The Breadth of Light

“No, Mr. Bond. I expect you to die,” said Auric Goldfinger as a steel-melting laser inched closer to 007’s favorite anatomical region. This secret-agent-slicing beam would probably only need to be a few thousand watts of continuous power, according to Martin Centurion (PhD ’05), a postdoc in Caltech’s Center for the Physics of Information, but the femtosecond lasers he works with routinely put out an unimaginable 10 gigawatts of pulsed power. (For Dr. Evil’s benefit, that’s ten BILL-ion watts.) It’s not that the lasers have gotten bigger—in fact, they’re quite a bit smaller these days—but power equals energy divided by time, and a femtosecond is 10-15, or one quadrillionth, of a second.

These ultrafast pulses might be ideal for communications and switching systems, and even optoelectronic computers, but for one serious drawback—the sheer intensity of the light induces a nonlinear phenomenon called the “Kerr effect” that alters the refractive index of the material the beam is passing through. The lasers used in fiber-optic systems are way too weak to be subject to this, but once you cross a certain threshold, the brighter the beam, the more the refractive index changes. The light consequently focuses inside the glass, heating the atoms along its path to a plasma—a fog of free electrons and ionized silicon and oxygen atoms. This might actually be a good thing, in that the plasma’s refractive index is negative and prevents further self-focusing, but forming the plasma drains the beam’s energy. And, of course, there’s the unfortunate side effect of eventually vaporizing the fiber.

A theoretical fix, called “nonlinearity management,” has been around for a decade, and now Centurion; fellow postdoc Mason Porter (BS ’98); Demetri Psaltis, the Myers Professor of Electrical Engineering; and mathematics professor Panayotis Kevrekidis of the University of Massachusetts at Amherst have actually demonstrated it. The basic idea is to alternate stretches of a material that focuses light of the given intensity with one that causes it to spread at that same intensity. As the beam passes through this “Kerr sandwich,” it alternately expands and contracts—it “breathes,” if you will—and the beam size remains relatively constant overall.

“Basically, the smaller the beam is, the faster it will expand in the air,” says Centurion, “and the higher the power, the faster it will focus in the glass, so you can play with these parameters to reach a balance. In our case, the beam diameter doubles after about four millimeters in air.”

The sandwich consisted of nine ordinary microscope slides, each about one millimeter thick, placed parallel to one another at one-millimeter intervals. The slides were used not only because they were handy, says Porter, but because “we also wanted to indicate that it didn’t need any special materials or circumstances to work.” The laser, focused to a beam less than 50 microns (millionths of a meter) in diameter, was shot in pulses of 160 femtoseconds each. If these pulses had gone through solid glass, plasma formation would have kicked in after about two millimeters. Instead, it emerged from the sandwich with essentially the diameter it had when it entered, with no plasma formation. It did lose more than half its power, however, due to internal reflections at each air-glass interface; further studies using slides with a nonreflective coating are already showing better results.

The work was published in the July 21 issue of Physical Review Letters. —DS