Layered Film That Stacks Up

Using pulsed laser deposition, ORNL researchers have grown perfect ferroelectric superlattices with surprising properties.

As recently as two years ago, most researchers seeking to build metal-oxide thin films with atomically abrupt interfaces were convinced that molecular beam epitaxy was the only technique that could provide the required control.

Model of an artificial oxide solid built from lanthanum manganite (LaMnO3) and strontium titanate (SrTiO3) building blocks. Synthesis as well as computer simulations and measurements of the electronic structure are being performed at ORNL. Visualization by Leon Petit.

Pulsed laser deposition—using laser light to vaporize target material, which is deposited on a substrate—was not considered suitable for this task. But MBE, which typically operates at much lower oxygen pressures than pulsed laser deposition, presents challenges when applied to insulating oxides because of defects formed that muddle attempts to characterize accurately the film's electronic properties.

In 2004, Ho Nyung Lee surprised his fellow researchers in the Thin Film and Nanostructured Materials Physics Group in ORNL's Condensed Matter Sciences Division. "Ho Nyung Lee used pulsed laser deposition to fabricate perovskite nanostructures with a degree of perfection we had not seen before," says Hans Christen. Unlike MBE, pulsed laser deposition forms these oxides without the need for separate heat treatment in oxygen after the film is grown.

So he wouldn't be shooting in the dark, Lee used reflection high-energy electron diffraction during film growth. "With the diffraction as his flashlight, he realized the importance of slowing down the process to precisely control the amount of material deposited for each crystalline unit of the film," Christen says. "Now, he can stack atomically smooth layers of different composition to create a perfect ferroelectric superlattice, with hundreds of individually controlled layers that together are about 200 nanometers thick."

Ferroelectric materials store electronic charge at their surfaces because of the asymmetric displacements of ions within their crystalline structure. The displaced ions give each layer positive and negative sides.

A ferroelectric lattice.

This polarization can be "switched" by applying a voltage across a ferroelectric crystal. If electrodes are applied to two sides of a crystal, a current pulse flows during such polarization reversal, making these materials potentially useful in data storage devices.

Lee, working with Christen, Matt Chisholm, Chris Rouleau, and Doug Lowndes, synthesized and characterized "asymmetric three-component ferroelectric superlattices" with a "strong polarization enhancement" that was the subject of a letter published in the January 27, 2005, issue of Nature magazine.

The superlattices described by the ORNL researchers in the Nature letter consist of dozens of repetitions of barium titanate (BaTiO₃), strontium titanate (SrTiO₃), and calcium titanate (CaTiO₃), stacked with atomic precision on top of conducting, perfectly flat strontium ruthenate (SrRuO₃) layers. Each repetition is 3 to 10 nm thick.

Barium titanate, accounting for only a fraction of the total material in the superlattice, is the only ferroelectric compound among the constituents. Yet, partly because "strain" is maintained—that is, the crystalline film is forced to grow in alignment with the SrTiO₃ substrate—the ORNL superlattice has 50% greater polarization than similarly grown pure BaTiO₃.

ORNL researchers can deposit thin films of a complex oxide simultaneously onto 11 substrates, each heated to a different temperature.

This feature—for the first time—confirms theoretical predictions made by two theorists at Rutgers University. Although containing excellent ferroelectric properties, the ORNL superlattice is not the best available ferroelectric material for applications such as memory devices.

"The purposes of this study were to prove we could make a perfect ferroelectric film using pulsed laser deposition, to determine the mechanisms on an atomic scale that influence the material's properties, and to learn how to control and modify the material's properties," Christen says. "Such information could allow us to engineer future ferroelectric films for specific applications and would ultimately help us understand atomic-scale mechanisms in other oxide nanostructures."

The critical issue is the material's behavior directly at each interface between layers. "Our data revealed that the specific interface structure and local asymmetries played an unexpected role in the polarization enhancement," Christen says.

Researchers are now ready to go beyond tailoring ionic displacements at interfaces to changing the electronic configuration in materials such as lanthanum manganite. Aided by excellent tools and strong collaborations, the big picture is coming into focus, helping ORNL researchers improve the behavior of their thin films.

Web site provided by Oak Ridge National Laboratory's Communications and External Relations
ORNL is a multi-program research and development facility managed by UT-Battelle for the US Department of Energy
[ORNL Home] [Communications] [Privacy and Security Disclaimer]