TECHNOLOGY TRANSFER This article also appears in the Oak Ridge National Laboratory Review (Vol. 26, No. 1), a quarterly research and development magazine. If you'd like more information about the research discussed in the article or about the Review, or if you have any helpful comments, drop us a line. Thanks for reading the Review. SUPERCOMPUTING CENTER DEDICATED AS PART OF CRADA WITH INTEL In the political community and now in the academic community, there can be no doubt that Tennessee has a seat at the table among the leaders of the world." That's how Billy Stair, senior policy advisor to Tennessee Governor Ned McWherter, characterized the importance of ORNL's Center for Computational Sciences (CCS) at its November 10, 1992, dedication ceremony. Ed Masi, president of Intel Corporation, echoed Stair's optimism, declaring "While supercomputing is advancing in cycles of two to two-and-a-half years, there is a chance to dazzle the world and to provide Tennessee with a unique opportunity to become a global player in supercomputer technology." The CCS is one of only two Department of Energy high-performance computing research centers dedicated to exploring applications of state-of-the-art computer systems to areas of scientific, economic, and environmental importance. The other center is at Los Alamos National Laboratory. DOE's latest efforts in supercomputing come in response to the presidential initiative on high-performance computing, which is the result of the High-Performance Computing Act of 1991, cosponsored by Albert Gore, former Tennessee senator and now vice president of the United States. The dedication of the CCS was accompanied by the long-awaited startup of the Paragon XP/S supercomputer, custom-designed for ORNL by the Intel Corporation of Beaverton, Oregon. Installation of the Paragon is the latest phase of a three-year cooperative research and development agreement (CRADA) between ORNL and Intel in support of the CCS. Researchers will use the Paragon to build detailed models of the world's climate, predict movement of hazardous waste in groundwater, and design state-of-the-art metals and ceramics on the molecular level. Later, ORNL scientists will collaborate with university researchers to investigate other complex scientific problems called "grand challenges," such as mapping the human genome and superconductor modeling. Many of these problems require manipulating huge amounts of data--so many data, in fact, that they simply couldn't be addressed in sufficient detail with less powerful computers. The Paragon meets this avalanche of data head-on with a concept called "massively parallel processing." In other words, instead of routing all of the data through a single processor, the Paragon divides its work among 2048 smaller processors. The analytical power supplied by this computational juggernaut is equal to that of 15,000 typical desktop workstations, enabling the Paragon to add, subtract, multiply, or divide 150 billion times every second, making it one of the fastest computers in the world. The center will be the intellectual home for a collaborative effort among three DOE facilities--ORNL, Ames Laboratory, and Brookhaven National Laboratory--and seven major universities, including UT, Vanderbilt University, Rice University, State University of New York at Stony Brook, Texas A&M University, and the University of South Carolina. ORNL researchers will also work with Sandia National Laboratories on supercomputer-based mathematics and science education programs. To ensure that collaborators across the country have access to the Paragon, it will also be connected to the proposed National Research and Education Network, a federal computer network linking high-performance computing resources nationwide. --Jim Pearce NEW OAK RIDGE CRADAS FOR SDI OPTICAL SYSTEMS The high-tech weaponry of the nation's Strategic Defense Initiative Organization (SDIO) owes much of its accuracy to optical systems that use light to locate, track, and intercept targets. In 1988, the SDIO asked ORNL to help private industry find the best ways to manufacture high-precision optical components. The result was the Optics Manufacturing Operations Development and Integration Laboratory (MODIL), an interdisciplinary project involving ORNL, the Oak Ridge Y-12 Plant, and the Oak Ridge K-25 Site. It enables private companies to keep up with the latest in manufacturing techniques to meet stringent SDIO requirements and deadlines for delivering components. Because of its mission to work with industry, the Optics MODIL (through Martin Marietta Energy Systems, Inc.) has entered into several CRADAs with industrial partners. In space-bound surveillance systems, high-precision mirrors track enemy missiles and reflect the image to detectors. The detectors, in turn, signal interceptor systems to fire an optically guided missile to destroy the target missile. Baffles within the optical systems act as light traps, absorbing stray light so that false readings are minimized; the detector "sees" only the light coming from its target. Scientists and technicians at the Optics MODIL are now working with Martin Marietta Missile Systems, based in Orlando, Florida, through a CRADA to develop quicker, more efficient, and less costly methods to make better baffles and mirrors from beryllium, a commercially available metal. Optical baffles must be lightweight to help minimize launch costs, sturdy enough to endure the stress of lift-off, and resistant to flaking when handled to maintain their surface texture. "Surface features, one of the most important aspects of baffles, influence optical performance and fragility," says Roland Seals, a project manager at the Optics MODIL. Baffles, with their porous surface texture, keep unwanted light from bouncing or scattering onto the mirror and into the detector of the optical system. The scattered light should be evenly dispersed, with low but equal amounts of energy in all directions. "Surface texture of baffle components influences the amount and distribution of light scatter within the optical system," Seals said. "We are optimizing the coating and texturing process to get a sturdy material that still has excellent optical characteristics." Improving the processes for manufacturing beryllium mirrors is another goal of the Optics MODIL group. The scientists hope to eliminate some time-consuming steps while increasing quality. To make the mirrors, technicians first machine the unit to a precise shape having the exact amount of curvature needed to properly reflect and guide light through the optical system to detectors. After machining, a reflective coating is applied in a vacuum chamber through a process called sputtering. An ion beam bombards a piece of beryllium, called a target, knocking off the outer layer of atoms. These atoms are deposited onto the body of the mirror. Diamond-tipped tools are then used to machine the finish to a precise smoothness. "We are trying to eliminate the need for the mirror-polishing phase for optical systems required by SDIO systems," Seals said. "The SDIO program has strict deadlines, and the polishing process is time consuming and very difficult to predict. The sputtering and machining processes we are working on will help private contractors meet those deadlines." The precise methods developed and tested at the Optics MODIL will also increase manufacturers' ability to repeat the processes the same way every time. According to Seals, this repeatability, which has proved to be nearly impossible for private manufacturers, is crucial to the SDIO program because thousands of optical components are used. --Wayne Scarbrough SECOND OPTICS MODEL CRADA WITH UTOS Energy Systems has also signed a CRADA with United Technologies Optical Systems (UTOS) to determine the best procedures for making high-precision mirrors from silicon carbide, a widely used industrial ceramic compound. The mirrors will be prototypes of those to be tested for use on the high-tech weaponry of SDI. SDI tracking and surveillance systems use light to detect enemy launch sites on the earth and to focus on warheads as they arc through space. Meticulously machined mirrors, formed to have an exact curvature, are situated within the SDI optical systems to guide the light to detectors that signal interceptor systems to destroy enemy targets. The joint work on the new mirrors is being performed by the Optics MODIL. This is the second CRADA that Energy Systems has signed with UTOS involving the Optics MODIL. Keith Kahl, an ORNL researcher and a project manager at the Optics MODIL, said that until now, silicon carbide has been an underutilized material for making optical surfaces, such as high-precision mirrors, because of its brittle nature. "Optical surfaces need to be as smooth as possible," he said. "Brittle materials often leave cracked or pitted surfaces and subsurface damage after being machined." To machine the material to a desired shape, it is placed on a lathe and then ground using a wheel that is surfaced with very fine, almost dustlike, diamond grit. The grinding wheel is positioned at an angle against the material's surface so that a very thin layer is peeled away as the lathe turns. Silicon carbide has specific properties that make it attractive as a mirror-producing material. Because it is very strong, certain grades of the compound can be used to manufacture stiff, lightweight structures. The UTOS-made silicon-carbide material that Kahl and his colleagues are using at the Optics MODIL poses an additional challenge in that it is actually a two-phase material. It has a structure that in some ways resembles a microscopic filter of silicon carbide whose spaces are filled with a softer silicon. "The cutting depth must be kept very shallow so that this brittle, two-phase material can essentially be ground away by the diamond grit without causing any cracking," Kahl said. The Energy Systems and UTOS researchers also hope to lower costs and reduce production time with their advanced manufacturing methods. Milling techniques available at the Optics MODIL should enable technicians to more quickly produce a mirror that is very close to a desired figure before final finishing is needed. Current commercial techniques yield first-stage-production mirrors that are within about 3% of their final, desired shape. Optics MODIL techniques are expected to bring that figure down to around 0.1%. This vast improvement may allow technicians and scientists to eliminate the lengthy and expensive polishing phase of mirror production, which will further reduce manufacturing costs and time. --Wayne Scarbrough DIAMOND TOOLS EVALUATED FOR SDIO MIRRORS The high-precision mirrors used on SDIO's tracking and surveillance weaponry must be machined to near perfection in terms of shape and reflective finish. Intercepting high-speed missiles at great distances requires bright, distortion-free images. To obtain a smooth, uniform surface that is devoid of microscopic flaws, technicians rely on high-accuracy natural diamond tools. Therefore, the quality of a mirror's surface depends on the quality of the tool. To evaluate new high-accuracy natural diamond tools, Energy Systems has teamed with Contour Fine Tooling, Inc., a private manufacturer of high-quality diamond tools. The work is being performed at the Optics MODIL under a CRADA. "We are now able to produce tools better than our ability to measure within our facilities," said Allen Lake, a representative of Contour Fine Tooling, Inc. "We guarantee the waviness of the tools' edge to within 10 millionths of an inch. At the Optics MODIL, we have been able to inspect the edge to around 5 millionths of an inch, so we're well within our specifications." Art Miller, manager of the MODIL's Productivity Validation Test Bed, in which manufacturing equipment and tools are tested using methods unavailable to many private manufacturers, explained that if waves or bumps are present on the edge of the tool, these flaws will be imprinted into the material being machined. "The Optics MODIL," he says, "has the most accurate commercially available diamond-turning machine equipment in this country, and it is being used to evaluate this new diamond tool." Miller and colleagues at the Optics MODIL will be cutting sample mirrors to demonstrate the accuracy of the tools. "We will evaluate the tool's edge, the produced mirror, and the way the mirror scatters light, which is one indicator of surface quality." If the new Contour Fine Tooling tools are successful, he added, the reliability of the diamond-turning process and the quality of the resulting mirrors should be economically improved. --Wayne Scarbrough CRADA TO DEVELOP CERAMIC MACHINING TECHNIQUES The Oak Ridge Y-12 Plant and ORNL have begun a new collaborative research effort with a Delaware firm to help the company develop new, more efficient means of manufacturing ceramic composites for automotive use. The CRADA is expected to help Lanxide reduce future manufacturing costs. This collaboration with the Lanxide Corporation of Newark, Delaware,uses the precision machining capability developed at the Y-12 Plant in the manufacture of nuclear weapons components as well as the extensive expertise of ORNL in the development and analysis of advanced materials. Lanxide makes components from a broad range of proprietary composites of ceramic and metal. These composites are lightweight but very strong materials that have ideal properties for many applications. However, because of their hardness and wear resistance, they can be difficult to machine. The purpose of the collaborative effort is to develop cost-effective machining techniques for these composites. Work also will be done on establishing process control and material characterization techniques. The high cost of machining is considered to be a principal barrier to the use of ceramic-containing composites in the automotive industry. Initial work will be conducted at the Y-12 Plant during establishment of the Ceramic Manufactur-ability Center in the High Temperature Materials Laboratory (HTML). The HTML, which is open to industrial users, houses a unique collection of state-of-the-art equipment for analyzing and studying ceramic materials. The Ceramic Manufacturability Center, which is being established under a cooperative program for Cost-Effective Machining of Ceramic Components, is its most recent addition. Cooperative research and development projects under CRADAs such as this one with Lanxide and earlier agreements with Coors Ceramics Company and the Detroit Diesel Corporation will help U.S. industry to maintain a position of leadership in the machining of precision components and manfacturing advanced materials. It is expected that additional CRADAs will be forthcoming from other U.S. companies. Specific objectives of this project include improving the accuracy and consistency of critical workpiece dimensions that are generated by processes such as threading, drilling, grinding, honing, cutting, broaching, turning, and milling. The initial focus of the project will be development of techniques and tooling for the cost-effective machining of metal matrix composite connecting rods and brake calipers and rotors. The existing Lanxide machining processes will be characterized and test bed activities will be conducted in Oak Ridge to demonstrate the feasibility of applying Y-12 Plant manufacturing technology. Test pieces will be provided by Lanxide for the evaluation and feasibility demonstration. Characterization of machined test pieces will be shared by Lanxide and Oak Ridge. Lanxide Corporation represents the world's largest development and commercialization effort devoted to ceramic and metallic composites, according to Marc S. Newkirk, president and chief executive officer.CRADA with GM on Nickel Aluminide Components ORNL and General Motors (GM) Corporation are working under a CRADA to develop longer-lasting, heat-resistant assemblies for heat-treating furnaces used in producing automotive parts. The collaboration focuses on using nickel aluminide alloys developed at ORNL to manufacture assemblies consisting of trays, support posts, and fixtures. These assemblies will be used to hold automotive components as they are being heat treated in specialized furnaces. The goals of the CRADA for GM are a more energy-efficient manufacturing process for producing automotive parts, an increase in component throughput, and a reduction in cost stemming from longer tool life. To achieve these goals, the ORNL and GM researchers must develop an improved casting process, characterize and modify the alloy to optimize its manufacturability and performance under typical heat-treating furnace operating conditions, and test and evaluate specimens and prototype parts. ORNL, B&W Team Up on Fuel Studies A collaborative study between ORNL and the Babcock and Wilcox, Inc. (B&W), Alliance Research Center in Alliance, Ohio, may help electric utilities increase the efficiency of some power plants while reducing pollution. Researchers from the two organizations have teamed up to study the combustion of certain coal-derived solid fuels, called chars. They hope to use the results to determine the effectiveness of using chars as fuels in steam-driven power plants. Char is a residue from the production of coal liquids by mild gasification. The liquids are being investigated for use as supplemental engine fuels. However, after gasification, most of the coal's fuel energy value remains in the char, says Stuart Daw of ORNL's Engineering Technology Division. "We want to find an effective use for the chars," he adds, "so that no waste is produced." Not all chars are the same, and with different types of char come various burning characteristics. Understanding these differences will aid utility operators in selecting the best type of char for use in steam plants. "Several different techniques are available for mild gasification of coal," Daw says. "Different types of char are produced depending on which technique and which parent coal is used. We want to see how one char differs from another, particularly in the way they burn," he said. "The longer a char takes to burn in the combustion chamber, the greater chance there is for some of it to escape, which means some of the energy value is lost." Also, he said, a chars that burns uniformly, without hot and cool spots, could result in reduced output of pollutants, such as nitrogen oxides. The information gathered during these studies will be put into an existing data base for comparing the burning abilities of chars and other solid fuels. "This information," Daw says, "can help us determine if char products can compete with other solid fuels on the market," Another goal of the cooperative effort is to identify better ways to produce practical alternative fuels by identifying the mild gasification technique that yields the optimum split of liquid fuel and char. Additionally, B&W will use the information generated in the studies to determine how char fuels will perform in pressurized fluidized-bed combustors in power plants. Fluidized beds are one type of boiler in steam-producing power plants. Some operate at normal, or ambient, atmospheric pressure, and others are pressurized to about 10 times that amount. Daw said that a good foundation for determining chars' burning characteristics in ambient-atmosphere fluidized beds already exists. B&W will use the new data to extend that foundation to see how chars burn in a pressurized environment. This approach, Daw said, is a notable step in developing more efficient, less polluting boilers. Pressurized fluidized beds are more efficient than conventional boilers in converting the energy potential of coal into electricity because they produce both steam and pressurized gases to drive several turbines. Fluidized beds help reduce polluting emissions by treating them at the source. The boilers contain limestone that traps much of the sulfur released when coal or char burns. This method of treatment eliminates the need for flue-gas scrubbers, which are expensive cleaning mechanisms required by conventional boilers. The studies were funded as part of DOE's fossil energy research and developmment program. --Wayne Scarbrough CRADA ON MICROBES TO REMOVE URANIUM A technology that uses microorganisms to remove uranium and other toxic heavy metals from waste streams is the goal of a CRADA between Energy Systems and Ogden Environmental and Energy Services Company, Inc., of Fairfax, Virginia, working through its German subsidiary. This is the first international CRADA involving a national laboratory. The environmental remediation technology will involve use of bioreactor columns containing microorganisms (e.g., bacteria or fungi) selected for their ability to remove uranium, arsenic, and other heavy metals from waste streams. The microorganisms will be immobilized within beads the size of pinheads. The beads will be suspended in the bioreactors through which aqueous wastes containing dissolved metals will be pumped. As a result, the metal contaminants will then be adsorbed onto the microbial biomass. After the technology is developed at ORNL, Ogden will demonstrate its use in the remediation of contaminated water in flooded uranium mines in eastern Germany. The mines were formerly operated for the East German and Soviet governments by WISMUT (a private corporation that was once part of the government). Ogden's German subsidiary, Ogden Umwelt und Energie, will assist with management of the project. If successful, the technology may be used for various cleanup projects at DOE sites in the United States, including Oak Ridge. It may also "enhance the competitiveness of the U.S. environmental industry in the international market," says Clyde Frank, deputy assistant secretary for Technology Development for the Department of Energy. It has long been known that certain microorganisms adsorb heavy metals, but only recently have researchers considered the possibility of using this capability for waste management and environmental cleanup. Brendlyn Faison, principal investigator for the CRADA and a researcher in ORNL's Chemical Technology Division, has been successful in identifying organic material that adsorbs strontium and cesium from waste streams. For the CRADA she and fellow division researchers Jeanne Bonner, Gene Bloomingburg, John Norman, Brian Davison, and Mark Reeves along with Howard Adler, former Biology Division director now with Oak Ridge Associated Universities, will try to identify naturally existing microorganisms that can remove heavy metals from metal-contaminated water samples at ORNL that simulate the contents of the German pond. These microorganisms, she said, will not be modified by genetic manipulation. "Our role at first will be to identify the best medium to accomplish the removal of the heavy metals under the conditions at the German demonstration site," Faison said. "The medium probably will be a patented gel developed by Charles Scott and his colleagues at ORNL. Such gels are made from substances from natural sources such as seaweed." Up to 37 liters (10 gallons) of beads, held in a column less than a meter in diameter and more than a meter tall, can reduce the metal in 3700 liters (1000 gallons) from an initial concentration of 50 parts per million to no more than 50 parts per billion. According to Faison, the water leaving the column may not have to be handled as waste. Thus, the residual waste material would be only a fraction of the volume of the original waste stream and could be handled in one of two ways. By altering their chemical environment, the organisms could be forced to release the metal, which could then be retrieved. Or, the beads could be discarded and replaced with new ones. Because the microorganism and gel material in the beads are mostly water, the discarded material could be dried, reducing the mass by more than 80%, or it could be incinerated, leaving only metal compounds. Different microorganisms have an affinity for different families of heavy metals. Thus, the gel beads containing bacteria that remove uranium may not be equally effective in removing arsenic. However, by combining microorganisms on gel beads, a bioreactor could be tailored to remove several waste constituents at the same time. Ogden's program manager for this CRADA is senior vice president Kenneth Darnell. Principal participants on Ogden's team include Luke Williams, the project manager, and Leslie Dole, director of technology in Ogden's office in Oak Ridge and a former ORNL researcher. The costs of the $2-million, three-year CRADA will be shared by Ogden and DOE's Office of Technology Development. Technology Transfer Awards for Hypochlorite Removal Process Four Energy Systems employees have been presented awards for excellence in technology transfer by the Federal Laboratory Consortium (FLC). They were recognized for inventing, developing, licensing, and commercializing a process for removing potentially toxic chlorine from waste streams. Alicia L. Compere and William L. Griffith, both of ORNL's Chemistry Division, William P. Huxtable of the company's Engineering organization, and John Googin of the Oak Ridge Y-12 Plant Development Division received recognition at the FLC's recent National Technology Transfer Meeting in Indianapolis, Indiana. The FLC consists of representatives from more than 700 research and development laboratories and centers representing 16 government agencies. Consortium participants seek to enhance the transfer of federal technology results to domestic users in industry, state, and local governments. The recipients were among 38 winners selected by the FLC from among 3500 entries. They were honored for developing a simple, safe method for catalytically dechlorinating wastewater streams. The resulting patented Cl2EAN OUTTM process under pilot development employs a catalyst to convert toxic hypochlorite (chlorine bleach) to salt and oxygen (see schematic at right). The inventors of the process also received the International Hall of Fame's Advanced Technology Award. DOE has exclusively licensed the invention to R&D Solutions, Inc., based in Oak Ridge. The president of this company is Chet Thornton of ORNL's Plant and Equipment Division. R&D Solutions, Inc., received the Hall of Fame's Environmental Award for its efforts in developing the process. Cl2EAN OUTTM is expected to contribute to a safer environment because current studies show that it has the potential to decrease stream concen-trations of chlorine, which can be toxic to aquatic organisms. It also could be used to dechlorinate swimming pools and cooling towers for building air-conditioning units. Compere, Griffith, and Huxtable developed the hypochlorite degradation process in the early 1980s, assisted by Googin, a senior corporate fellow of Energy Systems. (keywords: supercomputing, optical systems, ceramic machining, ceramics, alternative fuels, bioremediation, chlorine, dechlorination) ------------------------------------------------------------------------ Please send us your comments. Date Posted: 1/26/94 (ktb)