Now that we’re in the age in which everything is moving at the speed of light, it’s no wonder that our technological advances in photonics are racing to keep up. Although the technology is improving in record time, the challenge is for companies in the industry to meet production and standardization needs.
When the wave of microchip technology hit the market in the 1970s, it revolutionized the industry. One aspect of this revolution was the need to produce mass quantities of a family of entirely new products that had to be delivered to the market in as short a period of time as possible. The microchip industry has had more than 20 years to evolve to a point at which giant corporations like Intel (Santa Clara, CA) make the devices by the millions.
Capital equipment vendors, such as Applied Materials (Santa Clara, CA), Novellus (San Jose, CA), and KLA Tencor (San Jose, CA), have had ample time to develop the sophisticated systems necessary to fabricate these products. Many of these large, successful companies began as small start-ups in Silicon Valley. They were nurtured on process technology and tasked with the challenge of building the equipment that would make Moore’s Law a reality. More than 20 years later, they have evolved into world-class suppliers of precision automation technology for chip manufacturers.
We now are seeing a similar revolution taking place in the optoelectronics industry. Optoelectronic microchips not only carry information through electrical currents but also transmit and receive signals through coherent light waves generated by semiconductor lasers. The voracious demand for the optoelectronic components that constitute the core building blocks of telecommunication networks has created a severe global shortage of these devices. This is largely due to the fact that optoelectronic devices currently used in the field by telecommunication network providers are so new that there simply has not been enough time to design them with high-volume production in mind. There also has not been enough time to develop machines that can produce these devices at a rate of thousands per day.
In today’s fast-paced global market, however, no one can wait for this industry to go through decades of evolution before it reaches the level of maturity of the semiconductor electronics industry. Photonics companies are attempting to squeeze 20 years of technical evolution into two or three years. Equipment vendors are uniquely positioned to answer this need and to take the leadership role in blazing the trail that will lead to economical, high-volume production of fiber-optic components. But there are challenges in achieving this goal.
One problem is that so many types of optoelectronic components are finding their way so rapidly from R&D laboratories to the manufacturing floors that there is little or no process or component design standardization. This makes the design of fully automated machines very challenging because the essence of automated manufacturing is performing the same task over and over on the same type of product.
Another challenge is that, unlike electronic devices, optoelectronic devices have optical fibers hanging from them. These fibers are very fragile, they require submicron directional alignment, and they are awkward to deal with. As a result, the devices need to be handled manually by skilled technicians at many stages of the manufacturing process. This leads to a situation in which your process becomes operator dependent, thus lowering yield levels and throughput.
Consequently, the key to building high-volume-production photonics assembly lines and factories is not necessarily the mechanization of the same old manual process, but rather the definition of new automation strategies and machine platforms that can make the manufacturing process independent of the operator and highly repeatable. Once you achieve repeatability, it is possible to gain control over your process and ramp up both your production capacity and yield levels. This results in an economical operation. The next-generation automated fiber-optic component production lines are based on this concept.
For the fiber-optic component industry to thrive in this emerging optically enabled world, we need to learn from the lessons of the microelectronics experience gained since the 1970s: By rapidly embracing automation technology as a yield-enhancement tool, photonics can leap past the early years of its evolution to the long-term benefit of all.
(By Robert Deuster, Newport Corp,. OEmagazine, February, 2001)