BC SealBoston College Magazine Fall 2003
current issue
features
prologue
Linden Lane
Q and A
Works and Days
Letters to the Editor
BCM Home
Archives
Contact BCM
Coming Events
. Linden Lane
.

The accidental technologist

.
John Fourkas (left) and Michael Previte. By Lee Pellegrini

A CHEMIST'S FORAY INTO THE FUTURE OF DATA STORAGE

The world owes penicillin to a moldy petri dish, microwave ovens to a melted chocolate bar. Add to those accidental discoveries the X-ray machine, nylon, Teflon, Velcro, safety glass, and cornflakes, and you see that serendipity is a force like gravity or electromagnetism, able to redirect a scientist's work as swiftly as a dam burst alters the course of a river. In the summer of 1999, serendipity paid a visit to the laboratory of BC's Professor John Fourkas. A young chemist whose work on the dynamics of liquids had already won him important honors (an Alfred P. Sloan Research Fellowship and a Camille Dreyfus Teacher-Scholar Award, among them), Fourkas found himself diverted, purely by chance, into a quite different area of research: three-dimensional optical memory.

The field is one of high stakes and huge technical challenges. Unlike the familiar two-dimensional world of CDs and DVDs, in which lasers write and read data on a flat surface, 3-D memory technology will burrow beneath the surface of a disc. Lasers will store data on dozens or hundreds of different planes at different depths, and consumers will have to learn yet another Greek prefix: Beyond kilobytes, megabytes, and gigabytes, 3-D optical discs will store terabytes—trillions of bytes—of data. Hundreds of movies or an entire college library will sit on a disk the size of a CD. That, at least, is the technology's promise. After years of trying, academic and commercial researchers have yet to devise a 3-D optical memory system that is both practical and inexpensive. But the technique that Fourkas and his lab team stumbled across that summer's day may be one of the strongest contenders yet.

It started with a chemistry experiment gone bad. Fourkas and some graduate students were preparing to investigate the properties of deeply supercooled liquids, materials that can remain in a liquid state even at temperatures below their normal freezing point. "What we were looking for was a deeply supercooled liquid that would be ideal to study at room temperature," Fourkas recalls. Trying different materials, he and his students would pepper a sample with luminescent "tracer" molecules, shine a laser through the tip of a microscope, then observe how bits of the sample behaved, using the tracers as a guide. One class of materials they tested was a slightly altered form of phenolphthalein, a useful, if unpronounceable, staple of childhood chemistry sets. The materials were a dud. "After we shined our laser on them for a while, we found that they started to give off light, drowning out the light from the tracers," says Fourkas. "So we abandoned them."

Months later, Michael Previte, then a graduate student in Fourkas's lab, was attempting to fix a calibration problem in the lab's sophisticated microscope and laser apparatus. Needing a material that would shine under the microscope, he pulled out a sample of the phenolphthalein-like substance. "I knew that focusing the laser on one place in the material would create a bright fluorescent spot. So, to help calibrate the microscope stage, I made a diagonal line of bright spots across the screen," Previte recounts. What he hadn't anticipated was that the spots would continue to emit light even when he later scanned them with the laser at much lower power. "For want of a better term, it looked cool," he says.

Previte showed the string of glowing dots to his mentor. Was it useful for anything? "No, I don't think so," Fourkas replied. Only later that night did the significance of the find begin to dawn on Fourkas: A technique that could create fluorescent spots at precise locations and then read them back might find a use in three-dimensional optical memory.

Of the many ingenious approaches that have emerged for storing data in three dimensions, all have major drawbacks. Some methods store complex images in the form of holograms, but they require such fine control over the movement of lasers and discs that they work only under laboratory conditions. A technique that zaps tiny bubbles into fused silica needs an expensive high-powered laser and produces layers that are relatively far apart. Another general approach—creating luminescent dots in organic materials—has been hampered by problems of its own. But, as Fourkas soon learned, he and Previte had just solved several of them.

One by one, the advantages of the phenolphthalein-like materials became clear. Unlike other luminescent materials being tried elsewhere, they worked with relatively low-power, low-cost lasers. They could be made cheaply and easily; the same molecular family includes Devcon 5-Minute Epoxy — which, unbeknownst to the manufacturer, stores data beautifully. And over time the materials held on to the data like, well, glue. Lasers could read the luminescent dots repeatedly without either the background growing brighter or the dots growing much dimmer, a liability of other inexpensive materials. In fact, even after a million reads—a pounding that is unlikely to be administered in actual use—the dots had lost only 10 percent of their luminosity. "Suddenly," says Fourkas, "we realized this material was pretty special."

Over the next year, Fourkas and his crew began to find out why that was so. It turned out that molecules in the material were becoming luminescent because of chemical changes caused by the laser. Those changes occurred only when the molecules absorbed three photons, or particles of light, simultaneously. To guarantee that three photons would arrive at once, the molecules had to be bombarded with many photons over a short period of time. Writing data, then, was a matter of focusing the laser on a certain point and delivering a short but intense burst of light. Reading was different. For those same molecules to give off light again, they needed to absorb only two photons. That meant the data could be read with relatively weak pulses of light that posed no risk of accidentally writing new spots. All of this added up to an approach that was promising enough to earn Fourkas a $344,000 grant from the Air Force Office of Scientific Research in April 2001.

Since then, Fourkas and company have been inching closer to the goal of building a practical 3-D optical memory system. "The most exciting moment for me," says Christopher Olson, a graduate student who worked on the project, "was when I realized how, theoretically, our storage density was orders of magnitude higher than audio CDs, and several-fold higher than today's DVDs." The team has succeeded in writing 25 layers of data in a sample of material, which—scaled up to disc size—would equal 87 gigabytes of data, or 130 times what a CD holds. Eventually, says Fourkas, they should be able to break the terabyte barrier, squeezing 150 layers onto each side of a disc.

But adding layers is just one item on a rather extensive to-do list. To succeed commercially, discs will have to be readable with low-power commodity lasers—to which end Fourkas is mapping out a technique that uses only one photon of light instead of two. Discs will also need to be writable much faster than the present snail's pace of 10,000 bits per second. And, ideally, the discs should be rewritable, which they currently are not. Answers to those quandaries might present themselves as the researchers get a firmer grasp on the chemical changes that cause the materials to fluoresce in the first place. "We have to understand what happens at the microscopic level in order to rationally improve the materials," Fourkas says.

There's one more item on that to-do list: licensing the technology as quickly as possible. Thanks to a paper that Fourkas, Olson, and Previte (now a postdoctoral fellow at MIT) published in the November 2002 issue of Nature Materials, their work has generated a lot of press. Now Fourkas is hoping that some farsighted company will step in and help nurse the technology to commercial viability. It's a question of priorities, he explains: "Of all the projects I'm working on, this may be the one with the broadest potential impact on the public. Yet it's moving toward a realm that's more engineering than science. Once it gets to a certain point, I'd like to hand it off and watch it from a distance." In other words, he is a chemist first, and chemistry beckons.

David Brittan


David Brittan is a freelance writer and editor who lives in Newburyport, Massachusetts.


Photo: John Fourkas (left) and Michael Previte. By Lee Pellegrini


. . .
  » 
. .  
  » 
     
  » 
     
  » 
     
  » 
. .  
  » 
     
  » 
     
  » 
     
  » 
. .  
  » 
     
  » 
     
  » 
     
  » 
. .  
  » 
     
  » 
     
  » 
     
  » 
. .  
  » 
     
  » 
     
  » 
     


.    
  »
.    
  »
.    
  »
.    
  »
Alumni Home
BC Home