A technology for converting Gaussian laser beams into collimated flat-top beams with nearly 100% efficiency is poised to emerge from the laboratory. A refractive optical conversion system first reported by IBM Almaden (San Jose, CA) researchers almost two years ago was exhibited as a "new-product concept" for a refractive beam shaper by Newport (Irvine, CA) during the Conference on Lasers and Electro-Optics (CLEO; Long Beach, CA) exhibition in May. The purpose of the exhibit presentation was to gauge user interest in the technology as well as in various potential device applications and parameters.
The technology is based on a Keplerian beam reshaping system using two convex aspheres with an intermediate focus (see Laser Focus World, December 2000, p. 9). It originally arose out of two different optics projects at the IBM Almaden laboratories. A holographic data storage project needed a uniform illumination source for its liquid-crystal spatial-light modulator to provide uniformly high signal-to-noise ratio across the illuminated surface.1 And an interferometric lithography project that sought to characterize photoresists well beyond typical dimensions also required a spatially broad and uniform beam.
"The key thing about both of those applications was that we were trying to illuminate relatively large areas, and both of them depend very essentially on preserving the coherence of the laser beam," said IBM researcher John Hoffnagle. Hoffnagle fielded questions about the device in the Newport booth at the CLEO exhibition, along with Michael Jefferson, also a researcher at IBM and co-author of the original research report.2 "You can't use the kind of randomizing, coherence-destroying approaches to illumination that are often used in lithography. We wanted to make real propagating plane waves—or as close to plane waves as we could—and to have uniform illumination over a fairly large area."
The problem of making laser beam shapes more useful for illumination has been around ever since lasers were invented, Hoffnagle said (see figure). Numerous solutions have been proposed but combining both efficient performance and ease of manufacture has proven elusive. The concept of refractive laser-beam reshaping using aspheric lenses is more than three decades old with a simple coaxial optical arrangement based on a variant of a Galilean beam-expanding telescope. Actually achieving the potentially high conversion efficiency in a cost-effectively manufacturable form has had to wait in part for improvements in asphere technology, however.
Keplerian design
"The technology of making aspherics has improved dramatically in recent years and that's helpful," Hoffnagle said. "But even so, it's not exactly a trivial thing to do." An important innovation by the IBM researchers in making the device more manufacturable was to switch from the Galilean telescope design to a Keplerian design in which all of the aspheric surfaces are convex.
"When you are polishing those surfaces, you are always polishing the top of the hill," Hoffnagle said. "Everything is acceptable to the polishing tool and there are no strange surfaces or reflection points or things of that sort. Attention to detail like that is what makes this really doable."
Another important factor involved designing the system for an output intensity profile that would propagate effectively. A lot of the earlier literature refers to generating top hat intensity distributions in which the intensity is designed to be absolutely uniform out to some predetermined radius and then to drop discontinuously to zero. "We wanted to make beams for our purposes that would be plane waves that would propagate, and it's pretty clear that a step function is not going to exhibit that property," Hoffnagle said. So the IBM researchers chose an output intensity profile with a controlled roll off at the edge that enabled computation of diffraction intervals to evaluate propagation, while at the same time preserving the property of very good uniformity and efficient use of laser energy with in the uniform part of the beam.
Visitors to the CLEO exhibit asked questions about making flat tops very small for material processing applications like marking, scribing, cutting, and drilling. In the meantime, the researchers have also been working in the laboratory to more fully characterize the emerging wavefront. "We've established that in fact it is a plane wave to about λ/5," he said.3 They've also calculated, using commercial ray-tracing software, that even though the optics demonstrated at CLEO were designed to be used at 532 nm, those same optics will produce diffraction-limited beam quality over a range from 250 to 1500 nm.
With all that has taken place over the past three decades, in some ways the developments are just beginning, Hoffnagle said. "My feeling about this particular system is that there has been a great deal of development from the first proposal of aspheric Gaussian-flat top conversion back in the 1960s to the point where we now have something that actually works pretty well," he said. "I think that the next steps will come when we get this out of the lab and it starts to become part of what you might call the optical designers toolkit."
REFERENCES
1. H. J. Coufal, D. Psaltis, and G. T. Sincerbox (Eds.), Holographic Data Storage, Springer, 2000.
2. J. A. Hoffnagle and C. M. Jefferson, Appl. Optics 39(30): 5488 (Oct. 20, 2002).
3. J. A. Hoffnagle and C. M. Jefferson, Proc. SPIE 4443, 115 (2001).
Hassaun A. Jones-Bey | Senior Editor and Freelance Writer
Hassaun A. Jones-Bey was a senior editor and then freelance writer for Laser Focus World.