Laser Makes History's Fastest Holes

October 07, 1999

SEARCH THE GAZETTE

HARVARD GAZETTE ARCHIVES

By William J. Cromie
Gazette Staff

Laser pulses (red) melt atoms inside a piece of glass as it moves from right to left, creating a channel that serves as a new type of fiber-optic cable. See an animated demonstration below.

Visit the Mazur Group web page
Visit the microexplosions web page.

The problem was to see how small a hole could be drilled inside a piece of glass, that is, without cracking the surfaces.

In the 1970s, Harvard physicists actually drilled such holes by using a microscope to focus a one-trillionth of a second laser burst inside a thin piece of glass. That wasn’t too useful immediately, but the research eventually led to enormous practical applications, from boring and cutting metals and other materials to surgical operations.

Now the experimenters are at it again, this time using laser pulses a scant 100 millionth-billionth of a second (100 femtoseconds). An intense pulse of light enters the glass and causes microexplosions that blow out vanishingly small holes or minute cylinders. These are the fastest human-made holes in history, and they promise amazing new communications, data storage, and surgical capabilities.

"We’re well on our way to making new types of three-dimensional compact disks that store the equivalent of 100 CDs on one, and to writing optical communications devices that will be much faster than electronic devices," believes Eric Mazur, Gordon McKay Professor of Applied Physics.

"Femtosecond lasers also have the potential to be used for a range of medical purposes, from corrective eye surgery to removal of wrinkles," say Chris Schaffer, a graduate student in Mazur’s lab.

Download a Quicktime movie (5Mb, Quicktime plugin required) of the laser system.

Download an animated gif (2Mb, No plugin required) of the laser system.

Harvard patented a system for peppering the inside of a piece of glass with laser microexplosions about four years ago. These explosions leave microcraters, or damage spots, that serve as "ones" in the binary language of digital communication. Undamaged spots make "zeros," allowing data to be stored in much less space than is now possible. These dots can be written in multiple layers (100) compared to one layer on present CDs.

At first, "writing" ones and zeros with the laser was a slow process despite the extraordinarily short pulses. "With the first system we developed, the laser fired only 1,000 times a second; at that rate it would take about 25 years to write a 100-layer CD," Mazur notes. "But Chris [Schaffer] solved this problem recently and we can now do the job in one minute by firing the laser every 40 billionths of a second."

Harvard has patented this 3-D data storage technique. "An increasing number of private companies are interested in the system to create structures in transparent and other types of materials," Mazur points out.

Eric Mazur (left), Gordon McKay Professor of Applied Physics, and graduate student Chris Schaffer adjust a laser system that uses pulses 100 millionth-billionths of a second long to make experimental optical communications and data storage devices, and to do eye and skin surgery.

Glass Blowing

To give you an idea of how short a femtosecond is, there are the same number of femtoseconds in one second as there are minutes in the age of the universe, some 12 billion years. Made of titanium and sapphire, such lasers are built specially in Mazur’s lab. The pulses of dark red (800 nanometers) light they generate are focused into a cubic spot 25,000ths of an inch on the side. These spots can form data patterns inside glass as thin as one 500th of an inch. Individual layers of spots are spaced 2,500th of an inch apart.

Concentrating light energy into a space only 25,000ths of a cubic inch causes atoms in the glass to explode outward, creating a hole inside the glass. Because the light is concentrated in such a minute volume, very little energy is needed. "A typical explosive pulse generates about 10 billionths of a joule, or about the kinetic energy of a flying mosquito," Schaffer notes.

Schaffer and Mazur continue to study the details of exactly what happens, and how the result varies with different materials and laser light frequencies. They have also moved the glass as the laser fires to create a continuous cylinder within glass that can serve as a light channel, or wave guide.

Modulated light pulses moving through a wave guide carry information in the same way that changing voltages moving through a copper wire do. But light traveling through a fiber-optic cable can be modulated much faster that electric pulses.

"Light pulses may be 10,000 times shorter than electric pulses," Mazur says. "That capacity let’s you send 10,000 times as much information in the same time. Optical fibers carrying information across the ocean floor already are much cheaper than satellites that s and receive electronic signals."

At present, however, voice, television, and other types of information must be converted from electronic to light signals then back again because there are no optical switches, amplifiers, filters and other devices needed to make complete optical circuits.

For example, an optical fiber can carry several phone conversations simultaneously. However, when they arrive at a certain point, they must leave the fiber and be split into their respective destinations electronically.

"We are experimenting now with wave guides, written by the femtosecond laser, which intersect each other and will split out conversations carried by different frequencies of light," Mazur notes. "We might also use diffraction gratings to separate frequencies."

"There’s a big push to make communications all optical," Schaffer notes. "And I think we’ll see it in a couple of years."

The femtosecond lasers he, Mazur, and their colleagues built cost $60,000 each, but "costs will fall as commercial applications increase," they say.

Some of the applications are unexpected. The Irish company that makes Waterford crystal is interested in using the technique to engrave its products internally. It seems that workers break lots of glass when scratching the company mark on its surfaces.

Light That’s Skin Deep
Laser microexplosions can be induced in anything that’s transparent, including the cornea of the eye and human skin. (If you don’t think your skin is transparent, try shining a flashlight through the back of your hand.)

"We can focus light about one-hundredth of an inch below the surface of mouse, pig, or human skin," notes Schaffer.

This capability opens the door for using the laser to remove port wine stains, liver spots, and tattoos. Like other lasers, the femtosecond light should successfully treat unsightly bluish leg veins, benign tumors of blood vessels (hermangiomas), and be useful for removing hair and layers of wrinked skin.

Lasers presently employed for these jobs need to blast away the top layer of skin to achieve their cosmetic or medical goal, leaving a person prone to infection. But the femtosecond pulses can be focused right through the surface. Also, a different type of laser is now required for each removal or resurfacing job. "With the femtosecond laser, one light source could do all those jobs," Schaffer boasts.

Another possibility involves shaping the transparent cornea to correct near- or far-sightedness. Over the cornea at the front of the eye lies a clear, thin covering that is easily damaged, causing discomfort and raising infection risk. To avoid such problems, eye surgeons separate this layer, pull it up, lase the eye, than lower the flap of tissue back in place – a technique known as "flap and zap."

Because the flap is transparent, a femtosecond laser would shine right through it, so ophthalmologists at the University of Michigan are doing experiments to determine if laser microexplosions will be easier on the eyes.

Finally, the spot of a femtosecond light is much smaller than a human cell; therefore, it should provide a way to kill a malignant or damaged cell without harming adjacent healthy cells. No one is doing this yet, but Harvard biologists are considering it for the study of embryos. By zapping a few cells on a growing nonhuman embryo, they should be able to determine how each cell affects development.

As femtosecond lasers increase in availability and capability, it seems likely that both research and practical uses will multiply. Build the laser and they will come.