INQUIRY 2001

Researchers at Ames Laboratory’s Scalable Computing Lab have extended their investigation into communication technology for cluster computers thanks to a National Science Foundation Major Research Instrument grant awarded to Iowa State University’s Center for Physical and Computational Mathematics. The $300,000 MRI grant includes $190,000 from NSF and more than $100,000 in matching funds from ISU’s Institute for Physical Research and Technology, Iowa State University and the ISU departments of physics and chemistry.

Scalable Computing Lab researchers are using the MRI funds to improve interconnect solutions for cluster computers, personal computer or workstation networks that can operate at speeds comparable to today’s commercial parallel computers, but for a fraction of the cost. Making message-passing between computers in a cluster faster and more efficient is the primary goal of the MRI-supported research effort.

The SCL scientists believe cluster computers could be very appealing to the larger scientific and academic communities. "A university department or a research group can’t afford to buy a supercomputer, but they can afford to put a good cluster together," says Mark Gordon, director of Ames Lab’s Applied Mathematics and Computational Sciences Program and head of the SCL.

Gordon emphasizes that solutions to grand-challenge problems, such as the design of new materials and catalysts, the development of viable methods for environmental remediation, and the search for the origin of life will depend on state-of-the-art computational hardware and applications software to take advantage of modern computers. "The high-performance computing environment of the future will undoubtedly include scalable cluster computing," he says. (Scalable means the ability to increase, or "scale up," computer processing power to run the same job in less time.)

"In the case of clusters, we want to figure out how to best manage the hardware and get computers talking to each other with a minimum amount of time used in communicating and a maximum amount of time in actually doing the calculations," Gordon adds. "That’s what the MRI grant is all about."

When researchers at Ames Laboratory developed a compact laboratory furnace, it marked a huge leap forward in the ability to understand what happens to a material’s crystal structure as the material is heated and cooled. That success has led to the building of two additional furnaces for other DOE laboratories and a refined design intended to make the furnace easier to operate.

Used in conjunction with a powerful X-ray beam, such as that produced by the synchrotron at Argonne National Laboratory’s Advanced Photon Source, the furnace allows researchers to capture in a few seconds the patterns created as the crystal structure of the material diffracts the beam. By studying changes in the diffraction patterns as the material is heated and cooled in the furnace, scientists quickly gain an understanding of what happens when a material transitions from one type of solid to another or from solid to liquid and back again.

The original furnace is being used at the sector of the APS facility operated by the Midwest Universities Collaborative Access Team. According to scientist Matt Kramer, one of the developers of the furnace, two copies of that furnace are being built by Ames Lab’s Engineering Services Group.

"One is being built for Systems Research and Instrumentation, which is the developmental CAT for Argonne National Laboratory," says Kramer. "The other one is for the High Temperature Materials Research Laboratory at Oak Ridge National Laboratory."

The furnace’s fame isn’t limited to this country. Kramer’s group designed and built an even smaller unit for the European Synchrotron Radiation Facility, located in Grenoble, France. That collaborative effort was aided by ESRF researcher Larry Margulies, a former Ames Lab graduate student who worked with Kramer, senior materials scientist Bill McCallum, and assistant metallurgist Kevin Dennis on the original furnace design.

Even with the furnace’s success, Kramer’s group is looking to improve upon it by making it quicker to set up and easier to use.

"With the present furnace, it takes a substantial amount of time, up to 36 hours, to get things properly lined up," he says. "The new design we’re working on will be more modular and make it easier to position the sample in relationship to the beam, so we can spend more time doing research instead of setting things up. Also, if we need to swap out components, we don’t have to go through the entire alignment process again."

A material that rivals industrial diamond in hardness continues to amaze the researchers who developed it and attract interest from a variety of industrial sectors. The material represents a breakthrough technology that could have a substantial impact on the machining industry, which spends $300 billion each year in labor and overhead in the United States alone.

But hardness is only one of several unique properties possessed by the boron-aluminum-magnesium alloy, nicknamed "BAM." Preliminary tests show it stands up well in cutting both concrete and stainless steel, giving it a definite advantage over diamond-coated cutting tools in slicing up steel-reinforced concrete.

"Diamond does a good job of cutting concrete and masonry," says researcher Bruce Cook, "but wears quickly when cutting steel due to the chemical reaction between the carbon in the diamond and the iron in the workpiece. Our alloy seems to cut both concrete and steel equally well without much wear."

Though Cook and fellow researchers Alan Russell and Joel Harringa expected BAM to perform well as a cutting tool, they were surprised to find that the material cuts without getting hot. Even when the cutting edge is spinning off red-hot ribbons of lathe-turned stainless steel (somewhere around 700 degrees Celsius, or 1,292 degrees Fahrenheit), the tool stays cool enough to touch with a bare fingertip.

"There’s very little heat transport, and we think that’s due to fine grain size and the complex crystal structure within the material," says Cook. Bruce Harmon, Ames Lab deputy director and director of the Lab’s Condensed Matter Physics Program, and graduate student Younben Lee are also conducting further calculations on the fundamental electronic structure. "We’re interested to see if there’s a relationship between the electrical transport and thermal transport that may help explain how ultrahard materials perform," Cook adds.

Cook’s research team has also boosted the scale of production of the material, in part to keep up with requests for samples. Button-sized samples have been replaced by much larger wafers, nearly 2 inches in diameter and about three-quarters of an inch thick. But given BAM’s low density (about 2.5 grams/cm3), the wafers lack the expected heft of conventional tungsten carbide used to slice through concrete and steel.

Since discovery of the material was first announced in October 1999, more than 100 companies and research facilities have requested samples.

While in-house testing continues, including recent successful trials in machining titanium, Cook hopes to learn as much or more from the various companies that have received samples. For example, a manufacturer of tooling for industrial woodworking equipment returned a sample that had been shaped using electric spark erosion. And a producer of industrial gases and surface coatings is investigating the material as an extremely durable coating for various industrial applications.

"They may not realize it, but these companies are providing us with valuable information," Cook says. "It gives us a much broader perspective. At the same time, most of the research and development is being conducted by scientific staff so they can supply test data to support the various applications. It really helps our understanding of the material."

Ames Laboratory and Iowa State University’s Institute for Physical Research and Technology hosted the second meeting of the Midwest Forensics Resource Center in May. Just one year earlier representatives from the two organizations had met with Iowa crime lab officials to discuss the types of research, services and training that would be most helpful to their facilities and to suggest funding sources for the proposed center.

This year’s meeting reflected how much the effort to establish the MFRC had grown. Attendees represented eight Midwestern state crime labs; three universities; the Federal Bureau of Investigation; the National Institute of Justice; the Department of Energy; and the Bureau of Alcohol, Tobacco and Firearms.

The MFRC will be located at Ames Laboratory and serve as a central point for regional training, education and research in forensic science. It will draw on the expertise of faculty and staff members at Iowa State University, the Institute for Physical Research and Technology, Ames Laboratory and other professionals throughout the region.

To carry out its mission of helping regional crime labs develop more efficient and reliable methods of analyzing crime-scene evidence, the MFRC will focus on four major goals:

• Develop customized training for regional partners.

• Conduct short-term, case-related forensics research projects.

• Develop a forensics curriculum for ISU and associated universities.

• Conduct long-term research into new and improved forensics techniques.