Atom-scale Images Give Materials Researchers New Tool For Developing Advanced Ceramics

A paper published in the April 15, 2004, issue of the scientific journal Nature describes research that would represent a valuable advantage in the development of strong and heat-resistant materials for a variety of applications.

The work, by ORNL researchers Stephen Pennycook of the Condensed Matter Sciences Division, Gayle Painter and Paul Becher of the Metals and Ceramics Division and visiting researcher Naoya Shibata, reveals, in world-record 0.7 angstrom resolution, the preferred location of atoms within a silicon nitride ceramic.

Where specific atoms reside is key to the properties of the materials. The atom-scale images match, almost exactly, the positions predicted by theoretical calculations.

"With this new confidence in our theories, we will, in the near future, model materials on a computer screen and predict their properties without having to actually fabricate and characterize a large number of samples, which is very expensive and difficult," Pennycook said.

The images of silicon nitride were made with ORNL's 300-kilovolt Z-contrast scanning transmission electron microscope (STEM), aided by an emerging technology called aberration correction, which uses computer technology to correct errors introduced to the images by imperfections in the electron lenses. Shibata, a fellow of the Japan Society for the Promotion of Science, produced the images, which were then refined with technology provided by Pixon LLC of Setauket, N.Y.

Silicon nitride is of great interest to materials researchers because it is strong and lightweight. However, it is also intrinsically brittle, so researchers are constantly searching for ways to make it tougher and less brittle and thus more suitable for applications that require strong, heat-resistant and light-weight components.

One way to toughen the material is to induce the growth of whisker-like grains that act much like reinforcing rods in concrete. Researchers know how to form the whisker-like grains in the silicon nitride by adding certain rare-earth "doping" agents such as lanthanum oxide. However, slight changes in the doping agents result in variations in the properties of the materials. The ability to predict and manipulate the structure of these materials at the atomic level will aid researchers in developing the ceramic materials with the most desirable properties.

Silicon nitride ceramics, like many ceramic materials, are made by first compressing powders into a desired shape, which still contains a large amount of pores. To eliminate the pores, the material is sintered--essentially baked--at very high temperature, which combined with oxide powders produces a dense ceramic. The resulting silicon nitride ceramic also contains a very thin amorphous, glassy film, which surrounds all the silicon nitride grains. The properties of the ceramic depend on how the doping agents eventually situate themselves in the silicon nitride ceramic. In the past, researchers seeking the best properties have had to try different combinations until they arrive at the best material.

"Rare-earth elements like lanthanum and lutetium have quite different effects," said Becher. "You get different looking microstructures with different properties. Lanthanum will produce long, slender reinforcing grains, while lutetium produces fatter grains. The real question was, why do these elements cause these changes?

"The theoretical calculations led by Painter predicted that these elements had different preferences for locating themselves at the silicon nitride grain surfaces. Those like lanthanum were seen to want to go to the grain surfaces, causing long, thin grains to form. On the other hand, lutetium was predicted to be less likely to locate next to the grain surface, which allows the grains to grow fatter.

"We know that the particular microstructure we can obtain and the nature of the amorphous film strongly affect the properties of the silicon nitride. So the knowing 'the why' is critical to the development of new materials," Becher said.

But determining how accurate the theory was required finding where specific elements like lanthanum resided. Because of the presence of the amorphous films around each silicon nitride grain, "it is very difficult to see these dopant atoms in a microscope," Pennycook said, adding that this was a "good problem" for his world-record holding Z-contrast STEM. Shibata, who arrived last April, proved to be up to the task.

Shibata's Pixon-enhanced images corresponded to the theoretical predictions of ORNL's Painter so closely that Pennycook and Becher, who are both ORNL corporate fellows, believe researchers will, in the future, be able to confidently design optimum materials by computer using advances in theory and the understanding gained by atomic scale analysis, significantly speeding the development of new advanced ceramic materials.

"Now we know, at the atomic level, why things are happening," Becher said. "This will allow researchers to create materials that are much tougher and stronger. And those materials will be found in tomorrow's advanced microturbines and auxiliary power systems for aircraft and trucks."

Co-authors on the paper are William A. Shelton of the Computer Sciences and Mathematics Division and Tim Gosnell of Pixon. The work is sponsored by the DOE Office of Science, Basic Energy Sciences, Division of Materials Sciences and Engineering.

ORNL is currently constructing an advanced materials characterization laboratory that will further the application of aberration-correction technologies to atom-scale microscopy.

###

Oak Ridge National Laboratory is a multiprogram research facility managed by UT-Battelle for the Department of Energy.