ENGINES – Miniature maximization . . .
Improving efficiency and performance of tiny engines like those used in remote-controlled planes is the focus of a report that may thrust the technology into this century. These single-cylinder engines, which have potential for unmanned aerial systems and portable power, can be improved by using stronger, more heat-resistant materials and tighter tolerances between the piston wall gap, according to a study led by Mike Kass of Oak Ridge National Laboratory. "Miniature engine development hasn't advanced much beyond the designs that were introduced in the 1950s," said Kass, who noted that profit margins are too low to support the tools necessary to advance the technology. While typically relegated to the hobby industry, these engines can also help meet power needs for in-flight sensors and future hybrid versions that would offer greater flexibility. The report will appear in SAE International. [Contact: Ron Walli, (865) 576-0226; email@example.com]
MATERIALS – Transient doping success . . .
Using a scanning tunneling microscope tip, scientists at the University of Rostock, Germany, and Oak Ridge National Laboratory have accomplished something thought to be virtually impossible. "We devised an approach to change electronic properties of one-dimensional systems without introducing dopant atoms that effectively cut the chain into smaller pieces," said Paul Snijders of ORNL's Materials Science and Technology Division. This discovery, known as transient doping and detailed in a paper published in Physical Review Letters, was accomplished using only electrons. That makes it especially intriguing to co-author Snijders and colleagues who note that the doping of materials, which changes the number of electrons, has led to significant – and often unexpected -- discoveries of colossal magnetoresistance and unconventional superconductivity. The paper is available at http://link.aps.org/doi/10.1103/PhysRevLett.111.156801 [Contact: Ron Walli, (865) 576-0226; firstname.lastname@example.org]
BIG DATA – Powerful pattern discovery . . .
Analyzing vast amounts of information to identify suspicious activities could become easier with EAGLE, a technology developed by a team led by Oak Ridge National Laboratory's Rangan Sukumar. The software that scales on Cray's Urika allows users to retrieve and visualize massive relationship-graphs, simplifying the process of recognizing patterns that are sometimes difficult to find by just querying data. For example, the team studied a doctor referral database of connections between providers who referred patients to another during a one-year period. "Our algorithms discover situations such as 80 percent of patients who visit Doctor A also visit Clinic B, which is owned by a friend of Doctor A, that may warrant an investigation," Sukumar said. Other potential relationship-analytics applications for EAGLE include social network analysis, fraud detection and forensics, semantic-integration for drug design and clinical pathway analysis. [Contact: Ron Walli, (865) 576-0226; email@example.com]
VEHICLES – ORNL inverter a boost for EVs . . .
Less expensive, lighter and more efficient inverters could put hybrid electric vehicles on the highway to improved viability. While batteries receive a lot of attention, Oak Ridge National Laboratory inventor Gui-Jia Su noted that inverters, which convert direct current into alternating current, play an equally important role in powering hybrid electric vehicles. The patent-pending ORNL inverter is more compact, reduces battery losses, improves operating conditions and reliability, and can be operated in high-temperature conditions. The inverter also significantly reduces undesirable motor torque ripples, which increase or decrease output torque as the output shaft rotates. In addition to uses in hybrid electric vehicles, Su envisions the inverter being used in industrial motor drives. [Contact: Ron Walli, (865) 576-0226; firstname.lastname@example.org]
AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert! system.