Open the hood of a classic automobile and you'll find an engine, a few hoses, and wires—and lots of empty space. The form and function of everything will be easy to see, not only because there is no obscuration, but also because the hardware itself is relatively simple. Do the same to a modern car and you will likely be confronted with an intricate mass of cables, tubes, and other equipment that almost hides the engine. This is a shame, for the modern automobile engine has developed a more complex form that mirrors the technological revolution inside, where sophistication does not come at the expense of reliability, but adds to it.
The first truly production-worthy pieces of leading-edge optical lithographic equipment—wafer steppers—were like classic cars: spare, simply built, and easily accessible to technicians. The modern wafer stepper or scanner has gone through generations of refinement far beyond even those of the automobile. Two decades ago, the ability to expose integrated-circuit features approximately 2 µm in size was standard; since then, the printable feature size has plunged by a factor of 15 in production lithographic equipment to 0.13 µm. Reliability and throughput in wafers per hour have also advanced. Apply that sort of improvement to an automobile and the result would be a faster, more reliable vehicle that gets more than 200 miles per gallon of gas.
Open the environmental chamber that houses the modern wafer stepper or scanner and you'll find incredible complexity. Built to submicron alignment tolerances, the optical train itself may weigh half a ton. The once-visible illuminator, projection lenses, and wafer translation stages are almost hidden inside massive support structures required for stability; precision wafer and photomask alignment systems, wafer transporters, and photomask positioners must fit wherever possible. The well-publicized development of 157-nm lithography—built around the fluorine excimer laser—will ensure even higher complexity, keeping design engineers scrambling to maintain reliability and throughput.
These three articles cover the emerging technology of 157-nm lithography, from the general to the specific. First, Don Golini and coworkers describe the application of magnetorheological finishing to the fabrication of calcium fluoride optical components crucial to 157-nm optics. The second article, by James McClay and colleagues, outlines a program to tackle the many remaining technical problems facing designers of steppers and scanners. The concluding article, by Charlene Smith, illustrates how modified-silica glass photomasks can be made to transmit at the shorter wavelength, avoiding a tricky material problem. All show that there are many miles left in optical lithography.
John Wallace | Senior Technical Editor (1998-2022)
John Wallace was with Laser Focus World for nearly 25 years, retiring in late June 2022. He obtained a bachelor's degree in mechanical engineering and physics at Rutgers University and a master's in optical engineering at the University of Rochester. Before becoming an editor, John worked as an engineer at RCA, Exxon, Eastman Kodak, and GCA Corporation.