Solid state laser for microlithography

Details Solid state laser for microlithography Solid state laser for microlithography

1999-08-05

2002-12-24

Scott, Jr., Leon (Department: 2828)

Coherent light generators

Particular beam control device

Nonlinear device

Reexamination Certificate

active

06498801

ABSTRACT:

FIELD OF INVENTION
This invention relates to solid state lasers and particularly to diode-pumped solid state lasers that utilize phase conjugating Stimulated Raman Scattering (SRS) to generate the desired beam wavelengths suitable for microlithography and other applications.
BACKGROUND TO THE INVENTION
The field of microlithography and the lasers needed to etch electronic microchips is burgeoning. At the present state of technology, one of the limiting parameters to smaller microchips and higher component density on such chips is the wavelength of the coherent light etching the components on the substrate. The quality of the chip circuitry and the economic viability of the production methods depend on a number of factors arising from the laser used for etching. These factors are the laser's emission wavelength, the laser's beam divergence and beam profile, the beam's pulse duration, the beam's pulse repetition rate, and the laser's overall compactness and safety.
The laser's emission wavelength determines the resolution that can be achieved on the individual chip. The shorter the laser's emission wavelength, the higher the resolution that can be achieved. Since higher resolutions lead to higher component densities on the chip, this translates to more valuable chips.
The laser's beam divergence and profile also affect the viability of the production method. A beam divergence that is too high, in conjunction with other beam parameters, renders a laser unsuitable for microlithographic applications. For the intended application, it is desirable to bring the laser's beam divergence closer to the diffraction-limited, plane-wave, TEMOO single mode.
A high pulse repetition rate coupled with shorter pulse durations lowers the probability of damaging the manufactured components. Thus, it is desirable to have higher pulse repetition rates and, within certain limits, shorter pulse durations. The laser system's overall compactness and work safety factor relate simply to the cost of chip fabrication plants.
Today's mature excimer lasers can directly generate in the sub-200 nm wavelength range of the ultraviolet (UV) spectrum. However, these lasers have some unfortunate drawbacks. These excimer lasers, such as those outlined in U.S. Pat. No. 5,586,134 issued to Palash et al. and in U.S. Pat. No. 5,018,161 issued to Sandstrom et al., require the use of dangerous halogen materials such as xenon fluoride (XeF), krypton fluoride (KrF), argon fluoride (ArF) and others. The use of such gases necessitates large gas processing, storage and circulation technologies. Also, using such lasers in a fabrication plant requires changing the air inside the plant entirely every several hours for safety purposes. These lasers are bulky, complex, potentially hazardous, and expensive. Furthermore, these excimer lasers cannot generate at pulse repetition rates much higher than 1 kHz, and their beam mode is quite far from the desired TEM
00
single mode.
On the other hand, solid state lasers do not suffer from the safety, repetition rate, and beam mode drawbacks of excimer lasers. Solid state lasers do not use dangerous gases and this eliminates the added expense and potential harm such gases can cause. Not only that, but solid state lasers can be pulsed at multi-kHz repetition rates that are much higher than that of excimer lasers. Furthermore, the beam quality of solid-state lasers can be close to the ultimately possible diffraction limit. However, solid state lasers do not directly emit in the desired UV range. The optical frequency, and thereby the wavelength, of the solid state laser's emission must therefore be converted to the desired wavelength range to make them suitable for microlithography.
Such a conversion can be achieved by directly multiplying the optical frequency, a method known as harmonic generation. Harmonic generation is the most common method of frequency conversion (see e.g. W. Koechner. “Solid-State Laser Engineering”, 3
rd
Edition, Springer-Verlag, Berlin, 1992). However, this method does not allow the generation of wavelengths other than the initial wavelength divided by an integer factor.
Fortunately, there are other methods that allow the production of UV emission from solid state lasers. One method that is utilized in U.S. Pat. No. 5,742,626 issued to Mead et al. is the use of an optical parametric oscillator (OPO). This method is based on a nonlinear process that includes producing two independent, separated beams from the initial single beam. This approach suffers from an unfortunate practical shortcoming. The method necessitates combining two separated beams to produce the desired UV output beam. A sum frequency generator would combine the OPO-produced beam with the fifth harmonic of the input laser beam. While combining such beams is theoretically and experimentally possible, achieving the necessary efficiency to make the method useful will be very difficult, if not practically impossible. One reason for this is that it is very difficult to achieve the precision required to align independent, separated beams with the necessary parallelism. Also, the pulses produced by the laser and by the OPO have different temporal shapes, further complicating the efficient mixing of the beams. Also, the OPO's jitter and the substantially broadened spectrum preclude efficient mixing. The above reasons therefore render the method and apparatus of Mead et al. to be impractical.
Another method for frequency conversion that produces different wavelengths from fixed wavelength solid state lasers is the use of Stimulated Raman Scattering (SRS). Raman scattering is a process in which light is scattered at frequencies which are the sum and the difference between the incident frequency and the oscillation/vibration frequencies of the scattering material. When a scattering material is irradiated by a monochromatic light that has a frequency which does not correspond to any of the absorption lines of the material, frequency shifted components of the light can be detected in the scattered radiation. These shifted components have shifts independent of the irradiation frequency but characteristic of the scattering material. A laser beam, when passed through a scattering material—or Raman medium—produces shifted lines on the low frequency side and shifted lines on the high frequency side. The shifted lines on the low frequency side are called the Stokes-shifted lines and the shifted lines on the high frequency side are called the anti-Stokes shifted lines. Thus, laser light with frequency v scattered on a Raman medium consists of not only the initial frequency v but also the v−v
1
frequency (Stokes-shifted with a longer wavelength) and the v+v
1
frequency (anti-Stokes shifted with a shorter wavelength) where v
1
is characteristic of the Raman medium.
From the above, it can therefore be seen that SRS can produce different wavelengths of light given an input beam and a judicious choice of a Raman medium. The Raman medium could be gaseous, liquid or solid. The Raman threshold and other physical conditions for frequency shifting are very different for the three types of media and this leads to substantial differences in designs of gas, liquid and solid state Raman lasers. Most known practical designs of Raman lasers use gas Raman cells. U.S. Pat. No. 4,633,103 issued to Hyman et al. uses SRS in gas cells to produce yellow light for laser guide-star applications. U.S. Pat. No. 5,796,761 issued to Cheung et al. utilizes SRS in solid Raman cells to produce a laser beam in the visible range for similar applications. U.S. Pat. No. 5,673,281 issued to Byer also proposes using an SRS frequency conversion laser, this time using a solid SRS cell, for guide star applications. However, none of the methods above can convert laser light into the UV-range with a sufficient efficiency to be suitable for microlithography.
The patent by Hyman et al. describes a classic gas-cell based Raman laser where Raman cells are placed outside the laser resonator. N

Affiliated with

Also associated with

Rating

Solid state laser for microlithography does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Solid state laser for microlithography, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Solid state laser for microlithography will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2992932