JPH0344641B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for detecting alignment marks for performing alignment before pattern printing in an apparatus for printing an actual element pattern of a mask onto a wafer.

Conventionally, this type of device was disclosed in Japanese Patent Application Laid-Open No. 48-64884, for example.
No. 49-28363, etc.
However, in the above case, since the half mirror is located near the object that has the mark, there is a risk that the light that has passed through the half mirror may be irradiated and reflected around the lens barrel, etc., and enter the object side or the photoelectric detection side as stray light. . Therefore, no improvement in mark detection accuracy can be expected. This drawback is the same as above. Furthermore, the above method cannot accommodate changes in the placement position of the alignment mark.

An object of the present invention is to solve the above-mentioned difficulties, and in order to achieve the object, an alignment mark detection method according to the present invention forms an image of a light source at the pupil position of an objective lens, and Illuminating the object surface with the light beam from the light source via a bending mirror placed near the objective lens,
The reflected light generated on the object surface is received by a filter arranged at a position conjugate with the pupil position of the objective lens via the bending mirror and the objective lens, and the alignment mark on the object surface is detected in dark field by the filter. The method comprises making the objective lens and the bending mirror movable relative to the light source section and the filter, moving the objective lens and the bending mirror according to the position of the alignment mark, and moving the objective lens and the bending mirror in the vicinity of the alignment mark. On the other hand, the positional deviation of the image with respect to the pupil position of the objective lens that occurs when the objective lens and the bending mirror are moved, and the deviation in the conjugate relationship between the pupil position of the objective lens and the filter are calculated as follows: The present invention is characterized in that the correction is performed by moving a part of the optical members arranged in the optical path between the light source section and the filter.

The figure shows an alignment mark detection device for a printing device according to an embodiment of the present invention. Usually, a plurality of alignment marks are often placed on a wafer to ensure accurate alignment, and at least two are required to constrain two degrees of freedom in the parallel direction and one degree of freedom in rotation.

In the figure, 10 is a wafer containing a semiconductor, and 1
0a and 10b are first and second alignment marks on the wafer 10, respectively. 30 is a mask, 3
0a and 30b are alignment marks similar to those on the wafer. Reference numeral 50 denotes a parallel movement table for fixing and moving the wafer 10, and is capable of linear movement in the X and Y directions and rotational movement in the R direction.
101, 102, . . . are actual circuit element patterns already printed on the wafer 10. 301,
302 are formed on the mask 10 and are actual circuit element patterns to be printed on the wafer 10 from now on.

The systems indicated by numbers 11a to 25a are the first systems for detecting alignment marks 10a and 30a.
photoelectric conversion mechanism, and an equivalent second photoelectric conversion mechanism 1
1b to 25b are provided for another alignment mark.

Hereinafter, similar parts will be explained without the symbols a and b.

19 is the microscope objective lens, 18 is the objective lens 1
This aperture is provided to coincide with the front focal position of lens 9, which is also the "entrance pupil" position of objective lens 19. 17 is a total reflection mirror, 16 is a semi-transparent mirror, 15 is a lens, 14 is a field stop, 13 is an aperture stop, and 12 is a lens.
Formed over the opening. Brightness aperture 13 is entrance pupil 1
The size of the secondary light source image formed on 8 is determined. Field stop 14 determines the area of wafer 10 that is to be illuminated. Without the field stop 14, areas outside the effective field of the microscope would be illuminated excessively, and scattered light would occur outside the effective diameter, resulting in poor precision. 15
is a lens for forming an image of the light source 11 at the front focal position of the objective lens 19, that is, at the entrance pupil position 18.

Here, instead of creating an image of the light source so as to cover the entire diameter of the entrance pupil 18 determined by the numerical aperture of the microscope objective lens 19, so-called partially coherent illumination is performed in which a light source is created much smaller than the diameter of the pupil and used for illumination. Do this. Incidentally, when the diameter of the pupil is R and the diameter of the image of the light source is r, r/R has a value in the range of 0.2 to 0.7.

20 is a relay lens, and 21 is a scanner. The scanner 21 is arranged to coincide with the imaging plane of the wafer 10 formed by the objective lens 19 and the relay lens 20. (The optical path is indicated by a broken line.) The structure of the scanner 21 itself may be either a transmission type or a reflection type, and in any case, the scanner 21 can sample photoelectric information in an arbitrary area on an object. becomes possible. A specific detection method is, for example, Japanese Patent Application Laid-open No. 18472/1983.

Numerals 22 and 24 each represent a pupil imaging lens, and the relay lens 20 and the imaging lens 22 combine to form an image of the pupil 18.
After forming an image on the power filter 23, the image is again formed on the photodetector 25 using the imaging lens 24.
(The optical path is shown by a thin solid line.) The filter 23 has a stopper in the center for removing specularly reflected light. The size of this stopper is determined by the size of the light source imaged on the pupil 18 and the combined pupil imaging magnification of the lens group. Further, the position of the photodetector 25 is the same as the position of the pupil.

With the above optical arrangement, the light from the partially coherent light source formed on the entrance pupil 18 passes through the microscope objective lens 19 and then exits with its chief ray parallel to the optical axis. Of the light reflected by the reflective wafer 10, the image of the light source by specularly reflected light matches the original light source image on the pupil 18, and the part where the image of the light source by specularly reflected light is not formed is Detection reflected light arrives.

The specularly reflected light and the detection reflected light that have passed through the pupil 18 surface pass through the relay lens 20 and the imaging lens 22 of the pupil 18, and the specularly reflected light forms an image on a stopper at the center of the filter 23 and is blocked. Light other than the specularly reflected light passes through the filter 23 and is received by the photodetector 25 via the lens 24. In this way, the alignment mark information is transmitted to the photo detector 2.
5, and also removes flare components such as specular reflection light, so information with sufficiently high contrast is obtained, and therefore interference thin film effects in the silicon dioxide layer and photoresist layer of the wafer are completely ignored. can. In addition, when using the inclined part of the wafer as an alignment mark, it is preferable that the mark has a structure including many inclined parts. In addition, the part to be detected by the photo detector 25 is
is activated and selected.

Furthermore, in this device, an objective lens 19 and an aperture 18
The total reflection mirror 17 is integral and can be moved in parallel.
It can be moved closer or further away from the semi-transparent mirror 16 depending on the placement position of the alignment mark or the size of the wafer. However, when the position of the objective lens 19 is moved, the light source (pinhole of the diaphragm 13) and the entrance pupil 18 do not necessarily satisfy the mutual relationship. In this case, optical members 12, 13, 14, 1 are arranged in the optical path between the light source 11 and the light source image on the pupil 18.
The position of lens 15 of lenses 5 and 16 is adjusted by moving it by a minute amount.

Furthermore, although the entrance pupil (diaphragm 18) and the filter 23 are mutually interlocked via the lenses 20 and 22, in this case as well, the movement of the objective lens 19 may prevent them from achieving an accurate interlinkage relationship. However, also in this case, the optical members 16, 20, 21, which are lined up in the optical path between the filter 23 and the light source image on the pupil 18,
It is assumed that the position of the lens 22 among the lenses 22 is moved to adjust the mutual relationship. Therefore, no matter where the alignment mark is placed, it is possible to form a light source image on the entrance pupil 18 of the objective lens 19, and to maintain the entrance pupil 18 and the filter 23 in a conjugate manner. Alignment marks can be detected in dark field.

A baking device with such an arrangement performs baking with the mask and wafer in contact, or the mask and wafer are placed at a very small distance of several tens of microns.
Burning is performed from a distance. Although the lighting device for printing is not shown in FIG. 1, the lighting device for printing is provided above the mask 30 as is well known.

Mechanisms designated by numerals 11 to 25 are removed from the printing optical path during printing, and are moved toward each other to positions as shown in the figure when alignment marks are detected. Incidentally, in a conventional apparatus, the mask is fixed to the main body of the printing apparatus, and the wafer is moved and aligned, and this method is also followed in this embodiment.

First, the objective lens 19, the diaphragm 18, and the mirror 17 approach each other so that they are in positions where they can see the alignment marks. The illumination light from the light source 11 illuminates the surface of the diaphragm 18 in a partially coherent manner, and further illuminates the mask alignment mark and the wafer alignment mark through the objective lens 19. The specularly reflected light, which is the principal ray of the illumination light beam vertically reflected from the peripheral surface of the alignment mark and the alignment mark surface, and the detection reflected light are blocked by the filter 23 and received and detected by the photodetector 25. A servo mechanism (not shown) operates based on the detected information representing the difference in position between the wafer and the mask, and the parallel movement table 50 is translated in the X direction and Y direction and rotated in the R direction to obtain difference information. The position of the wafer is shifted until a predetermined condition is satisfied.

Note that it is also possible to arrange a semi-transparent mirror between the lens 20 and the scanner 21 to derive the luminous flux and observe it directly with the eye.

As described above, in this embodiment, since the total reflection mirror is skillfully used, it is possible to reliably prevent stray light from entering. Further, when changing the placement position of the alignment mark, the number of movable parts for maintaining the optical characteristics of the alignment mark can be reduced as much as possible.

[Brief explanation of drawings]

The figure shows an embodiment of the present invention. 10... Wafer, 10a, 10b... Alignment mark, 11... Light source, 12, 15... Lens, 17
...Total reflection mirror, 18...Aperture, 19...Objective lens.

Claims (1)

[Scope of Claims] 1. An image of the light source is formed at the pupil position of the objective lens, and the object surface is illuminated with the light beam from the light source via the objective lens and a bending mirror disposed near the objective lens. Then, the reflected light generated on the object surface is received by a filter arranged at a position conjugate with the pupil position of the objective lens via the bending mirror and the objective lens, and the alignment mark on the object plane is detected by the filter in a dark field. A method for detecting the alignment mark by making the objective lens and the bending mirror movable relative to the light source section and the filter, and moving the objective lens and the bending mirror according to the position of the alignment mark. The positional deviation of the image with respect to the pupil position of the objective lens that occurs when the objective lens and the bending mirror are moved, and the deviation of the conjugate relationship between the pupil position of the objective lens and the filter An alignment mark detection method comprising: moving a part of an optical member lined up in an optical path between the light source section and the filter for correction.