JPH0430175B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a reduction projection exposure apparatus, and in particular to an apparatus for measuring the overlay offset of patterns with high precision and automatically correcting the overlay offset based on the results. The present invention relates to a possible reduction projection exposure apparatus.

(Prior Art) Generally, a reduction projection exposure apparatus is equipped with a pattern detection function for detecting the position of a pattern on a wafer.

FIG. 1 is a schematic perspective view showing an example of the configuration of a reduction projection exposure apparatus equipped with a pattern detector. In the figure, 1 is an XY stage, 2 is a mercury lamp,
3 is a condenser lens, 4 is a reduction projection lens, 5
is a laser length measuring device, 6 and 7 are pattern detectors, 8 is a reticle, 8A is a reticle window, 9 is a wafer, 10
11 is a servo motor, 11 is an alignment pattern, 12 is an exposure illumination optical axis, and 22 is a CPU (computing unit).

During operation, light emitted from the mercury lamp 2 and passed through the condenser lens 3 passes through the pattern of the reticle 8 through the reduction projection lens 4 and then onto the XY stage 1.
The image is projected onto the wafer 9 fixed above and exposed.

The current position of the XY stage 1 is detected by a laser length measuring device 5, and is accurately positioned by a servo motor 10 with an accuracy of approximately ±0.1 μm.

In addition, in this apparatus, in order to realize accurate overlay of the later reticle pattern on the reticle pattern of the previous layer, the alignment pattern 11 on the wafer 9 is observed through the reduction projection lens 4 and the window 8A provided in the reticle 8. Pattern detectors 6, 7 are provided.

As a result, the amount of relative deviation between the reticle 8 and the alignment pattern 11 on the wafer 9, that is, the amount of offset, is detected, and after being calculated by the arithmetic processing unit 22,
By adding correction to the target value of XY stage 1, it is possible to achieve accurate overlay of approximately ±0.2 μm.

Using a reduction projection exposure apparatus as shown in Fig. 1,
When measuring the interval between patterns on a wafer formed in the m-th layer and the n-th layer, (1) For example, the position of the XY stage 1 when the pattern detector 6 recognizes the position of the pattern in the m-th layer is measured. (2) Then, while measuring the amount of movement with a laser, the pattern position of the nth layer can be recognized.
Move the XY stage 1, detect the position of the XY stage 1 at that time in the same way, and (3) calculate the difference between the positions of the XY stage in (1) and (2) above, that is, the amount of movement of the stage 1, as the pattern interval. The following approach was taken.

(Problem to be Solved by the Invention) However, in the conventional method described above, the measured value includes an error in position measurement of the stage by the laser length measuring device 5 and an error in pattern detection twice, so 0.2μ
It was not possible to measure with an accuracy of less than m.

However, in recent LSI production, for example, 1 megabit DRAM production, higher overlay accuracy is required, and the conventional method of measuring pattern spacing on a wafer has the problem of not being able to evaluate offsets. Ta.

The purpose of the present invention is to measure with high precision the overlay offset of the n-th layer pattern with respect to the m-th layer on the wafer 9, and for the second and subsequent wafers, the detected offset amount is used to measure the overlap offset of the n-th layer pattern with respect to the m-th layer on the wafer 9. An object of the present invention is to provide a reduction projection exposure apparatus capable of automatically correcting offset by performing feedback during superimposition of objects and adding the feedback to a target value for adjusting the relative positions of a reticle and an object to be processed.

(Means for Solving the Problem) In order to achieve the above object, in the present invention, within the same image (field of view) with one scan of detection light,
The reduction projection exposure apparatus is equipped with a function to measure the distance between the m-th layer pattern and the n-th layer pattern on the wafer and find out how much offset (error) this distance has with respect to the known device design value. Furthermore, the amount of offset between the two patterns obtained as described above is memorized, and the amount of offset is fed back to the pattern matching system during subsequent pattern matching to adjust the relative position of the reticle and the workpiece. The reduction projection exposure apparatus has a function of reducing the offset by adding it to the target value (for example, the target coordinates of the XY stage).

(Function) According to the present invention, by scanning once or by simultaneously recognizing the patterns of each separate layer within the same image, it is possible to move the stage and its position in order to detect the offset of the pattern. Since no measurement is required, stage position measurement errors are not involved in pattern offset detection, allowing highly accurate offset detection.

Then, by feeding back the detected offset amount to the pattern matching system and adding it to the target value for relative position adjustment of the reticle and the workpiece,
The offset (alignment error) between each layer pattern can be made extremely small and its correction (reduction) can be easily performed.

(Embodiment) The present invention is a reduction projection exposure apparatus as shown in FIG. 1, in which a detection optical system as shown in FIG. 2 is added. In the figure, the same reference numerals as in FIG. 1 represent the same or equivalent parts.

Further, 12A is a detection optical axis, 13 is an illumination lamp for detection light, 14 is a half mirror, 15 is a reflection mirror, 16 is an m-th layer and is a pattern formed on the wafer 9, and 17 is an n-th layer, which is the same wafer 9. 18 is a position detection/slit plate driving device, 19 is a slit plate, 20 is a photomultiplier, 21 is an amplifier, 23 is an arithmetic processing unit, 24,
25 is a lens, and 26 is a memory.

The light emitted from the detection light illumination lamp 13 is reflected by the half mirror 14, enters the detection light shaft 12A, passes through the reflection mirror 15, the reticle 8, and the reduction projection lens 4, and is projected onto the wafer 9 at the m-th layer. The formed pattern 16 and the nth layer (where n≠m)
The pattern 17 formed on the same wafer 9 is illuminated at the same time.

The reflected light from the two patterns 16 and 17 travels backward through the reduction projection lens 4 and the reticle 8. Then, it is further reflected by the reflection mirror 15 and proceeds along the detection optical axis 12A, and images of the two patterns 16 and 17 are formed on the slit plate 19 by the lens 24.

A minute slit is bored in the slit plate 19, and the light passing therethrough is focused by a lens 25 onto the light receiving surface of the photomultiplier 20. The light incident on the photomultiplier 20 is converted into an electrical signal according to the amount of light, and the amplifier 21
The data is supplied to the arithmetic processing unit 23 via.

On the other hand, the position detection/slit plate driving device 18 is
By suitable means (not shown), the slit plate 1 is
9 is driven in the vertical direction in the figure within that plane. At that time, the amount of movement is input to the arithmetic processing device 23.

Therefore, by driving the slit plate 19, for example, first, after aligning the slits of the slit plate 19 with the image of the pattern 16 formed in the m-th layer, the slit plate 19 is further driven to match the image of the pattern 16 formed in the m-th layer. When the slits of the slit plate 19 are aligned with the image of the pattern 17 formed by the layers, as shown in FIG.
A signal output that changes over time according to the movement of the slit plate 19 is obtained from the amplifier 21.

The arithmetic processing unit 23 calculates the images of the two patterns 16 and 17 based on the input signal from the amplifier 21 and the signal input from the position detection/slit plate driving device 18 using a method that will be described in detail later. Calculate each center and the distance K between them. Then, data α of the distance between the two patterns 16 and 17, which was planned in the device design, is compared with the detected value K.

As described above, it can be seen in which direction and how much offset the n-th layer, which has been superimposed on the m-th layer, has.

FIG. 3 shows a flowchart of the arithmetic processing device. That is, the position signal input from the position detection/slit plate driving device 18 and the amplifier 21
The waveform of the data f(x) shown in the figure is formed by the light amount detection signal input from the
S1). The waveform f(x) is passed through a smoothing circuit to form a waveform f(A) (step S2). Furthermore, the waveform f
Pass (A) through a differential circuit to find its inflection point (steps S3, S4, S5).

In this case, the inflection points of the waveform of the pattern 16 formed in the m-th layer are x3 and x4, and the inflection points of the waveform of the pattern 17 formed in the n-th layer are x1, x2
Assume that

Next, calculate (x3+x4)/2 and (x1+x2)/2 to find XR and XL (step S6,
7).

The values XL and XR correspond to the respective patterns 17,
It can be seen as 16 position coordinates, so XR−
If XL=K is determined (step S8), this becomes the interval between patterns 16 and 17. Therefore, the value obtained by subtracting the pattern interval α planned in the device design from K, that is, K-α=β, is the pattern 16.
and 17 superimposed offsets (step
S9).

The detection result β obtained as described above is stored in the memory 26 (step S10), and the next time overlay exposure is performed, this data is fed back to the pattern alignment system as the overlay correction value β, and is automatically Add correction to the target coordinates of the XY stage. In this way, subsequent pattern overlay will be performed accurately and no offset will occur unless other conditions or environments change.

FIG. 4 shows details of the arithmetic processing unit 23 used to realize the above-described flowchart. In the figure, 23a is a smoothing processing circuit, 23b is a differentiation circuit, 23c is an inflection point search circuit, and 23d is a symmetry calculation circuit. Its operation is clear from the above description with respect to FIG. 3, and therefore will not be described here.

In addition, in the above explanation, the detection system for measuring the offset amount (reflection mirror 15, position detection device 1
8, photomultiplier 20, etc.), but as can be easily seen from the functional comparison, it is also possible to replace the offset detection system with pattern detectors 6, 7, etc. shown in Fig. 1. Good too. Specifically, the slits placed in front of the pattern detectors 6 and 7 may be moved in the same manner as described for the slit 19 in FIG. 2, and the distance of movement may be measured.

In addition, in the above explanation, the case where the detector is a photomultiplier was used as an example, but the photomultiplier is a one-dimensionally arranged photoelectric conversion array (for example,
It can be replaced with a semiconductor photo array) or a video camera. In this case, the slit 19, the position detection device, and the drive device 18 can be omitted.

Furthermore, although the so-called through-the-lens method in which the detection lights of the patterns 16 and 17 pass through the reduction projection lens 4 has been described above, the present invention basically allows It has a function that can detect the interval between two patterns on the wafer, and if necessary, this detected value can be used for the next reticle overlay, and the next overlay detected value, that is, the target coordinates of the stage. As long as the reduction projection exposure apparatus has a function that allows automatic correction, it can be applied regardless of the configuration of the detection system.

Therefore, the present invention is also applicable to a so-called off-axis detection method in which detection (light reflected from a pattern) does not pass through the exposure illumination optical axis 12.

(Effects of the Invention) As explained in detail above, according to the present invention, the distance between two patterns can be measured while the XY stage is fixed, making it easy to measure the distance between patterns with high precision of about 0.1 μm. Can be done. That is,
The distance between the pattern formed in the m-th layer and the pattern formed in the n-th layer can be measured with the same degree of accuracy. Therefore, the overlay offset between the mth layer and the nth layer can be measured with the required accuracy.

The offset data measured accurately in this way is fed back to the positioning system of the XY stage, that is, the means for adjusting the relative positions of the reticle and the workpiece, during the next overlapping exposure of the second and subsequent wafers. target value (e.g.
Since the target coordinates of the XY stage) are added, the target value of the relative position adjustment means (target coordinates of the XY stage) is automatically corrected, and it is possible to sufficiently reduce the overlay offset.

[Brief explanation of drawings]

FIG. 1 is a schematic perspective view of a reduction projection exposure apparatus suitable for applying the present invention, FIG. 2 is a schematic configuration diagram of main parts showing an embodiment of the present invention, and FIG. 3 is a schematic perspective view of a reduction projection exposure apparatus suitable for applying the present invention. FIG. 4 is a block diagram of an arithmetic processing device used in an embodiment of the present invention, and FIG. 5 is a waveform diagram obtained corresponding to two patterns. . 1...XY stage, 4...reduction projection lens,
8... Reticle, 9... Wafer, 16... Pattern formed in the m-th layer, 18... Position detection/
Slit plate drive device, 19...Slit plate, 20
...Photomultiplier, 23...Arithmetic processing unit, 26...Memory.

Claims (1)

[Claims] 1. An XY stage for fixing a workpiece thereon, means for driving the XY stage on the XY plane, means for detecting the position of the XY stage, and images of a plurality of reticles. sequentially, adjusting the relative positions of the reticle and the object to be processed, and the exposure means that performs exposure by reducing the projection onto the object to be processed, so that the subsequent reticle pattern is superimposed on the reticle pattern of the previous layer. a reduction projection exposure apparatus having a reticle pattern overlay means for simultaneously forming optical images of two patterns formed on a workpiece by different reticle patterns within the same field of view; means for converting optical images of the two patterns into electrical signals; means for measuring the spacing between the two patterns based on the shape of the obtained electrical signals; and converting the obtained measured value of the pattern spacing into its design value. means for calculating an offset amount by comparing with the reticle pattern; and feeding back the offset amount to the reticle pattern superimposing means and adding it to the target value for the relative position adjustment, so that the subsequent reticle pattern is adjusted to the reticle pattern of the previous layer. 1. A reduction projection exposure apparatus comprising means for correcting overlay offset.