JP5933273B2 - Measuring method of surface shape of sample by scanning white interferometer - Google Patents

Description

  The present invention relates to a method for measuring the surface shape of a sample mixed with a dielectric or metal using a scanning white interferometer, and in particular, when measuring the surface shape of a sample mixed with a dielectric or metal. The present invention relates to a method for correcting a phase change of reflected light in a scanning white light interferometer.

  In this specification, the term “surface shape of the sample” is meant to include the concept of the height, level difference, film thickness, and surface roughness of the sample.

  A scanning white light interferometer is a device that uses non-coherent white light as a light source and can measure the surface shape of a sample in a non-contact manner using a Michelson-type or Mirau-type iso-optical path interferometer. Used for measuring the surface shape of a wafer or the like. The principle of the scanning white interferometer is shown in FIG. 1 of the accompanying drawings. Reference numeral 1 denotes a light source, which is a high-intensity white light source. 2 is a filter for white light from the light source 1, 3 is a beam splitter, and 4 is a Michelson interferometer. The Michelson interferometer 4 includes an objective lens 4a, a beam splitter 4b, and a mirror 4c. The Michelson interferometer 4 is provided with a piezo actuator 5 that vertically scans the Michelson interferometer 4. In FIG. 1, reference numeral 6 denotes a CCD camera that forms a light receiving element, and reference numeral 7 denotes a sample holder that supports a sample 8.

  In the apparatus configuration shown in FIG. 1, an interferometer is formed under the objective lens 4a of the optical microscope, and an interference waveform is obtained by scanning the objective lens 4a or the mirror 4c. That is, an interference waveform at each pixel of the CCD camera 6 can be obtained by photographing the intensity of light as a moving image with the CCD camera 6 while scanning the objective lens 4a. Since the peak position of the obtained interference waveform corresponds to the height of the surface of the sample, the surface height can be obtained in the photographed area by obtaining the peak position for each pixel.

  If there is no phase change due to reflection in the two interfering optical paths, the lights strengthen each other with an optical path difference of 0, resulting in the peak of the interference waveform. However, there is not only the sample surface but also a phase change at the time of reflection by a mirror, and in general, the interference waveform is not always maximized when the optical path difference is 0 (the envelope is maximized). Even if there is a phase change at the time of reflection, if the sample is a single substance, the amount of phase change is constant over the entire surface of the sample, so there is no effect on the relative height relationship within the surface, and the phase from the interference waveform is changed to the surface. There is no problem even if the height is calculated. The method using the phase of the interference waveform causes a problem with a plurality of samples made of substances having different complex refractive indexes (see Non-Patent Documents 1 and 2).

  Since the envelope of the interference waveform is not affected by the phase change of the reflected light and becomes maximum when the optical path difference is 0, the surface height may be obtained from the “peak position of the envelope”. The use of “the peak position of the envelope” is effective for a sample made of a plurality of substances having different complex refractive indexes. However, since the envelope is like connecting the apexes of a plurality of peaks in the interference waveform, the width of the spread is wide, so the accuracy of calculating the peak position is low. That is, the method of using the envelope curve is less accurate in calculating the surface height than the method of using the phase of the interference waveform. Therefore, the present inventor previously proposed a method of connecting these two methods to complement each other (see Patent Document 1).

  The interval at which data is collected while scanning the objective lens or mirror is usually 1/5 or less of the interference period. One half of the interference period is called the Nyquist interval, which is 0.4 times or less. As the collection interval is narrower and the number of data is larger, the calculation accuracy of the surface height is improved. However, since there is a problem that the collection time becomes longer, an attempt has been made to obtain the surface height by increasing the collection interval.

  Since the original interference waveform cannot be obtained if the collection interval is wider than the Nyquist interval, the surface height cannot be calculated by a normal method, but several methods for overcoming it have been devised (Patent Document 2, Non-Patent Document). Reference 1). The present inventor also previously proposed a method of calculating the phase and envelope of the collected waveform using the Hilbert transform and obtaining the surface height therefrom (see Patent Documents 3 and 4).

  The waveform obtained when this collection interval is expanded beyond the Nyquist interval (hereinafter referred to as the acquisition waveform) is different from the interference waveform, but its envelope and phase can be calculated using the Hilbert transform, and the envelope is It matches the interference waveform with high accuracy, and its phase is connected to the envelope with a certain relationship, and the phase can be calculated with high accuracy from the beginning, so the surface height can be determined with high accuracy from the phase of the acquired waveform. Can do.

  The phase of the collected waveform whose collection interval is wider than the Nyquist interval is also affected by the phase change of the reflected light on the sample surface, as in the case of the normal interference waveform. In a normal interference waveform, the difference between the “scan position where the phase is 0 = the peak position of the peak of the interference waveform” and the “peak position of the envelope” is the same as the surface height of the sample if the sample surface material is the same. However, since the difference is different for different surface materials, the surface height (= “envelope peak position”) for each surface material from the measured “scanning position where the phase is 0”. ) Was not difficult to calculate.

  However, in the collected waveform when the collection interval is wider than the Nyquist interval, the difference between the “scan position where the phase is 0” and the “peak position of the envelope” is not a constant value, but depends on the sample surface height. Therefore, it is difficult to calculate the surface height by correcting the phase change due to the surface reflection from the “scanning position where the phase becomes 0”.

Japanese Patent Application No. 2011-153435 Japanese Patent No. 2679876 Japanese Patent Application No. 2011-152999 Japanese Patent Application No. 2011-153264

Toru Yoshizawa, “Latest optical 3D measurement”, 2006, Asakura Shoten, Chapter 5, 2 Optical Interferometry, p. 71 Takuma Doi et al., Optics, 20 (9) 603-606, 1991.

  As described above, the calculation accuracy of the sample surface height is improved by using the phase of the waveform collected in the white interferometer. However, if there is a metal on the surface, the phase of the reflected light will change and the sample will be mixed with different substances However, there is a problem that the surface shape cannot be calculated correctly.

  Therefore, the present invention provides a method for measuring the surface shape of a sample with a scanning white interferometer that solves such problems and solves the problems associated with making the collection interval wider than the Nyquist interval in order to shorten the data collection time. It is intended to provide.

In order to achieve the above object, according to the present invention, a beam splitter and a mirror are arranged under the objective lens, and a Michelson type interferometer including the sample surface is constructed, and the distance to the sample or Scanning white interferometer that scans the distance to the mirror with a piezo actuator, records the resulting interference waveform with a CCD camera, records it as movie file data, and measures the surface shape of the sample based on the movie file data In the method of measuring the surface shape of the sample by
Using a CCD camera, the data collection interval is wider than the Nyquist interval (half the interference period), and a large number of video file data is measured.
The slope of the linear relationship between the “envelope peak position” and the “scan position where the phase is 0” of the collected waveform calculated using the Hilbert transform from the waveform thus measured and collected is defined as a.
The expression (“scan position with phase 0” − “envelope peak position”) / (− a + 1) + “envelope peak position” is expressed in units of the collected data array index, which is expressed as By determining each corresponding pixel on the surface and comparing each other, determine whether the sample surface of each pixel is the same type or different from the sample surface of other pixels,
A straight line representing the relationship between “envelope peak position” vs. “scanning position with phase 0” and a straight line “scanning position with phase 0” = “peak position of envelope” intersect at the decimal part of the above formula. In addition, the average value is calculated over a large number of pixels of the same substance for the decimal part, and the average value is determined from a large number of data for the slope a.
The surface shape of the sample is measured by determining the “peak position of the envelope” representing the correct sample surface height from the measured value of “scanning position where the phase is 0” without depending on the phase change of the reflected light. It is said.

  When the data collection interval becomes wider than the Nyquist interval, the collected waveform is different from the interference waveform and is referred to as “collected waveform” in this specification. Since the phase of the collected waveform is also affected by the phase change due to the surface reflection, the calculated value of the sample surface height is affected. The present invention relates to a method for calculating a correct surface height without being affected by the phase change.

Since the envelope obtained from the acquired waveform using the Hilbert transform matches the envelope of the original interference waveform with high accuracy, the peak position is used as a reference. The “peak position of the envelope” corresponds to the sample surface height.
The phase of the acquired waveform is also calculated using the Hilbert transform, and the scanning position where the phase becomes 0 is obtained. If the complex refractive index of the sample surface material is different, the amount of phase change due to reflection is different, the phase of the interference waveform is different accordingly, and the phase of the collected waveform is also different. Is also different.

  If there is a phase change in the interference waveform due to a difference in phase change amount at the time of reflection on the sample surface, a phase change of a constant multiple appears in the collected waveform. Then, in the acquired waveform, a relational diagram of “envelope peak position” vs. “scan position where phase is 0” is shown, and an index of the acquired data array is a unit (the scan position at the time of data acquisition is 1, 2, 3,... N When expressed as), the decimal part differs depending on the amount of phase change due to reflection on the sample surface. Therefore, if this is obtained for each pixel corresponding to the sample surface and compared with each other, it can be determined whether the substance in each pixel is the same or different.

  In the relational diagram of “envelope peak position” vs. “scanning position where phase is 0” regarding the collected waveform, the influence of the phase change of the interference waveform (that is, the difference in the phase change amount due to the sample surface reflection) appears in the slope a. It appears as a certain amount of shift in the vertical axis direction. In the relationship diagram, the straight line representing the relationship between “peak position of envelope” vs. “scanning position with phase 0” and the straight line “scanning position with phase 0” = “peak position of envelope” It intersects with the decimal part. Therefore, an average value is obtained over a large number of pixels of the same substance for the decimal part, and an average value is obtained from a large number of data for the above-mentioned slope a, and using them, the “peak position of the envelope” vs. “ The relationship diagram of “scanning position with phase 0” is accurately determined (that is, the slope a and y intercept of the linear equation are accurately determined), and the relationship is used from the measured value of “scanning position with phase 0”. The "peak position of the envelope", that is, the correct sample surface height can be calculated with high accuracy.

  Therefore, from the “scanning position where the phase is 0” which can be calculated with high accuracy, the “envelope peak position” representing the correct sample surface height can be obtained without depending on the phase change of the reflected light. Therefore, the correct surface shape can be obtained with high accuracy even on the surface of a sample in which substances having different amounts of phase change due to reflection are mixed.

According to the present invention, the following effects can be obtained.
In a waveform collected at intervals wider than the Nyquist interval, a is the slope of the linear relationship between the “peak position of the envelope” and the “scanning position where the phase is 0” calculated using the Hilbert transform.

("Scanning position where phase is 0"-"peak position of envelope") / (-a + 1)
+ “Peak position of envelope”

When the array index is expressed in units, the fractional part corresponds to the phase change amount of the reflected light on the sample surface, so that the substance of the pixel can be specified from the value (in the embodiment, whether it is silicon or copper). Can be identified).
A linear relationship between the “peak position of the envelope” and the “scanning position where the phase is 0” (y intercept determined by the inclination a and the decimal part) is obtained from a large number of measurement data, and the measured value “phase is 0 The calculated value “envelope peak position” (corresponding to the height of the sample surface) is obtained from the linear relationship from the “scanning position”, and the measured value “scanning position with phase 0” is the measured value “envelope Since the measurement accuracy is originally higher than the “peak position”, the sample surface height can be obtained with high accuracy.
Since the above-described linear relationship of the y-intercept determined by the value is adopted by using the fractional part that differs depending on the substance, the correct sample surface height can be obtained by removing the difference in the phase change amount of the reflected light on the sample surface.
-In order to avoid the influence of the phase change of the surface reflected light, it was necessary to obtain the surface height from the envelope in the past. However, in the present invention, the measurement accuracy is improved by one digit or more because it is obtained from the phase of the collected waveform To do.

Schematic which shows the structural example of the scanning-type white interferometer which can be used when implementing this invention. The graph which shows the example of calculation of the interference waveform and collection waveform in case the phase difference (phi) = 0.45 (pi) of the light which interferes in optical path difference 0, and the surface height h = 0 of a sample. The graph which shows the envelope and phase which were calculated | required from the collection waveform in the graph of FIG. The graph which shows the measurement example of the collected waveform, its Hilbert transformation ((circle)), and an envelope. The graph which shows the phase of the collection waveform computed from the collection waveform of FIG. 4, and its Hilbert transform. The graph which shows the example of a measurement of "the peak position of an envelope" and "the scanning position where a phase is 0". The graph which plotted the data of FIG. 6 by "the peak position of an envelope" versus "the scanning position where a phase is 0". Graph showing an example in which ("scanning position where phase is 0"-"peak position of envelope") / 9 + "peak position of envelope" is calculated from the data of Fig. 6 and plotted against x (pixel) . The graph which plotted the decimal part of the data of FIG. The graph which shows what carried out the moving average of the data of FIG. FIG. 8 is a diagram for enlarging FIG. 7 to show a method for obtaining a “envelope peak position” corresponding to the measurement value of “scanning position with phase 0” with high accuracy. The graph which shows the example of the height of the surface of the sample calculated | required with the method of this invention. The figure which expands and shows the silicon | silicone area | region of FIG.

The present invention will be described below with reference to the accompanying drawings.
The interference waveform in the white interferometer is the phase difference between two interfering lights at the scanning position s of the mirror 4c or the objective lens 4a, the surface height h of the sample 8, the wavelength λi, and the optical path difference 0 in the apparatus of FIG. If φ, it is expressed by the following formula (1).
Σ [1 + cos {2π (s−h) / (λi / 2) + φ}] / N (1)
^λi
The total is obtained by changing the wavelength λi, and the total is divided by N. The scanning position where the optical path difference is 0 at the surface height h = 0 of the sample 8 is set to 0 of s.

FIG. 2 is a dotted line graph showing an example in which the center wavelength is 550 nm, the bandwidth is 80 nm, the wavelength λi is changed by 0.1 nm from 510 nm to 590 nm, summed, and divided by the total number N = 800 to calculate the interference waveform. φ = 0.45π and h = 0. The scanning position is waveform data every 1 nm. When φ = 0 and h = 0, the maximum peak of the interference waveform comes to the scanning position 0. However, when φ is different, the phase of the interference waveform is different and the peak of the interference waveform is shifted. However, the position of the envelope itself does not change (see Patent Document 1). That is, the phase of the interference waveform is affected by the phase change of the reflected light on the surface of the sample 8, but the envelope is not affected.

  An example in which the above interference waveform is collected at intervals of 240 nm is shown by ▪ in FIG. The collection interval is 1.75 times the Nyquist interval. The collected data forms a kind of waveform. It can be subjected to Hilbert transform to calculate the envelope and phase of the collected waveform. FIG. 3 shows the envelope (◯) obtained from the waveform indicated by ■ in FIG. 2 and the phase of the collected waveform (× in □). The envelope of the collected waveform matches the envelope of the original interference waveform with high accuracy, and the “peak position of the envelope” corresponds to the surface height h of the sample 8. The relationship between the “scanning position where the phase is 0” and the “peak position of the envelope” are approximately linear, and the surface height h can be calculated from the “scanning position where the phase is 0”. Since the “scanning position where the phase is 0” can be calculated with higher accuracy than the “peak position of the envelope”, the surface height h obtained therefrom is also calculated with higher accuracy (see Patent Document 4).

  As can be seen from FIG. 2, when the phase of the interference waveform changes, the acquired waveform also changes, and the phase also changes. However, the envelope of the collected waveform does not change. The amount of change in the phase of the collected waveform is larger by a certain factor than the amount of change in the phase of the interference waveform. This can be confirmed by using the formula (1) and calculating the results shown in FIGS. 2 and 3 while changing the parameters. This can be confirmed from the experimental results shown below. For this reason, a certain phase change of the surface reflected light of the sample 8 changes the phase of the interference waveform and changes the phase of the collected waveform by a certain value. Accordingly, the amount of change in the phase of the collected waveform is also determined according to the value of the complex refractive index of the surface material. When such a change in the phase of the collected waveform is obtained for each pixel, it can be determined whether the material of each pixel is the same or different from the magnitude of the value.

If there is a phase change in the interference waveform due to a difference in phase change amount at the time of reflection on the surface of the sample 8, a phase change of a constant multiple appears in the collected waveform. In the acquired waveform, the slope in the relationship diagram of “envelope peak position” vs. “scan position with phase 0” is a (“scan position with phase 0” − “peak position of envelope”) / (− a + 1)
+ “Peak position of envelope” (2)

Is expressed as a unit of the collected data array index (the scanning position at the time of data collection is expressed as 1, 2, 3,... N), the decimal part depends on the amount of phase change due to reflection on the sample surface. Different. Therefore, if this is calculated | required in each pixel corresponding on the surface of the sample 8, and it mutually compares, it can be determined whether the substance in each pixel is the same kind or a different kind.

This will be described below with examples of measurement results.
Fig. 4 shows the measurement result of the collected waveform (●), the result of applying Hilbert transform to it (○), and the envelope calculated from them (+ in □), and Fig. 5 shows the phase. It was. The optical filter 2 used has a center wavelength of 550 nm, a passband width of 80 nm, an interference period of 275 nm, and a Nyquist interval of 137.5 nm. The data collection interval is 247.5 nm, which is 1.8 times the Nyquist interval. Since the collected data includes noise of the camera 6 and the like, the envelope data also includes noise. The envelope data is smoothed by moving average processing or the like, and the peak position is calculated. The phase changes linearly unless it is far from the peak position of the envelope. The data in FIG. 5 changes every 2p, but if connected in phase, it is connected in a straight line, and is fitted with a linear expression to calculate the scanning position where the phase is zero. There are a plurality of “scanning positions with a phase of 0”, but those close to the “peak position of the envelope” are adopted. The “scan position with phase 0” can be calculated with higher accuracy than the “peak position of the envelope”.

Collected data is handled as an array, and when the number of frames to be collected is n, the scanning position (corresponding to time) is represented by an index from 1 to n for each pixel of x and y as follows.

D (1, x, y), D (2, x, y), ..., D (i, x, y), D (i + 1, x, y), ... D (n, x, y) )

  The horizontal axis in FIGS. 4 and 5 is an index of the scanning position of this array, and the interval of 1 corresponds to the collection interval of 247.5 nm.

FIG. 6 shows an example of “envelope peak position” and “phase 0 scan position” calculated for each pixel. A value in the x direction in a row of a pixel of y is shown. Sample 8 has a shape as shown in FIG. 12 and is obtained by forming a copper thin film on a silicon substrate. The area around x = 365 is the end of the copper thin film, and the film thickness increases as x increases from there. In FIG. 6, the direction of the vertical axis is opposite to that in FIG. It can be seen that the “scan position with phase 0” can be calculated with higher accuracy than the “peak position of the envelope”.
In FIG. 6, the “peak position of the envelope” increases linearly with respect to x in the silicon region due to the inclination of the sample 8. It can also be seen that the “scanning position where the phase is 0” also changes linearly with a negative slope. It can be said that the absolute value of the inclination of the “scanning position where the phase is 0” is larger than that of the “peak position of the envelope”, and the sensitivity to the change in the surface height h of the sample 8 is high. The fact that it can be calculated with high accuracy and that the sensitivity to the surface height is large is an advantage of using the “scanning position with phase 0”.

  FIG. 7 shows the data of FIG. 6 plotted as “envelope peak position” vs. “phase zero scan position”. Since the “peak position of the envelope” cannot be calculated with high accuracy, the data varies in the horizontal axis direction. However, a linear relationship between the “scan position where the phase is 0” and the “peak position of the envelope” can be seen (many straight lines with negative slopes drawn in FIG. 7 according to the measurement data). In each of the silicon region and the copper region, a repetitive linear relationship shifted by 1 with respect to the “peak position of the envelope” (the array index is a unit) is arranged (see Patent Document 4). In this example, the slope of these straight lines is −8.0. It can be seen that the linear relationship is shifted between the silicon region and the copper region. This represents the difference in the amount of phase change of reflected light between silicon and copper.

FIG. 8 is a plot from the data of FIG. 6 in which the slope a is −8.0 in equation (1), that is, the equation (3) is plotted on the vertical axis.

("Scanning position where phase is 0"-"peak position of envelope") / 9
+ “Peak position of envelope” (3)


In this way, there is one jump on the vertical axis, but otherwise there is no dependency on the horizontal axis.

  In FIG. 9, the decimal part of the data of FIG. 8 is plotted. Data in which the decimal part is 0.9 in the copper region in FIG. 8 is represented as −0.1 in FIG. The difference in the vertical axis values between silicon and copper can be clearly seen. The variation in the value on the vertical axis is due to the low calculation accuracy of the “peak position of the envelope”.

  FIG. 10 shows a result of moving average processing of the data of FIG. For each pixel, if the value on the vertical axis is greater than 0.27, for example, silicon can be determined, and if it is smaller, it can be determined that the pixel is different (similar or similar). The average value on the vertical axis in the silicon region of FIG. 10 is 0.43, and in the copper region is 0.097. These values correspond to the decimal part of the intersection of the straight lines in FIG.

  FIG. 11 shows an example in which a part of the silicon region in FIG. 7 is enlarged. The inclination a of the straight lines A, B, and C in FIG. 11 was set to −8.0 as described above. This can be determined from the average value of a large number of data. As described above, the decimal part of the intersection of the straight line “scanning position where phase is 0” = “peak position of envelope” and straight lines A, B, C is 0.43.

From the measurement value “envelope peak position” and the measurement value “scan position where phase is 0” at each pixel, it is specified which of the straight lines A, B, and C in FIG. Then, using the relationship of the identified straight line, the calculated value “peak position of the envelope” is obtained by calculation from the measured value “scanning position where phase is 0” at the pixel. By using the straight line relationship, the variation of the measurement value “peak position of the envelope” is removed. The calculated value “peak position of the envelope” corresponds to the height h of the surface of the sample 8.
If it is determined to be copper, the height h of the surface of the sample 8 is calculated in the same manner as described above using a straight line whose inclination is −8.0, which is the same as that of silicon, but whose decimal part is 0.097.

Hereinafter, a method of calculating the surface height h of the sample 8 will be specifically described with an example of silicon in FIG. The measurement value of “envelope peak position” is xm, the measurement value of “scanning position where phase is 0” is ym, and a reference value xms is created by “integer part of xm + 0.43”. 0.43 is the decimal part of the equation (3) in silicon as described above. The straight line in FIG. 11 is determined as follows. In this example, the slope of the straight line is −8.0 as described above. Using this straight line, the “peak position of the envelope” xc corresponding to the measured value ym of “scanning position with phase 0” is obtained by calculation.

(1) If ym> −8 (xm−xms) + xms + 4.5, use y = −8 {xc− (xms + 1)} + (xms + 1) straight line
xc = {ym− (xms + 1)} / (− 8) + (xms + 1) “Envelope peak position” is obtained.
Expressing the above in the example of 42 ≦ xm <43, if ym is above the dotted line “B + 4.5” in FIG. 11, the straight line C is used to obtain xc from ym.

(2) If −8 (xm−xms) + xms−4.5 <ym <−8 (xm−xms) + xms + 4.5, use y = −8 (xc−xms) + xms straight line
xc = (ym−xms) / (− 8) + xms “peak position of envelope” is obtained.
Expressing the above in the example of 42 ≦ xm <43, if ym is above the dotted line “B−4.5” and below the dotted line “B + 4.5” in FIG. 11, xc is obtained from ym using straight line B.

(3) If ym <−8 (xm−xms) + xms−4.5, use the straight line y = −8 {xc− (xms−1)} + (xms−1)
xc = {ym− (xms−1)} / (− 8) + (xms−1) “Peak position of envelope” is obtained.
Expressing the above in the example of 42 ≦ xm <43, if ym is below the dotted line “B-4.5” in FIG. 11, xc is obtained from ym using straight line A.

  The obtained xc corresponds to the height h of the surface of the sample 8. Since the measurement value of the “envelope peak position” varies greatly, the above-described case classification is required. In the case where the surface of the sample 8 is copper, the reference value xms may be made by “integer part of xm + 0.097” using 0.097 of the decimal part of the above-described formula (3) with copper. The method for obtaining the surface height later is the same as in the case of silicon.

  An example of the surface height h of the sample 8 thus obtained is shown in FIG. As described above, a copper thin film is attached on a silicon substrate, and the vicinity of x = 365 is the end of the copper thin film, and the film thickness becomes thinner as it approaches the end. Data in the x direction at a certain y. Since the scanning method of the camera used is an interlace method, the data in the y direction is discontinuous in the y direction with the collection time being shifted every other line, so the final calculation data of the surface height h of the sample 8 is the y direction. The discontinuities are eliminated by moving averages two by two. The horizontal axis is 640 pixels and corresponds to 900 μm. Since the surface height is calculated based on the “envelope peak position”, the copper step is correctly calculated without being affected by the phase change of the surface reflected light.

  FIG. 13 shows an enlarged view of the silicon region of FIG. The sample surface height can be calculated with high accuracy with a noise peak-to-peak of several nanometers, even though the data collection interval is 1.8 times the Nyquist interval. In order to avoid the influence of the phase change of the surface reflected light, it has been conventionally necessary to obtain the surface height from the envelope. However, in the present invention, the calculation accuracy is obtained by calculating from the phase of the collected waveform. improves.

1: Light source 2: Filter 3: Beam splitter 4: Michelson interferometer 4a: Objective lens 4b: Beam splitter 4c: Mirror 5: Piezo actuator 6: CCD camera 7: Sample holder 8: Sample

Claims (1)

A beam splitter and mirror are placed under the objective lens, and a Michelson-type interferometer is constructed including the sample surface. The distance to the sample or the distance to the mirror is scanned with a piezo actuator, resulting in interference. In the method for measuring the surface shape of a sample with a scanning white interferometer, the waveform is photographed with a CCD camera and recorded as moving image file data, and the surface shape of the sample is measured based on the moving image file data.
Using a CCD camera, the data collection interval is wider than the Nyquist interval (half the interference period), and a large number of video file data is measured.
The slope of the linear relationship between the “envelope peak position” and the “scan position with phase 0” of the acquired waveform calculated using the Hilbert transform from the waveform thus measured and acquired is a, and “envelope” The intersection of “the peak position of” and “the scanning position where the phase is 0” is the fractional part,
The expression (“scan position with phase 0” − “envelope peak position”) / (− a + 1) + “envelope peak position” is expressed in units of the collected data array index, which is expressed as By determining each corresponding pixel on the surface and comparing each other, determine whether the sample surface of each pixel is the same type or different from the sample surface of other pixels,
A straight line representing the relationship between “envelope peak position” vs. “scanning position with phase 0” and a straight line “scanning position with phase 0” = “peak position of envelope” intersect at the decimal part of the above formula. , The average value is calculated over a large number of pixels of the same substance for the decimal part, the average value is also determined from a large number of data for the slope a,
From the measured value of the “scanning position where the phase is 0” using the obtained decimal part and the above inclination a, the “peak position of the envelope” representing the correct sample surface height is obtained without depending on the phase change of the reflected light. And measuring the surface shape of the sample with a scanning white light interferometer.

Images