JP2000249511A - Plane interferometer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform the absolute correction of the flatness of a plane that is installed vertical to gravity by calibrating the secondary residue (residue that is fitted by a secondary function) of the absolute flatness error of a surface to be inspected and at the same time calibrating the secondary constituent (constituent that is fitted by the secondary function) of the absolute flatness error of the surface to be inspected. SOLUTION: Measurement light entering a Fizeau surface 2a is partially reflected and returns to the inside of an interferometer 1. The remaining light is transmitted and enters vertically to a surface 3a to be inspected of an object 3 to be inspected, and returns the same light path and interferes with reflection light in the interferometer 1, thus forming an interference fringe. The interference fringe is converted into an electrical signal by an image pick-up element, is subjected to operation analysis to measure the relative flatness of the surface 3a to be inspected for the Fizeau surface 2a. Then, in an arithmetic unit, a first calibration means calibrates the secondary residue of the absolute flatness error of the surface to be inspected as the surface accuracy error of the surface to be inspected, and a second calibration means calibrates the secondary constituent of the absolute flatness error of the surface to be inspected as the error constituent of the symmetrical shape (plane) of the surface to be inspected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a plane interferometer for calibrating flatness.

[0002]

2. Description of the Related Art Conventionally, as a method of calibrating flatness, two out of three flat plates having unknown flatness are adopted as a Fizeau surface and a test surface of a Fizeau interferometer, and interference measurement is performed. 2
Three sets of interferometric data are obtained by exchanging combinations of sheets, and these interferometric data are substituted into an equation for determining flatness to obtain an unknown number (three-plane alignment).

In the case where the measurement is performed in a state where the plane is set in parallel to the gravitational force, as described in Japanese Patent Application Laid-Open No. 6-281427, the reaction force against the gravitational force of the support portion has a directional distortion. For the first time, a special hold is applied to give
Calibration of flatness becomes possible.

[0004]

However, when the measurement is performed in a state where the plane is set perpendicular to the gravity, it is unavoidable that the deflection of the plane is added as an unknown number, and the three-plane alignment as described above is performed. Was not applicable.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned drawbacks of the related art, and has an object to enable absolute calibration of flatness even when measurement is performed in a state where a plane is installed perpendicular to gravity. And

[0006]

A first means for solving the above-mentioned problem is to convert a measurement light beam emitted from a light source into a measurement wavefront having the same geometric shape as the surface of a test object. By irradiating the test surface of the test object, the measurement wavefront reflected from the test surface and the reference wavefront emitted from the light source and converted to have a predetermined surface shape interfere with each other, By subjecting the interference fringes generated by the interference to photoelectric detection and analysis and processing, the surface accuracy error of the surface to be inspected (the undulation error component that does not include the error component of the geometric shape among the deviation from the geometric shape of the surface shape). , Wherein the geometric shape is a plane, and a secondary residual of an absolute flatness error of the test surface (an absolute flatness error is expressed by a quadratic function) as a surface accuracy error of the test surface. First calibration means for calibrating the fitted residual) and the geometry of the surface to be inspected A second calibration means for calibrating a quadratic component (a component fitted with a quadratic function) of an absolute flatness error of the surface to be inspected as a shape (plane) error component. This is a plane interferometer (claim 1).

According to this means, both the secondary residual of the absolute flatness error of the test surface and the secondary component of the absolute flatness error of the test surface are calibrated, so that the flatness of the test surface is corrected. Can be easily calibrated.

Incidentally, "error component of geometric shape", "surface accuracy error of surface to be inspected (surface accuracy error of plane)", "second-order residual of absolute flatness error of surface to be inspected", " The exact meaning of the "second order component of the absolute flatness error of the surface" will be described later in the section "Embodiments of the Invention" using mathematical expressions.

[0009] A second means for solving the above-mentioned problems is as follows.
The measurement light beam emitted from the light source is converted into a measurement wavefront having the same geometric shape as the test surface of the test object, and is irradiated on the test surface of the test object, and the measurement reflected from the test surface is measured. The wavefront and the reference wavefront emitted from the light source and converted to have a predetermined surface shape interfere with each other, and an interference fringe generated by the interference is photoelectrically detected and analyzed, thereby obtaining a surface of the surface to be measured. Accuracy error (of the deviation of the surface shape from the geometric shape,
An interferometer for measuring a waviness error component that does not include an error component of the geometric shape, wherein the geometric shape is a plane, and a surface accuracy error of the measurement wavefront is a quadratic of an absolute flatness error of the measurement wavefront. First calibration means for calibrating a residual (residual obtained by fitting the absolute flatness error with a quadratic function); and an absolute flatness of the test surface as an error component of the geometric shape (plane) of the test surface. A second interferometer comprising second calibration means for calibrating a second-order component of the error (a component fitted with a quadratic function).

In this means, after the reference surface is once calibrated absolutely using a ref prototype, the reference surface is
Since it is possible to use a dummy long mirror instead of the surface to be calibrated for the secondary component calibration, it is possible to improve the accuracy of the absolute calibration of the secondary component of the surface to be calibrated.

A third means for solving the above-mentioned problem is as follows.
The first means or the second means, wherein the second calibration means performs calibration using a secondary residual calibrated by the first calibration means. (Claim 3).

This makes it possible to simplify the second calibration means without lowering the accuracy of the absolute calibration of the flatness of the surface to be inspected.

A fourth means for solving the above-mentioned problem is as follows.
In any one of the first means to the third means, the second calibration means includes a shift means for shifting the test object so that the interference fringes do not disappear, and And tilt measuring means for measuring the tilt of the test object (claim 4).
It is.

In this means, when the second calibration means performs calibration, the test object is shifted so that the interference fringes do not disappear. Therefore, even if the test object is shifted, the measurement is performed. Can continue. In addition, since a means for measuring the amount of tilt of the test object from the shift direction of the test object is provided, the second calibration means can perform calibration in a form that incorporates a tilt error.

As another method for measuring the amount of tilt, there is a method that enables a tilt interference measurement by providing a deflecting means for deflecting the measurement wavefront in the direction of shift of the test object. According to this method, a part of the light beam emitted from the light source of the plane interferometer is deflected by the deflecting means in the shift direction of the test object (in most cases, the direction perpendicular to the traveling direction of the light beam). The amount of tilt of the test object can be grasped by causing the light to reflect off the reflection surface attached to the test object and interfere with the reference wavefront. By doing so, the tilt amount can be detected by sharing the components of the plane interferometer without providing a special tilt detecting device, and the device configuration can be simplified.

Note that, as in the case where the tilt deflection generated relatively between the monitor optical system and the plane Fizeau plate cannot be ignored compared with the accuracy required for the absolute calibration,
If the stability of the tilt measuring means is a problem, provide a stability measuring means for measuring the stability of the tilt measuring means,
It is preferable to correct the tilt measurement value based on the measurement value. Furthermore, if the interference measurement using the measurement wavefront is used as the stability measurement means, the device configuration can be simplified.

Further, by adding at least one optical element provided perpendicularly to the shift direction to at least one of the tilt measuring means and the stability measuring means, the tilt, And it is preferable to measure at least one of the stability.

In addition, the tilt of the test object and the stability of the tilt measuring means can be measured at the same time, so that even when there is a relative displacement between the optical elements due to vibration or the like, the tilt can be measured. This is preferable because the influence can be eliminated.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram for explaining the basic principle of the present invention, and is a schematic diagram of an example of the first embodiment of the present invention. In FIG. 1, 1 is an interferometer,
2 is a plane Fizeau plate, 2a is a Fizeau surface, 3 is a test object,
3a is a test surface, 4 is a reference surface, 5 is a monitor optical system, and 6 is an auxiliary mirror surface.

The plane wave emitted from the interferometer 1 is "F"
Is perpendicularly incident on the Fizeau surface 2a of the planar Fizeau plate 2 marked with the symbol. He-N having a wavelength of 633 nm is used for the measurement light.
Generally, an e-laser light source is used.
In order to avoid interference with the internal optical system, it is necessary to provide a wedge angle on the plane Fizeau plate 2 or to provide an AR coating (antireflection film) on one of the parallel plane plates.
In FIG. 1, the former means is employed for the test object 3 and the latter means is employed for the plane Fizeau plate 2. In the following figures, whichever means is adopted, it is illustrated by a parallel plane.

Next, a part of the measurement light incident on the Fizeau surface 2 a is reflected as it is, and returns to the inside of the interferometer 1. The remaining measurement light is transmitted, enters perpendicularly onto the test surface 3a of the test object 3 with the symbol "W", and returns to the same optical path in the opposite direction. Then, the reflected light interferes with the inside of the interferometer 1 to form an interference fringe. This interference fringe is converted into an electric signal by an image sensor (not shown), and the information is subjected to arithmetic analysis processing by an arithmetic unit (not shown), whereby the Fizeau surface
The relative flatness of the test surface 3a with respect to a can be measured.

As described above, since the interference measurement is a relative measurement to the last, the Fizeau surface 2a serving as a reference surface is formed with high precision as much as possible, but cannot be an absolute plane. It is an object of the present invention to calibrate the error that the reference plane has with respect to the absolute plane (referred to as “absolute flatness error”).
For this purpose, an interference measurement device disclosed in Japanese Patent Application Laid-Open No. H5-222337 is used. This measurement method,
This will be briefly described with reference to FIG.

In FIG. 18, the rotation averaged data measured while applying rotation to the surface to be inspected and averaged in the arithmetic unit is
It is subtracted from the reference data in the direction of the rotation angle of 0 °. As a result, the absolute flatness error of the Fizeau surface is canceled, and the non-rotationally symmetric component of the test surface is absolutely obtained. The above procedure is called ABS-0 in the figure. This ABS
The value of −0 is applicable from a plane having no axis to the surface to be tested to a rotationally symmetric aspherical surface having an axis.

Next, the surface to be inspected is shifted laterally, and the measurement is performed.
Subtract from the data in degrees. As a result, the absolute flatness error of the Fizeau surface is offset, and the absolute flatness error of the test surface is laterally shifted and subtracted. The laterally shifted subtraction data is further corrected using the above-described non-rotationally symmetric component, so that the laterally shifted subtraction data of the rotationally symmetric component of the test surface is absolutely obtained.

From this data, a rotationally symmetric component is absolutely obtained by utilizing the properties of the contour line of the rotationally symmetric component, and is added to the non-rotationally symmetric component. It becomes measurable. The above procedure is illustrated in the figure.
It is called ABS-3. Since the ABS-3 is used in combination with a lateral shift, the application of an aspheric surface is limited to a mild aspheric surface. In addition, in order to perform ABS-3 accurately, a means for accurately grasping the lateral shift amount is required.

Hereinafter, it will be described that the error of the secondary component is mixed in the absolute flatness error of the surface to be detected due to the lateral shift.
For example, Z1 = C _01T (X ² + Y ² ) + C ₀₂ (X ² + Y ² ) ² + C ₀₃ (X ² + Y ² ) ³ +..., (1), when the lateral shift amount s is given in the X-axis direction, the absolute flatness error of the test surface excluding the non-rotationally symmetric component is Z2 = C _01T {( X−s) ² + Y ² } + C ₀₂ {(X−s) ² + Y ² } ² + C ₀₃ {(X−s) ² + Y ² } ³ + ...… (2) You. Thus, lateral shift subtraction data of the test surface is, Z3 = Z1-Z2 = 2sX <C 01T + C 02 [(X 2 + Y 2) + {(Xs) 2 + Y 2}] + ···> ... It is represented by (3). It has been confirmed by simulation that components other than the constant term in <> can be ignored. In the end, the component of Z3 = 2C _01T sX… (4) is regarded as the tilt error, and the tilt error (alignment error) of the interferometer Mixed in.

This alignment error is an unavoidable error because the interferometer is a relative measurement, and is an apparent aberration error caused by misalignment.
The work of removing this inside the arithmetic unit is called alignment error correction. Since the tilt component represented by the equation (4) is also removed by this alignment error correction, the secondary coefficient _{C01T of} the absolute flatness error cannot be accurately obtained.

[0028] However, "the components other than the constant term, is confirmed from the simulation can be ignored" in other words the "C ₀₂ or more coefficients are exactly can be measured is ascertained from simulation" are. This component is represented by Z4 = C ₀₂ (X ² + Y ² ) ² + C ₀₃ (X ² + Y ² ) ³ +... (5), and a second-order component that cancels this component in rms. Z5 = C _01H (X ² + Y ² ) (6) is also uniquely determined by the arithmetic processing. That is, Z6 = C _01H (X ² + Y ² ) + C ₀₂ (X ² + Y ² ) ² + C ₀₃ (X ² + Y ² ) ³ +... The quadratic component obtained by quadratic fitting is zero, and mathematically speaking, the residual (second narrow residual in a narrow sense) obtained by quadratic fitting the absolute flatness error excluding the non-rotationally symmetric component of the equation (1), that is, “ Second-order residual of the absolute flatness error of the test surface excluding the non-rotationally symmetric component.

A component obtained by adding the above-described non-rotationally symmetric component to the narrowly defined second-order residual is a broadly-defined second-order residual, and this undulation error component is defined as “absolute surface precision error of a plane”. . This can be absolutely calibrated by the method of ABS-3. A component obtained by rotationally averaging the second-order residual in a broad sense in the arithmetic device is a non-rotationally symmetric component.

The error component represented by the following equation: Z6 = C _01T (X ² + Y ² ) −C _01H (X ² + Y ² ) (8) is referred to as “error component of the geometric shape of the plane”. Is defined. Numerically, it is “a quadratic component obtained by quadratic fitting the absolute flatness error of the formula (1) excluding the non-rotationally symmetric component”. Also, due to the nature of the non-rotationally symmetric component, the absolute flatness error including the actual non-rotationally symmetric component is equal to the quadratic component obtained by quadratic fitting. The calibration of the secondary component can be performed by accurately grasping the tilt component at the time of shifting the surface to be inspected as described above.

The calibration of the second-order residual (absolute surface precision error of the plane to be inspected) and the secondary component (error component of the geometric shape of the plane to be inspected) in a broad sense can be performed at the same time. It is also possible to do it. First, the case where the operations are performed simultaneously will be described with reference to FIG.

The test object 3 is rotated around the rotation axis to collect data, and the above-described ABS-3 processing is performed. This AB
By the process of S-3, a second-order residual in a broad sense (absolute surface accuracy error of the test plane) is calibrated.

The calibration of the secondary component (error component of the geometric shape of the plane to be inspected) can be performed simultaneously with the execution of ABS-3. That is, when the lateral displacement of the ABS-3 is performed by the moving means (not shown) of the test object 3 in the direction of the arrow in the drawing, the reference mirror 4 is moved to the holding means 7 of the test object 3 in the direction of the lateral displacement. And installed vertically. Thereby, when the test object 3 is shifted laterally, the tilt in the lateral shift direction generated on the test surface 3 a can be made equal to the tilt in the gravity direction generated on the reference mirror 4. This reference mirror 4
Is monitored by the monitor optical system 5.

The monitor optical system 5 itself is mounted on a stable base common to the plane Fizeau plate 2, but the tilt of the tilt generated between the monitor optical system 5 and the plane Fizeau plate 2 is relatively large. If the accuracy required for absolute calibration cannot be neglected as compared with the absolute calibration, as one means, the measurement light emitted from the interferometer 1 is used to simultaneously or individually measure the vibration of the monitor optical system 5 with the measurement of the surface 3a to be measured. Need to be measured. An auxiliary mirror 6 for that purpose is provided so as to be integrated with the monitor optical system. The monitor optical system may be a second interferometer or a collimator.

Note that "g ↓" shown in the figure represents the direction of gravity, and it is essential that the direction of rotation and lateral displacement of the surface to be measured be perpendicular to the axis of gravity. In addition, the test surface 3
The absolute flatness of a refers to the surface to be inspected in a state where the surface is subjected to surface distortion due to the reaction force of the holding means 7 against gravity,
Now, focusing on only one axis, the tilt error in the absolute space, which occurs in the reference mirror 4 before and after the lateral displacement, can be referred to as Tz. In addition, the alignment shift calculated by the alignment error correction of the difference data before and after the lateral shift is also Ta. Further, the tilt shake of the monitor optical system 5 before and after the lateral shift can be set to Tm. At this time, the relationship Ta− (Tz−Tm) = C01T (X2 + Y2) −C01H (X2 + Y2) (9) holds, so that the secondary component (error component of the geometric shape of the plane to be measured) is satisfied. Absolute calibration becomes possible.

The actual tilt component generated by the lateral shift is not guaranteed to be only the X tilt, and the Y tilt is also generated at the same time. Further, since the measurement area of the test surface visually recognized by the aperture of the measurement light is different due to the lateral shift, even after the alignment error correction of the lateral shift difference data due to the non-rotationally symmetric component of the test surface 3a, ( An error is added to the tilt component to be obtained by Ta in equation (9).

As a countermeasure for this, first, the lateral shift is made to have symmetry with respect to the sampling of the image sensor. The same consideration is given to the measurement area on the surface to be inspected by masking or black coating. Then, the tilt error due to the non-rotationally symmetric component of the surface to be measured is also calculated by A
The laterally shifted subtraction data is corrected using the second-order residual (non-rotationally symmetric component) of the test surface obtained by BS-0.

Due to these considerations, it is only necessary to extract a one-axis tilt error component in the lateral shift direction as a tilt component to be monitored.

Hereinafter, a second embodiment of the plane interferometer of the present invention will be described with reference to FIGS. This embodiment is an example of individually calibrating the above-described second-order residual (absolute surface accuracy error of a test plane) and secondary component (error component of a geometric shape of a test plane) in a broad sense. . In the following drawings, components having the same reference numerals as those shown in the preceding drawings indicate the same components. In FIG. 2, reference numeral 30 denotes a ref prototype, and 30a denotes a ref prototype surface.

In this embodiment, first, the ref prototype 30 is prepared separately from the test object, and the above-described ABS-3 is applied to the ref prototype. Thereby, the secondary residual of the Fizeau surface 2a is calibrated simultaneously with the secondary residual of the ref prototype surface 30a.
That is, the second-order residual of the measurement wavefront (absolute surface accuracy error of the measurement wavefront) is calibrated. At this time, in order to withstand the absolute calibration, the condition of the Fizeau surface must not be changed by adjusting the environmental conditions, or the Fizeau surface must be subtracted (ref. Calibration of the surface is assumed.

In the case where the existence of an error factor which gives a bias to the phase error when performing the interference measurement and which is applied to the measurement light can be neglected between the interference arm of the Fizeau surface and the test surface, the Fizeau surface can be rephrased as the measurement wavefront. Good. When the wavefront incident on the Fizeau surface has good coherency of the measurement light,
Since the interference measurement is a relative phase difference measurement between the Fizeau surface and the test surface, it is not necessary to use a perfect plane wave. Therefore,
A more accurate expression of “measurement wavefront” is “measurement wavefront that is considered by virtually resetting the phase by the Fizeau surface”.

Next, as shown in FIGS. 3 to 5, a long mirror 300 is used as a dummy element instead of the ref prototype.
The secondary component of the dummy surface 3a (error component of the geometric shape of the test plane) is calibrated.

At this time, as a means common to FIG. 1, a reference mirror 4 is installed integrally with the long mirror 300 on a long mirror holding means (not shown), and the monitor optical system 5 controls the direction of the lateral shift. The tilt of the long mirror 300 is monitored in parallel.
At this time, the tilt of the monitor optical system 5 itself is changed by the measuring light emitted from the interferometer 1 by the long mirror 3.
The measurement is performed simultaneously with or separately from the measurement of the dummy surface 300a. Therefore, the auxiliary mirror 6 is provided in the monitor optical system. The monitor optical system may be a second interferometer or a collimator. This monitor optical system corresponds to "means for measuring the stability of the tilt measuring means" described in "means for solving the problem".

Unlike FIG. 1, as is apparent from FIG.
Since the dummy surface 300a is larger than the effective luminous flux of the Fizeau plane 2a, it is necessary to increase the stroke of the moving means for laterally shifting and to accurately grasp and control the amount of laterally shifting.

FIG. 6 is an explanatory diagram of the wavefront synthesis of the data of the dummy surface measured in FIGS. By dividing the horizontal shift amount by the interferometer magnification “mm / pixel”,
The number of pixels laterally displaced equivalently is calculated in the arithmetic unit, and by returning this amount by this amount in the arithmetic unit, as shown in the figure, the surface accuracy error (undulation error: second-order residual in a broad sense) of the dummy surface 300a is reduced. Will be reproduced.

In this superimposed area (hereinafter referred to as "overlapping area"), the surface accuracy error of the reflex surface 30a of the reflex prototype 30 or the Fizeau surface 3a is superimposed.
Correcting the bias in the overlap region using the second-order residual (absolute surface accuracy error) of the reflex surface 30a or the Fizeau surface 3a, which is obtained in advance in S-3, is more accurate. This is necessary for calibration of the secondary component.

By using the long mirror 300, it is possible to improve the detection sensitivity of the secondary component including the deflection due to its own weight. Using the secondary component of the long mirror, the Fizeau surface, and hence the surface to be inspected, can be improved. The secondary component can also be obtained with high accuracy. Further, by using the long mirror 300 in which the secondary component is calibrated, it is possible to improve the accuracy in performing squareness of two surfaces.

FIG. 7 is a diagram showing a third example of the embodiment of the present invention. In this embodiment, the basic configuration and the arrangement of the devices are the same as in FIG. 3, but an interferometer is used instead of the monitor optical system 5 for monitoring the tilt of the reference mirror 4 from the laterally displaced direction of the movable mirror. The difference is that a deflecting means 8 for deflecting the measurement light from No. 1 and performing interference measurement of the tilt of the reference mirror 4 is used. As the deflecting means 8,
Although a right-angled triangular prism is used, a reflecting mirror may be used.

The deflecting means 8 itself is a flat Fizeau plate 2
If the tilt deflection generated between the deflecting means 8 and the plane Fizeau plate 2 cannot be neglected compared to the accuracy required for absolute calibration, As another means, it is necessary to measure the deflection of the deflecting means 8 at the same time as or separately from the measurement of the dummy surface 300a using the measurement light emitted from the interferometer 1. An auxiliary mirror 6 for that purpose is provided so as to be integrated with the deflecting means 8. It goes without saying that the deflecting means 8 and the auxiliary mirror 6 may be formed as one optical element.

In order to reduce the angle accuracy of these optical elements, the reference mirror 4 is variable so that the angle can be adjusted. Also in this case, the reference mirror 4 is the long mirror 3
It is attached to a long mirror holding means (not shown) so as to move integrally with 00.

FIG. 17 shows a modification of the auxiliary mirror and the deflecting means shown in FIG. 7, in which a right-angled triangular prism 8 as a deflecting means is attached to the auxiliary mirror 6. In this case, a part of the light from the interferometer 1 in FIG. 7 is reflected by the auxiliary mirror 6 and interferes with the reflected light from the Fizeau surface 2a. The other part is reflected at right angles on the prism surface and enters the reference mirror 4. Then, the reflected light is reflected again at a right angle by the prism surface and interferes with the reflected light from the Fizeau surface 2a.

In the third embodiment, the broadly defined 2
The method for calibrating the secondary residual (absolute surface precision error of the test plane) and the secondary component (error component of the geometric shape of the test plane) is the same as the method in the second embodiment.

Hereinafter, the second embodiment described above,
Various modifications of the third embodiment will be described. FIG. 8 shows an example in which two reference mirrors 4 are provided on both end surfaces of the long mirror 300, and an improvement in the monitoring accuracy of the tilt of the long mirror 300 can be expected due to the averaging effect. FIG. 8A shows a case in which the monitor optical system 5 is adopted, in which the function of the monitor optical system 5 is increased to two for one auxiliary mirror 6. FIG. 8 (b) shows a case where a deflecting means is employed.
In this example, the number of reference mirrors can be adjusted, as in FIG.

FIG. 9 shows a case where two monitor optical systems 5 are employed, and shows another example in which the function of the monitor optical system 5 is increased with respect to the auxiliary mirror. As with the measurement arrangement of FIG.
The averaging effect can be expected to improve the monitoring accuracy of the long mirror tilt. FIG. 9A shows an example in which the number of monitor optical systems is increased to two symmetrically with respect to the long mirror 300 while the number of reference mirrors 4 remains unchanged. FIG. 9 (b)
The same measuring element as in FIG. 9A is used as in FIG.
This is an example in which the number is symmetrically increased to two. In any measurement, a deflection unit may be employed instead of the monitor optical system 5.

FIG. 10 is symmetrical with respect to the long mirror 300.
This is an example in which the number of auxiliary mirrors 6 is increased to two, and is attached to the monitor optical system 5 via mirror holding means 9. The function of the monitor optical system 5 may be increased or decreased in accordance with the increase or decrease in the number of the reference mirrors 4.

Although the number of the monitor optical system 5 is one in the figure, two laser measuring devices are provided in the direction of gravity, and the amount of change in the difference between the measured values of the laser measuring devices is changed by two lasers. By dividing by the distance between the length measuring devices, it is possible to use the monitor optical system for the tilt of the long mirror 300 as an alternative.

FIG. 11 shows an example in which the functions of the auxiliary mirror 6 and the deflecting means 8 are achieved by a right-angled triangular prism 68 having the same shape as in FIG. This optical element includes an auxiliary mirror surface 68a.
And the deflecting mirror surface 68b are formed at 45 degrees, and the deflecting mirror surface 68b is coated with a high reflection surface. An antireflection film is formed on a surface parallel to the reference mirror 4. The auxiliary mirror surface 68a may be a half mirror.
It is also necessary to use a light-shielding plate (not shown) in combination in order to perform masking as needed and to avoid three-beam interference. FIG.
Similarly to the above, adjustment of the reference mirror 4 is enabled.

Hereinafter, a fourth embodiment of the present invention will be described. The embodiment shown in FIG. 12 consists of almost the same arrangement of measuring instruments as in FIG.
Instead of this, the end surface of the long mirror 300 is finished to a high-precision surface and used as a reference surface 40. In order to monitor the reference plane 40 from a direction parallel to the lateral shift, an angle is set so as to be orthogonal to the lateral shift direction and the dummy surface 300a. Then, the second interferometer 10 is provided so that the Fizeau surface 20a of the plane Fizeau plate 20 forms interference fringes with the reference surface 40.

On the other hand, the monitor optical system 5 monitors the tilt of the Fizeau surface 20a. With the above arrangement, the monitor optical system 5 can indirectly measure the tilt of the reference surface 40, that is, the dummy surface 300a via the plane Fizeau plate 20.

In place of the second interferometer 10, a second
Monitor optical system may be adopted. At this time, it is necessary to prepare an auxiliary mirror surface instead of the Fizeau surface 20a, and set the second monitor optical system and the auxiliary mirror surface integrally.

In the fourth embodiment, in a broad sense, 2
The method for calibrating the secondary residual (absolute surface precision error of the test plane) and the secondary component (error component of the geometric shape of the test plane) is the same as the method in the second embodiment.

FIG. 13 is a diagram showing a fifth embodiment of the present invention. This embodiment has a measurement device arrangement similar to that of the embodiment shown in FIG. 12, and the reference plane 40, the plane Fizeau plate 20, and the second interferometer 10 are exactly the same as those in FIG. It is. The auxiliary mirror surface 6 is integrated at a position orthogonal to the auxiliary mirror surface 6 so as to be integrated therewith.
1 is prepared, and instead of monitoring the Fizeau surface 20a with the monitor optical system, the auxiliary mirror surface 6 is formed by the second interferometer.
By monitoring 1, it is possible to indirectly measure the stability of the second monitor optical system composed of the second interferometer and the plane Fizeau plate 20 via the auxiliary mirror surface 61. .

In the fifth embodiment, 2 in a broad sense
The method for calibrating the secondary residual (absolute surface precision error of the test plane) and the secondary component (error component of the geometric shape of the test plane) is the same as the method in the second embodiment.

FIG. 14 shows a sixth embodiment of the present invention. The monitor optical system 5 and the reference plane 40 have the same arrangement as in the embodiment shown in FIG.
By providing the 5-degree bending mirror 80, the measurement light of the monitor optical system 5 is deflected in the direction of gravity, and the Fizeau surface 20a of the flat Fizeau plate 20 is monitored. Second interferometer 10 and Fizeau surface 20a of plane Fizeau plate 20
Are the same as those in FIG.

In the case where the simultaneous measurement of the measurement of the dummy surface 300a and the auxiliary mirror surface 6 and the measurement of the reference surface 40 and the auxiliary mirror 6 are not required, as shown in the drawing, the first interferometer 1 is used. An alternative is possible by shifting to the position of the second interferometer 10. Also, the first interferometer 1
Alternatively, it is possible to switch the optical path in the middle.

Also in the sixth embodiment, a method of calibrating the second-order residual (absolute surface precision error of the plane to be inspected) and the second-order component (error component of the geometric shape of the plane to be inspected) in a broad sense. Is
This is the same as the method in the second embodiment.

FIG. 15 is a diagram showing a seventh embodiment of the present invention. The measuring method in this embodiment is shown in FIG.
Similarly to the embodiment shown in FIG.
13 is the same as the measurement method of the embodiment shown in FIG. 13 except that the measurement light of the second interferometer 10 is horizontally deflected from the direction of gravity by providing 0. Also, the first
The relationship between the interferometer 1 and the second interferometer 10 is the same as that described with reference to FIG.

FIG. 16 is a diagram showing an eighth embodiment of the present invention. The measuring method in this embodiment is shown in FIG.
5 is similar to the measuring method of the embodiment shown in FIG. 5, but the measuring light emitted from the interferometer 1 is split by the half mirror 81 in two directions after passing through the Fizeau surface 2a. The measurement light reflected and divided in the direction of gravity interferometrically measures the long mirror 300 and the mirror surface of the auxiliary optical element 62.
On the other hand, the transmitted measurement light interferometrically measures the reference surface 40 and the mirror surface of the auxiliary optical element 62 via the bending mirror 82. It is difficult to perform both interference measurements at the same time, and a shutter (not shown) may need to avoid three-beam interference. Note that the auxiliary optical element 62 is not required as long as the perpendicularity of the bending mirror 82 is sufficiently accurate.

Also in the seventh and eighth embodiments, the second-order residual (absolute surface precision error of the plane to be inspected) and the second-order component (error of the geometric shape of the plane to be inspected) are broadly defined. The method of calibrating the component) is the same as the method in the second embodiment.

When the stability of the tilt of the test object and the tilt measuring means described above is measured by the monitor optical system or the interferometer, it is preferable to perform each measurement at the same time. This makes it possible to eliminate the influence of shaking due to relative vibration or the like even if the measurement data has such an influence.

[0071]

As described above, when the plane interferometer according to the present invention is employed, not only the second-order residual of the plane surface accuracy error but also the absolute calibration of the second-order component can be performed. Further, by using a long mirror whose secondary component is calibrated by the plane interferometer according to the present invention, it is possible to improve the accuracy of perpendicularity of two surfaces.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining a basic principle of the present invention and is a schematic diagram of an example of a first embodiment of the present invention.

FIG. 2 is a schematic diagram of an example of a second embodiment of the present invention.

FIG. 3 is a schematic diagram of an example of a second embodiment of the present invention.

FIG. 4 is a schematic diagram of an example of a second embodiment of the present invention.

FIG. 5 is a schematic diagram of an example of a second embodiment of the present invention.

FIG. 6 is an explanatory diagram of wavefront synthesis of data of a dummy surface measured in the second embodiment.

FIG. 7 is a schematic diagram of an example of a third embodiment of the present invention.

FIG. 8 is a schematic diagram of a modification of the third embodiment of the present invention.

FIG. 9 is a schematic diagram of a modification of the third embodiment of the present invention.

FIG. 10 is a schematic diagram of a modification of the third embodiment of the present invention.

FIG. 11 is a schematic diagram of a modified example of the third embodiment of the present invention.

FIG. 12 is a schematic diagram of an example of a fourth embodiment of the present invention.

FIG. 13 is a schematic diagram of an example of the fifth embodiment of the present invention.

FIG. 14 is a schematic diagram of an example of the sixth embodiment of the present invention.

FIG. 15 is a schematic diagram of an example of a seventh embodiment of the present invention.

FIG. 16 is a schematic diagram of an example of the eighth embodiment of the present invention.

FIG. 17 is a view showing a modification of the auxiliary mirror and the deflecting means.

FIG. 18 is a diagram showing an interference measurement method described in the publication.

[Explanation of symbols]

1: Interferometer, 2: flat Fizeau plate, 2a: Fizeau surface,
DESCRIPTION OF SYMBOLS 3 ... Test object, 3a ... Test surface, 4 ... Reference surface, 5 ... Monitor optical system, 6 ... Auxiliary mirror (surface), 8 ... Deflection means, 9 ... Mirror holding means, 20 ... Plane Fizeau plate, 20a ... Fizeau surface, 30: Ref prototype, 30a: Ref prototype, 40: Reference surface, 61: Mirror surface, 68: Triangular prism, 68
a: auxiliary mirror surface, 68b: deflection mirror surface, 80: bending mirror, 81: half mirror, 82: bending mirror, 300: long mirror, 300a: dummy surface

Continued on the front page (72) Inventor Takahiro Yamamoto 3-2-3 Marunouchi, Chiyoda-ku, Tokyo Inside Nikon Corporation (72) Inventor Yoshikazu Sugiyama 3-2-2, Marunouchi, Chiyoda-ku, Tokyo Nikon Corporation F term (reference) 2F064 AA09 DD01 DD09 EE05 FF01 GG12 GG13 GG22 HH03 HH08 JJ01 2F065 AA47 AA54 DD04 EE00 FF52 FF61 JJ03 JJ26 LL12 LL17 LL30 LL46 MM03 MM14 MM24 MM28 QQ

Claims (4)

[Claims] 1. A light beam for measurement emitted from a light source is converted into a measurement wavefront having the same geometric shape as the surface of the object to be measured and radiated onto the surface of the object to be measured. The reflected measurement wavefront and the reference wavefront emitted from the light source interfere with each other, and the surface accuracy error of the interference fringe test surface caused by the interference (of the deviation from the geometric shape of the surface shape, the geometric shape The undulation error component does not include the undulation error component), wherein the geometrical shape is a plane, and the surface accuracy error of the test surface is a quadratic residual of the absolute flatness error of the test surface. First calibration means for calibrating the difference (residual obtained by fitting the absolute flatness error with a quadratic function); and an error component of the absolute flatness error of the test surface as an error component of the geometric shape (plane) of the test surface. Second calibration means for calibrating the secondary component (the component fitted with the quadratic function) A plane interferometer, comprising: 2. A measuring light beam emitted from a light source is converted into a measuring wavefront having the same geometric shape as the surface to be inspected of the object, and is irradiated onto the surface to be inspected of the object. The measurement wavefront reflected from the light source, the reference wavefront emitted from the light source and converted to have a predetermined surface shape interfere with each other, by performing an interference fringe photoelectrically detected and analyzed by the interference processing, An interferometer for measuring a surface accuracy error of a test surface (a swell error component that does not include a geometric error component among deviations of the surface shape from the geometric shape), wherein the geometric shape is a flat surface. And the secondary residual of the absolute flatness error of the measurement wavefront (the absolute flatness error is 2
First calibration means for calibrating the residual fit by a quadratic function, and a quadratic component (a quadratic function) of the absolute flatness error of the test surface as an error component of the geometric shape (plane) of the test surface. A second calibrating means for calibrating the fitted component). 3. The plane interferometer according to claim 1, wherein said second calibration means performs calibration using a second-order residual calibrated by said first calibration means. A plane interferometer characterized by performing. 4. One of claims 1 to 3
The plane interferometer according to claim, wherein the second calibration means shifts the test object so that the interference fringes do not disappear, and the shift direction of the test object from the direction of the shift. A plane interferometer comprising: tilt measuring means for measuring tilt.