New materials with high melting temperatures and excellent chemical stability are required for the next generation of jet engines to burn hotter, quieter, faster, and more efficiently.
Lightweight nickel and aluminum alloys seem to be the likely candidates, but creating the right mixture of metals in the lab and then testing the exact properties of that combination can be expensive and time-consuming—so materials researchers have turned to supercomputers.
Of all the possible combinations of metal "species" you might bring together for a given application, says Long-Qing Chen, "computer models could help you narrow that down to two or three that you might want to try experimentally." Chen, professor of materials science and engineering at Penn State, works on a microscopic scale, his computer models crunching complex equations to generate images of crystals embedded within crystals—M.C. Escher-like designs that underlie both the elegance and practicality of modeling in materials science.
"The beautiful images that come from my models are similar to the images you see under a microscope when you're doing an actual experiment," says Chen.
The smaller geometric shapes revealed in Chen's images, locked in a matrix like a puzzle, are known as microstructures: solid precipitates that develop in both metals and ceramics. "Everyone is interested in microstructures, because that's what controls the properties of materials."
In the case of alloys, when a certain amount of one metal is added to another metal, a solid precipitates out, like salt crystals precipitating in a glass of salt-rich water, except that the solid forms a microstructure within another solid. These embedded microstructures can serve as barriers that prevent the metal from cracking or breaking along lines of weakness.
"If you add more of one type of species, it might change the size or the spacing of the microstructure," Chen explains, which could affect the strength, ductility, or thermodynamic properties of the material. Chen and his research team, working as part of NASA's Ultra-Efficient Engine Technology program, are using their models to predict the chemical and mechanical properties of single-crystal nickel-based superalloys: metal mixtures that are able to retain their strength even after being exposed to temperatures higher than 1,200 degrees F over long periods of time. Two major goals of the program are to develop efficient propulsion technologies to reduce fuel consumption by up to 15 percent, with an accompanying drop in CO2 emissions; and to select and configure materials for engines that can slash nitrogen oxide emissions generated during aircraft landing and take-off.
"Chen's contributions in many respects change the ways we address these problems today and pave a road to a massive use of modern supercomputing in materials characterization and design," says Armen Khachaturyan, a leading computational materials scientist who chairs Rutgers University's department of ceramic and materials engineering.
Long-Qing Chen, Ph.D., is professor of materials science and engineering, College of Earth and Mineral Sciences, 102 Steidle Bldg., University Park, PA 16802; 814-863-8101; lqc3@psu.edu. His research is funded by NASA, the National Science Foundation, and Alcoa.