The @MATEC Archives

Mary Jane Willis
@matec.gif (37291 bytes)  

Star Date 2366 Log Entry – Today we were successful in using our submicroscopic robots to remove individually diseased cells from a blood vessel of one of our crew. We have now proven that these "nanites" are capable of performing intracellular surgery. (

Science fiction or reality? If we look at what is being accomplished with the development of micromachines, we may be closer to this technology than we think. The predecessors of the nanites are already in development and in some cases, in production for commercial use. You might even have one in your car. The technology of little machines so small that they can attach themselves to human cells is called MEMS or MicroElectroMechanical Systems.

MEMS are microscopic sensors and actuators that are currently being developed for a variety of applications from the automotive to the biomedical industries. The most popular MEMS produced today is the micromachined pressure sensor. Such devices are used in the medical industry as blood pressure sensors and in the automotive industry to help increase the fuel efficiency of U.S. cars. These sensors are small, cheap, and easy to replace, making them an easily disposable device. Another commercial use of MEMS is the accelerometer used in automotive air bag systems. These micromachined accelerometers provide the same function as the previously used crash sensors but at a more competitive cost. Texas Instruments has developed a MEMS process that creates a large array of digitally controlled mirrors on a single silicon chip. These chips can be used in projection displays, rear projection televisions and 600 dpi color printers.

Future applications of MEMS include tiny heaters that achieve temperatures as high as 1000 C; sensors and pumps that analyze one’s body chemistry and deliver a precise amount of life-saving drug; spectrometers that fit in the palm of the hand; probes that stimulate particular brain cells enabling the blind to see; and tiny robotic hands that move biological tissue or cells.

The development of MEMS has been relatively fast thanks to the technology used in the semiconductor manufacturing industry. Building a micromachine requires essentially the same technology as building a microprocessor. Cleanrooms are required to prevent contamination due to the product’s size, a photolithography process is needed to pattern the polysilicon and silicon dioxide layers, and an etch process is required to create anchor points and free-standing devices between the various layers.

Two primary differences between manufacturing a micromachine and manufacturing an integrated circuit are the thickness of the layers and the amount of material that is removed in each layer. Micromachines require layers that are measured in microns rather than angstroms required for semiconductor devices. Because these devices are machines, they must be able to rotate, oscillate, and pump fluids. Such actions require free-standing or floating devices; therefore, enough material must be removed to allow for movement.

One micromachining process, referred to as "bulk micromachining", removes most of the silicon wafer by physical or chemical etching, leaving a single-crystal microstructure. This process is used for optical structures and single-layer wafer-based structures. In another process, "surface micromachining", an oxide layer is grown on the silicon substrate, then the desired microstructure material such as polysilicon is deposited on top. The sacrificial oxide layer is laterally etched away in order to produce free-standing microstructures on the silicon substrate. The actual structure to be built will determine the number of polysilicon / silicon dioxide layers required. For instance, a three-level polysilicon process is required to build a linear comb-drive actuator linked to multiple rotating gears. These multi-layer polysilicon processes are referred to as MUMPS or Multi-User MEMS. (Visit for several examples of such devices.)

As with all new products, there are problems. Micromachines have the same properties as their larger counterparts. Moving parts produce friction and thus heat. This heat has to be dissipated which can be difficult for such a small surface area. The friction can also cause wear, and for such small components it doesn’t take much wear to affect the operation of the device. Other problems include interfacing these devices with their electronic control signals, layers sticking together due to surface tension created in wet etch processing, and rotors that aren’t round due to the CAD packages that use many sided polygons for circles. As with all commercial products, mass production is required for cost effectiveness. To date this has been difficult due to low yields and the lack of batch processing for such products. Designing a generic package has also proven to be a problem. It’s not like one can just pop these micromachines into a board. MEMS need to be incorporated into the environment in which they are to do work. Such environments could be a petri dish, a human brain, or a tunneling electron microscope. However, as with most technologies, given the time, the money and the people, such problems can be overcome.

The MCNC MEMS Technology Applications Center projects a 10 – 20% annual growth rate for the MEMS industry, with a potential of a greater than $8 million market by the year 2001. Other projections are showing a $10 – 20 million market by the end of this decade. These projections are due to the potential that these components have on impacting the commercial and the defense markets. Many applications already exist and many more are just being discovered. The basic technology for making micromachines already exists, so it’s just a matter of time before these microdevices are injected into every aspect of our lives.



Banks, Danny. "Microsystems, Microsensors & Microactuators: An Introduction". Microsoft Internet Explorer. 1996.

Sniegowski, J. J. "Moving the world with surface micromaching". Solid State Technology. February 1996. Pp 83 – 89.

Gabriel, Kaigham J. "Engineering Microscopic Machines". Scientific America. September, 1995. Pp. 150-153.

Pottenger, Michael. Eyre, Beverly. Kruglick, Ezekiel. Lin, Gisela. "MEMS: The maturing of a new technology". Solid State Technology., September 1997. Pp. 89 – 96.


For more information on MEMS visit the following websites:


Biography on Mary Jane Willis

Mary Jane Willis is an instructor of electronics and manufacturing at Albuquerque TVI. She has been at TVI since 1978. In addition to teaching, Mary Jane works closely with industry in identifying and developing training needs. Through a National Science Foundation grant she has been able to meet and work with instructors throughout the United States who are pursuing the implementation of semiconductor manufacturing technology into their curriculum.

Whenever a term break comes, MJ takes off into the backcountry. She loves hiking and backpacking in the canyons of Utah and the mountains of New Mexico and Colorado. When time allows, she heads to the water for kayaking or rafting trips. For daily relaxation, Mary Jane enjoys walking, cooking, and gardening.