Science Communications - Publicity for Technologya

Imagination for Microelectronics

They can see, hear, taste, touch and smell better than our natural senses. They can go places we wouldn’t dream of going. They react faster than you ever imagined. And they surround us at every moment of the day. Measured in microns, or millionths of a meter, Micro Electro Mechanical Systems, better known by the acronym MEMS, are the tiniest machines ever manufactured in commercial quantities. They’re in ink jet printer heads, automobile airbags, cell phones, laptop computers and projection TVs to name a few applications. And soon they will be appearing with increasing frequency inside our bodies.

As a matter of principle, MEMS do three things: first, a sensor detects a non-electrical force such as pressure, motion, airflow, heat, sound, or magnetism; second, a transducer converts those forces into electrical signals; and third, in response to those electrical signals an actuator performs a mechanical action of some sort or other.

Just how each of those things gets done is a matter of intelligence, imagination and innovation on the part of MEMS designers and manufacturers.


Although today MEMS is a $5 billion-plus global industry, it remains a little known fact that some of the earliest MEMS patents originated in Pittsburgh. For about a decade, beginning in 1965, Dr. Harvey Nathanson, in conjunction with Robert Wickstrom, Dr. William Newell and a team of 18 engineers, scientists and technicians at the Westinghouse Research Labs in Pittsburgh, developed a number of silicon-friendly MEMS devices, including vibrating beam MEMS, microwave relay and contact MEMS, field emission MEMS, accelerometer MEMS and TV Projection MEMS.

Patented in 1968, Nathanson’s first vibrating beam MEMS device, called a “resonating gate transistor,” solved a stubborn tuning problem for the young microelectronics industry. The one millimeter-long device responded to an extremely narrow range of electrical input signals and transmitted only those within the selected range to an output circuit, while ignoring all other frequencies, thus serving as a frequency filter for integrated circuits. This breakthrough enabled the development of an array of MEMS devices, such as those cited above as well as others.

Electromechanically speaking, Nathanson’s resonant gate transistors employed an electrostatic field between the suspended end of a conductive cantilever and a silicon substrate, to which the other end of the cantilever was affixed. When vibrated (think buzzed) by an electrostatic force, the suspended end of the cantilever would alternately move closer to and further from the substrate, rapidly decreasing and increasing the resistance of a small “sense” transistor. The cantilever’s length and shape gave it a fixed frequency, just as the varying length and thickness of each string on a piano gives each note its own rate of vibration. When the varying electrostatic field was on and the cantilever’s resonant frequency was the same as the input frequency, the cantilever would vibrate. Any other frequency would be ignored.

If getting all those elements and phenomena to work together at any size seems complicated, doing it in the mid-1960s in a device that would fit on the head of a pin is truly cause for amazement. Included among the devices enabled by the resonant gate transistor was an accelerometer that employed the same principles in a very different way. Nathanson’s accelerometer measured changes in motion by exploiting the principle that, when started or stopped suddenly, each micro-scale cantilever would move a well- understood microscopic distance – much the way your body reacts to a sudden start or stop when you’re driving. Similarly, the cantilever beams in Nathanson’s accelerometer would move a different distance at different rates of acceleration. The correlation between the rate of acceleration (or deceleration) and the response of all the cantilevers in the array is known as the “acceleration signature.” The closing of switches beneath the array would be an indicator of, for example, the desired safe launch of an artillery shell. Today, inertial sensors in automobile airbags work on the same principle, although they employ a different set of materials to do the sensing.

Perhaps of greater long-term consequence to the MEMS industry is the fact that in the process of developing a manufacturing method for their new devices, Nathanson’s group innovated a method for undercutting layers of silicon wafers using masks and sacrificial layers, which patterned the micro-scale features to be etched. Upon completion of the etching of the sacrificial layer, the mask was disposed of. At that time, sacrificial layers were considered to be revolutionary. Today, they are a mainstay of MEMS manufacturing.

In a recent interview, Nathanson, who is retired in Pittsburgh and works as a consultant to Northrop Grumman Corporation, Westinghouse’s corporate successor, commented on the richness of MEMS technology. “The beauty of MEMS and the use of photolithographic techniques is that the kinds of mechanical structure you can get are almost only limited by your imagination.” Today, in Pittsburgh, evidence of the truth of Nathanson’s reference to imagination in the MEMS industry abounds. In the academic sphere, researchers are investigating the development and application of new sensing and actuation materials. Product and process developers are inventing new ways of incorporating them into device manufacturing. Bio-MEMS people are looking for ways to power and communicate with implanted or body-worn devices. And market-driven, private-sector companies are incorporating MEMS devices into commercial products and taking them to market. Examples of those activities in the Pittsburgh area follow.


Piezoelectric crystals generate electricity when put under mechanical stress; when they are bent or squeezed, for instance. Conversely, piezo crystals expand when put under an electrical charge and return to their original size when the charge is removed. Chances are, the shrill beep that awakens you in the morning is powered by a piezoelectric crystal. Same thing with your cell phone’s ring tone and the smoke alarm in that unreachable spot on the ceiling in the hall.

Professor William Clark of the University of Pittsburgh works in the field of “smart materials.” Dr. Clark has developed a way of using the reciprocal force-charge generation properties of piezoelectric materials to develop a patented tunable piezoelectric resonator capable of handling frequencies in the gigahertz range. While piezoelectric materials are mainstays for timing devices in electronics, they are limited by the fact that their resonant frequency can be adjusted only by changing the size, shape or composition of the piezo element. Clark’s invention employs an electronic feedback system that allows the piezo element to shift frequencies by changing its effective stiffness. The device utilizes the piezoelectric material as both a sensor and actuator. Such a component would be key to enabling the further miniaturization of microelectronic devices, he stated.

Cousins of piezoelectric materials, piezoresistive materials change their thermal and electrical conductivity when put under mechanical stress. In another example of using smart materials in MEMS, Professor Gary Fedder of Carnegie Mellon University is using piezoresistive materials for bioimplantable bone sensors to measure bone stress inside the body. Because orthopedic surgeons have no way of measuring the strength of replaced, implanted, reconstructed or bio-engineered bones, such a device is of great benefit to recovering patients. The sensor, which is between 1 and 3 millimeters square, employs an array of piezoresistive silicon pillars that are sized to allow bone cells to interlock. It is powered by an external coil, something like that used in an MRI, but more compact, and communicates by means of a coil antenna to transmit signals to a signal processor outside the body.

As their name suggests, the physical state of phase-change materials can be altered by forces, such as heat or light. For example, candle wax changes phases from solid to liquid and back, when it melts and re-solidifies. Rewriteable compact discs use phase change materials that are responsive to laser light. Professor James Bain of Carnegie Mellon University is investigating the use of “phase- change” materials for data storage, which he suggests would eventually become part of a new generation of computer devices called Memory Intensive Self-Configuring Integrated Circuits. In the case of the materials with which Bain and his colleagues -- professors Fedder, T.E. (Ed) Schlesinger and Larry Pileggi -- are working, the precise manipulation of the rates of melting and cooling of a special class of compounds, called chalcogenides, are used to switch the material’s phase from crystalline to amorphous and back again. The process results in either a high reflectivity conductor (crystalline) or a low reflectivity insulator (amorphous) material. The materials’ two sets of materials properties would serve as a mechanism for recording binary code, as well as making and breaking electrical connections. A MEMS probe tip, similar to that used in scanning probe microscopes, would melt and re-melt the materials.

MEMS in the Marketplace

Close relatives of piezoelectric materials, pyroelectric materials transform heat into electricity. In an outstanding example of the integration of existing science and advanced technology, Bridge Semiconductor has developed a low-cost thermal imaging module that allows firefighters to see through smoke. It also enables night drivers to see five times as far as with headlights. The device is based on the integration of several technologies: 1) A century-old, but little used, component called a Fabry- Perot resonator, which in this case, captures and intensifies invisible, long infrared waves in a mirrored heat trap and transfers the amplified heat to: 2) a proprietary Bridge Infrared Focal Plane Array, which uses “pyroelectric” materials to convert the heat into electricity, which is transferred to; 3) a proprietary Bridge signal processor that converts the electrical signals into a black and white video signal. Bridge CEO, Joshua Ziff states that expensive thermal imaging modules based on different technologies are already pioneering the high-end automobile and first responder market sectors. Bridge plans on entering and expanding those markets when it introduces its lower-cost product later this year.

In another example of using imagination and advanced technology to apply well- known physical principles to existing technologies, Dr. Michele Migliuolo, CEO of Verimetra, has developed a method of growing a nanoscale thin-film array of proprietary metal composite thermocouples on cardio catheter tips for use in the thermal ablation of heart tissue to control irregular heartbeats. Migliuolo explained the technology and its benefits: “One of the ways atrial fibrillation is treated is by the localized heating of the heart muscle with a catheter tip to interrupt the problematic electrical pathway. Currently there is no precise indication of the temperature right at the interface of the hot tip of the catheter and the heart muscle. This is due to the fact that the tip is metal, the heart muscle is tissue, it’s all in blood and things are moving. In response to this problem, we designed an array of multiple temperature sensors for this hot tip. So the doctor can have a true indication of how hot the area being burned is, reducing the risk of post-operative complications such as internal bleeding or strokes.”

In an example of using advanced materials to further improve revolutionary products, Dr. Metin Sitti of Carnegie Mellon’s Robotics Laboratory is collaborating with Dr. Ragu Appasamy to improve the capsule endoscope or pill camera, a device that Appasamy uses in his UPMC Southside gastroenterology practice. The PillcamTM, a patented and trademarked product of Israeli company Given Imaging, is a jellybean-sized MEMS camera that, after being swallowed, takes pictures of a patient’s small intestines at the rate of two exposures per second. The picture signals are transmitted by radio waves to a body-worn monitor. When the pill cam’s journey is complete, the doctor can view either a movie or a set of still images of the camera’s voyage. Appasamy’s favorable experience with it has prompted him to imagine making the device ambulatory and incorporating MEMS-scale instruments for purposes of taking biopsies and delivering drugs. A means of ambulation, controllable from outside the body, would enable a doctor to stop the camera and examine areas of concern in real- time and in greater detail. The added ability to take biopsies during the pill cam’s journey would, in many cases, obviate the need for exploratory surgery. On the engineering side of this effort, Sitti, whose area of interest is micro-nano robotics, is working with a material called “shape memory alloy wire” to make legs for the pill cam. Sitti describes a configuration of MEMS-scale wires, springs and pulleys that would extend the pill cam’s legs when heated by an external power source but collapse into a retracted position at all other times.

Ellen McDevitt, Executive Director of the MEMS Industry Group, which is headquartered here in Pittsburgh, recently reported having attended a DARPA conference in which two Pittsburghers were cited as having had profound impact on the MEMS industry. One was Harvey Nathanson, who was cited for his singular impact during the industry’s infancy in the 1960s and ‘70s. The other was Dr. Ken Gabriel, Founder of both Akustica and the MEMS Industry Group.

Akustica has patented a manufacturing process that integrates MEMS into standard silicon technology. Marketed under the brand name “Sensory Silicon,” the process employs the repeated application and removal of sequential metallic and sacrificial layers of material to create intricate surface features into which micro-machined metallic mesh layers are embedded and sealed to form micro-speakers and microphones that are not attached components, but intrinsic features of the CMOS chip.

Gabriel elaborated upon the motivating force behind his vision for Akustica: “If you were to take your cell phone and travel back in time and show it to Alexander Graham Bell, the only two parts he would recognize would be the microphone and the speaker. “We are focused on acoustic applications because they represent a large, existing, growing market that needs a new solution, because the old technology of doing electret condenser microphones cannot be shrunk anymore. Nor can they provide the features necessary to meet the voice quality demands for voice recognition, voice messaging, voice command and VOIP (voice over internet protocol). So for us it’s a huge opportunity. There are more than 1.5 billion microphones made every year and that business opportunity is a very good match for our technology. Our CMOS MEMS platform is going to enable Akustica to deliver new ways of capturing and reproducing sound for all the new audio and voice applications for the future.”

Located in downtown Pittsburgh, Body Media is not a MEMS manufacturer but a MEMS consumer. Its patented product line of body-worn sensing devices for monitoring health fitness and wellness includes an armband monitor - branded as bodybuggTM - which is used in health club wellness programs. In addition, BodyMedia is developing an armband ambulatory heart monitor for research purposes. The armband heart monitor does electrocardiograms while patients go about their everyday activities. The company’s system patent specifies the incorporation of certain devices into their patented products, including accelerometers, galvanic skin response sensors and heat flux sensors along with an electronic data feedback monitor, some of which are filled by MEMS today and all of which may appropriately be filled by MEMS in the future.

Summarizing his thoughts about the significance of MEMS, Dr. Nathanson noted, “To me, integrated circuits, up until the onset of MEMS, were basically calculators. They were extremely sophisticated. They could do millions and millions of calculations per second. And they could also do all the things you can do with a computer. But they’re blind and deaf. They’re not dumb, but they’re blind and deaf. One of the things that MEMS can do is provide the sensor package to the real world. MEMS can provide the eyes and ears - something that has been missing. And they provide it at the scale and with the same processes as the integrated circuit. This is all in the size of the chip.” Surely, Nathanson’s vision is borne out by what’s happening in Pittsburgh today. What is more, from the look of things, imagination may be the key ingredient.

This article first appeared as a TEQ cover story. You can read the original on the Pittsburgh Technology Council’s website.

©Copyright 2005 Thomas P. Imerito/ dba Science Communications

Bookmark and Share

©2009 Science Communications