Solid-state UV laser to improve step-and-scan

04/01/1997

Solid-state UV laserto improve step-and-scan

Aculight Corp., Bellevue, WA, has been awarded a $560,000 contract from the Air Force Wright Laboratory in Dayton, OH, to develop a solid-state 10 kHz pulse repetition rate, multiwatt output power, 193-nm ultraviolet (UV) laser. The laser is intended to be used in 0.18-micron microlithography equipment to fabricate advanced microcircuits. Other possible applications include jet engine combustion diagnostics and short range optical communication.

The laser concept is based on frequency-mixing the harmonics of a 1-micron laser with the mid-infrared output from a periodically-poled lithium niobate optical parametric oscillator. One of the objectives of the contract is to develop a nonlinear optical wavelength converter that can be eventually scaled to 5 W at 193 nm. "The goal is to get several watts of 193-nm output power from a solid-state laser," explains Roy Mead, principal research scientist. "We are now doing the earliest experiments, generating milliwatts of power at first, but we are looking to scale that up as soon as we can."

The challenge is to make the system work at a low repetition rate of relatively few pulses/second at the right energy of pulses to get several watts at 10 kHz. Aculight is looking to generate fractions of a millijoule of energy out of the laser at low repetition rate. "First we want to make that low repetition rate step efficient," says Mead. "The efficiency is important so that when we scale up the device it won`t require huge lasers in order to make it work. Then, we`ll move on to higher repetition rates and high average power heating of the crystals that allow us to convert the long wavelengths to short wavelengths." Heating of those crystals may be a significant problem. They can be damaged by the high power of the laser light.

The solid-state materials used would be those like neodynium yag (Nd:YAG), which is now being used for various semiconductor applications, including probing for memory repair, but at very low power and longer wavelengths than 193 nm. "The advantages of the solid state are longer lifetimes, lower operating costs, higher repetition rates - 10 kHz as opposed to a limit of perhaps 1 kHz for the excimers - and easier narrowbanding," says Mead.

Applications for the new laser include stepper and step-and-scan tools. Higher repetition rates for step-and-scan tools will allow for more pulses in a given exposure area for improved dose control. "Being able to narrow the laser to less than a half picometer would allow the use of all refractive optics as are used now at i-line and 248 nm. The best bandwidth that`s been achieved with reasonable power in an argon fluoride laser is about 5 picometers at a few watts of output power. That`s too broad to use with an all-refractive lens system as all of the steppers are using now. A solid-state light source that would be capable of much narrower bandwidth would enable extension of the step-and-repeat lens technology that`s used now."

Development of the solid-state laser is at an early stage. Mead projects that high power devices could be under test by 1998, and production models could be available in 2001. "We`ve actually produced 193-nm light by this means at relatively low pulse energy." - L.S.