Tech Students win Sandia MEMS Competition
written by Kippra D. Hopper
A researcher squints through the eyepiece of an electron microscope, peering into the world of the small. This is a place where an ant’s foot appears the size of a camel’s hoof, and an antennae looks like a tree limb. Texas Tech researchers and students are familiar with this world unseen to the human eye and are creating micro machines that function in the micro world.
Called microelectromechanical systems, or MEMS, these mechanical devices are measured in micrometers, and if Texas Tech University researchers and students are successful in their designs, the university may be vying for the title of creating the world’s smallest clock mechanism and drive system. Seventy of these micromechanical clocks, with hands that move, will fit into the area of a penny. The College of Engineering research team also is vying for the world’s smallest movable micro chain; these achievements have garnered the attention of Sandia National Laboratories in Albuquerque, New Mexico.
Students from Texas Tech University’s Electrical and Computer Engineering and Mechanical Engineering departments have won the first MEMS design competition sponsored by Sandia National Laboratories. The winning entry is a combination of four individual designs that include the micromechanical clock, the micro chain, a torsion micro mirror and a micron-sized atomic force microscope. Now, team members led by student Phillip Beverly and faculty adviser Tim Dallas, Ph.D., assistant professor of electrical and computer engineering and associate director of Texas Tech’s Nano Tech Center, await the news of whether their designs can function after being fabricated at Sandia.
MEMS devices are used in pressure sensors, gyroscopes, air bag sensors, projection televisions and video projectors, as well as biomedical devices. In research laboratories since the 1980s, MEMS devices began to materialize as commercial products in the mid-1990s. They are used to make pressure, temperature, chemical and vibration sensors, light reflectors and switches as well as accelerometers for airbags, vehicle control, pacemakers and electronic gaming . The technology also is used to make ink jet print heads.
The devices are built onto semiconductor chips through a micromachining fabrication process. The Texas Tech engineering team will have its designs fabricated for free, using Sandia’s fabrication process, called SUMMiT V™, the world’s most advanced silicon surface micromachining fabrication process. Beverly, of Arlington, Texas, and Dallas visited Sandia this summer to present their ideas, titled, “The 18mm² Classroom,” and to tour the facilities at the national lab. Contest judge and University of Utah Professor Bob Huber chose the Texas Tech team’s design based on the use of Sandia’s SUMMiT V™ strengths, usefulness of the design for educational demonstrations, and uniqueness of design.
“What we’re trying to do, hopefully, is to design some new devices based upon the fabrication sequence that Sandia is capable of carrying out. Sandia has a standardized process, and we’re trying to see what we can build using their process sequence. These are all devices made on silicon wafers that are 6 or 8 inches in diameter and they’re made by building layers of silicon and other materials with spacer material between the layers to create the structures,” explains Dallas.
The Texas Tech team designed polysilicon MEMS using a customized AutoCAD-based layout system. With the computer-aided drawing software, the devices are designed and constructed layer by layer using additive and subtractive processes along with sacrificial layers that are removed in the final processing step to release the moving structures
The micromechanical clock was the first device designed by the Texas Tech team, which includes faculty members Jordan Berg, Ph.D., associate professor of mechanical engineering, and Richard Gale, Ph.D., professor of electrical and computer engineering, as well as 10 other students. This mechanical time piece offers a system to read the actual hands of the clock through a microscope. Dallas said that if the clock works, it will be the smallest clock ever designed and fabricated. “The clock provides a good test to see how much friction can be overcome based on our design. There are a lot of gears and the gears have to change levels in order to get power to the second hand, minute hand and hour hand.”
The micromechanical clock will measure time like a regular clock, with the difference being its smaller size, Beverly comments. “This is basically the output part of the clock. A Torsional Ratcheting Actuator is used to drive a 13-gear drive system with three additional gears used to rotate the second, minute and hour hands. We have to accurately spin our motor in order to keep accurate time but that’s easily done because of the nature of the device and how we can generate very accurate pulse generators that this runs on.”
Precisely controlling the frequency of the voltage going into the motor is readily done, said Dallas. “Because there are really not that many mechanical devices out there at this scale, there is still a lot to be learned about durability and liability. This technology really is at the point of answering the question of how reliable it is for any length of time.”
The micro mirror is the best example to explain the SUMMiT V™ process and how MEMS are created. In all video projectors is a small computer chip made up of millions of separate small mirrors. What the human eye can see on the chip is a black and white image, created when one mirror is turned one way and another mirror is turned the other way. The reflection of light off the mirrors creates the image. On a chip from Texas Instruments that Dallas exhibits, the surface is made up of 2.2 million small mirrors that flip back and forth on hinges, operating like teeter-totters, bringing an image of contrasts in gray scale. The intensity of color is dependent upon the length of time the mirror is in the bright position and how long it is in the dark position for each one of the colors, he says.
Part of the process of fabrication is a patterning technique called photolithography, in many ways a three-dimensional photographic process in which emulsion after a chemical interaction stays or leaves the surface of photographic film, leaving behind an image in blacks, whites and grays. “The SUMMiT V™ MEMS fabrication involves process sequences that layer poly silicon, or a poly crystal in a version of silicon. There are many different types of arrangements of silicon atoms. Sandia’s technology uses poly crystal and silicon. Silicon dioxide, or silicon with oxygen atoms, provides the spacer material, or the sacrificial layers, that can be removed collectively. Poly silicon is used to bring electrical energy to the device. Upon that layer, we build other layers to create these complicated structures,” Dallas says.
"For the micro mirror technology, the material on the silicon surface is mostly aluminum and silicon dioxide. On the next layer, spacer 1, there is separation between where the hinge is that moves the mirror and the surface. Like when you’re making cookies, you’ve got to stamp them out at some point to separate them, so we cut out basically each mirror. That’s just one major step. This is all encased in the spacer material. The next step is to remove that. It releases everything, so it’s called the sacrificial layer and the result is the three-dimensional patternable material," Dallas explains.
One major issue with these micro-devices is that some of the forces that are important at the macro scale are not very important at the micro scale, Dallas emphasized. His research involves one of these forces, called sticktion. “One of the forces is known as sticktion. A macro example of sticktion would be if you have a cold glass of water with condensation on the glass and it is sitting on top of a smooth glass table, the glass will adhere to the surface of the table. If this happens on the micro scale, then parts could be permanently fused together,” he says.
“Sticktion is a big deal in MEMS devices. It comes down to keeping out the pathogens to keep these safe from the environment. A lot of work goes into making sure this is reliable and preserved from dust and humidity, the two main pathogens. Dust and humidity will kill these devices. They are created in clean rooms, or laboratories that are free of pathogens. With the devices being so small, if you just got one little piece of dust between the mirrors, you’d kill many. These devices are actually very durable because the devices are hermetically sealed to protect them from pathogens. As long as the device is protected in its outside cover, you could throw it against the wall and not hurt the micro mirrors. However, if you pull off the outside glass package, you could just wipe all the germs onto the mirrors with one stroke of your finger, and the mirrors would all be gone,” Dallas says.
One major problem that could show up in the fabrication process is friction. “I think one of the things that probably will happen with some of the designs is that minimum feature sizes are going to be a problem and friction is the other problem. Especially with the atomic force microscope,” Beverly says, “a lot of surfaces are touching each other in order for the device to work, and each of those parts has to slide without friction.” Dallas, whose research work also involves friction, offers the example of the Torsioning Ratcheter Actuator, which if not strong enough will not overcome the friction and therefore will not create tension. “Immediately it’s a fatal defect with this and you could never be able to turn it successfully,” Dallas said. “It’s like a glass door that’s jammed. Once you’ve got something stuck in there, the glass door isn’t going to open anymore,” Beverly said.
Some of the fabrication tools that are used to make computer chips and also the idea of building up layer by layer originated with the transistor industry, Dallas said. Modern transistors numbering about 100 million go into a Pentium processor in a computer. “The rest of the process is getting electricity and information into and out of the device. The more transistors you have, the more wiring levels you need. The transistors basically take electrons in and put electrons out. In these MEMS devices, sometimes electrons come in and mechanical motion comes out. With these accelerometers, you move and use external forces to get motion from mechanical systems, and those mechanical systems convert that energy into electrons as a signal that goes to your computer, and for example, in a vehicle signals the air bag to go off. There are magnetic, electric, chemical and biological sensors that are MEMS devices, and there are many different applications for the technology,” Dallas explained.
Huber, the judge for the Sandia competition, noted that the design tools and production facilities needed for a real learning experience in the MEMS field are too expensive for all but the wealthiest schools to provide. The Sandia program brings these facilities within reach of many more schools. “The students respond with some super designs,” he said. Sandia is a multiprogramming laboratory for the U.S. Department of Energy’s National Nuclear Security Administration. Sandia has major research and development responsibilities in national security, energy and environmental technologies, and economic competitiveness.
With its MEMS design competition win, Texas Tech also was awarded membership in the international MEMS organization, MANCEF. Other competing student teams from the University of Oklahoma in Norman, Oklahoma; Albuquerque’s Technical-Vocational Institute in New Mexico, and the University of North Carolina in Charlotte also will have their designs fabricated by Sandia. Institutions must be members of Sandia’s MEMS University Alliance for their students to participate. Members receive course materials structured to help start or further develop their own MEMS program, licenses for Sandia’s MEMS design software and other benefits. The University Alliance members receive MEMS parts to use in their curriculum.
The College of Engineering at Texas Tech offers MEMS courses in the fall and spring semesters and has done so for the past five years through a National Science Foundation grant. “The idea is that when we do get the parts back from Sandia, they will be part of the next assignment for these other classes too, to hook up these devices and test them out. We need to test durability and whether the devices will operate at all. That’s kind of the first question that we brought in is whether anything moves,” Dallas says.
“The clock is the most straightforward designs of these devices because the others will be more difficult to operate,” he notes. “The clock will definitely serve a purpose. Hopefully it will run and keep running and that will offer more demonstrations and testing of that MEMS device. With Sandia fabricating parts, we can enhance the educational experience for subsequent MEMS classes. Students now actually not only get to design some parts, but they also actually get experience actuating previously designed and built products.”
More than 100 process steps are completed by full-time teams of Sandia workers to build the MEMS devices. The process is expensive, and typically, Sandia would charge $10,000 or more for one module. Sandia has at least a $100 million laboratory to accomplish the fabrication of these prototypes, Dallas explains. Part of the Alliance program involves Sandia providing universities with software to plug into the computer-aided drawing programs, as well as with on-site training for faculty advisers.
The Texas Tech research team can expect the fabrication process to take two to three months. In the meantime, Beverly, Dallas and other faculty researchers and their students will be awaiting the news about whether their micro machines will work and are looking forward to the possibility of being in the Guinness Book of World Records. Successfully designing the smallest mechanical clock and the smallest moveable chain is a giant feat in the universe of the small.