Compact sources seek fresh industrial niche

26 Nov 2002

New advances in diode-pumped femtosecond sources based on Cr:LiSAF and ytterbium-doped crystals could help them break into industrial applications. Michael Hatcher asks whether these compact devices can succeed where Ti:sapphire lasers have so far failed.

From Opto & Laser Europe December 2002

If you were asked to name the source most suited to producing femtosecond pulses of light, chances are that you would choose the Ti:sapphire laser. The crystal's thermal properties and its wide fluorescence band make it the obvious choice, especially for laboratory researchers who are looking for a flexible system to experiment with.

However, the large size and relative complexity of Ti:sapphire lasers are issues that mean they have yet to find any serious industrial applications. Although some smaller systems - such as Spectra-Physics' one-box Hurricane laser or Coherent's new "hands-free" Chameleon - now exist, they are primarily designed for scientific use and require a skilled operator. Only US company Clark-MXR's RS-2001 Ti:sapphire workstation stands out as having been specifically developed with industrial materials-processing applications in mind.

Bulky systems The reasons that Ti:sapphire systems are so bulky - the RS-2001 is about the same size as a walk-in wardrobe - are twofold. They need to be pumped in the green, and the crystal's lasing threshold demands a high-power source. Since no suitable diode lasers operate in this region, the Ti:sapphire crystal emission is generally seeded with an argon-ion or frequency-doubled YAG laser.

For femtosecond lasers to achieve large-scale commercial deployment, they almost certainly need to be diode-pumped. And that means finding crystals that can be pumped in the red or near-infrared, and have a relatively low lasing threshold.

Recently, a new breed of diode-pumped source has begun to look very promising. Over the last decade researchers have identified a number of crystals capable of delivering diode-pumped femtosecond pulses. They are now busily building laser systems around those crystals, the most obvious advantage of which is size: the new lasers based on ytterbium- and chromium-doped crystals are about the size of a shoebox.

The first such crystal to be identified was colquiriite - better known as the chromium-doped LiSAF crystal (chemical formula Cr³⁺:LiSrAlF₆) - which was discovered by Steve Payne and colleagues at Lawrence Livermore National Laboratory in the US. Cr:LiSAF development was rapid in the early 1990s, especially at Imperial College, London, and at ETH Zurich under Ursula Keller. Her semiconductor saturable absorption modelocking (SESAM) devices were ideal for producing ultrashort pulses from the lasers, and the ETH Zurich spin-off company Time-Bandwidth Products soon released a commercial version of the Cr:LiSAF laser.

However, the laser proved to be unstable. Before long, Time-Bandwidth took it off the market. "There was nothing fundamentally wrong with the Cr:LiSAF systems, but the problem was with the diodes - they simply weren't reliable enough," Keller told OLE.

The reason for this was that Cr:LiSAF, which emits at 860 nm, has an absorption band between 600 and 680 nm in the red. Although red diodes did exist to pump the crystal, diode manufacturers concentrating on the important telecoms wavelengths had not developed red diodes in favour of their near-infrared cousins. However, with the emergence of DVD read-write modules, the likes of Sony, Sanyo and Hitachi have now developed far more reliable narrow-stripe red diode-lasers with good beam quality.

Researchers in Wilson Sibbett's group at the University of St Andrews, UK, have taken advantage of this new breed of red diodes to make a portable femtosecond laser based on a Cr:LiSAF crystal. Ben Agate and colleagues' Z-cavity design produces 122 fs pulses with 35 mW average power output.

The narrow-stripe diodes make the optical system much simpler, with the entire battery-powered set-up fitting onto a piece of A4-sized paper. Importantly from a commercial viewpoint, the system is also highly energy-efficient: at 4%, its overall electrical-optical efficiency is thought to be the highest of any reported femtosecond laser. Alan Kemp, who has worked on the St Andrews project, says that three factors combined to produce such high efficiency.

"Using high-beam-quality diodes gave us excellent mode overlap so that we could utilize the pump power much more efficiently. We also minimized losses in the cavity design and improved the efficiency of the pump-coupling optics," Kemp explained.

The St Andrews team has also frequency-doubled the LiSAF output into the blue. Accessing this region should make the laser suitable for applications in biomedical imaging. And the project doesn't stop there: the next stage will involve frequency-tripling the output to access the ultraviolet and probe genes implicated in skin cancer development. Collaborating with Sir David Lane at Ninewalls Hospital in Dundee, Agate plans to study p53 - a gene known to suppress cancer growth - using the tripled 290 nm output.

However, as with Ti:sapphire sources, the drawbacks of Cr:LiSAF lasers mean that they are unlikely to find uses in materials processing. According to Keller, the crystal's high upper-state absorption means that high power output is not an option.

With uncertainties over Cr:LiSAF lasers, attention has turned to ytterbium-doped crystals. At the Laboratoire Charles Fabry de l'Institut d'Optique in Paris, France, Patrick Georges is at the forefront of novel laser-crystal development in this area. His recent innovations include the Yb:BOYS (Yb³⁺:Sr₃Y(BO₃)₃) and Yb:GdCOB (Ca₄GdO(BO₃)₃) lasers, which showed enough promise for Austria-based High Q Laser to develop into products.

Like other ytterbium-based lasers, the GdCOB and BOYS crystals absorb in the near infrared at around 980 nm, which is ideal for pumping with diode lasers developed for optical communications components. Despite absorbing and emitting at the same wavelengths, these particular crystals have a wider fluorescence bandwidth than the conventional YAG host, and can therefore produce shorter pulses. Pulses of just 69 fs from Yb:BOYS and 89 fs from Yb:GdCOB have been measured, compared with 340 fs from Yb:YAG.

So can portable femtosecond lasers challenge Ti:sapphire sources? Typical Ti:sapphire laser users - scientific researchers - look unlikely to be convinced, as they want the flexibility that is offered by Ti:sapphire systems. More promising, however, looks the possibility of these sources breaking into new applications that have so far been resistant to femtosecond technology.

Commercial breakthrough High Q Laser sells both Cr:LiSAF and ytterbium-based systems. Its chief executive, Daniel Kopf, says that chromium-based systems are inevitably limited to scientific applications such as biomedical imaging by their relatively low output-power when pumped with narrow-stripe diodes. But he believes that ytterbium sources will make a commercial breakthrough in industries that Ti:sapphire lasers have yet to crack. "I think we could see industrial deployment in 1-2 years. Systems are being shipped now and are going into development," he said.

Kopf cites micromachining as a major market that ytterbium-based femtosecond lasers could penetrate. But for all the potential of such sources in niche industrial and scientific applications, it is clear that the Ti:sapphire laser will remain the workhorse of choice in ultrafast laser technology. As Kopf says: "If tunability is important, the Ti:sapphire wins every time."

But Kopf is adamant that diode-pumped sources will soon be making their mark, both in the market for replacing Ti:sapphire systems, and - more importantly - in fresh industrial applications in which femtosecond lasers have yet to make an impact.