Scientists Show How Large-Scale Order and Small-Scale Disorder Unite in Materials Used in Ultrasound, Sonar

Scientists Show How Large-Scale Order and Small-Scale Disorder Unite in Materials Used in Ultrasound, Sonar
November 01, 2002
PHILADELPHIA – Chemists at the University of Pennsylvania have shown how atom-scale changes in certain materials can dramatically affect their ability to interchange mechanical and electrical signals. This finding, reported in the Oct. 31 issue of Nature, may permit scientists to engineer these materials to be more sensitive, durable and environmentally friendly.

Crystals of these materials, known as piezoelectrics, are already used to amplify mechanical signals – generally sound waves – in applications ranging from microphones to naval sonar to medical ultrasound devices. Car manufacturers are interested in piezoelectrics' potential to tune engine performance according to minute-by-minute driving conditions. New piezoelectric materials could boost the performance of these and other systems that exchange electrical and mechanical energy.

"With piezoelectrics, you essentially have atomic-level disorder embedded in an ordered lattice," said principal researcher Andrew M. Rappe, associate professor of chemistry at Penn. "We have now shown the effects of this disorder on materials' piezoelectric properties, which may let us harness it to optimize these properties."

The work may also help scientists design piezoelectric materials free of lead, which is found in all current-generation piezoelectrics. This would yield environmental benefits while rendering the materials lighter in weight.

Since all known high-performance piezoelectrics are structurally and chemically complex, the search for better piezoelectrics has often relied on trial-and-error synthesis and testing. Rappe and co-workers developed computer models of the relationship between atomic structure and piezoelectric properties, providing guidance for the systematic design of new piezoelectrics.

"Many technologically important materials are complex, and this complexity is often essential to producing technologically valuable behavior," said Rutgers University physicist Karin M. Rabe, an expert in the computational modeling of oxide materials who was not involved in this work. "Dr. Rappe's work may suggest new ways of designing piezoelectric materials with improved properties controlled by local atomic arrangements."

Rappe and his colleagues looked at the structure of a piezoelectric material called PZT (Pb(Zr,Ti)O3), which is an ordered lattice of lead and oxygen inhabited by randomly arrayed atoms of titanium and zirconium. Rappe's team showed that distortions of the material at the atomic level – distortions that grant a material its piezoelectric properties – can be predicted by the arrangement of titanium and zirconium atoms.

"A simple statistical interplay of well-known chemical interactions obtained through first-principles calculations, coupled with disorder, can give rise to rich local structure and to complex phase behavior in a solid solution," Rappe and his co-authors, postdoctoral researcher Ilya Grinberg and graduate student Valentino R. Cooper, write in Nature.

PZT exhibits six distinct phases as its titanium/zirconium mix ranges from all titanium to all zirconium; Rappe's work helps explain why piezoelectric properties are most potent when the material is near a phase transition, a situation often seen with a near 50/50 split between the two metals. This state best allows PZT to maintain a variable polarization that can change instantaneously with the application of mechanical energy.

Rappe is part of a team of piezoelectric researchers, based in Penn's Laboratory for Research on the Structure of Matter, that includes Dawn A. Bonnell, I-Wei Chen, Peter K. Davies and Takeshi Egami, all professors of material science and engineering; Jay Kikkawa, assistant professor of physics; and John M. Vohs, professor of chemical and biomolecular engineering. The work described in Nature was supported by the Office of Naval Research and the National Science Foundation, with computational support from the Center for Piezoelectric Design and the U.S. Army Engineer Research and Development Center.