aspects, either rotating parts or vibrating parts. Contaminating particles can interfere with mechanical motion either by physical constraint (a microscopic ìshoe in the doorî) or by changing balance conditions and degrading (or shifting) vibration resonances.
Another consequence of the higher surface-to-volume ratio is that the device physics is more dominated by surface effects than bulk effects. NEMS devices may have a substantial fraction of the total atoms at or near the surface.3 At this point, mechanical aspects such as tensile strength change from bulk effects to surface effects. Particles or thin films, such as deposited from airborne molecular contamination,4 can interfere with surface physics effects. MEMS or NEMS developers face some of the same cleanliness issues as those in wafer fabrication, with the added complication that most of the mechanical devices are not planar, making it harder to reach all surfaces. Since in some cases it has been demonstrated that supposed surface contamination can be actually a positive surface attribute,5 one can envision the possibility of ìcontaminationî used as a protective measure on microscopic devices.
Note: Thanks to Dr. Panos Datskos, Research Staff Member at Oak Ridge National Laboratory and a Research Associate Professor at the University of Tennessee, for his helpful contributions to and review of this column.
References
1 B. Kanegsberg, E. Kanegsberg. ìContamination in and out of the Cleanroom,î A2C2, (November, 2003).
2 N. Lavrik, P. Datskos. ìNano-mechanics Weighs In,î Physics World, (April 2004); available at http://physicsweb.org/article/world/17/4/3.
3 M. L. Roukes. ìNanoelectromechanical systems face the future,î Physics World, (February, 2001); available at http://physicsweb.org/article/world/14/2/8.
4 B. Kanegsberg, M. Chawla. A2C2, op. cit. (February-June, 2001).
5 B. Kanegsberg, M. Chawla. A2C2, op. cit. (January-February, 2002).
