Figure 1. As part of his work with fiber optic sensors in smart structures, William Spillman worked with Dryver Huston at the Univ. of Vermont to develop this vibrating plate test of a distributed fiber optic sensor network.
When Deep Space I left Earth's atmosphere in 1998 and headed for a rendezvous with the asteroid 9969 Braille, William B. Spillman, Jr. of BF Goodrich (now associate director of the Carilion Optical Sciences and Engineering Research Center at Virginia Polytechnic Institute) had already developed his definition of a smart structure. It spanned many disciplines and technology sectors, and -- by Spillman's definition -- the deep space probe met the criteria.
"Deep Space 1 had a control system they called a remote agent," Spillman said. "The spacecraft autonomously navigated (monitored its environment) to fulfill its mission. It was presented with simulated failures of its subsystems, for which it discovered and developed workarounds (monitored its own state and acted to optimize itself). And of course, it did this 125 million miles from Earth. Pretty neat."
Spillman's group came up with the novel definition of smart structures in 1996, and it indicated a shift away from the fiber-optic-sensor-dominated view of smart structures. According to Spillman's group, a smart structure is "a nonbiological physical structure that has: (1) a definite purpose, (2) means and imperative to achieve that purpose, (3) a biological pattern of functioning."
Imparting biological functionality to both micro and macro systems has long been the goal of researchers in many fields, from neural networks to autonomous vehicles to smart structures. To achieve this kind of functionality, the field of smart structures is expanding to include input from material, actuator, and system engineers, as well as semiconductor manufacturers and chemists.
Figure 2. The benefits of vibration control on a structure can be studied and analyzed on a test platform, such as this one in the smart structures lab at Sandia.
When the term "smart structures" was originally coined by Eric Udd of McDonnell Douglas (now president of Blue Road Research) in the late 1980s, most development concentrated on fiber optic sensors embedded in civil structures. Today, companies like Blue Road Research, FISO, and Smartec, along with learning institutions around the world, continue trying to bring fiber optic strain and pressure sensors to the commercial arena. Even as examples of embedded fiber optic sensors gain acceptance -- for example, they were used in the construction of the world's largest dam on China's Yangtze River -- active development in embedded fiber optic sensors is taking a back seat to new research in monolithic smart structure systems that not only sense their environment, but react to it through integrated actuation.
"At a recent meeting, one interesting thing that was said by an organizer associated with the (U.S.) government was that enough work had been done in the area of fiber sensors," said Richard Claus, director of the Fiber Electro-Optics Research Center at Virgina Polytechnic Institute. "They didn't say that fiber sensors were done, but that they had been developed.
"A lot of groups, including Virginia Tech," Claus continued, "have spent a lot of time and money on fiber sensors -- how they can be used, multiplexing, and other issues, and the consensus was that additional work in that area was probably not going to lead to order-of-magnitude improvements in performance. So, (the organizer) said the smart structures community might be better served if it turned to more significant problems in the actuator and integration area -- making things smaller, more efficient, etc. So, not necessarily because of that [statement] -- but maybe in resonance with it -- our group has spent a lot of time moving away from fiber sensors."
Figure 3. Finite Element codes and active control can be used to predict and eliminate vibration, such as that shown for the upper mirror of an x-ray lithography machine.
This growing area of smart material multifunctional sensor and actuator development is generally composed of four main elements: the sensor, actuator, control electronics, and an attached structure, which supports the device or provides a structure against which the device can apply force. For Claus, the changing development goals mean shifting away from the "Frankenstein" approach illustrated by the fiber optic sensor systems, in which sensors and actuators were either not present or separate components in a single system, and moving toward a completely integrated approach that more closely mimics biological structures.
"Compare a bridge to a hand," Claus said. "A hand is very integrated: it has sensors, actuators, a protective, permeable shell. It can sweat to cool itself and regulate temperature. It's a very smart structure. That bridge is a long way from your hand." Development directions
This kind of complete integration is not easy to achieve. Researchers are developing each of the four smart structures elements, but scientist Marc Regelbrugge at Rhombus Consultants Group said the different groups are often quite detached from each other.
Figure 4. A researcher works on a commercial lithography machine, which is a microcircuit manufacturing device that can benefit from a Sandia-designed system that minimizes vibration.
"I'm a mechanical engineer, so I approach the subject from mechanical actuation, but others approach it from control, materials, or sensing," Regelbrugge said. "One of the things we're seeing is that there are several groups that are starting to focus more specifically on how all the elements get integrated. That's a very difficult problem and it turns out to be very specific to system development." Like many optical applications, system development of smart structures tends to rely on a single application per project. Although larger lessons will come from such work, further efforts to meld the four elements into a modular and single design can only help this emerging market.
Regelbrugge concentrates on increasing the output force and power densities of actuation materials designs. These macroscale devices measure between 10 cm3 and 100 cm3, using piezoelectric, metal, ceramic, and polymers. "We're looking to develop actuators that provide higher motion, forces, and power densities than the actuators currently available," Regelbrugge said. "Our targets in those ranges are motors or devices that are driven by solid state phenomenon -- things like pumps... Most prototypes we're working on target precision motion of components. Some are associated with optical systems either terrestrial or space based, and other kinds of precision positioning devices."
Some people believe that smart, precision positioners, which can also take the form of vibrational dampers by counteracting high-frequency movement, may eventually become a large market for smart structures. Several universities and the Sandia National Lab. have spent several years working on precision positioners, including magnetically levitated platinums for weapon component protection and vibrational control, and dampeners to allow faster stage movement within a microlithography machine that makes microchips.
Spillman said other researchers are working on smart polymers such as carbon nanotube, PVDF, and other electroactive polymers. "The polymers are used in the same ways human muscles are," Spillman said. "They have little force but big displacement, and if you can combine them in the right way you have something like a muscle."
Implantable devices are another area receiving considerable attention. According to Spillman, many applications found in hospitals, health care facilities, and the military could use small implantable devices capable of monitoring various bodily functions. Implanted sensors could have their data read noninvasively by external magnetic induction devices analogous to the "tricorders" of the TV series Star Trek. This technology has already been demonstrated in composite structures. The implanted sensors could also be used for the control and injection of pharmaceuticals, such as insulin for diabetics. Claus added that smart structures capable of detecting various diseases or conditions could be placed directly on the skin. A change in color of the device could indicate that the patient is suffering from a particular condition, greatly reducing the costs of diagnostics and testing.
Smart structures may one day include completely chemical structures. In one sense, some proteins are smart structures: they have a purpose (selective bonding to other molecules), chemical bonds that provide a way to achieve that purpose, and they certainly have a "biological pattern of functioning."
What will happen to smart structures in the future? Claus believes economics will ultimately determine whether or not the field of smart structures expands to include biochemistry. "If people find that drug delivery through smart implants are more economical, and that area takes off more than vibration controlthen that's what will play," he said.
"There was a prediction a long time ago at a VT meeting held by DuPont that said the real first applications of smart structures would be in toys, and I think that's probably true," Claus continued. "Think about the transformer toys that change color when they're put in hot water or cold water. Kids really buy them."
Richard O. Claus
Fiber Electro-Optics Research Ctr.
Virginia Polytechnic Institute
Phone: (1) 540/231-7203
Rhombus Consultants Group, Inc.
Mountain View, CA
Phone: (1) 650/691-1142
FISO Technologies Inc.
Phone : (1) 418/688-8065
Blue Road Research
Phone: (1) 503/667-7772
Phone: 41 91 993 09 24
William B. Spillman
Director of Research
Optical Sciences and Eng. Research Ctr.
Virginia Polytechnic Institute
Phone: (1) 540/231-1891
R. Winn Hardin
R. Winn Hardin is a science and technology writer based in Jacksonville, FL.