Dr. Walter Merrill, Associate for Business Development, John H. Glenn Research Center, Cleveland, OH
- Monday, 18 December 2006
NASA Tech Briefs: What are MEMS?
Dr. Walter Merrill: To be quite literal, they are systems composed of both electronic and mechanical subsystems, and are micro in the sense that they are small and have feature sizes that are measured in microns (millionths of a meter). Combining sensors with actuators yielding to this micro domain would be a good example of a MEMS device.
NTB: What are the challenges faced by MEMS developers?
Dr. Merrill: The first is packaging. We understand how to make the little die using the integrated-chip-like process. This is not the end of the story, however. You have to be able to put that little die into some sort of package and connect it to the outside world. So, you want to be connected to the real world, but at the same time you want to protect your device from the real-world environment. This is sort of a dual problem where the packaging becomes more of a challenge (it also tends to be a cost driver).
Because it is still a relatively new field, and although there are some exceptions, reliability is a concern. In general, there have not been many of these devices that have been fielded. The field is becoming more mature, but reliability numbers are still a little hard to come by.
Standards are something of an issue as well. In one sense, when you are developing integrated chips, the standards are a little easier to come by because you’re making transistors and resistors, and the domain of what you are making is contained. In MEMS devices you can be making anything; therefore, standards are much more of a problem for the industry.
NTB: How is NASA utilizing MEMS technology?
Dr. Merrill: There is the big world of MEMS, and then there is the world that is of interest to NASA, which perhaps is a bit narrower. We often are interested in devices that will operate in harsh or extreme environments, such as on the lunar or the Martian surface where you get extremes in temperature. We are also interested in extremely hot conditions where you would want a MEMS device to operate in or as part of a rocket engine or a gas-turbine engine. Combustion processes in each of those can result in very high temperatures.
My particular work for the past five years has been to look at MEMS devices that are made out of a different material – silicon carbide. This material can in fact operate at 500°C and higher.
We also are looking at harsh conditions from an entirely different perspective. We have done some work locally with the Cleveland Clinic Foundation that has a well-known bio-MEMS department. They are actually looking at MEMS devices that will go into the body. If you look at internal body chemistry, if you try to introduce a foreign device the body is quite good at trying to get rid of it by dissolving or encapsulating it. We have had some success with this type of application by developing a transducer that will provide an ultrasonic image of a vascular or arterial pathway during a catheterization process. We have also developed a “bone-pressure sensor” that measures how successfully bones knit back together, typically after back surgery.
NTB: What are the overarching goals of the technology?
Dr. Merrill: Certainly from the NASA perspective, size, weight, and mass are always considerations. Goals of the technology include new functionally and new capabilities, as well as existing capabilities that you can put somewhere that you couldn’t put before.
From the commercial point of view, cost is important. By making use of the mass-production processes – as is the case in the IC industry – and churning out airbag sensors, it now becomes affordable to put them in your car. If you could make sensors small, you could consider not just putting one sensor into a location, you could envision putting in a whole array.
The industry has looked to MEMS to provide not just one of these features, but has really looked to accomplish them all.