JPS63148112A - Tilt angle measuring instrument - Google Patents

Abstract

PURPOSE:To measure a tilt angle without destruction nor contacting by outputting an image signal which indicates the end part shape waveform of a pattern according to a secondary electron beam, and comparing the end part shape waveform with a stored simulation waveform and determining the tilt angle. CONSTITUTION:A simple 10 to be measured such as a semiconductor wafer where the resist pattern of an LSI, etc., is formed is mounted on the mount base of an SEM main body 1. Electron beam flux 4 emitted by an electron gun 3 is scanned by a scanning coil part 8 and projected on the sample 10 through an objective 11. A secondary electron emitted from the surface of the sample 10 is detected by a secondary electron detector 12, which outputs an image signal IS to a tilt angle measurement part 2 through an amplifier 13. The CPU 22 of the measurement part 2 compares the image signal indicating the end part waveform signal of the pattern obtained from the signal IS with the simulation waveform stored in a waveform storage part previously to determine the tilt angle. Thus, the tilt angle of the end part of the fine regist pattern is measured automatically without destruction nor contacting.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention provides a method for automatically measuring the inclination angle of the edge of a minute resist pattern formed on a semiconductor wafer. This invention relates to an angle measuring device.

(Prior Art) Conventionally, the measurement of the inclination of the edge of a resist pattern formed on a semiconductor wafer with respect to the wafer plate surface is an important part of inspection of a resist process.

In this type of tilt inspection, the pattern is cut perpendicularly, and the cut surface is 8 EM (Scanning
Observe with an ectron microscope (scanning electron microscope) and find the inclination on a photograph, or
Directly trace the undulations of the surface with a contact probe.
I'm looking for the slope.

However, the method of cutting and inspecting the former sample is
The problem is that it requires a destructive test and the sample fee is high. On the other hand, the latter method using a probe cannot measure when the pattern widths are arranged at intervals of 1 to 2 μm. However, all of these methods have the disadvantage that the inspection efficiency is extremely low. Non-destructive tilt angle to the surface,
It is an object of the present invention to provide an inclination angle measuring device that can perform non-contact and highly efficient measurement.

[Structure of the invention]

(Means and operations for solving the problem) A means for outputting an image signal indicating a pattern edge shape waveform based on a secondary electron beam, and a means for storing a shimieresis waveform to be compared; The end shape of the pattern is inspected non-contactly and non-destructively by means of comparing the end shape waveform and the shimmy erosion waveform and determining the inclination angle based on the comparison result.

(Example) Hereinafter, the present invention will be described in detail based on an example with reference to the drawings.

Figure 1 is a configuration diagram of the dimension measuring device of this embodiment.
g Electron Microscope; Hereinafter, simply abbreviated as 8EM. ) Main body (1) and this S
An inclination angle measurement unit (2) that measures the inclination angle of the end of a specific resist pattern portion captured by the EM main unit (1)
It consists of The main body (1) includes an electron gun (3) that emits electrons by a power source (not shown), and an electron gun (3) that emits electrons by a power source (not shown).
) and a condenser lens (5) that reduces the electron beam flux (4) emitted from the
and a sweep signal generator that generates a sweep signal SS for scanning the electron beam (4) based on the clock signal PS output from the reference signal generator (6). (7) and a magnification for outputting a control signal C8I to a scanning coil section (8), which will be described later, in combination with the sweep signal SS output from the sweep signal generation section (7) by setting a magnification changeover switch (not shown). Switching part (9)
, a scanning coil section (8) that controls the scanning direction and width of the electron beam (4) based on the control signal C8I, and further reduces the electron beam (4) and places the electron beam (4) on the measurement sample (10). ) and an objective lens (11) that irradiates the measurement sample (1
a secondary electron detector (12) that collects secondary electrons emitted from the secondary electron detector (12); an amplification section (13) that amplifies the signal from the secondary electron detector (12); 13) and the sweep signal SS output from the sweep signal generator (7), the CRT (Catho
An image signal amplifier (15) for displaying an image on a display tube (14), and a CRT (15) for displaying an enlarged image of a specific part of a measurement sample (10) held on a mounting table (not shown). 14). On the other hand, the tilt angle measurement section (2) includes a timing control section (16) that outputs a control signal ST based on the clock signal PS from the reference signal generation section (6) and the sweep signal SS from the sweep signal generation section (7). ), and an analog-digital (A/D) converter (
17) and a display image memory unit (
18), and a measurement image memory unit (19) that stores the digital value output from the input ρ conversion unit (17) based on the control signal SM output from the timing control unit (16).
and a graphic image for displaying a plurality of cursors on the CRT (14) superimposed on a still image based on the image from the SFM main unit (1) or the image data stored in the display image memory unit (18). a memory section (2o);
The cursor setting section (2o) is connected to the input side of this graphic image memory section (2o) and is used to move the cursor in various ways.
21), a CPU section (22) that reads data from the measurement image memory section (19) and the graphic image memory section (20), and has various arithmetic processing functions and storage functions;
A waveform storage section (22a) in which a simulation waveform, which will be described later, is stored, a measurement image memory section (19), and a display image memory section (18).

Data of 9 copies (20) of graphic image memos are transferred to CRT.
(14) A display control unit (23) that performs various controls for displaying numbers, a measurement image memory unit (19), a display image memory unit (r8), and a graphic image memory unit (20).
) and a digital-to-analog (D/A) converter (24) that is connected to the output side of the display controller (23) and converts digital data into analog and outputs and displays it on CR,T (14). has been done. The display control section (20) is also connected to the image signal amplifier (15), and controls to display the image data stored in the display image memory section (18) in CR,'r (14). It outputs a signal MA.

Next, the operation of the inclination angle measuring device configured as described above will be described in detail.

First, a measurement sample (10) such as a semiconductor wafer on which a resist pattern is formed, such as an LSI, is placed on the mounting table of the SEM main body (1). Therefore, the electron beam flux (4) emitted from the electron gun (3) passes through the condenser lens (5).
)..., and the control signal C8I output from the magnification switching section (9) causes the scanning coil section (8) to
Raster scanning is performed in the Y direction, the object lens (11) further reduces the size, and the measurement sample (10) is irradiated. Then,
Secondary electrons are emitted from the surface of the measurement sample (10). This secondary electron is collected by a secondary electron detector (12) and converted into an electrical signal. The electrical signal output from this secondary electron detector (12) is amplified by an amplifier (13),
It is output as an image signal signal S to an image signal amplifier (15). The image signal amplifier (15) combines the sweep signal SS output from the sweep signal generator (7) and the image signal generator S, and displays the combination as an image on the CRT (14). On the other hand, the image signal Is is controlled by the timing control section (16
), the A/D converter (1
7), the display image memory section (
18), one screen worth of data is stored. Display image memory section (
18), for example, as shown in Figure 2, 512X512
It has a capacity of 8 bits. The stored data is transferred to D by signals MC and MD of the display control section (23).
The /A conversion section (24) converts the control signal into an analog signal, and the control signal M is displayed as a still image on a CRT (4J). 8 to the graphic image memory section (20) and the D/A converter section (20).
4) Output the iC through the CRT (14) and connect the two wires to the CRT (25a).

(25b)-t' is set so as to sandwich the resist pattern P (see FIG. 3).

At this time, the positions of the cursors (25a) and (25b) are written into the graphic image memory section (20) using the signal KS output from the cursor setting section (21), and are stored in the display image memory section (18). Displayed completely overlapping. On the other hand, the CPU section (22) uses the cursor (25a), (25b) from the graphic image memory section (20).
) is read, and the address data is sent to the sweep signal generator (7) using the signal SA. The sweep signal generating section (7) is composed of kerfuls (25a) and (25b).
The electron beam beam (4) scans only the designated area. This time as well, the image signal Is is converted into a digital value by the human/D conversion section (17) and stored in the measurement image memory section (19). This measurement image memory section (19) stores data at a density several times to several tens of times higher than that of the display image memory section (18) for scanning lines. Collect about 2048 points. However, data is not collected over the entire screen, and the cursor (25a
), only any location specified in (25b).

The image data of the portions specified by the cursors (25a) and (25b) are sequentially transferred to the measurement image memory section (1).
9), but usually Si (Signal/N
o1se) For signals with poor ratios, integration processing is performed (blocks (26) and (27) in Figure 4). this is,
This method involves scanning the same location repeatedly, adding the data at the same point one after another, and finally standardizing by dividing by the number of additions. On the other hand, for signals with a good S/N ratio, the integration process is not performed and the sampling is completed once (block (26) in Figure 4).Next, noise is removed (block (28) in Figure 4). ).As noise removal methods, the smoothing method (block (29) in Figure 4) and F
FT (star ast Fourier Transrorm
) method. When removing noise using the above FFT method, the data imported from the measurement image memory section (19) (see Fig. 5 (a)) is subjected to Fourier transformation (block (30) in Fig. 4), and frequency analysis is performed to eliminate noise. Cut high frequency components to remove (block (3t) in Figure 4)
.

Then, the data from which the high frequency components have been cut is subjected to inverse Fourier transform to reproduce the waveform (block (32) in Figure 4).
). Through this processing, noise can be removed from the original waveform, as shown in FIG. 5(b).

On the other hand, the smoothing method uses one line of data (0<j≦n)
This method calculates an arbitrary point by multiplying several data points before and after an arbitrary point by a coefficient, adding the products, and dividing by the sum of the coefficients for normalization.

That is, when considering k-pixel smoothing, if the third data is d (j), the data of the points before and after %d (J) are multiplied by the coefficient ai (0≦i≦k), This method calculates the value of d(j) by taking the total sum, dividing it by the sum of coefficients, and normalizing it.

Through this processing, a waveform (FIG. 6(b)) with noise removed compared to the original waveform (FIG. 6(a)) is obtained.

As shown in FIGS. 7 and 8, for the waveform from which noise has been removed by any of the above methods, the discrimination area for measuring the inclination angle is specified using the cursors (25a) and (25b) (the fourth Figure block (33)).

The discrimination area can be identified by the fact that the voltage value of the waveform corresponding to the inclination angle measurement site, that is, the resist pattern part (area P in Figure 3) is larger than other parts. The maximum value (51) and minimum value (52) at one side edge shown in FIG. 4 are determined (block (34) in FIG. 4). Therefore, the maximum value (51) and the minimum value (
52) arbitrarily 2 points between (56), (57)
and specify the range for linear approximation (block (35) in Figure 4). Next, these two points (56), (5
7) Using the least squares method, the regression line (5
9) (block (36) in Figure 4).

Furthermore, the average of the data between the minimum value (52) and the point (57), that is, the data of the flat part, is determined to determine the average level (61) of the base (block (37) in Figure 4) 0
Next, the regression line (59) and the base mean level (
Find the intersection (53) of 61). In the same way, the regression line (74) at the other side edge and the average level (75) at the base are determined, and their intersection point (76) is calculated (
Figure 4 block (38)). The above intersection (53), (
The position of 76) is displayed on the CRT (14), and the interval (number of pixels) between the two is calculated, multiplied by the size per pixel determined by the magnification switching unit (9), and converted into a size (
Figure 4 block (39)). Then, click the cursor (2)
sa), (zsb) spans multiple scanning lines, the same process is repeated for another line (
Figure 4 block (40)). Based on the dimensions indicating the width of the resist pattern P obtained for each line, the inclination angle α of the resist pattern P shown in FIG.
Find β. This resist pattern P is formed on a silicon substrate WS.

It is assumed that the thickness H and width of this resist pattern P are known. By the way, when an electron beam of radius r is irradiated onto a wafer at an accelerating voltage of several keV, the electron beam penetrates a depth d (middle o, sr) below the surface.
It is said that the amount of secondary electron beam emitted is approximately proportional to the cross-sectional area of the sphere of the electron beam cut by the surface. Therefore, the cross-sectional area of the electron beam cut by the end of the resist pattern P, that is, the amount of secondary electron emission, becomes a function of the inclination angles α and β.

Therefore, by storing in advance the shimmering waveforms (see Fig. 10) of various inclination angles and thicknesses H in the waveform storage section (22a) and determining the degree of coincidence with the detected waveform shown in Fig. 8, The actual inclination angles α and β of the resist pattern P can be determined. Therefore, first, the thickness H1 width D (determined by the process described above) and the inclination angle are initialized in the waveform storage section (22a) (block (41) in FIG. 4).

Then, based on the above initial values, a
In other words, a waveform whose amplitude is almost equal to the detected waveform is calculated (block (42) in FIG. 4).

For this purpose, the maximum value (51) and minimum value (52) of the detected waveform shown in FIG.
It works best by matching the Next, the simulation waveform is linearly approximated in the same manner as the detected waveform to obtain the regression line L1 (block (43) in Figure 4).
Find the difference in slope ΔL from block (4) in Figure 4.
4)) Then, the absolute value of this slope difference ΔL is 1ΔL
l Determine whether the force μ is smaller than the constant value CL. Then, when the constant value CL is small, the inclination angle θ at this time is
Let this be the inclination angle α of the resist pattern P. However, when it is larger than the constant value CL, block (4) in Figure 4
Returning to the process of 2), a different inclination angle θ is initialized again, and the same process is performed (block (45) in Figure 4).
). Thus, the above processing of block (44) is repeated until the inclination angle α is determined. Next, regarding the inclination angle β,
Regression line (74) and maximum and minimum values P3. P4 is obtained using the same procedure as for the inclination angle α. The obtained inclination angles α and β are displayed on the CRT (t4).

As described above, the inclination angle measurement device of this embodiment (according to which, for example, the inclination angle of a resist pattern edge formed on a silicon wafer and having a line width of about 1 μm) with respect to the wafer plate surface can be measured in a non-destructive manner. Moreover, it can be determined non-contact, with high precision and high efficiency.

Note that the width of the resist pattern P in the above embodiment may be measured visually by a measurer while looking at the CRT (14). Furthermore, the inclination angle measurement pair is not limited to resist, and may be of any type.

Furthermore, in order to increase secondary electron emission, the wafer side may be tilted by a certain amount in advance with respect to the electron beam, and the tilt may be finally canceled out.

〔Effect of the invention〕

According to the inclination angle measuring device of the present invention, for example, the inclination angle of a pattern edge formed on a silicon wafer and having a line width of about 1 μm can be determined non-destructively and non-contact, with high precision and high efficiency. be able to.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram of an inclination angle measuring device according to an embodiment of the present invention, FIG. 2 is an explanatory diagram showing division of an image signal obtained by the inclination angle measuring device of FIG. Figure 3 is a diagram showing cursor settings on a CRT, Figure 4 is a flowchart showing the tilt angle measurement procedure using the tilt angle measuring device in Figure 1, and Figures 5 and 6 are image signals before noise removal and noise removal. Graphs showing subsequent image signals, FIGS. 7 and 8 are graphs for explaining how to determine the pattern width, FIG. 9 is a cross-sectional view of the resist pattern, and FIG. 10 is a diagram showing simulation waveforms. (1)...SEM main body (image signal output means). (22)...CPU section (tilt angle calculation means). (22a)...Waveform storage section. L-111-i ■ -Mo -□--Yo Figure 2 P Figure 3 Figure 5 Figure 7 y Giant shoulder blade Figure 8 Figure 9 Figure 10

Claims (1)

[Claims] An image signal that outputs an image signal indicating an end shape waveform of the pattern based on a secondary electron beam obtained when an electron beam is irradiated onto a measurement sample on which a pattern whose end inclination angle is to be measured is attached. an output means; a waveform storage means storing a simulation waveform outputted from the image signal output means in each case where the inclination angle of the end portion of the pattern is different; and a simulation waveform stored in the waveform storage means; An inclination angle calculation means for comparing the end shape waveform of the pattern outputted from the image signal output means and calculating an inclination angle of the end part of the pattern based on the comparison result. Angle measuring device.