JP5698963B2 - Surface shape measurement method - Google Patents

Description

  The present invention relates to a surface shape measurement method for measuring a surface shape of one measurement target part in a plurality of times and connecting the surface shape measurement data to obtain one surface shape data.

  As a method for measuring the surface shape of a measurement object with high accuracy, an optical surface shape measurement method using optical interference such as laser light or white light is known. In such an optical surface shape measuring instrument using optical interference, a high-magnification objective lens is used to increase the horizontal resolution of the measurement. Often millimeters or less. Therefore, there are cases where the region to be measured cannot be measured by a single measurement. When the area to be measured exceeds the measurement field of view of the surface shape measuring device in this way, generally, the measurement surface area is scanned and measured multiple times, and the obtained multiple surface measurement data are joined together. A stitching method (aperture synthesis method) that obtains a surface shape measurement result having an area larger than the measurement field of view of the surface shape measuring apparatus is used (for example, see Patent Document 1). Conventional stitching methods perform complex calculations to correct lens distortion, reference surface shape error, precision stage movement error, and other measurement errors, and correct not only the translation direction but also the rotation direction. Then, the measurement data are joined together (see, for example, Non-Patent Document 1).

JP-A-4-290905

"Study on measurement of whole functional surface by connecting 3D shape measurement data group", 2004 JSPE Spring Conference Annual Lecture Proceedings (p.667-668)

  However, the conventional stitching method requires, for example, about 30% to 50% of overlap areas, which are measurement surface areas that overlap each other, particularly in order to correct the rotational direction. There is a problem that the number of times of measurement for measuring increases.

  In addition, in the measurement of fine surface shapes that measure in nanometer order, even a temperature change of just 1 degree of the device temperature and ambient temperature has a great influence on the measurement result due to the thermal expansion of the measuring device or measurement object. However, the conventional method has a problem in that the measurement time increases as the number of measurements increases, and it is easily affected by temperature changes during the measurement.

  Optical interferometers have a considerable amount of optical errors due to lens distortion and reference surface distortion. Although this optical error is often smaller than the error due to the precision stage, the error accumulates every time data is connected, and may eventually become a large error that cannot be ignored. In other words, if there is distortion due to optical errors in the overlap area, not only the translation direction but also the rotation direction will be matched to match the distortion, and the synthesized measurement data will be compared to the original surface shape. The entire state is inclined by the amount of the distortion angle. If the next measurement data is further connected to the measurement data in the tilted state, the tilt error further increases. If this process is repeated, errors that were small at the beginning may be accumulated and become large errors. As described above, the conventional method has a problem that the rotational direction correction mainly performed to correct the movement error of the precision stage or the like causes another large accumulated error depending on the case. FIG. 16 is an example of measurement when a cumulative error occurs in planar measurement. When rotation correction is not performed, the error is 5 nm as a whole, but when rotation correction is performed, an error of 100 nm or more occurs.

  Accordingly, an object of the present invention is to solve the above-described problems of the prior art. Specifically, when measuring a measurement area that exceeds the measurement field of view of a surface shape measuring instrument, it is possible to easily combine multiple measurement results, reducing the overlap area and reducing the overall measurement time. An object of the present invention is to provide a surface shape measuring method capable of reducing the generation of a certain cumulative error.

  That is, the present invention is a surface shape measurement method for measuring the surface shape by joining the surface shape measurement data measured in multiple times, the first step of installing the measurement object on the horizontal plane moving stage, A second step of obtaining a first surface measurement data by measuring a three-dimensional surface shape of the first measurement surface region of the measurement object with an optical surface shape measuring machine; and A third step of moving the horizontal plane moving stage so that a second measurement surface region having an overlap region overlapping with a part and a measurement field of view of the optical surface shape measuring machine coincide with each other; and the second measurement A fourth step of measuring a three-dimensional surface shape of a surface region with the optical surface shape measuring instrument to obtain second surface measurement data, and a surface measurement corresponding to the part of the first surface measurement data Data and second surface measurement data A fifth step of translating one or both of the first surface measurement data and the second surface measurement data so that the surface measurement data corresponding to the overlap region in the overlap, and the translated first And a sixth step of generating combined surface measurement data by combining the one surface measurement data and the second surface measurement data.

  In this surface shape measurement method, it is possible to suppress the occurrence of cumulative errors caused by optical distortion, in particular, by performing only translation without correcting the rotation direction at the time of joining.

  Preferably, the overlap region may have an area of 10% or less of a measurement visual field of the optical surface shape measuring instrument.

  More preferably, the overlap area may be an area of 1% or less of a measurement visual field of the optical surface shape measuring instrument.

  Since reducing the overlap region does not substantially affect the accuracy of splicing with respect to the surface shape measurement method in the present invention, substantially sufficient performance can be obtained even with such a very small area. Can do. On the other hand, since the overlap region is reduced, the number of measurements can be reduced, and the time required for the entire measurement can be reduced.

  More preferably, the parallel movement can be a movement only in the height direction.

  When the horizontal error of the surface shape measuring device is sufficiently small with respect to the required accuracy, the measurement can be further speeded up by making the parallel movement only in the height direction.

  Specifically, the first surface measurement data and the second surface measurement data are generated so that the first surface measurement data and the second surface measurement data are continuously connected. It is possible to generate synthetic surface measurement data that combines surface measurement data of measurement surface regions that overlap each other by weighted averaging processing of the surface measurement data.

  Normally, even if the parallel movement is performed, the overlapping portion of the first surface measurement data and the second surface measurement data does not completely coincide with each other. It is necessary to generate and synthesize new surface measurement data. At this time, by averaging after performing weighting according to the position of the overlapping portion, the surface shape is likely to represent the surface shape of the original measurement object from the two surface measurement data, and the first It is possible to generate composite surface measurement data in which the surface measurement data and the second surface measurement data are continuously connected.

  More specifically, the optical surface profilometer can be a white interferometer.

  Further, after the sixth step, a seventh step of repeatedly performing the third to sixth steps using the already measured measurement surface region as the first measurement surface region can be provided.

  By doing in this way, it becomes possible to repeat the measurement and the connection of the measurement data and measure an arbitrary area until all the measurement regions to be measured can be measured.

  Hereinafter, a surface shape measuring method according to the present invention will be described with reference to the accompanying drawings.

It is the schematic of one Example of the surface shape measuring apparatus used for the surface shape measuring method of this invention. It is a figure which shows the relationship between the moving direction of a horizontal surface moving stage, and the camera visual field of the CCD camera of a white interferometer. The scale calibration of the camera field of the white interferometer is shown. It is a flowchart which shows the measurement step of the surface shape measuring method of this invention. The state which the 1st measurement surface area | region and the 2nd measurement surface area | region overlap with an overlap area | region is shown. The surface measurement data before translation is shown. FIG. 7 shows combined surface measurement data obtained by combining the two surface measurement data in FIG. 6. It is a graph showing the state of disagreement of the surface measurement data in the overlap region and the combined surface measurement data after performing the weighted average. It is a table | surface which shows the measurement conditions in the example of a measurement by the surface shape measuring method of this invention. It is a graph which shows the result of having measured the aspherical lens among the measurement examples by the surface shape measuring method of this invention. It is a graph which shows the comparison between each measurement conditions about the measurement result of an aspherical lens. It is a graph which shows the difference of the measurement result of a 1% overlap area | region, and the measurement result of a stylus type surface shape measuring machine about the measurement result of an aspherical lens. It is a graph which shows the comparison between measurement conditions about the measurement result of an aspherical lens among the measurement examples by the surface shape measuring method of this invention. It is a graph which shows the comparison between measurement conditions about the measurement result of a chromium vapor deposition plane. It is a graph which shows the stitching accompanying the conventional rotation correction. It is a graph which shows the example of a measurement when a cumulative error generate | occur | produces by the rotation correction by a conventional method.

  A surface shape measuring apparatus 10 shown in FIG. 1 includes a white interferometer 11, a horizontal plane moving stage 12, and a data calculation unit 13 that performs data calculation of measurement data. The white interferometer 11 and the horizontal plane moving stage 12 are fixed to the apparatus base 21. The white interferometer 11 includes an interference objective lens 15 incorporating a reference mirror, a measurement actuator 16, an optical head 17, and a CCD camera 19. The interference objective lens 15 is selected to have an appropriate magnification in accordance with the object to be measured. In particular, when it is desired to increase the horizontal resolution of the measurement, a high magnification lens is selected. The interference objective lens 15 is attached to the measurement actuator 16, and the measurement actuator 16 precisely scans the interference objective lens 15 in the vertical direction during interference fringe measurement to change the distance between the interference objective lens 15 and the measurement object 30. Used to make. The measuring actuator 16 has a built-in linear scale so that the amount of change by the actuator can be measured accurately. The measurement actuator 16 is further attached to the optical head 17. The optical head 17 includes a light source 18, an optical system that appropriately guides light from the light source 18 to the interference objective lens 15, a reference mirror built in the interference objective lens 15, and the measurement object 30. An optical system that accurately forms an interference fringe due to reflected light on the imaging surface of the CCD camera is incorporated. In this apparatus, a high-intensity white LED light source is used as the light source. Since generation of heat is suppressed as compared with conventional halogen lamps, slight thermal expansion is particularly effective for fine surface shape measurement that causes measurement errors. The white interferometer 11 is fixed to the apparatus base 21 via the Z-axis drive unit 20. The horizontal plane moving stage 12 positioned below the white interferometer 11 and fixed to the apparatus base 21 is a high straightness XY axis stage that can be driven in the XY axis direction. The high straightness XY axis stage used in this example is a stage that is also used in the micro shape measuring machine ET series manufactured by Kosaka Laboratory Ltd., and the straightness is 0.1 μm and 5 mm when moving 100 mm. It is a high straightness stage that guarantees 5 nm when moving. In this surface shape measuring device, a white interferometer is adopted as the optical surface shape measuring device, but the present invention is not limited to this, and other optical surface shape measuring devices such as a phase shift method and laser beam scanning are used. It is also possible to use the surface shape measuring device used.

  As shown in FIG. 2, the CCD camera 19 provided in the white interferometer 11 is installed such that the movement directions X and Y of the high straightness XY axis stage coincide with the long side direction Xa and the short side direction Ya of the camera visual field 40. Yes. The position adjustment of the CCD camera 19 is performed when, for example, a measurement object having a characteristic fine protrusion is placed on a high straightness XY axis stage and moved in the X axis direction on the high straightness XY axis stage. This is done by adjusting the rotation direction of the CCD camera 19 so that the protrusions appear to move in the Xa direction. At this time, it is important to match as much as possible. For example, when the fine protrusion moves from one end of the camera visual field 40 to the other end, the position adjustment of the CCD camera 19 is performed so that the one-pixel shift does not occur in the Ya direction. I do. Further, the length Xc in the X direction and the length Yc in the Y direction of the camera field of view are calibrated using high-precision linear scales Xs and Ys. In this way, by adjusting the angle and scale of the camera field of view of the CCD camera in advance, there is no need to consider the enlargement / reduction of the surface measurement data in the XY direction, the positional deviation correction, and the rotation process. A measurement result with sufficient accuracy can be obtained by connecting measurement data by translation only in the height direction (Z direction).

  Next, a surface shape measuring method using the surface shape measuring apparatus 10 will be described based on the flowchart of FIG. At the start of measurement, first, the measurement object 30 is placed on the horizontal plane moving stage 12 (S01). It is necessary to securely fix the measurement object 30 so that the measurement object 30 is not displaced due to vibration during measurement.

  Next, the surface shape of the first measurement surface region 31 is measured by the white interferometer 11 (S02). The interference fringes measured by sequentially moving the interference objective lens up and down by the measurement actuator 16 are picked up by the CCD camera 19 and sent to the data calculation unit 13 together with the height information at the time of measurement of the measurement actuator 16. Data processing is performed in the data calculation unit 13 and stored as first surface measurement data.

  Next, the horizontal plane stage 12 is moved so that the second measurement surface region 32 and the measurement visual field of the white interferometer 11 coincide (S03). Here, the second measurement surface region 32 is a measurement surface region including an overlap region 33 that overlaps the first measurement surface region 31. As will be described later, the size of the area of the overlap region hardly affects the accuracy of joining the surface measurement data that overlap each other. In terms of averaging the noise included in the surface measurement data, it is better that the overlap region is large. However, in practice, the overlap region has no practical problem even if it is 10% of the measurement field of view. Even as a line (width of one pixel in a CCD camera), sufficient joining accuracy can be secured. Therefore, in the present embodiment, the measurement is usually performed with the overlap region as 1%. However, depending on the shape and material of the measurement object in the overlapping region, there are cases where measurement cannot be performed with a white interferometer, or even if measurement can be performed, noise increases and connection cannot be performed with sufficient accuracy. In such a case, the overlap region is appropriately set large so that the joining can be performed with sufficient accuracy.

  Next, the surface shape of the second measurement surface region 32 is set by the white interferometer 11 as in the case of the first measurement surface region 31, and the second surface measurement data is generated and stored in the data calculation unit 13 ( S04).

  Next, a portion corresponding to the overlap region 33 of the first surface measurement data 34 and the second surface measurement data 35 is compared, and a positional deviation at the time of measurement of both data is obtained. Here, in the present invention, rotation correction around the X, Y, and Z axes as in the prior art is not performed, and only translation in the X, Y, and Z directions is performed. When translating data in the X or Y direction, calculations such as matching processing of feature points in each surface measurement data are required. However, the use of a high-straightness stage and accurate adjustment of the CCD camera are required. In addition, by performing calibration in advance, in practice, sufficient measurement accuracy can be obtained without performing parallel movement in the X direction and the Y direction. Regarding the height direction (Z direction), although it depends on the accuracy of the measurement actuator 16 of the white interferometer 11, it is usually not possible to guarantee nanometer-order accuracy. Translation in the direction (Z direction) is required. In this embodiment, the amount of displacement (Zoffset) only in the height direction (Z-axis direction) is calculated and connected. Specifically, the amount of deviation in the height direction between the first surface measurement data 34 and the second surface measurement data 35 is obtained from the difference in the height direction of the overlap region, and the second surface measurement data is correspondingly obtained. The 35 height direction data are translated in parallel so that the residual between the data in the overlap region 33 is minimized. In this embodiment, the second surface measurement data 35 is translated in the height direction so as to match the first surface measurement data 34. However, the first surface measurement data 34 is translated and the second surface measurement data 35 is translated. Even if it matches the surface measurement data 35 or both surface measurement data are translated and combined, there is no essential difference and the obtained result is the same.

  Next, the first surface measurement data and the second surface measurement data translated in the height direction are combined to generate one combined surface measurement data 36. At this time, normally, the two surface measurement data in the overlap region 33 do not completely coincide with each other, and the data is shifted as shown in FIG. Such a shift is mainly caused by optical distortion such as lens distortion or reference surface distortion, and can be reduced to some extent by correcting such optical distortion before measurement. However, since there is still no perfect match, some operation is required for joining. In this embodiment, weighted average is adopted as one means. As a result of the weighted averaging process, the double-sided measurement data are smoothly joined as shown in FIG. In the present embodiment, this optical distortion correction is not particularly described, but the correction before measurement is appropriately performed as necessary.

  The area to be measured varies depending on the situation, and it is rather rare that the measurement of the entire area to be measured by two measurements as described above is completed. In such a case, the measurement surface area that has already been measured is replaced with the first measurement surface area, and the measurement surface area that partially overlaps the second measurement surface area is measured and synthesized. By repeating the above, it is possible to measure all areas to be measured. For example, in the measurement example described later, this process is repeated 20 to 40 times to measure a predetermined area.

10 to 14 show the measurement results when actually measuring according to the method of the above embodiment. Hereinafter, measurement examples will be described.
First, a measurement example in which the surface shape of an aspheric lens having a width of 4.5 mm and a convex portion of 30 μm is measured will be described. From the results shown in FIGS. 10 and 11, a difference of about 10 nm is seen when comparing the 50% overlap region and the 1% overlap region. Further, as shown in FIG. 12, the difference between the measurement with the 1% overlap region and the measurement with the stylus type surface shape measuring instrument is about 20 nm, but 0.1% for the measurement of the convex shape of 30 μm. %, It can be said that the performance is sufficient. By the way, the difference between these measurements includes not only differences due to measurement methods, but also effects due to differences in measurement locations and sensitivity factors. It is considered that they match with higher accuracy. The maximum measurement resolution in the height direction of the stylus type surface shape measuring machine used for comparison is 0.1 nm, and the reproducibility of the step measurement is within 2 nm (1σ).

  The difference between the measurement result of the present invention in the 1% overlap region shown in FIG. 14 and the measurement result of the stylus type surface shape measuring machine is 5 nm or less, which shows a very good agreement. Compared to the measurement example of the aspherical lens, it shows a better match than the glass surface of the aspherical lens because the chromium deposition surface has a higher light reflectivity, and the noise during measurement by the optical interferometer method is higher. This is due to the small size and the fact that there is almost no influence of the displacement of the measurement position between the two measurements because of the flat surface.

  In addition, in order to measure the 4.5 mm area, it is necessary to perform 40 measurements in the 50% overlap area and to connect the 40 surface measurement data, but in the 1% overlap area, it is 20 times half. Measurement is enough. That is, the measurement can be completed in about half the time compared to the conventional case where an overlap region of 30% to 50% is required. This not only brings about the effect that the measurement time can be shortened, but also shortens the measurement time, thereby reducing environmental changes such as temperature during that time and contributing to improving the accuracy of the measurement results.

DESCRIPTION OF SYMBOLS 10 Surface shape measuring device 11 White interferometer 12 Horizontal surface moving stage 13 Data calculating part 15 Interference objective lens 16 Measuring actuator 17 Optical head 18 Light source 19 CCD camera 20 Z-axis drive part 21 Device base 30 Measurement object 31 1st Measurement surface region 32 Second measurement surface region 33 Overlap region 34 First surface measurement data 35 Second surface measurement data 36 Composite surface measurement data 40 Camera field of view

Claims (6)

It is a surface shape measurement method for measuring the surface shape by connecting the surface shape measurement data measured in multiple times,
A first step of installing a measurement object on a horizontal plane moving stage whose straightness fixed with respect to the apparatus base is 5 nm or less when moving 5 mm ;
A second measurement unit obtains first surface measurement data by measuring the three-dimensional surface shape of the first measurement surface region of the measurement object with an optical surface shape measuring instrument fixed in the horizontal plane direction with respect to the apparatus base. And the process of
The first part and the second measuring surface region and the optical surface profile measuring instrument measurement field of view and is the horizontal moving stage the optical surface to match the having overlapping area overlapping the measurement surface area A third step of moving relative to the shape measuring machine ;
A fourth step of obtaining the second surface measurement data by measuring the three-dimensional surface shape of the second measurement surface region with the optical surface shape measuring instrument;
The first surface measurement data and the first surface measurement data so that the surface measurement data corresponding to the part of the first surface measurement data and the surface measurement data corresponding to the overlap region in the second surface measurement data overlap. A fifth step of translating one or both of the two surface measurement data only in the height direction ;
A sixth step of synthesizing the translated first surface measurement data and the second surface measurement data to generate combined surface measurement data;
A surface shape measuring method having   The surface shape measuring method according to claim 1, wherein the overlap region has an area of 10% or less of a measurement visual field of the optical surface shape measuring instrument.   The surface shape measuring method according to claim 1, wherein the overlap region has an area of 1% or less of a measurement visual field of the optical surface shape measuring instrument. Generation of the composite surface measurement data is a weighted average for the first surface measurement data and the second surface measurement data so that the first surface measurement data and the second surface measurement data are continuously connected. surface shape measuring method according to any one of claims 1 to 3 is the generation of synthetic surface measurement data synthesizing surface measurement data of the measurement surface areas overlap each other by the processing. The surface shape measuring method according to any one of claims 1 to 4 , wherein the optical surface shape measuring instrument is a white interferometer. The method according to any one of claims 1 to 5 , further comprising a seventh step of repeating the third to sixth steps after the sixth step, with the already measured measurement surface region as the first measurement surface region. The surface shape measuring method according to item .

Images