JP3960862B2 - Height measurement method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method of measuring surface information of a sample by scanning the sample with light through an optical system of an optical microscope, and more particularly to a height measuring method using a confocal scanning optical microscope.
[0002]
[Prior art]
A confocal scanning optical microscope illuminates a sample in a dot-like manner, collects transmitted or reflected light from the sample on a pinhole, and then detects the intensity of light transmitted through the pinhole with a photodetector. . By performing such light detection over the entire sample surface, a two-dimensional image of the sample surface can be obtained.
[0003]
In this case, when the relative position (Z) between the objective lens and the sample is changed in the optical axis direction, the luminance (I) changes, and when the light from the light source is focused on the sample surface (when the sample surface is in focus). , I is maximized (hereinafter, the relationship between Z and I is referred to as “IZ curve”).
[0004]
In height measurement using a confocal scanning optical microscope, the Z at which I is maximum on the I-Z curve is the height at that point. In this case, in order to improve the measurement accuracy of the height, it is necessary to change Z finely and detect the maximum value of I, which takes time.
[0005]
Therefore, in the method disclosed in Japanese Patent Application Laid-Open No. 9-68413, the IZ curve is approximated by an expression such as a quadratic expression or a Gaussian function using I at a plurality of Zs. The Z at which I is maximized in this approximate expression is calculated from the coefficient of the approximate expression, and is changed to the Z at which I is maximized by the IZ curve to obtain the height at that point. Thereby, the height information is obtained with high accuracy in a wide step (= small number of captured sheets).
[0006]
[Problems to be solved by the invention]
Here, the Z step is determined in consideration of the shape of the IZ curve and the accuracy and time required for the measurement. However, the shape (spread) of the IZ curve is determined by the surface shape of the sample. When the surface is a scattering surface, the IZ curve spreads more than the mirror surface (the spread is specular depending on the surface state at the maximum). About twice as much). At this time, if the spread of the IZ curve is uniform and known in the screen, the step width of Z can be increased corresponding to the spread of the IZ curve, but the surface shape is unknown and I -If the spread of the Z curve is unknown or if there is a distribution in the screen, the Z step width cannot be changed.
[0007]
Also, the extraction of I used for calculating the approximate expression is
(1) Obtain the number of extraction points (N), and extract the maximum value of I and I at N points before and after that.
(2) Obtain a threshold value and extract I that exceeds the threshold value.
Therefore, if the step of Z is fixed, in the extraction method of I above,
(1) Since the portion where the change in I near the peak of the IZ curve is small is sampled, the influence of noise is large.
(2) The number of extracted data points increases and the calculation takes time.
This causes problems.
[0008]
The present invention has been made to solve the above-described problems. Even if the shape of the I-Z curve changes due to the surface shape of the sample, the Z step is not changed, and the coefficient of the approximate expression is calculated. An object of the present invention is to provide a height measuring method for measuring the height at high speed and accurately by calculating the coefficient of the approximate expression of the I-Z curve without increasing the number of I used and suppressing the influence of noise. To do.
[0009]
[Means for Solving the Problems]
In order to solve the above problems, a height measurement method according to the present invention is a height measurement method using a confocal scanning optical microscope, and changes a relative position between a sample and an objective lens at a predetermined interval. Measuring the luminance at a plurality of positions, and evaluating the influence of noise using luminance data at at least three consecutive positions including the maximum luminance among the luminance measurement results at the plurality of positions, Based on the evaluation result, an approximate curve is obtained and the peak position of the luminance is calculated.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram showing a schematic configuration of a confocal scanning optical microscope system applied to a height measuring method according to an embodiment of the present invention. In this embodiment, the surface information is acquired by two-dimensionally scanning the sample using the optical system of the confocal scanning optical microscope 16.
[0011]
The confocal scanning optical microscope 16 shown in FIG. 1 reflects the scanning laser light emitted from the laser light source 18 by the mirror 20 and enters the scanning mechanism 24 through the half mirror 22.
[0012]
The scanning mechanism 24 is connected to a computer 28 via a scanning control unit 26, and is driven and controlled based on a scanning control signal P1 output from the scanning control unit 26 in response to a command from the computer 28. The scanning mechanism 24 focuses the scanning laser light on the sample 36 on the stage 34 through the objective lens 32 set on the revolver 30 based on the scanning control signal P1, and in this state the scanning laser light. Is scanned in the X and Y directions on the sample 36 in the same manner as the raster scan of the television.
[0013]
Reflected light reflected from the sample 36 in the sample scanning by the scanning laser light is guided to the half mirror 22 through the objective lens 32 and the scanning mechanism 24, and is reflected by the half mirror 22 toward the photodetector 40 side.
[0014]
The reflected light reflected by the half mirror 22 passes through a pinhole 38 disposed at a position conjugate with the condensing position of the objective lens and then enters the photodetector 40. The photodetector 40 converts the incident reflected light into an electrical signal corresponding to the amount of light and outputs it to the image processing unit 42.
[0015]
The image processing unit 42 incorporates an image memory 42a composed of, for example, 512 pixels × 512 pixels × 8 bits (256 gradations). The image memory 42 a is connected to the photodetector 40, and can store an electric signal output from the photodetector 40. Further, the image memory 42a is connected to a Z-direction movement control circuit 44 that controls the movement of the stage 34 in the Z direction (that is, the optical axis direction of the scanning laser beam) and scans the scanning laser beam in the Z direction. ing. A count value obtained by counting the number of movements of the stage 34 based on the signal output from the Z-direction movement control circuit 44 is stored in the image memory 42a.
[0016]
The stage 34 is controlled to move in the Z direction by a predetermined amount based on the Z control signal P2 output from the Z direction movement control circuit 44 in response to a command from the computer 28. At this time, the amount of movement per stage 34 is controlled by the computer 28.
[0017]
Further, the setting of the measurement range, the setting of the moving amount of the stage 34 within each measurement range, the control of the image display and the microscope system, and the like are set by the observer via the monitor 46 connected to the computer 28.
[0018]
In the confocal scanning microscope system configured as described above, after the observer places the sample 36 on the stage 34, the micro spot focused on the sample 36 is scanned in the XY directions under the control of the computer 28. To do. At the same time, the stage 34 is moved and controlled in the Z direction at each measurement point (x, y) to perform focusing control on the sample 36. At this time, whether or not the sample 36 is in focus is determined while viewing the image displayed on the monitor 46.
[0019]
Next, the observer sets parameters related to the measurement operation. First, after the measurement range L of the sample 36 and the position Z ₀ of the stage 34 where measurement is started are set by the computer 28, the amount of movement Δ of the stage 34 in one Z scan is set. In the present embodiment, the movement amount Δ is a movement amount when the position of the sample 36 is moved with respect to the focal position of the objective lens 32.
[0020]
When the measurement range L and the amount of movement Δ per stage 34 are set, the number N of movements of the stage 34 is determined according to the relationship L / Δ ≦ N. By the way, since the count value of the number of movements of the stage 34 is stored in the image memory 42a, the number of movements N of the stage 34 is limited to the number of gradations 255 or less of the image memory 42a.
[0021]
When the measurement with respect to the sample 36 is started after setting the measurement range L, the movement amount Δ, and the movement number N as described above, the relative position Z ₁ , Z _{2 ,.} electrical signals _{I 1,} I 2 in, ... _{I n} are detected.
[0022]
A height measurement method according to an embodiment of the present invention applied to the confocal scanning optical microscope system configured as described above will be described with reference to the flowchart of FIG.
First, the luminance is sampled by performing Z scanning, and the maximum luminance value, the luminance at five points before and after that, and the value of the Z counter at which the luminance is maximum are stored (step S1 including step S2 to step S1). The specific contents of steps S2 to S9 are as follows.
[0023]
Initial setting at the start of measurement is performed (step S2). As a specific initial setting, the Z stage is moved to Z ₀ , the counter is reset (0 is substituted for k), the initial luminance value I ₀ is taken, and the value of I ₀ is set to the maximum luminance value _Mc . Store. Next, the Z stage is moved by Δ, the counter value k is incremented, and the luminance Ik is captured (step S3).
[0024]
The luminance I _k is compared with the maximum luminance value M _c (step S4). If I _k is larger than M _c , I _{k is set} to M _c , and the previous luminance L _{1 is set} to M _a1. The previous luminance L ₂ is stored in M _a2 and k is stored in M _d (step S5), and the process proceeds to step S8. In step S4, if I _k is a value equal to or less than M _c , it is determined whether the counter value is k = M _d +2 (step S6). If k = M _d +2, the luminance I _k is set. to mb2, and stores each _{L 1} to _{M b1} (step S7), and the process proceeds to step S8. If not k = M _d +2 in step S6, the process proceeds to step S8.
[0025]
Then, in step S8, 1 previous luminance _{L 1} 2 one to the previous brightness _{L 2,} respectively store the luminance _{I k} 1 single Previous luminance _{L 1.} Next, when the counter value reaches the end value N, luminance sampling and extraction of five points around the maximum value are completed (step S9). If the counter value has not reached the end value in step S9, the processing from step S3 is repeated.
[0026]
When the luminance sampling is completed, the coefficients of the approximate curve are obtained by the method of least squares using the data M _a2 , M _a1 , M _c , M _b1 , and M _b2 around the maximum value (step S10). The magnitude of the influence of noise is determined according to the quadratic coefficient a of the approximate curve obtained in step S10 (step S11). The magnitude of the influence of noise in this case will be described with reference to FIG. 3. FIG. 3 is a diagram for explaining the influence of noise. In FIG. 3, sample A is an IZ curve when the influence of noise is small, and sample B is an IZ curve when the influence of noise is large.
[0027]
In step S11, when the coefficient a is negative (a <0), the influence of noise is small because the quadratic function is convex upward as shown in the 5-point approximate curve of the sample A in FIG. The peak position is calculated using the coefficient obtained in step S10 (step S12). However, in step S11, when the coefficient a is not negative and positive, it is convex downward like the 5-point approximate curve of the sample B, and when it is 0, it is a straight line. Therefore, in such a case, the coefficients of the approximate expression are calculated using M _a2 , M _c , and M _b2 without using the luminances M _a1 and M _b1 on both sides of the maximum value M _c (step S13). Based on this coefficient, the peak position is calculated (step S14). By doing in this way, an approximate curve as shown in the three-point approximate curve of sample B is obtained.
[0028]
According to the method as described above, even if the Z scanning range is widened, the stored data is 6 frames, and the memory consumption can be reduced.
[0029]
In the above description, in order to simplify the description, a <0 (convex downward) is used as a condition for determining the influence of noise. However, the approximate expression may be too wide even when convex upward. Therefore, it is preferable to use a more appropriate threshold value (<0).
[0030]
If there is no noise and the peak position is calculated using the result of step S10, the result is always within ± 1/2 of the Z step. Therefore, in step S10, the calculation is performed up to the peak position. In step S11, when the calculated peak position is in the range of ± 1/2 × Δ, the peak position calculated in step S10 in step S12. In other cases, it is better to obtain the peak position in the procedure after step S13 (note that noise is unavoidably added to the sampled luminance, so the reference in step S11 is somewhat broadened, and ± Δ or ± 2 × Δ may also be used).
[0031]
If the luminance I _1, I 2 in all positions of the Z scanning range in step S4 if there is room in the memory, a _... I _n to store in the memory, the processing in step S8 from step S5 is unnecessary.
[0032]
In this case, the maximum luminance value and N points before and after that are extracted in step S10. In this case, the number of extraction points may be three or more, and is not limited to three or five points.
[0033]
Further, when it is evaluated that the noise is large in step S11, the re-extraction of the luminance data in step S13 is not performed from the luminance extracted in step S10, but in all of the Z scanning ranges temporarily stored in step S4. luminance I _1, I 2 at position _... before and after sandwiching the maximum value I _n, may be extracted 5 points every other.
[0034]
Thus, the luminance I _1, I 2 at all positions of the Z scanning range, _... If you to store the I _n the memory, processing time of sampling the luminance (step S1) is reduced more, fast The degree of freedom of noise evaluation and luminance data re-extraction is increased such that sampling can be performed and the number of luminances used for peak position calculation can be made constant regardless of the influence of noise.
[0035]
The following invention is extracted from each of the above embodiments.
The height measurement method according to the present invention is a height measurement method using a confocal scanning optical microscope, and measures the luminance at a plurality of positions while changing the relative position of the sample and the objective lens at a predetermined interval. Then, of the luminance measurement results at the plurality of positions, the influence of noise is evaluated using luminance data at at least three consecutive positions including the maximum luminance, and an approximation curve is calculated based on the noise evaluation results. The luminance peak position is calculated and calculated.
[0036]
In the above height measurement method, when the approximation curve is obtained, if the influence of noise is small, an approximation curve is obtained using luminance data at at least three consecutive positions including the maximum luminance, and the noise When the influence is large, the approximate curve is obtained by using luminance data excluding at least the luminance data at a position adjacent to the position of the maximum luminance among the measured luminance data.
[0037]
In each of the height measurement methods described above, when obtaining the approximate curve, a width of an approximate expression is used as an evaluation criterion for the noise. Further, the recalculation of the approximate curve is characterized by using three points of the center and both ends among the extracted five points.
[0038]
The present invention is not limited to the above-described embodiments. Of course, various modifications can be made without departing from the scope of the present invention.
[0039]
【The invention's effect】
Even if the shape of the IZ curve changes depending on the surface shape of the sample, the Z step is not changed, the number of I used for calculating the coefficient of the approximate expression is not increased, the influence of noise is suppressed, and the I- By calculating the coefficient of the approximate expression of the Z curve, the height can be measured quickly and accurately.
[Brief description of the drawings]
FIG. 1 is a diagram showing a schematic configuration of a confocal scanning optical microscope system applied to a height measuring method according to an embodiment of the present invention.
FIG. 2 is a flowchart showing the operation of a height measurement method according to an embodiment of the present invention applied to a confocal scanning optical microscope system.
FIG. 3 is a view for explaining the height measuring method of the present invention.
[Explanation of symbols]
16 ... confocal scanning optical microscope 18 ... laser light source 20 ... mirror 22 ... half mirror 24 ... scanning mechanism 26 ... scanning control unit 28 ... computer 30 ... revolver 32 ... objective lens 34 ... stage 36 ... sample 38 ... pinhole 40 ... Photodetector 42 ... Image processing unit 42a ... Image memory 44 ... Z direction movement control circuit 46 ... Monitor

Claims (4)

A height measurement method using a confocal scanning optical microscope,
While changing the relative position of the sample and the objective lens at a predetermined interval, measure the brightness at multiple positions,
Among the measurement results of luminance at the plurality of positions, the influence of noise is evaluated using luminance data at at least three consecutive positions including the maximum luminance ,
A height measurement method, comprising: calculating an approximate curve based on the noise evaluation result and calculating a luminance peak position. In the height measurement method according to claim 1, when obtaining the approximate curve,
When the influence of noise is small, obtain an approximate curve using the luminance data used for noise evaluation,
A height measuring method characterized in that, when the influence of noise is large, an approximate curve is obtained using luminance data excluding at least luminance data at a position adjacent to the position of the maximum luminance among the measured luminance data .   The height measurement method according to claim 1, wherein a second order coefficient of an approximate curve is obtained by a least square method using brightness data at at least three consecutive positions including the maximum brightness, and according to the second order coefficient. A method for measuring the height, characterized by determining the magnitude of the influence of the noise.   4. The height measurement method according to claim 3, wherein when the influence of the noise is determined, if the quadratic coefficient of the approximate curve is negative, it is determined that the influence of the noise is small, and When the coefficient is positive, it is determined that the influence of the noise is large.

Images