JP2954381B2 - Pattern inspection method and apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a visual inspection apparatus for detecting a defect in a pattern to be inspected, and more particularly to a pattern detecting method and apparatus suitable for visual inspection of a pattern such as a semiconductor wafer or a liquid crystal display.

[0002]

2. Description of the Related Art Conventionally, this type of inspection apparatus is disclosed in
As described in Japanese Patent No. 192943, while moving the pattern to be inspected at a constant speed, an image of the pattern to be inspected is detected by an image sensor such as a line sensor, and the position of the detected image signal and the image signal delayed by a certain time are detected. The mismatch is recognized as a defect by correcting the deviation at predetermined time intervals and comparing them. As a line sensor,
One-dimensional CCD line sensor and recently time delay integration type
(Time Delay Integration) CCD Image Sensor
(Hereinafter, referred to as a TDI image sensor) is used to detect an image of a target pattern via an objective lens having a relatively high magnification. The TDI image sensor is
It has a structure in which a plurality of one-dimensional image sensors are arranged two-dimensionally, and the output of each one-dimensional image sensor is delayed for a predetermined time and added to the output of an adjacent one-dimensional image sensor that has imaged the same target position. By doing so, the amount of detected light is increased.

[0003]

However, in such a configuration, each layer of the target multilayer pattern overlaps,
As a result of having high steps and irregularities, there is a problem that the depth of focus is insufficient with a high-magnification objective lens, and only a pattern in a narrow range formed on an image sensor is detected as an image, and most of the other components are blurred. Therefore, according to the conventional method, only a part of the focused pattern is accurately inspected, and most of the other patterns that are not focused have low defect detection sensitivity, so that highly reliable inspection cannot be realized. Had issues.

An object of the present invention is to perform a highly reliable inspection of a single-layer or multilayer pattern, thereby realizing accurate and accurate image detection and highly sensitive comparison even for a pattern having a high step. To provide an appearance inspection method and apparatus.

It is another object of the present invention to provide an extremely high-speed appearance inspection method and apparatus.

[0006]

For this reason, the present invention has achieved the above object by realizing the following concept.

It has a structure in which a plurality of one-dimensional image sensors are arranged two-dimensionally, and the output of each one-dimensional image sensor is imaged at the same position of an object by delaying the output by a predetermined time. And an image sensor called a TDI image sensor having a function of adding the above.

The tilting direction is a direction perpendicular to the longitudinal direction of the internal one-dimensional image sensor with the center of the TDI image sensor as a fulcrum.

The amount of tilt is an amount corresponding to the unevenness of the object.

The target or the TDI image sensor is moved, and the TDI image sensor is driven in synchronization with the position.

The target image is detected by
This is compared with the detected pattern or the reference pattern in parallel with the positional deviation correction.

[0012]

According to the above means, each one-dimensional image sensor inside the TDI image sensor forms an image at a slightly different position (Z position) in a direction perpendicular to the optical axis. Further, each one-dimensional image sensor can detect the same position (X position) in the direction parallel to the optical axis without shifting. For this reason, even if the target has a high step or unevenness, the target forms an image on any one-dimensional image sensor surface, and as a result, a clear, large-amplitude signal output is obtained. The presence or absence of a defect is reflected in the value of any one-dimensional image sensor output. The clear large-amplitude signal and other blurred small-amplitude signals are added by the signal addition function of the TDI image sensor. These added signals are compared by the above-described means. However, the output signal of one of the one-dimensional image sensors capturing the defect has a large amplitude without blurring and a large difference from a normal part. Then, the presence of a defect can be detected. Actually, each one-dimensional image sensor causes a slight magnification error (Y position) due to the image formation in the direction perpendicular to the moving direction, but this is not a problem because it is canceled by the means. In addition, high-speed inspection can be performed by correcting positional deviation in parallel and comparing. As a result, even in a multilayer pattern in which each layer overlaps and has irregularities, high-precision pattern detection focused on each layer and high sensitivity can be compared in a short time. Therefore, the defect detection performance can be significantly improved as compared with the related art.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. FIG. 1 is a pattern inspection apparatus showing one embodiment of the present invention. In the figure, 5 is a time delay integration type (Time
This TDI image sensor has a structure in which a plurality of one-dimensional image sensors are arranged two-dimensionally, and the output of each one-dimensional image sensor is delayed by a predetermined time and the same position of the target is detected. Is added to the output of the adjacent one-dimensional image sensor that has taken the image to increase the amount of detected light. The concept of this type image sensor of the present invention and the like utility model (real
No. 61-26373 ). TDI image sensors of this type include those manufactured by Darsa Corporation of Canada and those manufactured by Reticon Corporation of the United States. The TDI image sensor 5 is arranged at an angle θ with respect to a plane perpendicular to the optical axis, as shown in FIG. The tilt direction is
Plural internal 1 with the center of TDI image sensor as a fulcrum
The direction is orthogonal to the longitudinal direction (Y direction) of the three-dimensional image sensor. The tilt amount is an amount corresponding to the unevenness of the target. For example, since a multilayer pattern of an LSI wafer has irregularities of about 1 to 3 μm, an amount obtained by multiplying the square of the magnification M of the objective lens 4 used, that is, an amount obtained by multiplying 1600 times by a 40 × objective lens is 1.6. Tilt the TDI image sensor by ~ 4.8 mm. Of course, this amount also depends on the performance of the auto-focusing mechanism (not shown) used, and it is preferable to provide a gradient only in a range covering this. That is, when the focusing accuracy is not good depending on the target, the tilt amount θ is also slightly increased. The scanning of the one-dimensional image sensor, for example, inside the TDI image sensor arranged at such an inclination is made coincident with the scanning in the Y direction, whereby the LSI wafer 1 which is the pattern to be inspected is made one-dimensional via the objective lens 4. The LSI wafer 1 can be detected by the XY table 6 and the T
The direction orthogonal to the main scanning of the DI image sensor 5, that is, X
By moving in the direction, the pattern to be inspected can be detected as a two-dimensional image. The LSI wafer 1 is illuminated by the illumination lamp 3. The XY table 6 has a linear scale 8 mounted thereon, and can accurately detect the actual position of the wafer. Timing generation circuit 9
, A start timing signal indicating a pixel is generated from the output signal of the linear scale 8, and the TDI image sensor 5
Is driven by this start timing signal every time it moves a certain distance. For example, when the pixel size is 0.15 μm, a start timing signal is generated every time the wafer moves by 0.15 μm in the X direction, and the image sensor is driven.

According to the above configuration, as shown in FIG. 2, the one-dimensional image sensors 5-1 to 5-m inside the TDI image sensor are slightly different in position (Z) in the direction perpendicular to the optical axis. Position). One-dimensional image sensor 5-k forms an image on the upper surface of layer A at time Ti. When the LSI wafer 1 moves in the X direction, the same position on the upper surface of the A layer is imaged on the adjacent one-dimensional image sensor at time Tj. In this way, the same position (X position) of the target is detected without being shifted by each one-dimensional image sensor. For this reason, even if the target has a high step or unevenness, the target forms an image on any one-dimensional image sensor surface. As a result, a clear, large-amplitude signal output is obtained, and the presence or absence of a defect is reflected in the value of any one-dimensional image sensor output. The signals of all these image planes are added by the signal addition function of the TDI image sensor. In the case where the target is a three-layer pattern as shown in FIG. 3A, if there is a pattern defect in the lowermost layer, conventionally, for example, only the intermediate layer is focused, and as shown in FIG. Although only the pattern of the layer has a large contrast of the detection signal waveform, in the present invention, all three layers show the same contrast. Therefore, in the case where there is a defect in the lower layer, the contrast is conventionally small as shown in FIG. 3 (b) and the difference from the normal part is not clear, but in the present invention, the difference from the normal part is as shown in FIG. 3 (c). Can be clarified.

In FIG. 1, the TDI image sensor 5 outputs, for example, eight plural image signals in parallel. These plurality of output signals are output from the delay memory 7 to the wafer 1 by one.
Delay by the amount of time to move for chips. Thereby, TD
The output signal of the I image sensor 5 and the output signal of the delay memory 7 correspond to the image signals of the adjacent chips 2a and 2b. The plurality of output signals are compared in parallel for each channel, and defects are detected at high speed. Next, processing contents for one channel will be described. Image signal edge detection circuit 11
Input to a and 11b to detect the pattern edge of the pattern to be inspected. The edge detection circuits 11a and 11b are equipped with a differential operator for detecting, for example, a dark pattern edge. Then, the detected pattern edges are binarized by the binarization circuits 12a and 12b. Next, 22 is a mismatch detection circuit,
The configuration is as shown in FIG. That is, the mismatch detection circuit
Reference numeral 22 denotes a TD for converting the image signal from the output of the
Shift register for delaying one scan of I image sensor
26 × 26 pixels and 2 × 7 × 7 pixels (arbitrary range) by serial-in / parallel-out shift registers 27a-27g
Cut out a dimensional local image. The outputs of the other binarization circuit 12b are the same as the shift registers 28a to 28c and 2
9, the output of which is synchronized with the center position of the two-dimensional local image. Next, the shift register
EXOR circuits 30a to 30n take the exclusive OR of the output of 29 and each bit output of the local image to detect unmatched pixels, and the counters 31a to 30n calculate the number of unmatched pixels. Each of the counters 31a to 31n is cleared to zero every N scans of the TDI image sensor 5, and the value is read immediately before the counter to read out the area M × N of the product of the number M of pixels of the TDI image sensor 5 and the number N of scans. The unmatched pixels within are indicated. Each bit output of the local image is ± 3 with respect to the output of the shift register 29 in the X and Y directions.
Since the pixels are shifted by one pixel within the range of pixels, the counters 31a to 31n count the number of mismatched pixels in each shift amount when the ± 3 pixel input pattern is shifted in the X and Y directions. . Therefore, if a counter having a characteristic value of the number of mismatched pixels such as a minimum value or a minimum value is checked, it is possible to perform optimal alignment with small image mismatch. Of course, a counter in which the number of mismatched pixels is smaller than a predetermined value may be obtained, and a counter that satisfies the counter may be determined as the optimum position.

As described above, when the counters 31a to 31n are X,
When the number of mismatched pixels in each shift amount when shifting the ± 3 pixel input pattern in the Y direction is counted, the value counted by the minimum value detection circuit 32 is read, and a counter having a minimum value or a minimum value is selected. The TDI image sensor 5 outputs shift amounts 34a and 34b for scanning in the Y direction and shift amounts 33a and 33b for scanning in the X direction perpendicular to the Y direction.

Reference numerals 23a and 23b denote delay circuits, which are the number M of pixels of the internal one-dimensional image sensor of the TDI image sensor 5.
And the register in an area M × N corresponding to the product of the number of scans of the TDI image sensor 5 and the number of scans N required for alignment, so that the outputs from the binarization circuits 12a and 12b are delayed. I have.

Numeral 24 denotes an alignment circuit. As shown in FIG. 5, shift amounts 33a and 33b for scanning the TDI image sensor 5 in the X direction from the minimum value detection circuit 32 are input to a selection circuit 35. In this case, the selection circuit 35 selects an optimum shift position from the outputs of the one delay circuit 23a and the shift registers 36a to 36f for delaying by one scan of the TDI image sensor 5, and inputs the selected shift position to the registers 37a and 37b. Is done. The selection circuits 38a and 38b select the optimum shift position of the TDI image sensor 5 in the Y direction from the shift amounts 34a and 34b when the TDI image sensor 5 scans in the Y direction from the minimum detection circuit 32. . Therefore, a local image corresponding to the shift position where the amount of mismatch is minimum or minimum is extracted from the outputs of the selection circuits 38a and 38b. From the output of the other delay circuit 23b, the shift register 3
Using the outputs of 9a to 39c and 40, the shift register 29
An image is synchronously extracted at a position delayed by the same amount as the output of (see FIG. 4). In this state, the local images output from the selection circuits 38a and 38b are at the optimal shift positions without positional deviation with respect to the local images output from the shift register 40.

When the optimal shift position is determined in this way, the defect judging circuit 25 regards the inconsistency which appears in common as a defect. FIG. 6 shows the defect determination circuit 25 described above. The figure shows the case of two shift positions. The EXOR circuit detects a mismatch between the output of the selection circuit 38a and the output of the shift register 40, and at the same time, the output of the output shift register 40 of the selection circuit 38b. The EXOR circuit 43 detects the inconsistency, and the AND of the output of the EXOR circuit 42 is calculated by the AND circuit 44. Then, the decision unit 45 decides whether or not the output signal of the AND circuit 44 is defective according to the size and outputs it.

As described above, the edge detection circuit 11, the binarization circuit 12, the mismatch detection circuit 22, the delay circuit 23, the positioning circuit 24, the defect determination circuit 25, etc.
It is possible to detect a defect for each (Channel). In this embodiment, the processing is completely independent for each channel.
Can be shared. FIG. 7 shows an example of an inspection apparatus having a common alignment. In the pattern inspection apparatus shown in FIG.
For the channels 2 to 8, the number of unmatched pixels for each channel is calculated by the counter 31 as shown in FIG. 9 in the unmatched pixel number detection circuit 48 for each channel, and the unmatched pixel number is calculated for the channel 1 as shown in FIG. Are summed by an adder 46, and the output of the adder 46 is processed by a minimum value detection circuit. In this case, since a wider range of images can be handled, the accuracy of positional deviation correction is improved. This is particularly effective for inspecting places where the pattern density is small. Channels 1, 2,
7 and 8 are configured as described above, and otherwise, the delay circuit 2
3. Mismatch detection may be omitted as only the alignment circuit 24 and the defect determination circuit 25.

In the above embodiment, the binarization circuit 12
We explained defect detection using digitized images,
In FIGS. 1 and 7, the binarization circuit 12 may be set to thru to detect defects using the grayscale image itself. (FIGS. 1, 7
Are connected like dotted lines. In this case, the binarization circuit 12
Are connected only to the mismatch detection circuit 22. In this case, a region having a large difference in shading is regarded as a defect.
With such a configuration, a signal of a defect different from that of a normal portion can be detected with higher accuracy. Further, in practice, each one-dimensional image sensor causes a slight magnification error (Y position) due to image formation in a direction orthogonal to the moving direction, but this error is also canceled out by the comparison, which is a problem. No. in this way,
Although the TDI image sensor is intended to increase the amount of detected light, it is extremely effective in increasing the depth of focus. As a result, even when a multi-layer pattern in which each layer overlaps and has irregularities is formed, the focus of each layer can be reduced. Highly accurate pattern detection and high sensitivity can be compared. In particular, fine defects can be detected in any layer. Therefore, the defect detection performance can be significantly improved as compared with the related art.

Although the details have been described with reference to the embodiments, the individual components can be realized by existing techniques. In addition, although the method of inspecting using one TDI image sensor has been described, the present invention can be applied to a method of comparing and inspecting images detected simultaneously using two TDI image sensors.

Further, a model pattern is created from a detected image, and the pattern is compared with a circuit pattern of a chip to be inspected, as described in the method proposed by the present invention, that is, as described in JP-A-61-65444. May be used to detect defects. Also, in the case of the method of comparing using design data, it is not necessary to specially consider the difference in blur caused by the positional relationship of each layer in the Z direction, and only the vertical relationship of each layer is considered. Accurate comparisons can be made with uniform and simple processing.

[0024]

As described above, even in a multilayer pattern in which each layer overlaps and has irregularities, it is possible to detect a focused high-precision pattern and compare high sensitivity, thereby dramatically improving defect detection performance as compared with the conventional method. Thus, it is possible to provide a visual inspection method and apparatus which are described above. In particular, fine defects can be detected in any layer. Further, high-speed appearance inspection can be realized.

[Brief description of the drawings]

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

FIG. 2 is a diagram showing an image forming relationship by a TDI image sensor.

FIG. 3 is a diagram illustrating an example of a detection signal waveform.

FIG. 4 is a diagram illustrating a mismatch detection circuit;

FIG. 5 is a diagram showing an alignment circuit.

FIG. 6 is a diagram illustrating a defect determination circuit.

FIG. 7 is a diagram illustrating a pattern inspection apparatus according to another embodiment.

FIG. 8 is a diagram illustrating a mismatch detection circuit;

FIG. 9 is a diagram illustrating a mismatched pixel number detection circuit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Wafer 2 ... Chip 3 ... Illumination light 4 ... Objective lens 5 ... TDI image sensor 6 ... Stage 7 ... Delay memory 8 ... Linear scale 9 ... Timing generation circuit 11 ... Edge detection circuit 12 ... Binarization circuit 22 ... Mismatch detection Circuit 23 ... Delay circuit 24 ... Alignment circuit 25 ... Defect determination circuit 26-29 ... Shift register 30 ... EXOR circuit 31 ... Counter 32 ... Minimum value detection circuit 36,37,39,40 ... Shift register 45 ... Determiner 46 ... Adder 47: mismatch detection circuit 48: mismatch detection pixel number detection circuit

──────────────────────────────────────────────────続 き Continuation of the front page (72) Takashi Hiroi 292, Yoshida-cho, Totsuka-ku, Yokohama-shi Inside Hitachi, Ltd. Production Engineering Laboratory (56) References JP-A-59-19294 (JP, A) JP-A-1 -162471 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G01N 21/88 G01B 11/24 H01L 21/66

Claims (10)

(57) [Claims] An inspection object on a mounting surface of a sample stage, on which a plurality of patterns to be originally formed in the same shape are regularly arranged, a plurality of one-dimensional image sensors are two-dimensionally arranged. A method for detecting a defect in a pattern by imaging with a time delay integration type CCD image sensor having a structure and comparing the image of the pattern obtained by imaging with a comparative image stored in a storage means. While relatively moving the time-delay-integration type CCD image sensor and the object to be inspected in a direction substantially perpendicular to the longitudinal direction of the one-dimensional image sensor, the sample is moved in a direction orthogonal to the longitudinal direction of the one-dimensional image sensor. An image of the pattern obtained by capturing an image in synchronization with the movement by the time delay integration type CCD image sensor installed at an angle to the mounting surface of the table is referred to as the comparison image. Pattern inspection method characterized by inspecting defects of said pattern by compare. A plurality of one-dimensional image sensors are two-dimensionally arranged on a specimen to be inspected on a mounting surface of a sample stage, on which a plurality of patterns which should have the same shape are regularly arranged. A method for detecting a defect in the pattern by scanning and imaging in a direction substantially perpendicular to the longitudinal direction of the one-dimensional image sensor with a time delay integration type CCD image sensor having a structure, wherein the sample is illuminated, The time-delay integrating type C, in which the image of the sample obtained is tilted in the scanning direction with respect to the optical axis of the detection optical system via the detection optical system, and the light-receiving surface is installed.
A pattern inspection method, wherein an image is captured by a CD image sensor, and an image obtained by the imaging is compared with a comparison image stored in a storage unit to inspect the pattern for defects. 3. The sample moves in a direction perpendicular to a longitudinal direction of the one-dimensional image sensor with respect to the time delay integration type CCD image sensor, and a distance between the sample and the sample gradually changes in the moving direction. 3. The pattern inspection method according to claim 2, wherein the image is taken in synchronization with the relative movement by the time-delay-integration type CCD image sensor having a light-receiving surface installed so as to perform the operation. 4. The CCD camera according to claim 1, wherein the sample stage is the time delay integration type CCD.
The image sensor is installed so as to continuously move in a direction substantially perpendicular to the longitudinal direction of the one-dimensional image sensor, and the distance between the sample stage and the sample stage gradually changes in the continuously moving direction. 3. The pattern inspection method according to claim 2, wherein said time delay integration type CCD image sensor having a light receiving surface picks up an image in synchronization with said continuous movement. 5. A method in which a plurality of patterns which should have the same shape are provided.
A plurality of one-dimensional image sensors
Inspection by imaging with a solid-state image sensor in which sensors are arranged two-dimensionally
The inspection object is attached to the solid-state imaging device.
On the other hand, one that is almost perpendicular to the arrangement of the one-dimensional image sensors
Moving relatively continuously in the direction,
Synchronize the two-dimensional image of the same part of the inspection object
Sequentially detected by a plurality of one-dimensional image sensors arranged in
The image signals obtained by the sequential detection are output from the solid-state image sensor before.
Inspecting the pattern for defects by outputting in parallel in synchronization with the relative continuous movement and comparing and processing the image signals output in parallel with the comparison image stored in the storage means; A pattern inspection method characterized by the following. 6. A solid-state imaging device comprising:
The pattern inspection method according to claim 5, wherein the pattern inspection method is a D image sensor. 7. A time-delay integration type CCD image sensor having a structure in which a plurality of one-dimensional image sensors are two-dimensionally arrayed to inspect an object to be inspected in which a plurality of patterns which should have the same shape are regularly arranged. And a method for detecting a defect in the pattern by comparing the image of the pattern obtained by the imaging with a comparison image stored in a storage means, wherein the time-delay integrating CCD image sensor and the object to be inspected are compared. While moving relatively in a direction substantially perpendicular to the longitudinal direction of the one-dimensional image sensor, the time-delay integrating CCD image sensor images the object to be inspected in synchronization with the movement. A plurality of image signals obtained by imaging are output in parallel from the time delay integration type CCD image sensor, and the plurality of image signals output in parallel are compared with the comparative image. Pattern inspection method characterized by inspecting defects of serial pattern. 8. A sample table means having a mounting surface on which a sample can be mounted and movable in a plane, and detection for obtaining an image of an inspection object mounted on the mounting surface of the sample table means. An optical system means, and a structure in which a plurality of one-dimensional image sensors for capturing an image obtained by the detection optical system means are two-dimensionally arranged, and a sensor surface is arranged with respect to an optical axis of the detection optical system. A time-delay-integration type CCD image sensor means inclined and installed in a direction substantially perpendicular to the longitudinal direction of the one-dimensional image sensor; a storage means for storing a comparison image;
Control means for controlling the timing of imaging by the time-delay-integrated CCD image sensor means in accordance with the amount of movement of the two-dimensional image sensor in a direction orthogonal to the longitudinal direction; and the time-delay-integrated CCD controlled by the control means A defect detection unit for detecting a defect in the pattern by comparing an image of the pattern obtained by imaging the sample with an image sensor unit with a comparison image stored in the storage unit. Inspection equipment. 9. A sample table means having a mounting surface on which a sample can be mounted and movable in a plane, and detection for obtaining an image of an inspection object mounted on the mounting surface of said sample table means. Optical system means, time delay integration type CCD image sensor means having a structure in which a plurality of one-dimensional image sensors for capturing an image obtained by the detection optical system means are two-dimensionally arranged, and storage for storing a comparison image Means for controlling the timing of imaging of the time-delay-integrated CCD image sensor means in accordance with movement of the mounting surface of the sample stage means in a direction substantially perpendicular to the longitudinal direction of the one-dimensional image sensor; Comparing the image of the pattern obtained by imaging the sample with the time-delay-integration type CCD image sensor controlled by the control means with a comparison image stored in the storage means to detect a defect in the pattern; Missing to detect A defect detection device comprising: a detection unit that outputs in parallel the image signal of the pattern obtained by imaging the sample by the time delay integration type CCD image sensor unit; A pattern inspection apparatus for inspecting a defect of the pattern by comparing the output image signal with a comparison image stored in the storage unit. 10. A sample stage means having a mounting surface on which a sample can be mounted and movable in a plane, and detection for obtaining an image of an inspection object mounted on the mounting surface of said sample stage means. Optical system means, time delay integration type CCD image sensor means having a structure in which a plurality of one-dimensional image sensors for capturing an image obtained by the detection optical system means are two-dimensionally arranged, and storage for storing a comparison image Means for controlling the timing of imaging of the time-delay-integrated CCD image sensor means in accordance with the amount of movement of the mounting surface of the sample stage means in a direction substantially perpendicular to the longitudinal direction of the one-dimensional image sensor. Comparing the image of the pattern obtained by imaging the sample with the time-delay-integration type CCD image sensor means controlled by the control means with a comparison image stored in the storage means to determine a pattern defect. Missing to detect A defect detection device comprising: a detection unit configured to detect a mismatch amount between the two images by comparing an image signal output from the time delay integration type CCD image sensor unit with a comparison image stored in the storage unit. And a delay time control means for controlling a delay time of an image signal output from the time delay integration type CCD image sensor means based on the amount of mismatch detected by the mismatch detection means. Pattern inspection equipment.