Improving the Response of Piezoelectric Materials

Housed within the Center for Dielectric Studies at Penn State, the Center of Excellence in Piezoelectric Materials and Devices focuses on electromechanically active materials development, processing, and prototyping.

Piezoelectricity is a property of certain crystals, ceramic materials, and polymers. A piezoelectric material changes its shape when a voltage is applied, and conversely, produces an electric charge when pressure is applied. Piezoelectric materials are used for the production of sound, as in ultrasonic waves and sonar; as actuators for printers, pumps, and fine tuning of scientific instruments, such as atomic force microscopes; and as sensors, such as accelerometers. Because piezoelectric materials retain their performance at small scales, they have become important components of many electronic devices and microelectromechanical systems (MEMS).

Over the past 20 years, MEMS have become widely used in industry for the miniaturization of devices. In a recent estimate by Yole Développement, a French market research company, the demand for MEMS devices is expected to grow from $6.5B in 2009 to $16B by 2015. Piezoelectrics are also used in medical ultrasound, nondestructive testing, and a variety of sensors, actuators, and resonators. Improving the response, that is, making the mechanical movement larger or the driving voltage lower could provide large benefits in the areas of industry, health, and defense.

Imagine what you could do

Flavio Griggio, Ph.D. candidate in materials science and engineering, with his advisor Prof. Susan Trolier-McKinstry. Griggio studies thin film piezoelectrics to fabricate high frequency array transducers.
Flavio Griggio, Ph.D. candidate in materials science and engineering, with his advisor Prof. Susan Trolier-McKinstry. Griggio studies thin film piezoelectrics to fabricate high frequency array transducers.

If you could make a material in the form of a thin film with a very high piezoelectric or pyroelectric coefficient, then you could imagine doing a number of very interesting things with it. You might be able to create very tiny medical ultrasound devices that could be swallowed in a pill to take snapshots of internal organs. Or develop autonomous energy harvesting sensors that run on the energy of vibrations. You could make a device that combined with a CMOS electric drive made very fast, very low loss switches for radar, handsets, and cell phones. Or make cheap thermal imaging systems to avoid, for instance, deer on a dark road. You could make ultra small vacuum pumps on electronic chips for a handheld mass spectrometer that could test the air for hidden explosives at airports. Or even make a flying robotic insect with flapping wings that could take photos of people planting roadside bombs without being seen.

In the Center of Excellence in Piezoelectric Materials and Devices at Penn State (CPMD), all of those devices are being imagined, developed, and in some cases, prototyped. Crucial to success in any of these applications, is a fundamental understanding of how piezoelectric materials behave when they are reduced from the bulk material to a thin film that may be only a few nanometers to a micrometer in thickness.

The fundamental things apply

“It’s pretty fundamental work,” says Susan Trolier-McKinstry, director of the CPMD and professor of ceramic science and engineering, about trying to improve the piezoelectric response in ceramic thin films. Part of it comes down to what happens when the grain size of the ceramic materials is shrunk. In bulk ceramics, grain sizes are often relatively large, 1-10 microns. With large grain sizes, the domain walls – the boundaries separating regions of similar polarity – are freer to move and contribute to the piezoelectric effect. In thin films, the grain sizes are only 50-200 nm, and the grain boundaries act as potential pinning sites, hindering the mechanical motion.

One way to increase the piezoelectric response in thin films, is to increase the grain size. This can be done by providing a set of well ordered nucleation sites, the points at which the grains begin to grow. Trolier- McKinstry also suspects a secondary effect – a small amount of liquid-phase sintering, which would allow molecules to move more freely – also helps increase grain size.

One of her graduate students, Flavio Griggio, is working on trying to determine just what is taking place when piezoelectric thin films are processed. He recently returned from a three-week visit to Oak Ridge National Lab where he has been mapping the domain wall mobility at a very fine scale, on the order of 50 nm. “We are trying over time to determine which of all the possible pinning sites are the worst ones in terms of properties,” Trolier-McKinstry says. With that understanding it should be possible to optimize the response by eliminating the worst sites.

A transducer array for miniaturized ultrasound.
A transducer array for miniaturized ultrasound.

Improving the figure of merit for energy harvesting devices

Several years ago, Trolier-McKinstry was approached by a company that wanted her to design a sensor for condition-based maintenance, a sort of just-in-time form of monitoring that tells when equipment and structures needs repair or replacement. She recalls designing a smallish sensor for them, about 6 mm on a side, controlled by an integrated circuit. But when her sensor was placed in the final device, the power supply took up most of the space. “You’d really like to move away from where the system volume is governed by the batteries and not by the guts of what you’re trying to do,” she remarks. The solution is energy harvesting, pulling energy out of the environment.

There are many ways to extract energy from the environment: small solar cells can draw power directly from sunlight, while thermoelectric generators can scavenge energy from waste heat, like in the engine of a car. But other places that might use sensors – in homes and office buildings, on the underside of bridges, in a chemical plant to sniff out toxic gases, in many industrial settings – thermal energy and sunlight are not readily available. In these situations, a piezoelectric system of harvesting energy might be the only solution, by taking mechanical energy, such as vibrations, and turning it into an electrical charge.

For most condition-based maintenance, if a sensor could take a reading every couple of hours or once a day, and send out a signal to a receiver, that would meet the demands for monitoring. This system would need the appropriate type of sensor coupled with a low power integrated circuit, a slow trickle charge battery, a wireless communication package, and a material capable of charging the battery over time.

“It’s a hard materials problem,” Trolier-McKinstry admits. “The sorts of materials we use for high piezoelectric coefficients don’t necessarily have high energy harvesting figures of merit, which describes how a material actually behaves in an operating device.” Raising the figure of merit depends on finding the right combination of materials constants.

So far, she and her colleagues in the Center of Excellence have managed to increase the figure of merit for piezoelectric materials in devices by 70 percent relative to their base composition. An alternative materials system has shown some preliminary results, that if confirmed could increase the figure of merit by a factor of four relative to aluminum nitride, the material currently considered the gold standard for piezoelectric energy harvesting. That could be good enough to eliminate the wires or bulky batteries that need to be replaced in current sensing devices.

Printing on (and with) thin films

Researchers in the CPMD are searching for a damage free approach to patterning thin films for MEMS. The materials that are the most interesting as piezoelectric materials are composites, called complex oxides, which typically have at least four different kinds of atoms. The chemistry of these atoms can be easily damaged in the type of patterning processes that take place in the semiconductor industry foundries, which are increasingly being used to produce MEMS devices.

To solve the damage challenge, Trolier-McKinstry’s group is working to develop a method to print structures directly on the thin film with high fidelity. To this end, they are exploring microcontact printing as a means to not only limit the damage caused by etching, but also to avoid a large fraction of the expense involved in the several photolithography and patterning steps necessary to create microstructures in MEMS. Microcontact printing uses a stamp, usually with a raised design in the shape of a device, which is inked with a precursor of the chosen material and printed on a substrate. Microcontact printing has been used with organic materials for more than a decade, but rarely for piezoelectric materials. If this is possible, she believes that industry could put piezoelectric MEMS in systems where it would have been cost prohibitive in the past.

This piezoelectric flextensional actuator for a MEMS device was made using silicon micromanufacturing techniques.
This piezoelectric flextensional actuator for a MEMS device was made using silicon micromanufacturing techniques.

The main industrial use for piezoelectric MEMS at present is in inkjet printing. Traditionally, a bulk piezoelectric called PZT (lead zirconate titanate) was used to control the print head. Recently, led by manufacturing companies in Japan, a transition is underway to thin films. The advantage of using thin films is that smaller PZT actuators allow the spray nozzles to be placed much more closely together. The latest industrial size ink jet printers are capable of printing at the speed of newspaper presses, opening up the possibility of patterning all kinds of consumer products, from books, newspapers and magazines to organic, flexible LED displays to dishware.

Another large piezoelectric MEMS application is frequency filtering for communications. Currently, companies such as TDK, Epcos, and Avago are using aluminum nitride based materials to produce filters for cell phones, PDAs and television. Hundreds of millions of these filters are produced each year. But if piezo thin films could be developed that are powerful enough to make actuators that moved useable distances at low enough voltage, and are compatible with CMOS electronics, a whole new field of CMOS electronics could open up – sometimes referred to as “more than Moore.” First among these are super fast, low loss switches for communications devices and radar.

In addition, one of the long-term goals of the CPMD and of the active materials field in general is to develop a thermal imaging system that is low enough in price that it could be placed on the dashboard of every car to aid with night time vision. In central Pennsylvania, which has one of the highest deer densities in the world, this would provide a huge benefit to the public.

Pumps on chips and flying devices

The Department of Defense and DARPA, its advanced research arm, fund a number of futuristic projects that seem right out of Jules Verne. Trolier-McKinstry has worked on several of those projects involving smart materials and devices. The first of these was through defense contractor Northrop Grumman, who was trying to develop a miniature mass spectrometer for use in threat situations to analyze chemicals and toxins. “Mass spectrometers tend to be desk sized.” she explains.“Northrop had miniaturized everything but the pump into something that was 1 cm x 2.5 cm by ½ mm thick. Pretty small. We worked on the piezoelectric vacuum system they needed, but the project never went to completion. However, it’s still possible to do.”

Center of Excellence in Piezoelectric Materials and Devices

The Army and DARPA have large research efforts in developing small robots for unobtrusive surveillance of the battlefield. One effort involves small flying machines that can carry a payload of a miniature camera and transmitter. The group the CPMD works with at the Army Research Lab has been the leader in this effort. The Army does the majority of the modeling and mechanical design and all of the fabrication, while her group works on optimizing the piezoelectric response for insect-like motion using PZT actuators on the wings. “How do you optimize the response so instead of driving at 10 volts, you’re driving at five volts?” she posits.

Her work with the Army is funded through the Department of Defense’s new National Security Science and Engineering Faculty Fellows Program, of which she is one of six inaugural fellows. The grant of up to $3 million for each fellow will allow Trolier-McKinstry and her group the freedom to pursue fundamental research in areas of crucial importance for national security.

Susan Trolier-McKinstry, Ph.D., professor of ceramic science and engineering in the Department of Materials Science and Engineering, is the director of the W.M. Keck Smart Materials Integration Laboratory and the Center of Excellence in Piezoelectric Materials and Devices. Visit her website at http://www.mri.psu.edu/faculty/stm.

The Center of Excellence in Piezoelectric Materials and Devices is a new center housed within the Center for Dielectric Studies. The center director is Prof. Susan Trolier-McKinstry and the associate director is Prof. Thomas Shrout.

Companies - aixACCT GmbH, Bridge Semiconductor, Ethicon EndoSurgery

Website - http://www.mri.psu.edu/Centers/CDS/profile.asp

Brochure - http://www.mri.psu.edu/centers/cds/Media/Brochures/CPMDBrochure.pdf

This article was featured in Focus on Materials - Fall 2010.