JP2007139821A - Relative position measuring method and relative position measuring device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a relative position measuring method and a relative position measuring device for measuring relative positions of a plurality of observation samples, in which relative positions such as superposed positions of a plurality of observation samples to be compared can be directly measured with high accuracy. <P>SOLUTION: The device is equipped with a movable stage that can simultaneously mount a plurality of observation samples whose relative positions are to be measured and a laser interferometer that precisely measures each position of the plurality of observation samples to be measured. The method includes a step of measuring each position of the plurality of observation samples with the laser interferometer, a step of acquiring images of patterns on the plurality of observation samples, and a step of calculating a relative position of the pattern on the image of each observation sample based on the acquired image and the measured position of each observation sample. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a relative position measuring method and a relative position measuring apparatus for measuring relative positions of a plurality of samples to be observed.

  In recent years, masks for LSIs for semiconductors are required to have not only the dimensional accuracy of patterns in the mask but also the overlay accuracy of patterns between masks to be overlaid.

  In the conventional method, as shown in FIG. 9A, a position accuracy reference pattern is prepared as design data, and a reference mask is created. Then, as shown in FIG. 9 (b), the position coordinate measuring device performs measurement using the created reference mask and stores the measured value. Periodically measure using the reference mask, compare the measured value with the stored measured value, and the position coordinate measuring device is stable, that is, the measurement reproduction accuracy is ensured. Confirm.

  Next, as shown in FIG. 9 (c), a reference pattern is periodically drawn by a drawing device, and the drawn reference pattern is measured by a position coordinate measuring device that ensures the above-mentioned measurement reproduction accuracy. Then, it is confirmed that the drawing apparatus is stable as compared with the design data, that is, the drawing reproduction accuracy is ensured.

  As described above, the conventional method measures the reference mask to confirm the stability of the position coordinate measuring apparatus, and confirms the stability of the pattern drawn by the drawing apparatus with the position coordinate measuring apparatus whose stability has been confirmed. Since it was an indirect method, especially when creating a pattern to be overlaid with multiple real masks, it is a major problem that both cannot be measured directly to confirm stability (measurement reproduction accuracy) or accuracy. was there.

  In order to solve these problems, the present invention simultaneously mounts a plurality of samples to be compared on one stage, and measures the positions of the plurality of samples to be observed with a laser interferometer, respectively. Measure the shape of the pattern on the sample to be observed by scanning the charged particle beam to acquire and measure the image, and directly measure the relative position such as the overlay position of multiple samples to be compared. Like to do.

  In the present invention, a plurality of samples to be compared are simultaneously mounted on one stage, and the shape of the pattern on the plurality of samples to be observed is measured with a laser interferometer, respectively. By acquiring and measuring an image by scanning a charged particle beam or the like, it is possible to directly and accurately measure a relative position such as an overlay position of a plurality of samples to be compared. .

  In the present invention, a plurality of samples to be compared are simultaneously mounted on one stage, and the shape of the pattern on the plurality of samples to be observed is measured with a laser interferometer, respectively. For example, a charged particle beam beam was scanned to obtain an image and measured, and a relative position such as a superimposed position of a plurality of samples to be compared was directly measured with high accuracy.

  FIG. 1 shows a system configuration diagram of the present invention. Here, a case where one of the two specimens 5 to be observed is mounted on the main stage 1 and the remaining one is mounted on the substage 2 on the main stage 1 will be described in detail below. As a method of mounting three or more samples 5 to be observed on the stage 1, one sample 5 to be observed is mounted on the main stage 1 and all other samples to be observed 5 are mounted on the substage 2. Alternatively, one sample 5 to be observed is mounted on the main stage 1, the other remaining first sample 5 to be observed is mounted on the substage 2, and another substage (not shown) provided on the substage 2 is mounted. It is sufficient to repeat the mounting of the remaining second sample 5 to be observed.

FIG. 1A shows a schematic structural diagram of the mechanism section.
In FIG. 1A, a main stage 1 is a stage (X, Y, Z, θ) for mounting and moving a plurality of specimens 5 to be observed. And two sub-stages 2 are mounted.

  The sub-stage 2 is mounted on the main stage 1 and is a stage (X, Y, Z, θ) for mounting and moving the sample 5 to be observed. This is for adjusting the mounted sample 5 to be matched with the sample 5 mounted on the main stage 1 (see FIG. 3).

  The laser interferometer 3 is a known device for measuring the position precisely, and here, as shown in the figure, a stand (mask holder) (or an object to be observed) for fixing an object to be observed 5 mounted on the main stage 1. The X and Y directions of the sample 5 itself) and the X and Y directions of the table (mask holder) (or the sample 5 itself) for fixing the sample 5 to be observed mounted on the substage 2 are always measured accurately. is there.

  The electron optical system 4 is a well-known one that generates a narrowed electron beam, scans the surface of the sample 5 to be observed, and detects a signal emitted, reflected or absorbed at that time. , Scanning the plane while irradiating the sample 5 mounted on the main stage 1 and the sample 5 mounted on the substage 2 independently of each other and squeezing the electron beam narrowly, This is for detecting the reflected or absorbed signal and simultaneously generating images of the sample 5 to be observed.

  The sample 5 to be observed is a sample to be measured whose relative position is to be measured. Here, one sample is mounted on the main stage 1 and the other sample is mounted on the substage 2.

FIG. 1B shows a configuration example of a PC (personal computer) that controls the mechanism of FIG.
In FIG. 1B, a PC 11 is a personal computer and performs various controls according to a program. Here, a check unit 12, a measurement unit 13, a determination unit 14, a mask pattern data 15, a measurement data 16, It comprises correction data 17, a display device 18, an input / output device 19, and the like.

  The checking means 12 checks the error of the substage 2 with respect to the main stage 1 and adjusts the stage system (see FIG. 3).

  The measuring means 13 measures the position and size of the pattern on the image of the sample 5 to be observed mounted on the main stage 1 and the sample 5 to be observed mounted on the substage 2 (see FIG. 4 and the like). .

  The determination unit 14 compares the patterns on the images of the plurality of samples 5 to be observed with each other based on the positions and dimensions of the patterns on the image of the sample 5 to be observed measured by the measurement unit 13. (See FIGS. 4 and 5, etc.).

  The mask pattern data 15 is design data (data such as positions and dimensions of wiring patterns and through holes) when the sample 5 to be observed is a mask.

  The measurement data 16 is data of a pattern (data such as position and size) measured by the measurement unit 13 on the image.

  The correction data 17 stores various correction data (substage 2 correction data in FIG. 3, shift amount in FIG. 5 and the like).

The display device 18 is a display that displays an image or the like.
The input / output device 19 is used for inputting instructions and outputting measurement results.

  FIG. 2 shows a detailed structural diagram of the present invention. This shows an example of the detailed structure of stage 1 and sub-stage 2 in FIG.

  In FIG. 2, a mask holder 21 is provided on the main stage 1 as shown in the figure, and the sample 5 to be observed is mounted on the mask holder 21.

  The substage 2 is provided on the main stage 1 as shown in the figure, and is movable in the X, Y, Z, and θ directions.

  The mask holder 22 is provided on the substage 2 as shown in the figure, and the sample 5 to be observed is mounted on the mask holder 22.

  The electron optical system 4 is provided in association with the observed sample 5 mounted on the mask holder 21 and the mask holder 22, respectively, and scans the surface while irradiating a narrowed electron beam, and emits, reflects, or absorbs it. It is a well-known one for detecting the detected signal and displaying the image on the display device 18 respectively. Here, as shown in the figure, while irradiating an observation sample 5 mounted on the main stage 1 and an observation sample 5 mounted on the substage 2 independently and simultaneously with a narrowed electron beam. Planar scanning is performed, and each image is displayed on the display device 18.

  Next, the adjustment of the substage 2 will be described in detail according to the order of the flowchart of FIG. 3 under the structure of FIG. 1 and FIG.

FIG. 3 is a flowchart for explaining the operation of the present invention (stage system adjustment).
In FIG. 3, in S <b> 1, the mask (1) and the mask (2) having the same pattern are attached to the main stage 1 and the sub stage 2. As shown in FIGS. 1 and 2, this is a sample 5 to be observed. Here, a mask (1) and a mask (2) having the same pattern are attached (mounted) to the main stage 1 and the substage 2, respectively. .

  In S2, the area coordinates in the laser interferometer 3 are instructed to move the main stage. This is because the PC 11 in FIG. 1 has the same pattern masks (1) and (2) mounted on the main stage 1 and the substage 2 in S1 in order to adjust the main stage 1 and the substage 2. The main stage 1 is moved so that the coordinates read by the laser interferometer 3 become an area with a stage adjustment pattern.

  In S3, an image is acquired. This is because the main stage 1 is moved in S2 and the finely-squeezed electron beam emitted from the electron optical system in a state where the stage adjustment pattern on the mask (1) and the mask (2) is set to a certain area. The area is planarly scanned with a beam, and the emitted secondary electrons are detected and amplified to acquire respective images.

  In S4, the images are overlapped, and the error of the position at that time is measured. This is because the position error is measured by, for example, measuring Δx11 and Δy11 by superimposing the image from the mask (1) acquired in S3 and the image from the mask (2) and measuring the dimensional difference at that time. To do.

  In S5, it is determined whether or not the process is finished. This is for all of the reference patterns formed on the mask (1) and mask (2) of the same pattern (for example, a total of five reference patterns near the center and the four corners of the masks (1) and (2)). It is determined whether the error is measured by performing S2 to S4. In the case of YES, since measurement of the errors of all the reference patterns on the mask (1) and the mask (2) has been completed, the process proceeds to S6. In the case of NO, S2 and subsequent steps are repeated.

  In S6, it is determined whether the error is within a predetermined range. It is determined whether the error when superposed in S4 is within a predetermined range (within an allowable error obtained in advance by experiment). In the case of YES, since it has been found that the adjustment of the substage 2 with respect to the main stage 1 is completed (unnecessary), x, y, θ of the substage 2 at that time is stored in S8, and the process ends. On the other hand, in the case of NO in S6, since it was found that the overlay error is not within the predetermined value and is large, in S7, the substage (x, y, θ) is adjusted so that the overlay error becomes small, and S2 Repeat thereafter.

  1 and 2, the mask (1) (observed sample 5) mounted on the main stage 1 and the mask (2) (observed sample 5) mounted on the substage 2 are constructed. ) And the substage 2 can be adjusted so that the error when the images acquired from the above are superimposed within a predetermined range. Thereby, the error when the images acquired from the mask (1) mounted on the main stage 1 and the mask (2) mounted on the substage 2 are superimposed is within a predetermined range (within an allowable value). It was adjusted to fit in. Note that the error being within the predetermined range may be such that the error is within the predetermined range around 0 (zero), or may be within the predetermined range around the predetermined value.

  At this time, the laser interferometer 3 determines the position coordinates of the mask (1) itself and the mask (2) itself, or the mask holder 21 that holds the mask (1) and the mask holder 22 that holds the mask (2), respectively. The measurement is performed, and the main stage 1 is moved so that the predetermined area coordinates are obtained (the same applies hereinafter).

FIG. 4 is a flowchart for explaining the operation of the present invention (actual mask measurement).
In FIG. 4, a mask (1) and a mask (2) are set in S11. This is because the mask (1) and mask (2) to be superimposed (for example, the mask (1) (wiring pattern) in FIG. 6A described later) and the mask (2) (contact in FIG. 6B). The hole pattern)) is set (mounted) on the main stage 1 and the substage 2 shown in FIGS.

  In S 12, the mask (1) is adjusted to the reference point on the main stage 1. This is performed by moving the main stage 1 to match the reference point on the mask (1) mounted on the main stage 1 using the laser interferometer 3 (for example, aligning it with the center of the image). The positions of the mask (1) measured by the laser interferometer 3 at the time of the alignment are X and Y.

  In S13, the mask (2) is adjusted to the reference point in the substage 2. Similarly, the laser interferometer 3 is used to align the reference point on the mask (2) mounted on the substage 2 by moving the substage 2 (for example, align it with the center of the image). The position of the mask (2) measured by the laser interferometer 3 at the time of the alignment is defined as x and y.

  By the above S11 to S13, the position (X, Y) of the reference point of the mask (1) mounted on the main stage 1 and the position (x, y) of the reference point of the mask (2) mounted on the substage 2 Is measured by the laser interferometer 3 and calibrated.

  In S14, the process moves to a region where the overlay accuracy is required (or required). In S11 to S13, after calibrating at the reference point of the mask (1) and the reference point of the mask (2), the measurement object is moved to the region where the overlay accuracy is strictly required or the region where the requirement is required (here, Move on main stage 1).

  S15 checks two images. This is because after moving in S14, an image of the mask (1) mounted on the main stage 1 is acquired, and an image of the mask (2) mounted on the substage 1 is acquired, and the overlay state of both is checked. To do. For example, the state of a non-defective product shown in FIG. 6D described later or a defective product shown in FIG. 6E is checked (described later with reference to FIG. 6).

  S16 outputs. This is checked in S15 and the result is output (for example, the dimension in the case of a defective product is output).

  As described above, the main stage 1 is moved using the laser interferometer 3 to align the reference point of the mask (1) mounted on the main stage 1, and then the substage 2 is moved to move the substage. After aligning with the reference point of the mask (2) mounted on the stage 2, the mask (1) is moved to the area subject to overlay check (the area with high overlay accuracy, designated area) by the main stage 1 And the image of the mask (2) are superposed and the difference between them is checked (see FIG. 6). If the product is a non-defective product or a defective product, the difference (error) The position and the like can be output. Thereby, the image of the mask (1) mounted on one main stage 1 and the image of the mask (2) mounted on the substage 2 provided on the main stage 1 are simultaneously acquired and compared. Thus, it is possible to measure and determine the degree of overlay with extremely high accuracy and good reproducibility. That is, the environmental conditions (mask (1) and mask (2) temperature, etc.) are made the same, and images are acquired and compared simultaneously in parallel, and in particular, the quality of pattern superposition and the like is determined, and high accuracy and high reproduction are achieved. It is possible to make a determination while maintaining the property.

  FIG. 4 is a flowchart for explaining the operation of the present invention (actual mask measurement while checking the electron optical system and the like).

  In FIG. 4, S21 is adjusted to the position of the reference pattern of the masks (1) and (2). This is done by using the laser interferometer 3 and moving the main stage 1 so that the reference pattern of the mask (1) or mask (2) is at the center of the image, for example.

  In step S22, an image is acquired. This obtains an image (an image including a reference pattern) of the mask (1) mounted on the main stage 1, and an image (an image including the reference pattern) of the mask (2) mounted on the substage 2. get.

  In S23, the image patterns of both are superposed and the shift amount of the position is obtained. This is done by superimposing the image of the reference pattern of the mask (1) mounted on the main stage 1 acquired in S22 and the image of the reference pattern of the mask (2) mounted on the substage 2 at the position. Obtain the shift amount of. Here, the shift amounts of the images of both reference patterns are Sx1 and Sy1.

  In step S24, it is determined whether the process is finished. This determines whether or not the processing from S21 to S23 is completed for all the reference patterns. If YES, the process proceeds to S25. If NO, return to S21 and repeat.

S25 stores the shift amount obtained in S23.
The above-described S21 to S5 are executed at regular time intervals of S29 or every time the work area is changed, so that the reference pattern of the mask (1) mounted on the main stage 1 and the mask mounted on the substage 2 ( It is possible to sequentially measure the shift amount with respect to the reference pattern of 2).

  S26 is compared with the previous time. In this case, if there is a previously stored shift amount, the stored shift amount is compared with the current shift amount. If there is no previous shift amount stored, S26 is skipped.

  In S27, it is determined whether the shift amount is within a predetermined range. In S26, the previous and current shift amounts are compared to determine whether the difference is within a predetermined range. If YES, the process proceeds to S28. Even if the difference between the previous time and the current time is within a predetermined range, if it exceeds the predetermined allowable value compared with the first shift amount, it is determined NO in S27, and an alarm is issued to notify the administrator in S30. On the other hand, in the case of NO in S27, since the shift amount of the previous difference is greater than or equal to a predetermined value, an alarm is issued in S30 to notify the administrator.

  In S28, an actual pattern is measured. For this, actual pattern measurement such as overlaying of the mask (1) and the mask (2) is performed by S11 to S16 of FIG.

  In step S29, it is determined whether the work area changes every fixed time. In the case of YES, S21 and subsequent steps are executed. If NO, the process ends.

  As described above, the image of the reference pattern of the mask (1) and the mask (2) is compared at regular time intervals or every time the work area is changed, and an alarm is generated to notify the administrator when the shift amount is greater than or equal to a predetermined amount. On the other hand, the actual pattern measurement (FIG. 4) is performed when the shift amount is equal to or less than a predetermined value, so that the reference pattern on the mask (1) mounted on the main stage 1 and the mask (2) mounted on the substage 2 are measured. It is possible to perform overlay check on the pattern of the mask (1) and the mask (2) and the like with good accuracy and good reproducibility while monitoring that the shift amount with respect to the reference pattern is within a predetermined range.

FIG. 6 is an explanatory diagram of the present invention.
FIG. 6A shows an example of the mask (1) (wiring pattern).

FIG. 6B shows an example of the mask (2) (contact hole pattern).
FIG. 6C shows an example of a state in which the mask (1) and the mask (2) are overlapped. 6 (d) and 6 (e) show an enlarged view of the dotted dotted ellipse on the left side.

FIG. 6D shows a state in the case of a non-defective product. In the case of this good product,
・ The absolute value of the difference between a and b is within the specified value. ・ The absolute value of the difference between c and d is within the specified value.

FIG. 6E shows a state in the case of a defective product. In the case of this defective product,
・ The absolute value of the difference between a and b is greater than or equal to the specified value. ・ The absolute value of the difference between c and d is greater than or equal to the specified value.

  As described above, in the case of the wiring pattern of the mask (1) in FIG. 6A and the contact hole pattern of the mask (2) in FIG. (C)), a, b, c, and d in the figure are measured on the image, respectively, and when the absolute value of the difference between a and b and the difference between c and d is within a specified value, When the value is equal to or greater than the value, it can be determined as a defective product. At this time, the mask (1) is mounted on the main stage 1 of FIGS. 1 and 2, and the mask (2) is mounted on the substage 2 of FIGS. 1 and 2, and the method of FIG. 4 or FIG. It is possible to measure and determine good and defective products at high speed, high accuracy, and good reproducibility.

FIG. 7 is an explanatory diagram of the present invention (direct confirmation of accuracy of an actual LSI mask).
FIG. 7A shows an example of an LSI actual mask. An example is shown in which there is a DRAM area in the upper left, a Logic area in the upper right, and an MCP area in the lower center.

  7B shows the mask (1) (wiring pattern), and FIG. 7C shows the mask (2) (contact hole pattern). These masks (1) and (2) are obtained by enlarging the portion below the center of the actual LSI mask shown in FIG.

  FIG. 7D shows a state in which the mask (1) and the mask (2) overlap. When the contact hole pattern of the mask (2) overlaps the wiring pattern of the mask (1), a, b, c, and d are measured as shown in (d) or (e) of FIG. Accordingly, it is possible to determine a good product or a defective product at high speed, high accuracy, and good reproducibility.

  FIG. 8 is an explanatory diagram of the present invention. This shows an example in which alignment measurement is performed for each mask pattern region after the reference points of the masks (1) and (2) are aligned.

  FIG. 8A shows an example of an LSI actual mask. An example is shown in which there is a DRAM area in the upper left, a Logic area in the upper right, and an MCP area in the lower center.

  FIG. 8B shows an example of the mask (1), and FIG. 8C shows an example of the mask (2). Here, the reference points (reference patterns) are provided at three locations (upper center, left center, and lower right) here. Here, after matching the three reference points (reference patterns) of the mask (1) and the mask (2) (S12 and S13 in FIG. 4), each region (DRAM region, Logic region, MCP region) The alignment measurement of the internal pattern is performed.

  In order to further improve the accuracy, for example, three reference patterns are provided for each region, and after the reference patterns of the mask (1) and the mask (2) are matched (S12 in FIG. In step S13, overlay measurement of the pattern in the area is performed to determine whether the pattern is good or bad ((d) and (e) in FIG. 6).

  In the present invention, a plurality of samples to be compared are simultaneously mounted on one stage, and the shape of the pattern on the plurality of samples to be observed is measured with a laser interferometer, respectively. Relative position measurement method and relative position for measuring the relative position, such as the overlay position of a plurality of samples to be compared, with high accuracy by scanning the charged particle beam for measurement The present invention relates to a measuring device.

It is a system configuration diagram of the present invention. It is a detailed structure figure of the present invention. It is an operation explanation flowchart (stage system adjustment) of the present invention. It is operation | movement description flowchart (actual mask measurement) of this invention. It is operation | movement description flowchart (actual mask measurement, checking an electron optical system etc.) of this invention. It is explanatory drawing of this invention. It is explanatory drawing of this invention. It is explanatory drawing of this invention. It is explanatory drawing of a prior art.

Explanation of symbols

1: Main stage 2: Substage 3: Laser interferometer 4: Electro-optical system 5: Sample to be observed (mask)
11: PC (PC)
12: Check means 13: Measuring means 14: Determination means 15: Mask pattern data 16: Measurement data 17: Correction data 18: Display device 19: Input / output devices 21, 22: Mask holder

Claims (10)

In a relative position measurement method for measuring the relative position of a plurality of samples to be observed,
A stage capable of simultaneously loading and moving a plurality of samples to be measured at relative positions;
A laser interferometer that accurately measures each position for a plurality of samples to be measured of the relative position,
Measuring the positions of the plurality of samples to be observed with the laser interferometer,
Acquiring images of patterns on the plurality of specimens to be observed;
And calculating the relative position of the pattern on the image of each observed sample based on the acquired image and the measured position of each observed sample.   The image acquisition means for acquiring images of the patterns on the plurality of samples to be observed at the same time is provided, and images of patterns on the plurality of samples to be observed are acquired at the same time. Relative position measurement method.   3. An image is generated by scanning an electron beam with respect to images of patterns on the plurality of specimens to be detected and detecting signals emitted, reflected, or absorbed. Relative position measurement method.   The plurality of 2 are set as the plural, and two electron optical systems are provided for the image of the pattern on the two observed samples, and the two observed samples are emitted by the respective electron beam beams emitted from the two electron optical systems. 4. The relative position measuring method according to claim 1, wherein the image is generated by respectively detecting the signals emitted, reflected, or absorbed by scanning each of the above.   A plurality of samples to be observed were sequentially mounted by repeatedly mounting one sample to be observed on the one stage, providing a sub-stage on the one stage, and mounting another sample to be observed. The relative position measuring method according to claim 1, wherein:   One of the samples to be observed is mounted on the one stage, a substage is provided on the one stage, and all the other samples to be observed are mounted. The relative position measuring method described in 1.   When a sub-stage is provided on the one stage, the sub-stage is adjusted to align the sample to be observed mounted on the stage with the sample to be observed mounted on the sub-stage. The relative position measuring method according to claim 5 or 6.   A reference pattern for relative position measurement is provided at any position on the plurality of samples to be observed, and the relative positions of the patterns on the plurality of samples to be observed are measured based on the reference patterns. The relative position measuring method according to any one of claims 1 to 7.   The relative position measurement according to any one of claims 1 to 8, wherein the images generated from the plurality of samples to be observed are overlapped, and the position and dimensions are measured under the overlapped state. Method. In a relative position measuring device that measures the relative position of a plurality of samples to be observed,
A stage capable of simultaneously loading and moving a plurality of samples to be measured at relative positions;
A laser interferometer that precisely measures each position of a plurality of samples to be measured at the relative positions;
Means for measuring the positions of the plurality of samples to be observed with the laser interferometer,
Means for acquiring images of patterns on the plurality of samples to be observed;
A relative position measuring apparatus comprising: means for calculating a relative position of a pattern on an image of each observed sample based on the acquired image and the measured position of each observed sample.

Images