ORNL is developing and evaluating technologies for producing highly energy-efficient, low-emission vehicles and for reducing traffic congestion.

Driving the Transportation Revolution

Early next century, you buy your next car, a near-zero-emissions vehicle. The body of your car is made of a lightweight carbon-fiber composite, not heavy steel. It gets 70 miles per gallon, thanks to a hybrid-electric engine powered by diesel fuel combined with electricity from a battery or fuel cell. You are thrilled with your new car's awesome computing power and the information and control technologies that guide you around heavy traffic and help you avoid accidents. You watch the road ahead while looking through the head-up display (HUD), a virtual dashboard projected on the bottom of the windshield that stays in your peripheral vision. Clicking on the steering wheel control panel, you turn the HUD's speed and gas level readings into a color-coded map that shows you the fastest route to a new restaurant. That evening after dinner, you transform the HUD into a "night vision" thermal-imaging display so that the scenes on the way home look amazingly clear.

The internal-combustion automobile, called "the product of the century" by Time magazine, has driven the U.S. economy. Mass-produced, affordable cars, pioneered by Henry Ford, created the middle class. Cars and trucks spurred the growth of cities and suburbs and generally improved our quality of life. The automobile industry is having an important economic impact in Tennessee, which ranks fourth in the nation in automotive manufacturing.

Lean, Clean Cars Needed

Although clearly the automotive industry contributed greatly to an improved quality of life for many in this century, 20th-century vehicles may actually threaten our quality of life in the 21st century. Because of the travel boom, we face unwelcome increases in traffic congestion, dependence on foreign oil, fuel consumption, and emissions of pollutants (despite improvements in fuel economy). These rising emissions may threaten the health of both people and the environment. Additionally, the 26% of U.S. greenhouse gases emitted to the atmosphere as a result of transportation-related activities could ultimately influence the global climate's stability.

To help address these problems, the Department of Energy is sponsoring research for the U.S. Partnership for a New Generation of Vehicles (PNGV), whose chief goal is more-energy-efficient, emission-free transportation vehicles. With funding from DOE and other agencies, ORNL is playing a role in developing safe vehicles that will emit virtually no pollutants and that will travel three times as far as today's cars, buses, and trucks, using the same amount of fuel. These "smart" vehicles will offer advanced information technologies to make driving safer and more efficient. Below is a sampling of ORNL's research related to new information, materials, propulsion, and emissions-control technologies that may advance the transportation revolution.

These technologies will be further developed at the National Transportation Research Center (NTRC), a collaborative effort among DOE, ORNL, the University of Tennessee, and The Development Corporation of Knox County. The official groundbreaking for the NTRC building, to be completed by 2000 in the Pellissippi Corporate Center in Knox County, took place April 8, 1999. The center, whose director is ORNL's Bob Honea, will take advantage of local transportation research expertise to solve complex national problems and to attract transportation-related firms to the region.

Testing Driver Response to Information Systems

Ron Harris has installed cameras, sensors, computers, and other equipment in this Dodge Intrepid, which is serving as the DOE Driver Research Vehicle.

Drivers of future cars may be deluged with information from "intelligent transportation systems" (ITS). For example, they may see HUD displays of maps and verbal messages and hear computer voices telling them how to get around traffic to reach a destination faster or how to avoid a collision in time.

In 1999, ORNL is seeking to determine how 40 drivers respond to such automobile-based information technologies, using a DOE research vehicle and a driving simulator. The research vehicle, a 1999 Dodge Intrepid, has been outfitted with sensors, instruments, and computers integrated and installed by Ron Harris of ORNL's Instrumentation and Controls (I&C) Division. The reactions of the drivers will be studied by human-factors expert Dan Tufano of ORNL's Computer Science and Mathematics Division (CSMD). The research is supported by ORNL's Laboratory Directed Research and Development Program. "New information delivery systems are designed to make it easier for the driver to navigate through traffic to a final destination and to operate the car more safely to avoid accidents," Tufano says. "Ironically, some information systems may distract and startle drivers, making driving less safe. To evaluate the effectiveness and safety of these systems, we will be collecting data on the responses of drivers in various highway situations, using both the research vehicle and a driving simulator."

Drivers of the research vehicle will wear physiological monitors linked wirelessly to the car's data acquisition system. The monitors will measure the driver's heartbeat rate, skin conductivity, and muscle tension, all of which signal the extent of a person's nervous reaction.

Wheel, steering wheel, and global positioning system sensors will indicate the speed, direction, and location of the vehicle on the road at any given time. Six miniature video cameras will allow the researchers to see the driver's hands and face and the forward and rear roadway scenes. An integrated data acquisition and storage system in the car's trunk will collect the vehicle, roadway, driver, and warning system data and "time stamp" it. Thus, the researchers will have information on events occurring inside and outside the car at any one point in time.

Here, Ron Harris connects the device controlling the lane tracker system with a computer.

The test car has a radar headway collision warning system and adaptive cruise control, a left blind spot collision warning system, and a video-based lane tracker and roadway departure warning system. These systems collect data and, through beeping tones and flashing lights, warn the driver of an imminent collision that can be avoided by a quick response. They also can be hooked up to the car's throttle so they can automatically adjust the accelerator to help the driver avoid an accident.

The vehicle will also be equipped with an ITS data bus, a computer network that handles information from the engine, vehicle, navigation and collision avoidance warning systems, and communications devices, such as a cell phone or pager. A filter made possible by the bus prioritizes messages so that the driver receives the most urgently needed information first. For example, messages needed to help you avoid an accident and remind you to take the next exit precede the message about the cancellation of tomorrow's staff meeting.

Night Vision Seen for Drivers

ORNL researchers are working on technologies to help people see more clearly in the dark. Panos Datskos and Slo Rajic, both of the Engineering Technology Division (ETD), have developed the world's first uncooled infrared photon detector, which could help improve the vision of people driving at night. The researchers have solved the problem of detecting photons of infrared light by measuring the mechanical stresses these photons induce in the microstructure of semiconducting material. A micromechanical quantum detector bends in proportion to the stresses, and the amount of bending indicates the presence and intensity of the light.

Today's infrared photon devices must be chilled by liquid nitrogen, but the ORNL invention does not require cooling to cryogenic temperatures. Because the new device will not require cooling equipment, it will cost less, weigh less, and use less electricity than today's infrared photon detectors. The uncooled photon detector offers the sensitivity and speed of cooled infrared photon detectors but not the associated increase in cost, size, and complexity.

Advanced Materials for Cars and Trucks

April McMillan and Felix Paulauskas examine the result of using microwave energy to bond two glass-fiber-reinforced composite parts.

The next wave in carmaking?
Steel is strong but it's also heavy. A car will go farther on less fuel if built using materials lighter than steel. Replacing steel body and chassis components with components made of the same carbon-fiber composite used in aircraft is expected to reduce vehicle weight by as much as 60%, significantly increasing vehicle fuel economy. Currently, production of carbon fibers is too expensive and slow for them to be used widely to replace steel as the primary material used in new cars and trucks. The problem: the high cost of the carbon-bearing starting material (precursor), the energy needed to heat it to make fibers, and the large ovens and other capital equipment used in its manufacture. Today pitch, or polyacrylonitrile (PAN) precursor, is converted to carbon fibers by thermal pyrolysis, a slow, energy-consuming process, combined with stressing to achieve the right properties.

In research for PNGV, Felix Paulauskas of ETD is working with industrial collaborators (AKZO, Amoco, Hexcel, Zoltek) to use microwave heating instead of less-energy-efficient thermal processing to increase the speed and reduce the cost of producing carbon fibers. They have already demonstrated that microwave-assisted processing of PAN precursors is a viable alternative to conventional thermal processing for manufacturing carbon fiber.

"Microwave technologies offer the potential to accelerate processing of precursors to produce carbon fibers with the appropriate properties," Paulauskas says. "Our early studies show that a properly designed and implemented microwave energy delivery system may enable a fourfold increase in the production speed, from 60 minutes to 15 minutes. Our economic studies show it has the potential to reduce fiber price by approximately 20% and the amount of energy required by approximately 15 to 20%."

Because the microwave units are smaller and cheaper and only several units are required to replace the massive single ovens now being used, maintenance downtime, capital equipment costs, and plant space requirements will be greatly reduced. Widespread implementation of this technology could replace about 50 to 70% of the conventional carbon-fiber processing line with very inexpensive equipment.

Coming unglued.
Adhesives will be used to join vehicle parts made of carbon-fiber composite materials. The problem is that components so bonded cannot be easily disassembled for repair without cutting or otherwise damaging the assembly. What's the solution? Microwave-reversible bonding can nondestructively take apart the adhesive bonds between components so they can be repaired or replaced.

"In our research," says Paulauskas, "we have identified suitable adhesives that can be taken apart by microwaves at low temperatures. We are developing adhesives that can be debonded at even higher temperatures."

Working with Barbara Frame of ETD and April McMillan of ORNL's Metals and Ceramics (M&C) Division, Paulauskas has shown that microwave-reversible bonding can facilitate the repair and maintenance of adhesive-bonded components made of plastic, carbon-fiber composites, and other nonmetallic materials. "By making repairs and upgrades more affordable," Paulauskas says, "microwave-reversible bonding could extend the life cycle of cars."

Better brakes and radiators.
Lightweight carbon composites could be even more valuable for components of automobiles and computers if they could be made so that more heat flows through them more quickly. For example, by increasing the thermal conductivity of carbon materials, they could be used to make safer automobile brakes. Today's brakes when applied may overheat or develop hot spots that can cause annoying vibrations.

Tim Burchell shows carbon-fiber composite samples of various shapes produced and studied at ORNL.

In ORNL's M&C Division, Tim Burchell and James Klett have developed novel carbon-carbon composite preform materials with the increased thermal conductivity needed to make safer brakes. They have developed processing technologies that will reduce the time for and cost of fabricating these materials. They have developed novel carbon foams with improved heat-flow properties for automotive applications.

The brake products developed from carbon-carbon composite preform materials using the ORNL fabrication process have 3 to 5 times greater thermal conductivity than conventional brake materials. The ORNL technique cut processing time requirements almost in half.

Klett developed a new low-density carbon foam that has a very high thermal conductivity. It transfers heat so rapidly that if you hold the foam in your hand and press an ice cube on top of the foam, your hand feels cold almost immediately.

Infrared images of ice melting on carbon foam developed at ORNL as a result of a discovery by James Klett.

"The key to the foam's conductivity is its unusual graphite crystal structure," Klett says. "If you filled a bag of marbles with foam and then took the marbles out leaving air pockets in the foam where the marbles were, you'd have a similar skeletal structure. This foam conducts heat almost as well as aluminum but at one-fifth the weight. Our foam is only 25% dense, so we hope to increase its thermal conductivity further by finding a way to fill the marble-shaped air pockets with graphite."

James Klett demonstrates the rapid melting of an ice cube applied to a carbon foam sample held in his hand. The sample transfers heat from his hand rapidly to the ice cube, causing the cube to melt quickly and his hand to feel suddenly cold. Such carbon foam could prove useful for making car radiators.

This foam can be easily fabricated into complex, three-dimensional geometries. It could be used to make smaller, lighter car radiators and other automotive parts.

Lighter electrodes for fuel cells.
One way to decrease emissions and increase fuel economy in cars is to power them with electricity from fuel cells (after gas stations start offering hydrogen or methanol for your fuel tank). Like an electric battery, a fuel cell has positive and negative electrodes (bipolar plates) with an electrolyte between. In a proton exchange membrane (PEM) fuel cell, hydrogen fuel is supplied to the negative electrode, and oxygen from the air is pulled into the positive electrode. During cell operation, the fuel is oxidized, and the hydrogen nuclei, or protons, produced in the reaction are transported through the polymer electrolyte to the negative electrode, while electrons freed by the reaction provide the current. A car operating on hydrogen and powered by a fuel cell would likely be considered an efficient, "zero emissions" vehicle.

The problem with using today's PEM fuel cells to power cars is that their bipolar plates, which are made of machined graphite, are too heavy, too brittle, and too costly for use in automobiles. The solution is to make bipolar plates from a carbon-fiber composite, which is lighter, tougher, and cheaper.

Ted Besmann, Klett, and Burchell, all of the M&C Division, have developed a method for making composite plates. "We mix chopped-up carbon fibers with a phenolic resin in a water solution and pour it into a mold having a screen on the bottom that is under a vacuum," Besmann says. "The water is pulled through the screen, leaving the fibers behind. The resulting shape is the bipolar plate. We call this slurry molding. We cure the plate to activate the resin, so all the fibers are glued together to give sufficient strength for handling. Then we stamp channels and holes into the plate to make it an electrode."

The next step is called chemical vapor infiltration. "In production we will be able to put thousands of plates into a furnace and flow methane over them at 1400°C," Besmann says. "Carbon from the methane will be deposited on the fibers, penetrating as much as half a millimeter below the surface of the fibrous material. That's why we call it chemical vapor infiltration rather than chemical vapor deposition. The deposited carbon will fill the pores, sealing the surface."

Why is this important? A fuel cell is really a series of cells, or a Dagwood-sandwich-like stack of bipolar plates with electrolytes between. The cells will not work if hydrogen and oxygen leak from one cell to another, so it is essential that the porous plate surfaces be sealed.

If carbon-fiber composite plates can be made to perform as well as graphite, they may make useful components for automotive fuel cells because, besides being tougher, they are only half as heavy and will cost perhaps one-fifth as much as machined graphite, meeting PNGV goals.

Increasing Efficiency of Electric Buses

One of Chattanooga's electric buses has an ORNL soft-switching inverter that makes the vehicle lighter and more energy efficient.

To reduce air pollution, noise levels, and dependence on imported oil, Chattanooga, Tennessee, allows only electric shuttle buses on one downtown route. Electric buses are cleaner and quieter than their diesel counterparts. As a partner in Chattanooga's Clean Cities Initiative to improve the environment and as part of a PNGV project, ORNL researchers are demonstrating that new technology can increase the energy efficiency of electric buses manufactured by Advanced Vehicle Systems (AVS) of Chattanooga.

The electrical energy for each AVS bus comes from a battery pack, which supplies 300 volts of direct current (dc). The dc must be converted by an inverter to alternating current (ac) to drive the vehicle's ac induction motor. The problem with conventional inverters is that they use a "hard-switching" technique—their solid-state transistor switches operate at full load voltages and currents as they open and close up to 20,000 times per second to create an alternating current. As a result, they waste electrical energy, generate heat, wear out components quickly, and produce high voltage spikes in the motor and significant electromagnetic interference (EMI) that can disrupt operations of other electronic devices.

One of the Chattanooga buses has a new generation of inverter technology developed at ORNL by Fang Peng, Gui Ja Su, Cliff White, George Ott, Matt Scudiere, Laura Marino, and Curt Ayers, all in ETD. This "soft-switching inverter" is more efficient, more compact, and more reliable than conventional inverters and eliminates the problems of voltage spikes and EMI.

ORNL's auxiliary resonance tank soft-switching inverter has small components that temporarily divert electrical current from the main switches so that no power is lost when they are turned on and off. The device also has light, inexpensive "sinks" to absorb operating heat so that components can be placed closer together safely. As a result, the device weighs only one third as much and occupies only one-tenth as much space as the newest conventional inverter. Thus, it is ideal for electric cars and buses because its lighter weight will increase vehicle energy efficiency.

Catalyst Candidates for Auto Emission Control

Smog and ground-level ozone could be reduced in large cities if nitrogen oxides could be removed from the exhaust from internal-combustion and diesel engines. Past research suggests that nitrogen oxide (NO) can be captured using an elemental metal and metal oxide if the surface interactions are right. The actual mechanism for NO removal remains unknown. Trying to understand this mechanism, ORNL researchers Steve H. Overbury and David R. Mullins have been studying a candidate catalyst for emission control, especially in future lean-burn engines. It contains a metal—rhodium (Rh)—and a metal oxide that has multiple oxidation states, cerium oxide (CeO_y; y = 1.5-2). They prepared Rh-CeO_y materials as thin films and found that this form is promising for laboratory study to determine how surface interactions actually capture NO from exhaust. Using surface-sensitive spectroscopic techniques to characterize the thin films, the researchers obtained unprecedented detail on interactions among NO, rhodium, and cerium oxide. Such insights should help researchers design improved catalysts for advanced automotive engines.

Characterizing Diesel Particle Exhaust

Besides their high emissions of nitrogen oxide, advanced diesel engines that are being developed to use 35% less fuel per mile than today's gasoline-burning, spark-ignition engines present another environmental problem—airborne particulates that are hazardous to humans inhaling them. Diesel exhaust particles are generally less than 2.5 micrometers (µm) in diameter—a potentially dangerous size range because the smaller they are, the easier it is for them to sneak past the body's filters and land in the lungs. They also reduce visibility in populated areas.

John Storey and others in ORNL's ETD have developed an electrostatic method for capturing diesel particulates from a test diesel en-gine's exhaust so that their structure and makeup can be analyzed. He has also developed ways to measure the varying sizes of agglomerates of particles for experiments that will determine the effect of changes in diesel fuel combustion and exhaust aftertreatment (using catalysts) on particle sizes.

Micrograph of diesel particulate showing the semicrystalline, "turbostratic" structure in which sheets of atoms in rows are turned in different directions.

Ted Nolan, Karren More, and others in the M&C Division have used the Hitachi HF-2000 transmission electron microscope and other characterization tools to determine the structures of diesel particulates. They found that particles measuring 20 to 40 nanometers may range from a noncrystalline (amorphous) structure to a semicrystalline (turbo-stratic) structure in which atoms are lined up in sheets, but the sheets are not oriented the same as they are in graphite. "It's like tearing off all the sheets from a pad of paper," says Nolan, "and then laying them on top of each other while turning them in different directions."

Storey is also working with DOE's Lawrence Berkeley National Laboratory to build and calibrate real-time scatterometers for both labs. These instruments will be used to scatter light off particulates in diesel exhaust to measure their relative sizes. It is important to know the fractions of particulates of various sizes in the exhaust because new particulate standards will limit the total mass of particle sizes under 2.5 µm. In addition to monitoring emissions compliance, these rapid-response instruments could also provide information to engine manufacturers and service facilities to guide the development of cleaner diesel engines and the adjustment of current ones.

Computer Simulations:
Designing Cleaner, More Efficient, Safer Cars

A computer code developed largely at ORNL that won the 1998 Gordon Bell Prize for the fastest application in high-performance computing was used to perform a first-principles simulation of the magnetic behavior of 1024 atoms of iron (shown here). This code will be used to simulate the electronic structure of materials to enable the design of better catalytic converters and lighter batteries for electric cars.

A prize-winning computer code written at ORNL for use on powerful supercomputers may advance transportation research. This code is being used to simulate the electronic structure and properties of materials to enable the design of better catalytic converters, lighter batteries for electric cars, and better coatings on aircraft turbine blades. The code, which ORNL Corporate Fellow Malcolm Stocks helped develop, won the 1998 Gordon Bell Prize for the fastest application in high-performance computing. It was the first code to run at greater than one teraflop, or more than one trillion calculations per second. Using the code, Stocks and his collaborators (including ORNL's Don Nicholson, Xaioguang Zhang, and Bill Shelton) performed a first-principles simulation of the magnetic behavior of 1024 atoms of iron, using increasingly powerful Cray T3E supercomputers.

By crunching numbers on ORNL's powerful parallel supercomputers, Laboratory researchers are reducing the need to crunch metal by crashing cars together to determine how well their materials hold up in a collision. By making billions of calculations per second and creating visualizations of mounds of data using these computers, they have developed a nonviolent method for designing the lightweight, fuel-efficient cars of tomorrow that are at least as safe as the heavier steel cars of today.

ORNL researchers are modeling an Audi A8, an all-aluminum car made of a lightweight material that may be used in future cars. A computer model of a car and its components, combined with a model of the lightweight material making up the car, help researchers analyze how well the material will hold up in a variety of collisions.

Researchers at ORNL, in collaboration with the National Highway Traffic Safety Administration and George Washington University, are developing detailed computer models of a variety of vehicles after disassembling them and scanning in the parts. In the past few years, they have completed models of the Ford Taurus and Explorer, both among the top-selling vehicles in the United States. Researchers are now modeling an Audi A8, an all-aluminum car that is one of the first to use a lightweight material that may be used extensively in future cars.

"We use a computer model of a car and its components combined with a model of the lightweight material used in the car to analyze how well the material will hold up in a wide variety of crashes," says model developer Srdan Simunovic of the Computer Science and Mathematics Division. "We can substitute different materials in individual parts in our model and compare the results to determine which material performs best during a collision between cars.

"Our development of the parametric finite-element model has enabled us to tune the grid—in which the vehicle is divided into hundreds of small sections—according to the kind of crash we're going to simulate and number of computer resources available. This innovative solution has made car crash simulation more manageable for the computer."

ORNL researchers will soon be using a recently acquired IBM RS/6000 SP supercomputer for PNGV studies to support development of advanced transportation vehicles and alternative fuel technologies. The initial IBM system is configured to perform 100 gigaflops (a billion calculations per second), or a tenth of a teraflop. It will be upgraded to 400 gigaflops later this year and to 1 teraflop in the middle of 2000. The new machine will be more than seven times faster than ORNL's Intel Paragon XP/S 150, which in 1995 was the world's fastest computer.

Reducing Congestion:
Truck Rollover Warning System

Most of us have at some time been delayed by a traffic jam caused by an overturned truck. Truck rollover crashes are responsible for at least $3 billion a year in losses associated with deaths and injuries, property damage, lost productivity, and lost time because of traffic backups.

To address this problem, an ORNL-led team has designed a system to warn truck drivers who are at risk of rollover in time for them to take corrective action. This prototype system will be implemented, tested, and evaluated by Scott Stevens of the Energy Division and Phil Spelt of CPMD, in collaboration with representatives of the Tennessee Department of Transportation, the Transportation Research Center at the University of Tennessee, and U.S. Xpress Enterprises, a Chattanooga trucking company.

This warning device could help drivers avoid a truck rollover.

"Our system of on-board sensors and a computer will collect data and determine the instantaneous roll stability of three tractor-trailer rigs," Stevens says. "We will deploy roadside beacons at selected sharp curves or ramps in Tennessee to broadcast curve characteristics to oncoming vehicles. Our smart test trucks will receive the broadcasts, estimate the risk of rollover, and warn the drivers if they are approaching the curve at a speed that is likely to cause rollover. An alarm will be sounded in time for the drivers to take corrective action. The on-board instruments will also collect data about the drivers' response to highway design features and various traffic situations." It is estimated that 4000 of the annual 15,000 truck rollovers could be prevented with a rollover alert system.

ORNL is helping to drive the transportation revolution by evaluating intelligent transportation systems, developing better ways to make lightweight materials for vehicles, improving efficiency in electric vehicles, developing and testing methods for reducing and characterizing emissions, evaluating the safety of lightweight vehicles in collisions, and developing information systems to reduce traffic congestion by lowering the risk of truck rollovers.