JPH04232802A - Measuring method for micro-size - Google Patents

Abstract

PURPOSE:To measure the micro-size without being affected by the fluctuation of the reflectance with simple software by using the approximate size determined from the pattern width of the preset intensity level as a reference to measure the correct size. CONSTITUTION:The reflected light of the laser light 130 focused to a micro-spot diameter and deflected on the face of a measured object 12 is reflected by a beam splitter 110 to becomes the reflected light 135, and the reflected light intensity at each deflection position is detected by a light reception section 140 synchronously with the deflection control voltage 125. A reflected light intensity pattern generation section 13 stores the data of the reflected light intensity in one period of deflection, a data procession section 150 constituted of a CPU processes the data of the reflected light intensity pattern and calculates the size. The reflected light intensity pattern is converted into binary values by the preset intensity slice level, the pattern width (half-value width) is calculated 14, the reflected light intensity calculated 15 from the approximate size of the measured object 12 thus obtained is used as a reference to calculate the correct edge position, and it is converted 19 into the actual size.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dimension measuring method for measuring minute dimensions of 1 micrometer or less using laser light.

0002

[Background Art] With the recent progress in precision processing technology, micromachining in the submicron region of 1 micrometer or less has become possible, and the need for precise measurement of dimensions, shapes, etc. of workpieces has increased. It's increasing. Optical measurement, which allows non-contact measurement, is often used to measure minute dimensions in the submicron region.

The method most often used in optical dimension measurement is the half-width detection method, in which the object to be measured is illuminated with a white light source and the reflected light is detected by an image sensor. %, the pattern width (half width) is detected. FIG. 2 shows a conventional half-width detection method. FIG. 2(a) shows the shape of the object to be measured 12 whose dimensions are measured, and 21 is the dimension part where the dimensions are measured, and the edge 20
0, 210, and the dimension between the edges is G. 2
Reference numeral 2 denotes a base material portion, which is present on both sides of the dimension portion 21. here,
When the object to be measured 12 is a magnetic head, the dimension portion 21 is a gap made of glass, and the base material portion 22 is a track made of ferrite. In the object to be measured having the above configuration, the dimensional portion 21 and the base material portion 22 have different reflectances, and the reflectance of the dimensional portion 21 is Rm, and the reflectance of the base material portion 22 is Rs. In the following description, it is assumed that Rm<Rs. Further, it is assumed that the surfaces of the dimension portion 21 and the base material portion 22 have substantially no level difference and are within the depth of focus of the illumination light.

A waveform 23 in FIG. 2(b) is a video signal obtained when the image sensor detects the reflected light from the object 12 illuminated with a white light source, and has a V-shaped pattern. The horizontal axis of the graph is the pixel address of the image sensor,
The vertical axis is the reflected light intensity. A line 24 is a slice level for binarizing the waveform 23. Generally, an intensity value of 50% (intermediate intensity between the maximum intensity and the minimum intensity) is used as the slice level. A waveform 25 in FIG. 2(c) is a binarized waveform when the video signal waveform 23 is binarized. Intensities higher than the intensity level of the line 24 have a level of 1 when binarized, and conversely, intensities lower than the level of the line 24 have a level of 0. Falling portion 26 and rising portion 27 of the binarized waveform 25
Detects the pixel addresses of , and calculates the number of pixels included between them. This number of pixels is called the half-width, and the half-width is converted into dimensions.

[0005] The inventor of the present invention has also proposed a dimension measurement method using optical deflection of a laser beam condensed to a minute spot diameter, and this method is described in detail in Japanese Patent Application No. 1983-51617. There is. This is done by irradiating the object to be measured 12 with a laser beam condensed to a minute spot diameter, deflecting the laser beam onto the surface of the object to be measured 12 using an acousto-optic deflection element, and The reflected light intensity pattern is detected and the dimensions are measured from the minimum intensity level of the reflected light intensity pattern.

FIG. 3 shows a conventional method for measuring dimensions using optical deflection of a laser beam. A waveform 31 in FIG. 3A is a reflected light intensity pattern detected when the laser beam is deflected on the surface of the object to be measured 12, and is a V-shaped pattern. The reflected light intensity pattern represents the relationship between the deflection position on the surface of the object to be measured and the reflected light intensity at that point.The horizontal axis of the graph is the deflection position of the laser beam, and the vertical axis is the reflected light intensity. , the reflected light intensity 33 is the maximum intensity Vm, and the reflected light intensity 32 is the minimum intensity Vn. At this time, Vn/Vm, which is the ratio of the minimum intensity Vn to the maximum intensity Vm, is calculated. When the dimensions are small, the difference between the minimum strength Vn and the maximum strength Vm is small, and when the dimensions are large, the difference becomes large. FIG. 3B is a graph showing the relationship between the dimension G and the reflected light intensity ratio Vn/Vm, where the horizontal axis is the dimension and the vertical axis is the intensity ratio. As shown in the figure, the two are in a proportional relationship in a specific size range, and the size can be measured from the intensity ratio Vn/Vm.

[0007]

In the conventional half-width detection method, as the dimensions of the object to be measured become smaller, the intensity changes at the falling and rising portions of the video signal 23 become broader. When performing binarization by setting a slice level of 50% intensity, even a slight variation in the slice level causes a large variation in the number of binarized pixels, reducing the reliability of dimension measurement. Furthermore, as the dimensions become smaller, the change in the half-width with respect to the change in dimensions becomes smaller, resulting in a problem that the measurement sensitivity decreases. Further, in the method using optical deflection of a laser beam, accurate dimension measurement cannot be performed when the reflectance of the object to be measured changes. This is because the minimum intensity Vn changes particularly when the reflectance of the dimensional portion 21 changes. The purpose of the present invention is to solve the above problems and to
It is an object of the present invention to provide a method for accurately measuring the dimensions of an object without being affected by changes in reflectance of an object to be measured.

[0008]

[Means for Solving the Problems] In order to achieve the above object, the present invention comprises the method shown below. A laser beam focused on a minute spot diameter is irradiated onto the surface of an object to be measured that has edges and whose dimensions between the edges are to be measured, and the light is deflected to reflect light from the object. In a method of measuring minute dimensions in which the dimensions of the object to be measured are measured by detection, a reflected light intensity pattern representing the relationship between the deflection position on the object to be measured detected in one cycle of light deflection and the intensity of reflected light is used. In contrast, the pattern width at a preset intensity level is detected, the approximate dimensions of the object to be measured are calculated from the pattern width, and the approximate edge positions of the object to be measured corresponding to the dimensions are calculated. The reflected light intensity Ve is set at
setting an intensity range for each region with an intensity level lower than the intensity, and approximating the intensity data of the reflected light intensity pattern within the intensity range with first and second straight lines;
The intersection point of the first and second straight lines is calculated to determine an accurate edge position of the object to be measured, and the dimensions of the object to be measured are measured from the amount of light deflected between the edges.

[0009]

[Operation] At each deflection position of the laser beam, the intensity of reflected light from the object whose dimensions are to be measured is detected, and a reflected light intensity pattern is created for one period of deflection. The reflected light intensity pattern is binarized using a slice level having a preset intensity level, and the pattern width is calculated. This putter
When calculating the dimensions of the workpiece from the information on the width of the pattern,
Since the conversion sensitivity from the width of the ring to dimensions is low, the accuracy of the dimensions obtained is not very high and is only an approximate dimension. The reflected light intensity Ve corresponding to the approximate edge position of the object to be measured is calculated from this approximate dimension, and the accurate edge position is analytically calculated based on the intensity Ve. In order to detect the edge position, the state in which the edge is irradiated with the intensity peak position of the laser beam is detected. Since the laser beam has a Gaussian intensity distribution, the intensity change rate of the reflected light is modulated before and after the edge is irradiated with the intensity peak point. Therefore, by detecting feature points where the rate of change in intensity of the reflected light intensity pattern is modulated, accurate edge positions can be determined. For this reason, a region of high intensity and a region of low intensity are set around the intensity Ve, and the reflected light intensity data in each region is approximated by a straight line. Since these two straight lines have different slopes, there is an intersection point, and the edge is irradiated with the intensity peak point of the laser beam at the intersection point. In this way, accurate edge positions are analytically determined from information on approximate edge positions, and dimensions are measured from the amount of deflection between the edges.

0010

Embodiments Examples of the present invention will be described below with reference to the drawings. FIG. 1(a) is a system block diagram when performing minute dimension measurement according to the present invention, and FIG. 1(b) is a flowchart explaining the method of data processing of reflected light intensity patterns according to the present invention.
This is a diagram. FIG. 4 is a diagram illustrating a reflected light intensity pattern detected by deflection of a laser beam, in which the horizontal axis of the graph represents the deflection position on the object to be measured, and the vertical axis represents the reflected light intensity. The dimension measurement method will be explained with reference to FIGS. 1 and 4. In FIG. 1A, 10 is a laser light source, which is made of, for example, a He--Ne laser tube or a semiconductor laser, and emits a laser beam 100. 11 is a deflection optical system that performs optical deflection (scanning) of the laser beam 100, and includes an acousto-optic deflection element (hereinafter abbreviated as AO).
105, a beam splitter 110, an objective lens 115 that focuses the laser beam into a minute spot diameter, and other optical elements such as lenses and mirrors (not shown). A deflection control unit 120 controls the deflection operation of the AO 105 using a deflection control signal 125. Deflection control signal 125
is a ramp wave signal whose voltage changes continuously, and deflects the laser beam to a position according to the voltage. Reference numeral 12 denotes an object whose dimensions are to be measured, as shown in (a) in FIG. Laser light 130 that is focused into a minute spot diameter and deflected onto the surface of the object to be measured 12 is reflected, passes through the objective lens 115, is reflected by the beam splitter 110, and becomes reflected light 13.
5 and is detected by the light receiving unit 140. The light receiving section 140 is P
IN photodiode, current-voltage converter, and A/D
It consists of a converter and detects the intensity of reflected light at each deflection position in synchronization with the deflection control voltage 125. Reference numeral 13 denotes a reflected light intensity pattern creation section, which is composed of a memory, and a light receiving section 140.
The reflected light intensity at each deflection position detected in the above is stored in a memory circuit, and the data of the reflected light intensity in one period of deflection is stored. Reference numeral 150 denotes a data processing section which is composed of a CPU and processes data of the detected reflected light intensity pattern to calculate dimensions.

Next, a data processing method will be explained using the data processing flowchart shown in FIG. 1(b). The data processing described below is performed using a microprocessor.
This is a numerical calculation based on A waveform 41 in FIG. 4 is a reflected light intensity pattern, which is similar to the waveform 31 shown in FIG. 3 and is a V-shaped pattern. Reference numeral 14 in FIG. 1B indicates pattern width detection, which detects the pattern width of the reflected light intensity pattern at a preset slice intensity level. A line 42 in FIG. 4 is the pattern width Ph. The slice level set at this time is the 50% intensity level. The pattern width Ph is the object to be measured 1
It is not a direct dimension of 2, but a value larger than the dimension,
The pattern width Ph is converted into a dimension using a dimension conversion coefficient. The dimensions calculated at this time are not exact dimensions and generally include errors. This is because when the dimensions become smaller, the change in pattern width Ph becomes smaller with respect to the change in dimensions, and the conversion sensitivity decreases. Therefore, the putter
Only approximate dimensions can be obtained from the ring width Ph. However, this pattern width Ph does not depend on the reflectance of the object to be measured;
If the diameter of the irradiated beam spot is constant, it has a value specific to the dimension.

15 is edge strength setting, pattern width Ph
The reflected light intensity level at the approximate edge position corresponding to the approximate dimension determined from is set. A line 43 in FIG. 4 is a pattern width Pe corresponding to the approximate dimension, and the reflected light intensity level Ve at that time is calculated. Since the amount of deflection per unit deflection control voltage is determined by the deflection optical system 11, the above-mentioned pattern width Pe can be easily determined from approximate dimensional values. 44 is the intensity corresponding to one (left) edge, 45 is the intensity corresponding to the other (right) edge,
Generally the two strengths are equal. An accurate edge is determined based on this strength Ve, and accurate dimensions are measured. 16 is an intensity range setting, which sets a calculation range for performing various calculations to determine accurate edges. Range 46 in Figure 4
is a region where the intensity is higher than the Ve intensity for determining the left edge position, and the range 47 is a region where the intensity is also lower. The range 48 is a region with a higher intensity than the Ve intensity for determining the right edge position, and the range 49 is a region with a lower intensity, and the above four regions are set. Note that if the reflected light intensity pattern is symmetrical with respect to the minimum intensity position,
An intensity range may be set for either side.

Reference numeral 17 in FIG. 1(b) is a linearization calculation, in which the above-mentioned intensity ranges 46, 47, 4 set in the intensity range setting 16 are
The intensity data of the reflected light intensity pattern in 8 and 49 is linearly approximated. At this time, the least squares method is used for linearization. A straight line in an intensity range higher than the reflected light intensity Ve is called a first straight line, and a straight line in an intensity range lower than the reflected light intensity Ve is called a second straight line. Reflected light intensity pattern V
In a certain intensity range around the e intensity, the intensity change can be considered sufficiently linear. At this time, two straight lines can be set for one edge (left side) and the other edge (right side). 18 is an intersection calculation, which determines the intersection of the first straight line and the second straight line determined in the linearization operation 17. Since the slope of the first straight line is greater than the slope of the second straight line, there is an intersection between the two straight lines. This intersection position is measured object 1
2 accurate edge position. This edge position is given by the voltage value of the laser beam deflection control signal 125. 19 is a dimension calculation, in which the voltage difference between the deflection control voltages of the laser beams corresponding to the two edge positions determined in the intersection point calculation 18 is converted into an actual dimension using a dimension conversion coefficient.

Next, a specific method of calculating a reflected light intensity pattern according to the present invention will be explained. FIG. 5(a) shows the state of deflection of the laser beam on the surface of the object to be measured 12, and FIG. 5(b) shows the relationship between the intensity at each deflection position of the reflected light intensity pattern. 5(c) shows the relationship between the pattern width and dimensions at the 50% intensity level. In FIG. 5(A), the object to be measured is the magnetic head shown in FIG. 2, 50 is a track portion made of a magnetic material, 51 is a gap made of glass, and has edges 53 and 54, and the dimension between the edges is measured. do. 52
is a laser beam to be irradiated, and its intensity distribution is Gaussian, and has a symmetrical distribution with a peak intensity of 520 as the center. Here, the laser beam 52 is deflected from point A to point B. The diameter of the beam spot of the laser beam 52 is 13.5 of the peak intensity value.
% intensity point, it is ~1.5 μm, but since the gap dimension G is ~0.5 μm, when the gap 51 is irradiated with the laser beam 52, only a part of it is irradiated. That will happen.

In the S1 state, the right end of the laser beam 52 is irradiated just before the edge 53, and the reflected light is on the track 50.
The intensity of the reflected light is maximum. When the deflection progresses from the S1 state to the S2 state, the peak of the laser beam 52
The dark strength position coincides with the edge 53. The S3 state is a state in which the peak intensity position of the laser beam 52 is irradiated to the center of the gap 51, and the S4 state is a state in which the laser beam 52 is irradiated as in the S2 state.
The state where the peak intensity position of is irradiated on the edge 54, S5
This state is the same as the S1 state, in which the left end portion of the laser beam 52 is irradiated immediately after the edge 54. Although the figure shows a typical state of deflection, the actual deflection is performed in finer steps, for example, about 0.01 .mu.m, which is less than 1/100 of the beam diameter of the laser beam 52. Reflectance R of track portion 50
s and the reflectance Rm of the gap portion 51 are different, and Rs>Rm
Therefore, the reflected light intensity pattern detected in one period of deflection is
Although the pattern has a V-shaped waveform 41 as shown in FIG. 5(b), the reflected light intensity becomes minimum in the S3 state. In the S2 state and the S4 state, the rate of change of the reflected light intensity is modulated depending on the polarization state before and after the S2 state and the S4 state. The dimension measuring method according to the present invention analytically detects the S2 state and the S4 state from the reflected light intensity pattern, and determines the dimension from the amount of deflection from the S2 state to the S4 state.

The S2 state and the S4 state are states in which the reflected light intensity changes characteristically. In the case of the S2 state, the intensity difference between the left and right sides of the laser beam 52 at its peak intensity position is maximum, and as the deflection progresses from the S2 state to the S3 state, the left part (on the track) The part on the right (the part that is illuminated by the gap) has a large decrease in intensity, and the part on the right (the part that is illuminated in the gap) has a slight increase in intensity. Therefore, the rate of change in reflected light intensity differs before and after the S2 state. The S4 state is the opposite of the S2 state. This rate of change is not very large, and the edge positions 53 and 54 cannot be determined directly from the reflected light intensity pattern alone. Therefore, in the present invention, the approximate edge positions are determined in advance from the pattern width information of the reflected light intensity pattern. The edge position 53 is determined based on that position.
, 54 are determined.

FIG. 5(c) shows a reflected light intensity pattern waveform 41.
An example of the relationship between the pattern width Ph and the dimension G at the 50% intensity level is shown. The horizontal axis of the graph is the dimension, the vertical axis is the pattern width, and the voltage used to control the deflection is shown. As is clear from the figure, as the dimensions become smaller, the change in pattern width Ph becomes smaller. When the dimension is about 0.5 μm, the pattern width Ph changes by only 30% of the change in dimension, so it becomes difficult to directly calculate the exact dimension. At this time, the conversion coefficient of the relationship shown in FIG. 5(c) is determined in advance, and the approximate dimension Gn is calculated from the pattern width Ph. Next, the pattern width corresponding to the approximate dimension Gn is determined to calculate the approximate edge position. The pattern width at this time is 5 in Figure 5 (b).
7. In this way, the approximate edge positions 58 and 5
9 can be determined. The reflected light intensity at the approximate edge position at this time is Ve.

Next, a method for determining accurate edge positions from information on approximate edge positions will be explained. FIG. 6(a) shows the relationship between the dimension G and the intensity of reflected light at the edge positions 53 and 54, and FIG. 6(b) shows calculation of the edge position by the linear approximation method. In Fig. 6 (a), the horizontal axis of the graph is the dimension G, and the vertical axis is the reflected light intensity Vt at the edge position, when the maximum intensity of the reflected light intensity pattern is normalized to 1 and the minimum intensity to 0. Shown as normalized strength. As is clear from the diagram,
The reflected light intensity Vt changes in proportion to the dimension G, for example 0.
A dimension of 5 μm results in a Vt intensity of about 15%. However, even if the dimensions are the same, the Vt intensity varies depending on the beam diameter of the irradiated laser beam. The intensity Ve corresponding to the approximate edge position obtained by measurement is generally different from the intensity Vt corresponding to the exact edge position shown above. As described above, the intensity Ve includes errors and therefore varies by several percent with respect to the true intensity Vt. In order to absorb this variation and determine accurate edge positions, the reflected light intensity data is subjected to the following linearization approximation process.

FIG. 6B shows an analysis method for the left edge. The same applies to the right edge. Although the ranges 46 and 47 have been explained with reference to FIG. 4, an intensity range 46 higher than the Ve intensity and an intensity range 47 lower than the Ve intensity are set. When Vt intensity is 15%, Ve intensity is 13%
Suppose that we obtain At this time, the range 46 is from 16% to 3
Set the intensity range to 0%, range 47 is from 10% to 5
Set it in the range up to %. However, the aforementioned strength range varies depending on the size. Since the reflected light intensity changes almost linearly in each of the above intensity ranges, the first straight line shown by line 62 is used for range 46, and the second straight line shown by line 64 is used for range 47. decide. The linearization calculation at this time uses the least squares method. The first straight line 64 has a greater slope than the second straight line 64, and the two straight lines intersect at a point 66. This intersection is a characteristic point of the change in reflected light intensity in the S2 state and corresponds to an accurate edge position. For the right edge, the intersection 68 is calculated in the same manner, and the dimension G is calculated from the deflection control voltage when deflecting between the intersection 66 and the intersection 68.

In the above description, in order to accurately detect the edge position, it is necessary to deflect the laser beam by a minute step distance. For this purpose, AO105, which is capable of stable deflection, is used. In order to measure dimensions according to the present invention, a laser beam focused to a diameter of 1.5 μm is
Deflect in m steps. The deflection optical system used at this time is described in detail in the patent application publication by the inventor mentioned in the prior art section, so it will be omitted in this application. Also,
In the above description, an example was described in which the reflected light intensity is approximated by a straight line, but the approximation may be performed using a quadratic, tertiary, or quartic polynomial, and the intersection of the polynomials may be calculated.

[0021]

[Effects of the Invention] As described above, according to the present invention, it is possible to accurately detect edges by analyzing reflected light intensity based on approximate dimension values based on half-width information whose dimension conversion sensitivity is not very good. It is possible to perform accurate submicron dimension measurements with an accuracy of 0.01 μm without using special hardware and with simple software, without being affected by changes in the reflectance of the measured object. .

[Brief explanation of the drawing]

FIG. 1 is a block diagram illustrating the dimension measurement method of the present invention; FIG. 1 (A) shows the overall configuration, and FIG. 1 (B) shows a flowchart of data processing for dimension measurement. It is a diagram.

FIG. 2 is a diagram illustrating a conventional technique using a half-width detection method.

FIG. 3 is a diagram illustrating a conventional technique for calculating dimensions from the intensity of a reflected light intensity pattern caused by deflection of laser light.

FIG. 4 is a diagram illustrating a method for processing a reflected light intensity pattern according to the present invention.

FIG. 5 is a diagram explaining the relationship between the optical deflection of a laser beam and the reflected light intensity pattern according to the present invention, FIG. 5(a) is a state diagram of the deflection of the laser beam, and FIG. 5(b) 5 is a diagram illustrating the relationship between the half-width of the reflected light intensity pattern and the approximate edge, and FIG. 5C is a diagram illustrating the relationship between the dimension and the half-width.

FIG. 6 is a diagram illustrating a method for determining an edge position according to the present invention; FIG. 6(a) is a diagram illustrating the relationship between dimensions and reflected light intensity at an edge position; FIG. 6(b) is an edge diagram by linear approximation It is a figure explaining the method of determining a position.

[Explanation of symbols]

10 Laser light source 13 Reflected light intensity pattern creation section 14 Pattern width detection 15 Edge strength setting 16 Intensity range setting 17 Linearization operation 18 Intersection calculation 19 Dimension calculation 41 Reflected light intensity pattern 62 First straight line 64 Second straight line

Claims (1)

[Claims] 1. A laser beam condensed into a minute spot diameter is irradiated onto the surface of an object to be measured which has edges and whose dimensions between the edges are to be measured, and the light is deflected. In a micro-dimension measurement method in which the dimensions of the object to be measured are measured by detecting the reflected light from For the intensity pattern, the pattern width at a preset intensity level is detected, the approximate dimensions of the object to be measured are calculated from the pattern width, and the approximate dimensions of the object to be measured corresponding to the dimensions are determined. Set the reflected light intensity Ve at the approximate edge position of the intensity data of the reflected light intensity pattern within the intensity range is approximated by first and second straight lines, and the intersection point of the first and second straight lines is calculated to determine the intensity data of the reflected light intensity pattern within the intensity range. 1. A method for measuring minute dimensions, comprising: determining accurate edge positions of the object; and measuring the dimensions of the object from the amount of light deflected between the edges.