Down to Earth

Victor Viola, Distinguished Professor of Chemistry, and Kris Kwiatowski, Senior Scientist in Chemistry, both at Indiana University Bloomington, and Lai Wan Woo (right to left), who served as a computer consultant in the design of the Indiana Silicon Sphere detector array (ISiS), are in front of the central control panels at the Indiana University Cyclotron Facility. In the foreground, the monitor screen shows a line drawing of the ISiS detector assembly. --credit

Although it might seem paradoxical to study matter on an atomic scale to learn about a stellar system that can possess a mass millions of times greater than Earth, Viola and Kwiatkowski's studies on nuclei provide important data that might someday help describe how exploding supernovae turn into neutron stars--stars so fantastically dense that they can contain the mass of the sun within a ten-kilometer radius. These stars are the likely source of pulsars, oscillating intergalactic radio signals detected by astronomers. Viola and Kwiatkowski are specifically interested in the phase transition that occurs when the gaseous nuclear matter at the core of an exploding supernova condenses into liquid, resulting in the formation of a neutron star--in effect, a gigantic nucleus composed primarily of neutrons. Given that direct observation of neutron-star formation in the core of an exploding supernovae is out of the question (the closest examples to date are 150,000 light years away in the neighboring Magellanic Clouds), the conundrum is obvious. How do you recreate the gas-to-liquid condensation phase of an exploding star in a lab?

"The answer is you can't," Viola says. "The necessary conditions are too extreme. It just wouldn't work." Viola's approach to the problem hinges on the fact that atomic nuclei behave very much like a liquid. "Their protons and neutrons hang together much like a microscopic drop of water," he explains. "Scientists who study hydrodynamics have an equation of state (EOS) that describes the heating and eventual vaporization of water. In a very simple sense, if you have a pan of water heating on the stove, the EOS describes how much energy it takes to get to the boiling point. Presumably, the liquid-to-gas phase transition of atomic nuclei is a similar process in reverse. We can learn about one from the other."

Viola and his team are attempting to determine the EOS for hot nuclei by studying the conditions under which nuclei expand and vaporize. By bombarding a gold nucleus with simple, projectiles such as protons and helium-3 particles, they can heat it to temperatures more than a thousand times that found in the sun's interior. This intense heat causes the nucleus to expand, explode, and subsequently fragment into many smaller clusters--the opposite of the gas-to-liquid phase transition that occurs in exploding supernovae, but, Viola presumes, is theoretically similar and reversible. The main part of this work was done at the Saturne accelerator in Saclay, France, and at the Alternating Gradient Synchrotron accelerator at Brookhaven National Lab in Long Island, New York.

The Indiana Silicon Sphere detector array (ISiS), a device for measuring type, energy, and direction of explosion-like nuclear events - credit

Much as forensic scientists reconstruct a bomb explosion from the resulting debris, Viola and his team reconstruct nuclear explosions. They study the fragmentation process, recording data on the energies, directions, and sizes of many fragments to learn as much as possible about the original environment that created the event. The results of a series of such experiments, when announced several years ago, provided the first direct evidence of the expansion of nuclear matter and offered theorists working on models of phase transition a significant piece of evidence. "There are two theories and several theorists who have predicted that intense heating could cause nuclear expansion, but it has been very difficult to prove experimentally," Viola says. "There have been many inferences about this process going back twenty-five years, but the instrumentation just wasn't available."

The instrument that finally made it possible to confirm nuclear expansion, the Indiana Silicon Sphere, was made to order for the type of nuclear reactions Viola and Kwiatkowski want to observe for their nuclear EOS research. At first glance, the sphere looks as if it could play a role in a 1950s Hollywood sci-fi film, largely because its spherical casing is riddled with cables Viola jokingly calls "spaghetti." The cables connect each of its 162 telescope-like detector devices to electronic data processing modules, which collect and eventually transfer the data to computers. The sphere itself consists of eight rings, each composed of eighteen telescope housings or "cans" (a component the scientists call a "telescope" even though it is used like a microscope--to view subatomic objects). "Getting seventeen of these cans to fit together was easy, but we needed eighteen of them to fit perfectly in each ring," Viola recalls. "That wasn't so easy. The precision required in constructing this device was on the order of a few thousandths of an inch."

Each component serves to collect electrons as they are knocked off the atoms as the nuclear particles pass through them. The first is composed of a freon-like gas (not an ozone destroyer), the second is a wafer of silicon about 0.02-inch thick (material for computer chips and responsible for the ISiS sensitivity) , and the third is a crystal of cesium iodide about 1-inch thick (heavy table salt). These materials work in combination to provide three separate electronic signals, enabling the researchers to identify each detectable fragment's atomic number, its kinetic energy, and the angle relative to the beam at which it was ejected from the hot nucleus.

Gary Fleener, Shop Supervisor John Dorsett, and Kenny Bastin (l­r), from the Indiana University Chemistry Mechanical Instrument Services, display the components made in their shop that were used on the Indiana Silicon Sphere detector arrays (ISiS). The arc bars (foreground) were developed with a computerized numerical control milling machine. The cans were all formed, bent, and welded by hand, using an inert gas welding process. --credit

In an experiment, a nuclear particle accelerator beams high-speed particles to a target nucleus positioned inside the sphere. During the bombardment process, the target is surrounded by the regularly spaced telescopes. "This means we can look at all the pieces of the nuclear explosion at one time," Viola explains. "Before we had this detector, we could only infer that nuclear expansion was taking place by looking at only a few pieces of evidence at a time. That would be analogous to a forensic scientist trying to reconstruct a bomb explosion from one piece of debris. We've been building silicon telescopes for a long time, but what has changed is the ability to put lots and lots of these silicon pieces together, and, subsequently, to have the power to analyze very complex results." Once they have plotted and analyzed the data, Viola and Kwiatkowski have the information they need to reconstruct the complex sequence of events that caused the target nucleus to expand and burst into fragments. It is a small but important piece of the puzzle that might one day link the behavior of exploding nuclei on Earth to exploding supernovae in space.

Viola is quick to clarify that the detector array, however sophisticated, would be useless without the extraordinary leaps in computer technology that have occurred in the last fifteen years. "We understood how to build this detector almost twenty years ago, but we simply couldn't have handled the amount of data it produces. The detector can give us five hundred pieces of information simultaneously, all of which are taken on a time scale of a microsecond. So, we have many, many thousands of events recorded per second," Viola explains. "Without the increasing speed, memory, and low cost of computers, we couldn't possibly manage the data."

It is clear from Viola's recollections that designing and building the silicon sphere was in some ways as exciting as the confirmation of the nuclear expansion phenomenon. "It was a nerve-wracking experience for all of us, particularly for the students who had their Ph.D. theses riding on this very complex detector," he recalls. "No one knew how it would work."

Viola gives enormous credit to project manager Kwiatkowski, whose design maximized the efficient use of silicon--the expensive, high-precision detection material that gives the apparatus its special detection capabilities. He also credits instrumentation technicians like John Dorsett and his staff in the Department of Chemistry Mechanical Instrument Shops. Dorsett's shop machined the telescope cans with such precision that all eighteen do fit in each of the rings that form the sphere--a feat that would have cost the project an incredible amount of money in an outside shop. The group also benefitted from the expertise of the Chemistry Electronics Instruments shop and the technical staff at the Indiana University Cyclotron Facility. But some parts could not be made on site. Consequently, large amounts of Kwiatkowski's time and energy went into locating quality vendors who could manufacture the required parts with an extremely high level of precision. "It was a very frustrating experience," Viola recalls. "When you do something like this, you learn why the military overspends so much. You discover that you can't always depend on vendors." One advantage that came out of the project, Viola observes, was computerized machining capability in the Department of Chemistry machine shop.

Compared to the half-dozen other detector arrays in existence, the ISiS was unquestionably the least expensive to build. "We built ours for about $700,000. The French counterpart, which has some advantages and some disadvantages compared to ours, cost about $3 million," Viola says. "Yet I think most people consider these the two leading silicon-based devices." He makes frequent appreciative references to the team of dedicated and creative students, including Erin Renshaw, Ph.D. '94, now a software engineer at Microsoft; Kevin Morely, Ph.D. '94 and David Bracken, Ph.D. '96, who are now staff scientists at Los Alamos National Laboratory; and former postdoctoral fellow Janusz Brzychczyk, all of whom contributed many ideas, innovations, and long hours to the design and building process. In addition, David Ginger, B.S. '97, now a Marshall and NSF Fellow at Cambridge University, consultant Lai Wan Woo, and chemistry postdoctoral fellows Wen-chien Hsi and Gang Wang have carried on the work of the original team.

Schematic diagram of a hydrogen ion (99 percent the velocity of light) colliding with a gold nucleus. Successive time stops (in seconds) show the initial heating of the nucleus via multiple interactions of protons, neutrons, and mesons in the core of the nucleus, after which the system cools by expanding and emitting particles. Eventually, the expansion leads to an explosion of the hot nucleus. --credit

Viola remembers with admiration how David Bracken thought up an improbable solution to a tough design problem: how to seal the plastic skeleton or cage surrounding the inner sphere containing the target, which must be maintained at high vacuum, from the detectors that are immersed in gas. "This was the part of the design that we thought would give us the most trouble," he recalls. "Whatever covering we used had to be very thin, because, during the explosion, the particles come out of the nucleus in such low energies that a covering with any appreciable mass would stop them. Bracken came up with this ingenious technique of stretching a piece of very, very thin polypropylene over it, which is, essentially, like Saran Wrap."

Viola points out that this kind of mechanical ingenuity is harder to come by in students these days, probably because young people are more inclined to explore computers than to explore under the hoods of their cars, as they did before the computer revolution. "Computers are critical to our work, but they don't necessarily give students the hands-on experience they need for a project like ours," Viola observes. "Computers won't do us any good if we don't know how to design and build experimental devices that give us good data in the first place." While he agrees that computer experience with powerful design software programs can provide students with some design expertise, Viola emphasizes that such programs can't provide essential hands-on experience. "The way to do it, I'm convinced, is through undergraduate research projects," he says. "Give the students access to the laboratory and good equipment and let them work. The more of that we can do, the more technically trained people we will have."

Today, Viola is thrilled with the continuing success of the Indiana Silicon Sphere. "So far we've been lucky. It's continued to function although we've moved it all over the world. Right now, it's sitting on Long Island." In the summer of 1998, Viola and his team will conduct a second round of experiments begun a year and a half ago at the Brookhaven National Laboratory on Long Island. These studies will extend their research using antiprotons as bombarding particles, which they believe may prove even more effective than protons and helium-3 particles for heating nuclei.

When asked what role he thinks the Linda and Jack Gill Center for Instrumentation and Measurement might play in the future of nuclear research on the Bloomington campus, Viola, a faculty member since 1980 and a former director of the IU Cyclotron Facility, emphasizes that the center is crucial to the future of small group research. "Chemists tend to work in small groups, and our nuclear science group is quite small. We exist in the interface between physics and chemistry," he explains. "There are quite a few small group research projects in areas such as condensed matter physics and, certainly, in biology and geology." Funding constraints, Viola explains, will make it harder for small groups to maintain enough independence and flexibility to take on complex and innovative research projects such as the Indiana Silicon Sphere. "The Gill Center will allow these orts of projects to continue," Viola says.

Future detector arrays, he says, will undoubtedly be funded and designed by consortiums. "It's a kind of loss, I think, that science is going this way--always bigger and bigger. In some cases, it will mean that the people who make the crucial decisions about a research project are going to be far removed from the kind of hands-on ingenuity that enabled a student like David Bracken to solve a design problem with a piece of polypropylene."

For small-group investigations to survive, they will need strong technical support, "not just for the computer, but for electronics too," Viola explains. "We have a great tradition of instrumentation development on this campus. Our project is one, but there are many others--the many successes at the IU Cyclotron, for example. This is quite remarkable for a university without an engineering school. But the computer is taking us into a new age of complexity, where we won't survive without engineering and technical support."

Viola hopes the Gill Center will help to provide the College of Arts and Sciences with crucial state-of-the art technology--in effect, the background that researchers will need to be successful ten or twenty years from now in an increasingly competitive environment. "I think it's very simple," Viola concludes. "Universities that provide the technical back-up for their researchers are the ones that are going to succeed."

Return to the Table of Contents