JP2012104846A - Narrowband laser device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a narrowband laser device which allows for stabilization control of the spectral purity width (spectrum index value) without having any impact on the control of central wavelength.SOLUTION: The spectral purity width E95 of laser light output from a laser device for amplification is measured by spectral purity width measuring means. Discharge timing after starting discharge in a laser device for oscillation and before starting discharge in a laser device for amplification is controlled so that the spectral purity width E95 thus measured falls within the allowable width E95±dE95 of a target spectral purity width E95.

Description

  The present invention relates to a narrow-band laser device, and in particular, in a narrow-band excimer laser device or a narrow-band F2 laser device as a light source of a reduction projection exposure apparatus used for manufacturing a semiconductor, the spectral purity of the laser light The present invention relates to an apparatus for controlling spectral index values such as width.

The prior art of the narrow-band laser apparatus used as the light source of the reduced projection exposure apparatus will be described below for each item.
(Light source for exposure)
As semiconductor integrated circuits are miniaturized and highly integrated, improvement in resolving power is demanded in semiconductor exposure apparatuses. For this reason, the wavelength of light emitted from the exposure light source is being shortened, and a gas laser device is used as the exposure light source in place of the conventional mercury lamp. As the current gas laser apparatus for exposure, a KrF excimer laser apparatus that emits ultraviolet light with a wavelength of 248 nm and an ArF excimer laser apparatus that emits ultraviolet light with a wavelength of 193 nm are used. As a next-generation exposure technology, it is conceivable to apply immersion technology to ArF exposure, which shortens the apparent wavelength of the exposure light source by filling the space between the exposure lens and the wafer with liquid and changing the refractive index. ing. In ArF immersion, the apparent wavelength is as short as 134 nm. As a next-generation exposure light source, an F2 laser device that emits ultraviolet light having a wavelength of 157 nm is prominent, and F2 laser immersion exposure may be adopted. F2 immersion is said to shorten the wavelength to 115 nm.
(Exposure optics and chromatic aberration)
A projection optical system is adopted as an optical system of many semiconductor exposure apparatuses. In the projection optical system, chromatic aberration correction is performed by combining optical elements such as lenses having different refractive indexes. At present, there are no optical materials other than synthetic quartz and CaF2 suitable for use as the lens material of the projection optical system in the wavelength range of 248 nm to 115 nm of the laser wavelength which is an exposure light source. For this reason, as the projection lens for KrF excimer laser, an all-refractive type monochromatic lens composed only of synthetic quartz is adopted, and as the projection lens for ArF excimer laser, an all-refractive type composed of synthetic quartz and CaF2 is adopted. Partially achromatic lens is used. However, since the spontaneous amplitudes of KrF and ArF excimer lasers are as wide as about 350 to 400 pm, the use of these projection lenses causes chromatic aberration and lowers the resolution. Therefore, it is necessary to narrow the spectral line width of the laser light emitted from the gas laser device before chromatic aberration can be ignored. For this reason, in the laser device, a narrowband module having a narrowband element (etalon, grating, etc.) is provided in the laser resonator, and the spectral linewidth is narrowed.
(Spectral purity range)
The imaging performance of the exposure apparatus is greatly influenced not only by the full width at half maximum of the spectral waveform of the laser light but also by the bottom component of the spectral waveform. Therefore, a new index value of a spectrum called a so-called spectral purity range has been introduced. This spectral purity range is evaluated by, for example, a spectral width that includes 95% of the total energy.
In order to guarantee the quality of the integrated circuit, it is required to suppress the spectral purity range to 0.5 pm or less, for example.
(Reason to stabilize spectral purity range)
However, in recent years, it has begun to be said that even if the spectral purity range is a value that is significantly narrower than the value designed in the optical system, the quality of the integrated circuit may deteriorate. This is described in Patent Documents 1 (US6721340) and 2 (Japanese Patent Laid-Open No. 2001-267673). For this reason, the spectral purity range needs to be controlled so as to be stable within a certain predetermined allowable range (hereinafter referred to as stabilization control as appropriate).
(Prior art of spectral purity range control)
Patent Documents 1 and 2 describe the stabilization control of the spectral purity range. In this Patent Document 2, a wavelength detector is provided, and a high-speed tuning mechanism is provided in the narrow-band unit. Based on the detected wavelength, a high-speed tuning mechanism is used to oscillate the wavelength minutely and at high speed for each pulse. Thus, an invention is described in which the apparent spectral purity range is controlled to fall within the allowable range. Here, “apparent spectral purity range control” refers to control for artificially obtaining a spectral purity range corresponding to the amplitude by oscillating the center wavelength at each moment and integrating the time.

US6721340 JP 2001-267673 A

However, in the prior art described in Patent Document 2, the center wavelength also changes accompanying with the control of the spectral purity range. For this reason, it is difficult to independently perform the center wavelength control for matching the center wavelength to a desired value and the spectrum line width control for keeping the spectrum purity width within a predetermined allowable width. For this reason, the following problems occur.
1) Although it is desirable to control the center wavelength for each pulse, there is a problem that this is complicated control.
2) In addition, in the situation where the center wavelength is stable, the accuracy of the center wavelength control is not a problem, but it is necessary to dynamically control the wavelength when an instruction to change the target wavelength is issued from the exposure apparatus. In some cases, the accuracy of center wavelength control may be affected.
3) Further, in the initial stage of burst oscillation, a chirping phenomenon in which the center wavelength is greatly shifted occurs.
The present invention has been made in view of such a situation, and it is possible to control the stabilization of the spectral purity width (spectral index value) without affecting the control of the center wavelength, so that the apparent spectral purity width can be controlled. An object of the present invention is to solve the above-mentioned problems that occur due to control.

The first invention is
An oscillation laser device that pulse-oscillates seed light that has been narrowed by repeatedly performing a laser gas discharge operation in the oscillation chamber and a seed that is discharged between the electrodes having the laser gas in the space in the amplification chamber. In a narrow-band laser device including an amplification laser device that amplifies light and outputs the amplified laser light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
The discharge timing from the start of the discharge by the oscillation laser device to the start of the discharge by the amplification laser device is controlled so that the measured spectrum index value is within an allowable range of the target spectrum index value. And a control means.
The second invention is
An oscillation laser device that pulse-oscillates seed light that has been narrowed by repeatedly discharging laser gas in the oscillation chamber, and the seed light that is amplified by discharging the laser gas in the amplification chamber is amplified. In a narrow-band laser device including an amplification laser device that outputs laser light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
And control means for controlling the spectrum index value of the seed light so that the measured spectrum index value is within an allowable range of the target spectrum index value.
The third invention is
An oscillation laser device that pulse-oscillates seed light that has been narrowed by repeatedly discharging laser gas in the oscillation chamber, and the seed light that is amplified by discharging the laser gas in the amplification chamber is amplified. In a narrow-band laser device including an amplification laser device that outputs laser light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
The discharge timing from the start of the discharge by the oscillation laser device to the start of the discharge by the amplification laser device is controlled so that the measured spectrum index value is within an allowable range of the target spectrum index value. And a control means for controlling the spectral index value of the seed light.
4th invention is 1st invention or 3rd invention,
Control by the control means is performed within a range of allowable synchronization width in which the energy of the laser beam output from the amplification laser device is equal to or higher than an allowable level among the allowable width of the target spectrum index value. To do.
The fifth invention is the first invention or the third invention,
Pulse stretching means for extending the laser pulse waveform of the seed light is further provided;
By extending the pulse waveform of the seed light by the pulse stretching means, the allowable synchronization width in which the energy of the laser light output from the amplification laser device is greater than or equal to an allowable level is increased.
The sixth invention is the second invention or the third invention,
The control means is characterized by changing the time from the start of discharge in the oscillation laser device to the rise of the laser pulse to control the spectral index value of the seed light.
The seventh invention is the second invention or the third invention,
The control means changes the pulse waveform of the seed light by changing the molar concentration or partial pressure of the fluorine molecule F2 in the oscillation chamber (hereinafter simply referred to as “the concentration of the fluorine molecule F2”). It is characterized by that.
The eighth invention is the second invention or the third invention,
The control means changes the pulse waveform of the seed light by changing the total gas pressure in the oscillation chamber.
The ninth invention is the second invention or the third invention,
The oscillation laser device performs a main discharge by applying a voltage according to a charging voltage between a pair of electrodes,
The control means is characterized in that the pulse waveform of the seed light is changed by changing the charging voltage.
The tenth invention is the second invention or the third invention,
The oscillation laser device is:
A charging circuit including a peaking capacitor electrically parallel to the pair of discharge electrodes and a second capacitor electrically parallel to the preceding stage of the peaking capacitor, and storing in the second capacitor; The generated charge is transferred to the peaking capacitor, and a voltage corresponding to the charging voltage of the peaking capacitor is applied to the pair of electrodes, and discharging is performed.
The pulse waveform of the seed light is changed by changing the capacitance of the peaking capacitor or / and the second capacitor or / and the capacitance ratio of the second capacitor to the peaking capacitor. .
The eleventh invention is the second invention or the third invention,
The oscillation laser device is:
A charging circuit including a preionization capacitor arranged in parallel with a pair of discharge electrodes is provided, and preionization is performed between the pair of electrodes according to a charging voltage of the preionization capacitor,
The pulse waveform of the seed light is changed by changing the capacitance of the preionization capacitor.
The twelfth invention is the second invention or the third invention,
The oscillation laser device includes:
An output coupler for returning the light generated in the oscillation chamber to the oscillation chamber with a predetermined reflectance;
The pulse waveform of the seed light is changed by changing the reflectance of the output coupler.
The thirteenth invention is the second invention or the third invention,
The control means changes the time from the start of discharge in the oscillation laser device to the rise of the laser pulse, and controls the spectral index value of the seed light,
When the rise time of the pulse waveform of the seed light changes, from the start of discharge in the oscillation laser device to the start of discharge in the amplification laser device according to the change in the rise time of the pulse waveform The discharge timing is changed to control the discharge timing to coincide with a desired synchronization timing.
The fourteenth invention is
An oscillation laser device that pulse-oscillates seed light that has been narrowed by repeatedly discharging laser gas in the oscillation chamber, and the seed light that is amplified by discharging the laser gas in the amplification chamber is amplified. In a narrow-band laser device including an amplification laser device that outputs laser light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
Control means for controlling the narrowing performance of the oscillation laser device so that the measured spectrum index value falls within the allowable range of the target spectrum index value.
The fifteenth invention
An oscillation laser device that pulse-oscillates seed light that has been narrowed by repeatedly discharging laser gas in the oscillation chamber, and the seed light that is amplified by discharging the laser gas in the amplification chamber is amplified. In a narrow-band laser device including an amplification laser device that outputs laser light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
The discharge timing from the start of the discharge by the oscillation laser device to the start of the discharge by the amplification laser device is controlled so that the measured spectrum index value is within an allowable range of the target spectrum index value. And a control means for controlling the band narrowing performance of the oscillation laser device.
In a sixteenth aspect based on the fourteenth aspect or the fifteenth aspect,
The oscillation laser device includes:
Wavefront changing means for changing the wavefront of the light generated in the oscillation chamber;
By changing the wavefront of the light generated in the oscillation chamber by the wavefront changing means, the bandwidth narrowing performance of the oscillation laser device is changed.
In a seventeenth aspect based on the fourteenth aspect or the fifteenth aspect,
The oscillation laser device includes:
An enlargement ratio changing means for changing an enlargement ratio of the light generated in the oscillation chamber;
The expansion rate changing means changes the expansion rate of the light generated in the oscillation chamber, thereby changing the band narrowing performance of the oscillation laser device.
In an eighteenth aspect based on the fourteenth aspect or the fifteenth aspect,
The oscillation laser device includes:
Beam width changing means for changing the beam width of the light generated in the oscillation chamber is provided,
The beam narrowing performance of the oscillation laser device is changed by changing the beam width of the light generated in the oscillation chamber by the beam width changing means.
The nineteenth invention
An oscillation laser device that pulse-oscillates seed light that has been narrowed by repeatedly discharging laser gas in the oscillation chamber, and the seed light that is amplified by discharging the laser gas in the amplification chamber is amplified. In a narrow-band laser device including an amplification laser device that outputs laser light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
Control means for controlling the propagation speed of the acoustic wave generated by the discharge in the oscillation chamber so that the measured spectrum index value falls within an allowable range of the target spectrum index value. .
The twentieth invention is
An oscillation laser device that pulse-oscillates seed light that has been narrowed by repeatedly discharging laser gas in the oscillation chamber, and the seed light that is amplified by discharging the laser gas in the amplification chamber is amplified. In a narrow-band laser device including an amplification laser device that outputs laser light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
The discharge timing from the start of the discharge by the oscillation laser device to the start of the discharge by the amplification laser device is controlled so that the measured spectrum index value is within an allowable range of the target spectrum index value. And control means for controlling the propagation speed of the acoustic wave generated by the discharge in the oscillation chamber.
The 21st invention is the 19th invention or the 20th invention,
Oscillation frequency detection means for detecting the oscillation frequency of the seed light oscillated by the oscillation laser device;
Laser gas temperature changing means for changing the temperature of the laser gas in the oscillation chamber,
Based on the relationship between the oscillation frequency of the seed light, the temperature of the laser gas in the oscillation chamber and the spectral index value, the laser gas temperature is changed according to the detected oscillation frequency of the seed light, and the measured spectral index value Is controlled to be within the allowable range of the target spectrum index value.

The inventor has determined that the spectral purity range E95 (representative spectrum index value) of laser light output from the laser device (two-stage laser device) 2 shown in FIG. It has been found that it is determined by the discharge timing from the start to the start of discharge in the amplification chamber 30 and the spectral purity range E95 of the laser light (seed light) output from the oscillation chamber 10. It has been found that parameters such as the discharge timing and the spectral purity range of the seed light are parameters that can be controlled independently of the center wavelength control.
In addition, in order to eliminate the fluctuation of the spectral purity range due to various fluctuation factors and stabilize within a certain allowable range, the spectral purity range is monitored (measured) with the spectral purity range E95 detector (spectral purity range measuring means) It was found that if the spectral purity range fluctuates from the target spectral purity range, the spectral purity range E95 may be controlled to be the target spectral purity range.
That is, the stabilization control of the spectral purity range is mainly performed by
1) The discharge timing is changed (first invention).
2) The spectral purity range of the seed light output from the oscillation laser device 100 is changed (second invention).
3) The spectral purity range of the seed light output from the oscillation laser device 100 is changed, and the discharge timing is also changed (third invention).
It is realized by three types of means. Among these, 3) has the greatest effect of stabilization control.
The “control of the apparent spectral purity range” described in the prior art changes the wavelength for each pulse, and therefore, depending on the stabilization control of the spectral purity range, the best on the wafer for each pulse. The focus position changes and exposure is performed. Thus, since independent center wavelength control cannot be performed, the problem of the best focus position shift arises.
On the other hand, according to the control means 1), 2), 3) (first invention, second invention, third invention) of the present invention, the spectral purity range can be substantially increased or decreased. On the other hand, the center wavelength can be controlled independently. For this reason, the center wavelength does not change and the position of the best focus does not change while stabilizing the spectral purity range. Since an optimum spectral shape for the projection lens of the exposure apparatus 3 is obtained and there is no position shift of the best focus, the imaging performance of the projection lens can be maintained.
In the first invention, the spectral purity width E95 of the laser beam output from the amplification laser device 300 is measured by the spectral purity width measuring means, and the measured spectral purity width E95 is an allowable width E950 ± of the target spectral purity width E950. The spectral purity range E95 is stabilized and controlled by controlling the discharge timing from the start of the discharge by the oscillation laser device 100 to the start of the discharge by the amplification laser device 300 so as to be within dE95.
That is, the two-stage laser system shown in FIG. 1 differs from the one-stage laser system in that the spectral purity width depends on the discharge timing from the start of discharge in the oscillation chamber 10 to the start of discharge in the amplification chamber 30. E95 can be controlled.
FIG. 4 shows how the spectral purity range E95 changes according to the discharge timing dt.
The dt on the horizontal axis of the graph of FIG. 4 is the discharge timing, that is, the time from the start of discharge in the oscillation chamber 10 to the start of discharge in the amplification chamber 30. The vertical axis on the left side of the graph of FIG. 4 is the spectral purity range E95. As shown in the spectral purity range characteristic L1, the spectral purity range E95 decreases as the discharge timing dt increases (the discharge timing is delayed). I understand. The reason for this will be described with reference to FIG.
FIG. 5 shows a pulse waveform L3 of the seed light. As shown in FIG. 5, the oscillating laser beam serving as the seed light has a temporal spectral purity width E95 distribution, and the spectral purity width E95 decreases as it goes backward of the laser pulse waveform. Therefore, the spectral purity width E95 of the amplified laser light is determined depending on which part of the seed light pulse waveform having the temporal spectral purity width E95 distribution shown in FIG. For example, when the discharge timing dt is delayed and the discharge is synchronized with the latter half of the pulse waveform of the seed light, the seed light having a narrow spectral purity range E95 is amplified, and as a result, the spectral purity of the amplified laser light The width E95 becomes thinner. On the contrary, when the discharge timing dt is advanced and the discharge is synchronized with the first half of the pulse waveform of the seed light, the seed light having a wide spectral purity range E95 is amplified, and as a result, the amplified laser light The spectral purity range E95 becomes thicker.
By using such characteristics, in the first invention, when the monitored (measured) spectral purity range E95 increases, the discharge timing dt is delayed to decrease the spectral purity range E95, and the spectral purity range E95 decreases. In such a case, control is performed so as to increase the spectral purity range E95 by advancing the discharge timing dt.
If the timing of starting discharge in the amplification laser device 300 is matched with the first half of the seed light pulse waveform L3 output from the oscillation laser device 100, light having a wide spectral width in the first half of the seed light pulse waveform L3 is amplified. In contrast, if the second half of the seed light pulse waveform L3 is matched, light having a narrow spectral width in the second half is amplified (see FIG. 5).
On the left vertical axis in FIG. 4, the target value of the spectral purity range is set to E950, and the allowable range is set to E950 ± dE95.
For example, it is assumed that the laser devices 100 and 300 are operated on the curve L1 with the discharge timing dt set to dt0 so that the spectral purity range matches the target value E950.
Here, when the actually measured spectral purity range E95 is wide (when E95 increases), that is, when the curve L1 changes from the curve L1 to the curve L1 (a), as indicated by the arrow LA, the discharge timing dt Is delayed to change from dt0 to dt2. By delaying the discharge timing dt from dt0 to dt2 in this way, the spectral purity range becomes narrow and can be returned to the original target value E950.
Conversely, when the actually measured spectral purity range E95 is narrowed (when E95 is reduced), that is, when the spectrum purity range E1 is changed from the curve L1 to the curve L1 (b), as shown by the arrow LB, the discharge timing dt To change from dt0 to dt1. Thus, by advancing the discharge timing dt from dt0 to dt1, the spectral purity range becomes wide and can be returned to the original target value E950.
However, as shown in FIG. 4, the control width of the spectral purity range (vertical axis in FIG. 4) is related to the control width of the discharge timing dt (horizontal axis in FIG. 4), and the laser output becomes an allowable level or more. It is desirable that it is within a range (vertical axis in FIG. 4) corresponding to the allowable synchronization width (horizontal axis in FIG. 4) (fourth invention).
In the second invention, the spectral purity width E95 of the laser beam output from the amplification laser device 300 is measured by the spectral purity width measuring means, and the measured spectral purity width E95 is the allowable width E950 ± of the target spectral purity width E950. By controlling the spectral purity width E95 of the seed light output from the oscillation laser device 100 so as to be within dE95, the spectral purity width E95 of the laser light output from the amplification laser device 300 is controlled to be stabilized. .
In a sixth invention subordinate to the second invention, the spectral purity width E95 of the seed light is controlled by changing the time from the start of discharge in the oscillation laser device 100 to the rise of the laser pulse, and the amplification laser Stabilization control is performed on the spectral purity range E95 of the laser beam output from the apparatus 300.
That is, as another method for controlling the spectral purity range E95, there is a means for controlling the spectral purity range E95 of the seed light itself. There are three methods for controlling the spectral purity range E95 of the seed light as follows.
2) -1 Method for controlling the rising of the pulse waveform of the seed light (second invention and sixth invention subordinate thereto).
2) -2 Method for controlling narrowband performance (14th invention).
2) -3 Method for controlling propagation speed of acoustic wave (19th invention).
Regardless of which method is used, the spectral purity range E95 itself of the output seed light changes, so that the amplified laser light also changes correspondingly.
First, the second invention and the sixth invention subordinate thereto will be described.
FIG. 15 shows the waveform of the laser pulse of the seed light output from the oscillation laser device 100 as a graph with the horizontal axis representing time and the vertical axis representing laser output.
When the time from the start of discharge in the oscillation laser device 100 to the rise of the pulse is delayed by reducing the small signal gain, the laser pulse waveform changes from the waveform shown by the broken line to the waveform shown by the solid line in FIG. To do. In this way, the spectral purity range can be narrowed by increasing the time from the start of discharge to the rise of the laser pulse waveform, that is, by shifting the laser pulse to the latter half of the time. This is because, as shown in FIG. 11, the photon after passing through the wavelength selection element (LNM16) several times is amplified and the pulse rises, and the spectrum purity width decreases as the number of round trips increases. . On the other hand, by increasing the small signal gain and shifting the laser pulse to the first half, the spectral purity width is widened by the same principle (the spectral purity width increases as the number of round trips decreases).
By utilizing the above characteristics, in the second and sixth inventions, by controlling the rise of the laser pulse waveform of the seed light output from the oscillation laser device 100 (see FIG. 15), The wavelength line width of the spectrum waveform is changed as shown in FIGS. Since the spectral purity width E95 of the seed light changes in this way, the spectral purity width E95 of the laser light amplified and output by the amplification laser device 300 also changes accordingly.
Specifically, as shown in FIG. 17, by changing the fluorine molecule F2 concentration in the oscillation chamber 10, the pulse waveform of the seed light is changed, thereby controlling the rise of the laser pulse, and the spectral purity range. E95 is controlled to be stabilized (sixth invention).
Parameters that can change the laser pulse waveform of the seed light and also change the spectral purity range E95 include the total gas pressure, charging voltage, capacitor capacity of the charging circuit, capacity ratio, reserve, in addition to the F2 concentration. There are the capacity of the ionizing capacitor, the output fogger (OC) reflectivity, etc. (the seventh invention, the eighth invention, the ninth invention, the tenth invention, the eleventh invention, the twelfth invention).
In addition, by implementing the second invention, when the rise time of the pulse waveform of the seed light changes, the oscillation laser device 100 starts discharging according to the change dt of the rise time of the pulse waveform. It is desirable to change the discharge timing from the start to the start of discharge in the amplification laser device 300 so that the discharge timing coincides with a desired synchronization timing (13th invention).
According to the thirteenth aspect of the present invention, it is possible to suppress a decrease in laser light output and to effectively control the spectral purity range.
In the fourteenth invention, the spectral purity width E95 of the laser beam output from the amplification laser device 300 is measured by the spectral purity width measuring means, and the measured spectral purity width E95 is an allowable width E950 ± of the target spectral purity width E950. By controlling the narrowing performance of the oscillation laser device 100 so as to be within dE95, the spectral purity range E95 of the laser light output from the amplification laser device 300 is controlled to be stabilized.
Specifically, the oscillation laser device 100 is provided with wavefront changing means for changing the wavefront of the light generated in the oscillation chamber 10, and the wavefront changing means changes the wavefront of the light generated in the oscillation chamber 10. By changing, the band narrowing performance of the oscillation laser device 100 is changed, and the spectral purity range E95 of the laser beam output from the amplification laser device 300 is controlled to be stabilized (the sixteenth invention).
Other parameters that can control the spectral band width E95 by controlling the band narrowing performance of the oscillation laser device 100 include a light magnification factor and a light beam width (17th and 18th inventions).
In the nineteenth aspect of the invention, the spectral purity width E95 of the laser beam output from the amplification laser device 300 is measured by the spectral purity width measuring means, and the measured spectral purity width E95 is the allowable width E950 ± of the target spectral purity width E950. By controlling the propagation speed of the acoustic wave generated by the discharge in the oscillation chamber so as to be within dE95, the spectral purity range E95 of the laser beam output from the amplification laser device 300 is controlled to be stabilized.
Specifically, an oscillation frequency detecting means for detecting the oscillation frequency of the seed light oscillated by the oscillation laser apparatus 100 and a laser gas temperature changing means for changing the temperature of the laser gas in the oscillation chamber 10 are provided. Based on the relationship between the oscillation frequency of light, the temperature of the laser gas in the oscillation chamber 10 and the spectral purity range E95, the laser gas temperature is changed according to the oscillation frequency of the detected seed light, and the amplification laser device 300 Stabilization control is performed on the spectral purity range E95 of the laser beam to be output (21st invention).
The principle applied to the nineteenth invention will be described.
As the gas temperature in the oscillation chamber 10 changes, the spectral purity range E95 changes. This is because the acoustic wave generated by the discharge changes the particle density distribution on the laser optical path and changes the laser wavefront. The gas temperature T [K] is between the acoustic wave propagation velocity v and
v ∝ (T) ^1/2
The relationship is established. Therefore, when the gas temperature is changed, the propagation speed of the acoustic wave changes, the particle density distribution on the laser optical path changes, the laser wavefront changes, and finally the spectral purity range E95 changes.
In addition, since the change in the oscillation frequency affects the acoustic wave, the relationship between the gas temperature and the spectral purity range E95 described above also changes according to the oscillation frequency (FIG. 43).
As described above, since the relationship between the gas temperature and the spectral purity range E95 varies depending on the oscillation frequency, in the nineteenth aspect (the twenty-first aspect), for example, the database shows the correlation between the gas temperature and the spectral purity range E95 (see FIG. 44; L16, L17, and L18) are stored, the correlation corresponding to the current oscillation frequency is read, and the spectral temperature range E95 is controlled by changing the gas temperature based on the read correlation. Therefore, the actual spectral purity range is controlled as the laser gas temperature.
Next, the third invention will be described.
As another method for controlling the spectral purity range E95, there is means 3) that combines the control of the discharge timing in 1) and the control of the spectral purity range E95 of the seed light in 2). In this control means 3) (third invention), it is expected that the synchronization allowable range is expanded when the spectral purity range E95 is controlled within the allowable range. Here, the allowable synchronization width is a discharge timing range in which the energy of the laser beam is equal to or higher than an allowable level. Specifically, the allowable synchronization width will be described with reference to FIG. In FIG. 4, the right vertical axis represents the laser output, and L2 (in the case of the MOPO system) and L2 ′ (in the case of the MOPA system) are laser light output characteristics. The allowable synchronization width refers to the range of the discharge timing dt for the amplified laser light output to fall within a ratio of, for example, 80% of the peak output. If the synchronization tolerance is deviated, the laser output is greatly reduced. Compared to the laser output characteristic L2 of the MOPO system, L2 ′ is a laser output characteristic when the MOPA system is used. As can be seen from a comparison of the laser output characteristics, the MOPA system has an allowable synchronization width compared to the MOPO system. Is small.
For example, when the spectral purity width E95 is considerably narrowed, first, the spectral purity width E95 of the seed light itself is narrowed by controlling the spectral purity width E95 of the seed light, and then the discharge timing of 1) is controlled. If the synchronization timing (FIG. 5) is adjusted to the latter half of the seed light pulse waveform, the spectral purity range E95 can be made considerably narrower. Conversely, when the spectral purity range E95 is considerably thickened, first, the spectral purity range E95 of the seed light itself is increased by the control of 2), and then the synchronization timing (FIG. 5) is set by the control of 1) as the seed light pulse. By matching the first half of the waveform, the spectral purity range E95 can be considerably increased. The order of the seed light E95 control and the discharge timing control can be controlled in reverse to the above description.
FIG. 56 is a timing chart used to explain the effect when the control for changing the discharge timing and the control for changing the pulse waveform (changing the pulse rising timing) are used, with the horizontal axis representing time and the vertical axis Is a seed light output (intensity), and shows a pulse waveform of the seed light.
FIG. 56 (a) is a diagram for explaining a case where the spectral purity range E95 is reduced by delaying the discharge timing and further delaying the pulse waveform of the seed light, and FIG. 56 (b) shows the discharge timing. It is a figure explaining the case where the spectral purity range E95 is enlarged by speeding up the pulse waveform of seed light earlier.
FIG. 56A shows an effect obtained by combining the effect described in FIG. 5 and the effect described in FIG. That is, as shown by an arrow 3A, by delaying the discharge timing, the seed light wavelength portion to be amplified in the seed light pulse waveform L3 has a spectral purity width from a portion L31 where the spectral purity width becomes thicker. The process shifts to the thinned portion L32 (the effect described in FIG. 5). Further, as indicated by an arrow 3B, the pulse waveform of the seed light is delayed from L3 to L3 ′, thereby shifting to a portion L32 ′ in which the spectral purity width is further narrowed in the pulse waveform L3 ′ (in FIG. 15). Explained effect).
The same applies to the case where the spectral purity range E95 is increased. As shown by the arrow 13A in FIG. 56 (b), by increasing the discharge timing, the seed light wavelength to be amplified in the pulse waveform L3 of the seed light. The portion shifts from the portion L32 where the spectral purity width is narrowed to the portion L31 where the spectral purity width is thick (the effect described in FIG. 5). Furthermore, as shown by an arrow 13B, the pulse waveform of the seed light is advanced from L3 to L3 ″, so that the pulse waveform L3 ″ shifts to a portion L31 ″ where the spectral purity width becomes thicker (described in FIG. 15). Effect).
As a result, the control width of the discharge timing dt, that is, the synchronization allowable width where the laser output is equal to or higher than the allowable level is further expanded from 3C to 3D, and the controllability is dramatically improved.
Next, the effects of the fifteenth invention combining the first invention and the fourteenth invention, and the twentieth invention combining the first invention and the nineteenth invention will be described.
FIG. 57 shows the effect of combining the control for changing the discharge timing and the control for changing the spectral purity range of the seed light (control for changing the band narrowing performance or control for changing the propagation speed of the acoustic wave). It is a figure used for explanation, and the horizontal axis indicates the discharge timing dt and the vertical axis indicates the spectral purity range E95 as in FIG. 4 described above.
A characteristic L1 in FIG. 57 corresponds to the characteristic L1 shown in FIG.
In order to increase the spectral purity range E95 so as to coincide with the target spectral purity range E950, first, control is performed to increase the spectral purity range of the seed light. As a result, the characteristic L1 changes to a characteristic L1A in which the spectral purity range becomes larger. Further, control is performed to decrease the discharge timing dt in order to increase the spectral purity range E95. As a result, the discharge timing changes in the direction of decreasing on the characteristic L1A.
In order to reduce the spectral purity range E95 so as to coincide with the target spectral purity range E950, first, control is performed to reduce the spectral purity range of the seed light. As a result, the characteristic L1 changes to a characteristic L1B having a smaller spectral purity range. Furthermore, control is performed to increase the discharge timing dt in order to reduce the spectral purity range E95. As a result, the discharge timing changes in the direction of increasing on the characteristic L1B.
Therefore, the control width 1A of the spectral purity width E95 when only the control for changing the spectral purity width of the seed light is combined with the control for changing the spectral purity width of the seed light and the control for changing the discharge timing dt. Comparing the control width 1B in this case, it can be read that the control width (vertical axis) of the spectral purity range E95 is expanded if the two controls are combined within the same synchronization allowable width (horizontal axis).
In this way, by performing control combining both controls, the range in which the spectral purity range E95 can be swung within the same allowable synchronization range is further expanded, and the controllability is greatly improved.
Further, if the control of stretching the discharge pulse of the oscillation chamber 10 is combined with the first invention, the second invention, and the third invention (15th invention and 20th invention) (the fifth invention), the pulse waveform of the seed light Becomes longer, the allowable synchronization width is expanded (see FIG. 14). For this reason, the control range of the spectral purity range E95 can be further increased. Further, since the allowable synchronization width is widened, the output of the amplified laser beam can be suppressed to a small amount with respect to the change of the discharge timing dt, and the laser output is easily stabilized. Become.
As described above, according to the present invention, it is possible to control the stabilization of the spectral purity range (spectral index value) without affecting the control of the center wavelength, and thus control the apparent spectral purity range. It is possible to solve various problems of the prior art.

FIG. 1 is a configuration diagram of a laser system according to the present embodiment. 2A and 2B are configuration diagrams of each chamber and its vicinity. FIGS. 3A and 3B are diagrams showing the configuration of the power supply in an electric circuit. FIG. 4 is a diagram showing the relationship between the discharge timing and the spectral purity range. FIG. 5 is a diagram for explaining that the spectral purity width is determined by the pulse waveform of the seed light and the synchronization timing. FIG. 6 is a flowchart showing a spectral purity range measurement subroutine. FIG. 7 is a flowchart showing a main routine of the spectral purity range stabilization control. FIG. 8 is a configuration diagram of the monitor module. FIG. 9 is a diagram showing the correlation between the measured spectrum index value and the true value. FIG. 10 is a flowchart showing a subroutine for measuring the spectral purity range. FIGS. 11A to 11J are diagrams used to explain the relationship between the number of round trips, the laser pulse waveform, and the spectral purity range. FIG. 12 is a diagram for explaining the relationship between the number of round trips and the spectrum waveform. FIG. 13 is a flowchart showing a discharge timing control subroutine (Example 1). FIG. 14 is a diagram for explaining the effect of pulse stretching. FIG. 15 is a diagram for explaining the change of the spectrum by the pulse waveform control. FIG. 16 is a graph showing the relationship between fluorine concentration, spectral purity range, and laser output. FIG. 17 is a diagram showing the relationship between the fluorine concentration and the laser pulse waveform. FIG. 18 is a flowchart of a fluorine concentration control subroutine (Example 4). FIG. 19 is a flowchart of a subroutine for controlling the total gas pressure (Example 5). FIG. 20 is a flowchart of the charging voltage control subroutine (Embodiment 6). FIGS. 21A and 21B are diagrams illustrating a method of changing the reflectance of the output coupler. 22 (a), 22 (b), and 22 (c) are diagrams showing changes in pulse waveforms due to capacitor capacitance. FIG. 23 is a graph showing the relationship between the peaking capacitor capacity and the spectral purity range. FIG. 24 is a flowchart of a subroutine (Example 9) for controlling the temperature of the peaking capacitor. FIGS. 25A, 25B, and 25C are diagrams for explaining the effect when the discharge timing is fixed with respect to the change in the pulse waveform of the seed light. FIGS. 26A, 26B, and 26C are diagrams for explaining the effect when the discharge timing is changed in accordance with the change in the pulse waveform of the seed light. FIG. 27 is a flowchart of a control subroutine (Embodiment 8) for changing the discharge timing in accordance with the change in the pulse waveform of the seed light. FIG. 28 is a diagram showing the relationship between the radius of curvature of the grating and the spectral purity range. FIG. 29 is a diagram showing a grating bending mechanism. FIG. 30 is a flowchart showing a spectral purity range control subroutine (Embodiment 11) by wavefront correction. FIG. 31 is a diagram illustrating a laser wavefront control system using a deformable mirror. FIG. 32 shows a line type deformable mirror. FIG. 33 is a diagram showing the relationship between the radius of curvature of the laser light wavefront and the spectral purity range. FIG. 34 is a flowchart showing a spectral purity range control subroutine (Embodiment 12) by wavefront correction. FIGS. 35A and 35B are diagrams for explaining spectral purity range control by changing the enlargement ratio (magnification). FIG. 36 is a flowchart showing a subroutine (embodiment 13) for spectral purity range control by changing the enlargement ratio (magnification). FIG. 37 is a flowchart showing a control subroutine for increasing the enlargement ratio. FIG. 38 is a flowchart showing a control subroutine for reducing the enlargement ratio. FIGS. 39A and 39B are diagrams showing the relationship between the enlargement ratio, the spectrum width, and the output of the oscillation laser device. FIG. 40 is a diagram for explaining the spectral width control by the slit width. FIG. 41 is a diagram showing the relationship between the slit width and the spectral width. FIG. 42 is a flowchart showing a subroutine (Example 14) of spectral purity range control by slit control. FIG. 43 is a diagram showing how the oscillation frequency and the spectral purity range change depending on the gas temperature. FIG. 44 is a graph showing the relationship between the gas temperature and the spectral purity range. FIG. 45 is a diagram illustrating a configuration example of a temperature sensor. FIG. 46 is a diagram illustrating a configuration example of a temperature sensor. FIG. 47 is a diagram illustrating a configuration example of a temperature sensor. FIG. 48 is a diagram illustrating a configuration example of a temperature sensor. FIG. 49 is a diagram illustrating a configuration for changing the gas temperature. FIG. 50 is a diagram illustrating a configuration for changing the gas temperature. FIG. 51 is a diagram illustrating a configuration for changing the gas temperature. FIG. 52 is a diagram illustrating a configuration for changing the gas temperature. FIG. 53 is a flowchart (embodiment 15) showing a subroutine of spectral purity range control by gas temperature control. FIG. 54 is a diagram showing a configuration example of a narrowband module using a wavefront corrector. FIGS. 55A and 55B are diagrams showing a configuration example of the wavefront corrector. 56 (a) and 56 (b) are diagrams for explaining the effect of a combination of discharge timing control and pulse waveform control. FIG. 57 is a diagram for explaining the effect of a combination of discharge timing control and seed light spectral purity range control. FIG. 58 is a flowchart showing a spectral purity range control subroutine (Embodiment 16) combining pulse waveform control and discharge timing control. FIG. 59 is a flowchart showing a spectral purity range control subroutine (Embodiment 18) combining the spectral purity range control of seed light and the discharge timing control. FIG. 60 is a flowchart showing a spectral purity range control subroutine (Example 17) combining pulse waveform control and discharge timing control. FIG. 61 is a flowchart showing a spectral purity range control subroutine (Example 19) in which the spectral purity range control of seed light and the discharge timing control are combined. 62A, 62B, and 62C are diagrams for explaining amplification in the case of the MOPA method. 63A, 63B, and 63C are diagrams illustrating amplification in the case of the MOPO method. FIG. 64 is a diagram showing the relationship between the charging voltage, the laser gas total pressure, and the laser pulse waveform.

Embodiments of a narrow-band laser device according to the present invention will be described below with reference to the drawings.
(Spectrum index value)
First, the meaning of terms used in this specification will be described. In this specification, the term “spectral index value” is used to mean including spectral line width, spectral purity width, contrast loss, spectral standard deviation, and white OTF (Optical Transfer Function).
The spectral line width is the full width of the light intensity value of the spectral waveform of the laser beam, and in particular, the full width at half maximum FWHM (Full Width at Half Maximum) obtained by cutting the spectral waveform by half the peak value is often evaluated.
The spectral purity width is the total width of the portion occupied by a certain proportion of energy centered on the center wavelength λ 0 of the total spectral energy, and is often evaluated with a spectral width that contains 95% of energy in particular. In this specification, the spectral purity range is specifically referred to as “E95”. Regarding the spectral purity range E95, the following equation (1) is established, where the wavelength is λ and the center wavelength is λ0.

Contrast loss is an index value given to chromatic aberration by the spectral waveform of laser light. The product of chromatic aberration function p (λ) and spectral waveform g (λ) representing the amount of chromatic aberration for each wavelength in the optical system of the exposure apparatus is the wavelength. Is the integrated value of. Regarding contrast loss, the following equation (2) holds.
CL = ∫g (λ) · p (λ) dλ (2)
The spectrum standard deviation σ is an index value defined by the following expressions (3) and (4).
λ0 = (∫λ · g (λ) dλ) / (∫g (λ) dλ) (3)
σ2 = ∫ [g (λ + λ0) −g (λ0)] 2dλ (4)
The white OTF is an index value that can be obtained from the following equation (5) based on the OTF of monochromatic light and the spectrum waveform.

In the equation (5), Rλ (u, v) is each monochromatic OTF, and Rw (u, v) is white OTF. Wλ is a weight for each wavelength (wavelength intensity distribution).
Hereinafter, the spectral index value will be described by using the spectral purity range as a representative, but the present invention can also be applied to control of other spectral index values.
(2 stage laser system)
Next, an outline of a two-stage laser system as a premise of the present invention will be described.
In recent years, there has been a demand for higher output of a narrow-band laser. A well-known way of accomplishing this is a two-stage laser system. The two-stage laser system is an oscillation laser device that generates narrow-band narrow-width light (seed light; seed laser light) and an amplification laser device that amplifies the intensity of the seed light (seed laser light). A laser system consisting of The two-stage laser system has a feature that, unlike a one-stage laser having only one laser chamber, the output can be increased by an amplification laser device against a decrease in output due to narrowing of the band.
There are two types of two-stage laser systems, the MOPO method and the MOPA method, depending on the means of amplification.
MOPO is an abbreviation for Master Oscillator and Power Oscillator. It is a system in which not only the oscillation chamber that constitutes the oscillation laser device but also the amplification chamber has a resonator. Laser oscillation is possible. On the other hand, MOPA is an abbreviation for Master Oscillator and Power Amplifer, and is a system in which an amplifying chamber is not equipped with a resonator. Without seed light, laser light cannot be extracted.
Hereinafter, unless otherwise specified, light output from the oscillation laser device is referred to as “seed light”, and light output from the amplification laser device is referred to as “laser light”.
62 (a), (b), and (c) are diagrams showing the state of amplification performed by the MOPA method. The waveform of the seed light, the gain curve of the amplification device, and the waveform of the laser light are respectively associated with the time axis. It shows.
In the MOPA method, as shown in FIG. 62, only the portion where the gain curve of the seed light and the amplification laser device overlaps (the portion indicated by the oblique lines) is amplified, and the spectral purity range E95 component of the seed light is amplified as it is. .
On the other hand, the state of amplification performed by the MOPO method is shown in FIG. 63 similar to FIG.
In the MOPO method, as shown in FIG. 63, only the E95 component of the seed light in the portion where the gain curve and the seed light of the amplification laser device first overlap (the portion indicated by hatching) is amplified and the laser light is output. . As described above, in the MOPA method, only the overlapped portion becomes laser light, so that the synchronization allowable width, that is, the time (discharge timing) from the start of discharge in the oscillation laser device to the start of discharge in the amplification laser device. Allowable width is shorter than MOPO method.
(Description of the MOPO laser system according to this embodiment)
Next, the configuration of the laser system of the embodiment will be described using the MOPO method as a representative.
FIG. 1 is a configuration diagram of a laser system according to an embodiment. FIG. 1 shows a MOPO type two-stage laser apparatus. 2A is a diagram showing the configuration of the oscillation chamber shown in FIG. 1 and the vicinity thereof, and FIG. 2B is a diagram showing the configuration of the amplification chamber shown in FIG. 1 and the configuration in the vicinity thereof. .
The laser system according to the embodiment mainly includes a two-stage laser apparatus 2, and an exposure apparatus 3 is provided at the subsequent stage. The two-stage laser apparatus 2 mainly includes an oscillation laser apparatus (OSC) 100 that oscillates seed light that has been narrowed by repeatedly discharging laser gas in the oscillation chamber 10, and an amplification chamber. It comprises an amplification laser device (AMP) 300 that amplifies the seed light by discharging the laser gas within 30 and outputs the amplified laser light. As described above, the oscillation laser device 100 generates the narrow-band seed light, and the amplification laser device 300 amplifies the seed light output from the oscillation laser device 100.
The spectral characteristics of the entire two-stage laser apparatus 2 are determined by the spectral characteristics of the seed light output from the oscillation laser 100. The laser output (energy or power) of the two-stage laser device 2 itself is determined by the amplification capability of the amplification laser device 300.
Laser light output from the amplification laser device 300 is input to the exposure device 3, and the input laser light is used for exposure of an exposure target such as a semiconductor wafer.
The oscillation laser device 100 includes an oscillation chamber 10, a charger 11, an oscillation high voltage pulse generator 12, a gas supply / exhaust unit 14, a cooling water supply unit 15, an LNM 16, a front mirror 17, The first monitor module 19 and the discharge detector 20 are configured.
The amplification laser device 300 includes an amplification chamber 30, a charger 31, an amplification high voltage pulse generator 32, a gas supply / exhaust unit 34, a cooling water supply unit 35, a rear side mirror 36, and an output. A mirror 37 and a second monitor module 39 are included. Note that the optical resonator formed by the mirrors 36 and 37 is not necessarily an unstable resonator, and the stable resonator and any mirror may be a planar etalon type resonator.
Since the oscillation laser device 100 and the amplification laser device 300 have the same parts, the same parts will be described below as a representative of the oscillation laser apparatus 100.
As shown in FIG. 2A, inside the oscillation chamber 10, a pair of electrodes 10a and 10b, that is, cathodes, which are separated by a predetermined distance, are parallel to each other in the longitudinal direction, and are opposed to the discharge surface. An electrode 10a and an anode electrode 10b are provided.
An example of a power source for applying a voltage to these electrodes 10a and 10b is shown in FIG. FIG. 3A shows a power supply and the inside of the chamber by an electric circuit.
FIG. 3A shows a circuit including a step-up transformer Tr1 in addition to the magnetic pulse compression circuit. Instead of using the circuit of FIG. 3A, the circuit of FIG. 3B may be used. FIG. 3 (b) is a circuit including a reactor L1 for charging the main capacitor C0 instead of the step-up transformer of FIG. 3 (a). Note that the circuit of FIG. 3B is not operated to be boosted by the step-up transformer, and other operations are the same as those of the circuit of FIG. Further, since the configuration and operation of the power source of the oscillation laser device 100 and the power source of the amplification laser device 300 are the same, description of the power source of the amplification laser 300 is omitted. Reference numerals in parentheses () shown in FIGS. 3 (a) and 3 (b) indicate components of the amplification laser device 300.
The circuit configuration and operation will be described below with reference to FIG.
The power source includes a charger 11 and an oscillation high voltage pulse generator 12.
The oscillation high-voltage pulse generator 12 is a two-stage magnetic pulse compression circuit using three magnetic switches SR1, SR2 and SR3 composed of saturable reactors. The magnetic switch SR1 is provided to reduce the switching loss in the solid switch SW, and is also called magnetic assist. For example, a semiconductor switching element such as an IGBT is used for the solid switch SW.
A peaking capacitor Cp is arranged electrically in parallel with the pair of discharge electrodes 10a and 10b, and a capacitor C2 is arranged electrically in parallel before the peaking capacitor Cp. A capacitor C1 is arranged.
In the present embodiment, the magnetic switches SR2, SR3 and the capacitors C1, C2 constitute a two-stage capacitance transfer type circuit.
The voltage of the charger 11 is adjusted to a predetermined value V, and the main capacitor C0 is charged according to the voltage value V. At this time, assuming that the solid switch SW is off, when the main capacitor C0 is fully charged and the solid switch SW is turned on, the voltage applied to both ends of the solid switch SW is mainly the magnetic switch SR1. Take on both ends. When the time integration value of the charging voltage V0 of the main capacitor C0 applied to both ends of the magnetic switch SR1 reaches a limit value determined by the characteristics of the magnetic switch SR1, the magnetic switch SR1 is saturated and becomes conductive. Then, a current flows through the loop of the main capacitor C0, the magnetic switch SR1, the primary side of the step-up transformer Tr1, and the solid switch SW. At the same time, current flows through the secondary side of the step-up transformer Tr1 and the loop of the capacitor C1, the charge stored in the main capacitor C0 is transferred to the capacitor C1, and the capacitor C1 is charged. When the time integral value of the voltage V1 in the capacitor C1 reaches a limit value determined by the characteristics of the magnetic switch SR2, the magnetic switch SR2 is saturated and becomes conductive. Then, a current flows through the loop of the capacitor C1, the capacitor C2, and the magnetic switch SR3, the electric charge stored in the capacitor C1 is transferred to the capacitor C2, and the capacitor C2 is charged. When the time integral value of the voltage V2 in the capacitor C2 reaches a limit value determined by the characteristics of the magnetic switch SR3, the magnetic switch SR3 is saturated and becomes conductive. Then, current flows through the loop of the capacitor C2, the peaking capacitor Cp, and the magnetic switch SR3, and the charge stored in the capacitor C2 is transferred to the peaking capacitor Cp, and the peaking capacitor Cp is charged.
As shown in FIG. 3A, a preionization means including a first electrode 91, a dielectric tube 92, and a second electrode 93 is provided in the oscillation chamber 10. The preionization capacitor Cp ′ is disposed electrically in parallel with the pair of discharge electrodes 10a and 10b. Preionization is performed between the pair of electrodes 10a and 10b according to the charging voltage of the preionization capacitor Cp ′.
Corona discharge for preionization occurs on the outer peripheral surface of the dielectric tube 92 starting from a point where the dielectric tube 92 in which the first electrode 91 is inserted and the second electrode 93 are in contact with each other. As the peaking capacitor Cp is charged, the voltage Vp increases. When the voltage Vp reaches a predetermined voltage, corona discharge is generated on the outer peripheral surface of the dielectric tube 92. This corona discharge generates ultraviolet rays on the outer periphery of the dielectric tube 92, and the laser gas between the pair of electrodes 10a and 10b is preionized. As the charging of the peaking capacitor Cp further proceeds, the voltage Vp of the peaking capacitor Cp increases. When this voltage Vp reaches a certain value (breakdown voltage) Vb, the laser gas between the pair of electrodes 10a, 10b is broken down and main discharge is started. This main discharge excites the laser medium. Thereby, in the case of the oscillation laser device 100, seed light is generated, and in the case of the amplification laser 300 (or amplifier), the injected seed light is amplified. Due to the main discharge, the voltage of the peaking capacitor Cp decreases rapidly and eventually returns to the state before the start of charging. Such a discharge operation is repeatedly performed by the switching operation of the solid-state switch SW, whereby pulse laser oscillation is performed. The switching operation of the solid switch SW is performed based on an external trigger signal. The external controller that sends out this trigger signal is, for example, a synchronous controller 8 to be described later.
In the capacity transfer type circuit shown in FIG. 3 (a), if the inductance of each stage is set to be smaller as it goes to the subsequent stage, a pulse compression operation is realized in which the pulse width of the current pulse flowing through each stage is sequentially narrowed. Is done. As a result, a strong discharge with a short pulse is realized between the pair of electrodes 10a and 10b (between the pair of electrodes 30a and 30b).
As described above, a high voltage pulse is applied to the pair of electrodes 10a and 10b by the power source constituted by the charger 11 and the oscillation high voltage pulse generator 12. When a high voltage pulse is applied to the electrodes 10a and 10b, a discharge is generated between the electrodes 10a and 10b, and the laser gas sealed in the oscillation chamber 10 is excited by this discharge.
The above is the configuration and operation of the charger 11 and the oscillation high-voltage pulse generator 12 of the oscillation laser device 100.
The gas supply / exhaust unit 14 shown in FIG. 1 includes a gas supply system that supplies laser gas into the oscillation chamber 10 and a gas exhaust system that exhausts the laser gas in the oscillation chamber 10.
The gas supply system of the gas supply / exhaust unit 14 supplies laser gas into the oscillation chamber 10. As a result, the laser gas is sealed in the oscillation chamber 10.
When the laser system shown in FIG. 1 is a fluorine molecular (F2) laser system, the gas supply / exhaust unit 14 includes a fluorine (F2) gas and a buffer gas made of helium (He), neon (Ne), or the like. Are supplied to the oscillation chamber 10. When this laser system is a KrF excimer laser system, the gas supply / exhaust unit 14 is composed of krypton (Kr) gas and fluorine (F2) gas, helium (He), neon (Ne), and the like. Buffer gas is supplied to the oscillation chamber 10. When this laser system is an ArF excimer laser system, the gas supply / exhaust unit 14 is made of argon (Ar) gas and fluorine (F2) gas, helium (He), neon (Ne), and the like. Buffer gas is supplied to the oscillation chamber 10. The supply and exhaust of each gas is controlled by opening and closing each valve provided in the gas supply / exhaust unit 14.
Inside the oscillation chamber 10, as shown in FIG. 2A, a cross flow fan 10c is provided. Laser gas is circulated in the chamber 10 by the cross flow fan 10c and sent between the electrodes 10a and 10b.
As shown in FIG. 2A, a heat exchanger 10d is provided inside the oscillation chamber 10. Cooling water is supplied from the cooling water supply unit 15 to the heat exchanger 10d. Accordingly, the heat exchanger 10d performs exhaust heat in the oscillation chamber 10. The supply of the cooling water to the heat exchanger 10d is controlled by opening and closing the valve of the cooling water supply unit 15 (FIG. 1).
Windows 10e and 10f are provided on the laser beam output portion on the optical axis of the laser beam in the oscillation chamber 10. The windows 10e and 10f are made of a material having transparency to laser light, such as CaF2. The two windows 10e and 10f are arranged with their outer surfaces parallel to each other and installed at a Brewster angle to reduce reflection loss with respect to the laser beam, and the linear polarization direction of the laser beam is relative to the window surface. It is installed to be vertical.
The pressure sensor Pl monitors the gas pressure in the oscillation chamber 10. A signal indicating the gas pressure detected by the pressure sensor P 1 is input to the utility controller 5. The temperature sensor Tl monitors the temperature in the oscillation chamber 10. A signal indicating the temperature detected by the temperature sensor T 1 is input to the utility controller 5.
The utility controller 5 uses the detection signal of the pressure sensor P1 as a feedback signal to generate a gas flow rate adjustment signal that indicates opening and closing of each valve of the gas supply / exhaust unit 14 and its opening (or gas flow rate). -Outputs to the exhaust unit 14. When the gas flow rate adjustment signal is input to the gas supply / exhaust unit 14, the gas supply / exhaust unit 14 controls the opening and closing of the valves. Thereby, the gas composition and gas pressure in the oscillation chamber 10 are adjusted to desired values.
The laser output varies depending on the temperature of the laser gas in the oscillation chamber 10. Therefore, the utility controller 5 opens and closes the valve of the cooling water supply unit 15 and its opening (or cooling) in order to adjust the laser gas in the oscillation chamber 10 to a desired temperature using the detection signal of the temperature sensor T1 as a feedback signal. A coolant flow rate adjustment signal for instructing (water flow rate) is generated and output to the coolant supply unit 15. When a cooling water flow rate adjustment signal is input to the cooling water supply unit 15, the opening / closing of each valve is controlled by the cooling water supply unit 15. Thereby, the flow rate of the cooling water supplied to the heat exchanger 10d in the oscillation chamber 10, that is, the amount of exhaust heat is adjusted.
An LNM (narrow band module) 16 is provided outside the oscillation chamber 10 and on the optical axis of the laser beam on the window 10e side (see FIG. 2A). A front mirror 17 is also provided outside the oscillation chamber 10 and on the optical axis of the laser beam on the window 10f side (see FIG. 2A). The LNM 16 includes, for example, an optical element such as a magnifying prism and a grating (diffraction grating) that is a wavelength selection element. The LNM 16 may be configured by an optical element such as an etalon that is a wavelength selection element and a total reflection mirror. The optical element in the LNM 16 and the front mirror 17 constitute a laser resonator.
The first monitor module 19 monitors (measures) laser beam characteristics such as the energy of the laser light that has passed through the front mirror 17, the output line width, and the center wavelength. The first monitor module 19 generates a signal indicating the center wavelength of the laser light and outputs this signal to the wavelength controller 6. The first monitor module 19 measures the energy of the laser beam and outputs a signal indicating this energy to the energy controller 7. The functions of the electrodes 30a and 30b, the cross flow fan 30c, the heat exchanger 30d, and the windows 30e and 30f of the amplification chamber 30 shown in FIG. 2B are the same as those of the oscillation chamber 10 shown in FIG. The functions are the same as those of the corresponding components, that is, the electrodes 10a and 10b, the cross flow fan 10c, the heat exchanger 10d, and the windows 10e and 10f.
Further, a charger 31 provided in the amplification laser device 300, an amplification high voltage pulse generator 32, a gas supply / exhaust unit 34, a cooling water supply unit 35, a second monitor module 39, a pressure sensor P2, and a temperature sensor T2. Are the corresponding components provided on the above-described oscillation laser device 100 side, that is, the charger 11, the oscillation high voltage pulse generator 12, the gas supply / exhaust unit 14, the cooling water supply unit 15, and the first. The functions of the monitor module 19, the pressure sensor P1, and the temperature sensor T1 are the same. On the other hand, the amplification laser device 300 is provided with an unstable resonator described below in place of the laser resonator made of the LNM 16 or the like provided in the oscillation laser 100.
That is, a rear side mirror 36 is provided on the optical axis of the laser beam on the window 30e side (see FIG. 2B) outside the amplification chamber 30, and is also outside the amplification chamber 30. An output mirror 37 is provided on the optical axis of the laser beam on the window 30f side (see FIG. 2B). The rear mirror 36 and the output mirror 37 constitute an unstable resonator. The reflection surface of the rear side mirror 36 is a concave surface, and a hole through which laser light passes from the rear side of the mirror to the reflection surface side is provided at the center. The reflective surface of the rear side mirror 36 is provided with an HR (High Reflection) coat. The reflecting surface of the output mirror 37 is a convex surface, and an HR (High Reflection) coat is applied to the center thereof, and an AR (Anti Reflection) coat is applied to the periphery of the center. As the rear-side mirror 36, a mirror substrate on which an AR coating is applied only to a portion corresponding to the hole may be used instead of using a mirror having a hole in the center. Further, a stable resonator may be used instead of an unstable resonator.
A beam propagation unit 42 including a reflection mirror is provided between the front mirror 17 of the oscillation laser device 100 and the rear side mirror 36 of the amplification laser 300. The laser light (seed light) transmitted through the front mirror 17 is guided to the rear side mirror 36 by the beam propagating unit 42. Further, the laser light guided to the beam propagation part 42 passes through the hole of the rear side mirror 36 via the beam propagation part 42, and enters the amplification chamber 30. The laser light incident on the amplification chamber 30 passes through the amplification chamber 30 and is reflected by the central portion of the output mirror 37. The laser light reflected by the output mirror 37 passes through the amplification chamber 30 and is reflected around the hole of the rear side mirror 36. Further, the laser light reflected by the rear-side mirror 36 passes through the amplification chamber 30, passes around the center of the output mirror 37, and is output. In the amplification chamber 30, discharge is performed between the pair of electrodes 30 a and 30 b in the same manner as the oscillation chamber 10. When discharge occurs when the laser light passes through the discharge part of the amplification chamber 30, that is, between the electrodes 30a and 30b, the power of the laser light is amplified.
A signal indicating the laser beam characteristics monitored by the first monitor module 19 and the second monitor module 39 is input to the wavelength controller 6. The wavelength controller 6 generates a signal for changing the wavelength selected by the wavelength selection element (grating, etalon, etc.) in the LNM 16 so that the center wavelength of the laser beam is a desired wavelength, and outputs this signal to the driver 21. The selection wavelength of the wavelength selection element changes, for example, by changing the incident angle of the laser light incident on the wavelength selection element. The driver 21 adjusts the optical elements in the LNM 16 (for example, an enlargement prism, a total reflection mirror, a grating, etc.) so that the incident angle of the laser beam to the wavelength selection element changes based on the signal received from the wavelength controller 6. Control the posture angle. Note that the wavelength selection control of the wavelength selection element is not limited to that described above. For example, when the wavelength selection element is an air gap etalon, the air pressure (such as nitrogen) in the air gap in the LNM 16 may be controlled, or the gap interval may be controlled.
The energy controller 7 receives a signal indicating the laser beam characteristics (laser light energy) monitored by the first monitor module 19 and the second monitor module 39. Note that the exposure apparatus 3 may be provided with an output monitor 51 that monitors the energy of the laser beam, and a signal monitored by the output monitor 51 may be directly input to the energy controller 7. Further, the signal monitored by the output monitor 51 of the exposure apparatus 3 may be input to the controller 52 of the exposure apparatus 3, and the signal may be sent from the controller 52 to the energy controller 7 on the laser apparatus 2 side. .
The energy controller 7 generates a signal indicating the next charging voltage so as to set the pulse energy to a desired value, and outputs this signal to the synchronous controller 8.
The discharge detectors 20 and 40 detect the discharge start time in the chambers 10 and 30, respectively.
A signal indicating the next charging voltage input from the energy controller 7 and a signal indicating each discharge start time detected by the discharge detectors 20 and 40 are input to the synchronous controller 8. The synchronous controller 8 controls the charging voltage of the charger 11 based on the next charging voltage value input from the energy controller 7 and each discharge start time detected by the discharge detectors 20 and 40.
If the discharge timing of the oscillation chamber 10 and the discharge timing of the amplification chamber 30 are shifted, the laser light (seed light) output from the oscillation chamber 10 is not efficiently amplified in the amplification chamber 30. Therefore, the laser beam (seed light) output from the oscillation chamber 10 is discharged in the amplification chamber 30 at a discharge timing when the discharge region (excitation region) between the pair of electrodes 30a and 30b in the amplification chamber 30 is filled. It needs to be discharged. In order to realize this, the synchronous controller 8 amplifies the timing for outputting the trigger signal to the solid-state switch SW of the oscillation high voltage pulse generator 12 so that the discharge timing amplified efficiently can be obtained. A delay time for outputting a trigger signal to the solid state switch SW of the high voltage pulse generator 32 is determined. By changing the delay time, the synchronization timing changes.
The above is the configuration of the MOPO laser system.
(Description of MOPA method)
Although the MOPO method has been described above, the MOPA method configuration is obtained by removing the rear mirror 36 and the output mirror 37 constituting the laser resonator of the amplification chamber 30 in FIG. In the case of the MOPA method, discharge is started in the amplification chamber 30 at the timing when the seed light enters the amplification chamber 30. The seed light is amplified by passing through a laser medium excited to an upper level by discharge. The number of times the seed light passes through the laser medium in the amplification chamber 30 may be one. For further amplification, mirrors are arranged at both ends of the amplification chamber 30, and the laser in the amplification chamber 30 is provided. The seed light may be configured to pass through the medium several times.
The above is the laser system that is the premise of the configuration of the present invention. The present invention can be applied to any of two types of laser systems, the MOPO method and the MOPA method. However, the MOPO method will be explained below unless otherwise specified.
Hereinafter, after explaining the knowledge of the present invention, specific control examples (each example) will be described.
(Spectrum index values of exposure equipment and exposure light source)
As described above, the imaging performance of the exposure apparatus 3 is affected by the spectral performance of the laser light from the exposure light source, that is, the two-stage laser apparatus 2. In order to maintain imaging performance, it is necessary to stabilize and control at least one of the spectral index values (spectral line width, spectral purity width E95, contrast loss, spectral standard deviation, white OTF (Optical Transfer Function)). . Here, the stabilization control of the spectrum index value is to control the spectrum index value so as to be within the allowable range of the target spectrum index value.
(Reason for necessity to stabilize spectral purity range E95 (necessity))
When the spectral purity width is increased, the imaging performance deteriorates in the exposure apparatus 3 due to the chromatic aberration of the projection lens. On the other hand, as described above, even if the spectral purity range is much narrower than the value designed by the optical system, if the spectral purity range is not within a certain allowable range, It is said that the quality may deteriorate (see Patent Document 1). For this reason, it is said that the spectral purity range needs to be stabilized and controlled within a value having a certain allowable range.
(Fluctuation factor of spectral purity range E95)
On the other hand, the spectral purity range fluctuates over a long period of time due to changes in the optical elements in the LNM 16 or the chambers 10 and 30 with time, thermal load due to oscillation, and the like. Further, since the thermal load changes according to changes in the oscillation pattern, that is, the frequency, duty, number of pulses, and pause time, the spectral purity range changes depending on the laser operation pattern.
(Control means for spectral purity range E95)
Therefore, the present inventor has determined that the spectral purity range E95 (representative spectrum index value) of the laser light output from the laser device 2 to the outside is started after the oscillation chamber 10 starts discharging and then the amplification chamber 30. To be determined by the discharge timing until the discharge is started and the spectral purity range E95 of the laser beam (seed beam) output from the oscillation chamber 10, and the discharge timing and the spectral purity of the seed beam are determined. It has been found that the parameter of width is a parameter that can be controlled independently of the central wavelength control.
In addition, in order to eliminate the fluctuation of the spectral purity range due to the above-mentioned fluctuation factors and stabilize within a certain allowable range, the spectral purity range is monitored (measured) by a spectral purity range E95 detector (spectral purity range measuring means) described later. If the target spectral purity range fluctuates, the control means using each actuator described later may be controlled so that the spectral purity range E95 becomes the target target value. Obtained knowledge.
The stabilization control of the spectral purity range is mainly
1) Change the discharge timing.
2) The spectral purity range of the seed light output from the oscillation laser device 100 is changed.
3) The spectral purity range of the seed light output from the oscillation laser device 100 is changed, and the discharge timing is also changed.
It is realized by three types of means. Among these, 3) has the greatest effect of stabilization control.
The “control of the apparent spectral purity range” described in the prior art changes the wavelength for each pulse, and therefore, depending on the stabilization control of the spectral purity range, the best on the wafer for each pulse. The focus position changes and exposure is performed. Thus, since independent center wavelength control cannot be performed, the problem of the best focus position shift arises.
On the other hand, according to the control means 1), 2) and 3) of the present invention, the spectral purity range can be substantially increased or decreased, while the center wavelength can be controlled independently. it can. For this reason, the center wavelength does not change and the position of the best focus does not change while stabilizing the spectral purity range. Since an optimum spectral shape for the projection lens of the exposure apparatus 3 is obtained and there is no position shift of the best focus, the imaging performance of the projection lens can be maintained.
The control means 1) will be described.
In the two-stage laser system shown in FIG. 1, unlike the one-stage laser system, the spectral purity range E95 is set in accordance with the discharge timing from the start of discharge in the oscillation chamber 10 to the start of discharge in the amplification chamber 30. Can be controlled.
FIG. 4 shows how the spectral purity range E95 changes according to the discharge timing dt.
The dt on the horizontal axis of the graph of FIG. 4 is the discharge timing, that is, the time from the start of discharge in the oscillation chamber 10 to the start of discharge in the amplification chamber 30. The vertical axis on the left side of the graph of FIG. 4 is the spectral purity range E95. As shown in the spectral purity range characteristic L1, the spectral purity range E95 decreases as the discharge timing dt increases (the discharge timing is delayed). I understand. The reason for this will be described with reference to FIG.
FIG. 5 shows a pulse waveform L3 of the seed light. As shown in FIG. 5, the oscillating laser beam serving as the seed light has a temporal spectral purity width E95 distribution, and the spectral purity width E95 decreases as it goes backward of the laser pulse waveform. Therefore, the spectral purity width E95 of the amplified laser light is determined depending on which part of the seed light pulse waveform having the temporal spectral purity width E95 distribution shown in FIG. For example, when the discharge timing dt is delayed and the discharge is synchronized with the latter half of the pulse waveform of the seed light, the seed light having a narrow spectral purity range E95 is amplified, and as a result, the spectral purity of the amplified laser light The width E95 becomes thinner. On the contrary, when the discharge timing dt is advanced and the discharge is synchronized with the first half of the pulse waveform of the seed light, the seed light having a wide spectral purity range E95 is amplified, and as a result, the amplified laser light The spectral purity range E95 becomes thicker. Using such characteristics, when the monitored (measured) spectral purity range E95 is increased, the discharge timing dt is increased to decrease the spectral purity range E95, and when the spectral purity range E95 is decreased, discharging is performed. Control may be performed so as to increase the spectral purity range E95 by decreasing the timing dt.
Next, the control means 2) will be described.
As another method for controlling the spectral purity range E95, there is means for controlling the spectral purity range E95 of the seed light itself. There are three methods for controlling the spectral purity range E95 of the seed light as follows.
2) -1 A method for controlling the rise of the pulse waveform of the seed light.
2) -2 A method of controlling the narrowing performance.
2) -3 A method for controlling the propagation speed of acoustic waves.
Regardless of which method is used, the spectral purity range E95 itself of the output seed light changes, so that the amplified laser light also changes correspondingly.
Next, the control means 3) will be described.
Furthermore, as another method for controlling the spectral purity range E95, there is a means for combining the control of the discharge timing in 1) and the control of the spectral purity range E95 of the seed light in 2). In this control means 3), it is possible to expect an increase in the allowable synchronization width in controlling the spectral purity range E95 within the allowable range. Here, the allowable synchronization width is a discharge timing range in which the energy of the laser beam is equal to or higher than an allowable level. Specifically, the allowable synchronization width will be described with reference to FIG. In FIG. 4, the right vertical axis represents the laser output, and L2 (in the case of the MOPO method) and L2 ′ (in the case of the MOPA method) are laser light output characteristics. The allowable synchronization width refers to the range of the discharge timing dt for the amplified laser light output to fall within a ratio of, for example, 80% of the peak output. If the synchronization tolerance is deviated, the laser output is greatly reduced. Compared to the laser output characteristic L2 of the MOPO system, L2 ′ is a laser output characteristic when the MOPA system is used. As can be seen from a comparison of the laser output characteristics, the MOPA system has an allowable synchronization width compared to the MOPO system. Is small.
For example, when the spectral purity width E95 is considerably narrowed, first, the spectral purity width E95 of the seed light itself is narrowed by controlling the spectral purity width E95 of the seed light, and then the discharge timing of 1) is controlled. If the synchronization timing (FIG. 5) is adjusted to the latter half of the seed light pulse waveform, the spectral purity range E95 can be made considerably narrower. Conversely, when the spectral purity range E95 is considerably thickened, first, the spectral purity range E95 of the seed light itself is increased by the control of 2), and then the synchronization timing (FIG. 5) is set by the control of 1) as the seed light pulse. By matching the first half of the waveform, the spectral purity range E95 can be considerably increased. The order of the seed light E95 control and the discharge timing control can be controlled in reverse to the above description.
If the above-described control means 1), 2), and 3) are combined with a technique to be described later that stretches the discharge pulse of the oscillation chamber 10, the pulse waveform of the seed light becomes longer, so that the allowable synchronization width is expanded. . For this reason, the control range of the spectral purity range E95 can be further increased. Further, since the allowable synchronization width is widened, the output of the amplified laser beam can be suppressed to a small amount with respect to the change of the discharge timing dt, and the laser output is easily stabilized. Become.
(Spectral purity range E95 detector (spectral purity range measurement means))
In order to execute the control in the control means 1), 2), and 3), the value of the actual spectral purity width E95 of the laser beam output from the amplification laser device 300 is required. The spectral purity range E95 of the laser light output from the amplification laser device 300 is detected (measured) by the second monitor module 39 as a spectral purity range E95 detector (spectral purity range measuring means). FIG. 8 shows a configuration diagram of the second monitor module 39. Since the configuration of the first monitor module 19 is the same, the description thereof is omitted.
The second monitor module 39 includes a beam splitter 391, an etalon spectroscope 393, and a photodiode 392.
The etalon spectroscope 393 measures spectral index values such as the spectral purity range E95, and the photodiode 392 measures the laser output intensity. The etalon spectroscope 393 includes a beam diffusing unit 394 such as a diffusing plate or a lens array, an etalon 395, a lens 396, and a sensor array 397. As the sensor array 397, for example, a line sensor in which a plurality of photodiode arrays are arranged one-dimensionally can be used. In this case, the plurality of line sensors are arranged in the order of channels (ch: integer).
In the second monitor module 39, a part of the laser light is sampled by the beam splitter 391 and incident on the etalon spectroscope 393. The laser light incident on the etalon spectroscope 393 is diffused by the beam diffusion means 394 and incident on the etalon 395. The laser light that has passed through the etalon 395 enters the lens 396. A sensor array 397 is installed on the focal plane of the lens 396. Therefore, when the laser light is transmitted through the lens 396, interference fringes are generated on the sensor array 397. From the fringe data on the sensor array 397, linear data of the wavelength and light amount of the laser beam is obtained as a spectrum waveform, and the spectrum purity range E95 is calculated.
In the embodiment, the etalon spectroscope 393 is used. However, as a form of the spectroscope, an angle dispersive optical element may be used. For example, a Czerny-Turner type spectroscope, a spectroscope using a plurality of gratings, or a multipass spectroscope may be used.
The spectroscope has an inherent response characteristic, that is, an instrument function. The measured spectrum waveform is the result of convolution integration of the true spectrum waveform with the device function. Therefore, in order to obtain a true spectrum waveform, the measured spectrum waveform may be deconvolved with an apparatus function. However, since this calculation takes time, the correlation between the spectral purity range actually measured by the spectroscope and the true spectral purity range is stored in advance, and the true spectral purity range is obtained by calculation. It is desirable. FIG. 9 illustrates the correlation L4 between the actual spectral index value (spectral purity range E95) measured by the etalon spectroscope 393 and the true spectral index value (spectral purity range E95) measured by the high resolution spectrometer. ing. The true spectral purity range corresponding to the spectral purity range actually measured by the etalon spectroscope 393 can be obtained from the correlation L4 shown in FIG. However, since the correlation may change minutely, it is necessary to calibrate with a high-resolution spectrometer placed outside regularly.
Next, the main routine of the spectral purity range stabilization control will be described with reference to the flowchart shown in FIG.
That is, as shown in FIG. 7, simultaneously with laser oscillation, the process proceeds to step 101, where a “spectral purity range E95 measurement” subroutine is executed, and the second monitor module 39 measures the spectral purity range E95. The spectral purity range E95 is measured for each pulse. However, the spectral purity range E95 may be evaluated by an average value over n pulses or a moving average value in consideration of the calculation time. In this case, it is desirable to perform a calibration process with a certain interval in order to check whether the measured value is deviated from the actual value. The specific contents of the “spectral purity range E95 measurement” subroutine will be described later (step 101).
The value of the target spectral purity range E95 is set to E950, and the first allowable range for the target spectral purity range E950 is set to E950 ± dE95 (S) (first control threshold dE95 (S)). The first allowable range E950 ± dE95 (S) for the target spectral purity range E950 is set according to the specifications required by the exposure apparatus 3. It is necessary to perform control so that the upper limit value E950 + dE95 (S) required by the exposure apparatus 3 exceeds the lower limit value E950−dE95 (S) and does not fall outside the first allowable width range. is there. Therefore, a second control threshold value dE95 having a predetermined margin (dE95 (S) −dE95), that is, a second allowable width E950 ± dE95 with respect to the target spectral purity range E95 is set. The range of the second control threshold dE95 is 0 ≦ dE95 <dE95 (S). When dE95 = 0, if the measured value of the spectral purity range E95 slightly deviates from the target value E950, the E95 actuator described later operates and stabilization control is executed so that the measured value E95 matches the target value E950. Will be.
After the actual spectral purity range E95 is measured, whether or not the absolute value of the difference between the measured value E95 and the target value E950 is less than or equal to the second control threshold dE95, that is, the measured spectral purity range E95 is 2. It is calculated whether or not it is within the allowable width E950 ± dE95 (step 102).
If the absolute value of the difference between the measured value E95 and the target value E950 is equal to or smaller than the second control threshold dE95, that is, | E95−E950 | ≦ dE95, the stabilization control of the spectral purity range E95 is not executed (in step 102). Decision Yes). On the other hand, if the absolute value of the difference between the measured value E95 and the target value E950 exceeds the second control threshold value dE95, that is, | E95−E950 |> dE95 (determination No in step 102), then measurement is performed. It is determined whether or not the absolute value of the difference between the value E95 and the target value E950 is below the first control threshold dE95 (S) (| E95−E950 | <dE95 (S)) (step 103). As a result, when the absolute value of the difference between the measured value E95 and the target value E950 is equal to or greater than the first control threshold value dE95 (S) (No in Step 103), an error signal is sent to the exposure apparatus 3. The laser oscillation is stopped or the shutter existing between the exposure apparatus 3 and the two-stage laser apparatus 2 so as to prevent the laser beam whose spectral purity width deviates from the first allowable width from entering the exposure apparatus 3. Or close.
On the other hand, when the absolute value of the difference between the measured value E95 and the target value E950 is less than the first control threshold dE95 (S) (Yes in step 103), the measured value E95 is made to match the target value E950. Then, the process proceeds to a subroutine of “stabilization control by the E95 actuator” described later, and the E95 actuator operates to execute the stabilization control (step 104).
The contents of the “stabilization control by the E95 actuator” subroutine in step 104 will be described in each embodiment described later. If the spectral purity range E95 falls within the allowable range as a result of the processing of the “stabilization control by the E95 actuator” subroutine, the process returns to the main routine of FIG.
FIG. 6 shows a “spectral purity range E95 measurement subroutine”.
After the subroutine is started, the spectrum waveform is measured by the second monitor module 39 described with reference to FIG. 8 (step 201). The measured spectrum waveform is deconvolved to calculate a true spectrum waveform (step 202). Next, an average value or a moving average value of the spectral purity range E95 is obtained by calculation (step 203). Next, it is determined whether or not to calibrate the spectrum index value (spectral purity range) (step 204). When it is determined to calibrate the spectrum index value (spectral purity range), the spectrum index value (spectral purity range) is determined. Is calibrated and the spectral purity range E95 is recalculated according to the calibration value (step 205). If it is determined not to calibrate the spectrum index value (spectral purity range), the process returns to the main routine of FIG. 7 as it is (step 206).
FIG. 10 shows a “spectral purity range E95 measurement subroutine” in a form different from FIG.
After the subroutine is started, the spectrum waveform is measured by the second monitor module 39 described with reference to FIG. 8 (step 301). Next, an average value or a moving average value of the spectral purity range E95 is obtained by calculation (step 302).
Next, in order to shorten the calculation time, a true value corresponding to the measured value is obtained based on the correlation L4 between the measured value and the true value of the spectral purity range E95 described in FIG. In this process, the correlation between the etalon spectroscope 393 actually used for measurement and a high resolution spectroscope, which is a different standard, is measured in advance, and the measurement result is correlated with the correlation L4. Is stored in the memory (step 303). Next, it is determined whether or not to calibrate the spectrum index value (spectral purity range) (step 304). When it is determined to calibrate the spectrum index value (spectral purity range), the spectrum index value (spectral purity range) is determined. Is calibrated and the spectral purity range E95 is recalculated according to the calibration value (step 305). If it is determined not to calibrate the spectrum index value (spectral purity range), the process returns to the main routine of FIG. 7 as it is (step 306). As a calibration method in this case, calibration may be performed using the above-described high-resolution spectrometer, or calibration may be performed by calculation from the resolution result of the exposure apparatus 3.
(E95 actuator)
Each E95 actuator that executes the subroutine of “stabilization control by E95 actuator” in step 104 in FIG. 7 will be described in the following embodiments (each control example).
(Stabilization control of spectral purity range E95 by (Example 1: control of discharge timing (control means 1)))
In this embodiment, the spectral purity range E95 of the laser beam output from the amplification laser device 300 is measured by the spectral purity range measuring means, and the measured spectral purity range E95 is the allowable width E950 ± of the target spectral purity range E950. By controlling the discharge timing from the start of the discharge by the oscillation laser device 100 to the start of the discharge by the amplification laser device 300 so as to be within dE95, the spectral purity range E95 is stabilized and controlled. is there.
In the case of the two-stage laser apparatus 2 shown in FIG. 1, the spectral performance can be changed by changing the discharge timing of the two laser apparatuses 100 and 200 as described above. Reference 1 (Japanese Patent Application No. 2002-046328) describes a technique for controlling the spectral line width to 0.2 pm or less using this characteristic. However, the technique described in this reference is a control for setting the spectral line width to 0.2 pm or less, and does not describe a control method for stabilizing the spectral line width within a predetermined allowable width.
Hereinafter, an embodiment in which the spectral purity range is stabilized and controlled using the characteristic that the spectral performance is changed by changing the discharge timing will be described. In general, when the discharge timing is changed, the laser output also changes. However, particularly in the MOPO system, this influence is relatively small, which is advantageous in terms of stabilizing the laser output. In order to further suppress the change in output of the laser output, it is desirable to control the applied voltage, gas pressure, and the like.
First, the principle of controlling the spectral performance by the discharge timing will be described.
In a narrow-band laser, a wavelength dispersion element is inserted in a laser resonator that oscillates in a broadband to narrow the wavelength line width Δλ and the spectral purity width E95. The wavelength dispersion element is a prism, a grating, an etalon, or the like. Since only the selected wavelength is returned to the laser resonance axis by the wavelength dispersion element, only the selected wavelength is amplified and extracted as laser light. By passing through this wavelength dispersion element, light in a wavelength region outside the selected wavelength is removed, so that the greater the number of passes, the narrower the wavelength line width Δλ and the spectral purity width E95. This number of passes is called the number of round trips.
FIG. 11 shows the relationship between the number of round trips, the laser pulse waveform, and the spectral purity range E95.
FIGS. 11A to 11D are views of the laser optical axis when the number of round trips is 0, 1, 2, and 3 as viewed from the side of the chamber 10, and FIG. It is a side view which shows the laser optical axis which overlap | superposed (a)-(d). FIG.11 (k) is a top view of the chamber 10 corresponding to Fig.11 (a)-(e).
11 (f), (g), (h), (i), and (j) are laser pulses corresponding to FIGS. 11 (a), (b), (c), (d), and (e), respectively. The waveform, the wavelength line width Δλ, and the spectral purity width E95. In FIGS. 11F to 11J, the horizontal axis represents time (ns), the left vertical axis represents the laser intensity, and the right vertical axis represents the wavelength line width Δλ and the spectral purity width E95. In FIGS. 11 (f) to 11 (j), the wavelength line width Δλ and the spectral purity range E95 are indicated by x.
FIG. 11A shows the laser optical axis when the photons generated in the chamber 10 are output from the chamber 10 without passing through the wavelength dispersion element (LNM16) (the number of round trips is 0). 11 (f) shows the laser pulse waveform corresponding to FIG. 11 (a), the wavelength line width Δλ, and the spectral purity width E95.
As shown in FIGS. 11A and 11F, the light output from the chamber 10 becomes an ASE (Amplified spontaneous emission) component. The oscillation laser device 100 is provided with an output cover 60 that returns the light generated in the oscillation chamber 10 into the oscillation chamber 10 with a predetermined reflectance. The output coupler 60 has a reflectance of approximately 30%. For this reason, 70% of the ASE generated in the chamber 10 is output to the outside. The remaining 30% of the reflected ASE light generated in the chamber 10 returns to the chamber 10 to become a laser light. Although the ASE output intensity is significantly smaller than the laser output intensity, since the ASE does not pass through the wavelength dispersion element (LNM16), the wavelength line width Δλ and the spectral purity width E95 are wide (see FIG. 11 (f)). .
FIG. 11 (b) shows the laser optical axis when light that has passed through the wavelength dispersion element (LNM 16) once is output from the chamber 10 (the number of round trips is one), and FIG. The laser pulse waveform corresponding to FIG. 11B, the wavelength line width Δλ, and the spectral purity width E95 are shown.
As shown in FIGS. 11 (b) and 11 (g), the ASE light shown in FIG. 11 (a) is reflected by the output coupler 60 and returned into the chamber 10, and passes through the wavelength dispersion element (LNM 16). The wavelength-selected light is returned into the chamber 10 and output to the outside of the chamber 10. The laser beam output at this time is delayed by the time for which the light travels once in the resonator. Since it passes through the wavelength dispersion element (LNM16) once, the wavelength line width Δλ and the spectral purity width E95 are reduced accordingly (see FIG. 11G).
Similarly, FIG. 11C and FIG. 11H show the case where the number of round trips is two. Similarly, FIG. 11D and FIG. 11I show a case where the number of round trips is three.
Thus, as the number of round trips increases, the output start time is further delayed and the spectral purity range E95 is decreased (see FIGS. 11 (h) and 11 (i)).
FIG. 11 (e) is obtained by superimposing the laser optical axes shown in FIGS. 11 (a) to 11 (d). FIG. 11 (j) shows the laser pulse waveform corresponding to FIG. 11 (a). The wavelength line width Δλ and the spectral purity range E95 are shown. The actually observed laser pulse waveform has the shape shown in FIG. As shown in FIG. 11 (e), it can be seen that the wavelength line width Δλ and the spectral purity width E95 become narrower toward the latter half of the laser pulse waveform. In FIGS. 11A to 11E, for convenience of explanation, the laser optical axis reciprocating in the chamber 10 is drawn while being shifted, but actually, they are overlapped on the same optical axis.
The number of round trips shown in FIGS. 11A to 11E and the spectrum waveform of the laser beam corresponding to them are collectively shown in FIG. In FIG. 12, (a) to (e) correspond to (a) to (e) in FIG. 11, respectively. Since the spectral purity range E95 is the width of the region where 95% of the total energy is included, if the optical component of the first half of the laser pulse waveform corresponding to FIGS. The purity range E95 becomes wider.
FIGS. 11F to 11J show the case where the small signal gain of the laser medium is large. When the small signal gain is small, the first half of the laser pulse is not observed because the output intensity is weak, so the first half components of (f) and (g) do not appear, and the rise of the laser pulse waveform is delayed.
Thus, in the narrow-band laser, the values of the spectral line width Δλ and the spectral purity width E95 differ depending on whether the laser pulse waveform is in the first half or the second half. In the first half of the laser pulse waveform, the spectral line width Δλ and the spectral purity width E95 are large, and in the second half of the laser pulse waveform, they are narrowed. In a system that requires synchronization of both laser devices, such as the two-stage laser device 2, as described with reference to FIG. 5, the timing for starting discharge in the amplification laser device 300 is the seed output from the oscillation laser device 100. When matched with the first half of the optical pulse waveform L3, light having a wide spectral width in the first half of the seed optical pulse waveform L3 is amplified, and conversely, when aligned with the latter half of the seed optical pulse waveform L3, the latter half thereof. Thus, light having a narrow spectral width is amplified. Therefore, as described with reference to FIG. 4, the spectral line width Δλ and the spectral purity range E95 become narrower as the discharge timing dt is delayed.
On the left vertical axis of FIG. 4, the target value of the spectral purity range is set to E950, and the second control threshold described above is set to dE95 (the second allowable width is set to E950 ± dE95). Thus, the allowable upper limit of the spectral purity range is E950 + dE95, and the allowable lower limit value is E950−dE95.
For example, it is assumed that the laser devices 100 and 300 are operated on the curve L1 with the discharge timing dt set to dt0 so that the spectral purity range matches the target value E950.
Here, when the actually measured spectral purity range E95 is wide (when E95 increases), that is, when the curve L1 changes from the curve L1 to the curve L1 (a), as indicated by the arrow LA, the discharge timing dt Is delayed to change from dt0 to dt2. By delaying the discharge timing dt from dt0 to dt2 in this way, the spectral purity range becomes narrow and can be returned to the original target value E950.
Conversely, when the actually measured spectral purity range E95 is narrowed (when E95 is reduced), that is, when the spectrum purity range E1 is changed from the curve L1 to the curve L1 (b), as shown by the arrow LB, the discharge timing dt To change from dt0 to dt1. Thus, by advancing the discharge timing dt from dt0 to dt1, the spectral purity range becomes wide and can be returned to the original target value E950.
FIG. 13 shows a flowchart of the first embodiment. FIG. 13 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). (Within ± dE95 (S)) (determination Yes in step 103), the subroutine shown in FIG. 13 is entered.
The processing of the subroutine shown in FIG. 13 is executed by E95, wavelength controller 6 and synchronous controller 8 shown in FIG.
When the subroutine shown in FIG. 13 starts, first, it is determined whether the spectral purity range E95 has become wider or narrower than the target value, that is, whether E95> E950 or not, E95, wavelength controller 6 (FIG. 1). ) (Step 401).
As a result of this determination, if E95> E950, the spectral purity range is wider than the target value, so the value of the discharge timing dt for narrowing the spectral purity range E95 to match the target value ( dt = dt0 + dt ′) is calculated (step 402). On the other hand, when E95> E950 is not satisfied, the spectral purity range is narrower than the target value. Therefore, the discharge timing dt (dt = dt0− for widening the spectral purity range E95 to match the target value). dt ') is calculated (step 403).
The value of the control interval dt ′ may be calculated from the slope k (see FIG. 4) of the correlation curve L1 between the spectral purity range E95 and the discharge timing dt (dt ′ = (| E95−E950 |) / k). In addition, the control may actually be performed in advance, an optimum value that allows smooth control to be obtained, and fixed to that value.
Next, it is determined whether or not the calculated discharge timing dt deviates from the allowable synchronization width (see FIG. 4) (limit detection) (step 404).
When the calculated discharge timing dt deviates from the allowable synchronization width (limit detection), the laser output is greatly reduced, and the original meaning as the two-stage laser device 2 is not made. On the other hand, if the discharge timing dt is within the allowable synchronization range, the laser output only changes slightly. In this case, the charging voltage and gas pressure of the amplification laser device 300 are simultaneously controlled as necessary. This guarantees a laser output above a certain level.
For this reason, by shifting the discharge timing by the control interval dt ', if it falls outside the allowable synchronization range (limit detection), the stabilization control cannot be performed any further, so switch to another E95 control method or control Laser signal oscillation is stopped by sending a disable signal to the main controller 4 (step 405).
If the discharge timing is within the allowable synchronization range (the limit is not detected), the stabilization control is possible, and the process proceeds to the next step 406 to send a command signal for causing the synchronous controller 8 to perform the stabilization control. In response, the synchronous controller 8 changes the discharge timing dt by the control interval dt ′ (step 406).
As described above, in the present embodiment, the control for obtaining the synchronization allowable width in which the energy of the laser beam output from the amplification laser device 300 is equal to or higher than the allowable level among the allowable width E950 ± dE95 with respect to the target spectral purity width E950. Stabilization control is performed within the range.
Next, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered in the same manner as step 101 in FIG. 7, and the actual spectral purity range E95 after changing the discharge timing dt is measured (step 407), it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (step 408). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 409).
On the other hand, if the measured spectral purity range E95 does not fall within the second allowable range (E950 ± dE95) as a result of the determination in step 408, the process proceeds to step 401 again, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 409).
(Example 2: Stabilization control of spectral purity range E95 by discharge timing control with pulse stretch control (control means 1))
In the present embodiment, pulse stretching means for extending the laser pulse waveform of the seed light output from the oscillation laser device 100 is further provided. By extending the pulse waveform of the seed light by this pulse stretching means, an amplification laser is provided. After enlarging the synchronization allowable range in which the energy (output) of the laser beam output from the apparatus 300 is equal to or higher than the allowable level, the discharge timing is controlled in the same manner as in the first embodiment,
The spectral purity range E95 is stabilized and controlled.
Extending the laser pulse waveform by changing the laser discharge characteristics is called (discharge) pulse stretching (control). By applying this pulse stretch control to the oscillation laser device 100, the pulse length of the seed light can be increased and the allowable synchronization width shown in FIG. 4 can be increased.
FIG. 14 is a graph similar to FIG. 4, and shows the relationship between the discharge timing dt and the laser output when pulse stretch control is performed (with pulse stretch) and when pulse stretch control is not performed (without pulse stretch) ( (Laser output characteristics) L6 and L2 are shown in comparison, and discharge timing dt and spectral purity range E95 when pulse stretch control is performed (with pulse stretch) and when pulse stretch control is not performed (without pulse stretch) (Spectral purity range characteristics) L7 and L1 are shown in comparison.
Regarding the spectral purity range characteristics, there is no significant change between the characteristic L7 when the pulse stretch control is performed (with pulse stretch) and the characteristic L1 when the pulse stretch control is not performed (without the pulse stretch), but with respect to the laser output characteristics. When pulse stretch control is performed (with pulse stretch) and the characteristic L2 when pulse stretch control is not performed (without pulse stretch), a large change occurs, and when pulse stretch control is performed (pulse It can be seen that the range of the allowable synchronization width (the width in which the laser output is equal to or greater than the allowable level (for example, 80% of the peak output)) is greatly expanded in the case of stretching. As a result, the controllable width of the discharge timing dt is also increased. As described above, by expanding the allowable synchronization width, the range in which the energy of the laser beam output from the amplification laser device 300 becomes equal to or higher than the allowable level (for example, 80% of the peak output) is expanded, and the discharge timing dt is changed. In this case, there is an advantage that the fluctuation of the laser output can be further reduced.
For this reason, when the pulse timing is applied when the discharge timing control (control means 1) is executed, the control range of the discharge timing dt is widened and becomes more practical. The pulse stretching is realized by pulse stretching means for extending the laser pulse waveform of the seed light.
Next, the contents of the pulse stretch means will be described.
In a high repetition laser, methods for performing discharge pulse stretching include an oscillating current method, a current superposition method, and a simple spiker sustainer method.
The oscillating current method is described in, for example, Reference 2 (Japanese Patent Laid-Open No. 2001-156367). In the oscillating current method, the circuit constant is determined so that the period of the oscillating current flowing between the discharge electrodes 10a and 10b is shortened and the peak value of the current is increased, and the first ½ period of the oscillating current and at least the subsequent period In one half cycle, the laser gas is excited and the laser oscillation operation is continued for pulse stretching.
The current superposition method is described in, for example, Reference 3 (Japanese Patent No. 3427889). The current superposition method will be described with reference to FIG. 3. A primary current for injecting energy from the magnetic pulse compression circuit to the discharge electrodes 10a and 10b via the peaking capacitor Cp, and charging of the peaking capacitor at the final stage of the magnetic pulse compression circuit. And a secondary current for injecting energy from the capacitor C2 to the discharge electrodes 10a and 10b, and the oscillation period of the secondary current is set to be longer than the oscillation period of the primary current. Pulse stretching is performed by performing a laser oscillation operation of one pulse in the first half cycle of the discharge oscillation current waveform in which the polarity of the next current is reversed and at least two half cycles following the discharge half-cycle.
The simple spica sustainer method is described in, for example, Reference 4 (Japanese Patent Laid-Open No. 2000-353839). In the simple spiker sustainer method, a spiker circuit for applying discharge and a sustainer circuit for sustaining discharge are configured and pulse stretched.
In the second embodiment, any of the above-described pulse stretching means may be used. When any of the pulse stretching means is used, the synchronization allowable range shown in FIG. 4 is widened, and the spectral purity range E95 is controlled to be stabilized. The above is effective in expanding the control range of the discharge timing dt.
The details of the control with pulse stretching added are the same as those in the flowchart of FIG.
Example 3 Stabilization Control of Spectral Purity Width E95 by Controlling Rise of Pulse Waveform of Seed Light (Control Unit 2) -1)
In the third and subsequent embodiments, the spectral purity width E95 of the laser light output from the amplification laser device 300 is measured by the spectral purity width measuring means, and the measured spectral purity width E95 is an allowable width of the target spectral purity width E950. The spectral purity range E95 of the laser beam output from the amplification laser device 300 is stabilized by controlling the spectral purity range E95 of the seed beam output from the oscillation laser device 100 so as to be within E950 ± dE95. It is something to control.
In the third embodiment, the spectral purity range E95 of the seed light is controlled by changing the time from the start of the discharge in the oscillation laser device 100 to the rise of the laser pulse, and is output from the amplification laser device 300. The spectral purity range E95 of the laser beam is controlled to be stabilized.
FIG. 15 shows the waveform of the laser pulse of the seed light output from the oscillation laser device 100 as a graph with the horizontal axis representing time and the vertical axis representing laser output.
When the time from the start of discharge in the oscillation laser device 100 to the rise of the pulse is delayed by reducing the small signal gain, the laser pulse waveform changes from the waveform shown by the broken line to the waveform shown by the solid line in FIG. Change. Thus, the spectral purity range can be narrowed by lengthening the time from the start of discharge to the rise of the laser pulse waveform, that is, by shifting the laser pulse to the latter half of the time. The reason for this is as described with reference to FIG. 11. This is because the photon after passing through the wavelength selection element (LNM16) several times is amplified and the pulse rises. As the number of round trips increases, the spectral purity width decreases. It is. On the other hand, by increasing the small signal gain and shifting the laser pulse to the first half, the spectral purity width is widened by the same principle (the spectral purity width increases as the number of round trips decreases).
By using the characteristics as described above to control the rise of the laser pulse waveform of the seed light output from the oscillation laser device 100, the wavelength line width of the spectrum waveform of the seed light is changed to that shown in FIGS. It can be changed as shown in e). Since the spectral purity width E95 of the seed light changes in this way, the spectral purity width E95 of the laser light amplified and output by the amplification laser device 300 also changes accordingly.
As will be described later, parameters that can change the laser pulse waveform and also change the spectral purity range E95 include the concentration of F2, total gas pressure, charging voltage, and output fogger (OC) reflectance. . When these parameters change, not only the spectral purity range E95 but also the laser output changes. For this reason, in the one-stage laser apparatus, the variable range of each parameter is limited, the control range of the spectral purity range E95 is narrow, and the parameter having a large influence on the laser output such as the charging voltage has a spectral purity range. It was difficult to control E95. On the other hand, in the two-stage laser apparatus 2, since the laser output can be controlled by the amplification laser apparatus 300, it is assumed that the output of the seed light output from the oscillation laser apparatus 100 is slightly changed by each parameter change. However, a change in the output of the laser beam output from the amplification laser device 300 does not cause a problem. Therefore, it is possible to control the stabilization of the spectral purity range E95 by controlling each parameter without being limited to the laser output.
(Example 4: Stabilization control of spectral purity range E95 by changing pulse waveform of seed light by changing fluorine F2 concentration to control pulse rising (control means 2) -1))
In this embodiment, the seed light pulse waveform is changed by changing the fluorine molecule F2 concentration in the oscillation chamber 10, thereby controlling the rise of the laser pulse and stabilizing the spectral purity range E95. It is.
FIG. 16 shows the relationship L8 between the concentration NF2 (%) of the fluorine molecule F2 in the oscillation chamber 10 and the spectral purity range E95 (au), the concentration NF2 (%) of the fluorine molecule F2 and the laser output (seed). The relationship L9 with the light intensity) E (au) is shown. As shown in FIG. 16, by reducing the fluorine molecule F2 concentration in the oscillation chamber 10, the spectral purity range E95 can be narrowed. This is also described in Reference 5 (Japanese Patent Application No. 4-312202). By utilizing this phenomenon and controlling the fluorine molecule F2 concentration, it becomes possible to stabilize and control the spectral purity range E95.
FIG. 17 shows the laser pulse waveform of the seed light when the fluorine molecule F2 concentration is changed. The horizontal axis in FIG. 17 is time (ns), and the vertical axis is laser intensity (au). The laser pulse waveform of the seed light when the fluorine molecule F2 concentration is 0.06% is indicated by a broken line, and the laser pulse waveform of the seed light when the fluorine molecule F2 concentration is 0.14% is indicated by a solid line. Moreover, the part corresponding to (a)-(d) of FIG. 12 is shown in FIG.
As shown in FIG. 17, when the fluorine molecule F2 concentration is high, the time from the start of discharge until the output seed light is observed is shortened, that is, the seed light is output in a short time. Become. Therefore, when the fluorine molecule F2 concentration is high, the components (a) and (b) in the first half of the pulse become strong, and the spectral purity width E95 of the seed light is as shown in (a) and (b) of FIG. Become wider.
The reason why the rise of the pulse is accelerated by increasing the concentration of the fluorine molecule F2 is that the small signal gain g in the laser medium is increased by increasing the number density of the fluorine molecule F2. When the reflectivity of the output coupler is R1, the reflectivity of the other mirror constituting the resonator is R2, the absorption coefficient in the laser medium is α, and the length of the laser medium is L, the inside of the resonator is reciprocated once. In this case, the net gain (increase rate) G of the light intensity is expressed by G = R1 · R2 · EXP {2 (g−α) L}. Here, it is assumed that the condition G> 1 where the light intensity is amplified is satisfied. When the initial fluorescence intensity is I0, the intensity I output to the outside after n round trips is I = (1-R1) · I0 · G ^ n−A. A represents the loss received by the propagation system such as the monitor module after output. In order to observe the rising of the laser pulse exceeding 1 [mJ / cm2], for example, as the laser output outside, I> 1 is required. In other words, (1-R1) · I0 · [R1 · R2 · EXP {2 (g−α) L}] ^ n −A> 1 is a condition for starting the laser pulse. Therefore, when the small signal gain g is large, the above conditions are satisfied with a small number of round trips, so that laser oscillation is started in a short time.
Based on the above principle, by increasing the fluorine molecule F2 concentration, the gain increases, the laser output can be obtained in a short time, and the spectral purity range E95 becomes wide. Thereby, the spectral purity range E95 can be controlled to be stabilized by controlling the fluorine molecule F2 concentration.
FIG. 18 shows a flowchart of the fourth embodiment. FIG. 18 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). (Within ± dE95 (S)) (determination Yes in step 103), the subroutine shown in FIG. 18 is entered.
18 is executed by E95, the wavelength controller 6, the main controller 4, and the utility controller 5 shown in FIG.
When the subroutine shown in FIG. 18 starts, first, the current fluorine molecule F2 concentration value NF2 in the oscillation chamber 10 is detected. The fluorine molecule F2 gas is originally reduced by reacting with the material of the electrodes 10a and 10b by discharge, and the laser output is also reduced accordingly. Therefore, fluorine molecular F2 gas is periodically injected to stabilize the laser output. For example, the amount of fluorine molecule F2 gas consumed per unit shot is measured in advance, and fluorine molecule F2 gas is injected into the oscillation chamber 10 according to the number of shots based on the measured value. Therefore, in order to easily detect the fluorine molecule F2 concentration, it is desirable to record the injection amount and the gas discharge amount of the fluorine F2 gas from the time of laser gas exchange, and calculate from the values of the injection amount and the discharge amount. Alternatively, a concentration meter for measuring the concentration of the fluorine molecule F2 may be attached to the oscillation chamber 10, and the concentration of the fluorine molecule F2 may be detected by reading the value of this concentration meter (step 510).
Next, whether the spectral purity range E95 is wider or narrower than the target value, that is, whether or not E95> E950 is calculated by E95 and the wavelength controller 6 (FIG. 1) (step 501). ).
As a result of this determination, if E95> E950, the spectral purity range is wider than the target value, so the value of the decrease amount dNF2 of the fluorine molecule F2 concentration necessary to narrow the spectral purity range E95. Is calculated and the reduced concentration value is
NF2 = NF2-dNF2
Is calculated (step 502).
On the other hand, when E95> E950 is not satisfied, the spectral purity range is narrower than the target value, and therefore the value of the increase amount dNF2 of the fluorine molecule F2 concentration necessary to widen the spectral purity range E95 is calculated. The increased concentration value is
NF2 = NF2 + dNF2
Is calculated (step 503). The value of the increase / decrease amount dNF2 is obtained from, for example, the correlation L8 between the spectral purity range E95 and the fluorine molecule F2 concentration shown in FIG.
Calculate it.
The spectral purity range E95 varies depending on the values of the total gas pressure and the charging voltage in addition to the fluorine molecule F2 concentration. For this reason, in advance, the relational expression between the spectral purity range E95, the charging voltage V, the fluorine concentration NF2, the total gas pressure TP,
E95 = f (V, NF2, TP)
, And the charging voltage V and the total gas pressure TP at the time of spectral purity range measurement are detected, and these detected values and the fluorine F2 concentration NF2 detected in the above step 510 are expressed by the above relational expression (E95 = f (V, NF2, TP)), and the relationship between the spectral purity range and the fluorine concentration,
E95 = h (NF2)
It is desirable to calculate And from this relational expression,
dNF2 = h-1 (| E95-E950 |)
And the change amount dNF2 of the fluorine concentration corresponding to the displacement | E95−E950 | of the spectral purity range may be calculated. Here, x = h−1 (y) is an inverse function of y = h (x). In addition, an optimal value that allows the control to be performed smoothly by actually performing a control operation in advance may be obtained and fixed to that value.
Next, it is determined whether or not the value obtained by changing the concentration increase / decrease amount dNF2 thus obtained is within the range where the fluorine molecule F2 concentration can actually be varied. This is because if the fluorine molecule F2 concentration is extremely reduced or increased, the gain decreases or the glow discharge becomes unstable and laser oscillation cannot be performed. Therefore, a control range of the fluorine molecule F2 concentration capable of laser oscillation is set, and it is determined whether or not the value changed by the concentration increase / decrease amount dNF2 is out of this concentration control range (limit detection) (step 504). If the limit is detected as a result of this determination, it is determined that the spectral purity range E95 can no longer be stabilized by controlling the fluorine molecule F2 concentration, and the control is switched to another E95 control method or the uncontrollable signal is The laser oscillation is stopped by sending to the controller 4 (step 505).
If the value changed by the density increase / decrease amount dNF2 is within the above-mentioned density control range (the limit is not detected), stabilization control is possible, and the process proceeds to the next step 506 to cause the main controller 4 to perform the stabilization control. In response, the main controller 4 outputs a command signal to the utility controller 5, and the utility controller 5 operates the gas supply / exhaust unit 14 in response to the command signal. By operating the gas supply / exhaust unit 14, gas is injected into the oscillation chamber 10 or exhausted from the oscillation chamber 10, and the fluorine molecule F2 concentration changes by the concentration increase / decrease amount dNF2. Specifically, in the gas supply / exhaust unit 14, an air valve or a mass flow controller of an F2 / Ne gas line out of an air valve of a gas exhaust line connected to the oscillation chamber 10 and a gas intake line connected to the oscillation chamber 10 is provided. Be controlled. When narrowing the spectral purity range, open the valve of the gas exhaust line so that the fluorine concentration is reduced by dNF2 and exhaust by dp1 (= TP · (dNF2 / NF2)), then only dp1 is Ar / Ne. Inject gas. Conversely, when increasing the spectral purity range, open the valve of the gas exhaust line so that the fluorine concentration increases by dNF2, exhaust by dp2 (= dp1 / 0.01x), and then dilute by x% by dp2. Injected F2 / Ne gas (dp2 = dp1 / 0.01x). At this time, although the gas pressure temporarily decreases, in order to eliminate this, gas exhaust and gas injection may be performed simultaneously using a mass flow controller (step 506).
A preferred embodiment of gas injection will be described using an ArF excimer laser as an example. Gas for replenishment is stored in two cylinders 1 and 2. A gas mixture of Ar gas and Ne gas is sealed inside the cylinder 1, and a gas mixture of F 2 gas, Ar gas and Ne gas is sealed inside the cylinder 2. In both cylinders, the partial pressure ratio of Ar gas and Ne gas is substantially the same, and the partial pressure ratio is adjusted and mixed to a ratio suitable for the operation of the ArF excimer laser. The F2 gas in the cylinder 2 has a high partial pressure exceeding the range suitable for the operation of the ArF excimer laser. Therefore, when F2 gas is to be injected into the oscillation chamber, it is injected from the cylinder 2. Since the F2 gas in the cylinder 2 has a high partial pressure, the F2 gas having an appropriate partial pressure can be supplied by being diluted with the gas in the oscillation chamber. When gas other than F2 gas is to be injected, it is injected from the cylinder 1. Even if the injection and exhaust of the laser gas are repeated, the partial pressure ratio between the Ar gas and the Ne gas inside the oscillation chamber can be maintained at a substantially constant value, and the long-term stable operation of the laser can be enabled.
Next, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered in the same manner as in step 101 of FIG. 7, and the actual spectral purity range E95 after changing the fluorine molecule F2 concentration is measured ( Step 507), it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (Step 508). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 509).
On the other hand, if it is determined in step 508 that the measured spectral purity range E95 is not within the second allowable range (E950 ± dE95), the process proceeds to step 510 again, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 509).
The processing of the subroutine described above is performed, and the output intensity of the seed light changes as the fluorine molecule F2 concentration changes (see L9 in FIG. 16). As described above, this is the amplification laser apparatus 300. It is possible to stabilize the laser output by controlling the discharge parameters (total gas pressure, charging voltage, etc.).
(Example 5: Stabilization control of spectral purity range E95 by changing pulse waveform of seed light by changing total gas pressure to control pulse rise (control means 2) -1))
In the present embodiment, the pulse waveform of the seed light is changed by changing the total gas pressure in the oscillation chamber 10, thereby controlling the rise of the laser pulse and stabilizing the spectral purity range E95. is there.
FIG. 64 shows a laser pulse waveform of the seed light when the total gas pressure in the oscillation chamber 10 is changed. In FIG. 64, the horizontal axis represents time, and the vertical axis represents laser output (seed light intensity). The laser pulse waveform of the seed light when the total gas pressure is low is indicated by a broken line, and the laser pulse waveform of the seed light when the total gas pressure is high is indicated by a solid line.
As shown in FIG. 64, by lowering the total gas pressure in the oscillation chamber 10, the rise of the laser pulse is delayed and the latter half of the pulse waveform is strengthened. Conversely, when the total gas pressure is increased, the laser rises faster and the first half of the pulse waveform becomes stronger. This is because when the total gas pressure is increased, the discharge resistance is increased, so that the discharge start voltage is increased, and hence the injection energy from the power supplies 11 and 12 to the laser medium is increased. This is also because the density of the excimer molecule itself increases and the gain increases. As described in the fourth embodiment, the pulse waveform also changes due to the change in gain. Therefore, the relationship between the total gas pressure TP, the spectral purity range E95, and the laser output E is similar to the relationship L8, L9 between the fluorine gas molecule F2 concentration, the spectral purity range E95, and the laser output E shown in FIG. Present.
FIG. 19 shows a flowchart of the fifth embodiment. FIG. 19 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). Within ± dE95 (S) (Yes at step 103), the subroutine shown in FIG. 19 is entered.
The subroutine processing shown in FIG. 19 is executed by E95, wavelength controller 6, main controller 4 and utility controller 5 shown in FIG.
When the subroutine shown in FIG. 19 is started, first, the current total gas pressure TP in the oscillation chamber 10 is detected (step 610).
Next, whether the spectral purity range E95 is wider or narrower than the target value, that is, whether or not E95> E950 is calculated by E95 and the wavelength controller 6 (FIG. 1) (step 601). ).
As a result of this determination, when E95> E950, the spectral purity range is wider than the target value, so the value of the decrease dTP of the total gas pressure TP necessary to narrow the spectral purity range E95. Is calculated and the reduced value is
TP = TP−dTP
Is calculated (step 602).
Conversely, when E95> E950 is not satisfied, the spectral purity range is narrower than the target value, and therefore the value of the increase dTP of the total gas pressure TP necessary for widening the spectral purity range E95 is calculated. , The increased value is
TP = TP + dTP
Is calculated (step 603). The value of the increase / decrease amount dTP may be calculated from, for example, the correlation L8 between the spectral purity range E95 and the total gas pressure similar to the correlation between the spectral purity range E95 and the fluorine molecule F2 concentration shown in FIG.
The spectral purity range E95 varies depending on the value of the fluorine molecule F2 concentration and the charging voltage in addition to the total gas pressure. For this reason, in advance, the relational expression between the spectral purity range E95, the charging voltage V, the fluorine concentration NF2, the total gas pressure TP,
E95 = f (V, NF2, TP)
, And the charging voltage V and the fluorine F2 concentration NF2 at the time of measuring the spectral purity range are detected, and these detected values and the total gas pressure TP detected in the above step 610 are expressed by the above relational expression (E95 = f (V, NF2, TP)), and the relationship between the spectral purity range and the total gas pressure,
E95 = i (TP)
It is desirable to calculate And from this relational expression,
dTP = i-1 (| E95-E950 |)
And the change amount dTP of the total gas pressure corresponding to the displacement amount | E95−E950 | of the spectral purity range may be calculated. Here, x = i−1 (y) is an inverse function of y = i (x). In addition, an optimal value that allows the control to be performed smoothly by actually performing a control operation in advance may be obtained and fixed to that value.
Next, it is determined whether or not the value obtained by changing the total gas pressure increase / decrease amount dTP obtained in this way is within the range in which the total gas pressure can actually be swung. This is because when the total gas pressure is extremely low, laser oscillation cannot be performed due to a decrease in gain, and when the total gas pressure is extremely high, the pressure limit value of the oscillation chamber 10 is reached and the gas is physically increased. This is because the pressure cannot be increased.
Therefore, the control range of the total gas pressure is set, and it is determined whether or not the value changed by the total gas pressure increase / decrease amount dTP is out of the total gas pressure control range (limit detection) (step 604). If the limit is detected as a result of the determination, it is determined that the spectral purity range E95 can no longer be stabilized by controlling the total gas pressure, and the control is switched to another E95 control method, or an uncontrollable signal is sent to the main controller 4. The laser oscillation is stopped either (step 605).
If the value changed by the total gas pressure increase / decrease amount dTP is within the above-mentioned total gas pressure control range (the limit is not detected), stabilization control is possible, and the process proceeds to the next step 606 where the main controller 4 controls the stabilization control. In response to this, the main controller 4 outputs a command signal to the utility controller 5, and the utility controller 5 operates the gas supply / exhaust unit 14 in response to the command signal. By operating the gas supply / exhaust unit 14, gas is injected into the oscillation chamber 10 or exhausted from the oscillation chamber 10, and the total gas pressure TP changes by the total gas pressure increase / decrease amount dTP. Specifically, in the gas supply / exhaust unit 14, the air valve of the gas exhaust line connected to the oscillation chamber 10 and the air valve of the Ar / Ne gas line among the gas intake lines connected to the oscillation chamber 10 are controlled. . When the spectral purity range is narrowed, the gas is exhausted by opening the valve of the gas exhaust line so that the total gas pressure is reduced by dTP.
Conversely, when increasing the spectral purity range, the gas is injected by opening the valve of the Ar / Ne gas line so that the total gas pressure increases by dTP (step 606).
Here, a desirable embodiment of gas injection will be described using an ArF excimer laser as an example. Gas for replenishment is stored in two cylinders 1 and 2. A gas mixture of Ar gas and Ne gas is sealed inside the cylinder 1, and a gas mixture of F 2 gas, Ar gas and Ne gas is sealed inside the cylinder 2. In both cylinders, the partial pressure ratio of Ar gas and Ne gas is substantially the same, and the partial pressure ratio is adjusted and mixed to a ratio suitable for the operation of the ArF excimer laser. The F2 gas in the cylinder 2 has a high partial pressure exceeding the range suitable for the operation of the ArF excimer laser. Therefore, when F2 gas is to be injected into the oscillation chamber, it is injected from the cylinder 2. Since the F2 gas in the cylinder 2 has a high partial pressure, the F2 gas having an appropriate partial pressure can be supplied by being diluted with the gas in the oscillation chamber. When gas other than F2 gas is to be injected, it is injected from the cylinder 1. Even if the injection and exhaust of the laser gas are repeated, the partial pressure ratio between the Ar gas and the Ne gas inside the oscillation chamber can be maintained at a substantially constant value, and the long-term stable operation of the laser can be enabled.
Next, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered in the same manner as in step 101 in FIG. 7, and the actual spectral purity range E95 after changing the total gas pressure is measured (step 607), it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (step 608). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 609).
On the other hand, as a result of the determination in step 608, when the measured spectral purity range E95 does not fall within the second allowable range (E950 ± dE95), the process proceeds to step 610 again, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 609).
As the above-described subroutine processing is performed and the total gas pressure changes, the output intensity of the seed light changes (see L9 in FIG. 16). By controlling the discharge parameters (total gas pressure, charging voltage, etc.), it is possible to stabilize the laser output.
(Embodiment 6: Stabilization control of spectral purity range E95 by changing pulse voltage of seed light by changing charging voltage to control pulse rising (control means 2) -1))
1, 2, and 3, in the oscillation laser device 100, the main discharge is performed by applying a voltage according to the charging voltage of the power supplies 11 and 12 between the pair of electrodes 10 a and 10 b. Is called. In the present embodiment, the pulse voltage of the seed light is changed by changing the charging voltage, thereby controlling the rise of the laser pulse and stabilizing the spectral purity range E95.
FIG. 64 shows a laser pulse waveform of the seed light when the charging voltage V of the charger 11 corresponding to the voltage applied to the electrodes 10a and 10b in the oscillation chamber 10 is changed. In FIG. 64, the horizontal axis represents time, and the vertical axis represents laser output (seed light intensity). The laser pulse waveform of the seed light when the charging voltage V is low is indicated by a broken line, and the laser pulse waveform of the seed light when the charging voltage V is high is indicated by a solid line.
As shown in FIG. 64, by lowering the charging voltage V, the rise of the laser pulse is delayed and the latter half of the pulse waveform becomes stronger. Conversely, when the charging voltage V is increased, the laser rises earlier and the first half of the pulse waveform becomes stronger. The reason for this is that by increasing the voltage charged to the capacitors of the power supplies 11 and 12, the charging speed from the capacitor to the electrodes 10a and 10b is increased at the start of discharge, and the discharge start voltage is also increased. This is because the injection energy from 12 to the laser medium increases and the gain also increases. As described in the fourth embodiment, the pulse waveform also changes due to the change in gain. Therefore, the relationship between the charging voltage V and the spectral purity range E95 and the laser output E exhibits the same behavior as the relationships L8 and L9 between the fluorine gas molecule F2 concentration, the spectral purity range E95 and the laser output E shown in FIG. .
FIG. 20 shows a flowchart of the sixth embodiment. FIG. 20 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). Within ± dE95 (S) (Yes at step 103), the subroutine shown in FIG. 20 is entered.
20 is executed by the E95, the wavelength controller 6, the main controller 4, the energy controller 7, and the synchronous controller 8 shown in FIG.
When the subroutine shown in FIG. 20 starts, first, the current charging voltage V of the power supplies 11 and 12 (charger 11) for applying a voltage to the discharge electrodes 10a and 10b in the oscillation chamber 10 is detected (step 710). .
Next, whether the spectral purity range E95 is wider or narrower than the target value, that is, whether or not E95> E950 is calculated by E95, the wavelength controller 6 (FIG. 1) (step 701). ).
As a result of this determination, if E95> E950, the spectral purity range has become wider than the target value, so the value of the reduction amount dV of the charging voltage V required to narrow the spectral purity range E95 is The calculated and reduced value is
V = V−dV
Is calculated (step 702).
Conversely, when E95> E950 is not satisfied, the spectral purity range is narrower than the target value, so the value of the increase amount dV of the charging voltage V required to widen the spectral purity range E95 is calculated. The increased value is
V = V + dV
Is calculated (step 703). The value of the increase / decrease amount dV may be calculated from, for example, the correlation L8 between the spectral purity range E95 and the charging voltage similar to the correlation between the spectral purity range E95 and the fluorine molecule F2 concentration shown in FIG.
The spectral purity range E95 varies depending on the fluorine molecule F2 concentration and the total gas pressure in addition to the charging voltage. For this reason, in advance, the spectral purity range E95, the relationship between the charging voltage V, the fluorine concentration NF2, the total gas pressure TP
E95 = f (V, NF2, TP)
And the total gas pressure TP and the fluorine F2 concentration NF2 at the time of measuring the spectral purity range are detected, and these detected values and the charging voltage V detected in the step 710 are represented by the above relational expression (E95 = f (V, NF2, TP)) and the relational expression between spectral purity range and charging voltage,
E95 = g (V)
It is desirable to calculate And from this relational expression,
dV = g-1 (| E95-E950 |)
And the change amount dV of the charging voltage corresponding to the displacement amount | E95−E950 | of the spectral purity range may be calculated. Here, x = g−1 (y) is an inverse function of y = g (x). In addition, an optimal value that allows the control to be performed smoothly by actually performing a control operation in advance may be obtained and fixed to that value.
Next, it is determined whether or not the value changed by the charging voltage increase / decrease amount dV thus obtained is within a range where the charging voltage can actually be varied. This is because in the actual oscillation laser device 100, the chargeable range is limited by the performance of the power supplies 11 and 12.
Therefore, the control range of the charging voltage is set, and it is determined whether or not the value changed by the charging voltage increase / decrease amount dV is out of the charging voltage control range (limit detection) (step 704). When the limit is detected, it is determined that the spectral purity range E95 can no longer be stabilized by the control of the charging voltage, and the control is switched to another E95 control method or an uncontrollable signal is sent to the main controller 4. Then, laser oscillation is stopped (step 705).
If the value changed by the charging voltage increase / decrease amount dV is within the above-mentioned charging voltage control range (no limit is detected), stabilization control is possible, and the process proceeds to the next step 706 to perform stabilization control on the main controller 4. In response to this, the main controller 4 outputs a command signal to the synchronous controller 8 via the energy controller 7, and the synchronous controller 8 receives the command signal. The charger 11 is operated to change the charging voltage V by the charging voltage increase / decrease amount dV (step 706).
Next, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered in the same manner as in step 101 of FIG. 7, and the actual spectral purity range E95 after changing the charging voltage is measured (step 707). ), It is determined whether or not the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95) (step 708). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 709).
On the other hand, if it is determined in step 708 that the measured spectral purity range E95 is not within the second allowable range (E950 ± dE95), the process proceeds to step 710 again, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 709).
The processing of the subroutine described above is performed, and the output intensity of the seed light changes as the charging voltage changes (see L9 in FIG. 16). As described above, the discharge of the amplification laser apparatus 300 is performed as described above. The laser output can be stabilized by controlling parameters (total gas pressure, charging voltage, etc.).
(Example 7: Stabilization control of the spectral purity range E95 by changing the pulse waveform of the seed light by changing the reflectance of the output turnbuckle to control the pulse rise (control means 2) -1))
In this embodiment, as shown in FIGS. 11 and 21, the output laser 60 is provided in the oscillation laser device 100 for returning the light generated in the oscillation chamber 10 into the oscillation chamber 10 with a predetermined reflectance. It is assumed that In the present embodiment, the pulse waveform of the seed light is changed by changing the reflectance of the output cover 60, thereby controlling the rise of the laser pulse and stabilizing the spectral purity range E95.
The laser pulse waveform can be changed by changing the reflectance of the output coupler 60 of the laser resonator in the oscillation laser device 100 shown in FIG. If the reflectivity of the output coupler 60 is increased, the latter half of the pulse becomes stronger. Conversely, if the reflectivity is decreased, the first half of the pulse becomes stronger. Therefore, for example, if the output coupler 60 is designed with a reference value of 30% reflectivity and the spectral purity range E95 is thick, the spectral purity range E95 can be increased by increasing the reflectivity from the reference value to 40%. Can be kept within the allowable range. When the spectral purity width E95 is narrowed, the spectral purity width E95 can be widened by reducing the reflectance from the reference value to 20%. Here, the numerical value of the reflectance is merely an example, and is not limited thereto.
FIGS. 21A and 21B illustrate means for changing the reflectance of the output coupler 60. FIG. FIG. 21 is a view of the output coupler 60 as viewed from the oscillation chamber 10 side.
In FIG. 21A, the beam irradiation surface of the output coupler 60 is formed in a horizontally long direction in the figure, and the reflectance increases, for example, in the left direction 61A in the figure and reflected in the right direction 61B in the figure. The reflectance distribution is formed so that the rate is reduced, and the reflectance of the laser beam irradiation position 62 is changed by sliding the beam irradiation surface of the output coupler 60 left and right in the drawing by the slide mechanism 61. The structure of letting it be illustrated.
In FIG. 21 (b), the beam irradiation surface of the output coupler 60 is formed in the circumferential direction, and the reflectance is increased, for example, in the left circumferential direction in the drawing, and the reflectance is increased in the right circumferential direction in the drawing. The reflectance distribution is formed so as to decrease, and the rotation mechanism 63 rotates the beam irradiation surface of the output coupler 60 in the left and right rotation directions 63A and 63B in the drawing, thereby reflecting the irradiation position 62 of the laser beam. The structure of changing the rate is illustrated.
In FIG. 21, a distribution in which the reflectance continuously changes is formed on the beam irradiation surface of the output coupler 60. However, a film in which the reflectance is changed in stages is formed. The reflectivity may be changed stepwise.
(Embodiment 8: Stabilization of spectral purity range E95 by performing control (control means 3) to change the spectral purity range of seed light output from oscillation laser apparatus 100 and to change the discharge timing accordingly) control)
In this embodiment, the spectral purity range E95 of the laser beam output from the amplification laser device 300 is measured by the spectral purity range measuring means, and the measured spectral purity range E95 is the allowable width E950 ± of the target spectral purity range E950. The discharge timing from the start of the discharge by the oscillation laser device 100 to the start of the discharge by the amplification laser device 300 is controlled so that it falls within dE95, and the seed light output from the oscillation laser device 100 is controlled. By controlling the spectral purity range E95, the spectral purity range E95 of the laser light output from the amplification laser device 300 is controlled to be stabilized.
Specifically, when the rise time of the pulse waveform of the seed light changes, the laser device for amplification 300 starts after the oscillation laser device 100 starts discharging according to the change dt of the rise time of the pulse waveform. The control is performed by changing the discharge timing until the discharge is started and matching the discharge timing with a desired synchronization timing.
FIG. 25 is a diagram for explaining the effect when the discharge timing is fixed when the rise of the seed light pulse waveform is changed, and FIG. 26 is the case where the rise of the seed light pulse waveform is changed. It is a figure explaining the effect at the time of changing discharge timing according to the change.
25 (a), (b), and (c), the time axis on the horizontal axis is the same, and the spectral purity range E95 of the laser beam output from the amplification laser device 300 and the oscillation laser device 100, respectively. 2 shows a pulse waveform (vertical axis; seed light output (intensity)) of the seed light output from, and a pulse waveform (vertical axis; laser light output (intensity)) of the laser light output from the laser apparatus 300 for amplification. . The same applies to FIGS. 26 (a), (b), and (c). The pulse waveform indicated by the broken line is the one before the delay, and the pulse waveform indicated by the solid line is the one after the delay.
As shown in FIG. 25, the spectral purity width E95 of the seed light changes by changing the rising edge of the pulse waveform of the seed light (FIG. 25B). However, in this case, when the discharge timing is fixed without shifting the discharge start timing in the amplification laser device 300 (FIG. 25C), the value of the spectral purity width E95 of the laser beam is as shown in FIG. Hardly changes as indicated by P1 (FIG. 25A). In addition, changing the rising edge of the pulse waveform of the seed light may deviate from the allowable synchronization width, and the laser output decreases (see FIG. 25C). However, in the case of the MOPA method, as described with reference to FIG. 62, the portion where the seed light and the gain curve of the amplification laser device 300 overlap is amplified, so that the spectral purity range E95 is maintained even when the discharge timing is fixed. Changes to avoid a decrease in laser output. Therefore, the control shown in FIG. 25 for keeping the discharge timing fixed when the rising edge of the pulse waveform of the seed light is changed is effective for the MOPA method.
On the other hand, as shown in FIG. 26, if the discharge timing is changed by the same dt in accordance with the change dt of the pulse rise time of the seed light (FIG. 26B) (FIG. 26C), the synchronization is performed. Since the timing does not shift, the spectral purity width E95 of the amplified laser light changes from P2 to P3 under the influence of the spectral purity width E95 of the seed light (FIG. 26A), and the laser There is an advantage that the optical output does not change (FIG. 26 (c)). For this reason, when the rising of the pulse waveform of the seed light shown in FIG. 26 is changed, the control of changing the discharge timing in accordance with the change is effective for both the MOPO and MOPA systems.
FIG. 27 shows the control described in FIG. 18, that is, the control of changing the fluorine molecule F2 concentration to change the laser pulse waveform of the seed light (changing the pulse rise time of the seed light), and the discharge according to the change. A processing procedure combined with control for changing timing is shown in a flowchart. In FIG. 27, description of steps 510 and 501 to 509 that are common to FIG. 18 is omitted.
In FIG. 27, step 511 for executing the process of “changing the discharge timing by the change amount dt of the pulse rise time of the seed light” is inserted between step 506 and step 507 in FIG. Different from FIG.
That is, in this embodiment, the fluorine molecule F2 concentration is changed by the concentration increase / decrease amount dNF2, thereby changing the laser pulse waveform of the seed light and changing the pulse rise time of the seed light by dt (step 506). Then, the process of changing the discharge timing by the change amount dt of the pulse rise time of the seed light is executed (step 511).
According to the eighth embodiment, a decrease in laser light output can be suppressed as compared with the fourth embodiment described with reference to FIG.
In FIG. 27, the control described in FIG. 18, that is, the control of changing the fluorine molecule F2 concentration to change the laser pulse waveform of the seed light (changing the pulse rise time of the seed light), and the change However, similarly, the control described in FIG. 19, that is, the total gas pressure is changed to change the laser pulse waveform of the seed light (seed light pulse). The control for changing the rise time) and the control for changing the discharge timing according to the change may be combined. Similarly, similarly, the control described in FIG. 20, that is, the charge voltage is changed. Control to change the laser pulse waveform of the seed light (change the pulse rise time of the seed light) and discharge according to the change It may be combined with the control of changing the imming, and the pulse waveform of the seed light is changed (the pulse rise time of the seed light is changed) by changing the reflectance of the output cover 60 described in FIG. Control) and control of changing the discharge timing in accordance with the change may be combined.
(Example 9: Stabilization control of spectral purity range E95 by changing the pulse waveform of seed light in accordance with the capacitance and capacitance ratio of the magnetic compression circuit capacitor)
In this embodiment, as shown in FIG. 3, it is assumed that the oscillation laser device 100 is provided with a magnetic compression circuit. That is, a charging circuit (magnetic) including a peaking capacitor Cp disposed in parallel with the pair of discharge electrodes 10a and 10b and a second capacitor C2 disposed in parallel in front of the peaking capacitor Cp. A compression circuit) is provided in the oscillation laser device 100, the charge stored in the second capacitor C2 is transferred to the peaking capacitor Cp, and a voltage corresponding to the charging voltage of the peaking capacitor Cp is applied to the pair of electrodes 10a and 10b. It is assumed that a discharge is performed by this. In this embodiment, the pulse waveform of the seed light is changed by changing the capacitance of the peaking capacitor Cp or / and the second capacitor Cp or / and the capacitance ratio C2 / Cp of the peaking capacitor Cp to the second capacitor. This stabilizes the spectral purity range E95.
For example, as shown in FIG. 22C, by increasing the capacitance ratio C2 / Cp, pulse stretching is performed and the pulse waveform of the seed light becomes longer.
First, the principle of the relationship between the capacitance of the capacitor and the pulse waveform of the seed light will be described.
The laser pulse waveform depends on the discharge current waveform, and the discharge current waveform depends on the voltage waveform between the electrodes 10a and 10b. In particular, the laser oscillation is continued even after the first half cycle of the oscillating current when the peak value of the current is large. The voltage at which discharge starts between the main discharge electrodes 10a and 10b (referred to as the breakdown voltage Vb) depends on the rise of the voltage applied between the main discharge electrodes 10a and 10b, and the rise time is fast. Discharge start voltage Vb increases (overvoltage occurs). For example, in order to increase the applied voltage rapidly, if the capacitance of the second capacitor C2 is increased with respect to the capacitance of the peaking capacitor Cp (conversely, if Cp is decreased), the voltage rises sharply. Become. However, the larger the capacitance of the second capacitor C2 relative to the capacitance of the peaking capacitor Cp, the sharper the rise of the voltage is desirable. On the other hand, the larger the capacitance of the second capacitor C2, the more for oscillation. Since the energy required for driving the entire laser device 100 increases and the efficiency of the oscillation laser device 100 decreases, the capacity ratio C2 / Cp has a limit, and there is a case where the capacity ratio is controlled. It is necessary to change within the range.
Further, if the period after the first 1/2 period of the oscillating current flowing between the main discharge electrodes 10a, 10b is shortened, the laser oscillation is continued in the second and subsequent 1/2 periods. This is because if this period is long, spatial concentration of discharge occurs in the latter half of one half period, and the necessary uniform excitation cannot be performed efficiently. The parameters that determine the period of the 1/2 cycle after the second are the capacitance and stray inductance in the loop (discharge current circuit) formed by the peaking capacitor Cp and the main discharge electrodes 10a and 10b. It is proportional to the period. Therefore, in order to shorten the period, the capacitance of the peaking capacitor Cp may be reduced.
FIG. 22 shows changes in (a) voltage between electrodes 10a and 10b, (b) discharge current, and (c) seed light pulse waveform (vertical axis; laser intensity) when the capacitance of the capacitor is changed. . The horizontal axes in FIGS. 22A, 22B, and 22C are a common time axis. In FIG. 22, the broken line is the waveform before the capacitance change, and the solid line is the waveform after the capacitance change.
As described above, when the capacitance of the peaking capacitor Cp decreases and the capacitance ratio C2 / Cp increases, the discharge voltage rises suddenly (changes from a broken line to a solid line in FIG. 22A), and discharge starts. It can be seen that the voltage Vb also increases (see FIG. 22A). As a result, the peak value of the discharge current increases, and the discharge current vibrates for a long time (see FIG. 22B). Along with this, the laser pulse waveform of the seed light also becomes longer (see FIG. 22C). The longer laser pulse means that the latter half of the pulse becomes stronger and the components with narrow spectral purity range increase.
Therefore, in the charging circuit (magnetic compression circuit) of the oscillation laser device 100, the capacitance of the peaking capacitor Cp is reduced, or / and the capacitance of the second capacitor C2 is increased, or / and the capacitance ratio C2 / Cp is increased. As a result, the spectral purity range E95 of the seed light is narrowed. Conversely, by increasing the capacitance of the peaking capacitor Cp or / and decreasing the capacitance of the second capacitor C2 or / and decreasing the capacitance ratio C2 / Cp, the spectral purity width E95 of the seed light is widened.
The capacity of the capacitor can be adjusted by the temperature of the capacitor. For example, as the temperature increases by 20 ° C., the capacitance of the capacitor decreases by 10%.
Therefore, for example, the temperature control temperature of the peaking capacitor Cp is designed to be changeable to 40 ° C. with air cooling, 20 ° C. with water cooling, and 60 ° C. without air cooling.
The relationship L10 between the temperature (° C.) of the peaking capacitor Cp and the spectral purity range E95 (au) is as illustrated in FIG.
That is, as shown in FIG. 23, when the spectral purity range E95 becomes thick, the capacity decreases if the temperature of the peaking capacitor Cp is increased by turning off the air cooling (40 ° C. → 60 ° C.). As a result, the laser pulse width is increased, and the spectral purity width E95 can be narrowed. On the other hand, when the spectral purity range E95 is narrowed, if the temperature of the peaking capacitor Cp is lowered by water cooling (40 ° C. → 20 ° C.), the capacity increases and the laser pulse width decreases. The spectral purity range E95 can be widened (see FIG. 23). For example, when the capacitance ratio C2 / Cp = 5.6 / 8, the capacitance of the peaking capacitor Cp can be changed from 5nF to 6.2nF, so that the capacitance ratio C2 / Cp can be controlled in the range of 0.63 to 0.78. It becomes.
FIG. 24 shows a flowchart of the ninth embodiment. FIG. 24 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). Within ± dE95 (S) (Yes at step 103), the subroutine shown in FIG. 20 is entered.
The processing of the subroutine shown in FIG. 24 is executed by E95, wavelength controller 6 and main controller 4 shown in FIG.
When the subroutine shown in FIG. 24 starts, first, the current temperature Tcp of the peaking capacitor Cp is detected (step 810).
Next, whether the spectral purity range E95 is wider or narrower than the target value, that is, whether or not E95> E950 is calculated by E95 and the wavelength controller 6 (FIG. 1) (step 801). ).
As a result of this determination, if E95> E950, the spectral purity range is wider than the target value, and therefore the value of the increase dTcp in the peaking capacitor temperature Tcp necessary to narrow the spectral purity range E95. Is calculated and the increased value is
Tcp = Tcp + dTcp
Is calculated (step 802).
When E95> E950 is not satisfied, the spectral purity range is narrower than the target value. Therefore, the reduction value dTcp of the peaking capacitor temperature Tcp necessary to widen the spectral purity range E95 is calculated and reduced. Value is
Tcp = Tcp−dTcp
Is calculated (step 803). The value of the increase / decrease amount dTcp may be calculated from the correlation L10 between the temperature of the peaking capacitor and the spectral purity range E95 shown in FIG.
Next, it is determined whether or not the value changed by the temperature increase / decrease amount dTcp of the peaking capacitor Cp thus obtained is within the range where the temperature can actually be varied. This is because the energy efficiency at the time of driving the oscillation laser device 100 changes due to the change in the capacitance of the peaking capacitor Cp due to the temperature increase / decrease. It is.
Therefore, the control range of the peaking capacitor temperature is set, and it is determined whether or not the value changed by the temperature increase / decrease amount dTcp is out of this temperature control range (limit detection) (step 804). If the limit is detected, it is determined that the spectral purity range E95 can no longer be stabilized by controlling the temperature of the peaking capacitor, and the control is switched to another E95 control method, or the uncontrollable signal is sent to the main controller 4 Then, laser oscillation is stopped (step 805).
If the value changed by the temperature increase / decrease amount dTcp is within the temperature control range (the limit is not detected), stabilization control is possible, and the process proceeds to the next step 806 to cause the main controller 4 to perform the stabilization control. In response to this, the main controller 4 changes the temperature of the peaking capacitor Cp by the temperature increase / decrease amount dTcp (step 806).
Next, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered in the same manner as in step 101 of FIG. 7, and the actual spectral purity range E95 after changing the temperature Tcp of the peaking capacitor Cp is measured. (Step 807), it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (Step 808). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 809).
On the other hand, if it is determined in step 808 that the measured spectral purity range E95 is not within the second allowable range (E950 ± dE95), the process proceeds to step 810 again, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 809).
In the flowchart shown in FIG. 24, the case where the temperature of the peaking capacitor Cp is controlled is exemplified. Similarly, the temperature of the second capacitor C2 may be controlled, and the capacitance ratio C2 / Cp changes. Alternatively, the temperature of both capacitors may be controlled. Alternatively, control of the temperature of the peaking capacitor Cp, control of the temperature of the second capacitor C2, and control of the capacitance ratio C2 / Cp (temperature of both capacitors) may be appropriately performed.
Further, instead of controlling the temperature of the capacitor, the capacitance may be controlled by changing the number of connections in the case of a door knob type capacitor. For example, when a 5.6nF peaking capacitor CpF is composed of 28 x 200pF capacitors, the capacitance can be changed by changing the number of connections in the range of 25 to 31, and the temperature can be changed. The same effect can be obtained.
(Example 10: Stabilization control of spectral purity range E95 by changing the pulse waveform of the seed light according to the capacity of the preionization capacitor)
In the present embodiment, as shown in FIG. 3, the oscillation laser apparatus 100 includes a charging circuit including a preionization capacitor Cp ′ arranged in parallel with a pair of discharge electrodes 10a and 10b. It is assumed that preliminary ionization is performed between the pair of electrodes 10a and 10b according to the charging voltage of the capacitor Cp ′. In this embodiment, by changing the capacitance of the preionization capacitor Cp ′, the pulse waveform of the seed light is changed, and the stabilization control of the spectral purity range E95 is performed.
Similar to the laser pulse waveform shown in FIG. 22C, the laser pulse waveform also changes when the value of the capacity of the preionization capacitor Cp ′ is changed. For example, if the capacity of the preionization capacitor Cp ′ is reduced, the current flowing to the preionization capacitor Cp side is increased accordingly, and as a result, the laser pulse waveform becomes longer. Along with this, the intensity of the latter half of the pulse is increased, and the spectral purity range E95 is reduced. Conversely, when the capacity of the preionization capacitor Cp ′ increases, the current flowing to the preionization capacitor Cp side decreases accordingly, and as a result, the laser pulse waveform becomes shorter. Along with this, the intensity of the first half of the pulse is increased, and the spectral purity range E95 is increased. The capacity of the preionization capacitor Cp ′ can be controlled by temperature in the same manner as the peaking capacitor Cp described above. Also, the capacitance can be controlled by changing the number of connected capacitors constituting the preionization capacitor Cp ′.
A specific embodiment is realized by a similar process in which “peaking capacitor Cp” is replaced with “preliminary ionization capacitor Cp ′” in the flowchart of FIG.
In the ninth and tenth embodiments described above, the pulse waveform of the seed light is subjected to pulse stretching. However, the ninth and tenth embodiments may be combined with the pulse waveform control based on the fluorine concentration in the fourth embodiment. In this case, the effect that the allowable synchronization width can be further increased is obtained.
Similarly, the ninth and tenth embodiments in which the pulse waveform of the seed light is subjected to pulse stretching may be combined with the pulse waveform control based on the total gas pressure in the fifth embodiment.
Similarly, the ninth and tenth embodiments in which the pulse waveform of the seed light is pulse stretched may be combined with the pulse waveform control by the charging voltage in the sixth embodiment.
Similarly, the ninth and tenth embodiments, in which the pulse waveform of the seed light is subjected to pulse stretching, may be combined with the pulse waveform control based on the output turnover reflectivity of the seventh embodiment.
Further, as described in the eighth embodiment, control is performed to change the spectral purity width of the seed light by pulse stretching the laser pulse of the seed light (FIG. 22C) and control to change the discharge timing. May be implemented.
(Example 11: Stabilization control of spectral purity range E95 by narrowband performance control (wavefront control by the curvature radius of the grating, wavefront control by the wavefront corrector))
In this embodiment, the spectral purity range E95 of the laser beam output from the amplification laser device 300 is measured by the spectral purity range measuring means, and the measured spectral purity range E95 is the allowable width E950 ± of the target spectral purity range E950. By controlling the band narrowing performance of the oscillation laser device 100 so as to be within dE95, the spectral purity range E95 of the laser light output from the amplification laser device 300 is controlled to be stabilized.
In the eleventh embodiment, the oscillation laser device 100 is provided with wavefront changing means for changing the wavefront of the light generated in the oscillation chamber 10, and the wavefront of the light generated in the oscillation chamber 10 by the wavefront changing means. Is changed to change the band narrowing performance of the oscillation laser device 100, and the spectral purity range E95 of the laser light output from the amplification laser device 300 is controlled to be stabilized.
First, the principle applied to the present embodiment will be described.
By modifying the wavefront of the laser light, the spectral performance changes. By utilizing this, the spectral purity range E95 can be controlled.
That is, in the laser resonator, the wavefront of the laser beam has divergence (expansion) and curvature due to various factors. For example, when a slit is disposed in the laser resonator, the light after passing through the slit becomes a spherical wave due to diffraction by the slit. In addition, the wavefront may be distorted by the aberration of the optical element itself disposed in the laser resonator. For example, in a transmission type optical element such as a prism expander used as a band narrowing element,
(a) Internal refractive index distribution is not completely uniform
(b) The polished surface of the prism is distorted
For this reason, the wavefront of the laser light that has passed through the optical element has a convex or concave curvature. When laser light having a wavefront having such a curvature is incident on a flat-shaped grating, the wavelength selection performance by the grating is degraded. In other words, when the incident wavefront of the laser beam on the grating has a curvature, the laser beam is incident on each groove of the grating at a different angle, so the wavelength selection characteristic of the grating is lowered and the band is narrowed. The spectral line width of the laser beam is increased.
Therefore, by bending the grating itself so as to coincide with the wavefront of the laser light incident on the grating, the reflected wavefront can be corrected to prevent the spectral line width from widening.
FIG. 29 is a sectional view showing a configuration example of a curvature generator that generates a curvature in a grating, that is, a grating bending mechanism.
The grating bending mechanism of FIG. 29 is provided in the narrowband module (LNM) 16.
A side surface of the grating 161 is supported by a grating support portion 162. A spring 163 is disposed at the center of the surface of the grating 161 opposite to the laser light (seed light) incident surface. One end face of the spring 163 is arranged to be pressed against the grating 161. The other end surface of the spring 163 is in contact with the pressing member 164. The spring 163 is arranged to be stretchable along the moving direction of the push member 164. The push member 164 has a tapered inclined surface 164a, and the tip of the adjustment bolt 165 is in contact with the inclined surface 164a. The adjustment bolt 165 is in contact with the push member 164 in such a positional relationship that the push member 164 moves in accordance with the direct movement of the adjustment bolt 165. The adjustment bolt 165 is connected to the rotation shaft of the stepping motor 166, and the adjustment bolt 165 moves linearly according to the rotational drive of the stepping motor 166.
Therefore, when the stepping motor 166 is rotationally driven and the adjustment bolt 165 moves linearly with respect to the inclined surface 164a of the pressing member 164, the pressing member 164 contracts the spring 164 in accordance with the linear movement direction (left in the figure). Direction) or the direction of extending the spring 164 (right direction in the figure). As a result, the central portion of the grating 161 is pushed and pulled, and the radius of curvature of the light incident surface of the grating 161 changes. The spring 163 is provided to make the control interval fine.
FIG. 28 shows a relationship L11 between the radius of curvature of the grating 161 and the spectral purity range E95. A curve L11 shown in FIG. 28 is a curve having a minimum value of the spectral purity width at a certain radius of curvature. Therefore, in the case of actual control, is the control range (SA) on the side where the radius of curvature is smaller than the minimum point? It is desirable to perform control by determining one of the control ranges (SB) on the side where the radius of curvature is larger than the minimum point. Since a region having a large inclination on the curve L11 is easier to control, it is more effective in this example to use the control range (SA) on the side having a larger inclination.
FIG. 30 shows a processing procedure of the eleventh embodiment in which the spectral purity range E95 is stabilized and controlled by adjusting the radius of curvature of the grating 161. FIG. 30 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). Within ± dE95 (S) (Yes at step 103), the subroutine shown in FIG. 30 is entered.
The processing of the subroutine shown in FIG. 30 is executed by E95 and the wavelength controller 6 shown in FIG.
When the subroutine shown in FIG. 30 starts, first, the current bending amount x of the grating 161 in the narrowband module (LNM) 16 is detected (step 910).
Next, whether the spectral purity range E95 is wider or narrower than the target value, that is, whether or not E95> E950 is calculated by E95, the wavelength controller 6 (FIG. 1) (step 901). ).
As a result of this determination, when E95> E950, the spectral purity range is wider than the target value, and therefore corresponds to the difference E95-E950 in the spectral purity range necessary for narrowing the spectral purity range E95. The control amount dx of the bending amount of the grating 161 is calculated, and the value changed by the control amount dx is
x = x + dx
Is calculated (step 902).
When E95> E950 is not satisfied, the spectral purity range is narrower than the target value. Therefore, the bending amount of the grating 161 corresponding to the difference E950-E95 in the spectral purity range necessary to widen the spectral purity range E95. The control amount dx is calculated and the value changed by the control amount dx is
x = x-dx
Is calculated (step 903). The value of the control amount dx may be calculated from, for example, the correlation L11 between the radius of curvature of the grating 161 and the spectral purity range E95 shown in FIG.
Next, it is determined whether or not the value changed by the control amount dx of the bending amount of the grating 161 obtained in this way is within a controllable range.
Therefore, it is determined whether or not the value changed by the control amount dx is out of the control range (limit detection) (step 904). If the limit is detected as a result of this determination, the value of the grating 161 is no longer present. It is determined that the spectral purity range E95 cannot be stabilized by adjusting the curvature radius (bending amount), and the laser oscillation is stopped by switching to another E95 control method or sending an uncontrollable signal to the main controller 4. (Step 905).
If the value changed by the control amount dx is within the control range (the limit is not detected), stabilization control is possible, and the process proceeds to the next step 906, where a drive command signal is sent to the driver 21, for example in FIG. The stepping motor 166 of the grating bending mechanism shown is driven to change the bending amount of the grating 161 by the control amount dx (step 906).
Next, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered as in step 101 of FIG. 7, and the actual spectral purity range E95 after changing the bending amount x of the grating 161 is measured. (Step 907), it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (step 808). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 909).
On the other hand, if the measured spectral purity range E95 is not within the second allowable range (E950 ± dE95) as a result of the determination in step 908, the process proceeds to step 910 again, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 909).
In the flowchart of FIG. 30 described above, it is assumed that the wavefront of light is changed (corrected) by changing the radius of curvature of the grating 161. However, the radius of curvature of another element in the LNM 16, such as a tuning mirror, is described. The wavefront may be corrected in the same manner by changing.
Further, as described in Reference 6 (Japanese Patent Application No. 9-119631), the LNM 16 is provided with wavefront correction means for correcting and emitting the wavefront of the laser light, and the wavefront correction characteristic of the wavefront correction means is set. The spectral purity range E95 may be controlled so as to be changed.
FIG. 54 shows a configuration example of the narrowband module 16 including the wavefront corrector 169.
As shown in FIG. 54, the narrowband module (LNM) 16 is a component including a beam expander 168, a grating 161, a rotary actuator 167 that changes the attitude of the grating 161, and a wavefront corrector 169. It is configured.
The wavefront corrector 169 has a function of correcting and emitting the wavefront of the incident laser light. The laser light incident on the narrowband module 16 is first incident on the beam expander 168 after its wavefront is corrected by the wavefront corrector 169. The beam expander 168 expands the beam width of the laser light. Further, the laser light is incident on the grating 161 and diffracted, whereby only the laser light having a predetermined wavelength component is folded back in the same direction as the incident light. The laser light reflected by the grating 161 is incident on the wavefront corrector 169 after the beam width is reduced by a beam expander (for example, prism) 168. In the wavefront corrector 169, the wavefront is corrected so that the wavefront of the laser light becomes the same plane wave as when the laser beam enters the narrowband module 16. The laser beam whose wavefront is corrected is emitted from the band narrowing module 16 and is incident on the oscillation chamber 10.
The stabilization control of the spectral purity range E95 is performed by changing the wavefront correction characteristic of the wavefront corrector 169 in the narrowband module 16 configured as described above.
FIG. 55 shows a configuration example of the wavefront corrector 169.
FIG. 55A is a configuration example of a wavefront corrector 169 that changes the wavefront by controlling the temperature distribution of the transmissive optical element substrate.
In general, the refractive index of optical materials such as CaF2 varies with temperature. Therefore, a refractive index distribution can be generated by intentionally giving a temperature distribution to the optical element.
Therefore, as shown in FIG. 55A, a heating / cooling device 1691 capable of heating and cooling, such as a thermoelectric element, is installed on each of the four side surfaces of the substrate 1692. The temperature of the substrate 1692 in the vicinity where the heating / cooling device 1691 is installed is detected by the temperature sensor 1691a, and the temperature of each heating / cooling device 1691 is controlled based on the detection value of the temperature sensor 1691a so that the substrate 1692 has a predetermined temperature distribution. By controlling, a desired refractive index profile is given to the substrate 1692 and the wavefront of the laser beam is changed.
FIG. 55B illustrates a wavefront corrector 169 configured by a convex lens 1695, a concave lens 1696, a moving stage 1693 that moves the convex lens 1695 in the optical axis direction, and a pulse motor 1694 that drives the moving stage 1693. ing.
As shown in FIG. 55B, the incident convex wavefront or concave wavefront is converted into a plane wave according to the relative position of the convex lens 1695 and the concave lens 1696 in the optical axis direction. When the distance between the convex lens 1695 and the concave lens 1696 is large, the convex wavefront can be converted into a plane wave. When the distance between the convex lens 1695 and the concave lens 1696 is small, the concave wavefront can be converted into a plane wave. In this way, by driving the pulse motor 1694 to move the moving stage 1693 and changing the distance between the convex lens 1695 and the concave lens 1696, the wavefront of the laser light can be changed.
(Example 12: Stabilization control of spectral purity width E95 by narrowband performance control (wavefront control by deformable mirror))
In the eleventh embodiment described above, the case where the wavefront is controlled according to the curvature radius of the grating or by the wavefront corrector has been described. Next, in order to control the wavefront of the light generated in the oscillation chamber 10 as well. An embodiment using a deformable mirror will be described.
FIG. 31 shows a configuration of a laser wavefront control system using a deformable mirror.
As shown in FIG. 31, a deformable mirror 70 as an end mirror is disposed on the front side of the oscillation chamber 10. In the present embodiment, the shape of the total reflection surface of the deformable mirror 70 is shaped to change (correct) the laser wavefront.
A laser resonator is formed between the grating 161 disposed on the rear side of the oscillation chamber 10 and the deformable mirror 70 on the front side. On the optical path between the oscillation chamber 10 and the deformable mirror 70 on the front side, a beam splitter 71 having an incident angle of 45 degrees is disposed. The beam splitter 71 functions as an output coupling mirror.
That is, part of the light output from the oscillation chamber 10 is reflected by the output coupling mirror 71, extracted as output light (seed light) of the oscillation laser device 100, and input as injection light to the amplification laser device 100. Is done.
On the other hand, the light that has passed through the output coupling mirror 71 is incident on the output coupling mirror 71 again with the reflected wavefront changed by the deformable mirror 70. The light transmitted through the output coupling mirror 71 passes through the oscillation chamber 10 and is amplified. The reflected light from the output coupling mirror 71 enters the wavefront detector 72. The light transmitted through the oscillation chamber 10 and amplified is narrowed by the beam expander 168 and the grating 161 formed by a prism, and the diffracted light is transmitted through the oscillation chamber 10 again and amplified. R Then, again, the light that has been transmitted through the oscillation chamber 10 and amplified is incident on the output coupling mirror 71 and laser-oscillated.
The wavefront of laser light generated in the oscillation chamber 10 is ideally a cylindrical wavefront. The wavefront detector 72 detects the radius of curvature R of the wavefront of the laser light generated in the oscillation chamber 10. The wavefront detector 72 may monitor the oscillation laser light, and may detect the wavefront of light by a guide laser such as another visible light. As a method for detecting the wavefront of a laser beam, a sharing interferometer, a Hartmann-Shack method, or the like is generally known. The wavefront detector 72 can be configured based on these detection principles.
The deformable mirror 70 is mechanically pushed and pulled by a plurality of portions of the reflecting surface by an actuator 73. FIG. 32 is a cross-sectional view taken along the line AA of FIG. 31, showing a line-type deformable mirror 70 and actuators 73 provided at three points of the reflective surface of the deformable mirror 70. For example, a piezo element is used as the actuator 73. By expanding and contracting each actuator 73 (piezo element), each part of the reflecting surface of the deformable mirror 70 is pushed and pulled, and the radius of curvature of the deformable mirror 70 is adjusted. FIG. 32 illustrates the case where the number of actuators 73 is three. However, the number of actuators 73 is not limited to this, and the number of actuators 73 is arbitrary. Accurate wavefront correction is possible. The actuator 73 that pushes and pulls the reflecting surface of the deformable mirror 70 is not limited to a piezo element, and any actuator can be used. For example, in addition to the piezo element, heat due to temperature change can be used. An actuator that pushes and pulls the reflecting surface of the deformable mirror 70 using expansion may be used.
The controller 74 controls the curvature radius r of the deformable mirror 70 via the actuator 73 based on the curvature radius R of the wavefront of the light detected by the wavefront detector 72 and the detected value of the spectral purity range E95. To do.
FIG. 33 shows a relationship L12 between the radius of curvature of the wavefront of the laser beam detected by the wavefront detector 72 and the spectral purity range E95. Similarly to FIG. 28, the curve L12 shown in FIG. 33 is a curve having a minimum value of spectral purity width at a certain radius of curvature. It is desirable to perform control by determining either the control range (SA) or the control range (SB) on the side having a larger radius of curvature than the minimum point. Since a region having a large inclination on the curve L12 is easier to control, it is more effective in this example to use the control range (SA) on the side with a larger inclination.
FIG. 34 shows the processing procedure of the twelfth embodiment for stabilizing the spectral purity range E95 by adjusting the radius of curvature of the laser beam wavefront. FIG. 34 corresponds to the subroutine of “stabilization control by E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). Within ± dE95 (S) (Yes in step 103), the subroutine shown in FIG. 34 is entered.
The processing of the subroutine shown in FIG. 34 is executed by E95, the wavelength controller 6 and the controller 74 shown in FIG.
When the subroutine shown in FIG. 34 is started, first, the current curvature radius R of the wavefront of the laser beam generated in the oscillation chamber 10 is detected by the wavefront detector 72 (step 1010).
Next, whether the spectral purity range E95 is wider or narrower than the target value, that is, whether E95> E950 or not is calculated by E95 and the wavelength controller 6 (FIG. 1) (step 1001). ).
As a result of this determination, when E95> E950, the spectral purity range is wider than the target value, and therefore corresponds to the difference E95-E950 in the spectral purity range necessary for narrowing the spectral purity range E95. The control amount dR of the radius of curvature of the laser light wavefront is calculated, and the value changed by the control amount dR is
R = R + dR
Is calculated (step 1002).
When E95> E950 is not satisfied, the spectral purity range is narrower than the target value. Therefore, the curvature of the laser light wavefront corresponding to the difference E950-E95 in the spectral purity range necessary to widen the spectral purity range E95. The radius control amount dR is calculated, and the value changed by the control amount dR is
R = R−dR
Is calculated (step 1003). The value of the control amount dR may be calculated, for example, from the correlation L12 between the radius of curvature of the laser light wavefront and the spectral purity range E95 shown in FIG.
Next, the amount of change dr of the radius of curvature of the deformable mirror 70 necessary for changing the radius of curvature of the laser light wavefront by the amount of control dR is calculated. Then, it is determined whether or not the radius of curvature of the deformable mirror 70 that has been changed by the control amount dr thus obtained is within a controllable range.
Therefore, it is determined whether or not the value changed by the control amount dr is out of the control range (limit detection) (step 1004), and if the limit is detected as a result of this determination, it is no longer deformable. Adjustment of the radius of curvature of the mirror 70 (the radius of curvature of the laser light wavefront) determines that the spectral purity range E95 cannot be stabilized and switches to another E95 control method or sends an uncontrollable signal to the main controller 4 Then, laser oscillation is stopped (step 1005).
If the value changed by the control amount dr is within the control range (the limit is not detected), stabilization control is possible, and the process proceeds to the next step 1006 to send a drive command signal to the controller 74 of the deformable mirror 70. Through the actuator 73, the curvature radius r of the deformable mirror 70 is changed by the control amount dr, and the curvature radius R of the wavefront of the laser light generated in the oscillation chamber 10 is changed by the control amount dR. (Step 1006).
Next, as in step 101 of FIG. 7, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered, and the radius of curvature r of the deformable mirror 70 (the radius of curvature R of the laser light wavefront) is changed. The actual spectral purity range E95 after the measurement is measured (step 1007), and it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (step 1008). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 1009).
On the other hand, if the measured spectral purity range E95 does not fall within the second allowable range (E950 ± dE95) as a result of the determination in step 1008, the process proceeds to step 1010 again, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 1009).
In this embodiment, an ideal case is assumed where the laser light wavefront is a cylindrical wavefront. However, in reality, the wavefront of the laser light is distorted due to the temperature distribution of the optical element in the laser resonator and the influence of the acoustic wave due to the discharge, and the line-type deformable mirror 70 has a highly accurate wavefront. It may be difficult to perform control. Therefore, a more accurate control of the spectral purity range E95 is performed by mounting a two-dimensional type deformable mirror instead of a line type, and controlling the laser light wavefront with higher accuracy. May be.
In the eleventh embodiment, the bending amount (curvature radius) of the grating 161 is detected, and the bending amount (curvature radius) of the grating 161 is changed based on the detection result. As in the twelfth embodiment, instead of detecting the bending amount (curvature radius) of the grating 161, the wavefront detector detects the wavefront of the laser beam, and based on the detection result, the bending amount (curvature radius) of the grating 161 is detected. It is also possible to implement such as changing.
Example 13 Stabilization Control of Spectral Purity Width E95 by Control of Narrow Bandwidth Performance (Expansion Ratio (Magnification) Change)
In the above-described Embodiments 11 and 12, the case where the band narrowing performance of the oscillation laser device 100 is controlled by changing the wavefront of the light generated in the oscillation chamber 10 has been described. In order to control the band narrowing performance of the laser device 100 for laser, an embodiment in which the magnification ratio of the light generated in the oscillation chamber 10 is changed will be described.
The principle applied to this embodiment will be described.
FIG. 35A shows the positional relationship between the components of the narrowband module (LNM) 16 of the oscillation laser apparatus 100 and the optical axis.
The band narrowing module (LNM) 16 of the oscillation laser device 100 includes a prism 168a (hereinafter abbreviated as “prism a” as appropriate) and a prism 168b (hereinafter referred to as “prism b” as appropriate) in order from a location close to the rear window 10e. The grating 161 is disposed. The beam expander 168 is configured by the prisms 168a and 168b.
As shown by arrows in FIG. 35A, when the postures of the prisms 168a and 168b are changed, the light expansion rate by the beam expander 168 of the oscillation laser device (MO) 100 is changed, and thereby the oscillation chamber The spectral width of the light generated within 10 changes.
FIG. 39A shows a relationship L13 between the magnification M and the spectral width Δλ.
As can be seen from the relationship L13 in FIG. 39A, the spectral width Δλ decreases as the light expansion rate M by the beam expander 168 increases. On the contrary, when the light enlargement ratio M by the beam expander 168 decreases, the spectral width Δλ increases.
Therefore, when the spectral purity width or a spectrum index value instead of this is measured (detected), and the current spectral width Δλ obtained as the measured value (detected value) is wider than the target spectral width Δλ0, the beam When the magnification M of the light by the expander 168 is increased and the current spectral width Δλ is narrower than the target spectral width Δλ0, the spectral width is allowed to the target spectral width by decreasing the expansion ratio. Can be stabilized, and the spectral performance is stabilized.
On the other hand, when the prism groups 168a and 168b are used as the beam expander 168, it is necessary to keep the center wavelength constant. Depending on the adjustment of the attitude of the plurality of prisms 168a, 168b, it is possible to keep the center wavelength the same and change only the magnification. According to the present embodiment, the prism a is used for adjusting the magnification ratio, and the prism b is used mainly for adjusting the wavelength. By adjusting the posture of each, the central wavelength and the magnification ratio can be controlled. (FIG. 35A). Here, in the prior art, as shown in FIG. 35 (b) as a reference example, only the orientation of the prism b is adjusted in order to control only the center wavelength. Is clear.
FIG. 39B shows a relationship L14 between the enlargement factor M and the seed light output (intensity) output from the oscillation laser device 100. FIG.
As can be seen from FIG. 39 (b), the output of the seed light may slightly change when the enlargement ratio M changes with the adjustment of the postures of the plurality of prisms 168a, 168b. However, the change in the output of the seed light is sufficiently compensated by adjusting the voltage applied between the electrodes 10 a and 10 b in the oscillation chamber 10 and the gas pressure in the chamber 10.
FIG. 36 shows the processing procedure of the thirteenth embodiment for stabilizing the spectral purity range E95 by adjusting the magnification (magnification) M. FIG. 36 corresponds to a subroutine of “stabilization control by E95 actuator” (step 104) in the main routine of FIG. In the following processing, the prism a is used for magnification adjustment, and the prism b is used mainly for wavelength adjustment.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). (Within ± dE95 (S)) (Yes at step 103), the subroutine shown in FIG. 36 is entered.
The processing of the subroutine shown in FIG. 36 is executed by the E95 and wavelength controller 6 shown in FIG.
When the subroutine shown in FIG. 36 is started, first, it is determined whether the spectral purity range E95 is wider or narrower than the target value, that is, whether E95> E950 or not, E95, wavelength controller 6 (FIG. 1). ) (Step 1101).
As a result of this determination, when E95> E950, the spectral purity range is wider than the target value, and therefore corresponds to the difference E95-E950 in the spectral purity range necessary for narrowing the spectral purity range E95. Then, the enlargement rate change amount + ΔM per step is calculated, and the postures of the prisms a and b are adjusted accordingly (step 1102). The processing of step 1102 is executed as a subroutine shown in FIG.
First, the enlargement rate change amount + ΔM per step corresponding to the difference E95-E950 in the spectral purity range necessary for narrowing the spectral purity range E95 is calculated, and the value changed by this enlargement rate change amount + ΔM is calculated as follows:
M = M + ΔM
The prism a is rotated to adjust the posture of the prism a so that the calculated enlargement factor M is obtained. The value of the magnification rate change amount + ΔM may be calculated from the correlation L13 between the magnification rate M and the spectral purity range E95 (Δλ) shown in FIG. 39A, for example (step 1201).
However, if the posture of the prism a is changed in the above 1201, the oscillation wavelength is shifted, and at the same time, the posture of the prism b is adjusted to control the center wavelength. That is, the prism b is rotated to adjust the posture of the prism b to return the center wavelength to the original target center wavelength (step 1202).
Next, a part of the seed light output from the oscillation laser device 100 is cut out, and the first monitor module 19 detects the center wavelength λ of the seed light and measures the spectral purity range E95.
Next, the target center wavelength λo required by the exposure apparatus 3 is compared with the detection center wavelength λ, and whether or not the detection center wavelength λ exceeds the allowable width Δ with respect to the target center wavelength λ0, that is,
| Λ− λo |> Δ
Is determined (step 1204).
As a result, when the detection center wavelength λ exceeds the allowable width Δ of the target center wavelength λ0, the orientation of only the prism b is adjusted to change the wavelength so that the center wavelength becomes λo ( Step 1205) is returned to Step 1104 of FIG. 36 (Step 1206). On the other hand, if the detection center wavelength λ1 does not exceed the value obtained by adding the allowable width Δ to the target center wavelength λ0, the process returns to step 1104 in FIG. 36 as it is (step 1206).
On the other hand, as a result of the determination in step 1101 of FIG. 36, when E95> E950 is not satisfied, the spectral purity range is narrower than the target value, so that the spectral purity range necessary for widening the spectral purity range E95 is obtained. An enlargement rate decrease amount −ΔM per step corresponding to the difference E95−E950 is calculated, and the postures of the prisms a and b are adjusted accordingly (step 1103). The processing in step 1103 is executed as a subroutine shown in FIG.
First, the enlargement rate decrease amount -ΔM per step corresponding to the difference E95-E950 in the spectral purity range necessary for widening the spectral purity range E95 is calculated, and the value changed by this enlargement rate change amount -ΔM. But,
M = M−ΔM
The prism a is rotated to adjust the posture of the prism a so that the calculated magnification M is obtained (step 1301). Hereinafter, the processing of steps 1302 to 1306 is the same as the processing of steps 1202 to 1206 described with reference to FIG.
Next, in Step 1104 of FIG. 36, as a result of rotating the prism a and the prism b, whether their attitude angles deviate from the allowable angle width (upper limit value to lower limit value) (for example, if it is the lower limit value, the lower limit value is set). It is judged whether or not). For example, a limit switch for detecting limit angles (upper limit value and lower limit value) of the posture angle of the prism a and the prism b is provided, and it is determined whether or not the limit (limit angle) is detected by the limit switch (step 1104). .
When the attitude angles of the prisms a and b are detected as limit angles (limit detection), it is determined that the spectral purity range E95 can no longer be stably controlled by adjusting the prisms a and b (magnification factor M). The laser oscillation is stopped by switching to the E95 control method or sending an uncontrollable signal to the main controller 4 (step 1105).
If the attitude angles of the prisms a and b do not detect the limit angle (the limit is not detected), stabilization control is possible, and the process proceeds to the next step 1107 as it is.
In step 1107, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered in the same manner as in step 101 of FIG. 7, and the actual spectral purity range E95 after changing the postures of the prisms a and b is determined. It is measured (step 1107), and it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (step 1108). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 1109).
On the other hand, if the measured spectral purity range E95 does not fall within the second allowable range (E950 ± dE95) as a result of the determination in step 1108, the process proceeds again to step 1101, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 1109).
The above control content is a case where the center wavelength and the magnification ratio are controlled by adjusting the postures of the pair of prisms a and b, but the same control can be performed even if the combination of optical components is arbitrary. For example, by adjusting the posture of the prism and the grating, by adjusting the posture of the prism and the rotating mirror, and by adjusting the posture of the rotating mirror and the grating, similarly, the center wavelength and the magnification are increased. The rate can be controlled. Further, an afocal optical system may be arranged so that the magnification can be varied, and the same control may be performed.
When adjusting the attitude of the prism, as shown in FIG. 39 (b), if the attitude angle becomes too large, the output of the seed light is drastically reduced. Possible reasons are that the reflection loss increases when the angle of incidence on the prism deviates from the optimum region of the film, and the vignetting increases when the optical path shifts. In addition, changing the magnification may change the beam quality. Therefore, if necessary, the shaping optical module may compensate for the optical quality of the seed light and the seed light output.
In particular, in the two-stage laser apparatus 2, unlike the one-stage laser apparatus, the beam quality lowered by the oscillation laser apparatus 100 can be adjusted by the amplification laser apparatus 300 to perform compensation.
Example 14 Stabilization Control of Spectral Purity Width E95 by Control of Narrow Bandwidth Performance (Beam Width Change)
Next, an embodiment in which the beam width of the light generated in the oscillation chamber 10 is changed in order to control the narrowing performance of the oscillation laser device 100 will be described.
The principle applied to this embodiment will be described.
FIG. 40 shows the positional relationship between the optical axis of the oscillation laser device 100 and the slit. FIG. 40 is a lateral view of the oscillation chamber 10 (vertical direction) shown in FIG. 2 as viewed from above, and the narrow band module (LNM) 16 of the oscillation laser device 100 includes a rear window 10e. A prism 168a, a prism 168b, and a grating 161 are arranged in this order from a closer location.
A (rear side) slit 80 is disposed between the rear side window 10 e and the prism 168. A (front side) slit 81 is disposed between the front side window 10 f and the front mirror 17. That is, slits 80 and 81 are disposed in the laser resonator. The slits 80 and 81 are configured such that the slit width W in the horizontal direction (vertical direction in the figure) of the oscillation chamber 10 is changed as indicated by arrows. The controller 82 drives and controls the slit width W of the slits 80 and 81 via a driver (not shown).
When the slit width W (lateral slit width) of the slits 80 and 81 in the laser resonator changes, the divergence of the light in the lateral direction changes, and the spectral width changes.
FIG. 41 shows a relationship L15 between the slit width W of the slits 80 and 81 and the spectral width Δλ.
As can be seen from the relationship L15 in FIG. 41, when the slit width W increases within the range of W0 to W1, the spectral width Δλ increases. Conversely, when the slit width W is reduced, the spectral width Δλ is reduced.
Therefore, when the spectral width of spectrum or a spectrum index value instead thereof is measured (detected), and the current spectral width Δλ obtained as the measured value (detected value) is wider than the target spectral width Δλ0, the slit When the width W is reduced and the current spectral width Δλ is narrower than the target spectral width Δλ0, the spectral width is increased with respect to the target spectral width by increasing the slit width W.
It can be stabilized within the allowable range, and the spectral performance is stabilized.
The slit whose slit width W changes may be arranged as the slits 80 and 81 on both the rear side and the front side of the oscillation chamber 10, or may be arranged as the slit 80 only on the rear side. The slit 81 may be disposed only on the front side.
On the other hand, when the slit width W changes, the output of the seed light may change.
However, in the two-stage laser device 2, it is only necessary that the output of the amplification laser device 300 is stable, and slight fluctuations in the output of the oscillation laser device 100 do not cause a problem. Further, the change in the output of the seed light is sufficiently compensated while maintaining the spectral performance by adjusting the voltage applied between the electrodes 10a and 10b in the oscillation chamber 10 and the gas pressure in the chamber 10. The Furthermore, the light beam profile and beam divergence also change due to the change in the slit width W. If these are problematic, the beam profile and beam divergence can be reduced by placing a shaping optical module in the propagation system. It becomes possible to keep it almost constant.
The slits 80 and 81 may change the slit width W from only one side with respect to the optical axis, or may change the slit width W from both sides of the optical axis. However, the controllability is superior when the slit width W is changed from both sides of the optical axis.
In FIG. 40, the slits 80 and 81 whose slit width W changes in the horizontal direction have been described. However, a slit whose width W changes in the vertical direction, that is, the vertical direction in FIG. Similarly, the spectral width can be controlled. However, since the spectral performance change with respect to the change in the slit width W is smaller in the horizontal direction than in the case where the slit width W is changed in the vertical direction, the slits 80 and 81 (FIG. 40) that change in the vertical direction are used. It is desirable to use it in terms of control performance.
As can be seen from the characteristic L15 in FIG. 41, when the slit width W becomes narrower than a certain threshold value W0, the spectral performance deteriorates. The reason for this is considered to be that the theoretical resolution of the grating 161 is lowered because the beam width that can be irradiated to the grating 161 is narrowed. If the slit width W is narrower than the threshold value W0, the output of the seed light is greatly reduced. Therefore, there is little advantage of controlling in the region where the slit width is narrower than the threshold value W0.
Further, in a region where the slit width W is larger than a certain threshold value W1 (> W0), there is almost no change in the spectral performance with respect to the change in the slit width. The reason for this is considered to be that in this region, the slit width is larger than the beam width. Therefore, there is little advantage of controlling in the region where the slit width is wider than the threshold value W1.
Therefore, it is desirable to control the spectral width by changing the slit width W in the region between the threshold values W0 to W1. Further, in this region W0 to W1, the spectral width Δλ monotonously changes (increases) with respect to the change (increase) in the slit width W, which is desirable in terms of control characteristics.
FIG. 42 shows a processing procedure of the fourteenth embodiment in which the spectral purity range E95 is controlled to be stabilized by adjusting the slit width W. FIG. 42 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). (Within ± dE95 (S)) (determination Yes in step 103), the subroutine shown in FIG. 42 is entered.
The processing of the subroutine shown in FIG. 42 is executed by E95, the wavelength controller 6 shown in FIG. 1, and the controller 82 shown in FIG.
When the subroutine shown in FIG. 42 is started, first, it is determined whether the spectral purity range E95 has become wider or narrower than the target value, that is, whether E95> E950 or not, E95, wavelength controller 6 (FIG. 1). ) (Step 1401).
As a result of this determination, when E95> E950, the spectral purity range is wider than the target value, and therefore corresponds to the difference E95-E950 in the spectral purity range necessary for narrowing the spectral purity range E95. The slit width reduction amount -ΔW per step is calculated, and the slit widths W of the slits 80 and 81 are adjusted accordingly. The value of the slit width reduction amount −ΔW may be calculated from, for example, the correlation L15 between the slit width W and the spectral purity width E95 (Δλ) shown in FIG. 41 (step 1402).
On the other hand, if it is determined in step 1401 that E95> E950 is not satisfied, the spectral purity range is narrower than the target value. Therefore, the difference E95− between the spectral purity ranges necessary to widen the spectral purity range E95. A slit width increase amount + ΔW per step corresponding to E950 is calculated, and the slit width W of the slits 80 and 81 is adjusted accordingly (step 1403).
Next, as a result of changing the slit width W of the slits 80 and 81 by the slit width decrease amount -ΔW or the slit width increase amount + ΔW, the slit width is determined from the allowable slit widths W0 to W1 (upper limit value W1, lower limit value W0). It is determined whether it has deviated (for example, if it is the lower limit value W0, it is below the lower limit value W0). For example, limit switches for detecting the limit angles W0 and W1 of the slit widths of the slits 80 and 81 are provided, and it is determined whether or not the limit (limit angle) is detected by the limit switch (step 1404).
If the slit width W of the slits 80 and 81 exceeds the limit angle W1 or falls below the limit angle W0 (limit detection), it is determined that the spectral purity range E95 can no longer be stabilized by adjusting the slit width W. The laser oscillation is stopped by switching to the E95 control method or sending an uncontrollable signal to the main controller 4 (step 1405).
If the slit width W of the slits 80 and 81 does not exceed the limit angle W1 and does not fall below the limit angle W0 (no limit is detected), stabilization control is possible, and the process proceeds to the next step 1407 as it is. .
In step 1407, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered in the same manner as in step 101 of FIG. 7, and the actual spectral purity range E95 after changing the slit width W is measured ( Step 1407), it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (Step 1408). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the shaping optical module compensates for the optical quality of the seed light and the seed light output. (Step 1411), the process returns to the main routine shown in FIG. 7 (Step 1409).
On the other hand, if the measured spectral purity range E95 does not fall within the second allowable range (E950 ± dE95) as a result of the determination in step 1408, the process proceeds to step 1401 again, where the spectral purity range E95 is the second. This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the shaping optical module compensates for the optical quality of the seed light and the seed light output (step 1411). Returning to the routine (step 1409).
Note that the compensation processing in step 1411 may be executed as necessary, and may be omitted.
In particular, in the two-stage laser apparatus 2, unlike the one-stage laser apparatus, it is possible to adjust and compensate for the beam quality deteriorated in the oscillation laser apparatus 100 by the amplification laser apparatus 300.
Example 15 Stabilization Control of Spectral Purity Width E95 by Controlling Propagation Speed of Acoustic Wave Generated by Discharge in Oscillation Chamber
In this embodiment, the spectral purity range E95 of the laser beam output from the amplification laser device 300 is measured by the spectral purity range measuring means, and the measured spectral purity range E95 is the allowable width E950 ± of the target spectral purity range E950. By controlling the propagation speed of the acoustic wave generated by the discharge in the oscillation chamber so as to be within dE95, the spectral purity range E95 of the laser beam output from the amplification laser device 300 is controlled to be stabilized. is there.
In the fifteenth embodiment, an oscillation frequency detecting means for detecting the oscillation frequency of the seed light oscillated by the oscillation laser apparatus 100 and a laser gas temperature changing means for changing the temperature of the laser gas in the oscillation chamber 10 are provided. Based on the relationship between the oscillation frequency of the seed light, the temperature of the laser gas in the oscillation chamber 10 and the spectral purity range E95, the laser gas temperature is changed according to the detected oscillation frequency of the seed light, and the amplification laser device 300 Is to stabilize and control the spectral purity range E95 of the laser beam output from the laser beam.
First, the principle applied to the present embodiment will be described.
As the gas temperature in the oscillation chamber 10 changes, the spectral purity range E95 changes. This is because the acoustic wave generated by the discharge changes the particle density distribution on the laser optical path and changes the laser wavefront. The gas temperature T [K] is between the acoustic wave propagation velocity v and
v ∝ (T) 1/2
The relationship is established. Therefore, when the gas temperature is changed, the propagation speed of the acoustic wave changes, the particle density distribution on the laser optical path changes, the laser wavefront changes, and finally the spectral purity range E95 changes.
In addition, since the change in the oscillation frequency affects the acoustic wave, the relationship between the gas temperature and the spectral purity range E95 described above also changes according to the oscillation frequency.
FIG. 43 is a graph showing how the spectral purity range E95 changes according to the change of the oscillation frequency, and shows each characteristic when the gas temperature is changed to 20 ° C., 40 ° C., and 60 ° C. As can be seen from FIG. 43, when the oscillation laser device 100 oscillates at a certain frequency, if the gas temperature in the oscillation chamber 10 is changed, the spectral purity range E95 changes accordingly, The target value can be set.
FIG. 44 shows the relationship between the gas temperature (° C.) and the spectral purity range E95 for each oscillation frequency. The characteristic L16 shows the relationship between the gas temperature (° C) when the oscillation frequency is 3.6 kHz and the spectral purity range E95, and the characteristic L17 shows the gas temperature (° C) and spectrum when the oscillation frequency is 3.7 kHz. The relationship between the purity range E95 and the characteristic L17 indicate the relationship between the gas temperature (° C.) and the spectral purity range E95 when the oscillation frequency is 4 kHz.
Thus, since the relationship between the gas temperature and the spectral purity range E95 changes depending on the oscillation frequency, when actually controlling, the correlation between the gas temperature and the spectral purity range E95 at each frequency (L16, L17, L18) is stored, the correlation corresponding to the current oscillation frequency is read, and the spectral temperature range E95 is controlled by changing the gas temperature based on the read correlation. Therefore, the actual spectral purity range is controlled as the laser gas temperature.
Next, a configuration example of the temperature sensor T1 that detects the laser gas temperature used in this embodiment will be described.
As a measuring instrument for detecting the gas temperature in the oscillation chamber 10, a thermocouple, a resistance temperature detector, or the like can be used. Further, a fiber thermometer or an infrared thermometer may be used.
45 to 48 show how the temperature sensor T1 is attached to the oscillation chamber 10. FIG.
In FIG. 45, the temperature sensor T 1 is attached to the sheath 91 protruding inward of the oscillation chamber 10. The detection signal of the temperature sensor T1 is input to the utility controller 5 (FIG. 1). The sheath 91 is preferably made of a material that does not react with the laser gas, has a high thermal conductivity, and is made as thin as possible in order to improve the sensitivity of the temperature sensor T1 to the temperature. The sheath 91 is sealed from the outside of the chamber 10 by O-ring or welding, and attached to the chamber partition wall 90 so as to be in contact with the laser gas inside the chamber 10. If the temperature sensor T1 itself is made of a material that does not react with the laser gas, the sheath 91 is unnecessary.
FIG. 46 illustrates the mounting position of the temperature sensor T1.
The temperature sensor T1 is preferably disposed in the chamber partition wall 90 at positions 90A and 90B in the vicinity of the electrodes 10a and 10b. Further, the temperature sensor T1 may be arranged at the positions 90C and 90D in the vicinity of the cross flow fan 10c.
In FIG. 45, the temperature sensor T1 is in contact with the laser gas via the sheath 91. However, as shown in FIG. 47, the temperature sensor T1 is attached to the chamber partition wall 90 so as to contact the laser gas. The temperature may be detected as the temperature of the laser gas. However, it is desirable to cover the periphery of the temperature sensor T1 with a heat insulating material 92 for stable temperature measurement.
Further, as shown in FIG. 48, the response to temperature can be further improved by thinning the chamber partition wall 90 where the temperature sensor T1 is mounted.
The above is the case where the laser gas is detected by the temperature sensor T1, but it is also possible to measure the pressure in the oscillation chamber 10 and calculate the temperature from the pressure.
Next, a configuration example of means for changing the laser gas temperature will be described.
As described with reference to FIG. 2, the oscillation chamber 10 generally includes a heat exchanger 10d for cooling the laser gas.
FIG. 49 shows a configuration in which the gas temperature is changed by adjusting the flow rate of the cooling water flowing through the heat exchanger 10d. A valve 15b is provided on the cooling water supply path 15a for supplying cooling water to the heat exchanger 10.
That is, when the gas temperature is detected by the temperature sensor T1, the temperature detection signal is sent to the utility controller 5. Inside the utility controller 5, the cooling water flow rate required to achieve the target laser gas temperature is calculated using the laser gas temperature as a feedback signal and PID control or the like, and the valve opening corresponding to the cooling water flow rate is calculated. Arithmetic processing is executed. Then, the utility controller 5 sends a valve opening signal to the valve 15 b of the cooling water supply unit 15. Thereby, the valve opening degree of the valve 15b is adjusted, and the cooling water with a necessary flow rate is supplied to the heat exchanger 10d.
FIG. 50 shows a configuration example in which a temperature controller 93 is provided on the cooling water supply path 15a. In the case of FIG. 50 as well, in the same way as in the configuration example of FIG. 49, the utility controller 5 performs an arithmetic process for calculating the operation amount of the temperature controller 93 necessary to achieve the target laser gas temperature, and the operation signal Is sent to the temperature controller 93 of the cooling water supply unit 15. Thereby, the temperature controller 93 is operated and the temperature of the cooling water flowing through the cooling water supply path 15a is adjusted. According to the configuration example of FIG. 50, the response of control for bringing the laser gas temperature close to the target value is improved.
51 and 52 show a configuration example in which the laser gas temperature is changed by the heater attached to the oscillation chamber 10. In this case as well, as in FIGS. 49 and 50, an operation signal is sent from the controller to the heater. By doing so, the temperature of the laser gas is controlled.
In FIG. 51, a heater 94 such as a mantle heater or a ceramic heater is mounted outside the partition wall 90 of the oscillation chamber 10, and the temperature of the laser gas inside the chamber 10 is controlled by adjusting the power supplied to the heater 94. Is done.
Desirably, a heater 95 is mounted inside the oscillation chamber 10 as shown in FIG. In FIG. 52, inside the chamber partition wall 90 and in the vicinity of the electrode 10a,
A heater 95 such as a cartridge heater is mounted. It is desirable to cover the heater 95 with a sheath so as not to react with the laser gas.
Next, a configuration example of means for detecting the oscillation frequency will be described.
Usually, the stepper scanner determines the oscillation frequency of the laser. At that time, specifically, when the exposure apparatus 3 instructs the oscillation laser apparatus 100 to specify the value of the oscillation frequency (for example, 2100 Hz), a laser oscillation trigger signal (for example, a rectangular wave) is sent into the oscillation laser apparatus 100. May oscillate at that timing.
In the former case, since the actual oscillation frequency is instructed from the exposure apparatus 3, the controller inside the oscillation laser apparatus 100 can know the oscillation frequency currently oscillated. In the latter case, since only a trigger signal is sent from the exposure apparatus 3, it is impossible to know the oscillation frequency currently oscillated by the controller inside the oscillation laser apparatus 100.
Therefore, in the latter case, the controller in the oscillation laser device 100 needs to have a function of calculating the current oscillation frequency from the interval of the trigger signal and the count value of the number of trigger signals within a certain time. . However, if the control based on the oscillation frequency before the change is delayed after the oscillation frequency is changed, the spectral purity range E95 may deviate from the allowable range dE95.Therefore, the oscillation frequency to be changed must be changed before the oscillation frequency is changed. It is preferable to be able to obtain.
FIG. 53 shows the processing procedure of the fifteenth embodiment for stabilizing the spectral purity range E95 by adjusting the laser gas temperature. FIG. 53 corresponds to a subroutine of “stabilization control by E95 actuator” (step 104) in the main routine of FIG.
That is, as described with reference to FIG. 7, in the main routine, when the value of the spectral purity range E95 measured by the second monitor module 39 becomes wider than the second allowable range E950 ± dE95 (first allowable range E950). Within ± dE95 (S) (Yes at step 103), the subroutine shown in FIG. 53 is entered.
53 is executed by the E95, wavelength controller 6, utility controller 5, and main controller 4 shown in FIG.
When the subroutine shown in FIG. 53 starts, first, the main controller 4 detects (recognizes) the oscillation frequency f of the seed light oscillated by the oscillation laser device 100. This transmission frequency f is sent to E95 and the wavelength controller 6 (step 1510).
Next, whether the spectral purity range E95 is wider or narrower than the target value, that is, whether E95> E950 or not is calculated by E95 and the wavelength controller 6 (step 1501).
As a result of this determination, if E95> E950, the spectral purity range is wider than the target value, so the gas temperature T required to narrow the spectral purity range E95 is calculated. Specifically, the relationship between the gas temperature T and the spectral purity range E95 for each oscillation frequency shown in FIG. 44 is stored in the database, and in order to obtain the target spectral purity range E950 at the currently detected oscillation frequency f. Is read out. The correspondence between the gas temperature T and the spectral purity range E95 draws various curves as exemplified by the characteristics L16 to L18 in FIG. If it is monotonically increasing or monotonically decreasing within the control range of the gas temperature, there is only one gas temperature T corresponding to the target spectral purity range E950, but if the correlation has a minimum value or a maximum value, There are two or more gas temperatures T corresponding to the spectral purity range E950. In such a case, any gas temperature T may be selected. However, in order to perform the control in a short time, it is desirable to select the gas temperature T closest to the currently detected gas temperature. (Step 1502).
On the other hand, if E95> E950 is not satisfied as a result of the determination in step 1501, the spectral purity range is narrower than the target value. Therefore, the gas temperature T required to widen the spectral purity range E95 is set in the same manner. It is calculated (step 1503).
Next, it is determined whether or not the calculated gas temperature T exceeds a controllable range (limit detection) (step 1504).
If the calculated gas temperature T exceeds the controllable range (limit detection), it is determined that the spectral purity range E95 can no longer be stabilized by adjusting the laser gas temperature, and switching to another E95 control method, The laser oscillation is stopped by sending an uncontrollable signal to the main controller 4 (step 1505).
If the calculated gas temperature T does not exceed the controllable range (no limit is detected), stabilization control is possible, and the process proceeds to the next step 1506 as it is.
In the next step 1506, a control command signal is sent to the utility controller 5 via the main controller 4, and the utility controller 5 controls the laser gas temperature via gas temperature changing means such as the cooling water supply unit 15. That is, control is performed to match the target gas temperature T with the current laser gas temperature detected by the temperature sensor T1 as a feedback amount (step 1506).
In the next step 1507, the “spectral purity range E95 measurement” subroutine (see FIGS. 6 and 10) is entered in the same manner as in step 101 of FIG. 7, and the actual spectrum after changing the gas temperature in the oscillation chamber 10 is obtained. The purity range E95 is measured (step 1507), and it is determined whether or not the measured spectral purity range E95 is within the second allowable range (E950 ± dE95) (step 1508). As a result, if the measured spectral purity range E95 falls within the second allowable range (E950 ± dE95), the process returns to the main routine shown in FIG. 7 (step 1509).
On the other hand, as a result of the determination in step 1508, if the measured spectral purity range E95 is not within the second allowable range (E950 ± dE95), the process proceeds to step 1510 again, where the spectral purity range E95 is the second This subroutine is repeated so as to be within the allowable range. Then, when the spectral purity range E95 falls within the second allowable range, the process returns to the main routine of FIG. 7 (step 1509).
(Examples 16 and 17; control (control means 3)) that changes the spectral purity range of the seed light output from the oscillation laser device 100 (changes the pulse waveform) and also changes the discharge timing. Stabilization control of spectral purity range E95)
In the first embodiment (FIG. 13) described above, stabilization of the spectral purity range E95 of the laser beam is controlled by changing the discharge timing. In the above-described Examples 3 to 7, 9, and 10, the pulse waveform of the seed light is changed, thereby changing the rising timing of the pulse waveform of the seed light, thereby changing the spectral purity width of the seed light. Thus, the stabilization of the spectral purity range E95 of the laser beam was controlled. In the sixteenth and seventeenth embodiments, control combining both of these controls is performed. According to the present embodiment, by performing the above-described control in combination, the control width (synchronization allowable width) of the discharge timing dt is expanded and the controllability is further improved by a synergistic effect.
FIG. 56 is a timing chart used to explain the effect when the control for changing the discharge timing and the control for changing the pulse waveform (changing the pulse rising timing) are used, with the horizontal axis representing time and the vertical axis Is a seed light output (intensity), and shows a pulse waveform of the seed light.
FIG. 56 (a) is a diagram for explaining a case where the spectral purity range E95 is reduced by delaying the discharge timing and further delaying the pulse waveform of the seed light, and FIG. 56 (b) shows the discharge timing. It is a figure explaining the case where the spectral purity range E95 is enlarged by speeding up the pulse waveform of seed light earlier.
FIG. 56A shows an effect obtained by combining the effect described in FIG. 5 and the effect described in FIG. That is, as shown by an arrow 3A, by delaying the discharge timing, the seed light wavelength portion to be amplified in the seed light pulse waveform L3 has a spectral purity width from a portion L31 where the spectral purity width becomes thicker. The process shifts to the thinned portion L32 (the effect described in FIG. 5). Further, as indicated by an arrow 3B, the pulse waveform of the seed light is delayed from L3 to L3 ′, thereby shifting to a portion L32 ′ in which the spectral purity width is further narrowed in the pulse waveform L3 ′ (in FIG. 15). Explained effect).
The same applies to the case where the spectral purity range E95 is increased. As shown by the arrow 13A in FIG. 56 (b), by increasing the discharge timing, the seed light wavelength to be amplified in the pulse waveform L3 of the seed light. The portion shifts from the portion L32 where the spectral purity width is narrowed to the portion L31 where the spectral purity width is thick (the effect described in FIG. 5). Furthermore, as shown by an arrow 13B, the pulse waveform of the seed light is advanced from L3 to L3 ″, so that the pulse waveform L3 ″ shifts to a portion L31 ″ where the spectral purity width becomes thicker (described in FIG. 15). Effect).
As a result, the control width of the discharge timing dt, that is, the synchronization allowable width where the laser output is equal to or higher than the allowable level is further expanded from 3C to 3D, and the controllability is dramatically improved.
FIG. 58 shows the processing procedure of the sixteenth embodiment. FIG. 58 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
FIG. 58 shows processing contents obtained by combining Example 1 (discharge timing control) and Examples 3 to 7, 9, and 10 (pulse waveform control). The description overlapping with the description of the flowchart already described is omitted, and the corresponding part is pointed out and replaced with the description. The pulse waveform control will be described with reference to the fourth embodiment (FIG. 18).
Steps 1601, 1605, 1607, 1608, and 1609 in FIG. 58 are the same as steps 401, 405, 407, 408, and 409 in the first embodiment (FIG. 13).
In step 1602 of FIG. 58, the processing of steps 502 and 506 for delaying the rise of the pulse waveform of the seed light is executed as in the flowchart of the fourth embodiment (FIG. 18). Or the process similar to the step corresponding to the flowchart of other Examples 5-7, 9, 10 is performed.
In Step 1603 of FIG. 58, the processing of Steps 503 and 506 for increasing the rise of the pulse waveform of the seed light is executed as in the flowchart of the fourth embodiment (FIG. 18). Or the process similar to the step corresponding to the flowchart of other Examples 5-7, 9, 10 is performed.
In step 1612 of FIG. 58, the processing of steps 402 and 406 for increasing the interval of the discharge timing is executed as in the flowchart of the first embodiment (FIG. 13).
In step 1613 of FIG. 58, similarly to the flowchart of the first embodiment (FIG. 13), the processing of steps 403 and 406 for decreasing the discharge timing interval is executed.
In step 1604 of FIG. 58, it is determined whether or not a limit is detected with respect to the discharge timing (FIG. 13; step 404), and F2 concentration which is a parameter for changing the pulse waveform (or other examples 5 to 7, It is determined whether or not a limit has been detected for the corresponding parameters in FIGS. 9 and 10 (FIG. 18; step 504 (or the corresponding steps in the flowcharts of other embodiments 5-7, 9, 10)).
In the flowchart of the sixteenth embodiment of FIG. 58, the limit detection determination process (step 1604) is performed after the seed light pulse is changed and the discharge timing is changed, and the control is performed when the limit is detected. Although the impossible process (step 1605) is performed, as shown in FIG. 60, after changing the pulse waveform of the seed light, the limit detection is determined, and then the process of changing the discharge timing is performed. In this way, it is possible to omit the uncontrollable process.
FIG. 60 shows a flowchart of the seventeenth embodiment which is a modification of the flowchart of the sixteenth embodiment shown in FIG.
In FIG. 60, the processing of step 1701 and the processing of steps 1707, 1708, and 1709 are executed in the same procedure as the corresponding step 1601 of FIG. 58 and the processing of steps 1607, 1608, and 1609. , Steps 1707 to 1709 are different from those in FIG.
That is, in FIG. 60, the process of step 1701 is performed, and in step 1702, the process of delaying the rise of the pulse waveform of the seed light is executed as in step 1602 of FIG. 58. Next, the pulse waveform is changed. It is determined whether or not a limit has been detected for the F2 concentration (or the corresponding parameter in the other embodiments 5 to 7, 9, and 10), which is a parameter to be executed (step 1704). As a result, if the limit is detected, control is not disabled, but a process for increasing the discharge timing interval is executed (step 1712) as in step 1612 of FIG. To be migrated. If no limit is detected, the process proceeds to step 1707 as it is without controlling the discharge timing.
Similarly, after the processing in step 1701, in step 1703, processing for increasing the rise of the pulse waveform of the seed light is executed in the same manner as in step 1603 in FIG. 58. Next, parameters for changing the pulse waveform are executed. It is determined whether or not a limit has been detected with respect to the F2 concentration (or the corresponding parameters in the other examples 5 to 7, 9, and 10) (step 1705). As a result, if the limit is detected, control is not disabled, but a process of reducing the discharge timing interval is executed (step 1713) as in step 1613 of FIG. To be migrated. If no limit is detected, the process proceeds to step 1707 as it is without controlling the discharge timing.
In FIG. 60, after changing the pulse waveform (steps 1702 and 1703), it is determined whether or not a limit is detected (steps 1704 and 1705), and the discharge timing is changed according to the result. (Steps 1712 and 1713), on the contrary, after changing the discharge timing, it is determined whether or not a limit has been detected, and control is performed according to the procedure of changing the pulse waveform according to the result. May be.
According to the embodiments 16 and 17 described above, by controlling the control for changing the discharge timing and the control for changing the pulse waveform (changing the pulse rising timing), the control of the discharge timing dt is achieved by a synergistic effect. The width (synchronization allowable width) is expanded, and the controllability is further improved. Furthermore, the control according to the second embodiment, that is, the embodiment in which the seed light is pulse stretched may be combined. As described with reference to FIG. 14, when the seed light is pulse stretched and the pulse width of the seed light is increased, the allowable synchronization width is further increased, and the controllability can be further improved.
In the flowcharts of the fourth embodiment (examples 5 to 7, 10, and 11) in which the pulse waveform is controlled by changing the parameters such as the fluorine concentration described above, the laser oscillation is stopped when the limit is detected. In the fourth embodiment, the control is performed (step 505 in FIG. 18 in the case of the fourth embodiment). As in FIG. 60, when the limit is detected, the control (step for changing the discharge timing) is performed. 1712, 1713).
(Examples 18 and 19; changing the spectral purity range of the seed light output from the oscillation laser device 100 (changing the band narrowing performance or changing the acoustic wave propagation speed), and changing the discharge timing) (Stabilization control of spectral purity range E95 by performing control (control means 3)))
In the first embodiment (FIG. 13) described above, stabilization of the spectral purity range E95 of the laser beam is controlled by changing the discharge timing. In the above-described Examples 11 to 14 and 15, the narrowing performance of the oscillation laser device 100 is changed, or the propagation speed of the acoustic wave generated by the discharge in the oscillation chamber 10 is changed. The stabilization of the spectral purity range E95 of the laser beam was controlled by changing the spectral purity range of the seed beam. In the eighteenth and nineteenth embodiments, control in which both these controls are combined is performed. According to the present embodiment, by performing the above-described control in combination, the control range of the spectral purity range E95 is expanded by the synergistic effect, and the controllability is further improved.
FIG. 57 shows the effect of combining the control for changing the discharge timing and the control for changing the spectral purity range of the seed light (control for changing the band narrowing performance or control for changing the propagation speed of the acoustic wave). It is a figure used for explanation, and the horizontal axis indicates the discharge timing dt and the vertical axis indicates the spectral purity range E95 as in FIG. 4 described above.
A characteristic L1 in FIG. 57 corresponds to the characteristic L1 shown in FIG.
In order to increase the spectral purity range E95 so as to coincide with the target spectral purity range E950, first, control is performed to increase the spectral purity range of the seed light. As a result, the characteristic L1 changes to a characteristic L1A in which the spectral purity range becomes larger. Further, control is performed to decrease the discharge timing dt in order to increase the spectral purity range E95. As a result, the discharge timing changes in the direction of decreasing on the characteristic L1A.
In order to reduce the spectral purity range E95 so as to coincide with the target spectral purity range E950, first, control is performed to reduce the spectral purity range of the seed light. As a result, the characteristic L1 changes to a characteristic L1B having a smaller spectral purity range. Furthermore, control is performed to increase the discharge timing dt in order to reduce the spectral purity range E95. As a result, the discharge timing changes in the direction of increasing on the characteristic L1B.
Therefore, the control width 1A of the spectral purity width E95 when only the control for changing the spectral purity width of the seed light is combined with the control for changing the spectral purity width of the seed light and the control for changing the discharge timing dt. Comparing the control width 1B in this case, it can be read that the control width (vertical axis) of the spectral purity range E95 is expanded if the two controls are combined within the same synchronization allowable width (horizontal axis).
In this way, by performing control combining both controls, the range in which the spectral purity range E95 can be swung within the same allowable synchronization range is further expanded, and the controllability is greatly improved.
FIG. 59 shows the processing procedure of the eighteenth embodiment. FIG. 58 corresponds to the subroutine of “stabilization control by the E95 actuator” (step 104) in the main routine of FIG.
FIG. 59 shows the processing contents obtained by combining Example 1 (control of discharge timing) and Examples 11 to 14 and 15 (control of the spectral purity range of passceed light). The description overlapping with the description of the flowchart already described is omitted, and the corresponding part is pointed out and replaced with the description. The control of the spectral purity range of the seed light will be described with reference to Example 11 (FIG. 30).
Steps 1801, 1805, 1807, 1808, and 1809 in FIG. 59 are the same as steps 401, 405, 407, 408, and 409 in the first embodiment (FIG. 13).
In Step 1802 of FIG. 59, the processing of Steps 902 and 906 for reducing the spectral purity width of the seed light is executed, as in the flowchart of Example 11 (FIG. 30). Or the process similar to the step corresponding to the flowchart of other Examples 12-14, 15 is performed.
In Step 1803 of FIG. 59, the processing of Steps 903 and 906 for increasing the spectral purity range of the seed light is executed, as in the flowchart of Example 11 (FIG. 30). Or the process similar to the step corresponding to the flowchart of other Examples 12-14, 15 is performed.
In step 1812 of FIG. 59, the processes of steps 402 and 406 for increasing the interval of the discharge timing are executed as in the flowchart of the first embodiment (FIG. 13).
In step 1813 of FIG. 59, the processing of steps 403 and 406 for reducing the interval of the discharge timing is executed as in the flowchart of the first embodiment (FIG. 13).
In step 1804 of FIG. 59, it is determined whether or not a limit is detected with respect to the discharge timing (FIG. 13; step 404), and the curvature radius of the grating (or other parameters that change the spectral purity range of the seed light) It is determined whether or not a limit has been detected with respect to the corresponding parameters in Examples 12-14, 15 (FIG. 30; step 904 (or corresponding steps in the flowcharts of other Examples 12-14, 15)).
In the flowchart of Example 18 in FIG. 59, when the limit is detected by performing the limit detection determination process (step 1804) after changing the spectral purity range of the seed light and changing the discharge timing. Although the uncontrollable process (step 1805) is performed, as shown in FIG. 61, after changing the spectral purity range of the seed light, the limit detection is determined, and then the discharge timing is changed. It is also possible to omit the uncontrollable process by performing the process.
FIG. 61 shows a flowchart of the nineteenth embodiment, which is a modification of the flowchart of the eighteenth embodiment shown in FIG.
In FIG. 61, the processing of step 1901 and the processing of steps 1907, 1908, and 1909 are executed in the same procedure as the corresponding processing of step 1801 and steps 1807, 1808, and 1809 of FIG. The processing procedure performed between steps 1907 to 1909 is different from that in FIG.
That is, in FIG. 61, the process of step 1901 is performed, and the process of reducing the spectral purity range of the seed light is executed in step 1902 as in step 1802 of FIG. 59. Next, the spectral purity of the seed light is performed. It is determined whether or not a limit has been detected with respect to the curvature radius of the grating (or the corresponding parameter in the other examples 12 to 14 and 15), which is a parameter for changing the width (step 1904). As a result, if a limit is detected, control is not disabled, but a process for increasing the discharge timing interval is executed (step 1912), as in step 1812 in FIG. To be migrated. If no limit is detected, the process proceeds to step 1907 as it is without controlling the discharge timing.
Similarly, after the processing of step 1901, in step 1903, processing for increasing the spectral purity width of the seed light is executed in the same manner as in step 1803 of FIG. 59. Next, the spectral purity width of the seed light is increased. It is determined whether or not a limit has been detected for the curvature radius of the grating, which is a parameter to be changed (or a corresponding parameter in the other examples 12 to 14, 15) (step 1905). As a result, if a limit is detected, control is not disabled, but a process for reducing the discharge timing interval is executed (step 1913), as in step 1813 in FIG. To be migrated. If no limit is detected, the process proceeds to step 1907 as it is without controlling the discharge timing.
In FIG. 61, after changing the spectral purity range of the seed light (steps 1902 and 1903), it is determined whether or not a limit is detected (steps 1904 and 1905), and the discharge timing is changed according to the result. (Steps 1912 and 1913), on the contrary, after changing the discharge timing, it is determined whether or not the limit is detected, and the spectral purity range of the seed light is changed according to the result. You may make it control by the procedure to make.
According to the above-described Examples 18 and 19, the control for changing the discharge timing and the control for changing the spectral purity range of the seed light (the control for changing the band narrowing performance or the control for changing the propagation speed of the acoustic wave) As a result of the combination, the range in which the spectral purity range E95 can be swung within the same synchronization permissible range is further expanded by the synergistic effect, and the controllability is dramatically improved.
In the flowcharts of Embodiment 11 and the like (Examples 12 to 14 and 15) in which the spectral purity range of the seed light is controlled by changing parameters such as the radius of curvature of the grating described above, the laser beam is detected when a limit is detected. Although processing that cannot be controlled is performed such as stopping oscillation (step 905 in FIG. 30 in the case of Example 11), the discharge timing is changed when a limit is detected as in FIG. You may make it switch to the control (step 1912, 1913) to perform.
(Industrial applicability)
Although the present invention is premised on the two-stage laser apparatus 2, various controls of the present embodiment can be applied to a one-stage laser apparatus.

4 ... Main controller 10 ... Oscillation chamber 30 ... Amplification chamber 100 ... Oscillation laser device 300 ... Amplification laser device

Claims (21)

An oscillation laser device that outputs seed light that has been narrowed by passing the laser light through the wavelength dispersion element several times, and an amplified laser that amplifies the seed light by discharging laser gas in the amplification chamber. In a narrow-band laser device provided with an amplification laser device or an amplification device that outputs light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
Discharge timing from the start of discharge by the oscillation laser device to the start of discharge by the amplification laser device or the amplification device so that the measured spectrum index value falls within the allowable range of the target spectrum index value A narrow-band laser device comprising: control means for controlling An oscillation laser device that outputs seed light that has been narrowed by passing the laser light through the wavelength dispersion element several times, and an amplified laser that amplifies the seed light by discharging laser gas in the amplification chamber. In a narrow-band laser device provided with an amplification laser device or an amplification device that outputs light,
Spectrum index value measuring means for measuring a spectrum index value of laser light output from the amplification laser device or the amplification device;
A narrow-band laser apparatus comprising: a control unit that controls a spectrum index value of the seed light so that a measured spectrum index value falls within an allowable range of a target spectrum index value. An oscillation laser device that outputs seed light that has been narrowed by passing the laser light through the wavelength dispersion element several times, and an amplified laser that amplifies the seed light by discharging laser gas in the amplification chamber. In a narrow-band laser device provided with an amplification laser device or an amplification device that outputs light,
Spectrum index value measuring means for measuring a spectrum index value of laser light output from the amplification laser device or the amplification device;
Discharge timing from the start of discharge by the oscillation laser device to the start of discharge by the amplification laser device or the amplification device so that the measured spectrum index value falls within the allowable range of the target spectrum index value And a control means for controlling the spectral index value of the seed light. Control by the control means is performed within a range of allowable synchronization width in which the energy of the laser beam output from the amplification laser device is equal to or higher than an allowable level among the allowable width of the target spectrum index value. The narrow-band laser device according to claim 1 or 3. Pulse stretching means for extending the laser pulse waveform of the seed light is further provided;
By extending the pulse waveform of the seed light by the pulse stretching means, an allowable synchronization width in which the energy of the laser light output from the amplification laser device or the amplification device is equal to or higher than an allowable level is expanded. The narrow-band laser device according to claim 1, 2 or 3. 3. The control means for controlling the spectral index value of the seed light by changing the time from the start of discharge by the oscillation laser device to the rise of the laser pulse. 3. The narrow-band laser device according to 3. 4. The narrowed structure according to claim 2, wherein the control means changes the pulse waveform of the seed light by changing the molar concentration or partial pressure of fluorine molecules F2 in the oscillation chamber. Banded laser device. 4. The narrow-band laser device according to claim 2, wherein the control means changes a pulse waveform of the seed light by changing a total gas pressure in the oscillation chamber. The oscillation laser device performs a main discharge by applying a voltage according to a charging voltage between a pair of electrodes,
4. The narrow-band laser device according to claim 2, wherein the control means changes a pulse waveform of the seed light by changing a charging voltage. The oscillation laser device is:
A charging circuit including a peaking capacitor electrically parallel to the pair of discharge electrodes and a second capacitor electrically parallel to the preceding stage of the peaking capacitor, and storing in the second capacitor; The generated charge is transferred to the peaking capacitor, and a voltage corresponding to the charging voltage of the peaking capacitor is applied to the pair of electrodes, and discharging is performed.
The pulse waveform of the seed light is changed by changing the capacitance of the peaking capacitor or / and the second capacitor or / and the capacitance ratio of the second capacitor to the peaking capacitor. The narrow-band laser device according to claim 2 or 3. The oscillation laser device is:
A charging circuit including a preionization capacitor arranged in parallel with a pair of discharge electrodes is provided, and preionization is performed between the pair of electrodes according to a charging voltage of the preionization capacitor,
4. The narrow-band laser device according to claim 2, wherein a pulse waveform of the seed light is changed by changing a capacity of the preionization capacitor. The oscillation laser device includes:
An output coupler for returning the light generated in the oscillation chamber to the oscillation chamber with a predetermined reflectance;
4. The narrow-band laser device according to claim 2, wherein a pulse waveform of the seed light is changed by changing a reflectance of the output coupler. The control means changes the time from the start of discharge in the oscillation laser device to the rise of the laser pulse, and controls the spectral index value of the seed light,
When the rise time of the pulse waveform of the seed light changes, from the start of discharge in the oscillation laser device to the start of discharge in the amplification laser device according to the change in the rise time of the pulse waveform 4. The narrow-band laser apparatus according to claim 2, wherein control is performed to change the discharge timing to match the discharge timing with a desired synchronization timing. An oscillation laser device that outputs seed light that has been narrowed by passing the laser light through the wavelength dispersion element several times, and an amplified laser that amplifies the seed light by discharging laser gas in the amplification chamber. In a narrow-band laser device provided with an amplification laser device or an amplification device that outputs light,
Spectrum index value measuring means for measuring a spectrum index value of laser light output from the amplification laser device or the amplification device;
Narrowband characterized by comprising control means for controlling the narrowband performance of the oscillation laser device or amplification device so that the measured spectral index value falls within the allowable range of the target spectral index value Laser equipment. An oscillation laser device that outputs seed light that has been narrowed by passing the laser light through the wavelength dispersion element several times, and an amplified laser that amplifies the seed light by discharging laser gas in the amplification chamber. In a narrow-band laser device provided with an amplification laser device or an amplification device that outputs light,
Spectrum index value measuring means for measuring a spectrum index value of laser light output from the amplification laser device or the amplification device;
Discharge timing from the start of discharge by the oscillation laser device to the start of discharge by the amplification laser device or the amplification device so that the measured spectrum index value falls within the allowable range of the target spectrum index value And a control means for controlling the narrowing performance of the oscillation laser device. The oscillation laser device includes:
Wavefront changing means for changing the wavefront of the light generated in the oscillation chamber;
16. The narrow-band performance of the oscillation laser device is changed by changing a wavefront of light generated in the oscillation chamber by the wavefront changing means. Narrow band laser device. The oscillation laser device includes:
An enlargement ratio changing means for changing an enlargement ratio of the light generated in the oscillation chamber;
16. The narrowing performance of the oscillation laser device is changed by changing an enlargement ratio of light generated in the oscillation chamber by the enlargement ratio changing means. The narrow-band laser device described. The oscillation laser device includes:
Beam width changing means for changing the beam width of the light generated in the oscillation chamber is provided,
16. The narrowing performance of the oscillation laser device is changed by changing a beam width of light generated in the oscillation chamber by the beam width changing means. The narrow-band laser device described. An oscillation laser device that outputs seed light that has been narrowed by passing the laser light through the wavelength dispersion element several times, and an amplified laser that amplifies the seed light by discharging laser gas in the amplification chamber. In a narrow-band laser device provided with an amplification laser device or an amplification device that outputs light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
Control means for controlling the propagation speed of the acoustic wave generated by the discharge in the oscillation chamber so that the measured spectrum index value falls within the allowable range of the target spectrum index value. Narrow band laser device. An oscillation laser device that outputs seed light that has been narrowed by passing the laser light through the wavelength dispersion element several times, and an amplified laser that amplifies the seed light by discharging laser gas in the amplification chamber. In a narrow-band laser device provided with an amplification laser device or an amplification device that outputs light,
Spectrum index value measuring means for measuring a spectrum index value of the laser beam output from the amplification laser device;
The discharge timing from the start of the discharge by the oscillation laser device to the start of the discharge by the amplification laser device is controlled so that the measured spectrum index value is within an allowable range of the target spectrum index value. And a control means for controlling the propagation speed of the acoustic wave generated by the discharge in the oscillation chamber. Oscillation frequency detection means for detecting the oscillation frequency of the seed light oscillated by the oscillation laser device;
Laser gas temperature changing means for changing the temperature of the laser gas in the oscillation chamber,
Based on the relationship between the oscillation frequency of the seed light, the temperature of the laser gas in the oscillation chamber and the spectral index value, the laser gas temperature is changed according to the detected oscillation frequency of the seed light, and the measured spectral index value 21. The narrow-band laser device according to claim 19 or 20, wherein control is performed so as to fall within an allowable range of the target spectrum index value.

Images