JPH02165003A - Position detecting method - Google Patents

Abstract

PURPOSE:To obtain approximately constant level of a detecting output without adversely influencing detecting time and detecting accuracy by detecting a peak value of an output signal of a beam detector thereby to form a reference signal, dividing said output signal of the beam detector by said reference signal thereby to obtain a signal and synchronously detecting said signal. CONSTITUTION:Light beams radiated from a beam source comprised of a light source 1, a condensing lens 2 and the like are irradiated onto an object to be tested via a half mirror 3 and an objective lens 4. The light beams are irradiated, for example, to the vicinity of a linear mark 6 on the object 5 which mark is drawn for detecting the position of the object. The reflecting light from the mark 6 moves upwards through the objective lens 4 and half mirror 3, going through a very small gap 7a of a slit plate 7 to be received by a photoelectric converter 8. The converted signal passes through a preamplifier 9, an AGC circuit 14 and a synchronous detection circuit 10 and is output to an indicating meter 13. In this case, if the signal is not larger than the allowable value at the circuit 14, a limiter circuit 22 outputs a predetermined value, and no division by the value not larger than the allowable value is executed by a divider 23. Therefore, even if the light amount is changed, the detecting output does not change.

Description

[Detailed description of the invention] [Purpose of the invention] (Industrial application field) The present invention relates to a method for detecting one position.

(Prior Art) There are several methods for detecting the relative position between two members. Among them, there are a vibration type position detection method and a differential slit type position detection method. These position detection methods are applied to photoelectron microscopes and the like.

Since the vibration type position detection method and the differential slit type position detection method can easily obtain high detection sensitivity, they have recently been used as position detection methods in various automatic positioning processes such as semiconductor manufacturing processes.

The principle of the vibration-type position detection method will be briefly explained using an example in which it is used in a photoelectron microscope. Its basic configuration is as shown in FIG. That is, in the figure, 1 is a light source, 2 is a focusing lens, 3 is a half mirror 14 is an objective lens, and 5 is an object to be inspected with a mark 6 attached thereto.

7 is a slit plate, 8 is a photoelectric converter, 9 is a preamplifier, 1
0 is a synchronous detection circuit, 11 is an oscillator, 12 is a resonator, 13
indicates an indicator meter. In this example, a vibrator 12 is used as a mechanism for vibrating the slit plate 7.

The light reflected by the mark 6 passes through the half mirror 3 and the slit plate 7 and is detected by the photoelectric converter 8. Since the slit plate 7 is vibrating, the waveform of the output signal of the photoelectric converter 8 differs depending on the position of the mark 6. Photoelectric converter 8
The relationship between the output signal waveform and the position ffi x of mark 6 is shown in FIGS. 12 (Sl) to (S9). 1st
As can be seen from FIG. 2, the waveform of the output signal is a synchronizing signal with a frequency equal to the vibration frequency or a frequency multiple thereof.

Then, when the position of the mark 6 coincides with the vibration center position of the slit, that is, at point (Ss), the signal becomes a signal with twice the frequency, and the fundamental frequency component becomes zero. In this position detection method, the fundamental frequency component is detected by the synchronous detection circuit 10. The figure on the right shows how the fundamental frequency component aw changes with respect to a change in the position X of the mark 6. The way this fundamental frequency component changes (output characteristic curve) varies depending on the mark width, slit width, vibration amplitude, etc., but it is zero at the point (Ss) where the position of mark 6 coincides with the vibration center of the slit. , the sign is reversed before and after that. Therefore, the relative position of the slit plate 7 and the object to be inspected 5 can be detected from the output of the synchronous detection circuit 10. In this method, since the output of the synchronous detection circuit 10 changes as described above as the relative positions of the two members change, detection using a zero meter is not only possible.

It is also advantageous when driving a servo system.

By the way, the output of the synchronous detection circuit 10, that is, the shape of the output characteristic curve, depends on the width of the mark, the contrast of the mark, and the scanning (
Vibration) varies depending on amplitude and brightness of lighting, etc. Therefore, in order to obtain correct position information, it is necessary to perform calibration each time. Otherwise, only inaccurate position detection will be possible. For example, if we use this position detection method to detect a position in a semiconductor manufacturing process typified by a light exposure process,
In the semiconductor manufacturing process, exposure is repeated across a wide variety of processing steps, so the contrast of the mark varies each time due to differences in the process and resist conditions.
As shown in FIG. 13, the outputs aft, af2. af3. af
, 1 peak value PI rP2. P3. The P4 and slope will be different.

If the output forms differ in this way, accurate position detection cannot be performed since position detection is performed based on the level value of the output.

Therefore, in order to eliminate such problems.

One possible method is to input contrast information of marks due to differences in processes into a computer in advance and correct the output based on this information. However, with this method, since the contrast information of the mark is based on the pre-M1, highly accurate correction cannot be performed. moreover.

The above output also becomes unstable due to fluctuations in illuminance. Therefore, if sufficient illuminance stability cannot be obtained, the peak value and slope of the output will still change. Since such changes over time are difficult to predict, it becomes difficult to correct the output with high precision.

On the other hand, the same can be said about the differential slit type position detection method. FIG. 14 shows an example of aligning a mask and a wafer in a reduction projection exposure apparatus using a differential slit type position detection method. In the figure, 31 is a mask as a first member, 32 is a reduction projection lens, 33 is a wafer as a second member, and 34 is a moving table. The position signal detection system is
Laser light source 41. AOD element 43. AOD driver 44
, oscillator 45. Half mirror 471 reflection mirror 47',
Photoelectric converter 51. Preamplifier 52 and synchronous detection circuit 5
Consists of 3rd class. In addition.

Although not shown, an exposure light source is arranged above the mask 31.

A pair of marks 35 for position detection are provided on the lower surface of the mask 31, and a pair of marks 36 for position detection are provided on the upper surface of the wafer 33. As shown in FIG. 15(a), the mask mark 35 is formed by separating a pair of rectangular patterns by a distance g1 in a direction perpendicular to the longitudinal direction of the patterns. Similarly, the wafer mark 36 is as shown in FIG. 15(b).

Distance between a pair of rectangular patterns fI2 (g2-Fg])
are formed with a distance of

In the above configuration, the light beam emitted from the laser light source 41 is irradiated onto the mask mark 35 through the AOD element 43 and the mirrors 47 and 47'. Here, the amount of beam deflection by the AOD element 43 is determined appropriately, and the mask marks 35 are alternately illuminated. The light beam passing through the mark 35 is irradiated onto the wafer 33 via the projection lens 32.

The reflected light from the wafer 33 is transmitted to the projection lens 32. mask 31
The light is then received by the photoelectric converter 51 via the half mirror 47. The detection signal of the photoelectric converter 51 is transmitted to the synchronous detection circuit 5
3, the waves are detected in synchronization with the alternate irradiation. here,
When one side (area A) of the mask mark 35 is illuminated,
The output characteristics with respect to positional deviation obtained by the photoelectric converter 51 when the other side (area B) is illuminated are as shown in FIG. 16(a). As can be seen from the figure, the output maximum (peak) positions of area A and area B are different, and the two light amounts are balanced at the intersection point Po.

Therefore, an AOD element 43 is installed in the optical path of the laser irradiation light, and a predetermined high frequency signal is applied to this element 43 by an AOD driver 44. As a result, the direction of the irradiation beam changes periodically, and the direction of the irradiation beam changes periodically.

Area B will be illuminated alternately. in this case.

Considering only the AC component of the output signal of the photoelectric converter 51,
It becomes maximum when the difference between the solid line and the broken line shown in FIG. 16(a) is maximum, and becomes zero at the balance position Po. Therefore, the output characteristics obtained by performing synchronous detection are shown in Figure 16(c).
If the alignment is performed at the balance position Po where the two curves intersect, not only zero point servo is possible, but also ideal alignment with good linearity is possible. In addition, since synchronous detection type position detection is possible without applying mechanical vibration, it is less susceptible to effects such as drift, and highly accurate position detection is possible.

However, the above-described position detection method also suffers from the same problems as the above-described vibration-type position detection method. That is, in the semiconductor manufacturing process, exposure is repeated across a wide variety of processing steps.

The contrast of the mark differs each time due to differences in the process, resist conditions, etc., and in the end, the result is almost the same as that shown in FIG. 13.

For this reason, recently, in order to achieve more accurate position detection, signal AGC (auto gain control) has been implemented.

It has become necessary to keep the slope of the output characteristic curve constant. A simple AGC method is to always approach the mark from one direction and control the gain so that the peak value of the output characteristic curve is constant. However, with this method, it is necessary to perform detection while scanning the mark for a certain period of time.

This leads to a delay in position detection time, resulting in a reduction in the throughput of the device. Furthermore, it is difficult to respond to changes in the signal over time while detecting the peak value of the output characteristic curve, and it is difficult to perform accurate position detection in a short time.

Conventionally, the method of performing AGC in real time is to detect the effective value and peak value of the AC component of the signal.

A method of adjusting the gain so that its value is always constant has been considered. However, the detection signal in the vibration-type position detection method includes not only the fundamental wave component but also higher-order frequency components, and has a complex signal waveform, with the output components greatly varying depending on the mark position. Therefore, with such a method, it is difficult to perform AGC with high accuracy without losing position information.

On the other hand, in the differential slit type position detection method.

The detection signal is a rectangular wave with only the fundamental wave component.

The magnitude of the AC component appears with mark position information. In this case, if the gain is adjusted so that the effective value and peak value of the AC component are constant, one position information itself will be lost, and therefore one position detection itself will become impossible.

(Problems to be Solved by the Invention) As described above, with the conventional position detection method, it is difficult to perform highly accurate position detection in a short time.

Therefore, the present invention provides the following advantages:
Another object of the present invention is to provide a position detection method that can keep position detection output characteristics constant even if the level of an input position signal changes.

[Structure of the Invention (Means for Solving the Problem) As a result of extensive research, the present inventors have discovered that the peak value of a position signal modulated by vibration etc. can serve as a standard representing the level of the position signal. I found it.

That is, one aspect of the present invention is to provide a mark section on a first member, irradiate a beam obtained through the mark section to a mark member or a second member having a slit, and emit a beam obtained from the second member. detecting with a beam detector and relatively vibrating the beam and the second member;
In the position detection method, the output signal of the beam detector is synchronously detected in synchronization with this vibration, and the relative position of the first member and the second member is detected using the detected output. A reference signal is created by detecting the peak value of the output signal of the beam detector, and a signal obtained by dividing the output signal of the beam detector by this reference signal is synchronously detected.

(Function) - With method h, even if the level of the modulated position signal changes temporarily, the position signal is divided by the reference signal created based on the peak value that represents it. This is because the signals obtained are synchronously detected.

The level of the synchronous detection output is approximately constant. Therefore, position detection is possible without impairing detection time or detection accuracy.

(Example) The method of the present invention will be explained below with reference to one drawing.

FIG. 1 is a schematic diagram of an example in which the present invention is applied to a vibration type position detection method.

A forward beam emitted from a beam irradiation source consisting of a light source 1, a focusing lens 2, etc. is irradiated onto an object to be inspected 5 (first member) via a half mirror 3 and an objective lens 4. For example, a linear position detection mark 6 is formed on the object 5 to be inspected, and the light beam is irradiated near the mark 6. The reflected light from the mark 6 due to the light beam irradiation is as follows.

The light travels upward through the objective lens 4 and the half mirror 3, passes through the minute gap 7a of the slit plate 7 (second member), and is received by the photoelectric converter 8.

The detection signal converted by the photoelectric converter 8 is sent to a preamplifier 9.
AGC circuit 14. The signal is output to the indicator meter 13 through the synchronous detection circuit 1o. here.

The other configurations, signal processing principles, etc. except for the AGC circuit 14 are the same as those described in the conventional example.

A detailed explanation thereof will be omitted. What is different from conventional methods?

AGC circuit] 4 is arranged to perform signal processing.

FIG. 2 is a block diagram showing a schematic configuration of the AGC circuit 14. The input signal is branched into two, one of which passes through the bypass filter 20 and enters the divider 231; Let this be H. The other signal enters a divider 23 via a peak detector 21 and a limiter circuit 22. Let this be P. Divider 2
3 performs H/P calculation and outputs the result. This becomes the output of the AGC circuit 14.

The operation of this AGC circuit 14 will be explained using FIG.

The input signal has various waveforms depending on the position of the mark 6, but in FIG. 3, (S3) and (S
4), only the signal at the position corresponding to point (Ss) is shown. First, consider the position (S3).

As described above, the level of the input signal varies as shown by the solid line, broken line, and dashed-dotted line in accordance with the change in the amount of light due to the difference in the reflectance of the mark part. When this signal passes through the )1 pass filter 20, it becomes the waveform shown in (2) [bypass output]. On the other hand, peak detector 2
1 detects and holds the peak value of the input signal every time it arrives, and sends out the DC output shown in (2) [Peak detector output]. Here, the level ratio of each signal with respect to the change in light amount has a constant relationship. That is, the amplitude (P-P value) of (2) [bypass output] changes in accordance with the change in the amount of light, and the level of (2) [peak detector output] also changes in accordance with the change in the amount of light. Therefore, the result of division by the 1 divider 23 is an AC component waveform that does not depend on changes in the amount of light, as shown in (2) [divider output]. Mark position (S, i
) and (Ss) can be considered in the same way, and the output of the divider by 7 becomes an AC component waveform that does not depend on changes in the amount of light. Although omitted in the figure, the same effect can be obtained at mark positions (S6) and (S?). Mark position (Sl) (S2), (S8), (
In the signal obtained by Sg), since the DC component becomes small, it is conceivable that the denominator of the divider 23 becomes less than the allowable value, but in this embodiment, the limiter circuit 22 solves this problem. That is, the limiter circuit 22 outputs a constant value when the value is less than the allowable value, and the divider 23 does not perform division by less than the allowable value.

FIG. 4 shows the output characteristics as a result of performing synchronous detection processing on the signal that has passed through the AGC circuit 14 configured in this manner. In this figure, solid lines, broken lines, and dashed-dotted lines indicate differences in input signals (differences due to changes in light amount) corresponding to FIG. 3.

As can be seen from this figure, even if there is a difference in the amount of light, there is no difference in the detection output in the range of mark positions (S3) to (S7), which are particularly problematic in position detection. In other words, even if there is a difference in the amount of light, the detection output does not change within the detection range that should be considered as one problem due to the effect of AGC.

Therefore, in the position detection method described above, even if there is a change in the input signal due to a difference in the amount of light, automatic gain adjustment (AGC) is not performed.
This means that a position detection signal is obtained. Therefore, tremendous effects can be obtained when applied to positioning in semiconductor manufacturing equipment and the like.

Note that the present invention is not limited to the above-mentioned example.

FIG. 5 is a schematic configuration diagram showing an example in which the method of the present invention is applied to a reduction projection exposure apparatus that employs a differential slit type position detection method for mask/wafer alignment.

In this example as well, the other (1M) components except for the AGC circuit 14 are the same as those shown in the conventional example, so the same numbers are given and the explanation thereof will be omitted.

In this example, changes in the input signal and detection signal corresponding to those shown in FIG. 3 described above are as shown in FIG. 6. As for the waveform of the input signal, unlike the case of FIG. 3, all AC components are rectangular wave components only.

However, the level of the peak value changes according to the change in light intensity, and by dividing ■[bypass output] by ■[peak detector output], AGC
The effect works. In other words, the output signal obtained is only a rectangular wave component that is not related to changes in the amount of light.

Therefore, in this example as well, even if the human input signal level changes due to a difference in the amount of light, a position detection signal subjected to automatic gain control (AGC) can be obtained, making it possible to perform highly accurate position detection and alignment.

FIG. 7 shows a modification of the AGC circuit, and this AGC circuit 1
4a is different from the AGC circuit shown in FIG.
The reason is that a subtracter 25 is added to subtract a predetermined offset amount determined in advance from the signal inputted to the denominator. The purpose of adding the subtracter 25 in this way is to cope with the case where the input signal includes an offset as shown in FIG. Possible causes of this offset include the influence of external light that cannot be completely blocked and noise generated in electrical circuits. This offset amount is measured in advance, and a subtraction circuit 25 subtracts the offset amount from the output component of the peak detector 21. As a result, the component input to the denominator of the 1 divider 23 is as follows.

No offset will be included. Therefore, it is possible to prevent the accuracy of position detection from decreasing due to the amount of offset.

Note that the offset amount may be measured by measuring the level of the signal (human input signal) while cutting off the input light. Further, in order to perform this in real time, it is possible to apply brightness modulation at a frequency sufficiently higher than the WR dynamic number. FIG. 9 shows an example of an input signal waveform when brightness modulation is applied.

By measuring Vl-V2 in synchronization with the modulation signal, the amount of offset can be determined.

Although the shape of the mark pattern on each side has been described as a single linear pattern, which is commonly used, the shape of the mark pattern may also be a lattice shape as shown in FIG. 10.

Further, a slit plate for vibrating the light beam may be disposed on the light incidence side, that is, between the beam irradiation source and the mark. Furthermore, it is also possible to use transmitted light that has passed through the mark and the sample instead of reflected light as the detection light. Further, the beam irradiation source may emit an electron beam instead of light. In this case, the electron beam may be deflected as a means for vibrating the beam. moreover.

As the detection beam, reflected electrons, secondary electrons, X-rays, etc. may be used. Two other modifications may be made without departing from the gist of the present invention.

[Effects of the Invention] As detailed above, according to the present invention, deterioration of position detection characteristics caused by differences in reflectance of marks and samples can be prevented by relatively simple means, and detection time and detection accuracy are impaired. The position detection characteristics can be kept constant without any problems. Therefore, it is possible to detect and align mark positions with high precision for samples that have gone through a wide variety of processes, and this is extremely effective in, for example, semiconductor manufacturing processes.

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram for explaining an example in which the method of the present invention is applied to position detection of a vibrating photoelectron microscope. Figure 2 is a block diagram showing the specific configuration of the AGC circuit, Figure 3 is a signal waveform diagram to explain the AGC effect, and Figure 4 is the change in characteristic curve due to the AGC circuit when the level of the input signal changes. FIG. FIG. 5 is a schematic configuration diagram for explaining another application example of the method of the present invention, FIG. 6 is a signal waveform diagram for explaining the AGC effect on the same side, and FIG. 7 is a modified example of the AGC circuit. 8 is a waveform diagram of an input signal including an offset, FIG. 9 is a waveform diagram of an input signal when brightness modulation is applied, FIG. 10 is a plan view showing a modified example of the mark pattern, and FIG. 16 to 16 are diagrams for explaining the conventional method. DESCRIPTION OF SYMBOLS 1... Light source, 2... Focusing lens, 3... Half mirror 4... Objective lens, 5... Inspection object (sample),
6... Position detection mark, 7... Slit plate, 8...
...Photoelectric converter, 9...Preamplifier, 10.53...
・Synchronous detection circuit. 11.45... Oscillator, 12... Vibrator, 13...
・Indication meter, 14, 14a-AGC circuit, 20-/-
Ibus filter, 21...Peak detector, 22.
...Limiter circuit, 23...Divider, 25...Subtractor. 26... Offset setting circuit, 31... Mask, 3
2... Projection lens, 33... Wafer, 41... Laser light source. 43... Acousto-optic deflection element (AOD). Applicant's representative Patent attorney Takehiko Suzue

Claims (6)

[Claims] (1) A mark part is provided on the first member, a beam obtained through the mark part is irradiated onto a mark member or a second member having a slit, and the beam obtained from the second member is sent to a beam detector. At the same time, the beam and the second member are vibrated relatively, the output signal of the beam detector is synchronously detected in synchronization with this vibration, and the detected output is used to detect the beam and the second member. In the method of detecting the relative position with respect to the second member, a reference signal is generated by detecting the peak value of the output signal of the beam detector, and the output signal of the beam detector is divided by this reference signal. A position detection method characterized in that the obtained signals are synchronously detected. (2) The position detection method according to claim 1, wherein light is used as the beam. (3) The position detection method according to claim 1, wherein the reference signal is a signal having a level obtained by subtracting a predetermined offset amount from the peak value. (4) The position detection method according to claim 3, wherein the predetermined offset amount is a signal component that is mixed in regardless of the position detection signal. (5) A first mark portion consisting of a pair of rectangular patterns or a diffraction grating pattern is provided on the first member at a predetermined distance l_1, and a pair of rectangular patterns are provided at a predetermined distance l_2 on the second member. Alternatively, a second mark section consisting of a diffraction grating pattern is provided, and the first mark section is irradiated with a light beam alternately at a constant period, so that the light beam is emitted from the first mark section via the second mark section. The obtained light beam is detected by a photoelectric converter, the output signal of this photoelectric converter is synchronously detected in synchronization with the period of light beam irradiation, and the first member and the second member are connected based on this detection output. In the position detection method for determining the relative positional deviation amount of the photoelectric converter, a reference signal is created by detecting the peak value of the output signal of the photoelectric converter, and the output signal of the photoelectric converter is divided by this reference signal. A position detection method characterized by synchronously detecting a signal. (6) The reference signal is a signal having a level obtained by subtracting an offset amount of a disturbance light component unrelated to position detection, a component determined by a circuit, etc. from the peak value of the output signal of the photoelectric converter. position detection method.