JPH03231420A - Projection optical device - Google Patents

Abstract

PURPOSE:To enable an optimum image surface to be detected simply and readily by receiving light which is reflected from a light-sensitive substrate, etc., and is returned through a projection optical system and a mask with a photometric element at a required conjugated position and then by detecting an average focusing surface position. CONSTITUTION:Rays from a pupil EP of a projecting optical system 10 of an illumination distribution unification means 2 and from a secondary light source on an irradiation surface at a conjugate position enable a substrate 11 to be irradiated through a reticle blind 5 in conjugate relation, a pattern surface of a mask 9, the mask 9, and a projection optical system 10. Then, a reflected light from a wafer 11 reverses the same light path and then enters the pupil EP and a photometric element 15 at a conjugate position. In this case, when the wafer 11 is located at a focusing position of the optical system 10 and the wafer 11 and the mask 9 are in a conjugate relationship, reflected light image is formed again at a pattern of the mask 9 and all reflection lights pass through the mask 9. Thus, by controlling a stage 12 through a focus controller 16 so that the amount of light entering an element 15 reaches the maximum, an average optimum image surface including curve and slant can be detected simply and readily in first attempt without pattern printing, development, and photometry.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a projection exposure apparatus used in the manufacture of semiconductor integrated circuits, etc., and particularly relates to an improvement in focal position detection means of a projection optical system. .

[Conventional technology]

As a conventional focus position detection means for this type of apparatus, there is a so-called gap sensor that detects the distance between the projection optical system and the photosensitive substrate. This can be done, for example, by entering a light beam onto the photosensitive substrate obliquely with respect to the optical axis of the projection optical system and detecting the distance by detecting the position of the reflected light, or by blowing air onto the photosensitive substrate to remove the air. There have been methods of detecting intervals by detecting pressure changes. However, it is known that the focal position of the projection optical system changes due to changes in atmospheric pressure, air temperature, the temperature of the projection optical system, humidity, or an increase in the temperature of the projection optical system due to illumination light. The above-mentioned gap sensor alone cannot deal with these factors, and has been used in combination with some kind of correction means to correct for the above-mentioned changes.

[Problem to be solved by the invention]

In the above-mentioned conventional technology, the means for correcting changes in focus position stores in advance the amount of change in focus position for each cause of variation, measures the causes of variation, estimates the focus position, and applies correction. ing. Therefore, there is a problem in that it cannot cope with changes caused by causes other than the fluctuation characteristics stored in advance, such as long-term drift of the device, and requires periodic calibration.

Further, the focal position is generally measured only at the center position of the exposed image plane, and the focal position correction is also performed with respect to the center position. This is because the focal position is measured by moving the photosensitive substrate in the optical axis direction to find the best image surface.
Since it is complicated to measure at multiple points within the image plane while sequentially exposing patterns, the value at the center position is representative. However, in reality, there is a curvature of the field of view, and this curvature, like the focal position, changes due to various causes. Therefore, when considering the entire exposure area, there is a problem in that if the focus position is corrected only at the center position of the image plane, there is a possibility that defocus will occur in the peripheral area.

[Means to solve problems]

In order to solve this problem, the present invention includes a light source l that illuminates a mask 9 on which a predetermined pattern is formed, and an image of the pattern that is projected onto a photosensitive substrate or a light reflective substrate 11 in a predetermined image formation state. In a projection optical device equipped with a projection optical system lO, the amount of reflected light reflected from the photosensitive substrate or light reflective substrate 11 and returned to the light source 1 side via the projection optical system 10 and the mask 9 is transmitted to the projection optical system. A photometric element 15 detecting at a position substantially conjugate with the pupil of the system 10, and a substrate 1
Focal position detection detects the average focal plane position of a predetermined area within the field of view of the projection optical system IO based on the photometric output of the photometric element 15 obtained when the distance between the mask 1 and the mask 9 is relatively changed. 8 means 16, and the mask 9 has lines and spaces with a line width near the resolution limit of the projection optical system 10.

[For production]

In the present invention, reflected light of illumination light from a photosensitive substrate or a light reflective substrate is measured through a projection optical system and a mask. In addition, in this type of projection optical device, in order to prevent projection magnification errors due to defocus, the image side of the projection optical system, that is, the side of the photosensitive substrate or light reflective substrate is a telecentric optical system.

Therefore, when the photosensitive substrate or the light reflective substrate and the mask are in a conjugate position via the projection optical system, that is, when the photosensitive substrate or the light reflective substrate is at the focal point of the projection optical system, as shown in FIG. Reflected light from a photosensitive substrate or a light reflective substrate such as Al is re-imaged on the pattern of the mask, and all of the reflected light passes above the mask (towards the light source).

On the other hand, when the photosensitive substrate or the light reflective substrate and the mask are not in a conjugate position, the reflected light from the photosensitive substrate or the light reflective substrate is not imaged on the pattern of the mask, as shown in 31st ffl(b). First, some light rays are blocked by the mask pattern and do not pass above the mask (towards the light source). By utilizing the above phenomenon, the position of the photosensitive substrate where the amount of reflected light is maximum can be detected as the focal position. This method is
Since the exposure light beam is used as it is, an accurate focal position can be obtained, and the correction deviation that occurs in the conventional method does not occur.

In addition, in the present invention, the light receiving section for reflected light is installed on a plane that is almost conjugate with the pupil of the projection optical system, and since it can receive reflected light from the entire exposure area, curvature, tilting, etc. of the image plane can be avoided. The most average image plane can be detected.

〔Example〕

FIG. 1 is a schematic diagram of a projection optical device according to a first embodiment of the present invention. The light source l is a mercury lamp, a laser light source, etc., and the light beam emitted from the light source l is a fly lamp.
The illuminance distribution uniforming unit 2 such as an eye lens is irradiated with the light. A secondary light source generated on the exit side of the illuminance distribution uniformizing section 2 passes through a semi-transmissive mirror 3 and is irradiated onto a reticle blind 5 via a lens system 4. The light flux from the reticle blind 5 is irradiated onto a reflecting mirror 7 via a lens system 6, and the reflected light is irradiated onto a mask 9 via a lens system 8.

The light beam from the mask 9 is projected and exposed onto a wafer 11 as a photosensitive substrate via a projection optical system IO.

At this time, the secondary light source generated on the exit surface of the illuminance distribution uniformizing means 2 is the lens system 4. 6. Projection optical system l by 8
It is conjugate with the pupil (exit) plane (Ep) of
1 are in a conjugate relationship by the lens systems 6 and 8 and the projection optical system 10, respectively.

Next, a method for detecting a focal point position according to the present invention will be explained. The reflected light from the wafer 11 is transmitted to the projection optical system IO and the mask 9.
The light is reflected by the semi-transmissive mirror 3 and received by the reflected light sensor 15. A signal from the reflected light sensor 15 is sent to a focus controller 16. focus controller 1
6 sends a signal to the motor 17 that drives the stage 12 in the optical axis direction, moves the stage 12 slightly in the optical axis direction, and finds the position where the output of the reflected light sensor 15 is maximum. At this time, if the light-receiving surface of the reflected light sensor 15 is located at a position substantially conjugate to the pupil of the projection optical system 10, it can receive reflected light from the entire exposure area of the wafer 11. Therefore, if the image plane is curved or tilted, the above method allows the wafer 11 to be placed on the image plane where the error is minimized, that is, on the average image plane.

Further, since the reticle blind 5 is located at a position conjugate with the pattern surface of the mask 9, it functions to block the exposure light beam when only a part of the pattern on the mask is exposed. Even when only the peripheral pattern of the mask 9 is exposed using this reticle blind 5, the reflected light from the wafer 11 returns only from the exposure area limited by the reticle blind 5, so the average in the exposure area image plane can be detected.

Furthermore, although not shown, if the stage 12 has a function (wafer leveling stage section) that tilts the stage 12 from a plane perpendicular to the optical axis, the tilt of the wafer 11 or The tilt of the image itself can be canceled and the best image plane can be obtained. Further, since the phenomenon in which the reflected light from the wafer 11 is blocked by the pattern of the mask 9 is used, areas where the pattern has many edges, that is, areas where the pattern is fine, can be measured sensitively. For this reason, a mask with a mixture of fine parts and rough parts of the pattern has the advantage that it is possible to detect an image plane with emphasis on the fine parts of the pattern. However, since the above method uses exposure light, the sea urchin/X11 is exposed even during the detection of the focal position. Therefore, if the focus positioning operation is completed in a fraction of the exposure time or less, the remainder can be exposed at the best image plane, making it practical. The exposure time is at most 0.1 to 0.00 mm in the case where the light source 1 uses the g-line or i-line of a mercury lamp. Since the time is about 2 seconds, the motor 17 must move the stage 12 up and down at high speed to search for the best image plane. Alternatively, insert an ND Lafurter that can be taken in and out at high speed into the optical path, and place the wafer 11 while searching for the best image plane.
Possible means include reducing the exposure light.

However, with the above method, a sufficient signal cannot be obtained when the reflectance of the wafer 11 is low. For this reason, the projection optical system 1
A possible method is to measure the focal position using the highly reflective surface 18 using a mechanism that keeps the wafer 11 at a constant distance from the wafer 11, and to align the wafer 11 with the focal position measured at that time. The method will be briefly explained below.

When a new wafer 11 is loaded onto the stage 12,
The stage 12 moves the highly reflective surface 18 below the projection optical system 10. Here, in the same manner as described above, the reflected light sensor 15 is
The focal position is detected using the projection optical system 10.
The distance between the high reflection surface 18 and the high reflection surface 18 is measured. This is a conventional mechanism for keeping the wafer 11 and the projection optical system IO at a constant distance, that is, the projector 13. This is done using the light receiver 14. The projector 13 includes a light source such as an LED and a condensing lens, and projects a light beam onto the highly reflective surface 18 obliquely from above. The light receiver 14 is SP
It consists of a light receiving element such as D and a condensing lens, and receives reflected light from the high reflection surface 18. Since the reflected light from the high reflection surface 18 is shifted depending on the position of the high reflection surface 18 in the optical axis direction of the projection optical system 10, the position of the high reflection surface 18 can be measured from the amount of shift.

Using this mechanism, actual exposure is performed so that the wafer 11 is positioned at the same position as the focal position determined by the high reflection surface 18.

That is, the highly reflective surface 18 that is the maximum in the reflected light sensor 15
The light receiver 14 or the focus controller 1 is set so that when the height position of
Calibrate 6. According to this method,
It is only necessary to detect the focal position using the reflection light sensor 15 once per wafer, and throughput is also improved. Furthermore, it is known that the projection optical system 10 absorbs the exposure light beam and the focal position varies, but even if one wafer is exposed at the same focal position as described above, the variation is small and does not pose a problem. Even in the case of a problem, a method can be considered in which interpolation is performed from the previous data and the correction is gradually made while the wafer 11 is being exposed. Another method for keeping the projection optical system 10 and the wafer 11 at a constant distance is to use an air micrometer, but this method can be used in the same manner as in this embodiment.

The above method is an example in which the reflected light sensor 15 is located at a position that is approximately conjugate with the pupil position of the projection optical system 10. In this case, although the average image plane can be detected, the amount of curvature or inclination cannot be detected. I don't know the amount. For this reason, it is also possible to arrange several sensors at positions substantially conjugate with the image plane and detect the respective focal positions near the center of the image or separately from the periphery.

According to this method, since the magnitude of the curvature, etc. can be known, a warning can be issued if the curvature becomes larger than an allowable value due to absorption of the exposure light beam by the projection lens 10, etc.

The above method shows an example in which focus position detection is performed using a mask that is actually used. Although this is possible in principle, in the case of a mask with few patterns, the amount of reflected light changes only slightly even if the wafer deviates from the focal position, so it cannot always be used. However, if you prepare a dedicated mask, you can easily find the focal position compared to the conventional method of determining the focal position by test exposing the wafer II, so it is a big advantage even if you use it only for measurement or adjustment. There is. The second embodiment of the present invention is one such example and is an example used for adjusting a focus position correction mechanism of a stepper.

FIG. 2 is a diagram showing a schematic configuration of a projection optical device according to a second embodiment of the present invention. The basic configuration is the same as that in FIG. First, the focal position correction mechanism will be explained. The focal point is the surrounding temperature and humidity.

It fluctuates due to environmental changes such as atmospheric pressure, temperature rise of the lens due to absorption of exposure light, long-term drift, etc.

This embodiment shows a method of measuring these factors, predicting changes in focal position, and correcting them. Regarding environmental changes, the environmental sensor 19 measures the temperature, humidity, atmospheric pressure, etc., and calculates the average image plane change using a table or formula determined in advance through simulation calculations or experiments. Regarding the absorption of exposure light, the amount of light incident on the projection lens IO is determined in advance by a light amount sensor 20 installed on the stage 12, and information on opening and closing of the shutter 21 is obtained from the shutter control circuit 22 that drives the shutter 21, and the focus is determined. A controller 16 calculates the changes in the image plane. This calculation is performed by determining in advance the change characteristic of the focal position of the projection optical system 10 due to exposure light absorption, and by storing this characteristic in a table or mathematical formula. The method of determining the above fluctuation characteristics according to the present invention will be explained below. First, the light amount sensor 2o
The amount of exposure light incident on the projection lens IO is measured. Next, the change in focal position relative to the amount of light is measured. After the shutter 21 is closed for a sufficient time and the focal position becomes sufficiently stable, the shutter 21 is opened.

Thereafter, as in the first embodiment, the position of the stage 12 is continued to be controlled so that the output of the reflected light sensor 15 is maximized. At this time, the projection optical system IO absorbs the exposure light and the focal position gradually changes, but due to the passage of time and the relationship with the position of the stage 12 when controlled as described above, the curvature of field is The average change characteristics of the focal position including changes are determined. In this method, only the change in the focal position needs to be known, so that the mask 9 and wafer 11 can be selected to be convenient for the measurement. First, regarding the mask 9, let the pattern existence rate (transmittance) be X, and the amount of light reflected from the base at best focus be α·X. α is a suitable proportionality constant. At this time, considering a state in which the focal position is completely shifted, the reflected light at the position on the mask 9 returns as a uniform light amount over the entire surface. At this time, the amount of light reflected from the base to the reflected light sensor 15 is α・x”x=α・X
It becomes 2. The larger the difference between the signals between best focus and complete defocus, the higher the resolution of the focus position. Therefore, when finding the X that maximizes α(x-x2), we get
x-=0.5 (50%).

Further, as mentioned above, the finer the size and shape, the higher the resolution. However, if the resolution is less than the resolution of the projection optical system 10, the image will be blurred and a signal with good resolution will not be obtained. From the above, it can be said that a line-and-space mask near the resolution limit of the projection optical system 10 is preferable. moreover,
If the astigmatism of the projection optical system 10 is also taken into account, a mask with orthogonal lines and spaces, a checkerboard pattern, or a pattern in the circumferential direction and radial direction drawn on the entire surface as shown in the example in FIG. 4 can be said to be optimal. . Next, regarding the wafer 11, as mentioned above, the higher the reflectance, the higher the resolution.
For example, a flat wafer coated with aluminum is desirable.

According to the above method, a reference pattern is exposed on a conventional wafer, a resist pattern formed on the wafer is observed,
Compared to the method of finding the focal point position by measurement, this method has the advantages of being able to collect data continuously over time, easily obtaining the average image surface of the entire image surface, and not requiring time to read the exposed pattern. be. Furthermore, it is known that the change in focus position due to absorption of exposure light also occurs due to absorption of exposure light reflected from wafer 11, and that the reflectance of wafer 11 is also a factor that determines the amount of change in focus position. Therefore, when correcting the focal position while storing the fluctuation characteristics as described above, it is also necessary to know the reflectance of the wafer 11. The reflected light sensor 15 can also be used as a reflectance sensor for measuring the reflectance of the wafer 11. Further, the second embodiment can be used to investigate not only the focus variation due to the absorption of exposure light described above, but also characteristics such as atmospheric pressure change and focus change, temperature change and focus position change, etc. In addition, periodically during actual use, for example,
By loading a measurement mask and checking the focal position about once a day, it is possible to calibrate changes in focal position due to long-term drift of the device.

In all of the embodiments described above, the mask 9 and the wafer 11 have a conjugate relationship, but in the case of the first embodiment, if the pattern of the mask 9 is small, it cannot be detected with high resolution. Furthermore, the second embodiment has the disadvantage that it requires a special measurement mask. For this purpose, a measurement mask having a pattern as shown in FIG. 4 is placed on a surface conjugate to the mask 9, for example, the surface on which the reticle blind 5 is installed, and the measurement mask and wafer 11 are connected using the same principle as described above. Another possibility is to take a conjugate relationship with . This measurement mask can be inserted into and taken out of the optical path, and is used by inserting it into the optical path only during measurement. Normally, the mask 9 is used without being loaded, but if there are few patterns on the mask 9, it is possible to use the measurement mask through the transparent part of the mask 9 even if the mask 9 is loaded.
As in the first embodiment, it is possible to perform exposure while correcting the focal position using the high reflection surface 18 for each wafer. However, although the method of inserting the measurement mask at the conjugate point of the mask 9 must always keep it conjugate to the mask 9, the conjugate may shift due to subtle changes in temperature or atmospheric pressure of the device. For this reason, it is necessary to create a configuration in which conjugate shift does not occur as much as possible, and further calibrate the conjugate with the mask 9. For example, using the mask 9 with a highly reflective pattern on the entire surface, the above-mentioned mask 9 and the wafer 1
1, the conjugate position (reticle) is set so that the output of the reflected light sensor 15 is maximum.
One possible method is to move the measurement mask at the blind position in the optical axis direction.

In the above embodiment, the focal position was detected by the output of the reflected light sensor 15, but the stage 12 was detected based on the same principle.
It is also possible to detect the focal position using the upper illuminance sensor 20. In other words, the reflected light sensor 15 takes advantage of the fact that the reflected light is blocked by the pattern of the mask due to the focal position shift and does not pass upward, but the blocked light is reflected downward again and is received by the illuminance sensor 20. It can be used for focal position detection. In this case, the position where the output of the illuminance sensor 20 is minimum is the focal position.

〔Effect of the invention〕

As described above, according to the present invention, pattern baking, development,
The focal position can be detected without any measurement work, and the average optimum image plane including the curvature, inclination, etc. of the image plane over the entire exposure area can be detected in one measurement.

[Brief explanation of drawings]

FIG. 1 is a diagram showing a schematic configuration of a projection optical device according to a first embodiment of the present invention, FIG. 2 is a diagram showing a schematic configuration of a projection optical device according to a second embodiment of the present invention, FIG. 3 is a diagram showing the principle of focus position detection, and FIG. 4 is a diagram showing an example of a pattern of a dedicated mask. [Explanation of symbols of main parts] ■...Light source, 2...Illuminance distribution uniformization means, 3...
Semi-transparent mirror, 5... Reticle blind, 9... Mask, 10... Projection optical system, 11... Wafer, 12
...stage, 13 emitter, 14... receiver, 15
...Reflected light sensor, 16-...Focus controller, 17...Motor, 18...High reflective surface, 19
...Environmental sensor 20...Light sensor, 21...
・Shutter, 22...Shutter control circuit

Claims (2)

[Claims] (1) A projection optical device comprising an illumination light source that illuminates a mask on which a predetermined pattern is formed, and a projection optical system that projects an image of the pattern onto a photosensitive substrate or a light reflective substrate in a predetermined image formation state. In this step, an amount of reflected light reflected from the photosensitive substrate or the light reflective substrate and returning to the illumination light source side via the projection optical system and the mask is detected at a position substantially conjugate with the pupil of the projection optical system. and the average focal plane position of a predetermined area within the field of view of the projection optical system based on the photometric output of the photometric means obtained when the distance between the substrate and the mask is relatively changed. What is claimed is: 1. A projection optical device comprising: focal position detection means for detecting a focal position. (2) The projection optical device according to claim 1, wherein the mask has lines, ANDs, and spaces with a line width near the resolution limit of the projection optical system.