JP2000275021A - Apparatus for measuring shape of face - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a face shape-measuring apparatus which can highly accurately measure a face shape with the use of a phase shift method even when the face vibrates or the like when the face is measured. SOLUTION: A reflecting light from an area 5a of a reference face of a Fizeau plane member 5 and a reflecting light from a face 6 to be measured are made to interfere, and a phase distribution is measured. A face shape of the face 6 to be measured is measured with the use of a phase shift interference method. In this case, a reflecting light from an area 5b and a reflecting light from a reflecting plane member 8 as a standard face are made to interfere and, a phase distribution is measured at the same time. A phase measurement error is obtained from the latter phase distribution, based on which the former phase distribution is corrected. Since the measured value is corrected on the basis of the error calculated when the standard face is simultaneously measured, even an effect of errors due to vibrations or the like can be eliminated, and the face shape can be correctly measured.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a surface shape measuring apparatus for measuring a surface shape used for an optical element such as a lens or a mirror with high accuracy, and more particularly to a phase shift interferometer. More specifically, the present invention relates to a surface shape measuring apparatus which can remove an error due to vibration or the like during measurement and can perform accurate measurement.

[0002]

2. Description of the Related Art In recent years, with the demand for high-precision optical instruments,
Optical elements such as lenses and mirrors that constitute such equipment tend to be highly accurate. Therefore, a surface shape measuring device for measuring the surface shape of the optical element is required to have the same high accuracy.

An interferometer such as a Fizeau interferometer is widely used for measuring the surface shape, and a well-known phase shift method is used for measuring the phase distribution of interference fringes with high accuracy.

[0004]

However, when the phase distribution of the interference fringes is measured by the phase shift method, the phase distribution is affected by external vibrations and errors in the driving device for generating the phase shift. There is a problem that an error occurs in the shift amount and an error occurs in the phase measurement.

Since the phase measurement error due to the vibration and the like becomes larger as the width of the phase distribution of the detected interference fringes becomes wider, it is necessary to measure a surface shape having a large shape error or to use an aspherical surface shape using a spherical interferometer. This is particularly problematic when measuring the temperature.

The present invention has been made in order to solve such a problem. Even if there is vibration during measurement or there is an error in the amount of phase shift, these effects can be compensated for and the phase shift can be compensated. An object of the present invention is to provide a surface shape measuring device capable of measuring a surface shape with high accuracy using a method.

[0007]

The gist of the present invention is to provide a means for measuring a phase measurement error generated by vibration or the like,
Based on the measured error, the data obtained by measuring the surface to be measured is corrected to obtain accurate measurement data of the surface to be measured, and to measure the surface to be measured when the measured error is small. By adopting the data as the measurement data, accurate measurement data of the test surface is obtained.

[0008] That is, the first to solve the above-mentioned problems.
Means for irradiating a part of the light beam emitted from the light source to the surface to be measured as a measurement light beam, and a part of the measurement light beam reflected by the surface to be measured and a part of the light beam emitted from the light source, and A surface profile measuring apparatus for causing a reference light beam having a wavefront of the type to interfere with each other and measuring a phase distribution of an interference fringe (measurement fringe) generated by the interference by a phase shift interferometry, wherein (a) reflection is performed on the surface to be measured Means for forming an error detection measurement light beam having a constant phase relationship with the measured light beam, and (b) means for forming an error detection reference light beam having a constant phase relationship with the reference light beam, c) means for causing the measurement light beam for error detection and the reference light beam for error detection to interfere with each other, and observing the phase distribution of interference fringes (error detection interference fringes) caused by the interference simultaneously with the phase distribution of the measurement interference fringes. (D) Measurement of the phase distribution of the interference fringes for error detection. Means for obtaining a phase measurement error from the constant data, and (e) means for using the obtained phase measurement error to correct the measurement error of the shape of the test surface obtained from the measurement interference fringes. A surface shape measuring device (claim 1).

In this means, the phase distribution of the interference fringes for error detection is measured simultaneously with the phase distribution of the measured interference fringes. In the means for creating the error detection measurement light beam, a light beam from a surface having a known surface shape, such as a reference reflection surface, can be used as the error detection measurement light beam. Therefore, in this case, as described later in the description of the embodiments of the invention, a phase shift method is applied to the interference fringes for error detection, and a known surface is obtained from the data by fitting calculation or the like. By calculating the error of the above, the amount of the phase measurement error can be obtained. Therefore, by correcting the data of the phase distribution of the measured interference fringes based on the amount of the phase measurement error, the shape of the test surface can be accurately measured.

[0010] Even when a light beam from a surface whose surface properties are not known is used as a measurement light beam for error detection, this surface is made substantially parallel to the reference surface, as described later in the section of the embodiment of the invention. By applying the phase shift method above, the shape of this surface can be determined in advance. As a result, this surface becomes a known surface, and by applying the above-described method, the shape of the test surface can be accurately measured.

A second means for solving the above-mentioned problem is as follows.
While irradiating a part of the light beam emitted from the light source to the surface to be measured as a measurement light beam, the measurement light beam reflected by the surface to be measured and a part of the light beam emitted from the light source and having a predetermined wavefront A surface shape measuring apparatus for causing a reference light beam to interfere with each other and measuring a phase distribution of an interference fringe (measurement interference fringe) generated by the interference by a phase shift interferometry, wherein (a) the measurement reflected on the surface to be measured Means for forming an error detection measurement light beam having a constant phase relationship with the light beam, (b) means for forming an error detection reference light beam having a constant phase relationship with the reference light beam, and (c) the error Means for causing the measurement light beam for detection and the reference light beam for error detection to interfere with each other, and observing the phase distribution of interference fringes (error detection interference fringes) caused by the interference simultaneously with the phase distribution of the measurement interference fringes; Is it the measured data of the phase distribution of the interference fringes for error detection? Means for obtaining a phase measurement error from the data, and (e) when the amplitude of the obtained phase measurement error is smaller than a predetermined value, the measurement data of the shape of the test surface obtained from the measurement interference fringes, A surface shape measuring device (claim 2).

In this means, the phase measurement error is obtained from the measurement data of the phase distribution of the error detecting interference fringe in the same manner as in the first means. When the amplitude of the phase measurement error is smaller than a predetermined value, it is determined that the phase measurement error is small and can be ignored, and the measurement data of the shape of the test surface obtained from the measured interference fringes at that time is used as the shape of the test surface. To be adopted. Therefore, the shape of the test surface can be accurately measured without performing complicated correction calculations.

A third means for solving the above-mentioned problem is as follows.
The second means includes means for intentionally adding an error to the amount of phase shift to be given by the phase shift interferometry (claim 3).

In this means, the state where the amplitude of the phase measurement error is smaller than a predetermined value, as described in the second means, is created by intentionally adding an error to the phase shift amount and performing the measurement. Then, the measurement data of the shape of the test surface obtained from the measured interference fringes at that time is adopted as the shape of the test surface. This increases the frequency of occurrence of a state in which the amplitude of the phase measurement error is smaller than the predetermined value, so that the second means can be used effectively.

[0015] A fourth means for solving the above problems is as follows.
Any one of the first means to the third means, wherein a surface shape error of the means for forming the error detection reference light beam and the surface shape error of the means for forming the error detection measurement light beam is a desired surface to be measured. The present invention is characterized in that it is sufficiently small with respect to the shape measurement accuracy (claim 4).

In this means, since the shape error of the surface serving as a reference for error detection is sufficiently small with respect to the desired measurement accuracy of the surface shape to be detected, the detection accuracy of the phase measurement error can be reduced to the desired measurement surface. The detection accuracy of the phase measurement error necessary for obtaining the shape measurement accuracy can be made sufficiently high. Therefore, the shape of the test surface can be accurately measured.

In each of the above means, it is preferable that the means for forming the reference light beam also serves as the means for forming the reference light beam for error detection, as will be described later in the embodiments of the invention. Accordingly, it is not necessary to separately provide a unit for forming the error detecting reference light beam, so that the apparatus can be inexpensive, and the phase relationship between the reference light beam and the error detecting reference light beam can be easily maintained constant. be able to.

In each of the above means, it is preferable that the means for forming the measurement light beam for error detection includes a reflecting member fixed to a holder on the surface to be detected. Accordingly, the phase relationship between the measurement light beam and the error detection measurement light beam can be easily maintained constant.

Further, in each of the above-mentioned means, it is preferable that the light beam for error detection and the light beam for surface to be measured are light beams emitted from the same light source. In this way, only one light source is required, and the apparatus can be inexpensive, and the wavelength of the light for error detection and the wavelength of the light for measurement of the surface to be measured are surely the same. Can be.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. 1 and 2 are views showing an example in which the present invention is applied to a plane measurement using a Fizeau interferometer. FIG.
FIG. 2 is a diagram showing the entire configuration of the apparatus, and FIG. 2 is a diagram showing the Fizeau plane member and the surface to be inspected in more detail.

The linearly polarized light beam emitted from the laser light source 1 is converted in beam diameter by a beam expander 2 and is incident on a polarization beam splitter 3 (hereinafter, referred to as "PBS"). The plane of polarization of the light beam is chosen to be reflected by the PBS3. The light beam reflected by the PBS 3 enters the Fizeau plane member 5 via the quarter-wave plate 4. The light beam transmitted through the region 5a of the reference surface of the Fizeau plane member 5 is incident on the surface 6 to be measured as a measurement light beam. The measurement light beam reflected by the test surface 6 is again incident on the PBS 3 via the Fizeau plane member 5 and the quarter-wave plate 4. Since the measurement light beam passes through the quarter-wave plate twice in a round trip and the polarization plane is rotated 90 degrees, PBS
3 is transmitted. The measurement light beam transmitted through the PBS 3 has its beam diameter converted by the beam diameter conversion optical system 9 and enters the two-dimensional image detector 10.

On the other hand, the reflected light on the reference surface area 5a of the Fizeau plane member 5 is incident on the PBS 3 via the quarter-wave plate 4 as a reference light flux. The reference light beam passes through the quarter-wave plate twice in a round trip and the polarization plane is rotated 90 degrees.
Transmit through S3. The beam diameter of the reference light beam transmitted through the PBS 3 is converted by the beam diameter conversion optical system 9 and enters the two-dimensional image detector 10, where interference fringes between the measurement light beam and the reference light beam are detected.

Note that the beam diameter conversion optical system 9 is
Is also formed on the two-dimensional image detector 10, and the distortion is suppressed in order to accurately know the shape of the surface 6 to be detected. By correcting the abscissa of the interference fringes using the design value or the measured value of the distortion, the coordinates on the test surface 6 and the coordinates on the two-dimensional image detector 10 can be accurately related.

In the present embodiment, the Fizeau plane member 5 has a reference surface area 5 a used for measuring the surface shape of the surface 6 to be measured.
The configuration is such that the light beam is incident on a wider area than that of the reference plane, and the part incident on the area 5b outside the area 5a on the reference surface is used for the error detecting light beam. Thus, the phase relationship between the reference light beam and the reference light beam for error detection is kept completely constant.

The error detecting reference light beam reflected by the reference surface region 5b passes through the quarter wavelength plate 4, passes through the PBS 3, and
The beam diameter is converted by the beam diameter conversion optical system 9 and is incident on the two-dimensional image detector 10. On the other hand, the measurement light flux for error detection transmitted through the area 5 b of the reference surface is incident on the reflection plane member 8 provided on the test surface holder 7. The measurement light flux for error detection reflected by the reflection plane member 8 enters the PBS 3 again through the Fizeau plane member 5 and the quarter-wave plate 4. PBS3
The measurement light flux for error detection which has passed through the
Then, the beam diameter is converted into the two-dimensional image detector 10, and the interference fringes of the measurement light beam for error detection and the reference light beam for error detection are detected.

The reflecting flat member 8 is preferably formed so as to have a sufficiently small surface shape error with respect to the shape measurement accuracy of the surface 6 to be measured, that is, to have a sufficient flatness, but this is not satisfied. Even in this case, as will be described later, it is possible to calculate and correct the shape error of the reflection plane member 8 from the absolute plane. Also, the reflection plane member 8
Is mounted on the same test surface holder 7 as the test surface 6, the phase relationship between the measurement light beam and the error detection measurement light beam is always kept constant. However, in order to make the number of interference fringes of the error detecting light beam variable, the reflecting plane member 8 has a mechanism capable of adjusting the tilt with respect to the surface 6 to be measured.

The output from the two-dimensional image detector 10 is taken into a computer (not shown), and the phase distribution of interference fringes is calculated using a phase shift interferometry. In the present embodiment, a phase shift is realized by slightly moving the test surface holder 7 in the optical axis direction by a piezo element (not shown).

Next, a method for correcting a surface shape measurement error will be described. In the phase shift interferometry, the phase difference between the measurement light beam and the reference light beam is moved by 2π or more, and the intensity of the interference fringe for each π / 2 is measured. There are various methods for calculating the phase φ in terms of the number of samples and the calculation formula. For example, the calculation formula of formula (1) is well known. tanφ = [(B ₀ + B ₄ ) / 2−B ₂ ] / (B ₁ −B ₃ ) (1) where B ₀ is the intensity of the interference fringe in the initial phase to be obtained, and B ₁ is + π / _2, B 2 is + π, B ₃ is + 3π /
2, B ₄ is the intensity of interference fringes when a phase shift of 2π is given.

Further, by generalizing the equation (1), when there is a linear shift in the phase shift amount, that is, when the phase shift amount in B _n is u × (nπ / 2), the phase φ is expressed by the equation (1). It can be calculated using 2). tan (φ + uπ) sin (πu / 2) = [(B ₀ + B ₄ ) / 2−B ₂ ] / (B ₁ -B ₃ ) (2) where u is a linear coefficient of a phase shift amount. It is. Equation (2)
And u = 1 corresponds to equation (1). A linear shift in the amount of phase shift occurs when the direction of the phase shift is different from the normal direction of the surface to be inspected (for example, when the surface to be inspected is a spherical surface),
This is a case where there is a linear shift in the displacement of the piezo element that gives a phase shift.

When the phase is measured in accordance with the phase calculation formula (1) or (2), an error occurs in the amount of phase shift due to the influence of external vibration or a driving error of the piezo element.
As a result, errors occur in the phase measurement. The phase φ ′ calculated by the phase shift method is approximately expressed by Expression (3) when the shift amount of the phase shift from an integral multiple of π / 2 is small. φ ′ = φ + a · cos (2φ) + b (3) where φ is the true value of the phase. Further, the amplitude a and the constant term b change depending on the state of vibration during measurement and the driving error of the piezo element, and have different values each time measurement is performed.

According to the present invention, the error term of the equation (3) is detected in-situ for each test surface measurement by measuring the phase distribution of the interference fringes formed by the error detecting light beam, and the test surface is measured. Is to correct the shape measurement error.

The measuring method will be described step by step. Let the coordinates of the pixel on the two-dimensional image detector be (X, Y). A: The test surface holder 7 is aligned such that the number of interference fringes between the measurement light beam for measuring the test surface and the reference light beam is as small as possible. The tilt of the reflection plane member 8 is adjusted so that the number of interference fringes due to the error detecting light beam is 0.5 or more and one pitch of the interference fringes is 4 pixels or more of the two-dimensional image detector. The reason why the number of interference fringes due to the error detecting light beam is 0.5 or more is to enable observation of a phase range of π or more, and one pitch of the interference fringes is set to 4 of the two-dimensional image detector.
The reason why the number of pixels is equal to or larger than the number of pixels is to obtain a sufficient resolution. : Execute the phase shift interferometry in this state. : From the phase distribution φ ′ _A (X, Y) of the interference fringes of the error detecting light beam obtained in step (1), the amplitude a and the constant term b in equation (3) are obtained.

Considering that the reflecting portion of the reflecting flat member 8 is a plane, according to equation (3), the phase distribution of the interference fringes of the error detecting light beam is expressed as follows. φ ′ _A (X, Y) = C ₀ + C _X X + C _Y Y + a · cos {2 (C ₀ + C _X X + C _Y Y)} + b (4) By fitting the above equation to the phase distribution of the interference fringes, C ₀ , C _X , C _Y
At the same time, the values of the amplitude a and the constant term b are obtained.

[0034] Incidentally, as impart a known tilt planar reflecting member 8, after a known and C _x, C _y in (4) to perform a fitting of the equation (4), and a b may be obtained.

The "amplitude of the phase measurement error" referred to in this specification is this a. When the amplitude a is small: It is determined that the influence of the phase shift error on the acquired data is small, and no correction is performed. In the case where the amplitude a is large: For the phase distribution φ ′ (X, Y) of the interference fringe of the measurement light beam for measurement of the surface to be measured and the reference light beam measured in Solve to find the true phase distribution φ (X,
Y).

In order to perform more accurate correction, it is desirable to minimize the linear error of the phase shift amount. That is, the phase shift amount is accurately adjusted to an integral multiple of π / 2, and the phase calculation is performed using Equation (1), or the linear coefficient u of the phase shift is accurately measured to obtain Equation (2).
It is desirable to perform the phase calculation using

If the surface shape error between the Fizeau plane member 5 and the reflection plane member 8 cannot be ignored, it is necessary to correct the influence of the surface shape error. The measurement procedure in that case is shown below. ': The test surface holder 7 is aligned so that the number of interference fringes between the measurement light beam for measuring the test surface and the reference light beam is minimized. ': Adjust the tilt of the reflecting plane member 8 so that the interference fringes due to the error detection light beam become a single fringe. ). Since the interference fringes become a single fringe, the phase width Δφ in the image is sufficiently small, and the change at each point of the error term acos (2φ) + b is sufficiently small, and the constant component and Can be considered. This error term is subtracted as a constant in the process' described later. The phase shift interferometry is executed in the state of ':', and the phase distribution φ ' _B (X, Y) of the interference fringes due to the error detecting light beam is measured. ': A phase distribution E (X, Y) is obtained by removing the constant component and the tilt component from the phase distribution φ' _B (X, Y). That is, D ₀ + D _x X + D _y Y is fitted to the phase distribution φ ′ _B (X, Y), and the residual term at that time is replaced with the phase distribution E (X, X, Y). The reason why the constant component and the tilt component are removed is that these components are separated at the time of fitting performed later, so that only components that cannot be separated at that time are determined.

': The tilt of the reflection plane member 8 is adjusted so that the number of interference fringes due to the error detecting light beam is 0.5 or more and one pitch of the interference fringes is 4 pixels or more of the two-dimensional image detector. . ': Execute phase shift interferometry in this state. Then, the phase distribution φ ′ _A (X, Y) of the interference fringes of the error detecting light beam is obtained. ': Δφ' (X, Y ) = φ ' to calculate the _{A (X, Y) -E (} X, Y), determining the value of the amplitude a and the constant term b in Formula (3).

Φ ′ _A (X, Y) obtained by ′ is φ ′ _A (X, Y) ≒ C _0A + C _XA X + C _YA Y + E (X, Y) + a · cos {2 ( C _0A + C _XA X + C _YA Y)} + b (5) Therefore, Δφ ′ (X, Y) is represented by the following equation, and the amplitude a and the constant term b are obtained by fitting. Δφ ′ (X, Y) = C _0A + C _XA X + C _YA Y + a · cos {2 (C _0A + C _XA X + C _YA Y)} + b (6) When the amplitude a is small: It is determined that the influence of the phase shift error on the acquired data is small, and no correction is performed. When the amplitude a is large For the phase distribution φ ′ (X, Y) of the interference fringe of the measurement light beam for measuring the surface to be measured and the reference light beam measured by “:”, the nonlinear expression of the formula (3) is used for each pixel. Solve the equation to find the true phase distribution φ
Find (X, Y).

In the present embodiment, the measurement error of the shape of the test surface is corrected by obtaining the amplitude a and the constant term b of the equation (3). However, the measurement is performed a plurality of times, and the value of the amplitude a is sufficiently small. May be obtained.
Further, an error may be intentionally added to the driving of the test surface holder by the piezo to create a state where the amplitude a becomes small, and the data at that time may be acquired.

According to the present embodiment, it is possible to measure a plane having a large shape error and a surface to be inspected which has been subjected to aspheric processing with a small sag amount on the plane with high accuracy.

Next, a description will be given of a second embodiment in which the present invention is applied to a spherical interferometer using a Fizeau lens. In this embodiment, only the Fizeau lens, the test surface, and the test surface holder are different from those of the first embodiment, and therefore those portions are the same as those of the first embodiment using FIG. Will be described with reference to FIG.

The beam diameter of the linearly polarized light beam emitted from the laser light source 1 is converted by the beam expander 2 and
It is incident on S3. The polarization plane of the light beam L is selected to be reflected by the PBS 3. The light beam reflected by the PBS 3 enters the Fizeau lens 11 via the quarter-wave plate 4, and is converted into a spherical wave. The light beam transmitted through the reference spherical surface 11a of the Fizeau lens 11 is used as a measurement light beam on the surface 12 to be measured.
Incident on. The measurement light beam reflected by the test surface 12 again enters the PBS 3 through the Fizeau lens 11 and the quarter-wave plate 4. The measurement light beam passes through the quarter-wave plate twice in a reciprocating manner and the polarization plane is rotated by 90 degrees, so that it passes through the PBS 3. P
The beam diameter of the measurement light beam transmitted through the BS 3 is converted by the beam diameter conversion optical system 9 and is incident on the two-dimensional image detector 10.

On the other hand, the reference surface 11a of the Fizeau lens 11
Is reflected as a reference light beam through a quarter-wave plate 4,
It is incident on PBS3. The reference light beam passes through the quarter-wave plate twice in a round trip and the polarization plane is rotated by 90 degrees.
Through. The beam diameter of the reference light beam transmitted through the PBS 3 is converted by the beam diameter conversion optical system 9 and enters the two-dimensional image detector 10, where interference fringes between the measurement light beam and the reference light beam are detected.

In this embodiment, the reference plane member 13 for forming the error detecting light beam is fixed to the Fizeau lens 11, and the light beam also enters the reference plane 13 a of the reference plane member 13. Thus, the phase relationship between the reference light beam and the reference light beam for error detection is kept completely constant.

The error detecting reference light beam reflected by the reference plane 13a passes through the 1/3 wavelength plate 4, passes through the PBS 3, is converted in beam diameter by the beam diameter converting optical system 9, and is subjected to the two-dimensional image detector 10 Incident on. On the other hand, the measurement light flux for error detection transmitted through the reference plane 13a is incident on the reflection plane member 15 provided on the surface holder 14 to be measured. The measurement light flux for error detection reflected by the reflection plane member 15 is again transmitted to the reference plane member 1.
The light enters the PBS 3 via the 3/4 wavelength plate 4. PBS3
The measurement light flux for error detection which has passed through the
Then, the beam diameter is converted into the two-dimensional image detector 10, and the interference fringes of the measurement light beam for error detection and the reference light beam for error detection are detected.

The reflecting plane member 15 has a sufficiently small surface shape error with respect to the shape measurement accuracy of the surface 12 to be measured.
That is, it is preferable that the reflection flat member 15 is formed so as to have a sufficient flatness. However, even if the flatness is not satisfied, it is possible to calculate and correct the shape error of the reflection plane member 15 from the absolute plane as described above. It is possible. Further, since the reflection flat member 15 is mounted on the same test surface holder 14 as the test surface 12, the phase relation between the measurement light beam and the error detection measurement light beam is always kept constant. However, in order to make the number of interference fringes of the error detecting light beam variable, the reflecting plane member 15 has a mechanism capable of adjusting the tilt with respect to the surface 12 to be measured.

Next, a method of correcting a surface shape measurement error will be described. What is different from the plane measurement of the first embodiment is that
The point is that the normal direction and the phase shift direction at each point on the test surface are different, that is, there is a linear shift in the phase shift amount. The detection of the phase measurement error using the error detecting light beam and the like are the same as in the first embodiment. That is, the amplitude a of the error and the constant term b are obtained by the above-mentioned steps or the steps of '~'.

However, the nonlinear equation to be solved at the time of correcting the measurement error of the shape of the test surface is changed to the following equation. φ ′ = φ + {a · cos (2φ) + b} · cos θ (7) where θ is the angle between the normal line at each point on the surface to be measured and the optical axis (phase shift direction).

In order to perform more accurate measurement, equation (2) is used for the phase calculation equation, and the linear coefficient u is expressed as u = u ₀ cos
It is desirable to be θ. Here, u ₀ is a linear coefficient at θ = 0 (phase shift direction), and if the amount of phase shift is accurately adjusted, u ₀ = 1.

In the present embodiment, the amplitude a and the constant term b of the equation (3) are obtained to correct the measurement error of the shape of the test surface.
The measurement may be performed a plurality of times to obtain the test surface shape data when the value of the amplitude a is sufficiently small. Further, an error may be intentionally added to the driving of the test surface holder by the piezo to create a state where the amplitude a becomes small, and the data at that time may be acquired.

According to the present invention, it is possible to measure a spherical surface having a large shape error and an aspherical surface having a small aspherical amount with high accuracy.

Next, a third embodiment in which the present invention is applied to a spherical interferometer using a Fizeau lens will be described.
FIG. 4 shows the configuration of the Fizeau lens, the test surface, and the test surface holder. As the error detecting light beam, the reflected light from the reference spherical surface 11a of the Fizeau lens and the reflected light from the reflective spherical member 17 provided on the test surface holder 16 are used.

Since the reference light beam and the error detection reference light beam are formed by the same Fizeau lens 11, the phase relationship between the reference light beam and the error detection reference light beam is always kept constant. Further, since the reflective spherical member 17 and the test surface 12 are mounted on the same test surface holder 16, the phase relation between the measurement light beam and the error detection measurement light beam is always kept constant.

The positional relationship between the reflective spherical member 17 and the test surface 12 in the optical axis direction is such that when alignment is performed so that the number of interference fringes of the light beam for measurement on the test surface is minimized, the interference fringe of the light beam for error detection is reduced. Are adjusted so that the number of stripes is 0.5 or more and one pitch of interference fringes is 4 or more pixels of the two-dimensional image detector.

The measuring method will be described step by step. Let the coordinates of the pixel on the two-dimensional image detector be (X, Y). ": The test surface holder 16 is aligned so that the number of interference fringes between the measurement light beam for measuring the test surface and the reference light beam is minimized.": The phase shift interferometry is executed in this state. Phase distribution φ ' _{A of the} interference fringes for error detection obtained in ":"
From (X, Y), the tilt component and the focus component are removed, the amplitude a of Expression (3) is obtained, and it is determined whether the value of a is sufficiently smaller than the target accuracy. The removal of the focus component is performed, for example, by defining θ as the angle between the optical axis and the surface normal, and D (sin ² θ +
sin ⁴ θ / 4) is performed by fitting D as a parameter. The steps of ":" and "" are repeated until it is determined that the value of a is sufficiently small, and the phase distribution of the measured interference fringes when the value of a is sufficiently small is adopted as the measurement data.

The phase distribution of the interference fringes of the light beam for error detection is obtained by setting the linear coefficient of the phase shift amount to u = u ₀ cos θ _A.
It is calculated by the phase calculation formula of Expression (2). Here, θ _A is the angle between the light beam incident on the reflective spherical member 17 and the optical axis.

If the surface shape error between the reference spherical surface and the reflective spherical member of the Fizeau lens cannot be ignored, it is necessary to correct the influence of the surface shape error. In that case, the phase distribution of the interference fringes measured interference fringes of advance error detecting light beam is arranged so that the fringe color, extracting a component caused by the shape error, the phase distribution at the time of error detection phi _'A ( X, Y).

Also in the present embodiment, it is possible to correct the shape data of the surface to be detected by obtaining the amplitude a and the constant term b of the equation (3) from the phase distribution of the interference fringes of the light beam for error detection. In this case, for the phase distribution φ ′ (X, Y) of the interference fringe of the light beam for measuring the surface to be measured, the nonlinear phase equation (7) is solved for each pixel to obtain the true phase distribution φ (X, Y). ). φ ′ = φ + {a · cos (2φ) + b} · (cos θ / cos θ _A ) (7) where θ is a normal line and an optical axis (phase shift direction) at each point on the surface 12 to be measured. Angle.

Further, an error may be intentionally added to the driving of the test surface holder by the piezo to create a state where the amplitude a becomes small, and the data at that time may be obtained.

The present invention relates to a Null lens, a zone plate,
The present invention is also applicable to an interferometer using a Null element having a reference plane portion and an aspherical portion. The correction of the test surface measurement error and the like are the same as in the second embodiment.

FIG. 5 shows a configuration in the case of an interferometer using a Null lens. Null lens 21 and Fizeau plane member 22
Is integrated. Thus, the phase relationship between the reference light beam and the reference light beam for error detection is kept completely constant.

The light beam transmitted through the reference plane 22a of the Fizeau plane member 22 enters the Null lens 21 as a measurement light beam, is converted into a predetermined wavefront, and is incident on the surface 23 to be measured.
On the other hand, the light beam reflected by the reference plane 22a is used as a reference light beam. The reflected light from the reference plane 22a of the Fizeau plane member and the reflected light from the reflective plane member 15 provided on the test surface holder 14 are used as the error detection light flux.
Since the test surface 23 and the reflection flat member 15 are mounted on the same test surface holder 14, the phase relationship between the measurement light beam and the error detection measurement light beam is always kept constant. However, in order to make the number of interference fringes of the error detecting light beam variable, the reflecting plane member 15 has a mechanism capable of adjusting the tilt with respect to the surface 23 to be measured.

FIG. 6 shows a configuration in the case of an interferometer using a zone plate. The transmitted diffracted light converted into a predetermined wavefront by the pattern forming portion 31a of the zone plate 31 is incident on the surface 23 to be measured as a measurement light beam. On the other hand, zero-order light reflected from the zone plate 31 is used as a reference light beam. In the error detection light beam, a portion 3 where the pattern of the surface on which the pattern of the zone plate 31 is formed is not formed.
The reflected light from 1b and the reflected light from the reflective flat member 15 provided on the test surface holder 14 are used.

Since the part 31b of the zone plate 31 is used for forming the error detecting light beam, the phase relationship between the reference light beam and the error detecting reference light beam is kept completely constant. Further, since the test surface 23 and the reflection flat member 15 are attached to the same test surface holder 14, the phase relationship between the measurement light beam and the error detection measurement light beam is always kept constant. However, in order to make the number of interference fringes of the error detecting light beam variable, the reflecting plane member 15 has a mechanism capable of adjusting the tilt with respect to the surface 23 to be measured.

FIG. 7 shows a configuration in the case of an interferometer using a Null element consisting of a reference plane portion and an aspherical portion. Null element 4
The light beam transmitted through the reference plane portion 41a is converted into a predetermined wavefront by the aspheric surface portion 41b as a measurement light beam,
Incident on. On the other hand, the light beam reflected by the reference plane portion 41a is used as the reference light beam. The reflected light from the reference plane portion 41a of the Null element 41 and the reflected light from the reflecting plane member 15 provided on the test surface holder 14 are used as the error detection light flux.

Since a part 41a of the Null element 41 is used for forming the error detecting light beam, the phase relationship between the reference light beam and the error detecting reference light beam is kept completely constant. The test surface 23 and the reflection flat member 15 are the same as the test surface holder 1.
4, the phase relationship between the measurement light beam and the error detection measurement light beam is always kept constant. But,
In order to make the number of interference fringes of the error detecting light beam variable, the reflecting plane member 15 has a mechanism capable of adjusting the tilt with respect to the surface 23 to be measured.

The amplitude a in the equation (3) may be obtained from the PV value of the phase distribution of the interference fringes of the error detecting light beam after the tilt component is removed. At this time, the sign of the amplitude a is
In the phase distribution of the interference fringes before the tilt component removal, the determination is made based on the sign of the phase after the tilt component removal in the pixel where φ ′ ≒ 0. When correcting the measurement data of the surface to be measured, the constant term b =
The nonlinear equation may be solved with 0.

The formulas (1) and (2) have been described as examples of the phase shift method, but the present invention can be applied to the case where the phase is calculated by another formula. In particular, it has been confirmed that when there is a linear error in the phase shift amount, the error correction of the present invention functions more effectively when the phase shift amount is -π to π than when it is 0 to 2π. The phase calculation formula in this case is as follows. tanφ · sin (πu / 2) = [B ₀ − (B ₋₂ + B ₂ ) / 2] / (B ₁ −B ₋₁ ) (8) where B ₀ is the interference pattern in the initial phase to be obtained. of the intensity, B _-2 is - [pi], B _-1 is -π / 2, B ₁ is + [pi /
2, B ₂ is an intensity of the interference fringes when given a phase shift of + [pi.

In the embodiments described above, only the influence of the translation in the optical axis direction between the reference surface and the test surface has been considered.
If the tilt must be considered, the correction is performed by expressing the amplitude a and the constant term b as a linear function of the coordinates (X, Y), with the coordinates in the direction orthogonal to the optical axis as (X, Y). Specifically, the amplitude a and the constant term b at a plurality of locations of the error detecting reflective member arranged around the surface to be measured are obtained, and the coordinates (X,
The fitting is performed as a linear function of Y), and the function a
(X, Y) and b (X, Y) are obtained, and correction is performed for each coordinate on the test surface.

The present invention can be applied to a point-diffraction-interferometer (JP-A-2-228505) based on a wavefront diffracted by a pinhole. In this case, the configuration of the test surface holder and the like is the same as that of FIG. 4 in the third embodiment. In this case as well, data obtained when the value of the amplitude a in Expression (3) becomes small may be obtained, or the phase measurement error may be corrected using the detected amplitude a and the constant term b. Is also good.

Further, the wavefront forming means disclosed in Japanese Patent Application No. 268576/1998 discloses a method of forming a wavefront between a non-spherical surface to be measured and a reference prototype. The present invention can also be applied to a case where a wavefront is formed and a comparative measurement is performed between a test aspheric surface and a reference prototype. By applying the present invention, high-precision measurement can be performed without using only one stripe, so that a test surface having a large amount of aspherical surface can be measured with high accuracy.

[0073]

As described above, in the first and second aspects of the present invention, the lens, the mirror, and the like are removed by removing the influence of vibration and the like on the phase shift method. A surface shape measuring device capable of measuring with high accuracy can be provided.

Particularly, even when the width of the phase distribution of the interference fringes is wide, high-precision measurement is possible. Therefore, it is possible to measure a plane and a sphere having a large shape error and an aspheric surface having a small aspheric amount with high accuracy. Will be possible.

In the invention according to the third aspect, the frequency of occurrence of a state where the amplitude of the phase measurement error is smaller than the predetermined value increases, and the second means can be used effectively.

According to the fourth aspect of the present invention, the detection accuracy of the phase measurement error can be made sufficiently higher than the detection accuracy of the phase measurement error required to obtain a desired measurement accuracy of the shape of the test surface. The shape of the test surface can be accurately measured.

[Brief description of the drawings]

FIG. 1 is a diagram showing a basic configuration of an embodiment of the present invention.

FIG. 2 is a diagram showing an outline of a Fizeau surface portion and a test surface portion of the surface shape measuring apparatus for planar measurement according to the first embodiment of the present invention.

FIG. 3 is a diagram showing an outline of a Fizeau surface portion and a test surface portion of a spherical shape measuring surface shape measuring apparatus according to a second embodiment of the present invention.

FIG. 4 is a diagram showing an outline of a Fizeau surface portion and a test surface portion of a spherical shape measuring surface shape measuring apparatus according to a third embodiment of the present invention.

FIG. 5 is a diagram showing an outline of a Fizeau surface portion and a test surface portion of a surface shape measuring apparatus using a Null lens.

FIG. 6 is a view showing an outline of a Fizeau surface portion and a test surface portion of the surface shape measuring device using a zone plate.

FIG. 7 is a diagram showing an outline of a Fizeau surface portion and a test surface portion of a surface shape measuring apparatus using a Null element including a reference plane portion and an aspherical surface portion.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Laser light source, 2 ... Beam expander, 3 ... Polarization beam splitter, 4 ... 1/4 wavelength plate, 5 ... Fizeau plane member, 5a ... Reference surface area, 5b ... Outside reference surface area,
6 ... test surface, 7 ... test surface holder, 8 ... reflection flat member,
9: beam diameter conversion optical system, 10: two-dimensional image detector, 1
Reference numeral 1 denotes a Fizeau lens, 11a denotes a reference surface, 12 denotes a test surface,
Reference numeral 13: Reference plane member, 13a: Reference plane, 14: Test surface holder, 15: Reflection plane member, 16: Test surface holder, 17: Reflection spherical member, 21: Null lens, 22: Fizeau plane member, 22a ... Reference plane, 23 ... test surface, 3
1: Zone plate, 31a: Pattern forming part, 31b
… Part where no pattern is formed, 41… Null element,
41a: Reference plane portion, 41b: Aspheric portion

Claims (4)

[Claims] 1. A part of a light beam emitted from a light source is radiated to a surface to be measured as a measurement light beam, and the measurement light beam reflected by the surface to be measured and a part of a light beam emitted from the light source are included. A surface profile measuring apparatus for causing a reference light beam having a predetermined wavefront to interfere with each other and measuring a phase distribution of an interference fringe (measurement fringe) generated by the interference by a phase shift interferometry, wherein (a)
Means for forming an error detection measurement light beam having a constant phase relationship with the measurement light beam reflected by the test surface, and (b) an error detection reference light beam having a constant phase relationship with the reference light beam. Means for forming, and (c) causing the measurement light beam for error detection and the reference light beam for error detection to interfere with each other, and changing the phase distribution of interference fringes (error detection interference fringes) caused by the interference to the phase distribution of the measurement interference fringes. Means for simultaneously observing, (d) means for obtaining a phase measurement error from measured data of the phase distribution of the error detection interference fringes, and (e) using the obtained phase measurement errors, Means for correcting a measurement error of the shape of the test surface. 2. A method according to claim 1, further comprising: irradiating a part of a light beam emitted from the light source to the surface to be measured as a measurement light beam, and measuring a light beam reflected by the surface to be measured and a part of the light beam emitted from the light source. A surface profile measuring apparatus for causing a reference light beam having a predetermined wavefront to interfere with each other and measuring a phase distribution of an interference fringe (measurement fringe) generated by the interference by a phase shift interferometry, wherein (a)
Means for forming an error detection measurement light beam having a constant phase relationship with the measurement light beam reflected by the test surface, and (b) an error detection reference light beam having a constant phase relationship with the reference light beam. Means for forming, and (c) causing the measurement light beam for error detection and the reference light beam for error detection to interfere with each other, and changing the phase distribution of interference fringes (error detection interference fringes) caused by the interference to the phase distribution of the measurement interference fringes. Simultaneously observing means, (d) means for obtaining a phase measurement error from measurement data of the phase distribution of the error detection interference fringes, and (e) when the amplitude of the obtained phase measurement error is smaller than a predetermined value, Means for adopting measurement data of the shape of the test surface obtained from the measured interference fringes as the shape of the test surface. 3. The surface shape measuring apparatus according to claim 2, further comprising means for intentionally adding an error to a phase shift amount to be given by the phase shift interferometry. 4. One of claims 1 to 3
The surface shape measuring apparatus according to the item, wherein the means for forming the error detection reference light beam, and the surface shape error of the means for forming the error detection measurement light beam, the desired surface shape measurement accuracy with respect to the desired Surface shape measuring device characterized by being sufficiently small.