JP3126722B2 - Laser device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a discharge excitation type laser device used as a light source for material processing, projection exposure, and the like.

[Conventional technology]

The discharge excitation type laser device is used for material processing such as marking, cutting, welding, etc. In addition, large-scale integration 8 (LSI)
It is used as a light source for optical lithography for the production of the circuit pattern.

For material processing, CO ₂ , excimer laser, or the like is used. In optical lithography, a reduction projection method is mainly used. In this reduction projection method, transmitted light of an original (reticle) pattern illuminated by an illumination light source is transmitted to a photosensitive material on a semiconductor substrate by a reduction projection optical system. Projection forms a circuit pattern. The resolution of this projected image is limited by the wavelength of the light source used. Therefore, in order to improve the resolution, the wavelength of the light source has been gradually shortened from the visible region to the ultraviolet region in recent years. Conventionally, g-line (wavelength 463 nm) and i-line (wavelength 3) generated from a high-pressure mercury lamp as a light source in the ultraviolet region.
65 nm) was used. But the line width of the smallest pattern is
When the size is less than 0.25 μm, which is required for 64 MB, the wavelength of the i-line has already reached the limit for shortening the wavelength. To overcome this technical limitation, Deep Ultra Viole
t; hereafter referred to as Deep-UV). Excimer lasers in particular have high output and high efficiency, and KrF (wavelength 246 nm) and ArF (wavelength 193 nm) depend on the composition of the medium gas.
For example, strong oscillation can be obtained at short wavelengths. Deep above
-In the UV region, the glass and crystal materials constituting the reduction projection lens system are very limited, so that it is difficult to correct the chromatic aberration used in the reduction projection lens system using a mercury lamp.
Therefore, instead of correcting the lens system for chromatic aberration, a wavelength selection element such as an etalon is provided in the laser resonator, and the spectral width of the output light is reduced to such an extent that the chromatic aberration of the lens material can be ignored. With this method, the output which was several nm in the spectral width in the case of natural oscillation can be narrowed to several pm and narrowed.

[Problems to be solved by the invention]

Such a narrow band excimer laser is a discharge excitation type laser device, and as shown in FIG.
A laser gas such as KrF is filled, and this laser gas is discharged and excited by electrodes 6 and 7 disposed vertically along the longitudinal direction of the laser chamber 1 to perform laser oscillation. In the drawing, a portion 11 surrounded by a broken line is a discharge excitation region.
The opposite surfaces 6a and 7a of the electrodes 6 and 7 are consumed by the discharge, and the discharge width is reduced.
WA changes. At the same time, the beam width of the output laser light 4 changes. This poses a problem in terms of stabilizing the beam width. In the same figure, the same figure (b) corresponds to the C of the same figure (a).
FIG. 3C is a longitudinal sectional view of FIG. When used for processing, this change in the beam width of the laser degrades the transverse mode, changes the beam condensing property, and causes a change in output, which poses a practical problem.
In particular, when used as a light source for optical lithography, this change in beam width causes the following inconveniences in practical use of a narrow-band excimer laser.

That is, it is well known that a wavelength selection element used for narrowing a band has an angular dispersion characteristic. For example, when a diffraction grating (grating) is used as a wavelength selection element, high-order diffraction is used to reduce the spectrum width, so that the angular dispersion at the operating point is large and the divergence angle of the laser directly affects the spectrum width. That is, when the divergence angle is large, the spectrum width is widened. Therefore, when the discharge width changes and the gain region changes, the beam divergence of the laser light changes, so that the spectrum width also changes greatly. In order to avoid this fluctuation, a conventional aperture 8 (FIG. 8A) is provided to stabilize the gain width. However, in the conventionally used so-called Chang type electrode or the like, when the output is increased when the electrode is consumed, the discharge width inevitably increases greatly, so that the gain limitation by the aperture 8 makes it difficult to control the output.

Here, a mechanism for increasing the discharge width will be described.
Hereinafter, in the drawings, the electrodes are represented by the CC section of FIG. 8 (b). Townsend theory is known as a macroscopic phenomenology for explaining the discharge phenomenon, and is useful for understanding the discharge phenomenon. According to this theory, a gas containing a halogen gas used in an excimer laser is called a negative gas, and captures electrons generated by impact ionization (ionization coefficient: α) of electrons during discharge due to a large electron affinity of halogen. (Electron attachment coefficient: η) and apparent ionization coefficient (α-η) are reduced to facilitate concentration of discharge. FIG. 9 shows the relationship between these coefficients and the electric field strength E. P is a constant for normalization.

As apparent from FIG. 9, the ionization coefficient α strongly depends on the electric field strength E, and the electron attachment coefficient η hardly depends on the electric field strength E. Therefore, the apparent ionization coefficient (α−η) has a certain electric field strength E As described above, the value rapidly increases and greatly depends on the electric field intensity E. That is, since the parameter (α-η) for driving the discharge changes according to the electric field intensity distribution on the electrode surface, the discharge width is greatly affected by the electric field intensity distribution on the electrode surface. For this reason, in order to secure a discharge width, it is necessary to make the electrode shape of the discharge excitation type laser device such that the electric field intensity region that is uniform in the electrode width direction can be large. Conventionally, based on the analytical solution of electric field calculation under ideal conditions, Chan
g, modified Chang type electrodes and other shapes have been adopted.

Here, FIG. 10 (a) shows the potential distribution of the electric field calculation in the case of the Chang type electrode. In FIG. 1A, L1... Are equipotential lines. From the figure, the cathode at the negative voltage potential (upper electrode)
It can be seen that the potential distribution between the anode 6 and the anode (lower electrode) 7 is largely bent by the metal plate on which the anode 7 is mounted, the current return wiring 10 and the insulating member 9. That is, it is understood that the electric field ideally deviates from the Chang type electric field due to such wiring and insulation in the actual configuration.

FIG. 10 (b) shows the isoelectric field intensity line L2 near the cathode electrode 6.
And (c) shows the electric field intensity distribution along the surface of the cathode electrode 6. The electric field strength on the surface of the electrode 6 clearly shows that the electric field intensity on the surface of the electrode 6 is less than one-third of the distance from the electrode center point A to the right, and there is little intersection with the uniform electric field intensity line. It can be seen that is formed. As shown by the line L3 in FIG. 3C, the electric field intensity first rises very slowly from the center point of the electrode (point A) to the right end point (point B) of the electrode, and rapidly rises toward the end point. You can see the situation. Also, FIG.
In the vicinity, there are shown equal electric field strength lines L4, and FIG. 7E shows a change in electric field strength along the anode electrode surface. The electric field strength at the center point is E ₀ . As shown in these figures, the electrode center point (A
It can be seen that the electric field intensity is uniform in the first section of about 4 mm from (point), and then gradually weakens toward the right end point (point B) of the electrode, and then rapidly weakens after leaving about 12 mm (lines L4, L5). reference). This has a tendency opposite to that of the cathode 6 described above. Although the discharge occurs at the center of the electrode, it is considered that the factor limiting the width of the discharge is on the anode 7 side where the electric field intensity is large at the center.

The shape of the electrode at the central part of the depleted electrode was measured 1 × 10 ⁸ shots after the start of discharge, and the electric field intensity distribution at the central part of the depleted electrode was calculated by a simulation analysis technique using the shape measurement data. The simulation results are shown in FIG. 11 with the horizontal axis representing the rightward distance from the electrode center point A and the vertical axis representing the rate of change ΔE / E ₀ of the electric field intensity E with respect to the electric field intensity E ₀ at the central point A. FIG. 2A shows the cathode 6, and FIG. 1B shows the anode 7. The white circles indicate before the start of discharge, and the black circles indicate after 1 × 10 ⁸ shots. As is clear from these figures, after 1 × 10 ⁸ shots, the wear has progressed, and the uniform electric field portion has spread in the central portion from the start. The width of this uniform electric field portion substantially corresponds to the observed beam width. This is considered to be due to the fact that the current was concentrated on the uniform electric field portion, which is a strong electric field, and the wear proceeded, and as a result, a wide uniform electric field portion was formed.
As described above, in the conventional electrode, when the operation of the laser device is started, the electrode is worn over time, the uniform electric field portion is expanded, and the beam width of the output laser light is changed accordingly. . This is because the discharge width is smaller than the width of the electrode.

The present invention has been made in view of such circumstances,
It is an object of the present invention to provide a laser device capable of reducing a change in a beam width due to electrode consumption.

[Means for solving the problem]

Therefore, in the present invention, in a laser device that discharges between two electrodes disposed vertically facing each other along the longitudinal direction of the laser chamber and excites a laser gas in the laser chamber to generate laser light, The width of the ground potential electrode of the two electrodes coincides with the length of the major axis, and the vertical center axis of the ground potential electrode in a cross section perpendicular to the longitudinal direction of the laser chamber coincides with the minor axis. Outer contour lines of the cross section facing the other electrode are formed in an elliptical shape.

[Action]

That is, according to such a configuration, the discharge width becomes a constant width defined by the size of the electrode width. Therefore, the beam width of the output laser light becomes constant, and the laser device operates stably.

〔Example〕

Hereinafter, embodiments of the laser device according to the present invention will be described with reference to the drawings. In the embodiment, a narrow-band oscillation excimer laser is assumed as the laser device. The device of this embodiment has basically the same configuration as that shown in FIG. 8, and uses electrodes 12 and 13 described later instead of the electrodes 6 and 7. The laser chamber 1 is filled with a laser gas such as KrF, and the laser gas is discharged by an electrode 12 (cathode electrode) and an electrode 13 (anode electrode) arranged vertically along the longitudinal direction of the laser chamber 1. Excitation causes laser oscillation. The laser light is resonated by an optical resonator composed of a laser chamber 1, a front mirror 2, and a grating 5 as a wavelength selection element, and is output from the front mirror 2 as an effective laser light 4. The grating 5 narrows the oscillation light band and functions as a rear mirror, and has a so-called Littrow arrangement. In the drawing, a portion 11 surrounded by a broken line is a discharge excitation region, 8 is an aperture, 9 is an insulating member, and 10 is an anode current return wiring. WA indicates a discharge width. The electrode 13 is characterized in that a line facing the electrode 12 in a cross section perpendicular to the longitudinal direction of the laser chamber 1, that is, an outer line of a cross section, is made elliptical. The cross sections of the electrodes 12, 13 are symmetrical left and right, and thus the potential distribution during operation of the laser device is symmetrical with respect to the center axis of the cross section. So electrode 1
FIG. 1 shows the potential distribution as representative of only the right portions of 2 and 13.

In the figure, the upper side is a cathode electrode 12 having an arc-shaped opposing surface 12a having a radius of 13 mm, and the lower side is an anode electrode 13 having an electrode width of 8 mm and an opposing surface 13a formed in an elliptical shape as described above. . In this embodiment, the anode electrode 13 is formed such that the lateral width thereof matches the length of the major axis of the ellipse, and the central axis of the cross section matches the minor axis of the ellipse. The electrode interval between the cathode electrode 12 and the anode electrode 13 is 25 mm. That is, the width of the electrode is smaller than the electrode interval.

FIG. 3A shows a simulation analysis result when the minor / major axis ratio (r = minor axis / major axis) of the ellipse of the opposing surface 13a of the anode electrode 13 is r = 1, that is, when the opposing surface 13a is an arc. FIG. 4B shows a simulation analysis result when the minor axis ratio r = 2, and FIG. 7C shows a simulation analysis result when the minor axis ratio r = 4. In the drawings up to FIG. 5, (a), (b) and (c) respectively show the minor axis ratio r =
Cases 1, 2, and 4 are shown respectively. The tendency of the entire equipotential line is similar to the conventional one shown in FIG. 10 (a), but the equipotential line near the anode 13 is pushed up by the rod-shaped electrode and the electric potential at the upper end portion (central axis portion) of the electrode is increased. The slope is increasing. This situation can be clearly seen in FIG.

FIG. 3 shows an electric field intensity distribution diagram, FIG. 4 shows details of the electric field intensity distribution near the anode electrode 13, and FIG. 5 shows how the electric field intensity E changes on the surface of the electrode 13. As is apparent from these figures, in the case of the perfect circle of (a), the electric field intensity is large at the upper end portion (center portion) of the electrode and becomes weaker and sharper toward the right end, and has a sharp distribution. In the case of the ellipse (short-major axis ratio of 1: 2) in (b), the electric field distribution has a flat uniform electric field intensity width of about 6 mm with a slight depression at the upper end.
Looking further at the absolute value of the electric field strength at the upper end (a),
It can be seen that the strength is stronger in the order of (b) and (c), and as the elliptical shape becomes flatter, the electric field intensity at the center becomes smaller, and the electric field at the shoulder portion of the uniform electric field intensity region becomes stronger on the contrary.

Now, it is assumed that the operator turns on the operation switch to start the operation of the laser apparatus, and discharge starts from the state of the ellipse (minor-major axis ratio 1: 2) shown in FIG. The electric field is fairly uniform, and the discharge spreads over almost the entire elliptical portion 13a of the electrode 13.
13a Progress simultaneously over the entire surface. As can be seen from FIG. 5 (b), more discharge current locally flows to the central portion where the electric field is strong, so that this portion is consumed more quickly. Eventually, the state approaches (c), the electric field near the center point A becomes weak, and wear is reduced. This time, the electric field in the central peripheral portion D is strengthened, and more discharge current flows in the peripheral portion D where the electric field is strong, and the peripheral portion D is consumed more quickly. Thus, the state returns to the state of FIG. In this way, the shape of the discharge portion is settled in a certain stable shape between (b) and (c). Since the self-shape maintaining mechanism by such a kind of negative feedback mechanism operates, it is possible to stably operate for a long period of time without changing the discharge width WA while maintaining a uniform electric field region. The above was clarified from the simulation results.

On the other hand, experimentally, when the minor axis ratio r is small, the electric field intensity E is strong near the center point A of the electrode and rapidly decreases toward the periphery of the central portion, so that the discharge concentrates in a narrow central region. on the other hand,
If the ratio of the minor axis to the major axis r is excessively increased, the electric field intensity E at the central peripheral portion is increased.
And discharge tends to concentrate on the peripheral portion of the electrode.
That is, there is a certain optimum value of the minor axis ratio r in which the discharge is not locally concentrated. In the actual case, when the ratio of minor axis to major axis is r = 2, the electric field intensity becomes constant over almost the entire area of the opposing surface 13a of the electrode 13, and the discharge width is almost the same as the electrode width.
WA was able to be obtained. As described above, when the opposing surface 13a of the electrode 13 is formed with an optimum ratio of the minor axis to the major axis, and the laser operation is performed, the discharge spreads over almost the entire elliptic portion 13a, and the wear proceeds simultaneously over the entire elliptic portion 13a. Even if there is a locally strong electric field, more discharge current flows there, and it is consumed faster. Eventually, a kind of negative feedback mechanism that weakens the electric field in that part and reduces wear is activated. For this reason, the uniform electric field region is kept constant for a long time, and stable operation without changing the discharge width can be performed. These experimental results show a good agreement with the above simulation results.

In this embodiment, a constant electric field intensity is obtained over the entire surface of the elliptical electrode portion 13a at the minor axis ratio r = 2, and a discharge width WA equal to the electrode width is obtained. Since it varies depending on the width of the electrode, the distance between the electrodes, the geometrical arrangement, the dielectric constant of the insulating portion, and the properties of the discharge medium gas, it is necessary to find an optimum value according to the situation.

In the embodiment, the discharge width WA having the same size as the electrode width is obtained only on one side of the anode electrode 13, but as shown in FIG. 6, the electrodes 12 'and 13' on both sides have the same size as the electrode width. It is also possible to obtain the discharge width WA, and the same effect as that of the embodiment can be obtained.

In the embodiment, the opposing surface has an elliptical shape, but the present invention is not limited to this, and it is sufficient that the electric field intensity distribution on the opposing surface can be uniform even if the opposing surface is not elliptical. As an example, the opposing surface 14a may be a combination of a circular arc and a straight line as in the electrode 14 in FIG. 7 (a), and the opposing surface 15a may be an ellipse as in the electrode 15 in FIG. 7 (b). A combination of straight lines may be used. Further, the facing surface may be formed by a curve of a smooth functional system. In these cases, if the straight lines and the straight lines or the curved lines and the straight lines are connected smoothly, electric field concentration is unlikely to occur, and it is possible to prevent creeping discharge and unnecessary discharge.

〔The invention's effect〕

As described above, according to the present invention, it becomes possible to stably operate a discharge excitation type laser device without changing the discharge width over a long period of time. Therefore, if the present laser device is used for material processing, stable material processing without change in beam width and transverse mode can be performed for a long period of time, which is extremely useful in practical use. Also, by using a wavelength selection element inside or part of the cavity of this laser device, a narrow-band oscillation laser with extremely small fluctuation of the spectrum width can be realized. For a long period of time and is extremely useful in practice.

[Brief description of the drawings]

FIG. 1 is a diagram showing the distribution of equipotential lines near the electrodes used in the embodiment of the laser device according to the present invention, and FIG. 2 is a diagram showing the equipotential lines near the anode electrode shown in FIG. FIG. 3 is a diagram showing the distribution of electric field intensity, FIG. 3 is a diagram showing the distribution of electric field intensity in the device of the embodiment, FIG. 4 is a diagram showing the distribution of electric field intensity near the anode electrode shown in FIG. FIG. 4 is a diagram showing a state of a change in electric field intensity on the surface facing the anode electrode shown in FIG. 4, FIG. 6 and FIG. 7 are diagrams showing the shape of the electrode in another embodiment, and FIG. FIG. 9 is a diagram showing the overall configuration of the apparatus, and FIG. 9 is a graph showing the electric field intensity dependence of the ionization coefficient, adhesion coefficient, and apparent ionization coefficient during excimer laser gas discharge, and FIGS. 10 (a), (b), and (c). ,
(D) and (e) respectively show the potential distribution near both electrodes, the electric field intensity distribution near the cathode electrode, the change in electric field intensity at the surface facing the cathode electrode, and the electric field intensity near the anode electrode when the electrodes according to the prior art are used. FIG. 11 is a graph showing the distribution and the change in the electric field strength on the surface facing the anode electrode, and FIG. 11 is a graph showing the rate of change in the electric field strength on the electrode surface before and after the laser operation when the conventional electrode is used. 1 ... laser chamber, 2 ... front mirror, 4 ...
Laser light, 5 ... Grating, 12 ... Cathode, 13 ... Anode.

Continuation of the front page (56) References JP-A-63-229870 (JP, A) JP-A-64-84676 (JP, A) JP-A-2-192188 (JP, A) JP-A-64-20680 (JP) , A) JP-A-1-302786 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H01S 3/03-3/038

Claims (3)

(57) [Claims] 1. A laser device for generating a laser beam by discharging between two electrodes disposed vertically opposed along a longitudinal direction of a laser chamber to excite a laser gas in the laser chamber. The width of the ground potential electrode of the two electrodes coincides with the length of the major axis, and the vertical center axis of the ground potential electrode in a cross section perpendicular to the longitudinal direction of the laser chamber coincides with the minor axis. A laser device, wherein a contour line facing the other electrode of the contour lines of the cross section is formed in an elliptical shape. 2. The laser device according to claim 1, wherein the ratio of the major axis length to the minor axis length of the ellipse is in the range of 1 to 4. 3. A laser device which discharges between two electrodes disposed vertically opposed along a longitudinal direction of a laser chamber to excite a laser gas in the laser chamber to generate laser light, For only the ground potential electrode of the two electrodes, the width of the ground potential electrode is substantially equal to the discharge width, and the width of the ground potential electrode substantially matched to the discharge width is set between the two electrodes. A laser device characterized by being shorter than the distance.