Ames Laboratory, U.S. Department of Energy, Ames, Iowa Ames Laboratory, U.S. Department of Energy, Ames, Iowa
Science and Technology at the Ames Laboratory
Summer 1995
At a recent conference on sensors, a scientist described an implantable glucose sensor for diabetics that had seven different coatings, each designed to screen out a chemical species that might interfere with the accurate measurement of blood sugar. The coatings worked, but not surprisingly the sensor's response time left something to be desired. In fact, someone in the back of the room was heard to mutter, "By the time the guy knows he has a sugar problem, he'll be dead."
The glucose sensor illustrates the central problem in the design of chemical sensors: achieving selectivity without incurring a penalty that will ultimately scuttle the design. This requirement is so tough that, excluding biological molecules that can be detected by natural antibodies, there are few practical chemical sensors around.
At the Ames Laboratory Guojun Liu, a postdoctoral fellow in the Environmental Technology Development Program, and Mankit Ho, a graduate research associate in the Processes and Techniques Program, are working on two clever means of making highly selective coatings for two sensors, a hydrogen sensor and an organic-vapor sensor. The first coating, which is an alloy of palladium and nickel, is based on some fairly well-known chemistry. The second coating, made of a cornstarch derivative called cyclodextrin, is much more experimental. "That's our blue sky work," says Marc Porter, an Ames Lab associate and associate professor of chemistry at Iowa State University (ISU). Porter oversees both projects with Stanley Burns, an ISU professor of electrical and computer engineering and senior investigator at ISU's Microelectronics Research Center (MRC).
The palladium-based sensor is one of several projects at Ames Lab that address problems at the Department of Energy's Hanford Site in southeastern Washington, where roughly 50 percent by volume of the nation's high-level radioactive wastes from weapons production are stored. Roughly 50 of the 177 million-gallon underground storage tanks at the site are on a watch list because they contain potentially explosive chemicals, including hydrogen gas (see Inquiry, Summer 1992).
The hydrogen levels in the tanks are currently being monitored with an electrochemical sensor, but this instrument has a detection limit of 7 parts per thousand and a response time of several minutes. The Lab's prototype sensor, in contrast, has a detection limit in the parts-per-million range and returns a reading more quickly.
The affinity of palladium for hydrogen achieved notoriety in 1989 when electrochemists Martin Fleischman and Stanley Pons claimed they had succeeded in fusing the nuclei of hydrogen atoms with a palladium cathode. "Not that we think there's cold fusion going on in our palladium coating," comments Liu. Unlike cold fusion, the hydrogen chemistry relevant to the sensor's operation is well understood and manageable.
The sensor signals the presence of hydrogen by changing its vibrating frequency. The palladium-based alloy is deposited on a piezoelectric resonator, a thin disk of a material that can be made to vibrate by applying a varying voltage across it. The disk has a preferred vibrating frequency, called its resonance frequency, which can be detected electronically.
If a resonator coated with the palladium-based alloy is exposed to hydrogen, the hydrogen dissolves in the palladium, the disk becomes heavier, and it resonates at a lower frequency. "It's like an organ pipe," Porter explains. "The longer the pipe, the deeper the pitch."
What is new about the Ames Lab sensor is a neat trick for getting rid of interfering signals. "The beautiful thing about this sensor is that its response to hydrogen is a bulk phenomenon, but its response to interfering gases is a surface phenomenon," says Liu. If the coating on a resonator is 120 nanometers thick, hydrogen will cause a comparatively big change in the resonance frequency; but if the coating is only 30 nanometers thick, the change will be smaller. On the other hand, the response to interfering gases is nearly independent of the coating thickness. This difference allows interfering signals to be subtracted from the raw signal to give a pure hydrogen signal.
The Ames Lab group hopes eventually to use special microminiature resonators now under development at the MRC for the hydrogen sensor. The fundamental resonance frequency of these devices will be roughly 100 times higher and their sensitivity to changes in mass comparably greater.
Liu estimates that with the new resonators the sensor's limit of detection will be in the parts-per-billion range, which is not bad for a chemical system that was performing in the parts-per-thousand range only yesterday. "That's what the Department of Energy pays us for," he says with a laugh. The second Ames Lab sensor, the one that detects organic vapors, is based on a completely different concept. Molecules have size and shape as well as chemistry. If a resonator coating has open spaces, some molecules will fit comfortably into them but other molecules will be excluded.
Conceivably such a coating could be made mechanically with a scanning tunneling microscope or other instrument capable of moving one atom at a time. But a smarter way to go about it is to find a molecule with a ready-made hole that can be induced to self-assemble into a monolayer.
One such molecule is cyclodextrin, a doughnut-shaped molecule formed when an enzyme snips cornstarch molecules at particular locations. Because the inside of the doughnut is hydrophobic, or not attracted to water, in water solutions it is a comfortable bolt-hole for other hydrophobic molecules. A variety of small molecules readily bind within the cyclodextrin molecule, a property that has been exploited to make aroma molecules used in the food industry stay pungent longer.
The diameter of the cyclodextrin cavity depends on the number of glucose molecules the cyclodextrin contains: alpha cyclodextrin has six glucose molecules, beta cyclodextrin has seven and gamma cyclodextrin has eight. As a general rule, small, straight-chain molecules fit best in alpha cyclodextrin cavities; small cyclic molecules fit best in beta cyclodextrin cavities; and large, complex molecules fit best in gamma cyclodextrin cavities. It is this size-selecting feature of the molecules the Ames Lab work seeks to exploit.
The immediate goal of the work is to get the cyclodextrins to line up neatly on a substrate, such as a gold film, with all their cavities pointing skyward. One way to make the cyclodextrin stick to gold is to add sulfur atoms to the bottom of the cyclodextrin molecule.
"The affinity of sulfur for gold and other metals has been known for a long time," Ho explains, and chemists have recently found many ways to exploit this affinity. If the sulfur is stuck directly to the cyclodextrin molecules, however, the cyclodextrin doesn't form a nice monolayer. Some molecules point skyward, but others aim off at an angle. As a result, guest molecules can't glide easily into all of the cavities, and the chemical selectivity of the monolayer is reduced.
The monolayer is imperfect because the sulfur atoms don't deposit at random on the gold surface. Instead they have certain preferred sites, and there is a mismatch between the spacing of these sites and the dimensions of cyclodextrin molecules with sulfur feet.
Ho solved this problem by building a cyclodextrin molecule with legs. The legs, which end in sulfur feet, are composed of ethylene (-CH2CH2-), which makes them long enough and flexible enough that all of the feet can be planted in preferred sites. Unlike other chemical conversions the group has tried, this one consistently produces well-ordered monolayers. And because the layers are nearly perfect, they are maximally selective.
To demonstrate selectivity, Ho coated piezoelectric resonators with alpha cyclodextrin and with beta cyclodextrin. When the resonators were exposed to hexane, a small molecule, their resonance frequencies decreased by the same amount. When they were exposed to tetrachloroethylene, a medium-sized molecule, the resonance frequency of the resonator coated with beta cyclodextrin dropped more than the resonance frequency of the resonator coated with the alpha cyclodextrin. The tetrachloroethylene couldn't fit in the alpha cyclodextrin's cavity.
Dr. David Walt, Chairman of the Chemistry Department at Tufts University, says, "Although palladium-based sensors have been used many times for looking at hydrogen, this is the first application that I know of where there's a separation between the bulk versus the surface effect. And although there have been quite a few attempts, both successful and unsuccessful, at using cyclodextrins in sensors, the ability to engineer cyclodextrin so that it forms ordered monolayers is a significant advance. Both of these sensors are refinements on previous concepts, but those of us in the sensor community know that what it really takes to get sensors to work are refinements. Dr. Porter's group has done what is necessary to get these designs to the next level of development."
For more information contact:
Marc Porter, 515-294-6433
porter@ameslab.gov
Imagine sitting down at a computer and designing anything from a coffee cup to a car -- all to your personal specifications. Then imagine being able to send this personalized design instantly to a manufacturer anywhere in the world, where a variety of computer-controlled tools, machines and robots create unique components from raw materials, assemble the parts and ship the customized product to your home, all in the same day.
Sound like science fiction? To some researchers at Ames Laboratory, it's more science than fiction.
"The basic technology already exists to do all this and more," says Ames Lab Associate Metallurgist Tim Ellis. "There just needs to be some fine-tuning of the processes involved to make it a reality." The process that Ellis and Senior Metallurgist Rohit Trivedi are helping to fine-tune is stereolithography (SLA), a method of rapid prototyping.
Stereolithography is best suited to creating many unique components rather than hundreds of thousands of copies of one component, as in mass production systems. In a typical system an ultraviolet laser is fired into a tub of light-sensitive liquid acrylate polymer. Also in the tub is a moveable, computer-controlled stage, or table, which is positioned just under the surface to cover it with a thin layer (usually less than 0.1 millimeter thick) of the polymer. As the laser follows a cross section of the object, the polymer is set, or solidified, wherever the laser is aimed; the rest remains liquid. Once one layer is set, the table descends just enough to be covered by another layer of liquid exactly the same thickness as the next slice of the object's computer model. By this means, a solid, three-dimensional object is created layer by layer.
This is not the end of production, however. The new polymer object is used to create a piece of tooling, such as a mold or a die, which in turn is used to produce the desired parts. "The problem is that the material the prototype is made of is garbage," says Ellis. "This polymer is relatively gooey stuff; at 200 degrees Celsius (392 F) it just melts into a lump of garbage and loses all dimensional shape and form."
The polymer's low melting temperature limits the types of materials and processes that can be used to make the molds. Many industries spray-form molds because the high cooling rate of this process allows relatively low-temperature application. Also, a low-melting-point metal alloy can be used to ensure that the dimensions and shape of the original polymer part are accurately reproduced.
But Rohit Trivedi says low-melting-point metal alloys must be used with caution; they affect the quality of the mold. "There's some contradiction," says Trivedi. "You want an alloy with a very low melting point so as not to melt the polymer, but once you form a mold you want it strong and able to withstand higher temperatures.
"One of the problems with existing technology is that the alloys being used to make the molds are not giving consistent results, and sometimes the molds fail after only a few runs," Trivedi says.
"What we are trying to do is develop new materi- als but use the same technology," says Trivedi, who this year received the Minerals, Metals and Materials Society's Bruce Chalmers Award for his contributions to the field of solidification science. He adds that Ames Lab is the only facility in the nation concentrating on improving SLA through a materials approach.
A common material used in the spray forming of molds is zinc-aluminum. "The problem is that zinc is a very soft metal," says Ellis. "It doesn't have very good mechanical properties, so the molds don't last very long - maybe ten or a hundred shots, and that's it."
"The zinc-aluminum alloy is a low-melting-temperature alloy but has a couple of problems," says Trivedi. "One is an oxidation problem, which can lead to cracking in the mold. The other is that it's not very strong. So we designed a new alloy that is twice as strong and doesn't oxidize."
Trivedi and Ellis have applied for a patent on their new zinc-aluminum alloy, which they doped with some rare-earth metals. "The alloy we've developed may allow a thousand or two thousand shots per mold without breaking. There's a large number of products or parts that are needed in this country with less than ten thousand copies," says Ellis.
Another way to make the molds last, adds Ellis, is to form the original component from a stronger material; a stronger component, in turn, should allow stronger molds to be made that would have better mechanical properties.
To pursue this possibility, the Ames Lab team built their own stereolithography machine. "Ours is a very versatile piece of equipment," says Ed Wanat, a research assistant at the Lab who helped assemble the laser, computer and other equipment that make up the SLA machine. "We could have bought one commercially, but it wouldn't have had the flexibility we need to perform all the tests with a large variety of materials in different forms and phases."
The first materials to be tested will be reactive gases. "We are trying to look at about 15 or 20 different materials, some copper-based alloys, some iron-based alloys and many organometallic compounds," says Trivedi.
"We can get every element of the periodic table in the form of a reactive gas," Ellis adds.
The reactive gas will be hit with two lasers rather than one. The energy at the lasers' intersection will be exactly the amount needed to break the specific bonds between the metal and the organic molecules. The metal atoms will be deposited, while the rest will be removed as a gas. "The lasers are able to act like a scalpel and surgically break only the bonds between the atoms we wish without affecting the other bonds. Breaking those requires different energies or wavelengths," says Ellis.
If successful, the modified stereolithography process may be used either to fabricate entire components or to build only the component's outer shell, depending on the component's size. "The critical part is to form the shell, which is next to the mold," says Trivedi. "We will use the laser to form the shell and then back it up with specially designed alloy powders, which will fuse and become very strong."
The key to that approach, says Trivedi, is choosing a powder composition that holds up well to high-temperature processes for producing molds.
Trivedi adds that no matter what technique or material is used to make stronger molds, he is confident their research is headed in the right direction. "We feel that with so much advancement in materials, this is where the breakthrough will come from. By focusing on materials innovations, we might get excellent technology in an economical way."
And while you may not yet be able to have your custom-designed coffee mug or car, you may soon see a greater variety of product styles available. "In a larger sense, the whole drive and focus is to bring product differentiation to the individual consumer," says Ellis.
For more information contact:
Tim Ellis, 515-294-1366, ellis@ameslab.gov
Rohit Trivedi, 515-294-1189, trivedi@ameslab.gov
It consists of lipid molecules that look like miniature tadpoles, with a head group at one end and a tail at the other. The tadpoles are arranged in a double layer with their tails in the center and their heads facing outward. Embedded in this bilayer are occasional squat protein molecules, like frogs among the tadpoles.
This oddball structure is one on whose proper functioning life depends. It selects which of the many substances found in living organisms may pass into the cell and which may not. All of life is based on the ability of molecules to organize themselves into supramolecular structures like the cell membrane.
A new instrument called a liquid-surface x-ray reflectometer gives Ames Laboratory scientists unprecedented ability to study these structures. The reflectometer, one of only a handful in the world, was designed by Mechanical Design Engineer Terry Herrman to the specifications of Associate Physicist David Vaknin. It is being used primarily to probe the structure of organic monolayers on water and of supramolecular structures built of these monolayers.
If chemists can learn how to make supramolecular structures, they will be able to tap powerful biological processes, such as self-assembly and molecular recognition. The harnessing of these processes could change chemistry as radically as the harnessing of coal tar in the nineteenth century.
Coaxing a collection of organic molecules into forming a well-defined structure is not easy, however. One way to make the problem more manageable is to shave a dimension. It is easier to understand a two-dimensional structure than it is to understand a three-dimensional one. A second way is to use water as the supporting surface. Water has an extremely clean surface, and it can be made almost perfectly flat. Barring levitation, layers of organic molecules on water are as close as one can get to layers in isolation.
But there's a hitch. If you spread a layer on water, how do you study it? A single layer of lipid molecules on water is invisible to the unaided eye. And the many methods for probing the surfaces of solids at the molecular level, such as x-ray photoelectron spectroscopy, secondary-ion mass spectrometry and low-energy electron diffraction, are ill-suited for studying the surfaces of liquids. This is where the liquid-surface reflectometer comes in.
Using a pocket laser, a piece of a silicon wafer and a sheet of white paper, Vaknin demonstrates how it works. A nearly horizontal pencil of laser light, reflected off the polished silicon surface at a grazing angle, forms a small red spot on a sheet of white paper. "Now, watch this," Vaknin says, squeezing a few drops of ethanol from a washbottle onto the silicon. As he trips the overhead light, the rounded blips of an interference pattern magically appear above the red spot. For a moment, the pattern holds steady, but then the blips begin to dance -- sliding, merging and re-emerging. Seconds later they vanish, leaving only the red spot.
"You get the pattern because light reflected from the ethanol-silicon interface interferes with light reflected from the ethanol-air interface," Vaknin explains. "From the spacing of the interference fringes, you can determine the thickness of the ethanol film. The pattern jumps around because the ethanol is evaporating and the film is getting thinner. It disappears when the film is gone.
"An x-ray reflectivity study of a monolayer is based on the same physics," Vaknin continues. "The x-rays reflected from the air-monolayer interface interfere with those reflected from the monolayer-water interface. The difference is that a monolayer is much thinner than the ethanol film. The ethanol, at a guess, is thousands of nanometers thick, but a true monolayer is the diameter of one molecule, which means it is typically between one and six nanometers thick.
"To measure an object, you need the right size ruler," says Vaknin. We wouldn't see anything if we bounced the laser light off a monolayer, because the laser's wavelength is too long. The reflectometer operates at a wavelength of 0.15 nanometer, however, and you can measure a distance of one nanometer in units of 0.15 nanometer."
The geometry of x-ray reflection also differs from that of reflection at visible wavelengths. Consider a diver submerged at night who is trying to flash a light at a friend in a rowboat. If the rowboat has drifted away from the diver so that he must shine the light at a shallow angle to the surface (48 degrees or less), all of the light will be reflected back into the water. Even if the beam is pointed at the rowboat, the friend will see nothing. But if the rowboat is nearly overhead and the angle between the beam and the surface is closer to the perpendicular (larger than 48 degrees), some of the light will cross the surface, and the friend will see it.
For x-rays the situation is flipped; radiation arriving through the air rather than through the water is totally reflected. And, if the wavelength is 0.15 nanometer, the critical angle is 0.15 degree rather than 48 degrees.
Then there is an additional complication. The variations in reflectivity caused by the interference of light reflected from different depths show up only at angles larger than the critical angle, but the x-ray reflectivity also plummets at these angles. "At 3.5 degrees, which is only 3.35 degrees above the critical angle, only one out of every 10 million particles of light, or photons, that strike the surface is reflected back," says Vaknin. "For this reason, we spend most of our time collecting photons at angles between 0.1 degree and 3 degrees of the surface. This is the whole range of angles over which we are measuring reflectivities.
"Just to give you some idea how shallow the angles are," he adds, "even though the incident beam passes through a slit only 0.1 millimeter wide, its footprint on the surface is typically about 40 millimeters long. The lower the sun on the horizon, the longer your shadow becomes, and our 'sun' is very low on the horizon."
The technology of working at grazing angles to liquid surfaces is not trivial, which is why until recently physicists measured the reflectivity of liquids only at the critical angle. The first reflectometer capable of making measurements of liquid surfaces at a wider range of angles was built in 1982 by J. Als-Nielson of the Riso National Laboratory in Denmark and Peter S. Pershan of Harvard University. In 1989 Als-Nielsen asked Vaknin, who had been working with neutron scattering at Brookhaven National Laboratory, to come to Denmark to build a neutron reflectometer. "We built the first fixed-wavelength neutron reflectometer, and that's how I was introduced to reflectivity," says Vaknin.
The Ames Lab reflectometer may be only months old, but it has already been used to study several different supramolecular structures. For example, Vaknin is currently doing reflectivity studies of films that may eventually be used to create light-collecting antennae for artificial photosynthetic systems. His studies assist research on these devices being done by Walter Struve, a senior chemist at Ames Lab and professor of chemistry at Iowa State University(ISU), Brian Gregory, a postdoctoral fellow in the Fundamental Interactions Program at Ames Lab, and Therese Cotton, an Ames Lab associate and professor of chemistry at ISU.
In its current form the light-harvesting array has two layers: a lipid layer and a layer of a water-soluble pigment, such as a charged porphyrin. Porphyrin is a compound closely related to a stuctural component of chlorophyll. The idea is that the porphyrin molecules will absorb visible light, transforming it into electronic excitation. The molecules will then pass this energy like a bucket brigade to a central site, where it will fuel chemical reactions.
The lipid layer, which rides the interface between air and water, immobilizes the porphyrin layer by holding it at the interface. The headgroups of the lipids, which are negatively charged phosphate groups, dangle into the water. The porphyrin molecules, which are flat squares like floor tiles, have four positively charged nitrogen atoms at their corners. The attraction between the negatively charged phosphate groups and the positively charged nitrogens overrides the porphyrin's natural tendency to aggregate and induces it to spread out under the lipid instead. One of the questions the reflectivity studies should answer is whether the porphyrin molecule aligns itself with the lipid layer flat side up or edgewise.
Recently, pushing his reflectometer to its limits, Vaknin has been looking at the interference pattern created by the molecules in the lipid layer rather than by the interfaces between molecular layers. Interference patterns of this type can be interpreted to yield the spatial arrangement, or packing, of the molecules in the layer. Although Vaknin emphasizes that the resolution is poor, what he sees is consistent with the standard lipid packing arrangement, in which the molecules form staggered rows.
Vaknin will soon be able to do this more delicate kind of study with considerably less effort. The reflectometer was designed for eventual installation at the Advanced Photon Source, a new synchrotron facility at the Argonne National Laboratory near Chicago that is dedicated to the production of extremely brilliant x-ray beams for research. "Our x-ray source at the Lab is just a regular rotating anode," Vaknin explains. "We get 105 or 106 good photons per second. At the synchrotron, we'll get 1010 good photons per second, which is a huge improvement. With the anode, we can see only the gross changes from one layer to the next. At the synchrotron we will be able to look routinely at the structure within a single layer."
"Dr. Vaknin's project for studying liquid surfaces with x-ray and neutron reflectivity is an excellent example of a recent approach to experimental studies of liquids and their interfaces," comments Pershan, who is the Gordon McKay Professor of Applied Physics at Harvard. "These techniques are currently leading to new understanding of age-old phenomena."
For more information contact:
David Vaknin, 515-294-6023
vaknin@ameslab.gov
Last spring Ames Lab educators became matchmakers when they introduced a group of students from the Ames Christian School to three-dimensional geometry on a super-computer. Fascinated, the students fell for it, and the relationship began.
The students were participating in the Lab's SuperKids program (see Inquiry, Spring 1993). Taking advantage of the interest kids have in computer technology, SuperKids helps students acquire and sharpen their skills in coordinate geometry through exploratory activities with super-computers.
Special animation software called Wireman lets students view and manipu-late objects in three-dimensional space on a computer screen, which helps them gain a more practical knowledge of how to work with geometric shapes and the Cartesian coordinate system (x, y and z axes).
"SuperKids familiarizes students with supercomputers and provides an academically challenging and creative way to improve math learning skills," says Chris Ohana, acting coordinator for the Lab's Office of Educational Programs. "Students experience personal success as their computer skills and understanding of three-dimensional geometry increase and they are able to produce their own computer-animated 3-D movies."
Helping students improve their abilities to communicate mathematically and understand spatial relationships, SuperKids promotes the standards of the National Council of Teachers of Mathematics (NCTM), a group that fosters change in mathematics teaching.
Revamping SuperKids to better implement the NCTM standards, Lab educators changed the program from a weekly, Saturday morning event with an extracurricular flavor to a two-week session with daily meetings. The new format allowed the class of upper elementary and middle school students from the Ames Christian School to experience SuperKids as an integrated part of their math curriculum.
"Because we had the students five days a week for three hours a day, we were able to take this group a lot further than we've taken other groups, both in terms of the activities we did and, more importantly, in terms of the concepts we covered," says Kelli Kerry, a Lab graduate assistant and an instructor for the SuperKids program. "Meeting every day also gave the students a lot more time on the computer."
"Continuity is a big factor," adds Rich Valde, also a Lab graduate assistant and an instructor for the SuperKids program. "With the daily program, the students don't have that week break where they might forget what they did the time before."
Ohana notes that the revised SuperKids program also incorporated a teacher-training component. "Since we were working with a single classroom, the teachers assisted to allow for greater reinforcement of learning. This wasn't the case in the past when we had students from different classrooms and schools."
Perhaps the most exciting aspect of the new look for SuperKids was the emphasis on hands-on activities to help the students visualize complex geometry concepts and transfer those concepts to computers.
In one activity, the students formed a human grid and played a navigation game with balls to sharpen their skills in using the x, y and z axes to move within an environment. The activity also encouraged students to communicate mathematically by requiring them to use the language of the Cartesian coordinate system.
While one student called out a series of x, y and z coordinates, the other students responded, moving the ball to each point in the grid. To incorporate the concept of changing perspective, Kerry and Valde had the student who was calling directions lie down on the floor. The x, y and z axes were no longer the same for the stu-dents forming the human grid as they were for the student lying on the floor, and they had to adjust their perspective to match his.
The caramel construction game was another activity that promoted the use of x, y and z terminology and movement along the x, y and z axes. Students were given 10 caramels each and then divided into teams of two. One person in each team built a caramel structure, keeping it shielded from their partner. Using the language of the Cartesian coordinate system, the builders for each team told their partners how to create the structure. After removing the shields, the teams saw how successful they had been.
Kerry and Valde agree that the increased use of hands-on activities enhanced the SuperKids program. "The activities helped us see where any misconceptions were and how we could better direct the students," says Kerry. "The computer is wonderful, and the Wireman program is fantastic for teaching three-dimensional geometry, but the computer is still a two-dimensional environment simulating a three-dimensional environment," she reminds us. "Without the activities, you could miss what the students are not understanding."
Valde adds, "It's exciting that a program like SuperKids is available so students can start building in their minds an understanding of what happens as you move around objects and how they might look from different perspectives. The hands-on activities help students as they go to the computer, but it bridges back the other way too. What's happening on the computer applies to the real world and how things exist."
On the wall above John McClelland's desk proudly hangs a drawing of Alexander Graham Bell's photophone. If not for Bell, McClelland might not be where he is today.
"Bell was interested in the photoacoustic effect as a means to communicate using light waves, but he also recognized it could be used for optical spectroscopy, which is what we use it for today," says McClelland, an Ames Lab physicist.
McClelland has transformed Bell's idea for spectroscopy into his own business, MTEC Photoacoustics. MTEC produces McClelland's award-winning device, the photoacoustic detector, which uses light and sound for the chemical analysis of solids in Fourier-transform infrared (FTIR) spectrometers.
During analysis, a modulated infrared beam from the FTIR spectrometer strikes the sample. The light is absorbed and produces heat that alters the pressure of the surrounding gas. The changing pressure is detected as sound by a microphone and is then converted to an absorption spectrum, which scientists read to identify and quantify the sample.
The photoacoustic detector was developed when McClelland noticed a need to improve photoacoustic spectroscopic instrumentation. "Initially we were trying to do our spectroscopy in the visible and ultraviolet region of the spectrum, which technically is quite difficult to work in," says McClelland.
In that region, says McClelland, the absorption coefficients for the photoacoustic signal are usually quite high, which makes it very difficult to notice any change in the coefficients. "But if you move into the mid-infrared region, the coefficients are lower," he says, which makes it easier to notice changes and provides an area "rich in in-formation" for chemical spectrometric analysis.
But before he could take full advantage of this phenomenon, McClelland had to face a problem similar to the one Bell encountered about 100 years earlier. "Alexander Graham Bell started out with the photoacoustic technique, but the instrumentation just wasn't good enough to do anything. The type of instrument that we needed was not available commercially, so in order to do our research we had to develop the equip-ment," says McClelland.
The first photoacoustic detectors were designed and built with a little help from private industry. In the early 1980s, McClelland, then also at Ames Lab, was doing a little research and development work with IBM. "They wanted to buy photoacoustic detectors, but there really weren't any good ones available," he says. "So I designed our detector and went to IBM, and they ordered 10."
Realizing the commercial demand for his improved detector, McClelland used some of his own money and produced 25 instead of only the 10 IBM had ordered. "Other companies were starting to get interested because we had really improved on the performance of previous detectors," McClelland says.
His new design was recognized by Research and Development Magazine as one of the 100 most significant new technical products of 1985. This award went to both Ames Laboratory and McClelland's newly formed MTEC Photoacoustics.
But the infant company needed help, which it found at Iowa State University. "With MTEC and their need for space outside the laboratory and access to support facilities, we were able to convince the university to provide space for a business incubator," says Dan Williams, division director of Planning and Technology Application. MTEC became the first occupant of ISU's new, on-campus business incubator in 1985. "I felt that type of support system for spin-off businesses was critical," says Williams.
Now fully self-reliant, MTEC boasts over 700 units sold worldwide to a broad variety of industries, including the automobile, electronic, pharmaceutical, textile, paint and coating industries.
Such wide industrial contact allows MTEC to give back to the Ames Lab that spawned the company. "Through customer inquiries and interactions we've learned about research opportunities that the Lab otherwise wouldn't have known about," says McClelland.
When the space shuttle blasts off in July, Ames Lab scientists will be closely monitoring the mission. The shuttle will be carrying an experiment designed in part by Rohit Trivedi, an Ames Lab senior scientist and an Iowa State University (ISU) distinguished professor of engineering.
The experiment, housed in a "smart can" inside the shuttle's cargo bay, will assist scientists in understanding how particles in a hot liquid metal behave as the metal cools and solidifies in microgravity.
The experiment consists of four crucibles, each containing an ingot of a tin-cadmium alloy in which the particles are evenly distributed. The metal ingots will be melted in space and then cooled. Once the shuttle returns from space, the distribution of the particles within the resolidified ingots will be examined by ultrasound and by optical and electron microscopy. Because particle arrangement affects a material's mechanical properties, the information gleaned from the experiment could lead to the development of stronger and tougher metal composites on Earth.
Additional support for the project was provided by ISU's Institute for Physical Research and Technology, the Iowa Space Grant Consortium and the University of Iowa.
Ames Lab's commitment to Cooperative Research and Development Agreements (CRADAs) continues to grow. The Lab is currently involved in 14 such agreements, representing approximately $3.5-million-worth of research and development effort. Additional CRADAs are in various stages of negotiation.
Two recent CRADAs with American Superconductor Corp. (ASC) will use the Lab's materials expertise to develop ways to improve the performance of high-temperature superconducting wires and tape. One project, which focuses on reducing energy losses associated with alternating current, will help ASC, the Electric Power Research Institute, the Pirelli Cable Corp. and other DOE national labs build a prototype high-temperature superconductor cable for underground power transmission.
Developing new materials for use in a variety of automotive sensors is the goal of an Ames Lab CRADA with General Motors Corp. "The thrust of this research is either to discover new materials with desired properties or to optimize existing ones," says Paul Canfield, an Ames Lab physicist working on the GM project.
Technology developed under the CRADA could also have application in other fields, including electricity generation and temperature control.
Winning top honors at Ames Lab's Adventures in Supercomputing (AiS) Expo in April just wasn't enough for an all-girl team from North Polk High School in Alleman, IA. In June the girls proved they were the best in the nation when they captured first place for their supercomputing project on particle physics at DOE's National AiS Expo in Washington, D.C.
Barbara Helland, Ames Lab acting program director for Applied Mathematical Sciences and the 1995 national AiS coordinator, says the North Polk team went to Washington, D.C., well prepared for the national competition. "It was a very close competition, but the students worked hard and did well, gaining a much greater awareness of computational science in the process. They have become more technically and scientifically literate and so are better prepared for the future."