Adding Fiber Pays Off for Industry

Hank Hogan, Contributing Editor, hank@hankhogan.com

For Electro Scientific Industries Inc. (ESI) of Portland, Ore., getting fiber into its diet has been one key to success.

Among other things, the company makes a line of micromachining systems for the semiconductor, flat panel display, photovoltaic and other industries. Some of these systems are powered by fiber lasers, a technology in which a fiber doped by rare earth elements, and typically pumped by a diode, acts as the lasing medium. The choice of when to go with this approach over an alternative depends in part on what is being machined and where the system will be used.

“We deploy fiber lasers in places where we require, predominantly, high average power, and we want industrial robustness that extends well beyond a year,” said Jeffrey Albelo, ESI’s general manager for interconnect and micromachining.

As the ESI example shows, fiber lasers are being used for cutting, marking and other materials processing. They also are showing up in lidar and other applications. Their efficiency, compactness, relatively high power and ruggedness often make them the laser technology of choice – provided the wavelengths offered match the wavelengths needed.

Recent advances promise new fiber laser capabilities, including the ability to engineer polarization modes not possible before. A look at these and other developments reveals some of what is in store for the technology, the industry and applications.

Going down … and up

With increasing peak power and repetition rates, fiber lasers are moving into applications that were handled in the past by other technologies at ESI. In part, according to Albelo, that is because fiber lasers are not only more robust but generally cheaper on a per-watt basis than YAG lasers, the other solid-state laser technology used by the company. The company does a lot of beam conditioning in its products, so another plus is that fiber lasers are very amenable to software manipulation.

One of ESI’s products, for example, puts two fiber lasers to use, combining them with precise positioning to increase machining speed. The wavelength of the lasers is about 1 μm, which is appropriate for the metallic features the system is designed to handle.

But today, fiber lasers aren’t the answer to all of the company’s needs. When asked what he wants, Albelo replied, “The answer is very easy: ultraviolet.”

In general, he said, shorter wavelengths allow more efficient coupling of laser energy into the material being processed. Thus, a fiber laser with power in the ultraviolet would be helpful, as would one with a polarized beam and a higher repetition rate. Of course, Albelo wants these advances without sacrificing any of the current advantages offered by the technology.

The request for lower wavelengths is one that the industry is aware of and is trying to meet, said William S. Shiner, vice president of industrial products at high-power fiber laser leader IPG Photonics Corp. in Oxford, Mass.

“We’re already at 532, which is a visible green. We’ve already frequency-doubled that to the UV. We just haven’t released product yet,” he noted.

Although efforts aimed at creating an efficient UV fiber laser are under way, the technology has undergone growth in other areas. Steve Norman is chief technology officer at SPI Lasers in Southampton, UK, a wholly owned subsidiary of the Ditzingen, Germany-based Trumpf Group. He noted that his company’s products have seen a steady increase in power over the past five years.

“We’ve power-scaled from 50 to 100 W to 200 to 400 W. We’re now in volume production with 400-W products,” he said.

The latter power can be ganged together to create multikilowatt systems, with that power being used to speed materials processing. The trade-off for achieving higher power in this way involves an increase in the number of modes. However, in many high-power cutting and welding applications, a multimodal beam is preferred, in part because it produces a more uniform distribution of energy on the workpiece.

There’s an app lab for that

Besides power or wavelength, other parameters that can be tailored to an application involve pulse characteristics. Marking plastic, for example, may work best by using relatively low pulse energies delivered at a higher repetition rate and by using a predetermined waveform. All of these settings can differ from those that are best when working metal.

Figuring out the right laser setup can be a challenge. That’s one reason SPI runs application labs in both the UK and in Santa Clara, Calif. Equipped to do cutting, welding, scribing and analysis, these sites do research and development to prove new application feasibility and to optimize processing parameters.

A fiber laser cuts silicon. Courtesy of SPI Lasers.

Meeting this type of need lies behind a Portland, Ore., micromaterials processing facility established by ultrafast fiber laser maker Fianium, which also is based in Southampton. The applications lab is located within and run by Summit Photonics, which offers third-party photonics engineering services.

The partnership between the two was announced last year, but the site was only fully equipped and operational in May of this year. Colin Seaton, global vice president of sales and marketing at Fianium, noted that much of the initial investigations into materials processing with picosecond lasers has been done with “scientific-type setups,” systems that offer vendors such as Fianium an opening.

Processing silicon for solar cells could benefit from visible picosecond fiber lasers, as shown here with this run of the effects of 532-nm picosecond pulses on silicon-nitride-coated photovoltaic-grade c-silicon. Pulse energy increases from upper left to lower right. Courtesy of Brian Baird, Summit Photonics.

“We’ve got the benefit of cost, complexity and reliability that the fiber laser brings, compared to those other types of systems,” he said.

As for the applications that might be done at the lab, Summit’s managing director, Brian Baird, noted that there is a broad area of interest in picosecond processing of semiconductors and thin films. He gave a talk at Photonics West this year in which he pointed out that wafer scribing, silicon micromachining, laser marking and thin-film trimming are all possible areas where the technology could be used.

“Those areas requiring high repetition rates and good precision in the beam quality and pulse energy delivery are prime candidates for being in picosecond laser products,” he said.

Remote sensing

There are, of course, industrial uses of fiber lasers that don’t involve materials processing. One example can be found in lidar, an application being pursued by Keopsys SA of Lannion, France. The company’s products are used for laser ranging systems that measure distance to a point and wind speeds of the air, or do 3-D scanning of an object. This is done using wavelengths that range from 532 nm to 2 μm.

With the right settings for power, repetition rate and pulse waveform, fiber lasers can mark metals or plastics, as has been done with the parts shown here. Courtesy of SPI Lasers.

Fiber lasers are well suited for lidar for a number of reasons, said Keopsys’ vice president and founder Jean-Marc Delavaux. “The two most important are they’re lightweight and [have] low energy consumption.”

Keopsys touts its V-groove side-pumping approach, which leads to a one-step coupling process of the diode pump laser into the double-cladding gain fibers used in the company’s lasers. Keopsys believes that this gives its products a 90 percent coupling efficiency, which it asserts is significantly higher than would be possible otherwise. That, in turn, helps lower power consumption and improves performance in lidar and other applications.

Putting polarization to work

Finally, recent research has shown that fiber lasers in the future may have another knob that can be adjusted. Researchers from the University of Dayton in Ohio recently reported on a fiber laser with an adjustable polarization output. In a May 10, 2010, Optics Express paper, Qiwen Zhan, associate professor of electro-optics, and his team described a specially designed laser cavity. By combining that with a uniaxial c-cut calcite crystal, they created a cylindrical vector beam with cylindrical polarization symmetry.

Such polarization is different from the traditional variety, which is typically either linear or circular. This nontypical polarization, and the ability to adjust it, could prove useful in optical tweezers and in micromachining, Zhan said.

In both cases, cylindrical vector polarization has been shown to yield better results than the traditional variety. Also, in both cases, the best polarization depends upon the task, so being able to adjust parameters as needed is important.

Zhan said the choice of fiber is critical in the group’s research because the fiber determines how many polarized modes can be supported in the laser cavity. The crystal is key to selecting and producing the reconfigurable vectorial output. Experimental results are shown of a complex vectorial mode created using a fiber laser (left), with numerical simulation results (right). (a) The intensity profile of the output mode; (b)-(e) the beam intensity profiles after a linear polarizer (transmission axis indicated by white arrow), show the different modes. Courtesy of Qiwen Zhan, University of Dayton.

In their demonstration device, the researchers showed that modes with radial, azimuthal and generalized cylindrical vector polarizations could be generated by translating one lens within the laser cavity. They also showed that more complicated vectorial vortex output modes could be created by introducing some angular misalignments into the setup.

Transforming these lab results into products will require improving the output power to something over 1 W in continuous-wave operation, Zhan said. His research plans also call for exploring Q-switch, mode-locking, frequency conversion and other operation modes. He also intends to investigate more complicated vectorial modes.

Eventually this could lead to a new type of device, and it is one that Zhan believes he can build, given the right funding.

“The ultimate goal is to achieve a compact all-fiber high-power fiber laser producing reconfigurable vectorial mode outputs. I already have a strategy to achieve this,” he said.