JP2010073863A - Method for adjusting sensitivity of spectroscope and spectroscope - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve nonuniformity of sensitivity by adjusting the sensitivity of an optical sensor array wherein sensitivity has became ununiform. <P>SOLUTION: An amplification stage laser 11 of two-stage laser equipment configured by an oscillation stage laser 10 and the amplification stage laser 11 is singly oscillated to generate light having a spectral width wider than a free spectral width (FSR) of an etalon 1b. This light is made incident to a spectrum measuring instrument 1, and then incident to a sensor surface of a CCD sensor 1d through the etalon 1b, etc. A compensation factor-constant calculating portion 2b of a CCD controller 2 calculates a compensation factor-compensation constant for compensating degradation for each channel of the CCD sensor 1d, and based on this constant, gain of a variable gain means 2c is controlled, and sensitivity is compensated for each channel of the CCD sensor 1d to uniformize the sensitivity. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a method for adjusting the sensitivity of a spectrometer used for spectrum measurement / measurement of a narrow-band oscillation excimer laser apparatus and the like, and the spectrometer.

Spectrometers that measure the wavelength distribution of light and the like are used in various fields. For example, they are used for spectrum measurement of output laser light of a narrow-band excimer laser device that is a light source for semiconductor lithography.
The wavelength measurement technique of the excimer laser that oscillates in a narrow band is described in Patent Document 1, for example. As described in the same document, in the excimer laser device for exposure, the center wavelength is measured by a monitor module using an etalon or the like. In this monitor module, a part of the laser light is branched by a beam splitter, and the wavelength and the like are measured from a fringe (interference fringe) waveform obtained by passing the branched light through an etalon.

FIG. 6 shows a schematic diagram of the optical system of the spectrometer.
This figure shows a case where the spectrum of the output laser light of a twin chamber (two-stage system) laser device provided with, for example, the oscillation stage laser and the amplification stage laser shown in FIG. 6A is measured.
In order to respond to the demand for higher output, the twin chamber (two-stage type) laser apparatus has an oscillation stage laser 10 that produces laser light with high light quality (spectral performance, etc.) and small output, and an amplification stage that amplifies the laser light. It is comprised by the laser 11 (refer patent document 2, 3 grade | etc., About the laser apparatus of a twin chamber).
In the figure, a monitor module 12 is provided in a twin chamber type laser device composed of an oscillation stage laser 10 and an amplification stage laser 11, and a part of the output laser beam is branched by a beam splitter 13, and the branched light is monitored. This is taken into the spectrum measuring instrument 1 provided in the module 12.

The laser light (measurement light) taken into the spectrum measuring instrument 1 is condensed by the incident lens 1a and incident on the etalon 1b as shown in FIG. By concentrating the light emitted from the etalon 1b by the imaging system lens 1c, concentric circles (fringe patterns) having a radius corresponding to the wavelength of the laser light appear. By placing a line sensor such as a CCD sensor at the position of the fringe pattern, a light intensity distribution as shown in FIG. 6C can be obtained. For example, the oscillation center wavelength of the laser beam can be calculated from the radius indicating the peak of the light intensity.
JP-A-6-188502 JP 2001-24265 A JP 2004-342964 A

In a narrow-band oscillation excimer laser device, a highly accurate spectrum measuring instrument (spectrometer) is required to stably measure the spectral line width. In addition, durability that does not deteriorate measurement accuracy due to long-term excimer laser measurement is required.
However, due to long-term ultraviolet measurement, the deterioration of the optical sensor array such as a CCD sensor has occurred, and the non-uniformity of sensor sensitivity has a problem of affecting the measurement accuracy.
FIG. 7 is a diagram for explaining deterioration of the optical sensor that occurs when the fringe pattern is measured by an optical sensor array such as a CCD sensor.
The fringe pattern has strong and weak light intensity depending on the position. When this is measured by the optical sensor array, the intensity of the incident light is different for each channel. For this reason, as shown in FIG. 5A, the sensitivity of the sensor channel in the portion where strong light is incident is more rapidly deteriorated than the other portions.
For this reason, as shown in FIG. 5B, the deterioration progresses rapidly at the fringe peak position where the light intensity is strong, and the sensitivity of the optical sensor becomes non-uniform.

When the line width is measured at the sensor portion where the sensitivity is lowered, the shape changes with respect to the original fringe waveform, and thus an incorrect line width is measured.
When measured with a degraded optical sensor, for example, as shown in FIGS. 8A and 8B, the peak value is measured lower than the original fringe waveform, or the line width is measured smaller.
Since the position of the fringe changes depending on the wavelength, the shape of the fringe changes variously depending on the sensitivity of the sensor when the wavelength is changed. Along with this change, the measured value may also show a large value with respect to the original line width, or conversely, a small value, so the wavelength dependency of the line width measurement value becomes large due to sensor deterioration. was there.
The present invention has been made in view of the above circumstances, and an object of the present invention is to adjust the sensitivity of a photosensor array having non-uniform sensitivity and to eliminate the sensitivity non-uniformity. An adjustment method and a spectrometer having such a function are provided.

In the present invention, the above problem is solved as follows.
(1) In a sensitivity adjustment method for a spectroscope that detects light that has passed through a spectroscopic means with a photosensor array and measures a spectrum based on the output of the photosensor array, the free spectral width (FSR: free spectral range) of the spectroscopic means Light having a wider spectral width is incident on the spectroscopic means, the light passing through the spectroscopic means is detected by the photosensor array, and the sensitivity to incident light in each channel of the photosensor array is determined based on the detection result. The correction value of each channel is obtained based on this sensitivity. Based on this correction value, the gain of each channel of the optical sensor array is adjusted.
For example, as shown in FIG. 2, in the optical sensor array 1d, when the sensitivity of some channels deteriorates to 0.7, 0.9, and 0.6, this deterioration can be corrected for each channel. Gain is obtained, and this gain is applied to each channel to correct the sensitivity and make the sensitivity uniform.
(2) In a spectroscope including a spectroscopic unit, a photosensor array, and a light source that emits light having a spectrum width wider than a free spectral width (FSR) of the spectroscopic unit, gain of each channel of the photosensor array An adjusting means for adjusting is provided. The adjusting means causes light from the light source to enter the optical sensor array, obtains sensitivity to incident light in each channel of the optical sensor array, obtains a correction value of gain of each channel based on the sensitivity, and sets the correction value to the correction value. Based on this, the gain of each channel of the photosensor array is adjusted.
(3) In the above (2), the adjustment means calculates the optical sensitivity and offset of each channel with respect to any one of the average output, output integration, measurement time, or number of pulses of each channel of the optical sensor array. Thus, the sensitivity and offset of each channel are obtained, and the correction value of each channel is obtained based on the optical sensitivity and offset of each channel and the average value, maximum value, or minimum value thereof.
(4) In the above (3), the spectroscope is attached to the pulse laser device, and measures the spectral line width of the laser light emitted from the pulse laser device.

  In the present invention, the light that has passed through the spectroscopic means is detected by the optical sensor array, the correction value of each channel is obtained based on the sensitivity to the incident light in each channel of the optical sensor array, and based on this correction value, the optical sensor array Since the gain of each channel is adjusted, non-uniformity of the sensitivity of the optical sensor can be eliminated. For this reason, it is possible to stably measure the spectral line width in long-term excimer laser measurement.

FIG. 1 is a diagram showing a configuration of an embodiment of the present invention. FIG. 1 shows a configuration when measuring a spectrum of an output laser beam of a twin chamber type (two-stage type) laser apparatus, and FIG. Indicates the overall configuration of the system, and FIG. 5B shows the gain adjustment operation of the CCD sensor.
As shown in FIG. 1 (a), the laser device is composed of an oscillation stage laser 10 that has a narrow-band means and generates low-power laser light with high optical quality, and an amplification stage laser 11 that amplifies the laser light. And power supplies 10a and 11a for generating high voltage pulses, respectively.
A monitor module 12 is provided on the output side of the laser device, a part of the output laser light of the amplification stage laser 11 is branched by the beam splitter 13, and the branched light is taken into the spectrum measuring instrument 1 provided in the monitor module 12. .
As shown in FIG. 1B, the spectrum measuring instrument 1 is provided with an incident system lens 1a, an etalon (spectral means) 1b, an imaging system lens 1c, and a CCD sensor (light sensor array) 1d. As described above, the laser beam branched by the splitter 13 is collected by the incident lens 1a and enters the etalon 1b, which is a spectroscopic means. The light emitted from the etalon 1b is condensed by the imaging system lens 1c, and a fringe image (interference fringe) is formed on the CCD sensor 1d.

The fringe waveform measured by the spectrum measuring instrument 1 is sent to the CCD controller 2. The CCD controller 2 controls the CCD sensor 1d and the like of the spectrum measuring instrument 1 and captures the fringe waveform measured by the CCD sensor 1d. The spectrum calculation unit 2a outputs E95 of the output laser beam (the oscillation spectrum centering on the oscillation wavelength). Calculation is performed such as a wavelength width in which 95% of the included light energy exists.
During normal operation, the laser controller 14 synchronously controls the oscillation stage laser 10 and the amplification stage laser 11 to output a predetermined laser beam.
Further, when adjusting the sensitivity of the CCD sensor 1d in which the sensitivity of each channel is non-uniform and eliminating the non-uniformity of sensitivity, the power source 10a of the oscillation stage laser 10 is turned off and the power source 11a of the amplification stage laser 11 is turned off. The amplifier stage laser 11 is oscillated alone to generate light having a wide spectral width.
This light is incident on the sensor surface of the CCD sensor 1d, and the correction coefficient / constant calculation unit 2b of the CCD controller 2 calculates a correction coefficient / constant for correcting deterioration for each channel of the CCD sensor 1d, and is variable based on this. The gain of the gain means 2c is controlled, the sensitivity is corrected for each channel of the CCD sensor 1d, and the sensitivity is made uniform.

That is, when correcting the sensitivity of the CCD sensor 1d, the laser controller 14 oscillates the amplification stage laser 11 as described above, and the spectrum width wider than the free spectrum width (FSR) of the etalon 1b, for example, from the amplification stage laser 11. The light Lm (m = 1 to n) is emitted.
The light Lm (m = 1 to n) is captured by the beam splitter 13 and enters the CCD sensor 1d via the etalon 1b and the like as shown in FIG. 1B.
The CCD controller 2 scans the CCD sensor 1d and takes in the output Pmi of each channel.
The correction coefficient / constant calculation unit 2b of the CCD controller 2 calculates a correction coefficient Ai and a correction constant Bi. Then, the gain of the variable gain means 2c provided on the output side of the CCD sensor 1d is adjusted so that the sensitivity of each channel of the CCD sensor is made uniform.
The correction coefficient / constant calculation unit 2b and the variable gain means 2c constitute an adjusting means for adjusting the gain of each channel of the CCD sensor. The variable gain means 2c may use hardware such as an amplifier, or may be realized by software in the CCD controller 2.

Calculation of the correction coefficient Ai and the correction constant Bi is performed as follows.
As described above, the single-stage oscillation of the amplification stage laser is performed, and the light L1 to Ln (hereinafter referred to as Lm, where m = 1) having a spectrum width wider than the free spectrum width (FSR) of the etalon 1d and different light amounts. ˜n) enter the spectrum measuring instrument 1. This light Lm is applied to the sensor surface of the CCD sensor 1d.
The sensor channel output at that time is Pmi (i = 1 to 1024 ch), and the average output is Vm. Here, the number of channels i of the CCD sensor 1d is 1024.
The correction coefficient / constant calculation unit 2b of the CCD controller 2 calculates the photosensitivity a _i and the offset b _i shown in the following equation (1) for all channels i.
P _mi = V _m × a _i + b _i (Relation between light input and sensor output) (1)
Here, a _i and b _i are coefficients obtained by a general equation of a regression line expressed by the following equations (2) and (3). V _m , V, and P _m are expressed by the following equations (4) to (6).
a _i = Σ (V _m −V) (P _mi −P _m ) / Σ (V _m −V) ² (2)
b _i = P _m −a _i V (3)
V _m = ΣP _mi / 1024, m = 1 to n, i = 1 to 1024 (4)
_{V = ΣV m / n ... (} 5)
P _m = ΣP _mi / 1024 (6)

According to the following equations (7) to (10), the correction coefficient Ai and the correction constant Bi are set so that the sensitivity ai and the offset bi of all the channels become the sensitivity average a and the offset average b. To calculate.
a = Σa _i / 1024 (7)
b = Σb _i / 1024 (8)
A _i = a / a _i (9)
B _i = b− (a / a _i ) / b _i (10)
Correction coefficients A _i and correction constants B _{i for} each channel are calculated and determined in the CCD controller 2. The calculated correction coefficient and correction constant are transmitted to the variable gain means 2c of the CCD sensor 1d, and the measured light quantity is made uniform by performing gain adjustment for changing the gain of each channel of the sensor.
For example, as shown in FIG. 2, when the sensitivity of some channels of the CCD sensor 1d is 0.7, 0.9, or 0.6 due to deterioration (the sensitivity is 1 when there is no deterioration). ), The gain of the variable gain means 2c is adjusted to correct the sensitivity lowered by the above deterioration as A, B, and C. Thereby, the sensitivity of the sensor is made uniform.

In the above embodiment, the average output Vm is used in the correction coefficient / constant calculation. However, other values such as the irradiation time, the number of irradiation pulses, and the integrated output may be used.
Further, any number of light input levels (number of n) may be used, and the accuracy increases as the level increases. Furthermore, since the correction coefficient / correction constant is intended to make the sensitivity and offset of each channel uniform, it may not be based on the sensitivity average and the offset average, for example, sensitivity MAX and sensitivity MIN.
The light input to the spectrum measuring instrument is not limited to the laser device, and other light sources may be used. The number of channels of the CCD sensor is not limited to 1024, and any number of channels may be used.

FIG. 3 is a diagram showing an example of variation in sensitivity of each channel of the CCD sensor.
FIG. 3A shows an example of each channel output Pm when the light input is L1 and L2, and V1 and V2 indicate average outputs of the channels of all the sensors when the light input is L1 and L2, respectively. As shown in the figure, it can be seen that there is a variation in sensitivity for each channel.
FIG. 3B shows an example of the sensor output Pm of each channel 1 to 1024 with respect to the average output V1 to Vm of each channel, and the broken line shows the average of each channel.
As shown in the figure, the sensitivity is different for each channel and the values of the coefficients a _i and b _i are different. As described above, the correction coefficient A _i and the correction constant B _i for each channel are obtained. By correcting, the sensitivity of the sensor can be made uniform.

FIG. 4 is a diagram for explaining the effect of the sensitivity adjustment of the CCD sensor. FIG. 4A shows the variation in sensitivity of each channel before and after the sensitivity adjustment of the CCD sensor.
As shown in the figure, the output characteristics of each channel can be made almost flat by adjusting the sensitivity of this embodiment.
FIG. 4B shows an E95 error with respect to the wavelength. Note that, as described above, E95 is a wavelength width in which 95% of the light energy exists, and the E95 error is a difference from the value of E95 measured by a reference measuring instrument. As shown in FIG. 4B, the E95 error value is greatly improved after sensitivity adjustment.
As described above, it is possible to reduce the wavelength dependence of the measurement error by equalizing the sensitivity of the deteriorated sensor as described in the present embodiment.

FIG. 5 is a flowchart showing the sensitivity correction process of the CCD sensor. The gain adjustment process will be further described with reference to the flowchart of FIG.
The laser controller 14 turns off the power of the oscillation stage laser 10, stops the oscillation stage laser 10, and causes the amplification stage laser 11 to oscillate alone (step S1 in FIG. 2). As a result, the amplification stage laser 11 enters a laser oscillation mode having a broad spectrum, and outputs light Lm (m = 1 to n) having a spectrum width wider than the free spectrum width (FSR) of the etalon 1b of the spectrum measuring instrument 1.
On the other hand, zero adjustment is performed so that the sensor output of each channel i of the CCD sensor 1d when there is no light input becomes 0 (step S2).
Next, m is set to 1 (step S3), and light Lm having a light quantity with which the output average of all the channels i of the sensor is Vm is emitted from the laser device and incident on the spectrum measuring instrument 1 through the beam splitter 13. Then, this light is irradiated onto the sensor surface of the CCD sensor 1d via the etalon 1b, and the output Pmi of each channel i is measured (step S4).
It is determined whether m> n (step S5). If m> n, m is set to m + 1 (step S6), and the process returns to step 4 where m = 2, m = 3, m = 4,. In this manner, the above operation is repeated while increasing m, and if m> n, the process proceeds to step S7.

Using the output Pmi of each channel with respect to the output average Vmi obtained in step 4 above, for each channel of the CCD sensor, the slope ai and the offset bi in the equation (1) when the horizontal axis is Vm and the vertical axis is Pm. Is calculated by the method of least squares (step S7).
Next, the average a of the inclination ai and the average b of the offset bi of all the channels i of the CCD sensor are calculated by the equations (7) and (8) (step S8).
A correction coefficient Ai and a correction constant Bi for which the slope ai and the offset bi of each channel i are the average values a and b are calculated for each channel i by the above equations (9) and (10) (step S9).
The correction coefficient Ai and the correction constant Bi are sent to the variable gain means 2c of the CCD sensor, the correction coefficient Ai and the correction constant Bi are adjusted, and the gain value and the offset value of each channel are changed (step S10).

It is a figure which shows the structure of the laser apparatus of an Example of this invention, and a vector detector. It is a figure explaining the sensitivity adjustment of each channel of a sensor. It is a figure which shows an example of the variation in the sensitivity of each channel of a CCD sensor. It is a figure explaining the effect by the sensitivity adjustment of a CCD sensor. It is a flowchart which shows the sensitivity correction process of the sensor of the Example of this invention. It is a schematic diagram of the optical system of the spectrometer used for a laser apparatus. It is a figure explaining degradation of a photosensor which arises when measuring a fringe pattern with photosensor arrays, such as a CCD sensor. It is a figure explaining the problem which arises when measuring with a deteriorated sensor.

Explanation of symbols

1 Spectral measuring instrument 1a Incident lens 1b Etalon (spectral means)
1c Imaging lens 1d CCD sensor (photo sensor array)
2 CCD controller 2a Spectrum calculation unit 2b Correction coefficient / constant calculation unit 2c Variable gain means 10 Oscillation stage laser 11 Amplification stage laser 10a, 11a Power supply 12 Monitor module 13 Beam splitter 14 Laser controller

Claims (4)

A method for adjusting the sensitivity of a spectroscope that detects light passing through a spectroscopic means with an optical sensor array and measures a spectrum based on the output of the optical sensor array,
Light having a spectrum width wider than the free spectral width (FSR) of the spectroscopic means is incident on the spectroscopic means, the light passing through the spectroscopic means is detected by the photosensor array, and an optical sensor is based on the detection result. Find the sensitivity to incident light in each channel of the array,
Based on the sensitivity of each channel above, find the correction value for each channel,
A method for adjusting the sensitivity of a spectroscope, wherein the gain of each channel of the photosensor array is adjusted based on the correction value. A spectroscope comprising spectroscopic means, an optical sensor array, and a light source that emits light having a spectral width wider than a free spectral width (FSR) of the spectroscopic means,
An adjustment means for adjusting the gain of each channel of the optical sensor array is provided,
The adjusting means causes light from the light source to enter the optical sensor array, obtains sensitivity to incident light in each channel of the optical sensor array, obtains a gain correction value for each channel based on the sensitivity,
A spectrometer which adjusts the gain of each channel of the photosensor array based on the correction value. The adjustment means calculates the sensitivity and offset of each channel by calculating the optical sensitivity and offset of each channel with respect to either the average output of each channel of the optical sensor array, the output integration, the measurement time, or the number of pulses. 3. The spectroscope according to claim 2, wherein a correction value for each channel is obtained based on the photosensitivity and offset of each channel and the average value, maximum value, or minimum value thereof. The spectroscope according to claim 3, wherein the spectroscope is attached to a pulse laser device and measures a spectral line width of laser light emitted from the pulse laser device.

Images