We all have seen the projections: The optical networking market is expected to grow at a rate of about 50 percent per year for the next five years. Implementation of these networks will require a huge volume of basic optical components, such as filters, mirrors, polarization optics, and prisms. These basic components deserve more attention in the network planners’ efforts to connect the world.
Beginning in the mid-1990s, the demand for increasing production efficiencies in semiconductor fabrication drove up demand for high-quality optics. The industry also moved to deep-ultraviolet (UV) excimer lasers for lithography and into UV regions for inspection systems. Coinciding with these technical advances was the start of the electronics industry’s biggest-ever building boom of wafer fabrication facilities.
The impact on manufacturers of laser optics was huge. Optics companies saw the opportunity to grow at unheard-of annual rates of up to 50 percent. But 50 percent growth in optics sales was not sufficient to draw capital investment into the industry at that time, so optics companies expanded using debt or internally generated funds. New companies were started but they were mostly undercapitalized. As a result, extremely long lead times and shortages were the order of the day.
After a two-year boom, the market for semiconductor fab equipment, following its normal strong cyclicality, petered out. Many optics companies struggled for survival. Some were weakened or bought out at very low valuations by stronger competitors.
Could this pattern happen again in the demand for dense-wavelength-division-multiplexing (DWDM) optics? If Intel’s Andy Grove is correct and “only the paranoid survive,” optics company managers have to consider this possibility as they respond to the requests for quotes (RFQs) for optics in annual quantities of hundreds of thousands, or even millions.
Differences in optics fabrication
So why can’t companies fabricating high-quality optics react to the market demand as fast as the companies that incorporate those optics into component packages or systems for the end applications? I think there are three major factors at work.
First, most optics companies remain privately owned and don’t have access to the public markets for capital. As a result, they have to be much more careful with what little capital they have. Expanding 50 percent to 100 percent per year is “betting the farm.”
Second, high-quality optical fabrication techniques remain heavily skill and time oriented. People who visit a very good optical fabrication shop for the first time always remark, “How can they achieve such high-quality products with that equipment?” Some equipment manufacturers have attempted automation, but with mixed results and questionable cost efficiency.
Third, only rarely can people make the leap to master optician after just a few weeks of training and a few months of experience. In most high-quality optical fabrication, the individual technician makes a much more important contribution to the success of a production run than the equipment he or she is using. It may take years just to develop the skill needed to polish quartz to 1 Å rmf surface roughness.
Some might ask if the tens and hundreds of thousands of optical components manufactured annually for camera and telescope applications could be put to use for DWDM products. Yes, high-volume, high-speed techniques are used in manufacturing lenses and prisms for these applications, but those optics can tolerate very large imperfections in material and surface quality while still functioning adequately within the limits of human vision. The specifications for DWDM optical components are at the opposite end of the quality spectrum. The efficiency requirements of these systems will not allow vision-grade optics. The challenge is to scale up the capabilities of fabrication processes to yield volumes of higher-quality optical surfaces.
Teaming up on DWDM
In the last 12 months there has been overwhelming interest by the DWDM systems integrators in coating technology and capacity. Several private and a few public coating companies have been acquired for truly astonishing valuations. Partnership with a much larger parent gives these companies improved access to capital for capacity expansion and process development. It also provides the optics company with better market vision in product development efforts. This kind of partnership minimizes the risk a private company incurs when it dedicates itself to a single industry.
So far, interest and acquisitions have focused, with a few exceptions, on a single element in the DWDM system: thin-film bandpass filters. There is significant cause for concern about the world’s capacity for high-quality micro lenses, prisms, beamsplitters, waveplates, polarization optics, and similar basic components, however. Will the industry direct capital and intellectual effort to the need for significant expansion in capacity for these optical components?
It is clear that appreciable capital will be necessary. If not forthcoming, these components may become the gating element in the system as other, more critical, issues are resolved.
(By Jim Higdon, CVI Laser, Inc. OEmagazine, March, 2001)