KR102305190B1 - Device for measuring 3d step height of sample surface - Google Patents

Abstract

The present invention relates to a device for measuring the 3D step of a sample surface. The 3D step measuring apparatus according to an embodiment of the present invention uses a signal pre-processing unit that pre-processes the interference fringe signal detected from the step measuring device using a preset conversion formula, and the amplitude value for each frequency calculated by the pre-processing unit for the interference fringe signal. A reliability detector that calculates a reliability value, a reliability analyzer that analyzes a reliability range with the reliability value of the interference fringe signal calculated by the reliability detector, and an interference fringe corresponding to a range of high and medium reliability in which reliability values are preset among the interference fringe signals. It includes an interference fringe signal restoration processor that interpolates the signal through band-pass HRIDCT and restores it with high resolution. In measuring the height of the surface of the sample even in a situation where it is difficult to measure the height, it is an apparatus that can determine the height of the surface of the sample by using the intensity of the interference fringe as a weight and averaging the height values. And for the interference fringes classified as low reliability, the height value of the sample surface is not determined from the interference fringe, but the height value is determined by interpolation using the height value of the sample surface measured from the surrounding interference fringes of medium reliability or higher. .

Description

DEVICE FOR MEASURING 3D STEP HEIGHT OF SAMPLE SURFACE

The present invention relates to an apparatus for determining the height of a sample surface by interpolating an interference fringe signal or a sample surface height value of a WLSI (White Light Scanning Interferometer) by a weighted average method, and determining the peak position of a WLSI interferometer. It relates to a device for measuring the 3D step of the sample surface that can determine the optimal height even in difficult situations.

In the manufacturing industry in many fields including the display field, measuring the height difference of the sample surface is one of the most important processes in manufacturing process management. Accordingly, a 3D step measuring instrument such as a White Light Scanning Interferometer (WLSI) is mainly used to measure the height of a sample or a sample surface in a non-destructive manner.

As an example, the WLSI irradiates white light at each height while moving the interference lens in the height direction, and the interference fringe signal ( Interferogram) is detected sequentially. Then, by sequentially analyzing and processing the detected interference fringe signal, the surface height is measured in a non-destructive manner, and measurement result data is derived. This WLSI detects the peak position of the interference fringe signal, that is, the time and position at which the amplitude of the interference fringe signal is the highest, derives the spatial position including the height of the sample surface, and arranges the result data in 2D or 3D shape. derive

However, the conventional WLSI is very sensitive to vibration, and when vibration occurs during height measurement, when the interference lens moves in the height direction, the distance between the interference lens and the sample surface does not change constantly, so it is difficult to measure the height.

In addition, since the conventional WLSI is greatly affected by the structural characteristics of the sample or the sample surface, there is a problem in that the reliability of the height measurement result of the sample or the samples having a structure with many inclinations or curves is greatly reduced.

1 is a view showing the measurement result of the height of a specific sample measured through the conventional WLSI.

As shown in FIG. 1 , in the case of a plane in which the sample or the sample surface is perpendicular to the detection direction of the interference fringe signal, since the reliability of the height value calculated from the measured interference fringe is high, a constant result can be obtained, but the measurement is discontinuous. A special algorithm is required to measure the height with high precision from the interferometric fringes.

Also, since the diffuse reflection ratio of the interference fringe signal is increased in the curved portion of the sample or the sample surface having a large slope, the error rate of the sample surface height value measured from the interference fringe signal detected in the inclined or curved portion is inevitably increased. Therefore, in the prior art, the height data of the sample surface was filtered with a specific filter (eg, Median Filter) and used for height measurement. had to be lower.

In the present invention, a method for evaluating reliability from interference fringe data is devised, and the height is determined by applying an appropriate height determination method based on this. We want to obtain reliable data by applying the decision method.

Specifically, the present invention restores an interferogram of WLSI to a high-quality signal through FFT (Fast Fourier Transform) and Band-pass HRIDCT (High Resolution Inverse Discrete Cosine Transform), and uses a weighted average to restore the sample surface It is to provide a 3D step measuring device that can be used as sample surface height measurement data by determining the height of the .

In addition, the present invention provides a standard for evaluating the reliability of an interference fringe signal, and for an interference fringe that is difficult to trust, interpolates from a high reliable height value in the vicinity to determine a height value, and 3D that can be used as sample surface height measurement data To provide a step difference measuring device.

The object of the present invention is not limited to the above-mentioned object, and other objects and advantages of the present invention not mentioned can be understood by the following description, and will be more clearly understood by the examples of the present invention. Moreover, it will be readily apparent that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

3D step measuring apparatus according to an embodiment of the present invention for achieving the technical problem as described above through the interference fringe signal detected from the step measuring device through FFT (Fast Fourier Transform) and Band-pass HRIDCT (High Resolution Inverse Discrete Cosine Transform) Reconstruct a high-quality signal and determine the location of the sample surface using a weighted average. and. The height value is determined by the interpolation method by preparing a standard that can predict erroneous measurement.

To this end, the 3D step measuring apparatus according to an embodiment of the present invention includes a signal preprocessing unit that pre-processes the interference fringe signal detected from the step measuring device using a preset conversion equation, and the amplitude value for each frequency calculated by the preprocessing unit for the interference fringe signal. A reliability detector that calculates a reliability value using The interference fringe signal restoration processing unit that performs band-pass HRIDCT on the interference fringe signal to restore high resolution, calculates the position weighted average by weighting the strength of the interference fringe signal having a value greater than or equal to the predetermined signal strength for high and medium reliability data. Since it is difficult to measure the height of the interference fringe signal corresponding to the low reliability range in which the reliability value of the interference fringe signal is set in advance, the height calculated from the surrounding high-reliability and medium-reliability fringes is difficult to measure. It may be configured to include a low-reliability data processing unit that interpolates using a value to determine the height of the corresponding position.

The apparatus for measuring the 3D step of the sample surface according to an embodiment of the present invention having the above technical characteristics can obtain information on the height of the sample surface even in the WLSI interference fringe, in which it is difficult to determine the height. In other words, even if vibration occurs during WLSI interference fringe measurement and there is a deviation in the interference fringe peak position, the weighted average of the interference fringe strength around the peak signal is weighted to determine the representative value of the sample surface height to reduce noise such as vibration. The height value can be determined sensitively.

In addition, even if the sample surface is inclined rather than perpendicular to the optical axis and thus an interference fringe produced by successive height changes, a representative value of the sample surface height can be determined by the weighted average method. In addition, since the sample surface is rough, the representative value of the sample surface height can be determined by the weighted average method even in the interference fringes formed by the surfaces of various heights.

In addition, it is expected that the tendency of thickness distribution for each location can be observed by determining the representative value of the sample surface height using the weighted average method in the interference fringes formed by the structure with the transparent thin film on the sample surface. In particular, the non-measurable area may determine the height of the location by interpolating from the height value of the surrounding valid area.

1 is a view showing a height measurement result of a specific sample measured through a conventional WLSI.
2 is a block diagram specifically illustrating a 3D step difference measuring apparatus according to an embodiment of the present invention.
FIG. 3 is a diagram for explaining a method of detecting an interference fringe signal of the WLSI scanner shown in FIG. 2 .
4 is a waveform diagram illustrating an interference fringe signal detected by the WLSI scanner shown in FIG. 2 and input to a signal preprocessor.
5 is an interference fringe signal obtained by reconstructing the high-reliability, medium-reliability, and low-reliability interference fringes shown in FIG. 4 at high resolution by the signal restoration processing unit.
FIG. 6 is a diagram illustrating an example in which a representative value of a sample surface height is determined by a position weighted average using the strength of the high reliability and medium reliability interference fringe signal shown in FIG. 5 as a weight.
FIG. 7 is a diagram illustrating a 3D step measurement result derived from the data calculating unit and the low-reliability data processing unit of FIG. 2 .

The terms or words used in the present specification and claims should not be construed as being limited to their ordinary or dictionary meanings, and the inventor may properly define the concept of the term in order to best describe his invention. Based on the principle that can be, it should be interpreted as meaning and concept consistent with the technical idea of the present invention. In addition, the configurations shown in the embodiments and drawings described in the present specification are only one of the most preferred embodiments of the present invention, and do not represent all the technical spirit of the present invention, so they can be substituted at the time of the present application It should be understood that various equivalents and modifications may be made.

Hereinafter, an apparatus for measuring a 3D step of a sample surface according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

2 is a block diagram specifically showing a 3D step difference measuring apparatus according to an embodiment of the present invention. 3 is a diagram for explaining a method of detecting an interference fringe signal of a WLSI scanner.

First, referring to FIG. 2 , the 3D step measurement apparatus 200 includes a signal preprocessor 210 , a reliability detector 220 , a reliability analyzer 230 , an interference fringe signal restoration processor 240 , and a data calculator 250 . ), and a low-reliability data processing unit 260 .

Specifically, the signal preprocessor 210 pre-processes the interference fringe signal detected from the step measuring device such as the WLSI scanner 100 using a preset conversion equation, thereby extracting the amplitude and phase components of the interferogram signal. do. Here, a step measuring device such as the WLSI scanner 100 detects an interference fringe signal as a surface height detection signal of a sample or a sample, which is a surface height measurement target.

Referring to FIG. 3 , the step measuring device such as the WLSI scanner 100 irradiates white light at a preset height to the surface of a sample or samples while moving the interference lens in the height direction. At this time, the WLSI scanner 100 or the like irradiates white light at each index point preset according to a position or a timer, and the optical path between the measurement light reflected from the measurement sample or the surface of the sample and the reference light reflected from the reference mirror. The interference fringe signal generated by the difference is detected. In this way, the interference fringe signal detected from the WLSI scanner 100 or the like is transmitted to the signal preprocessor 210 .

Accordingly, the signal pre-processing unit 210 pre-processes the interference fringe signal detected by the step measuring device in a method (eg, Discrete Fourier Transform method) according to the preset Equation 1 below, for each index (or , by frequency), calculate _{the amplitude (absolute value of f k} ) and phase ( _{declination angle of f k} ) of the interference fringe signal.

[Equation 1]

Here, k is the frequency of the interference fringe signal.

_{According to the Fourier coefficient f k} of the interference fringe signal detected by the index or frequency obtained by Equation 1 above, the intensity value of the interference fringe according to the position set for each index can be restored through the Inverse Discrete Fourier Transform of Equation 2 below. can

[Equation 2]

The reliability detector 220 calculates a reliability value for the interference fringe signal by using the amplitude value for each frequency calculated by the signal preprocessor 210 . That is, the reliability detector 220 may calculate the reliability value of the obtained interference fringe signal by applying a preset reliability calculation formula using the amplitude value for each frequency calculated by the signal preprocessor 210 .

In order to calculate the reliability value of the interference fringe signal, the reliability detector 220 calculates the reliability value C of the interference fringe signal using the preset Equation 3 below.

[Equation 3]

Here, A _k (Absolute Value (Amplitude) of f _k ) is an amplitude value for each index obtained by FFT (Fast Fourier Transform) of the interference fringe signal. In addition, A ₀ is the sum of the strengths of the measured interference fringe signals. Accordingly, when divided by the number of intensity values of the obtained interference fringe signal, the average intensity of the corresponding interference fringe signal may be obtained. Here, even if the brightness of the plane image state in which the interference fringe signal is detected is different for each pixel position, the _{normalized value using A o} is used as the classification standard, so that a consistent standard can be applied to the entire area.

The reliability analysis unit 230 analyzes the reliability range with the reliability value C of the interference fringe signal calculated by the reliability detection unit 220 . At this time, the reliability analysis unit 230 compares the reliability measurement value (C) of the interference fringe signal with a preset reliability index value, and analyzes the reliability range of the interference fringe signal into high reliability, medium reliability, and low reliability ranges, respectively, and can be classified.

4 is a waveform diagram illustrating an interference fringe signal detected by the WLSI scanner shown in FIG. 2 and input to a signal preprocessor.

Referring to Table 1 below together with FIG. 4 , the reliability analysis unit 230 determines that the reliability measurement value (C) of the preprocessed interference fringe signal is in a preset high reliability index value range (for example, in a range of 0.075 or more). If included, the corresponding interference fringe signal is classified and set as data with high reliability.

[Table 1]

And, if the reliability measurement value (C) of the preprocessed interference fringe signal is in a range of medium reliability index values, for example, 0.05 or more and less than 0.075, the reliability analysis unit 230 classifies and sets the interference fringe signal as medium reliability data. do.

In addition, when the reliability measurement value (C) of the preprocessed interference fringe signal is included in a preset index value range of low reliability, for example, less than 0.05, the reliability analysis unit 230 converts the interference fringe signal to low reliability. Classify and set as data of

Here, the reliability classification criterion value is a sample preset using at least one WLSI, for example, according to a result of calculating the surface of a column spacer or a photo spacer of a display panel. It may be preset in consideration of an experimental value or an average value.

In the present invention, an example in which the range of 0.075 to 0.05 is applied as the reliability classification reference value is presented, but the reliability classification reference value can be used and changed by presetting a specific sample or a recipe according to the state of the sample.

The interference fringe signal restoration processing unit 240 interpolates and restores the interference fringe signal according to a preset conversion equation according to a reliability classification range of the interference fringe signal classified into high reliability and medium reliability data.

FIG. 5 is an interference fringe signal obtained by reconstructing the high-reliability, medium-reliability, and low-reliability interference fringes shown in FIG.

The signal restoration processing unit 240 interpolates and transforms the interference fringe signal classified into high reliability and medium reliability data in a manner according to Equation 4 below to restore high resolution. Here, it is assumed that the signal restoration method according to Equation 4 used is a band-pass HRIDCT (High Resolution Inverse Discrete Cosine Transform). In addition, Equation 4 can be derived from Equations 1 and 2 under the condition that N is an even number and x _j is a real number.

[Equation 4]

The interference fringe signal restoration processing unit 240 restores the interference fringe signal according to the position represented by a continuous real number j with high resolution using the amplitude and phase for each frequency calculated by the signal preprocessing unit in the manner according to Equation 4 above. can do. In the present invention, an example in which k (frequency) is taken in the range of 85 to 140 is illustrated, but the applicable range may be changed and applied depending on the characteristics of the light source or the type and state of the sample.

FIG. 6 is a diagram illustrating an example in which a representative value of a sample surface height is determined by a position weighted average using the strength of the high reliability and medium reliability interference fringe signal shown in FIG. 5 as a weight. Specifically, FIG. 6 shows the values obtained by calculating the weighted average position of the high reliability and medium reliability interference fringe signals shown in FIG. 5 by using Equations 5 and 6 below, respectively.

Referring to FIG. 6 , the data calculating unit 250 calculates the interference fringe signal reconstructed in high resolution by the interference fringe signal restoration processing unit with respect to the interference fringe signal classified into high reliability and medium reliability data, respectively, in Equations 5 and 6 The height value is calculated according to the weighted average method.

[Equation 5]

Here, the d value is a dark level, which is set to prevent an interference fringe signal having a predetermined intensity or less from affecting the weighted average value. In this embodiment, a value of 0.02 is set for high reliability data.

When determining the location of the sample surface, the data calculator 250 may determine the height value less sensitive to noise by averaging the height values around the maximum value of the interference fringe and determining the height as a representative value of the heights.

Medium-reliability data are generally interference fringes formed by surfaces of several heights due to the slope or roughness of the sample surface, and are characterized by weaker strength than high-reliability interference fringe signals and spread over a wider range of height values.

[Equation 6]

Here, the d (Dark Level) value is an example of setting 0.009 for medium-reliability data, but it is preferable to set it to a small value compared to high-reliability data. The reason is that, as shown in FIG. 5 , the interference fringe signal of medium reliability has relatively weak strength and is spread over a wide range of height values.

On the other hand, the low reliability data processing unit 260 uses the height value measured at the location where the interference fringe classified as low reliability is located at the location where the interference fringe classified as high reliability or medium reliability is located. It is determined by interpolation by applying Equation 7.

Therefore, when erroneous measurement is expected by analyzing the reliability of the interference fringe signal, the height is not measured using the low-reliability fringe data, but interpolated from the height data measured from the surrounding high-reliability or medium-reliability fringes to determine the total data. can improve the reliability of

[Equation 7]

here,

이다.

am.

FIG. 7 is a diagram illustrating a 3D step measurement result derived from the data calculating unit and the low-reliability data processing unit of FIG. 2 .

Referring to FIG. 7 , the data calculating unit 250 measures the height of the sample surface by reconstructing the interference fringes to a high resolution at a location where the interference fringes having high reliability or higher reliability are measured, and then performing a weighted average to measure the height of the interference fringes with low reliability. At the location where the fringe data is measured, the 3D step measurement result in which the height value is determined by interpolation using the height data measured from the surrounding high- and medium-reliability interference fringes can be seen.

Through this method, the optimal height value can be obtained even when it is difficult to determine the height of the sample surface by the peak position of the interference fringe when the peak position error of the interference fringe occurs due to vibration or the geometric structure of the sample surface is complicated. It can be determined to measure the 3D step.

As described above, the 3D step difference measuring apparatus for the sample surface according to an embodiment of the present invention pre-processes the WLSI interference fringe signal according to a preset conversion equation to determine its reliability, and determines the reliability of the WLSI interference fringe signal according to the result of the determination. At the location where the interference fringes of reliability are measured, the interference fringe signal can be restored with high resolution and the height value of the sample surface can be determined by the weighted average of the height values. Accordingly, even when it is difficult to measure the height of the sample surface according to the peak position of the interference fringe signal at the corresponding position due to vibration or diffuse reflection caused by the complex geometry of the sample surface during measurement, it is possible to determine the optimal height value to measure the 3D step. can

It is to be understood that the foregoing embodiments are illustrative in all respects and not restrictive, and the scope of the present invention will be indicated by the appended claims rather than the foregoing detailed description. And all changes and modifications derived from the meaning and scope of the claims to be described later as well as equivalent concepts should be construed as being included in the scope of the present invention.

100: WLSI Scanner
200: 3D step measuring device
210: signal preprocessor
220: reliability detection unit
230: reliability analysis unit
240: interference fringe signal restoration processing unit
250: data output unit
260: low-reliability data processing unit

Claims (7)

a signal pre-processing unit for pre-processing the interference fringe signal detected from the step measuring device using a preset conversion equation;
a reliability detection unit for calculating a reliability value for the interference fringe signal by using the amplitude value for each frequency calculated by the preprocessor;
a reliability analysis unit for analyzing a reliability range with the reliability value of the interference fringe signal calculated by the reliability detection unit; and
an interference fringe signal restoration processing unit for performing band-pass HRIDCT on an interference fringe signal corresponding to a range of high reliability and medium reliability in which a reliability value is preset among the interference fringe signals to restore a high resolution;
and a data calculator configured to determine a height value by weighted average of high-resolution reconstructed signals of interference fringe signals corresponding to high and medium reliability ranges in which reliability values are preset among the interference fringe signals;
A device for measuring the 3D step of the sample surface.
The method of claim 1,
The signal preprocessor
The interference fringe signal detected by the step measuring device is pre-processed in a manner according to Equation 1 below, and the amplitude and phase of _{the Fourier coefficients f k} and f _{k of the interference fringe signal are extracted for each frequency (by index).} and
[Equation 1]

Here, k is the frequency component of the interference fringe signal, and the reliability of the measured data is evaluated using amplitude and phase information for each frequency (index) of the detected interference fringe signal, or the intensity value of the interference fringe for each set position or timing. to detect,
A device for measuring the 3D step of the sample surface.
3. The method of claim 2,
The reliability detection unit
Using Equation 2 below, the reliability measurement value (C) is calculated using the absolute value A _{k of the Fourier coefficient f k for each} frequency of the interference fringe signal preprocessed according to Equation 1 above,
[Equation 2]

The A ₀ is the sum of all interference fringe strengths measured while moving the z-axis, and even if the average strengths of the interference fringes are different depending on the positions of the x and y axes, the reliability of the interference fringes is calculated by normalizing to the sum or average strength,
A device for measuring the 3D step of the sample surface.
4. The method of claim 3,
The reliability analysis unit
When the reliability measurement value of the interference fringe signal is included in a preset index value range of high reliability, classifying and setting the interference fringe signal as high reliability data;
When the reliability measurement value of the interference fringe signal is included in the index value range of medium reliability, classifying and setting the interference fringe signal as data of medium reliability,
classifying and setting the interference fringe signal as low-reliability data when the reliability measurement value of the interference fringe signal is less than a preset low-reliability classification reference value;
A device for measuring the 3D step of the sample surface.
5. The method of claim 4,
The interference fringe signal restoration processing unit
The interference fringe signal classified as data with high and medium reliability is interpolated and transformed by the method (Band-pass HRIDCT) according to Equation 4 below to restore high resolution,
[Equation 4]

Restored under the condition that N is an even number and j is a continuous real number,
A device for measuring the 3D step of the sample surface.
5. The method of claim 4,
The data calculation unit
With respect to the interference fringe signal classified into the high-reliability data and the medium-reliability data, the weighted average positions of the high-reliability and medium-reliability interference fringe signals reconstructed at high resolution are calculated using Equations 5 and 6 below, respectively. to detect the height data,
[Equation 5]

Here, the d value is a dark level, which is set to prevent the interference fringe signal having a certain intensity or less from affecting the weighted average value. It was set to 0.02 for high reliability data,
[Equation 6]

Here, the d value is a dark level, which is set to prevent the interference fringe signal of less than a certain intensity from affecting the weighted average value. It is set to 0.009 for medium reliability data,
A device for measuring the 3D step of the sample surface.
5. The method of claim 4,
A low reliability data processing unit for interpolating the interference fringe signal whose reliability value calculated for the interference fringe signal is in the low reliability range using the height values measured from the surrounding high and medium reliability interference fringes;
[Equation 6]

The low-reliability data processing unit
With respect to the interference fringes classified as data with low reliability by the reliability analysis unit,
In determining the height of the position where the interference fringe signal is measured, the height is determined by interpolating using the height data including the height values calculated from the surrounding high and medium reliability interference fringes using Equation 6,
A device for measuring the 3D step of the sample surface.

Images