Coherent light generators – Particular beam control device – Tuning
Reexamination Certificate
1999-09-27
2001-05-29
Scott, Jr., Leon (Department: 2881)
Coherent light generators
Particular beam control device
Tuning
C372S057000, C372S102000
Reexamination Certificate
active
06240110
ABSTRACT:
BACKGROUND OF THE INVENTION
TECHNIQUES FOR LINE NARROWING
Techniques for decreasing the bandwidth of the output of lasers are well known. Several such techniques used on excimer lasers are discussed by John F. Reintjes at pages 44-50 in
Laser Handbook
, Vol. 5, North-Holland Physics Publishing, Elsevier Science Publishers B.V. These techniques include the utilization of gratings, including echelle gratings for wavelength selection. Use of beam expanding prisms ahead of the grating can increase the effectiveness of the grating.
A prior art narrow band KrF excimer laser is shown in FIG.
1
. The resonance cavity of excimer laser
2
is formed by output coupler
4
(which is a partially reflecting mirror) and echelle grating
16
. A portion of the laser output beam
20
(having a cross section of about 3 mm in the horizonal direction and 20 mm in the vertical direction) exits the rear of laser chamber
3
. This portion of the beam is expanded in the horizonal direction by prisms
8
,
10
and
12
and reflected by mirror
14
onto echelle grating
16
. Mirror
14
is pivoted to select the narrow band output for laser
2
. Grating
16
is arranged in Littrow configuration so that the selected narrow band of wavelengths is reflected back off mirror
14
and back through prisms
12
,
10
and
8
and into chamber
3
for amplification. Light at wavelengths outside the selected narrow band is disbursed so that this disbursed out-of-band light is not reflected back into the laser chamber. Total beam expansion for this laser is about 20X. The beam has a horizontal polarization (P-polarization for the prisms with the actual surface arranged vertically). Typical KrF lasers operating in a pulse mode may have a cavity length of about 1 m and produce pulses having a duration of about 15 to 25 ns. Thus, photons within the resonance cavity will make, on the average, about 3 to 5 round trips within the cavity. On each round trip, about 90 percent of the beam exits at the output coupler and about 10 percent is sent back for further amplification and line narrowing. The beam is repeatedly line narrowed as it passes through the line narrowing module.
With this prior art arrangement, the bandwidth of the KrF laser is reduced from its natural bandwidth of about 300 pm (full width half maximum or FWHM) to about 0.8 pm for KrF lasers and about 0.6 pm for ArF lasers.
Some applications of KrF lasers, however, require greater narrowing of the bandwidth. There is a need for smaller bandwidths such as FWHM values of 0.5 pm and 0.4 pm for KrF and ArF, respectively.
One prior art method is to add an etalon within the resonance cavity. In this case, the etalon is operated in a transmissive mode and the light is additionally line narrowed as it passes through the etalon. In such system one should use a relatively high finesse etalon, with a finesse value ƒ of about 1 or higher which produces sharp fringe patterns. The finesse value ƒ is determined by the equation:
ƒ=&pgr;
r
½
/(1−
r
)
where r is the reflectivity of the etalon surfaces. The dependence of etalon transmission spectrum on r is shown graphically in
FIG. 2
which is extracted from page 298 of
Fundamentals of Optics
by Jenkins and White, published by McGraw Hill. From
FIG. 2
, it is obvious why prior art transmissive etalons used for line narrowing have surfaces with reflectance of about 50% to 80% (see curves B and C of FIG.
2
).
FIG. 2
also shows that it would not be practical to use curve A-type low finesse etalon as it line-narrowing efficiency in this prior art arrangement would be very low. Prior art high finesse etalons used with diffraction gratings should enhance the line-narrowing capabilities provided by diffraction grating, and in general, improve the laser line-width. The major disadvantages of this technique are that the many reflections within the etalon tend to heat up the etalon producing distortions and that the tuning of the etalon synchronously with the grating does present a real technical challenge and is difficult to accomplish in practice.
F
2
gas molecular laser will likely be a successor to KrF and ArF excimer lasers as a production light source for next generation microlithography, enabling printing circuits with features as small as 50-70 nm. It is chosen for that purpose because of the very short wavelength of its output beam, which is about 157 nmn, as compared to currently used KrF and ArF excimer lasers, which produce beams with wavelengths of about 248 nm and 193 nm respectively. Narrowband versions of these lasers are currently used in semiconductor microlithography for printing circuits with smallest features in the range of 250 to 100 nm. The shorter wavelength of F
2
molecular laser permits much more tighter focusing and as a result, smaller features in the range of 50-70 nm can be printed.
However, all the available materials transparent at 157 nm have a significant dispersion in that region. Because of that, different portions of the laser beam, having slightly different wavelengths, will be focused by imaging lens onto different spots, so the image blurs. As a result, there is a tight requirement that the bandwidth of F
2
laser should be as small as possible. The estimations of possible lens designs show, that the bandwidth of F
2
laser should be about 0.2 to 0.5 pm at full-width half-maximum level (FWHM).
Unfortunately, the free running F
2
laser generates two lines, the stronger line at 157.630 nm and a weaker line at 157.523 nm. The separation between these two lines is about 107 nm which is much larger than the FWHM specification of 0.2 to 0.5 pm. Each line has a bandwidth of about 1.2 pm, which is also significantly higher than the required FWHM. Therefore, in order for F
2
laser to succeed in microlithography, its spectrum has to be line-narrowed. The weaker line should be suppressed as much as feasible, and the stronger line should be line-narrowed by at least ⅓ times.
One of the first choice for line-narrowing technique for F
2
laser might be diffraction grating-based technique, described above in
FIG. 1
, because this technique has been already proven in KrF and ArF excimer lasers to deliver line-narrowed pulses with pulse energy in the range of 5-10 mJ.
Unfortunately, when this technique is applied for F
2
laser line-narrowing, the pulse energy output becomes very small. There are a number of reasons for that. Because of line-narrowing requirements it is necessary to use high magnification prism beam expander in F
2
laser just like in KrF or ArF lasers. For example, the beam expansion coefficient for F
2
laser should typically be at least 20X. That requires at least three prisms to be used in beam expander as shown in FIG.
1
. Unfortunately, the absorption losses in prisms and losses from the surfaces of the prisms become very big as compared to KrF or ArF lasers. The limited number of materials which can be used for coatings at 157 nm wavelength greatly limits the quality of any anti-reflection coatings which can be applied to surfaces of the prisms in order to reduce the losses. Very high gain of F
2
laser produces a very strong amplified spontaneous radiation which increases divergence of the beam and broadens its spectrum. In addition to all that, the reflectivity of eschelle diffraction grating is also smaller at 157 nm as compared to 248 nm or 193 nm wavelengths of KrF or ArF lasers.
Because of all these factors, the typical pulse energy, which such a line-narrowed laser can produce is much less than 5 mJ and is not enough for microlithography applications. In addition to that the lifetime of anti-reflection coatings on optical components, such as prisms is too low, and correspondingly the cost of operation of such a system would be too high.
Therefore, there is a need to develop a more efficient line-narrowing scheme, which would have a higher lifetime of the components and preferably would provide higher pulse energy of line-narrowed light.
SUMMARY OF THE INVENTION
The present invention provides an F
2
laser having an etalon-bas
Cymer Inc.
Jr. Leon Scott
Ross, Esq. John R.
LandOfFree
Line narrowed F2 laser with etalon based output coupler does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Line narrowed F2 laser with etalon based output coupler, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Line narrowed F2 laser with etalon based output coupler will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2476118