JP2005300884A - Mask inspection apparatus and mask inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To carry out a mask defect inspection in a short period of time with high accuracy. <P>SOLUTION: The apparatus is equipped with: an inspection area operation section 51 to operate the inspection area around a segment functioning to seam drawn images from the drawing data of a mask; a pixel setting section 52 to divide the inspection area operated by the inspection area operation section 51 into a plurality of pixels; a reflected image deciding section 53 to take in the reflected image on the mask surface in the inspection area and to decide whether possibility of failure is present or not in the segment; and a segment deciding section 54 to take in the translucent image of the mask in the inspection area when possibility of failure is decided to be present in the segment by the reflective image deciding section 53 and to compare the taken-in image with a preliminarily taken in reference image by each pixel unit to decide whether failure in the segment is present or not. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a mask inspection apparatus and a mask inspection method for inspecting a mask defect in a segment serving as a joint in drawing when manufacturing a mask.

  In a video system device (CCD, LCD, CMOS sensor, etc.), it is formed with a periodic pattern in which minute openings are repeatedly arranged. In such a product, when unevenness occurs on the device in the device manufacturing stage, it appears as an imaging result or a video output result. One cause of the occurrence of this unevenness is a problem in lithography technology using a photomask (hereinafter simply referred to as “mask”). In other words, when manufacturing a device pattern by transferring it onto a wafer or glass substrate using a lithography technique, if the pattern is misaligned or misaligned on the original photomask, this misalignment is transferred as it is. As a result, in an image system device such as a CCD, this leads to non-uniformity of the opening serving as a light receiving portion.

  Visual unevenness (stripes, bands, spots, discoloration) on the photomask is guaranteed. Projector light (100,000,000 Lux) is projected onto the front and back of the photomask to open the pattern. A human being visually detects (sensory inspection) a difference in light density (brightness / darkness) caused by a change in luminance from the portion or a refracted light reflected on the pattern. Also disclosed is a method of detecting image density unevenness by processing an image obtained by photographing an inspection object (see, for example, Patent Document 1).

JP 2003-91728 A

  However, in visual (sensory test) determination, individual differences are greatly affected by differences in skill level, subjectivity of inspectors, and the like, and the determination also varies. In addition, in the method of detecting the unevenness from the image shading by performing the signal processing of the luminance change from the pattern opening using image processing or a line sensor, the image processing (calculation time) for detecting the unevenness from the image shading is large. There is a problem that more time is required to obtain the inspection result as the area to be inspected becomes larger.

  The present invention has been made to solve such problems. That is, the mask inspection apparatus of the present invention includes an inspection area calculation unit that calculates an inspection area centered on a segment that becomes a drawing joint from mask drawing data, and the inspection area calculated by the inspection area calculation unit includes a plurality of pixels. A pixel setting unit that divides into two, a reflected image of the mask surface in the inspection area, a reflection image determination unit that determines whether there is a possibility of failure in the segment, and a possibility of failure in the segment in the reflection image determination unit A segment determination unit that captures a transmission image of a mask in an inspection area and determines whether there is a defect in the segment by comparing with a reference image that has been captured in advance in pixel units. I have.

  Further, the mask inspection method of the present invention includes a step of calculating an inspection area centered on a segment serving as a drawing joint from mask drawing data, a step of dividing the calculated inspection area into a plurality of pixels, and an inspection area Capturing a reflection image of the mask surface in the step of determining whether there is a possibility of failure in the segment, and if it is determined that there is a possibility of failure in the segment, capturing a transmission image of the mask in the inspection area, And comparing with a reference image captured in advance on a pixel-by-pixel basis to determine whether there is a defect in the segment.

  In the present invention, when inspecting a mask, an inspection area centered on a segment serving as a joint of drawing is obtained from mask drawing data in advance, and inspection is limited to this inspection area. It becomes possible to efficiently perform the defect inspection of the segment portion having a large influence. In addition, in this inspection area, first, a reflection image of the mask surface is captured, and when it is determined that there is a possibility of a defect from the reflection image, a transmission image is captured, and the reference image and the pixel unit are compared. In the case where there is a possibility of failure, high-accuracy image comparison is performed, so that high-precision defect inspection can be performed in a short time.

  Therefore, according to the present invention, it is possible to greatly reduce the inspection time by limiting the inspection area of the mask, and to reduce detection variation as compared with visual (sensory inspection). It becomes possible to improve quality control and reliability.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram illustrating a mask inspection apparatus according to the present embodiment, FIG. 2 is a block diagram illustrating an internal configuration of a memory & arithmetic unit, and FIG. 3 is a schematic diagram illustrating a mask segment.

  As shown in FIG. 1, the mask inspection apparatus of this embodiment measures a stage 1 on which a mask M to be inspected is mounted, a servo motor 2 that moves the stage 1 in the XY direction, and a position in the XY direction of the stage 1. Stage interferometer 3, lamp 4 for irradiating light to mask M, first imaging device 10 for capturing a reflection image on the surface of mask M, second imaging device 20 for capturing a transmission image of mask M, setting of inspection area and image processing The memory & calculating part 5 which performs, the monitor 6 which displays a test result, and the control part 7 which controls each part are provided.

  Here, in order to accurately measure the position of the stage 1 that can be moved by the servo motor 2, the stage interferometer 3 is preferably a laser interferometer. The first imaging device 10 and the second imaging device 20 use a shutter function camera (CCD or CMOS sensor or the like). In addition, a revolver type wavelength interference filter 8 corresponding to a plurality of wavelengths is provided in front of the light receiving unit of the first imaging device 10. Further, each of the first imaging device 10 and the second imaging device 20 is provided with a magnification changing lens L. The lamp 4 that irradiates light on the mask surface preferably uses a high-intensity LED in order to realize illumination with high parallelism.

  The surface of the mask M mounted on the stage 1 is irradiated with light from the lamp 4, and a reflected image of the light can be captured by the first imaging device 10. The reflection image of the mask M captured by the first imaging device 10 can extract only a predetermined wavelength via the revolver type wavelength interference filter 8. The captured reflection image is sent to the memory & calculation unit 5 via the control unit 7.

  Further, the stage 1 is provided with a hole, and light emitted from the lamp 4 is transmitted below the stage 1 through the hole. Since the second imaging device 20 is disposed below the stage 1, the second imaging device 20 can capture a transmission image of the mask M. The captured transparent image is sent to the memory & calculation unit 5 via the control unit 7.

  The first imaging device 10 and the second imaging device 20 are arranged so as to capture a reflection image and a transmission image of the same area of the mask M. The area for capturing an image can be switched by moving the mask M together with the stage 1. The images captured by the first imaging device 10 and the second imaging device 20 are subjected to predetermined image processing by the memory & calculation unit 5. The captured image can be confirmed on the monitor 6, and the result of the image processing can be displayed on the monitor 6.

  The mask inspection apparatus according to the present embodiment is characterized in that visual unevenness (mainly streaky unevenness) mainly generated on the mask M is accurately detected in a short time. Here, the cause of the unevenness on the mask M is mostly caused by the mask drawing apparatus. In the mask drawing apparatus, the device pattern is divided into small areas of about 1 mm, and the device pattern is drawn for each area, and the boundary between the areas appears on the mask M as stripe-like unevenness.

  The dotted lines shown in FIG. 3 indicate streaky irregularities generated on the mask. FIG. 3A shows a mask drawing apparatus that continuously moves the stage, and FIG. 3B shows mask drawing that moves the stage in steps. This is the case of the device. Since the mask drawing apparatus moves the stage while scanning a beam having a predetermined shape, streaky unevenness occurs depending on the movement type of the stage. The position where the unevenness occurs coincides with a position (segment) that becomes a joint of mask drawing. In the present embodiment, this segment is detected from mask drawing data, an inspection area centered on the segment is set, and high-precision inspection centering on the portion where unevenness occurs is performed.

  In the mask inspection apparatus of this embodiment, the memory & calculation unit 5 detects a segment from the drawing data of the mask M, sets an inspection area centered on this segment, and determines a defect by image processing. In order to realize such processing, the memory & calculation unit 5 has a configuration as shown in FIG. That is, the memory & calculation unit 5 includes an inspection area calculation unit 51, a pixel setting unit 52, a reflected image determination unit 53, a segment determination unit 54, and a memory 55. Here, the inspection area calculation unit 51, the pixel setting unit 52, the reflected image determination unit 53, and the segment determination unit 54 are realized by a normal program, but may be configured by dedicated hardware.

  The inspection area calculation unit 51 detects the position of the segment from the drawing data of the mask M stored in the memory 55, and performs a process of calculating a predetermined inspection area centered on this segment. In addition, the pixel setting unit 52 performs processing for dividing the inspection area calculated by the inspection area calculation unit 51 into a plurality of pixels. The reflection image determination unit 53 acquires a reflection image of the mask surface in the inspection area from the first imaging device 10 and performs a process of determining whether there is a possibility of a defect in the segment. The segment determination unit 54 acquires a transmission image of the mask from the second imaging device 20 when the reflection image determination unit 53 determines that there is a possibility of a defect in the segment, and acquires a reference image (reference image) that has been captured in advance. ) And the pixel unit, and processing for determining whether or not there is a defect in the segment is performed.

  With such a configuration, the mask inspection area is limited to the center of the segment, and when the reflection image determination unit 53 determines that there is a possibility of a defect, the mask transmission image is compared with the reference image and the defect is highly accurate. Detection determination can be performed, and high-precision mask inspection can be performed in a short time.

  Next, the mask inspection method of this embodiment will be described. 4 to 8 are diagrams for sequentially explaining the mask inspection method of this embodiment, and each figure shows a flowchart and a schematic diagram corresponding to the flowchart. In addition, the code | symbol which is not illustrated by the following description shall refer FIG. 1, FIG.

First, as shown in FIG. 4, and retrieve the drawing data (step S1), the pixel pitch information using the Data Preparation System from the drawing data and drawing the segment information _{(coordinates: X 1, Y 1, X} 2, Y ₂₎ to extract and output in text format (step S2 and S3). The drawing data is acquired by the inspection area calculation unit 51 from the memory 55 or an external storage device (not shown). The drawing segment information includes XY coordinates corresponding to the segment start position and XY coordinates corresponding to the segment end position.

  Next, as shown in FIG. 5, calculation processing is performed based on the output coordinate value of the drawing segment information and the pixel pitch information (Xp, Yp), and the image capture range (Px1, Py1, Px2, Py2) that becomes the inspection area ) Is determined (steps S4 to S6). That is, the inspection area calculation unit 51 calculates a predetermined width centered on the segment position as the inspection area, and obtains the XY coordinates of the inspection area. For example, the width of the upper and lower 2 pixels (for a total of 4 pixels) centered on the segment is obtained from the pixel pitch information, and the area of the width centered on the segment is obtained as the detection area.

  Further, the magnification changing lens L is operated to divide the inspection area into a plurality of pixels. That is, the pixel setting unit 52 performs setting so that a plurality of pixels (lattice areas) having a predetermined size are allocated to the size of the pattern constituting the mask. For example, the magnification changing lens L is operated so that four pixels of 2 × 2 correspond to one opening pattern, and the pattern size and the pixel arrangement when the image is captured are set to match ( Step S5).

  Next, the (Px1, Py1, Px2, Py2) area set as the inspection area is divided into fields (step S7), and one image capture area (Pfx1, Pfy1, Pfx2, Pfy2, Pfpx, Pfpy) is determined. This field division is also performed by the inspection area calculation unit 51.

  Next, as shown in FIG. 6, in the field of the same size adjacent to the inspection area, a reference image (reference image) is captured using a region not including a segment as a reference region (step S8). The reference image is a transmission image captured from the second imaging device 20. The captured reference image is digitized as transmitted light intensity in units of pixels and stored in the memory 55.

  Next, as shown in FIG. 7, the first imaging device 10 captures a reflection image of the inspection area (Px1, Py1, Px2, Py2) region (steps S9 to S11). At this time, the first imaging device 10 switches the revolver-type wavelength interference filter 8 at the same position, captures several types of images corresponding to different wavelengths, and calculates the light intensity distribution in each single wavelength image (step). S12).

  In the present embodiment, an LED light source is used as the lamp 4 in order to realize illumination with high parallelism. Further, by capturing reflected images of different wavelengths, noise components caused by interference can be reduced, and uneven portions can be detected from light intensity differences. Here, as an example, a reflection image corresponding to three wavelengths of 600 nm, 500 nm, and 400 nm is acquired, and each light intensity distribution (light intensity distribution along a dashed line in the drawing of the captured image shown in the figure) is calculated.

  Next, a predetermined threshold value is determined from the calculated light intensity distribution, and the reflection image determination unit 53 determines whether or not there is a possibility of occurrence of unevenness. The threshold can be arbitrarily set as a parameter (PRM).

  In the example shown in FIG. 7, the light intensity at the wavelength of 400 nm does not exceed the threshold value, but the light intensity at the wavelength of 500 nm and 600 nm exceeds the threshold value. In the determination based on this threshold value, even if the light intensity of any one wavelength exceeds the threshold value, there is a possibility of unevenness, and if the light intensity of two or more wavelengths exceeds the threshold value, there is a possibility of unevenness. The threshold value and determination method may be determined according to conditions such as the pattern material and pattern thickness of the mask M.

  If it is determined in this determination that there is a possibility that unevenness has occurred, the process proceeds to the next transmission image determination. On the other hand, if it is determined that there is no possibility of occurrence of unevenness, it is determined that no defect has occurred in the field of this inspection area, and inspection of the next field is performed to shorten the inspection time. You may make it plan. Even if it is determined that there is no possibility of unevenness, the next transmission image determination may be performed as necessary to increase the certainty of inspection.

  Next, the second imaging device 20 captures a transmission image in the field of the same inspection area with respect to the field of the inspection area extracted as a candidate where unevenness has occurred. Then, as shown in FIG. 8, a process of comparing the light transmission intensity of the transmission image in the field of the inspection area captured by the second imaging device 20 and the light transmission intensity of the reference image captured in advance in units of pixels. Perform (step S14). This comparison process is performed by the segment determination unit 54. As a result of comparing the light transmission intensity of the inspection area and the light transmission intensity of the reference image in units of pixels, it is automatically determined whether or not the difference exceeds a predetermined threshold value. It is determined that This threshold value can be arbitrarily set as a parameter (PRM).

  The presence / absence of occurrence of unevenness of the mask M using such a reflection image and transmission image is sequentially determined in each field of the inspection area to detect the presence / absence of unevenness of the mask M as a whole. Then, the determination as the final mask defect is determined based on the number of occurrences of unevenness and the occurrence location. Since the determination of the mask defect differs depending on the specification of the mask M, the determination of pass / fail of the mask defect can be automated by setting a determination criterion in advance in the mask inspection apparatus.

  In this way, since the position of the segment where the mask unevenness is generated in advance is detected from the drawing data, and the pass / fail judgment is performed by the image processing by setting the inspection area around the segment, It is not necessary to perform image processing, and the inspection time can be shortened.

  In addition, in the inspection area, first, a reflection image is captured to determine the possibility of unevenness, and if there is a possibility of unevenness, a transmission image of the same inspection area is captured and numerically compared with the reference image in units of pixels. Highly accurate pass / fail judgment is possible. As a result, it is possible to perform mask inspection with accuracy and little variation.

  In addition, the numerical value of the light transmission intensity of the reference image and the numerical value of the light transmission intensity of the transmission image in the inspection area used in the above description are merely examples, and actually, a finer value is used. In the above embodiment, an example of inspecting streaky unevenness along the horizontal direction in the figure as shown in FIG. 3A has been described, but along the vertical and horizontal directions in the figure as shown in FIG. Even in the case of inspecting streaky irregularities, an inspection area centering on vertical and horizontal segments may be set to perform the same inspection.

It is a schematic diagram explaining the mask inspection apparatus which concerns on this embodiment. It is a block diagram explaining the internal structure of a memory & calculating part. It is a schematic diagram explaining the segment of a mask. It is FIG. (1) explaining the mask inspection method of this embodiment in order. It is FIG. (2) explaining the mask inspection method of this embodiment in order. It is FIG. (The 3) explaining the mask inspection method of this embodiment in order. It is FIG. (4) explaining the mask inspection method of this embodiment in order. It is FIG. (5) explaining the mask inspection method of this embodiment in order.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Stage, 2 ... Servo motor, 3 ... Stage interferometer, 4 ... Lamp, 5 ... Memory & calculating part, 6 ... Monitor, 7 ... Control part, 8 ... Revolver type wavelength interference filter, 10 ... 1st imaging device, DESCRIPTION OF SYMBOLS 20 ... 2nd imaging device, 51 ... Inspection area calculating part, 52 ... Pixel setting part, 53 ... Reflection image determination part, 54 ... Segment determination part, 55 ... Memory, L ... Magnification change lens, M ... Mask

Claims (6)

An inspection area calculation unit that detects a segment that becomes a joint when drawing the mask from the drawing data of the mask, and calculates an inspection area around the segment;
A pixel setting unit that divides the inspection area calculated by the inspection area calculation unit into a plurality of pixels;
A reflection image determination unit that captures a reflection image of the mask surface in the inspection area and determines whether there is a possibility of a defect in the segment;
When the reflection image determination unit determines that there is a possibility of a defect in the segment, the transmission image of the mask in the inspection area is captured, and the transmission image and the reference image captured in advance are compared in pixel units. And a segment determination unit that determines whether or not there is a defect in the segment. The mask inspection apparatus according to claim 1, wherein the pixel setting unit assigns the plurality of pixels in accordance with a size of a pattern constituting the mask. The reflection image determination unit captures the reflection image of light having a plurality of wavelengths, and determines the possibility of failure in the segment based on the reflection image of light of any wavelength. Item 2. A mask inspection apparatus according to Item 1. A step of calculating an inspection area centered on a segment that becomes a joint of drawing from drawing data of the mask;
Dividing the calculated inspection area into a plurality of pixels;
Capturing a reflection image of the mask surface in the inspection area and determining whether there is a possibility of failure in the segment;
When it is determined that there is a possibility of a defect in the segment, a transmission image of the mask in the inspection area is captured, and the transmission image and a reference image that has been captured in advance are compared in units of pixels to determine a defect in the segment. And a step of determining whether or not there is a mask inspection method. 5. The mask inspection method according to claim 4, wherein in the step of dividing the inspection area into a plurality of pixels, the plurality of pixels are assigned in accordance with a size of a pattern constituting the mask. In the step of determining the possibility of failure in the segment, a reflected image of light having a plurality of wavelengths is captured, and the possibility of failure in the segment is determined based on the reflected image of light having any wavelength. The mask inspection method according to claim 4.

Images