JPS63241307A - Method for detecting and correcting angle of rotation - Google Patents

Abstract

PURPOSE:To automatically detect and correct rotational shift, by storing and differentiating the image data obtained by scanning two regions by electron beam to detect the position of a characteristic point. CONSTITUTION:A scanning control part 5 and a stage control part 6 are controlled to perform the electron beam scanning of the first predetermined region containing one edge part of an object (wafer) 2 to be measured in a definite direction and subsequently moved by predetermined quantity to scan the second predetermined region by electron beam. The first and second image data obtained in two regions are stored in a memory 8 and differentiated. From the first and second differentiated image data, the positions of characteristic points where the data held between the max. and min. values become 0 and, from the difference between the positions of these two characteristic points and the predetermined moved quantity from the first region to the second region, the shift quantity of the angle of rotation of the object 2 to be measured is calculated to correct the feed quantity of the object 2 to be measured.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) This invention relates to a measurement device using a scanning electron microscope, and is particularly useful for detecting errors due to rotational misalignment and detecting errors caused by rotational misalignment, especially when determining all J of semiconductor wafers. The present invention relates to a rotation angle detection and correction method for correcting a deviation from a correct installation position.

(Prior art) In the conventional technology, the dimension al1 of the pattern of a semiconductor integrated circuit is
For one constant, a measuring device using a micrometer using an optical microscope is widely used. However, when the pattern size is about 2 μm or less, measurement becomes difficult due to resolution problems, so pattern size measurement is performed using a scanning electron microscope.

That is, as shown in FIG. 6, a semiconductor wafer is positioned and mounted on an x-y table, and the pattern-forming surface of the wafer is scanned with an electron beam to obtain image information. This image information is displayed on the screen, and an operator measures the dimensions by holding the measurement pattern image displayed on the image with a cursor or the like.

(Problem to be Solved by the Invention) The pattern image displayed by a scanning electron microscope may be caused by the rotation of the wafer relative to the If this happens, a measurement error S will occur as shown in FIG. 6 with respect to the cursor line used by the operator to insert the measurement pattern. In order to center such a tllj constant error, an operator usually rotates the displayed image and then pinches the measurement pattern between cursor lines to detect the dimensions. Additionally, when measuring the dimensions of a predetermined pattern on a wafer, if the wafer is rotating relative to the Ta.

The present invention enables highly accurate and automatic detection of rotational angle deviation during measurement using such a scanning electron microscope.
The purpose is to correct the feed amount of the object to be measured based on the results.

[Structure of the Invention] (Means for resolving the problem) In a method for detecting the rotation angle of an object to be measured of an electronic measuring device using a scanning microscope, The electron beam is scanned in a predetermined direction in a first region, and then moved by a predetermined distance to scan a predetermined second region. The first and second image data obtained in these two regions are stored and differentiated. From the differentiated first and second image data, the position of the feature point where the data value between the maximum value and the minimum value is O is detected, and the difference between the positions of these two feature points and the first The amount of rotation angle deviation of the object to be measured is calculated from the predetermined amount moved from the area to the second area, and the amount of feed of the object to be measured when performing an arbitrary measurement is corrected from this amount of rotation angle deviation.

(Operation) In the method of detecting and correcting the rotation angle as described above, the feature point where the data value between the maximum value and the minimum value of the first and second image data is 0 is the first and second image data. It almost coincides with the edge of the object to be measured in the second region. This is because when the electron beam is scanned, the edges of the object to be measured are tilted, and due to the tilt angle effect, more secondary electrons are emitted than other parts, resulting in the peak shown in Figure 7. Hail! By differentiating this, the point where the data value between the maximum value and the minimum value becomes 0 is defined as the edge of the object to be measured. Since the difference in distance between the first region and the second region is predetermined, the arc sine can be calculated from the difference in the position of the edge in the first region and the edge in the second region. The angle of rotational deviation can be detected. Using the rotation angle detected in this way, the object to be measured is placed at a predetermined measurement position.
By correcting the amount of movement of the -Y table, the n+ constant device can be operated automatically and with high precision.

(Example) An example of a wafer pattern measuring apparatus using the method of the present invention will be described below with reference to FIGS. 1 to 5.

Fig. 2 shows a configuration diagram of the apparatus of this example.The wafer (2) is positioned in the measurement position of the scanning electron microscope (1) via the bin 1.
1 is provided with an X-Y table (3) for positioning and mounting. A scan control unit (5) connected to the electron lens (4) in the scanning electron microscope (1) controls the scanning of the electron beam, and a table control unit (5) that controls the movement of the x-y table (3). 6) is provided. Further, the electron beam b is scanned onto the wafer (2), and a secondary electron detector (7) that captures the secondary electrons b2 generated by this scanning is provided with a storage memory (8) that stores the detected data, and a secondary electron detector (7) that captures the secondary electrons b2 generated by this scanning. A display section (9) for displaying images is connected. It is provided so that data in this storage memory (8) can be read. Further, this arithmetic control unit (lO) is connected to the table control unit (6) so that its position can be controlled or confirmed so that control signals can be output to the scan control unit (5) and the display unit (9). .

Next, the operation of the method of the present invention will be explained using the apparatus of this embodiment. FIG. 1 is a flowchart of the method in this embodiment. A wafer (2) is fixed on an x-y table (3) using a positioning reference bin 11 as shown in FIG. 3, using an orientation flat reference. After fixing, the scan control section (5) and table control section (6) are rotated 14 times so that the area 21 falls within the scanning area of the electron beam as shown in FIG. At this time, as shown in FIG. 4a, a 500x low magnification image of the orientation flat of Ueno (2) is displayed on the display section (9). Here, the displayed image is scanned with the electron beam in the horizontal direction, and the image signal is stored in the storage memory (8). Memory memory at this time (8
) is shown in FIG. 4.

The peak value e of this waveform data is due to the inclination at the edge of the orientation flat, so a tilt angle effect occurs, and the secondary electron b
This is because the release of 2 increases. Using this peak value e, the apex of the peak value e is defined as the edge of the orientation flat. Therefore, this waveform data a is differentiated to obtain a differential waveform a2 as shown in FIG. 4C. This means that if the peak value e is as shown in Figure 4, all you need to do is find the maximum value of the waveform, but in reality, the peak value may be changed due to some problem, as shown in waveform data C in Figure 4 d. Values greater than 0 may appear in the vicinity. Therefore, by differentiating this waveform data C, the peak value e is determined as a point where the value sandwiched between the maximum value and the minimum value of the waveform data c2 obtained by differentiating the waveform data C is 0. This method of determination is, when examined from the scanning direction of the electron beam b, the point of 0 value next to the maximum value, and vice versa, the point of 0 value next to the minimum value. The point obtained in this manner is defined as the edge of the orientation flat in the region 21.

Next, move the X-Y table (3) by a distance d.

At this time, for the area 22 displayed on the display section (9),
By scanning in the same manner as described above, the edge of the orientation flat in the area 22 is detected. At this time, the edge in area 21 and area 2
By determining the difference t between the edges of the two edges, the rotation angle θ of the wafer (2) is detected by taking the inverse sine of the movement ff1d of the x-Y table (3). In other words, the rotation angle θ can be detected by θ−arc tan ad /d. (However, α is μm per pixel.) Next, by correcting the amount of deflection of the display section (9) using this angle θ, a display like that shown in Figure 5a is obtained as shown in Figure 5. can be corrected to reduce measurement errors when cursor lines are aligned.

Also, move the X-Y table (
When moving and measuring 3), correct the angle θ using the following formula to the amount of movement of the X-Y table (3),
Highly accurate positioning is possible.

X = x cos θ - y sin θ Y - x sln θ + y cos θ In addition, the rotational deviation of wafer (2) does not necessarily mean that wafer (2)
Since the center of gravity of the electron beam is not shifted from the center of gravity, when the X-Y table (3) is moved, it will move with the shift as shown in Figure 5a, so the amount of deflection of the electron beam can be changed. By correcting by θ, an image as shown in FIG. 5 can be obtained.

In this embodiment, the magnification of the image was set to 500 times, but changing the magnification depending on the object to be measured does not have the same effect on the present invention.

[Effects of the Invention] As described above, by using the method of the present invention, it is possible to automatically detect and correct the rotational angle deviation, which was adjusted by an operator in the conventional technique.

In addition, automation eliminates variations in accuracy between operators and enables more precise adjustment.Furthermore, when working by moving the object to be detected, the movement can be performed with pre-compensation for pattern positional deviations due to rotational angle deviations. It made it possible.

[Brief explanation of drawings]

FIG. 1 is a flowchart of an embodiment of the apparatus using the method of the present invention, FIG. 2 is a block diagram showing the configuration of the apparatus, and FIG. 3 is a front view showing the state of the object to be detected.
Fig. 4 is a state diagram showing a similarly captured image, and a graph of the image data and image differential data of the image, and Fig. 5 is a state diagram showing the same before and after correction of the captured image. FIG. 6 is a state diagram showing a conventional method for measuring pattern dimensions.

Claims (1)

[Claims] A method for detecting the rotation angle of an object to be measured in an electronic measuring device using a scanning microscope includes an electron beam scanning step of scanning an electron beam in a fixed direction in a predetermined first region; An electron beam scanning movement step in which the electron beam is moved by a predetermined amount from the first region to scan the electron beam in a predetermined second region after completion, and the first and second image data obtained in the above two steps. an image data storage step for storing the image data, an image data differentiation step for differentiating the first and second image data, and maximum values from the results of the first and second image data obtained in the image data differentiation step, respectively. From the feature point detection step of detecting the location of the feature point where the data value between and the minimum value is 0, and the difference in the location of the feature point of the first and second image data obtained in this feature point detection step. A rotational deviation amount calculation step for calculating the rotational angle deviation amount of the object to be measured; and a feed amount correction step for correcting the feed amount of the object to be measured when performing an arbitrary measurement based on the calculation result of the rotational angle deviation amount calculation step. A rotation angle detection and correction method characterized by comprising: