Electric Crystals That Really Swing

The speakers and microphones in modern telephones depend on tiny crystals that change electricity into sound and vice-versa. A computer model of these crystal's molecular structure, reported in the current Nature, may one day allow scientists to custom design new crystals for supersensitive ultrasound devices that could eliminate the need for some kinds of exploratory surgery.

When a voltage is applied to a so-called piezoelectric crystal, atoms in the crystal lattice stretch, like an accordion, along the direction of the applied electric field. Also like an accordion, the crystal squeaks as it stretches--which is part of what makes them so useful. The reverse also happens: A sound wave can bend the crystal, generating an electric current. A crystal's flexibility is proportional to the volume of sound it can produce or detect.

Because scientists don't completely understand the piezoelectric response, new crystals often are discovered serendipitously. "They have to cook and look," says Ronald Cohen of the Geophysical Laboratory of the Carnegie Institution of Washington. Recently, a team concocted a winner: a batch of two lead-based crystals called PZN-PT and PMN-PT that can stretch about 20 times as much as other crystals. While these champion crystals will certainly boost the sensitivity of piezoelectric devices, scientists were perplexed by how the lead crystals worked so well.

Now they may have the beginnings of an answer. Cohen and collaborator Huaxiang Fu have developed a computer model that explains the large piezoelectric response of the lead crystals. Depending on its structure, a crystal stretches along one of two possible directions, or polarizations. As for PZN-PT and PMN-PT, their polarizations expand by very different amounts: one by 0.1% and the other by almost 2%. Cohen and Fu's new model shows that increasing the strength of the electric field can bump the polarization from one state to the other, and it's this switch between polarizations that dramatically increases the piezoelectric response.

There is probably more to the story than that, says David Vanderbilt, a physicist at Rutgers University in Piscataway, New Jersey. He points out that in real life, crystals are more disordered than Cohen and Fu assume in their model; this disorder may somehow contribute to how well the crystals work in the lab. "There is a lot of ferment in this field," Vanderbilt says, "it may take 2 or 3 years to completely understand the large response." Part of the excitement is that such flexible piezoelectric crystals would make more sensitive detectors and could improve the resolution of medical devices that use ultrasound.