Ultrafast LIGHT

In the 1980s, the titanium-doped sapphire (Ti:sapphire) laser changed the way scientists investigate the physical world. With pulse durations in the femtoseconds, scientists essentially could take the equivalent of a still photograph of extremely short-lived events, illuminating chemical interactions as they happened and creating a new understanding of how materials are formed from molecular constituents. These first ultrafast laser systems were large, complex optical systems that used tricky argon ion-beam lasers to pump the Ti:sapphire oscillators and multilayered engineered optics to compress the laser pulse to only a few femtoseconds in duration. In short, only laser jocks could make ultrafast laser systems work well.

By the mid-1990s, the need to make ultrafast lasers simpler to operate led to diode-pumped solid-state (DPSS) green pump lasers for the Ti:sapphire oscillators. Ultrafast laser systems that used to be several boxes including pump, oscillator, and power supplies now sat comfortably on an optical bench. This simplified system opened ultrafast laser technology to new users and new applications in material processing and biology, but widespread use required more development. The result was the "infamous black box," explains Marco Arrigoni, director of marketing at Coherent Inc. (Santa Clara, CA).

The simplification of ultrafast laser sources has resulted in steady market growth for these systems between 5% and 15% each year, according to industry insiders. Exceptions include high growth during the first years following the introduction of Ti:sapphire lasers in the 1980s, and then the upgrade to DPSS pump systems in the 1990s. In 2002, the ultrafast market was estimated at approximately $110 million, with sales split as follows: 25% continuous-wave pumps, 20% Q-switched pumps, 30% oscillators with optical parametric amplifiers (OPA), and 25% amplifiers plus OPA. The market is expected to grow at approximately 5% compound annual growth rate for the next few years, slightly below recent growth rates during the past five years.

The major reason for the dip in ultrafast laser sales has to do with who funds the purchases: namely, governments. Despite the advent of single box ultrafast laser sources to meet a growing variety of applications, these systems are still mainly used for research activities in chemistry, physics, and biology rather than industrial applications.

"There are cycles that ultrafast lasers go through," says Wilson Sibbett, director of research in the School of Physics and Astronomy at the University of St. Andrews (St. Andrews, Scotland). "For instance, [ultrafast laser] funding in Germany was good for the past several years, but now funding of pioneering or pure science is not as good, whereas the manufacturing side of lasers in the German market is quite good. The scientific [laser] market has always been determined by funding cycles of the National Science Foundation [NSF] and other organizations that fund research."

Sibbett alludes to other events as well: The NSF just finished doubling its budget during the past five years; Germany had an environmental initiative to replace gas lasers such as the argon ion laser, which led to additional DPSS pump sales for ultrafast oscillators; China is also finishing an effort to double R&D spending during the last five years, and there is a similar initiative in Japan.

"The saving grace of the global market has been the Asian market. The problem [for the global market] would be if China starts to produce a lot of product in their own country. That could really change market predictions," Sibbett says.

The next market transition for ultrafast lasers may come under the title "ultrafast Legos," according to Wayne Knox, director of the Institute of Optics at the University of Rochester (Rochester, NY). "Having just come off the telecom bubble, we're looking for new applications for some of the ideas we learned. A perfect example is the appearance of very high-power 980-nm pump diodes that cost 10 times less than they did a few years ago. Now we can use those as building blocks like Legos. We're all looking for a way to build scalable femtosecond sources—scalable in power, repetition rate, pulse width, and spectrum."

If these technologies are available today, the question arises—why are diode lasers not already used in commercial systems? St. Andrews' Sibbett may have the answer. "There's a lot of money that goes into developing product, and the last thing you want is for your product to become obsolete too soon." oe