JP2000294489A - Pattern overlapping method and aligner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform accurate pattern overlapping by a method wherein a difference between overlapping measuring data at points as reference image heights and points excluding the points as the reference image heights in a predetermined region is acquired, the points exceeding the threshold are eliminated, and a replacement is made to values to which linear components are annexed. SOLUTION: A resist coating is performed with a coat unit on a semiconductor wafer surface as a processing body (S1). Next, the semiconductor wafer is carried into an algner, and measuring is made on overlapping between patterns formed therein as an underlayer in the semiconductor wafer and patterns to be patterned from now on (S2). When this overlapping is measured, overlapping correction data are calculated (S3). Thus, values seemed as abnormal values are removed to enhance measuring precision. Next, an exposing processing is made for such the overlapped semiconductor wafers (S4). Continuously, a developing processing is made for the exposed semiconductor wafers (S5), and further the overlapping measure is made (S6). Here also, abnormal values are removed to enhance measuring precision.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for pattern superposition, and more particularly to a method for pattern superposition in an exposure step in a semiconductor manufacturing process.

[0002]

2. Description of the Related Art In recent years, with the increasing integration of semiconductor devices, reduction projection exposure techniques have been increasingly miniaturized. Therefore, the conditions for the pattern superimposition technique in the semiconductor manufacturing process, that is, the superimposition margin (position deviation tolerance) has become so severe that more than the ability of the apparatus is required. Here, the pattern superimposition is performed in the photolithography process to eliminate a lateral (horizontal) positional deviation between a pattern to be currently exposed and a target pattern as a base.

[0003] Conventionally, for example, when a measurement mark for overlay has become rough due to a trouble occurring in a previous process, or a foreign matter has adhered to the measurement mark, measurement is performed using the measurement mark and the overlay is performed. When the alignment is performed, the overlay accuracy is reduced. For the purpose of avoiding this, a data operation algorithm for removing such an abnormal value is used.

[0004]

However, the conventional outlier elimination algorithm is very simple,
Judgment for removing an abnormal value is made based on whether or not the deviation amount from the average value of the overlay measurement data exceeds the judgment reference value.

In such an algorithm, the pattern may actually be displaced during the superposition. Therefore, in order to accurately remove the abnormal value, it is necessary to set the judgment reference value to a considerably large value. The functions cannot be used effectively.

[0006] The present invention has been made in view of the above points, and an object of the present invention is to provide a pattern superimposing method capable of accurately performing pattern superimposition.

[0007]

Means for Solving the Problems In order to solve the above problems, the present invention has taken the following means. The present invention is a method of superimposing a pattern on the pattern of the pattern-formed object, after removing linear components from height data of a plurality of points in a plurality of predetermined regions in the object, A step of setting a point serving as a reference image height among the plurality of points, and overlay measurement data of points other than the reference image height in the predetermined area and the overlay measurement data of the point serving as the reference image height. Determining a difference between the predetermined regions, performing a threshold determination on the variation of the difference between the predetermined regions, and removing overlay measurement data at points exceeding the threshold; and removing the removed overlay. Measured data
Replacing the average value of the image height in the predetermined area with a value obtained by adding a linear component to the average value.

According to this configuration, the height data considered to be an abnormal value is removed and replaced with a value estimated as a true value.
Accurate overlay measurement can be performed, and pattern overlay can be performed with high accuracy.

In the pattern superposition method according to the present invention, it is preferable that the reference image height is height data of a point at which a specific point in the predetermined area has a smallest variation in the object.

In the pattern superimposing method according to the present invention, the threshold value in the threshold value judgment is determined based on the superimposition measurement data of the point serving as the reference image height and the reference image height in the plurality of predetermined regions. It is preferable that the difference is a predetermined amount from the average value of each difference between the overlay measurement data of points other than the point and the difference, and it is preferable that the difference can be changed depending on the processing content for the object to be processed.

The present invention is also an exposure apparatus for performing exposure and development in forming a pattern on an object to be processed, comprising: an overlay measuring means for performing measurement when the pattern is superimposed on the object to be processed; Exposure means for exposing the object after superimposition, and developing means for performing development processing on the object after exposure, wherein the superposition measurement means comprises: A reference image height setting unit configured to set a point that becomes a reference image height among the plurality of points after removing a linear component from height data of a plurality of points in a plurality of predetermined regions in the processing body; A difference between the overlay measurement data of points and the overlay measurement data of points other than the point at which the reference image height is obtained in the predetermined area is determined. Go Unnecessary data removing means for removing overlay measurement data at a point exceeding the threshold value; and replacing the removed overlay measurement data with a value obtained by adding a linear component to an average value of image heights in the predetermined area. And a data replacement unit.

According to this structure, the overlay measurement can be performed accurately, and the pattern can be superimposed with high accuracy.
Development can be performed.

[0013]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings. (Embodiment 1) FIG. 1 is a flowchart showing an exposure step using a pattern superposition method according to an embodiment of the present invention. FIGS. 2A and 2B are diagrams for explaining an algorithm of a pattern superposition method according to an embodiment of the present invention, wherein FIG. 2A is a plan view showing measurement points, and FIG. FIG.

As shown in FIG. 1, in the exposure step according to the present embodiment, first, a resist coating is performed on a surface of a semiconductor wafer which is an object to be processed by a coating unit (S1). Next, the semiconductor wafer is carried into an exposure apparatus, where overlay measurement between a pattern formed as a base on the semiconductor wafer and a pattern to be patterned is performed (S2).

In this overlay measurement, overlay correction data is calculated (S3). As a result, a value considered to be an abnormal value is removed to improve measurement accuracy. Then
An exposure process is performed on the semiconductor wafer on which the overlay process has been performed (S4).

Next, the exposed semiconductor wafer is developed using a predetermined developing solution (S5). Further, overlay accuracy measurement is performed on the developed semiconductor wafer (S6). Also in this case, the value considered to be an abnormal value is removed to improve the measurement accuracy.

In the pattern superposition method of the present invention performed in S3 and / or S6, first, a reference image height is obtained as processing 1. That is, in FIG. 2A, when the overlay measurement of a plurality of points is performed within a predetermined area (one shot) 1 on the semiconductor wafer W, a linear component is first removed from the obtained measurement data. Therefore, only non-linear components are left as residual components. Next, the variation in the image height of each point in the predetermined area 1 within the wafer surface is obtained. For example, with respect to the measurement point UR shown in FIG. In this way, the measurement points UL, CE, LL, L
The variation is determined for R. Then, the image height at the measurement point corresponding to the image height with the smallest variation is set as the reference image height. Here, CE is a reference image height.

In processing 2, abnormal values are removed. That is, when the overlay measurement of a plurality of points is performed in the predetermined area 1, the overlay measurement data (C) of the measurement point is set based on the reference image height obtained in the processing 1 in the predetermined area.
E) is subtracted from other measurement data (UL, CE, LL, LR). A certain criterion is set for the variation of the superimposed measurement data of each image height in the predetermined region obtained by the subtraction between the predetermined regions, and the measurement data exceeding the criterion is excluded (threshold value judgment). .

The criterion value (threshold value) is the amount of deviation from the average value obtained as shown in the following equation (1). This threshold value can be appropriately changed and set in accordance with the processing content for the object to be processed. This makes it possible to perform adaptive measurement control according to the processing content.

(Equation 1)

In process 3, the abnormal value is replaced with a calculated value. That is, the value removed in the process 2 is replaced with a value obtained by adding a linear component to the average value of image heights in the same shot. Note that the calculated value refers to a true value obtained by the calculation and a predicted value.

After removing the abnormal values in this way,
(Average value, wafer scaling, wafer rotation, wafer orthogonality, shot magnification, shot rotation, etc.)
By performing the exposure by feeding back the data of
It was confirmed that the overlay accuracy was clearly improved.

In the conventional method, the criterion value for removing an abnormal value is set in units of several μm, and the abnormal value removal did not function well. However, by applying the present algorithm, the abnormal value was removed. The removal can be performed with high precision, and a judgment reference value of about 0.03 μm in the process around the GATE and about 0.08 μm in the wiring process can be sufficiently dealt with. it can.

(Embodiment 2) Data obtained by overlay accuracy measurement of a certain product is subjected to abnormal value elimination and replaced with an ideal value, and then the overlay accuracy summary, for example, maximum value, minimum value, average value , 3σ, wafer scaling, wafer rotation, wafer orthogonality, shot magnification, shot rotation, and the like.

As a result, it is possible to more accurately judge the quality of the product in the overlay accuracy, and the value fed back to the exposure apparatus becomes closer to the true value, thereby improving the overlay accuracy in the lithography process.

The present invention is not limited to the first and second embodiments, but can be implemented with various modifications. Also,
Resist coating in the above embodiment, exposure,
There are no particular restrictions on the materials and conditions used in the development.

[0026]

As described above, the pattern superposition method of the present invention can perform the superposition measurement by removing the abnormal value, so that the pattern superposition can be performed accurately.

[Brief description of the drawings]

FIG. 1 is a flowchart showing an exposure step using a pattern superposition method according to one embodiment of the present invention.

FIG. 2 is a diagram for explaining an algorithm of a pattern overlapping method according to one embodiment of the present invention;
(A) is a plan view showing measurement points, and (b) is an enlarged view of (a).

[Explanation of symbols]

1: Measurement shot, UL, UR, CE, LL, LR: Overlay measurement point.

Claims (5)

[Claims] 1. A method of superimposing a pattern on the pattern of a pattern-formed object, comprising removing linear components from height data of a plurality of points in a plurality of predetermined regions on the object. Setting a point serving as a reference image height among the plurality of points; and overlay measurement data of points other than the reference image height in the predetermined area and overlay measurement data of the point serving as the reference image height. Determining a difference between the predetermined area and the threshold, and performing a threshold determination on the variation of the difference between the predetermined areas to remove overlay measurement data at points exceeding the threshold. Replacing the alignment measurement data with a value obtained by adding a linear component to the average value of the image heights in the predetermined area. 2. The pattern superimposing method according to claim 1, wherein the reference image height is height data of a point at which a specific point in the predetermined area has a smallest variation in the object. 3. The threshold value in the threshold value determination,
A predetermined amount of deviation from the average value of each difference between the overlay measurement data of the point serving as the reference image height and the overlay measurement data of points other than the point serving as the reference image height in the plurality of predetermined regions. The pattern superimposing method according to claim 1, wherein 4. A threshold value in the threshold value determination,
2. The pattern superposition method according to claim 1, wherein the pattern can be changed according to the processing content of the object. 5. An exposure apparatus for performing exposure and development in forming a pattern on an object to be processed, comprising: overlay measurement means for performing measurement when performing pattern overlay on the object to be processed; Exposure means for exposing the object after exposure, developing means for performing development processing on the object after exposure,
Wherein the overlay measurement means sets a point which becomes a reference image height among the plurality of points after removing a linear component from height data of a plurality of points in a plurality of predetermined regions in the object to be processed. Reference image height setting means, Find the difference between the overlay measurement data of the point that becomes the reference image height and the overlay measurement data of points other than the point that becomes the reference image height in the predetermined area, Unnecessary data removing means for performing a threshold determination on the variation between the predetermined regions and removing overlay measurement data at points exceeding the threshold; And a data replacement unit for replacing the average value of image heights in the image data with a value obtained by adding a linear component.