KR20010088997A - Superposition Measurement Method in Semiconductor Process - Google Patents

Abstract

PURPOSE: A method for measuring overlap status during semiconductor processes is provided to be capable of excluding measurement error caused due to deformed patterns by improving a prior method that simply designated only the centers of overlap patterns, and be capable of minimizing the deformation due to processes as well as saving costs and time taken to design overlap patterns of new style. CONSTITUTION: First, signal areas of images to be used are set and positions representing respective setting areas are set in outer bar patterns so that the center points of both edges are obtained under the state that both edge images of an outer bar are in a parallel movement relation. Then, the distance(D) between setting areas is calculated.

Description

Method of measuring overlap in semiconductor process {omitted}

The present invention relates to a method for measuring the degree of overlap in the lithography process of the semiconductor manufacturing process, and more particularly, even if the pattern for measuring the degree of overlap is modified due to the manufacturing process can be excluded the effect of the deformation The invention of the measurement algorithm.

In the semiconductor manufacturing process, the degree of overlap refers to the degree of overlap between a layer formed by a previous photo process and an ongoing photo process. In order to measure the degree of overlap using a mechanical device, the degree of overlap is formed by using a pattern of overlapping shapes such as a box-in-box and a bar-in-bar.

The overlapping measurement using the pattern is performed by measuring the relative positions of the outer box and the inner box formed in different layers as shown in FIG. 1. (A) is a cross-sectional view of a general overlapping pattern, and (B) is a plan view corresponding thereto.

For example, the inner box (①) is formed inside the outer box (②) already formed in the previous layer, and the relative position of the inner box and the outer box is measured by using the overlapping degree.

In the degree of overlap measurement equipment, the measurement of the degree of overlap is generally performed in the following order.

(1) After recognizing the edges of the inner box and the outer box, determine the center of the recognized edge. In this case, signal processing (③) of the brightness on the image at the upper, lower, left and right ends of the inner box and the outer box is generally performed through mathematical processing as possible by software, and finally the thickening point of both edges is determined.

(2) Calculate the relative distance between the centers of the inner box and the outer box and use this as the overlap between the two layers. In FIG. 2, Cin (x), Cin (y), Cout (x), and Cout (y) denote center coordinates (x direction and y direction) of the inner box and center coordinates (x direction and y direction) of the outer box, respectively. represents, Δ x, Δ Y denotes a distance (x direction, y direction) between each of the inner box and the outer box center. Finally overlap value is Δ X, Δ Y values.

In existing superimposition measuring equipment, when centering the inner box and the outer box, the basic premise is that the two signal shapes on both sides are symmetrically mirrored.

However, the outer box is deformed during the process (where deformation refers to physical deformation, or simply visually deformable), and assumes that the signal shape at the left and right sides (or top and bottom) is symmetrical to the mirror surface. Will not fit into. 3A illustrates a case in which a metal film is deposited on the outer box (4), and the deposition state of the metal film is different from the left and right sides of the outer box. In the case where the metal film is deposited with a gentle slope (6), the reflected light does not return, so that it appears dark. On the other hand, when the metal deposition becomes horizontal (⑤), the brightness on the plan view appears bright. 3 (c) shows brightness signals on the image of the outer box in a direction from ⓐ to ⓑ. In this case, since the left and right signals are not in the mirror plane symmetry relationship with respect to any point ⓒ, the center point obtained under the assumption that the mirror plane is symmetrical is very errored, and thus a reliable overlapping value cannot be obtained.

In the box-in-box overlap pattern, the box-in-bar overlap pattern may be used because it is very difficult to correct deformation caused by the process. However, even in the case of a box-in-bar pattern, it is impossible to completely eliminate the measurement error due to deformation. FIG. 4 shows a state (a) and a state (a) of a signal on an image in which an outer bar of a box in bar pattern is deformed by metal deposition.

In addition to the metal deposition process, the superimposition pattern is also deformed by a process such as CMP (chemical & mechanical polishing) process or etch-back. 5 shows an actual image of an outer bar deformed by position on a wafer. It can be seen that the overlapping patterns at the left end (a), the center (b), and the right end (c) in the wafer are different from each other in the shape of the pattern bar, especially the outer bar.

The pattern for measuring the degree of overlap has a size of approximately ˜20 × 20 μm ² and deformation of the pattern occurs by various processes (metal deposition, CMP, etc.). Processes other than lithography are mainly wafer-sized [~ 200 mm] (8), so deformations are collectively shaped throughout the overlap pattern. That is, in one overlapping pattern, the left and right (or top and bottom) edges are deformed in the same direction and about the same degree (⑦). In this case, the assumption that the deformation of the left and right edges is symmetrical to the mirror surface does not fit, and the right edge form has a form that can be obtained by simply moving the left edge in parallel.

In conclusion, in finding the center of the outer bar (or inner bar), the center point of the pattern deformed by the process is more simply assuming that the images of both left and right (or top and bottom) edges are in a parallel relationship. That's a reasonable way. 7 illustrates a cross-sectional view (a) and an image signal (b) of the deformed outer pattern, (c) illustrates a case where the center of the outer pattern is obtained assuming that both image signals are symmetrical to the mirror surface. D) describes a case where the center is obtained assuming that both image signals are in a parallel movement relationship with each other. The case (D) is more reliable in the process of finding the center point than the case (C).

1 is a cross-sectional view and a plan view of a general overlapping measurement pattern

2 is a method of calculating the degree of overlap

3 is a cross-sectional view and plan view of the overlapped measurement pattern modified by the manufacturing process and a graph of brightness intensity of the overlapping pattern

4 is a cross-sectional view and a plan view of the overlapping pattern in the form of a box-in-bar and the overlapping pattern brightness intensity graph

5 is an actual cross sectional view of a modified overlap pattern.

6 is a modified embodiment of the characteristics and overlap pattern of the manufacturing process

7 shows methods for determining a center in a modified overlap pattern.

8 is an example of a method of determining the center of a superimposition pattern under a translational relationship;

<Explanation of symbols for main parts of drawing>

①: inner box (mainly a photoresist and part of the overlapping pattern formed in the current photo process)

②: outer box (part of the overlapping pattern formed in the previous process)

③: brightness signal on the image of the overlapping pattern

④: metal film

⑤: the part where the metal film is deposited horizontally (bright part)

⑥: Metal film deposited obliquely (dark area)

⑦: outer bar

⑧: Shape of characteristic distribution of process at wafer size level

8 illustrates a method of measuring overlapping according to the present invention.

8A shows a plan view of the deformed outer bar, and FIG. 8B shows an image signal of the outer bar.

Assuming that both edge images of the outer bar are parallel to each other, the following two steps must be taken to find the center point of both edges.

(1) The setting of the signal area on the image to be used in the outer bar pattern and the position setting that will represent each set area.

8 (c) shows an example of setting the image signal area and setting the center value of the areas. Set areas at regular intervals (for example, 1µm) on specific areas of the image signal (signal peaks in the figure). When the image area near both edges of the outer bar is set, the center of the set area can be naturally determined (A, B), respectively.

(2) Calculate the distance (D) between the set areas.

8 (d) shows a method for obtaining the distance between the set regions. Moving the entire area set on the left in parallel to the right, when the left and right image signals are the best fit (when the matching, often the correlation is well done), the movement distance of the area at that time (D ), And considering the moving distance D and the center A of the area to be moved, the center of the outer bar can be obtained simply. In other words, the center C of the outer bar is represented by C = A + D / 2.

By improving the method of centering the overlap pattern simply by improving the method of centering the overlap pattern, the measurement error caused by the deformed pattern can be eliminated, and a new type of overlap pattern is designed to minimize the deformation caused by the process. This can save you money and time.

Claims (1)

In the superimposition measurement using a superimposition pattern in which square or bar-shaped patterns are arranged in a square, from the left (or top) right (or bottom) pattern when centering an outer pattern or an inner pattern With respect to the generated primary image signal or mathematically processed signal, the degree of overlap between the two patterns is determined by assuming that all or some regions of one image signal are parallel to each other with respect to the other image signal. If you calculate.