The technology was developed by Jonathan Woodward of ORNL's Chemical Technology Division and two contractor employees from Midwest Technical, Inc.--Lynette M. Stephan and Laurence J. Koran, Jr. The new technique resulted from a discovery made in 1992. At that time, the researchers were trying to use a protein-an enzyme called cellulase-to convert waste paper to sugar for fermentation into a liquid fuel called ethanol.
"We were surprised to find that cellulase easily digests and removes ink from newspapers under certain experimental conditions," says Woodward. "We found our technique separates high-quality ink-free pulp fibers from ink-covered fibers, creating a whitish recyclable paper product and a gray product that could be used to make egg boxes, insulation, or combustible fuel."
The enzyme, or biocatalyst, technology has three advantages over the traditional de-inking technology. First, it works on old newspapers that use water-based ink as well as oil-based ink, without releasing ink to the environment.
Second, because it requires no de-inking chemicals, it uses much less energy than traditional methods-as much as 2.3 million British thermal units less per oven-dried ton of paper. If the technique were used on the 8 million tons of newspapers collected annually for recycle, 0.018 quads of energy, or 13 million barrels of oil per year, would be saved.
Third, the ORNL technology is environmentally friendly because cellulase is an enzyme produced naturally by a fungus that is used in washing powders such as those for stone-washing jeans. Traditional de-inking technologies use potentially harmful chemicals, such as caustic sodium hydroxide, sodium silicate, and hydrogen peroxide.
The new biocatalytic technique uses the enzyme cellulase to treat cellulose fibers making up more than half of waste paper. Cellulase can completely break down cellulose into glucose sugar. However, under certain conditions discovered by the researchers, the action of cellulase results only in the desired separation and pulping of the fibers with little cellulose breakdown.
The apparatus used for the process is simple. It consists of two cylindrical glass containers, or bioreactors, connected by a tube through which water circulates. In each container is a cylindrical strainer made of 20-mesh plastic canvas, the porous material used for needlepoint.
Waste paper and cellulase are mixed together in the strainer in the first bioreactor. The mixing of the cellulase with the waste paper separates its large and small pulp fibers and triggers the release of ink from the paper surface. The released ink tends to stick to the small pulp fibers.
As water flows at 40°ree;C between the two containers, the small inked fibers pass through the pores of the plastic strainer and end up in the other bioreactor outside its strainer. The larger ink-free pulp fibers remain inside the strainer of the first container because they are too large to pass through the strainer pores. In this process, ink is almost completely removed from paper.
Woodward says that several paper companies that have shown interest in the new ORNL technique. Funding for the work that resulted in the discovery came from the Department of Energy's Office of Basic Energy Sciences. The development of the separation technology was supported by internal funding through the Laboratory Director's Research and Development Fund.--Carolyn Krause
A postage-stamp-size instrument for efficiently separating and identifying chemicals in liquids is being developed by ORNL scientists. Such a device could have widespread use in the environmental, manufacturing, health care, and pharmaceutical industries.
The ORNL researchers have also shown that microscopic devices can be used to carry out chemical reactions. They have used the device to separate chemicals in a liquid droplet in only 150 milliseconds. The work suggests that an entire chemical laboratory, including chemical containers, "beakers" for mixing chemicals, and analysis instruments, can be placed on a microchip.
The "chip" is a glass plate as thin as a microscope slide and about the size of a postage stamp. A winding, hairlike capillary channel, covering an area the size of a dime, is etched in the glass using standard micromachining techniques. The etched channels are closed by bonding a thin plate of glass over the top.
"Such a microchip laboratory could provide faster, cheaper, and more reliable chemical analyses for environmental monitoring, industrial process control, and medical diagnosis," says J. Michael Ramsey, one of the developers of the technology. "It could also be used by unskilled personnel to perform sophisticated chemica l analyses in remote locations."
The laboratory on a chip offers several potential advantages over conventional approaches to chemical analysis. Preparation of chemicals for analysis is automated, saving labor and protecting humans from unnecessary chemical exposure. Also, the amount of chemical reagents needed to stimulate chemical reactions on a chip is a millionth of the typical volume used in a laboratory setting, thus minimizing chemical waste. Finally, because the miniaturized device has no moving parts, it should be more reliable and inexpensive enough to be disposable.
Thanks to its greatly reduced size and weight, the microchip laboratory could be incorporated into hand-held devices for surveying waste sites and diagnosing a patient's disease in a physician's office. It could also be part of a small gadget that would be used in chemical process pipes to monitor and control production in a factory.
"One of our goals is to develop microdevices for chemical analysis that demonstrate the same advantages as microelectronics, including small size, low cost, high speed, reliability, and operational simplicity," says Ramsey. "We have designed and tested several types of microfabricated chemical separation devices. Our results show that it is possible to reduce instrument size from a few cubic feet to several cubic centimeters and that the performance of the miniaturized instruments is equivalent to or better than the conventional laboratory versions.
"We have also shown that chemical reactions can be performed rapidly in a very small volume using micromachined devices," Ramsey continues. "At ORNL results of a chemical analysis on a chip were obtained in 5 minutes using reagents in an amount equal to about 1/250 of a drop of water. In general, such a device would consume about 1 milliliter of reagent, or the equivalent of 20 drops of water, per year of operation.
"Engineers have not been able to design microscopic pumps and valves that work well enough for these applications," says Ramsey. "We do fluid pumping and valving using electric fields and osmotic forces to induce liquid to flow through a microscopic channel etched in glass. Charged molecules of different chemicals move at different speeds based on differences in charge and size. Uncharged molecules of different chemicals move at different rates based on differing attractions for material dispersed in the liquid or coating the channel walls.
"With electric field strengths as high as 1500 volts per centimeter, we have separated chemicals in liquids in as few as 150 milliseconds, which is one of the fastest speeds ever," Ramsey adds. "This demonstration shows that a miniature chemical device, much like a microelectronic device, can achieve greater speeds than devices of conventional size."
Ramsey says that chemistry labs on chips could be integrated into a miniature chemical factory. Just as several computers can work at the same time on different parts of a complex problem, chemical analysis microchips could work in parallel to synthesize and test new drugs.
"Test results might be used to influence the next sequence of compounds that is synthesized for tests," he says. "This approach could speed up the discovery of new drugs that are effective against disease."
Besides Ramsey, the developers of the microchip laboratory include Stephen C. Jacobson, Roland Hergenroder, Lance B. Koutny, and Alvin W. Moore, all of ORNL's Chemical and Analytical Sciences Division. This work has been funded by the ORNL Director's Laboratory Directed Research funds and by the Department of Energy's Office of Nonproliferation and National Security.-Carolyn Krause
Charles L. Britton, Jr., and Alan Wintenberg, both of ORNL's Instrumentation and Controls Division, have designed a radiation-resistant chip for storing information resulting from particle collisions measured in a high-energy physics experiment planned for the recently closed Superconducting Super Collider (SSC) project. They are now designing radiation-resistant circuits for robots that may someday be used to clean up contaminated tanks at DOE waste sites.
For this work in collaboration with Ken Read of ORNL's Physics Division and Lloyd Clonts, a University of Tennessee at Knoxville (UTK) graduate student, Britton and Wintenberg recently received a Technical Achievement Award from Lockheed Martin Energy Systems, which manages ORNL for DOE. They were cited for "superior contributions to the experimental physics community by developing the world's first radiation-hardened analog memory unit."
The researchers have also designed a less expensive chip not hardened against radiation for use in collider physics experiments at other accelerators. "Chips like these," Britton says, "could be used for storing television signals long enough to process them into improved TV images. By processing the stored signals in a few millionths of a second, TV pictures could be made clearer, sharper, and steadier."
The radiation-hardened chip was made by a special process for manufacturing complementary metal oxide semiconductor (CMOS) chips. Harris Semiconductor, Inc. of Melbourne, Florida, developed this process to meet military needs. Britton and his colleagues developed circuits using computer-aided design and drafting tools and sent the designs to Harris by electronic mail. Harris then manufactured about 200 chips to meet ORNL's specifications.
To determine each chip's resistance to radiation damage, the ORNL researchers subject them to gamma radiation from a cobalt-60 source. "These chips are supposed to be able to withstand as much as 10 million rads of radiation," Britton says. "By contrast, ordinary chips would be destroyed by about the same amount of radiation that kills a cockroach-20,000 to 30,000 rads, about 50 times as much as would kill a human."
The SSC, which was terminated by the U.S. Congress in 1993, was being designed to accelerate protons to extremely high energies traveling in opposite directions around a 54-mile ring in Texas. One object of research using the machine was to understand the most basic particles of matter and the glue that holds them together.
The proton beams were to be slammed together in the central tracker system of the eight-story-high gamma-electron-muon (GEM) detector being designed for the SSC. ORNL was asked by DOE to develop radiation-hardened chips to store information about the particles shooting out in all directions during head-on collisions of protons 62 million times a second in the opposing beams.
"The chips had to be hardened against radiation because of the highly energetic gamma rays and neutrons produced from proton collisions and other particle interactions," Britton says. "They should be able to operate even with a little radiation damage by recalibrating their memories."
"There was no practical way for the SSC to immediately process information on all the particles, their energies, and the directions and distances they travel," Wintenberg says. "A lot of the information would be of no interest because it would result from protons passing by each other or glancing off each other rather than directly colliding. So analog memory chips were needed in the SSC to store all electrical information as voltages until computers decide which information might be significant to reconstruct particle tracks, for example. Such voltages would then be converted to numbers for computer processing."
The ORNL chip consists of many capacitors, transistor switches, amplifiers, and associated circuitry. In the SSC the capacitors would have stored electrical signals as voltages from charged wires in gas-filled detectors. These wires would have picked up electrical charges produced when particles from proton collisions passed through each detector, stripping electrons from gas molecules. The voltages of interest stored on the chips would have been turned into numbers for computer use by analog-to-digital converters.
"Radiation-hardened chips will not be needed for the SSC now because of its demise," Read says, "but they may be used in a somewhat smaller collider to be constructed in Switzerland for experiments in which U.S. scientists will participate. Other uses would be for robots that work in high-radiation environments such as nuclear power plants or waste sites."
The ORNL researchers are now working with ORBIT Semiconductor to produce CMOS chips that have similar circuit designs but are not radiation hardened. These chips can be made for a much lower cost than the radiation-hardened ones.
These memory chips will be used for relativistic heavy-ion physics experiments involving ORNL physicists. Several applications are planned for the PHENIX experiment at the Relativistic Heavy Ion Collider Center Center under construction at Brookhaven National Laboratory in New York and an existing accelerator at the European Center for Particle Physics (CERN) in Geneva, Switzerland.
The CERN experiment is the WA98 experiment, one in a series of experiments aimed at creating a plasma of free quarks, the basic constituents of particles in atomic nuclei, and gluons, the particles that bind quarks together. It is believed that a superhot quark-gluon plasma was first formed during the Big Bang that created the universe and that, within several millionths of a second, the plasma cooled into the atoms making up the universe. Glenn Young, Frank Plasil, and others in ORNL's Physics Division have worked on these experiments since 1986.
"In the past," Britton says, "they have used conventional electronics developed at ORNL for these experiments. But for these new experiments they will use custom chips from ORNL for the first time."
The funding for developing radiation-hardened analog memory chips for the SSC came from the Department of Energy through the SSC Laboratory in Dallas, Texas. DOE's Office of Energy Research also has been supporting development of the other chips for the Brookhaven and CERN accelerator experiments.
Environmental scientists at ORNL are working to develop a fast-growing grass as a source of liquid fuel. The Biofuels Feedstock Development Program at ORNL has selected a native grass species for further studies in alternative-fuel development. The species is called switchgrass.
Scientists and politicians alike realize the importance of developing clean, renewable sources of energy. Today, much of the world depends on oil reserves that are concentrated in a handful of nations. Another problem is pollution, much of which stems from fossil-fuel use. Conversion of woody and herbaceous plants into clean-burning fuel offers an attractive possibility for meeting future energy demands, researchers say.
Switchgrass is a perennial warm-season grass that grows well in many areas of the country-even on fairly dry, nutrient-deficient land of marginal quality-and it has positive environmental attributes. Its natural range extends from Quebec through Mexico and into Central America.
Switchgrass has two growth forms. One is a leafy, thicker-stemmed variety common to lowland sites in the Southeast. The second form is shorter, fine-bladed, and more typical of grass in the western plains. Both were abundant on the native prairie when American pioneers arrived.
The ORNL program emphasizes the need for high-yield crops that can be grown economically over a wide area and with minimum impact on the environment. Switchgrass was selected as a model herbaceous crop because it has high, stable yields and relatively low production costs. Its other positive environmental attributes include relatively low need for pesticides and fertilizers and excellent soil conservation potential. These features could make switchgrass tremendously beneficial to owners of small farms in a depressed agricultural industry.
"One reason switchgrass was selected as a fuel alternative is that it is very similar to hay in the way it is raised, making its production compatible with traditional agricultural practices," says Sandy McLaughlin, an ORNL senior research staff member and a task leader in the ORNL biofuels program. "In fact, the equipment used for harvesting hay can also be used for switchgrass, helping to lower the front-end costs for farmers who want to grow switchgrass and sell it to conversion facilities."
McLaughlin says that 100 gallons of fuel could be produced per dry ton of harvested switchgrass. The amount of fuel produced per acre of harvested switchgrass could be as high as 500 gallons.
Research on developing fast-growing grasses and hardwoods for the ORNL program is paralleled by research at DOE's National Renewable Energy Laboratory in Golden, Colorado. There engineers work to develop more efficient ways to convert biomass to liquid and gaseous fuels. By the 21st century, energy crops could yield more than 230 million gallons of ethanol annually, or about 650,000 gallons of ethanol every 24 hours.
The research at ORNL is supported by DOE's Office of Transportation Technologies through the Energy Efficiency and Renewable Energy Program.
Research at Virginia Polytechnic Institute and State University has shown that deep switchgrass rooting patterns improve soil quality by increasing soil carbon content. This ability to transfer carbon dioxide from the air to the soil will help offset carbon dioxide emissions to the atmosphere, a significant global pollution problem.
Switchgrass protects against erosion, has low fertilizer and pesticide requirements, and provides cover for wildlife. But someday it may have a new claim to fame. Americans may eventually be talking about making gas from grass.--Angela Swatzell and Wayne Scarbrough
As conceived, the engine could provide high-level performance when desired, would cost less than conventional automobile engines, and should offer economy-minded drivers the possibility of 38 kilometers per liter, or 80 to 90 miles per gallon, of gasoline, for a lifetime mileage of at least 805,000 kilometers, or 500,000 road miles.
Carsten M. (Kit) Haaland of ORNL's Computing and Mathematical Sciences Division has invented a magnetohydrodynamic liquid metal (LM) engine that can "shrink" the engine combustion volume down to around 0.06 liters, which is much smaller than that found in existing commercial automobiles. The engine volume can expand by a factor of 10 to 0.6 liters in about one second when needed. The smaller engine volume will burn just enough fuel at high-efficiency steady cruising, whereas the larger engine volume will take a lightweight vehicle to 96 kilometers per hour, or 60 miles per hour, in just seven seconds.
Lockheed Martin Energy Systems has applied for a patent for the magnetohydrodynamic engine. Magnetohydrodynamics is the science of magnetic fields combined with the motion of liquids that conduct electricity, such as mercury or liquid sodium.
The output of the LM engine is electric power in the form of alternating current (ac) at variable frequencies that match the variable rotational frequencies of car wheels. The electrical output of the LM engine is fed by wires directly to ac motors that provide rotation in the driving wheels. Unlike a conventional electric generator or auto engine, the LM engine has no rotating parts. The electric power is generated by the back-and-forth motion of the LM engine in a strong magnetic field. The driving force for this motion comes from the same force that drives a conventional auto engine-the expansion force that results from burning fuel.
However, in the LM engine, the connection from expansion force to electric power generation is made without any mechanical devices such as connecting rods, crankshaft, pulleys, or belts. Thus, the engine has fewer moving parts and problems with friction.
"The LM engine can adjust its output to suit different driving requirements, ensuring superb vehicle performance and maximum fuel efficiency at any speed," Haaland says. "Because there are fewer moving parts in the engine, and because an auto powered by it won't need a clutch or transmission, the initial vehicle costs should be lower and the frequency of repairs should be greatly reduced.
"Because the engine displacement volume can expand quickly," Haaland says, "a driver will be able to push the accelerator pedal to the floor and get an exciting response from an engine that has suddenly increased its available power up to 10 times. The driver could then let up on the gas pedal to shrink the LM engine volume when the desired speed is reached. The LM engine could provide the driver up to 80 miles per gallon of fuel at a constant speed of 65 miles per hour, and up to 100 miles per gallon at a steady 55 miles per hour on a smooth level surface."
Haaland's part of a 1990 study supported by the Laboratory Director's Research and Development Fund resulted in the variable-size LM engine. He conceived the idea for a four-piston double-duct LM engine in 1992. The four-piston setup is used to drive the LM engine back and forth in opposite directions in two separate ducts. This arrangement eliminates vibration and instabilities in the LM motion that result from strong magnetic fields. It also allows a transformer to be conveniently connected to one side of the double-duct to step up voltages and step down currents to practical levels. A specially designed computer chip would control the operation of the LM engine and the vehicle.
The variable-size (variable-displacement) LM engine uses free pistons, with no connecting rods or solid metal attachments. The pistons are driven by combustion of a liquid fuel, such as gasoline, diesel fuel, liquid propane, or some combination. Each pair of pistons is linked by a duct containing liquid metal, the source of which is a nearby reservoir. The distance the pistons travel is limited by the quantity of liquid metal in the operating ducts: the more liquid metal, the less piston movement.
If the liquid metal is pumped out of the duct back into the reservoir, the piston stroke gets longer, resulting in more movement of the liquid metal through the magnetic field produced by nearby magnetic material; as a result, more electricity is produced to power the wheels. This arrangement allows the LM engine size to vary according to need.
Such an engine requires liquid metals that have high electrical conductivity, low density, and low viscosity. The most favorable liquid metals for this purpose are members of the alkali metal family, such as liquid sodium, liquid potassium, or a combination of these. Because they will burn spontaneously in air or water, the ducts carrying liquid metal in the engine must be designed to remain sealed against the chaotic forces of collision. To prevent any contact with air or water, the liquid metal is sealed permanently in two channels. Approximately two liters (slightly more than two quarts) of liquid metal would be used in a typical LM engine.
"There are no load-bearing sliding surfaces in the LM engine. The sideward push of pistons against cylinders in the usual engine is eliminated, and no crankshaft, connecting rods, wrist pins, and their associated bearings are required," Haaland says. "Thus, the engine will experience less load from internal friction. Stress points on connecting rods and wrist pins are eliminated.
"Because of these features, the LM engine could operate up to 300 cycles per second in a two-cycle mode with greatly improved scavenging, resulting in much smaller engine size per kilowatt. All these factors, including the use of air cooling instead of water cooling, result in less weight, improved fuel efficiency, lower cost, and longer life for the LM-engine-powered vehicle.
[ Next article | Search | Mail | Contents]