JP2002303588A - Pattern defect inspection device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a pattern defect inspection device reducing pseudo defect detection due to uneven colors and capable of improving sensitiveness of actual defect detection. SOLUTION: The pattern defect inspection device is provided with an analog- to-digital converter 2 detecting a gray level image signal of a pattern under inspection and converting it into a digital image signal, a pixel unit positioning part 11 calculating a density difference while shifting one of two images to be compared in pixel units, a local comparing part 20 comparing signals acquired from minutely delimited local adjacent portions, a data converting part 21 outputting 0 when there is no signal difference, and a comparing part 14 comparing signals of all right and left patterns based on outputs of the data converting part 21.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a defect inspection apparatus for a pattern to be inspected on a semiconductor wafer, a liquid crystal display or the like.

[0002]

2. Description of the Related Art In the manufacturing process of a semiconductor wafer process, defects are generated on a wafer surface due to particles of a manufacturing apparatus or process abnormalities. Defect inspection is performed in several steps in the manufacturing process for the purpose of detecting this defect at an early stage, specifying the generation process, and promoting the improvement of the abnormality generation process device. As this type of pattern defect inspection apparatus, an apparatus based on an image comparison method is generally known. In a defect inspection apparatus known as an image comparison method, a repeated portion where the same pattern is formed between semiconductor devices (chips) is defined as one region, and a signal at the same position between the regions is transmitted by about 0.5 μm. It is captured as an image signal in the minimum unit, and when the signal is different from the same pattern portion on the left and right, it is recognized that a defect exists at that location.

The principle of such a defect inspection apparatus using the image comparison method will be described with reference to FIGS. As shown in FIG. 3, a part where the same pattern such as a product chip formed on the semiconductor wafer 4 is formed is defined as one region (chip).
In general, image signals at the same position in similar regions existing on the left and right sides are compared, and if the value exceeds an arbitrarily set value, it is recognized that a defect has occurred. Since it is not possible to determine which defect is present only by the signal difference, a method is generally used in which the left and right areas A and C are compared, and a defect is present when a signal difference occurs in both areas.

[0004] While scanning the wafer 4 on the image signal, the image signal is captured in a minutely divided range 15 as shown in FIG. For example, first, the image signal of A is captured and stored in the memory, and then the image signal of B is captured. Each image signal is divided into the smallest units 1
5 is converted into a numerical value as a brightness signal, and compared with the values at the same position on the left and right (signal comparison 1 and signal comparison 2). If the value exceeds an arbitrarily set value, it is stored as a candidate for a defect occurrence position. . In the comparison between B and C, the values are also compared with the values at the same position on the left and right (signal comparison 3 and signal comparison 4). It is determined that there is.

[0005] If the color of the pattern and the surface of each area ABC are exactly the same, no signal difference occurs except for the portion where the defect occurs, and the defect can be detected with high sensitivity.
Actually, as shown in FIG. 5, on the surface of the wafer 4, a signal difference 25 occurs even if no defect occurs due to a difference in uniformity at the time of forming an insulating film and a uniformity of pattern processing. In some cases, the color unevenness is detected as a pseudo defect with the threshold value for detecting.

In a film forming process and a CMP process,
Due to the problem of uniformity within the wafer surface, a difference in film thickness occurs between chips. In the case of an insulating film or the like, the color difference on the surface is slightly different between regions due to the film thickness difference.
The inspection result may determine that this is a defect. Since this difference in color does not affect the semiconductor device, it is not necessary to recognize the defect as a defect, and such a defect is usually called a pseudo defect. Since the number of defects (actual defects) that may cause a product to be defective cannot be determined, it is necessary to lower the detection sensitivity to a level at which no pseudo defects occur.

As a technique for improving the above, a technique for performing brightness correction before comparing captured image data has been developed. FIG. 6 shows a conventional defect inspection apparatus for a pattern to be inspected (see JP-A-11-132959). In this example, the pattern to be inspected is a circuit pattern formed on the semiconductor wafer.

In FIG. 6, reference numeral 1 denotes an image sensor which outputs a gray-scale image signal corresponding to the brightness of light reflected from the semiconductor wafer 4 as a pattern to be inspected, that is, a gray scale. Reference numeral 2 denotes an A / D converter for converting a gray image signal obtained from the image sensor 1 into a digital image signal 9, 3 a delay memory for delaying the gray image signal, 4 a semiconductor wafer having a pattern to be inspected, and 5 an inspected circuit. The X direction, the Y direction, and the Z direction on which the semiconductor wafer 4 of the pattern is placed.
The stage moves in the direction and the θ direction (rotation), 6 is an objective lens for the semiconductor wafer 4, 7 is an illumination light source for illuminating the semiconductor wafer 4 of the pattern to be inspected, 8 is a semiconductor wafer that reflects illumination light and passes through the objective lens 6. 4 is a half mirror that irradiates the semiconductor wafer 4 and transmits reflected light from the semiconductor wafer 4.

Reference numeral 9 denotes a digital image signal from the area C in which the grayscale image signal has been converted by the A / D converter. In this manner, the illumination light from the illumination light source 7 is reflected, and the semiconductor wafer 4 is subjected to, for example, bright field illumination through the objective lens 6. Reference numeral 3 denotes a delay memory for storing and delaying the image signal 9 for one or more repeated cell pitches.

Reference numeral 11 denotes a pixel-based positioning unit for positioning the digital image signal 9 and the digital image signal 10 from the delayed B region, and detects a positional displacement amount at which the shading difference becomes minimum in pixel units. Then, one image is shifted based on the positional shift amount to align the two images. Here, the image is continuously detected by the image sensor, but the image is divided into, for example, 256 lines, and alignment is performed in this unit. 12
Is a brightness conversion unit that converts image signals having different brightnesses so as to match the brightnesses of the two image signals. Here, we ’ll perform a batch filtering operation on the entire image,
Brightness is matched.

Reference numeral 13 denotes a gradation conversion unit for converting the gradations of the image signals having different brightnesses so that the brightnesses of the image signals match each other. Here, the brightness is matched by performing a linear conversion with a gain and an offset for each pixel. The obtained image signals are compared in the comparing unit 14 and the mismatch is detected as a defect. In addition, 1
Reference numeral 6 denotes a comparison result input unit for the A region and the B region, and 17 denotes a test result determination unit for the B region.

Next, the operation will be described. The stage 5 is formed on the semiconductor wafer 4 by the image sensor 1 while scanning the stage 5 in the X direction with the illumination light converged by the objective lens 6 and moving the target area of the semiconductor wafer 4 of the pattern to be inspected at a constant speed. The brightness information (shade image signal) of the inspected pattern is detected.

When the movement for one row is completed, the apparatus moves to the next row at a high speed in the Y direction to perform positioning. That is, the inspection is performed by repeating the constant speed movement and the high speed movement. The A / D converter 2 converts the output (shade image signal) of the image sensor 1 into a digital image signal 9. Image registration in pixel units calculates the grayscale difference (difference between the value of each pixel of the image and the value of the corresponding pixel) while shifting one of the two images to be compared in pixel units, and determines that the grayscale difference is the minimum. In this case, the following positional deviation amount is obtained. The range for detecting the positional shift of the image is, for example, a maximum of ± 3 pixels, and is variable according to the pattern design rule. By shifting one image position by the obtained positional shift amount, the two images are aligned.

[0014]

However, if the brightness correction is performed as described above, the brightness of the real defect is also corrected, and in some cases, the signal of the real defect is weakened.
There has been a problem that detection may be difficult. The present invention has been made in order to solve the above-described problems, and an object of the present invention is to provide a pattern defect inspection apparatus that can reduce false defect detection due to color unevenness and improve the sensitivity of actual defect detection. I do.

[0015]

To achieve the above object, the present invention provides an A / D converter for detecting a gray image signal of a pattern to be inspected and converting it into a digital image signal, and two images to be compared. There is no signal difference at this time between pixel unit positioning means for calculating the light and shade difference while shifting one of the pixels in pixel units, and local comparing means for comparing signals obtained from a minutely divided local adjacent part. A data converter for outputting 0 at the time, and a comparator for comparing signals of all left and right patterns based on the output of the data converter are provided.

[0016]

Embodiments of the present invention will be described below with reference to the drawings. According to the present invention, color unevenness and surface roughness information during film formation are deleted by the following means to improve the detection sensitivity of actual defects. FIG. 1 shows a defect inspection apparatus for a pattern to be inspected according to the present invention. The same members or those having the same functions as those in FIG. 6 are denoted by the same reference numerals.

As shown in FIG. 1, according to the present invention, a pixel unit positioning unit 11 for positioning the digital image signal 9 and the delayed digital image signal 10 and a comparing unit 14 for comparing the image signals. , A local comparison unit 20 and a data conversion unit 21. Next, the operation will be described. The stage 5 is formed on the semiconductor wafer 4 by the image sensor 1 while scanning the stage 5 in the X direction with the illumination light converged by the objective lens 6 and moving the target region of the semiconductor wafer 4 of the pattern to be inspected at a constant speed. The brightness information (shade image signal) of the inspected pattern is detected.

When the movement for one row is completed, the head is moved to the next row at a high speed in the Y direction and positioned. That is, the inspection is performed by repeating the constant speed movement and the high speed movement. The A / D converter 2 converts the output (shade image signal) of the image sensor 1 into a digital image signal 9. The image registration in pixel units is to calculate the grayscale difference while shifting one of the two images to be compared in pixel units, and to obtain a positional shift amount that minimizes the grayscale difference. Up to this point, it is the same as the conventional one.

In the embodiment of the present invention, as shown in FIG. 2, in a certain region, for example, a region B, a signal from, for example, an adjacent local in a minutely divided range (local) 15 is compared with a local comparing unit 20. Will be compared. The data is converted into digital data by the data conversion unit 21, and 0 is output to the comparison unit 14 when there is no change and 1 is output when there is a change. As a result, a signal in a portion where a signal difference is likely to occur, such as an edge portion of the pattern defect 23, is emphasized. Reference numeral 17 denotes an inspection result determination unit for the B region, but the same applies to other regions.

By such means, it is possible to delete information on color unevenness and surface roughness at the time of film formation, thereby improving the detection sensitivity of actual defects. As described above, in the present embodiment, when a defect occurs, an abnormal part or a pattern such as a foreign substance or a pattern usually occurring in an empty space or a pattern not being formed in a portion where a pattern is to be formed. Paying attention to the fact that there is a local change in either the normal part, compare the image signal in the minimum unit time when capturing the image signal in one comparison target area, and if there is no constant signal difference, By setting all to 0, it is possible to reduce false defects due to color unevenness and improve the sensitivity of actual defect detection.

[0021]

As described in detail above, according to the present invention, local data comparison is performed in advance in the comparison area, and a signal having little change is set to a signal of 0.
By emphasizing signals in areas where signal differences are likely to occur, such as edges of patterns and defects, and then comparing the areas, pseudo defects due to color unevenness and surface roughness information during film formation are reduced. Defect detection sensitivity can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a pattern defect inspection apparatus according to the present invention.

FIG. 2 is an image diagram of a pattern defect inspection according to the present embodiment.

FIG. 3 is an image diagram (1) of a pattern defect in a conventional image comparison method.

FIG. 4 is an image diagram (2) of a pattern defect in a conventional image comparison method.

FIG. 5 is an image diagram (3) of a pattern defect in the conventional image comparison method.

FIG. 6 is a block diagram of a conventional pattern defect inspection device.

[Explanation of symbols]

2 ... A / D converter, 4 ... Inspection object, 11 ... Pixel unit alignment unit, 14 ... Comparison unit, 20 ... Local comparison unit, 21 ... Data conversion unit.

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2F065 AA02 AA49 AA56 BB02 CC19 CC25 DD04 FF04 JJ03 JJ26 LL00 MM03 MM04 PP12 QQ04 QQ24 QQ33 RR02 UU05 2G051 AA51 AB11 BB11 CA04 CB01 DA07 EA08 EA11 DJA12 EA11 DJA12 DJ11

Claims (1)

[Claims] An A / D converter for detecting a gray image signal of a pattern to be inspected and converting it into a digital image signal, and a pixel unit for calculating a gray difference while shifting one of two images to be compared in a pixel unit Positioning means, local comparing means for comparing signals obtained from minutely divided local adjacent parts, data converting means for outputting 0 when there is no signal difference at this time, and output of the data converting means And a comparing means for comparing signals of all the left and right patterns based on the data.